Abstract
Background:
Leukemia is a heterogeneous disorder, characterized by elevated proliferation of white blood cells. Various genetic studies have assessed the contributory roles of several single nucleotide polymorphisms with the development of leukemia. The role of genetic variation in the ARID5B and IKZF1 genes has previously been identified in various population groups; however, the role of these variants in the north Indian populations of Jammu and Kashmir is unknown.
Aim:
In this study, we explored the association of the newly identified genetic variants, rs10740055 of ARID5B and rs6964823 of IKZF1, with leukemic patients from Jammu and Kashmir of northern India.
Methods:
The variants were genotyped using TaqMan allele discrimination assays for 616 individuals (210 leukemic cases and 406 healthy controls). The association of each SNP with the disease was evaluated using logistic regression.
Results:
It was observed that the variants rs6964823 (IKZF1) and rs10740055 (ARID5B) showed significant associations with odds ratio (OR) and p-values of 1.5 (1.0-2.3 at 95% confidence interval [CI]) and 0.04; and 2.5 (1.5-4.1 at 95% CI) and 0.0002, respectively. We also evaluated the cumulative effect for both the variants by combining the risk genotypes and obtained and OR of 4.9.
Discussion:
It was found that the variants rs10740055 of ARID5B and rs6964823 of IKZF1 act individually and additively as risk factors in the development of leukemia in the populations of Jammu and Kashmir in Northern India.
Introduction
Leukemia is the group of heterogeneous malignancies of immature and dysfunctional hematopoietic progenitors that proliferate and accumulate in the bone marrow (Zahran et al., 2016). Disruption means disturbance in cellular processes (transcription, DNA replication), proliferation (increased number of cells), and differentiation (process by which less specialized cells become more specialized cell types). These changes alter the hematopoietic stem cells causing leukemia. (Agathocleous and Harris, 2013). In the United States, leukemia is ranked eighth for being the most commonly diagnosed cancer with ∼60,300 new cases and 24,370 deaths in 2018 (Siegel et al., 2018). According to the Population Based Cancer Registries, in India, males are mostly affected compared with females with a ratio of 1:2 (Bashar and Thakur, 2017). Incidence of leukemia in Jammu and Kashmir has shown a rapid increase from the previous decades (Wani et al., 2014). The population of northern part of Jammu and Kashmir State practice endogamy, thus preserving the gene pools that result in the increase of homozygosity. This factor has been documented as an inherited genetic factor that can contribute to the etiology of leukemia (Brisson et al., 2015).
Recent genome wide association studies of acute lymphoblastic leukemia (ALL) found variations in ARID5B (10q21.2), showing a strong association with the disease (Rudant et al., 2015). ARID5B acts as a transcription factor, and genetic variation in said gene affects the expression of the corresponding genes that might increase the risk of the ALL (Ge et al., 2018; Marke et al., 2018). However, transcription factor IKZF1 is also found to be associated with ALL, and regulates the differentiation of lymphoid lineage cells. Various genetic studies on ALL suggested that IKZF1 is either mutated or deleted (Marke et al., 2018). IKZF1 is emerging as a leading tumor suppressor gene in leukemia. In this study, we explored the association of ARID5B and IKZF1 gene with the prognosis of leukemia in the Jammu and Kashmir population. Our results suggested that these variants play a significant role in the etiology of leukemia.
Materials and Methods
Ethics statement
The Institutional Ethics Review Board (IERB) of Shri Mata Vaishno Devi University (SMVDU) approved the study under the notification number (SMVDU/IERB/16/41). All the details have been recorded in a predesigned pro forma and the written informed consent was obtained from each participant before conducting the study. All experimental procedures were conducted according to the guidelines and regulations set by the IERB, SMVDU.
Sampling
A total of 616 subjects were recruited for the study, out of which 210 were leukemic patients (cases) and 406 were healthy controls. All cases were histopathologically confirmed. The genomic DNA was isolated from the blood samples using Qiagen DNA Isolation kit (Cat. No. 51206). Agarose gel electrophoresis was used to analyze the quality of the genomic DNA and quantification was performed using UV spectrophotometer.
Genotyping
Genotyping of variants rs10740055 of ARID5B and rs6964823 of IKZF1 was performed using allele discrimination assay on MX3005p Agilent Real-Time PCR. Taqman Probe labeled with 2′-Chloro-7′phenyl-1-4 dichloro-6-carboxyl-flurorescin and Flurosceinamidite (Thermo Fisher Scientific) and Uracil-N-Glycoslyase Master Mix (Applied Biosystems) was used for genotyping. The volume of the total PCR was 10 μL, comprising 2.5 μL of Taqman UNG Master Mix, 0.25 μL of probe, 3 μL DNA (5 ng/μL), and 4.25 μL nuclease-free water added together to make the final volume. The thermal conditions adopted were hold for 10 min at 95°C, 40 cycles at 95°C for 15 s, and 60°C for 1 min. All the samples were run in a 96-well plate with three no template controls. The post-PCR detection system is used to measure allele-specific fluorescence. A total of 93 random samples each from leukemia were picked and regenotyped for cross-validation of the genotyping calls and concordance rate was 100%.
Statistical analysis
Statistical analysis of the data was done by using SPSS software (v.20; Chicago, IL). Chi-square was performed and genotyping frequencies were tested for total Hardy-Weinberg equilibrium. Logistic regression analysis was used to estimate odds ratio (OR) at 95% confidence interval (CI) and respective level of significance was estimated as p-value.
Results
This case-control association study included 616 samples (210 leukemic cases and 406 healthy controls). The clinical characteristic distributions of the cases and the controls are given in Table 1. The frequency of risk allele “A” observed for the variant rs10740055 of ARID5B gene was 0.441 in cases and 0.402 in controls. The frequency of risk allele “A” observed for the variant rs6964823 of IKZF1 gene was 0.364 in cases and 0.266 in controls. The allele frequency distribution for both the SNPs is summarized in Table 2. This study demonstrates that the variant rs6964823 of IKZF1 genes is significantly associated with leukemic risk with an allelic OR of 1.6 (1.2-2.0 at 95% CI) and p-value 0.005. To evaluate the effect of risk genotypes along with the heterozygous genotypes, which are underestimated in the dominant model (Gaye and Davis, 2017), the additive model was applied and the OR observed was 1.5 (1.1-2.1, 95% CI) with p-value 0.009 after correcting it with age, gender, body mass index (BMI), alcohol consumption, and smoking. The allelic OR observed for the variant rs10740055 of ARID5B gene was 1.2 (0.9-1.5 at 95% CI) and p = 0.194. The unadjusted OR observed was 2.9 (1.8-4.4 at 95% CI) and p = 8.2 × 10−7, and after correction the OR was 2.5 (1.5-4.1 at 95% CI) and p = 0.0002. The variant was found not to be associated with the additive model and OR observed was 1.1 (0.7-1.5 at 95% CI) and p = 0.637 after correction with age, gender, BMI, alcohol consumption, and smoking.
Clinical Characteristics for Cases and Controls
BMI, body mass index.
Allele Frequency Distribution Variants rs10740055 (ARID5B) and rs6964823 (IKZF1) Risk in Leukemia Risk
Corrected with age, gender, BMI, alcohol consumption, and smoking.
CI, confidence interval; HWE, Hardy-Weinberg equilibrium; OR, odds ratio.
Furthermore, the cumulative effect of risk genotypes of both the variants, that is, AA (homozygous)+AC (heterozygous) of ARID5B and AA (homozygous)+AG (heterozygous) of IKZF1, was analyzed against the genotype CC (homozygous) of ARID5B and genotype GG (homozygous) of IKZF1, which are nonrisk genotypes. The interaction analyses provide cumulative effect for risk allele XA_ARID5B versus CC_ARID5B+XA_IKZF1 versus GG_IKZF1 and the unadjusted OR observed was 3.7 (2.10-6.8) and p = 4.1 × 10−6, and after correction, the OR observed was 4.9 (2.4-9.9) and p = 8 × 10-6, signifying that the individuals with risk genotypes of both the variants have higher risk of leukemia in the studied population group.
Discussion
In this study, we attempted to explore the association of the variant rs6964823 of IKZF1 and rs10740055 of ARID5B independently as well as cumulatively in the ethnic population of northern India. These variants were previously evaluated in the Taiwanese population with ALL but both the variants did not show association with the disease (Lin et al., 2014). Study in Hispanic population found variant rs6964823 of IKZF1 to be associated with ALL (Hsu et al., 2015). In another study, variant rs10440055 of IKZF1 showed significant association with ALL among the Yemeni children (Al-absi et al., 2017), whereas variant rs10740055 of ARID5B showed association with ALL in the French children (Orsi et al., 2012) and French Canadians of European descent from the Quebec cohort (Healy et al., 2010). In population-based studies, ARID5B and IKZF1 have been found to be associated with ALL susceptibility in Indian children irrespective of ethnicity (Bhandari et al., 2016). Our study indicated that genetic variants rs10740055 of ARID5B and rs6964823 of IKZF1 pose as risk factors for leukemia in the population of Jammu and Kashmir.
IKAROS or IKZF1 function is associated with its ability to localize the pericentromeric-heterochromatin (PC-HC) in the nucleus. Binding of IKAROS protein to the upstream promoter region of the target gene leads to its recruitment with PC-HC (Olsson and Johansson, 2015), thereby, regulating the transcription for chromatin remodeling. ARID5B forms a complex with (Cys4-His-Cys3-Motif in plant Homeo-domain) finger protein 2 (PHF2) and the ARID5B-PHF2 complex, thereby inducing demethylation of the dimethylated “Lys-9” of histone H3 (H3K9me2) (Huo et al., 2017), leading to transcription activation of targeted genes. IKAROS is phosphorylated by CK2 kinase and dephosphorylated by protein phosphatase 1 (PP1) with interaction at a consensus PP1 binding motif, RVXF (Piazza et al., 2012) (Fig. 1). Since IKAROS binding peaks are present in the promoter of ARID5B, it advocates the regulatory role of IKAROS on ARID5B expression in ALL cells (Song et al., 2015). This inhibitory effect on IKAROS results in the decreased expression of ARID5B, thereby inducing relapse of leukemia that suggests its associative expression and regulation in leukemic progression (Olsson and Johansson, 2015). A previous study demonstrated the interdependent expression of IKZF1 with ARID5B and concluded that IKZF1 regulates the expression of ARID5B, inducing an “oncogenic effect” in ALL (Ge et al., 2018).
Hypothesized IKZF1-mediated regulation of promoter suppression and ARID5B expression. CK2 proto-oncogene phosphorylates IKZF1 inhibiting its promoter binding activity. IKZF1 inhibition leads to chromatin suppression and decrease in expression of ARID5B. This inhibitory effect on Ikaros results in the decreased expression of ARID5B, thereby inducing leukemia progression and relapse.
We also evaluated the cumulative effect of the risk genotypes by combining both the variants against the combination of nonrisk genotypes. This approach of analysis between the risk and nonrisk genotype combinations in a population showed highly significant associations. Previous studies have found that IKZF1 dysfunction (Yu et al., 2002) and mutations in ARID5B chromatin remodeling gene play a critical role in other types of cancers as well.
Therefore, as it indicates that both IKZF1 and ARID5B genes are regulators of cellular processes, any alteration in these genes may lead to the development of leukemic risk. Thus it is much needed to identify the effect of these genetic variants in molecular functioning of both IKZF1 and ARID5B and the associated downstream signaling cascade.
Conclusion
Our findings provide the evidence that the variants rs10740055 of ARID5B and rs6964823 of IKZF1 are associated independently and cumulatively with the predisposition of leukemic risk in northern India. Understanding the effects of these variants in different ethnic groups of India as well as other Asian population is crucial as they may confer risks of leukemia within populations and may act as predictive or progonostic biomarker of leukemia in the population of Jammu and Kashmir.
Footnotes
Acknowledgments
R.K. and A.B. thank the Department of Science & Technology, Government of India (DST/SSTP/J&K/459) for financial assistance. A.B. thanks G.R.B., S.V., R.S., D.B., B.S., and M.P. for their constant support in sampling, I.S. and V.S. for continuous assistance in data analysis, and J.S. for technical suggestions.
Author Disclosure Statement
No competing financial interests exist.