Association of APOB 3ʹ-VNTR alleles with type 2 diabetes, BMI, systolic and diastolic blood pressure Sajib AA1, Khan MAT2, Haque MN3, Kibria KMK4, Chowdhury AKA5, Yeasmin S6 Abstract Background: Apolipoprotein B (APOB) is a component of chylomicrons, low-density lipoproteins (LDL), very low-density 

Background: Apolipoprotein B (APOB) is a component of chylomicrons, low-density lipoproteins (LDL), very low-density lipoproteins (VLDL), intermediate density lipoproteins (IDL) and functions as the main protein for transporting cholesterol to peripheral cells. APOB gene has an AT-rich VNTR site at its 3ʹ-untranslated region (3ʹ-UTR). APOB 3ʹ-VNTR alleles with ≥36 repeats have been shown to be strongly associated with increased serum lipid levels, gallstone formation and coronary artery disease. Objectives: To investigate any possible association of APOB 3ʹ-VNTR alleles with type 2 diabetes, BMI, systolic and diastolic blood pressure. Materials and methods: APOB 3ʹ-VNTR region in the DNA of non-diabetic controls and type 2 diabetic patients were amplified by polymerase chain reaction (PCR) and the numbers of core repeat in the amplified products were determined. Frequencies of the APOB alleles and genotypes among the controls and the patients were calculated and statistical analyses were performed. Results and discussion: Here we report for the first time that APOB 3ʹ-VNTR alleles have different distribution frequencies among type 2 diabetic and non-diabetic individuals. We also observed higher body mass index (BMI), systolic and diastolic blood pressures in individuals who had at least one APOB 3ʹ-VNTR allele with ≥35 repeats. Conclusion: Our study might bridge among the genetic signature of APOB 3ʹ-VNTR, high APOB protein level in blood, diabetes and other co-morbidities.


Introduction
APOB protein is a component of chylomicrons, low-density lipoproteins (LDL), very low-density lipoproteins (VLDL), intermediate density lipoproteins (IDL) and functions as the main protein for transporting cholesterol to peripheral cells [1][2][3][4][5][6] .The plasma concentration of APOB protein reflects the total number of potentially atherogenic particles 2,4 and, therefore, is predictive of high risk of coronary heart disease that may not be otherwise detected from routine lipid profile 2,[7][8][9][10] .High APOB protein level in blood is correlated with cerebrovascular and coronary artery diseases as well as increased risk of diabetes 1,3,11,12 .Variable number of tandem repeats (VNTRs), also known as minisatellites, comprise a significant portion (~3%) of the human genome 13 .These are predominantly localized in the sub-telomeric region of chromosomes and have a core repeat unit of 10 to100 base pairs 6,14 .VNTRs are stably inherited in a Mendelian fashion across generations and DOI: http://dx.doi.org/10.3329/bjms.v17i1.35284considered as the most informative markers for genetic characterization 15 .Hyper-variable nature of many VNTR loci have found useful applications in genetic linkage analysis, forensic identification, paternity testing, anthropological research and phylogenetic studies 6,15,16 .One such VNTR is located 73bp upstream of the second polyadenylation signal at the 3ʹ-end of the human Apolipoprotein B (APOB) gene on the short arm of chromosome 2 15,17,18 .The APOB 3´-VNTR region consists of tandem repeats of AT-rich DNA sequences (14-16bp in length) 17 .Longer alleles of APOB 3ʹ-VNTR site has been shown to be associated with different disease conditions.APOB gene alleles with larger 3ʹ-VNTR repeat numbers occur more frequently in gallstone patients 19 .The most common cause of death in diabetes is atherosclerotic cardiovascular disease resulting from dyslipidemia 20 .APOB 3ʹ-VNTR alleles with >37 repeats have been shown to be strongly associated with increased lipid levels and coronary artery disease 18 .In this study, we investigated whether there is any association of the number of repeats at the APOB 3ʹ-VNTR locus with type 2 diabetes, body mass index (BMI), systolic and diastolic blood pressures.

Materials and methods
DNA samples: Blood samples and additional information (e.g., systolic and diastolic blood pressure, body weight and height, other disease conditions, family history, etc) were collected with written consent from 67 unrelated Bangladeshi individuals diagnosed to have type 2 diabetes and 46 non-diabetic controls.Males and females represented 56.31% and 43.69%, respectively, of the studied population.Blood glucose level was measured following overnight fasting and two hours after meal.DNA was extracted from whole blood using Genomic DNA mini kit (AGB100, ATP Biotech, Taiwan).Purity and concentration of extracted DNA samples were measured using Nano Drop UV-Vis Spectrophotometer (Thermo Fisher Scientific Inc.).Sequence amplification and detection: APOB 3ʹ-VNTR region of the studied samples were amplified by PCR using primers already described by Ruixing et al. 21in a thermal cycler (Gene Atlas G, Astec Co. Ltd.).About 20 to 50 ng of genomic DNA was used for amplification in a final reaction volume of 25 μl containing PCR buffer (EP0712, Thermo Scientific, USA), 1.0 µl of each primer (10 µM), 1 µl dNTPmix (10 mM) (R0191, Thermo Scientific, USA) and 1U Taq polymerase (EP0712, Thermo Scientific, USA).The cycle condition was as follows: an initial denaturation step at 94°C for 5 minutes, then 33 cycles-each with denaturation at 94°C for 30 seconds, annealing at 58°C temperature for 60 seconds, and elongation at 72°C for 60 seconds followed by a final extension at 72°C for 5 minutes.The amplified products were resolved in 2% agarose gel with 0.5x Tris Borate EDTA (TBE) buffer in a horizontal Agarose gel electrophoresis system along with DNA size markers (300003, GeneON).DNA bands were observed in a gel documentation system (WGD-30, Witeg) following staining with ethidium bromide in 0.5x TBE buffer and photographed with Wise Capture II TM software.

Allele size determination and statistical analysis:
Sizes of PCR products were determined using the software "Gel Analyzer" 22 by comparing their migration relative to DNA size markers (300003, Gene ON).PCR products of the patients and the controls were run side by side in gels.The number of core repeats in the amplified products was calculated using Microsoft Excel following the formula described by Ruixing et al. 21.The alleles were designated according to the number of core repeats.Frequencies of the APOB alleles and genotypes of the individuals and other statistical analyses were performed using GenAlEx 6.5 23 and GraphPad Prism ® software.Ethical approval was received from local ethics committee before data collection was initiated.

Results
All together 24 different alleles were identified in the studied population (Table 1).Despite the less number of non-diabetic individuals than the diabetic patients in this study, the former had more APOB 3ʹ-VNTR alleles (21 alleles) compared to the later (17 alleles).The smallest and the largest alleles had 17 and 47 repeats, respectively.Alleles at both ends of the distribution had relatively low frequencies.Among the non-diabetic individuals, allele with 35 repeats had the highest frequency (0.163) followed by allele with 39 (0.152) repeats.Among the diabetic patients allele with 37 repeats was the most common (0.209) followed by allele with 35 (0.164) repeats.In combined population data (diabetic and non-diabetic together), allele with 37 had the highest frequency (0.173) followed by allele with 35 (0.164) and 39 (0.128) repeats.In our study, APOB 3ʹ-VNTR alleles with 37 repeats showed statistically significant (p<0.001)difference in frequencies between the diabetic and non-diabetic individuals (Table 1).Overall distribution of APOB 3ʹ-VNTR allelic frequencies in these two groups was different.Combined frequencies of alleles with ≥35 repeats as well as genotypes of individuals with at least one allele with ≥35 repeats were more common among the diabetic individuals-although this difference was not statistically significant.We divided the diabetic and the non-diabetic studied population in two groups -one included individuals who had both APOB alleles with <35 repeats and the second group had at least one allele with ≥35 repeats.We did not compare the lipid profiles of diabetic and non-diabetic individuals in this study.But individuals having at least one allele with ≥35 repeats had higher Body mass index (BMI), systolic and diastolic blood pressures (in both males and females) (Figure 1).

Discussion
This is the first study to investigate any possible association of the number of repeats at the APOB 3ʹ-VNTR locus with type 2 diabetes, body mass index (BMI), systolic and diastolic blood pressures.Earlier studies have found alleles with 35, 37 and 39 repeats to be the most common ones in different populations worldwide 16,18,[24][25][26][27][28][29][30][31][32] .In two different Indian subpopulations 18,29 , allele with 35 repeats was found to be the most prevalent.In our study, allele with 35 repeats had the highest frequency (0.163) among the non-diabetic individuals, whereas the diabetic patients had the allele with 37 repeats as the most common (0.209).APOB 3ʹ-VNTR alleles with relatively larger number of repeats have association with a number of disease conditions.Alleles with higher repeats (≥36 repeats) were shown to be strongly correlated with coronary heart disease 18,[33][34][35] .APOB 3ʹ-VNTR alleles with higher repeats are also significantly associated with higher levels of total cholesterol (TC) and low density lipoprotein-cholesterol (LDL-C) We also observed that individuals having at least one allele with ≥35 repeats had higher Body mass index (BMI), systolic and diastolic blood pressures (in both males and females).These differences, however, were not statistically significant.
In this study we did not measure APOB level in serum to correlate with 3ʹ-VNTR repeat numbers in diabetic and non-diabetic individuals.Earlier studies have reported high APOB protein level in serum of individuals with larger APOB 3ʹ-VNTR alleles 18,36,37 .In Canadian aboriginal people, high level of plasma APOB protein was found to be associated with type 2 diabetes as well as a better predictor of risk compared to LDL or HDL cholesterol 12 .Hashemi et al. found that serum APOB protein levels in diabetic children (aged 9-18 years) and healthy children with diabetic parents were significantly higher than the healthy children of similar age with nondiabetic parents 20 .Atherosclerotic coronary artery disease (CAD) is the most common co-morbidity in type 2 diabetes and the risk of death by CAD among diabetic individuals is greater by 3-folds compared to the nondiabetics 38 .CAD can at least partially be attributed to abnormalities in lipid and lipoprotein metabolism 39 .Our observation that the APOB 3ʹ-VNTR alleles with larger repeats are more prevalent in type 2 diabetic individuals might make the bridge among the genetic Figure 1: Association of APOB 3ʹ-VNTR alleles with BMI, systolic and diastolic blood pressure.A. Irrespective of the disease condition, individuals with one or both APOB 3ʹ-VNTR alleles with ≥35 repeats had higher BMI values, although none of these differences were statistically significant.B and C. In both males and females, individuals with one or both APOB alleles with ≥35 repeats had higher systolic and diastolic blood pressure values, although none of these differences were statistically significant.signature of APOB 3ʹ-VNTR, high APOB level in blood, diabetes and other co-morbidities.The mechanism through which 3ʹ-VNTR polymorphism influence serum lipid level is not clearly elucidated yet 18 .Since these VNTR repeats are present in the 3ʹuntranslated region (3ʹ-UTR) of APOB gene, these are not supposed to play roles in the structure and/or function of APOB protein.
Alleles with the higher repeat numbers may be in linkage disequilibrium with other genes which may have some effect on the regulation of gene expression 5 .High repeat numbers may also increase stability/half-life of APOB mRNA or enhance translation through interaction with other proteins or factors.For example, APOB 3ʹ-VNTR region is ATrich 17,40 .Human antigen R (HuR)/ELAV1 protein is a key player in both mRNA turnover and translation regulation processes in eukaryotic cells and binds to sites called AU-rich elements (ARE) 41 .There are different opinions on the actual consensus motif of AREs.Barreau et al. 42 suggested that AREs are just sequence elements of 50-150 nucleotides which are rich in adenosine (A) and uridine (U) bases.Proteins like HuR may play role in increased expression of APOB alleles with high number of AT-rich repeats.
This and other possibilities may be explored to decipher the role of APOB 3ʹ-VNTR repeat number with its expression level and associated diseases.Conclusion APOB 3ʹ-VNTR alleles have different distribution frequencies among type 2 diabetic and non-diabetic individuals.We observed that individuals harboring at least one allele with ≥35 repeats have higher body mass index (BMI), systolic and diastolic blood pressures.Our study might bridge among the genetic signature of APOB 3ʹ-VNTR, high APOB protein level in blood, diabetes and other co-morbidities.The mechanism behind the influence of the 3ʹ-VNTR polymorphism on serum lipids is not known yet.Molecular interaction between APOB 3ʹ-VNTR site and different regulatory proteins may be explored to decipher the role of APOB 3ʹ-VNTR repeat number on its expression level and associated diseases.Declaration There is no known conflict of interest.