Cloning and characterization of a novel cellobiase gene, cba3, encoding the first known β-glucosidase of glycoside hydrolase family 1 of Cellulomonas biazotea

Gene 493 (2012) 52–61

Contents lists available at SciVerse ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Cloning and characterization of a novel cellobiase gene, cba3, encoding the first known β-glucosidase of glycoside hydrolase family 1 of Cellulomonas biazotea Anthony K.N. Chan a, Yule Y. Wang a, K.L. Ng a, Zhibiao Fu b, W.K.R. Wong a,⁎ a b

Division of Life Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China GlaxoSmithKline, Microbial and Cell Culture Development, 709 Swedeland Rd., King of Prussia, PA 19406, USA

a r t i c l e

i n f o

Article history: Accepted 15 November 2011 Available online 27 November 2011 Received by A.J. van Wijnen Keywords: Cellulomonas biazotea Escherichia coli Glycoside hydrolase family 1 (GH1) Glycoside hydrolase family 3 (GH3) Leaderless cellobiase Non-classical protein secretion

a b s t r a c t A novel cellobiase gene, designated cba3, was cloned from Cellulomonas biazotea. Although cellobiase genes of C. biazotea were previously cloned, published and/or patented, they encoded β-glucosidases all belonging to glycoside hydrolase family 3 (GH3); the new Cba3 cellobiase was identified to be a glycoside hydrolase family 1 (GH1) member, which represents the first discovered GH1 β-glucosidase of C. biazotea. Escherichia coli transformants expressing recombinant Cba3 were shown to grow readily in minimal media using cellobiose as the sole carbon source, supporting the conclusion that Cba3 is a genuine cellobiase. The full-length cba3 gene was revealed by sequencing to be 1344 bp long. Cba3 deletants lacking either the N-terminal 10 amino acids or the C-terminal 10 residues were found to be biologically inactive, supporting the importance of both ends in catalysis. Like other GH1 β-glucosidases, Cba3 was shown to contain the highly conserved NEP and ENG motifs, which are crucial for enzymatic activity. Despite lacking a classical N-terminal signal peptide, Cba3 was demonstrated to be a secretory protein. The findings that Cba3 is a cellobiase, and that it was expressed well as an extracellular protein in E. coli, support the potential of Cba3 for use with other cellulases in the hydrolysis of cellulosic biomass. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Enzymatic hydrolysis of biomass has been a vigorous research topic arousing global interest as a major goal is the development of a costeffective process to produce ethanol, a useful biofuel, utilizing cellulosic residues of municipal, forestry, and agricultural origins as substrates (Ragauskas et al., 2006). To achieve the goal, a number of approaches involving the use of cellulase saccharification in conjunction with fermentation (Öhgren et al., 2007), fermenting organisms expressing transgenic cellulase genes (Wong et al., 1988), and modified cellulolytic microbial strains with enhanced fermentative ability have been Abbreviations: aa, amino acid(s); Ap, ampicillin; bp, base pair(s); Bgl, β-glucosidase (s); Cel, cellobiase(s); C. biazotea, Cellulomonas biazotea; CIAP, calf intestinal alkaline phosphatase; E. coli, Escherichia coli; Eng, endoglucanase; Exg, exoglucanase; GH1, glycoside hydrolase family 1; GH3, glycoside hydrolase family 3; hr, hour(s); IPTG, isopropyl β-D-thiogalactopyranoside; kb, kilobase pair(s); kDa, kilodalton(s); MUG, 4methylumbelliferyl β-D-glucopyranoside; MUGase, enzyme capable of hydrolyzing MUG; NCBI, National Centre for Biotechnology Information; min, minute(s); pNP, pnitrophenol; nt, nucleotide(s); ORF, open reading frame; PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; pNPG, p-nitrophenyl β-Dglucopyranoside; pNPGase, enzyme capable of hydrolyzing pNPG; RBS, ribosomal binding site; rpm, revolutions per minute. ⁎ Corresponding author at: Division of Life Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China. Tel.: + 852 23587299; fax: + 852 23581552. E-mail address: [email protected] (W.K.R. Wong). 0378-1119/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2011.11.027

contrived (Blanchard et al., 2010). However, regardless of the approach employed, the largest stumbling block is the availability of a practical enzymatic process for cellulose hydrolysis (Merino and Cherry, 2007), which requires a cooperative action among three different types of cellulase: endoglucanase (Eng; EC 3.2.1.4), exoglucanase (Exg; EC 3.2.1.91), and cellobiase (Cel; EC 3.2.1.21). In addition, the hydrolysis is expected to be operated on a large scale, and thus cost-effective provision of cellulases exhibiting high specific activities is imperative. The application of recombinant DNA technology has enabled the cloning and characterization of a large number of cellulase genes from a variety of cellulolytic microorganisms since early 1980s. The availability of host systems for the export of recombinant cellulases into the culture medium facilitates not only the production of the enzymes, but also the studies of their reconstitution and interaction. Although Eng and Exg have been commonly expressed as extracellular enzymes (Fu et al., 2006; Lam et al., 1997, 1998, Wang et al., 2010b, 2011; Wong and Lam, 2000; Wong et al., 1988), the attainment of Cel as secreted or excreted products is not so common. The low frequency is attributable to the fact that cellulolytic microbes are commonly found to produce cell-bound Cel (Bhat and Bhat, 1997; Wakarchuk et al., 1984), thus hindering the progress of developing recombinant strains expressing extracellular Cel. As a result, the deficiency of Cel activity in cellulase preparations presents a major cause of cellobiose inhibition in cellulose saccharification (Mosier et al., 1999).

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Our laboratory has been involved in research and application of microbial cellulases for a long time (Fu et al., 2005, 2006; Gilkes et al., 1984; Lam et al., 1997, 1998; Lau and Wong, 2001; Skipper et al., 1985; Wang et al., 2010b, 2011; Warren et al., 1986; Wong and Chan, 1998; Wong and Lam, 2000; Wong and Sutherland, 1997; Wong et al., 1986, 1988, 1998). Over these years, we have cloned and expressed the cex and cenA genes from Cellulomonas fimi, which encode an Exg and an Eng, respectively, in a wide range of host systems (Fu et al., 2006; Lam et al., 1997, 1998; Wang et al., 2010b, 2011; Wong and Lam, 2000; Wong and Sutherland, 1997; Wong et al., 1988). To search for Cel that might work co-operatively with Exg and Eng, we have put our focus on Cellulomonas biazotea due to the reasons that i) it secretes high levels of Cel activity to the culture medium (Lau and Wong, 2001; Saratale et al., 2010; Wong and Chan, 1998; Wong et al., 1998), and ii) it is closely related to C. fimi (Wood and McCrae Sheila, 1979). Our attempt to demystify the Cel complex of C. biazotea has resulted in the characterization of a couple of secretory Cel: Cba (Wong and Chan, 1998; Wong et al., 1998) and Cba2 (Lau and Wong, 2001), which will be employed with other Cel, as well as Exg and Eng of C. fimi for reconstitution studies. Concurrently, we have also engineered a number of bacterial systems for efficient extracellular production of heterologous proteins (Wong and Lam, 2000; Wong and Sutherland, 1997). All these research results may pave the way for the development of a costeffective process for the production of recombinant cellulases for various industrial applications. In this communication, we report the cloning and characterization of a novel Cel gene, cba3, from C. biazotea. Unexpectedly, the Cba3 enzyme was identified to be the first representative of glycoside hydrolase family 1 (GH1) in C. biazotea, whereas the previously characterized Cba (Wong and Chan, 1998; Wong et al., 1998) (GenBank ID: AAC38196.1) and Cba2 (Lau and Wong, 2001; Okura et al., 2009) (GenBank ID: DM144958.1) enzymes were shown to be members of glycoside hydrolase family 3 (GH3). Furthermore, based on the comparison of Cba3 with Cba, Cba2 and β-glucosidases of other microorganisms, we discuss the analysis of the physical and enzymatic properties of Cba3. Lastly, despite the absence of a classical secretion signal peptide, we provide evidence to support that Cba3 is a bona fide secretory protein.

53

(Ipswich, MA, USA). All PCR and sequencing primers (Table 1) were purchased from Invitrogen (Carlsbad, CA). The Phusion® High-Fidelity DNA polymerase was purchased from Finnzymes Oy (Keilaranta, Finland). PCR was carried out with the help of the Bioline BioTaq PCR™ kit (London, UK), otherwise specified. All other chemicals were purchased from Sigma-Aldrich Corporation (St. Louis, Mo, USA) unless otherwise specified. 2.3. Construction of cloning and expression vectors The cloning and expression vector, pM, has been previously described (Wong et al., 1998). A pM derivative, designated pMB, was constructed to facilitate the cloning experiments described below by the insertion of a BamHI restriction site. In brief, PCR was first amplified with primers KBtac1 and Hterm (Table 1), using vector pM as the template. The PCR product and pM were then restricted with KpnI and PstI. The digested product and the larger fragment of pM were purified, and ligated to form the pMB vector (Fig. 1A). Plasmid DNA was prepared using the alkaline lysis method described previously (Sambrook and Russel, 2001). 2.4. Construction of the C. biazotea DNA library

The C. biazotea ATCC 486 strain from which the cba (Wong and Chan, 1998; Wong et al., 1998) and cba3 (this study) genes were cloned was obtained from the American Type Culture Collection (Rockville, MD, USA). Escherichia coli JM101, which was employed as the host for the cloning, expression and maintenance of plasmids, has been previously described (Sambrook and Russel, 2001).

C. biazotea genomic DNA was prepared from an overnight culture grown in 2× YT medium supplemented with 0.2% (w/v) glucose essentially the same as described previously (Wong et al., 1998). In brief, the overnight culture (500 ml) was pelleted, and resuspended in 57 ml of TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0). The cell suspension was treated with 6 ml of 10% SDS, 6 mg of proteinase K and 240 mg of lysozyme, followed by incubation at 37 °C for 3 hr. The cell lysate was extracted with an equal volume of phenol:chloroform (1:1) (pH 8.0), and centrifuged at 13,000 ×g and 4 °C for 30 min. Genomic DNA in the upper aqueous layer was precipitated by adding sodium acetate (pH 5.2) to a final concentration of 5 mM and an equal volume of isopropyl alcohol, followed by centrifugation at 13,000 ×g and 4 °C for 30 min. The DNA pellet was washed once with 25 ml of 70% ethanol, dried and resuspended in 8 ml of TE buffer. The genomic DNA was further purified by CsCl equilibrium density gradient ultracentrifugation as described previously (Sambrook and Russel, 2001). The purified genomic DNA of C. biazotea was partially restricted with PstI as described previously (Wong et al., 1998). The partially digested fragments were resolved on a low-melting-point agarose gel. DNA fragments with sizes between 0.6 kb and 23 kb were retrieved and ligated with the pM vector linearized with PstI. The ligation mixture was transformed into E. coli JM101 cells using the CaCl2 method (Sambrook and Russel, 2001). The transformants were first selected on 2× YT plates supplemented with Ap, and then screened on MUG plates. The Bgl+ clones were kept for further characterization.

2.2. Media, chemicals and growth conditions

2.5. Molecular cloning of the cba3 gene and its derivatives

Both C. biazotea and E. coli strains were cultured in 2× YT medium at 30 °C and 250 rpm. For selection of transformants, the medium was supplemented with 100 μg ml − 1 of ampicillin (Ap). When solid medium was required, Bacto agar was added to a final concentration of 1.5% (w/v). The 4-methylumbelliferyl β-D-glucopyranoside (MUG) agar plates for selecting β-glucosidase (Bgl) positive clones contained at a final concentration of 1 mM MUG, 0.1 mM isopropyl β-Dthiogalactopyranoside (IPTG), and 100 μg ml− 1 of Ap. Cellobiose minimal medium was prepared by adding 1% (w/v) cellobiose and 0.1 mM IPTG to the M9 medium (Ausubel et al., 2001), in which glucose had been omitted. Tryptone, yeast extract and Bacto agar were purchased from Oxoid Limited (Hampshire, UK). IPTG was purchased from USB Corporation (Cleveland, USA). The Quick-Stick DNA ligase was purchased from Bioline (London, UK). Other DNA modifying enzymes and restriction enzymes were purchased from New England Biolabs

One recombinant construct, designated pM9178, was shown to contain a 4.3-kb PstI insert expressing Bgl activity (Sections 3.1 and 3.2). The 4.3-kb PstI insert was first subcloned into the PstI site of the pMB vector to generate construct pMB9178, which was then employed to facilitate subsequent cloning experiments using the unique BamHI and XhoI sites (Fig. 1A). To generate construct pM9178dSS (Fig. 1A), plasmid pM9178 was completely digested with StuI and SmaI, followed by retention and self-religation of the largest 6.2-kb restriction fragment. To generate the pMBcba3-P construct, a PCR product, designated pRBS, was first amplified with primers KBtac2 and RBSrev2 (Table 1), using vector pMB as the template. Another PCR product, designated cba3-X, was then amplified with primers cba3-0 and XhoI-rev (Table 1), using construct pM9178 as the template. The pRBS and cba3-X products were then subjected to overlapping extension PCR using KBtac2 and XhoI-rev (Table 1) as the primers to obtain an intermediate, pcba3-X.

2. Material and methods 2.1. Bacterial strains

54

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Table 1 Primers employed in this study. Primer

Orientation

Sequence (5′ to 3′)a

Cloning and mutagenesis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

KBtac1 Hterm KBtac2 RBSrev2 cba3-0 cba3-33nts XhoI-rev XhoI-f P-cba3-TGA cba3dC10aab Cba3E169A-b Cba3E351A-b Cba3E169A-f Cba3E351A-f PstI-cba3b

Forward Reverse Forward Reverse Forward Forward Reverse Forward Reverse Reverse Reverse Reverse Forward Forward Reverse

GGggtaccGGATCCTTACTCCCCATCCCCCTGTTGA CCAAGCTTAAAAAAAAGCCCGCTCAT GGGGTACCggatccTTACTCC CATAATTTTTTCCTCCTGTGTGAAATTGTTAT CACAGGAGGAAAAAATTATGCAGACCGGTCCGAGCCA CACAGGAGGAAAAAATTATGGGCTTCCTGTTCGGTGCCAG CCGTCctcgagCGCCTGC GCAGGCGctcgagGACGG AAActgcagTCACGCCGTCGTCCCGCG TTTctgcagTCAGTCGGCGTACCAGTCGAACGA CGCGTTCACCGGGCACCAGTG CGCCGTGATCAGGATCGGCGG GGTGCCCGGTGAACGCGCCGAACGTCGTCACGCTC CCGATCCTGATCACGGCGAACGGCTGCTCGTACGGCA AAActgcagTCACGCCGTCGT

DNA sequencing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

uptacp RAUp pM9178f1 pM9178f2 pM9178f3 cba3F1 cba3F2 cba3F3 cba3F4 MFSf1 MFSf2 pTR MFSb cba3B1 cba3B2 cba3B3 cba3B4 cba3B5 cba3b6 cba3b7 cba3b8

Forward Forward Forward Forward Forward Forward Forward Forward Forward Forward Forward Reverse Reverse Reverse Reverse Reverse Reverse Reverse Reverse Reverse Reverse

GGCGGTGAAGGGCAATCA TTCACACAGGAGGAAAAAATTATG CAGCGCCCAGAACAGGTC CGGGGTCGGCGGTCGA GCAGGGTCTGGACGAGG TGAGCAGGCGGCCCATGGCT TACCGGTTCTCGATCGCGTG CGGTCCTGCCGCTGTCGGGC CCTCGACTCCCACCTGGGCG CTCAACCCGGTGTGGGGCGC GCGCTCCCCCGCGAG CTAATGCACCCAGTAAGGC GGCCGCGAGCGTCAG CCCAGGCGAGCACCATC GTGAAGCCGTCGGCCCA CCGGTTCCACAGCGCGT GGTTCGCCGGGCCGCT CCCGGGCCGCTCGTG TGACGGCGGTCCGCGA GAGCACTGCCCGCCC GGCGGGGTCGGGTG

a

Restriction sites used in the cloning experiments are shown in lowercase. Intended mutations in the constructs: pMBcba3E169A and pMBcba3E351A, are underlined.

Subsequent to restriction with BamHI and XhoI, the pcba3-X fragment was cloned into the pMB9178 plasmid restricted with the same two enzymes to result in construct pMBcba3-P (Fig. 1A). To generate the pMBdN10aacba3-P construct, a PCR product, designated dN10, was first amplified with primers cba3-33nt and XhoIrev (Table 1), using pM9178 as the template. Another PCR product, designated pdN10, was then generated by overlapping extension of products pRBS and dN10 with the help of primers KBtac2 and XhoIrev (Table 1). The resultant product, pdN10, was restricted with BamHI and XhoI, and then cloned into the pMB9178 plasmid restricted with the same two enzymes to obtain construct pMBdN10aacba3P (Figs. 1A and 2). To generate the pMBcba3 construct, a PCR product, designated Xcba3-P, was amplified with primers XhoI-f and P-cba3-TGA (Table 1), using pMBcba3-P as the template. The resultant product, X-cba3-P, was restricted with XhoI and PstI, and then cloned into the pMBcba3-P plasmid to obtain construct pMBcba3 (Fig. 1A). To generate the pMBcba3dC10aa construct, a PCR product, designated dC10, was first amplified with primers XhoI-f and cba3dC10aab (Table 1), using pMBcba3 as the template. The resultant product, dC10, was restricted with XhoI and PstI, and then cloned into the pMBcba3 plasmid restricted with the same two enzymes to obtain construct pMBcba3dC10aa (Figs. 1A and 2). To generate the pMBcba3E169A construct (using Phusion® HighFidelity DNA polymerase), a PCR product, designated 169a, was first amplified with primers XhoI-f and Cba3E169A-b (Table 1), using pMBcba3 as the template. Another PCR product, 169b, was then

amplified with primers Cba3E169A-f and PstI-cba3b (Table 1), using pMBcba3 as the template. By overlapping extension of products 169a and 169b, using primers XhoI-f and PstI-cba3b (Table 1), an intermediate product, 169c, was obtained. Subsequent to restriction with XhoI and PstI, the 169c fragment was cloned into the pMBcba3 plasmid to result in construct pMBcba3E169A (Fig. 1A). To generate construct, pMBcba3E351A (using Phusion® HighFidelity DNA polymerase), a PCR product, designated 351a, was first amplified with primers XhoI-f and Cba3E351A-b (Table 1), using pMBcba3 as the template. Another PCR product, 351b, was then amplified with primers Cba3E351A-f and PstI-cba3b (Table 1), using pMBcba3 as the template. By overlapping extension of products 351a and 351b, using primers XhoI-f and PstI-cba3b (Table 1), an intermediate product, 351c, was obtained. Subsequent to restriction with XhoI and PstI, the 351c fragment was cloned into the pMBcba3 plasmid to result in construct pMBcba3E351A (Fig. 1A). All of the constructs were confirmed by DNA sequencing with the help of specifically designed primers (Table 1). Sequencing reactions were performed using the ABI PRISM® BigDye® Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). The cycling reactions were performed employing the ERICOMP TwinBlockTm system (Ericomp Inc., San Diego, CA). 2.6. Assays for Bgl expression The MUG plate assay and the pNPG assay for the detection and the quantification of Bgl activity, respectively, were performed as

A.K.N. Chan et al. / Gene 493 (2012) 52–61

55

A

B

Fig. 1. Localization of the cba3 gene carried on plasmid pMB9178 and its derivatives. (A) Schematic representation of the various DNA constructs. The gray rectangular boxes represent the genomic inserts of C. biazotea, whereas the dotted line boxes denote deletions. All the boxes are drawn to scale. The black solid lines on both sides of the boxes represent either the pM or pMB vector; both vectors are 3.9 kb long (not to scale). The gray bar above the pMBcba3-P construct indicates the location of the cba3 gene. Restriction sites shown are: B = BamHI; K = KpnI; P, P1 and P2 = PstI; S1 to S5 = SmaI; X = XhoI. The ATG start codons and TGA stop codons of the cba3 gene and its deletants are indicated. The ATG codons of the cba3 gene and its deletants were fused precisely to the 3′ end of a sequence containing the E. coli consensus RBS (AGGAGGAAAAAATT; RBS underlined) carried on the pMB vector. The ompA leader sequence originally present in the pM/pMB vector was deleted in all the plasmid derivatives. The MUGase phenotypes of the plasmid clones are summarized on the right side. (B) Western blot analysis of culture lysates of C. biazotea and an E. coli transformant containing the pMBcba3 plasmid using antibodies directed against the Cba3 protein.

previously described (Lau and Wong, 2001; Wong et al., 1998). One unit of pNPGase activity is defined as the amount of enzyme that releases 1 μmol of p-nitrophenol per min at 37 °C. 2.7. Cellular fractionation Cellular fractionation of a culture resulted in any combination of the following three components: cytoplasmic fraction, periplasmic fluid and culture supernatant, depending upon the need of the investigation. The culture supernatant was prepared by spinning down the cells of a culture at 13,000 ×g for 5 min, followed by collecting the liquid medium. The periplasmic and cytoplasmic components were prepared according to the protocol described previously (Sambrook and Russel, 2001) with slight modifications. In brief, cells were pelleted from 1 ml of culture diluted to an OD550 of 2. The cell pellet was resuspended in 100 μl of freshly prepared cell lysis buffer [1 mg/ml lysozyme, 20% (w/v) sucrose, 30 mM Tris–Cl (pH 8.0), and 1 mM EDTA (pH 8.0)], followed by incubation on ice for 10 min. The cell mixture was centrifuged at 13,000 ×g for 1 min, and the supernatant collected was the periplasmic component. The remaining cell pellet was then resuspended in 100 μl of 0.1 M Tris–Cl (pH 8.0). The cell suspension was frozen at −80 °C for 5 min and thawed at 37 °C for 5 min, followed by vigorously vortexing. This freeze–thaw-vortex cycle was repeated twice. The cell mixture was then centrifuged at 13,000 ×g for 5 min, and the supernatant collected was the cytoplasmic fraction.

2.8. Antibodies for Western blot analysis Recombinant Cba3 containing a C-terminal 6× His tag was first expressed in E. coli, followed by purification using a commercial Co2+resin column according to the manufacturer's instructions (Clontech Laboratories, Inc., USA). The purified Cba3 was used to raise anti-Cba3 antibodies in rabbits as described previously (Fu et al., 2005). To prepare for an antiserum against the E. coli MazG protein, a GST-MazG fusion protein was first expressed in E. coli, followed by cleavage of the fusion protein with thrombin to obtain the MazG protein as instructed by the supplier (Amersham Bioscience AB, Uppsala, Sweden). The purified MazG was then used to raise anti-MazG antibodies as described (Fu et al., 2005). The antisera were employed as the primary antibodies in Western blot analysis as described previously (Fu et al., 2005). 3. Results 3.1. Shotgun cloning of DNA fragments encoding β-glucosidase activities from C. biazotea Agarose gel electrophoresis of genomic DNA fragments of C. biazotea partially restricted with PstI revealed a continuous DNA smear on the gel with the smallest products detected at the 0.6-kb region (data not shown). The result suggested that shotgun cloning of the partial PstI DNA digest could result in recombinant clones encoding various Cel components. In our first attempt in employing this cloning strategy, we identified a single Cel + clone comprising the 2.5-kb cba gene

56

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Fig. 2. Primary structure of the cba3 gene harbored in plasmid pM9178. Only the segment covering nucleotides 1700 to 3200 of the 4.3 kb insert of pM9178 is shown. The complete coding sequence of the cba3 gene is highlighted in gray color. The putative start codon, ATG, of the cba3 ORF is denoted by a thick arrow, whereas the deduced aa sequence of Cba3 accompanied by single-letter aa code is indicated. The putative RBS of cba3 and the conserved NEP and ENG motifs of Cba3 are underlined. Amino acid residues E169 and E351, presumably serving as an acid–base catalyst and an attacking nucleophile in the active site of Cba3, are indicated by a solid square and a solid triangle, respectively. The deletions in the pMBdN10aacba3-P and pMBcba3dC10aa constructs shown in Fig. 1A are denoted. The GenBank accession number of the cba3 gene sequence is “JF727823”.

among the 18 β-glucosidase-positive (Bgl +) transformants obtained (Wong and Chan, 1998; Wong et al., 1998). Despite the attainment of only one Cel gene, we believed that the approach was viable and that other Cel gene determinants would be obtainable if larger numbers of transformants were screened in future trials. In our next cloning exercise (reported herein) employing the same strategy and the pM vector described previously (Wong et al., 1998), altogether 19,091 E. coli transformants, which were 8 times as many as that obtained in the first trial (Wong et al., 1998), were screened using the MUG plate assay (Wong et al., 1998). This time, 134 Bgl + transformants were obtained. PstI analysis of their plasmid constructs revealed the presence of 7 different patterns, with one of them shown to be the same as that of pBZ4.7, which harbored the

2.5-kb cba gene and was the only Cel + clone obtained in the first cloning attempt (Wong et al., 1998). The much larger number of Bgl + transformants attained in this trial, as expected, resulted correspondingly in more, 3 in total, Cel + clones. The amino acid (aa) sequences deduced from the DNA sequencing results of the 3 Cel+ clones were used to compare with those of other cel genes published in the literature (Lau and Wong, 2001; Okura et al., 2009; Wong and Chan, 1998; Wong et al., 1998). The comparison revealed that majority of the 5 Cel+ clones belonged to GH3, one of the two families of Bgl enzymes classified according to aa similarities (Cantarel et al., 2009). However, one plasmid construct, designated pM9178, was found to encode a Cel enzyme with structural features, which were distinctly different from those of the GH3 members. The

A.K.N. Chan et al. / Gene 493 (2012) 52–61

unique structural and functional properties of this Cel, designated Cba3, as well as its gene determinant, cba3, were characterized in detail as reported below. 3.2. Localization of the cba3 gene Construct pM9178 was shown by electrophoresis to harbor a PstI insert with size of around 4.3 kb. Restriction analysis of pM9178 revealed that there were more than 3 SmaI sites (the exact number was shown by DNA sequencing to be 5) present at the leftward 2 kb of the 4.3-kb PstI insert (Fig. 1A). This 2-kb fragment appeared to contain essential structure of the cba3 gene as deletion of it from pM9178 with StuI and SmaI complete digestion, thus removing sequences found within the unique StuI site located on the pM vector and the fifth SmaI (S5) site present on the 4.3-kb insert (Fig. 1A). The deletion resulted in a complete inactivation of the cba3 gene in the smaller construct, pM9178dSS, to express Bgl activity (Fig. 1A). The result suggested the likeliness that part of the cba3 ORF, which appeared to be essential for the biological function of Cba3, was located at the rightward 2.3 kb of the insert carried on pM9178SS (Fig. 1A). Therefore, it was decided that this 2.3-kb fragment was fully sequenced. 3.3. Characterization of the cba3 gene Sequencing of the rightward 2.3-kb fragment on pM9178SS was facilitated using the universal primers: uptacp and RAUp (Table 1), which primed at a site just upstream from the StuI/fifth SmaI (S5) fusion point or the 5′ end of the truncated insert (Fig. 1A). From the first 400 bp determined, it was apparent that this entire 400-bp sequence belonged to part of an ORF, which was presumably the coding sequence of the cba3 gene. Sequencing of the rest of the rightward 2.3-kb insert on pM9178SS (Fig. 1A) revealed the residual nucleotide (nt) sequence of cba3. The ORF ended at a putative stop codon, TGA, which was about 1.3 kb downstream from the S5 site (Fig. 1A). The putative start codon, ATG, of the ORF was revealed to be located just about 30 nt upstream from the S5 site when the leftward 2-kb PstI (P1)/S5 sequence (Fig. 1A) was also determined. Two pieces of data support that this ATG is indeed the start codon of the cba3 gene. First, a putative Shine–Dalgarno sequence: 5′-GGAGG-3′ was found 8 nt upstream from this ATG codon (Fig. 2). Second, a deletion of 1835 bp from the P1 site to a spot 32 bp downstream from the A of the ATG codon resulted in a deletion derivative, pMBdN10aacba3-P (see Section 2.5 for more details of its construction), which was shown to express a Bgl − phenotype (Figs. 1A and 2). However, if the same deletion was allowed to cease just upstream of the ATG codon, thus deleting 1805 bp from the P1 site, the resulting deletant, pMBcba3-P, expressed a Bgl + phenotype (Fig. 1A). To localize the TGA codon of the cba3 gene, two other deletants: pMBcba3 and pMBcba3dC10aa (Figs. 1A and 2), were constructed. Deletion of DNA sequences downstream from the putative TGA stop codon of the cba3 gene in construct pMBcba3-P showed no detectable adverse effects on Bgl expression conferred by the new derivative, pMBcba3 (Fig. 1A). However, if the last 10 aa-coding codons of the cba3 gene in pMBcba3 were deleted (Fig. 2), the mutation would result in a defective deletant, pMBcba3dC10aa, expressing a Bgl − phenotype (Fig. 1A). The native and recombinant Cba3 products were then purified respectively from the cell lysates of the native host, C. biazotea, and an E. coli transformant harboring the pMBcba3 deletant, in which the cba3 gene was embraced precisely by the putative start (ATG) and stop (TGA) codons (Figs. 1A and 2). Western blot analysis showed that the two protein products apparently shared the same size (Fig. 1B), supporting the results of the deletion experiments that the ATG and TGA codons embracing the cba3 gene in construct pMBcba3 (Fig. 1A) represent the genuine start and stop codons, respectively, of the cba3 gene.

57

From the deletion analysis (Fig. 1A) and DNA sequence information (Fig. 2), the cba3 gene was shown to possess an ORF of 1341 bp, with its ATG start codon identified at 1806 nt of the PstI insert and its TGA stop codon located 1341 nt downstream from A of the start codon (Figs. 1A and 2). Therefore, the deduced Cba3 product is composed of 447 aa with a predicted Mr of 48,000. The cba3 gene exhibits a high GC content of 72.8%; such a high value is typical of Cellulomonas DNA. 3.4. Relationship of Cba3 with other Bgl enzymes There have been three C. biazotea Bgl enzymes reported so far, including Cba (Wong et al., 1998), Cba2 (Lau and Wong, 2001) and Cba3 (this paper). Despite their origin from the same bacterium, the three enzymes do not share much similarity in aa composition and structure. The aa sequences of them were then compared with those available in the CAZy database (Cantarel et al., 2009). Immediately, an unequivocal conclusion was drawn from the comparison: the three C. biazotea Bgl enzymes belong to two different glycosyl hydrolase families, in which Cba3 (GenBank ID: JF727823.1) is a member of family 1 (GH1), whereas Cba (GenBank ID: AAC38196.1) and Cba2 (GenBank ID: DM144958.1) belong to family 3 (GH3). The conclusion is well supported by the high degrees of similarity in aa sequences between Cba3 and other GH1 members (Table 2), and among Cba and other GH3 members (Table 3). The results of the comparison provide also the first evidence that the C. bizaotea Bgl complex consists of both GH1 and GH3 enzymes. From the comparison, it is also interesting to note that the GH1 enzymes (Table 2) are all significantly smaller in size (over 30%) than their GH3 counterparts (Table 3). The variation in molecular size among the three C. bizaotea Bgl enzymes also supports this observation. Cba3, which is shown to have a Mr of 48,000 (Fig. 2), as expected, is substantially smaller than the previously studied Cba (Wong et al., 1998) and Cba2 (Lau and Wong, 2001), both of which have a Mr larger than 85,000. 3.5. Catalytic properties of Cba3 Alignment of the aa sequence of Cba3 and those of other GH1 enzymes helps identify the essential aa residues involved in catalysis. Like other GH1 members, Cba3 was revealed to possess the highly conserved NEP and ENG motifs, which comprise residues 168–170 and 351–353, respectively, of the Cba3 sequence (Figs. 2 and 3). The Glu residues at positions 169 and 351 were shown to be essential for catalytic function in other GH1 members; Glu 169 might serve as an acid–base catalyst, whereas Glu 351 was presumably the attacking nucleophile (Moracci et al., 1996). To demonstrate that Glu 169 and Glu 351 in Cba3 were indeed critical for catalysis, these two residues were individually replaced by an Ala residue, thus resulting in two Cba3 derivatives, Cba3-Ala 169 and Cba3-Ala 351, which were encoded by plasmids pMBcba3E169A and pMBcba3E351A, respectively (Figs. 1A and 2). Despite just a single aa replacement, the catalytic function of both Cba3-Ala 169 and Cba3-Ala 351 was totally impaired, supporting the speculation that both Glu 169 and Glu 351 play an essentail role in the catalytic function of Cba3. In addition to the more common hydrolytic activities towards aryl-glucosides and β-galactosides (data not shown), Cba3 was also shown to possess the less common Cel activity. It was observed that E. coli JM101 transformant harboring the pMBcba3 construct (Fig. 1A), in which the non-cba3 gene sequences had been deleted, was able to grow in a minimal medium supplemented with cellobiose as the sole carbon and energy source (Fig. 4). A GH3 member, Cba, which was encoded by the cba gene previously cloned from C. biazotea (Wong et al., 1998), was also shown to be able to facilitate its E. coli host to utilize cellobiose as a carbon and energy source (Wong et al., 1998).

58

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Table 2 GH1 β-glucosidases that share high similarities with Cba3a.

1 2 3 4 5 6 7 8 9 10

Species

GenBank ID accession/versionc

Identities (%)a

Protein size (kDa)b

References

Saccharopolyspora erythraea Amycolatopsis mediterranei Streptomyces griseus Rhodococcus erythropolis Streptomyces avermitilis Rhodococcus equi Kitasatospora setae Streptomyces coelicolor Streptomyces griseus Arthrobacter aurescens

YP_001103497.1 YP_003762312.1 YP_001821883.1 YP_002766308.1 NP_826775.1 YP_004006815.1 BAJ26623.1 NP_626770.1 YP_001822943.1 YP_948659.1

60 58 60 55 57 57 60 55 56 54

47.0 48.7 49.8 50.5 48.7 52.3 48.7 49.2 50.4 52.8

Oliynyk et al., 2007 Zhao et al., 2010 Ohnishi et al., 2008 Sekine et al., 2006 Ikeda et al., 2003 Letek et al., 2010 Ichikawa et al., 2010 Bentley et al., 2002 Ohnishi et al., 2008 Mongodin et al., 2006

a The results were obtained by searching Cba3 (GenBank ID: JF727823.1) in the NCBI Blastp database with default parameters. A brief account of the ten most similar GH1 β-glucosidases is provided. b The protein sizes were predicted using the ProParam tool (http://au.expasy.org/tools/protparam.html) available in the ExPASy Proteomics Server. c The protein sequences (not shown) for the comparison were retrieved from NCBI, using their GenBank accession numbers.

Table 3 GH3 β-glucosidases that share high similarities with Cbaa.

1 2 3 4 5 6 7 8 9 10

Species

GenBank ID accession/versionc

Identities (%)a

Protein size (kDa)b

References

Arthrobacter aurescens Sorangium cellulosum Saccharopolyspora erythraea Streptomyces bingchenggensis Streptomyces ambofaciens Clavibacter michiganensis Clavibacter michiganensis Amycolatopsis mediterranei Kitasatospora setae Arthrobacter aurescens

YP_950314.1 YP_001617546.1 YP_001103532.1 ADI10567.1 CAJ88068.1 YP_001221782.1 YP_001709426.1 YP_003770711.1 BAJ26087.1 YP_946463.1

54 45 44 43 44 44 45 42 44 41

87.5 87.6 86.2 88.8 85.5 87.1 89.7 88.8 86.5 88.3

Mongodin et al., 2006 Schneiker et al., 2007 Oliynyk et al., 2007 Wang et al., 2010a Barona-Gómez et al., 2006 Gartemann et al., 2008 Bentley et al., 2008 Zhao et al., 2010 Ichikawa et al., 2010 Mongodin et al., 2006

a The results were obtained by searching Cba (GenBank ID: AAC38196.1) in the NCBI Blastp database with default parameters. A brief account of the ten most similar GH3 β-glucosidases is provided. b The protein sizes were predicted using the ProParam tool (http://au.expasy.org/tools/protparam.html) available in the ExPASy Proteomics Server. c The protein sequences (not shown) for the comparison were retrieved from NCBI, using their GenBank accession numbers.

3.6. Secretion of Cba3 The GH3 members of C. biazotea, Cba and Cba2, reported previously were both shown to be secretable proteins (Lau and Wong, 2001; Wong and Chan, 1998; Wong et al., 1998), which are infrequently found among Cel enzymes. Although sequencing of the cba2 gene (GenBank ID: DM144958.1) revealed that Cba2 carries an N-terminal signal peptide, interestingly, such a signal is apparently absent from Cba, of which the primary structure was deduced from the cba gene sequence (Wong et al., 1998). The results suggested that Cba2 and Cba might utilize different secretion pathways in C. biazotea. The above postulation gains support from the present study. Cba3, which was expressed as a secretory as well as an extracellular protein in E. coli (Fig. 5), was shown to be devoid of a signal peptide as deduced from the nt sequence of the cba3 gene (Fig. 2). An analysis of the various Bgl genes available in the literature (Tables 2 and 3) may prove that the unusual secretory Bgl enzymes such as Cba and Cba3, which do not possess a signal peptide for secretion, are more common than expected in nature. 4. Discussion Bgl enzymes of both GH1 and GH3 have been shown to be present in a number of microbial species; however, it was not known whether C. biazotea produced both families of enzymes until the identification of Cba3 in this study. Cba3, a new Bgl, or more specifically, a new Cel, is the first GH1 representative identified in C. biazotea. The availability of Cba3 has offered us a valuable opportunity to better understand the organization of the Cel complex, the possible interaction among different Cel components of the two families in substrate hydrolysis,

and the co-operation of the Cel members with other cellulase components in biomass degradation. It was noted that Bgl enzymes of GH1 (Table 2) were in general more than 30% smaller than their GH3 counterparts (Table 3). The newly identified Cba3, which has a size of 48 kDa, obeys the mentioned “rule” and is 40–50% smaller than its two homologous GH3 counterparts, Cba (Wong and Chan, 1998; Wong et al., 1998) and Cba2 (Lau and Wong, 2001). Despite the lack of supporting evidence, it is intriguing to hypothesize that the two families of Cel enzymes might have complementary roles to play in substrate hydrolysis. Considering the complexity of cellulosic biomass available in the natural environment, it is not difficult to envisage that the three types of cellulases produced by the cellulolytic microbes are highly diversified to ensure a thorough hydrolysis of the cellulose substrates. As far as Bgl are concerned, GH1 and GH3 together comprise a wide variety of enzymes exhibiting broad substrate specificities (Dodd et al., 2010; Hill and Reilly, 2008). Thus, it is tempting to postulate that the GH3 Cel, e.g. the formerly characterized Cba (Wong and Chan, 1998; Wong et al., 1998) and Cba2 (Lau and Wong, 2001), might work together with the GH1 components, e.g. the Cba3 enzyme introduced in this study, to facilitate C. biazotea to efficiently hydrolyse cellodextrins, of which the existence may impose strong product inhibition on the complex process of cellulose hydrolysis (Holtzapple et al., 1990). The comparison of Cba3 sequence with other Bgl sequences available in the CAZy database (Cantarel et al., 2009) facilitated the unequivocal classification of Cba3 as a GH1 enzyme. Further, based on the observation that E. coli transformants expressing Cba3 activity grew well in minimal media supplemented with cellobiose as the sole carbon source (Fig. 4), Cba3 was concluded to be a Cel. The availability of both GH1 and GH3 Cel may help us better understand their substrate specificities and catalytic activities as a whole. The presence of both families of enzymes may

A.K.N. Chan et al. / Gene 493 (2012) 52–61

59

Fig. 3. Amino acid sequence alignment of Cba3 and other GH1 Bgl enzymes. Cba3 is aligned with the top ten most similar Bgl (Table 2) using ClustalW (Thompson et al., 1994) and the alignment was analyzed and edited using Jalview (Waterhouse et al., 2009). The numbers flanking the sequences represent the aa positions of the corresponding sequences. Similar aa residues are shaded, the darker the highlight, the higher levels of the identity shown among the residues. The putative NEP and ENG motifs are boxed, and the predicted general acid/base catalyst (■) and catalytic nucleophile (▲) of GH1 Bgl, respectively, are indicated.

also help us clarify whether the size discrepancy observed between them has any relationship with their substrate specificities. On the other hand, it is possible to develop Cel enzymes with enhanced catalytic efficiencies (Scott et al., 2010) and improved stability (Hill et al., 2010) for cellulose hydrolysis through protein engineering. It is intriguing to note that both the N- and C-termini of Cba3 are critical for enzymatic activity (Fig. 1A). Therefore, an understanding of how the terminal deletions affect the biological function of Cba3 may shed light on the development of stable Cba3 derivatives. Moreover, the potential active site of Cba3 was readily identified by sequence comparison with other GH1 Bgl available in the database. More potent Cba3 derivatives may then be engineered through replacement of the conserved aa residues in the active site.

Not only was Cba3 identified to be a Cel, but it was also shown to be a secretory protein. Probably due to a low level of secretion of the native Cba3, the detection of it in the culture medium of C. biazotea has been unsuccessful. Nevertheless, the secretory nature of Cba3 was clearly demonstrated by the secretability of its recombinant counterpart in E. coli. Two lines of evidence support this conclusion. First, extracellular Cba3 activity was detected throughout the exponential growth of an E. coli transformant harboring construct pMBcba3, which contained the full-length cba3 gene (Fig. 5A and B). The result was substantiated by the detection of Cba3, but not the intracellular protein, MazG, in both the periplasmic and extracellular samples of an overnight culture of the transformant (Fig. 5C). Second, the same transformant was able to grow in a minimum medium supplemented with cellobiase as the sole carbon source (Fig. 4),

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Viable cell counts on ampicillin plates Extracellular pNPGase activitiy

9.6

2.5 2.0

CFU/ml (Log10)

Absorbance at 550 nm

A

pM pMBcba3

3.0

1.5 1.0

5 4

9.2

3 2

8.8

1

0.5 8.4

0 0

12

24

36

48

60

72

84

0

97 122 145

2

thus supporting the conclusion that Cba3 was exported to the supernatant to hydrolyse the disaccharide to an absorbable product, glucose. However, the absence of a secretion signal peptide on either the deduced Cba3 sequence (Fig. 2) or the native Cba3 product (reflected by the same molecular size shared between the two products; Fig. 1B) supports the idea that Cba3 secretion is through a non-classical route instead of the common Sec pathway. In fact, other secretory enzymes shown to be devoid of an N-terminal signal peptide have previously been identified (Bendtsen et al., 2005), and a homologous counterpart of Cba3, Cba of GH3, has also been shown to be secretory without employing an obvious signal peptide (Wong et al., 1998). Interestingly, another homologous GH3 member, Cba2, has been shown to be a standard secretory protein possessing a signal peptide (Lau and Wong, 2001; Okura et al, 2009) (GenBank ID: DM144958.1). Therefore, it appears that C. biazotea adopts at least two different modes of secretion for the export of Bgl enzymes. The fact that Cba2 exists as a major Bgl in the culture medium (Lau and Wong, 2001) supports the postulation that the classical Sec pathway may be employed as the primary route for protein secretion in C. biazotea. It will be intriguing to study whether Cba2 plays as a forerunner in the hydrolysis of cellodextrins, presumably the larger oligosaccharides. The resulting intermediates may then be processed by other secretory Cel components, e.g. Cba3, which are less efficiently exported through a non-classical pathway to complete the hydrolysis. Despite the detection of high levels of Cel activity in the culture medium during the log phase of growth of C. biazotea (Wong et al., 1998), only the Cel components equipped with a signal peptide, e.g. Cba2 (Lau and Wong, 2001) were readily secreted and detectable in the culture medium. Therefore, a practical problem for the use of native Cel enzymes for cellulose hydrolysis is the deficiency of those members, which employ a less efficient non-classical pathway for their delivery to the extracellular milieu. A much higher level of Cba3 obtainable from E. coli than from C. biazotea (Fig. 1B), however, supports the argument that recombinant DNA technology offers a viable approach for the production of even the non-classical secretory Cel enzymes. Our development of various efficient extracellular host systems (Fu et al., 2006; Lam et al., 1997, 1998; Wang et al., 2010b, 2011; Wong and Lam, 2000; Wong and Sutherland, 1997; Wong et al., 1988), on the one hand, may facilitate the production, thereby the understanding of the catalytic properties and substrate specialties of recombinant GH1 and GH3 Cel of C. biazotea. Our studies along this line may hopefully result in useful information on the complementary roles played by the two families of Cel enzymes in the hydrolysis of oligosaccharides. On the other hand, the E. coli and Bacillus subtilis systems engineered in our laboratory for the expression of extracellular heterologous proteins may offer a facile approach for the efficient production of the various

8

12

16

20

24

0

Post-induction time (hr)

Post-induction time (hr) Fig. 4. Ability of E. coli JM101 transformants harboring the pM and pMBcba3 constructs to utilize cellobiose as the sole carbon source. The transformants were grown in M9 minimal medium supplemented with cellobiose and IPTG (for induced expression of the cba3 gene). Data shown are mean± SEM values from different experiments (n= 3).

4

pNPGase (U/L)

60

B Anti-Cba3 0

1

4

8

12

16

20

24

CP

Supernatant samples from different post-induction time (hr)

C

SN

PP

CP Anti-Cba3

Anti-MazG

Fig. 5. Expression of Cba3 by E. coli JM101 transformant harboring the pMBcba3 construct. (A) Detection of extracellular Cba3 activities in a time-course study conducted under IPTG induction. Cba3 activities are presented as pNPGase activities, whereas viable cell counts are denoted as colony forming units (CFU). Data shown are mean ± SEM values from different experiments (n = 3). (B) Western blot analysis of culture media of the transformant. Supernatant samples (200 μl of each) from a time-course experiment described above were concentrated and resolved by SDS-PAGE, followed by incubation with antiCba3 antibodies. CP represents a cytoplasmic sample prepared from the transformant (see below). (C) Cellular fractionation. The samples of the following 3 fractions: supernatant (SN), periplasm (PP) and cytoplasm (CP) were prepared from a 24-hr culture of the transformant grown under IPTG induction. The samples were subjected to Western blot analysis using anti-Cba3 (upper panel) and anti-MazG (lower panel) antibodies. The MazG protein was employed as an intracellular marker. The same quantity of proteins was employed in the following 3 lanes: SN and PP of the upper panel, and CP of the lower panel. To obtain a band of comparable quantity, a reduced sample size (5% of the other three) was employed in the CP lane of the upper panel.

cellulase components. The application of these systems will hopefully facilitate studies of synergism involving the recombinant cellulases, and the subsequent development of a cost-effective process for the saccharification of cellulosic biomass. 5. Conclusions 1. A novel Cel gene, cba3, which is 1344 bp in length, was cloned and characterized from C. biazotea. 2. Comparison of the deduced Cba3 sequence with the information available for Bgl enzymes in the CAZy database unequivocally supports that Cba3 is a GH1 member, which represents the first known GH1 Bgl of C. biazotea. 3. Deletion of 10 codons from either the 5′- or the 3′- end of the cba3 gene totally abolished the biological function of Cba3, suggesting that these regions are important for enzymatic activity of Cba3. 4. Like other GH1 Bgl enzymes, Cba3 was revealed to possess the highly conserved NEP (residues 168–170) and ENG (residues 351–353) motifs, in which the two Glu (E) residues were shown to be essential for catalytic function. 5. Although Cba3 is a secretory protein, it appears that it does not possess a typical secretion signal for its delivery in C. biazotea.

A.K.N. Chan et al. / Gene 493 (2012) 52–61

Acknowledgments The authors wish to thank Dorothy Y.Y. Lam and Asma Ali for their technical assistance in the cloning and sequencing of the cba3 gene.

References Ausubel, F.M., et al., 2001. Curr. Protoc. Mol. Biol. John Wiley & Sons, Inc, Hoboken, NJ, USA. Barona-Gómez, F., Lautru, S., Francou, F.X., Leblond, P., Pernodet, J.L., Challis, G.L., 2006. Multiple biosynthetic and uptake systems mediate siderophore-dependent iron acquisition in Streptomyces coelicolor A3(2) and Streptomyces ambofaciens ATCC 23877. Microbiology. 152, 3355–3366. Bendtsen, J.D., Kiemer, L., Fausbøll, A., Brunak, S., 2005. Non-classical protein secretion in bacteria. BMC Microbiol. 5, 58. Bentley, S.D., et al., 2002. Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature. 417, 141–147. Bentley, S.D., et al., 2008. Genome of the actinomycete plant pathogen Clavibacter michiganensis subsp. sepedonicus suggests recent niche adaptation. J. Bacteriol. 190, 2150–2160. Bhat, M.K., Bhat, S., 1997. Cellulose degrading enzymes and their potential industrial applications. Biotechnol. Adv. 15, 583–620. Blanchard, J., Petit, E., Fabel, J., Leschine, S., 2010. Methods and compositions for improving the production of products in microorganisms. US20100028966. Cantarel, B.L., Coutinho, P.M., Rancurel, C., Bernard, T., Lombard, V., Henrissat, B., 2009. The Carbohydrate-Active EnZymes database (CAZy): an expert resource for glycogenomics. Nucleic Acids Res. 37, D233–D238. Dodd, D., Kiyonari, S., Mackie, R.I., Cann, I.K., 2010. Functional diversity of four glycoside hydrolase family 3 enzymes from the rumen bacterium Prevotella bryantii B14. J. Bacteriol. 192, 2335–2345. Fu, Z.B., Ng, K.L., Lam, T.L., Wong, W.K.R., 2005. Cell death caused by hyper-expression of a secretory exoglucanase in Escherichia coli. Protein Expr. Purif. 42, 67–77. Fu, Z.B., Ng, K.L., Lam, C.C., Leung, K.C., Yip, W.H., Wong, W.K.R., 2006. A two-stage refinement approach for the enhancement of excretory production of an exoglucanase from Escherichia coli. Protein Expr. Purif. 48, 205–214. Gartemann, K.H., et al., 2008. The genome sequence of the tomato-pathogenic actinomycete Clavibacter michiganensis subsp. michiganensis NCPPB382 reveals a large island involved in pathogenicity. J. Bacteriol. 190, 2138–2149. Gilkes, N.R., et al., 1984. Isolation and characterization of Escherichia coli clones expressing cellulase genes from Cellulomonas fimi. J. Gen. Microbiol. 130, 1377–1384. Hill, A.D., Reilly, P.J., 2008. Computational analysis of glycoside hydrolase family 1 specificities. Biopolymers. 89, 1021–1031. Hill, C., Lavigne, J., Whissel, M., Tomashek, J.J., 2010. Modified beta-glucosidases with improved stability. US20100093040 . Holtzapple, M., Cognata, M., Shu, Y., Hendrickson, C., 1990. Inhibition of Trichoderma reesei cellulase by sugars and solvents. Biotechnol. Bioeng. 36, 275–287. Ichikawa, N., et al., 2010. Genome sequence of Kitasatospora setae NBRC 14216T: an evolutionary snapshot of the family Streptomycetaceae. DNA Res. 17, 393–406. Ikeda, H., et al., 2003. Complete genome sequence and comparative analysis of the industrial microorganism Streptomyces avermitilis. Nat. Biotech. 21, 526–531. Lam, T.L., Wong, R.S.C., Wong, W.K.R., 1997. Enhancement of extracellular production of a Cellulomonas fimi exoglucanase in Escherichia coli by the reduction of promoter strength. Enzyme Microb. Technol. 20, 482–488. Lam, K.H., Chow, K.C., Wong, W.K.R., 1998. Construction of an efficient Bacillus subtilis system for extracellular production of heterologous proteins. J. Biotechnol. 63, 167–177. Lau, A.T.Y., Wong, W.K.R., 2001. Purification and characterization of a major secretory cellobiase, Cba2, from Cellulomonas biazotea. Protein Expr. Purif. 23, 159–166. Letek, M., et al., 2010. The genome of a pathogenic rhodococcus: cooptive virulence underpinned by key gene acquisitions. PLoS Genet. 6 (9). Merino, S.T., Cherry, J., 2007. Progress and challenges in enzyme development for biomass utilization. Adv. Biochem. Eng. Biotechnol. 108, 95–120. Mongodin, E.F., et al., 2006. Secrets of soil survival revealed by the genome sequence of Arthrobacter aurescens TC1. PLoS Genet. 2 (12), e214.

61

Moracci, M., Capalbo, L., Ciaramella, M., Rossi, M., 1996. Identification of two glutamic acid residues essential for catalysis in the beta-glycosidase from the thermoacidophilic archaeon Sulfolobus solfataricus. Protein Eng. 9, 1191–1195. Mosier, N.S., Hall, P., Ladisch, C.M., Ladisch, M.R., 1999. Reaction kinetics, molecular action, and mechanisms of cellulolytic proteins. Adv. Biochem. Eng. Biotechnol. 65, 23–40. Öhgren, K., Bura, R., Lesnicki, G., Saddler, J., Zacchi, G., 2007. A comparison between simultaneous saccharification and fermentation and separate hydrolysis and fermentation using steam-pretreated corn stover. Process Biochem. 42, 834–839. Ohnishi, Y., et al., 2008. Genome sequence of the streptomycin-producing microorganism Streptomyces griseus IFO 13350. J. Bacteriol. 190, 4050–4060. Okura, T., Nishimoto, T., Chaen, H., Fukuda, S., 2009. Beta-glucosidase, its preparation and use for producing beta-glucosidases. JP2009034028-A/1 . Oliynyk, M., et al., 2007. Complete genome sequence of the erythromycin-producing bacterium Saccharopolyspora erythraea NRRL23338. Nat. Biotechnol. 25, 447–453. Ragauskas, A.J., et al., 2006. The path forward for biofuels and biomaterials. Science 311, 484–489. Sambrook, J., Russel, D., 2001. Molecular Cloning: a Laboratory Manual, third ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Saratale, G.D., Saratale, R.G., Lo, Y.C., Chang, J.S., 2010. Multicomponent cellulase production by Cellulomonas biazotea NCIM-2550 and its applications for cellulosic biohydrogen production. Biotechnol. Prog. 26, 406–416. Schneiker, S., et al., 2007. Complete genome sequence of the myxobacterium Sorangium cellulosum. Nat. Biotechnol. 25, 1281–1289. Scott, B.R., Liu, C., Lavigne, J., Tomashek, J., 2010. Novel beta-glucosidase enzymes. US20100304438 . Sekine, M., et al., 2006. Sequence analysis of three plasmids harboured in Rhodococcus erythropolis strain PR4. Environ. Microbiol. 8, 334–346. Skipper, N., et al., 1985. Secretion of a bacterial cellulase by yeast. Science 230, 958–960. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positionspecific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. Wakarchuk, W.W., Kilburn, D.G., Miller Jr., R.C., Warren, R.A.J., 1984. The preliminary characterization of the β-glucosidases of Cellulomonas fimi. J. Gen. Microbiol. 130, 1385–1389. Wang, X.J., et al., 2010a. Genome sequence of the milbemycin-producing bacterium Streptomyces bingchenggensis. J. Bacteriol. 192, 4526–4527. Wang, Y.Y., et al., 2010b. Efficient Bacillus subtilis promoters for graded expression of heterologous genes in Escherichia coli. Res. J. Biotechnol. 5, 5–14. Wang, Y.Y., et al., 2011. Enhancement of excretory production of an exoglucanase from Escherichia coli with phage shock protein A (PspA) overexpression. J. Microbiol. Biotechnol. 21, 637–645. Warren, R.A., et al., 1986. Sequence conservation and region shuffling in an endoglucanase and an exoglucanase from Cellulomonas fimi. Proteins. 1, 335–341. Waterhouse, A.M., Procter, J.B., Martin, D.M.A., Clamp, M., Barton, G.J., 2009. Jalview Version 2 — a multiple sequence alignment editor and analysis workbench. Bioinformatics. 25, 1189–1191. Wong, W.K.R., Chan, W.K., 1998. Cellobiase obtained from Cellulomonas biazotea. UKGB2289050B . Wong, W.K.R., Lam, K.H., 2000. Bacterial expression system. US006146848 . Wong, W.K.R., Sutherland, M.L., 1997. Excretion of heterologous proteins from E. coli. US5646015 . Wong, W.K.R., Gerhard, B., Guo, Z.M., Kilburn, D.G., Warren, A.J., Miller Jr., R.C., 1986. Characterization and structure of an endoglucanase gene cenA of Cellulomonas fimi. Gene 44, 315–324. Wong, W.K.R., et al., 1988. Wood Hydrolysis by Cellulomonas fimi endoglucanase and exoglucanase coexpressed as secreted enzymes in Saccharomyces cerevisiae. Nat. Biotechnol. 6, 713–719. Wong, W.K.R., Ali, A., Chan, W.K., Ho, V., Lee, N.T.K., 1998. The cloning, expression and characterization of a cellobiase gene encoding a secretory enzyme from Cellulomonas biazotea. Gene 207, 79–86. Wood, T.M., McCrae Sheila, I., 1979. Synergism between enzymes involved in the solubilization of native cellulose. Hydrolysis of Cellulose: Mechanisms of Enzymatic and Acid Catalysis: Adv. Chem. ACS. , pp. 181–209. Zhao, W., et al., 2010. Complete genome sequence of the rifamycin SV-producing Amycolatopsis mediterranei U32 revealed its genetic characteristics in phylogeny and metabolism. Cell. Res. 20, 1096–1108.