Amino acid changes during transition to a vegan diet supplemented with fish in healthy humans

Abstract

Purpose: To explore whether changes in dietary protein sources can lower plasma branched-chain amino acids (BCAAs), aromatic amino acids and sulfur amino acids (SAAs) that are often elevated in the obese, insulin-resistant state and in type 2 diabetes.

Methods: Thirty-six subjects (mean age 31 ± 2 years) underwent a voluntary abstinence from meat, poultry, eggs, and dairy products for 6 weeks, while enriching the diet with fish, in fulfillment of a religious fast. Subjects were assessed 1 week before the fast (V1), 1 week after initiation of the fast (V2) and in the last week of the fast (V3). Thirty-four subjects completed all three visits.

Results: Fasting plasma BCAAs decreased at V2 and remained low at V3 (P < 0.001 for all). Valine showed the greatest decline, by 20 and 19 % at V2 and V3, respectively. Phenylalanine and tryptophan, but not tyrosine, also decreased at V2 and V3. The two proteinogenic SAAs, methionine and cysteine, remained stable, but the cysteine product, taurine, decreased from 92 ± 7 μmol/L to 66 ± 6 (V2; P = 0.003) and 65 ± 6 μmol/L (V3; P = 0.003). A progressive decline in plasma glutamic acid, coupled with an increase in glutamine, was observed. Plasma total and LDL cholesterol decreased at V2 and V3 (P < 0.001 for all).

Conclusion: Changing dietary protein sources to plant- and fish-based sources in an ad libitum setting lowers the plasma BCAAs that have been linked to diabetes risk. These findings point to habitual diet as a potentially modifiable determinant of fasting plasma BCAA concentrations.

Keywords: Body mass index; Branched-chain amino acids; Egyptian Orthodox Christians; Lean mass; Mass spectrometry; Sulfur amino acids.

Fig. 1

Frequency of consumption of different food groups, presented as number of servings per week (a–c), or per 2 days (d, e). Data represent median, 25th–75th percentiles, with individual data plotted. For all food groups, intake at V2 and V3 differed significantly from V1 (P < 0.001). Where there is overlap between the percentile lines and most or all data points (e.g., 1-A and 1-B at V2 and V3), individual data dots are obscured by the percentile lines. V1 baseline, V2 first week after diet change, V3 last week of diet change

Fig. 2

Fasting plasma branched-chain (a–c) and aromatic (d–f) amino acids at V1 (baseline), V2 (first week after diet change), and V3 (last week of diet change). Data represent mean (SEM). *P < 0.01; **P < 0.001 for pairwise comparisons versus V1, with repeated measures ANOVA P < 0.05

Fig. 3

Fasting plasma sulfur amino acids and related metabolites at V1 (baseline), V2 (first week after diet change), and V3 (last week of diet change). Data represent mean (SEM). *P < 0.05; **P ≤ 0.004 for pairwise comparisons versus V1, with repeated measures ANOVA P < 0.05

Fig. 4

Changes in glutamine and glutamic acid. Data represent mean. **P < 0.001 for pairwise comparisons versus V1, with repeated measures ANOVA P < 0.05