Impact of the rs822393 Variant on Adiponectin Levels and Metabolic Parameters after Weight Loss Secondary to a High-Fat Hypocaloric Diet with Mediterranean Pattern Open Access

In recent years, there has been growing evidence to support the notion that weight loss plays a vital role in improving metabolic parameters and overall health. This is particularly relevant for individuals with metabolic syndrome, a condition characterized by a cluster of metabolic abnormalities such as obesity, insulin resistance, hypertension, and dyslipidemia [1]. Insulin resistance is the main cornerstone, and some interventional studies have reported that high intakes of saturated fat worsen, whereas high intakes of monounsaturated fats improved insulin resistance [2].

The effects of the rs822393 variant of ADIPOQ gene on metabolic parameters such as insulin resistance and adiponectin levels following weight loss through dietary intervention are still uncertain, as there are conflicting findings in the literature [3]. In this medium-term hypocaloric diet (9 months) with a Mediterranean partner increased adiponectin levels in non-T-allele carriers [3]. Adiponectin is an adipokine that plays a crucial role in regulating glucose and lipid metabolism. However, the impact of weight loss on adiponectin levels and the factors that influence these levels during weight loss interventions are not well understood [4]. Serum adiponectin levels are low in patients with obesity [5], and hypoadiponectinemia is related with insulin resistance and metabolic syndrome [6]. On the other hand, this adipokine enhances energy expenditure and fatty acid oxidation with a decrease of free fatty acids, improves insulin sensitivity with better glucose levels, and also possesses anti-inflammatory and antioxidant properties, protecting cardiovascular health [7]. Adiponectin is encoded by ADIPOQ gene on chromosome 3 at q27 [8], and this molecule has a lot of actions [4‒7].

In clinical practice, dietary interventions are modulated by gene background of the individuals [9]. Additionally, studies examining the impact of weight loss on adiponectin levels have yielded conflicting results. Some studies have shown an increase in adiponectin levels following weight loss interventions for obesity, while others have not observed such changes [9]. One genetic variant in ADIPOQ gene is rs822393 (−4522C/T), and this variant is located in the proximal promoter region. This single nucleotide polymorphism (SNP) rs822393 influences adiponectin promoter activity and generates low circulating levels of serum adiponectin and a high insulin resistance status [10]. Nutritional intervention studies are scarce in the literature; for example, an interventional study with a Mediterranean partner hypocaloric diet [11] reported a better response of serum adiponectin levels and HDL cholesterol in non-T-allele carriers than T-allele carriers. This diet had a moderate caloric restriction of 500 calories per day for 3 months with a contribution of monounsaturated fats that accounted for 50% of the caloric contribution from total fat amount. In other study with a similar design with two different dietary fat profiles and a similar caloric restriction [12], authors reported greater metabolic benefits in patients who do not carry the T allele. Finally, a study of caloric restriction lasting for 9 months demonstrated similar results as previous studies [3].

It would be interesting to assess whether increasing the percentage of total fat in the diet, compared to previous studies and with a Mediterranean diet pattern, would persist in producing beneficial metabolic effects. This is especially important to consider given the high consumption of fat in current diets [13]. Understanding the relationship between genetic variants, such as rs822393, and metabolic parameters and adiponectin levels is important in developing personalized interventions for obesity. Identifying individuals who may be more responsive to weight loss interventions based on their genetic profile could lead to more targeted and effective treatment strategies. The aim of this study was to evaluate the role of rs822393 of ADIPOQ on adiponectin levels and metabolic parameters after body weight loss with a high-fat hypocaloric diet with Mediterranean pattern during 3 months.

In this study, a cohort of participants with obesity were recruited, who underwent a dietary intervention aimed at promoting weight loss. Participants’ genetic profiles were analyzed for the presence of the rs822393 variant of ADIPOQ gene and metabolic parameters. The Ethics Committee (HCUVA Committee PI19/2021) approved the study protocol, and written informed consent from participants was obtained in all participants. A continuous consecutive methodology of sampling was realized to enroll 283 Caucasian patients with obesity taken care of at tertiary hospital in Spain.

The following were the eligibility criteria: patients with a body mass index greater than 30 kg/m² and an age over 30 years. On the other hand, the exclusion criteria included evidence of previous cardiovascular or cerebrovascular disease, severe renal or hepatic disorders, active alcohol consumption exceeding 20 g/day), uncontrolled hypothyroidism, diabetes mellitus, receiving drugs known to influence lipid levels such as fibrates and statins, hormonal therapy glucocorticoids and anti-inflammatory drugs), or glucose levels (such as sulfonylureas, thiazolidinedione insulin GLP-1 receptor antagonists, SGLT2 inhibitors, DPP-IV inhibitors, metformin.

At the start and after 3 months of dietary intervention, measurements were taken for various parameters related to adiposity including weight, body mass index, total fat mass, and waist circumference. Blood samples were collected after an overnight fast of at least 10 h and stored at −80°C until analysis. The following biochemical parameters were analyzed: total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, C-reactive protein (CRP), fasting glucose levels, insulin levels, leptin levels, and total adiponectin levels. Calculations included determining LDL cholesterol values, identifying insulin resistance using the homeostasis model assessment (HOMA-IR), and calculating the ratio of adiponectin/leptin.

During the 12-week study, the participants followed a high-fat calorie-restricted diet that aimed to provide approximately 500 kcal/day less than their estimated total energy expenditure. The high-fat diet consisted of carbohydrates accounting for 40% of the macronutrient composition, proteins contributing making up 20%, and fats contributing to 40%. In terms of dietary fat quality, it comprised 60% monounsaturated fatty acids, 30% saturated fatty acids, and 10% polyunsaturated fatty acids. To ensure understanding and address any concerns or questions regarding the prescribed diet, all participants had two individual sessions with a registered dietitian at the beginning of the trial. These sessions lasted for around 45 min each and included detailed explanations about following the diet plan as well as clarifying any doubts. The recommended food choices adhered to Mediterranean dietary patterns were legumes, vegetables, poultry, fish, and fresh fruit along with using olive oil for cooking purposes. As part of this intervention program, sugar-sweetened drinks and alcoholic beverages with high alcohol content were strictly prohibited. All enrolled individuals were instructed to keep a record of their daily food intake for three nonconsecutive days, including one weekend day. These sets of dietary records were collected 2 weeks before randomization to establish baseline diet data and then monthly during the randomization period to collect intervention diet data. Compliance was monitored weekly through phone calls by the dietitian in order to enhance adherence. The records were analyzed using a computer-based data evaluation system Dietsource^® (Nestle, Geneva, Switzerland), with national composition food tables serving as reference sources [14].

After removing buffy coats from blood samples, they were stored in tubes coated with EDTA at a temperature of −80°C. Genomic DNA was then extracted from 200 μL of the buffy coat using the QIAamp^® DNA blood kit and the standard salting out method. Using Beacon Designer 5.0 (PREMIER Biosoft International^®, LA, CA, USA), oligonucleotide primers and probes were designed. A PCR reaction mixture containing 60 μL was prepared by combining 2 μL of genomic DNA, 10 μL of a 10x Buffer Reaction (MyTaq^™ DNA polymerase, BIOLINE, LA, CA, USA), 1.5 μL each of forward and reverse primers and an additional amount of MyTaqDNA polymerase. PCR reaction was carried out using primer forward: 5′- ACG​TTG​GAT​GAA​AGC​ATG​ACA​CGG​AGC​TTC -3′ and reverse 5′- ACG​TTG​GAT​GAA​CCC​TCA​CCC​ATG​TCA​GC -3′ in a 2 μL final volume (Termociclador Life Technologies, LA, CA, USA). The DNA was heated to 90°C for 3 min to denature it, followed by 50 cycles of denaturation at 90°C for 30 s, and annealing at 56.1°C for 1 min. The forward sequence and reverse sequence of the internal standard GAPDH were used in RT-PCR. It has been used as internal standard for RT-PCR (GAPDH) with a forward sequence: GTCTCCTCTGACTTCAA and reverse sequence: ACCACCCTGTTGCTGTA. Patients with growth in both strands were classified as heterozygous, while those with growth in only one strand were classified as homozygous. The thermal cycler software categorized patients into homozygous wild type, heterozygous, or homozygous mutant. Each reaction included a negative control and control samples representing all genotypes. Hardy-Weinberg equilibrium was assessed using a statistical test called χ², which showed that the ADIPOQ gene variant was in Hardy-Weinberg equilibrium with a p value of 0.27.

After at least 10 h of fasting state, biochemical measurements were conducted using an automated analyzer, the COBAS INTEGRA 400^® Analyzer (Roche Diagnostic, Montreal, Canada). Total cholesterol, HDL cholesterol, triglycerides, CRP, fasting glucose, and insulin levels were determined. LDL cholesterol was calculated using the Friedewald formula: LDL cholesterol = total cholesterol−HDL-cholesterol−triglycerides/5 [15]. Insulin resistance was assessed using the homeostasis model assessment method: HOMA-IR = glucose x insulin/22.5 [16]. All adipokines were measured using enzyme immunoassay (RandD systems, Inc., Minneapolis, MN, USA). Serum levels of adiponectin, resistin, and leptin were measured with normal ranges of 8.65–21.43 μg/mL, 4–12 ng/mL, and 10–100 ng/mL respectively [17, 18]. The adiponectin/leptin ratio was calculated directly from these values. Additionally, CRP levels were determined by immunoturbidimetry with a normal range of 0–7 mg/dL.

To calculate body mass index, it has been obtained from height and weight measurements using a stadiometer (Omron, LA, CA, USA) and an electrical scale (Omron, LA, CA, USA), respectively. BMI was computed as the ratio of body weight in kilograms to the square of height in meters. Waist circumference was determined using a non-stretchable measuring tape from the midpoint between the lateral iliac crest and lowest rib with accuracy up to 0.5 cm. Total fat mass was assessed through impedance analysis with an accuracy level of 5 g by employing impedanciometry equipment (EFG BIA 101 Anniversary, Akern, Florence, Italy) [19]. Patients were at rest for 8 h before the BIA and did not consume alcohol, caffeine, or smoke.

Blood pressure readings were taken twice at 10-min intervals utilizing a digital sphygmomanometer (Omron, LA, CA, USA). All participants had rested beforehand, and they were advised to abstain from caffeine and physical activity for 8 previous hours.

The required sample size for detecting differences in adiponectin levels of at least 10 ng/dL after a dietary intervention, with a power of 90% and significance level of 5%, was determined to be 275 patients with obesity in each diet group. The analysis was conducted using a dominant genetic model, indicating rs822393 T-allele as the risk allele (CC vs. CT + TT). Variable distribution was analyzed using the Kolmogorov-Smirnov test. Descriptive statistics such as mean, standard deviation, and genotype frequencies were calculated for the data, followed by inferential analysis using χ² test, Student’s t test, and non-parametric tests. To reduce type I error in association analysis with multiple testing, Bonferroni correction was applied. ANCOVA (covariance analysis) adjusted for age, sex, BMI was used to assess the interaction between the gene and each dietary intervention on dependent variables. Statistical analyses were performed using IBM SPSS version 23.0 software package (SPSS Inc., Chicago, IL, USA). A significance level of p < 0.05 was considered statistically significant.

A total of 283 Caucasian patients with obesity were included in the study. These patients had three different genotypes: CC (59.0%), CT (33.6%), and TT (7.4%). The allelic frequencies for C and T were 0.89 and 0.20, respectively. The average age of all participants was 49.1 ± 3.7 years, ranging from 33 to 64 years old. The mean BMI was calculated as 36.0 ± 0.8 kg/m². A majority of patients were females (71%), while males were 29%. For comparison purposes, the wild genotype group showed similar ages than those carrying mutant type (wild type [CC] vs. mutant type [CT + TT]) (48.9 ± 4.1 years vs. 49.9 ± 6.1 years: ns). Gender distribution was similar in both genotype groups (wild type [CC] vs. mutant type [CT + TT]), males (27.5% vs. 26.3%) and females (72.5% vs. 73.7%).

All patients (n = 283) successfully completed the 12 weeks of follow-up period with dietary intervention, without any dropouts. Prior to intervention, a baseline evaluation of their dietary intake revealed an average daily energy consumption of approximately 2016.3 ± 318.8 kcal/day with the next macronutrient distribution: carbohydrate intake was 50.5% of total calories (259.9 ± 31.1 g/day), fat intake was 29.5% of total calories (70.1 ± 9.1 g/day), and protein intake was 20.0% of calories (88.1 ± 10.3 g/day). The distribution of basal dietary fats was 60.2% from saturated fats, 24.0% from monounsaturated fats, and 15.8% from polyunsaturated fats. No differences were detected in basal physical activity between both genotype groups (161.1 ± 21.3 min/week vs. 165.8 ± 18.9 min/week; p = 0.43). After dietary intervention, patients reached the goal recommendations (1,465.9 ± 129.8 kcal/day); carbohydrates 39% (140.1 ± 22.3 g/day) of the macronutrient composition, proteins 20% (75.1 ± 7.3 g/day), and fats contributing to 41% (62.2 ± 4.3 g/day). In relation with dietary fat quality, it was 61% monounsaturated fatty acids, 28% saturated fatty acids, and 11% polyunsaturated fatty acids. Physical activity remained similar in both genotype groups (168.9 ± 23.3 min/week vs. 169.8 ± 21.9 min/week; p = 0.49).

As demonstrated in Table 1, there were no differences in adiposity measurements and blood pressure levels, both before and after the dietary intervention in both genotypes. Following the hypocaloric diet, body weight, BMI, fat mass, waist circumference, and systolic blood pressure decreased in both genotypes, secondaries to dietary treatment. After dietary intervention, patients with both genotypes (CC vs. CT + TT) responded with similar improvement in body weight (−3.3 ± 0.9 kg vs. −3.7 ± 1.0 kg; p = 0.38), BMI (−0.9 ± 0.1 kg/m² vs. −1.1 ± 0.2 kg/m²; p = 0.31), fat mass (−2.8 ± 0.9 kg vs. −2.3 ± 0.8 kg; p = 0.31), waist circumference (−3.3 ± 0.9 cm vs. −3.8 ± 0.6 cm; p = 0.42), and systolic blood pressure (−13.8 ± 3.1 mm Hg vs. −11.2 ± 2.1 mm Hg; p = 0.36). Diastolic blood pressure remained unchanged in both genotypes.

In general, no significant differences in basal biochemical parameters were found for the variant rs822393 of the ADIPOQ gene, using a dominant model (CC vs. CT + TT) (Table 2). However, T-allele carriers showed lower levels of HDL at both baseline and posttreatment than non-T-allele carriers. After dietary intervention with this fat-enriched hypocaloric diet, patients in both genotype groups showed a significant improvement in total cholesterol levels (CC vs. CT + TT) (−18.1 ± 3.1 mg/dL vs. −21.2 ± 2.4 mg/dL; p = 0.41), LDL cholesterol (−17.1 ± 2.1 mg/dL vs. −18.3 ± 2.2 mg/dL; p = 0.41), insulin levels (−5.0 ± 0.8 mU/L vs. −4.6 ± 0.4 mU/L; p = 0.33), and HOMA-IR (−2.2 ± 0.3 units vs. −2.4 ± 0.3 units; p = 0.33). Moreover, HDL cholesterol levels (CC vs. CT + TT) (delta: 8.9 ± 1.1 mg/dL vs. 1.7 ± 0.8 mg/dL; p = 0.02) improved only in non-T-allele carriers after body weight loss.

Table 3 presents the results of serum adipokines and the adiponectin/leptin ratio. After the dietary intervention, there were significant improvements in leptin levels for both genotype groups. However, T-allele carriers had lower basal and post-intervention levels of adiponectin as well as a lower adiponectin/leptin ratio compared to non-T-allele carriers. After implementing dietary interventions (CC vs. CT + TT), there was a statistically significant increase in serum adiponectin levels in non-T-allele carriers (43.1 ± 5.9 ng/dL vs. 2.8 ± 3 0.0 ng/dL; p = 0.01) and adiponectin/leptin ratio (1.37 ± 0.1 units vs. 0.17 ± 0.08 units; p = 0.02).

In this interventional study involving Caucasian subjects with obesity, a significant association has been observed between the rs822393 polymorphism of the ADIPOQ gene and changes in metabolic response following a high-fat hypocaloric diet with Mediterranean pattern. Among non-T-allele carriers, there was a notable increase in HDL cholesterol levels as well as adiponectin levels and ratio of adiponectin to leptin. On the other hand, T-allele carriers exhibited lower levels of HDL cholesterol, adiponectin, and ratio of adiponectin to leptin when compared to non-T-allele carrier.

The literature on the rs822393 variant of the ADIPOQ gene is limited and has yielded contradictory findings. One study, called The Healthy Lifestyle in Europe by Nutrition in Adolescents Study found a correlation between this genetic variant and HDL cholesterol levels in young individuals, which is agreed with the current study involving adult participants [10]. Interestingly, these associations with lipid profile have not been replicated in other research studies. However, some studies have identified a potential link between rs822393 and cardiovascular risk, suggesting its possible involvement in modulating the lipid profile [20]. Overall, the rs822393 polymorphism of the ADIPOQ gene appears to have an impact on metabolic response and lipid profile, particularly HDL cholesterol levels [21]. This association of rs822393 with HDL cholesterol levels has been described in different ethnicities, and Rayma et al. [22] had shown the same low HDL cholesterol levels in the T-allele carriers found by us in the current sample of Caucasian patients, but in their case in Asian subjects with obesity. Another metabolic discovery in our research is the presence of reduced levels of adiponectin in individuals who carry the T allele. This relationship has been documented previously in the CARDIA (Coronary artery development in young adults) study [23]. The decreased adiponectin levels may be attributed to rs822393 being an intronic variant that can potentially alter alternative-splicing patterns [24]. Perhaps, the observed metabolic findings, specifically low levels of HDL cholesterol and serum adiponectin levels, may be interconnected. This is because adiponectin has the ability to regulate HDL cholesterol levels by activating lipoprotein lipase and ATP-binding cassette transporter A1, and finally, this fact can increase hepatic production of ApoA1 [24, 25], too.

The effects of dietary interventions and the interaction with this genetic variant of ADIPOQ gene have been barely evaluated. A prior 3-month study, involving a caloric intake of 1,500 calories per day and adherence to a Mediterranean diet that consisted of 50% monounsaturated fats, yielded noteworthy enhancements in adiponectin levels (around 20 ng/dL), the ratio of adiponectin to leptin (around 0.5 points), and HDL cholesterol levels (around 6 mg/dL) [11]. In another intervention study of 12 weeks, similar metabolic outcomes were observed with two different hypocaloric diets, one diet enriched with polyunsaturated fatty acids and the other with monounsaturated fatty acids [12]. The amount of the effect on the 3 variables was similar to the previous study. In both diets, the quantity of fat was around 30% of energy, less than the present study, targeted at 40% of caloric intake. In a recent design, the participants followed a 1,000 calorie per day diet for a duration of 9 months [3]. The diet consisted of a higher proportion of monounsaturated fats, accounting for approximately 25% of the total energy intake as fats. The amount of changes in HDL cholesterol, adiponectin, and ratio adiponectin-leptin was similar that the above-mentioned studies. Perhaps the quantity of total fat (40% present study) in the diet and perhaps its percentage of monounsaturated fats (60% present study) have a predominant role in the response of adiponectin to body weight loss after these low-calorie diets. In the present study, the increase in HDL cholesterol was 9 mg/dL, the increase in adiponectin was 43 mg/dL and the increase in the adiponectin-leptin ratio was 1.5 points. These three increases were higher than the previous increases described in the literature [3, 11, 12]. Besides, previous research has shown that a diet enriched in unsaturated fats can lead to higher levels of adiponectin compared to diets high in protein or carbohydrates [26]. Additionally, a study found that supplementation with unsaturated fatty acids for 3 months resulted in a significant increase in adiponectin levels [27], too. In animal models, modifying the proportion of saturated/monounsaturated/polyunsaturated fatty acids in the diet was able to change the expression of adiponectin when following a high-fat diet [28]. Even among individuals who are not obese, research has shown that the response of postprandial adiponectin can vary depending on the amount of fat in a typical meal [29]. Additionally, the GOLDN study found a significant link between a specific genetic variant of the ADIPOQ gene and obesity as well as metabolic-related factors when dietary modifications were taken into account [30].

In this study design, the intervention controlled dietary trial with high retention and adherence was a major strength. However, there were some limitations to this study, too. First, only small changes in HDL cholesterol and adiponectin levels have been detected. Additionally, only one specific SNP of the ADIPOQ gene was evaluated, while other SNPs may also be relevant to these findings or perhaps there are interactions between different SNPs or linkage imbalances. Second, interactions between ADIPOQ variants and gene-environmental factors were not considered in this study design which could modify the associations between these variants and metabolic changes. Third, these findings are limited to metabolic alterations rather than cardiovascular events; however, an intriguing relationship between the adiponectin/leptin ratio exists that shows promise for identifying individuals at increased metabolic risk [31]. Specifically, carriers of the T allele showed an increase in this ratio before and after dietary intervention indicating greater metabolic risk compared to those without this allele. Finally, the self-reported dietary intake and physical exercise is not reliable, and it might include bias of under- or over-reporting.

In summary, individuals with obesity and without the T allele of rs822393 experienced improvements in adiponectin levels, adiponectin/leptin ratio, and HDL cholesterol after following a high-fat hypocaloric diet with a Mediterranean pattern. Furthermore, this improvement is superior to that reported in the literature using a diet very rich in monounsaturated fats. As such, it may be worthwhile to genotype patients for the rs822393 variant before initiating a diet and to implement a fat-enriched hypocaloric diet in order to reach a better metabolic response. This prediction is more significant than considering the dietary fat profile alone.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional and/or National Research Committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. And it was approved by our local committee (Hospital Clinic University of the Valladolid Committee PI19/2021). Written Informed consent was obtained from all individual participants included in the study.

Daniel de Luis declares that he has no conflict of interest. David Primo declares that he has no conflict of interest. Olatz Izaola declares that she has no conflict of interest. Daniel Rico declares that he has no conflict of interest. Juan Jose Lopez declares that he has no conflict of interest.

The authors have no funding sources to declare. No funding was received for the development of this article in any of its stages.

Daniel de Luis designed the study, realized statistical analysis, and wrote the article. Juan Jose Lopez realized anthropometric evaluation and revised the article. Daniel Rico realized biochemical evaluation and revised the article. Olatz Izaola realized anthropometric evaluation and control of dietary intake. She contributed to the acquisition of data for the work and revised it critically for important intellectual content. She approved the version to be published and accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. D. Primo realized biochemical evaluation and genotype and wrote the article. Daniel de Luis, Juan Jose Lopez, Daniel Rico, and D. Primo contributed to the analysis of data for the work and revised it critically for important intellectual content. Daniel de Luis, Juan Jose Lopez, Daniel Rico, Olatz Izaola, and D. Primo approved the version to be published and accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.