Engineering of glycoside hydrolase family 7 cellobiohydrolases directed by natural diversity screening

Abstract

Protein engineering and screening of processive fungal cellobiohydrolases (CBHs) remain challenging due to limited expression hosts, synergy-dependency, and recalcitrant substrates. In particular, glycoside hydrolase family 7 (GH7) CBHs are critically important for the bioeconomy and typically difficult to engineer. Here, we target the discovery of highly active natural GH7 CBHs and engineering of variants with improved activity. Using experimentally assayed activities of genome mined CBHs, we applied sequence and structural alignments to top performers to identify key point mutations linked to improved activity. From ∼1500 known GH7 sequences, an evolutionarily diverse subset of 57 GH7 CBH genes was expressed in Trichoderma reesei and screened using a multiplexed activity screening assay. Ten catalytically enhanced natural variants were identified, produced, purified, and tested for efficacy using industrially relevant conditions and substrates. Three key amino acids in CBHs with performance comparable or superior to Penicillium funiculosum Cel7A were identified and combinatorially engineered into P. funiculosum cel7a, expressed in T. reesei, and assayed on lignocellulosic biomass. The top performer generated using this combined approach of natural diversity genome mining, experimental assays, and computational modeling produced a 41% increase in conversion extent over native P. funiculosum Cel7A, a 55% increase over the current industrial standard T. reesei Cel7A, and 10% improvement over Aspergillus oryzae Cel7C, the best natural GH7 CBH previously identified in our laboratory.

Keywords: cellobiohydrolase; cellulase; cellulose; enzyme improvement; genomics; protein engineering; rational design.

Figure 1

Cladogram of 1856 GH7 enzymes homologous to T. reesei Cel7A depicting known natural diversity and initial activity screen.A, all genes for which expression was attempted are highlighted with circles; green circles represent successfully expressed genes, blue circles represent genes that were successfully synthesized but failed to express, pink circles represent genes that failed to synthesize on one or two attempts, and black circles represent GH7 CBHs included as benchmarks. B, heat map for the activity of the 57 GH7 CBHs selected for further testing. The four columns represent four combinations of growth conditions (T. reesei host grown on glycerol, which suppresses background cellulase production, or lactose, which induces background cellulase production) and assayed substrate (PASC- phosphoric acid swollen cellulose, or solids from DMR- deacetylated mechanically refined corn stover). The data are sorted by the first column because this assay targets the intrinsic ability of the enzyme alone to depolymerize cellulose. Enzyme activities are represented as mg/ml glucose released in the reaction. Asterisks denote the enzymes selected for more detailed characterization. The full data, along with specific GH7 identification, are available in Data S3 and visualized alternatively in Fig. S14. CBHs, cellobiohydrolases; GH7, glycoside hydrolase family 7.

Figure 2

Activity screens of natural GH7 CBHs. Conversion of deacetylated mechanically refined corn stover solids for the three highest performing natural GH7 CBHs compared to well-characterized CBHs TrCel7A and PfCel7A. Each assay was performed in a trinary cocktail of cellobiohydrolase, purified Acidothermus cellulolyticus Cel5A endocellulase, and Aspergillus niger β-d-glucosidase at loadings of 28, 1.8, and 0.5 mg enzyme/g glucan, respectively. Solids loading was ∼1% and all assays were performed at pH 5.0 and 50 °C, except for that of Talaromyces aculeatus Cel7A, which was performed at 60 °C. Numerical data for this figure are provided in Table S3. Each assay was performed in biological triplicate, and error bars represent the standard deviation. The resulting extents of conversion are reported as the percent glucan converted. Statistical significance was determined via unpaired t test, indicating that the extent of conversion at 95 h is significantly different (p < 0.05) for AoCel7C and TtCel7A compared to PfCel7A (p values for both of <0.001), whereas TaCel7A (p = 0.379) was not. CBHs, cellobiohydrolases; GH7, glycoside hydrolase family 7.

Figure 3

Differences in GH7 CBH loops potentially relevant for enzyme performance.A, the catalytic domain of PfCel7A (PDB 4XEB) (20) on the surface of cellulose, with the A1 loop and “top loop” highlighted. The molecular configuration was generated via alignment of the PfCel7A catalytic domain with that of TrCel7A bound to an extracted chain on the surface of a cellulose microfibril that was generated in a previous molecular dynamics study (20). B, besides the variation in the length of the A1 loop, four instances of single residue variations near the binding tunnel entrance (left side) are shown. The tertiary structures of PfCel7A (PDB 4XEB) (20) and TrCel7A with cellononaose bound (PDB 4C4C) (37) are shown in orange and gray, respectively. Coordinates for the TtCel7A side chains shown are taken from a homology model generated via SWISS-MODEL (73). Cellulose binding subsite −7 (tunnel entrance) through −4 are shown according to the standard convention for carbohydrate binding sites (74). CBHs, cellobiohydrolases; GH7, glycoside hydrolase family 7; PDB, Protein Data Bank.

Figure 4

Activity screens of engineered GH7 CBHs and comparisons to PfCel7A.A, conversion of deacetylated mechanically refined corn stover solids for single, double, and triple mutants of PfCel7A. Each assay was performed in a trinary cocktail of cellobiohydrolase, purified Acidothermus cellulolyticus Cel5A endocellulase, and Aspergillus niger β-d-glucosidase at loadings of 28, 1.8, and 0.5 mg enzyme/g glucan, respectively. Solids loading was ∼1% and all assays were performed at pH 5.0 and 50 °C. Numerical data for this figure are provided in Table S5. Each assay was performed in biological triplicate, and error bars represent the standard deviation. The resulting extents of conversion are reported as the percent glucan converted. Statistical significance was determined via unpaired t test, indicating that the extent of conversion at 95 h is significantly different (p < 0.05) for all of the single, double, and triple mutants as compared to PfCel7A. B, relative conversion extent at 95 h for natural and engineered CBH variants compared to PfCel7A. The three most active natural GH7 CBHs from the current study, as well as single, double, and triple mutants of PfCel7A demonstrate comparable or superior performance to well-characterized PfCel7A. Numerical data for this figure are provided in Table S6. Asterisks overlaid on the bar graph indicate the statistical significance from unpaired t test; ‘ns’ indicates not significant (p > 0.05), ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001. CBHs, cellobiohydrolases; GH7, glycoside hydrolase family 7.