Targeting the EZH2-PPAR Axis Is a Potential Therapeutic Pathway for Pancreatic Cancer

Enhancer of zeste homolog 2 (EZH2) is abnormally highly expressed in pancreatic cancer (PC). However, it is not ideal to treat PC by inhibiting EZH2. This study reported that the combined use of pan-peroxisome proliferator-activated receptor (PPAR) agonist could significantly improve the anti-PC effect of EZH2 inhibitor. In vitro, PC cell lines PANC-1 and AsPC-1 were cultured, and MTT and flow cytometry were performed to observe the effects of pan-PPAR agonist bezafibrate and EZH2 selective inhibitor GSK126 on cell viability and apoptosis. In vivo, CDXs of PANC-1 and AsPC-1 were established to observe the effects of bezafibrate and GSK126 on bearing tumors. Western blotting was performed to detect the protein expressions of H3K27me3, β-catenin, p-β-catenin, cyclin D1, c-Myc, and cleaved caspase 3 in vitro and in vivo. The results showed that bezafibrate significantly improved the effects of GSK126 on proliferation inhibition and apoptosis promotion in vitro and the growth suppression of CDX tumors in vivo. It also significantly enhanced the effects of GSK126 on upregulating the expression level of p-β-catenin and that of cleaved caspase 3 in vitro and in vivo. In parallel, downregulation of the expression levels of H3K27me3, β-catenin, cyclin D1, and c-Myc was also observed in vitro or in vivo. These results suggest that the combination of bezafibrate and GSK126 has synergistic effects on PC, and the molecular mechanism may be related to the enhanced inhibition of the Wnt/β-catenin signaling pathway. We believe that targeting the EZH2-PPAR axis is a potential therapeutic pathway for PC.


Introduction
The incidence of pancreatic cancer (PC) ranks 14th in the world, and it has become the 7th leading cause of cancer death in the world due to its extremely high degree of malignancy [1]. The incidence of PC in China has been increasing year by year, and it ranked the 6th cause of cancer death in the country in 2015 [2]. PC has a short course of disease and rapid development. Its early characteristics are always not obvious, and it is already at an advanced stage at the time of diagnosis. In addition, it is difficult to treat in the late stage, which makes high clinical mortality rate of PC [3]. At pres-ent, surgical resection is the only effective method for patients with PC to obtain a cure and long-term survival, while most patients with PC lose the opportunity for surgery due to the late disease stage [4]. What is more, the molecular mechanism of PC is still unclear, which limits the drug development and precise treatment of PC [4,5].
Gene expression patterns determine the fate of cells, and abnormal gene transcriptions affect the phenotype of cancers [6]. Epigenetic regulation of transcription is a complex dynamic and plastic process, and epigenetic disorder is one of the basic characteristics of cancers [7]. The family of polycomb group genes (PCGs) is a group of important epigenetic regulators that suppress transcription. Polycomb repressive complex 2 (PRC2) is one of the core protein complexes of PCGs including enhancer of zeste 2 (EZH2), suppressor of zeste 12 (Suz12), embryonic ectoderm development (Eed), and the histone binding proteins RbAp46/RbAp48 and mediates gene silencing by regulating chromatin structure [8]. EZH2 is an enzyme catalytic subunit of PRC2 and mediates methylation of Lys27 in histone 3 (H3K27me), including mono-H3K27me (H3K27me1), di-H3K27me (H3K27me2), and tri-H3K27me (H3K27me3) [9,10]. This methylated H3K27me3 chromatin mark is commonly associated with silencing of differentiation genes in organisms ranging from plants to flies to humans [11]. More and more evidences show that EZH2 is involved in the abnormal transcription of cancer cells and is related to the poor prognosis of a variety of human malignancies. Overexpression of EZH2 leads to increases in H3K27me3, which inhibits the genes related to tumor suppression and cell differentiation (among numerous others) [12]. Inhibiting EZH2 has potential therapeutic effects on a variety of cancers [13]. Huang et al. have confirmed the abnormal overexpression of EZH2 in PC and found that the EZH2 inhibitors have certain therapeutic effects on PC, while the effects were limited [14]. In the preliminary experiments, we found that the anti-PC effect of EZH2 inhibitor GSK126 can be significantly improved by combining bezafibrate, an agonist of pan-peroxisome proliferator-activated receptors (PPAR). This study reported the therapeutic effect of GSK126 combined with bezafibrate on PC in vitro and in vivo.

Cell
Culture. Human normal pancreatic ductal epithelial cell line of H6C7 and human PC cell lines of PANC-1 and AsPC-1 were all purchased from Procell (Wuhan, HB, CHN). The cells of PANC-1 were grown in DMEM with 10% calf bovine serum and 1% penicillin-streptomycin at 37°C with 5% CO 2 (v/v). The cells of H6C7 and AsPC-1 were grown in RPMI-1640 with 10% calf bovine serum and 1% penicillin-streptomycin at 37°C with 5% CO 2 (v/v). Medium was replaced two to three days, and the cells were passaged when the cell adherence area reached 80% of the culture dish.

Experimental Groups and Treatments
In Vitro. Cell lines of PANC-1 and AsPC-1 were carried out as independent experiments and grouped as follows: (1) control group (control), in which the cells were treated with vehicle in medium for 120 h; (2) bezafibrate treatment group (Bez), in which the cells were treated with different concentrations of bezafibrate (0, 0.125, 0.25, 0.5, 1, and 2 mM) in medium for 120 h; (3) GSK126 group (GSK126), in which the cells were treated with different concentrations of GSK126 (0, 2.5, 5, 10, 20, and 40 μM) in medium for 120 h; and (4) bezafibrate combination with GSK126 group (Bez+GSK126), in which the cells were treated with bezafibrate and GSK126 together in medium for 120 h.
2.5. Apoptosis Assay. Trypsin-EDTA was used to obtain single-cell suspension after incubating for 10 min, and then, the cells were washed with chilled D-Hanks (pH = 7:2~7:4) after centrifugation and incubated in Annexin-V binding buffer for 15 min at room temperature, containing Annexin-V-FITC. Flow cytometry (Becton Dickinson, USA) was used to quantify the fluorescence of Annexin-V-FITC with a minimum of 10,000 cells counted for each group [15].
2.6. Animals. 18 to 22 g (4-6 weeks) adult male Balb/c nude mice were obtained from Nanchang University Laboratory Animal Center. Mice were allowed to access food and water ad libitum under a 12 h dark/light cycle at 22°C to 25°C. Experiments were carried out according to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised in 1996) and approved by the Ethics Committee of Nanchang University (No. 2019-0402).

Cell-Derived Xenograft (CDX).
Lidocaine was applied to the skin of the left forelimb scapula of nude mice after skin preparation by iodophor; then, 100 μl cell suspension (1 × 10 6 /ml) was extracted by a microinjector and injected into the subcutaneous area of the scapula of nude mice (P 0 ). The nude mice were fed routinely, and the tumor formation was continuously observed. When the bearing tumor growth reached 1000 mm 3 , CO 2 euthanasia was performed. In aseptic environment, the tumor was completely peeled off and cut into tissue blocks of uniform size (2 × 2 × 2 mm ). After skin preparation with iodophor and local anesthesia with lidocaine, single tissue blocks were inoculated subcutaneously into the subcutaneous area of the scapula of normal nude mice (P 1 ). When the bearing tumor growth reached 1000 mm 3 , normal adult mice were inoculated according to the above method (P 2 ). CDXs of P 3 were used for formal experiments.

Experimental Groups and Treatments
In Vivo. CDXs of PANC-1 and AsPC-1 were carried out as independent experiments. The P 3 mouse bearing tumors with tumor volume of about 100-200 mm 3 were selected and divided into the following groups: (1) control group (control), in which the mice were treated with vehicle daily; (2)

Results
3.1. EZH2 Is Highly Expressed in Cells of PANC-1 and AsPC-1. Western blotting was used to detect the protein expressions of EZH2, H3K27ac, and H3K27me3 in cells of H6C7, PANC-1, and AsPC-1. As shown in Figures 1(a)-1(d), compared with H6C7, the proteins of EZH2 and H3K27me3 were significantly upregulated in PANC-1 and AsPC-1, while H3K27ac was significantly downregulated (P < 0:05). The qPCR was used to detect the mRNA transcription level of EZH2, and the result showed that the relative mRNA levels of EZH2 were significantly upregulated in PANC-1 and AsPC-1 (Figure 1(e)), which is consistent with the results of western blotting.

Effects of Combination of Bezafibrate and GSK126 on Cell
Viability of PANC-1 and AsPC-1. Different concentrations of bezafibrate or GSK126 were used to treat PANC-1 and AsPC-1 cells. As shown in Figure 2, when the concentration of bezafibrate reached 2 mM, the cell viability was significantly inhibited. GSK126 could significantly inhibit the viability of cells in a dose-dependent manner, and the IC50 of PANC-1 and AsPC-1 was 16 μM and 24 μM, respectively. What is important, the combination of bezafibrate and GSK126 could inhibit cell viability in a dosedependent manner. Their corresponding CI plot analysis showed that the combination of bezafibrate and GSK126 inhibited cell viability synergistically at the majority of concentrations. The combined drug treatment of 1 mM bezafibrate and 20 μM GSK126 marked a transition from

PPAR Research
drug concentrations that prevented the growth of cancer cells only to a concentration that effectively prevented the growth of cancer cells. Therefore, these concentrations were examined further.

Effects of Combination of Bezafibrate and GSK126 on
Apoptosis in Cells of PANC-1 and AsPC-1. As shown in Figure 3, bezafibrate alone could not significantly promote the apoptosis of PANC-1 or AsPC-1 compared with the control (P > 0:05), while GSK126 alone treatment could signifi-cantly increase the apoptosis levels of PANC-1 and AsPC-1 (P < 0:05 vs. the control group). The combination of bezafibrate and GSK126 could enhance the proapoptotic effects of GSK126 on PANC-1 or AsPC-1 significantly (P < 0:05 vs. the GSK126 group).

Effects of Combination of Bezafibrate and GSK126 on
Protein Expressions in Cells of PANC-1 and AsPC-1. As shown in Figure 4, bezafibrate alone has no significant effect on H3K27me3 (P > 0:05 vs. the control group), while

PPAR Research
PANC-1 or AsPC-1 (P > 0:05 vs. the control group), while GSK126 treatment alone could significantly inhibit the growth of bearing tumors of PANC-1 or AsPC-1 (P < 0:05 vs. the control group). When comparing with the group of GSK126, bezafibrate in combination with GSK126 could more significantly inhibit the growth of bearing tumors of PANC-1 or AsPC-1 (P < 0:05). Figure 6, bezafibrate treatment has no obvious effects on methylation of histone H3, and GSK126 treatment could significantly prevent the methylation of histone H3 and downregulate the expression level of H3K27me3 (P < 0:05 vs. the control group). The combination of bezafibrate and GSK126 could further downregulate the expression level of H3K27me3 (P < 0:05 vs. the GSK126 group). Both bezafibrate and GSK126 could significantly upregulate the expressions of p-β-catenin and cleaved caspase 3 and downregulate the expressions of β-catenin, cyclin D1, and c-Myc (P < 0:05 vs. the control group), and the combination of bezafibrate and GSK126 could further affect the expressions of above proteins accordingly (P < 0:05 vs. the GSK126 group).

Discussion
In this study, we found that EZH2 was abnormally overexpressed in PC cell lines, suggesting that EZH2 is closely related to PC. GSK126, a selective inhibitor of EZH2, can significantly inhibit the proliferation and promote the apoptosis of PC cells. The pan-PPAR agonist bezafibrate alone has no significant effect on the proliferation and apoptosis of PC, but its combined use with GSK126 could further enhance the anti-PC cancer effect, suggesting that GSK126 and bezafibrate have synergistic effects at certain concentrations. To our knowledge, this is the first report on the synergistic effect of GSK126 and bezafibrate on PC.
Peroxisomes were first discovered in the liver of rats and participated in cellular functions such as fatty acid metabolic transport and ROS detoxification [18]. Peroxisome proliferators can induce the proliferation of peroxisomes, and the receptors of proliferators are generally referred to as PPARs, which are involved in activating/inhibiting target genes [19]. PPARs contain three subtypes including γ, α, and β/δ, which are widely expressed in human cells [20]. PPARs play important roles in the regulation of cell differentiation, development, metabolism, and tumorigenesis. Pro12Ala polymorphism in PPAR γ was found to be a risk factor for gastric   [21]. PPAR α and γ have been found to be associated with the risk of breast cancer [22]. Recent studies have shown that the expression levels of PPAR pathway genes MED1, PRKCA, and PRKCB in PC are higher than normal tissues, suggesting that PPAR may contribute to susceptibility to PC [23]. In mammalian cells, PPARs and the Wnt/β-catenin signaling pathway have the opposite manner. PPAR agonists can inhibit nuclear translocation of β-catenin, thus limiting the regulation of the Wnt/β-catenin signaling pathway, and conversely, the inhibition of the Wnt/β-catenin signaling pathway can induce PPAR activation [24,25]. Wnt/β-catenin signal plays an important role in cell fate, epithelial-mesenchymal transition signal transduction, and embryonic development. Its dysfunction involves a variety of diseases, including PC [26]. PPAR agonists are commonly used in the treatment of diabetes. Since the target organ of both diabetes and PC is   [27,28]. In this study, we observed that bezafibrate alone, an agonist of pan-PPAR, had certain inhibitory effects on PC cells PANC-1 and AsPC-1 in vivo and in vitro with no significance. Western blotting results showed that bezafibrate could significantly inhibit the protein expressions of β-catenin, cyclin D1, and c-Myc and increase the expression of p-β-catenin in PANC-1 and AsPC-1. These suggested that activation of PPARs has a limited therapeutic effect on PC, and its limited effect may be related to its ability to inhibit the activity of the Wnt/βcatenin pathway in PC cells.
Studies have shown that the EZH2-PRC2 complex controls multiple target genes in the Wnt pathway at the epigenetic level [29]. In cervical cancer, EZH2 activates the Wnt/β-catenin pathway through epigenetic silencing of GSK-3β and TP53, thereby promoting cervical cancer cell proliferation and tumor formation [30]. In the mechanism of EZH2-mediated activation of Wnt/β-catenin signal to promote carcinogenesis, EZH2 acts as a transcription repressor, inhibiting Wnt antagonists by reducing the acetylation of H3K27, thereby promoting the development of different types of cancers [31,32]. Conversely, EZH2 knockout can induce G2/M phase arrest of breast cancer and regulate the expressions of cyclin D1 and β-catenin [33]. By inhibiting EZH2 to interfere with the Wnt/β-catenin pathway, the progression of glioma can be inhibited [34]. Inhibition of EZH2 can also reduce the expression of apoptosis-inhibiting genes by inhibiting mTOR activity and aggravate cisplatin-induced apoptosis [35]. In this study, the EZH2 inhibitor GSK126 can significantly inhibit the methylation level of histone H3, prevent the proliferation of PC cells in vitro and in vivo, and promote the expressions of apoptosisrelated protein to aggravate apoptosis of cancer cells. Meanwhile, GSK126 could significantly downregulate the expression of Wnt/β-catenin pathway key protein β-catenin and promote its phosphorylation. It should be noted that β-catenin is bound to and phosphorylated by a destructive complex that resides within the cytoplasm. Upon phosphorylation, β-catenin leaves the complex to be further ubiquitinated by beta-transducin repeat-containing proteins and degraded by the proteasome [36]. Combining with bezafibrate can significantly enhance the anti-PC effects of GSK126 and further inhibit the Wnt/βcatenin signaling pathway. We speculate that the synergistic anti-PC effects of GSK126 and bezafibrate are related to their ability to synergistically inhibit the activity of the Wnt/βcatenin pathway.
In conclusion, we have proved that bezafibrate, an agonist of pan-PPAR, can significantly enhance the proliferation inhibition and apoptosis promotion of GSK126, a selective inhibitor of EZH2, on PC cells in vivo and in vitro. The combination of bezafibrate and GSK126 enhanced the inhibition of Wnt/β-catenin pathway activity, which may be the molecular mechanism of bezafibrate and GSK126 in synergistic anti-PC. The above results suggest that targeting the EZH2-PPAR axis may be a potential treatment for PC. However, the biggest limitation of this study is inadequate exploration in the possible protein interaction between EZH2 and PPAR, which will be the focus of future research.

Data Availability
The datasets generated and analyzed during the study are available from the corresponding author on request.