Next-generation DNA sequencing reveals that low fungal diversity in house dust is associated with childhood asthma development

Abstract

Dampness and visible mold in homes are associated with asthma development, but causal mechanisms remain unclear. The goal of this research was to explore associations among measured dampness, fungal exposure, and childhood asthma development without the bias of culture-based microbial analysis. In the low-income, Latino CHAMACOS birth cohort, house dust was collected at age 12 months, and asthma status was determined at age 7 years.The current analysis included 13 asthma cases and 28 controls. Next-generation DNA sequencing methods quantified fungal taxa and diversity. Lower fungal diversity (number of fungal operational taxonomic units) was significantly associated with increased risk of asthma development: unadjusted odds ratio(OR) 4.80 (95% confidence interval (CI) 1.04–22.1). Control for potential confounders strengthened this relationship. Decreased diversity within the genus Cryptococcus was significantly associated with increased asthma risk (OR 21.0, 95% CI 2.16–204). No fungal taxon (species, genus, class) was significantly positively associated with asthma development, and one was significantly negatively associated. Elevated moisture was associated with increased fungal diversity, and moisture/mold indicators were associated with four fungal taxa. Next-generation DNA sequencing provided comprehensive estimates of fungal identity and diversity, demonstrating significant associations between low fungal diversity and childhood asthma development in this community.

Practical implications: Early life exposure to low fungal diversity in house dust was associated with increased risk for later asthma developmen tin this low-income, immigrant community. No individual fungal taxon (species, genus, or class) was associated with asthma development, although exposure to low diversity within the genus Cryptococcus was associated with asthma development. Future asthma development studies should incorporate fungal diversity measurements, in addition to measuring individual fungal taxa. These results represent a step toward identifying the aspect(s) of indoor microbial populations that are associated with asthma development and suggest that understanding the factors that control diversity in the indoor environment may lead to public health recommendations for asthma prevention in the future.

Figure 1

Figure 1(A, B, C, D, E). Fungal diversity associations with asthma development and with the presence of moisture indicators. (A, B) Rarefaction analysis for fungi in case and control homes (n = 12 for asthma cases, n = 26 for controls), with values summarized in a bar graph. (C, D) Rarefaction analysis for fungi in houses with two or more qualitative moisture indicators (n=13 for homes with two or more moisture indicators, n = 25 for homes with fewer than two moisture indicators), with values summarized in a bar graph. Moisture indicators included peeling paint, water damage, rotting wood, musty odor, water leak in the kitchen, or visible mold growth. (E) Bar graph summary of rarefaction analysis for measured moisture at three threshold values (17, 21, and 24). The maximum moisture reading anywhere in the living area or child’s sleeping area was used. For the three thresholds of 17, 21, and 24, the respective numbers of homes with low moisture were n = 23, 26, and 33 and the respective numbers of homes with high moisture were n = 15, 12, and 5. Rarefaction curves appear in Figure S4. Error bars represent standard errors. All OTUs are defined at 97%

Figure 2

Figure 2A, B, C, D. Pearson correlation coefficients and graphs comparing maximum moisture content of any wall in the home compared to number of fungal OTUs in floor dust. Graphs are (A, B) stratified by homes without (n = 27) and with (n = 11) visible mold growth and (C, D) stratified by asthma control (n = 26) and case (n = 12) homes. All OTUs are defined at 97% similarity. Sequences were normalized to 450 sequences per sample. The respective Pearson correlation coefficient p-values were (A) 0.009, (B) 0.22, (C) 0.04, and (D) 0.81.

Figure 3. Negative confounding of the relationship between low fungal diversity and asthma development by measured home moisture

The unadjusted models are represented in solid grey arrows and the adjusted model is represented in open red arrows. (A) Negative confounding by high measured wall moisture (>24) reduced the association seen in unadjusted models for low fungal diversity and asthma development. This was because measured wall moisture had a positive association with asthma development but a negative association with low fungal diversity. With adjustment for measured wall moisture, the association between low fungal diversity and asthma development increased by 34% and that between measured wall moisture and asthma development increased by 127%. UOR = unadjusted odds ratio (solid grey), AOR = adjusted odds ratio (open red), * = p < 0.05.