Metabolically normal obese people are protected from adverse effects following weight gain (original) (raw)
Subjects
The flow chart of study subjects is shown in Figure 4. Between August 2010 and February 2014, a total of 68 potential subjects were screened for this study, 36 of whom were considered eligible. Subjects were recruited through the Volunteers of Health Database at Washington University School of Medicine and by local postings. Eleven subjects discontinued the study because of relocation (n = 1), withdrawal of consent (n = 9), and nonstudy-related surgery (n = 1); 5 subjects were excluded from the final analyses because they did not achieve the predetermined acceptable 5%–7% weight gain. Therefore, data were collected and final analyses were performed on a total of 20 subjects. All parts of this study were conducted at the Clinical Research Unit of Washington University School of Medicine. IHTG content was used to identify subjects who were either MNO (IHTG <5.6%, _n_ = 12, age = 43 ± 10 years, 2 males, 5 African-Americans, 1 Native-American, and 6 subjects of mixed European descent) or MAO (IHTG >10%, n = 8, age = 52 ± 7 years, 4 males, 2 African-Americans, and 6 subjects of mixed European descent), because increased IHTG content is a robust marker of inappropriate fat distribution and metabolic dysfunction (3, 7, 37). All subjects completed a comprehensive medical evaluation. To minimize potential weight gain–induced health risks to the study subjects, those with extreme obesity (BMI ≥40 kg/m2), IHTG content greater than 25%, diabetes, poorly controlled obesity comorbidities (e.g., blood pressure >150/100), or other serious diseases were excluded. No subject smoked tobacco or took medications that could affect the study’s outcome measures.
Study flow chart. A total of 68 subjects were assessed for eligibility, 36 of whom were considered eligible. Eleven subjects withdrew from the study, and 5 subjects did not achieve the acceptable 5%–7% weight gain, thus data were collected and analyses performed on 12 MNO and 8 MAO subjects.
Study design
Body composition. Body FM and FFM were determined using dual-energy x-ray absorptiometry (Lunar iDXA; GE Healthcare). VAT and s.c. abdominal adipose tissue volumes were quantified using MRI (Siemens; Analyze 7.0 software; Mayo Clinic, Mayo Foundation), and IHTG content was determined using magnetic resonance spectroscopy (Siemens) (38).
Hepatic, skeletal muscle, and adipose tissue insulin sensitivity. The subjects were admitted to the Clinical Research Unit at Washington University School of Medicine on the evening before the clamp procedure. At 1900 hours, the subjects were served a standard meal and then fasted until study completion the next day. At 0500 hours the following morning, one catheter was inserted into a forearm vein to infuse stable isotopically labeled tracers, dextrose and insulin, and a second catheter was inserted into a radial artery in the contralateral hand to obtain blood samples. At 0600 hours, a primed (22.5 μmol/kg), continuous (0.25 μmol/kg/min) infusion of [6,6-2H2]glucose (Cambridge Isotope Laboratories) was started, followed by a continuous (6 nmol/kg FFM/min) infusion of [U-13C]palmitate (Cambridge Isotope Laboratories) bound to albumin at 0800 hours. After infusion of the glucose tracer for 3.5 hours and palmitate tracer for 1.5 hours (basal period), a 2-stage hyperinsulinemic-euglycemic clamp procedure was started and continued for 6 hours. Insulin was infused at a rate of 7 mU/m2 BSA/min (initiated with a priming dose of 28 mU/m2/min for 5 min and then 14 mU/m2/min for 5 min) during stage 1 (3.5–6.5 h) and at a rate of 50 mU/m2 BSA/min (initiated with a priming dose of 200 mU/m2/min for 5 min and then 100 mU/m2/min for 5 min) during stage 2 of the clamp procedure (6.5–9.5 h). These 2 insulin infusion rates were chosen to evaluate hepatic and adipose tissue insulin sensitivity (low-dose insulin infusion to submaximally suppress endogenous glucose production and adipose tissue lipolysis) and skeletal muscle insulin sensitivity (high-dose insulin infusion to stimulate muscle glucose uptake) (39). The infusion rates of glucose and palmitate were reduced by 50% during stage 1 and turned off during stage 2 of the clamp procedure to account for expected decreases in hepatic glucose production and lipolytic rates. Euglycemia (~100 mg/dl) was maintained by variable infusion of 20% dextrose, which was enriched to 2.5% with [6,6-2H2]glucose to help ensure a constant glucose tracer-to-tracee ratio (TTR). REE was determined during the basal period of the clamp procedure by using online expiratory gas exchange analyses (TrueOne 2400; Parvo Medics). Blood samples were collected before beginning the tracer infusion to determine the background plasma substrate TTR and every 10 minutes during the final 30 minutes of the basal period and in stages 1 and 2 of the clamp procedure to determine glucose and insulin concentrations and substrate kinetics.
VLDL apoB100 kinetics. Approximately 1 week after the hyperinsulinemic-euglycemic clamp procedure, the subjects were readmitted to the Clinical Research Unit to assess VLDL apoB100 kinetics. At 1900 hours, the subjects were served a standard meal and then fasted until study completion the next day. At 0500 hours the following morning, a catheter was inserted into a forearm vein to infuse the stable isotopically labeled leucine tracer. A second catheter was inserted into a contralateral hand vein, which was heated to 55°C by using a thermostatically controlled box to obtain arterialized blood samples. At 0600 hours, a primed (4.2 μmol/kg) constant (0.062 μmol/kg/min) infusion of [5,5,5-2H3]leucine (Cambridge Isotope Laboratories) was started and maintained for 12 hours (4). Blood samples were obtained before the start of the tracer infusion and at 5, 15, 30, 60, 90, and 120 minutes and then every hour for 10 hours after starting the tracer infusion to determine the leucine TTR in plasma and in VLDL apoB100.
Adipose tissue biopsies. Adipose tissue biopsies were obtained during the basal stage of the clamp procedure at ~0700 hours. After anesthetizing the skin and underlying tissues by percutaneous injection of lidocaine, abdominal s.c. adipose tissue was aspirated through a 4-mm liposuction cannula (Tulip Medical Products) connected to a 30-cc syringe from the periumbilical area. Tissue samples were immediately rinsed with ice-cold saline and frozen in liquid nitrogen before being stored at –80°C for subsequent RNA extraction.
High-calorie diet and weight gain. After baseline studies were completed, the subjects were instructed to consume 1,000 kcal/day more than their estimated baseline total energy requirements (calculated as 1.25 times the measured REE) (40), but maintain the same relative macronutrient composition as their baseline diet, until they gained ~6% (acceptable range of 5% to 7%) of their initial body weight. This additional food intake was achieved by having subjects eat specific menu choices from among 5 fast-food restaurant chains (Burger King, Kentucky Fried Chicken, McDonald’s, Pizza Hut, and Taco Bell). The subjects were required to keep daily food records and meet individually and weekly with the study dietitian to help ensure dietary compliance. Four consecutive days of food records were collected, validated, and analyzed at baseline and at ~4 weeks after the subjects started the high-calorie diet to assess total daily energy intake and dietary macronutrient composition. The subjects were also seen every week for a medical examination and blood tests to ensure medical safety. After the subjects achieved the targeted 5%–7% gain in body weight, they were kept weight stable for more than 2 weeks before repeating the same procedures conducted at baseline. To help maintain weight stability, the subjects were instructed to reduce their energy intake by approximately 200 kcal/day. At each weekly visit, the subjects were weighed, and dietary recommendations were adjusted as needed to keep their weight stable. Once the final studies were completed, the subjects were enrolled in a weight-loss program, supervised by our study dietitian and behavioral therapist, until they lost at least as much weight as they gained during the study.
Analyses of samples and calculations. Plasma glucose concentration was measured by an automated glucose analyzer (Yellow Springs Instruments Co.). Plasma insulin concentration was measured by using electrochemiluminescence technology (Elecsys 2010; Roche Diagnostics). Plasma adiponectin and leptin concentrations were determined by performing ELISA and a radioimmunoassay, respectively (both from EMD Millipore). Plasma samples for BCAAs were prepared according to the manufacturer’s instructions (Phenomenex) for analyses by gas chromatography–mass spectrometry (41). The glucose, palmitate, and leucine TTR in plasma and the leucine TTR in VLDL particles were determined by using gas chromatography–mass spectrometry (3, 4). VLDL was separated from plasma by density-gradient ultracentrifugation (42). Plasma VLDL apoB100 concentrations were measured by using an immunoturbidimetric kit (Wako).
HOMA-IR was measured from the subjects’ fasting plasma glucose levels and insulin concentrations (43). Isotopic steady-state conditions were achieved during the final 30 minutes of the basal period and stages 1 and 2 of the clamp procedure, and Steele’s equation for steady-state conditions was used to calculate substrate kinetics (44). Glucose Ra in plasma was calculated by dividing the glucose tracer infusion rate by the average plasma glucose TTR during the last 30 minutes of the basal period and in stages 1 and 2 of the clamp procedure. During the clamp procedure, the endogenous glucose production rate was calculated by subtracting the dextrose infusion rate from total glucose Ra. Glucose Rd from plasma was assumed to equal total glucose Ra. Palmitate Ra was calculated by dividing the palmitate tracer infusion rate by the average plasma palmitate TTR obtained during the final 30 minutes of the basal period and in stage 1 of the clamp procedure. Hepatic and adipose tissue insulin sensitivity was assessed as the relative decrease in glucose and palmitate Ra, respectively, from basal to low-dose insulin infusion (stage 1) of the clamp procedure. Skeletal muscle insulin sensitivity was determined as the relative increase in glucose Rd from basal to high-dose insulin infusion (stage 2) of the clamp procedure (7).
The fractional turnover rate (FTR) of VLDL apoB100 (in pools/h) was determined by fitting the TTR of leucine in plasma and in VLDL apoB100 to a multicompartmental model, as previously described (42). The rate of VLDL apoB100 secretion into plasma (in nmol/l/min) was calculated by multiplying the FTR of VLDL apoB100 (in pools/min) by the steady-state concentration of VLDL apoB100 (in nmol/l) and the plasma volume (in liters). Plasma VLDL apoB100 clearance (in ml/min) was calculated by multiplying the FTR of VLDL apoB100 (in pools/h) by the plasma volume (in liters), divided by 60.
Total RNA was isolated from frozen s.c. adipose tissue samples by using QIAzol and an RNeasy Mini Kit (QIAGEN). Microarray analyses were performed with the GeneChip Human Gene 1.0 ST Array (Affymetrix). PAGE was performed as previously described (22, 23). Gene sets were obtained from the gene set enrichment analysis (GSEA) database (gene ontology [GO] gene sets, C5 collection; http://www.broad.mit.edu/gsea/msigdb/msigdb_index.html), and Z scores and P values were calculated for each gene set. A P value of less than 0.05 was considered statistically significant. All data were analyzed using the R statistical software package (http://www.bioconductor.org). Microarray datasets used in this study have been deposited in the NCBI’s Gene Expression Omnibus database (GEO GSE62832). Real-time PCR was performed on key enzymes involved in adipose tissue lipogenesis (FADS1, FADS2, and ELOVL6). Gene expression was determined using an ABI 7500 real-time PCR system (Invitrogen) and SYBR Green Master Mix (Invitrogen). The expression of each gene was determined by normalizing the Ct value of each sample to the housekeeping control gene, ribosomal protein (36B4). Primer details are listed in Supplemental Table 1 (supplemental material available online with this article; doi:10.1172/JCI78425DS1).
Statistics
The statistical analyses were based on a hierarchy of outcomes: (a) skeletal muscle insulin sensitivity was considered the primary outcome; (b) hepatic and adipose tissue insulin sensitivity and VLDL apoB100 kinetics were considered secondary outcomes; and (c) all other variables (body composition, other cardiometabolic variables, and adipose tissue gene expression) were considered exploratory analyses. All datasets were tested for normality according to the Shapiro-Wilks method, and non-normally distributed variables were log transformed for analysis and back transformed for presentation. The effect of weight gain was determined by repeated-measures ANCOVA, with the intervention as the within-subjects factor (before vs. after overfeeding) and the group as the between-subjects factor (MNO vs. MAO), and adjusting for sex and race as covariates. When significant interactions between the intervention and group were found, a 2-tailed Student’s t test was used to evaluate the effect of treatment within each group (paired Student’s t test) and differences between groups at baseline and after intervention (independent Student’s t test). Results are presented as the mean with SDs (for normally distributed variables) or 95% CIs (for non-normally distributed variables), unless otherwise indicated. A P value of 0.05 or less was considered statistically significant. Analyses were performed using SPSS software, version 19 (SPSS Inc.).
On the basis of the interindividual variability of insulin-mediated suppression of glucose and palmitate Ra and insulin-mediated stimulation of glucose Rd, assessed by using the hyperinsulinemic-euglycemic clamp procedure in a large cohort of nondiabetic obese subjects we had studied previously (39, 45), we estimated that 8 subjects in each group (MNO and MAO) would be needed to detect between-group differences in insulin sensitivity of ≥19% in hepatic tissue, ≥25% in adipose tissue, and ≥29% in skeletal muscle, with a β value of 0.20 (i.e., 80% power) and an α value of 0.05. We estimated that 15–20 subjects would need to be recruited in each group to ensure that an adequate number of subjects completed the study, after allowing for dropouts and subjects who were not able to achieve the required 5%–7% weight gain.
Study approval
Subjects provided written informed consent before participating in this study, which was approved by the Human Research Protection Office of the Washington University School of Medicine.