A bacterial pioneer produces cellulase complexes that persist through community succession

Abstract

Cultivation of microbial consortia provides low-complexity communities that can serve as tractable models to understand community dynamics. Time-resolved metagenomics demonstrated that an aerobic cellulolytic consortium cultivated from compost exhibited community dynamics consistent with the definition of an endogenous heterotrophic succession. The genome of the proposed pioneer population, 'Candidatus Reconcilibacillus cellulovorans', possessed a gene cluster containing multidomain glycoside hydrolases (GHs). Purification of the soluble cellulase activity from a 300litre cultivation of this consortium revealed that ~70% of the activity arose from the 'Ca. Reconcilibacillus cellulovorans' multidomain GHs assembled into cellulase complexes through glycosylation. These remarkably stable complexes have supramolecular structures for enzymatic cellulose hydrolysis that are distinct from cellulosomes. The persistence of these complexes during cultivation indicates that they may be active through multiple cultivations of this consortium and act as public goods that sustain the community. The provision of extracellular GHs as public goods may influence microbial community dynamics in native biomass-deconstructing communities relevant to agriculture, human health and biotechnology.

Fig. 1. Cultivation of cellulolytic consortium at 15 l scale.

a, Relative abundance of dominant populations (≥1%) at the end of the cultivation. Detailed genome information and average coverage are provided in Supplementary Tables 1 and 2. Single DNA samples were isolated from the 15 l culture on each indicated day for metagenomic sequencing. b, CMCase (red) and xylanase (green) activity measurements obtained by daily sampling of the 15 l cultivation. Enzymatic assays are reported as the mean of technical replicates (n = 3) and error bars represent standard error of the mean. c, Daily relative abundances (calculated using the average number of reads of binned scaffolds from time-series metagenomic data for the 15 l cultivation) of the Paenibacillaceae 1 population (red) during a 14-day cultivation of the consortium grown with microcrystalline cellulose.

Fig. 2. Genome analysis.

a, Population genome of ‘Ca. R. cellulovorans’ recovered from metagenomics data from the 15 l cultivation. The genome was dispersed on 114 scaffolds (blue), with 2,814 predicted CDS (coding DNA sequences) in forward (red) and reverse (green) and average (orange) coverage. N50 is the shortest sequence length that includes 50% of the assembled genome, summing from the largest contig. b, Maximum-likelihood phylogenetic tree based on 86 concatenated amino acid sequences that are conserved in the Paenibacillaceae (Supplementary Table 6). c, Molecular organization of multidomain GH genes from the population genome of ‘Ca. R. cellulovorans’ arranged in a 17-kb gene cluster.

Fig. 3. Analysis of GH complexes.

a, Separation of cellulase and xylanase complexes eluted from an anion-exchange chromatography column at 260 mM NaCl and visualized by 2D BN–PAGE. Complexes containing S-Layer proteins (S), CMCase (C) and xylanase (X) activity were separated in the first dimension according to their indicated masses by BN–PAGE. Protein staining was accompanied by zymography with gels embedded with CMC and xylan. b, Subunits of the native complexes were separated in a second dimension by SDS–PAGE (8%) and identified as CelABC and XynA by proteomics. The XynA molecular weight (~80 kDa) is indicative of a truncation of the full-length protein (100 kDa). Zymography with CMC revealed the activity of bands corresponding to CelC and CelA. c, GH complexes enriched by affinity digestion were separated by BN–PAGE and protein stains were accompanied by zymography with gels embedded with CMC and xylan. d, Native complexes were separated by SDS–PAGE into subunits CelABC and visualized by protein and CMCase activity staining. e, SDS–PAGE was also performed without initial heat denaturation, and three abundant individual complexes with different compositions of CelABC were identified by proteomics. Detailed proteomics data are provided in Supplementary Figs. 8 and 9. Images were cropped for clarity. Each gel is representative of five gels performed on multiple protein preparations. Gels stained with Coomassie and analysed by zymography were run in parallel in the same electrophoretic cell to ensure comparability. The gels in this figure are from one individual protein preparation for each of the two purification techniques described in the text. The gel images were cropped for clarity and the original gel images are provided in Supplementary Fig. 14.

Fig. 4. Cellulolytic activities of CelABC.

a, Saccharification of phosphoric acid swollen cellulose (PASC) and Avicel for the culture supernatant, affinity digested preparation (AD) and recombinant CelABC individually expressed in E. coli (see Methods for details). Enzymatic assays are reported as the mean of technical replicates (n = 3) and error bars represent standard error of the mean. b, CMCase-specific activities of the supernatant, AD fraction and in E. coli-expressed CelABC. Enzymatic assays are reported as the mean of technical replicates (n = 3) and error bars represent standard error of the mean. c, SDS–PAGE (8–16% gradient; stained with Coomassie Brilliant Blue) of the AD fraction and E. coli-expressed CelABC. The gel depicted is representative of three gels that displayed very similar results. The gel images were cropped for clarity; the original gel images are provided in Supplementary Fig. 15.