The Effect of Season on the Classroom Microbiome of Elementary Schools in the Northeastern United States

초록

Introduction: It is established that allergens are an important trigger of asthma exacerbations, but microbial exposures also play an important role. Next-generation sequencing has enabled the ability to reconcile conflicting studies on environmental endotoxin exposure and asthma by characterizing microbial community composition using sequencing. Seasonal influences on the school microbiome are not well understood. Methods: As a part of the School Inner-City Asthma Intervention Study (SICAS-2, NCT02291302) we collected vacuumed dust samples from schools of asthmatic children. 669 school settled dust samples from 251 locations (including classrooms, gyms, and cafeterias) of 41 schools underwent successful metagenomics sequencing on the Illumina NovaSeq 6000 platform. At the time of nucleic acid extraction, we added spike-in halophiles (Zymo Research, Cat#D6320) in order to estimate absolute microbial abundance. Total species richness and inverse Simpson index were calculated by `microbiome` R package. Season was modeled as a 4-level categorical variable using meteorological season in a mixed effects model using the lme4::lmer function (microbial feature ∼ season + (1|school/class)). Beta-diversity was calculated with the Morisita-Horn (M-H) index. Statistical significance of beta diversity was tested in two ways. Nested permutational analysis of variance (PERMANOVA) with the vegan::adonis2 function was used to estimate the effect of season on microbial community structure. The M-H index between microbiome samples obtained from classrooms, with the fall sample as a reference point, was used to assess the global shift in the microbiome. Results: The average sequencing depth was 4.6 million microbial reads. Univariate analyses on alpha diversity indices suggested a trend towards decreased species richness during winter (p=0.097). A significant decrease in total microbial load (cells/mgsample) was observed in winter (p=0.027). PERMANOVA on beta diversity showed that ...