CD36 Gene Variants in Early Prediction of Type 2 Diabetes Mellitus

Abstract

Type 2 diabetes (T2DM) is a noncommunicable disease affecting huge populations in India and abroad. Single-nucleotide polymorphisms (SNPs) in CD36, a macrophage scavenger receptor, have been implicated in the pathogenesis of T2DM and its complications. Eleven SNPs in the CD36 gene and their association with 100 each of control subjects and T2DM patients were investigated in the present study. The haplotype analysis of a few significant SNPs was carried out in individuals from families with diabetic history to evaluate its utility in disease prediction. Polymerase chain reaction-restriction fragment length polymorphism was used for genotyping. Ten families with a family history of diabetes were identified and blood samples were collected from as many family members as possible. Genotyping of three SNPs, namely rs1761667 (G>A) in exon 1 A, rs3211938 (T>G) in exon 10, and rs3212018 (16 bp del) in exon 14, was performed in all samples. Our results suggested that individuals having a GATTC1 haplotype might be at risk of developing T2DM (p<0.001) and, therefore, might be susceptible to related complications. Moreover, the presence of A, G, and G alleles of SNPs rs1761667 (G>A), rs3211938 (T>G), and rs1984112 (T>G) tends to have increased BMI, respectively. Such studies may be helpful for disease prediction in individuals at risk of T2DM. The predictive potential of CD36 variants can be explored with more families in the study population to use this as a genetic marker.

Introduction

Type 2 diabetes mellitus (T2DM) is one of the most prevalent diseases across the world, characterized by hyperglycemia resulting from a combination of insulin resistance, its inadequate secretion, and excessive or inappropriate glucagon secretion. T2DM is related to both genetic and nongenetic components, for example, ethnicity, lack of proper diet, and physical activity. Several common factors are found to be prevalent with T2DM in Asian countries. India is a developing country with fast industrialization and modern lifestyle. The current estimate of diabetics in India is more than 63 million and is expected to increase in the next 20 years if preventive programs are not put in place. Untreated diabetes leads to various complications such as nephropathy, retinopathy, kidney failure, atherosclerosis, and cardiac failure. Identification of genetic factors responsible for different disease conditions would help to target people at risk and take timely action to delay the onset of disease.

The CD36 gene is located in chromosome 7 at locus q11.2 with 17 exons and 16 introns, which cover approximately 36 Kb region of DNA (Silverstein et al., 1992). CD36 is a plasma membrane glycoprotein constituted by a single 88-kDa polypeptide chain of 472 amino acid residues (Greenwalt et al., 1992; Gautam and Banerjee, 2010). CD36 is widely distributed in a variety of cell types especially on macrophages, monocytes, platelets, adipocytes, etc. CD36 plays a major role in adhesion of ligand, uptake of fatty acids, and signal transduction. In obesity and hyperlipidemia, these ligands affect inflammatory and insulin signaling pathways in a pathophysiological response using CD36 (Kennedy et al., 2011). In macrophages, the uptake of fatty acids especially the oxidized form of low-density lipoprotein (Ox-LDL) leads to foam cell formation, plaque formation, and ultimately atherosclerosis. Since lipid metabolism is closely associated with diabetes and, in turn, atherosclerosis, the study of the CD36 locus was undertaken. Several studies on gene variants were undertaken (Banerjee et al., 2010; Gautam et al., 2011, 2013a, 2013b; Rać et al., 2012), which led to our investigation of selected CD36 single nucleotide polymorphisms (SNPs) and haplotypic association with T2DM in a North Indian population.

Materials and Methods

Study design and participants

The case-control study was approved by the Institutional Ethics Committee (King George's Medical University, Lucknow, India), and informed consent was obtained from all participants. Medical records of all patients were reviewed to ascertain diabetes-associated complications. The study subjects selected according to the inclusion and exclusion criteria (Gautam et al., 2013a) were >40 years of age.

Technical information

SNP selection and genetic analysis

Genomic DNA was extracted from blood samples of T2DM patients and controls (n=100 each), using the salting-out method with slight modifications and checked on 1% agarose gels (Miller et al., 1988; Gautam et al., 2011). The quality and quantity of DNA were estimated using a biophotometer (Eppendorf).

Eleven SNPs, namely rs1984112, A>G in exon 1A; rs1761667, −31118G>A in 5′ promoter region of exon 1A; −178A>C in promoter region; rs1527479, C>T in intron IB; 478C>T in exon 4; AC dinucleotide del in exon 5; rs3211938, 1264T>G in exon 10; Ins 1159A at 317 codon in exon 11 (CI962235); rs1527483, 25444C>T in intron 11; rs3212018, 16 bp del in exon 14; and rs1049673 G>C in exon 15, were identified in the CD36 gene after surveying literature and SNPdb (Table 1). The sample size for each SNP was determined using QUANTO software (v.1.2.1) after genotyping 100 control individuals and minor allele frequencies (MAFs) were calculated. SNPs with MAF≥0.01 were included in the haplotype analysis. SNPs were genotyped using polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) (Banerjee et al., 2010).

Table 1.

Position and Minor Allele Frequency of Eleven SNPs in a North Indian Population

rs number	Type of polymorphism	Site of polymorphism	MAF
Pro90 Ser mutation	478 C>T	Exon 4	<0.01
Dinucleotide deletion	DelAC	Exon 5	<0.01
rs1527483	C>T	25444 Intron 11	0.01
rs1761667	−31118 G>A	−31118 Promoter 5′ region of exon 1	0.11
rs 3211938	1264T>G	Exon 10	0.03
SNP	−178A>C	Promoter region	<0.01
rs 1527479	C>T	Intron IB	0 .03
rs 1984112	A>G	Exon 11	0.32
CI962235	Ins 1361A	At 317 codon in exon 10	<0.01
rs3212018	16 bp del	Exon 14	0.01
rs1049673	G>C	Exon 15	<0.01

MAF, minor allele frequency; SNP, single-nucleotide polymorphism.

Statistical analysis

Comparison of all clinical parameters between controls and T2DM patients was carried out by the t-test. Allelic and genotypic frequencies along with carriage rates of polymorphic alleles in patient and control groups were evaluated by Pearson's chi-square (χ²) test/Fisher's exact test. The Hardy-Weinberg equilibrium at individual locus was assessed by chi-square (χ²) statistics using SPSS software (v. 15.0). All p-values<0.05 were considered statistically significant. The odds ratio (OR) at 95% confidence interval (CI) was determined to describe the strength of association by the logistic regression model.

Haplotype analysis

Haplotype analysis was performed for all six SNPs whose MAF ≥0.01. The SNPs taken were G>A (rs1761667), C>T (rs1527483), A>G (rs1984112), C>T (rs1527479), T>G (rs3211938), and 16 bp del (rs3212018). Multivariate logistic regression analysis was carried out to analyze haplotypes. Pairwise linkage disequilibrium (LD) based on D′ statistic for the six SNPs was calculated using SHEsis software. LD blocks were defined using their default algorithm (Shi and He, 2005).

Pedigree analysis

Ten families with a family history of diabetes were identified and blood samples were collected from the available family members and DNA was extracted. The three SNPs, namely rs1761667 (G>A) in exon 1 A, rs3211938 (T>G) in exon 10, and rs3212018 (16 bp del) in exon 14, which showed significant association with T2DM in the North Indian population (Banerjee et al., 2010; Gautam et al., 2011, 2013a, 2013b), were genotyped in all individuals by PCR-RFLP. The biochemical profile of each individual was correlated with their genotypes.

Results

Prevalence of the disease was taken as 3.0% and power of the study as 80.0%. The biochemical analysis in patients (n=450) and controls (n=400) has been reported earlier (Gautam et al., 2011).

Genetic analysis

Out of eleven SNPs studied, five, namely −178A>C in promoter region; 478C>T in exon 4; AC dinucleotide del in exon 5; Ins 1159A at 317 codon in exon 11 (CI962235); and rs1049673 G>C in exon 15, were monomorphic (MAF<0.01) and therefore were not analyzed further (Table 1). Polymorphic studies of the remaining six SNPs (rs1984112, rs1761667, rs1527479, rs3211938, rs1527483, and rs3212018) in controls and patients (n=100 each) showed them to be in the Hardy-Weinberg equilibrium.

The GA heterozygous genotype of the promoter polymorphism G>A (rs1761667) in the CD36 gene was more prevalent in T2DM patients and showed an association with T2DM in the North Indian population (Banerjee et al., 2010). However, TG genotype of SNP T>G (rs3211938) had a mild significance (p=0.045) and seemed to confer susceptibility to T2DM (Gautam et al., 2013a). We also analyzed CI962235 (Ins1159A) and rs3212018 (16 bp del) SNPs, but did not observe any significant association (Gautam et al., 2013b).

Haplotype analysis

Analysis of six SNPs taken together, that is, rs1984112 (A>G), rs1761667 (G>A), rs1527479 (C>T), rs3211938 (T>G), rs1527483 (C>T), and rs3212018 (16 bp del) showed thirty-four different combinations of haplotypes. Among them, the G A T T C 1* showed the highest significant association (p<0.001) with T2DM (OR: 6.606 [2.443-17.861]). Moreover, few other haplotypic combinations also showed significant association with T2DM (p<0.05) (Table 2).

Table 2.

Haplotype Combinations of Six SNPs and Their Association with Controls and T2DM Patients

Haplotype	Patients (freq)	Controls (freq)	χ²	p-value	Odds ratio (95% CI)
A A C G C 1	0.00 (0.000)	0.25 (0.001)	—	—	—
A A C G C 2	0.00 (0.000)	5.20 (0.026)	5.494	0.019	—
A A C G T 2	0.00 (0.000)	2.48 (0.012)	2.603	0.107	—
A A C T C 1	7.91 (0.040)	7.08 (0.035)	0.019	0.891	1.075 (0.382-3.023)
A A T G C 1	2.08 (0.010)	1.59 (0.008)	0.046	0.830	1.254 (0.158-9.957)
A A T G T 1	0.00 (0.000)	1.44 (0.007)	—	—
A A T T C 1	27.94 (0.140)	18.53 (0.093)	1.734	0.188	1.518 (0.813-2.833)
A G C G C 1	14.24 (0.071)	11.70 (0.059)	0.163	0.686	1.179 (0.530-2.624)
A G C G C 2	0.00 (0.000)	5.10 (0.026)	5.393	0.020	—
A G C G T 1	0.01 (0.000)	5.31 (0.027)	4.751	0.029	0.001 (0.000-0.024)
A G C T C 1	0.00 (0.000)	2.28 (0.011)	1.693	0.193	0.001 (0.000-0.017)
A G C T T 1	0.00 (0.000)	0.01 (0.000)	—	—	—
A G T G C 1	17.36 (0.087)	9.24 (0.046)	2.301	0.129	1.876 (0.823-4.278)
A G T G C 2	6.94 (0.035)	7.59 (0.038)	0.066	0.798	0.872 (0.305-2.489)
A G T T C 1	5.89 (0.029)	31.20 (0.156)	20.285	<0.001	0.156 (0.063-0.386)
G A C G C 1	1.48 (0.007)	2.67 (0.013)	0.398	0.528	0.528 (0.070-3.969)
G A C G C 2	0.00 (0.000)	1.92 (0.010)	—	—	—
G A C T C 1	0.00 (0.000)	0.00 (0.000)	—	—	—
G A T G C 1	0.00 (0.000)	16.83 (0.084)	17.393	<0.001	0.000 (0.000-0.002)
G A T G T 1	0.00 (0.000)	2.57 (0.013)	2.697	0.101	—
G A T T C 1^*	28.64 (0.143)	4.73 (0.024)	17.639	<0.001	6.606 (2.443-17.861)
G A T T C 2	0.00 (0.000)	4.72 (0.024)	4.981	0.026	—
G G C G T 1	0.00 (0.000)	1.00 (0.005)	—	—	—
G G T G C 1	20.13 (0.101)	9.62 (0.048)	3.549	0.059	2.118 (0.956-4.691)
G G T G C 2	0.00 (0.000)	2.00 (0.010)	—	—	—
G G T G T 1	0.00 (0.000)	4.49 (0.022)	4.735	0.029	—
G G T T C 1	4.41 (0.022)	38.76 (0.194)	32.381	<0.001	0.089 (0.033-0.244)
G G T T T 1	0.00 (0.000)	1.70 (0.009)	—	—	—
A A T T C 2	3.54 (0.018)	0.00 (0.000)	3.422	0.064	—
A A T T T 1	12.20 (0.061)	0.00 (0.000)	12.066	<0.005	—
A A T T T 2	3.21 (0.016)	0.00 (0.000)	3.104	0.078	—
A G C G T 2	0.02 (0.000)	0.00 (0.000)	—	—	—
A G T G T 2	2.49 (0.012)	0.00 (0.000)	2.397	0.122	—
A G T T C 2	4.18 (0.021)	0.00 (0.000)	4.046	0.044	—
G G C G C 1	31.63 (0.158)	0.00 (0.000)	32.978	<0.001	—
G G C G C 2	2.63 (0.013)	0.00 (0.000)	2.536	0.111	—
G G C T T 1	3.08 (0.015)	0.00 (0.000)	2.976	0.085	—

Significance of bold values is to show the haplotypes whose p-values are <0.05.

Haplotype with significant p-value and high odds ratio.

The LD patterns according to the position of variants within the gene, which showed the strongest statistical significance of LD between two SNPs, were rs3211938 (T>G): rs1527483 (C>T) (frequency=0.725; p<0.01); however, a few other SNP combinations also showed a high significant association with T2DM [rs1527479 (C>T): rs3211938 (T>G) (frequency=0.702; p<0.01) and rs1761667 (G>A): rs3211938 (T>G) (frequency=0.728; p<0.01)] (Fig. 1).

FIG. 1.

Linkage Disequilibrium (D′) of six single-nucleotide polymorphisms in the CD36 gene in a North Indian population with T2DM. Pairwise LD in subjects is represented as dark gray squares for strong LD, and light gray squares for little or no LD, linkage disequilibrium; T2DM, type 2 diabetes mellitus.

Pedigree analysis

The three SNPs reported earlier showed similar results in family studies. Out of 10, one large family has been described, in which subjects II-1, II-3, II-4, II-8, and II-9 were diabetic at the time of sample collection (Fig. 2). The individuals II-1, II-3, and II-9 were related and all were between 50 and 65 years of age (Fig. 2). Their random glucose levels were 41, 114, 102, 124, and 132 mg/dL, respectively. Most of the individuals of the family were heterozygous GA and TG for the SNPs rs1761667 (G>A) and rs 3211938 (T>G), respectively. Those individuals who have the A allele of rs1761667 and the G allele of rs3211938 have a higher BMI than other members, for example, II-4, II-8, and II-9. It was also observed that some individuals who were not yet diabetic but possessed similar alleles showed a higher BMI (II-5, II-7, III-3, and III-5).

FIG. 2.

Pedigree analysis: A family with four generations (I, II, III, IV); individuals II1, 3,4,8, and 9 are diagnosed with T2DM; * Individuals with GA genotype of rs1761667 showing increased BMI; I-1 is deceased; NA, not available for sampling. Circle, female; square, male; open symbols, normal, healthy individuals; dark symbols, patients with T2DM; BMI, body mass index.

Discussion

CD36 recognizes and binds to specific Ox-LDL and initiates an intracellular signaling cascade in macrophages. Its uptake finally leads to the formation of plaques and atherosclerosis (Gautam and Banerjee, 2010). A substantial number of studies have investigated the association of CD36 with certain disorders such as abnormal serum LDLs (Goyenechea et al., 2008; Morii et al., 2009), fatty acids (Ma et al., 2004), coronary artery disease (Doris, 2002; Kuang et al., 2004; Ma et al., 2004; Yun et al., 2007), left ventricular mass (Arnett et al., 2009; Hall et al., 2011), metabolic syndrome (Leprêtre et al., 2004; Noel et al., 2010), insulin resistance (Heilbronn et al., 2007; Kashyap et al., 2009), low adiponectin levels (Leprêtre et al., 2004), and stroke (Ikram et al., 2009). Recent studies on a Chinese population showed the association of CD36 gene variants with triglycerides and high-density lipoprotein (Chien et al., 2012). Another study reported a significant polymorphic association of the CD36 gene with essential hypertension and systolic blood pressure in a Chinese population (Liu et al., 2013). Our results suggested that individuals having a GA genotype might be susceptible to T2DM and may be at risk of developing related complications (Banerjee et al., 2010). A few of our other studies have also suggested the role of CD36 variants in the development of T2DM (Gautam et al., 2013a, 2013b).

These alleles may also act as a genetic marker in early age diagnosis of T2DM such as subject III-5 in our family study. At present, he is only 14 years of age and has heterozygous genotypes GA and TG of rs1761667 (G>A) and rs3211938 (T>G), respectively (Fig. 2), and an increased BMI, which might indicate a high risk of developing T2DM in the future.

Moreover, the presence of the G allele in the TG genotype of SNP T>G (rs3211938) and the G allele of SNP T>G (rs1984112) appeared to be indicators of increased BMI and, therefore, can be used as a markers for early prediction of T2DM. Another interesting finding from haplotype studies was that individuals with G A T T C 1 (six SNP loci) (Fig. 1) showed a highly significant (p<0.001) association with a six to seven times higher risk of developing T2DM in the near future.

The genotype results led to the conclusion that CD36 variants have a high predictive potential on determining the disease susceptibility in the North Indian population. However, further follow-up studies are essential to determine what changes occur at molecular levels for these SNPs, which may help in correct diagnosis and treatment of T2DM.

Footnotes

Acknowledgments

The authors would like to thank the Department of Biotechnology (DBT), the Department of Science & Technology (DST), the Indian Council of Medical Research, New Delhi, India, and the Centre of Excellence, Higher Education, Government of Uttar Pradesh, Lucknow, India, for funding the work. SG would like to acknowledge the Rajiv Gandhi National Fellowship, University Grants Commission, New Delhi, India.

Author Disclosure Statement

No competing financial interests exist.

References

Arnett

, Li

, Tang

, et al. (2009) Genome-wide association study identifies single-nucleotide polymorphism in KCNB1 associated with left ventricular mass in humans: the HyperGEN Study. BMC Med Genet, 10:43.

Banerjee

, Gautam

, Saxena

, et al. (2010) Association of CD36 gene variants rs1761667 (G>A) and rs1527483 (C>T) with Type 2 diabetes in North Indian population. Int J Diabetes Mellit, 2:179-183.

Chien

K-L

, Hsu

H-C

, Liu

P-H

, et al. (2012) Common sequence variants in CD36 gene and the levels of triglyceride and high-density lipoprotein cholesterol among ethnic Chinese in Taiwan. Lipids Health Dis, 11:174.

Doris

(2002) Hypertension genetics, single nucleotide polymorphisms, and the common disease: common variant hypothesis. Hypertension. 323-331.

Gautam

, Agrawal

, Bid

, Banerjee

(2011) Preliminary studies on CD36 gene in type 2 diabetic patients from north India. Indian J Med Res, 134:107-112.

Gautam

, Banerjee

(2011) The macrophage Ox-LDL receptor, CD36 and its association with type II diabetes mellitus. Mol Genet Metab, 102:389-398.

Gautam

, Pirabu

, Agrawal

, Banerjee

(2013a) CD36 gene variants and their association with type 2 diabetes in an Indian population. Diabetes Technol Ther, 15:680-687.

Gautam

, Agarwal

, Banerjee

(2013b) Study of CI962235 (Ins1361A), rs3212018 (16 bp del) and rs1049673 (G>C) CD36 gene Polymorphisms in T2DM patients of North India. J Med Sci, 13:439-445.

Goyenechea

, Collins

, Parra

, et al. (2008) CD36 gene promoter polymorphisms are associated with low density lipoprotein-cholesterol in normal twins and after a low-calorie diet in obese subjects. Twin Res Hum Genet, 11:621-628.

10.

Greenwalt

, Lipsky

, Ockenhouse

, et al. (1992) Membrane glycoprotein CD36: a review of its roles in adherence, signal transduction, and transfusion medicine. Blood, 80:1105-1115.

11.

Hall

, Mayosi

, Rahman

, et al. (2011) Common variation in the CD36 (fatty acid translocase) gene is associated with left-ventricular mass. J Hypertens, 29:690-695.

12.

Heilbronn

, Gregersen

, Shirkhedkar

, et al. (2007) Impaired fat oxidation after a single high-fat meal in insulin-sensitive nondiabetic individuals with a family history of type 2 diabetes. Diabetes, 56:2046-2053.

13.

Ikram

, Seshadri

, Bis

, et al. (2009) Genomewide association studies of stroke. N Engl J Med, 360:1718-1728.

14.

Kashyap

, Ioachimescu

, Gornik

, et al. (2009) Lipid-induced insulin resistance is associated with increased monocyte expression of scavenger receptor CD36 and internalization of oxidized LDL. Obesity (Silver Spring), 17:2142-2148.

15.

Kennedy

, Kuchibhotla

, Westfall

, et al. (2011) A CD36-dependent pathway enhances macrophage and adipose tissue inflammation and impairs insulin signalling. Cardiovasc Res, 89:604-613.

16.

Kuang

, Febbraio

, Wagg

, et al. (2004) Fatty acid translocase/CD36 deficiency does not energetically or functionally compromise hearts before or after ischemia. Circulation, 109:1550-1557.

17.

Leprêtre

, Linton

, Lacquemant

, et al. (2004) Genetic study of the CD36 gene in a French diabetic population. Diabetes Metab, 30:459-463.

18.

Liu

, Meng

, Yang

(2013) Association study of CD36 single nucleotide polymorphisms with essential hypertension in the Northeastern Han Chinese. Gene, 527:410-415.

19.

, Bacci

, Mlynarski

, et al. (2004) A common haplotype at the CD36 locus is associated with high free fatty acid levels and increased cardiovascular risk in Caucasians. Hum Mol Genet, 13:2197-2205.

20.

Miller

, Dykes

, Polesky

(1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 16:1215.

21.

Morii

, Ohno

, Kato

, et al. (2009) CD36 single nucleotide polymorphism is associated with variation in low-density lipoprotein-cholesterol in young Japanese men. Biomarkers, 14:207-212.

22.

Noel

, Lai

, Mattei

, et al. (2010) Variants of the CD36 gene and metabolic syndrome in Boston Puerto Rican adults. Atherosclerosis, 211:210-215.

23.

Rać

, Suchy

, Kurzawski

, et al. (2012) Polymorphism of the CD36 gene and cardiovascular risk factors in patients with coronary artery disease manifested at a young age. Biochem Genet, 50:103-111.

24.

Shi

, He

(2005) SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res, 15:97-98.

25.

Silverstein

, Baird

, Lo

, Yesner

(1992) Sense and antisense cDNA transfection of CD36 (glycoprotein IV) in melanoma cells. Role of CD36 as a thrombospondin receptor. J Biol Chem, 267:16607-16612.

26.

Yun

, Song

, et al. (2007) CD36 polymorphism and its relationship with body mass index and coronary artery disease in a Korean population. Clin Chem Lab Med, 45:1277-1282.