SEARCH

SEARCH BY CITATION

Keywords:

  • fungal transformation;
  • phenol oxidase;
  • fluorescence

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

In silico analyses of several laccase promoter sequences have shown the presence of many different responsive elements differentially distributed along the promoter sequences. Analysis of Pleurotus ostreatus laccase promoter poxa1b extending around 1400-bp upstream of the start codon showed the presence of several putative response elements, such as 10 metal-responsive elements. Development of a system for in vivo analysis of P. ostreatus laccase promoter poxa1b by enhanced green fluorescent protein expression was carried out, based on a polyethylene glycol–mediated procedure for fungal transformation. Quantitative measurement of fluorescence expressed in P. ostreatus transformants grown in the presence and in the absence of copper sulfate was performed, demonstrating an increase in expression level induced by the metal.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

Twelve putative laccase genes have been identified in the recently sequenced Pleurotus ostreatus genome (http://www.jgi.doe.gov/sequencing/why/50009.html), one of which is annotated as a ferroxidase-like. The promoter regions of all the 11 P. ostreatus laccase genes, extending 500-bp upstream of the start codon, have been analyzed, revealing the presence of several putative response elements, differentially distributed along the promoter sequences (Piscitelli et al., 2011). All the analyzed P. ostreatus laccase promoters contain putative metal-responsive elements (MREs) with sequence homology to those reported in ascomycetous yeast. Cis-acting MREs have been discovered in multiple copies in the Saccharomyces cerevisiae metallothionein promoter where they are essential for efficient metal-inducible transcription (Thiele, 1992). Sequences similar to MREs have been also found in several laccase promoters of basidiomycetous fungi such as the promoter region of the gene coding for the major laccase isoenzyme LAP2 from Trametes pubescens (Galhaup et al., 2002), the promoter region of the copper-inducible LAC2 laccase from Gaeumannomyces graminis (Litvintseva & Henson, 2002), the promoter region of the strongly copper-induced lac4 gene from Pleurotus sajorcaju (Soden & Dobson, 2003), and the promoters of three laccase genes (lacA, lacB, and lacC) from Trametes sp. AH28-2 (Xiao et al., 2006).

The presence of putative MREs in P. ostreatus laccase promoters is consistent with the observation that the level of laccase activity production by the fungus increases substantially in copper-supplemented cultures and the copper induction on expression of POX isoenzymes acts at the level of gene transcription (Palmieri et al., 2000). It is worth noting that poxa1b mRNA was the most abundant induced transcript at all of the growth times analyzed. Analyses of the region P. ostreatus poxa1b promoter extending around 500-bp upstream of the ATG had allowed individuation of four putative MREs (Piscitelli et al., 2011), all being recognized by fungal proteins as shown by electromobility shift assays (Faraco et al., 2003). MRE-like sequences involved in formation of complexes with fungal proteins have been identified by footprinting analyses of the poxa1b promoter that showed the occurrence of a large protected region including a1bMRE2 and a1bMRE3 sites with opposite orientations (Faraco et al., 2003). Besides increasing expectation of their roles in regulation of laccase expression, no physiological function of these putative MREs could be confirmed, because of lack of appropriate promoter assay systems in basidiomycetes. Indeed, development of an efficient transformation system of the fungus P. ostreatus is needed to perform in vivo analysis of these laccase promoter elements, in view of their mutagenesis for laccase overproduction.

In this work, a system for enhanced green fluorescent protein (GFP) expression under the control of laccase promoter poxa1b in P. ostreatus was developed, based on a polyethylene glycol (PEG)–mediated fungal transformation procedure. Analysis of effect of copper sulfate addition to fungal growth medium on fluorescence expression driven by poxa1b promoter in P. ostreatus showed an increase in expression level induced by the metal.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

Strain and culture media

Pleurotus ostreatus dikaryotic strain #261 (ATCC 66376) was used as the host strain for transformation experiments. Maintenance of the strain was performed on PDY [2.4% potato dextrose (Difco, Detroit, Michigan), 0.5% yeast extract (Difco), 1.5% agar (Difco)] medium at 28 °C.

Liquid cultures of P. ostreatus transformants were prepared pre-inoculating 75 mL of PDY broth in 250-mL Erlenmeyer flasks with six agar plugs (11 mm diameter) of P. ostreatus mycelium, from 7-day-old agar culture, in a temperature-controlled incubator at 28 °C on rotary shaker (at 125 rpm). The cultures were prepared by inoculating 80 mL of PDY broth in 250-mL Erlenmeyer flask with 8 mL of 5-day-old pre-inocula. When indicated, CuSO4 (150 μM) was added to the cultures.

Vector construction

The GenBank accession number of the sequence of the P. ostreatus laccase poxa1b gene (Giardina et al., 1999) reported in this paper is AJ005017.

To prepare the vector pEGFPea1b (Fig. 1a) designed to study the poxa1b promoter through enhanced GFP gene expression in P. ostreatus, the poxa1b terminator and the poxa1b promoter were amplified by PCR using plasmid vectors selected from the P. ostreatus genomic library (Giardina et al., 1995, 1996, 1999) as templates and the gene-specific oligonucleotides Termpoxa1bXbaI/Termpoxa1bPstI and Prompoxa1SacIrev/Prompoxa1SacIfw (Table 1) as primers, respectively. The amplified fragment of poxa1b terminator was subjected to hydrolysis with the restriction enzymes XbaI and PstI and ligated into the XbaI-/PstI-digested pUC13 vector, giving the vector pA1BTERM.

Table 1. List of the primers used in the amplification experiments (oligonucleotides of the rows 1–6) and oligonucleotides for intron/exon fragment preparation (oligonucleotides of the rows 7–8)
 OligonucleotideNucleotide sequences
  1. Nucleotides in bold are complementary to the sequences to be amplified. Nucleotides in italics are recognized or leaved by restriction enzymes, nucleotides in lower-case type belong to intron sequences

1Termpoxa1b XbaIGCTCTAGATTATAATTTGACTT
2Termpoxa1b PstICATTGGTTCTGCAGCCAGCGT
3Prompoxa1 SacIAATTGCGAGCTCGGGGAGATC
4Prompoxa1 EcoRIGGAATTCGGCGCATCAATACC
5EGFP 3dirCGGGATCCGTGAGCAAGGGCGAGGAG
6EGFP 5revGCTCTAGACTTGTACAGCTCGTCCATG
7EGFP 1dir C ATGCCACCTGgtatgcatacttccccctgcggcccgtattgctgctaacctgcatttctccagCCGCG G
8EGFP 1revGATCCCGCGGctggagaaatgcaggttagcagcaatacgggccgcagggggaagtatgcatacCAGGTGGCATGAGCT
image

Figure 1. Map of the vectors pEGFPea1b (a) and pEGFPCBX (b), adopted to study the poxa1b promoter through enhanced GFP gene expression in Pleurotus ostreatus with and without cotransformation with pTM1 vector, respectively.

Download figure to PowerPoint

An intron/exon fragment was prepared by annealing of the synthetic oligomers EGFP1dir and EGFP1rev having complementary sequences including poxc gene intron number XIX flanked by two amino acids at the 5′ end and three at 3′ end (Giardina et al., 1996) and the sticky ends features of the restriction enzymes SacI and BamHI, followed by digestion by SacI and BamHI.

The egfp gene was amplified by PCR using the plasmid vector pEGFP-C1 (Clontech Laboratories, Inc., CA) as template and the gene-specific oligonucleotides EGFP3dir/EGFP5rev as primers (Table 1). The plasmid vector pA1BTERM was subjected to hydrolysis by the endonucleases SacI and XbaI, and ligation reaction among the amplified egfp gene, the intron/exon fragment, and the linearized pA1BTERM vector was carried out. The vector thus obtained was subjected to SacI/EcoRI hydrolysis and ligated to the amplified poxa1b promoter fragment after SacI/EcoRI digestion, giving the pEGFPea1b vector.

The vector pEGFPCBX (Fig. 1b) was constructed by cloning the DNA fragment resulting from NotI/SphI hydrolysis of pTM1 into pEGFPea1b vector. This fragment includes the gene cbxR and its own promoter and terminator. To include the desired restriction sites NotI/SphI within pEGFPea1b, this vector was hydrolyzed by the enzymes SphI–EcoRI, and an oligonucleotide whose sequence contains the polylinker EcoRI–NotI–SphI was then ligated. Ligation between the DNA fragment excised from pTM1 and pEGFPea1b hydrolyzed by NotI and SphI was then carried out.

Transformation of P. ostreatus

Liquid cultures of P. ostreatus for protoplasting were set up by inoculating 60 mL YMG broth [1% glucose, 0.4% yeast extract (Difco), 1% malt extract] in 250-mL cotton plugged Erlenmeyer flasks with six agar plugs (11 mm diameter) of P. ostreatus mycelium, grown on PDA [2.4% potato dextrose (Difco)] medium. The inocula were incubated in a temperature-controlled incubator at 28 °C on a rotary shaker (at 120 rpm). The biomass was homogenized and retrieved in a 100-mL single flask. Recovered mycelium was incubated for 5 h in a temperature-controlled incubator at 28 °C on rotary shaker (at 120 rpm). The biomass was transferred in two 50-mL Falcon conical tubes. The samples were washed twice with deuterium-depleted water and twice with 0.5 M sucrose by centrifuging at 450 g for 8 min. The pellets were recovered into one tube. Enzyme digestion solution consisting of 200 mg of lysing enzyme from Trichoderma harzianum (Sigma-Aldrich SRL, Milano, Italy) and 20 mg of chitinase from Trichoderma viride (Sigma-Aldrich SRL) was dissolved by ultrasonic machine in 10 mL of 0.5 M sucrose and filtered by 0.22-μm PVDF membrane (Millipore S.p.A., Vimodrone, Italy). Enzyme digestion solution was added to the sample that was incubated at 31 °C for 3 h on a rotary shaker (at 50 rpm). Next, 0.5 M sucrose was added to the sample up to 50 mL. The sample was centrifuged at 450 g for 8 min and washed twice with STC [0.5 M sucrose, 0.05 mM Tris–HCl (pH 8.0) solution with 18.2% sorbitol and 2.22% CaCl2 anhydrate] to remove enzymatic solution. Protoplasts were resuspended in 4 mL of STC solution.

For transformation, 200 μL of this protoplast solution was gently mixed with 15 μL of heat-denaturated λ phage DNA (0.3 γ/λ; Fermentas) and transforming DNA (1 μg of pTM1 or 1 μg of pTM1 and 5 μg of pEGFPea1b or 1 μg pEGFPCBX). Samples were incubated on ice for 40 min. Then, 1 mL of PTC [0.5 M sucrose, 0.05 mM Tris–HCl (pH 8.0) solution with 40% PEG#4000 (Sigma-Aldrich SRL), 17.2% sucrose, 8.88% CaCl2 anhydrate] was added. The sample was mixed gently at RT, then incubated at RT for 20 min. Protoplast solution (600 μL) was spread on regeneration medium (1% glucose, 0.4% yeast extract, 1% malt extract, 17.1% sucrose, 1.5% agar) containing 2 μg mL−1 of carboxin (Sigma-Aldrich). Plates were incubated at 28 °C.

Extraction of intracellular proteins

Pleurotus ostreatus 7-day-old liquid cultures prepared as described in the first paragraph of this section in the presence of 2 μg mL−1 of carboxin were filtered through sterilized cotton lint to retrieve suspended mycelia. Recovered mycelium was frozen and then lyophilized. Mycelium was crushed in porcelain mortar and then suspended in the extraction buffer containing 100 mM Tris–HCl pH 7.5, 2.5 mM EDTA, 7 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, and 1% Triton (Sigma-Aldrich). After centrifuging at 15 000 g at 4 °C for 15 min, supernatant was recovered for further assays. Protein concentration was determined by the method of Lowry et al. (1951), using the BioRad Protein Assay (BioRad Laboratories S.r.l., Segrate, MI – Italy), with bovine serum albumin as standard.

Intracellular GFP determination

The crude supernatant was diluted to 0.05 mg of protein per mL with the extraction buffer above reported, and a fluorescence spectrum (500–600 nm) was determined using a 460 nm excitation wavelength with a LS 50B Fluorescence Spectrometer (Perkin-Elmer). Maximum fluorescence occurred at 520 nm. The entity of fluorescence emission was measured as difference between spectrum area recorded between 500 and 550 nm for the transformant and that of the control sample (nontransformed fungus). The experiments were performed in three replicates, and reported values are representative of two experiments.

Fluorescence microscopy

Pleurotus ostreatus mycelia were grown on microscope coverslips and observed in a NIKON ECLIPSE TE 2000-U microscopic system with appropriate fluorescein isothiocyanate filters (Nikon Corporation, Tokyo, Japan). Normal phase-contrast images of each sample were used as controls. The digital image was further processed using photoshop 5.0 (Adobe).

Detection of the introduced sequence in the transformants

Chromosomal high-molecular weight DNA from P. ostreatus was prepared as described by Raeder & Broda (1988). Amplification experiments were carried out on 50 ng of genomic DNA in a 50 μL total volume, using the gene-specific oligonucleotides EGFP 3dir and EGFP 5rev (Table 1) as primers and Taq DNA polymerase (Invitrogen, Carlsbad, CA). Polymerase chain reaction (PCR) conditions consisted of 30 cycles of 94 °C (1 min), 58 °C (45 s), and 72 °C (2 min) plus an additional final chain elongation step at 72 °C for 10 min.

Southern hybridization analysis of the transformants

Genomic DNA from the transformants was isolated (Raeder & Broda, 1988), digested with the restriction enzymes EcoRI, BamHI, and PstI (Promega, Italy), and after electrophoresis on 0.8% agarose gel, transferred to a Hybond-NX nylon membrane (GE Healthcare). The membrane was hybridized using the PCR-amplified egfp sequence as radioactive probe, as previously described (Palmieri et al., 2000).

Detection of egfp transcript in the transformants

Total RNAs were extracted from lyophilized mycelia of transformants using Qiagen RNeasy Plant (Qiagen, Italy) and following manufacturer's instructions. Reverse transcription reaction was performed using MultiScribe™ Reverse Transcriptase (Applied Biosystems, Branchburg, NJ) and the oligonucleotide dT-NotI as primer. Products of the PCR experiments, performed using the gene-specific oligonucleotides EGFP3dir/EGFP5rev (Table 1), were analyzed on 1% agarose gel.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

In silico analysis of poxa1b, poxc, and poxa3 promoters

Analysis of the P. ostreatus poxa1b, poxc, and poxa3 promoter regions extending around 1400-bp upstream of the ATG was performed searching for the putative response elements heat shock element (HSE, repeated NGAAN motif; Mager & De Kruijff, 1995), NIT2 binding site (TATCT; Marzluf, 1997), antioxidant response element (ARE, TGACNNNGC; Soden & Dobson, 2003), putative response elements PRE (ATATC and TGGGT motifs; Soden & Dobson, 2003), MRE (TGCRCNC; Thiele, 1992), xenobiotic responsive elements (XRE TNGCGTG; Xiao et al., 2006), Cre-A-binding site (GCGGGG; Litvintseva & Henson, 2002), and stress-responsive element (STRE, CCCCT; Galhaup et al., 2002). Several putative response elements were identified differentially distributed along the promoter sequences (Fig. 2). The highest number (10) of putative MREs was identified within the poxa3 and poxa1b promoters, in the latter case consistently with previous data of poxa1b transcription induction by copper addition to fungal growth medium (Palmieri et al., 2000).

image

Figure 2. Distribution of putative cis-acting elements in the promoter regions of Pleurotus ostreatus laccase genes poxc, poxa1b, and poxa3 around 1400-bp upstream of the start codons: (a) TATA box; (b) CAAT box; (c) GC box; (image) HSE; (image) NIT2; (image) ARE; (image) MRE; (image) XRE; (image) Cre-A-binding site; (image) STRE.

Download figure to PowerPoint

Expression plasmids and P. ostreatus transformation

Considering the high number of putative MREs present in poxa1b promoter, it can be considered as a good candidate as regulated promoter to be used to drive homologous recombinant expression in the fungus P. ostreatus. To develop a system for in vivo analysis of poxa1b promoter and its metal regulation, the gene-encoding GFP was adopted as reporter gene putting its expression under the control of 1400-bp-long poxa1b promoter region. GFP of the jellyfish Aequorea victorea emits fluorescence as a result of its intrinsic chromophore structure, not requiring any substrate or cofactor (Chalfie et al., 1994), and it represents a versatile reporter gene (Cubitt et al., 1995).

The vector pEGFPea1b for in vivo analysis of P. ostreatus laccase promoters was constructed using the gene coding for enhanced GFP (EGFP). A P. ostreatus poxa1b promoter region of 1336 bp was used as cis-regulatory element to drive expression of EGFP. An intron/exon fragment containing an intron/exon sequence of the poxc gene was included between the poxa1b promoter and the egfp gene, considering previous results showing that efficient GFP expression in Agaricus bisporus and Coprinus cinereus (Burns et al., 2005) and Phanerochaete chrysosporium (Ma et al., 2001) requires introns.

A homologous selection marker, the mutant gene cassette CbxR, encoding a modified iron–sulfur protein Ip subunit of succinate dehydrogenase with an aminoacid substitution (His239 to Leu) and conferring resistance to systemic fungicide carboxin (Honda et al., 2000), was adopted.

Cotransformation with pTM1 vector conferring carboxin resistance and pEGFPea1b vector containing egfp gene under the control of poxa1b promoter region was carried out, by adopting an adapted version of transformation protocol reported by Salame et al. (2010).

Moreover, an unique vector containing both the mutant gene cassette CbxR and the reporter cassette poxa1b promoter-egfp gene was constructed and adopted for transformation.

Transformants were firstly screened for carboxin resistance. The carboxin-resistant colonies were subjected to at least four rounds of selection by transferring on fresh selection medium. Around 50 carboxin-resistant transformants were obtained per μg of pTM1 DNA per 107 viable protoplasts in a transformation with pTM1 and pEGFPea1b, and five carboxin-resistant transformants were obtained per μg of pEGFPCBX DNA per 107 viable protoplasts in a transformation with this vector. Hence, cotransformation with vectors containing gene cassette CbxR and the reporter cassette poxa1b promoter-egfp gene allowed a 10-fold higher transformation efficiency than transformation with an unique vector containing both cassettes. This could be ascribed to the larger size of the latter construct. The carboxin-resistant transformants were further analyzed for checking the presence of egfp and fluorescence emission.

Molecular analyses

Carboxin-resistant transformants were analyzed by PCR to verify the presence of the transforming DNA. The PCR was performed on genomic DNA extracted from transformants mycelia. Six of the nine analyzed transformants showed the expected 0.7-kb target band, indicating the presence of the egfp gene in the transformants (Fig. 3).

image

Figure 3. Electrophoretic analysis of PCR products obtained on genomic DNA extracted from Pleurotus ostreatus transformants. The analysis was performed on 1% agarose gel. Lane 1: DNA Molecular Weight Marker III (Roche Applied Science, Indianapolis, IN); lane 2: PCR products obtained on genomic DNA extracted from nontransformed mycelium; lanes 3–9: PCR products obtained on genomic DNA extracted from mycelium cotransformed with the vectors pEGFPea1b and pTM1 (clone n.1, n.5, n.6, n.7, n.8, n.10); lanes 9–11: PCR products obtained on genomic DNA extracted from mycelium containing the vector pEGFPCBX (clone n.2a, n.2a*, n.43).

Download figure to PowerPoint

Southern hybridization analysis of the transformants 5 and 43 was carried out to analyze the mode of integration of the transforming DNA (Fig. 4). The non transformed mycelium does not show any hybridization. The transformants 5 and 43 showed a different pattern of bands. The transformant 43 showed single bands in each digestion. For the transformant 5, several bands of various sizes were observed. These results demonstrated that the introduced sequence was integrated ectopically into the chromosomal DNA with one or more copy numbers in these transformants.

image

Figure 4. Southern hybridization analysis of the transformants. Lanes 1, 2, and 3: EcoRI-, BamHI-, PstI-digested DNA extracted from the transformant 5; lanes 4, 5, and 6: EcoRI-, BamHI-, PstI-digested DNA extracted from the transformant 43; lanes 7, 8, and 9: EcoRI-, BamHI-, PstI-digested DNA extracted from the nontransformed mycelium.

Download figure to PowerPoint

Transcription of egfp in the transformants 5 and 43 was demonstrated by RT-PCR (Fig. 5).

image

Figure 5. Electrophoretic analysis of RT-PCR products obtained on total RNA extracted from Pleurotus ostreatus transformants. The analysis was performed on 1% agarose gel. Lane 1: DNA Molecular Weight Marker III (Roche Applied Science); lane 2: PCR products obtained on RNA extracted from nontransformed mycelium; lanes 3–4: PCR products obtained on RNA extracted from the transformants 43 and 5, respectively.

Download figure to PowerPoint

EGFP visualization

Detection of fluorescence was performed in vivo on 2 days grown transformants mycelia on microscopic slides. In Fig. 6, phase-contrast micrographs of transformants (a) and the corresponding images under UV light (b) are shown. Nontransformed mycelium did not show any fluorescence. Scanned images show a positive fluorescence emission with respect to untransformed control. Fluorescence emission extended to entire hyphae, especially to clamps connection. Similar phenomenon was also observed when poxc promoter-driven reporter plasmid was used for transformation (to be published elsewhere).

image

Figure 6. Phase-contrast images (a) and fluorescence micrographs (b) of transformed mycelium containing the plasmids pEGFPa1b and pEGFPCBX (1: transformant 5; 2: transformant 43). The green color in the mycelium is due to GFP.

Download figure to PowerPoint

Intracellular GFP determination

The P. ostreatus transformants 1, 5, 2, and 43 were analyzed for intracellular fluorescence emission by measuring emission of fluorescence of intracellular protein extracts from 7-day-old mycelium in comparison with the control (nontransformed mycelium; Fig. 7). The entity of fluorescence emission was measured as difference between spectrum area recorded between 500 and 550 nm for the transformant and that of the control sample (nontransformed fungus). The expression of GFP in each of the transformants has proved stable over a 6-month period of repeated subculturing on selective media (data not shown).

image

Figure 7. Bar graph showing the fluorescence emission intensity at 500–550 nm of some of the transformants (transformants number 1, 5, 2, and 43), after subtracting the background fluorescence emission of control (nontransformed fungus).

Download figure to PowerPoint

Difference in intracellular fluorescence emission was revealed for different transformants that could be ascribed to the different copy numbers and loci of exogen DNA integration within the fungal genome. Variation in GFP concentration among independent fungal transformants has been observed by other authors (Chalfie et al., 1994; Cubitt et al., 1995).

Comparison of intracellular fluorescence emission by transformants growth in the presence and in the absence of copper sulfate showed that metal addition causes an increase in green fluorescence driven by the poxa1b promoter, up to fourfold (20 000 fluorescence unit per 0.05 mg of proteins).

It is worth noting that an induction of transcription from a particular promoter sequence was hereby demonstrated by quantitative measurement of fluorescence emission for the first time in basidiomycetes.

GFP has been successfully expressed in the homobasidiomycetes Schizophyllum commune (Lugones et al., 1999), P. chrysosporium (Ma et al., 2001), A. bisporus, and C. cinereus (Burns et al., 2005). A number of factors such as inactivation of transforming DNA by preferential methylation (Mooibroek et al., 1990), inactivation of gene expression of AT-rich sequences (Schuren & Wessels, 1998; Scholtmeijer et al., 2001), need of introns for mRNA accumulation (Lugones et al., 1999; Scholtmeijer et al., 2001; Burns et al., 2005) seem to hamper transgene expression of GFP in basidiomycetes. Moreover, in a few manuscripts so far reported on transformation of P. ostreatus with the GFP gene, green fluorescence after transformation was unstable (Li et al., 2006), or no quantitative measurement of protein expression was reported (Ding et al., 2011).

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

In this manuscript, a transcriptional induction of a laccase promoter was demonstrated in P. ostreatus by enhanced GFP expression, based on a PEG-mediated procedure for fungal transformation. The promoter of poxa1b was chosen among the different P. ostreatus laccase promoters, because it contains the highest number of putative MREs sites and poxa1b transcript is the most copper-affected among the P. ostreatus laccase transcripts.

Cotransformation with pTM1 vector conferring carboxin resistance and pEGFPea1b vector containing egfp gene under the control of poxa1b promoter region was carried out and compared to transformation with the unique pEGFPCBX vector containing both carboxin resistance cassette and poxa1b promoter-egfp gene cassette. The presence of egfp gene was demonstrated in most of the carboxin-resistant transformants. Southern hybridization analysis of the transformants 5 (cotransformed with pTM1 and pEGFPea1b vectors) and 43 (transformed with the unique pEGFPCBX vector) showed that the introduced sequence was integrated ectopically into the chromosomal DNA with one or more copy numbers. Transcription of egfp in the transformants 5 and 43 was also demonstrated. An intracellular fluorescence emission up to around 5,000 (Units per 0.05 mg of protein) in comparison with the nontransformed mycelium was measured. No significant difference of fluorescence emission was observed comparing pEGFPea1b and pEGFPCBX transformants. However, a less transformation efficiency was achieved using the bigger pEGFPCBX vector. By analyzing intracellular fluorescence emission by transformants growth in the presence of copper sulfate, an increase in green fluorescence was revealed up to 20 000 fluorescence unit per 0.05 mg of proteins, providing in vivo demonstration of susceptibility of poxa1b laccase promoter to the metal.

The developed system allowed both in vivo demonstration of copper-induction of expression driven by poxa1b promoter and its quantitative analysis. This will allow investigation of the role of putative metal response elements present in this promoter.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References

The authors are grateful to Prof. Giovanni Sannia, Department of Chemical Sciences, University of Naples ‘Federico II’, Prof. Yitzhak Hadar and Mr. Tomer M. Salome of The Hebrew University of Jerusalem, Israel, and Prof. Ursula Kües and Dr Martin Ruhl of Georg-August-University Göttingen, Germany, for their assistance and helpful discussions to make simple and efficient transformation protocol in P. ostreatus.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgments
  8. References