Highly Expressing SCARA5 Promotes Proliferation and Migration of Esophageal Squamous Cell Carcinoma

Background Thrombospondin type 1 domain-containing 7A (THSD7A) was reported to play a procancer role in esophageal squamous cell carcinoma (ESCC). The aim of the study was to screen the downstream functional genes of THSD7A and explore their functions in ESCC, based on the reported research into THSD7A function and on gene microarrays. Methods We adopted quantitative reverse-transcription polymerase chain reaction (qRT-PCR) and Celigo high-content screening (HCS) technology to screen the downstream genes of THSD7A. The expression level of target genes was examined by PCR, western blot, and immunohistochemistry (IHC). The effects of these target genes on ESCC malignant biological behavior were performed in vivo and in vitro. The Kaplan-Meier (K-M) survival analysis and Cox regression were used to analyze the prognostic significance of target genes in ESCC patients. Experiments in the literature on liver cancer (LC) were repeated to verify the functions of these genes in different tumors. We further explored the cancer-promoting mechanism of target genes in ESCC by sequencing of the genes' exons. Results Scavenger receptor class A member 5 (SCARA5) was proved to be the downstream driving gene of THSD7A. SCARA5 promoted cell proliferation and migration but inhibited apoptosis in ESCC. IHC results confirmed that SCARA5 expression in ESCC exceeded that in normal tissues. The K-M survival analysis indicated that SCARA5 expression quantity was not related to prognosis, but tumor volume and T classification were both the independent prognostic factors. Repetition of experiments in LC in the literature confirmed that SCARA5 had exactly opposite functions in EC and LC. Conclusion SCARA5 was related to the development and occurrence of ESCC. Our findings suggested that it was a potentially diagnostic individualized therapeutic target for ESCC in the future and that its application could possibly be combined with that of upstream THSD7A gene.


Introduction
Esophageal cancer (EC) has been reported as the 8th most common malignancy, as well as the 6th most common cause of cancer deaths in the world [1]. Every year, approximately 350,000 new cases of EC are diagnosed globally, and 200,000 people die of EC annually [2]. EC morbidity and mortality rates in China are both the highest worldwide, accounting for more than 50% of worldwide toll. China also has the highest incidence of squamous cell cancer, which comprises >90% of its EC cases [3]. Therefore, it is important to explore candidate genes that could have significance to the clinical diagnosis, as well as the treatment of esophageal squamous cell carcinoma (ESCC) at the molecular level [4]. Our previous study found that the thrombospondin type 1 domain-containing 7A (THSD7A) gene plays a procancer role in ESCC and participates in important signaling pathways, including mammalian target of rapamycin (mTOR). However, whether there are functional target genes downstream of THSD7A in ESCC has not been reported. We found that scavenger receptor class A member 5(SCARA5), located on chromosome 8p21.1, is the downstream driving gene of THSD7A in ESCC [5]. SCARA5 has significant class A scavenger receptor characteristics [6]. Studies on SCARA5 are mainly focused on tumors, inflammatory responses, and cardiovascular disease (CVD) [7][8][9][10][11][12][13][14]. SCARA5 expression is significantly downregulated in various tumor tissues including liver cancer (LC), rectal cancer, and glioblastoma [15,16] and might affect cell proliferation and invasion abilities by the focal adhesion kinase (FAK) signaling pathway and epithelial-mesenchymal transition (EMT) [5,[17][18][19][20]. SCARA5 has been proved to be related to tumor occurrence and development. It is a tumor suppressor in various tumors, including LC and breast cancer (BC). Tumorrelated studies have shown the clinical application and research prospects of this gene, suggesting that it might play the important role in the treatment of tumors. However, the expression level of SCARA5 in ESCC and the gene's effect on the occurrence and development of this disease have not yet been reported. Based on the abovementioned findings, quantitative reverse-transcription polymerase chain reaction (qRT-PCR), western blot, Celigo high-content screening (HCS), exon sequencing, and functional experiments were used to screen and verify multiple functional downstream genes of THSD7A, including SCARA5. Through HCS experiments, we determined that SCARA5, as the downstream driving gene of THSD7A, has a critical role in promoting ESCC. The combination of IHC with clinicopathological and prognostic analyses revealed the significantly enhancing SCARA5 expression in cancer tissues compared to the adjacent. Our repetition of experiments mentioned in the previous studies confirmed that SCARA5 functioned differently in different tumors. We used exon sequencing and wild-type (WT) gene overexpression to rule out the possibility of functional differences caused by mutations. These experiments laid a foundation for further study of SCARA5.

Patients and Samples.
A tissue microarray for IHC analysis of SCARA5 was provided by Shanghai Outdo Biotech Co., Ltd (Shanghai, China). The microarray contained tissues from 100 ESCC cases and 80 paired normal adjacent tissues. The 100 patients included 74 men and 26 women. Patients' median age was 68 years (48-82 years). Patient's tumor staging and grade scores followed the seventh American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) and tumor, node, and metastasis (TNM) classification of EC. All of the clinical and pathological parameter data for the analysis were obtainable for the 70 of 100 cases of ESCC. Of the other 30 cases, 14 had no gross information available, while the other 16 cases had no tumor volume information available. Patients did not receive any therapy before esophagectomy.

Cell
Culture. Three human EC cell lines were included. Eca109 and TE-1 were obtained from the Shanghai Cell Bank of Chinese Academy of Sciences, and EC9706 was from China Center for Type Culture Collection (Wuhan, China). All cells were maintained in Dulbecco's Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS) (Gibco, Karlsruhe, Germany) containing 1% penicillin and streptomycin solution. The cells were grown at 37°C, 5% CO 2 , and humidified atmosphere.
2.3. Transfection of shRNA and sgRNA Vectors. Short-hairpin and single-guide ribonucleic acid (shRNA and sgRNA) sequences against target genes and scramble sequences and lentiviral vectors of target genes containing hU6, ubiquitin, multiple cloning site (MCS), internal ribosome entry site (IRES), and enhanced green fluorescent protein (eGFP) were both constructed and synthesized by GeneChem (Shanghai, China). We transfected the ESCC cells (5 × 10 4 /ml) with lentivirus-mediated target gene shRNA, sgRNA, and negative control. The cells were collected after transduction for 72 h and then used for qRT-PCR and western blot analysis.

Quantitative Reverse-Transcription PCR (qRT-PCR).
We isolated total RNA from samples of Eca109, EC9706, and TE-1 cells using TRIzol reagent (Invitrogen, Carlsbad, California, USA). We detected RNA purity and concentration using NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). RNA was reversely transcribed into complementary deoxyribonucleic acid (cDNA) using a PrimeScript RT-PCR kit (TaKaRa, Dalian, China). We used an IQ5 RT-PCR system (Bio-Rad, Laboratories, Hercules, CA, USA) with SYBR Select Master Mix (Boster, Wuhan, China) to conduct qRT-PCR per manufacturers' instructions. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal reference for normalization. We evaluated the relative expression of target gene messenger ribonucleic acid (mRNA) via the 2-ΔΔCt method. The primers were shown in Table 1S.
2.5. Immunohistochemistry (IHC). Subsequently, we washed the samples twice in xylene (5 min per wash); then with graduate ethanol, 95%, 70%, 50%, and 30%; and then in phosphate-buffered saline (PBS) in distilled water for 5 min on a shaker, step by step. We used 3% hydrogen peroxide to inactivate endogenous peroxidase. For the next step, the samples were heated in sodium citrate buffer (pH 6.0) in a microwave oven maintaining at 95°C for 15 min to retrieve the antigen. We incubated the tissue sections at 4°C, overnight, with rabbit anti-SCARA5 primary antibody (1 : 200; Abcam, Cambridge, UK). After washing for three times with PBS, we added biotin-labeled secondary antibody (ZSGB-BIO, Beijing, China) and incubated for 15 min. After another wash, we used 3,3′-diaminobenzidine (DAB) to visualize immunoreactivity and hematoxylin for counterstaining. Staining was scored by the percentage (0, <5%; 1, 5-25%; 2, 26-50%; and 3, >50%) and positive staining intensity (0, none; 1, weak; 2, moderate; and 3, strong). The patients were 2 Journal of Immunology Research divided into two, the low expression group (total score, ≤6) and the high expression group (total score, 6-9), according to the total score.  19.0 (IBM Corp., Armonk, NJ, USA). Pearson's chi-square and/or Fisher's exact tests were used for the correlation between SCARA5 expression and patients' clinicopathological characteristics. t-test was used to calculate discrepancies between groups. On the basis of a log-rank test and Cox's proportional hazard regression model, a Kaplan-Meier (K-M) survival curve was used to analyze the remarkable between-patient differences in prognosis. Values were expressed as means ± standard deviations (SDs). A P < 0:05 value of two-sided was considered as significant level.

Results
3.1. Screening the Downstream Genes of THSD7A. In our previous study, we reported that THSD7A plays an important role in ESCC. 37 candidate genes listed in Table 1 were revealed after we compared the results from the Affymetrix human GeneChipPrimoView (Affymetrix, Santa Clara, CA, USA) and PubMed database. All of these genes are rarely reported in esophageal cancer.

Detecting the Downregulation Rate of the Candidate
Genes via shRNA-THSD7A. Transfection of shRNA-THSD7A into Eca109 resulted in 86.5% knockdown efficiency of the target gene (Figure 1(a)). Further qRT-PCR detection showed that the 37 candidate genes were downregulated to different degrees. We divided candidate genes into three categories according to degree of downregulation: a >50% group, a 40-50% group, and a <40% group. Genes in the <40% group were considered to have no significant change in expression. We then performed functional screening of genes in the >40% group and found that 11 were significantly downregulated (5 genes, >50%; 6 genes, 40%-50%) and expressed as downstream genes of THSD7A. Therefore, we carried out further functional screening of these 11 genes, as detailed in Table 2. 3.3. Screening Downstream Functional Genes of THSD7A via HCS. We designed a mixture of three interference targets per gene with significantly different degrees of downregulation, and then, we transfected the mixtures into the Eca109 cells. Changes in transfected cells' proliferation ability was observed using the Celigo imaging cytometer. We selected those with genes that significantly affected cell proliferation function, and then, we analyzed proliferation ratio of the Eca109 cells in the KD group compared to the control. A ratio of ≥2 was defined as strongly positive, >1.5 as positive, and<1.5 as negative. The results showed two strongly positive genes and nine negative genes, as indicated in Figures 1(b) and 1(c). We separately compared Celigo screening results of two strongly positive genes (Figures 1(d) and 1(e)).

Functional Verification of Single
Target of SCARA5 and SMCO4. As two downstream genes of THSD7A, SCARA5 and single-pass membrane protein with coiled-coil domain 4 (SMCO4) showed significant effects on Eca109 proliferation in the Celigo assay with the mixed interference sequence. Next, we separately packaged three interference  Journal of Immunology Research sequences of each gene into vectors and transfected them into Eca109. The effect of each sequence on the Eca109 proliferation was detected via Celigo cytometer. The results of this single-target experiment showed that of the six interfering targets of the two candidate genes, sh-SCARA5 (PSC38272)and sh-SMCO4 (PSC37528) behaved as strongly positive targets, with proliferation rates double or more (Figures 1(f)-1(j) and Table 3). We selected the two most significant sequences for further in vitro functional experiments.    The results showed that SCARA5 was increased in the Eca109 and EC9706 cells and lowly expressed in the TE-1 cells (Figure 2(a)). Gene expression abundance in the cell was considered high when the ΔCt value ≤12 and low when the ΔCt value ≥16.
3.6. Detecting the Interference Efficiency of shRNA-SCARA5 (PSC38272) Sequence via qRT-PCR and Western Blot. After transfection of sh-SCARA5 (PSC38272) lentivirus vector into the Eca109 cells for 72 h, qRT-PCR detection found that SCARA5 at mRNA level in Eca109 of KD group was significantly inhibited (P < 0:05), and knockdown efficiency reached 80.9%. Moreover, western blot analysis showed that sh-SCARA5 (PSC38272) was an effective target due to its knockdown effect on exogenous expression of the SCARA5 gene (Figures 2(b) and 2(c)).

ShRNA-SCARA5 Inhibited Proliferation of the Eca109.
We used Celigo fixed-point real-time monitoring technology to detect the effect of SCARA5 on Eca109 cell growth. After transfecting Eca109 cells with shRNA lentivirus for 3 days, we spread the cells into 96-well plate, 2000 cells per well, and continuously monitored them for 5 days using the Celigo cytometer. The results showed that the absolute number of the Eca109 cells and the multiple of change in cell number were obviously reduced in the KD group compared to the control; that is, the proliferation rate of cells was significantly inhibited after SCARA5 knockdown (P < 0:01) The MTT results showed that for cells in the KD group, the light absorption rate of 490 nm and the multiple of change in light absorption rate decreased significantly after 72 h of transfection compared to the control, and cell proliferation activity was significantly inhibited (P < 0:01) (Figures 2(g) and 2(h)).
We conducted a clone formation experiment to detect the effect of SCARA5 on Eca109 proliferation. After 15 days, we counted the number of cell clones and found the number of Eca109 cell colonies in the KD group significantly reducing compared to the control (P < 0:01). The Eca109 cloneforming ability was significantly inhibited after SCARA5 knockdown (Figures 2(i)-2(k)).
The results of all of the above experiments suggested that SCARA5 can significantly affect the cell proliferation ability of the Eca109 cells.    The cell scratch test showed that after transfecting with shRNA-SCARA5 for 24 h, the scratch width of the Eca109 cells of the KD group was significantly larger than the control, and the difference was statistically significant (P < 0:05 ) (Figures 2(l) and 2(m)). Transwell experiment showed that the number of migrated cells in the KD group was significantly lower than in the control (P < 0:05) (Figures 2(n)-2(p)), which showed that SCARA5 had a significant influence on the migration ability of the Eca109 cells.
3.9. ShRNA-SCARA5 Inhibited Eca109 Cell Cycle in S and G2/M Phases. We detected cell cycle changes via FCM. The results showed that the percentage of cells in S phase and G2/M phase in the KD group was significantly higher than in the control (P < 0:05). SCARA5 might have affected cell proliferation function by acting on cell S phase and G2/M phase (Figures 2(q) and 2(r)).

ShRNA-SCARA5 Promoted Apoptosis of the Eca109
Cells. After transfecting Eca109 cells with shRNA-SCARA5 for 48 h, we detected the change in apoptosis rate via FCM. The results showed that the apoptosis rate in the KD group was significantly higher than in the control (P < 0:05), meaning that SCARA5 could significantly affect apoptosis of the Eca109 cells (Figures 2(s) and 2(t)).
3.11. ShRNA-SCARA5 Inhibited Subcutaneous Tumorigenicity of the Eca109 Cells In Nude Mice. First, the qRT-PCR test showed that SCARA5 gene knockdown efficiency in the KD group reached 83.3%. Then, we performed a subcutaneous tumorigenesis experiment using nude mice. In vivo imaging showed that both the fluorescence intensity and    Journal of Immunology Research expression range of the KD group were significantly reduced compared to the control (P < 0:05, Figures 3(a)  and 3(b)). Tumor-forming experiment showed that the tumor body in the KD was smaller and lighter than the control (P < 0:05, Figures 3(c)-3(e)). The in vivo imaging experiment showed that the total and average radiant efficiencies of the KD group were weaker than those of the control (Figures 3(f)-3(k)). Therefore, shRNA-SCARA5 could effectively suppress Eca109 cell subcutaneous tumorigenicity in nude mice. This further verified the cancer-

Expression Intensity of SCARA5 in ESCC Tissues Was
Remarkably Increased Compared with Normal Mucosal Tissues. We conducted IHC to detect expression levels and features of SCARA5 in ESCC. As seen in Figure 4(a), only cell membrane and cytoplasm were immunostained positive. We analyzed differences between the 100 cases of ESCC and the 80 paired normal adjacent tissues and found no obvious difference in baseline characteristics among the high and the low SCARA5 expression groups. The other clinicopathological parameters, including clinical stage, differentiation, tumor location, T/N classification, tumor volume, gross classification, vascular invasion, nerve invasion, and baseline parameters (sex and age), were not with correlation with the expression intensity of SCARA5 (Table 4). 11 Journal of Immunology Research 3.13. Expression Intensity of SCARA5 Was Not Related to Prognosis. The correlation between SCARA5 intensity and the prognoses of ESCC patients were performed using the K-M survival curves as well as Cox proportional-regression analysis. Medium follow-up duration was 19 months (range, 1-107 months). Patients' 1-, 3-, and 5-year overall survival rates were 63%, 28%, and 19%, respectively (Figure 4(b)). Prognoses did not significantly differ between patients with high and low SCARA5 expression (Figure 4(c) and Table 5), while the K-M univariate analysis indicated that sex, T/N classification, tumor volume, clinical stage, and lon-gest diameter of tumor were associated with prognosis (Figures 4(d)-4(i) and Table 5); Cox multivariate analysis indicated that tumor volume and T classification were both independent prognostic factors ( Table 5).
3.14. shRNA-SCARA5 Inhibited Proliferation of the EC9706 and TE-1 Cells. In the above experiments, SCARA5 was proven to play cancer-promoting role in the Eca109 cells. However, the authors of all SCARA5 tumor-related studies published so far believe that SCARA5 functions as a tumor suppressor gene [15][16][17][18][19][20][21][22][23]. In order to verify SCARA5  Journal of Immunology Research function in ESCC cells, a MTT assay was performed on ESCC cell lines EC9706 and TE-1. After SCARA5 was knocked down, the results showed that the EC9706 and TE-1 KD groups significantly decreased in light absorption rate and the multiple of change of 490 nm of cells compared to the control, and the cell proliferation ability was significantly inhibited (P < 0:05; Figures 2(u)-2(x)). These experimental results were consistent with those of the Eca109 cells.

Functional Verification of SCARA5 in Liver Cancer
Cells. Studies published to date on SCARA5's role in tumors have all involved in multiple tumors: the tumor suppressor function of SCARA5 was first discovered in LC, and the functional studies thereof were comprehensive. In view of the significant differences in SCARA5 findings in EC, we validated some of the proliferation and migration assays (clone formation, MTT, Transwell, and western blot) in the literature on LC. Clone formation results showed that the number of cell colonies in the Hep 3B and Huh-7 overexpression groups was significantly lower than the control (P < 0:05; Figures 5(a)-5(i)). MTT data showed that cell proliferation rate was significantly suppressed in both the Hep 3B and  and Huh-7 cell proliferation and migration but promoted those in EC and LC.

Exon Sequencing of SCARA5 in the Three ESCC Cell
Lines. To explore why SCARA5 function in ESCC was opposite to that in other tumors, we performed exon sequencing of the Eca109, EC9706, and TE-1 cells and compared the exon sequences of WT SCARA5 with those in the PubMed gene database. We found 11 exon mutations in the three cell lines, and 4 of the 11 were identical across all three lines (Figure 6(a) and Table 6). Since SCARA5 was a protooncogene in all three lines, we speculated that one or more of the four common mutations might be responsible for the functional changes in the gene. By comparing codons, we found that two of the four were noncoding mutations, a third was a synonymous mutation, and the last was a missense mutation. The missense mutation was at position 30 of exon 5, resulting in the conversion of methionine to isoleucine. Therefore, we speculated that this site mutation might be the cause of SCARA5 functional changes in the three ESCC cell lines.
3.17. Detecting the Proliferation Function of WT SCARA5 in the ESCC Cell Lines. We constructed a WT SCARA5 overexpression vector to determine whether SCARA5 could play a cancer-promoting role in EC due to mutation, and then, we used a MTT assay to detect proliferation capacity. In Eca109, EC9706, and TE-1 cell lines, we observed no significant changes in the WT overexpression group compared to the control (Figures 6(b)-6(g)). Therefore, our results showed that WT SCARA5 also played a cancer-promoting role in ESCC cells, negating our previous assumption that mutations had led to functional changes.

Detecting SMCO4 Proliferation in the ESCC Cell Lines.
We performed MTT and clone formation experiments on the basis of sh-SMCO4 knockdown. The MTT assay indicated no significant differences of absorptivity or variation ratio of the Eca109 cells between the KD and the control (Figures 7(a) and 7(b)). Clone formation showed that the number of Eca109 cell colonies was not significantly reduced in the KD and the control. The above experiments indicated that SMCO4 had no significant influence on Eca109 proliferation (Figures 7(c) and 7(d)).  Journal of Immunology Research a a a a a a a SCARA5-5-30-G a t t t t g a a a a a a (a)

Discussion
In this study, in order to clarify THSD7A's cancerpromoting function, mechanism, and suspected downstream genes in ESCC, we performed preliminary screening and validation of THSD7A's downstream driving genes [24]. We are the first to explore the clinical significance, functions, and mechanisms of the downstream gene SCARA5 of THSD7A in ESCC. So far, SCARA5 has been previously proved as a candidate tumor suppressor for hepatocellular carcinoma (HCC), BC, and other tumors [5,15,[25][26][27][28][29]. However, critical SCARA5 in ESCC remains unclear. Therefore, we analyzed the clinical significance and function of SCARA5 in ESCC by using ESCC samples in vivo and in vitro, and we explored the possible causes of SCARA5 tumorigenicity. As can be seen from the existing literature, SCARA5 is associated with conditions that include tumors, inflammation, CVD, hemophilia, retinopathy, bone lesions, connective-tissue disease, and pregnancy [10,12,25,[30][31][32][33]. Most studies on SCARA5 are tumor related, involving BC, colorectal cancer (CRC), LC, and other cancers [26,29,34]. These studies suggest that SCARA5 might act a particular role in tumor progression. However, it is worth pondering that most SCARA5 tumor studies have shown this gene significant effect in tumor inhibition, quite different from the findings of this study.
We began our study by identifying 11 genes, including SCARA5 and SMCO4, as THSD7A downstream genes by using qRT-PCR. Then, we used Celigo HCS technology for preliminary screening of mixed targets and secondary screening of single targets and found that SCARA5 and SMCO4 might be proto-oncogenes in ESCC [35,36]. Further functional verification tests confirmed that of the 11 genes involved in ESCC, SCARA5 was the only oncogene. The reason for the difference between our results and those of previous studies might have been the false-positive rate of HCS [37]. It can be inferred from our results that the upstream oncogene THSD7A not only affected the mTOR, extracellular signal-regulated kinase (ERK), and adenosine monophosphate-activated protein kinase (AMPK) signaling pathways but might also upregulate the expression of SCARA5 to promote cancer in ESCC [24].
In this study, Eca109 proliferation and migration abilities were significantly inhibited after we knocked down SCARA5 using shRNA. In addition, the results of our mouse subcutaneous-transplantation tumor model experiment were similar to those of our in vitro cell proliferation experiment. Proliferation and migration are important markers of malignancy, and SCARA5 can alter these properties [5,19]. Our data suggested that SCARA5 might play a cancerpromoting role in ESCC, which was contrary to the conclusions of most published SCARA5-related tumor studies. Therefore, we repeated the proliferation experiments in which SCARA5 was first found to be an inhibitor in HCC, and we also confirmed the experimental data of tumor inhibition function in LC. The reasons for our results might be Bold formatting indicates that 4 of the 11 exon mutations found in the three cell lines were identical across all three lines.
17 Journal of Immunology Research as follows: first, SCARA5 might have been mutated in ESCC, leading to the transformation of a tumor suppressor gene into a tumor promoter gene [38,39]. Second, the results of multiple tumor tissue samples detected in the ATLAS database show high expression of SCARA5 in tumors including BC and CRC. Third, EC, especially ESCC, is different from adenoepithelial and other tumors from different tissue sources. Fourth, a recent study on atherosclerosis found that the proliferating and migrating abilities of aortic smooth muscle cells were significantly inhibited after SCARA5 KD [12]. Moreover, a tumor study showed that elevated ferritin levels in SCARA5 transport might promote tumor metabolism and growth [21]. The above conclusions are consistent with the results of our study. To some extent, this also indicates the possibility of SCARA5-promoting cancer functions. In addition, after SCARA5 knockdown, the ratio of apoptosis rate to cell cycle arrest in S and G2/M phases was significantly increased, which were also common in tumor progression [5,27].
To confirm the cancer-promoting effect of SCARA5 in EC cell lines, based on our multiple proliferation related experiments in vivo and in vitro on the Eca109cells, we also conducted cell proliferation (MTT) experiments on the EC9706 and TE-1 cell lines. Our findings were consistent with those of the Eca109 experiments, which showed the cancer-promoting effect of SCARA5.
Using an ESCC microarray, we found that SCARA5 was significantly overexpressed in this cancer. However, no significant difference between SCARA5 expression and other clinicopathological information was found, including TNM stage, classification, and prognosis. This might have been due to the limited sample size; a larger study is needed to explain the clinical significance of SCARA5. It should be noted that SCARA5 expression was significantly different across a variety of tissues, including tumors and stem cells. On the one hand, SCARA5 expression is downregulated relative to normal tissues in 18 tumors, including esophageal adenocarcinoma (EAC), osteosarcoma, BC, and CRC [22]. On the other hand, it is upregulated in BC stem cells including MDA-MB-231 in triple-negative BC (TNBC) cell lines [23]. More interestingly, contrary to the previous findings, another tumor study found SCARA5 expression in glioma cells, neuroblasts, and BC [21]. These significant differences reflect that SCARA5 expression and biological function might be independent in different tissues. Other gene studies have reported differences in gene expression and function in different tumors. DEAD-box RNA helicase 3 (DDX3) and ubiquitin-specific peptidase 22 (USP22) genes differ significantly in expression and function according to tumor type [38,39]. The reasons for this difference might include the following: first, genetic mutations might fundamentally alter the function of genes. Second, the gene proteins in different

18
Journal of Immunology Research tumors might show different binding states. Third, other factors such as viruses might have an effect [39]. To investigate the first possibility, we found the mutation sites of three ESCC lines by exon sequencing, constructed a WT SCARA5 overexpression group and a mutant group to compare against one another in a proliferation experiment, and found that WT SCARA5 also played a role in promoting cancer. Therefore, we concluded that mutations could be ruled out, while the latter two possibilities need to be further explored.
In addition, it should be noted that SCARA5 metastasis was inconsistent between animal and cellular experiments. The results from clinical samples showed that SCARA5 was not associated with lymph node metastasis. But the Eca109 cell line was tested for SCARA5 phenotype after siRNA knockdown, and the results showed that SCARA5 could promote cell migration, suggesting that the gene might be necessary for in vitro migration. One possibility is that this result was due to the limitation of our clinical sample size; another is that the single intervention of the cell experiment was significantly different from the multivariate confounding of clinical samples.
In addition to significant differences between ESCC and matched normal control tissues, we found no significant association between SCARA5 expression and clinical features. Several limitations of this study must be acknowledged: first, the relatively small sample size of ESCC organizations might have led to biased or inadequate conclusions. Second, considering that the specificity and correctness of primary antibodies might lead to nonreproducibility, all of the primary antibodies should be fully evaluated before experimental use. Third, we used a tissue microarray to detect the phenotype, which might not have reflected the true heterogeneity of SCARA5 expression in individual tumors. Therefore, our current observations need to be confirmed in more cases.

Conclusions
We found that SCARA5 was the downstream gene of THSD7A and promoted proliferation and migration of ESCC cells. Moreover, we observed that SCARA5 was expressed differently between ESCC and normal esophageal tissues, but no significant phenotypic associations were found between SCARA5 expression and lymph node metastasis, invasion, or prognosis. Our study suggested that SCARA5 might play an important role in the canceration process and serve as a therapeutic and prognostic target for ESCC.