Functional genetic variants in complement component 7 confer susceptibility to gastric cancer

Study population

This case-control study contains 471 gastric cancer patients and 471 healthy controls. All patients were recruited from North China University of Science and Technology Affiliated Tangshan Renmin Hospital and Affiliated Tangshan Gongren Hospital from January 2011 to May 2015. Healthy individuals were from a large population underwent physical examinations in same area during the same period. We also detected the C7 expression level in serum of 70 healthy control and 70 gastric cancer patients. This study was approved by Institutional Review Board of North China University of Science and Technology (2019021). All subjects signed an informed consent form.

Potentially functional SNPs filtering

Based on NCBI dbSNP (https://www.ncbi.nlm.nih.gov/snp/) and Ensembl (http://asia.ensembl.org/index.html) database, we screened out all SNPs in the promoter (2,000 bp upstream of the transcription revelation site) and 3′ untranslated region (3′UTR) of C7 with minor allele frequency (MAF) over than 0.05 in Chinese population. Alibaba 2.1 (http://gene-regulation.com/pub/programs/alibaba2/index.html) tool was then used to predict the binding ability of transcription factors and SNP info Web server (https://manticore.niehs.nih.gov/) to predict the miRNA binding changes of SNPs in 3′UTR.

Genotyping of genetic variants

Peripheral blood DNA was extracted by using TIANamp Blood DNA Kit (TIANGEN, Beijing, China) according to manufacturer’s instructions. PCR-restriction fragment length polymorphism (PCR-RFLP) analysis were applied for genotyping. The target DNA fragment containing C7 rs1376178 or rs1061429 was amplified with primer pairs, rs1376178 PF (5′-GCTAGAATCAATGCAAAGCTATGCG-3′)/PR (5′-TCAGATCACTGTGTTGGAAAGTT-3′) and rs1061429 PF (5′-GCTAGAATCAATGC AAAGCTATGCG-3′)/PR (5′-AAGGAAAAGCTGTCCAGTGC-3′), respectively. PCR was performed in 6 µL PCR reaction mixture with 1 ×Ftaq PCR Mix, 20 ng genomic DNA and 0.1 µM each primer. The thermal cycling conditions for both C7 rs1376178 and rs1061429 variants were 3 min at 94 °C followed by 30 s at 94 °C, 30 s at 60 °C, and 30 s at 72 °C for 30 cycles, and then a final extension 3 min at 72 °C. PCR products for C7 rs1376178 C > A (125 bp) and C7 rs1061429 C > A (242 bp) were digested by Hha I (NEB, Ipswich, MA, USA) and Nco I (NEB, Ipswich, MA, USA) and then was separated on 3% agarose gel. For quality assurance, approximately 10% of the samples were randomly selected for re-genotyping and all results were in 100% concordance.

Bioinformatic analysis of C7 expression

To analyze the levels of the C7 mRNA expression in GC, the GEPIA databases were analyzed. GEPIA (http://gepia.cancer-pku.cn/) is a comprehensive and interactive web resource for analyzing cancer data, which includes 9,736 tumors and 8,587 normal samples from Genotype-Tissue Expression (GTEx) and The Cancer Genome Atlas (TCGA) (Tang et al., 2017). Additionally, we also extracted eQTL data from the GTEx database, where differences in gene levels under different SNPs were examined. P- value with < 0.05 was considered as statistical significance.

Enzyme linked immunosorbent assay (ELISA)

The serum samples from 70 normal individuals and 70 gastric cancer patients were used to detect the level of C7 protein by using enzyme-linked immunosorbent assay (ELISA). Human Complement (C7) kit were purchased from Cusabio company (Wuhan, China). We conducted the C7 expression analysis in serum according to manufactory’s instructor.

Cell culture

Gastric cancer cell line (BGC823) was purchased from Cell Bank of Type Culture Collection of the Chinese Academy of Sciences Shanghai Institute of Biochemistry and Cell Biology. BGC823 cells were cultured in RPMI-1640 (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Waltham, MA, USA) and 1% antibiotics (100 U/ml penicillin and 100 µg/mL streptomycin) in an atmosphere of 5% CO2 at 37 °C.

Plasmid constructure and luciferase reporter gene assay

A 1,615 bp DNA fragment containing rs1376178 site in the promoter of C7 was inserted into pGL3-basic plasmid (Promega, Madison, WI, USA) to conduct luciferase reporter gene assay. The PCR primers with Kpn I and Xho I (NEB, Ipswich, USA) cutting site adaptor were 5′-GGGTACCCTTTCCCACTTCCAGTGGTGC-3′) and 5′-CCGCTCGAG CTGAGATTTAGCTCCTACCCC-3′. The final plasmids with rs1376178 C or A allele were designed as pGL3_rs1376178C and pGL3_rs1376178A, respectively. A 1,268 bp fragment with rs1061429 site in the 3′ untranslated region (UTR) was cloned into psiCHECK2 plasmid (Promega, Madison, WI, USA). The PCR primer pairs with Xho I and Not I (NEB, Ipswich, USA) sites were 5′-CCGCTCGAGTGCGAGGA AGAAGGGTTT-3′ and 5′-GAATGCGGCCGCTGGGACTGTATCCACAGAA-3′. The constructors with rs1061429 C or A allele were named as psiCHECK2_rs1061429C and psiCHECK2_rs1061429A, respectively.

BGC823 gastric cancer cells were seeded in 24-well plates at a density of 2  × 10⁵ cells/well. As the cells reached 60–70% confluent, 300 ng of each pGL3-containing plasmids and 5ng pRL-SV40 (Promega, Madison, WI, USA) were co-transfected into the cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). For psiCHECK2-containing plasmids, 20 ng of plasmids was transfected into gastric cancer cells with or without 30pmol of miR-591 mimic (GenePharma, Shanghai, China). Gastric cancer cells were then collected 24 h after transfection. The activity of luciferase and renilla reporter gene activity were measured using GloMax 20/20 Luminometer (Promega, Madison, WI, USA).

Statistical analysis

The statistical analyses in our study were conducted by SPSS 23.0 (SPSS Inc., Chicago, IL, USA). The Hardy-Weinberg equilibrium (HWE) of C7 polymorphisms in controls were assessed by χ2 test. Differences of basic characteristics between cases and control subjects were evaluated by χ2 test. The association of C7 genotypes with the susceptibility to gastric cancer was evaluated by unconditional logistic regression with OR (95% CI) after adjusted by age, sex, smoking status and drinking status. The interaction between gene and environment was analyzed by epiR program in R platform (version 3.6.1). We defined smokers as they smoked more than 100 cigarettes in their lifetime. Drinkers were categorized as the individuals took at least 12 drinks on one occasion during the previous year according to international guide for monitoring alcohol consumption and harm of WHO. P value < 0.05 was regarded as statistical significance. To explore the associations of C7 expression level and the prognosis of gastric cancer, we performed survival analysis by Kaplan–Meier online program (https://kmplot.com/).

Subject characteristics

The distribution of select characteristics of all subjects were shown in Table 1. The study involved 471 gastric cancer patients and 471 healthy controls. There was no statistically significant difference in distribution of age and gender between cases and controls (age: P = 1.00; gender P = 0.44). Regard to the distribution of smoking and drinking status, there was no significant difference between cases and controls (smoking: P = 0.32; drinking P = 0.25).

Association of C7 gene polymorphisms with gastric cancer risk

After predicting potential regulatory functional SNPs, we found that C7 rs1376178 C > A variant enhanced the binding capability to transcription factor STAT1 and C7 rs1061429 C > A allele created a binding site with has-miR-591 in 3′ UTR. Genotype distributions of rs1376178 and rs1061429 polymorphisms in controls were conformed to Hardy-Weinberg equilibrium (HWE) (Table 2).

The relationship between each genetic variant and the susceptibility to gastric cancer was shown in Table 3. After adjusted by gender, age, drinking and smoking status, non-conditional logistic regression analysis showed that the distribution of C7 rs1376178 AA and CA genotypes was statistically different between cases and controls (P < 0.001). The individuals with C7 rs1376178 AA or CA genotype had a higher risk of GC with OR (95% CI) of 2.09 (1.43–3.03) and 1.88 (1.35–2.63), respectively. For C7 rs1061429 C > A polymorphism, AA genotype was associated with the elevated risk for developing gastric cancer (OR = 2.16, 95% CI [1.37–3.38]).

Stratification analysis of C7 variants with GC risk

To further analyze the effect of age, gender, smoking and drinking on the association of C7 variants (rs1376178 and rs1061429) with the risk of GC, we performed stratification analysis (Table 4). Gender stratification analysis showed that C7 rs1376178 AA genotype was associated with the elevated risk of gastric cancer among males (OR = 2.88, 95% CI [1.81–4.58], P < 0.001), but not among females (OR = 1.06, 95% CI [0.55–2.06], P > 0.05). When stratified by age, individuals with C7 rs1376178 AA had an increased risk of gastric cancer in both groups with OR (95% CI) of 2.19 (1.26–3.78) for the youngers and 2.07 (1.23–3.48) for the elders. Our data also showed that the risk of gastric cancer was associated with the rs1376178 AA regardless of smoking and drinking status (OR = 2.05, 95% CI [1.31­-3.21], P = 0.002 for nonsmokers; OR = 2.40, 95% CI [1.19–4.87], P = 0.015 for smokers; OR = 1.94, 95% CI [1.28–2.93], P = 0.002 for nondrinkers; OR = 3.55, 95% CI [1.37–9.17], P = 0.009 for drinkers).

Stratification analysis of C7 rs1061429 polymorphism was showed in Table 5. Our data suggested that individuals with rs1061429 AA had an increased risk of gastric cancer among youngers (OR = 2.84, 95% CI [1.39–5.80], P = 0.004) and non-smokers (OR = 2.79, 95% CI [1.63–4.77], P < 0.001), but not among elders (OR = 1.81, 95% CI [1.00–3.28], P = 0.050) and smokers (OR = 1.03, 95% CI [0.43–2.5], P = 0.947). The stratification analysis by gender or drinking status showed that rs1061429 AA genotype was contributed to the risk of gastric cancer regardless of gender and drinking status (OR = 1.94, 95% CI [1.13–3.33], P = 0.017 for males; OR = 2.87, 95% CI [1.26–6.54], P = 0.012 for females; OR = 1.92, 95% CI [1.18–3.13], P = 0.008 for nondrinkers; OR = 5.23, 95% CI [1.34–20.40], P = 0.017 for drinkers). Use R package epiR to build Logistic regression gene-environment interaction mode, we found that rs1061429 had no additive and multiplicative interaction with drinking or gender to affect the risk of gastric cancer (P > 0.05).

The expression of C7 in gastric cancer and its effect on prognosis and clinical-pathological characteristics of gastric cancer

Furthermore, we explore the potential function of gene C7 in gastric cancer, based on GEPIA data, we analyzed the differential expression of C7 in gastric cancer tissues and adjacent normal tissues and found that the level of C7 mRNA in gastric cancer tissues (n = 408) was significantly depressed when compared with that in adjacent normal tissues (n = 211) (Fig. 1A). To verify the result from GEPIA data, we further compared the expression of C7 in serum between normal individuals and gastric cancer patients and found a significant differential expression of C7 (P = 0.043) (Fig. 1B).

We also analyzed the impact of C7 expression on the prognosis of gastric cancer using Kaplan–Meier online program and demonstrated that higher expression of C7 was related with poor overall survival time (OS) and post-progression survival time (PPS) of gastric patients with HR (95% CI) of 1.29 (1.09–1.53) and 1.51 (1.19–1.91), respectively (Figs. 1C and 1D). We also evaluated the association of C7 expression level with TNM stage. Our data didn’t show any correlation between C7 expression and TNM stage (Fig. S1 ).

C7 genotypes and clinical-pathological characteristics

We used Kendall’s rank correlation tests to estimate the correlation between these two SNPs and clinicopathological characteristics. Our results showed that rs1061029 C > A variant has a positive correlation with lymph node metastasis, but the correlation strength was low (P = 0.022, Kendall’s Tau-b = 0.104). We didn’t find any correlation between rs1376178 C > A variant and TNM stage, tumor size, lymph node metastasis, distant metastasis (Table 6).

The effect of C7 polymorphisms on the regulatory activity of C7

To substantiate the association between the identified SNPs and GC risk, using the genotype and gene expression data of 324 normal gastric tissues in the GTEx to analyze, the eQTL data results showed that the mRNA expression of C7 was significantly related to the rs1061429 genotype (P < 0.001) (Fig. 2A). However, there is no eQTL data related to rs1376178 in GEPIA database.

We then measured the C7 protein level in serum of 70 individuals. For rs1061429 variant, the expression of C7 was significantly lower in individuals with AA genotype when compared with those with CC genotype (P = 0.029) (Fig. 2B). For rs1376178 polymorphism, our data didn’t show any effect on C7 expression (Fig. 2C).

Our luciferase reporter assay showed that pGL3 _rs1376178C exhibited 2.86-fold higher luciferase activity than pGL3_rs1376178A (P < 0.001) (Fig. 3A). We also found that psiCHECK2_rs1061429A had 1.34-fold increased luciferase activity in comparison to psiCHECK2_rs1061429Cwhen co-transfected with miR-591 (P = 0.0012) (Fig. 3B). There was no effect of rs1061429 polymorphism on the reporter gene activity without additional miR-591.