Genomic Characterization of Candidate Division LCP-89 Reveals an Atypical Cell Wall Structure, Microcompartment Production, and Dual Respiratory and Fermentative Capacities

Abstract

Recent experimental and bioinformatic advances enable the recovery of genomes belonging to yet-uncultured microbial lineages directly from environmental samples. Here, we report on the recovery and characterization of single amplified genomes (SAGs) and metagenome-assembled genomes (MAGs) representing candidate phylum LCP-89, previously defined based on 16S rRNA gene sequences. Analysis of LCP-89 genomes recovered from Zodletone Spring, an anoxic spring in Oklahoma, predicts slow-growing, rod-shaped organisms. LCP-89 genomes contain genes for cell wall lipopolysaccharide (LPS) production but lack the entire machinery for peptidoglycan biosynthesis, suggesting an atypical cell wall structure. The genomes, however, encode S-layer homology domain-containing proteins, as well as machinery for the biosynthesis of CMP-legionaminate, inferring the possession of an S-layer glycoprotein. A nearly complete chemotaxis machinery coupled to the absence of flagellar synthesis and assembly genes argues for the utilization of alternative types of motility. A strict anaerobic lifestyle is predicted, with dual respiratory (nitrite ammonification) and fermentative capacities. Predicted substrates include a wide range of sugars and sugar alcohols and a few amino acids. The capability of rhamnose metabolism is confirmed by the identification of bacterial microcompartment genes to sequester the toxic intermediates generated. Comparative genomic analysis identified differences in oxygen sensitivities, respiratory capabilities, substrate utilization preferences, and fermentation end products between LCP-89 genomes and those belonging to its four sister phyla (Calditrichota, SM32-31, AABM5-125-24, and KSB1) within the broader FCB (Fibrobacteres-Chlorobi-Bacteroidetes) superphylum. Our results provide a detailed characterization of members of the candidate division LCP-89 and highlight the importance of reconciling 16S rRNA-based and genome-based phylogenies.IMPORTANCE Our understanding of the metabolic capacities, physiological preferences, and ecological roles of yet-uncultured microbial phyla is expanding rapidly. Two distinct approaches are currently being utilized for characterizing microbial communities in nature: amplicon-based 16S rRNA gene surveys for community characterization and metagenomics/single-cell genomics for detailed metabolic reconstruction. The occurrence of multiple yet-uncultured bacterial phyla has been documented using 16S rRNA surveys, and obtaining genome representatives of these yet-uncultured lineages is critical to our understanding of the role of yet-uncultured organisms in nature. This study provides a genomics-based analysis highlighting the structural features and metabolic capacities of a yet-uncultured bacterial phylum (LCP-89) previously identified in 16S rRNA surveys for which no prior genomes have been described. Our analysis identifies several interesting structural features for members of this phylum, e.g., lack of peptidoglycan biosynthetic machinery and the ability to form bacterial microcompartments. Predicted metabolic capabilities include degradation of a wide range of sugars, anaerobic respiratory capacity, and fermentative capacities. In addition to the detailed structural and metabolic analysis provided for candidate division LCP-89, this effort represents an additional step toward a unified scheme for microbial taxonomy by reconciling 16S rRNA gene-based and genomics-based taxonomic outlines.

Keywords: candidate phyla; environmental genomics; metagenomic bins; single-cell genomics.

FIG 1

Maximum-likelihood phylogenetic tree based on the concatenated protein alignment of 120 single-copy markers, highlighting the phylogenetic position of LCP-89 genomes. Reference taxa are either type strains of cultured microorganisms or genomes of relevant uncultured bacterial phyla recovered using single-cell genomics or genome-resolved metagenomics. MAGs and SAGs obtained as part of this study are shown in boldface, with LCP-89 genomes in red boldface. The concatenated alignment used to construct the protein tree was generated using the GTDB-Tk. The tree was obtained using RaxML. Bootstrap values (from 100 replicates) are shown for nodes with bootstrap support of more than 50. Accession numbers are provided in parentheses.

FIG 2

Maximum-likelihood phylogenetic trees based on 16S rRNA genes, highlighting the phylogenetic position of LCP-89 genomes. Reference taxa are type strains of cultured microorganisms, genomes of relevant uncultured bacterial phyla recovered using single-cell genomics or genome-resolved metagenomics, and 16S rRNA amplicons recovered in culture-independent 16S rRNA gene diversity surveys. MAGs and SAGs obtained as part of this study are shown in boldface, with LCP-89 genomes in red boldface. The tree was generated using SILVA-aligned sequences and obtained using FastTree. Bootstrap values (from 100 replicates) are shown for nodes with bootstrap support of more than 50. Accession numbers are provided in parentheses.

FIG 3

Cartoon depicting the predicted cell wall structure of the following: members of LCP-89; phyla KSB1, Calditrichota, and SM23-31; candidate phylum AABM5-125-24; Dehalococcoidia class of Chloroflexi; and a typical pseudomurein-containing archaeal cell wall. Note the absence of a peptidoglycan layer in LCP-89 members, as opposed to its presence in all other sister phyla. LCP-89 members, as well as members of sister phyla KSB1, Calditrichota, and SM23-31, are predicted to have an external N-glycosylated S-layer. The predicted absence of peptidoglycan in LCP-89 cell walls is similar to its predicted absence in cell walls of the Dehalococcoidia class of Chloroflexi, albeit Dehalococcoidia cell walls lack an outer membrane with LPS. A typical pseudomurein-containing archaeal cell wall is depicted for comparative purposes. IM, inner membrane; OM, outer membrane.

FIG 4

Metabolic reconstruction of candidate phylum LCP-89 as predicted by collectively analyzing 3 SAGs and 1 MAG belonging to the phylum. All possible substrates potentially supporting growth are shown in blue, while predicted final products are shown in purple. The inner membrane is depicted as a phospholipid bilayer interrupted by membrane proteins color coded as follows: components of the predicted respiratory chain are shown in green, the ATPase complex is shown in red, transporters of the phosphotransferase system are shown in orange, ABC transporters are shown in brown, and secondary transporters are shown in gray. The bacterial microcompartment (BMC) is depicted by an octahedral structure showing all reactions predicted to occur inside the BMC. α-KG, α-ketoglutarate; Amt, ammonium channel transporter; Asp, aspartic acid; DHAP, dihydroxyacetone phosphate; E-I, enzyme I of the PTS; E-IIA-C, subunits A, B, and C of enzyme II of the PTS; Fru, fructose; Fru-1,6-PP, fructose-1,6-bisphosphate; Fum, fumarate; GAP, glyceraldehyde-3-phosphate; GluC, glucose; l-Ald, lactaldehyde; Man, mannose; NrfAH, cytochrome c nitrite reductase (NH3 forming) [EC 1.7.2.2]; OAA, oxaloacetate; P, permeases of the ABC transporter; 1,2-PD, 1,2-propanediol; P-ald, propionaldehyde; Prop-CoA, propionyl-coenzyme A; Pyr, pyruvate; PPP, pentose phosphate pathway; Q, quinone; Rha, rhamnose; SBP, substrate binding protein of the ABC transporter; Succ, succinate; SDH, succinate dehydrogenase; TCA, tricarboxylic acid cycle; Xyl, xylose; Xylu, xylulose.