Structure of Gene Coding for the Fruit Body-Specific Hydrophobin Fbh1 of the Edible Basidiomycete Pleurotus ostreatus

Mycologia, 96(1), 2004, pp. 75–82. q 2004 by The Mycological Society of America, Lawrence, KS 66044-8897 Structure of gene coding for the fruit body-specific hydrophobin Fbh1 of the edible basidiomycete Pleurotus ostreatus Marı́a M. Peñas Joseba Aranguren Lucı́a Ramı́rez Antonio G. Pisabarro1 most highly expressed in fungi, this expression being developmentally regulated (de Groot et al 1996, Asgeirsdóttir et al 1998, 1999, Lugones et al 1998, Schuurs et al 1998, Segers et al 1999, Peñas et al 2002). These two characteristics make hydrophobin genes interesting models for the study of fungal development and for isolating strong tissue-specific gene promoters. The presence of many different hydrophobins in a given organism raises questions about their functions and the evolution of this gene family (Wösten 2001). It has been shown that Schizophyllum commune hydrophobin SC3 (Schuren and Wessels 1990) lowers the surface tension of the culture medium, letting the hyphae escape from the liquid phase and initiate aerial growth (Wösten and Wessels 1997). Fruit body-specific hydrophobins, on the other hand, are involved in the hyphal adhesion during basidiocarp formation, protection against desiccation and regulation of gas exchange in this structure (Wessels 2000). Two commercial varieties (florida and ostreatus, differing in size, fruiting temperature, and pileus structure and color) of the white-rot, edible mushroom Pleurotus ostreatus have been used to study fruit bodyspecific hydrophobins in this species. Peñas et al (1998) reported the characterization of Fbh1 (Fruit Body Hydrophobin 1) from var. florida, and Asgeirsdóttir et al described POH1 (P. ostreatus Hydrophobin 1) in var. ostreatus (Asgeirsdóttir et al 1998). Both proteins contain 113 amino acids and have two cysteine clusters spaced as expected for Class I hydrophobins (Kershaw and Talbot 1998). Expression of fbh1 and POH1 was limited to fruit bodies; their transcripts were absent in monokaryotic or dikaryotic mycelia. However, Peñas et al (2002) showed that another hydrophobin gene (vmh3), not allelic to fbh1, is expressed in both vegetative mycelium and fruit bodies. The occurrence of various hydrophobins in fruit bodies has been reported in the bracket mushroom S. commune, where SC1, SC3 and SC4 are simultaneously present, albeit at different places (van Wetter et al 2000), and in the button mushroom Agaricus bisporus, where two hydrophobins (ABH1 and ABH2) are detected in different parts of the mature fruit body (Lugones et al 1996, de Groot et al 1999). In this context, the identification of two fruit body-specific hydrophobins in P. ostreatus raises the Departamento de Producción Agraria, Universidad Pública de Navarra, E-31006 Pamplona, Spain Abstract: Two fruit body-specific hydrophobins (Fbh1 and POH1) have been identified in two different strains of the edible basidiomycete Pleurotus ostreatus. Comparison of their nucleotide and amino acid sequences yielded similarity values (59% and 66%, respectively) smaller than those found for alleles of the same hydrophobin gene but higher than those found for different hydrophobin genes in P. ostreatus var. florida (Peñas et al 2002). In this paper, we have addressed the question of Fbh1 and POH1 allelism by studying the structure of the gene fbh1 and by a classical genetic analysis to compare it with that of POH1. The structure of both genes is similar, as revealed by the similarity of their promoters and leader peptide sequences and by the conserved position of their introns. Furthermore, the allelism analysis revealed that both genes segregated as alleles when present in the same hybrid. These results suggest an allelic condition for POH1 and fbh1 and stress the importance of the similarity of fbh1/POH1 promoter and leader sequences. Furthermore, we have identified various microsatellite-like regions in this gene that can be used for strain and species typing in the future. Key words: allele polymorphism, gene structure, microsatellites INTRODUCTION Hydrophobins (Schuren and Wessels 1990) are cysteine-rich, abundantly expressed fungal proteins whose peculiar physico-chemical characteristics (Wösten and Wessels 1997) make them attractive candidates for possible use in several biotechnological applications (Wessels 1997, Scholtmeijer et al 2001, Palomo et al 2003). Hydrophobin genes are among the Accepted for publication May 30, 2003. 1 Corresponding author. E-mail: gpisabarro@unavarra.es 75 76 MYCOLOGIA question whether to consider them as products of duplicate loci, or as alleles of the same locus. This question is more pertinent because sequence similarity at either amino acid (59%) or nucleotide (66%) level between POH1 and Fbh1 is lower than that observed for alleles of a given hydrophobin gene (85.4% and 85.8%) but higher than that observed for different hydrophobin genes (19.6% and 40.8%) in P. ostreatus var. florida (Peñas et al 2002). To clarify this question, we show here that gene fbh1 is structurally similar to POH1 (Asgeirsdóttir et al 1998) but different from other P. ostreatus hydrophobin genes (Peñas et al 2002) and demonstrate that fbh1 and POH1 behave as allelic by means of a genetic segregation analysis. MATERIALS AND METHODS Fungal strains and culture conditions. Pleurotus ostreatus ( Jacq. ex Fr) Kummer var. florida (strain N001) (Larraya et al 1999, 2000) and var. ostreatus (strain N007) (Asgeirsdóttir et al 1998) are the dikaryotic strains used in this work. The two nuclei present in the dikaryotic strain N001 were separated by protoplasting and two monokaryotic strains (protoclones) called PC9 and PC15, containing each one of the two nucleus types, were constructed as explained elsewhere (Larraya et al 1999). The two monokaryotic protoclones of P. ostreatus N001 are deposited in the Spanish Type Culture Collection under accession numbers CECT20311 (PC9) and CECT20312 (PC15). To study the conservation of the promoter in different Pleurotus species and P. ostreatus varieties, a collection of strains from our laboratory (Larraya et al 1999) and others kindly supplied by Sylvan Inc. (Kittanning, Pennsylvania) was used. Spores produced by dikaryotic strains were collected, monokaryotic cultures started and mating types determined as described elsewhere (Larraya et al 1999). The florida 3 ostreatus hybrid strain (N015) was constructed by crossing monokaryon MA005 derived from N001 to monokaryon MG001 derived from N007. Mycelia were cultured on solid Eger Medium (20 g L21 malt extract, 15 g L21 agar) (Eger 1976) or in SMY (10% sucrose w/v, 10% w/v malt extract, 0.4% w/v yeast extract, pH 6.5) (Katayose et al 1986) for liquid cultures and incubated without shaking at 24 C in the dark. Molecular techniques. DNA was purified as described by Larraya et al (1999). For RFLP analysis, the corresponding DNA was digested, following the recommendations of the enzyme suppliers. Hybridizations were performed at 65 C as described by Church and Gilbert (1984). The two fbh1 alleles present in P. ostreatus N001 were PCR-amplified using genomic DNA as template and primers that included the fbh1 start and stop codons (underlined): forward primer 59-ATGTTCTCCATCCGCATC-39 (positions 1–18 from fbh1 translation start point), reverse primer 59-TTAGAGGTTGAGGTTAATG-39 (positions 557 to 539 from fbh1 translation start point) (Peñas et al 1998). PCR mixes were incubated 5 min at 95 C and subjected to 30 cycles of denaturation (95 C, 1 min), annealing (55 C, 1 min) and extension (72 C, 2 min). The amplified fragments were cloned into a pGEM-T vector (Promega, Madison, Wisconsin). For PCR amplification of the fbh1 promoter region in different P. ostreatus strains, these two primers were designed after the sequence of fbh1-1: forward primer 59-CAAACCCCGAATCACGTCC-39 (positions 2547 to 2529 from fbh1 translation start point), reverse primer 59-CGTCGAGCACAGTTGAGTCC-39 (positions 2207 to 2226 from fbh1 translation start point). PCR conditions were: PCR mixes were incubated 5 min at 95 C and subjected to 40 cycles of denaturation (95 C, 1 min), annealing (60 C, 1 min) and extension (72 C, 1.5 min). Two hundred ng of template DNA were used per reaction. The protocols described by Sambrook et al (1989) and Dieffenbach and Dveksler (1995) were followed for general molecular techniques. DNA sequence analysis and nucleotide sequence accession number. Sequencing reactions were performed with an ABI Prism BigDye Terminator Kit (Applied Biosystems, Foster City, California). Sequencing electrophoreses were made with ABI Prism 377 equipment. Sequence alignments and similarity search were carried out with programs CLUSTAL W (Thompson et al 1994), and BLASTN, BLASTX and Pairwise BLAST (Altschul et al 1990) at the National Center for Biotechnology Information (NCBI) Website. The genomic sequences obtained from protoclone PC9 (containing the allele fbh1-1 plus intergenic regions, 3695 bp) and in protoclone PC15 (containing the allele fbh1-2 from the translation start to stop points, 543 bp) have been deposited in the EMBL and GenBank nucleotide sequence databases under accession numbers AJ319663 and AJ416753, respectively. RESULTS Cloning and sequence analysis of the fhb1 gene. In a previous paper the cDNA sequence coding for P. ostreatus fruit body-specific hydrophobin Fbh1 was reported (Peñas et al 1998). This cDNA was used as probe in a genomic highly stringent Southern-blot analysis of P. ostreatus genomic DNA digested with different restriction enzymes (see FIG. 1) to determine the copy number of the corresponding gene. fbh1 was a single copy gene in P. ostreatus N001 genome and no cross-hybridization with other sequences occurred under these experimental conditions (FIG. 1). Digestion of dikaryon N001 genomic DNA with restriction enzyme HindIII yielded two RFLP alleles (FIG. 1, lane 3). To clone the gene fbh1, a genomic library was constructed by complete digestion of P. ostreatus N001 genomic DNA using HindIII and cloning of the products ranging in size from 3 to 4.5 Kbp into a pBluescript-SK vector. The library was screened using fbh1 cDNA as probe, and a 3.7 Kbp HindIII clone containing the fbh1 structural gene as well as upstream and downstream regions was isolated and sequenced. This allele was named fbh1-1. The sequence of genomic clone fbh1-1 was com- b a 55 (AATAAT) 129 20 (AATATA) 55 53 53 53 119 21 (AACAAA) 90 80 — — 33 342 342 342 351 351 336 336 342 327 327 396 82 Different genes are separated by horizontal lines. Genes fbh1, vmh1, vmh2 and vmh3 were isolated from P. ostreatus var. florida. Genes POH1, POH2, and POH3 were isolated from var. ostreatus. 0.63 0.65 0.82 0.64 0.64 0.83 0.78 0.97 0.11 0.11 0.62 5.46 5.46 4.68 4.78 4.85 5.77 5.77 8.13 7.48 5.21 8.17 25 25 25 28 28 24 24 24 23 23 23 11.10 11.12 11.01 11.67 11.80 11.22 11.20 11.21 11.15 11.15 13.11 113 113 113 116 116 111 111 113 108 108 131 30 (AATATT) 155 61 64 53 85 68 59 53 55 51 52 54 72 72 54 70 69 64 53 55 56 56 49 53 52 58 153 48 153 48 162 45 47 47 83 48 83 48 83 48 47 47 159 48 51 51 45 38 38 76 76 76 38 38 78 Isoelectric point Signal peptide Poly(A) signal position 39 UTR Introns Length (nt) of: Exons ORF 59 UTR Genea,b TABLE I. Summary of the characteristics of P. ostreatus hydrophobin genes and proteins No. of amino acid residues Molecular mass (kDa) FIG. 1. Identification of the fbh1 alleles present in P. ostreatus var. florida. Genomic DNA isolated from dikaryon N001 and from protoclones PC9 and PC15 was digested with different restriction enzymes and probed with fbh1 cDNA. Lanes 1–3, PC9, PC15, and N001, respectively, digested with HindIII. Lanes 4–11, N001 digested with HindIII 1 EcoRI (4), EcoRI (5), EcoRI 1 BamHI (6), BamHI (7), BamHI 1 BglII (8), BglII (9), BglII 1 HindIII (10) and BamHI 1 HindIII (11). An identical RFLP pattern was obtained when POH1 cDNA was used as probe (data not shown). pared with that of the fbh1 cDNA previously determined by Peñas et al (1998) to determine the intron and exon regions. The matching between the cDNA sequence and the genomic clone was perfect and let us determine that the fbh1-1 allele (TABLE I) contained three introns that bore the conserved splice borders (59GT, 39AG) and the internal splice signals (PyGCTAAC) previously described in filamentous fungi (Gurr et al 1987) and in other P. ostreatus hydrophobin genes (Peñas et al 2002). A 667-bp DNA region was sequenced upstream of fbh1-1 translation start point. This region included the 82 bp 59 transcribed but untranslated region (59-UTR) previously described in this gene (Peñas et al 1998). The putative transcription start site was placed at position 282 from the translation start by comparison between the genomic DNA and cDNA sequences. Putative CAAT and TATA boxes were found at positions 2403 and 2111, respectively, from the translation start. A region of 2470 bp downstream of the translation stop codon was sequenced, which included the 155 bp 39UTR sequence previously reported (Peñas et al 1998). A putative polyadenylation signal (AATATT) was found at position 1679 from the translation start (30 bp downstream from the translation stop codon). To check if other proteins in addition to Fbh1 were coded in the 3.7 Kbp fragment analyzed, the complete sequence (3.7 Kbp) was used as query in a BLASTX (all possible nucleotide translations versus protein sequences) search for homologous proteins in the databases. No other positive similarity results were found outside the hydrophobin coding region, 77 FBH-1 GENE STRUCTURE 90 90 90 266 266 129 129 135 242 242 111 PLEUROTUS OSTREATUS fbh1-1 fbh1-2 POH1 vmh1-1 vmh1-2 vmh2-1 vmh2-2 POH3 vmh3-1 vmh3-2 POH2 ET AL: Hydropathy index PEÑAS 78 MYCOLOGIA suggesting that fbh1 could be the only gene present in this DNA fragment (data not shown). The G1C content of the 3.7 Kbp sequenced was 55.0%. This value was higher at the coding region, where G1C content rose to 56.9% in introns and 64.3% in exons. A 64-bp AT-rich stretch was found between the putative CAAT and TATA boxes containing 10 nearly identical repetitions of the motif ACTTT (positions 2345 to 2281 from fbh1 translation start point). The G1C content of this region was reduced concomitantly (21.5%). In the region coding for fbh1, the codon usage was biased: 64% of codons end with a pyrimidine, and 85% of them had a C at this position. In the case of codons ending with a purine, most of them had a G at the third position. To clone the second fbh1 allele present in P. ostreatus N001 (fbh1-2), PCR amplifications were performed; DNA from protoclones PC9 or PC15 (monokaryons containing each one of the two nuclei present in dikaryon N001) as template and oligonucleotides corresponding to the ends of the translated sequence as primers were used. This amplification strategy would produce a DNA fragment that included exclusively the translated exons and intervening introns; the untranslated upstream and downstream regions as well as the intergenic regions, on the contrary, would not be recovered. Monokaryotic protoclone PC9 bore the allele fbh1-1 described above, whereas protoclone PC15 carried the second allele (fbh1-2), which showed some minor differences with respect to fbh1-1 (TABLE I). The GC content in the coding and noncoding (introns) regions was identical in both alleles. Sequence similarity between the two fbh1 alleles scored 96% at the nucleotide and 99% at the aminoacid level (data not shown). Sequence and structure comparison between genes fbh1 and POH1. Fbh1 was isolated from P. ostreatus var. florida. Another fruit-body specific hydrophobin (POH1) independently was isolated from P. ostreatus var. ostreatus (Asgeirsdóttir et al 1998). To check the level of cross-hybridization between the fbh1 and POH1 cDNAs, a Southern blot containing both of them digested with PstI was probed independently with fbh1 and POH1 cDNAs (FIG. 2). PstI was chosen because a restriction site for this enzyme is present in the sequence of fbh1 cDNA and absent in that of POH1. Furthermore, the splitting of fbh1 cDNA into two moieties allowed the distinction between the 59 and 39 ends of the cDNA and consequently between the regions coding for the N-terminal and C-terminal protein ends. Both probes cross-hybridized, although with different intensities. Furthermore, probe POH1 recognized exclusively the larger PstI fragment of the fbh1 cDNA corresponding to its 59 end. This result FIG. 2. Cross-hybridization between fbh1 and POH1 probes. Southern blot containing POH1 (lane 1) and fbh1 (lane 2) cDNAs digested with PstI and probed with fbh1 (left panel) and POH1 (right panel) probes. suggested that the homology between fbh1 and POH1 was localized mainly at the 59 end of their cDNAs. To test this hypothesis, the complete genomic sequence that includes allele fbh1-1 (3.7 Kbp) was used as query in a BLASTN (nucleotide versus nucleotide) database search for homologous sequences using a word size of 10 nt. Under these conditions, only a short region corresponding to the first 86 bp of POH1 sequence was retrieved (86% identity), indicating that sequence similarity between the two genes decreased substantially beyond this point. No other nucleotide sequences with relevant similarity to any portion of the 3.7 Kbp fragment were found in the databases. To determine the similarity between the nucleotide sequences of fbh1-1 and POH1, a Pairwise BLAST was made using the 3.7 Kbp fragment and the sequence of the POH1 gene (accession number AJ225060) using the default blasting conditions. In this case, the region of similarity between the two sequences was extended up to position 118 in POH1 cDNA and included the coding region of the first exon and the first 28 bp of the first intron of both genes. When the genomic regions coding for either fbh11 or fbh1-2 were used in paired comparisons with the cDNA of POH1, similar results were obtained: The global nucleotide similarity level between fbh1 and POH1 was 66% (data not shown). When amino acid sequences deduced from the nucleotide sequences of the three genes were compared, a nearly complete conservation of the leader peptide sequence was found (FIG. 3). In the mature protein, however, the similarity level dropped to a value of 59%. Published POH1 sequence corresponds exclusively to the translated portion of the gene (Asgeirsdóttir et al 1998). To study the conservation of the fbh1 promoter in other P. ostreatus strains, a PCR experiment was performed to amplify a 341 bp fragment PEÑAS ET AL: PLEUROTUS OSTREATUS FIG. 3. Comparison of the amino acid sequences of the Fbh1/POH1 alleles. The shadowed rectangles indicate amino acid polymorphisms. Dashes denote missing amino acids. The amino acids underlined correspond to the signal peptide sequence. The triangles above the sequence of Fbh1-1 indicate intron positions in Fbh1-1 and Fbh1-2. The black arrow indicates position of intron 1 in the POH1 as predicted by the sequence alignment program. of the promoter sequence comprising the putative CAAT and ACTTT boxes described above in different P. ostreatus strains. FIGURE 4 shows that this sequence is present in all P. ostreatus strains tested, indicating that this promoter region of fbh1 may be conserved in all of them. Allelism analysis. To determine if fbh1 and POH1 were alleles of the same locus, the presence of sequences homologous to POH1 in the genome of P. ostreatus N001 was investigated. DNA samples from dikaryon N001 and from protoclones PC9 and PC15 were fully digested with a number of different restriction enzymes in single and double digestions, Southern blotted and probed with fbh1 and POH1 (fbh1 and POH1 cDNAs, respectively). The two probes highlighted identical hybridization patterns (data not shown), indicating that no other P. ostreatus N001 genomic sequences in addition to those correspond- FBH-1 GENE STRUCTURE 79 FIG. 5. Segregation of fbh1/POH1 alleles. Genomic DNA isolated from hybrid dikaryon N015 (lane 1), its two parental monokaryons (MA005, lane 2; and MG001, lane 3) and of seven monokaryons derived from it (lanes 4–10) was digested with EcoRI and probed with POH1 cDNA. An identical RFLP pattern was obtained when fbh1 cDNA was used as probe (data not shown). ing to the fbh1 gene were similar to POH1. Consequently, the portion of POH1 cDNA with little similarity to fbh1 (that is, downstream of position 118 in the sequence of POH1) failed to detect other genomic sequences in P. ostreatus N001. To confirm the allelism of fbh1 and POH1, compatible monokar yons derived from var. florida (MA005) and var. ostreatus (MG001) were mated to construct hybrid dikaryon N015, which later was placed under fruiting and sporulation conditions. Genomic DNA from dikaryon N015 and from 24 monokaryons randomly selected among its progeny was purified, digested with EcoRI, blotted and analyzed with fbh1 and POH1 probes. Both probes revealed two hybridization bands of 8.5 and 5.0 Kb, which segregated as alleles in the collection of monokaryons (FIG. 5 corresponds to the experiment probed with fbh1 cDNA). The analysis of the popu- FIG. 4. Structure of the fbh1 gene and analysis of its promoter. (A) Scheme of the gene. The black box represents the conserved ACTTT motif, vertical lines are the CAAT and TATA boxes, white boxes represent the 59 and 39 UTRs and dashed boxes represent the exons. The arrows represent the primers used for the amplification of the promoter sequence. (B) PCR amplification of the promoter sequence in different P. ostreatus strains and Pleurotus species. Lane 1, positive control; 2–7, P. ostreatus N001–N006 (Larraya et al 1999); 8, P. ostreatus HK35; 9, P. pulmonarius; 10, P. colombinus; 11, P. cornucopiae strain 3040; 12, P. eryngii; 13, P. sapidus; 14, P. cornucopiae strain 608; 15, P. saja-caju strain 1; 16, P. saja-caju strain 2; 17, P. ostreatus N001 protoclone PC9; 18, P. ostreatus N001 protoclone PC15. 80 MYCOLOGIA lation showed that the 8.5 Kb fragment was provided by MG001 while the 5.0 Kb band originally was present in MA005. A slight difference in the hybridization intensity was observed between the 8.5 and 5.0 Kb bands depending on the probe used: The 8.5 Kb band was fainter when probed with fbh1, while the 5.0 Kb band was fainter when probed with POH1. DISCUSSION Two different fruit body-specific hydrophobins have been isolated from the edible basidiomycete P. ostreatus: Fbh1 from var. florida (Peñas et al 1998) and POH1 from var. ostreatus (Asgeirsdóttir et al 1998). A high redundancy of hydrophobin genes has been reported in several mushrooms (Wessels 2000), including P. ostreatus (Peñas et al 2002) and, in this scenario, the question about allelism versus gene duplication arises. Furthermore, this question was more pertinent in the case of Fbh1 and POH1 because the similarity between both sequences is lower than that previously reported for alleles of the same hydrophobin gene in P. ostreatus (Peñas et al 2002). In this paper we have addressed this question, and our results suggest that both proteins are products of alleles of the same gene. Two kinds of evidence support this assertion: (i) the structure of genes fbh1 and POH1 is highly similar, as revealed by the conservation of their coding sequence sizes, promoter regions, leader peptides and number and position of introns and exons (TABLE I); and (ii) fbh1 and POH1 appear to behave as alleles when both sequences are present in the same dikaryon (FIG. 5). Both fbh1 and POH1 are single-copy genes in P. ostreatus as revealed by genomic Southern analysis. The RFLP analysis showed that probes corresponding to either of the two genes revealed DNA fragments of the same size with all the restriction enzymes tested. The three Fbh1/POH1 hydrophobins have 113 amino acid residues. The two alleles of fbh1 have a 96% (nucleotide) and 99% (aminoacid) sequence similarity; however, sequence comparison of Fbh1-1 and Fbh1-2 with POH1 produced amino acid identities of 59 and 58%, respectively, and 67% homology (including conservative substitutions) (FIG. 3). Sequence conservation is higher in the signal peptide region (68% identity, 76% homology) than in the mature protein (53% identity, 64% homology). Moreover, the amino acid sequences flanking the leader peptide processing site are absolutely conserved in the three proteins, suggesting that they use the same export mechanism. In addition to that, the length of the leader peptides in Fbh1 and POH1 is identical. The leader peptide size seems to be con- served between alleles and to differ between different hydrophobin genes (TABLE I). Two major deletion/insertion events can be observed when Fbh1 and POH1 are compared (FIG. 3): The two Fbh1 variants have two extra amino acids preceding the first Cys residue and lack three in the stretch between the third and fourth Cys residues. The connectivity proposed for the four disulfide bonds that can be formed in hydrophobins (i.e., Cys 1–2, 3–4, 5–6 and 7–8, [Wessels 1997]) suggests a four-looped structure for these proteins (Wessels 2000, Wösten and de Vocht 2000). The insertion (Fbh1 versus POH1) corresponds to the amino acids flanking the splicing site of intron-1 upstream of the first protein loop, and the deletion event occurs in exon-3 within the major protein loop. An additional minor deletion/insertion event can be observed at the protein C-terminus, where an Asn residue present in Fbh1 is missing in POH1. Intron number and position is similar in genes fbh1 and POH1 (FIG. 3). P. ostreatus hydrophobin genes can be grouped into two classes according to their intron number (TABLE I): Genes vmh1 and vmh3 contain two, whereas genes fbh1/POH1, POH2 and vmh2/POH3 contain three introns. This second class of genes can be further divided into two groups because not all vmh2/POH3 introns are equivalent to those present in fbh1/POH1 and POH2. Only one equivalent intron is present in all P. ostreatus hydrophobin sequences. This intron limits the final hydrophobin exon that is conserved in size in all P. ostreatus hydrophobins, which includes the eighth conserved Cys residue. The conservation of the last hydrophobin exon could represent one of the oldest features of this gene family because other genes coding for hydrophobins in higher basidiomycetes, such as A. bisporus ABH1 and ABH2 (Lugones et al 1996) and S. commune SC1, SC2 and SC3 (Asgeirsdóttir 1994), contain an exon with similar characteristics. The allelism analysis described in this paper indicates that fbh1 and POH1 behave as alleles when they are present in the same individual (FIG. 5). In A. bisporus, it has been shown that genes ABH1 and ABH2 are arranged in a closely linked tandem (Lugones et al 1996). The possibility that fbh1 and POH1 were two different genes ordered in a similar way can be ruled out because no evidence of a sequence homologous to POH1 (in addition to fbh1) was found in the 3.7 Kb genomic region studied in this work. Were POH1 sequence outside this genomic fragment, segregation of its RFLP signals would not have been compatible with that of an allele of fbh1, or secondar y hybridization signals would have appeared. This never has been the case. Two additional features of gene fbh1 deserve spe- PEÑAS ET AL: PLEUROTUS OSTREATUS cial consideration: the structure of fbh1 promoter and the occurrence of microsatellite-like motifs in the region sequenced. Hydrophobin genes are highly expressed in a developmentally regulated manner. Three features of the fbh1 promoter sequence can be responsible for the high expression level. First, the presence of a putative CAAT and a canonical TATA boxes separated by an AT-rich motif (ACTTT box) that markedly reduces the local DNA dissociation temperature. In genes from filamentous fungi, the CAAT consensus box, when present, usually is found between 60–120 bp upstream the major transcription start point (Gurr et al 1987). The distant position of the putative fbh1 CAAT box (-403), however, is not exceptional because a similar position has been reported in gene Aa-Pri2 coding for a fruiting-initiation-specific hydrophobin in Agrocybe aegerita (Santos and Labarère 1999). Second, the region flanking the initial ATG (ACACAATGTT) conforms with the consensus sequence postulated by Kozak, where nucleotide in position-3 is always a purine (preferentially adenine) (Kozak 1981). Third, the codon usage in fbh1 that is biased as described for highly expressed genes in filamentous fungi (Gurr et al 1987). Four microsatellite-like sequences are prominent in the fbh1 sequence described here. The AT-rich region already described above that comprises 10 nearly identical ACTTT motifs: the occurrence of short repetitive regions of 4–7 nucleotides in intron-2 (sequence element GTACCTT occurs twice in fbh1-1 while only once in fbh1-2, element ACCAAAC occurs three times in fbh1-1 and once in fbh1-2, and the short tandem repeated element ACTA is found twice in fbh1-2 but once in fbh1-1); a 12-times repeated AGG motif at sequence positions 2351–2389; and a five times repeated GTCATA motif at sequence positions 3538–3568. These microsatellite-like elements can be used in species and variety differentiation within the genus Pleurotus. ACKNOWLEDGMENT This work was supported by research project BIO99–0278 of the Comisión Nacional de Ciencia y Tecnologı́a. LITERATURE CITED Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215:403– 410. Asgeirsdóttir SA. 1994. Proteins involved in emergent growth of Schizophyllum commune. Groningen, The Netherlands: University of Groningen. ———, de Vries OM, Wessels JG. 1998. Identification of three differentially expressed hydrophobins in Pleuro- FBH-1 GENE STRUCTURE 81 tus ostreatus (oyster mushroom). Microbiol 144:2961– 2969. Church GM, Gilbert W. 1984. Genomic sequencing. Proc Natl Acad Sci USA 81:1991–1995. de Groot PW, Roeven RT, Van Griensven LJ, Visser J, Schaap PJ. 1999. Different temporal and spatial expression of two hydrophobin-encoding genes of the edible mushroom Agaricus bisporus. Microbiol 145:1105–1113. ———, Schaap PJ, Sonnenberg AS, Visser J, Van Griensven LJ. 1996. The Agaricus bisporus hypA gene encodes a hydrophobin and specifically accumulates in peel tissue of mushroom caps during fruit body development. J Mol Biol 257:1008–1018. Dieffenbach CW, Dveksler GS. 1995. PCR Primer. A laboratory manual. Cold Spring Harbor, USA: Cold Spring Harbor Laboratory Press. Eger G. 1976. Pleurotus ostreatus-breeding potential of a new cultivated mushroom. Theor Appl Genet 47:155– 163. Gurr SJ, Unkles SE, Kinghorn JR. 1987. The structure and organization of nuclear genes in filamentous fungi. In: Gurr SJ, Unkles SE, Kinghorn JR, eds. Gene structure in eukaryotic microbes. Oxford: IRL Press. p 93–139. Katayose Y, Shishido K, Ohmasa M. 1986. Cloning of Lentinus edodes mitochondrial DNA fragment capable of autonomous replication in Saccharomyces cerevisiae. Biochem Biophys Res Commun 138:1110–1115. Kershaw MJ, Talbot NJ. 1998. Hydrophobins and repellents: proteins with fundamental roles in fungal morphogenesis. Fungal Genet Biol 23:18–33. Kozak M. 1981. Composition and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Nucl Acids Res 12:857–872. Larraya L, Peñas MM, Pérez G, Santos C, Ritter E, Pisabarro AG, Ramı́rez L. 1999. Identification of incompatibility alleles and characterization of molecular markers genetically linked to the A incompatibility locus in the white rot fungus Pleurotus ostreatus. Curr Genet 34: 486–493. ———, Pérez G, Peñas MM, Baars JJ, Mikosch TS, Pisabarro AG, Ramı́rez L. 1999. Molecular karyotype of the white rot fungus Pleurotus ostreatus. Appl Environ Microbiol 65:3413–3417. Larraya LM, Pérez G, Ritter E, Pisabarro AG, Ramı́rez L. 2000. Genetic linkage map of the edible basidiomycete Pleurotus ostreatus. Appl Environ Microbiol 66:5290– 5300. Lugones L, Bosscher JS, Scholtmeyer K, de Vries OMH, Wessels JGH. 1996. An abundant hydrophobin (ABH1) forms hydrophobic rodlet layers in Agaricus bisporus. Microbiol 142:1321–1329. ———, Wösten HAB, Wessels JGH. 1998. A hydrophobin (ABH3) specifically secreted by vegetatively growing hyphae of Agaricus bisporus (common white button mushroom). Microbiol 144:2345–2353. Palomo JM, Peñas MM, Fernández-Lorente G, Mateo C, Pisabarro AG, Fernández-Lafuente R, Ramı́rez L, Guisán JM. 2003. Solid phase handling of hydrophobins: immobilized hydrophobins as a new tool to study lipases. Biomacromol 4:204–210. 82 MYCOLOGIA Peñas MM, Asgeirsdóttir SA, Lasa I, Culiañez-Macià FA, Pisabarro AG, Wessels JGH, Ramı́rez L. 1998. Identification, characterization, and in situ detection of a fruitbody-specific hydrophobin of Pleurotus ostreatus. Appl Environ Microbiol 64:4028–4034. ———, Rust B, Larraya LM, Ramı́rez L, Pisabarro AG. 2002. Differentially regulated vegetative mycelium specific hydrophobins of the edible basidiomycete Pleurotus ostreatus. Appl Environ Microbiol 68:3891–3898. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor, USA: Cold Spring Harbor Laboratory. Santos C, Labarère J. 1999. Aa-Pri2, a single-copy gene from Agrocybe aegerita, specifically expressed during fruiting initiation, encodes a hydrophobin with a leucine-zipper domain. Curr Genet 35:564–570. Scholtmeijer K, Wessels JG, Wösten HA. 2001. Fungal hydrophobins in medical and technical applications. Appl Microbiol Biotechnol 56:1–8. Schuren FHJ, Wessels JGH. 1990. Two genes specifically expressed in fruiting dikaryons of Schizophyllum commune: homologies with a gene not regulated by matingtype genes. Gene 90:199–205. Schuurs TA, Dalstra HJ, Scheer JM, Wessels JG. 1998. Positioning of nuclei in the secondary mycelium of Schi- zophyllum commune in relation to differential gene expression. Fungal Genet Biol 23:150–61. Segers GC, Hamada W, Oliver RP, Spanu PD. 1999. Isolation and characterization of five different hydrophobin-encoding cDNAs from the fungal tomato pathogen Cladosporium fulvum. Mol Gen Genet 261:644–652. Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl Acid Res 22:4673–4680. van Wetter MA, Wösten HA, Wessels JG. 2000. SC3 and SC4 hydrophobins have distinct roles in formation of aerial structures in dikaryons of Schizophyllum commune. Mol Microbiol 36:201–210. Wessels JGH. 1997. Hydrophobins: proteins that change the nature of fungal surface. Adv Microbial Physiol 38:1– 44. ———. 2000. Hydrophobins, unique fungal proteins. Mycologist 14:153–159. Wösten HA. 2001. Hydrophobins: multipurpose proteins. Annu Rev Microbiol 55:625–46. ———, de Vocht ML. 2000. Hydrophobins, the fungal coat unravelled. Biochim Biophys Acta 1469:79–86. ———, Wessels JGH. 1997. Hydrophobins, from molecular structure to multiple functions in fungal development. Mycoscience 38:363–374.