Primary sequence and enzymic properties of two modular endoglucanases, Cel5A and Cel45A, from the anaerobic fungus Piromyces equi

Ruth Y. Eberhardt1, Harry J. Gilbert2 and Geoffrey P. Hazlewooda,1

Laboratory of Molecular Enzymology, The Babraham Institute, Babraham, Cambridge CB2 4AT, UK1
Department of Biological and Nutritional Sciences, The University of Newcastle upon Tyne, Newcastle upon Tyne NE1 7RU, UK2

Author for correspondence: Ruth Y. Eberhardt. Tel: +44 1223 496478. Fax: +44 1223 496023. e-mail: ruth.eberhardt{at}bbsrc.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Two endoglucanase cDNAs, designated cel5A and cel45A, were isolated from a cDNA library of the anaerobic fungus Piromyces equi. Sequence analysis revealed that cel5A has an open reading frame of 5142 bp and encodes a 1714 amino acid modular enzyme, Cel5A, with a molecular mass of 194847 Da. Cel5A consists of four catalytic domains homologous to family-5 glycosyl hydrolases, two C-terminal dockerins and one N-terminal dockerin. This is the first report of a complete gene containing tandem repeats of family-5 catalytic domains. The cDNA cel45A has an open reading frame of 1233 bp and encodes a 410 amino acid modular enzyme, Cel45A, with a molecular mass of 44380 Da. The catalytic domain, located at the C terminus, is homologous to the family-45 glycosyl hydrolases. Cel45A is the first family-45 enzyme to be described in an anaerobe. The presence of dockerins at the N and C termini of Cel5A and at the N terminus of Cel45A implies that both enzymes are part of the high-molecular-mass cellulose-degrading complex produced by Piromyces equi. The catalytic domain nearest the C terminus of Cel5A and the catalytic domain of Cel45A were hyperexpressed as thioredoxin fusion proteins, Trx-Cel5A' and Trx-Cel45A', and subjected to biochemical analysis. Trx-Cel5A' has a broad substrate range, showing activity against carboxymethylcellulose, acid-swollen cellulose, barley ß-glucan, lichenin, carob galactomannan,p-nitrophenyl ß-D-cellobiopyranoside and xylan. Trx-Cel45A' is active against carboxymethylcellulose, acid-swollen cellulose and the mixed linkage glucans, barley ß-glucan and lichenin.

Keywords: anaerobic fungi, cellulase, glycosyl hydrolases (families 5 and 45), endoglucanases, Piromyces equi

Abbreviations: CMC, carboxymethylcellulose; CMCase, carboxymethylcellulase; DNSA, dinitrosalicylic acid reagent; pNPC, p-nitrophenyl ß-D-cellobiopyranoside

The EMBL accession numbers for the sequences reported in this paper are AJ277482 and AJ277483.

a Present address: Finnfeeds International, PO Box 777, Marlborough SN8 1XN, UK.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Cellulose, the most abundant molecule in the biosphere, is a polymer of ß-1,4-linked glucose molecules. Chains of cellulose form insoluble fibrils, which are maintained by inter- and intramolecular hydrogen bonds and van der Waals interactions (Béguin & Lemaire, 1996 ). Three classes of enzymes are required for the complete hydrolysis of cellulose: endoglucanases (EC 3.2.1.4), which cleave ß-1,4-glycosidic bonds randomly within the cellulose chain; cellobiohydrolases (EC 3.2.1.91), which cleave cellobiose from the ends of the cellulose chain; and ß-glucosidases (EC 3.2.1.21), which convert cellobiose and other low molecular mass cellodextrins into glucose. Synergism between these three types of enzyme has been widely reported (Béguin, 1990 ; Leschine, 1995 ).

The sequences of over 1500 glycosyl hydrolases are known and have been assigned to at least 77 families on the basis of hydrophobic cluster analysis and amino acid sequence similarities. Twelve of these families contain endoglucanases (Henrissat, 1991 , 1998 ; Henrissat & Bairoch, 1993 , 1996 ). Many glycosyl hydrolases are modular enzymes consisting of one or more catalytic domains joined to non-catalytic domains via linker sequences rich in proline and hydroxy amino acids (Gilkes et al., 1991 ; Tomme et al., 1995 ; Warren, 1996 ).

Herbivores do not produce polysaccharide hydrolases, so rely on symbiotic relationships with micro-organisms (mostly bacteria and fungi), which inhabit their gastrointestinal tracts, to digest plant material. Anaerobic fungi were first isolated by Orpin (1975) , from the rumen of a sheep. Since then they have been recovered from the digestive tracts of many different species of herbivores, including both ruminants and non-ruminants, where they are believed to be responsible for the digestion of 40–70% of the ingested plant material (Li & Heath, 1993 ; Trinci et al., 1994 ). At least five genera of anaerobic fungi have been described to date including Neocallimastix, Orpinomyces, Piromyces, Caecomyces and Anaeromyces. The polysaccharide hydrolases produced by anaerobic fungi are amongst the most active that have been described to date; they are capable of degrading a wide range of polysaccharides and can completely solubilize both amorphous and highly crystalline cellulose (Li & Heath, 1993 ; Selinger et al., 1996 ; Wubah et al., 1993 ). Unlike aerobic fungi, anaerobic fungi produce large, multienzyme cellulase–hemicellulase complexes, similar to the cellulosome of Clostridium thermocellum (Ali et al., 1995 ; Dijkerman et al., 1997 ; Fanutti et al., 1995 ; Hazlewood & Gilbert, 1998 ; Teunissen et al., 1993 ; Wilson & Wood, 1992 ).

The anaerobic fungus Piromyces equi was isolated from the caecum of a pony (Munn, 1994 ; Orpin, 1981 ). The cellulose–hemicellulose degrading system of P. equi consists of a large multienzyme complex, which accounts for up to 90% of the cellulase, mannanase and xylanase activities produced by the fungus. This complex consists of at least ten polypeptides ranging from 50 to 190 kDa, including a 97 kDa putative scaffolding protein (Ali et al., 1995 ; Fanutti et al., 1995 ; Hazlewood & Gilbert, 1998 ). Catalytic components of the complex are all modular enzymes and contain one, two or three copies of a highly conserved 40-residue non-catalytic sequence that is believed to function as a dockerin domain (Fanutti et al., 1995 ). These putative dockerin domains are located at either the N terminus, the C terminus or between catalytic domains of P. equi enzymes (Hazlewood & Gilbert, 1998 ).

Although genes encoding xylanases and mannanases from P. equi have been reported (Fanutti et al., 1995 ; Millward-Sadler et al., 1996 ), no endoglucanase genes have been described to date. In this study we aim to characterize the endoglucanase components of the P. equi multienzyme complex. Here we describe the elucidation of the primary structures and enzymic properties of two endoglucanases from P. equi, Cel5A and Cel45A. Cel5A has four similar family-5 catalytic domains and shows cellulase, xylanase and mannanase activities. Cel5A is the first anaerobic fungal enzyme to be reported which has both N-terminal and the C-terminal dockerin domains within the same polypeptide. Cel45A is the first family-45 endoglucanase to be isolated from an anaerobic organism.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Microbial strains, vectors and culture conditions.
Escherichia coli strains SURE 2 (Stratagene), XL1-Blue and BL21(DE3) were cultured in Luria broth (LB) at 37 °C; the medium was supplemented with ampicillin (50 µg ml-1) to select against non-recombinants. To induce expression of genes under the control of the T7lac promoter, E. coli strains were grown to mid-exponential phase, 0·5 mM IPTG was added, and the bacteria cultured for a further 3 h. Vectors used in this study were {lambda}ZAPII (Stratagene), pBluescript SK- (Stratagene) and pET32a (Novagen). P. equi was cultured under anaerobic conditions at 37 °C in a medium containing rumen fluid (Kemp et al., 1984 ) and supplemented with Avicel (0·5%) and soluble xylan (0·1%).

General recombinant DNA methodology.
Agarose-gel electrophoresis, Southern hybridization, Northern hybridization and the transformation of E. coli were performed as described by Sambrook et al. (1989) . Plasmid DNA was prepared by an adaptation of the alkaline lysis method of Birnboim & Doly (1979) or by using Qiagen resin columns (Qiagen), following the protocol recommended by the manufacturer. Restriction endonucleases and other DNA-modifying enzymes were used in accordance with the manufacturer’s instructions. Total RNA was extracted from a 2 l culture of P. equi by grinding fungal mycelia to a powder in liquid nitrogen, resuspending in GTC [4 M guanidine thiocyanate, 25 mM sodium citrate, 0·5 mM EDTA pH 7·5, 0·5% (w/v) sodium lauryl sarcosine] and homogenizing. The resulting solution was centrifuged at 2000 g for 10 min. CsCl was added to the supernatant to a final concentration of 0·1 g ml-1, transferred to tubes containing 5·7 M CsCl, 0·1 M EDTA and centrifuged at 80000 g for 24 h. The supernatant was discarded and the RNA pellet resuspended in TE (10 mM Tris, 1 mM EDTA) pH 7·6, 0·1% SDS. The RNA was further purified by chloroform/n-butanol extraction and ethanol precipitation, before resuspension in 1 mM EDTA pH 7·6. A cDNA library was constructed in {lambda}ZAPII from P. equi mRNA as described previously (Ali et al., 1995 ) and screened for carboxymethylcellulase (CMCase) activity by plating onto NZY medium in a soft agar overlay containing 0·5 mM IPTG. After 16 h incubation at 37 °C, the plaques which had formed were overlaid with molten agar containing 0·5% carboxymethylcellulose (CMC), 50 mM K2HPO4 and 12 mM citric acid pH 6·5. After a further 4 h incubation at 37 °C, plates were stained with Congo red and destained with 1 M NaCl (Teather & Wood, 1982 ) to identify positive clones. Clones expressing CMCase activity were excised and recircularized to form pBluescript phagemids.

Nucleotide sequencing.
Nucleotide sequencing was performed on plasmid DNA using a PRISM ready reaction dye-deoxy terminator cycle sequencing kit (Applied Biosystems). To sequence cel5A, nested deletions were generated with exonuclease III using the double-stranded nested deletion kit (Amersham-Pharmacia). cel45A was sequenced by chromosome walking: synthetic oligonucleotide primers were designed according to known sequence. Sequences were compiled and ordered using software written by the Genetics Computer Group at the University of Wisconsin.

Construction of pRE1 and pRE2.
The primers 5'-CGGGATCCGAACCAACTGGTAATATTCGTG-3' (primer A1) and 5 ' - A C T A C C G C T C G A G T T A T G G T T C TTCTTCAATAG-3' (primer A2) were used to amplify the region of cel5A encoding the catalytic domain nearest the C terminus (between nucleotides 3790 and 4884) to construct pRE1. To construct pRE2, the region of cel45A encoding the catalytic domain (between nucleotides 535 and 1260) was amplified by PCR using the primers 5'-CGGGATCCACTGGTTCGGGTGGTAACAGC-3' (primer B1) and 5'-GCCGCCGCTCGAGAGATAGGGATATACTGG-3' (primer B2). The reaction conditions for PCR were: denaturation 30 s at 94 °C, annealing 45 s at 50 °C, elongation 75 s at 72 °C, for 25 cycles. The amplified DNA sequences were digested with BamHI and XhoI and ligated into the thioredoxin fusion expression vector, pET32a, digested with the same restriction enzymes to form pRE1 and pRE2, encoding Trx-Cel5A' and Trx-Cel45A', respectively.

Purification of fusion proteins.
E. coli strains harbouring the plasmids pRE1 and pRE2 were grown to mid-exponential phase in 3 l (pRE1) or 6·7 l (pRE2) LB medium containing 50 µg ampicillin ml-1 at 37 °C. IPTG was added to each culture to a concentration of 0·5 mM and cultures were grown for a further 3 h at 37 °C (pRE1) or 30 °C (pRE2). Cells were pelleted by centrifugation at 10000 g for 10 min and then resuspended in 75 ml 100 mM NaCl, 20 mM Tris pH 8·0. The cells were then lysed by sonication. Triton X-100 was added to the pRE2 extract to a final concentration of 0·1% (v/v), followed by an incubation at 4 °C for 20 min. Cell debris was removed by centrifugation at 10000 g for 20 min and the resulting cell-free extract was incubated with 5 ml TALON metal affinity resin (Clontech) for 2 h at 4 °C. Unbound proteins were removed by 5 min centrifugation at 150 g, four washes with 100 mM NaCl, 20 mM Tris/HCl pH 8·0 and two washes with 10 mM imidazole, 100 mM NaCl, 20 mM Tris/HCl pH 8·0. To elute bound proteins, the TALON resin was applied to a column and 5 ml 50 mM imidazole, 100 mM NaCl, 20 mM Tris/HCl pH 8·0 was added. Five 1 ml eluates were collected; elution was repeated with 5 ml 100 mM imidazole, 100 mM NaCl, 20 mM Tris/HCl pH 8·0 to obtain a further five eluates. Eluates of Trx-Cel5A' and Trx-Cel45A' were checked by SDS-PAGE as described by Laemmli (1970) . The eluates containing fusion protein were pooled and dialysed twice against 10 mM Tris pH 8·0 to remove imidazole, Trx-Cel45A' was further purified by anion-exchange chromatography using a 2 ml BioQ column (Bio-Rad) on an automated Biologic workstation (Bio-Rad) and eluted in 10 mM Tris/HCl pH 8·0 with a gradient of 0–500 mM NaCl.

Assays.
Protein concentrations of Trx-Cel5A' and Trx-Cel45A' were determined by dye binding (Bradford, 1976 ). Enzyme activity was assayed in 50 mM K2HPO4, 12 mM citric acid (PC buffer) pH 6·5 in the presence of 1 mg BSA at 37 °C (Trx-Cel5A') or 60 °C (Trx-Cel45A'), unless otherwise stated. Reducing sugar release was measured with the dinitrosalicylic acid reagent (DNSA) (Miller, 1959 ). All assays were performed for 5, 10, 15, 30 and 60 min to determine the initial rate of activity. Substrates tested were CMC (BDH), acid-swollen cellulose (Sigmacell, treated according to Wood, 1971 ), bacterial microcrystalline cellulose [harvested from Acetobacter hansenii, washed with 4% (v/v) NaOH, neutralized with HCl and washed in distilled H2O], Avicel (Merck), soluble birchwood xylan, barley ß-glucan (Megazyme), lichenin (Cetraria islandica; Sigma), laminarin (Laminaria digitata; Sigma), carob galactomannan (Megazyme), debranched arabinan (Megazyme) and potato galactan (Megazyme). Concentrations of substrate used ranged from 0·1% to 1·125% w/v. Cellobiohydrolase activity was measured using 100 µl 1 mM p-nitrophenyl ß-D-cellobiopyranoside (pNPC) (Sigma) in a 1 ml reaction volume, stopping the reaction with 500 µl 1 M Na2CO3 and measuring p-nitrophenyl release spectrophotometrically at 400 nm. Assays to determine pH optima were performed in citric acid, Na2HPO4 buffer (pH 4·0–7·6) or sodium barbitone, HCl buffer (pH 7·9–8·1). Temperature inactivation studies were carried out by preincubating enzyme and buffer at a given temperature for 0, 2, 5, 10, 15, 20, 30, 45 and 60 min, before assaying for activity against CMC as described above. One unit of enzyme activity was defined as the quantity of enzyme required to liberate 1 µmol product in 1 min.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Isolation of cel5A and cel45A
A cDNA library in {lambda}ZAPII, consisting of 1·8x107 clones, was constructed as described previously using mRNA isolated from Pir. equi mycelia cultured in medium containing soluble xylan and Avicel (Ali et al., 1995 ). Twenty-nine clones producing CMCase activity were isolated from this library. Restriction mapping and Southern hybridization indicated that the fungal cDNAs contained in these clones were derived from three different genes. Nucleic acid sequencing of the 5' and 3' ends of representatives of each of the three groups revealed sequences characteristic of enzymes in glycosyl hydrolase families 5, 6 and 45. Initial biochemical studies revealed that the family-5 enzyme had a broad substrate range (data not shown), so this enzyme was chosen for further study. The only family-45 enzymes described to date are from aerobic fungi and the aerobic bacterium Pseudomonas fluorescens (Davies et al., 1993 ; Gilbert et al., 1990 ; Saloheimo et al., 1994 ; Schauwecker et al., 1995 ; Schülein, 1997 ; Sheppard et al., 1994 ), which makes the Pir. equi family-45 enzyme particularly interesting for further study. The longest cDNAs encoding enzymes from families 5 and 45 (pKPC1 and pKPC28, respectively) were estimated by restriction mapping to be 5·7 kb and 1·5 kb, respectively. Northern blot analysis was performed subsequently using nucleic acid probes derived from the two cDNA sequences, in order to deduce the size of the full-length transcripts of these genes. Pir. equi produces RNA transcripts of 5·5 kb and 1·7 kb which hybridize to these regions, confirming that the cDNAs contained in pKPC1 and pKPC28 are complete (Fig. 1).



View larger version (52K):
[in this window]
[in a new window]
 
Fig. 1. Northern blot analysis of cel5A and cel45A expression in P. equi. (a) Ethidium bromide stained 1% agarose/2·2 M formaldehyde gel. Lanes: 1, RNA molecular mass markers (Sigma); 2, P. equi RNA. (b) Northern blot probed with the amplified region of cel5A cDNA encoding a catalytic domain, labelled with [{alpha}-32P]dCTP. (c) Northern blot probed with the amplified region of cel45A cDNA encoding the catalytic domain, labelled with [{alpha}-32P]dCTP.

 
Nucleotide sequence
A set of nested deletions was created with exonuclease III to facilitate the sequencing of the cDNA derived from pKPC1, designated cel5A. The complete nucleotide sequence of the cDNA derived from pKPC28, designated cel45A, was determined in both strands by chromosome walking. Translation of the nucleotide sequences of cel5A and cel45A revealed ORFs of 5142 bp and 1230 bp, respectively, encoding polypeptides of Mr 194848 and Mr 44380 (EMBL accession numbers AJ277482 and AJ277483). The assignment of the proposed translation-initiation codons was based on the following observations: (i) the presence of translational stop codons in all three ORFs upstream of the proposed translation-initiation codon; (ii) the absence of ATG sequences upstream of the ORF; (iii) a very high A+T content preceding the proposed translation-initiation codon, as found previously in cDNAs encoding cellulases, xylanases and mannanases from anaerobic fungi (Fanutti et al., 1995 ; Li et al., 1997a ). The non-coding regions of cel5A and cel45A had A+T contents of 90% and 88% respectively, compared with 62% and 57% for the ORFs. The codon utilization of cel5A and cel45A is very similar to that of other polysaccharide hydrolase genes in Piromyces, Neocallimastix and Orpinomyces (Black et al., 1994 ; Chen et al., 1997 ; Denman et al., 1996 ; Gilbert et al., 1992 ; Li et al., 1997a ; Millward-Sadler et al., 1996 ; Zhou et al., 1994 ); 16 of the 61 amino acid codons are absent from cel5A and 20 are absent from cel45A. There is a marked preference for T in the wobble position; 47% of cel5A and 51% of cel45A codons end in T, whereas only 12% (cel5A) and 9% (cel45A) end in G.

Molecular architecture
The deduced primary structures of Cel5A and Cel45A, encoded by cel5A and cel45A, respectively, revealed that the N-terminal regions conformed to typical signal peptides, which is consistent with the observation that endoglucanases are components of the extracellular cellulase system of P. equi. Cel5A has three copies of a highly conserved approximately 40-residue sequence, one between residues 23 and 57, and a further two between residues 1631 and 1708. Cel45A has three repeats of this 40-residue sequence between residues 22 and 152. This highly conserved 40-residue sequence is present in many of the polysaccharide hydrolases isolated from anaerobic fungi and is believed to function as a dockerin domain, which mediates binding of the catalytic subunits of the fungal multienzyme cellulase complex to a scaffolding protein (Dalrymple et al., 1997 ; Fanutti et al., 1995 ; Hazlewood & Gilbert, 1998 ; Li et al., 1997b ). Fungal dockerins have been previously found in one, two or three copies at either the N terminus, the C terminus or between two catalytic domains (Fig. 2). Cel5A is the first reported example of an enzyme with dockerins at two separate locations within the molecule. This may have arisen via the fusion of two endoglucanase genes encoding proteins with dockerin domains. Alternatively, the presence of dockerins at both ends of the enzyme could provide a means for linking complexes together to form clusters analogous to the polycellulosome of C. thermocellum (Mayer et al., 1987 ; Shoham et al., 1999 ).



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 2. Schematic representation of the molecular architecture of Cel5A and Cel45A. The locations of the signal peptides (horizontal lines), dockerin domains (diagonal lines), linker sequences (white), family-5 catalytic domains (vertical lines) and family-45 catalytic domains (wavy lines) are shown.

 
Based on sequence homology with known glycosyl hydrolases, Cel5A has four putative catalytic domains, spanning residues 106–462, 495–851, 880–1236 and 1269–1625, which are more than 99% identical to each other on the nucleic acid level. Comparison of the catalytic domains of Cel5A with sequences in the SWISS-PROT database revealed homology with the family-5 endoglucanases (Fig. 3). Aylward et al. (1999) recently reported the incomplete sequence of celD from Neocallimastix patriciarum. The protein encoded by this sequence consists of three family-5 catalytic domains followed by two dockerin domains and shows a high degree of similarity with the C-terminal part of Cel5A. In an earlier study, Xue et al. (1992) demonstrated that the full-length N. patriciarum celD transcript was 5·5 kb, which is the same length as the cel5A transcript produced by P. equi. This suggests that the full-length cel5A from P. equi is of a similar size to the full-length N. patriciarum celD. In addition to N. patriciarum celD, there are a number of other genes encoding family-5 endoglucanases from anaerobic fungi in the EMBL database including celB from Orpinomyces sp. PC-2 (accession no. U57818), Orpinomyces joyonii celA (accession no. U59432), celB2 (accession no. AF015249) and celB29 (accession no. AF015248), celB from N. patriciarum (accession no. Z31364) and celA from Neocallimastix frontalis (accession no. U38843). Orpinomyces sp. PC-2 celB, O. joyonii celB29 and N. patriciarum celB are all complete; their proposed start codons are preceded by A+T-rich DNA typical of the non-coding regions of anaerobic fungi (Li et al., 1997b ; Zhou et al., 1994 ). Furthermore, all three genes encode modular enzymes consisting of a single family-5 catalytic domain at the C terminus followed by two dockerin domains at the N terminus. The gene encoded by O. joyonii celA consists of a C-terminal catalytic domain followed by a repeated domain of unknown function (Liu et al., 1997 ). N. frontalis celA and O. joyonii celB2 are both incomplete. N. frontalis celA encodes a protein which consists of a 37-residue linker, followed by a family-5 catalytic domain and two dockerins (Fujino et al., 1998 ). O. joyonii celB2 encodes a protein which consists of a family-5 catalytic domain followed by a single C-terminal dockerin. Based on comparison with the sequence of P. equi cel5A, we believe that the proposed start codon of celB2 actually encodes residue 13 of the catalytic domain. Further support for this view comes from the fact that there is no stop codon upstream of the proposed start codon, and the DNA upstream of this codon is not A+T rich, suggesting that it may be part of a longer ORF. It is possible that both N. frontalis celA and O. joyonii celB2 are part of much larger genes. The presence of a single band at 5·5 kb on a Northern blot of P. equi RNA (Fig. 1), when probed with a truncated variant of cel5A encoding only a single catalytic domain (nucleotides 3790–4884), implies that P. equi does not produce any other enzymes homologous to the family-5 glycosyl hydrolases. Genes encoding xylanases with duplicated catalytic domains have been isolated from P. equi and N. patriciarum (Fanutti et al., 1995 ; Gilbert et al., 1992 ), but cel5A is the first complete cDNA to be cloned from an anaerobic fungus encoding an endoglucanase with multiple catalytic domains. The presence of reiterated catalytic domains within an enzyme could enable the domains to act synergistically; alternatively, the repetition could provide a means for increasing expression of the particular catalytic domain. The high degree of similarity between Cel5A and the other family-5 cellulases from anaerobic fungi reinforces the view that the anaerobic fungi are extremely closely related to each other and have evolved from a common ancestral plant-cell-wall-degrading fungus. An alternative hypothesis is that lateral transfer of genes has occurred between anaerobic fungi; the evidence for horizontal transfer of genes from rumen-inhabiting bacteria to fungi suggests that anaerobic fungi are capable of the uptake of genetic material (Chen et al., 1997 ; Gilbert et al., 1992 ; Hazlewood & Gilbert, 1998 ; Li et al., 1997b ; Zhou et al., 1994 ).



View larger version (80K):
[in this window]
[in a new window]
 
Fig. 3. Comparison of the amino acid sequences of the putative catalytic domains of P. equi Cel5A (between residues 106 and 462, 495 and 851, 880 and 1236, and 1269 and 1625) with the sequences of the catalytic domains of Clostridium longisporum CelA (SWISS-PROT P54937) (Mittendorf & Thomson, 1993 ), Ruminococcus albus EGA (SWISS-PROT P23660) (Poole et al., 1990 ), N. patriciarum CelB (SWISS-PROT Q12647) (Zhou et al., 1994 ) and C. thermocellum EGE (SWISS-PROT P10477) (Hall et al., 1988 ); conserved residues are shown in bold.

 
Cel45A has a putative C-terminal catalytic domain located between residues 195 and 410. Comparison of the sequence of this catalytic domain with sequences in the SWISS-PROT database revealed homology with family-45 endoglucanases (Fig. 4.). Cel45A is the first family-45 endoglucanase to be reported from an anaerobic organism. This suggests that Cel45A has not evolved by horizontal transfer of genes from rumen bacteria but may have originated from an aerobic fungal gene.



View larger version (64K):
[in this window]
[in a new window]
 
Fig. 4. Comparison of the amino acid sequence of the putative catalytic domain of P. equi Cel45A (between residues 197 and 405) with the sequences of the catalytic domains of Ustilago maydis EG1 (SWISS-PROT P54424) (Schauwecker et al., 1995 ), Humicola insolens EGV (SWISS-PROT P43316) (Davies et al., 1993 , 1995 ; Schülein, 1997 ), Pseudomonas fluorescens EGB (SWISS-PROT P18126) (Gilbert et al., 1990 ) and Fusarium oxysporum Kfam1 (SWISS-PROT P45699) (Sheppard et al., 1994 ); conserved residues are shown in bold.

 
Characterization of Cel5A' and Cel45A'
To determine the biochemical properties of a single putative catalytic domain of Cel5A (located between residues 1269 and 1625) and the putative catalytic domain of Cel45A (located between residues 195 and 410), fusion genes consisting of the DNA sequences encoding these domains, fused to the thioredoxin gene, were constructed using the pET32a vector. Purified Trx-Cel5A' and Trx-Cel45A' appeared as single bands on SDS-PAGE; their apparent molecular masses of 56 kDa (Trx-Cel5A') and 45 kDa (Trx-Cel45A') were consistent with their predicted molecular masses. The initial rates of activity of Trx-Cel5A' and Trx-Cel45A' on various substrates were determined (Table 1). Trx-Cel5A' showed cellulase, xylanase and mannanase activities; CMC, acid-swollen cellulose, barley ß-glucan, lichenin, xylan, galactomannan and pNPC were all hydrolysed, with specific activities of between 0·1 and 59·4 U (mg protein)-1. No hydrolysis was observed with laminarin, arabinan, galactan, bacterial microcrystalline cellulose or Avicel as substrates. Most of the family-5 enzymes reported to date are endoglucanases, but the family also includes exoglucanases (Mackenzie et al., 1997 ; Vazquez de Aldana et al., 1991 ), mannanases (Hilge et al., 1998 ) and some enzymes which are primarily endoglucanases but with a broad substrate specificity. P. equi Cel5A shows more sequence similarity to the family-5 endoglucanases than to the exoglucanases or the mannanases. Trx-Cel45A' hydrolysed CMC, acid-swollen cellulose and the mixed-linkage glucans, lichenin and barley ß-glucan efficiently, with specific activities of between 30·0 and 97·4 U (mg protein)-1. No hydrolysis was observed with laminarin, xylan, galactomannan, galactan, arabinan, bacterial microcrystalline cellulose, Avicel, or pNPC as substrates. The family-45 enzyme, endoglucanase B, from Pseudomonas fluorescens subspecies cellulosa, has activity against CMC, acid-swollen cellulose, barley ß-glucan, lichenin and Avicel (Gilbert et al., 1990 ). The lack of activity of Cel45A' against Avicel may be explained by the fact that, unlike EGB, Cel45A' has no cellulose-binding domain. Trx-Cel5A' and Trx-Cel45A' are both more active against barley ß-glucan than CMC; a likely reason for this is that CMC is highly substituted with methoxy side chains which may interfere with enzyme activity. The specific activities of these enzymes against acid-swollen cellulose are less than the specific activities against CMC; acid-swollen cellulose has a more highly ordered structure than CMC and is therefore less accessible to these enzymes.


View this table:
[in this window]
[in a new window]
 
Table 1. Substrate specificity of Trx-Cel5A' and Trx-Cel45A'

 
The pH profiles of Trx-Cel5A' and Trx-Cel45A' were examined between pH 4·0 and 8·0 using CMC as a substrate. Trx-Cel5A' is active over a narrow pH range (Fig. 5), with an optimum of around pH 5·1; approximately 78% of maximum activity is seen at pH 6·4, but activity falls off sharply at lower pH values. Trx-Cel45A' is active over a broader pH range, with an optimum of around pH 6·5 and approximately 65% of maximum activity between pH 5·2 and 7·9. The effect of temperature on Trx-Cel5A' and Trx-Cel45A' activity on CMC was determined. The initial rate of hydrolysis of CMC by Trx-Cel5A' was highest at 45 °C. Preincubation at 50 °C and at 55 °C caused a decrease in Trx-Cel5A' activity; an almost complete loss of Trx-Cel5A' activity was seen after 20 min preincubation at 55 °C. All further assays with Trx-Cel5A' were performed at 37 °C since the enzyme was unstable at higher temperatures. The initial rate of Trx-Cel45A' activity increased with increasing temperature between 40 °C and 70 °C. Trx-Cel45A' was more thermostable than Trx-Cel5A'; the enzyme was stable for at least an hour at 65 °C, but after 20 min incubation at 70 °C only 10% of the original enzyme activity remained. For this reason all assays with Trx-Cel45A' were performed at 60 °C.



View larger version (10K):
[in this window]
[in a new window]
 
Fig. 5. pH profiles of Trx-Cel5A' and Trx-Cel45A'. Trx-Cel45A' ({blacksquare}) and Trx-Cel5A' ({bullet}) activities against CMC were determined at 37 °C and 60 °C, respectively, in citric acid/Na2HPO4 buffer (pH 4·0–7·6) and sodium barbitone/HCl buffer (pH 7·9–8·1). Assays were performed for 5, 10, 15, 30 and 60 min to determine the initial rate of activity at each pH.

 
The data presented in this report provide further insights into the complex nature of the cellulase systems expressed by anaerobic fungi. It is apparent that the cellulase–hemicellulase complex produced by these microbial eukaryotes includes family-45 enzymes, dockerin sequences can be present at both the N and C termini of the catalytic components of the complex, and evidence is presented which indicates that certain cellulases produced by these fungi have evolved by extensive duplication of an ancestral core sequence.


   ACKNOWLEDGEMENTS
 
The authors gratefully acknowledge the financial support of BBSRC and Finnfeeds International. We thank Dr Peter White, Jo Durgan, Keith Davidson, Judith Laurie and Babraham Technix for their assistance.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Ali, B. R. S., Zhou, L., Graves, F. M., Freedman, R. B., Black, G. W., Gilbert, H. J. & Hazlewood, G. P. (1995). Cellulases and hemicellulases of the anaerobic fungus Piromyces constitute a multiprotein cellulose-binding complex and are encoded by multigene families. FEMS Microbiol Lett 125, 15-21.[Medline]

Aylward, J. H., Gobius, K. S., Xue, G.-P., Simpson, G. D. & Dalrymple, B. P. (1999). The Neocallimastix patriciarum cellulase, CelD, contains three almost identical catalytic domains with high specific activities on avicel. Enzyme Microb Technol 24, 609-614.

Béguin, P. (1990). Molecular biology of cellulose degradation. Annu Rev Microbiol 44, 219-248.[Medline]

Béguin, P. & Lemaire, M. (1996). The cellulosome: an exocellular, multiprotein complex specialized in cellulose degradation. Crit Rev Biochem Mol Biol 31, 201-236.[Abstract]

Birnboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7, 1513-1522.[Abstract]

Black, G. W., Hazlewood, G. P., Xue, G.-P., Orpin, C. G. & Gilbert, H. J. (1994). Xylanase B from Neocallimastix patriciarum contains a non-catalytic 455-residue linker sequence comprised of 57 repeats of an octapeptide. Biochem J 299, 381-387.[Medline]

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 72, 248-254.[Medline]

Chen, H., Li, X.-L. & Ljungdahl, L. G. (1997). Sequencing of a 1,3-1,4-ß-D-glucanase (lichenase) from the anaerobic fungus Orpinomyces strain PC-2: properties of the enzyme expressed in Escherichia coli and evidence that the gene has a bacterial origin. J Bacteriol 179, 6028-6034.[Abstract]

Dalrymple, B. P., Cybinski, D. H., Layton, I., McSweeney, C. S., Xue, G.-P., Swadling, Y. J. & Lowry, J. B. (1997). Three Neocallimastix patriciarum esterases associated with the degradation of complex polysaccharides are members of a new family of hydrolases. Microbiology 143, 2605-2614.[Abstract]

Davies, G. J., Dodson, G. G., Hubbard, R. E. & 7 other authors (1993). Structure and function of endoglucanase V. Nature 365, 362–364.[Medline]

Davies, G. J., Tolley, S. P., Henrissat, B., Hjort, C. & Schülein, M. (1995). Structures of oligosaccharide-bound forms of the endoglucanase V from Humicola insolens at 1·9 resolution. Biochemistry 34, 16210-16220.[Medline]

Denman, S., Xue, G.-P. & Patel, B. (1996). Characterization of a Neocallimastix patriciarum cellulase cDNA (CelA) homologous to Trichoderma reesei cellobiohydrolase II. Appl Environ Microbiol 62, 1889-1896.[Abstract]

Dijkerman, R., Op den Camp, H. J. M., Van der Drift, C. & Vogels, G. D. (1997). The role of the cellulolytic high molecular mass (HMM) complex of the anaerobic fungus Piromyces sp. strain E2 in the hydrolysis of microcrystalline cellulose. Arch Microbiol 167, 137-142.

Fanutti, C., Ponyi, T., Black, G. W., Hazlewood, G. P. & Gilbert, H. J. (1995). The conserved noncatalytic 40-residue sequence in cellulases and hemicellulases from anaerobic fungi functions as a protein docking domain. J Biol Chem 270, 29314-29322.[Abstract/Free Full Text]

Fujino, Y., Ogata, K., Nagamine, T. & Ushida, K. (1998). Cloning, sequencing, and expression of an endoglucananse gene from the rumen anaerobic fungus Neocallimastix frontalis MCH3. Biosci Biotechnol Biochem 62, 1795-1798.[Medline]

Gilbert, H. J., Hall, J., Hazlewood, G. P. & Ferreira, L. M. A. (1990). The N-terminal region of an endoglucanase from Pseudomonas fluorescens subspecies cellulosa constitutes a cellulose-binding domain that is distinct from the catalytic centre. Mol Microbiol 4, 759-767.[Medline]

Gilbert, H. J., Hazlewood, G. P., Laurie, J. I., Orpin, C. G. & Xue, G.-P. (1992). Homologous catalytic domains in a rumen fungal xylanase: evidence for gene duplication and prokaryotic origin. Mol Microbiol 6, 2065-2072.[Medline]

Gilkes, N. R., Henrissat, B., Kilburn, D. G., Miller, R. C. & Warren, R. A. J. (1991). Domains in microbial ß-1,4-glycanases: sequence conservation, function, and enzyme families. Microbiol Rev 55, 303-315.

Hall, J., Hazlewood, G. P., Barker, P. J. & Gilbert, H. J. (1988). Conserved reiterated domains in Clostridium thermocellum endoglucanases are not essential for catalytic activity. Gene 69, 29-38.[Medline]

Hazlewood, G. P. & Gilbert, H. J. (1998). Structure–function relationships in the cellulase–hemicellulase system of anaerobic fungi. In Carbohydrases from Trichoderma reesei and Other Microorganisms, pp. 147-155. Edited by M. Claeyssens, W. Nerinckx & K. Piens. Cambridge: The Royal Society of Chemistry.

Henrissat, B. (1991). A classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J 280, 309-316.[Medline]

Henrissat, B. (1998). Glycosidase families. Biochem Soc Trans 26, 153-156.[Medline]

Henrissat, B. & Bairoch, A. (1993). New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J 293, 781-788.[Medline]

Henrissat, B. & Bairoch, A. (1996). Updating the sequence-based classification of glycosyl hydrolases. Biochem J 316, 695-696.[Medline]

Hilge, M., Gloor, S. M., Rypniewski, W., Sauer, O., Heightman, T. D., Zimmermann, W., Winterhalter, K. & Piontek, K. (1998). High-resolution native and complex structures of thermostable ß-mannanase from Thermomonospora fusca – substrate specificity in glycosyl hydrolase family 5. Structure 6, 1433-1444.[Medline]

Kemp, P., Lander, D. J. & Orpin, C. G. (1984). The lipids of the anaerobic rumen fungus Piromonas communis. J Gen Microbiol 130, 27-37.[Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.[Medline]

Leschine, S. B. (1995). Cellulose degradation in anaerobic environments. Annu Rev Microbiol 49, 399-426.[Medline]

Li, J. & Heath, I. B. (1993). Chytridiomycetous gut fungi, oft overlooked contributors to herbivore digestion. Can J Microbiol 39, 1003-1013.[Medline]

Li, X.-L., Chen, H. & Ljungdahl, L. G. (1997a). Monocentric and polycentric anaerobic fungi produce structurally related cellulases and xylanases. Appl Environ Microbiol 63, 628-635.[Abstract]

Li, X.-L., Chen, H. & Ljungdahl, L. G. (1997b). Two cellulases, CelA and CelC, from the polycentric anaerobic fungus Orpinomyces strain PC-2 contain N-terminal docking domains for a cellulase-hemicellulase complex. Appl Environ Microbiol 63, 4721-4728.[Abstract]

Liu, J.-H., Selinger, L. B., Hui, Y.-H., Molony, M. M., Cheng, K.-J. & Beauchemin, K. A. (1997). An endoglucanase from the anaerobic fungus Orpinomyces joyonii: characterization of the gene and its product. Can J Microbiol 43, 477-485.[Medline]

Mackenzie, L. F., Brooke, G. S., Cutfield, J. F., Sullivan, P. A. & Withers, S. G. (1997). Identification of Glu-330 as the catalytic nucleophile of Candida albicans exo-ß-(1,3)-glucanase. J Biol Chem 272, 3161-3167.[Abstract/Free Full Text]

Mayer, F., Coughlan, M. P., Mori, Y. & Ljungdahl, L. G. (1987). Macromolecular organization of the cellulolytic enzyme complex of Clostridium thermocellum as revealed by electron microscopy. Appl Environ Microbiol 53, 2785-2792.

Miller, G. L. (1959). Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal Chem 31, 426-428.

Millward-Sadler, S. J., Hall, J., Black, G. W., Hazlewood, G. P. & Gilbert, H. J. (1996). Evidence that the Piromyces gene family encoding endo-1,4-mannanases arose through gene duplication. FEMS Microbiol Lett 141, 183-188.[Medline]

Mittendorf, V. & Thomson, J. A. (1993). Cloning of an endo-(1->4)-ß-glucanase gene, celA, from the rumen bacterium Clostridium sp. (‘C. longisporum’) and characterization of its product, CelA, in Escherichia coli. J Gen Microbiol 139, 3233-3242.[Medline]

Munn, E. A. (1994). The ultrastructure of anaerobic fungi. In Anaerobic Fungi: Biology, Ecology, and Function, pp. 47-105. Edited by D. O. Mountfort & C. G. Orpin. New York: Marcel Dekker.

Orpin, C. G. (1975). Studies on the rumen flagellate Neocallimastix frontalis. J Gen Microbiol 91, 249-262.[Medline]

Orpin, C. G. (1981). Isolation of cellulolytic phycomycete fungi from the caecum of the horse. J Gen Microbiol 123, 287-296.[Medline]

Poole, D. M., Hazlewood, G. P., Laurie, J. I., Barker, P. J. & Gilbert, H. J. (1990). Nucleotide sequence of the Ruminococcus albus SY3 endoglucanase genes celA and celB. Mol Gen Genet 223, 217-223.[Medline]

Saloheimo, A., Henrissat, B., Hoffrén, A.-M., Teleman, O. & Penttilä, M. (1994). A novel, small endoglucanase gene, egl5, from Trichoderma reesei, isolated by expression in yeast. Mol Microbiol 13, 219-228.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schauwecker, F., Wannner, G. & Kahmann, R. (1995). Filament-specific expression of a cellulase gene in the dimorphic fungus Ustilago maydis. Biol Chem Hoppe-Seyler 376, 617-625.[Medline]

Schülein, M. S. (1997). Enzymatic properties of cellulases from Humicola insolens. J Biotechnol 57, 71-81.[Medline]

Selinger, L. B., Forsberg, C. W. & Cheng, K.-J. (1996). The rumen: a unique source of enzymes for enhancing livestock production. Anaerobe 2, 263-284.

Sheppard, P. O., Grant, F. J., Oort, P. J., Sprecher, C. A., Foster, D. C., Hagen, F. S., Upshall, A., McKnight, G. L. & O’Hara, P. J. (1994). The use of conserved cellulase family-specific sequences to clone cellulase homologue cDNAs from Fusarium oxysporum. Gene 150, 163-167.[Medline]

Shoham, Y., Lamed, R. & Bayer, E. A. (1999). The cellulosome concept as an efficient microbial strategy for the degradation of insoluble polysaccharides. Trends Microbiol 7, 275-281.[Medline]

Teather, R. M. & Wood, P. J. (1982). Use of Congo red–polysaccharide interactions in enumeration and characterization of cellulolytic bacteria from the bovine rumen. Appl Environ Microbiol 43, 777-780.[Medline]

Teunissen, M. J., Hermans, J. M. H., Huis in ’t Veld, J. H. J. & Vogels, G. D. (1993). Purification and characterization of a complex bound and a free ß-1,4-endoxylanase from the culture fluid of the anaerobic fungus Piromyces sp. strain E2. Arch Microbiol 159, 265-271.

Tomme, P., Warren, R. A. J. & Gilkes, N. R. (1995). Cellulose hydrolysis by bacteria and fungi. Adv Microb Physiol 37, 1-81.[Medline]

Trinci, A. P. J., Davies, D. R., Gull, K., Lawrence, M. I., Nielsen, B. B., Rickers, A. & Theodorou, M. K. (1994). Anaerobic fungi in herbivorous animals. Mycol Res 98, 129-152.

Vazquez de Aldana, C. R., Correa, J., San Segundo, P., Bueno, A., Nebreda, A. R., Mendez, E. & del Rey, F. (1991). Nucleotide sequence of the exo-1,3-ß-glucanase-encoding gene, EXG1, of the yeast Saccharomyces cerevisiae. Gene 97, 173-182.[Medline]

Warren, R. A. J. (1996). Microbial hydrolysis of polysaccharides. Annu Rev Microbiol 50, 183-212.[Medline]

Wilson, C. A. & Wood, T. M. (1992). The anaerobic fungus Neocallimastix frontalis: isolation and properties of a cellulosome-type enzyme fraction with the capacity to solubilize hydrogen-bond-ordered cellulose. Appl Microbiol Biotechnol 37, 125-129.

Wood, T. M. (1971). The cellulase of Fusarium solani. Purification and specificity of the ß-(1-4)-glucanase and the ß-D-glucosidase components. Biochem J 121, 353-362.[Medline]

Wubah, D. A., Akin, D. E. & Borneman, W. S. (1993). Biology, fiber-degradation, and enzymology of anaerobic zoosporic fungi. Crit Rev Microbiol 19, 99-115.[Medline]

Xue, G.-P., Gobius, K. S. & Orpin, C. G. (1992). A novel polysaccharide hydrolase cDNA (celD) from Neocallimastix patriciarum encoding three multi-functional catalytic domains with high endoglucanase, cellobiohydrolase and xylanase activities. J Gen Microbiol 138, 2397-2403.[Medline]

Zhou, L., Xue, G.-P., Orpin, C. G., Black, G. W., Gilbert, H. J. & Hazlewood, G. P. (1994). Intronless celB from the anaerobic fungus Neocallimastix patriciarum encodes a modular family A endoglucanase. Biochem J 297, 359-364.[Medline]

Received 17 February 2000; revised 11 May 2000; accepted 17 May 2000.