Cloning of Phanerochaete chrysosporium leu2 by Complementation of Bacterial Auxotrophs and Transformation of Fungal Auxotrophs

Abstract

A Phanerochaete chrysosporium cDNA library was constructed in an expression vector that allows expression in both Escherichia coli and Saccharomyces cerevisiae. This expression vector, λYES, contains the lacZ promoter for expression in E. coli and the GAL1 promoter for expression in yeast. A number of genes were cloned by complementation of bacterial amino acid auxotrophs. The cDNA encoding the β-isopropylmalate dehydrogenase from P. chrysosporium was characterized further. The genomic clone (gleu2) was subsequently isolated and was used successfully as a selectable marker to transform P. chrysosporium auxotrophs for LEU2. Protoplasts for transformation were prepared with readily obtained conidiospores rather than with basidiospores, which were used in previous P. chrysosporium transformation procedures. The method described here allows other genes to be isolated from P. chrysosporium for use as selectable markers.

Lignin is a random polymer of phenylpropanoid units that constitutes up to 30% of woody biomass. The structural heterogeneity of lignin makes it resistant to most forms of microbial attack. The predominant degraders of lignin are basidiomycete fungi, and the best-characterized of these is Phanerochaete chrysosporium. To depolymerize the complex polymer, P. chrysosporium secretes members of two isozyme families, the lignin peroxidases and the manganese peroxidases, which catalyze oxidative cleavage of lignin. Since the discovery of these enzymes more than a decade ago, research into lignin degradation has intensified due to potential applications in lignocellulosic utilization and due to interest in this fundamental process.

The isolation of cDNAs and genomic clones encoding the peroxidases has enabled researchers to use molecular genetic techniques to study lignin degradation. A critical component necessary for the use of these techniques is the development of a transformation system. In 1985, genetic transformation was reported for only 10 fungi (15), including five ascomycetes, one phycomycete, and four ascogenous yeast species but no basidiomycetes. Since then, transformation systems have been developed for a few basidiomycetes. However, the progress has been slow and has required prodigious effort. There are many ways to develop a transformation system, but a common strategy is to isolate a wild-type gene corresponding to a gene for which the host is deficient. This strategy is slow due to the low number of genes isolated from lignin-degrading basidiomycetes. Another factor that has contributed to this slow development is the difficulty in obtaining basidiospores or conidiospores; this in turn creates difficulty in obtaining auxotrophs as hosts for transformation and in obtaining protoplasts (6). Other factors include the presence of nonspecific nucleases and possible methylation of heterologous DNA sequences (17).

In 1989, Alic et al. (4) reported the first transformation of P. chrysosporium. These researchers used a gene from a related fungus, ade2 from Schizophyllum commune, to transform an ADE2 mutant of P. chrysosporium. Members of the same group later used the S. commune ade5 (3) and ura3 (1) genes, the ura5 gene from the ascomycete Podospora anserina (1), and the ade1 gene from P. chrysosporium (5) to transform the corresponding P. chrysosporium auxotrophs.

Other than the ade1 gene (5), homologous transformation of P. chrysosporium has been limited. Only four P. chrysosporium genes that can potentially act as selectable markers have been cloned to date. These include the glyceraldehyde-3-phosphate dehydrogenase (gdp) gene, ade1, ura3, and trpC (1, 5, 13, 20). These genes have been isolated by probing genomic libraries with complementing genes from related fungi or, in the case of trpC, by complementation of Escherichia coli.

The present paper described the construction of a P. chrysosporium cDNA library in a novel expression vector, λYES (8). This vector allowed us to complement both E. coli and yeast auxotrophs and to clone a number of cDNAs involved in biosynthetic pathways, including the leu2 gene encoding β-isopropylmalate dehydrogenase. The cDNA was then used to isolate the genomic clone which was used to transform P. chrysosporium LEU2 auxotrophs. This cDNA library should be useful in cloning other biosynthetic genes for use in transformation of P. chrysosporium and related wood-degrading fungi.

MATERIALS AND METHODS

Strains and phage.

P. chrysosporium BKM-F-1767 (= ATCC 24725) was maintained on malt extract slants. Two P. chrysosporium auxotrophs for leu2, Leu1 and Leu5, were obtained from Michael H. Gold, Oregon Graduate Institute of Science and Technology. E. coli AB1157 (= ATCC 29055) was obtained from Ronald Porter, The Pennsylvania State University. Other E. coli strains were obtained from the E. coli Genetic Stock Center of Yale University (New Haven, Conn.). All E. coli auxotrophs were maintained on Luria broth complete medium plates. λYES and λKC were obtained from Andrew Buchman, The Pennsylvania State University.

Materials.

A Packagene lambda DNA packaging system was obtained from Promega (Madison, Wis.). pBluescript II and λ-Dash were obtained from Stratagene (La Jolla, Calif.). Universal and reverse sequencing primers and a Sequenase sequencing kit were obtained from U.S. Biochemicals (Cleveland, Ohio). An RNA ladder and restriction enzymes were obtained from Gibco BRL (Gaithersburg, Md.). A Puregene kit was obtained from Gentra Systems (Minneapolis, Minn.). Kanamycin and ampicillin were obtained from Sigma Chemical Co. (St. Louis, Mo.). Novozyme was obtained from Calbiochem-Novabiochem (La Jolla, Calif.). Cellulase was obtained from Solvay Enzymes (Elkhart, Ind.). All other reagents were reagent grade.

Construction of λYES · cDNA library.

Primary cultures of P. chrysosporium BKM-F-1767 were grown in minimal medium to induce amino acid biosynthesis. This medium consisted of basal III/glucose medium (25) containing 10 times more ammonium tartrate than it normally does. Poly(A) RNA was isolated from 2-day-old cultures (26) and was used to construct a cDNA library in λYES as described by Elledge et al. (8). The λYES cDNA library was packaged by using a Packagene lambda DNA packaging kit according to the Promega protocol. The primary library consisted of more than 10⁶ PFU with an insertion frequency of 40%.

Complementation of amino acid auxotrophs.

Auxotrophic strains of E. coli were first infected with λKC, which conferred kanamycin resistance and the cre recombinase, by using the procedure of Elledge et al. (8). Lysogens were selected based on kanamycin resistance, grown in liquid culture, and infected with the λYES cDNA library at a concentration of 6 × 10⁶ phage/10⁹ bacterial cells. The cre recombinase catalyzes site-specific recombination at directly repeated lox sites in the linear phage to give a circular λYES · cDNA plasmid. Cells were plated onto minimal M9 medium that contained ampicillin and was selective for the amino acid of interest. Complement colonies were grown in a liquid culture supplemented with ampicillin, and plasmid DNA was isolated by the rapid alkaline method.

Analyses of leu2 cDNA and deduced amino acid sequence.

Plasmid λYES cDNA that complemented E. coli CV514, a leuB auxotroph, was digested with EcoRI. The cDNA insert was ligated into pBluescript II. Dideoxy sequencing was carried out for both strands (19) by using commercial primers and primers synthesized with a model Oligo1000 DNA synthesizer (Beckman Instruments, Inc., Fullerton, Calif.).

The isolated cDNA insert was labeled with ³²P by random priming by using an Amersham Rediprime kit and was used to probe a Northern blot containing 3 μg of poly(A) RNA as previously described (26).

The nucleotide sequence homologies of the putative leu2 cDNA and other species were determined by using the BLASTN search method and the GenBank database (National Institutes of Health).

Amino acid sequences for bacterial leuB and fungal leu2 gene products of different strains were downloaded from GenBank (National Institutes of Health). These sequences were aligned by using the Clustal W multiple-alignment program (23). The amino acid sequence deduced from our putative leu2 cDNA was then aligned with the other amino acid sequences by using the same program.

Cloning of genomic leu2.

The genomic library of P. chrysosporium in λ-Dash (14) was screened with ³²P-labeled leu2 cDNA. Phage DNA was isolated after three rounds of screening. Southern blot analyses indicated that none of the 12 clones screened contained the entire gene. The library was constructed from Sau3A-digested genomic DNA. It appeared that a hot spot for Sau3A was present in the coding region. To ligate the two halves together, we engineered a unique HindIII site into each half with a single-base (silent) mutation of C to T at nucleotide 751 by PCR (Fig. 1). This altered the sequence ₇₄₆AAGCTC₇₅₁ to ₇₄₆AAGCTT₇₅₁; the codon CTT still encoded a leucine. The PCR strategy used is described in the legend to Fig. 1.

FIG. 1.

Genomic leu2 construct. A 5′ fragment and a 3′ fragment of genomic leu2 with overlapping sequences were cloned individually into pBluescript II (solid line). A HindIII site was engineered into the overlapping segments by PCR amplification. A PCR primer (5′-GTCGCGAACCCCAAGAAGCTTAACGGCGTGAT-3′) with a 1 base mismatch at nucleotide 751 (C to T) and its complement were used to introduce the HindIII site (solid rectangles). The other PCR primers (solid circles) were complementary to Bluescript sequences. The PCR products were ligated into the EcoRV site of pBluescript II and then cut with HindIII and other restriction enzymes as described above to yield the entire gene in pBluescript II.

Transformation of P. chrysosporium LEU2 auxotrophs.

Protoplasts were prepared from Leu1 and Leu5 conidiospores. Leu1 and Leu5 were grown on malt agar slants (24) at 37°C to induce conidiospore formation. After 3 to 4 days of growth, conidiospores were scraped into sterile water and filtered through sterile glass wool. The filtered conidiospores were added to 40 ml of modified Vogel’s medium (4) at a concentration of 5 × 10⁵ spores/ml. The spore suspension was incubated at 37°C with shaking at 150 rpm until the spores were swollen and spherical, approximately 4 h. The swollen spores were isolated by centrifugation at 800 × g for 10 min and resuspended in 1 ml of MgOSM (4) containing 10 mg of Novozyme per ml and 10 mg of cellulase per ml. The suspension was then incubated at 37°C with gentle shaking for 3 to 5 h until protoplasts were formed. Protoplasts were isolated by centrifugation at 800 × g for 10 min and washed twice with SorbOSM (4). After the final wash, the protoplasts were gently resuspended in SorbOSM containing 40 mM CaCl₂ and stored on ice overnight to establish competency.

A total of 2 × 10⁶ protoplasts were transformed with 1 to 10 μg of genomic leu2 DNA in pBluescript II (pBS gleu2) as follows. DNA was diluted to the appropriate concentration with 10 mM Tris-HCl–1 mM EDTA–40 mM CaCl₂ (pH 8). Competent protoplasts (100 μl of a suspension containing 2 × 10⁷ protoplasts per ml) in SorbOSM-CaCl₂ were incubated with 60 μl of the diluted DNA on ice for 10 min. An equal volume of 44% polyethylene glycol was added below the protoplast-DNA mixture, and the solution was incubated on ice for an additional 10 min. After mixing, the transformation mixture was kept on ice and diluted 10-fold into minimal medium (18) containing 0.5 M MgSO₄. Dilutions of the transformation mixture were plated along with minimal medium top agar (4) onto minimal media protoplast regeneration plates (4) and incubated at 37°C.

Individual transformants were visualized after 5 or 6 days, isolated by restreaking on sorbose plates (10), and maintained on malt agar slants (24). Genomic DNA was isolated by using a Puregene kit according to the manufacturer’s protocol. DNA was digested with HindIII, and Southern blotting was carried out by using standard methods. Hybridization was carried out with leu2 cDNA labeled with ³²P by random priming.

Nucleotide sequence accession number.

The sequence of P. chrysosporium leu2 cDNA has been deposited in the GenBank database under accession no. AF050668.

RESULTS

The λYES vector contains an ampicillin selectable marker and the lacZ promoter for expression in E. coli (8). It also contains the ura3 selectable marker and GAL1 promoter for expression in Saccharomyces cerevisiae. Due to the ease of transformation and the higher transformation frequency with E. coli than with S. cerevisiae, our initial attempts to clone by complementation were with E. coli. To test the quality of the library, the λYES · cDNA library was used to infect E. coli AB1157, an auxotroph for arginine, proline, leucine, threonine, and histidine. Plating onto selective media (containing all amino acids except the one of interest) resulted in isolation of transformants for arginine, histidine, leucine, and threonine, which reflected favorably on the size and fidelity of the library. Isolation of the plasmid DNA from the respective transformants and retransformation indicated that all of the transformants from the second round were prototrophs (data not shown).

Leucine complementation was further examined with E. coli CV514, which contains a lesion in the leuB gene encoding β-isopropylmalate dehydrogenase (22). Strain CV514 was infected with the λYES cDNA library and plated onto minimal medium plates containing ampicillin and lacking leucine. Plasmid DNAs were isolated from five complement colonies and used to retransform competent CV514 cells. Four of the five plasmids were capable of complementing the leuB auxotroph in this second round of transformation. One of these plasmids was characterized further. It contained a 1.3-kb insert that was used to probe a Northern blot of P. chrysosporium poly(A) RNAs isolated from primary cultures grown in minimal medium. A band was visualized at approximately 1.3 kb (Fig. 2).

FIG. 2.

Northern blot analysis of P. chrysosporium poly(A) RNA probed with leu2 cDNA. Northern blot analysis was carried out as described in Materials and Methods. The RNA size markers were determined by electrophoresing an RNA ladder on the same gel and staining with ethidium bromide.

The cDNA insert was subcloned into plasmid Bluescript II and used for double-stranded dideoxy sequencing. The complete cDNA sequence is shown in Fig. 3. The G+C content of the coding region was 63%. In the 3′ noncoding region, the G+C content was approximately 40%. These values are consistent with the G+C contents of other P. chrysosporium genes (9). A BLASTN search of the National Institute of Health GenBank database revealed high degrees of nucleotide sequence homology with β-isopropylmalate dehydrogenase genes from a variety of species. A total of 10 of the top 12 matches encoded this enzyme. The isolated cDNA encoded a protein with an expected molecular mass of 38 kDa. The amino acid sequence showed a high degree of conservation with the amino acid sequences of β-isopropylmalate dehydrogenases from a variety of bacterial and fungal sources (Fig. 4).

FIG. 3.

Nucleotide and predicted amino acid sequences of leu2 cDNA from P. chrysosporium. The cDNA sequence has an open reading frame starting at nucleotide 26. The 3′ noncoding region starts at nucleotide 1166 and includes a poly(A) tail.

FIG. 4.

Alignment of P. chrysosporium leu2 amino acid sequence with β-isopropylmalate dehydrogenase sequences from bacterial and fungal species. The sequences were aligned by using the Clustal W multiple-alignment program (23). Conserved amino acids are indicated by asterisks. Amino acids that have strongly related characteristics, as determined with a Gonnet Pam250 matrix, are indicated with colons. Less closely related amino acids are indicated with dots. Amino acid sequences were downloaded from GenBank (National Institutes of Health). Abbreviations: E. coli, Escherichia coli; Salmonella, Salmonella typhimurium; Thiobac, Thiobacillus ferrooxidans; Spirulina, Spirulina platensis; H. polym, Hansenula polymorpha; S. cer, Saccharomyces cerevisiae; Y. ohm, Yamadazyma ohmeri; A. niger, Aspergillus niger; N. crassa, Neurospora crassa; Ph. chr, Phanerochaete chrysosporium

The complete leu2 gene was isolated and subcloned into pBluescript II (pBS gleu2). As described above, we were not able to clone out the entire gene in one fragment. Figure 1 shows the strategy used for ligating the two halves of the gene to yield the complete leu2 gene. This plasmid was used to transform protoplasts of P. chrysosporium LEU2 auxotrophs (Leu1 and Leu5). Both Leu1 and Leu5 lack β-isopropylmalate dehydrogenase activity (16). Protoplasts were prepared from Leu1 and Leu5 conidiospores. Counting of the conidiospores and resulting protoplasts with a hemocytometer revealed that approximately 40% of the conidiospores formed protoplasts; but only 2 to 13% of these protoplasts were viable. However, increasing the number of protoplasts did not increase the transformation frequency (data not shown). Transformation with linearized DNA also did not increase the transformation frequency compared with transformation with circular DNA. The transformation frequency varied with each experiment but was on the order of 2 to 10 transformants/μg of DNA for both Leu1 and Leu5. No colonies were observed with either Leu1 or Leu5 protoplasts that were not treated with pBS gleu2.

Southern blot analysis of the fungal transformants showed that the leu2 gene had been integrated into the genome of the P. chrysosporium LEU2 auxotrophs. Digestion with HindIII in nontransformed Leu1 and Leu5 yielded, as predicted, one band at approximately 17 kb (Fig. 5). Because a HindIII site was engineered into the transforming leu2 DNA, it was predicted that two bands would be obtained from HindIII-digested DNAs of transformants. As predicted, when DNAs from transformants (three transformants from both Leu1 and Leu5) were digested with HindIII, they yielded at least two bands. Some of the transformants produced bands with much greater intensity, which indicated that multiple integrations had occurred (Fig. 5). Ectopic integration of the leu2 gene would have resulted in the presence of the 17-kb wild-type copy upon digestion with HindIII in addition to the two bands of different sizes from the transformed copy. In contrast, homologous recombination would have yielded only two bands, with the sizes adding up to 17 kb. A closer examination of Leu5-A revealed the presence of two such bands. Figure 5 shows that bands at approximately 11 and 6 kb were observed after HindIII digestion, which is consistent with homologous recombination.

FIG. 5.

Southern blot of Leu1 and Leu5 auxotrophs and three transformants of each. Genomic DNA was digested with HindIII, and the Southern analysis was carried out as described in Materials and Methods. A single HindIII site was engineered into the transforming leu2 gene. This site is not present in the wild-type gene. Transformants of both Leu1 and Leu5 produced additional bands when they were probed with leu2 cDNA. Transformant Leu 5-A produced only a faint wild-type band, and this may indicate that homologous recombination occurred. The other transformants exhibited multiple ectopic integrations of the transforming DNA.

DISCUSSION

Using genes from other fungi to probe screen genomic libraries has been useful for isolating a relatively small number of biosynthetic genes from P. chrysosporium (1, 5, 13). An alternative method that does not require a probe gene from a closely related species is screening for genes by complementation. This method allows cloning of genes with similar functions from heterologous systems. However, in the absence of a high-frequency transformation system, cloning by complementation is problematic. Nevertheless, Schrank and coworkers were able to use complementation of an E. coli auxotroph to isolate the trpC gene from P. chrysosporium (20). These workers successfully used a genomic library of P. chrysosporium to clone the trpC gene (20), but their success is most likely the exception, not the rule. Using genomic libraries for cloning by complementation requires overcoming the difficulties of promoter recognition and interference from introns. These problems are alleviated by using a cDNA expression library with the appropriate host promoters. The λYES system of Elledge and coworkers (8) is especially well-suited for this purpose.

We constructed a P. chrysosporium cDNA expression library in the shuttle vector λYES (Yeast-E. coli Shuttle). This phage vector contains the ampicillin selectable marker and the lacZ promoter for expression in E. coli. It also contains the ura3 selectable marker and GAL1 promoter for expression in S. cerevisiae. The unique construction of λYES permits expression and selection in both E. coli and S. cerevisiae. Furthermore, subcloning from λYES is simplified by site-specific recombination at directly repeating lox sites to give circular DNA plasmids (8). The cDNA library which we constructed is a fairly representative library. In one experiment, four of the five amino acid deficiencies in E. coli AB1157, including the leucine deficiency, were complemented by our library. This indicates that other genes expressed under primary minimal medium growth conditions should also be present. We chose to focus on the leucine biosynthetic pathway to develop a transformation system for two reasons. First, there are only four unique steps in the pathway, making isolation of genes to match auxotrophs less complicated. Second, Molskness et al. (16) had previously isolated P. chrysosporium leucine amino acid auxotrophs for three of the four steps in the pathway.

Both heterologous (1, 3, 4) and homologous (1, 2, 5) transformations have been achieved with P. chrysosporium (see reference 9 for a review). In all of these cases, basidiospores were used to obtain protoplasts for transformation. Basidiospore formation requires closely controlled physiological conditions, including carbon source, nitrogen levels, and light (11). Alternatively, protoplasts can be prepared from mycelia treated with lytic enzymes (12). However, protoplasts formed in this manner are heterogeneous in size and number of nuclei (27). In this work, the readily obtained conidiospores were used to prepare protoplasts. Preparation of protoplasts from conidiospores has been described with other fungal species (6, 7, 21), but to our knowledge this is the first account of protoplasting and transformation with conidiospores from P. chrysosporium. Although only one-half of the conidiospores formed protoplasts, the protoplast regeneration frequency was similar to the 5% frequency seen with protoplasts from basidiospores (4). Use of conidiospores greatly simplifies and accelerates the transformation procedure.

Southern blot analysis in which the leu2 cDNA was used as a probe revealed that the intensity of the DNA bands was much higher in some transformants than in others. This is consistent with multiple integrations into the genome, which have also been observed by Alic et al. (3–5). With the exception of Leu5-A, when all of the transformants were digested with HindIII, they yielded the wild-type copy at 17 kb in addition to other copies corresponding to the transforming DNA. The restriction digest of Leu5-A is most consistent with homologous recombination. Homologous recombination of the gleu2 DNA into the wild-type chromosomal copy should result in two bands totaling 17 kb in size upon HindIII digestion. This is because we engineered a HindIII site into gleu2. Indeed, digestion of Leu5-A did result in the loss of the 17-kb band and the appearance of two bands totaling 17 kb in size. The very faint band observed at 17 kb was most likely due to partial digestion. Previous work by Alic et al. (2) indicated that homologous recombination is a rare event. These workers used a plasmid containing the P. chrysosporium ura3 gene interrupted by the S. commune ade2 gene to transform a P. chrysosporium ade2 auxotroph. Homologous recombination into the ura3 locus should result in ura3 knockout. Only a small percentage of ade2 prototrophs were uracil auxotrophs. In fact, transformants which had undergone homologous recombination were obtained only by positive selection with 5-fluoroorotic acid. Although one of six is not statistically significant, our results warrant further investigation and suggest that with our leu2 construct, homologous recombination occurs at a higher frequency than the frequency observed by Alic et al. (2).

The use of leu2 as a selectable marker and the transformation system described here should facilitate studies on the expression of P. chrysosporium genes, particularly the genes involved in lignin degradation. The transformants exhibited no loss of leucine prototrophy after repeated streaking, suggesting that they are mitotically stable. Our initial three attempts at fruiting the transformants to test for meiotic stability were unsuccessful. However, Alic et al. (4) showed that their transformants of P. chrysosporium were both meiotically and mitotically stable. The transformation frequency in the experiments described here was low, on the order of 10 transformants per μg of DNA. This is on the same order of magnitude that was observed by other researchers working with P. chrysosporium (5, 9). There is no apparent reason for this low transformation frequency. Our present efforts are aimed at increasing this frequency through more efficient protoplasting, the use of basidiospores, electroporation, and additional purification of the transforming DNA.