Capturing single cell genomes of active polysaccharide degraders: an unexpected contribution of Verrucomicrobia

Abstract

Microbial hydrolysis of polysaccharides is critical to ecosystem functioning and is of great interest in diverse biotechnological applications, such as biofuel production and bioremediation. Here we demonstrate the use of a new, efficient approach to recover genomes of active polysaccharide degraders from natural, complex microbial assemblages, using a combination of fluorescently labeled substrates, fluorescence-activated cell sorting, and single cell genomics. We employed this approach to analyze freshwater and coastal bacterioplankton for degraders of laminarin and xylan, two of the most abundant storage and structural polysaccharides in nature. Our results suggest that a few phylotypes of Verrucomicrobia make a considerable contribution to polysaccharide degradation, although they constituted only a minor fraction of the total microbial community. Genomic sequencing of five cells, representing the most predominant, polysaccharide-active Verrucomicrobia phylotype, revealed significant enrichment in genes encoding a wide spectrum of glycoside hydrolases, sulfatases, peptidases, carbohydrate lyases and esterases, confirming that these organisms were well equipped for the hydrolysis of diverse polysaccharides. Remarkably, this enrichment was on average higher than in the sequenced representatives of Bacteroidetes, which are frequently regarded as highly efficient biopolymer degraders. These findings shed light on the ecological roles of uncultured Verrucomicrobia and suggest specific taxa as promising bioprospecting targets. The employed method offers a powerful tool to rapidly identify and recover discrete genomes of active players in polysaccharide degradation, without the need for cultivation.

Figure 1. Optimization of cell probing conditions with fluorescently labeled laminarin.

Flow cytometric dot plots of heat-killed and live freshwater samples incubated for various lengths of time with 4 or 40 µM fluorescently labeled laminarin. Red polygons indicate gates used to count putative laminarin-positive cells.

Figure 2. Flow-cytometric sort gates (A) and taxonomic composition (B) of single amplified genomes (SAGs) generated from coastal bacterioplankton using various fluorescent probes.

Bacterioplankton were probed with (from top to bottom): 1) nucleic acid stain SYTO-9, targeting high- and low-nucleic acid content cells (HNA and LNA cells) representing a random subset of the entire microbial assemblage; 2) fluorescently-labeled laminarin; 3) fluorescently-labeled xylan; 4) 5-cyano-2,3-ditolyltetrazolium chloride (ETS-active cells) and 5) carboxyfluoresceindiacetate (esterase-active cells). Gates used for cell sorting are indicated in blue.

Figure 3. Phylogenetic composition of Verrucomicrobia SAGs.

(A) Maximum likelihood phylogenetic analysis of the SSU rRNA gene sequences. Bootstrap (1000 replicates) values ≥50 are displayed. Each phylotype, indicated in red (coastal) or blue (freshwater) is formed by SAGs with ≥99% SSU rRNA gene sequence similarity. Five SAGs from the most abundant putative polysaccharide degrader phylotype in the coastal sample were selected for whole genome sequencing (red star). (B) Phylotype relative abundances in SAG libraries generated using various fluorescent probes. (nd) = not detected in a SAG library.

Figure 4. Comparative analysis of genes encoding hydrolytic enzymes in prokaryote genomes.

The bar chart indicates the genome-wide frequency of glycoside hydrolase genes in various microbial groups, average ± standard deviation. The number of publicly available genomes found in the IMG database (as of February 2012) for each taxonomic group is provided in parentheses. The average enrichment of glycoside hydrolases was also estimated for the Bacteria domain. The small pie chart shows the number and composition of genes involved in polysaccharide hydrolysis in the Verrucomicrobia SAG AAA168-F10. The large pie chart shows CAZy families of glycoside hydrolase genes detected in SAG AAA168-F10. Each glycoside hydrolase family is indicated as GH-xxx, according to CAZy database nomenclature .

Figure 5. Evidence for the laminarinase gene in the single amplified genome AAA168-F10.

(A) Active site, including the catalytic residues responsible for laminarin hydrolysis, derived from Conserved Domain Protein, SWISS-MODEL, and PROSITE databases. (B) Neighbor-joining phylogenetic tree of amino acid sequences, applying the Kimura evolutionary model and indicating bootstrap values above 50.