Association between the MTNR1B, HHEX, SLC30A8, and TCF7L2 single nucleotide polymorphisms and cardiometabolic risk profile in a mixed ancestry South African population

Single nucleotide polymorphisms of the TCF7L2, HHEX, SLC30A8, MTNR1B, SLC2A2 and GLIS3 genes are well established candidate genes for cardiometabolic diseases (CMDs) across different ethnic populations. We investigated their association with CMDs in a mixed ancestry population of South Africa. rs10830963, rs1111875, rs11920090, rs13266634, rs7034200 and rs7903146 SNPs were genotyped by quantitative real time PCR in 1650 participants and Hardy–Weinberg equilibrium (HWE) analyses performed on the SNPs. Diabetes, obesity, hypertension and cardiometabolic traits were compared across genotypes of SNPs in HWE. Linear and logistic regressions adjusting for age, gender and body mass index were used to determine the risk of T2DM, obesity and hypertension. rs7903146 (p = 0.055), rs1111875 (p = 0.465), rs13266634 (p = 0.828), and rs10830963 (p = 0.158) were in HWE. The rs10830963 recessive genotype was able to predict FPG, insulin and HOMA-IR, while the rs1111875 recessive genotype was able to predict total cholesterol, triglyceride, LDL cholesterol and FPG. The rs7903146 recessive genotype was able to predict SBP and LDL cholesterol. The recessive genotypes of MTNRIB and HHEX SNPs were associated with T2DM traits in the study population and could partially explain the high prevalence of T2DM. Further studies are required to confirm these findings and establish candidate genes in the African population.


Methodology Study design and population
This cross sectional study used data from the Cape Town Vascular and Metabolic Health (VMH) study consisting of 1969 male and female participants.They were recruited during a community based survey involving mixed ancestry South Africans residing in the township of Bellville South in Cape Town as previously described 24 .The mixed ancestry South African population is a well-defined, multiracial ethnic group that comes from a combination of ethnic backgrounds including indigenous South Africans, Caucasians, Griquas and Asians 25 .Ethical approval was obtained from the Research Ethics Committees of the Cape Peninsula University of Technology and Stellenbosch University (NHREC: REC-230 408-014 and N14/01/003) and the study was conducted in accordance with the principles of the Declaration of Helsinki.The city and community authorities granted permission for the study to be conducted in their municipality and all study participants gave written informed consent before they were included.

Data collection and sampling
Data including demographic characteristics, personal and family medical history, ongoing treatment and smoking habits were collected by trained fieldworkers using a standardized questionnaire on a password-protected personal digital assistant.Physical examination including body weight in kilogram (kg), height in centimetres (cm), waist circumference (WC), hip circumference (HC), systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured in triplicate as previously described 24 with the lowest values used for analysis.After physical examination, blood samples were collected by trained registered nurses from all participants and stored at − 80 °C for biochemical analysis and SNP genotyping.This was closely followed by an oral glucose tolerant test (OGTT) with a 75 g of glucose load in accordance with the World Health Organization (WHO) recommendations.The OGTT test was not performed in patients with known T2DM.

Biochemical analyses
Fasting plasma glucose (FPG) was measured by enzymatic hexokinase method (Beckman Coulter), insulin by a paramagnetic particle chemiluminescence assay (Beckman Coulter) and glycated haemoglobin (HbA1c) by high performance liquid chromatography (Biorad Variant Turbo, BioRad).Total cholesterol (TC) and high-density lipoprotein cholesterol (HDL-C) were measured by enzymatic immune-inhibition, triglycerides (TG) by glycerol phosphate oxidase-peroxidase and low-density lipoprotein cholesterol (LDL-C) by enzymatic selective protection-End Point (Beckman Coulter).A modified Jaffe-Kinetic method was used to measure serum creatinine (Beckman AU, Beckman Coulter, South Africa).Highly sensitive c-reactive protein (hs-CRP) was measured by Enzyme Linked Immuno-Sorbent Assay (ELISA) (BIOMATIK) according to the manufacturers protocol.

DNA extraction and genotyping
Deoxyribonucleic Acid (DNA) was extracted using the salt extraction method 26 and the concentration/purity determined using the Nanodrop™ One (Thermo Fisher Scientific, Waltham, MA, USA).All samples with a ratio of absorbance at 260 nm and 280 nm between 1.7 and 2.2 were considered good quality and used for SNP genotyping by real time quantitative polymerase chain reaction (RT-qPCR).Six SNPs (rs10830963, rs1111875, rs11920090, rs13266634, rs7034200 and rs7903146) were genotyped with a TaqMan® Universal PCR Master www.nature.com/scientificreports/Mix, according to the manufacturer's protocol.Enough PCR master mix for the requisite number of reactions was made in a 1.5 ml tube as shown in Table 1, pulse vortexed and then briefly centrifuged.Afterwards, 8 µL of the master mix were pipetted into each of the 96-well plate's wells. 2 µL of the template DNA sample (5 ng/µL concentration) were then pipetted into the appropriate well.For quality control purposes, a non-template control was also included in each PCR run, with 2 µL of ddH 2 0 used in place of template DNA.When all components of the PCR had been pipetted into the appropriate wells, the 96-well plate was covered with MicroAmp optical caps, and the PCR run on the Applied Biosystems® Quant Studio™ 7 Flex Real-time PCR system (ThermoFisher Scientific) as shown in Fig. 1.
For each SNP, the three genotypes were determined from the PCR amplification output as follows: a sample was homozygous for the dominant allele when amplification curves were observed only in the VIC channel, homozygous for the recessive allele if an amplification signal in the ROX channel was observed, whilst amplification in both channels indicated a heterozygous genotype.Reagents and primers (TaqMan® SNP Genotyping Assays) were sourced from ThermoFisher Scientific.
The accuracy of SNP genotyping using RT-qPCR was confirmed by comparing genotyping results of 20 randomly selected samples sequenced by Inqaba Biotechnical Industries.

Statistical analysis
Data were analysed using the IBM Statistical Package for Social Sciences (SPSS) version 29 software.Categorical variables were reported as counts (percentages) and continuous variables reported as median (25-75th percentiles).Hardy-Weinberg equilibrium (HWE) was tested for all the six SNPs and those in HWE were used for further analysis.Diabetes, obesity, hypertension, dyslipidemia and cardiometabolic traits (SBP, DBP, TC, HDL-C, TG, LDL-C, hs-CRP, FPG, insulin, HOMA-IR, HOMA-B and HbA1c ) were compared across genotypes of SNPs in HWE using the chi-square goodness-of-fit test for categorical variables and median test for continuous variables given that the variables were skewed.Linear regression analyses was performed to test the interactions between cardiometabolic traits and the genotypes of the SNPs investigated while adjusting for age, gender and  www.nature.com/scientificreports/

Association between risk genotype and cardiometabolic traits/diseases
Clinical characteristics were compared between the genotypes of the four SNPs in HWE.The results showed that pulse (p = 0.005), fasting plasma glucose (p = 0.01), insulin (p = 0.003), HbA1c (p = 0.037), HOMA-IR (p = 0.006), HOMA-B (p = 0.021) and total cholesterol (p = 0.043) were significantly higher in carriers of the recessive GG genotype of rs10830963 SNP when compared to the wild CC genotype (Table 4).Results of the rs1111875 SNP   www.nature.com/scientificreports/revealed that individuals with the recessive TT genotype had lower cotinine values compared to those with the dominant CC genotype and the difference was statistically significant (p = 0.014).Moreover, the number of participants with high TC levels were significantly higher amongst carriers of the TT genotype when compared to carriers of the CC genotype (p = 0.03).There was a higher number of individuals with hypertension (p = 0.034) and dyslipidemia characterized by high triglycerides (p = 0.018) amongst carriers of the TT recessive genotype of the rs7903146 SNP compared to carriers of the CC genotype (Table 5).

Linear and logistic regression analysis for the association between single nucleotide polymorphisms and cardiometabolic traits
Linear regression analyses were carried out to determine whether carrying the recessive genotypes of the SNPs investigated could increase the risk of quantitative traits of cardiometabolic diseases while adjusting for age, gender, and BMI (Table 6).The results showed that expression of the recessive genotype of rs10830963 SNP may be suggestive of an increased risk of dysglycemia characterized by increased fasting plasma glucose (β = − 0.  HOMA-IR 1.5 (0.9-2.3) 1.5 (0.9-2.5) 1.4 (0.9-2.9) 0.909 1.5 (0.9-2.4) 1.5 (0.9-2.www.nature.com/scientificreports/p = 0.001 compared to the homozygote dominant genotype).The recessive allele of the rs13266634 SNP was not associated with cardiometabolic traits risks, while the recessive genotype of rs7903146 SNP increased the risk of high blood pressure characterized by high systolic blood pressure (β = − 4.679, p = 0.022 compared to the heterozygote genotype).Logistic regression analyses adjusting for age, gender and BMI were carried out to evaluate the association between the recessive genotypes of the SNPs and T2DM, obesity, hypertension and dyslipidemia characterized by high TC, high TG, low HDL-C and high LDL-C (Table 6).From these results, the odds of hypercholesterolemia reduced by 33% in carriers of the homozygote dominant genotype (95% CI = 0.49-0.92,p = 0.013) and 34% in carriers of the heterozygote genotype (95%CI = 0.48-0.90,p = 0.009) when compared to carriers of the risk genotype of rs1111875 SNP.Moreover, the odds of high LDL-C reduced by 40% in carriers of the heterozygote genotype (95% CI = 0.39-0.93,p = 0.022) of the rs7903146 SNP when compared to carriers of the risk genotype (Table 7).

Discussion
This study examined the association between six SNPs and cardiometabolic risk profile in a mixed ancestry South African population.The TT recessive genotype of the rs7903146 SNP was significantly higher in patients with hypertension when compared to the CC genotype.While there was no association with diabetes, obesity and dyslipidemia, the SNPs of the TCF7L2 gene (rs7903146), HHEX gene (rs1111875) and MTNR1B gene (rs10830963) were associated with cardiometabolic traits.Individuals with the recessive genotype of rs10830963 SNP had significantly higher fasting plasma glucose (FPG), insulin and HOMA-IR values relative to carriers of the dominant genotype.More participants with the rs1111875 recessive genotype had high total cholesterol and LDL cholesterol compared to those with the dominant genotype.After adjusting for covariates (age, gender, and BMI), the rs10830963 recessive genotype was associated with FPG, insulin and HOMA-IR.The rs1111875 recessive genotype was associated with total cholesterol, triglyceride, LDL cholesterol and fasting glucose, whilst the rs7903146 recessive genotype was associated with systolic blood pressure and LDL cholesterol.
People living in the same environment experience different risk for T2DM and other metabolic diseases, partly due to their genetic differences.Single nucleotide polymorphisms, found in at least one percent of the population, are the most common type of genetic variation that exists, and they occur when one nucleotide in a DNA sequence is replaced by another.Several SNPs have been used as biological markers to identify genes that are associated with diseases.The candidacy gene approach and GWAS have led to the identification of several gene variants associated with T2DM via different metabolic pathways.For instance, the rs10830963 SNP of the Melatonin receptor 1B gene and rs13266634 of the Solute carrier family 2-member 2 encode for proteins which affect insulin secretion and glucose homeostasis pathways, leading to the development of T2DM 20,30 .On the other hand, rs7903146 and rs1111875 are SNPs of the TCF7L2 and HHEX genes which encode transcription factors involved in the wnt signaling pathways.While rs7903146 mainly affects the insulin sensitivity pathway by the regulation of genes involved in lipid and glucose metabolism 31 , rs1111875 is involved in beta cell function via the reduction of cell mass and decrease in beta-cell secretory capacity 18 .Our study reported similar findings on the association between cardiometabolic traits and two of the studied SNPs (rs10830963 and rs1111875).The rs10830963 SNP was observed to affect the insulin resistance pathway, with no effect in beta cell dysfunction.Insulin resistance resulted in significantly higher fasting plasma glucose and insulin levels in the carriers of the risk genotypes.The association of the rs1111875 recessive genotype with dyslipidemia and hyperglycemia in our study suggests the SNP affects insulin sensitivity rather than insulin secretion pathways.While no association was observed between rs10830963 and rs1111875 with these cardio-metabolic diseases, their association with traits of T2DM could be indicative that these SNPs are possible risk factors for T2DM.The absence of association with disease state has also been reported in studies carried out in Indian 32 and European populations 33 .Our study failed to replicate findings on the association between rs7903146 SNP of the TCF7L2 gene and T2DM and its  traits including glucose and insulin.The recessive TT genotype of the rs7903146 SNP is a well-known risk factor for T2DM in Caucasian and Asian populations.Moreover, studies in sub-Saharan Africa reported an association between rs7903146 and increased risk for T2DM in Ghana 34 and South Africa 35 .Our study was carried out in a mixed ancestry population which is genetically more diverse compared to the black African, Caucasian and Asian populations, and could be responsible for the difference in the findings.Moreover, our study did not find an association between rs13266634 SNP of the SLC30A8 gene and diseases states (diabetes, obesity and hypertension) or traits (plasma glucose, insulin, blood pressure or lipid profile), whose association with T2DM has been previously reported in Asian 36 and European populations 37 .As such, it is plausible that these SNPs confer risk to T2DM in specific populations and may not necessarily be of concern in the mixed ancestry South Africa population.Our study had some limitations, and this included its cross-sectional nature, which made it difficult to infer causality of disease.Moreover, the discordant findings reported in our study could be because of our study design, which differed from other studies which often investigate associations between SNPs and cardiometabolic diseases using case-control designs.Due to funding constraints, the genotypes were confirmed by sequencing in less than 1.5% of the samples, which is relatively small.Despite these limitations, the large sample size and the similar genetic background of our study participants significantly added to the growing pool of data on the genetics of T2DM in the mixed ancestry population.

Conclusion
Recessive genotypes of the rs1111875 SNP of the HHEX gene and the rs10830963 SNP of the MTNR1B gene may increase the risk of T2DM in the mixed ancestry South African population.This may be through coding for proteins which affect the insulin signaling pathway without an effect on beta cell function.Other studies confirming such findings in the same population as well as other populations are warranted to fully understand the genetic predisposition to T2DM in the South African population.

Table 1 .
Components of the PCR master mix.The 2× genotyping mix contains AmpliTaq Gold® DNA Polymerase, dNTPs, ROX™ passive reference and buffer components optimized for tight endpoint fluorescence clusters, reproducible allelic discrimination, and bench top stability, whilst the 40× SNP assay mix contains forward and reverse primers specific for the SNP and two TaqMan® MGB probes for detecting alleles.ddH 2 0double distilled water.

Volume per reaction (μL)
Figure 1.PCR cycling conditions for the SNP genotyping assay.

Table 3 .
Determination of single nucleotide polymorphism in hardy Weinberg equilibrium.Significant values are in bold.

Table 4 .
Clinical characteristics across genotypes of rs10830963 and rs1111875 single nucleotide

Table 6 .
Association between single nucleotide polymorphisms genotypes and cardiometabolic traits.