A Reagent for Amine-Directed Conjugation to IgG1 Antibodies

Article abstract of DOI:10.1002/anie.202013911

Functionalized antibodies are an indispensable resource for diagnosis, therapy and as a research tool for chemical biology. However, simpler and better methodologies are often required to improve the labeling of antibodies in terms of selectivity and scal

Full text of DOI:10.1002/anie.202013911

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Reagent for Amine-Directed Conjugation to IgG1 Antibodies
Authors: Anders Märcher, Johan Palmfeldt, Marija Nisavic, and Kurt
Vesterager Gothelf
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202013911
Link to VoR: https://doi.org/10.1002/anie.202013911

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
A Reagent for Amine-Directed Conjugation to IgG1 Antibodies
Anders Märcher, Johan Palmfeldt, Marija Nisavic, Kurt V. Gothelf*
Abstract: Functionalized antibodies are an indispensable resource
for diagnosis, therapy and as a research tool for chemical biology.
However, simpler and better methodologies are often required to
produces more homogeneous conjugates, it is also more work
intensive and often gives rise to poor expression yields.[13]
Site-directed labelling is a strategy that overcomes some of
improve the labeling of antibodies in terms of selectivity and scalability. the obstacles mentioned above.[14,15] Here, a directing reagent
Herein, we report the development of an easily available chemical
reagent that allows site-directed labeling of native human IgG1
antibodies in good yield and mono-labeling selectivity. The
salicylaldehyde moiety of the reagent reacts with surface exposed
lysine residues to transiently form an iminium ion, and this positions a
semi-reactive ester into proximity of a second lysine residue that
reacts with the ester to form an amide. Interestingly, it appears that
the formation of the iminium ion also has a significant activating effect
of the ester. We use flow cytometry and bio-layer interferometry to
confirm that the labeled antibodies retain antigen binding.
forms a complex with a chosen protein of interest (POI). The
complex increases the local concentration of a reactive part of the
reagent enabling reaction with the POI. The size of the probe
restricts the labelling, such that it preferentially occurs in vicinity
of the complexation site. Hamachi and coworkers have pioneered
this method for protein labeling, primarily with focus on ligand
directed approaches. Over the last decade they have developed
several probes for site-directed labeling for
a variety of
proteins.[16–18] The high specificity of the ligand directed approach,
allows for protein labeling in vivo, such as visualization of live
neurons.[19]
Targeting more universal sites such as metal binding sites
or disulfides, allows for the utilization of the same reagent or
strategy on different proteins.[15] Recently, we, and Forte et al.,
published two related approaches for disulfide directed labeling of
proteins.[20,21] The metal chelator tris-nitrilotriacetic acid has been
used in different approaches for targeting metal-binding sites.[22,23]
In one case, a metal-targeting strategy was used to prepare
deoxyribonucleic acid (DNA) functionalized antibodies with a yield
of around 23%.[24]
In 2018 and 2020 Rai and co-workers reported on a new
method for lysine-directed labeling of histidine or lysine residues
respectively.[25,26] The probes entail a salicylaldehyde directing
group, a linker of different lengths and a reactive moiety for either
histidine or lysine. In both cases they report the modification of
one native immunoglobulin G1 (IgG1), in which both heavy chain
(HC) and light chain (LC) is labeled. In one case a Trastuzumab
FAB is modified with a conversion of 30-41%, no yields are given
for native IgGs.
Introduction
By conjugating synthetic molecules to proteins their functional
and therapeutic properties can be changed and optimized. This
inspires the development of methods for precise engineering of
native proteins. Antibody conjugates are especially interesting,
since functionalized antibodies have been used extensively for
diagnostics, in vivo imaging, therapy and as a tool for molecular
biology.[1–5] Conjugation to antibodies is, however, far from trivial.
Full size antibodies are large proteins, around 150 kDa, with more
than 80 lysine residues,[6] and they are dimeric structures that
consist of two HC and two LC. The two HCs, and LCs, have the
same amino acid sequence, respectively, meaning that multiple
different residues that are chemically identical are found on each
antibody. Conventionally, functionalized antibodies are prepared
by a random global lysine labelling with N-hydroxysuccinimide
(NHS) esters[7,8] or by cysteine conjugation via reduction of
solvent accessible disulfide bonds and reaction with
maleimides.[9] For both methods there is limited control of
attachment sites and stoichiometry, which means the methods
produce nonhomogeneous conjugates. It has been shown for
antibody drug conjugates (ADC) that heterogeneous conjugates
Results and Discussion
Inspired by the above described work we have designed a new,
smaller type of lysine directed labelling reagent (LDLR) 1a for
labeling of IgG1 antibodies (Figure 1). In addition to the proximity
directing effect the reagent is also activated by interaction with the
directing group. The LDLR reagent is envisioned for straight
forward labeling with low time demand and to produce antibody
conjugates with higher control of labeling amount than for
common NHS labeling strategies. We focus on IgG1 antibodies,
since it is by far the therapeutically most relevant subtype, both in
development and in the clinic.[27] The LDLR 1a is easily prepared
in three steps from readily available starting materials. It entails a
salicylaldehyde that can in a rapid and reversible reaction form an
iminium ion with surface exposed lysine residues.[26,28] It has been
reported that the reaction of salicylaldehyde with amines under
aqueous conditions leads to the preferred formation of the
iminium/phenolate (Figure 1b) rather than the imine/phenol
structure.[29] The main hypothesis of this work is that the p-phenol
ester is activated by formation of the iminium ion through
[*]
Anders Märcher, Prof. Dr. Kurt V. Gothelf
Department of Chemistry and Interdisciplinary Nanoscience Centre
(iNANO), Aarhus University
Gustav Wieds Vej 14, 8000 Aarhus (Denmark)
Email: kvg@inano.au.dk
Johan Palmfeldt
Department of Clinical Medicine, Aarhus University
Brendstrupgårdsvej 21A, 8200 Aarhus N (Denmark)
Marija Nisavic
Department of Chemistry and Department of Clinical Medicine,
Aarhus University
Brendstrupgårdsvej 21A, 8200 Aarhus N (Denmark)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
means poorer therapeutic indices and higher clearance rates.[10–
12]
Another conjugation method is the use of site-directed
mutagenesis for incorporation of cysteines. While this strategy
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
Figure 1. a) Synthetic route used to prepare the LDLR 1a. The LDLR is prepared in 27% yield in three synthetic steps, starting from mono azide functionalized
TEG. b) Illustration of the concept of the LDLR. A lysine residue on the surface of a protein of interest reacts in a rapid and reversible reaction to form an iminium
ion. A nucleophilic lysine residue in proximity to the formed iminium ion attacks the now labile intramolecular ester in the LDLR and is labeled with the small azide
tag. The tag can be further functionalized with a DBCO reagents of interest, we have used a 20 K Peg, DNA and a fluorophore.
stabilization of the negative charge of the leaving group, and
thereby increases the reactivity towards nucleophilic lysines in
proximity of the formed iminium ion. The salicylaldehyde iminium
ion leaving group will in time be hydrolyzed. We further imagined
that 1a would only react when the iminium ion is formed, as the
reactivity and concentration of the ester is too low for the reaction
to take place if the labeling is not facilitated by proximity and
thereby an increased local concentration.
reaction with a 20 K DBCO-PEG. The products were analyzed by
non-reducing and reducing SDS-PAGE (Supporting information,
Fig S2 and S3). The best conditions were determined to be 2.5
eq. LDLR 1a at pH 8.5. These conditions provided predominately
mono labeled HC and the highest degree of mono and double
labeled Cetuximab (approximately 39% and 15%, respectively).
The LDLR 1a is designed with an azide tag on the amide
forming part of the reagent. The azide tag is used for further
conjugation in strain-promoted alkyne azide cycloadditions
(SPAAC) with a functionalized dibenzocyclooctyne (DBCO).
Importantly numerous DBCO reagents with functionalities of
interest are commercially available.
With the LDLR 1a at hand we wanted to test and optimize
labeling conditions for IgG1s. Cetuximab, one of the first
therapeutic IgG1s, was chosen as the test substrate. In a first
attempt 10 µM Cetuximab was treated with 4 eq. of LDLR 1a
overnight (ON), at pH 8.5 and rt. Excess 1a was removed by
molecular weight cut off (MWCO) filtration and the azide-tag now
attached to the protein was reacted with 10 eq. 20 K DBCO-PEG
ON in a SPAAC. The product was then analyzed by reducing
SDS-PAGE (Figure 2a). Unless otherwise noted, the same
procedure of conjugation with 20 K DBCO-PEG and SDS-PAGE
gel analysis was used in the following experiments. We were
excited to see that labeling did occur using such low number of
equivalents. In other studies, mentioned above, where the same
guiding moiety is used, much harsher conditions were required
with up to 25 eq. of reagent and in some cases 20% DMSO.[25,26]
Furthermore, the labeling from the LDLR 1a occurred
predominantly once on the heavy chain. Diminutive amounts of
LC labeling and double labeled HC was also observed (Figure 2a).
Next the labeling conditions were optimized to achieve the
highest degree of double and mono labeled antibody, while
keeping the degree of higher labeled antibody constructs low.
Mono and double labeled antibody was targeted to produce
conjugates with high homogeneity.
Figure 2. a) Initial labeling of Cetuximab with 1a, analyzed with reducing SDS-
PAGE. Lane 1: Protein ladder, Lane 2: Native Cetuximab, Lane 3: Cetuximab
labeled with the 1a, reacted with 10 eq. 20 K PEG-DBCO. Labeling conditions:
Cetuximab (10 µM), pH 8.5, 100 mM NaCl, 4 eq. 1a, ON. b) Investigation of the
activation of the LDLR 1a analyzed with reducing SDS-PAGE. Lane 1: Protein
Ladder, Lane 2: Native Cetuximab, Lane 3: Cetuximab labeled with the 1a, Lane
4: Cetuximab labeled with 1b. c) Confirmation of site-directed labeling of 1a,
analyzed with reducing SDS-PAGE. Lane 1: Protein ladder, Lane 2: Cetuximab
labeled with the 1a, Lane 3: Cetuximab labeled with 1d, Lane 4: Cetuximab
labeled with 1c, Lane 5: Native Cetuximab. Labeling conditions: Cetuximab (10
µM), pH 8.5, 100 mM NaCl 2.5 eq. reagent 1a/1b/1c/1d, ON. d) Structure and
reactions of reagents 1a, b, c and d prepared to investigate the mechanism of
the LDLR labeling. Reagents are shown with their proposed reaction with a
protein except for 1d, which is not expected to react in any way. For full gel
image see Fig. S1, S4 and S5 in the Supporting information.
Labeling with the LDLR 1a was investigated at pH 7.5 and
8.5 as the labeling is dependent upon the protonation state of the
lysines. Different equivalents (1, 2.5, 5 or 10 eq.) of 1a were
incubated with the protein ON. Excess reagent 1a was removed
by MWCO filtration, and the conjugates were subjected to ON
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
attached are observed for the NHS labeling as well as a more
prominent light chain labeled band. As antibodies are dimeric
structures, HC with multiple labels will equal even greater
complexity of number of labels on the native protein. Other studies
have also shown that producing antibody conjugates with high
homogeneity and yield is very difficult with NHS-ester based
labeling methods.[32,33]
We next challenged the methodology by other IgG
antibodies. Four other IgG1 antibodies (Rituximab, Trastuzumab,
Belimumab and Pertuzumab), one IgG2 (Panitumumab) and one
IgG4 (Nivolumab) were chosen for further studies. The antibodies
were all subjected to the LDLR 1a using the optimized protocol
and were then analyzed by non-reducing SDS-PAGE (Fig 4a).
Reference non-reducing and reducing SDS-page of the native
antibodies are shown in the Supporting information (Fig. S10-
S11). As noticed all IgG1s and Nivolumab (IgG4) were
successfully labeled, whereas Panitumumab (IgG2) showed no
labeling. Still, the methodology allowed for similar labeling of five
IgG1s requiring no further optimization for each protein. The
optimized LDLR conditions labeled all IgG1 antibodies with similar
conversion for mono and double labeled conjugates (36-40% and
12-17% respectively). Also, it was found by reductive SDS-PAGE
that the label was almost exclusively attached to the HC of the
antibodies (Supporting information, Fig. S12).
The observation that no labeling occurs on Panitumumab is
interesting. We additionally attempted labeling IgG2 Kappa using
the LDLR, which resulted in trace labeling (Supporting information,
Fig. S13). The primary difference between IgG1 and IgG2
antibodies is in the hinge region. The IgG2 hinge region is shorter
and much more rigid than the IgG1 hinge region, consisting of
twelve amino acids residues with four disulfide linkages contrary
to the fifteen amino acids with two disulfide linkages in IgG1s.[34]
A lysine residue in the hinge region of Cetuximab was the first
modifications sites identified with MS (data not shown), which
might explain why the tested IgG2 antibodies labels poorly. The
IgG4 hinge region is much more comparable to the IgG1 in terms
of rigidity.
Encouraged by the promising results we wanted to
investigate if the LDLR 1a is indeed both directed and activated
by a neighboring amine. To explore this three new reference
reagents 1b, 1c and 1d were synthesized. In reagent 1b the
salicylaldehyde moiety is conserved, but the positions of the
aldehyde and phenol hydroxyl group are swapped (Figure 2b and
2d). As the aldehyde is no longer placed in the para position to
the intramolecular ester, this reagent should be much less
activated upon iminium formation. For this reagent no labeling of
Cetuximab is observed. Since the salicylaldehyde is conserved in
1b it is expected that the directing effect is conserved through
formation of the iminium ion. Therefore, the lack of reactivity must
be explained by a less reactive ester moiety, which may be
explained by the lack of activation upon iminium ion formation. But
it is also likely that the ester in reagent 1b is to some degree
deactivated by the iminium ion formation, since this positions the
electron donating phenolate in conjugation with the leaving group
of the ester. In reagent 1c the hydroxyl group is removed, which,
as described by Neri and coworkers,[26] shifts the equilibrium
towards the aldehyde. Furthermore, the species that is formed
with the first lysine is most likely the neutral imine, which is not
activated relative to the aldehyde. In agreement with this
assumption, no coupling to Cetuximab is observed for reagent 1c.
The same observation is made for reagent 1d which is prepared
with an ethyl ester instead of an aldehyde and thus there can be
no proximity guiding. It has been shown by computational studies
that the electron withdrawing capacity of an ethyl ester was within
the same range as an aldehyde,[30,31] and we rationalized that if
probe 1d did not label unguided, then neither would reagents 1a-
c. Based on these reference experiments we conclude that both
the directing and activating effect of the salicylaldehyde is of
importance for the successful labeling with reagent 1a.
Figure 4. a) Different IgGs labeled with the LDLR 1a and analyzed by non-
reducing SDS-PAGE. Labeling conditions: Antibody (10 µM), pH 8.5, 100 mM
NaCl, 2.5 eq. 1a. Full antibodies are seen in each lane. Band 1 is native antibody,
band 2 is antibody with one modification and band 3 is antibody with two
modifications. Lane 1: Protein ladder, Lane 2: 1a labeled Cetuximab, Lane 3:
1a labeled Trastuzumab, Lane 4: 1a labeled Rituximab, Lane 5: 1a labeled
Pertuzumab, Lane 6: 1a labeled Belimumab, Lane 7: Panitumumab (IgG2)
attempted labeled with 1a, Lane 8: 1a labeled Nivolumab (IgG4). Lane 1-5 are
IgG1s. Control with native unlabeled antibodies is found in Fig. S10 in the
Supporting information. For a full gel image see Fig. S7 in the Supporting
information. b) Quantification of band 1, 2 and 3 in each lane of. Each band is
the conversion of each antibody to that amount of labels. Band percentage is
an average of the shown gel and two more replicates found in the supporting
information Fig. S8-S9.
Figure 3. Labeling comparison of an NHS-ester and the LDLR 1a analyzed by
reducing SDS-PAGE. Lane 1: Protein ladder, Lane 2: Native Cetuximab, Lane
3: Cetuximab labeled with 1a, Lane 4: Cetuximab Labeled with Azide-NHS.
Labeling conditions: Cetuximab (10 µM), pH 8.5, 100 mM NaCl, 2.5 eq. reagent,
ON. For full gel image see Fig. S6 in the Supporting information.
Next the LDLR labeling was compared to the conventional
NHS labeling using the azide-NHS ester shown in Figure 3. Both
reagents were used to label Cetuximab, and the antibodies were
analyzed by reducing SDS-PAGE (Figure 3a). The overall
conversion of antibody was higher for the NHS-ester. However,
the LDLR 1a produces conjugates with considerable higher
homogeneity, as bands with more than three or more labels
All the LDLR 1a labeled antibodies, without the 20k PEG
DBCO modification, were then prepared for MS studies by in gel
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
Figure 5. Illustration of the labeling of IgG1 antibodies using the LDLR 1a. General optimized conditions are given, which provide 36-39% and 12-17% mono
labeled and double labeled antibody respectively. b) Flow cytometry data using Cy3-antibody conjugates prepared using the LDLR reagent. Control is in all cases
pretreatment of the cells with native unconjugated antibody. Cell types used: Cetuximab (A431, EGFR+), Pertuzumab and Trastuzumab (SKBR3, HER2+),
Rituximab (Ramos, CD20+).
trypsin digestion. Labeling sites were identified and major labeling
sites were located. In general 3-4 sites were found on each
antibody (Data shown in Supporting information, pages S14-16).
Multiple lysine residues in the same small peptide sequence were
found to have been modified across different IgG1 antibodies.
Two such examples are the hinge region lysine residue found in
the sequence CDKTH and a lysine residue found in the top of the
heavy chain in the sequence TKNQ. They were both found in
three of the antibodies. In general, most of the labeling sites
identified were found on at least two of the antibodies. Common
for most labeling sites is that they have another lysine residue in
close proximity. However, in some cases, no lysine residue can
be found in proximity in the sequence, but here it is likely that a
lysine residue is still closely positioned in the 3D structure of the
overexpress the target epitope for each antibody was then
selected. Each cell type was treated with the respective antibody-
Cy3 conjugate and analyzed by flow cytometry. As a control in all
cases a cell population was pretreated with native antibody to
block the epitopes. Control samples were used to ensure that
binding, to the cells, of the antibody-Cy3 conjugates remained
through the original epitope. In all cases the antibody-Cy3
conjugates retained binding, and binding could be minimized by
pre-blocking the expressed epitopes (Figure 5).
Bio-layer interferometry (BLI) was used to verify whether or
not Belimumab retained binding to its epitope B-cell activating
factor (BAFF). However, BLI requires clean conjugate without
residues of the native unmodified antibody. To achieve pure
antibody-LDLR conjugate,
a DNA-antibody construct was
protein. Analysis of crystal structures of Fab domains of Rituximab, prepared using the LDLR methodology. Belimumab was labeled
Cetuximab and Belimumab showed that these labelling sites
indeed have a nearby lysine residue in the 3D structure
(Supporting information Fig. S19-S21). We know from earlier
studies, using the same salicylaldehyde guiding moiety for protein
labeling, that labeling of the LC was also observed.[26] We were
however, pleased to see that for the LDLR 1a reagent most sites
were located on the HC (Data shown in Supporting information
pages S14-16).
After establishing the efficacy of the LDLR 1a for site-
directed labeling of IgG1s we wanted to investigate if the
antibodies retained affinity for their native antigen. Flow cytometry
was used to investigate this, for four of the LDLR-antibody
with LDLR 1a using optimized conditions. A DBCO-DNA strand
was then reacted with labeled Belimumab ON and the
Belimumab-DNA conjugate was purified by anion exchange
HPLC. Impressively, DNA-antibody conjugate prepared in 10
nmol scale could be purified in a 36 and 14% yield for the mono
and double labeled antibody respectively, when using this
strategy (Supporting information, Fig. S15-S16). Belimumab-DNA
(monolabeled) and native Belimumab was then used for BLI. The
size of the DNA labeled Belimumab has changed because of the
attached DNA, which will affect the readout of BLI, meaning that
the binding cannot be directly compared to unlabeled Belimumab.
However, it is clear that, as expected, LDLR 1a labeled
Belimumab retains affinity for BAFF (Supporting information, Fig.
S17-S18)
conjugates
(Cetuximab,
Rituximab,
Pertuzumab
and
Trastuzumab). For this purpose, Cy-3 conjugates were prepared
of these antibodies. The standard LDLR protocol was used for
each antibody, followed by reaction with Cy3-DBCO in a SPAAC
reaction overnight. Excess Cy3-DBCO is removed by molecular
weight cut off (MWCO) filtration, using an Amicon™ 30 K filter with
PBS. The Cy3-antibody conjugates were then analyzed by
reducing SDS-PAGE, to ensure that Cy3 was attached to the
antibodies (Supporting information, Fig S14). As expected,
predominantly HC labeling had occurred. A cell type that
Conclusion
A new reagent 1a was designed, which is not only directed, but
also activated by a neighboring lysine through iminium ion
formation for reaction with the first lysine. Three reference
reagents were prepared and their lack of reactivity support that
labeling with LDLR 1a is both directed and activated upon iminium
ion formation. The LDLR 1a is efficient for insertion of a reactive
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
Adv. 2017, 7, 24828–24832.
azide handle onto native human IgG1s. This handle was applied
for conjugation to DBCO functionalized cargos. This allowed for
fabrication of antibody PEG, fluorophore and DNA conjugates, but
it could in theory be used for most functionalities with a DBCO
handle. The methodology is simple and mild, since it only requires
the addition of one component in low stoichiometry at standard
conditions and requires no elaborate laboratory techniques. Also,
it produces conjugates with better homogeneity than the
commonly used NHS-ester. Generally, the conversions of mono
and double labeled antibodies are around 38% and 14%,
respectively. The methodology exhibited robustness as five
different IgG1s and one IgG4 antibody was labeled with similar
conversion. Reducing SDS-PAGE and MS studies confirm that
labeling occurs primarily on the HC. From MS studies it was also
found that generally 3-4 major labeling sites were found for each
antibody. All five labeled IgG1s were found to retain epitope
binding with either flow cytometry or BLI.
[14]
M. R. Mortensen, M. B. Skovsgaard, A. H. Okholm, C. Scavenius,
D. M. Dupont, C. B. Rosen, J. J. Enghild, J. Kjems, K. V. Gothelf,
Bioconjug. Chem. 2018, 29, 3016–3025.
M. R. Mortensen, M. B. Skovsgaard, K. V. Gothelf, ChemBioChem
2019, 20, 2711–2728.
[15]
[16]
[17]
[18]
[19]
Y. Koshi, E. Nakata, M. Miyagawa, S. Tsukiji, T. Ogawa, I.
Hamachi, J. Am. Chem. Soc. 2008, 130, 245–251.
T. Tamura, Z. Song, K. Amaike, S. Lee, S. Yin, S. Kiyonaka, I.
Hamachi, J. Am. Chem. Soc. 2017, 139, 14181–14191.
T. Tamura, T. Ueda, T. Goto, T. Tsukidate, Y. Shapira, Y.
Nishikawa, A. Fujisawa, I. Hamachi, Nat. Commun. 2018, 9, 1–12.
S. Wakayama, S. Kiyonaka, I. Arai, W. Kakegawa, S. Matsuda, K.
Ibata, Y. L. Nemoto, A. Kusumi, M. Yuzaki, I. Hamachi, Nat.
Commun. 2017, 8, DOI 10.1038/ncomms14850.
T. Nielsen, A. Märcher, Z. Drobňáková, M. Hučko, M. Štengl, V.
Balšánek, C. Wiberg, P. F. Nielsen, T. E. Nielsen, K. V. Gothelf, et
al., Org. Biomol. Chem. 2020, 18, 4717–4722.
N. Forte, I. Benni, K. Karu, V. Chudasama, J. R. Baker, Chem. Sci.
2019, 10, 10919–10924.
C. B. Rosen, A. L. B. Kodal, J. S. Nielsen, D. H. Schaffert, C.
Scavenius, A. H. Okholm, N. V. Voigt, J. J. Enghild, J. Kjems, T.
Tørring, et al., Nat. Chem. 2014, 6, 804–809.
S. H. Uchinomiya, H. Nonaka, S. H. Fujishima, S. Tsukiji, A. Ojida,
I. Hamachi, Chem. Commun. 2009, 5880–5882.
A. L. B. Kodal, C. B. Rosen, M. R. Mortensen, T. Tørring, K. V.
Gothelf, ChemBioChem 2016, 17, 1338–1342.
S. R. Adusumalli, D. G. Rawale, U. Singh, P. Tripathi, R. Paul, N.
Kalra, R. K. Mishra, S. Shukla, V. Rai, J. Am. Chem. Soc. 2018, 140,
15114–15123.
S. R. Adusumalli, D. G. Rawale, K. Thakur, L. Purushottam, N. C.
Reddy, N. Kalra, S. Shukla, V. Rai, Angew. Chemie - Int. Ed. 2020,
59, 10332–10336.
[20]
[21]
[22]
Acknowledgements
[23]
[24]
[25]
The work is funded by the Novo Nordisk foundation (CEMBID)
(Grant Number NNF17OC0028070).
Conflict of interest
[26]
The authors declare no conflict of interest.
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
V. Irani, A. J. Guy, D. Andrew, J. G. Beeson, P. A. Ramsland, J. S.
Richards, Mol. Immunol. 2015, 67, 171–182.
A. Dal Corso, M. Catalano, A. Schmid, J. Scheuermann, D. Neri,
Angew. Chemie - Int. Ed. 2018, 57, 17178–17182.
J. Crugeiras, A. Rios, E. Riveiros, J. P. Richard, J. Am. Chem. Soc.
2009, 131, 15815–15824.
G. S. Remya, C. H. Suresh, Phys. Chem. Chem. Phys. 2016, 18,
20615–20626.
M. Shahamirian, H. Szatylowicz, T. M. Krygowski, Struct. Chem.
2017, 28, 1563–1572.
L. Wang, G. Amphlett, W. A. Blättler, J. M. Lambert, W. Zhang,
Protein Sci. 2005, 14, 2436–2446.
J. Wiener, D. Kokotek, S. Rosowski, H. Lickert, M. Meier, Sci. Rep.
2020, 10, 1–11.
G. Vidarsson, G. Dekkers, T. Rispens, Front. Immunol. 2014, 5, 1–
17.
Keywords: IgG1 • antibody • Site-directed • labeling •
fluorophore
[1]
E. B. Ehlerding, H. J. Lee, D. Jiang, C. A. Ferreira, C. D. Zahm, P.
Huang, J. W. Engle, D. G. McNeel, W. Cai, Am. J. Cancer Res.
2019, 9, 53–63.
[2]
[3]
[4]
P. O. Krutzik, J. M. Irish, G. P. Nolan, O. D. Perez, Clin. Immunol.
2004, 110, 206–221.
R. Pal, H. Kang, H. S. Choi, A. T. N. Kumar, Clin. Cancer Res. 2019,
25, 6653–6661.
A. Younes, N. L. Bartlett, J. P. Leonard, D. A. Kennedy, C. M.
Lynch, E. L. Sievers, A. Forero-Torres, N. Engl. J. Med. 2010, 363,
1812–1821.
[5]
[6]
H. Andresen, C. Grötzinger, K. Zarse, O. J. Kreuzer, E.
Ehrentreich-Förster, F. F. Bier, Proteomics 2006, 6, 1376–1384.
D. Ayoub, W. Jabs, A. Resemann, W. Evers, C. Evans, L. Main, C.
Baessmann, E. Wagner-Rousset, D. Suckau, A. Beck, MAbs 2013,
5, 699–710.
[7]
[8]
[9]
A. A. Wakankar, M. B. Feeney, J. Rivera, Y. Chen, M. Kim, V. K.
Sharma, Y. J. Wang, Bioconjug. Chem. 2010, 21, 1588–1595.
S. Knutson, E. Raja, R. Bomgarden, M. Nlend, A. Chen, R.
Kalyanasundaram, S. Desai, PLoS One 2016, 11, 1–25.
S. O. Doronina, B. E. Toki, M. Y. Torgov, B. A. Mendelsohn, C. G.
Cerveny, D. F. Chace, R. L. DeBlanc, R. P. Gearing, T. D. Bovee, C.
B. Siegall, et al., Nat. Biotechnol. 2003, 21, 778–784.
C. A. Boswell, E. E. Mundo, C. Zhang, D. Bumbaca, N. R. Valle, K.
R. Kozak, A. Fourie, J. Chuh, N. Koppada, O. Saad, et al.,
Bioconjug. Chem. 2011, 22, 1994–2004.
[10]
[11]
[12]
Y. T. Adem, K. A. Schwarz, E. Duenas, T. W. Patapoff, W. J.
Galush, O. Esue, Bioconjug. Chem. 2014, 25, 656–664.
J. R. Junutula, H. Raab, S. Clark, S. Bhakta, D. D. Leipold, S. Weir,
Y. Chen, M. Simpson, S. P. Tsai, M. S. Dennis, et al., Nat.
Biotechnol. 2008, 26, 925–932.
[13]
J. P. M. Nunes, V. Vassileva, E. Robinson, M. Morais, M. E. B.
Smith, R. B. Pedley, S. Caddick, J. R. Baker, V. Chudasama, RSC
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202013911
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
Anders Märcher, Johan Palmfeldt,
Marija Nisavic, Kurt V. Gothelf*
Page No. – Page No.
A Reagent for Amine-Directed
Conjugation to IgG1 Antibodies
Double up: A method for site-directed conjugation to antibodies is developed, in
which a first lysine both directs and activates the reagent for reaction with a
neighboring lysine.
6
This article is protected by copyright. All rights reserved.