Acute effects of a single bout of high-intensity strength and endurance exercise on cognitive biomarkers in young adults and elderly men: a within-subjects crossover study

Abstract

Background: Although evidence for exercise-induced changes in neurocognitive biomarkers is emerging, research examining acute responses to different exercise regimes across sex and age is lacking. This study investigated serum concentrations of three neurocognitive biomarkers (i.e., Klotho, brain-derived neurotrophic factor (BDNF), and glycosylphosphatidylinositol-specific phospholipase D1 (GPLD1)) after acute strength and aerobic exercise, along with skeletal muscle gene expression.

Methods: In a within-subjects crossover design, blood samples of 19 young women, 20 young men, and 14 elderly men were taken before, immediately, 3 h and 24 h after one bout of strength training (ST) and high-intensity interval training (HIIT). Muscle biopsies were taken from a subgroup (n = 22) before, 3 h and 24 h after ST and HIIT for gene expression analyses. Time changes and baseline levels, including the influence of sex and age, were analyzed using a multilevel model and Welch's analysis of variance, respectively. Biomarker levels were adjusted for exercise-induced plasma volume changes.

Results: Serum concentration of all biomarkers increased after ST and HIIT but were not affected by sex or age. While serum Klotho and BDNF levels peaked immediately after exercise in all groups, serum GPLD1 levels were highest at 3 h (young groups only). Age was a determining factor for baseline measures; young men had higher and lower resting serum Klotho and BDNF concentration, respectively, than elderly men. Muscle gene expression of Klotho increased after both exercise modes, and BDNF and GPLD1 expression was reduced within 24 h.

Conclusions: Circulating levels of biomarkers linked to brain health can acutely be increased by one bout of ST or HIIT. This increase might be related to altered gene expression of these proteins in skeletal muscle. Ultimately, this could have beneficial implications for the management of mental and neurocognitive impairments.

Keywords: BDNF; Cognition; Exercise; GPLD1; Gene expression; HIIT (high-intensity interval training); Klotho; Serum; Skeletal muscle; Strength training.

Fig. 1

Flow chart. ^a 1 young woman withdrew, but participated in HIIT. ^b 2 young women, 1 young man, and 1 elderly man withdrew, but participated in ST. ST = strength training, HIIT = high-intensity interval training. Muscle biopsies were taken from a subgroup of volunteers at baseline (1st visit only), 3 h and 24 h after ST and HIIT (n = 22, 6 young women, 9 young men, 7 elderly men)

Fig. 2

Structure of levels used in the analysis. Observations t₁ = baseline (pre), t₂ = immediately after exercise (post), t₃ = 3 h after exercise, t₄ = 24 h after exercise. Modified from Monsalves et al. [54]; http://creativecommons.org/licenses/by/4.0/

Fig. 3

Baseline serum levels of α-Klotho (A), BDNF (B) and GPLD1 (C). BDNF brain-derived neurotrophic factor, GPLD1 glycosylphosphatidylinositol specific phospholipase D1. Data are presented as means with error bars representing standard deviation. Welch’s ANOVA with Dunnett’s T3 post hoc test to adjust for multiple testing was used for detecting possible differences in sex (young women vs. young men) and age (young men vs. elderly men). *p < 0.01

Fig. 4

Serum levels of α-Klotho at baseline, immediately, 3 h, and 24 h after exercise. Data are presented as estimated means with error bars representing 95% confidence intervals based on a random intercept model with time, group, and time x group as fixed factors. Young women n = 17 (ST), n = 16 (HIIT); young men n = 20 (ST), n = 19 (HIIT); elderly men n = 14 (ST), n = 13 (HIIT). ST = strength training, HIIT = high-intensity interval training. Pairwise comparisons using Šidák adjustment for multiple testing resulted in the following significant differences (see Table 2 for exact p-values and effect sizes): main effect of time: ** p < 0.001 vs. baseline within all 3 groups; * p < 0.01 vs. baseline within all 3 groups; # p < 0.05 vs. baseline within young women and elderly men; ## p < 0.001 vs. baseline within young men and young women

Fig. 5

Serum levels of BDNF at baseline, immediately, 3 h, and 24 h after exercise. BDNF brain-derived neurotrophic factor. Data are presented as estimated means with error bars representing 95% confidence intervals based on a random intercept model with time, group, and time x group as fixed factors. Young women n = 18 (ST), n = 17 (HIIT); young men n = 20 (ST), n = 19 (HIIT); elderly men n = 14 (ST), n = 13 (HIIT). ST = strength training, HIIT = high-intensity interval training. Pairwise comparisons using Šidák adjustment for multiple testing resulted in the following significant differences (see Table 2 for exact p-values and effect sizes): main effect of time: * p < 0.01 vs. baseline within all 3 groups; ** p < 0.001 vs. baseline within all 3 groups; # p < 0.05 vs. baseline within elderly men; main effect of group: $p < 0.01 young men vs. elderly men; $$ p < 0.001 young men vs. elderly men

Fig. 6

Serum levels of GPLD1 at baseline, immediately, 3 h, and 24 h after exercise. GPLD1 glycosylphosphatidylinositol specific phospholipase D1. Data are presented as geometric means with error bars representing non-symmetric 95% confidence intervals due to data transformation (natural logarithm). Data are based on a random intercept model with time, group, and time x group as fixed factors. Young women n = 17 (ST), n = 16 (HIIT); young men n = 20 (ST), n = 19 (HIIT); elderly men n = 14 (ST), n = 12 (HIIT). ST strength training, HIIT high-intensity interval training. Pairwise comparisons using Šidák adjustment for multiple testing resulted in the following significant differences (see Table 2 for exact p-values and effect sizes): main effect of time: * p < 0.05 within young women only; # p < 0.01 within young men only; main effect of group:$ p < 0.05 young men vs. elderly men

Fig. 7

Relative mRNA expression of α-Klotho (A), BDNF (B) and GPLD1 (C) at baseline (pre), 3 h, and 24 h after exercise. mRNA messenger RNA, BDNF brain-derived neurotrophic factor, GPLD1 glycosylphosphatidylinositol specific phospholipase D1, ST strength training, HIIT high-intensity interval training. Data are presented as estimated means and 95% confidence intervals (CIs) based on a random intercept model with time as fixed factor (geometric means and non-symmetric 95% CIs for BDNF due to data transformation (natural logarithm)). Klotho n = 21 (ST), n = 22 (HIIT), BDNF n = 21 (ST), n = 22 (HIIT), GPLD1 n = 20 (ST), n = 21 (HIIT). Pairwise comparisons were adjusted for multiple testing (Šidák adjustment). *p < 0.05, **p < 0.001