A novel functional polymorphism of GSTM3 reduces clear cell renal cell carcinoma risk through enhancing its expression by interfering miR‐556 binding

Abstract Dysregulation of glutathione‐S‐transferase M3 (GSTM3) has been related to clear cell renal cell carcinoma (ccRCC) in our former study. GSTM3 plays a pivotal role of detoxification and clearance of reactive oxygen species (ROS) in tumour tissues. This study aimed to examine: (1) the associations between GSTM3 single nucleotide polymorphisms (SNPs) and risk of ccRCC, and (2) the potential molecular mechanism accounting for its effects. 5 SNPs in 3′UTR of GSTM3 were initially genotyped in 329 cases and 420 healthy controls. A SNP‐rs1055259 was found to be significantly associated with the susceptibility of ccRCC (OR = 0.59, 95% CI = 0.41‐0.92; P = .019). The minor allele of rs1055259 (G allele) was associated with RCC risk. This SNP was predicted to affect microRNA (miR)‐556 binding to 3′UTR of GSTM3 mRNA. To determine the functional impact, plasmid constructs carrying different alleles of rs1055259 were created. Compared to rs1055259 A‐allele constructs, cells transfected with rs1055259 G‐allele construct had higher transcriptional activity and were less responsive to miR‐556 changes and gene expression. Elevated GSTM3 expression in G‐allele cells was associated with ROS activity and ccRCC development. Taken together, this study indicated that a functional polymorphism of GSTM3 ‐rs1055259 reduced susceptibility of RCC in the Chinese population. It influenced GSTM3 protein synthesis by interfering miR‐556 binding, subsequently suppressed ROS activity and ccRCC progression.

many problems of these novel strategies. 2 Little is known about the advancement of specific prognostic or risk molecular markers for RCC.
Recent genome-wide association studies (GWAS) indicated that several loci mapped on 2p21, 2q22.3, 11q13.3, 12p11.23 and 12q24.31 were associated with RCC risk significantly. [3][4][5] However, the genetic risk factor identified in Caucasian population is not so consistent with that in Asian population. 6 The causal genetic factors for RCC are still unclear.
In our previous study, we have screened the differentiated genes between metastasis and primary ccRCC cells using cDNA microarray.
Glutathione-S-transferase Mu3 (GSTM3), a GSTM-class subunit, was identified as a notably down-regulated gene. It both down-regulated in metastatic vs primary ccRCC cells and primary ccRCC cells vs adjacent normal renal tissues. 7 We also studied the tumour suppressor role of GSTM3 in the progression of ccRCC and a polymorphism rs1332018 which was significantly associated with the postoperative prognosis of ccRCC. 8 In the current study, we report a novel SNP in ccRCC-rs1055259 in the 3 0 untranslated region (UTR) of GSTM3 significantly reduced the ccRCC risk and influence the binding of miR-556 to the 3 0 UTR of GSTM3. This might be the potential mechanism of rs1055259 impact the renal carcinoma risk.  Table 1. Cases and controls were frequency matched on gender and healthy controls were older than patients. All the patients were pathologically

| Blood samples collection and ccRCC cells primary culture
Ethylene diamine tetraacetic acid (EDTA) anticoagulated peripheral blood samples were collected from patients before nephrectomy and healthy controls. Genomic DNA was extracted using a Relax-Gene Blood DNA System (TIANGEN biotech, Beijing, China) according to the manufacturer's instruction. The fresh surgical specimens of ccRCC from 17 patients were harvested and immediately immersed in ice-cold PBS containing 1% penicillin/streptomycin and 0.5% glutamine (Beyotime, Shanghai, China). Primary cell culture was performed within 60 min after surgery as previously described. 7 As a variation of phenotype might occur at higher passages, we chose the ccRCC cultures at the fourth passage for functional experiments.
All SNPs were genotyped using matrix-assisted laser desorption

| Bioinformatic prediction of candidate SNPs
The SNP-flanking region RNA was online analysed using RNAfold   2.5 | Quantitative real-time PCR analysis of miR-556 The expression level of miR-556 was detected using RNA-tailing and primer-extension real-time PCR according to the instructions of All-in-One miRNA Detection Kit (Fulengen, Guangzhou, China). miR-556 forward primer: 5 0 -AGAAGCAATTCTAGAAG-3 0 ; the reverse primer was miRNA universal adaptor PCR primer (Fulengen).

| Luciferase reporter assay
To evaluate the binding of miR-556 to 3 0 UTR of GSTM3, 3 reporter constructs carrying 2 copies of rs1055259-A allele, G allele or mutant sequences at the 3 0 UTR of luciferase gene were created.
Firstly, three~100-bp DNA sequences centred at rs1055259 (A or G allele) or mutant sequence were synthesized; then, 2 tandem copies of these sequences were cloned into pGL-3 Basic vector (Promega, Madison, WI, USA) using restriction enzyme sites-Sal I and BamH I.
The schematic diagram for vector construction is shown in Figure   away, and the invaded cells were stained with ponceau. Finally, the images were captured with a phasecontrast microscope (Olympus IX51, Tokyo, Japan). The invaded cells were quantified by counting 10 independent visual fields to determine the average invasion ratio.

| Cell proliferation and invasion assay
Each assay was performed in triplicate.

| Association of SNP rs1055259 genotypes with miR-556 binding and GSTM3 expression
Bioinformatical analyses suggested that rs1055259 was located at miR-556 binding site, the 3 0 UTR of GSTM3. The A allele was predicted to bind more stably with miR-556 than G allele, but the binding site in mutant sequence was completely destroyed (Figure 2B). To evaluate the different affinity of A and G allele to miR-556, 3 luciferase reporter constructs were created. In ccRCC cells transfected with rs1055259-A allele construct, miR-556 mimic attenuated the luciferase activity by 45.43% while compared with control group; however, miR-556 mimic only reduced the luciferase activity by 25.61% in rs1055259-G allele group ( Figure 2C).
Antagomir-miR-556 increased the luciferase activity by 20.14% in A allele group than that by 14.96% in G allele group ( Figure 2D).
Intriguingly, CoCl2 treatment reduced the luciferase activity by 31.22% in A allele group and 20.59% in G allele group ( Figure 2E).

| rs1055259 variant reduced ROS activity
To evaluate the biological effect of rs1055259 on GSTM3 activity in vivo, we measured the ROS activity in ccRCC cells from patients with AA, AG and GG genotypes. As shown in Figure 4,

| rs1055259 variant suppressed the proliferation and invasion of ccRCC cells
It has been proven that GSTM3 is a tumour suppressor in ccRCC. 8 To confirm the relationship between rs1055259 and progression of  While the ROS scavenger-NAC was added in the concentration of 1 mmol/L, the number of invasion cells was obviously decreased by 18.17% in AA group and 23.12% in AG group, but no significant change in GG group ( Figure 5A,C). Similarly, the cell viability for proliferation ability investigation in AA group was the highest (standardized as 100%) than AG and GG groups (64.50% and 55.39%, respectively). In addition to NAC, the cell viability was notably reduced by 43.52%, 36.43% and 30.71% in AA, AG and GG groups, respectively ( Figure 5B). Taken  To the best of our knowledge, the functional polymorphism-rs1055259 has not been reported in other investigations. This is the first study to demonstrate the potential function of rs1055259 in GSTM3 expression and ccRCC development.
Another intriguing finding was that rs1055259 might be related to ROS activity, by fine tuning the GSTM3 expression. As is known to all, cytosolic glutathione S-transferases (GSTs) are members of phase II metabolic isozymes super family. They protect cells against electrophilic damage via catalysing the conjugation of ROS and glutathione. 9,10 Dysfunction of GSTs has been implicated in the progression of RCC, because of the defect of anti-oxidant capacity. 11 Our study showed that the ROS activity (reflected by DCF immunofluorescence) was highest in ccRCC cells with rs1055259 AA genotype than AG and GG genotypes. However, rs1055259 variant was proven associated with GSTM3 expression. Namely rs1055259 might modulate the ROS activity via elevating GSTM3 expression.
Thus, the subjects carrying A allele are more prone to ccRCC because their ROS activity tends to be higher against those with G allele.
Generous studies revealed that epigenetic inactivation of GSTM3 or low GSTM3 expression correlated with tumorigenesis and advanced AJCC stage of cancer. 12,13 These reports, together with our previously study, 7 suggested that GSTM3 might function as a suppressor in tumour progression. Elevated ROS activity also played an important role in cancer cell survival and development. 14 Thus, we detected the relationship between rs1055259, ROS activity and ccRCC development. In transwell assay, we found that the invasion cells per visual field were the most in AA group and then AG and GG groups. After ROS scavenger-NAC was added, the numbers of invasion cells were obviously decreased. Similarly, the cell viability in AA group was higher than AG and GG group. With NAC+, the cell viability was notably reduced by more than 30%.  [15][16][17] Several studies revealed that unregulated and augmented ROS could initiate redox-linked signalling responses and irreversible injuries in RCC. 18,19 Moreover, changes of cellular redox balance in RCC might also be associated with alterations of glutathione S-transferase (GST) expression and phenotype. 18 Therefore, it is necessary to clarify underlying mechanisms between the important anti-oxidant enzyme-GSTM3 in ccRCC, 7,8 thiol level, ROS activity and redox signalling responses. The further investigations should be carried out.
In summary, we demonstrated that rs1055259 in GSTM3 3 0 UTR affected the miR-556 binding and GSTM3 expression in ccRCC for the first time. The G allele of rs1055259 predisposes hosts to downregulated GSTM3 and reduced ccRCC susceptibility. rs1055259 variant genotype was a significant genetic protect factor. These findings will be helpful in diagnosis prediction and surveillance of ccRCC, as well as in the advancement of targeted therapy for this renal malignancy. The current study also indicated that it warranted further investigations of GSTM3 expression in ccRCC.

CONFLI CTS OF INTEREST
The authors have declared that no conflicts of interest exist.