2021 Volume 96 Issue 4 Pages 187-191
Colorectal cancer (CRC), which includes the development of cancer from the colon or rectum, is one of the highly prevalent cancers in the populations of Jammu and Kashmir (J&K) in India. However, case–control genetic association studies on CRC are lacking in this population. Various genome-wide association studies have previously shown that single-nucleotide polymorphisms (SNPs) of the AT-rich interaction domain 5B (ARID5B) gene located on chromosome 10q21.2 contribute substantially to the development of colorectal cancer. The association between ARID5B and CRC risk in north Indian population groups is still unknown. To understand the role of ARID5B SNPs in CRC in the population of J&K, we designed a case–control study to investigate the association of the cancer susceptibility variant rs10740055 of ARID5B with CRC in the population of J&K. The study included 180 cases and 390 healthy controls. Genotyping of the rs10740055 variant was performed by RT-PCR using the TaqMan assay technique. Hardy–Weinberg equilibrium of the variant was assessed using the chi-squared test. The allele- and genotype-specific risks were estimated by odds ratios (ORs) with 95% confidence intervals (CIs). The rs10740055 variant showed a higher risk for colorectal cancer with an OR of 3.35 (1.99–5.65 at 95% CI) and P = 0.000005 corrected for age, gender, ethnicity, BMI, alcohol intake and smoking. Our results indicate that the A allele of rs10740055 imparts risk to the population and also that a larger sample size is needed for further statistical validation. The association of other variants in other ARID family genes should also be tested as their role cannot be ruled out.
Colorectal cancer (CRC) is the third most common malignancy in the world (Bray et al., 2018), with diverse malignant conditions based on its fundamental biology, assimilated genomic changes and clinical response to targeted therapy (Nguyen and Duong, 2018). In 2018, there were 900,000 reported deaths due to CRC and around 1.8 million new cases globally (Bray et al., 2018). The estimated number of CRC cases worldwide is 484,224 males and 396,568 females (Nguyen and Duong, 2018). The incidence of CRC in India has increased steadily since the middle of the last century, and 30 per 1,000,000 adults are diagnosed with CRC annually in India (Haggar and Boushey, 2009). The estimated death rate of males and females due to CRC is 1.16% and 0.53%, respectively (Sharma, 2020). The Union Territory of Jammu and Kashmir (J&K) is located under the foothills of the Himalayas and has long been considered as a cancer hotspot. Based on recent hospital-based data, the incidence rate of CRC in the Kashmir region is 16.8%, second only to stomach cancer, which has an incidence rate of 25.2% (Qurieshi et al., 2016). In the Jammu region, we found a similar trend, with CRC being the second most common cancer after stomach cancer (unpublished hospital-based data).
The predominant population groups of J&K are endogamous (Bhasin and Khanna, 1992; Sharma et al., 2017) and diverse populations (Sharma et al., 2018). This factor has been recognized as a major contributing factor in the development of CRC, which may thus be hereditary. The pathogenesis of CRC is known to result from the interplay of multiple environmental and heredity factors. Various studies have reported that the high incidence of gastric cancer, esophageal cancer and CRC in the Kashmiri population is due to the presence of certain chemicals, such as nitroso compounds, amines and nitrates, in their food items (Mir and Dar, 2009; Qadir et al., 2020).
Besides the environmental factors, germline and somatic variation play a key role in both susceptibility to, and the outcome of, CRC. Genome-wide association studies (GWASs) have added greatly to our understanding of the genetic predisposition in the development of CRC (Shaik et al., 2015). Single-nucleotide polymorphism (SNP)-based studies have further assisted in the understanding of cancer susceptibility and the response to personalized treatment (Deng et al., 2017). Two of the GWASs have associated the 10q21.2 locus with leukemia in the Caucasian population (Papaemmanuil et al., 2009; Treviño et al., 2009). This locus harbors the AT-rich interaction domain 5B (ARID5B) gene, which belongs to the ARID family of genes and encodes a transcription factor that plays a pivotal role in cell growth and expansion. Studies in Taiwanese and Yemeni populations have reported a significant association between the SNP variant rs10740055 of ARID5B and lymphoblastic leukemia (Lin et al., 2014; Al-Absi et al., 2017). To the best of our knowledge, there are limited reports on the genetic predisposition to CRC in North Indian populations (Sharma et al., 2021a, 2021b). The present study aimed to explore the association of variant rs10740055 of ARID5B with the risk of CRC in the J&K region of North India.
We have previously reported a significant association between the rs10740055 variant and an increased risk of developing leukemia in the J&K population (Bhat et al., 2019a). Nevertheless, the relationship between this variant and susceptibility to CRC still remains unexplored in the study population. Therefore, we tried to determine the role of rs10740055 in CRC in the population of the J&K region.
The clinicopathological details of cases and controls are given in Table 1. Locus and genotyping information of the rs10740055 variant are presented in Table 2.
| Characteristic | Cases (n = 180) | Controls (n = 390) | P-value |
|---|---|---|---|
| Age (± SD), years | 59.21 (± 12.7) | 58.36 (± 13.11) | < 0.05 |
| BMI (± SD), kg/m2 | 20.75 (± 0.869) | 24.21 (± 3.246) | 0.0001 |
| Gender (%) | Male = 110 (61%) | Male = 270 (69%) | |
| Female = 70 (39%) | Female = 120 (31%) | ||
| Alcohol habits | Non-drinker = 52 | Non-drinker = 179 | |
| Drinker = 128 | Drinker = 211 | ||
| Smoking habits | Smoker = 106 | Smoker = 82 | |
| Non-smoker = 74 | Non-smoker = 308 |
| Gene | SNP | Chromosome | Position | Location | Major/minor allele | HWE in controls |
|---|---|---|---|---|---|---|
| ARID5B | rs10740055 | 10 | 10q21.2 | Intron 3 | Major (risk) | 0.851 |
The allelic frequency distribution between cases and controls is summarized in Table 3. The observed frequency of allele “A” was 0.47 in cases and 0.40 in controls. The unadjusted allelic odds ratio (OR) of 1.3 (95% confidence interval (CI) 1.02–1.69, P = 0.01) suggests a significant association between rs10740055 and colorectal cancer with “A” being the risk allele. Under the dominant model (AA + AC vs. CC), the estimated risk of the heterozygous genotype after adjustments for age, gender, ethnicity, BMI, alcohol consumption and smoking was also statistically significant (OR = 3.35 [95% CI 1.99–5.65], P = 0.000005) (Table 3). The variant rs10740055 followed the Hardy–Weinberg equilibrium (HWE) (P-value = 0.851; Table 2) in the studied population. The power of this study was 94.6%.
| Gene/SNP | Allele frequency in cases | Allele frequency in control | Allele OR at 95% CI | P-value | Dominant OR* | P-value* |
|---|---|---|---|---|---|---|
| ARID5B | C = 53.1 | C = 59.7 | 1.3 (1.02–1.69) | 0.01 | 3.35 (1.99–5.65) | 0.000005 |
| rs10740055 | A = 46.9 | A = 40.3 |
To unravel the genetic predisposition of CRC in the population of J&K, we studied the rs10740055 variant of the ARID5B gene in 180 cases and 390 controls. This variant was recently found to be associated with an increased risk of CRC in some GWASs (Wang et al., 2013; Urayama et al., 2018). Based on the existing data and our own finding that rs10740055 increases leukemia risk (Bhat et al., 2019a), we hypothesized that this variant might also increase CRC risk in the J&K population. Our assumption was further based on a recent report showing a significant association between an ARID1A nucleotide variant and gastric cancer in the population of the Kashmir region (Bhandari et al., 2016). The ARID5B variant shows significant association with colorectal cancer, and the unadjusted OR observed was 1.3 (1.02–1.69 at 95% CI), whereas, when adjusted for confounding factors (age, gender, ethnicity, BMI, alcohol intake and smoking), risk increases with the OR observed as 3.35 (95% CI 1.99–5.65), suggesting that variant rs10740055 is contributing to the risk of colorectal cancer. ARID5B helps in the induction of mitochondrial membrane potential, electron transport chain-encoding genes and oxidative metabolism (Cichocki et al., 2018). It also plays an important role in the development of B-cell progenitors (Leong et al., 2017). The AT-rich interaction domain (ARID) family of proteins to which ARID5B belongs has been found to be involved in the regulation of transcription, cell differentiation, cell expansion and cell development (Patsialou et al., 2005). Previous work suggests that ARID family proteins are closely involved in cancer-related signaling pathways, extremely transformed or differentially expressed in tumor tissues, and act as a prognostic tool for cancer prediction or therapeutic consequence (Lin et al., 2014). ARID5B also reportedly regulates transcription of androgen receptor genes through histone methylation in the promoter region (Yamakawa et al., 2018).
Our in silico analyses of the rs10740055 variant with the online ESE finder (Cartegni et al., 2003) and Snipa (Arnold et al., 2015) tools predicted a change in the splicing factor binding sites of exonic splicing enhancers (ESE) intronic site and SF2/ASF (IgM-BRCA1), suggesting that the locus plays an important role in epigenetic regulation (Supplementary Fig. S1). Besides this, the rs10740055 variant showed a direct regulatory effect (Supplementary Fig. S2); nucleotide change in any of the exonic, intronic and UTR regions might affect the neighboring variants and interrupt the functioning of the DNA segment. The difference in SRp40 binding sites is small, but could potentially affect the folding pattern, interactions and function of the protein. The 4-bp upstream migration of the binding region may affect the protein’s covalent and non-covalent bond interaction and its functional efficacy, thus hampering its activity. The LD plot in Supplementary Fig. S2 identifies variants neighboring rs10740055, which may also be linked to disease development. However, in silico data cannot be considered as conclusive evidence to support any biological function of an SNP, and experimental evidence is necessary for validation.
To the best of our knowledge, no study has yet illuminated the role of ARID5B gene polymorphism(s) in colorectal cancer. This preliminary study therefore highlights a possible correlation between the ARID5B rs10740055 polymorphism and susceptibility to colorectal cancer.
In conclusion, colorectal cancer risk is associated with several factors such as genetic, environmental and lifestyle. Our findings suggest that variant rs10740055 on chromosome 10 in intron 3 of the ARID5B gene contributes to the increased risk of colorectal cancer in the J&K population, and particularly in older individuals with the habits of alcohol consumption and smoking. Even though our data showed a highly significant association between CRC and rs10740055 polymorphism, the small sample size and genotyping of only one gene of the ARID family do limit the scope of this study. Therefore, future studies involving more genes and single-nucleotide variants of the ARID gene family, and with an increased sample size, are necessary to understand better the role of these genes in the initiation and progression of CRC in the J&K population. Such studies should provide new targets for cancer therapy.
This study was approved by the Institutional Ethical Review Board (IERB) of Shri Mata Vaishno Devi University. All experimental procedures were conducted in accordance with the guidelines and regulations set by the IERB.
Sample collectionA total of 570 samples (180 CRC cases and 390 healthy controls) were recruited in this study. Samples of all cases were obtained from hospitals and clinics in J&K. Details of each participant were recorded in a pre-designed proforma, and written informed consent was obtained from each participant before the study was conducted.
Blood samplesA blood sample (2–3 ml of venous blood) was collected in a BD Vacutainer EDTA tube from cases and controls after prior consent. Details of both cases and controls are given in Table 1.
DNA extractionGenomic DNA was isolated from the venous blood samples using a Qiagen FlexiGene DNA Kit (catalog no. 51206).
GenotypingVariant rs10740055 of ARID5B was genotyped using a TaqMan assay, following our previously published protocol (Bhat et al., 2019b; Verma et al., 2020). The quality of genotyping was checked by repeating genotyping in 100 samples, and the concordance rate was found to be 100%.
Statistical analysisStatistical analysis was mainly performed using the Statistical Package for the Social Sciences (SPSS version 20; SPSS). A comparative analysis of clinical parameters was performed. A chi-squared test was performed, and genotype frequencies were tested for total HWE. The logistic regression model was used to estimate the OR at a 95% CI. Statistical significance was considered when P ≤ 0.05. P.S. software was used to calculate the power of the study. The effect of the variant on splicing was analyzed using the web tools Human Splicing Finder (HSF) 3.1 and ESE finder (Cartegni et al., 2003), and Snipa v3.3 (Arnold et al., 2015) was used to evaluate the effect of the sentinel variant on neighboring SNPs.
B. S. and R. K. acknowledge the Department of Science and Technology, Government of India for a grant (DST/SSTP/J&K/459) and the Indian Council of Medical Research (5/10/15/CAR-SMVDU/2018-RBMCH) for financial support. We thank Dr. Varun Sharma for critical suggestions.
