The Associations between Diet and Socioeconomic Disparities and the Intestinal Microbiome in Preadolescence

Abstract

The intestinal microbiome continues to shift and develop throughout youth and could play a pivotal role in health and wellbeing throughout adulthood. Environmental and interpersonal determinants are strong mediators of the intestinal microbiome during the rapid growth period of preadolescence. We aim to delineate associations between the gut microbiome composition, body mass index (BMI), dietary intake and socioeconomic status (SES) in a cohort of ethnically homogenous preadolescents. This cohort included 139 Arab children aged 10-12 years, from varying socioeconomic strata. Dietary intake was assessed using the 24-h recall method. The intestinal microbiome was analyzed using 16S rRNA gene amplicon sequencing. Microbial composition was associated with SES, showing an overrepresentation of Prevotella and Eubacterium in children with lower SES. Higher BMI was associated with lower microbial diversity and altered taxonomic composition, including higher levels of Collinsella, especially among participants from lower SES. Intake of polyunsaturated fatty acids was the strongest predictor of bacterial alterations, including an independent association with Lachnobacterium and Lactobacillus. This study demonstrates that the intestinal microbiome in preadolescents is associated with socioeconomic determinants, BMI and dietary intake, specifically with higher consumption of polyunsaturated fatty acids. Thus, tailored interventions during these crucial years have the potential to improve health disparities throughout the lifespan.

Keywords: dietary intake; microbiome; obesity; school age; socioeconomic status.

Figure 1

Socio-demographic characteristics and dietary intake (A) Correlation matrix of Spearman’s correlation coefficients across the study variables. Positive correlations are colored in green and inverse correlations are colored in orange. (B) A scatter plot of the correlation between BMIZ score and SES score. (C) A scatter plot of the correlation between intake of PUFAs (%Kcal) and SES score. X in the figure represents non-significant correlation (p > 0.05 adjusted for FDR).

Figure 2

Bacterial diversity and composition. (A) A scatter plot of the correlation between alpha diversity, evaluated by Simpson’s evenness index, increased and intake of protein (%Kcal/day). (B) Box plot of alpha diversity, evaluated by Simpson’s evenness index according to tertiles consumption of PUFAs (%Kcal/day) (T1 = lowest tertile, T2 =middle tertile, T3 = upper tertile). (C) Principal component analysis (PcoA) of the Unweighted UniFrac distance showing a significant separation based on the SES score and intake of PUFAs. (D) A heatmap of the multivariable model describing the top 50 associations between the independent variables and bacterial features. Positive associations are colored in red, while inverse associations are colored in blue. The color gradient represents the strength of the association (the effect size), with darker colors representing the stronger associations. The effect size was calculated by the following formula: (−log(qval)*SIGN (coeff)).

Figure 3

Low socioeconomic status and the intestinal microbiome. Box plots of alpha diversity according to BMIZ score categories. (A) Alpha diversity was measured by the Shannon’s diversity index, and (B) richness. (C) Principal component analysis (PcoA) of the weighted UniFrac showing a significant separation of the BMIZ score categories and (D) intake of protein measured by the unweighted UniFrac distance. (E) A heatmap of the multivariable model describing the top associations between the independent variables and bacterial features. Positive associations are colored in blue, while inverse associations are colored in yellow. The color gradient represents the strength of the association, with darker colors representing stronger associations. The effect size was calculated by the following formula: (−log(qval)*SIGN (coeff)).

Figure 4

High socioeconomic status and the intestinal microbiome. Box plots of alpha diversity, measured by Simpson’s evenness, according to tertiles intake of (A) dietary protein and (B) PUFAs (T1 = lowest tertile, T2 = middle tertile, T3 = upper tertile). (C) Principal component analysis (PcoA) of Unweighted UniFrac showing the intake of PUFAs among participants from high SES. (D) A heatmap of the multivariable model describing the top associations between the independent variables and bacterial features. Positive associations are colored in purple, while inverse associations are colored in green. The color gradient represents the strength of the association observed, with darker colors representing stronger associations. The effect size was calculated by the following formula: (−log(qval)*SIGN (coeff)).