New STAT3 inhibitor through biotransformation of celastrol by Streptomyces olivaceus CICC 23628

2.1 Biotransformation and metabolites identification

In this study, 400 mg of celastrol (CEL) was fed into Streptomyces olivaceus CICC 23628 cultures. After culturing for 4 days, the incubation supernatants were extracted with ethyl acetate three times. The extracts were concentrated and then separated by repeated column chromatography and semi-preparative HPLC to yield 2 metabolites, CEL-1 (51 mg) and CEL-2 (32 mg) (Scheme 1). Their structures were identified by HR‑ESI‑MS and NMR data analyses. The ¹H NMR and ¹³C NMR spectral data of the new metabolites are listed in Table 1.

Close

Scheme 1 Biotransformation of celastrol (CEL) by S. olivaceus.

Export to PPT

Metabolite CEL-1 was isolated as a red powder. The molecular formula of CEL-1 was established as C₂₉H₃₈O₅ by the [M + H]⁺ ion at m/z 467.2811 (calcd for C₂₉H₃₉O₅, 467.2831) in positive ion mode of HR-ESI-MS, indicating a 16 amu mass increase than CEL. The 1D NMR spectra showed a new OH-bearing methine carbon signal at δ_C 73.8 ppm and the disappearance of the methyl carbon signal compared with CEL. The new CH₂ proton signals at δ_H 3.66 (d, J = 10.5 Hz, 1H) and δ_H 3.48 (d, J = 10.4 Hz, 1H) had direct ¹H–¹³C correlations with δ_C 73.8 ppm in HSQC spectra. In the HMBC experiment, the new proton signal δ_H 3.66 show long-range ¹H–¹³C correlations with C-29 (δ_C 180.4), C-20 (δ_C 45.9), and C-21 (δ_C 24.9), and another new proton signal δ_H 3.48 show correlations with C-29 (δ_C 180.4), C-19 (δ_C 25.7), suggesting the hydroxylation occurred at C-30. Thus, CEL-1 was characterized as 3,30-dihydroxy-24-nor-2-oxo-1(10),3,5,7-friedelatetraen-29-oic acid. (Fig. 1).

Close

Fig. 1 Key HMBC and NOESY correlations of compound CEL-1 and CEL-2.

Export to PPT

Metabolite CEL-2 was isolated as a red powder. The HR-ESI-MS of CEL-2 showed a [M + Na]⁺ ion at m/z 461.2663 (calcd for C₂₈H₃₈NaO₄, 461.2665), indicating a 12 amu mass decrease than CEL. The ¹³C NMR and DEPT data of CEL-2 showed the presence of two new oxygenated carbon signals at δ_C 72.0 (quaternary carbon) and 71.2 (secondary carbon) ppm, along with the disappearance of the carboxyl carbon signal compared with the CEL data. The ¹H NMR spectra indicate that CEL-2 loses a methyl signal peak. The signals of carbon and proton in A-D rings remain roughly the same with the substrate, which reveals the structural changes of the E ring containing the carboxyl group. First of all, the new carbon signal at δ_C 72.0 was assigned to C-20 based on HMBC correlations between δ_H 1.82 (m, 1H, H-18), 1.16 (m, 1H, H-22a), 1.60 (m, 1H, H-22b) and δ_C 72.0 ppm. Combined with the absence of a carboxyl signal, it can be inferred that the decarboxylation reaction takes place in C-20 of the substrate, replaced by a new hydroxyl group. Secondly, the new CH₂ proton signal at δ_H 3.46 (d, 1H), which had direct ¹H–¹³C correlations with δ_C 71.2 ppm, showed long-range ¹H–¹³C correlations with C-20 (δ_C 72.0) in HMBC, indicated that δ_C 71.2 should be assigned to C-30. This new hydroxyl group at C-30 reflected the absence of a methyl signal. The HR-MS spectrum data also conform with the structural changes. The NOE effects between H-30 (δ_H 3.46) and α-CH₃-27 (δ_H 2.91) suggested the presence of C-20 hydroxyl group with a β-configuration. Therefore, CEL-2 was determined as a novel friedelanes nor-triterpene, which is named: 3,20β,30-trihydroxy-24,29-dinor-1(10),3,5,7-friedelatetraen-2-one. (Fig. 1).

2.2 Cytotoxic assay in vitro

The inhibition rates of all the derivatives and their parent compound were evaluated against three human cancer cell lines (A549, HCT-116, HepG2) by CCK-8 assay. According to the in vitro results (Table 2), the anti-proliferation effect of CEL-2 was higher than celastrol (CEL) in all of the screened cell lines, but CEL-1 showed no effect below the maximum test concentration (12.5 μM). Therefore, CEL-2, as a potential candidate compound, deserves further study.

2.3 Effects of CEL-2 on cell-cycle distribution and apoptosis

Celastrol has been reported to induce cell-cycle arrest and apoptosis in different tumor cell lines (Moreira et al., 2019, Yao et al., 2019, Zhong et al., 2019). CEL-2 as celastrol’s derivative was then selected to characterize its effect on cell-cycle distribution and apoptosis of HCT-116 cells to explore its anti-tumor mechanisms. As shown in Fig. 2A, CEL-2 could induce cell apoptosis in a concentration-dependent manner. Furthermore, CEL-2 also had a retardation effect on the cell-cycle distribution of HCT −116 compared with the control conditions (DMSO) (Fig. 7B).

Close

Fig. 2 Cell apoptosis (A) and Cell-cycle (B) analysis of CEL-2 on HCT-116. Cells were treated with CEL-2 at distinct concentrations (0.661, 1.650, 3.300, 5.000 μM) for 48 h, analyzed by flow cytometry. The figures were representative of three independent experiments.

Export to PPT

2.4 CEL-2 inhibited the phosphorylation of STAT3 at site pTyr705 and the downstream genes

There is more evidence showing that STAT3 activity has a critical role in carcinogenic processes by regulating the cell cycle and apoptosis (Shao et al., 2017). To investigate whether the anti-tumor effects of CEL-2 were related to the inhibition of the STAT3 pathway, we evaluated the level of p-STAT3 and STAT3 stimulated by IL-6 in HCT-116, as well as the expressions of STAT3-targeted genes (Survivin and Mcl-1). As observed in Fig. 3, CEL-2 effectively decreased the level of p-STAT3 at site pTyr705 in a concentration-dependent manner, and the inhibitory effect of CEL-2 was stronger than CEL at 0.75 μM. However, the total level of STAT3 was not changed with CEL-2 treatment. Meanwhile, CEL-2 down-regulated the level of Survivin and Mcl-1 as STAT3′s downstream genes (Fig. 4). These results indicated that CEL-2 could effectively inhibit the activity of STAT3, which may be the underlying mechanism that CEL-2 exerts its anti-colorectal tumor effect.

Close

Fig. 3 CEL-2 inhibited STAT3 phosphorylation at site pTyr705 in HCT-116. The HCT-116 were serum-starved overnight, then left untreated or treated with CEL (0.75 μM) and CEL-2 (0.5, 0.75, 1.0 μM) for 6 h, followed by stimulation by IL-6 (25 ng/mL). The cells were harvested at 30 min and analyzed by Western blot assays. Compared with DMSO (control) group, *P＜0.05; Compared with IL-6 (model) group, ^#P＜0.05, ^##P＜0.01.

Export to PPT

Close

Fig. 4 CEL-2 inhibited STAT3′s downstream genes (Survivin and MCL-1) in HCT-116.

Export to PPT

2.5 CEL-2 had no effects on STAT3′s typical upstream kinases JAK2

As a signaling pathway closely related to the occurrence and development of tumors, the phosphorylation of STAT3 is mediated by JAK2, one of the typical STAT3 upstream tyrosine kinases (Mengie Ayele et al., 2022). A corresponding western blot was carried out to evaluate whether the inhibitory effect of CEL-2 on p-STAT3 is regulated by upstream JAK2. It was evident in Fig. 5 that CEL-2 had no effects on both the levels of p-JAK2 and total JAK2, illustrating that CEL-2 might inhibit p-STAT3 by directly targeting STAT3 protein.

Close

Fig. 5 CEL-2 did not affect STAT3 upstream tyrosine kinase (JAK2) in HCT-116.

Export to PPT

2.6 SPR analysis of CEL-2 to recombinant human STAT3 protein

Surface plasmon resonance (SPR) analysis was used to reveal the direct interaction between CEL and STAT3 (Ye et al., 2020, Xu et al., 2022). In order to test the hypothesis that CEL-2 inhibits the phosphorylation of STAT3 by targeting STAT3 directly, we conducted SPR analysis to evaluate the interaction at the molecular level and used CEL as a positive control. We observed that CEL-2 could also interact with the recombinant human STAT3 protein and the response unit (RU) values were proportional to compound concentrations within the selected ranges (Fig. 6A). However, CEL-1 had no binding affinity with rhSTAT3 in the SPR experiment. According to the fitting calculation results, CEL-2 showed a slightly stronger binding affinity with a lower K_D value (the equilibrium dissociation constant) of 40.96 µM, when compared with CEL (K_D = 60.38 µM) (Fig. 6B).

Close

Fig. 6 SPR analysis of CEL-2 (A) and CEL (B) with the recombinant human STAT3 protein. Compound concentrations varied from 3.91 to 250 µM. Time is measured in seconds (s).

Export to PPT

2.7 Effects of CEL-2 on colorectal cancer organoids

Organoid model, which simulates the growth morphology of solid tumors, was emerging as a better drug screening platform than conventional established 2D cell lines (Kondo and Inoue 2019). Four colorectal cancer organoid cultures (CCO-1, CCO-2, CCO-3, and CCO-4) from humans were used to investigate the activity of CEL-2 further and assess its potential in clinical application. The inhibitory effect of CEL-2 on human colorectal cancer organoids (CCOs) was evaluated. The commonly used drug oxaliplatin (l-OHP) in clinical was chosen as a control. Their cytotoxicity was also tested on colorectal normal organoids (CNO). As the results show, CEL-2 has a much more robust activity than the positive drug l-OHP on all the colorectal cancer organoids (Fig. 7, Table 3). More importantly, CEL-2 showed lower toxicity to CNO. While l-OHP gradually emerges drug resistance in the clinical application (Martinez-Balibrea et al., 2015), CEL-2 could be used as an alternative drug candidate.

Close

Fig. 7 The inhibition effect of compound CEL-2 on human colorectal cancer organoids (CCOs) and colorectal normal organoid (CNO) was evaluated. Organoids were treated with CEL-2 and oxaliplatin (l-OHP, positive drug) at different concentrations for 6 days and then cell viability was determined utilizing CCK8. (A) Representative microscopy images of organoids incubated with CEL-2 at the indicated concentrations for 6 days. And the dose-dependent effect of CEL-2 on organoids viability. (B) Representative microscopy images of organoids incubated with l-OHP at the indicated concentrations for 6 days. And the dose-dependent effect of l-OHP on organoids viability. Scale bar, 100 μm.

Export to PPT

4.1 General procedures

NMR spectra were recorded on a Bruker AV-600 spectrometer in C₅D₅N solution with TMS as the internal standard and chemical shifts were expressed in δ (parts per million). High-resolution mass spectra (HR-MS) were obtained using a Water Q-Tof micro mass spectrometry. All the solvents used for extraction and isolation were of analytical grade. HPTLC was performed on precoated silica gel GF₂₅₄ plates. Separation and purification were carried out by column chromatography on silica gel (200–300 mesh) and preparative HPLC (Essentia Prep LC-16P, SHIMADZU, Japan). Silica gel was purchased from Qingdao Marine Chemical Group Co., PR China.

Celastrol was purchased from Nanjing jingzhu bio-technology Co.,ltd, with over 98 % purity. Streptomyces olivaceus CICC 23628, Rhizopus chinensis CICC 40335, and Penicilium griseofulvum CICC 40,293 were purchased from the China Center of Industrial Culture Collection, China. Bacillus megaterium CGMCC 1.1741 and Streptomyces griseus ATCC 13,273 were obtained from courtesy of Prof. Jian Zhang of China Pharmaceutical University.

4.2 Analytical and preparative scale biotransformation, isolation, and identification of products

Cultures were grown by a two-stage procedure in 50 mL of soybean meal glucose medium held in 250 mL culture flasks. The soybean meal glucose medium contained (in g/L) 20 glucose, 5 yeast extract, 5 soybean meal, 5 NaCl, and 5 K₂HPO₄ in distilled water and was adjusted to pH 7.0 with 6 N HCl before being autoclaved at 121 °C for 15 min. Cultures were incubated with shaking at 180 rpm at 28 °C. A 10 % inoculum derived from 48-h-old stage I cultures was used to initiate stage II cultures, which were incubated for 24 h before receiving 10 mg of substrates in 1 mL of ethanol, and incubations were conducted as before. Substrate controls consisted of a sterile medium and substrates incubated under the same conditions but without microorganisms. Cultures were incubated for 5 days and extracted with an equal volume of EtOAc. The organic phase was concentrated and spotted on silica gel HPTLC plate, which was developed by chloroform/methanol (10:1, v/v). The results were visualized by spraying with H₂SO₄ (10 % in ethanol) and heating at 120 °C for 1–2 min.

Using 24-h-old stage II cultures, substrate (300 mg) was distributed evenly among thirty 250 mL culture flasks. Substrate-containing cultures were incubated for 5 days and then extracted with equal EtOAc three times. The organic solvent layer was evaporated to dryness. The crude extracts were subjected to silica gel column chromatography eluted with a solvent system of chloroform/methanol (99:1 to 95:5) to afford Fraction 1 and 2, respectively. Fraction 1 was purified by preparative HPLC with acetonitrile: H₂O = 45:55, 3 mL/min. Metabolites CEL-1 (51 mg) were isolated. Fraction 2 was purified again by silica gel column chromatography eluted with a solvent system of chloroform/methanol (95:5 to 90:10) to afford CEL-2 (32 mg). The purity of CEL-1 and CEL-2 are determined by HPLC-ELSD, both of which are higher than 98 %. The structures were identified based on their MS and NMR.

4.3 Reagents for pharmacological experiments

DMEM-high glucose, RPMI 1640 and PBS were purchased from HyClone. Penicillin-Streptomycin (100 × ) (P/S) solution, 0.25 % trypsin, and Fetal Bovine Serum (FBS) were purchased from Gibco. DMSO was obtained from Sigma. Cell Counting Kit-8 (CCK-8), 20 × TBS, and PMSF were purchased from Meilunbio®. SDS-PAGE Running Buffer, RIPA Lysis buffer, Western Transfer Buffer, 1 M Tris-HCL, 10 % SDS, 30 % Acr-Bis, TEMED, BCA kit, Cell-cycle and Apoptosis Kit were purchased from Beyotime. Super ECL Detection Reagents were purchased from US Everbright® Inc. Prestained Protein Ladder was purchased from Thermo Scientific. Recombinant Human IL-6 (C009) was purchased from Novoprotein. β-actin (C4) sc-47778 was obtained from SANTA. HRP Goat anti-Mouse IgG was obtained from ImmunoWay. HRP Goat anti-Rabbit IgG was obtained from LABLEAD. STAT3 (79D7) Rabbit mAb (4904 T) and p-STAT3 (Y705) Rabbit mAb (9145 T) were obtained from CST. Survivin (EP2880Y) Rabbit mAb (ab76424), MCL1 (Y37) Rabbit mAb (ab32087), JAK2 (EPR108 (2)) Rabbit mAb (ab108596) and p-JAK2 (E132) Rabbit mAb (ab32101) were obtained from Abcam. Reagents and buffer solution for SPR were obtained from GE Healthcare. Recombinant human STAT3 proteins were obtained from Cusabio.

4.4 Cytotoxic assay in vitro

The cytotoxic activities of compounds were evaluated against A549, HCT-116 (cultured in RPMI 1640), and HepG2 (cultured in DMEM-high glucose) by the CCK-8 method. For this assay, 100 μL (5 × 10³/mL) cells per well were seeded in 96-well plates and allowed to incubate for 24 h. Then, the compounds with different concentrations (0.02、0.1、0.5、2.5、12.5 μM) were added. After 24 h of incubation, CCK-8 solution (10 %) was added, and the plates were incubated again for another 4 h at 37 ℃. The OD values at 450 nm were immediately read by a microplate reader (Tecan Infinite M200 Pro). Subsequently, the IC₅₀ values were calculated by Graphpad Prism 5. Three independent experiments were performed. Data are presented as the mean ± SD (n = 3).

4.5 Western blot analysis

HCT-116 cells were seeded (1 × 10⁷/mL, 2 mL/orifice) in 6-well plates and incubated for 24 h. The cells were serum-starved. Then celastrol or CEL-2 with specific concentrations was added for 6 h incubation. After 6 h, cells were stimulated by IL-6 (25 ng/mL). The cells were harvested after 30 min, and then the proteins were extracted with lysis buffer and quantified by the BCA method. Each sample was separated on 10 % SDS-PAGE and transferred to a PVDF membrane (Millipore). The membranes were blocked and sequentially incubated with primary and secondary antibodies diluted in 5 % bovine serum albumin (BioSharp). Subsequently, the bands were detected in a gel imaging analysis system ((Bio-Rad. ChemiDoc XRS).

4.6 SPR analysis

Purified rhSTAT3 protein (optimum pH = 5.5) was immobilized on sensor chip CM5 (GE Healthcare) and carried out at 25 ℃ on Biacore T200 instruments (GE Healthcare). Approximately 14,937 RU of STAT3 was amino coupled to a CM5 Chip (according to the manufacturer’s protocol), and another cell was left blank for reference subtraction. HBS-EP buffer was used as the running buffer. The test compounds (10 mM, DMSO) were diluted to 3.91 μM, 7.81 μM, 15.63 μM, 31.25 μM, 62.5 μM, 125 μM and 250 μM using PBS buffer with 5 % DMSO. The operating conditions: contact time: 80 s, flow rate: 30 μL/min, dissociation time: 180 s. The ratio of the association and dissociation rate constants was determined as the affinity (equilibrium constants, K_D). The K_D (equilibrium constant) values were calculated as k_d (dissociation rate constant)/k_a (associated rate constant) for each interaction and determined by globally fitting with Biacore T200 evaluation software 2.0 (GE Healthcare).

4.7 Cell-cycle and apoptosis assay

HCT-116 cells (log phases) were seeded in 6-well plates and preincubated at 37 ℃ for 24 h. Then cells were further incubated with DMSO (blank control) or CEL-2 at different concentrations for 48 h. After the respective period of incubation, cells were harvested softly and washed twice with cold PBS and collected by centrifugation. For apoptosis analysis, after centrifugation, the cells were resuspended with 1 × binding buffer solution and partial shipments to 100 μL. Then PI and Annexin V-FITC were added, and the mixture was incubated at room temperature for 15 min in the darkness. Finally, the mixture was diluted with 1 × binding buffer and analyzed by flow cytometry (Agilent Technologies, NovoCyte 3000) at 488 nm. NovoExpress software was used to analyze and generate the double dispersion point diagram. For cell-cycle analysis, the cells were fixed in 70 % ethanol at 4 ℃ overnight.

After washing with PBS, cells were suspended in PBS containing 50 mg/mL PI and 100 mg/mL RNase A, incubated at 37 ℃ for 30 min, and protected from the light. Then the fluorescence intensity was measured at 488 nm by flow cytometry.

4.8 Activity verification on colorectal cancer organoids