High Density Lipoprotein in Bangladeshi Adults: Characterization and the Basis of its Variation

Abstract

Background and objectives: High-density lipoprotein cholesterol (HDL-c) constitutes a vital cardioprotective factor, yet low levels are prevalent among the Bangladeshi population, and its biochemical and genetic determinants remain poorly characterized. This study aimed to determine the prevalence of low HDL-c and associated components of dyslipidemia, along with their demographic and biochemical factors, among healthy Bangladeshi adults. The main focus of the study was to determine the effect of twelve single nucleotide polymorphisms (SNPs) in genes involved in lipid dynamics, i.e., ApoA1 (–75 G/A and +83 C/T), ApoB (7673C/T [rs693], 10108A/G [rs1801701], 12669G/A [rs1042031]), ABCA1 (–565C/T, 1051G/A, 2868G/A), PON1 (163T/A [rs854560, L55M] and 575A/G [rs662, Q192R]), and CETP (–629C/A, 277C/T [Taq1B]) was explored. Methodology: This cross-sectional study recruited 409 healthy adults from different areas of the Dhaka division. Participants were free from diabetes, hypertension, kidney, liver, or other chronic diseases. After obtaining informed consent, demographic measurements and clinical histories were recorded. Fasting blood samples (5 mL) were collected following aseptic procedures and processed for biochemical analyses using automated spectrophotometric instruments. Measurements of serum lipids, specifically total cholesterol (TC), triglycerides (TG), and HDL-c, were conducted by spectrophotometric end-point methods. Low-density lipoprotein cholesterol (LDL-c) was subsequently derived via the Friedewald formula. Additionally, Apolipoprotein A1 (ApoA1) and Apolipoprotein B (ApoB) concentrations in serum were quantified using immunoturbidimetry on an automated platform. A column-based genomic DNA extraction kit was used to extract DNA from homogenized whole blood leukocytes, and genotyped SNPs in the ApoA1, ApoB, ABCA1, PON1, and CETP genes using Polymerase Chain Reaction-Restriction Fragment Length Polymorphism (PCR-RFLP) followed by agarose-gel electrophoresis. We categorized the participants based on median HDL-c levels or according to reference values into lower HDL-c and higher HDL-c groups following NCEPATPIII guidelines for statistical analysis. Logistic or multiple linear regressions and Fisher’s exact tests were employed to assess associations between genotypes, allele frequencies, lipid profiles, and demographic variables. Results: The median value of serum HDL-c of the total participants was 34.0 (95%CI: 33.0– 35.0) mg/dL. Females exhibited significantly higher levels of HDL-c than males [37.1 (35.5– xvii 38.5) mg/dL vs. 31.6 (30.7–32.6) mg/dL, p < 0.001]. Lower levels of HDL-c (< 40 mg/dL for males and < 50 mg/dL for females) were prevalent in 91.9% of the participants with similar prevalence across genders (90.4% vs 93.7%, p = 0.229). Among the participants, elevated levels of triglycerides (>150 mg/dL) were found in 34.0%, total cholesterol (>200 mg/dL) in 21.0%, and LDL-c (>130 mg/dL) in 8.8%. In contrast to HDL-c, gender-based differences were observed in the prevalence of other lipid abnormalities. Elevated TG levels were significantly more common among males (40.2%) than females (26.8%, p = 0.005). Conversely, females exhibited a higher prevalence of elevated LDL-c (12.6% vs 5.5%, p = 0.013) and increased ApoA1 concentrations (62.6% vs 50.2%, p = 0.012). Their concentrations were also differed between genders, males with HDL-c below the median (< 31.6 mg/dL) showed higher TG (156 vs 114 mg/dL, p < 0.001) and BMI (24.65 vs 23.05 kg/m², p = 0.002) with lower ApoA1 (108 vs 132 mg/dL, p < 0.001), whereas females with lower levels of serum HDL-c (< 37.1 mg/dL) had lower levels of serum TC (159 vs 178 mg/dL, p < 0.001), LDL-c (102 vs 114 mg/dL, p < 0.001), and ApoA1 (123 vs 148 mg/dL, p < 0.001) along with higher TG (128 vs 103 mg/dL, p < 0.001). Logistic regression statistics identified increased TG and decreased ApoA1 as significant predictors of lower HDL-c across both genders. Across the gene loci studied, ApoA1 (–75 G/A and +83 C/T), ApoB (7673C/T, 10108A/G, 12669G/A), ABCA1 (–565C/T, 1051G/A, 2868G/A), PON1 (163T/A and 575A/G), and CETP (–629C/A and 277C/T), majority of the subjects exhibited wild-type homozygous genotypes. Heterozygous genotypes were less frequent than wild homozygous, and mutant homozygous genotypes were rare. The genotype distributions for most loci conformed to Hardy–Weinberg equilibrium (HWE). Exceptions were noted for ABCA1 2868G/A (p = 0.034), where the distribution deviated from HWE. We found no significant differences in HDL-c and ApoA1 or other lipid variables across the ApoA1 –75G/A and +83C/T genotypes among the overall subjects or males and females. In multivariable linear regression for HDL-c (n = 392), after adjusting for age, BMI, TG and LDL-c neither the –75GG (β = 0.806, p = 0.277) nor the +83CC (β = –1.212, p = 0.330) genotype showed any significant association with HDL-c in the overall sample. Within the male subgroup (n = 210), associations for –75GG (β = –0.665, p = 0.462) and +83CC (β = 1.535, p = 0.325) were non-significant. In contrast, in females the –75GG genotype was linked to a 2.78 mg/dL xviii increase (β = 2.783, p = 0.023) and the +83CC genotype to a 4.28 mg/dL decrease (β = –4.281, p = 0.031) in HDL-c. For serum ApoA1, the overall and male subgroup models showed no significant associations with either –75GG (overall: β = 0.629, p = 0.796; males: β = –2.401, p = 0.459, females: β = 4.606, p = 0.220 ) or +83CC (overall: β = –1.151, p = 0.778; males: β = 4.375, p = 0.434, females: β = –7.796, p = 0.201). Thus, ApoA1 –75 and +83 variants do not independently affect serum ApoA1 levels, they modulate HDL-c in a gender-specific manner, with significant associations observed only in females. Both the ApoB 12669G/A genotype distribution (p = 0.019) and allelic frequencies (p = 0.016) differed between normal and elevated LDL-c groups. Additionally, the ApoB 12669 A allele frequency was higher in subjects with elevated ApoB levels (p = 0.042). Logistic regression revealed the carriers of ApoB 10108 GA+AA had higher risk of elevated TG (OR = 3.36, 95% CI: 1.20–9.45, p = 0.021). Conversely, the ApoB 12669 GA+AA genotype was protective for elevated LDL-c (OR = 0.33, 95% CI: 0.12–0.87, p = 0.023) and elevated ApoB levels (OR = 0.51, 95% CI: 0.27–0.99, p = 0.045). None of the ABCA1 –565C/T, 1051G/A and 2868G/A SNPs differed in genotype distribution and allele frequency between the two groups of HDL-c (P>0.05). The ABCA1 –565TT, 1051AA, and 2868 GA+AA genotypes were not associated with low HDL-c. This was evident from the adjusted OR statistics: 0.80 (95% CI: 0.49 - 1.31, p = 0.186) for –565TT; 1.46 (95% CI: 0.74 - 2.89, p = 0.403) for 1051AA; and 1.12 (95% CI: 0.49 - 1.63, p = 0.809) for 2868 GA+AA. The median (95%CI) of the PON1 arylesterase (PON1-ARE) was 2.50 (2.41 – 2.56) kU/L in the total subjects and higher in males compared to females 2.56 (2.49 – 2.67) vs 2.41 (2.28 – 2.53, p = 0.028). PON1-ARE was highest in 163TT and 575GG genotypes, followed by heterozygous 163TA and 575AG, homozygous 163AA and 575AA. The 163TT, TA and TT genotypes and T, A alleles were almost similar in the two groups of HDL-c (p > 0.05). Similarly, the 575 AA, AG and GG genotypes and A, G alleles showed no difference between the two HDL-c groups (p >0.05). Logistic regression statistics revealed no association of 163T/A and 575A/G with HDL-c [OR(95%CI): 1.06 (0.68 – 1.65), p = 0.804; 0.90 (0.59 – 1.38), p = 0.630]. No significant differences in genotype distribution or allele frequencies of CETP –629CA and 277CT SNPs were observed between HDL-c groups (p > 0.05). For the 277C/T polymorphism, the combined 277(CC+CT) genotypes differed significantly from the TT genotype between HDL-c groups (p = 0.011, OR = 0.37, 95% xix CI = 0.18–0.78). Higher HDL-c was observed in –629AA (p = 0.023) and CA+AA (p = 0.043) carriers compared to CC carriers. Similarly, higher HDL-c was observed in 277TT (p = 0.002) and CT+TT (p = 0.019) carriers compared to CC genotype. Finally, multiple linear regression statistics revealed negative effects of -629CC (β = –1.106, p = 0.038) and 277(CC+CT) (β = – 0.963, p = 0.016) on serum HDL-c levels. Conclusion: Low levels of HDL-c are exceedingly prevalent among the Bangladeshi population and are associated with male gender, elevated TG and decreased ApoA1 levels, potentially contributing to a higher risk of atherosclerotic cardiovascular disease (ASCVD) in this group. The ApoA1 –75GG showed positive and +83CC showed a negative significant impact on HDL-c only in females. The ApoB 10108G/A polymorphism was associated with elevated TG, and the 12669G/A variant was linked to elevated LDL-c and ApoB levels, although ApoB gene polymorphisms did not affect serum HDL-c levels. No association was found between circulating HDL-c levels and the common ABCA1 genotypes –656CT (heterozygous), 1051GA (heterozygous), and 2868GG (wild homozygous). PON1 163T/A and 575A/G polymorphisms contributed to circulating PON1 esterase activity but were not associated with serum HDL-c. Notably, the CETP –629CC, 277CC, and 277CT genotypes were associated with low levels of HDL-c in the Bangladeshi population, suggesting that screening for CETP gene variants may serve as a valuable biomarker for diagnosing low HDL-c levels and potentially guiding interventions to reduce ASCVD risk in this population.