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

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.

Engineering cellulases for conversion of lignocellulosic biomass

Lignocellulosic biomass is a renewable source of energy, chemicals and materials. Many applications of this resource require the depolymerization of one or more of its polymeric constituents. Efficient enzymatic depolymerization of cellulose to glucose by cellulases and accessory enzymes such as lytic polysaccharide monooxygenases is a prerequisite for economically viable exploitation of this biomass. Microbes produce a remarkably diverse range of cellulases, which consist of glycoside hydrolase (GH) catalytic domains and, although not in all cases, substrate-binding carbohydrate-binding modules (CBMs). As enzymes are a considerable cost factor, there is great interest in finding or engineering improved and robust cellulases, with higher activity and stability, easy expression, and minimal product inhibition. This review addresses relevant engineering targets for cellulases, discusses a few notable cellulase engineering studies of the past decades and provides an overview of recent work in the field.

Model of Processive Catalysis with Site Clustering and Blocking and Its Application to Cellulose Hydrolysis

Interactions between particles moving on a linear track and their possible blocking by obstacles can lead to crowding, impeding the particles' transport kinetics. When the particles are enzymes processively catalyzing a reaction along a linear polymeric substrate, these crowding and blocking effects may substantially reduce the overall catalytic rate. Cellulose hydrolysis by exocellulases processively moving along cellulose chains assembled into insoluble cellulose particles is an example of such a catalytic transport process. The details of the kinetics of cellulose hydrolysis and the causes of the often observed reduction of hydrolysis rate over time are not yet fully understood. Crowding and blocking of enzyme particles are thought to be one of the important factors affecting the cellulose hydrolysis, but its exact role and mechanism are not clear. Here, we introduce a simple model based on an elementary transport process that incorporates the crowding and blocking effects in a straightforward way. This is achieved by making a distinction between binding and non-binding sites on the chain. The model reproduces a range of experimental results, mainly related to the early phase of cellulose hydrolysis. Our results indicate that the combined effects of clustering of binding sites together with the occupancy pattern of these sites by the enzyme molecules play a decisive role in the overall kinetics of cellulose hydrolysis. It is suggested that periodic desorption and rebinding of enzyme molecules could be a basis of a strategy to partially counter the clustering of and blocking by the binding sites and so enhance the rate of cellulose hydrolysis. The general nature of the model means that it could be applicable also to other transport processes that make a distinction between binding and non-binding sites, where crowding and blocking are expected to be relevant.

Enzyme Synergy in Transient Clusters of Endo- and Exocellulase Enables a Multilayer Mode of Processive Depolymerization of Cellulose

Biological degradation of cellulosic materials relies on the molecular-mechanistic principle that internally chain-cleaving endocellulases work synergistically with chain end-cleaving exocellulases in polysaccharide chain depolymerization. How endo-exo synergy becomes effective in the deconstruction of a solid substrate that presents cellulose chains assembled into crystalline material is an open question of the mechanism, with immediate implications on the bioconversion efficiency of cellulases. Here, based on single-molecule evidence from real-time atomic force microscopy, we discover that endo- and exocellulases engage in the formation of transient clusters of typically three to four enzymes at the cellulose surface. The clusters form specifically at regular domains of crystalline cellulose microfibrils that feature molecular defects in the polysaccharide chain organization. The dynamics of cluster formation correlates with substrate degradation through a multilayer-processive mode of chain depolymerization, overall leading to the directed ablation of single microfibrils from the cellulose surface. Each multilayer-processive step involves the spatiotemporally coordinated and mechanistically concerted activity of the endo- and exocellulases in close proximity. Mechanistically, the cooperativity with the endocellulase enables the exocellulase to pass through its processive cycles ∼100-fold faster than when acting alone. Our results suggest an advanced paradigm of efficient multienzymatic degradation of structurally organized polymer materials by endo-exo synergetic chain depolymerization.

Heterologous production of cellulose- and starch-degrading hydrolases to expand Saccharomyces cerevisiae substrate utilization: Lessons learnt

Selected strains of Saccharomyces cerevisiae are used for commercial bioethanol production from cellulose and starch, but the high cost of exogenous enzymes for substrate hydrolysis remains a challenge. This can be addressed through consolidated bioprocessing (CBP) where S. cerevisiae strains are engineered to express recombinant glycoside hydrolases during fermentation. Looking back at numerous strategies undertaken over the past four decades to improve recombinant protein production in S. cerevisiae, it is evident that various steps in the protein production "pipeline" can be manipulated depending on the protein of interest and its anticipated application. In this review, we briefly introduce some of the strategies and highlight lessons learned with regards to improved transcription, translation, post-translational modification and protein secretion of heterologous hydrolases. We examine how host strain selection and modification, as well as enzyme compatibility, are crucial determinants for overall success. Finally, we discuss how lessons from heterologous hydrolase expression can inform modern synthetic biology and genome editing tools to provide process-ready yeast strains in future. However, it is clear that the successful expression of any particular enzyme is still unpredictable and requires a trial-and-error approach.

Machine learning reveals sequence-function relationships in family 7 glycoside hydrolases

Family 7 glycoside hydrolases (GH7) are among the principal enzymes for cellulose degradation in nature and industrially. These enzymes are often bimodular, including a catalytic domain and carbohydrate-binding module (CBM) attached via a flexible linker, and exhibit an active site that binds cello-oligomers of up to ten glucosyl moieties. GH7 cellulases consist of two major subtypes: cellobiohydrolases (CBH) and endoglucanases (EG). Despite the critical importance of GH7 enzymes, there remain gaps in our understanding of how GH7 sequence and structure relate to function. Here, we employed machine learning to gain data-driven insights into relationships between sequence, structure, and function across the GH7 family. Machine-learning models, trained only on the number of residues in the active-site loops as features, were able to discriminate GH7 CBHs and EGs with up to 99% accuracy, demonstrating that the lengths of loops A4, B2, B3, and B4 strongly correlate with functional subtype across the GH7 family. Classification rules were derived such that specific residues at 42 different sequence positions each predicted the functional subtype with accuracies surpassing 87%. A random forest model trained on residues at 19 positions in the catalytic domain predicted the presence of a CBM with 89.5% accuracy. Our machine learning results recapitulate, as top-performing features, a substantial number of the sequence positions determined by previous experimental studies to play vital roles in GH7 activity. We surmise that the yet-to-be-explored sequence positions among the top-performing features also contribute to GH7 functional variation and may be exploited to understand and manipulate function.

Enhancing the catalytic activity of a GH5 processive endoglucanase from Bacillus subtilis BS-5 by site-directed mutagenesis

Processive endoglucanases possess both endo- and exoglucanase activity, making them attractive discovery and engineering targets. Here, a processive endoglucanase EG5C-1 from Bacillus subtilis was employed as the starting point for enzyme engineering. Referring to the complex structure information of EG5C-1 and cellohexaose, the amino acid residues in the active site architecture were identified and subjected to alanine scanning mutagenesis. The residues were chosen for a saturation mutagenesis since their variants showed similar activities to EG5C-1. Variants D70Q and S235W showed increased activity towards the substrates CMC and Avicel, an increase was further enhanced in D70Q/S235W double mutant, which displayed a 2.1- and 1.7-fold improvement in the hydrolytic activity towards CMC and Avicel, respectively. In addition, kinetic measurements showed that double mutant had higher substrate affinity (K_m) and a significantly higher catalytic efficiency (k_cat/K_m). The binding isotherms of wild-type EG5C-1 and double mutant D70Q/S235W suggested that the binding capability of EG5C-1 for the insoluble substrate was weaker than that of D70Q/S235W. Molecular dynamics simulations suggested that the collaborative substitutions of D70Q and S235W altered the hydrogen bonding network within the active site architecture and introduced new hydrogen bonds between the enzyme and cellohexaose, thus enhancing both substrate affinity and catalytic efficiency.

Functional analysis of chimeric TrCel6A enzymes with different carbohydrate binding modules

The glycoside hydrolase (GH) family 6 is an important group of enzymes that constitute an essential part of industrial enzyme cocktails used to convert lignocellulose into fermentable sugars. In nature, enzymes from this family often have a carbohydrate binding module (CBM) from the CBM family 1. These modules are known to promote adsorption to the cellulose surface and influence enzymatic activity. Here, we have investigated the functional diversity of CBMs found within the GH6 family. This was done by constructing five chimeric enzymes based on the model enzyme, TrCel6A, from the soft-rot fungus Trichoderma reesei. The natural CBM of this enzyme was exchanged with CBMs from other GH6 enzymes originating from different cellulose degrading fungi. The chimeric enzymes were expressed in the same host and investigated in adsorption and quasi-steady-state kinetic experiments. Our results quantified functional differences of these phylogenetically distant binding modules. Thus, the partitioning coefficient for substrate binding varied 4-fold, while the maximal turnover (kcat) showed a 2-fold difference. The wild-type enzyme showed the highest cellulose affinity on all tested substrates and the highest catalytic turnover. The CBM from Serendipita indica strongly promoted the enzyme's ability to form productive complexes with sites on the substrate surface but showed lower turnover of the complex. We conclude that the CBM plays an important role for the functional differences between GH6 wild-type enzymes.

pH profiles of cellulases depend on the substrate and architecture of the binding region

Understanding the pH effect of cellulolytic enzymes is of great technological importance. In this study, we have examined the influence of pH on activity and stability for central cellulases (Cel7A, Cel7B, Cel6A from Trichoderma reesei, and Cel7A from Rasamsonia emersonii). We systematically changed pH from 2 to 7, temperature from 20°C to 70°C, and used both soluble (4-nitrophenyl β- d-lactopyranoside [pNPL]) and insoluble (Avicel) substrates at different concentrations. Collective interpretation of these data provided new insights. An unusual tolerance to acidic conditions was observed for both investigated Cel7As, but only on real insoluble cellulose. In contrast, pH profiles on pNPL were bell-shaped with a strong loss of activity both above and below the optimal pH for all four enzymes. On a practical level, these observations call for the caution of the common practice of using soluble substrates for the general characterization of pH effects on cellulase activity. Kinetic modeling of the experimental data suggested that the nucleophile of Cel7A experiences a strong downward shift in pK_a upon complexation with an insoluble substrate. This shift was less pronounced for Cel7B, Cel6A, and for Cel7A acting on the soluble substrate, and we hypothesize that these differences are related to the accessibility of water to the binding region of the Michaelis complex.

The dissociation mechanism of processive cellulases

Cellulase enzymes deconstruct recalcitrant cellulose into soluble sugars, making them a biocatalyst of biotechnological interest for use in the nascent lignocellulosic bioeconomy. Cellobiohydrolases (CBHs) are cellulases capable of liberating many sugar molecules in a processive manner without dissociating from the substrate. Within the complete processive cycle of CBHs, dissociation from the cellulose substrate is rate limiting, but the molecular mechanism of this step is unknown. Here, we present a direct comparison of potential molecular mechanisms for dissociation via Hamiltonian replica exchange molecular dynamics of the model fungal CBH, Trichoderma reesei Cel7A. Computational rate estimates indicate that stepwise cellulose dethreading from the binding tunnel is 4 orders of magnitude faster than a clamshell mechanism, in which the substrate-enclosing loops open and release the substrate without reversing. We also present the crystal structure of a disulfide variant that covalently links substrate-enclosing loops on either side of the substrate-binding tunnel, which constitutes a CBH that can only dissociate via stepwise dethreading. Biochemical measurements indicate that this variant has a dissociation rate constant essentially equivalent to the wild type, implying that dethreading is likely the predominant mechanism for dissociation.

Modeling the activity burst in the initial phase of cellulose hydrolysis by the processive cellobiohydrolase Cel7A

The hydrolysis of cellulose by processive cellulases, such as exocellulase TrCel7A from Trichoderma reesei, is typically characterized by an initial burst of high activity followed by a slowdown, often leading to incomplete hydrolysis of the substrate. The origins of these limitations to cellulose hydrolysis are not yet fully understood. Here, we propose a new model for the initial phase of cellulose hydrolysis by processive cellulases, incorporating a bound but inactive enzyme state. The model, based on ordinary differential equations, accurately reproduces the activity burst and the subsequent slowdown of the cellulose hydrolysis and describes the experimental data equally well or better than the previously suggested model. We also derive steady‐state expressions that can be used to describe the pseudo‐steady state reached after the initial activity burst. Importantly, we show that the new model predicts the existence of an optimal enzyme‐substrate affinity at which the pseudo‐steady state hydrolysis rate is maximized. The model further allows the calculation of glucose production rate from the first cut in the processive run and reproduces the second activity burst commonly observed upon new enzyme addition. These results are expected to be applicable also to other processive enzymes.

Predominant Nonproductive Substrate Binding by Fungal Cellobiohydrolase I and Implications for Activity Improvement

Enzymatic conversion of the most abundant renewable source of organic compounds, cellulose to fermentable sugars is attractive for production of green fuels and chemicals. The major component of industrial enzyme systems, cellobiohydrolase I from Hypocrea jecorina (Trichoderma reesei) (HjCel7A) processively splits disaccharide units from the reducing ends of tightly packed cellulose chains. HjCel7A consists of a catalytic domain (CD) and a carbohydrate-binding module (CBM) separated by a linker peptide. A tunnel-shaped substrate-binding site in the CD includes nine subsites for β-d-glucose units, seven of which (-7 to -1) precede the catalytic center. Low catalytic activity of Cel7A is the bottleneck and the primary target for improvement. Here it is shown for the first time that, in spite of much lower apparent k_cat of HjCel7A at the hydrolysis of β-1,4-glucosidic linkages in the fluorogenic cellotetra- and -pentaose compared to the structurally related endoglucanase I (HjCel7B), the specificity constants (catalytic efficiency) k_cat /K_m for both enzymes are almost equal in these reactions. The observed activity difference appears from strong nonproductive substrate binding by HjCel7A, particularly significant for MU-β-cellotetraose (MUG₄ ). Interaction of substrates with the subsites -6 and -5 proximal to the nonconserved Gln101 residue in HjCel7A decreases K_m,ap by >1500 times. HjCel7A can be nonproductively bound onto cellulose surface with K_d ≈2-9 nM via CBM and CD that captures six terminal glucose units of cellulose chain. Decomposition of this nonproductive complex can determine the rate of cellulose conversion. MUG₄ is a promising substrate to select active cellobiohydrolase I variants with reduced nonproductive substrate binding.

Engineering enhanced cellobiohydrolase activity

Glycoside Hydrolase Family 7 cellobiohydrolases (GH7 CBHs) catalyze cellulose depolymerization in cellulolytic eukaryotes, making them key discovery and engineering targets. However, there remains a lack of robust structure–activity relationships for these industrially important cellulases. Here, we compare CBHs from Trichoderma reesei (TrCel7A) and Penicillium funiculosum (PfCel7A), which exhibit a multi-modular architecture consisting of catalytic domain (CD), carbohydrate-binding module, and linker. We show that PfCel7A exhibits 60% greater performance on biomass than TrCel7A. To understand the contribution of each domain to this improvement, we measure enzymatic activity for a library of CBH chimeras with swapped subdomains, demonstrating that the enhancement is mainly caused by PfCel7A CD. We solve the crystal structure of PfCel7A CD and use this information to create a second library of TrCel7A CD mutants, identifying a TrCel7A double mutant with near-equivalent activity to wild-type PfCel7A. Overall, these results reveal CBH regions that enable targeted activity improvements.

Cellobiohydrolases (CBHs) are critical for natural and industrial biomass degradation but their structure–activity relationships are not fully understood. Here, the authors present the biochemical and structural characterization of two CBHs, identifying protein regions that confer enhanced CBH activity.

Thermoactivation of a cellobiohydrolase

We have measured activity and substrate affinity of the thermostable cellobiohydrolase, Cel7A, from Rasamsonia emersonii over a broad range of temperatures. For the wild type enzyme, which does not have a Carbohydrate Binding Module (CBM), higher temperature only led to moderately increased activity against cellulose, and we ascribed this to a pronounced, temperature induced desorption of enzyme from the substrate surface. We also tested a "high affinity" variant of R. emersonii Cel7A with a linker and CBM from a related enzyme. At room temperature, the activity of the variant was similar to the wild type, but the variant was more accelerated by temperature and about two-fold faster around 70 °C. This better thermoactivation of the high-affinity variant could not be linked to differences in stability or the catalytic process, but coincided with less desorption as temperature increased. Based on these observations and earlier reports on moderate thermoactivation of cellulases, we suggest that better cellulolytic activity at industrially relevant temperatures may be attained by engineering improved substrate affinity into enzymes that already possess good thermostability.

Correlation of structure, function and protein dynamics in GH7 cellobiohydrolases from Trichoderma atroviride, T. reesei and T. harzianum

BACKGROUND

The ascomycete fungus Trichoderma reesei is the predominant source of enzymes for industrial conversion of lignocellulose. Its glycoside hydrolase family 7 cellobiohydrolase (GH7 CBH) TreCel7A constitutes nearly half of the enzyme cocktail by weight and is the major workhorse in the cellulose hydrolysis process. The orthologs from Trichoderma atroviride (TatCel7A) and Trichoderma harzianum (ThaCel7A) show high sequence identity with TreCel7A, ~ 80%, and represent naturally evolved combinations of cellulose-binding tunnel-enclosing loop motifs, which have been suggested to influence intrinsic cellobiohydrolase properties, such as endo-initiation, processivity, and off-rate.

RESULTS

The TatCel7A, ThaCel7A, and TreCel7A enzymes were characterized for comparison of function. The catalytic domain of TatCel7A was crystallized, and two structures were determined: without ligand and with thio-cellotriose in the active site. Initial hydrolysis of bacterial cellulose was faster with TatCel7A than either ThaCel7A or TreCel7A. In synergistic saccharification of pretreated corn stover, both TatCel7A and ThaCel7A were more efficient than TreCel7A, although TatCel7A was more sensitive to thermal inactivation. Structural analyses and molecular dynamics (MD) simulations were performed to elucidate important structure/function correlations. Moreover, reverse conservation analysis (RCA) of sequence diversity revealed divergent regions of interest located outside the cellulose-binding tunnel of Trichoderma spp. GH7 CBHs.

CONCLUSIONS

We hypothesize that the combination of loop motifs is the main determinant for the observed differences in Cel7A activity on cellulosic substrates. Fine-tuning of the loop flexibility appears to be an important evolutionary target in Trichoderma spp., a conclusion supported by the RCA data. Our results indicate that, for industrial use, it would be beneficial to combine loop motifs from TatCel7A with the thermostability features of TreCel7A. Furthermore, one region implicated in thermal unfolding is suggested as a primary target for protein engineering.