Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2015 Jan 13;24(3):408–419. doi: 10.1002/pro.2632

Structure, activity, and stability of metagenome-derived glycoside hydrolase family 9 endoglucanase with an N-terminal Ig-like domain

Hiroyuki Okano 1, Eiko Kanaya 1, Masashi Ozaki 1, Clement Angkawidjaja 1,2, Shigenori Kanaya 1,*
PMCID: PMC4353366  PMID: 25545469

Abstract

A metagenome-derived glycoside hydrolase family 9 enzyme with an N-terminal immunoglobulin-like (Ig-like) domain, leaf-branch compost (LC)-CelG, was characterized and its crystal structure was determined. LC-CelG did not hydrolyze p-nitrophenyl cellobioside but hydrolyzed CM-cellulose, indicating that it is endoglucanase. LC-CelG exhibited the highest activity at 70°C and >80% of the maximal activity at a broad pH range of 5–9. Its denaturation temperature was 81.4°C, indicating that LC-CelG is a thermostable enzyme. The structure of LC-CelG resembles those of CelD from Clostridium thermocellum (CtCelD), Cel9A from Alicyclobacillus acidocaldarius (AaCel9A), and cellobiohydrolase CbhA from C. thermocellum (CtCbhA), which show relatively low (29–31%) amino acid sequence identities to LC-CelG. Three acidic active site residues are conserved as Asp194, Asp197, and Glu558 in LC-CelG. Ten of the thirteen residues that form the substrate binding pocket of AaCel9A are conserved in LC-CelG. Removal of the Ig-like domain reduced the activity and stability of LC-CelG by 100-fold and 6.3°C, respectively. Removal of the Gln40- and Asp99-mediated interactions between the Ig-like and catalytic domains destabilized LC-CelG by 5.0°C without significantly affecting its activity. These results suggest that the Ig-like domain contributes to the stabilization of LC-CelG mainly due to the Gln40- and Asp99-mediated interactions. Because the LC-CelG derivative lacking the Ig-like domain accumulated in Escherichia coli cells mostly in an insoluble form and this derivative accumulated in a soluble form exhibited very weak activity, the Ig-like domain may be required to make the conformation of the active site functional and prevent aggregation of the catalytic domain.

Keywords: metagenome, GH family 9, endoglucanase, immunoglobulin-like (Ig-like) domain, crystal structure, thermal stability

Introduction

Cellulose is the major component of plant cell walls and is one of the most abundant organic molecules found in nature. Cellulase degrades glycosidic bonds of cellulose and is widely used for industrial purposes, such as fiber processing, food processing, and bioethanol production.1–6 The complete degradation of cellulose requires the synergistic action of three types of cellulases, endo-1,4-β-d-glucanase (endoglucanase; EC 3.2.1.4), exo-1,4-β-d-glucanase (cellobiohydrolase; EC 3.2.1.91 and 3.2.1.176), and β-glucosidase (EC 3.2.1.21). These cellulases and other enzymes that hydrolyze glycosidic bonds are collectively termed glycoside hydrolases (GH). According to the CAZy database (http://www.cazy.org/), GHs are classified into 133 families (GH families 1–133) based on the difference in their amino acid sequences. Endoglucanases, which participate in the decomposition of cellulose polymer by hydrolyzing the β-1,4-d-glycosidic linkages in cellulose, are classified into 12 families (GH families 5, 6–9, 12, 44, 45, 48, 51, 74, and 124).

GH family 9 is one of the well-known families of endoglucanases, although it includes several cellobiohydrolases and other carbohydrate-degrading enzymes as well. To date, over 100 GH family 9 enzymes have been characterized. For 12 of them, the crystal structures are available. These structures can be divided into three groups based on the difference in the number and types of the domains. The first group includes the structures of Cel9M from Clostridium cellulolyticum (PDB ID 1IA6),7 CelT from C. thermocellum (2YIK),8 EF-EG2 from earthworm Eisenia fetida (3WC3),9 and NtEgl from termite Nasutitermes takasagoensis (1KSC),10 that only contain the catalytic domain. The second group includes the structures of Cel9A from Alicyclobacillus acidocaldarius (AaCel9A) (3EZ8),11 CelD from C. thermocellum (CtCelD) (1CLC),12 and cellobiohydrolase CbhA from C. thermocellum (CtCbhA) (1UT9),13 that contain an N-terminal immunoglobulin-like (Ig-like) domain besides the catalytic domain. The third group includes the structures of Cel9G from C. cellulolyticum (1G87)14 and endo/exocellulase E4 from Thermomonospora fusca (1TF4)15 that contain a C-terminal family 3 carbohydrate-binding module (CBM3) besides the catalytic domain. The structures of the catalytic domains of these GH family 9 enzymes are characterized by the (α/α)6-barrel fold with three acidic active site residues (two aspartate and one glutamate residues). These two aspartate residues activate the water molecule that acts as a nucleophile by deprotonating it, whereas the glutamate residue acts as a general acid (proton donor).16 These aspartate residues bind to the catalytic water molecule, in such a way that they share this water molecule. CtCbhA contains N-terminal CBM4, X11, and X12 modules, CBM3, and a dockerin module, in addition to the Ig-like and catalytic domains. However, the CtCbhA derivative containing only the Ig-like and catalytic domains is enzymatically active and the crystal structure of CtCbhA has been determined using this derivative.13 It has been reported for this derivative that deletion of the Ig-like domain inactivates the enzyme.17 However, the role of the Ig-like domain remains to be fully understood.

A novel GH family 9 enzyme, termed leaf-branch compost (LC)-CelG, has been isolated from LC of EXPO Park, Japan, using a metagenomic approach.18 LC-CelG is composed of 577 amino acid residues and contains a putative signal peptide (Residues 1–19) at the N-terminus. LC-CelG without this signal peptide consists of an N-terminal Ig-like domain (Residues 20–132) and a C-terminal catalytic domain (Residues 133–577). It shows the highest amino acid sequence identity of 42% to GH family 9 enzyme from Microcoleus sp. PCC 7113 (accession No. K9WM66). It shows relatively low amino acid sequence identities to CtCelD (31%), AaCel9A (31%), and CtCbhA (Ig-like and catalytic domains; 29%), for which the crystal structures are available. Therefore, it would be informative to examine whether LC-CelG has a similar structure to those of other GH family 9 enzymes and loses activity by removal of the Ig-like domain.

In this study, we overproduced LC-CelG in Escherichia coli, and purified and characterized it. We showed LC-CelG is a thermostable endoglucanase. We determined its crystal structure. This structure resembles those of other GH family 9 enzymes. We constructed the LC-CelG derivatives lacking the Ig-like domain or the interactions between the Ig-like and catalytic domains and characterized them. Based on these results, we discuss the role of the Ig-like domain.

Results and Discussion

Amino acid sequence of LC-CelG

The amino acid sequence of LC-CelG without a putative signal peptide (Residues 20–577) is compared with those of CtCelD, AaCel9A, and CtCbhA, for which the crystal structures are available, in Figure 1. CtCelD, AaCel9A, and CtCbhA are composed of 649, 537, and 813 amino acid residues, respectively. The three acidic active site residues of GH family 9 enzymes are conserved as Asp194, Asp197, and Glu558 in LC-CelG. These residues are conserved as Asp198, Asp201, and Glu555 in CtCelD, Asp143, Asp146, and Glu515 in AaCel9A, and Asp383, Asp386, and Glu795 in CtCbhA. According to the cocrystal structure of AaCel9A with the substrate (cellotetraose),19 Asp143, Asp146, Tyr150, Phe221, Gly298, Tyr300, Trp343, Ile400, Trp401, His461, Arg463, His485, Gln487, Tyr511, Glu515, and Tyr519 form the substrate binding pocket. All of these residues, except for Phe221, Ile400, and Tyr519 are conserved in LC-CelG. Phe221, Ile400, and Tyr519 are replaced by Trp270, His450, and Asp562 in LC-CelG, respectively. These results suggest that the substrate binding mechanism and catalytic mechanism of LC-CelG are similar to those of AaCel9A.

Figure 1.

Figure 1

Open in a new tab

Alignment of amino acid sequences of LC-CelG, CtCelD, AaCel9A, and CtCbhA. The accession numbers are KF626654 for LC-CelG, CAA28255 for CtCelD, ACV59481 for AaCel9A, and ABN51651 for CtCbhA. The amino acid sequence of LC-CelG without a putative signal peptide (Residues 20–577) and the corresponding regions of other three proteins are shown. The amino acid residues, which are conserved in all four proteins, are denoted with white letters and highlighted in black. The amino acid residues, which are conserved in two or three different proteins, are highlighted in gray. The ranges of the secondary structures of LC-CelG (βa–βh strands for the Ig-like domain, and β1–β6 strands and α1–α13 helices for the cellulase domain) are shown above its sequence based on its crystal structure. The residues that form the catalytic site (Asp194, Asp197, and Glu558), substrate binding site, Zn site (Cys150, Asp166, His167, and His193), Ca1 site (Asp225, Glu232, Asn235, Val237, Asp239, and Asp242), and Ca2 site (Asp351, Asp353, Asp356, Asp357, and Val395) are indicated by “*,” “+,” “z,” “1,” and “2,” respectively, above the LC-CelG sequence. The residues that form the Ca3 site of CtCelD and AaCel9A are indicated by “3” below the sequences. The position, at which N-terminal Ig-like domain of LC-CelG is truncated to construct His-ΔIg-CelG, is shown by solid arrow head above the LC-CelG sequence. Likewise, Gln40 and Asp99 of LC-CelG, which are mutated in this study, are indicated by open arrow heads.

The number of the metal binding sites varies for different proteins. It is the highest for CtCelD, which has one zinc binding site (Zn site) and three calcium binding sites (Ca1–Ca3 sites). Of these sites, only the Ca2 site is conserved in AaCel9A and CtCbhA. The Zn and Ca3 sites are only conserved in AaCel9A, whereas the Ca1 site is only conserved in CtCbhA. As a result, AaCel9A has the Zn, Ca2, and Ca3 sites, and CtCbhA has the Ca1 and Ca2 sites. The Zn site is formed by Cys155, Cys173, His174, and His197 in CtCelD, and Cys104, Cys121, His122, and His142 in AaCel9A. Corresponding residues in LC-CelG are Cys150, Asp166, His167, and His193, suggesting that the Zn site is conserved in LC-CelG. The Ca1 site is formed by water, Glu236, Asn239, Ile241, Asp243, and Asp246 in CtCelD, and Asp421, Glu428, Asn431, Tyr433, Asp435, and Asp438 in CtCbhA. The corresponding residues in LC-CelG are Asp225, Glu232, Asn235, Val237, Asp239, and Asp242, suggesting that the Ca1 site is conserved in LC-CelG. The Ca2 site is formed by Thr356, Ser358, Asp361, Asp362, and Asp401 in CtCelD, Asp302, Glu304, Asp307, Glu308, and Ala344 in AaCel9A, and Asp557, Tyr559, Asp562, Asp563, and Gly617 in CtCbhA. The corresponding residues in LC-CelG are Asp351, Asp353, Asp356, Asp357, and Val395, suggesting that the Ca2 site is conserved in LC-CelG. The Ca3 site is formed by Ser465, Asp468, and Val470 in CtCelD, and Ser520, Asp523, and Ile525 in AaCel9A. The corresponding residues in LC-CelG are Val514, Ser517, and Val519, suggesting that the Ca3 site is not conserved in LC-CelG.

Overproduction and purification of LC-CelG and His-LC-CelG

For characterization and structural analysis, LC-CelG without a signal peptide (Residues 20–577) was overproduced in E. coli either in a non-His-tagged or a His-tagged form. LC-CelG in a non-His-tagged form with Met at the N-terminus is simply designated as LC-CelG, whereas LC-CelG with a His-tag at the N-terminus is designated as His-LC-CelG. On induction for overproduction, LC-CelG and His-LC-CelG accumulated in E. coli cells in a soluble form. Both proteins were purified to give a single band on sodium dodesyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) (data not shown). The amount of the protein purified from 1 L culture was typically 3 mg for LC-CelG and 4 mg for His-LC-CelG. The N-terminal amino acid sequence of LC-CelG was determined to be Met-Leu-Ala-Gly-, indicating that LC-CelG contains the entire region of LC-CelG without a signal peptide. The molecular mass of LC-CelG was estimated to be 60 kDa by gel filtration chromatography. This value is comparable to the calculated one (62.6 kDa), indicating that LC-CelG exists as a monomer.

Activity of His-LC-CelG

To examine whether LC-CelG is an endoglucanase like CtCelD and AaCel9A or a cellobiohydrolase like CtCbhA, hydrolyses of CM-celluloase, and p-nitrophenyl cellobioside with His-LC-CelG were analyzed at 50°C in 100 mM sodium phosphate (pH 7.0). The concentrations of the enzyme, CM-cellulose, and p-nitrophenyl cellobioside were 0.01 mg mL−1, 1% (w/v), and 1 mM, respectively. His-LC-CelG hydrolyzed CM-cellulose, but did not hydrolyze p-nitrophenyl cellobioside (data not shown), indicating that LC-CelG is an endoglucanase.

The pH dependence of the enzymatic activity of His-LC-CelG was analyzed by measuring the activity at 60°C and various pHs ranging from 4.0 to 10.0 using CM-cellulose as a substrate. As shown in Figure 2(A), His-LC-CelG exhibited >80% activity at a broad pH range (pH 5.0–9.0). The temperature dependence of the enzymatic activity of His-LC-CelG was also analyzed by measuring the activity at pH 7.0 and various temperatures ranging from 40 to 90°C using CM-cellulose as a substrate. As shown in Figure 2(B), His-LC-CelG exhibited the highest activity at 70°C. The specific activity of His-LC-CelG determined at 70°C and pH 7.0 was 50.1 ± 0.3 units mg−1. It has been reported that AaCel9A exhibits maximal activity at pH 5.5 and 70°C and its specific activity in this condition is approximately 80 units mg−1.20 This result suggests that His-LC-CelG is slightly less active than AaCel9A and as stable as AaCel9A. However, AaCel9A exhibits approximately 50% of the maximal activity at pH 4.5 and 7.0, suggesting that the pH range suitable for activity of His-LC-CelG is broader than that of AaCel9A.

Figure 2.

Figure 2

Open in a new tab

Optimum pH and temperature for activity of His-LC-CelG. The pH (A) and temperature (B) dependencies of the enzymatic activity of His-LC-CelG are shown. The activity was determined at 60°C at the pH indicated (A) or at pH 7.0 and the temperatures indicated (B) using 1% (w/v) CM-cellulose as a substrate, as described in Materials and Methods. The buffers used to analyze the pH dependence of the activity were 100 mM sodium citrate (pH 4.0–6.5), 100 mM sodium phosphate (pH 6.0–8.0), and 100 mM Glycine-NaOH (pH 8.0–10.0). The experiment was performed at least twice, and errors from the average values are indicated by vertical lines.

It is noted that the pH and temperature dependencies of LC-CelG were similar to those of His-LC-CelG (data not shown), indicating that attachment of an N-terminal His-tag does not significantly affect the activity of LC-CelG.

Stability of His-LC-CelG

To analyze the stability of His-LC-CelG, thermal denaturation of this protein was analyzed at pH 7.0 in the presence of 5 mM CaCl2 by monitoring the change in circular dichroism (CD) values at 222 nm. Thermal denaturation of this protein was irreversible in this condition. However, the thermal denaturation curve of this protein was reproducible, unless the protein concentration, the pH, and the rate of the temperature increase (scan rate) were significantly changed. The thermal denaturation curve of His-LC-CelG is shown in Figure 3. The midpoint of the transition of this curve, T1/2, is 81.4°C (Table I). The denaturation temperature of AaCel9A has been reported to be 77.9°C,17 indicating that His-LC-CelG is more stable than AaCel9A by 3.5°C. The T1/2 value of His-LC-CelG is higher than its optimum temperature for activity (70°C), probably because the active site is locally denatured before the overall structure is denatured on thermal denaturation.

Figure 3.

Figure 3

Open in a new tab

Thermal denaturation curves of His-LC-CelG and its derivatives. The thermal denaturation curves of His-LC-CelG (thin solid line), His-ΔIg-CelG (thick solid line), His-Q40A-CelG (thin dashed line), His-D99A-CelG (thick dotted line), and His-Q40A/D99A-CelG (thick dashed line) are shown. These curves were obtained in the presence of 5 mM CaCl2 at pH 7.0 by monitoring the change in CD values at 222 nm as described in Materials and Methods.

Table I.

Activities and Stabilities of His-LC-CelG and its Mutants

Protein Specific activitya (Units mg−1) Relative activityb (%) T1/2c (°C) ΔT1/2d (°C)
His-LC-CelG 34.4 ± 0.4 100 81.4 ± 0.2 –
His-Q40A-CelG 32.2 ± 0.9 94 81.4 ± 0.2 0
His-D99A-CelG 36.3 ± 1.2 106 79.5 ± 0.1 −1.9
His-Q40A/D99A-CelG 33.2 ± 0.1 97 76.4 ± 0.2 −5.0
His-ΔIg-CelG 0.4 ± 0.2 1.1 75.1 ± 0.1 −6.3
Open in a new tab
a

The specific activity was determined at pH 7.0 and 50°C using CM-cellulose as a substrate as described in Materials and Methods. Each experiment was performed at least twice and the average value is shown together with the error.

b

The relative activity was calculated by dividing the specific activity of the protein by that of His-LC-CelG.

c

The temperature of the midpoint of the thermal denaturation transition, T1/2, was determined from the thermal denaturation curves shown in Figure 3.

d

ΔT1/2 = T1/2 determined—81.4°C.

It is noted that the T1/2 value of LC-CelG was comparable to that of His-LC-CelG (data not shown), suggesting that attachment of a His-tag does not significantly affect the stability of LC-CelG.

Crystal structure of LC-CelG

To examine whether the structure of LC-CelG is similar to those of other GH family 9 enzymes, the crystal structure of LC-CelG was determined at 2.15 Å resolution. The asymmetric unit consists of two molecules (A and B). Because LC-CelG exists as a monomer in solution, the intermolecular interaction observed in the crystals is probably artifact of the crystal packing. For both molecules, the electron density for N-terminal methionine residue and a part of the Ig-like domain (Leu20-Asp29) is not observed, probably due to structural disorder. The structures of molecules A and B are highly similar to each other with a root-mean-square deviation (RMSD) value of 0.30 Å for 548 Cα atoms. We used the structure of molecule A in this study.

The overall structure of LC-CelG is shown in Figure 4(A), in comparison with the structure of CtCelD (PDB code 1CLC) as a representative of the structures of GH family 9 enzymes. The structure of LC-CelG resembles that of CtCelD with a RMSD value of 1.58 Å for 482 Cα atoms. It also resembles those of AaCelA and CtCbhA with RMSD values of 1.49 Å for 490 Cα atoms and 1.92 Å for 503 Cα atoms, respectively. The structure of the catalytic domain belongs to a typical (α/α)6-barrel fold, which contains 12 long α-helices forming the central (α/α)6-barrel, two short α-helices, and six antiparallel strands forming two β-sheets. The Ig-like domain consists of eight antiparallel strands forming two β-sheets. The steric configurations of the three acidic catalytic residues (Asp194, Asp197, and Glu558) are nearly identical to those of CtCelD [Fig. 4(A)], AaCel9A [Fig. 4(B)], and CtCbhA (figure not shown). Asp194 and Asp197 form hydrogen bonds with a water molecule that may function as a nucleophile. The distance between Asp194 or Asp197 and this water molecule is 2.6 or 2.7 Å, respectively.

Figure 4.

Figure 4

Open in a new tab

Crystal structure of LC-CelG. (A) A stereo view of the overall structure of LC-CelG. The structure of LC-CelG is superimposed on that of CtCelD (PDB code 1CLC). For the LC-CelG structure, the Ig-like and catalytic domains are colored cyan and green, respectively. One zinc and two calcium ions (Ca1 and Ca2) are shown as orange and yellow spheres, with the numbers of the calcium ions indicated. Three active site residues (Asp194, Asp197, and Glu558) are indicated by green stick models, in which the oxygen and nitrogen atoms are colored red and blue, respectively. The entire CtCelD structure, including one zinc and three calcium ions (Ca1–Ca3), is colored gray. Three active site residues (Asp198, Asp201, and Glu555) are indicated by gray stick models. They are labeled in parentheses. (B) The structure around the substrate binding pocket. The structure around the substrate binding pocket of LC-CelG is superimposed on those of AaCel9A (PDB code 3H3K) and CtCbhA (PDB code 1RQ5) in complex with cellotetraose. The residues forming the substrate binding pocket of LC-CelG are indicated by green stick models, the corresponding residues and cellotetraose in the AaCel9A structure are indicated by gray stick models, and the corresponding residues and cellotetraose in the CtCbhA structure are indicated by yellow stick models. In these stick models, which are labeled with the same colors, the oxygen and nitrogen atoms are colored red and blue, respectively. The six subsites are labeled from −4 to +2. The water molecule is shown as red sphere. Dashed lines represent hydrogen bonds. (C–E) Electron density around the binding sites of zinc ion (C), and calcium ions Ca1 (D) and Ca2 (E). The zinc and calcium ions are shown as orange and yellow spheres, respectively. The water molecule is shown as red sphere. The residues co-ordinated with these metal ions are indicated as shown in (B). In (C) and (E), the 2Fo–Fc maps contoured at the 2.0σ and 4.0σ levels are shown in magenta and blue, respectively. In (D), the 2Fo–Fc map contoured at the 1.5σ level is shown in orange. In these figures, the Fo–Fc omit maps contoured at the 5.0σ level are also shown in red. Numbers represent the coordinate bond lengths (Å). (F) The structure around Gln40 and Asp99. Gln40 and Asp99 located in the Ig-like domain are shown as cyan stick models. The residues that are located in the catalytic domain and form hydrogen bonds with Gln40 (Ala494, Ser503, and As504) or salt bridge with Asp99 (Arg545) are shown by green stick models. In these stick models, the oxygen and nitrogen atoms are colored red and blue, respectively.

Substrate binding pocket

The structure of LC-CelG around the substrate binding pocket is compared with those of AaCel9A (PDB code 3H3K) and CtCbhA (PDB code 1RQ5) in complex with cellotetraose in Figure 4(B). The structures of the AaCel9A-cellotetraose and CtCbhA-cellotetraose complexes represent those of the enzyme-product and enzyme-substrate complexes, respectively, because cellotetraose binds to subsites −4 to −1 in the AaCel9A-cellotetraose structure, whereas it binds to subsites −2 to +2 in the CtCbhA-cellotetraose structure. The AaCel9A-cellotetraose structure also contains a glucose, which binds to subsite +1. Subsites −2 to +1 identified in the AaCel9A-cellotetraose structure are nearly identical to those identified in the CtCbhA-cellotetraose structure. The steric configurations of the amino acid residues that form the substrate binding pocket of LC-CelG highly resembles those of AsCel9A and CtCbhA, suggesting that these residues interact with substrate or product like those of AaCel9A and CtCbhA. Namely, Tyr201 OH, Gly347 N, Tyr349 OH, Glu558 Oɛ1, and Asp562 Oδ1 may form hydrogen bonds with glucosyl unit in subsite −1, Glc(−1). Trp394 Nɛ1 may form hydrogen bond with Glc(−2) and Trp451 may facilitate its binding by stacking interaction. His450 and His532 may facilitate binding of Glc(−4) by stacking interactions. His510 and Arg512 may form hydrogen bonds with Glc(+1). Trp270 and Tyr554 may facilitate its binding by stacking interactions. However, it remains to be determined whether Gln534 Nɛ forms hydrogen bond with Glc(−3) as does Gln487 Nɛ of AaCel9A, because Gln534 Nɛ is located far from Gln487 Nɛ.

According to the structure of the AaCel9A-cellotetraose complex, Asp143, and Asp146, which correspond to Asp194 and Asp197 of LC-CelG, respectively, form hydrogen bonds with the O1 atom of Glc(−1). However, the cellotetraose-free structure of this protein indicates that a water molecule, which may function as a nucleophile, binds to the position where this O1 atom binds.19 This O1 atom is not present when Glc(−1) and Glc(+1) are covalently linked as shown in the CtCbhA-cellotetraose structure but is present when the glycoside bond between Glc(−1) and Glc(+1) is cleaved as shown in the AaCel9A-cellotetraose structure. This result suggests that the catalytic water molecule bound to Asp194 and Asp197 of LC-CelG keeps binding as the O1 atom of Glc(−1) after cleavage of the C1=;O4 bond between Glc(−1) and Glc(+1).

Metal binding sites

Comparison of the amino acid sequence of LC-CelG with those of CtCelD, AaCel9A, and CtCbhA suggests that LC-CelG has the Zn, Ca1, and Ca2 sites (Fig. 1). The 2Fo–Fc maps of LC-CelG at the regions corresponding to the Zn, Ca1, and Ca2 sites of CtCelD are shown in Figure 4(C–E), respectively. Because no metal ions are present in the crystallization conditions of the protein and the coordination geometry of zinc ion is different from that of calcium ion, these peaks could be identified as the zinc and calcium (Ca1 and Ca2) ions. No clear density peak was detected at the region corresponding to the Ca3 site of CtCelD, indicating that LC-CelG does not have this site. The Fo–Fc omit maps of LC-CelG around the Zn, Ca1, and Ca2 sites are also shown in Figure 4(C–E), respectively. These maps show clear peaks, indicating that the metal ions are present in the Zn, Ca1, and Ca2 sites. The zinc ion is tetracoordinated with Cys150 SH, Asp166 Oδ2, His167 Nδ1, and His193 Nε2 with the coordinate bond lengths of 2.1, 2.0, 1.9, and 2.1 Å, respectively, in the Zn site [Fig. 4(C)]. The calcium ions are hexacoordinated with Asp225 Oδ1, Glu232 O, Asn 235 Oδ, Val237 O, Asp239 Oδ1, and Asp242 Oδ2 with the coordinate bond lengths of 2.4, 2.7, 2.6, 2.2, 2.7, and 2.5 Å, respectively, in the Ca1 site [Fig. 4(D)], and heptacoordinated with Asp351 Oδ1, Asp351 Oδ2, Asp353 O, Asp356 Oδ1, Asp357 Oδ2, Val395 O, and one water molecule with the coordinate bond lengths of 2.4, 2.3, 2.3, 2.7, 2.2, 2.4, and 2.5 Å, respectively, in the Ca2 site [Fig. 4(E)]. The electron density peak of the calcium ion in the Ca1 site is weaker than those of other two metal ions, probably because the calcium ion binds to the Ca1 site weakly and is partly dissociated from this site when the protein was dialyzed against metal-free buffer for crystallization and biochemical characterization. However, the enzymatic activity determined in the presence of 10 mM CaCl2 was comparable to that determined in the absence of metal ions (data not shown), suggesting that binding of the calcium ion to the Ca1 site does not significantly affect the activity. The zinc and calcium ions probably bind to the Zn and Ca2 sites too tightly to be dissociated from these sites on dialysis against metal-free buffer.

Interface between the Ig-like and catalytic domains

The interface between the Ig-like and catalytic domains of LC-CelG is formed by many hydrophilic and hydrophobic interactions, like those of CtCelD, AaCel9A, and CtCbhA. However, hydrogen bonds formed between Gln40 in the Ig-like domain and the loop between α11- and α12-helices in the catalytic domain are the only interactions conserved in AaCel9A and CtCbhA. Gln40 is conserved as Gln13 in AaCel9A and Gln218 in CtCbhA. It is replaced by Ser58 in CtCelD. However, Ser58 forms hydrogen bonds with the loop between α11- and α12-helices as does Gln40 of LC-CelG. The hydrogen bonds formed between the side chain of Gln40 and the loop between α11- and α12-helices are shown in Figure 4(F). The distances between Gln40 Nε and Ala494 O, Gln40 Oε, and Ser503 N, and Gln40 Oε and Asn504 N are 2.9, 3.0, and 2.9 Å, respectively. In addition, the salt bridge between Asp99 and Arg545, which are located in the Ig-like and catalytic domains, respectively, is conserved in CtCelD and CtCbhA. This salt bridge is shown in Figure 4(F). The corresponding salt bridge is formed between Glu103 and Lys542 in CtCelD and between Asp267 and Lys782 in CtCbhA. This salt bridge is not conserved in AaCel9A, because AaCel9A lacks the loop containing the acidic partner of this salt bridge (Fig. 1).

Importance of Ig-like domain and interactions between Ig-like and catalytic domains for activity and stability of LC-CelG

The role of the Ig-like domain remains to be fully understood. It has been reported that removal of this domain of CtCbhA results in a complete loss of enzymatic activity.17 It has also been reported that removal of one (Asp264-Tyr676) or two (Thr230-Gly661 and Asp262-Gly661) hydrogen bonds between the Ig-like and catalytic domains destabilize CtCbhA by 8.7 and 6.0°C, respectively, without significantly affecting the activity.17 However, it remains to be determined whether removal of the Ig-like domain inactivates other GH family 9 enzymes as well. Likewise, it remains to be determined whether removal of other interactions between the Ig-like and catalytic domains, which are conserved in GH family 9 enzymes, destabilizes the protein without significantly affecting its enzymatic activity. The hydrogen bonds between Thr230 and Gly661, Asp262 and Gly661, and Asp264 and Tyr676 in CtCbhA are not conserved in LC-CelG.

To examine whether removal of the Ig-like domain affects the activity and stability of His-LC-CelG, His-ΔIg-CelG, in which the Ig-like domain of His-LC-CelG is removed, was constructed. Likewise, to examine whether removal of Gln40-mediated hydrogen bonding interactions and salt bridge between Asp99 and Arg545 affects the activity and stability of His-LC-CelG, His-Q40A-CelG, His-D99A-CelG, and His-Q40A/D99A-CelG, in which Gln40, Asp99, and both of these residues of His-LC-CelG are replaced by Ala, respectively, were constructed. All mutant proteins were purified to give a single band on SDS-PAGE (data not shown). The production levels of these proteins in E. coli cells were comparable to that of His-LC-CelG. However, His-ΔIg-CelG accumulated in E. coli cells mostly in an insoluble form and only partly in a soluble form, whereas other proteins accumulated in E. coli cells in a soluble form. The proteins were purified from those accumulated in E. coli cells in a soluble form. As a result, the amount of the protein purified from 1 L culture was typically 0.3 mg for His-ΔIg-CelG and 4 mg for other proteins, which is comparable to that of His-LC-CelG.

The far-UV CD spectra of all mutant proteins measured at pH 7.0 and 25°C are shown in comparison with that of His-LC-CelG in Figure 5. The spectra of all mutant proteins, except for that of His-ΔIg-CelG, are nearly identical to that of His-LC-CelG, indicating that the mutations at Gln40 and Asp99 do not significantly affect the structure of His-LC-CelG. The spectrum of His-ΔIg-CelG is similar to that of His-LC-CelG in shape, but is different from that of His-LC-CelG in depth of a broad trough at 210–220 nm. The trough of His-ΔIg-CelG is deeper than that of His-LC-CelG, suggesting that the helical content of His-ΔIg-CelG is higher than that of His-LC-CelG. This result is consistent with the helical contents of His-LC-CelG (36.6%) and its catalytic domain (45.8%) calculated from the crystal structure of LC-CelG, suggesting that removal of the Ig-like domain does not significantly alter the structure of the catalytic domain of His-LC-CelG.

Figure 5.

Figure 5

Open in a new tab

CD spectra of His-LC-CelG and its derivatives. The far-UV CD spectra of His-LC-CelG (thin solid line), His-ΔIg-CelG (thick solid line), His-Q40A-CelG (thin dashed line), His-D99A-CelG (thick dotted line), and His-Q40A/D99A-CelG (thick dashed line) were measured at pH 7.0 and 25°C, as described in Materials and Methods.

The enzymatic activities of the mutant proteins were determined at pH 7.0 and 50°C using CM-cellulose as a substrate. His-ΔIg-CelG exhibited 1% of the activity of His-LC-CelG, whereas other mutant proteins exhibited comparable activities to that of His-LC-CelG (Table I). His-ΔIg-CelG exhibited poor activity even at 37°C. These results indicate that removal of the Ig-like domain greatly reduces the activity of His-LC-CelG, whereas mutations at Gln40 and Asp99 do not significantly affect its activity.

The stability of the mutant proteins was analyzed at pH 7.0 in the presence of 5 mM CaCl2 by monitoring the change in CD values at 222 nm as the temperature was increased. Thermal denaturation of these proteins was irreversible in the condition examined. However, the thermal denaturation curves of these proteins were reproducible, unless the protein concentration, the pH, and the rate of the temperature increase were significantly changed. The thermal denaturation curves of the mutant proteins are shown in Figure 3. The T1/2 values determined from these thermal denaturation curves are summarized in Table I. The T1/2 value of His-ΔIg-CelG is lower than that of His-LC-CelG by 6.3°C. The T1/2 value of His-Q40A/D99A-CelG is also lower than that of His-LC-CelG by 5.0°C. The difference in T1/2 values between His-Q40A/D99A-CelG and His-LC-CelG accounts for 80% of that between His-ΔIg-CelG and His-LC-CelG. This result suggests that the Ig-like domain contributes to the stabilization of His-LC-CelG mainly due to Gln40- and Asp99-mediated interactions. However, the T1/2 values of His-Q40A-CelG and His-D99A-CelG are identical to and only 1.9°C lower than that of His-LC-CelG, respectively. These results suggest that neither Gln40-mediated interactions nor Asp99-mediated interaction significantly contribute to the stabilization of His-LC-CelG, but the interactions at both sites co-operatively contribute to it.

Possible role of Ig-like domain

Removal of the Ig-like domain reduces the activity and stability of His-LC-CelG by 100-fold and 6.3°C, respectively (Table I). Similar result has been reported for CtCbhA.17 Removal of this domain completely inactivates CtCbhA and destabilizes it by 10.3°C. Removal of the Ig-like domain of His-LC-CelG may result in the loss of multiple interactions between the Ig-like and catalytic domains and thereby affect the conformation of the active site, as proposed for CtCbhA.17 Because a long loop between α12- and α13-helices containing one of the catalytic residues, Glu558, and a number of the residues that form the substrate binding pocket directly contact the Ig-like domain, the conformation of this loop may be changed by removal of the Ig-like domain. Thus, the Ig-like domain is probably required to make the conformation of the active site and substrate binding pocket functional. Molecular dynamics simulation of AaCel9A has suggested that residues within the Ig-like domain are dynamically correlated with residues in the substrate binding pocket and catalytic site.21 A possibility that the Ig-like domain is directly involved in substrate binding has been excluded, because CtCbhA has an ability to hydrolyze p-nitrophenyl cellobioside and completely loses this ability by removal of the Ig-like domain.17 The cellobioside portion of this substrate binds to the subsites −1 and −2, which are located far from the Ig-like domain. Double mutations at Gln40 and Asp99 that destabilize His-LC-CelG by 5.0°C do not significantly affect its activity (Table I), probably because these mutations are not enough to change the conformation of the active site and substrate binding pocket.

The finding that the Ig-like domain contributes to the stabilization of His-LC-CelG and CtCbhA suggests that these enzymes acquire this domain to adapt to high-temperature environment. However, thermostable GH family 9 enzymes do not always contain this domain, because GH family 9 enzymes without an Ig-like domain have been isolated from various organisms that adapt to low, moderate, and high temperature environments.22–24

It has been reported that removal of the Ig-like domain greatly reduces the expression level of AaCel9A in E. coli cells, probably because AaCel9A lacking this domain is not correctly folded.20 Similar result was obtained in this study. His-ΔIg-CelG accumulated in E. coli cells mostly in an insoluble form, probably because removal of the Ig-like domain affects folding of His-LC-CelG. His-LC-CelG accumulated in E. coli cells in a soluble form assumes a native-like structure as revealed by CD spectroscopy. However, this protein exhibits very weak activity. These results suggest that the Ig-like domain of GH family 9 enzyme is probably required to prevent aggregation of the catalytic domain due to exposure of interface between the Ig-like and catalytic domains and/or its incorrect folding.

Materials and Methods

Construction of plasmids

For construction of plasmid pET-LC-CelG and pET-His-LC-CelG used to overproduce LC-CelG without a signal peptide (Residues 20–577) with Met and His-tag at the N-terminus, respectively, the gene encoding LC-CelG was amplified by polymerase chain reaction (PCR) using the fosmid vector harboring the pre-LC-CelG gene, which was previously constructed,18 as a template. The sequences of the PCR primers were 5′-GGACATATGCTGGCCGGATCCGTCC-3′ for 5′-primer and 5′-TCGGAATTCCGCCCTCACAGCGAC-3′ for 3′-primer, where underlines represent the NdeI site for 5′-primer and EcoRI site for 3′-primer. The resultant DNA fragment was digested with NdeI and EcoRI, and ligated into the NdeI-EcoRI sites of pET25b and pET28a (Novagen, Madison, WI) to generate plasmid pET-LC-CelG and pET-His-LC-CelG, respectively.

Plasmid pET-His-ΔIg-CelG used to overproduce His-ΔIg-CelG was constructed using KOD-plus-Mutagenesis kit (Toyobo Co., Osaka, Japan). The sequences of the primers for inverse PCR were 5′-GTCTACCGCGACGTGCTCTACGCGG-3′ for the forward primer and 5′-CATATGGCTGCCGCGCGGCACCAGG-3′ for the reverse primer. Plasmid pET-His-LC-CelG was used as a template. Plasmids pET-His-Q40A-CelG, pET-His-D99A-CelG, and pET-His-Q40A/D99A-CelG used to overproduce His-Q40A-CelG, His-D99A-CelG, and His-Q40A/D99A-CelG, respectively, were also constructed using KOD-plus-Mutagenesis kit (Toyobo). The primers for inverse PCR were designed in such a way that the CAG codon for Gln40 and GAC codon for Asp99 are changed to the GCC codon for Ala. Plasmid pET-His-LC-CelG was used as a template.

PCR was performed with Gene Amp PCR system 2400 (Applied Biosystems, Tokyo, Japan) using KOD DNA polymerase (Toyobo) according to the protocol by the supplier. PCR primers were synthesized by Hokkaido System Science (Sapporo, Japan). The nucleotide sequence was confirmed by an ABI Prism 3100 DNA sequencer (Applied Biosystems).

Overproduction and purification

Overproduction of LC-CelG and its derivatives using E. coli BL21(DE3) transformants was performed at 37°C in NZCYM medium. When the absorbance of the culture at 600 nm reached around 0.5, isopropyl-β-d-thio galactopyranoside was added to the culture medium and cultivation was continued for an additional 4 h. Cells were harvested by centrifugation at 8000g for 10 min, suspended in 10 mM Tris-HCl (pH 8.0) containing 1 mM DTT, disrupted by sonication lysis, and centrifuged at 30,000g for 30 min. The supernatant was collected, and the protein in the supernatant was precipitated by the addition of ammonium sulfate to 50% saturation. Subsequently, the pellet was dissolved in 10 mM Tris-HCl (pH 8.5) containing 1 mM DTT, 1 mM CaCl2, and 1 mM ZnCl2, and dialyzed against 10 mM Tris-HCl (pH 8.5) containing 10 mM 2-mercaptoethanol and 1mM CaCl2 at 4°C overnight. This solution was then incubated at 60°C for 1 h for heat treatment and centrifuged at 15,000g for 10 min. The supernatant was collected and subjected to different purification procedures for non-His-tagged and His-tagged proteins. For purification of the non-His-tagged protein, the supernatant was loaded onto a HiTrap Q HP column (GE Healthcare, Tokyo, Japan) equilibrated with 10 mM Tris-HCl (pH 8.5). The protein was eluted from the column by linearly increasing the NaCl concentration from 0 to 1M. The fractions containing the protein were collected, dialyzed against 10 mM Tris-HCl (pH 8.0) containing 50 mM NaCl, and loaded onto a Hi-Load 16/60 Superdex 200 pg column (GE Healthcare) equilibrated with the same buffer. The fractions containing the protein were collected and used for further studies. For purification of the His-tagged protein, the supernatant was dialyzed against 10 mM Tris-HCl (pH 8.0) containing 1 mM DTT, 10 mM imidazole, and 0.3M NaCl, and loaded onto a Ni Sepharose 6 Fast Flow column (GE Healthcare) equilibrated with the same buffer. The protein was eluted from the column by linearly increasing the imidazole concentration from 10 to 300 mM. The fractions containing the protein were collected, dialyzed against 10 mM Tris-HCl (pH 8.0) and used for further studies.

Determination of enzymatic activity

The enzymatic activity was determined at the temperatures indicated by the dinitrosalicylic acid (DNS) stopped method25 using CM-cellulose as a substrate. The reaction mixture (100 µL) contained 100 mM sodium phosphate (pH 7.0) and 1% (w/v) CM-cellulose (low viscosity grade, Sigma-Aldrich Co., St. Louis, MO). The enzymatic reaction was initiated by adding an appropriate amount of the enzyme and terminated by adding 10 µL of 10% SDS and boiling for 5 min. The reaction time was 10 min. The resultant solution was mixed with 300 µL of the DNS solution prepared as described previously,25 boiled for 5 min, and cooled on ice. After centrifugation at 17,000g for 5 min, an aliquot of the supernatant (100 µL) was withdrawn, diluted twice by distilled water, and measured for absorption at 500 nm (A500). The amount of reducing sugars released from the substrate was determined using glucose as a standard. One unit of enzymatic activity was defined as the amount of enzyme that produced 1 µmol of reducing sugars per min.

Measurement of CD spectra

The far-UV (200–260 nm) CD spectrum of the protein was measured at 25°C on a J-725 spectropolarimeter (Japan Spectroscopic, Tokyo, Japan). The protein was dissolved in 10 mM Tris-HCl (pH 7.0). The protein concentration was 0.1 mg mL−1 and a cell with an optical path length of 2 mm was used. The mean residue ellipticity, [θ], which has the units of deg cm2 dmol−1, was calculated using an average amino acid molecular mass of 110 Da.

Thermal denaturation

The thermal denaturation curve of the protein was obtained by monitoring the change in CD values at 222 nm as the temperature was increased. The protein was dissolved in 10 mM Tris-HCl (pH 7.0) containing 5 mM CaCl2. The protein concentration and optical path length were 0.1 mg mL−1 and 2 mm, respectively. The rate of temperature increase was 1.0°C min−1. Thermal denaturation of the proteins was irreversible in the condition examined. The temperature of the midpoint of the transition, T1/2, was calculated by curve fitting of the resultant CD values versus temperature data on the basis of a least-square analysis.

Crystallization

Prior to crystallization, LC-CelG was dialyzed against 10 mM Tris-HCl (pH 8.0) and concentrated to 10.3 mg mL−1 using an ultrafiltration system Centricon (Millipore, Billerica, MA). The crystallization condition was screened using crystallization kits from Hampton Research (Alise Viejo, CA) (Crystal Screen I and II) and Emerald Biostructures (Bainbridge Island, WA; Wizard I, II, III, and IV) by the sitting-drop vapor-diffusion method at 4 and 20°C. Drops were prepared by mixing the protein solution with the reservoir solution at equal volume (1 µL) and were vapor-equilibrated against 100 µL reservoir solution. Cuboid-shaped single crystals of LC-CelG appeared after a few months in Crystal Screen I No. 6 [0.1M Tris-HCl (pH 8.5), 0.2M magnesium chloride hexahydrate and 30% (w/v) polyethylene glycol 4000] at 20°C.

X-ray diffraction data collection and structure determination

X-ray diffraction dataset of LC-CelG were collected at −173°C without cryoprotectant using a wavelength of 0.9 Å with the beam line BL44XU of SPring-8 (Hyogo, Japan). Diffraction dataset was indexed, integrated, and scaled using the HKL2000 program suite.26 The structure was determined by the molecular replacement method using MOLREP27 in the CCP4 program suite.28 The crystal structure of CtCelD [PDB code 1CLC] at 1.9 Å resolution was used as a starting model. Automatic model building was done using ArpWarp.29 Structural refinement was performed using REFMAC30 of the CCP4 program and the model was corrected using COOT.31 The statistics for data collection and refinement are summarized in Table II. The figures were prepared using PyMol (http://www.pymol.org).

Table II.

Data Collection and Refinement Statistics of LC-CelG

 Crystal LC-CelG
Wavelength (Å) 0.900
Space group P212121
Cell parameters
 a, b, c (Å) 84.631, 89.913, 151.157
 α = β = γ (°) 90.00
Molecules/asymmetric unit 2
Resolution range (Å) 50.00–2.15 (2.19–2.15)a
Reflections measured 941,321
Unique reflections 63,355
Redundancy 14.9 (15.0)
Completeness (%) 99.9 (100.0)a
Rmerge (%)b 13.3 (68.7)a
Average I/σ (I) 28.0 (2.6)a
Refinement statistics
 Resolution limits (Å) 77.28–2.15
No. of atoms
 Protein/water 4239/410
 Rwork (%)/Rfree (%)c 18.7/24.6
Rms deviations from ideal values
 Bond lengths (Å) 0.011
 Bond angles (°) 1.190
Average B factors (Å2)
 MonomerA/B 47.6/39.6
 Zn2+/Ca2+1/Ca2+2 in monomer A 40.6/79.3/46.0
 Zn2+/Ca2+1/Ca2+2 in monomer B 34.3/56.8/33.6
 Water 44.3
Ramachandran plot statistics
 Preferred regions (%) 95.3
 Allowed regions (%) 4.5
Open in a new tab
a

Values in parentheses are for the highest-resolution shell.

b

Rmerge = ∑|Ihkl−<Ihkl>|/∑ Ihkl, where Ihkl is an intensity measurement for reflection with indices hkl and <Ihkl> is the mean intensity for multiply recorded reflections.

c

Free R-value was calculated using 5% of the total reflections chosen randomly and omitted from refinement.

Protein data bank accession number

The coordinates and structure factors for LC-CelG have been deposited in the PDB under ID code 3X17.

Acknowledgments

The synchrotron radiation experiments were performed at Osaka University beam line BL44XU at SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2014A6915). We thank Dr. Y. Koga for helpful discussions. The authors do not have a conflict of interest to declare.

References

  1. Wilson DB. Cellulases and biofuels. Curr Opin Biotechnol. 2009;20:295–299. doi: 10.1016/j.copbio.2009.05.007. [DOI] [PubMed] [Google Scholar]
  2. Lynd LR, van Zyl WH, McBride JE, Laser M. Consolidated bioprocessing of cellulosic biomass: an update. Curr Opin Biotechnol. 2005;16:577–583. doi: 10.1016/j.copbio.2005.08.009. [DOI] [PubMed] [Google Scholar]
  3. la Grange DC, den Haan R, van Zyl WH. Engineering cellulolytic ability into bioprocessing organisms. Appl Microbiol Biotechnol. 2010;87:1195–1208. doi: 10.1007/s00253-010-2660-x. [DOI] [PubMed] [Google Scholar]
  4. Yanase S, Yamada R, Kaneko S, Noda H, Hasunuma T, Tanaka T, Ogino C, Fukuda H, Kondo A. Ethanol production from cellulosic materials using cellulase-expressing yeast. Biotechnol J. 2010;5:449–455. doi: 10.1002/biot.200900291. [DOI] [PubMed] [Google Scholar]
  5. Yamada R, Nakatani Y, Ogino C, Kondo A. Efficient direct ethanol production from cellulose by cellulase- and cellodextrin transporter-co-expressing Saccharomyces cerevisiae. AMB Express. 2013;3:34. doi: 10.1186/2191-0855-3-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Brethauer S, Wyman CE. Review: continuous hydrolysis and fermentation for cellulosic ethanol production. Bioresour Technol. 2010;101:4862–4874. doi: 10.1016/j.biortech.2009.11.009. [DOI] [PubMed] [Google Scholar]
  7. Parsiegla G, Belaïch A, Belaïch JP, Haser R. Crystal structure of the cellulase Cel9M enlightens structure/function relationships of the variable catalytic modules in glycoside hydrolases. Biochemistry. 2002;41:11134–11142. doi: 10.1021/bi025816m. [DOI] [PubMed] [Google Scholar]
  8. Kesavulu MM, Tsai JY, Lee HL, Liang PH, Hsiao CD. Structure of the catalytic domain of the Clostridium thermocellum cellulase CelT. Acta Crystallogr D Biol Crystallogr. 2012;68:310–320. doi: 10.1107/S0907444912001990. [DOI] [PubMed] [Google Scholar]
  9. Arimori T, Ito A, Nakazawa M, Ueda M, Tamada T. Crystal structure of endo-1,4-β-glucanase from Eisenia fetida. J Synchrotron Radiat. 2013;20:884–889. doi: 10.1107/S0909049513021110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Khademi S, Guarino LA, Watanabe H, Tokuda G, Meyer EF. Structure of an endoglucanase from termite, Nasutitermes takasagoensis. Acta Crystallogr D Biol Crystallogr. 2002;58:653–659. doi: 10.1107/s0907444902002366. [DOI] [PubMed] [Google Scholar]
  11. Pereira JH, Sapra R, Volponi JV, Kozina CL, Simmons B, Adams PD. Structure of endoglucanase Cel9A from the thermoacidophilic Alicyclobacillus acidocaldarius. Acta Crystallogr D Biol Crystallogr. 2009;65:744–750. doi: 10.1107/S0907444909012773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chauvaux S, Souchon H, Alzari PM, Chariot P, Beguin P. Structural and functional analysis of the metal-binding sites of Clostridium thermocellum endoglucanase CelD. J Biol Chem. 1995;270:9757–9762. doi: 10.1074/jbc.270.17.9757. [DOI] [PubMed] [Google Scholar]
  13. Schubot FD, Kataeva IA, Chang J, Shah AK, Ljungdahl LG, Rose JP, Wang BC. Structural basis for the exocellulase activity of the cellobiohydrolase CbhA from Clostridium thermocellum. Biochemistry. 2004;43:1163–1170. doi: 10.1021/bi030202i. [DOI] [PubMed] [Google Scholar]
  14. Mandelman D, Belaich A, Belaich JP, Aghajari N, Driguez H, Haser R. X-Ray crystal structure of the multidomain endoglucanase Cel9G from Clostridium cellulolyticum complexed with natural and synthetic cello-oligosaccharides. J Bacteriol. 2003;185:4127–4135. doi: 10.1128/JB.185.14.4127-4135.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Sakon J, Irwin D, Wilson DB, Karplus PA. Structure and mechanism of endo/exocellulase E4 from Thermomonospora fusca. Nat Struct Biol. 1997;4:810–818. doi: 10.1038/nsb1097-810. [DOI] [PubMed] [Google Scholar]
  16. Zhou W, Irwin DC, Escovar-Kousen J, Wilson DB. Kinetic studies of Thermobifida fusca Cel9A active site mutant enzymes. Biochemistry. 2004;43:9655–9663. doi: 10.1021/bi049394n. [DOI] [PubMed] [Google Scholar]
  17. Kataeva IA, Uversky VN, Brewer JM, Schubot F, Rose JP, Wang BC, Ljungdahl LG. Interactions between immunoglobulin-like and catalytic modules in Clostridium thermocellum cellulosomal cellobiohydrolase CbhA. Protein Eng Des Sel. 2004;17:759–769. doi: 10.1093/protein/gzh094. [DOI] [PubMed] [Google Scholar]
  18. Okano H, Ozaki M, Kanaya E, Kim JJ, Angkawidjaja C, Koga Y, Kanaya S. Structure and stability of metagenome-derived glycoside hydrolase family 12 cellulase, LC-CelA, a homolog of Cel12A from Rhodothermus marinus. FEBS Open Bio. 2014;4:936–946. doi: 10.1016/j.fob.2014.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Eckert K, Vigouroux A, Lo Leggio L, Moréra S. Crystal structures of A. acidocaldarius endoglucanase Cel9A in complex with cello-oligosaccharides: strong −1 and −2 subsites mimic cellobiohydrolase activity. J Mol Biol. 2009;394:61–70. doi: 10.1016/j.jmb.2009.08.060. [DOI] [PubMed] [Google Scholar]
  20. Eckert K, Zielinski F, Lo Leggio L, Schneider E. Gene cloning, sequencing, and characterization of a family 9 endoglucanase (CelA) with an unusual pattern of activity from the thermoacidophile Alicyclobacillus acidocaldarius ATCC27009. Appl Microbiol Biotechnol. 2002;60:428–436. doi: 10.1007/s00253-002-1131-4. [DOI] [PubMed] [Google Scholar]
  21. Liu H, Pereira JH, Adams PD, Sapra R, Simmons BA, Sale KL. Molecular simulations provide new insights into the role of the accessory immunoglobulin-like domain of Cel9A. FEBS Lett. 2010;584:3431–3435. doi: 10.1016/j.febslet.2010.06.041. [DOI] [PubMed] [Google Scholar]
  22. Ueda M, Ito A, Nakazawa M, Miyatake K, Sakaguchi M, Inouye K. Cloning and expression of the cold-adapted endo-1,4-β-glucanase gene from Eisenia fetida. Carbohydr Polym. 2014;101:511–516. doi: 10.1016/j.carbpol.2013.09.057. [DOI] [PubMed] [Google Scholar]
  23. Kurokawa J, Hemjinda E, Arai T, Kimura T, Sakka K, Ohmiya K. Clostridium thermocellum cellulase CelT, a family 9 endoglucanase without an Ig-like domain or family 3c carbohydrate-binding module. Appl Microbiol Biotechnol. 2002;59:455–461. doi: 10.1007/s00253-002-1048-y. [DOI] [PubMed] [Google Scholar]
  24. Mingardon F, Bagert JD, Maisonnier C, Trudeau DL, Arnold FH. Comparison of family 9 cellulases from mesophilic and thermophilic bacteria. Appl Environ Microbiol. 2011;77:1436–1442. doi: 10.1128/AEM.01802-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Miller GL, Blum R, Glennon WE, Burton AL. Measurement of carboxymethylcellulase activity. Anal Biochem. 1960;2:127–132. [Google Scholar]
  26. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997;276:307–326. doi: 10.1016/S0076-6879(97)76066-X. [DOI] [PubMed] [Google Scholar]
  27. Vagin A, Teplyakov A. Molecular replacement with MOLREP. Acta Crystallogr D Biol Crystallogr. 2010;66:22–25. doi: 10.1107/S0907444909042589. [DOI] [PubMed] [Google Scholar]
  28. Collaborative Computational Project, Number 4. The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr. 1994;50:760–763. doi: 10.1107/S0907444994003112. [DOI] [PubMed] [Google Scholar]
  29. Langer G, Cohen SX, Lamzin VS, Perrakis A. Automated macromolecular model building for X-ray crystallography using ARP/wARP version 7. Nat Protoc. 2008;3:1171–1179. doi: 10.1038/nprot.2008.91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr. 1997;53:240–255. doi: 10.1107/S0907444996012255. [DOI] [PubMed] [Google Scholar]
  31. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr. 2004;60:2126–2132. doi: 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES