Introduction
Hypoxia, characterized by decreased intracellular oxygen levels, is associated with
the development of malignancy, and resistance to radiation and chemotherapy, and predicts
poor outcomes in various tumor types.[1 ]
[2 ] Hypoxia-inducible factors (HIFs), the mast regulator in the response to hypoxia,
consist of three inducible α-subunits (HIF-1α, HIF-2α, and HIF-3α) and one constitutively
expressed β-subunit (HIF-1β or arylhydrocarbon receptor nuclear translocator).[3 ]
[4 ]
[5 ] Under normal oxygen conditions, HIF-α subunits undergo rapid hydroxylation at specific
proline residues by HIF-prolyl hydroxylases, which was subsequently degraded via the
von Hippel–Lindau tumor suppressor gene product (pVHL)-mediated ubiquitin–proteasomal
pathway.[6 ]
[7 ] Upon hypoxic stabilization, HIF-α accumulates and forms transcriptional complexes
with HIF-1β translocates to the nucleus to bind to hypoxia response elements (HREs)
and induce the expression of downstream genes involved in cell survival, proliferation,
angiogenesis, and metabolic reprogramming.[8 ]
[9 ] The HIF pathway is a positive regulator of the malignant phenotype and regulates
multiple aspects of tumorigenesis, making it a critical target for antitumor therapy.
In the tumor hypoxic microenvironment, HIF plays a critical role in enhancing glucose
metabolism and vascular endothelial growth factor (VEGF) expression to facilitate
angiogenesis, enabling cells to adapt to low oxygen levels. HIF signaling promotes
angiogenesis by inducing VEGF transcription, facilitating endothelial cell migration
to hypoxic regions.[10 ] Thus, targeting the HIF–VEGF axis emerged as a promising therapeutic strategy for
cancer treatment, with significant efforts focused on developing HIF inhibitors. PT2399,
also known as belzutifan and MK-6482, is a small-molecule inhibitor that binds directly
to HIF-2α and disrupts its interaction with HIF-1β has been approved for treating
patients with VHL disease associated with renal cell carcinoma or pancreatic neuroendocrine
tumors since 2022.[11 ]
[12 ] Antitumor drugs topotecan suppressed HIF-1α and resulted in a decrease in VEGF expression
in both in vivo and in vitro assays.[13 ] The proteasome inhibitor anticancer drug bortezomib (PS-341) was reported to repress
HIF-1α protein expression and nuclear accumulation by inhibiting both PI3K/Akt/mTOR
and MAPK pathways in prostate cancer.[14 ]
[15 ]
[16 ] Additionally, various compounds, including acriflavine, (E )-phenoxyacrylic amide derivatives, chalcone derivatives, and benzofuran derivatives,
have been identified as potential HIF inhibitors with antiangiogenic properties, offering
potential for the development of novel cancer therapeutics.[17 ]
[18 ]
[19 ]
[20 ] Herein, we screened our compound collections and identified HST3782, a 3-hydroxy-8-azabicyclo[3.2.1]octane
bridged compound, as a novel HIF inhibitor with interesting scaffold. HST3782 effectively
suppressed HIF-targeted gene expression, including VEGFa and VEGFR, and potently inhibited
the angiogenesis and vasculature in zebrafish models, demonstrating its potential
for further development.
Results and Discussion
To identify novel scaffolds targeting the HIF pathway, we developed a dual-luciferase
screening assay using our compound collection. Initially, primary clear cell adenocarcinoma
cell line 786-O cells that harbored mutated VHL-1 and exhibited constituted active
HIF-2α signals were transiently transfected with the HER-Luc2 vector and renilla control
vector. The collection of about 100 molecules was screened with a concentration of
10 μmol/L and the luciferase signals were measured 48 hours after drug incubation.
The HRE luciferase signals were normalized with the control renilla signals. Among
the compounds tested, HST3782 emerged as a potent inhibitor at 10 μmol/L. Subsequently,
it was tested in a multiple-dosed assay along with PT2385, a validated and commercially
available HIF-2α inhibitor.[21 ] The result showed that HST3782 dose-dependently inhibits HIF signaling with an IC50 of 1.028 μmol/L.
Targeting the HIF pathway represents a potential new strategy for refractory triple-negative
breast cancer therapy.[22 ] Next, we assessed the effects of HST3782 on the HIF pathway in the triple-negative
breast cancer cell line SUM159 ([Fig. 1A ]). These cells were transfected with HRE-Luc2 and renilla vectors under both normoxic
(21% O2 ) and hypoxic (1% O2 ) conditions.[22 ] Consistent with previous reports, hypoxia-activated the HIF pathway, resulting in
a significant increase in HRE reporter signals. Treatment with 10 and 20 μmol/L HST3782
led to a reduction in HIF signaling by 74 and 80%, respectively; whereas the HIF signal
inhibitor acriflavine at 5 μmol/L suppressed HIF signaling by 90%.
Fig. 1 Screening and the characterization of HST3782 as a HIF signaling inhibitor. (A ) The inhibition of HRE dual-luciferase reporters of HST3782 in SUM159 cells. (B ) The downregulation of HIF-2α regulated gene in 786-O cells. Data are expressed as
the mean ± standard deviation (n = 10). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 versus the control group (cells treated with the same volume of DMSO instead
of a drug), analyzed by two-tailed Student's t- test. ACF, acriflavine.
After demonstrating the inhibitory effect of HST3782 against the HIF reporter, a quantitative
real-time PCR was conducted to assess the expression of HIF downstream genes in 786-O
cells with additional HST3782 treatment. The result showed that 10 μmol/L HST3782
effectively inhibited the expression of VEGFa, VEGFR-1, HIF-2α, BNIP3, EDM1, and Serpine
1 after 24 hours of administration, with a further increase in potency observed at
20 μmol/L ([Fig. 1B ]). Concurrently, the reference compound PT2385 also repressed the expression of these
genes in the same experiment, thereby validating the assay conditions. Angiogenesis
is primarily mediated by VEGF signaling, initiated when VEGF ligands bind to their
cognate membrane-bound receptors (VEGFR) on the surface of endothelial cells.[23 ] HST3782 downregulated HIF-targeted genes, suggesting its potential to inhibit angiogenesis.[24 ]
[25 ]
Next, we explored a synthetic route of HST3782 as illustrated in [Scheme 1 ]. The phenyl-1,3,4-triazole core (5 ) was generated in a diazonium reaction starting with 2,6-dichloroaniline (1 ) and methyl-2-chloro-2-(2-(2,6-dichlorophenyl)hydrazineylidene)acetate (2 ), followed by cyclization in the presence of freshly prepared isobutyraldehyde oxime
(4 ). The methyl ester group in 5 was hydrolyzed by sodium hydroxide solution to yield the carboxylic acid 6 . It was then condensed with 8-azabicyclo[3.2.1]octan-3-one, leading to the amide
compound 7 , the ketone group of which was reduced to give the final products, HST3782 and 8b , as a pair of endo–exo isomers, which can be easily separated by chromatography ([Fig. 2 ]). The conformations of isomers were confirmed through two-dimensional nuclear overhauser
effect (2-D NOE) analysis of 2D 1H-1H nuclear overhauser effect spectroscopy (NOESY).
It was observed that the H26 hydrogen atom of HST3782 with a chemical shift of around 3.93 ppm, is coupled to
the four hydrogens on C22 and C23 (chemical shifts of around 1.73 (2H) and 1.97 (2H)) and did not show a NOE correlation
with hydrogens on C18 and C19 . This indicates that the configuration of HST3782 is endo. In contrast, the 2D-NOESY
of exo isomer, 8b , displayed NOE correlations between H25 (chemical shift = 3.98) and hydrogens on C18 and C19 (chemical shift = 1.5). In addition, the 1 H NMR (nuclear magnetic resonance spectra) of HST3782 and 8b exhibited differences in the chemical shift and coupling constant of hydrogen on
the bridged ring ([Fig. 2C ]).
Scheme 1 The synthetic route of HST3782 and its isomer. Reagents and conditions: (a ) NaNO2 , HCl, MeOH, r.t., 12 hours; (b ) NH2 OH•HCl, toluene, r.t., 30 minutes; (c ) TEA, toluene, 120°C, 2 hours, 42.1%; (d ) 2 mol/L NaOH, MeOH/H2 O, r.t., 2 hours, 92.0%; (e ) TBTU, DIPEA, DCM, r.t., 12 hours, 50.1%; (f ) NaBH4 , MeOH, r.t., 2 hours, 17.1%.
Fig. 2 (A ) Structure, (B ) 2D-NOESY spectra, and (C ) 1 H NMR spectra of HST3782 and 8b .
Next, the in vivo efficacy of HST3782 against angiogenesis was evaluated in the well-established zebrafish
model. AB-strain zebrafish were bred and fertilized eggs were collected. Healthy zebrafish
embryos at 6-hour postfertilization (hpf) were carefully selected under a dissecting
microscope and distributed into 24-well plates, with 10 embryos per well. In multidosed
toxicity evaluation, both HST3782 and compound 8b showed no obvious toxicity ([Fig. 3A ]), and the mortality rates after 96 hours of drug exposure were less than 10%, even
at a maximum concentration of 500 μmol/L. The malformation rates were assessed from
24 to 96 hours, and concentration-dependent and time-cumulative effects were illustrated
([Fig. 3B ]). Subsequent angiogenesis experiments were conducted at concentrations of 31.25,
62.5, and 125 μmol/L, with the highest experimental concentration representing 1/10
to 1/3 of the EC50 value that does not cause malformation. Subsequently, the intersegmental vessels
(ISV) number was counted for the blank control group and drug-treated groups at 48
hpf, corresponding to the 24-hour drug administration period. At this stage, only
the ISV can be counted, while the subintestinal vessels (SIV) have not yet developed.
As illustrated in [Fig. 3C ], HST3782 inhibited the dorsal ISV development by 12 and 14%, respectively, at 62.5
and 125 μmol/L; whereas 8b showed no significant effect. At 48 hours, the areas of SIV were analyzed. As shown
in [Fig. 3D ], HST3782 significantly and dosed-dependently inhibited abdominal SIV development,
with inhibition rates being 26, 45, and 56%, respectively, at 31.25, 62.5, and 125
μmol/L, whereas its exo isomer 8b exhibited reduced inhibitory effects. Both the results of zebrafish ISV and SIV development
demonstrated that HST3782 exhibited inhibitory effects on zebrafish vasculature and
the exo isomer 8b has reduced activity.
Fig. 3 Anti-angiogenetic efficacy of HST3782 in zebrafish models. (A ) The representative image of vascular suppression in the control group, 8b or HST3782 at the concentration of 31.25, 62.5, and 125 μmol/L, at the indicated
time, respectively (magnification: 4 ×). The yellow dashed box represents the position
of ISV and SIV. The malformation rate (B ), ISV number (C ), and SIV area (D ) were analyzed at the indicated time. Data are expressed as the mean ± standard deviation
(n = 10). *p < 0.05, **p < 0.01, ***p < 0.001 versus the DMSO control group, analyzed by two-tailed Student's t- test. ISV, intersegmental vessels; SIV, subintestinal vessels.
Experimental Section
General Conditions
All reagents and solvents were used directly as purchased from commercial sources.
Flash chromatography was performed using silica gel (200–300 mesh). All reactions
were monitored by thin-layer chromatography, using silica gel plates with fluorescence
F254 and UV light visualization. 1 H NMR and 13 C NMR spectra were recorded on a Bruker AV-400 spectrometer (Bruker, Billerica, United
States). Coupling constants (J ) are expressed in hertz (Hz). Chemical shifts (δ) of NMR are reported in parts per
million (ppm) units relative to an internal control (trimethylsilane). Electrospray
ionization mass spectrometry were recorded on an Agilent 1200 HPLC-MSD mass spectrometer
(Agilent, Santa Clara, United States).
Synthesis of Methyl 1-(2,6-dichlorophenyl)-5-isopropyl-1H -1,2,4-triazole-3-carboxylate (Compound 5)
2,6-dichloroaniline (1.4 g, 9 mmol, 1 equiv.) was dissolved in MeOH (8 mL). Sodium
nitrite aqueous solution (1.2 g, 18 mmol, 2 equiv.) was added slowly. The resulting
mixture was stirred for 15 minutes and adjust the pH to 3-4 with sodium acetate. Methyl
2-chloro-3-oxobutanoate (1.1 mL, 9 mmol, 1 equiv.) was added. The mixture was continuously
stirred for 12 hours at room temperature, quenched with water, and then extracted
with ethyl acetate three times. The combined organic layers were washed with saturated
NaCl solution and concentrated to give compound 2 as a yellow solid, which was used directly without further purification.
To a solution of isobutyraldehyde (0.6 g, 9 mmol, 1 equiv.) in toluene (30 mL) was
added hydroxylamine hydrochloride (0.6 g, 9.9 mmol, 1 equiv.). The mixture was stirred
at room temperature for 30 minutes. Compounds 2 and Et3 N (2.5 mL) were added. The reaction was refluxed at 120°C for 2 hours. After completion
of the reaction, toluene was removed by vacuum rotary evaporation. The residue was
washed with saturated NaCl solution, extracted with EtOAc, and purified by column
chromatography (ethyl acetate:petroleum ether = 1:5) to give compound 5 (120 mg, 42.1%) as a yellow solid.
Synthesis of 1-(2,6-dichlorophenyl)-5-isopropyl-1H -1,2,4-triazole-3-carboxylic Acid (Compound 6)
Compound 5 (500 mg, 1.5 mmol) was dissolved in MeOH, and sodium hydroxide aqueous solution (2 mol/L,
20 mL) was added. The mixture was stirred at room temperature for 2 hours. After completion
of the reaction, the pH was adjusted to 3 to 4 with hydrochloric acid until a solid
precipitate appeared, which was isolated by suction filtration, giving compound 6 (437 mg, 92.0%) as a white solid.
Synthesis of 8-(1-(2,6-dichlorophenyl)-5-isopropyl-1H -1,2,4-triazole-3-carbonyl)-8-azabicyclo[3.2.1]octan-3-one (Compound 7)
Compound 6 (400 mg, 1.3 mmol, 1 equiv.) was dissolved in dichloromethane at 0°C, followed by
the addition of TBTU (429 mg, 1.3 mmol, 1 equiv.) and DIPEA (N, N-Diisopropylethylamine;
1.4 mL, 6 equiv.). The mixture was stirred at room temperature for 12 hours, washed
with an ammonium chloride aqueous solution upon completion of the reaction, and extracted
with dichloromethane. The organic layers were concentrated. The residue was purified
by column chromatography (ethyl acetate:petroleum ether = 1:2) to give compound 7 (271 mg, 50.1%) as a white solid.
Synthesis of HST3782 and 8b
Compound 7 (200 mg, 0.5 mmol, 1 equiv.) was dissolved in MeOH. Sodium borohydride was added
slowly. The reaction was stirred at room temperature for 2 hours and quenched with
ammonium chloride. The mixture was extracted with EtOAc. The organic layers were concentrated.
The residue was purified by column chromatography (ethyl acetate:petroleum ether = 1:2)
to give HST3782 and 8b , as a pair of endo–exo isomers.
HST3782, chemically named “endo-(1-(2,6-dichlorophenyl)-5-isopropyl-1H -1,2,4-triazol-3-yl)(3-hydroxy-8-azabicyclo[3.2.1]octan-8-yl)methanone.” 1 H NMR (600 MHz, dimethylsulfoxide [DMSO]-d
6 ) δ 7.83 (d, J = 8.2 Hz, 2H), 7.76–7.70 (m, 1H), 4.68 (d, J = 2.5 Hz, 1H), 4.61 (dt, J = 6.8, 3.0 Hz, 1H), 4.56 (dt, J = 6.7, 3.0 Hz, 1H), 3.98–3.95 (m, 1H), 2.71 (p, J = 6.8 Hz, 1H), 2.28–2.20 (m, 2H), 2.00 (ddt, J = 13.7, 8.7, 4.3 Hz, 2H), 1.91–1.79 (m, 2H), 1.78–1.70 (m, 2H), 1.21 (dd, J = 8.0, 6.8 Hz, 6H). 13 C NMR (151 MHz, DMSO) δ 162.98, 158.19, 156.63, 133.90 (2C), 132.11, 130.07 (2C),
63.34, 55.43, 51.63, 40.70, 38.80, 28.77, 26.99, 26.20, 21.39, 21.34 (2C). MS-ESI
(m /z ) calcd. for C19 H23 Cl2 N4 O2 [M + H]+ 409.1120, found: 409.1237.
8b , chemically named “exo-(1-(2,6-dichlorophenyl)-5-isopropyl-1H -1,2,4-triazol-3-yl)(3-hydroxy-8-azabicyclo[3.2.1]octan-8-yl)methanone.” 1 H NMR (600 MHz, DMSO-d
6 ) δ 7.84 (d, J = 8.2 Hz, 2H), 7.73 (dd, J = 8.7, 7.7 Hz, 1H), 4.63 (h, J = 3.0 Hz, 2H), 4.60 (d, J = 6.1 Hz, 1H), 3.99 (tq, J = 11.4, 6.0 Hz, 1H), 2.72 (hept, J = 6.8 Hz, 1H), 1.97-1.81 (m, 4H), 1.73 (qd, J = 8.7, 7.5, 2.8 Hz, 2H), 1.56-1.47 (m, 2H), 1.22 (t, J = 6.8 Hz, 6H). 13 C NMR (151 MHz, DMSO) δ 163.04, 158.14, 156.88, 133.88, 133.83 (2C), 132.09, 130.08
(2C), 62.24, 55.40, 51.93, 42.75, 40.88, 28.75, 27.00, 26.21, 21.40, 21.36 (2C). MS-ESI
(m /z ) calcd. for C19 H23 Cl2 N4 O2 [M + H]+ 409.1120, found: 409.1197.
Hypoxia Response Element Dual Luciferase Reporter Assay in 786-O cells
786-O cells (ATCC, Madison, United States) were seeded in 96-well plates at a density
of 7,000 cells/well and transfected when they reached 80 to 85% confluency. The transfections
use 0.5 μg HRE-Luc2 vector (Promega, #E4001, Madison, United States) mixed with 0.05
μg Renilla-pGL DNA (Promega, #E1741, Madison, United States) and 0.825 μL lipofectamine
2000 transfection reagent (Thermo Fisher Scientific, #11668030, Agawam, United States)
in a 10 μL Opti-MEM medium per well. The mixtures were cultured for 20 minutes. The
complexes (10 μL) were directly added to cells in a reduced serum medium. Approximately
6 hours of posttransfection, PT2385 (TargetMol Chemicals, #T7848, Shanghai, China)
or HST3782 was added. At 24 hours of posttransfection, a dual luciferase reporter
assay kit (Promega #E1960, Madison, United States) was used to determine the luciferase
activity of the cells according to the manufacturer's introduction to assess the activation
of HIF signaling.
Hypoxia Response Element Dual Luciferase Reporter Assay in SUM159 cells
SUM159 cells (kindly provided by Prof. Haiquan Lv from Shandong University) were stably
transfected with reporter plasmid pHRE-SV-Firefly and control plasmid pSV-Renilla
using PolyJet (SignaGen, Rockville, United States). 1 × 104 transfected SUM159 cells were seeded in each well of 96-well plate and cultured under
21% or 1% O2 in the presence of indicated doses of HST3782, acriflavine (TargetMol Chemicals,
T19832, Shanghai, China), or DMSO (as control) for 3 days. The ratio of Firefly/Renilla
luciferase activity was determined by using the dual luciferase reporter assay kit
on Centro LB960 microplate luminometer (Berthold, Ringtunveien, Norway).
RNA Isolation and Quantitative Real-Time Polymerase Chain Reaction
786-O cells (ATCC, Madison, United States) were seeded in 12-well plates, treated
with PT2385 (10 μmol/L), HST3782 (0 μmol/L), or HST3782 (20 μmol/L) in a reduced serum
medium for 24 hours, and then washed with phosphate buffer saline. Total RNA was isolated
by Trizol (Beyotime, #R0016, Shanghai, China). RT-PCRs were performed in triplicate
using standard SYBR green reagents. The primer sequences used for quantitative PCR
are as follows: erythropoietin (EPO)_fwd, AACAATCACTGCTGACACTT; EPO_rev, AGAGTTGCTCTCTGGACAGT;
VEGFa_fwd, AGGGCAGAATCATCACGAAGT; VEGFa_rev, AGGGTCTCGATTGGATGGCA; Serpine1_fwd, ACCGCAACGTGGTTTTCTCA;
Serpine1_rev, TTGAATCCCATAGCTGCTTGAAT; HIF-2α_fwd, CGGAGGTGTTCTATGAGCTGG; HIF-2α_rev,
AGCTTGTGTGTTCGCAGGAA. VEGFR-1_fwd, TTTGCCTGAAATGGTGAGTAAGG; VEGFR-1_rev, TGGTTTGCTTGAGCTGTGTTC;
BNIP3_fwd, CAGGGCTCCTGGGTAGAACT; BNIP3_rev, CTACTCCGTCCAGACTCATGC; EDM1_fwd, GATCACGTTCCTGAAAAACACG;
EDM1_rev, GCTCTCCGTCTGGATGCAG.
Antiangiogenesis in Zebrafish
Transgenic zebrafish embryos with the background of the AB strain expressing Tg (Fli:
EGFP) were used in the experiment according to a reported study.[27 ] Healthy zebrafish embryos at 6 hpf were selected under a dissecting microscope and
placed in 24-well plates at a density of 10 embryos per well. For embryo toxicity
testing, the experimental groups included a DMSO control group and six working concentrations
(15.6, 31.25, 62.5, 125, 250, and 500 μmol/L) for each compound. Drug administration
started at 24 hpf. Mortality and deformity rates of embryos were recorded at 24, 48,
72, and 96 hours, respectively. For vasculature development evaluation, the experimental
groups included a solvent control group (3.2% DMSO) and three working concentrations
(31.25, 62.5, and 125 μmol/L) for each compound, with three replicates for each concentration
gradient. After 24 and 48 hours of exposure, zebrafish were captured with a fluorescence
microscopy system (Nikon Ci-S), and the number of ISV and area of SIV were measured.
Data Analysis
The results are represented as means with respective standard errors. Data analysis
was performed with GraphPad Prism Software (GraphPad Software Inc., version 8). Student's
t -test or one-way analysis of variance with Brown–Forsythe was utilized for statistical
analysis. Significant differences were indicated by asterisks considering p -values < 0.05.