Restriction of essential amino acids dictates the systemic metabolic response to dietary protein dilution

Abstract

Dietary protein dilution (DPD) promotes metabolic-remodelling and -health but the precise nutritional components driving this response remain elusive. Here, by mimicking amino acid (AA) supply from a casein-based diet, we demonstrate that restriction of dietary essential AA (EAA), but not non-EAA, drives the systemic metabolic response to total AA deprivation; independent from dietary carbohydrate supply. Furthermore, systemic deprivation of threonine and tryptophan, independent of total AA supply, are both adequate and necessary to confer the systemic metabolic response to both diet, and genetic AA-transport loss, driven AA restriction. Dietary threonine restriction (DTR) retards the development of obesity-associated metabolic dysfunction. Liver-derived fibroblast growth factor 21 is required for the metabolic remodelling with DTR. Strikingly, hepatocyte-selective establishment of threonine biosynthetic capacity reverses the systemic metabolic response to DTR. Taken together, our studies of mice demonstrate that the restriction of EAA are sufficient and necessary to confer the systemic metabolic effects of DPD.

Fig. 1. Dietary amino acids are required for the systemic metabolic effects of dietary protein dilution.

a Serum urea levels of mice in response to a 3-week treatment with diets containing 20% energy from protein (20P), 5% energy from protein (5P), and 5% energy from protein and 15% energy from amino acids to match that of 20P. Data are mean and SEM; n = 5 individual mice per group. Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than 20P: *P < 0.05, **P < 0.01, ***P < 0.001. Different than 5P: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. b Feed efficiency of mice as in (a). c Energy expenditure over the different day phases of mice as in (a). d The rate of O₂ consumption (VO₂) over the different day phases of mice as in (a). e The rate of CO₂ production (VCO₂) over the different day phases of mice as in (a). f Scatter plot of energy expenditure (EE) versus body weight of mice as in (a). g Physical activity as assessed by laser beam breaks across three physical dimensions (sumXYZ) over the different day phases of mice as in (a). h Serum fibroblast growth factor 21 (FGF21) levels of mice as in (a). i Insulin sensitivity index during fasting (ISI(f)) of mice as in (a).

Fig. 2. Dietary essential amino acid restriction, independent from non-essential amino acid or carbohydrate supply, dictates the systemic metabolic response to dietary protein dilution.

a Hepatic portal vein amino acid levels in response to refeeding a low-protein diet following a week of diet adaptation. Data (mean and SEM) are represented as a proportion of the control group fed a normal protein diet (20%P). n = 5/group derived from samples of individual mice. Data were analysed by Student’s t-test. Different than 20%P, *P < 0.05. b Liver tissue amino acid levels in response to refeeding a low-protein diet following a week of diet adaptation. Data are represented as a proportion of the control group fed a normal protein diet (20%P). n = 5/group derived from samples of individual mice. Data were analysed by Student’s t-test. Different than 20%P, *P < 0.05. c Nutrient source breakdown as % contribution to total energy for the experimental groups. EAA: essential amino acids. NEAA: non-essential amino acids. d Urinary urea output of mice in response to a 3-week treatment with diets as per the protocol of SF1A containing nutrient energy sources as in (c). Data are mean and SEM; n = 5 individual mice per group. Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than 20P: *P < 0.05, **P < 0.01, ***P < 0.001. Different than 5P: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. e Feed efficiency of mice as in (d). f Energy expenditure of mice as in (d). g Serum fibroblast growth factor 21 (FGF21) levels of mice as in (d). h Blood glucose levels during an intraperitoneal glucose tolerance test (ipGTT) of mice as in (d). i Plasma insulin levels during an intraperitoneal glucose tolerance test (ipGTT) of mice as in (d). j Insulin sensitivity index during fasting (ISI(f)) of mice as in (d).

Fig. 3. Certain essential amino acids including threonine and tryptophan are sufficient and necessary for the systemic metabolic effects of dietary protein dilution.

a Feed efficiency of mice in response to a 3-week treatment with diets containing 18% from amino acids (normal amino acid; NAA), 4.5% essential AA (LEAA; as of diet E in Fig. 2c), and LEAA supplemented with either lysine, threonine, and tryptophan (LEAA + KTW), phenylalanine, histidine, and methionine (LEAA + FHM), or isoleucine, leucine, and valine (LEAA + ILV), all with other AA equally adjusted to give 18% AA in total. Data are mean and SEM (n = 6 individual mice per group). Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than diet NAA: *P < 0.05, **P < 0.01, ***P < 0.001. Different than diet LEAA: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. b Energy expenditure of mice as in (a). c Serum fibroblast growth factor 21 (FGF21) levels of mice as in (a). d Insulin sensitivity index during fasting (ISI(f)) of mice as in (a). e Feed efficiency of mice in response to a 3-week treatment with diets containing 18% from amino acids (AA; NAA), 4.5% essential AA (LEAA; as of diet E in Fig. 2c), and diet singly with restricted amounts of lysine (LK), threonine (LT), and tryptophan (LW), all with other AA equally adjusted to give 18% AA in total. Data are mean and SEM; n = 6 individual mice per group. Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than diet NAA: *P < 0.05, **P < 0.01, ***P < 0.001. Different than diet LEAA: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. f Energy expenditure of mice as in (e). g Serum FGF21 levels of mice as in (e). h ISI(f) of mice as in (e). i Feed efficiency of mice in response to a 3-week treatment with diets containing 18% from amino acids (normal amino acid; NAA), 4.5% AA (LAA; as of diet B in Fig. 2c), and LAA supplemented with threonine and tryptophan while keeping total AA at 4.5% (LAA(TW)). Data are mean and SEM (n = 5 individual mice per group). Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than NAA: *P < 0.05, **P < 0.01, ***P < 0.001. Different than LAA: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. j Energy expenditure of mice as in (e). k Serum FGF21 levels of mice as in (e). l ISI(f) of mice as in (e).

Fig. 4. The systemic metabolic response to dietary AA restriction is conserved in mature male and female mice.

a The change in body, fat, and lean mass of 6-month-old male (shown left) mice in response to an 8-week treatment with diets containing 18% from amino acids (normal amino acid; NAA), 4.5% AA (LAA; as of diet B in Fig. 2c), 4.5% essential AA (LEAA; as of diet E in Fig. 2c), and a diet low in threonine but with matching total AA to NAA (LT). Data are mean and SEM (n = 5 individual mice per group). Data were analysed by one-way ANOVA with Holm–Sidak post-hoc tests. Different than NAA: ^aP < 0.05, ^aaP < 0.01, ^aaaP < 0.001. Different than LAA: ^bP < 0.05, ^bbP < 0.01, ^bbbP < 0.001. Different than LEAA: ^cP < 0.05, ^ccP < 0.01, ^cccP < 0.001. b The change in body, fat, and lean mass of 6-month-old female (shown left) mice treated as in (a). c Tissue weights of mice at the end of the treatment as in (a). d Tissue weights of mice at the end of the treatment as in (b). e Feed efficiency during the 8-week treatment of mice as in (a). f Feed efficiency during the 8-week treatment of mice as in (b). g Energy expenditure of mice as in (a). h Energy expenditure of mice as in (b). i Serum fibroblast growth factor 21 (FGF21) levels of mice as in (a). j Serum FGF21 of mice as in (b). k Insulin sensitivity index during fasting (ISI(f)) of mice as in (a). l ISI(f) of mice as in (b).

Fig. 5. Threonine restriction is a common feature of other models of systemic AA restriction and retards obesity-induced metabolic dysfunction in mice.

a Hepatic portal vein serum amino acid (AA) concentrations in Slc6a19 knockout (−/−) or wildtype (+/+) littermate mice in the refed state on a standard control diet. Data are mean and SEM; n = 5 individual mice per group group. Data were analysed by a Student’s t-test. Different than +/+: *P < 0.05. b Plasma FGF21 levels from Slc6a19 knockout (−/−) or wildtype (+/+) littermate mice in the refed state following intraperitoneal administration of the amino acids threonine and tryptophan (TW), phenylalanine and histidine (FH), or vehicle (VEH: saline solution) on a standard control diet. Data are mean and SEM; n = 5 individual mice per group. Data were analysed by a two-way repeated measures ANOVA. Different than +/+: *P < 0.05, **P < 0.01, ***P < 0.001. Different than VEH: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. c Blood glucose levels during a 4-week treatment of New Zealand Obese mice fed diets containing 18% from amino acids (normal amino acid; NAA) or a diet low in threonine but with matching total AA to NAA (LT). Data are mean and SEM; n = 8 individual mice per group. Data were analysed by one-way repeated measures ANOVA. Different than NAA: *P < 0.05, **P < 0.01, ***P < 0.001. d Serum triglyceride (TG) levels at the end of mice at the end of treatment as in (c). Data were analysed by a Student’s t-test. Different than NAA: *P < 0.05, **P < 0.01, ***P < 0.001. e Serum fibroblast growth factor 21 (FGF21) levels of mice at the end of treatment as in (c).

Fig. 6. Liver-derived fibroblast growth factor 21 is necessary for the systemic metabolic remodelling with dietary threonine restriction.

a Serum fibroblast growth factor 21 (FGF21) levels of Fgf21fl/fl mice at the end of an 8-week treatment with diets containing 18% from amino acids (AA; NAA) or low threonine with other AA equally adjusted to give 18% AA in total (LT); with pre-treatment with adeno-associated viruses to express Cre-recombinase (AAV-CRE) or green fluorescent protein (AAV-GFP) in an hepatocyte-selective manner. Data are mean and SEM (n = 6 NAA × AAV-GFP; n = 7 LAA AAV-GFP; n = 6 LT × AAV-CRE; n = 8 LT × AAV-CRE). Data were analysed by two-way ANOVA with Holm–Sidak post-hoc tests. Different than NAA: *P < 0.05, **P < 0.01, ***P < 0.001. Different than AAV-GFP: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. b Feed efficiency during the 8-week treatment of mice as in (a). c Energy expenditure of mice as in (a). d Insulin sensitivity index during fasting (ISI(f)) of mice as in (a).

Fig. 7. Enforced hepatic threonine biosynthetic capacity reverses the systemic metabolic effects to dietary threonine restriction.

a Feed efficiency of mice in response to a 3-week treatment with diets containing 18% from amino acids (normal amino acid; NAA; yellow bars) and a diet with restricted amounts of threonine (LT; green bars), following prior treatments with adeno-associated viruses to transduce the liver to express yeast threonine biosynthetic enzymes (AAV-yTHR1 + THR4) or a negative control (AAV-GFP). Data are mean and SEM (N = 6 individual mice per group). Data were analysed by two-way ANOVA with Holm–Sidak post-hoc tests. Different than diet NAA: *P < 0.05, **P < 0.01, ***P < 0.001. Different than AAV-GFP: ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. b Energy expenditure of mice as in (a). c Serum fibroblast growth factor 21 (FGF21) levels of mice as in (a). d Insulin sensitivity index during fasting (ISI(f)) of mice as in (a).