Genotypic and haplotype analysis of Interleukin-6 and -18 gene polymorphisms in association with clinicopathological factors in breast cancer.

Cytokines are multifunctional glycoproteins that play a vital role in the tumor microenvironment and progression of breast cancer. Genetic polymorphisms may influence the immune responses restrained by pro- and anti-inflammatory cytokine expression in tumors. Hence, the present study evaluated the contribution of Interleukin (IL) 6 (rs1800797, rs1800796, and rs1800795) and IL18 (rs1946518, rs187238, and rs549908) genotypes and their haplotypes to the risk, progression of breast cancer in South Indian population. The polymorphisms of IL6 -597G > A, -572C > G & -174G > C, and IL18 -607C > A, -137G > C, 105A > T were genotyped through PCR-RFLP and As-PCR assays in the blood DNA of 600 subjects. We have performed haplotype, LD, univariate, multivariate logistic regression, and Kaplan-Meier analyses for the obtained data. The frequency of AA genotype & A-allele of IL6 -597G > A, and CC genotype & C-allele of IL6 -174G > C polymorphism was higher in breast cancer patients and was found to be significantly associated with late (advanced) stage, metastasis, etc. Further, IL18 -607C > A, -137G > C, and 105A > T polymorphisms were found to be associated with lobular carcinoma subtype, PgR -ve, and HER2 +ve breast cancer patients. In survival analysis, we have observed that the C-allele of IL6 -174G > C polymorphism to be significantly associated with 5 years of overall survival in breast cancer subjects. All SNPs of the IL6 and IL18 genes showed perfect LD; the G-C-C, A-G-G, and A-C-C haplotype combinations of IL6 gene conferred 2.09, 2.25, and 4.72 folds risk for breast cancer respectively. Hence, our results suggest the importance of genotypic and haplotype analysis of IL6 and IL18 gene variants in the progression and risk prediction of breast cancer.

survival analysis, we observed that the C allele of rs 1800795 was found to be significantly associated with 5years overall survival in breast cancer subjects. Conclusions: Overall, our results suggest the importance of genotypic and haplotype analysis of IL6, and IL18 gene variants in progression and risk identification of breast cancer.

I. Introduction
Breast cancer (BC) a strong interaction between genetic and epigenetic factors cause dynamic changes in the genome leading to uncontrolled cell growth, ability to invade and metastasize [1,2]. Recent experimental evidences implicate the role of cytokines in breast cancer development and progression. Interleukins (IL) are a class of cytokines primarily expressed by leukocytes and are low molecular weight protiens, which act on many target cells often in an additive, synergistic or antagonistic manner and are considered as key intercellular mediators that control survival, growth, differentiation, and the effector cell function. In addition, they are also involved in the pathology of BC surveillance system and other complex diseases [3][4][5].
IL6 and IL18 belong to the IL1 cytokine family and are known to regulate the immune reactions, inflammation and promote tumor growth by up-regulating angiogenic and anti-apoptotic proteins [4]. The genes for IL6 and IL18 are encoded on human chromosome 7p15.3 and 11q23.1 respectively. It has been reported that several important polymorphic sites in the IL6 and IL18 genes, including three in the promoter regions (-597G>A, -572C>G & -174G>C) of IL6 and two in promoter and one in exon regions (-607C>A, -137G>C & 105A>T) of IL18 play an important role in breast cancer progression [6][7][8]. Furthermore, the single nucleotide polymorphisms (SNPs) have been considered to be important biomarkers in cancer screening, staging, grading and risk assessment of the disease [9].
Despite the limitations in individual SNP analysis, haplotype blocks in the human genome which are formed as a result of linkage disequilibrium, can be used to identify its association with human diseases like cancer [10]. Hence possible that a haplotype effect, rather than an individual effect of SNPs on clinicopathological status, might explain some of the prognostic and survival information. The aim of the present study is to perform a comprehensive analysis of the possible prognostic importance of genotypic and haplotypes of the

Study Design and Participants
A total of 300 breast cancer patients and 300 healthy cancer-free control subjects with similar socioeconomic and geographic backgrounds were included in the study after signed written informed consent. The patients with Breast cancer were recruited from the MNJ Institute of Oncology and Regional Cancer Centre (MNJIO&RCC), Hyderabad during the period of 2010-2016. The histo-clinicopathological information of the patients was obtained from their medical records in the hospital. All the samples were coded and maintained confidentially. The study has been approved by the Institutional and Hospital review boards of Osmania University and MNJIO&RCC-Hyderabad, Telangana state, INDIA.

Polymorphism and SNP Genotyping analysis
Genomic DNA was extracted from peripheral blood using the Bio-serve DNA Blood Mini kit (Bio-Serve, INDIA), according to the manufacturer's instructions [11]. The genotypes of IL6 (-597G>A, -572C>G and -174G>C) and IL18 (-607C>A, -137G>C and 105A>T) gene polymorphisms were determined using PCR-RFLP (Polymerase chain reaction-Restriction Fragment Length Polymorphism) and As-PCR (Allele Specific-Polymerase Chain Reaction) methods respectively. PCR primers, RFLP enzymes, size of the PCR-RFLP products and conditions are shown Table 1. As a genotyping quality assessment purpose, we have randomly selected and repeated 10% of the samples. The samples were found to be 100% concordant in two independent assays. forward primer (F); reverse primer (R); melting temperature (mT); restriction endonuclease (RE); base pair (bp).

Statistical analysis
χ2 (chi-squraed) test and the odds ratio (OR) for allele and genotype frequencies of the studied polymorphism between cases and healthy controls were determined. All the p values were two sided, and the level of significance was taken as p<0.05. The different statistical tools used to analyse the quantitative and qualitative variables in the present study are IBM SPSS version 20.0, SNPStats (http://bioinfo.iconcologia .net/index.php? module ¼Snp stats) [12], HAPLOVIEW (http://www.broad.mit.edu/mpg/haploview) [13], Generalized Linkage disequilibrium and Graph Pad Prism [14].

Comparison of demographic characteristics among patients with breast cancer and controls
The detailed demographic and baseline clinicopathological characteristics of study subjects are illustrated in Fig. 1. In the present study the age of the breast cancer patients was within a range of 30-75 years, and their average age at diagnosis was 47.98 years (SD:±10.80) while the age of the controls at recruitment ranged from 35-68 years with a mean age of 46.34 years (SD:±7.97). An interesting finding that there was a significantly greater proportion of subjects are with higher BMI, mixed diet, history of smoking habit and alcohol consumption in the BC group (37.0%, 85.56%, 18.34% and 40.4%) than in the carcinoma-free control group (15.7%, 71.0%, 9.0% and 19.0%) respectively (p <0.001), while there was a no significant difference observed in age, area of living and occupation (p>0.05) of the patients and controls as shown in Fig.1A. The frequency distributions of age at late menarche >14 years (27.0%), nulliparous (8.34%), lactation (9.0%) and history of parental consanguinity (37.0%) were high in breast cancer patients compared to controls. Moreover, there was no significant difference between the two groups in the frequency distribution in terms of menstrual cycles, and menstruation status (p>0.05). The familial incidence of the breast cancer was found to be 15.0% while the same of the other carcinomas was found to be 19% in breast cancer patients as shown in Fig.1B.
The clinicopathological characteristics of breast cancer cases are shown in Fig. 1C for some of the patients there was a missing data on receptor status (8.67%) respectively.

Association Analysis of Alleles and Genotypes with Risk of Breast cancer
The genotypic and allele frequencies of 6 SNPs in the IL6 (-597G>A,-572C>G, -174G>C) and IL18 (-607C>A, -137G>C, 105A>T) genes in 300 breast cancer patients and 300 controls groups adjusted with age by multivariate logistic regression analyses are summarized in Table 2. The frequency of A-allele and the GA and AA genotypes of IL6 -597G>A polymorphism have shown a significant association (p<0.01) with breast cancer.
The frequency of G-allele and GG genotype of IL6 -572C>G polymorphism was found to be less in breast cancer cases when compared to controls. Further analysis revealed that the C allele of IL6 -174G>C polymorphism was found to be significantly associated with breast cancer patients when compared to the healthy controls.
The genotype frequency of IL18 -607C>A polymorphism did not revealed any significant association with breast cancer .The C-allele of -137C>A gene polymorphism was found to be similar in both controls and breast cancer patients, though the CC genotype was observed to be significantly associated with risk for development of breast cancer (p<0.04). Further TT genotype frequency of IL18 105A>T polymorphism was marginally higher in breast cancer cases compared to controls are showed in Table 2. The genotype frequencies of

Association of genotypes with Clinicopathological risk Factors of Breast cancer
Analyses of association between genotypic and clinico-pathological features adjusted by age with multivariate logistic regression analyses in breast cancer patients revealed that the AA genotype of IL6 -597G>A polymorphism was found to be associated with TNM staging (T2-T4 late stage); GG genotype of IL6 -572C>G polymorphism was found to be significantly associated with late (advanced) stage of breast cancer patients. In addition, the frequency of AA genotype of IL6 -597G>A, GG genotype of IL6 -572C>G and CC genotype of IL6 -174G>C gene polymorphism were found to be significantly associated with positive metastasis status in breast cancer patients as summarized in Table 2a. However, analysis of polymorphic variants (-607C>A, -137G>C and 105A>T) of IL18 gene polymorphisms with respect to histological subtype revealed that the AA and CA genotype of -607C>A polymorphism and AT+TT genotype of 105A>T polymorphisms had significant association with lobular carcinoma subtype in breast cancer patients. Furthermore, CC and GC genotypes of -137G>C polymorphism was found to be significantly associated with PgR -ve and HER2 +ve breast cancer patients respectively (Table 2b).

HapMap analysis of different ethnic groups:
We have compared the allele frequencies of IL6 and IL18 in different ethnic groups (Asian-JPT, Asian-HCB, Japanese-JPT, and European CEU) using HapMap dataset and found that the frequency of G allele of -597, the C allele of -572 and C allele of -174; C allele of -607, G allele of -137 and A allele of 105 were similar to that of our population (Fig. 2).  Table 3). Analysis of clinicopathological variables for haplotypes did not show any significant association with breast cancer ( Table 4).   (Table   5). However, the analysis of clinicopathological variables with haplotypes revealed that A-G-A and A-C-T haplotypes to be associated with histological type of breast cancer (ductal) (p<0.019). In addition, we also found strong association of C-C-T haplotype with axillary lymph node and HER2/neu receptor status while C-C-A with metastasis status respectively as shown in Table 6.
While, the SNPs of IL6 -572 C>G, IL6 -174G>C also showed a strong LD with IL18 -607C>A and IL18 -137G >C. Similarly, the SNPs IL18 -607C>A and IL18 -137G >C have also shown a strong LD with 105A>T with Pearson's correlation coefficient (D) of in red colour Table 7.  Table 7. Linkage disequilibrium analysis between the Six Interleukin SNPs analyzed in Breast cancer patients.
SNPs    Table 9 and Fig 3 A-F).

Transcription factor binding sites (TFBS) and Pre-mRNA secondary structures predictions
The prediction of transcription factor binding sites (TFBSs) for IL6 gene (-597G>A, -572C>G and -174G>C) polymorphisms revealed that G-allele of -597G>A polymorphism has NF-1, Sp1 and AP-2alph sites whereas A-allele has TBP, YY1 and NF-1. Further it is revealed that the C-allele of -572C>G polymorphism has binding site for Sp1, whereas G-allele has loss of Sp1 binding site. Furthermore, G allele of 174G>C polymorphism has no binding site while C -allele has binding site for NF-1 respectively as depicted in Figure 4. Similarly, the IL18 gene -607C>A polymorphism has shown that C allele has binding sites for C/EBP beta, MEB-1 sites whereas A allele has binding sites for C/EBP beta, HOXA4, MEB-1 sites. Further, the prediction of transcription factor binding sites of -137G>C polymorphism has shown that G allele has binding site for Nf-kappa B whereas C allele has an additional binding site for C/EBP beta and NF1 sites as respectively depicted in Fig. 4. The pre-mRNA secondary structures of IL18 105A>T polymorphism had shown that T-allele (-77.65 kcal/mol) has higher entropy and is less stable than A-allele (-76.08 kcal/mol) as deduced from the minimum free energy was shown in Fig. 4.

Protein-protein interaction (PPI) analysis
For protein interaction data, the present study utilized a human PPI dataset from the Search Tool for the Retrieval of Interacting genes/proteins. The obtained image contained several colored lines supporting their interactions, the more lines between two peptides and the more evidence to support their relationship. Our analysis revealed that the IL6 protein is found to be directly interacting with IL18, MMP1, 3, 9 and TIMP1 genes, while, MMP1 and MMP9 proteins are strongly interacting with each other and are co-expressed along with TIMP1 proteins as shown in Fig.5.

DISCUSSION
Multiple molecular mechanisms are known to be involved in the development of breast cancer.
Inflammation within the tumor microenvironment, one of the key mechanisms of carcinogenesis has been correlated with increased invasiveness and poor prognosis in many types of cancer including breast cancer [17]. Different cytokines are known to have diverse roles in breast cancer initiation and metastasis [18]. The cytokines are critical mediators of the inflammatory response [4]. The present study aimed to evaluate the association of IL6 (-597G>A, -572C>G, -174G>C) and IL18 (-607C>A, -137G>C, 105A>T) gene polymorphic variants with the epidemiological and clinicopathological characteristics in south Indian Breast cancer patients. The genetic variations of IL6 gene are known to influence the transcriptional activity by affecting the expression, leading to the susceptibility and progression of breast cancer [19]. Three single-nucleotide polymorphisms -597G>A, -572C>G and -174G>C have been identified in the promoter region of the IL6 gene, which might have an effect on IL6 transcription and plasma IL6 levels in multiple human diseases [20].
IL6 -597G>A polymorphism was reported to be associated with serum IL6 levels and as a susceptibility factor in several diseases such as multiple myeloma [21], rheumatoid arthritis [22], Sepsis [23], cervical cancer [24] and chronic obstructive pulmonary disease [25]. The genotype and allele frequency distribution of IL6 -597G>A polymorphism in controls and breast cancer patients revealed that the AA genotype and A-allele frequency was higher in breast cancer patients compared to controls and has conferred 3.97 and 2.53 folds risk for breast cancer respectively. Several groups have shown an association between AA genotype IL6 -597G>A polymorphism with elevated serum IL6 levels and incidence of chronic inflammation with poor survival in multiple cancers. IL6 -572C>G polymorphism is another important promoter polymorphism associated with serum IL6 levels and susceptibility to several diseases like OSCC [26], hypertension [27], T2DM [28]. It has also been suggested that individuals carrying the C allele of IL6 -572C>G polymorphism have an increased risk for breast cancer. Similarly, the genotype and allele frequency distribution of IL6 -572C>G polymorphism in controls and breast cancer patients revealed that the CG and GG genotypes had a protective role against breast cancer. Further, G allele frequency was high in controls and was associated with reduced risk, indicating C-allele carriers are at an increased risk for breast cancer.
Another functional polymorphism in the promoter region of the IL6 gene at position -174G>C has been reported to be associated with diseases, such as OSCC [26], Ovarian [29], Cervical and breast cancer [30,31]. Earlier studies by Lagmay et al., have shown that the CC genotype of IL6 promoter -174G>C polymorphism to be associated with elevated high-risk for neuroblastoma [32] and prostate cancer [33]. A recent study suggested that individuals carrying the -174 C allele have an elevated risk for breast cancer [34]. The genotype and allele frequency distribution of IL6 -174G>C polymorphism in controls and breast cancer patients revealed that the frequency of CC genotype and C allele were significantly elevated in breast cancer cases and conferred 2.52 and 1.91 fold risk for breast cancer development respectively.
The IL18 gene promoter -607C>A and -137G>C polymorphisms and 105A>T polymorphism in exon 4 are found to influence the expression and may affect the blood plasma level of IL18 [35]. IL18 gene (-607C>A, -137G>C and 105A>T) polymorphisms have been associated with susceptibility to several diseases like CHD, Systemic lupus erythematosus (SLE), esophageal squamous cell carcinoma [36], prostate [37], colorectal [38], ovarian cancer [39], nasopharyngeal (rare type of head and neck) carcinoma [40] and breast cancer [41]. In contrary, no significant association was found when IL18 -607C>A, IL18 -137G>C gene polymorphisms were studied in patients with Crohn disease or ulcerative colitis and other cancers. Recently, similar results were found in patients with oral cavity cancer where there was no significant association of IL18 -607C>A polymorphism even after adjusting for age [42]. Similarly, in the present study the genotype and allele frequency distribution of IL18 -607C>A polymorphism in controls and breast cancer patients revealed that there was no significant association of IL18 -607C>A polymorphism with breast cancer.
Another single nucleotide polymorphism in the promoter region of the IL18 gene at position -137 G>C has been reported to be associated with a variety of diseases, such as Chronic Lymphocytic and Chronic Myelogenous [43], cervical [44], breast cancer [41] and unfavourable clinical outcome in patients affected by high-risk neuroblastoma, trauma inflammation etc [45]. In the present study, the genotype and allele frequency distribution of of the IL18 gene may be associated with the pathogenesis of asthma in Japanese patients [46,47]. However in the present study there was no significancant association between IL18 105A>T polymorphism and breast cancer in different genetic models studied.
Over expression of IL6 and IL18 genes have been found to be positively associated with the clinicopathological characteristics of several malignancies [26,48,49]. In the present study SNPs of IL6 (-597G>A, consequently creating imbalance in the vital cytokines system that results in excessive tumor progression and breast cancer development. To our best knowledge, present study is first to report the combined effect of three polymorphisms and haplotypes (A-C-C, A-G-G and G-C-C) along with the clinicopathological parameters which showed a significant association of IL6 polymorphism with BC. Overall, our results revealed that the polymorphisms in the promoter region of IL6, and IL18 genes when correlated with clinicopathological characteristics and survival rate have shown significant effects on the risk and progression of breast cancer, substantiated by in-silico analysis.

Conclusions
The study is focussed to understand the clinical importance of IL6 and IL18 gene variants in progression of breast cancer and to help in identifying individuals at risk of developing breast cancer. Interleukins are considered as promising targets for therapy due to their strong involvement in tumor pathology and progression at both molecular and clinicopathological levels. Therefore, understanding cancer subtypes, functional epigenomics, transcriptomic programs and signalling pathways regulating ILs expression is critical for developing breast cancer therapeutics, not only to treat but also to prevent cancer. Further such a comprehensive approach may reveal highly prominent candidate molecular markers in future for breast cancer diagnosis and prognosis.
Limitation of the study: To the best of our knowledge this is the first study reporting on the combined effects of SNPs of IL6 and IL18 genes in correlation with clinicopathological variables along with LD, survival rate and Insilico analysis. However, our study has several limitations. Firstly, a small study that was analysed in South Indian population, because we restricted the study subjects to individuals of South Indian ethnicity; it is uncertain whether these results can be generalized to other populations. Second, there were few patients with missing data on hormonal receptor status, which may bias the results indicating an association with advanced disease status. Third, our LD and survival analysis included only 600 samples and this may have limited the power of the pooled results.
Therefore, collaborative studies on different populations are necessary to corroborate our findings. Administrative support; data analysis and interpretation; article writing, editing, and final approval of the article.

Conflicts of Interest:
The authors declare no conflicts of interest.