A gemcitabine-based conjugate with enhanced antitumor efficacy by suppressing HIF-1α expression under hypoxia

Article abstract of DOI:10.1016/j.bmc.2021.116214

Hypoxia is one of the unique features of tumor physiology. Hypoxia inducible factor (HIF-1α), as a major transcription factor in response to hypoxia, has been considered as a promising tumor-specific target for anticancer therapy. The formation of a hypox

Full text of DOI:10.1016/j.bmc.2021.116214

Bioorg. Med. Chem. 41 (2021) 116214
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
A gemcitabine-based conjugate with enhanced antitumor efficacy by
suppressing HIF-1α expression under hypoxia
,
b
,
,
,
,b,*
Zichen Xua 1, Bin Zhanga 1, Zhixin Liaoa *, Shaohua Goua
a Pharmaceutical Research Center and School of Chemistry and Chemical Engineering, Southeast University, Nanjing 211189, China
b Institute of Nanjing Junruo Biomedicine, Nanjing 211100, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Hypoxia is one of the unique features of tumor physiology. Hypoxia inducible factor (HIF-1α), as a major
Hypoxia
transcription factor in response to hypoxia, has been considered as a promising tumor-specific target for anti-
cancer therapy. The formation of a hypoxic microenvironment in tumors can decrease the curative effect of
cytotoxic chemotherapeutic drugs. To promote the antitumor efficacy of chemotherapy by suppressing hypoxia,
HIF-1
α
Gemcitabine
Ameliorate tumor hypoxia
Cancer therapy
we designed and prepared a novel gemcitabine-based drug conjugate (GEM-5) containing a HIF-1
α
inhibitor (YC-
М under hypoxia) and
remarkably induced the apoptosis of A2780 cells in vitro. Additionally, western blot analysis demonstrated that
1). As expected, GEM-5 showed excellent antiproliferative activity (IC50 = 0.03
μ
GEM-5 significantly down-regulated the expression of HIF-1
α
and up-regulated the expression of tumor sup-
pressor p53. More importantly, GEM-5 effectively inhibited tumor growth in the A2780 xenograft mouse model
and significantly ameliorated tumor hypoxia in vivo. This novel, simple, and effective strategy for overcoming
tumor hypoxia and enhancing the antitumor effect of chemotherapeutic drugs has great potential in cancer
therapy.
1. Introduction
promotes polyubiquitination of HIF-1α, followed by degradation by the
26S proteasome.11,12 However, once the tissue is hypoxic, binding of
The heterogeneous tumor cell growth and dysfunctional vasculature
result in the formation of hypoxic regions that rely on anaerobic meta-
bolism to survive.1–3 Hypoxia spreads heterogeneously within the tumor
tissue and eventually leads to the formation of solid tumors.4 As re-
ported, the normal oxygen level in most mammalian tissues is 2–9%.
However, hypoxia is defined by <2% of tissue oxygen levels, which
occurs in various pathological conditions including cancer progres-
sion.4,5 Nowadays, hypoxia has been regard as a huge obstacle to cancer
therapy, as it can contribute to tumor proliferation, metastasis, angio-
genesis and resistance to therapeutics.6–8
pVHL to HIF-1α is inhibited, resulting in accumulation of HIF-1α and its
dimerization with the constitutive HIF-1β subunit. Association of HIF-1
α
with the HIF-1β subunit leads to the formation of HIF-1, causing the
expression of hypoxia-responsive element (HRE), and therefore acti-
vates transcription of a number of downstream target genes.13,14 HIF-1
α
helps tumors adapt to hypoxia, subsequently making tumors more
aggressive and resistant to chemotherapy and radiation, thereby
resulting in poor patient prognosis.15–17 Recent studies have found that
remarkable overexpression of HIF-1α was measured in many human
cancers, such as lung cancer, ovarian cancer, hepatocellular carcinoma,
Hypoxia-inducible factor 1 (HIF-1), a transcription factor that re-
breast cancer, and so on.18,19 Therefore, HIF-1
α has been considered to
sponds to a hypoxic environment is a heterodimer composed of HIF-1
α
be a promising target for the development of novel therapeutics against
cancer.
and HIF-1β subunits.9,10 Under the normoxic condition, proteins with an
oxygen-dependent prolyl-4-hydroxylase domain (PHD) covalently
Chemotherapy has been regarded as the critical therapy for various
human cancers.20 However, There is growing evidence that hypoxic
solid tumors display decreased drug accumulation and worse thera-
peutic effect after anticancer drug treatment.21,22 For instance,
modify a region in HIF-1α, by hydroxylating proline residues. Hydrox-
ylated HIF-1
α subsequently forms hydrogen bonds with side chains on
the von Hippel–Lindau tumor suppressor protein (pVHL), which in turn
* Corresponding authors at: Pharmaceutical Research Center and School of Chemistry and Chemical Engineering, Southeast University, Nanjing 211189, China (S.
Gou).
E-mail addresses: zxliao@seu.edu.cn (Z. Liao), 2219265800@qq.com (S. Gou).
1
These authors contributed equally to this work.
https://doi.org/10.1016/j.bmc.2021.116214
Received 22 March 2021; Received in revised form 6 May 2021; Accepted 7 May 2021
Available online 11 May 2021
0968-0896/© 2021 Elsevier Ltd. All rights reserved.

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
abnormal angiogenesis in the hypoxic region of tumors reduces the
bioavailability of drugs.23 Besides, cancer cells adapted to hypoxia
microenvironment can up-regulate the activity of P-glycoprotein, and
thus increasing the resistance of cancer cells to chemotherapy drugs.24,25
studies.30,31 The intermediates (GEM-1, GEM-2, and GEM-3) were pre-
pared according to the reported methods.33,34 Then GEM-5 conjugate
was obtained via esterification of intermediate GEM-3 and YC-2 in the
presence of TBTU and TEA, and subsequent deprotection of the tert-
butoxycarbonyl group under trifluoroacetic acid. The chemical struc-
tures of intermediates and GEM-5 were characterized by 1H and 13C
NMR spectroscopy together with HRMS spectrometer (Figures S1-S11 in
Supporting information). The purity of target compound GEM-5 was
assayed by HPLC technique (Figure S12 in Supporting information).
In view of the above considerations, we herein report a novel HIF-1
α
α
inhibitor- gemcitabine conjugate (GEM-5), which can deliver a HIF-1
inhibitor and a chemotherapy drug unit to hypoxic tumor cells. Gem-
citabine (2′,2′-difluoro-2′-deoxycytidine, dFdC, GEM), a pyrimidine
nucleoside analogue, is widely used as the hydrochloride salt to treat
pancreatic, non-small cell lung, ovarian, bladder, and breast can-
cers.26,27 YC-1 is a potent HIF-1
α inhibitor that has been considered as
2.2. Antiproliferative activity assay
one of the best potential drug candidates due to its a variety of potent
biological activities like anti-platelet and anti-vascular.28,29 Besides, the
In order to investigate the effect of the GEM-5 conjugate against
human cancer cells, A549 (lung), MCF-7 (breast) and A2780 (ovarian)
together with human normal liver cells LO2 were first incubated with
GEM-5 under normoxic condition and evaluated via MTT assay, using
GEM and YC-1 as references. The corresponding IC50 values obtained
after 72 h exposure are summarized in Table 1. As shown in Table 1, YC-
HIF-1
α
inhibition of YC-1 has been verified in our previous reports.30,31
The obtained GEM-5 is expected to have the ability of releasing GEM
and YC-1 for their respective biological actions that can not only carry
the chemotherapy drug warhead into the tumor cells but also have a
functional molecule to target the tumor cells and inhibit HIF-1
hypoxia.
α under
Table 1
2. Results and discussion
Cytotoxic effects of GEM, YC-1 and GEM-5 under normoxia.
Compound
IC50
(
μ
М) α
2.1. Chemistry
A549
MCF-7
A2780
LO2
The detailed procedure of synthesizing GEM-5 conjugate is shown in
Scheme 1. As the ester is easily affected by acid or esterase, it can be
cleaved to release active biological species in cancer cells.32 YC-1 as a
GEM
2.31 ± 0.02
>80
6.35 ± 0.09
>80
1.25 ± 0.05
>80
11.87 ± 0.68
>80
YC-1
GEM-5
0.37 ± 0.13
2.60 ± 0.03
0.13 ± 0.02
4.43 ± 0.45
HIF-1α targeting moiety was modified by reacting with succinic anhy-
α
Values are the mean ± SD of three independent measurements in duplicates.
dride to generate YC-2, which was prepared as reported in our previous
Scheme 1. Reagents and conditions: (i) TBDMSCl, imidazole, pyridine, rt; (ii) Boc2O, DMAP, TEA, 1,4-dioxane, rt; (iii) triethylamine trihydrofluoride, THF, rt; (iv)
succinic anhydride, DMAP, THF, 70 ℃; (v) TBTU, TEA, rt; (vi) CF3COOH, DCM, 0 ℃.
2

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
1 displayed negligible cytotoxicity against three cancer cells under
normoxia, while GEM-5 showed stronger activity than YC-1 and GEM
against all the tested cancer cells. In particular, GEM-5 was about 9.6
times more effective than GEM against human ovarian cancer (A2780)
normoxic or hypoxic conditions respectively. These results revealed that
in A2780 cells, GEM-5 evidently arrested the S phase of the cell cycle
under either normoxic or hypoxic conditions.
cells, with an IC50 value of 0.13 μM in comparison with 1.25 μM of GEM.
2.5. Western blot analysis of GEM-5
Besides, GEM-5 was 6.2 and 2.4 fold more potent than GEM against
human lung cancer (A549) and breast cancer (MCF-7) cell lines,
For the sake of investigating the mechanism of apoptosis induced by
GEM-5, western blot analysis was made in A2780 cells under hypoxia,
using GEM as positive control. The results in Fig. 3 showed that the
respectively, with IC50 values of 0.37 and 2.60 μM. It is worth noting that
GEM-5 possessed the selectivity between cancer cells and normal cells
when compared its cytotoxicity toward cancer cells with that toward
human normal LO2 liver cells.
inhibitory effect of GEM on HIF-1
centration of 1
decreased by GEM-5 under hypoxic condition in a dose dependent
α
was quite weak even at the con-
μ
M. As expected, the level of HIF-1α was significantly
In contrast to the cytotoxicity under normoxia, antitumor activity of
GEM-5 was further evaluated on A549 and A2780 cells under hypoxic
condition due to its potent antitumor activity under normoxia. As shown
in Table 2, under hypoxic condition, YC-1 exhibited improved cytotoxic
manner as YC-1, with about 50% decrease at 1 μM. It has been proposed
that p53 activation promotes cancer cell killing.35,36 Moreover, activa-
tion of p53 pathway can restrict malignant transformation by triggering
apoptosis or cell cycle arrest.37 Therefore, we further evaluated the
expression level of p53. As shown in Fig. 3, the level of p53 was up-
regulated in a dose-dependent manner when cells were co-incubated
with GEM or GEM-5. However, YC-1 has no effect on the expression
of p53. In brief, these results indicated that GEM-5 induced apoptosis by
activity in A549 (IC50 = 25.62
μ
M) and A2780 (IC50 = 20.35
μM) cells,
owing to the intrinsic function of inhibition of HIF-1
α
. The cytotoxic
activities of GEM-5 and GEM towards A549 were weaker than that
under normoxia. Besides, the cytotoxicity of GEM against A2780 cancer
cells under hypoxic condition was similar to that under normoxia.
Encouragingly, GEM-5 exhibited significantly promoted cytotoxicity in
down-regulating HIF-1
α and simultaneously increasing the expression of
A2780 cells, with IC50 values of 0.03 μM in comparison to 1.38 μM of
p53 protein.
GEM. Taken together, these results indicated that both the level of ox-
ygen and the cell types have great impacts on the in vitro anticancer
activity of GEM-5. Notably, GEM-5 exhibited the best antitumor activity
toward A2780 cells under hypoxic condition. Therefore, it is significant
for us to explore the antitumor mechanism of GEM-5 further.
2.6. In vivo antitumor activity
To validate the efficacy of GEM-5 inhibiting tumor growth in vivo,
A2780 cell xenograft mouse models were established by subcutaneously
injecting A2780 cells in the logarithmic phase into the right armpit of
the mice. After the model was well-established, mice were randomly
divided into four groups with 5 mice in each treated group: (1) vehicle
treated group (physiological saline injection as control), (2) GEM treated
group (125 mg/kg), (3) GEM-5 treated group (125 mg/kg, equal mass
dose to GEM), (4) GEM-5 treated group (271 mg/kg, equal molar dose to
GEM), and treated with above-mentioned formulations once a week for
28 days in the entire observation period. As shown in Fig. 4, tumor
volume and body weight of tumor-bearing mice were monitored every 3
days for 28 days. At the end of the experiments, the tumor volumes
(Fig. 4c) in mice treated with GEM-5 was much smaller than the tumor
volumes in mice treated with PBS. Compared with that of the PBS group,
the tumor volume after 28 days treatment was 29.37% for GEM, 46.09%
for GEM-5 (125 mg/kg), or 36.61% for GEM-5 (271 mg/kg), showing
that GEM-5 procured better tumor growth inhibitory efficacy than GEM.
The tumor inhibitory rate (TIR) was calculated from tumor weight
(Fig. 4d). Compared with that of the PBS group, the TIR of GEM-5 (271
mg/kg) is 64.6%, which is slightly lower than that of GEM (69.70%), but
higher than that of GEM-5 (125 mg/kg) (55.1%). Notably, none of the
compounds significantly affected the animal body weight, illustrating no
obvious toxicity of GEM-5 for tumor therapy (Fig. 4e).
2.3. Induced apoptotic cell death of GEM-5
According to the results of cytotoxicity assay, apoptosis of GEM-5
against A2780 cells was studied by an Annexin VFITC/PI assay under
normoxic or hypoxic condition, using GEM as positive control. As
showed in Fig. 1, under normoxic condition (Fig. 1a and c), GEM ach-
ieved an apoptosis rate of 39.44% (5.02% early and 34.42% late
apoptosis) after 72 h incubation. Meanwhile, the apoptotic population
rose to 52.67% (3.85% early and 48.82% late apoptosis) after treatment
with 0.5 μM of GEM-5. As demonstrated in Fig. 1b and c, when the
hypoxia was conducted, the apoptotic percentage of GEM reached
44.51% (25.00% early and 19.51% late apoptosis), which was similar to
that under normoxia. In contrast, the enhanced apoptosis rate of GEM-5
was much greater than GEM with the population of apoptotic cells rising
to 80.89% (48.48% early and 32.41% late apoptosis). The above results
confirm the effect of GEM-5 as an antitumor agent under hypoxia.
2.4. Effect on cell cycle arrest of GEM-5
In order to investigate the effect of GEM-5 on cell cycle arrest, the
cycle distribution of A2780 cells treated with 0.5 μM of GEM-5 for 24 h
2.7. Detection of hypoxia in vivo
under hypoxic and normoxic conditions was analyzed by flow cytometry
with untreated cells as negative control and GEM-treated cells as posi-
tive control. The obtained data (Fig. 2) obviously indicated that GEM-5
arrested the cell cycle at the S phase (63.02% under normoxia and
72.64% under hypoxia), when compared to the untreated control group
(20.11% under normoxia and 34.79% under hypoxia). Also, GEM
arrested the cell cycle at the S phase by 51.54% and 60.08% under
Finally, the capability of GEM-5 to ameliorate tumor hypoxia in vivo
was further studied using immunofluorescence staining assay, in which
the tumor cell nuclei and hypoxia areas were stained with 4,6-diami-
dino-2-phenylindole (DAPI) (blue) and antipimonidazole antibody
(green), respectively. As shown in Fig. 5, mice tumor sections of the
intravenous administration of GEM-5 (especially at the dose of 271 mg/
kg) showed a significantly weakened green fluorescence (the greener
pimonidazole-stained fluorescent regions are, the more severe hypoxia
of tumor tissue is) compared to the control and GEM treated groups,
respectively. These data demonstrate that tumor hypoxia can be suc-
cessfully alleviated by GEM-5 due to YC-1 conjugated.
Table 2
Cytotoxic effects of GEM, YC-1 and GEM-5 under hypoxia.
Compound
IC50
(
μ
М)α
A549
A2780
LO2
GEM
14.65 ± 0.24
25.62 ± 1.41
3.03 ± 0.31
1.38 ± 0.02
20.35 ± 1.37
0.03 ± 0.038
4.51 ± 0.31
26.58 ± 2.06
3.04 ± 0.19
YC-1
3. Conclusions
GEM-5
α
Values are the mean ± SD of three independent measurements in duplicates.
In summary, Gem-5, our newly prepared bi-functional cancer
3

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
Fig. 1. Flow cytometry analysis for apoptosis of A2780 cells induced by GEM and GEM-5, at the same concentration of 0.5 μM for 72 h under either normoxic (a) or
hypoxic (b) conditions. The cells were harvested and labeled with annexin-V-FITC and PI, and analyzed by flow cytometry. (c) Bar graph indicating ratio of the cell
population in different phases after staining with annexin V/PI.
therapeutic agent conjugate, is composed of HIF-1
α
inhibitor YC-1 and
Jiangsu KeyGEN BioTECH company (China). 1H and 13C NMR spectra
were recorded in DMSO‑d6 or Methanol‑d4 with Bruker 300 MHz or 500
MHz spectrometer. Mass spectra were measured by an Agilent 6224 ESI/
TOF MS instrument.
antitumor drug gemcitabine. Interestingly, Gem-5 showed stronger
antiproliferative activity than gemcitabine and YC-1 against all the
tested cancer cells under either normoxic or hypoxia condition. Espe-
cially, the cytotoxicity and the apoptosis level of GEM-5 in A2780 cells
were dramatically increased with the decrease of the O2 concentration.
Besides, GEM-5 could significantly down-regulate the expression of HIF-
4.2. Synthesis and characterization
1α and up-regulate the expression of tumor suppressor p53 under the
4.2.1. Synthesis of GEM-1
hypoxic condition. These advantages of GEM-5 resulted in effective
inhibition on tumor growth in the A2780 xenograft mouse model and
exhibited low toxicity in vivo. More importantly, Gem-5 significantly
ameliorated tumor hypoxia compared to gemcitabine. Consequently,
our study has provided a new and promising strategy for constructing
To a solution of Gemcitabine HCl (3.0 g, 10 mmol) and imidazole
(0.82 g, 12 mmol) in 20 mL anhydrous pyridine, TBDMS-Cl (1.81 g, 12
mmol) dissolved in anhydrous pyridine (6 mL) was added slowly. The
reaction mixture was stirred at room temperature for 4 h, then solvents
were removed in vacuo, and the resulting crude was dissolved in 50 mL
ethyl acetate, washed with brine (2 × 20 mL), dried over Na2SO4,
concentrated and recrystallized from ethyl acetate to give GEM-1 (1.73
g, 45.2%). 1H NMR (300 MHz, DMSO‑d6) δ 7.62 (d, J = 7.5 Hz, 1H), 7.36
(s, 2H), 6.28 (d, J = 6.5 Hz, 1H), 6.13 (t, J = 7.7 Hz, 1H), 5.76 (d, J = 7.5
Hz, 1H), 4.12 (m, 1H), 3.95 (d, J = 11.6 Hz, 1H), 3.87–3.80 (m, 3H),
0.90 (s, 9H), 0.09 (s, 6H) ppm. HRMS (m/z) (ESI): calcd for
anticancer agents with HIF-1
α inhibitor to achieve more effective
chemotherapeutic therapy in hypoxic tumor tissue.
4. Experimental section
4.1. Reagents and instruments
C
15H24F2N3O4Si [Mꢀ H]ꢀ : 376.1582, found: 376.1620.
All chemical reagents and solvents were purchased commercially
from Aladdin, Energy Chemical or Adamas-beta and used without
further purification, unless noted specifically. The purity of GEM-5 used
in the biological studies was 98.06% (measured by HPLC, Waters
4.2.2. Synthesis of GEM-2
GEM-1 (1.13 g, 3.0 mmol) dissolved in dry 1,4-dioxane (6 mL) was
first mixed with DMAP (49 mg, 0.4 mmol) and TEA (10 mL), then the
mixture was stirred at 0 ◦C, DBDC (6.82 g, 31.5 mmol) in 1,4-dixane (15
mL) was added dropwise. The resulting solution was stirred at room
e2695). The GAPDH, HIF-1α, p53 antibodies were purchased from Santa
Cruz Biotechnology or Imgenex. All cancer cell lines were obtained from
4

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
Fig. 2. Flow cytometry analysis for cell cycle arrest of A2780 cells induced by GEM and GEM-5, at the same concentration of 0.5 μM for 24 h under either normoxic
(a) or hypoxic (b) conditions. The cells were fixed and stained with PI to analyze DNA content by flow cytometry. (c) Bar graph indicating the ratio of cells in
different cycles.
Fig. 3. (a) The expression levels of HIF-1
α and p53 in A2780 cells induced by gemcitabine, GEM-5 and YC-1 at the indicated concentrations, determined by western
blot analysis. (b) Signals obtained from western blot and normalized with GAPDH. Data were presented as the mean ± SD of three independent experiments.
temperature overnight. Solvents were evaporated under reduced pres-
sure. The crude product was purified by flash column chromatograph
over silica gel with petroleum ether/ethyl acetate (v/v, 5:1) to give
GEM-2 (1.38 g, 67.5%). 1H NMR (500 MHz, DMSO‑d6) δ 8.10 (d, J = 7.7
Hz, 1H), 6.96 (d, J = 7.7 Hz, 1H), 6.30 (t, J = 8.23 Hz, 1H), 5.27–5.21
(m, 1H), 4.35–4.32 (m, 1H), 3.98 (dd, J = 3.4, 11.95 Hz, 1H), 3.90 (dd, J
= 3.75, 11.95 Hz, 1H), 1.50 (s, 18H), 1.45 (s, 9H), 0.89 (s, 9H), 0.09 (d,
J = 4.24 Hz, 6H) ppm. HRMS (m/z) (ESI): calcd for C30H50F2N3O10Si
[M+H]+: 678.3228, found: 678.3133.
4.2.3. Synthesis of GEM-3
To a solution of GEM-2 (1.35 g, 2.0 mmol) in anhydrous THF (20
mL), triethylamine trihydrofluoride (1.61 g, 10 mmol) was added. The
reaction mixture was stirred at room temperature overnight. After
completion of reaction, the solvent was removed under reduced pres-
sure, and then ethyl acetate (150 mL) was added the reaction, washed
with brine three times. The organic phase was dried over Na2SO4, and
concentrated to dryness under reduced pressure. The residue was sub-
jected to flash column chromatograph over silica gel with petroleum
ether/ethyl acetate (v/v, 1:1) to give GEM-3 (0.76 g, 68.0%). 1H NMR
5

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
Fig. 4. Antitumor activity of GEM-5 on A2780 xenograft models. Nude mice implanted with A2780 cells in the flank were dosed via injecting intravenously of GEM
(125 mg/kg), GEM-5 (125 mg/kg), GEM-5 (271 mg/kg) and vehicle (equivalent volume of normal saline injection as control) once a week. The mice were sacrificed
and made mathematical statistics. (a, b) Mice and tumor pictures obtained after sacrifice of the mice treated with GEM and GEM-5. (c) The RTV (relative tumor
volume) (±SEM) is graphed with error bars representing the standard deviation. (d) Weight of the excised tumors of each group. (e) The average body weights
(±SEM) were plotted after sacrifice of the mice treated with GEM and GEM-5.
(500 MHz, DMSO‑d6) δ 8.24 (d, J = 7.7 Hz, 1H), 6.96 (d, J = 7.6 Hz, 1H),
6.28 (t, J = 8.4 Hz, 1H), 5.24–5.19 (m, 1H), 4.25–4.26 (m, 1H), 3.80 (dd,
J = 12.7 Hz, 2.7 Hz, 1H), 3.69 (dd, J = 12.7 Hz, 3.8 Hz, 1H), 1.50 (s,
18H), 1.46 (s, 9H), 1.19 (s, 1H) ppm. HRMS (m/z) (ESI): calcd for
C
24H34F2N3O10 [Mꢀ H]ꢀ : 562.2291, found: 562.2568.
4.2.4. Synthesis of GEM-4
YC-2 (81.0 mg, 0.2 mmol) dissolved in dry DMF (5 mL) was mixed
with TBTU (96.3 mg, 0.3 mmol) and TEA (30.0 mg, 0.3 mmol), then the
mixture was stirred at room temperature. After stirring 10 min, a solu-
tion of GEM-3 (141 mg, 0.25 mmol) in anhydrous DMF (2 mL) was
added. The resulting mixture was heated to 50 ◦C and stirred overnight.
After completion of reaction, the solvent was removed under reduced
pressure. The residue was subjected to flash column chromatograph
over silica gel with dichloromethane/ethyl acetate (v/v, 10:1) to give
GEM-4 (104 mg, 55.6%). 1H NMR (300 MHz, Methanol‑d4) δ 8.28 (d, J
= 7.7 Hz, 1H), 8.13 (d, J = 8.2 Hz, 1H), 7.52 (d, J = 8.5 Hz, 1H), 7.42 (t,
J = 7.6 Hz, 1H), 7.25 (dt, J = 13.5, 7.6 Hz, 6H), 7.00 (d, J = 3.3 Hz, 1H),
6.95 (d, J = 7.6 Hz, 1H), 6.64 (d, J = 3.3 Hz, 1H), 6.28 (t, J = 8.4 Hz,
1H), 5.66 (s, 2H), 5.24 (s, 2H), 5.20–5.18 (m, 1H), 4.24–4.25 (m, 1H),
3.77 (dd, J = 2.7 Hz, 12.7 Hz, 1H), 3.70 (dd, J = 3.8 Hz, 12.7 Hz, 1H),
2.65 (s, 4H), 1.50 (s, 18H), 1.48 (s, 9H) ppm. HRMS (m/z) (ESI): calcd
for C47H54F2N5O14 [M+H]+: 950.3635, found: 950.3590.
4.2.5. Synthesis of GEM-5
To a solution of Boc-protected GEM-4 (95.0 g, 0.1 mmol) in anhy-
drous CH2Cl2 (10 mL), trifluoroacetic acid (2 mL) was added dropwise at
0 ◦C, then the solution was stirred for 4 h. After completion, the reaction
mixture was adjusted pH 9 with saturated NaHCO3 solution. The
mixture was added to water (50 mL) and extracted with CH2Cl2 (2 × 50
mL), and then the organic phase was dried over anhydrous Na2SO4 and
subjected to flash column chromatograph over silica gel with dichloro-
methane/methanol (v/v, 15:1) to give GEM-5 (38 mg, 58.1%). 1H NMR
(300 MHz, DMSO‑d6) δ 8.29 (d, J = 7.7 Hz, 1H), 8.14 (d, J = 8.2 Hz, 1H),
7.52 (d, J = 8.5 Hz, 1H), 7.42 (t, J = 7.6 Hz, 1H), 7.25 (dt, J = 13.5, 7.6
Hz, 6H), 7.01 (d, J = 3.3 Hz, 1H), 6.95 (d, J = 7.6 Hz, 1H), 6.63 (d, J =
Fig. 5. Representative immunofluorescence images of tumor sections after
hypoxia staining. The nuclei and hypoxia areas were stained with DAPI (blue)
and antipimonidazole antibody (green), respectively. The experiments were
performed three times, and the results of the representative experiments
are shown.
6

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
3.3 Hz, 1H), 6.25 (t, J = 8.4 Hz, 1H), 5.66 (s, 2H), 5.24 (s, 2H),
5.21–5.19 (m, 1H), 4.54 (s, 1H), 4.21 (d, J = 8.4 Hz, 1H) 3.77 (dd, J =
2.7 Hz, 12.7 Hz, 1H), 3.69 (dd, J = 3.8 Hz, 12.7 Hz, 1H), 2.64–2.65 (m,
4H) ppm. 13C NMR (75 MHz, DMSO‑d6) δ 173.79, 172.40, 160.08,
149.53, 149.00, 147.65, 143.74, 140.79, 137.76, 135.46, 129.07,
128.04, 127.72, 127.41, 126.45, 122.17, 121.50, 120.72, 113.25,
110.77, 108.43, 95.21, 92.91, 81.90, 79.86, 59.14, 58.27, 52.43, 29.08,
29.05 ppm. HRMS (m/z) (ESI): calcd for C32H30F2N5O8 [M+H]+:
650.2057, found: 650.2066.
Fischer Scientifics Ltd.). A monoclonal anti-GAPDH antibody was used
as a loading control.
4.6. Antitumor activity in vivo
The in vivo cytotoxic activity of GEM-5 was investigated using a
human ovarian cancer cell line (A2780) in BALB/c nude mice. Five-
week-old female BALB/c nude mice (16–18 g) were purchased from
Shanghai Ling Chang biotechnology company (China); tumors were
induced by a subcutaneous injection in their right armpit region of 107
cells in 0.1 mL of sterile PBS. Animals were randomly divided into four
groups. When the tumors reached a volume of 100–150 mm3 in all mice
on day 19, the first group was injected with an equivalent volume of 5%
dextrose via a tail vein as the vehicle control mice. No. 2 group was
treated with gemcitabine at dose of 125 mg/kg body weight once a week
for 4 weeks. No. 3 and No. 4 groups were treated with GEM-5 at the
doses of 125 or 271 mg/kg body weight once a week for 4 weeks,
respectively. Gemcitabine HCl was dissolved in vehicle. GEM-5 was
dissolved in a small amount of DMF, and then diluted with Tween 80 and
5% dextrose injection. The final solution contains DMF:Tween 80:5%
dextrose injection = 10:2:88. Tumor volume and body weight were
recorded every three days after drug treatment. All mice were sacrificed
after 4 weeks of treatment and the tumor volumes were measured with
electronic digital calipers and determined by measuring length (A) and
width (B) to calculate volume (V = AB2/2).
4.3. Antiproliferative assay
The cytotoxic activity of target compound (GEM-5) was evaluated
with cancer cell lines (A549, MCF-7 and A2780) and normal cell lines
(LO2) by MTT assay under normoxic (20% O2, 5% CO2, and N2 75%) or
hypoxic (1% O2, 5% CO2, and N2 94%) condition, with GEM and YC-1 as
positive controls. All tested compounds were dissolved in DMSO to a
final concentration of 2 mmol/L and then subsequently diluted in cul-
ture medium at final concentration of 1.25, 2.5, 5, 10, 20, 40, 80 μmol/L,
respectively. About 5 × 104 cells/mL cells, which were in the logarith-
mic phase, were seeded in each well of 96-well plates and incubated for
12 h at the indicated condition. Compounds at seven different concen-
trations were then added to the test well and the cells were incubated at
◦
37 C under the indicated condition (normoxia or hypoxia) for 72 h.
Fresh MTT was added to each well at a terminal concentration of 5 mg/
mL and incubated with cells at 37 ◦C for 4 h. The formazan crystals were
dissolved in 100 mL DMSO each well, and an enzyme labeling instru-
ment was used to read absorbance with 570/630 nm double wavelength
measurement. Cytotoxicity was examined on the percentage of cell
survival compared with the negative control. The final IC50 values were
calculated by the Bliss method (n = 5). All of the tests were repeated in
three times.
4.7. Hypoxia immunofluorescence assay
Hypoxia marker pimonidazole HCl (Hypoxyprobe-1 plus kit, Hypo-
xyprobe Inc., USA) was used for tissue staining of hypoxia according to
the protocol provided with the kit. Briefly, tumor sections were incu-
bated with antipimonidazole antibody (FITC-MBb1) (dilution 1:100,
Hypoxyprobe Inc.) and horseradish peroxidase linked to rabbit anti-
FITC secondary antibody (dilution 1:100) following the kit manufac-
turer’s instructions. Nuclei were stained with DAPI (blue) and hypoxia
areas were stained with antipimonidazole antibody (green). Images
were obtained by confocal microscopy.
4.4. Cell apoptosis assay by flow cytometry
A2780 cells were seeded at the density of 2 × 106 cells/mL of the
DMEM medium with 10% FBS on 6-well plates to the final volume of 2
mL. The plates were incubated overnight and then treated with equiv-
alent concentrations of GEM-5 and GEM for 72 h. Briefly, after incu-
bation under normoxic (20% O2, 5% CO2 and 75% N2, at 37 ◦C) or
hypoxic condition (1% O2, 5% CO2 and 94% N2, at 37 ◦C) for 72 h, cells
were collected and washed with PBS twice, and then resuspended in
binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2, pH 7.4) at
Declaration of Competing Interest
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.
a concentration of 1 × 106 cells/mL. The cells were stained with 5
μL of
FITC Annexin V (BD, Pharmingen) and 5
μL propidium iodide (PI)
staining using annexin-V FITC apoptosis kit followed; 100 μL of the
Acknowledgements
solution was transferred to a 5 mL culture tube and incubated for 30 min
at RT (25 ◦C) in the dark. The apoptosis ratio was quantified by system
software (Cell Quest; BD Biosciences).
The authors are grateful to Jiangsu Province Hi-tech Key Laboratory
for Biomedical Research for financial aids to this research. We appre-
ciate Jiangsu KeyGen Biotech Company for in vivo tests.
4.5. Western blot analysis
Appendix A. Supplementary material
After the treatment with the indicated concentration of each sample
for 24 h under hypoxia, cells were collected and lysed in lysis buffer
(100 mM Tris-Cl, pH 6.8, 4% (m/v) sodium dodecylsulfonate, 20% (v/v)
glycerol, 200 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl
fluoride, and 1 g/mL aprotinin). Lysates were collected by centrifuga-
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmc.2021.116214.
References
◦
tion at 13000 rpm for 20 min at 4 C. Proteins from cell lysates were
separated on the SDS-PAGE and transferred onto polyvinylidine
difluoride (PVDF) membrane (Amersham Biosciences). The membrane
was blocked with PBST containing 5% non-fat dry milk for 1 h and
1 Thomas S, Harding MA, Smith SC, et al. CD24 is an effector of HIF-1–driven primary
tumor growth and metastasis. Cancer Res. 2012;72:5600–5612.
2 Bergers G, Hanahan D. Modes of resistance to antiangiogenic therapy. Nat Rev
Cancer. 2008;8:592–603.
further incubated with monoclonal anti-human HIF-1α antibody (Santa
3 Zeng W, Liu P, Pan W, et al. Hypoxia and hypoxia inducible factors in tumor
metabolism. Cancer Lett. 2015;356:263–267.
Cruz Biotechnology, USA) or anti-p53 antibody (Imgenex, USA) over-
night at 4 ◦C under gentle shaking. After that, the membrane was
incubated with the secondary antibody (1:2000) for 1 h at RT (25 ◦C).
Protein blots were detected with chemiluminescence reagent (Thermo
4 Brown JM, Wilson WR. Exploiting tumour hypoxia in cancer treatment. Nat Rev
Cancer. 2004;4:437–447.
5 Evans SM, Koch CJ. Prognostic significance of tumor oxygenation in humans. Cancer
Lett. 2003;195:1–16.
7

Z. Xu et al.
Bioorganic & Medicinal Chemistry 41 (2021) 116214
6 Chang Q, Jurisica I, Do T, et al. Hypoxia predicts aggressive growth and spontaneous
metastasis formation from orthotopically grown primary xenografts of human
pancreatic cancer. Cancer Res. 2011;71:3110–3120.
22 Jahanban-Esfahlan R, Guardia M, Ahmadi D, et al. Modulating tumor hypoxia by
nanomedicine for effective cancer therapy. J Cell Physiol. 2018;233:2019–2031.
23 Jain RK. Normalization of tumor vasculature: An emerging concept in antiangiogenic
therapy. Science. 2005;307:58–62.
7 Dewhirst MW, Cao Y, Moeller B. Cycling hypoxia and free radicals regulate
angiogenesis and radiotherapy response. Nat Rev Cancer. 2008;8:425–437.
8 Goel S, Ni D, Cai W. Harnessing the power of nanotechnology for enhanced radiation
therapy. ACS Nano. 2017;11:5233–5237.
24 Lucien F, Pelletier PP, Lavoie RR, et al. Hypoxia-induced mobilization of NHE6 to the
plasma membrane triggers endosome hyperacidification and chemoresistance. Nat
Commun. 2017;8:15884.
¨
9 Hockel M, Vaupel P. Tumor hypoxia: definitions and current clinical, biologic, and
25 Comerford KM, Wallace TJ, Karhausen J, et al. Hypoxia-inducible factor-1-
dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res. 2002;62:
3387–3394.
molecular aspects. J Natl Cancer Inst. 2001;93:266–276.
10 Harris AL. Hypoxia—a key regulatory factor in tumour growth. Nat Rev Cancer.
2002;2:38–47.
26 Eckel F, Schneider G, Schmid RM. Pancreatic cancer: a review of recent advances.
Expert Opin Invest Drugs. 2006;15:1395–1410.
11 Bruick RK, McKnight SL. A conserved family of prolyl-4-hydroxylases that modify
HIF. Science. 2001;294:1337–1340.
27 Moysan E, Bastiat G, Benoit JP. Gemcitabine versus modified gemcitabine: A review
of several promising chemical modifications. Mol Pharmaceutics. 2013;10:430–444.
28 Xiao J, Jin C, Liu Z, et al. The design, synthesis, and biological evaluation of novel
YC-1 derivatives as potent anti-hepatic fibrosis agents. Org Biomol Chem. 2015;13:
7257–7264.
12 Lia B, Sherb D, Kelly L, et al. Catalytic promiscuity in the biosynthesis of cyclic
peptide secondary metabolites in planktonic marine cyanobacteria. Proc Natl Acad
Sci USA. 2010;107:10430–10435.
13 Semenza G. Defining the role of hypoxia-inducible factor 1 in cancer biology and
therapeutics. Oncogene. 2010;29:625–634.
29 Sun HL, Liu YN, Huang YT, et al. YC-1 inhibits HIF-1 expression in prostate cancer
14 Koyasu S, Kobayashi M, Goto Y, et al. Regulatory mechanisms of hypoxia-inducible
factor 1 activity: Two decades of knowledge. Cancer Sci. 2018;109:560–571.
15 Aebersold DM, Burri P, Beer KT, et al. Expression of hypoxia-inducible factor-1a: A
novel predictive and prognostic parameter in the radiotherapy of oropharyngeal
cancer. Cancer Res. 2001;61:2911–2916.
cells: contribution of Akt/NF-κB signaling to HIF-1
α accumulation during hypoxia.
Oncogene. 2007;26:3941–3951.
30 Xu ZC, Zhao J, Gou SH, et al. Novel hypoxia-targeting Pt(IV) prodrugs. Chem
Commun. 2017;53:3749–3752.
31 Zhang B, Huang XC, Wang HS, et al. Promoting antitumor efficacy by suppressing
hypoxia via nano self-assembly of two irinotecan-based dual drug conjugates having
16 Schito L, Semenza GL. Hypoxia-inducible factors: Master regulators of cancer
progression. Trends Cancer. 2016;2:758–770.
a HIF-1α inhibitor. J Mater Chem B. 2019;7:5352–5362.
17 Moon E, Brizel D, Chi JTA, et al. The potential role of intrinsic hypoxia markers as
prognostic variables in cancer. Antioxid Redox Signaling. 2007;9:1237–1294.
18 Chen C, Lou T. Hypoxia inducible factors in hepatocellular carcinoma. OncoTarget.
2017;8:46691–46703.
32 Huang P, Wang DL, Su Y, et al. Combination of small molecule prodrug and
nanodrug delivery: Amphiphilic drug-drug conjugate for cancer therapy. J Am Chem
Soc. 2014;136:11748–11756.
33 Bazzanini R, Gouy MH, Peyrottes S, et al. Synthetic approaches to a mononucleotide
prodrug of cytarabine. Nucleosides Nucleotides Nucleic Acids. 2005;24:1635–1649.
34 Sk UH, Kambhampati SP, Mishra MK, et al. Enhancing the efficacy of Ara–C through
conjugation with PAMAM dendrimer and linear PEG: A comparative study.
Biomacromolecules. 2013;14:801–810.
19 Kim BW, Cho H, Chung JY, et al. Prognostic assessment of hypoxia and metabolic
markers in cervical cancer using automated digital image analysis of
immunohistochemistry. J Transl Med. 2013;11:185.
20 Kievit FM, Zhang M. Cancer nanotheranostics: Improving imaging and therapy by
targeted delivery across biological barriers. Adv Mater. 2011;23:H217–H247.
21 Qiu GZ, Jin MZ, Dai JX, et al. Reprogramming of the tumor in the hypoxic niche: The
emerging concept and associated therapeutic strategies. Trends Pharmacol Sci. 2017;
38:669–686.
35 Kastan MB. Wild-type p53: Tumors can’t stand it. Cell. 2007;128:837–840.
36 Levesque AA, Eastman A. p53-based cancer therapies: is defective p53 the Achilles
heel of the tumor? Carcinogenesis. 2007;28:13–20.
37 Lujambio A, Akkari L, Simon J, et al. Non-cell-autonomous tumor suppression by
p53. Cell. 2013;153:449–460.
8