Low protein diets produce divergent effects on energy balance

Diets deficient in protein often increase food consumption, body weight and fat mass; however, the underlying mechanisms remain poorly understood. We compared the effects of diets varying in protein concentrations on energy balance in obesity-prone rats. We demonstrate that protein-free (0% protein calories) diets decreased energy intake and increased energy expenditure, very low protein (5% protein) diets increased energy intake and expenditure, whereas moderately low protein (10% protein) diets increased energy intake without altering expenditure, relative to control diet (15% protein). These diet-induced alterations in energy expenditure are in part mediated through enhanced serotonergic and β-adrenergic signaling coupled with upregulation of key thermogenic markers in brown fat and skeletal muscle. The protein-free and very low protein diets decreased plasma concentrations of multiple essential amino acids, anorexigenic and metabolic hormones, but these diets increased the tissue expression and plasma concentrations of fibroblast growth factor-21. Protein-free and very low protein diets induced fatty liver, reduced energy digestibility, and decreased lean mass and body weight that persisted beyond the restriction period. In contrast, moderately low protein diets promoted gain in body weight and adiposity following the period of protein restriction. Together, our findings demonstrate that low protein diets produce divergent effects on energy balance.

. Diet composition. Table S2. Effects of low protein diets on organ weights and energy digestibility. Table S3. Plasma concentrations (nmol/mL) of essential and nonessential amino acids in obesity-prone rats after 14 days of protein restriction. Table S4. Feed efficiency during protein restriction and realimentation periods in obesityprone rats. Table S5. The host, dilution and supplier of primary and secondary antibodies for immunoblotting. Table S6. The primer sequence (forward, F, and reverse, R), location on template (base pairs, bp), amplicon size, and GenBank accession numbers for target and reference genes used for qPCR in the current study. Figure S2. Effects of low protein diets on energy balance. (a) Dark period energy intake, (b) light period energy intake, (c) mean energy expenditure (EE) in the dark period, (d) mean energy expenditure in the light period, (e) body fat percentage, (f) body lean percentage, (g) total daily heat production using lean mass as covariate, (h) dark period heat production using lean mass as covariate and (i) light period heat production using lean mass as covariate in obesity-prone rats. Following feeding different dietary treatments for 14 days, all animals were provided the control diet (15% protein) for 14 days during a realimentation phase. Dotted line separates the restriction and realimentation phases. Values are mean ± SEM, n=13-16/group. *P < 0.05 vs 15P, 15% protein; 10P, 10% protein; 5P, 5% protein; 0P, 0% protein. Figure S3. Effects of low protein diets on energy intake. Cumulative hourly caloric intakes for (a) day 4, (b) day 7, (c) day 11, (d) day 14, (e) day 18, (f) day 21, (g) day 25, and (h) day 28 in obesity-prone rats. Following feeding different dietary treatments for 14 days, all animals were provided the control diet (15% protein) for 14 days during a realimentation phase. Black bar denotes the dark period. Values are mean ± SEM, n=13-16/group. *P < 0.05 vs 15P, 15% protein; 10P, 10% protein; 5P, 5% protein; 0P, 0% protein.

General Maintenance and Husbandry
The general maintenance and husbandry was according to our previously published procedures 1 . Briefly, the Comprehensive Lab Animal Monitoring System (CLAMS®, Columbus Instruments; Columbus, OH, USA) was started at 1100 h every day and stopped at 0900 h in the following day (22h measurement). General maintenance and husbandry including calibration of the sensors, filling of the feeders and water bottles were carried out between 0900 to 1100 h. Further, during this period, body weight and body composition measurements, intraperitoneal glucose tolerance test (IPGTT) and drug injections were conducted. Throughout the study period, the animals had ad libitum access to fresh food provided on alternate days and water except the days before IPGTT, drug injections and meal test when food was taken away but not water for 16 h (overnight fasting).

Measurements of Food Intake and Energy Expenditure
Food intake and energy expenditure were recorded daily by the CLAMS® system throughout the study as we reported previously 1 . Briefly, the rats had access to powdered food through a center feeder assembly resting on a balance, and the balance weights were recorded at ~30 sec intervals. The energy intake of individual rats was measured by multiplying the food intake (g) by physiological fuel value of diets (4.4 kcal/g). The total energy expenditure was measured by open circuit indirect calorimetry with negative flow controllers providing fresh-air to all chambers at 2 L/min. Air was sampled from each cage for 5 sec following a 55 sec stabilization period, and after every two cages, a reference air measurement was recorded. Each day before starting the CLAMS ® , the O2 and CO2 sensors were calibrated with a calibration gas (20.50% O2, 0.5% CO2) at 5-10 psi. Oxygen consumption rate (VO2 ml/kg body weight/h), carbon dioxide production rate (VCO2 ml/kg body weight/h) and respiratory exchange ratio (RER) were measured and the total energy expenditure was computed as calorific value (CV) × VO2, where CV = 3.815 + 1.232 × RER 2 and data were expressed as kcal/(kg lean mass/h).

Blood Sampling and Tissue Harvesting
Following our previous procedures 1,3 , blood samples were collected on ice into tubes containing ethylenediaminetetraacetic acid (EDTA; 1.5 mg/ml blood), protease inhibitor cocktail (10 µl/ml blood; Sigma-Aldrich, Oakville, ON, Canada) and dipeptidyl peptidase IV inhibitor (DPP-IV inhibitor; 10 µl/ml blood; Millipore Corporation, Temecula, CA, USA). Samples were immediately centrifuged at 4°C (4,000 rpm for 15 min), plasma was collected and stored at -80°C until further analyses. Following the last blood sampling, the animals were euthanized by using sodium pentobarbital (120 mg/kg IP; Euthanyl®, Bimeda-MTC, ON, Canada) and interscapular brown adipose (BAT), liver, leg skeletal muscle, kidney and heart tissues were collected, rinsed in sterile phosphate buffer saline, weighed and immediately snap-frozen in liquid nitrogen, and stored at -80°C until further analyses. Liver lipid content was measured using a biopsy probe of the Minispec LF110® NMR Analyzer (Bruker Corporation, Milton, ON, Canada).

Measurement of Plasma Amino Acid Concentrations
Plasma amino acid concentrations were measured as described previously 4 . Briefly, plasma amino acid concentrations were measured for terminal samples collected at 120 min (except cystine, proline and hydroxy proline), by using a fluorometric high performance liquid chromatography method involving pre-column separation of amino acid derivatives using a derivatizing agent, o-phthaldialdehyde. Amino acids were separated by gradient elution from a Supelcosil LC-18 column (15 cm ×4.6 mm, 3 µm) and quantified with amino acid standards (Sigma-Aldrich, Oakville, ON, Canada).