Molecular characterization of the phenol oxidase (pox2) gene from the ligninolytic fungus Pleurotus ostreatus

Authors


  • Editor: Geoffrey Gadd

Correspondence: Tarek A.A. Moussa, Department of Botany, Faculty of Science, Cairo University, Giza 12613, Egypt. Tel.: +20 101 531 738; fax: +20 237 229 253; e-mail: tarekmoussa@cu.edu.eg

Abstract

The gene (pox2) encoding a phenol oxidase from Pleurotus ostreatus, a lignin-degrading basidiomycete, was sequenced and the corresponding pox2-cDNA was also synthesized, cloned and sequenced. The isolated gene consisted of 2674 bp, with the coding sequence interrupted by 19 introns and flanked by an upstream region in which the putative metal-responsive elements (MREs) were determined in the promoter region (849 bp), where MRE 1, 2, 3 and 4 were located in positions −20, −60, −236 and −297. A functional TATA consensus sequence was recognized in position −85, while CAAT and its inversion consensus sequences were recognized in positions −284, −554, −689 and −752. The putative GC box consensus sequences were recognized in positions −181 and −460, and xenobiotic-responsive elements in positions −107, −277 and −390. The isolation of a second cDNA (pox2-cDNA), the nucleotide sequence of pox2, was found to contain an ORF of 1665 bp capable of coding for a protein of 533 amino acid residues. Northern blot analysis revealed that strong transcriptional induction was observed in the copper-supplemented cultures for the pox2 gene.

Introduction

White rot fungi are the most active microorganisms degrading lignin, a complex aromatic biopolymer that is extremely recalcitrant to degradation (Kirk & Farrell, 1987). These fungi produce different oxidative enzymes, with a broad substrate specificity, which can also be used to degrade a vast range of toxic aromatic pollutants (Hammel, 1995; Rodriguez et al., 1999).

The existence of multiple genes encoding different laccase isoenzymes has been demonstrated in several fungi (Yaver & Golightly, 1996; Mansur et al., 1997; Smith et al., 1998; Giardina et al., 1999). Moreover, laccase gene expression depends on the culture conditions, and differentially regulated systems to control laccase production have been reported (Yaver et al., 1996; Collins & Dobson, 1997; Munöz et al., 1997; Mansur et al., 1998). Among the various inducers tested, copper ions considerably increase laccase gene transcription in several fungi (Collins & Dobson, 1997; Karahanian et al., 1998; Palmieri et al., 2000; Soden & Dobson, 2001; Galhaup et al., 2002).

Laccases are phenol oxidases, which reduce oxygen to water and simultaneously perform a one-electron oxidation of many aromatic substrates (Leonowicz et al., 2001). The substrate range of these enzymes can be extended to include nonphenolic lignin subunits in the presence of readily oxidizable primary substrates, which can act as electron-transfer mediators (Bourbonnais et al., 1997). Laccases belong to the group of blue copper oxidases and contain four copper atoms/molecule, distributed among three different copper-binding sites (Solomon et al., 1997). The structure and organization of laccase copper-binding sites are shared by other multicopper proteins, which have physiological roles not apparently related to phenol oxidase activity (copper tolerance, iron transport and sporulation) (Solano et al., 2001).

Fungal multicopper oxidases are receiving increasing interest as potential industrial enzymes in applications such as detoxification of toxic phenolic compounds and azo dyes (Husain & Jan, 2000), enzymatic bleaching of kraft pulp (Bourbonnais et al., 1995) and delignification (Youn et al., 1995) because these oxidases catalyze the oxidation of phenols.

In this study, the author designed primers to amplify the pox2 gene (DNA and cDNA) and also for the promoter regions that extend 5′-upstream of the initiation codon ATG to analyze and determine the putative sequences that control the transcription of the gene. Also, Northern blot analysis to determine the effect of copper supplementation on the regulation of the transcription was performed.

Materials and methods

Bacterial and fungal strains and plasmid

For standard bacterial cloning, Escherichia coli DH5α (Hanahan, 1983) was grown in Luria–Bertani (LB) medium (Sigma) supplemented with 10 μg mL−1 of ampicillin. The white rot fungus Pleurotus ostreatus (Jacq.) NRRL0366 (oyster mushroom) was maintained by periodic transfer at 4 °C on potato dextrose agar (3.9%, Oxoid, UK) plates in the presence of 0.5% yeast extract (Oxoid). The pGEM®-T Easy Vector system I (Promega, Madison) was used in subcloning of pox2-cDNA.

Cultivation of P. ostreatus

Incubations were carried out at 28±2 °C by inoculating 100 mL of potato dextrose broth containing 0.5% yeast extract and/or supplemented with different concentrations of CuSO4·5H2O solution in 500-mL flasks with three discs of P. ostreatus. The cultures were incubated in the dark on a rotary shaker (100 r.p.m. min−1). At different incubation times, the mycelia were harvested by filtration and kept at −40 °C until used.

Isolation of genomic DNA

DNA was isolated using the mixer mill isolation protocol. The mycelia were ground in liquid N2 and suspended in a DNA isolation buffer [50 mM Tris-HCl (pH 7.9), 250 mM NaCl, 10 mM EDTA (pH 8.0) and 0.5% sodium dodecyl sulfate (SDS) 0.5%] with a metal ball. The tubes were placed in a mixer mill (4 min at 15 Hz, followed by 20 s at 20 Hz). The tubes were then centrifuged at 1050 g for 15 min at room temperature (RT), and the supernatant was transferred to new Eppendorf tube. Then, 10 μL RNAse (10 mg mL−1) was added and incubated for 10 min at 65 °C and for 30 min at 37 °C, followed by the addition of 600 μL of PCI and centrifugation at 1050 g for 10 min at RT. The supernatant was transferred to a new Eppendorf tube and c. 250 μL chloroform was added and centrifuged at 1050 g for 10 min at RT. The supernatant was transferred to a new Eppendorf tube and c. 600 μL isopropanol was added and incubated for 30 min at RT, followed by centrifugation at 1050 g for 10 min at 4 °C. The pellet was washed using 70% EtOH (c. 100 μL), and then centrifuged at 1050 g for 5 min at 4 °C. The pellet was dissolved in 100 μL water and heated at 60 °C for 20 min for complete dissolution. A final centrifugation at 1050 g for 5 min at 4 °C was performed. About 90 μL was transferred to a new Eppendorf tube.

Isolation of total RNA and PCR for pox2

RNA was isolated using RNA isolation solution (Omega Bio-Tek Inc.). The cDNA sequence of pox2 was analyzed by reverse-transcribed PCR from total RNA using Superscript (Invitrogen). DNA and cDNA were used for PCR amplification of the pox2 gene in 50-μL reaction volumes containing 1 μL template, 4 μL dNTPs, 5 μL buffer 3 (containing MgCl2), 2 μL of each primer and 1 μL expand high fidelity PCR enzyme (Roche Applied Science). The primers were synthesized at Microsynth Co. (Switzerland) (Table 1).

Table 1.   List of primers used in this study
TargetPrimer sequenceProduct size (nt)
GenomiccDNA
pox2 promoter
Pox2.1 (Fw)CGGTTGCGGAGGTCGTAG849
Pox2.2 (Rv)ATGTTTCCAGGCGCACGG  
pox2-DNA
Pox2.2 (Fw)ATGTTTCCAGGCGCACGG26741665
Pox2.3 (Rv)AACGAAAACCTTTCGACGTG  
pox2-cDNA
Pox2.2 AscI (Fw)ATTGGCGCGCCATGTTTCCAGGCGCACGG26741665
Pox2.3 AscI (Rv)ATTGGCGCGCCAACGAAAACCTTTCGACGTG  

cDNA cloning and transformation

The PCR product was eluted from the gel using the MicroEluteTM gel extraction kit (Omega Bio-Tek Inc.) and digested using AscI restriction enzyme. The pGEM-T easy vector (Promega) was digested using AscI. The digested PCR product was ligated to the vector using a ligation kit (BioLabs, UK). The cloned gene was transformed inside E. coli DH5α cells (Stratagene) following standard procedures (Ausubel et al., 1992). In a precooled 15-mL tube, 200 or 400 μL of E. coli cells were transferred, followed by the ligation reaction, and mixed gently. The mixture was incubated for 20–30 min on ice, then heat shocked for 90 s at 42 °C and cooled for 2 min on ice. The mixture was made up to 1 mL with LB and different volumes of the cells were plated on LB containing ampicillin, and incubated at 37 °C overnight.

Miniprep plasmid isolation

The isolated colonies indicating positive transformation were selected and transferred to tubes containing 3 mL LB medium with ampicillin. They were incubated overnight at 37 °C with shaking. The cultures were centrifuged in Eppendorf tubes at 225 g at 4 °C for 5 min, and then the pellet was suspended in 300 μL of buffer A (5 mM Tris-HCl, 10 mM EDTA and 400 μg mL−1 RNAse A). Three hundred microliters of buffer B was added (0.2 M NaOH, 1% SDS), mixed by inverting and kept at RT for 5 min. Three hundred microliters of buffer C was added (2.55 M potassium acetate, pH 4.8) and kept on ice for 5 min. The mixture was centrifuged at 1050 g for 10 min at 4 °C. The supernatant was transferred to a new Eppendorf tube and 600 μL isopropanol was added, incubated at RT for 15–20 min. The mixture was centrifuged at 1050 g for 10 min at 4 °C. The pellet was washed using 70 μL of 70% ethanol, dried and resuspended in 20 μL water.

Sequencing of the pox2 gene

Nucleotide sequences were determined using the ABI Prism Big Dye Terminator Cycle Sequencing kit (Applied Biosystems) on ABI automated sequencers (ABI 3100), carried out at the Biocenter, Innsbruck Medical University, Austria. The genomic nucleotide sequence for the pox2 gene is available on the GenBank database with accession no. AB474261 assigned to the cDNA. Nucleotide and amino-acid sequence similarity searches used the blast method (Altschul et al., 1990) from the National Center for Biotechnology Information databases.

Northern blot analysis

Ten micrograms of total RNA was electrophoresed on 1.2% (w/v) agarose–2.2 M formaldehyde gels and blotted onto Hybond N membranes (Amersham Biosciences). The hybridization probes used in this study were generated by PCR using the oligonucleotides 5′-ATGTTTCCAGGCGCACGG-3′ and 5′-AACGAAAACCTTTCGACGTG-3′.

Results

Isolation and analysis of the laccase (phenol oxidase) genomic sequence were performed. Using a PCR method, which involved two pairs of gene-specific primers, three PCR products were obtained, which showed strong homology to known basidiomycete laccase genes. Based on the sequence analysis, two fragments were formed: one was about 2674 bp, which is the coding sequence of the pox2 gene, and the other fragment was about 849 bp and extended to the 5′-noncoding region. Thus, the entire structure of the pox2 gene of P. ostreatus could be determined; the coding sequence was 2674 bp long, interrupted by 19 introns varying in size from 48 to 58 bp (Fig. 1), when compared with the third PCR fragment Pox2-cDNA sequence. All of the introns splice junctions corresponded to the GT–AG rule (Fig. 1). Putative regulatory sites such as metal-responsive elements (MREs) and xenobiotic-responsive elements were identified in the pox2 promoter region, which, extending about 849 bp upstream of the start codon, are shown in Fig. 2. The putative MREs were determined in the promoter region, where MRE 1, 2, 3 and 4 were located in positions −20, −60, −236 and −297, while xenobiotic-responsive elements were located in positions −107, −277 and −390. A functional TATA consensus sequence was recognized in position −85, while CAAT and its inversion consensus sequences were recognized in positions −284, −554, −689 and −752. The putative GC box consensus sequences were recognized in positions −181 and −460 (Fig. 2).

Figure 1.

Figure 1.

 Nucleotide sequence of the Pleurotus ostreatus pox2 gene and the deduced amino acid sequence of POX2. Introns are shown as lower-case letters and indicated as INS. The putative signal peptide is underlined. The forward and reverse primers are also underlined.

Figure 1.

Figure 1.

 Nucleotide sequence of the Pleurotus ostreatus pox2 gene and the deduced amino acid sequence of POX2. Introns are shown as lower-case letters and indicated as INS. The putative signal peptide is underlined. The forward and reverse primers are also underlined.

Figure 2.

 Nucleotide sequence of the Pleurotus ostreatus pox2 promoter region, extending 850 bp upstream of the start codon (ATG). Transcription-initiation site is indicated by vertical arrow. The putative TATA box, GCs box, CAAT box, MREs and xenobiotic-resposive elements (XREs) are underlined. Primers used for amplification are indicated as bold and underlined.

First-strand cDNA was reverse transcribed from mRNA of a 5-day-old mycelium culture. An amplification experiment was performed using the primer AscI. The fragment which counted for the pox2 cDNA, was cloned using pGEM-T easy vector. The nucleotide sequence of pox2 was found to contain an ORF of 1665 bp (Fig. 3), capable of coding for a protein of 533 amino acid residues. The alignment of sequence analysis of this fragment with the previously determined nucleotide sequence led to the definition of the gene (pox2-cDNA, accession no. AB474261).

Figure 3.

Figure 3.

 Multiple sequence alignment of nucleotide sequences of phenol oxidase (pox2-cDNA), phenol oxidase 2 (pox2), bilirubin oxidase (box) and laccase (lacc) of Pleurotus ostreatus. The primers used for amplification are underlined. Alignment was performed using clustalw 2.0 software (http://www.ebi.ac.uk/Tools/clustalw2/index.html).

Figure 3.

Figure 3.

 Multiple sequence alignment of nucleotide sequences of phenol oxidase (pox2-cDNA), phenol oxidase 2 (pox2), bilirubin oxidase (box) and laccase (lacc) of Pleurotus ostreatus. The primers used for amplification are underlined. Alignment was performed using clustalw 2.0 software (http://www.ebi.ac.uk/Tools/clustalw2/index.html).

Figure 3.

Figure 3.

 Multiple sequence alignment of nucleotide sequences of phenol oxidase (pox2-cDNA), phenol oxidase 2 (pox2), bilirubin oxidase (box) and laccase (lacc) of Pleurotus ostreatus. The primers used for amplification are underlined. Alignment was performed using clustalw 2.0 software (http://www.ebi.ac.uk/Tools/clustalw2/index.html).

The isolated gene codes for a protein of 533 amino acids. The encoded amino acid sequence is reported in Fig. 4. The multiple alignment of the deduced amino acid sequence of POX2 shared significant homology with phenol oxidase 2 (POX2) 100%, bilirubin oxidase (BOX) 97% and laccase (Lacc) 96% of P. ostreatus in the Databank (Table 2). The copper-binding domain structure found in other laccase genes is conserved in the P. ostreatus laccase protein (Fig. 4).

Figure 4.

 Multiple sequence alignment of the predicted amino acid sequence of phenol oxidase (POX2-cDNA), Phenol oxidase 2 (POX2), bilirubin oxidase (BOX) and laccase (Lacc) of Pleurotus ostreatus. Four potential copper-binding domains are indicated as bold and underlined. Four Cys residues involved in the formation of two disulfide bridges are shaded and underlined. Alignment was performed using clustalw 2.0 software (http://www.ebi.ac.uk/Tools/clustalw2/index.html).

Table 2.   Comparison of the phenol oxidase 2 (pox2) cDNA sequence (accession no. AB474261) with published sequences of this gene and other related genes from Pleurotus ostreatus
GenePercentage homologyNucleotide accession no.
nt levelaa level
Pox2100100Z34848
Box9397AB020026
lacc9396AY450404

The results of the Northern blot experiments are shown in Fig. 5, which also shows the results of the quantification analysis of the intensity of each band. Strong transcriptional induction was observed in the copper-supplemented cultures for the pox2 gene. The amount of mRNA decreased after 2 days and there was an increase from the third day onwards for copper supplementations. Pleurotus ostreatus pox2 expression was found to be upregulated by copper supplementations.

Figure 5.

Pleurotus ostreatus pox2 expression is upregulated during copper supplementations. Total RNA was isolated from P. ostreatus following growth for 5 days with copper starvation (0 μM) and different copper concentration conditions, and subjected to Northern blot analysis.

Discussion

Although laccase production in white rot fungi is known to be influenced by a number of factors, little work has been carried out to study the regulation of laccase gene expression at the molecular level (Kalisz et al., 1986; Thurston, 1994). The positions of the 19 introns showing the canonical splicing junctions are conserved. The intron sequences were based on comparison with the published sequences (Kojima et al., 1990; Saloheimo et al., 1991; Coll et al., 1993b; Morohoshi, 1993) and consensus sequences for 59 splicing GT(AG)(AT)GT and 39 splicing (CT)AG junctions present in filamentous fungi (Ballance, 1986).

In the 5′-flanking region of the pox2 promoter, several sequences have been identified that match closely the consensus of regulatory elements, in particular, a MRE (Thiele, 1992), and a xenobiotic-responsive element (Fujisawa-Sehara et al., 1988). In the poxa1b promoter, a GC-rich region, homologous to the core binding site for transcription factor Sp1, decreases the binding affinity of the adjacent MRE, affecting its interactions with fungal protein factors (Faraco et al., 2003).

Nucleotide sequences of the poxc and poxa1b promoter regions, extending about 400 nt upstream of the start codon (ATG), have been analyzed, and multiple putative regulatory sites such as MREs, xenobiotic-responsive elements and heat-shock elements have been identified in them. The sequences of all MREs are similar to the core MRE consensus sequence (5′-TGCRCNC-3′) identified in metallothionein (mt) gene promoters (Thiele, 1992). Other laccase promoters have been reported to contain multiple putative MRE sites (Karahanian et al., 1998; Mansur et al., 1998; Klonowska et al., 2001; Galhaup et al., 2002).

Multiple putative MREs elements show nucleotide sequences similar to the core MRE consensus sequence TGCPuCXC, which is known to be involved in the metal response of mt genes in higher eukaryotes (Hagen et al., 1988; Greco et al., 1990; Hill et al., 1991). Metal-regulated gene transcription systems play important roles in metal homeostasis and detoxification (Kägi & Shäffer, 1988). The best-characterized example of a metal-regulated transcription system is that of the mt genes. In mt promoters from higher eukaryotes, multiple copies of MREs constitute the cis-acting sequences responsible for heavy-metal induction of mt gene expression (Culotta & Hamer, 1989). The role of metallothioneins in protection from metal toxicity correlates with the ability of several metal ions, including zinc, copper, cadmium and others, to activate mt gene transcription (Hamer, 1986). Mechanisms of metal regulation have so far been elucidated for mt gene transcription systems (Andersen et al., 1987, 1990; Mueller et al., 1988; Zhou & Thiele, 1991) and for some other metal-responsive transcription systems (Jin & Ringertz, 1990; Williams & Morimoto, 1990; Carri et al., 1991; Merchant et al., 1991). Regulation of mt genes occurs via a metal-regulatory protein that functions both as a metal receptor and as a transcription factor.

Laccase genes have been isolated from several basidiomycetes (Kojima et al., 1990; Saloheimo et al., 1991; Giardina et al., 1995; Berka et al., 1997). The sequences of these genes display a common pattern in that they encode polypeptides of c. 520 to 550 amino acid residues including an N-terminal signal peptide (Coll et al., 1993a; Giardina et al., 1995; Eggert et al., 1998).

In addition, the single cysteine residue and 10 histidine residues involved in binding the four catalytic cupric ions found in each laccase molecule are conserved, together with a small amount of sequence around the four regions in which the copper ligands are clustered (Thurston, 1994; Eggert et al., 1998). It is in the copper-binding amino acid residues and their general distribution in the polypeptide chain that the laccases are all similar (Coll et al., 1993a; Giardina et al., 1995; Eggert et al., 1998). Alignment of the polypeptide sequence derived from lac1 with the sequences derived from other basidiomycete laccase genes shows that the domain structure of the Lac1 protein is conserved. Lac1 showed the conserved sequences in the single cysteine residue and 10 histidine residues. The N-terminal lac1 sequence is separated from the C-terminal catalytic domain by a hinge region (Thurston, 1994). The latter appears to be duplicated, but is typically rich in serine residues.

Laccases are copper-containing oxidases that catalyze the four-electron oxidation of a variety of phenolic compounds and a simultaneous four-electron reduction of oxygen to water. The PCR strategy used in this study is based on the use of degenerate primers corresponding to the consensus sequences conserved in the copper-binding regions in the N-terminal domains of known basidiomycete laccases (Kojima et al., 1990; Messerschmidt & Huber, 1990; Saloheimo et al., 1991; Coll et al., 1993a; Morohoshi, 1993; Perry et al., 1993; Thurston, 1994).

In this study, Northern blot analyses clearly revealed that copper had a marked effect on induction of pox2 gene transcription. In addition, the pox2 transcript was the most abundant transcript in the copper-supplemented cultures at all of the times analyzed. Pleurotus ostreatus pox2 expression was found to be upregulated by copper supplementations. Collins & Dobson (1997) have found that the expression of laccase in Trametes versicolor was regulated at the level of gene transcription by copper and nitrogen. As the concentration of copper or nitrogen in fungal cultures was increased, an increase in laccase activity corresponding to increased laccase gene transcription was observed. Zhao & Kwan (1999) used HN medium supplemented with copper to study the effects of physiological parameters on laccase expression in Lentinula edodes. The addition of copper sulfate to P. ostreatus growth medium causes a marked increase of total laccase activity and a transcription induction of poxc and, mostly, poxa1b genes (Palmieri et al., 2000).

Acknowledgements

I thank Prof. Dr. Hubertus Haas, Division of Molecular Biology, Biocenter, Innsbruck Medical University, Innsbruck, Austria for providing facilities for this work.

Ancillary