UCSF

Objective

Exercise is one of the most potent strategies available to support cognitive health with age, yet substantial variability exists. Sexual dimorphism is evident for brain and immune functioning, the latter being implicated as important pathway for exercise. We examined the moderating role of sex on the relationship between physical activity and systemic inflammatory and brain health outcomes in support of more personalized approaches to behavioral interventions.

Methods

Our discovery cohort included 45 typically aging women matched on age (±5y) and education (±2y) to 45 men (mean age = 72.5; Clinical Dementia Rating = 0) who completed self-reported current physical activity (Physical Activity Scale for Elderly), blood draw, neuropsychological evaluation, and brain MRI. An independent sample of 45 typically aging women and 36 men who completed the same measures comprised a replication cohort. Plasma was analyzed for 11 proinflammatory cytokine and chemokine markers via MesoScale Discovery.

Results

Discovery cohort: Reported physical activity did not differ between sexes (150 vs. 157, p = 0.72). There was a significant interaction between sex and physical activity on chemokine markers MDC, MIP-1b, MCP-4, and eotaxin-3 (ps < 0.03), with a similar trend for MCP-1 and INFγ (ps < 0.09). Men who reported greater activity demonstrated lower inflammatory markers, an effect attenuated-to-absent in women. An interaction between sex and physical activity was also observed for parahippocampal volumes (p = 0.02) and cognition (processing speed and visual memory; ps < 0.04). Again, the beneficial effect of physical activity on outcomes was present in men, but not women. Replication cohort analyses conferred a consistent effect of sex on the relationship between physical activity and immune markers; models examining neurobehavioral outcomes did not strongly replicate. Across cohorts, post-hoc models demonstrated an interaction between sex and activity-related inflammatory markers on total gray matter volume and visual memory. Men with higher inflammatory markers demonstrated poorer brain structure and function, whereas inflammatory markers did not strongly relate to neurobehavioral outcomes in women.

Conclusions

Greater physical activity was associated with lower markers of inflammation in clinically normal older men, but not women - an effect consistently replicated across cohorts. Additionally, men appeared disproportionately vulnerable to the adverse effects of peripheral inflammatory markers on brain structure and function compared to women. Immune activation may be a male-specific pathway through which exercise confers neurobehavioral benefit.