Uncoupling Proteins Variants Are Linked to Hypercholesterolemia and Abdominal Obesity in Metabolically Unhealthy Women

Abnormal or excessive fat accumulation leads to overweight and obesity defined by body mass index (BMI) [1]. Recent studies have shown that individuals classified with normal weight (NW) or excess weight (EW) according to BMI, do not always show the same risk of cardiovascular diseases (CVD) and metabolic syndrome [2].

Indeed, there are NW subjects who present a metabolically unhealthy (MUH) phenotype and a risk of CVD 1.5 to 3 times higher than individuals with NW and metabolically healthy (MH) phenotype [3]. Moreover, subjects with EW can also be MH. Note that “healthy” implies a low risk for cardiometabolic diseases but does not mean no risk for other diseases [3].

Moreover, uncoupling proteins (UCPs) 1, 2 and 3, members of a family of anion-carrier protein located in the inner mitochondrial membrane, are involved in exothermic reaction that dissipate metabolic energy as heat [4]. The UCP1 gene is mostly expressed in brown adipose tissue (BAT), UCP2 is expressed in the adipose tissue (white and brown fat), skeletal muscle, liver, pancreas (particularly in beta cells), brain, kidneys, heart, and intestine, while UCP3 is mainly expressed in skeletal muscle [5]. The uncoupling, or proton leak regulates process like thermogenesis (UCP1), regulation of energy expenditure (UCP1, UCP2, and UCP3), lipid metabolism (UCP2 and UCP3), reduction of mitochondrial reactive oxygen species (ROS) production (UCP1, UCP2, and UCP3), as well as the regulation of insulin secretion by pancreatic beta cells, glucose metabolism, and neuronal activity (UCP2) [4].

The genetic variants -3826A/G (rs1800592) in UCP1, Ala55Val (C/T) (rs660339) in UCP2, and -55C/T (rs1800849) in UCP3 may reduce the activity of these UCPs, thereby decreasing energy expenditure by increasing the coupling of oxidative phosphorylation, which ultimately enhances susceptibility to obesity and related traits [6]. However, results from studies that analyzed associations between UCP1-3 variants and metabolic abnormalities differ among different populations. While some analyses have reported associations between one or more genetic variants with obesity and metabolic disorders, others did not find any association [7‒10].

The role of the genetic variants UCPs has not been studied in adults classified by metabolic phenotypes. Thus, this study aims to evaluate whether the UCP1 3826A/G (rs1800592), UCP2 Ala55Val (rs660339), and UCP3 55C/T (rs1800849) variants are associated with metabolic disturbances variables in Mexican MH and MUH women.

This cross-sectional study was conducted at the Institute of Translational Nutrigenetics and Nutrigenomics of the University of Guadalajara, Jalisco, México. The recruitment of participants took place between 2016 and 2018. Posters were created inviting adults to participate in the project. Of 365 unrelated Mexican-Mestizos who were enrolled, 137 did not provide complete biochemical data or enough genomic deoxyribonucleic acid (DNA) (Fig. 1).

A total of 228 women aged 18–65 years were included and classified by BMI as NW (BMI from 18.5 to <25 kg/m²), and excess weight (EW; BMI from 25 to <40 kg/m²). Subjects with any medication prescribed for any chronic disease that affect the cardiovascular, liver, kidney, or pancreatic functions, as well as those with type 2 diabetes mellitus were excluded. Pregnant or breastfeeding women were also not included.

Before enrollment, participants were informed about the research procedures and those who agreed signed written informed consent. This study was approved by the Research and Ethics Committees of the University of Guadalajara (Register number: CI/019/2010) and was conducted based on the ethical guidelines of the 2013 Declaration of Helsinki [11].

After an 8–10 h fasting, the anthropometric measurements of the female participants were recorded. All measurements were performed without footwear and with light clothing. Body composition, including protein mass, fat mass, and percent body fat, as well as body weight, were determined by tetrapolar electrical bioimpedance (InBody 3.0, Biospace Co., Seoul, Korea). The height measurement was determined with a stadiometer with a precision of 1 mm (Rochester Clinical Research, Inc., New York, NY, USA). BMI was calculated as weight in kilograms (kg) divided by height in square meters (m²). Participants were classified according to their BMI as NW (18.5 to <25 kg/m²) or EW if they were overweight or obese (25 to <40 kg/m²) [12].

Waist and hip circumferences were measured using a Lufkin Executive^® Thinline 2 mm tape (Lufkin Executive Thinline, W606PM, MD, USA). Waist circumference was measured at the narrowest point between the bottom of the ribs and the top of the hip bones (iliac crests), while hip circumference was measured at the widest part of the buttocks [13]. Abdominal obesity was diagnosed when the waist circumference was ≥88 cm [14].

Systolic and diastolic blood pressure was evaluated using a LifeSource digital sphygmomanometer (LifeSource, Milpitas, CA, USA), after an 8–10 h fasting where participants were asked for avoid food, caffeine, and strenuous physical activity. Women rest for at least 15 min and were seated in a chair with their back in contact with the chair back, their arms resting on a horizontal surface, and their legs uncrossed [15]. Two measurements were averaged.

The individuals were classified into two metabolic phenotypes: MH or MUH according to ATP-III criteria and the homeostasis model assessment of insulin resistance (HOMA-IR). The following criteria were used to categorize these phenotypes as unhealthy: blood pressure ≥130/85 mm Hg, triglycerides (TGs) ≥150 mg/dL, HDL-c <50 mg/dL, fasting glucose ≥100 mg/dL, and HOMA-IR >2.5 [16]. Thus, participants were recorded with one or no abnormalities as MH and those with two or more abnormalities as metabolically unhealthy. Accordingly, four groups were created: (1) NW-MH, (2) NW-MUH, (3) EW-MH, and (4) EW-MUH.

After 8–10 h of fasting, blood samples were taken and centrifuged to obtain the serum. A Vitros 350 analyzer (Ortho-Clinical Diagnostics, Johnson and Johnson Services Inc., Rochester, NY, USA) was used to measure glucose, TGs, total cholesterol (TC), and HDL-c by dry chemistry. Low-density lipoprotein cholesterol (LDL-c) was calculated using the Friedewald formula [17] whenever TGs levels were <400 mg/dL (LDL-c = TC – [HDL-c + TGs/5]). Very low levels of lipoproteins cholesterol (VLDL-c) was calculated as follows: [TC – (LDL-c + HDL-c)]. The cut-off points used to define alterations in the lipid profile were those established by the American College of Cardiology and the American Heart Association 2018 [18]: hypertriglyceridemia ≥150 mg/dL, hypoalphalipoproteinemia <50 mg/dL, and elevated LDL-c ≥100 mg/dL. Hypercholesterolemia was defined as ≥200 mg/dL, according to the National Cholesterol Education Program (NCEP) III guidelines [14].

Insulin concentrations were determined using an ELISA assay (Monobind Inc, Lake Forest, CA, USA) according to instructions from the supplier. Insulin resistance was estimated according to the HOMA-IR and calculated as follows [17]: (fasting insulin [μU/mL] × fasting glucose [mg/dL])/405, a value >2.5 indicates the presence of insulin resistance. Also, the glucose and triglyceride (TG) index was calculated as the Ln (fasting TGs [mg/dL] × fasting glucose [mg/dL])/2 [19].

The High Pure PCR Template Preparation kit (Roche Diagnostics, Mannheim, Germany) was used to extract blood genomic DNA which was then diluted to 20 ng/µL. UCPs variants were identified by allelic discrimination using TaqMan^® probes (assay number C___8866368_20 UCP1, C____903746_1_ UCP2, C___8751325_1_ UCP3; Drug Metabolism Assay, Applied Biosystems, Foster City, CA, USA). A LightCycler^® 96 real-time PCR system (Roche Diagnostics, Mannheim, Germany) was used to perform the experiments under the following conditions: 95°C for 10 min and 40 cycles of denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min. UCPs genotyping were verified using negative and positive controls of DNA samples corresponding to the three possible genotypes in each 96-well plate [20]. A total of 20% of all samples were analyzed in duplicate.

The statistical power was evaluated according to the calculation of sample size, performed with an estimated error margin of 5%, a confidence interval (CI) level of 95%, and an expected prevalence of metabolically healthy obese subjects of 19% reported in a previous study [16]. The Shapiro-Wilks test was used to determine normal distribution of quantitative variables. All variables were log-transformed to better approximate a normal distribution. To analyze differences between subjects with NW or EW and MH or MUH phenotypes and their UCPs genotypes, the two-way analysis of covariance (ANCOVA) was used. Variables were adjusted by age presented as mean and standard error of the mean. Multiple testing adjustments were performed by Bonferroni test. The χ² test was used to compare categorical variables and calculate the Hardy-Weinberg equilibrium.

All associations were performed using logistic regression adjusted for age. The odds ratios (ORs) with 95% confidence interval (CI) were calculated to assess the associations of the UCP1 3826A/G, UCP2 Ala55Val, and UCP3 55C/T variants with anthropometric and biochemical variables according to their metabolic phenotype and by allele frequency (A vs. G for UCP1, and T vs. C for UCP2 and UCP3), under dominant (AG + GG vs. AA for UCP1 and CT + TT vs. CC for UCP2 and UCP3) recessive (AA vs. AG + GG for UCP1, and TT vs. CT + CC for UCP2 and UCP3), co-dominant 1 (AG vs. GG for UCP1 and CT vs. TT for UCP2 and UCP3), and co-dominant 2 (AA vs. GG UCP1 and CC vs. TT for UCP2 and UCP3) models, respectively [21]. All statistical analyses were performed using SPSS v. 28.0 software (IBM Corp., Armonk, NY, USA) and a p value <0.05 was considered statistically significant.

In this population, the mean age of female participants was 37.1 ± 11.1 years. General characteristics of the sample, according to their metabolic phenotype are shown in Table 1. Significant differences were found for most of the variables, except for TC and LDL-c.

The distribution of the UCP1 -3826A/G, UCP2 Ala55Val, and UCP3 55C/T variants was similar in subjects with MH and MUH phenotypes in dominant, recessive, or co-dominant models (Table 2). There were no significant differences between the frequencies in the MH and MUH subject.

Metabolic phenotype and genotype data for the three UCP variants were analyzed in a two-way ANCOVA. For all three genetic variants, subjects had differences in anthropometric variables (weight, BMI, muscle mass, fat mass and waist circumference, and VLDL-c levels) (p < 0.001) according to their phenotypes (online suppl. Tables 1–3; for all online suppl. material, see https://doi.org/10.1159/000543484).

In the UCP1 -3826A/G variant, significant differences in muscle mass were observed in relation to their genotype, the phenotype, and genotype interaction (online suppl. Table 1). In contrast, in the UCP2 Ala55Val, and UCP3 55C/T variants no significant differences were found related to the genotypes or the interaction between phenotype and genotype (online suppl. Tables 2, 3).

Association analyses were performed for the variables that had significant differences for the genotype. The UCP1 3826A/G variant was associated with high TC levels in NW-MUH subjects. There were no significant differences in the analysis of muscle mass. Otherwise, the UCP2 Ala55Val variant was linked to abdominal obesity in EW-MUH subjects (Table 3).

In the present study, the associations between the UCP1 -3826A/G (rs1800592), UCP2 Ala55Val (rs660339), and UCP3 -55C/T (rs1800849) variants with metabolic disturbances were evaluated according to metabolic phenotype in Mexican women. In subjects with the UCP1 -3826A/G variant, association with increased risk of hypercholesterolemia was found in subjects with NW-MUH. The UCP2 Ala55Val variant in EW-MUH subjects was associated with abdominal obesity, and the UCP3 -55C/T variant was not associated with metabolic phenotype or any variables according to their genotypes.

The UCP1 -3826A/G is one of the most studied genetic variants, with controversial results in different populations [10]. To knowledge, no previous studies have reported this variant in adult Mexicans. The present findings suggest that the G risk allele in UCP1 -3826A/G, is associated with higher TC serum concentrations; however, previous studies among Asian and European populations have reported no significant differences [7, 22, 23].

In a study on Turkish population, TC levels of 150 subjects with obesity were analyzed according to BMI categories where individuals with GG genotype had higher TC concentrations vs. subjects with the AA genotype (p = 0.027) [24]. Conversely, in Mexican adolescents, UCP1 -3826A/G variant was not associated with serum lipid levels, but was linked to a higher fat mass percentage (p = 0.002) [25].

The UCP1 -3826A/G variant is located in the regulatory region of the gene, and carriers of the G allele exhibit reduced expression of UCP1 mRNA, suggesting functional significance for this single nucleotide variant [26]. The UCP1 -3826A/G is likely to disrupt the transcription factor binding sites, thus reducing UCP1 expression of to some extent. UCP1 deficiency elevates the expression of the adipose stearoyl-CoA desaturase gene (SCD), which is related to sterol regulatory element binding protein-1c (SREBP-1c), resulting in increased availability of MUFA for TG and cholesteryl ester (CE) synthesis. The major products of SCD, palmitoleic acid and oleic acid, are key substrates for the formation of complex lipids such as phospholipids, TG, cholesterol esters, wax esters, and alkyl-2,3-diacylglycerols [23, 24]. This could explain why the genetic variant has an influence on lipid metabolism although further studies are required.

In these results, women with NW-MUH had a higher risk of hypercholesterolemia. The -3826A/G variant in UCP1 may favor the condition of unhealthy subjects in this group, not observed in EW subjects due to their chronic inflammation and altered metabolic parameters [27].

Moreover, in the present study UCP2 Ala55Val variant increases the risk of abdominal obesity in EW-MUH subjects. In Spain, those with the Val/Val genotype had higher waist circumference values compared to Ala/Ala and Ala/Val genotypes (p = 0.002) [28]. Also, Astrup et al. found lower 24-h energy expenditure and fat oxidation in homozygous risk allele carriers in a healthy Danish population [29].

The UCP2 Ala55Val, is a missense variant characterized by C/T substitution in exon 4 resulting in a conservative amino acid change from alanine to valine. There is no evidence that this variant causes a functional change in the protein [28]. However, this variant is closely linked with −866 G/A variant in the promoter region, and with a 45 bp insertion/deletion variant in the 3′untranslated region. The Ala55Val variant may not be directly responsible for causing disease but could instead reflect the effects of other associated variants [30]. Additionally, metabolically unhealthy individuals have greater abdominal circumference, related to higher metabolic alterations, indicating that fat distribution is crucial for a healthy metabolic profile, even in excess weight [31].

Regarding UCP3, investigations have not produced definitive results. One study found that in Spanish children and adolescents, there were no significant differences in BMI, glucose levels, insulin levels, HOMA index, or physical activity between individuals with and without obesity who carried the risk allele [32]. On the other hand, in Caucasian individuals with metabolic syndrome carrying the -55C/T UCP3 gene variant, concentrations of TC and LDL cholesterol were lower compared to those with the wild-type [33]. Similarly, in a Spanish population, carriers of the risk allele demonstrated a reduced risk of obesity when recreational energy expenditure was taken into account [34].

In this research, no significant differences were found for the -55C/T variant in UCP3. However, there is a study on Mexican population in which this genetic variant was associated with higher levels of blood pressure and hypertriglyceridemia [35]. However, this data was not presented in the study, as they were variables used for classifying subjects by metabolic phenotypes.

The associations observed between specific UCP variants and distinct metabolic disturbances suggest that these genetic markers could serve as potential indicators for early identification of individuals at risk for developing metabolically unhealthy phenotypes. This highlights the possibility of utilizing UCP genotyping as part of a precision nutrition approach to guide personalized lifestyle and therapeutic interventions aimed at mitigating cardiovascular risks and related conditions.

These findings further underscore the intricate interplay between genetic variants and metabolic traits, emphasizing the critical roles of other genes such as FTO, LEPR, FABP, and ADRB2 in regulating obesity and lipid metabolism. Variants in LEPR, particularly rs1137101, have been implicated in abdominal obesity and metabolic syndrome, while the FTO gene, specifically the rs9939609 variant, has been consistently associated with increased body fat and related metabolic disorders. Additionally, research on FABP and ADRB2 elucidates the complex genetic landscape that influences lipid accumulation and energy balance, suggesting that multiple genetic factors converge to impact obesity risk and lipid profiles [36].

Ethnic differences in allele frequencies and the complex processes involving metabolic abnormalities require further studies to elucidate molecular mechanisms, associations with metabolic phenotypes, lipid and lipoprotein-related diseases, and diet. Follow-up studies that analyze the interaction of key genetic variants with diet and thermogenesis are necessary. Finally, a greater understanding of gene influences on UCPs could help develop predictive tools for preventive intervention based on metabolic phenotype in terms of health maintenance and disease prevention.

A key strength of this study is the classification by metabolic phenotypes, using multiple variables to ensure homogeneous groups for comparison. Furthermore, to our knowledge, this is the first study to examine UCP variants in relation to BMI and metabolic phenotype classification, potentially paving the way for similar studies in different populations.

However, the study has some limitations, including the focus on the population of Western Mexico, which may restrict the generalizability of the findings. Larger sample sizes are needed in future studies to allow for better subgrouping and adjustments. Additionally, it is crucial to account for physical activity levels that will be essential in future analyses to provide more comprehensive insights.

Future research should focus on the interactions between genetic variants and environmental factors, such as diet and physical activity, to define preventive strategies tailored to diverse populations, considering the complex interplay of environmental, nutritional, and genetic factors. Additionally, increasing the sample size and analyzing the results by biological sex are suggested to corroborate the findings.

These findings underscore the role of the UCP1 -3826A/G and UCP2 Ala55Val variants in influencing specific aspects of metabolic health in women. The results associated with the UCP1 variant suggest the potential as a target for early cardiovascular risk intervention, while the UCP2 variant highlights the contribution to central obesity and metabolic syndrome. These results emphasize the necessity of integrating genetic screening into personalized strategies for managing metabolic disorders. Further research is needed to confirm these associations and investigate their mechanisms across diverse populations.

The authors thank the University of Guadalajara, and the subjects who participated in the study.

The study was reviewed and approved by the Research Ethical Committee of the University of Guadalajara (Registration number: CI/019/2010). The study was conducted according to the guidelines of the Declaration of Helsinki. Written informed consent was obtained from all subjects involved in the study.

The authors have no conflicts of interest to declare.

This research was funded in part by PROINPEP 2018 Universidad de Guadalajara, grant for E.S.-R., B.V. and E.M.-L. We also benefit of the support of the Programme for Strengthening Research and Postgraduate Studies 2020 (Programa de Fortalecimiento de la investigación y el posgrado 2020; REC/0342/2020) grant to E.M.-L.

E.S.-R and E.M.-L. designed the study, E.S.-R. and N.T.-C. collected and processed the data, W.C.P., B.V., and E.S.-R. analyzed the data, E.S.-R., W.C.P., N.T.-C., and E.M.-L. participated in data interpretation and manuscript preparation, B.V. and E.M.-L. funding acquisition. All authors read and approved the final manuscript.

The data that support the findings of this study are not publicly available due to confidentiality and privacy, data involves sensitive information about individuals, such as medical or personal data, it is crucial to protect the privacy of participants. Additional inquiries can be contacting to the corresponding author (E.M.-L. [email protected]).