A randomized clinical trial (ClinicalTrials.gov, Number NCT04315376) was conducted at the Institute of Translational Nutrigenetics and Nutrigenomics of the University of Guadalajara. The study was performed from February 2018 to February 2019. Eighty-three participants were recruited through flyers and social media invitations, however 61 met the inclusion criteria (18 male and 43 female). The inclusion criteria were mestizos participants from West Mexico, age between 25 to 50 years with obesity (BMI ≥30 kg/m2), with a waist circumference greater than 80 centimeters in women and 90 centimeters in men, a sedentary lifestyle according to the World Health Organization (WHO) , and subjects without history of medication for at least 1 year. Non-inclusion criteria were pregnant or breastfeeding women, diagnosis of diabetes, cardiovascular disease, and cancer, tobacco and alcohol (consumption ≥ 40 g of alcohol per day for men and ≥20 g for women) consumption, and muscle or joint injury. All the participants signed a written informed consent. This study was approved by the Ethics and Biosafety Committee of the Health Sciences Center, University of Guadalajara (Registration number CI-08518) and was carried out according to the Declaration of Helsinki (2013).
The intervention consisted a 4-month follow up period. The participants were randomly assigned to the diet and exercise program intervention (diet-exercise group), or only diet program intervention (diet group). Also, we defined the baseline measures as Time 0, and the final measures (4th month) as Time 4.
The participants who performed the diet-exercise intervention, completed an Astrand-Ryhming submaximal test described by Astrand  on RECK MOTOmed viva2 cycle ergometer to estimate the VO2max before the exercise intervention program (50-75 W for woman and 50-100 W for man), and to design the moderate-intensity structured exercise program based on the heart rate, which was monitored during the test in each individual. Once the individuals completed the training program, the final Astrand-Ryhming test was performed considering the individuals as trained, therefore, the measured watts were greater according to the Astrand protocol (75-100 W for women and 100-150 W for men).
Exercise program intervention
A personal trainer certified by the American College of Sports Medicine designed a progressive three-phase moderate-intensity exercise program: 1) conditioning phase: 45 minutes/day, 3 days per week for five weeks (~65%HR), 2) progression phase: 1 hour/day, 4 days a week for eight weeks (~70-75%HR), and 3) maintenance phase: 1 hour/day, 5 days per week for three weeks (~75%HR). The main exercises consisted of improving aerobic (walking and jogging), speed (functional exercise circuits and short races), and resistance performance using dumbbells (weighing less than 5 kilograms). During all sessions, breathlessness and fatigue were measured with the Borg CR10 grade scale (0-10). The personal trainer supervised three sessions per week within the sports facilities of the University Center to ensure that all exercises were performed with an appropriate technique. For the two unsupervised training days, participants were given the detailed training program to perform in the house or park.
The nutrition service carried out the dietary intervention and follow-up evaluation each month. The validated 24-hour diet record and the three-day dietary food record questionnaire (two days during the week and one day of the weekend) were used to collect the dietary information. Participants were instructed to provide the correct description of their habitual food intake. The nutritionist showed food scale models from Nasco® to enhance the accuracy of the portion sizes based on Mexican food composition tables. The Nutritionist Pro™ software (Axxya Systems, Woodinville, WA) was used to estimate the energy intake and food consumption quantification by a trained nutritionist. The nutritional intervention consisted of a hypocaloric-diet 20% total energy expenditure reduction estimated by the Mifflin-St.Jeor formula  using the current weight, with a nutrient distribution as follows: 50% carbohydrates, 30% lipids, and 20% proteins.
Anthropometric and biochemical measurements
Height and weight were measured after an 8-12 hour fast and with participants wearing light clothes. Height measurements were taken using a stadiometer (Rochester Clinical Research, New York, USA). Waist circumference was measured using a Lufkin Executive® tape and body composition by electrical bioimpedance (InBody 370, Biospace Co. Seoul, Korea) every month of the intervention period.
Peripheral blood samples were taken after an 8-12 hour fast and immediately centrifuged at 3500rpm to obtain serum. Biochemical variables were performed using a dry chemistry analyzer (Vitros 250 Analyzer, Ortho-Clinical Diagnostics, Johnson & Johnson Services, Inc., Rochester, NY, USA). Low-density lipoprotein cholesterol (LDL-c) was calculated using the Friedewald formula except when triglycerides levels were higher than 400 mg/dL. The atherogenic index was calculated using the formula [Total cholesterol (mg/dL) / HDL-c (mg/dL)], and for cardiovascular risk, values >3 in women and >3.5 in men, were considered. Serum insulin levels were determined using Insulin Model ELISA kit, Catalog: CT-600101A (International Diagnostics, S.A de C.V) following the supplier's instructions. The homeostatic model assessment of insulin resistance (HOMA-IR) was calculated as described by Matthews . A HOMA-IR value >2.5 was considered as insulin resistance. The biochemical variables were measured every month during the intervention period.
ASC mRNA expression and cytokine levels
The ASC mRNA expression from peripheral blood and cytokine levels from serum samples were measured before the start (Time 0) and at the end of the intervention (Time 4). Total RNA was extracted from 5ml of peripheral blood samples, and the quantification and purity were measured by Nanodrop 2000 UV (Thermo Fisher Scientific, Waltham, MA). The cDNA synthesis was performed using 1 µg of total RNA according to standard techniques  of Invitrogen™, Carlsbad, CA.
Quantitative real-time PCR was performed using Hs00203118_m1 TaqMan® probes (Applied Biosystems, Foster City, CA) in a Light Cycler 96 system considering standard PCR conditions (Roche, Mannheim, Germany) to analyze ASC mRNA relative expression by 2-ΔΔCt method. The amplification reactions were performed in duplicate using the ACTβ gene (Hs01060665_g1, Applied Biosystems, Foster City, CA) as a constitutive gene to normalize the samples.
The quantification of pro-inflammatory and anti-inflammatory cytokines was performed using Bio-Plex Pro™ Human cytokine Standard 17-Plex, Group I kit (Cat. 10014905, Bio-Rad-Laboratories, Hercules, CA) following the supplier's instructions and read immediately using the MAGPIX™ analyzer. The IL-18 cytokine was measured with Interleukin-18 Human ELISA kit (Biovendor, Brno, Czech Republic) in accordance to the manufacturer´s instructions.
The sample size was calculated using the mean difference formula for clinical trials in order to detect statistical differences in ASC gene expression in peripheral blood using the data from Dang WT, 2015 study . To achieve a statistical power of 80% and an alpha of 5%, a sample size of 13 participants in each study group was required. However, considering the predicted loss of participants during the study, more than 13 individuals were included per study group. Shapiro-Wilk test was used to evaluate data normality and the Levene’s test to verify the homogeneity of the variables. Quantitative variables are expressed as mean ± standard deviation (SD) or standard error of the mean (SEM) when the analysis was adjusted for co-variables. Statistical differences between groups were analyzed using the unpaired Student’s t-test, and paired Student's t-test or Wilcoxon test to evaluate changes between baseline and final time depending on the normality of the variables, and we used repeated measures ANOVA to observed differences over time. Pearson and Spearman correlation coefficients were considered normality dependent. A P-value <0.05 was considered statistically significant. All statistical analyses were performed using the software SPSS v.22.0 (IBM, Chicago, IL).