SYNJ2 Variant Rs9365723 is Associated with Colorectal Cancer Risk in Chinese Han Population

Abstract

Purpose

Colorectal cancer (CRC) is the third most common cancer and fourth leading cause of cancer mortality, and twin studies have shown that approximately 35% of the variation in susceptibility to CRC involves inherited genetic differences. We sought to investigate potential genetic associations between some single nucleotide polymorphisms (SNPs) and the risk of CRC in the Chinese Han population.

Methods

We conducted a case-control study including 269 cases and 309 controls. Sixteen SNPs associated with CRC risk were selected from previous genome-wide association studies and genotyped using Sequenom MassARRAY technology. Odds ratios and 95% confidence intervals (CIs) were calculated by unconditional logistic regression adjusting for age and gender.

Results

Using the chi-squared test we found that rs9365723 was associated with CRC risk (p = 0.012). With genetic model analysis, the genotype A/G-G/G (OR = 1.50; 95% CI 1.02-2.21; p = 0.038) of rs9365723 showed an increased risk of CRC in the dominant model. Furthermore, we found that rs9365723 was associated with an increased risk only for colon cancer but not rectal cancer (p = 0.009 and p = 0.414, respectively).

Conclusions

Our results, combined with previous studies, suggest that rs9365723, located on SYNJ2, is associated with the risk of CRC in a Chinese population. Thus, SYNJ2 may play a role in the development of CRC, especially colon cancer.

Keywords

Colorectal cancer Single nucleotide polymorphism (SNP)SYNJ2 Case-control study

Introduction

Colorectal cancer (CRC) is the third most common cancer and fourth leading cause of cancer mortality (1). It affects over one million people each year worldwide (2). During the past few decades, the incidence and mortality of CRC have been increasing rapidly in China: according to a 2007 investigation from 38 registered areas of China, CRC accounted for approximately 391,000 new cases and 186,800 deaths.

Although the recent development of new drugs and more advanced therapeutics have remarkably improved overall survival, and even if the 5-year survival rate is around 90% for patients who are diagnosed at stage I (3), the prognosis of patients with CRC at an advanced stage is still disappointing. CRC is a very complex disease in which environmental and genetic risk factors play a role. Unhealthy dietary habits and lifestyle, including high-fat and low-fiber diets, may contribute to the development of CRC. Genetic factors are also important: twin studies have shown that approximately 35% of the variation in susceptibility to CRC involves inherited genetic differences (4, 5).

Genome-wide association studies using large sets of cases and controls have proven to be an effective strategy to identify common single nucleotide polymorphisms (SNPs) associated with cancer risk (6). For the present study, we selected 16 SNPs from 15 chromosomal regions which were either associated with CRC in previous studies or were located in close proximity to previously reported risk SNPs: rs1912453 (1q23.3), rs11903757 (2q32.3), rs4591517 (3p24.3), rs367615 (5q21.3), rs647161 (5q31.1), rs2057314 (6q22.1), rs9365723 (6q25.3), rs39453 (7p15.3), rs12548021 (8p12), rs10114408 (9q22.32), rs1665650 (10q25.3), rs10774214 (12p13.32), rs3217901 (12p13.32), rs59336 (12q24.21), rs4939827 (18q21.1), and rs961253 (20p12.3) (7-8-9-10-11). Among these variants, rs4939827 and rs961253 have been researched in Chinese (12-13-14), but the results were inconsistent across studies. Moreover, few studies have reported the association between other SNPs and CRC susceptibility in Chinese. This case-control study focused on the risk of 16 SNPs in a Chinese population and the differences between our results and previous association studies in other populations.

Materials and methods

Study Subjects

The present study included 269 CRC cases and 309 controls. All patients were consecutively recruited from the First Affiliated Hospital of Xi'an Jiaotong University, China. The diagnosis of CRC was confirmed by histopathological examination. The control subjects were without a positive personal history of malignancy or chronic disease. Both patients and controls were Han Chinese. We excluded patients who underwent radiotherapy or chemotherapy as well as controls with chronic disease. All participants in the current study were not blood related.

Clinical Data and Demographic Information

We used a standard epidemiological questionnaire to collect personal data through in-person interviews including gender, age, occupation history, tobacco history and other lifestyle factors, as well as clinical and pathological investigations in a custom-designed questionnaire. After the study participants had signed informed consent forms, 5 mL of peripheral venous blood was collected in EDTA-coated tubes and stored at −80°C. The study was approved by the ethics committee of the First Affiliated Hospital of Xi'an Jiaotong University.

SNP Selection and Genotyping

All of the 16 SNPs we selected were previously reported to be most associated with CRC, with minor allele frequencies >5% in the HapMap Chinese Han Beijing population. DNA was extracted from whole-blood samples by means of a GoldMag-Mini whole blood genomic DNA purification kit (GoldMag Co. Ltd.). The extracted DNA was quantified using a NanoDrop 2000 spectrophotometer (Thermo Scientific). The Sequenom MassARRAY Assay Design 3.0 software was used to design primers for amplification and extension reactions (15, 16). Genotyping was performed by Sequenom MassARRAY RS1000 according to the manufacturers’ instructions.

Statistical Analysis

Statistical analyses were conducted with the SPSS 17.0 statistical package. All p values presented in this study are 2-sided, and a p value <0.05 was considered statistically significant. The differences in the characteristics of the case and control groups were compared using the chi-squared test (for categorical variables) and Student's t-test (for continuous variables). The Hardy-Weinberg equilibrium of all selected SNPs in controls was tested using the chi-squared test (17). The chi-squared test was also used to detect statistical differences in the distributions of genotypes between cases and controls. In addition, we compared allele frequencies in different tumor locations of CRC. The effects of the polymorphisms on the risk of CRC were expressed as odds ratios (ORs) with 95% confidence intervals (95% CIs), computed using unconditional logistic regression analysis with adjustments for age and sex (18).

SNP stats, a website software tool (http://bioinfo.iconcologia.net/SNPstats_web), was used to test the associations between all SNPs and the risk of CRC under different genetic models (codominant, dominant, recessive, overdominant and additive) (19). The most common control homozygote was used as a reference.

Results

This study included 268 CRC cases (101 women, 167 men; mean age 58.4 years) and 309 controls (113 women, 196 men; mean age 49.1 years). The clinical characteristics of cases and controls are summarized in Table I.

Table I

General characteristics of the study population

Variable	Cases, N (%)	Controls, N (%)	P value
Age (mean ± SD)	58.4 ± 11.67	49.1 ± 7.89	<0.001^#
Sex			0.782^*
Male	167 (62.1)	196 (63.4)
Female	101 (37.5)	113 (36.6)
Missing	1 (0.4)	-
Tumor site
Colon	159 (59.1)
Rectum	107 (39.8)
Others	3 (1.1)
Total	269	309

p values were calculated with Student's t-test.

p values were calculated with 2-sided chi-square test.

No significant difference was found between patients and controls in the distribution of sex (p = 0.782). The incidence of CRC was higher in men than women (62.1% and 37.5%, respectively; p<0.01). The case group included a higher proportion of colon cancer than rectum cancer (59.1% and 39.8%, respectively).

Table II shows the basic characteristics of all SNPs in the study population. The minor allele frequency of SNPs among cases and controls was consistent with the genotyping data for the Chinese Han Beijing population in the HapMap database. The SNPs were in Hardy-Weinberg equilibrium in controls (p>0.05), except for rs1077421, which was excluded from subsequent analyses. Our study established that rs9365723 was significantly associated with increased risk of CRC, with an OR of 1.34 (95% CI 1.06-1.70; p = 0.012). No significant associations were detected between CRC risk and the other SNPs.

Table II

Basic information of candidate SNPs and associations with colorectal cancer risk

SNP ID	Chromosome position	Base change	MAF (cases)	MAF (controls)	HWE P	OR	95% CI	P value
rs1912453	1q23.3	T/C	0.395	0.393	0.943	1.01	0.80-1.28	0.836
rs11903757	2q32.3	T/C	0.041	0.042	0.440	0.98	0.55-1.75	0.270
rs4591517	3p24.3	C/T	0.165	0.177	0.889	0.92	0.68-1.25	0.856
rs367615	5q21.3	C/T	0.447	0.432	0.895	1.06	0.84-1.34	0.854
rs647161	5q31.1	C/A	0.314	0.275	0.054	1.20	0.93-1.55	0.146
rs2057314	6q22.1	T/C	0.498	0.456	0.242	1.20	0.95-1.51	0.276
rs9365723	6q25.3	A/G	0.450	0.378	0.093	1.34	1.06-1.70	0.012^*
rs39453	7p15.3	T/C	0.354	0.338	0.463	1.07	0.84-1.37	0.834
rs12548021	8p12	A/G	0.217	0.201	0.580	1.10	0.83-1.47	0.480
rs10114408	9q22.32	A/T	0.185	0.180	0.252	1.03	0.77-1.40	0.711
rs1665650	10q25.3	G/A	0.327	0.304	0.663	1.12	0.87-1.43	0.624
rs10774214	12p13.32	C/T	0.336	0.320	0.015^#	1.07	0.84-1.37	0.112
rs3217901	12p13.32	G/A	0.457	0.500	0.078	0.84	0.67-1.06	0.263
rs59336	12q24.21	A/T	0.397	0.373	0.691	1.10	0.87-1.40	0.423
rs4939827	18q21.1	C/T	0.296	0.249	0.246	1.26	0.97-1.64	0.172
rs961253	20p12.3	C/A	0.105	0.100	0.575	1.06	0.72-1.55	0.720

SNP = single nucleotide polymorphism; HWE = Hardy-Weinberg equilibrium; OR = odds ratio; CI = confidence interval.

site with HWE p≤0.01 is excluded.

p value ≤0.05 indicates statistical significance.

We further assessed the association between SNPs and CRC risk using genetic models by unconditional logistic regression analysis with adjustments for age and gender. The results showed that the G allele of rs9365723 increased the CRC risk (Tab. III). In the dominant model, the GG and AG genotypes increased the CRC risk more than 1.5-fold (OR = 1.50; 95% CI 1.02-2.21; p = 0.0038) after adjusting for age and sex.

Table III

Genotypic model analysis of relationship between rs9365723 and colorectal cancer risk

Model	Genotype	Controls	Cases	OR (95% CI)^#	P value
	A/A	126 (40.9%)	77 (28.8%)	1.00
Codominant	A/G	131 (42.5%)	139 (52.1%)	1.62 (1.07-2.44)	0.071
	G/G	51 (16.6%)	51 (19.1%)	1.25 (0.74-2.13)
Dominant	A/A	126 (40.9%)	77 (28.8%)	1.00	0.038^*
	A/G-G/G	182 (59.1%)	190 (71.2%)	1.50 (1.02-2.21)
Recessive	A/A-A/G	257 (83.4%)	216 (80.9%)	1.00	0.820
	G/G	51 (16.6%)	51 (19.1%)	0.95 (0.59-1.52)

OR = odds ratio; CI = confidence interval.

adjusted for sex and age.

p value ≤0.05 indicates statistical significance.

Given the biological differences between the histological forms of CRC, we stratified by subtype. In this restricted analysis, the statistically significant association between rs9365723 and CRC was stronger among patients with colon cancer (OR = 1.43; 95% CI 1.09-1.88; p = 0.009) than among all patients with CRC (Tab. IV). The results of the analysis by genetic models are listed in Table V. In the dominant model, the GT and GG genotypes of rs9365723 were associated with a 1.70-fold increased colon cancer risk (OR = 1.70; 95% CI 1.06-2.73; p = 0.027).

Table IV

Basic information of candidate SNPs and associations with rectum and colon cancer

	Rectum				Colon
SNP ID	HWE (p value)	OR	95% CI	P value	HWE (p value)	OR	95% CI	P value
rs1912453	0.943	0.99	0.72-1.36	0.969	0.943	1.02	0.77-1.35	0.628
rs11903757	0.440	1.23	0.60-2.54	0.073	0.440	0.83	0.41-1.71	‐
rs4591517	0.889	0.73	0.47-1.14	0.303	0.889	1.08	0.76-1.53	0.907
rs367615	0.895	1.39	1.02-1.91	0.107	0.895	0.88	0.67-1.16	0.531
rs647161	0.054	1.16	0.83-1.64	0.591	0.054	1.23	0.92-1.65	0.149
rs2057314	0.242	1.19	0.87-1.63	0.222	0.242	1.21	0.92-1.58	0.195
rs9365723	0.093	1.22	0.89-1.67	0.414	0.093	1.43	1.09-1.88	0.009^*
rs39453	0.463	0.83	0.59-1.16	0.430	0.463	1.29	0.97-1.71	0.232
rs12548021	0.580	1.06	0.72-1.55	0.518	0.580	1.16	0.83-1.62	0.542
rs10114408	0.252	1.05	0.70-1.57	0.947	0.252	1.03	0.72-1.46	0.586
rs1665650	0.663	1.09	0.78-1.52	0.708	0.663	1.13	0.85-1.51	0.712
rs10774214	0.015^#	1.22	0.88-1.68	0.063	0.015^#	0.98	0.73-1.31	0.474
rs3217901	0.078	0.75	0.55-1.03	0.221	0.078	0.90	0.69-1.19	0.513
rs59336	0.691	1.13	0.82-1.56	0.382	0.691	1.07	0.81-1.42	0.703
rs4939827	0.246	1.29	0.91-1.81	0.346	0.246	1.25	0.92-1.69	0.324
rs961253	0.575	0.92	0.54-1.57	1.000	0.575	1.17	0.76-1.81	0.530

SNP = single nucleotide polymorphism; HWE = Hardy-Weinberg equilibrium; OR = odds ratio; CI = confidence interval.

site with HWE p≤0.01 is excluded.

p value ≤0.05 indicates statistical significance.

Table V

Genotypic model analysis of relationship between rs9365723 and colon cancer risk

Model	Genotype	Controls	Cases	OR (95% CI)^#	P value
Codominant	A/A	126 (40.9%)	42 (26.6%)	1.00	0.063
	A/G	131 (42.5%)	85 (53.8%)	1.82 (1.10-3.01)
	G/G	51 (16.6%)	31 (19.6%)	1.44 (0.76-2.71)
Dominant	A/A	126 (40.9%)	42 (26.6%)	1.00	0.027^*
	A/G-G/G	182 (59.1%)	116 (73.4%)	1.70 (1.06-2.73)
Recessive	A/A-A/G	257 (83.4%)	127 (80.4%)	1.00	0.990
	G/G	51 (16.6%)	31 (19.6%)	1.00 (0.58-1.75)

OR = odds ratio; CI = confidence interval.

adjusted for sex and age.

p value ≤0.05 indicates statistical significance.

Discussion

In this study, we investigated the association between 16 SNPs and the risk of CRC in a Chinese population. We found for the first time that rs9365723 was associated with a significantly increased risk of CRC. Stratified analyses indicated that rs9365723 was more likely to increase the risk of colon cancer.

The rs9365723 SNP is located in synaptojanin 2 (SYNJ2), an important member of the inositol polyphosphate phosphatase family (20). SYNJ2 removes the 5-position phosphate from phosphoinositides, thought to be involved in membrane trafficking and signal transduction pathways. Using a causative mutation in the SYNJ2 gene, Manji et al. identified SYNJ2 as a critical regulator of hair cell survival that is essential for maintaining normal hearing in mice (21). Recent studies demonstrated that SYNJ2 SNPs were associated with agreeableness and symptoms of depression in the elderly (22) and with cognitive abilities (23). Although the functions of the SYNJ2 gene were poorly understood, we discovered that SYNJ2 may provide a potential mechanism of colorectal tumorigenesis. Additionally, we observed that rs9365723 was associated with an increased risk of colon cancer but not rectal cancer, and it could therefore be hypothesized that SYNJ2 plays a more important role in the development of colon cancer than rectal cancer. Therefore, the biological functions of SYNJ2 are of great interest and need further investigation.

Our results indicated that rs4939827, from chromosome 18q21.1, did not increase the risk of CRC in Chinese. Previous evidence relates rs4939827 to the risk of CRC. For example, several genome-wide association studies showed that rs4939827 conferred an increased risk of CRC (24, 25). Several studies showed similar results in the Chinese population (12-13-14). However, rs4939827 was associated with a significantly decreased risk of CRC in Korean (26) and Croatian (27) populations. These conflicting results might be attributed to ethnic differences among current studies. Therefore, these findings need to be validated in further large and more ethnically diverse population-based studies.

We also found no correlation between rs961253 (located in proximity to bone morphogenetic protein [BMP] genes) and CRC risk. Associations between rs961253 and CRC risk have also been inconsistent across studies (13, 28-29-30). For example, several studies identified this association in Chinese (13, 28), southern Europeans (29) and African Americans (30), but Thean et al. failed to demonstrate the association in Singapore Chinese (31). It is currently believed that BMP signaling is related to CRC development through an increase in stem cell numbers near colorectal crypt bases, thus elevating the number of cells susceptible to tumor-causing mutations (32).

These conflicting results might be attributable to ethnic differences among the current studies and differences in the number of specimens. Moreover, certain dietary and lifestyle factors are undoubtedly major risk factors for CRC. Therefore, additional studies involving a larger number of cases are needed to assess the predictive value of rs961253 and rs4939827 for CRC. More importantly, functional study of the polymorphism in addition to epidemiological replication is crucial to clarify whether the association is causative.

In conclusion, to our knowledge this was the first attempt to provide powerful new evidence for the relationship between rs9365723 and susceptibility to CRC, especially colon cancer. Our findings suggest that SYNJ2 variants may play an important role in the development of CRC.

Footnotes

Acknowledgment

We are grateful to all patients and individuals in the study who made this work possible. We would also like to thank the clinicians and other hospital staff who contributed to the data collection for this study.

Financial support: This work is supported by the National 863 High-Technology Research and Development Program (No.2012AA02A519).

Conflict of interest: The authors have no conflicts of interest to report.

References

Jemal

Bray

Center

Ferlay

Ward

Forman

Global cancer statistics. CA Cancer J Clin 2011; 61 (2): 69–90

Tenesa

Dunlop

New insights into the aetiology of colorectal cancer from genome-wide association studies. Nat Rev Genet 2009; 10 (6): 353–358

Cui

Okada

Jang

. Common variant in 6q26-q27 is associated with distal colon cancer in an Asian population. Gut 2011; 60 (6): 799–805

Lichtenstein

Holm

Verkasalo

. Environmental and heritable factors in the causation of cancer—analyses of cohorts of twins from Sweden, Denmark, and Finland. N Engl J Med 2000; 343 (2): 78–85

Goel

Boland

Recent insights into the pathogenesis of colorectal cancer. Curr Opin Gastroenterol 2010; 26 (1): 47–52

Fletcher

Houlston

Architecture of inherited susceptibility to common cancer. Nat Rev Cancer 2010; 10 (5): 353–361

Jiao

Hsu

Berndt

. Genome-wide search for gene-gene interactions in colorectal cancer. PLoS ONE 2012; 7 (12): e52535

Fernandez-Rozadilla

Cazier

Tomlinson

et al; EPICOLON Consortium. A colorectal cancer genome-wide association study in a Spanish cohort identifies two variants associated with colorectal cancer risk at 1p33 and 8p12. BMC Genomics 2013; 14 (1): 55

Jia

Zhang

Matsuo

et al; Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO); Colon Cancer Family Registry (CCFR). Genome-wide association analyses in East Asians identify new susceptibility loci for colorectal cancer. Nat Genet 2013; 45 (2): 191–196

10.

Peters

Jiao

Schumacher

. Identification of genetic susceptibility loci for colorectal tumors in a genome-wide meta-analysis. Gastroenterology 2013; 144 (4): 799–807

11.

Houlston

Webb

Broderick

et al; COGENT Study; Colorectal Cancer Association Study Consortium; CoRGI Consortium. Meta-analysis of genome-wide association data identifies four new susceptibility loci for colorectal cancer. Nat Genet 2008; 40 (12): 1426–1435

12.

Song

Zhu

. A common SMAD7 variant is associated with risk of colorectal cancer: evidence from a case-control study and a meta-analysis. PLoS ONE 2012; 7 (3): e33318

13.

Xiong

. Risk of genome-wide association study-identified genetic variants for colorectal cancer in a Chinese population. Cancer Epidemiol Biomarkers Prev 2010; 19 (7): 1855–1861

14.

Choi

Lee

. Replication study of SNP associations for colorectal cancer in Hong Kong Chinese. Br J Cancer 2011; 104 (2): 369–375

15.

Gabriel

Ziaugra

Tabbaa

SNP genotyping using the Sequenom MassARRAY iPLEX platform. Curr Protoc Hum Genet 2009 Chapter 2: Unit 2.12.

16.

Thomas

Baker

Debiasi

. High-throughput oncogene mutation profiling in human cancer. Nat Genet 2007; 39 (3): 347–351

17.

Adamec

[Example of the use of the nonparametric test. Test X2 for comparison of 2 independent examples]. Cesk Zdrav 1964; 12613–619 [Article in Czech].

18.

Bland

Altman

Statistics notes. The odds ratio. BMJ 2000; 320 (7247): 1468

19.

Solé

Guinó

Valls

Iniesta

Moreno

SNPStats: a web tool for the analysis of association studies. Bioinformatics 2006; 22 (15): 1928–1929

20.

Nemoto

Arribas

Haffner

DeCamilli

Synaptojanin 2, a novel synaptojanin isoform with a distinct targeting domain and expression pattern. J Biol Chem 1997; 272 (49): 30817–30821

21.

Manji

Williams

Miller

. A mutation in synaptojanin 2 causes progressive hearing loss in the ENU-mutagenised mouse strain Mozart. PLoS ONE 2011; 6 (3): e17607

22.

Luciano

Lopez

de Moor

. Longevity candidate genes and their association with personality traits in the elderly. Am J Med Genet B Neuropsychiatr Genet 2012; 159B (2): 192–200

23.

Lopez

Harris

Luciano

. Evolutionary conserved longevity genes and human cognitive abilities in elderly cohorts. Eur J Hum Genet 2012; 20 (3): 341–347

24.

Broderick

Carvajal-Carmona

Pittman

et al; CORGI Consortium. A genome-wide association study shows that common alleles of SMAD7 influence colorectal cancer risk. Nat Genet 2007; 39 (11): 1315–1317

25.

Tenesa

Farrington

Prendergast

. Genome-wide association scan identifies a colorectal cancer susceptibility locus on 11q23 and replicates risk loci at 8q24 and 18q21. Nat Genet 2008; 40 (5): 631–637

26.

Hong

Park

Kim

. Colorectal cancer‐susceptibility single nucleotide polymorphisms in Korean population. J Gastroenterol Hepatol 2015; 30 (5): 849–857

27.

Kirac

Matošević

Augustin

. SMAD7 variant rs4939827 is associated with colorectal cancer risk in Croatian population. PLoS ONE 2013; 8 (9): e74042

28.

Zheng

Wang

Zhu

. The SNP rs961253 in 20p12.3 is associated with colorectal cancer risk: a case-control study and a meta-analysis of the published literature. PLoS ONE 2012; 7 (4): e34625

29.

Fernandez-Rozadilla

Palles

Carvajal-Carmona

et al; EPICOLON Consortium. BMP2/BMP4 colorectal cancer susceptibility loci in northern and southern European populations. Carcinogenesis 2013; 34 (2): 314–318

30.

Wilkens

Stram

. Generalizability and epidemiologic characterization of eleven colorectal cancer GWAS hits in multiple populations. Cancer Epidemiol Biomarkers Prev 2011; 20 (1): 70–81

31.

Thean

Teo

. Association of Caucasian-identified variants with colorectal cancer risk in Singapore Chinese. PLoS One 2012; 7 (8): e42407

32.

Zhang

Tong

W-G

. BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-β-catenin signaling. Nat Genet 2004; 36 (10): 1117–1121