Gut microbiome is associated with asthma and race in children with food allergy

Abstract

Background: The composition of the gut microbiome has been associated with development of atopic conditions such as food allergy (FA) and asthma. African American or Black children with FA have higher rate of asthma compared to their White counterparts.

Objective: We sought to investigate whether the diversity and relative abundance (RA) of gut microbiota is different between children with FA from different racial backgrounds living in the same cities. Furthermore, we aimed to understand whether the difference in the gut microbiota is associated with asthma in children with FA.

Methods: We analyzed and compared the stool microbiome of a cohort of Black and White children with FA by shotgun genomic sequencing.

Results: A total of 152 children with IgE-mediated FA enrolled onto FORWARD (Food Allergy Outcomes Related to White and African American Racial Differences); 30 Black and 122 White children were included. The RA of several bacteria was associated with race and asthma. Most notably the RA of Bacteroides thetaiotaomicron, Chlamydia thrachomatis, Parabacteroides goldsteinii, and Bacteroides eggerthii were significantly higher, while the RA of Bifidobacterium sp CAG:754, Parabacterium johnsonii, Bacteroides intestinalis, and Bifidobacterium breve were significantly lower in stool samples of Black children compared to White children. Asthma was associated with lower RA of B breve, Bifidobacterium catenulatum, Prevotella copri, Veilloella sp CAG:933, and Bacteroides plebius, and higher RA of 3 Bacteroides species.

Conclusions: The observed variations in the gut microbiota of Black and White children such as differences in the Bacteroides and Bifidobacterium species along with their association to history of asthma in our cohort is indicative of their potential role in the higher rate of asthma observed among Black children with FA.

Keywords: Black; Microbiome; asthma; food allergy; race; relative abundance (RA).

Figure 1.

Alpha diversity indices among children with FA. Chao 1 (A) and Shannon (B) diversity indices stratified by race. Diversity among races was significantly different for the Shannon index (Wilcoxon rank-sum test, p = 0.02). Spearman correlation was used to assess the association between alpha diversity and age (C-D). An increase in Shannon index was significantly correlated with age (R = 0.34, p < 0.001). No significant differences were observed in the Chao 1 index.

Figure 2.

Beta diversity of the microbiome in children with FA by race and age. (A) Microbial composition at the family level of taxa with relative abundance >5% shown and compared in two races (left side) and in different age groups (right side). (B) Principal coordinate analysis (PCoA) was calculated using Bray-Curtis distances. Ellipses assume a multivariate t-distribution with a 95% CI.

Figure 3.

Differentially abundant (DA) taxa at species level in children with food allergy. Regression model of race controlling for age. The values shown are DA taxa for Black subjects compared to White. Differential abundance was determined by beta binomial regression with a false discovery rate cutoff of < 0.05. The center dots represent the calculated regression coefficients. The left and right error bars represent the lower and upper 95% confidence intervals respectively. Red bars indicate significant positive associations and blue bars represent significant negative associations. Subjects included in the model were between 3 and 12 years of age.

Figure 4.

Differentially abundant (DA) taxa at species level in children with food allergy. Regression models of atopic comorbidities controlling for race and age. The values shown are DA taxa for subjects with either asthma or eczema compared to those without. Differential abundance determined by beta binomial regression with a false discovery rate cutoff of < 0.05. The center dots represent the calculated regression coefficients. The left and right error bars represent the lower and upper 95% confidence intervals respectively. Red bars indicate significant positive associations and blue bars represent significant negative associations. Subjects included in the model were between 3 and 12 years of age.