Haspin knockdown can inhibit progression and development of pancreatic cancer in vitro and vivo

Journal Pre-proof Haspin knockdown can inhibit progression and development of pancreatic cancer in vitro and vivo Xu Han, Yun Ren, Zhiyao Wang, Qingwu Liao, Wei Chen PII:

S0014-4827(19)30465-3

DOI:

https://doi.org/10.1016/j.yexcr.2019.111605

Reference:

YEXCR 111605

To appear in:

Experimental Cell Research

Received Date: 16 May 2019 Revised Date:

14 August 2019

Accepted Date: 3 September 2019

Please cite this article as: X. Han, Y. Ren, Z. Wang, Q. Liao, W. Chen, Haspin knockdown can inhibit progression and development of pancreatic cancer in vitro and vivo, Experimental Cell Research (2019), doi: https://doi.org/10.1016/j.yexcr.2019.111605. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.

Haspin knockdown can inhibit progression and development of pancreatic cancer in vitro and vivo Running title: Haspin promotes pancreatic cancer Xu Han1, MD, Yun Ren2#, MS, Zhiyao Wang1, MD, Qingwu Liao1, MD, Wei Chen2,3*, MS #

These authors contributed equally to this work.

1

Department of Pancreatic Surgery, Zhongshan Hospital, Fudan University; 180

Fenglin Road, Shanghai 200032, China 2

Department of Anesthesia, Zhongshan Hospital affiliated with Fudan University,

Shanghai, China; 180 Fenglin Road, Shanghai 200032, China 3

Department of Anesthesia, Qingpu Branch of Zhongshan Hospital affiliated with

Fudan University, Shanghai, China; No. 1158 Gongyuan East Road, Shanghai 201700, China *

Correspondence to: Wei Chen

Department of Anesthesia, Zhongshan Hospital affiliated with Fudan University, Shanghai, China; Department of Anesthesia, Qingpu Branch of Zhongshan Hospital affiliated with Fudan University, Shanghai, China E-mail: [email protected] Phone number: +86-21-64041990

Abstract Background: Pancreatic cancer is one of the most aggressive and lethal malignancies and it is the eighth most common cause of death from cancer worldwide. The purpose of this study was to investigate the role of GSG2 (HASPIN) in the development and progression of pancreatic cancer. Material and methods: GSG2 expression was detected by immunohistochemistry in tumor tissue samples, and by qRT-PCR and western blot assay in human pancreatic cancer cell lines. Cell proliferation was evaluated by MTT assay. Giemsa staining was used for analyzing colony formation. Cell cycle and cell apoptosis were determined using Fluorescence activated Cells Sorting. Wound healing assay and transwell assay were applied for examining cell migration. The molecular mechanism was investigated by human apoptosis antibody array. Tumor-bearing animal model was constructed to verify the effects of GSG2 on pancreatic cancer in vivo. Results: GSG2 expression was upregulated in pancreatic cancer tissues and human pancreatic cancer cell lines: PANC-1 and SW1990. Higher expression of GSG2 in tumor samples was associated with poorer prognosis. GSG2 knockdown suppressed cell proliferation, colony formation, metastasis and promoted cell apoptosis, which was also verified in vivo. In addition, GSG2 knockdown blocked the cell cycle in G2. It was also found that downregulation of GSG2 inhibited Bcl-2, Bcl-w, cIAP, HSP60 and Livin expression as well as promoted IGFBP-6 expression. Conclusion: This study revealed that GSG2 upregulation was associated with pancreatic cancer progression. GSG2 knockdown inhibited cell proliferation, colony formation and migration, blocked cell cycle at G2 phase, and induced cell apoptosis. Therefore, GSG2 might serve as a potential therapeutic target for pancreatic cancer therapy and a market for prognosis.

Key words: GSG2; pancreatic cancer; apoptosis; metastasis;

Introduction Pancreatic cancer is a serious problem that it is a disease of near uniform lethality, insights into molecular pathogenesis are urgently needed (1). It is one of the most aggressive and lethal malignancies and ranks as the eighth most common causes of death from cancer worldwide (2, 3). Despite many efforts to treating, it still has worst prognosis of all major malignant tumors with a 5-year survival rate of < 5% and a median survival of < 6 months (4, 5). Therefore, it will be of great importance to find effective treatment strategies and explore the mechanisms responsible for improving the methods to save the pancreatic cancer patients. GSG2 (HASPIN) is a threonine-protein kinase that was for a long time considered an inactive pseudokinase due to low degree of structural homology of GSG2 with the 'classical' protein kinases (6, 7), thereby creating a recognition motif for docking of the chromosomal passenger complex that is crucial for the progression of cell division. It is shown that GSG2 over-expression plays an important role in the cancer cells compared to normal cells (8) and it results in defective mitosis cells (9). Huertas D. et al. reported the anti-tumor activity of GSG2 kinase inhibitor (CHR-6494) using mice xenograft model (10). However, the role of GSG2 in pancreatic cancer remains unknown and was rarely reported. In this study, for the first time, we found that the expression of GSG2 was upregulated in pancreatic cancer tissues. In addition, the knockdown of GSG2 could significantly inhibit the proliferation and metastasis of pancreatic cancer, indicating that GSG2 may be a potential therapeutic target of pancreatic cancer.

Materials and methods Patients and tissue specimens Pancreatic cancer tissues and adjacent tissues were purchased from Shanghai Outdo Biotech Co., Itd and the chip lot number was XT14-056. There were 170 samples from 99 patients in the chip and the clinical characteristics of each patient were collected. Informed written consent were obtained from patients of all tissue samples. Ethical approval was obtained from the ethics committee of Zhongshan Hospital

affiliated Fudan University. Cell lines and cell culture The human pancreatic cancer cell lines PANC-1 and SW1990 were purchased from Cell Resource Center, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). SW1990 cells were cultured in Leibovitz Medium (Thermo Fisher Scientific, Waltham, MA, USA) with 10% FBS and PANC-1 cells were cultured in 90% DMEM (GIBCO, Thermo Fisher Scientific, Waltham, MA, USA) with 10% FBS at 37°C in a humidified atmosphere containing 5% CO2. Immunohistochemical staining Pancreatic cancer and adjacent tissue sections from pancreatic cancer tissue chip were collected. The tissue sections were deparaffinized, repaired with citrate antigen, blocked and then incubated with the GSG2 antibody (1:200, Bioss) at 4°C overnight in incubator. Tissue sections were stained with DAB, and again stained with hematoxylin. Images were captured using a photomicroscope and analyzed. Target gene RNA interference lentiviral vector preparation Designing RNA interference target sequences using GSG2 gene as a template to construct a target gene RNA interference 5′ - CCACAGGACAATGCTGAACTT - 3′ lentiviral vector (Shanghai Bioscienceres, Co., Lt., Shanghai, China). Synthesis of single-stranded DNA oligo containing interference sequences, the synthesized single-stranded DNA oligo dry powder was dissolved in an annealing buffer (final concentration: 100 M), and a water bath at 90°C for 15 min. After naturally cooling to room temperature, a double strand with a sticky end was formed, then directly ligating into the digested lentiviral vector through its restriction sites at both ends. Lentivirus containing shGSG2 or shCtrl were transfected into prepared competent cells. Then the cells were culture for 72 h at 37°C and the expression of green fluorescent protein (GFP) was observed under the microscope to evaluate the transfection efficiency. RNA Extraction and qRT-PCR ShGSG2 expression levels were measured by qRT-PCR. Total RNA from each group (shCtrl and shGSG2) was collected with trizol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction, and cDNA was obtained. PCR reaction

system containing cDNA, primer, Premix Taq enzyme (TaKaRa, Dalian, China), and sterile water. PCR conditions were set as 95°C 10 s, 1 cycle; 95°C 5 s, 60°C 30 s, 45 cycle. The relative expression of RNA was calculated with the 2-∆∆CT method, using GAPDH as an internal control. Nanodrop 2000/2000C spectrophotometric was used to analysis the relative levels of GSG2 mRNA. The primer sequences of GSG2 in this study were as follows: forward primer, 5′ GGAAGGGGTGTTTGGCGAAGT

-

3′;

reverse

primer,

5′

-

TGAGGAGCAAGGGAGGGTAAG - 3′. The primer sequences of GAPDH in this study were as follows: forward primer, 5′ - TGACTTCAACAGCGACACCCA - 3′; reverse primer, 5′ - CACCCTGTTGCTGTAGCCAAA - 3′. Western blot analysis In order to detect protein expression levels of GSG2, PANC-1 and SW1990 cells were collected and lysed by using RIPA lysis buffer (Cell Signal Technology, Danvers, MA) on ice according to the manufacturer’s instruction. Then, the total cellular proteins were subjected to SDS-PAGE (10%) for western blot (WB) analysis. After transferring to polyvinylidene difluoride (PVDF) membranes, blots were incubated with 5% BSA (Gibco, US) in Trisbuffered saline containing 0.5% Tween 20 (TBST) for 60 min and incubated overnight at 4°C on a rocker with the following primary antibodies: anti-GSG2 antibody (1:1,000, Abcam, CA, USA), anti-p-H3 (S10) antibody (1:300, Abcam, CA, USA), anti-p-H3 (S28) antibody (1:2,000, Abcam, CA, USA), anti-p-H3 (T3) antibody (1:500, Abcam, CA, USA), and anti-GAPDH antibody (1:3,000, Bioworld Technology, Inc, Minnesota, USA). Following washing three times with TBST for 5 min, and then PVDF membranes were incubated with horseradish peroxidase (HRP) conjugated goat anti-rabbit IgG polyclonal secondary antibody (1:3,000, Beyotime Biotechnology, Shanghai, China) at room temperature for 1 h. Using Amersham’s ECL plusTM western blotting system kit for color developing. Signals were detected with enhanced chemiluminescence, using GAPDH as the internal standard (Kodak). MTT Assay MTT assay was carried out to investigate the cell proliferation ability. PANC-1 and

SW1990 cells were first seeded to 96-well plates and incubated at 37°C for 1 day, 2 days, 3 days, 4 days and 5 days. Added MTT solution (Sigma, MO, USA) per well incubated for another 4 h. Afterwards, discarded the culture medium with MTT and added DMSO, shaking for 10 min at room temperature. The absorbance was measured at 490 nm with a standard micro plate reader and the cell viability was calculated. Colony formation assay ShRNA lentivirus infected PANC-1 and SW1990 cells were cultured for 5 days, cells in logarithmic phase were digested by trypsin, resuspended, counted and then seeded into a six-well plate, and cultured for 8 days to form colonies. Then the cells were stained with Giemsa for 20 min, and photographed with a digital camera to statistics the number of cell clones. Detection of cell cycle by Fluorescence activated Cells Sorting (FACS) When the cell growth reached about 70% coverage, the cells were digested with trypsin and centrifuged at 1,200 rmp for 5 min, washed by PBS precooled at 4°C once and then fixed with 70% ethanol, precooled at 4°C, for 1 h. After centrifuging at 1500 rmp for 5 min, the fixed solution was discarded and PBS washed cells once. Cell staining solution was proportioned at the ratio of 40 × PI (2 mg/ml):100 × RNase (10 mg/ml):1 × PBS = 25:10:1,000. 1 ml of staining solution was taken to suspend the cells. Fluorescence activated Cells Sorting (Millipore, Darmstadt, Germany, USA) was used to detect cell cycle distribution. 300 mesh screen was filtered in the flow tube. The cell passage rate was 200-350 cells/s. Wound healing assay The PANC-1 and SW1990 cells were plated in a 96-well plate for culturing. When the cell density reached 90%, the cells were replaced with the conditioned medium for scratching test and the slide was pushed up to form a scratch. Added 0.5% FBS, and took a photo. Then the cells were incubated at 37°C, 5% CO2 incubator, for 24 h and 48 h, taken a fluorescence microscope to take photos. The cell migration rate of each group was calculated based on the pictures after the scratches. Transwell migration experiment

PANC-1 and SW1990 cells’ suspension were prepared, and adjusted the cell density to 5 × 105 cells/mL. 50,000 cells were inoculated in transwell, and added cell medium containing 20% FBS to the lower chamber of 24-well plate to examine the cell migration ability. After 48 h later, removed the transwell chamber, washed with PBS, and fixed in methanol for 30 min. Stained by 0.1% crystal violet for 20 min, the cells were observed under the microscope, counted and pictured. Detection of cell apoptosis by Fluorescence activated Cells Sorting (FACS) ShRNA lentivirus infected PANC-1 and SW1990 cells were plated in a six-well plate (2 mL/well). After 6 days of culture, cell suspensions were centrifuged at 1,300 rmp for 5 min, then the supernatants were discarded, and the cells were washed with 4°C pre-cooled D-Hanks (pH = 7.2-7.4). 10 µL Annexin V-APC were added to stain for 15 min in the dark. The percentage of cells phases was measured using FACScan (Millipore, Darmstadt, Germany, USA) to assess the apoptotic rate, and results were analyzed. Human apoptosis antibody array To detect the changes of related genes expression in the human apoptosis signaling pathway after RNA interference with the GSG2 gene in PANC-1 cells, the PANC-1 cells were lysed by lysis buffer. The membranes were blocked by 2 mL 1 × Blocking Buffer at room temperature for 30 min, incubated with cell lysis at 4°C overnight, added 1 × Biotin-conjugated Anti-Cytokines and incubated at 4°C overnight, then incubated in 1 × Streptavidin-HRP at room temperature for 2 h, and finally, the signals of membranes were detected using a chemiluminescence imaging system. Tumor-bearing animal model Four-week-old female BALB/c nude mice were purchased from Shanghai SLAC Laboratory Animal Co. LTD. All mice were housed under standard housing conditions, and divided into two groups randomly, each group containing 5 mice. All mice were. All the animal experimental protocols were approved by the Animal Care Committee of Zhongshan Hospital. The logarithmic growth phase SW1990 cells infected with shRNA lentivirus were digested by trypsin, and then resuspended, mice were given a suspension of SW1990

cells (4 × 106 cells/mL) 200 uL subcutaneously. Total 10 mice were divided into two groups (shGSG2 and shCtrl group). All mice were raised for 39 days after subcutaneous injection. On the last day of breeding, the mice were anesthetized by intraperitoneal injection of 0.7% pentobarbital sodium at a dose of 10 uL/g and the nude mice were placed in a flat-box in a small animal multi-spectral living imaging system for imaging (Lumina LT, Perkin Elmer, Massachusetts, USA), and the excitation wavelength and reception wavelength were 468 nm and 562 nm. Then all mice were sacrificed by cervical vertebrae. The tumors were removed, arranged on the whiteboard, measured its volume with calipers, weighted, and saved in -80°C with liquid nitrogen. Ki67 staining The sections of mice tumor tissues were fixed with 4% paraformaldehyde at room temperature, added 0.3% TritonX-100, then incubated with Ki67 antibody (1:200, Abcam, CA, USA) at 4°C overnight. Washed by PBS, sections were incubated with horseradish peroxidase (HRP) conjugated goat anti-rabbit IgG secondary antibody (1:400, Abcam, CA, USA) for 2 h at room temperature in the dark, then counterstained with DAPI, mounted with glycerin, and observed by microscope. Statistical analysis QRT-PCR was analyzed by 2−∆∆CT method. All the cell experiments in this study were independently repeated for 3 times. For statistical analysis, all data obtained in this study were analyzed using Prism 6 for Windows software (GraphPad Software, San Diego, USA). Statistically significant differences between studied groups were evaluated using the unpaired Student’s t-test and Fisher’s exact test. P values less than 0.05 were considered statistically significant.

Results Expression of GSG2 in pancreatic cancer tissues To investigate the role of GSG2 in the development and progression of pancreatic cancer, the expression levels of GSG2 were detected in clinical pancreatic cancer tissues and compared with para-carcinoma tissues by immunohistochemical staining.

As shown in Figure 1A, the results indicated a cytoplasmic localization of GSG2 and demonstrated that the expression of GSG2 in pancreatic cancer tissues was significantly up-regulated compared with that in para-carcinoma tissues, which was also shown in Table 1. Moreover, the relationship between expression of GSG2 in pancreatic cancer and clinicopathological characteristics was statistically analyzed, and the results indicated there was a significant correlation between the expression of GSG2 and the pathological grade (P < 0.05) (Table 2). The Spearman rank correlation analysis further suggested that GSG2 expression was positively correlated with the pathological grade of pancreatic cancer (P < 0.05) (Table 3). Finally, Kaplan-Meier survival analysis was performed and its results demonstrated that the high expression of GSG2 was significantly associated with low overall survival of pancreatic cancer (Figure 1B). These results demonstrated that the expression of GSG2 was significantly associated with the development and progression of pancreatic cancer and might serve as a marker for pancreatic cancer prognosis. Knockdown efficiency of GSG2 by shRNA lentivirus infection in pancreatic cancer cells ShRNA targeting GSG2 was cloned into a lentiviral vector with green fluorescent protein (GFP). Then, GSG2-shRNA lentivirus or shCtrl lentivirus (as negative control) were transfected into PANC-1 and SW1990 cell lines. As shown in Figure 2A, the fluorescence intensity was observed under microscope 72 h after transfection, which revealed that the cell infection efficiency was > 80%. Then we detected the expression of GSG2 in PANC-1 and SW1990 cell lines by qRT-PCR. The GSG2 mRNA levels in pancreatic cancer cell lines appear as Figure 2B, which indicated that the level of GSG2 mRNA was reduced by 66.1% in PANC-1 cell and 64% in SW1990 cell compared to shCtrl group, respectively (P < 0.05). Western blot assay demonstrated that GSG2 protein expression was down-regulated post-infection by shGSG2 as compared to shCtrl group (Figure 2C). Combined with the above results, GSG2 knockdown cell model was successfully constructed and used for subsequent experiments. Knockdown of GSG2 inhibited the proliferation and colony formation capacity

of pancreatic cancer cells and blocked the cell cycle MTT assay was applied for deciphering cell growth curve to explore the effect of GSG2 on cell growth. As illustrated in Figure 3A, cell proliferation was significantly inhibited in shGSG2 group relative to the negative control cells as assessed (P < 0.05). Furthermore, the number of cell colony was markedly decreased after shGSG2 transfection compared to shCtrl transfection in PANC-1 cell and SW1990 cell (Figure 3B) (P < 0.05). Besides, downregulation of GSG2 decreased the percentage of PANC-1 and SW1990 cells in G1 phase and increased the percentage of cells in G2 phase (Figure 3C) (P < 0.05). The western blot assay results indicated that GSG2 knockdown inhibited the phosphorylation at Threonine 3 of H3 (pH3T3) and there was no changes of H3S10 and H3S28 phosphorylation, suggesting that GSG2 might block the cell cycle of pancreatic cancer cells in G2 phase by inhibiting the phosphorylation of H3T3 (Figure 3D). Knockdown of GSG2 suppressed pancreatic cancer cells migration At the same time, for the sake of further investigating GSG2 knockdown influence on pancreatic cancer metastasis, wound healing assay and transwell assays were applied in this study. The migration rate of PANC-1 was decreased by 11% in shGSG2 group compared with shCtrl group after 48 h (P < 0.05), and rates of SW1990 were decreased by 9% and 19% in shGSG2 group compared with shCtrl group after 24 and 48 h, respectively (Figure 4A) (P < 0.05). As we expected, the similar results appeared in transwell assay that the migratory cells per field of PANC-1 cells and SW1990 cells were distinctly decreased in shGSG2 group compared with shCtrl group (Figure 4B) (P < 0.05). Both of these two assay indicated that knockdown of GSG2 significantly suppressed pancreatic cancer cell migration in vitro. Mechanism study of GSG2 affecting on cell apoptosis in PANC-1 cells The results of Annexin V-APC staining, applied by FACS, were shown in Figure 5A, and it was obviously that, compared with the negative control, the percentage of cell apoptosis in shGSG2 group doubled in PANC-1 cells and increased 5-fold in SW1990 cells respectively (P < 0.05), suggesting that downregulation of GSG2 expression levels promoted PANC-1 and SW1990 cell apoptosis. In addition, the regulatory

mechanism of GSG2 in the tumor development of pancreatic cancer was preliminarily investigated in PANC-1 cells. In PANC-1 cells after interference with the GSG2, the proteins in the Human Apoptosis Antibody Array were detected, of which IGFBP-6 protein was significantly upregulated, and the expression levels of Bcl-2, Bcl-w, clAP-2, HSP60 and Livin proteins were significantly downregulated (Figure 5B-D) (P < 0.05), indicating that knockdown of GSG2 inhibited these proteins, so these would probably be the new targets of GSG2 for treating pancreatic cancer. The effects of GSG2 knockdown on tumor progression in vivo The xenograft models in nude mice were constructed to demonstrate GSG2 effects on pancreatic cancer after investigating the role of GSG2 in pancreatic cancer cell lines. SW1990 cells, successfully transfected with either shGSG2 or shCtrl lentivirus, were subcutaneously injected into nude mice. After 39 days, 15 mg/mL D-Luciferin and 0.7% pentobarbital sodium were injected intraperitoneally at an amount of 10 µL/g, respectively, and the fluorescence in mice was observed and photographed (Figure 6A). The levels of fluorescence expression in the bodies of anaesthetized mice measured and showed in Figure 6B: the levels in the shGSG2 group were obviously lower than those in the negative control group (P < 0.05), which suggested the tumor growth was significantly suppressed by shGSG2. Moreover, tumors volume from mice in shGSG2 group was smaller relative to that in shCtrl group (Figure 6C, D) (P < 0.05), and as expected, the tumors weight in shGSG2 group was lighter (Figure 6E) (P < 0.05), indicating that tumors grew more slowly in the shGSG2 group compared with shCtrl group. Also we proved that the expression levels of Ki67, a marker of cell proliferation, were downregulated by shGSG2 using immunohistochemical staining (Figure 6F). Taken together, these data indicated that downregulation of GSG2 by shGSG2 could have an inhibitory effect on pancreatic cancer development in vivo.

Discussion Previous studies indicated that there were some targets associated with the pancreatic cancer. Germline BRCA2 gene mutation was thought to be accounting for the highest

proportion of known causes of inherited pancreatic cancer (11). PALB2 played an important role in increasing the risk of pancreatic cancer (12). AKT1-PI3K-MTOR was investigated as therapeutic targets in cancer (13). Although there were many reports indicating that the treatment of pancreatic cancer has advanced a lot, targets with the potential to be used for pancreatic cancer therapy were still needed to be discovered. GSG2 was mainly located in the concentrated chromosomes during mitosis, centrosomes after nuclear envelope breakdown (NEBD), spindle microtubules during metaphase and midody during telophase. At present, the only known substrate of GSG2 was histone H3 (9). It has been reported that GSG2 kinase modulated mitosis in cells by directly phosphorylating histone H3 at Thr-3 because it was indicated to regulate the proliferation of cell growth and play a central role in modulating the growth of cancer cells (14). Based on these, GSG2 kinase inhibitors was regarded as anti-cancer drugs. Despite the efforts for finding new strategies to treat pancreatic cancer more effectively, rarely knew the relationship between GSG2 kinase and pancreatic cancer. In this study, we provided evidences for a novel link between GSG2 and pancreatic cancer. Immunohistochemical analysis revealed that GSG2 was significantly upregulated in pancreatic cancer and correlated with malignant grade and overall survival. Therefore, GSG2 knockdown cell models were constructed. Results of qRT-PCR and western blot showed that GSG2 expression in PANC-1 and SW1990 cells was successfully downregulated by shGSG2 compared with shCtrl. Further study showed that GSG2 knockdown inhibited proliferation and colony formation capacity of pancreatic cancer cells. And it was also demonstrated that downregulation of GSG2 blocked pancreatic cancer cells in G2 phase by decreasing the phosphorylation level of H3T3. Wound healing assay and transwell assays indicated that knockdown of GSG2 could inhibit pancreatic cancer cell migration. Moreover, we found that knockdown of GSG2 induced the apoptosis of PANC-1 and SW1990 cells. We also preliminarily investigated the regulatory mechanism of GSG2 in the tumor development of

pancreatic cancer. Downregulation of GSG2 remarkably promoted the expression of IGFBP-6, and inhibited Bcl-2, Bcl-w, cIAP-2, HSP60 and Livin expressions in pancreatic cancer cells as compared with shCtrl group. There was a study demonstrated that miR-506 inhibited pancreatic cancer cell growth in vitro and in vivo and enhanced cell apoptosis and chemosensitivity through down-regulating Bcl-2

pathway (15). It was verified that down-regulating Bcl-w by histone deacetylase inhibitor (HDACI) could inhibit the progress and development of pancreatic tumor (16, 17). G.L.V.de Oliveira et al. found cIAP-2 was an inhibitor of apoptosis protein that promoted the growth of tumor (18). Heat shock protein 90 (HSP-90) expression was not associated with any of the clinicopathological parameters examined, hut the staining intensity of HSP90 was significantly associated with gastric tumor size (19). Nevertheless, heat shock protein 60 (HSP60) knockdown promoted pancreatic cancer cells proliferation, induced cell apoptosis and cycle arrest (20). Livin had been reported it was involved in tumor development through the inhibition of caspases (21). Moreover, Insulin-like Growth Factor Binding Protein (IGFBPs) was one mechanism that could modify IGF activities through high affinity interactions (22). These results showed that GSG2 played a critical role in the development and progression of pancreatic cancer. Frankly speaking, this study was still limited by the small number of clinical specimens and lacking of mechanism research. Further research was needed to discover the prognostic significance of GSG2 in pancreatic cancer and the role of GSG2 in different tumor cells. In conclusion, this was the first research that linked GSG2 to pancreatic cancer progression. Our study suggested that GSG2 was over-expressed in pancreatic cancer and regulated pancreatic cancer cell proliferation and migration. These results will deepen the understanding about causal roles of GSG2 in pancreatic cancer progression and provide a potential therapeutic target.

Acknowledgement This study was supported by National Natural Science Foundation of China (81702304).

Author contribution Xu Han and Yun Ren conceived and coordinated the study, designed, performed and analyzed the experiments, wrote the paper. Zhiyao Wang and Qingwu Liao carried out the data collection, data analysis, and revised the paper. Wei Chen supervised the study. All authors reviewed the results and approved the final version of the manuscript.

Competing interests The authors declare no conflicting of interests Reference 1. Bhattacharyya S., Pradhan K., Campbell N., Mazdo J., Vasantkumar A., Maqbool S., et al. Altered hydroxymethylation is seen at regulatory regions in pancreatic cancer and regulates oncogenic pathways. Genome Res. 2017, 27(11): 1830-1842. 2. Tyagi N., Marimuthu S., Bhardwaj A., Deshmukh S. K., Srivastava S. K., Singh A. P., et al. p-21 activated kinase 4 (PAK4) maintains stem cell-like phenotypes in pancreatic cancer cells through activation of STAT3 signaling. Cancer Lett. 2016, 370(2): 260-7. 3. Scaiewicz V., Sorin V., Fellig Y., Birman T., Mizrahi A., Galula J., et al. Use of H19 Gene Regulatory Sequences in DNA-Based Therapy for Pancreatic Cancer. J Oncol. 2010, 2010: 1-10. 4. Ying H., Dey P., Yao W., Kimmelman A. C., Draetta G. F., Maitra A., et al. Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev. 2016, 30(4): 355-85. 5. Zhang X., Shi S., Zhang B., Ni Q., Yu X. and Xu J. Circulating biomarkers for early diagnosis of pancreatic cancer: facts and hopes. Am J Cancer Res. 2018, 8(3): 332-353. 6. Kim J. E., Lee S. Y., Jang M., Choi H. K., Kim J. H., Chen H., et al. Coumestrol Epigenetically Suppresses Cancer Cell Proliferation: Coumestrol Is a Natural Haspin Kinase Inhibitor. Int J Mol Sci. 2017, 18(10). 7. Han L., Wang P., Sun Y., Liu S. and Dai J. Anti-Melanoma Activities of Haspin

Inhibitor CHR-6494 Deployed as a Single Agent or in a Synergistic Combination with MEK Inhibitor. J Cancer. 2017, 8(15): 2933-2943. 8. Cuny G. D., Ulyanova N. P., Patnaik D., Liu J. F., Lin X., Auerbach K., et al. Structure-activity relationship study of beta-carboline derivatives as haspin kinase inhibitors. Bioorg Med Chem Lett. 2012, 22(5): 2015-9. 9. Amoussou N. G., Bigot A., Roussakis C. and Robert J. H. Haspin: a promising target for the design of inhibitors as potent anticancer drugs. Drug Discov Today. 2018, 23(2): 409-415. 10. Huertas D., Soler M., Moreto J., Villanueva A., Martinez A., Vidal A., et al. Antitumor activity of a small-molecule inhibitor of the histone kinase Haspin. Oncogene. 2012, 31(11): 1408-18. 11. Li M., Mou Y., Hou S., Cao D. and Li A. Response of germline BRCA2-mutated advanced pancreatic acinar cell carcinoma to olaparib: A case report. Medicine (Baltimore). 2018, 97(45): e13113. 12. Takeuchi S., Doi M., Ikari N., Yamamoto M. and Furukawa T. Mutations in BRCA1, BRCA2, and PALB2, and a panel of 50 cancer-associated genes in pancreatic ductal adenocarcinoma. Sci Rep. 2018, 8(1): 8105. 13. Mohammed S., Van Buren G., 2nd and Fisher W. E. Pancreatic cancer: advances in treatment. World J Gastroenterol. 2014, 20(28): 9354-60. 14. Opoku-Temeng C., Dayal N., Aflaki Sooreshjani M. and Sintim H. O. 3H-pyrazolo[4,3-f]quinoline haspin kinase inhibitors and anticancer properties. Bioorg Chem. 2018, 78: 418-426. 15. Li J., Wu H., Li W., Yin L., Guo S., Xu X., et al. Downregulated miR-506 expression facilitates pancreatic cancer progression and chemoresistance via SPHK1/Akt/NF-kappaB signaling. Oncogene. 2016, 35(42): 5501-5514. 16. Koutsounas I., Giaginis C., Patsouris E. and Theocharis S. Current evidence for histone deacetylase inhibitors in pancreatic cancer. World J Gastroenterol. 2013, 19(6): 813-28. 17. Damaskos C., Garmpis N., Karatzas T., Nikolidakis L., Kostakis I. D., Garmpi A., et al. Histone Deacetylase (HDAC) Inhibitors: Current Evidence for Therapeutic

Activities in Pancreatic Cancer. Anticancer Res. 2015, 35(6): 3129-35. 18. de Oliveira G. L., Ferreira A. F., Gasparotto E. P., Kashima S., Covas D. T., Guerreiro C. T., et al. Defective expression of apoptosis-related molecules in multiple sclerosis patients is normalized early after autologous haematopoietic stem cell transplantation. Clin Exp Immunol. 2017, 187(3): 383-398. 19. Giaginis C., Daskalopoulou S. S., Vgenopoulou S., Sfiniadakis I., Kouraklis G. and Theocharis S. E. Heat Shock Protein-27, -60 and -90 expression in gastric cancer: association

with

clinicopathological

variables

and

patient

survival.

BMC

Gastroenterol. 2009, 9: 14. 20. Zhou C., Sun H., Zheng C., Gao J., Fu Q., Hu N., et al. Oncogenic HSP60 regulates mitochondrial oxidative phosphorylation to support Erk1/2 activation during pancreatic cancer cell growth. Cell Death Dis. 2018, 9(2): 161. 21. Altieri B., Sbiera S., Della Casa S., Weigand I., Wild V., Steinhauer S., et al. Livin/BIRC7 expression as malignancy marker in adrenocortical tumors. Oncotarget. 2017, 8(6): 9323-9338. 22. James P. L., Stewart C. E. and Rotwein P. Insulin-like growth factor binding protein-5 modulates muscle differentiation through an insulin-like growth factor-dependent mechanism. J Cell Biol. 1996, 133(3): 683-93.

Figure legend Figure 1 Expression of GSG2 in pancreatic tissues. (A) Negative staining in adjacent normal tissues and representative immunohistological characteristics with high expression of GSG2 in pancreatic cancer tissues (200 ×). (B) The results of Kaplan-Meier survival analysis in clinical samples. GSG2, germ cell associated 2; tumor, pancreatic cancer tissues; normal, adjacent normal tissues. Figure 2 GSG2 knockdown using shRNA in PANC-1 and SW1990 cells. (A) Transfection efficiency was examined by fluorescence imaging in PANC-1 and SW1990 cells (200 ×). (B) GSG2 mRNA expression levels were determined by qRT-PCR in PANC-1 and SW1990 cells. (C) GSG2 protein expression levels were detected by western blot assay in PANC-1 and SW1990 cells. GSG2, germ cell associated 2; shRNA, short hairpin RNA; shGSG2, cells transfected with GSG2-targeting shRNA; shCtrl, cells transfected with control shRNA. (n ≥ 3). *** P< 0.001. Figure 3 GSG2 knockdown inhibited proliferation of PANC-1 and SW1990 cells and blocked the cell cycle. (A) The effect of GSG2 knockdown on the viability of PANC-1 and SW1990 cells was investigated by MTT assay. Comparing with the shCtrl group, the growth rates of PANC-1 cells and SW1990 cells were both slowed down in the shGSG2 group. (B) Photomicrographs of Giemsa-stained PANC-1 and SW1990 cells showed the number of cell clone in 6-well plates at 8 days post-seeded into 6-well plate. GSG2 knockdown inhibited human pancreatic cancer cells colony formation. (C) The cell cycle was analyzed by FACS. Downregulation of GSG2 expression decreased the percentage of G1 phase and increased the percentage of G2 phase. (D) Western blot detected the phosphorylation level of H3T3 after knockdown of GSG2. GSG2, germ cell associated 2; shRNA, short hairpin RNA; shGSG2, cells transfected with GSG2-targeting shRNA; shCtrl, cells transfected with control shRNA. (n ≥ 3). **P < 0.01, ***P < 0.001. Figure 4 GSG2 knockdown inhibited metastasis of pancreatic cancer cells. (A) Microscope photo of PANC-1 and SW1990 cells presented the migration rate of PANC-1

and

SW1990

cells

after

shRNA or

shGSG2

transfection.

(B)

Photomicrographs of Giemsa-stained PANC-1 and SW1990 cells, and the number of migratory PANC-1 and SW1990 cells per field was counted. GSG2, germ cell associated 2; shRNA, short hairpin RNA; shGSG2, cells transfected with GSG2-targeting shRNA; shCtrl, cells transfected with control shRNA. (n ≥ 3). ***P < 0.001. Figure 5 Mechanism study of GSG2 affecting on cell apoptosis in PANC-1 cells. (A) The percentage of apoptosis and representative histograms of the FACS results were detected by FACS in shCtrl and shGSG2 groups. (B) The results of human apoptosis antibody array in PANC-1 cells after Lv-shGSG2 infection. Antibody spots exhibiting signal markedly enhancement were indicated in red boxes while the significantly attenuated signal were indicated in green boxes. (C) The positional distribution of 43 apoptosis markers in the human apoptosis antibody array. (D) Plot histogram of GSG2-related signaling molecules in PANC-1 cells by SignaLink 2.0 analysis. GSG2, germ cell associated 2; shRNA, short hairpin RNA; shGSG2, cells transfected with GSG2-targeting shRNA; shCtrl, cells transfected with control shRNA. (n ≥ 3). *P < 0.05, **P < 0.01. Figure 6 The effects of GSG2 knockdown on tumor progression in vivo. (A) Fluorescence images of mice models in shCtrl and shGSG2. (B) The levels of fluorescent expression in shGSG2 group were significantly decreased than that in negative control group. (C) The representative images of tumors in subcutaneous xenograft mice models. (D-E) Changes of the tumor volume and weight. (F) Changes of Ki67 expression levels in shCtrl and shGSG2 groups were investigated by immunohistochemical staining (100 × and 200 ×). GSG2, germ cell associated 2; shRNA, short hairpin RNA; shGSG2, cells transfected with GSG2-targeting shRNA; shCtrl, cells transfected with control shRNA. (n ≥ 3). *P < 0.05.

Table 1. Expression patterns in pancreatic cancer tissues and para-carcinoma tissues revealed in immunohistochemistry analysis. *** P < 0.001 GSG2 expression Low High

Tumor tissue Cases Percentage 26 29.9% 61 70.1%

Para-carcinoma tissue Cases Percentage 38 100% 0 /

P value 0.000***

Table 2. Relationship between GSG2 expression and tumor characteristics in patients with pancreatic cancer. ** P < 0.01 Features

No. of patients

All patients Age (years) ≤ 59 > 59 Gender Male Female Grade I/II III/IV Tumor size ≤ 4cm > 4cm Lymph node NO. ≤8 >8 Lymph node NO.(positive) 0 ≥1 T stage T1 T2 T3 N stage N0 N1 AJCC stage 1 2 4

87

GSG2 expression low high 26 61

P value 1.000

27 30

14 16

13 14

55 32

18 8

37 24

0.450

0.001** 58 29

24 2

34 27 0.930

49 37

15 11

34 26 0.490

6 36

4 18

2 18 0.960

44 36

12 10

32 26

3 65 18

1 19 6

2 46 12

0.801

0.962 46 36

13 10

33 26 0.726

34 49 2

10 14 0

24 35 2

Table 3. Relationship between GSG2 expression and tumor characteristics in patients with pancreatic cancer. ** P < 0.01

Grade

Pearson related

GSG2 0.355

significance（two-tailed）

0.001**

N

87