Molecular cloning, gene expression analysis and structural modelling of the cellobiohydrolase I from Penicillium occitanis

Enzyme and Microbial Technology 46 (2010) 74–81

Contents lists available at ScienceDirect

Enzyme and Microbial Technology journal homepage: www.elsevier.com/locate/emt

Molecular cloning, gene expression analysis and structural modelling of the cellobiohydrolase I from Penicillium occitanis Fatma Bhiri a , Ali Gargouri b , Mamdouh Ben Ali c , Hafedh Belghith b , Monia Blibech a , Semia Ellouz Chaabouni a,∗ a b c

Unité Enzymes et bioconversions, Ecole Nationale d’Ingénieurs de Sfax, route soukra km 3.5, BP «1173», 3038 Sfax, Tunisia Laboratoire de Génétique Moléculaire des Eucaryotes, Centre de Biotechnologie de Sfax, BP «K», 3038 Sfax, Tunisia Laboratoire d’ Enzymes et de Métabolites des Procaryotes, Centre de Biotechnologie de Sfax, route Sidi Mansour, BP «K», 3038 Sfax, Tunisia

a r t i c l e

i n f o

Article history: Received 16 March 2009 Received in revised form 3 October 2009 Accepted 6 October 2009 Keywords: Penicillium occitanis cDNA and genomic libraries Cellobiohydrolase Genomic organisation Cellobiose inhibition 3D structure

a b s t r a c t The filamentous fungus Penicillium occitanis produces a complete set of cellulolytic enzymes needed for efficient solubilization of native cellulose. Cellobiohydrolase I (CBHI), the most abundant cellulolytic enzyme produced by this micro-organism, has been purified and characterized. In this report, the cDNA encoding this enzyme, isolated from a cDNA bank of P. occitanis, and the equivalent genomic sequence have been cloned. DNA sequencing revealed that the cbh1 gene is intronless and has an open reading frame of 1587 bp encoding a putative polypeptide of 529 amino acids. This polypeptide has a predicted molecular mass of 52.5 kDa and consists of a fungal cellulose binding module (CBM) and a catalytic module, linked together by a serine–threonine-rich region. Northern blot analysis showed that cbh1 mRNA expression is partially constitutive since, besides being highly induced by cellulose, it is slightly repressed by glucose. Comparative investigation of different cellobiohydrolases I 3D structures by molecular modelling showed that poor hydrogen bonding, together with a more open configuration of the active site account for the weak binding and the relative insensitivity of P. occitanis CBHI to product inhibition. © 2009 Elsevier Inc. All rights reserved.

1. Introduction Cellulose is an insoluble polysaccharide composed of long linear chains of ␤-1,4-linked glucose units. It is the most abundant renewable biomass on earth since its microbial breakdown creates the potential for the production of energy [1]. Cellulolytic enzymes can be divided into three types: endo-␤-1,4-glucanase (EC 3.2.1.4), exo-␤-1,4-glucanase (cellobiohydrolase, EC 3.2.1.91), and ␤-glucosidase (EC 3.2.1.21). They are collectively known as cellulases and act in a synergistic manner to facilitate complete cleavage of ␤-1,4-glycosidic bonds of the cellulose to produce glucose [1]. Cellulases are used in waste recycling processes and in the processing of cellulose-rich raw materials for food, detergent, paper and textiles industries. Recently, cellulases gained significant commercial importance due to their potential applications in biofuel production [2]. Because of the low activity of endo-␤-1,4glucanases to hydrolyze crystalline cellulose, exo-type cellulases such as cellobiohydrolases (CBHs) are necessary in hydrolyzing

Abbreviations: cbh1, cellobiohydrolase I; CBM, cellulose binding module; GHF, glycoside hydrolase family. ∗ Corresponding author. Tel.: +216 74 274 418; fax: +216 74 275 595. E-mail address: [email protected] (S. Ellouz Chaabouni). 0141-0229/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.enzmictec.2009.10.002

crystalline cellulose. These exo-acting enzymes possess tunnel-like active sites, which can only accept a substrate chain via its terminal regions [3]. Thus, CBH enzymes act by threading the cellulose chain through the tunnel, where successive cellobiose units are removed in a sequential manner. Cellobiohydrolases I (CBHI) are modular enzymes consisting of a minimum of one catalytic module and one cellulose binding module (CBM) connected by a proline/serine/threonine-rich linker [3,4]. CBHs from different microbial sources belong to families 6 and 7 of glycoside hydrolases [5]. The most characterized members of the family 7 are cellobiohydrolase Cel7A from Trichoderma reesei and cellobiohydrolase Cel7D from Phanerochaete chrysosporium. The structure of both CBHs consists of two ␤-sheets that pack faceto-face to form a ␤-sandwich [6,7]. The cellobiohydrolase Cel7A from T. reesei is composed of long loops, on one face of the sandwich, that form a cellulose binding tunnel of 50 Å. The catalytic residues are glutamate 212 and 217, which are located on opposite sides of the active site, separated by an intervening distance consistent with a double-displacement retaining mechanism [3]. The mechanism of action, the kinetic parameters and the enzyme–ligand interactions of enzymes belonging to family 7 of glycoside hydrolases are well characterized [6]. The fungus Penicillium occitanis has been shown to possess a high capacity for the production of cellulases [8,9] having high cel-

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

lulose degradation efficiency [10]. Two cellobiohydrolases (CBHI and CBHII) and two ␤-glucosidases have been purified from this fungus and their properties were characterized [11]. Compared with other P. occitanis cellulases, the amount of CBHI secreted is much higher (50% of total proteins). CBHI of P. occitanis was described as an enzyme producing cellobiose from cellulose. This enzyme shares some biochemical properties with the cellobiohydrolases of glycoside hydrolases family 7 [11]. CBHI was also reported to act synergistically with cellobiohydrolases II and to be inhibited by cellobiose [11]. But, compared to P. chrysosporium, T. reesei and Talaromyces emersonii cellobiohydrolases I, P. occitanis CBHI exhibited less pronounced product inhibition [11–13]. Interestingly, this enzyme showed a mannanase activity using the Locust Bean Gum as substrate (unpublished data). Cellobiohydrolases genes have been cloned and characterized from a variety of fungal sources [14–17]. However, there are no reports of gene sequences coding for extracellular cellulolytic enzymes from P. occitanis. As an initial step toward elucidating the genetic basis for the production of cellulases from this organism, we report here the cloning and the characterization of the cbh1 cDNA and its corresponding gene from a genomic bank of P. occitanis. We also describe the similarity between the deduced CBHI protein and other fungal cellobiohydrolases. In order to explain the origin of the resistance to cellobiose inhibition, exhibited by this enzyme, structural modelling of the P. occitanis CBHI has been performed based on the X-ray crystallographic structure of T. emersonii cellobiohydrolase I. 2. Materials and methods

75

2.5. Construction and screening of the cDNA library The cDNA library was constructed using 4 ␮g of mRNA according to the manufacturer’s instructions of the cDNA Synthesis System kit (Amersham). Library screening was done with the RT-PCR amplified fragment described previously, as a probe. 2.6. Construction and screening of the genomic library Total genomic DNA of CL100 was prepared according to Aifa et al. [20], partially digested with Sau3AI and fractionated by sucrose gradient (10–40%) according to Hopwood et al. [21]. The fragments sizing from 6 to 9 kb length were isolated and cloned in pUC18 vector linearized at the BamH1 site. The recombinant clones were selected on Luria broth plates containing 100 ␮g/ml of ampicillin and 12.5 ␮g/ml of tetracycline. 2.7. Northern and Southern blots RNA, prepared according to Aifa et al. [20], was denatured with formamide and size fractionated by electrophoresis in 1.5% agarose formaldehyde gel. Restriction enzyme digestions, Northern and Southern hybridizations were performed as described by Sambrook et al. [22] using N+ -Hybond. 2.8. Primer extension Poly(A)+ RNA were isolated as described in the Quick Prep Micro mRNA Purification Kit (Amersham). Two picomoles of a 5 ␥32 P labelled primer (5 GGTTTCAGCAGTATAAGT-3 ) located at 87 nucleotides from the initiating ATG of the cbh1 gene, were mixed with 4.5 ␮g of polyA+ cellulose-induced RNA in the presence of 5× annealing buffer (25 mM Tris, pH 8.3; 375 mM KCl and 5 mM EDTA), denatured at 75 ◦ C for 2 min, incubated at 45 ◦ C for 30 min and then gradually cooled down to 37 ◦ C. After an ethanol precipitation step, the extension was performed in 20 ␮l final volume containing 15 units of AMV (Amersham) in 1× reverse transcriptase buffer (50 mM Tris–HCl, pH 8.3; 50 mM NaCl; 8 mM MgCl2 and 1 mM DTT), 5 mM DTT, 1.5 mM dNTP. After 1 h at 42 ◦ C, the reaction was stopped by addition of a solution containing 95% formamide, 20 mM EDTA, 0.05% bromophenol blue and 0.05% xylene cyanol FF.

2.1. Strains 2.9. Sequencing and sequence analysis P. occitanis CL100 and Pol6 were provided by Professor Tiraby, CAYLA Company, Toulouse—France. The Pol6 strain is a hypercellulolytic mutant selected by Jain et al. [8] after eight rounds of mutagenesis from the CL100 mother strain. The CT1 strain is a hyperpectinolytic and a fully constitutive mutant selected after a single round of mutagenesis from the same parental strain [18]. Escherichia coli strain: Top 10 F ((F lacIq Tn10 (TetR )) mcrA (mrr-hsdRMS-mcrBC) 80lacZM15 lacX74 recA1 araD139 (ara-leu)7697 galU galK rpsL (StrR )endA1 nupG; Invitrogen) was used as a host for the pUC18 and pMOSblue T cloning vectors.

The nucleotide sequence was carried out on both strands with both Thermosequenase Cycle Sequencing Kit (Amersham) and the BigDye Terminator version. 3.1 Cycle Sequencing Kit using an automated ABI Prism 3100-Avant Genetic Analyser (Applied Biosystems Inc.). The cbh1 gene sequence reported in this article has been submitted to the GenBank under accession number AY690482. 2.10. Amino acid sequence analysis and homology modelling

2.2. Vectors pMOSblue T-vector (Amersham) was used for the cloning of PCR fragments; ␭MOS Elox plasmid (Amersham) was used for the construction of the cDNA library; pMOS Elox derived from the excision of the ␭MOS Elox after infection of the BM 25.8 strain and pUC18 was used for the construction of the genomic library.

Multalin software was used to generate the alignment of CBHI sequences [23]. Rendering of the alignment figure including the prediction of the secondary structures was performed with the ESPript program [24]. Putative signal sequences were identified using the SignalP prediction software [25]. 2.11. Computer-aided model building of the tertiary structure of the CBHI

2.3. Media and growth conditions Potato dextrose agar (PDA, Merck Co.) was used for the propagation and the storage of the fungal strains. The liquid medium of Mandels and Weber [19] modified by Ellouz Chaabouni et al. [9] was also used. The carbon source is 2% cellulose (Avicel PH 101 Fluka, Switzerland) or 2% glucose. The cultures were grown in Erlenmeyer flasks (100/500 ml) at 30 ◦ C. Luria broth medium was used for the cultivation of bacterial strains.

The automated protein structure homology-modelling server, SWISS-MODEL [26] was used to generate the 3D model. The Deep View Swiss PDB Viewer software from EXPASY server (available at http://www.expasy.org/spdbv) was used to visualize and analyze the atomic structure of the model. Molecular modelling of P. occitanis CBHI was analyzed based on the X-ray crystallographic structure of the cellobiohydrolase of T. emersonii (pdb accession code 1Q9H). Finally, PyMOL [27], the Molecular Graphics System was used to render figures.

2.4. Amplification of the cbh1 cDNA

3. Results and discussion

Reverse transcription was performed for 60 min at 37 ◦ C on cellulose-induced polyA+ mRNA as a template with an oligo-dT primer [5 -GGGATCCGCGGCCGC(T15 )]. The mRNA extracted from glucose grown culture, was used as a control. Based on the sequences of fungal cellobiohydrolases I present in the database, primers were designed for the amplification of the cbh1 cDNA. The primer sequences are as follows: P1: 5 -TGCGGTCTCAACGGCGCCCTCTA-3 , P2: 5 -ATGGACGCCGACGGTGG-3 and P3: 5 -GAGATGGATATCTGGGAGGCCAA3 (sense primers corresponding to the peptides CGLNGALY, MDADGG and EMDIWEAN, respectively), P4: 5 -GGGATAGGTGCTGTCGAGCCACAACA-3 , P5: 5 -CCICCA/GCAT/CTGICC-3 and P6: 5 -AAGGCATTGCGAGTAGTAGTCGTT-3 (antisense primers corresponding to the peptides MLWLDSTYP, GQCGG and YYSQCL, respectively). The amplification protocol consisted of an initial denaturing cycle of 30 s at 94 ◦ C followed by a 90 s annealing step at 55 ◦ C and finally a 4 min polymerisation cycle at 72 ◦ C.

3.1. Isolation of the cbh1 cDNA and analysis of its sequence In order to clone the cbh1 cDNA of P. occitanis, we performed RT-PCR reactions on poly(A)+ RNA extracted from the hypercellulolytic Pol6 mutant grown on cellulose and glucose using different primer combinations: P1–P4, P2–P4, P3–oligo-dT, P2–P5 and P2–P6 (Fig. 1B). The amplified fragments were tested in Southern blot by a CBHI probe obtained by PCR from a Trichoderma species. Only the cellulose-induced RNA allowed the amplification of 1 kb fragment using the P3 and the oligo-dT primers, which strongly hybridized

76

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

Fig. 1. (A) Analysis of the RT-PCR products. Reactions were performed with induced (lines 1, 2, 3, 4 and 5) and uninduced (lines 1 , 2 , 3 , 4 and 5 ) poly(A)+ RNA; using the following primers: P1–P4 (lines 1 and 1 ); P2–P4 (lines 2 and 2 ); P3–oligo-dT (lines 3 and 3 ); P2–P5 (lines 4 and 4 ) and P2–P6 (lines 5 and 5 ). (a) Staining with ethidium bromide. (b) Hybridization with the Tricoderma reesei cbh1 probe. (B) Localization of the primers used for the RT-PCR. The primer sequences are: P1 (5 -TGCGGTCTCAACGGCGCCCTCTA3 ), P2 (5 -ATGGACGCCGACGGTGG-3 ), P3 (5 -GAGATGGATATCTGGGAGGCCAA-3 ), P4 (5 -GGGATAGGTGCTGTCGAGCCACAACA-3 ), P5 (5 -CCICCA/GCAT/CTGICC-3 ) and P6 (5 -AAGGCATTGCGAGTAGTAGTCGTT-3 ).

with the CBHI probe (Fig. 1A). This fragment was then cloned in pMOSblue T cloning vector and sequenced. The identity of the fragment as a CBHI sequence was then established by nucleotidic sequencing and Blast study. The cbh1 amplified fragment was used as a probe to screen a cDNA library (2 × 103 clones) constructed in the ␭ MOS Elox phagic vector. The complete nucleotide sequence of the cbh1 cDNA was determined by combining partial sequences obtained from various positive overlapping clones. The nucleotide sequence of cbh1 contains a single open reading frame (ORF) consisting of 1587 bp and coding for a putative polypeptide of 529 amino acids. By comparison with the N-terminal amino acid region of other known CBHI sequences, a putative leader peptide of 25 amino acids is predicted using the SignalP software. The secretory precursor is thought to be processed at a specific cleavage site between the A 25 and Q 26 residues, resulting in the formation of a mature enzyme composed of 504 amino acids with a molecular weight of 52.5 kDa. The difference between the calculated molecular weight (52.5 kDa) and that of the purified enzyme (60 kDa) [11] is probably due to posttranslational modifications of the protein. In fact, the CBHI was found to be a glycoprotein containing about 20% of carbohydrates, which is in agreement with the results obtained. The codon usage in the P. occitanis cbh1 showed that this gene contains 55.14% (G + C). Like in the pectin lyase gene (pnl1) of the same fungus [28], cytosine residues are preferred at the third position of codons and used in 49.15% of the cases (data not shown). As in other cellulase genes [14,29,30], there is a bias against NTA codons.

The localization of the 5 -ends of the transcripts by primer extension analysis revealed heterogeneity in the transcription start sites. Three major transcription start sites located at 29, 32 and 35 nucleotides upstream of the ATG were identified (Fig. 2). The corresponding 5 -untranslated mRNA sequences are thus 29, 32 and 35 nucleotides long. Multiple start sites seem to be common among filamentous fungal genes [35], whereas the start point of higher eukaryotic genes is usually 30 nucleotides from the TATA-box [36]. Cellulase genes of T. reesei are repressed in the presence of glucose by the Catabolite Repression Element (CREI) which together with CREA of the Aspergilli is the only known repressor of cellulase and hemicellulase genes [17]. Putative regulatory CRE binding sites in the cbh1 promoter have been identified and investigated in both Trichoderma and Penicillium species [16,17,37]. The P. occitanis cbh1 upstream sequence was found to contain up to seven CRE consensus binding sites, suggesting their participation in repression of this gene in response to glucose. Cellulase promoters are also positively regulated by Activators of Cellulase Expression (ACEI and ACEII) factors [38,39]. Putative sequences involved in the binding of the ACEI and ACEII were also identified in the P. occitanis cbh1 promoter [37].

3.2. Isolation of the cbh1 gene

3.4. Copy number of cbh1 gene

In order to isolate the whole cbh1 gene and to study its regulation, a genomic bank covering approximately the totality of the fungal genome was prepared in pUC18 plasmid using DNA extracted from the wild type strain (CL100) of P. occitanis. Approximately 13,000 clones of this library were replicated and hybridized with the CBHI probe obtained previously (see above): 80 clones were positives. The presence of the cbh1 insert in these clones was checked by Southern blot hybridization and by sequencing. The average size of these inserts is around 5 kb, covering the encoding sequence and its 3 and 5 flanking regions. When compared with the cDNA sequence, the P. occitanis cbh1 gene was found intronless. The same result was obtained for the Aspergillus aculeatus, Penicillium janthinellum and Penicillium funiculosum cellobiohydrolase

To estimate the copy number of the cbh1 gene in P. occitanis, the genomic DNA was digested with various restriction enzymes (AvaI, AccI, SacI, BglI, SpeI, HindIII and EcoRV). There are no internal sites for AvaI, SacI, BglI, SpeI and HindIII in the probe sequence but one site for EcoRV and AccI. The digested genomic DNA was subjected to gel electrophoresis, blotted to hybond membrane and then hybridized with the whole sequence of the cbh1 cDNA used as probe. A single intense band was seen in most lanes except in some instances where the digestion is partial or when the restriction enzyme cut once in the interior of the probe sequence (Fig. 3). This result suggests that only one gene encodes the CBHI protein. The second band detected in lines 3–6 is probably due to the presence of cbh1 homologous genes, like

genes [16,31,32] whereas all the other fungal cbh1 genes sequenced so far have their nucleotide sequences interrupted by introns at varying positions [33,34]. 3.3. Structure of the 5 -non-coding region of cbh1 gene

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

77

Fig. 2. Nucleotide and deduced amino acid sequence of the cbh1 gene. The TATA-box is underlined. The AATAAA polyadenylation signal and the CA-polyadenylation site are in bold and underlined. The peptides used to design the primer sequences tested in the RT-PCR are shown by arrows. The putative signal sequence is underlined. Presumed transcription initiation sites are indicated by stars. Motifs for the binding of Activators of Cellulase Expression: ACEI (lower-case letters) and ACEII (italicized and underlined letters) are shown. The Catabolite Repression Element (CREA) binding sites are also indicated (bold capitalized and underlined letters). The catalytic residues are circled. Potential N-glycosylation sites are boxed. This sequence has been submitted to the GenBank under accession number AY690482.

the endoglucanase I gene, which shares an identity of 45% with the cbh1 gene of T. reesei [40]. Previous works suggested the existence of a single-copy cbh1 gene in several Trichoderma strains [41] or multiple copies of the cbh1 gene for P. chrysosporium and P. janthinellum [16,33].

3.5. Analysis of the cbh1 gene expression Northern hybridization of RNA isolated from the hypercellulolytic mutant Pol6 was performed using the cbh1 cDNA as a probe. The RNA from the hyperpectinolytic mutant CT1 and the parental strain CL100 was also extracted and used as controls in the same experiment.

Fig. 4 shows that a transcript of 1.9 kb hybridizes with this probe. The estimated size of the cbh1 mRNA is in good agreement with that of the isolated cDNA. As shown in the same figure, the cbh1 RNA expression is induced by cellulose and repressed by glucose in CL100 and CT1 strains. However, the Pol6 mutant showed a very high induction of cbh1 transcripts on cellulose and a significant expression on glucose. The cbh1 expression on glucose is even stronger than that observed on cellulose grown CL100 or CT1 strains. Such a result has not been yet described in other filamentous fungi. The partial constitutivity of Pol6 strain could be due to the multiple rounds of mutagenesis which have probably affected the Pol6 expression regulation. These mutations could affect one or more trans-regulating factors since sequencing of cbh1 promoter from the wild and the mutant strains shows an identity of 100%.

78

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

Fig. 3. Southern blot hybridization of Penicillium occitanis DNA. Genomic DNA (20 ␮g) was cut with AvaI (1), AccI (2), SacI (3), BglI (4), SpeI (5), HindIII (6) and EcoRV (7) and then analysed on 1% agarose gel and probed with the whole CBHI sequence. There are no internal sites for AvaI, SacI, BglI, SpeI and HindIII in the probe sequence but one site for EcoRV and AccI.

3.6. Amino acid sequence data of P. occitanis CBHI Analysis of the deduced amino-peptide sequence using a Blast search revealed the presence of putative conserved domains on the CBHI protein suggesting its belonging to glycosyl hydrolase family 7 (GH7).

Fig. 4. Northern analysis of total RNA extracted from the CL100 wild type strain, the hypercellulolytic Pol6 and hyperpectinolytic CT1 mutants grown on cellulose (C) or glucose (G) for 3 days. The loaded quantity of RNA (20 ␮g/lane) was controlled by staining the gel with ethidium bromide prior to blotting.

The C-terminal domain of P. occitanis CBHI showed a high degree of similarity with the “Cellulose Binding Modules” (CBMs) of family 1. The three-dimensional structure of T. reesei CBHI–CBM belonging to family 1 and determined by NMR spectroscopy provides the key residues involved in the interaction between CBM and cellulose. The structure suggests that aromatic moieties of three conserved Tyr side chains of the CBM (Y524, Y525 and Y498) form hydrophobic interactions with the cellulose molecule, and the side chains of three amino acids (Q500, N522, and Q527) stabilize the

Fig. 5. Protein sequence analysis and secondary structure alignment with ESPript program of cellobiohydrolases from glycoside hydrolase family 7. Residues identical in all sequences are printed in white on a red background. Residues identical or with a conservative substitution in at least four of the seven sequences are printed in red on a yellow background. A low consensus value (50%) was used in the multiple sequence alignment. A residue that is highly conserved appears as an uppercase letter in the consensus line. A residue that is weakly conserved appears as a lower-case letter in the consensus line. A position with no conserved residue is represented by a dot in the consensus line. Clusters of homologous symbols are indicated in the consensus line: IV (!), BDENZQ (#), LM ($) and FY (%). The secondary structure element from the known Talaromyces emersonii cellobiohydrolase (1Q9H) 3D structure is indicated at the top of the alignment. ␣-Helices, ␩-helices, ␤-sheets and strict ␤-turns are denoted ␣, ␩, ␤ and TT, respectively. The catalytic residues (E234, D236 and E239) are shown by green arrows. The CBM and linker sequences are boxed. Penicillium occitanis CBHI exhibited an identity of 93%, 71%, 62%, 61%, 57% and 57% with Penicillium funiculosum (AJ312295), Talaromyces emersonii (pdb 1Q9H), Penicillium janthinellum (Q06886), Aspergillus aculeatus (O59843), Trichoderma reesei (E00389) and Phanerochaete chrysosporium (M22220), respectively. The sequence numbering refers to the Penicillium occitanis CBHI. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of the article.)

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

79

Fig. 6. (A) Structure model of the Penicillium occitanis CBHI catalytic module. (red) Catalytic residues; (yellow) disulphide bridges. (B) Surface representation of the structures of CBHI of Trichoderma reseei (a), Phanerochaete chrysosporium (b) and Penicillium occitanis (c). The active site and the catalytic residues are shown in red and yellow, respectively. The substrate analog (a nanomer of glucose residues) is drawn in yellow. The apparent (Ki ) values for cellobiose are 0.02, 0.18 and 2 mM for Trichoderma reseei, Phanerochaete chrysosporium and Penicillium occitanis, respectively. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of the article.)

cellulose–CBM interaction by forming hydrogen bond with the glucose residues in the cellulose chains [6,42]. In the cellobiohydrolases I CBM, all these aromatic residues are conserved (Y524, Y525 and Y498). The presence of tyrosyl or other aromatic residues in the binding face is typical for carbohydrate–protein interactions. However, the aromatic residue, at the position 498, is present as a tryptophan in the Penicillium species and P. chrysosporium (Fig. 5). This residue is demonstrated to have a high affinity towards cellulose in the endoglucanase I (EGI) compared to the CBHI of T. reesei. The mutation Y498W in the CBHI increased its CBM affinity but did not change its enzymatic activity [43]. The CBHI–CBM is preceded by a putative linker peptide of about 37 amino acids, rich in serine and threonine residues. The linker peptide probably ensures an optimal interdomain distance between the catalytic and the cellulose binding modules and/or protects the enzyme against proteolytic attack since it is a flexible and highly O-glycosylated region [44]. Koch et al. [16] reported that the CBHI-type enzymes are characterized by 22 conserved cysteine residues and a catalytic site involving two glutamic acid residues. The fact that these residues are conserved in P. occitanis CBHI supports the likelihood that this enzyme is a CBHI-type cellulase. Differences in access to the active site appear to influence the kinetics of these enzymes on small substrate. In fact, compared to cellobiohydrolase Cel7A from T. reesei, binding of the natural product cellobiose is about five times weaker in cellobiohydrolase Cel7D from P. chrysosporium CBHI (Ki = 0.18 mM versus 0.02 mM) and even 100 times weaker in P. occitanis CBHI (Ki = 2 mM versus 0.02 mM)

[11–13]. Such Ki value with P. occitanis CBHI has not previously been reported for enzymes in family 7 of glycoside hydrolases. The strong cellobiose inhibition of T. reesei cellobiohydrolase I was clearly reduced (Ki = 0.3 mM) by the deletion of the tip of the loop forming the active site tunnel roof [13]. Despite the sequence conservation between T. reesei cellobiohydrolase I and P. occitanis CBHI in this region (G267-Y274), the latter enzyme exhibited less pronounced product inhibition without any need to delete this region. 3.7. Structure prediction of the catalytic domain of P. occitanis CBHI In order to investigate the CBHI resistance to cellobiose inhibition at a molecular level, a 3D model of CBHI was constructed, on the basis of the X-ray crystallographic structure of the cellobiohydrolase of the thermophilic fungus T. emersonii (pdb accession code 1Q9H). CBHI from P. occitanis shares 71% of identity with that of T. emersonii. As shown in Fig. 6A, P. occitanis CBHI folds into an antiparallel ␤-sheet jellyroll architecture typical of the characteristic fold of GH7 family. This fold is present in an increasingly large number of proteins and was first described for the plant lectins such as concanavalin A [45]. The core catalytic module of the protein is arranged in two large antiparallel ␤-sheets stacking on top each other in a sandwich like manner, and are highly curved, forming convex and concave surfaces which consist of seven antiparallel strands each, that are bent and create a cleft crossing one side

80

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81

Table 1 Distances (Å) between some sub-sites and the substrate analog in the active sites of Phanerochaete chrysosporium, Trichoderma reesei and Penicillium occitanis cellobiohydrolases. The residue numbering corresponds to that of Penicillium occitanis. Residues

Phanerochaete chrysosporium

Trichoderma reesei

Penicillium occitanis

Asp 195 Tyr 167 Glu 234 Trp 63 Asn 163

4.38 2.63 4.39 3.05 3.12

3.96 2.58 4.48 2.75 2.73

4.53 2.72 4.5 3.30 3.24

of the protein where the substrate is bound. Many of the side chains in the ␤-sheets are hydrophobic, and interactions between these residues appear to hold the ␤-sandwich in position. With the exception of four ␣-helices and two pairs of short ␤-strands, the rest of the protein consists almost entirely of loops connecting the ␤-strands. The loops extending from the ␤-sandwich are stabilized by the presence of nine disulphide bonds which are located between residues 44–50, 75–96, 86–92, 160–425, 194–232, 198–231, 252–278, 260–265 and 283–359. The cleft on the concave side of the molecule defines the oligosaccharide substrate binding site. It is lined with mainly aromatic residues on its walls and with acidic residues at the bottom. The catalytic residues are located in the same ␤-strand where there is a strict alternation of polar and nonpolar side chains, the first pointing toward the surface of the protein where they are able to interact with the substrate, the latter toward the hydrophobic interior. Three catalytic acidic amino acids are conserved between the members of glycoside hydrolase family 7 [35]. These catalytic residues were observed in the P. occitanis CBHI (E234, D236 and E239). Their positions in P. occitanis CBHI are identical to those in T. reesei cellobiohydrolase I, suggesting that they should perform the same function in these enzymes. In the acid/base reaction mechanism, Glu234 should be the nucleophile and Glu239 the proton donor. Asp236 is likely to be involved in maintenance of the appropriate pKa values for the other catalytic residues, assuring the correct ionization state of the active site during catalysis, as has been suggested for T. reesei cellulases [46]. Comparison of the active site of cellobiohydrolase I from T. reesei, P. chrysosporium and P. occitanis showed that all the interactions made with cellobiose are conserved, and the three molecules shared together the same sub-sites. This observation takes more importance when the interaction distances between the analog of substrate and each sub-site are compared. In fact, among the three enzymes, the distances are larger in P. occitanis CBHI compared to P. chrysosporium and T. reesei cellobiohydrolases I, suggesting a more open configuration of the active site. The differences in distances are estimated to be about 0.02–0.6 Å between P. occitanis CBHI and T. reesei cellobiohydrolase I. As a consequence, the affinity between cellobiose and the active site may be affected (Table 1). Thus, poor hydrogen bonding, together with the more open CBHI active site configuration (Fig. 6B), provides a reasonable explanation for why cellobiose binds less tightly to this enzyme than to T. reesei and P. chrysosporium cellobiohydrolases making it less sensitive to product inhibition. 4. Conclusion In this paper, the isolation of the cDNA encoding cellobiohydrolase I from the hypercellulolytic P. occitanis strain was described. The whole cbh1 gene was subsequently cloned from a genomic bank constructed in this fungus. Southern analysis of genomic DNA indicated that there is only a single intronless gene encoding for the cellobiohydrolase

I in this strain. Unlike the hyperpectinolytic mutant CT1 and the parental strain CL100, the hypercellulolytic Pol6 mutant was shown to be partially constitutive since its cbh1 mRNA, besides being highly induced by cellulose, is slightly repressed by glucose. In many aspects, the P. occitanis CBHI protein is a typical cellulase: it has a modular structure with a CBM in the C-terminus joined to the N-terminus catalytic core by a linker rich in serine and threonine residues. Comparative modelling of the structure of P. occitanis, P. chrysosporium and T. reesei cellobiohydrolases I suggested that the resistance to product inhibition exhibited by P. occitanis CBHI is due to the poor hydrogen bonding to cellobiose associated with a more open configuration of the active site. Acknowledgements We express our gratitude to Prof. G. Tiraby and Dr. H. Durand (Cayla Company, France) for kindly supplying the Penicillium occitanis strains used in this work. N. Fradi is thanked for supplying the Trichoderma CBHI probe. We also gratefully acknowledge Prof. R. Ellouz for his continual support and his interest for the subject. This work was supported by Ministère de l’Enseignement Supérieur, de la Recherche Scientifique et de la Technologie. References [1] Enari TM. Microbial cellulases. I. In: Fagarty NM, editor. Microbial enzymes and biotechnology. London: Applied Science Publisher; 1983. p. 183–223. [2] Bhat MK. Cellulases and related enzymes in biotechnology. Biotechnol Adv 2000;18:355–83. [3] Divne C, Ståhlberg J, Teeri TT, Jones TA. High-resolution crystal structures reveal how a cellulose chain is bound in the 50 Å long tunnel of cellobiohydrolase I from Trichoderma reesei. J Mol Biol 1998;275:309–25. [4] Tomme P, Warren RA, Miller Jr RC, Kilburn DG, Gilkes NR. Cellulose-binding domains: classification and properties. In: Saddler JN, Penner M, editors. Enzymatic degradation of insoluble polysaccharides. Washington: American Chemical Society; 1995. p. 63–142. [5] Henrissat B, Claeyssens M, Tomme P, Lemesle L, Marnon J-P. Cellulase families revealed by hydrophobic cluster analysis. Gene 1989;81:83–95. [6] Kraulis PJ, Glore GM, Nilges M, Jones TA, Pettersson G, Knowles J, et al. Determination of the three dimensional structure of the C-terminal domain of the cellobiohydrolase I from Trichoderma reesei. A study using nuclear magnetic resonance and hybrid distance geometry-dynamical simulated annealing. Biochemistry 1989;28:7241–57. ˇ IG, Ubhayasekera W, Henriksson H, Szabó I, Pettersson G, Johansson G, [7] Munoz et al. Family 7 cellobiohydrolases from Phanerochaete chrysosporium: crystal structure of the catalytic module of Cel7D (CBH 58) at 1.32 Å resolution and homology models of the isozymes. J Mol Biol 2001;314:1097–111. [8] Jain S, Parriche M, Durand H, Tiraby G. Production of polysaccharidases by a cellulase–pectinase hyperproducing mutant (Pol6) of Penicillium occitanis. Enzyme Microb Technol 1990;12:691–6. [9] Ellouz Chaabouni S, Belguith H, Hsairi I, M’rad K, Ellouz R. Optimisation of cellulases production by Penicillium occitanis. Appl Microbiol Biotechnol 1995;43:267–9. [10] Hadj-Taieb N, Chaabouni-Ellouz S, Kammoun A, Ellouz R. Hydrolytic efficiency of Pencillium occitanis cellulase: Kinetic aspects. Appl Microbiol Biotechnol 1992;37:197–201. [11] Limam F, Ellouz-Chaabouni S, Ghrir R, Marzouki N. Two cellobiohydrolases of Penicillium occitanis mutant Pol 6: purification and properties. Enzyme Microb Technol 1995;17:340–6. [12] Tuohy MG, Walsh DJ, Murray PG, Claeyssens M, Cuffe MM, Savage AV, et al. Kinetic parameters and mode of action of the cellobiohydrolases produced by Talaromyces emersonii. Biochem Biophys Acta 2002;1596:366–80. [13] Ossowski IV, Ståhlberg J, Koivula A, Piens K, Becker D, Boer H, et al. Engineering the Exo-loop of Trichoderma reesei cellobiohydrolase, Cel7A. A comparison with Phanerochaete chrysosporium Cel7D. J Mol Biol 2003;333:817–29. [14] Shoemaker S, Schweickart V, Lander M, Gelfand D, Kwok S, Myambo K, et al. Molecular cloning of exocellobiohydrolase I from Trichoderma reesei strain L27. Bio/Technology 1983;1:691–6. [15] Sims P, James C, Broda P. The identification, molecular cloning and characterisation of a gene from Phanerochaete chrysosporium that shows strong homology to the exo-cellobiohydrolase I gene from Trichoderma reesei. Gene 1988;74:411–22. [16] Koch A, Weigel CTO, Schultz G. Cloning, sequencing and heterologous expression of a cellulase-encoding cDNA (cbh1) from Penicillium janthinellum. Gene 1993;124:57–65.

F. Bhiri et al. / Enzyme and Microbial Technology 46 (2010) 74–81 [17] De Vries RP, Visser J. Aspergillus enzymes involved in degradation of plant cell wall polysaccharides. Microbiol Mol Biol Rev 2001;65:497–522. [18] Hadj-Taieb N, Ayadi M, Trigui S, Bouabdallah F, Gargouri A. Hyperproduction of pectinase activities by a fully constitutive mutant (CT1) of Pencillium occitanis. Enzyme Microb Technol 2002;30:662–6. [19] Mandels M, Weber J. The production of cellulases. Adv Chem Ser 1969;95:391–412. [20] Aifa MS, Sayadi S, Gargouri A. Heterologous expression of Lip H8 of Phanerochaete chrysosporium in Aspergillus niger. Biotechnol Lett 1999;21: 849–53. [21] Hopwood DA, Bibb MJ, Chater KF, Kieser T, Bruton CJ, Kieser HM, et al. Genetic manipulation of Streptomyces. A laboratory manual; 1985. pp. 152–3. [22] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Harbor, NY: Spring; 1989 (ED). [23] Corpet F. Multiple sequence alignment with hierarchical clustering. Nucl Acids Res 1988;16(22):10881–90. [24] Gouet P, Robert X, Courcelle E. ESPript/ENDscript: extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res 2003;31:3320–3. [25] Nielsen H, Engelbrecht J, Brunak S, Von Heijne G. A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Int J Neural Syst 1997;8:581–99. [26] Schwede T, Kopp J, Guex N, Peitsch MC. SWISS-MODEL: an automated protein homology-modelling server. Nucleic Acids Res 2003;31:3381–5. [27] DeLano WL. The PyMOL molecular graphics system DeLano scientific. CA, USA: San Carlos; 2002. [28] Trigui-Lahiani H, Gargouri A. Cloning, genomic organisation and mRNA expression of a pectin lyase gene from a mutant strain of Penicillium occitanis. Gene 2007;388:54–60. [29] Penttilä M, Lehtovaara B, Bhikhabbai R, Knowles J. Homology between cellulase genes of Trichoderma reesei: complete nucleotide sequences of endoglucanase I gene. Gene 1986;45:253–63. [30] Teeri TT, Lehtovaara P, Kauppinen S, Salovuori I, Knowles J. Homologous domains in Trichoderma reesei cellulolytic enzymes: gene sequence and expression of cellulobiohydrolase II. Gene 1987;51:43–52. [31] Takada G, Kawaguchi T, Sumitani J, Arai M. Cloning, nucleotide sequence, and transcriptional analysis of Aspergillus aculeatus No.F-50 cellobiohydrolase I (cbhI) gene. J Ferment Bioeng 1998;85:1–9. [32] Alcocer MJ, Furniss CS, Kroon PA, Campbell M, Archer DB. Comparison of modular and non-modular xylanases as carrier proteins for the efficient secretion of heterologous proteins from Penicillium funiculosum. Appl Microbiol Biotechnol 2003;60:726–32.

81

[33] Covert SF, Vanden Wymelenberg A, Cullen D. Structure, organization and transcription of a cellobiohydrolase gene cluster from Phanerochaete chrysosporium. Appl Environ Microbiol 1992;58:2168–75. [34] Wey TT, Hseu TH, Huang L. Molecular cloning and sequence analysis of the cellobiohydrolase I gene from Trichoderma koningii G-39. Curr Microbiol 1994;28:31–9. [35] Nunberg JH, Meade JH, Cole G, Lawyer FC, McCabe P, Schweickart V, et al. Molecular cloning and characterization of the glucoamylase gene of Aspergillus awamori. Mol Cell Biol 1984;4:2306–15. [36] Breathnach R, Chambon P. Organization and expression of eukaryotic split genes coding for proteins. Annu Rev Biochem 1981;6:336–45. [37] Suto M, Tomita F. Induction and catabolite repression mechanisms of cellulase in fungi. J Biosci Bioeng 2001;92:305–11. [38] Saloheimo A, Aro N, Ilmen M, Penttilä M. Isolation of the ace 1 gene encoding a Cyc2 -His2 transcription factor involved in regulation of activity of the cellulase promoter cbh1 of Trichoderma reesei. J Biol Chem 2000;275:5817–25. [39] Aro N, Saloheimo A, Ilmen M, Penttilä M. ACEII, a novel transcriptional activator involved in regulation of cellulase and xylanase gene of Trichoderma reesei. J Biol Chem 2001;276:24309–14. [40] Penttila M, Lehtovaara B, Bhikhabbai R, Knowles J. Homology between cellulose genes of Trichoderma reesei: complete nucleotide sequences of endoglucanase I gene. Gene 1986;45:253–63. [41] Morawetz R, Gruber F, Messner R, Kubicek CP. Presence, transcription and translation of cellobiohydrolase Trichoderma species. Curr Genet 1992;21:31–6. [42] Linder M, Mattinen M, Kontteli M, Lindeberg G, Ståhlberg J, Drakenberg T, et al. Identification of functionally important amino acids in the cellulose-binding domain of Trichoderma reesei cellobiohydrolase I. Protein Sci 1995;4:1056–64. [43] Linder M, Lindeberg G, Reinikainen T, Teeri TT, Pettersson G. The difference in affinity between two fungal cellulose-binding domains is dominated by a single amino acid substitution. FEBS Lett 1995;372:96–8. [44] Nidetzky B, Steiner W, Claeyssens M. Cellulose hydrolysis by the cellulases from Trichoderma reesei: adsorptions of two cellobiohydrolases, two endocellulases and their core proteins on filter paper and their relation to hydrolysis. Biochem J 1994;303:817–23. [45] Edelman GM, Cunningham BA, Reeke Jr GN, Becker JW, Waxdal MJ, Wang LJ. The covalent and three-dimensional structure of concanavalin A (X-ray crystallography/sequence/2-A resolution/binding sites/lectin). Proc Nat Acad Sci 1972;69:2580–4. [46] Ståhlberg J, Divine C, Koivula A, Piens K, Claeyssens M, Teeri TT, et al. Activity studies and crystal structures of catalytically deficient mutants of cellobiohydrolase I from Trichoderma reesei. J Mol Biol 1996;264:337–49.