ITGA1 polymorphisms and haplotypes are associated with gastric cancer risk in a Korean population

Abstract

AIM: To evaluate the association between the genetic polymorphisms and haplotypes of the ITGA1 gene and the risk of gastric cancer.

METHODS: The study subjects were 477 age- and sex-matched case-control pairs. Genotyping was performed for 15 single nucleotide polymorphisms (SNPs) in ITGA1. The associations between gastric cancer and these SNPs and haplotypes were analyzed with multivariate conditional logistic regression models. Multiple testing corrections were carried out following methodology for controlling the false discovery rate. Gene-based association tests were performed using the versatile gene-based association study (VEGAS) method.

RESULTS: In the codominant model, the ORs for SNPs rs2432143 (1.517; 95%CI: 1.144-2.011) and rs2447867 (1.258; 95%CI: 1.051-1.505) were statistically significant. In the dominant model, polymorphisms of rs1862610 and rs2447867 were found to be significant risk factors, with ORs of 1.337 (95%CI: 1.029-1.737) and 1.412 (95%CI: 1.061-1.881), respectively. In the recessive model, only the rs2432143 polymorphism was significant (OR = 1.559, 95%CI: 1.150-2.114). The C-C type of ITGA1 haplotype block 2 was a significant protective factor against gastric cancer in the both codominant model (OR = 0.602, 95%CI: 0.212-0.709, P = 0.021) and the dominant model (OR = 0.653, 95%CI: 0.483-0.884). The ITGA1 gene showed a significant gene-based association with gastric cancer in the VEGAS test. In the dominant model, the A-T type of ITGA1 haplotype block 2 was a significant risk factor (OR = 1.341, 95%CI: 1.034-1.741). SNP rs2447867 might be related to the severity of gastric epithelial injury due to inflammation and, thus, to the risk of developing gastric cancer.

CONCLUSION: ITGA1 gene SNPs rs1862610, rs24321

43, and rs2447867 and the ITGA1 haplotype block that includes SNPs rs1862610 and rs2432143 were significantly associated with gastric cancer.

Key Words: Integrin, ITGA1, Gastric cancer, Polymorphism, Haplotype

Core tip: There are few studies addressing the role of the integrin α 1 subunit in the development of gastric cancer. To the best of our knowledge, this study is the first to show that ITGA1 gene single nucleotide polymorphisms and haplotypes are associated with gastric cancer risk.

INTRODUCTION

Gastric cancer is the second most common cancer in South Korea and represents the second leading cause of cancer death for both men and women worldwide[1,2]. Approximately one million new cases of stomach cancer are estimated to have occurred (989000 cases, 7.8% of the total), currently making it the fourth most common malignancy in the world, following cancers of the lung, breast and colo-rectum[2]. Epidemiological studies have provided evidence that a high intake of salt and nitrite-rich foods and Helicobacter pylori (H. pylori) infection are associated with a high incidence of gastric cancer in South Korea[3-7].

The risk of developing gastric cancer is estimated to be increased 2-6 fold in patients with H. pylori infection[8]. The risk of gastric cancer among individuals infected with H. pylori is influenced by bacterial virulence. The most widely studied H. pylori virulence factors are the cag (cytotoxin-associated gene) antigens[9]. Compared to individuals infected with cagA-negative H. pylori strains, those infected with cagA-positive H. pylori strains show a higher risk of developing gastric cancer[10]. To introduce cagA into host cells, the cagL protein of H. pylori binds to integrins on the basolateral surface of gastric epithelial cells[11,12].

Integrins are members of a family of heterodimeric cell-surface proteins that mediate cell-matrix and cell-cell interactions. The 18 integrin α-subunits and 8 β-subunits together form at least 25 different integrins[13]. Integrins mediate signaling events that are essential for stable cell adhesion, spreading, migration, survival, proliferation and differentiation. Several integrins, including α1β1, bind to extracellular matrix proteins present in the basal membrane of mature vessels[14,15]. The tumor progression and metastasis of various cancers are associated with integrins[16,17].

The ITGA1 gene, located on chromosome 5q11.2, encodes the integrin α1 subunit, which is involved in the adhesion of gastric cancer cells to the peritoneum. The adhesion of integrin α1-positive gastric cancer cells to the extracellular matrix is a critical process in peritoneal dissemination[18,19]. There are few studies addressing the roles of integrins in the development of gastric cancer. An association with an increased risk of gastric cancer has only been reported for the ITGA2 C807T polymorphism in a Chinese population[20]. As the level of integrin α1β1 is up-regulated in association with inflammation of the gastrointestinal tract mucosa, which is the first step in gastric carcinogenesis[21], it is possible that the integrin α1 subunit plays an important role in gastric cancer development.

The purpose of this study was to evaluate the association between the genetic polymorphisms and haplotypes of the ITGA1 gene and the risk of gastric cancer.

MATERIALS AND METHODS

Study subjects

This subjects included in this study consisted of 477 newly diagnosed gastric cancer patients and an equal number of age- (within 3 years) and sex-matched controls. The diagnoses of the gastric cancer patients were all histologically confirmed at Chungbuk National University Hospital and Eulji University Hospital, which are located in a geographically central region of South Korea. Controls were selected from individuals receiving routine medical examinations in these hospitals, and individuals with a previous diagnosis of any type of cancer were excluded. Trained interviewers used a structured questionnaire including questions about demographic factors, smoking habits, alcohol consumption and dietary habits to interview all subjects who provided written informed consent. Peripheral blood samples were collected from all subjects. This study was approved by the institutional review boards of Chungbuk National University Hospital, South Korea (IRB No. 2011-09-071).

Selection of single nucleotide polymorphisms in ITGA1

At the International HapMap Project website (http://hapmap.ncbi.nlm.nih.gov/), tag SNPs were selected using a cut-off minimum minor allele frequency in the JPT population of 0.05 and pairwise tagging (r² = 1-0.8). SNPs that significantly deviated from Hardy-Weinberg equilibrium were discarded.

Genomic DNA was extracted from whole blood using the QuickGene-810 nucleic acid isolation system (Fujifilm, Tokyo, Japan) and the QuickGene DNA Whole Blood Kit (Kurabo, Osaka, Japan), in accordance with the manufacturer’s instructions. DNA was stored at 4 °C until use. SNP genotyping was performed using a GoldenGate Genotyping Assay with VeraCode technology (Illumina, San Diego, CA, United States). A custom GoldenGate assay was designed for the analysis of the selected SNPs in the ITGA1 gene. Those SNPs were then assessed for suitability for the GoldenGate genotyping platform, and the analysis was carried out on the validated SNPs. The average call rate was 99.2%. Genotyping was carried out by Macrogen (Seoul, South Korea).

Statistical analysis

The study power was calculated using the “case-control for discrete traits” mode in the Genetic Power Calculator[22]. The following parameters were applied: risk allele frequency -0.4, alpha error -0.01, and disease prevalence -0.1%. The power of a codominant model was 0.7768 when the heterozygous OR was set to 1.5. For a dominant model, when the OR for a genotype with one or 2 risk allele(s) was taken as 2, the power was 0.8821. When a value of 2 was input for the OR for a genotype with 2 risk allele(s), the power of a recessive model was 0.8182.

Testing for deviation from the HWP was performed for each SNP in both cases and in controls using Pearson’s χ² test. D values were measured using Lewontin’s method for all combinations of biallelic loci[23,24], and linkage disequilibrium blocks were structured using Haploview version 4.2 (Daly Lab at the Broad Institute Cambridge, MA, United States). Haplotype blocks were constructed and statistically compared between cases and controls with SNP Analyzer version 2.0 (ISTEC Inc., Goyang, South Korea).

Student’s t test was used to compare continuous variables between patients and control subjects. Associations between gastric cancer and the investigated SNPs and haplotypes were estimated via the OR and their corresponding 95%CI derived from multivariate conditional logistic regression models, after adjusting for potential confounding factors such as age, sex, smoking history, and alcohol intake. The genotypes of major homozygotes, heterozygotes and minor homozygotes were coded as 0, 1, and 2 in the codominant model, 0, 1 and 1 in the dominant model, and 0, 0 and 1 in the recessive model, respectively. Multiple testing corrections were carried out using Benjaminin and Hochberg’s methods for controlling the false discovery rate (FDR)[25]. A two-sided adjusted P value of < 0.05 was considered statistically significant. FDR Q values were calculated separately for the SNPs and haplotypes based on these numbers. Gene-based association tests were performed using the versatile gene-based association study (VEGAS) method[26]. For these statistical analyses, SAS version 9.2 (SAS Institute, Cary, NC, United States) was employed.

RESULTS

Patient characteristics are summarized in Table 1. No significant difference was observed between the distributions of the age, sex, and smoking and drinking habits of the cases and controls.

Table 1 Characteristics of the study subjects n (%).

Variables	Controls (n = 477)	Cases (n = 477)	OR (95%CI)
Age (yr) mean ± SD	57.8 ± 10.2	58.7 ± 9.9
Sex
Males	301 (63.1)	301 (63.1)
Females	176 (36.9)	176 (36.9)
Smoking status
Non-smokers	225 (47.6)	194 (41.0)	1.00 (reference)
Smokers	248 (52.4)	279 (59.0)	1.64 (0.95-2.84)
Alcohol intake status
Non-drinkers	194 (40.7)	189 (39.6)	1.00 (reference)
Drinkers	283 (59.3)	288 (60.4)	1.18 (0.71-1.76)

Table 2 lists and provides the frequencies of the 15 selected SNPs in the study subjects. None of the polymorphisms were significantly deviated from Hardy-Weinberg equilibrium. All the minor allele frequencies of the cases and controls were greater than 10%.

Table 2 Frequency of ITGA1 polymorphisms in cases and controls.

SNP	Chromosomal position	Amino acid change	Genotype case/control				Case		Control
							Frequency	HWE1	Frequency	HWE1
rs13188662	2686006	-	AA	AG	GG	N	0.280	0.573	0.276	0.597
			249/253	186/186	40/38	475/477
rs11740785	2707341	-	AA	AC	CC	N	0.241	0.866	0.229	0.259
			279/290	166/156	32/31	477/477
rs1820167	2713715	-	AA	AG	GG	N	0.435	0.806	0.420	0.904
			151/162	237/229	89/83	477/477
rs1862610	2722239	-	CC	AC	AA	N	0.369	0.861	0.387	0.484
			172/205	223/192	82/80	477/477
rs2432143	2725674	-	TT	TC	CC	N	0.104	0.671	0.146	0.658
			382/346	87/121	8/10	477/477
rs2447867	2751733	C/C	CC	TC	TT	N	0.490	0.742	0.430	0.769
			123/155	241/229	113/89	477/473
rs4865745	2770258	-	TT	TC	CC	N	0.270	0.892	0.268	0.124
			253/247	186/198	35/28	474/473
rs13163497	2773367	-	GG	AG	AA	N	0.110	0.409	0.108	0.515
			375/381	97/89	4/7	476/477
rs1904163	2780355	-	CC	TC	TT	N	0.298	0.196	0.272	0.698
			238/245	184/187	48/33	470/465
rs1466445	2789486	-	CC	TC	TT	N	0.460	0.783	0.455	0.696
			139/142	233/229	101/100	473/471
rs16880453	2789866	-	GG	GC	CC	N	0.466	0.914	0.465	0.424
			133/130	235/243	100/98	468/471
rs2452864	2796757	-	TT	TC	CC	N	0.367	0.874	0.369	0.368
			190/183	224/230	63/59	477/472
rs1275659	2828018	-	AA	AG	GG	N	0.257	0.185	0.278	0.864
			256/247	192/189	26/37	474/473
rs1871186	2828974	-	TT	TC	CC	N	0.221	0.723	0.213	0.674
			287/296	166/157	22/23	475/476
rs988574	2835169	E/G	TT	TC	CC	N	0.180	0.723	0.183	0.674
			319/309	141/155	15/9	475/473

¹P value for deviation from Hardy-Weinberg Equilibrium (HWE). SNP: Single nucleotide polymorphism.

The haplotype linkage disequilibrium blocks and haplotype frequencies for ITGA1 are shown in Figure 1. D values were measured using Lewontin’s method. Four block haplotypes were constructed using Haploview version 4.2. The common haplotypes (frequency > 10%) in each block accounted for 84.2%, 99.8%, 91.6% and 99.9% for the cases and 85.7%, 99.8%, 91.2% and 99.9% for the controls.

Figure 1 Haplotype linkage disequilibrium blocks and haplotype frequencies for ITGA1. A: Linkage disequilibrium (LD) blocks among ITGA1 polymorphisms. Black squares indicate a statistically significant allelic association between a pair of single nucleotide polymorphisms, as measured by the D statistic; darker gray indicate higher values of D; B: Haplotype frequencies of ITGA1 polymorphisms in cases (top) and controls (bottom).

The observed associations between the genetic polymorphisms in the ITGA1 gene and the risk of gastric cancer are shown in Table 3. In the codominant model, the OR of 1.517 obtained for SNP rs2432143 (95%CI: 1.144-2.011; P = 0.003; FDR Q = 0.045) was statistically significant, even after controlling the FDR, and that for rs2447867, of 1.258 (95%CI: 1.051-1.505; P = 0.012; FDR Q = 0.090), was marginally significant. In the dominant model, the rs1862610 and rs2447867 polymorphisms were not statistically significant risk factors for gastric cancer, displaying ORs of 1.337 (95%CI: 1.029-1.737; P = 0.029; FDR Q = 0.217) and 1.412 (95%CI: 1.061-1.881; P = 0.018; FDR Q = 0.217), respectively. Only the rs2432143 polymorphism was marginally significant in the recessive model, exhibiting an OR of 1.559 (95%CI: 1.150-2.114; P = 0.004; FDR Q = 0.060).

Table 3 Association between ITGA1 polymorphisms and gastric cancer in a case-control study of a Korean population.

SNP	Chromosomal position	Codominant			Dominant			Recessive
		OR (95%CI)	P value1	Q2	OR (95%CI)	P value1	Q2	OR (95%CI)	P value1	Q2
rs13188662	2686006	1.040 (0.840-1.281)	0.161	0.483	1.060 (0.811-1.379)	0.689	0.866	1.060 (0.660-1.690)	0.811	0.963
rs11740785	2707341	1.069 (0.869-1.313)	0.528	0.965	1.106 (0.848-1.442)	0.457	0.866	1.032 (0.630-1.692)	0.899	0.963
rs1820167	2713715	1.066 (0.884-1.286)	0.503	0.964	1.115 (0.846-1.468)	0.440	0.866	1.043 (0.751-1.447)	0.801	0.963
rs1862610	2722239	1.151 (0.965-1.372)	0.118	0.483	1.337 (1.029-1.737)	0.029	0.217	1.029 (0.740-1.429)	0.866	0.963
rs2432143	2725674	1.517 (1.144-2.011)	0.003	0.045	1.800 (0.603-5.371)	0.292	0.883	1.559 (1.150-2.114)	0.004	0.060
rs2447867	2751733	1.258 (1.051-1.505)	0.012	0.090	1.412 (1.061-1.881)	0.018	0.217	1.303 (0.966-1.756)	0.083	0.415
rs4865745	2770258	1.016 (0.829-1.246)	0.875	0.965	0.967 (0.750-1.247)	0.795	0.863	1.269 (0.759-2.122)	0.363	0.927
rs13163497	2773367	1.021 (0.768-1.357)	0.884	0.965	1.064 (0.781-1.449)	0.693	0.866	0.571 (0.167-1.952)	0.371	0.927
rs1904163	2780355	1.157 (0.943-1.420)	0.161	0.483	1.104 (0.849-1.436)	0.461	0.866	1.593 (0.984-2.577)	0.058	0.415
rs1466445	2789486	1.013 (0.845-1.213)	0.890	0.965	1.032 (0.778-1.368)	0.829	0.883	1.000 (0.736-1.358)	1.000	1.000
rs16880453	2789866	1.000 (0.832-1.201)	1.000	1.000	0.979 (0.734-1.305)	0.883	0.883	1.025 (0.752-1.398)	0.874	9.632
rs2452864	2796757	0.986 (0.816-1.191)	0.885	0.965	0.947 (0.728-1.233)	0.687	0.883	1.056 (0.728-1.532)	0.775	9.632
rs1275659	2828018	1.136 (0.919-1.404)	0.237	0.592	1.522 (0.899-2.575)	0.117	0.585	1.095 (0.841-1.427)	0.500	9.632
rs1871186	2828974	1.043 (0.841-1.293)	0.701	0.965	1.072 (0.828-1.388)	0.597	0.866	0.957 (0.533-1.716)	0.881	9.632
rs988574	2835169	0.985 (0.772-1.256)	0.901	0.965	0.927 (0.707-1.215)	0.581	0.866	1.667 (0.729-3.808)	0.225	0.843
VEGAS statistics (P)		23.986 (0.105)			16.823 (0.364)			18.732 (0.260)

¹P values for logistic analysis of three alternative models (codominant, dominant and recessive); ²False discovery rate Q value. When the lower P value generated by the dominant and recessive models was applied for every single nucleotide polymorphism (SNP), the value of the versatile gene-based association study (VEGAS) statistic was 29.622 (P = 0.037).

When the P values for the minor alleles of the codominant, dominant and recessive models were subjected to the VEGAS test, no significant gene-based associations were found. However, when the lower P value generated by the dominant and recessive models was input for every SNP, the value of the test statistic was 29.622, which was statistically significant (P = 0.037).

Four haplotype blocks were constructed using SNP Analyzer version 2.0. These blocks were evaluated for an association with the risk of gastric cancer (Table 4). The C-C type of ITGA1 haplotype block 2 was marginally significant in the codominant model (OR = 0.602, 95%CI: 0.212-0.709; P = 0.021; FDR Q = 0.063) and was a significant protective factor against gastric cancer in the dominant model (OR = 0.653, 95%CI: 0483-0.884; P = 0.006; FDR Q = 0.018). In the dominant model, the A-T type of ITGA1 haplotype block 2 was a significant risk factor (OR = 1.341, 95%CI: 1.034-1.741; P = 0.027; FDR Q = 0.045). No haplotype block was found to be significant in the recessive model.

Table 4 Association between ITGA1 haplotypes and gastric cancer.

Haplotypes		Codominant			Dominant			Recessive
		OR (95%CI)	P value1	Q2	OR (95%CI)	P value1	Q2	OR (95%CI)	P value1	Q2
ITGA1	AAA	0.771 (0.510-1.165)	0.414	0.973	0.860 (0.666-1.112)	0.250	0.750	0.819 (0.555-1.210)	0.316	0.913
Haplotype	GAG	1.039 (0.643-1.678)	0.973	0.973	1.030 (0.799-1.328)	0.819	0.819	1.026 (0.644-1.636)	0.913	0.913
block 1	ACA	0.992 (0.559-1.760)	0.768	0.973	1.088 (0.839-1.410)	0.525	0.787	0.957 (0.544-1.683)	0.879	0.913
ITGA1	CT	0.982 (0.688-1.407)	0.640	0.640	1.072 (0.800-1.437)	0.641	0.641	0.911 (0.679-1.223)	0.536	0.536
Haplotype	AT	1.316 (0.686-1.407)	0.086	0.129	1.341 (1.034-1.741)	0.027	0.045	1.121 (0.784-1.603)	0.532	0.536
block 2	CC	0.602 (0.212-0.709)	0.021	0.063	0.653 (0.483-0.884)	0.006	0.018	0.661 (0.233-1.872)	0.433	0.536
ITGA1	CCT	1.023 (0.707-1.480)	0.677	0.794	0.934 (0.705-1.236)	0.631	0.916	0.819 (0.555-1.210)	0.316	0.913
Haplotype	TGC	0.973 (0.641-1.475)	0.314	0.794	0.986 (0.761-1.278)	0.916	0.916	1.026 (0.644-1.636)	0.913	0.913
block 3	TGT	1.418 (0.446-4.507)	0.794	0.794	1.084 (0.782-1.505)	0.627	0.916	0.957 (0.544-1.683)	0.879	0.913
ITGA1	TA	0.938 (0.641-1.370)	0.907	0.907	0.928 (0.658-1.310)	0.671	0.671	0.997 (0.765-1.299)	0.981	0.981
Haplotype	CA	0.983 (0.536-1.803)	0.803	0.907	1.079 (0.832-1.400)	0.567	0.671	0.952 (0.523-1.733)	0.873	0.981
block 4	TG	1.619 (0.698-3.756)	0.320	0.907	0.925 (0.708-1.209)	0.569	0.671	1.685 (0.730-3.888)	0.217	0.981

¹P values for logistic analysis of three alternative models (codominant, dominant and recessive). The P value for haplotype associations were calculated using single nucleotide polymorphisms Analyzer™ 2.0 software; ²False discovery rate Q value.

DISCUSSION

The present study focused on the association of genetic polymorphisms and haplotypes of the ITGA1 gene with gastric cancer risk. It has been suggested that the integrin α1 subunit could be involved in gastric cancer carcinogenesis. Integrins on gastric epithelial cells have been reported to serve as a portal for the entry of H. pylori cagA[11]. Additionally, the integrin α1 subunit is involved in the adhesion and dissemination of gastric cancer cells to the peritoneum[18], and an ITGA2 polymorphism has been reported to be associated with an increase in the risk of gastric cancer[20]. However, to our knowledge, no previous study has examined the association between ITGA1 polymorphisms and the risk of gastric cancer.

The SNPs rs1862610, rs2432143 and rs2447867 were significantly associated with an increase in the risk of gastric cancer. After controlling the FDR, only SNP rs2432143 in the codominant model was statistically significant. In a gene-based association test, the ITGA1 gene was found to be significantly associated with gastric cancer.

The C-C type of ITGA1 haplotype block 2, which includes rs1862610 and rs2432143 in intron 1 of the ITGA1 gene, was found to be a significant protective factor and the A-T type to be a risk factor for gastric cancer. This statistical significance was maintained after controlling the FDR. However, the precise molecular mechanism related to these SNPs is not clear. Based on SNP function prediction using computational methods, SNPs rs1862610 and rs2432143 are not predicted to be involved in any structural or functional changes in the integrin α1 subunit. However, we cannot rule out the possibility that these SNPs are either associated with the stability of ITGA1 mRNA, or in linkage disequilibrium with an as yet unknown functional polymorphism affecting the expression or function of the integrin α1 subunit.

We used public databases of SNPs related to gastric cancer and assessed the potential functions of selected SNPs with SNP function prediction software. Among the 15 selected SNPs, only two were located in exons, and one was non-synonymous. The potential function was not predicted for any of these SNPs, except for rs2447867, which was predicted to be an exonic splicing enhancer (ESE). ESEs are clinically significant because synonymous point mutations in ESEs that were previously thought to be silent mutations can lead to exon skipping and the production of a non-functional protein. As loss of integrin α1β1 has been observed in some other malignancies[27], non-functional integrin α1β1 could be associated with gastric cancer.

The increased expression of integrin molecules by epithelial cells during inflammation of the underlying lamina propria is probably an adaptive response to prevent extensive epithelial cell sloughing caused by inflammatory mediators. Loss of epithelial integrity due to a decrease in the function of integrin results in more severe injury of the epithelium[21]. At these sites of tissue injury, bone marrow-derived cells are recruited, and these cells can be a potential source of malignancy[28]. Because chronic infection with H. pylori also induces repopulation of the stomach with bone marrow-derived cells, there is a possibility that a non-functional integrin α1 subunit and H. pylori infection would have a synergistic effect in increasing the risk of gastric cancer. The major limitation of the present study is that we did not test for the presence of antibodies against H. pylori and the cagA antigen in the sera of the case and control subjects.

The OR obtained for SNPs rs1862610, rs2432143, and rs2447867 were all below 1.6, while the OR for the ITGA2 C807T polymorphism in relation to gastric cancer in a Chinese population is 1.57[20]. These relatively small values can be explained by the promiscuity and redundancy of integrins: one integrin can bind several different ligands, and many different integrins can bind to the same ligand[29]. Therefore, if an integrin is not functioning, other integrins can compensate for at least some of its function.

In conclusion, the ITGA1 gene SNPs rs2432143 and rs2447867 and the ITGA1 haplotype block that includes SNP rs2432143 are significantly associated with gastric cancer risk.

COMMENTS

Background

Integrins mediate signaling events that are essential for stable cell adhesion, cell spreading, migration, survival, proliferation and differentiation. Several integrins, including α1β1, bind to extracellular matrix proteins present in the basal membranes of mature vessels. Tumor progression and the metastasis of various cancers are associated with integrins. The ITGA1 gene, located on chromosome 5q11.2, encodes the integrin α1 subunit, which is involved in the adhesion of gastric cancer cells to the peritoneum. Adhesion of integrin α1-positive gastric cancer cells to the extracellular matrix is a critical process in peritoneal dissemination. As integrin α1β1 is up-regulated during inflammation in the gastrointestinal tract mucosa, which is the first step in gastric carcinogenesis, it is possible that the integrin α1 subunit plays an important role in the development of gastric cancer. It has been suggested that the integrin α1 subunit could be involved in gastric cancer carcinogenesis. Integrins on gastric epithelial cells have been reported to serve as a portal for the entry of Helicobacter pylori (H. pylori) cagA. As integrin α1β1 is up-regulated during inflammation in the gastrointestinal tract mucosa, which is the first step in the gastric carcinogenesis, it is possible that the integrin α1 subunit plays an important role in the development of gastric cancer.

Research frontiers

There are few studies addressing the role of integrins in the development of gastric cancer. An association with an increased risk of gastric cancer has only been reported previously for the ITGA2 C807T polymorphism in a Chinese population. No earlier study has focused on the association of ITGA1 gene single nucleotide polymorphisms (SNPs) and haplotypes with gastric cancer risk.

Innovations and breakthroughs

To the best of the authors’ knowledge, this present study is the first to suggest a significant association of the genetic polymorphisms and haplotypes of ITGA1 gene with an increased gastric cancer risk.

Applications

Integrins on gastric epithelial cells have been reported to serve as a portal of entry for H. pylori cagA, and loss of epithelial integrity due to a decrease in the function of integrins results in more severe injury of the epithelium. Studies are needed addressing the interaction of non-functional integrin α1 subunit and H. pylori infection in increasing the risk of gastric cancer.

Peer review

This paper is focused on the ITGA1 polymorphisms and haplotypes, and gastric cancer risk in a Korean population. The results showed the SNPs rs1862610, rs2432143, and rs2447867, and the ITGA1 haplotype block which includes SNPs rs1862610 and rs2432143 were significantly associated with gastric cancer. It is interesting.