Decolourization of azo dyes by Pleurotus ostreatus, a white-rot fungus capable of lignin depolymerization and mineralization, is related to the ligninolytic activity of enzymes produced by this fungus. The capacity of P. ostreatus to decolourize the azo dye Orange II (OII) was dependent and positively co-linear to Mn2+ concentration in the medium, and thus attributed to Mn2+-dependent peroxidase (MnP) activity. Based on the ongoing P. ostreatus genome deciphering project we identified at least nine genes encoding for MnP gene family members (mnp1–9), of which only four (mnp1–4) were previously known. Relative real-time PCR quantification analysis confirmed that all the nine genes are transcribed, and that Mn2+ amendment results in a drastic increase in the transcript levels of the predominantly expressed MnP genes (mnp3 and mnp9), while decreasing versatile peroxidase gene transcription (mnp4). A reverse genetics strategy based on silencing the P. ostreatus mnp3 gene by RNAi was implemented. Knock-down of mnp3 resulted in the reduction of fungal OII decolourization capacity, which was co-linear with marked silencing of the Mn2+-dependent peroxidase genes mnp3 and mnp9. This is the first direct genetic proof of an association between MnP gene expression levels and azo dye decolourization capacity in P. ostreatus, which may have significant implication on understanding the mechanisms governing lignin biodegradation. Moreover, this study has proven the applicability of RNAi as a tool for gene function studies in Pleurotus research.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Pleurotus ostreatus is a commercially important edible white-rot basidiomycete known as the oyster mushroom. The ligninolytic system of Pleurotus species has been found to be mainly composed of the lignin-modifying enzymes (LMEs) laccase, aryl-alcohol oxidase and two types of peroxidases: Mn2+-dependent peroxidase (MnP) and versatile peroxidase (VP). All of these enzymes may function separately or in cooperation (Cohen et al., 2002a; Stajićet al., 2009). Versatile peroxidase has been suggested to be a MnP-LiP (lignin peroxidase) ‘hybrid’, based on its ability to oxidize different substrates in the presence or the absence of Mn2+, its crystal structure and theoretical molecular models (Camarero et al., 1999; Ruiz-Dueñas et al., 2001; Martínez, 2002). In recent years, different Mn2+ peroxidases, designated MnP1–4, have been purified and characterized from Pleurotus species and their corresponding genes (mnp1–4) were sequenced and analysed (Asada et al., 1995; Ruiz-Dueñas et al., 1999; Giardina et al., 2000; Irie et al., 2000). The peroxidases MnP1, MnP2 and MnP4 were characterized as VPs while MnP3 was characterized as a MnP (Cohen et al., 2001).
Manganese peroxidases are the most common ligninolytic peroxidases produced by almost all white-rot basidiomycetes and by various litter-decomposing fungi. The remarkable degradative potential of MnP makes this enzyme an attractive versatile biocatalyst for biotechnological and environmental applications, e.g. in pulping and bleaching of lignocellulose, in removing of hazardous wastes or in certain organic syntheses (Hofrichter, 2002; Cohen et al., 2002a; Hammel and Cullen, 2008; Stajićet al., 2009). Effective harnessing of these applications is dependent on establishing a comprehensive understanding of the enzyme's properties and mode of action. The ability of lignin-degrading white-rot fungi (WRF) to degrade a wide range of synthetic chemicals, including dyes (many of which are recalcitrant to biodegradation) has been reported and linked to the ligninolytic process, i.e. the activity of LMEs that take part in the lignin depolymerization process, which is non-specific in nature. These enzymes play significant roles in dye metabolism due to the structural similarity of most commercially relevant dyes to lignin (sub)structures amenable to be transformed by LMEs. So far, WRF have been found to be the most efficient microorganisms in degrading synthetic dyes, and thus are currently being investigated in the development of bioremediation solutions for these polluting xenobiotics (Wesenberg et al., 2003; Asgher et al., 2008; Faraco et al., 2009). Additionally, on the basis of the accumulated evidence, several methods involving dyes have been reported as screening methodologies for investigation of the ligninolytic process. The use of dyes offers a number of advantages over conventional natural substrates because they are stable, soluble and affordable with high molar extinction rates and low toxicity. These dyes can be applied in simple, quick and quantitative spectrophotometric assays, thereby serving as a basis for simple phenotypic-based assays for the functionality of the ligninolytic system (Glenn and Gold, 1983; Platt et al., 1985; Field et al., 1993; Shrivastava et al., 2005; Hernández-Luna et al., 2008; Lucas et al., 2008). In this study, the azo dye Orange II (OII) was chosen as a model compound because it is a representative of a large class of extensively used commercial dyes, and its complex structure presents several characteristics (such as its aromatic structure, azo linkage and a sulfonic group) that are relevant to many xenobiotics and contaminants. This dye was previously shown to be decolourized by P. ostreatus (Knapp et al., 1995). Moreover, its decolourization by the white-rot basidiomycetes strain F29 (Knapp et al., 1997) and Bjerkandera sp. (Mielgo et al., 2003; López et al., 2004) was found to be positively affected by Mn2+ amendments, which may suggest the involvement of Mn2+ peroxidases. Decolourization of OII servers as a measure of its degradation, as a higher UV-VIS absorbance correlates with a higher dye concentration (Mielgo et al., 2001; 2003).
Mn2+ ions are naturally present in wood and lignocellulose residues. Several studies have shown the effect of Mn2+ on enhancing lignin degradation and mineralization by Pleurotus species, suggesting the importance of MnP in the process (Camarero et al., 1996; Kerem and Hadar, 1997; Cohen et al., 2001; 2002b). Cohen and colleagues (2001) used semi-quantitative RT-PCR to monitor the relative expression levels of the four mnp genes as affected by Mn2+ amendment. That study described a reduction in the abundance of VP genes transcript and an increase in mnp3 transcript, which were co-linear with the changes observed in the MnP enzymes' activity profiles. These results have indicated the importance of MnP in lignin degradation and that transcriptional regulation plays a role in the process. Nevertheless, most of the information regarding the significance of Mn2+-dependent peroxidase in this process has been derived from hypotheses based on indirect findings.
The feasibility of affecting gene expression in P. ostreatus by genetic manipulation is an invaluable tool for the dissection of the LMEs functionality in this fungus. Honda and colleagues (2000) developed a PEG-CaCl2-meditated method for transformation and recombinant gene expression system in P. ostreatus, based on the homologous drug-resistant marker gene CbxR, which confers dominant resistance to the systemic fungicide carboxin. This system has since been used for the homologous expression of recombinant Mn2+ peroxidase genes (Irie et al., 2001a; Tsukihara et al., 2006). However, homologous recombination has been shown to occur only rarely, rendering gene knockout studies difficult (Honda et al., 2000; Irie et al., 2001a).
RNA interference (RNAi) is a post-transcriptional gene-silencing phenomenon in which double-stranded RNA (dsRNA) triggers the degradation of homologous mRNA, thereby diminishing or abolishing gene expression. Three fundamental components of the RNA silencing machinery have been identified: Argonaute, Dicer and RNA-dependent RNA polymerase (RdRP) (Tomari and Zamore, 2005; Nakayashiki et al., 2006; Nakayashiki and Nguyen, 2008). These have been identified in a wide range of fungi including ascomycetes, basidiomycetyes and zygomycetes, many of which harbour multiple RNA silencing components in the genome, whereas a portion of ascomycete and basidiomycete fungi apparently lacks most of or whole of the components. In addition to being recognized as a principal mechanism for controlling eukaryotic gene expression, the natural occurrence of this phenomenon has also been harnessed as a tool for elucidating gene function. As an efficient reverse genetics approach this technique involves the introduction or production of dsRNA molecules homologous to the gene being targeted for silencing in the cell of interest. RNAi has been proven effective in most eukaryotes, including vertebrates, plants, invertebrates, protists and fungi (Nakayashiki et al., 2006; Nakayashiki and Nguyen, 2008).
The P. ostreatus (monokaryon PC15) genome sequencing project has been recently completed by the DOE JGI (http://genome.jgi-psf.org/PleosPC15-1). The availability of the genome sequence, and the fact that the fungus is amenable to genetic modifications makes P. ostreatus accessible for comprehensive functional genomics studies. This has prompted us to study the involvement of MnPs in the degradation of aromatic substrates, using available and modified tools for gene manipulation in this fungus. To do so we facilitated a reverse genetics strategy of silencing the P. ostreatus mnp3 gene using an RNAi-based approach, in combination with a comprehensive analysis of the expression levels of MnP gene family members. Consequently, we determined the effects of silencing mnp3 on fungal growth, levels of MnP gene family expression in response to Mn2+ amendment, and the significance of Mn2+-dependent peroxidases for the functionality of P. ostreatus ligninolytic system as evaluated by OII decolourization.
Orange II decolourization is Mn2+ dependent
The capacity of the white-rot fungus P. ostreatus strain PC9 to decolourize OII was evaluated both on solid media and in liquid culture, in the presence of Mn2+ at several concentrations ranging 0–270 µM. Mn2+ concentration in the non-amended medium was determined by atomic absorption spectroscopy and was found to be less than 0.1 µM. On solid medium, linear growth rate was not affected by the Mn2+ amendments, yet decolourization was apparent only at concentrations above 8.1 µM, and its intensity was increased with elevation of Mn2+ concentration in the medium (Fig. 1A). In the absence of Mn2+ no visible changes in OII colour intensity were observed even after 30 days of incubation. Media containing Mn2+ concentrations higher than 54 µM showed formation of dark precipitation foci of MnO2 (López et al., 2007). Consequently, for further analysis on solid media we used Mn2+ at a concentration of 27 µM. Decolourization was also monitored in liquid culture for 10 days (Fig. 1B), showing a similar dependency on Mn2+ concentration. The fungal dry weight in liquid culture was similar in all treatments (average of 256 mg/flask after 10 days of incubation, with a ±4% deviation). OII decolourization did not occur in non-inoculated media, at any of the examined Mn2+ concentrations, even after 30 days of incubation. Since this reaction depends on the presence of Mn2+ and there was no decolourization in the absence of Mn2+, it was attributed to Mn2+-dependent peroxidase activity. Accordingly, OII decolourization can serve as a differential phenotypic assay for the expression level of these enzymes and was used here to assess mnp3 silenced strains.
P. ostreatus harbours more than one Mn2+-dependent peroxidase
The P. ostreatus genome sequencing project has revealed the existence of at least nine non-allelic genes coding for MnP gene family members (Table 1; http://genome.jgi-psf.org/PleosPC15-1). To date, only four of these genes (mnp1–4) have been studied (Cohen et al., 2001; 2002b). Of these known genes, only mnp3 encodes a Mn2+-dependent peroxidase, whereas the others encode VPs (Mn2+-independent peroxidases). We designated the additional five genes mnp5–9 (Table 1). The deduced protein sequences of mnp5–9 indicate that MnP6, 7, 8 and 9 are Mn2+-dependent peroxidases, whereas MnP5 is most likely a VP (Asada et al., 1995; Ruiz-Dueñas et al., 1999; Giardina et al., 2000; Irie et al., 2000; Cohen et al., 2001). In order to study the effect of Mn2+ on the transcription profile of P. ostreatus MnP gene family members we used relative real-time PCR quantification analysis. The fungus was grown for 7 days in either a liquid medium amended with 27 µM Mn2+ (+Mn treatment) or a non-amended medium (−Mn treatment). The results presented in Fig. 2 show the relative expression of the nine different MnP gene family members. The primers used for real-time PCR analyses (Table 1) were verified to be gene-specific, and examination of melting curves indicated highly specific amplification of the respective cDNAs (data not shown). The endogenous control gene used was β-tubulin, and the calibrator was the −Mn treatment. Transcripts of all the nine mnp genes were detected in both Mn2+-amended and non-amended cultures. However, Mn2+ in the medium affected the transcript abundance level of the mnp genes analysed in different manners. The transcripts levels of mnp3 and mnp9 were about 200-fold higher when Mn2+ was present in the medium; conversely, mnp4 transcript abundance was about 70-fold higher in the non-amended medium. The transcript levels of all the other genes were increased by only 1.4- to 3-fold in the presence of Mn2+, and were therefore not considered to be substantially induced. In addition, absolute quantification of the basal expression levels of the MnP gene family members, based on evaluation deduced from the real-time PCR amplification plots, indicate that, relative to the treatment, the transcript levels of mnp3, 4 and 9 were at least eightfold higher than those of the other mnp genes. Similar results were observed on the basis of semi-quantitative RT-PCR. The abundance of the β-tubulin transcript was not affected by the presence of Mn2+. We therefore concluded that the presence of Mn2+ can affect different mnp genes in opposing ways.
Table 1. Sequences of oligonucleotides used in this study.
In the case of data bank entries, GenBank and JGI P. ostreatus genome database are followed with accession numbers and protein ID numbers respectively. Note: referenced is the source of the primary sequences, all sequences were compared with JGI genome database which served as basis for primers design.
Orange II decolourization capacity is inhibited in strains harbouring mnp3 RNAi constructs
To study the effect of reduced Mn2+-dependent peroxidases gene expression, we transformed P. ostreatus with pTMS1, a construct designed to induce RNAi-based gene silencing of mnp3. Construct design was initially based on mnp3, a gene that was later found to be highly similar to mnp9 in terms of Mn2+-dependent gene regulation. pTMS1 (Fig. 3) was designed to express a single-stranded RNA that forms a loop (linker region), corresponding to basepairs 100–1 of the mnp3 cDNA antisense strand, with a double-stranded stem, corresponding to basepairs 750–101 of the mnp3 cDNA antisense, annealed to basepairs 101–750 of the mnp3 sense strand. Formation of a hairpin RNA (hpRNA) structure, including the inverted repeat stem with a strong dsRNA secondary structure (a single 650-bp-long double-stranded stem in which the free energy of the structure is −1315.3 kkal mol−1) was verified by RNA secondary structure prediction tools (http://www.genebee.msu.su/genebee.html).
The modified transformation procedure developed during the course of this study proved to be efficient and reproducible, yielding over 20 transformants per µg vector DNA; occurrence of abortive transformants was rare.
Thirty pTMS1 transformants were isolated, alongside one randomly chosen control strain (TC3) transformed with pTM1 (Honda et al., 2000). All the transformants were grown in the presence of carboxin, and were found stable as they remained resistant to carboxin after three transfers in medium lacking the fungicide.
In order to determine whether pTMS1 transformants were affected in their ability to decolourize OII, the different strains were grown on solid GP medium containing OII, amended with 27 µM of Mn2+, under carboxin selection pressure. Growth and decolourization, after 10 days of incubation, were monitored and compared with the untransformed wild-type strain PC9 (Fig. 4A). TC3 exhibited similar growth (±3%) and decolourization (±5%) rates as PC9. pTMS1 transformants showed marked variability in their growth and OII decolourization capabilities. Of these, four strains, designated TS1, TS9, TS24 and TS30, which grew similarly to the wild-type (88–94%), but showed high inhibition (90–99%) in their OII decolourization capacity in comparison to PC9 and TC3, were selected for further examination (Fig. 4B).
OII decolourization by the four selected strains was measured in liquid GP medium amended with 27 µM Mn2+ for 10 days (Fig. 4C). The control strain, TC3, exhibited similar decolourization levels as the PC9 wild-type strain throughout the experiment. In contradiction, during the first 8 days of incubation, the pTMS1 transformants (TS1, TS9, TS24 and TS30) exhibited a significant lower rate of OII decolourization (albeit, at different levels). After this period extensive decolourization was observed in all strains. All the tested strains produced a similar biomass (239 ± 24 mg/flask) over the mentioned time period. Based on these results, which were consistent in both solid and liquid cultures, it can be concluded that Mn2+-dependent peroxidase expression is indeed reduced in the examined strains, as a consequence of the genetic manipulation performed.
A control experiment was conducted to examine the effect of Mn2+ on the ability of the MnP silenced transformants to decolourize the anthraquinone dye Acid blue 62, whose decolourization by P. ostreatus has been attributed to laccase (Faraco et al., 2009). We found that decolourization of acid blue 62 was independent of Mn2+ amendments, and both wild-type P. ostreatus and the transformants decolourized the dye similarly, with or without Mn2+ (Supporting Information; Fig. S1). This is in agreement with our conclusion that the inhibition of OII decolourization by the pTMS1 transformants is indeed due to specific silencing of Mn2+-dependent peroxidases genes.
To further verify the link between the transformation procedure and the phenotypic effects observed, we confirmed that the transformants analysed did, in fact, harbour the RNAi insert. Our analysis was based on the fact that carboxin resistance (CbxR) is conferred by a point-mutated P. ostreatus sdi1 gene, whose source is an allelic sdi1 fragment from P. ostreatus strain #261-22 (Irie et al., 1998a; Honda et al., 2000; GenBank AB009845). Analysis of the PC9 sdi1 allele (This study; GenBank FJ598647) revealed a single-nucleotide alteration, which results in the presence of a unique SphI recognition site at position 724 bp in the PC9 sdi1 allele, which is not present in the CbxR-resistance-conferring allele of sdi1 (Fig. 3, marked by ‘*’). Accordingly, confirmation of the integrative nature of the constructs used was performed by PCR-based amplification of sdi gene fragment(s) followed by SphI-digestion, in order to confirm the presence of the CbxR gene; and a segment corresponding to the sdi1 promoter linked to the mnp3 cDNA antisense, in order to confirm the presence of the hpRNA expressing mnp3 silencing construct (Fig. 3). On the basis of the resulting amplicons and SphI restriction profile (Fig. 5), the presence of the hpRNA expressing mnp3 silencing construct in the relevant transformants was confirmed (For further explanation see Supporting Information).
MnP gene family expression is altered in strains harbouring mnp3 RNAi constructs
In order to determine whether alterations in MnP gene family transcript levels accompany the changes observed in OII decolourization we used relative real-time PCR quantification analysis. The selected strains (TS1, TS9, TS24, TS30 and the TC3 control) were grown for 7 days in liquid culture, in the presence of carboxin, in either a medium amended with 27 µM Mn2+ (+Mn treatment) or a non-amended medium (−Mn treatment). The results presented in Fig. 6 show the relative expression of the nine different members of the MnP gene family in the five selected strains. The overall abundance of the endogenous control gene (β-tubulin) transcript was found to be similar in all the strains in all the tested treatments. As TC3 exhibited similar gene expression levels to the PC9 wild-type strain, we used the former, in the −Mn treatment, as our calibrators for relative quantifications. The results obtained (Fig. 6) indicate that the predominantly expressed MnP gene family members expression in the pTMS1 transformants varied in intensity in response to Mn2+ amendment, similarly to the calibrator (TC3) and PC9 (Fig. 2), but at different levels. Specifically, relatively to the calibrator, mnp3 and mnp9 gene expression, though not substantially affected in the −Mn treatment (less than twofold difference), were strongly inhibited in the +Mn treatment which exhibited 24- to 79-fold higher levels in the calibrator. Therefore, mnp3 and mnp9 were not even remotely upregulated at levels comparable to the control in response to Mn2+ amendment. mnp4 gene expression was not substantially affected in the −Mn treatment (less than 2.2-fold difference); however, in the +Mn treatment it was 8- to 24-fold higher in the calibrator; mnp8 gene expression was not substantially affected in the −Mn treatment (less than 2 fold difference), but in the +Mn treatment it was 4- to 9-fold higher in the calibrator. Thus, in contrast to the striking results obtained in the case of mnp3, 4 and 9, the transcript abundance of mnp1, 2, 5, 6 and 7 was not substantially affected (less than fourfold difference), generally showing the same trend as the calibrator both in the −Mn and +Mn treatments.
In an attempt to explain the ‘off-target’ effects of the mnp3 silencing construct on the other MnP gene family members, a Smith-Waterman local alignment of the silencing construct against the various mnps transcripts was performed. The sequences were found to have potential cross-hybridization RNA fragments, according to the basic rules set by Yamada and Morishita (2005), thus providing a probable explanation for the silencing of other gene family members.
These results verify that transformation of P. ostreatus with an mnp3 cDNA hpRNA expressing construct (pTMS1) induces strong silencing of the two Mn2+-dependent peroxidase genes, mnp3 and mnp9, and increases the reduction in mnp4 transcript in the +Mn treatment, without substantially affecting the expression of the other MnP gene family members. Moreover, the silencing level of the mnp3 gene (Fig. 6) strongly correlates with the level of inhibition in OII decolourization capacity of these strains both on solid and in liquid media (Fig. 4A and 4C respectively). The combined results from analysing the fungal response to Mn2+ amendments, as expressed by OII decolourization capacity, along with the changes in MnP gene family expression levels, support our conclusion linking the phenotypic effects observed with MnP gene silencing.
Previous studies demonstrated that Mn2+ amendment greatly enhances the ligninolytic degradative capabilities of P. ostreatus. This was presumably due to an increase in the activity and transcript levels of Mn2+-dependent peroxidase (Camarero et al., 1996; Kerem and Hadar, 1997; Cohen et al., 2001; 2002b). Similarly, MnP expression in Phanerochaete chrysosporium was shown to be positively regulated by Mn2+ (Gettemy et al., 1998; Kersten and Cullen, 2007). To evaluate the molecular genetic basis for the enhanced decolourization of OII in the presence of Mn2+, we identified the components of the P. ostreatus MnP gene family and followed their transcription profile.
Prior to this study, only four known members of this gene family (mnp1–4) have been studied (see Table 1). Annotation of the recently sequenced P. ostreatus genome revealed that this gene family is comprised of at least nine members, five of which (mnp5–9) have not been previously studied (Table 1; http://genome.jgi-psf.org/PleosPC15-1). Due to the fact that these genes were found to be closely related, the use of specific primers, designed on the basis of non-conserved regions, was required to study their transcription profile. Our results confirm that all nine genes (mnp1–9) are transcribed, and that Mn2+ differentially regulates the abundance of the MnPs (mnp3 and mnp9) and VP (mnp4) transcripts, all of which have a substantially higher level of basal transcript abundance than the other family members. In Mn2+-containing cultures the transcript levels of mnp3 and mnp9 were drastically increased relative to the non-amended medium, and a concomitant marked reduction in mnp4 transcript abundance was measured (Fig. 2). mnp3 encodes a Mn2+-dependent peroxidase, and the deduced amino acid sequence of mnp9 indicates that this gene also encodes a Mn2+-dependent peroxidase, whereas mnp4 is known to encode a VP (Asada et al., 1995; Ruiz-Dueñas et al., 1999; Giardina et al., 2000; Irie et al., 2000; Cohen et al., 2001; http://genome.jgi-psf.org/PleosPC15-1).
Multiple LME isoenzymes, encoded by various, and apparently redundant structurally related genes, is common among wood-associated fungi. For example, P. chrysosporium, Ceriporiopsis subvermispora and Phlebia chrysocreas have at least 5, 4 and 7 different genes encoding MnPs respectively (Kersten and Cullen, 2007; Gutiérrez et al., 2008; Morgenstern et al., 2008), while P. ostreatus, Coprinopsis cinerea and Laccaria bicolor have at least 7, 17 and 11 different genes encoding laccases respectively (Kilaru et al., 2006; Courty et al., 2009; Pezzella et al., 2009). Studies of the catalytic properties of various LME isoenzymes reflected distinct differences in both the culture conditions and substrate specificity associated with their transcription and kinetic constants. For example, seven isoforms of MnPs were isolated from C. subvermispora cultures in salt medium, whereas four isoenzymes were fractioned in extracts derived from wood chips. The requirement for Mn2+ by each of these MnPs varied on the basis of the nature of the aromatic substrate (vanillylacetone, o-dianisidine, p-dianisidine, guaiacol) added to the reaction mixture (Urzúa et al., 1995). Phanerochaete chrysosporium MnP gene expression is regulated by nitrogen levels, Mn2+ as well as by heat shock, agitation and other factors. While mnp1–3 are expressed under various culture conditions, mnp4 and mnp5 seem to be actively transcribed only when the fungus is grown on wood-containing soil samples, and wood pulp respectively (Gettemy et al., 1998; Kersten and Cullen, 2007).
The fact the P. ostreatus has a large number of distinct genes encoding transcripts of various isoenzymes belonging to the MnP gene family points to the potential redundancy and/or diversity of these enzymes, and may imply that complex and versatile strategies are employed by this fungus for the degradation of aromatic and recalcitrant compounds such as amorphic lignin and azo dyes. Moreover, these enzymes were shown to be differentially regulated by Mn2+, and it is also probable that they may prove to have different substrate specificity and catalytic properties.
RNAi has been proven effective for gene expression downregulation in most eukaryotes, including vertebrates, plants, worms, protists and fungi (Nakayashiki and Nguyen, 2008). This method has proven especially useful when homologous recombination, necessary for gene targeting using knockout techniques, occurs at low frequency (Matityahu et al., 2008; Nakayashiki and Nguyen, 2008; Kemppainen et al., 2009), such as in the predominant random ectopic integration of transformed DNA into Pleurotus (Honda et al., 2000; Irie et al., 2001a). Furthermore, unlike conventional gene disruption methods, RNAi confers only partial reduction (knock-down) but not complete loss (knockout) of gene expression. The incomplete gene suppression frequently results in phenotypic variations that may provide more detailed information concerning the function of the silenced gene (Nakayashiki and Nguyen, 2008). Within the framework of the P. ostreatus genome deciphering project, our preliminary results identified genes encoding orthologues of key components of the RNA-mediated gene silencing machinery (Argonaute, Dicer and RdRP) (Tomari and Zamore, 2005; Nakayashiki et al., 2006; Nakayashiki and Nguyen, 2008), supporting the feasibility of RNAi-mediated gene expression regulation in this fungus (expressed sequence tags have also been found for many of these genes) (http://genome.jgi-psf.org/PleosPC15-1). We therefore employed an RNAi-based approach for gene knock-down of Mn2+-dependent peroxidases, followed by evaluation of OII decolourization and MnP gene family transcription abundance levels in the corresponding strains, demonstrating that the expression of Mn2+-dependent peroxidases can be directly associated with P. ostreatus ligninolytic degradative capabilities. Since the data regarding the existence of mnp5–9 were not available until the final stages of this project, our strategy focused on targeting mnp3, which was the only gene known to encode for Mn2+-dependent peroxidase in P. ostreatus at the time.
To date, plasmid constructs expressing hpRNA or intron containing hpRNA are the most prevalent and reliable platforms to induce RNAi in fungi. These types of vectors have been successfully used to demonstrate RNAi using model genes and to explore gene function in a wide range of fungal species. In filamentous fungi specifically, this strategy has been shown to induce the most efficient, systemic, trans-acting and stable silencing (Nakayashiki and Nguyen, 2008). With this in mind, pTMS1, the vector designed to induce RNAi-based gene silencing of mnp3, was constructed according to the basic principle of expressing an hpRNA.
Another advantage of RNAi, is that, in contrast to the gene knockout approach, it facilitates a trans-acting effect of reducing gene expression in the transformed strains without having the need for mating, production of fruiting bodies and spores, for isolation of transformant monokaryons, in order to obtain observable silencing effects, which may have been necessary using knockout techniques (Nakayashiki and Nguyen, 2008). In this respect, silencing via RNAi can result in the simultaneous suppression of expression of an entire gene family, thus avoiding compensatory effects of redundant gene family members. Simultaneous interference with homologous family members by using dsRNA has been demonstrated in trypanosomes and Drosophila. In fact, in most protozoan parasites and fungi, this approach is significantly more efficient than constructing multiple knockout strains (Nakayashiki and Nguyen, 2008).
These characteristic RNAi effects could be clearly seen in strains harbouring mnp3 RNAi constructs. Integration of the mnp3-silencing cassette caused a substantial reduction in both the expression levels of mnp3, 9 and 4 and in OII decolourization capacity (Figs 4 and 6). Nonetheless, and even though sdi1 expression signals are considered to be constitutive (Irie et al., 1998b; 2001a,b; Honda et al., 2000; Tsukihara et al., 2006), both effects were not total, as the silenced strains still exhibited decolourization, pronounced in the late stages of incubation. This phenomenon is most likely due to the incomplete suppression of the expression of the MnP genes, which allowed gradual accumulation of MnPs to sufficient levels for decolourization. Indeed, the silencing level of the mnp3 gene positively correlated with the level of inhibition of OII decolourization capacity of these strains (Figs 4 and 6). Similar ‘off-target effects’ were described by Matityahu et al. 2008 when attempting to silence the P. chrysosporium MnSOD1 gene. There, the MnSOD1 RNAi construct also indirectly led to the partial silencing of MnSOD2 and 3, two other genes that belong to the MnSOD gene family.
In this study, the outcome of our attempt to silence mnp3 clearly illustrates the potential multigene silencing capacity of RNAi (Fig. 6). These effects can be attributed to the nature of the hpRNA construct we employed (Fig. 3), which contained numerous regions that are highly conserved among various members of the MnP gene family. Due to this sequence similarity, and as may be expected, the homologous nontargeted members of this multiple gene family were simultaneously silenced, along with the ‘true target’–mnp3. This phenomenon may have been to our advantage, since it resulted in silencing of mnp9, in addition to mnp3 (Fig. 6). Whether or not mnp9 and mnp3 truly have redundant or compensatory functions has yet to be determined.
Taken together, we established the first direct association between Mn2+-dependent peroxidases (mnp3 and mnp9) level of gene expression and azo dye decolourization capacity in P. ostreatus. This conclusion was obtained here on the basis of an RNAi reverse genetics approach, now providing a proven platform for further dissection of gene family function in P. ostreatus.
Fungal and bacterial strains and growth conditions
Pleurotus ostreatus monokaryon strain PC9 (Spanish Type Culture Collection accession number CECT20311), which is a protoclone derived by dedikaryotization of the dikaryon commercial strain N001 (Larraya et al., 1999), was used throughout this study. Fungal strains were grown and maintained in YMG medium [1% w/v glucose, 1% w/v malt extract (Difco), 0.4% w/v yeast extract (Difco)] (Irie et al., 2001b) or in GP medium [2% w/v glucose, 0.5% w/v peptone (Difco), 0.2% w/v yeast extract (Difco), 0.1% w/v K2HPO4, 0.05% w/v MgSO4·7H2O] (Irie et al., 2001a); Mn2+ was added as MnSO4 as specified. Solid culture was performed in 9 cm diameter Petri dishes containing 15 ml media solidified with 1.5% w/v agar. Liquid cultures were maintained in stationary 100 ml Erlenmeyer flasks containing 10 ml media. Cultures were incubated at 28°C in the dark. The inoculum for all growth conditions was one disk (5 mm diameter) of mycelium, from the edge of a freshly grown colony in solid culture, positioned at the centre of the Petri dish or the flask. The azo dye Orange II [4-(2-hydroxy-1-naphthylazo)benzenesulfonic acid sodium salt] and the fungicide carboxin (Sigma-Aldrich) were added to a final concentration of 100 mg l−1 and 8.5 µM (LD50 = 0.7 µM), respectively, as specified. Mycelial growth in solid culture was evaluated by measuring colony area, and in liquid culture biomass production was measured as dry weight (oven dried to constant weight at 65°C). Escherichia coli JM109 cells (Invitrogen) were used for standard cloning procedures according to the manufacturer's protocol.
Analysis of Orange II decolourization
In solid culture, OII decolourization capacity was estimated according to the visually decolourized area, as measured from the centre of the inoculation point. In liquid culture, 200 µl of the media was centrifuged (4720 g, 10 min, room temperature) and mixed with 800 µl phosphate buffer (0.1 M, pH 7.0). OII dye concentration in the media was quantified according to the absorption reading of the solution at λmax 483 nm, using a BioMate 3 spectrophotometer (Thermo Spectronic), according to a standard curve. Non-inoculated medium amended with OII was used as a control.
Nucleic acid manipulation and analysis
Molecular manipulations were carried out on the basis of standard protocols (Sambrook et al., 1989). Commercial kits were used according to the manufacturer's instructions. Genomic DNA was isolated using the ZR Fungal/Bacterial DNA Kit (Zymo Research). RNA was extracted from 7-day-old cultures, grown in liquid GP medium, either non-amended or amended with 27 µM of Mn2+, using the RNeasy Plant Mini Kit (Qiagen) along with the manufacturer's RLC lysis buffer. cDNA synthesis was performed using the SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen). Plasmid DNA purification was performed using the QIAprep Spin Miniprep Kit (Qiagen). PCR was performed on an Eppendorf Mastercycler Gradient Thermocycler (Eppendorf), using BIO-X-ACT Short Mix (BIOLINE), with the primers detailed in Table 1. DNA endonuclease restriction and ligation were performed using restriction enzymes and T4 DNA Ligase from Fermentas. Isolation and purification of DNA fragments from agarose gel or PCR amplification was performed using the kit Wizard SV Gel and PCR Clean-Up System (Promega). Cloning to plasmids was performed using the kit pGEM-T Vector System II (Promega). Full sequencing was performed for all the DNA fragments, plasmids and real-time PCR amplicons, in the Center for Genomic Technologies at the Hebrew University of Jerusalem. Real-time PCR relative quantification analysis by the ΔΔCT method (Livak and Schmittgen, 2001) was performed on an ABI Prism 7000 Sequence Detection System and software (Applied Biosystems), using SYBR Green PCR Master Mix (Applied Biosystems), with the primers detailed in Table 1 and an annealing temperature of 63°C, according to the manufacturer's default operating procedures.
Construction of pTMS1: an mnp3 silencing vector
The construct for RNAi silencing of mnp3 was designed with inverted 650 bp repeats, corresponding to the first 650 bp of mnp3 cDNA (This study; GenBank FJ594281), separated by a 100 bp loop (linker region) corresponding to mnp3 cDNA 100–1 bp, under the control of sdi1 expression signals, which was ligated into the pTM1 plasmid (Honda et al., 2000), harbouring the carboxin resistance gene, CbxR, under the control of sdi1 expression signals; the resulting plasmid was designated pTMS1 (Fig. 3; see also Results). All the primers used in the construction procedure are listed in Table 1 and indicated in Fig. 3; these primers were designed with unique restriction sites at their ends, to facilitate directional cloning of the corresponding fragments. Specifically, mnp3 cDNA was amplified using primers MnP3C1 and MnP3C2. sdi1 promoter was amplified from pTM1 using primers sdi1PF and Sdi1PR, which adapts the restriction sites SacII and NheI respectively. The antisense (+linker region) segment, corresponding to mnp3 cDNA 750–1 bp, was amplified from mnp3 cDNA using primers MnP3CAS750 and MnP3CAS1, which adapts the restriction sites NheII and AscI respectively. The sense segment, corresponding to mnp3 cDNA 101–750 bp, was amplified from mnp3 cDNA using primers MnP3CS101 and MnP3CS750, which adapts the restriction sites AscI and AgeI respectively. sdi1 terminator was amplified from pTM1 using primers sdi1TF and Sdi1TR, which adapts the restriction sites AgeI and SphI respectively. Each of the amplified segments was separately cloned into a pGEM-T vector (Promega). The promoter, antisense (+linker region), sense and terminator segments were then extracted via endonuclease digestion with the corresponding restriction enzymes specified above; and plasmid pTM1 was digested with the restriction enzymes SphI and SacII. All the digested segments were isolated and purified from agarose gel, and then mixed together in a ligation reaction. The resulting ligation mixture was transformed into E. coli using the pGEM-T Vector System II (Promega); and plasmids were successively purified and screened via digestion to locate the desired plasmid. This plasmid was designated pTMS1 (Fig. 3) and it was entirely sequenced to confirm that all the segments were cloned correctly (both in location and orientation).
Transformation was performed based on the PEG/CaCl2 protocol previously adapted for P. ostreatus (Honda et al., 2000; Irie et al., 2001b). The fungicide carboxin was used as a selective marker, and conferring resistance was achieved via introduction of the carboxin-resistance marker gene, CbxR, which was constructed by introducing a base substitution in the sdi1 gene (allele of strain #261-22) coding sequence (Honda et al., 2000). Competent protoplasts were produced by digestion of P. ostreatus mycelium, from liquid culture in YMG medium, with lytic enzymes. The lytic enzymes solution was comprised of 2% w/v Lysing enzymes from Trichoderma harzianum (Sigma-Aldrich) and 0.2% w/v Chitinase from Trichoderma viride (Sigma-Aldrich), in 0.5% w/v sucrose as an osmotic stabilizer. The protoplasts were washed in STC solution (18.2% w/v sorbitol, 50 mM CaCl2, 50 mM Tris-HCl pH 8.0, 0.5 M sucrose), and adjusted to a final concentration of 5 × 107 protoplasts ml−1. Then 200 µl of protoplasts were mixed with 10 µl of plasmid DNA (200 ng µl−1), 30 µl of single-strand λ phage carrier DNA (500 µg ml−1, after denaturation at 95°C for 5 min and immediate transfer to ice), and 15 µl of heparin (5 mg dissolved in 1 ml of STC solution). After 40 min of incubation on ice, 1 ml of PTC solution (40% w/v PEG#4000, 50 mM CaCl2, 50 mM Tris-HCl pH 8.0, 0.5 M sucrose) was added, and the mixture was incubated for 20 min at room temperature. The mixture was then plated on selective YMG solid regeneration medium, containing 0.5 M sucrose and carboxin at a final concentration of 8.5 µM. Transformants were isolated after 10 days of incubation at 28°C. The transformants stability was verified by three successive transfers (inoculated from the edge of a 7-day-old colony) to solid medium without carboxin, and then returning and maintaining them in selective solid culture amended with carboxin.
Screening and evaluation of P. ostreatus Mn2+-dependent peroxidase silenced transformants
Thirty pTMS1 transformants were isolated, alongside one randomly chosen control strain (TC3) transformed with pTM1. These transformants were analysed for both phenotypic and genotypic properties. Phenotypic analysis was performed in three stages: (i) selection of strains growing on solid medium under carboxin selection pressure; (ii) evaluation of their OII decolourization capacity on solid medium (which allowed for a large scale screen), followed by selection of strains exhibiting similar growth but high inhibition of their OII decolourization capability compared with the wild-type (Fig. 4A and 4B); and (iii) quantitative time-course assay of OII decolourization and biomass production by the selected strains in liquid culture (Fig. 4C). Genotypic analysis was performed at two levels: (i) confirmation of the constructs integration into the selected strains genome, by isolation of genomic DNA which served for specific PCR detection of the CbxR (sdi1) gene, using primers R1 and R3, followed by SphI-digestion, in order to confirm the presence of the CbxR gene; and a segment corresponding to the sdi1 promoter linked to the mnp3 cDNA antisense, using primers pTMS1F3628 and pTMS1R4191, to confirm the presence of the hpRNA mnp3 silencing construct (Table 1; Figs 3 and 5; see Results and Supporting Information for more details); (ii) evaluation of MnP gene family expression levels by extraction of RNA from liquid cultures of the selected strains, either non-amended or amended with 27 µM Mn2+, followed by relative real-time PCR quantification analysis, in order to confirm silencing effects of Mn2+-dependent peroxidases gene expression (Table 1 and Fig. 6).
We are deeply grateful to Professor Takashi Watanabe and Professor Yoichi Honda, Research Institute for Sustainable Humanosphere, Kyoto University, Japan, for generously providing plasmid pTM1. We also thank the Joint Genome Institute (US Department of Energy) and the Pleurotus Genome Consortium for access to the P. ostreatus genome sequence prior to publication. This work was supported in part by the Joint Israeli-Lower Saxony research projects financed by the Lower Saxony Ministry of Science and Culture.