Immunology 1994 81 171-176

Thl/Th2 profiles in tuberculosis, based on the proliferation and cytokine

response of blood lymphocytes to mycobacterial antigens

H.-M. SURCEL,* M. TROYE-BLOMBERG,t S. PAULIEt G. ANDERSSON,4 C. MORENO,* G. PASVOL§ &

J. IVANYI* *MRC Tuberculosis and Related Infections Unit, Royal Postgraduate Medical School, Hammersmith Hospital,
London, U.K. tDepartment of Immunology, Stockholm University, Sweden, $Department of Medical Microbiology and
Clinical Immunology, University of Lund, Sweden, and § Unit of Infectious Diseases and Tropical Medicine,
St Mary's Hospital Medical School, Northwick Park Hospital, London, U.K.

SUMMARY
Proliferation and cytokine production profiles by blood mononuclear cells in response to in vitro
stimulation with mycobacterial antigens were compared in patients with active tuberculosis and in
sensitized healthy controls. Interleukin-4 (IL-4) and interferon-y (IFN-y) were detected at single-cell
level using the ELISPOT assay. Patients showed significantly (P < 0-01) increased numbers of IL-4-
secreting cells and decreased thymidine incorporation, but no significant difference in IFN-y-
producing cells in response to the 38,000 MW or 19,000 MW antigens and their immunodominant
peptide epitopes. Pronounced individual variations were found in both patient and control groups,
when comparing the responsiveness to the mycobacterial extract, two protein antigens and five
synthetic peptides. None of the antigens or peptides tested showed preferential stimulation of either
IL-4- or IFN-y-secreting T cells, and proliferation was not correlated with either IL-4 or IFN-y
production. In particular, cytokine responsiveness was of similar frequency in subjects who did or did
not show positive proliferation, indicating that the latter test was not fully representative of the active
T-cell repertoire. It is concluded that the demonstrated Th2 type of profile in response to two
prominent mycobacterial antigens may play a role in the mechanisms of defective host resistance in
tuberculosis.

INTRODUCTION cytokine profiles that they secrete upon antigen stimulation.6

Thi cells characteristically secrete interleukin-2 (IL-2) and
Resistance to tuberculosis (TB) depends crucially on antigen- interferon-y (IFN-y), whereas Th2 lymphocytes produce typi-
specific T-cell mediated activation of macrophages which are the cally IL-4, IL-5 and IL-10, which enhance antibody synthesis of
major effectors of cell-mediated killing of intracellular patho- B cells6 and play a role in allergic diseases.7 CD4+ T cells with an
genic mycobacteria. Lymphocyte proliferation in vitro against intermediate cytokine profile (ThO) have also been described.8
the 'purified protein derivative' (PPD), containing a large The secreted cytokines of Th I and Th2 cell types can mutually
variety of antigens, is decreased in tuberculosis patients.'A An regulate and inhibit each other's functions.9 Therefore, the fine
inverse relationship between T-cell proliferation and antibody balance between the secreted cytokines is important for the
levels to Mycobacterium tuberculosis antigens has been consid- resulting nature of host resistance against the pathogen. The
ered to be an indicator of defective resistance in tuberculosis Thl/Th2 diversification on the basis of cytokine secretion has
patients.3'5 recently been examined in parasitic and viral diseases9 and in
The profile of secreted cytokines in vitro is taken as indicative leprosy.'0 IL-2 and IFN-y coding mRNA was most evident in
of T-lymphocyte function in vivo, but the possible relationships tuberculoid leprosy, whilst mRNA for IL-4, IL-5 and IL-10 was
between antigen specificity and T-lymphocyte function have so predominant in lepromatous leprosy. These findings indicate
far not been examined. Murine CD4+ T cells have been divided that the Th 1 phenotype can be associated with certain forms of
into at least two different subsets (Thl and Th2), based on the disease as well as with protection.
The present study investigated ThI/Th2-like cell profiles in
patients with active tuberculosis. Peripheral blood mononuclear
Received 26 July 1993; revised 18 August 1993; accepted 21 October cells (PBMC) were stimulated in vitro with purified 38,000 MW
1993. and 19,000 MW antigens of M. tuberculosis, both representative
Abbreviations: BCIP/NBT, 5-bromo-4-chloro-3-indolyl phosphate of lipoproteins" which have previously been found to induce the
in dimethylformade/p-nitroblue tetrazolium chloride; Fmoc, g-fluorenyl most frequent and highest titres of serum antibodies in tubercu-
methoxy carbonyl; HOBT/PyBOP, 1-hydroxybenzotriazone/benzotria- losis.'2"3 Association of anti-38,000 MW antibody levels with
zone-l-yl-oxy-TRIS-pyrrolidine-phosphonium hexafluoro phosphate. the severity of multibacillary disease'4 made this antigen
Correspondence: Dr H.-M. Surcel, National Public Health Institute, particularly suitable for the study of Th2-like response profiles
Box 310, 90100 Oulu 10, Finland. in the human.
17i
172 H.-M. Surcel et al.
The immunogenic and genetically permissive T-cell epitopes Lymphocyte cultures
of the 38,000 MW and 19,000 MW antigens have been mapped PBMC were isolated from citrated blood by centrifugation on
using the proliferative assay.'5-'7 Corresponding synthetic pep- Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). The PBMC
tides were used to explore the relationships between epitope were suspended in RPMI-1640 (Gibco, Paisley, U.K.), supple-
specificity and T-cell function in this study. mented with 2 mM glutamine, 100 U/ml penicillin, 100 gg/ml
streptomycin and 5 % A+ heat-inactivated human serum. The
MATERIALS AND METHODS PBMC (1 x 105/well) were cultured in triplicate wells of 96-well
Subjects round-bottomed microtitre plates, in the presence of the
Nineteen patients (mean age 39-8+21-5 years) with active appropriate antigen, in a total volume of 200 p1, for 6 days in a
tuberculosis were diagnosed by routine clinical, bacteriological humidified 5% CO2 atmosphere at 37°. [3H]thymidine was
and histological parameters. Thirteen of the patients had active present in the cultures for the last 18 hr and the proliferation was
pulmonary tuberculosis, of which 10 had been confirmed as assayed from the incorporated radioactivity measured by using
smear positive. Six patients had extrapulmonary tuberculosis: a liquid scintillation counter. The results are expressed as
pleural (P7 + P1), spinalosteomyelitis (P2), jejunal (P8), cervical geometric mean counts per minute (c.p.m.) of triplicate cultures
abscess (P5) or cervical adenopathy (P23). These patients had or in terms of the stimulation index (SI). The SI was calculated
neither suspicion nor evidence of human immunodeficiency as a ratio of the mean proliferative response (c.p.m.) in the
virus (HIV) infection. Blood was drawn either before or within presence or absence of antigen. SI>3 was considered as a
2 weeks of the onset of chemotherapy. PPD-positive healthy positive response.16 Minimal antigen concentrations, which in
laboratory donors ranging in age from 23 to 58 years, who had preliminary experiments gave maximal lymphocyte prolifera-
no history of previous clinical tuberculosis, served as controls. tive responses, were used throughout this study. Thus, MTSE,
Patients and healthy donors had similar sex, race and age 38,000 and 19,000 MW proteins were used in a final concentra-
distribution, but were not matched. tion of 10 pg/ml, and the synthetic peptides were used at 25 pg/
ml.
Antigens
Mycobacterium tuberculosis, strain H37Rv (Difco, Detroit, MI), Cytokine-producing cells
soluble extract (MTSE) was prepared by mechanical disrup- Cytokine production was induced by incubating the isolated
tion. 13 PBMC (106/ml) in RPMI-1640 supplemented with 5% fetal calf
The 38,000 MW protein was isolated from the culture serum (FCS) in the presence or absence of phytohaemagglutinin
supernatant of M. tuberculosis, strain H37Rv, using 70-90% (PHA; 10 pg/ml; Wellcome, Beckenham, U.K.) or the appro-
ammonium sulphate and alcohol precipitation, followed by gel priate mycobacterial antigen.
filtration.18 Protein eluted from the DE52 gel bed with 0-1-0-2 M IFN-y- and IL-4-secreting cells were determined using a
NaCl showed only a single band of 38,000 MW in an over- modification of an ELISPOT assay described earlier.20 Briefly,
loaded sodium dodecyl sulphate-polyacrylamide gel electro- nitrocellulose-bottomed 96-well Millititer HA plates (Millipore
phoresis (SDS-PAGE) stained with Coomassie blue (results not Co., Bedford, MA) were coated with 100 p1 of monoclonal
shown). antibody (mAb) 82-4 for IL-4 (Department of Immunology,
The 19,000 MW recombinant protein expressed in M. University of Stockholm, Sweden) or mAb 7-B6-1 for IFN-y
smegmatis'9 was obtained as a gift from Dr T. Garbe (MRC, (Chromogenix AB, M6lndal, Sweden) at a concentration of 15
London, U.K.). pg/ml in phosphate-buffered saline (PBS), and incubated over-
night at 4°. Unbound antibodies were removed by four
Synthetic peptides successive washings with PBS.
Peptides comprising 16-20 amino acids, representing the Freshly isolated PBMC, suspended in RPMI-1640 medium
sequence of the previously determined T-cell immunogenic containing 5% FCS, were either incubated in round-bottomed
epitopes of the 38,000 MW and 19,000 MW protein antigens of 96-well plates for 72 hr before transfer to anti-IFN-y antibody-
M. tuberculosis, were synthesized by simultaneous multiple coated nitrocellulose-bottomed plates, or added directly to the
peptide synthesis, as described previously.'6 Briefly, the Fmoc anti-IL-4-coated wells (1 x 105 PBMC/well) in the presence of
methodology was used, with a trialkoxy-diphenyl-methylester antigen and incubated for 20 hr in a humidified 5% CO2
(Rink) resin and HOBT/PyBOP-activated coupling. After clea- atmosphere at 37°. After incubation, the cells were removed by
vage with trifluoroacetic acid (TFA), the crude peptides were washing the wells four times with PBS containing 0-05% Tween-
purified by gel filtration (Sephadex G-15, Pharmacia, 25% 80 (PBS-T). Biotin-conjugated mAb 12-1 for IL-4 (Department
aqueous acetic acid). Homogeneity was confirmed by reverse- of Immunology, University of Stockholm, Sweden) or mAb 1-
phase high-performance liquid chromatography (HPLC) (Zor- DIK for IFN-y (Chromogenix AB, Molndal, Sweden) was
bax ODS, Jones Chromatography, Hengoed, U.K., 0-1% TFA added (100 p1) to each well at optimal dilution (1-3 pg/ml) and
in water/acetonitrile) and amino acid sequences incubated for 3 hr at room temperature. Then the wells were
were determined by automated Edman degradation. The washed four times with PBS-T and exposed to 100 p1 streptavi-
sequences of peptides were derived from the 38,000 MW din-alkaline phosphatase (Mabtech AB, Stockholm, Sweden)
antigen: 38-A (1-20) MKIRLHLLAVLTAAPLLLA, 38-I for 1 hr. Unbound conjugate was removed by washing
(65-83) FNLWGPAFHERYPNVTITA, 38-G (350-369) thoroughly with PBS and, finally, 100 p1 of BCIP/NBT
DQVHFQPLPPAVVKLSDALI, or 19,000 MW antigen: 19- substrate solution (Bio-Rad Laboratories, Hercules, CA) was
6A (50-64) GAASGPKWIDGKDQN and 19-7 (61-80) added and incubated until the appearance of blue spots in the
VTGSVVCTTAAGNVNIAIGG. Each peptide used in these wells (30-40 min). The colour development was stopped by
experiments was from a single batch. extensive washing under running tap water and, after drying, the
 T-cell subsets in tuberculosis 173
number of spots was scored using a dissection microscope 10,0
MTSE 38,000 MW 19,000 MW
(Nikon, SMZ-2B, Tokyo, Japan). The mean error of the
duplicated counts for the ELISPOT assay was less than 20% of
the counted spot numbers. 10
A~~~~~~
C.)
HLA typing
HLA-DR genotypes were determined from genomic DNA co
D
0)
isolated from anti-coagulated whole peripheral blood by the C

RFLP typing system described by Bidwell et al.2'

a)UL) <I AI3 AA
A5AxlA
CD
Statistical analysis 1000,
MTSE 38,000 MW 19,000 MW
Statistical analysis of the lymphocyte proliferative responses 0

and cytokine levels was carried out using the Student's t-test.
Frequencies of positive responders among total groups were 6
100 - § A AA
z
compared between patients and controls using the x2 test with
Yates' correction.
RESULTS
I A
Lymphocyte proliferation and cytokine secretion responses to A A A
MTSE, 38,000 and 19,000 MW proteins < II Axi AXi A x5 A x4 A x6 A x3
P C P C P C
Lymphocyte proliferative reactivity to mycobacterial antigens
Figure 2. Frequency of cytokine-secreting cells following antigenic
was studied with PBMC from 19 patients with active tuberculo-
stimulation of 105 PBMC in vitro. Each symbol represents individual
sis and from 15 healthy controls. Figure 1 shows that PBMC values determined by ELISPOT assay ofcells from tuberculosis patients
from tuberculosis patients had slightly decreased proliferative (P) and sensitized healthy controls (C). Horizontal bars are mean values
responses (c.p.m.) against MTSE and 38,000 and 19,000 MW of cytokine-secreting cells.
protein compared with healthy individuals, but the differences
did not reach statistical significance. The high thymidine uptake
values in response to PHA represented the positive control and Table 1. Summary of results on anti-mycobacterial T-cell responses in
indicated good viability of the cultured PBMC. tuberculosis (P) and control (C) subjects
T-lymphocyte activation was further studied by analysing
IL-4 and IFN-y production at single-cell level using ELISPOT
assays in PBMC stimulated with individual mycobacterial % of subjects with elevated response
antigens. Figure 2 shows the numbers of IL-4- and IFN-y-
secreting cells after subtracting the spontaneous production of Proliferationt IL-4: IFN-yt
Stimulation
IL-4 (1-6+2-7/105 for controls and 2-3+2-9/105 for patients) with P C P C P C
and IFN-y (5-0+11-9/105 for controls and 0-9+1-4/105 for
patients). The average numbers of cytokine-secreting cells were Extract
generally elevated in TB patients following the stimulation of MTSE 95 93 79*** 20 69 70
PBMC with MTSE containing a wide range ofantigens, and this Antigens
was confirmed using the 38,000 and 19,000 MW proteins. The
38,000 MW 53 73 58*** 7 50 33
numbers of IL-4-producing cells were significantly elevated in 19,000 MW 47 80 74*** 13 49 38
Peptides
b 4;0 - El Patients 38-A 13 38 38 8 33 25
0 Controls 38-G 7 33 33 13 27 15
E 38-I 22 53 22 7 13 15
3
19-6A 47 33 50 13 13 38
c
0

i00 _ _ _: aL 19-7 40 61 40 8 33 18
0.

~0
'O . II
.- 4 *** P < 0-01 was obtained comparing the frequencies of positive
responders between tuberculosis patients and controls using x2 test with
Yates correction.
t ) rz () f °0)U 3 0 U) t SI>3.
t >2 positive cells after subtracting the background value.
PHA MTSE 38,000 MW 19,000 MW Medium

Figure 1. Proliferative responses (geometric mean + SD) of PBMC to PBMC from TB patients compared with controls in response to
PHA, MTSE or purified antigens. The ratio of responder/total tested the 38,00 MW antigen (6-8+7-8 versus 2-2+3-0, P<0-025).
individuals is given at the bottom of each column. Responders were Moreover, the 19,000 MW antigen induced more prominent
defined when thymidine counts (c.p.m.) in the presence of antigen were activation of IFN-y-producing cells in TB patients compared
three times above their individual values in the absence of antigen. with controls (mean number of spots 12-9 + 22-7 versus
174 H.-M. Surcel et al.
38,000 MW 38-A 38-G 38-1 19,000 MW 19-6A 19-7 HLA-DR
CK~ OK~ OD, AK A~,'jKA
Patients %3 <:@ 3\qk\V3 \Hi if\ ad Hi \ifN
P17 N
P6 N
P26 D RD 6
P15 NC
P3 ND N ND NCD
P4 ND ND ND NE
P21 NI
P28 D=l R 3,I I
P19 NI
P24 A __ DFR 8,13
P5 vND
P7 IIND
P2 DANDY ND
P9 DR5,3
P23 I DR 3
P8 I] DR5
DR 5
Controls
C5 V/VZVA I DR 4,10
ClI D ND DR 2,7
C2 I - DR 3,7
C3 --V4 ~7 a DR 1,2
C4 71'2- 11 I DR 3,7
C7
C9
-III I II II III 1UAI II I.{r
AI I1 lI ' =
ND
- DR3,8
CIO
C5I
-- 02- - -.- -
I I I I I I
ItZ7
-
I
I
I I
7AV7
I - -..
I VA
- IDR 2,7
- DR4,6
C16 -V4. I V-4 7~21. I .1 I E I VA I m - DR 1,3
Cs I DR 11,15
C13 1E 1EFI MA I I I I I N DR 3,7
DR 1,2
C14 -1. Il I l] I .1 1 - 1, m I
LP IFN-y IL-4
* >10 >50 >5
CM 3-10 2-50 2-5
E <3 <2 <2

Figure 3. Individual variation in T-cell responsiveness to immunodominant synthetic peptides. PBMC were cultured in the presence of
whole protein (38,000 and 19,000 MW) or peptides (38-A, G, I and 19-6A,7), derived from the 38,000 and 19,000 MW antigens. The
results for each stimulant are shown in the sequence of proliferation (LP, stimulation index), IFN-y- and IL-4-producing cells/105
stimulated PBMC. The symbols correspond to high (13), low (U) or no (0) response. ND, not done.

5-9+8-2, P<0005). No correlation was found between pro- Lymphocyte proliferation and cytokine secretion response to
liferation and numbers of IL-4 or IFN-y producing cells synthetic peptides
following antigenic stimulation.
Subsequently, all subjects were classified as either res- In view of the pronounced variations of proliferative responses
ponders or non-responders. The IFN-y or IL-4 response was in both TB patients and controls, it was of interest to evaluate
considered positive when the counted number of spots/l x 105 the occurrence of individual variations in the secretion of IL-4
PBMC in the presence of antigen minus the number of and IFN-y cytokines. It was of particular interest to find out if
spots/l x 105 PBMC in the absence of antigen was more than cytokine profiles would associate with distinct immunodomi-
two. nant epitopes, which have previously been identified for both
The IL-4 response to MTSE was significantly stronger in the the 38,000 MW and 19,000 MW antigens.'5"'7
patients than in the controls (79% versus 20%, P < 0-0 1). Of the TB patients showed positive proliferation (SI > 3) to several
TB patients 58% responded to 38,000 MW and 74% to 19,000 synthetic peptides less frequently than healthy controls (see Fig.
MW antigens with a positive activation of IL-4-producing cells, 3 and Table 1), but the relatively small size of groups in relation
while 7% of the controls showed a positive IL-4 response to the to low frequency of responders did not warrant further
38,000 MW and 13% to the 19,000 MW (P<0-01, Table 1). statistical evaluation. However, peptide-stimulated IL-4- and
MTSE induced a strong IFN-y response, both in the patients IFN-y-producing cell counts did not correlate with proliferation
and in controls (69% versus 70%, P>0 1). 49-50% of the in individual TB patients or control subjects. In response to
patients responded to the 38,000 MW and/or 19,000 MW synthetic peptides, positive proliferation (SI > 3) correlated with
antigens with a positive activation of IFN-y-producing cells, significant IL-4 and/or IFN-y production in only five out of 22
while 33% of the controls showed a positive IFN-y response to tested TB cases (Fig. 3). Thus, IL-4- and/or IFN-y-secreting cells
the 38,000 MW and 38% to the 19,000 MW antigen (P>0 1, were demonstrably at similar frequencies in patients who did or
Table 1). did not show antigen-stimulated cell proliferation.
 T-cell subsets in tuberculosis 175
None of the five peptides tested stimulated PBMC to argued that this result is due to differences in the HLA-DR
produce one cytokine (either IL-4 or IFN-y) in preference to the genotype among donors, as shown by Pfeiffer et al.,27 who
other, and this was the case both in controls and patients. The reported that the MHC genotype controlled the selective
synthetic peptides induced IL-4 24 times and IFN-y 27 times in activation of Thl - and Th2-type cells in response to the collagen
all tested subjects. IV peptide. We think this explanation for our data to be
unlikely; although HLA-DR typing was carried out in the
present work, mainly in the healthy subjects, there was no
DISCUSSION evident association of any of the HLA-DR haplotypes with
Defective proliferative responsiveness and deficient IL-2- and either IFN-y or IL-4 production. Our results are thus in
IFN-y secreting capacity to mycobacteria have previously been agreement with those of Haanen et al.28 who showed that human
reported in patients with severe tuberculosis.'"3 In the present T-cell clones secreted either IFN-y or IL-4, irrespective of HLA
study, lymphocyte proliferative reactivity to MTSE, the 38,000 haplotype. A corresponding result is represented by our finding
MW and the 19,000 MW protein did not significantly differ that IL-4 and IFN-y production to individual peptides was often
between patients and controls. Lower proliferative reactivity but not always mutually exclusive. It remains to be established
was pronounced in the case of the 38-G peptide, which has whether double producers (IFN-y and IL-4) comparable to the
previously been reported to cause relative anergy in tuberculosis ThO phenotype described for the mouse8 are present in the
patients. 16 PBMC of tuberculosis patients.
Characterization of functional T-cell subsets against iso- Higher IL-4 production in tuberculosis patients can explain
lated mycobacterial antigens was based on the enumeration of the increased antibody levels that have been found previously in
IL-4- and IFN-y-producing cells following stimulation with active tuberculosis.'4 Moreover, IL-4 together with other cyto-
synthetic peptides derived from the 38,000 MW and 19,000 MW kines may also act as an influential down-regulator of Th 1-type
antigens, which have been shown previously to be immunogenic responsiveness in infections,29 especially if it is secreted at an
in human.'5 17 Interestingly, cytokine production was marked, early stage of the immune response.30 3' Thus it is tempting to
even in cases where PBMC proliferation to a given antigen was speculate that IL-4 production may be involved in the loss of
not detectable. This result underlines the importance of measur- protective host response and linked to the pathogenesis of
ing several parameters when studying T-cell activation and is tuberculosis.
thus in accordance with comparable analyses of responses ACKNOWLEDGMENTS
directed to malaria antigens.22 When studying the T-cell reac-
tions in the course of experimental M. tuberculosis infection, We are grateful to Dr Peter Klouda from the U.K. Transplant Support
T-cell proliferation was an insufficient indicator of antigen Service, Bristol, U.K. and Mr Paul Brookes from the Department of
Immunology, RPMS, London, U.K. for HLA typing. We also thank
recognition, and elevated levels of IFN-y were secreted from Mr A. Hills, MRC Tuberculosis and Related Infections Unit, London,
apparently non-proliferating T cells.23 According to our results, U.K., for preparing the synthetic peptides.
the numbers of IFN-y-secreting cells did not differ significantly
between patients and controls. Thus, the results are not in REFERENCES
accordance with the suggestion that deficient IFN-y production
is involved with active tuberculosis.' 3In fact, increased secre- 1. ONWUBALILI J.K., SCOTT G.M. & ROBINSON J.A. (1985) Deficient
immune interferon production in tuberculosis. Clin. exp. Immunol.
tion of IFN-y was found slightly more frequently in patients 59,405.
than in controls after 3 days of antigen stimulation. Convention- 2. Toossi Z., KLEINHENZ M.E. & ELLNER J.J. (1986) Defective
ally, IFN-y is measured in 5-7-days culture supernatants, and interleukin 2 and responsiveness in human pulmonary tuberculosis.
IFN-y levels are found to be at least marginally decreased in TB J. exp. Med. 163, 1162.
patients.' 3 These results may reflect different kinetics of IFN-y 3. HUYGEN K., VAN VOOREN J.-P., TURNEER M., BOSMANs R., DIERCKX
production between patients and controls. In vivo-activated P. & DE BRUYN J. (1988) Specific lymphoproliferation, gamma
T cells in the TB patients may result in faster detection of interferon production, and serum immunoglobulin G directed
cytokine production in vitro. against a purified 32 kDa mycobacterial protein antigen (P32) in
Earlier studies based on the characterization of PPD- patients with active tuberculosis. Scand. J. Immunol. 27, 187.
stimulated human T-cell clones claimed that IL-4 production is 4. RIBERA E., ESPANOL T., MARTINEZ-VAZQUES M., OCANA I. &
ENCABO G. (1990) Lymphocyte proliferation and gamma-inter-
not involved in immune reactivity against M. tuberculosis feron production after 'in vitro' stimulation with PPD. Chest,
infection.2426 We have demonstrated here that PBMC from TB 97, 1381.
patients responded to MTSE with IL-4 reactivity more fre- 5. HAVLIR D.V., WALLIs R.S., BooM W.H., DANIEL T.M., CHERVENAK
quently than PBMC from healthy subjects. This result was K. & ELLNER J.J. (1991) Human immune response to Mycobacterial
confirmed using the 38,000 MW and 19,000 MW antigens. Even tuberculosis antigens. Infect. Immun. 59, 665.
though low, the increase in the number of IL-4-producing cells 6. MOSMANN C.N.L., CHERWINSKI H., BOND M.W., GIEDLIN M.A. &
after antigen stimulation was convincing and comparable to our COFFMAN R.L. (1986) Two types of murine helper T cell clone. I.
experience using tetanus toxoid as T-cell stimulating antigen (El Definition according to profiles of lymphokine activities and
Ghagali et al. manuscript in preparation). Whether this is a secreted proteins. J. Immunol. 136, 2348.
reflection of antigen-specific IL-4-producing cells being few, or 7. CORRIGAN C.J. & KAY A.B. (1992) T cells and eosinophils in the
pathogenesis of asthma. Immunol. Today 13, 501.
that they are down-regulated by other factors or that they are 8. STREET N.E., SCHUMACHER J.H., FONG T.A.T., BASS H., FIOREN-
located somewhere else is not known. TINO D.F., LEVERAH J.A. & MOSMANN T.R. (1990) Heterogeneity of
An interesting result obtained in this study was that PBMC mouse helper T cells: evidence from bulk cultures and limiting
from different individuals responded to several synthetic pep- dilution cloning for precursors of Thl and Th2 cells. J. Immunol.
tides but without preference for either IFN-y or IL-4. It could be 144, 1629.
176 H.-M. Surcel et al.
9. SHER A., GAZZINELLI R.T., OSWALD I.P., CLERICI M., KULLBERG that positively identifies serologically well-defined and ill-defined
M., PEARCE E.J. et al. (1992) Role of T-cell derived cytokines on the HLA-DR and DQ alleles, including Dew 10. Transplantation,
downregulation of immune responses in parasitic and retroviral 45, 640.
infection. Immunol. Rev. 127,183. 22. TROYE-BLOMBERG M., RILEY E.M., KABILA L., HOLMBERG M.,
10. YAMAMURA M., UYEMURA K., DEANS R.J., WEINBERG K., REA T.H., PERLMANN H., ANDERSSON U., HEUSSER C.H. & PERLMANN P. (1990)
BLOOM B.R. & MODLIN R.L. (1991) Defining protective responses Production by activated human T cells of interleukin 4 but not
to pathogens: cytokine profiles in leprosy lesion. Science, 254, 277. interferon-y is associated with elevated levels of serum antibodies to
11. YOUNG D.B. & GARBE T.R. (1991) Lipoprotein antigens of activating malaria antigens. Proc. natl. Acad. Sci. U.S.A. 87, 5484.
Mycobacterium tuberculosis. Res. Microbiol. 142, 55. 23. ORME I.M., MILLER E.S., ROBERTS A.D., FURNEY S.K., GRIFFIN
12. HEWITT J.A.R.M., COATES D.A., MITCHISON D.A. & IVANYI J. J.P., DOBos K.M., CHI D., RIVOIRE B. & BRENNA P.J. (1992) T
(1982) The use of murine monoclonal antibodies without purifica- lymphocytes mediating protection and cellular cytolysis during the
tion of antigen in the serodiagnosis of tuberculosis. J. Immunol. course of Mycobacterium tuberculosis infection. J. Immunol.
Meth. 55, 205. 148, 189.
13. JACKETT P.S., BOTHAMLEY G.H., BATRA H.V., MISTRY A., YOUNG 24. DEL PRETE G.F., DE CARLI M., MASTROMAURO C., BIAGIOTTI R.,
D.B. & IVANY1 J. (1988) Specificity of antibodies to immunodomi- MACCHIA D., FALAGIANI P., RIcCI M. & ROMAGNANI S. (1991)
nant mycobacterial antigens in pulmonary tuberculosis. J. clin. Purified protein derivative of Mycobacterium tuberculosis and
Microbiol. 26, 2313. excretory-secretory antigen(s) of Toxoplasma canis expand in vitro
14. BOTHAMLEY G.H., RUDD R., FESTENSTEIN F. & IVANYI J. (1992) human T cells with stable and opposite (type 1 T helper ao type 2 T
Clinical value of the measurement of Mycobacterium tuberculosis- helper) profile of cytokine production. J. cdin. Invest. 88, 346.
specific antibody in pulmonary tuberculosis. Thorax, 47, 270. 25. BARNES P.F., ABRAMS J.S., Lu S., SIELING P.A., REA T.H. & MODLIN
15. FAITH A., MORENO C., LATHIGRA R., ROMAN E., FERNANDEZ M., R.L. (1993) Patterns of cytokine production by mycobacterium-
BRETT S., MITCHELL D.M., IVANYI J. & REES A.D.M. (1991) reactive human T-cell clones. Infect. Immun. 61, 197.
Analysis of human T-cell epitopes in the 19,000 MW antigen of 26. BooM W.H., WALLIS R.S. & CHERVENAK K.A. (1991) Human
Mycobacterium tuberculosis: influence of HLA-DR. Immunology, Mycobacterium tuberculosis-reactive CD4+ T-cell clones: hetero-
74, 1. geneity in antigen recognition, cytokine production, and cytotoxi-
16. VORDERMEIER H.-M., HARRIS D.P., FRISCIA G., ROMAN E., SURCEL city for mononuclear phagocytes. Infect. Immun. 59, 2737.
H.-M., MORENO C., PASVOL G. & IVANYI J. (1992) T cell repertoire 27. PFEIFFER C., MURRAY J., MADRI J. & BOTTOMLY K. (1991) Selective
in tuberculosis: selective anergy to an immunodominant epitope of activation of Thl- and Th2-like cells in vivo. Response to human
the 38-kDa antigen in patients with active disease. Eur. J. Immunol. collagen IV. Immunol. Rev. 123, 65.
22, 2631. 28. HAANEN J.B.A.G., DE WAAL MALEFIJT R., RES P.C.M., KRAAKMAN
17. HARRIS D.P., VORDERMEIER H.M., FRISCIA G., ROMAN E., SURCEL E.M., OTTENHOFF T.H.M., DE VRIES R.R.P. & SPITS H. (1991)
H-M., PASVOL G., MORENO C. & IVANYI J. (1993) Genetically Selection of a human T helper type 1-like T cell subset by
permissive recognition of adjacent epitopes from the 1 9-kDa mycobacteria. J. exp. Med. 174, 583.
antigen of Mycobacterium tuberculosis by human and murine T 29. APPELBERG R., ORME I.M., DE SOUSA M.I.P. & SILVA M.T. (1992) In
cells. J. Immunol. 150, 5041. vitro effects of interleukin-4 on interferon-y-induced macrophage
18. ESPITIA C., CERVERE I., GONZALEZ R. & MANCILLA R. (1989) A 38- activation. Immunology, 76, 553.
kDa Mycobacterium tuberculosis antigen associated with infection. 30. LIEW F.Y., LI Y., SEVERN A., MILLOTT S., SCHMIDT J., SALTER M. &
Its isolation and serologic evaluation. Clin. exp. Immunol. 77, 373. MONCADA S. (1991) A possible novel pathway of regulation by
19. GARBE T., HARRIS D., VORDERMEIER M., LATHIGRA R., IVANYI J. & murine T helper type-2 (Th2) cells of a Thl cell activity via the
YOUNG D. (1993) Expression of the Mycobacterium tuberculosis 19- modulation of the induction of nitric oxide synthase on macro-
Kilodalton antigen in Mycobacterium smegmatis: Immunological phages. Eur. J. Immunol. 21, 2489.
analysis and evidence of glycosylation. Infect. Immun. 61, 260. 31. ABEHSIRA-AMAR O., GIBERT M., JOLIY M., THEZE J. & JANKOVIC
20. KABILAN L., ANDERSSON G., LOLLI F., EKRE H.-P., OLSSON T. & D.L. (1992) IL-4 plays a dominant role in the differential develop-
TROYE-BLOMBERG M. (1990) Detection of intracellular expression ment of ThO into Thl and Th2 cells. J. Immunol. 148, 3820.
and secretion of interferon-y at the single-cell level after activation 32. VAN DER POUW-KRAAN T., DE JONG R. & AARDEN L. (1993)
of human T cells with tetanus toxoid in vitro. Eur. J. Immunol. Development of human Thl and Th2 cytokine responses: the
20, 1085. cytokine production profile of T cells is dictated by the primary in
21. BIDWELL J.L., BIDWELL E.A., SAVAGE D.A., MIDDLETON D., vitro stimulus. Eur. J. Immunol. 23, 1.
KLOUDA P.T. & BRADLEY B.A. (1988) A DNA-RFLP typing system