Gradual Recovery of Building Plumbing-Associated Microbial Communities after Extended Periods of Altered Water Demand during the COVID-19 Pandemic

Abstract

COVID-19 pandemic-related building restrictions heightened drinking water microbiological safety concerns post-reopening due to the unprecedented nature of commercial building closures. Starting with phased reopening (i.e., June 2020), we sampled drinking water for 6 months from three commercial buildings with reduced water usage and four occupied residential households. Samples were analyzed using flow cytometry and full-length 16S rRNA gene sequencing along with comprehensive water chemistry characterization. Prolonged building closures resulted in 10-fold higher microbial cell counts in the commercial buildings [(2.95 ± 3.67) × 10⁵ cells mL^-1] than in residential households [(1.11 ± 0.58) × 10⁴ cells mL^-1] with majority intact cells. While flushing reduced cell counts and increased disinfection residuals, microbial communities in commercial buildings remained distinct from those in residential households on the basis of flow cytometric fingerprinting [Bray-Curtis dissimilarity (d_BC) = 0.33 ± 0.07] and 16S rRNA gene sequencing (d_BC = 0.72 ± 0.20). An increase in water demand post-reopening resulted in gradual convergence in microbial communities in water samples collected from commercial buildings and residential households. Overall, we find that the gradual recovery of water demand played a key role in the recovery of building plumbing-associated microbial communities as compared to short-term flushing after extended periods of reduced water demand.

Keywords: COVID-19 pandemic; flow cytometry; premise plumbing; stagnation; water quality.

Figure 1

Water demand (cubic meters per month) at three commercial building (COM, ●) and four residential household (RES, ▲) sites from January 2019 to November 2020. The dashed lines indicate the closure of commercial building sites due to COVID-19 policy interventions, commercial building sites reopening for research activity, and commercial building sites reopening for the fall 2020 academic semester. The red shaded area indicates the 6 month sampling period (June–November 2020).

Figure 2

(A) Bubble plot showing the proportion of intact cells (depicted by size) with phenotypic diversity and evenness indices of TP_0–30 samples that were collected during the first week of building reopening from the commercial building and residential household sites. Panels from the left to right represent measures related to (i) all samples (TP_0–30), (ii) first draw samples (TP₀), and (iii) final samples (TP₃₀). (B) NMDS plot illustrating differences in microbial community composition among TP_0–30 samples of commercial building and residential household sites estimated using Bray–Curtis distances derived from flow cytometric fingerprinting data. Ellipses drawn at a 95% confidence limit. (C and D) Bray–Curtis dissimilarity-based dbRDA biplot using flow cytometric and 16S rRNA gene sequencing data illustrating the relationship among the site type, water chemistry parameters, and microbial community structure of TP₀ and TP₃₀ samples of the commercial building (orange) and residential household (gray) sites. Water quality parameters that significantly explained differences in community composition based on PERMANOVA analysis are shown as black arrows. Colinear variables are shown in brackets, with “POS” and “NEG” denoting positive and negative correlations at Pearson’s correlation coefficients of >0.70 and <0.70, respectively.

Figure 3

(A) Maximum likelihood phylogenetic tree showing the grouping of Mycobacterium ASVs with 16S rRNA gene sequence of Mycobacterium reference strains. Bootstrap analysis of 1000 replicates was performed, and bootstrap values were reported as percentages (depicted by size). (B) Scatter plot showing the absolute abundance on a log scale of 22 Mycobacterium ASVs detected in the TP₀ and TP₃₀ samples of the commercial buildings and residential households. Mycobacterium ASVs classified at the species level had >99% relatedness with reference Mycobacterium 16S rRNA strains (Table S6) and were grouped as slow growers (red) or rapid growers (blue).

Figure 4

(A) Monthly flush profiles based on TCC measurements that were averaged for individual time points within each month across all commercial building and residential household sites. Bars represent the standard deviation, indicating the dispersion of individual TCC measures in relation to the mean. (B) Bubble plot showing intact cell proportions (depicted by size) with the phenotypic diversity index (D₂) and evenness of monthly samples that were collected from the commercial building and residential household sites over 6 months. Panels from the top to bottom represent measures related to first draw samples (TP₀), and subsequent samples (TP_5–30), respectively.

Figure 5

Linear regression slope estimated for Bray–Curtis dissimilarity distances of all time point combinations over the flush duration for commercial building (orange) and residential household (gray) sites. The red diamond represents the mean calculated across the linear regression slope estimates for each month.

Figure 6

(A and B) NMDS plot illustrating differences in microbial community composition between samples of commercial buildings (orange) and residential households (gray) sites, using Bray–Curtis dissimilarity distances on flow cytometric fingerprinting and 16S rRNA gene sequencing data, respectively. The ellipse is drawn at a 95% confidence limit. (C) NMDS plot based on Bray–Curtis dissimilarity distances on 16S rRNA sequencing data illustrating differences in microbial community composition between TP₃₀ samples of commercial buildings (●) and residential households (▲) between months. (D) Structure-based Bray–Curtis distance to the group centroid of commercial building and residential household microbial communities grouped by month. The red diamond represents the mean centroid calculated within each month.