Metabolomics analysis reveals altered metabolites in lean compared with obese adolescents and additional metabolic shifts associated with hyperinsulinaemia and insulin resistance in obese adolescents: a cross-sectional study

Abstract

Introduction: Hyperinsulinaemia and insulin resistance (IR) are strongly associated with obesity and are forerunners of type 2 diabetes. Little is known about metabolic alterations separately associated with obesity, hyperinsulinaemia/IR and impaired glucose tolerance (IGT) in adolescents.

Objectives: To identify metabolic alterations associated with obesity, hyperinsulinaemia/IR and hyperinsulinaemia/IR combined with IGT in obese adolescents.

Methods: 81 adolescents were stratified into four groups based on body mass index (lean vs. obese), insulin responses (normal insulin (NI) vs. high insulin (HI)) and glucose responses (normal glucose tolerance (NGT) vs. IGT) after an oral glucose tolerance test (OGTT). The groups comprised: (1) healthy lean with NI and NGT, (2) obese with NI and NGT, (3) obese with HI and NGT, and (4) obese with HI and IGT. Targeted nuclear magnetic resonance-based metabolomics analysis was performed on fasting and seven post-OGTT plasma samples, followed by univariate and multivariate statistical analyses.

Results: Two groups of metabolites were identified: (1) Metabolites associated with insulin response level: adolescents with HI (groups 3-4) had higher concentrations of branched-chain amino acids and tyrosine, and lower concentrations of serine, glycine, myo-inositol and dimethylsulfone, than adolescents with NI (groups 1-2). (2) Metabolites associated with obesity status: obese adolescents (groups 2-4) had higher concentrations of acetylcarnitine, alanine, pyruvate and glutamate, and lower concentrations of acetate, than lean adolescents (group 1).

Conclusions: Obesity is associated with shifts in fat and energy metabolism. Hyperinsulinaemia/IR in obese adolescents is also associated with increased branched-chain and aromatic amino acids.

Keywords: Energy metabolism; Hyperinsulinaemia; Insulin resistance; NMR metabolomics; Obesity.

Fig. 1

Plasma concentrations during an oral glucose tolerance test (OGTT; left panel) and area under the curve (AUC; right panel) of a insulin and b glucose in lean adolescents with normal insulin (NI; black, n = 21), adolescents with obesity and NI (green, n = 18), adolescents with obesity and high insulin (HI; blue, n = 20), and adolescents with obesity and HI in combination with impaired glucose tolerance (IGT; red, n = 23) (Color figure online)

Fig. 2

Plasma levels during an oral glucose tolerance test (OGTT; left panel) and area under the curve (AUC; right panel) of a valine, b leucine, c isoleucine and d tyrosine in lean adolescents with normal insulin (NI; black, n = 21), adolescents with obesity and NI (green, n = 18), adolescents with obesity and high insulin (HI; blue, n = 20), and adolescents with obesity and HI in combination with impaired glucose tolerance (IGT; red, n = 23). P-values for group effect (n = 4, mixed model) and corresponding Benjamini-Hochberg-adjusted P-values (Q-values) are presented. All metabolites also showed a significant insulin response effect (NI vs. HI) (P < 0.05) (Color figure online)

Fig. 3

Plasma levels during an oral glucose tolerance test (OGTT; left panel) and area under the curve (AUC; right panel) of a serine, b glycine, c myo-inositol and d dimethylsulfone in lean adolescents with normal insulin (NI; black, n = 21), adolescents with obesity and NI (green, n = 18), adolescents with obesity and high insulin (HI; blue, n = 20), and adolescents with obesity and HI in combination with impaired glucose tolerance (IGT; red, n = 23). P-values for group effect (n = 4, mixed model), time × group interaction and corresponding Benjamini-Hochberg-adjusted P-values (Q-values) are presented for metabolites identified as discriminative via the univariate statistical approach (serine, dimethylsulfone and myo-inositol). Serine and dimethylsulfone also showed a significant insulin response effect (NI vs. HI) (P < 0.05). Glycine was identified as discriminative via the multivariate statistical approach and significant models were obtained at 30, 60, 90 and 120 min (Color figure online)

Fig. 4

Plasma levels during an oral glucose tolerance test (OGTT; left panel) and area under the curve (AUC; right panel) of a o-acetylcholine, b glutamate, c alanine and d pyruvate in lean adolescents with normal insulin (NI; black, n = 21), adolescents with obesity and NI (green, n = 18), adolescents with obesity and high insulin (HI; blue, n = 20), and adolescents with obesity and HI in combination with impaired glucose tolerance (IGT; red, n = 23). P-values for group (n = 4, mixed model) and corresponding Benjamini-Hochberg-adjusted P-values (Q-values) are presented for metabolites identified as discriminative via the univariate statistical approach (o-acetylcarnitine, glutamate and alanine). O-acetylcarnitine, glutamate and alanine also showed a significant obesity effect (lean vs. obese) (P < 0.05). Pyruvate was identified as discriminative via the multivariate statistical approach and a significant model was obtained at 60 min. Pyruvate also showed a close to significant time × obesity interaction (P = 0.0551) (Color figure online)

Fig. 5

Plasma levels during an oral glucose tolerance test (OGTT; left panel) and area under the curve (AUC; right panel) of acetone in lean adolescents with normal insulin (NI; black, n = 21), adolescents with obesity and NI (green, n = 18), adolescents with obesity and high insulin (HI; blue, n = 20), and adolescents with obesity and HI in combination with impaired glucose tolerance (IGT; red, n = 22). P-values for group (n = 4, mixed model) and corresponding Benjamini-Hochberg-adjusted P-values (Q-values) are presented. Acetate also showed a significant obesity effect (lean vs. obese) (P < 0.05) (Color figure online)