SYPL1 Inhibits Apoptosis in Pancreatic Ductal Adenocarcinoma via Suppression of ROS-Induced ERK Activation

Synaptophysin-like 1 (SYPL1) is a neuroendocrine-related protein. The role of SYPL1 in pancreatic ductal adenocarcinoma (PDAC) and the underlying molecular mechanism remain unclarified. Here, after analyzing five datasets (GSE15471, GSE16515, GSE28735, TCGA, and PACA-AU) and 78 PDAC patients from Sun Yat-sen University Cancer Center, we demonstrated that SYPL1 was upregulated in PDAC and that a high level of SYPL1 indicated poor prognosis. Bioinformatics analysis implied that SYPL1 was related to cell proliferation and cell death. To validate these findings, gain-of-function and loss-of-function experiments were carried out, and we found that SYPL1 promoted cell proliferation in vitro and in vivo and that it protected cells from apoptosis. Mechanistic studies revealed that sustained extracellular-regulated protein kinase (ERK) activation was responsible for the cell death resulting from knockdown of SYPL1. In addition, bioinformatics analysis showed that the expression of SYPL1 positively correlated with antioxidant activity. Reactive oxygen species (ROS) were upregulated in cells with SYPL1 knockdown and vice versa. Upregulated ROS led to ERK activation and cell death. These results suggest that SYPL1 plays a vital role in PDAC and promotes cancer cell survival by suppressing ROS-induced ERK activation.


INTRODUCTION
Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease with a dismal 5-years survival rate of <10% in the USA (1) and is predicted to be the second leading cause of cancer-specific death by 2030 (2). Surgical resection can be performed in only ∼20% of PDAC patients, whereas most patients suffer from locally advanced or metastatic diseases due to the insidious onset and early metastasis of PDAC. Even after resection with curative intent, nearly 80% of patients experience recurrence (3). To improve clinical outcomes, it is vital to deepen the understanding of PDAC mechanisms and identify new therapeutic targets.
Synaptophysin-like 1 (SYPL1), which belongs to the synaptophysin (SYP) family, was originally regarded as a neuroendocrine-related protein (4) and is expressed in both neuronal and non-neuronal tissues (5). SYPL1 is reported as a component of transport vesicles, which associate with insulin-responsive glucose transporter type 4 (GLUT4)-containing vesicles in adipocytes (6) and podocyte exosome-enriched fraction in urine (7). In addition, SYPL1 may be a regulator of the NF-κB pathway in a genome-wide siRNA screen (8). Recently, an immunohistochemistry (IHC)based study showed that SYPL1 was a prognostic factor of poor prognosis of hepatocellular carcinoma and was related to epithelial-mesenchymal transition (9). In a bioinformatics-based study, SYPL1 was predicted to play a vital role in papillary thyroid carcinoma (10). However, the mechanism by which SYPL1 promotes the initiation and progression of tumors and the role of SYPL1 in PDAC remain unclarified.
In this study, significant upregulation of SYPL1 was detected in PDAC patient tumor tissue, which indicated poor prognosis. Knockdown of SYPL1 inhibited proliferation, induced apoptosis of tumor cells, and vice versa. Mechanistically, knockdown of SYPL1 increased reactive oxygen species (ROS) and sustainably activated extracellular-regulated protein kinases (ERKs). We demonstrated that SYPL1 promoted PDAC progression through its regulation of the ROS/ERK pathway.

Patients and Tissue Specimens
This study was approved by the ethics committee of Sun Yat-sen University Cancer Center (SYSUCC). Tissue samples were collected from 78 PDAC patients who underwent curative resections in SYSUCC from March 2008 to November 2017. Preoperative chemotherapy or radiotherapy was not performed. The last follow-up data were from October 14, 2019. Overall survival (OS) was defined as the duration from surgery to death or last follow-up. The clinical features of the patients are summarized in Table 1. After excision, tissue samples were formalin fixed and paraffin embedded.
of Sciences (Shanghai, China), and the immortal human pancreatic duct epithelial cell line HPDE6-C7 was provided as a gift from Professor Dongxin Lin from SYSUCC. Cells were cultured at 37 • C in a humidified atmosphere of 5% CO 2 . The culture medium was used as recommended and was supplemented with 10% fetal bovine serum (Gibco, California, USA). Cell lines with SYPL1 stably silenced or overexpressed were established using lentiviruses (iGeneBio, GuangZhou, China). We also used siRNA (RiboBio, Guangzhou, China), which

Cell Proliferation and Colony Formation Assay
Cell proliferation was assessed using the Cell Counting Kit-8 assay (CCK8, Dojindo, Tabaru, Japan) following the instructions. Cells were seeded in 96-well plates (1,000-2,000 cells/well). The incubation time of the CCK8 solution was 2 h.
Assessments of colony formation were performed as described before (11). Cells were seeded in 6-well plates (500 cells/well) and cultured for 2 weeks. KeyGen Biotech, Nanjing, China) for 2 h. Cells were trypsinized and washed twice with phosphate-buffered saline (PBS). We used the Annexin V-APC/PI Apoptosis Detection Kit (KGA1030-100, KeyGen Biotech, Nanjing, China) to assess apoptosis according to the manufacturer's instructions. Data were acquired using flow cytometry. For the ROS assay, the ROS fluorescence probe-DHE (KGAF019) was obtained from KeyGen BioTech (Nanjing, China) and flow cytometry was applied.

Animal Experiments
All animal experiments were approved by the ethics committee of SYSUCC. Six-weeks-old Crj:BALB/c female athymic nude mice were purchased from Vital River (Beijing, China). A xenograft model was established by subcutaneously injecting 5 × 10 6 cells into the left flank of the nude mice. We measured the size of the tumor every 7 days using calipers (volume = 0.5 × length × width × width). After 5-7 weeks, mice were euthanized, and xenografts were removed, weighed and preserved.

Statistical and Bioinformatics Analysis
Gene set enrichment analysis (GSEA) was performed using software downloaded from https://www.gsea-msigdb.org/gsea/ index.jsp (16,17). Gene set variation analysis (GSVA) was carried out using the R package "GSVA" (18). GSEA and GSVA were used in pathway analysis. Details of the statistical analysis were described in our previous study (19). Graph Pad Prism 7.0 software, R 3.5.1 and SPSS 23 were used.

SYPL1 Is Upregulated in PDAC
We analyzed the gene expression of SYPL1 in GEO (GSE15471, GSE16515, and GSE28735), TCGA and GTEx. SYPL1 was elevated in tumor tissue at the level of transcription ( Figure 1A). Forty-two tissue sections from SYSUCC contained both PDAC and pancreatic acini. IHC results confirmed the upregulation of SYPL1 in PDAC at the protein level (Figures 1B,C). SYPL1 was also elevated in PDAC cell lines compared to its expression in the HPDE6-C7 pancreatic duct epithelial cell line based on WB and qPCR assays (Figures 1D,E).

Explorations of KEGG Pathways Related to SYPL1
Five datasets (GSE15471, GSE16515, GSE28735, TCGA, and PACA-AU) were analyzed. Pathway activity was calculated using GSVA. Correlation coefficients (r) between SYPL1 and the activity of KEGG pathways were calculated. Twenty-six pathways with r > 0.1 in 5 datasets or pathways with r < −0.1 in 5 datasets were identified as pathways related to SYPL1 ( Figure 3A). Two groups was defined based on SYPL1 expression (upper third vs. bottom third) in each dataset. The activities of the cell cycle pathway, DNA replication pathway and p53 pathway were higher in the SYPL1-high group, which indicated that SYPL1 was related to the proliferation and survival of PDAC cells (Figures 3B-D,  Supplementary Figures 1,2).

Altered SYPL1 Expression Affects Cell Proliferation in vitro and in vivo
We infected BXPC-3 and PANC-1 cells with lentivirus carrying shRNA targeting SYPL1 (shSYPL1) or control shRNA (shCon). We infected the two cell lines with lentivirus carrying SYPL1 expression vector (overSYPL1) or control empty vector ( Figure 4A). We also used siRNA to knockdown SYPL1 (Supplementary Figure 3). The CCK8 assay showed that knockdown of SYPL1 significantly reduced the proliferation of BXPC-3 and PANC-1 cells, while overexpression of SYPL1 did the opposite (Figure 4B). In the colony formation assay, we found that knockdown of SYPL1 significantly reduced the colony formation ability of BXPC-3 and PANC-1 cells and that overexpression of SYPL1 did the opposite (Figure 4C,  Supplementary Figure 3). To assess the effects of SYPL1 on cell proliferation in vivo, a subcutaneous tumor model in BALB/c

Knockdown of SYPL1 Promoted Apoptosis by Activating ERK
To explore the regulatory mechanisms of SYPL1, we searched for genes closely correlated with SYPL1 in pancreatic adenocarcinoma datasets of GEPIA. We found that the expression of KRAS was positively correlated with SYPL1 (r = 0.7), and similar results were seen in other datasets (Figures 6A,B), which implied that SYPL1 was related to the MAPK pathway. Phosphorylated ERK (pERK), phosphorylated P38 (p-P38) and phosphorylated JNK (pJNK) were assessed using WB. pERK was downregulated in BXPC-3 cells overexpressing SYPL1, while no obvious change was observed in pJNK or p-P38 ( Figure 6C). In addition, upregulated pERK and downregulated ERK were observed in shSYPL1 cells, which indicated that ERK was sustainably activated in cells with SYPL1 knockdown (Figure 6D, Supplementary Figure 4). To clarify whether downregulation of SYPL1 inhibited proliferation and promoted apoptosis via sustainable ERK activation, selumetinib (AZD6244, Selleck Chemicals, Houston, TX, USA, 0.4 µM for BXPC-3, 2 µM for PANC-1), a MEK inhibitor, was used ( Figure 6D). After shSYPL1 cells were treated with selumetinib, no obvious change was observed in cell proliferation (Supplementary Figure 3). However, selumetinib significantly reduced the number of apoptotic cells in the shSYPL1 cell line (Figure 6E), which meant that knockdown of SYPL1 resulted in apoptosis through sustainable ERK activation.

Knockdown of SYPL1 Activates ERK by Increasing ROS
Several studies reported that sustainable activation of ERK, which promoted cell death, resulted from increased ROS (34)(35)(36)(37)(38)(39). In Figure 3A, GSVA scores of both the pentose phosphate pathway (PPP) and peroxisome were positively correlated with the RNA level of SYPL1, which indicated that SYPL1 was involved in oxidative stress. GSEA on TCGA reconfirmed the relationship between SYPL1 and PPP, which was a major source of NADPH ( Figure 7A). Increased GSVA scores of PPP were observed in SYPL1-high samples (Figure 7B). Glucose-6-phosphate dehydrogenase (G6PD) and phosphogluconate dehydrogenase (PGD) are NADPH-producing enzymes in the PPP (40). Generally, SYPL1 was positively correlated with both G6PD and PGD at the transcriptional level (Figures 7F,G). The GSVA score of peroxisomes was higher in SYPL1-high samples ( Figure 7C). Peroxisomes have protective measures to counteract oxidative stress (41). At the transcriptional level, SYPL1 positively correlated with genes reported to possess antioxidant activity in the peroxisome gene set (Figure 7H) (41)(42)(43)(44)(45)(46). In addition, a gene set named "GO_ANTIOXIDANT_ACTIVITY, " which contained components that can trap free radicals, was analyzed. The GSVA score of antioxidant activity was positively correlated with SYPL1, and a higher GSVA score was seen in SYPL1-high patients (Figures 7D,E). The results above suggested that SYPL1 was likely to help cope with oxidative stress. To validate this hypothesis, we used flow cytometry to assess intracellular ROS. The ROS level of SYPL1-silenced BXPC-3 cells was upregulated compared to that of shCon BXPC-3 cells, while PANC-1 cells overexpressing SYPL1 had decreased ROS levels ( Figure 7I). Then, BXPC-3 and PANC-1 cells were treated with hydrogen peroxide (H 2 O 2 ), which was reported to increase intracellular ROS and mimicked the effect of downregulated SYPL1 on intracellular ROS (47). Upregulation of pERK and increased apoptotic cells were found in H 2 O 2 -treated cells, and Selumetinib significantly reduced the number of apoptotic cells (Figures 7J,K, Supplementary Figure 4). Overall, knockdown of SYPL1 upregulated ROS, which led to activation of ERK and cell death.

DISCUSSION
The present study demonstrates that SYPL1, which is upregulated in tumor tissue and PDAC cell lines at the transcriptional level and protein level, is an independent factor associated with poor prognosis. Data analysis based on five datasets (GSE15471, GSE16515, GSE28735, TCGA, and PACA-AU) shows that SYPL1 is associated with the proliferation and survival of cancer cells. SYPL1 is silenced or upregulated in BXPC-3 and PANC-1 cells. Through CCK-8, colony formation and subcutaneous xenotransplanted tumor models in nude mice, we found that downregulation of SYPL1 inhibits tumor growth in vitro and in vivo, while overexpressed SYPL1 promotes cell proliferation. Data analysis and flow cytometry show that SYPL1 protects tumor cells from apoptosis. The prosurvival effects of SYPL1 result from the suppression of sustainably activated ERK by intracellular ROS. To our knowledge, this is the first study to demonstrate the enhancing effects of SYPL1 on cell proliferation and survival in tumors. This is also the first study that reports the expression of SYPL1 in PDAC and clarifies the mechanism of its anti-apoptotic effects. The MAPK-ERK pathway, which requires delicate regulation of its spatiotemporal activity, is a double-edged sword in tumorigenesis. Activated ERK promotes cell survival as a result of its oncogenic potential. Paradoxically, aberrant activation of ERK is reported to promote cell death (48)(49)(50). The proapoptotic effect of the MAPK pathway was first reported in 1996. In that study, Taxol-induced apoptosis depended on Raf-1 activation (51). Later, an increasing number of studies found that DNA-damaging agents and antitumor compounds induced cell death by activating ERK, which could be rescued using a MEK inhibitor (51)(52)(53)(54). In addition, constitutive activation of ERK by death-associated protein kinase 1 or Raf-1 can lead to cell death without other stimuli (55). In this study, by combining bioinformatics analysis and experiments, we demonstrated that downregulation of SYPL1 led to increased apoptosis and ERK activation and vice versa. After cells were treated with selumetinib, a MEK inhibitor, ERK was inactivated, and the number of apoptotic cells was decreased, which demonstrated that knockdown of SYPL1 activated ERK, resulting in cell death. ROS are a group of chemical reactive molecules with vital roles in cell proliferation and differentiation and include hydroxyl, superoxide, and H 2 O 2 (56). However, excessive ROS result in oxidative damage to DNA and protein and induce cell death (57). Some studies have reported that ERK activation is involved in this process and that MEK inhibitors suppress ROS-induced cell death (58,59). Usually, ERK activation is precisely controlled and not prolonged. Sustained activation of ERK is required to induce cell death (60). A review of ERK-induced cell death concluded that ROS-mediated sustained activation of ERK was a crucial mechanism in this process (48). Here, we found that SYPL1silenced cells had higher levels of ROS and vice versa. Increased ROS contributed to ERK activation and cell apoptosis.
Several antioxidant systems, such as NADPH, SOD, and catalase, assist cells in maintaining redox balance and preventing excessive ROS (61). PPP is a main source of NADPH, which is an ROS scavenger (40). G6PD and PGD are enzymes generating NADPH in the PPP. In addition, there are antioxidant enzymes in peroxisomes, such as SOD1, catalase, PRDX5, and GSTK1 (41). These protect cells from oxidative stress. We analyzed five datasets at the transcription level and found that SYPL1 expression positively correlated with the expression of PPPrelated and peroxisome-related antioxidant genes, especially G6PD, PGD, SOD1, and CAT, which was a reflection of the antioxidant status. Generally, SYPL1 expression positively correlated with antioxidant activity, which was consistent with the upregulation of ROS in SYPL1-silenced cells.
We demonstrated that SYPL1 promoted cell proliferation in vitro and in vivo. Decreased tumor sizes of shSYPL1 cells were observed in a subcutaneous tumor model in BALB/c nude mice. However, in IHC, the expression of SYPL1 did not correlate with the tumor sizes of patients in SYSUCC, which may result from insufficient sample size and sampling error. In addition, PDAC is characterized by desmoplastic stroma, which comprises up to 80% of the tumor mass (62). The subcutaneous xenotransplanted tumor model may not fully reflect the tumor-stroma interaction in the pancreas (63). As a result, the stroma, which comprised the major part of the PDAC in patients, was different in the subcutaneous tumor model in nude mice, which may account for the discrepancy between clinical data and the mouse model. In addition, the effects of SYPL1 on the stroma will be studied.
In summary, this study demonstrated that SYPL1 is upregulated in PDAC, indicating poor prognosis and promoting cell proliferation as well as survival. The mechanism of its antiapoptotic effect is suppression of ROS-induced ERK activation.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the ethics committee of Sun Yat-sen University Cancer Center. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the ethics committee of Sun Yat-sen University Cancer Center. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

AUTHOR CONTRIBUTIONS
SL and YS: study concept and design. YS, XS, and FD: acquisition, analysis, or interpretation of data. YS and XS: drafting of the manuscript. YS, JiW, and JuW: critical revision of the manuscript for important intellectual content. YS, CH, and RW: statistical analysis. SL: obtained funding. XH, YM, FX, KX, and SS: administrative, technical, or material support. SL and JuW: study supervision. All authors have read and approved the final version of the manuscript.