Effect of weight loss on circulating fatty acid profiles in overweight subjects with high visceral fat area: a 12-week randomized controlled trial

Study subjects were recruited through advertisements posted in Seoul. Based on the data screened by the Clinical Nutrition Lab, Yonsei University, overweight subjects [25.0 kg/m² ≤ body mass index (BMI) < 30 kg/m²] were referred to the Department of Internal Medicine, Yonsei University Severance Hospital. After their health and basic blood tests, including serum glucose, were rechecked, individuals who met the study criteria were recommended for study participation. The inclusion criteria were as follows: age between 20 and 60 years; absence of pregnancy or breastfeeding; stable body weight (body weight change < 1 kg in the 3 months before screening); BMI between 25 and 30 kg/m²; high L4 VFA (L4 VFA ≥ 100 cm²); absence of hypertension, type 2 diabetes, cardiovascular disease, and thyroid disease; and no use of medication affecting body weight, energy expenditure, or glucose control for 6 months prior to screening. Subjects were excluded if they had a history of Cushing syndrome, malignancy, or liver disease, including chronic viral hepatitis, autoimmune hepatitis, primary biliary cirrhosis, and drug-induced liver disease. Male and female subjects whose alcohol consumption was > 40 and > 20 g/day, respectively, and subjects with a history of intentional weight reduction in the 6 months before the current study were excluded. Finally, those who consented to participate in the program were included in this study. The purpose of the study was carefully explained to all participants, and written consent was obtained prior to their participation. The protocol was approved by the Institutional Review Board of Yonsei University and Yonsei University Severance Hospital in accordance with the Helsinki Declaration.

The basic framework of the present study was based on a previous study [12]. A 12-week, placebo-controlled, randomized study was conducted with 80 high VFA, nondiabetic overweight subjects (L4 VFA ≥ 100 cm²). Subjects were divided into two groups: a weight-loss group with 12 weeks of mild calorie restriction (a 300 kcal/day intake reduction) or a control group with no treatment (NCT02992639; http://​www.​clinicaltrials.​gov/). Randomization was achieved by computer-generated block randomization (placebo:test = 1:1).

The subjects in the weight-loss group followed a 12-week weight-loss program comprising a 300 kcal/day reduction of their usual caloric intake. Basically, the subjects were instructed to remove 1/3 of a bowl of rice (approximately 100 kcal; a bowl of rice is approximately 300 kcal per meal, according to the food composition table from the Rural Development Administration of Korea [13]) from one meal a day for easier application of the 300 kcal/day deficit. Moreover, based on each subject’s reported food intake, an individualized and nutritionally balanced diet plan was produced by a trained nutritionist to achieve the goal of losing a minimum of 3% of initial body weight. The instructions included food choices, cooking methods, reductions in snack consumption frequency, low-calorie substitutions for high-calorie foods, low-fat foods, and limitations on simple sugar consumption. The subjects in the control group were instructed to maintain their usual dietary and physical activity habits during the study period.

At baseline, the subjects’ usual dietary intakes were assessed via 24-h recall and semiquantitative food frequency questionnaires completed with the nutritionist’s assistance. In addition, a standardized 3-day dietary record was obtained from each participant at week 6 and 12. These records were completed at home following detailed instructions from the nutritionist. A computerized version of the Korean Nutrition File (Can-Pro 3.0; The Korean Nutrition Society, Seoul, Korea) was used to determine the macronutrient contents of foods and total daily energy intakes. On the same days as the dietary records, 3-day physical activity records were completed at home. The total energy expenditure (TEE) (kcal/day) was calculated from activity patterns, including the basal metabolic rate (BMR), physical activity for 24 h, and the specific dynamic action of food. Each subject’s BMR was calculated using the Harris-Benedict equation.

Detailed information was previously published [14]. Briefly, all data were acquired at baseline and week 12. Body weight (Inbody370; Biospace, Cheonan, Korea), height, waist circumference, and blood pressure (BP) were measured. BMI was calculated in units of kilogram per square meter (kg/m²). The abdominal fat distribution was measured at L4 by computed tomography (CT). The scanning parameters were a slice thickness of 1 mm at 200 mA and 120 kVp, with a 48-cm field of view. Abdominal adipose tissue was determined using the attenuation range from − 150 to − 50 Hounsfield units in the CT images. The body composition of the study participants was measured via dual-energy X-ray absorptiometry (DEXA) to determine the fat percentage, fat mass, and lean body mass.

Details were provided previously [14]. Blood samples were collected after an overnight fast of at least 12 h. Venous blood specimens were collected in EDTA-treated tubes and serum tubes. The blood samples were centrifuged to obtain plasma and serum samples, which were then stored at − 70 °C. The levels of serum-fasting glucose, insulin, triglycerides (TGs), total cholesterol, and high-density lipoprotein (HDL) cholesterol were measured [15]. The IR and low-density lipoprotein (LDL) cholesterol levels were calculated using the equations of homeostatic model assessment (HOMA) and the Friedewald formula, respectively: HOMA-IR = [fasting insulin (μIU/mL) × fasting glucose (mg/dL)]/405; Friedewald formula: LDL cholesterol = total cholesterol – [HDL cholesterol + (TG/5)].

The details of the GC-MS procedure have been previously published [16]. Briefly, plasma samples (100 μL) were placed into PTEE screw-capped Pyrex tubes. An internal standard compound containing heptadecanoic acid (500 μL, 25 ppm in n-hexane) and methanol (2 mL) were added to the samples. After vortexing, acetyl chloride (100 μL) was added slowly, and the tubes were then heated at 95 °C for 1 h. After cooling on ice, 6% potassium carbonate (5 mL) was added to the tubes. The tubes were centrifuged (4 °C, 3000 rpm, 5 min), and the clear n-hexane top layer (100 μL), including the fatty acid methyl esters (FAME), was then transferred into a vial prior to GC-MS analysis.

GC-MS analyses were performed on an Agilent Technologies 7890 N gas chromatograph coupled with an Agilent Technologies 5977A quadrupole mass selective spectrometer with a triple-axis detector (Agilent, Palo Alto, CA, USA) in the electron ionization (70 eV) and full scan monitoring (m/z 50–800) modes. The injection volume of the samples was 1 μL in the splitless mode. The fatty acid methyl esters of all samples were separated on a VF-WAX column (Agilent Technologies, Middelburg, Netherlands) with helium as a carrier gas. The temperature program was as follows: (1) initial temperature of 50 °C for 2.3 min, (2) increase to 175 °C at 50 °C/min, and (3) increase to 230 °C at 2 °C/min. The metabolites in the samples were identified by comparing their relative retention times and mass spectra with those of authentic reference standards. The relative levels of metabolites were calculated by comparing their peak areas to those of the internal standard compound.

The statistical analyses were conducted with SPSS version 23.0 software (IBM/SPSS, Chicago, IL, USA). Skewed variables were logarithmically transformed. An independent t-test compared continuous variables between the control and weight-loss groups. A paired t-test was used to compare continuous variables at baseline and follow-up within each group. For the comparison of categorical variables, a chi-squared test was conducted. A general linear model test was performed to adjust the baseline values of the changed values. A Pearson’s correlation coefficient was used to examine the relations among variables. A heat map was created using Multiple Experiment Viewer (MeV) version 4.9.0 (http://​www.​tm4.​org/​) to visualize and evaluate the relations among variables in the study population. The results are expressed as the mean ± standard error (SE). A two-tailed P < 0.05 was considered statistically significant.

The age, sex distribution, smoking, and drinking did not significantly differ between the two groups (Table 1). At baseline, no significant differences were observed in all clinical and biochemical parameters between the control and weight-loss groups; after the 12-week intervention, the BMI, serum insulin levels, HOMA-IR score, and serum TG levels were significantly decreased in the weight-loss group compared with the control group (Table 1). In the control group, the subjects showed slight but significant weight gain (weight and BMI; all P < 0.001) during the intervention, whereas the subjects in the weight-loss group showed significant reductions in not only the weight and BMI (both P < 0.001) but also the waist circumference (P < 0.001), systolic BP (P = 0.049), insulin levels (P = 0.001), and HOMA-IR score (P = 0.003) following the intervention (Table 1). In addition, these reductions in weight, BMI, and waist circumference in the weight-loss group were greater than those in the control group after adjusting for the baseline values (all P < 0.001) (Table 1). The serum insulin levels, HOMA-IR score, and serum TG levels also showed significantly greater decreases in the weight-loss group than in the control group after adjusting for the baseline values (P = 0.020, P = 0.034, and P = 0.022, respectively) (Table 1). The estimated TEE (2144.6 ± 45.3 kcal/d vs. 2201.3 ± 59.3 kcal/d; P = 0.519), total calorie intake (TCI) (2105.8 ± 40.0 kcal/d vs. 2200.5 ± 50.3 kcal/d; P = 0.159), and percentage of macronutrient intakes (%CHO:%PRO:%FAT = 61:15:22 in both groups) did not significantly differ between the two groups at baseline (data not shown). The subjects’ overall compliance with the weight-loss plan was good. The estimated TCI of the weight-loss group at 12-week follow-up was, on average, 1949.4 kcal (%CHO:%PRO:%FAT = 59:19:22); and the types of dietary fats [saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), and polyunsaturated fatty acids (PUFAs)] ingested by the study participants did not demonstrate any changes during the intervention (data not shown).

Table 2 presents the effects of a 12-week mild calorie-restricted diet on the percent of body weight change, abdominal fat area measured by CT, and body composition measured by DEXA. Comparison of the percent of body weight changes (differences from baseline) revealed that the weight-loss group had greater reductions in the percent of body weight change than the control group (0.63 ± 0.15 vs. -3.44 ± 0.34; P < 0.001). Regarding the results of CT, no significant differences were observed in the L4 abdominal fat area (whole fat area and VFA), L4 subcutaneous fat area, or visceral-to-subcutaneous area ratio (VSR) at baseline between the control and weight-loss groups. At follow-up, the L4 VFA in the weight-loss group was lower than in the control group (P < 0.001). The VFA decreased by 17.7 cm² (15%) from baseline in the weight-loss group (P < 0.001), and the VSR decreased by 0.09 from baseline in the weight-loss group (P < 0.001). Significantly greater reductions in the L4 VFA and VSR were observed in the weight-loss group compared with the control group (all P < 0.001). No statistically significant shifts in the VFA, subcutaneous fat area, or VSR were identified in the control group (Table 2). With regard to the DEXA results, there were no significant differences in the fat percentage, fat mass, or lean body mass between the two groups at both baseline and follow-up (Table 2). After the intervention, significant decreases in the fat percentage and fat mass were observed in both the control and weight-loss groups; however, greater reductions in the fat percentage and fat mass were observed in the weight-loss group than in the control group after adjustment for the baseline values (P = 0.006 and P < 0.001, respectively). The control group showed a significant increase in the lean body mass, whereas the weight-loss group showed a significant decrease in the lean body mass at the 12-week follow-up; these changes were significantly different between the two groups (P < 0.001) (Table 2).

Table 3 presents the effects of a 12-week mild calorie-restricted diet on plasma fatty acid profiles. At baseline, no significant difference was observed in the SFAs, MUFAs, n-6 PUFAs, and n-3 PUFAs between the control and weight-loss groups. The total SFAs, including palmitic acid (C16:0) and stearic acid (C18:0), decreased from baseline in the weight-loss group; in addition, these reductions were significantly greater than those in the control group after adjusting for the baseline values (P = 0.027, P = 0.033, and P = 0.021, respectively). At follow-up, the weight-loss group showed significantly lower levels of total SFAs (P = 0.029), lauric acid (C12:0) (P = 0.032), and palmitic acid (C16:0) (P = 0.028) than the control group. Similarly, the total MUFAs, including palmitoleic acid (C16:1, n-7) and oleic acid (C18:1, n-9), decreased from baseline in the weight-loss group, and these decreases were significantly greater than those in the control group after adjustment for the baseline values (P = 0.017, P = 0.006, and P = 0.018, respectively). At follow-up, the weight-loss group showed significantly lower levels of total MUFAs (P = 0.021), palmitoleic acid (C16:1, n-7) (P = 0.018), and oleic acid (C18:1, n-9) (P = 0.024) than did control group. Although the total n-6 and n-3 PUFA levels did not significantly change, eicosadienoic acid (C20:2, n-6) and dihomo-γ-linolenic acid (C20:3, n-6) decreased from baseline in the weight-loss group. Both showed greater reductions in the weight-loss group than in the control group after adjusting for the baseline values (P < 0.001 and P = 0.003, respectively). At follow-up, the weight-loss group showed significantly lower levels of eicosadienoic acid (C20:2, n-6) (P < 0.001) and dihomo-γ-linolenic acid (C20:3, n-6) (P = 0.001) than the control group (Table 3).

C16 Δ9-desaturase activity significantly decreased from baseline in the weight-loss group, and this reduction was greater in the weight-loss group than in the control group after adjusting for the baseline values (P = 0.014). The activity of Δ5-desaturase significantly increased from baseline in the weight-loss group, and this increase was greater in the weight-loss group than in the control group after adjusting for the baseline values (P = 0.001). In addition, at follow-up, Δ5-desaturase activity was significantly higher in the weight-loss group than in the control group (P < 0.001). Elongase activity significantly decreased after the intervention (Table 3).

Overall, 22 variables showed significant changes in the weight-loss group compared with the control group after adjusting for the baseline values [Table 1: weight, BMI, waist circumference, insulin, HOMA-IR, and TGs; Table 2: whole fat area, VFA, VSR, fat percentage, fat mass, and lean body mass; Table 3: SFAs, palmitic acid (C16:0), stearic acid (C18:0), MUFAs, palmitoleic acid (C16:1, n-7), oleic acid (C18:1, n-9), eicosadienoic acid (C20:2, n-6), dihomo-γ-linolenic acid (C20:3, n-6), C16 Δ9-desaturase, and Δ5-desaturase]. A correlation analysis between changes in L4 VFA and changes in these variables was performed. Of the total study subjects (n = 75), the changes in L4 VFA demonstrated significant positive correlations with changes in weight (r = 0.596, P < 0.001), BMI (r = 0.596, P < 0.001), waist circumference (r = 0.317, P = 0.006), insulin (r = 0.259, P = 0.025), fat percentage (r = 0.241, P = 0.037), fat mass (r = 0.381, P = 0.001), lean body mass (r = 0.464, P < 0.001), and dihomo-γ-linolenic acid (C20:3, n-6) (r = 0.235, P = 0.042) and a significant negative correlation with changes in Δ5-desaturase activities (r = −0.277, P = 0.016) (Fig. 1).