IMMUNOLOGY

CORE NOTES
MEDICAL IMMUNOLOGY 544
FALL 2011

Dr. George A. Gutman

SCHOOL OF MEDICINE
UNIVERSITY OF CALIFORNIA, IRVINE
(Copyright) 2011 Regents of the University of California

TABLE OF CONTENTS
CHAPTER 1
CHAPTER 2
CHAPTER 3
CHAPTER 4
CHAPTER 5
CHAPTER 6
CHAPTER 7
CHAPTER 8
CHAPTER 9
CHAPTER 10
CHAPTER 11
CHAPTER 12
CHAPTER 13
CHAPTER 14
CHAPTER 15
CHAPTER 16
CHAPTER 17
CHAPTER 18
CHAPTER 19
CHAPTER 20
CHAPTER 21
CHAPTER 22
CHAPTER 23
APPENDIX

GLOSSARY
INDEX

INTRODUCTION...................................................................................... 3
ANTIGEN/ANTIBODY INTERACTIONS .............................................. 9
ANTIBODY STRUCTURE I .................................................................. 17
ANTIBODY STRUCTURE II................................................................. 23
COMPLEMENT ...................................................................................... 33
ANTIBODY GENETICS, ISOTYPES, ALLOTYPES, IDIOTYPES..... 45
CELLULAR BASIS OF ANTIBODY DIVERSITY:
CLONAL SELECTION.................................................................. 53
GENETIC BASIS OF ANTIBODY DIVERSITY................................... 61
IMMUNOGLOBULIN BIOSYNTHESIS ............................................... 69
BLOOD GROUPS: ABO AND Rh ......................................................... 77
CELL-MEDIATED IMMUNITY AND MHC ........................................ 83
CELL INTERACTIONS IN CELL MEDIATED IMMUNITY .............. 91
T-CELL/B-CELL COOPERATION IN HUMORAL IMMUNITY ...... 105
CELL SURFACE MARKERS OF T-CELLS, B-CELLS
AND MACROPHAGES............................................................... 111
CELL INTERACTIONS IN HUMORAL RESPONSES: THE
CARRIER EFFECT...................................................................... 117
LYMPHOID TISSUE STRUCTURE.................................................... 123
ONTOGENY OF THE IMMUNE SYSTEM ........................................ 129
TOLERANCE........................................................................................ 135
AUTOIMMUNITY: BREAKDOWN OF SELF-TOLERANCE .......... 143
IMMUNODEFICIENCY ....................................................................... 147
IMMEDIATE HYPERSENSITIVITY: ALLERGY ............................. 153
VACCINATION .................................................................................... 163
TUMOR IMMUNOLOGY .................................................................... 169
1. PRECIPITIN CURVE .................................................................... 175
2. LABELING OF ANTIBODIES ..................................................... 177
3. OUCHTERLONY ANALYSIS ..................................................... 179
4. ABSORPTION AND AFFINITY PURIFICATION...................... 181
5. RADIOIMMUNOASSAY (RIA)................................................... 185
6. EQUILIBRIUM DIALYSIS ........................................................... 189
7. CROSS-REACTIVITY .................................................................. 195
8. COMPLEMENT FIXATION ASSAY........................................... 197
9. SOUTHERN BLOTTING.............................................................. 199
10. GENETICS OF INBREEDING ..................................................... 201
11. MIXED LYMPHOCYTE REACTION.......................................... 205
12. PLAQUE-FORMING CELL ASSAY............................................ 207
13. HYBRIDOMAS ............................................................................. 209
................................................................................................................ 213
................................................................................................................ 225
1

CHAPTER 1
INTRODUCTION

Resistance to infectious diseases relies on both INNATE and ADAPTIVE modes of

immunity. While both are effective and significant, a major focus of this course is
ADAPTIVE IMMUNITY, that mode of immunity which exhibits SPECIFICITY and
MEMORY.
Two systems of adaptive immunity protect all vertebrates, namely CELLULAR and
HUMORAL immunity. The consequences of both of these classes of immune responses
may be harmful as well as beneficial, and are mediated by cells of the highly distributed
LYMPHOID SYSTEM.
INNATE versus ADAPTIVE IMMUNITY
INNATE IMMUNITY
The body's first line of defense against pathogenic organisms (including bacteria, fungi
and viruses) is the physical barrier provided by the skin, by the epithelium and mucus
secretions of the alimentary tract and lungs, etc. This level of protection, however, is
relatively non-specific; it distinguishes little, for example, between the bacterial organisms
Staphylococcus and Streptococcus, or between the viral agents causing polio and smallpox.
A next level of defense is manifested by a variety of cells and serum molecules which may
promote ingestion and killing of potentially infectious organisms, cells including
macrophages, neutrophils and dendritic cells, and molecules including complement and
defensins. These modes of protection are present in all healthy individuals, and are essentially
unchanged following repeated challenges by the offending pathogens - that is to say they do
not display memory, and are collectively referred to as INNATE IMMUNITY. Mediators of
innate immunity contribute to the complex process of development of INFLAMMATION.
However, as we will discuss later, the mechanisms of non-specific inflammation overlap and
interact extensively with those mediating adaptive immune responses, which will be clearly
illustrated in the modes of action of dendritic cells, macrophages and complement.
Although relatively non-specific, innate immunity is highly effective and centrally
important to our well-being, as evidenced by the consequences of damage to this system by
trauma (e.g. wounds which damage epithelium and may get infected, ionizing radiation
which can inhibit the inflammatory response) or by disease (e.g. emphysema, which causes
greatly increased sensitivity to bacterial infection in diseased lungs). In fact, as we will see,
loss of effective innate immunity can have as deadly consequences as any loss of adaptive
immune function.
ADAPTIVE IMMUNITY
What is classically meant when referring to the "immune system", however, is not the
non-specific manifestations of innate immunity, but the complex system of immune reactions
known as ADAPTIVE IMMUNITY (including both humoral and cellular immunity, defined

below), which display the closely related features of SPECIFICITY and MEMORY. This
course, and these notes, will attempt to define the nature of, and the molecular basis of both
of these features.

The discipline of immunology can be approached from two distinct perspectives:

1) Historical/Medical outlook: Resistance to Infectious Disease. The adaptive immune
system can confer specific resistance to many infectious diseases, e.g., smallpox. This
example illustrates two key features of immune reactions, namely specificity and
memory. Having recovered from smallpox (or having been vaccinated) makes one
resistant ("immune") to being infected with smallpox later - the immune system
exhibits memory. Resistance to smallpox, however, does not make a person resistant
to measles, mumps, diphtheria or other diseases caused by unrelated organisms - the
immune system thus shows specificity.
2) Biological approach: Recognition of "Self" Versus "Non-self". Quite apart from its
importance in resistance to infectious disease, the immune system has been of
tremendous interest to biologists interested in the nature and mechanisms of
immunological specificity, one aspect of which can be regarded as an organism's ability
to distinguish self from non-self. The single-celled amoeba, for instance, ingests food
by phagocytosis; how does it distinguish a particle of food from one of its own
pseudopods? (Note that the same question may be asked of a macrophage.)

"Self"

"Non-self"

Figure 1-1
For multicellular organisms the problem becomes even more complex -- they must be
capable of recognizing their own diversity of normal cell types as self while at the same time
retaining the ability to recognize foreign particles and cells and reacting against them.
The biological concept of specificity, particularly in the context of cell surface
recognition, extends into many other areas, e. g. that of control of cell proliferation and
differentiation. As a result, the problem of immunological specificity has attracted study by
many scientists whose basic interests lay in the areas of differentiation and tumor biology.

Lets define the following two terms in the context of adaptive immunity:
IMMUNITY - Acquired resistance to infectious disease displaying specificity at the
molecular level. Weve already noted that many factors other than the adaptive
immune system contribute to resistance to disease, for instance the barrier to
microorganisms provided by our skin and other membranes and phagocytic cells
("innate immunity"). These are not acquired, however, nor do they exhibit the
specificity required by this definition, and therefore they are not by themselves
considered an expression of the "adaptive immune system".
IMMUNE RESPONSE - Reactivity against a target displaying specificity at the
molecular level. The targets of such reactivity may be disease-producing organisms, or
may be completely harmless substances such as foreign red blood cells or foreign
serum proteins.
One major criterion for effective reactivity under normal
circumstances is that the target be "foreign" to the responding organism, although we
shall see that immune responses may be directed against self components as well.
We shall also learn of many factors that may affect the magnitude of immune responses
to various targets.
ROLES OF THE IMMUNE SYSTEM
Resistance to infectious disease. From a medical or evolutionary standpoint, this is highly
beneficial, and obviously a central role of the immune system. Deficiency in the ability to
mount effective immune responses leads to increased susceptibility to infection by bacteria,
fungi and viruses. (The largely discredited idea that the immune system also effectively
seeks out and destroys cells which are undergoing neoplastic transformation, i.e. "immune
surveillance", is discussed in Chapter 23.)
However, the immune system does not always act in a manner beneficial to the organism;
some immune responses result in considerable harmful effects and may be fatal, as
illustrated by the following examples:
Allergy. Immune responses to food and to plant and animal products in our environment
may result in the various manifestations of allergies. Hay fever and allergies to foods and
animal products are very common, and while they often are not very serious, they may
sometimes be life-threatening, such as in the case of severe asthmatic reactions or
anaphylactic shock.
Autoimmunity . The normal ability of the immune system to distinguish self and non-self
can be disrupted by a variety of influences, resulting in damaging and potentially lethal
reactivity to normal "self" components. Rheumatoid Arthritis (RA) and Systemic Lupus
Erythematosus (SLE) are two examples of many such autoimmune reactions.
Graft rejection; the rejection of foreign tissues and organ transplants is a normal
consequence of immunological specificity; however, the ultimate result of immune rejection
of a heart or liver transplant, for instance, may be fatal. Much research is stilled aimed at
discovering more effective methods to prevent immune rejection of grafts, while at the same
time maintaining the recipient's ability to resist infectious organisms.

THE LYMPHOID SYSTEM: ORGAN OF IMMUNITY

We will discuss later the many different cell types which are directly or indirectly
involved in immune responses. One cell type, however, the LYMPHOCYTE, is centrally
involved in all adaptive immune responses.
No single, localized organ is responsible for immune reactivity, but rather it involves
a wide variety of organs which include lymph nodes, spleen, Peyer's patches, tonsils,
thymus and bone marrow. These are collectively known as the LYMPHOID SYSTEM, by
virtue of the fact that they all contain large numbers of the white blood cells known as
lymphocytes. Each has a unique structure and role in immune responses, which we will
examine in some detail later (Chapter 16).
TWO SYSTEMS OF IMMUNITY PROTECT VERTEBRATES:
HUMORAL AND CELL-MEDIATED (CELLULAR ") IMMUNITY
Immune responses can generally be categorized as either humoral or cellular.
HUMORAL IMMUNE RESPONSES are those mediated by antibodies in various body
fluids ("humors"), including blood, saliva and the mucous secretions of the lungs and
intestinal tract (the nature and structure of antibodies will be discussed in Chapters 3 and 4,
and the cellular basis for humoral immunity in Chapter 15). In general, the humoral response
offers protection from infections caused by organisms which are extracellular, a category
which includes most bacteria as well as many of their toxic products (e.g. diphtheria and
tetanus toxins). Hay fever and food allergies are examples of humoral responses which are
harmful to the host.
CELL-MEDIATED IMMUNE RESPONSES are not mediated simply by antibodies, but
require the direct participation of immunologically reactive cells (to be discussed in Chapters
11 and 12). Cell-mediated immunity is responsible, in general, for resistance to infectious
organisms which are primarily intracellular. This includes resistance to viral infections as
well as to certain bacteria (e.g., the Mycobacterium responsible for tuberculosis). Immunity
to fungal infections and graft rejection are also largely the responsibility of the cellular
immune system.
The ability of immune serum to transfer humoral immunity promoted early studies which
identified the relevant effector molecules, the ANTIBODIES. In the next few chapters we
will examine the structure and function of antibodies, which are the mediators of humoral
immune responses, and proceed in later chapters to examine the features of cell-mediated
immune responses.

CHAPTER 1, STUDY QUESTIONS:

1.

What are the defining differences between INNATE and ADAPTIVE immunity?

2.

What are the differences between HUMORAL and CELL-MEDIATED immunity?

3.

What are some of the biological and medical consequences of immune reactions?

CHAPTER 2
ANTIGEN/ANTIBODY INTERACTIONS

See APPENDIX (1) THE PRECIPITIN CURVE; (2) LABELING OF ANTIBODIES

The defining characteristic of HUMORAL immune responses (which distinguishes
them from CELL-MEDIATED responses), is their ability to be transferred by serum, and
the proteins within serum which are responsible for such immunity are ANTIBODIES.
We can formulate intriguingly circular definitions for antibodies and ANTIGENS, and
note that the universal property of antibodies is their ability to specifically bind their
cognate antigens. The consequences of such binding, however, can vary considerably,
depending on the nature of the particular antigen and antibody involved.
We distinguish the PHYSICAL and the BIOLOGICAL PROPERTIES of
antibodies, and the properties of ANTIGENICITY versus IMMUNOGENICITY, and
introduce the concept of ADJUVANTS, substances which are capable of increasing
immunogenicity.

We'll begin by defining three important terms:

ANTIBODY - The molecule present in serum and other body fluids which mediates
humoral immunity, and which can bind specifically to an antigen. Serum which
contains antibodies (directed against one or more antigens) is termed an antiserum.
ANTIGEN - A molecule which can be specifically bound by an antibody (typically a
protein or carbohydrate recognized as "foreign").
EPITOPE (= antigenic determinant = "antigenic specificity") - The minimum
target structure on an antigen which is bound by a particular antibody molecule. A
particular antigen molecule may (and generally does) bear many different epitopes or
determinants, each of which can be a target for antibody binding.
(NOTE: Antibodies themselves can serve as antigens; human antibodies, for instance, are
"foreign" to rabbits, and can elicit rabbit antibodies to human antibody molecules. As we
will see later, the use of antibodies as antigens has been an extremely powerful tool for
understanding antibody structure and genetics.)

DEFINING HUMORAL IMMUNITY

Experimentally defining a humoral immune response involves demonstrating that such
immunity can be transferred by serum (or other fluids). The example below (Fig, 2-1)
illustrates some key features of humoral immunity.
Live
Pneumococcus

DIES

Naive
Killed
Pneumococcus
Naive

Immune

Live
Pneumococcus

wait
2
weeks

TRANSFER
SERUM

SURVIVES
Immune
"Active Immunity"

Live
Pneumococcus

SURVIVES

Naive
"Passive Immunity"

Figure 2-1
If a mouse is injected with a sufficient dose of live Pneumococcus bacteria, it will die
of infection within a few days. If, however, it has previously been injected with killed
organisms, not only does it not succumb to infection, but it will survive a subsequent
injection of a normally lethal dose of this organism; such a mouse has been immunized, and is
therefore said to be immune to Pneumococcus. Although not illustrated here, we can further
demonstrate that this resistance is specific the immune mouse will retain normal
susceptibility to some other organism to which it had not previously been exposed. Such
specificity establishes that the immunity we see is a result of the mouses adaptive immune
response.
Question: Does this resistance represent humoral immunity?
To find out, we take serum from the immune mouse and inject it into a non-immune
recipient, then inject a lethal dose of Pneumococcus. We find that this recipient survives this
treatment; serum from an immune mouse transfers immunity to a nave recipient. This
demonstrates that immunity to this organism is mediated by humoral immunity. (NOTE:
This does not, however, mean that resistance to all bacterial infections is mediated by
humoral immunity. As we will see in Chapter 12, transferring serum from a mouse which is
immune to another bacterium, Listeria (which is an intracellular pathogen), does not confer
resistance to nave recipients; such immunity is therefore not humoral.)
This illustration also serves to define two distinct modes of adaptive immunity,
namely ACTIVE IMMUNITY and PASSIVE IMMUNITY. Immunization of the mouse in
the second line of Fig. 2-1 results in a state of "active" immunity; the animal's own immune
system is responsible for resistance to the subsequent bacterial challenge. On the other hand,
transfer of serum, as in line 3 above, results in a state of "passive" immunity in the recipient;
such immunity is the result of the presence of transferred antibody (see below). The animal's
own immune system does not participate at all, and this immunity lasts only as long as
sufficient levels of antibody are present.
10

The substance present in immune serum which is responsible for transferring

immunity is antibody. In addition to transferring resistance to infection, these serum
antibodies can carry out a variety of other functions. For example, if immune serum is mixed
with a suspension of Pneumococcus, the bacteria will be seen to rapidly "clump" together.
This effect is known as agglutination, and is one of the many ways in which antibodies can
be detected and quantitated.
The various effects that antibodies may exhibit can be generally categorized as
physical effects, which depend only on the physical nature of the antibody and antigen, or
biological effects, which additionally depend on the particular biological properties of the
target antigen or other biologically active molecules which are involved.
PHYSICAL EFFECTS OF ANTIBODY
Agglutination. "Clumping" of a particulate antigen, e.g. bacteria or SRBC (sheep red
blood cells). Agglutination of red blood cells is a technique which has been widely
used in clinical and basic research as well as in the clinical laboratory, and is called
HEMAGGLUTINATION. Many soluble antigens can be made effectively particulate
by coating them onto SRBC or latex or other particles; the resulting clumping by
antibody is known as passive agglutination.
Precipitation. Interaction of antibody with a soluble antigen to form an insoluble
complex, e.g., with BSA (bovine serum albumin).
In liquid - the precipitate can be recovered by centrifugation and analyzed (see
APPENDIX 1, THE PRECIPITIN CURVE). If either the antigen or antibody is
radioactively labeled (see APPENDIX 2, LABELLING OF ANTIBODIES), it can be
used in a RadioImmunoPrecipitation (RIP) assay, first developed in the 1950s.
In agarose - if the antigen-antibody interaction takes place in a semi-solid medium
such as agarose, the resulting precipitate can be easily visualized. This is of special
significance in a configuration known as Ouchterlony Analysis (see APPENDIX 3,
OUCHTERLONY ANALYSIS).
Precipitation and agglutination are both consequence of cross-linking of antigens by
antibody into large complexes. The ability of antibodies to carry out this process implies that
each antibody can bind at least two antigen molecules, and that it can only occur if the
antigen molecule has two or more epitopes (determinants ") which can be recognized by that
antibody.
Binding. If an antigen is bound to a solid matrix (latex particles or a plastic dish, for
example), and if the antibody is labeled in some way (with a visible, radioactive or
enzyme molecule), binding of the antibody to its antigen can be easily and sensitively
measured.
If a radioactive label is used, the assay is called a solid-state
RadioImmunoAssay (RIA). With an enzyme-based label, on the other hand, it
becomes an Enzyme-Linked ImmunoSorbent Assay (ELISA). These solid state
assays (particularly ELISA's) have largely replaced precipitation and agglutination
assays in a wide variety of clinical and research applications.

11

BIOLOGICAL EFFECTS OF ANTIBODY

Protection from infectious disease. We have already seen in the Pneumococcus
example (Figure 2-1) how this manifestation of antibody can be assayed by transferring
serum from one animal to another.
Immobilization. An antibody directed against components of the flagellae of motile
bacteria or protozoa can cause these flagellae to stop moving. This results in the loss of
the organisms' ability to move around, and this loss of motility can be detected by
microscopic examination.
Cytolysis. If the target antigen is an integral component of the membrane of certain
sensitive cells, antibodies may cause disruption of the membrane and death of the cell.
This requires the participation of a collection of other serum components collectively
known as COMPLEMENT (see Chapter 5), and binding of these components to
antibodies is referred to as Complement Fixation.
If the antigen target is a red blood cell, this effect is known as hemolysis, which can
be readily detected visually. In the case of a bacterial cell target, the effect is referred
to as bacteriolysis.
If the target is a nucleated cell the effect is referred to as cytotoxicity, and may be
measured by release of a radioactive label incorporated into the cell (such as 51Cr),
exclusion of "vital" dyes such as Trypan Blue, or any of several other measures of cell
viability.
Opsonization. If the target antigen is particulate (e.g. a bacterium, or an antigencoated latex particle), bound antibodies may greatly increase the efficiency with which
the particles are phagocytosed by macrophages and other "scavenger" cells. This
improvement of phagocytosis is known as opsonization, and may be facilitated even
further by the presence of complement. As will be discussed later, opsonization is the
result of antibodies increasing the degree to which antigenic particles will "stick" to
phagocytic cells. This phenomenon has therefore been referred to as immune
adherence, and depends on the presence in the membranes of white blood cells of
specific receptors either for antibody (FcR, or "Fc-receptors") or for complement (CR,
or "complement receptors"), both of which will be discussed later (see Chapter 14, for
example).
ONE COMMON DEFINING PROPERTY OF ANTIBODIES:
ALL ANTIBODIES EXHIBIT SPECIFIC BINDING TO ANTIGEN
Different antibodies may show various combinations of effects; some antibodies may
precipitate but not interact with complement (and therefore not show cytolysis), some may be
opsonizing but not be capable of agglutination. The single common feature of all antibodies,
however, is that of specific recognition and binding to antigen. All other effects, physical or
biological, are secondary consequences of this specific binding. The structure of antibodies
and the basis of their ability to specifically bind antigen are the subjects of the next two
chapters (Chapters 3 and 4).

12

ANTIGENS, IMMUNOGENS AND HAPTENS

We have been discussing "antigens" as molecules (1) which can elicit antibody
production upon injection into an appropriate host; and (2) to which these antibodies can then
bind. The difference between these two properties is an important one which we will now
make explicit by defining two related but distinct terms:
IMMUNOGEN. A molecule which can elicit the production of specific antibody upon
injection into a suitable host.
ANTIGEN. A molecule which can be specifically recognized and bound by an
antibody.
These definitions imply that an immunogen must be an antigen, but an antigen is not
necessarily an immunogen. Lets illustrate this in the following table:
Substance

Molecular weight

Immunogen?

Antigen?

1) BSA
2) DNP
3) DNP10-BSA

"68,000"
~200
"70,000"

4) "clarified" BSA

68,000

yes
no
DNP - yes
BSA - yes
no (see Note)

yes
yes
yes
yes
yes

If we take a conventional preparation of purified bovine serum albumin (BSA) and

inject it into a mouse (line 1 in the table above), the mouse will produce antibodies which
will bind to BSA. BSA is therefore both an immunogen and an antigen.
If we take the small organic molecule dinitrophenol (DNP) and inject it into a mouse
(line 2), no antibodies will be produced which can bind DNP. DNP is therefore not
immunogenic; we will deal with its antigenicity shortly.
We can chemically couple DNP molecules to the protein BSA, yielding DNP-BSA. If
we inject this material into a mouse (line 3), we see that antibodies to BSA are elicited (as we
would expect), but also find antibodies which will bind specifically to the DNP groups on
BSA; we can further demonstrate that these anti-DNP antibodies will also bind free DNP (or
DNP coupled to any other molecule). Therefore, DNP-BSA is both immunogenic and
antigenic (with respect to both the DNP groups and the BSA itself), and the free DNP is also
antigenic, even though we have shown it is not immunogenic. DNP is an example of a
HAPTEN, a small molecule which is not immunogenic unless it is coupled to a larger
immunogenic CARRIER molecule, in this case BSA. (Such a hapten/carrier system will be
used in Chapter 14 to illustrate the mechanisms of cell interactions required to generate
humoral immune responses).
We can further demonstrate that the immunogenicity of BSA depends on the presence of
aggregates of BSA molecules. If we take a sample of our BSA and centrifuge it at high speed
we can remove any aggregated material, leaving behind only single, monomeric BSA
molecules in solution. If we immediately inject this "clarified" BSA into a mouse we find
that it does not elicit the production of antibodies (as seen in line 4); this monomeric BSA is
13

therefore not immunogenic (nor can it serve as an effective carrier for a hapten). It is still
antigenic, however, which we can show by reacting it with the anti-BSA antibodies which we
made against the non-clarified BSA (in line 1, for example).
(NOTE: "Clarified" BSA not only fails to induce antibody formation, but can induce a state
of TOLERANCE to BSA, defined as the specific inability of the mouse to respond to
subsequent injections of normally immunogenic BSA. The mechanism of such tolerance will
be discussed in Chapter 18.)
REASONS FOR LACK OF IMMUNOGENICITY
Substances may lack immunogenicity for a variety of reasons:
1) Molecular weight too low. Haptens, for example, are not immunogenic until they are
coupled to a high molecular weight carrier. There is no simple cutoff for required
molecular weight, however; we have already seen that even the 68,000 mw of BSA is
not sufficient to be immunogenic unless the molecules are aggregated into even larger
complexes.
2) Not foreign to host. The adaptive immune system normally responds only to "foreign"
substances. A sheep, for instance, will normally not make antibodies against its own
red blood cells (SRBC), although SRBC are highly immunogenic in mice. (The basis
of normal SELF-TOLERANCE is covered in Chapter 18).
3) Some molecules are intrinsically poor immunogens for reasons which are not well
understood. Lipids, in general, are poor immunogens, probably because they do not
have a structure rigid enough to be stably bound by antibodies. Nucleic acids are also
relatively weak immunogens, although they are nevertheless common targets for
antibodies present in various autoimmune diseases (discussed in Chapter 19)
HOW TO INCREASE IMMUNOGENICITY: ADJUVANTS
(See also CHAPTER 22)
An ADJUVANT is any substance which, when administered together with an antigen,
increases the immune response to that antigen. One of the most widely used adjuvants (in
animals but not in humans) is FREUND'S ADJUVANT, which consists of mineral oil, an
emulsifying agent, and killed Mycobacterium (the organism which causes tuberculosis). A
solution of the desired antigen in water or saline is homogenized with this oil mixture,
resulting in a water-in-oil emulsion which is injected into the recipient. Its powerful
adjuvant properties result from several factors:
1) The antigen is released from the emulsion over an extended period of time, causing a
continuous and more effective stimulation of the immune system. (Antigen given in
soluble form will typically be cleared in a matter of hours or days, whereas it can
persist for weeks or months in a depot created by the adjuvant.)

14

2) The Mycobacteria contain substances which non-specifically stimulate the immune

system, resulting in a higher level of response to the specific antigen. One of these
substances which has been extensively studied is Muramyl Dipeptide (MDP).
Although Freund's Adjuvant is not used in humans, other forms of adjuvant can be
used, such as alum precipitation of antigen, by which a soluble antigen is precipitated
together with aluminum hydroxide, resulting in particles of the salt coated with antigen. A
soluble antigen is thus converted to a particulate form, and again is released from the mixture
over an extended period of time. Substances such as purified MDP and others are also being
used to develop effective adjuvants which are less toxic, and therefore potentially usable in
humans (see Chapter 22)

CHAPTER 2, STUDY QUESTIONS:

1.

Define ANTIBODY, ANTIGEN, IMMUNOGEN and HAPTEN.

2.

How would you determine if a particular immune response is a HUMORAL

response?

3.

Describe assays which could be used to measure AGGLUTINATION,

PRECIPITATION, HEMOLYSIS and OPSONIZATION.

4.

Describe two antibody assays which require no antibody function other than specific
binding to an antigen.

5.

Define and distinguish ACTIVE versus PASSIVE immunity.

15

16

CHAPTER 3
ANTIBODY STRUCTURE I
See APPENDIX: (3) OUCHTERLONY ANALYSIS; (6), EQUILIBRIUM DIALYSIS;
(7) CROSS-REACTIVITY
Electrophoretic separation of serum proteins identifies the GAMMA-GLOBULIN
fraction as containing the majority of antibodies. Three terms which are often
confusingly interchanged are defined and distinguished (GAMMA-GLOBULIN,
IMMUNOGLOBULIN, ANTIBODY), as are two terms describing antibody/antigen
binding, AFFINITY and AVIDITY.
All antibodies are made up of one or more IgG-like subunits, each of which has
exactly two antigen-combining sites. The affinity of these sites for their antigen (defined
as the Keq of the binding reaction) is highly heterogeneous in any normal immune
response. While the avidity of an antibody (its ability to form stable complexes with
antigen) does depend on its intrinsic affinity, it also increases dramatically with an
increasing number of combining sites per antibody.

In order to determine the structure of antibodies, we first must have a way of isolating
these molecules in relatively pure form. Well begin by describing the general process of
serum fractionation, then go on to analyze the nature of antigen-antibody binding.
The many components of normal serum can be separated from one another by various
means:
Salt precipitation. Ammonium sulfate [(NH4)2SO4] as well as a variety of other salts
can be used to precipitate serum components; different proteins will precipitate at different
concentrations of salt, providing a convenient means of separating them. The fraction
containing most of the antibody activity generally precipitates at relatively low salt, at about
30-40% of saturated ammonium sulfate. This is a very widely used experimental method for
fractionation of serum components (and proteins in general).
Ethanol precipitation. Ethanol can also be used to precipitate serum components,
which come out of solution at different concentrations and under different conditions of ionic
strength, pH and temperature. This is a more elaborate procedure to carry out than salt
fractionation, but is the basis for Cohn Fractionation, which in modified form remains a
standard procedure for preparing serum protein fractions for clinical use more than sixty
years after its original description in the 1940s.
Electrophoresis. Different serum proteins migrate at varying rates in an electric field,
a property which can be used to separate them. While this procedure can be adapted for use
on a preparative scale, it is most commonly used for analysis.

17

A typical pattern generated by electrophoresis of a serum sample (e.g. on a filter paper strip)
is shown in Figure 3-1.

albumin

globulins

1 2

ELECTROPHORESIS OF NORMAL SERUM

Figure 3-1
Several important points emerge from this pattern:
1) Most serum proteins carry a negative charge, and therefore tend to migrate from the
point of origin (labeled "O") toward the anode, the positively charged electrode.
2) Four major peaks are seen in this example; these are named (from the anodal, or
positive, side) the albumin peak (which is by far the largest), followed by four globulin
peaks, 1- and 2-globulin, -globulin and -globulin.
3) This pattern is deceptively simple; serum actually contains hundreds of known proteins.
Thus, "-globulin" is not a single protein, but a mixture of many components which all
happen to migrate in a particular region on electrophoresis.
4) Most (but not all) antibodies migrate in the -globulin region.
5) The -globulin peak is markedly broader than the others, reflecting the high degree of
heterogeneity of the antibodies it contains. This heterogeneity is so great that some
antibody molecules in fact migrate in the positions characteristic of -globulin or globulin.
6) The -globulin peak is generally centered near the origin, labeled "O"; this reflects the
fact that antibodies as a group are relatively neutral, i.e. less highly charged than most
other serum components.

18

Three easily confused terms are all commonly used to refer to antibody molecules,
gamma-globulins, immunoglobulins and antibodies. To avoid this confusion let's explicitly
define each of them:
GAMMA-GLOBULIN -- Any molecule which migrates in the gamma-globulin peak
on electrophoresis. Most, but not all, antibodies are in this category, although the term is
often used to refer to antibodies in general. (Other serum components migrate in this region
as well; therefore, strictly speaking, not all gamma-globulins are antibodies.)
IMMUNOGLOBULIN -- A family of molecules (to which all antibodies belong) with
similar structures and physical properties. We shall see that these involve homologous amino
acid sequences, similar "domain" structures and similar quarternary structures (the ways in
which different polypeptide chains are joined into a larger functional unit).
ANTIBODY -- A molecule belonging to the Immunoglobulin family, with binding
specificity for a particular antigen. While all antibodies are immunoglobulins, most but not
all antibodies are gamma-globulins.
Note that our definition of "antibody" requires knowledge of the binding specificity
of the molecule. If one is dealing with an "antibody" molecule whose specificity is
not known, or is irrelevant, it is more accurate to refer to it simply as an
"immunoglobulin". (Common usage of these terms varies considerably, however.)
ANALYSIS OF THE ANTIBODY COMBINING SITE:
VALENCY, AFFINITY AND AVIDITY
If we immunize a rabbit with DNP-BSA, we can obtain an antiserum which contains
antibodies to both the hapten and the carrier protein. This antiserum will precipitate DNPBSA in addition to DNP-KLH (Keyhole Limpet Hemocyanin, an unrelated protein carrier).
If we attach DNP to SRBC (sheep red blood cells) or to latex particles, we can show that the
antiserum is capable of showing agglutination (and possibly hemolysis in the case of SRBC).
We can use these antibodies to the DNP hapten in order to learn about antibody
structure and function. Specifically, we will ask two questions:
1)
2)

How many hapten molecules can a single antibody molecule bind (i.e how many
combining sites does it have, or what is its valency)?
What is the strength of binding of the hapten to its combining site(s) on the
antibody molecule (i.e. what is the affinity of the combining site)?

We have previously made the prediction that in order for an antibody molecule to be
capable of precipitation or agglutination it must have at least two combining sites, in order to
permit cross-linking of the antigen into large, insoluble complexes. We can determine the
actual number of combining sites of our anti-DNP antibodies, as well as their affinity, by
several techniques; one of them, EQUILIBRIUM DIALYSIS, is discussed more fully in
APPENDIX 6, and we will use the results of such an analysis as the basis for our discussion
below.

19

RABBIT IgG ANTIBODIES HAVE TWO HAPTEN-COMBINING SITES

The structure of rabbit IgG antibodies represents the basic structure of all antibodies
and we can show by equilibrium dialysis that each anti-DNP antibody molecule can bind
exactly two DNP molecules. Thus, our minimum prediction of at least two combining sites is
fulfilled. Other kinds of antibodies can be shown to have more than two combining sites
(IgM and some IgA), but we will see that such antibodies are always made up of multiple
units of the basic "IgG-like" structure, each of which bears precisely two combining sites.
CONVENTIONAL ANTIBODIES ARE HETEROGENEOUS
WITH RESPECT TO AFFINITY
The DNP hapten is bound to each combining site by non-covalent forces, and the
strength of this binding is measured by the equilibrium constant of the binding reaction,
known as the AFFINITY. The antiserum we describe above contains anti-DNP antibodies
with many different affinities, typically ranging from 105 to 1010. (Antibodies certainly exist
with affinities outside this range, but such values are difficult to determine accurately due to
technical limitations.)
This antibody heterogeneity is a hallmark of the immune response, and has many
practical and theoretical implications (see discussions of Clonal Selection and Affinity
Maturation [Chapter 7], and Isotype Switching [Chapter 9]). The broadness of the gammaglobulin peak on serum electrophoresis (which we have already described) is one
consequence of this heterogeneity; in fact, a sharp, narrow gamma-globulin peak
(representing a homogeneous protein) is a pathological sign of a myeloma or other
monoclonal gammopathy. However, homogeneous antibodies known as HYBRIDOMAS,
or MONOCLONAL ANTIBODIES can be generated experimentally, and are important in
many research and clinical applications (see APPENDIX 13).
ANTIBODY AVIDITY: ABILITY TO FORM STABLE COMPLEXES WITH ANTIGEN
AFFINITY is a thermodynamically defined term representing the strength of
interaction of a single combining site with its hapten. Naturally produced antibodies always
have two or more sites, however, so that affinity does not tell the whole story with respect to
antigen-binding. A bivalent anti-DNP antibody, for example, can simultaneously bind to two
DNP haptens on a single BSA molecule, resulting in a much more stable complex than if it
only bound to a single site.
AVIDITY, on the other hand, is the term used to describe the ability of an antibody to
form stable complexes with its antigen. Avidity, of course, depends partly on affinity; all
other things being equal (which they rarely are), one IgG antibody with a higher affinity for
DNP than another will also have a higher avidity. However, various other factors also play a
role, such as the number and spacing of the epitopes on the antigen, the distance between the
combining sites on the antibody, and properties such as the "flexibility" of the particular
antibody molecule.
Avidity does not have a formal thermodynamic definition, and is most commonly
used only in a relative context (by demonstrating that one antiserum may exhibit a higher or
20

lower avidity than another). Nevertheless, in discussing the interaction of an intact antibody
(which is at least bivalent) with a conventional antigen (which is almost always highly
multivalent), one must almost always think in terms of "avidity" rather than "affinity". This
is of particular importance when considering the biological effectiveness of antibodies which
have more than two combining sites, such as serum IgM and some IgA.

CHAPTER 3, STUDY QUESTIONS:

1.

Define the terms ANTIBODY, IMMUNOGLOBULIN and GAMMA-GLOBULIN.

2.

How is EQUILIBRIUM DIALYSIS carried out, and what can it measure?

3.

Define and distinguish antibody AFFINITY and AVIDITY.

21

22

CHAPTER 4
ANTIBODY STRUCTURE II
See APPENDIX: (4) AFFINITY CHROMATOGRAPHY; (5) RADIOIMMUNOASSAY
The IgG-like subunit of all human antibodies consists of two identical LIGHT
CHAINS and two identical HEAVY CHAINS, and proteolytic digestion can be used to
establish the different functions of distinct portions of antibody molecules. The nine
different CLASSES and subclasses (ISOTYPES) of human immunoglobulins exhibit
different physical and biological properties, determined by the amino acid sequence of the
CONSTANT REGION of the heavy chains. On the other hand, the VARIABLE
REGION of heavy and light chains together form the ANTIGEN-BINDING SITES of
antibodies.
The DOMAIN organization of antibodies, and an understanding of the
COVALENT and NON-COVALENT interactions involved in their structure and in their
interaction with antigen, form the basis for understanding their distinct roles in
PRIMARY and SECONDARY IMMUNE RESPONSES, and the key features of
IMMUNOLOGICAL MEMORY.

ANTIBODY HEAVY AND LIGHT CHAINS: PROTEOLYTIC FRAGMENTS

In order to examine the structure of antibodies, we will again start with a "typical" IgG
molecule, such as the rabbit antibody we discussed at in the last chapter. A schematic
diagram of an IgG molecule, together with its proteolytic fragments, is shown below:

PAPAIN
H

Fd

Fab

Fab

Fc

L
H
L
H

PEPSIN

Fd'

Intact IgG

(Fc degraded)
F(ab')2

Figure 4-1

23

 Intact IgG has a molecular weight of ~150,000.

Each IgG molecule consists of two identical heavy chains (mol.wt.~50,000) and two
identical light chains (mol.wt.~25,000). The two heavy chains are linked to each other
by one or more disulfide bonds (shown as lines in the diagram), and each light chain is
linked to one heavy chain by a single disulfide bond. Immunoglobulin heavy and light
chains, as we will see later, are encoded by evolutionarily related members of a large
multigene family.
This molecule can be broken down into smaller fragments by a wide variety of
enzymatic and chemical treatments. Of particular importance has been the use of two
proteolytic enzymes, papain, and pepsin. Treatment with papain yields two kinds of
fragments, a pair of F(ab) fragments, and one Fc fragment. Treatment with pepsin yields a
single fragment, the F(ab')2 (the Fc region is degraded by this enzyme).
If we begin with an antibody molecule of known antigen specificity, we can study the
fragments generated in this way not only with respect to their physical properties, but also for
their ability to interact with antigen and complement. Their properties are as follows:
Fab

size ~50,000 Da
binds antigen (ab = antigen-binding)
does not show precipitation or agglutination
will not fix complement or opsonize

F(ab)2

size ~100,000 Da
binds antigen
can show precipitation and agglutination
will not fix complement or opsonize

Fc

size ~50,000 Da
cannot bind antigen
is crystallizable
does not fix complement (except very poorly)

[Fd

that portion of the heavy chain which is included in the Fab

fragment]

Thus we see that proteolytic cleavage can separate IgG into fragments with different
functional properties. The F(ab) and F(ab')2 fragments are capable of binding antigen (they
contain the antigen binding sites), but only the F(ab')2 can exhibit agglutination or
precipitation; this is because these functions require bivalent binding for cross-linking. Each
F(ab) fragment has one antigen-combining site that requires the presence of both the heavy
chain and the light chain. In addition, we see that without the Fc piece, complement fixation
does not take place.
The Fc piece, on the other hand, does not bind antigen; and while it is required for
complement fixation, it cannot significantly bind complement on its own. The fact that it can
be made to form crystals was recognized very early as a striking property of the Fc (hence its
name). Antibodies in general do not form crystals, as a result of their characteristic
heterogeneity. The Fc fragment does not share the major source of this heterogeneity
(variation in the antigen combining site) since it does not contain a binding site.

24

IMMUNOGLOBULIN CLASS (ISOTYPE) IS DETERMINED BY

THE HEAVY CHAIN
So far we have discussed the properties only of a "typical" rabbit IgG molecule. If we
look at human immunoglobulins, we see that a variety of different forms exist, and this
variety is typical of mammals in general. Five major classes of immunoglobulin are known,
IgG, IgM, IgA, IgD and IgE; in addition, there are four subclasses of IgG, namely IgG1,
IgG2, IgG3 and IgG4, and two subclasses of IgA, namely IgA1 and IgA2, yielding a total of
nine different classes and subclasses (="isotypes"; see Chapter 6) of human immunoglobulin.

Class

PROPERTIES OF HUMAN IMMUNOGLOBULINS

serum
subconc. sedim.
# binding H-chain
classes
mg/ml coeff.
size
sites
class

Biological
properties

IgG

12

7S

150 KDa

(1, 2,

3, 4)

IgM

19S

900 KDa

10

IgA

7,9,11S

(160 KDa)n

2,4,6

(
1, 2)

IgD

0.03

7S

180 KDa

B-cell surface Ig

IgE

0.0003

8S

200 KDa

"reaginic" Ig

C-fixing,
placental X-fer

C-fixing,
B-cell surface Ig
secretory Ig

The class (and subclass) of an immunoglobulin is determined by the kind of heavy chain it
bears. Nine different heavy chains define the nine classes and subclasses (isotypes) of
human Ig.
All immunoglobulins share the same pool of light chains; there are two types of light
chains, kappa and lambda. Thus, an IgG1 molecule has two gamma-1 heavy chains,
and may have. Likewise for an IgM molecule which has mu heavy chains, and may
have either kappa or lambda light chains, but not both...etc.
All immunoglobulins share the same basic "IgG-like" structure consisting of two linked
heavy chains and two light chains. IgG, IgD and IgE have precisely this structure, and
differences in their molecular masses (seen in the table above) are due to differences in
the size of the heavy chain polypeptides and their carbohydrate content.
We can represent this basic structure by the general empirical formula, H2L2. Thus, a
particular IgG1 molecule may be represented as either 12 2, or 12 2 , depending on
whether it contains kappa or lambda chains.
IgM and some IgA consist of multiple IgG-like subunits. IgM consists of a pentamer of
five such subunits, each of which consists of two mu chains and two light chains (either
kappa or lambda); it therefore contains ten antigen combining sites, and is the largest of
the immunoglobulins. IgA can exist either as a monomer or dimer (rarely a trimer) of
25

the basic H2L2 subunit, and therefore can bear two, four or six combining sites. IgM
and polymeric IgA have an additional polypeptide associated with them, the J-chain
(for "joining"). The J-chain is thought to participate in the process of polymerization of
these molecules. (See also Chapter 9, Ig BIOSYNTHESIS)
DIFFERENT Ig ISOTYPES HAVE DIFFERENT BIOLOGICAL FUNCTIONS
The structures of the different immunoglobulin heavy chains are quite different from
one another, in amino acid sequence, in chain length and in carbohydrate content. These
differences result in marked differences in biological properties of the different isotypes
(classes and subclasses) of Ig.
IgM Most efficient Ig at complement fixation.
Produced early in immune responses.
Serum IgM bears an additional polypeptide, the J-piece (for joining), as does
polymeric IgA.
Serve as membrane receptors of B-cells (as IgMs, an IgG-like monomer.
IgG

Most abundant serum Ig.

Various subclasses can all fix complement (via classical or alternate pathway).
Can be transferred across the placenta into the fetal circulation.
Produced late in immune responses.
Opsonizing activity via Fc receptors present on macrophages and PMNs.

IgA

While IgA is the second-most abundant serum Ig, it is most characteristically a

secretory immunoglobulin, and is the most abundant Ig in exocrine secretions
(saliva, bile, mucus of the gut and respiratory tract, milk, etc.).
IgA in serum is mostly an 2L2 monomer; IgA in secretions is polymeric (mostly
dimer, i.e. [2L2]2, with some trimer), includes an S-piece (for secretory),
and like IgM also bears the J-piece.

IgD

Antigen-binding receptor on B-cells (together with IgMs).

Rare in serum; no other known biological function.

IgE

Extremely low levels in serum.

Reaginic" antibody, responsible for allergic reactions following binding to
surface of tissue mast cells (see Chapter 21).
While antibodies are generally very stable, IgE is an exception and is
characteristically heat labile.

HOMOGENEOUS IMMUNOGLOBULINS: MYELOMA PROTEINS

We have already noted that heterogeneity is one of the most striking features of
antibodies; this is understandable in that the primary function of antibodies is to bind to a
wide variety of eiptopes. This heterogeneity, however, makes it difficult to carry out
conventional structural studies on these molecules, particularly by amino acid sequencing or
crystallization, which both require a homogeneous protein.
Our understanding of immunoglobulin structure resulted largely from the existence of
MYELOMA PROTEINS, homogeneous forms of immunoglobulins produced by tumors of
antibody-forming plasma cells. Since the tumors are monoclonal in origin, so are their
protein products; in any particular patient, the monoclonal protein (also referred to as "Mcomponent") is of a single heavy chain class and bears a single kind of combining site. One
26

reflection of their homogeneity is the fact that myeloma proteins are often crystallizable,
unlike conventional Ig. In some cases the homogeneous protein consists of free light chain
without an associated heavy chain; these free light chains are rapidly cleared by the kidneys
and appear in the urine as BENCE-JONES PROTEINS. In other cases a tumor might
produce only heavy chains which accumulate in serum, with no associated light chain.
A variety of pathological conditions can result in the presence of monoclonal Ig in
serum, collectively referred to as Monoclonal Gammopathies. While some of these
conditions are benign, many represent malignant tumors (cancers) of antibody-forming cells.
Depending on their clinical presentation, these malignancies may be termed Multiple
Myeloma, Plasmacytoma, Immunocytoma, Waldenstrom's Macroglobulinemia, Heavy Chain
Disease or Light Chain Disease. We will use the general term myeloma protein to refer to
any such monoclonal immunoglobulin.
IMMUNOGLOBULIN HEAVY AND LIGHT CHAINS
HAVE VARIABLE AND CONSTANT REGIONS
Amino acid sequencing studies of Bence-Jones Proteins first showed that light chains consist
of a variable region and a constant region. If several kappa-type B-J proteins (each from a
different patient), are subjected to amino acid sequencing, one finds that the amino terminal
half of the polypeptide (about 110 amino acids) shows substantial variation from one protein
to another, while the carboxy-terminal half is essentially constant (except for allelic variants
discussed in Chapter 6). Similarly, the heavy chains of different myeloma proteins have an
amino-terminal variable region (also about 110 amino acids long), while the remainder of the
polypeptide remains constant within a given isotype.
Constant Region

Variable Region

L-chain
H-chain
NH2 -terminus

COOH-terminus

CDR's

Figure 4-2
The antigen combining site of an antibody is made up of the variable regions of one light
chain and one heavy chain. As can be seen in the diagram above, the two variable regions
(VH and VL) are located in the F(ab) regions, precisely where we would expect them to be if
they are responsible for antigen binding.
Within the variable regions, typically comprising 105-110 amino acids, some positions show
more sequence variation than others. The highest degree of variability between different
monoclonal proteins exists in three small sub-regions within the entire V-region, which were
named the Hypervariable Regions; the four segments outside these hypervariable regions are
termed "framework" regions, and show considerably less diversity (although they are still
variable). The hypervariable regions form the antibody combining site in the native three27

dimensional structure of the antibody, and are now commonly known by the more descriptive
name of COMPLEMENTARITY-DETERMINING REGIONS (CDRs)
THREE FAMILIES OF V-REGIONS EXIST, one for kappa light chains, one for lambda
light chains, and the third for heavy chains. In spite of their variable nature, one can readily
determine from its amino acid sequence (even from the first few N-terminal residues) to
which of these three families a given V-region belongs, kappa, lambda or heavy chain.
However, in the case of VH regions, one cannot predict which of the heavy chain isotypes it
will be associated with; all heavy chains, regardless of class or subclass, draw from a single
pool of VH-regions. We will examine the cellular and genetic basis for this phenomenon in
later chapters.
IMMUNOGLOBULIN HEAVY AND LIGHT CHAINS CONSIST OF GLOBULAR
"DOMAINS "
Up to now we have diagrammed Ig molecules as consisting simply of linear polypeptides
connected to one another by disulfide bonds. However, the polypeptides making up Ig
molecules (and proteins in general) are not normally stretched out like pieces of spaghetti, but
are folded into complex three-dimensional structures. In the case of a light chain, the Vregion and C-region are each folded into separate, compact globular DOMAINS, the two of
which are connected to each other by a short stretch of extended polypeptide chain. The
heavy chain likewise has a single V-region domain, but has several C-region domains, a total
of three in the case of IgG. The diagram in Figure 4-3 incorporates this more sophisticated
view of immunoglobulin:
VH
VL

"hinge"
region

C HI
CL

C HII
C HIII

Figure 4-3
Every heavy or light chain is made up of one V-region domain and one or more C-region
domains. In the case of light chains these are named VL and CL (V and C for kappa
chains, etc.). In the case of heavy chains, these are termed VH for the variable domain,
and CHI, CHII etc. for the constant domains; for a particular heavy chain, mu for
instance, the constant domains become CI, CII, etc.
28

Light chains have a single constant domain, heavy chains have three (IgG, IgD, IgA) or
four (IgM, IgE) constant domains.
Each domain consists of a compact globular unit, and they are linked to one another by
short stretches of extended polypeptide chain.
Different domains share varying degrees of amino acid sequence similarity and a high
degree of similarity in their three dimensional structures. These patterns of similarity
indicates that the different domains, and the three different immunoglobulin gene
families (encoding the heavy, kappa and lambda chains), are all the product of a series
of gene duplications of a primordial gene resembling a single domain.
Heavy chains have a stretch of extended polypeptide chain which is not part of any domain,
between the CHI and CHII domains. This segment is known as the HINGE REGION,
and is the region containing the cysteine residues involved in the H-H and H-L
disulfide linkages.
COVALENT AND NON-COVALENT FORCES STABILIZE ANTIBODY STRUCTURE
Many kinds of molecular interactions contribute to the extraordinary stability of
antibody structure. We have already seen that the chain structure is held together by a series
of covalent disulfide linkages between the two heavy chains, and between each light chain
and one heavy chain.
In addition, powerful non-covalent interactions exist between various adjacent pairs
of domains, specifically between VL and VH, between CL and CHI, and between the two CHIII
domains; these are indicated by heavy lines between these regions in Figure 4-3. Thus, even
if all the disulfide bonds holding the chains together are broken, the overall structure of the
Ig molecule will generally be maintained.
Each domain forms an extremely stable globular structure, held together by powerful
non-covalent forces as well as a single internal disulfide bond. The compactness of these
structures confers a degree of resistance to proteolytic cleavage; it is for this reason that much
proteolytic cleavage of immunoglobulins (as for instance with papain) takes place in the
extended chains between domains.
ANTIBODIES BIND ANTIGENS BY POWERFUL NON-COVALENT FORCES
The basis of antibody-antigen binding is steric complementarity between the
combining site and the epitope or antigenic determinant ("lock and key" concept). The
combining site itself is made up by both VL and VH, most directly involving the hypervariable
regions or CDRs ("complementarity determining regions") of both.
A variety of non-covalent forces are involved in this binding:
1) Van der Waals interactions. These are very weak and extremely short-range forces
(they decrease with the seventh power of the distance), but they become important
when many such interactions over the large area of an interactive surface are added
together.
2) Hydrophobic interactions. Exclusion of water molecules between hydrophobic
surfaces can also be a major contributor to antigen-antibody binding.

29

3) Hydrogen bonds and ionic interactions can also provide considerable stabilization
of antibody-antigen binding, in those cases where the epitope is capable of
participating in such interactions.

Ab
Ag1

GOOD FIT

Ab
Ag2

Ab
Ag3

NO FIT

MODERATE FIT
(CROSS-REACTIVE,
LOWER AFFINITY)

Figure 4-4
The single most critical factor in the strength of binding between antibodies and their
antigens is the degree of complementarity ("goodness of fit") between the antibody
combining site and the epitope (Fig. 4-4, left). A very slight change in the shape of either one
can turn extremely strong binding into no binding at all (Fig. 4-4, center). Alternatively, a
different kind of shape variation might turn strong binding into weaker binding, without
eliminating it altogether; this is part of the basis of cross-reactivity between different but
related antigens (Fig. 4-4, right).
IMMUNOLOGICAL MEMORY: PRIMARY VERSUS SECONDARY RESPONSES
One of the hallmarks of immune responses is their ability to display memory; the
immune system in some way "remembers" what antigens it has seen previously, and responds
more effectively the second time around. This is the basis, for instance, of acquired resistance
to smallpox - having once recovered from the disease (or having been vaccinated), a persons
immune system responds more rapidly the next time it is exposed to the virus, and eliminates
it before the disease process can be initiated.
Having discussed the variety of known Ig isotypes and the concept of affinity, we can
now identify several of the key features of immunological memory , illustrated in Figure 4-5
below, which shows the results of immunizing an animal with a "typical" antigen. The graph
on the left represents the results of a primary immunization; plotted on the X-axis is the time
after immunization in weeks, while the Y-axis represents a measure of the amount of
antibody in the serum (which could be determined by any of the techniques we have
described, agglutination, precipitation, ELISA, etc.). In the graph on the right we see the
results following re-immunizing the same animal with the same antigen at some later date,
termed a secondary immunization. The differences between the two graphs represent the
features which define immunological memory.

30

Serum
Ab

IgG
IgM

1o Ag

IgM

IgG

weeks

2 Ag

weeks

PRIMARY AND SECONDARY HUMORAL RESPONSES

Figure 4-5
Outlined in the table below are the most important features which distinguish primary
and secondary humoral immune responses.

PRIMARY RESPONSE

SECONDARY RESPONSE

SLOW
LOW Ab LEVELS
SHORT-LIVED

FAST
(kinetics of
HIGH Ab LEVELS clonal selection)
LONG-LIVED
MAINLY IgG
HIGH AFFINITY

MAINLY IgM
LOW AFFINITY

(Ig structure)

The first three features are a consequence of the cellular events in immune responses,
while the last two relate to the physical properties of the antibodies produced.
1) Speed. In a typical primary response we see a lag of about 4 days, followed by a slow
rise of antibody to a peak at about two weeks. In the secondary, the lag is only two
days, and antibody levels rise very rapidly to a peak within 4-6 days.
2) Antibody levels. The secondary response reaches peak antibody levels which are very
much higher than in the primary.
3) Duration. Primary responses not only appear more slowly, they disappear fairly
rapidly. Antibody titers in secondary responses remain at high levels for longer periods
of time, and in humans may persist for many years.
4) Ig class. The majority of the antibody in a primary response is IgM, with some IgG
appearing later in the response; in a secondary, IgG is the predominant antibody
throughout the response. IgM also appears during a secondary response, but at levels
31

and with a time course comparable to the primary and is therefore swamped out by the
higher levels of IgG. IgM does not exhibit immunological memory: it does not appear
more rapidly or at higher levels in secondary responses, and does not undergo
significant affinity maturation.
5) Affinity. The average affinity of antibody in a secondary response is higher than in a
primary response. Even during the course of a particular secondary response, it can be
shown that the affinity of "late" antibody is higher than that of "early" antibody. This
progressive increase in the average affinity of antibody is known as AFFINITY
MATURATION.
The net result of immunological memory is an immune response of greatly increased
effectiveness. The antibody is made more rapidly, it is made in higher amounts, it lasts
longer, and it is made with higher affinity so that it binds more effectively to its target. The
mechanisms involved in all of these phenomena will become clearer when we have discussed
the CLONAL SELECTION THEORY (see Chapter 7).
NOTE: It is important to keep in mind that immune responses vary greatly in their
time course, intensity and persistence. While the figure above illustrates the antibody
responses to a "typical" antigen in a "typical" organism, any particular response may be very
different in many of its features. One antigen may result in a much slower or more rapid
response than another, or may be unable to produce a secondary response at all.
Nevertheless, the fundamental differences between primary and secondary responses are
important and instructive generalizations.

CHAPTER 4, STUDY QUESTIONS:

1.

Describe the structure and biological properties of the pepsin- and papain-generated
PROTEOLYTIC FRAGMENTS of antibodies.

2.

What are the distinguishing structural and biological features of each of the five
CLASSES of human Ig?

3.

What are the common structural features of all antibodies?

4.

Why did the discovery of VARIABLE and CONSTANT regions of immunoglobulin

heavy and light chains pose such a serious puzzle for molecular biologists?

5.

How do the globular DOMAINS of immunoglobulins contribute to the remarkable

stability of antibodies?

6.

What features of SECONDARY antibody responses make them more effective than
PRIMARY responses?

32

CHAPTER 5
COMPLEMENT
See APPENDIX (8) COMPLEMENT FIXATION ASSAY
The complex of serum proteins known as COMPLEMENT plays key roles in the
lytic and inflammatory properties of antibodies. The CLASSICAL pathway is initiated
by antigen-antibody complexes (via complement components C1, C4, and C2), while the
activation of the ALTERNATE pathway (via components B, D and P), and the
MBLECTIN ("mannan-binding lectin") pathway may be initiated by other substances
independently of adaptive immune responses; all three pathways share those complement
components involved in the inflammatory and lytic consequences, namely C3, C5, C6,
C7, C8 and C9. The INFLAMMATION which is a consequence of complement fixation
is illustrated by the manifestations of SERUM SICKNESS, and complement is also seen
to be central to the normal process of clearing immune complexes, which is important in
preventing IMMUNE COMPLEX DISEASE.
In the late 19th century, a researcher named Jules Bordet, following the earlier results of
Richard Pfeiffer, was investigating the lysis of the bacterium Cholera vibrio (the agent which
causes cholera) by immune sera, and found that the ability of an immune serum to lyse its
targets was lost upon heating (e.g., at 56 C for 30 min). This ability to cause lysis was also
lost after simple storage of the serum for a few days at room temperature.
Bordet showed that such heating did not destroy the antibodies, however, since the addition
of a small amount of normal, non-immune serum, to the heat-inactivated antiserum fully
restored its capacity to specifically lyse cholera targets. Thus, the ability of immune serum to
lyse bacteria depends not only on antibodies specific for C. vibrio, but also on a non-specific
heat-labile substance found in normal serum.
This substance became known as
COMPLEMENT, since it "complements" the activity of the antibodies which are still present
in heat-inactivated antisera.
COMPLEMENT - A group of serum proteins which can be activated
(= "fixed") by antigen-antibody complexes or other substances, which may
result in lysis of a microbial target, or a variety of other biological effects
important in both innate and adaptive immunity. (The majority of these
proteins are produced by the liver.)
Before going into the details of the components of the complement cascade and their
activation, lets preview the various biological effects which can be attributed to the action of
complement, and identify those complement components or complexes which are responsible
for these effects.

33

BIOLOGICAL EFFECTS OF COMPLEMENT

A)

Cytolysis
[C5b6789] (Note: the bar identifies an activated complex)
Destruction of target cells by lysis of the cell membrane. This is termed cytotoxicity in
the case of nucleated cells, hemolysis for red blood cells, or bacteriolysis in the case of
bacteria. (NOTE: Not all bacterial and eukaryotic cells are susceptible to
complement- dependent lysis).

B)

Anaphylotoxin activity (= "vasoactive" or "phlogistic")

[C3a, C5a]
Stimulation of mast cells to release histamine and other substances, resulting in
increased capillary permeability and local accumulation of fluid in the tissue.

C)

Chemotaxis
[C5a, C5b67]
Attraction of polymorphonuclear neutrophils (PMN's) to a local site of inflammation.

D)

Opsonization (= "immune adherence")

[C3b]
Facilitation of phagocytosis by macrophages or PMN's via cell-surface receptors
specific for complement components ("complement receptors", or "CRs")

E)

Tissue damage
[C5b6789; PMN's]
Both the lytic complex and the inflammatory PMN's can cause considerable damage to
normal tissues, for instance in an Arthus Reaction or in Immune Complex Disease.

The consequences of complement activation can be categorized into two general classes:

Facilitating antibody function; destruction and removal of foreign material.

-Target cell lysis; the lytic membrane attack complex (MAC) can be produce by all
three of the complement pathways discussed below.
-Removal of immune complexes ("immune clearance"); this is a critically important
function facilitated by the presence of receptors ("CR") for various complement
components on the surface of leukocytes and erythrocytes. The special process by
which soluble (i.e. small) immune complexes are normally cleared from serum relies
on the presence on erythrocytes of CR1 complement receptors which are capable of
binding C4b (this process is discussed later in this chapter).

Development of inflammation; increased circulation and accumulation of fluids and

cells all contribute to the cardinal signs of inflammation (heat, redness, swelling and
pain); these functions are mediated directly and indirectly by proteolytic products of the
complement cascade (C3a, C5a).

THREE PATHWAYS FOR COMPLEMENT FIXATION

The process of complement fixation requires specific protein/protein interactions, it
involves proteolytic cleavages and conformational changes of proteins, and new biological
activities are generated as a result.
Three distinct (although related) mechanisms are known which can initiate the complement
cascade, the Classical Pathway, the Alternate Pathway, and the more recently recognized
MBLECTIN Pathway. The central event in all three of these modes of complement
activation is the cleavage of component C3. The pathways differ only in the mechanism by
which they achieve this cleavage, and we will consider them in turn.

34

CLASSICAL PATHWAY: ANTIBODY-DEPENDENT COMPLEMENT FIXATION

This pathway is initiated by antigen/antibody complexes and requires heat-sensitive
complement components. An outline of the components and events in complement fixation
by the classical pathway is shown in Figure 5-1. Lets examine in some detail the reactions
involved.
1)

Ag + Ab

(S + A

AgAb

SA)

An antibody/antigen complex is formed.

Conventional designations for Ag and Ab are S (for antigenic Site) and A (for Antibody).
While many different forms of antigen can fix complement and cause its various biological
consequences, the concept of lysis is only meaningful for membrane-bound antigens;
therefore, in the discussion below, we will assume we are dealing with an antigen in a
membrane susceptible to lysis (as, for example, an erythrocyte or a gram-negative bacterial
cell).
Ca++

2) a) C1q, C1r, C1s

complex.)
b) SA + C1qrs

C1qrs

(Spontaneous; occurs in serum without AgAb

SA C1qrs

Requires either a single IgM or 2 adjacent IgG's.

The antigen-antibody complex binds C1qrs and activates the esterase activity
associated with C1s.
(For simplicity in the discussion below, we will leave out the SA (or AgAb), understanding
however that it is required for the initiation of this pathway and is associated with the early
complexes. The bar over the C1qrs is a conventional way of indicating an activated complex.)
3) C1qrs + C4

C1qrs/4b

new active complex

C4a

(minimally vasoactive)

C1s (in the complex) binds soluble C4 and cleaves it into C4a and C4b; C4b becomes
covalently bound to the complex or to the nearby membrane (where it has minimal
opsonizing activity); C4a is released, and while it is somewhat vasoactive, its low level
of activity is probably not biologically important.
Some amplification of the process occurs here, since dozens of C4b molecules can be
generated and bound to the membrane.

35

CLASSICAL COMPLEMENT PATHWAY

C1q,C1r,C1s
++

Ca

C1qrs
Ag/Ab
C1qrs *
Amplification

C4
C4a

C1qrs/4b
++

Mg

C2
C2a

C1qrs/4b2b
C3
C3a

Amplification

near

Vasoactive

far

C3b

C4b2b3b

Opsonizing

C5
Vasoactive,
Chemotactic

C5a
Amplification

C4b2b3b5b
C6, C7
C4b2b3b5b67
C5b67 (soluble)

Chemotactic

C5b67
(membrane)
C8
C5b678

osmotic "holes"
C9

C5b6789

large holes, LYSIS

"MAC": Membrane Attack Complex

Figure 5-1

36

Mg++

4) C1qrs/4b

C2

C1qrs/4b2b +
new active
complex

C2a
no known activity

The esterase activity of C1s acts again, this time to cleave C2 into C2b (which is bound
by 4b in the complex) and C2a (which is released). (NOTE: The names of 2a and 2b
are reversed from the usage you may find in older literature as well as some current
sources. Using the present nomenclature, all those fragments which bind to the
complex are named "b", while those that are released in soluble form are named "a".)
____
At this point the presence of the AgAbC1qrs complex is no longer necessary for the
cascade to continue (even though it's still present), and we'll omit it in the diagrams
below.
____
______
5) C4b2b
+
C3

C4b2b3b
+
C3a
"C3 convertase"
opsonizing (3b)
vasoactive
The C4b2b complex has an enzymatic activity called C3 convertase, indicating it can
cleave C3 into C3a and C3b. This is a key step in this pathway, involving biological
amplification and the generation of molecules with new biological activities.
Many hundreds of C3 molecules can be cleaved by a single C4b2b complex,
resulting in considerable amplification. The resulting C3b molecules can either be
bound by the complex or can be released and bind elsewhere on the membrane (C3b
molecules which are not bound in one place or the other are rapidly destroyed.)
C3b has powerful opsonizing activity, whether it is bound in the complex or to some
other site on the membrane.
C3a is strongly vasoactive. It is also chemotactic for mast cells.
______
________
6) C4b2b3b
+
C5

C4b2b3b5b +
C5a
vasoactive
chemotactic for PMNs (and mast cells)
Cleavage of C5 results in a new complex and another biologically active molecule,
C5a, which is both vasoactive and chemotactic.
______
______
7) C4b2b3b5b +
C6, C7 C4b2b3b5b67
or...

C5b67 binds to nearby sites on the membrane,

in soluble form is chemotactic for PMNs

The C5b67 complex may be released and bind to a nearby site in the membrane, and in
soluble form also exhibits chemotactic activity. The resulting C4b2b3b complex left
behind can then sequentially bind additional C5, C6 and C7 molecules, which results in
another important site of amplification in the complement cascade.

37

The final reactions of the complement cascade produce the membrane-destroying

complexes:
8)

____
C5b67

C8

9)

_____
C5b678

C9

_____
C5b678
osmotic "holes", leakage
______
C5b6789
Membrane Attack Complex,MAC,
large "holes", lysis

The formation of large, open channels in target cell membranes results in the loss of
hemoglobin from erythrocytes (visible as HEMOLYSIS) or leakage of cytoplasmic
contents and death of nucleated cells (CYTOTOXICITY).
ALTERNATE PATHWAY OF COMPLEMENT FIXATION
The biochemist Louis Pillemer discovered in the 1950's that complement fixation could be
triggered by the yeast polysaccharide Zymosan in the absence of antibody, by a process
which has become known as the Alternate Pathway. The initial steps of this process are
quite different from those of the classical pathway and involve several unique serum
components, namely factor P (for "properdin"), and factors named B and D. The mechanism
is outlined below, and the initial step relies on the fact that very small amounts of soluble C3b
are normally present in serum, due to low levels of spontaneous C3 cleavage (tickover),
which may not be C4-dependent.

++

B + C3b

D + Mg

Ba

BbC3b

PBbC3b

very
unstable

"C3 convertase"
unstable

can be
protected and
rendered stable by
Zymosan or other
"activating surfaces"
on microbes

decays

C3
C3a
PBb(C3b) 2
can fix C5, C6 etc.,
and form lytic
complex

Figure 5-2

38

The initial steps of the alternate pathway result in the formation of an unstable complex
which has "C3 convertase" activity (namely PBbC3b). This complex is formed continuously
at low levels and rapidly degrades, but it may be bound and stabilized by "activating
surfaces" such as zymosan-containing yeast cell walls, LPS-containing gram-negative
bacteria, teichoic acid-containing gram-positive bacteria, and others. The result of
stabilization is that these C3-convertase complexes can carry out the remainder of the
complement fixation cascade in a manner identical to what we have outlined above for the
classical pathway.
Thus, fixation of complement by the alternate pathway can yield all the variety of
biological activities we have already seen -- chemotaxis, opsonization, anaphylotoxic activity
and lysis (only if the membrane is susceptible to lysis, of course), in addition to all the
inflammatory sequelae.
Initiation of the alternate pathway can be triggered by the presence of a variety of
microorganisms (as mentioned above) and parasites.
In addition, aggregated
immunoglobulin may also trigger the alternate pathway, even isotypes of Ig which are
incapable of fixing complement by the classical pathway (such as human IgG4 and IgA).

RELATIONSHIP BETWEEN CLASSICAL AND ALTERNATE PATHWAYS

Ag/Ab
complexes

CLASSICAL
C4b2b
PBbC3b
ALTERNATE
"Activating
surfaces"

Both are
"C3 convertases";
can cleave C3, then
fix C5, C6, etc.

Figure 5-3

Figure 5-3 illustrates the relationship between the two complement pathways which we
have just discussed. Both the classical and alternate pathways, using different mechanisms,
generate complexes which have C3 convertase activity. After that point the two pathways are
identical - each complex can generate and bind a new C3b molecule, then proceed to fix C5,
C6 etc. It is worth noting that while the initiation of the alternate pathway is dependent on
Mg++ (although this may not be absolute), the early steps of the classical pathway are
dependent on both Mg++ and Ca++ (see steps 2 and 4 in the classical pathway diagram).
MBLECTIN COMPLEMENT PATHWAY
A third mechanism for the initiation of complement fixation has been described which
depends on the presence of another normal serum protein known as the mannan-binding
lectin, or MBLECTIN. This protein is capable of binding to microbial carbohydrates
containing terminal mannose residues, and consequently binding two other proteins, MASP-1
and MASP-2 (mannan-binding lectin-associated serum protease-1 and -2). The resulting
complex has C4-convertase activity (i.e. it can bind and cleave C4), and the remainder of the
39

complement cascade (C2, C3, C5 etc.) is activated just as in the case of the classical and
alternate pathways. This distinct complement pathway is diagrammed in Figure 5-4.

MBLECTIN
+ mannan

MBLECTIN
mannan
MASP-1/MASP-2

"C4-convertase"
(can cleave C4,
then fix C2, C3 etc.)

Figure 5-4
SIGNIFICANCE OF THE NON-CLASSICAL COMPLEMENT PATHWAYS:
RAPID, NONSPECIFIC ACTION
Since they do not require the presence of antibody, both the ALTERNATE and
MBLECTIN pathways can be initiated much more rapidly than the classical pathway upon
initial exposure to a microorganism. There is no need to wait for several days while antibody
formation is initiated, and it is thus a manifestation of "innate" immunity. On the other hand,
these pathways do not have the exquisite specificity conferred by antibodies, and they are
limited to recognizing only certain kinds of microbial structures for triggering (these are both
characteristic features of innate immunity). Nor do they exhibit any form of memory.
Nevertheless, the alternate pathway, at least, can augment the biological effectiveness of
specific antibodies (produced by adaptive immunity) in at least two ways. First, as
mentioned above, aggregated Ig's (for instance, clustered on the surface of a microorganism)
may trigger the alternate pathway, regardless of whether the classical pathway has also been
initiated. Second, the C3b molecules generated by the classical pathway can promote the
formation of the alternate pathway's "C3 convertase", PBbC3b, again resulting in increased
complement fixation.
BIOLOGICAL EFFECTIVENESS OF COMPLEMENT DOES NOT
DEPEND SOLELY ON LYSIS
While complement fixation is generally thought of as culminating in lysis, we have already
noted that only a limited variety of bacteria (mostly gram-negative organisms) are susceptible
to such lysis (and not all nucleated cells, for that matter). This does not mean, however, that
such cells are completely resistant to the consequences of complement fixation. A grampositive bacterium can still be effectively opsonized (and consequently phagocytosed and
destroyed) as a consequence of complement fixation by either pathway. In addition, the
inflammatory sequelae of complement fixation (increased blood flow and accumulation of
scavenger cells) all tend to increase the effectiveness of microbial destruction.

40

The final outcome following introduction of a pathogen into an organism will depend on
many factors, including its susceptibility to complement dependent lysis and opsonization
and its ability to trigger the alternate pathway of complement, as well as on the nature of the
adaptive immune response which it generates (depending on its degree of immunogenicity
and the isotype distribution of the resulting antibodies) and possible previous exposures of
the immune system to the same or similar pathogens (immunological memory). Overall, the
opsonizing and inflammatory effects of complement appear to be more significant than
lysis in providing protection against infectious organisms.
REGULATION OF THE COMPLEMENT CASCADE
We have thus far mentioned only those factors which activate complement components
and drive the reactions in the direction of biological effectiveness. If these were the only
relevant factors, then an initial complement fixation event would result in an uncontrolled
inflammatory response and the consumption of all the complement available in the system;
clearly this does not normally occur.
A variety of factors exist which are responsible for the inactivation of the various
complement products which are biologically active. Some of these factors are known,
including ones which inactivate the C1 complex, the C3a, C3b, C4a and C5a fragments, and
the C5b67 complex. These inhibitory factors are critical to the normal balance of the
complement system. For example, a hereditary deficiency in C3b-inactivator (called
"Factor I"), results in excessive breakdown of C3 by the BbC3b complex and greatly depletes
the serum levels of C3, which in turn results in high susceptibility to recurrent bacterial
infections. A great deal is known about mechanisms which regulate complement activation,
although we will not discuss them further here.
IMMUNE COMPLEXES: COMPLEMENT-MEDIATED TISSUE DAMAGE
One of the consequences of antibody binding to its antigen is the elimination of the Ag/Ab
complex by phagocytic cells; removal of foreign antigens is obviously one of the main goals
of the immune response. If antigen/antibody complexes are formed in the circulation,
however, they may have other consequences which can be quite harmful to the host.
Antigen/antibody complexes in the bloodstream can bind to the basement membranes of
blood vessels and kidney glomeruli; at these sites they can fix complement which results in
damage to tissues. This is the basis for IMMUNE COMPLEX DISEASE. The antigen may
be associated with a pathogen (a virus, for instance), it may be clinically introduced (with
blood transfusions or with drug or antibody therapy), or may be a normal "self" component
(in the case of AUTOIMMUNE DISEASES such as RHEUMATOID ARTHRITIS and
LUPUS ERYTHEMATOSUS).
Immune complexes are harmful only if they fall within a limited range of sizes. If the
complexes are very large, they are efficiently removed by phagocytic cells (e.g. liver Kupfer
cells and lung macrophages); if they are very small, they are effectively soluble and are not
trapped in basement membranes (and these small complexes are normally removed by
erythrocytes, discussed below). Immune complexes which fall between these two extremes,
however, may become trapped in the basement membranes of the vasculature in various
organs where they can fix complement and cause tissue damage.

41

The consequences of deposition of complexes within tissues include local

inflammation and tissue necrosis. These are the consequences of the activity of the various
complement components which we have already outlined, and there can be more systemic
symptoms such as fever and malaise.
Let's illustrate this with a classic example known as One-Shot Serum Sickness. A man
having suffered a severe tetanus-prone wound is given a single injection of horse serum
containing anti-tetanus toxoid. After about a week, he complains of fever, malaise and
rashes, which spontaneously subside after another week; and he remains well after that. To
understand the progress of this reaction, we will refer to Figure 5-5 below:

free
horse
proteins
a

serum
levels

immune
complexes

b
free
anti-horse
antibodies
0

weeks 1

ONE-SHOT SERUM SICKNESS

Figure 5-5
During the first week following the injection, the horse proteins are removed from the
circulation at a slow rate determined by their biological properties; this is referred to as
Biological Clearance ("a"). At the beginning of the second week, they begin disappearing at
a more rapid rate; this is due to the production of anti-horse antibodies and a resulting
Immune Clearance of the complexes ("b"). Later on (in the third week) we can find
appreciable levels of free circulating anti-horse antibodies, but only after the horse proteins
themselves are gone.
During the period of immune clearance, at a time when both horse protein and anti-horse
antibodies are present, complexes may be formed which are of the appropriate size to be
trapped in basement membranes; and the results include the rash (local necrosis) as well as
the fever and other systemic manifestations. In this example the disease is self-limiting, since
there is no continuing source of antigen -- when the horse serum proteins are cleared, the
symptoms disappear. However, if there is a continuing source of antigen, immune complex
disease may present a more serious chronic problem; for example, the antigen may be
associated with a chronic viral infection (e.g. hepatitis), or may be a normal tissue component
(in the case of AUTOIMMUNITY).

42

This illustrates the danger of any therapy which makes use of foreign proteins, including
anti-toxin therapy, and blood transfusions in general. This example is called "one shot"
because it involves a single primary exposure to the antigen, and the disease symptoms are
therefore delayed a week or two while induction of antibody formation takes place; if
circulating antibody is already present at the time of injection, the disease will appear much
more rapidly. It can be particularly dangerous to treat someone intravenously with a protein
to which he has already been exposed, (with a second course of horse serum anti-toxin, for
instance) due to the presence of pre-existing antibody and the induction of a rapid and
powerful secondary antibody response.
C1 & C4 COMPLEMENT DEFICIENCIES:
INADEQUATE CLEARANCE OF IMMUNE COMPLEXES
We can learn much by examining the clinical consequences of congenital complement
deficiencies. Deficiencies of components of the membrane attack complex (C5-9), as well as
the alternate pathway components B, D and P have the expected results of increasing
susceptibility to bacterial infections. More surprising, however, is the fact that deficiencies
of C1 and C4 result in increased susceptibility to the development of autoimmune disease
(SLE, or Systemic Lupus Erythematosus, which will be discussed later), and associated
immune complex disease.
This reflects the fact that C1 and C4, unlike C3 and the later components of the classical
pathway, can be efficiently fixed by soluble Ag/Ab complexes, and these components are
therefore critically important in causing the rapid removal of such complexes from the
circulation. This process depends largely on the presence of membrane receptors for C4b on
erythrocytes (which express the complement receptor CR1). Soluble immune complexes
which contain C4b therefore bind efficiently to RBCs, which carry them to the liver where
the complexes are stripped off by macrophages, and the RBCs are returned unharmed to the
circulation. When these early complement components are absent, small immune complexes
which are normally harmless (because they are rapidly removed) may grow in size and
eventually be deposited in tissues, and the alternate pathway components can then promote
the inflammatory consequences we have already discussed. The resulting tissue damage can
feed a continuing cycle of formation of further immune complexes and induction of
autoimmune antibodies (see Chapters 18 and 19 on TOLERANCE and AUTOIMMUNITY).
The importance of this mechanism for clearing immune complexes results from the fact
that normal serum always contains small amounts of immune complexes, mostly due to the
spontaneous formation of autoantibodies to various self components. In the presence of the
normal complement-dependent clearing mechanism, these autoantibodies and immune
complexes are harmless. When this mechanism is defective, these immune complexes may
grow and fuel a continuing cycle of tissue damage and autoimmunity (see Chapter 19).

43

CHAPTER 5, STUDY QUESTIONS:

1.

What is required for the initiation of complement fixation by each of the

CLASSICAL, ALTERNATE and MBLECTIN pathways? What do the three
pathways have in common?

2.

Which proteolytic fragments of complement components have known autonomous

biological activities?

3.

What are the clinical consequences of the congenital absence of various complement
components?

4.

What medical situations are likely to result in ONE-SHOT SERUM SICKNESS?

5.

What is the process by which soluble immune complexes are normally cleared from
the blood?

44

CHAPTER 6
ANTIBODY GENETICS: ISOTYPES, ALLOTYPES, IDIOTYPES
See APPENDIX: (3) OUCHTERLONY; (4) AFFINITY CHROMATOGRAPHY
Human immunoglobulins are made up of LIGHT and HEAVY chains encoded by a
total of 14 CONSTANT REGION GENES organized into three gene families, namely
ONE KAPPA gene, FOUR LAMBDA genes and NINE HEAVY CHAIN genes. Allelic
variants exist for the kappa locus and six of the nine heavy chain loci.
The nine heavy chains define nine human CLASSES and SUBCLASSES of human
immunoglobulin. Together with the five different light chains, these genes represent 14
distinct ISOTYPES which are present in all normal human sera. ALLOTYPES, or allelic
variants within the constant regions, are known to exist for some of these isotypes, and
are inherited in a Mendelian co-dominant fashion. IDIOTYPES consist of the unique
combination of VH and VL which characterize a particular immunoglobulin's combining
sites.

HUMAN IMMUNOGLOBULINS
L-chains
H-chains
kappa

isotypes:
allotype
variants:

Km

lambda

1 2 3 4

--

--

-- G1m G2m G3m G4m Em --

--

--

--

2
A2m

Figure 6-1

Immunoglobulins all have light and heavy chains, but the light and heavy chains of a
particular Ig molecule may be different from those of other molecules. We have already seen
that different kinds of heavy chains define the class (and subclass) of antibody to which they
belong. We have also learned of the two varieties of light chains that exist, kappa and
lambda. These and other structural differences between Ig polypeptides can be organized
into three categories of differences, which we call isotypic, allotypic, and idiotypic. These
are commonly distinguished serologically, i.e., by the use of antibodies which recognize
different specificities on target immunoglobulins.

45

ISOTYPES -- Distinct forms of light or heavy chains which are present in all members of
a species, encoded at distinct genetic loci. Kappa and lambda are isotypes of light chains.
Mu (), delta (), gamma-1 (1), etc. are isotypes of heavy chains. All isotypes can be
readily found in all normal sera.
ALLOTYPES -- Genetic variants within the C-region sequences of particular isotypes that
are inherited in an allelic manner ("allelic type"). Different members of a species will
therefore differ from one another with respect to which particular alleles of a given isotype
they received from their parents. Km1 and Km2 are allotypes of humans kappa chains;
G1m(4) and G1m(17) are allotypes of human gamma-1 chains. The presence of particular
allotypes, like isotypes, can be readily detected in those normal sera in which they are
present.
IDIOTYPE -- An antigenic specificity (epitope) which distinguishes a particular
combination of VH and VL (the antigen recognition site) from all others. Thus, a
particular monoclonal immunoglobulin (a myeloma protein, for example) will bear an
idiotype different from any other. Unlike isotypes or allotypes, particular idiotypes can
generally be detected (with very rare exceptions) only in sera from myeloma patients.
This is because any particular idiotype will be represented only at extremely low levels
among the many thousands of kinds of combining sites present in serum immunoglobulin,
even in specific immune responses.
Lets go through a series of experimental analyses of human myeloma proteins and BenceJones proteins. We will use classical serological approaches to demonstrate:
(1) The existence of two light chain isotypes ( and );
(2) The presence of two chain allotypes (Km1 and Km2;
(3) The existence of two chain isotypes (1 and 2).

EXPERIMENTAL DEFINITION OF LIGHT CHAIN ISOTYPES: KAPPA AND

LAMBDA
Well begin with a series of eight purified Bence-Jones proteins, each from a different patient
(labeled A through H), and carry out an antigenic analysis to illustrate the properties of
isotypes and allotypes. Let's take four of these proteins (A,B,C and E in this example) and
use each them to immunize rabbits, resulting in four antisera. We will then take each of the
four antisera (labelled Ab) and test each of them by double diffusion in agarose (see
APPENDIX 3, OUCHTERLONY analysis) against each of the eight original proteins. In the
table below, a plus indicates that the antibody does show a line of precipitation with the
particular B-J protein, and a minus that it does not.

46

Bence-Jones Proteins
Ab

Rab A

Rab B

Rab C

Rab E

Table 6-1
Clearly, while we have four different antisera, we see only two different patterns of reactivity,
since antisera made against A and B show the same pattern as each other, and those against C
and E are also the same. The results indicate that these Bence-Jones proteins can be
separated into two "kinds": one group (A, B, D, F and G) reacts only with one pair of
antibodies (A and B), the other group (C, E and H) reacts only with the other pair (C and
E). We could call these two kinds of B-J proteins types I and II, or types a and b, but they
have actually been named Kappa and Lambda after their discoverers, Korngold and Lipari.
Thus we find that all B-J proteins (and, in fact, all human immunoglobulin light chains) can
be classified as either kappa or lambda type. Now we ask whether these two varieties
represent isotypes or allotypes of light chains? To answer this question we must use our
antisera to test a large number of normal human sera; when we do this we find that every
normal serum contains both kappa chains and lambda chains. Therefore, these two
types of light chains fulfill the definition of isotypes. This implies that there are two
separate genetic loci which encode these two light chain types - they are not encoded by two
different alleles at a single locus.
Kappa and lambda light chains are both found in immunoglobulins of all mammals. In
humans about 60% of all light chains are of kappa type, and 40% are lambda. In mice and
rats, on the other hand, about 95% of light chains are kappa type, while the vast majority of
light chains of horses and cattle are lambda-type. The biological significance of these
differences is not known.
The various heavy chain classes and subclasses which have been listed in Chapter 4 are all
isotypes of human heavy chains. Unlike light chain isotypes, however, heavy chain isotypes
are known to differ from one another in a variety of important biological functions -- IgM is
the most efficient at complement fixation, IgA is most efficiently secreted into exocrine
fluids, etc. (see Chapter 4). [It should be noted that different vertebrate species have different
numbers (and kinds) of heavy chain isotypes. Rabbits, for example, have only a single kind
of IgG (i.e. they have no gamma-chain subclasses), while the major serum immunoglobulin
in chickens is termed IgY.]

47

EXPERIMENTAL DEFINITION OF KAPPA CHAIN ALLOTYPES (InV/Km)

Lets analyze this collection of B-J proteins further. We now take our Rabbit-anti-B (which
is one of our anti-kappa reagents) and test it again in Ouchterlony against four of the kappa
type chains, A, B, D and F. This time, however, we will put the four antigens in nearby wells
to examine their antigenic relatedness (as opposed to simply testing each one independently
for its ability to precipitate with the antibody). The results are shown below:

F
RB
Figure 6-2

While Rabbit anti-B reacts with all four antigens, it clearly detects more epitopes on two of
them, B and F, than on the other two. That is to say, in addition to detecting an epitope
common to all of them, the antibody is detecting at least one epitope present on B and F
which is not present on A and D.
In order to make the analysis clearer, we will absorb the antiserum, removing those
antibodies which react with all four proteins and leaving only those antibodies reacting
specifically with B and F. We do this by passing the antiserum over an "immunoadsorbent"
column which contains antigen A covalently coupled to a matrix (see AFFINITY
CHROMATOGRAPHY in APPENDIX 4). All those antibodies which can bind to B-J
protein A" will do so, all the others will pass through the column and will be recovered.
The recovered antiserum (which passes through the column), we call RB[abs A] . When we
test this absorbed antiserum in the same manner as before, we see the results below:
B

F
RB
[abs A]

Figure 6-3
This absorbed antiserum now reacts only with B-J proteins B and F, and not at all with A and
D. It therefore defines two different forms of kappa chains, which, when they were
discovered, were originally called Inv+ and Inv-.
48

Now we need to ask once again, are these variants isotypes or allotypes of kappa chains?
When we use this specific antiserum to test a large number of normal human sera, we find
that only some sera show a reaction. Furthermore, we find that the presence or absence of
this epitope (defined by Rab B[abs A]) can be accounted for by the segregation of a pair of
Mendelian alleles at a single locus (this information comes from family studies). This Inv
factor is therefore not an isotype, but fulfills the definition of an allotype.
Thus, every individual human has kappa chains (an "isotype"), but the kappa chains in any
individuals serum may exist as either or both of at least two allelic forms ("allotypes"); some
people have only Inv+ kappa chains, others have only Inv- kappa chains, and some (who are
heterozygotes) have both. Modern nomenclature replaces the name Inv with Km (for Kappamarker), and a total of three allelic epitopes are now known to be defined by the Km locus,
named Km(1), Km(2) and Km(3). Our anti-Inv antibody (Rab B[abs A]) represents an
anti-Km(1) antibody, and by a similar set of immunizations and absorptions we can produce
specific antisera detecting Km(2) and Km(3). Using these antibodies to study human
populations, we can review the relationship between genotype and phenotype for the three
alleles defined by this set of antisera, illustrated as follows:
Phenotype versus Genotype for Kappa Allotypes
Genotype

Phenotype

Km(1/1)
Km(1/2)
Km(1/3)
Km(2/2)
Km(2/3)
Km(3/3)

Km(1+)
Km(1+2+)
Km(1+3+)
Km(2+)
Km(2+3+)
Km(3+)

Since kappa chain alleles are codominantly expressed in all people (as is the case for all
immunoglobulin genes), an individual who is heterozygous for the Km(1) and Km(2) genes
will will always show a positive result when his serum is typed using either reagent. Thus,
there are three alleles at the single kappa locus, and a total of six unique genotypes and
phenotypes.
Many of the human heavy chains also show allelic variants. "Gm" is a general term for
allotypes of human gamma chains (meaning "Gamma marker"). "G1m", for example,
specifies allelic markers of gamma-1 heavy chains which exist in several allelic forms, two of
which are known as G1m(4) and G1m(17). Similarly, G3m specifies allelic markers of
gamma-3 heavy chains, etc. As another example, while all normal human sera contain both
IgA1 and IgA2 (which are therefore isotypes), the alpha-2 heavy chain present in IgA2 can
exist as either of two allelic variants, which are known as A2m(1) and A2m(2).
Immunoglobulin allotypes have been known since the 1950's, with the discovery of rabbit
allotypes by Oudin and of human allotypes by Grubb. Most of what we know of the
remarkable organization of immunoglobulin genes, culminating in their characterization in
the early 1980s by recombinant DNA techniques (which we will discuss in Chapter 8), was
based on studies of the classical genetics of immunoglobulin allotypes in humans, rabbits and
mice.

49

EXPERIMENTAL DEFINITION OF ISOTYPES OF HUMAN LAMBDA CHAINS (Oz)

Let's carry out for a third time the kind of analysis we just described for kappa chains, this
time with the lambda chain Bence-Jones proteins in the panel above. We will use Rabbit
anti-E (antiserum against one of the lambda chains) and test it against proteins C, E and H
(all lambda chains), with the following results:

H
RE
[absC]

RE

Figure 6-4

Figure 6-5

On the left is the pattern shown by the original antiserum, and on the right the pattern seen
after the antiserum has been absorbed with protein C, as we had done before with the antikappa antiserum. We see that not all the lambda proteins are the same, i.e., proteins E and H
have at least one epitope that protein C lacks. We have thus defined a variant of human
lambda chains which we call Oz; proteins E and H are Oz+ while protein C is Oz-.
Once again we ask if this represents an allelic variant of lambda chains, as Km is for kappa
chains. When we test a large panel of normal human sera, we find that every serum contains
Oz+ lambda chains. (Every serum also contains Oz- lambda chains, which we can determine
by further analysis). Therefore we are not dealing with an allotype, but a new isotypic
variant. While our earlier analysis had indicated the existence of at least two genetic loci for
human light chains, one for kappa and one for lambda, we now have to hypothesize at least
two distinct loci for lambda chains, one for each the two variants we have just defined (Oz+
and Oz-).
Since the original discovery of the Oz marker, two other isotypic serological markers have
been found for human lambda chains, namely Mcg and Kern. These three specificities are
found in four different combinations which define a total of four genetic loci encoding human
lambda chains, named lambda-1 through lambda-4. No allelic variants have yet been
described for any of these four lambda isotypes, however.

50

EXPERIMENTAL DEFINITION OF IDIOTYPES:

INDIVIDUALLY SPECIFIC EPITOPES
We will now examine the results of one additional set of analyses, this time following
immunization of a rabbit not with a purified light chain, but with an entire intact myeloma
protein. If we start (for example) with a protein which has a kappa light chain and a gamma1 heavy chain (i.e., a G1 protein) we can predict the presence of specific antibodies to the
four distinct varieties of known epitopes in the resultant antiserum:
anti-kappa. The rabbit should make antibodies which react with all kappa chains
(anti-isotype), just as it did when we immunized with purified kappa chain.
anti-G1. We also expect to find antibodies which will react with all G1 heavy chains,
defining the G1 isotype.
anti-Km. We might also expect to have anti-allotype antibodies. If the kappa chain
happened to be Km(1), for instance, the rabbit should produce anti-Km(1), equivalent
to the antibody we originally called "anti-Inv" in the example above.
anti-G1m. Similarly, we might expect the rabbit to recognize whatever G1m allotypic
markers are present on the heavy chain. If the heavy chain happened to be G1m(17),
for example, the rabbit could make antibodies to this epitope. (NOTE: While we need
to consider this possibility, anti-Gm antibodies are not easily produced by heterologous
immunizations, as in this example.)
Each of these categories of antibodies can be detected by Ouchterlony precipitation using an
appropriate variety of myeloma proteins as antigens. Each of these four represents a separate
and distinct population of antibodies, and each can be removed by appropriate absorption (as
we have described above). What happens when we absorb (remove) all antibody activity to
the four kinds of specificities listed above, then test the resultant absorbed antiserum against
the original G1 immunogen (as well as a variety of other myeloma proteins)? Our
prediction is that there should be no detectable remaining antibody, since we have removed
antibodies to all those epitopes we know about.
However, what we actually find is that the absorbed antiserum still reacts with the
original G1, although it does not react with any other myeloma protein (of any isotype or
allotype composition).
This antibody, therefore, defines an idiotype, or idiotypic
epitope/specificity. It is "individually specific", reacting with the original immunogen but not
with any other. Nor will it react with isolated light or heavy chains, even from the same
protein, but it requires the presence of the intact molecule. Such an antibody is binding to an
epitope created by the combination of a particular VH and VL domain, that is, the unique
antigen-combining site of the immunizing myeloma protein. This idiotype will in general not
be detectable in normal sera; even though the particular combination of VH and VL may be
present on some normal serum Ig molecules, they will only be a few molecules among many
thousands of different combining sites present in normal Ig and would be extremely difficult
to detect.

51

CHAPTER 6, STUDY QUESTIONS:

1. Define and distinguish antibody ISOTYPE, ALLOTYPE and IDIOTYPE.
2. Given a newly produced antiserum directed against a human Bence-Jones protein, how
would you determine if it recognized ISOTYPE or ALLOTYPE specificities?
3. How might you produce an anti-G3-specific antiserum?
antiserum?
4. How would you produce an anti-IDIOTYPE antiserum?

52

An anti-Km(2)-specific

CHAPTER 7
CELLULAR BASIS OF ANTIBODY DIVERSITY: CLONAL
SELECTION
The specificity of humoral immune responses relies on the huge DIVERSITY of
antigen combining sites present in antibodies, diversity which is generated in an antigenindependent fashion. The process by which antigen (during an immune response)
stimulates the clonal expansion and differentiation of antibody-forming cells of
predetermined specificity is known as CLONAL SELECTION.
The basic principles of clonal selection are illustrated in this chapter by two classic
experiments, and are applied to our understanding of some of the fundamental features of
the adaptive immune response, namely TOLERANCE, MEMORY and AFFINITY
MATURATION, as well as the presence of NATURAL ANTIBODIES.

A humoral immune response typically generates antibodies with a wide diversity of

combining sites; even a simple hapten will elicit a highly heterogeneous population of
antibodies. In order to provide a framework for understanding how the diversity of the
antibody response is generated, we can set out three facts regarding antibody specificity
which are based on experimental findings:
1) An animal can produce antibodies to many different epitopes. This has been
known since the early days of immunology, based simply on the variety of distinct
molecules that antibodies can distinguish. How many antibody combining sites exist in
nature? For a start lets say there are at least one million different combining sites
(although this is certainly a gross underestimate).
2) A single antibody-secreting cell produces only one kind of antibody. All molecules
of antibody produced by a single plasma cell are identical with respect to isotype,
allotype and idiotype specificities. The heterogeneity of antibody responses, therefore,
reflects the heterogeneity of antibody-forming cells, and not that of the products of each
individual cell. This explains the homogeneity of myeloma proteins, since they are the
products of clonally derived tumor cells. (NOTE: One important consequence of this
fact is allelic exclusion, which will be discussed later [see Chapter 9].)
3) The specificity of antibodies is determined by the primary structure (i.e., the
amino acid sequence) of the light and heavy chain variable regions. Antibody
diversity cannot be explained by different folding patterns of molecules with identical
amino acid sequences (as had erroneously been proposed in the 1940's).

Lets examine the problem of the generation of immune responses at the cellular level. How
does a particular antibody-forming cell know which of the million possible combining sites
it is supposed to produce? Two general kinds of schemes can be invoked to explain this,
namely those of instruction and selection. These two schemes differ in the assumptions they
make about the hypothetical precursors of the antibody-forming cells ("AFCP"s).

53

AFCP

AFC
D

AFCP

AFC

C
D

Ag-D

Ag-D
D

F
G

INSTRUCTION

SELECTION
Figure 7-1

Under INSTRUCTIONAL schemes, the precursor of an antibody-forming cell is not

precommitted, but has the potential of making any one of a million different antibodies. This
multipotential precursor cell, following exposure to antigen, somehow recognizes which of
the million possible antibodies is needed and proceeds to differentiate into an antibodyforming cell producing that particular antibody; proliferation also takes place, expanding the
number of antibody-forming cells over the number of precursors. Under such a scheme every
precursor cell can potentially respond to any antigen.
SELECTIONAL schemes, however, hypothesize that the precursors of antibody-forming
cells are themselves precommitted to producing antibody of a particular specificity. Only a
small fraction of the precursor cells is capable of responding to any particular antigen. The
consequences of antigen exposure are the same as described for the instructional scheme,
i.e. differentiation into mature antibody-forming cells, and multiple rounds of proliferation
which greatly expand the number of antibody-forming cells.
CLONAL SELECTION
One formulation of the selectional class of schemes is known as CLONAL SELECTION, and
was developed in the 1950's by Burnet based on work of Jerne and Talmadge. This
hypothesis has become the foundation of modern immunology, and states the following:
1) Each Antibody-Forming Cell Precursors (AFCP) is precommitted to making antibody
of a particular specificity, even before it ever encounters a target antigen.

54

2) Each AFCP bears membrane-bound immunoglobulin displaying the combining site it

is capable of making, in effect, a "sample of its wares".
3) This membrane Ig acts as a specific receptor molecule for the particular epitope to
which it is directed. Upon binding to an AFCP's surface Ig, the antigen stimulates that
cell to differentiate into an antibody-forming cell, and to proliferate (clonal
expansion).
Two central features of Burnet's Clonal Selection Hypothesis should be emphasized. First, it
makes the simplifying assumption that the only molecule that can specifically recognize an
epitope is antibody itself. By hypothesizing the presence of antibody pre-existing on the
surface of the AFCP, Burnet avoided the problem of how a non-committed precursor cell
"looks" at antigen to determine which of the many possible antibodies is needed. Second, it
states that the development of the immunological repertoire is antigen-independent. The
differentiation which results in the appearance of a particular AFCP must be a
developmentally controlled process, and must be able to take place in the absence of any
antigen stimulation.
EXPERIMENTAL CONFIRMATION OF CLONAL SELECTION
Let's examine two key experiments, carried out in the late 1960's, which provided
compelling evidence for the idea that AFCPs are precommitted to a particular antigen. In one
case specific AFCPs are killed as a result of binding a specific radioactive antigen, in the
other the AFCPs are killed as a result of their proliferative response to their particular
antigen.
Hot Antigen Suicide. Ada and Byrt (1969) reasoned that if AFCPs bear surface receptors
which can specifically bind antigen, then these cells should be susceptible to being killed if
presented with a toxic form of antigen. They used an ADOPTIVE TRANSFER system,
consisting of injecting normal spleen cells together with antigen into a lethally irradiated
recipient. Testing the recipient's spleen for the presence of antibody-forming cells showed a
high response at the end of seven days.

SPLEEN
CELLS
ANTIGEN
NORMAL SPLEEN DONOR
900R

TEST SPLEEN
FOR Ab-FORMING
CELLS AT DAY 7
IRRADIATED RECIPIENT

ADOPTIVE TRANSFER
Figure 7-2

55

They then took a sample of the antigen (Salmonella flagellin, or "Fla") and made it highly
radioactive by coupling it with Iodine-125. This radioactive antigen was incubated overnight
with the normal spleen cells before transfer into the irradiated recipient. The rationale is that
if precursor cells are precommitted to a particular antigen, and if they have cell surface
receptors specific for the antigen, then those cells precommitted to the antigen Fla, and only
those cells, should bind the radioactive antigen and thus commit "suicide" due to the
radioactivity.
The results they obtained are summarized below:

1)

Ag used for
preincubation
none

2)

Ag transferred
with cells
none

Resulting 7-day
Ab response
none

none

Fla

high

Fla

low

3)

125

I-Fla

4)

125

I-Fla

Fla + BGG

low anti-Fla
high anti-BGG

Spleen cells incubated with "hot" Fla can no longer transfer responsiveness to Fla, although
their response to an unrelated antigen is (BGG) unaffected.
These results show that the precursors of antibody-forming cells for the anti-Fla response can
be eliminated by incubating the spleen cells with highly radioactive Fla, but that this
treatment does not affect the precursors for the anti-BGG response. The precursors of antiFla antibody-forming cells are a different population from the anti-BGG precursors, and this
precommitment is reflected in their ability to specifically bind the antigen to their surface.
These are the two key elements of Burnet's Clonal Selection Theory.
3

H-TdR Suicide. Mishell and Dutton (1966) took a slightly different approach when they
developed a system in which they could generate an in vitro primary antibody response.
Under a carefully specified set of culture conditions, they incubated normal mouse spleen
cells together with antigen, and after five days they were able to demonstrate the presence of
large numbers of antibody-forming cells (PFC's, for Plaque-Forming Cells) in the culture
dish (line 1 in Figure 7-3).

They then used highly radioactive tritium-labeled thymidine (3H-TdR) to selectively kill
those cells undergoing proliferation. (Thymidine is a DNA precursor which is taken up and
incorporated only by those cells synthesizing DNA, i.e., proliferating cells; "resting" cells are
not affected by this treatment.) By treating with "hot" thymidine during different 24-hour
periods, they were able to show that the precursors of antibody forming cells were
proliferating (and therefore could be killed by 3H-TdR) between 24 and 48 hours after
initiation of the culture with the antigen SRBC (sheep red blood cells; see line 3); treating the
cells during the first 24 hours after initiation had no affect on the resulting response (seen in
line 2).

56

Ab RESPONSE
at DAY 5 (PFCs)

1)

days of culture
SRBC

2)

SRBC

3)

0
SRBC

4)

SRBC

SRBC
BRBC

SRBC
+BRBC

= Period of 3 H-TdR treatment

3 H-TdR SUICIDE OF ANTIGEN-SPECIFIC AFC PRECURSORS

Figure 7-3
The key part of the experiment is in line 4. A 3H-TdR pulse between 24 and 48 hours should
eliminate the response to the original SRBC challenge (as in line 3). However, if AFCPs are
not precommitted, then it should not affect the response to a second dose of SRBC given at
24 hr (see line 2), nor should it effect the response to a different antigen (BRBC, burro red
blood cell) given at the same time. But what they found was that while the response to
BRBC was high, as expected, the response to the second dose of SRBC was very low.
These results show that the population of precursor cells which proliferate in a specific
response to antigenic challenge are different for the closely related antigens SRBC and
BRBC (burro red blood cells). Following the "hot" pulse at 24-48 hr, which kills all those
precursors which had responded to SRBC, there is no other remaining population of
precursors which can respond to SRBC (as there would be under an "instructional" scheme),
although the precursors for a different antigen have not been effected by this treatment.
These two sets of experiments, and many more in later years, confirmed the basic features of
CLONAL SELECTION--separate precommitted populations of precursor cells respond (by
proliferation and differentiation) to different antigens, and each have antigen-specific
receptors on their surface.

57

CLONAL SELECTION AND TOLERANCE

One of the most striking features of Burnet's Clonal Selection theory was its simple
explanation of the phenomenon of SELF-TOLERANCE, an organism's normal lack of ability
to make immune responses against its own "self" components, while simultaneously being
able to respond to any "foreign" antigen Burnet hypothesized that during the development of
an antibody-forming cell precursor, it goes through a short-lived stage in which exposure to
its specific antigen will result in its own death. This would ensure that no precursor can
develop which recognizes antigens ordinarily present in the organism, but all other precursors
would develop normally. This represents tolerance by the mechanism of CLONAL
ABORTION, and was the most attractive explanation of the phenomenon of tolerance for
many years.
The phenomenon of tolerance will be discussed at greater length in Chapter 18, and we will
find that while clonal abortion is known to take place and to play an important role, it clearly
fails to account for some of the features of natural or experimental tolerance. It can be shown
in many cases, for instance, that some clonally specific precursor cells specific for "self"
antigens do exist in normal animals; therefore, the fact that they do not ordinarily respond and
become antibody-forming cells must be explained on some basis other than clonal abortion.
CLONAL SELECTION AND MEMORY
Antigen-dependent proliferation of clonally precommitted cells has at least two important
consequences.
1) Production of a large number of effector cells of that particular specificity (Clonal
Expansion). A single antigen-reactive precursor may give rise to hundreds of
thousands of antibody-forming cells.
2) Production of an expanded number of memory cells. These memory cells are
themselves antigen-reactive cells, which can respond to antigen stimulation by
proliferation and differentiation into antibody-forming cells. However, they differ from
the original "virgin" precursors in several ways, including that they are considerably
more numerous., and that they respond much more rapidly to antigenic stimulation,
The increased speed of response of memory cells, as well as their higher numbers, is the
basis of immunological memory. This will be discussed in more detail in Chapter 9 (see
Figure 9-3).
CLONAL SELECTION AND AFFINITY MATURATION
We have already noted that secondary (memory) responses produce antibody of higher
affinity than primary responses. This is one manifestation of the phenomenon of affinity
maturation which can also be readily understood in the context of clonal selection. The
heterogeneity of antibody responses is a reflection of the heterogeneity of AFCPs. Many
different precursors will exist, for example, which can recognize the hapten DNP, each with
its own characteristic affinity. Lets assume, for the argument, the existence of seven
different precursor cells with affinities for DNP ranging from 104 to 1010. In a typical

58

immune response there will be enough antigen to trigger all seven (they will all differentiate
and proliferate), resulting in an average antibody affinity of about 107 (see the left hand side
of Figure 7-4).

4
5
6

DNP-BSA

average

K=10 7

average
K=10

DNP-BSA

9
10

10

EARLY

LATE

AFFINITY MATURATION BY ANTIGEN-DRIVEN SELECTION

Figure 7-4
As the response progresses, one of its consequences is the elimination of antigen (by
"processing", opsonization and phagocytosis, etc.). Before the antigen is completely
eliminated, however, the small amount that still remains may continue to trigger available
precursor cells. However, if the amount of remaining antigen is limiting and there is not
enough to trigger the entire population of available precursors, those precursors for higher
affinity antibody will be selectively triggered. If, in the example above, only those precursors
in the range of 106 to 1010 are re-stimulated later in the response, they will be selectively
expanded and the average affinity of serum antibody will increase toward 108 (shown on the
right-hand side of Figure 7-4).
Thus, affinity maturation is a logical consequence of the heterogeneity of precommitted
AFCPs, and of the fact that antigen can continue to trigger precursor cells as long as it is
present. (The process of somatic hypermutation, discussed in Chapter 9, increases the
diversity of antigen-combining sites during immune responses, thus contributing to even
more effective affinity maturation.)
"NATURAL" ANTIBODIES?
As we have seen, antibodies are produced following stimulation by specific antigens. The
total immunoglobulin in our serum, which typically amounts to 10-15 mg/ml (1-1.5 gm/dl), is
the result of extensive exposure to many immunogenic substances in our food and on the
59

microorganisms which inhabit and infect us, as well as from deliberate vaccinations.
Experimental animals which are raised under germ-free and antigen-free conditions have
little or no detectable serum Ig; if they are subsequently exposed to a conventional
environment, however, their Ig levels rapidly rise to reach the normal range.
Are there any antibodies which can truly be called "natural" and which are not the result of
specific antigen stimulation? Perhaps, but only in a limited sense. It is important to
recognize that immune responses often include non-specific components -- in addition to
resulting in a specific antibody response, antigenic stimulation may result in non-specific
stimulation of nearby bystander B-cells, a phenomenon known as POLYCLONAL
ACTIVATION (we will discuss some of the mechanisms involved later in Chapters 12 and
15.) This is thought to reflect the non-specific stimulation of pre-existing memory cells
(among others) and, therefore, will tend to be biased toward the products of previous immune
responses. In general, the term "natural antibody" simply means antibody which is not the
result of a specific and known antigenic stimulation.

CHAPTER 7, STUDY QUESTIONS:

1.

How does CLONAL SELECTION differ from instructional theories in explaining

antibody diversity?

2.

How do the experiments of Ada and Byrt (hot antigen suicide) and Mishell and
Dutton (hot thymidine suicide) support Clonal Selection? How do they differ from
one another?

3.

Does either of the above experiments exclude the possibility that a single AFCP may
be capable of producing two different antibodies with different specificities? Does
this situation occur in nature?

4.

Why is affinity maturation a logical consequence of Clonal Selection?

60

CHAPTER 8
GENETIC BASIS OF ANTIBODY DIVERSITY
SEE APPENDIX (9) SOUTHERN BLOTTING
Underlying the development of antibody diversity is a unique pattern of gene
organization and molecular events. THREE FAMILIES OF IMMUNOGLOBULIN
GENES exist in mammals, one encoding HEAVY chains, another KAPPA chains, and
the third LAMBDA chains. Each of these clusters contains one or more constant region
genes and a number of variable region gene segments. The formation of a complete
variable region of a light or heavy chain requires the joining of two or three separate
genetic elements by a process of GENE REARRANGEMENT; a separate DNA
rearrangement in the heavy-chain complex is required for subsequent CLASSSWITCHING. Both germ-line and somatic events contribute to antibody diversity,
including
COMBINATORIAL
JOINING,
SOMATIC
MUTATION
and
COMBINATORIAL ASSOCIATION. This gene organization and the requirement for
gene rearrangement is unique to immunoglobulins, with the single exception of the
related family of genes encoding the T-CELL RECEPTORS, whose functions will be
discussed later.
The problem of the genetic basis of antibody diversity is one which was hotly debated
through the 1960's and 70's, and only through the application of recombinant DNA
technology have its essential features become well understood.
The basic problem can be set out as follows: if there are one million different antibodies
which the immune system can produce, and if each of them has a unique primary structure
(i.e., amino acid sequence), are there then a million genes required for their production, or
can a smaller number of genes be modified in some systematic way to account for the total
diversity? These two possibilities define the essential features of two competing hypotheses,
GERM LINE generation of diversity on the one hand versus SOMATIC generation of
diversity on the other.
We can simplify the problem by recognizing that the antibody combining site (which is, of
course, responsible for specificity) is made up of two elements, the VH and VL. If as few as
one thousand VL domains combine randomly with the same number of VH domains, we could
account for one million different combining sites (103X103=106).
If we use the kappa light chain system as an example (recognizing that the same arguments
can be applied to the other V-region families), we can define the extreme forms of GERM
LINE and SOMATIC theories as follows:
GERM-LINE THEORY -- For every kappa-chain V-region there exists one unique
germ-line gene. A particular antibody-forming cell selects one of these and expresses it
in unmodified form.
SOMATIC THEORY -- Only a single germ-line gene exists for all kappa-chain Vregions. A particular antibody-forming cell expresses this gene following a process of
somatic mutation, which results in each cell expressing a different version of this gene.

61

Decades of genetic studies, culminating in the cloning and sequencing of immunoglobulin

genes, have shown that significant features of both theories are correct. Many genes exist
for V-regions of immunoglobulins, and these genes are somatically modified in a variety of
ways in the course of their expression in antibody-forming cells.
STRUCTURE AND EXPRESSION OF IMMUNOGLOBULIN GENES
Three families of immunoglobulin genes exist, each on a separate chromosome. One includes
kappa genes, another lambda genes, and the third includes all the heavy chain genes. Each
family consists of a series of V-regions genetically linked (but still separated by long
stretches of DNA) to one or more C-regions.
We can diagram the three human Ig clusters as follows:
Human
chromosome #

kappa:
V1

V2

V(n)

V2

V(n)

Vh2

Vh(n)

lambda:
V1

J C1 J C2 J C3 J C7

22

heavy:
Vh1

Dh

Jh

C3 C1

C1 C2 C4 C C2

14

Figure 8-1
Lets examine in some detail the structure of the kappa chain complex and the processes
involved in its expression, referring to Fig. 8-2 below. This complex consists of a large
number of V-region genes (about 55) genetically linked (by a long stretch of DNA) to a
single copy of the constant region gene. An additional cluster of five short gene segments
called J-segments is located a few thousand base pairs upstream (5 direction) of the C-region
gene, and each of these codes for the last 13 amino acids of the variable region. (NOTE:
Dont confuse these "J-segments" with the "J-chain," the polypeptide attached to IgM and
polymeric IgA, which we will discuss in Chapter 9.) This is the "germ-line" configuration,
present in germ cells (sperm and eggs) as well as all somatic cells other than Ig-producing
cells, and is shown at the top of Figure 8-2.
The first step in expression of this gene is DNA REARRANGEMENT, involving the joining
of one V-region and one J-segment, each chosen at random in any given B-cell. The result is
the structure in line 2 of the figure, and the DNA which was originally between the selected V
and J genes is cut out and lost in the form of a closed circular molecule; other V-region

62

MOLECULAR BASIS OF KAPPA GENE EXPRESSION

Germ-line configuration:
V1

J1 J2 J3J4 J5

V3 . . .

V2

3'

5'

1) Germ-line
DNA

Following DNA rearrangement:

V1

V2 J3

3'

5'

2) Rearranged
DNA

transcription of mRNA

Following transcription:

V2 J3

5'

3) Precursor
mRNA

3'

Following RNA splicing:

V2 J3 C

5'

3'ut

AAAAA

4) Mature
mRNA

3'

Following translation:
L

NH 2

COOH

5) Precursor
PEPTIDE

COOH

6) Mature
PEPTIDE

Following peptide processing:

V

NH 2

Figure 8-2

63

gene segments which happen to reside outside this excised segment (to the left, or 5, of the
V-region which is used (for example, V1 in line 2 of Figure 8-2) are retained, although they
are no longer relevant to expression, described below. This process is unique to
immunoglobulin (and T-cell receptor) genes.
The process of TRANSCRIPTION starts at the beginning of the rearranged V-region and
continues past the end of the C-region, resulting in an immature mRNA whose structure is
shown in line 3. The large intervening sequence (intron) between the J-segments and the
C-region is removed by the process of RNA SPLICING, resulting in the mature mRNA
shown in line 4. It includes a 200 nucleotide region at its 3' end which is not translated, and a
3 poly-A tail. The structure of this transcript and the processes by which it was produced
(transcription and RNA splicing) are characteristic of most eukaryotic genes.
TRANSLATION of this message occurs on ribosomes associated with the rough
endoplasmic reticulum (RER), and the resulting polypeptide has the structure shown in
line 5. It is identical to the known structure of immunoglobulin kappa chains with a
single exception -- it has an additional 13 or so amino acids at its amino terminal end, known
as the LEADER or SIGNAL sequence. This sequence is required for the transport of the
polypeptide across the endoplasmic reticulum membrane into the lumen of the ER, and is
cleaved off when the polypeptide moves into the cisterna of the rough ER. The presence of
the leader sequence and its proteolytic cleavage are general features of all secreted proteins
in eukaryotes and prokaryotes.
The structure and mechanism of expression of lambda chains and heavy chains are similar to
what we have just described for kappa chains--all have J-SEGMENTS, all show DNA
REARRANGEMENT, TRANSCRIPTION, RNA SPLICING, TRANSLATION and
proteolytic cleavage of the LEADER POLYPEPTIDE. Heavy chain gene structure is
somewhat more complex, however, as there also exists an additional cluster of gene segments
(known as D, for diversity, segments) which each encodes four amino acids between the
V-region cluster and the J-segments. DNA rearrangement for H-chains thus involves two
events, joining of a V with a D, and joining of the D with a J-segment. Transcription and the
other processes discussed above take place as they do for kappa genes.
In each case, the end result is a polypeptide whose amino acid sequence has been determined
by three or four separate genetic elements, and which is incorporated into the final
immunoglobulin molecule by processes which will be discussed in Chapter 9.
ALTERNATE SPLICING IN B-CELLS
One unusual phenomenon still needs to be explained, i.e., the simultaneous synthesis of IgM
and IgD by a single B-cell. This is the only example of a normal cell simultaneously
producing two different kinds of immunoglobulin. The explanation derives from the fact that
mu () and delta () constant region genes are adjacent to one another in the heavy chain gene
complex, as illustrated in Figure 8-3. Using the same rearranged heavy chain V/D/J complex,
a B-cell can make two kinds of mRNA--it can transcribe from the V-region through the end
of the C gene and make IgM, or it can transcribe all the way through the C gene and make
IgD by splicing out the C region together with the intervening sequence during RNA
splicing. Two different mRNAs can thus be made from a single gene complex. It should be
emphasized that alternate splicing of RNA is a mechanism used by many other genes to
generate diverse protein products.
64

mu-chain mRNA

DJ

C
AAAAAAA

V DJ

(Rearranged DNA)
L

DJ

delta-chain mRNA

AAAAAAA

Dashed lines in RNA indicate an intervening

sequence removed during RNA processing

SIMULTANEOUS SYNTHESIS OF IgM AND IgD IN B-CELLS

BY ALTERNATE RNA SPLICING
Figure 8-3
This process accounts for two important phenomena. First, IgM and IgD synthesis in B-cells
can occur simultaneously and continuously (it is not a transient result of a switch from mu to
delta). And second, it has been shown that the mu and delta chains are produced from the
same chromosome, and not from the two different allelic copies. This is an extension of
allelic exclusion which will be discussed in Chapter 9, and is known as haplotype exclusion.
Alternate splicing of mRNA also accounts for another important feature of immunoglobulin
expression, namely the choice of whether a secreted versus a membrane-bound form of Ig is
produced. All classes of Ig can be produced either as membrane-bound molecules expressed
on the surface of B-cells, or as secreted molecules released into the serum or extracellular
space. This difference is determined by which of two alternate exons is selected to be present
at the 3' end of the heavy chain mRNA, which in turn depends on alternate mRNA splicing.
Regulation of such mRNA splicing is therefore an important element in the differentiation of
the B-cell lineage.

HEAVY CHAIN CLASS SWITCHING: A SECOND DNA REARRANGEMENT

While antibody-secreting cells begin their existence by secreting IgM, the progeny of an
individual cell can go on to produce IgG, IgE or IgA. This phenomenon is known as CLASS
SWITCHING and occurs by another DNA rearrangement event (shown in Figure 8-4,
below), which relocates the already rearranged V/D/J complex from its original position near
the C gene to a position close to one of the other heavy-chain C-regions (3 in this
example). This results in a new transcription unit and the synthesis of a heavy chain with
same V-region but a new C-region.

65

Transcript before switch: IgM

C

L V DJ

C 3

DNA rearrangement ("switch"):

C 3

L V DJ

Transcript after switch: IgG3

MOLECULAR BASIS OF CLASS SWITCHING

Figure 8-4
This process accounts for two of the key features of class switching. First, as mentioned
above, the VH-region remains the same. Since switching does not change the light chains,
this means that the specificity of the antibody (and its idiotype) cannot change. Second, class
switching is unidirectional and irreversible. A cell can switch from one C-region to another
situated to the right of the first (5-->3' direction), but cannot go backwards (3-->5
direction). The order of switching is therefore defined by the order of the heavy chain
constant region genes on the chromosome. [NOTE: CH genes are each made up of a number
of exons and introns which are not shown in this schematic.]
SOURCES OF ANTIBODY DIVERSITY: SUMMARY
Lets return to our original question of the Generation of Diversity and examine the genetic
basis for the existence of a million or more antibody specificities. Well use what we now
know of human kappa chain gene structure and expression as our primary example.
1) About 50 V
-genes exist in the human genome. This is a fairly large number, much
higher than somatic theories predicted, although smaller than predicted by most germline theorists.
2) Five different J
-segments exist in the human genome. Thus, if every V-region can
be used with any J-segment, a total of about 250 (50 x 5) different combinations can be
made. (In the case of heavy chains, the existence of several D-segments increases this
number even more.) This process is referred to as COMBINATORIAL JOINING.
3) DNA rearrangement is imprecise. The joining of V and J (and the joining of D
segments to V and J) is a deliberately error-prone process, resulting in random
nucleotide substitutions and deletion/insertions at the site of rearrangement. Therefore,
the joining of a particular V-region and a particular J-segment can yield different results
in different cells in which it occurs. If this increases the number of potential V-regions
by about ten-fold, we now have about 2500 possible kappa chain V-regions (250 x 10).
4) Somatic mutation occurs in rearranged V-regions. Expressed kappa chain genes
have been cloned and sequenced from many antibody secreting cells, and in most cases

66

they have been found to differ from any existing germ-line V-region in sequence
positions other than the site of joining of V and J. These somatic mutations tend to be
localized to the hypervariable regions, and occur by a local relaxation of the normal
processes of error-correction in newly synthesized DNA. The presence of these
mutations may increase the number of possible V-regions by another ten-fold or more,
resulting in at least 25,000 different kappa V-region sequences (2500 x 10).
5) Heavy and light chains associate in random combinations. If a comparable number
of different H-chain V-regions can be produced (which is an underestimate), then there
are potentially some 6x108 different antibody combining sites (25,000 X 25,000), a
number considerably larger than the one million we initially set out to explain. This
random association of H- and L-chains is referred to as COMBINATORIAL
ASSOCIATION.
It has therefore become clear that the historical debate between proponents of GERM-LINE
and SOMATIC theories of antibody diversity is no longer meaningful, and that elements of
both theories are important in the generation of antibody diversity. Points 1, 2, and 5 above
fall within the framework of the germ-line theory, while points 3 and 4 quite clearly represent
somatic processes.

T-CELL RECEPTORS: Ig-LIKE GENE ORGANIZATION

AND GENE REARRANGEMENTS
The organization of immunoglobulin V- and C-region segments, and the processes by which
they are rearranged, are clearly well-adapted for the generation of the huge diversity which so
important for the humoral immune system. Such gene rearrangements are, in fact, unique to
the immunoglobulin genes, with the single exception of the T-cell receptor (TCR), whose
roles in immune responses we will discuss later (see Chapters 14, 15 and 18). These
molecules, although quite distinct from immunoglobulins, are evolutionarily related to them,
and share many important properties with them. They are a highly diverse family of
heterodimeric membrane receptors capable of specifically interacting with antigens, their
polypeptide chains include variable and constant regions, and the genes encoding them are
organized and rearranged in a manner very similar to those of immunoglobulins.
There are, however, three critical differences in the structure and expression of TCR's
compared with Ig: 1) Each TCR molecule bears only a single combining site (it does not
exhibit the bivalency characteristic of all antibody molecules); 2) Somatic mutation, an
important component in the generation of antibody diversity, does not occur during TCR
expression. 3) Unlike immunoglobulins which can exist as either soluble or membranebound forms, TCR's function only as membrane-bound molecules. In later chapters well
discuss the unique biology and antigen-recognition functions of the TCR, and it will become
clear why these differences between Ig and TCR are centrally important.

67

RAG-1 AND RAG-2 RECOMBINASE

Despite the important differences between immunoglobulin and T-cell antigen receptors, they
both express unique antigen-combining sites which, as described above, can be assembled
only as a result of V(D)J recombination of germ-line DNA. In both cases this process
requires the participation of two proteins known as RAG-1 and RAG-2 (for "Recombination
Activating Genes"). These recombinase enzymes first appeared during the evolution of
cartilagenous fish, and are present only in those organisms capable of carrying out adaptive
immune responses, namely the "higher" vertebrates, including fish, reptiles, birds,
amphibians and mammals. Genetic knock-out of the RAG proteins in mice results in the
complete inability to produce either immunoglobulins or T-cell receptors, and consequently
the absence of both mature B-cells and T-cells. Such knock-outs have been widely used to
manipulate and study the vertebrate immune system.

CHAPTER 8, STUDY QUESTIONS:

1.

Describe the overall architecture of the three mammalian IMMUNOGLOBULIN

GENE FAMILIES.

2.

Describe the molecular processes involved in expression of a kappa light chain gene.
What processes are unique to immunoglobulins? In what ways does expression of a
heavy chain differ from that of light chains?

3.

What are the various processes which contribute to antibody diversity?

4.

In what ways can DNA rearrangements be used to help define and diagnose
lymphocyte tumors?

5.

What would be the expected phenotypic consequence of a null mutation in the human
RAG-1 or RAG-2 gene?

68

CHAPTER 9
IMMUNOGLOBULIN BIOSYNTHESIS

Although the process by which a functional gene for immunoglobulin HEAVY and
LIGHT CHAINS is formed is highly unusual, the SYNTHESIS, POSTTRANSLATIONAL PROCESSING and SECRETION of these glycoproteins all occur
by conventional pathways. The J-CHAIN is required for polymeric Ig and the
SECRETORY PIECE for the unique process by which secretory IgA is transported into
exocrine fluids. Knowing the molecular and cellular mechanisms involved in Ig
production, we can understand the importance of ALLELIC EXCLUSION and the
structural differences between SECRETED versus MEMBRANE-BOUND FORMS of
Ig, and provide a rational basis for the existence of CLASS SWITCHING.

Most circulating immunoglobulin is produced by plasma cells, which are highly specialized
and efficient antibody "factories". The biosynthesis of immunoglobulins has much in
common with the biosynthesis of other secreted proteins. Several aspects of this process,
however, are unusual, and some features are quite unique to immunoglobulins.
The structural components of Ig molecules we need to account for are as follows:

Light Chains
Heavy Chains
Carbohydrate (on H-chain only)
J-Chain (attached to H-chain of IgM and polymeric IgA)
Secretory piece (attached to H-chain of secretory IgA)

Lets follow the process of synthesis and assembly of an immunoglobulin molecule in a

plasma cell, referring to the diagram (Fig. 9-1) below.
FEATURES OF IMMUNOGLOBULIN SYNTHESIS
1) Conventional mRNA molecules are produced separately for the light and heavy chains,
each containing uninterrupted coding sequences for V and C-regions. While the
structure of the mRNA is conventional, some of the mechanisms involved in its
synthesis (the DNA rearrangements discussed earlier) are quite unique to
immunoglobulins. Translation of the message occurs on the ribosomes of the rough
endoplasmic reticulum (RER).

69

IMMUNOGLOBULIN BIOSYNTHESIS
mRNA
L

C
AAAAA

NUCLEUS

RER

Peptides

pre-mRNA

L
C

CYTOPLASM

H- H

H-L

H-H-L
L-H-H-L

Assembly

CHO
GOLGI
Post-Golgi
vesicles

secreted
IgG

Secretion

IgM
(or IgA)
+J-Chain)

Membrane IgG
(or IgMS)

secreted
IgM

Figure 9-1

70

2) Nascent polypeptide chains are translocated across the membrane into the cisternae of
the RER. As is the case for any secreted protein, this process requires the presence of
the leader polypeptide or "signal sequence" at the amino terminus, which is
enzymatically cleaved immediately after translocation.
3) Assembly of the heavy and light chains into a typical IgG-like subunit (H2L2) by
formation of disulfide bonds is a spontaneous process - no specific enzymes are
thought to be required, although a member of the HSP70 class of molecular
"chaperonins" (the Heavy Chain Binding Protein, or BiP) is known to facilitate the
proper folding of the heavy chains prior to their association with light chains. In
normal plasma cells one sees balanced synthesis of H and L-chains, or a slight excess
of L-chains (which are degraded intracellularly). Myeloma cells, on the other hand,
often show markedly unbalanced synthesis with a large excess of L-chains secreted into
the circulation which are then passed into the urine as Bence-Jones proteins. Some
myelomas produce only light chains or only heavy chains; in the latter case (HeavyChain Disease) the H-chains are often defective, and typically have a deleted hinge
region.
4) Assembled H2L2 molecules move through the RER to the Golgi apparatus, and into the
post-Golgi vesicles. These vesicles then fuse with the external cell membrane and
release their contents, resulting in secretion of Ig from the plasma cell.
5) Carbohydrate is added to H-chains in a well-defined order, beginning on nascent chains
while they are still attached to ribosomes, and continuing throughout the process of
movement through the cell right up to the point of secretion. Carbohydrate is added to
asparagine residues which are part of a specific recognition sequence, namely an
asparagine separated by one amino acid residue from a serine or threonine:
Ser
Asn -XXX
CHO

Thr

In normal antibody molecules carbohydrate is attached only to H-chain, and only in the
constant region, because this recognition sequence is normally present only in CH.
Carbohydrate is a universal part of Ig heavy chains, and is thought to stabilize the threedimensional structure of the Fc. [Note: In some myeloma proteins and other
monoclonal immunoglobulins one sometimes finds VH or VL-associated carbohydrate
in those cases where, through mutation, a recognition sequence happens to be present in
the V-region. Carbohydrate will be attached to any accessible recognition sequence by
the responsible enzyme system.]
6) Just prior to release of the immunoglobulin from the cell, two more events occur:
a)

The addition of a large final lump of carbohydrate. (The function of H-chain

associated carbohydrate is still not fully understood, although it has been shown
that its absence may affect the binding of Ig to Fc-receptors.)

71

b) In the case of IgM and polymeric IgA, the addition of the J-chain. The J-chain is
appears to be required for the process of polymerization, although it can be
experimentally removed from polymeric Ig without disrupting its structure.
NOTE: Not all Ig produced by a cell is necessarily secreted - some may be retained in
the cell membrane and function as the B-cell's antigen receptor. The structural
differences between secreted and membrane-bound forms of Ig are discussed below.
SECRETION OF IgA INTO EXOCRINE FLUIDS
We have accounted for all of the structural components of immunoglobulins except for the Spiece (secretory piece) associated with secretory IgA. The S-piece is not synthesized by the
plasma cell which produced the immunoglobulin, but is added during the process of secretion
into exocrine fluids. Lets look at the example of secretion of an IgA molecule into the gut in
Figure 9-2 (although secretory IgA also appears in other exocrine secretions such as saliva
and bile).

(IgA) 2

LUMEN

Sf
(IgA)2
(IgA)2

(IgA)2

(IgA)2

EXTRACELLULAR SPACE

TRANSPORT OF EXOCRINE IgA

Figure 9-2
1) Polymeric IgA [mostly (IgA)2, with a small amount of (IgA)3], is secreted by plasma
cells into the extracellular fluids of the lamina propria of the gut.
2) Epithelial cells lining the gut have S-piece precursor (labeled Sp in the figure above) in
their abluminal membranes. This Sp behaves as a membrane receptor and binds
polymeric IgA.
3) Polymeric IgA thus bound is internalized, transported to the luminal side of the cell,
and is released into the lumen of the gut following proteolytic cleavage of the Sp into
two fragments.

72

4) Secretory IgA now has conventional S-piece (one portion of the Sp) covalently linked to
its H-chain; the rest of the Sp molecule remains in the epithelial cell membrane (as Sf,
or S-fragment) where it is internalized and degraded.
5) Only polymeric IgA (and to a slight extent IgM) is transported in this manner,
presumably because other Ig molecules simply do not bind to the Sp receptor on the
epithelial cell. Monomeric IgA, the major serum form of IgA, is not secreted.
6) The presence of IgA in exocrine secretions can be of considerable importance in
protection against infectious organisms. The normal route of entry of polio virus, for
example, is the intestinal tract, and secretory IgA in this location confers a high degree
of protection.
Weve now seen two processes involving immunoglobulins which are both referred to as
"secretion", and it is important not to confuse them.
Secretion of Ig from plasma cells: Mature immunoglobulin molecules are "secreted"
from the plasma cell into the extracellular space by the conventional process of
exocytosis of post-Golgi vesicles. This same process is responsible for the release from
cells of all "secreted" proteins.
Exocrine secretion of IgA: Secretion of polymeric IgA into the mucus of the gut and
other exocrine fluids is a distinct process of "secretion". This process, as we have seen,
depends on the participation of the "secretory piece", or "S-piece", which is added to IgA
by secretory epithelial cells. This process is unique to immunoglobulins, and IgA in
particular.
KEY FEATURES OF ANTIBODY PRODUCTION
1) One antibody-forming cell (AFC) cell produces only one kind of antibody.
a) only one light chain isotype ( or 1 or 2, etc).
b) only one heavy chain isotype.
c) only one kind of VH and one kind of VL; therefore, only a single idiotype.
One consequence of these features is that antibody molecules are always symmetrical
with respect to their light and heavy chains.
(NOTE: There are some exceptions to the strict one-cell-one-antibody rule. One is
the transient simultaneous synthesis by virgin B-cells of both IgM and IgD - the
mechanism of this dual synthesis has been discussed in Chapter 8. Another
situation which hardly qualifies as an exception is discussed under point [4],
below. Neither exception violates the basic principle of the symmetry of Ig
molecules, or rules [a] and [c] above.)

73

2) Allelic exclusion. While diploid cells have two copies of every immunoglobulin gene,
only one of the two is expressed in a given B-cell or plasma cell for each of the light
and heavy chain (the other allele is either rearranged aberrantly and cannot be
expressed, or is not rearranged at all). This is an unusual and important feature of
immunoglobulins (and the TCR); although recent work has shown that there exist a
number of other genes that also exhibit this behavior.
One of the consequences of allelic exclusion is that it ensures the symmetry of antibody
molecules. Expression of both copies of the kappa locus (for example) would result in
production of Ig moleculeswith two different V regions, and therefore two different
combining sites (remembering that the assembly of light and heavy chains is a random
process). Such asymmetric molecules would have impaired biological activity (see the
earlier discussion of affinity and avidity in Chapter 3), and would substantially dilute
the pool of active bivalent molecules.
3) Heavy chain class switching. A particular antibody-forming cell can switch from
production of IgM to IgG, from IgG to IgA, etc. The light chain does not change, nor
does the VH; only the CH changes, so that the original combining site (and therefore
idiotype) is now associated with a molecule of a different class or subclass.
Switching is unidirectional--an IgG-producing cell cannot go back to production of
IgM, for instance. The molecular basis of this irreversibility was discussed in the
Chapter 8.
Very rare myelomas have been found which produce two or more classes of
immunoglobulin (double producers). These proteins generally differ only in their CH
and are considered to represent the results of class switching in these cells, the
myeloma cells having gotten "stuck" during this process.
4) Membrane-bound versus secreted immunoglobulins. A virgin B-cell bears IgM
(and possibly IgD) in its membrane; following stimulation it begins to secrete IgM into
its surrounding environment. These two forms of IgM are structurally different. The
mu heavy chain of membrane-bound IgM has an amino acid sequence at its carboxyterminal end which anchors the molecule into the membrane; the secreted form of IgM
has a different C-terminal sequence which lacks a membrane anchoring region. The
membrane-bound form of IgM is also incapable of associating with J-chain and
forming its normal pentameric structure; membrane IgM, therefore, exists exclusively in
monomeric form (H2L2), also known as IgMS or "sub-unit" IgM.
The two forms of mu chain are synthesized via an alternative splicing scheme
analogous to that which allows simultaneous IgM and IgD synthesis (as was discussed
in Chapter 8). Similar differences exist for secreted versus membrane-bound forms of
IgG, IgA and IgE heavy chains.
A virgin B-cell produces only the membrane-bound form of IgM. As it develops into
an antibody-secreting plasma cell it may transiently produce both membrane and
secreted IgM, but it eventually produces only the secreted form. Likewise, a memory
B-cell which synthesizes only membrane-bound IgG will shift to producing exclusively
the secreted form of IgG as it develops into a plasma cell following secondary antigen
stimulation.

74

Figure 9-3 shows the relationships between "virgin" B-cells (BV, IgM-bearing), "memory" Bcells (BM, shown as IgG-bearing, but can also be IgA or IgE-bearing), and antibody-forming
cells (AFC, typically with little or no membrane-bound Ig). This illustrates some of the
features of primary and secondary immune responses we have already discussed in Chapter 7
(e.g., more rapid and efficient generation of AFC's from memory cells than from virgin Bcells), but also adds those membrane phenotypes of the participating cells which are a
consequence of the molecular processes we have just covered (class switching from IgM to
IgG, A or E). Note that the membrane-bound IgM (and IgD) which characterizes virgin Bcells is lost during the generation of either AFCs or memory cells.
IgM

IgD

BV

BV

primary
response

= "virgin" B-cell

AFC = antibody-forming cell

BM

= "memory" B-cell
= differentiation
= with class switching
and somatic hypermutation

IgG
AFC

AFC

AFC

AFC

BM

= with somatic
hypermutation

IgG

BM

secondary response

Secreted Ab (IgM)

IgG
AFC

AFC

AFC

AFC

AFC

AFC

IgG

BM

IgE

IgG

BM

BM

BM

Secreted Ab (IgG)

IMMUNOGLOBULIN EXPRESSION BY B-CELLS AND THEIR PROGENY

Figure 9-3
5) Somatic Hypermutation. The V-regions of membrane Ig on memory cells typically
express sequences different from those of the nave B-cell from which they were
derived, resulting in a marked increase in overall diversity which is reflected in the
serum antibody response. This is the result of the accumulation of point mutations
through a process called somatic hypermutation. As indicated by the broken lines in
the figure above, this process occurs only during the generation of B-memory cells
following isotype switching, and therefore affects only IgG, IgA and IgE.
The mechanism involves the inactivation of the normal process of DNA editing, which
exists to maintain sequence integrity during DNA replication. This inactivation,
however, takes place only over a limited region within and near the V-regions of both
heavy and light chains, excluding the constant regions. The point mutations which
occur are, of course, random, but the only ones we will observe are the minority which
happen to increase antibody affinity, thereby allowing the variant B-cells to be selected
by antigen for continued proliferation and differentiation.

75

WHY CLASS SWITCHING?

B-cells start out with IgM and IgD on their surface; after antigenic triggering they lose their
IgD and become IgM secretors. Most of them will subsequently switch from IgM to
production of some other class of immunoglobulin during the course of an immune response.
Why, then, do immune responses not start immediately with the final mixture of
immunoglobulin isotypes? Why start every clone with IgM and go through the complex
process of class switching?
There is clearly an advantage to ending up with a variety of isotypes represented in serum,
because each will have its distinct set of biological activities. But the question still remains,
why start everything off with IgM and only switch afterwards?
We can better understand this by recalling the phenomenon of affinity maturation:
a)

As a consequence of clonal selection, the affinity of antibodies produced early in a

response will always be lower than those produced later on (remember that the
generation of the antibody repertoire is random).

b) Low affinity antibodies will tend to be less biologically effective than higher affinity
ones because they bind less well to their target antigens.
c)

The immune response can compensate for the low affinity of early antibodies by
making polymeric IgM which will have a very high avidity, even with very low affinity
combining sites.

d) However, IgM is very expensive for a cell to produce, it requires six times more raw
materials and energy per molecule than IgG. Once the immune response has had time
to become established using IgM antibodies, the process of affinity maturation can
allow the expensive IgM to be replaced with more economical IgG-like antibodies, also
gaining the additional benefit of a wide diversity of biological effector functions.

CHAPTER 9, STUDY QUESTIONS:

1.

List each of the structural components which make up the various isotypes of human
immunoglobulins. Describe the processes by which each is incorporated into a
mature immunoglobulin.

2.

What are the two distinct processes which are referred to as immunoglobulin
"secretion"?

3.

Why is allelic exclusion such a critically important feature of AFC's?

76

CHAPTER 10
BLOOD GROUPS: ABO AND Rh

The success of human blood transfusions requires compatibility for the two major
blood group antigen systems, namely ABO and Rh. The ABO system is defined by two
red blood cell antigens, A and B, whose presence or absence is determined by three
alleles (A, B, O) segregating at a single genetic locus. An unusual feature of this system
is the presence of serum IgM antibodies in healthy adults to whichever antigen (A or B) is
absent from that individual's cells. The presence or absence of Rh antigens on red blood
cells is determined by two alleles at another locus, Rh. Rh INCOMPATIBILITY
between mother and infant may result in ERYTHROBLASTOSIS FETALIS, which can
be prevented by passive immunization of the mother with anti-Rh antibodies
(RHOGAM).

Many lives are saved throughout the world each year through the use of blood transfusions,
by preventing death from loss of blood due to trauma, and by allowing performance of
surgical procedures which would otherwise be impossible. However, it was not until the
early nineteen hundreds that routine blood transfusion between humans became possible,
following the discovery that genetically determined differences exist between the blood of
different individuals. These differences must be identified and compatibility ensured before
transfusions can be safely carried out.
Many genetic systems controlling blood groups in humans are known, and they have been of
considerable importance in our understanding of human genetics, as well as in both clinical
and forensic medicine. We will discuss briefly the major features of the two most important
of these systems, those controlling the ABO blood groups and the Rh factor.
ABO BLOOD GROUPS
A and B antigens
The ABO blood groups are defined by the presence of two alternative antigens on red blood
cells, determined by three alternative alleles at a single genetic locus. Two basic rules
governing this system are as follows:
1) The blood "type" is defined by the presence of two red blood cell antigens, "A" and
"B." RBCs of type A have the A antigen on their surface, those of type B have antigen
B, type AB red cells bear both antigens, while type O cells bear neither antigen.
2) "Natural" antibodies called isoagglutinins exist in an individual's serum, directed
against whichever of the A and B antigens is not present on that person's red cells (we
will examine the source of these antibodies later).

77

The presence of ABO antigens and antibodies (isoagglutinins) in the four blood types is
summarized below:
BLOOD
TYPE

RBC
ANTIGENS

SERUM
ANTIBODIES

anti-B

40%

anti-A

10%

AB

A and B

none

none
anti-A
and anti-B

FREQUENCY

5%
45%

The success of blood transfusions depends on ensuring the compatibility of the blood types
between donor and recipient. If the recipient has antibodies to the infused red cells, these red
cells will be rapidly destroyed, resulting in a potentially lethal transfusion reaction. Type A
blood given to a type B recipient, for instance, can result in such a reaction, since the
recipient's serum contains anti-A antibodies.
Blood typing & crossmatching
The presence of these antigens and antibodies can be readily detected by the agglutination
reaction; mixing type A plasma (which contains anti-B antibodies) with type B red cells, for
instance, results in agglutination of the red cells which can be easily observed. The blood
type of any given individual can be determined in this manner, carrying out the agglutination
reaction with a set of standard antibody-containing sera. Before a blood transfusion is given,
in addition to choosing donor blood only of the same ABO type, direct crossmatching of
donor and recipient is also generally carried out. This involves mixing donor RBCs with the
recipients serum to guarantee that the original typing of donor and recipient was correct, and
to detect any possible unexpected agglutination reactions. (This is important because the
complexity of the ABO system is considerably greater than we are discussing here, and many
other relevant blood group systems exist as well.)
The primary cause of ABO mismatched transfusion reactions results from destruction of
donor red cells by the recipient's antibodies; the reaction between donor antibodies and
recipient cells is of less importance, since the small amount of antibody contained in the
transfusion is generally diluted to harmless levels in the recipient. As a result, a type O
individual (whose red cells bear neither antigen) is classically referred to as a universal
donor, since his blood can fairly safely be given to a recipient with any ABO type;
conversely, a person of blood type AB has been called a universal recipient, since his blood
will contain no AB antibodies to damage any transfused red cells. Modern terminology more
accurately refers to type O as a universal red cell donor, and to type AB as a universal
plasma donor In general, however, every effort is made to assure complete compatibility

78

between donor and recipient ABO blood groups (as well as the Rh type which we will
discuss below) provided that an appropriate donor is available.

The safety of blood transfusions which are not perfectly matched can be enhanced by
transfusing packed red blood cells only, without the antibody-containing plasma; type O red
blood cells are safe when given to any recipient, and the potential damage done by the donor's
anti-A and anti-B antibodies may be avoided in this manner. But two relevant points should
be noted. First, transfusions of whole human blood are actually now fairly uncommon, and
the vast majority of transfusions involve the use of separated blood components such as
packed RBCs, plasma, platelets, leukocytes or purified plasma proteins. Second, perfect
matching of donor and recipient for ABO and Rh, even for transfusion of packed RBCs, is
still the norm, except in unusual or emergency situations.
Structure of ABO antigens; "natural antibodies?"
Why are there so-called "natural" antibodies to A and B blood group antigens? A description
of the nature and distribution of these antigens will help answer this question. The blood
group substances A and B represent two modified forms of a "stem" carbohydrate present on
red blood cells and other tissues. Their structures are shown below (where GLU is
glucosamine, GAL is galactose or galactosamine, FUC is fucose, and NAc represents an Nacetyl group):
[RBC]---O- GLU - GAL
|
|
NAc FUC

"stem" carbohydrate (="H" antigen)

[RBC]---O- GLU - GAL - GAL

|
|
NAc FUC

substance B

[RBC]---O- GLU - GAL - GAL

|
|
|
NAc FUC NAc

substance A

These same carbohydrates are also a common component of many foods we eat and many
microorganisms in our intestinal tract. The immune system is therefore constantly exposed
to these antigens, and responds by making an effective humoral response. Since the immune
system does not in general respond to antigens which are a normal part of "self" (see Chapter
18, TOLERANCE), a type B individual does not make antibodies to the B blood group
substance, although the response to the type A antigens is robust. The net result is the
production of antibodies, mostly of the IgM class, to whichever of these substances is not
present on an individuals red blood cells. (It should be noted that such complex
carbohydrates are a typical example of thymus-independent antigens, which generally elicit
only IgM antibodies, discussed in Chapters 13 and 15.)
It is important to remember that the A and B blood group substances are present not only on
red blood cells, but also in virtually every other tissue. They are therefore important
transplantation antigens and must be taken into account together with HLA tissue-typing
(see Chapter 11) when organ transplantation is performed. This is particularly important

79

given the fact that substantial levels of anti-A and anti-B antibodies (isoagglutinins) may be
normally present, depending on the recipient's blood type.
Genetics of ABO
The presence of A and B carbohydrates in our tissues is determined by three alleles at a single
genetic locus. One allele encodes an enzyme which produces the A substance, another the B
substance; and when both of these alleles are present in a heterozygote both carbohydrates are
made. The third allele, O, behaves essentially as a "null" allele, producing neither A nor B
substance. Thus, while the ABO system yields only four blood types (phenotypes), there are
six possible genotypes:

Genotype

Blood Type
(Phenotype)

A/A
A/O
B/B
B/O
A/B
O/O

A
A
B
B
AB
O

Only a single genotype can produce the phenotype AB, namely the heterozygous state A/B.
Likewise, type O individuals must be homozygous O/O. However, type A or type B
individuals can be either homozygous or heterozygous, the O allele being effectively
recessive since it does not contribute either of the two antigens.
The inheritance of the ABO blood groups follows simple Mendelian rules. For instance, a
homozygous type A mother and a type AB father (below, left) can yield only two kinds of
offspring, type A (genotype A/A) or type AB (genotype A/B). A heterozygous type A and a
heterozygous type B, on the other hand (below, right), can yield four genotypes and four
corresponding phenotypes.
A/A X

A/B

A/A
0.50

A/B
0.50

A/O X
A/B
0.25

A/O
0.25

B/O
B/O
0.25

parents
O/O
0.25

offspring
frequency

Rh BLOOD GROUPS
Genetics
While many blood group systems are known other than the ABO system, the Rh system is of
special importance,. This was originally defined by a rabbit antibody directed against the red
blood cells of Rhesus monkeys, an antibody which turned out to be capable of distinguishing
between the red blood cells of different human individuals. In simple terms, this system is
defined by the presence or absence of a single red blood cell antigen, representing the two
blood types Rh+ and Rh-. These are determined by two alleles at a single locus, which
segregate independently of the ABO blood group locus. Thus an Rh+ individual may be
homozygous (+/+) or heterozygous (+/-), while an Rh- individual must be homozygous (-/-).
80

The Rh- blood type is relatively uncommon, representing less than 15% of the population.
Since Rh segregates independently of ABO, one can readily calculate the frequency of any
given combination of ABO type and Rh type. If type A represents 40% of the population,
and Rh- only 15%, then the frequency of type A, Rh- individuals is given by:
0.40 x 0.15 = 0.06
or ~6% of the population.
So-called "natural" antibodies to Rh do not exist in humans, as they do for the AB antigens.
However, Rh+ cells infused into an Rh negative recipient can give rise to a strong antibody
response, mainly of the IgG class, which can result in dangerous reactions to subsequent
transfusions.
Blood typing and cross-matching are therefore important to ensure
compatibility for the Rh factor as well as ABO. However, unlike the A and B antigens, the
Rh antigens are present only on red blood cells. Therefore, while they are important for
blood transfusion, they do not normally play a role in organ transplantation, and Rh typing
of organ donors and recipients therefore not a significant consideration.
Rh-incompatibility; RhoGAM Therapy
The Rh factor assumes a special importance in maternal-fetal interactions. A mother who is
Rh- can bear an Rh+ child if the father is Rh+ (either homozygous or heterozygous). Since
there are no natural anti-Rh antibodies, this generally poses no special risk for the first
pregnancy. At the time of birth, however, tissue damage resulting from the separation of the
placenta from the uterine wall can result in a significant amount of fetal blood entering the
maternal circulation; which may stimulate a strong IgG anti-Rh response in the mother.
If the same mother then bears a second Rh+ child, the existing anti-Rh antibodies can cross
the placenta during the pregnancy and destroy fetal red blood cells. The ensuing damage to
various organs results in the potentially dangerous condition Erythroblastosis Fetalis (also
known as Hemolytic Disease of the Newborn, or HDN). This can be diagnosed prenatally by
carrying out amniocentesis, and examining the amniotic fluid for the presence of free
hemoglobin and its degradation products. Various approaches can be used during and after
birth to rescue the infant, including exchange transfusion, complete replacement of the
infant's blood to remove the anti-Rh antibodies and provide undamaged red blood cells.
However, the production of anti-Rh antibodies in an Rh- mother can often be prevented by
administering anti-Rh immune globulin (e.g. RhoGAM) into the mother, typically at around
28 weeks of gestation and again within 72 hours of the birth of her Rh+ baby. By
mechanisms which are still not fully understood, these antibodies greatly reduce the
likelihood of sensitization of the mother's immune system by the Rh+ erythrocytes. If this
procedure, developed in the 1960s, is successfully carried out during each Rh+ pregnancy,
anti-Rh antibodies are not produced by the mother, and subsequent pregnancies will not be at
risk.
While Rh incompatibility is of considerable clinical significance, it should be noted that not
all untreated incompatible pregnancies result in disease. Only a small fraction of

81

incompatible pregnancies actually result in the production of maternal anti-Rh antibodies,

and in only a fraction of these cases is there significant damage to the newborn.
What about ABO incompatibility?
If maternal antibodies to red blood cells can damage the developing fetus, why is
incompatibility for ABO blood groups not as dangerous as Rh-incompatibility, particularly
since ABO isoagglutinins normally exist in mothers which could potentially damage the
infant even during a first pregnancy? The answer lies in the isotype of antibody produced in
the two cases. Anti-Rh-antibodies are mainly IgG (typical of anti-peptide responses) which is
capable of crossing the placenta and entering the fetal circulation. The natural antibodies
(isoagglutinins) to A and B blood group substances, however, are mostly of the IgM class
(typical of anti-carbohydrate responses) and therefore do not cross the placenta.
IgG antibodies to the A and B blood group antigens may develop in some individuals, and the
resulting ABO incompatibility actually accounts for about two thirds of all discernable cases
of HDN; such cases, however, are generally very mild and require little or no treatment.
Thus, while ABO incompatibility is actually much more common than Rh incompatibility, it
is much less likely to cause significant d
isease. In fact, it appears that the presence of ABO incompatibility between mother and fetus
may confer some protection against the development of clinical Rh incompatibility.

CHAPTER 10, STUDY QUESTIONS:

1.

How many potential genetic crosses exist in which parents with identical ABO
phenotypes can produce children who differ from one another in their phenotypes?

2.

Is the Rh antigen a T-DEPENDENT or T-INDEPENDENT ANTIGEN? (See

Chapter 15).

3.

Maternal-fetal ABO incompatibility, which is much more common than Rh

incompatibility, may protect against the development of anti-Rh antibodies in Rhincompatible pregnancies. Suggest an explanation for this phenomenon.

82

CHAPTER 11
CELL-MEDIATED IMMUNITY AND MHC
See APPENDIX (10) INBREEDING; (11) MIXED LYMPHOCYTE REACTION (MLR)
Graft rejection is a manifestation of CELL-MEDIATED IMMUNITY (CMI), as
can be demonstrated by adoptive transfer experiments; antibodies generally contribute
little to the rejection of grafted tissues. Simple rules predict the acceptance or rejection of
alloografts between inbred mice. Although these specific rules are not valid in humans
(or other outbred populations), a common fundamental principle holds, namely, that a
graft will be rejected if the host's immune system recognizes a foreign antigen on the
graft. The most important antigens contributing to graft rejection in vertebrates are those
encoded by the MAJOR HISTOCOMPATIBILITY COMPLEX (MHC), a cluster of
genes known as H-2 in mice and HLA in humans; within this complex are encoded
several CLASS I and CLASS II molecules, with distinct structures and biological
functions. The basic rules of immune recognition and reactivity are illustrated by an
analysis of GRAFT-VERSUS HOST (GvH) reactions.
TRANSPLANT REJECTION: CELL-MEDIATED IMMUNITY
The greatest barrier to our ability to transplant tissues and organs between individuals is
rejection by the immune system: grafts between genetically different individuals are generally
rejected. Using skin grafts between mice as a model system, lets examine the basic
immunological and genetic features of such a rejection reaction.
If we graft a patch of skin from one mouse of strain Y onto another mouse of strain X (an
allograft), the graft will heal and may remain viable for a week or more, but at the end of
about 15 days it will have been rejected -- only a bit of scar tissue will remain at the site (as
indicated in Figure 11-1 below). This is known as first set rejection (i.e. primary immune
response). A simultaneous graft from another strain X mouse (an isograft or syngeneic graft,
not shown in the figure) will heal and remain viable and intact indefinitely.
STRAIN "Y" SKIN
"Y" SKIN
REJECTED
IN 15 DAYS

X
STRAIN "X" MOUSE

BOTH REJECTED:
"Y" IN 5 DAYS,
"Z" IN 15 DAYS

SAME MOUSE,
TWO NEW GRAFTS

FIRST-SET vs. SECOND-SET GRAFT REJECTION:

IMMUNOLOGICAL MEMORY IN CELL-MEDIATED IMMUNITY

Figure 11-1

83

We then take the same mouse which has rejected the first graft, and place on it two more
grafts, one from strain Y and another from a third, unrelated mouse strain Z. The Z graft is
rejected as was the original Y graft, in about 15 days. The new Y graft, however, is rejected
in five days, much more rapidly than the first time. This is known as second set rejection
(i.e. secondary immune response).
Graft rejection is an immunological phenomenon, defined by the following properties.
i)

It shows specificity. Graft Y (an allograft) is rejected while the isograft X is not
effected.

ii) It displays immunological memory. Graft Y is rejected in "second set" fashion the
second time around, while the simultaneous grafting of skin from strain Z is rejected in
"first set" fashion (this is also a reflection of immunological specificity).
iii) It is systemic. While rejection occurs in a particular location (at the site of the grafted
skin in this example), the ability to reject a graft is not localized, either in first set or
second set tempo. The first or second graft of Y skin could have been placed anywhere
on the recipient (apart from certain immunologically "privileged" sites; see Chapter
18), and the result would have been the same. The systemic nature of immunological
responsiveness is a consequence of the continuous and rapid movement of cells of the
immune system throughout the body, which we will discuss more fully in Chapter 16.
GRAFT REJECTION: MANIFESTATION OF CELL-MEDIATED IMMUNITY
The above experiment showed us that graft rejection is an immunological process, and we
can now ask whether it is a manifestation of HUMORAL or CELL-MEDIATED immunity
(i.e. is it mediated by antibodies or not). To answer this question we carry out an
ADOPTIVE TRANSFER experiment, transferring either serum or spleen cells from a strain
X mouse which has already rejected a Y skin graft, into two untreated strain X recipients.
We then graft each of the two recipients with strain Y skin, and observe how long it takes for
the grafts to be rejected. These are "first set" grafts, and we would normally expect them to
be rejected in about 15 days.
REJECTED "Y" SKIN

STRAIN "X" MOUSE

IMMUNE TO "Y" SKIN

REJECT
15 DAYS

NAIVE RECIPIENTS

SERUM
SPLEEN
CELLS

ADOPTIVE TRANSFER OF CMI

Figure 11-2

84

REJECT
5 DAYS

We find, however, that the mouse treated with immune spleen cells rejects the graft in
"second set" fashion (in about five days) while the one treated with immune serum rejects it
in the expected "first set" fashion (fifteen days). Thus, we find that immunity to the skin graft
has been transferred by spleen cells but not by serum. This defines graft rejection as a
CELLULAR immune response, or a manifestation of cell-mediated immunity (CMI).
(If we had been studying a different kind of immune response, namely protective immunity to
Pneumococcus in mice, we would have gotten a different result [See Fig. 2-1]. In that case
immunity would have been transferred by either immune spleen cells or immune serum,
which would define it as HUMORAL immunity. Humoral and cellular immunity are both
transferable by spleen cells, but only humoral immunity is transferable by serum or other
antibody-containing fluids.)
With some exceptions ("hyperimmune rejection" of kidney grafts, for instance) circulating
antibody contributes little to rejection of grafted tissues. In some instances, in fact, the
presence of antibody can protect a graft from rejection, a phenomenon known as
immunological enhancement (and such antibodies are referred to as "enhancing"
antibodies). The mechanism of this phenomenon is complex, and may involve "hiding" of
histocompatibility antigens from the host's cellular response when they are bound with
antibody.
GENETIC RULES OF TRANSPLANTATION
Let's examine (Figure 11-3) the pattern of rejection or acceptance of skin grafts between three
inbred strains of mice, one of strain C57Bl/6, another of strain DBA/2, and the third an F1
hybrid between these two inbred strains. The arrows indicate the direction of the graft; an
"A" indicates the acceptance of a particular graft while an "R" indicates rejection. (NOTE:
The rules shown in this figure hold for inbred mice, but not for outbred organisms, including
humans.)

R
DBA/2

C57Bl/6

R
R

A
"A"=accepted
"R"=rejected

(DBAxC57)F 1

TISSUE GRAFT REJECTION IN INBRED MICE

Figure 11-3
The key features of immunological graft rejection are summarized below:
1) A graft is rejected if a foreign ("non-self") antigen is recognized. Grafts between two
unrelated inbred strains, or between two individuals of a non-inbred species, will be
rejected.

85

2) Grafts from an F1 hybrid onto a parental inbred strain are rejected. The antigens of
both parental strains are expressed on the F1, and each will be seen as foreign by the
other parental strain.
3) Grafts from either inbred parent strain onto an F1 will be accepted. No foreign
antigen can be recognized in such a graft.
4) The antigens triggering graft rejection are cell surface glycoproteins. Most of these
are present on all tissues of an organism, and are known as HISTOCOMPATIBILITY
ANTIGENS.
5) There are many genes which encode histocompatibility antigens (more than 30 named
genes in mice). Any gene encoding a polymorphic protein is potentially a
histocompatibility locus. (NOTE: Only polymorphic loci can contribute to allograft
rejection, since a monomorphic gene product will not be foreign to any individual of
the species.)
6) In any given vertebrate species, one single histocompatibility "gene" predominates in
promoting graft rejection.
This "gene" is known generally as the MAJOR
HISTOCOMPATIBILITY COMPLEX (MHC), and has different specific names in
different species.
HUMAN

HLA (Human Leukocyte Antigen)

MOUSE

H-2 (Histocompatibility locus # 2)

The MHC is, in fact, not a single gene, but is actually a closely linked complex of many genes,
dozens or more--hence its name. These genes are highly polymorphic in most species,
accounting for the fact that allografts in humans (and other non-inbred species) are almost
always rejected.
The many histocompatibility genes outside of the MHC are collectively known as minor
histocompatibility loci. Differences at the MHC will always cause rapid graft rejection,
regardless of the status of minor loci. Differences at minor loci, however, even many of them
simultaneously, will not cause as rapid rejection as a difference in the MHC itself.
MAJOR HISTOCOMPATIBILITY COMPLEX (MHC) OF MICE AND HUMANS
A simplified diagram of the H-2 complex of mice compared with the HLA complex of
humans is shown below:

Class:

H-2
HLA
Class:

II

II

Figure 11-4

86

Three regions can be distinguished in the H-2 complex, each of which is itself a complex of
multiple gene loci:
The K and D regions encode Class I molecules. Class I molecules of the MHC are
expressed on all mouse cells and the major targets for recognition and rejection of foreign
grafts. Skin grafts between strains of mice differing only in the I-region (Class II) will be
rejected only slowly, if at all, while differences in K or D will result in rapid rejection. In
humans, Class II differences play a larger role in inducing rejection, which will be discussed
later.
The I-region encodes Class II molecules. Class II molecules of the MHC are expressed
only on some cells, and are required for the process of antigen presentation to "helper" T-cell
(discussed in Chapters 12 and 15). In mice these antigens are referred to as Ia antigens (for
"I-region Antigens"), a term sometimes loosley applied to human Class II molecules.
The I region of mice was shown many years ago to contain genes which control the ability to
generate immune responses to particular antigens, genes were called "Ir genes" or "Immune
Response Genes;" this is the feature of the I-region that originally gave it its name. It was
subsequently shown that these "Ir" genes are the Class II genes themselves, and the
mechanism by which they regulate immune responsiveness involves their participation in the
process of antigen presentation (see Chapters 12 & 15).
The HLA complex of humans is organized slightly differently, but in a homologous fashion:
The A, B and C regions of HLA encode Class I molecules, which are expressed on
all nucleated cells. While Class I molecules are the major targets for graft rejection,
Class II molecules can play a much more important role in human graft rejection than
they do in mice. Tissue typing for Class I antigens in humans is traditionally carried
out by using antibodies specific for the various allelic forms; these antigens are
therefore referred to as Serologically Determined, or SD antigens.
The D-region of HLA encodes Class II molecules (which are expressed only on some
cells), and includes several defined subregions, namely DP, DQ and DR (as seen
below). These genes are capable of regulating certain kinds of immune responses
(typically autoimmune conditions including "rheumatic" diseases), in a manner
analogous to the I-region of mouse H-2. Antigens of the D-region of HLA are
traditionally not detected by antibodies (although there are now exceptions), but are
determined using the MIXED LYMPHOCYTE REACTION (MLR) (see APPENDIX
11). D-region antigens are therefore referred to as Lymphocyte-Determined (LD)
antigens, to distinguish them from the SD antigens of the A, B and C regions.
Shown below in Fig. 11-5 is a more detailed diagram of H-2 and HLA, showing a few
additional features of each, as well as some of the relationships between the two:

87

class:

H-2

II

[III]

I-A, I-E

[S]

[C2, C4, B]

DP, DQ, DR

HLA
class:

II

"type":

LD

SD

SD

SD

MAJOR HISTOCOMPATIBILITY COMPLEX OF MOUSE AND HUMAN

Figure 11-5
Several new features are evident:
The D-region of H-2 is seen as containing two sub-regions, D and L, both of which
encode Class I antigens. The D-region of HLA, which encodes Class II antigens, can
ce seen to contain three subregions, namely DP, DQ, and DR, mentioned above.
An additional region is shown in both H-2 and HLA which codes for the C2, C4 and B
components of complement (this region is called the S-region in mouse). These genes
are also referred to as Class III genes, although the reason for these complement
component genes (and several other unrelated genes) being located in the middle of the
MHC remains a puzzle.
While there are obviously differences in the organization of the MHC of mouse and
man, their common features are the most important. They both contain genes
determining Class I and Class II antigens; they are both highly polymorphic; they both
encode glycoproteins which are the major targets for graft recognition and rejection
(Class I) as well as mediators of cell cooperation in immune responses (Class II).
GRAFT-VERSUS-HOST REACTION (GvH)
The rules of transplantation as outlined above (see Fig. 11-3) tell us that a skin graft from a
C57Bl/6 mouse onto a (C57Bl/6xDBA/2)F1 recipient will be accepted indefinitely, with no
resulting harm to either graft or host. However, if we transfer spleen cells in this same
genetic combination, we see something quite different--the recipient will rapidly sicken and
die. This phenomenon is the result of an immune reaction carried out by the grafted cells
against histocompatibility antigens present in the recipient, and is known as a Graft-versusHost (GvH) reaction. The GvH reaction is of great importance in many human tissue and
organ grafts, and we need to understand the principles underlying its production.
In order for a GvH reaction to occur after tissue or organ transplantation, the following three
conditions must hold:
1) The grafted tissue must contain immunocompetent cells. Grafted skin cannot cause
GvH simply because it does not contain cells capable of carrying out immune reactions.
Spleen, however, does contain appreciable numbers of immunocompetent cells, as does
lymph node, and both can therefore initiate a GvH reaction in an experimental situation.
While neither spleen nor lymph node cells would normally be transferred between
humans, transfer of other lymphocyte-containing tissues such as blood or bone marrow

88

may be capable of initiating GvH reactions. In addition, some organ grafts may contain
sufficient numbers of lymphocytes to initiate GvH reactions, notably liver, intestine and
lung.
2) The graft must be capable of recognizing foreign antigens on host tissue. Spleen
cells from an F1 mouse injected into a parental strain recipient, for instance, will not give
rise to a GvH, as the host tissues bear no antigen which is foreign to the F1 graft. In this
example, in fact, the grafted spleen cells would be rapidly destroyed (since these cells do
bear antigens foreign to the hosts immune system). In the human situation, only a graft
between identical twins would fail to be able to recognize any foreign antigens in the
host.
3) The recipient must be incapable of rejecting the grafted tissue. In the mouse example
cited above (C57Bl/6 spleen cells into an F1 recipient), the host cannot reject the
transplanted spleen cells because no foreign antigens are recognized (this is the same
reason it cannot reject a skin graft in that genetic combination). However, other reasons
may also exist for failure to reject the graft. For example, injection of C57Bl/6 spleen
cells into a newborn DBA/2 mouse will result in fatal GvH. In this instance the newborn
mouse cannot reject the graft because its immune system is not sufficiently developed
such a graft would not result in a GvH if given to an adult DBA/2 mouse because it
would be rapidly rejected. In humans, however, the immune system of a newborn is
perfectly capable of rejecting a foreign graft (see Chapter 17). A third example is that
where the host immune system is defective due to a congenital immunodeficiency or
immunosuppression, in either an experimental or clinical setting. If an adult DBA/2
mouse is subjected to sublethal X-irradiation, or treatment with anti-lymphocyte serum (a
powerful immunosuppressant), injecting C57Bl/6 spleen cells will result in a GvH
reaction. Likewise, transfusion of whole blood into a human infant suffering from a
congenital T-cell immunodeficiency may result in a fatal GvH [see Chapter 20]).
CLINICAL SIGNIFICANCE OF GvH
GvH reactions have been very useful to immunologists studying the genetics and mechanisms
of the cellular immune response, but they have also been of considerable clinical
significance. A classic example is that of bone marrow/stem cell transplantation which
has long been potentially dangerous due to the presence of immunocompetent cells that can
carry out a GvH reaction. Methods for removing such cells prior to transplantation and for
controlling the resulting GvH have become much more sophisticated and effective in recent
years, but stem-cell transplantation remains a fairly high-risk procedure.
Another example is that of transfusions of whole blood, which, of course, contain
immunocompetent T-cells. Most commonly, such transfusions pose little danger of
producing GvH disease, since the recipients immune system will recognize and reject the
foreign T-cells (Rule 3, above, does not hold).
However, if the recipient is
immunodeficient, as in the example above, a fatal GvH reaction may ensue from what would
otherwise be a harmless transfusion. In such cases the risk, if it is recognized, can be
eliminated by X-irradiation of the transfused blood, which eliminates the lymphocytes
without damaging red blood cells or other important blood components.
Whole blood transfusion may also pose a significant risk for GvH if they happen to be from
an HLA-homozygous donor into a recipient heterozygous for the same haplotype as the
89

donor. This is ordinarily a rare occurrence in most populations which are highly polymorphic
for HLA, but occurs more frequently in some populations with more restricted heterogeneity,
particularly in the case of transfusions between first-degree relatives (parent/offspring, and
siblings).

CHAPTER 11, STUDY QUESTIONS:

1.

Describe the cellular basis for the difference between "first set" and "second set"
GRAFT REJECTION?

2.

Why do the simple RULES OF TRANSPLANTATION between inbred mice

(Fig. 11-3) not apply to humans? What rules do apply to humans?

3.

How many human chromosomes carry genes encoding HISTOCOMPATIBILITY

ANTIGENS?

4.

What are the differences between CLASS I and CLASS II MHC molecules?

5.

Draw a diagram showing the rules for the ability of cell transfers between inbred
strains to generate GvH REACTIONS (analogous to Figure 11-3).

90

CHAPTER 12
CELL INTERACTIONS IN CELL MEDIATED IMMUNITY
SEE APPENDIX (11) MIXED LYMPHOCYTE REACTION
The diverse manifestations of CELL-MEDIATED IMMUNITY (CMI) include
several mechanisms by which target cells may be killed. The induction of one of these
mechanisms, namely killing by CYTOTOXIC T-CELLS, involves the participation of
two distinct, antigenically precommitted lymphocytes (TH- and TC -cellCELLS) as well
as non-specific antigen-presenting cells (APCs, typified by DENDRITIC CELLS and
MACROPHAGES). TH cells can be subdivided into two major sub-types, TH1 cells
which promote inflammatory reactions and CMI, and TH2 cells which largely provide
"help" to B-cells in generating humoral immunity.
T-cell receptors recognize antigenic peptides only when they are associated with
MHC molecules, which accounts for the phenomenon of MHC-RESTRICTED
RECOGNITION. The existence of two DISTINCT CELLULAR PATHWAYS by which
CLASS I and CLASS II molecules become associated with antigenic peptides is
described. Resistance to the intracellular bacterium LISTERIA serves to illustrate a
distinct mode of lymphoid cell interactions and cell-mediated effector functions which
does not depend on TC cells.
All those immune reactions which can not be transferred by serum or other antibodycontaining fluids are referred to as CELL MEDIATED immunity, or CMI. CMI is initiated
by various kinds of T-cells, and both T-cells and other cell types (notably macrophages) may
act as effector cells in such reactions. Because of its characteristic slow time course, CMI is
also referred to as DELAYED-TYPE HYPERSENSITIVITY or DTH (although this term is
often used to refer specifically to one particular type of skin reaction described below). In
Chapter 14 we will discuss some of the characteristic surface markers of known T-cell
subsets. But remember that all T-cells (like B-cells, but unlike macrophages or any other
cells) are clonally precommitted, and have antigen-specific receptors in their
membranes. Remember also that T and B-cells use different molecules for antigenrecognition.
Examples of CMI Reactions
1) Graft rejection. The major barrier to successful tissue and
organ transplantation is immune rejection. Antibodies are
not the primary mediators of such rejection, although they
can be involved in various ways (as we will see later).

Effector T-cells
TH1/TC

2) Resistance to certain infectious diseases. While humoral

immunity is important for protection against many bacterial
infections, immunity to certain pathogens, notably fungi and
intracellular parasites (including viruses and intracellular
bacteria) depends primarily on the cellular immune system.

TH1/Tc

3) DTH skin reaction. The tuberculin skin test is often used as

the classic example of delayed-type hypersensitivity reactions
or cell-mediated immunity. It is one manifestation of the mechanism
of graft rejection and immune resistance to intracellular parasites.

91

TH1

4) Graft-versus-host reaction. GvH has been discussed in

Chapter 11, and is of both experimental and clinical interest.

TH1/TC

5) Contact dermatitis. Sensitivity to poison oak, certain drugs, and

TH1/TC
other environmental substances (e.g., cosmetics) involve CMI skin
reactions against a reactive hapten coupled to an endogenous carrier.
While the carrier is normally regarded by the immune system as"self,
the conjugate may highly immunogenic. One way to distinguish contact
dermatitis from IgE-mediated allergy ("immediatehypersensitivity") is
by comparing the time course of development of the reactions.
6) Mixed lymphocyte Reaction (MLR). This in vitro manifestation
of cell-mediated immunity has been of considerable importance in
MHC tissue-typing in preparation for transplantation, and is described
in more detail in APPENDIX 11.

TH1

(NOTE: The more recently defined TH17 cell is discussed in Chap. 14.)
THREE MECHANISMS FOR CELL-MEDIATED TARGET CELL KILLING
1)

TC -mediated lysis (which is the principle topic of this chapter)

Tc
Tc

Tc cell recognizes
target antigen

Target cell killed

Figure 12-1
Activated TC cells specific for a cell-surface antigen, bind to the target and cause lysis, as
seen in Figure 12-1 (these cells are Ag-specific, clonally committed T-cells). This binding
requires ASSOCIATED RECOGNITION of the foreign antigen together with self class I
MHC antigens; this phenomenon is discussed in more detail later in this chapter.
2)

NK-killing

NK
NK

NK cell binds to
target cell

Target cell killed

Figure 12-2

92

Another cell which can kill target cells is the NK or "Natural Killer" cell. NK cells were first
detected using in vitro assays, through their ability to kill certain tumor cell targets
(schematized in Figure 12-2). Unlike TC cells, which utilize antigen-specific receptors to
recognize I MHC Class I-associated peptides on target cells, NK cells kill targets based on
their expression of abnormally low levels of MHC-I molecules (regardless of the particular
peptides associated with them), or abnormal carbohydrates. Tumor cells which underexpress MHC molecules, or virus-infected cells which may down-regulate MHC-I expression
(thereby avoiding TC-based killing) may thus become targets for NK killing.
NK cells are not induced by prior exposure to the tumor cells, but are constitutively present.
They are members of a population of white blood cells known as Large Granular
Lymphocytes (LGL), originally named "null" lymphocytes due to their lack of the membrane
markers (notably Ig and TCR/CD3) which defined B-cells and T-cells.
3)

Antibody Dependent Cell-mediated Cytoxicity (ADCC).

PMN

NK

NK

Ab binds to
target antigen

Killer cells bind

via Fc receptors

Target cell killed

Figure 12-3
NK cells can also kill antibody-coated targets by a different mechanism deriving from the
fact that NK cells bear Fc receptors, resulting in a mode of killing which is known as ADCC
(Figure 12-3). Macrophages and neutrophils (as indicated in the figure) also bear Fc
receptors and may also mediate this type of target cell killing. ADCC is a striking example of
the complex relationships between innate and adaptive immunity, which in this case results in
target cells being killed in a specific fashion by cells which are inherently non-specific.
In the remainder of this chapter we will discuss in more detail the basis of Tc-mediated
target cell lysis, the first of these three mechanisms, illustrated in Figure 12-1.
ASSOCIATED RECOGNITION IN T-CELL KILLING:
MHC-RESTRICTED RECOGNITION
One of the most striking and important findings to emerge from half a century of research in
cellular immunology is that cytotoxic T-cells, in order to kill their targets, must
simultaneously recognize not only a foreign antigen, but also self-MHC antigens. We
can illustrate this remarkable phenomenon with a diagram of the original experiment carried
out by Doherty and Zinkernagel using mice infected with the LCM (lymphocytic
choriomeningitis) virus, as shown in Figure 12-4.

93

Infected mice develop cell-mediated immunity which can be detected by an in vitro assay.
Spleen cells from such immune mice can kill LCM-infected target cells in culture; they are
clearly LCM-specific, since they will not kill uninfected cells; this specificity is illustrated by
the results shown in the first two lines of results in the figure.
Infect with LCMV,
wait ten days
Remove spleen, test for
target killing in vitro
by spleen Tc cells

BALB/c

Target Cells

LCMV
infected?

H-2

Target Cells
killed?

BALB/c

NO

BALB/c

YES

(BALB/c X C57Bl/6)F1

YES

d/b

(BALB/c X AKR)F1

YES

d/k

C57Bl/6

YES

Tc must see
viral antigens
Tc must also see
self H-2 antigens

MHC-RESTRICTED RECOGNITION BY KILLER T-CELLS

Figure 12-4
But the specificity for LCMV obviously doesn't tell the whole story, since the immune
BALB/c spleen cells will not kill cells derived from a different strain of mice, even if they are
LCMV-infected in this example they will kill only BALB/c or F1 targets (shown in lines 35 in the figure). This was the first demonstration that cytotoxic T-cells, in order to kill their
targets, must recognize LCMV antigens on the target cell in association with "self" MHC
molecules. More precisely, they must recognize the viral antigens associated with self Class I
molecules of the MHC complex (namely K, D or L-molecules of mouse H-2, or A, B or C
molecules of human HLA). This phenomenon is known as MHC-Restriction or MHCAssociated Recognition.
CELL COOPERATION IN CELL MEDIATED IMMUNITY
MHC-associated recognition in target cell killing is part of a more general phenomenon
which reflects the mechanism by which the T-cell receptors of all T-cells recognize their
specific antigens. The diagram on the next page (Figure 12-5) illustrates the central features
of the induction of CMI, including MHC-associated recognition.
The induction of CMI requires the cooperation of three cell types, an effector cell (in this
case a cytotoxic T-cell, TCcell), a helper cell (TH1) and an accessory cell (e.g. dendritic cell,
macrophage or other antigen-presenting cell). Antigen (a viral peptide in this example) must
also participate, illustrated by the solid triangles. The final result of the cell/cell interactions
and differentiation events described below is the killing of a virus-infected target cell.

94

"ASSOCIATED RECOGNITION" IN CELL-MEDIATED IMMUNITY

Cell cooperation in the generation of effector T-cells
directed against viral antigens.

ANTIGEN PRESENTING CELL

II

II
4

II

7
28

28

TD

(TH1 )

VIRUS-INFECTED
CELL

7
28

T H2

T H1
TC

Stimulation
phase

IL-4, etc.
activation of
macrophages:
IFN-, MIF/MCF, etc.

IL-2
(B-cell)

TC

Effector
phase

= viral antigen

= CD4

= CD8

Free
Virus

= B7/CD28
= MHC Class I
= MHC Class II

TC

no
recognition

28

"killer"

TC

= TcR
8

= IL-2 recep.

"precursor"

= proliferation.

Target Cell
Recognition

Target Cell
Death

Figure 12-5

95

Some key features of these interactions are as follows:

The antigen-presenting cell ("accessory cell") must have acquired the foreign antigen (an
LCMV peptide) by capturing it from the surrounding tissue fluid (by phagocytosis or
pinocytosis) and expresses it on its cell surface in association with MHC Class II
molecules.

A helper T-cell, the TH1 cell, clonally precommitted to an LCMV antigen, binds to the
accessory cell. Its TCRs must recognize LCMV antigens together with MHC Class II
molecules, and this interaction is therefore described as being Class II restricted. The
TH1 cell bound in this manner now also recognizes and binds the B7 molecules on the
antigen-presenting cell with its own CD28 membrane molecules. Specific binding of
both these T-cell membrane receptors (TCR and CD28) results in this TH-cell being
triggered to secrete the cytokine IL-2 (Interleukin-2). among others.

Meanwhile, a potential target cell must have been infected with the virus, and expresses
LCMV antigens in association with MHC Class I molecules.

A clonally precommitted TC cell recognizes LCMV antigen on the target cell by virtue of
its antigen-specific receptors (TCR), but in this case complexed with MHC Class I; this
interaction is therefore described as being Class I-restricted. As a result of this
interaction, the TC cell is stimulated to express receptors on its surface for the cytokine
IL-2.

IL-2 produced by the TH1 cell is bound by the nearby TC cell (now expressing IL-2
receptors), which is thereby triggered into its final stage of differentiation, becoming an
effective killer cell. (This TC cell, on being triggered to express the IL-2 receptor, may
also itself secrete some IL-2 which can bind to and activate its own receptors in an
autocrine process). This TC can then bind to an LCMV-infected target cell, recognizing
LCMV antigens together with Class I molecules, and causes the lysis of the target. This
represents a second Class I restricted interaction.

The 2-Signal Principle. These phenomena illustrate the general principle that all T-cells
require two distinct molecular signals in order to be triggered to differentiate and
proliferate, and ultimately carry out their effector roles in immune responses; one signal
is antigen-specific, the other is not. In the case of TH cells, the first signal is produced by
occupancy of their TcRs (when they bind to their specific antigen together with MHC
Class II on the APC), and the second ("costimulatory") signal is given by CD28 binding
to B7. In the case of TC cells, the first signal is provided by occupancy of their TcRs by
antigen/MHC Class I, and the second signal is given by IL-2 binding to the newly
expressed IL-2-receptors. We will see later that B-cells also require two distinct
molecular signals for activation.

If the TH cell in the above example were a TH2-type cell (illustrated in dotted lines at the top
of Figure 12-5), its interaction with the Class II-presented antigen would result in this cells
producing IL-4 (among other cytokines), which can stimulate a B-cell to differentiate into an
antibody-producing cell; T-cell/B-cell cooperation in humoral responses will be discussed in
Chapter 15.
Also illustrated in this diagram is the interaction between the accessory cell and another TH1
cell (labeled TD, in the upper left), but in this case the T-cell produces macrophage
chemotactic factors (MCF/MIF) as well as macrophage-activating factors such as IFN-.
This interaction similarly requires recognition of antigen together with self Class II (i.e. it is
Class II restricted). The production of this mix of cytokines leads to other manifestations of
96

CMI such as the tuberculin skin reaction (a classic example of DTH), and is an important part
of the mechanism of resistance to infection by intracellular bacteria such as Listeria
(discussed below).
IMPORTANCE OF MHC-RESTRICTED RECOGNITION IN T-CELL TRIGGERING
T-helper cells, which are clonally precommitted, antigen-specific cells, recognize their
cognate antigens only in association with MHC Class II molecules, and therefore are
triggered to differentiate and produce relevant lymphokines only when they are presented
with antigen by an authentic antigen-processing cell (macrophage, dendritic cell, etc.). The
importance of this phenomenon may lie in controlling the location and nature of such
interactions, since not all cells bear Class II antigens. It would be pointless, for instance, for
a TH-cell to attempt cooperation with a muscle cell, even if there were antigen on its surface,
since the muscle cell would be incapable of triggering the T-cell with the required
"costimulatory" signals (e.g. by expressing B7). Such an interaction cannot be attempted,
however, since muscle cells lack Class II antigens.
T-cell killing, on the other hand, is restricted with respect to Class I antigens, which are
present on all nucleated cells. The significance of this restriction of TC responses can be
readily understood, since one of its effects is to assure that TC cells can only be triggered by
membrane-bound antigens. It would be counterproductive to permit TC cells to be triggered
by soluble antigen, say, in the example of an LCMV infection. First, inactivation of free
virus can be readily accomplished by antibody, and does not require the participation of TC
cells; in fact, a Tc cell cannot generally do any harm to a free viral particle. Second, such
stimulation would divert TC cells from their critical role in killing virus-infected cells, which
cannot be effectively dealt with by circulating antibody.
Although MHC-restricted recognition was first described in mice, precisely the same
phenomenon regulates immune responsiveness in humans. In this case, of course, Class I
antigens of the HLA complex are the A, B and C antigens, and Class II is represented by the
DP, DQ and DR antigens (sometimes also loosely but incorrectly referred to as Ia antigens,
by analogy with mouse H-2).
CYTOTOXIC T-CELLS KILL BY INDUCING APOPTOSIS
Cytotoxic T-cells kill their targets by inducing apoptosis (programmed cell death) using at
least two mechanisms, as illustrated in Figure 12-6. As we have already seen, the TC cell
binds to the target via its T-cell receptors, which recognizes MHC-Class I-associated
antigenic peptide. (The participation of CD8 is also illustrated here, but not a variety of other
membrane adhesion molecules which are also known to be important.) Subsequent binding
of the target cells membrane-associated Fas molecule by the T-cells Fas-ligand is one wellknown mechanism for inducing apoptosis. In addition, the T-cell releases the contents of
specialized granules which contain two proteins, perforin and granzyme. The perforin inserts
itself into the target cell membrane and forms large pores. The granzyme then enters the
target cell through these pores and initiates a cascade of signaling events which ends by
inducing the cell to undergo apoptosis. [NOTE: Under experimental conditions many target
cells can be killed by perforin alone, in a manner similar to the effect of the membrane-attack
complex (MAC) of complement. This simple lytic mechanism, however, appears not to be a
significant factor in T-cell killing in vivo.]

97

8
I

TC

FL F

apoptosis

Target
= Perforin
= Granzyme
F = Fas
FL = Fas Ligand

APOPTOSIS-DEPENDENT KILLING OF TARGET CELLS

BY CYTOTOXIC T-CELLS
Figure 12-6
CLASS I AND CLASS II ANTIGEN ASSOCIATION REPRESENT TWO DISTINCT
INTRACELLULAR PATHWAYS FOR ANTIGEN PROCESSING
As discussed above, antigens (or, more precisely, antigen-derived peptides) can be associated
with either Class I or Class II MHC molecules; these two forms of MHC association result in
different modes of T-cell recognition and different biological consequences. One must then
ask the question, how does a cell decide whether to express a particular peptide in
association with either Class I or Class II MHC molecules?
The general rule to remember is that endogenous peptides (i.e. those derived from proteins
synthesized within the cell) become associated with Class I MHC molecules, whereas
exogenous peptides (those taken up by phagocytosis or pinocytosis) become associated
with Class II MHC molecules. This implies that separate intracellular pathways exist for
the expression of the two classes of MHC molecules, and these are illustrated schematically
in Figure 12-7.
Exogenous proteins may enter an antigen-presenting cell either by phagocytosis or some other
form of endocytosis, and are incorporated into lysosomal vesicles in which they are degraded
to small peptides. These lysosomes fuse with other vesicles which have emigrated from the
ER and contain newly synthesized Class II molecules, and in the acidic environment created
by the lysosome the Class II molecules can bind the peptides in their specialized groove. The
small black dot shown in the new Class II molecules indicates the presence of the so-called
CLIP peptide, part of a system responsible for preventing the Class II molecule from
accepting a peptide before fusion of its vesicle with a lysosome, as well as for directing these
vesicles toward the lysosomal compartment.

98

lysosome

Exogenous
protein

MHC Class II
presentation

fusion
phago/pinocytosis,
RME

exocytosis

Class II
vesicle

Golgi/exocytosis

Endogenous
protein
proteasome

TAP

Class I
in ER

MHC Class I
presentation

SEPARATE INTRACELLULAR PATHWAYS FOR CLASS I AND

CLASS II PRESENTATION OF PEPTIDES
Figure 12-7
Endogenous proteins which are present in the cytoplasm (i.e. those synthesized within the
cell itself) may encounter a proteasome which cleaves them into smaller peptides and feeds
these peptides to the TAP transporter which pumps them into the lumen of the endoplasmic
reticulum. Once in the ER, these endogenous peptides can be bound by newly synthesized
MHC Class I molecules, and these complexes are transported via the normal secretory
pathway in vesicles which eventually fuse with the plasma membrane. Its important to
recognize that while our focus in this discussion has been how viral peptides come to be
exposed on the surface of an infected cell, this process encompasses all endogenous peptides,
not just those of viral origin. Thus, the MHC Class I molecules on the surface of any cell will
present a sampling of all the peptides produced inside that cell, normal or viral.
Thus, in cells which lack Class II molecules (non-antigen-presenting cells), only endogenous
peptides are displayed on its surface, and only in association with Class I MHC molecules
(remember that virtually all nucleated cells bear Class I). In those cells which express Class
II as well as Class I (e.g. macrophages, dendritic cells and B-cells), endogenous peptides are
associated only with Class I (and can serve as targets for T-cell mediated killing), while
exogenous peptides are associated only with Class II (and can serve to "present" antigen to
helper T-cells). (NOTE: While this separation of Class I and Class II pathways generally
holds and is important, there are intracellular mechanisms, not discussed here, which can
cause some exogenous peptides to become associated with MHC Class I, and vice versa.)
One other important feature of MHC-peptide interactions should be kept in mind, namely that
the peptide-binding grooves of MHC molecules, whether Class I or Class II, are already
occupied with a peptide at the time they appear on the cell surface. Peptides become
associated with MHC molecules during the process of assembly of these molecules, and do

99

not generally associate with pre-existing MHC molecules on a cell membrane (except under
experimental or otherwise unusual conditions). As an illustration of this principle, mutant
cells which lack a functional TAP transporter system express few if any Class I molecules on
their surface, resulting in one form of a disease known as Bare Lymphocyte syndrome. In
the absence of suitable peptides to occupy their binding grooves, MHC molecules are
degraded and never appear on the cell surface.
It is also important to recognize that antigenic peptides from the same protein or viral particle
may be associated with both Class I and Class II molecules on the same cell. A macrophage
may, for example, be infected with LCMV; it will therefore express Class I-associated
LCMV peptides (as does any infected cell), in addition to Class II-associated peptides (from
phagocytosed virus particles). In this case the antigen-presenting accessory cell may itself be
a target for cell-mediated killing.
SUPERANTIGENS: PATHOLOGICAL TRIGGERING OF T-CELLS
The mechanism which we have described for T-cell MHC-restricted antigen recognition and
triggering has provided a variety of opportunities for subversion of the immune system by
pathogenic microorganisms. An example of this is the production by various microbes of
SUPERANTIGENS, one of which is the staphylococcal protein Toxic Shock Syndrome
Toxin-1 (TSST-1) whose mode of action is illustrated in Figure 12-8.

APC

APC

II
7

no specific
TcR/Ag recognition

28

II

TSST-1

28

TH
TH
IL-2,
TNF...

TSST-1 AS "SUPERANTIGEN"
Figure 12-8
Superantigens have the unique property of being able to simultaneously bind to both the Tcell receptor and MHC Class II, thus effectively cross-linking a CD4+ T-cell with a Class IIbearing cell (e.g. an antigen-presenting dendritic cell or macrophage). This cross-linking
takes place regardless of whether the TcR has any specificity for the particular ClassII-bound
peptide presented by the APC, but nevertheless results in activation of the T-cell and
production of high levels of lymphokines. Any particular superantigen will be capable of
binding anywhere from 2% to 20% of all TcRs (depending on which V domain is expressed
in the TCR), thus triggering a significant proportion of all the bodys T-cells. (Remember
that normal immune responses engage only a very small proportion of T-cells, a small
fraction of one percent, and that lymphokines are normally present at substantial
concentrations only locally, very close to the site where they are produced.) The resulting
high systemic level of lymphokines can be extremely toxic and is the cause of Toxic Shock
100

Syndrome (as in the illustrated example) and other diseases. Other microbial superantigens
include several of the staphylococcal enterotoxins as well as some virally encoded membrane
proteins.
ALLOGRAFT REJECTION: AN EXCEPTION TO THE RULE??
The above examples and discussion illustrate the important general rule that T-cells must
recognize self MHC molecules in order to be triggered to either provide "help" (in the case of
TH) or to kill target cells(for TC). If such recognition of self Class I MHC molecules were a
simple absolute requirement for CMI, we should fail to see graft rejection between different
strains of mice or between most unrelated humans, which in either case may not share any
identical MHC molecules. This, of course, is not the case, and rejection of allografts takes
place primarily as a result of the recognition of foreign Class I molecules in the absence of
self recognition; it could therefore be regarded as an exception to this rule. The
explanation of this dilemma resides in the complex pattern of differentiation which T-cells
undergo within the thymus, which we will discuss in some detail when we cover the
phenomenon of TOLERANCE in Chapter 18; this process involves both positive and
negative selection of T-cells based on the nature of their T-cell antigen receptors, one result
of which is their ability to directly recognize many foreign MHC molecules.
It is also worth keeping in mind, however, that allografts are really experimental artifacts;
transfer of tissues or organs between individuals (at least in higher organisms such as
vertebrates) does not normally take place in nature, and cell-mediated immunity has not
evolved for the purpose of dealing with such transplants.
IMMUNE RESPONSE (Ir) GENES AND MHC RESTRICTION
An understanding of MHC-linked immune response genes has become increasingly important
in understanding the basis of the striking association of specific HLA haplotypes with a
variety of human diseases, particularly the large family of rheumatoid diseases involving
autoimmunity and inflammation. Immune response ("Ir") genes were first described as genes
which controlled the levels of antibody produced against certain T-dependent antigens. Some
strains of mice, for instance, respond with high levels of antibody to the synthetic polypeptide
TGAL, while mice of other strains respond poorly or not at all. The ability to make antiTGAL antibody could be shown to be controlled by a single genetic locus, and was an early
example of an Ir gene locus. Many other examples of Ir genes have since been described.
Most of these genes have been shown to be located in the Class II region of the MHC, and in
fact are the Class II genes themselves.
Why would MHC Class II genes behave as immune response genes? In what way could
polymorphisms of Class II molecules encourage or prevent the generation of immune
responses to specific antigens? As weve already seen, the generation of helper T-cells for
any given antigen requires T-cell recognition of "processed" antigen coupled to a Class II
molecule on the surface of an antigen presenting cell. Its been found that there exist certain
allelic forms of Class II molecules (remember that they are highly polymorphic) which
fortuitously cannot combine appropriately with a particular antigenic peptide; in such a case,
an organism which has only this ineffective version of Class II will not be able to mount an
effective immune response to this particular peptide. This phenomenon does not result in

101

general immunodeficiency, since it is antigen-specific, and responses to any other antigen

will therefore be normal. A Class II allele which combines poorly with one antigen wlll be
capable of combining very well with another.
IMMUNOLOGICAL DEFENSE AGAINST INTRACELLULAR BACTERIA:
LISTERIA MODEL
T-cells can also participate in other modes of cell-mediated immunity which do not involve
TC-mediated target cell lysis. One example is the mechanism of resistance to intracellular
parasites, this one not involving T-cell killing, is illustrated by the classic experiments of
Mackaness using the bacterium Listeria monocytogenes, a gram-positive organism which is
an animal and human pathogen. While this organism is effectively phagocytosed by
macrophages, it continues growing inside these cells unless the macrophage has been
activated by an antigen-specific T-cell. This is illustrated in Figure 12-9, below.
Sensitized
T-cell

live Listeria
killed Listeria
Infect with
Listeria

TH1
IL-12
4

IFN-

Macrophage with
live Listeria

"Activated" macrophage;
Listeria killed

T-CELLS ACTIVATE MACROPHAGES TO KILL LISTERIA

Figure 12-9
The T-cell (a TH1-class cell) is activated by bacterial antigen through the mechanisms
we have already discussed. This sensitized T-cell can then recognize the affected
macrophage by its MHC Class II-associated Listeria antigens, and begins secreting
interferon-gamma which activates the macrophage. The macrophage then becomes capable
of killing the Listeria organisms it had already ingested. The macrophage, in turn, can
produce IL-12 which further stimulates the T-cell to produce IFN-
[NOTE: There are different mechanisms by which intracellular pathogens may
survive inside mammalian cells. For example, Listeria enters and is able to multiply
within the cell's cytoplasm itself, whereas Mycobacterium (see below) remains inside
the cell's endocytic vesicles where it proliferates. The principles described here, of
the requirement for macrophage activation by T-cells, nevertheless hold for both
situations.]

102

Let's examine a hypothetical set of experiments, illustrated in Figure 12-10, which should
clarify the distinct roles of T-cells (which are antigen-specific) and macrophages (which are
not antigen-specific) in this disease model.

Immunize with
Listeria

CHALLENGE RECIPIENTS WITH:

TRANSFER:

serum

Listeria
alone

Mycobacteria
alone

Listeria
+ Mycobacteria

infected

infected

infected

infected

infected

infected

immune

infected

immune

macrophages

lymphocytes

IMMUNITY TO LISTERIA IS CELL-MEDIATED

Figure 12-10

Immune resistance to Listeria can be adoptively transferred by using lymphocytes, but not by
using macrophages, because activated macrophages are relatively short-lived.
A mouse adoptively immunized in this manner (by transer of immune lymphocytes) is
resistant to a subsequent infection by Listeria, but not resistant to Mycobacterium
tuberculosis, another intracellular bacterial parasite. However, if the recipient mouse is
challenged with Mycobacterium together with Listeria, it will not develop an infection with
either one. The TH1 memory response in this example is specific only for Listeria
antigens, but once the macrophages are activated, they are capable of non-specifically
killing either organism. These experiments in mice illustrate the principles by which human
cell-mediated immunity can provide protection against tuberculosis and other intracellular
bacteria.

103

CHAPTER 12, STUDY QUESTIONS:

1.

Describe the original Doherty and Zinkernagel experiment (cytoxicity against LCM
virus-infected target cells) which defined the phenomenon of MHC-RESTRICTED
RECOGNITION.

2.

Diagram the various CELL-TO-CELL INTERACTIONS which occur during the

generation of a cytotoxic T-cell response. What are the cell surface molecules
involved in each interaction?

3.

With respect to MHC/PEPTIDE ANTIGEN PRESENTATION, describe the fate of a

viral protein: a) following its production during a viral infection; b) following
injection of the purified protein into a mouse.

4.

Immune resistance to LISTERIA INFECTION and to a VIRAL INFECTION are both

the result of cell-mediated immunity (CMI); in what ways do they differ?

104

CHAPTER 13
T-CELL/B-CELL COOPERATION IN HUMORAL IMMUNITY
See APPENDIX (12) PLAQUE-FORMING CELL ASSAY
Adoptive transfer experiments illustrate the use of antibodies to cell surface
markers in distinguishing the roles of different cell types in the CELL/CELL
COOPERATION required for humoral immune responses. While the THYMUS is not
itself a site of immune reactivity, it is the source of T-CELLS (helper TH-cells) which
are required to cooperate with B-CELLS (the precursors of antibody-forming cells) to
generate antibody responses to THYMUS-DEPENDENT (TD) ANTIGENS, as well as
those T-cells (effector T-cells") responsible for CELL-MEDIATED IMMUNITY.
Humoral responses to THYMUS-INDEPENDENT (TI) ANTIGENS, however, do not
require T-cell help. While B-cells in mammals are produced in the bone-marrow, a
distinct organ exists for their production in birds (the BURSA). One widely utilized
animal model for immunodeficiency is the athymic or NUDE MOUSE. The congenital
absence of a thymus in these mice results in the absence of T-cells, and consequently an
almost complete lack of cell-mediated immunity and of humoral responses to TD
antigens.
THE THYMUS IS REQUIRED FOR THE DEVELOPMENT OF IMMUNE
RESPONSIVENESS
The thymus is an organ containing large numbers of lymphocytes, which in humans
surrounds the top part of the heart. Until the 1950s nothing was known of its function,
although its histology clearly made it part of the lymphoid system. Classical kinds of
experiments to determine its function by surgical removal in adult animals gave no clear
results no physiological defects became apparent and the organ was apparently not missed.
One unusual feature which has been recognized since ancient times is that the thymus starts
out as a fairly large organ in very young animals (including humans) which continues to grow
through early life, but then undergoes a process of involution or progressive degeneration and
decrease in size, beginning at about the time of puberty. In older adult animals, the thymus is
often little more than a small bit of connective tissue.
Our present understanding of thymic function developed only around 1960 through the
experiments of J.F.A.P. Miller, who showed that the thymus was critical for the development
of immune competence. For unrelated reasons, he had removed the thymus from mice within
24 hours of birth (neonatal thymectomy), and found that such mice grew poorly, suffered
from a continuous series of viral and other infections, and often died before reaching
adulthood, a condition known as runting syndrome. However, runting only developed if
thymectomy was carried out within the first 24 hours of birth -- thymectomy in older mice
had little or no effect, consistent with the results of the earlier experiments referred to above.
The cause of runting in these neonatally thymectomized mice was shown to be due to the fact
that they are severely deficient in their ability to mount immune responses both to infectious
agents (which accounts for their susceptibility to infections) and to experimental antigens.

105

This immunodeficiency includes both humoral immunity (poor antibody response to some,
but not all, antigens), and cell-mediated immunity (lack of ability to reject skin grafts) and
can be corrected by a thymus transplant. Thus, while the thymus is not required for immune
responses, it is necessary for the development of immune responsiveness. We will see later
that this is because T-cells are produced in the thymus and exported to the peripheral
lymphoid tissues, and it is only in the periphery that they normally respond to antigen
challenge.
SYNERGY IN IMMUNE RESPONSES: ANTIBODY FORMATION
REQUIRES T-CELLS AND B-CELLS
One researcher who followed up on Miller's findings was Henry Claman, who in 1966
published the results of his studies on the role of the thymus in humoral antibody responses.
He used an adoptive transfer system enabling him to mix cells from different sources to
examine the antibody response to SRBC. He transferred thymus cells, bone marrow cells, or
both, together with the antigen SRBC, into lethally irradiated recipients.
900R

THY cells
B.M. cells
SRBC
TEST SPLEEN
FOR PFCs
ON DAY 7

ADOPTIVE TRANSFER DEMONSTRATING T/B CELL COOPERATION

Figure 13-1
He measured the response not by assaying circulating antibody, but by removing the spleen
and determining the number of antibody-forming cells it contained, using the newlydeveloped hemolytic Plaque-Forming Cell (PFC) assay (see APPENDIX 12). An example
of his results is shown below:
Lymphoid
Cells transferred

PFC's per spleen

none
thymus cells
bone marrow cells
thymus + bone marrow

20
50
60
550

These results show very little response with either thymus cells or bone marrow cells alone (a
normal response to SRBC would yield 50,000 or more PFC's per spleen), but there was tenfold higher response with both cell populations together than with either alone. This was the
first evidence of synergy, or cell cooperation in immune responses. While Claman
concluded that the humoral response required participation of cells from both thymus and
bone marrow (which later became known as T-cells and B-cells respectively), he was unable
to determine which of the two sources gave rise to the antibody-forming cells themselves.

106

B-CELLS BECOME ANTIBODY-FORMING CELLS -- T-CELLS ARE "HELPERS"

Mitchell and Miller were soon able to answer this question by developing a modified and
more complex adoptive transfer system, and by using anti-H-2 antibodies to distinguish the
different cell types. Instead of doing a short-term transfer into irradiated hosts, they injected
bone marrow cells into mice that had previously been thymectomized as adults and lethally
irradiated. The injected cells rescued the animals from radiation death by completely
restoring their hemopoietic system, except for T-cells, which could not be regenerated in the
absence of the thymus. After three weeks (to allow hemopoietic restoration to take place),
the animals were immunologically equivalent to neonatally thymectomized animals -- they
had the same immunological deficiency, which could be corrected by transplanting thymus
tissue, or, as we will see below, by providing another source of competent thymus-derived
cells. Such animals are known as ATxBM mice (for Adult Thymectomized, Bone-Marrow
restored).
The second improvement Mitchell and Miller were able to make was to use lymphocytes
from the thoracic duct lymph (the major lymphatic vessel) instead of thymus cells. These
thoracic duct lymphocytes (TDL's) contained more mature thymus-derived cells than
thymus tissue itself, and resulted in a much higher adoptive response. Their adoptive transfer
system is illustrated below:
900R

TDL + SRBC

B.M. cells

7 days

3wks

TEST SPLEEN
FOR PFCs

"ATxBM"

"ATx"

ADOPTIVE TRANSFER IN STABLE CHIMERAS

Figure 13-2
Using this new adoptive transfer system, and measuring the number of PFC's seven days after
immunization, they obtained the kinds of results shown below:
Treatment of ATxBM Hosts

Spleen PFC's

SRBC alone

200

TDL + SRBC

11,000 (!!)

While the ATxBM animals responded almost not at all to SRBC, addition of TDL's restored
their response to levels comparable (although not quite equal) to those of normal, intact
animals.

107

The third advantage of this system turned out to be the fact that they could get this high level
of responsiveness using cells derived from genetically different donors. They constructed an
adoptive transfer with the following MHC types:
Strain
ATxBM HOST
BONE MARROW
TDL

H-2 Haplotype

C57Bl/6
(C57Bl/6 X BALB/c)F1
C57Bl/6

(H-2 )
b/d
(H-2 )
b
(H-2 )

Note that in this particular combination all cells bear H-2 antigens of the H-2b haplotype, but
only the bone-marrow cells (and therefore their progeny) bear H-2d antigens.
In order to exploit the MHC difference between the two cell populations, Mitchell and Miller
added one additional step to the assay procedure: after removing the spleen cells for assay,
but before placing them in the PFC assay system, they treated the cells with anti-H-2 antisera
plus complement. This treatment killed any cells bearing those H-2 antigens recognized by
the antibody, and such cells would therefore not be seen in the PFC assay. They used two
antisera, one detecting H-2b, the other H-2d. If the antibody-forming cells are derived from
bone-marrow, they should be killed by both antisera; if they are derived from TDL, they
should be killed only by anti-H-2b. The results they actually obtained are illustrated below:
PFC per spleen

Treatment of cells
none
anti-H-2b
anti-H-2d

11,000
550
450

These results show that these antibody forming cells can be killed by both anti-H-2 antisera,
and therefore the antibody-forming cells must have been derived from the bone marrow, and
not from the TDL's. Thus, they demonstrated that while both "B cells" (bone-marrow
derived) and "T cells" (thymus-derived) are required to generate a humoral response to
SRBC, only the B-cells develop into antibody-forming cells, while the T-cells perform a
"helper" function. The nature of this "helper" function will be discussed in Chapter 15.
THYMUS-DEPENDENT VERSUS THYMUS-INDEPENDENT ANTIGENS
Not all humoral responses are absent in neonatally thymectomized animals; T-cell deprived
animals can respond quite well to a number of antigens, which is the basis for distinguishing
between thymus-dependent and thymus-independent antigens.
Thymus-dependent (TD) antigens generally include most protein antigens and most
cell surface antigens (which are commonly glycoproteins). Such molecules are
generally capable of inducing not only primary (IgM) responses, but also memory
responses, i.e. class switching to IgG, IgA and IgE isotypes.
Thymus-independent (TI) antigens include polysaccharides with highly repetitive
epitopes; since many antigens in bacterial cell walls (and other microorganisms) bear
108

such structures, this is clearly a biologically significant category. These antigens

commonly induce only primary (IgM) responses with little or no memory or class
switching. (See Chapter 15 for more detail on TI antigens.)
THYMUS AND BURSA IN BIRDS:
SEPARATE PRIMARY LYMPHOID ORGANS FOR T CELLS AND B CELLS
In addition to having a thymus analogous in structure and function to that of mammals,
members of the Class Aves (birds) have another lymphoid organ called the Bursa of
Fabricius, located near the cloaca. Removal of the embryonic thymus before hatching results
in chickens with no cellular responsiveness (they cannot reject skin grafts), and deficient
humoral responses to many, but not all antigens; this is similar to what we have seen
following neonatal thymectomy in mice, and in congenitally athymic (nude) mice (see
below). Removal of the embryonic Bursa, on the other hand, results in a lack of ability to
make any humoral response, while the ability to reject grafts (and carry out other cellmediated immune responses) is unaffected.
Both the Thymus and Bursa are considered to be Central Lymphoid Organs in birds; Tcells mature in the thymus and B-cells in the Bursa, although they actually carry out their
immune responses in other (peripheral) organs. There is no distinct organ in mammals
homologous to the Bursa of birds, and mammalian B-cells are produced largely within the
bone marrow itself.
[It is interesting to note that the chicken was the first system in which the dichotomy between
humoral and cellular responsiveness was made clear, and the term B-cell, in fact, was
originally coined to refer to "Bursa-derived" cells.]
GENETIC THYMIC DEFICIENCY: THE "NUDE" MOUSE
Surgical removal of the neonatal thymus provided the earliest demonstration of that organ's
importance in the development of the immune system. Soon afterwards, however, a genetic
animal model of thymic insufficiency was discovered, namely mice homozygous for the
recessive mutation nu (for "nude").
"Nude" mice, whose genotype is nu/nu, lack body hair (hence the name), and also lack a
functional thymus. They show all the features of neonatally thymectomized mice, i.e.,
"runting" and poor general health, inability to make humoral responses to many (but not all)
antigens, and lack of ability to reject skin grafts and carry out other cell-mediated immune
reactions. They lack functional T-cells in the blood and peripheral lymphoid tissues.
Implanting a fetal thymus (e.g. under the kidney capsule) completely restores the normal
level of T-cells as well as the ability to generate all normal immune responses.
The nude (nu/nu) mouse has been widely used for studies on T-cell function; it is a more
reliable model than neonatally thymectomized animals, since this surgery is difficult to carry
out and may result in a small amount of thymic tissue being left behind. The nude mouse has
also been a valuable tool for the study of heterologous tumors, especially human tumors. In
the absence of cell-mediated immunity, the foreign tumors are not rejected, as they would be
in a normal animal, and they can be grown and transferred from mouse to mouse. Decades of
109

studies in tumor biology have relied on the ability of human tumor cells to grow in nude mice
as a defining feature of their oncogenic potential. The human counterpart of the mouse nu
gene has recently be shown to be the whn gene, and homozygosity for rare mutations at this
locus result in a combination of hairlessness and immunodeficiency very similar to the
condition seen in the nu/nu mouse

CHAPTER 13, STUDY QUESTIONS:

1.

How was the role of the THYMUS in immunity first discovered?

2.

Describe the experiments of Mitchell and Miller which defined the roles of T and B
cells, and state succinctly what their conclusions were.

3.

Define TI versus TD antigens. Give some examples of each which are of significance
in people.

4.

Why are there so many NUDE (nu/nu) MICE in the world?

110

CHAPTER 14
CELL SURFACE MARKERS OF T-CELLS, B-CELLS AND
MACROPHAGES

An understanding of the distinct families of molecules present on different cells of

the immune system provides the tools for distinguishing these cell types in both diagnosis
and therapy, as well as understanding the molecular basis for many lymphoid cell
functions. Diagnostic MARKERS for T-cells include the T-cell receptor (TcR) with its
associated CD3 signaling complex as well as CD5 and the E-receptor (CD2). The
presence of membrane immunoglobulin (mIg), which functions as antigen receptor, is
diagnostic for B-cells. Complement receptors (CR) and Fc receptors (FcR) which can
mediate opsonization, and MHC Class II molecules which are important in antigen
presentation, are present on B-cells and macrophages (as well as dendritic cells). Two
major classes of T-cells are distinguished by the presence of either CD4 (on TH and Treg)
or CD8 (on TCcell). TH cells can be further subdivided into TH1,TH2 and TH17, which
produce different combinations of cytokines and have distinct physiological roles.

We know that there are at least three major cell types required for immune responses, namely
the T-cell, B-cell and macrophage. Conventional microscopy is only of limited usefulness in
distinguishing these cells, and is of no use whatsoever in distinguishing various lymphocytes
one from another (T-cells from B-cells, or different subpopulations of either).
Different classes of cells can readily be distinguished, however, by virtue of the fact that they
express unique combinations of molecules in their membranes. Our knowledge of the
different cell types involved in immune responses is a direct result of the development of
reagents to distinguish these various cells by their cell surface markers. We will see that
such markers include molecules distinguished either by antibodies directed against them, or
else by their ability to bind various other molecules or cells.
SOME MARKERS ON HUMAN T-CELLS, B-CELLS, AND MACROPHAGES

T-cell

MHC
Ig
TcR(&CD3)
FcR
CR
ClassII
__________________________________________

B-cell

Macrophage

Table 14-1

111

Ig. One defining characteristic of B-cells is the presence of membrane immunoglobulin,

which T-cells lack. These immunoglobulins are IgMS (2L2 IgG-like subunits) and IgD,
they are anchored in the cell membrane through a specialized C-terminal domain of the heavy
chain, and they function as the cells' antigen-specific receptors. Memory B-cells bear
membrane-bound isotypes other than IgM, representing whatever Ig the cell will begin
secreting upon antigenic stimulation.
TcR. The T-cell antigen receptor (TcR) is the principle defining marker of all T-cells. This
molecule is used by the T-cell for specific recognition of MHC-associated peptide antigens,
discussed in Chapter 12. Also associated with the TcR is a complex of proteins known as
CD3, which participate in the transduction of an intracellular signal following TcR binding to
its cognate MHC/antigen complex.
FcR. Various cells including B-cells, PMNs and macrophages have molecules on their
surface known as Fc-Receptors, which are able of binding IgG through its Fc region.
Aggregated or antigen-bound IgG binds much more strongly to these receptors than free,
soluble IgG, so that B-cells or macrophages taken directly from an animal will normally have
little or no Fc-bound Ig on their surface. The function of Fc-receptors on macrophages is
easy to understand in that they mediate opsonization. The function of these receptors on Bcells is less clear, and most likely is involved in regulating humoral responses. When
determining if a cell bears surface Ig (to identify B-cells), one must always be careful to
account for the possibility of passive Fc-binding (especially to monocytes and/or
granulocytes), which may give misleading results.
CR. B-cells and macrophages can also bind various complement components, either in free
form or as part of an immune complex, by their complement receptors (CR), of which there
are at least five known forms. One form of CR known as "CR1" binds C3b and C4b, and is
expressed on macrophages, PMNs, B-cells and erythrocytes. As is the case for Fc receptors,
the presence of CR on macrophages clearly enhances opsonization (mainly through binding
of C3b), while the significance of CR1 on B-cells is less clear. The ability of CR1 to bind
C4b is the basis for its role in the normal clearance of soluble immune complexes by
erythrocytes (see Chapter 5). T-cells lack complement receptors.
MHC Class II. Class II MHC molecules in humans include DP, DQ and DR. (In the mouse
they are known as Ia molecules, or I-region antigens). Unlike MHC Class I, Class II
molecules are not expressed on all cells; they are present on B-cells, macrophages and other
antigen-presenting cells, but not on cells of most non-lymphoid tissues. They are not
expressed on "resting" T-cells, but may be induced at low levels when the T-cell is activated.
T-cell receptors on "helper"-type (CD4+) T-cells recognize specific antigenic peptides only
when the peptides are associated with MHC Class II molecules on the surface of an antigenpresenting cell (discussed in Chapters 12 and 15).

112

ANTIBODIES WHICH DISTINGUISH HUMAN T-CELL SUBPOPULATIONS

T-cells consist of a variety of subpopulations, each of which can carry out one or more
specific immune functions. One important way in which these subpopulations can be
distinguished is by the use of the two cell surface markers CD4 and CD8.
Distinct Classes of T-Cells
Class of
Cells

TH

TC

Cell
Function

TH1

Helper cell for CMI, stimulates TC maturation and activity;

recruits macrophages in generating DTH response.

TH2

Classical T-helper; stimulates B-cell proliferation and

differentiation.

TH17

T cell with pro-inflammatory role, facilitates activity of

other cells, including Tc, NK and macrophages

Treg

Regulatory T-cell; suppresses development or

execution of both B-cell and T-cell immune responses

(CD4+)

(CD8+)

Cytotoxic T-cell; kills antigen-bearing target cells;

acts as effector cell in graft rejection and other cell-mediated
immune responses
Table 14-2

Mature T-cells express only one or the other of these two cell surface antigens, whose
molecular roles in cell cooperation we have discussed in Chapter 12. One class of T-cells is
CD4+CD8- and includes the "helper" or TH cells which may be involved in cooperative
interactions with either B-cells (TH2) or with other T-cells (TH1), or play other proinflammatory roles (TH17). Treg or regulatory T-cells, which play an important role in
tolerance (see Chapter 18), also express this CD4+phenotype. The second major category of
T-cells defined by these two markers (CD4-CD8+) encompasses the "cytotoxic" or TC cells
which have important effector functions in graft rejection and other cell-mediated immune
reactions.
The helper class of T-cells can be divided into three sub-categories, based not on cell surface
markers but on the lymphokines they produce and the kinds of immune reactions in which
they participate. TH2 cells are the classical TH helper cells originally defined by their role in
providing B-cell "help" in humoral immune responses. TH1 cells, on the other hand, are
113

"inflammatory" T-cells which provide "help" to other T-cells (e.g. in development of graft
rejection) or directly promote inflammatory reactions by their actions on macrophages and
PMN's (e.g. in skin DTH reactions). The roles of TH1 and TH2 cells are discussed in more
detail in Chapters 12 and15. A more recently defined class of CD4+CD8- TH cell is the TH17
cell, which produces a mix of cytokines including IL-17. This pro-inflammatory cell
stimulates the differentiation and activity of a variety of other cells important in protective
immune resoponses, including NK cells, PMNs and macrophages.
A wide variety of antibodies have been developed which distinguish human T-cell
subpopulations. These have been the result of the development of HYBRIDOMA technology
(see APPENDIX 13), and a few of the commercially available antibodies are listed below:

"CD"
Name

Anti-human
monoclonal
Antibodies

Distribution
on Human
Cells

CD2

(OKT-11, Leu-5)

All T ("E-receptor")

CD3*

(OKT-3, Leu-4)

All T

CD4*

(OKT-4, Leu-3)

TH ,Treg (+ macrophage)

CD5

(OKT-1, Leu-1)

All T

CD8*

(OKT-8, Leu-2)

TC

*You should be familiar with the distribution of these three markers and their
biological significance.
Table 14-3

The CD4 and CD8 molecules (and the antibodies directed against them) are of particular
importance in clinical medicine. The loss of CD4-bearing T-cells, for example, is associated
with disease progression in AIDS, and determination of the CD4:CD8 ratio is a standard tool
in the evaluation of HIV-positive patients. CD4 itself also plays an important role in HIV
infection, as it is the major receptor by which the virus enters those cells which it infects (Tcells and macrophages).
The monoclonal antibodies to human lymphocyte antigens outlined above have been
tremendously useful in diagnostic medicine for evaluating the immune status of patients. Use
of these reagents can yield information about the relative abundance of different functional Tcell subsets, which is much more useful than knowing only the total number of lymphocytes,
or even the relative number of T-cells and B-cells. In addition, such reagents can be used
therapeutically for specific removal of a particular T-cell subset, as, for example, the use of
anti-CD4 to inhibit TH cell function in autoimmune disease.
114

A note on nomenclature: The confusing naming systems historically associated with

many cell surface markers in different species over the years has been made more
rational by the adoption of a standard nomenclature system. If cell surface antigens
in different species (which often are assigned unrelated names as they are discovered)
are determined to be homologous by molecular characterization, they are assigned a
standard name in the CD series ("Complex of Differentiation"). Thus, CD3
molecules in human, mouse, rat, etc all have similar sequences and structures.
Despite such structural homology, however, identical CD antigens do not necessarily
have the same tissue distribution or function in different species. For example, CD5
in humans is present on all T-cells, whereas CD5 in mice (originally called Ly-1) is
present only on the helper subpopulation of T-cells. This is just one example of
many differences in the patterns of differentiation of T-cells between humans and
other mammalian species.
FUNCTIONAL MARKERS FOR LYMPHOCYTE SUBSETS:
MITOGENS AND OTHERS
While the presence of one or another cell surface antigen has long been the most widely used
method for distinguishing different lymphoid cells, other approaches have also been used, for
example the use of mitogens to activate different classes of lymphocytes.
A mitogen is any substance which stimulates a non-cycling cell to undergo mitosis, and many
mitogens are members of a large group of carbohydrate-binding molecules known as lectins.
Concanavalin A (ConA) is one such lectin, which selectively stimulates T-cells to divide in
both humans and rodents; the proliferative response of a lymphocyte population to ConA can
therefore be used as a crude measure of the proportion of T-cells it contains. Another lectin,
Pokeweed Mitogen (PWM), is selective for rodent B-cells and can therefore be used
experimentally as a measure of the proportion of B-cells in a population.
Other functional characteristics can also be used to distinguish various populations of
lymphocytes. The ability to transfer a DTH response, for instance, is the defining
characteristic of TD (or TDTH) cells (which are mostly TH1), and can be used to detect their
presence. As another example, macrophages/monocytes (and granulocytes) can be readily
distinguished from lymphocytes by virtue of their ability to adhere tightly to glass or plastic
surfaces, as well as by their relative resistance to the toxic effects of ionizing radiation. For
example, adherence can be used to selectively remove monocytes from a mixed population of
blood leukocytes, leaving behind mostly lymphocytes. Exposing blood leukocytes to
sufficient doses of ionizing radiation, on the other hand, will selectively kill lymphocytes, and
the surviving cells will be mostly monocytes and granulocytes.

115

CHAPTER 14, STUDY QUESTIONS:

1.

Describe those cell surface markers which characterize and distinguish B-CELLS, TCELLS AND MACROPHAGES.

2.

What functional roles do FcR and CR play in the immune system?

3.

Describe the distinct roles of TH1 and TH2 cells in immune responses. Which cell
markers do they share? Which markers may be used to distinguish them?

116

CHAPTER 15
CELL INTERACTIONS IN HUMORAL RESPONSES:
THE CARRIER EFFECT
A primary humoral immune response requires the cooperation of three cell types,
an ANTIGEN-PRESENTING CELL (APC, typically a dendritic cell or macrophage), a
T-HELPER CELL (TH2), and a B-CELL. The APC/T-CELL interaction involves the
recognition of MHC Class II-bound peptide by the T-cell receptor, and the participation
of the accessory molecule CD4. The same principles apply to secondary humoral
responses, except that the B-cell can also take on the role of APC, presenting processed
antigen to memory T-cells; thus, only two cells (B-cell and memory T-cell) need
participate in this interaction. This two-cell interaction is the basis for understanding the
experimental phenomenon of the CARRIER EFFECT.
"THREE-CELL INTERACTION" IN IMMUNE RESPONSES
The generation of immune responses generally requires the participation of three classes of
lymphoid cells, an antigen-presenting cell (APC), a helper cell and an effector cell. We have
seen one example of this in the generation of cell-mediated immunity outlined in Chapter 12;
in that case the APC was a macrophage or dendritic cell, the helper was a TH1 cell, and the
effector was a TC cell.
The same principle holds in the cell interactions required for the development of humoral
responses. At the top of Figure 15-1 is shown such an interaction, involving an APC, a TH2
cell (a virgin TH2 cell, as opposed to a memory cell, discussed further below), and a B-cell.
This interaction is similar in most respects to the one we have already seen in Chapter 12.
The APC bears surface antigen associated with Class II MHC molecules which is recognized
by the TH cell; the T-cell receptor provides the antigen-specific recognition, and the
CD4/Class II contact stabilizes this interaction. The antigen-presenting cell provides a
"costimulatory" signal to the T-cell in the form of its membrane B7 molecule (which is
recognized and bound by CD28 on the TH-cell) and the TH cell then responds by secreting IL4. The cell which responds to this IL-4 is a nearby B-cell, which bears membrane Ig
receptors (specific for LCMV) which have already been occupied and cross-linked by
antigen; this B-cell responds to IL-4 by differentiating into an antibody-forming cell and
undergoing clonal expansion.
Memory TH-cells (TH(M)-cells) can participate in a second kind of interaction, illustrated at the
bottom of Figure 15-1. This two-cell interaction results from the fact that B-cells are
capable of acting as efficient APC's for memory TH cells. (Its worth noting that we might
have predicted this, based on the fact that B-cells constitutively bear Class II molecules on
their surface.) In this interaction, the B-cell plays two roles. First, it acts as an APC, by
presenting Class-II-associated viral antigen to the TH(M)-cell and expressing the costimulatory
molecule B7. Second, it then acts as an effector cell by responding to the IL-4 secreted by the
T-cell, and it undergoes differentiation and clonal expansion. Thus, this interaction can still
be conceptually regarded as a "three-cell" interaction (APC/Helper/Effector), despite the fact
that only two distinct cells participate.

117

Antigen phagocytosed by macrophage

H
H

Antigen
"processing"

Ag-presentation:
(Ag/ClassII +
costimulation)

C
H

APC

7
C

28

II

CD

TH(V)-CELL

H
H

Antigen
bound by Ig
receptors

B (V) -CELL

T-cell help:
IL-4

("C"=carrier determinant,
"H"=hapten determinant)

Differentiation
to AFC

ANTIGEN PRESENTATION BY MACROPHAGE TO TH(V) -CELL

Ag-presentation
(Ag/Class II +
costimulation)

C
H
H

C
H

Antigen
bound by Ig
receptors

Antigen
"processing"

II

B (M) -CELL

Differentiation
to AFC

28

TH(M) -CELL

CD4

T-cell help:
IL-4

ANTIGEN PRESENTATION BY B(M) -CELL TO TH(M) -CELL

CELL INTERACTIONS IN HUMORAL IMMUNITY

Figure 15-1

118

T/B-CELL COOPERATION
The special importance of B-cells as APCs for memory T-cells stems from the fact that they
are clonally precommitted; those rare memory B-cells (perhaps one in a thousand) which
happen to be specific for an epitope on a particular antigen can very effectively bind that
antigen by their membrane Ig receptors. Having bound the antigen, the B-cell then
internalizes it (by receptor-mediated endocytosis, not phagocytosis), associates its degraded
peptides with newly synthesized Class II MHC molecules, and displays these new class II
molecules on its surface. Those memory TH cells which happen to be clonally precommitted
to carrier epitopes on the antigen (also rare cells) can recognize the Class II-associated
carrier determinants via their T-cell receptors. (Why such B-cell-dependent antigen
presentation plays only a small role in primary responses is not immediately obvious, since
B-cells' advantages as APCs should in principle still hold; it may be that prior to clonal
expansion and affinity maturation they are too rare or bind antigen too poorly to play a more
effective role.)
The signaling which then follows is the same as we have already seen in CMI, namely the
transmission of a co-stimulatory signal by the APC (B-cell), which triggers the helper T cell
to produce IL-4, which then in turn stimulates the B-cell (which bears IL-4 receptors). Thus,
two signals are necessary for triggering a B-cell to differentiate into an effector (=Absecreting) cell: first, its antigen-specific Ig receptors must be occupied and cross-linked;
second, it must receive appropriate signals (e.g. IL-4) from a helper cell. As we have seen,
the requirement for two signals is analogous to that of both TH1 and TC cells in CMI.
For antibody production to T-dependent antigens, T-cell/B-cell interaction is required for
both primary and secondary responses although the nature of the interactions appears to
differ in the two cases (as we have seen in Figure 15-1). We will illustrate some of the
consequences of this model as applied to secondary humoral responses by two experiments
which, more than 40 years ago, defined the CARRIER EFFECT.

EXPERIMENTAL ANALYSIS OF THE CARRIER EFFECT

The key features of the T-cell/B-cell interaction for humoral immune responses, which we
will illustrate using the Carrier Effect, are as follows:
1) Two classes of lymphocyte are required, TH-cells and B-cells.
2) T and B-cells recognize different epitopes, or antigenic determinants, on the
same antigen molecule. Those recognized by B-cells we call hapten determinants,
those recognized by T-cells are carrier determinants.
We can illustrate these requirements with two simple experiments in mice, using the DNP
(dinitrophenyl) group as a hapten, and three unrelated (i.e. non-cross-reactive) proteins as
carriers, namely bovine gamma globulin [BGG], bovine serum albumin [BSA] and chicken
ovalbumin [OVA]
In order to understand these experiments, its important to keep in mind two basic features.
119

1. We are measuring only antibody (humoral) responses to the hapten DNP.

Humoral immune responses against the carrier will also take place, but we will ignore
them. Cell-mediated responses against the carrier may also be generated, but we will
not concern ourselves with these either.
2. We are measuring only secondary, memory responses, which result in high
levels of IgG antibody. Primary humoral responses to DNP will certainly be generated,
but these will produce only low levels of IgM and, again, we will ignore them.
As will be illustrated in the experiments below, generation of these secondary responses
requires the presence of both memory B-cells and memory T-cells, each of which must
have been generated by a previous primary immunization.
EXPERIMENT 1: CARRIER EFFECT
The host must be primed to both hapten and carrier to generate a response, and must
be challenged with both on the same molecule.
We immunize ("prime") mice with various antigens, then challenge ("boost") two weeks
later. We then test their serum for the presence of anti-DNP antibodies one week after the
boost.
Prime
Challenge
Test for anti-DNP Ab
_________________________________________________________
1)
2)

DNP alone
DNP-BGG

DNP-BGG
DNP-BGG

3)

DNP-BGG

DNP-BSA

4)

DNP-BGG

BGG

5)

DNP-BGG

DNP-BSA + BGG

As can be seen in line 3, priming with DNP-BGG will not result in an anti-DNP response to a
boost with DNP on another carrier (BSA). This defines the carrier effect; the host's immune
status toward the carrier affects the response to the hapten.
EXPERIMENT 2: SEPARATE CARRIER AND HAPTEN PRIMING
The host must have been primed to both the carrier and the hapten used for challenge;
however, the priming can be carried out separately for the two kinds of epitopes.
We will give the mice two primary immunizations; these can be given simultaneously (in two
different sites) or a week apart. We will then continue as before, challenging after two weeks
then testing for the presence of anti-DNP antibody one week later.

120

Primary Immunizations
#1
#2
1)
2)
3)

DNP-OVA
DNP-OVA
DNP-OVA

4)
5)

DNP-BGG
DNP-BGG
etc.

Challenge

Anti-DNP
Antibody

BGG

DNP-OVA
DNP-BGG
DNP-BGG

OVA
OVA

OVA
DNP-OVA

In order to produce a secondary anti-DNP response after challenge with DNP-BGG, the
animal must have been primed to both DNP and BGG (line 2); however, as seen in line 3, the
animal can be separately immunized to DNP (on any carrier) and BGG (without DNP).
Therefore, the recognition of hapten and carrier determinants must be executed by separate
cell populations, since they can be separately primed.
By carrying out additional experiments not shown here, one can demonstrate that those
primed cells required for recognition of the carrier are T-cells (i.e. they can be killed with
antibodies to the mouse T-cell-specific antigen Thy-1), while those cells recognizing the
hapten are B-cells (not affected by anti Thy-1 treatment).
The basic feature of T-cell B-cell cooperation that is demonstrated here is that two different
cells need to recognize different antigenic specificities on the immunogen in order to
cooperate. The names "hapten" and "carrier" in this context refer simply to those epitopes
recognized by B-cells and T-cells, respectively, but some clarification may be useful. The
epitope recognized by the B-cells does not necessarily have to be a hapten in the strict
sense of the term (i.e. a small chemical moiety which can be coupled to different carriers).
Any suitable target, including portions of a carrier protein molecule, could be recognized
by B-cells and serve as the hapten portion of the hapten/carrier phenomenon; the use of a
moveable classic hapten such as DNP simply makes the design and interpretation of such
experiments much easier.
The carrier specificity, however, has more limited possibilities. In theory, T-cells could use
DNP as a "carrier" specificity and could then cooperate with B-cells for an antibody response
to some other (or the same) "hapten" determinant present on the BSA molecule. However,
DNP and other typical haptens cannot be recognized by T-cells, because they are not peptides
and therefore cannot bind to the peptide-binding groove of an MHC molecule (see Chapter
12). Thus, while B-cells using their cell surface immunoglobulin can bind to both peptide
antigens, either free or within a larger protein, and haptens such as DNP, T-cells, which use
their TCR for antigen recognition, are largely restricted to recognizing only peptides
intimately associated with a suitable MHC molecule.
TWO KINDS OF T-INDEPENDENT ANTIGENS: TI-1 AND TI-2
The cell interactions shown in Figure 15-1 and illustrated by the Carrier Effect are required
for humoral responses to T-dependent (TD) antigens (see Chapter 13). As already mentioned,
however, there are some antigens which do not require T-cell help to generate a humoral
121

immune response, and these are known as T-independent (TI) antigens. Two general classes
of T-independent antigens can be distinguished, which have been named TI-1 and TI-2.
TI-1 antigens are exemplified by lipopolysaccharide (LPS), a component of the cell wall of
Gram-negative bacteria. LPS is mitogenic for B-cells, and can therefore efficiently activate
any B-cell with LPS-binding antigen receptors, even if the LPS is present only at very low
concentrations, which results in an antibody response. In effect, the mitogenic activity of
LPS replaces the need for the second signal normally provided by the TH cell.
TI-2 antigens, on the other hand, are not mitogenic, but typically bear highly repetitive
epitopes (for example, high molecular weight carbohydrates). The exceptionally high degree
of Ig-receptor cross-linking produced by such antigens seems to provide a very high level of
Signal 1, and reduces or eliminates altogether the requirement for Signal 2 (normally
provided by IL-4 and other lymphokines). However, in the absence of antigen-specific T-cell
help, humoral immune responses to TI antigens of either type generally do not exhibit class
switching, and are therefore largely restricted to triggering IgM responses (as, for example,
the ABO blood group antigens discussed in Chapter 10).

CHAPTER 15, STUDY QUESTIONS:

1.

Describe the different cell/cell interactions involved in PRIMARY versus

SECONDARY humoral responses.

2.

What are the critical cellular requirements for generating an IgG ANTI-HAPTEN
HUMORAL RESPONSE in response to the injection of DNP coupled to a carrier?

3.

Define and distinguish TH1 and TH2 cells.

4.

Define and describe the two classes of T-independent antigens.

122

CHAPTER 16
LYMPHOID TISSUE STRUCTURE
Beginning with mouse lymph node as the prototype, we examine some fundamental
features of all secondary LYMPHOID TISSUES, particularly the segregation of TDEPENDENT REGIONS containing T-cells (DIFFUSE CORTEX) from TINDEPENDENT REGIONS containing B-cells (PRIMARY FOLLICLES). The process
of MIGRATION (the exit of mature B-cells from the bone marrow, the movement of Tcell precursors from bone-marrow to thymus, and the emigration of mature T-cells from
the thymus) is distinguished from that of RECIRCULATION (the process by which both
B- and T-cells continually enter lymphoid tissues from the blood and return to the blood
via the lymph). The formation of GERMINAL CENTERS in pre-existing primary
follicles is seen to be an antigen-dependent and T-cell-dependent process.
Other peripheral lymphoid tissues (e.g. spleen, Peyer's patches), and lymphoid
tissues of other species including humans, although differing in details, all show the same
basic features of structural organization and cell movement.

Immune responses are not carried out in any single organ, but in a wide variety of structures
collectively known as LYMPHOID TISSUE. Lymphoid tissue can be generally categorized
as CENTRAL (or PRIMARY), versus PERIPHERAL (or SECONDARY). Central
lymphoid tissues are those which act as a source of immunocompetent cells; these cells then
migrate to the peripheral lymphoid tissues which are the sites of immune responses.
Central
Thymus
Bone Marrow

Peripheral
Lymph Node
Spleen
Gut-Associated lymphoid tissue (GALT)
Tonsils
Appendix
Peyer's Patches

The lymphoid tissues we will be discussing are broadly included in the term
Reticuloendothelial System, or RES. The RES also includes other regions rich in
phagocytic cells, notably those in the vasculature of the liver and lungs.
MOUSE LYMPH NODE
We will examine the structure of a typical mouse lymph node (diagrammed on next page) to
illustrate several key points: Lymphoid tissue is not just a "bag" of lymphoid cells, but
consists of a highly ordered structure; it is also a dynamic structure, maintained by a
continuous movement of cells into and out of the tissue (via MIGRATION and
RECIRCULATION); characteristic anatomical changes can be observed during the
generation of an immune response.

123

The structure of a mouse lymph node can be divided into two areas, CORTEX and
MEDULLA.
Primary follicles:
collections of densely packed
small lymphocytes

FOLLICLES

Secondary Follicles:

CORTEX

follicles containing a germinal center

DIFFUSE CORTEX

MEDULLA

Region between follicles, less densely

packed small lymphocytes; some
plasma cells and macrophages

Large numbers of plasma cells and "cords" of macrophages;

also lymphocytes, especially medium and large

Figure 16-1

AFFERENT LYMPHATIC
PRIMARY
FOLLICLE
GERMINAL CENTER
MANTLE

SECONDARY
FOLLICLE

DIFFUSE
CORTEX

H.E. VENULE

MEDULLA

EFFERENT LYMPHATIC

ARCHITECTURE OF A MOUSE LYMPH NODE

Figure 16-2

124

As fluids from various tissues are collected in the lymphatics, they enter the afferent vessels
of a lymph node, percolate through the cortex and through the medulla, and are collected in
the single efferent lymphatic through which they leave the node; this fluid eventually ends up
in one of the major lymphatics (e.g., thoracic duct) and is dumped into the venous circulation.
The Primary Follicles of the cortex are made up primarily of B-lymphocytes (with occasional
T-cells) and are characteristic of a resting or unstimulated node. Upon antigenic stimulation,
normally following entry of antigen via the afferent lymphatic fluid, a region of intense
proliferation develops within the follicle, known as a Germinal Center; the germinal center
displaces the remaining densely packed B-cells into a peripheral "mantle", and the follicle is
now known as a Secondary Follicle. In addition to containing proliferating cells, the
germinal center also is a site of cell death, and one diagnostic feature is the presence of
macrophages which contain phagocytosed cell debris ("Tingible Body Macrophages").
The Diffuse Cortex contains mainly T-lymphocytes. In neonatally thymectomized mice (also
in nu/nu athymic mice and congenitally athymic humans) this area is virtually empty of
lymphocytes, and for this reason became known as a THYMUS-DEPENDENT AREA
(TDA) of lymphoid tissue. In young mice (and continuously, although to a lesser extent in
adult animals) mature T-cells which develop within the thymus leave that organ, enter the
circulation and colonize the T-dependent areas of lymph nodes and other peripheral lymphoid
tissues; this process is one manifestation of lymphocyte MIGRATION.
A second consequence of neonatal thymectomy is the lack of development of germinal
centers upon antigenic stimulation (and, of course, a lack of immune responsiveness to Tdependent antigens). Thus, although germinal centers develop within B-cell areas (primary
follicles), they are T-dependent in their development.
LYMPHOCYTE RECIRCULATION
Lymphocytes leave the lymph node continuously through the efferent lymphatic, enter the
blood circulation, and re-enter the lymph nodes through specialized vessels known as High
Endothelial Venules (HEVs); this process is known as RECIRCULATION. If the thoracic
duct (a major lymphatic vessel) of a mouse is cannulated and the lymph (and the cells within
it) is removed for a period of a week, the diffuse cortex becomes emptied, just as if the
animal had been neonatally thymectomized. This is because T-cells recirculate more rapidly
than B-cells and are therefore more readily depleted; a longer period of cannulation will
eventually remove the B-cells as well.
The entry of lymphocytes into lymph nodes is a highly specific process, involving recognition
by lymphocytes of receptor molecules on the endothelial cells of the HEVs; erythrocytes,
granulocytes and other cells are not capable of carrying out this process. As their name
implies, the HEVs are characterized by high cuboidal endothelial cells instead of the
squamous cells commonly lining other vessels, and they are present in some other peripheral
lymphoid tissues (e.g. Peyer's patches) as well. Recirculation of lymphocytes, in fact, is a
universal characteristic of peripheral lymphoid tissues. (NOTE: Recirculation of
lymphocytes also occurs through the spleen, although this organ does not contain HEVs. In
this case lymphocytes enter the spleen by a morphologically distinct route in the "marginal
sinus" which surrounds the follicles.)

125

A diagrammatic representation of lymphocyte migration and recirculation through lymph

nodes is shown below. Keep in mind that these processes occur in all peripheral lymphoid
tissues (spleen, Peyer's patches, etc.), and that all of them contain follicles as well as thymusdependent areas.
thoracic
duct

THYMUS

(T-cells)

bloodstream

MIGRATION

RECIRCULATION

efferent
lymphatic

(B-cells)
(pre-T
cells)

Lymph Node
BONE
MARROW

diffuse cortex,
follicles

LYMPHOCYTE MIGRATION, AND RECIRCULATION THROUGH LYMPH NODES

Figure 16-3
One additional migration pathway which should be mentioned is the one by which bone
marrow-derived pre-T cells enter the thymus (this is also shown in the diagram). These are
the cells which following entry into the thymus differentiate into mature T-cells, undergoing
positive and negative selection to ensure that only suitable T-cell receptors are expressed, and
eventually migrate back out into the periphery. This process occurs at a high rate during the
early development of the immune system, but persists at a lower rate throughout the life of
the animal despite the overall involution of the thymus. (See Chapter 18, TOLERANCE for
a more detailed discussion.)
TIME COURSE OF EVENTS IN A "TYPICAL" IMMUNE RESPONSE
Day 0

Antigen introduced in a local site (e.g., skin of the foot).

Day 1

Antigen found within the draining lymph node (e.g., popliteal), having been
carried in by activated dendritic cells in the tissues, as well as taken up by
resident dendritic cells in the diffuse cortex (also by phagocytic cells of the
medulla). These cortical dendritic cells are highly efficient antigen-presenting
cells, and it is here that the initial interaction involving antigen-specific TH and
B-cells is thought to occur.

Day 1-2 Lymph node enlarges, due to increased entry of cells from circulation and
decreased exit via lymphatics (i.e. inflammatory response).
Day 2

Wave of T-cell mitosis followed by B-cell mitosis.

Day 4

Antibody-forming cells begin to appear in diffuse cortex.

Day 4-5 Early germinal center formation, site of class-switching and somatic mutation.

126

Day 5-6 Early antibody-forming cells exit lymph node via efferent lymphatic, colonize
distant lymphoid tissues (spleen, other lymph nodes); large numbers of
antibody-producing plasma cells begin to appear in medulla.
The following points should be noted:
The mitoses seen in the local node represent clonal expansion of antigen-specific
precommitted T-cells and B-cells, as well as non-specific proliferation resulting
from the many lymphokines present.
The rapid recirculation of lymphocytes (particularly of T-cells) allows the recruitment
of antigen-specific cells to the local site. The enlargement of the node represents
the result of such recruitment, in addition to a great deal of non-specific
accumulation of cells and fluid.
Although the initial response is generated within the lymph nodes which drain the
local site of antigen deposition, the antibody-forming cells which are produced
rapidly spread to other sites, and memory cells rapidly enter the recirculating pool
as well as colonize other peripheral lymphoid tissues; the net result is effective
systemic immune responsiveness and memory.
Germinal Centers are a major site for the generation of memory B-cells, isotype
switching and somatic generation of diversity for immunoglobulins. While
follicles are defined as regions of B-cell localization, it is important to remember
that germinal centers are both T-dependent and antigen-dependent in their
formation.
As has already been discussed, there exists a class of antigens ("T-independent
antigens") which do not depend on T-cell function to generate an antibody
response. The response to such antigens does not lead to germinal center
formation, and the histological events in such responses will not be discussed here.

CHAPTER 16, STUDY QUESTIONS:

1.

Follow a lymphocyte from the bloodstream through its RECIRCULATION

pathway(s) back to the blood; identify where it goes and what specific cell
interactions must take place during this process. How do T and B-cells differ in this
regard?

2.

What cellular and molecular events take place in the GERMINAL CENTER? Do all
immune responses result in GERMINAL CENTER FORMATION?

127

128

CHAPTER 17
ONTOGENY OF THE IMMUNE SYSTEM

HEMATOPOIETIC STEM CELLS originate in the yolk sac of the developing

embryo, migrate early into the FETAL LIVER, and later into the BONE MARROW,
which is the only normal site of hematopoiesis in the adult. The generation of
immunocompetent cells from hematopoietic precursors in the bone marrow is a process
which continues throughout the life of an individual. Human infants are born with a
functioning immune system, and are additionally protected by transplacentally acquired
maternal IgG for the first few months of life. Exchange of cells and immunoglobulins
between mother and fetus takes place during gestation, which may result in MATERNAL
IMMUNIZATION to HLA and blood group antigens, and the pathological consequences
of Rh INCOMPATIBILITY (discussed in Chapter 10).

A discussion of the "development" of an organ system would normally consist simply of a

description of its embryological origins. However, as we have already noted, the entire
hematopoietic system (including the immune system) is in a continuous state of regeneration
throughout life, and is therefore continuously being "developed." It is largely in this context
that we will discuss the ontogeny of the immune system. We will also discuss some
important aspects of the immunological relationships between mother and fetus; keep in
mind, however, that this relationship is still only poorly understood, and is much more
complex than is being discussed here.
ORIGIN OF HEMATOPOIETIC STEM CELLS
All cells of the hematopoietic system are continuously being generated from a single kind of
precursor cell known as the Hematopoietic Stem Cell. This stem cell is capable of
unlimited mitotic cell division, more specifically asymmetric division which results in two
classes of products. One class includes cells in various stages of differentiation, eventually
yielding each of the mature cells of the blood and immune system including lymphocytes
(both B- and T-cells), granulocytes, monocytes, red blood cells and platelets. A second class
of product is represented by new stem cells identical to the parent cells. The stem cell is
therefore said to exhibit the property of self-renewal; in fact, self-renewal is a defining
property of stem cells. This continuous regeneration of immunologically competent cells has
many important consequences, not the least of which is the fact that whatever processes are
required to maintain the normal state of TOLERANCE (see Chapter 18) must take place not
only during embryological development, but continuously throughout life.

129

YOLK
SAC

STEM
CELLS

EMBRYONIC/
FETAL LIVER

PRE-T

Embryonic/
Fetal pathways
Adult
pathways

THYMUS T-CELLS

STEM
CELLS

PERIPHERAL
LYMPHOID
TISSUE
B-CELLS

BONE
MARROW

CELL MIGRATION IN FETUS AND ADULT

Figure 17-1
In normal human adults, the generation of all cells of the hematopoietic system, with one
important exception, is restricted to the bone marrow. Weve already discussed this
exception in Chapter 13; while B-cells (and most other blood cells) are produced within the
bone marrow, mature T-cells are produced exclusively within the thymus, from precursors
("pre-T-cells") which themselves are bone-marrow-derived and have entered the thymus from
the blood.
The question of the origin of the immune system therefore can, in one sense, be reduced to
the question of the origin of stem cells. As shown in Figure 17-1, the first stem cells (and the
first blood cells) appear early in the course of human embryological development (at about
two weeks of gestation) in the blood islands of the yolk sac. As the developing blood vessels
begin making connections with the embryo itself, stem cells move into the developing fetal
liver, the first hematopoietic organ of the embryo, and transiently into the spleen. By the
time of birth, neither the liver nor the spleen remains a site of hematopoiesis in humans; the
stem cells have migrated into the bone marrow, which remains the normal site of generation
of all blood cells throughout life. This movement of stem cells from the yolk sac to the
embryonic liver, and then to the bone marrow, adds two new paths to the patterns of cell
migration we have already discussed in Chapter 16.
[NOTE: In certain pathological conditions, some anemias for example, development of
blood cells may take place at sites other than the bone marrow, a condition referred to
as "extramedullary hematopoiesis". In other species (mice, for example) the normal
spleen may retain its hematopoietic role throughout life; it is for this reason that mouse
spleen contains not only B- and T-cells characteristic of normal peripheral lymphoid
tissue, but also hematopoietic stem cells capable of rescuing a lethally irradiated animal
from hematopoietic death.]

130

IMMUNOLOGICAL STATUS OF THE NEWBORN

Fetal Development of Immunity
Mice are born at an immunologically very immature stage of development; few B-cells are
found in the peripheral lymphoid tissue, and almost no T-cells. This explains the fact that
removal of the thymus immediately after birth results in a profound and permanent T-cell
deficiency in mice (as we discussed in Chapter 13), and accounts for the sensitivity of
newborn mice to developing GvH.
Humans, on the other hand, are born with considerably more mature immune systems; the
newborn human is capable of generating effective immune responses, both humoral and
cellular, although not necessarily at adult levels. T-cell and B-cell areas of peripheral
lymphoid tissue are already largely populated, and show the same basic architecture seen in
adult tissues. (NOTE: One characteristic feature of newborn human peripheral lymphoid
tissue is the presence of primary follicles. These are rare or absent later in life, due to the
normal antigenic stimulation of these tissues, and the consequent development of germinal
centers and secondary follicles).
Maternal IgG in the Newborn
If one examines the level and nature of immunoglobulins in the serum of a newborn human,
an interesting situation becomes apparent. While the total Ig in newborn serum is at a level
close to that of a normal adult, almost all of it is IgG of maternal origin. This results from
the fact that IgG (but not other classes) can be transported across the placenta, passing from
the maternal circulation to that of the fetus via a transport mechanism involving specific Fcreceptors in the placenta. This process begins around the 22nd week of gestation and
continues to term.
total serum Ig
maternal IgG

100

% of adult levels

endogenous Ig

0
0

9 months

SERUM IMMUNOGLOBULIN IN THE NEWBORN

Figure 17-2
As shown in the figure above, the only immunoglobulin normally present at substantial levels
in the newborn is maternal IgG (this can be determined by examination of allotypes). Small
amounts of IgM and trace amounts of IgA are also present; since these immunoglobulins do
131

not cross the placental barrier, they must be of fetal origin; their presence in the newborn
circulation at high levels (detected by its presence in umbilical cord blood) is considered a
sign of intrauterine infection and a resulting fetal immune response.
Mature B-cells and T-cells are already present at the time of birth, but it is only after birth and
exposure to environmental antigens that they normally begin to generate appreciable immune
responses. As a result, the levels of endogenous (as opposed to maternal) serum
immunoglobulin begin to rise during the first few months after birth, IgM rising earliest,
followed by IgG and IgA.
While endogenous synthesis of immunoglobulins is already underway at birth, it takes several
months to reach levels which can effectively replace the protection conferred by the passively
acquired maternal IgG. This maternal IgG disappears with a normal half-life of 2-3 weeks,
and (of course) is not replenished. A low point in total serum Ig is typically reached at about
4-5 months of age, which is also the time at which humoral immunodeficiencies may become
clinically evident (see Bruton's Agammaglobulinemia in Chapter 20).
NOTE: The newborn infant is also protected by maternal IgA which it acquires from its
mother's milk (particularly the early form known as colostrum). However, while this
IgA plays an important role by protecting against infection by gut-localized pathogens,
this IgA does not enter the infant's circulation.
Maternal/Fetal Interactions:
The developing fetus can be regarded as a graft of "foreign" tissue onto the mother; it is
clearly a histoincompatible graft, since at least some HLA antigens (those of paternal
origin) will be foreign to the mother. If this is so, why is the fetus not recognized as foreign
and rejected? In fact, the fetus is generally recognized as foreign by the mother's immune
system, but is nevertheless not rejected. There are several (and still poorly understood)
reasons for this.
First, the placenta itself may act as a filter for anti-HLA antibodies; maternal antibodies
directed against paternal HLA antigens present in the fetal component of the placenta may be
bound by the placental tissue. The placenta is not harmed by these antibodies, but it prevents
their passage into the fetal circulation where they might be harmful, thus effectively
neutralizing the mother's humoral immune response to the fetus. Second, this "coating" of
antibody on paternal antigens present in the placenta may "hide" the foreign HLA antigens
and prevent recognition and damage by the mother's cell-mediated immune response (in the
manner of enhancing antibodies; see Chapters 11 and 23). Third, cells of the outermost layer
of the placenta, the trophoblast, do not express the HLA Class I proteins which are present on
all other nucleated cells, thus reducing the generation of, and the target for, anti-HLA cellmediated responses by the mother. And fourth, the state of pregnancy induces a state of
moderate immunosuppression of the mother (by various mechanisms), which has the effect
of further discouraging anti-fetal responses.

132

While the placenta does a relatively good job of keeping the maternal and fetal sides of the
circulation separate, several materials of immunological importance can cross the placental
barrier to various extents, as outlined in Figure 17-3, below.

RBC, WBC,
placental cells
MOTHER

FETUS

anti-HLA Abs
anti-HLA Tc cells

IgG
MATERNAL-FETAL EXCHANGE
Figure 17-3
One example, as we have already seen, is that of maternal IgG, which is efficiently
transported into the fetal circulation before birth; this IgG is critically important for
protection of the newborn during its first few months of life. Another example is that of
small numbers of cells of fetal origin (red blood cells as well as nucleated cells) which enter
the maternal circulation, presumably through microscopic defects in the placental barrier.
These cells are of interest for at least two reasons. First, they may eventually stimulate
substantial anti-HLA antibody responses in the mother (already mentioned above),
particularly after multiple pregnancies; in fact, multiparous women are an important source of
the anti-HLA antibodies which are used in histocompatibility typing. Second, these rare cells
can be identified and isolated from the mother's circulation; their use for prenatal testing of
genetic disease, which would eliminate the need for the more expensive and hazardous
process of amniocentesis, is currently under development. (Red blood cells may also trigger
maternal antibody responses directed against blood group antigens, although such a response
is far stronger after the appearance of the larger number of red blood cells which enter the
maternal circulation at the time of birth and the separation of the placenta. The development
and significance of maternal anti-Rh responses have been discussed in Chapter 10).

133

CHAPTER 17, STUDY QUESTIONS:

1.

Follow a MEMORY T- or B-CELL back in time through its development; what are
the membrane markers and anatomical localizations associated with each of its
identifiable precursors?

2.

How would you use heavy and light chain Ig ALLOTYPES to distinguish the origins
of serum Ig in a newborn? How would your results be different if you looked at a 6month-old or 2-year-old child?

3.

Why are multiparous women not a good source for ANTI-Gm ANTIBODIES

134

CHAPTER 18
TOLERANCE
TOLERANCE (which is antigen-specific) must be distinguished from immune deficiency
(non-specific).
Its most important manifestation, the maintenance of SELFTOLERANCE, was originally explained by the Clonal Selection theory as the result of
CLONAL ABORTION of self-reactive clones. However, potentially self-reactive T- and
B-cells do exist in normal individuals, leading to the recognition of the importance of
active processes in the maintenance of tolerance, such as SUPPRESSION. The dynamic
and competing balance between immunity and tolerance is discussed, together with three
useful model systems, INDUCED CLONAL ABORTION, RECEPTOR BLOCKADE,
and ANTIGEN SEQUESTRATION. Two general categories of tolerance are recognized,
CENTRAL TOLERANCE which is generated during development of T-cells within the
thymus, and PERIPHERAL TOLERANCE generated by a variety of mechanisms outside
this central lymphoid organ.
One of the defining features of the immune system is its ability to distinguish self from nonself--that is, it must be capable of mounting a reaction against any foreign antigen, but not
respond to substances normally present in the organism itself. This essentially describes the
phenomenon of natural tolerance.
TOLERANCE Lack of ability of an organism to mount an
immune response to a specific antigen.
Antigen-specificity is a key part of the definition of tolerance, and distinguishes it from
the phenomena of immunosuppression or immunodeficiency in which immune
reactivity is diminished to all antigens.
EARLIEST ANALYSIS OF TOLERANCE: DIZYGOTIC CATTLE TWINS
The beginning of our modern understanding of tolerance dates back to the experiments of
Ray Owen in the 1940's. He was studying the genetics of blood groups in cattle, and found
that dizygotic (i.e. non-identical) twins were often chimeric, showing mixtures of genetically
distinct blood groups in their circulation. This was because such cattle twins often shared a
single placenta which resulted in a mixing of embryonic blood between them. Each had
erythrocytes of its own blood groups, plus a varying number of cells of its twin's blood
groups which persisted throughout life. (NOTE: This is very different from most human
dizygotic twins which do not share a common placenta.)
Owen noted a remarkable feature of this chimerism, namely that neither twin made
antibodies against the blood cells of the other, which they would have done had they been
exposed only as adults to the foreign blood cells. In addition, he was able to show that
neither twin would reject a tissue graft from the other as adults, which they also would
certainly do under normal circumstances. He hypothesized that exposure of the immature
immune system to a foreign antigen resulted in specific tolerance to that antigen. This
concept was soon confirmed by Medawar's experiments in mice, and was a central part of
Burnet's Clonal Selection theory published in 1959.

135

As we have mentioned earlier, Burnet in 1959 proposed the mechanism of CLONAL

ABORTION as an explanation of normal self-tolerance. This simply represents the
elimination, at an early stage of development, of all cells specifically precommitted to any
antigen which is present at that stage. In the case of B-cells (remember that the existence of
T-cells was not known at the time) this concept is fairly straightforward, and clonal abortion
is now accepted as being an important mechanism for establishing and maintaining B-cell
tolerance. However, it is more difficult to understand how clonal abortion might operate in
the case of T-cells, given the necessity of their recognizing antigen strictly in an MHCassociated fashion (as we discussed in Chapters 14 and 15).
POSITIVE AND NEGATIVE SELECTION OF T-CELLS WITHIN THE THYMUS
A key to understanding the normal functioning of T-cells, as well as their self-tolerance, is
the process of positive and negative selection which they undergo during their maturation
within the thymus, as shown in Figure 19-1, below.

Epithelial
cell

Strength of
interaction

Result

Strong

Death

Intermediate

Emigration

positive

TC

negative
selection

selection

- +
CD4 8
CD8

Weak/none

++

CD4 8

Thymus
Blood

CD4+8-

Immature T-cell
"double positive"
(majority of thymocytes)

Death

no positive
selection

(Class II-based
TH selection for TcRs
occurs in the same
way as illustrated
above for Class I...)

Pre-T cell
-CD4 8 ("double negative")

Thymic T-Cell Education: Positive & Negative Selection

Figure 18-1
Bone-marrow-derived pre-T-cells, which bear neither CD4 nor CD8, enter the thymus from
the bloodstream, develop into immature T-cells which are both CD4 and CD8 positive, then
lose one or the other marker to become mature CD4+ (helper) or CD8+ (cytotoxic)
effector cells. As part of this process, they undergo selection based on the interaction of their
newly acquired T-cell receptors (TcR) with either Class I (for CD8+cells) or Class II (for
CD4+) MHC expressed on epithelial or dendritic cells within the thymus. These MHC
molecules are associated with a normal complement of self peptides. The case of a CD8+

136

T-cell is illustrated in the figure, and is shown interacting with MHC Class I on an epithelial
cell. If its TcR happens to interact strongly with a Class I molecule presenting some selfpeptide, it is eliminated as a potentially auto-reactive cell; this process is termed negative
selection, and is analogous to conventional clonal abortion which operates on developing
B-cells. However, if this cell never encounters a Class I molecule in the thymus to which its
TcR is capable of binding at all, then the T-cell is also eliminated, since it will not be able to
carry out MHC-associated recognition of foreign antigens when necessary.
Only if the cell experiences a TcR/MHC interaction which is at some intermediate level (not
too strong, not too weak), will the T-cell eventually be allowed to emigrate from the thymus
as a mature CD8+ T-cell; this is the process termed positive selection. It is among this
successful population of T-cells, selected to recognize self MHC at some intermediate
level, that cells exist which may recognize a foreign antigenic peptide in self-MHC strongly
enough to proliferate to generate an immune response when triggered to do so (as we have
discussed in Chapters 14 and 15). Although Figure 19-1 shows only a TcR/Class I
interaction, a similar process of selection takes place within the thymus involving CD4+ Tcells interacting with MHC Class II on dendritic cells.
Thus, Clonal Abortion (i.e. negative selection) is a major contributor to establishing the
normal state of tolerance among both B-cells and T-cells. Because this mode of tolerance is
generated during the development of T-cells within the thymus (a "central" lymphoid organ),
it is referred to as CENTRAL TOLERANCE. However, it is important to recognize that
while this process of elimination of self-reactive clones does take place, it is not completely
efficient. In fact, T-cells and B-cells with receptors capable of recognizing and responding
to self antigens can readily be found in the blood and lymphoid organs of normal individuals.
The recognition that clonal abortion is not a sufficient explanation for the normal state of
self-tolerance led to the realization that active processes must be involved, including the
process referred to as SUPPRESSION. The generation of immunological suppressor or
"regulatory" cells, as well as a variety of other mechanisms which can maintain tolerance by
processes taking place outside the thymus (outlined below), are collectively referred to as
PERIPHERAL TOLERANCE.
TOLERANCE MAY BE CAUSED BY ACTIVE SUPPRESSION
Let's illustrate an example of tolerance mediated by "suppression" using the male-specific
antigen of mice. The H-Y antigen, controlled by a gene on the Y-chromosome, is present on
a variety of tissues in male mice, but not in females. In certain strains (C57Bl/6 for instance)
this can be demonstrated by grafting skin from males onto females and observing that the
grafts are rejected, although relatively slowly. Grafts in the opposite direction, from females
onto males, are accepted permanently.
Female mice may be rendered tolerant to the H-Y antigen if they are injected soon after birth
with spleen cells from male mice of the same strain (as expected from Owen's results
discussed above). Such females will not reject male skin grafts as adults.
We can carry out a simple transfer experiment to test the hypothesis that this experimentally
induced tolerance is due simply to clonal abortion, i.e. the elimination of clonally
precommitted cells reactive to H-Y antigen. As illustrated in Fig 19-2, below, we transfer
spleen cells from a tolerant female to a normal newborn female, in exactly the same manner
137

we did with male cells, then ask if the recipient female can subsequently reject male skin as
an adult. If tolerance is due simply to clonal abortion, we should not induce a tolerant state in
the recipient, since all we are transferring are cells lacking reactivity to H-Y. The recipient's
own immune system should develop the normal number of H-Y reactive cells and should be
normally responsive.
Spleen cells from
tolerant donor
NO REJECTION;
TOLERANT!

Newborn

Test adult with

skin graft

Transfer of tolerance to H-Y antigen

Figure 18-2
What we find, however, is that the recipient mouse grows up to be tolerant to H-Y and cannot
reject male skin as an adult. Therefore, clonal abortion (even though it may indeed have
occurred in the tolerized female) cannot fully explain the tolerant state, and cannot explain at
all the transfer of tolerance as described above. We must hypothesize an active process
which has become known as SUPPRESSION, carried out by cells known as SUPPRESSOR
CELLS. More recent work has established that such suppressor function is carried out by a
population of CD4+ T-cells originally referred to as suppressor T-cells (TS), which are now
more commonly referred to as regulatory T-cells (Treg), and which act to prevent the
maturation of antigen-precommitted T or B-cells into effector cells in an antigen-specific
fashion. Research on the functions and diversity of Treg cells has, in recent years, become an
exciting and tremendously important area of cellular immunology. The mechanism(s) by
which Treg carry out their functions is not fully understood, although it is known to involve
the production of inhibitory cytokines in response to antigenic stimulation.
IMMUNE REACTIONS REPRESENT A DYNAMIC BALANCE
BETWEEN IMMUNITY AND SUPPRESSION
The immune system's response to any antigenic challenge can be regarded as being a dynamic
balance between immunity and tolerance, as illustrated below:
Treg
TOLERANCE

TH
ANTIGENIC CHALLENGE

IMMUNITY

Following any given antigenic challenge, the generation of helper cells pushes the system
toward an effective immune response (both cellular or humoral) while the generation of
regulatory cells pushes toward tolerance induction. The net result may be immunity,
tolerance, or some balance between the two.
What are the factors which determine whether a particular antigenic challenge will result in
tolerance, immunity or neither? While the mechanisms involved in these processes are not

138

fully understood, a fair bit is known about the phenomenology, as illustrated in the table
below:
Influencing
Factor

To Induce
Immunity:

To Induce
Tolerance:

Nature of
antigen

Aggregated,
insoluble

Monomeric,
soluble

Dose of
antigen

Moderate
dose

Very high or
low doses

Route of
immunization

Subcutaneous,
intramuscular

Intravenous

Adjuvant

Use of adjuvant

No adjuvant

Table 18-1
While none of these factors is absolute in its behavior, this table outlines those features which
are generally considered when one is attempting to generate a particular kind of immune
response. Note that the general properties of adjuvants (Freund's, for example) maximize the
induction of immunity while minimizing the induction of tolerance; Freund's converts a
soluble antigen into insoluble form (by creating a water-in-oil emulsion), and it moderates the
effects of high doses of antigen by releasing it over a period of time.
CLONAL ABORTION: EXPERIMENTAL MODELS
We have already discussed two experimental examples of tolerance induction in vitro by the
mechanism of Clonal Abortion in Chapter 7. Ada and Byrt were able to eliminate antigenspecific clones by incubation of precursor cells with radioactive antigen, and Mishell and
Dutton accomplished a similar end by killing proliferating cells with radioactive thymidine.
Effective clonal abortion can also be induced experimentally in whole animals by a variety of
drugs which selectively kill proliferating cells.
If the cytotoxic drug Cytoxan
(cyclophosphamide) for instance, is administered to a mouse for several days following
injection of SRBC, one finds not only that the mouse fails to make an antibody response to
SRBC, but that it is specifically unresponsive to subsequent doses of this antigen for a period
of several weeks. Those specific cells which have undergone clonal expansion in response to
the initial dose of SRBC have been selectively eliminated by the drug treatment, effectively
creating a state of tolerance by clonal abortion (as in the in vitro Mishell-Dutton
experiments). It has not been possible to effectively adopt such treatments to humans,
however, due to the severe toxicity of such drugs.
TOLERANCE BY RECEPTOR BLOCKADE
As indicated in the table above, a high dose of non-aggregated, soluble antigen tends to
induce tolerance rather than immunity. We have already mentioned one example of this
phenomenon in Chapter 2; injection of a high dose of "clarified" BSA into mice not only fails

139

to induce an immune response, but creates a state of tolerance to this antigen which lasts as
long as the levels of circulating BSA are kept high. Part of the reason for this is that although
such an antigen is very poorly processed by accessory cells (and therefore poor at triggering a
T-cell response), it is capable of directly binding to the Ig receptors on B-cells without crosslinking these receptors. Occupation of receptors on B-cells, without concomitant signaling
by T-helper cells, paralyzes the B cells and renders them anergic; not only do they fail to
be triggered by this soluble antigen, but they are no longer capable of being triggered by
subsequent appropriate T cell signaling so long as their receptors are occupied in this way.
Thus, while these specific B-cells have not been killed they are anergic, and a state of
tolerance exists so long as they remain unresponsive and have not been replaced by newly
produced B-cells.
Figure 18-3 illustrates a hapten-specific B-cell which has its Ig receptors occupied, but not
cross-linked, by an excess of soluble antigen. Compare this diagram with that in Figure 15-1.
In the absence of MHC Class II presentation there is no T-cell help generated; furthermore,
even if there were active T-helper cells available, such a B-cell would not be triggered, since
cross-linking of its antigen receptors ("Signal I") is a prerequisite for activation.

Excess of soluble antigen,

receptor occupancy without
cross-linking.

H
H

C
H

C
H

poor
antigen
processing

C
H

C
II

Tolerance

B -CELL

H
H

B-CELL TOLERANCE BY RECEPTOR BLOCKADE

Figure 18-3
Natural tolerance to at least some serum proteins may be maintained by this mechanism. For
example, in certain patients who have very low natural levels of serum IgA, transfusion of
exogenous IgA may result in an "autoimmune" anti-IgA response; this IgA is treated by the
immune system as a foreign protein. This does not ordinarily happen in normal patients who
receive blood transfusions, at least in part because potentially autoreactive B-cells are kept
paralyzed by the normally high levels of circulating IgA.
ANTIGEN SEQUESTRATION AND "PRIVILEGED SITES"
An animal's normal lack of immune response to "self" antigens is not always the result of
tolerance as define above. There are proteins in the eye lens and in spermatozoa, for

140

instance, which are highly immunogenic when injected into normal recipients (either male or
female). Why, then, does an animal's immune system not normally make a response to these
antigens?
It appears to be at least partially due to the fact that the immune system is not normally
exposed to these antigens; they are sequestered, or localized in areas which have no
lymphatic drainage. As a result there is neither immunity nor tolerance induction until the
antigen is introduced to the immune system experimentally (or, as we will see later,
accidentally [see Chapter 19]). Upon such exposure the immune system responds normally,
even though the antigen is "self" and the resulting immune response can be highly damaging
to the organism.
Areas within an organism which have no lymphatic drainage are referred to as
immunologically "privileged sites", since antigens normally present or experimentally
introduced into those areas do not generally stimulate an immune response. Thus, foreign
tumor cells can be grown in the cheek pouch of a hamster, although in other locations they
would be rapidly rejected. In humans, the eye, brain, testes and embryo have been recognized
as immunologically privileged sites.
The immune protection of these "privileged sites", however, is never perfect, and antigen
sequestration is never the sole contributor to the lack of immune responses to antigens in
such regions. There also exists some degree of true peripheral tolerance mediated by Treg
cells, even for those antigens which are largely restricted to privileged sites.

CHAPTER 18, STUDY QUESTIONS:

1.

Define positive and negative selection in the process of T-cell maturation within
the thymus. Compare this maturation process with that of B-cells; in what ways are
they similar or different?

2.

Describe how you might attempt to induce a state of tolerance in an intact animal
using each of three approaches: Hot antigen suicide or 3HTdR suicide (see
Chapter 7), or cytoxan-induced tolerance.

3.

What might be some problems associated with carrying out each of the procedures
from Question 2 in humans?

4.

What are the differences between CENTRAL and PERIPHERAL TOLERANCE?

141

142

CHAPTER 19
AUTOIMMUNITY: BREAKDOWN OF SELF-TOLERANCE
A breakdown of the normal balance between immunity and self-tolerance can result
in AUTOIMMUNITY, through the generation of an excessive level of T-cell helper
function, a deficiency in the normal process of T-cell suppression, or an escape by
autoreactive T-cells or B-cells from the normal process of clonal abortion. Several
human diseases (and animal models) illustrate the key features of autoimmune states
(THYROIDITIS, ENCEPHALOMYELITIS, MYASTHENIA GRAVIS, RHEUMATOID ARTHRITIS and SLE), and the significance of IMMUNE COMPLEXES in
the pathophysiology of autoimmune disease.
An immune response to "self" antigens defines a state of autoimmunity. Clinically,
autoimmune states may be mild or symptom-free, but in other cases may result in severe and
fatal afflictions.
Autoimmune responses may result from two general causes. In one case, antigens normally
"hidden" from the immune system ("sequestered" antigens) may be released into the
circulation and trigger a response by the immune system. In this case no state of peripheral
tolerance need exist (see Chapter 18), and the immune system is capable of generating an
effective response on exposure to the antigen. Examples of clinical or experimentally
induced autoimmunity to such sequestered antigens include those directed against
thyroglobulin, eye lens protein and spermatozoa antigens. (Keep in mind, however, that
sequestration of antigens within some anatomical compartment is never absolute, and that at
least some degree of central and/or peripheral tolerance to such antigens is often
demonstrable).
A different situation exists when the normally effective tolerant state of the immune system
to "self" antigens is for some reason abrogated. The final result of such a breakdown of
tolerance, however, is the same as in the case above: a damaging and potentially fatal
autoimmune reaction.
Breaking of the normal tolerant state of an organism's immune system may result from a
variety of influences:
i) Decrease in Treg activity.
ii) Increase in TH activity (e.g., with adjuvants).
iii) Immune response to foreign antigens which happen to cross-react with "self" (for
example, the experimental induction of rabbit anti-AChR, and streptococcalinduced rheumatic fever in humans)
The mechanisms which maintain a normal state of tolerance are certainly not 100% efficient,
and we have already noted in Chapter 18 that T and B-cells with potentially autoreactive
antigen receptors can be found in healthy individuals. Low levels of autoantibodies can also

143

be commonly found in humans and other organisms, much more frequently than one can find
clinically significant autoimmune disease. There are, however, many clinical and
experimental situations in which autoimmune processes play a key role in active tissue
destruction and disease.
We discuss below a few examples of human autoimmune diseases and identify their key
features. For some of them we will also present experimental autoimmune diseases which
have been central to our understanding of the underlying immunological basis of the human
diseases. We will discuss little of the details of clinical treatment of different human
autoimmune diseases, but will note that it often includes generalized immunosuppression by
steroids or other drugs.
HASHIMOTO'S THYROIDITIS. Autoimmune damage to the thyroid, primarily by the
humoral response, results in decreased function and clinical hypothyroidism. It is
characterized by the presence of circulating antibodies to thyroglobulin and thyroid
peroxidase as well as to microsomal proteins and other components of thyroid cells. A
mononuclear infiltrate is also typically present within the thyroid. The etiology of this
disease is generally unclear, but may include initial damage to the organ from either a viral
infection or trauma. Treatment includes hormone replacement, and, as with most
autoimmune diseases, there is a significant genetic component which includes contributions
by HLA linked genes.
AUTOIMMUNE THYROIDITIS OF MICE
When healthy mice are injected with homogenized mouse thyroid tissue together with
complete Freund's adjuvant, they will develop autoantibodies to thyroglobulin as well
as other antigens normally hidden inside cells of the thyroid. There will also develop a
cellular infiltrate of the thyroid characterizing T-cell mediated immunity, and
progressive damage to the organ resulting in hypothyroidism. This is considered an
example of autoimmunity by deliberately induced release of sequestered antigen
(despite the fact that low amounts of thyroglobulin are normally present in the
circulation.)
GRAVES DISEASE. This is also an autoimmune thyroid disorder, but is characterized by
hyperthyroidism. Autoantibodies are produced which are directed against the receptor for
TSH (Thyroid Stimulating Hormone) which is expressed on thyroid follicular cells, and these
antibodies stimulate the chronic overproduction of thyroid hormone. This is an example of
an antibody acting as an agonist for its target molecules (i.e. they stimulate TSH activity), as
opposed to an antagonist (e.g. anti-AcChR antibodies which block receptor function see
below). Treatment in such cases most commonly involves reducing thyroid function by
surgery or radioactive iodine.
ACUTE DISSEMINATED ENCEPHALOMYELITIS (ADEM). This disease may follow
infection by (or rarely vaccination with) a variety of viral pathogens, including measles,
rubella and influenza. Cellular infiltrates representing a cellular immune response to myelin
basic protein are evident, very similar to EAE, but without a humoral response. A wide
variety of neurological defects can result from this autoimmune reactivity, which may be
fatal.

144

MULTIPLE SCLEROSIS is another human disease characterized by inflammatory

demyelination in the central nervous system (CMI involving CD8+ T-cells specific for myelin
basic protein, or MBP). Its causes are unknown, although the appearance of high levels of
antibodies to measles and other viruses suggests a connection with viral infection.
EXPERIMENTAL ALLERGIC ENCEPHALOMYELITIS (EAE) OF MICE
Mice immunized with brain tissue or with the basic protein of myelin in Freund's
adjuvant develop a progressive and often fatal degeneration of myelinated nerves. It is
typically characterized by the presence circulating antibodies to myelin as well as
mononuclear infiltration of nervous tissue. However, the disease can be transferred to
healthy mice by adoptive transfer of immune lymphocytes, but not by transfer of serum,
indicating that only the cellular immune component is pathogenic. As with
experimental autoimmune thyroiditis, this represents an immune response to a normally
sequestered antigen following deliberate breaking of normal tolerance by use of an
adjuvant.
MYASTHENIA GRAVIS. This disease of unknown etiology is associated with circulating
antibodies against the AcCh receptor, resulting in progressive muscle weakening. It can be
treated with anti-cholinesterase drugs together with immunosuppressive therapy. An
interesting histological aspect of this disease is the frequent presence of structures resembling
germinal centers within the thymus (remember that the normal thymus contains neither Bcells nor germinal centers). Thymectomy is often beneficial in treating this disease for
reasons which arent fully clear, although cells producing anti-AcCH antibodies have been
found within these ectopic germinal centers.
RABBIT AUTOANTIBODY TO ACETYLCHOLINE RECEPTOR (AcChR)
In the 1970s, researchers immunized rabbits with purified acetylcholine receptor
protein from the ray, Torpedo, in order to produce antibodies to help in studying this
molecule. Much to their surprise, the rabbits rapidly developed a progressive and fatal
weakening of their muscles, which was found to be the result of autoantibodies directed
against their own AcCh receptor molecules. This represents autoimmunity due to a
response to a cross-reactive antigen; the rabbits immune systems responded to the
foreign protein of Torpedo, which happened to be closely enough related to their own
receptor that it broke the normal state of tolerance.
Two other human autoimmune disorders are of particular importance, namely Rheumatoid
Arthritis and Systemic Lupus Erythematosus. These are examples of a larger group of
related disorders known as the Rheumatic Diseases, and are part of the group of "nonorgan-specific" autoimmune diseases, to distinguish them from those considered "organspecific" (which include the diseases discussed above).
RHEUMATOID ARTHRITIS (RA)
One common feature of RA is the presence of autoantibodies directed against the Fc fragment
of IgG (as well as other "self" components, nuclear proteins and other intracellular
molecules). The anti-IgG autoantibodies are known as rheumatoid factor, and are mostly
IgM (but may be IgG). The resulting immune complexes may be deposited in various sites
within the vasculature and the joints. In either location they can cause tissue damage through

145

the fixation of complement and the attraction of PMN's; these processes result in the
synovitis and vasculitis which (among other features) characterize this disease.
SYSTEMIC LUPUS ERYTHEMATOSUS (SLE)
SLE is characterized by the presence of autoantibodies to a wide variety of autoantigens,
including DNA, RNA, histones, ribonucleoproteins and other nuclear and cellular elements
(often with rheumatoid factor present as well). These antibodies are associated with a pattern
of damage to an equally wide variety of tissues and organs including skin lesions (erythema)
and damage to the vocal cords which results in a characteristic hoarseness (lupus = "wolf").
In both RA and SLE (as well as many other autoimmune diseases), the etiological connection
between the autoantibodies and the disease state is far from clear; did an initial breakdown of
normal tolerance result in production of autoantibodies which cause the tissue damage, or did
some other process initially cause tissue damage which then caused the subsequent
breakdown of tolerance?

AUTOIMMUNITY AND IMMUNE COMPLEX DISEASE

We've already examined (in Chapter 5) the pathological consequences of circulating immune
complexes, which can result in damage to the vasculature in general, and to the kidney in
particular. It is important to remember that the presence of circulating immune complexes
is a common feature of many autoimmune diseases, most notably the non-organ specific
ones such as RA and SLE; these autoimmune complexes exhibit the same sequelae as those
formed with heterologous antigens, i.e. fixation of complement and local tissue damage. The
kidney is an especially important target of autoimmune-mediated damage. The presence of
immunoglobulins on the basement membranes of the kidney is a diagnostic feature of distinct
autoimmune states, depending on whether they show a "smooth" distribution (representing
antibodies directed against the basement membrane itself, as in Goodpasture's syndrome) or a
granular "lumpy-bumpy" distribution (representing the deposition of circulating antigenantibody complexes, as in RA and SLE). We have also discussed in Chapter 5 the crucial
role that complement plays in the normal removal of circulation immune complexes, and that
defects in this clearance system may predispose to the development of autoimmune disease.

CHAPTER 19, STUDY QUESTIONS:

1.

For three human autoimmune diseases, describe an underlying mechanism which is

known from an animal model.

2.

What relationships exist between the complement cascade and autoimmune disease?

146

CHAPTER 20
IMMUNODEFICIENCY

CONGENITAL immunodeficiencies, while generally uncommon, serve to illustrate

many of the key features of the development and function of the immune system,
particularly those involving the CENTRAL LIMB.
The ACQUIRED
immunodeficiencies, particularly those of IATROGENIC origin, are much more common
and are of broad clinical importance, fueling continuing searches for more effective and
selective immunosuppressive drugs.

The critical importance of the immune system to our everyday health and well-being becomes
especially obvious when one observes the results of deficiencies in immune function. These
immunodeficiencies can be classified into two major categories:
CONGENITAL ("PRIMARY") IMMUNODEFICIENCIES - Victims are born with
these diseases, which are the result either of inherited or developmental defects.
ACQUIRED ("SECONDARY") IMMUNODEFICIENCIES - These are acquired as
secondary results of various disease states, due either to the disease processes
themselves or the therapy used to treat them.
We can also classify immunodeficiency states with respect to which of the three "limbs" of
the immune response is affected, that is whether they represent defects in the AFFERENT,
CENTRAL or EFFERENT limb. Remember that the Afferent limb represents antigen
processing carried out by macrophages and related cells, the Central limb involves the
triggering and proliferation of clonally precommitted T-cells and B-cells, and the Efferent
limb is the effector limb, involving the various effector T-cells (TC Treg TDTH ...) and the
biological consequences of antibody binding (which may include complement fixation,
phagocytosis and allergic responses).
We will briefly go over a few examples of immunodeficiencies, keeping in mind that these
are intended only as illustrations of the importance of various elements of the immune
system. This list is far from complete, and the discussion of each example will cover only a
few of its most striking features.
CONGENITAL DEFICIENCIES OF THE EFFERENT LIMB
CHRONIC GRANULOMATOUS DISEASE
Syndrome: Granulocytes and monocytes carry out their normal functions of phagocytosis, but
are incapable of killing the organisms they phagocytose due to a deficiency of the enzyme
NADPH oxidase, required to produce the "oxidative burst". Patients are susceptible to
various microorganisms which are normally of low virulence, particularly with
Staphylococcus aureus and gram-negative bacteria.

147

Inheritance: This disorder can be caused by several different genetic defects, one of
which is controlled by an X-linked gene; symptoms appear at about two years of age.
Treatment: antibiotic therapy for infections.
COMPLEMENT DEFICIENCIES
A variety of genetic deficiencies of complement components are known, which we will not
review here. Some may have only very mild consequences, but they are often associated with
differing degrees of increased susceptibility to bacterial infections. Interestingly, deficiencies
of some of the early complement components also appear to be associated with increased
susceptibility to development of lupus and other autoimmune states, underlining the
importance of the early components of complement in the normal process of clearance of
immune complexes (discussed in Chapter 5).
CONGENITAL DEFICIENCIES OF THE CENTRAL LIMB
The congenital defects of T-cell and B-cell lineages are the most biologically interesting and
informative of the immunodeficiencies, although the clinically significant ones are extremely
rare. As we review four examples of such conditions, it will be useful to refer to Figure 21-1,
which briefly outlines some key features of the process of differentiation of hemopoietic
cells, focusing on the lymphocytic lineage.
RBC
Platelets
Granulocytes
Monocytes

Ag
CR

CR

c
Hematopoeitic
stem cell

Pre-B
Common
Lymphocyte
Precursor

IgM...

Ab-forming cells

THY

TH,C,S

TH,TC,Treg ...
T-effector cells

Pre-T

CR = complement receptor;
, = membrane IgM and IgD
c = cytoplasmic chain

LYMPHOCYTE DIFFERENTIATION
Figure 20-1
X-LINKED INFANTILE HYPOGAMMAGLOBULINEMIA (BRUTON'S SYNDROME)
Syndrome:
Extreme susceptibility to bacterial infections (more so than to viruses or fungi) starting
at ~4-6 months of age.
Serum IgG is extremely low, other Ig isotypes absent.

148

B-cells (membrane Ig-bearing) are absent, but CR-bearing pre-B-cells may be present.
Faulty Ig heavy chain gene rearrangements; D/J rearrangement may occur, but no
V/D/J.
T-cell numbers and function are normal.
No follicles or germinal centers, or plasma cells.
Inheritance: X-linked recessive gene, frequency ~1 in 105. This gene has been
identified as encoding a tyrosine kinase (Brutons Tyrosine Kinase, or BTK) which is
expressed selectively in developing B-cells.
Treatment: Includes chronic treatment with gamma-globulin (intravenous IgG for
passive immunization) and antibiotics.
This genetic defect prevents B-cell precursors from developing into functional B-cells;
replacement therapy with human IgG is very effective. Since the development of IV-Ig, these
patients have been surviving well past their fourth decade. Nevertheless, live vaccines are
dangerous for these patients and should be avoided.
CONGENITAL THYMIC APLASIA (DiGEORGE SYNDROME)
Syndrome:
Hypocalcemic tetany is evident within 24 hrs. of birth (due to a deficiency of PTH, or
parathormone, which is normally produced by the parathyroid and regulates
potassium and calcium metabolism).
Repeated infections with viruses and fungi, (also bacteria); Candida and Pneumocystis
carinii are characteristic.
No functional thymus (hypoplasia or aplasia).
Few or no T-cells.
B-cells present, but with variable function (serum Ig levels are also variable).
Primary follicles present in lymphoid tissues but without germinal centers, and empty
thymus-dependent areas.
Inheritance: This disease is most commonly sporadic (not inherited) due to de novo
deletion of a region of chromosome 22q11 (remember the distinction between "congenital"
and "inherited"). This deletion results in a developmental defect of the 3rd and 4th
pharyngeal pouches which give rise to both the thymus and parathyroid. Often associated
with malformations of the heart and face, DiGeorge is considered as one manifestation of
the larger complex of diseases collectively known as cardiovelofacial syndrome. The
frequency of the deletion is ~1 in 104, the appearance of severe immunodeficiency is less
common.
Treatment: Successful treatment may be carried out with fetal thymus transplant; the
transplant need not contain lymphocytes, only a small amount of thymic epithelium. Any
therapy must consider the danger of GvH reaction; for this reason one must not use adult
thymus tissue, and any transfused blood must be first X-irradiated. No live vaccines should
be given.

149

This disease is the result of the absence of T-cell development through failure to provide an
appropriate environment for differentiation. The patients lymphocytes themselves (pre-T
cells) are perfectly capable of developing into mature cells, and if the appropriate
environment is provided (by a thymic transplant) they will colonize it and develop
normally.
SEVERE COMBINED IMMUNODEFICIENCY (SCID)
Syndrome:
Overwhelming infections in first year of life.
No functional T- or B-cells (abnormal B-cells may be present).
Thymus and peripheral lymphoid tissues empty or severely depleted of lymphocytes.
Inheritance: Several different known genetic defects can cause this disease, including
both X-linked and autosomal recessive forms, with an overall frequency around 1 in
106
Treatment: The only successful treatment is replacement of the hemopoietic stem cells
by bone marrow transplant. Prevention of GvH is a key to success (by removing cells
of the T-cell lineage).
This genetic defect is expressed in the common lymphocyte precursor of T and B-cells
indicated in the diagram above, and prevents their further development. If a source of
competent stem cells is provided (e.g. by bone marrow transplantation), they will colonize
the thymus and peripheral lymphoid tissues in a normal fashion.
SELECTIVE IgA DEFICIENCY
Syndrome:
Very low serum levels of IgA (<0.05 mg/ml), normal levels of other isotypes.
Normal cell-mediated immunity.
Increased susceptibility to viral and bacterial sinopulmonary infections, although it may
often be asymptomatic. Can also be associated with autoimmune or allergic states.
Inheritance: Genetic basis variable and poorly understood; frequency up to 1 in 300.
Treatment: No specific treatment other than antibiotic therapy. IgA should not be
administered, as it can trigger an anti-IgA autoimmune or allergic responses (see
Chapter 21).
This is the most common of several isotype-specific Ig deficiencies; it is not known whether
it is a defect in the precursors of IgA-producing cells, or in the (unknown) mechanism by
which their differentiation is regulated.

150

ACQUIRED IMMUNODEFICIENCIES
A. SECONDARY TO DISEASE
1)

Many infectious diseases result in more or less general immunosuppression. In

the case of Human Immunodeficiency Virus (HIV), the agent which causes
Acquired Immunodeficiency Disease (AIDS), its pathogenicity is a direct
consequence of its severe depression of immune responsiveness.

2)

Malignancies can often result in immunosuppression, either by generally

interfering with normal physiological functions, or through the production of
factors which specifically suppress immune functions (this may be a significant
aspect of the natural biology of malignancy).

3)

Renal failure can cause the loss of large amounts of serum immunoglobulins into
the urine, resulting in humoral immunodeficiency.

4)

Enteropathies can lead to loss of immunoglobulin through the gut, with similar
results.

B. IATROGENIC. ("Caused by the healer"; referring to a condition which is the result of

therapeutic treatment.)
This category includes the most common immunodeficiency conditions which most
physicians will encounter. They are the result of various forms of therapy which have
either as their goal or as a major side effect the suppression of immune responsiveness.
Increased susceptibility to infections is an important consequence of
immunosuppression which must always be considered, and balanced against the
therapeutic benefits of a particular treatment. Some therapeutic treatments which can
have this result are:
1)

Corticosteroids. Prednisone is widely used for both its anti-inflammatory effect

and immunosuppressive capability; lymphocytes in general are very sensitive to
steroids.

2)

Cytotoxic drugs.
Many anti-tumor drugs (such as azathioprine and
cyclophosphamide) are strongly immunosuppressive as well, and may also be
used intentionally for this purpose. Susceptibility to infections may therefore be a
major side effect of anti-tumor therapy, in a patient who may already be
immunosuppressed by the presence of the tumor itself.

3)

Anti-Lymphocyte Antibodies. Sera from horses immunized with human

thymocytes contain effective anti-T-cell antibodies, and have been used
extensively since the 1950's to inhibit rejection of transplanted organs; such
preparations are known as Antilymphocyte Serum (ALS) or Anti-Lymphocyte
Globulin (ALG). Their use is much more limited now than in the past, as a result
of the increasing availability of more selective monoclonal antibodies (see Chapter
14 and APPENDIX 13) and the development of new classes of
immunosuppressive drugs (such as cyclosporin and FK-506, mentioned below).

151

4)

Ionizing Radiation. X-rays or gamma-rays, often used in tumor therapy, also

destroy whatever lymphoid tissue happens to be in their path. While ionizing
radiation generally kills proliferating cells in a highly selective manner (which is
the basis for its anti-tumor effectiveness), resting lymphocytes are unusually
sensitive to such radiation.

CYCLOSPORIN A AND FK506 (Tacrolimus)

Since the early 1980s Cyclosporin A, a fungal peptide, and FK506, a bacterial macrolide,
have served as remarkably effective immunosuppressive agents in human transplantation.
They inhibit T-cell function to a greater extent than B-cells or phagocytic cells, selectively
blocking TH cell function by interfering with the production of and response to IL-2, but
without killing the cells. While they are less effective in inhibiting established immune
responses, they have nevertheless been tremendously useful in prolonging kidney, liver and
heart transplants. Their molecular target, however, is not restricted to cells of the immune
system, and they exhibit a variety of often serious side effects, including nephrotoxicity.
Although they have not completely replaced steroids and cytotoxic drugs in clinical
immunosuppressive therapy, cyclosporin and FK506 have allowed the use of lower doses of
these drugs, resulting in far less severe generalized immunosuppression and other side effects
in the management of transplant recipients.

CHAPTER 20, STUDY QUESTIONS:

1.

For each of the congenital immunodeficiency diseases discussed, identify the relevant
biological defect and describe how it leads to the symptoms of the disease.

2.

What situations may result in iatrogenic immunodeficiency states?

152

CHAPTER 21
IMMEDIATE HYPERSENSITIVITY: ALLERGY

Allergies represent TYPE I reactions according to the Gell and Coombs

classification. Most are caused by IgE ANTIBODIES which are capable of binding to
Fc-receptors for IgE on tissue MAST CELLS. Cross-linking of these membrane-bound
IgE's by specific antigen results in mast cell DEGRANULATION; this process releases
HISTAMINE and a variety of other effector molecules, which in turn results in the
myriad symptoms of allergy. (rash, hay fever, asthma etc.) Passive cutaneous
anaphylaxis (PCA) in the guinea pig and the Prausnitz-Kstner (P-K) skin reaction in
humans provide models for understanding the underlying mechanism of allergic
reactions. Management of allergies begins with allergen avoidance, and includes the use
of a variety of drugs and allergen-specific DESENSITIZATION.

One of the many manifestations of antibody, specifically antibody of the IgE isotype, is the
mediation of allergic reactions. We are all familiar with such reactions, whether they appear
as hay fever, asthma, sensitivity to penicillin or other drugs, or skin reactions due to
substances in food or cosmetics. This class of humoral immune reactions is referred to as
immediate hypersensitivity, since the time required for its development is much shorter
than typically required for cellular immune reactions (delayed-type hypersensitivity) or
other kinds of humoral reactions.
Any substance which can elicit an allergic response is referred to as an ALLERGEN.
However, an allergen can only be effective in causing an allergic reaction if the recipient has
been previously sensitized; that is, there must be present, in his tissues, antibodies of the IgE
class directed against the allergen. An allergen can be a component of any number of
microorganisms, plants or animals, or may be a synthetic compound. Allergic reactions may
be elicited by exposure through the air (pollen), contact with the skin (cosmetics), ingestion
(natural or artificial food products), or injection (drugs or insect bites).
GELL AND COOMBS CLASSIFICATION OF IMMUNE REACTIONS
Before describing the mechanisms involved in allergic reactions, we will review a general
scheme of classification of immune reactions proposed almost fifty years ago by Gell and
Coombs. While this scheme is quite dated in many respects, it remains widely used and can
be helpful in understanding the relationships between different immune reactions,
Under this scheme, all immune reactions are classified into one of four headings known as
Types I, II, III, and IV.
Type I reactions include all of immediate hypersensitivity, the allergic reactions we
will cover in detail in this chapter. They are mostly IgE-mediated (although other Ig
classes may sometimes participate), and their rapid onset, typically within minutes of
exposure to antigen, is characteristic.

153

Type II reactions result from antibody binding to membrane-bound Ag resulting in

complement-mediated cytotoxicity or opsonization/inflammation. This class of
reactions may be typified by hemolytic anemia (resulting from auto-antibodies directed
against red blood cells), hemolytic disease of the newborn (HDN, see Chapter 10), and
certain drug reactions.
Type III reactions occur when Ab binds to soluble Ags to form immune complexes,
which can cause any of several kinds of immune complex disease (see Chapter 5).
While IgG, IgM and complement may be involved, as in Type II, the key difference is
that Type III reactions are the result of deposition of circulating antigen-antibody
complexes in tissues, as opposed to the binding of antibodies to antigens which are an
integral part of a target cell membrane.
Type IV reactions include all cell-mediated reactions (Cell-Mediated Immunity or
CMI). Delayed-type hypersensitivity (DTH) is a synonym for CMI, and alludes to the
slower development of such reactions compared with antibody-mediated reactions. For
example, the tuberculin skin test to PPD (Purified Protein Derivative) develops over a
period of a few days, compared with a few minutes for Type I reactions and a few hours
for Type III.
GELL AND COOMBS CLASSIFICATION OF IMMUNE REACTIONS
Skin Reaction
Example
(Cellular Infiltrate)

Typical
Time of
Onset

Allergy skin test

(eosinophils & PMNs,
late phase only)

1-20 min

Reaction

Description

Participating
Antibody

TYPE I

Immediate hypersensitivity
(e.g. allergy)

IgE
(plus Mast
Cell)

TYPE II

Ab to cellassociated antigens
(e.g., hemolytic
anemia)

IgG/IgM

----

TYPE III

Ab to soluble Ags,
immune complexes
(e.g., serum sickness)

IgG(IgM)

Arthus reaction

TYPE IV

Delayed-type
hypersensitivity
(DTH);
Cell-mediated
immunity (CMI)
(e.g. graft rejection)

---

7-10 hrs

(neutrophils [PMNs])
---

Tuberculin test
(mononuclear cells,
i.e. lymphocytes and
macrophages)

1-3 days

These distinct kinds of immune reactions can be distinguished not only by the time course of
their development, but also by the nature of the cellular infiltrate present in the sites of a
typical reaction. We have seen (Chapter 5) that one of the consequences of complement
fixation is the release of factors which are chemotactic for PMN's; thus we typically find
these cells in Type III reactions (Arthus reaction). As we will see later in this chapter, one of

154

the mediators of allergic reactions is a factor which is chemotactic for eosinophils--thus the
characteristic infiltrate of the late phase of Type I reactions (e.g. allergy skin test). The
mechanisms of cell-mediated immunity have been discussed in Chapter 12, and we have seen
that lymphocytes and macrophages are both effector cells of such reactions, and that
macrophage-chemotactic factors are known mediators. Thus, mononuclear cells (a term
encompassing macrophages and lymphocytes) characterize the inflammation at the site of a
tuberculin skin test (Type IV reaction). We can therefore see a clear connection between the
nature of the cellular infiltrate identified by the pathologist, and what is known of the cellular
and molecular mechanisms of immune reactions.
IgE: REAGINIC ANTIBODY
IgE antibodies are also known as reagins or reaginic antibody. They have several features
which distinguish them from other classes of Ig:
Low serum levels

IgE serum concentrations are lower than any other class of Ig, typically
in the range of nanograms per ml. (But remember that only IgE which
is bound to mast-cells can initiate allergic reactions.)

Heat labile

IgE is very sensitive to heat, and can be inactivated by moderate heating

which has little effect on other Ig classes (for instance, heating at 56C
for 30 minutes, typically carried out to inactivate complement).

"Homocytotropic"

This refers to the ability of an animal's Ig to bind to its own mast cells
and initiate allergic reaction. IgE is the major mediator of allergy,
although in humans IgG4 can also carry out this function with lower
efficiency.
[This term exists only to distinguish IgE from
"heterocytotropic" antibody, which can initiate allergic reactions only in
a species other than the one from which the antibody was obtained.
Rabbit IgG, for instance, can cause allergic responses when transferred
to guinea pigs, although it does not do so in the rabbit. Heterocytotropic antibodies are mostly of experimental interest.]

SEQUENCE OF EVENTS IN ALLERGIC RESPONSES

Allergic immune responses can be separated into two components, a period of sensitization
followed by the allergic reaction itself.
Sensitization Phase
1) Exposure of the immune system to the immunogenic allergen results in the induction
of IgE antibody, with the participation of T-cells, B-cells and macrophages. This
induction of Ig synthesis must precede the allergic reaction itself, which is outlined in
Figure 18-1.
2) IgE antibody binds to tissue mast cells through its Fc piece (see Figure 18-1). Mast
cells have Fc receptors on their surface with an extremely high affinity for IgE
antibodies. The stage is now set for the initiation of an allergic response.

155

Ag

Ag

IgE produced and

binds to Mast Cell

Allergen binds
to IgE

SENSITIZATION
PHASE

IgE is cross-linked,
triggers Mast Cell

Ag

Mast Cell
degranulates,
releases mediators

REACTION PHASE

IgE AND ALLERGIC REACTIONS

Figure 21-1
Reaction Phase
1) Re-exposure to the allergen results in the binding of the allergen to mast-cell-bound
IgE antibody. Cross-linking of IgE molecules on the mast cell surface by multivalent
antigen is required for the subsequent reactions.
2) The large, basophilic granules present in the mast cells are released into the tissues; this
process of degranulation releases a variety of pharmacologically active compounds;
some directly from the granules (e.g. histamine and heparin), others newly synthesized
by the cell soon afterwards (see below).
Mediator

Effect

Histamine, heparin, proteases

Contraction of the smooth muscle of bronchi, gut

and venules; capillary dilation & vascular
permability; increased mucus secretion

Prostaglandins, leukotrienes,

Bronchoconstriction; asthma; increased

mucus secretion and venule permeability.

Kinins

Vasodilation, histamine-like activity

Cytokines, chemokines,
interleukins (IL-4,8)

Inflammation, tissue remodeling

ECF-A
(Eosinophil Chemotactic
Factor ofAnaphylaxis)

Eosinophilia

156

These and other active compounds can lead to such conditions as asthma and rhinitis,
generally in response to allergens in pollen, dust or animal dander. Allergens in cosmetics
can lead to rashes, and allergens in food, drugs or injected animal poisons may result in
rashes or in vascular or gastrointestinal anaphylaxis. These conditions may range from
very mild to severe, and are potentially fatal (for instance, in the case of severe asthma or
vascular anaphylaxis).
The nature and severity of allergic reactions depends on the route of exposure and the degree
of sensitization, it varies considerably from one species to another, and within humans may
vary tremendously between different individuals; there are clearly very significant genetic
factors which are not understood, but they include at least some HLA- linked components.
EXPERIMENTAL MODELS OF ANAPHYLAXIS
Vascular Anaphylaxis in the Guinea Pig
The guinea pig is extremely sensitive to the development of severe allergic reactions, and
was for many years the system of choice for studying allergic responses and the physiology of
IgE.
Immunize
with OVA
5 min

3 weeks

(sensitization)

anaphylaxis,
death

intravenous OVA
(challenge)

Vascular Anaphylaxis in the Guinea Pig

Figure 21-2

A guinea pig can be sensitized by intramuscular injection of an antigen, say OVA

(ovalbumin). Its immune system responds by producing antibody to OVA, including (but not
exclusively) IgE. Some of this circulating IgE will be fixed onto mast cells in various tissues,
including the vasculature and respiratory tract. Three weeks later, the same animal can be
challenged either with an intravenous dose of OVA or by exposure to an aerosol containing
OVA. Following IV injection, the animal will rapidly develop severe vascular shock and die
within a few minutes (the combination of venule constriction and capillary dilation results in
pooling of blood in the peripheral circulation and a drastic drop in blood pressure). If
exposed to the aerosol, it will equally rapidly die from bronchial constriction, an
experimental model for human asthma.
Passive Cutaneous Anaphylaxis (PCA)
The ability to carry out an anaphylactic reaction can be transferred from a sensitive animal to
a normal one by transferring serum (this defines it as a humoral immune response). This can

157

be done in such a way as to cause a local rather than systemic reaction, as illustrated by PCA
(Figure 18-3). Serum is collected from a guinea pig sensitized to OVA (as described above),
and a small amount is injected intradermally into a normal recipient.
Serum from
OVA-immune g.p.,
intradermal

Blue patch,
PCA
5 min

24 hr

"latent
period"

(sensitization)

OVA + blue dye,

intravenous
(challenge)

Passive Cutaneous Anaphylaxis in the Guinea Pig

Figure 21-3
After 24 hours (the latent period required for the IgE antibodies to bind to the surface of
local mast cells), OVA together with the dye Evans Blue is injected intravenously. Within
minutes, a blue patch becomes visible on the animal's skin where the immune serum had
been injected. This is because local degranulation of the mast cells has resulted in capillary
dilation, and the leakage of fluids from blood into the intercellular spaces of the skin is
readily visible due to the presence of blue dye.
Prausnitz-Kstner (P-K) Reaction
This is a human version of passive cutaneous anaphylaxis, and is primarily of experimental
rather than routine clinical interest. Serum from an "atopic" (allergic) patient is injected
intradermally into a non-allergic recipient. Twelve hours later (this is the latent period) a
small amount of the allergen is injected into the same region. Within minutes the allergic
reaction becomes visible as a wheal and flare. The "wheal" is a region of swelling resulting
from the leakage of fluid into the extracellular space (edema), the red "flare" surrounding it is
the result of increased blood flow.
Atopic serum,
intradermal

Allergen,
intradermal
12 hr

5 min

Skin, cross-section

Sensitized skin

(sensitization)

(challenge)

5 min

Increased blood
flow, redness

..
Prausnitz-Kustner Reaction
Figure 21-4

158

Edema, swelling
(pale center,
surrounded by
redness)

EARLY-PHASE VERSUS LATE-PHASE ALLERGIC REACTIONS

Allergic reactions are referred to as Immediate Hypersensitivity because of their
typically rapid onset. However, there are often important long-term consequences of the
immediate reaction which are collectively referred to as the Late-Phase Reaction. In the case
of an allergic skin reaction (Prausnitz-Kstner, for example) the early or immediate phase
occurs within a few minutes. However, over the course of subsequent hours and days a
cellular infiltrate dominated by eosinophils will develop, which also includes neutrophils,
macrophages and TH2 cells (keep in mind that we have only mentioned a few of the many
inflammatory and chemotactic factors released by mast cells during degranulation).
In the case of asthmatic reactions, the late-phase response may be manifested by a
chronic inflammation and hypersensitivity of the bronchi, with important diagnostic and
therapeutic implications. Cromolyn sodium, mentioned below, can be given prophylactically
to inhibit mast-cell degranulation and prevent the initial reaction. In the presence of an
ongoing chronic late-phase reaction, however, administration of steroids may be indicated for
their anti-inflammatory properties.
EVALUATION OF THE ALLERGIC STATE
In managing allergic patients, it would be useful to have a simple and accurate laboratory test
to determine the specificity and severity of any allergy. While no ideal test exists, one can
monitor IgE levels in such patients, as well as skin sensitivity to defined antigens, with
varying degrees of usefulness.
Determination of serum IgE levels in allergic patients can be carried out by several assays.
One of them, the RadioImmunoSorbent Test (RIST), measures total serum IgE; however,
these levels correlate very poorly with the degree of clinical allergy. Somewhat better is the
RadioAllergoSorbent Test (RAST), which measures the levels of circulating IgE antibody
directed against a particular allergen; still, the correlation is not good.
The reasons why these tests are less than satisfactory are not fully clear, but reflect the fact
that there exists a high degree of variability among different people, not only in the levels of
IgE (relative to other isotypes) in humoral immune responses, but also in the efficiency of its
fixation to tissue mast cells, the sensitivity of mast cells to degranulation, and the sensitivity
of various tissues to the effector molecules and the inflammatory response.
Skin testing can be very useful in determining to which allergens a particular patient is
sensitive. This involves injecting small amounts of various purified antigens intradermally,
and measuring the severity of the wheal and flare reaction over a period of about half an
hour. (This is an active rather than a passive cutaneous anaphylaxis reaction).
Test diets are also very helpful in identifying food allergies. The patient is placed on a diet
free of the most common sources of allergens (wheat and related grains, dairy products, eggs,
etc.), and various foods are added one at a time over a period of weeks or months to
determine which may be responsible for the allergic response.

159

TREATMENT OF THE ALLERGIC STATE

1) Avoidance of allergens is the single most important and effective element in managing
allergic states. This may include eliminating known allergens from ones diet (nuts,
milk, eggs, for instance), removing them from ones home (wool carpets, feather
pillows, pets and plants, for instance), and, of course, avoiding drugs to which one is
sensitive (penicillin and its derivatives, for instance). Eliminating the major known
allergens from ones environment can not only avoid reactions to those substances, but
can also reduce overall sensitivity to other allergens.
2) Desensitization may be quite effective, but not in all cases. This involves deliberate
immunization with small but increasing amounts of a particular purified allergen over a
period of months or years. It is thought that the effectiveness of this procedure results
from the induction of high levels of IgG antibodies, which can prevent allergic
reactions by competing for the allergen and preventing it from reaching mast-cell
bound IgE.
Desensitization has been useful for pollen and dust allergens as well as for some
animal danders, but its effectiveness is unpredictable and varies between individuals.
It is not generally useful for food or drug allergies (although a protocol has recently
been introduced for penicillin allergies). Desensitization also carries with it a certain
degree of risk, as each injection may itself induce a more or less severe allergic
reaction. The procedure must be carried out under careful supervision.
3) Drugs can be useful in both preventing and treating allergic reactions.
Antihistamines inhibit binding of histamine to its receptors (but will not effect the
target tissue response once histamine is bound).
Corticosteroids are anti-inflammatory (useful for the late-phase response), and can
also inhibit histamine synthesis.
Cromolyn sodium is widely used as a prophylaxis for excercise and allergy-induced
asthma, although its mechanism of action is unclear. While it is known to stabilize
mast cells, this is probably not the basis for its efficacy.
Isoproterenol, salbutamol and epinephrine can be used to counteract the effects of
various of the mediators released by mast cells; unlike the drugs mentioned above, they
can be effective in controlling an ongoing allergic reaction. Prompt treatment of a
systemic anaphylactic reaction (to a food, drug or insect bite, for example) can be lifesaving.
PENICILLIN SENSITIVITY: SELF-PROTEINS MAY ACT AS CARRIERS
Type I allergic reactions to penicillin are not uncommon, and can be life-threatening. The
mechanism of generation of such reactions is of interest. The beta-lactam ring of this
antibiotic is chemically highly reactive, and can spontaneously and covalently bind to serum
proteins. The result is, in effect, the production of a hapten-carrier complex which may be
immunogenic (despite the fact that the carrier molecules are "self"); they are capable of
160

generating a humoral antibody response (including IgE) to the penicillin hapten, and of
triggering an allergic reaction once IgE is present. (Penicillin bound to red blood cells may
also serve as a target for antibody-dependent cell destruction and hemolytic anemia, an
example of a Type II immune reaction.)
Other drugs and chemicals can also bind to "self" proteins, and give rise to allergic reactions
(humoral immunity) in addition to delayed type hypersensitivity (cell-mediated immunity) to
small molecules which would not by themselves be immunogenic. DTH skin reactions to
metals such as nickel in jewelry are generated in this manner, producing Contact Dermatitis.
WHY IgE?
While the dangers and discomfort associated with IgE-mediated reactions are well known,
the physiological benefits derived from IgE are less clear; surely the evolution of the IgE
system did not occur in order for us to enjoy the benefits of hay fever and asthma. One
important clue seems to be the association of extraordinarily high levels of IgE (thirty or
more times normal levels) with certain helminthic infections, and IgE-mediated contraction
of the smooth muscle of the gut may obviously promote expulsion of large numbers of the
parasites.
It has also been shown that eosinophils can cause damage to some of these parasitic worms,
but the situation remains puzzling. Most of the excess IgE is not specific for the parasites,
and the eosinophilia associated with such infections affords only a limited level of protection.
In short, the damaging effects of the IgE system are still understood far better than the
positive role it must have had in mammalian evolution.

CHAPTER 21, STUDY QUESTIONS:

1.

Is the IgE antibody bound to a particular mast cell monospecific? Is it monoclonal?

2.

Would heating an antiserum at 56C for 30 minutes affect its ability to transfer an
ARTHUS reaction? A PRAUSNITZ-KSTNER REACTION? CONTACT
DERMATITIS?

3.

What is the rationale for allergy DESENSITIZATION?

161

162

CHAPTER 22
VACCINATION
VACCINATION has had a huge impact on increasing human life expectancy.
ACTIVE IMMUNIZATION confers long-lasting immunity by inoculation with killed or
attenuated organisms, toxoids or purified antigens. PASSIVE IMMUNIZATION by
transfer of antibodies has the advantage of more rapidly establishing protection, but it is
short-lived and may carry a risk of inducing serum sickness. The route and schedule of
immunization, physical nature of the vaccine antigen, and immune status of the host all
contribute to the relative effectiveness of any vaccination protocol, in ways which are still
not fully understood.

HISTORY
Awareness of acquired resistance to infectious diseases has existed since ancient times. This
knowledge led to the development in China, several centuries ago, of the technique of
variolation, the deliberate inoculation of small amounts of material from smallpox pustules
into healthy individuals. Having recovered from the disease induced by this treatment, which
generally was milder and less dangerous than the naturally acquired disease, the inoculated
individuals had a life-long immunity to smallpox. Variolation was brought from China to the
West, and was increasingly widely used in Europe and the New World through the eighteenth
century. It remained, however, a dangerous procedure with an appreciable risk of death, and
provided a source of infection for spread to others.
This technique was replaced almost overnight following Jenner's publication of the results of
his technique of vaccination. He had learned of the fact that milkmaids who had handled
cows which were suffering from cowpox developed a high degree of resistance to the related
human disease, smallpox. He then was able to show that inoculation of cowpox material into
humans, while not causing serious disease, was in fact capable of producing resistance to
smallpox.
The distinction between variolation and vaccination is an important one. In vaccination, the
fact that the inoculated material is not a virulent human pathogen makes this technique much
safer than variolation, and it derives its name from the species which originally provided this
material (vaccus = cow).
MODES OF IMMUNIZATION
A)

ACTIVE IMMUNIZATION

Active immunization, of which Jenner's original vaccination is an example, involves the

inoculation of immunogenic material to induce an immune response (and therefore
immunological memory) in the recipient. Active immunity takes longer to develop than
passive (see below), but also lasts much longer, and may often be life-long.

163

Various forms of antigenic material may serve as a vaccine:

i) Killed organisms. The viruses which cause rabies, influenza and polio (the latter
in the case of the Salk vaccine) can be collected, killed by treatment with heat or
chemicals, and used as effective vaccinating agents. The bacteria responsible for
cholera, whooping cough (pertussis) and typhoid fever can be used in the same
manner.
ii) Attenuated organisms. Live viruses, but in a weakened or "attenuated" form,
provide effective vaccination for measles, mumps and polio (Sabin vaccine), and
more recently for influenza (in a form administered by nasal mist). Attenuated
bacterial vaccines also exist, typified by those for anthrax and for tuberculosis
(BCG, Bacille de Calmette-Guerin, an attenuated form of the organism which
causes bovine turberculosis). The advantage of attenuated organism over killed
ones is that they can set up active (although hopefully harmless) infections and
provide more effective stimulation of protective immune responses.
Attenuation of viruses may be achieved by growth in vitro in the laboratory and
selection of genetic variants with limited pathogenic potential. While the use of
vaccinia (cowpox virus) also falls into the category of "attenuated viruses", it actually
represents the fortuitous existence in nature of a cross-reactive but non-pathogenic
organism. Similarly, attenuation of bacteria may be achieved by allowing bacterial
cultures to "age" during laboratory culture with or without deliberate selection of
variants.
A key factor in producing any viral vaccine is the development of a cell culture system
which allows the growth of the virus in the laboratory. The development of the Salk
vaccine in the 1950's, for example, followed hard on the heels of the discovery that
monkey kidney cells in culture could be used to grow the virus.
iii) Toxoids. In the case of diphtheria and tetanus infections, the real danger in the
disease comes not from the presence of the organisms themselves but from the
potent toxins which they produce. Effective immunity can be induced by
immunization with chemically modified toxins, or toxoids, which are no longer
toxic but still highly immunogenic (and, of course, cross-reactive with the native
toxins).
iv) Purified antigens ("subunit vaccines"). Vaccines for meningococcus (Neisseria
meningitidis), pneumococcus (Streptococcus pneumoniae) and the Hepatitis B
virus each consist of purified antigens from these organisms, polysaccharide for
the first two and protein for the last. In those cases where effective antigens can
be identified and purified they have an advantage over attenuated vaccines in not
posing any risk of infection, and over killed organism vaccines in being less likely
to cause severe inflammatory reactions.
A relatively recent development in subunit vaccines can overcome the limitations
of immune responses to carbohydrates, which are typically restricted to IgM.
Carbohydrates can be chemically coupled to immunogenic protein carriers, and the
resulting conjugate vaccines have been developed for pneumococcus and
meningococcus (among others) which can stimulate stronger responses showing
class switching and greater memory, and are effective in younger children.

164

v) Naked DNA. This relatively recent development in vaccination comes from the
surprising finding that direct inoculation of DNA encoding a protein results in a
strong and long-lived immune response to that protein, both humoral and cellmediated. It is thought that the DNA transfects local APCs resulting in expression
of the encoded protein in the context of both MHC Class I and Class II. One
attractive feature of such a protocol is that it can yield a strong cell-mediated
response without having to resort to the use of live virus vaccines. However,
DNA vaccination in humans has not yet progressed past the clinical trial stage.
B)

PASSIVE IMMUNIZATION

Injection of antibody to a pathogen can provide very rapid, although short-lived, resistance to
infection, and is referred to as passive immunization. Passive immunization is generally
used when there is no time to wait for the development of active immunity (see below), or
when no effective active vaccine exists.
i) Human antibodies. Normal human IgG, prepared from pools of many
individual donors, contains significant levels of antibody to measles and
hepatitis viruses. High levels of protective antibody for tetanus can be obtained
from immunized donors, and anti-Zoster antibodies from the serum of patients
collected during recovery from an infection. In each case, these antibodies can
be administered to recipients who are at high risk for acquiring the disease.
ii) Heterologous antibodies. Horse antibodies to diphtheria toxin or to the toxins
of snake and spider venoms have been very effective in neutralizing the effect of
these dangerous molecules. The use of heterologous serum, of course,
introduces a substantial risk of inducing serum sickness or an allergic reaction.
C)

ADOPTIVE IMMUNITY

We have already discussed adoptive immunity in experimental systems, illustrated by the

transfer of immune reactivity to non-immune (and/or irradiated) recipients using
immunocompetent cells, typically spleen cells. Obviously, such adoptive transfer of reactivity
cannot be carried out in humans due to histocompatibility barriers, and adoptive immunity
essentially does not exist in human medicine. This term is sometimes applied in a somewhat
different context, i.e. describing the hematopoietic restoration/replacement which results
from bone marrow transplantation, as, for example, in the case of Severe Combined
Immunodeficiency (see Chapter 20).
SOME FACTORS AFFECTING IMMUNIZATION
i) Adjuvants (see also Chapter 2). The most effective experimental adjuvant known,
Freund's Adjuvant, cannot be used in clinical medicine because of its severe side
effects (inflammation, pain, and fever). Some human vaccines can be rendered more
effective, however, by precipitating the antigen together with an aluminum
hydroxide salt, a procedure known as alum precipitation; diphtheria toxoid is used
in this form. Synthetic adjuvants usable in humans are currently being developed,
some involving synthetic versions of the biologically active molecules of Freund's
(including muramyl dipeptide).

165

In some cases one can take advantage of the natural adjuvant properties of certain
vaccines, notably pertussis. The so-called triple vaccine, (DPT), consists of alumprecipitated diphtheria toxoid, killed pertussis organisms and tetanus toxoid. In this
case, the pertussis organisms act as an adjuvant (much as Mycobacterium does in
Freund's), which increases the immune response to the two purified protein antigens.
ii) Route of immunization. While most vaccinations are introduced through the skin,
either by scarification (e.g. smallpox) or by injection (e.g. Salk polio vaccine and
many others), the Sabin polio vaccine is one notable exception. The attenuated viral
organisms are administered orally, and they set up a chronic infection in the gut,
stimulating a local IgA response. Since the normal mode of entry of polio virus is
through the gut, this antibody response is precisely in the place where it should do
the most good. (It should be noted, however, that the Sabin vaccine is no longer
recommended for use in the U.S.) Similarly, one form of the influenza vaccine
(Flumist) is administered as an intranasal mist, mimicking the normal route of entry
of the infectious organism.
iii) Dose of antigen. The dose and time course of human vaccinations is largely
determined empirically; whatever works is used.
iv) State of the host. The effectiveness of active immunization naturally depends on
the ability of the host to mount a normal immune response. It can be dangerous,
however, to introduce any live vaccine into a host with a T-cell deficiency, since
even an attenuated organism can give rise to a lethal infection in such an
environment.
In the case of an immunologically healthy host, the degree of urgency may determine
if passive or active immunization is appropriate. For tetanus, active immunization
with the toxoid is generally used and is effective for ten years or more. However, in
the case of a very severe wound, or a tetanus-prone wound which is several days old
before being presented to the physician, passive immunization may be administered
for immediate protection, together with active immunization for longer-lasting
immunity. Similarly for rabies, passive immunization may be added to the standard
active immunization in the case of a particularly severe rabies-prone wound, or one
close to the head (since the brain is a major target of the virus).
OTHER ISSUES ASSOCIATED WITH IMMUNIZATION
A) Antigenic variation. Influenza virus, for example, can rapidly alter its antigenic
structure by mutation, so that it is no longer recognized by antibodies made against the
original virus. Influenza can therefore give rise to repeated infection with variants of
the same organism.
B) Antigenic competition. Two antigens given at the same site can sometimes each
interfere with the immune response to the other. In general, therefore, different
immunizations are given at different times and/or at different sites. But remember that
DPT is a notable exception to this rule, and other routine multiple vaccination protocols
exist (such as MMR).

166

C) Maternal immunoglobulin. The presence of specific antibodies at the time of

vaccination may interfere with its success. Measles vaccine, for example, should not be
given before 15 months of age, since the presence of maternal IgG antibodies may
prevent active immunization (remember RhoGAM, discussed in Chapter 10).

ACTIVE IMMUNIZATIONS
Diseases

Vaccine Type

Disease Organism

General
*Poliomyelitis
Measles
Mumps
Rubella

"MMR"

(German Measles)

*Diphtheria
*Pertussis
(whooping cough) DPT

*Tetanus
*Hepatitis B
Pneumococcal (PCV)
Meningitis
Hib
*Influenza
Varicella (chicken pox)

killed virus (Salk)

Polio virus
attenuated virus (Sabin) [not currently recommended]
attenuated virus
Measles virus (Rubeola)
attenuated virus
Mumps virus
attenuated virus
toxoid

Rubella virus
Corynebacterium diphtheriae

killed bacteria
toxoid
purified protein (recombinant )
conjugated carbohydrate
conjugated carbohydrate
conjugated carbohydrate
killed virus (injected)
attenuated virus (nasal mist)
attenuated virus

Bordatella pertussis
Clostridium tetani
Hepatitis B virus
Streptococcus pneumoniae
Neisseria meningitidis
Hemophilus influenzae type b
Influenza virus

live virus (Vaccinia)

killed virus
attenuated virus
killed organisms
killed or attenuated bacteria
killed bacteria
killed bacteria
killed bacteria
attenuated bacteria (BCG)
killed bacteria

Smallpox virus (Variola)

Rabies virus
Yellow fever virus
Rickettsia typhi
Salmonella typhi
Salmonella paratyphi
Vibrio cholera
Yersinia pestis
Mycobacterium tuberculosis
Rickettsia rickettsii

attenuated virus

Varicella-zoster virus

Varicella-zoster virus

Special
*Smallpox
Rabies
Yellow fever
Typhus
Typhoid fever
Paratyphoid
Cholera
Plague
Tuberculosis
Rocky Mountain
Spotted Fever
Shingles

*Key examples

167

CHAPTER 22, STUDY QUESTIONS:

1.

Describe at least one example each of clinical vaccination by (a) live organisms,
(b) killed organisms, or (c) purified antigen. What are the advantages and
disadvantages of each?

2.

What are the benefits and risks of active versus passive immunization?
examples.

3.

How might one account for the phenomenon of antigenic competition?

168

Give

CHAPTER 23
TUMOR IMMUNOLOGY

While the importance of the immune system's role in normally preventing the
development of tumors is questionable at best, it is clear that immune responses against
tumors often do develop, and can be detected and studied in both experimental and
clinical situations. This chapter introduces some of the concepts required to understand
the complex relationship between the immune system and tumor development, and a
selection of the complex terminology which has characterized this field. Three key points
are: (1) Tumor cells often do express new antigens ("neo-antigens") which are potential or
actual targets for immune recognition. (2) Despite this fact, the immune system is often
ineffective in eliminating tumors or preventing their growth. (3) Various means may be
possible for modifying anti-tumor immune responses to render them more effective, and
to use immunological approaches for both diagnosis and treatment.

NEOPLASIA:
Appearance of a tumor (from the Latin word for swelling) results from ABNORMAL
PROLIFERATION of cells, through the loss or modification of normal growth control. Cells
which normally do not divide (e.g. muscle or kidney cells) may start proliferating, or cells
which normally do proliferate (e.g. basal epithelial cells or hemopoeitic cells) may begin
dividing in an uncontrolled fashion.
If the growth of a tumor remains localized, it can often be removed surgically (if it is
accessible), and is therefore relatively harmless; or benign. In some cases, however, cells
from a growing tumor may be capable of intruding into adjacent normal tissue ("invasive
growth"), and may leave the original site and begin to proliferate in a new location
(METASTASIS). These properties distinguish "MALIGNANT" tumors from benign ones.
It is important to understand and reconcile two seemingly contradictory properties of tumors
namely their MONOCLONAL ORIGIN versus their HETEROGENEITY. While tumors are
almost invariably of monoclonal origin, mutation and chromosomal instability commonly
generate a substantial degree of heterogeneity within tumor populations (see
IMMUNOMODULATION and IMMUNOSELECTION, below). This fact has important
implications for both diagnosis and treatment of cancers.
ORIGINS OF NEOPLASIA:
At least three major factors are known to initiate tumor development; regardless of which of
the three is the cause (or if the cause is unknown), the consequence is uncontrolled growth,
resulting either from the abnormally high expression of genes which stimulate cell
proliferation (ONCOGENES), or defective expression of genes which normally control
proliferation (TUMOR SUPPRESSOR GENES):

169

1) CHEMICAL CARCINOGENS: Many natural as well as man-made compounds

present in our foods and environment are known to be capable of inducing tumors, i.e.
they are carcinogenic (tobacco smoke and char-broiled steaks, for example, both
contain complex mixtures of carcinogens).
2) IONIZING RADIATION: X-rays, gamma rays and ultraviolet (UV) radiation can all
induce the appearance of cancers by their ability to cause changes in gene expression;
We are exposed to these through sunlight (UV), normal background radiation (gamma
rays), and clinical sources (gamma rays and X-rays).
3) VIRUSES: Many viruses carry their own oncogenes, i.e. genes whose products can
stimulate neoplastic transformation of the infected cell; these genes may be expressed
during a viral infection, and in some cases may be integrated into the genome of an
infected cell and be expressed at a later time.
TUMOR-SPECIFIC ANTIGENS (TSA) and
TUMOR-ASSOCIATED ANTIGENS (TAA):
As mentioned above, many tumors can be shown to express cell surface antigens which are
not expressed in the normal progenitor cells before the neoplastic transformation event.
These antigens have been categorized based on their nature and distribution, resulting in a
complex collection of acronyms, some of which are defined here:
1) Chemical or radiation-induced tumors each generally express a unique neo-antigen,
different from other tumors induced by the same or different agent. These have been
termed Tumor-Specific Transplantation Antigens, or TSTA.
2) Tumors induced by the same virus express antigens shared between different tumors.
These consist of membrane-expressed virally encoded antigens, and have been termed
Tumor-Associated Transplantation Antigens, or TATA (since they are not, strictly
speaking, tumor specific).
3) Oncofetal antigens: These are TATAs which are more or less selectively expressed on
tumors, but are also shared with some normal fetal or embryonic tissues. Examples
include carcinoembryonic antigen (CEA, shared with healthy fetal gut tissue), and
alpha-fetoprotein (AFP, also present in the serum of healthy infants, but decreasing by
one year of age). Expression of such antigens by tumors is thought to reflect their
reversion to a less fully differentiated state.
IMMUNOLOGICAL SURVEILLANCE:
In its classical form, this term refers to the idea that an important role of the adaptive immune
system, particularly cell-mediate immunity (CMI), is to destroy newly appearing neoplastic
cells as a result of the new antigens they tend to express. However, considerable
experimental evidence has failed to support this idea, at least in its simplest form. Most
importantly, immunosuppressed hosts do not generally show the expected high frequency of
tumors predicted by this theory, a phenomenon which has been most strikingly illustrated in
the nu/nu thymus-deficient mouse model, extensively studied for precisely this reason. Those

170

tumors which have been found more frequently in immunosuppressed hosts are usually those
induced by viruses, and are best understood simply as the defective immune systems
inability to control the viral infection as it normally would. The importance of the adaptive
immune system in providing natural protection against neoplasia is therefore highly
questionable at best, although NK ("natural killer") cells are still considered as possibly
playing such a role. Nevertheless, as we will see, there has been considerable interest in
exploiting those anti-tumor immune responses which are known to occur, and direct them
toward more effective cancer therapy.

FACTORS WHICH LIMIT ANTI-TUMOR IMMUNE RESPONSES:

If tumors express antigens which can be recognized by the immune system, why are they not
normally eliminated, as if they were a foreign tissue graft? Several more or less wellcharacterized processes have been shown to contribute to the ability of tumors to continue to
grow despite the possible or demonstrated presence of a potentially destructive immune
response.
TOLERANCE: A state of immunological tolerance may be promoted by the presence
of very high levels of tumor antigens, particularly soluble forms which may be shed
into the serum by tumor cells (see also Blocking Factor, below).
IMMUNOMODULATION/IMMUNOSELECTION:
Immunomodulation: Antibody binding to a membrane antigen on either normal or
neoplastic cells may result in the disappearance of that antigen from the cell surface
either by endocytosis (internalization) or by shedding. Once a tumor cell has lost its
neoantigen, it becomes invisible to the immune system.
Immunoselection: Variant cells can appear in a tumor population which have lost a
particular TATA or TSTA through a mutational event; such a variant will have a
selective advantage in the face of an even slightly effective immune response (either
humoral or cellular), and cells with the antigen-negative phenotype will therefore
tend to progressively take over the population. Note that such a process depends on
the presence of heterogeneity in a tumor cell population.
ENHANCING ANTIBODIES, IMMUNOSTIMULATION.
Enhancing antibodies:
Remember that most nucleated cells (other than
lymphocytes, which constitute an important exception) are relatively resistant to
complement-mediated lysis, and destruction of grafted tissue is largely the role of
cell-mediated immunity. However, if antibodies are present which can bind to neoantigens, they may effectively "hide" these antigens which might otherwise serve as
targets for T-cell-mediated killing, and thus "enhance" the survival of such cells.
Immunostimulation: Antibodies binding to tumor cells may also, in some cases,
actually stimulate cell growth, presumably through the generation of receptormediated proliferative signals. We have previously discussed the fact that antibody
directed against a cell surface receptor may act as an agonist with respect to the
normal signaling pathway for that receptor (see Chapter 19, AUTOIMMUNITY).

171

BLOCKING FACTORS: Soluble factors have been described in the serum of tumorbearing animals as well as in patients which can inhibit an existing immune response
from affecting the tumor. The best characterized blocking factors have turned out
to consist of circulating antigen-antibody complexes containing tumor antigens,
which may blind or divert the immune response in ways which are still poorly
understood.
IMMUNOSUPPRESSION: Tumor-bearing animals and human tumor patients often
exhibit a substantial degree of generalized immunosuppression, resulting directly or
indirectly from inhibitory cytokines and other substances secreted by the tumor
(Hodgkin's lymphoma is one notable example).
CONCOMITANT IMMUNITY: A mouse bearing a progressively growing tumor may
reject a new inoculum of the same tumor at a different site; this rejection is a
manifestation of concomitant immunity, an immune rejection reaction occurring at
one site, co-existing with the progressive growth of an antigenically identical tumor
elsewhere in the organism. This phenomenon illustrates the importance of tumor
mass in determining the ultimate outcome of a battle between a tumor and the
immune system; a small focus of tumor cells is more susceptible to immune killing
than a large and well-vascularized tumor.
ACTUAL AND POTENTIAL APPLICATIONS OF IMMUNOLOGICAL
PRINCIPLES TO THE CANCER PROBLEM:
IMMUNODIAGNOSIS
1) Tumor-associated antigens in serum. Detection of TSTA/TATA in serum to detect the
presence of tumors which are undetectable by conventional methods, or to measure the
overall tumor mass and its response to therapy. Assays for determining serum levels of
CEA and PSA (prostate specific antigen), for example, are widely used for both
screening and evaluation of the success of therapy.
2) Antibody-based therapy. Use of radioactively labeled antibodies to TSTA, to allow
detection and localization of otherwise invisible metastases. While the use of
radiolabeled antibodies for such purposes has been proposed and studied for decades, it
has not yet advanced to routine clinical application.
IMMUNOTHERAPY
1) Humoral: A number of monoclonal antibodies which target tumor antigens have been
FDA-approved for therapy, currently representing the most widely used form of
immunotherapy. Many are used as unconjugated naked antibodies, and their
effectiveness derives either from their ability to target tumor cells for destruction by
fixing complement and opsonization, or by serving as antagonists for cell surface
receptors important for cell proliferation or angiogenesis (growth of blood vessels).

172

Several decades of research have been devoted to the potential use of anti-tumor
antibodies conjugated with toxins (e.g. diptheria toxin, or the plant toxin ricin), or
highly radioactive isotopes, thus serving as targeting agents to deliver these toxic
molecules to tumor cells. Only two conjugated mAbs, both radioactively labeled, have
thus far been approved for human use as therapeutics, and are being studied for use in
imaging (to detect otherwise invisible centers of growth). A number of other
radioactive mAb conjugates are in various stages of development.
2) Cell-mediated (Lymphokine-Activated Killer [LAK] cells). Lymphocytes removed
from a tumor-bearing patient can be treated in vitro with lymphokines (e.g. IL-2), and
the resulting "activated" cells re-injected into the patient, in the hope that the tumor
cells will be killed by tumor-specific killer T-cells present in the transfused cell
population. These lymphocytes can be either obtained from the patients blood (which
hopefully contains some tumor-specific T-cells), or else extracted from a surgically
excised tumor (tumor-infiltrating leukocytes or TILs), presumably consisting of a
more highly enriched tumor-specific population. Results in humans have not been as
positive as in experimental animals, but attempts continue to developing this approach
into an effective therapeutic.

CHAPTER 23, STUDY QUESTIONS:

1.
2.
3.
4.

Define the concept of Immunological Surveillance. What is the strongest evidence

regarding its validity?
Define and give examples of the terms Tumor-Specific Antigen, and TumorAssociated Antigen.
Describe at least three mechanisms by which tumors are known to avoid immune
destruction.
Describe two possible approaches to using antibodies for anti-tumor therapy.

173

174

APPENDIX
1. PRECIPITIN CURVE
Lets examine a typical precipitin curve, in which we take a fixed amount of antigen
(say BSA), and add to it increasing amounts of antibody. We then measure the quantity of
the resultant precipitate.

Ag
Ab

Antigen
excess

Equivalence

Antibody
excess

Amount
of ppt.

Amount of added Ab

Figure A1-1
We find that the precipitate increases as we add more antibody (as expected), but that
after a certain point the total precipitate decreases again. This is due to the formation of
soluble complexes of two different kinds, one in the region of large antigen excess, the other
in region of large antibody excess; these are illustrated in the diagrams above the precipitin
curve. The antibody-antigen ratio near the peak of the curve is known as the region of
equivalence, which results in the maximum formation of large, insoluble complexes. Since
the binding of antibody with antigen is a reversible process, changes in the concentration of
either component can drive the reaction in one direction or the other.

175

176

2. LABELING OF ANTIBODIES (AND OTHER PROTEINS)

The ability to label proteins with radioactive markers led to the development in the
1950s of the RIA (RADIOIMMUNOASSAY), which provided an extremely sensitive
approach to detecting and quantitating antibodies and antigens. Other methods for labeling
proteins have also been widely used, including fluorescent dyes and enzymes.
Radiolabeling - RIA. While radioactive isotopes of iodine (125I and 131I), are not the only
radioactive markers used for proteins (3H tritium and 35S sulfur have also served, among
others), they have been the most widely used. A variety of different reaction schemes can be
used to attach either free iodide or small iodine-containing molecules to antibodies, resulting
in a highly radioactive but still biologically active antibody molecule. These radioactive
antibodies can be detected in extremely small quantities, either in precipitates or bound to a
solid state antigen. For many years RadioImmunoAssays (RIA, see APPENDIX 5) have
served as sensitive and widely used assays in both clinical and research settings. Beginning
in the 1970s, however, RIA's have progressively been replaced by equally sensitive and less
expensive Enzyme-Linked Immunoadsorbent Assays (ELISA, see below, and APPENDIX 5).
Fluorescent labeling - IF. Fluorescent dyes can be coupled to antibodies, and the binding of
such labeled antibodies to bacteria (or other particulate antigens), or to antigens in
histological sections, can be readily detected by fluorescence microscopy, a technique known
as IMMUNOFLUORESCENCE (IF). The two classical dyes used have been Fluorescein
(which fluoresces green) and Rhodamine (red), although a large number of newer dyes such
as Texas Red and others have since been added to this list.
One major feature of IF is its ability to determine the localization of antibodies and
antigens within histological sections; such studies were the first, for example, to demonstrate
that plasma cells were the major site of antibody production. More recently, IF has been a
major contributor to the power of Fluorescence-Activated Cell Sorters (FACS), machines
which can analyze and separate complex populations of cells, one cell at a time. The basis of
such separation is often the labeling of one or another cell type in a complex mixture with a
fluorescent antibody specific for that cell type, but cells can also be sorted on the basis of
size, DNA content and a variety of other physical and biological parameters.
Electron-dense labels. A variety of electron-dense markers can be attached to antibodies,
including Ferritin (which has an extremely high content of iron atoms) and colloidal gold
particles. The resulting antibodies can be visualized in the electron microscope just as
fluorescent antibodies can localized by light microscopy, but, of course, with much higher
resolution.
Enzyme labeling - ELISA. A variety of enzyme molecules can be coupled to antibodies,
which can then be sensitively detected by their associated enzymatic activity. The resulting
ELISAs (ENZYME-LINKED IMMUNOSORBENT ASSAY) are very widely used, and have
replaced RIA's in most modern applications. Commonly used enzymes have included
horseradish peroxidase, alkaline phosphatase, , -galactosidase, and, more recently,
luciferase.
ELISA's can equal or exceed the sensitivity of RIA's, and have the major advantage of
not requiring the use of radioactivity with its associated costs and hazards. The detection
method often involves the addition of a colorless substrate which the enzyme converts into a
colored (or fluorescent) product, and the results can be tabulated photometrically in fractions
177

of seconds per sample as opposed to minutes per sample for typical RIAs. Another detection
method which has become increasingly widely applied involves the use of enzymes which
cause their substrates to emit photons of visible light (luminescence).
Such
luciferase/luciferin based assays can be considerably more sensitive than the more traditional
colorimetric assays.
Enzyme-based histochemistry. Enzyme-labeled antibodies can also be used for microscopy
in the place of fluorescent-labeled ones (see IF, above). In such cases a substrate is used
which yields a visible and insoluble product which is deposited at the site of enzyme (and
therefore antibody) localization. The fact that a single enzyme molecule can yield many
molecules of product makes such immunohistochemistry potentially more sensitive than
immunofluorescence, and the resulting microscopic preparation is more permanent and easier
to photograph than fluorescent material.

178

3. OUCHTERLONY ANALYSIS
The Ouchterlony assay was widely used in myriad research and clinical contexts for
many years after its development in 1948 , although it has been largely superceded by other
assays which are more sensitive and suitable for quantitation (see APPENDIX 5).
Nevertheless, this technique provides a very useful tool for illustrating and clarifying the
principles of antibody heterogeneity and specificity, which is why we cover it in some detail
here.
In the Ouchterlony assay, the antigen and antibody solutions are placed in nearby wells
cut out of a thin layer of agarose, and allowed to stand for a few hours or a day or two.
During that time they diffuse toward each other, and where they meet they will form a visible
line of precipitation.
The pattern in which adjacent lines cross one another yields considerable information
about the antigenic relationships between different antigens. Lets illustrate this with a
pattern generated by a rabbit antiserum made against whole human serum, using three
purified protein antigens as targets, namely HSA (human serum albumin), BSA (bovine
serum albumin) and HTf (human transferrin). The well labelled Ab contains the
antiserum, and the resulting pattern is shown below.
nonidentity

HSA

HTf

HSA
X

partial
identity

complete
identity

BSA

HTf

Ab

BSA

Figure A3-1
Several conclusions can be drawn from the pattern of precipitation shown here:
1)

The antiserum (in the well labelled Ab) contains antibodies against all three
antigens, since each one shows a precipitin line.

2)

The lines produced by the two adjacent wells containing HTf join completely,
in a pattern of complete identity, or simply "identity." This indicates that the
antigens in the two wells (which in this case we know are exactly the same)
are antigenically indistinguishable by this antiserum. We cant tell from the
pattern whether the antiserum is detecting just one epitope or twenty, but we
do know it is detecting all of them in both wells.

3)

The two antigens HSA and HTf show a pattern of non-identity - the precipitin
lines cross each other without joining at all. This indicates the two antigens
are antigenically unrelated, they have no epitopes in common which are

179

recognized by this antiserum. This is not unexpected, since these two

molecules are not related in structure or function.
4)

The two antigens HSA and BSA show a pattern of partial identity - the lines
join together, but not completely; there is a "spur" of the HSA line over the
BSA line. This indicates that the two antigens are related, but not identical,
with respect to this antiserum. More specifically, it means that there are at
least two epitopes recognized by the antibody on HSA, one of which is also
present on BSA. This is illustrated in the two cartoons on the right, showing
HSA as bearing two epitopes ("X" and Y), while BSA bears only one of the
two (X).

We can be more precise by stating that there are two classes of epitopes that the
antiserum detects on HSA, and only one of the two classes is present on BSA. That is, what
is shown as epitope X in the cartoons might actually be, say, eight separate epitopes which
are all present on both HSA and BSA, while what is indicated as Y might be three epitopes
present
only
on
HSA.

180

4. ABSORPTION AND AFFINITY PURIFICATION

Affinity Purification of anti-DNP Antibody
Antigens and antibodies can be readily analyzed and purified by affinity
chromatography, a technique widely used in chemistry. When applied to antigen-antibody
systems, this procedure is often referred to as immunoadsorption.
Suppose we immunize a rabbit with DNP-OVA (where the protein ovalbumin is the
carrier for the hapten DNP), and collect the resulting antiserum, which contains antibodies
both to DNP and to the OVA protein carrier. We can separate the two kinds of antibody by
preparing an immunoadsorption column as illustrated in Figure A4-1. We start by covalently
coupling DNP alone, or DNP-BSA (DNP coupled to an unrelated carrier) to an insoluble
matrix such as cellulose, and place it in a column. We pass our rabbit antiserum over this
column; and the antibodies to DNP will bind to the DNP while everything else (including the
anti-OVA antibodies) flows through. We refer to this pass-through fraction as an
absorbed antiserum; the anti-DNP antibodies have been removed (absorbed) leaving
behind only anti-OVA antibodies, together with all the normal serum proteins, of course.

3M Urea
anti-DNP-OVA
antiserum

DNP
OVA

DNP

DNP

DNP

OV
A

DNP

D
NP

DNP

D
NP

DNP

D
NP

OV
A

DNP
DNP

DNP

OV
A

DNP
DNP

OVA

DNP

OVA

DNP

(+serum
proteins)

(only)

DNP

DNP
immunoadsorbent
column

Pass-through:

Eluate:

"Absorbed" antiserum

"Affinity-purified" Ab

Affinity Purification of anti-DNP Antibody

Figure A4-1

181

We can then recover the anti-DNP antibodies by releasing them from the column,
using either a solution of free DNP (which will compete with the bound DNP for the
antibodies' combining sites), or a denaturing agent such as urea or guanidinium salts. After
removing the eluting agent by dialysis, this affinity purified fraction contains pure antiDNP antibodies. If, on the other hand, we use a column containing insolubilized OVA, we
can recover a pure preparation of anti-OVA antibodies in the same manner.
Absorption and Affinity Purification of HSA antiserum
This general procedure can be used to modify the specificity of antisera in more subtle
ways. An antiserum prepared in a rabbit against human serum albumn (HSA), for example,
will cross-react with bovine serum albumin (BSA), since the two proteins are closely related.
Since distinct populations of antibodies bind to different epitopes, and binding of antibody to
its antigen is reversible, we can use affinity chromatography to separate and purify the
antibody populations as illustrated in Figure A4-2.
We begin with an antiserum generated by immunizing a rabbit with purified human
serum albumin (HSA). When we analyze this antiserum A by Ouchterlony (bottom left)
we see that there are two distinct populations of antibody present, one which binds to both
HSA and BSA (bovine serum albumin), and another which binds only to HSA.
We then pass this antiserum over a column containing covalently bound BSA. All of
those antibodies capable of binding to BSA will be retained on the column, and everything
else will pass through, yielding preparation B, which we refer to as an absorbed
antiserum (more specifically, HSA[absBSA]. Ouchterlony analysis of this fraction
(bottom center) shows the presence of a single population of antibodies which bind only
HSA (since all those antibodies capable of binding BSA have been retained in the column).
We can now recover the antibodies bound to the column by washing it with 3M
guanidium. We can remove the guanidium by dialysis and test this affinity purfied
antibody fraction by Ouchterlony (bottom right). The resulting pattern tells us there is a
single population of antibodies which bind to both HSA and BSA.
All three of these preparations can be useful in research and clinical settings, and it will
be instructive to distinguish their properties.
Whole antiserum
Whole antiserum can be used without further treatment in many assays, such as
precipitation (as we have seen), agglutination, etc., so long as cross-reactivity between related
molecules is not important. For example, a researcher might want to detect the presence of
HSA in a set of samples which do not contain BSA or other cross-reactive proteins; in such a
case absorption and affinity purification are unnecessary. While only a small percentage of
the total protein in whole serum is specific antibody, the presence of other proteins does not
interfere in most assays.

182

3M guanidinium

HSA

BSA
HSA

"A"

BSA

HSA

BSA

H
SA

H
SA

BSA

H
SA

HSA

BSA

Original
antiserum

H
SA

BSA
HSA

BSA

H
SA

BSA

BSA

H
SA

HSA

HSA

HSA
(+serum
proteins)

(pure Ab)

Pass-through:
Absorbed antiserum

Eluate:
Affinity-purified Ab

"B"

"C"

HSA

BSA

HSA

HSA

BSA

Partially
cross-reactive

Non-cross-reactive,
"Monospecific"
(but not monoclonal)

Completely
cross-reactive

ABSORPTION AND AFFINITY PURIFICATION OF HSA ANTISERUM

Figure A4-2

183

 Absorbed antiserum
If a forensic immunologist needs a reagent to determine if a blood stain is human or
animal, then absorption of the original antiserum becomes necessary. Fraction B in our
example will do the job in many assays (Ouchterlony, for example), showing a positive
reaction only with human albumin.
Affinity-purified antibody
There are a number of applications, however, for which affinity purification of antibody
is either required or highly desirable. One example would be in developing an RIA or ELISA
(see APPENDIX 5) to measure mammalian serum albumin, In this case we need to be able
to label only the antibody molecules (with radioactivity or enzyme), and it is impossible to do
this in the presence of the large excess of irrelevant proteins in whole or absorbed antiserum.
Our affinity-purified fraction C, however, contains only specific antobody molecules. This
fraction could be labelled and would provide a reagent to detect mammalian serum albumin
by RIA or ELISA, reacting equally well with human and bovine albumin (and presumably
other related species). [QUESTION: How would you go about preparing an affinity-purified
antibody for use in an RIA specific for human serum albumin?]
Another example would be for the therapeutic use of antibodies which are to be injected
into a patient. The vast majority of irrelevant protein in either whole serum or absorbed
antiserum would expose the recipient to a far greater risk of SERUM SICKNESS than
affinity-purified antibody (see Chapter 5, COMPLEMENT)

184

5. RADIOIMMUNOASSAY - RIA
Principles of the assay
The development of the radioimmunoassay in the 1950s revolutionized the field of
immunochemistry in both research and clinical applications. The RIA relies on the one
property of antibody function which is a universal feature of all antibodies, namely the ability
to bind specifically to an antigen. The degree of binding can be very accurately quantitated
by making one of the components radioactive, in this example the antibody.
Our sample assay is designed to determine the concentration of insulin in some
unknown samples (blood serum, urine, cell culture supernatants, etc.). We begin by coating a
glass or plastic tube with a set quantity of insulin in such a way as to be permanently fixed in
place. As illustrated in the top row of Figure A5-1, we then add a fixed amount of antiinsulin antibody which we have tagged with the radioactive isotope 125Iodine. Most of the
antibody will bind to the insulin stuck to the tube, and we then rinse the tube with saline to
wash away all unbound material. We can now measure the amount of radioactivity in the
tube, representing the amount of bound antibody; we determine that there are 5,000 counts
per minute of radioactivity in the tube, which represents maximum binding.
We now carry out the same exercise (shown in the second row of figures), but this time
we add a large amount of soluble insulin (inhibitor) together with the radioactive antibody.
In this case, the antibody can bind either to the fixed or soluble insulin, and since there is
more soluble than fixed insulin, most of the antibody will bind to the soluble form. This time
when we rinse the tube with saline, we wash away most of the radioactive antibody together
with the soluble insulin. The small amount of radioactivity which still remains behind,
namely 400 counts per minute, represents some antibody (or contaminating radioactive
material) sticking nonspecifically to the assay tube, or background binding.
This defines the principles of the RIA specific binding of antibody to antigen can be
sensitively measured, and this binding can be inhibited by the addition of soluble antigen. A
small amount of soluble antigen will inhibit binding only slightly, a large amount will inhibit
more, and we can use this relationship to determine the concentration of inhibitor in any
unknown sample..
Inhibition curves
If we have a series of experimental tubes to which we add increasing amounts of inhibitor,
we can generate inhibition curves such as those shown at the bottom of Figure A5-1. One
common practice is to make serial dilutions of the inhibitor from an original stock (say 1:2,
1:4, 1:8, etc.) and add a fixed volume of each dilution to each assay tube. There are a variety
of methods for plotting such data, and the one we will use is shown at the right, plotting
increasing inhibition (from 0 to 100%) on the Y-axis, versus increasing dilution of inhibitor
on the X-axis. Using our standard volume of a low dilution (i.e. a large amount) of inhibitor,
we see 100% inhibition. As we increase the dilution (i.e. decrease the amount) of the
inhibitor, we see less and less inhibition, until the curve finally flattens out at 0% inhibition.
In this example we see 50% inhibition at a dilution of about 1:40, and we refer to this curve
as exhibiting a titer of 1:40

185

RadioImmunoAssay (RIA)
+

125

wash

I-Ab

Count:
5,000 cpm

ABOVE: Maximum
binding
BELOW: Maximum
inhibition
("background" binding)

Insulin bound
to tube

125

wash

I-Ab

Count:
400 cpm

+ soluble
insulin

ASSAY: Fixed quantity of bound Ag (insulin= )

125
Fixed quantity of I-Ab (= )
Variable quantity of soluble inhibitor (insulin=

Maximum total binding = 5000 cpm.

Maximum inhibition yields 400 cpm bound (=100% inhibition).
Maximum specific binding = 4600 cpm (=0% inhibition).

4
3
2
1
2

16

32

amount of inhibitor (ng)

5
100

percent inhibition

net CPM (x10 -3) bound

net CPM (x10 -3) bound

VARIOUS WAYS TO PLOT RESULTS...

4
3
2
1
1:10

20

40

80

dilution of inhibitor

160

80
60
40
20

1:10

20

40

80

dilution of inhibitor

Note: Display and interpretation of data are identical for

RIA and the more commonly used ELISA
Figure A5-1

186

160

How do we use this assay to determing the concentration of insulin in an unknown

sample? Consider the inhibition curves shown below in Figure A5-2, where our standard
insulin solution shows a titer of 1:40, while the unknown sample has a titer of 1:160. We
conclude that the unknown sample (in its original, undiluted form) contains insulin at a
concentration four times greater than our standard. If our standard insulin stock has a
concentration of 20 pmol/L (picomoles per liter), which we can determine by looking at the
manufacturers label, then the unknown sample must be at 80 pmol/L.

percent inhibition

100
80

unk

std

60
40
20

1:10

20

40

80

160

320

dilution of inhibitor

Figure A5-2
ENZYME-LINKED IMMUNOADSORBENT ASSAY (ELISA) versus RIA
As mentioned in APPENDIX 2, RIAs have been superceded in many of their
applications by Enzyme-Linked Immunosorbent Assays (ELISA), which can be as sensitive
than RIAs while avoiding the hazards and expense associated with the use of radioactive
materials. We have introduced this class of binding assays using the RIA as the primary
example because it is conceptually simpler, but the principles of the ELISA are the same as
those described above, and the data are plotted and interpreted in precisely the same manner
as RIAs. In fact, there is no way to know simply by examining an inhibition curve whether
it was generated in an ELISA or RIA.

187

188

6.

ANALYSIS OF THE ANTIBODY COMBINING SITE BY

EQUILIBRIUM DIALYSIS

Equilibirum dialysis was the first technique used to determine the thermodynamic
properties of antigen-antibody binding. It revealed that this reaction was a non-covalent and
reversible reaction, that the basic IgG-like unit had exactly two combining sites, and it
determined the values of the equilibrium constant (affinity) of the binding reaction.
Requirements for the Analysis
1)

We must have a pure preparation of antibody, not just a gamma-globulin

preparation or an ammonium sulfate fraction; the reason for this is that we must
know the molar concentration of antibody. We must, therefore, also know the
molecular weight of IgG antibody, which is about 150,000.

2)

We must carry out the analysis using a monovalent, low molecular weight antigen
such as a hapten. The reasons for this are twofold: first, the hapten must be small
enough to pass across a dialysis membrane (the antibody, of course, is too large);
and second, it must be monovalent to fulfill the assumptions we will make regarding
independent binding to multiple antibody combining sites on the antibody.

Dialysis Apparatus
The apparatus consists of two small chambers connected only through a semipermeable
membrane, as diagrammed below. In one chamber we place a known amount of total
antibody (AbT), in the other a known amount of total hapten (HT); the starting condition is
shown on the left of the figure below.
membrane

AbH
Ab T

HT

H'

Ab H'

Start

End
Figure A6-1

The two chambers are then allowed to reach thermodynamic equilibrium, which may take
one or two days. During that time, the hapten will diffuse across the membrane, and some of
it will be bound by the antibody in the other chamber, as indicated on the right side of the
diagram.

189

The key point is that the concentration of free hapten on both sides of the membrane will
be the same at equilibrium, indicated as H'. In addition, there will be free antibody (Ab) and
antibody bound to hapten (AbH).
Since HT = 2H' + AbH, we can determine the value of AbH (HT is known, and H' is
measured). Then, since AbT = Ab + AbH, we can also determine Ab, or the concentration of
free antibody.
Analysis
The experimental data are collected by measuring the concentration of free hapten at
equilibrium, having started with a series of different initial hapten concentrations. The basic
rationale is that at any given hapten concentration, a higher affinity antibody will bind a
higher proportion of hapten.
We can rewrite the reaction as a series of reactions, as follows:
Ab

+ H

AbH

K1

AbH + H

AbH2

K2

AbH2 + H

AbH3

K3

etc...
Each reaction represents the binding of one additional hapten molecule to an antibody
molecule, up to some unknown maximum (which depends on the total number of sites
actually present on each antibody molecule); each reaction is governed by an equilibrium
constant, K1, K2, etc.
Assumptions: (1) All binding sites on a single Ab molecule have the same
affinity, and (2) binding of H by different combining sites on the same antibody
molecule is independent; i.e. binding of H to one site does not help or hinder
binding of H to another. If these assumptions are correct, then all the reactions
above can be represented by a single reaction with a single equilibrium constant,
where "S" represents a single antibody-combining site.
S

+ H

SH

By the definition of the equilibrium constant:

Keq

[SH]
= -------[S][H]

190

Keq

This can be shown to be equivalent to the following equation:

r
---- = Kn - Kr
c
where:

r
c
K
n

=
=
=
=

moles H bound per mole of antibody

concentration of free H, or [H]
Keq, defined above
number of binding sites per Ab molecule (or [S] / [Ab])

This equation is that of a straight line of the form Y = b + mX, where Y is replaced with
R/c, the constant b is replaced with Kn, and mX is replaced with -Kr. If we plot our data, we
expect to get the straight line labeled "theoretical", as shown below:

r
c- = Kn - Kr

theoretical

r
c

slope = -Keq,
where Keq is affinity

actual

X-intercept = n,
no. of sites per Ab
(if r/c =0, then r=n)
K av

Figure A6-2
Two important values can be determined from this straight line; first, the slope (actually
the negative slope) gives us Keq, the equilibrium constant, which we have defined as the
affinity; second, the X-intercept gives us the value of n, the number of combining sites per
antibody molecule. In this graph, the theoretical line is drawn so that it crosses the X-axis at
a value of 2.
So far we have been dealing with a theoretical straight line; but when we actually carry
out such an analysis with the rabbit anti-DNP antibodies and plot the results, we get not a
straight line, but a curve such as shown in the figure above (labeled "actual").

The curvature indicates that the antibody preparation is heterogeneous with respect
to affinity. There are antibodies with high and low affinities, and everything in
between. (The degree of curvature can, in fact, be used as a measure of the degree of
antibody heterogeneity.)

191

If we extrapolate the curve to the X-axis, we can estimate a value for n of 2; this
means that there are two combining sites for DNP on each IgG antibody molecule.

If we carry out this analysis with homogeneous IgG anti-DNP antibodies (for instance
with a human DNP-binding myeloma, or an anti-DNP hybridoma), we do get a
straight line with an X-intercept of 2, and a slope which defines the affinity of that
particular antibody (as shown by the "theoretical" line in Figure A5-2).

The shape of the curve can be understood intuitively. At very high values for r/c, that is,
at very low levels of free hapten, those antibodies with high affinities will more effectively
compete for what little hapten is present; therefore, binding by the high affinity antibodies
predominates and the slope of the curve is high (the curve is steep) in that region. At low
values of r/c there is plenty of hapten to go around, and the lower affinity antibodies can
contribute a larger proportion of the binding; the curve is shallower in this region,
representing a lower overall affinity.
One can define the "average affinity" (Kav) of a heterogeneous antibody population as
the slope of the curve where r =1, that is, at the point where half of the available sites are
occupied; this is indicated as Kav in Figure A6-2.
To better understand the relationship between these experimental data and the "reallife" functions of antibodies, lets define two terms which are distinct, but easily confused,
namely AFFINITY and AVIDITY.
AFFINITY: The strength of binding of a single antibody combining site with its
epitope; the equilibrium constant of the binding reaction. Conventional antibodies
have measurable affinity constants in the range of l04 to 1010; higher and lower values
certainly exist in nature, but are generally difficult to measure accurately. Many
methods other than equilibrium dialysis exist for determining affinity constants, but
they are all measures of the interaction of a single combining site with its antigen.
AVIDITY: The strength of binding of a multivalent antibody to a multivalent
antigen; a measure of the ability of antibodies to form stable complexes with their
antigens. The avidity of an antibody depends not only on the affinity, but also on the
valency of both antibody and antigen, and on various physical properties of both.
Avidity does not have a standard thermodynamic definition, although it can be
defined in a relative and ad hoc manner within a specific given experimental context.

192

The difference between affinity and avidity may be illustrated by the following diagram,
showing the two-step binding of a bivalent antibody to a multivalent antigen molecule:

Ag

k2

k1

Ag

Ag

Figure A6-3
The first step of binding occurs at a rate (shown as k1 above) determined by the affinity
of the combining site for its epitope (note that we are now discussing reaction rates, not
equilibrium constants). The second step, however, occurs with a much higher rate constant,
k2; once the first site has bound to the antigen, the likelihood of the second site binding is
much higher, since the second site is held in close proximity to another epitope. This
cooperativity between binding sites on an antibody molecule makes the effective binding to
antigen much stronger than would be predicted simply by the equilibrium constant; avidity is
the term used to describe this greater strength of binding.
Antigens in nature are often polyvalent and highly repetitive structures. The strength of
binding by antibodies to such antigens (microbial cell wall components, for instance) is much
higher than the antibodies' affinities would predict, because of the effect of cooperative
binding. This effect becomes even more important in the case of polymeric IgA, and of IgM
which can have up to 10 combining sites in a single antibody molecule (although all of the
sites may not necessarily be capable of binding antigen simultaneously).

193

194

7.

CROSS-REACTIVITY

Weve seen examples of serological cross-reactivity illustrated by Ouchterlony

precipitation patterns. Adding the use of RIA allows us to distinguish two different modes of
cross-reactivity.
Figure A7-1 shows the kind of cross-reactivity we have already seen, due to the presence
of multiple distinct epitopes which may be unique or shared between antigens. At the left we
see the structural basis for such cross-reactivity - antigen A bears two epitopes labeled X
and Y, antigen B also has epitope X, together with a different epitope Z. Using an
antiserum prepared against antigen A shows the Ouchterlony pattern we recognize as
partial identity - a shared line of precipitation due to the presence of epitope X on both
antigens, but with a spur of A over B because epitope Y is present only on antigen
A.

RIA Inhibition:
Antigens:

("A" bound to plate,

125I-labelled anti-A)

Ouchterlony:
100

% inhibition

A
X

A
50

B
0
10 20 40 80 160
dilution of inhibitor

Cross-Reaction due to Multiple Distinct Epitopes

Figure A7-1
If we carry out an analysis by inhibition of RIA (at the right), we also see evidence of
partial cross-reactivity. The assay consists of antigen A bound to the tube incubated with
affinity-purified radiolabeled antibody to the same antigen. Thus we have labeled antibodies
to both epitopes X and Y. If we inhibit with a sufficient amount of purified antigen A,
we can inhibit 100% of the antibody binding, as expected. However, when we use antigen
B as inhibitor, we find that no matter how much we add we can only inhibit 50% of the
binding. This is because some of the anti-A antibody (50% in this example) is directed
against epitope Y. Since antigen B doesnt have this epitope, it cannot inhibit this
binding. (Note that the putative presence of epitope Z on antigen B is irrelevant in our
example, since there are no antibodies to this epitope in the system.)

195

Lets consider another situation, where two antigens each have one epitope which is
related, but not quite identical, to that on the other (indicated as X and X below). In this
case the pattern we see in Ouchterlony is one of complete identity, despite the fact that we
expect antibodies to epitope X to bind less strongly to X. But as long as the antibodies bind
X sufficiently strongly to cause precipitation, Ouchterlony analysis will not distinguish the
two antigens - they appear identical.

RIA Inhibition:

100

A
A

("A" bound to plate,

125I-labelled anti-A)

Ouchterlony:

% inhibition

Antigens:

X'

50

B
0
10 20 40 80 160 320
dilution of inhibitor

Cross-Reaction due to Related Epitopes

Figure A7-2
When we analyze these antigens by RIA, however, we see something different. Antigen
B can, indeed, completely inhibit the binding of anti-X to epitope X, but since its own
epitope X is recognized by the antibody with lower affinity (therefore resulting in lower
avidity), it has a more difficult job inhibiting. This is reflected in the inhibition curve for
antigen B which shows a shallower slope than that for A, and a greater amount of antigen
necessary to finally achieve complete inhibition.
Recognizing this property of cross-reactive antigens is of practical importance in
interpreting RIA results. One explicit assumption in using RIA to determing the
concentration of an antigen (as, for example, insulin in APPENDIX 5), is that the antigen in
the unknown samples is identical to that in the standard. If we see an inhibition curve which
is not identical in shape to that of the standard, then we know that the material in the
unknown sample is structurally different from that in the standard, and we cannot draw any
conclusion regarding relative concentrations.

196

8. COMPLEMENT FIXATION ASSAY (see Chapter 5, COMPLEMENT)

Many different methods exist to measure the levels and degree of activation of the various
complement components, most of which are used only in highly specialized labs devoted to
complement research. The classical Complement Fixation (CF) assay, however, has long
been widely used by biologists of all kinds as a simple and sensitive way to detect and
quantitate a wide variety of antigens and antibodies. While it is still sometimes used in both
research and clinical situation, many of its applications have been replaced by ELISA or other
more sensitive and quantifiable methods.
Lets illustrate the principles by constructing a complement-fixation assay to detect a
hypothetical viral antigen (VA) in several unknown serum samples.
We begin by making "sensitized" erythrocytes, which are simply RBC coated with
anti-red blood cell antibodies; we call these EA (Erythrocyte + Antibody). In the
absence of complement these cells remain intact, but they will be lysed if exposed to
any source of complement; these EAs are our indicator cells.
Next, we prepare a standard source of complement, usually fresh guinea pig serum
diluted appropriately. The dilution is such that it contains just enough complement to
completely lyse a standard amount of EA.
We also prepare a standardized antibody to the antigen, by immunizing a rabbit with
purified VA. Before use the rabbit antiserum is heat-treated (56 C for 30 min) to
remove its inherent complement activity.
We then take a series of test samples (for example, heat-treated serum samples from
various patients), mix them with a standard amount of our antibody, and incubate the
mixture for an hour (allowing the antibodies to react with the antigen, if it is present).
To this mixture we then add a standard amount of the guinea pig serum as a source of
complement, and after a short time (to allow the AgAb complexes to interact with the
complement) we add a standard amount of the sensitized erythrocytes (EA). We
incubate the samples at 37 C for a half hour and examine them visually (or
spectrophotometrically) for lysis; a suspension of intact RBC is cloudy, but it
becomes transparent when the cells are lysed.
The protocol would be as shown on the table below (where "unk" refers to an "unknown" or
test sample).
Tube 1 tells us that the erythrocytes have been properly sensitized and that our guinea pig
serum (as a source of complement) is effective. Tube 2 shows us that the antibody alone
does not interfere with the complement-dependent lysis of the sensitized RBC. In tube 3, the
antibody will form complexes with the added VA protein, and these complexes will consume
("fix") the complement; therefore, there will not be enough complement left to subsequently
lyse the indicator cells. Repeating this test with two unknown samples in tubes 4 and 5 tell
us that sample No. 2 must contain the viral antigen VA, while sample No. 1 does not. We
carry out one additional negative control, by re-testing unk sample number 2 in the absence
of specific antibody. As expected (tube #6), no complement is fixed, and the indicator cells
are lysed.

197

Tube
No.

Sample

Result

Interpretation

1)

(saline alone

C)

EA

lysis

neg control

2)

(Ab + saline

C)

EA

lysis

neg control

3)

(Ab + VA

C)

EA

no lysis

pos control

4)

(Ab + unk No. 1

C)

EA

lysis

no VA present

5)

(Ab + unk No. 2

C)

EA

no lysis

VA present

6)
(
unk No. 2
+
C) + EA
lysis
neg control, OK
--------------------------------------------------------------------------------------------------------7)
(Ab + unk No. 3
+
C) + EA
no lysis
VA present??
8)

unk No. 3

C)

EA

no lysis

anti-complement
activity

However, the last two tubes (7 and 8) illustrated a possible complication in interpreting these
results. Tube 7 shows no lysis, suggesting the presence of viral antigen in sample #3 (as was
the case with sample #2). However, when we run this same sample again in the absence of
specific Ab, there is still no lysis, although we expect the same result as in our negative
controls (tubes 2 and 6). This shows that serum #3 inhibits complement activity by itself, a
property referred to as "anti-complementary;" we therefore can say nothing about the
presence or absence of viral antigen in this sample unless we find a way to remove this
activity. Knowing of this possibility, it is clear that we must repeat this control test for every
positive sample (as we did for sample #2 in tube 6) to confirm that the positive result is not
the result of anti-complementary activity. (Such activity might be due to the presence of
unrelated immune complexes in the serum, the presence of antibodies which happen to
recognize components of the guinea pig complement, or the presence of drugs which inhibit
complement.)
Despite these cautions, complement fixation assays are still used in a variety of specialized
applications. They can provide very sensitive tests for any antigen to which an antibody is
available, or, by changing the variables, can be used to test for the presence of antibody to a
standard antigen. They can also be used to quantitate the level of antigen (or antibody) by
carrying out the tests with varying dilutions of the unknown, and accurately determining the
level of lysis of the indicator cells by colorimetric means.

198

9.

ANALYSIS OF KAPPA-CHAIN GENE REARRANGEMENTS

BY SOUTHERN BLOTTING

Determining the nature of gene rearrangements in tumor cell populations can be a

powerful tool to aid in diagnosis of a variety of cancers. Lets examine how the Southern
blotting technique can be applied to examining kappa-chain rearrangements in cells of the Bcell lineage.
At the top of Figure A9-1 (#1, germline) is a schematic representation of the genomic
configuration of the human kappa-chain gene complex, showing the constant region with its
five associated J-segments at the right, and one of the many V-regions at the left. Also
indicated are the positions of EcoRI sites, short DNA sequences which are specifically
cleaved by this restriction enzyme. If this un-rearranged configuration of DNA were treated
with EcoRI, the only fragment containing C sequences would be the 2.5kb fragment
indicated.
2.5kb
E

V36

E J1J2J3J4J5

#1: germline

*
E=

EcoRI site

3kb
E

J3J4J5

V36

*
#2: rearranged (V36/J3)

... Vk143
2kb
E

E J2J3J4J5

V143

*
#3: rearranged (V143/J2)

Germline and Two Rearranged Configurations of Human Ck

Figure A9-1
If we look at the configuration of this region of DNA in B-cells which have rearranged
their -chain complex, we might see any one of a large number of different results, two of
which are illustrated in the figure. In the first case (line #2), the rearrangement has removed
the DNA containing the EcoRI site just to the left (3 side) of J1, so that the the C region
now happens to be on an EcoRI fragment 3kb in length. In the second case (line #3), the Vk
gene segment that was used happens to have an EcoRI site near its 5 end, so the C gene is
now on a fragment 2kb long. The hundreds of different possible Vk/Jk rearrangements will
yield many different results; in unrearranged DNA the Ck gene segment will exist only on a
2.5kb EcoRI fragment, in rearranged DNA it could end up on any sized fragment.
How can we use this information to analyze cell populations and help diagnose B-cell
tumors? Lets use the Southern blotting technique to examine DNA from various sources.
We extract DNA from a cell population, treat it with EcoRI, then separate the resulting
fragments according to size by electrophoresis in an agarose gel. We then transfer the
separated fragments from the gel onto a sheet of nitrocellulose, incubate the sheet with a

199

radioactive DNA probe which will hybridize with its complementary sequences, and we can
determine the location of the radioactivity by autoradiography.
A schematic showing the results of such an experiment is shown in Figure A9-2.
Various samples of EcoRI-treated have been separated in an electric field from top to bottom,
the smaller fragments moving the farthest. The size of fragments at each position is indicated
on the left.

Kb

Lane #:
3
2

4
#1: germline band

3
2.5

#2,3: single monoclonal

rearrangements
(+ unrearranged allele)
#4: polyclonal
rearrangements
(+ unrearranged allele)

Southern Blot Hybridized with C Probe

Figure A9-2
Sample #1 consists of DNA from a cell population which has not rearranged Ck, for
example skin, liver, kidney, brain, or any source other than cells of the B-cell lineage. Since
no rearrangement has taken place, the only place we see C-hybridizing sequences is at the
position of the germ-line 2.5kb fragment.
Samples #2 and #3 have been prepared from two different plasmacytomas, Ig-producing
human tumors. Since cancers are monoclonal in their origin, all the plasma cells in a
particular tumor have undergone exactly the same rearrangement(s). In this case, each tumor
has rearranged one of its two copies of C, but not the other, resulting in the appearance of
two bands in each lane, the familiar 2.5kb germ-line band (non-rearranged) and a single,
unique new band (rearranged).
Sample #4 is what we would expect with DNA prepared from a population of normal Bcells, taken (for example) from spleen, lymph node or blood. Each B-cell in the population
has rearranged at least one C gene, some have rearranged both. But each cell has rearranged
in a unique fashion, resulting in a heterogeneous mix of many different fragments which
contain the C segment; this is seen as the smear of radioactivity in the right-hand lane.
Since some of the B-cells have rearranged only one of their two alleles, all the nonrearranged alleles contribute to the visible band at the germ-line position.
We can apply this approach, for example, to help distinguish a B-cell tumor from more
benign forms of lymphoid hyperplasia in a particular patient. In the case of a tumor,
Southern blotting will show a pattern indicating monoclonal rearrangement (one or two
discrete non-germline bands as in Lanes #2 and #3 above), whereas a pattern of polyclonal
rearrangement (as in Lane #4) would indicate non-cancerous form of B-cell hyperplasia. The
same approach can be used to help in the diagnosis of T-cell tumors, in that case using probes
specific for T-cell receptor sequences rather than immunoglobulins.

200

10. GENETICS OF INBREEDING

"...the development of inbred strains has constituted probably the greatest advance in all
cancer research." (W.E. Heston, 1963)
"The introduction of inbred strains into biology is probably comparable in importance with
that of the analytical balance into chemistry." (Hans Grneberg, 1952)

IMPORTANCE OF INBRED STRAINS OF MICE

One of the major obstacles in biological research of all kinds is the genetic variation inherent
in most natural populations of organisms. Individual humans as well as experimental
animals vary widely in their physiological properties and biological responses, and much of
this variation (although not all of it) is genetically determined. One particular manifestation
of such variability was recognized by cancer biologists more than a century ago, namely the
inability to reliably transfer growing tumors from one mouse in a laboratory colony to
another. It was also recognized that occasional successful transplants generally involved
closely related animals, and depended on their degree of genetic relatedness; this is the basis
for the special relevance of inbreeding to cancer research alluded to by Heston, above.
Shortly after the turn of the twentieth century, the geneticist C.C. Little developed the earliest
inbred strains of mice, which for the first time provided genetically uniform mammals for
research. The importance of such animals cannot be overstated, as attested to by the two
quotations above. While their development was originally stimulated by their need in tumor
transplantation, inbred mice have provided essential tools for research in genetics, physiology
and behavior, as well as in all aspects of immunology. They have provided animal models
for the study of many human diseases, including amyloidosis, polyuria, diabetes, muscular
dystrophy and many others. In an immunological context, the use of inbred strains has been
indispensible for the success of adoptive transfer systems, of which we have discussed
several examples, as well as for the elucidation of the biological roles and chemical structure
of products of the Major Histocompatibility Complex (MHC).
PRODUCTION OF INBRED MICE: BROTHER-SISTER INBREEDING
We can illustrate the basic features of the process of inbreeding, and of the resulting
inbred strains, by the protocol originally used by Little. Lets begin with two randomly
chosen mice taken either from the wild, or, as was more commonly done, from the colonies
of mouse "fanciers" who kept them for amusement or for sale. The two mice we start with
may or may not be related, but in either case will differ in the allelic forms of genes they
carry at many genetic loci.
Lets consider a randomly chosen genetic locus which has two allelic forms, which we
name A and a. NOTE: We do not have to know what this locus is, nor do we need to be able
to distinguish its genotypes in order to understand the analysis that follows.

201

Suppose we happen to start with two individuals heterozygous at this locus; the
resulting cross can then be diagrammed as follows:
Parents:
Offspring (F1)
Frequencies:

Aa

X Aa

AA Aa aA aa
AA:
Aa:
aa:

25%
50%
25%

Using simple Mendelian genetics, we can see that there are three possible genotypes
for the offspring of such a cross, and their frequencies are given above. If we then randomly
choose two of these offspring to mate for the next (F2) generation, there are six possible
crosses which may result. These six possibilities, together with the probability of having
chosen each, are shown below:
probability of cross
next generation
---------------------------------------------------------------------------------AA
X
AA
1/16
offspring all AA, p=1/16
AA
X
Aa
1/4
AA
X
aa
1/8
offspring varied
Aa
X
Aa
1/4
Aa
X
aa
1/4
aa
X
aa
1/16
offspring all aa, p=1/16
For example, the probability of our having randomly chosen two animals of genotype
AA is simply 1/4 X 1/4, or 1/16 (= 6.25%). The probability of the second combination (AA
X Aa) is 1/4 X 1/2 X 2 = 1/4 (we multiply by 2 to take into account the fact that there are two
possible choices which will yield this cross, namely AA X Aa, and Aa X AA; the same
situtation holds for all crosses in which the two individuals differ from one another).
The first and last possibilities in this list (AA X AA, and aa X aa) are special cases:
for both of these crosses all offspring will be identical to their parents. If we add the
probabilities for these two cases, we find that there is a 12.5% likelihood (1/16+1/16= 1/8)
that we will happen to have picked a pair of identically homozygous animals in this
generation. If this happens, regardless of which of the two possibilities we have chosen, all
subsequent generations will be homozygous at this locus, as long as we keep to the
brother-sister mating scheme.
This is equivalent to stating that 12.5% of all genetic loci will be homozygous in this
generation, based simply on chance. Adhering to the brother-sister mating scheme ensures
that once we happen to have chosen two animals homozygous for a given locus, no new
heterozygosity can ever be introduced. Thus, as we continue this mating scheme, generation
after generation, the resulting animals progressively approach a condition of uniform
homozygosity at all loci. It is important to recognize that this analysis does not depend on
our knowing anything about any of the genetic loci involved, nor does it rely on any
deliberate selection on our part.

202

After 20 generations of such inbreeding, the overall homozygosity is about 98.5% (the
"coefficient of inbreeding" = 98.5%), and we are entitled to call this line an inbred strain.
Brother-sister inbreeding is continued throughout the maintenance of all inbred strains, and
the overall level of homozygosity continues to increase, until that point at which it is
counterbalanced by the appearance of spontaneous mutations (at a very low rate).
The oldest inbred strains of mice and rats date back to about 1909, and many have
gone through hundreds of generations of brother-sister mating since then. Commonly used
strains of mice include C57Bl/6, C3H, BALB/c, DBA/2 and AKR. There exist well over 200
strains (and many substrains) of inbred mice and scores of inbred rat strains, as well as inbred
strains of rabbits, chickens, hamsters and guinea pigs.
In addition to the random brother-sister mating scheme described above, additional
selection can be imposed during the early stages of inbreeding for the establishment of
desired traits. This has resulted in the production of inbred strains of mice and rats with high
(or low) incidence of various tumors, dental caries, blood pressure, and many other
physiological or behavioral characteristics.
PROPERTIES OF INBRED STRAINS
Inbred strains have several properties which distinguish them from conventional
laboratory or wild animals.
1)

All members of a particular strain are genetically uniform (or "isogenic"); they are as
identical to each other as human identical twins.

2)

Furthermore, they are homozygous at all loci (unlike human twins).

3)

Inbred strains breed true; that is, if two members of an inbred strain are mated, their
offspring are also members of that strain; they are uniformly homozygous and identical
to all other individuals of that strain.

Individuals of inbred strains are often (although not always) fairly delicate and of
limited fertility, which tends to make them difficult to maintain and expensive to use.

PROPERTIES OF F1 INBRED CROSSES

Individuals of two different strains are often mated to produce F1 crosses. For
example, a female C57Bl/6 mouse might be mated with a male of strain DBA/2, yielding an
F1 cross named (C57Bl/6xDBA/2)F1, or, more briefly, "B6D2F1".
Such F1 crosses between inbred strains posess important and useful properties:
1)
2)

Like inbred strains, they are also genetically uniform ("isogenic"); they are homozygous
at all those genetic loci for which the parents happen to be identical.
However, they are not homozygous at all loci, unlike their inbred parents - in this respect
they are more similar to human identical twins than are their inbred parents. In fact, they
are heterozygous at all those loci (but only at those loci) for which the two parental
strains carry different alleles.

203

3)

The property of being an F1 hybrid does not breed true; a cross between two B6D2F1
mice does not yield another B6D2F1 mouse, but a mixture of genetically heterogenous
offspring. A regular supply of F1 hybrids therefore requires the continuous maintenance
of both inbred parental strains.

4)

The products of F1 crosses between inbred strains are often very hardy and of higher
fertility, exhibiting what is called "hybrid vigor". This generally makes them easier and
less expensive to use, and such animals are widely used for many experiments which
require genetic uniformity, but not homozygosity.

204

11.

MIXED LYMPHOCYTE REACTION (MLR)

Allelic differences in the MHC complexes between two individuals can be detected
by transplant rejection, and by specific antibodies to the transplantation antigens. They can
also be detected by the MLR.
If lymphoid cells (spleen cells, for instance) from two different strains of mouse are
mixed and allowed to incubate in culture for a few days, each population will be triggered by
the foreign antigens of the other. One result of this triggering is a proliferative response
which can be detected by increased uptake of tritiated thymidine.
Cells

Thymidine uptake, CPM

BALB/c + BALB/c
BALB/c + C57Bl

800
28,400

This is referred to as a "two-way" MLR, since both BALB/c and C57B1 cells can
respond to the MHC antigens of the other. One can make this test more informative by
turning it into a "one-way" MLR . This is done by treating one of the two cell populations
with the drug Mitomycin C, or with a high dose of X-irradiation. In either case, the ability of
the cell population to proliferate is eliminated, but its ability to act as a stimulator is not
affected.
Cells
(*=mitomycin treated)

Thymidine uptake, CPM

_

BALB/c + BALB/c*
BALB/c + C57Bl
BALB/c*+ C57Bl
BALB/c*+ C57Bl*

800
27,200
21,700
950

BALB/c + (C57BlxBALB/c)F1
BALB/c + (C57BlxBALB/c)F1*
BALB/c*+ (C57BlxBALB/c)F1

21,500
22,900
1,100

Mitomycin treatment of either cell population in the first case reduces the response
(since fewer cells are proliferating), while treating both eliminates the response altogether. In
the second case, we are mixing BALB/c cells with F1 cells; this is already a one-way
response, since the standard rules apply, and the F1 cannot respond to the parental cells (the
F1 cells see nothing "foreign" on the parent). In this case, mitomycin-treating the F1 has no
effect, while treating the BALB/c parent cells eliminates the response.
The MLR differs from the transplantation reaction in one major way, however, in that
the antigens responsible for MLR are primarily Class II antigens of the MHC. Mixing cells
from mice differing only at K or D will yield weak MLR reactions, while cells differing only
in the I-region will give strong reactions. It is for this reason that the MLR has been of
considerable importance in tissue typing for human transplantion. Antibodies against
Class II antigens have generally been difficult to produce, and MLR has allowed transplants
to be matched for Class II antigens as well as Class I antigens (which are readily typed
serologically). It is for this reason that Class II antigens have generally been referred to as
LD ("lymphocyte determined"). antigens, as apposed to the Class I, SD ("serologically
determined") antigens (see Figure 11-5).

205

206

12. PLAQUE-FORMING CELL (PFC) ASSAY

The development in the 1960's of the PFC (plaque-forming cell) assay by Jerne represented a
major advance in the progress of cellular immunology. It provided, for the first time, a
simple and reliable method for determining the number of individual antibody-forming cells
(AFCs) in a cell population. The kinetics of antibody-forming cell production could then be
studied without having to rely solely on the levels of serum antibody, which is related only
very indirectly to the number of cells producing it.
This procedure relies on the use of red blood cells (usually SRBC) as antigen, and on their
ability to undergo lysis in the presence of antibody and complement. Cells from an immune
spleen are mixed with a suspension of SRBC in a warm agarose solution, which is spread in a
thin layer on a microscope slide and allowed to solidify. The slides are incubated for one
hour, and antibody to SRBC produced by any particular cell in this gel will diffuse and bind
to nearby SRBCs. At the end of the hour, the slide is incubated in a suitable dilution of fresh
guinea pig serum (as a source of complement), and any SRBCs which are coated with
antibody will undergo lysis. The result is a small clear circle (a "hemolytic plaque") on a
cloudy background, surrounding each AFC. The single Ab-producing cell in the middle of
each plaque is not visible except under a microscope. These clear plaques, schematically
illustrated in Figure A12-1 below (middle slide) can be readily counted, and provide an
accurate determination of the number of AFCs: each antibody-secreting cell produces a
single clear plaque.

No C;
continuous lawn of RBC

C only;
direct (IgM) plaques

C + anti-IgG;
indirect (IgG) plaques,
also including direct plaques

Figure A12-1

207

Under the conditions described above, only those cells which produce IgM will create a
visible plaque; IgG will not generally achieve high enough concentrations to yield two IgG
molecules binding sufficiently close together on an SRBC to lyse the cell. In order to detect
IgG-secreting cells, antibody to mouse IgG (in this example) is added to the complement
solution. Two or more molecules of this secondary antibody will then bind to each single
IgG bound to an RBC, and will result in lysis in the presence of complement.
The plaques produced in the absence of this secondary "developing" antibody are known as
direct plaques, and represent the number of IgM antibody producers (middle slide above).
The additional plaques developed in the presence of the anti-IgG antibody are known as
indirect plaques and represent IgG producing cells (seen in the bottom slide in the figure).
In the example shown above, there are 5 AFCs producing IgM anti-SRBC (direct plaques),
and an estimated 11 IgG-producing cells (i.e. 16 total indirect plaques, from which one must
subtract an estimated 5 IgM-producing cells determined from the slide above).
As described here, this technique can only be used to measure those antibody-forming cells
producing antibody to red blood cells (SRBC or other RBCs). However, it can be made more
generally useful by coupling any desired antigen to RBCs, then using these coated cells as
indicator cells for that antigen (similar to what is done for passive hemagglutination). AFCs
producing antibody specific for DNP or HGG, for instance, can be detected as plaqueforming cells by using RBCs coated with either the hapten or the protein antigen.

208

13. HYBRIDOMAS: MONOCLONAL ANTIBODIES

THE PROBLEM: LIMITATIONS OF CONVENTIONAL ANTIBODIES,
HETEROGENEITY AND BATCH-TO-BATCH VARIATION
Conventional antisera, as we have seen, contain complex and heterogeneous mixtures of
antibodies with different antigen-binding specificity, as well as bearing different heavy and
light chain isotypes. Immunization of a rabbit, for example, with a human kappa IgG3
myeloma protein will potentially result in formation of antibodies directed against kappa and
3 isotype epitopes, as well as allotypic and idiotypic epitopes on both heavy and light chains.
Note that the rabbit antibodies may themselves consist of a mixture of classes (e.g. IgG and
IgM) and light chain isotypes ( and ), but this heterogeneity is not relevant for our
discussion here.
In order to prepare a -specific reagent which could be used for a clinical laboratory
assay for chains, an elaborate and expensive procedure of absorption and/or affinity
purification must be carried out, and the resulting antibody extensively characterized for its
efficacy and specificity. Furthermore, when this one batch of antibody is used up, the entire
procedure must be repeated from scratch, with no guarantee that the next batch will be the
same as, or as good as the first, since every new rabbit may respond differently to
immunization.
The problems of antibody heterogeneity and batch-to-batch variation are even more
serious for those antibodies which are the result of fortuitous "natural" immunization rather
than controlled laboratory immunizations, for example those antibodies used for human Ig
allotype determination (e.g. from polytransfused patients) and for HLA-typing (e.g. from
polytransfused patients or multiparous women).
THE SOLUTION: MONOCLONAL ANTIBODIES
In order to resolve these problems it clearly would be desirable to create a source of
antibody which is monoclonal (and therefore monospecific), and can be produced in
unlimited quantities with no batch-to-batch variation. This is precisely what the development
of HYBRIDOMA technology in the mid-1970's has allowed us to do. In principle, we could
simply choose a particular antibody-forming cell producing the antibody we desire, and
"immortalize" it so that we can propagate it in culture and collect its secreted product
indefinitely. Unfortunately these cells cannot ordinarily be grown in culture (although some
success has been achieved in immortalizing such cells by transformation with Epstein-Barr
virus).
AFC's AND FUSION PARTNERS
We begin with spleen cells from a mouse repeatedly immunized (in this example) with
a human IgG3 myeloma protein. This population is enriched for AFC's producing the Ab
we want (anti-), although many other irrelevant antibodies will also be represented.
We then harvest cells of a plasmacytoma cell line, the "fusion partner", which has been
adapted to cell culture. Several such lines are commonly available, and they have several
important properties. First, they have the capacity of unlimited growth in culture.
Second, they have the required machinery for high-level expression and secretion of
immunoglobulin; the commonly used lines have also been selected for loss of expression of
209

their own immunoglobulin, to avoid interference with the specific Ab we want. Third, they
have been selected for loss of expression of one of the enzymes required for DNA synthesis
by the "salvage pathway". One commonly used fusion partner is the mouse cell line SP2/0,
which we will use in this example; it produces no Ig of its own, and is HAT-sensitive (see
below).
"HAT" SELECTION
Figure A13-1 outlines the metabolic pathways involved in the "HAT" selection scheme.
Cells can produce nucleosides required for DNA synthesis (and therefore cell proliferation)
by either of two pathways. First, they can build them up from simple carbon sources such as
glycine via the "de novo" pathway indicated as "1"; one step in this pathway requires the
enzyme Dihydrofolate Reductase (DHFR), as indicated. Alternatively, if preformed purines
and pyrimidines are made available to the cell, it can incorporate them via the "salvage
pathway" indicated as "2", one step of which requires the enzyme Thymidine Kinase (TK).

2
Preformed, or
"salvage" pathway
TK

Thymidine,
hypoxanthine

DNA

Loss-of-function mutant

"De novo"
pathway

DHFR

Glycine

Aminopterin-sensitive

Two pathways for synthesis

of DNA precursors
Figure A13-1

However, if a cell has been selected for loss of TK expression (indicated in the figure as
TK-), then the cell is completely dependent on the de novo pathway in order to synthesize
DNA and proliferate. [Selection for loss of other enzymes in this pathway, e.g.
Hypoxanthine-Guanine Phosphoribosyl Transferase, or HGPRT, can also be used for this
purpose]. Such a cell is now exquisitely sensitive to the drug Aminopterin, which is a potent
inhibitor of DHFR. This sensitivity is the basis for HAT selection, as discussed below.

210

MAKING HYBRIDOMAS: FUSION/SELECTION/SCREENING

Three basic steps are involved in the production of the hybridoma we desire.
1) FUSION: Having immunized a mouse with the purified IgG3 protein, we mix its
spleen cells (~108 cells) with cells of the fusion partner ("PC" in the figure; ~107 cells), in the
presence of an agent which facilitates cell fusion (typically either polyethylene glycol [PEG],
or Sendai virus). At low frequency (~10-5) individual AFC's (indicated as "B" in the figure)
will fuse with individual plasmacytoma cells.
2) HAT SELECTION: The mixture is then distributed into some 1000 separate culture
wells (in a number of 96-well plates), and covered with HAT medium. This medium
contains Aminopterin ("A"), which blocks the de novo synthesis pathway, as well as
Hypoxanthine and Thymidine ("H", "T") which allow competent cells to utilize the salvage
pathway. As indicated in the Figure A13-2,, the only cells capable of growing under these
conditions are the fusion products; the unfused plasmacytoma cells are TK- and are killed by
the drug, while the unfused AFCs inherently lack the ability to grow in culture (even though
they are TK+).
3) ANTIBODY SCREENING: After a week or two in culture, a few hundred (say 500)
of the wells will contain growing cells, most of them the product of a single fusion event.
Supernatants from these 500 wells are then assayed for the presence of Ab of the desired
specificity, identifying perhaps ten positive clones. These clones are transferred into larger
dishes, subjected to additional rounds of subcloning (to eliminate possible contamination by
irrelevant clones or unstable variants), and eventually propagated for collection of the
monoclonal antibody they produce, and frozen in liquid nitrogen for long-term storage.
HAT-sensitive
plasmacytoma

PC
TK

Activated B-cells
from immune
spleen

Fused hybrid

PC
TK

~107

~500

Dies in
HAT medium

Proliferates in
HAT medium

B
TK

Assay supernatants
Figure A13-2

211

B
TK

~108

Does not survive

in culture

In our example, the antibody screening might utilize an ELISA assay with purified
chain as the antigen; using a variety of different monoclonal chains could enable us to
identify mAbs with either isotypic or allotypic specificity. [NOTE: In fact, we could have
used purified chains as the original immunogen, possibly increasing the proportion of
useful clones recovered. Two considerations might have led us to immunize instead as
described, i.e. using the intact IgG3. First, the time and expense required for chain
purification are not necessary, since we can readily identify the desired clones even among a
large number of irrelevant ones. Second, we might be interested in simultaneously screening
the 500 growing clones for anti-3(G3)-specific Ab as well, which could be used in other
clinically or scientifically relevant assays.]
The final result consists of one or more stable cell lines, each producing a monoclonal
antibody specific for human kappa chains, with specificity for either isotype or allotype.
These Abs (or the cells producing them) can be shared by laboratories all over the world with
the assurance that each lab is using Ab of precisely the same specificity. The Ab produced
years later from the same cell line will be identical to the original (so long as rare mutants are
avoided), and one batch will have precisely the same specificity as any other batch.
APPLICATIONS
Monoclonal Abs have found widespread use in many areas of clinical and basic
relevance, including assays for many drugs and hormones, reagents for blood-typing and (to a
more limited extent) HLA-typing, and for defining and diagnosing the presence of different
serotypes of pathogenic viruses and bacteria. Weve already mentioned (in Chapter 13,
"CELL SURFACE MARKERS") the importance of those monoclonal antibodies defining
several important markers, e.g. CD3, CD4 and CD8, as well as many others. Such antibodies
can be used in the laboratory determination of lymphocyte subsets (CD4/CD8), for the
depletion of T-cells from potential bone marrow grafts (using anti-CD3, for example), and for
immunosuppressive therapy (e.g. anti-CD3).

212

GLOSSARY
ACTIVE IMMUNIZATION: Stimulation of an immune response by exposure to an
antigen, e.g. injection of a mouse with BSA resulting in anti-BSA antibodies,
immunization of people with Measles virus resulting in a protective cell-mediated
response. Active immunization results in generation of memory and, therefore, longlasting immunity, as opposed to PASSIVE IMMUNIZATION.
ADOPTIVE IMMUNITY:
TRANSER.

Immune responsiveness resulting from ADOPTIVE

ADOPTIVE TRANSFER: The transfer of immune responsiveness from one animal to

another by use of immunologically competent cells. This may result in permanent
establishment of immune responsiveness, ADOPTIVE IMMUNITY, as opposed to the
short-term immunity which follows passive transfer of antibodies. This term commonly
refers to experimental systems in inbred animals.
AFFERENT: "Bringing towards"; for example, in the context of an afferent lymphatic (a
vessel bringing lymph towards a lymph node) or in a more symbolic sense to describe
the general processes whereby antigen is brought into contact with the lymphoid system - thus: "the afferent limb of the immune response." (see EFFERENT.)
AFFINITY: The association constant for the equilibrium binding of antibody to an
antigenic determinant or hapten; sometimes expressed as the dissociation constant. The
higher the affinity of a given population of antibodies for a hapten, the lower the
concentration of free hapten needed for their binding sites to become occupied.
AFFINITY describes the strength of binding of a single antibody combining site with a
single hapten; (see AVIDITY)
AGGLUTINATION: The "clumping" of a particulate antigen resulting from antibody
cross-linking, which may be visible by the naked eye or under a microscope. (see
HEMAGGLUTINATION)
ALLELE: Alternative forms of a gene found at the same locus on homologous
chromosomes within or between individuals.
ALLELIC EXCLUSION:
The process by which only a single allele for an
immunoglobulin heavy or light chain (of the two which are present for each) is selected
for expression in a single B-cell. (Also for some T-CELL RECEPTORS)
ALLERGY: Hypersensitivity state dependent on IgE and mast-cell mediated reactions;
IMMEDIATE HYPERSENSITIVITY [In obsolete usage, refers to heightened immune
reactivity of any kind to antigen.]
ALLOANTIBODY: Antibody produced in one organism directed against ALLOTYPE
determinants of a genetically different individual of the same species. Examples of
human alloantibodies include anti-HLA antibodies resulting from transfusions, or antiRh antibodies which develop in an Rh-negative mother following her delivery of an Rhpositive baby.
ALLOGENEIC: see ALLOGRAFT
ALLOGRAFT: An ALLOGENEIC tissue graft, i.e. one performed between genetically
different individuals of a single species (for example, between two different strains of
inbred mouse, or between two unrelated humans). (see also ISOGRAFT and
XENOGRAFT).
213

ALLOTYPE: A genetic POLYMORPHISM or "allelic type" of immunoglobulin. Allotypes

can be distinguished by physicochemical and chemical criteria, but are most often
defined by antigenic specificities. These allotypic specificities vary between different
members of the same species, and are inherited in an allelic manner (which distinguishes
them from ISOTYPES).
ANAPHYLAXIS: An ALLERGIC reaction characterized by a drastic drop in blood
pressure. In clinical medicine this may result from injection of a drug (e.g. penicillin) or
ingestion of an allergen (e.g. peanuts) in sensitive individuals.
ANTIBODY: Immunoglobulins (Ig) produced by lymphoid cells of vertebrates with the
ability to specifically bind to antigen. In the absence of known antigen-binding
specificity, one should use the more general term immunoglobulin.
ANTIGEN: A molecule which is specifically recognized and can be bound by antibody. It
is customary to characterize the ability of molecules to react with antibody as their
antigenicity, while the capacity to actively induce an immune response on injection into
the body is referred to as the immunogenicity of the molecule. An antigen acting as an
inducer of any type of immune response is referred to as an immunogen.
ANTIGENIC DETERMINANT: see EPITOPE.
ANTIGEN-PRESENTING CELL (APC): A cell capable of presenting antigenic peptides
associated with MHC Class II to a T-cell, and providing the required co-stimulatory
signals (e.g. by expressing cell surface B7 which can be recognized by the T-cells
CD28). The so-called professional APCs constitutively express Class II, and include
DENDRITIC CELLS, MACROPHAGES and B-CELLS.
ARTHUS REACTION: An inflammatory (skin) reaction, initiated by antigen-antibody
complexes and complement, characterized by edema, hemorrhage, PMN infiltration, and
necrosis. This term most commonly refers to the experimentally induced skin reaction
which follows administration of passive antibody and antigen, but may also be used
more generally to refer to immune complex-mediated tissue damage (for example, in the
kidney glomeruli due to serum sickness or autoantibodies).
ATOPY: ALLERGY.
AUTOGRAFT: A tissue taken from one place in an organism and grafted into another
location in the same organism (e.g. skin grafts in a burn victim). Note that such grafts
are, by definition, ISOGENEIC.
AUTOIMMUNITY: Specific immunity, either humoral or cell-mediated, to constituents of
the body's own tissues (autoantigens). If such reactions result in tissue damage sufficient
to cause any clinical abnormality, an autoimmune disease is present.
AVIDITY: A measure of the ability of multivalent antibody to form stable complexes with
multivalent antigen. This depends on AFFINITY, on the valency of the antibody and
antigen, and on other features of antigen and antibody structure.
BARE LYMPHOCYTE SYNDROME: Immunodeficiency disorder caused by a genetic
defect which prevents the expression of MHC Class I ("Type I"), or Class II ("Type II").
B-CELL: Precursor of antibody-forming cells (AFCP). A lymphocyte derived from the
haemopoietic tissues of mammals (bone marrow or fetal liver) or the Bursa of Fabricius
in birds, and capable, on antigenic stimulation, of yielding a clone of antibody-forming
cells. While all cells actively secreting antibody belong to the B-cell lineage, the
unqualified term B-cell generally refers to the resting state of the cell prior to Ag

214

stimulation. B-cells may be defined by the property of bearing readily detectable surface
immunoglobulin (IgM and IgD) or other B-cell-specific surface antigens.
BCG: Bacille Calmette-Guerin, an attenuated strain of the bacterium Mycobacterium bovis
(the causative agent of bovine tuberculosis). This organism has been used both
experimentally and clinically as a non-specific stimulator of the immune system and is
used for vaccination against tuberculosis in many parts of the world.
BENCE-JONES PROTEIN: Monoclonal immunoglobulin light chains produced by
PLASMACYTOMA tumor cells and excreted into the urine.
BETA-2-MICROGLOBULIN (
2M): A small (12 kD) non-polymorphic protein member
of the immunoglobulin superfamily, one of the two polypeptide chains present in all
MHC Class I molecules.
BGG: Bovine gamma globulin. Often used as a carrier protein for hapten immunization.
BSA: Bovine serum albumin. Often used as a carrier protein for hapten immunization.
CAPPING: The aggregation of particular cell surface molecules into a single mass which
may subsequently be shed or internalized. "Capping" may be induced by crosslinking
membrane molecules, for example with an antibody.
CARCINOGENIC: Describing a substance or procedure known to be capable of causing an
increased incidence of malignancy; thus, a carcinogenic chemical. The word is almost
synonymous with oncogenic, though by common usage, chemicals are usually referred to
as being carcinogenic, whereas viruses capable of inducing tumors are usually termed
oncogenic viruses.
CARRIER: A protein involved in the induction of immune responses when coupled to a
hapten. The investigator measures antibody production to the hapten, but may be
interested in the role that the carrier protein plays in the cellular phenomena of
immunity. Carriers are usually themselves immunogenic, and in most situations the
injection of a hapten-carrier complex leads to antibody formation also against the carrier.
CELL MEDIATED IMMUNITY (CMI): Specific immunity which is dependent on Tcells but not on antibody; it is expressed in one form as the cutaneous delayed type
hypersensitivity (DTH) reaction and is important in defense against infection by viruses,
fungi, intracellular bacteria and some protozoa, and in rejection of tissue and organ
grafts.
CENTRAL LIMB: Central limb of the immune system, the part involving the triggering
and differentiation of antigen-specific lymphocytes.
CHEMOTAXIS: The directed movement of cells in a particular direction or to a particular
site, as for example the migration of lymphoid cells into a site of inflammation.
CHIMERA: An individual organism composed of two or more populations of cells with
different genotypes; usually experimentally created, as in adoptive transfers.
CLASS (of Ig): The class of an immunoglobulin molecule is defined by the ISOTYPE of its
heavy chain; IgG represents the class which bear gamma H- chains, IgA bear alpha Hchains, etc. Ig molecules of any given class (i.e. heavy chain class) may bear any of the
existing light chain isotypes (e.g. kappa or lambda).
CLASS I and II (MHC molecules): see MHC
CLONE: All the cells derived from a single progenitor cell by repeated mitosis, and all
having the same genetic constitution (apart from mutation or gene rearrangement).

215

CLONAL DELETION (=CLONAL ABORTION): The process by which auto-reactive Bor T-cells are destroyed early during their maturation.
CLONAL SELECTION: A hypothesis developed by Burnet in the 1950's by which antigenprecommitted precursors are selected by the presence of specific antigen to proliferate
and differentiate into immunocompetent cells. Originally proposed for an understanding
of B-cell function, but it holds as well for T-cells.
CMI: see CELL MEDIATED IMMUNITY
CODOMINANCE: The phenotypic expression of both alleles for a particular locus in a
heterozygote. For example, all Ig genes are codominantly expressed in an individual
(despite the phenomenon of allelic exclusion in individual B-cells).
COMBINATORIAL ASSOCIATION: In immunoglobulin biosynthesis, the random
association of light and heavy chain variable regions to form the antigen-binding site of
antibodies; a major contributor to diversity (also for T-CELL RECEPTORS).
COMBINATORIAL JOINING: In immunoglobulin gene rearrangement, the random
joining of V and J gene segments of light chains (V, D and J segments for heavy chains),
a major contributor to diversity (also for T-CELL RECEPTORS).
COMPLEMENT: An enzymatic complex of serum proteins that is activated by many
antigen-antibody reactions, and which is essential for antibody-mediated immune
hemolysis and bacteriolysis; it also plays a major part in clearance of immune complexes
and other biological reactions.
COMPLEMENT FIXATION: Initiation of the COMPLEMENT cascade, resulting in the
consumption of active complement. The Complement Fixation Assay is a tool which
may be used to detect and quantitate the presence of either antigen or antibody, relying
on the ability of (some) antigen/antibody complexes to consume complement.
CR: Complement receptor; a variety of receptors specific for different complement
components and their proteolytic products are present on a variety of cells of the immune
and hematopoietic system.
CYTOKINE: Any of a large number of secreted proteins mediating signals between
populations of leukocytes during immune responses. Includes lymphokines produced by
lymphocytes (e.g. IL-2 and IL-4), and monokines produced by monocytes (e.g. IL-1 and
IL-6)
CYTOPHILIC ANTIBODY: Antibody passively adsorbed on to the surface of a cell which
did not produce it (e.g. macrophages, B-cells), bound by cell receptors specific for the Fc
fragment of the antibody.
DELAYED-TYPE HYPERSENSITIVITY (DTH):
Any manifestation of CELL
MEDIATED IMMUNITY (e.g. TB skin test), alluding to the fact these reactions usually
take a day or more to develop, as opposed to antibody-mediated reactions or
HUMORAL IMMUNITY which may be manifested in minutes.
DENDRITIC CELL: Bone-marrow-derived cells with extensive finger-like processes.
Immature dendritic cells reside in tissues, take up pathogenic or other antigenic material
by phagocytosis, and migrate via afferent lymphatics to a lymph node where they mature
and present antigen to T-cells.
DETERMINANT: see EPITOPE.
DNP: The dinitrophenyl HAPTEN, or dinitrophenol.

216

DOMINANT: Describing an allelic trait which is phenotypically expressed in individuals

who are heterozygous for the particular allele (as well as in homozygotes); contrasted
with the RECESSIVE allelic form, expressed only in homozygotes.
DTH: (see DELAYED-TYPE HYPERSENSITIVITY)
EFFECTOR: Usually used as an adjective, e.g., effector cell, which carries out the final
stages of an immune reaction. Activated T-cells (for cellular immunity) and plasma cells
(for humoral immunity) can both be considered effector cells. (Other cell types can also
have effector roles in immune reactions.)
EFFERENT: "Carrying away"; efferent lymphatics carry lymph (and cells) away from
lymph nodes. The efferent limb of the immune response includes all the effector functions of antibodies and T-cells. (see AFFERENT.)
ELISA: Enzyme-Linked ImmunoSorbent Assay; any assay for antigen or antibody which
relies on detection of enzymed-linked molecules.
EPITOPE: The minimal portion of an antigen molecule which is recongized and bound by
an antibody; also referred to as an ANTIGENIC DETERMINANT. Epitopes on the
surface of proteins or carbohydrates are small, commonly being in the size range
equivalent to 3-6 amino acids.
Fab: "Antigen-binding" fragment of immunoglobulin molecules produced by proteolysis,
containing the entire light chain and the N-terminal portion of the heavy chain; includes
the variable regions of both chains.
Fc: "Crystallizable" fragment of immunoglobulin molecules produced by proteolysis,
contains only the C-terminal portion of the heavy chain constant region.
FcR: "Fc-receptor"; diverse family of receptors specific for the Fc region of different Ig
isotypes, present primarily on phagocytic cells (macrophages, PMN's) and B-cells.
FITC: Fluorescein isothiocyanate, a chemically reactive derivative of flourescein, readily
coupled to free amino groups of proteins to render them fluorescent.
FOLLICLE: Ovoid or spherical collection of densely packed lymphocytes identified in
histological sections of lymphoid tissues. A primary lymphoid follicle is a rounded
aggregation of small lymphocytes (mainly B-cells), and dendritic follicular cells; a
secondary lymphoid follicle is a more complex structure containing the above two
elements together with a nest of rapidly proliferating lymphocytes (mixed B and T cells)
known as a GERMINAL CENTER.
GENOTYPE: The genetic constitution of an individual which results in the expression of a
particular PHENOTYPE.
GERMINAL CENTER: (see FOLLICLE.) Region of cell proliferation within a secondary
follicle; germinal centers are antigen-dependent and T-dependent in their origin,
containing both T and B cells. Generation of B-memory cells and Ig somatic mutation
occur within germinal centers.
GRANULOCYTE: One of three varieties of blood leukocyte with a segmented nucleus and
cytoplasmic granules. Includes polymorphonuclear neutrophils (PMN), eosinophils and
basophils.
H-2: The major histocompatibility complex (MHC) of the mouse.

217

HAPLOTYPE: A combination of allelic forms of several loci inherited as a closely linked

unit on a chromosome. A haplotype of HLA, for instance, represents a particular combination of allelic forms of HLA-A, B, C, D, etc.
HAPTEN: A small molecule which by itself cannot stimulate antibody production but
which can be recognized and bound by antibody once the antibody has been formed (i.e.
a hapten is antigenic but not immunogenic). A hapten can be rendered immunogenic by
being coupled to a CARRIER. Landsteiner, in the early 1900's, introduced the use of
small chemically defined structures to add greater precision to the study of
immunochemistry.
HEMAGGLUTINATION: AGGLUTINATION involving red blood cells (RBC). ACTIVE
hemagglutination results from the binding of antibodies to antigens which are a normal
part of the RBC membrane; PASSIVE hemagglutination results from antigens which
have been artificially attached to RBC membranes.
HETEROGRAFT: A tissue or organ graft between two individuals of different species.
This older term has been replaced by xenograft.
HETEROTOPIC: see ORTHOTOPIC
HETEROTOPIC: see ORTHOTOPIC.
HETEROZYGOTE: An individual possessing two different alleles at a particular locus on
a pair of homologous chromosomes.
HEV: HIGH ENDOTHELIAL VENULE. A specialized small vein, originally known as a
POST CAPILLARY VENULE (PCV), which is found in the deep cortex of lymph
nodes. This vessel is lined by a high, cuboidal endothelium and is prominently involved
in the mechanism of recirculation of lymphocytes. The lining cells have special
biochemical features which allow the attachment and migration of B- and T-cells from
the circulation into the node.
HINGE REGION: That portion of the immunoglobulin heavy chain which lies between
CHI and CHII, and around which the Fab regions can typically flex.
HGG: Human gamma globulin.
HISTOCOMPATIBLE: Two individuals identical in those genes with determine the antigenicity of a tissue graft and will therefore accept tissue grafts made between them (see
MHC.)
HLA: The major histocompatibility complex (MHC) of humans (the name derives from
Human Leukocyte Antigen).
HOMOLOGOUS CHROMOSOMES: Chromosomes which pair during mitosis and
contain matching loci (although each may bear different alleles at a particular locus).
HOMOZYGOTE: An individual possessing two identical alleles at one particular locus on
a pair of homologous chromosomes.
HSA: Human serum albumin.
HUMORAL IMMUNITY: Specific immunity which is dependent upon antibodies present
in the blood and tissue fluids.
HYBRIDOMA:
An experimentally produced cultured cell line which secretes a
monoclonal, homogeneous antibody. This term generally refers to the product of cellular
fusion event between a transformed plasmacytoma in culture, and a normal, activated B-

218

cell. It can more broadly include fusions involving T-cells, macrophages or other
lymphoid cells.
HYPERSENSITIVITY: State of the previously immunized individual in which tissue
damage results from the immune reaction to a further dose of antigen. May be antibody
mediated (immediate) or cell-mediated immunity (delayed).
IATROGENIC: "Caused by the healer"; a condition resulting from an undesirable sideeffect of medical treatment.
IDIOTYPE: The antigenic determinant(s) of an antibody molecule unique to its combining
site, the particular combination of VH and VL.
IMMEDIATE HYPERSENSITIVITY: see ALLERGY
IMMUNE COMPLEX: Antigen-antibody complex. May be present in soluble form, especially in antigen excess, or as a precipitate, especially at equivalence; can be deposited in
tissues and cause considerable damage.
IMMUNOASSAY: Any method utilizing specific antigen-antibody reaction of biological
material (e.g., complement fixation test, hormone radioimmunoassay, radial
immunodiffusion test, passive agglutination test).
IMMUNODEFICIENCY: Deficiency in an organism's ability to mount an immune
response, may be either congenital or acquired.
IMMUNOGEN: (see ANTIGEN)
IMMUNOGLOBULIN: Member of a family of proteins each made up of light chains and
heavy chains linked together by disulfide bonds. All antibodies are immunoglobulins.
Present in serum and body fluids; on electrophoresis mostly migrate as gammaglobulins.
IMMUNOLOGICALLY COMPETENT CELL: Synonymous with Antigen Reactive
Cell. Clonally pre-committed lymphoid cell capable of generating an immune response
after activation by antigen, and includes both T- and B-cells.
INBREEDING: Breeding program which generally involves brother-sister matings for
many generations which tends to progressively decrease genetic variability. Fully inbred
strains (e.g., of mice or rats) are homozygous at every locus, and, in addition, individuals
within an inbred strain are as identical to each other as identical twins.
INNATE IMMUNITY: The ability of an organism to resist microbial infection by
mechanisms which do not exhibit memory or clonal specificity. Skin and mucus
barriers, phagocytic cells and COMPLEMENT are all part of the innate immune system.
INTERLEUKIN: (see CYTOKINE)
ISOGENEIC: see ISOGRAFT
ISOGRAFT: An ISOGENEIC or SYNGENIC tissue graft, i.e. one performed between
genetically identical individuals (for example, between two mice of the same inbred
strain). (see also ALLOGRAFT and XENOGRAFT).
ISOTYPE: In relation to immunoglobulins, refers to variants of particular heavy or light
chains which are present in all individuals of a species as opposed to being inherited in
allelic fashion (which distinguishes them from allotypes). Mu, gamma, and alpha
(among others) are isotypes of heavy chains, etc., and define the CLASS of Ig. Kappa
and lambda are isotypes of light chains.

219

J-CHAIN: "Joining chain"; the polypeptide chain covalently attached to the heavy chain of
secreted IgM and polymeric IgA during the process of their polymerization. (Not to be
confused with the J-SEGMENT).
J-SEGMENT: One of two (or three) genetic elements required to form a complete variable
region for antibody light (or heavy) chains during the process of gene rearrangement,
together with the V-region (and D-region) gene segment(s); not to be confused with the
J-CHAIN. (Also present in T-CELL RECEPTOR)
KLH: Keyhole limpet hemocyanin. Very large and highly immunogenic protein, may be
used as a carrier protein for hapten immunization, or alone to generate an antibody
response or a DTH skin reaction.
LAD (LEUKOCYTE ADHESION DEFICIENCY): A group of rare immunodeficiency
diseases caused by a genetic defect in any one of several adhesion molecules expressed
on lymphocytes or PMNs.
LEUKOCYTE: White blood cell, including LYMPHOCYTE, MONOCYTE and
GRANULOCYTE.
LOCUS: The position in a chromosome at which a particular gene is located.
LYMPHOCYTE: All cells of lymphoid tissue, blood, or bone marrow which are
characterized by the possession of a round nucleus, relatively little cytoplasm, and lack
of prominent vacuoles or granules. This was originally a morphological term, which
includes cells with many different functions, but it is often used in a more restricted
sense to refer specifically to T-cells and B-cells.
LYMPHOKINE: General name for soluble factors secreted by lymphocytes which have
biological effects on other cells (see CYTOKINE).
MACROPHAGE:
Large phagocytic cell, responsible for antigen processing and
presentation to T-cells; key component of afferent limb of the immune response.
Radioresistant and adherent cell, similar to blood monocyte and skin histiocyte.
MEDULLA: This word is used as an anatomical term to describe the inner portion of many
organs; in immunology, it usually refers to that deep portion of lymph nodes in which
extensive phagocytosis takes place, or the deep portion of the thymus.
MHC: Major Histocompatibility Complex, a series of closely linked genes encoding CLASS
I and CLASS II MHC molecules. Class I MHC molecules, in combination with bound
antigenic peptides, are targets for recognition and killing by cytotoxic T-cells; Class II
MHC molecules, in combination with bound antigen, are required for antigen
presentation to helper T-cells.
MONOCLONAL: Describing an antibody population derived from the progeny of a single
antibody-forming cell precursor. HYBRIDOMAS and myeloma proteins are both
examples of monoclonal immunoglobulins.
MONOCYTE: see MACROPHAGE
MONONUCLEAR CELLS:
Collectively referring to lymphocytes and
monocytes/macrophages, to distinguish them from leukocytes with segmented nuclei
(i.e. PMN's and other granulocytes).
MONOSPECIFIC: Describing an antibody population specific for a particular antigenic
determinant.

220

MULTIPLE MYELOMA: Plasma cell tumor typically characterized by production of

MYELOMA PROTEIN and local destruction of bone tissue.
MYELOMA PROTEIN: A monoclonal immunoglobulin molecule, which may be normal
or abnormal in its structure, secreted by a myeloma or other tumor of antibody-forming
cells (see PLASMACYTOMA). Myeloma protein heavy chains are found in serum, and
free light chains in both serum and urine (BENCE-JONES PROTEIN).
NATURAL KILLER CELLS: Lymphocytes which lack T-cell and B-cell markers, and
which are capable of killing target cells based on abnormal expression of MHC
molecules. Also referred to morphologically as large granular lymphocytes.
NK CELLS: see NATURAL KILLER CELLS
ONCOGENIC: Describing a substance or procedure known to be capable of causing an
increased incidence of malignancy; thus, a carcinogenic chemical. The word is almost
synonymous with carcinogenic, although by common usage, chemicals are usually
referred to as being carcinogenic, whereas tumor-causing viruses are usually termed
oncogenic.
OPSONIZATION: The facilitation of phagocytosis by macrophages or granulocytes
resulting from antibody and/or complement bound to a target (e.g. a bacterium).
ORTHOTOPIC: A term describing some tissue transplants. A graft is orthotopic if it is
placed into the same anatomical region as the organ which it is designed to replace,
otherwise it is heterotopic. Thus, human heart transplants are orthotopic, whereas
human thymus transplants into the abdominal muscle wall are heterotopic.
OVA: Ovalbumin, albumin of chicken (or other birds) eggs. Often used a carrier protein
for hapten immunization
PASSIVE IMMUNIZATION: Transfer of immunity by transfer of specific antibody.
Injection of antibody to tetanus toxoid, for example, results in short-term resistance to
tetanus infection, which decays with the half-life of serum immunoglobulin (~2-3
weeks).
PFC: see PLAQUE-FORMING CELL
PHAGOCYTOSIS: Internalization of extracellular material by the formation of pseudopods
and subsequent engulfment; macrophages and PMN's are phagocytic. (see also
PINOCYTOSIS and RECEPTOR-MEDIATED ENDOCYTOSIS).
PHENOTYPE: The physical or chemical manifestations of an organisms GENOTYPE. A
person whose phenotype is blood group A, for example, could be of genotype A/A or
A/O.
PINOCYTOSIS: Internalization of soluble extracellular material by formation of small
membrane vesicles. (see also PHAGOCYTOSIS and RECEPTOR-MEDIATED
ENDOCYTOSIS.
PLAQUE-FORMING CELL: Antibody-secreting cell visualized by its ability to cause
localized lysis of red blood cells (thus forming a plaque) in an experimental assay
system.
PLASMA CELL: The major immunoglobulin-secreting cell type and the terminally
differentiated cell of the B-cell series. It is present within lymphoid organs but rarely in
the blood (despite its name), and is characterized by plentiful basophilic cytoplasm with
much endoplasmic reticulum and a high content of RNA.

221

PLASMA: The fluid portion of anti-coagulated blood remaining after the "formed elements"
(cells and platelets) are removed. Commonly used anti-coagulants include heparin,
citrate, and EDTA (see also SERUM).
PLASMACYTOMA: A tumor derived from a plasma cell, with which it shares
morphological characteristics, often secreting either an intact or aberrant
MONOCLONAL IMMUNOGLOBULIN.
PMN: Polymorphonuclear neutrophil, the most abundant of the granulocytes in blood,
whose cytoplasmic granules stain with neutral dyes.
POLYMORPHIC: Describes a gene or its protein product which exists in distinct allelic
forms. The alpha chains of MHC Class I molecules (HLA-A, B and C), for example, are
highly polymorphic, although their 2-microglobulin component is not. Other examples
of polymorphism are the ABO and Rh blood groups, and ALLOTYPES of
immunoglobulin light and heavy chains.
POST CAPILLARY VENULE: see HEV.
PRECIPITATION: The formation of an insoluble complex from a soluble antigen by
antibody-binding.
RBC: Red blood cells. This abbreviation is often preceded by a letter indicated the species
from which they are derived; thus, SRBC are sheep erythrocytes, HRBC are from
human, BRBC from burros, etc.
RECEPTOR: Cell membrane constituent which binds a specific ligand (e.g. complement
receptors, Fc receptors, etc.)
RECEPTOR-MEDIATED
ENDOCYTOSIS:
Internalization
of
extracellar
macromolecules induced by binding to specific cell-surface receptors. (see also
PHAGOCYTOSIS and PINOCYTOSIS).
RECESSIVE: Referring to an allele which is phenotypically expressed only in individuals
who are homozygous for the particular allele, and not in those who are heterozygous at
the locus. (see DOMINANT).
RED PULP: That portion of the spleen in which erythrocytes are stored, phagocytosed and
degraded, in which phagocytosis of foreign material occurs, and where extensive
erythro-myelopoiesis occurs in some species (rodents, but not humans). Distinguished
from WHITE PULP.
RES: Reticuloendothelial system, including the lymphoid tissues and the phagocytic
compartments of the vasculature, especially the liver and lungs.
RIA: Radioimmunoassay; any antibody-based assay (for detecting and quantitating antigen
or antibody) which relies on the use of radioactively labeled antibody or antigen.
ROSETTES: A lymphoid cell which binds a ring of erythrocytes around itself, forms a
structure identified microscopically as a rosette. Human T-cells bind sheep red blood
cells via their CD2 molecules, forming E-rosettes (for "erythrocyte" rosette).
Erythrocytes coated with IgG antibody can similarly form rosettes with Fc-receptorbearing cells such as macrophages, which are known as "EA" rosettes.
SEROLOGICAL: Having to do with serum or its components; describing assays or
procedures which rely on the use of antibodies (ultimately derived from serum).

222

SERUM: The fluid portion of blood which remains after a blood clot is formed. Serum is
similar in its composition to PLASMA, except for the lack in plasma of those proteins
consumed or altered during the process of coagulation.
SRBC: Sheep red blood cells, a commonly used antigen in experimental systems because of
the variety of assays in which it can function (agglutination, lysis, PFC assays).
SUPERANTIGEN: A molecule capable of cross-linking an MHC Class II molecule with the
variable regions of a large proportion of TCRs (typically activating 2-20% of all T cells).
Such molecules, produced by various microorganisms, can result in conditions such as
Toxic Shock Syndrome.
SYNGENEIC: see ISOGRAFT.
T-CELL: Thymus derived lymphocyte, responsible for Cell-Mediated Immunity (CMI).
T-CELL RECEPTOR (TCR): The antigen-binding receptor of T-cells; encoded by a family
of genes distinct from (but homologous to) immunoglobulins, and produced by the same
processes of gene rearrangement. The TCR can recognize antigenic peptides only in the
context of MHC Class I or Class II molecules.
TAP ("Transporters associated with Antigen Processing"): Part of the machinery by which
peptides from endogenously produced proteins are translocated from the cytosol into the
ER. Within the ER they become associated with newly synthesized MHC Class I
molecules which are in turn transported to the plasma membrane.
TCR: see T-CELL RECEPTOR.
THY-1: A cell surface marker for mouse T-cells, originally known as "theta" antigen.
Different strains of mice express different allelic forms of this molecule, and anti-Thy-1
antibodies have been used to distinguish T-cells in adoptive transfer experiments.
TISSUE TYPING: The identification of histocompatibility antigens most commonly MHC.
HLA (human MHC) typing is carried out by observing the cytotoxic effect of specific
antisera on blood leukocytes collected from the prospective donor and recipient of
organs or tissues that are to be transplanted.
TOLERANCE: (Immunological) Inability to mount an immune response against a specific
antigen which is normally immunogenic, as in tolerance to self antigens. Development
of tolerance may follow contact with antigen in fetal or early postnatal life or, in adults,
administration of very high or very low doses of certain antigens, and can be expressed
at the level of T-cells or B-cells, or both.
VASOACTIVE: Having the ability to cause dilation of blood vessels, as for example
COMPLEMENT components C3a and C5a.
WHITE PULP: The portion of the spleen containing dense aggregation of lymphocytes,
namely lymphoid follicles, and periarteriolar white sheath; distinguished from RED
PULP.
XENOGENEIC: see XENOGRAFT
XENOGRAFT: A XENOGENEIC tissue graft, i.e. one performed between individuals of
different species. (see also ALLOGRAFT and ISOGRAFT).

223

224

INDEX
antigen-binding sites .............................. 23
antigenic competition........................... 166
antigenic determinant............................... 9
antigenic variation................................ 166
antigenicity............................................... 9
antigen-presenting cell ......................... 117
anti-HLA antibodies............................. 133
antisera ................................................... 47
APC...................................................... 117
appendix............................................... 123
Arthus reaction............................... 34, 154
asthma .................................................. 156
attenuated organisms.................... 163, 164
ATxBM ................................................ 107
autoimmune thyroiditis ........................ 144
autoimmunity ................................... 5, 143
automimmune diseases .......................... 41
avidity....................................... 19, 20, 192
B cell ............................................ 106, 125
bacteriolysis ........................................... 34
Bence-Jones proteins ................. 27, 46, 47
binding sites ......................................... 191
biological clearance................................ 42
blocking factors.................................... 172
blood transfusions .................................. 89
bone marrow ........................ 123, 129, 130
bone marrow transplants ........................ 89
BRBC..................................................... 57
bronchoconstriction.............................. 156
brother sister inbreeding....................... 201
Bruton's agammaglobulinemia..... 132, 148
BSA........................................................ 13
Bursa of Fabricius ........................ 105, 109
capillary dilation .................................. 156
carcinoembryonic antigen .................... 170
carcinogens........................................... 170
cardiovelofacial syndrome ................... 149
carrier effect ......................................... 119
CD antigens.......................................... 115
CD2 ...................................................... 111
CD3 ...................................................... 111
CD4 .............................................. 111, 114
CD5 ...................................................... 111
CD8 .............................................. 111, 114
CDR ....................................................... 28

A2m........................................................ 49
ABO blood groups ........................... 77, 82
absorbed antiserum ................................ 48
absorption............................................. 181
acetylcholine receptor .......................... 145
acquired immunodeficiencies .............. 147
Acquired Immunodeficiency
Disease (AIDS) ............... 151
activated macrophages ......................... 103
active immunity...................................... 10
active immunization............................. 163
acute disseminated encephalomyelitis . 144
adaptive immunity.................................... 3
ADCC .................................................... 93
ADEM.................................................. 144
adjuvant...................................... 9, 14, 165
adoptive transfer............... 55, 84, 106, 108
AFCP...................................................... 54
affinity.............................. 19, 20, 189, 192
affinity chromotography......................... 48
affinity maturation...................... 32, 53, 58
affinity purification .............................. 181
AFP ...................................................... 170
agglutination .......................................... 11
AIDS .................................................... 151
alkaline phosphatase ............................ 177
allelic exclusion ......................... 53, 65, 69
allergen................................................. 153
allergic reaction.................................... 165
allergy skin test .................................... 154
allograft rejection ........................... 83, 101
allotype............................................. 46, 49
alpha-fetoprotein .................................. 170
ALS ...................................................... 151
alternate splicing .................................... 64
alum precipitation .......................... 15, 165
anaphylotoxin......................................... 34
anergy................................................... 140
antibody.................................... 6, 9, 11, 19
antibody combining site ......................... 19
antibody dependent cell-mediated
cytoxicity............................................ 93
antibody heterogeneity ................... 20, 191
antigen................................................ 9, 13
antigen sequestration............................ 135

225

diffuse cortex ....................................... 123

DiGeorge syndrome ............................. 149
diphtheria ..................................... 164, 167
direct plaques ....................................... 208
diversity.................................................. 53
dizygotic cattle twins ........................... 135
DNA rearrangement ............................... 62
DNP................................................ 13, 120
domains .................................................. 28
DTH response ................................ 91, 115
EAE...................................................... 145
ECF-A .................................................. 156
electron microscope ............................. 177
electrophoresis ....................................... 17
ELISA .................................. 177, 184, 187
encephalomyelitis................................. 143
endogenous peptides .............................. 98
enhancing antibodies.............. 85, 132, 171
enteropathies ........................................ 151
enzyme labeling ................................... 177
enzyme-labeled antibodies ................... 178
eosinophil............................................. 156
eosinophilia .......................................... 156
epitope...................................................... 9
equilibrium constant..................... 189, 190
equilibrium dialysis........................ 19, 189
erythroblastosis fetalis...................... 77, 81
ethanol precipitation............................... 17
exchange transfusion.............................. 81
exogenous peptides ................................ 98
experimental allergic
encephalomyelitis.......... 145
eye lens protein .................................... 143
F(ab)....................................................... 24
F(ab')2 ..................................................... 24
F1 inbred crosses.................................. 203
Fc............................................................ 24
FcR......................................... 12, 111, 112
Fc-receptors.................................... 12, 112
fetal development................................. 131
fetal liver .............................................. 129
fetal thymus transplant ......................... 149
first set rejection..................................... 83
fluorescence-activated cell
sorter (FACS)................ 177
fluorescent labeling .............................. 177
Freunds adjuvant................................... 14
GALT ................................................... 123

CEA...................................................... 170
cell surface markers ............................. 111
cell-mediated
immunity (CMI).... 83, 85, 91, 105, 154
cellular immune responses ....................... 6
central limb .......................................... 147
central lymphoid organs....................... 109
chemotaxis ............................................. 34
chronic granulomatous disease ............ 147
circulating immune complexes ............ 146
class I MHC ............................... 83, 87, 96
class II MHC ...................... 83, 87, 96, 117
class switching ................................. 65, 69
classes of Ig...................................... 23, 25
classical complement pathway............... 35
class-switching....................................... 61
clonal abortion ....................... 58, 135, 139
clonal expansion............................... 55, 58
clonal selection..................... 32, 53, 54, 57
CMI .................................................. 83, 91
Cohn fractionation ................................. 17
combinatorial association................. 61, 67
combinatorial joining ....................... 61, 66
complement............................................ 33
complement deficiencies...................... 148
complement fixation assay................... 197
complement receptor (CR)............. 12, 112
complementarity determining
regions (CDR)......... 28
concanavalin A (ConA) ....................... 115
concomitant immunity ......................... 172
congenital immunodeficiencies............ 147
congenital thymic aplasia..................... 149
constant region ................................. 23, 27
constant region ....................................... 45
contact dermatitis ................................... 92
CR .................................... 12, 34, 111, 112
CR-bearing pre-B-cells ........................ 149
cross-matching ....................................... 78
cross-reactivity ..................................... 195
cyclosporin ........................................... 152
cytolysis.................................................. 12
cytotoxic drugs..................................... 151
cytotoxic T cells ..................................... 91
cytotoxicity............................................. 12
degranulation........................................ 156
delayed type
hypersensitivity (DTH) ........ 153, 154

226

gamma-globulin ..................................... 19
gene rearrangement ................................ 61
generation of diversity...................... 61, 66
germinal centers ................... 123, 125, 127
germ-line theory..................................... 61
Goodpasture's syndrome ...................... 146
graft rejection ............................... 5, 84, 91
graft versus host (GvH)...... 83, 88, 92, 149
H-2 ......................................................... 86
haplotype exclusion................................ 65
hapten............................................. 13, 119
Hashimotos thyroiditis........................ 144
heavy chain class switching ................... 65
heavy chains ............................... 23, 24, 61
hemagglutination.................................... 11
hematopoietic stem cells ...................... 129
hemolysis ....................................... 12, 154
heterologous antibodies ....................... 165
high endothelial venules (HEV)........... 125
hinge region............................................ 29
histamine .............................................. 156
histocompatibility antigens .................... 86
HIV....................................................... 151
HLA ....................................................... 86
homocytotropic .................................... 155
horseradish peroxidase......................... 177
hot antigen suicide ................................. 55
3
H-TdR suicide....................................... 56
Human Immunodeficiency
Virus (HIV) ........ 151
humoral immunity.............................. 6, 85
H-Y antigen.......................................... 137
hybridoma ............................................ 114
hyperimmune rejection .......................... 85
hypervariable regions ............................. 27
hypocalcemic tetany............................. 149
iatrogenic.............................................. 147
identity non, partial, complete........... 179
idiotype................................................... 46
IF, immunofluorescence....................... 177
IFN- ...................................................... 96
IgA, IgD, IgE, IgG,.......................... 25, 26
IgM........................................... 25, 26, 193
IL-17..................................................... 114
IL-2......................................................... 96
IL-4......................................................... 96
immobilization ....................................... 12
immune adherence ........................... 12, 34

immune clearance .................................. 42

immune complex disease ... 33, 34, 41, 154
immune complexes ................ 41, 143, 154
immune response (Ir) genes ................... 87
immunoadsorption ............................... 181
immunodeficiency................................ 135
immunodiagnosis ................................. 172
immunofluorescence (IF) ..................... 177
immunogen ........................................ 9, 13
immunoglobulin..................................... 19
immunoglobulin genes........................... 61
immunological enhancement ................. 85
immunological memory ............. 23, 30, 84
immunological surveillance ................. 170
immunologically privileged sites ... 84, 141
immunomodulation .............................. 169
immunoselection .................................. 169
immunosorbent assay (ELISA) .............. 11
immunosuppression ............................. 135
immunotherapy .................................... 172
in vitro primary antibody response ........ 56
inbreeding ............................................ 201
inflammation ............................................ 3
influenza............................................... 167
innate immunity ....................................... 3
intracellular parasites ............................. 91
intravenous........................................... 157
intravenous IgG (IVIG)........................ 149
ionizing radiation ......................... 152, 170
Ir genes ........................................... 87, 101
isoagglutinins ......................................... 77
isograft ................................................... 83
isotype .................................. 23, 25, 46, 49
J-chain .................................................... 69
J-segments.............................................. 64
kappa chain complex.............................. 62
kappa chains......................... 25, 46, 61, 66
Kern........................................................ 50
kidney................................................... 146
killed organisms ................................... 164
kinins.................................................... 156
Km.......................................................... 46
LAK ..................................................... 173
lambda chains....................... 25, 46, 50, 61
large granular lymphocytes .................... 93
latent period ......................................... 158
LCMV .................................................... 96
leader polypeptide .................................. 64

227

oncofetal antigens ................................ 170

oncogenes............................................. 169
one-shot serum sickness......................... 42
one-way MLR ...................................... 205
opsonization ............................. 12, 34, 112
ouchterlony............................................. 11
Ouchterlony.................................... 46, 179
papain..................................................... 24
passive agglutination.............................. 11
passive cutaneous anaphylaxis (PCA) . 153
passive immunity ........................... 10, 163
PCA...................................................... 153
pepsin ..................................................... 24
pertussis................................................ 167
Peyer's patches ..................................... 123
PFC .............................................. 106, 207
phagocytosis......................................... 147
pharyngeal pouches.............................. 149
plaque-forming cell (PFC) ........... 106, 207
PMNs ................................................... 154
pneumococcus........................................ 11
Pneumocystis carinii............................ 149
point of equivalence............................. 175
pokeweed mitogen ............................... 115
poliomyelitis ........................................ 167
polyclonal activation .............................. 60
post-translational processing.................. 69
Prausnitz-Kstner......................... 153, 158
precipitin curve .................................... 175
primary follicles ................................... 123
primary immunodeficiencies................ 147
proteolytic fragments ............................. 23
radioimmunoassay (RIA) ............... 11, 177
radiolabeling ........................................ 177
rashes.................................................... 157
reaginic antibody.................................. 155
reagins .................................................. 155
receptor blockade ......................... 135, 139
recirculation ......................... 123, 125, 126
renal failure .......................................... 151
RER........................................................ 64
reticuloendothelial system (RES)......... 123
Rh........................................................... 77
Rh incompatibility ............................... 129
rheumatoid arthritis................ 41, 143, 145
rheumatoid factor ................................. 145
rhinitis .................................................. 157
RhoGam therapy .................................... 81

lectins ................................................... 115

leukotrienes .......................................... 156
light chains ....................................... 23, 24
Listeria ................................................. 102
luciferase .............................................. 177
luminescence........................................ 178
lupus erythematosus............................... 41
lymph node........................................... 123
lymphatic.............................................. 125
lymphocyte-determined (LD)................. 87
lymphocytes ......................................... 154
lymphoid system ...................................... 6
lymphoid tissue .................................... 123
lymphokine-activated killer (LAK)...... 173
macrophage ............................ 91, 117, 154
major histocompatibility
complex (MHC) ............. 83, 86
malignancies ........................................ 151
marginal sinus ...................................... 125
mast cells.............................................. 153
maternal IgG ................................ 131, 167
maternal immunization ........................ 129
Mcg ........................................................ 50
memory .............................................. 4, 53
memory cells .......................................... 58
metastasis ............................................. 169
MHC ................................................ 83, 86
MHC class I & II.................... 96, 111, 112
MHC restricted recognition ............. 91, 94
mIg ....................................................... 111
migration ...................................... 126, 130
mitogens............................................... 115
mitomycin treatment ............................ 205
mixed lymphocyte reaction (MLR)........ 92
monoclonal gammopathy....................... 20
mononuclear cells ........................ 154, 155
mucus secretion.................................... 156
multiple sclerosis ................................. 145
muramyl dipeptide ................................. 15
myasthenia gravis......................... 143, 145
mycobacterium............................... 14, 166
myeloma................................... 20, 26, 192
natural antibodies ....................... 53, 59, 79
natural killer (NK) cell........................... 93
neo-antigens ......................................... 169
neoplasia .............................................. 169
neutrophils (PMN) ............................... 154
nude mouse .................................. 105, 109

228

RIA....................................... 177, 184, 185

RNA splicing ......................................... 64
rough endoplasmic reticulum................. 64
Sabin polio vaccine .............................. 166
salt precipitation..................................... 17
SCID .................................................... 150
second set rejection ................................ 84
secondary follicle ................................. 125
secondary immune responses........... 23, 30
secondary immunodeficiencies ............ 147
selective IgA deficiency....................... 150
self-renewal.......................................... 129
self-tolerance........................................ 135
sequestered antigens............................. 143
serologically setermined (SD)................ 87
serum sickness ............... 33, 163, 165, 184
severe combined immunodeficiency.... 150
signal ...................................................... 64
skin grafts............................................... 83
SLE ...................................................... 146
smallpox............................................... 167
somatic mutation.............................. 61, 66
somatic theory........................................ 61
Southern blot........................................ 199
specificity........................................... 4, 84
spermatozoa antigens ........................... 143
spleen ........................................... 123, 125
SRBC ..................................................... 57
SRS-A .................................................. 156
stable complexes .................................. 192
stem cells.............................................. 130
steroids ................................................. 151
subclasses............................................... 25
superantigen ......................................... 100
suppression........................................... 135
suppressor cells .................................... 138
syngeneic................................................ 83
systemic lupus erythematosus (SLE) ... 146
T cell ............................................ 106, 125
T cell receptor (TCR)............................. 67
TATA................................................... 170
TC ......................................................... 113
TC cell ...................................... 91, 94, 111
TC-mediated lysis................................... 92
TcR............................................... 111, 112
tetanus .......................................... 164, 167
TH1 & TH 2 ............... 91, 94, 96, 111, 113

TH17 ............................................ 111, 113

thoracic duct................................. 107, 125
thymectomy.......................................... 105
thymic transplant.................................. 150
thymus.......................................... 105, 123
thymus-dependent (TD) antigen .......... 105
thymus-dependent area......... 108, 123, 125
thymus-independent (TI) antigens122, 127
thymus-independent area ............. 108, 123
thyroglobulin........................................ 143
TILs ...................................................... 173
tingible body macrophages .................. 125
tissue mast cell ..................................... 155
tolerance..................... 14, 53, 58, 129, 135
tonsils ................................................... 123
toxoids.................................................. 164
transcription ........................................... 64
transfusion reaction ................................ 78
translation............................................... 64
transplantation........................................ 85
transplantation antigens ......................... 79
Treg .............................................. 111, 113
TSTA.................................................... 170
tuberculin skin test ......................... 91, 154
tumor .................................................... 169
tumor suppressor genes........................ 169
tumor-associated antigens .................... 170
tumor-infiltrating leukocytes................ 173
tumor-specific antigens ........................ 170
two-way MLR ...................................... 205
type I, II, III, IV reactions............. 153, 154
umbilical cord blood ............................ 132
universal donor & recipient.................... 78
vaccination ........................................... 163
valency ................................................... 19
variable region.................................. 23, 27
variolation ............................................ 163
vascular anaphylaxis ............................ 157
vasoactive............................................... 34
vasodilation .......................................... 156
venule permeability.............................. 156
viruses .................................................. 170
wheal and flare ..................................... 158
X-linked infantile
hypogammaglobulinemia................. 148
yolk sac ................................................ 130

229