Optimization of IEDDA bioorthogonal system: Efficient process to improve trans-cyclooctene/tetrazine interaction

Article abstract of DOI:10.1016/j.ejmech.2020.112574

The antibody pretargeting approach for radioimmunotherapy (RIT) using inverse electron demand Diels-Alder cycloaddition (IEDDA) constitutes an emerging theranostic approach for solid cancers. However, IEDDA pretargeting has not reached clinical trial. The major limitation of the IEDDA strategy depends largely on trans-cyclooctene (TCO) stability. Indeed, TCO may isomerize into the more stable but unreactive cis-cyclooctene (CCO), leading to a drastic decrease of IEDDA efficiency. We have thus developed both efficient and reproducible synthetic pathways and analytical follow up for (PEGylated) TCO derivatives, providing high TCO isomeric purity for antibody modification. We have set up an original process to limit the isomerization of TCO to CCO before the mAbs’ functionalization to allow high TCO/tetrazine cycloaddition.

Full text of DOI:10.1016/j.ejmech.2020.112574

European Journal of Medicinal Chemistry 203 (2020) 112574
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Optimization of IEDDA bioorthogonal system: Efficient process to
improve trans-cyclooctene/tetrazine interaction
a
,
b
,
c
,
,
b
,
c
,
2
a
,
b
,
c
,
d
,
1
ꢀ
ꢀ
Jean-Baptiste Bequignat
2, Nancy Ty a
, Aurelie Rondon
,
Ludivine Taiariol a c, Françoise Degoul a c, Damien Canitrot a
,
,
b,
,
b
,
, b,
c
Mercedes Quintana a c, Isabelle Navarro-Teulon d, Elisabeth Miot-Noirault a
,
,
b,
, b, c
Claude Boucheix e, Jean-Michel Chezal a c, Emmanuel Moreau a
,
b,
, b, c, *
a
ꢀ
ꢀ
ꢀ
ꢀ
Universite Clermont Auvergne, Imagerie Moleculaire et Strategies Theranostiques, BP 184, F-63005, Clermont-Ferrand, France
b Inserm, U 1240, F-63000, Clermont-Ferrand, France
c Centre Jean Perrin, F-63011, Clermont-Ferrand, France
d
ꢀ
ꢀ
Institut de Recherche en Cancerologie (IRCM), U1194 e Universite Montpellier e ICM, Radiobiology and Targeted Radiotherapy, 34298, Montpellier Cedex
5, France
e INSERM, UMR-S 935, Villejuif, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The antibody pretargeting approach for radioimmunotherapy (RIT) using inverse electron demand Diels-
Alder cycloaddition (IEDDA) constitutes an emerging theranostic approach for solid cancers. However,
IEDDA pretargeting has not reached clinical trial. The major limitation of the IEDDA strategy depends
largely on trans-cyclooctene (TCO) stability. Indeed, TCO may isomerize into the more stable but
unreactive cis-cyclooctene (CCO), leading to a drastic decrease of IEDDA efficiency. We have thus
developed both efficient and reproducible synthetic pathways and analytical follow up for (PEGylated)
TCO derivatives, providing high TCO isomeric purity for antibody modification. We have set up an
original process to limit the isomerization of TCO to CCO before the mAbs’ functionalization to allow high
TCO/tetrazine cycloaddition.
Received 21 February 2020
Received in revised form
11 June 2020
Accepted 11 June 2020
Available online xxx
Keywords:
Bioorganic chemistry
Pretargeting
Trans-cyclooctene
TCO isomerization
PEGylated linkers synthesis
© 2020 Elsevier Masson SAS. All rights reserved.
1. Introduction
However, RIT of solid tumors faces two major drawbacks, namely
poor mAb diffusion in tumors and hematotoxicity caused by their
Radioimmunotherapy (RIT) consists in treating tumors using
monoclonal antibodies (mAbs) radiolabeled with - or -emitting
slow blood clearance [1,2]. Recent advances have led to the devel-
opment of pretargeting approaches (PRIT) to limit irradiation of
healthy tissues (Fig. 1A) [3,4].
b
a
radionuclide, either directly (e.g. radioiodination on tyrosine resi-
dues) or through a radiolabeled prosthetic group (most notably
couplings between N-hydroxysuccinimidyl (NHS) esters and ly-
sines or hetero-Michael additions between maleimides and thiol
group of cysteine residues). RIT efficiency has been well established
in hematological diseases, such as Non-Hodgkin Lymphoma with
the approval in 2002 of ibritumomab tiuxetan (Zevalin®, Bayer).
Unlike RIT, PRIT relies on a specific recognition between a
functionalized mAb and a fast blood cleared radiolabeled probe,
injected 24e48 h after the modified mAb. However, only few PRIT
approaches have so far been investigated in clinical trials [5e7]. The
(strept)avidin-biotin system, which was the first PRIT system
assessed in humans, demonstrated important (strept)avidin-based
immunogenicity and non-specific interaction with endogenous
biotin. The bispecific antibodies/hapten system successfully
reached phase II/III but requires a long and expensive engineering
[8e14]. Copper-free click chemistry strategies offer an attractive
alternative due to their selectivity, fast reaction rate, ease and
modularity of ligation, particularly for the “second generation”
bioorthogonal reactions. These stand out by their fast reaction
* Corresponding author. UMR 1240, IMOST, Univ. Clermont Auvergne, INSERM, 58
rue Montalembert, F-63005, Clermont-Ferrand, France.
E-mail address: emmanuel.moreau@uca.fr (E. Moreau).
1
ꢀ
Present address: Universite Catholique de Louvain, Louvain Drug Research
Institute, ADDB, 73 avenue Mounier, 1200 Bruxelles, Belgique.
2
These authors contributed equally.
https://doi.org/10.1016/j.ejmech.2020.112574
0223-5234/© 2020 Elsevier Masson SAS. All rights reserved.

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
2
kinetics under physiological conditions (solvent, pH and tempera-
ture). Bioorthogonal chemistry is also characterized by the absence
of side reactions between bioorthogonal entities and biological
macromolecules, thus ensuring biocompatibility in living systems.
This led to the development of a dozen of reactions where the most
promising appeared to be strained-promoted alkyne-azide cyclo-
addition (SPAAC) and the inverse electron-demand Diels-Alder
(IEDDA) cycloaddition based on trans-cyclooctene (TCO)/tetrazine
(Tz) conjugation [15e18]. For PRIT applications, TCO moieties are
usually linked to mAb lysine residues, while Tz probes are attached
to a macrocycle chelator via polyethylene glycol (PEG) linkers
(Fig. 1B/C).
In 2010, Rossin et al. have first demonstrated the proof of
concept of IEDDA in living systems, via SPECT-CT imaging using
CC49 anti-TAG72 mAb (bearing z 7 TCO/mAb), and an [111In]In-
DOTA-Tz probe. Thereafter, IEDDA reaction based on TCO/Tz inter-
action has been extensively applied for either fluorescence assays,
SPECT/PET imaging or preclinical therapy studies and has proved to
background noise and high tumor to non-target organs ratios.
Thereby, imaging investigations demonstrated the possibility to
use IEDDA to specifically target tumors with low background noise
and high tumor to non-target organs ratios [19e24]. In addition,
using PRIT instead of RIT for tumor treatment was all the more
interesting since radiation doses received by non-targeted organs
were significantly lower with PRIT according to dosimetry calcu-
lations [25].
To the best of our knowledge, no clinical studies using pre-
targeting through IEDDA cycloaddition have been undertaken yet,
neither for imaging nor for therapy purpose. However, many pre-
clinical studies based on TCO/Tz interaction essentially focused on
the optimization of the Tz moiety and its linker to improve in vivo
reaction kinetic and Tz-based radioligand stability in biological
media (Fig. 1B) [26,27]. Concerning the TCO moiety, the first study
led by Robillard et al. also used a PEGylated linker (e.g. PEG12) [19],
PEGylation being a well-known process to reduce the lipophilicity
and immunogenicity of immunoconjugates while increasing their
blood half-lives [28]. But moving TCO groups away from the mAb
be
a powerful tool to specifically target tumors with low
Fig. 1. Principle of the pretargeting strategy (A) and major checkpoints to optimize the TCO/Tz IEDDA cycloaddition (BeC).

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
3
surface exposed them to serum protein bound copper (II) (e.g.
transcuprein, mouse serum albumin, ceruloplasmin) or thiol de-
rivatives, thus increasing TCO isomerization into its unreactive cis-
cyclooctene (CCO) isomer. Others authors have also developed
TCOs with exo- or endocyclic heteroatoms. Not only does hetero-
atom incorporation increase hydrophilicity but it also preserves (or
even increases) the reactivity of the cyclic dienophiles [29].
The pioneering teams in the development of the IEDDA system,
namely Robillard and Zeglis groups, obtained excellent in vivo re-
sults without PEGylated linker between TCO groups and their mAbs
(Fig. 1C) [30e32]. We also demonstrated, using fluorescent Tz
probes, that PEGylated mAb-TCO were less reactive towards Tz than
non-PEGylated mAb-TCO, whatever the length of the linker (e.g.
PEG4 and PEG12), either in vitro or in vivo [33,34]. However, Cook
et al. recently described the evaluation of a novel bifunctional
PEGylated dendrimeric TCO for in vivo PET imaging, resulting in a
two-fold improvement of the absolute tumoral uptake [35]. We
thereby wondered if one pitfall of the IEDDA system for further
clinical application was not the TCO group itself. The purpose of the
present study was to determine which critical step(s) could induce
TCO isomerization: (i) the introduction of the PEGylated linker
during the synthesis of PEGylated TCO derivatives; (ii) the grafting
process of TCO derivatives (PEGylated or not) on the mAb; (iii) and/
or in vitro and in vivo instability of the mAb-TCO immunoconju-
gates (either PEGylated or not). To answer these three questions,
we first systematically checked the isomerization rate of TCO dur-
ing all chemical syntheses by analytical HPLC prior to mAb func-
tionalization. However, TCO/CCO conversion can no longer be
accurately quantified by analytical HPLC once mAbs were func-
tionalized. Subsequently, we first quantified TCO/Tz interaction by
SDS-PAGE-gel electrophoresis, by reacting commercially available
fluorescent Tz probes (Tz-Cy3 and Tz-5-FAM) with Ts29.2 mAb
conjugates bearing both three different PEG lengths (PEG0, PEG4
and PEG12) and three different TCO/CCO isomeric ratios (90/10; 50/
50; 10/90). We then assessed the influence of TCO/CCO isomeric
purity by immunofluorescence assays, on HT29 colorectal cell line
expressing tetraspanin 8 (TSPAN8) antigen, using the correspond-
ing modified Ts29.2 mAb. The impact of both PEGylation and TCO
isomeric purity on mAb-PEGx-TCO stability (x ¼ 0, 4 and 12) were
finally evaluated by SDS-PAGE gel electrophoresis monitoring after
prior incubation of the different immunoconjugates at 37 ꢀC in PBS
or plasma.
Scheme 1. TCO derivative syntheses with high TCO/CCO ratio.
focused on the purification of the crude TCO-NHS obtained in the
first attempt. An original purification process on AgNO3-impreg-
nated silica gel was developed allowing to successively elute
CCOeNHS and impurities with ethyl acetate (control by TLC and
HPLC) while TCO-NHS was selectively retained by AgNO3. Then,
TCO-NHS was eluted with a mixture of ethyl acetate/triethylamine
(80/20, v/v), leading to a 100% isomeric purity of compound TCO-
NHS. This original quick and inexpensive method, based on the
synthetic procedure to obtain TCO from CCO, allowed to obtain
TCO-NHS with a 100/0 TCO/CCO ratio in a reproducible manner
(Fig. S3) [15,34]. Thereafter, to obtain PEGylated TCO derivatives 7a-
c, we investigated several synthesis methodologies to access
PEGylated amino acid precursors NH2PEGxCO2H 5a-c (Scheme 1).
These different strategies are explained in the supporting infor-
mation (Figs. S4eS7). We finally set an original process for the
synthesis of the PEGylated amino acid linkers 5a-c from the cor-
responding N3PEGxOH involving Jones oxidation of silyl ethers s9a-
c as key step (Scheme 2).
The PEGylated TCO-derivative 6a was then obtained by coupling
TCO-NHS with the PEGylated amino acid linker 5a, but with the
concomitant formation of its PEGylated CCO derivative 8a and
traces of TCO-NHS. Attempts to isolate 6a all proved to be unsuc-
cessful and systematically led to an increased isomerization rate
with poor recovery of the desired product regardless the purifica-
tion method used (washing protocols, purification by normal,
reversed phase or AgNO3-silica gel flash chromatography). There-
fore, once the starting TCO-NHS was almost totally consumed
(HPLC monitoring), the crude mixture had to be directly engaged in
the following step, using DSC (3 eq.) and catalytic DMAP (0.2 eq.) as
optimal experimental conditions, to afford the corresponding
activated ester 7a with an overall yield of 66% and 94% isomeric
purity (Fig. S8).
2. Results
2.1. Synthesis of pretargeting PEGylated components
The synthesis of PEGylated TCO derivatives started with the
photochemical conversion of the commercially available (Z)-
cyclooct-4-enol (CCO) into (E)-cyclooct-4-enol (TCO) (Scheme 1).
We succeeded in obtaining TCO with highly reproducible purity
rate (e.g. 100/0 TCO/CCO ratio) compared to a 70/30 ratio value
previously described by Rossin and co-workers (Fig. S1, A-B)
[21,36]. A one-pot two-step procedure was first applied to obtain
pure TCO-NHS and CCOeNHS compounds. Unfortunately, this
synthesis was not as reproducible as expected: most of crude TCO-
NHS batches usually contained CCOeNHS isomer in various pro-
portions (up to 22%) with more or less impurities (Fig. S1 C-D).
Moreover, conventional chromatographic separation of TCO-NHS
from CCOeNHS isomers on normal phase is complex due to their
similar polarity. To minimize TCO isomerization, we first tried to
perform direct CCOeNHS irradiation into TCO-NHS based on a
process recently developed by Bormans and co-workers [37]. Un-
fortunately, this reaction led only to TCO (2) probably due to the
photochemical cleavage of the carbamate bond (Fig. S2). Finally, we
This one-pot two-step method applied to 5a-c led to 7a-c with
yields ranging from 31 to 67% and with low TCO isomerization after
purification by preparative RP-HPLC followed by freeze-drying
(Scheme 1; Table 1, S1 and S2).
Using these experimental conditions afforded 7a-c with TCO/
CCO ratio superior to 90/10. CCOPEGxCO2H 8a-c and CCOPEGxNHS
9a-c were prepared as reference compounds to control the TCO
isomerization rate to CCO by analytical HPLC monitoring during the
synthesis of compounds 7a-c (Fig. S9). All our TCO- and CCO-
PEGxNHS proved to be stable and did not isomerize when
conserved at ꢁ20 ꢀC as stock solutions in acetonitrile for at least
one year.
2.2. Antibody functionalization
Biological experiments were then performed with Ts29.2-based
immunoconjugates directed against tetraspanin 8 antigen. The

ꢀ
4
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
Scheme 2. Synthesis of compounds NH2PEGxCO2H 5a-c (x ¼ 4, 8 and 12).
Table 1
Synthesis of compounds TCOPEGxCONHS 7a-c (x ¼ 4, 8 and 12).
Reactant
Step 1 (5a-c into 6a-c)
Step 2 (6a-c into 7a-c)
Yield (%)
Product
TCO/CCO ratio
Product
TCO/CCO ratio
5a
5b
5c
6a
6b
6c
100/0
100/0
94/6
7a
7b
7c
94/6
99/1
93/7
66
31
67
in vitro and in vivo proof-of-concepts of IEDDA pretargeting for
fluorescence imaging of colorectal cancer and peritoneal carcino-
matosis were recently published by our team [32,33]. Different
amounts of CCO derivatives (e.g; CCOeNHS 4 or 9a or 9c) were
added to the TCO derivatives (e.g; TCO-NHS 3 or 7a or 7c, respec-
tively) adducts so as to obtain fixed TCO/CCO ratios (e.g. 90/10, 50/
50 and 10/90). Control of TCO/CCO ratios was systematically per-
formed using analytical HPLC both before and after adding CCO
isomers (Scheme 3; Fig. S10, S11 and S12).
Scheme 3. Ts29-2 functionalization with mixtures of TCO- and CCOeNHS esters.
Ts29.2 was then modified by addition of the appropriate TCO-
NHS 3/CCOeNHS 4 mixture (ratio 1, 2 or 3) to afford the expected
modified Ts29-2-TCO immunoconjugates, namely Ts29.2(PEG0)
@1a, Ts29.2(PEG0)@2a or Ts29e2(PEG0)@3a. Similar procedures
were performed to produce Ts29.2(PEG4)@1b to Ts29.2(PEG12)@3c
by addition of the appropriate mixture of 7a/9a or 7c/9c for ratios
1e3 (Scheme 3). The mean number of TCO-CCO moieties grafted
per mAb ranged from 7.6 to 7.9 for Ts29.2(PEG0)@1a-Ts29.2(PEG0)
@3a and from 4.1 to 4.9 for both Ts29.2(PEG4)@1b-Ts29.2(PEG4)
@3b, Ts29.2(PEG12)@1c-Ts29.2(PEG12)@3c (n ¼ 3) (Table S3).
2.3. Immunofluorescence assays on HT29 cells
Assays were made with all Ts29.2-PEGx-TCO CCO (x ¼ 4, 8 and
12) (e.g. Ts29.2(PEG12)@1a to Ts29.2(PEG0)@3c, Fig. 2) as described
in supporting information.
Whatever the PEG linker used (PEG0, PEG4 or PEG12), we
observed, as expected, a strong decrease of the fluorescence in-
tensity located at the cell membrane in correlation with the
decrease of the TCO proportion. As expected, the mean fluores-
cence intensity, reflecting the TCO/Tz ligation, was higher with
TCO/CCO ratio of 90/10 whatever the length of PEG chain but
became weaker as the amount of CCO increases (e.g. TCO/CCO
isomeric purity ratio decreases) (Fig. 3). Quantification of the mean
fluorescence intensity confirmed this tendency for all PEG linkers.
It was observed that the decrease of the mean fluorescence in-
tensity is lower with PEG0 than PEG4 and PEG12 for all the corre-
sponding isomeric ratios (Figs. 2 and 3). The presence of PEGylated
linkers was suspected to induce a faster TCO isomerization.
Nevertheless, we can notice that for a 90/10 TCO/CCO ratio, a higher
mean fluorescence signal was obtained with PEG4. Then, the in-
fluence of PEGylation on Ts29.2-PEGx-TCO/CCO stability over time
was assessed using SDS-PAGE gel electrophoresis after prior in-
cubations of all immunoconjugates either in D-PBS (control, Fig. 4)
or in murine plasma, at 37 ꢀC for 1e24 h (Fig. 5).
Fig. 2. Assessment of the interaction between Ts29.2-TCO/CCO and Tz-Cy3 through
immunofluorescence assays.
In D-PBS at 37 ꢀC, a progressive decrease of fluorescence in-
tensity was observed as expected, as the CCO isomeric rate
increased whatever the PEG length. With PEG0/PEG4 immuno-
conjugates prepared from ratio 1 (e.g. TCO/CCO 90/10), the fluo-
rescence intensity, reflecting the interaction between Ts29.2-PEGx-
TCO and the tetrazine probe, slightly decreased in the first hours
then remained relatively stable between 5 h and 24 h. However, a
significant decrease was observed for ratios 2 and 3 regardless the
PEG length over the 24 h incubation period. Regarding PEG12 de-
rivatives, the fluorescence intensity decreased continuously over
24 h, whatever the starting ratio. We have checked that the incu-
bation between immunoconjugates and Tz-5-FAM before gel
electrophoresis gave similar results compared to incubation of the
gel in Tz bath after migration (Fig. S15).
Concerning the stability of the three immunoconjugates

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
5
3. Discussion
While immunopretargeting is not a recent approach, the
development of bioorthogonal chemistry has challenged this
concept by favouring dosimetry through the specific delivery of
therapeutic radiations to solid tumors while reducing the non-
specific uptake in non-targeted tissues. The most promising
method is based on the trans-cyclooctene (TCO)/tetrazine (Tz)
cycloaddition (IEDDA) which has already been the subject of
several publications for both optical imaging, PET, SPECT and
radioimmunotherapy (bꢁ and
a) [17,19,24,30,32,34,38e40]. More-
over, a recent review exhaustively identified 25 PubMed articles
reporting preclinical imaging and five therapy studies with full
mAbs as targeting vectors, since its first application in 2010 [7].
However, to the best of our knowledge, no clinical trials have yet
been undertaken with this concept, neither for imaging nor for
therapy purpose, despite these very promising in vivo results re-
ported in different models of cancer. This is quite surprising
considering that bioorthogonal chemistries, previously difficult to
synthesize, are nowadays commercially available. The Achilles’ heel
of the TCO/Tz tandem thereby seems to rely on the stability of the
TCO group itself. Indeed, if TCO isomerizes into its less-reactive
form cis-cyclooctene (CCO), the ligation with tetrazine becomes
very slow. To counteract this known isomerization issue, assess-
ments of either modifications on TCO structure [5,41] or increasing
of the number of TCO moieties conjugated on mAbs via the addition
of dendritic TCO were performed and led to an improvement of TCO
stability, but also highlighted that optimizations are still needed.
Other interesting strategies consisting in the reaction of Tz moieties
with constrained alkyne derivatives -which would prevent the
isomerization-are also under investigation but the ligation kinetic
is less efficient [5].
Despite potential isomerization, the bioconjugation of TCO with
antibodies remains an interesting therapeutic option for the po-
tential treatment of many cancers. We here thereby decided to
focus on the optimization of TCO derivatives and to investigate the
process of TCO/CCO isomerization supposedly occurring:(i) during
the synthesis of PEGylated TCO derivatives; (ii) and/or during the
grafting process of (PEGylated or not) TCO derivatives on the mAb;
(iii) and/or during in vitro and/or in vivo experiments.
The synthesis of PEGylated amino acid derivatives -not always
commercially available in Europe-is often expensive when PEG are
longer than four oligomeric units and not easy to synthesize. We
therefore developed several synthetic methods to obtain these
different PEGylated amino acids which could finally be extended to
different PEGylated derivatives longer than 12 oligomeric units.
During the synthesis of PEGylated TCO derivatives, we moni-
tored the isomerization of TCO by analytical HPLC at each step. We
have thus demonstrated that the precursor TCO-NHS can be easily
and efficiently obtained with 100% of isomeric purity via the use of
an AgNO3/silica cartridge. In addition, the optimized TCO-NHS
precursor was also employed in our team for the conjugation of
anti-CEA 35A7 mAb [33] as well as for clinical approved mAbs such
as Rituximab, Cetuximab and Panitunimab (data not shown).
This is quite interesting as most studies recommend to directly
conjugate TCO to the mAb, without any PEG linkers, in order to
improve the in vivo stability induced by the shielding effect of the
mAb against Cu-containing serum proteins supposedly isomerizing
TCO into non-reactive CCO [31]. This optimization could be helpful
to a wide variety of new applications, such as the recently devel-
oped “PeptoBrush concept”, which allow a higher TCO-loading on
an biodegradable co-polymer showing EPR-mediated tumor accu-
mulation than on antibodies for which it is generally advised to not
exceed 4 grafted entities (2<DAR<5) or the dendrimer-bearing TCO
conjugated site-specifically to the Fc fragment of mAb
Fig. 3. Quantification of the mean fluorescence intensity localized on the cell mem-
brane using ImageJ software. *P < 0.05: 0/0 vs 10/90 and 50/50 vs 90/10; * P < 0.001: 0/
0 vs 10/90, ** P < 0.0001: 0/0 vs 90/10 and 50/50, 10/90 vs 50/50 and 90/10, and 50/50
vs 90/10. (n ¼ 3 independent experiments).
incubated at 37 ꢀC in murine plasma, for each PEG length and
whatever the TCO/CCO ratio, fluorescence intensity significantly
decreased within 5 h of incubation compared to D-PBS. However,
from 5 h to 24 h, there were few differences except for PEG12
immunoconjugates because there are no reactivity after 5h of
contact in plasma. We found that the higher the percentage of CCO
before mAb grafting, the faster the fluorescence intensity
decreased, and consequently, the TCO/Tz interaction. Thereby, after
5 h of incubation in plasma, the presence of a PEGylated linker
seemed to dramatically increase TCO to CCO isomerization. Indeed,
for PEG12, a total loss of fluorescence intensity occurred within 5 h,
compared to 80% and 90% loss for PEG0 and PEG4 conjugates
respectively. These results suggested that introducing a long PEG
linker on mAb could be highly detrimental for in vivo IEDDA
cycloaddition between TCO and Tz as TCO groups would be more
easily exposed to plasma proteins. Conversely, Ts29.2(PEG0)@1a-
Ts29.2(PEG0)@3a conjugates would be potentially more stable
in vivo, which could be explained by the proximity of TCO from the
mAb surface.

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
6
Fig. 4. Influence of the amount of CCO in the reaction medium before Ts29.2 grafting on both Ts29.2-PEGx-TCO stability in D-PBS and interaction with Tz-5-FAM. * P < 0.05: 50/50 vs
90/10 and 10/90 vs 50/50; ** p < 0.001: 10/90 vs 90/10; ¤ p < 0.05: 24 h vs 1 h, 5 h vs 1 h, 24 h vs 5 h ** p < 0.001: 5 h and 24 h vs 1 h (n ꢂ 3 independent experiments).
Fig. 5. Influence of the amount of CCO in the reaction medium before Ts29.2 grafting on both Ts29.2-PEGx-TCO stability in mice plasma diluted in 50% D-PBS and interaction with
Tz-5-FAM. *P < 0.0001: 50/50 vs 90/10, 10/90 vs 90/10 and 10/90 vs 50/50; **p < 0.001: PEG4 vs PEG0, PEG12 vs PEG0 and PEG12 vs PEG4; ¤ p < 0.0001: 24h vs 1 h, 24 h vs 5 h and
5 h vs 1 h (n ꢂ 3 independent experiments).
[5,35,42e45]. Indeed, the concept of "high TCO loading" undoubt-
edly compensates for the non-reactivity of isomerized TCO by
increasing the possibility of ligation towards Tz. Moreover, ensuring
that TCO-NHS precursors are 100% isomerically pure before func-
tionalizing the co-polymer of the dendritic structure would be an
additional advantage to reduce the inactivation of moieties. With
this in mind, our method for controlling TCO isomerization rate
below 10% with PEGylated linkers prior to mAb grafting makes
sense. In our experiments no isomerization was observed during
the conjugation of TCO derivatives (PEGylated or not) on the mAb
(Fig. 3). We can even note that 90/10 and 50/50 ratios have given
almost similar results with no PEGylated spacers in D-PBS at 37 ꢀC.
From 5 h to 24 h of incubation in D-PBS, there are few differences
except for the PEG12 conjugate. The results highlight that the
control of isomerization rate prior to antibody conjugation is an
important issue and could be a helpful tool for further in-
vestigations to increase active TCO at the tumour level and slowing
its possible in vivo isomerization with or without PEGylated
spacers.
before and during the modification of the antibody, we then
questioned about our third hypothesis concerning the in vivo
deactivation of TCO. Within the immunoconjugate assays, for each
PEG length and whatever the TCO/CCO ratio we observed that
fluorescence intensity slightly decreased over 5 h of incubation in
plasma (Fig. 5). In addition, the higher the percentage of CCO before
the functionalization of the mAb, the lower the fluorescence in-
tensity. After 5 h of incubation in this media, the presence of a
PEGylated linker dramatically raised the isomerization of TCO into
CCO. Once again, we observed that the higher the percentage of
CCO before mAb’s functionalization the lower the fluorescence in-
tensity. The present study is consistent with our previous work,
confirming that Ts29.2(PEG0)@1a is the most appropriate conju-
gate for further PRIT studies, and brings new information about the
behaviour of Ts29.2 immunoconjugates in plasma, hence reflecting
their potential behaviour in vivo [33].
In vitro assays using Ts29-PEGx-TCO conjugates confirmed that a
fast TCO isomerization was correlated to an increased length of
PEGylated linkers, and more notably in murine plasma (superior to
90% for PEGylated conjugates after only 5 h of incubation). Despite
Since we are able to optimize and control the isomerization

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
7
this surprising results obtained in vitro after 5 h, we observed
in vivo IEDDA ligation 24 h following injection of mAb-TCO with
fluorescent Tz or [111In]In-DOTA-Tz probes [33,34].
using 60 F254 or alumina gel 60A þ F254 plates and visualized with
UV light (UV lamp Fisher Bioblock Scientific, 365 nm or 254 nm),
dragendorff revelator (1.6 g of bismuth (III) nitrate for 100 mL of
glacial acetic acid/H2O (1/4, v/v), KMnO4 (1.5 g for 100 mL of water)
or phosphomolybdic acid (8 g for 200 mL of ethanol). Uncorrected
melting points (mp) were measured on an IA9100 Digital Melting
Point Apparatus. Purifications by column chromatography were
performed on silica gel (Chromagel 60 ACC, 40e63 mm, Carlo Erba
Reagents). Purifications by preparative reversed-phase HPLC were
conducted on a Combiflash EZ prep system (Teledyne Isco) with a
Moreover, several in vivo PRIT investigations of using IEDDA
cycloaddition reported interactions between mAbs-TCO and Tz
injected 48 or 72 h after the immunoconjugate [34,44]. The authors
demonstrated that the interaction is nevertheless weaker than the
one observed at 24 h (i.e. 11 and 13 vs 21 %IA/g, respectively) [44].
For present results, we can hypothesize that the incubation of the
mAbs TCO in this ex vivo condition did not reflect entirely the
in vivo situation: plasma is not full blood, absence of fluid mech-
anism, ratio of mAbs-TCO/plasmatic protein was different (in vivo
RP-18 column (250 mm ꢃ 20 mm, pores 100 Å, particles 5
mM). All
the PEG derivatives used in synthesis were systematically dried by
co-evaporation with anhydrous toluene twice and flushed with
argon before use, unless otherwise stated. HT29 cell line, a colon
adenocarcinoma cell line expressing TSPAN8, and the non-
internalizing murine monoclonal antibody (mAb) directed against
tetraspanine 8, Ts29.2 (IgG2b), were provided by Dr. C. Boucheix
(Inserm, Villejuif, France). Cells were cultured in Dulbecco’s Modi-
fied Eagle F12 Medium supplemented with 10% fetal serum, 1%
penicillin/streptomycin, 1% geneticin, 1% hygromycin at 37 ꢀC with
5% CO2 in humidified environment. For the different experiments,
two fluorescent Tz probes were purchased at JenaBioscience (Ger-
many), namely 3-(p-benzylamino)-1.2.4.5-tetrazine-5-Fluorescein
(Tz-5-FAM, absorbance/emission wavelengths 492/517 nm, mo-
lecular weight (MW) 545.50 g/mol) and 3-(p-benzylamino)-1.2.4.5-
tetrazine-cyanine3 (Tz-Cy3, absorbance/emission wavelengths
550/570 nm, MW 908.07 g/mol). Those different fluorophores do
not alter the ability of Tz to bind TCO moieties due to the high af-
finity and specificity of Tz for TCO and due to their ability as organic
dyes to not interfere with biological functions. In addition, in a
same experiment, we always compared mAb immunoconjugates
using the same fluorescent Tz probe -e.g the most adapted to obtain
significant signal detection according to the laser capacity of the
device used-in order to avoid introduction of any bias. Signal
observation was made using a confocal imager (Leica SPE, ICCF
platform, Clermont-Ferrand, France). Statistical analysis was per-
formed using XLSTAT 2012 software. Continuous data were
expressed with the mean and standard deviation (SEM). To
compare continuous values of the different modified antibodies, a
one-way ANOVA or two-paired Student T-test were used. We
considered p < 0.05 statistically significant. Native monoclonal
antibody Ts29.2 was solvent-exchanged and concentrated on
Amicon Ultra-4 centrifugal filter units (50 kDa MW cut-off) using
50
in 50
m
g mAbs TCO diluted in 1.3 mL of blood and in vitro 5
mg diluted
mL of plasma). Additionally, potential mAb-TCO degradation
cannot be checked with this approach. Our in vitro experiments
were then performed under drastic conditions to demonstrate the
interest of controlling the isomerization rate before conjugation in
order to maximize the possibility of reaction towards Tz in vivo. The
results obtained cannot be considered as the reflect of in vivo
processes of TCO isomerization. This demonstrated a loss of in vivo
TCO/Tz interaction over time, thereby corroborating our in vitro
observations. In view of the recent results of Zeglis [44], this is not
only due to isomerization but also to other mechanisms that will
have to be highlighted in order to clinically transfer this concept of
pretargeting as soon as possible.
4. Conclusion
In conclusion, we here developed an efficient method to syn-
thesize (PEGylated)TCO derivatives with high isomeric purity and
monitoring the TCO:CCO ratio via HPLC before the conjugation of
mAbs. Our original method is easy to use, fast and highly repro-
ducible either for direct TCO activated ester precursors or for the
conjugation of PEGylated TCO derivatives. The AgNO3 impregnated
silica gel allows obtaining precursor 3 with a 100% of isomeric
purity while the optimized one-pot two-step synthesis process of
compounds 7a-c, 8a-c and 9a-c, resulting in excellent TCO isomeric
purity of at least 94% before mAbs conjugation. In vitro assays using
Ts29-PEGx-TCO conjugates confirmed thereafter that a fast TCO
deactivation was correlated to an increased length of PEGylated
linkers, and more notably in murine plasma (superior to 90% for
PEGylated conjugates after only 5 h of incubation).
This work represents a helpful tool for further investigations on
TCO/Tz reaction for PRIT experiments by allowing controlling the
TCO isomerization rate to obtain a TCO prosthetic group with a
TCO:CCO ratio as close as possible to 100:0 before mAb conjugation.
The process decried herein can be applied to a wide range of mAbs
thereby arousing a great interest for further SPECT/PET imaging
and/or PRIT applications using IEDDA cycloaddition, especially for
considering the GMP transfer of IEDDA bioorthogonal chemistry to
the clinic.
D-PBS (Gibco) to reach a final concentration between 5 and 10
mg/
mL before use, whereas TCO-grafted antibodies were purified on
Zeba desalting spin columns (40 kDa MW cut-off, 0.5 mL, Ther-
moFisher scientific). Concentrations of mAb-TCO conjugates were
determined with a MultiskanTM go spectrophotometer (Thermo-
Fisher Scientific) at 280 nm while the mean number of TCO grafted
by mAb was determined by MALDI-ToF mass spectrometry.
Syntheses. The synthesis of TCO (2), TCO-NHS (3), CCOeNHS (5)
and all others syntheses were developed in the supporting
information.
5. Experimental section
5.1. Materials for chemical syntheses
5.1.1. General procedure for syntheses of compounds (7a-c) and
(9a-c)
Unless otherwise mentioned, all manipulations were performed
under argon atmosphere; all reagents were purchased from the
following commercial suppliers: Sigma-Aldrich, Acros Organics,
Carlo Erba, TCI Europa, Alfa Aesar. Anhydrous DMF, anhydrous
triethylamine and anhydrous pyridine were purchased from Acros
Organics. THF was dried over a Pure Solv™ Micro Solvent Purifi-
cation System (Sigma-Aldrich) equipped with an alumina column.
Dichloromethane was distilled over calcium hydride. Reaction
progresses were monitored by thin layer chromatography (TLC)
To a solution of TCO-NHS (3) or CCOeNHS (4) (approx.
20e30 mg, 1 eq.) in anhydrous dichloromethane (1 mL) were
successively added anhydrous triethylamine (10 eq.) and a solution
of the appropriate PEGylated aminoacid (5a-c) (1.5 eq.) in anhy-
drous dichloromethane (1.5 mL) at room temperature and under
argon atmosphere. After completion of the reaction (HPLC moni-
toring), DMAP (0.2 eq.) and N,N0-disuccinimidyl carbonate (DSC) (3
eq.) were added. The resulting reaction mixture was stirred for 1 h,
then successively washed with a hydrochloric acid solution (0.5 M)

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
8
(2.5 mL), water (3 x 4 mL) and brine (6 mL), dried over MgSO4,
filtered and evaporated under reduced pressure to afford the
desired crude product as a pale yellow lacquer. Then, the crude
product was purified by preparative RP-HPLC with a water/aceto-
nitrile gradient, namely 50/50 to 20/80 over 15 min then 20/80 to 0/
100 over 5 min, providing after freeze-drying the desired activated
ester. Stock solutions of TCO-NHS (3), CCOeNHS (4), TCOPEGxNHS
esters (7a-c) and CCOPEGxNHS esters (9a-c) in DMSO for antibody
conjugation were prepared and stored at ꢁ20 ꢀC in the dark and
proved to be stable for at least a year.
69.9, 70.2, 70.5, 70.6 x 2, 70.7, 81.4, 117.7, 128.2, 129.0, 133.1, 134.9,
141.2, 152.8, 166.7, 166.9, 168.9; HRMS (ESIþ) m/z calculated for
C
47H76O19N3 986.5068 [MþH]þ; found 986.5064.
5.1.1.4. (Z)-2,5-dioxopyrrolidin-1-yl 1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl)-1-oxo-5,8,11,14-tetraoxa-2-azaheptadecan-
17-oate (9a). General procedure using CCOeNHS (4) (20 mg,
0.052 mmol, 1 eq.), anhydrous triethylamine (72 mL, 0.52 mmol, 10
eq.) and 5a (20.7 mg, 0.078 mmol, 1.5 eq.), followed 4 h after by
addition of DMAP (1.3 mg, 0.010 mmol) and DSC (42 mg,
0.156 mmol) (HPLC monitoring) afforded compound 9a as a color-
less oil (19.7 mg, 0.031 mmol, 60%). HPLC: TR ¼ 7.67 min
5.1.1.1. (E)-2,5-dioxopyrrolidin-1-yl 1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl)-1-oxo-5,8,11,14-tetraoxa-2-azaheptadecan-
17-oate (7a). General procedure using TCO-NHS (3) (20 mg,
(
lmax ¼ 265 nm), purity ¼ 98%. Revelator: UV and phosphomo-
lybdic acid. 1H NMR (500 MHz, CDCl3)
d 1.57e2.40 (m, 10H,
0.052 mmol, 1 eq.), anhydrous triethylamine (72
m
L, 0.52 mmol, 10
H2’’,H3’’, H4’’,H7’’,H8’’), 2.79 (s, 4H, Hb), 2.83 (t, J ¼ 6.5 Hz, 2H, H2),
3.58e3.64 (m,16 H, CH2eO), 3.78 (t, J ¼ 6.5 Hz, 2H, H3), 4.83 (m,1H,
H1’’), 5.55e5.71 (m, 2H, H5’’, H6’’), 6.92 (brs, 1H, NH), 6.98 (brs, 1H,
NH), 7.42 (d, J ¼ 9.0 Hz, 2H, H20, H60), 7.74 (d, J ¼ 9.0 Hz, 2H, H30,
eq.) and 5a (20.7 mg, 0.078 mmol), followed 4 h later by addition of
DMAP (1.3 mg, 0.01 mmol) and DSC (42 mg, 0.156 mmol) (HPLC
monitoring) afforded compound 7a as a colorless oil (21.7 mg,
0.034 mmol, 66%). HPLC: TR ¼ 7.21 min (lmax ¼ 265 nm),
purity ¼ 99%, TCO/CCO ¼ 94/6. Revelator: UV and phosphomolyb-
H5’); 13C NMR (125 MHz, CDCl3)
d 22.3, 24.8, 25.6, 32.1, 33.8, 33.9,
39.8, 65.7, 69.9, 70.2, 70.4, 70.5 x 2, 70.6, 70.7, 77.0, 116.3, 117.7,
128.2, 128.9, 129.5, 129.8, 141.2, 152.9, 166.8, 167.0, 169.1; HRMS
(ESIþ): m/z calculated for C31H44O11N3 634.2970 [MþH]þ; found
634.2966.
dic acid. 1H NMR (500 MHz, CDCl3)
d 1.56e2.40 (m, 10H, H2’’,H3’’,
H4’’,H7’’,H8’’), 2.81 (s, 4H, Hb), 2.84 (t, J ¼ 6.5 Hz, 2H, H2), 3,59-3,66
(m, 16 H, CH2eO), 3.79 (t, J ¼ 6.5 Hz, 2H, H3), 4.45 (m, 1H, H1’’),
5.47e5.65 (m, 2H, H5’’, H6’’), 6,85 (brs, 2H, NH), 7.42 (d, J ¼ 9.0 Hz,
2H, H20, H60), 7.75 (d, J ¼ 9.0 Hz, 2H, H30, H5’); 13C NMR (125 MHz,
5.1.1.5. (Z)-2,5-dioxopyrrolidin-1-yl 1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl)-1-oxo-5,8,11,14,17,20,23,26-octaoxa-2-
azanonacosan-29-oate (9b). General procedure using CCOeNHS (4)
CDCl3)
d 25.6, 31.0, 32.2, 32.5, 34.2, 38.6, 39.8, 41.1, 65.7, 69.8, 70.3,
70.5 (2C), 70.6 (2C), 70.7, 81.5, 117.7, 128.2, 129.0, 133.1, 134.9, 141.1,
152.8, 166.7, 166.9, 169.0; HRMS (ESIþ) m/z calculated from
(20 mg, 0.052 mmol), anhydrous triethylamine (72 mL, 0.52 mmol,
C
31H44O11N3 634.2970 [MþH]þ; found 634.2980.
10 eq.) and 5b (34.3 mg, 0.078 mmol, 1.5 eq.), followed 21 h after by
addition of DMAP (1.3 mg, 0.010 mmol) and DSC (42 mg,
0.156 mmol) (HPLC monitoring) afforded compound 9b as a
colorless oil (17.1 mg, 0.021 mmol, 41%). HPLC: TR ¼ 7.42 min
5.1.1.2. (E)-2,5-dioxopyrrolidin-1-yl 1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl)-1-oxo-5,8,11,14,17,20,23,26-octaoxa-2-
azanonacosan-29-oate (7b). General procedure using TCO-NHS (3)
(
lmax ¼ 265 nm), purity ¼ 99%. Revelator: UV and phosphomo-
(21 mg, 0.054 mmol), anhydrous triethylamine (75
m
L, 0.54 mmol)
lybdic acid. 1H NMR (500 MHz, CDCl3)
d 1.55e2.38 (m, 10H,
and 5b (35.8 mg, 0.081 mmol), followed 21 h later by addition of
DMAP (1.3 mg, 0.010 mmol) and DSC (42 mg, 0.156 mmol) (HPLC
monitoring) afforded compound 7b as a colorless oil (13,6 mg,
0.017 mmol, 31%). HPLC: TR ¼ 6.99 min (lmax ¼ 265 nm),
purity ¼ 97%, TCO/CCO ¼ 99/1. Revelator: UV and phosphomolybdic
H2’’,H3’’, H4’’,H7’’,H8’’), 2.80 (s, 4H, Hb), 2.85 (t, J ¼ 6,5 Hz, 2H, H2),
3.56e3.65 (m, 32 H, CH2eO), 3.80 (t, J ¼ 6,5 Hz, 2H, H3), 4.83 (m,
1H, H1’’), 5.59e5.71 (m, 2H, H5’’, H6’’), 6.92 (m, 1H, NH), 7.14 (brs,
1H, NH), 7.44 (d, J ¼ 9.0 Hz, 2H, H20, H60), 7.75 (d, J ¼ 9.0 Hz, 2H, H30,
H5’); 13C NMR (125 MHz, CDCl3)
d 22.3, 24.8, 25.6, 32.2, 34.0, 39.8,
acid. 1H NMR (500 MHz, CDCl3)
d
1.56e2.40 (m, 10H, H2’’,H3’’,
69.9, 70.2, 70.5 x 3, 70.6, 70.7, 77.0, 116.3, 117.8, 128.2, 128.9, 129.5,
129.8, 141.3, 153.0, 166.7, 166.9, 168.9; HRMS (ESIþ): m/z calculated
for C39H60O15N3 810.4019 [MþH]þ; found 810.4015.
H4’’,H7’’,H8’’), 2.80 (s, 4H, Hb), 2.86 (m, 2H, H2), 3.55e3.67 (m,16 H,
CH2eO), 3.81 (m, 2H, H3), 4.44 (m,1H, H1’’), 5.47e5.63 (m, 2H, H5’’,
H6’’), 6.92 (brs, 1H, NH), 7.18 (brs, 1H, NH), 7.43 (d, J ¼ 9.0 Hz, 2H,
H20, H60), 7.75 (d, J ¼ 9.0 Hz, 2H, H30, H5’). 13C NMR (125 MHz,
5.1.1.6. 2,5-dioxopyrrolidin-1-yl (Z)-1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl) ꢁ41-oxo-4,7,10,13,16,19,22,25,28,31,34,37-
dodecaoxa-40-azahentetracontanoate (9c). General procedure us-
ing CCOeNHS (4) (20 mg, 0.052 mmol, 1 eq.), anhydrous triethyl-
CDCl3)
d 25.6, 31.0, 32.2, 32.5, 34.2, 38.6, 39.8, 41.1, 65.7, 69.9, 70.2,
70.5 x 2, 70.6, 70.7 x 2, 81.3, 117.7, 128.2, 128.9, 133.0, 134.9, 141.3,
152.9, 166.7, 166.9, 168.9. HRMS (ESIþ) m/z calculated from
C
39H60O15N3 810.4019 [MþH]þ; found 810.4023.
amine (72 mL, 0.52 mmol, 10 eq.) and 5c (48.2 mg, 0.078 mmol, 1.5
eq.), followed 21 h after, DMAP (1.3 mg, 0.010 mmol) and DSC
(42 mg, 0.156 mmol) (HPLC monitoring) afforded compound 9c as a
colorless oil (24.7 mg, 0.025 mmol, 48%). HPLC: TR ¼ 7.13 min
5.1.1.3. (E)-2,5-dioxopyrrolidin-1-yl 1-(4-(((cyclooct-4-en-1-yloxy)
carbonyl)amino)phenyl)-1-oxo-5,8,11,14,17,20,23,26,29,32,35,38-
dodecaoxa-2-azahentetracontan-41-oate (7c). [31] General proced-
ure using TCO-NHS (3) (22 mg, 0.057 mmol, 1 eq.), anhydrous
(
lmax ¼ 265 nm), purity ¼ 99%. Revelator: UV and phosphomo-
lybdic acid. 1H NMR (500 MHz, CDCl3)
d 1.58e2.38 (m, 10H,
triethylamine (79
m
L, 0.57 mmol, 10 eq.) and 5c (52.8 mg,
H2’’,H3’’, H4’’,H7’’,H8’’), 2.80 (s, 4H, Hb), 2.86 (t, J ¼ 6.5 Hz, 2H, H2),
3.56e3.62 (m, 48 H, CH2eO), 3.81 (t, J ¼ 6.5 Hz, 2H, H3), 4.82 (m,
1H, H1’’), 5.61e5.70 (m, 2H, H5’’, H6’’), 6.83 (m, 1H, NH), 7.20 (brs,
1H, NH), 7.45 (d, J ¼ 8.5 Hz, 2H, H20, H60), 7.73 (d, J ¼ 8.5 Hz, 2H, H30,
0.086 mmol, 1.5 eq.), followed 21 h after by addition of DMAP
(1.4 mg, 0.011 mmol) and DSC (46.1 mg, 0.171 mmol) (HPLC
monitoring) afforded compound 7c as a colorless oil (37.6 mg,
0.038 mmol, 67%). HPLC: TR ¼ 6.78 min (lmax ¼ 265 nm),
purity ¼ 98%, TCO/CCO 93/7. Revelator: UV and phosphomolybdic
H5’); 13C NMR (125 MHz, CDCl3)
d 22.4, 24.7, 25.6, 32.3, 33.6, 33.9,
39.8, 65.8, 69.8, 70.3, 70.6, 70.7, 77.0, 117.9, 128.1, 129.2, 129.5, 129.7,
acid. 1H NMR (500 MHz, CDCl3)
d
1.56e2.40 (m, 10H, H2’’,H3’’,
141.3, 152.9, 166.6, 166.8, 168.7; HRMS (ESIþ): m/z calculated for
H4’’,H7’’,H8’’), 2.81 (s, 4H, Hb), 2.84 (m, 2H, H2), 3.58e3.64 (m,
48 H, CH2eO), 3.83 (m, 2H, H3), 4.45 (m, 1H, H1’’), 5.49e5.70 (m,
2H, H5’’, H6’’), 6.92 (brs, 1H, NH), 7.05 (brs, 2H, NH), 7.44 (d,
J ¼ 9.0 Hz, 2H, H20, H60), 7.75 (d, J ¼ 9.0 Hz, 2H, H30, H5’); 13C NMR
C
47H76O19N3 986.5068 [MþH]þ; found 986.5075.
5.2. HPLC monitoring
(125 MHz, CDCl3)
d
25.6, 31.0, 32.2, 32.5, 35.0, 38.6, 39.8, 41.1, 65.7,
Analyses and monitoring of reactions involving TCO and CCO

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
9
derivatives were performed by analytical RP-HPLC on an Agilent
1100 HPLC system using an Agilent Zorbax extend C-18 column
(Sigma Aldrich, France). Chambers were finally removed from the
labteks and blades were mounted in vectashield-DAPI mounting
medium. Signal observation was made using a confocal imager
(Leica SPE, ICCF platform, Clermont-Ferrand, France). Region of
(4.6 ꢃ 150 mm, 5
mm particles). Mobile phase consisted of a mixture
of 18 M deionized water (solvent A) and acetonitrile (solvent B,
U
HPLC PLUS-Gradient from Carlo Erba, France) both containing 0.1%
TFA. Elution was performed at 1 mL/min using a linear gradient
from 40 to 62.5% solvent B over 15 min followed by column
equilibration for 2 min after return to 40% solvent B in 3 min. So-
lutions of crude reaction mixtures and of purified compounds were
interest (ROI) of three fields per well, all Z-sections every 2 mm,
were randomly imaged and quantified on the entire image with
ImageJ software using an automated 3D-strategy. For more details
about the quantification method, please refer to our previous works
[7,33].
systematically filtered beforehand through a membrane (0.45
Macherey-Nagel) before being injected (20 L injection volume).
The chromatograms were recorded at 265 nm (lmax of TCO- and
CCOPEG4/8/12 carboxylic acids and NHS esters) and 286 nm
mm,
m
5.5. In vitro mAb-TCO/CCO immunoconjugates reactivity
,
About 5 mg of Ts29.2 without TCO; Ts29.2(PEG0)@1a;
3/7/11
(
lmax of TCO- and CCOeNHS). The TCO/CCO ratio was determined
Ts29.2(PEG0)@2a;
Ts29.2(PEG4)@2b;
Ts29.2(PEG12)@2c; Ts29.2(PEG12)@3c were added to D-PBS or
plasma 50% diluted in D-PBS and then incubated during 1 h, 5 h or
24 h at 37 ꢀC in a water bath with moderate agitation. After 1 h, 5 h
or 24 h of incubation, each mixture of mAb was added to Laemli 4X
Ts29.2(PEG0)@3a;
Ts29.2(PEG4)@3b;
Ts29.2(PEG4)@1b;
Ts29.2(PEG12)@1c;
using the peak area ratio of TCO and CCO derivatives, compounds
that were identified on the basis of their retention times when they
were previously injected separately.
5.3. MALDI-ToF
and D-PBS, so as to reach a final volume of 20
mL. Afterwards,
Determination of the mean number of TCO moieties per mAb.
Assessments were made using MALDI-TOF MS analyzes (Voyager
DE-Pro mass spectrometer, Sciex, USA). Sinapinic acid (Sigma,
France) was diluted in acetonitrile/water (30/70, v/v) with 0.1% TFA
at 10 mg/mL to obtain the matrix solution. A set of three serial
dilutions from 1 mg/mL of all mAb-PEGx-TCO conjugates in PBS was
prepared by mixing those latter with the matrix solution (2/1, v/v).
Acquisitions were performed in a positive linear mode and 600
shots were averaged for each spectrum. The average molecular
weight of mAb conjugates was obtained from the mass of the
[MþH]þ peak for the dilution set. Calibration settings corresponded
to a close external mode using IgG1 (AB Sciex, USA). The number of
TCO derivatives grafted per mAb was calculated using the molec-
ular weight difference between the mAb-PEGx-TCO conjugates and
the unmodified mAbs, net masses added depending on the modi-
fication (about 272, 520 and 872 Da for 3, 9a and 9c respectively).
Final concentrations of mAb-PEGx-TCO conjugates were measured
using a Multiskan GO UV/Vis microplate spectrophotometer at
samples were directly loaded without any reducing agent on 4e15%
SDS-PAGE acrylamide gels (Biorad, France) and after migration, gels
were incubated 5 min under gentle shaking with 0.002 mM of 3-(p-
benzylamino)-1.2.4.5-tetrazine-5-Fluorescein diluted in water (Tz-
5-FAM, absorbance/emission wavelengths 492/517 nm, 10e14
equivalents with respect to TCO/CCO). After 10 min wash in water,
gels were imaged with Chemidoc imager (Biorad, France) and then
colored with Simplyblue™ SafeStain following the manufacturer’s
instructions (Thermofisher Scientific, France). Major band was
quantified using ImageLab software, fluorescence intensity signal
being normalized by the total amount of protein loaded.
Funding sources
J-B. B received a PhD fellowship of Clermont-Auvergne Univer-
ꢀ
sity. A.R. received a PhD fellowship from FEDER and Region
^
Auvergne-Rhone-Alpes.
l
¼ 280 nm (Thermo Fisher Scientific, France).
Declaration of competing interest
5.4. Immunofluorescence assays for trans-cis cyclooctene
isomerization assay
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
8-well labtek chambers (Sigma Aldrich, France) were first
coated at 5
mg/cm2 with rat tail collagen I (Corning, USA) during
Acknowledgment
1 h at RT. After washing 3 times with 400
mL of D-PBS to remove
acetic acid, 1.105 A431-CEA-Luc cells, transfected for the expression
of carcino-embryonic antigen (CEA) and Luciferase were incubated
48 h in chambers. Cells were cultured in Dulbecco’s Modified Eagle
F12 Medium supplemented with 10% fetal serum, 1% penicillin/
streptomycin, 1% geneticin, 1% hygromycin at 37 ꢀC with 5% CO2 in
humidified environment. For immunofluorescence experiments,
cells were first incubated 30 min with D-PBS-BSA 5% at RT then
1.5 h at 37 ꢀC/5% CO2 with 0.03 nmol of the appropriate mixture of
mAb (Ts29.2 without TCO; Ts29.2@1a; Ts29.2@2a; Ts29.2@3a;
Ts29.2@1b; Ts29.2@2b; Ts29.2@3b; Ts29.2@1c; Ts29.2@2c;
Ts29.2@3c) diluted in D-PBS-BSA 2.5%. After washing the cells
twice with D-PBS to remove unfixed mAbs, they were incubated
45 min at 37 ꢀC/5% CO2 with either 1/1000 donkey anti-mouse
antibody labeled with Cyanine 3 (Jackson ImmunoResearch, USA)
diluted in D-PBS-BSA 2.5% or 10e14 equivalent of 3-(p-benzyla-
mino)-1,2,4,5-tetrazine-cyanine3 (Tz-Cy3, absorbance/emission at
550/570 nm, Jena Bioscience, Germany) with respect to the number
of TCO or CCO, diluted in D-PBS-BSA 2.5% - DMSO 0.5%. Cells were
then washed 3 times with D-PBS and fixed with 10% formalin
We warmly thank Pr Mohamed Saraka (ICCF of Clermont-Fd,
France) for the photochemical syntheses. Confocal imaging was
performed at the CLIC platform of Clermont-Ferrand (France).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2020.112574.
Abbreviations
IEDDA
RIT
Inverse electronic demand Diels-Alder
radioimmunotherapy
TCO
trans-cyclooctene
CCO
PRIT
SPECT
PET
cos-cyclooctene
pretargeted radioimmunotherapy
single photon emission computed tomography
positron emission tomography
polyethylene glycol
PEG

ꢀ
J.-B. Bequignat et al. / European Journal of Medicinal Chemistry 203 (2020) 112574
10
demand diels-alder click chemistry, Mol. Canc. Therapeut. 16 (1) (2017)
124e133.
TSPN8
tetraspanin8
SDS-PAGE sodium dodecyl sulfate polyacrylamide gel
electrophoresis
€
[25] R. Rossin, T. Lappchen, S.M. van den Bosch, R. Laforest, M.S. Robillard, Diel-
sealder reaction for tumor pretargeting: in vivo chemistry can boost tumor
radiation dose compared with directly labeled antibody, J. Nucl. Med. 54 (11)
(2013) 1989e1995.
References
€
[26] T. Lappchen, R. Rossin, T.R. van Mourik, G. Gruntz, F.J.M. Hoeben,
R.M. Versteegen, H.M. Janssen, J. Lub, M.S. Robillard, DOTA-tetrazine probes
with modified linkers for tumor pretargeting, Nucl. Med. Biol. 55 (2017)
19e26.
[1] J.-P. Pouget, C. Lozza, E. Deshayes, V. Boudousq, I. Navarro-Teulon, Introduc-
tion to radiobiology of targeted radionuclide therapy, Front. Med. 2 (12)
(2015) 1e12.
[2] J.-P. Pouget, I. Navarro-Teulon, M. Bardies, N. Chouin, G. Cartron, A. Pelegrin,
D. Azria, Clinical radioimmunotherapy-the role of radiobiology, Nat. Rev. Clin.
Oncol. 8 (12) (2011) 720e734.
[3] D.M. Goldenberg, C.-H. Chang, E.A. Rossi, W.J. McBride, R.M. Sharkey, Pretar-
geted molecular imaging and radioimmunotherapy, Theranostics 2 (5) (2012)
523e540.
[27] S. Kronister, D. Svatunek, C. Denk, H. Mikula, Acylation-mediated ‘kinetic
turn-on’ of 3-Amino-1,2,4,5-tetrazines, Synlett 29 (10) (2018) 1297e1302.
[28] A.P. Chapman, PEGylated antibodies and antibody fragments for improved
therapy: a review, Adv. Drug Deliv. Rev. 54 (4) (2002) 531e545.
ꢁ
ꢁ
ꢀ
[29] E. Kozma, I. Nikic, B.R. Varga, I.V. Aramburu, J.H. Kang, O.T. Fackler, E.A. Lemke,
Hydrophilic trans-cyclooctenylated noncanonical amino acids for fast intra-
cellular protein labeling, Chembiochem 17 (2016) 1518e1524.
[30] R. Rossin, S.M. van den Bosch, W. Ten Hoeve, M. Carvelli, R.M. Versteegen,
J. Lub, M.S. Robillard, Highly reactive trans-cyclooctene tags with improved
stability for dielsealder chemistry in living systems, Bioconjugate Chem. 24
(7) (2013) 1210e1217.
[4] M. Patra, K. Zarschler, H.-J. Pietzsch, H. Stephan, G. Gasser, New insights into
the pretargeting approach to image and treat tumours, Chem. Soc. Rev. 45 (23)
(2016) 6415e6431.
ꢀ
[5] E.J.L. Steen, P.E. Edem, K. Nørregaard, J.T. Jørgensen, V. Shalgunov, A. Kjaer,
M.M. Herth, Pretargeting in nuclear imaging and radionuclide therapy:
improving efficacy of theranostics and nanomedicines, Biomaterials 179
(2018) 209e245.
[31] S. Poty, R. Membreno, J.M. Glaser, A. Ragupathi, W.W. Scholz, B.M. Zeglis,
J.S. Lewis, The inverse electron-demand DielseAlder reaction as
a new
methodology for the synthesis of 225Ac-labelled radioimmunoconjugates,
ꢀ ꢀ
[6] C. Bailly, C. Bodet-Milin, C. Rousseau, A. Faivre-Chauvet, F. Kraeber-Bodere,
Chem. Commun. 54 (21) (2018) 2599e2602.
J. Barbet, Pretargeting for imaging and therapy in oncological nuclear medi-
cine, EJNMMI Radiopharm. Chem. 2 (2017) 6.
€
[32] R. Rossin, S.M. van Duijnhoven, T. Lappchen, S.M. van den Bosch,
M.S. Robillard, Trans-cyclooctene tag with improved properties for tumor
pretargeting with the dielsealder reaction, Mol. Pharm. 11 (9) (2014)
3090e3096.
[7] A. Rondon, F. Degoul, Antibody pretargeting based on bioorthogonal click
chemistry for cancer imaging and targeted radionuclide therapy, Bioconjugate
Chem. (2020), https://doi.org/10.1021/acs.bioconjchem.9b00761.
[8] P. Magnani, F. Fazio, C. Grana, C. Songini, L. Frigerio, S. Pecorelli, G. Mangili,
N. Colombo, C.D.A. Mariani, G. Paganelli, Diagnosis of persistent ovarian car-
cinoma with three-step immunoscintigraphy, Br. J. Canc. 82 (3) (2000)
616e620.
[33] A. Rondon, N. Ty, J.-B. Bequignat, M. Quintana, A. Briat, T. Witkowski,
B. Bouchon, C. Boucheix, E. Miot-Noirault, J.P. Pouget, J.M. Chezal, I. Navarro-
Teulon, E. Moreau, F. Degoul, Antibody PEGylation in bioorthogonal pre-
targeting with trans-cyclooctene/tetrazine cycloaddition: in vitro and in vivo
evaluation in colorectal cancer models, Sci. Rep. 7 (1) (2017), 14918.
[34] A. Rondon, S. Schmitt, A. Briat, N. Ty, L. Maigne, M. Quintana, R. Membreno,
B.M. Zeglis, I. Navarro-Teulon, J.P. Pouget, J.M. Chezal, E. Miot-Noirault,
E. Moreau, F. Degoul, Pretargeted radioimmunotherapy and SPECT imaging of
peritoneal carcinomatosis using bioorthogonal click chemistry: probe selec-
tion and first proof-of-concept, Theranostics 9 (22) (2019) 6706e6718.
[35] B.E. Cook, R. Membreno, B.M. Zeglis, Dendrimer scaffold for the amplification
of in vivo pretargeting ligations, Bioconjugate Chem. 29 (8) (2018)
2734e2740.
ꢀ ꢀ
[9] F. Kraeber-Bodere, J. Barbet, J.-F. Chatal, Radioimmunotherapy: from current
clinical success to future industrial breakthrough? J. Nucl. Med. 57 (3) (2016)
329e331.
[10] D.R. Stickney, L.D. Anderson, J.B. Slater, C.N. Ahlem, G.A. Kirk,
S.A. Schweighardt, J.M. Frincke, Bifunctional antibody: a binary radiophar-
maceutical delivery system for imaging colorectal carcinoma, Cancer Res 51
(24) (1991) 6650e6655.
[11] H.P. Kalofonos, M. Rusckowski, D.A. Siebecker, G.B. Sivolapenko, D. Snook,
J.P. Lavender, A.A. Epenetos, D.J. Hnatowich, Imaging of tumor in patients with
indium-111-labeled biotin and streptavidin-conjugated antibodies: pre-
liminary communication, J. Nucl. Med. 31 (11) (1990) 1791e1796.
[12] P.L. Weiden, H.B. Breitz, O. Press, J.W. Appelbaum, J.K. Bryan, S. Gaffigan,
D. Stone, D. Axworthy, D. Fisher, J. Reno, Pretargeted radioimmunotherapy
(PRIT) for treatment of non-Hodgkin’s lymphoma (NHL): initial phase I/II
study results, Biother. Radiopharm. 15 (1) (2000) 15e29.
[36] M. Royzen, G.P.A. Yap, J.M. Fox, A photochemical synthesis of functionalized
trans-cyclooctenes driven by metal complexation, J. Am. Chem. Soc. 130 (12)
(2008) 3760e3761.
€
[37] E.M.F. Billaud, E. Shahbazali, M. Ahamed, F. Cleeren, T. Noel, M. Koole,
A. Verbruggen, V. Hessel, G. Bormans, Micro-flow photosynthesis of new
dienophiles for inverse-electron-demand DielseAlder reactions. Potential
applications for pretargeted in vivo PET imaging, Chem. Sci. 8 (2) (2017)
1251e1258.
[13] G. Liu, S. Dou, Y. Liu, Y. Wang, M. Rusckowski, D.J. Hnatowich, 90Y labeled
phosphorodiamidate morpholino oligomer for pretargeting radiotherapy,
Bioconjugate Chem. 22 (12) (2011) 2539e2545.
[14] K.D. Bagshawe, Antibody-directed enzyme prodrug therapy (ADEPT) for
cancer, Expert Rev. Anticancer Ther. 6 (2006) 1421e1431.
[15] E.M. Sletten, C.R. Bertozzi, From mechanism to mouse: a tale of two bio-
orthogonal reactions, Acc. Chem. Res. 44 (9) (2011) 666e676.
[16] F.C.J. van de Watering, M. Rijpkema, M.S. Robillard, W.J.G. Oyen, O.C. Boerman,
Pretargeted imaging and radioimmunotherapy of cancer using antibodies and
bioorthogonal chemistry, Front. Med. 1 (44) (2014) 1e11.
[17] M.L. Blackman, M. Royzen, J.M. Fox, Tetrazine ligation: fast bioconjugation
based on inverse-electron-demand DielsꢁAlder reactivity, J. Am. Chem. Soc.
130 (41) (2008) 13518e13519.
[18] J. Dommerholt, F.P.J.T. Rutjes, F.L. van Delft, Strain-promoted 1,3-dipolar
cycloaddition of cycloalkynes and organic azides, Top. Curr. Chem. 374 (2)
(2016) 16.
[19] R. Rossin, P.R. Verkerk, S.M. van den Bosch, R.C. Vulders, I. Verel, J. Lub,
M.S. Robillard, In vivo chemistry for pretargeted tumor imaging in live mice,
Angew Chem. Int. Ed. Engl. 49 (19) (2010) 3375e3378.
[20] B.M. Zeglis, P. Mohindra, G.I. Weissmann, V. Divilov, S.A. Hilderbrand,
R. Weissleder, J.S. Lewis, Modular strategy for the construction of radio-
metalated antibodies for positron emission tomography based on inverse
electron demand dielsealder click chemistry, Bioconjugate Chem. 22 (10)
(2011) 2048e2059.
[21] R. Rossin, M.S. Robillard, Pretargeted imaging using bioorthogonal chemistry
in mice, Curr. Opin. Chem. Biol. 21 (2014) 161e169.
[22] R.M. Rahim, R. Kota, J.B. Haun, Enhancing reactivity for bioorthogonal pre-
targeting by unmasking antibody-conjugated trans-cyclooctenes, Bio-
conjugate Chem. 26 (2) (2015) 352e360.
[23] J.P. Meyer, J.L. Houghton, P. Kozlowski, D. Abdel-Atti, T. Reiner, N.V. Pillarsetty,
W.W. Scholz, B.M. Zeglis, J.S. Lewis, 18F-Based pretargeted PET imaging based
on bioorthogonal dielsealder click chemistry, Bioconjugate Chem. 27 (2)
(2016) 298e301.
[24] J.L. Houghton, R. Membreno, D. Abdel-Atti, K.M. Cunanan, S. Carlin,
W.W. Scholz, P.B. Zanzonico, J.S. Lewis, B.M. Zeglis, Establishment of the
in vivo efficacy of pretargeted radioimmunotherapy utilizing inverse electron
[38] B.M. Zeglis, K.K. Sevak, T. Reiner, P. Mohindra, S.D. Carlin, P. Zanzonico,
R. Weissleder, J.S.A. Lewis, Pretargeted PET imaging strategy based on bio-
orthogonal diels-alder click chemistry, J. Nucl. Med. 54 (8) (2013) 1389e1396.
[39] A. Yazdani, H. Bilton, A. Vito, A.R. Genady, S.M. Rathmann, Z. Ahmad,
N. Janzen, S. Czorny, B.M. Zeglis, L.C. Francesconi, J.F. Valliant, A bone-seeking
trans-cyclooctene for pretargeting and bioorthogonal chemistry: a proof of
concept study using 99mTc- and 177Lu-labeled tetrazines, J. Med. Chem. 59
(20) (2016) 9381e9389.
[40] K. Fujiki, Y. Kanayama, S. Yano, N. Sato, T. Yokokita, P. Ahmadi, Y. Watanabe,
H. Haba, K. Tanaka, 211At-labeled immunoconjugate via a one-pot three-
component double click strategy: practical access to a-emission cancer radi-
otherapeutics, Chem. Sci. 10 (7) (2018) 1936e1944.
[41] A. Darko, S. Wallace, O. Dmitrenko, M.M. Machovina, R.A. Mehl, J.W. Chin,
J.M. Fox, Conformationally strained trans-cyclooctene with improved stability
and excellent reactivity in tetrazine ligation, Chem. Sci.
5 (10) (2014)
3770e3776.
ꢀ
[42] E.J.L. Steen, J.T. Jørgensen, K. Johann, K. Nørregaard, B. Sohr, D. Svatunek,
A. Birke, V. Shalgunov, P.E. Edem, R. Rossin, C. Seidl, F. Schmid, M.S. Robillard,
J.L. Kristensen, H. Mikula, M. Barz, A. Kjær, M.M. Herth, Trans-cyclooctene-
functionalized PeptoBrushes with improved reaction kinetics of the tetrazine
ligation for pretargeted nuclear imaging, ACS Nano (2020), https://doi.org/
10.1021/acsnano.9b06905.
[43] S.M. Cheal, M. Patel, G. Yang, D.R. Veach, H. Xu, H.F. Guo, P.B. Zanzonico,
D.B. Axworthy, N.K. Cheung, O. Ouerfelli, S.M. Larson, A N-acetylgalactosa-
mino dendron-clearing agent for high-therapeutic index DOTA-hapten pre-
targeted radioimmunotherapy, Bioconjugate Chem. (2019), https://doi.org/
10.1021/acs.bioconjchem.9b00736.
[44] R. Membreno, B.E. Cook, K. Fung, J.S. Lewis, B.M. Zeglis, Click-mediated pre-
targeted radioimmunotherapy of colorectal carcinoma, Mol. Pharm. 15 (4)
(2018) 1729e1734.
€
[45] R. Membreno, O.M. Keinanen, B.E. Cook, K.M. Tully, K.C. Fung, J.S. Lewis,
B.M. Zeglis, Toward the optimization of click-mediated pretargeted radio-
immunotherapy, Mol. Pharm. 16 (5) (2019) 2259e2263.