Examining Different Analysis Protocols Targeting Hospital Sanitary Facility Microbiomes

Abstract

Indoor spaces exhibit microbial compositions that are distinctly dissimilar from one another and from outdoor spaces. Unique in this regard, and a topic that has only recently come into focus, is the microbiome of hospitals. While the benefits of knowing exactly which microorganisms propagate how and where in hospitals are undoubtedly beneficial for preventing hospital-acquired infections, there are, to date, no standardized procedures on how to best study the hospital microbiome. Our study aimed to investigate the microbiome of hospital sanitary facilities, outlining the extent to which hospital microbiome analyses differ according to sample-preparation protocol. For this purpose, fifty samples were collected from two separate hospitals-from three wards and one hospital laboratory-using two different storage media from which DNA was extracted using two different extraction kits and sequenced with two different primer pairs (V1-V2 and V3-V4). There were no observable differences between the sample-preservation media, small differences in detected taxa between the DNA extraction kits (mainly concerning Propionibacteriaceae), and large differences in detected taxa between the two primer pairs V1-V2 and V3-V4. This analysis also showed that microbial occurrences and compositions can vary greatly from toilets to sinks to showers and across wards and hospitals. In surgical wards, patient toilets appeared to be characterized by lower species richness and diversity than staff toilets. Which sampling sites are the best for which assessments should be analyzed in more depth. The fact that the sample processing methods we investigated (apart from the choice of primers) seem to have changed the results only slightly suggests that comparing hospital microbiome studies is a realistic option. The observed differences in species richness and diversity between patient and staff toilets should be further investigated, as these, if confirmed, could be a result of excreted antimicrobials.

Keywords: 16S rRNA gene sequencing; NGS; built environment; high-throughput DNA sequencing; hospital environment; hospital-acquired infections; microbial ecology; microbiome.

Figure 1

Specimen collection, preservation, and processing flow. The color of the boxes on the right of the image indicates how many samples from each workflow were ultimately sequenced with the corresponding primers.

Figure 2

(A) Overall prevalence of the 30 most important taxa. Colors allow a more straightforward distinction between the different levels of taxonomic resolution. (B) Prevalence of the five most important phyla in the two hospitals and (C) institutes/wards. (D) Differences in richness and (E) Fisher-alpha diversity across the different institutes/wards.

Figure 3

Principal coordinate analysis (PCoA) plot with Bray–Curtis (left) and Jaccard (right) dissimilarity highlighting operational taxonomic unit (OUT) differences and similarities between MH samples linked to the sampling site.