Cellulase Linkers Are Optimized Based on Domain Type and Function: Insights from Sequence Analysis, Biophysical Measurements, and Molecular Simulation

Abstract

Cellulase enzymes deconstruct cellulose to glucose, and are often comprised of glycosylated linkers connecting glycoside hydrolases (GHs) to carbohydrate-binding modules (CBMs). Although linker modifications can alter cellulase activity, the functional role of linkers beyond domain connectivity remains unknown. Here we investigate cellulase linkers connecting GH Family 6 or 7 catalytic domains to Family 1 or 2 CBMs, from both bacterial and eukaryotic cellulases to identify conserved characteristics potentially related to function. Sequence analysis suggests that the linker lengths between structured domains are optimized based on the GH domain and CBM type, such that linker length may be important for activity. Longer linkers are observed in eukaryotic GH Family 6 cellulases compared to GH Family 7 cellulases. Bacterial GH Family 6 cellulases are found with structured domains in either N to C terminal order, and similar linker lengths suggest there is no effect of domain order on length. O-glycosylation is uniformly distributed across linkers, suggesting that glycans are required along entire linker lengths for proteolysis protection and, as suggested by simulation, for extension. Sequence comparisons show that proline content for bacterial linkers is more than double that observed in eukaryotic linkers, but with fewer putative O-glycan sites, suggesting alternative methods for extension. Conversely, near linker termini where linkers connect to structured domains, O-glycosylation sites are observed less frequently, whereas glycines are more prevalent, suggesting the need for flexibility to achieve proper domain orientations. Putative N-glycosylation sites are quite rare in cellulase linkers, while an N-P motif, which strongly disfavors the attachment of N-glycans, is commonly observed. These results suggest that linkers exhibit features that are likely tailored for optimal function, despite possessing low sequence identity. This study suggests that cellulase linkers may exhibit function in enzyme action, and highlights the need for additional studies to elucidate cellulase linker functions.

Glycosylated linkers in multimodular lignocellulose-degrading enzymes dynamically bind to cellulose

“…We note that linkers in lignocellulose-degrading enzymes lack sequence conservation, which differs from some intrinsically disordered proteins that form secondary structure upon binding. However, despite this lack of sequence conservation, a recent study demonstrated that linker lengths, glycan positioning, and amino acid content are both conserved within a given GH family and are different between at least two GH families, namely GH6 and GH7 enzymes examined here (24). These results (24), when taken together with the present study, suggest that linker lengths and sequences may be optimized for a given binding affinity that is tuned based on the catalytic action of the overall enzyme.…”

“…However, despite this lack of sequence conservation, a recent study demonstrated that linker lengths, glycan positioning, and amino acid content are both conserved within a given GH family and are different between at least two GH families, namely GH6 and GH7 enzymes examined here (24). These results (24), when taken together with the present study, suggest that linker lengths and sequences may be optimized for a given binding affinity that is tuned based on the catalytic action of the overall enzyme. Nature has evolved multiple strategies for combining binding and catalytic function in single proteins and protein complexes for degradation of plant cell walls, and the results obtained here are primarily relevant for multimodular, lignocellulose-degrading enzymes that contain sufficiently long, disordered linkers to bind to cellulose and exhibit glycosylation.…”

“…It has long been thought that linkers in lignocellulose-degrading enzymes acted primarily as flexible connectors between subdomains and that the glycans on the linkers potentially impart rigidity, prevent proteolysis (28), or extend operating distances, the latter as predicted from simulation results (23,24). The primary alternative hypothesis regarding the role of linkers suggests that linkers are inchworm-like domains that dynamically store and dissipate energy associated with the catalytic step in, for example, processive hydrolysis of cellulose (20).…”

“…Taken together, the general consensus for linkers in lignocellulose-degrading enzymes is that they are flexible connectors between ordered, functional domains, with glycosylation to prevent proteolysis, but with little function beyond domain connection. However, linker modifications affect enzyme activity (14), and sequence analysis of linkers from GH Family 6 and 7 cellulases suggests that linkers between these two GH families exhibit significantly different average lengths (24). Even on this limited dataset, this finding suggests that these linkers may be optimized for specific function beyond domain connection.…”

“…Over the last decade, biophysical studies have suggested that linkers in lignocellulose-degrading enzymes are intrinsically disordered (20)(21)(22)(23)(24). There is little sequence conservation in linkers, which is a common feature of intrinsically disordered proteins, and they typically contain significant glycine, proline, serine, and threonine content.…”