Characterization of novel lignocellulose-degrading enzymes from the porcupine microbiome using synthetic metagenomics

Abstract

Plant cell walls are composed of cellulose, hemicellulose, and lignin, collectively known as lignocellulose. Microorganisms degrade lignocellulose to liberate sugars to meet metabolic demands. Using a metagenomic sequencing approach, we previously demonstrated that the microbiome of the North American porcupine (Erethizon dorsatum) is replete with genes that could encode lignocellulose-degrading enzymes. Here, we report the identification, synthesis and partial characterization of four novel genes from the porcupine microbiome encoding putative lignocellulose-degrading enzymes: β-glucosidase, α-L-arabinofuranosidase, β-xylosidase, and endo-1,4-β-xylanase. These genes were identified via conserved catalytic domains associated with cellulose- and hemicellulose-degradation. Phylogenetic trees were created for each of these putative enzymes to depict genetic relatedness to known enzymes. Candidate genes were synthesized and cloned into plasmid expression vectors for inducible protein expression and secretion. The putative β-glucosidase fusion protein was efficiently secreted but did not permit Escherichia coli (E. coli) to use cellobiose as a sole carbon source, nor did the affinity purified enzyme cleave p-Nitrophenyl β-D-glucopyranoside (p-NPG) substrate in vitro over a range of physiological pH levels (pH 5-7). The putative hemicellulose-degrading β-xylosidase and α-L-arabinofuranosidase enzymes also lacked in vitro enzyme activity, but the affinity purified endo-1,4-β-xylanase protein cleaved a 6-chloro-4-methylumbelliferyl xylobioside substrate in acidic and neutral conditions, with maximal activity at pH 7. At this optimal pH, KM, Vmax, and kcat were determined to be 32.005 ± 4.72 μM, 1.16x10-5 ± 3.55x10-7 M/s, and 94.72 s-1, respectively. Thus, our pipeline enabled successful identification and characterization of a novel hemicellulose-degrading enzyme from the porcupine microbiome. Progress towards the goal of introducing a complete lignocellulose-degradation pathway into E. coli will be accelerated by combining synthetic metagenomic approaches with functional metagenomic library screening, which can identify novel enzymes unrelated to those found in available databases.

Fig 1. Identification of four genes from the porcupine microbiome with putative cellulose- and/or hemicellulose-degrading activity using a metagenomic sequencing pipeline.

The bioinformatic pipeline (A) began with Illumina MiSeq data previously collected from a porcupine fecal DNA sequencing project [14]. Reads were checked for quality and trimmed, concatemerized via MegaHit [15], and open reading frames were identified using Prodigal [16]. Protein sequences of interest were identified by pHMMER [17] using various protein databases and were selected for matches of interest based on e-value selection. (B) Top putative microbial enzymes identified by the metagenomic sequencing pipeline; putative signal sequences are shown in orange and predicted conserved protein domains are shown.

Fig 2. Four putative cellulose-/hemicellulose-degrading enzymes identified from the porcupine microbiome.

A) Functional description of putative enzymes with corresponding e-values. The e-value is a measure of confidence, with lower values denoting higher confidence. B) The catabolic pathway that converts cellobiose to glucose. C) The catabolic pathway that converts hemicellulose to xylose.

Fig 3. Phylogenetic analysis of putative enzymes.

Phylogenetic analysis of A) α-L-arabinofuranosidase, B) β-glucosidase, C) β-xylosidase, and D) Endo-1,4-β-xylanase were created from primary amino acid sequence that were aligned using ClustalW and generated using RAxML 7.2.8. Sequences were taken from highly related proteins from the BLASTP database. Trees were replicated 100 times for statistical strength with bootstrap values presented at each node. Putative enzymes are highlighted in red.

Fig 4. Secretion assay of putative enzymes.

Log-phase BL21(DE3) E. coli containing a putative enzyme or empty vector control were treated with 0.1 mM IPTG to induce T7-promoter-dependent transcription. After 3 hours, A) cells were lysed, B) cells were treated with cold osmotic shock to extract periplasmic protein fraction, or C) the supernatant was filtered, and protein was precipitated by addition of TCA. All protein fractions were prepared for SDS-PAGE and analyzed by western blot with an anti-His-antibody.

Fig 5. Purification and characterization of β-glucosidase.

A) Putative β-glucosidase and positive control β-glucosidase DesR (marked with red arrows) were affinity purified by 6xHis purification. B) Activity of the putative β-glucosidase against p-NPG at pH 5.0, 6.0 and 7.0; positive control β-glucosidase DesR was only tested at pH 7.0. C) Activity of putative β-glucosidase against p-NP-N-acetylglucosamine at pH 5.0, 6.0, and 7.0. Negative control β-glucosidase DesR was only tested at pH 7.0.

Fig 6. Characterization of endo-1,4- β-xylanase.

A) Putative endo-1,4-β-xylanase was purified by 6xHis purification. B) 0.6 μg of endo-1,4-β-xylanase was combined with 100 μM of CMU-X₂ to assess enzyme activity at different pH values. Samples were assessed via fluorescence in sodium citrate buffer (pH 4, 5, 5.5, 6) or sodium phosphate buffer (pH 6.5, 7, 7.5). A reading of raw fluorescent units (450 nm) was taken every minute for 30 minutes at 37°C. C) Relative activity of endo-1,4-β-xylanase at different pH levels compared to pH 7. D) Thermostability of endo-1,4-β-xylanase was assessed. Samples containing 0.6 μg of endo-1,4-β-xylanase in 50 mM sodium phosphate buffer were incubated for 30 minutes at 37°C, 50°C, 60°C, or 70°C before addition of 100 μM of CMU-X₂ and fluorescence detection. Control is unincubated endo-1,4-β-xylanase. E) Michaelis-Menten plot was generated with initial reaction rate against substrate concentrations above at 37°C, pH 7. F) Lineweaver-Burk plot was used to generate kinetic constants K_M, V_max, and k_cat.