Working memory and hippocampal expression of BDNF, ARC, and P-STAT3 in rats: effects of diet and exercise

Mounting evidence suggests diet and exercise influence learning and memory (LM). We compared a high-fat, high-sucrose Western diet (WD) to a plant-based, amylose/amylopectin blend, lower-fat diet known as the Daniel Fast (DF) in rats with and without regular aerobic exercise on a task of spatial working memory (WM).

Rats were randomly assigned to the WD or DF at 6 weeks of age. Exercised rats (WD-E, DF-E) ran on a treadmill 3 times/week for 30 min while the sedentary rats did not (WD-S, DF-S). Rats adhered to these assignments for 12 weeks, inclusive of ab libitum food intake, after which mild food restriction was implemented to encourage responding during WM testing. For nine months, WM performance was assessed once daily, six days per week, after which hippocampal sections were collected for subsequent analysis of brain-derived neurotrophic factor (BDNF), activity-regulated cytoskeletal protein (ARC), and signal transducer and activator of transcription 3 (P-STAT3, Tyr705).

DF-E rats exhibited the best DSA performance. Surprisingly, the WD-S group outperformed the WD-E group, but had significantly lower BDNF and ARC relative to the DF-S group, with a similar trend from the WD-E group. P-STAT3 expression was also significantly elevated in the WD-S group compared to both the DF-S and WD-E groups.

These results support previous research demonstrating negative effects of the WD on spatial LM, demonstrate the plant-based DF regimen combined with chronic aerobic exercise produces measurable WM and neuroprotective benefits, and suggest the need to carefully design exercise prescriptions to avoid over-stressing individuals making concurrent dietary changes.