The Sp1-mediaded allelic regulation of MMP13 expression by an ESCC susceptibility SNP rs2252070

Metallopeptidase 13 (MMP13), a well-known and highly regulated zinc-dependent MMP collagenase, plays a crucial part in development and progression of esophageal squamous cell carcinoma (ESCC). Therefore, we examined associations between ESCC susceptibility and four haplotype-tagging single nucleotide polymorphisms (htSNPs) using a two stage case-control strategy. Odds ratios (OR) and 95% confidence intervals (95% CI) were computed by logistic regression model. After analyzing 1588 ESCC patients and frequency-matched 1600 unaffected controls, we found that MMP13 rs2252070 G > A genetic polymorphism is significantly associated with ESCC risk in Chinese Han populations (GA: OR = 0.63, 95% CI = 0.54–0.74, P = 1.7 × 10−6, AA: OR = 0.73, 95% CI = 0.66–0.81, P = 1.8 × 10−6). Interestingly, the rs2252070 G-to-A change was shown to diminish a Sp1-binding site in ESCC cells. Reporter gene assays indicated that the rs2252070 A allele locating in a potential MMP13 promoter has low promoter activities. After measuring MMP13 gene expression in sixty-six pairs of esophageal cancer and normal tissues, we observed that the rs2252070 A protective allele carriers showed decreased oncogene MMP13 expression. Results of these analyses underline the support of the notion that MMP13 might function as a key oncogene in esophageal carcinogenesis.

Esophageal squamous cell carcinoma (ESCC) is one of the most common malignant tumors in the world, showing a relatively high morbidity in China 1 . Overall, ESCC patients have poor prognosis unless exhaustive treatments, including radical surgery, chemotherapy, and radiotherapy, are given. Accumulated evidences indicate that cigarette smoking, heavy ethanol consumption, micronutrient deficiency as well as dietary carcinogen exposure are main environmental risk factors of ESCC 2,3 . Recent progress on genome-wide association studies (GWAS) and candidate gene studies indicate that genetic makeup also contributes to ESCC etiology [4][5][6][7][8][9][10] .
As a family of zinc-ion dependent endopeptidases, matrix metalloproteinases (MMPs) comprise more than 21 subtypes 11,12 . MMPs include collagenases and gelatinases, which mediate degradation of basement membranes and the extracellular matrix 11,12 . Matrix metalloproteinases 13 (MMP13), encoded by the MMP13 gene, is a well-known and highly regulated zinc-dependent MMP collagenase 11 . It has been reported that MMP13 is significantly overexpressed in ESCC tissues compared with normal esophageal epithelium 12 . Deregulated MMP13 expression might impact prognosis of ESCC patients through tumor invasion, vascular permeation, and lymph node metastasis [13][14][15] . MMP13, in combination with MMP7 and MMP9, are involved in early stage development of ESCC, and their co-expression predicts poor outcome for relatively early stage ESCC cases 14 . Moreover, MUC1 induces cancer cell metastasis by upregulating MMP13 in ESCC 15 . Due to the crucial role of MMP13 in ESCC development, it is essential to figure out the molecular mechanisms in fine-regulation of MMP13 expression during tumorigenesis.
In the MMP13 gene locus, multiple functional single nucleotide polymorphisms (SNPs) have been identified. One of the most studied SNPs is the MMP13 rs2252070 polymorphism. Yoon et al. firstly evaluated its biological function via in vitro reporter gene assays and electrophoretic mobility-shift assays (EMSA) and showed that this polymorphism was a functional variant 16 . In HepG2 cells, the MMP13 promoter with the rs2252070 A allele had approximately twice as much transcriptional activity as that with the G allele in the same position (P = 0.0037) 16 . EMSA using nuclear extracts prepared HepG2 cells demonstrated that MMP13 probes containing the A allele differed in their binding to nuclear factors from the probes containing the G allele 16 . The associations between this functional SNP and multiple malignancies were repeatedly investigated in different ethnic populations [17][18][19][20] . However, the details on how this genetic variant impacts MMP13 expression is still largely unknown. Considering the importance of MMP13 in cancer development, we selected 4 haplotype-tagging SNPs (htSNP) across the whole MMP13 locus and conducted three large independent hospital-based case-control studies to investigate the association between MMP13 genotypes and ESCC risk. In addition, to the best of our knowledge, we firstly examined the fine-regulation of MMP13 expression by rs2252070-mediaed allelic binding of Sp1 and its involvement in ESCC. To validate the biological function of MMP13 rs2252070 genetic variant in vivo, we detected the association between its genotypes and MMP13 mRNA expression levels in normal and cancerous esophagus tissues. SNP selection and genotyping. There are multiple SNPs in MMP13 covering a ~13 kb region of chromosome 11q22.3. As a result, we utilized an htSNP approach to examine the MMP13 polymorphisms globally 22 . Genotyped HapMap SNPs among Han Chinese population (HCB data, HapMap Rel 27, NCBI B36) with a minor allele frequency > 5% were included in the selection. The htSNPs were chosen in a ~17 kb region (~13 kb MMP13 locus and 2 kb up-stream as well as 2 kb down-stream regions of MMP13). Four htSNPs were finally selected with Haploview 4.2 software on a block-by-block basis, using a method described previously with the sample size inflation factor, Rh 2 , of ≥ 0.8 (Supplementary Table 2). MMP13 htSNPs (rs11225490, rs2252070, rs17099788 and rs3758854) were analyzed by the MassArray system (Sequenom Inc., San Diego, California, USA). A 5% blind, random sample of study subjects was genotyped in duplicates and the reproducibility was 99%.

Materials and Methods
EMSA. Synthetic double-stranded and 3′ biotin-labeled oligonucleotides corresponding to the Sp1 consensus binding sequence, MMP13 rs2252070G or rs2252070A sequences (Supplementary Table 3) and pure recombinant Sp1 protein (E639A, Promega) were incubated at 25 °C for 20 min using the Light Shift Chemiluminescent EMSA Kit (Pierce, Rockford, IL). The reaction mixture was separated on 6% PAGE, and the products were detected by Stabilized Streptavidin-Horseradish Peroxidase Conjugate (Pierce). Unlabeled probes at 100-fold molar excess were added to the reaction mixture before the addition of biotin-labeled probes in competition assays. Table 4) with XhoI and KpnI restriction sites were used to amplify multiple deletion fragments spanning 5′ -region of MMP13 (from − 186 bp to − 41 bp, relative to the transcription start site) from human genomic DNA carrying MMP13 rs2252070GG or rs2252070AA genotype. The PCR products were then digested with XhoI and KpnI (New England Biolabs) and ligated into an appropriately digested pGL3-Basic vector (Promega) containing the firefly luciferase gene as a reporter. The resultant plasmid, designated pMMP-G or pMMP-A. Complete DNA sequencing confirmed the orientation and integrity of these two reporter constructs.

MMP13 reporter gene constructs. Specific primer pairs (Supplementary
Dual luciferase reporter assays. KYSE30 and KYSE150 ESCC cells were transfected with both reporter constructs (pGL3-Basic, pMMP-G or pMMP-A) and pRL-SV40 (Luciferase Assay System; Promega). Dual luciferase activities were examined at 48 h after transfection as previously described 23,24 . For each reporter construct, three independent transfections were done, and each was performed in triplicate.
Statistical Analyses. Pearson's χ 2 test was used to examine the differences in demographic variables, smoking status, drinking status, and genotype distributions of MMP13 SNPs between ESCC cases and healthy controls. Unconditional logistic regression model was utilized to estimate associations between MMP13 genotypes and ESCC risk by odds ratio (OR) and their 95% confidence intervals (CIs). All ORs were adjusted for age, sex, drinking and smoking status, where it was appropriate. A P value of less than 0.05 was used as the criterion of statistical significance, and all statistical tests were two-sided. All analyses were performed with SPSS software package (Version 16.0, SPSS Inc., Chicago, IL).

Results
MMP13 rs2252070 is associated with ESCC risk in Chinese Han populations. No Significant differences were found between cases and controls for both case-control sets in terms of median age and sex distribution (both P > 0.05). This indicated that the frequency matching of age and sex was adequate (Supplementary Table 1). The genotype frequencies of MMP13 candidate SNPs (rs11225490 T > C, rs2252070 G > A, rs17099788 A > G and rs3758854 G > A) are summarized in Table 1. All observed genotype frequencies in either controls or cases conform to Hardy-Weinberg equilibrium. Distributions of the all genotypes were then compared among patients and controls. Frequencies of rs2252070 genotypes among patients differed significantly from those among controls (all P < 0.05). Logistic regression analyses revealed that rs2252070 SNP was significantly associated with ESCC risk (allelic OR = 0.70, 95% CI = 0.60-0.83, P = 2.1 × 10 −5 ) (Table 1). However, no statistically significant differences of other htSNPs were observed between cases and controls (all P > 0.05) ( Table 1). As a result, no additional analyses on these SNPs were performed in the next studies.
Associations between genotypes of MMP13 rs2252070 G > A SNP and ESCC risk were calculated using unconditional logistic regression analyses ( Table 2). The MMP13 rs2252070 A allele was found to be a protective allele. Individuals with the rs2252070 GA genotype had an OR of 0.65 (95% CI = 0.49-0.88, P = 0.004) for developing ESCC in Jiangsu Set, compared with individual having the rs2252070 GG genotype. Similarly, the rs2252070 AA genotypes had a significantly decreased risk for ESCC compared with the rs2252070 GG genotype (OR = 0.79, 95% CI = 0.65-0.95, P = 0.011). In Shandong Set, carriers of the rs2252070 GA or AA genotypes were significantly associated with decreased ESCC risk (OR = 0.65, 95% CI = 0.53-0.81, P = 8.8 × 10 −5 , or OR = 0.75, 95% CI = 0.65-0.86, P = 4.9 × 10 −5 ) ( Table 3). In the pooled analyses, we observed similar results (For GA genotype: OR = 0.63, 95% CI = 0.54-0.74, P = 1.7 × 10 −6 , For AA genotype: OR = 0.73, 95% CI = 0.66-0.81, P = 1.8 × 10 −6 ) ( Table 2). All ORs were calculated with adjustments of sex, age, smoking and alcohol drinking status. The risk of ESCC associated with the rs2252070 SNP was further investigated by stratifying for age, sex, smoking and alcohol drinking status using the combined data of two case-control sets ( Table 3). The variant genotypes of MMP13 rs2252070 (GA or AA) were consistently associated with a significantly decreased risk of ESCC in all subgroups.

MMP13 rs2252070 SNP mediated allele-specific Sp1 binding in ESCC cells. Since rs2252070 SNP
is located a potential Sp1 binding sequence of the MMP13 5′ -region (Fig. 1A), we then conducted EMSA to distinguish the differences in binding capacity between the rs2252070G and A alleles to Sp1 (Fig. 1B). As shown in Fig. 1B, we found that Sp1 protein bound only to the biotin-labeled oligonucleotide probe with the G allele but not the A allele probe. We also did competition assay using pure recombinant Sp1 protein and Sp1 antibody (Fig. 1C). Interestingly, there is a super-shift band after adding Sp1 antibody, which conforming that the binding protein with MMP13 G probe is Sp1. Allelic regulation of rs2252070 on MMP13 promoter activities. Because the rs2252070 SNP is located in the Sp1 binding sit of MMP13 promoter, we speculated that this polymorphism will influence gene expression of MMP13. Therefore, we examined the promoter activity of this region by two luciferase reporter constructs ( Fig. 2A) with different rs2252070 allele in ESCC KYSE30 and KYSE150 cells. Interestingly, the MMP13 rs2252070G allelic reporter construct (pMMP-G) showed significantly higher luciferase activities compared to the rs2252070A allelic reporter construct (pMMP-A) in KYSE30 cells (P < 0.01) (Fig. 2B). Moreover, KYSE150 cells transfected with pMMP-G showed significantly higher luciferase activities compared to cells expressing pMMP-A (P < 0.05) (Fig. 2C). This indicates that Sp1 could bind the rs2252070G allelic MMP13 promoter and prompt increased MMP13 expression.

Impacts of rs2252070 on MMP13 expression in vivo. Considering rs2252070 G-to-A change could
impact MMP13 promoter activity in cancer cells, we investigated whether there is an allele-specific effect of rs2252070 SNP on MMP13 expression in esophagus tissue specimens (Fig. 3)     To standardize transfection efficiency, we cotransfected pRL-SV40 with these reporter gene constructs. Fold-changes were calculated by defining the luciferase activity of cells co-transfected with pGL3-basic as 1. All experiments were performed in triplicates in three independent transfection experiments and each value represents mean ± SD. Compared with pGL3-Basic transfected cells, * P < 0.05; * * P < 0.01.
To further verify if Sp1 indeed promotes MMP13 expression, we knocked-down endogenous Sp1 with siRNAs and examine the expression of MMP13 in two ESCC cell lines. Decreased MMP13 expression was observed after Sp1 silencing (Fig. 4), suggesting that Sp1 enhance endogenous MMP13 expression in ESCC cells.

Discussion
In the current study, we systematically examined the impacts of SNPs in the MMP13 locus on ESCC susceptibility via a case-control design as well as gene expression of MMP13 in vitro and in vivo. After genotyping 4 htSNPs at the discovery stage, we identified one ESCC susceptibility genetic polymorphism (rs2252070) which were validated in another case-control sets. Reporter gene assays indicated that the ESCC susceptibility SNP rs2252070 a potential MMP13 promoter has a genotype-specific effect on MMP13 expression. Our observations support the hypothesis that genetic polymorphisms in oncogene regulatory elements might impact genetic susceptibility of ESCC.
ESCC frequently shows extensive local invasion or regional lymph node metastasis at diagnosis and, thus, is one of the most common aggressive diseases with poor outcome. Tumor invasion and metastasis require the actions of MMPs for degradation of extracellular matrix. MMP13, a well-known zinc-dependent MMP collagenase, has been identified as a essential MMP in facilitating ESCC development. Elevated MMP13 expression was not only observed in ESCC tissues but also associated with tumor invasion, vascular permeation, and lymph node metastasis [12][13][14][15] . Therefore, it is crucial to examine fine-regulation of MMP13 expression during esophageal carcinogenesis. Accumulated evidences demonstrated that the MMP13 rs2252070 polymorphism is a regulatory polymorphism in cells, but the detailed mechanisms are far from clear. Here, for the first time, we found that the rs2252070 G allele but not A allele could bind Sp1 and promote MMP13 expression in ESCC. This is consistent with our molecular epidemiology studies showing that MMP13 rs2252070 A allele is a protecting allele of ESCC in Chinese. That is, subjects carrying the MMP13 rs2252070 A allele without Sp1 binding have less oncogene MMP13 expression; therefore, these carriers show decreased risk to develop ESCC.
In all, we demonstrated that functional MMP13 rs2252070 SNP was associated with a significantly decreased risk of ESCC in Chinese Han populations. Functional analysis showed that the rs2252070A allele contributes to significantly decreased expression of MMP13 in vitro and in vivo in the target tissues, which is most likely due to a diminished Sp1 regulation. These findings constitute strong evidence in support of the notion that MMP13 might function as a key oncogene in esophageal carcinogenesis.