SAMHD1, positively regulated by KLF4, suppresses the proliferation of gastric cancer cells through MAPK p38 signaling pathway

ABSTRACT SAMHD1 was reported to be related with the development of tumors, while its function in gastric cancer (GC) has not been elucidated yet. Here, we investigated the role and mechanism of SAMHD1 in regulating the proliferation of GC, as well as the mechanism of its expression regulation. Our results revealed that SAMHD1 was downregulated in GC tissues and cell lines, which was correlated with tumor size, depth of invasion and TNM stage. Overexpression of SAMHD1 inhibited the proliferation, clone formation, DNA synthesis and cell cycle progression, while knockdown of SAMHD1 promoted the proliferation of GC cells in vitro and vivo. Meanwhile, SAMHD1 inhibited the activation of MAPK p38 signaling pathway. Moreover, SB203580, as a MAPK p38 inhibitor, could reverse the proliferation and activation of MAPK p38 signaling pathway caused by knockdown of SAMHD1 in GC cells. Additionally, transcription factor Krüppel-like factor 4 (KLF4) bound to the core promoter of SAMHD1, increasing its transcriptional expression in GC cells. In conclusion, SAMHD1 suppressed the proliferation of GC through negatively regulating the activation of MAPK p38 signaling pathway and was upregulated by KLF4 in GC cells.


Introduction
Gastric cancer (GC) is the fifth malignant tumor worldwide, with more than 1 million new cases in 2020 [1]. Since most of the patients with GC are in advanced stage after diagnosis, the mortality rate is high, with an estimated 769,000 deaths in 2020, making it the fourth most common cause of cancer-related deaths [1]. Gastrectomy and D2 lymph node dissection combined with neoadjuvant or perioperative chemotherapy may benefit patients with GC at locally advanced stage [2][3][4]. However, the 5-year survival rate is less than satisfactory [5], and targeted therapy is one of the most promising ways to break through the treatment bottleneck. Currently, Trastuzumab, as a human epidermal growth factor receptor 2 (HER2) inhibitor, is the most common targeted drug in GC, which combining with first line chemotherapy could improve the overall survival for the patients with HER2 + [2]. However, the percentage of patients with HER2 + was less than 20%, leading to limited application of Trastuzumab in treatment of GC [6]. Thus, discovering novel molecules regulating development of GC may provide more options for future targeted therapies.
Human sterile alpha motif and HD-domaincontaining protein 1 (SAMHD1) is a triphosphohydrolase that degrades deoxynucleoside triphosphates to the constituent deoxynucleoside and inorganic triphosphate, which blocks the replication of retrovirus [7] and DNA virus [8]. Additionally, SAMHD1 is also associated with a congenital inflammatory disease of the nervous system -Aicardi-Goutières syndrome [9][10][11]. Recent studies have shown that SAMHD1 involves in regulating the development of lung cancer [12,13] and cutaneous T-cell lymphoma [14]. SAMHD1 is downregulated in lung cancer [12] and cutaneous T-cell lymphoma [15,16]. Furthermore, overexpression of exogenous SAMHD1 inhibits the proliferation of cell lines derived from lung cancer [12] and cutaneous T-cell lymphoma [14]. However, whether SAMHD1 functions to suppress GC remains unclear. The immunohistochemistry analysis showed that SAMHD1 was downregulated in GC tissues, which was associated with tumor size, depth of invasion and TNM stage, suggesting that low expression of SAMHD1 might be an important cause of GC proliferation.
In this work, we elucidated the mechanism of SAMHD1 inhibiting GC proliferation and its transcriptional regulation.

Patients and specimens
Paired GC and adjacent normal tissues were from 58 patients diagnosed with GC in The First Affiliated Hospital of Anhui Medical University between January 2019 and December 2020. None of the patients received preoperative adjuvant chemoradiotherapy. This study was conducted in accordance with the Declaration of Helsinki (2013) [17].

Cell lines
MGC-803 and HGC-27 cells were generously gifted by Dr. Ping Wu (Anhui Medical University). AGS, MKN-45 and GES-1 cells were purchased from the Cell Bank of Type Culture Collection of Chinese Academy of Sciences. All cell lines were authenticated using STR profiling (Biowing, China).

Plasmids construction and sources
Genomic DNA was extracted using the Cellular Genomic DNA Extraction Kit (Tiangen, China). And the promoters of SAMHD1 with different lengths were amplificated using the primers in Table S1. The purified DNA fragments and pGL3 vector were doubly digested with the Kpn I and Hind III restriction enzymes (Takara, Japan). Finally, the targeted fragments were inserted into the pGL3 vector using the DNA Insert Ligation (Thermofisher, USA). The pGL3-285 mutant was constructed using Rapid Site-specific Mutation kit (Tiangen) according to the manufacturer's instruction.

RNA extraction and quantitative real-time PCR (qRT-PCR)
Total intracellular RNA was extracted using AxyPrep Total mRNA Mini kit (Corning, USA) and was reversely transcribed into cDNA using PrimeScript TM RT Master Mix (Takara) according to the manufacturer's instructions. The qRT-PCR was performed as previously described [18] and the primers were listed in Table S1.

Cell proliferation assay
Cell proliferation was detected as previously described [19]. Results were shown as a ratio relative to the control group, which was set as 1.

Clone formation assay
Cells were cultured in 6-well plates at a density of 200 cells/well. Following culture for 2 weeks, the cells were fixed with 4% paraformaldehyde for 15 min and then stained with crystal violet (Beyotime, China) for 15 min. Finally, the stained cells were photographed and counted.

Western blot
Western blot was performed as previously described [19]. The information of primary antibodies is listed in Table S2. The gray values of protein bands were calculated using Image J software (version 1.53).

Cell cycle analysis
Cell cycle analysis was performed as previously described [19]. The stained cells were detected by CytoFLEX flow cytometry (Beckman, USA) and CytExpert software (version 2.0, Beckman). The proportions of cells in G 1 /G 0 , S and G 2 /M phases were analyzed by Flow Jo software (version 7.6.1).

Ethynyldeoxyuridine (EdU) insert assay
EdU can replace thymidine and be inserted into the newly synthesized DNA during DNA replication. EdU acetylenyl and Azide Alexa Fluor 594 formed stable triazole rings, which could be detected at a 594-nm wavelength. In this study, we used BeyoClick TM EdU-594 kit (Beyotime) to detect DNA synthesis. Briefly, GC cells in the 12-well plates were cultured in medium containing 10 μM EdU for 2 h, and fixed with 4% paraformaldehyde for 15 min. After removing the supernatant, cells were treated with PBS with 0.3% Triton X-100 and then stained with the Click Additive Solution in 37°C away from the light for 30 min. Next, nucleus was labeled with Hoechst 33,342. Finally, the stained cells were analyzed by CytoFLEX flow cytometry (Beckman), and the percentages of EdU + cells were calculated using CytExpert software (version 2.0, Beckman).

Animal experiment
Female nude mice aged 4 weeks were obtained from the Shanghai SLAC Laboratory Animal Co., Ltd. and kept under specific pathogen-free conditions at 22°C with 55% humidity and a 12-h light/ 12-h dark cycle. HGC-27 cells (shNC or shRNA-1, 10 6 cells/mouse) were injected subcutaneously into the middle-upper inguinal regions of the nude mice. The body weight of mice and the volume of tumors were measured every 7 days. After 4 weeks, the mice were euthanized by cervical dislocation and the tumors were resected for further detection.

RNA-seq and bioinformatics analysis
R software (version 3.6.2) was used for bioinformatics analysis in this study. RNA-seq was performed in the AGS and HGC-27 cells with different expression of SAMHD1 by Beijing Biomarker Technologies Co., Ltd. For analysis of RNA-seq data, the "org.Hs.eg.db" and "enrichplot" packages were used for screening out the differentially expressed genes and KEGG analysis respectively, and the heatmap was drew using the "heatmap" package. The data of 473 GC tissues and 41 adjacent normal tissues was downloaded from TCGA, which was normalized using log 2 (FPKM+0.01). Meanwhile, the data of 111 GC tissues and 21 adjacent normal tissues was also downloaded from GSE54129 database, which was normalized using "limma" package. And the differentially expressed transcriptional factors were identified using Wilcoxon test: P< 0.05, log 2 (fold change)>1.

Dual-luciferase reporter assay
The effect of KLF4 on the promoter of SAMHD1 was assessed by dual-luciferase reporter assay (Promega) according to the manufacturer's instructions. Briefly, the cells were lysed with 1× PLB solution for 15 min. And 20 μl cell lysis was added into 100 μl LAR II reaction solution, and the firefly fluorescence was immediately detected by Luciferase activity detection instrument (GLOMAX MULTI, Promega). And then the reaction system was added with 100 μl Stop&Glo solution, and the renal fluorescence was also immediately detected. The ratio of firefly fluorescence vs renal fluorescence represented the transcriptional activity of pGL3 plasmids.

Immunochromatin precipitation (ChIP)
SimpleChIP® Plus Enzymatic Chromatin IP Kit (Magnetic Beads, CST, USA) was used for the ChIP assay according to the manufacturer's instructions. Briefly, cells were crosslinked with methanal and then the nuclear were extracted. Chromatin digestion was performed using ultrasonic cell disruptor. The anti-KLF4 antibodies were added into the DNA samples, which were rotated and incubated overnight. Next, 30 μl Magnetic Beads were used to bind the DNA-KLF4 complex, which was then purified through centrifugal column. The purified DNA was used for further PCR and the primers were listed in Table S1. Finally, the PCR products were separated in agarose gels and photographed under ultraviolet light.

Statistical analysis
The quantitative data obtained from three independent and repeated experiments are presented as the mean ± standard error, and the qualitative differences between two groups were analyzed by unpaired Student's t-test. The relationships between expression levels of SAMHD1 and clinicopathological features were determined by chisquare test, Fisher's exact test or Student's t-test. * P< 0.05 and ** P< 0.01 were considered to indicate a statistically significant difference.

SAMHD1 was downregulated in GC tissues and cell lines
SAMHD1 was down-regulated in GC tissues compared with adjacent normal tissues (Figure 1(a,  b)). Moreover, the expression levels of SAMHD1 protein were associated with the tumor size, depth of invasion (T stage) and TNM stage in patients with GC (Table 1). Meanwhile, the expression levels of SAMHD1 protein and mRNA in GC cells were also lower than that in GES-1 cells (Figure 1(c-e)).

Knockdown of SAMHD1 promoted the proliferation and clone formation of GC cells
To determine the effect of lower expression of SAMHD1 on the proliferation of GC, we established the HGC-27 and MGC-803 cells with stable knockdown of SAMHD1 (Fig. S1). As shown in Figure 2(a), the proliferation of HGC-27 cells in shRNA-1 and −2 groups was faster than that in shNC group. Additionally, compared with shNC group, the abilities of clone formation of HGC-27 cells were also higher in shRNA-1 and −2 groups (Figure 2(b, c)). Meanwhile, knockdown of SAMHD1 also promoted the proliferation and clone formation of MGC-803 cells (Figure 2(d-f)).

Overexpression of exogenous SAMHD1 suppressed the proliferation, clone formation, DNA synthesis and cell cycle progression of MKN-45 cells
To verify the role of SAMHD1 in inhibiting the development of GC, we overexpressed the SAMHD1 protein in MKN-45 cells (Figure 3(a)), which inhibited the cell proliferation (Figure 3(b)). Moreover, overexpression of SAMHD1 also suppressed the clone formation of MKN-45 cells (Figure 3(c, d)). Dysregulation of cell cycles is a key cause in the development of GC. However, whether SAMHD1 influenced the cell cycles of GC cells was unclear. As shown in Figure 3(e, f), compared with empty vector group, the MKN-45 cells with overexpression of SAMHD1 had lower percentages of S and G2/M phases and a higher percentage of G1/ G0 phase, suggesting that SAMHD1 suppressed the cell cycle progression of GC cells. DNA synthesis occurs in S phase and is important for tumor cell division. Overexpression of exogenous SAMHD1 also slowed the DNA synthesis of MKN-45 cells (Figure 3(g, h)). Collectively, overexpression of exogenous SAMHD1 suppressed the proliferation, clone formation, DNA synthesis and cell cycle progression of GC cells.

Knockdown of SAMHD1 accelerated the growth of subcutaneous transplantation tumors
To further explore the effect of SAMHD1 on the GC proliferation in vivo, we constructed the subcutaneous tumor model of nude mice (Figure 4(a)). As shown in Figure 4(b), the resected tumors in shRNA-1 group were larger than those in shNC group. There was no difference in the body weight of mice among shNC and shRNA-1 group (Figure 4  (c)). As shown in Figure 4(d), the tumors in shRNA-1 group grew faster than those in shNC group. Additionally, compared with shNC group, the tumors in shRNA-1 group maintained extremely lower expression of SAMHD1 (Figure 4(e, f)). To sum up, knockdown of SAMHD1 promoted the growth of subcutaneous transplantation tumors.

SAMHD1 inhibited the proliferation of GC cells via suppressing the MAPK p38 signaling pathway
Kodigepalli et al [20] found that SAMHD1 suppresses the proliferation of THP-1 cells through inhibiting the PI3K/AKT signaling pathway [20]. However, the mechanism of SAMHD1 inhibiting GC proliferation has not yet been elucidated. Thus, we performed the RNA-seq in HGC-27 and AGS cells with different expression of SAMHD1. After stable overexpression of SAMHD1 in AGS cells (FigureS2), 197 genes were upregulated and 596 genes were downregulated (FigureS3a), which were mainly enriched in PI3K/AKT and MAPK signaling pathway ( Figure 5(a)). Meanwhile, after stable knockdown of SAMHD1 in HGC-27 cells, 140 genes were upregulated and 294 genes were downregulated ( Figure S3b), which were mainly enriched in MAPK signaling pathway ( Figure 5(b)). Taken together, the MAPK was the common signaling pathway related to the expression levels of SAMHD1 in AGS and HGC-27 cells.
Extracellular regulated protein kinases (ERK) 1/ 2, c-Jun N-terminal kinase, p38 and ERK5 are the classical MAPK signaling pathways [21]. But, which one of MAPK signaling pathways was regulated by SAMHD1? Next, we discovered 55 common differentially expressed genes in HGC-27 and AGS cells (FigureS3c), which are presented in heatmap ( Figure 5(c)). Among them, MAP2K6 was the significant one related to expression of SAMHD1 ( Figure 5(c)). Additionally, overexpression of SAMHD1 decreased the expression of MAP2K6 mRNA and protein, while knockdown of SAMHD1 increased the expression of MAP2K6 ( Figure 5(d, e)). MAP2K6, also named as MKK6, is the critical one of upstream kinases in MAPK p38 signaling pathway [22,23]. As expected, SAMHD1 inhibited the activation of MAPK p38 signaling   pathway, and the opposite results happened after knockdown of SAMHD1( Figure 5(e)).
To determine the role of MAPK p38 signaling pathway in SAMHD1 inhibiting the GC proliferation, the HGC-27 cells with stable knockdown of SAMHD1 were treated with MAPK p38 inhibitor (SB203580). As shown in Figure 5(f), downregulation of SAMHD1 promoted the proliferation of HGC-27 cells, which was reversed by SB203580. Besides, SB203580 also suppressed the knockdown of SAMHD1 caused activation of MAPK p38 signaling pathway ( Figure 5(g)). Taken together, SAMHD1 inhibited the proliferation of GC via negatively regulating the MAPK p38 signaling pathway.

KLF4 bound to the core promoter of SAMHD1, increasing its transcriptional activity
To illuminate the mechanism of expression regulation of SAMHD1 in GC, we tried to find out transcriptional factors regulating expression of SAMHD1. Schematic diagrams of pGL3 plasmids containing of SAMHD1 promoter are presented in FigureS4a. Dual luciferase reporting assay showed that the core promoter region was within the first 285 bp before translation start site of SAMHD1 gene (FigureS4b). In TCGA and GSE54129 databases, there were 43 differentially expressed transcription factors in GC tissues (Figure 6(a)). Online analysis in UCSC and JASPAR websites showed that among those differentially expressed transcription factors, KLF4 was the only one that could bind to the core promoter of SAMHD1. KLF4 was downregulated in GC cells compared with GES-1 cells (Figure 6(b)). Additionally, the expression levels of SAMHD1 mRNA and protein were positively correlated with the expression levels of KLF4 protein in GC cells (Figure 6(c)). The binding sequence of KLF4 is presented in Figure 6(d), and the sequence potentially bound by KLF4 in the core promoter of SAMHD1 is presented in Figure 6(e). Moreover, we constructed the pGL3-285 mutant plasmid (Figure 6 (f)). As shown in Figure 6 (g, h), KLF4 increased the relative fluorescence activity of pGL3-285 wild type, but not pGL3-285 mutant in AGS and MKN-45 cells. ChIP assay proved the binding of KLF4 protein to the core promoter of SAMHD1  ( Figure 6(i)). To sum up, KLF4 increased the transcriptional activity of SAMHD1 via binding to its core promoter.

Overexpression of SAMHD1 inhibited the proliferation of GC cells caused by knockdown of KLF4
As shown in Figure 7(a-c), overexpression of KLF4 increased the expression levels of SAMHD1 mRNA and protein in AGS and MKN-45 cells. On the contrary, knockdown of KLF4 decreased the expression levels of SAMHD1 protein in AGS cells (Figure 7(d)). Additionally, overexpression of SAMHD1 rescued the knockdown of KLF4 induced decrease of SAMHD1 (Figure 7(d)). Moreover, knockdown of KLF4 promoted the proliferation of AGS cells, which was reversed partially by overexpression of SAMHD1 (Figure 7(e)).

Discussion
Studies show that SAMHD1 was downregulated in lung cancer tissues [12,13]. In this study, we firstly found that SAMHD1 expressed less in GC tissues, which was associated with tumor size, depth of invasion and TNM stage. SAMHD1 inhibited the proliferation of Hela (cervical cancer) [24], A549 (lung cancer) [12,13] and HuT78 (cutaneous T-cell lymphoma) [14] cells, consistent with our results in GC cells. Moreover, we found that knockdown of SAMHD1 promoted the growth of subcutaneous transplantation of GC cells. Gao et al [13] proved that overexpression of SAMHD1 inhibited the proliferation of A549 cells in vivo [13]. However, Kodigepalli et al [25] established the leukemia models of nude mice by injecting intravenously with THP-1 cells, and the result shows no significant difference in the tumor growth between the knockout of SAMHD1 group and control group [25]. The different role of SAMHD1 in solid and hematologic tumors may be due to the different modeling approaches and complex environments in vivo.
Karthik et al [20] found that SAMHD1 regulated the PI3K/AKT signaling pathway in THP-1 cells using the microarray analysis [20]. Besides, they also discovered other signaling pathways, which may be related to the expression of SAMHD1, including interferon, NF-κB, MAPK p38 and death receptor in THP-1 cells [20]. In this study, we chose the RNA-seq method because of its higher sensitivity. Besides, we not only selected the HGC-27 cells with stable knockdown of SAMHD1, but also AGS cells with stable overexpression of SAMHD1 to screen out the enriched signaling pathways. Further experiment determined that SAMHD1 inhibited the activation of MAPK p38 signaling pathway in GC cells. MAPK p38 is the critical one of signaling pathways regulating the proliferation of GC [26,27]. As expected, we discovered that SAMHD1 suppressed the GC proliferation via negatively regulating the MAPK p38 signaling pathway.
Methylation regulates expression of SAMHD1 in lung cancer [12], cutaneous T-cell lymphoma [15] and skin cutaneous melanoma [28]. The methylation levels of SAMHD1 promoter are negatively correlated with its expression in skin cutaneous melanoma [28]. DNA methylation inhibitor increases the expression of SAMHD1 mRNA and protein in lung cancer and cutaneous T-cell lymphoma [12,15]. However, based on the online analysis of the EWAS database (https://ngdc.cncb. ac.cn/ewas/datahub/index), the methylation levels of SAMHD1 promoter are positively correlated with its expression in GC, thus methylation can't explain the phenomenon about its less expression. Valverde et al [29] found that interferon regulatory factor 1 increases the expression of SAMHD1 through binding to its promoter [29]. In this study, we found that KLF4, as the one of differentially expressed genes in GC, upregulated the expression of SAMHD1. Additionally, KLF4 was downregulated in GC cells and knockdown of KLF4 decreased the expression of SAMHD1, suggesting that KLF4 at a low expression level was insufficient to positively regulate the transcription of SAMHD1 gene, leading to the decline of its expression in GC.
KLF4, as one member of KLF family, owns zinc fingers binding to specific DNA sequence [30]. KLF4 expresses widely in epithelial cells from the skin and gastrointestinal tract mucosa, but was downregulated in GC [31] and colorectal cancer [32]. It is well known that KLF4 inhibits the development of GC [33,34] and colorectal cancer [32]. Studies have shown that KLF4 suppresses the development of GC through regulating the expression of oncogenes or tumor suppressor genes. In detail, KLF4 upregulates the expression of caudal type homeobox 2 [35] and ghrelin [36], and downregulates the expression of C-X-C Motif Chemokine Ligand 8 [34], miR-106a [37,38], inhibitor of apoptosis-stimulating protein of p53 [39] and PODXL [30] in GC. In this study, SAMHD1 was another downstream gene upregulated by KLF4.
The chronic infection of Helicobacter pylori (HP) leads to gastric atrophic gastritis and intestinal metaplasia, which is closely related to the incidence of gastric antrum cancer. Cytotoxin-associated proteins A (CagA) is one of typical pathogenic factors of HP, and involves in the regulation of GC cell proliferation and metastasis. Zhao et al found that the expression levels of KLF4 in HP infected GC tissues are lower than that in GC tissues without HP infection [40]. HP infection suppresses the expression of KLF4 through increasing the methylation of its promoter [40]. Besides, CagA also inhibits the expression of KLF4 via upregulating its upstream miR-155 [41]. However, whether CagA regulates the expression of SAMHD1 directly or indirectly remains unknown. In chronic infection, virus evolve a mechanism to antagonize the antiviral ability of SAMHD1 in host. For example, the Vpx or Vpr of HIV recruits SAMHD1 protein to Cullin ring-finger ubiquitin ligase 4 E3 ubiquitin ligase complex, and then the ubiquitinated SAMHD1 protein is degraded by proteasome pathway [42,43]. Epstein-Barr Virus-Encoded Protein Kinase 4 phosphorylates SAMHD1, resulting in loss of its ability to block the replication of EBV [44]. Thus, it is unknown whether SAMHD1 involves in chronic infection of HP, which will be identified in the follow-up study.
In conclusion, our study revealed that SAMHD1 inhibited proliferation of GC cells via negatively regulating the MAPK p38 signaling pathway and KLF4 promoted the transcription of SAMHD1 in GC cells.

Ethical statement
The study was approved by the Ethics Committee of the First Affiliated Hospital of Anhui Medical University. The protocol for animal experiment was approved by Animal Use and Care Committee of Anhui Medical University.

Informed consent
All patients signed the informed consent for this study.