The −44 C/G (rs1800972) polymorphism of the β‐defensin 1 is associated with increased risk of developing type 2 diabetes mellitus

Abstract Background The aim of this study was to establish the association of two polymorphisms of the β‐defensin 1 gene (DEFB1, OMIM#602056) with the risk of developing type 2 diabetes mellitus (T2DM) in a group of Mexican patients. Methods The 5′UTR −20 G/A, and −44 C/G polymorphisms of DEFB1 gene were genotyped by 5′ exonuclease TaqMan assays in a group of 252 patients with T2DM and 522 healthy control. Results Under dominant and additive models adjusted for the risk factors, the C allele of the −44 C/G polymorphism was associated with increased risk of T2DM (OR = 1.63, 95% CI = 1.07–2.48, pCdom = 0.021 and OR = 1.42, 95% CI = 1.05–1.91, pCadd = 0.023, respectively). In addition, the linkage disequilibrium analysis showed that AC haplotype was associated with an increased risk of developing T2DM (OR = 4.39, p = 0.04). The in‐silico analysis showed that the −44 C allele produces a binding site for the transcription factor Ikaros (IK). Conclusion This study demonstrates that the C allele of −44 C/G polymorphism, as well as haplotype AC are associated with the presence of T2DM in the Mexican population. The variation in this polymorphism of the DEFB1 gene could increase the migration of the macrophages to pancreatic islets accelerate the β‐cell dysfunction in T2DM.

pro-inflammatory cytokines. In addition, the elevated glucose and palmitate concentrations increase chemokines release that induce granulocyte colony-stimulating factor and macrophage inflammatory protein-1 from human and mouse pancreatic islets both in vitro and in vivo (Ehses et al., 2007;Inoue et al., 2018).
Experimental studies have shown that the β-defensin 1 plays an important role in activation of the innate immune system. This molecule acts as a chemoattractant of T lymphocytes, immature dendritic cells, monocytes, macrophages, and mast cells by interaction with CCR6, as well as in the release of pro-inflammatory cytokines (Soruri, Grigat, Forssman, Riggert, & Zwirner, 2007;Yang et al., 1999). This molecule is expressed by epithelial cells of the skin, gut, respiratory and urogenital tissue, the pancreas and the kidney (Schnapp, Reid, & Harris, 1998;Zhao, Wang, & Lehrer, 1996). Nonetheless, as far we know, in the literature there is no evidence about the genetic basis and the function of β-defensin 1 in the pancreas, as well as in the pathophysiology of the T2DM. However, recent reports have showed that the β-defensin 1 protein expression, as well as the mRNA levels increased in HMC and HEK-293 cells cultured with high levels of glucose and insulin (Barnea, Madar, & Froy, 2008;Malik & Al-Kafaji, 2007).
According with the antecedents and considering that the T2DM affect more than 550 million people all over the globe (American Diabetes Association, 2013;Chen, Magliano, & Zimmet, 2011), the aim of this study was to establish the association of two polymorphisms (−20 G/A [rs11362] and −44 C/G [rs1800972]) of the DEFB1 gene with the risk of developing T2DM in a group of Mexican patients.

| Population study
The study included 774 Mexican Mestizos, 252 consecutive patients with T2DM (30-78 years old) and 552 healthy controls (26-78 years old). From July 2010 to July 2015, 252 patients (189 men and 63 women, with a mean age of 58.5 ± 10.7) were referred to the Instituto Nacional de Cardiologia Ignacio Chavez. The diagnosis of T2DM was made according to the American Diabetes Association (ADA) and American Heart Association (AHA) based on three criteria, (a) an HbA1c value ≥6.5%, (b) fasting plasma glucose levels ≥126 mg/dl, (c) 2-hr plasma glucose ≥200 mg/dl during an oral glucose tolerance test. The test was performed as described by the World Health Organization, using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water (American Diabetes Association, 2013;Fox et al., 2015). Comorbidities such as BMI (weight [kg]/[height [m] 2 ), gender, age, hypertension, dyslipidemia, smoking, and alcohol habit were recorded. Blood pressure was measured three times by trained personnel using sphygmomanometry. Hypertension was defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg or the use of oral antihypertensive therapy. Plasma lipid concentrations were determined after a 12-hr overnight fast, EDTA blood samples were drawn and centrifuged within 15 minutes after collection; plasma was separated into aliquots and were analyzed within 24-hr after blood sample collection. Cholesterol and triglycerides plasma concentrations were determined by enzymatic/colorimetric assays (Randox Laboratories, UK). The phosphotungstic acid-Mg2+ method was used to determine HDL-C concentrations. LDL-cholesterol (LDL-C) was estimated in samples with a triglyceride level lower than 400 mg/dl, using the Friedewald formula, modified by DeLong (DeLong, DeLong, Wood, & Lippel, 1986). We followed the National Cholesterol Education Project (NCEP) Adult Treatment Panel (ATP III) guidelines to defined dyslipidemia with the following levels: cholesterol > 200 mg/dl, LDL-C > 130 mg/dl, HDL-C < 40 mg/ dl, and triglyceride > 150 mg/dl (https://www.nhlbi.nih. gov/guidelines/cholesterol/atp3_rpt.htm). Individuals with type 1 diabetes (T1D), maturity-onset diabetes of the young (MODY), secondary diabetes due to hemochromatosis, cystic fibrosis, or pancreatitis or individuals with a family member with T1D were excluded from the study. The 522 healthy individuals (395 men and 63 women, with a mean age 53.6 ± 7.52) belong from the Genetics of Atherosclerosis Disease (GEA) Mexican study. All subjects were asymptomatic and apparently healthy individuals without a family history of premature CAD, atherosclerosis, and T2DM who were recruited during June 2009 to June 2013 among blood bank donors and with the assistance of brochures posted in social service centers. The exclusion criteria included the use of anti-dyslipidemic or anti-inflammatory drugs at the time of the study, congestive heart failure, and liver, renal, thyroid, or oncological disease (Posadas-Sanchez et al., 2017). In addition, control subjects included in this study had a coronary calcium score = 0 as determined by computed tomography, indicating the absence of subclinical atherosclerosis in these individuals. All subjects included in this study were | 3 of 7 MARTINEZ-RIOS ET Al. ethnically matched and considered Mexican mestizos if they had been born in Mexico and if their family had lived in the country for at least three generations. A Mexican mestizo is defined as someone born in Mexico and a descendant of both the original autochthonous inhabitants of the region and of individuals, mainly Spaniards of Caucasian and/or Black origin, who came to America during the sixteenth century. The study complies with the declaration of Helsinki and was approved by the Ethics Committee of the Instituto Nacional de Cardiología Ignacio Chávez (INCICH). All participants provided written informed consent.

| Functional prediction analysis
Analysis in silico was made with the SNP Function Prediction (https://snpinfo.niehs.nih.gov/cgi-bin/snpinfo/snpfunc.cgi) program. This program analyze the location of the SNP (e.g., 5′-upstream, 3′-untranslated regions, intronic) and its possible functional effects, such as amino acid changes in protein structure, transcription factor binding sites in promoter or intronic enhancer regions, and alternative splicing regulation by disrupting exonic splicing enhancers or silencers (Xu & Taylor, 2009).

| Statistical analysis
Considering that the continuous variables such as age, body mass index, blood pressure, glucose, total cholesterol, HDL-C, LDL-C, and triglycerides did not have normal behavior, they were analyzed using the Mann-Whitney U test. On the other hand, the categorical variables such as gender, alcohol consumption, and smoking habit were analyzed with the chi-square test or Fisher's exact test. The association of the polymorphisms with T2DM was analyzed by logistic regression using the following models: co-dominant, dominant, recessive, heterozygous, and additive. Models were constructed in order to identify the variables that better explain the risk of developing T2DM between the study groups. Furthermore, models were built which incorporated one variable at a time, whereas final models included variables with biological relevance, statistical significance or both. Confounding bias was accepted when changes in estimated odds ratios (ORs) were equal or >10%. When a principal effect model was reached, the effect modification was also tested and interaction terms were constructed between the polymorphisms and various variables; the terms were included in the model when the significance of the p-value was higher or equal to 0.05. All p-values were corrected (pC) by the Bonferroni test. pC values <0.05 were considered statistically significant, and all odds ratios (OR) are presented with 95% confidence intervals. The occurrence of the T2DM in our population was based in the OR values: OR = 1 does not affect odds of developing T2DM, OR > 1 is associated with higher odds of developing T2DM, and OR < 1 is associated with lower odds of developing T2DM. The linkage disequilibrium analysis (LD, D") of the polymorphisms as well as the haplotype construction was performed with Haploview version 4.1 (Broad Institute of Massachusetts Institute of Technology and Harvard University, Cambridge, MA, USA). The analysis of data was performed with SPSS version 18.0 (SPSS, Chicago, IL) statistical package. The statistical power to detect association of the polymorphisms with T2DM was 0.80 and was estimated with QUANTO software (https://biostats.usc.edu/software).

| Characteristics of the study population
Baseline characteristics of the T2DM patients and healthy controls included in the study are presented in Table 1.

| Allele and genotype frequencies
The observed and expected frequencies in the two polymorphic sites of the DEFB1 gene were in Hardy-Weinberg equilibrium. The 5′UTR −20 G/A (rs11362) polymorphism frequency was similar in patients with T2DM and healthy controls. However, under dominant and additive models adjusted for gender, age, BMI, hypertension, dyslipidemia, and alcohol consumption, the C allele of the 5'UTR −44 C/G (rs1800972) polymorphism was associated with increased risk of developing T2DM (OR = 1.63, 95% CI = 1.07-2.48, pC dom = 0.021 and OR = 1.42, 95% CI = 1.05-1.91, pC add = 0.023, respectively) ( Table 2).

| Functional prediction
The functional prediction analysis showed that the presence of the A allele of the DEFB1 5′UTR −20 G/A (rs11362) polymorphism produces a binding motif for the breast cancer transcription factor (BRCA). The analysis also revealed that the C allele of the DEFB1 5′UTR −44 C/G (rs1800972) polymorphism generates binding motifs for ikaros transcription factor (IK).

| DISCUSSION
In the present work, two polymorphisms (−20 G/A and −44 C/G) located in the 5′ untranslated region (UTR) of DEFB1 gene were analyzed in T2DM patients and healthy controls. The distribution of 5′UTR −20 G/A (rs11362) was similar in both groups. Nonetheless, the analysis of the 5′UTR −44 C/G (rs1800972) polymorphism showed that the C allele was associated with an increased risk of developing T2DM under dominant and additive model. To the best of our knowledge, this is one of the few studies that describe the association of these polymorphisms with T2DM. Agreement with our results, Nemeth et al., report that the −44 CC genotype of the 5′UTR −44 C/G polymorphism was associated with increased risk for developing diabetes (OR = 2.05) in a Hungarian population. In addition, the authors reported that the frequency of the −44 GG genotype was lower in patients with diabetes (2%) than in the healthy controls (9%), suggesting that this genotype has a protective role in the development of diabetes (Nemeth et al., 2014). The association of 5′UTR −44 C/G polymorphism with other inflammatory diseases support our findings.  systemic lupus erythematosus patients and 288 healthy controls. In this work, the authors reported that the −44 GC genotype was associated with increased risk of developing this disease (OR = 1.60) in a Brazilian population (Sandrin-Garcia et al., 2012). Our findings are consistent with those studies that report positive associations of the −44 C/G polymorphism with some inflammatory diseases. Additionally, we observed strong linkage disequilibrium (D′ > 0.94) between the two studied polymorphisms (−20 G/A and −44 C/G), and one haplotype (AC) was associated with increased risk of developing T2DM (OR = 4.39, p = 0.04). As can be seen, this haplotype bears the C allele, that was associated with the disease in the independent analysis of the polymorphisms. This result corroborates the participation of this allele as a risk factor in the developing of the T2DM. Nonetheless, we considered that our findings, as well as the reported by Nemeth et al. (2014), need to be corroborated in additional studies with a larger number of individuals in different populations. This could help to further define the true role of this polymorphism as a marker of risk for developing T2DM. Moreover, we explored in silico the functional effect of the DEFB1 5′UTR-20 G/A and DEFB1 5′UTR-44 C/G polymorphisms. Although the 5′UTR-20 G/A SNP was not associated with T2DM, the analysis in silico showed that the A allele produces a binding motif for the BRCA1 transcription factor. The BRCA1 tumor suppressor is a protein that has an important role in multiple biochemical and biological functions such as DNA repair, replication, cell cycle checkpoints, protein ubiquitination, transcriptional regulation, chromatin remodeling, and apoptosis (Takoaka & Miki, 2018). Recently, Kang et al. (2012) reported that BRCA1 regulates the expression of insulin-like growth factor receptor-1 (IGF-IR) in human breast cancer cell line (MCF7). In line with this data, Shukla et al. (2006) reported that the absence of BRCA1 results in increased expression of IRS-1 (insulin receptor substrate-1) and increased levels of IGF-1R in BRCA1-deficient mice, and in cultured human cells. On the other hand, our in-silico analysis showed that the −44 C allele produces a binding site for the Ikaros (IK) transcription factor; IK factor regulates the expression of several other transcription factors as Cdx4 that participates in the activation of the STAT4, and in the regulation of the expression of several pro-inflammatory cytokines (John & Ward, 2011). In addition, it has been recently showed that the IK factor down-regulates to IGF-IR, which has an important role as a risk factor for T2DM (Singleton & Feldman, 2001). Additional to this information, Sun et al. (2006) reported that the C allele of the 668 C/G (named also −44 C/G) polymorphism increased the transcriptional activity of β-defensin 1 promoter in luciferase assays, suggesting that this allele has an important role in the overexpression of the β-defensin 1. Accordingly, with these data, we suggested that the carriers of the −44 C allele could increase binding of the IK factor, and in consequence, (a) affect the regulation the insulin-like growth factor receptor (IGF-IR), and (b) heighten expression of pro-inflammatory cytokines. On the other hand, recent studies showed the β-defensin 1 has an important role as chemoattractant molecule of macrophages through its interaction with CCR6 (Soruri et al., 2007;Yang et al., 1999). In this context, it could be speculated that the β-defensin 1 may be involved in the increased infiltration of the macrophages to pancreatic islets, accelerating the β-cell dysfunction. However, the functional consequence of this polymorphism deserves to be specifically addressed in future studies.
We recognize that our study has limitations as the limited size of the sample in the groups of study. The functional effect of the polymorphisms only was determined by informatics tools, so experimental designs are needed in order to corroborate this functional effect. In spite of these limitations, our study contributes to a new argument in which the 5′UTR −44 C/G polymorphism may have a role as a risk factor for T2DM.
In summary, this study demonstrates that the C allele of 5′ UTR −44 C/G polymorphism is associated with the presence of T2DM in the Mexican population. In addition, it was possible to distinguish one haplotype (AC) associated with the presence of T2DM. Nonetheless, due to the number of individuals included in our study and the specific genetic characteristics of the Mexican population, we considered that additional studies in a larger number of individuals and in other populations with different ethnic origins could help to define the true role of this polymorphism as susceptibility marker for developing T2DM and others inflammatory events.