The Prolyl Isomerase Pin1 Is a Novel Target of 6,7,4′-Trihydroxyisoflavone for Suppressing Esophageal Cancer Growth

Esophageal cancer is the eighth most common cancer worldwide and 456,000 new cases or 3.2% of total cancer cases were estimated in 2012. This cancer is the sixth most common cause of cancer-related death with an estimated 400,000 deaths or 4.9% of total cancer-related deaths. Furthermore, the survival rate of esophageal cancer is very low with a mortality-to-incidence ratio of 0.88 (1). The incidence of esophageal cancer tends to be dependent on geographic location. For example, 16.7 new cases of esophageal cancer per 100,000 people were estimated in 2008 in China. In contrast, only 0.8 or 2.7 new cases per 100,000 people were projected in Greece or Canada, respectively (1). This tendency might be interpreted as a result of nutritional and lifestyle differences (2).

Epidemiologically, an inverse correlation has been reported between the intake of soy foods and risk of breast (3), prostate (4), stomach (5), and esophageal cancers (2). Isoflavones are major bioactive chemicals present in soy foods. Daidzein is a major isoflavone in soybeans, with amounts of approximately 100 to 300 mg per 100 g of soybeans (6). Even with the high amount of daidzein found in soybean, only 7% to 30% or 10% of daidzein was detected in urine or feces, respectively (7–9). This input/output error of daidzein reflects an efficient metabolism of daidzein. Indeed, daidzein can be readily converted to its metabolites, including equol, which is produced in the human intestine by many, but not all, individuals (9). Moreover, cytochrome P450 (CYP1A2), a major metabolic enzyme in liver, hydroxylates daidzein to produce 6,7,4′-trihydroxyisoflavone (6,7,4′-THIF) and 7,3′,4′-trihydroxyisoflavone (7,3′,4′-THIF; refs. 10–12). We and other research groups previously reported the biological activities of daidzein metabolites (13–16). In the case of equol, protective effects against cardiovascular disease and reduction of LDL cholesterol were reported (13). The anticancer activity of 6,7,4′-THIF was also observed in a colon cancer model (16). However, to the best of our knowledge, Pin1 has not been reported as a molecular target of 6,7,4′-THIF.

Pin1 is an evolutionarily conserved member of the peptidyl-prolyl isomerase (PPIase) family (17). The phosphor-Ser/Thr/Pro motif is a major motif in many cell types and is associated with growth and transformation (18, 19). This protein specifically isomerizes phosphor-Ser/Thr/Pro residues and regulates various signaling pathways (17, 19, 20). Pin1 comprises an N-terminal WW domain (amino acids 1–39) and a C-terminal PPIase domain (amino acids 45–163; ref. 17). Although the WW domain of Pin1 plays a role in binding to the phosphor-Ser/Thr/Pro motif, the main function of the PPIase domain is isomerization (20). Pin1 represses apoptosis by affecting apoptosis-related proteins, such as survivin (21), p63α (22), and death-associated protein Daxx (23). Cheng and colleagues reported increased tumor size that was induced by MIHA cells, an immortalized liver cell line overexpressing Pin1, and decreased tumor volume associated with Pin1-depleted PLC/PRF/5 human liver hepatoma cells (21). Pin1 was originally identified as a mitotic cell-cycle protein (17). Although Pin1 is known to regulate transcription, protein stability and localization of cyclin D1 during the G₁–S phase (24), G₂–M phase–related proteins, such as hBora, cdc25, and Wee1 were also reported to be downstream molecular targets of Pin1 (20). Pin1 plays an important role in cell cycle and aberrant cell-cycle changes have been observed in Pin1-defective cells (25–27).

In this study, we found that Pin1 is a direct target of 6,7,4′-THIF for the suppression of esophageal cancer cell growth. Upregulation of Pin1 levels has been observed clinically and is closely correlated with poor survival of esophageal cancer patients (28). Indeed, we confirmed an overexpression of the Pin1 protein in esophageal cancer cells compared with normal cells (Supplementary Fig. S1). Because of the inverse association between intake of soy isoflavones and the risk of esophageal cancer, this study could at least partially explain these clinical observations (2). Because the upregulated Pin1 activity in esophageal cancer is suppressed by 6,7,4′-THIF, the intake of soy isoflavones could have a positive effect on the prognosis of esophageal cancer.

6,7,4′-THIF was purchased from ChromaDex and DMEM was from Hyclone. FBS was obtained from Gemini Bio-Products. Penicillin/streptomycin was purchased from Gibco. The protein assay kit was from Bio-Rad Laboratories. The β-actin antibody was obtained from Sigma-Aldrich. Primary antibodies recognizing Pin1, cyclin D1, Bax, Bcl-xL, cleaved-PARP, CDK2, pRb, caspase-3, and cyclin E were purchased from Santa Cruz Biotechnology. Antibodies against phospho-c-Jun, Bcl-2, cyclin B1, and cyclophilin A were obtained from Cell Signaling Technology. The primary FKBP antibody was purchased from BD Biosciences and CNBr-Sepharose 4B beads were purchased from GE Healthcare.

KYSE 30, 450, and 510 esophageal cancer cells were cultured in RPMI1640 medium supplemented with 10% FBS (v/v). Neu/Pin1 and Pin1 murine embryonic fibroblasts (MEF) were cultured in DMEM supplemented with 10% (v/v) FBS. All the cell lines were cytogenetically tested and authenticated before freezing.

Sepharose 4B powder (0.3 g) was activated by suspension in 1 mmol/L HCl. A coupling solution (0.1 mol/L NaHCO₃, pH 8.3, and 0.5 mol/L NaCl) and 6,7,4′-THIF were then added and rotated overnight at 4°C. The mixture was washed with coupling buffer, and then transferred to 0.1 mol/L Tris-HCl buffer (pH 8.3). The excess of uncoupled 6,7,4′-THIF was removed with washing buffer (0.1 mol/L acetate pH 4.0 and 0.1 mol/L Tris-HCl pH 8.0) containing 0.5 mol/L NaCl.

The Pin1 protein (500 ng) was mixed with Sepharose 4B beads (as a negative control) or 6,7,4′-THIF–conjugated Sepharose 4B beads (100 μL) in reaction buffer [50 mmol/L Tris pH 7.5, 5 mmol/L EDTA, 150 mmol/L NaCl, 1 mmol/L dithiothreitol (DTT), 0.01% Nonidet P-40, 2 mg/mL BSA, 0.2 mmol/L phenylmethylsulfonyl fluoride (PMSF), and 1× protease inhibitor mixture]. After incubation, the mixture was washed five times with buffer (50 mmol/L Tris pH 7.5, 5 mmol/L EDTA, 150 mmol/L NaCl, 1 mmol/L DTT, 0.01% Nonidet P-40, and 0.02 mmol/L PMSF) and the proteins bound to the beads were detected by Western blotting.

For the ex vivo pull-down assay, Neu/Pin1 wild-type (WT) and Pin1 WT MEFs were disrupted with lysis buffer (50 mmol/L Tris pH 7.5, 5 mmol/L EDTA, 150 mmol/L NaCl, 1 mmol/L DTT, 0.5% Nonidet P-40, 0.02 mmol/L PMSF, and 1× protein inhibitor mixture). Then proteins (500 mg) were incubated with Sepharose 4B or 6.7.4′-THIF-Sepharose 4B beads.

Cell proliferation was analyzed using Cell Titer 96 Aqueous One Solution (Promega). Briefly, cells (3 × 10³) were seeded into each well of a 96-well plate. At 4 hours after seeding, cells were treated with 6,7,4′-THIF and incubated for the indicated time. For evaluating proliferation, 20 μL of MTS solution was added to each well and cells were incubated for 1 hour at 37°C in a 5% CO₂ incubator. The absorbance was measured at 492 nm.

Cells (8 × 10³) were suspended in BME supplemented with 10% FBS and 1% gentamycin and added to 0.3% agar with the indicated concentration of 6,7,4′-THIF in a top layer over a base layer of 0.6% agar with different doses of 6,7,4′-THIF. The cultures were incubated at 37°C in a 5% CO₂ incubator for 3 weeks, and then the colonies were counted under a microscope using the Image-Pro Plus Software (v.4) program (Media Cybernetics).

The cell cycle was measured in KYSE 450 and 510 esophageal cancer cells using a previous method (29) with slight modifications. The cells were exposed to various concentrations of 6,7,4′-THIF for 36 hours and then fixed with ethanol. The cells were stained with propidium iodide and the cell-cycle phase was determined by flow cytometry.

The effect of 6,7,4′-THIF on Pin1 PPIase activity was examined using a proline isomerase assay that is based on the specific cleavage of the phosphorylated peptide WFY(pS)PR-pNA (PEPTIDE 2.0) by trypsin as described previously (30). Briefly, the activity of Pin1 was evaluated by absorbance of p-nitroaniline at 390 nm measured every 20 seconds for a total of 100 seconds using a Beckman DU800 spectrophotometer and a temperature-controlled cuvette holder. Before measurements, the purified Pin1 protein was coincubated with 6,7,4′-THIF in sample buffer at 10°C for 10 minutes. Trypsin (working solution of 0.1 mg/mL from a stock solution of 50 mg/mL) in 1 mmol/L HCl and 2 mmol/L CaCl₂ was then added. The reaction was immediately started by addition of the peptide dissolved in 0.47 mol/L LiCl/TFE (i.e., trifluoroethanol, anhydrous) at a final concentration of 20 mg/mL. The total volume of the reaction mixture was 1 mL, including 35 mmol/L HEPES (pH 7.8), 44 nmol/L Pin1, 20 mmol/L peptide, 0.1 μg/mL trypsin, and various doses of 6,7,4′-THIF.

To examine PPIase activity in cells, Pin1 WT and Neu/Pin1 WT MEFs were exposed to 6,7,4′-THIF for 24 hours. The cells were disrupted with 35 mmol/L HEPES (pH 7.4; containing proteinase inhibitors) and equivalent concentrations of soluble proteins were used for experiments.

Cells were disrupted with lysis buffer (10 mmol/L Tris pH 7.5, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% Triton X-100, 1 mmol/L DTT, 0.1 mmol/L PMSF, 10% glycerol, and protease inhibitor cocktail tablet). The protein concentration was measured using the bicinchoninic acid assay (Pierce Biotechnology). The proteins were separated by SDS-PAGE. After separation, the proteins were transferred to Immobilon-P membranes (Millipore Corporation) and the membranes were blocked with 5% fat-free milk for 1 hour. The membranes were incubated with specific primary antibodies at 4°C overnight and proteins were detected by chemiluminescence after hybridization with a horseradish peroxidase (HRP)-conjugated secondary antibody.

To determine whether 6,7,4′-THIF could interact with Pin1, we performed an in silico docking assay using Schrödinger Suite 2015 (31). The Pin1 crystal structure (PDB ID: 3OOb; ref. 30) was downloaded from the Protein Data Bank (32) and this structure was prepared under the standard procedures of the Protein Preparation Wizard (Schrödinger Suite 2015). Hydrogen atoms were added consistent with a pH of 7 and all water molecules were removed. The 6,7,4′-THIF–binding site-based receptor grid of Pin1 was generated for docking.

The 6,7,4′-THIF compound was prepared for docking by default parameters using the LigPrep program. Then, the docking of 6,7,4′-THIF with Pin1 was accomplished with default parameters under the extra precision (XP) mode using the Glide program to obtain the best-docked representative structure.

Female athymic nude mice (6- to 7-week-old) were used to examine the effect of 6,7,4′-THIF on the growth of Neu/Pin1 WT or knockout (KO) MEFs. Mice were divided into six groups as follows: (i) a Neu/Pin1 WT–injected group treated with vehicle only (n = 12); (ii) a Neu/Pin1 WT–injected group treated with a low dose (5 mg/kg) of 6,7,4′-THIF (n = 12); (iii) a Neu/Pin1 WT–injected group treated with a high dose (25 mg/kg) of 6,7,4′-THIF (n = 12); (iv) a Neu/Pin1 KO–injected group treated with vehicle only (n = 12); (v) a Neu/Pin1 KO–injected group treated with a low dose (5 mg/kg) of 6,7,4′-THIF (n = 12); and (vi) the Neu/Pin1 KO–injected group treated with a high dose (25 mg/kg) of 6,7,4′-THIF (n = 12). Cells (1 × 10⁶/100 μL) suspended in serum-free DMEM were injected subcutaneously into a flank of each mouse. Vehicle (5% DMSO + 45 % PEG400 + 50% PBS) or 6,7,4′-THIF was administered intraperitoneally three times a week. Tumor volume was calculated from measurements of two diameters of the individual tumor base using the following formula: tumor volume (mm³) = (length × width × height × 0.52).

All data are presented as mean values ± SD of triplicate samples from at least three independent experiments. Differences between groups were assessed by ANOVA and the minimum level of significance was set at a P value less than 0.05.

To understand how 6,7,4-THIF interacts with Pin1, we docked the compound into the ATP-binding pocket of Pin1 using several protocols included in Schrödinger Suite 2015. On the basis of the final computational docking model result, we found that 6,7,4-THIF formed hydrogen bonds at both the WW and PPIase domains of Pin1 (Fig. 1A). This indicates that 6,7,4-THIF might be a potential inhibitor of Pin1. Images were generated with the UCSF Chimera software program (33).

We used pull-down assay to confirm the binding of 6,7,4′-THIF to the Pin1 protein. Results revealed a direct interaction of 6,7,4′-THIF with Pin1 in vitro (Fig. 1B) and in Pin1 WT MEFs (Fig. 1C). The human Pin1 protein comprises a WW domain (N-terminal: 1–39 aa), a linker region (40–44 aa), and a PPIase domain (45–168 aa). We examined the potential binding domains of Pin1 with 6,7,4′-THIF and found that 6,7,4′-THIF interacted with the WW domain as well as the PPIase domain (Fig. 1D). To investigate whether the binding of 6,7,4′-THIF with Pin1 was specific, we determined whether 6,7,4′-THIF could bind with other PPIase family proteins, including FKBP or cyclophilin A, in Neu/Pin1 and Pin1 MEFs using a pull-down assay. Results showed that 6,7,4′-THIF could not bind with FKBP or cyclophilin A, indicating that the binding of 6,7,4′-THIF with Pin1 is specific (Supplementary Fig. S1). Molecular docking data suggested that 6,7,4′-THIF directly interacts with Arg17, Ser18, and Lys97 in the WW domain and Asp112, Ser154, and Gln131 in the PPIase domain of Pin1. To verify the direct interaction of 6,7,4′-THIF with Pin1, we transfected an Arg17Ala (R17A) point mutant plasmid of Pin1 into Neu/Pin1 KO MEFs. Lysates from WT Pin1-transfected and R17A Pin1–transfected MEFs were incubated with 6,7,4′-THIF–conjugated Sepharose 4B beads for a pull-down assay. The results demonstrated that the band intensity was decreased in the R17A Pin1–transfected group (Fig. 1E). Overall, we confirmed that 6,7,4′-THIF directly binds the Pin1 protein on both the WW and PPIase domains.

Next, we evaluated the effect of 6,7,4′-THIF on Pin1 PPIase activity using an α-trypsin–coupled assay as described in Materials and Methods. Results indicated that 6,7,4′-THIF dose dependently inhibited Pin1 PPIase activity in vitro (Fig. 2A). Pin1 WT MEFs and Neu/Pin1 WT MEFs were also used to measure the activity of Pin1 ex vivo. MEFs were treated with various concentrations of 6,7,4′-THIF for 24 hours and 6,7,4-THIF also suppressed Pin1 PPIase activity in Pin1 WT MEFs (Fig. 2B) and Neu/Pin1 MEFs (Fig. 2C). The phosphorylation of c-Jun on Ser63/73-Pro motifs was reported as a target of Pin1 binding for activation of the cyclin D1 promoter (34). Therefore, we next determined whether 6,7,4′-THIF could affect c-Jun phosphorylation at Ser63 or cyclin D1 protein expression levels in esophageal cancer cells. c-Jun phosphorylation was decreased and also cyclin D1 protein expression level was also reduced by 6,7,4′-THIF (Fig. 2D). Overall, we confirmed that 6,7,4′-THIF could suppress Pin1 PPIase activity in vitro and in Neu/Pin1 WT and Pin1 WT MEFs. In addition, the phosphorylation of the c-Jun and cyclin D1 proteins was strongly decreased by 6,7,4′-THIF treatment.

Minoru and colleagues previously reported a positive correlation between Pin1 and cyclin D1 protein expression level and prognosis in esophageal cancer patients suggesting that Pin1 might be a suitable biomarker for esophageal cancer (35). Thus, we hypothesized that 6,7,4′-THIF might suppress tumorigenic activity in an esophageal cancer cell model because it could reduce Pin1 activity. First, we examined the expression of Pin1 in three esophageal cancer cell lines, including KYSE 30, 450, and 510 cells, and results indicated that Pin1 is highly expressed compared with normal HaCaT cells (Supplementary Fig. S2). The effect of 6,7,4′-THIF was measured in the same three esophageal cancer cell lines and results showed that 6,7,4′-THIF inhibited the proliferation of all three cell lines (Fig. 3A–C). Furthermore, 6,7,4′-THIF was much more potent at inhibiting growth of these 3 cell lines compared with another similar soy compound, 7,8,4′-THIF (Supplementary Fig. S3). In addition, we evaluated the effect of 6,7,4′-THIF on anchorage-independent growth of KYSE 450 and KYSE 510 esophageal cancer cells and results showed that colony formation was dose dependently decreased by 6,7,4′-THIF (Fig. 3D). These results further supported the idea that 6,7,4′-THIF exerts antitumorigenic activity in esophageal cancer cells by suppressing Pin1 activity.

An accumulation of data indicated that Pin1 increases cancer cell survival (21, 23, 36). For example, several pro- or antiapoptotic proteins are affected by Pin1 and include NF-κB (37), survivin (21), p63α (22), and Daxx (23). Moreover, knocking down Pin1 was reported to markedly increase apoptosis (36). Thus, we determined whether the inhibition of Pin1 activity by 6,7,4′-THIF could trigger apoptosis in esophageal cancer cells. Apoptosis of KYSE 30, 450, and 510 esophageal cancer cells was increased dose dependently (Fig. 4A) after treatment with 6,7,4′-THIF for 36 hours (Fig. 4A) and in a time-dependent manner after treatment with 20 μmol/L 6,7,4′-THIF (Fig. 4B). Next, we evaluated the effect of 6,7,4′-THIF on the level of several apoptosis-related proteins in esophageal cancer cells. The level of antiapoptotic proteins, Bcl-xL and Bcl-2, was diminished and the level of the proapoptotic Bax was increased by 6,7,4′-THIF treatment (Fig. 4C). Also, PARP cleavage was induced by 6,7,4′-THIF (Fig. 4C). Overall, our data demonstrate that 6,7,4′-THIF could induce apoptosis in esophageal cancer cells presumably by reducing Pin1 PPIase activity.

Pin1 was originally identified as a regulator of cell cycle and various cell-cycle–associated proteins are binding partners with the Pin1 protein (17). In particular, Pin1 transcriptionally modulates cyclin D1 through the c-Jun/AP-1 and β-catenin/TCF signaling pathways. Furthermore, the stability and subcellular localization of cyclin D1 are affected by Pin1 (24). Cyclin E and Myc are also regulated by Pin1-associated degradation in the G₁–S transition of the cell cycle. During the G₁–S transition phase, Pin1 can accelerate the FBXW7-mediated cyclin E and Myc degradation (17). We therefore investigated the effect of 6,7,4′-THIF on the cell cycle of esophageal cancer cells. Treatment with 6,7,4′-THIF significantly enhanced accumulation of KYSE 450 and 510 cells in S-phase in a dose-dependent manner (Fig. 5A). We next evaluated the effect of 6,7,4′-THIF on the expression of various cell-cycle proteins and found that the protein levels of cyclin B1 and CDK2 were diminished by 6,7,4′-THIF treatment in all three cell lines. Also, Rb phosphorylation was decreased by 6,7,4′-THIF (Fig. 5B). This result agrees with data from a previous study that showed direct regulation of Rb function by Pin1 (38). These data suggest that 6,7,4′-THIF could induce cell-cycle arrest as well as stimulate apoptosis by inhibiting Pin1 activity in esophageal cancer cells.

Our previous studies showed that several proteins were target molecules of 6,7,4′-THIF (14–16). Thus, to evaluate whether the sensitivity of 6,7,4′-THIF depends on differences in the Pin1 protein expression level, we compared the effect of 6,7,4′-THIF in Neu/Pin1 WT and Neu/Pin1 KO MEFs. Pin1 is a key enzyme for Neu-associated tumorigenic actions (39, 40) and thus Neu/Pin1 KO MEFs did not show significant proliferation under 10% serum conditions. The growth rate of Neu/Pin1 WT MEFs was very rapid compared with KO MEFs (Fig. 6A). After treatment with 6,7,4′-THIF, the proliferation of Neu/Pin1 WT MEFs was dose dependently attenuated, whereas proliferation of Neu/Pin1 KO MEFs was not affected. We next investigated the effect of 6,7,4′-THIF on anchorage-independent growth of Neu/Pin1 WT and KO MEFs. Similarly, Neu/Pin1 KO MEFs had little anchorage-independent colony formation compared with Neu/Pin1 WT MEFs (Fig. 6B) and 6,7,4′-THIF reduced colony formation only for Neu/Pin1 WT MEFs. Pin1 downstream molecules, phosphorylated c-Jun, and cyclin D1 were affected by 6,7,4′-THIF in only Neu/Pin1 WT MEFs (Fig. 6C). We also examined the effect of 6,7,4′-THIF on apoptosis in these MEFS. Results indicated that the antiapoptotic Bcl-xL and Bcl-2 expression levels were diminished and proapoptotic Bax and caspase-3 were increased only in Neu/Pin1 WT MEFs (Fig. 6D). The expression of cell-cycle proteins was similar (Fig. 6E). The expression of cyclin E, CDK2, and phosphorylated Rb was reduced by 6,7,4′-THIF only in Neu/Pin1 WT MEFs. Overall, the expression level of Pin1 dictates the effectiveness of 6,7,4′-THIF suggesting that Pin1 is a major target of 6,7,4′-THIF.

We measured tumor growth in mice after injection of Neu/Pin1 WT MEFs or Neu/Pin1 KO MEFs as described in Materials and Methods. Each cell line (1 × 10⁶ cells) was injected into the flank of each mouse and then vehicle alone or vehicle with 6,7,4′-THIF was intraperitoneally administered at the indicated concentrations. The body weight of mice was not significantly changed during the period of the study (Fig. 7A). Results indicated that tumor volume was significantly increased in the Neu/Pin1 WT–injected group and 6,7,4′-THIF reduced the tumor volume. However, the tumor volume in Neu/Pin1 KO–injected mice was not changed (Fig. 7B). Overall, Pin1 plays an important role in tumor growth and is a molecular target of 6,7,4′-THIF in vivo.

Soy isoflavones are well-known phytochemicals with a potential bioactivity against various chronic diseases, such as cancer, diabetes, and obesity (2, 5, 14, 16). Even though the actual bioactive compounds in soy isoflavones are their metabolites (12), only a few studies have begun to reveal the physiologic activities and identify their modes of action (14–16). In the case of esophageal cancer, Pin1 has been clearly correlated with prognosis and poor survival of patients (28). Because Pin1 regulates oncogenes, such as c-Jun, p63α, p63nerb B2 (22, 40), the poor prognosis for patients with esophageal cancer was assumed to be associated with Pin1-related oncogene activation. At least one previous study reported a reduced esophageal cancer incidence in subjects consuming large quantities of soy and other isoflavones (2). We hypothesized that this reduction in esophageal cancer is associated with the inhibition of Pin1 activity by soy isoflavones.

In this study, we found that 6,7,4′-THIF, a major metabolite of daidzein, suppressed esophageal cancer growth by targeting Pin1, which is overexpressed in these cells (Supplementary Fig. S2). The direct binding of 6,7,4′-THIF with Pin1 was shown by using several experimental models, including virtual screening and immunoprecipitation assays. Notably, 6,7,4′-THIF did not bind to Pin1 family proteins, FKBP or cyclophilin A (Supplementary Fig. S1), suggesting a selective and specific binding of 6,7,4′-THIF with Pin1. Virtual screening results indicated that 6,7,4′-THIF could bind at Arg17, Ser18, and Lys97 in the WW domain and Asp112, Ser154, and Gln131 at the PPIase domain. We point-mutated Arg17 in the WW domain to alanine to verify our virtual screening result (Fig. 1A) and the direct binding between 6,7,4′-THIF and Pin1 was shown to be weaker in R17A Pin1-transfected Neu/Pin1 KO MEFs (Fig. 1E). To evaluate the effect of 6,7,4′-THIF on Pin1 activity, we estimated the proline isomerase activity by using a phosphorylated peptide, WFY(pS)PR-pNA based on a previous study (30). The 6,7,4-THIF compound dose dependently suppressed PPIase activity of a recombinant Pin1 protein (Fig. 2A) and this reduction was confirmed in ex vivo experiments using Pin1 WT MEFs and Neu/Pin1 MEFs lysates (Fig. 2B and C).

Pin1 is known to be an effective target for anticancer drug development because of its modulation of cell cycle or apoptosis-related proteins (20, 25–27). In agreement with previous studies, we found that 6,7,4′-THIF inhibited Pin1 activity resulting in suppression of downstream molecules in KYSE 30, 450, and 510 esophageal cancer cells (Fig. 2D). In addition, a reduction in tumor size and induction of apoptosis were observed in Pin1-depleted PLC/PRF/5 cell–injected mice (21). Because 6,7,4′-THIF directly inhibits Pin1 activity, we examined its effect on cell growth, apoptosis, and cell cycle in several esophageal cancer cell lines (Figs. 3–5) and in vivo (Fig. 7). Notably, it was more effective than a similar soy isoflavone, 7,8,4′-THIF, which had little effect on proliferation of esophageal cancer cells (Supplementary Fig. S3).

In conclusion, we report that 6,7,4′-THIF exerts anticarcinogenic activity by acting as a Pin1 inhibitor. Overall, these results show that 6,7,4′-THIF could be a natural antiesophageal cancer agent.

No potential conflicts of interest were disclosed.

Conception and design: T.-G. Lim, S.-Y. Lee, F. Liu, K.W. Lee, Z. Dong

Development of methodology: Z. Dong

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T.-G. Lim, Z. Duan, F. Liu, K. Liu, S.K. Jung

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T.-G. Lim, S.-Y. Lee, Z. Duan, H. Chen, A.M. Bode

Writing, review, and/or revision of the manuscript: T.-G. Lim, H. Chen, A.M. Bode, K.W. Lee, Z. Dong

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S.-Y. Lee, M.-H. Lee, K. Liu, D.J. Kim, K.W. Lee

Study supervision: K.W. Lee, Z. Dong

This work was supported by The Hormel Foundation and National Institutes of Health grants CA166011 (to Z. Dong), CA187027 (to Z. Dong), and CA196639 (to Z. Dong) and by the Mid-career Researcher Program (2015R1A2A1A10053567) (to K.W. Lee) through the National Research Foundation (NRF) grant funded by the Ministry of Science, ICT & Future Planning, Republic of Korea and also was supported by a grant from the Korea Food Research Institute.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.