Reevaluation of the Phylogenetic Diversity and Global Distribution of the Genus " Candidatus Accumulibacter"

Abstract

"Candidatus Accumulibacter" was the first microorganism identified as a polyphosphate-accumulating organism (PAO) important for phosphorus removal from wastewater. Members of this genus are diverse, and the current phylogeny and taxonomic framework appear complicated, with most publicly available genomes classified as "Candidatus Accumulibacter phosphatis," despite notable phylogenetic divergence. The ppk1 marker gene allows for a finer-scale differentiation into different "types" and "clades"; nevertheless, taxonomic assignments remain inconsistent across studies. Therefore, a comprehensive reevaluation is needed to establish a common understanding of this genus, in terms of both naming and basic conserved physiological traits. Here, we provide this reassessment using a comparison of genome, ppk1, and 16S rRNA gene-based approaches from comprehensive data sets. We identified 15 novel species, along with "Candidatus Accumulibacter phosphatis," "Candidatus Accumulibacter delftensis," and "Candidatus Accumulibacter aalborgensis." To compare the species in situ, we designed new species-specific fluorescence in situ hybridization (FISH) probes and revealed their morphology and arrangement in activated sludge. Based on the MiDAS global survey, "Ca. Accumulibacter" species were widespread in wastewater treatment plants (WWTPs) with phosphorus removal, indicating process design as a major driver for their abundance. Genome mining for PAO-related pathways and FISH-Raman microspectroscopy confirmed the potential for PAO metabolism in all "Ca. Accumulibacter" species, with detection in situ of the typical PAO storage polymers. Genome annotation further revealed differences in the nitrate/nitrite reduction pathways. This provides insights into the niche differentiation of these lineages, potentially explaining their coexistence in the same ecosystem while contributing to overall phosphorus and nitrogen removal. IMPORTANCE "Candidatus Accumulibacter" is the most studied PAO, with a primary role in biological nutrient removal. However, the species-level taxonomy of this lineage is convoluted due to the use of different phylogenetic markers or genome sequencing approaches. Here, we redefined the phylogeny of these organisms, proposing a comprehensive approach which could be used to address the classification of other diverse and uncultivated lineages. Using genome-resolved phylogeny, compared to phylogeny based on the 16S rRNA gene and other phylogenetic markers, we obtained a higher-resolution taxonomy and established a common understanding of this genus. Furthermore, genome mining of genes and pathways of interest, validated in situ by application of a new set of FISH probes and Raman microspectroscopy, provided additional high-resolution metabolic insights into these organisms.

Keywords: diversity; metagenome-assembled genome; phylogeny; ppk1 gene; “Candidatus Accumulibacter”.

FIG 1

Comparison between genome- and ppk1-based phylogenies. The maximum-likelihood genome tree was created from the concatenated alignment of 120 single-copy marker gene proteins trimmed to 5,000 amino acids using GTDB-Tk and 100 bootstraps. Three Azonexus (formerly Dechloromonas) isolates (GenBank assembly accession numbers IMG taxon_id 637000088, GCA_000519045.1, and GCA_001551835.1) were used as an outgroup. The maximum-likelihood ppk1 gene tree was created from the alignment of the ppk1 genes extracted from the genomes, using 100 bootstraps. For NCBI GenBank genome accession numbers and leaf names, see Data File S1 (available at https://doi.org/10.6084/m9.figshare.17306771.v1). Gray boxes indicate the ppk1-based types nomenclature. Species representatives are indicated in blue. Of these, the ones with established or proposed “Candidatus” names are indicated in bold. Red asterisks indicate MAGs with full-length 16S rRNA gene sequences.

FIG 2

Maximum-likelihood (PhyML)16S rRNA gene phylogenetic tree of “Ca. Accumulibacter” and related species. 16S rRNA gene sequences retrieved from the MAGs are indicated in blue, the original 16S rRNA gene sequence (GenBank accession number AJ224937) retrieved from Hesselman et al. (3) is indicated in red. The species renamed in this study are indicated in bold blue. 16S rRNA gene partial sequences are indicated with a black asterisk. The alignment used for the tree applied a 20% conservational filter to remove hypervariable positions, giving 1,250 aligned positions. Coverage of the fluorescence in situ hybridization (FISH) probes designed in this study is indicated with black brackets and is based on the MiDAS4 database (29). Probe coverage of widely applied probes for the “Ca. Accumulibacter” clades is shown with yellow (PAO651), orange (Acc-I-44), and red boxes (Acc-II-444). Bootstrap values from 1,000 resamplings are indicated for branches with >70% (gray dot) and >90% (black) support. Species of the genus Dechloromonas were used as the outgroup. The scale bar represents substitutions per nucleotide base.

FIG 3

The 10 most abundant “Ca. Accumulibacter” species worldwide according to the MiDAS4 survey. (A) Mean relative abundance across different process configurations (C, carbon removal; N, nitrification; DN, denitrification; P, phosphorus removal). (B) Mean relative abundance in EBPR plants across different countries. Data originate from the global survey of microbial communities in WWTPs (29) and it is based on a V1 to V3 amplicon data set. Species marked in blue were wrongly classified as “Ca. Accumulibacter” according to the genome-based taxonomy.

FIG 4

Multicolor FISH micrograph of different “Ca. Accumulibacter” species in full-scale activated sludge. “Ca. Accumulibacter proximus” (green) was targeted by the Acc471 probe. “Ca. Accumulibacter propinquus” (blue) was targeted by the Acc1011 probe. “Ca. Accumulibacter regalis” (magenta) was targeted by the Acc635 probe. “Ca. Accumulibacter delftensis” and “Ca. Accumulibacter aalborgensis” (yellow) were targeted by the Acc470 probe. All bacteria (gray) were targeted with the EUBmix probe. Bar, 10 μm.

FIG 5

PAO metabolism-related functional potential of the “Ca. Accumulibacter” MAGs and closest relatives. The gene list follows the progression in the text. For the full list of gene names and associated KO numbers, see Data File S2 (available at https://doi.org/10.6084/m9.figshare.17306828.v1). The MAGs and genomes are ordered as in the genome tree in Fig. 1.