Association of AdipoQ (G>T) gene polymorphism with obesity and hypertension in North Indian postmenopausal women of Punjab

Introduction

According to a current report by Unni [1], which is dependent on third consensus meeting of the Indian menopausal society, 43 million are menopausal women in India amongst the total population of more than 1 billion and is further anticipated that by the year 2026, the population of menopausal women will be 103 million, amongst the total population of 1.4 billion.

Menopause is related to decrease in estradiol and a reduction in the estrogen to testosterone proportion. It brings about endothelial dysfunction and an increase in body mass index resulting in an increase in sympathetic activation. Sympathetic activation causes an increase in both renin release and angiotensin II (Ang II). Endothelial dysfunction is accompanied by a decrease in NO and an increase in endothelin, the two of which are responsible for salt sensitivity of blood pressure. An increase in endothelin and Ang II and a decrease in NO may contribute as a causative factor to oxidative stress. All of these factors contribute to an increase in renal vasoconstriction that causes hypertension [2] (Figure 1).

Figure 1:

Causative factors to hypertension in menopausal women. Adapted from Coylewright et al. [2].

The growing prevalence of obesity and hypertension is progressively familiar as one of the most reliable risk factors for the progression of cardiovascular diseases. Postmenopausal women (PMW) are generally found to have a higher mean body mass index (BMI), waist circumference (WC) and waist-hip ratio (WHR) that is increased rates of obesity as compared to premenopausal women.

It is commonly seen that genetic factors assume a part in the etiology of obesity. BMI- the WHO-adopted measure of obesity has a heritability estimate of 40–70 % [3].

Numerous genes as well as environmental factors (diet and lifestyle) show an impact on both obesity and hypertension, but more research is needed to better understand their etiology. Although the etiology of hypertension is not fully understood, it is believed that the interplay of genes and environmental factors is crucial to the pathological development of hypertension [4]. Several independent association studies across populations worldwide have shown that some ADIPOQ SNPs have obesogenic effects, albeit with inconsistent findings, which could be due to ethnic population differences, varying SNP selection criteria, and study power [5, 6].

Research has recently concentrated on the genetic predisposition and genetic polymorphism of obesity and hypertension. Several single-nucleotide polymorphisms (SNPs) in the ADIPOQ gene have been mentioned.

The ADIPOQ gene (OMIM: 605441, Gene ID: 9370) located in human 3q27 has a DNA length of approximately 16 kb. It comprises three exons and two introns. Fat cells secrete a specific hormone protein ADIPOQ. As a defensive component, it plays a vital role in anti-atherosclerosis, insulin resistance and anti-inflammation [7].

Various population-based studies have found an association between ADIPOQ (276G>T) gene polymorphism and CVD risk factors such as obesity and hypertension and have reported conflicting results. Many researchers observed a positive association of this polymorphism with obesity, HTN and associated risk [8], [9], [10]. Contrarily, some stated a proactive role [11], [12], [13] while some failed to detect any association [14], [15], [16].

With this set of background knowledge, inconsistent results and scanty literature for the association of obesity and hypertension with ADIPOQ (+265 G>T) polymorphism in North Indian postmenopausal women of the Punjabi population, we aimed to investigate their alleged association in this population.

Materials and methods

Results

A comparison of the obesity profile is shown in Table 1. The mean values of BMI<25 (22.23 ± 2.32; 22.45 ± 2.13) and BMI≥25 (30.21 ± 3.58; 30.15 ± 3.45) were almost the same in postmenopausal women with and without CVD, with no statistically significant difference observed among the two groups (p=0.476, p=0.881, respectively).

Mean values of WC<80 was again almost same in both the groups with insignificant difference (75.71 ± 2.94, 75.96 ± 2.40; [p=0.786]) whereas, a significant difference was found in the mean values of WC≥80 (98.15 ± 10.11, 95.90 ± 9.35, [p=0.013]) when the two groups were compared.

An insignificant difference was observed in the mean values of WHR<0.81 (0.79 ± 0.01, 0.80 ± 0.00, [p=0.411]). Contrarily, a significant difference was observed in the mean values of WHR≥0.81 (0.93 ± 0.04, 0.90 ± 0.03; [p=0.001]) when both the groups were compared.

In Table 2 the occurrence of homozygous wild type genotype (GG) in PMW with CVD was 49.4 % and in PMW without CVD was 43.8 %. The frequency of heterozygous genotype (GT) was found to be higher in women without CVD (53.5 %) than with CVD (30.2 %). However, the homozygous mutant type genotype (TT) distribution was significantly greater in PMW with CVD (20.4 %) than in PMW without CVD (2.7 %) (Figure 2). The frequency distribution of the genotypes and alleles of the AdipoQ (+276G>T) polymorphism was compared, and it was discovered that the frequency of the minor type allele (T) was higher in PMW with CVD (35.5 %) than in PMW without CVD (29.5 %), while the frequency of the wild type allele (G) was higher in PMW without CVD (70.5 %) than in PMW with CVD (64.5 %). Both the genotype (χ²=50.7, p=0.001) and allele (χ²=4.04, p=0.04) distributions showed statistically significant differences, with the T allele conferring risk (OR=1.31, 1.01–1.70) towards CVD predisposition.

Figure 2:

Distribution of allele and genotype frequencies of AdipoQ (+276 G>T).

AdipoQ (+276G>T) TT homozygotes presented a substantial risk for CVD predisposition of 9 folds under recessive model (TT vs. GT+GG) analysis (OR=9.17 [4.08–20.59]; p=0.0001), the dominant model (GT+TT vs. GG) (OR=0.79 [0.56–1.12]; p=0.228) demonstrated no significant difference in the distribution whereas, the co-dominant model (GT vs. GG+TT) (OR=0.37 [0.26–0.53]; p=0.0001) offered protection against CVD when PMW with and without CVD were compared.

Table 3 depicts that with BMI<25, G & T allele frequency was 62.2 % & 37.8 % in PMW with CVD and 71.9 % & 28.1 % in PMW without CVD respectively. A significant difference was found in both allelic (p=0.042) and genotypic (p=0.0001) frequency distribution. Under the analysis of the recessive model (TT vs. GT+GG), the TT genotype conferred a high risk for CVD predisposition of 9 folds (OR=9.56 [2.81–32.49]; p=0.001) whereas, the co-dominant model offered protection against CVD (OR=0.39 [0.22–0.69]; p=0.001). With BMI≥25, the frequency of G & T allele was 66.5 % & 33.5 % in PMW with CVD and 69.8 % & 30.2 % in PMW without CVD respectively. A significant difference was observed in the genotypic (p=0.0001) frequency distribution. However, the distribution was insignificant for allele frequency (p=0.447). TT genotype conferred a high risk for CVD predisposition of 8 folds (OR=8.44 [2.85–24.91]; p=0.001) whereas, GT genotype offered protection towards CVD (OR=0.37 [0.23–0.59]; p=0.001).

In the background in women with WC<80, an insignificant difference was found in allelic (p=0.404) as well as genotypic (p=0.235) frequency distribution. With WC≥80 frequency of the G & T allele was 86.3 % & 13.7 % in PMW with CVD and 85.5 % & 14.5 % in PMW without CVD respectively. Nevertheless, the distribution in frequency was statistically insignificant for allele (p=0.067) whereas, it was statistically significant for genotype (p=0.001). The TT genotype contributed an approximately eightfold increased risk in recessive model analysis (TT vs. GT+GG) (OR=8.30 [3.68–18.72]; p=0.001). However, the codominant model (GT vs. GG+TT) had a lower risk of CVD predisposition (OR=0.37 [0.25–0.54]; p=0.001).

For menopausal women with WHR≥0.81, under a recessive model, the TT genotype imparted an approximately eightfold increased risk (OR=8.28 [3.69–18.58]; p=0.001). The GT under the codominant model (GT vs. GG+TT) conferred a reduced risk (OR=0.47 [0.33–0.67]; p=0.001) towards CVD predisposition. At the background in women with WHR<0.81 frequency of G and T alleles could not be commuted because the frequency of TT genotype in cases and the fequency of GT and TT genotypes in controls were not observed.

Table 4 depicts that the frequency of the G & T allele in hypertensive PMW with CVD was 61.8 % & 38.2 % and in hypertensive PMW without CVD was 72.3 % & 27.7 % respectively. Comparison between the two groups revealed significant differences (p<0.05) in the allelic (p=0.010) and genotypic (p=0.001) frequency distribution. Under the recessive model (TT vs. GT+GG) TT genotype conferred ∼ 7 folds increased risk (p=0.001, OR=7.21 [2.77–18.80]) and GT genotype under the codominant model (GT vs. GG+TT) conferred reduced risk (p=0.001, OR=0.44 [0.27–0.72]) towards CVD predisposition. However, an insignificant difference (p>0.05) was observed under the dominant model (GT+TT vs. GG) analysis (p=0.93, OR=1.04 [0.66–1.66]).

Stratified analysis of hypertension with AdipoQ (+276G/T) polymorphism in Table 5 showed that in participants with SBP<140, the frequency of G and T alleles was 71.8 % & 28.2 % in hypertensive postmenopausal women with CVD and 68.4 % & 31.6 % in hypertensive postmenopausal women without CVD respectively. On comparing the two groups significant difference in the genotype frequency distribution (p=0.0001) and an insignificant difference in the allele frequency distribution (p=0.491)] was found. TT genotype under recessive model (TT vs. GT+GG) analysis imparted ∼10-fold high risk (OR=10.39 [8.26–47.59]; p=0.0008) towards CVD susceptibility. Both codominant model (GT vs. TT+GG) (OR=0.29 [0.16–0.50]; p=0.0001) and dominant model (GT+TT vs. GG) analysis (OR=0.47 [0.28–0.81]; p=0.008) conferred reduced risk towards CVD susceptibility. In participants with SBP≥140, the frequency of G and T alleles was 60.5 % & 39.5 % in hypertensive PMW with CVD and 73.2 % & 26.8 % in hypertensive PMW without CVD. Both allelic (p=0.002) and genotypic (p=0.0001) frequency distribution showed significant differences amongst both groups. TT genotype under recessive model (TT vs. GT+GG) analysis imparted ∼7 fold more risk (OR=7.09 [2.71–18.57]; p=0.0001) and GT genotype under codominant model (GT vs. TT+GG) analysis conferred reduced risk (p=0.014, OR=0.52 [0.32–0.86]) towards CVD susceptibility. Contrarily, an insignificant difference was observed in dominant model (GT+TT vs. GG) analysis (OR=1.23 [0.76–1.98]; p=0.453).

In participants with DBP<90, the frequency of G &T alleles was 64.1 % & 35.9 % in hypertensive PMW with CVD and 69.9 % & 30.1 % in hypertensive PMW without CVD respectively. A significant difference was seen in the genotypic (p=0.0001) whereas, an insignificant difference was found in the allelic (p=0.143) frequency distribution. TT genotype under the recessive model (TT vs. GT+GG) conferred ∼ 9 folds increased risk (OR=9.21 [3.45–24.57]; p=0.001) and GT genotype under codominant model (GT vs. GG+TT) conferred protection (OR=0.37 [0.23–0.59]; p=0.0001) for CVD. However, an insignificant difference was found under the dominant model (GT+TT vs. GG) analysis (OR=0.78 [0.50–1.21]; p=0.329) on comparing the two groups. In participants with DBP≥90 frequency of the T allele was found to be more in hypertensive PMW with CVD (35 %) as compared to hypertensive PMW without CVD (27.8 %) whereas, G allele was found to be more in hypertensive PMW without CVD (72.2 %) than in hypertensive PMW with CVD (65 %) but, the difference in the allelic (p=0.168) frequency distribution was not statistically significant. However, a significant difference was found in the genotypic frequency distribution (p=0.0001) amongst both groups. TT genotype under recessive model (TT vs. GT+GG) analysis imparted ∼9 fold increased risk (OR=9.01 [2.07–39.20]; p=0.001) and GT genotype under codominant model (GT vs. TT+GG) analysis showed protection (OR=0.41 [0.22–0.74]; p=0.005) for CVD. Contrarily, the dominant model (GT+TT vs. GG) analysis showed no significant difference when cases were compared with controls (OR=0.88 [0.49–1.57]; p=0.777).

Pearson’s correlation was used to determine whether diastolic blood pressure (DBP) had any link with BMI, WC, and WHR. Table 6 depicts correlation between obesity profile and DBP. Positive correlation was observed for BMI, WC with DBP. However, no significant association (r=0.075, p=0.225) was found between WHR and DBP.

Figure 3 depicts a significant positive correlation (r=0.144; p=0.019) between DBP and BMI in postmenopausal women with CVD. Similarly, a significant positive correlation (r=0.170; p=0.005) was observed between DBP and WC in postmenopausal women with CVD as shown in Figure 4.

Figure 3:

Scatter chart showing correlation between DBP and BMI in postmenopausal women with CVD.

Figure 4:

Scatter chart showing correlation between DBP and WC in postmenopausal women with CVD.

Discussion

Higher socioeconomic class women are more likely to be pre-overweight, overweight, and obese [22]. Obesity is a significant factor in the development of many diseases, especially the buildup of visceral fat which increases the risk of T2D, dyslipidemia and hypertension, contributing to CVD. Obesity prevalence in postmenopausal women may reach 40 % [23].

The body mass index (BMI) is usually used as an index to evaluate the degree of body fat. BMI has been proven to be consistently linked to an elevated risk of CVD. Nevertheless, this assessment does not take into consideration variations in body fat distribution and body fat mass, which can change significantly among populations and within a limited range of BMI [24]. Studies have reported inconsistent findings related to BMI and CVD, BMI primarily reflects overall obesity while WC and WHR are associated to central obesity, and surplus intra-abdominal fat (visceral fat) is linked to more of the risk of obesity related morbidity as compared to generalized obesity. Accordingly, measurement of WC and WHR can be viewed as alternatives to BMI. Dalton et al. in their study observed WHR to be the preferred measurement of obesity, to anticipate whether a CVD risk is present [25]. In our present study, we observed significantly increased mean levels of WC≥80 and WHR≥0.81 in PMW with CVD as compared to PMW without CVD. Whereas, an insignificant difference was found in the mean levels of BMI ≥25 amongst the two groups (Table 1). Various studies have shown that women have a two-fold increase in cardiovascular dysfunction with normal body BMI and increased WHR. They considered WHR a superior interpreter for evaluating CVD risk as compared to BMI [26].

AdipoQ (+276G>T) TT genotype presented a substantial risk for CVD predisposition of 9 folds under recessive model (TT vs. GT+GG) analysis. When PMW with CVD were compared with PMW without CVD, the difference in the allele frequency distribution was found to be significant. The frequency of the T allele carrier was higher in PMW with CVD as compared to PMW without CVD (Table 2). Similarly, Fillipi et al. revealed that the risk of CAD was more in T allele carriers of the Italian population [27]. TT genotype of AdipoQ (+276G>T) polymorphism of the Egyptian population was also associated with CVD risk-a study specified by Ghattas et al. [28]. Similarly, Gui et al. [29] and Tong et al. [30] reported that AdipoQ +276G>T was positively correlated with an increased risk of CAD in a Chinese population. Furthermore, Li and his associates in their study on Chinese Adolescents reported that genotype SNP +276T allele was associated to lower levels of serum adiponectin in a way showing the risk role of T carriers [31]. Mohammadzadeh et al. revealed the association of high risk of CAD with T allele, GT and TT genotype of SNP +276G>T among the population of Iranians. Furthermore, they also observed the frequency of the T allele to be significantly higher in female CAD subjects as compared to controls [32]. A study by Amrita et al. also revealed that Punjabi Indian women who carry the T variant of the AdipoQ gene are more likely to experience oxidative stress, which increases their risk of cardiovascular disease (CVD), compared to those who carry the G allele [33].

Contrarily, Bacci et al. declared that homozygous carriers of the +276T allele were at lower risk as compared to carriers of the G allele in the Caucasian population [34]. In American diabetic males, Qi et al. reported a protective effect of +276 SNP for CAD under the recessive model of inheritance with the odds ratio being 0.38 for coronary artery disease [35]. In addition, Qi et al. also reported ∼45 % lower association of CVD risk in females from Nurses’ Health Study for +276G>T polymorphism [14]. Fredriksson et al. revealed that T-allele carriers had a protective role against obesity in Swedish and Caucasian subjects [11]. A decreased risk of CHD was also observed under dominant and co-dominant models in Chinese Han populations by Zhang et al. [36]. Al-Daghri et al. reported that the AdipoQ variant was not associated with CAD in a Saudi diabetic population [37]. Similarly, no association of rs1501299 with CVD in Afro-Caribbean diabetic patients was also observed by Foucan et al. [38].

The distribution of adipose tissue is shown to be an important determinant of inflammation. Indulekha observed that lower levels of adiponectin are linked with T2D and visceral obesity because adiponectin acts as an insulin sensitizing antiatherogenic and anti-inflammatory adipokine secreted by AT [39]. According to Gatuszka-Bilińska & Grzeszczak obese and overweight patients with T allele carrier were found to have high levels of adiponectin [15]. Similarly, Fredriksson et al. revealed that T allele carriers of Swedish and Caucasian subjects are protected against the negative metabolic effects of obesity because they have higher levels of visceral adipo gene expression [11]. According to Qi et al. in their study conducted on females from Nurses’ Health Study, baseline characteristics of the females showed a significant difference in the mean levels of BMI however, +276G>T polymorphism was not significantly associated with adiposity [14]. Boumaiza et al. also observed a protective role of +276G>T SNP among the Tunisian population [12]. Contrarily, in the present study under recessive model analysis, the TT genotype of the obese Indian Punjabi population with CVD conferred ∼8 fold high risk towards CVD predisposition in all the indices of obesity. Furthermore, non-obese menopausal women with CVD also presented ∼9 folds higher risk towards CVD predisposition. This may be due to the increased frequency of the T allele in CVD PMW (37.8 %) as compared to non-CVD PMW (28.1 %) (Table 3). Similarly, Filippi et al. reported that the T allele was associated with obesity risk in the Italian population [27]. Another study by Mackawy et al. on Saudi subjects confirmed the same findings for the association between T allele and TT genotype of +276G>T SNP with higher risk of obesity [40]. Zaki et al. also reported an association of T allele and TT genotype with the risk of obesity in Egyptian adolescents [8]. Likewise, a significant association of +276G>T polymorphism with the risk of obesity was observed by Kaur et al. among the North Indian Punjabi population [9]. Contrarily, an insignificant difference of the T allele distribution was found among obese patients of the outpatient Cardiology Clinic in Ruda Ślaska, Poland and healthy controls by Gatuszka-Bilińska & Grzeszczak [15].

A vital organ for preserving the balance and health of the cardiovascular system is adipose tissue. It can react to and generate, a variety of signals that have an impact on the homeostasis of the heart rate and blood pressure [41]. According to Ford et al. and Flegal et al. another mechanism by which BP may be elevated, is the presence of obesity [23, 42]. The changes in estrogen/androgen ratio connected with menopause might add to postmenopausal hypertension. It brings about changes in endothelial function and more of the body mass index (BMI), increasing the rate of sympathetic activation which results in an increase in renin release and angiotensin II (AngII) [2]. Alteration in the function of vascular endothelium occurs early in the development of atherosclerosis.

Several studies have concentrated on the genetic predisposition and genetic polymorphism of HTN [43]. In the present study, on statistical comparison for AdipoQ (+276G>T) polymorphism, a significant association of HTN as a risk factor with CVD was observed between hypertensive women with and without CVD. TT genotype, under recessive model analysis for hypertensive women with CVD conferred ∼7 folds higher risk towards CVD predisposition when compared with the ones without CVD (Table 4). Further, stratified analysis revealed that hypertensive TT homozygotes with CVD (SBP≥140 and DBP≥90) conferred a 7–9 fold risk towards CVD susceptibility. Furthermore, normotensive TT homozygotes with CVD (SBP<140 and DBP<90) also conferred a 9∼10fold risk towards CVD predisposition (Table 5). Similarly, Demir et al. found 276 G/T SNP to be a risk factor for essential hypertension in a Turkish population [10]. Fan and his associates proposed the protective role of rs1501299 polymorphism in the hypertensive Caucasian subgroup in their study on meta-analysis [13]. Contrarily, Ghattas et al. found insignificant variation in the prevalence of hypertension between T and G allele carriers in Egyptian CAD subjects [28]. Platelet aggregation is reduced and vasodilation is promoted as a result of adiponectin’s increased AMPK activation and NO generation. Last but not least, adiponectin reduces the incidence of HTN as described earlier. No association of HTN and AdipoQ +276 G to T polymorphism was also observed by Yu et al. in their study on meta-analysis [16].

Being overweight and obesity are related to increased activity of the sympathetic nervous system especially in the kidneys causing increased secretion of renin which is attributed to hypertension. Rosano et al. observed that increased BMI and increased proportion of visceral adipose tissue were in strong correlation with hypertension and other risk factors for cardiovascular diseases [44]. Similarly, our results showed significant positive correlation between DBP with BMI and WC (Table 5) (Figures 3 and 4). On the contrary, Achie et al. observed an insignificant relationship between BMI and DBP [45]. Pimenta et al. suggested that menopause was accompanied by a slight increase in BP that might be contributed to aging, BMI and augmented CV risk factors [46]. Similarly, in recent findings by Amrita et al., the proportion of HTN was found to be higher in obese women of >45.1 years of age group [47]. On the contrary, Salini in her study group observed that hypertensive postmenopausal women exhibited low BMI, hence, BMI might not be a contributing factor for elevated blood pressure [48].

Conclusions

The novel findings of our study represented that PMW of the Indian Punjabi population who are homozygous for T allele at position +276 of AdipoQ gene are at increased risk of CVD than G allele carriers. Interestingly, the most stimulating part in our findings was that under the recessive model of inheritance, increased risk was also observed for normotensive PMW and in PMW with BMI<25. Obesity and HTN may partially overlap one another as a specific clinical sub phenotype with a common genetic background. There may be a direct or indirect association between genes and the disease and between genes and the phenotypes. Genes may influence the susceptibility to CVD in response to many environmental & conventional risk factors. Another reason for the presence of T variant of AdipoQ in these women can be dietary habits and lifestyle choices which can also act as potential confounding factors in the onset of the disease. These factors may also play a role in figuring out how genes and the environment interact. HTN is affected by a combination of environmental & genetic factors. Further, study is required to determine the major contribution of AdipoQ gene variations to the emergence of obesity & HTN as risk factors of CVD disorders in light of gene-gene interaction in order to clarify the pathological processes of CVD.