research papers
Synergic action of an inserted carbohydrate-binding module in a glycoside hydrolase family 5 endoglucanase
aInstitute of Biological Chemistry, Academia Sinica, Taipei 115, Taiwan, bDepartment of Chemistry, National Taiwan University, Taipei 115, Taiwan, and cInstitute of Biochemical Sciences, National Taiwan University, Taipei 115, Taiwan
*Correspondence e-mail: kotping@gate.sinica.edu.tw, shwu@gate.sinica.edu.tw
Most known cellulase-associated carbohydrate-binding modules (CBMs) are attached to the N- or C-terminus of the enzyme or are expressed separately and assembled into multi-enzyme complexes (for example to form cellulosomes), rather than being an insertion into the MtGlu5 from Meiothermus taiwanensis WR-220, a GH5-family endo-β-1,4-glucanase (EC 3.2.1.4), has a bipartite architecture consisting of a Cel5A-like with a (β/α)8 TIM-barrel fold and an inserted CBM29-like noncatalytic domain with a β-jelly-roll fold. Deletion of the CBM significantly reduced the catalytic efficiency of MtGlu5, as determined by isothermal titration using inactive mutants of full-length and CBM-deleted MtGlu5 proteins. Conversely, insertion of the CBM from MtGlu5 into TmCel5A from Thermotoga maritima greatly enhanced the substrate affinity of TmCel5A. Bound sugars observed between two tryptophan side chains in the catalytic domains of active full-length and CBM-deleted MtGlu5 suggest an important stacking force. The synergistic action of the and CBM of MtGlu5 in binding to single-chain was visualized by substrate modeling, in which additional surface tryptophan residues were identified in a cross-domain groove. Subsequent site-specific mutagenesis results confirmed the pivotal role of several other tryptophan residues from both domains of MtGlu5 in substrate binding. These findings reveal a way to incorporate a CBM into the of an existing enzyme to make a robust cellulase.
Here, by solving the it is shown thatKeywords: cellulases; cross-domain substrate binding; CBM insertion; MtGlu5; Meiothermus taiwanensis WR-220.
1. Introduction
Plant cell walls (PCWs) are composed of et al., 2019; Himmel et al., 2007; Ragauskas et al., 2006; Coughlan, 1985). The most abundant polysaccharide in PCWs, cellulose, consists of β-1,4-linked glucose, which can be degraded into mono/disaccharides before to bioethanol (Rubin, 2008; Demain et al., 2005). Deconstruction processes of PCWs are generally mediated by glycoside (GHs), which have been classified into different families in the carbohydrate-active enzymes (CAZy) database (Drula et al., 2022; Lombard et al., 2014). GHs that catalyse the hydrolysis of in PCWs, including cellulose and hemicellulose, also have industrial importance due to their wide applications (Gilbert, 2010; Himmel et al., 2007; Kuhad et al., 2011). Cellulases, which cleave the β-glycosidic bonds in cellulose, are mainly categorized into exo-glucanases, endo-glucanases and β-glucosidases. Exo- and endo-glucanases work together in a synergistic process in which exo-glucanases turn out cellobiose units from the end of crystalline regions of cellulose, while endo-glucanases work on amorphous regions of cellulose and create new ends for the exo-enzymes (Jalak et al., 2012; Sakon et al., 1997; Teeri, 1997).
and serve as an important resource for green materials and energy. Polysaccharide decomposition and utilization play a crucial role in nature, promoting progression of the carbon cycle. In industry, a limited supply of petroleum has increased the need for renewable energy sources such as biofuel products from PCWs (KaoThe et al., 2012; Sakon et al., 1997; Davies & Henrissat, 1995; Breyer & Matthews, 2001). On the other hand, noncatalytic carbohydrate-binding modules (CBMs) are found in many CAZymes and can provide specific interactions to help CAZymes act towards insoluble cellulose substrates (for example PCWs) in nature (Novy et al., 2019; Sidar et al., 2020; Hashimoto, 2006; Gilbert et al., 2013; Hervé et al., 2010; Boraston et al., 2004; Tomme et al., 1988). CBMs are grouped into three types based on their binding modes. Type A CBMs recognize and disrupt the surface of crystalline cellulose, type B CBMs bind individual polysaccharide chains in amorphous cellulose and type C CBMs have a lectin-like property that binds mono-, di- or trisaccharides (Gilbert et al., 2013; Boraston et al., 2004). CBMs commonly attach directly to CAZymes at their N- or C-terminal ends (Sidar et al., 2020; Ravachol et al., 2014; Urbanowicz et al., 2007) or are assembled into cellulosome complexes (Eibinger et al., 2020; Bayer et al., 1998; Schwarz, 2001), thereby enhancing the enzyme–substrate interactions (Gilbert et al., 2013; Venditto et al., 2016; Hervé et al., 2010; Luís et al., 2013).
(CD) of cellulases contains the active site for hydrolysis, and the mode of action of the CD is based on its three-dimensional structure. For example, exo-cellulases have tunnel-like CD structures that only bind to the termini of cellulose molecules, whereas endo-cellulases usually have groove or cleft CD structures that bind to amorphous regions of cellulose (HornMeiothermus taiwanensis WR-220, a thermophilic, heterotrophic, aerobic Gram-negative bacterium isolated from Wu-rai hot springs in north Taiwan, has an optimum growth temperature of approximately 55°C (Chen et al., 2002). Its importance is underscored by its industrial and biomedical applications, and it therefore has great potential for further exploration (Wu et al., 2017; Liao et al., 2013; Su et al., 2016; Lin et al., 2016). Several glycoside were found in M. taiwanensis WR-220 by whole-genome sequencing, among which is a newly identified endo-β-1,4-glucanase. This enzyme, named MtGlu5, belongs to subfamily 25 of glycoside hydrolase family 5 (GH5_25). GH5, which is also known as cellulase family A, is one of the largest GH families that hydrolyse major polysaccharide components in the biosphere (Aspeborg et al., 2012; Henrissat et al., 1989). MtGlu5 contains a Cel5A-like CD similar to TmCel5A from Thermotoga maritima (Pereira et al., 2010; Wu et al., 2011), but has an additional novel domain inserted into an internal site in the CD. This novel domain, which was supposed to be a CBM and was thus named MtCBM, shares low sequence identity with other known CBMs. Here, we determined crystal structures of full-length MtGlu5 and CBM-deleted MtGlu5 (dubbed ΔCBM) in apo and sugar-bound forms. We also characterized the biophysical properties of MtGlu5 and investigated the effects of the presence or absence of the inserted CBM on the function of the CD. Furthermore, we inserted the CBM into TmCel5A to make a functional chimeric enzyme. Most importantly, this type of inserted CBM in an intact CD of a cellulase, although found in some xylanases (Flint et al., 1997; Wu et al., 2021), had never been investigated in detail before. Our structural and functional study of MtGlu5 provides insight into a possible mode of action of the inserted CBM domain that could serve as a paradigm for thus far uncharacterized endo-cellulases with inserted CBMs from other microorganisms.
2. Methods
2.1. Cloning, expression and purification
The DNA fragment encoding residues 15–470 of MtGlu5 (from which the signal peptide was removed) was amplified from the chromosomal DNA of M. taiwanensis WR-220 by PCR. The plasmids for ΔCBM (residues 1–235 and 366–455 of MtGlu5), for site-directed mutagenesis of MtGlu5 and for GST-fused gMtCBM (residues 236–365 of MtGlu5) were constructed using the MtGlu5 gene as a template (Zeng, 1998). The gene fragment encoding TmCel5A was synthesized chemically, and the plasmid for TmCel5A-CBM expression was constructed using the TmCel5A and MtGlu5 genes. The primers used are listed in Supplementary Table S1. All of the resultant DNA fragments were cloned into pET-21 vector with a C-terminal His6 tag. The recombinant plasmids were transformed into Escherichia coli BL21 (DE3) cells for protein expression. Inoculated competent cells were cultured in TB medium containing 50 µg ml−1 carbenicillin until the OD600 reached 1.0. Protein expression was then induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG; 1 mM final concentration) and the cells were further incubated at 37°C for 6 h. All recombinant proteins were purified by immobilized metal ion-affinity in lysis buffer (20 mM Tris, 100 mM sodium chloride, 20 mM imidazole pH 8). The elution used a 20–200 mM gradient of imidazole in the same buffer. Further purification by FPLC for crystallization and assays used Superdex 75 columns (GE Healthcare).
2.2. Crystallization and data collection
The inactive E393Q mutant, dubbed iMtGlu5, was crystallized in the apo form by sitting-drop vapour diffusion by mixing equal volumes (1 µl) of protein solution (20 mg ml−1 in 20 mM Tris–HCl, 100 mM NaCl pH 8.0) and reservoir solution (0.1 M citric acid, 0.7 M ammonium sulfate, 20% PEG 400 pH 5.0) at room temperature. A sugar-bound MtGlu5–glucose, was obtained in an MRC plate (Hampton Research, catalogue No. HR3-104) by the microbatch method (Chayen, 1997) using equal volumes (1 µl) of protein solution (30 mg ml−1 with a small amount of carboxymethyl cellulose, CMC) and reservoir solution (0.16 M citric acid, 1.1 M ammonium sulfate, 10% PEG 400 pH 5.0) under 10 µl Al's oil (a 1:1 mixture of paraffin and silicone oil; Hampton Research, catalogue No. HR3-413). Apo-form CBM-deleted MtGlu5, dubbed ΔCBM, was crystallized in an Eppendorf tube with a very high protein concentration (approximately 1500 µM in 20 mM Tris, 100 mM NaCl pH 8.0) over a very long time (over a year) by evaporation. Sugar-bound crystals of ΔCBM were obtained by soaking ΔCBM crystals in a solution containing 10 mM cellobiose or cellopentaose. Glycerol was added as a cryoprotectant for all crystals prior to flash-cooling. The X-ray diffraction data sets were collected on beamlines 05A1, 15A1 and 13B1 at the National Synchrotron Radiation Research Center (NSRRC), Taiwan at 1.0 Å wavelength. All data sets were processed with DENZO in HKL-2000 (Otwinowski & Minor, 1997).
2.3. and refinement
The MtGlu5 was solved by the molecular-replacement method using MOLREP (Vagin & Teplyakov, 2010) with the TmCel5A structure (PDB entry 3mmu; Pereira et al., 2010) as a search model; the other structures were solved using the well refined iMtGlu5 structure. The resulting model was subjected to computational with REFMAC5 (Murshudov et al., 2011). Well ordered solvent molecules, sugar ligands and water molecules were located with Coot (Emsley et al., 2010). The stereochemical quality of the refined models was checked with MolProbity (Williams et al., 2018). Final refinements were carried out by Phenix (Liebschner et al., 2019) and some statistics are listed in Table 1. Molecular figures were produced with PyMOL (Schrödinger, USA). The coordinates and structure factors have been deposited in the Protein Data Bank with accession codes 7vt4, 7vt5, 7vt6, 7vt7 and 7vt8.
of i
|
2.4. Isothermal titration calorimetry
The binding constant was determined using a Microcal iTC200 calorimeter (Malvern Panalytical) in Tris buffer (20 mM Tris, 100 mM NaCl pH 8.0). The titration experiments were performed by injecting 2 µl aliquots of 50 µM CMC into a sample cell containing protein samples at 25–100 µM at 25°C. The stirring speed and reference power were set to 750 rev min−1 and 5 µcal s−1, respectively. The reference heat background was determined under the same conditions by injecting CMC into buffer without protein. Thermodynamic parameters were calculated using ΔG = −RTlnKa = ΔH − TΔS, and data analysis was performed by the MicroCal PEAQ-ITC software version 1.21.
2.5. assay
The reducing sugars produced from enzyme catalysis were determined using 3,5-dinitrosalicylic acid (DNS) reagent (Miller, 1959). The enzyme solution was mixed with carboxymethyl cellulose (CMC, 1%) and incubated for the given periods. The reactions were stopped by adding three times the volume of DNS reagent and boiling for 10 min. The absorbance at 540 nm was measured using a UV–Vis spectrophotometer (Infinite M1000 PRO, Tecan). The optimal conditions for MtGlu5 and ΔCBM were determined at a range of pH values and temperatures in CHC buffer (20 mM citrate/HEPES/CHES, 100 mM NaCl). assays on insoluble RAC substrates, which were produced from Avicel (Zhang, 2006), were performed under the optimal conditions for 30 min and then heated at 100°C for a further 10 min to stop the reaction, followed by adding three times the volume of DNS reagent and boiling for 10 min. After centrifugation to remove the pellet, the products were measured at an absorbance of 540 nm using a UV–Vis spectrophotometer. Glucose was used as a standard to produce the The activity was initially calculated as in IU, which is defined as micromoles of product per minute per micromole of protein, and presented as relative activity (the percentage of the activity of the wild-type MtGlu5 enzyme) for easy comparison with the wild type.
The kinetic parameters of MtGlu5 and its mutants towards the CMC substrate were obtained by fitting the initial reaction velocities at various substrate concentrations (which were varied from approximately 0.2 to 10 times the Km of wild-type MtGlu5) using the Michaelis–Menten equation in GraphPad 6.0. Each reaction was run at least in triplicate with seven substrate concentrations under the optimal conditions. The enzyme-kinetics assay was conducted according to previously described procedures (Liang et al., 2018).
2.6. Substrate-binding model
The model was created according to superimposition of MtGlu5 on TmCel5A in complex with cellotetraose and on CBM29-2 in complex with cellohexaose. Based on the cellotetraose and cellohexaose models, a polysaccharide chain composed of 14 glucose units was manually inserted into the MtGlu5 structure. The model was then optimized by energy minimization.
2.7. Thermal shift assay
The binding of iMtGlu5 and other mutants to CMC was verified by the intrinsic tryptophan and tyrosine fluorescence in a thermal shift assay using a Tycho NT.6 (NanoTemper Technologies) label-free differential scanning fluorimeter (Sierla et al., 2018). The thermal stability of the inactive protein samples was measured in Tris buffer (20 mM Tris, 100 mM NaCl pH 8.0) with and without CMC. Aromatic residues buried in the protein core were exposed to the buffer upon temperature increase and protein unfolding. The shift of the 350/330 nm ratio was monitored in a quick thermal ramp from 35 to 95°C. The inflection point (Ti) of protein unfolding was then determined.
3. Results and discussion
3.1. The of MtGlu5 shows an inserted CBM29-like domain
Analysis of the MtGlu5 sequence by BLASTp on the NCBI website (https://www.ncbi.nlm.nih.gov/) indicated that MtGlu5 belongs to the GH5 family but shares low overall identity with other cellulases in this family. However, using the multiple sequence alignment method of ClustalW (Thompson et al., 1994), a unique inserted domain was identified in the middle of MtGlu5 (Fig. 1a). Further analysis of this domain alone with BLASTp revealed that it is probably a CBM. A model of the CD region was successfully predicted by I-TASSER (https://zhanggroup.org/I-TASSER/; Roy et al., 2010) using TmCel5A (PDB entry 3mmu; Pereira et al., 2010), which has approximately 50% identity to MtGlu5, as a template, but the query coverage was only 64% and structure prediction failed for the new inserted domain. The model of the CD showed that Glu149 and Glu393 are the highly conserved general acid/base and residues in the GH5 family. Initial attempts to crystallize wild-type MtGlu5 failed. Instead, the inactive mutant E393Q (dubbed iMtGlu5) was first crystallized in a suitable form for The was solved by the molecular-replacement approach using TmCel5A (PDB entry 3mmu) as a search model. The CDs of two iMtGlu5 molecules were correctly located in the monoclinic P21 The initial map showed some density outside the CD regions, and amino-acid residues corresponding to the CBM were manually placed, followed by computational and new map calculations. In this way, the model was gradually improved (Table 1).
Subsequently, sugar-bound crystals of wild-type MtGlu5 were obtained by substrate co-crystallization with carboxymethyl cellulose (CMC). On analysis of the cell content, the Matthews coefficient indicated that there might also be two MtGlu5 molecules in the of the tetragonal I422 crystal. However, it turned out to contain only one protein molecule in the with a high solvent content of ∼70%. In the refined model of MtGlu5, the conserved CD has a (β/α)8 TIM-barrel fold and the inserted CBM shows a β-jelly-roll fold (Fig. 1b, left). Bound glucose was observed in the MtGlu5 crystal, whereas no sugar was seen in the iMtGlu5 crystal (Fig. 1b, right). Presumably, some CMC in the crystallization solution had been hydrolyzed by the active wild-type enzyme. Glucose was also found to be a product of CMC hydrolysis by TmCel5A (Basit & Akhtar, 2018). In the catalytic pocket, the glucose molecule is sandwiched between the indole side chains of Trp43 and Trp224, which are presumably engaged in stacking interactions. The binding site is adjacent to the two conserved catalytic residues Glu149 and Glu393 (Fig. 1c). It corresponds to the −2 or −3 sugar of a bound cellulose substrate, according to the general nomenclature. The other sugar units were not observed because they were considered to be leaving products and were not properly retained in the active site (Wu et al., 2011; Davies et al., 1997).
To find out the possible classification of MtCBM, DALI was used to search for similar structures (Holm & Sander, 1993; Holm, 2020). The search results showed that MtCBM is topologically similar to CBM29-2, which belongs to the type B CBMs (Charnock et al., 2002). Sequence alignment of MtCBM and CBM29-2 revealed several conserved residues, including three surface aromatic amino acids Trp249, Trp251 and Trp270 in MtCBM, which correspond to Trp24, Trp26 and Tyr46 in CBM29-2, respectively. Both the MtCBM and the CBM29-2 structures show a β-jelly-roll topology, and they superimpose quite well with a root-mean-square deviation (r.m.s.d.) of 2.73 Å for 142 pairs of Cα atoms (Figs. 1d–1f). The CBM29 family was first identified in Piromyces equi in 2001 (Freelove et al., 2001). However, no other protein from a microorganism has shown homology to CBM29s in the CAZy database until now. According to the information on structural homology provided by the DALI server, the three surface tryptophan residues of MtCBM are considered to play a key role in substrate binding, as are the corresponding aromatic residues in CBM29-2 (Charnock et al., 2002; Flint et al., 2004, 2005). Additionally, other non-aromatic sugar-binding residues in CBM29-2, such as Gln116, also have equivalents (Gln337) in MtCBM, which are presumably engaged in similar interactions. The above structural analysis suggests a close kinship of MtCBM to the CBM29 family. The precise placement of MtCBM among the other known CBMs, however, awaits its classification by CAZy.
3.2. The presence of the MtCBM domain greatly enhances the substrate affinity
To characterize the function of the presumed MtCBM (amino acids 236–365 of MtGlu5), binding constants were determined by isothermal titration (ITC). Inactive full-length MtGlu5 (iMtGlu5), inactive CBM-deleted MtGlu5 (iΔCBM) and GST-fused MtCBM (gMtCBM) were constructed, expressed and purified for ITC experiments. The inactive proteins carried the active-site mutations E393Q in iMtGlu5 and E263Q in iΔCBM to avoid interference in the ITC measurements. Despite numerous trials to purify MtCBM alone under various conditions, the protein stability was too low for any further experiments. Therefore, a GST-fusion protein, dubbed gMtCBM, was designed and the solubility problem was successfully solved. GST alone was also expressed and purified as a negative control. The binding profiles and calculated parameters for the four different proteins are shown in Figs. 2(a)–2(d) and Table 2. All data were fitted with a single-site binding model. Because the CMC titrant and its molar concentration of binding sites are unknown, the n value was modified to 1 together with adjusting the concentration of the titrant (Szabó et al., 2001; Carvalho et al., 2004; Campos et al., 2016). The data show that iMtGlu5 has a strong affinity for the recognition of long single polysaccharide chains with Kd = 5.71 × 10−6 (Fig. 2a). However, the data for iΔCBM showed no significant binding to the CMC substrate, confirming that the deleted domain is responsible for substrate binding (Fig. 2b). Moreover, the data for gMtCBM also demonstrated significant affinity, with Kd = 4.56 × 10−2, while in contrast GST displayed no binding (Figs. 2c and 2d). The calculated thermodynamic parameters indicated that the binding between MtCBM and the polysaccharide chain was enthalpically favorable, similar to most other protein–carbohydrate interactions, including those observed in type B CBMs.
‡No binding detected. §Binding too weak to quantify by ITC. |
Furthermore, to investigate the possible enhancement of the activity of other enzymes on incorporating MtCBM, we constructed a chimeric TmCel5A-CBM protein in which MtCBM was inserted into TmCel5A at an equivalent internal location to that in MtGlu5. For comparative purposes, TmCel5A alone was also expressed and purified. Again, to avoid interference, subsequent ITC measurements used the inactive E253Q mutant of TmCel5A, and the corresponding proteins were named iTmCel5A and iTmCel5A-CBM. As expected, significantly different binding profiles were observed (Figs. 2e and 2f): iTmCel5A displayed low affinity towards CMC, similar to iΔCBM, while in contrast the chimeric iTmCel5A-CBM exhibited a strong affinity like that of iMtGlu5 (Table 2). The results of the ITC experiment serve as evidence of synergy between the CD and CBM in MtGlu5, as well as in the chimeric TmCel5A-CBM. Full-length iMtGlu5 presents an approximately 105-fold higher affinity towards substrate than the individual iΔCBM or gMtCBM (Table 2 and Fig. 2). This synergistically cooperative binding between two domains has also been demonstrated in other CBM-associated proteins such as CBM29-1-2 and Starch Excess4 (SEX4), but is not always present (Freelove et al., 2001; Meekins et al., 2014; Fox et al., 2013; Fernandes et al., 1999; Várnai et al., 2013). Presumably, the inserted CBMs act in synergy with CDs just like the intact iMtGlu5 and the chimeric iTmCel5A-CBM, which display an impressively higher affinity for the soluble substrates.
3.3. Deletion of the CBM domain severely impairs the catalytic efficiency of MtGlu5
The overall structure of MtGlu5 suggests clear separation of the CD and CBM into two independent domains (Fig. 1b). To better determine the role of the noncatalytic MtCBM, the active-form proteins full-length MtGlu5 and CBM-deleted ΔCBM were assayed for enzymatic activity. After testing various conditions, subsequent measurements were performed in 20 mM CHC buffer pH 5.0 with incubation for 10 min at 60°C (Supplementary Fig. S2). The results using the insoluble substrate regenerated amorphous cellulose (RAC) are shown in Fig. 3(a). The enzymatic activity of ΔCBM towards the RAC substrate was reduced to 55% of that of MtGlu5 (P < 0.001). Similar results were obtained using the soluble substrate CMC (Supplementary Fig. S3). The kinetic data indicated that without the CBM domain, the Km of ΔCBM increased nearly ninefold and the kcat decreased by more than tenfold (Table 3; Supplementary Fig. S4a). As a comparison, we also measured the enzymatic activity of TmCel5A and TmCel5A-CBM against RAC and determined the Km and kcat towards CMC. As shown in Fig. 3(b), incorporation of the CBM tripled the overall activity of TmCel5A towards RAC (p < 0.0001), which was mainly a result of the decreased Km, as the kcat largely remained the same (Table 3; Supplementary Fig. S4b). Interestingly, the Km values of TmCel5A and ΔCBM are very similar, suggesting a shared low substrate affinity of the Nevertheless, the artificial TmCel5A-CBM had a much lower performance when compared with the natural MtGlu5, presumably due to a lack of optimization. In fact, the first attempt to insert MtCBM into TmCel5A at a site just ten residues away failed to improve its activity, providing evidence for the importance of precise binding-cleft alignment with the CD for an inserted CBM to be beneficial.
|
In most other studies, CBM deletion or translocation did not dramatically change the activities of endo-cellulases (Wu et al., 2018; Várnai et al., 2013). Likewise, TmCel5A had almost the same kcat as TmCel5A-CBM. However, ΔCBM showed a dramatically lower enzymatic activity than MtGlu5 (Fig. 3a, Supplementary Fig. S3). The lower enzymatic activity of ΔCBM is not due to structural damage to the catalytic center, the integrity of which was demonstrated by analyzing the hydrolyzed products using Both full-length and CBM-deleted MtGlu5 remain capable of hydrolysis (Supplementary Fig. S5). To investigate the origin of the disparity in catalytic efficiency between ΔCBM and MtGlu5, we crystallized ΔCBM and solved the structure by using the CD of MtGlu5 as a search model. The orthorhombic P212121 crystal contained one molecule of ΔCBM in its The apo-form structure of ΔCBM displayed an intact TIM-barrel fold (Supplementary Fig. S6). The of ΔCBM presents almost the same polypeptide conformation as observed in the CD of MtGlu5, with an r.m.s.d. of 0.28 Å for 260 pairs of Cα atoms (Fig. 3c, left). Both structures superimpose well with TmCel5A, with r.m.s.d.s of 0.56 and 0.54 Å for 263 and 261 Cα atom pairs, respectively (Fig. 3c, left). Moreover, ΔCBM crystals in complex with glucose and cellobiose were obtained by soaking with cellopentaose and cellobiose, respectively. The sugar molecules in both structures were also sandwiched between the Trp43 and Trp224 side chains just as in MtGlu5 (Supplementary Figs. S7 and S8). The presence of bound glucose rather than cellopentaose in the former crystal is also evidence that ΔCBM remains capable of catalysis. The much lower kcat value of ΔCBM was likely to be a result of loop perturbation. In MtGlu5 (and also in TmCel5A) the β6–α6 loop is 30 residues long and contains the substrate-binding Trp224 (Trp210). Direct elimination of the CBM from MtGlu5 resulted in a shorter connection between this loop and the α6 helix, the equivalent of which in TmCel5A has one additional turn (Fig. 3c, right). Although the overall fold remained intact, the lack of flexibility in the tightened-up loop would probably have reduced the catalytic efficiency of ΔCBM.
3.4. The complex model reveals a substrate-binding cleft lined with aromatic residues
According to the above structural data, a plausible substrate-binding model of MtGlu5 can be constructed. Inspection of the MtGlu5 structure shows a distinct cleft that traverses from the bound sugars on one side of the CD through the catalytic pocket to the CBM domain on the other side. It is presumed to be a possible substrate-binding groove with many surface tryptophan residues. MtGlu5 consists of two domains: a Cel5A-like TIM-barrel fold and a CBM29-2-like noncatalytic carbohydrate-binding module which is inserted into the intact CD and extends the binding groove (Fig. 4). To investigate the possible substrate-recognition mechanism, the structure of MtGlu5 was compared with those of TmCel5A and CBM29-2. Superimposition of TmCel5A in complex with cellotetraose onto the CD of MtGlu5 showed that the residues in the catalytic pocket overlaid very well, revealing highly conserved structures (Fig. 4a). Both of them share a similar substrate-binding cleft beside the catalytic pocket (Fig. 4b). Likewise, when CBM29-2 in complex with cellohexaose was superimposed on MtCBM, the three conserved aromatic residues on the hydrophobic surface of CBM29-2 overlaid well with MtCBM, except for Tyr46 and Trp270. The superimposition also revealed an ambiguous binding cleft in MtCBM through the three corresponding surface aromatic residues (Fig. 4c). Combined with the superimposition results of MtGlu5, TmCel5A in complex with cellotetraose and CBM29-2 in complex with cellohexaose, a plausible continuous binding groove emerged that is richly embedded with surface tryptophan residues (Fig. 4b). The affinity and kinetic data also indicated that both the CD and CBM of MtGlu5 contribute to its binding to polysaccharide substrates.
Although only a couple of bound sugar residues were seen in the CD of MtGlu5, their positions are consistent with those in TmCel5A. While no sugar was seen in the CBM, the conserved surface aromatic residues suggest a similar sugar-binding mode to those of CBM29s. A model was therefore generated by bridging the segments of bound sugars from TmCel5A and CBM29-2 with three additional residues, making a continuous polysaccharide chain that passes through the putative binding groove (Supplementary Fig. S9a). The model was subjected to energy minimization using the YASARA server (https://www.yasara.org/minimizationserver.htm; Krieger et al., 2002). In the CD of this model, the three important tryptophan residues Trp43, Trp224 and Trp188, as well as other active-site residues, for example His108, His109 and Tyr212, can form hydrogen bonds to the substrate in the catalytic pocket (Supplementary Fig. S9b). Additionally, MtCBM may make hydrophobic and van der Waals interactions with the substrate through the three crucial surface tryptophan residues Trp249, Trp251 and Trp270. The backbone of Trp249 and the side chains of Arg303 and Gln337 may also form direct hydrogen bonds to the substrate (Supplementary Fig. S9c). In this model, the putative binding groove is 12–14 in length from one side to the other. Moreover, numerous factors may contribute to carbohydrate binding. Some of the pivotal factors, for example stacking interactions, are provided by aromatic residues. The stacking interaction of aromatic residues against the pyranose ring of sugars consists of hydrophobic interactions, van der Waals interactions, hydrogen bonding and even CH–π interaction of aromatic residues in CBMs (Campos et al., 2016; Szabó et al., 2001; Carvalho et al., 2004; Venditto et al., 2016; Nagy et al., 1998; Flint et al., 2004; Asensio et al., 2013). Tryptophan is usually the most essential residue to provide stacking interactions with sugar residues in other known CBMs. However, in the complex structure of MtGlu5 and the proposed binding model tryptophan residues are not only present in the substrate-binding groove of CBM but also in the CD region. In other words, these surface aromatic residues are presumed to work together in binding to the polysaccharide substrate.
Similar cross-domain binding to the substrate has been found in, for example, the processive endoglucanase E4 (now Cel9A-68) from Thermomonospora fusca, which also shows an extended binding surface from E4CD to the C-terminal CBM3c (Sakon et al., 1997). Previous studies demonstrated that the type A CBM3c is essential for crystalline cellulose binding and catalytic processivity in Cel9A-68. However, the flat binding surface of CBM3c lacks several conserved aromatic residues which are important for substrate binding in CBM3a and CBM3b. Furthermore, mutagenesis of the binding-surface residues on CBM3c does not dramatically decrease the activity of Cel9A-68 (Li et al., 2007, 2010; Tormo et al., 1996). Unlike the processive endoglucanase Cel9A-68, neither MtGlu5 nor ΔCBM present processivity, indicating that the extended binding groove in the presence of a type B MtCBM does not contribute to processive hydrolysis (Supplementary Table S2; Irwin et al., 1993). Moreover, MtGlu5 presents synergistic cooperation between the CD and CBM in which each tryptophan side chain on the binding groove is essential for hydrolytic activity (see below). This is also different from Cel9A-68. Although the substrate-binding modes appear to be similar for MtGlu5 and Cel9A-68, the biochemical properties of the two enzymes are distinct, indicating that MtGlu5 has a novel binding mode compared with Cel9A-68.
3.5. Mutagenesis results support the key role of aromatic residues in substrate binding
By structural comparison with CBM29-2, the three surface tryptophan residues Trp249, Trp251 and Trp270 in MtCBM were presumed to participate in substrate binding. Here, this hypothesis was verified by site-directed mutagenesis. The glucanase activity towards RAC of the single Trp mutants W249A, W251A and W270A, the double Trp mutants W249/251A, W249/270A and W251/270A and the triple Trp mutant W249/251/270A in the CBM region of MtGlu5 were measured and compared with the wild-type enzyme (Fig. 5a). All of the single mutants retained ∼40% of the activity of MtGlu5. Moreover, the data for the double mutants and triple mutant showed that when Trp249 and Trp251 were mutated simultaneously, the activity was dramatically reduced by ∼80%. In either the single, double or triple mutants, altering Trp270 seems to have milder adverse effects, probably due to its distal location in the substrate-binding cleft. Similar results were obtained using CMC as the substrate (Supplementary Fig. S10a). Four other surface tryptophan residues, Trp43, Trp224, Trp188 and Trp192, are found in the putative binding groove of the CD of MtGlu5. These four residues, along with the nearby non-aromatic Glu216, which served as a negative control, were also tested by site-directed mutagenesis. Using RAC as the substrate, W43A, W224A and W188A showed a prominent reduction in activity to less than 50%, whereas the effects of W192A and E216A were not as significant (Fig. 5b, Supplementary Fig. S10b).
Further kinetic measurements showed that single Trp mutants in the CBM drastically increased the Km to 15–17 times that of the wild type (Table 3, Supplementary Fig. S4c), but the kcat was also increased by 2.2–3.6-fold. The reduced substrate affinity might facilitate product release, resulting in higher turnover numbers. However, the double and triple Trp mutants did not show a further increase in Km, whereas the kcat values were significantly reduced. The resulting kcat/Km values were comparable to that of ΔCBM, suggesting that these mutants virtually aborted the original function of the CBM. Again, mutants involving the distal Trp270 showed milder effects on the activity of the enzyme (Fig. 5a, Supplementary Fig. S10a). The Km values of the W43A, W188A, W192A and E216A mutants in the CD region did not increase as much as those of the CBM mutants described above, but the Km of W224A was eight times that of the wild type (Table 3, Supplementary Fig. S4d). On the other hand, only the kcat of W43A and W224A showed a prominent reduction, by threefold and fourfold, respectively. Both Trp43 and Trp224 participate in binding to sugar units on the nonreducing end of the polysaccharide substrate immediately adjacent to the active site, and their absence should slow the reaction (Fig. 1c, Table 3). The low kcat/Km value of W224A again underscores the key role of Trp224 in catalysis by MtGlu5. In comparison, the sustained activity of W188A shows that Trp188 is less important in substrate binding. In comparison, the activity of W192A and E216A was only slightly decreased. Trp192 is indeed located at the CD–CBM interface, and Glu216 is in the Trp224-containing β6–α6 loop. Both residues apparently do not interact directly with the substrate.
Interestingly, by testing with various other substrates, MtGlu5 was found to be more active towards β-glucan than CMC, indicating that the enzyme may also bind efficiently to mixed 1,3- and 1,4-linked glucose units (Fig. 5c). However, the enzyme showed only β-1,4-glucanase activity. It did not cleave branched or α-linked glucans and nonglucose substrates. Additionally, ligand-induced protein stabilization is a widely known phenomenon that has been well described (Pace & McGrath, 1980; Pantoliano et al., 2001). To further evaluate the importance of tryptophan residues in substrate binding, the thermal shift assay, or the inflection point of protein unfolding (ΔTi), was measured for each inactive mutant in the presence or absence of CMC. As shown in Table 4, stabilization of iMtGlu5 by CMC binding resulted in a ΔTi of 4°C. The diminished thermal shift of iΔCBM indicates that MtCBM plays a key role in substrate binding. Similar effects were seen for Trp249 and Trp251 mutations in the CBM region. Mutation of the distal Trp270 in CBM had a less prominent effect, and so did those of the other tryptophan residues Trp43, Trp188 and Trp224 in the CD region, suggesting minor but still essential roles in substrate binding (Table 4). The thermal shift data are consistent with the above measurements, while mutating Trp192 and Glu216 showed milder effects on the activity but even larger ΔTi values than the wild-type enzyme (Table 4), suggesting a close relationship between substrate binding and a stable functioning protein conformation.
|
4. Concluding remarks
In this study, we report a new type of endoglucanase with a novel inserted CBM that is in the middle of the intact CD. Unlike other known cellulases with CBMs at the termini, this inserted CBM in the newly identified MtGlu5 acts in synergy with the CD, as demonstrated by ITC experiments. Superimposition of MtGlu5 with homologous structures, namely TmCel5A and CBM29-2, and the subsequent construction of a substrate-binding model suggest that the polysaccharide chain interacts with a continuous groove extending from the CD to the CBM. The binding mode of the endoglucanase MtGlu5 is similar but not the same as that of Cel9A-68 from T. fusca. Six tryptophan residues in the groove play a key role in sugar binding: Trp249, Trp251 and Trp270 in the CBM, and Trp43, Trp224 and Trp188 in the CD. As demonstrated by mutagenesis experiments, the loss of these surface tryptophans significantly reduced the substrate affinity. We also showed that by extending the substrate-binding groove, the chimeric TmCel5A-CBM was endowed with a higher affinity for longer Notably, the inserted CBM is not a fixed attachment, as its orientation varies with respect to the CD in the monoclinic and tetragonal crystals of iMtGlu5 and MtGlu5. The CBMs of the two iMtGlu5 molecules differ by a rigid-body rotation of 22.5°, and they differ from that of wild-type MtGlu5 by rotations of 25.3° and 10.8° (Supplementary Fig. S11). While large-scale conformational changes are often observed in enzymes upon substrate binding (Arora & Brooks, 2007; Suzuki et al., 2019), the moderate flexibility of CBM in MtGlu5 probably allows switching between different orientations of polysaccharide chains. The above findings provide a new paradigm for studying other as yet uncharacterized cellulases with inserted CBMs, which might act against natural insoluble substrates more efficiently in potential industrial applications. This approach may be complementary to recent work on multi-enzyme complexes in cellulose deconstruction (McConnell et al., 2020).
Supporting information
Supplementary Figures and Tables. DOI: https://doi.org/10.1107/S2059798322002601/jc5049sup1.pdf
Acknowledgements
We thank Dr Meng-Ru Ho of the Biophysics Facility, Ms Hui-Ling Shih of the Protein Crystallization Facility and the Protein X-ray Crystallography Service at the Institute of Biological Chemistry, Academia Sinica. We thank Dr Shu-Chuan (Chris) Jao, Mr Kun-Hung Chen and Miss Yu Jin-Hsuan in the Biophysics Core Facility, funded by the Academia Sinica Core Facility and Innovative Instrument Project (AS-CFII108-111), for providing useful discussions on data acquisition and analysis in the iTC200, PEAQ-ITC and Tycho NT.6 experiments. We also thank the staff of beamlines 05A1, 15A1 and 13B1 at the National Synchrotron Radiation Research Center (Hsinchu, Taiwan) for assistance with X-ray data collection.
Funding information
The following funding is acknowledged: Ministry of Science and Technology, Taiwan.
References
Arora, K. & Brooks, C. L. (2007). Proc. Natl Acad. Sci. USA, 104, 18496–18501. Web of Science CrossRef PubMed CAS Google Scholar
Asensio, J. L., Ardá, A., Cañada, F. J. & Jiménez-Barbero, J. (2013). Acc. Chem. Res. 46, 946–954. Web of Science CrossRef CAS PubMed Google Scholar
Aspeborg, H., Coutinho, P. M., Wang, Y., Brumer, H. & Henrissat, B. (2012). BMC Evol. Biol. 12, 186. Google Scholar
Basit, A. & Akhtar, M. W. (2018). Biotechnol. Bioeng. 115, 1675–1684. CrossRef CAS PubMed Google Scholar
Bayer, E. A., Chanzy, H., Lamed, R. & Shoham, Y. (1998). Curr. Opin. Struct. Biol. 8, 548–557. Web of Science CrossRef CAS PubMed Google Scholar
Boraston, A. B., Bolam, D. N., Gilbert, H. J. & Davies, G. J. (2004). Biochem. J. 382, 769–781. Web of Science CrossRef PubMed CAS Google Scholar
Breyer, W. A. & Matthews, B. W. (2001). Protein Sci. 10, 1699–1711. CrossRef PubMed CAS Google Scholar
Campos, B. M., Liberato, M. V., Alvarez, T. M., Zanphorlin, L. M., Ematsu, G. C., Barud, H., Polikarpov, I., Ruller, R., Gilbert, H. J., Zeri, A. C. & Squina, F. M. (2016). J. Biol. Chem. 291, 23734–23743. CrossRef CAS PubMed Google Scholar
Carvalho, A. L., Goyal, A., Prates, J. A. M., Bolam, D. N., Gilbert, H. J., Pires, V. M. R., Ferreira, L. M. A., Planas, A., Romão, M. J. & Fontes, C. M. G. A. (2004). J. Biol. Chem. 279, 34785–34793. Web of Science CrossRef PubMed CAS Google Scholar
Charnock, S. J., Bolam, D. N., Nurizzo, D., Szabó, L., McKie, V. A., Gilbert, H. J. & Davies, G. J. (2002). Proc. Natl Acad. Sci. USA, 99, 14077–14082. CrossRef PubMed CAS Google Scholar
Chayen, N. E. (1997). Structure, 5, 1269–1274. CrossRef CAS PubMed Web of Science Google Scholar
Chen, M.-Y., Lin, G.-H., Lin, Y.-T. & Tsay, S.-S. (2002). Int. J. Syst. Evol. Microbiol. 52, 1647–1654. PubMed CAS Google Scholar
Coughlan, M. P. (1985). Biotechnol. Genet. Eng. Rev. 3, 39–110. CrossRef CAS Google Scholar
Davies, G. & Henrissat, B. (1995). Structure, 3, 853–859. CrossRef CAS PubMed Web of Science Google Scholar
Davies, G. J., Wilson, K. S. & Henrissat, B. (1997). Biochem. J. 321, 557–559. CrossRef CAS PubMed Web of Science Google Scholar
Demain, A. L., Newcomb, M. & Wu, J. H. D. (2005). Microbiol. Mol. Biol. Rev. 69, 124–154. Web of Science CrossRef PubMed CAS Google Scholar
Drula, E., Garron, M. L., Dogan, S., Lombard, V., Henrissat, B. & Terrapon, N. (2022). Nucleic Acids Res. 50, D571–D577. CrossRef CAS PubMed Google Scholar
Eibinger, M., Ganner, T., Plank, H. & Nidetzky, B. (2020). ACS Cent. Sci. 6, 739–746. CrossRef CAS PubMed Google Scholar
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501. Web of Science CrossRef CAS IUCr Journals Google Scholar
Fernandes, A. C., Fontes, C. M. G. A., Gilbert, H. J., Hazlewood, G. P., Fernandes, T. H. & Ferreira, L. M. A. (1999). Biochem. J. 342, 105–110. CrossRef PubMed CAS Google Scholar
Flint, H. J., Whitehead, T. R., Martin, J. C. & Gasparic, A. (1997). Biochim. Biophys. Acta, 1337, 161–165. CrossRef CAS PubMed Google Scholar
Flint, J., Bolam, D. N., Nurizzo, D., Taylor, E. J., Williamson, M. P., Walters, C., Davies, G. J. & Gilbert, H. J. (2005). J. Biol. Chem. 280, 23718–23726. CrossRef PubMed CAS Google Scholar
Flint, J., Nurizzo, D., Harding, S. E., Longman, E., Davies, G. J., Gilbert, H. J. & Bolam, D. N. (2004). J. Mol. Biol. 337, 417–426. CrossRef PubMed CAS Google Scholar
Fox, J. M., Jess, P., Jambusaria, R. B., Moo, G. M., Liphardt, J., Clark, D. S. & Blanch, H. W. (2013). Nat. Chem. Biol. 9, 356–361. CrossRef CAS PubMed Google Scholar
Freelove, A. C., Bolam, D. N., White, P., Hazlewood, G. P. & Gilbert, H. J. (2001). J. Biol. Chem. 276, 43010–43017. Web of Science CrossRef PubMed CAS Google Scholar
Gilbert, H. J. (2010). Plant Physiol. 153, 444–455. Web of Science CrossRef CAS PubMed Google Scholar
Gilbert, H. J., Knox, J. P. & Boraston, A. B. (2013). Curr. Opin. Struct. Biol. 23, 669–677. Web of Science CrossRef CAS PubMed Google Scholar
Hashimoto, H. (2006). Cell. Mol. Life Sci. 63, 2954–2967. Web of Science CrossRef PubMed CAS Google Scholar
Henrissat, B., Claeyssens, M., Tomme, P., Lemesle, L. & Mornon, J.-P. (1989). Gene, 81, 83–95. CrossRef CAS PubMed Web of Science Google Scholar
Hervé, C., Rogowski, A., Blake, A. W., Marcus, S. E., Gilbert, H. J. & Knox, J. P. (2010). Proc. Natl Acad. Sci. USA, 107, 15293–15298. Web of Science PubMed Google Scholar
Himmel, M. E., Ding, S.-Y., Johnson, D. K., Adney, W. S., Nimlos, M. R., Brady, J. W. & Foust, T. D. (2007). Science, 315, 804–807. Web of Science CrossRef PubMed CAS Google Scholar
Holm, L. (2020). Methods Mol. Biol. 2112, 29–42. CrossRef CAS PubMed Google Scholar
Holm, L. & Sander, C. (1993). J. Mol. Biol. 233, 123–138. CrossRef CAS PubMed Web of Science Google Scholar
Horn, S. J., Sørlie, M., Vårum, K. M., Väljamäe, P. & Eijsink, V. G. (2012). Methods Enzymol. 510, 69–95. CrossRef CAS PubMed Google Scholar
Irwin, D. C., Spezio, M., Walker, L. P. & Wilson, D. B. (1993). Biotechnol. Bioeng. 42, 1002–1013. CrossRef PubMed CAS Google Scholar
Jalak, J., Kurašin, M., Teugjas, H. & Väljamäe, P. (2012). J. Biol. Chem. 287, 28802–28815. CrossRef CAS PubMed Google Scholar
Kao, M.-R., Kuo, H.-W., Lee, C.-C., Huang, K.-Y., Huang, T.-Y., Li, C.-W., Chen, C.-W., Wang, A. H.-J., Yu, S.-M. & Ho, T.-H. D. (2019). Biotechnol. Biofuels, 12, 258. CrossRef PubMed Google Scholar
Krieger, E., Koraimann, G. & Vriend, G. (2002). Proteins, 47, 393–402. Web of Science CrossRef PubMed CAS Google Scholar
Kuhad, R. C., Gupta, R. & Singh, A. (2011). Enzym. Res. 2011, 280696. CrossRef Google Scholar
Li, Y., Irwin, D. C. & Wilson, D. B. (2007). Appl. Environ. Microbiol. 73, 3165–3172. CrossRef PubMed CAS Google Scholar
Li, Y., Irwin, D. C. & Wilson, D. B. (2010). Appl. Environ. Microbiol. 76, 2582–2588. CrossRef CAS PubMed Google Scholar
Liang, P.-H., Lin, W.-L., Hsieh, H.-Y., Lin, T.-Y., Chen, C.-H., Tewary, S. K., Lee, H.-L., Yuan, S.-F., Yang, B., Yao, J.-Y. & Ho, M.-C. (2018). Biochim. Biophys. Acta, 1862, 513–521. CrossRef CAS Google Scholar
Liao, J.-H., Ihara, K., Kuo, C.-I., Huang, K.-F., Wakatsuki, S., Wu, S.-H. & Chang, C.-I. (2013). Acta Cryst. D69, 1395–1402. Web of Science CrossRef CAS IUCr Journals Google Scholar
Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861–877. Web of Science CrossRef IUCr Journals Google Scholar
Lin, C.-C., Su, S.-C., Su, M.-Y., Liang, P.-H., Feng, C.-C., Wu, S.-H. & Chang, C.-I. (2016). Structure, 24, 667–675. CrossRef CAS PubMed Google Scholar
Lombard, V., Golaconda Ramulu, H., Drula, E., Coutinho, P. M. & Henrissat, B. (2014). Nucleic Acids Res. 42, D490–D495. Web of Science CrossRef CAS PubMed Google Scholar
Luís, A. S., Venditto, I., Temple, M. J., Rogowski, A., Baslé, A., Xue, J., Knox, J. P., Prates, J. A. M., Ferreira, L. M. A., Fontes, C. M. G. A., Najmudin, S. & Gilbert, H. J. (2013). J. Biol. Chem. 288, 4799–4809. Web of Science PubMed Google Scholar
McConnell, S. A., Cannon, K. A., Morgan, C., McAllister, R., Amer, B. R., Clubb, R. T. & Yeates, T. O. (2020). ACS Synth. Biol. 9, 381–391. CrossRef CAS PubMed Google Scholar
Meekins, D. A., Raththagala, M., Husodo, S., White, C. J., Guo, H.-F., Kötting, O., Vander Kooi, C. W. & Gentry, M. S. (2014). Proc. Natl Acad. Sci. USA, 111, 7272–7277. Web of Science CrossRef CAS PubMed Google Scholar
Miller, G. L. (1959). Anal. Chem. 31, 426–428. CrossRef CAS Web of Science Google Scholar
Murshudov, G. N., Skubák, P., Lebedev, A. A., Pannu, N. S., Steiner, R. A., Nicholls, R. A., Winn, M. D., Long, F. & Vagin, A. A. (2011). Acta Cryst. D67, 355–367. Web of Science CrossRef CAS IUCr Journals Google Scholar
Nagy, T., Simpson, P., Williamson, M. P., Hazlewood, G. P., Gilbert, H. J. & Orosz, L. (1998). FEBS Lett. 429, 312–316. CrossRef CAS PubMed Google Scholar
Novy, V., Aïssa, K., Nielsen, F., Straus, S. K., Ciesielski, P., Hunt, C. G. & Saddler, J. (2019). Proc. Natl Acad. Sci. USA, 116, 22545–22551. CrossRef CAS PubMed Google Scholar
Otwinowski, Z. & Minor, W. (1997). Methods Enzymol. 276, 307–326. CrossRef CAS PubMed Web of Science Google Scholar
Pace, C. N. & McGrath, T. (1980). J. Biol. Chem. 255, 3862–3865. CrossRef CAS PubMed Google Scholar
Pantoliano, M. W., Petrella, E. C., Kwasnoski, J. D., Lobanov, V. S., Myslik, J., Graf, E., Carver, T., Asel, E., Springer, B. A., Lane, P. & Salemme, F. R. (2001). J. Biomol. Screen. 6, 429–440. Web of Science CrossRef PubMed CAS Google Scholar
Pereira, J. H., Chen, Z., McAndrew, R. P., Sapra, R., Chhabra, S. R., Sale, K. L., Simmons, B. A. & Adams, P. D. (2010). J. Struct. Biol. 172, 372–379. Web of Science CrossRef CAS PubMed Google Scholar
Ragauskas, A. J., Williams, C. K., Davison, B. H., Britovsek, G., Cairney, J., Eckert, C. A., Frederick, W. J. Jr, Hallett, J. P., Leak, D. J., Liotta, C. L., Mielenz, J. R., Murphy, R., Templer, R. & Tschaplinski, T. (2006). Science, 311, 484–489. Web of Science CrossRef PubMed CAS Google Scholar
Ravachol, J., Borne, R., Tardif, C., de Philip, P. & Fierobe, H. P. (2014). J. Biol. Chem. 289, 7335–7348. CrossRef CAS PubMed Google Scholar
Roy, A., Kucukural, A. & Zhang, Y. (2010). Nat. Protoc. 5, 725–738. Web of Science CrossRef CAS PubMed Google Scholar
Rubin, E. M. (2008). Nature, 454, 841–845. CrossRef PubMed CAS Google Scholar
Sakon, J., Irwin, D., Wilson, D. B. & Karplus, P. A. (1997). Nat. Struct. Mol. Biol. 4, 810–818. CrossRef CAS Web of Science Google Scholar
Schwarz, W. H. (2001). Appl. Microbiol. Biotechnol. 56, 634–649. Web of Science CrossRef PubMed CAS Google Scholar
Sidar, A., Albuquerque, E. D., Voshol, G. P., Ram, A. F. J., Vijgenboom, E. & Punt, P. J. (2020). Front. Bioeng. Biotechnol. 8, 871. CrossRef PubMed Google Scholar
Sierla, M., Hõrak, H., Overmyer, K., Waszczak, C., Yarmolinsky, D., Maierhofer, T., Vainonen, J. P., Salojärvi, J., Denessiouk, K., Laanemets, K., Tõldsepp, K., Vahisalu, T., Gauthier, A., Puukko, T., Paulin, L., Auvinen, P., Geiger, D., Hedrich, R., Kollist, H. & Kangasjärvi, J. (2018). Plant Cell, 30, 2813–2837. CrossRef CAS PubMed Google Scholar
Su, S.-C., Lin, C.-C., Tai, H.-C., Chang, M.-Y., Ho, M.-R., Babu, C. S., Liao, J.-H., Wu, S.-H., Chang, Y.-C., Lim, C. & Chang, C.-I. (2016). Structure, 24, 676–686. CrossRef CAS PubMed Google Scholar
Suzuki, K., Maeda, S. & Morokuma, K. (2019). ACS Omega, 4, 1178–1184. CrossRef CAS PubMed Google Scholar
Szabó, L., Jamal, S., Xie, H., Charnock, S. J., Bolam, D. N., Gilbert, H. J. & Davies, G. J. (2001). J. Biol. Chem. 276, 49061–49065. PubMed Google Scholar
Teeri, T. T. (1997). Trends Biotechnol. 15, 160–167. CrossRef Google Scholar
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). Nucleic Acids Res. 22, 4673–4680. CrossRef CAS PubMed Web of Science Google Scholar
Tomme, P., Van Tilbeurgh, H., Pettersson, G., Van Damme, J., Vandekerckhove, J., Knowles, J., Teeri, T. & Claeyssens, M. (1988). Eur. J. Biochem. 170, 575–581. CrossRef CAS PubMed Web of Science Google Scholar
Tormo, J., Lamed, R., Chirino, A. J., Morag, E., Bayer, E. A., Shoham, Y. & Steitz, T. A. (1996). EMBO J. 15, 5739–5751. CrossRef CAS PubMed Web of Science Google Scholar
Urbanowicz, B. R., Catalá, C., Irwin, D., Wilson, D. B., Ripoll, D. R. & Rose, J. K. (2007). J. Biol. Chem. 282, 12066–12074. CrossRef PubMed CAS Google Scholar
Vagin, A. & Teplyakov, A. (2010). Acta Cryst. D66, 22–25. Web of Science CrossRef CAS IUCr Journals Google Scholar
Várnai, A., Siika-aho, M. & Viikari, L. (2013). Biotechnol. Biofuels, 6, 30. PubMed Google Scholar
Venditto, I., Luis, A. S., Rydahl, M., Schückel, J., Fernandes, V. O., Vidal-Melgosa, S., Bule, P., Goyal, A., Pires, V. M. R., Dourado, C. G., Ferreira, L. M. A., Coutinho, P. M., Henrissat, B., Knox, J. P., Baslé, A., Najmudin, S., Gilbert, H. J., Willats, W. G. T. & Fontes, C. M. G. A. (2016). Proc. Natl Acad. Sci. USA, 113, 7136–7141. CrossRef CAS PubMed Google Scholar
Williams, C. J., Headd, J. J., Moriarty, N. W., Prisant, M. G., Videau, L. L., Deis, L. N., Verma, V., Keedy, D. A., Hintze, B. J., Chen, V. B., Jain, S., Lewis, S. M., Arendall, W. B., Snoeyink, J., Adams, P. D., Lovell, S. C., Richardson, J. S. & Richardson, J. S. (2018). Protein Sci. 27, 293–315. Web of Science CrossRef CAS PubMed Google Scholar
Wu, B., Zheng, S., Pedroso, M. M., Guddat, L. W., Chang, S., He, B. & Schenk, G. (2018). Biotechnol. Biofuels, 11, 20. CrossRef PubMed Google Scholar
Wu, H., Ioannou, E., Henrissat, B., Montanier, C. Y., Bozonnet, S., O'Donohue, M. J. & Dumon, C. (2021). Appl Environ Microbiol, 87, e01714-20. CAS PubMed Google Scholar
Wu, T.-H., Huang, C.-H., Ko, T.-P., Lai, H.-L., Ma, Y., Chen, C.-C., Cheng, Y.-S., Liu, J.-R. & Guo, R.-T. (2011). Biochim. Biophys. Acta, 1814, 1832–1840. Web of Science CrossRef CAS PubMed Google Scholar
Wu, W.-L., Chen, M.-Y., Tu, I.-F., Lin, Y.-C., EswarKumar, N., Chen, M.-Y., Ho, M.-C. & Wu, S.-H. (2017). Sci. Rep. 7, 4658. CrossRef PubMed Google Scholar
Zeng, G. (1998). Biotechniques, 25, 206–208. Web of Science CrossRef CAS PubMed Google Scholar
Zhang, P. (2006). J. Fluoresc. 16, 349–353. CrossRef PubMed CAS Google Scholar
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.