Protein Cell

DOI 10.1007/s13238-016-0323-0 Protein & Cell

REVIEW
Antibody-drug conjugates: recent advances
in conjugation and linker chemistries
Kyoji Tsuchikama& , Zhiqiang An

Texas Therapeutics Institute, The Brown Foundation Institute of Molecular Medicine, The University of Texas Health Science
Center at Houston, Houston, TX 77054, USA
& Correspondence: Kyoji.Tsuchikama@uth.tmc.edu (K. Tsuchikama)
Received July 5, 2016 Accepted August 6, 2016

Protein & Cell

ABSTRACT chemotherapy (DeVita and Chu, 2008). Chemotherapy using
cytotoxic agents is a major treatment option, in addition to
The antibody-drug conjugate (ADC), a humanized or
surgical removal, radiation, targeted therapies using small
human monoclonal antibody conjugated with highly
molecules or monoclonal antibodies (An, 2010), and, more
cytotoxic small molecules (payloads) through chemical
recently, immunotherapy. Chemotherapy has been refined
linkers, is a novel therapeutic format and has great
through screening and development of small molecules that
potential to make a paradigm shift in cancer chemother-
can cause cell death selectively to cancer cells through
apy. This new antibody-based molecular platform enables
inhibiting microtubule function, DNA synthesis, or protein
selective delivery of a potent cytotoxic payload to target
function. Although chemotherapy has seen great success in
cancer cells, resulting in improved efficacy, reduced
treatment of cancer, especially leukemia, difficult issues
systemic toxicity, and preferable pharmacokinetics (PK)/
remain, such as the development of resistance mechanisms.
pharmacodynamics (PD) and biodistribution compared
Severe adverse effects derived from off-target cytotoxicity
to traditional chemotherapy. Boosted by the successes of
may worsen a patient’s quality of life, contributing to dis-
FDA-approved Adcetris® and Kadcyla®, this drug class
continuation of medication. This fact has discouraged clini-
has been rapidly growing along with about 60 ADCs cur-
cians and medicinal chemists from pursuing more highly
rently in clinical trials. In this article, we briefly review
potent cytotoxic agents for cancer treatment. In this context,
molecular aspects of each component (the antibody,
the use of highly cytotoxic agents conjugated with cell-tar-
payload, and linker) of ADCs, and then mainly discuss
geting molecules emerged as a potential clinical strategy. In
traditional and new technologies of the conjugation and
particular, antibody-drug conjugates (ADCs), humanized or
linker chemistries for successful construction of clini-
human monoclonal antibodies conjugated with cytotoxic
cally effective ADCs. Current efforts in the conjugation
small molecules through chemical linkers, could potentially
and linker chemistries will provide greater insights into
make a fundamental change in the way cancer chemother-
molecular design and strategies for clinically effective
apy is designed and administered (Chari et al., 2014; Perez
ADCs from medicinal chemistry and pharmacology
et al., 2014; Bouchard et al., 2014; Jain et al., 2015;
standpoints. The development of site-specific conjuga-
McCombs and Owen, 2015; Chudasama et al., 2016; Dia-
tion methodologies for constructing homogeneous ADCs
mantis and Banerji, 2016). This platform enables targeting
is an especially promising path to improving ADC design,
cancer cells and selective delivery of highly cytotoxic drugs,
which will open the way for novel cancer therapeutics.
resulting in a broad therapeutic window. Indeed, successful
clinical outcomes using ADCs have inspired scientists in the
KEYWORDS antibody-drug conjugates, cancer,
biomedical research community to further advance this new
chemotherapy, conjugation, linker, site-specific conjugation
platform towards next-generation cancer therapeutics. In this
article, we review molecular aspects of ADCs, successful
INTRODUCTION ADCs currently used in clinical application, and recent pro-
gress in the conjugation and linker technologies for suc-
Over the past half century, cancer management has
cessful construction of ADCs.
improved significantly along with the advancement of

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

BRIEF HISTORY OF ADC topoisomerase or RNA polymerase inhibition, leading to cell

death. Cytotoxic chemical agents that have high potency to
The concept of selective delivery of toxic agents to target cells
cancer cells but low off-target cytotoxicity are generally used
causing disease was originally proposed in 1913 by German
as payload. Chemical structures of Mylotarg®, Adcetris®,
physician and scientist Paul Ehrlich (Ehrlich, 1913). Forty five
and Kadcyla® are depicted in Fig. 2.
years later, his concept of targeted therapy was first demon-
strated in the form of an ADC, methotrexate conjugated to a
leukemia cell-targeting antibody (Mathe et al., 1958). In early CHOICE OF ANTIGEN AND PAYLOAD
studies, polyclonal antibodies were the main targeting agents.
Given the mechanism of action, the ideal antibody needs to
The first ADC human clinical trial was conducted using an anti-
have sufficient antigen affinity and specificity. However,
carcinoembryonic antigen antibody-vindesine conjugate in
antibodies with extremely high antigen affinity are known to
1983 (Ford et al., 1983), and a promising outcome was
lead to reduced efficiency of solid tumor penetration (Rud-
reported. Technological advancements in antibody engineer-
nick et. al., 2011). Thus, ADCs with high antigen affinity do
ing, including production of humanized antibodies, boosted
not necessarily lead to high clinical efficacy. In addition, cell-
studies on ADC. The first-generation ADCs consisting of chi-
surface antigens must be predominately expressed on target
meric or humanized antibodies, were tested in the 1990s.
cells with minimal expression on healthy cells to achieve
Finally, further significant efforts towards practical therapeu-
effective drug delivery and selective killing of tumor cells,
tics led to FDA-approved ADCs: gemtuzumab ozogamicin
Protein & Cell

which determines the therapeutic window. In this context,

(Mylotarg®) in 2000 for CD33-positive acute myelogenous
one may think tumor antigen density directly correlates to
leukemia (Sievers et al., 2001), brentuximab vedotin (Ad-
efficacy of ADCs. However, several studies suggest that the
cetris®) in 2011 for CD30-positive relapsed or refractory
correlation between antigen density and ADC efficacy
Hodgkin’s lymphoma and systemic anaplastic large cell lym-
depends on the type of cancer cells (Polson et al., 2011;
phoma (Younes et al., 2010), and trastuzumab emtansine
Kung Sutherland et al., 2013) due to varying internalization
(Kadcyla®) in 2013 for HER2-positive breast cancer (LoRusso
rate of each antigen after formation of a complex with an
et al., 2011; Verma et al., 2012). However, Mylotarg® was
ADC molecule. While important to select a cancer cell-
withdrawn from the market in 2010 due to a lack of clinical
specific antigen, the prediction of total efficacy of ADCs
benefit and high fatal toxicity rate compared to the standard
based on the antigen expression level remains elusive
chemotherapy (ten Cate et al., 2009). In spite of this setback,
(Damelin et. al., 2015).
ADC technologies have been rapidly evolving and about 60
Another important consideration is the limited number of
ADCs are currently in clinical trials (Diamantis and Banerji,
payload molecules that can be efficiently delivered into target
2016). In addition to immunotherapy with checkpoint inhibitors
cells. Only 1.56% of administered drug molecules can enter
(Postow et al., 2015), this emerging molecular platform for
target cells if the efficiency of each step in the ADC mechanism
chemotherapy is predicted to significantly increase its share of
is assumed to be 50% (biodistribution, binding to antigen,
the market as one of the most effective anti-cancer thera-
internalization, release of payload, intracellular stability of
peutics in the near future (Mullard, 2013).
payload, and payload binding to target) (Teicher and Chari,
2011). Indeed, the actual uptake is estimated to be much lower
than this assumption (<0.01% injected dose per gram of
STRUCTURE AND MECHANISM OF ACTION OF ADC
tumor) (Sedlacek et al., 1992). Thus, to maximize treatment
ADCs comprise monoclonal antibodies and cytotoxic agents efficacy using ADCs, cytotoxic potency of payload is required
(payloads) covalently conjugated through chemical linkers to be high enough to effectively eradicate target cells, ideally in
(Fig. 1A). In modern research and development of ADCs, the picomolar range. While important to select highly potent
humanized or fully human monoclonal antibodies (hmAbs) toxic agents as payload, ideal agents have inherent selectivity
are the first choice of delivery platform to secure high cell for target cancer cells. Certain types of noncancerous cells
target specificity, long circulating half life in human blood- may be capable of internalizing ADCs through nonspecific
stream (up to three weeks in the case of immunoglobulin G pinocytosis or fragment crystallizable (Fc) region receptor-
(IgG)), and minimal immunogenicity. A general mechanism mediated endocytosis (Lencer and Blumberg, 2005). Fur-
of action of ADCs is depicted in Fig. 1B. After ADC mole- thermore, payload may be released upon degradation into
cules are administered into the blood stream, the antibody circulating blood. Thus, payloads have primarily been selec-
component of the ADC recognizes and binds to cell-surface ted based on the above-mentioned consideration; antimitotic
antigens that are highly expressed in target cancer cells. agents, which are generally less toxic to noncancerous cells
Upon internalization of the ADC-antigen complex through than to cancerous cells, are payloads that have been mainly
endocytosis, the complex is processed within lysosomes, used in the FDA-approved ADCs and ADCs in clinical trials. In
which releases the cytotoxic payload (antimitotic agents in addition to calicheamicins (used in Mylotarg®), auristatins
general) in a bioactive form inside the cell. The released (used in Adcetris®), and maytansinoids (used in Kadcyla®),
payload disrupts DNA strands or microtubules, or exerts new classes of highly potent antimitotic compounds have also

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

A B
Conjugation site
lysine coupling, cysteine alkylation,
enzymatic reaction, etc.
(6)
(1) Cell death
Binding to
mAb cell-surface
humanized or Linker antigen
fully human cleavable or Payload
non-cleavable antimitotic agent
(5)
DNA or microtubule disruption
Key factors
- High potency - High cancer cell specificity (2)
- Low immunogenicity - Long circulating life Endocytosis of
- Low cytotoxicity to off-target cells ADC-antigen
complex

(4)
Release of
(3) active payload
Lysosomal degradation

Protein & Cell

Figure 1. Structure and mechanism of action of ADC. (A) A general structure of an ADC containing a humanized/human
monoclonal antibody (mAb), a cleavable/non-cleavable chemical linker, and a cytotoxic payload. The linker is covalently linked to the
mAb at the conjugation site. (B) A general mechanism of action of ADCs. The ADC binds to its target cell-surface antigen receptor
(Step 1) to form an ADC-antigen complex, leading to endocytosis of the complex (Step 2). The internalized complex undergoes
lysosomal processing (Step 3) and the cytotoxic payload is released inside the cell (Step 4). The released payload binds to its target
(Step 5), leading to cell death (Step 6).

been explored for ADC payloads: duocarmycins, maytansinoid-based ADCs and adverse toxicity (Drake and
pyrrolobenzodiazepine dimers (PBDs), amanitins, and tubu- Rabuka, 2015). Therefore, it is important to identify ADC
lysin analogs are such examples (Chari et al., 2014; Perez linkers with optimal linker stability for each combination of
et al., 2014). antigen, target tumor type, and payload. (2) At the same
time, the linker needs to possess the ability to be rapidly
cleaved and to release free and toxic payload once the ADC
THE CONJUGATION AND LINKER CHEMISTRIES
is internalized into the target tumor cell. (3) Another property
FOR ADC
to be considered in the linker design is hydrophobicity.
Though it is important to select optimal target-specific anti- Hydrophobic linkers coupled with hydrophobic payloads
bodies and potent payloads based on type of cancer cells, often promote aggregation of ADC molecules. For example,
the conjugation and linker chemistries are also crucial King and co-workers observed non-covalent dimerization of
components for successful construction of an ADC and the the monoclonal antibody BR96 conjugated with doxorubicin
major topics of this review. The linker moiety covalently through a multi-loading, hydrophobic dipeptide linker (King
tethers the antibody and payload components. Its molecular et al., 2002). Such molecules are unfavorable in the pursuit
design and properties are critical determinant factors for of therapeutically useful ADCs; aggregated proteins tend to
ADC efficacy in terms of pharmacokinetics (PK)/pharmaco- be rapidly sequestered in the liver and cleared by the retic-
dynamics (PD) and therapeutic window. To maximize these uloendothelial system, resulting in hepatotoxicity (Finbloom
parameters, various types of linkers have been developed et al., 1980). In addition, aggregated proteins are likely to
and evaluated in vitro and in vivo. Several criteria must be function as immunogenic substances, provoking undesired
met for successful ADC construction. (1) The linker needs to immune response during circulation in bloodstream. This
possess sufficient stability in plasma so that ADC molecules problem can be overcome by employing hydrophilic linkers
can circulate in the bloodstream and localize to the tumor containing negatively charged sulfonate groups (Zhao et al.,
site without premature cleavage. Instability of the linker 2011), polyethylene glycol (PEG) groups (Lyon et al., 2015),
causes premature liberation of the toxic payload and unde- or pyrophosphate diester groups (Kern et al., 2016).
sired damage to non-target healthy cells, which can lead to Based on the above-mentioned criteria, tremendous effort
systemic toxicity and adverse effects. However, a clinical has been put toward developing conjugation methods and
study revealed reverse correlation between linker stability of ADC linker structures. Chemical conjugation and enzymatic

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

O O
O S N
N O HO
H NHCO 2Me O
N S
N S NH N O
αCD33 mAb H H O O O O Cl
O
I O H αHER2 O N OMe
S O N O mAb
O O
HO O
O OMe OH O
OMe Et
O O O N
HO AcN H OH
MeO OMe
MeO
OH
Gemtuzumab ozogamicin (Mylotarg ®) Trastuzumab emtansine (Kadcyla ®)
for CD33-positive acute myelogenous leukemia treatment for Her2-positive breast cancer treatment
FDA-approved in 2000; withdrawn in 2010 FDA-approved in 2013

Et
O O OH
H H
O N N N
O O O N N
S H
N N O OMe O OMe O
N N
O H H
αCD30 mAb O
Protein & Cell

NH Brentuximab vedotin (Adcetris ®)

for CD30-positive relapsed or refractory Hodgkin's lymphoma treatment
O NH 2 FDA-approved in 2011

Figure 2. Structures of FDA-approved ADCs: Mylotarg®, Adcetris®, and Kadcyla® (blue: linker, red: payload).

conjugation are two methods for tethering the antibody and However, there are about 80 lysine residues on a typical
payload components that are currently in use. Linker struc- antibody and about 10 residues are chemically accessible
ture is categorized into two major classes based on the (Chari, 2008). Thus, this conjugation modality often gives
payload release mechanism: cleavable or non-cleavable multiple ADC species with variable DARs and conjugation
linker. Herein, we review modern conjugation methods and sites. In the case of a maytansinoid-type ADC, the average
ADC linker technologies in detail. DAR was 3.5–4 with distribution between 0–7 (Lazar et al.,
2005). As described above, DAR and its distribution critically
Chemical conjugation impact PK/PD and cytotoxicity of ADCs. Furthermore, some
lysine residues that are critical in antibody-antigen interac-
In ADC chemical conjugation, accessible amino acid resi-
tions may be modified, resulting in reduced binding affinity.
dues on the surface of the antibody undergo a controlled
As such, heterogeneous mixtures of ADCs constructed
reaction with a reaction handle installed on the linker.
using this conjugation method could potentially lead to a
Depending on the chemical conjugation method selected,
poor therapeutic index. While achievable as seen in the
this process affords a mixture of ADC species with variable
FDA-approved Kadcyla® and clinically tested ADCs, the
Drug-Antibody Ratios (DARs) and tethering sites. In general,
lysine-based conjugation requires effort to develop repro-
a broad distribution of DAR can lead to reduced efficacy, and
ducible manufacturing processes ensuring controlled DAR
thus the distribution needs to be tightly controlled. High DAR
and distribution within a target range (typically 3–4 as a
can increase not only potency but also the risk of aggrega-
major species).
tion, clearance rate, and premature release of the toxic
payload during circulation. This risk can be reduced by
Cysteine coupling
employing hydrophilic, sufficiently stable linkers. Overall, it is
crucial to identify an optimal DAR value with a controlled Cysteine-based conjugation methods rely on a specific
distribution for each ADC that can maximize the balance of reaction between cysteine residues of the antibody and a
efficacy, tolerability, and cytotoxicity profiles. thiol-reactive functional group installed on the payload
(Fig. 4A). In general, antibodies do not possess free thiols,
and all cysteine residues form disulfide bonds. In human
Lysine amide coupling
IgG1, which is most commonly used in modern ADCs, there
Amide coupling is a major ADC conjugation method con- are 4 interchain and 12 intrachain disulfide bonds. The 4
necting a payload and solvent accessible lysine residues on interchain disulfides, which are generally not critical for
the antibody using linkers containing activated carboxylic structural stability of IgG1, can be selectively reduced under
acid esters (Fig. 3). Amide coupling of an amine and an mild conditions to give 2, 4, 6, or 8 free thiols while keeping
activated carboxylic acid is one of the most reliable, high- the 12 intrachain disulfides intact. Due to the limited number
yielding chemical conversions in organic synthesis. of conjugation sites and the distinct reactivity of the thiol

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

O
-
O 3S
+ N O
NH 2 NH

O O Payload or another
O
reaction handle Average DAR: 3.5-4
DAR distribution: 0-7

Figure 3. Lysine amide coupling. An activated carboxylic acid moiety reacts with a lysine residue, which results in amide bond
linkage between mAb and the payload. Optimized conjugation conditions give an average drug-to-antibody ratio (DAR) value of 3.5–4
with distribution between 0–7.

group, cysteine-based conjugation is superior to lysine- recombinant HER2 (Kd = 0.1–0.5 nmol/L) comparable to that
based conjugation in terms of controlled DAR and hetero- of the parent trastuzumab. Another advantage of this method
geneity. As is the case with the lysine-based conjugation, is that the resulting aryl-cysteine conjugates are stable to-
this conjugation method was a major choice for ADC con- wards acids, bases, oxidants, and externally added thiols.

Protein & Cell

struction and used for Adcetris® and many other ADCs in While unique and intriguing, this method needs substantial
clinical trials. However, this modality still has room for modification or improvement of several critical factors for the
improvement to achieve better DAR and heterogeneity future clinical application (toxicity of palladium, workup
control: the above-mentioned simple cysteine conjugation strategies for complete removal of palladium, cost for palla-
can give a DAR distribution raging from 0 and 8. Junutula dium complexes, DAR control, etc.). Despite current limita-
and co-workers introduced two new cysteine residues (one tions, further work will provide researchers with profound
per heavy chain) for selective antibody attachment (Junutula insights into cysteine-based conjugation chemistry and
et al., 2008). This engineered cysteine technology, THIO- rational design of reagents for preparing ADCs that could
MAB, enables generation of highly homogeneous ADCs with have not been constructed with traditional methods.
a DAR of 2 (>90% homogeneity). ADCs constructed using
this technology have shown quite encouraging results (high
Non-natural amino acid incorporation by genetic engineering
efficacy and therapeutic window) in in vivo studies (Junutula
et al., 2008). Cysteine rebridging is another strategy that was Installation of non-natural amino acid residues with a reac-
recently developed to better control DAR and heterogeneity tion handle is a strategy that allows for a site-specific
of ADCs. Dibromomaleimide (Behrens et al., 2015; Bryden chemical conjugation, leading to strictly controlled DARs.
et al., 2014), dibromopyridazinediones (Maruani et al., Schultz and co-workers have developed protein expression
2015), and a 1,3-bis(p-toluenesulfonyl)propane-based core systems (bacteria, yeast, and mammalian cells) where p-
(Bryant et al., 2015) can accept two reduced cysteines acetylphenylalanine containing a carbonyl group is geneti-
derived from interchain disulfide bonds to afford a rebridged cally encoded by introducing a unique codon-tRNA syn-
antibody (Fig. 4B). These site-specific conjugations theo- thetase (Axup et al., 2012; Tian et al., 2014) (Fig. 5A).
retically provide many advantages in terms of structural Engineered antibodies containing p-acetylphenylalanine
stability, homogeneity, and well-controlled DAR (predomi- residues are produced using either of the expression sys-
nant at 4 in the case of dibromomaleimide) (Behrens et al., tems, and the carbonyl groups introduced react with alkox-
2015). Coupled with selection of proper linker structure and yamine-functionalized linkers to provide oxime-conjugated
payload, this method can potentially lead to ADCs with ADCs. Other examples are p-azidomethyl-L-phenylalanine
enhanced PK/PD and therapeutic efficacy. (Zimmerman et al., 2014) and N6-((2-azidoethoxy)car-
Recently, Buchwald and co-workers developed a rapid, bonyl)-L-lysine (VanBrunt et al., 2015) (Fig. 5B). The incor-
highly selective cysteine conjugation using aryl palladium porated azide groups are used for conjugation with alkyne-
complexes (Vinogradova et al., 2015) (Fig. 4C). The aryl functionalized linkers through the copper-catalyzed Huisgen
palladium reagents are readily prepared by mixing active cycloaddition (called “click chemistry” in general) to provide
palladium-phosphine complexes and various aryl halides. triazole-linked ADCs. Zimmerman et al. used this method to
The resulting complexes undergo a thiol arylation with conjugate monomethyl auristatin F (MMAF) with the trastu-
reduced cysteine residues of the antibody in a rapid and zumab, which afforded a potent ADC (Zimmerman et al.,
selective manner. They demonstrated the potential of this 2014). All functional groups in the antibody sequence are
new method in direct conjugation with trastuzumab and a tolerant of both conjugation reactions. Thus, DARs can be
palladium complex of vandetanib (a kinase inhibitor), which tightly controlled by adjusting the degree of non-natural
gave a linker-free ADC with a DAR of 4.4. Although the amino acid incorporation and fully using the reaction handles
obtained ADC lacked a linker, it retained binding affinity to incorporated for conjugation. Bioorthogonal conjugation of

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

A O

+ N O
SH Cys S
alkylation
O N

O
DAR distribution: 2, 4, 6, and 8
(pedominant at 2 using THIOMAB)

O
Br
B O N O
N
Br
S S S S O S S
Reduction of H H Cys
interchain disulfide rebridging
DAR: predominant at 4
Protein & Cell

C
PR3
+ X Pd
SH Cys-aryl S
coupling
X = Cl, Br, I, OTf Average DAR: 4.4

Figure 4. Cysteine coupling. (A) Maleimide alkylation. A maleimide moiety reacts with a reduced cysteine residue of a mAb
(distribution of DAR: 2, 4, 6, and 8 or predominant at 2 with THIOMAB technology). (B) Rebridging of interchain disulfide bonds. The
dibromo (or disulfonate) reagent reacts with the reduced interchain disulfides to provide rebridged mAbs (DAR: predominant at 4).
(C) Cysteine arylation using palladium complexes. Aryl-palladium complex reagents undergo aryl-thiol coupling, which affords mAbs
containing arylcysteines (average DAR: 4.4).

azide-incorporated antibodies can be achieved by using Transpeptidation using sortase

strained cyclooctyne-functionalized linkers that do not
Sortase A from Staphylococcus aureus recognizes the
require a cytotoxic, oxidative copper catalyst (Fig. 5C).
LPXTG (X: any amino acid) motif, cleaves the threonine-
However, the non-natural amino acid-based methodology
glycine (T-G) bond, and attaches an oligoglycine (oligo-G)-
generally requires special techniques and biological agents
containing molecule. Various cargo can be fused to the oligo-
for the genetic engineering process, and the incorporated
G for sortase A-mediated conjugation: peptides, proteins,
non-natural amino acid residues could potentially invoke
nucleic acids, and so on (Popp et al., 2009; Witte et al.,
undesired immunological response. Further efforts to solve
2012). For example, Ploegh and co-workers demonstrated
such issues will make this method truly practical and ver-
stoichiometric and site-specific conjugation of a biotinylated
satile in industrial production of ADCs.
class I MHC-restricted epitope to an antibody against the
C-type lectin DEC205 containing a LPETG-His6 sequence at
Enzymatic conjugation its C-terminus of the heavy chain (Swee et al., 2013). The
resulting conjugate retained epitope generation ability upon
Several enzymes have been used for conjugating the native
binding to dendritic cells and enabled monitoring of intra-
or genetically engineered antibody with the payload or for
cellular processes in vitro and in vivo. Beerli and co-workers
installing unique reaction handles on the antibody scaffold
demonstrated the potential of this powerful approach for
for the following chemical conjugation. These enzymes
stoichiometric site-specific ADC conjugation (Beerli et al.,
modify the antibody in a site- or amino acid sequence-
2015) (Fig. 6A). They introduced the recognition motif
specific manner. Furthermore, the reaction sites in native
LPETG to the C-termini of the light and heavy chains of
mAbs or handles that are genetically introduced are
various mAbs. Then, the small molecule payload
designed to specifically react with counterpart functional
monomethyl auristatin E (MMAE) containing penta-G was
groups. Thus, (chemo)enzymatic approaches generally
conjugated to the mAbs in the presence of sortase A. The
allow for site-specific conjugation leading to tightly controlled
resulting conjugates (DAR: approximately 3.2, monomer
DARs.

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

A O O
O p-acetyl-Phe O N N

H 2N
+ O
Oxime
formation

B
Copper N N
catalyst N N
N3 N3 + N N
Click
chemistry
p-azidomethyl-Phe
or
N6-((2-azidoethoxy)carbonyl)-Lys

C N N N N

Protein & Cell

H Copper- N N
free
N3 N3 + H H
Click
H chemistry
H H

Figure 5. Non-natural amino acid incorporation by genetic engineering into mAbs and subsequent chemical conjugation.
(A) Oxime ligation. (B) Copper-catalyzed or (C) strain-promoted (copper-free) azide-alkyne cyclization. The site-specific conjugation
method gives a defined DAR value depending on the number of non-natural amino acid residues that are genetically incorporated.

content: >96%) showed no adverse effect on antibody is genetically incorporated, resulting in site-specific antibody-
binding to the counterpart antigens. Further, these conju- drug conjugation. Another advantage of this LLQG-specific
gates exerted in-vitro cell killing activities comparable to the bacterial transglutaminase is that conjugation sites can be
corresponding conjugates generated by traditional ADC flexibly laid by inserting this short peptide motif within the
conjugation methods, including the FDA-approved ADCs antibody structure. They demonstrated the potential of this
Adcetris® and Kadcyla®. This method can also be used for strategy by preparing two ADCs that showed tightly con-
site-specific conjugation of the single-chain variable frag- trolled DARs (∼1.9) and comparable cytotoxicity and tolera-
ment (scFv) derived from mAbs (Madej et al., 2012). bility profiles.

Transpeptidation using microbial transglutaminase N-Glycan engineering

The use of bacterial transglutaminases is a powerful Asn297 (N297) within the Fc domain and the N-glycan on
approach for site-specific incorporation of the payload into this residue are conserved in all IgG classes, making these
the antibody (Fig. 6B). A transglutaminase derived from components attractive reaction sites for broadly applicable
Streptomyces mobaraensis catalyzes transpeptidation ADC conjugation. Zhou and co-workers developed incorpo-
where a primary amine-containing linker is covalently ration of an aldehyde group on the N-glycan terminus using
attached to the primary amide side chain of a specific glu- β-1,4-galactosyltransferase (GalT) and α-2,6-sialyltrans-
tamine (Q295) within deglycosylated antibodies, resulting in ferase (SialT) (Fig. 6C)(Zhou et al., 2014). These two
ADCs with a defined DAR of 2 (one conjugation site per enzymes introduce a sialic acid on each N-glycan terminus,
heavy chain) (Jeger et al., 2010; Dennler et al., 2014). An which is subsequently converted into an aldehyde group
N297Q mutation prior to this conjugation provides two more using NaIO4 under mild oxidation conditions. The aldehyde
reaction sites (DAR = 4). This method is quite advantageous groups generated are then used to conjugate aminooxy-
in terms of practical ADC production as the glycosidase and functionalized payloads. In their study, this conjugation
transglutaminase directly modify and conjugate native mAbs method gave an average DAR of 1.6, which was approxi-
with the payload, without the need for genetic engineering. mately the same number of sialic acid residues introduced
Strop and co-workers developed an alternative version using per antibody. Unfortunately, the oxidation step using NaIO4
a peptide sequence-specific transglutaminase (Strop et al., can oxidize methionine residues within the antibody, and
2013). This enzyme recognizes and utilizes LLQG motif that DAR distribution is wide due to low conversion. Another

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

A
Sortase
+ GGGGG

GTEPL LPETG G 5TEPL LPETG 5

DAR: 3.2

B Microbial
Q295 transglutaminase
H 2N NH 2 + H 2N HN NH

O O O O
deglycosylated mAb DAR: 1.8-2

C Gal Sial
GalT, SialT

O
N
Protein & Cell

NaIO4 H
H 2N H
O
N Average DAR: ~1.6
O
D
Endo-
glycosidase GalNAz
then N3 N3
GalT (Y289L)
N3 UDP

N N
N N
Strain-promoted N N
copper-free
click chemistry
DAR: 2

Figure 6. Site-specific (chemo)enzymatic conjugation. (A) Sortase-mediated conjugation. Sortase attaches oligoglycine-
functionalized linkers to LPETG tags on the mAb. (B) Microbial transglutaminase-mediated conjugation. The enzyme attaches an
ADC linker possessing a primary amine to Q295 of the heavy chain (DAR: 1.8–2, high homogeneity). (C) Conjugation using β-1,4-
galactosyltransferase (GalT) and α-2,6-sialyltransferase (SialT) (light green square: β-1,4-galactose, magenta circle: sialic acid). The
aldehyde groups installed react with alkoxyamine-functionalized linkers (average DAR: ∼1.6). (D) GlycoConnect technology using
endoglycosidase, galactosyltransferase, and N-azidoacetylgalactosamine (GalNAz, light blue square). The azide groups installed
react with strained cyclooctyne-functionalized linkers (DAR: 2, high homogeneity).

approach is to incorporate non-natural saccharides pos- this technology is that it gives consistent results regardless
sessing orthogonal reaction handles into the antibody. One of the heterogeneity of the N-Glycan forms, meaning that it
of the latest technologies based on this strategy is the Gly- can be used for any IgG isotypes with various N-glycosyla-
coConnect technology developed by van Delft and co- tion profiles.
workers (van Geel et al., 2015) (Fig. 6D). The glycan chain at
Asn297 was trimmed using the endoglycosidase Endo S2
Cleavable linkers
and then azide groups were introduced using a mutant
galactosyl transferase GalT(Y289L) and N-azidoacetyl- A major class of ADC linkers is the cleavable linker (Fig. 7).
galactosamine (GalNAz). The azide handles were used for a Cleavable linkers are designed to be cleaved by responding
strain-promoted click reaction with payloads, resulting in to an environmental difference between the extracellular and
stable and homogeneous ADCs with tightly controlled DARs intracellular environments (pH, redox potential, etc.) or by
(predominant at 2 in most cases). The biggest advantage of specific lysosomal enzymes. In most cases, the linkers in

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

A Hydrazone B Cathepsin B-cleavable CathepsinB

(lysosomal enzyme) O

H O O O O N
N N H H
Acidic N
N N
O pH H H
O
O
Val-Cit H 2N
N NH 2
H

C Disulfide D Pyrophosphate diester

O O
HS HO
S Intracellular P P Sequential
S reduction O O O O lysosomal
HO HO processing

Figure 7. Cleavable linkers. (A) Hydrazone linker. This linker is cleaved in the acidic environment (i.e., endosome and lysosome).

Protein & Cell

(B) Cathepsin B-cleavable peptide linker such as valine-citrulline-p-aminobenzyloxycarbonyl (PABC) and valine-alanine-PABC. The
PABC moiety enables release of free payload molecules in a traceless manner. (C) Disulfide-containing linker. The disulfide bond is
reduced by intracellular reducing molecules (e.g., glutathione) to release the payload. (D) Pyrophosphate diester. This stable,
hydrophilic linker is cleaved in lysosomes and free payload molecules are released.

this class are designed to release parental payload mole- Cathepsin B-responsive linker
cules after bond cleavage. Such traceless drug release
Cathepsin B is a lysosomal protease that is over-expressed
mechanisms allow researchers to estimate cytotoxic potency
in various cancer cells and involved in numerous oncogenic
of the conjugated payload based on known pharmacological
processes in humans (Gondi and Rao, 2013). Cathepsin B
parameters of the free payload.
has a relatively broad scope of substrate, but it preferentially
recognizes certain sequences such as phenylalanine-lysine
Hydrazone linker (Phe-Lys) and valine-citrulline (Val-Cit) and cleaves a pep-
tide bond on the C-terminal side of such sequences. In
Hydrazone, an acid-labile group, is used as a cleavable
particular, Val-Cit and Val-Ala linkers coupled with p-
linker that releases free drug through hydrolysis once an
aminobenzyloxycarbonyl (Val-Cit-PABC and Val-Ala-PABC)
ADC is transported to acidic endosomes (pH 5.0–6.0) and
are the most successful cleavable linkers for ADCs
lysosomes (pH about 4.8) (Fig. 7A). The chimeric antibody
(Dubowchik et al., 2002; Hartley, 2011) (Fig. 7B). Upon
BR96-doxorubicin conjugate (BR96-DOX) was developed
internalization through endocytosis and transportation to
with the hydrazone conjugation strategy. BR96-DOX was
lysosomes, cathepsin B selectively cleaves this linker and
advanced to a Phase II human clinical trial in metastatic
cytotoxic payloads are released from the ADC in a traceless
breast cancer (Tolcher et al., 1999). The toxicity profile of the
manner. The PABC moiety functions as a spacer between
conjugate was considerably improved compared to free
Val-Cit moiety and the payload, allowing cathepsin B to
doxorubicin administration. However, gastrointestinal toxicity
exhibit its full protease activity to the linker connected to a
was still prominent and clinical outcomes were not satisfying
bulky payload molecule such as doxorubicin (Dubowchik
due to its low tolerability. Another example is the anti-CD33
et al., 2002). This linker was used to construct the chimeric
antibody calicheamicin conjugate, Mylotarg® (Linenberger
anti-CD30 antibody-MMAE conjugate, or brentuximab
et al., 2001; Sievers et al., 2001). Mylotarg® showed
vedotin (Adcetris®) (Younes et al., 2010).
encouraging clinical results and was approved in 2000.
However, as mentioned earlier, it was withdrawn from the
Disulfide linker
market in 2010 due to a lack of clinically significant
improvement of patient outcome. Both unsuccessful ADCs Glutathione sensitive linker is another common cleavable
suffered from toxicities and low tolerability, which seems to linker (Fig. 7C). This strategy relies on the higher concen-
be attributed to lability of the hydrazone linker during circu- tration of reducing molecules such as glutathione in the
lation. Indeed, ADCs with the hydrazone linker undergo slow cytoplasm (1–10 mmol/L) (Wu et al., 2004) compared to the
hydrolysis under physiological conditions (pH 7.4, 37°C), extracellular environment (about 5 µmol/L in blood) (Mills
resulting in a slow release of the toxic payload (Laguzza and Lang, 1996). A disulfide bond is embedded within the
et al., 1989). linker and resists reductive cleavage in circulation. However,

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

S
S O
O
NH N
NH N +
H 3N O
O Lysosomal
processing -
O
O O3C
Lys-payload

O
O S
S +
H 3N
N
N Lysosomal CO 2-
processing
O
O Cys-payload

Figure 8. Non-cleavable linker. The chemical stability of the non-cleavable linker withstands proteolytic degradation. Cytosolic/
lysosomal degradation of the mAb moiety liberates the payload molecule linked to an amino acid residue derived from the degraded
mAb.
Protein & Cell

upon internalization, abundant intracellular glutathione demonstrate the potential of the pyrophosphate diester linker
reductively cleaves the disulfide bond to release the free for the future development of therapeutically practical ADCs.
payload molecule. To further enhance stability during circu-
lation, methyl groups are often installed next to the disulfide
Non-cleavable Linkers
bond (Saito et al., 2003). This class of linker has been
employed in Mylotarg® (Sievers et al., 2001), and more Non-cleavable linkers consist of stable bonds that are
recently, in several maytansine-based candidates in clinical resistant to proteolytic degradation, ensuring greater stability
trials (Widdison et al., 2015). than that of cleavable linkers. Non-cleavable linkers rely on
complete degradation of the antibody component of ADC by
cytosolic and lysosomal proteases, which eventually liber-
Pyrophosphate diester linker
ates a payload molecule linked to an amino acid residue
Recently, Garbaccio and co-workers developed a novel derived from the degraded antibody (Fig. 8). As such, when
cleavable linker with a pyrophosphate diester structure coupled with a non-cleavable linker, the payload structure
(Fig. 7D) (Kern et al., 2016). This anionic linker has greater must be carefully selected and designed so that payload can
aqueous solubility than traditional linkers and excellent cir- exert comparable or even better anti-tumor potency in such a
culatory stability. Furthermore, upon internalization, the modified form. For that purpose, it may be necessary to
pyrophosphate diester gets promptly cleaved through the examine PK/PD and toxicity profiles of all possible metabo-
endosomal-lysosomal pathway to liberate unmodified pay- lites of ADCs with non-cleavable linkers. A successful
load molecules. The authors speculate that the pyrophos- example of ADCs using a non-cleavable linker is the
phate diester goes through a two-step enzymatic linker humanized anti-HER2 antibody-maytansine conjugate tras-
cleavage that releases a payload-monophosphate molecule tuzumab emtansine (T-DM1, or Kadcyla®) (LoRusso et al.,
and then a free payload, although the enzyme(s) involved in 2011; Verma et al., 2012).
this process have not yet been identified. With this encour-
aging result, they set out to construct conjugates of the anti-
CONCLUDING REMARKS AND OUTLOOK
human CD70 antibody and various glucocorticoids using this
linker. The ADCs constructed showed great stability in In this article, we have reviewed the concept and clinical
human plasma (intact in vitro up to 7 days) and fast linker potential of ADCs and various conjugation/linker strategies
cleavage and release of free payload molecules in lyso- for constructing this new class of molecules (Table 1).
somes. Interestingly, each conjugate released a free payload Compared to traditional small molecule-based chemother-
at different rates, depending on the substituent group prox- apy, well-designed ADCs have several distinct features and
imal to the pyrophosphate moiety. This result suggests that clinical advantages, including preferable PK/PD and biodis-
the rate of release could be fine-tuned by further structural tribution (which are generally similar to that of native IgGs),
modifications. In addition, one of the ADCs containing fluti- broader therapeutic window, and flexibility of molecular
casone propionate exerted remarkable potency (EC50: 0.37 customization. As exemplified in the successes of the FDA
nmol/L) in CD70-positive 786-O cells, comparable to free approved Adcetris® and Kadcyla®, this new therapeutic
fluticasone propionate (EC50: 0.25 nmol/L). These results modality has huge potential for anti-cancer therapy and has

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

Table 1. Advantages and disadvantages of the conjugation and linker chemistries described

Strategy DARa Advantages Disadvantages

Chemical Lysine coupling 0–7 Simple process Distributed DAR
conjugation Used in FDA-approved Heterogeneous mixtures of products
and clinically tested Potential reduction of antigen binding
ADCs
Cysteine coupling 0, 2, Simple process Heterogeneous mixtures of products
4, Used in FDA-approved Increased clearance rate with high DAR
6, 8 and clinically tested
ADCs
THIOMAB 2 Defined DAR Requires genetic engineering
Homogeneity
Cysteine rebridging 4 Defined DAR Potential disulfide scrambling
Homogeneity
High structural stability
Non-natural amino 2 Defined DAR Requires special techniques and biological
acid Homogeneity agents

Protein & Cell

Potential immunogenicity
Sortase 3–4 Tightly-controlled DAR Requires incorporation of LPETG motif on
No adverse effect on the heavy chain
antibody binding
(Chemo) Microbial 2 Defined DARs Requires removal of N-glycan on N297
enzymatic transglutaminase Homogeneity
conjugation
Glycan engineering 2 Defined DARs Requires multiple steps (i.e., N-glycan
(GlycoConnect) Homogeneity trimming, glycosylation, and conjugation)
Hydrazone pH-responsive cleavage Premature cleavage during circulation
Val-Citb-PABCc, Stability during circulation Hydrophobicity
Val-Ala-PABCc Traceless release of
payload
Cleavable Disulfide Intracellular release of Potential premature cleavage during
Linker payload circulation
Pyrophosphate Stability during circulation Unknown mechanism of lysosomal
diester Hydrophilicity cleavage
Traceless release of
payload
Non-cleavable Stable linker without Stability during circulation An amino acid residue attached on the
Linker cleavage released payload
mechanism
a b c
DAR, Drug-to-antibody ratio; Cit, citrulline; PABC, p-aminobenzyloxycarbonyl.

attracted a great deal of attention from researchers and optimization for clinical application and eventual failure in
clinicians. Indeed, significant advances have been made in clinical trials. To overcome these problems, current efforts
ADC technologies, with about 60 ADCs currently in clinical are directed toward developing novel stable linkers (with or
trials. This emerging molecular platform is expected to without a payload release mechanism) and site-specific
become mainstream in anti-cancer therapeutics in the near conjugation methods enabling construction of homogeneous
future. Despite its potential, further understanding biochem- ADCs. Further investigations along this line will provide
ical, immunological, pharmacological, and molecular aspects greater insights and sophisticated strategies from medicinal
of ADCs must be pursued to better design and develop chemistry and pharmacology standpoints, leading to inno-
effective ADCs. While choice of target antigens and pay- vative cancer therapeutics in the future.
loads is important, antibody-payload conjugation methods
and linker chemistry are also crucial elements for producing ACKNOWLEDGEMENTS
successful ADCs. In particular, instability of the linker and
We gratefully acknowledge Dr. Yasuaki Anami and Dr. Yasuhiro
heterogeneity of the product (i.e., broad distribution of DARs)
Shimamoto for insightful discussions and proofreading the article.
often negatively impacts ADC efficacy and therapeutic win- We thank Dr. Georgina T. Salazar for editing the article. This work
dow, which often leads to difficulty or limitation in the

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

was supported in part by the Welch Foundation Grant AU00024 and Bryant P, Pabst M, Badescu G, Bird M, McDowell W, Jamieson E,
the CPRIT grant RP150551. Swierkosz J, Jurlewicz K, Tommasi R, Henseleit K et al (2015)
In vitro and in vivo evaluation of cysteine rebridged trastuzumab–
ABBREVIATIONS MMAE antibody drug conjugates with defined drug-to-antibody
ratios. Mol Pharm 12:1872–1879
ADC, antibody-drug conjugate; CD, cluster of differentiation; DAR,
Bryden F, Maruani A, Savoie H, Chudasama V, Smith MEB, Caddick
drug-antibody ratio; DOX, doxorubicin; Fc, fragment crystallizable;
S, Boyle RW (2014) Regioselective and stoichiometrically con-
FDA, U.S. Food and Drug Administration; Gal, β-1,4-galactose;
trolled conjugation of photodynamic sensitizers to a HER2
GalNAz, N-azidoacetylgalactosamine; GalT, β-1,4-galactosyltrans-
targeting antibody fragment. Bioconjugate Chem 25:611–617
ferase; HER2, human epidermal growth factor receptor 2; hmAb,
Chari RVJ (2008) Targeted cancer therapy: conferring specificity to
humanized/human monoclonal antibody; IgG, immunoglobulin G;
cytotoxic drugs. Acc Chem Res 41:98–107
MMAE, monomethylauristatin E; MMAF, monomethylauristatin F;
Chari RVJ, Miller ML, Widdison WC (2014) Antibody-drug conju-
PABC, para-aminobenzyloxycarbonyl; PBD, pyrrolobenzodiazepine;
gates: an emerging concept in cancer therapy. Angew Chem Int
PEG, polyethylene glycol; PK/PD, pharmacokinetics/pharmacody-
Ed Engl 53:3796–3827
namics; scFv, single-chain variable fragment; Sial, sialic acid; SialT,
Chudasama V, Maruani A, Caddick S (2016) Recent advances in the
α-2,6-sialyltransferase.
construction of antibody-drug conjugates. Nat Chem 8:114–119
Damelin M, Zhong W, Myers J, Sapra P (2015) Evolving strategies
for target selection for antibody-drug conjugates. Pharm Res
Protein & Cell

COMPLIANCE WITH ETHICS GUIDELINES

32:3494–3507
Kyoji Tsuchikama and Zhiqiang An declare that they have no conflict Dennler P, Chiotellis A, Fischer E, Brégeon D, Belmant C, Gauthier
of interest. L, Lhospice F, Romagne F, Schibli R (2014) Transglutaminase-
This article does not contain any studies with human or animal based chemo-enzymatic conjugation approach yields homoge-
subjects performed by the any of the authors. neous antibody-drug conjugates. Bioconjugate Chem 25:569–
578
DeVita VT, Chu E (2008) A history of cancer chemotherapy. Cancer
Res 68:8643–8653
Diamantis N, Banerji U (2016) Antibody-drug conjugates-an emerg-
OPEN ACCESS
ing class of cancer treatment. Br J Cancer 114:362–367
This article is distributed under the terms of the Creative Commons Drake PM, Rabuka D (2015) An emerging playbook for antibody-
Attribution 4.0 International License (http://creativecommons.org/ drug conjugates: lessons from the laboratory and clinic suggest a
licenses/by/4.0/), which permits unrestricted use, distribution, and strategy for improving efficacy and safety. Curr Opin Chem Biol
reproduction in any medium, provided you give appropriate credit to 28:174–180
the original author(s) and the source, provide a link to the Creative Dubowchik GM, Firestone RA, Padilla L, Willner D, Hofstead SJ,
Commons license, and indicate if changes were made. Mosure K, Knipe JO, Lasch SJ, Trail PA (2002) Cathepsin
B-labile dipeptide linkers for lysosomal release of doxorubicin
from internalizing immunoconjugates: model studies of enzymatic
REFERENCES drug release and antigen-specific in vitro anticancer activity.
Bioconjugate Chem 13:855–869
An Z (2010) Monoclonal antibodies—a proven and rapidly expand-
Ehrlich P (1913) Address in pathology, ON CHEMIOTHERAPY:
ing therapeutic modality for human diseases. Protein Cell 1:319–
delivered before the seventeenth International Congress of
330
Medicine. Br Med J 2:353–359
Axup JY, Bajjuri KM, Ritland M, Hutchins BM, Kim CH, Kazane SA,
Finbloom DS, Abeles D, Rifai A, Plotz PH (1980) The specificity of
Halder R, Forsyth JS, Santidrian AF, Stafin K et al (2012)
uptake of model immune complexes and other protein aggre-
Synthesis of site-specific antibody-drug conjugates using unnat-
gates by the murine reticuloendothelial system. J Immunol
ural amino acids. Proc Natl Acad Sci USA 109:16101–16106
125:1060–1065
Beerli RR, Hell T, Merkel AS, Grawunder U (2015) Sortase enzyme-
Ford CH, Newman CE, Johnson JR, Woodhouse CS, Reeder TA,
mediated generation of site-specifically conjugated antibody drug
Rowland GF, Simmonds RG (1983) Localisation and toxicity
conjugates with high in vitro and in vivo potency. PLoS ONE 10:
study of a vindesine-anti-CEA conjugate in patients with
e0131177
advanced cancer. Br J Cancer 47:35–42
Behrens CR, Ha EH, Chinn LL, Bowers S, Probst G, Fitch-Bruhns M,
Gondi CS, Rao JS (2013) Cathepsin B as a cancer target. Expert
Monteon J, Valdiosera A, Bermudez A, Liao-Chan S et al (2015)
Opin Ther Targets 17:281–291
Antibody-drug conjugates (ADCs) derived from interchain cys-
Hartley JA (2011) The development of pyrrolobenzodiazepines as
teine cross-linking demonstrate improved homogeneity and other
antitumour agents. Expert Opin Investig Drugs 20:733–744
pharmacological properties over conventional heterogeneous
Jain N, Smith SW, Ghone S, Tomczuk B (2015) Current ADC linker
ADCs. Mol Pharm 12:3986–3998
chemistry. Pharm Res 32:3526–3540
Bouchard H, Viskov C, Garcia-Echeverria C (2014) Antibody–drug
Jeger S, Zimmermann K, Blanc A, Grünberg J, Honer M, Hunziker P,
conjugates—a new wave of cancer drugs. Bioorg Med Chem Lett
Struthers H, Schibli R (2010) Site-specific and stoichiometric
24:5357–5363

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
Recent ADC conjugation and linker chemistries REVIEW

modification of antibodies by bacterial transglutaminase. Angew globulins from hamsters inoculated with such leukemia by
Chem Int Ed Engl 49:9995–9997 heterografts. C R Hebd Seances Acad Sci 246:1626–1628
Junutula JR, Raab H, Clark S, Bhakta S, Leipold DD, Weir S, Chen McCombs JR, Owen SC (2015) Antibody drug conjugates: design
Y, Simpson M, Tsai SP, Dennis MS et al (2008) Site-specific and selection of linker, payload and conjugation chemistry. AAPS
conjugation of a cytotoxic drug to an antibody improves the J 17:339–351
therapeutic index. Nat Biotechnol 26:925–932 Mills BJ, Lang CA (1996) Differential distribution of free and bound
Kern JC, Cancilla M, Dooney D, Kwasnjuk K, Zhang R, Beaumont glutathione and cyst(e)ine in human blood. Biochem Pharmacol
M, Figueroa I, Hsieh S, Liang L, Tomazela D et al (2016) 52:401–406
Discovery of pyrophosphate diesters as tunable, soluble, and Mullard A (2013) Maturing antibody-drug conjugate pipeline hits 30.
bioorthogonal linkers for site-specific antibody-drug conjugates. Nat Rev Drug Discov 12:329–332
J Am Chem Soc 138:1430–1445 Perez HL, Cardarelli PM, Deshpande S, Gangwar S (2014)
King HD, Dubowchik GM, Mastalerz H, Willner D, Hofstead SJ, Antibody–drug conjugates: current status and future directions.
Firestone RA, Lasch SJ, Trail PA (2002) Monoclonal antibody Drug Discov Today 19:869–881
conjugates of doxorubicin prepared with branched peptide Polson AG, Ho WY, Ramakrishnan V (2011) Investigational anti-
linkers: inhibition of aggregation by methoxytriethyleneglycol body-drug conjugates for hematological malignancies. Expert
chains. J Med Chem 45:4336–4343 Opin Investig Drugs 20:75–85
Kung Sutherland MS, Walter RB, Jeffrey SC, Burke PJ, Yu C, Popp MWL, Antos JM, Ploegh HL (2009) Site-specific protein
Kostner H, Stone I, Ryan MC, Sussman D, Lyon RP et al (2013) labeling via sortase-mediated transpeptidation. Curr Protoc

Protein & Cell

SGN-CD33A: a novel CD33-targeting antibody-drug conjugate Protein Sci. doi:10.1002/0471140864.ps1503s56
using a pyrrolobenzodiazepine dimer is active in models of drug- Postow MA, Callahan MK, Wolchok JD (2015) Immune checkpoint
resistant AML. Blood 122:1455–1463 blockade in cancer therapy. J Clin Oncol 33:1974–1982
Laguzza BC, Nichols CL, Briggs SL, Cullinan GJ, Johnson DA, Rudnick SI, Lou J, Shaller CC, Tang Y, Klein-Szanto AJP, Weiner
Starling JJ, Baker AL, Bumol TF, Corvalan JR (1989) New LM, Marks JD, Adams GP (2011) Influence of affinity and antigen
antitumor monoclonal antibody-vinca conjugates LY203725 and internalization on the uptake and penetration of Anti-HER2
related compounds: design, preparation, and representative antibodies in solid tumors. Cancer Res 71:2250–2259
in vivo activity. J Med Chem 32:548–555 Saito G, Swanson JA, Lee K-D (2003) Drug delivery strategy utilizing
Lazar AC, Wang L, Blättler WA, Amphlett G, Lambert JM, Zhang W conjugation via reversible disulfide linkages: role and site of
(2005) Analysis of the composition of immunoconjugates using cellular reducing activities. Adv Drug Deliv Rev 55:199–215
size-exclusion chromatography coupled to mass spectrometry. Sedlacek HH, Seemann G, Hoffmann D, Czech J (1992) Antibodies
Rapid Commun Mass Spectrom 19:1806–1814 as carriers of cytotoxicity. Contrib Oncol 43:1–145
Lencer WI, Blumberg RS (2005) A passionate kiss, then run: Sievers EL, Larson RA, Stadtmauer EA, Estey E, Löwenberg B,
exocytosis and recycling of IgG by FcRn. Trends Cell Biol Dombret H, Karanes C, Theobald M, Bennett JM, Sherman ML
15:5–9 et al (2001) Efficacy and safety of gemtuzumab ozogamicin in
Linenberger ML, Hong T, Flowers D, Sievers EL, Gooley TA, Bennett patients with CD33-positive acute myeloid leukemia in first
JM, Berger MS, Leopold LH, Appelbaum FR, Bernstein ID (2001) relapse. J Clin Oncol 19:3244–3254
Multidrug-resistance phenotype and clinical responses to gem- Strop P, Liu S-H, Dorywalska M, Delaria K, Dushin RG, Tran T-T, Ho
tuzumab ozogamicin. Blood 98:988–994 W-H, Farias S, Casas MG, Abdiche Y et al (2013) Location
LoRusso PM, Weiss D, Guardino E, Girish S, Sliwkowski MX (2011) matters: site of conjugation modulates stability and pharmacoki-
Trastuzumab emtansine: a unique antibody-drug conjugate in netics of antibody drug conjugates. Chem Biol 20:161–167
development for human epidermal growth factor receptor 2-pos- Swee LK, Guimaraes CP, Sehrawat S, Spooner E, Barrasa MI,
itive cancer. Clin Cancer Res 17:6437–6447 Ploegh HL (2013) Sortase-mediated modification of αDEC205
Lyon RP, Bovee TD, Doronina SO, Burke PJ, Hunter JH, Neff- affords optimization of antigen presentation and immunization
LaFord HD, Jonas M, Anderson ME, Setter JR, Senter PD (2015) against a set of viral epitopes. Proc Natl Acad Sci USA
Reducing hydrophobicity of homogeneous antibody-drug conju- 110:1428–1433
gates improves pharmacokinetics and therapeutic index. Nat Teicher BA, Chari RVJ (2011) Antibody conjugate therapeutics:
Biotechnol 33:733–735 challenges and potential. Clin Cancer Res 17:6389–6397
Madej MP, Coia G, Williams CC, Caine JM, Pearce LA, Attwood R, ten Cate B, Bremer E, de Bruyn M, Bijma T, Samplonius D,
Bartone NA, Dolezal O, Nisbet RM, Nuttall SD et al (2012) Schwemmlein M, Huls G, Fey G, Helfrich W (2009) A novel AML-
Engineering of an anti-epidermal growth factor receptor antibody selective TRAIL fusion protein that is superior to Gemtuzumab
to single chain format and labeling by Sortase A-mediated protein Ozogamicin in terms of in vitro selectivity, activity and stability.
ligation. Biotechnol Bioeng 109:1461–1470 Leukemia 23:1389–1397
Maruani A, Smith MEB, Miranda E, Chester KA, Chudasama V, Tian F, Lu Y, Manibusan A, Sellers A, Tran H, Sun Y, Phuong T,
Caddick S (2015) A plug-and-play approach to antibody-based Barnett R, Hehli B, Song F et al (2014) A general approach to
therapeutics via a chemoselective dual click strategy. Nat site-specific antibody drug conjugates. Proc Natl Acad Sci. USA
Commun 6:6645 111:1766–1771
Mathe G, Loc TB, Bernard J (1958) Effect on mouse leukemia 1210 Tolcher AW, Sugarman S, Gelmon KA, Cohen R, Saleh M, Isaacs C,
of a combination by diazo-reaction of amethopterin and gamma- Young L, Healey D, Onetto N, Slichenmyer W (1999)

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn
 REVIEW Kyoji Tsuchikama, Zhiqiang An

Randomized phase II study of BR96-doxorubicin conjugate in Witte MD, Cragnolini JJ, Dougan SK, Yoder NC, Popp MW, Ploegh
patients with metastatic breast cancer. J Clin Oncol 17:478–484 HL (2012) Preparation of unnatural N-to-N and C-to-C protein
van Geel R, Wijdeven MA, Heesbeen R, Verkade JMM, Wasiel AA, fusions. Proc Natl Acad Sci USA 109:11993–11998
van Berkel SS, van Delft FL (2015) Chemoenzymatic conjugation Wu G, Fang Y-Z, Yang S, Lupton JR, Turner ND (2004) Glutathione
of toxic payloads to the globally conserved n-glycan of native metabolism and its implications for health. J Nutr 134:489–492
mAbs provides homogeneous and highly efficacious antibody- Younes A, Bartlett NL, Leonard JP, Kennedy DA, Lynch CM, Sievers
drug conjugates. Bioconjugate Chem 26:2233–2242 EL, Forero-Torres A (2010) Brentuximab vedotin (SGN-35) for
VanBrunt MP, Shanebeck K, Caldwell Z, Johnson J, Thompson P, relapsed CD30-positive lymphomas. N Engl J Med 363:1812–
Martin T, Dong H, Li G, Xu H, D’Hooge F et al (2015) Genetically 1821
encoded azide containing amino acid in mammalian cells Zhao RY, Wilhelm SD, Audette C, Jones G, Leece BA, Lazar AC,
enables site-specific antibody-drug conjugates using click Goldmacher VS, Singh R, Kovtun Y, Widdison WC et al (2011)
cycloaddition chemistry. Bioconjugate Chem 26:2249–2260 Synthesis and evaluation of hydrophilic linkers for antibody-
Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, Pegram maytansinoid conjugates. J Med Chem 54:3606–3623
M, Oh D-Y, Diéras V, Guardino E et al (2012) Trastuzumab Zhou Q, Stefano JE, Manning C, Kyazike J, Chen B, Gianolio DA,
emtansine for HER2-positive advanced breast cancer. N Engl J Park A, Busch M, Bird J, Zheng X et al (2014) Site-specific
Med 367:1783–1791 antibody-drug conjugation through glycoengineering. Bioconju-
Vinogradova EV, Zhang C, Spokoyny AM, Pentelute BL, Buchwald gate Chem 25:510–520
SL (2015) Organometallic palladium reagents for cysteine Zimmerman ES, Heibeck TH, Gill A, Li X, Murray CJ, Madlansacay
Protein & Cell

bioconjugation. Nature 526:687–691 MR, Tran C, Uter NT, Yin G, Rivers PJ et al (2014) Production of
Widdison W, Wilhelm S, Veale K, Costoplus J, Jones G, Audette C, site-specific antibody-drug conjugates using optimized non-nat-
Leece B, Bartle L, Kovtun Y, Chari R (2015) Metabolites of ural amino acids in a cell-free expression system. Bioconjugate
antibody-maytansinoid conjugates: characteristics and in vitro Chem 25:351–361
potencies. Mol Pharm 12:1762–1773

© The Author(s) 2016. This article is published with open access at Springerlink.com and journal.hep.com.cn