EXO1 Plays a Carcinogenic Role in Hepatocellular Carcinoma and is related to the regulation of FOXP3

Exonuclease 1 (EXO1), a member of the RAD2 nuclease family, was first described as possessing 5' to 3' nuclease activity and 5' structure-specific endonuclease activity. Here, we show that EXO1 is significantly upregulated in HCC tumor tissues and that high EXO1 expression is significantly correlated with liver cirrhosis. We further demonstrate that EXO1 knockdown decreases proliferation and colony forming abilities of HCC cells in vitro and tumorigenicity in vivo, as well as decreases migration and invasive capabilities of HCC cells. Alternatively, EXO1 overexpression significantly increases the proliferation, colony forming ability, and migration and invasive capabilities of HCC cells in vitro. Additionally, we truncated a region upstream of the transcription start site (TSS) of EXO1 and used the region with the strongest transcriptional activity to predict that the transcription factor FOXP3 can bind to the EXO1 promoter. Bioinformatics analysis found that FOXP3 was positively correlated with EXO1 and luciferase reporter assays and RT-PCR confirmed that FOXP3 could enhance the transcriptional activity of EXO1. CCK-8 assays showed that depletion of FOXP3 further reduces cell proliferation ability after knocking down of EXO1 in vitro. Taken together, our findings indicate that EXO1 acts as an oncogene in HCC and its expression level is related to FOXP3 activity.


Introduction
Hepatocellular carcinoma (HCC) is the fifth most common cancer and the third leading cause of cancer-related deaths worldwide [1]. Most patients are initially diagnosed at an advanced stage in the disease and thus, the curative effect of hepatectomy is often unsatisfactory. The 5-year recurrence rate of HCC after hepatectomy exceeds 80%, and the 5-year survival rate is only 30%-70% [2,3]. Therefore, indepth study of the molecular mechanisms involved in the development of HCC is of great significance to develop effective therapeutic drugs and prevent the recurrence of HCC after hepatectomy.
Exonuclease 1 (EXO1) is a member of the Rad2/ XPG family of nucleases and was originally identified in Schizosaccharomyces pombe [4]. EXO1 possess 5' to 3' exonuclease activity and structure-specific endonuclease activity and plays an extremely important role in biological processes such as DNA replication, DNA mismatch repair (MMR), DNA double-strand break repair (DSB) and telomere maintenance [5][6][7][8][9]. Deletion of the EXO1 gene causes genomic instability and leads to impaired DNA damage repair and meiosis defects [7,[10][11][12][13]. However, studies have shown that abnormally high expression of EXO1 is related to the occurrence, development and prognosis of various malignant tumors. Kretschmer et al. reported for the first time that EXO1 is significantly elevated in breast ductal carcinoma and invasive ductal carcinoma [14]. A recent study by de Sousa et al. has shown that EXO1 expression is significantly elevated in glioma, and high EXO1 expression is an independent risk factor for poor prognosis of patients [15]. Dai et al. reported that EXO1 is upregulated in HCC specimens and Ivyspring International Publisher EXO1 overexpression may also be associated with poor prognosis in HCC patients [16]. In addition, several single nucleotide polymorphisms (SNPs) in the EXO1 gene are associated with tumor susceptibility in a variety of tumors including liver, gastric, ovarian, cervical, breast, and colorectal cancers as well as head and neck squamous cell carcinoma [17][18][19][20][21][22][23][24][25]. Among them, rs1047840 (K589E), a common SNP genotype, increases the risk for developing non-viral liver cancer, colorectal cancer and lung cancer [17,26,27]. However, the role of EXO1 in the invasion and metastasis of HCC and the upstream transcriptional regulation of EXO1 is still poorly understood.
Forkhead box P3 (FOXP3) is a member of the forkhead/winged-helix family of transcription factors, which was first identified in T-regulatory (Treg) cells as an essential protein for regulating immune system development and function [28]. FOXP3 is primarily known for its function in regulating the CD4 + CD25 + regulatory cells and for defining their immunoregulatory phenotype [29]. Like other transcription factors, FOXP3 can bind to numerous enzymes and microRNAs to up-or downregulate a large number of genes [30]. Dysregulation of FOXP3 has been reported in the context of various tumors types indicating that it could be a poor prognostic factor in colorectal cancer and bladder cancer [31,32], but a potential tumor suppressor gene in breast cancer [33,34]. Nevertheless, whether FOXP3 can regulate the transcription of EXO1 is poorly understood.
In this study, we discovered that EXO1 expression is significantly higher in HCC tumor tissues. It was also found that EXO1 could promote the proliferation of HCC cells in vitro and in vivo, as well as promote the migration and invasion of HCC cells in vitro. FOXP3 could lead to the upregulation of EXO1 at the transcriptional level, where it could act as an oncogene in HCC as well. Our research revealed the relationship between the abnormal expression of EXO1 and malignant biological characteristics of HCC and might lead to the development of novel anticancer therapeutics for HCC treatment.

Materials and Methods
Gene chip analysis and human HCC samples 99 HCC tumor tissues and matched non-tumor tissues from archived patients at the Hepatic Surgery Centre, Tongji Hospital of Huazhong University of Science and Technology (Wuhan, China) were collected between January 2017 and December 2018. The diagnosis of HCC was confirmed by pathological examination and the differentiation status was graded according to the Edmondson grading system. After surgical resection, samples were stored at -80˚C immediately. Written informed consent was obtained from all subjects, and this study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Tongji Hospital, Huazhong University of Science and Technology.
HCC tissues and adjacent non-tumor liver tissues were placed in a cryogenic tank with dry ice and were delivered to the Shanghai Biotechnology Corporation. The Agilent version 16.0 was applied to screen the gene expression profiles of DNA repair pathways between HCC tissues and their adjacent liver tissues.

Immunohistochemical (IHC) staining assays
Formalin-fixed paraffin-embedded sections were deparaffinized in xylene and rehydrated with an ethanol gradient. The sections were placed in citric acid antigen repair buffer (Wuhan Goodbio Technology Corporation, Wuhan, China) and antigen repair was performed by microwaving on a low setting for 10 min. Endogenous peroxidases were blocked by 3% H2O2. Sections were then incubated overnight with an anti-EXO1 antibody at 1:200 (abs115859, Absin, Beijing, China) at 4℃. Secondary antibodies were incubated and peroxidase activity was detected using the EnVision kit (Dako, Glostrup, Denmark). Hematoxylin (Sigma Aldrich, St. Louis, MO, USA) was used for nuclear counterstaining. IHC scores were calculated as the positive staining intensity scores multiplied by the stained positive cells scores. Overall scores of >6 and ≤6 were defined as high expression or low expression of EXO1, respectively.

Cell culture
The
The supernatant-containing lentivirus produced by transfected 293T cells was collected 72 hours later. Collected lentiviral supernatants were filtered through a 0.45 μm filter (Millipore, USA). Target cells were infected with filtered lentivirus and 8 μg/mL polybrene (Sigma-Aldrich) to generate stable cell lines; treatment with 8 μg/mL puromycin (Ann Arbor, MI, USA) for 7 days was used to screen the positive cells.

Western blot analysis
Total protein was extracted from the cells using RIPA lysis buffer. Protein concentrations were measured with a BCA Protein Assay Kit (Thermo Fisher Scientific, MA, USA). Protein lysates were separated by SDS-PAGE and transferred onto a polyvinylidene difluoride (PVDF) membrane (EMD Millipore, Billerica, MA, USA). The membrane was blocked in non-fat milk in Tris-buffered saline containing 0.1% Tween-20 for 2 h and incubated with primary antibody (EXO1, ab95012, Abcam, Cambridge, MA, USA; β-actin, Cell Signaling Technology, Danvers, MA, USA) at 4 °C overnight. Then, membranes were incubated with secondary antibodies and signals were detected by Bio-rad chemiluminescence system.

Cell proliferation assay
Cells were inoculated into 96-well plates with 1×10 3 cells/well, and 100 μL of 10% cell counting kit-8 (CCK-8, Dojindo, Kumamoto, Japan) solution was added into each well at 0, 1, 2, 3, 4, and 5 days after cell adherence. The optical density (OD) values were measured by an EXL-800 absorbance reader at a wavelength of 450 nm at the indicated time points (after Cell Counting Kit-8 solution was added to the wells for 2 hours).

Cell migration and matrigel invasion assays
Cell migration and invasion assays were performed using transwell chambers (Corning, USA) with or without Matrigel (BD Biosciences, CA, USA). About 1.5×10 4 -10×10 4 cells in medium without FBS were seeded on transwell chambers with or without Matrigel. Below the transwell chambers, DMEM containing 10% fetal bovine serum (FBS) was added. An automated cell counter (Nexcelom Bioscience, USA) was used to quantify the number of migratory or invasive cells. Six fields at a magnification of ×200 were randomly selected for counting stained cells.

Cell colony formation
EXO1-knockdown or -overexpressing HCC cells (HLF, Huh7, Bel7402) were plated in 6-well plates (1000 cells/well) and culture media was replaced every 3 days. After 2-3 weeks when clones were visible to the naked eye, cells were fixed with 4% polyformaldehyde for 15 minutes and stained with crystal violet (Sigma-Aldrich Corporation) for 15 minutes. Photographs of the 6-well plates were taken, and the differences were counted.

Xenograft tumorigenicity assays
Animal studies were approved by the Ethics Committee of Tongji Hospital, Huazhong University of Science and Technology. All animal experiments were performed under the guidelines of the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Testing, and Education by the New York Academy of Sciences, Ad Hoc Animal Research Committee. 5-week-old male BALB/c (nu/nu) mice were obtained from the HFK BioScience Corporation (Beijing, China). All the mice were bred under specific pathogen-free (SPF) conditions at the Laboratory Animal Center of the Tongji Hospital. For subcutaneous tumorigenesis assays, HLF vector-and HLF shEXO1-788-transfected cells (2×10 6 ) were suspended in 100 μL of Dulbecco's modified Eagle's medium (DMEM) and injected subcutaneously into the left and right back of nude mice, respectively. Nude mice were monitored after the injection and sacrificed at day 30 after cell inoculation. The final tumor volumes were calculated according to the equation: V (volume, mm 3 ) = 0.5 × L (length, mm) × W 2 (width, mm 2 ).

RNA extraction and quantitative real-time PCR
Total RNA was extracted using TRIzol Reagent (Takara, Dalian, China) and reverse transcription was performed using a reverse-transcription PCR kit (TIANGEN, Beijing, China) according to the manufacturer's instructions. Quantitative real-time PCR was carried out using SYBR Green PCR master mix (TIANGEN, Beijing, China) on a Bio-Rad CFX system (Bio-Rad, Hercules, CA, USA) with the following primers: The Ct values of EXO1 were equilibrated to those of the internal control GAPDH.

Luciferase reporter assay
The region 2000 bp upstream of the EXO1 translational start site was amplified from the genomic DNA of the Huh7 cell line by PCR. EXO1 promoter regions of different lengths were subcloned into pGL4.17 (Promega, Madison, WI, USA). The open reading frame of FOXP3 and other candidate genes was amplified by PCR and then cloned into pcDNA 3.1(+) (Invitrogen, Thermo Fisher Scientific, MA, USA) to produce a series of products such as pcDNA 3.1-FOXP3 and so on. pGL4.17 and pcDNA3.1 were used as controls. pRL-TK was purchased from Promega (Madison, WI, USA). About 1 × 10 5 Huh7 cells/well were seeded into a 24-well plate and co-transfected with 4 ng of the pRL-TK, 200 ng of pgl4.17 and 400 ng of pcDNA3.1 after adherence. Lipofectamine 2000 (Invitrogen) was used for transfection. Cell lysates were prepared using passive lysis buffer (Promega) 48 h after transfection. A luciferase activity was performed using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) according to the manufacturer's instructions. The relative luciferase activity was measured using a GloMax 20/20 Luminometer (Promega).

Data acquisition and processing
Data for bioinformatics analysis were collected from The Cancer Genome Atlas (TCGA), Oncomine and Gene Expression Omnibus (GEO) databases. The cor.test () function in the R (version 3.5.2; https:// www.r-project.org/)) was used to calculate and verify the correlation coefficients between the EXO1 and FOXP3. Then R packages ggplot2 (version 3.1.1) was used for plotting.

Statistical analysis
Statistical analyses were performed with IBM SPSS Statistics 21.0 (SPSS, Chicago, IL, USA) or Prism 6.0 (GraphPad Software, La Jolla, CA) software. The results were presented as the mean ± standard deviation (SD) or mean ± standard error of mean (SEM). Quantitative data were compared by a twotailed Student's t-test, One-way ANOVA and Mann-Whitney U test when applicable. Categorical data were analyzed by the χ 2 test or Fisher's exact test. Kaplan-Meier survival analysis (log-rank test) was used to compare HCC patient survival. A two-sided value of p<0.05 was considered statistically significant.

EXO1 is an upregulated DNA damage repair gene in HCC
In order to screen the expression of DNA damage repair genes, three samples of HCC tissues and adjacent non-tumor tissues were collected for whole-genome sequencing. Gene chip analysis showed that 33 DNA damage repair genes were expressed at higher levels in tumor tissues compared to their corresponding non-tumor tissues (Figure 1), and in tumor tissues, 8 of them were expressed at levels greater than 10 times the normal levels. The expression of EXO1 in HCC tumor tissues was 27.2 times higher than that in peri-cancerous liver tissues (Table 1). Hence, we chose to perform follow-up studies on EXO1 and focused on its functional roles and mechanisms EXO1 in driving HCC development and metastasis.

Clinical significance of EXO1 in HCC patients
First, we investigated EXO1 expression in tumor and non-tumor liver tissues in three different clinical studies from the Oncomine database ( Figure 2A). The results showed that the expression of EXO1 was significantly higher in tumor tissues than that in their adjacent non-tumor tissues. Next, the expression level of EXO1 was detected in 99 HCC samples from Tongji Hospital (Wuhan, China) by western blotting ( Figure  2C, D). These results showed that the expression of EXO1 in HCC tumor tissues was higher (79%) than that in adjacent non-tumor tissues (21%) ( Figure 2B). We further explored the relationship between clinicopathological features and the expression level of EXO1 in HCC patients. Ninety-nine HCC patients were divided into an EXO1 low-expression group or EXO1 high-expression group. The detailed information is summarized in Table 2. High expression of EXO1 was significantly associated with the presence of liver cirrhosis (p=0.005). However, the expression of EXO1 was not correlated with gender, age, or AFP level. We then used the TCGA database to analyze the relationship between EXO1 expression and long-term prognosis of HCC. Kaplan-Meier analysis revealed that the disease-free survival rate and overall survival rate of patients with high EXO1 expression were significantly lower than that of patients with low EXO1 expression ( Figure 2E). Subsequently, the expression level of EXO1 was assessed by immunohistochemistry (IHC) in 10 pairs of tumor tissues and their adjacent non-tumor tissues. As shown in Figure 2F, EXO1 in hepatocytes was localized to the cytoplasm, and the average IHC score of EXO1 in tumor tissues was significantly higher than that in adjacent non-tumor tissues. Taken together, these results indicate that EXO1 expression is significantly upregulated in HCC tissues and high expression of EXO1 plays an important role in the prognosis of HCC patients and may contribute to the progression of HCC.

EXO1 enhances proliferation and colony formation of HCC cells in vitro, as well as tumor growth in vivo
To study the tumorigenic ability of EXO1 in vitro and in vivo, western blot was used to detect the expression level of EXO1 in 10 different HCC cell lines and one immortal liver cell line, HL-7702 ( Figure 3A). The results showed that EXO1 was upregulated in HLF and Huh7 cell lines, but downregulated in the Bel-7402 cell line. Hence, three short hairpin RNAs (shEXO1-788, shEXO1-789, and shEXO1-331) specifically targeting EXO1 were stably transfected into HLF and Huh7 cell lines, and an empty vector was transfected into each cell line (abbreviate as vector) and used as negative controls. The EXO1-shRNA significantly reduced EXO1 protein expression in HLF and Huh7 cells compared with their control vectors in the western blotting results and this result was also verified by RT-PCR ( Figure  3B). EXO1 was also stably transduced into Bel-7402 cell lines and an empty vector was used as a control. Western blot and RT-PCR verified that EXO1 was significantly overexpressed in transfected Bel-7402 cells ( Figure 3C). Using functional assays, the tumorigenicity of EXO1 was investigated. The Cell counting kit-8 (CCK-8) assay showed that EXO1 knockdown in HLF and Huh7 cells significantly inhibited cell proliferation, while EXO1 overexpression in Bel-7402 cells markedly promoted cell proliferation compared with their control vectors (p<0.01, Figure 3D, E). In the colony formation assay, the colony-forming ability of EXO1-knockdown cells was markedly lower than that of their control vectors (p<0.01, Figure 3F), whereas EXO1-overexpression in the cells showed significantly higher colony-forming ability than in the control vectors (p<0.01, Figure 3G). For in vivo tumorigenicity assays, HLF vector (control) and HLF-shEXO1-788-transfected HCC cells were injected subcutaneously into nude mice ( Figure 3H). As expected, the final tumor weights and volumes of the HLF-shEXO1-788 cell-injected group were significantly reduced compared to that of the HLF vector cell-injected group (p=0.03, Figure 3I, J). These results indicate that EXO1 has a strong promotive effect on cell proliferation and colony formation in HCC cells, and even promotes HCC tumorigenicity.

EXO1 promotes tumor invasion and metastasis in vitro
Metastasis is one of the reasons for poor outcomes in HCC. To further explore whether EXO1 can affect the invasion and metastatic abilities of HCC cells, we performed transwell assays. As shown in Figure 4A, migratory abilities were markedly decreased when EXO1 expression was downregulated by shRNA in HLF and Huh7 cells compared to their respective controls. Additionally, as shown in Figure 4B, invasive abilities of HLF-shEXO1 and Huh7-shEXO1 were observably reduced. In contrast, the migratory and invasive abilities of Bel7402-EXO1 cells overexpressing EXO1 were much higher than their corresponding controls ( Figure 4C, D). Altogether, these results demonstrate that EXO1 plays an important role in the invasion and metastasis of HCC.

Prediction of upstream regulators of EXO1 via bioinformatics analysis
To elucidate the molecular mechanisms of EXO1 overexpression in HCC tumor tissues, we concentrated on the region 2 kb upstream of the transcription start site (TSS) of EXO1 ( Figure 5A). Introduction of this 2-kb region upstream of firefly luciferase in pGL4.17-basic yielded significantly greater reporter activity compared with the promoter-less construct in Huh7 cell lines (28-fold) ( Figure 5B). With serial truncations of the 2 kb sequence at 500-bp intervals, we generated a set of luciferase reporter constructs (500 bp, 1000 bp, 1500 bp, 2000 bp). The luciferase reporter assay also showed a remarkable rise in luciferase activity when the upstream sequence of TSS of EXO1 was 1000 bp long, and its luciferase activity was the strongest of the four sequences (p<0.05, Figure 5B). To further determine whether there is a region with higher transcriptional activity involved in the 1000-bp sequence, we continued to generate luciferase reporter constructs with truncations of the 1000-bp sequence at 200-bp intervals, which were 200 bp, 400 bp, 600 bp, and 800 bp upstream of the TSS of EXO1. As shown in Figure 5C, the relative reporter activity of the 1000-bp sequence was still the highest compared to other sequences. These results suggest that the 1000-bp region directly upstream of the TSS of EXO1 possesses the highest transcriptional activity, and transcription factors possibly bind to the 1000-bp segment of EXO1 promoter.

FOXP3 could influence the expression of EXO1 in HCC
To identify the transcription factors that could bind to the promoter region of EXO1, we used the PROMO and JASPA websites to make predictions. According to the scoring system provided by the website and gene function itself, we focused on four transcription factors named FOXP3, E2F2, GATA2 and HNF1A. Plasmids were constructed with four coding sequences of these transcription factors mentioned above using pcDNA3.1 as a vector. We detected the transcriptional activity of these four transcription factors in the Huh7 cell line by luciferase reporter assay. Interestingly, as Figure 6A shows, FOXP3 had a significantly high reporter activity level compared to the other transcription factors. Based on this crucial result, we performed the same luciferase reporter assay in the Hep3B and PLC cell lines. As shown in Figure 6B and 6C, compared with co-transfection of the pcDNA 3.1 vector and pGL4.17-1000, when cotransfected with pcDNA 3.1-FOXP3 and PGL 4.17-1000, there was strong reporter activity. These results strongly suggest that FOXP3 may be involved in the transcriptional regulation of the EXO1 gene. Through bioinformatics analysis in the GEO database GSE109211, we found that the expression of FOXP3 and EXO1 had a positive correlation in HCC ( Figure  6D). Hence, we transiently transfected the pcDNA3.1-FOXP3 plasmid into the three different HCC cell lines (Huh7, PLC, Hep3B) and transfected a pcDNA3.1 vector as a control. Consistently, the mRNA level of EXO1 increased with FOXP3 transfection ( Figure 6E

Depletion of FOXP3 enhances the proliferation effects of EXO1 in vitro
Due to the regulatory impact of FOXP3 on EXO1, we wanted to explore whether FOXP3 can affect the proliferation activity of EXO1 in HCC cells. RT-PCR showed that among the three siFOXP3 (siFOXP3-1, siFOXP3-2, and siFOXP3-3) knockdown sequences, FOXP3 was significantly downregulated by siFOXP3-1 in HLF and Huh7 cell lines (p<0.05, Figure 7A, B). Next, we transfected siFOXP3-NC into shEXO1-vector HCC cells and shEXO1-788 HCC cells, respectively, and transfected siFOXP3-1 into shEXO1-788 HCC cells. As shown in Figure 7C and D, CCK-8 assays indicated that compared with HCC cells with only EXO1 knockdown, FOXP3 and EXO1 double knockdown significantly decreased the proliferation ability of the HCC cells (p<0.05). These results suggest that depletion of FOXP3 further reduces cell proliferation ability after knocking down of EXO1 in vitro, suggesting that FOXP3 may upregulate EXO1, and depletion of FOXP3 may reduce its expression.

Discussion
It has been shown that DNA damage caused by the hepatitis B virus (HBV) infection is one of the main molecular mechanisms of hepatocellular carcinoma [35][36][37]. HBV-DNA can be integrated into host genomic DNA and cause insertion mutations or other genomic instabilities [38,39]. Studies have shown that DNA double strand breaks (DSBs) induced by DNA damage are potential targets for HBV-DNA integration [40][41][42]. The body repairs DSBs through two major repair pathways: homologous recombination (HR) and non-homologous end-joining (NHEJ) [43]. Bill et al. previously suggested that integration of duck hepatitis B virus (DHBV) DNA into the host hepatocyte genome may be accomplished by the body's NHEJ repair of DSBs, and the integration could occur at the sites of DNA damage [44]. EXO1 plays an important role in DNA damage repair, especially in HR and NHEJ after DSBs [45][46][47][48]. As HR and NHEJ may participate in the integration of HBV-DNA, the relationship between EXO1 and HBV related HCC is worth investigating. In the present study we observed that EXO1 was frequently overexpressed in HCC samples and the same result was also seen in the Oncomine database. Analysis of EXO1 prognosis from the TCGA database showed the overall survival rate and tumor-free survival rate of patients with high EXO1 expression were significantly lower than those of patients with low EXO1 expression. Thus, we concluded that EXO1 is highly expressed in HCC and is associated with poor prognosis.
The link between EXO1 and cancer is still under continuous research and much progress has been made. As mentioned earlier, several SNPs in the EXO1 gene are related to the occurrence of certain tumors [17][18][19][20][21][22][23][24][25]. Some EXO1-specific mutations such as E109K and A153V are thought to be located in the region required for nuclease activity, thereby deactivating protein function and affecting cancer susceptibility [49,50]. In the mouse model, the incidence of lymphoma in EXO1 null/null increased and the survival rate decreased, but the microsatellite instability (MSI) was not affected [13]. EXO1 has also been reported to be overexpressed in several other cancers as described in the introduction [15,16,[51][52][53]. In general, EXO1 is poorly expressed and increased levels of EXO1 may cause cellular damage due to its 5'-3' exonuclease activity, which may be contrary to previous studies [8,54]. This may be due to the complex role of EXO1 in DNA damage repair and its tissue-specific expression. In humans, EXO1 is highly expressed in the testis and moderately expressed in the thymus, colon, and placenta [5,55]. The liver has high mitochondrial activity and increased reactive oxygen species (ROS), which may lead to increased activation of the DNA damage repair pathway, resulting in a relatively high expression level of EXO1 in the liver [55]. Muthuswami et al. treated breast cancer cell line, MCF7, with different alkylating agents such as carboplatin, cyclophosphamide, etc. to induce DNA repair and found that EXO1 expression increased with increasing concentration of these alkylating agents [51]. de Sousa et al. observed that EXO1 knockdown supported a faster DNA-DSB restoration after irradiation (IR) exposure in glioblastoma cell line T98G [15]. All these, explain to some extent that EXO1 is involved in the DNA repair pathway under cancerous conditions. The overexpression of EXO1 in tumors may also increase the genetic instability and contribute to tumor initiation and progression through its involvement in recombinant events such as DSBs and telomere stabilization [56].
At present, the research on the role of EXO1 in HCC is still lacking. Previously, it was reported that SNPs K589E (rs1047840) and rs3754093 of EXO1 could increase and decrease the susceptibility of HCC, respectively [17,18]. In 2018, Dai et al. used clonogenic assays to find that the expression of EXO1 affects the HCC cell survival under irradiation, and EXO1 overexpression has a poor prognosis for HCC patients [16]. In other types of tumors, as previously mentioned, Muthuswami et al. and de Sousa et al. investigated the possible role of EXO1 in different types of tumors through the breast cancer cell line MCF7 and the glioblastoma cell line T98G, respectively [15,51]. In 2019, Luo et al. found the differential expression of EXO1 in five prostate cancer cell lines, and the EXO1 expression in the other four high-invasive/metastatic prostate cancer cell lines significantly increased compared with the lowinvasive/metastatic potential cell line LNCaP [53]. These researches above led us to explore the role of EXO1 in the malignant biological characteristics of HCC by using HCC cell lines in vitro and in vivo. In this study, we observed that knockdown of EXO1 reduced cell proliferation in vitro and in vivo, while overexpression of EXO1 enhanced it. Subsequently, transwell assays showed that EXO1 could markedly increase the cell motility of HCC cells in vitro, indicating that EXO1 can affect the migration and invasion of HCC cells. Altogether, these functional assays indicate that EXO1 may play a carcinogenic role in HCC. However, previous research reported that knockdown of EXO1 in mice exhibits a mild phenotype, which may be contrary to our experimental results [57]. This may be due to complex environmental factors in vivo as well as the multiple roles the EXO1 gene itself plays in DSBs. There are still some researches suggesting that the absence of EXO1 in mice will show MMR defects, apoptotic defects, and sterility [10,11]. Therefore, although we suggest that EXO1 may function as a tumorigenic gene in vitro, more in vivo experiments are needed to validate these conclusions.   EXO1 has been previously reported to be regulated by a variety of molecules and to bind to certain specific proteins [58,59]. Studies on DNA mismatch repair (MMR) have reported that EXO1 interacts with the mismatch repair factors mutL homolog 1 (MLH1), mutS homolog 2 (MSH2), mutS homolog 3 (MSH3), and proliferating cell nuclear antigen (PCNA) to excise the newly synthesized DNA sequences containing errors during the MMR process [60][61][62][63]. Checkpoint regulatory proteins 14-3-3 are widely reported as interaction partners of EXO1. Engels et al. identified in 2011 that in yeast and mammalian cells, 14-3-3 proteins interact with EXO1 in vivo to regulate the phosphorylation status of EXO1, thereby promoting EXO1-dependent fork progression after inhibiting replication [64]. In 2012, Andersen et al. demonstrated specific interactions between EXO1 and all the 7 isoforms of 14-3-3 through in vitro GST pull-down assays, and they demonstrated that the binding may involve a second unrecognized binding motif in EXO1, the amino acid S746. They also demonstrated that the 14-3-3 association may affect the nuclease activity of EXO1 [65]. In 2015, Chen et al. showed that 14-3-3 can interact with the central portion of EXO1, at least partially, by suppressing its association with PCNA, to negatively regulate EXO1 damage recruitment and thus prevent excessive DNA excision [66]. However, very little is known about the transcriptional regulation of EXO1. Muthuswami et al. discovered that transcription factors E2F and Myc may be involved in the regulation of EXO1 by using luciferase reporter assays [51]. In this study, we uncovered a new potential transcriptional regulator of EXO1, FOXP3, by truncating the promoter region of EXO1 and using bioinformatic prediction. FOXP3 has been proposed to function as a tumor suppressor in breast and prostate epithelial cells [67,68]. However, FOXP3 could also be a prognostic factor in breast cancer and play an important role in the progression of cervical cancer [69,70]. In a recent report from 2019, Ou et al. reported that FOXP3 silencing may be associated with the inhibition of cell proliferation and migration of HCC cells, as well as the induction of apoptosis [71]. By transiently transfecting FOXP3 into HCC cell lines, we observed that FOXP3 could promote the transcriptional activity of EXO1 by a luciferase reporter assay. We further found that EXO1 mRNA levels were significantly upregulated after transient transfection of FOXP3. Moreover, bioinformatics analysis revealed a positive correlation between the expression of EXO1 and FOXP3, and a CCK-8 assay in HCC cells proved that FOXP3 could enhance the proliferative effect of EXO1 in vitro. These results indicate that FOXP3 could upregulate the transcription of EXO1 and might further regulate the expression level of EXO1. This is consistent with the reported results of Ou et al. Further explorations are required to validate this perspective.
Nonetheless, this study has some limitations. Firstly, the size of the clinical samples in this study was relatively small, with only 99 cases. Therefore, the number of samples assessed needs to be expanded. Second, there were database limitations and the expression level of EXO1 and the prognosis of HCC patients need to be further verified by complete clinical follow-up. Finally, it is not uncommon that DNA repair genes are highly overexpressed in HCC. In addition to the possible explanations given above, X-Ray Repair Cross Complementing 2 (XRCC2) has also been reported to be up-regulated in a variety of tumors [72,73]. It is possible that DNA damage and repair processes are more active in tumor cells. RAD51B, RAD51C, RAD51D, and XRCC2 form the BCDX2 complex, which promotes the aggregation of RAD51 protein at HR sites [74]. The mechanism of EXO1 overexpression in HCC is complex and still needs to be further explored.
In summary, our research found that EXO1 was overexpressed in HCC and played a carcinogenic role in HCC, and that FOXP3 may upregulate the expression level of EXO1. For the first time, we have uncovered a possible transcriptional regulation mechanism of EXO1. This discovery provides clues for the relationship between DNA damage repair genes and HCC and might lead to the development of novel anti-cancer therapeutics for HCC treatment.