Switching on prodrugs using radiotherapy

Article abstract of DOI:10.1038/s41557-021-00711-4

Chemotherapy is a powerful tool in the armoury against cancer, but it is fraught with problems due to its global systemic toxicity. Here we report the proof of concept of a chemistry-based strategy, whereby gamma/X-ray irradiation mediates the activation of a cancer prodrug, thereby enabling simultaneous chemo-radiotherapy with radiotherapy locally activating a prodrug. In an initial demonstration, we show the activation of a fluorescent probe using this approach. Expanding on this, we show how sulfonyl azide- and phenyl azide-caged prodrugs of pazopanib and doxorubicin can be liberated using clinically relevant doses of ionizing radiation. This strategy is different to conventional chemo-radiotherapy radiation, where chemo-sensitization of the cancer takes place so that subsequent radiotherapy is more effective. This approach could enable site-directed chemotherapy, rather than systemic chemotherapy, with ‘real time’ drug decaging at the tumour site. As such, it opens up a new era in targeted and directed chemotherapy. [Figure not available: see fulltext.].

Full text of DOI:10.1038/s41557-021-00711-4

Articles
https://doi.org/10.1038/s41557-021-00711-4
Switching on prodrugs using radiotherapy
Jin Gengꢀ ꢀ1
,2,7
ꢀ✉
, Yichuan Zhangꢀ ꢀ , Quan Gao , Kevin Neumannꢀ ꢀ , Hua Dong , Hamish Porter ,
1,2,7
2
1
1
3
4
5,6
3
1
ꢀ✉
Mark Potterꢀ ꢀ , Hua Ren , David Argyle and Mark Bradleyꢀ ꢀ
Chemotherapy is a powerful tool in the armoury against cancer, but it is fraught with problems due to its global systemic toxic-
ity. Here we report the proof of concept of a chemistry-based strategy, whereby gamma/X-ray irradiation mediates the activa-
tion of a cancer prodrug, thereby enabling simultaneous chemo-radiotherapy with radiotherapy locally activating a prodrug. In
an initial demonstration, we show the activation of a fluorescent probe using this approach. Expanding on this, we show how
sulfonyl azide- and phenyl azide-caged prodrugs of pazopanib and doxorubicin can be liberated using clinically relevant doses
of ionizing radiation. This strategy is different to conventional chemo-radiotherapy radiation, where chemo-sensitization of the
cancer takes place so that subsequent radiotherapy is more effective. This approach could enable site-directed chemotherapy,
rather than systemic chemotherapy, with ‘real time’ drug decaging at the tumour site. As such, it opens up a new era in targeted
and directed chemotherapy.
any efforts have been made to improve the therapeu- including double-strand breaks2 , which in turn leads to reac-
2–28
tic index of anticancer agents. Chief among these is the tion with intercellular oxygen and, ultimately, cell death. This may
M
prodrug approach, which has a proven clinical track explain why hypoxic tissues/cancers are radiation-resistant—in
record of improving administration routes, enhancing selectivities the absence of oxygen, the DNA-derived free radicals are simply
1
29
and reducing systemic toxicities . The major therapeutically rele- removed by glutathione . Chemically it has been reported that
30
vant routes to prodrug activation are typically via enzymes such as gamma/X-rays can induce a variety of transformations , so revers-
capecitabine, an orally available form of 5-fluorouracil that under- ible addition-fragment chain-transfer polymerization can be initiated
31
goes three enzymatic reactions (carboxylesterase, Cyd deaminase with gamma radiation . In this Article, we seek to develop a robust
2
and dThdPase in liver or/and tumours ), or physiological param- strategy to activate cancer prodrugs by the application of clinically
eters such as pH (for example in the use of aldoxorubicin (INNO- relevant doses of gamma/X-ray irradiation (Fig. 1). This strategy
2
06), an albumin-binding prodrug that releases doxorubicin under is totally different to conventional chemo-radiotherapy radiation,
3–6
acidic conditions ). There are a number of prodrugs in the pipeline where chemo-sensitization of the cancer takes place so that subse-
and in clinical trials. For instance, Apadaz releases its parent drug quent radiotherapy is more effective, and would allow site-directed
7
(
hydrocodone) by enzymatic cleavage in the gastrointestinal tract . chemotherapy rather than systemic chemotherapy, allowing ‘real
Polyethylene glycol (PEG) ‘protected or blocked’ interleukin-2 (IL- time’ drug decaging at the tumour site in vitro and in vivo, and open-
2
) is a biologic prodrug with IL-2 liberated by in vivo release of ing up a new era in targeted and directed chemotherapy.
8
the PEG chains . Other, more academic prodrug stimuli have been
explored in a variety of models, including electrochemical activa- Results and discussion
9
10
tion of metal-based prodrugs , ultrasound-induced release or Chemical reactions under X-ray irradiation. A high-throughput
11
light-mediated activation .
screen was initially applied to identify chemically relevant functional
Concurrent chemotherapy and radiation has been shown to give
12
13
much better overall survival for the treatment of rectal , lung and
X-ray
14
Drug/probe
Drug/probe
breast cancers than either alone. The combination of radiotherapy
and hypoxia-activated prodrugs such as evofosfamide and SN30000
Mask
15–17
(
an analogue of tirapazamine) has also been reported . However,
OH
OH
O
O
OH
OH
O
O
the combined toxicities of the dual treatments are often prohibitive,
necessitating a reduction in the intensity of either or both treatment
modalities and a reduction in efficacy.
H
N
N
N
N
N
N
H2N
O
O
O
O
Linear accelerator technology for the precise delivery of radio-
therapy allows the treatment of a range of tumour types, including
breast, lung, head and neck, prostate, gastrointestinal and gynae-
O
S
O
Doxorubicin 15
Coumarin 9
NH2
Pazopanib 10
OH
NH2
18,19
cological cancers . Ionizing radiation triggers a series of reac-
tions in cells, generating a variety of reactive species such as free Fig. 1 | Drug/probe decaging via X-ray irradiation. Coumarin is used as
•–
•
electrons and reactive oxygen species (including O and OH) a probe model, and pazopanib and doxorubicin are used as anticancer
2
20,21
in the tumour area . This gives rise to severe DNA damage, agents.
1
2
EaStCHEM School of Chemistry, University of Edinburgh, Edinburgh, UK. Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences,
3
4
Shenzhen, China. The Royal (Dick) School of Veterinary Studies and Roslin Institute, University of Edinburgh, Edinburgh, UK. Department of Surgery,
Western General Hospital, Edinburgh, UK. Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/
Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Department of Radiation Oncology, National
5
6
Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital and Shen Zhen Hospital, Chinese Academy of Medical Sciences and Peking
7
✉
Union Medical College, Shenzhen, China. These authors contributed equally: Jin Geng, Yichuan Zhang. e-mail: jin.geng@siat.ac.cn; mark.bradley@ed.ac.uk
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
05

Articles
NATurE CHEmIsTry
Table 1 | Model reactions under X-ray irradiation
Compound
Reactiona
Number of productsb
Compound
Reactiona
Number of productsb
OH
H2N
NH2
N
-
N
-
O
OH
Y
>2
Y
>5
N
N
N
N
N
N
N
N
F3C
N
-
Y
3
P
O
HO
N3
Y
1
-
Y
1
-
OH
N3
O
O
O
N3
N
N
N
O
OH
OH
N3
I
N
N
N
Y
-
1
N
N
-
-
N3
HO
HO
F
F
F
O
N3
F
F
HO
OH
F
O
N
-
Y
2
HN
O
OH
O
S
O
N3
a
b
All reactions were carried out in PBS with concentration of 10 μM and under 60 Gy irradiation. N, not reacted; Y, reacted. Confirmed by HPLC chromatography (254 nm).
groups that are modified by gamma/X-ray irradiation, includ- of <10% (Fig. 2e). However, excitingly, 4-(hydroxymethyl)-2,3,
ing azides, disulfides, azobenzenes, triphenylphosphine, tetrazoles 5,6-tetrafluoroaryl azide 4 (10µM in PBS) was cleanly reduced to
and so on, with all reactions performed in an oxygen-free environ- 4-(hydroxymethyl)-2,3,5,6-tetrafluoroaniline 5 under X-ray radia-
ment in aqueous solution to mimic a hypoxic tumour environment tion (60Gy) (Fig. 2c). Mechanistically, it has been reported that
(
Table 1). We observed that molecules such as dipyridyl tetra- sulfonyl azides can undergo photolysis in alcohols and water via
zine and azobenzene did react under X-ray radiation (60Gy), radical-mediated chemistry to generate sulfonamides either via a
but the resulting reaction mixtures were too complex to be useful sulfonylamido radical or a nitrene. We suggest a similar mechanism
(
Supplementary Information). However, we also observed that a here, in which the irradiation source generates hydroxy radicals that
number of functional groups were converted cleanly and with high react with the sulfonyl azide to generate the sulfonylamido radical
efficiency, with 4-acetamidobenzenesulfonyl azide and some aro- (and nitrogen), which then abstracts a hydrogen radical from the
matic azides reacting under irradiation to give simple reaction mix- surrounding environment. An analogous mechanism would oper-
3
2–34
tures that could be fully characterized.
ate similarly for the arylazide conversion to the aniline (Fig. 2)
.
As seen in Fig. 2, 4-acetamidobenzenesulfonyl azide 1 (10µM
in PBS) gave rise to 4-acetamidobenzenesulfonamide 2 in high ActivationofafluorescenceprobeviaX-rayirradiation. Following
conversion with traces of sulfonic acid 3 (Fig. 2a). A model reac- these observations, we performed an initial proof of concept involv-
tion of 9-azidobenzoic acid 6 (10µM in PBS) under X-ray irradia- ing X-ray irradiation unmasking of 7-azido-4-methylcoumarin 8
tion afforded 4-aminobenzoic acid 7, although with a conversion (Fig. 3) to give 7-amino-4-methylcoumarin 9, allowing fluorescence
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
06

NATurE CHEmIsTry
Articles
a
NHAc
NHAc
NHAc
b
1
after irradiation
2
X-ray
H
+
H
OH
3
O
S
O
O
S
O
O
S
O
OH + N2
N3
NH
NH2
1
1
1
2
X-ray
OH + HN3
NHAc
NHAc
0
1
2
3
4
5
6
Retention time (min)
d
4
after irradiation
O
S
O
O
S
O
4
OH
3
5
c
HO
HO
HO
4
4
F
F
F
F
F
F
F
F
F
F
F
F
X-ray
H
+
H
OH
OH + N2
N3
NH
NH2
0
1
2
3
4
5
6
4
5
Retention time (min)
f
6
after irradiation
6
e HO
O
HO
O
HO
O
7
X-ray
H
6
6
+
H
OH
OH + N2
N3
NH
NH2
6
7
0
1
2
3
4
5
6
Retention time (min)
Fig. 2 | Model reactions of 1, 4 and 6 under X-ray irradiation. a, Mechanism of the reaction of 4-acetamidobenzenesulfonyl azide 1 under X-ray irradiation.
−
1
b, HPLC traces of 1 upon irradiation (20ꢀµM, 60ꢀGy, 6ꢀGyꢀmin ) to afford 2 (retention time R ꢀ=ꢀ2.61ꢀmin) and 3 (R ꢀ=ꢀ2.83ꢀmin), with full conversion of 1
T
T
T
(
R ꢀ=ꢀ4.89ꢀmin). c, Mechanism of the reaction of 4-(hydroxymethyl)-2,3,5,6-tetrafluoroaryl azide 4 under X-ray irradiation to result in the generation of
−
1
4
-(hydroxymethyl)-2,3,5,6-tetrafluoroaniline 5. d, HPLC traces of 4 (R ꢀ=ꢀ4.31ꢀmin) upon irradiation (20ꢀµM, 60ꢀGy, 6ꢀGyꢀmin ) to afford 5 (R ꢀ=ꢀ3.01ꢀmin).
T
T
e, Mechanism of the reaction of 4-azidobenzoic acid 6 under X-ray irradiation. f, HPLC traces of 6 (R ꢀ=ꢀ3.99ꢀmin) and irradiated 6 (20ꢀµM, 60ꢀGy,
T
−1
6
ꢀGyꢀmin ) that afforded poor conversion to 4-aminobenzoic acid 7 (R ꢀ=ꢀ1.79ꢀmin). HPLC chromatograms show UV absorbance at 254ꢀnm.
T
3
7
assessment of activation (λ
=345/443nm, Fig. 1b). Conversion azide hydrogen sulfate as the diazotransfer reagent (Fig. 4) and its
ex/em
of azide 8 (100µM in PBS) by irradiation was >90%, with a lin- reaction with X-ray irradiation examined. Reaction took place in
ear relationship with irradiation dose (Fig. 3b,c). When HeLa cells a radiation-dose-dependent manner at ambient temperature, yield-
were incubated with 7-azido-4-methylcoumarin 8 (100µM) for 4h ing the drug pazopanib 10 (and by-product 12; Fig. 4). The reac-
and irradiation applied, flow cytometry (Fig. 3d) also showed an
increase in fluorescent intensity with increased irradiation dose.
a
b
9
8
6
4
3
0 Gy
8 Gy
6 Gy
Activation of a pazopanib prodrug. Pazopanib (trade name,
Votrient) is an orally available, small-molecule tyrosine kinase
inhibitor of vascular endothelial growth factor (VEGF) receptors 1,
X-ray
H2N
24 Gy
12 Gy
6 Gy
3
5,36
2
and 3 . Here, we use pazopanib as one of the models for prodrug
N3
O
O
O
O
activation through X-ray irradiation. The sulfonyl azide prodrug
1 was generated from pazopanib 10 using imidazole-1-sulfonyl
8
9
Non-fluorescent
Fluorescent
0
Gy
1
0
2
4
6
8
10
Retention time (min)
Fig. 3 | X-ray irradiation-mediated activation of azido coumarin.
a, Irradiation of 7-azido-4-methylcoumarin 8 results in the generation of
c
1
d
00
100
80
0
6
Gy
Gy
7-amino-4-methylcoumarin 9 with fluorescence turn-on. b, HPLC traces
9
0
of the reaction mixture of 8 (R ꢀ=ꢀ4.57ꢀmin) after irradiation with 0ꢀGy to
36 Gy
T
6
0 Gy
6
0ꢀGy to afforded product 9 (R ꢀ=ꢀ3.21ꢀmin). c, Graph showing the yield
80
60
T
of 9 generated with different irradiation doses. The data are presented
as meanꢀ±ꢀs.d. (nꢀ=ꢀ3). d, Flow cytometry analysis (λ ꢀ=ꢀ360/450ꢀnm)
7
6
5
0
0
0
4
2
0
0
0
ex/em
of cells incubated with 8 (100ꢀµM), with 9 generated in cells following
irradiation of 0ꢀGy, 6ꢀGy, 36ꢀGy and 60ꢀGy. Forward versus side scatter
profiles were used to gate intact cellular materials. Flow cytometry
analyses are based on a population of 10,000 cells.
0
10 20 30 40 50 60
Irradiation dose (Gy)
0
1
2
3
4
5
1
0
10 10 10 10 10
Coumarin
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
07

Articles
NATurE CHEmIsTry
a
c
O
HSO4
N3
S
10 12
11
H
N
H
N
O
NH
60 Gy
8 Gy
36 Gy
N
N
N
N
N
N
N
N
N
4
N
N
iPrOH/H2O
K2CO3
O
S
O
O
S
O
Pazopanib (10)
11
NH2
N3
2
1
6
4 Gy
2 Gy
Gy
b
H
H
N
0 Gy
N
N
N
N
N
N
X-ray
N
N
10
+
N
N
N
3
4
5
6
7
O
S
O
O
S
O
1
1
12
Inactive
Retention time (min)
N3
Inactive
Active
OH
d
e 120
6
4
2
0
5
µM
0 µM
20 µM
0 µM
1
5 µM
8
4
0
0
0
0
0
0
10 µM
0
6
12
24
0
10
20
30
40
50
60
Irradiation dose (Gy)
Irradiation dose (Gy)
f
50
g
PBS
100
PBS
4
3
2
1
0
0
0
0
0
Pazopanib 10 (P = 0.326)
Prodrug 11 (P = 0.627)
PBS + X-ray (P = 0.797)
Pazopanib 10 + X-ray
8
6
0
0
Pazopanib 10
Prodrug 11
PBS + X-ray
Pazopanib 10
+ X-ray
40
(
P = 0.100)
Prodrug 11 + X-ray
P = 0.184)
2
0
0
Prodrug 11 + X-ray
(
1
5
9
13 17 21 25 29
Days
0
10
20
30
40
Days
Fig. 4 | Synthesis of the pazopanib prodrug 11 and analysis of the irradiation reaction in vivo. a, Synthesis of prodrug 11 from pazopanib (10, 83ꢀmM) in
H O/i-PrOH with imidazole-1-sulfonyl azide (1.5ꢀequiv.) and K CO (4ꢀequiv.) (18ꢀh, yield 65%). b, The reaction of prodrug 11 under irradiation afforded
2
2
3
products 10 (pazobanib) and 12. c, HPLC traces of the reaction mixture of prodrug 11 (R ꢀ=ꢀ5.44ꢀmin) after irradiation (0ꢀGy to 60ꢀGy) to afford product
T
1
0 (R ꢀ=ꢀ4.23ꢀmin) and 12 (R ꢀ=ꢀ4.83ꢀmin), with some remaining unreacted prodrug 11. d, The yield of pazopanib 10 generated from prodrug 11 (5, 10
−1
T
T
and 20ꢀµM) under irradiation from 6ꢀGy up to 60ꢀGy (6ꢀGyꢀmin ). The data are presented as meanꢀ±ꢀs.d. (nꢀ=ꢀ3). e, HUVEC viability upon treatment with
prodrug 11 (0, 5 and 10ꢀµM) and X-ray irradiation (0, 6, 12 and 24ꢀGy). The data are presented as meanꢀ±ꢀs.d. (nꢀ=ꢀ3). f, HT29 tumour-bearing BALB/c nude
mice were treated with prodrug 11 by intratumoral injection. At 4ꢀh post injection, mice were treated with or without 6ꢀGy of X-ray irradiation. Tumour
burdens were measured every other day using a caliper. The data are presented as meanꢀ±ꢀs.d. (nꢀ=ꢀ5). Statistical analysis was performed using one-way
analysis of variance (ANOVA) with Dunnett post-test compared to PBS-treated mice at day 23. g, Survival curves for HT29 tumour-bearing mice. Mice
3
were aged until moribund or the tumour volume reached 2,000ꢀmm (nꢀ=ꢀ5).
tion of prodrug 11 (5µM, 10µM and 20µM in PBS) reached over
As desired, HUVEC cells treated with prodrug 11 and X-ray
9
0% conversion following X-ray irradiation (60Gy). Interestingly irradiation behaved in a similar manner to cells treated with pazo-
we observed that the amount of pazopanib 10 increased when panib 10, that is, showing lower morbidities, reduced metastasis and
irradiation doses increased, but reached its highest concentration tubule formation (Supplementary Figs. 28–30). Successful activa-
at 24Gy (Fig. 4), whereupon 11 started to decompose, generat- tion of the caged prodrug molecules was confirmed in an animal
ing the sulfonic acid by-product 12 (Supplementary Figs. 21 and model (Fig. 4f). Mice treated with the combination of prodrug 11
2
3). Importantly, pazopanib 10 (20µM in PBS) remained unreac- and X-ray irradiation (purple line, Fig. 4f) showed comparable effi-
tive under radiation (60Gy) (Supplementary Fig. 25). To evaluate cacy as the drug, with prolonged survival times compared to control
the efficacy of the combination of radiotherapy and activation of groups treated with drug or irradiation individually (Fig. 4g).
the prodrug, human umbilical vein endothelial cells (HUVEC)
were incubated with pazopanib 10 or prodrug 11 (between 5µM to Activation of a caged doxorubicin prodrug. Based on the model
38
2
0µM) and irradiated with different doses (from 0Gy to 60Gy) . reaction shown in Fig. 2c, where 4 was neatly reduced to a single
We observed that cell viability was dependent on both irradiation product, 5, we synthesized (p-azido)-2,3,5,6-tetrafluorobenzyl
dose and drug concentration. Activation of prodrug 11 under irra- oxycarbonyl-substituted doxorubicin 13 as an additional prodrug
3
9
diation generated pazopanib 10, giving high levels of cytotoxicity model with another mode of activation, examining the reduc-
due to the effect of irradiation and the liberated drug (Fig. 4e).
tion of phenyl azide to aniline and subsequent decaging of the
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
08

NATurE CHEmIsTry
Articles
OH
OH
OH
OH
OH
OH
a
HO
O
OH
O
O
O
OH
O
O
O
OH
OH
O
O
F
F
F
F
X-ray
Decarboxylation
+
O
O
OH
O
O
O
O
OH
O
O
O
O
O
NH2
13
14
15
Active
5
OH
OH
OH
Inactive
NH
NH
NH2
O
O
F
F
F
F
F
F
F
F
N3
NH2
b
c
d
Untreated
0.5 µM
10 µM
1 µM
5
15
14
13
5
µM
6
4
3
2
0 Gy
8 Gy
6 Gy
4 Gy
120
60
8
4
0
0
0
4
2
0
0
0
12 Gy
6
0
Gy
Gy
0
10 20 30 40 50 60
Irradiation dose (Gy)
0
6
12
24
36
48
60
2
3
4
5
6
Irradiation dose (Gy)
Retention time (min)
e
f
g
PBS
PBS
PBS + X-ray (P = 0.132)
Doxorubicin 15
Prodrug 13
NS NS
Doxorubicin 15 (*P = 0.019)
Prodrug 13 (P = 0.374)
NS
2
2
1
1
5
100
80
2
,500
,000
,500
,000
Doxorubicin 15 + X-ray
(**P = 0.009)
PBS
****
****
****
0
5
0
5
0
Doxorubicin 15
Prodrug 13
PBS + X-ray
Doxorubicin 15
2
Prodrug 13 + X-ray
6
4
2
0
0
0
0
(
*P = 0.011)
1
1
+
X-ray
Prodrug 13
X-ray
+
5
00
0
0
10
20
30
40
50
1
5
9
13 17 21 25 29 33 37
Days
Days
CK
CK-MB
LDH
Fig. 5 | Doxorubicin prodrug activation by X-ray irradiation. a, Reaction of the doxorubicin prodrug 13 and mechanistic pathway for liberation of
doxorubicin, 15. b, HPLC traces for the reaction mixture of prodrug 13 after irradiation with 0ꢀGy to 60ꢀGy, showing unreacted prodrug 13 (R ꢀ=ꢀ5.40ꢀmin),
T
intermediate 14 (R ꢀ=ꢀ4.36ꢀmin), doxorubicin 15 (R ꢀ=ꢀ3.14ꢀmin) and 5 (R ꢀ=ꢀ3.03ꢀmin). c, The yield of doxorubicin 15 generated from prodrug 13 (10ꢀµM)
T
T
T
−1
under irradiation (6ꢀGy up to 60ꢀGy, 6ꢀGyꢀmin ). Owing to the poor solubility of doxorubicin, the irradiation mixture was extracted into an organic solvent
prior to HPLC analysis. The data are presented as meanꢀ±ꢀs.d. (nꢀ=ꢀ3). d, HeLa cells were incubated with prodrug 13 (0.5, 1, 5 and 10ꢀµM) for 4ꢀh, followed
by X-ray irradiation (from 0ꢀGy to 60ꢀGy), and cell viability was measured using an MTT assay after 24ꢀh of incubation at 37ꢀ°C. The data are presented
as meanꢀ±ꢀs.d. (nꢀ=ꢀ6). e, HeLa tumour-bearing BALB/c nude mice were treated with prodrug 13 by intratumoral injection. At 4ꢀh post injection, mice were
treated with or without 6ꢀGy of X-ray irradiation. Tumour burdens were measured every other day using a caliper. The data are presented as meanꢀ±ꢀs.d.
(
nꢀ=ꢀ5). Statistical analysis was performed using one-way ANOVA with Dunnett post-test compared to PBS-treated mice at day 33. f, Survival curves
3
for HT29 tumour-bearing mice. Mice were aged until moribund or the tumour volume reached 2,000ꢀmm (nꢀ=ꢀ5). g, Levels of biochemical markers CK,
CK-MB and LDH in plasma 48ꢀh after PBS, prodrug 13 and doxorubicin 15 treatments. The data are presented as meanꢀ±ꢀs.d. (nꢀ=ꢀ5). Statistical analysis
was performed using one-way ANOVA with Dunnett post-test compared to PBS-treated mice, ****Pꢀ<ꢀ0.001; NS, not significant.
drug doxorubicin (Fig. 5). Prodrugs 11 and 13 were found to be concentration (IC ) values fall upon irradiation as the active drug
5
0
stable in a broad variety of biological conditions, including whole is generated (Fig. 5d). This drove us to investigate the anticancer
blood. We observed that prodrug 13 (10 µM) was also decaged efficacy of the prodrug system in animals. In a tumour-bearing
in a radiation-dose-dependent manner, reaching 50% conversion mice model, we found that tumour growth was dramatically
with irradiation of 60 Gy (Fig. 5c) to give three products: doxo- inhibited, with overall survival prolonged by the combined treat-
rubicin 15 (R = 3.14 min), tetrafluoroaniline 5 (R = 3.03 min) ment of X-ray and prodrug 13. Importantly, evaluations of body
T
T
and the relatively unstable ‘linker-doxorubicin product’ 14 weight changes and key organ histological abnormalities (Fig. 5g
3
6
(
R = 4.36 min) .
and Supplementary Figs. 42–44) showed that the prodrugs dis-
T
The prodrug was non-toxic when used at concentrations played no gross toxicities; indeed, they reduced the heart toxicity
up to 100 μM, and, as expected, the half-maximum inhibitory usually found with doxorubicin4
0,41
.
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
09

Articles
Conclusion
We have developed a gamma/X-ray-mediated strategy for the activa-
tion of prodrugs based on the reduction of sulfonyl azide and phenyl
azide. The reactions are clearly mediated via free-radical chemis-
try and through the reductive loss of nitrogen , although the exact
reductive mechanism is unclear and will be the subject of future
studies. For the decaging of doxorubicin from the prodrug, decaging
was achieved through a two-step reaction: radical reduction of phe-
nyl azide and subsequent 1,6 self-immolation linker cleavage.
The findings here open the way to a whole new area of ‘ioniz-
ing irradiation’-mediated chemistry—not just for prodrug activa-
tion, but for switching on drug delivery from implanted devices to
the generation of a whole suite of tunable activation chemistries.
The local conversion of an inactive prodrug to an active drug via
NATurE CHEmIsTry
1
1
1
1
2
2
6. Takakusagi, Y. et al. Radiotherapy synergizes with the hypoxia-activated
prodrug evofosfamide: in vitro and in vivo studies. Antioxid. Redox Signal.
2
8, 131–140 (2018).
7. Nytko, K. J. et al. ꢁe hypoxia-activated prodrug evofosfamide in
combination with multiple regimens of radiotherapy. Oncotarget 8,
23702–23712 (2017).
8. Karzmark, C. J. Advances in linear accelerator design for radiotherapy,
advances in linear accelerator design for radiotherapy. Med. Phys. 11,
42
1
05–128 (1984).
9. Wroe, L. M. et al. Comparative analysis of radiotherapy linear accelerator
downtime and failure modes in the UK, Nigeria and Botswana. Clin. Oncol.
32, e111–e118 (2020).
0. Shtarkman, I. N., Gudkov, S. V., Chernikov, A. V. & Bruskov, V. I. X-ray- and
heat-induced generation of hydrogen peroxide and hydroxyl radicals in
aqueous solutions of ꢄ-amino acids. Biophysics 53, 1–7 (2008).
1. Riley, P. A. Free radicals in biology: oxidative stress and the eꢅects of ionizing
radiation. Int. J. Radiat. Biol. 65, 27–33 (2009).
concurrent radiotherapy will lead to much more tolerable treatment 22. Sutherland, B. M., Bennett, P. V., Sutherland, J. C. & Laval, J. Clustered
DNA damages induced by X-rays in human cells. Radiat. Res. 157,
regimes without the systemic toxicities observed in conventionally
delivered chemotherapies.
6
11–616 (2002).
2
3. Sutherland, B. M., Bennett, P. V., Sidorkina, O., Laval, J. & Clustered, D. N. A.
Damages induced in isolated DNA and in human cells by low doses of
ionizing radiation. Proc. Natl. Acad. Sci. USA 97, 103–108 (2000).
Online content
Any methods, additional references, Nature Research reporting sum- 24. Lomax, M. E., Folkes, L. K. & O’Neill, P. Biological consequences of
radiation-induced DNA damage: relevance to radiotherapy. Clin. Oncol. 25,
maries, source data, extended data, supplementary information,
5
78–585 (2013).
acknowledgements, peer review information; details of author contri-
butionsandcompetinginterests;andstatementsofdataandcodeavail-
ability are available at https://doi.org/10.1038/s41557-021-00711-4.
2
5. Valerie, K. et al. Radiation-induced cell signaling: inside-out and outside-in.
Mol. Cancer Ter. 6, 789–801 (2007).
26. Sahin, E. et al. Telomere dysfunction induces metabolic and mitochondrial
compromise. Nature 470, 359–365 (2011).
2
2
2
7. O’Neill, P. & Wardman, P. Radiation chemistry comes before radiation
biology. Int. J. Radiat. Biol. 85, 9–25 (2009).
Received: 23 March 2020; Accepted: 22 April 2021;
Published online: 10 June 2021
8. Azzam, E. I., Jay-Gerin, J.-P. & Pain, D. Ionizing radiation-induced metabolic
oxidative stress and prolonged cell injury. Cancer Lett. 327, 48–60 (2012).
9. Nordsmark, M. et al. Prognostic value of tumor oxygenation in 397 head and
neck tumors aꢃer primary radiation therapy. An international multi-center
study. Radiother. Oncol. 77, 18–24 (2005).
References
1
.
Rautio, J. et al. Prodrugs: design and clinical applications. Nat. Rev. Drug
Discov. 7, 255–270 (2008).
2.
Miwa, M. et al. Design of a novel oral ꢀuoropyrimidine carbamate,
capecitabine, which generates 5-ꢀuorouracil selectively in tumours by
enzymes concentrated in human liver and cancer tissue. Eur. J. Cancer 34,
3
0. Kuzmin, G. N., Knatko, M. V. & Kurganov, S. V. Light and X-ray-induced
chemistry of methane on TiO . React. Kinet. Catal. Lett. 23, 313–317 (1983).
2
3
1. Barner-Kowollik, C., Vana, P., Quinn, J. F. & Davis, T. P. Long-lived
intermediates in reversible addition-fragmentation chain-transfer (RAFT)
polymerization generated by γ radiation. J. Polym. Sci. Pol. Chem. 40,
1274–1281 (1998).
3
.
.
Krasnovskaya, O. O. et al. ꢁiourea modiꢂed doxorubicin: a perspective
pH-sensitive prodrug. Bioconjug. Chem. 30, 741–750 (2019).
1
058–1063 (2002).
4
Ohwada, J. et al. Synthesis and biological activities of a pH-dependently
activated water-soluble prodrug of a novel hexacyclic camptothecin analog.
Bioorg. Med. Chem. Lett. 19, 2772–2776 (2009).
3
2. Yang, Y. et al. Photodecomposition of thienylsulfonyl azides: generation and
spectroscopic characterization of triplet thienylsulfonyl nitrenes and
3
-thienylnitrene. J. Phys. Chem. A 123, 9311–9320 (2019).
5
.
.
Swiꢃ, L. P., Cutts, S. M., Rephaeli, A., Nudelman, A. & Phillips, D. R.
Activation of adriamycin by the pH-dependent formaldehyde-releasing
prodrug hexamethylenetetramine. Mol. Cancer Ter. 2, 189–198 (2003).
Bentebibel, S.-E. et al. A ꢂrst-in-human study and biomarker analysis of
NKTR-214, a novel IL2Rβγ-biased cytokine, in patients with advanced or
metastatic solid tumors. Cancer Discov. 9, 711–721 (2019).
3
3
3
3
3. Reagan, M. T. & Nickon, A. ꢁe photolysis of sulfonyl azides in isopropyl
alcohol. J. Am. Chem. Soc. 90, 4096–4105 (1968).
4. Dermer, O. C. & Edmison, M. T. Orientation in aromatic substitution by the
benzenesulfonimido radical. J. Am. Chem. Soc. 77, 70–73 (1955).
5. Bukowski, R. M., Yasothan, U. & Kirkpatrick, P. Pazopanib. Nat. Rev. Drug
Discov. 9, 17–18 (2010).
6
7
.
.
Mustafa, A. A., Rajan, R., Suarez, J. D. & Alzghari, S. K. A review of the
opioid analgesic benzhydrocodone-acetaminophen. Cureus 10, e2844 (2018).
Charych, D. et al. Modeling the receptor pharmacology, pharmacokinetics
and pharmacodynamics of NKTR-214, a kinetically-controlled interleukin-2
6. Sleijfer, S. et al. Pazopanib, a multikinase angiogenesis inhibitor, in patients
with relapsed or refractory advanced soꢃ tissue sarcoma: a phase II study
from the European Organisation for Research and Treatment of Cancer–Soꢃ
Tissue and Bone Sarcoma Group (EORTC study 62043). J. Clin. Oncol. 27,
8
(
IL2) receptor agonist for cancer immunotherapy. PLoS ONE 12,
3
126–3132 (2009).
e0179431 (2017).
3
3
3
7. Stevens, M. Y., Sawant, R. T. & Odell, L. R. Synthesis of sulfonyl azides via
9
1
.
Norman, D. J. et al. Electrodrugs: an electrochemical prodrug activation
strategy. Chem. Commun. 54, 9242–9245 (2018).
15
diazotransfer using an imidazole-1-sulfonyl azide salt: scope and N NMR
labeling experiments. J. Org. Chem. 79, 4826–4831 (2014).
8. Zhu, X.-D. et al. Antiangiogenic eꢅects of pazopanib in xenograꢃ
hepatocellular carcinoma models: evaluation by quantitative
contrast-enhanced ultrasonography. BMC Cancer 11, 307 (2011).
9. Matikonda, S. S. et al. Mechanistic evaluation of bioorthogonal decaging with
trans-cyclooctene: the eꢅect of ꢀuorine substituents on aryl azide reactivity
and decaging from the 1,2,3-triazoline. Bioconj. Chem. 29, 324–334 (2018).
0. Doroshow, J. H. Eꢅect of anthracycline antibiotics on oxygen radical
formation in rat heart. Cancer Res. 43, 460–472 (1983).
0. Bezagu, M. et al. In situ targeted activation of an anticancer agent using
ultrasound-triggered release of composite droplets. Eur. J. Med. Chem. 142,
2–7 (2017).
1
1. Hossion, A. M. L., Bio, M., Nkepang, G., Awuah, S. G. & You, Y. Visible light
controlled release of anticancer drug through double activation of prodrug.
ACS Med. Chem. Lett. 4, 124–127 (2012).
1
1
2. Sauer, R. et al. Preoperative versus postoperative chemoradiotherapy for rectal
cancer. New Engl. J. Med. 351, 1731–1740 (2004).
4
4
4
3. Rosenzweig, K. E. & Gomez, J. E. Concurrent chemotherapy and radiation
therapy for inoperable locally advanced non-small-cell lung cancer. J. Clin.
Oncol. 35, 6–10 (2017).
1. Olson, R. D. & Mushlin, P. S. Doxorubicin cardiotoxicity: analysis of
prevailing hypotheses. FASEB J. 4, 3076–3086 (1990).
2. Benati, L. et al. Radical reduction of aromatic azides to amines with
triethylsilane. J. Org. Chem. 71, 5822–5825 (2006).
1
1
4. Alvarado-Miranda, A. et al. Concurrent chemo-radiotherapy following
neoadjuvant chemotherapy in locally advanced breast cancer. Radiat. Oncol.
4
, 24–28 (2009).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
5. Mao, X. et al. An agent-based model for drug-radiation interactions in the
tumour microenvironment: hypoxia-activated prodrug SN30000 in
multicellular tumour spheroids. PLoS Comput. Biol. 14, e1006469 (2018).
published maps and institutional affiliations.
© The Author(s), under exclusive licence to Springer Nature Limited 2021
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | VOL 13 | AUGUST 2021 | 805–810 | www.nature.com/naturechemistry
8
10

NATurE CHEmIsTry
Articles
Cell viability against prodrug 13 before and after X-ray irradiation. HeLa cell
viability was evaluated using an MTT assay as described above. The cells were
treated with prodrug 13 (0.5, 1, 5 and 10µM) for 4h, and irradiated with X-rays at
Methods
Animals. BALB/c mice and BALB/c nude mice (female, 6–8weeks old) were
provided by Guangdong Medical Laboratory Animal Centre. ꢁe housing
conditions for the animals comprised a barrier environment, with an ambient
temperature of 24°C and relative humidity of 50%. All the animals were
maintained in a 12h:12h light–dark cycle. All animal experiments were performed
under the Guide for Care and Use of Laboratory Animals and were approved by
the Institutional Animal Care and Use Committee (IACUC) of Shenzhen Institutes
of Advanced Technology (SIAT).
0
, 6, 12, 24, 36, 48 and 60Gy. Cells treated with 50% DMSO in DMEM were used
as a negative control.
Statistical analysis. Data are presented as mean± s.d. Dunnett’s t-tests were
used to determine whether the variance between two groups is similar. One-way
ANOVA was applied for comparison of multiple groups. Statistical analysis was
performed using GraphPad Prism. P<0.05 was considered statistically significant.
X-ray source. We used a linear accelerator (Clinac iX from Varian Medical
Reporting Summary. Further information on research design is available in the
Systems) generating X-rays with a nominal energy of 6MV (with a Bremsstrahlung
Nature Research Reporting Summary linked to this Article.
−
1
distribution of 2–6MeV), a dose rate of 6Gy min with samples treated at a depth
of 1,015mm from the tungsten target and a build-up of 15mm of solid water. The
linear accelerator used was a Varian Medical Systems VitalBeam medical linear
Data availability
−
1
All relevant data supporting the findings of this study are available within the paper
and its Supplementary Information files.
accelerator. A dosage rate of 6Gymin was used for all experiments.
Monitoring model reactions under X-ray irradiation. Selected model
compounds (chemical structures are provided in Table 1 and Supplementary Fig. 1)
were dissolved in DMSO to give stock solutions of 100mM and diluted in PBS
Acknowledgements
We thank the ERC (ERC-2013-ADG 340469 ADREEM) for funding M.B., J.G., K.N.
and Y.Z. J.G. acknowledges support from the National Natural Science Foundation of
China (22071263), the Natural Science Foundation of Guangdong Province, China
(
20ml) to give a final concentration of 10μM. When the solutions were degassed,
this was done by bubbling Ar for 30min, followed by X-ray irradiation (0 to 60Gy,
to 10min). The reaction mixtures after irradiation were analysed by HPLC.
0
(2020A1515010994), Guangdong Province Zhujiang Talents Program (2019QN01Y127),
Shenzhen Fundamental Research Program (JCYJ20200109110215774) and ‘Hundred
Talents Program’ of the Chinese Academy of Sciences. Y.Z. acknowledges support
from the National Natural Science Foundation of China (22001261) and the China
Postdoctoral Science Foundation (2020M672873). Q.G. acknowledges support from the
China Postdoctoral Science Foundation (2020M682976).
Reaction of coumarin azide 8 under X-ray irradiation. A stock solution
of coumarin azide 8 (1M in DMSO) was diluted in PBS (20ml) to a final
concentration of 100μM. The solution was degassed by bubbling Ar for 30min
before X-ray irradiation (0 to 60Gy). The fluorescence intensity of the reaction
mixtures was analysed. Product 9 was isolated by prep-HPLC and characterized by
1
H-NMR and high-resolution mass spectrometry (HRMS).
Author contributions
M.B. and J.G. conceived, designed and directed the project. Y.Z., K.N., H.D. and H.P.
conducted the compound syntheses, X-ray activation and characterizations. Y.Z. and J.G.
conducted the MTT, flow cytometry, transwell and wound healing assays. Y.Z. and Q.G.
conducted the in vivo experiments. J.G., Y.Z., M.P., H.R., D.A. and M.B. co-wrote the
manuscript. All authors analysed the data, contributed to the scientific discussion and
revised the manuscript.
Reaction of coumarin azide 8 in live cells under X-ray irradiation. HeLa cells
4
were seeded in 24-well plates at a density of 5×10 cells per well and incubated
overnight. The cells were then treated with coumarin azide 8 (100μM) 1h before
irradiation (0, 6, 36 and 60Gy) and analysed by flow cytometry using a DAPI filter
(λex/em =360/450nm) and confocal microscopy.
Reaction of prodrug 11 under X-ray irradiation. A stock solution of prodrug 11
100mM in DMSO) was diluted in PBS (20ml) to a final concentration of 20μM.
(
Competing interests
The solution was degassed by bubbling Ar for 30min before X-ray irradiation (0 to
The authors declare no competing interests.
6
0Gy). The reaction mixture was analysed by FTIR, and products 10 and 12 were
1
isolated by prep-HPLC and characterized H-NMR and HRMS.
Additional information
Reaction of prodrug 13 under X-ray irradiation. A stock solution of doxorubicin
prodrug 13 (100mM in DMSO) was diluted in PBS (20ml, with 0.1%, vol/vol
Triton X100) to give a final concentration of 20μM. The solution was degassed by
bubbling Ar for 30min before X-ray irradiation (0 to 60Gy). The reaction mixture
was analysed by FTIR, and products 15 and 5 were isolated by prep-HPLC and
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41557-021-00711-4.
Correspondence and requests for materials should be addressed to J.G. or M.B.
Peer review information Nature Chemistry thanks Peng Chen and the other,
anonymous, reviewer(s) for their contribution to the peer review of this work.
1
characterized H-NMR and HRMS. (Note that compounds 15 and 5 co-eluted on
the HPLC.)
Reprints and permissions information is available at www.nature.com/reprints.
NAꢁꢂRꢃ CHꢃMꢄSꢁRY | www.nature.com/naturechemistry

μ