Novel antibody-drug conjugate with UV-controlled cleavage mechanism for cytotoxin release

Article abstract of DOI:10.1016/j.bioorg.2020.104475

Antibody-drug conjugates (ADCs) are being developed worldwide with the potential to revolutionize current cancer treatment strategies. However, off-target toxicity caused by the instability of linkers remains one of the main issues to be resolved. Developing a novel photocontrol-ADC with good stability and photocontrolled release seemed to be an attractive and practical solution. In this study, we designed, for the first time, a novel ultraviolet (UV) light-controlled ADC by carefully integrating the UV-cleavable o-nitro-benzyl structure into the linker. Our preliminary work indicated that the ADC exhibited good stability and photocontrollability while maintaining a targeting effect similar to that of the naked antibody. Upon irradiation with UV light, the ADC rapidly released free cytotoxins and exerted significant cytotoxicity toward drug-resistant tumor cells. Compared to those of the unirradiated cells, the EC₅₀ values of ADCs increased by up to 50-fold. Furthermore, our research confirmed that the degradation products of unirradiated ADC, Cys-1a, were relatively less toxic, thus potentially reducing the off-target toxicity caused by nonspecific uptake of ADCs. The novel design strategy of UV light-controlled ADCs may provide new perspectives for future research on ADCs and promote the development of photocontrol systems.

Full text of DOI:10.1016/j.bioorg.2020.104475

Journal Pre-proofs
Novel antibody-drug conjugate with UV-controlled cleavage mechanism for
cytotoxin release
Jiaguo Li, Dian Xiao, Fei Xie, Wei Li, Lei Zhao, Wei Sun, Xiaohong Yang,
Xinbo Zhou
PII:
S0045-2068(20)31773-9
DOI:
https://doi.org/10.1016/j.bioorg.2020.104475
YBIOO 104475
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
25 July 2020
9 November 2020
11 November 2020
Please cite this article as: J. Li, D. Xiao, F. Xie, W. Li, L. Zhao, W. Sun, X. Yang, X. Zhou, Novel antibody-drug
conjugate with UV-controlled cleavage mechanism for cytotoxin release, Bioorganic Chemistry (2020), doi:
https://doi.org/10.1016/j.bioorg.2020.104475
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Inc.

Article
Novel antibody-drug conjugate with UV-controlled
cleavage mechanism for cytotoxin release
Jiaguo Li1, 2, ‡, Dian Xiao2, ‡, Fei Xie , Wei Li , Lei Zhao , Wei Sun *, Xiaohong Yang *, and Xinbo
Zhou *
2
2
2
1,
1,
2
,
1
School of Pharmaceutical Sciences, Jilin University, Changchun 130021, China;
2
National Engineering Research Center for the Emergency Drug, Beijing Institute of Pharmacology and Toxicology,
Beijing 100850, China; li_jiaguo@126.com (J.L.); be_xiaodian@163.com (D.X.); xiefei0058@163.com (F.X.);
a_moon1096@163.com (W.L.); leizhao-hit@hotmail.com (L.Z.);
‡
These authors contribute equally to this work
*
Correspondence: wsun@jlu.edu.cn (W.S.); xiaohongyang88@126.com (X.Y.); zhouxinbo@bmi.ac.cn (X.Z.); Tel.:
+
86-010-6693-0674 (X.Z.)
Received: date; Accepted: date; Published: date
Abstract: Antibody-drug conjugates (ADCs) are being developed worldwide with the potential to
revolutionize current cancer treatment strategies. However, off-target toxicity caused by the instability of
linkers remains one of the main issues to be resolved. Developing a novel photocontrol-ADC with good
stability and photocontrolled release seemed to be an attractive and practical solution. In this study, we
designed, for the first time, a novel ultraviolet (UV) light-controlled ADC by carefully integrating the UV-
cleavable o-nitro-benzyl structure into the linker. Our preliminary work indicated that the ADC exhibited good
stability and photocontrollability while maintaining a targeting effect similar to that of the naked antibody.
Upon irradiation with UV light, the ADC rapidly released free cytotoxins and exerted significant cytotoxicity
toward drug-resistant tumor cells. Compared to those of the unirradiated cells, the EC50 values of ADCs
increased by up to 50-fold. Furthermore, our research confirmed that the degradation products of unirradiated
ADC, Cys-1a, were relatively less toxic, thus potentially reducing the off-target toxicity caused by nonspecific
uptake of ADCs. The novel design strategy of UV light-controlled ADCs may provide new perspectives for
future research on ADCs and promote the development of photocontrol systems.
Keywords: antibody-drug conjugate; photocontrol; ultraviolet radiation; o-nitro-benzyl;
1
. Introduction
In recent years, antibody-drug conjugates (ADCs) have been developed worldwide with the potential to
1
-3
revolutionize current cancer treatment strategies . Structurally, they are formed through conjugation of an
antibody to a highly active cytotoxin via a linker. Currently, nine ADCs have been approved, and more than 80
are at different phases of clinical trials4, 5
.
Essentially, as a prodrug, ADCs are ideally highly stable in the circulatory system and release cytotoxins
in the tumor microenvironment. Therefore, the linker, similar to the trigger of the ADC, is always the key in
6
-8
ADC design . To date, three generations of ADCs have been developed: first-generation ADCs that contain
9
10
acid-labile linkers, such as hydrazone and carbonate bonds, represented by the drugs Mylotarg and Besponsa ;
and second- and third-generation ADCs that predominantly contain enzyme-cleavable linkers, such as
1
1
Adcetris . Although the continuous development of ADCs has effectively improved their therapeutic index to
a certain extent, off-target toxicity caused by the instability of linkers remains one of the main issues to be
resolved. For example, mainstream enzyme-cleavable ADCs mainly utilize cathepsin, which is present in all
1
2
mammalian tissue cells for drug release , but the degradation behaviors of off-target ADC linker structures can
lead to toxic events, thereby limiting their further clinical applications (Figure 1a). A study has shown that more
5
than 23 of the 55 conventional ADCs fail because of poor therapeutic indexes . Therefore, the development of
linker technology with a novel drug release mechanism is urgently needed.
In recent years, photocontrol technology has been widely studied as a noninvasive approach, which allows
control of active substance release with high spatiotemporal precision13-17. The principle behind this technology
is that a photoremovable protecting group (PPG) is linked to an active drug via a covalent bond to mask its
activity. However, the bond can be cleaved by external irradiation to release the active component from the

parent drug. However, certain intrinsic shortcomings of the current photocontrol technology still exist, such as
lack of targeting effects and its proneness to cause unavoidable toxic adverse effects, which are due to
nonspecific systemic distribution of the drug. These factors limit their applications for highly toxic cytotoxic
drugs. To resolve this issue, a combination photocontrolled drug release system with an effective targeted ADC
appears to be an attractive research topic.
Figure 1. Release mechanism of cytotoxin MMAE and UV Light-controlled ADC cell uptake process. (a) The amide bond
between the citrulline residue and the p-aminobenzylcarbamate portion of VC-based ADC was cleaved non-specifically by
cathepsin in lysosomes. Subsequent to undergo a 1,6-elimination process released free MMAE into the tumor cells. PAB:
p-aminobenzyl; (b) photolysis of UV light-controlled ADC upon 365nm UV irradiation with subsequent hydrolysis to
2
produce free MMAE and CO into the tumor cells expressing target antigen. DMNB: 4,5-dimethoxy-2-nitrobenzyl.
Currently, there are only a few reports on photocontrolled ADCs. Nani et al. introduced the structure of
cyanine to the ADC linker and designed two photocontrolled ADCs18, 19 (Figure 2). Upon irradiation with near-
infrared (NIR) light (650–900 nm), the ADCs effectively released the small-molecule cytotoxin, CA-4, and
duocarmycin in the irradiated tumor areas in a site-specific manner. Photocontrolled ADCs based on NIR
exhibit good tissue penetration, but their structures are complex. In addition, they tend to self-aggregate and be
photounstable20, 21, which to a certain extent limits their applications in biological research and their further
development as drugs. Overall, this work has provided us with a reference point to expand our ADC research.

O
NMe2
OMe
OMe
Cl
N
N
O
OMe
O
O
O
OMe
O3HS
O
SO3
N
EtN
Me
Me
Me
Me
Me
Me EtN
Me
Me
Me
N
N
N
N
N
(CH2)4SO3
O3S(H2C)4
(CH2)4SO3
O3S(H2C)4
O
Cy-Pan-CA4
CyEt-Pan-Duo
Figure 2. Structure of two near-infrared light activated ADCs that have been reported.
Herein, we report, for the first time, a type of ADC that utilizes ultraviolet (UV) light to achieve
photocontrolled drug release. In this design, we carefully introduced a UV light-controlled o-nitrobenzyl group
to replace the p-aminobenzyl (PAB) in linkers. In addition, this would help to maximize maintaining the
optimum pharmacological properties of ADCs (Figure 1b). Currently, the o-nitrobenzyl system and its
derivatives are among the most commonly used photosensitive protective groups (PPGs) in biological
research22-25, and some studies have shown that these PPGs have almost no cytotoxicity after photolysis26,27
.
Our novel type of ADC can be delivered to tumor sites by utilizing the targeting function of the antibody. Upon
brief UV irradiation, the o-nitrobenzyl group formed a highly active diradical, followed by dehydrogenation of
the carbon atom in the γ-position25-27, after which the cytotoxic drug was cleaved and released (Figure 1B). 1b),
finally achieving precise temporal and spatial control. In this study, we synthesized an ADC using the highly
active cytotoxic drug monomethyl auristatin E (MMAE) (EC50 = 0.18 nM) and the well-studied antibody mil40,
which is a biosimilar of trastuzumab and is in the clinical research stage (CTR20180362)28-30. Our preliminary
work revealed that our ADC possessed some promising features, such as good stability and targeting effect. In
addition, it rapidly released drugs and effectively destroyed drug-resistant breast and gastric cancer cells only
after irradiation with UV light. Our work presented herein preliminarily validated the feasibility of
photocontrolled ADCs for achieving drug release upon UV light irradiation. The novel design strategy of UV
light-controlled ADCs may provide new perspectives for future research on ADCs and promote the
development of photocontrol systems.
2
. Results and Discussion
2
.1. Chemistry
As shown in Scheme 1, the linker-MMAE portion, i.e., compounds 1a and 1b, was first synthesized, and
4
-Hydroxy-3-methoxyacetophenone and ethyl 4-bromobutyrate were used as starting materials. The
substitution reaction was performed under the catalysis of potassium carbonate at 50°C to obtain compound 3,
followed by a nitration reaction in acetic anhydride and nitric acid to obtain compound 4. Compound 5 was
obtained by a two-step reaction: compound 4 was hydrolyzed in a hot acidic solution of acetate and water, and
4
the ketone of the resulting intermediate was further reduced to a hydroxyl group by NaBH to produce
compound 5. Compound 6 was obtained via a condensation reaction between Boc-protected ethylenediamine
and compound 5. Compound 7 was obtained after removal of the Boc group in an acidic environment. The
naked amino groups were reacted with the respective chain structures that had different maleimide-containing
linkers (8a and 8b) to produce bifunctional linkers 9a and 9b. The bifunctional linkers were reacted with bis(p
nitrophenyl) carbonate on one end to produce active intermediates 10a and 10b, and these were further
conjugated with the toxin MMAE to obtain compounds 1a and 1b.

O
N
OH
OH
O
O
O
O
O
O
O
a
O
b
O
O
c,d
e
H
O
O
OH
N
O
O2
O
O2N
O2N
N
H
O
O
HO
O
3
O
O
O
O
4
5
2
6
NO2
OH
OH
O
O
O
O
O
O
O
g
O
O
h
O
O
f
O
O
O
H
H
N
H
+
N R O N
N
N
N
R
N
R
N
H
NO2
N
NO2
O2N
NH2
O
O
O
O
H
O
O
O
8
8
a
R1=
7
O
b
R2=
N
O
O
10a R1=
9a R1=
R2=
H
O
O
O
1
0b R2=
N
H
O
O
9b
N
H
O
O
H
O
H
N
O
O
N
O
O
O
N
N
NN
N
HN
N
H
N
O
O
O
N
O
O
O
O
NH OH
R
N
O
NO2
O
O
O
H
O
NH
OH
O
MMAE
i
1a R1=
O
1b
R2=
N
H
O
O
O
H
O
O
O
O
N
N
O
H
O
N
N
S
N
N
O
O
O
O
N
NO2
O
H
O
NH
OH
O
O
N
H
O
O
O
O
N
N
O
H
N
O
N
11
+
N R
N
NO2
O
O
O
O
Mil40
j
O
H
O
NH
OH
O
1
a
R1=
R2=
O
O
O
H
1b
O
O
N
N
N
O
O
O
O
O
N
N
H
N
H
O
O
O
O
O
S
N
N
O
N
O
NO2
H
H
O
NH
OH
O
O
12
Scheme 1. Synthesis of ADC 11 and 12. Reagents: (a) anhydrous DMF, K
2
CO
3
, Ethyl 4-bromobuty-rate, 50°C, 6 h; (b)
, EtOH, NaHCO , (e) tert-butyl(2-
HNO , Ac O, 0°C, 5 h, (c) HCl, H O, AcOH, reflux, 1 h, (d) NaBH
3
2
2
4
3
aminoethyl)carbamate, EDCI, HOBT, DIPEA, DMF, r.t., overnight, (f) HCl, r.t., 3 h, (g) DIPEA, anhydrous DMF, r.t.,
overnight, (h) bis(4-nitrophenyl)carbonate, anhydrous DMF, DIPEA, r.t., overnight, (i) anhydrous DMF, HOBT, DIPEA,
2
r.t., 12 h. (j) monoclonal antibody (Mil40), TECP, NAC, H O, r.t.
2
.2. Conjugation to antibody
Compounds 1a and 1b were each conjugated to an antibody to produce ADCs. To expedite our targeted
studies, we selected the well-studied humanized monoclonal antibody mil40 that targets HER2, which is a
marketed biosimilar of Herceptin. The intermolecular disulfide bond of the mil40 antibody was reduced by
TECP to free the sulfhydryl groups, the position of which enabled the linkage of the maleimide-containing
compounds 1a and 1b. The reaction was terminated by NAC. Finally, ADCs 11 and 12 were obtained (Scheme
1
).
2
.3. In vitro characterization
After conjugating the antibody to the novel linker-MMAE described above, we performed quality control
work to characterize ADC 11 and ADC 12. Hydrophobic interaction chromatography (HIC) analysis of the two
ADCs allowed resolution of the conjugates into several major peaks corresponding to 0, 1, 2, 3, 4, 5 and 6 drug
molecules per antibody (Figure 3a), with an average DAR of approximately 3.4 and 2.6 for ADC 11 and 12,
respectively. The proportion of unconjugated naked antibodies was less than 10%, and was 4.1% and 9.7% for
ADC 11 and ADC 12, respectively. At the same time, liquid chromatography-mass spectrometry (UPLC-Q-
TOF-MS) analysis of native ADCs showed that ADCs 11 and 12 contained 2.7 drugs per antibody (Figure 3a).
The DAR value was slightly lower than that detected by HIC, which might be due to the inhibition of ionization
signal by the conjugated linker payloads. According to size exclusion chromatography (SEC) analysis, ADCs
1
1 and 12 were composed nearly exclusively (>98.5% purity) of single monomeric species (Figure 3a).
The stability of new UV-controlled linkers in the circulatory system is a critical factor in determining the
efficacy and toxicity of ADCs. Therefore, we tested herein the stability of the linker-MMAE in human serum.
ADCs 11 and 12 were incubated in 50% human plasma at 37°C, and the amount of released MMAE was
monitored by LC-MS. The standard curve of MMAE revealed that the linear range of MMAE was 0.25 nM to
2
50 nM (Figure 3b). It has been calculated that approximately 3.28 nM and 1.87 nM MMAE were released
from 100 nM of ADCs 11 and 12. In other words, less than 1% of the total MMAE (approximately 340 nM for

ADC 11 and 240 nM for ADC 12) was released during 6 days in human plasma (Figure 3b). This result
suggested that the new UV-controlled linkers were very stable in human plasma.
Under normal conditions, degradation of ADCs in lysosomes will release the structural components of
conjugates containing amino acid residues of the antibody, linkers, and cytotoxins31-33, such as compounds Cys-
1
a and Cys-1b (Figure 1b). To mimic these structures, we reduced the reactive maleimide double bonds of 1a
and 1b with an excess of L-cysteine to obtain Cys-1a and Cys-1b, respectively. These compounds were mixed
with PBS buffer and incubated at 37°C for seven days. Aliquots were collected at different time points prior to
HPLC analysis and then stored at -80°C. The peak areas of the compounds at each time point were plotted
against the sampling time (Figure 3c). The results showed that there were no significant changes in the peak
areas of the two compounds after seven days. This indicated that our linker-MMAE did not break, that no
cytotoxins were released and that the compounds had good stability in PBS.
We next investigated whether the antibody in the ADC retains its original antigen-binding capacity at the
protein and cellular levels. We first performed an enzyme-linked immunosorbent assay (ELISA) on ADCs 11
and 12 to assess their affinities for the HER2 protein. The results showed that the EC50 values of ADC 11 and 12
were 0.015 nM and 0.017 nM, respectively, which were similar to that of naked antibody mil40 (EC50 = 0.007
nM) (Figure 3d). This indicated that there was no discernable effect on the affinity of the antibody toward HER2
after conjugation.
In addition to satisfying the condition of stability in the circulatory system, the linker must be capable of
rapidly and efficiently releasing cytotoxic compounds at the target tissue to achieve the function of destroying
tumor cells. Therefore, we next investigated the photorelease kinetics of compounds 1a and 1b. Studies have
shown that wavelengths shorter than 300 nm may cause damage to tissues and proteins. In contrast, UV
irradiation with wavelengths longer than 320 nm results in less DNA damage and is harmless at low doses34, 35
In line with this concept, we performed our experiments using a UV crosslinker that emitted a wavelength of
65 nm (40 W), which has been shown to be capable of cleaving PPG 4,5-dimethoxy- 2-nitrobenzyl (DMNB).
.
3
We irradiated 40 μM solutions of compounds 1a and 1b in methanol, and the cytotoxin release results are shown
in Figure 3e. Upon irradiation, both compounds showed a rapid release of MMAE that reached the highest
platform within 10 min. In addition, this photocleavage experiment could also serve as a reference for
determining irradiation time in subsequent cell cytotoxicity experiments.
In general, the released small molecule cytotoxins also require sufficient stability at the used wavelength
of UV light. Otherwise, the cytotoxins would be degraded immediately after its release or even before the bond
to the PPG is cleaved. We next evaluated the UV stability of the MMAE. MMAE solutions (1 mg/ml) in
methanol were irradiated at 365 nm (40 W). Under these conditions, MMAE showed excellent stability within
3
0 minutes (Supplementary Figure S1).
2
.4 Cell binding and endocytosis
At the cellular level, in view of the HER2-targeting property of the antibody, we selected BT474 breast
cancer cells that expressed high levels of HER2 for flow cytometry experiments to evaluate the binding of
compounds to the cell-surface HER2 antigen. We found that the mean fluorescence intensity signals in the
experimental group, in which ADCs 11 and 12 were added, were much higher than those in the blank control
cell group and were similar to those obtained with naked antibody mil40. However, in the negative control
MCF-7 cells, which had low HER2 expression, the average fluorescence intensity did not significantly increase
after the addition of naked antibody mil40 and compounds (Figure 4a). These results indicated that ADCs 11
and 12 specifically bound to the HER2 antigen on the surface of BT474 cells. In addition, the results of confocal
microscopy showed that cells incubated with the green fluorescence-stained ADC 12 and naked antibody mil40
at 30 min at 4°C displayed a similar pattern, where they were bound to the HER2 antigen on the surface of the
BT474 cell membrane. After incubation at 37°C for 24 h, scattered staining was observed within the cells, and
the intracellular signals colocalized with that of lysosomes appeared yellow, indicating that ADC 12 was
internalized and transported to the lysosomes (Figure 4b).

Figure 3 In vitro characterization of novel UV-controlled linkers and ADCs. (a) HIC, SEC and UPLC-Q-TOF-MS assays
of ADC 11 and 12. (b) Stability assays of linker-MMAE. ADC 11 and 12 were incubated in 50% human plasma at 37 °C
for seven days and analyzed by LC-MS. (c) Stability assays of linker-MMAE. Compounds Cys-1a and Cys-1b in pH = 7.4
PBS were incubated at 37 °C for seven days and analyzed by HPLC. (d) Affinity tests of ADC for HER2 protein by ELISA.
EC50 value of mil40 was 0.007 nM, EC50 value of ADC 11 and 12 was 0.015 and 0.017 nM. (c) and (e) UV-cleavage tests
on the linker-MMAE. 40 μM compound solutions in methanol were irradiated at 365 nm for 10 min and analyzed by HPLC.

Figure 4. Cell binding and endocytosis of the novel UV-controlled ADCs. (a) The ability of ADC 11 and 12 to bind to the
highly expressed HER2 antigen on BT474 cell membrane was examined by flow cytometry. MCF-7 cells with low
expression of HER2 antigen on cell membrane as negative control. (b) Traffic of positive control antibody mil40 or ADC
1
2 to lysosome in BT474 cells visualized by confocal microscopy. Mil40 or ADC 12 stained by Alexa 48-labeled second
antibody was shown in green, DAPI-stained nuclei was shown in blue, lysosome markers CD107b stained by Alexa Fluor
47-labeled anti-human CD107b antibody was shown in red. Co-localization of signals for mil40 or ADC 12 with lysosome
markers was shown as yellow. Magnification bar, 50 μm.
6

2
.5. Cytotoxicity of ADCs with or without UV irradiation
Figure 5. Her2 expression and Cytotoxicity. (a) Her2 expression was tested in two Herceptin-resistant cell lines. Cells not
stained with FITC was shown in purple peaks, isotype control-FITC was shown in green peaks, anti-Her2 Ab-FITC was
shown in pink peaks. (b) Cytotoxicity of ADCs with or without UV irradiation. Two-way ANOVA analysis, * p < 0.05, **
p < 0.01, *** p < 0.001, **** p < 0.0001.
In view of the good stability and photocleavability of the linker-MMAE payload, as well as the fact that
ADC possessed an antigen-binding capacity similar to that of a naked antibody, we next verified the feasibility
of the design strategy for UV light-controlled ADCs through cell proliferation experiments. Regarding cell line
selection, we considered that the mil40 monoclonal antibody could exhibit the same tumor cell-killing effect as
Herceptin. Therefore, we herein utilized the Herceptin-resistant cell lines BT474-HerDR and N87-HerDR,
which can block the action of this antibody, to objectively assess the true toxicity of the novel linker MMAE
after irradiation. We first verified the expression of HER2 antigen in these two cell lines using flow cytometry.
The results showed that nearly 100% of the cells were HER2-positive. In addition, the fluorescence signals
were intense, suggesting a high abundance of HER2 expression in each cell (Figure 5a).

After incubation with ADCs 11 and 12 for 2 h, the cells were irradiated with UV light at 365 nm (40 W),
and a cell growth assay was performed. The irradiation time of the cells could be selected based on the
photocleavage experiment to allow sufficient release of cytotoxins after UV irradiation; however, the timing in
which the cells could tolerate UV light must also be considered. We conducted relevant experiments and
determined that the longest UV irradiation time tolerated by BT474-HerDR cells was 3 min (Supplementary
Figure S2). First, we found that without UV irradiation, the activity of ADCs declined more than 10 times more
than MMAE. This may indicate that off-target toxicity caused by nonspecific uptake of ADC was significantly
reduced. Furthermore, it was exciting to find that the activities of ADC 11 (EC50 = 0.04 nM) and 12 (EC50
.04 nM) were greatly increased after 3 min of UV irradiation. Compared to those of the unirradiated
compounds, the EC values increased by 50-fold, and the inhibition rates were significantly higher under UV
=
0
5
0
irradiation than those of the unirradiated compounds (p < 0.0001, Figure 5b, Table 1). We also performed the
same tests on N87-HerDR cells and determined that the irradiation time for N87-HerDR cells was 5 min. After
UV irradiation, the EC50 values of ADC 11 and 12 were 0.24 nM and 0.23 nM, respectively, which were
approximately 15-fold higher than those of unirradiated cells (Table 1). Similarly, we found that the inhibition
rate was significantly higher under UV irradiation than without (p < 0.01, Figure 5b, Table 1). Nontargeted
effects of ADC 11 and 12 were assessed on MCF7 breast cancer lines lacking HER2 expression. Both ADCs
tested showed minimal antiproliferative activity in MCF7 cells with EC50 values of 88.64 nM and 45.10 nM,
respectively, whereas free MMAE showed potency equal to that observed on the HER2-positive lines (Table
1
). The cytotoxicity of ADC 11 and 12 on HER2-negative cells was more than 100-fold lower than that on
HER2-positive cells. These results indicated that our ADCs exhibited good UV controllability and antigen
selectivity, the off-target toxicity was greatly reduced without UV irradiation and the tumor cell-killing effects
were rapidly activated after a short UV irradiation.
Table 1. Cell proliferation inhibition activities of two ADCs with or without UV irradiation in two Herceptin-resistant cell
lines and Herceptin-negative cell lines.
BT474-HerDR Cell
BT474-Ratio
(UV-/UV+)
N87-HerDR Cell
N87-Ratio
(UV-/UV+)
MCF 7 Cell
Test compound
EC50 (nM)
EC50 (nM)
EC50 (nM)
—
—
—
—
mil40
7316.59
0.18
9668.35
0.45
>666.67
0.64
MMAE
1
1(UV-)
1(UV+)
2(UV-)
2(UV+)
2.03
3.55
5
5
0.75
14.79
88.64
1
0.04
0.24
1
2.17
3.95
4.25
17.17
45.10
1
0.04
0.23
Radiation by UVP Crosslinker (360nm, 40W). BT474-HerDR cells was irradiated for 3min, N87-HerDR Cells was
irradiated for 5min.
2
.6. Off-target toxicity of ADCs
An interesting observation was that unirradiated ADCs 11 and 12 exhibited less cytotoxicities than
MMAE. In fact, when ADCs were not exposed to UV light, they might be endocytosed (Figure 4) and degraded
by the lysosomes into Cys-1a and Cys-1b (Figure 6a). In our latest research, we detected Cys-linker-MMAE as
3
6
one of the degradation products from ADCs in cells by LC-MS/MS . Therefore, we speculated that the two
degradation products may be less toxic. The advantage of the low toxicity of the degradation products was the
reduction of the toxic adverse effects of the ADCs because of early degradation in the circulation or nonspecific
uptake. Thus, we first tested the cell proliferation inhibition activities of Cys-1a and Cys-1b. The results
indicated that compared with MMAE in the three cells, the cytotoxicity of Cys-1a and Cys-1b obviously
decreased, with a maximum reduction of approximately 47 times (Figure 6b). It was exciting to find that the
unirradiated Cys-1a and Cys-1b exhibited less cytotoxicity to N87-HerDR cells and BT474-HerDR cells, with

EC50 values similar to those of ADC 11 and ADC 12. This outcome primarily achieved our design goal of
reducing off-target toxicity by a low-toxicity prodrug form, Cys-linker-MMAE (Figure 6b).
Moreover, the lipid-water partition coefficient, MlogP, of Cys-1a and Cys-1b predicted by ADMET
predictor 8.0 software was -2.624 and -4.336, respectively (Table 2). Compared with MMAE (1.911), a
remarkable reduction was observed. Other permeability parameters such as S + MDCK exhibited the same
downward trend (Table 2). These experimental results indicated that the decline in the permeability of Cys-
linker-MMAE contributed to the reduction of cytotoxicity and a lower bystander effect.
Next, tubulin polymerization assays were performed on these two compounds to analyze further the reason
underlying the reduced toxicity. Both Cys-1a and Cys-1b exhibited a concentration-dependent inhibition of
tubulin polymerization (Figure 6c), and the calculated EC50 was 6.28 μM and 7.17 μM, respectively. Compared
with that of MMAE, their inhibitory effects were reduced by 6- and 7-fold, respectively (Table 2), confirming
that the compounds exhibited lower toxicities than MMAE. Comprehensive results of tubulin polymerization
assays and cytotoxic experiments indicated that the compound Cys-1a could be used as a good lead compound
to provide a structural basis for subsequent optimization to further reduce the toxicity of degradation products.
Figure 6. Off-target toxicity of ADCs. (a) ADCs were degraded into Cys-1a and Cys-1b without UV. (b) Effect of Cys-1a
and Cys-1b on tubulin polymerization in vitro. Purified tubulin protein at 2 mg/mL in a reaction buffer was incubated at 37
°
C in the presence of 0.1% DMSO, test compounds at 1.25, 2.5, 5, or 10 μM. Polymerizations were followed by an increase
in fluorescence emission at 460 nm over a 60 min period at 37 °C.
Table 2. Inhibitory activity of Cys-1a and Cys-1b on Tubulin Polymerization
Test
Inhibition of tubulin
polymerization
EC50 (μM)
BT474-HerDR Cell
N87-HerDR Cell
compound
MlogP a
S+MDCK b
EC50 (nM)
EC50 (nM)
MMAE
1.191
21.818
0.18
0.45
1.11

Cys-1a
Cys-1b
-2.624
-4.336
4.34
8.53
1.59
16.54
5.62
6.28
7.17
4.303
a
b
MlogP: Oil-water partition coefficient of compounds. Higher value indicates a better fat solubility. S+MDCK: Used to
predict the apparent permeability coefficient of compounds in Madin-Darby canine kidney cell models. Higher values
indicate greater membrane permeability.
2
.7. In vivo imaging of fluorescein-labeled ADC
The in vivo distribution and tumor targeting ability of Mil40 and ADC 11 were evaluated in a BALB/c
nude mouse NCI-N87 xenograft model by an optical imaging system. We used DyLight 680 NHS ester-labeled
Mil40 and ADC 11 to administer the tail vein of mice at a dose of 20 mg/kg, and the intense fluorescence signal
in the tumor site was clearly visualized within 6 h. Subsequently, the tumor-located image was clearly
maintained for approximately 15 days, and eventually the fluorescence signal disappeared on the 26th day
(Figure 7a and 7c). The experimental results demonstrated that our novel UV light-controlled ADC 11 had a
long half-life similar to that of the monoclonal antibody Mil40. This indicated that the novel UV light-controlled
ADC maintained the excellent drug-like properties of the marketed ADCs, such as excellent pharmacokinetic
properties and significant tumor targeting effect in vivo. In addition, the fluorescence intensity in the liver was
higher, while no significant fluorescence accumulation was observed in other normal tissues and organs (Figure
7
b). As for the fluorescence in the liver, it was supposed to be the result that ADC 11 needed to be metabolized
here. However, at the same time, it is worth noting that hepatotoxicity might be one of the main considerations
in further studies. Although our design of Cys-linker-MMAE might reduce toxicity to some extent, the real
safety of UV-controlled ADCs has yet to be verified by additional in vitro and in vivo experiments. After all, it
was a difficult lesson that the first ADC Mylotarg was once withdrawn from the market because of serious
hepatotoxicity.

Figure 7. In vivo imaging of fluorescein-labeled ADC 11. (a) In vivo fluorescence imaging of DyLight 680 NHS Ester-
labeled Mil40 and ADC 11 in NCI-N87 BALB/c nude mice. ADC 11 has similar tumor targeting ability and fluorescence
quenching time as the Mil40. (b) Higher fluorescence accumulation in tumor as compared to other nomal organs support
tumor selectivity of ADC 11. (c) Quantitative analysis of fluorescent signals. ADC 11 and Mil40 have similar fluorescence
in mice, and the fluorescence signal gradually decreases over time.
3
. Materials and Methods
3
.1. Chemistry
Reagents and solvents were purchased from Innochem, Energy Chemical, Aladdin, Adamas-beta, TCI,
Ark Pharm, Acros and Alfa Aesar and were used without additional purification. Mil40 is a biosimilar of
trastuzumab, which was provided by Zhejiang Haizheng Pharmaceutical Co., Ltd. Reactions were followed by
thin-layer chromatography (Dexin Biotechnology Co., Ltd, Yantai, China). Visualization was accomplished
with 254 nm UV light. Nuclear magnetic resonance (NMR) spectra were obtained on a JNM-ECA-400.
Chemical shifts are reported in ppm, and TMS is used as the internal standard. Coupling constants J are given
in Hertz. Spin multiplicities are reported as the following abbreviations: s (singulet), d (doublet), dd (doublet
doublet), t (triplet), q (quadruplet), and m (multiplet). Mass spectrometry (MS) was performed on a triple
quadrupole mass spectrometer using electrospray ionization, API 3000 (AB Sciex, Concord, ON, Canada).
3
.1.1. Synthesis of ethyl 4-(4-acetyl-2-methoxyphenoxy) butanoate (3)
A mixture of 1-(4-hydroxy-3-methoxyphenyl)ethan-1-one (15 g, 90.27 mmol) and ethyl 4-bromobu-
tanoate (21.13 g, 108.33 mmol) in anhydrous DMF (200 mL) and K CO (24.95 g, 180.52 mmol) was added.
2 3

Then, the mixture was heated to 50℃ for 6 h. After the reaction was cooled to room temperature, the mixture
was filtered, and the filtrate was partitioned between water (500 mL) and ethyl acetate (500 mL × 3). The
combined organic layers were washed with brine, dried over anhydrous sodium sulfate, and concentrated to
give a crude product. The crude product was purified by column chromatography (petroleum ether/ethyl acetate
1
=
5:1) to give ethyl 4-(4-acetyl-2-methoxyphenoxy) butanoate (23.2 g, yield 92%) as a white solid; H NMR
(400 MHz, DMSO-d6): δ 7.60 (dd, J = 8.4, 2.0 Hz, 1H), 7.43 (d, J = 2.0 Hz, 1H), 7.05 (d, J = 8.4 Hz, 1H), 4.09
–
4.04 (m, 4H), 3.81 (s, 3H), 2.52 (s, 3H), 2.46 (t, J = 7.3 Hz, 2H), 2.02 – 1.95 (m, 2H), 1.18 (t, J = 7.1 Hz, 3H).
+
+
ESI m/z (M + H) calculated for C15
21 5
H O 281.13; found 281.14.
3
.1.2. Synthesis of ethyl 4-(4-acetyl-2-methoxy-5-nitrophenoxy) butanoate (4)
A mixture of 70% HNO
3
(100 mL) and acetic anhydride (20 mL) was cooled to 0℃. A suspension of 3
(5.0 g, 17.8 mmol) in acetic anhydride (15 mL) was added dropwise, and the mixture was stirred for 5 h at 0℃.
After the reaction was completed, the solution was poured into ice water and cooled in the refrigerator for 2 h.
The resulting solid was collected by filtration, washed with water (40 mL × 2) to remove excess acid, and dried
to afford ethyl 4-(4-acetyl-2-methoxy-5-nitrophenoxy) butanoate (4.44 g, yield 77%) as a light yellow solid;
1
3
H NMR (400 MHz, DMSO-d6): δ 7.63 (s, 1H), 7.23 (s, 1H), 4.13 (t, J = 6.4 Hz, 2H), 4.07 (q, J = 7.1 Hz, 2H),
.93 (s, 3H), 2.51 (d, J = 2.4 Hz, 3H), 2.46 (t, J = 7.3 Hz, 2H), 2.03 – 1.96 (m, 2H), 1.18 (t, J = 7.1 Hz, 3H).
+
+
7
ESI m/z (M + H) calculated for C15H20NO 326.12; found 326.12.
3
.1.3. Synthesis of 4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy) butanoic acid (5)
Compound 4 (4.3 g, 13.2 mmol) was added to a hot acidic solution of acetic acid (20 mL) and water (60
mL), followed by the addition of concentrated hydrochloric acid (5 mL). The mixture was stirred at reflux for
h. The solution was cooled to room temperature when it became completely transparent. The precipitate was
1
filtered and collected and washed with water (200 mL × 2) to afford the crude product used directly in the next
reaction. The crude product was dispersed in an aqueous solution (150 mL), and sodium hydrogencarbonate
(2.45 g, 29.2 mmol) was added in batches. Ethanol (9 mL) was added dropwise to remove the foam. Then,
within one hour, sodium borohydride was added, and the pH was maintained in the range of 9-11 by the addition
of sodium bicarbonate. After 2 h, the reaction was acidified to pH=2 with 8 N HCl. The mixture was filtered,
washed with water (200 mL × 2), and dried to afford 4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy)
butanoic acid (3.21 g, yield 81%) as a yellow solid; 1H NMR (400 MHz, DMSO-d6): δ 12.19 (s, 1H), 7.53 (s,
1
H), 7.35 (s, 1H), 5.49 (d, J = 4.4 Hz, 1H), 5.28 – 5.22 (m, 1H), 4.05 (t, J = 6.5 Hz, 2H), 3.90 (s, 3H), 2.38 (t,
-
-
J = 7.3 Hz, 2H), 1.98 – 1.91 (m, 2H), 1.36 (d, J = 6.2 Hz, 3H). ESI m/z (M - H) calculated for C13H16NO
7
2
98.09; found 298.10.
3
.1.4. Synthesis of tert-butyl (2-(4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy)butaname-do) ethyl)
carbamate (6)
To a solution of 5 (1.7 g, 5.68 mmol) in anhydrous DMF (30 mL), HOBt (920 mg, 6.82 mmol), EDCI
1.63 g, 8.53 mmol), and DIPEA (1.1 g, 8.53 mmol) were added sequentially to the reaction solution. Then,
(
tert-butyl (2-aminoethyl) carbamate (1.09 g, 6.82 mmol) was added, and the mixture was stirred at room
temperature overnight. After the reaction was completed, the mixture was treated with water (100 mL) and
extracted with ethyl acetate (50 mL × 3). The combined organic layers were washed with brine, dried over
anhydrous sodium sulfate, and concentrated to give a crude product, which was purified by column
chromatography
(ethyl
acetate)
to
give
tert-butyl
(2-(4-(4-(1-hydroxyethyl)-2-methoxy-5-
nitrophenoxy)butanamedo) ethyl) carbamate (1.54 g, yield 61%) as a light yellow solid; 1H NMR (400 MHz,
DMSO-d6) δ 7.88 (t, J = 5.4 Hz, 1H), 7.52 (s, 1H), 7.35 (s, 1H), 6.80 (t, J = 5.5 Hz, 1H), 5.49 (d, J = 4.4 Hz,
1
1
H), 5.28 – 5.22 (m, 1H), 4.03 (t, J = 6.4 Hz, 2H), 3.90 (s, 3H), 3.06 (dd, J = 12.1, 5.9 Hz, 2H), 2.96 (dd, J =
+
2.1, 6.0 Hz, 2H), 2.22 (t, J = 7.4 Hz, 2H), 1.97 – 1.91 (m, 2H), 1.36 (t, J = 2.9 Hz, 12H). ESI m/z (M + H)
+
calculated for C20
H
32
N
3
O
8
442.22; found 442.21.
3
.1.5. Synthesis of N-(2-aminoethyl)-4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy) butan-amide (7)

To a suspension of 6 (1.5 g, 3.4 mmol) in methanol (20 mL) was added a solution of HCl (6 N, 20 mL).
The reaction mixture was stirred at room temperature for 5 h, concentrated, and dried under vacuum to afford
a crude product of N-(2-aminoethyl)-4-(4-(1-hydroxyethyl)-2-methoxy-5- nitrophenoxy)-butanamide as a
yellow liquid, which was used directly in the next reaction.
3
.1.6. Synthesis of 6-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)-N-(2-(4-(4-(1-hydroxyethyl)-2-methoxy- 5-
nitrophenoxy)butanamido)ethyl) hexanamide (9a)
To a suspension of 7 (700 mg, 2.05 mmol) in anhydrous DMF (20 mL) was added compound 8a (694.8
mg, 2.26 mmol). DIPEA (317.9 mg, 2.46 mmol) was added to the reaction solution, and then the mixture was
stirred at room temperature overnight and evaporated under reduced pressure to give the crude product. The
crude product was purified by column chromatography (dichloromethane/methanol = 50:1 to 10:1) to give 9a
(
(
(
955 mg, yield 87%) as a light yellow solid; 1H NMR (400 MHz, DMSO-d6): δ 7.89 (s, 1H), 7.80 (s, 1H), 7.52
s, 1H), 7.35 (s, 1H), 7.00 (s, 2H), 5.49 (d, J = 4.4 Hz, 1H), 5.28 – 5.22 (m, 1H), 4.02 (t, J = 6.4 Hz, 2H), 3.90
s, 3H), 3.37 (d, J = 7.1 Hz, 2H), 3.07 – 3.06 (m, 4H), 2.22 (t, J = 7.4 Hz, 2H), 2.00 (t, J = 7.4 Hz, 2H), 1.96 -
+
1
.91 (m, 2H), 1.49 – 1.42 (m, 4H), 1.36 (d, J = 6.2 Hz, 3H), 1.19 – 1.12 (m, 2H). ESI m/z (M + H) calculated
+
for C25
H
35
N
4
O
9
535.24; found 535.22.
3
.1.7.
Synthesis
of N-(16-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)-4,14-dioxo-7,10-dioxa-3,13-di-
azahexadecyl)-4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy) butanamide (9b)
Compound 9b was synthesized by employing the experimental protocols described for 9a. The yield was
5
1
3
4
1% as a light yellow solid. 1H NMR (400 MHz, DMSO-d6) δ 8.04 (t, J = 5.6 Hz, 1H), 7.90 (s, 2H), 7.52 (s,
H), 7.35 (s, 1H), 7.00 (s, 2H), 5.49 (d, J = 4.4 Hz, 1H), 5.28 – 5.22 (m, 1H), 4.03 (t, J = 6.4 Hz, 2H), 3.90 (s,
H), 3.60 – 3.56 (m, 4H), 3.46 (d, J = 2.8 Hz, 4H), 3.40 – 3.37 (m, 2H), 3.15 – 3.11 (m, 2H), 3.08 – 3.07 (m,
+
H), 2.34 – 2.27 (m, 4H), 2.25 – 2.21 (m, 2H), 1.98 – 1.93 (m, 2H), 1.36 (d, J = 6.2 Hz, 3H). ESI m/z (M + H)
+
calculated for C29
H
42
N
5
O
12 652.28; found 652.29.
3
.1.8. Synthesis of 1-(4-(4-((2-(6-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)hexanamido)ethyl)-amino)- 4-
oxobutoxy)-5-methoxy-2-nitrophenyl)ethyl (4-nitrophenyl) carbonate (10a)
Bis(4-nitrophenyl) carbonate (1.08 g, 3.56 mmol) was added to a suspension of compound 9a (950 mg,
1
.78 mmol) and DIPEA (690 mg, 5.34 mmol) in anhydrous DMF (20 mL). Then, the mixture was stirred at
room temperature overnight and evaporated under reduced pressure to give the crude product, which was
purified by column chromatography (dichloromethane/methanol = 50:1 to 10:1) to give 10a (522.5 mg, yield
4
7
3
1
2%) as a light yellow solid; 1H NMR (400 MHz, DMSO-d6) δ 8.31 – 8.27 (m, 2H), 7.90 (s, 1H), 7.80 (s, 1H),
.59 (s, 1H), 7.55 – 7.51 (m, 2H), 7.18 (s, 1H), 7.00 (s, 2H), 6.27 (q, J = 6.5 Hz, 1H), 4.06 (t, J = 6.4 Hz, 2H),
.96 (s, 3H), 3.36 (d, J = 7.1 Hz, 2H), 3.07 – 3.06 (m, 4H), 2.23 (t, J = 7.4 Hz, 2H), 2.00 (t, J = 7.3 Hz, 2H),
+
.97 –1.92 (m, 2H), 1.72 (d, J = 6.5 Hz, 3H), 1.49 – 1.41 (m, 4H), 1.19 – 1.11 (m, 2H). ESI m/z (M + H)
+
calculated for C32
H
38
N
5
O
13 700.24; found 700.25.
3
.1.9. Synthesis of 1-(4-((1-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)-3,13,18-trioxo-7,10-dioxa-4,14,17-
triazahenicosan-21-yl)oxy)-5-methoxy-2-nitrophenyl)ethyl (4-nitrophenyl) carbonate (10b)
Compound 10b was synthesized by employing the experimental protocols described for 10a. The yield
was 44% as a light yellow solid; 1H NMR (400 MHz, DMSO-d6) δ 8.29 (d, J = 9.1 Hz, 2H), 8.04 (t, J = 5.6
Hz, 1H), 7.90 (s, 2H), 7.59 (s, 1H), 7.53 (d, J = 9.0 Hz, 2H), 7.19 (s, 1H), 7.00 (s, 2H), 6.27 (dd, J = 12.7, 6.1
Hz, 1H), 4.07 (t, J = 6.3 Hz, 2H), 3.96 (s, 3H), 3.60 – 3.56 (m, 4H), 3.49 (d, J = 5.0 Hz, 2H), 3.36 – 3.33 (m,
4
2
H), 3.13 (dd, J = 11.3, 5.7 Hz, 2H), 3.08 (s, 4H), 2.34 – 2.27 (m, 4H), 2.23 (t, J = 7.5 Hz, 2H), 1.99 – 1.94 (m,
+
+
H), 1.72 (d, J = 6.4 Hz, 3H). ESI m/z (M + H) calculated for C38
H
45
N
6
O
16 817.28; found 817.29.
3
.1.10. Synthesis of compound 1a
To a solution of 10a (85.76 mg, 0.12 mmol) in anhydrous DMF (15 mL), HOBT (15.06 mg, 0.11 mmol),
DIPEA (28.43 mg, 0.22 mmol), and MMAE (80 mg, 0.11 mmol) were added sequentially, and then the mixture

was stirred at room temperature for 12 h. After the reaction was completed, the mixture was evaporated under
reduced pressure to give the crude product, which was purified by column chromatography
(dichloromethane/methanol = 70:1 to 15:1) to give 1a (87 mg, yield 62%) as a yellow solid; 1H NMR (400
MHz, DMSO-d6) δ 7.87 (d, J = 12.9 Hz, 2H), 7.76 (s, 1H), 7.61 (d, J = 8.4 Hz, 1H), 7.52 – 7.44 (m, 2H), 7.26
–
7.21 (m, 4H), 7.14 (t, J = 7.0 Hz, 1H), 6.96 (s, 2H), 6.11 – 6.06 (m, 1H), 5.39 (d, J = 4.6 Hz, 1H), 5.32 (d, J
=
4.7 Hz, 1H), 4.77 – 4.48 (m, 2H), 4.47 – 4.31 (m, 2H), 4.04 – 3.81 (m, 8H), 3.54 –3.43 (m, 2H), 3.37 – 3.32
(m, 2H), 3.24 – 3.17 (m, 5H), 3.15 – 3.08 (m, 4H), 3.02 (s, 5H), 2.98 – 2.89 (m, 3H), 2.73 (t, J = 17.3 Hz, 2H),
2
3
.19 (t, J = 6.5 Hz, 3H), 2.08 (s, 2H), 1.97 (t, J = 7.4 Hz, 3H), 1.91 (s, 3H), 1.80 – 1.61 (m, 4H), 1.58 – 1.52 (m,
+
H), 1.46 – 1.37 (m, 5H), 1.16 – 1.09 (m, 2H), 1.02 – 0.92 (m, 6H), 0.87 – 0.67 (m, 17H). ESI m/z (M + H)
+
calculated for C H N O 1278.7232; found 1278.7231.
6
5
100
9
17
3
.1.11. Synthesis of compound 1b
Compound 1b was synthesized by employing the experimental protocols described for 1a. The yield was
6% as a light yellow solid; 1H NMR (400 MHz, DMSO-d6) δ 8.03 (s, 1H), 7.91 (d, J = 12.3 Hz, 3H), 7.65 (d,
6
J = 8.1 Hz, 1H), 7.61 – 7.47 (m, 2H), 7.32 – 7.25 (m, 4H), 7.17 (t, J = 6.7 Hz, 1H), 7.00 (s, 2H), 6.20 – 6.08
(m, 1H), 5.44 (s, 1H), 5.36 (s, 1H), 4.81 – 4.58 (m, 2H), 4.50 – 4.36 (m, 2H), 4.07 – 3.84 (m, 8H), 3.63 – 3.52
(m, 6H), 3.46 (s, 5H), 3.26 – 3.20 (m, 5H), 3.19 – 3.12 (m, 6H), 3.08 (s, 5H), 3.02 – 2.93 (m, 3H), 2.76 (t, J =
1
0
1
7.1 Hz, 2H), 2.36 – 2.25 (m, 5H), 2.22 (s, 2H), 2.16 – 1.89 (m, 6H), 1.78 (s, 4H), 1.64 – 1.43 (m, 5H), 1.06 –
+
+
.96 (m, 6H), 0.93 – 0.71 (m, 17H). ESI m/z (M + H) calculated for C69
H
107
N
10
O
20 1395.7658; found
395.7637.
3
.2. Bioconjugation and Purification
Antibody-conjugated drugs 11 and 12 were synthesized according to the following method.
3
.2.1. Antibody reduction
Humanized anti-HER2 IgG1 antibody mil40 (5 mg/mL) was dissolved in pH 7.5, 20 mM L-histidine
buffer. Reductant tris(2-carboxyethyl) phosphine hydrochloride (TECP; 2.3 equivalents) was added, and then
the mixture was stirred at 25℃ for 2 h.
3
.2.2. Antibody conjugation
To a solution of reduced mil40 was added maleimide-containing compound 1a or 1b (8 equivalents) in ice-
cold dimethylacetamide (DMAC) (5% v/v). The mixture was stirred at 25°C for 50 min.
3
.2.3. Termination reaction
The reaction solution was added excess N-acetyl-L-cysteine (10 equivalents) to quench. After stirring for
0 min, the mixture was acidified to pH=5.5 with 0.3M AcOH.
3
3
.2.4. Purification and analysis
The mixture was allowed to be buffer exchanged using AKTA pure 25 (GE Healthcare, USA) and
Sephadex G-25 Medium buffer exchange columns to afford conjugates 11 or 12. The conjugates were
concentrated by ultrafiltration centrifugation and sterile filtered through a 0.2 μm filter under sterile conditions.
Then, the samples were stored at -80°C before use for analysis and testing. The drug-to-antibody ratio (DAR)
was determined by using HIC HPLC (Agilent 1260 HPLC; Wilmington, DE, USA); a UV-visible
spectrophotometer (UV-2450, SHIMADZU, Janpan) was used to determine the concentration of conjugates.
3
.3. Linker-MMAE Stability Assays In Vitro
ADCs 11 and 12 (100 nM) were incubated with 50% human plasma. The reaction mixture was incubated
at 37 ℃ in a constant temperature incubator. After 48, 96, 120, and 144 h, aliquot samples (50 μL) were taken,

and methanol (250 μL) was added to the samples. The solution was centrifuged, and the released MMAE was
analyzed by LC-MS.
Compounds 1a and 1b were dissolved in DMSO (1 mg/ml), followed by the addition of L-cysteine PBS
solution (0.095 mg/ml, pH 7.4). The final DMSO concentration was controlled at 10%, and the mixture was
stirred at room temperature for 10 min, after which HPLC analysis revealed complete conversion to Cys-1a and
Cys-1b. The reaction mixture was incubated at 37 ℃ in a constant temperature incubator. After 0, 5, 10, 15,
and 30 min and 1, 2, 4, 8, 12, 24, 48, 72, and 168 h, aliquot samples (300 μL) were taken and frozen at -20 ℃.
All samples were melted in sequence, and analysis was performed by HPLC.
3
.4. UV-cleavage tests on the linker-MMAE In Vitro
Compounds were dissolved in methyl alcohol (40 μM), and the solutions were irradiated at 365 nm by
using an ultraviolet crosslinker (CL-1000 L UVP Crosslinker, 5×8 W, Analytikjena, USA). Every minute from
0
to 10 min, aliquot samples (400 μL) were taken and analyzed by HPLC.
3
.5. ADC Affinity Assay
The HER2 affinity of ADC was measured by ELISA. In brief, HER2 antigen (Her2 protein, Human,
Recombinant, Sino Biological Inc., Beijing, China) was diluted to a final concentration of 1 μg/mL with coating
solution and plated on 96-well plates (100 μL for each well) at 4 ℃ overnight in a wet box. The coating solution
was removed, and the remaining protein-binding sites were blocked by incubation with 10% fetal bovine serum
for 2 h at 37°C. Reactions were initiated by adding gradient concentrations of ADC. After reacting for 1 h, the
plate was washed with PBS (containing 0.2% Tween) three times. Then, 100 μL of goat anti-human IgG Fc
cross-absorbed secondary antibody conjugated to HRP (Thermo Fisher Scientific, USA) was added to each well
and incubated for 45 min at room temperature. Then, the plate was washed three times with PBS (containing
0
.2% Tween). TMB was added to develop color (100 μL for each well) and protected from light for 1 min.
Finally, 2 M H SO was used to stop the reaction, and the absorbance of each well was read at 450 nm with a
multifunction microplate reader.
2
4
3
.6. Flow Cytometry Assay
MCF-7 and BT474 cells obtained from ATCC (Manassas, VA, USA) were resuspended in 200 μl PBS
(
pH 7.4, containing 2% fetal bovine serum) and incubated with 5 µg/ml mil40 or ADC at a density of 5 x 105
cells/ml for 30 min at 4°C. After being washed twice with PBS, the cells were further incubated with 10 μg/ml
PE-labeled anti-human IgG Fc (BioLegend, USA) for 30 min at 4°C. Then, the cells were washed twice with
PBS. The fluorescence intensity of PE was measured using a NovoCyte flow cytometer (ACEA Bioscience,
Hangzhou, China). Data were analyzed by using FlowJo software. For BT474-HerDR cells and N87-HerDR
cells provided by Yicon (Beijing) Medical Science And Technology Co., Ltd., fluorescence intensity
measurements can be performed only after incubation with FITC anti-human HER2 antibody (BioLegend,
USA). FITC mouse IgG was used as an isotype control (BioLegend, USA).
3
.7. Fluorescence microscopy
4
BT474 cells (5×10 /dish) were plated in new confocal dishes, cultured in RPMI-1640 medium with 10%
fetal bovine serum at 37°C and 5% CO2, and then incubated with 1 ml (10 μg/ml) ADC or positive control
antibody mil40 for 30 min on ice or 24 h at 37°C. Lysosomal protease inhibitors (25 nM) were added to all the
dishes. The cells were washed with prechilled PBS, fixed and permeabilized with 0.5% BSA and 0.5% Triton
X-100 PBS solution at 37°C for 5 min. After washing with prechilled PBS, ADC or mil40, the lysosomal
compartment was detected using 3 μg/mL Alexa Fluor 488-labeled goat anti-human IgG (BioLegend, USA)
and 3 μg/mL Alexa Fluor 647-labeled anti-human CD107b antibody (Invitrogen, Carlsbad, California, USA),
respectively. Next, the nuclear compartment was stained with 10 μg/ml DAPI at 4°C for 30 min. The cells were
washed three times with prechilled PBS, and fluorescence images were taken using a confocal microscope
(Leica, TCS SP8).
3
.8. Cell Proliferation Assay

To assess the cytotoxicity of test compounds on 3 cell lines in the presence or absence of ultraviolet
radiation, cells were seeded in 96-well plates (2500 per well) and allowed to adhere overnight at 37°C in a
humidified atmosphere of 5% CO . The medium was then removed from the plates, and the cells were treated
2
in their recommended growth media containing varying concentrations of test compounds. When the cells were
incubated with the drug for 2 h, the cell culture plates were taken and irradiated to the corresponding time with
an ultraviolet crosslinker (this step is only for compounds that required UV irradiation). After 96 hours of
incubation, CellTiter 96® AQueous One Solution (Promega, Madison, Wisconsin, USA) was added to the
plates and incubated for 3 hours at 37°C. EC50 values were calculated using GraphPad Prism 5.0.
3
.9. Tubulin Polymerization Inhibition Assay
A tubulin polymerization assay kit (BK011P, Cytoskeleton, USA) was used to detect tubulin
polymerization. All experimental steps followed the operating manual. First, 5 μL of inhibitors at various
concentrations were incubated in 96-well plates at 37°C for 1 min. Then, 50 μL of the tubulin reaction solution
containing 10 mg/ml porcine brain tubulin (>99% pure), 0.5 mM EGTA, 2 mM MgCl , 60% glycerol and 100
2
mM GTP was added. The increase in fluorescence was measured immediately by excitation at 360 nm and
emission at 420 nm in a multimode plate reader (PerkinElmer, USA).
3
.10. In vivo fluorescence imaging experiment
In vivo fluorescence imaging experiments were implemented using the NCI-N87 BALB/c nude mouse
xenograft model. Mil40 and ADC 11 were labeled with DyLight 680 NHS ester according to the manufacturer’s
operating manual (DyLight Amine-Reactive Dyes, Thermo Fisher Scientific). NCI-N87 cells were injected in
the right dorsum of the mice subcutaneously. When the tumor volumes reached approximately 300-400 mm3,
fluorescently labeled compounds were injected via the tail veins at a dose of 20 mg/mg in each mouse.
Isoflurane-anesthetized mice were placed in the optical imaging chamber of an IVIS SPECTRUM mouse live
imager (PerkinElmer, USA) at corresponding time points for in vivo distribution observation. Tumors and
organs were dissected 30 hours after the injection to observe organ uptake. The images were analyzed by Living
Image software.
4
. Disscusion
Although several admirable works considering photocontrolled ADCs using near infrared light (650 - 900
nm) have been reported, we developed novel UV-light (360 nm) control ADCs for the following points. First,
compared with the complex structure of near infrared light-controlled ADCs, UV light-controlled ADCs have
a more reasonable and simple structure and retain the structure of the marketed ADCs. Only a minor structural
modification has been made by replacing PAB with the commonly used PPG group o-nitrobenzyl. Therefore,
from the point of chemistry and pharmacology, our novel ADCs could maintain the excellent drug-like
properties of the marketed ADCs, such as good plasma stability (t1/2 > 6 days), long half-life in vivo (nearly the
same as that of Mil40) and significant tumor targeting effect in vivo. In addition, research on the toxicity and
membrane permeability of Cys-1a/1b suggested that the off-target toxicity might be lower for our ADCs. Off-
target toxicity was one of the main obstacles to the development of ADC, and the first ADC Mylotarg was
withdrawn once from the market because of serious off-target toxicity. However, we also admit that more in
vitro and in vivo experiments are needed to verify our assumption. The Cys-linker-MMAE degradation product
3
6
was detected in cells in our recent study .
Second, concerning the clinical application of UV light-controlled ADCs, we must admit that there are
still some difficulties at present. Although UVA irradiation (wavelengths above 320 nm) causes less direct DNA
damage than UVB or UVC exposure, a high dosage of UVA is toxic and can lead to oxidative stress, photoaging
and immunosuppression37, 38. Moreover, UV blue light cannot penetrate patient skin to reach deeply into the
tumor area37, 38. Only depths of approximately 100 μm can be reached37, 38. There are several possible solutions
in future clinical applications: the required light could be transmitted to the site of interest via optical fibers or
2
5
endoscopic probes. Moreover, in selected cases, adjacent areas could be illuminated during the surgery . At
the same time, it should be noted that some works have been done on the application of UV-light-based therapy.
3
9
For instance, psoralen plus UV-A (PUVA) therapy has been applied in dermatology since the 1970s .
Moreover, UV-light controlled small molecule prodrugs are also continuously developed, such as

2
5
photoactivatable prodrugs of the antimelanoma agent vemurafenib and photoactivatable caged prodrugs of
4
0
VEGFR-2 kinase inhibitors . We think that these novel UV light-controlled ADCs are a good complement to
the application of UV light for medical/therapeutic purposes.
5
. Conclusions
This article reports, for the first time, the use of UV light to control drug release by ADCs. The feasibility
of such a design was also preliminarily validated. Throughout the study, our ADCs exhibited good stability and
photocontrollability while maintaining a targeting effect similar to that of a naked antibody. Brief UV light
irradiation resulted in rapid release of small-molecule cytotoxins and effective killing of Herceptin-resistant
N87-HerDR and BT474-HerDR cells. Compared with those of unirradiated ADCs, the inhibitory activities of
irradiated ADCs increased by 15- and 50-fold in N87-HerDR and BT474-HerDR cells, respectively.
Furthermore, the degradation product of unirradiated ADCs, Cys-1a, exhibited lower toxicity and had the
potential to reduce the toxic adverse effects caused by nonspecific uptake of ADCs. The potential applications
of this novel type of UV light-controlled ADC include the use of fiber optic or endoscopic probes to transmit
the required light to the site of action or by irradiating the lesion during surgery. ADCs can also be used as cell
tool drugs. However, there are some limitations in this study that require further investigation. The first is the
optimization of the linker structure. Compound 1a can serve as a good lead compound, and its structure can be
modified to further reduce its toxicity. Second, the binding of ADCs and cells was only studied under ideal
conditions at 2 h, which was useful to elucidate the basic properties of UV-controlled ADCs. However, the
situation after ADCs enter the cells is more complex, and more work is needed in the future to thoroughly
understand this process. Overall, the novel design strategy for ADCs proposed in this study provides a new tool
for future mechanistic research on ADCs and new perspectives on the development of novel types of ADCs for
therapeutic applications.
Author Contributions: conceptualization, D.X.; methodology, J.L.; software, L.Z.; validation, W.S., X.Y. and X.Z.; formal
analysis, J.L. and F.X.; investigation, D.X. and W.L.; resources, X.Z.; data curation, J.L. and W.L.; writing—original draft
preparation, J.L.; writing—review and editing, W.S., X.Y. and X.Z.; supervision, X.Y. and X.Z.; project administration,
X.Z.; funding acquisition, X.Z.”
Funding: This research was funded by the National Science and Technology Major Project for Major New Drugs Innovation
and Development under grant no. 2018ZX09711003-009 and Chinese National Natural Science Foundation under grant no.
8
1872736.
Acknowledgments: The authors gratefully acknowledge research support from Dan Jiang laboratory for the structural
confirmation.
Conflicts of Interest: The authors declare no conflict of interest.
References
1
.
Beck, A.; Haeuw, J. F.; Wurch, T.; Goetsch, L.; Bailly, C.; Corvaïa, N., The next generation of antibody-
drug conjugates comes of age. Discovery Medicine 2010, 10 (53), 329-39.
Leal, M.; Sapra, P.; Hurvitz, S. A.; Senter, P.; Wahl, A.; Schutten, M.; Shah, D. K.; Haddish-
2
.
Berhane, N.; Kabbarah, O., Antibody-drug conjugates: an emerging modality for the treatment of cancer. Annals
of the New York Academy of Sciences 2015, 1321 (1), 41-54.
3
6
4
.
Sievers, E. L.; Senter, P. D., Antibody-Drug Conjugates in Cancer Therapy. Annual Review of Medicine 2013,
4 (3), 15-29.
Lyon, R., Drawing lessons from the clinical development of antibody-drug conjugates. Drug Discovery
.
Today: Technologies 2018, 30, 105-109.
Coats, S.; Williams, M.; Kebble, B.; Dixit, R.; Tseng, L.; Yao, N. S.; Tice, D. A.; Soria, J. C., Antibody-
5
.
Drug Conjugates: Future Directions in Clinical and Translational Strategies to Improve the Therapeutic Index.
Clinical cancer research : an official journal of the American Association for Cancer Research 2019.
6
3
7
.
Jain, N.; Smith, S.; Ghone, S.; Tomczuk, B., Current ADC Linker Chemistry. Pharm Res 2015, 32 (11), 3526-
540.
.
Lu, J.; Feng, J.; Aiping, L.; Ge, Z., Linkers Having a Crucial Role in Antibody–Drug Conjugates.
International Journal of Molecular Sciences 2016, 17 (4), 561-.

8
.
Tsuchikama, K.; An, Z., Antibody-drug conjugates: recent advances in conjugation and linker chemistries.
Protein & Cell 2018, 9 (1), 33-46.
Sorokin, P., Mylotarg approved for patients with CD33+ acute myeloid leukemia. Clinical journal of oncology
nursing 2000, 4 (6), 279-80.
0. Ricart; D., A., Antibody-Drug Conjugates of Calicheamicin Derivative: Gemtuzumab Ozogamicin and
Inotuzumab Ozogamicin. Clinical Cancer Research 2011, 17 (20), 6417-6427.
1. Deng, C.; Pan, B.; O'Connor, O. A., Brentuximab vedotin. Clinical Cancer Research An Official Journal of the
American Association for Cancer Research 2013.
2. Fitzgerald, D. J.; Wayne, A. S.; Kreitman, R. J.; Pastan, I., Treatment of hematologic malignancies with
immunotoxins and antibody-drug conjugates. Cancer Research 2011, 71 (20), 6300-6309.
3. None, Photolabile protecting groups and linkers. Journal of the Chemical Society Perkin Transactions 2002, 2
2), 125-142.
4. Fodor, S.; Read, J.; Pirrung, M.; Stryer, L.; Lu, A.; Solas, D., Light-directed, spatially addressable
parallel chemical synthesis. Science 1991, 251 (4995), 767-773.
5. Mcgall, G. H.; Barone, A. D.; Diggelmann, M.; Fodor, S. P. A.; Gentalen, E.; Ngo, N., The Efficiency of
Light-Directed Synthesis of DNA Arrays on Glass Substrates. Journal of the American Chemical Society 1997, 119
22), 5081-5090.
6. Guillier, F.; Orain, D.; Bradley, M., Linkers and cleavage strategies in solid-phase organic synthesis and
combinatorial chemistry. (Chem. Rev. 2000, 100, 2091. Published on the web may 6, 2000). Chemical Reviews 2000,
9
.
1
1
1
1
(
1
1
(
1
1
00 (6), 2091.
1
7. Pelliccioli, A. P.; Wirz, J., Photoremovable protecting groups: reaction mechanisms and applications.
Photochemical & Photobiological Sciences Official Journal of the European Photochemistry Association & the European
Society for Photobiology 2002, 1 (7), 441-458.
1
8. Rogerr., N.; Alexander P., G.; Tadanobu, N.; Tsuyoshi, Y.; Joseph, I.; Hisataka, K.; Martin J., S., In
Vivo Activation of Duocarmycin–AntibodyConjugates by Near-Infrared Light. Acs Central Science 2017, 3 (4),
3
1
29.
9. Nani, R. R.; Gorka, A. P.; Nagaya, T.; Kobayashi, H.; Schnermann, M. J., Near-IR Light-Mediated
Cleavage of Antibody-Drug Conjugates Using Cyanine Photocages. Angewandte Chemie International Edition
2
015, 127 (46), 13635.
2
0. Xu, Y.; Li, Z.; Malkovskiy, A.; Sun, S.; Pang, Y., Aggregation Control of Squaraines and Their Use as
Near-Infrared Fluorescent Sensors for Protein. Journal of Physical Chemistry B 2010, 114 (25), 8574-8580.
1. Mcrae, E. G.; Kasha, M., Enhancement of Phosphorescence Ability upon Aggregation of Dye Molecules.
Journal of Chemical Physics 1958, 28 (4), 721-722.
2. Kaplan, J. H.; Forbush, B.; Hoffman, J. F., Rapid photolytic release of adenosine 5'-triphosphate from a
protected analog: utilization by the sodium:potassium pump of human red blood cell ghosts. Biochemistry 1978,
2
2
1
7 (10), 1929-1935.
2
3. Engels, J.; Schlaeger, E. J., Synthesis, structure, and reactivity of adenosine cyclic 3′,5′-phosphate benzyl
triesters. Journal of Medicinal Chemistry 1977, 20 (7), 907-911.
2
2
2
4. Pinchuk, B.; Horbert, R.; Döbber, A.; Kuhl, L.; Peifer, C., Photoactivatable Caged Prodrugs of VEGFR-
Kinase Inhibitors. Molecules 2016, 21 (5), 570.
5. Horbert, R.; Pinchuk, B.; Davies, P.; Alessi, D.; Peifer, C., Photoactivatable Prodrugs of Antimelanoma
Agent Vemurafenib. Acs Chem Biol 2015, 10 (9), 2099-107.

2
6. Shin, W. S.; Han, J.; Kumar, R.; Lee, G. G.; Sessler, J. L.; Kim, J.-H.; Kim, J. S., Programmed activation
of cancer cell apoptosis: A tumor-targeted phototherapeutic topoisomerase I inhibitor. Scientific Reports 2016, 6,
2
2
9018.
7. Ibsen, S.; Zahavy, E.; Wrasdilo, W.; Berns, M.; Chan, M.; Esener, S., A Novel Doxorubicin Prodrug
with Controllable Photolysis Activation for Cancer Chemotherapy. Pharmaceutical Research 2010, 27 (9), 1848-
1
2
860.
8. Li, H.; Yu, C.; Jiang, J.; Huang, C.; Yao, X.; Xu, Q.; Yu, F.; Lou, L.; Fang, J., An anti-HER2 antibody
conjugated with monomethyl auristatin E is highly effective in HER2-positive human gastric cancer. Cancer
Biology & Therapy 2016, 17 (4), 346-354.
2
9. Jiang, J.; Dong, L.; Wang, L.; Wang, L.; Zhang, J.; Chen, F.; Zhang, X.; Huang, M.; Li, S.; Ma, W.,
HER2-targeted antibody drug conjugates for ovarian cancer therapy. European Journal of Pharmaceutical Sciences
2
016, 93, 274-286.
3
0. Yao, X.; Jiang, J.; Wang, X.; Huang, C.; Li, D.; Xie, K.; Xu, Q.; Li, H.; Li, Z.; Lou, L., A novel
humanized anti-HER2 antibody conjugated with MMAE exerts potent anti-tumor activity. Breast Cancer Research
Treatment 2015, 153 (1), 123-133.
1. Erickson; K., H., Antibody-Maytansinoid Conjugates Are Activated in Targeted Cancer Cells by Lysosomal
Degradation and Linker-Dependent Intracellular Processing. Cancer Research 2006, 66 (8), 4426-4433.
2. Lewis Phillips, G. D.; Li, G.; Dugger, D. L.; Crocker, L. M.; Parsons, K. L.; Mai, E.; Blattler, W. A.;
&
3
3
Lambert, J. M.; Chari, R. V.; Lutz, R. J.; Wong, W. L.; Jacobson, F. S.; Koeppen, H.; Schwall, R. H.;
Kenkare-Mitra, S. R.; Spencer, S. D.; Sliwkowski, M. X., Targeting HER2-positive breast cancer with
trastuzumab-DM1, an antibody-cytotoxic drug conjugate. Cancer Res 2008, 68 (22), 9280-90.
3
3. Doronina, S. O.; Toki, B. E.; Torgov, M. Y.; Mendelsohn, B. A.; Cerveny, C. G.; Chace, D. F.;
DeBlanc, R. L.; Gearing, R. P.; Bovee, T. D.; Siegall, C. B.; Francisco, J. A.; Wahl, A. F.; Meyer, D. L.;
Senter, P. D., Development of potent monoclonal antibody auristatin conjugates for cancer therapy. Nature
biotechnology 2003, 21 (7), 778-84.
3
4. Matsumura, Y.; Ananthaswamy, H. N., Toxic effects of ultraviolet radiation on the skin. Toxicol Appl
Pharmacol 2004, 195 (3), 298-308.
5. Bliman, D.; Nilsson, J. R.; Kettunen, P.; Andréasson, J.; Gr?tli, M., A Caged Ret Kinase Inhibitor and its
Effect on Motoneuron Development in Zebrafish Embryos. Scientific Reports 2015, 5 (1), 13109.
6. Wang, Y. M.; Liu, L. Q.; Fan, S. Y.; Xiao, D.; Xie, F.; Li, W.; Zhong, W.; Zhou, X. B., Antibody-Drug
Conjugate Using Ionized Cys-Linker-MMAE as the Potent Payload Shows Optimal Therapeutic Safety. Cancers
3
3
2
3
2
3
1
3
020, 12 (3).
7. Matsumura, Y.; Ananthaswamy, H. N., Toxic effects of ultraviolet radiation on the skin. Toxicol Appl Pharm
004, 195 (3), 298-308.
8. D'Orazio, J.; Jarrett, S.; Amaro-Ortiz, A.; Scott, T., UV Radiation and the Skin. Int J Mol Sci 2013, 14 (6),
2222-12248.
9. Morison, W. L.; Parrish, J. A.; Fitzpatrick, T. B., Controlled study of PUVA and adjunctive topical therapy
in the management of psoriasis. Br J Dermatol 1978, 98 (2), 125-32.
4
2
0. Pinchuk, B.; Horbert, R.; Dobber, A.; Kuhl, L.; Peifer, C., Photoactivatable Caged Prodrugs of VEGFR-
Kinase Inhibitors. Molecules 2016, 21 (5).





Antibody-drug conjugates are hopeful to revolutionize current cancer treatment strategies, but still
restrained by off-target toxicity.
Developing novel photocontrol-ADC to release highly active cytotoxin precisely in the tumor site
upon UV irradiation seemed an attractive subject, but still challenging.
The first ultraviolet (UV) light-controlled ADCs were designed by tactfully replacing p-aminobenzyl
with UV-cleavable o-nitro-benzyl, exhibiting good photo-controllability and lower off-target toxicity.
The present strategy may provide a new idea for future designing of ADCs and promote the
development of photocontrol systems.

Declaration of interests
☒
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.
☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests: