Analysis of SNP rs1800796 Association with Risk of Rheumatoid Arthritis in Pakistani Population: A Case Control Study

Haseeb Nisar1, Usman Pasha1, Kiran Hanif1, Rizwan Abid1, Rehan Ahmad Khan Sherwani2, Haja N. Kadarmideen3 and Saima Sadaf1* 1Institute of Biochemistry and Biotechnology, University of the Punjab, Lahore-54590, Pakistan 2College of Statistical and Actuarial Sciences, University of the Punjab, Lahore-54590, Pakistan 3Department of Applied Mathematics and Computer Science, DTU Compute, Denmark Article Information Received 23 August 2019 Revised 20 October 2019 Accepted 30 October 2019 Available online 25 June 2020


INTRODUCTION
R heumatoid arthritis (RA), a chronic autoimmune disorder, is affecting 0.5-1.0 % of the world population (Sokka, 2007). It is generally believed that an imbalance between the pro-and anti-inflammatory cytokines causes joints' inflammation, deformities, destruction, and the disability in the affected RA patients (Firestein, 2003); interaction of several genetic, epigenetic and environmental factors also contribute to this effect (Choi et al., 2006). Genetic factors that are involved in the disease pathogenesis include loci for genes of human leukocyte antigen (HLA) and non-HLA genes like TNF-α, IL-1 β, CTLA4, IL-23 receptor, PADI4 and STAT4 (Kurko, 2013). Amongst the non-HLA genes, interleukins (ILs, a subset of large cytokine family), appear to be essential in activating the various immune responses, such as inflammation.
ILs, after their production, generally proceeds to the target cells with the help of IL-specific receptors and activate numerous signaling cascades including those related to the inflammatory responses (Assier et al., 2010).
Genetic predisposition of several pro-inflammatory cytokines has been reported to be associated with increased risk or pathogenesis of RA in different population groups. More importantly, single nucleotide polymorphisms (SNPs) in genes of IL-6 [a key mediator of inflammation in tissues and/or synovial fluids of RA patients] have been reported in British, Turkish, Polish, Spanish, Chinese and Iranian subjects in the individual retrospective studies. Some of the reported SNPs displayed protective role whereas others were found associated with increased risk of RA in different ethnic groups.
Since genetic predisposition is population dependent, the screening of population for identification of SNPs in RA-associated immune-modulatory cytokines and/or IL genes may serve as useful "genetic marks" for the early diagnosis of RA. However, substantial inconsistency in the SNPs data found due to different ethnic backgrounds, O n l i n e

F i r s t A r t i c l e
varied number of sample size and the uncorrected physiological status (Dar et al., 2016;Zhai et al., 2012;Dai et al., 2014;HuaLee et al., 2015;Cai et al., 2014;Chen et al., 2012;Lee et al., 2012) has prompted us to perform systematic literature search on key activators/ mediators of inflammation in RA, such as IL-6. In the present study, we therefore have assessed the association of polymorphism in the promoter region of IL-6 (-572 G/C) with RA, firstly in overall population and later in Asian and Caucasian ethnicities; independent case control studies, recently added in the databases, were included in the analysis. The probable association of IL-6 -572G/C polymorphism in Pakistani RA patients versus healthy subjects (n=376) was thereafter evaluated using real-time PCR and restriction fragment length polymorphism (PCR-RFLP) analysis. Finally, using "STRING" online software, we have provided an analysis for the interconnection between different genetic polymorphisms of ILs and their combinatorial effect on the RA development and/or progression. This, to our knowledge, is the first study from Pakistan screening RA patients for genetic predisposition of IL-6 -572G/C polymorphism.

Literature search-identification of eligible studies and data extraction
We performed a web-based search strategy using the PubMed, Springer-link and Google Scholar as databases, with keywords/terms: "genetic variation", "polymorphisms", "IL-6", "rheumatoid arthritis" and "arthritis", for the identification of the studies examining association between IL-6 polymorphism and RA. The studies containing the overlapping data and/or data of family-based linkage analysis were excluded while the one, meeting the following two-fold criteria were included: 1) original research articles (excluding reviews) reporting the gene polymorphism of IL-6 -572G/C (rs1800796), 2) Case control studies having sample size ≥100 each and odds ratio (OR) with confidence interval (CI) 95%, when used in research validations.
The information extracted from the research articles were: authors' name, year of publication, country of origin, ethnicity of the patients enrolled, number of reported cases vs controls, and the number of genotypes and allelic frequency.

Statistical analysis
All statistical analyses were performed using STATA software version 14.1 (USA). Two-sided p-values <0.05 were considered as statistically significant. The strength of the association between IL-6 572 G/C rs1800796 and RA risk was assessed by the ORs along with their corresponding 95% CI. Four genetic models (Allelic, Dominant, Recessive and Additive) were used while calculating the ORs. Subgroup analyses were also performed by ethnicity as either Asian or Caucasian. Additionally, variations between studies were determined using Chi-squared based Q statistic test (Wu and Li, 1999). The effect of heterogeneity was quantified using I 2 statistic test and was categorized as: high (I 2 > 50%), middle (25% <I 2 < 50%) and low (I 2 < 25%) (Higgins and Thompson, 2002). Random effect model was used when a p-value <0.10 and I> 50% were the outcome, indicating significant heterogeneity. Fixed effect model was used in the cases where no heterogeneity existed. Genotype distribution of each polymorphism in controls was checked for Hardy Weinberg equilibrium (HWE) by the chi-square test. Publication bias was investigated using both Funnel plot and Egger linear regression test. A p-value <0.05 indicated a significant publication biasness.

Subjects of case-control study
Two hundred patients (35 males and 165 females; mean age ±44.1 years) from Punjab, Pakistan diagnosed with RA, in accordance with the 2008 Classification Criteria of the American College of Rheumatology, along with 176 healthy subjects (55 males and 121 females; mean age±45.1) of the same geographic and ethnic background, were enrolled in this study. The patients were recruited from the Department of Rheumatology, Sheikh Zayed Hospital, Lahore, Pakistan and the blood samples were collected with the informed consent while maintaining the privacy and confidentiality of the study subjects. The study design was duly approved by the Ethical Review Board of the University of the Punjab, Lahore (Bioethics-138/17).

Genotyping
Genomic DNA from the collected blood samples was isolated using Gene Jet Genomic DNA Purification Kit (ThermoFisher Scientific) and its quantification was done on Nano Drop 2000 spectrophotometer. A pair of forward and reverse primers viz F-IL-6 5´-GGAGACGCCTTGAAGTAACTGC-3´,R-IL-6 5´-GAGTTTCCTCTGACTCCATCGCAG-3´ was used to amplify a 163 bp promoter region of IL-6, by performing real time-PCR on Bio-Rad CFX-96 system (USA) (Shahid et al., 2019). The amplicons were digested with BsrBI restriction enzyme (recognition sequence: CCG ↓ CTC) in accordance with the manufacturer's instructions followed by analysis on 2 % agarose gel to identify the -572 G/C polymorphism in IL-6 promoter. Genotype and allele frequencies of RA patients, for IL-6 -572 G/C SNP, were compared with the control group using two-sided O n l i n e

rs1800796 polymorphism and RA susceptibility in reported literature
The literature search, based on title and abstract details, identified a total of 62 articles. After critical analysis, 6 studies (published between 2006-2017) involving 1185 RA patients and 1119 controls, met our inclusion criteria whereas 56 studies were excluded being reviews, metaanalysis or due to the SNPs other than rs1800796 (Fig.  1). The ethnicity-based analysis was thereafter performed for Caucasians and Asian populations; the list of studies along with details of study subjects is summarized as Table  I. The genotype frequencies of both cases and controls were extracted from each study and are listed in Table II. The genotype distribution of controls met the HWE. The results of meta-analysis revealed association between RA and rs1800796 polymorphism (IL-6 -572 G/C) in allelic (OR=0.8130, 95CI=0.708-0.932, p=0.003), co-dominant (OR=0.797, 95%CI=0.667-0.952, p=0.013) and dominant model (OR=0.784, 95%CI=0.617-0.997, p=0.048) in all study subjects. Furthermore, in subgroup analysis significant association was found in Asian population under the allelic and co-dominant model (Table II).

Heterogeneity and publication bias
A significant degree of heterogeneity (p<0.1) was noticed in most of the comparison models (Table II) and in order to identify the heterogeneity source, we performed meta-regression by ethnicity, year of publication and sample size. Publication bias could be a problem in this analysis because of the false number of positive studies. Generally, funnel plots are used to detect the publication bias but due to small number of studies included in our study, it was difficult to correlate. Egger's regression test, however, showed no evidence of publication bias in any of the genetic models (P>0.05).

Genotyping analysis
Prior to genotyping, the laboratory parameters for each patient such as rheumatoid arthritis factor (RA Factor), erythrocyte sedimentation rate (ESR), nodules formation (NF) and Disease Activity Score 28 (DAS28) were calculated and the results are summarized as Table  III. 82 % of the enrolled patients were females with mean age 44 years, reflecting at least 4-times high burden of disease in elderly women than men. Following amplification and digestion of 163 bp fragment, we were able to see the two bands of 102and 61 bps in case of dominant (GG) genotype; 163-, 102-and 61 bp fragments for heterozygous (GC); and a single 163 bp fragment for the recessive (CC) genotype ( Fig. 2 right panel). The IL-6 -572 G/C SNP genotype distribution followed the HWE in the control group. Frequency of G-allele distribution in the patient and the control groups was 75 and 70 % whereas that of C-allele was 25 and 30 %, respectively (P=0.803, OR=0.880, 95% CI= 0.323-2.396). Significant association was not observed in the genotype frequencies when the cases and the control groups were compared, as shown in Table IV

O n l i n e F i r s t A r t i c l e
CI=1.019-1.081), with reference to the gender.

DISCUSSION
IL-6 (MW: approx. 26 kDa), a pleiotropic cytokine, is involved in the regulation of T-and B-cells' proliferation, differentiation and/or maturation and is responsible for activating the host immune system. Owing to the involvement of IL-6 in diverse repertoire of biological functions, we predicted the functional protein networks of IL-6 by the STRING database, which revealed its interconnectivity with other inflammatory cytokines viz. IL-17A, 1L-17F, IL-18 and IL-22 etc., via NF-κB1 and STAT1, STAT3 proteins (Fig. 3, upper panel). Earlier studies have documented that IL-6 differentiates the T-lymphocytes into T H 17 cells, which in turn produce IL-17 (Firestein, 2003). Since T H 17 cells are considered to be the key players in inducing damage to the tissues during the course of inflammatory and other autoimmune O n l i n e

F i r s t A r t i c l e
SNP rs1800796 and its Association with RA Risk disorders (such as RA), the role of IL-6 in the development and/or progression of RA seems vital (Fig. 3, lower panel). This further suggests that the genetic and/or epigenetic alterations affecting the IL-6 expression may contribute in dysregulating the functions of STAT1, STAT3 and NFKB proteins, which directly/indirectly are involved in stimulating the expression of a wide variety of genes linked with inflammatory responses. Amongst the several polymorphisms of IL-6 gene, the two most commonly reported (i.e., -174 and -572 G/C) lie in the promoter region (Muniz et al., 2013;Arman et al., 2012). This region gives us an opportunity to predict the RA related loci. Preliminary review of the previous research studies revealed contradictory data relating to the role of rs1800796 SNP in RA development and/or progression. Huang et al., 2007, for instance, reported significant association of IL-6 -572 G/C polymorphism with RA whereas others (Muniz et al., 2013;Arman et al., 2012;Li et al., 2014;Lu et al., 2009;Srirangan and Choy, 2010;Fishman et al., 1998;Terry et al., 2000) described insignificant (Table I). We, therefore, performed critical assessment of the impact of proposed association in these case-control studies. Interestingly, the association between RA and IL-6 -572G/C polymorphism was found in all study subjects cumulatively, in the allelic, co-dominant and dominant models. In the subgroup analysis, the allelic and co-dominant models displayed significant association in the Asian ethnic group but not in the Caucasians. Overall, the results identified the IL-6 -572 G/C polymorphism as the hotspot region to be involved in the disease activity. When analyzed in the Pakistani RA patients and the controls (n=376), enrolled in this study, the differences in the genotype and allelic frequencies were found insignificant. Though our findings are in good agreement with those reported in case of Turkish and Mexican populations (Table I and  genotypes predominated in different population groups; genotype CC in the Chinese and Taiwan populations; GC in Egyptian; and GG in Turkish, Mexican and Pakistani populations. This heterogeneity advocates the importance of registering patients from different ethnic backgrounds along with their detailed clinical information in genomewide association studies of RA. The systematic literature search followed by casecontrol experimentation is of great significance as adding up new research data from Pakistan, for the first time. However, more research with significantly large number of patients originating from diverse ethnicities needs to be performed on IL-6 SNPs. More so, one or two SNPs cannot qualify as casual for polygenic complex diseases or be considered fully predictive of RA, primarily because this complex disease involves a large number of genes and environmental factors as potential risk factors. Nonetheless, we have provided estimation of SNP effect sizes with 95 % CI that may be used in developing polygenic risk scores for RA.

CONCLUSION
While earlier studies demonstrated that IL-6 rs1800796 polymorphism is associated with RA susceptibility in different genetic models, our casecontrol study involving Pakistani subjects (Asian ethnicity) revealed non-significant association between IL-6 rs1800796 gene polymorphism and RA risk; the GG genotype group, however, was found more susceptible to RA than the counterpart CC group. Further studies on next generation sequencing data along with detailed functional characterization of this variant seems imperative.

Compliance with ethical standards
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This study was supported by a grant from Higher Education Commission, Government of Pakistan (Grant No. 8488/2017). Authors are thankful to all RA patients who volunteered to provide the blood samples for this study. Informed consent was obtained from all individual participants included in the study.

Conflict of interest statement
Authors of this manuscript declare no financial / competing conflicts of interest.