Microbiota of the indoor environment: a meta-analysis

Abstract

Background: As modern humans, we spend the majority of our time in indoor environments. Consequently, environmental exposure to microorganisms has important implications for human health, and a better understanding of the ecological drivers and processes that impact indoor microbial assemblages will be key for expanding our knowledge of the built environment. In the present investigation, we combined recent studies examining the microbiota of the built environment in order to identify unifying community patterns and the relative importance of indoor environmental factors. Ultimately, the present meta-analysis focused on studies of bacteria and archaea due to the limited number of high-throughput fungal studies from the indoor environment. We combined 16S ribosomal RNA (rRNA) gene datasets from 16 surveys of indoor environments conducted worldwide, additionally including 7 other studies representing putative environmental sources of microbial taxa (outdoor air, soil, and the human body).

Results: Combined analysis of subsets of studies that shared specific experimental protocols or indoor habitats revealed community patterns indicative of consistent source environments and environmental filtering. Additionally, we were able to identify several consistent sources for indoor microorganisms, particularly outdoor air and skin, mirroring what has been shown in individual studies. Technical variation across studies had a strong effect on comparisons of microbial community assemblages, with differences in experimental protocols limiting our ability to extensively explore the importance of, for example, sampling locality, building function and use, or environmental substrate in structuring indoor microbial communities.

Conclusions: We present a snapshot of an important scientific field in its early stages, where studies have tended to focus on heavy sampling in a few geographic areas. From the practical perspective, this endeavor reinforces the importance of negative "kit" controls in microbiome studies. From the perspective of understanding mechanistic processes in the built environment, this meta-analysis confirms that broad factors, such as geography and building type, structure indoor microbes. However, this exercise suggests that individual studies with common sampling techniques may be more appropriate to explore the relative importance of subtle indoor environmental factors on the indoor microbiome.

Fig. 1

Bacterial community distance within and between indoor surfaces. A subset of studies from similar indoor environments was analyzed (Colorado kitchen surfaces, Colorado restroom surfaces, South Korea restroom and kitchen surfaces, and North Carolina kitchen and restroom surfaces), and figures show the density of unweighted UniFrac pairwise distances (a) within restrooms, within kitchens, and between restrooms and kitchen, as well as (b) within toilets, within fridges, and between toilets and fridges. Results indicate that the bacterial OTUs found on these surfaces tend to be more similar to each other than between surfaces

Fig. 2

Sources tracking of indoor environments. A subset of samples from each of the studies (see Table 1) was analyzed using the SourceTracker algorithm to apportion microbial sources for different “sinks” of indoor settings. Prominent sources were outdoor air, skin, soil, and laboratory kits (a), although the likelihood of identifying sources varied strongly by study. b Those studies that were more likely to have sources identified were those that originally included source environment samples (using their own sampling and laboratory methods—denoted by asterisk in the figure), contained samples that were more strongly sourced from skin, or targeted the same variable region as those in the source samples

Fig. 3

Principal coordinate analysis (PCoA) of bacteria in the 16 “sink” studies in this meta-analysis. Communities are compared using the unweighted UniFrac distance metric. a Studies cluster generally by study identity, and the top ten indicator taxa (triangles) are indicative of human-associated bacteria as well as outdoor-derived taxa. b Bacterial community composition also tend to group by the matrix type (the physical sample type) as well as the way the building is used (c)

Fig. 4

The proportion of sequences assigned to a reference database varies by study. Datasets from the human body (denoted by asterisk in the figure) and buildings with heavy dispersal from the human body tended to be better represented in the GreenGenes database, while soil and general outdoor sources were not as well represented. The numbers in parentheses show the number of samples in each study

Fig. 5

Difference between taxonomic and phylogenetic distance methods. Points from four studies of similar indoor environments (restroom and kitchen surfaces) are colored by study and analyzed by the Canberra community distance (a) and unweighted Unifrac (b). Similarly, points colored by sequencing protocols (including different primers and platforms) differ according to the Canberra distance (c) and the unweighted Unifrac (d)

Fig. 6

Closed and open-reference OTU picking yielded similar β-diversity results. Three pairs of studies were separately analyzed using the two OTU picking methods and then compared using Procrustes analysis. Each point is the result of open-reference OTU picking, and each arrowhead is the same sample from closed-reference OTU picking. A significant Procrustes statistic indicates that the results from β-diversity analysis are strongly correlated. The same sample across the two methods are linked with an arrow. a California dairy and neonatal intensive care unit, both near Davis, California; b North Carolina homes in and near Raleigh, North Carolina, and Boulder, Colorado residential kitchen surfaces; c Oregon classroom air and surface samples from Eugene. Although the California Dairy study appears to be different between the two methods (a), the Dairy site was statistically distinct from the paired NICU study regardless of OTU picking method

Fig. 7

Example of taxonomic bias observed in technical control samples compared to environmental samples. The composition of each of the pooled a dust, b kit controls, and c surfaces samples is shown as a donut. The “kit microbiome” displayed higher abundances of the bacterial phylum Tenericutes (green slice in the donut chart indicated by arrows). Per-sample abundance of Tenericutes is represented by green bars displayed across all panels. Some samples in the North Carolina homes study showed similar levels of Tenericutes compared to the kit controls (c, far right); this implies some level of contamination in non-control environmental samples from this study, which the authors identified and removed in the original study [9]. Donut and bar charts were generated using the Phinch data visualization framework [63]