Functional diversity of carbohydrate-active enzymes enabling a bacterium to ferment plant biomass

Abstract

Microbial metabolism of plant polysaccharides is an important part of environmental carbon cycling, human nutrition, and industrial processes based on cellulosic bioconversion. Here we demonstrate a broadly applicable method to analyze how microbes catabolize plant polysaccharides that integrates carbohydrate-active enzyme (CAZyme) assays, RNA sequencing (RNA-seq), and anaerobic growth screening. We apply this method to study how the bacterium Clostridium phytofermentans ferments plant biomass components including glucans, mannans, xylans, galactans, pectins, and arabinans. These polysaccharides are fermented with variable efficiencies, and diauxies prioritize metabolism of preferred substrates. Strand-specific RNA-seq reveals how this bacterium responds to polysaccharides by up-regulating specific groups of CAZymes, transporters, and enzymes to metabolize the constituent sugars. Fifty-six up-regulated CAZymes were purified, and their activities show most polysaccharides are degraded by multiple enzymes, often from the same family, but with divergent rates, specificities, and cellular localizations. CAZymes were then tested in combination to identify synergies between enzymes acting on the same substrate with different catalytic mechanisms. We discuss how these results advance our understanding of how microbes degrade and metabolize plant biomass.

Figure 1. C. phytofermentans growth on pectic A–E, hemicellulosic F–J, and glucan K–L.

Polysaccharides: homogalacturonan A, rhamnogalacturonan I B, galactan C, arabinan D, arabinogalactan II E, xylan F, arabinoxylan G, glucomannan H, galactomannan I, xyloglucan J, carboxymethylcellulose K, starch L. Growth was measured as OD₆₀₀ every 15 minutes. Each point is the mean of six cultures; red lines show one standard deviation.

Figure 2. mRNA expression of all 171 CAZymes during growth on pectins A–C, hemicelluloses D–E, glucans F–H, and raw corn stover I relative to expression on glucose.

Expression was quantified as log₂(RPKM) with significantly differentially expressed genes on a given polysaccharide shown as triangles and unchanged genes as circles. The 56 purified CAZymes are red and others are blue.

Figure 3. CAZymes clustered based on gene expression patterns (clusters A–I) show that some genes respond to multiple carbon sources while others are substrate-specific.

mRNA expression changes (log₂ expression ratios relative to glucose) for all 92 CAZyme genes differentially expressed on at least 1 polysaccharide relative to glucose were separated into nine clusters using K-means. Plot centers are expression on glucose and concentric rings show log₂ up-regulation on the following carbon sources: cellobiose (Cb), filter paper cellulose (Cl), starch (Sa), xylose (Xo), xylan (Xy), arabinan (Ar), galacturonic acid (Ga), homogalacturonan (Hg), galactan (Gl), galactomannan (Gm), raw corn stover (Co). Gene membership of clusters is shown in Table S9.

Figure 4. Cleavage A, binding B, and CAZy database classification C of purified enzymes.

A Polysaccharide cleavage was quantified as nmol reducing sugar released per milligram enzyme per minute: >160 red, 80–160 orange, 40–80 yellow, 20–40 green, <20 gray. B Binding to insoluble polysaccharides was quantified as percentage enzyme bound to substrate: >30% red, 20–30% orange, 15–20% yellow, 10–15% green, <10% gray. C CAZy database classifications: glycoside hydrolases (GH), carbohydrate esterases (CE), polysaccharide lyases (PL), and carbohydrate binding domains (CBM). Among 56 purified CAZymes, only the 32 enzymes for which activities were found are shown.

Figure 5. Members of the same CAZy family vary in polysaccharide cleavage activities and CAZymes can by potentiated by other enzymes.

A Variation in cleavage activities of GH10 enzymes on xylan. B GH5 and C GH26 family members differ in their activities and substrate specificities on amorphous cellulose (red), glucomannan (green), xyloglucan (violet), galactomannan (yellow), mannan (gray). Enzyme activities in A–C are nmol reducing sugar released per milligram enzyme per minute. D–G CAZyme mixtures have higher activities than the individual enzymes. D Cphy1163 and Cphy3367 alone and together on amorphous cellulose. E Cphy2105, Cphy3009, and Cphy3207 alone and the latter two enzymes plus Cphy2105 on xylan. F Cphy1719 and Cphy1071 alone and together on glucomannan. G Cphy1687, Cphy2567, and Cphy3310 alone and the latter two enzymes plus Cphy1687 on homogalacturonan. In D–G, enzyme activities are shown as reducing sugar (nmol) produced by individual and combined enzymes. The fraction of the reducing sugar produced by the mixed enzymes that exceeds the sum of the individual enzymes is shown in green.

Figure 6. Model of polysaccharide degradation and metabolism by C. phytofermentans. CAZymes (shown as the number of enzymes in CAZy families) are based on purified activities and are intra- or extracellular based on putative secretion signals.

Metabolic enzymes are shown as NCBI numbers and are proposed based on mRNA expression. Rhamnose transport and assimilation is based on pathway from . Abbreviations are D-galacturonic acid (GA), L-rhamnose (R), D-mannose (M), D-glucose (Gc), D-galactose (G), D-xylose (X), L-arabinose (A), fructose (F), phosphate (P), pentose phosphate pathway (PPP), dihydroxyacetone-phosphate (DHAP), glyceraldehyde-3-phosphate (G3P). For each substrate, the number of significantly up-regulated extracellular solute binding proteins (ESB) and ABC transporters (ABC) are shown. Shaded regions show metabolism of glucose (green), mannose (blue), xylose and arabinose (yellow), rhamnose (orange), and galacturonic acid (red).