Characterizing and contrasting the microbial ecology of laboratory and full-scale EBPR systems cultured on synthetic and real wastewaters

Abstract

The anthropogenic discharge of phosphorus (P) into surface waters can induce the proliferation of cyanobacteria and algae, which can negatively impact water quality. Enhanced biological P removal (EBPR) is an engineered process that can be employed to efficiently remove significant quantities of P from wastewater. Within this engineered system, the mixed microbial consortium (MMC) becomes enriched with polyphosphate accumulating organisms (PAOs). To date much knowledge has been developed on PAOs, and the EBPR process is generally well understood; nonetheless, the engineered process remains underutilized. In this study, investigations were conducted using qPCR and Illumina MiSeq to assess the impacts of wastewater (synthetic vs. real) on EBPR microbial ecology. While a strong relationship was demonstrated between EBPR metrics (P:C; influent VFA:P) and excellent P removal across diverse EBPR systems and MMCs, no such correlations existed with the specific MMCs. Moreover, MMCs exhibited distinct clusters based on substrate, and qPCR results based on the putative PAO Accumulibacter did not correlate with BLASTN eubacterial results for either Accumulibacter or Rhodocyclaceae. More critically, PAO-based sequences aligned poorly with Accumulibacter for both eubacterial and PAO primer sets, which strongly suggests that the conventional PAO primers applied in FISH and qPCR analysis do not sufficiently target the putative PAO Accumulibacter. In particular, negligible alignment was observed for PAO amplicons obtained from a MMC performing excellent EBPR on crude glycerol (an atypical substrate). A synthetic wastewater-based MMC exhibited the best observed BLASTN match of the PAO amplicons, raising concerns about the potential relevance in using synthetic substrates in the study of EBPR.

Keywords: Accumulibacter; Crude glycerol; Enhanced biological phosphorus removal, EBPR; Illumina MiSeq; Phosphorus accumulating organisms, PAOs; Volatile fatty acids, VFAs; qPCR.

Figure 1

Influent wastewater P:C ratio against effluent P for an array of laboratory-scale EBPR reactors (this study and (Coats et al. 2011a, Coats et al. 2011b, Coats et al. 2015, Horgan et al. 2010, Winkler et al. 2011)).

Figure 2

Influent wastewater VFA:P ratio against effluent P for an array of laboratory-scale EBPR reactors (this study and (Coats et al. 2011a, Coats et al. 2011b, Coats et al. 2015, Horgan et al. 2010, Winkler et al. 2011)).

Figure 3

Bivariate plot of the first two principal component (Dim 1 and Dim 2) scores for the 16S rRNA gene sequencing results using the EUB primer set at the genus level and the P removed. Samples have been grouped by system (distinguished by color) with the numbers corresponding to their index in an ordered list (i.e., sorted by operational day following an arbitrary system ordering: 1 to 7 are S-EBPR operational days 72 to 283; 8 to 14 are V-EBPR operational days 72 to 283; 15 to 21 are G-EBPR operational days 72 to 283; and 22 to 28 are R-EBPR operational days 72 to 283. The centroid for each group in the Dim 1-Dim 2 space is depicted as a square in the associated color and the ellipses denote the 95% confidence regions for the associated centroids in the Dim 1-Dim 2 space.

Figure 4

Heat map showing the relative abundance of phylotypes identified at the genus level using the EUB primer set. Unidentified phylotypes (denoted as “unknown”) were aggregated by the most specific taxonomic level at which they were identified (the “*” is used to indicate any identified taxon name). Note that the adopted color scheme is nonlinear at 0% relative abundance to differentiate phylotypes which were not detected (N. D.) in a sample from those that were. To conserve space, some taxon names have been abbreviated (Ca.: Candidatus; ^a: _incertae_sedis; ^b: _genera_incertae_sedis; and ^c: sensu stricto). The phylotypes with less than 2.5% relative abundance in all samples were aggregated and denoted “Phylotypes < 2.5%” followed by the number of phylotypes so categorized in parentheses at the bottom of the heat map.