AT-I and AG490 were purchased from Selleck (Houston, TX, USA). Stock solutions of AT-I (100 mM) and AG490 (10 mM) were dissolved in dimethyl sulfoxide (DMSO). Antibodies against HK2, PKM2, PFK, JAK2, phospho-JAK2, STAT3, and phospho-STAT3 were purchased from Cell Signaling Technology (Beverly, MA, USA). Antibodies against caspase-3, PARP, cleaved caspase-3, cleaved PARP, Bcl-2, Bax, and β-actin were purchased from Abcam (Cambridge, UK).

The human CRC cell lines HCT116 and SW480 were purchased from the Shanghai Institute for Biological Sciences (Shanghai, China). Human normal colon mucosal epithelial cell line NCM460 was obtained from INCELL (San Antonio, TX, USA). These cells were cultured in Mycos’ 5A (HCT116), L15 (SW480) and DMEM (NCM460) media (Invitrogen, Carlsbad, CA, USA), respectively, supplemented with 10% FBS (Invitrogen, Carlsbad, CA, USA), 100 U/mL penicillin, and 100 μg/mL streptomycin in a humidified atmosphere of 95% air and 5% CO₂ at 37°C.

Cell viability was evaluated using a Cell Counting Kit-8 (CCK-8) assay kit (Dojindo Laboratories Tokyo, Japan). Briefly, the cells were seeded into 96-well tissue culture plates at a density of 1 × 10⁴ cells/well and treated with the indicated concentrations of AT-I. After 24 or 48 h of incubation, 10 μL of CCK-8 solution was added and then the cells were incubated for a further 1 h. The absorbance of cells at 450 nm was measured using a microplate reader (Bio-Rad Laboratories, Richmond, CA, USA).

CRC cells were seeded in 96-well plates (5000 cells per well). After incubation for 24 h, the cells were pretreated with DMSO or AT-I for 24 h. Thereafter, 50 μM EdU (Ribobio, Guangzhou, China) was added to each well and following incubation at 37°C for 2 h, the cells were fixed with 4% formaldehyde for 30 min and then incubated with 2 mg/mL glycine for 5 min. After five washes with PBS, the cells were incubated with 100 μL of a 1× Apollo reaction cocktail for 30 min, following which, 1× Hoechst 33342 (5 μg/mL) was used to stain the nuclei for 30 min at room temperature. Cells were visualized in three fields of view/well under a fluorescence microscope (Carl Zeiss). The EdU incorporation rate was expressed as the ratio of EdU-positive cells to total Hoechst 33342-positive cells.

CRC cells were treated with AT-I at the indicated concentrations and then transferred to a six-well cell culture plate at a density of 5 × 10³ cells/well. After further incubation for 14 days, the cells were fixed with 4% paraformaldehyde for 15 min and then stained with 0.1% (wt/vol) crystal violet. After incubation for 10 min, the cells were washed with PBS and then photographed using a digital camera. Colonies containing more than 50 cells were counted.

Cells were stained with DAPI (Sigma-Aldrich, St Louis, MO, USA) to evaluate nuclear changes associated with apoptosis. Briefly, the CRC cells were seeded at a density of 1 × 10⁵ cells/well in six-well tissue culture plates, and then treated with the indicated concentrations of AT-I for 24 h. Thereafter, the cells were washed with PBS, followed by fixation with 4% paraformaldehyde for 30 min. After washing with PBS, the cells were stained with DAPI for 5 min. Samples were then mounted on a glass slide and the nuclear morphologies were observed using a BX51 fluorescence microscope (Olympus, Tokyo, Japan).

Apoptosis-mediated cell death was determined using an FITC Annexin V Apoptosis Detection Kit (BD Biosciences, San Jose, CA, USA) following the manufacturer’s guidelines. Briefly, CRC cells (1 × 10⁵ cells per well) were seeded in six-well tissue culture plates. After exposure to AT-I at the indicated concentrations for 24 h, both the attached and floating cells were collected. The cells were washed with PBS, then resuspended in 100 μL of binding buffer and stained with Annexin V-FITC and PI at room temperature for 15 min in the dark. Subsequently, 400 μL of binding buffer was added to each sample and the samples were examined using a FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Analysis of the data was performed using FlowJo software.

CRC cells (1 × 10⁵ cells/well) were seeded into six-well plates and treated with the indicated concentrations of AT-I for 24 h. The supernatant was collected and centrifuged to remove the cell debris. The glucose and lactate levels in the supernatant were measured using a glucose assay kit and a lactate assay colorimetric kit (Abcam, Cambridge, UK), respectively, according to the manufacturer’s instructions. Glucose consumption and lactate production were calculated based on a standard curve and normalized by cell numbers.

The human STAT3 expression plasmid pCMV3-STAT3-FLAG and empty vector pCMV3-C-FLAG were purchased from Sino Biological Inc. (Beijing, China). The CRC cells were transfected with the STAT3 overexpression vector or empty vector using Lipofectamine 2000 (Invitrogen, Grand Island, NY, USA) following the manufacturer’s instructions. After a 48 h transfection, the cells were treated with AT-I for 24 h and then harvested for analysis.

Total cell lysates were obtained by lysing with RIPA buffer (Beyotime, Nanjing, China) containing protease inhibitors. The protein concentration was determined using the Bradford assay (Bio-Rad Laboratory, Inc., Hercules, CA, USA). Proteins from each sample were separated by electrophoresis and then transferred to a PVDF membrane. Following blocking for 1 h at room temperature, the membrane was incubated with a 1:1000 dilution of primary antibodies overnight at 4°C. The following day, the membrane was washed and incubated with HRP-conjugated secondary antibodies (1:5000; Abcam, Cambridge, UK). Protein bands were developed using ECL Prime Western Blotting Detection reagent (GE Healthcare, Buckinghamshire, UK), and images were obtained using an ECL chemiluminescence instrument (Tanon Science and Technology Co., Ltd., Shanghai, China). Quantification of protein band density was performed using Image J software. Signals were densitometrically quantified and normalized to β-actin expression.

Autodock vina 1.1.2 was utilized for molecular docking to further study the interaction mode between AT-I and JAK2. The crystal structure of the JAK2 (PDB ID: 4BBE) was obtained from RCSB Protein Data Bank. The 3D structure of AT-I was constructed by ChemBio3D Ultra 14.0 software. The docking study was performed using the AutoDock Tools to generate the docking input files. Prior to docking, non-polar hydrogen atoms were merged and rotatable bonds were defined for the ligand. The search grid of the JAK2 site was identified as center_x: 3.581, center_y: -11.802, and center_z: -1.188 with dimensions size_x: 15, size_y: 15, and size_z: 15. AutoDock Tools and PyMol were used to visually analyze the docking results.

The complex of JAK2 and AT-I constructed by the docking was used as the initial structure for the MD simulations. Amber18 package was used for MD. JAK2 and AT-I were edited with the ff14SB force field and the GAFF force field, respectively. Each simulation system was immersed in a truncated octahedral box of TIP3P explicit water, and the distance between the proteins and margin was extended 10 Å. To relax the structure, the system was energy-minimized during 5,000 minimization steps. After that, the system was gradually heated up to 300 K over 50 ps. This procedure was followed by 50 ps of NPT simulation at 300 K and 1 atm pressure to equilibrate the system. After equilibration, the systems were used to run 50ns long MD simulations. A time step of 2fs was used and coordinates of the system were saved every 20 ps. The resulting trajectory data were viewed and analyzed using the VMD software. Binding free energy was calculated by Molecular mechanics/generalized Born surface area (MM/GBSA).

The animal experiments were approved by the Animal Ethics Committee of China Medical University (No. CMU2019321). For xenograft studies, male BALB/c nude mice (six weeks old and 18-20 g each) were injected subcutaneously with HCT116 (5 × 10⁶ cells/100 μl) into the right flank. When the diameter of the tumor reached to approximately 5 mm, mice were randomly divided into control and treated group (6 mice each group). Control group of mice was intraperitoneally injection with vehicle (0.9% sodium chloride plus 0.1% DMSO) and treated group was administrated with AT-I (50 mg/kg body weight) once daily for 3 weeks. The body weight and tumor volume were measured every 3 days after treatment begin. Tumor volumes were calculated using the formula: tumor volume (mm³) = (long diameter × short diameter ²)/2. At the end of the study, tumors were removed and then weighed.

For histological examination, dissected tumor tissues were fixed with 4% paraformaldehyde, embedded in paraffin and sectioned at 5 μm thicknesses. Levels of apoptosis in tumor tissue were determined using TdT-mediated dUTP-biotin nick end-labeling (TUNEL) assay with the in-situ Apoptosis Detection Kit (KeyGEN BioTECH, Nanjing, China) according to the manufacturer’s instructions. For immunohistochemical analysis of p-JAK2, p-STAT3 and HK2, the deparaffinized and hydrated sections were incubated with 0.3% H₂O₂ to block endogenous peroxidase activity and performed with sodium citrate buffer in a microwave for 20 min for antigen retrieval. Then the sections were incubated with the indicated antibodies overnight at 4°C followed by incubation with HRP-conjugated secondary antibody. 3,3′-Diaminobenzidine and hematoxylin were used to detect the immunocomplexes and nuclear counterstaining respectively. Finally, the tissue sections were captured with an Olympus digital camera attached to a light microscope.

All results were obtained from experiments that were repeated at least three times. For comparison of the two groups, statistical significance was analyzed with Student’s t-test. Statistical significance for multiple groups was performed using one-way ANOVA with Tukey’s post hoc test. Values were expressed as the mean ± standard error (SD). Statistical significance was set at P < 0.05.

We initially investigated the potential anti-cancer properties of AT-I in CRC cell lines. We cultured two human CRC cell lines, HCT116 and SW480, in a medium containing 0-200 μM of AT-I for 24 and 48 h, and then evaluated cell viability using a CCK-8 assay. The results revealed that AT-I inhibited CRC cell viability in dose- and time-dependent manners, with half maximal inhibitory concentrations (IC50) of 126.8 μM and 97.19 μM for a 24 h treatment, and 98.49 μM and 70.44 μM for a 48 h treatment, respectively ( Figure 1B ). In addition, we also tested the cytotoxic effect of AT-I on normal cells NCM460. As shown in Figure 1B , AT-I exhibited as less cytotoxic against NCM460, with the IC50 values of more than 200 μM after 24 and 48 h treatment. To determine whether the decrease in viable cell density was due to an anti-proliferative effect of AT-I, we further measured the effect of AT-I on DNA replication using an EdU incorporation assay. The results showed that the numbers of EdU-positive cells were significantly reduced in both the SW480 and HCT116 lines after AT-I treatment for 24 h ( Figure 1C ). In addition, the anti-proliferative activity of AT-I was further evaluated using a colony formation assay. As shown in Figure 1D , there was a significant reduction in colony number after treatment with AT-I. Collectively, these findings indicate that AT-I inhibits CRC cell proliferation but has low cytotoxicity towards normal cells.

To determine whether AT-I-mediated inhibition of CRC cell proliferation is associated with apoptosis, we used DAPI staining to examine nuclear condensation (a characteristic of apoptosis). The results revealed prominent chromatin condensation and nuclear fragmentation in the AT-I-treated CRC cells ( Figure 2A ). To confirm the induction of apoptosis in response to AT-I treatment, we also assessed apoptosis using flow cytometric analysis after double labeling with Annexin V and PI. The results showed that AT-I treatment for 24 h increased the percentage of apoptotic CRC cells in a dose-dependent manner ( Figure 2B ). Given that activation of caspase-3 and cleavage of PARP are well-known molecular markers of apoptosis (Elmore, 2007), we next examined the levels of caspase-3 and PARP in CRC cells treated using AT-I. As shown in Figures 2C, D , AT-I triggered caspase-3 and PARP-1 cleavage as seen by the reduced expression of procaspase-3 and full length PARP and increased expression of cleavage of caspase-3 and PARP in a dose-dependent manner. An imbalance between the pro-apoptotic protein Bax and anti-apoptotic protein Bcl-2 would result in caspase-3 activation and subsequent apoptosis (Plati et al., 2011), and we found that AT-I dose-dependently downregulated Bcl-2 and upregulated Bax expression levels compared with the control group ( Figures 2C, D ). Collectively, these findings indicated that AT-I could efficiently induce apoptosis in CRC cells.

Given that cancer cell proliferation and apoptosis are closely linked with glycolysis (Matsuura et al., 2016), we examined whether AT-I can be used to induce a metabolic shift in CRC cells. We found that after AT-I treatment, there was a significant dose-dependent decrease in glucose consumption by the CRC cells ( Figure 3A ). In addition, there was a significant decrease in lactate secretion in the AT-I-treated CRC cells ( Figure 3B ). Therefore, we examined the expression of three key rate-limiting glycolytic enzymes (HK2, PFK, and PKM2) in CRC cells treated with AT-I. Western blot analysis revealed that HK2 expression decreased in response to the AT-I treatment, whereas the expression of PFK and PKM2 remained unaltered ( Figures 3C, D ). Collectively, these findings indicate that AT-I may suppress glycolysis in the CRC cells by reducing HK2 expression.

Recent studies have shown that STAT3 is a transcription factor that regulates the transcription of HK2 (Jiang et al., 2012). Thus, we hypothesized that AT-I may regulate STAT3 activation in CRC cells. Phosphorylation of the Tyr705 residue of STAT3 appears to be an essential prerequisite for converting STAT3 to an active form (Levy and Lee, 2002). We found that AT-I suppresses the phosphorylation of STAT3 at Tyr705 in a dose-dependent manner without altering total STAT3 protein levels ( Figures 4A, B ). Considering the significant role of JAK2 in phosphorylation of the Tyr705 residue in STAT3 (Rawlings et al., 2004), we evaluated the activation of JAK2 in response to AT-I in CRC cells. As shown in Figures 4C, D , AT-I inhibited JAK2 phosphorylation in a dose-dependent manner. We therefore speculated as to whether JAK2 acts as a target of AT-I. To verify this hypothesis, we performed a computer docking analysis. The results showed that AT-I docks onto the binding site of human JAK2 ( Figure 4E ). AT-I was located at the hydrophobic site, surrounded by residues Leu-855, Phe-860, Val-863, Ala-880, Met-929, Leu-932, and Leu-983, forming strong hydrophobic binding. Importantly, we observed one key hydrogen bond interaction between AT-I and Leu-932 (bond length = 2.4 Å), which was the main interaction between AT-I and JAK2 ( Figure 4F ). All these interactions facilitated the anchoring of AT-I to the binding site of JAK2. Then we used MD simulations to evaluate the dynamic stability of the complexes of AT-I with JAK2. As shown in Figure 4G , the RMSD of docked AT-I-JAK2 complex and JAK2 apo all remain stable after a 25ns MD run. By contrast, the complex RMSD was more stable and lower than that of JAK2 apo. The total binding free energy between AT-I and JAK2 was -27.7kcal/mol, indicating a well binding between them. Collectively, these results indicated that AT-I undergoes a strong interaction with JAK2 and this binding interaction may increase the structural stability of JAK2. To confirm the role of JAK2 in AT-I-induced STAT3 inactivation, we pretreated CRC cells with AG490 (10 μM), a JAK2 inhibitor, followed by treatment with AT-I. We accordingly found that the AT-I-induced decrease in STAT3 phosphorylation was enhanced by pretreating the CRC cells with AG490 ( Figures 4H, I ). We therefore conclude that AT-I inhibits JAK2 activity and thereby functions as an inhibitor of the JAK2/STAT3 signaling pathway.

To further clarify the role of STAT3 in the regulatory effect of AT-I on glycolysis, we overexpressed STAT3 by transient transfection of a vector overexpressing STAT3 into CRC cells. After 24 h transfection, the expression of total STAT3 and STAT3 phosphorylation was increased remarkably in CRC cells ( Figure 5A ). Moreover, we found that when overexpression STAT3, AT-I-induced changes in the expression levels of HK2 were markedly reversed ( Figures 5B, C ). Further, we also observed that STAT3 overexpression rescued the deficient glucose consumption and lactate production in the AT-I-treated CRC cells ( Figures 5D, E ). These results indicated that AT-I-induced glycolysis suppression via decreased HK2 expression in CRC cells is partly mediated by inactivation of STAT3.

In addition to mediating glycolysis, STAT3 activation is also involved in the proliferation and anti-apoptosis of cancer cells [24]. We therefore wondered whether STAT3 is involved in AT-I-induced CRC cell apoptosis. After exogenously overexpressing STAT3 in CRC cells, the reduced cell viability caused by AT-I was partially reversed ( Figure 6A ). Moreover, the overexpression of STAT3 also reduced AT-I-induced apoptosis ( Figure 6B ). Consistent with these results, the effects of AT-I on apoptosis-related protein expression were significantly reversed after the overexpression of STAT3 in CRC cells ( Figures 6C, D ). These results indicated that STAT3 plays an important role in the anti-cancer effects mediated by AT-I in CRC cells.

In order to characterize the anti-cancer effects of AT-I in vivo, tumors of xenograft formed by the HCT116 cell were used. As shown in Figure 7A , during the three week treatment, AT-I significantly decreased tumor volume, when compared with the control group. Moreover, tumor weight of mice in AT-I treated group was markedly lower than that of the mice in the control group at the end of the treatment ( Figure 7B ). However, we observed that no significant difference in the body weight of mice between the control and AT-I-treated groups ( Figure 7C ), suggesting that AT-I has no obvious toxicity. Additionally, we detected apoptosis in xenografts mice using a TUNEL assay. We observed that AT-I increased the number of TUNEL-positive cells in the AT-I-treated group confirming that AT-I could induce apoptosis in vivo ( Figure 7D ). Consistent with the in vitro results, immunohistological analysis revealed that AT-I downregulated the expression of HK2, p-JAK2 and p-STAT3 in tumor tissuses of the xenograft mice ( Figure 7D ). Collectively, these in vivo data suggested that AT-I could induce CRC suppression through inhibiting JAK2/STAT3-dependent regulation of HK2 in tumors.