Refining gut microbial signatures of depressive symptoms through meta-analysis across population studies: effects of sex and age groups

Background

Gut microbiota involvement in mental health issues, including depressive symptoms, has becoming increasingly apparent as demonstrated by increasing number of recent high-quality (pre-)clinical evidence in the gut microbiome-brain axis field. However, the underlying functional mechanisms remain elusive, in part due to methodological limitations attributed to the complexity of microbial ecology and phenotypic diversity. Looking into the often-overlooked biological moderators, namely sex and age, may help uncovering the underlying biology given evidence of gut microbial involvement in inflammation pathways and hormone metabolism.

Objectives

(1) Identify gut microbial features robustly linked to depressive symptom across diverse population, sex, and age groups; (2) Estimate the effects of demographic factors in driving associations identified in the first objective, with a focus on sex differences

Methods

In this on-going large-scale meta-analysis study, we are conducting a harmonized meta-analysis of eight large-scale population-based cohorts (N = 23,344) including from Europe and the USA. Depressive symptom as an outcome is quantified using standardized and validated instruments (CES-D or PHQ-9). Microbial taxonomic profiles are estimated from 16S rRNA and/or shotgun metagenomic reads by mapping sequences to the Greengenes2 database (v2024.09) in a harmonized manner. Associations within cohort are estimated using linear regression with stepwise adjustments for sociodemographic, lifestyle factors, and medication use. Stratified analysis by sex and age group is conducted. Cohort-level summary statistics of microbial features that are present in at least 80% of the included population cohort studies are synthesized using weighted meta-analyses. All p-values are FDR corrected.

Results

In a subset of our European cohorts (N = 1,324; mean age = 62.67; 57.85% female; mean CES-D = 4.96), we discovered that gut microbial diversity was associated with higher depressive symptom scores in older population (β = 0.50, p = 0.01). We also identified 36 and 17 out of 240 bacterial genera that were associated with higher and lower depressive symptom scores, respectively. In sex-stratified analysis with per group z-score scaling, we observed that the associations with gut microbial diversity, alongside some genera, for instance Enterococcus_B, Streptococcus, and Blautia_A_141781, were amplified in female but diminished in male. Interestingly, a small subset of genera showed stronger effects in male compared to female, such as Lactobacillus. Age-stratified analysis with four age categories (<= 60, N = 455; 61 to 67, N = 641; 68 to 75, N = 181; > 75, N = 47) demonstrated that some effects, for instance, Butyrococcus_A_77030, were specific only to certain age groups. Our current analysis unveiled the presence of sex and age difference in gut microbiome relationships with depressive symptoms which might be meaningful for building functional interpretation and hypothesis. While the preliminary subset has narrow age and cohort variance, expanded analyses are currently underway to quantify the effects of age and population group more robustly.

References

No references