Funding agencies: This research was supported by the INSERM, Aquitaine Region and Ajinomoto 3ARP Research Program, European Community's Seventh Framework Programme (FP7-People2009-IEF-251494), and Fondation Recherche Médicale. F.J.B.S. is recipient of a research contract from the National System of Health (Instituto de Salud Carlos III; CP07/00283) and of a BAE from Instituto de Salud Carlos III (BA09/90066). This work was supported by INSERM, Aquitaine Region and Ajinomoto 3ARP research program (to DC), European Community's Seventh Framework Programme FP7-People2009-IEF-251494 (DC and EB) and Fondation Recherche Médicale. F.J.B.S. is recipient of a research contract from the National System of Health (Instituto de Salud Carlos III; CP07/00283) and of a BAE from Instituto de Salud Carlos III (BA09/90066).
Disclosure: The authors declared no conflict of interest.
Author contributions: E.B. and F.J.B.S. carried out experiments and analyzed data; M.E., T.L.L., I.B., S.C., and A.D. carried out experiments; G.M. analyzed data; E.B. and D.C. wrote the manuscript; all the authors edited and approved the manuscript; D.C. conceived and supervised the study.
High-protein diets favor weight loss and its maintenance. Whether these effects might be recapitulated by certain amino acids is unknown. Therefore, the impact of leucine supplementation on energy balance and associated metabolic changes in diet-induced obese (DIO) mice during and after weight loss was investigated.
DIO C57BL/6J mice were fed a normocaloric diet to induce weight loss while receiving or not the amino acid leucine in drinking water. Body weight, food intake, body composition, energy expenditure, glucose tolerance, insulin, and leptin sensitivity were evaluated. Q-PCR analysis was performed on muscle, brown and white adipose tissues.
DIO mice decreased body weight and fat mass in response to chow, but supplementation with leucine did not affect these parameters. During weight maintenance, mice supplemented with leucine had improved glucose tolerance, increased leptin sensitivity, and lower respiratory quotient. The latter was associated with changes in the expression of several genes modulating fatty acid metabolism and mitochondrial activity in the epididymal white and the brown adipose tissues, but not muscle.
Leucine supplementation might represent an adjuvant beneficial nutritional therapy during weight loss and maintenance, because it improves lipid and glucose metabolism and restores leptin sensitivity in previously obese animals.
Diet composition is particularly important in determining metabolic changes that might lead to obesity, insulin resistance, and the development of diabetes and cancer . On the other hand, appropriate dietary or nutritional interventions might be also efficacious in favoring body weight loss and improving metabolic performance. In particular, there is now rather convincing evidence that a higher protein intake favors loss of fat mass while preserving lean mass during states of negative energy balance in humans . Surprisingly, though, little is known on whether the beneficial weight-control effects characterizing high-protein diets might be due to certain types of amino acids. Over the years, several studies have focused on the ability of branched-chain amino acids (BCAA) to help preserve muscle mass under different pathological conditions . Among the BCAA, leucine is the most effective in inducing protein synthesis in muscle by stimulating the mammalian target of rapamycin (mTOR) pathway . A lot of information is available on the use of leucine as a supplement in clinical setting, to help preserve muscle mass during wasting syndromes, as well as an anabolic enhancer consumed by athletes in the attempt to increase their muscle mass . Nevertheless, only recent studies carried out in rodents have shown that this amino acid might have an important role in the regulation of energy balance and metabolism [6-9]. In particular, quite a few studies were published lately reporting evidence of decreased body weight gain and fat mass, increased energy expenditure, and improved lipid and glucose metabolism in animals supplemented with leucine while consuming a high-fat diet (HFD) [8-11]. In contrast, other evidence has clearly shown that circulating levels of BCAA are increased both in obese humans and animals [12-15] while dropping after weight loss [16, 17]. Additionally, elevated BCAA levels prognosticate future risk of developing diabetes . These phenomena might be explained by the inhibition of mitochondrial enzymes involved in BCAA catabolism, particularly in the white adipose tissue (WAT), with concomitant increase in fatty acid oxidation and changes in mitochondrial redox status that might in turn lead to insulin resistance [15, 16, 19]. Thus, although some controversy exists about the role of BCAA in obesity, no study has so far investigated whether leucine supplementation might recapitulate the well-known beneficial effects of high-protein diets during weight loss and weight maintenance. Here, we therefore determined the overall impact of leucine supplementation on energy balance and associated changes in glucose and lipid metabolism in diet-induced obese (DIO) mice during and after body weight loss induced by dietary intervention.
Six-week-old male C57BL/6J mice (Janvier, Le Genest-Saint-Isle, France) were housed individually in standard plastic rodent cages and maintained at 22°C on a 12-h light-dark cycle (lights off at 1 P.M.). After 1 week of acclimation, mice had ad libitum access to a HFD containing 60% calories from fat, 20% from protein and 20% from carbohydrate (5.24 kcal/g, D12492, Research Diets, New Brunswick, NJ) for 8 weeks. Thereafter, animals were placed on standard chow pellets containing 30% calories from fat, 20% from protein, 50% from carbohydrate (3.2 kcal/g, Standard Rodent Diet A03, SAFE, Augy, France) for 21 weeks and evenly distributed by food intake, body weight, and body composition in two groups receiving either water or water supplemented with 1.5% (wt./vol.) l-leucine (Sigma-Aldrich, St. Louis, MO) as in .
All experiments were conducted in strict compliance with the European Union recommendations (2010/63/EU) and were approved by the French Ministry of Agriculture and Fisheries (animal experimentation authorization n° 3309004) and by the local ethical committee. At the end of the study, 36-weeks old animals were sacrificed during the light phase by cervical dislocation, tissues collected in ice-cold isopentane and dry ice, trunk blood centrifuged at 3,000 rpm for 15 min, and tissues and plasma samples stored in −80°C until further analysis. At time of sacrifice, pancreatic islets were isolated for subsequent in vitro studies. Number of animals used for each experiment is detailed in the figure legends.
Body composition analysis
A nuclear echo magnetic resonance imaging whole-body composition analyzer (Echo MRI 900; EchoMedical Systems, Houston, TX) was used to assess body fat and lean mass in conscious mice. MRI analysis was carried out at week 8 of exposure to the HFD (baseline of the study, just before the animals were switched to the standard chow diet) and at week 8, 15, and 21 of exposure to chow.
Mice were individually housed in metabolic chambers (TSE systems GmbH, Bad Homburg, Germany) in which fluid, food intake, in cage locomotor activity, and gas exchanges can be monitored. The experiment was carried out as in , after 18 weeks of exposure to chow.
Glucose tolerance test (GTT) and insulin tolerance test (ITT)
Animals were injected with 2 g/kg of d-glucose (Sigma Aldrich, St. Louis, MO) for the GTT or with 0.5 U/kg of insulin (Humulin, Lilly, France) for the ITT. For the GTT, animals were fasted for 16 h, whereas for the ITT animals were fasted and leucine-deprived for 7 h, as in . Blood samples were taken from the tail vein and glucose concentration was measured using glucose sticks (OneTouch Vita, Lifescan France, Issy les Moulineaux, France). At time 0 of the GTT, a blood sample was also collected, centrifuged at 3,000 rpm for 15 min, and the obtained plasma stored at −80°C for subsequent measurement of insulin. HOMA index was calculated using the formula (fasting glucose mmol/L × fasting insulin mU/L)/22.5. GTT and ITT were performed after 7 and 8 weeks, respectively, on HFD, as well as after 13 and 14 weeks on chow.
Leptin food intake study
The study was conducted as in . The experiment was carried out after 15 weeks of exposure to chow and it was performed twice, with a within-subject design in which mice received both leptin [mouse recombinant leptin, 5 mg/kg, i.p., obtained from Dr. A.F. Parlow, National Hormone and Pituitary Program (Torrance, CA)] and its vehicle phosphate-buffered saline (PBS) in counterbalanced order.
Hormonal and lipids analysis
Plasma high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol were determined with Abcam ELISA kits (Abcam, Paris, France) and free fatty acids with an Abcam colorimetric assay kit. Plasma leptin and insulin were, respectively, measured using ELISA kits from Bertin Pharma (Paris, France) and Mercodia (Uppsala, Sweden) following the manufacturer's instructions. Hepatic triglycerides were extracted adapting published methods . Briefly, samples were weighed and homogenized with a TissueLyser (Qiagen, Courtaboeuf, France) in PBS. Homogenate was extracted with a 2:1 (v:v) chloroform–methanol solution. Samples were centrifuged at 6,500 rpm at 4°C for 5 min. The organic phase was transferred and nitrogen-dried. Samples were finally resuspended in PBS and quantified using a triglyceride colorimetric assay kit (Cayman Chemicals, Ann Arbor, MI).
The intestine (proximal jejunum) was rinsed in ice-cold 0.1M PBS and immediately frozen. Glucose-6-phosphatase (G6Pase) and phosphoenolpyruvate carboxykinase (PEPCK) activities were assayed as described previously .
Islets isolation and in vitro glucose-stimulated insulin secretion
All reagents used were from Sigma-Aldrich. Pancreatic islets were isolated by collagenase digestion method, as in . Briefly, pancreas was inflated with Hanks solution containing 0.8 mg/mL of collagenase, 5.6 mM glucose, and 0.05% bovine serum albumin, pH 7.4, then the tissue was removed and kept at 37°C for 6–8 min. After tissue digestion, the islets were manually collected and left recovering in RPMI culture media for 18–20 h. For static incubation experiments, groups of size-matched five islets were first incubated for 1 h at 37°C in 0.5 mL Krebs-bicarbonate buffer solution (in mM): 115 NaCl, 5 KCl, 2.56 CaCl2, 1 MgCl2, 24 NaHCO3, 15 HEPES, and 5.6 glucose, supplemented with 0.05% of bovine serum albumin and equilibrated with a mixture of 95% O2:5% CO2, pH 7.4. The medium was then replaced with 0.5 mL fresh buffer containing 3 mM glucose or 11 mM glucose and further incubated for 1 h. The islets were subsequently put at 4°C for 15 min to stop insulin secretion and the media were collected, centrifuged at 1,200 rpm for 10 min at 4°C, and stored at −20°C for measurement of insulin content by ELISA (Mercodia).
Quantitative real-time PCR (q-PCR)
Tissue samples (epididymal white and brown adipose tissue (BAT), soleus, and gastrocnemius muscle) were homogenized in Trizol (Fermentas, Fisher Scientific SAS, Illkirch, France) and RNA was isolated using a standard chloroform/isopropanol protocol. RNA was processed and analyzed following an adaptation of published methods . Q-PCR reactions were done in duplicate for each sample, using transcript-specific primers, cDNA (4 ng) and LightCycler 480 SYBR Green I Master (Roche) in a final volume of 10 μL. For the determination of the reference gene, the Genorm method was used. Relative expression analysis was corrected for PCR efficiency and normalized against two reference genes (see Supporting Information Table S1). The relative level of expression was calculated using the comparative (2−ΔΔCt) method. Primers sequences are reported in Supporting Information Table S1.
All values are reported as mean ± SEM. Statistics were performed using Statistica version 9 (Statsoft, Maisons-Alfort, France). Body weight loss, GTT and ITT glucose curves as well as energy expenditure, respiratory quotient, and food intake response after leptin injection were analyzed with repeated measures ANOVA, and LSD Fisher post hoc test was applied when appropriated. An ANOVA was carried out also for the analysis of body composition. Q-PCR, in vitro islets data, and hormonal and lipid measurements were analyzed with a two-tailed t-test. P values less than 0.05 denote statistical significance.
Seven-week-old male C57BL/6J mice were initially fed a 60% HFD for 8 weeks. At the end of the 8-weeks exposure to the HFD, the mice had gained around 56% of their initial body weight, reaching a mean weight of 36 ± 0.9 g (Figure 1a, first point in gray in the figure). At this time, the animals were all switched to a chow diet, which rapidly induced loss of body weight and fat mass (Figure 1a and 1b), and evenly distributed in two groups, supplemented (chow-leu) or not (chow-water) with the BCAA leucine in drinking water at the time of the diet change. Chow water- and leucine-supplemented mice had a similar daily leucine intake from the diet (chow-water: 73 ± 8.2 mg/day vs. chow-leu: 77 ± 1.3 mg/day, P > 0.05), but leucine-supplemented mice ingested an additional 67 ± 3.8 mg of leucine daily via water consumption, almost doubling their total daily leucine intake. Leucine supplementation, however, did not affect body weight, fat mass, or lean mass changes eventually induced by placing the animals on a standard chow (Figure 1a–1c), nor did it affect the mean daily food intake (chow-water daily FI: 4.68 ± 0.1 g vs. chow-leu daily FI: 4.56 ± 0.1 g; P > 0.05, N = 9–11 per group) during the entire period of the study.
Indirect calorimetry analysis done after 18 weeks in chow showed that both groups of mice had similar energy expenditure and respiratory quotient changes in response to the light/dark cycle (Figure 1d and 1e). However, chow-leu mice had a significantly lower respiratory quotient compared with the chow-water group during the light period (Figure 1e, P < 0.05), suggesting an increase in the use of fatty acids rather than carbohydrates as energy substrates. Although, plasma levels of free-fatty acids, HDL-, and LDL-cholesterol as well as hepatic triglycerides content were comparable between the two groups (see Table 1). Subsequent molecular analysis carried out on epididymal WAT and on BAT, revealed significant changes in several molecular markers of mitochondrial function and fatty acid metabolism. In particular, leucine-supplemented mice had significantly increased mRNA expression levels of the mitochondrial marker cyclooxygenase-III (Cox-III) and uncoupling protein 1 and 3 (UCP-1 and UCP-3) in the WAT together with a trend toward increased expression (P = 0.08) for cyclooxygenase-IV (Cox-IV) (Figure 2a). These changes were associated with increased expression of genes participating in fatty acid metabolism (Figure 2a), including stearoyl-CoA desaturase 1 (SCD-1), encoding for the enzyme controlling the synthesis of unsaturated fatty acids , as well as acetyl-CoA carboxylase (ACC) (P = 0.07), and fatty acid synthase (FAS), which are involved in the synthesis of fatty acids , and carnitine palmitoyltransferase-1 (CPT-1), the enzyme mediating the transport of long-chain fatty acids across the mitochondrial membrane so that they can undergo β-oxidation  (Figure 2a). Similarly, as compared with the chow-water group, the BAT of leucine-supplemented mice had increased mRNA expression levels of Cox-III, UCP-3, CPT-1, and peroxisome proliferator-activated receptor-alpha, the latter a transcription factor and a major regulator of lipid metabolism, whose activation leads to increased β-oxidation  (Figure 2b). Conversely, no significant molecular changes were observed in the muscle (Figure 3).
Table 1. Lipids and leptin levels in chow-water and chow-leu mice
To then determine whether leucine supplementation might affect glucose metabolism in vivo, fasting glucose and insulin were assessed and GTT and ITT studies were carried out. Fasting glucose levels were significantly higher in HFD as compared with both chow-water and chow-leu groups (glucose-HFD: 254 ± 10 mg/dL, chow-water glucose: 189.3 ± 10.3 mg/dL, chow-leu glucose: 150.9 ± 12.3 mg/dL, P < 0.05). Of note, fasting glucose levels were significantly lower in chow-leu mice as compared with the chow-water group (P < 0.05). However, fasting plasma insulin and resulting HOMA index did not significantly differ between the two groups (chow-water insulin: 51.2 ± 13.1 mU/L vs. chow-leu insulin: 48.6 ± 5.3 mU/L, P > 0.05; chow-water HOMA: 24 ± 6.7 vs. chow-leu HOMA: 18.9 ± 2.9, P > 0.05, N = 9–11 per group). As compared with the GTT carried out after 8 weeks of HFD consumption, the subsequent chronic exposure to the standard chow diet significantly improved glucose tolerance in both chow-water and chow-leu animals (Figure 4a and 4b). Importantly, leucine supplementation further and significantly ameliorated glucose tolerance (Figure 4a and 4b).
Differently, although exposure to chow significantly improved insulin sensitivity during an ITT as compared with the insulin-induced glucose changes obtained in the same animals while on HFD, leucine supplementation showed no further effect (Figure 4c and 4d).
In vitro studies evidenced a comparable response to glucose challenge in islets of Langerhans from chow-water and chow-leu mice, with no differences in glucose-stimulated insulin secretion (Figure 4e). In addition, subsequent analysis of the activity of G6Pase and PEPCK in the jejunum did not show any differences between the two groups of mice in intestinal gluconeogenesis (data not shown), a mechanism known to affect insulin sensitivity in response to protein ingestion .
Diet-induced obesity is classically associated with resistance to the well-known anorectic actions of leptin . It is, therefore, expected that weight loss might restore responsiveness to leptin.
Circulating leptin levels were comparable between water- and leucine-supplemented mice (see Table 1), in accordance with the similar fat mass in the two groups (Figure 1b). To therefore test whether chronic exposure to chow and supplementation with leucine might affect leptin-related responses, we assessed the ability of leptin to acutely decrease food intake in chow-water and chow-leu mice (Figure 5). Interestingly, while the administration of the hormone did not affect food intake in the chow-water group, it significantly decreased food intake in leucine-supplemented animals (Figure 5), suggesting that chronic supplementation with leucine increases sensitivity to the anorectic action of leptin.
In humans, diets high in protein are known to favor weight loss and its maintenance under states of negative energy balance . However, whether these effects might depend upon specific amino acids is at present unclear. Here, we therefore tested the effects of leucine supplementation on body weight loss and its maintenance in hyperglycemic, DIO mice that initially lost weight through the exposure to a chow diet. Our data show that a twofold increase in dietary leucine intake did not lead to differences in body weight, food intake, or body composition after weight loss and during its maintenance, while it increased fatty acid oxidation and improved glucose tolerance and leptin sensitivity.
Differently from what initially expected, leucine supplementation during body weight loss did not affect food intake or body weight. Nor did it impact the physiological weight gain, typical of aging animals, that was observed starting from week 10 on chow (Figure 1a). When leucine supplementation is associated to food restriction in lean, nonobese rodents, it leads to decreased body weight and fat mass and to preservation of lean mass . Differently, in this study DIO animals had ad libitum access to the chow diet, which may have accounted for the lack of effect of the BCAA, particularly on body composition. However, it is also possible that if animals had underwent a less dramatic body weight loss than the one observed in our study, beneficial effects of leucine supplementation on body composition could have been unveiled. Nevertheless, chow-fed, leucine-supplemented mice had a lower respiratory quotient as compared with the chow-water group, a phenomenon that could be explained taking into account the ketogenic nature of leucine and that was associated with changes in the expression of molecular markers regulating mitochondrial activity, fatty acid synthesis, and oxidation in both white and BATs. These data suggest a shift in fuel substrates caused by the chronic supplementation with leucine and are consistent with a recent clinical study reporting decreased respiratory quotient and increased fat oxidation in obese subjects undergoing leucine supplementation without calorie restriction for a period of 4 weeks . In vitro studies carried out on LT3-L1 adipocytes and in C2C12 muscle cells have also shown that leucine increases fatty acid oxidation, as determined by palmitate oxidation, possibly by increasing the activity of SIRT-1 , a NAD+-dependent deacetylase whose activation leads to increased mitochondrial biogenesis and metabolism . While we found no changes in the expression levels of genes known to regulate fatty acid metabolism in skeletal muscle tissue (Figure 3), we did see an increase in SCD-1, ACC-1 (P = 0.07), FAS, and CPT-1 mRNA expression in the WAT and increased PPARα and CPT-1 mRNA levels in the BAT, overall suggesting a specific adipose tissue response to leucine supplementation, with an increased fatty acid flux characterized by increased expression of molecular markers of both fatty acid synthesis (ACC-1, FAS, and SCD-1) and oxidation (PPARα and CPT-1). This phenomenon could possibly also help explain why leucine-supplemented mice did not show a change in the amount of fat mass, even tough they used more fat as energy substrate in vivo. To understand the underlying mechanism of our findings, further studies looking at in vivo fatty acid oxidation versus synthesis would be necessary.
Notably, the abovementioned molecular changes were also accompanied by changes in the expression levels of markers of mitochondrial function in the adipose tissue. In particular, leucine-supplementation led to an increase in Cox-III and UCP-3 mRNA expression in the BAT and to an increase in Cox-III, Cox-IV (P = 0.08), UCP-3 and UCP-1 in the WAT. The almost fourfold increase in the mRNA expression of UCP-1, a classical brown adipocyte marker used often as an indicator of thermogenesis , was particularly striking. Although we did not find changes in energy expenditure in vivo, further studies should investigate the impact of leucine on adipocyte differentiation and/or transdifferentiation in vitro and determine its ability to eventually favor changes in white adipocytes phenotype.
Leucine is known to have beneficial effects on glucose tolerance, lipid metabolism, and insulin sensitivity when chronically administered to HFD fed mice during weight gain [8, 9, 11, 36]. In our study, previously obese, glucose intolerant mice chronically fed with chow and supplemented with leucine during body weight loss and maintenance showed increased glucose tolerance and lower fasting glucose levels. Although the administration of the dose of glucose for the GTT was based on the weight of the mouse, which might have further increased differences between animals' GTT response before and after their weight loss, we did not find any improvement in the ITT, the fasting insulin levels or the calculated HOMA during leucine supplementation. A failure to find distinct results in the ITT does not necessarily mean that differences in insulin sensitivity might not contribute to the changes observed in glucose tolerance. In addition, leucine has been long known to have a stimulating effect on insulin secretion [37, 38]. We therefore wondered whether insulin turnover could be increased due to specific leucine action on the islets of Langerhans, thus leading to an increased glucose tolerance. In vitro experiments with isolated islets showed that this, however, was not the case. Though, this negative result does not exclude a direct action of leucine on the pancreatic β-cell in vivo, while the amino acid is being supplemented through water. Future studies should therefore be carried out to further investigate these aspects.
We also report here that chronic leucine supplementation improved leptin sensitivity in treated mice, because acute administration of the hormone significantly reduced food intake only in the chow-fed group chronically receiving the amino acid. Leptin activates the AMP-activated protein kinase in muscle, thereby stimulating fatty acids oxidation and glucose uptake . This evidence could indicate that increased leptin sensitivity at tissue level might participate to the metabolic and molecular changes observed in the leucine-supplemented mice.
Finally, it should be mentioned that leucine is known to affect energy balance by acting onto the mTOR pathway and other intracellular cascades in both the hypothalamus and the brainstem [6, 7, 40]. Thus, at present, we cannot exclude that chronic leucine supplementation might affect central nervous system circuits eventually leading to changes that might have in turn determined the increased fatty acid oxidation and molecular modifications in peripheral tissues observed in this study.
Overall, the present findings suggest that chronic leucine supplementation might represent an adjuvant beneficial nutritional therapy during body weight loss and maintenance, because its administration improves lipid and glucose metabolism and helps restore leptin sensitivity in previously obese animals. Taking into account that molecular and metabolic changes induced by leucine supplementation were consistent with increased metabolic capacity, it would be interesting to assess whether such changes might be protective against re-exposure to a HFD. Lastly, future clinical studies should be encouraged in order to verify the efficacy and the metabolic benefit of such nutritional approach in obese patients undergoing weight loss and weight maintenance.
We thank G. Drutel, A. Brochard, and the personnel of the q-PCR Platform of the NeuroCentre Magendie, N. Aubailly and the technical staff of the housing and experimental animal facility of the NeuroCentre Magendie.