M. oleifera seeds (Yunnan Tianyou Technology Development Co., Ltd; Dehong, Yunnan, China) were hulled, crushed, and passed through a 30-mesh sieve. The resulting M. oleifera seed powder was processed using a Soxhlet extractor. The powder was sealed, petroleum ether was added for deoiling, and the product was heated to reflux for more than 12 h until deoiling was complete. Ten grams of dry deoiled M. oleifera seed powder was added to a glass beaker, purified water was added at a solvent ratio of 1:60 (g/ml), the pH was adjusted to 5 with acetic acid, and then the powder was placed in a water bath at 30°C with mechanical stirring for 9 h, and finally centrifuged (4,000 rpm, 5 min) to collect the supernatant. The supernatant was extracted three times with 1:3 volume of dichloromethane, all fractions were combined, and then concentrated under reduced pressure with a rotary evaporator at 30°C, and protected from light until a white solid was produced. The white solid was subjected to recrystallization with dichloromethane. After five recrystallizations, MIC-1 with a purity of 98% was obtained, and the total yield was 4.70%. The resulting MIC-1 was analyzed by ¹H-¹³C nuclear magnetic resonance spectroscopy and high-resolution mass spectrometry. The data obtained by the analysis were consistent with those previously reported (Gueyrard et al., 2010) and the structure of MIC-1 was confirmed to be correct (Figure 1A).

Thirty-four cancer cell lines (LoVo, PC-12, A375, 786-O, SK-Hep-1, SH-SY5Y, SK-MEL-1, ZR-75-1, T-47D, Hep-G2, Hs578T, Caco-2, SK-BR-3, SK-OV-3, HCT116, MDA-MB-468, SW480, PC-3, HGC27, MDA-MB-231, HT29, Caki-1, TF-1, MCF-7, PANC1, HuH-6, A431, MGC80-3, SGC-7901, MDA-MB-453, 769-P, ACHN, OSRC-2, and SK-NEP-1) and human renal tubular epithelial (HK2) cells were purchased from the Kunming Institute of Zoology, Chinese Academy of Sciences (Kunming, Yunnan, China). The cells were cultured in DMEM high-glucose medium, DMEM/F12, or RPMI 1640 medium (HyClone, Novato, CA, United States) supplemented with 10% fetal bovine serum (FBS) (BI, Kibbutz Beit-Haemek, Israel) and 1% penicillin–streptomycin (Solarbio, Beijing, China) and cultured in a cell incubator at 37°C and with 5% CO₂.

Cells from 30 cancer cell lines were separately seeded in 96-well plates at a density of 1 × 10⁴ cells/well. Once the cells had adhered, they were treated with 10 μM MIC-1 for 48 h, with untreated cells serving as the control. After 48 h, 20 μl of 50 mg/ml MTT reagent was added to each well. After incubating for 4 h in the dark, the supernatant was removed, 200 μl of DMSO was added to each well, followed by shaking for 10 min. The absorbance of each well was measured at 492 nm using a microplate reader (Molecular Devices Co., CA, United States).

Cells of six renal cancer cell lines (786-O, 769-P, Caki-1, ACHN, OSRC-2, SK-NEP-1) were separately seeded in 96-well plates at a density of 1 × 10⁴ cells/well. Once they had adhered, the cells were treated with MIC-1 (0, 1, or 10 µM) for 48 h. Cell viability was then evaluated using the MTT assay.

The 786-O, 769-P, and HK2 cell lines (1 × 10⁴ cells/well) were separately seeded in 96-well plates for 24 h and treated with MIC-1 (0, 1, 2, 4, 8, or 16 µM) for 24 h or 48 h. Cell viability was then evaluated using the MTT assay.

Untransfected and PTP1B-overexpressing 786-O cells (1 × 10⁴ cells/well) were separately seeded in 96-well plates for 24 h and treated with MIC-1 (0, 2, or 4 μM) for 48 h. Cell viability was then evaluated using the MTT assay.

The cell lines 786-O and 769-P were separately seeded in 6-well plates at a density of 3 × 10³ cells/well. Once they had adhered, the cells were treated with MIC-1 (0, 1, 2, and 4 μM) for 24 h or 48 h. After 24 or 48 h, the supernatant was removed, 3 ml of normal medium was added to each well, and then the cells were incubated for 12 days. The medium was changed every 2 days. After 12 days, the supernatant was removed, after which the cells were washed twice with PBS, fixed in methanol for 10 min, stained with 0.1% crystal violet (1 ml) for 15–30 min, washed with distilled water, air-dried, and imaged. A total of 600 µl of 10% glacial acetic acid was added and mixed well. The absorbance was measured at 562 nm using a microplate reader.

Untransfected and PTP1B-overexpressing 786-O cells (3 × 10³ cells/well) were separately seeded in 6-well plates for 24 h and treated with MIC-1 (0, 2, or 4 μM) for 48 h. Colony numbers were then evaluated as described above.

The cell lines 786-O and 769-P were separately seeded in 6-well plates at a density of 1 × 10⁵ cells/well. Once the cells had adhered, a sterile micropipette tip was used to make a scratch in the cell monolayer. After removing the supernatant, the cells were treated with different concentrations of MIC-1 (0, 1, or 2 μM), and imaged at 0 and 24 h under an inverted microscope to observe and record wound healing. The cell migration rate was then measured and analyzed using ImageJ software and GraphPad Prism 5.0 software (San Diego, CA, United States). The same procedure was followed to compare differences in cell migration between untransfected and PTP1B-overexpressing 786-O cells.

Untransfected and PTP1B-overexpressing 786-O cells (1 × 10⁵ cells/well) were seeded separately in 6-well plates for 24 h and treated with MIC-1 (0, 1, or 2 μM) for 24 h. The cell migration rate was then evaluated as described above.

A serum-free 786-O and 769-P cell suspension (5 × 10⁵ cells/ml; 100 µl) containing different concentrations of MIC-1 (0, 1, or 2 µM) was separately added to the upper chamber of a Transwell insert. Then, 600 µl of medium containing 20% FBS was added to the lower chamber of the Transwell. After 24 h of incubation, the supernatant was discarded, the cells were washed twice with precooled PBS, fixed in methanol for 20 min, and finally stained with 0.1% crystal violet for 30 min. The unmigrated cells in the upper layer of the chamber were gently cleared with a cotton swab. After washing three times with precooled PBS, the migrated cells were photographed under an inverted microscope in five randomly selected fields. The number of cells in each image was counted using ImageJ software. The data are presented as percentage of cell migration relative to control.

Matrigel and serum-free culture medium were mixed in a 1:6 ratio. Then, 100 µl of the mixture was added to the upper chamber of a Transwell insert and incubated for 24 h. After washing the Matrigel with serum-free medium, 100 µl of a serum-free 786-O and 769-P cell suspension (5 × 10⁵ cells/ml) containing different concentrations of MIC-1 (0, 1, or 2 µM) was separately added to the upper chamber of the transwell insert, while 600 µl of medium containing 20% FBS was added to the lower chamber. After 24 h of incubation, the supernatant was removed, the cells were washed twice with precooled PBS, fixed in methanol for 20 min, and stained with 0.1% crystal violet for 30 min. Unmigrated cells in the upper layer of the chamber were gently washed off with a cotton swab, while the invading cells were photographed under an inverted microscope in five randomly selected fields. The number of cells in each image was counted using ImageJ software. The data are presented as percentage of cell migration relative to control.

The cell lines 786-O and 769-P in the logarithmic growth phase were seeded in a 6-well plate at 2 × 10⁵ cells/well. Once they had adhered, the cells were treated with MIC-1 (0, 1, 2, or 4 μM) for 48 h. The cells were then collected, washed three times with precooled PBS, centrifuged, and the supernatant removed. The cells were resuspended in 100 µl of binding buffer containing Annexin V-FITC and propidium iodide (PI), mixed gently, and incubated for 15 min at room temperature in the dark. The cell signals were detected using a flow cytometer (BD, FACSCalibur, CA, United States), and the apoptosis rate was assessed by FlowJo software (Treestar, United States).

To determine the cell cycle distribution, cells were seeded and treated as above. After 48 h, the cells were collected and fixed in 70% ethanol at 4°C overnight, centrifuged, the supernatant discarded, the cells were washed again with precooled PBS, and then resuspended in 400 µl of PI staining solution containing RNase. After incubating at 37°C for 30 min in the dark, the cells were analyzed using a flow cytometer, and the cell cycle distribution was analyzed using FlowJo software (Treestar).

The cell lines 786-O, 769-P, HCT-116, Hep-G2, and A431 (1 × 10⁶) in the logarithmic growth phase were seeded in a 5-cm dish. Once they had adhered, the cells were treated with MIC-1 (0, 2, or 4 μM) for 48 h. Total cellular protein was then extracted with RIPA lysis buffer (Solarbio) and quantified by the BCA method. Equal amounts of protein were separated by 8% or 10% SDS-PAGE, transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, United States), blocked with 5% bovine serum albumin (BSA) (Solarbio) for 30 min, and then incubated overnight at 4°C with primary antibodies targeting Bax, Bcl-2, CDK2, K-Ras, and cyclinD1 (Cat. #sc-7480, sc-7382, sc-6248, sc-30, and sc-8396, respectively; 1:1,000 dilution; Santa Cruz Biotechnology, CA, United States), PTP1B and p-Src (Tyr529) (Cat. #ab244207, 32,078; 1:2000 dilution; Abcam, Cambridge, MA, United States), Src, p-Src (Tyr416), cyclinA2, Raf, p-Raf, ERK1/2, p-ERK1/2, and β-actin (Cat. #2109, 59,548, 4,656, 9422S, 9,431, 4,695, 4,370, and 3,700, respectively; 1:2000 dilution; Cell Signaling Technology, Danvers, MA, United States). The following day, the membranes were incubated for 1 h with horseradish peroxidase (HRP)-conjugated secondary IgG antibody (1:10,000, R&D Systems, United States). Signals were detected using the ECL Western blotting detection system (FluorChem E Protein Simple).

Untransfected and PTP1B-overexpressing 786-O cells were processed as above to detect protein expression levels.

The tumor tissue protein was extracted and the concentration was determined. Protein expression levels were determined as described above.

The cell lines 786-O and 769-P in the logarithmic growth phase were seeded in a 6-well plate at 2 × 10⁵ cells/well. Once they had adhered, the cells were treated with MIC-1 (0, 1, or 2 μM) for 24 h. The amount of matrix metalloproteinase (MMP)-2 and MMP-9 released from cell culture supernatants were measured by an ELISA kit (Mei Mian Biotechnology Co., Ltd., Jiangsu China) according to the kit instructions.

The SDF file of the MIC-1 molecular structure was downloaded from the PubChem database, and uploaded to the PharmMapper server (http://59.78.96.61/pharmmapper). The search was started by selecting a library of 2,241 human disease-related target proteins as the screening object, selecting the allowable conformational change parameters, and setting the maximum number of conformations for each molecule to 100. Target proteins were ranked according to the comprehensive match value Z′-score.

The in vitro inhibitory activity of MIC-1 against PTP1B was determined using a PTP1B Inhibitor Screening Assay Kit (ab139465, Abcam). In brief, a phosphate standard curve was generated by adding different volumes of PTP1B assay buffer (100, 97.5, 95, 90, 80, and 70 μl) and different volumes of 100 μM phosphate standard solutions (0, 2.5, 5, 10, 20, and 30 μl) to each well. Next, 10 μl of different concentrations of MIC-1 (10, 20, 40, 80, 160, or 320 μM) or the positive control suramin (1, 5, 10, 20, 40, or 80 μM) were mixed with 35 μl of PTP1B assay buffer and 5 μl of PTP1B diluent, giving a total reaction volume of 50 μl. A “time zero” (10 μl of dimethyl sulfoxide [DMSO], 35 μl of PTP1B assay buffer, and 5 μl of PTP1B diluent) and a “control well” (10 μl of DMSO and 40 μl of PTP1B assay buffer) were also prepared. Next, 50 μl of PTP1B substrate was added to each well and incubated at 37°C for 30 min. Then, 25 μl of red assay reagent was added to stop the reaction, and, after 20 min, absorbance was measured at 620 nm using a microplate reader. The optical density (OD) value was converted to nmol PO₄ ²⁻ using a standard curve. Activity was calculated as the percent (%) of Control: % Activity = [Test sample (nmol PO₄ ²⁻)—“time zero” (nmol PO₄ ²⁻)] [Control (nmol PO₄ ²⁻)—“time zero” (nmol PO₄ ²⁻)] × 100.

Aliquots (10 μl) of different concentrations of MIC-1 (10, 20, 40, 80, 160, or 320 μM) were mixed with 50 μl of TC-PTP (40 μg/ml), and the mixture was incubated at room temperature for 10 min. Next, 50 μl of 1 mM para-nitrophenyl phosphate (pNPP) was added to the mixture, followed by incubation at 37°C for 30 min. Finally, 50 μl of NaOH (3 M) was added to stop the reaction. The absorbance was measured at 405 nm using a microplate reader.

The binding affinity of MIC-1 and PTP1B (Sino Biological, Cat. 10,304-H07E) was determined with a Biacore S200 system (GE Healthcare, Uppsala, Sweden). PTP1B was immobilized on the Fc2 channel of the CM5 chip using an amine-coupling kit (GE Healthcare) following the manufacturer’s protocol. PTP1B was diluted to 100 μg/ml with a 10 mM sodium acetate solution (pH 4.0). Kinetics and affinity were determined at a flow rate of 30 μl/min using PBS-P buffer. MIC-1 association and dissociation rates were measured for 90 s each. The KD values were calculated using the kinetics and affinity analysis option of the Biacore S200 Evaluation Software Version 1.1 (GE Healthcare).

The molecular docking analysis was performed using the CDOCKER program of Discovery Studio (DS) 4.5 software (Accelrys Software Inc., San Diego, CA, United States). The X-ray crystal structure of PTP1B (PDB ID: 1NWE) was retrieved from the Protein Data Bank (PDB). The macromolecular structure of the PTP1B protein was selected as the receptor, and the protein molecules were optimized and prepared by adding hydrogen atoms, dewatering molecules, and applying the CHARMm force field. The binding site of the protein molecule was determined by the binding site of the known original ligand and the target, and the binding pocket radius range was automatically generated. The structure of MIC-1 was converted into a three-dimensional (3D) molecular formula, and the “Prepare Ligand” module was used to add hydrogen atoms and perform energy optimization operations. The parameters were set to default values. After molecular docking, the 10 best small molecule–protein binding conformations were obtained. The binding free energy and binding mode simulations (two-dimensional [2D] and 3D graphs) of the ligand-receptor interaction were also generated after the CDOCKER process.

The PTP1B overexpression cell line was generated using the piggyBac (PB) transposon system. The cell line 786-O (4 × 10⁵ cells/well) was seeded into 6-well plates and the PTP1B cDNA vector was transfected into the cells using Lipofectamine 3,000 Reagent (Thermo Fisher Scientific, Waltham, MA, United States). After selection with 1 μg/ml puromycin, clones expressing PTP1B were established.

Total RNA was extracted from PTP1B-overexpressing cells using Trizol reagent. The PCR reaction was carried out in a programmable thermal cycler. The PTP1B primer pair used was PTP1B-F (5′-GGC​CAT​TTA​CCA​GTT​GAC​CA-3′) and PTP1B-R (5′-ATG​ACG​ACA​CCC​CTG​CTT​TT-3′), which was expected to generate a 185-bp product. The GAPDH housekeeping gene primer pair GAPDH-F (5′-GCT​CTC​TGC​TCC​TCC​TGT​TC-3′) and GAPDH-R (5′-TTC​CCG​TTC​TCA​GCC​TTG​AC-3′) was expected to generate a 273-bp product.

Twenty-four 6-week-old male BALB/C nude mice were purchased from Changzhou Cavans Laboratory Animal Co., Ltd. After 1 week of adaptive feeding, 786-O cells (5 × 10⁶) were subcutaneously injected into each mouse. Once the tumor volume had reached 100 mm³, the mice were randomly divided into four groups, including a control group, a sunitinib (25 mg/kg) group, a MIC-1 low-dose (25 mg/kg) group, and a MIC-1 high-dose (50 mg/kg) group. Gavage was carried out once every 2 days, and the tumor volume, body weight, food intake, and water intake were recorded. After 3 weeks, the mice were euthanized with CO₂, and the tumor tissues were excised and stored at −80°C until use. The animal experiment protocols were reviewed and approved by the Institutional Animal Ethical Committee of Yunnan Agricultural University (IACUC-20190824-12).

Data presented as bar graphs are the means ± standard error of the mean (SEM) of at least three independent experiments. Statistical significance was evaluated by Student’s t-tests, one-way ANOVA, or two-way ANOVA using GraphPad Prism 5.0. p < 0.05, p < 0.01, or p < 0.001 was considered statistically significant.

To determine the anticancer activity of MIC-1, cells from 30 cancer cell lines were treated with MIC-1 (0 or 10 μM) for 48 h. The results showed that MIC-1 significantly inhibited the growth of 26 of the 30 cancer cell lines (LoVo, PC-12, A375, 786-O, SK-Hep-1, SH-SY5Y, SK-MEL-1, Hep-G2, Hs578T, Caco-2, SK-BR-3, SK-OV-3, HCT116, MDA-MB-468, SW480, MDA-MB-453, ZR-75-1, T-47D, HGC27, MGC80-3, MCF-7, PANC1, MDA-MB-231, HT29, PC-3, and Caki-1), but had no effect on the growth of the other four cell lines (TF-1, HuH-6, A431, and SGC-7901). MIC-1 exerted the strongest growth inhibitory effects on 786-O RCC cells, with the growth inhibitory rate reaching 88.69 ± 2.26% (Figure 1B).

To further investigate the effects of MIC-1 on the growth of different renal cancer cells, 786-O, 769-P, Caki-1, ACHN, OSRC-2, and SK-NEP-1 cells were treated with MIC-1 (0, 1, or 10 μM) for 48 h. Compared with the control, treatments with low doses of MIC-1 significantly inhibited the growth of 786-O cells (growth inhibition rate = 36.3 ± 9.3%) and SK-NEP-1 cells (growth inhibition rate = 8.52 ± 1.34%), but not the growth of the other four renal cancer cells. Meanwhile, high-dose MIC-1 treatment significantly inhibited the growth of ACHN, SK-NEP-1, 769-P, OSRC-2, and 786-O cells, with growth inhibition rates of 35.92 ± 3.79%, 51.01 ± 2.99%, 59.99 ± 7.38%, 69.46 ± 13.23%, and 85.57 ± 2.11%, respectively (Figure 1C). These results indicated that MIC-1 can inhibit the growth of RCC cells.

To investigate whether the growth inhibitory effects of MIC-1 on 786-O and 769-P RCC cells were dose- and time-dependent, we treated 786-O and 769-P cells with different concentrations of MIC-1 (0, 1, 2, 4, 8, or 16 μM) for 24 or 48 h. The results showed that MIC-1 treatment decreased the viability of 786-O and 769-P cells in a dose- and time-dependent manner (Figures 1D,E).

The colony formation assay results indicated that MIC-1 (0, 1, 2, and 4 μM) significantly decreased the colony-forming ability of 786-O and 769-P cells compared with that of the control group. At 24 h, the colony-formation rate of 786-O cells decreased from 100.00 ± 1.66% to 87.40 ± 1.71%, 27.50 ± 0.71%, and 22.95 ± 0.22%, respectively; At 48 h, the colony-formation rate of 786-O cells decreased from 100.00 ± 3.61% to 69.44 ± 2.80%, 26.58 ± 0.44%, and 18.34 ± 0.15%, respectively (Figure 1F). Meanwhile, at 24 h, the colony-formation rate of 769-P cells decreased from 100.00 ± 6.07% to 87.73 ± 8.58%, 57.19 ± 4.70%, and 18.99 ± 2.04%, respectively; At 48 h, the colony-formation rate of 769-P cells decreased from 100.00 ± 3.58% to 70.84 ± 2.42%, 51.13 ± 4.56%, and 11.94 ± 0.55%, respectively (Figure 1G).

To determine whether MIC-1 was toxic to normal renal cells, we evaluated the effects of MIC-1 on the growth of renal tubular epithelial cells (HK2 cells) using the MTT assay. Even when the dose was as high as 8 μM, MIC-1 did not negatively influence the viability of HK2 cells (Figure 1H). These results indicated that MIC-1 could significantly inhibit the growth of RCC cells, but was not toxic to normal renal cells.

As most patients with advanced RCC experience distant metastases (Yu et al., 2020), we assessed the effects of MIC-1 on the migration and invasion of 786-O and 769-P cells. The results showed that exposure to MIC-1 (0, 1, and 2 μM) significantly inhibited the migration of 786-O and 769-P cells. The cell migration rate of 786-O cells decreased from 100.00 ± 3.17% to 88.00 ± 4.63% and 47.35 ± 5.31% in the wound healing assay, and from 100.00 ± 6.44% to 86.47 ± 2.49% and 63.08 ± 3.67% in the Transwell assay (Figures 2A,C). The cell migration rate of 769-P cells decreased from 100.00 ± 1.85% to 72.49 ± 3.00% and 34.73 ± 7.05% in the wound healing assay, and from 100.00 ± 4.38% to 91.39 ± 6.44% and 78.15 ± 7.19% in the Transwell assay (Figures 2B,C). In addition, compared with the control cells, MIC-1 (0, 1 and 2 μM) significantly inhibited the invasive ability of 786-O and 769-P cells. The invasion rate of 786-O cells decreased from 100.00 ± 7.21% to 61.92 ± 6.82% and 48.95 ± 3.79% (Figure 2D); the invasion rate of 769-P cells was decreased from 100.00 ± 2.62% to 74.55 ± 2.18% and 55.31 ± 3.78% (Figure 2D).

Metastasis could be simplified into two phases: translocation to a distant tissue and colonization. The initial step requires the cell with the ability to degrade and move through extracellular matrix of the surrounding tissues. Among the MMPs family, MMP2 and MMP9 are involved in the breakdown of extracellular matrix in normal processes, so they are closely associated with the cancer cell migration. To further clarify the effects of MIC-1 on the migration of 786-O and 769-P cells, we detected the expression of MMP-2 and MMP-9 by ELISA. The results showed that the expression of MMP-2 and MMP-9 released from the cells treated with MIC-1 was markedly decreased compared to control cells (Figures 2E,F). These results indicated that MIC-1 inhibited the migratory and invasive of RCC cells by suppressing the expression of MMP-2 and MMP-9.

To determine whether the inhibitory effect of MIC-1 on RCC cell growth was associated with apoptosis and cell cycle arrest, we used flow cytometry to evaluate the apoptosis rate and cell cycle distribution of 786-O and 769-P cells following treatment with different doses (0, 2, or 4 µM) of MIC-1 for 48 h. The results showed that MIC-1 induced the apoptosis of 786-O and 769-P cells in a dose-dependent manner. Compared with the control group (18.24 ± 0.55%), the apoptotic rate of 786-O cells treated with MIC-1 increased to 31.85 ± 1.33% and 54.44 ± 0.83%, respectively (Figure 3A), while the apoptotic rate of 769-P cells treated with MIC-1 increased from 19.10 ± 0.20% to 24.95 ± 0.52% and 29.42 ± 0.86%, respectively (Figure 3B). In addition, MIC-1 significantly increased the protein expression of the proapoptotic proteins Bax, decreased the expression of the antiapoptotic protein Bcl-2, and increased the Bax/Bcl-2 ratio in 786-O cells and 769-P cells (Figure 3C).

Furthermore, after MIC-1 treatment, the percentage of 786-O cells in the G2 phase of the cell cycle increased from 5.42 ± 1.52% in the control cells to 12.54 ± 1.79% and 25.32 ± 0.45%, respectively, while the percentage of cells in the S-phase decreased from 21.55 ± 1.58% in the control cells to 15.46 ± 0.67% and 6.87 ± 0.84%, respectively (Figure 3D). The percentage of 769-P cells in the G1 phase of the cell cycle increased from 58.43 ± 1.81% in the control cells to 69.13 ± 0.33% and 65.20 ± 0.12%, respectively (Figure 3E). We further determined the expression levels of cell cycle-related proteins in 786-O cells using western blotting, and found that MIC-1 downregulated the expression level of CDK2 and cyclinA2. In addition, MIC-1 significantly downregulated the expression level of cyclinD1 in 769-P cells (Figure 3F). These results suggested that MIC-1 inhibits the proliferation of RCC cells by inducing apoptosis and cell cycle arrest.

Computer-aided drug screening is a fast and accurate method to predict drug-protein interactions. To investigate the molecular mechanism underlying the MIC-1-related inhibitory effects on the growth and migration of 786-O cells, we used the PharmMapper network database to conduct reverse virtual screening of MIC-1 target proteins. Among the 10 target proteins with the highest Z′-score filtered from the target library of 2,241 human disease-related proteins—platelet glycoprotein Ib alpha chain (GPIb-alpha), histo-blood group ABO system transferase, CD209 antigen, peptidyl-prolyl cis-trans isomerase (PPIase) FKBP1B, PTP1B, glucosamine-6-phosphate isomerase (GNPDA), phenylalanine-4-hydroxylase (PAH), proto-oncogene serine/threonine-protein kinase pim-1, beta-hexosaminidase subunit beta (HEXB), and sorbitol dehydrogenase (SDH) (Supplementary Table S1), PTP1B was found to be the most closely related to cancer occurrence and development (Bollu et al., 2017).

To investigate the effects of MIC-1 on the activity of PTP1B, an in vitro enzyme activity assay was carried out. The results showed that MIC-1 significantly inhibited the activity of PTP1B in a dose-dependent manner, with an IC₅₀ of 45.28 μM. The IC₅₀ of the positive control suramin was 17.59 μM (Figure 4A). T-cell protein tyrosine phosphatase (TC-PTP) shared the greatest sequence similarity with PTP1B (74% identity) in the catalytic domain and had identical active sites. Therefore, to verify that MIC-1 exerted selective inhibitory effect on PTP1B, we tested the enzyme activity of TC-PTP under MIC-1 treatment. We found that MIC-1 exerted only weak inhibitory activity against TC-PTP (IC₅₀ = 256.66 μM) (Figure 4B).

Next, we performed a binding assay to determine whether MIC-1 could directly bind to PTP1B. The results showed that PTP1B directly interacted with MIC-1 and that MIC-1 had a relatively high affinity for PTP1B, with an equilibrium dissociation constant (KD) of 2.0 × 10⁻⁵, an association rate constant (k _on) (1/ms) of 3.73×10², and a dissociation rate constant (k _off) (1/s) of 7.44 × 10⁻³ (Figure 4C).

To predict the molecular binding mode of MIC-1 to the PTP1B protein, we used the molecular docking program CDDDock. Molecular docking analysis showed that MIC-1 and PTP1B have a high binding capacity, with a molecule-protein binding energy of −50.0513 kcal/mol. Analysis of the specific binding mode indicated that MIC-1 fit well into the hydrophobic pocket of the active site of the PTP1B protein (Figure 4D). In addition, clear molecular interaction patterns were found in the 2D simulation diagram of the interaction between MIC-1 and PTP1B (Figure 4E). The results showed that seven hydrogen bonds and seven hydrophobic interactions could be formed between MIC-1 and PTP1B. The sulfur atom in the MIC-1 isothiocyanate functional group formed a hydrogen bond with the Tyr20 and Arg254 residues of PTP1B, while the nitrogen atom in the isothiocyanate group also formed a C hydrogen bond with the Gly259 residue of PTP1B. The aromatic ring and glycosyl side chain of MIC-1 can formed three sets of pi hydrogen bonds with the Gln262 and Asp48 residues of PTP1B. The H of the MIC-1 glycosyl side chain formed a hydrogen bond with the Asp48 residue of PTP1B, and the methyl group of the glycosyl tail formed three sets of alkyl–alkyl interactions with Tyr46, Val49, and Ala217 of PTP1B. In addition, MIC-1 also formed Van der Waals interactions with the Arg24, Phe182, Ile261, and Ile2194 residues of PTP1B. These results indicated that MIC-1 could serve as a selective inhibitor of PTP1B.

Studies have shown that PTP1B promotes the proliferation of colon cancer, breast cancer, and lung cancer by phosphorylating and activating Src (Zhu et al., 2007; Arias-Romero et al., 2009; Liu et al., 2015), a tyrosine kinase containing an SH3 domain. Activated Src then further activates the downstream Ras/ERK signaling pathway, thereby promoting tumor cell growth and migration (Bollu et al., 2017). To further investigate the influence of MIC-1 on the PTP1B-dependent Src/Ras/Raf/ERK signaling pathway, we analyzed the expression of proteins involved in this pathway. Compared with untreated control cells, MIC-1 decreased the expression level of p-Src (Tyr416), K-Ras, p-Raf, and p-ERK1/2 (Figure 5A). These results indicated that MIC-1 blocked Src activation by inhibiting the activity of PTP1B, thereby suppressing the activation of the downstream Ras/Raf/ERK signaling pathway.

To further confirm that MIC-1 inhibits the proliferation and migration of 786-O cells by regulating the PTP1B-dependent Src/Ras/Raf/ERK signaling pathway, we established cell lines overexpressing PTP1B. We found that the mRNA and protein expression levels of PTP1B, cell proliferation rate, cell migration rate, p-Src (Tyr416) and p-ERK1/2 expression levels in 786-O cells overexpressing PTP1B were significantly higher than that in normal 786-O cells. In addition, the expression level of p-Src (Tyr529) in 786-O cells overexpressing PTP1B was lower than that in normal 786-O cells (Figures 5B–F). However, compared with untreated 786-O cells overexpressing PTP1B, MIC-1 significantly decreased the cell proliferation rate, cell migration rate, the expression of p-Src (Tyr416) and p-ERK1/2, and increased the expression of p-Src (Tyr529) (Figures 5C–F). These results indicated that MIC-1 inhibited the growth and migration of RCC cells through the PTP1B/Src/Ras/Raf/ERK signaling pathway.

To further reveal how MIC-1 inhibited the growth of 786-O cells in vivo, we generated a 786-O cell xenograft tumor model in nude mice. We found that, compared with the control group, MIC-1 (25 and 50 mg/kg body weight [BW]) treatment markedly suppressed the growth of 786-O cell xenograft-derived tumors in a dose-dependent manner. In addition, the inhibitory effect of low-dose MIC-1 (25 mg/kg BW) on tumor growth was equivalent to that of the positive control drug sunitinib (25 mg/kg BW) (Figures 6A–C). However, there was no significant difference in BW, water intake, and food intake between the mice of the different groups (Figures 6D–F). Next, we measured the expression levels of apoptosis-related proteins in tumor tissues. Compared with the control group, MIC-1 treatment increased the expression of Bax, decreased Bcl-2 expression, and significantly increased the Bax/Bcl-2 ratio (p < 0.05) (Figure 6G). These results indicated that MIC-1 could inhibit the growth of 786-O cells in vivo.

To investigate the regulatory effects of MIC-1 on PTP1B-related signaling pathways in non-renal cancer cells, the cancer cell lines (HCT- 116, Hep-G2, and A431) were treated with MIC-1 (0 or 4 μM) for 48 h, and then the expression of Src/Ras/Raf/ERK signaling pathway-related proteins were detected. The results showed that, compared with untreated control cells, MIC-1 exerted no effects on the expression of p-Src (Tyr416), K-Ras, and p-ERK1/2 in HCT-116 cells, Hep-G2 cells, or A431 cells (Figures 7A–D). These results indicated that MIC-1 had no effect on the PTP1B-dependent Src/Ras/Raf/ERK signaling pathway in non-renal cancer cells.