Analysis of expressed sequence tags from the ectomycorrhizal basidiomycetes Laccaria bicolor and Pisolithus microcarpus

Authors

  • Martina Peter,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    2. These authors contributed equally to this work
    Search for more papers by this author
  • Pierre-Emmanuel Courty,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    2. These authors contributed equally to this work
    Search for more papers by this author
  • Annegret Kohler,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • Christine Delaruelle,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • David Martin,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • Denis Tagu,

    1. INRA Rennes, Unité Mixte de Recherche BiO3P, BP 35327, 35653 Le Rheu Cedex, France;
    Search for more papers by this author
  • Pascale Frey-Klett,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • Sébastien Duplessis,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • Michel Chalot,

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author
  • Gopi Podila,

    1. Dept. of Biological Sciences, University of Alabama, Huntsville, AL 35899, USA;
    Search for more papers by this author
  • Francis Martin

    1. Unité Mixte de Recherche INRA–UHP 1136 ‘Interactions Arbres/Microorganismes’, Centre de Recherches de Nancy, 54280 Champenoux, France;
    Search for more papers by this author

Author for correspondence: Martina Peter Tel: +33 383 39 40 80 Fax: +33 383 39 40 69 Email: peter@nancy.inra.fr

Summary

  • • In an effort to discover genes that are expressed in the ectomycorrhizal basidiomycetes Laccaria bicolor and Pisolithus microcarpus, and in P. microcarpus/Eucalyptus globulus ectomycorrhizas, we have sequenced 1519 and 1681 expressed sequence tags (ESTs) from L. bicolor and P. microcarpus cDNA libraries.
  • • Contig analysis resulted in 905 and 806 tentative consensus sequences (unique transcripts) in L. bicolor and P. microcarpus, respectively. For 36% of the ESTs, significant similarities to sequences in databases were detected. The most abundant transcripts showed no similarity to previously identified genes. Sequence redundancy analysis between different developmental stages indicated that several genes were differentially expressed in free-living mycelium and symbiotic tissues of P. microcarpus.
  • • Based on sequence similarity, 11% of L. bicolor unique transcripts were also detected in P. microcarpus. Similarly, L. bicolor and P. microcarpus shared only a low proportion of common transcripts with other basidiomycetous fungi, such as Pleurotus ostreatus and Agaricus bisporus. Such a low proportion of shared transcripts between basidiomycetes suggests, on the one hand, that the variability of expressed transcripts in different fungi and fungal tissues is considerably high. On the other hand, it might reflect the low number of GenBank entries of basidiomycetous origin and stresses the necessity of an additional sequencing effort.
  • • The present ESTs provide a valuable resource for future research on the development and functioning of ectomycorrhizas.

Introduction

Most tree species in temperate and boreal forests live in symbiosis with ectomycorrhizal fungi. These fungi play a crucial role in forest tree health by improving nutrient acquisition, drought tolerance and pathogen resistance of their plant hosts. In return, autotrophic plants provide their heterotrophic fungal partners with carbohydrates (Smith & Read, 1997). In the ectomycorrhizal association, fungal hyphae colonize absorbing fine roots of trees, develop a mantle around these and penetrate between outer root cells forming the so-called Hartig net. Extraradical hyphae spread into the surrounding soil, take up nutrients and deliver them to the ectomycorrhizal organ in which nutrient exchange takes place.

Many aspects of the ectomycorrhizal symbiosis have been extensively studied both in ecological and physiological respects (Smith & Read, 1997). The formation and functioning of the symbiosis include major changes in cellular and tissue morphology (Peterson & Bonfante, 1994), as well as in the biochemistry and physiology of the partners (Botton & Chalot, 1999; Hampp et al., 1999; Martin et al., 1999). Emerging genomic tools such as expressed sequence tags (ESTs) and the cDNA array technology provide a new approach to the understanding of ectomycorrhizal development and functioning at the molecular level (Martin, 2001). These techniques allow to rapidly identify genes and to perform large-scale functional analyses of thousands of them (Ewing et al., 1999; Skinner et al., 2001). Because of the wide spectrum of genes and signals involved, these genomic tools are well suited to study molecular events in symbiotic interactions (Györgyey et al., 2000; Voiblet et al., 2001; Podila et al., 2002).

At present, there is, however, a lack of a coordinated development of resources aimed at sequencing the genome of ectomycorrhizal fungi or producing genetic tools that are useful to the scientific community (Martin, 2001). That is, the production of large numbers of ESTs or insertion and expression tagged lines, as are available for several nonsymbiotic model fungi. There is a relatively small (1642 ESTs published, GenBank release 032103), albeit slowly accumulating, body of EST sequence data derived from a number of ectomycorrhizal fungi in public (dbEST at the National Centre for Biotechnology Information [NCBI]) and in various local databases. It includes ESTs for Hebeloma cylindrosporum (NCBI dbEST & Sentenac et al. personal communication), Tuber borchii (Lacourt et al., 2002; Polidori et al., 2002), Laccaria bicolor (Podila et al., 2002), Amanita muscaria (U. Nehls et al., personal communication), and Paxillus involutus (Johansson et al., personal communication). To speed the discovery of novel genes and functions involved in ectomycorrhiza development, we have developed ESTs databases of 4-d-old Pisolithus microcarpus/Eucalyptus globulus ectomycorrhizas (Tagu & Martin, 1995; Voiblet et al., 2001).

In the project described here, we constructed cDNA libraries and generated ESTs from the free-living mycelium of P. microcarpus and of another ectomycorrhizal model species, L. bicolor. In addition, cDNA clones from P. microcarpus/E. globulus ectomycorrhizas at different developmental stages (4-, 12-, 21-d-old) were partially sequenced to increase our current EST database of Pisolithus ectomycorrhiza. Because the growth and mycorrhiza development conditions were distinct, we expected to identify a large number of novel genes. The gene diversity of the two ectomycorrhizal basidiomycetes was assessed throught EST sequencing, assembly and analysis. These data provide the basis for an initial glimpse into the overall metabolism and biology of fungal symbionts as revealed by expressed transcripts.

Materials and Methods

Strains and culture conditions

The ectomycorrhizal basidiomycetes Laccaria bicolor (Maire Orton) isolate S238N (Di Battista et al., 1996) and Pisolithus microcarpus (Coker & Mass.) Cunn. nom prov. isolate 441 (formerly P. tinctorius 441; Martin et al., 2002) were grown and maintained on Pachlewski medium agar plates (Nehls & Martin, 1995). For cDNA library construction of pure culture mycelium, the isolates were transferred onto cellophane-covered agar plates containing high sugar (20 g l−1 glucose, 5 g l−1 maltose; L. bicolor) or low sugar (1 g l−1 glucose; P. microcarpus) Pachlewski medium, and were grown for 3 weeks before harvesting the proliferating hyphal tips at the periphery. Ectomycorrhizas of P. microcarpus and E. globulus ssp. bicostata Kirkp. were synthesized using two previously described in vitro systems (vertically or horizontally oriented agar plates; Malajczuk et al., 1990; Burgess et al., 1996).

Isolation of total RNA and construction of cDNA libraries

Total RNA was isolated from snap-frozen (liquid nitrogen) and grounded fungal tissues using the RNeasy Plant Mini kit (Qiagen, Valencia, CA, USA) or according to Logemann et al. (1987). cDNA libraries of pure culture mycelium of L. bicolor and P. microcarpus were constructed from total RNA using the SMART cDNA synthesis kit in λTriplEx2 (Clontech, Palo Alto, CA, USA). The resulting cDNA was packed into phages using the Gigapack III Gold packaging kit (Stratagene, La Jolla, CA, USA). Aliquots of the libraries were amplified, followed by in vivo excision of the pTriplEx2 phagemid according to the manufacturer's instructions. Owing to the large proportion of ESTs coding for rRNA genes in the library of L. bicolor mycelium, a second cDNA library was constructed used poly(A)+ RNA. The mycelium was grown on low sugar Pachlewski medium for 10 weeks, sampled and immediately frozen in liquid nitrogen. Total RNA was extracted using the Qiagen Plant Mini kit and poly(A)+ RNA was enriched from 13 µg of total Dnase-digested RNA, using the Qiagen Oligotex kit. cDNA library was constructed in λTriplEx2 as previously described.

Several cDNA libraries were constructed of P. microcarpus/E. globulus ectomycorrhizas. Three cDNA libraries were set-up from total RNA of ectomycorrhizas formed in the horizontal Petri dish system (Malajczuk et al., 1990) and harvested after 4 d of contact. These libraries and their construction have already been described (Tagu et al., 1993; Voiblet et al., 2001). Three additional libraries were constructed within the framework of the present study, as described in the previous section, from total RNA of ectomycorrhizas formed in the vertical Petri-dish system (Burgess et al., 1996) and harvested after 4-, 12-, and 21-d of contact between the symbionts.

DNA sequencing

Aliquots of the pTriplEx2 phagemid libraries were used for infecting E. coli BM25,8 cells of OD600 1.0 and were subsequently plated on Luria-Bertani (LB) agar containing ampicillin. About 5000 bacterial clones from the various libraries were randomly collected, inoculated into 96-well plates containing selective LB media, and grown overnight without agitation at 37°C. Glycerol was added to a final concentration of 40%. Backup plates were created and stored at −80°C. Aliquots of 3 µl were PCR (in 50 µl) amplified using FORNAT (5′-AAGCGCGCCATTGTGTTGGTACCC-3′) and REVALEX (5′-CGGCCGCATGCATAAGCTTGCTCG-3′) as primers (present in the pTriplEx2 vector arms) (Kohler et al., 2003). The PCR included 95°C for 3 min, 95°C for 60 s, 60°C for 30 s, 72°C for 3 min for 30 cycles and a final extension at 72°C for 15 min (GeneAmp System 9700; Perkin Elmer, Boston, MA, USA). Five µl of each reaction were analysed on a 1% agarose gel and stained with ethidium bromide to control the size and quality of the PCR products. Excess primers and nucleotides were removed by ultrafiltration using the Montage 96-well plate system (Millipore MAHV N45). Purified PCR products were subjected to nucleotide sequencing on either a multicapillary sequencer CEQ 2000XL (Beckman Coulter, Fullerton, CA, USA) or on a ABI Prism 310 Genetic Analyser (Applied Biosystems, Foster City, CA, USA). We used the FORNAT primer, 50 bp upstream of the 5′ end of cDNA insert, and either the CEQ Dye-labelled Dideoxy-Terminator Cycle Sequencing kit (Beckman Coulter), or the BigDye Terminator Cycle sequencing kit and the POP-4 matrix (Applied Biosystems) according to the manufacturer's instructions.

The average length of readable sequences was 580 bp on the CEQ 2000XL sequencer (P. microcarpus ESTs), and 390 bp on the ABI Prism 310 genetic analyser (L. bicolor ESTs). Sequences obtained from mycelial cultures of L. bicolor S238N were designated in the database with an ‘Lb’ at the beginning of the sequence identification number (Lb01-Lb15, cDNA library constructed using total RNA; Lb16-Lb30, cDNA library constructed using mRNA). Sequences obtained from mycelial cultures of P. microcarpus were designated in the database with a ‘P’, whereas ESTs from 4-, 12-, and 21-d-old-ectomycorrhizas of P. microcarpus/E. globulus were named ‘EP4’, ‘EP12’, and ‘EP21’, respectively. To this set of novel sequences, we added ESTs generated in previously published studies (Nehls & Martin, 1995; Tagu & Martin, 1995; Voiblet et al., 2001). ESTs from randomly sampled cDNAs of 4-d-old P. microcarpus/E. globulus ectomycorrhizas, which were produced in the horizontal agar-plate system (Tagu & Martin, 1995; Voiblet et al., 2001), were designated with an ‘St’, ‘un’, or no prefix. ESTs from cDNA cloned using the suppressive subtractive hybridization (SSH) procedure (Voiblet et al., 2001) or differential screening of cDNA clones (Nehls & Martin, 1995) were designated with an ‘EgPtd’ and ‘ud’, respectively.

Sequence processing and annotation

All sequence outputs obtained from the automated sequencers were scanned visually to confirm overall quality of peak shape and correspondence with base calls. Sequence data were uploaded in SEQUENCHER (version 3.1.1) (Gene Codes Corporation, Ann Arbor, MI, USA) programme for Macintosh. Leading and trailing vector and polylinker sequences, and sequence ends with more than 3% ambiguous base calls were removed by SEQUENCHER filters. ESTs with less than 100 bp sequence information were eliminated. Edited sequences were exported as FASTA text files for further processing. The EST sequences were deposited in the GenBank dbEST at the NCBI (GenBank accession no. AW600807-AW600908, AW731605-AW731617, BE704426-BE704449, BF707467-BF707495, BF942500-BF942695, L41693-L41726, and CB009716-CB012283). A Mac OS X-compatible programme, MacESTtools (available from http://mycor.nancy.inra.fr/PoplarDB), was used for batch execution of BLASTN and WU-BLASTX against the nonredundant (nr) nucleic acid sequence databases at the Baylor College of Medecine Web server (Worley et al., 1995). This software was also used to retrieve best matches from the output BLAST html files and generate datatables. Sequences with an E-value < 1.0e−4 were considered to identify known genes or have partial similarity to known genes. Finally, MacESTtools was used to upload BLAST results in a searchable MySQL database containing raw sequences and BLAST results (http://mycor.nancy.inra.fr/EctomycorrhizaDB/). The web site also provides the opportunity to search the EST sequences using NCBI BLAST. The functions of ESTs were assigned based on the BLASTX search and annotated manually following the Munich Information Center for Protein Sequences (MIPS) role categorization (http://mips.gsf.de/proj/yeast/catalogues).

Assembly of the individual ESTs into groups of tentative consensus sequences (TCs), representing unique transcripts, was performed using the contig routine (80% identity over 40 nt length) of SEQUENCHER. The degree of amino acid-sequence similarity between ESTs of various basidiomycetous species was evaluated using the NCBI tBLASTX algorithm in a stand-alone NCBI BLAST with executables for Apple Macintosh downloaded from ftp://ftp.ncbi.nih.gov/blast/executables/. The matrix BLOSUM62 and default settings of BLASTALL was used to BLAST against locally created EST databases of L. bicolor, P. microcarpus, and databases of ESTs of Agaricus bisporus, Hebeloma cylindrosporum and Pleurotus ostreatus retrieved from GenBank (release 121302).

Generation and analysis of cDNA arrays

Microarrays were produced separately for either L. bicolor or P. microcarpus by spotting up to 1600 PCR-amplified cDNA inserts onto nylon membranes using the BioGrid arrayer (BioRobotics, Cambridge, UK) according to the manufacturer's instruction (Eurogentec, Saraing, Belgium). All unique transcripts were spotted at least twice on membranes. For each fungal species, total RNA was extracted from free-living mycelium grown on low sugar Pachlewski medium for 3 weeks. The transcript populations were amplified, labelled, and hybridized to nylon microarrays as described (Lacourt et al., 2002). Phosphorimages of hybridized membranes were analysed in XDOTREADER (Cose, Paris, France) to obtain background-subtracted raw spot intensity values.

Results and Discussion

cDNA library evaluation, cDNA sequencing and contig analysis

Plasmid DNA from 4000 clones derived from mycelial cDNA libraries of P. microcarpus 441 and L. bicolor S238N, and from three cDNA libraries of 4-, 12-, and 21-d-old P. microcarpus/E. globulus ectomycorrhizas was PCR amplified. Up to 90% of the amplified clones contained a cDNA insert. The inserts had an average size of c. 600 bp and 800 bp for L. bicolor and P. microcarpus, respectively, with a size range between 100 and 2400 bp. A total of 1519 L. bicolor clones and 1681 clones of P. microcarpus and P. microcarpus/E. globulus ectomycorrhizas were successfully sequenced from the 5′ end. We removed contaminants corresponding to mitochondrial and nuclear rRNA genes, as well as of mitochondrial DNA (L. bicolor, 19%; P. microcarpus, 12%). In addition, we have identified and eliminated E. globulus sequences (193 ESTs, 19%) from the ectomycorrhiza databases based on macroarray analyses (Voiblet et al., 2001; Courty et al., unpublished results), comparison to ESTs of P. microcarpus mycelium, and BLASTN analysis against the NCBI plant EST database. The remaining 1244 L. bicolor and 1304 P. microcarpus sequences were further organized in contigs to allow common clones to be identified. They represented 905 and 806 tentative consensus sequences (TCs), or unique transcripts, of L. bicolor and P. microcarpus, respectively (Table 1). The number of ESTs in clusters ranged between two and 75, with two and eight clusters including more than 15 ESTs (> 1% of all transcripts) in L. bicolor and P. microcarpus EST sets, respectively. Estimations suggest that the genome of filamentous fungi, including ectomycorrhizal fungi, averages 20–40 Mb, with a complement of about 8000 genes (Kupfer et al., 1997; Le Quéréet al., 2002). Using this approximation of the gene number in an ectomycorrhizal fungus, the present TC sets corresponded to 10% of the total expected complement of genes for L. bicolor and P. microcarpus. The sequence redundancy (EST in clusters/total ESTs) reached 40% for both species and implies that continued sequencing of random cDNA from our libraries still has the potential to uncover novel sequences.

Table 1.  The number of expressed sequence tags (ESTs) collected using different sampling strategies from Laccaria bicolor mycelium, Pisolithus microcarpus mycelium, and Pisolithus microcarpus/Eucalyptus globulus ectomycorrhizas
Origin of ESTsESTsSequence identity (WU-BLASTX/GenBank) of ESTs
TotalTCsaClustersSingletonsNo matchLowModerateHigh
[1e−10 < x < 1e−4][1e−20 < x ≤ 1e−10][x ≤ 1e−20]
  • a

    TCs, tentative consensus sequences (= unique transcripts).

  • b

    Only EST originating from P. microcarpus. Discrimination among plant and fungal origin of ESTs were based on macroarray hybridization intensities (Voiblet et al., 2001; P. E. Courty et al., unpublished), comparison with known ESTs of P. microcarpus mycelium, and BLASTN analysis against the GenBank Plant EST database.

  • c

    c SSH, suppression subtractive hybridization; DS, differential screening.

Laccaria mycelium, random124490515075560%11%12%17%
Pisolithus mycelium, random 550336 6327366% 9% 8%17%
Pisolithus/Eucalyptus
Mycorrhizas, totalb 754539 6747237%10%11%42%
 4 d, random 472347 3930836%10%12%42%
 4 d, SSH/DSc 129 56 18 3824%15% 8%53%
 12 d, random  77 71  3 6860% 6%14%20%
 21 d, random  76 66  6 6039% 6%12%43%
Pisolithus totalb130480614765950%10%10%30%

Functional characterization of expressed genes

To identify potential homologues to L. bicolor and P. microcarpus genes, ESTs were compared to sequences deposited in protein databases using the BLASTN and WU-BLASTX algorithm (Worley et al., 1995). A total of 497 of 1244 L. bicolor ESTs (40%) corresponded to genes with significant similarity to GenBank entries, including genes of known function as well as hypothetical proteins of other organisms. The remaining genes (60%) did not have significant matches within the GenBank databases (Table 1). Among the 1304 ESTs analysed of P. microcarpus, 647 (50%) did not show any similarity to previously identified genes within the NCBI database. The proportion of known genes in the mycelial library was 34%, whereas it was higher in the symbiotic tissues (40–64%). Unknown genes represented 50% to 65% of ESTs analysed in other filamentous fungi (Ospina-Giraldo et al., 2000; Skinner et al., 2001; Lee et al., 2002), whereas in plants, typically only 20–25% of ESTs fall into this category (Ronning et al., 2003; Kohler et al., 2003). This might reflect the low number of GenBank entries of fungal origin and emphasizes the value of the present sequencing effort. The number of proteins with known matches, found in both species and including all isoforms of the same proteins (e.g. hydrophobins, metallothioneins), was 680. A complete list of genes identified with their BLAST E-value is posted on the EctomycorrhizaDB Web site (http://mycor.nancy.inra.fr/EctomycorrhizaDB/).

In silico transcriptional profiling

Digital analysis of gene expression can be performed by counting the number of ESTs for a given gene within an EST population from which transcript abundance can be inferred (Ewing et al., 1999). This presumes that an important number (thousands) of ESTs is generated by randomly collecting cDNA clones. Because EST data is inherently noisy, formulas have been developed to estimate the statistical significance of a detected differential gene expression between different EST populations (Audic & Claverie, 1997). We performed digital analysis of the gene expression by comparing randomly sampled ESTs between the mycelia of the two ectomycorrhizal species L. bicolor and P. microcarpus, and between mycelial and ectomycorrhizal ESTs of P. microcarpus. Although the number of ESTs collected for these comparisons, ranging between c. 600 and 1200, seemed to be enough to allow digital profiling (cf. Audic & Claverie, 1997), the results may change with further sequencing.

Transcripts expressed in P. microcarpus and L. bicolor mycelium In the free-living mycelia of L. bicolor and P. microcarpus grown on Pachlewski agar medium, the majority of the most abundant transcripts, i.e. TCs with five or more sampled ESTs, were novel genes for which no function could be assigned (Table 2). Eight of the 15 most expressed genes of L. bicolor, for example, were only found in this species so far. The second most common transcript was already found in L. bicolor isolate D170 and was assigned a ras-related protein (Podila et al., 2002), but it showed low similarity to other known ras GTPases from basidiomycetes (e.g. Suillus bovinus GenBank no. AF250024; P. microcarpus, GenBank no. AF329890). Six additional ESTs (TC_Lb01B20, TC_Lb05D07, histone-like protein, TC_Lb02B23, TC_Lb24H10, and TC_Lb25B08) were expressed in both isolates S238N and D170 of L. bicolor (Podila et al., 2002; G. Podila, unpublished data). Among these 15 highly expressed transcripts, only one TC (CipC related protein) was also detected in P. microcarpus. TC_Lb02B23 showed weak amino acid sequence similarity (E-value 1.0e−11) to a Pleurotus ostreatus EST (GenBank n° AT002909; Lee et al., 2002), whereas four transcripts (TC_Lb05D07, TC_Lb21F10, CipC related protein; and a ras-related protein) were similar to Hebeloma cylindrosporum ESTs (GenBank no. BM077997, BM078018, BU964309 and BM078038). Among prominent transcripts with known function, two were involved in carbon metabolism (cytochrome C oxidase; 1,4-benzoquinone reductase) and one coded for a ribosomal protein.

Table 2.  Most abundant transcripts in Laccaria bicolor S238N free-living mycelium, in Pisolithus microcarpus 441 free-living mycelium, and in Pisolithus microcarpus/Eucalyptus globulus ectomycorrhizas as determined by clustering of expressed sequence tags (ESTs)
TCa#Protein homologue (species)Blast E-valueEST abundance (%)
L. bicolor myceliumP. microcarpus myceliumP. microcarpus mycorrhizab
  1. EST abundance = (ESTs in a cluster/total ESTs) × 100. Statistical significant (P < 0.05) differential expression based on the formula of Audic & Claverie (1997) between corresponding tissues is indicated with an asterisk (EST abundance in bold face vs EST abundance marked with an asterisk). Comparison of transcripts between L. bicolor and P. microcarpus were based on nucleotide- and amino acid-sequence similarity. aTC, tentative consensus sequence; unique transcript bEST randomly collected from all mycorrhiza libraries, and only clones originating from P. microcarpus. Discrimination among plant and fungal origin of ESTs was based on macroarray hybridization intensities (Voiblet et al., 2001; P. E. Courty et al. unpublished), comparison with known ESTs of P. microcarpus mycelium, and BLASTN analysis against the GenBank Plant EST database. cno significant similarity to GenBank entries (E-value = 1.0e−4). dTranscripts likely originating from double-stranded RNA mycovirus (Osaki et al., 2002) expressed in P. microcarpus.

 L. bicolor mycelium    
Lb01B20Hypothetical proteinc1.90.0*0.0*
Lb02A14Ras related protein (Laccaria bicolor)1.3e−561.30.0*0.0*
Lb05D07Hypothetical protein1.10.0*0.0*
Lb26D11Arginine-rich, histone-like protein (Parechinus angulosus)1.0e−061.00.0*0.0*
Lb11E01Hypothetical protein, mucin-like (Homo sapiens)1.1e−050.90.00.0
Lb24H10Hypothetical protein0.80.00.0
Lb05E05Hypothetical protein0.80.00.0
Lb02B23Hypothetical protein0.60.00.0
Lb01F03Hypothetical protein0.60.00.0
Lb10F12Hypothetical protein0.50.00.0
Lb25B08Hypothetical protein0.50.00.2
Lb21F10Hypothetical protein0.50.20.0
Lb18H08Hypothetical protein0.50.00.0
Lb21G12CipC related protein (Emericella nidulans)6.0e−140.50.00.0
Lb23D07Hypothetical protein0.40.00.0
Lb05E1040S ribosomal protein S12 (Blumeria graminis)2.8e−240.40.00.0
Lb10G01Proline-rich, LEA-like protein (Arabidopsis thaliana)9.4e−060.40.00.0
Lb17F10MAR-binding protein AHM1 (Triticum aestivum)4.3e−080.40.00.0
Lb03F15Cytochrome C oxidase subunit 1 (Agrocybe aegerita)1.4e−720.40.00.0
Lb18D021,4-Benzoquinone reductase (Phanerochaete chrysosporium)3.8e−450.40.00.0
 P. microcarpus mycelium    
ud240Hypothetical protein0.0*8.74.6
11A2Metallothionein-related protein (Agaricus bisporus)4.3e−050.0*5.51.3*
7A6Hypothetical protein0.0*3.10.3*
7A3Hypothetical protein0.0*1.61.0
P062A07SnodProt1 (Neurospora crassa)4.2e−230.0*1.50.0*
5C4Hydrophobin HydPt-3 (Pisolithus tinctorius)2.7e−530.0*1.51.0
EP1202B17RNA-dependent RNA polymerase (Helicobasidium mompa dsRNA mycovirus)d9.0e−480.0*1.30.5
P012F05Hypothetical protein0.0*1.30.3
P072E09Hypothetical protein0.0*0.90.0*
7A7Elongation factor 1-gamma (Artemia sp.)2.5e−200.1*0.70.8
P013F06Hypothetical protein0.0*0.70.0
P031A04Hypothetical protein (similar to hypothetical protein 7A3)0.0*0.70.3
P013H04Cysteine-rich protein0.0*0.70.2
P063A06Hypothetical protein0.0*0.50.3
P061B07Hypothetical protein0.0*0.50.3
P063H10Hypothetical protein0.0*0.50.0
P063G11Hypothetical protein0.0*0.50.0
P014G02Hypothetical protein0.0*0.50.2
P063G11Hypothetical protein0.00.50.0
10C5Hypothetical protein0.00.50.2
 P. microcarpus mycorrhizab    
ud240Hypothetical protein0.0*8.74.6
5A8Hydrophobin HydPt-2 (Pisolithus tinctorius)6.4e−380.0*0.0*2.6
11A2Metallothionein-related protein (Agaricus bisporus)3.5e−050.0*5.5*1.3
5C4Hydrophobin HydPt-3 (Pisolithus tinctorius)2.1e−530.0*1.51.0
EgPtdB57SRAP 172.6e−500.0*0.0*1.0
7A3Hypothetical protein0.0*1.61.0
st54Transmembrane FUN34 protein (Schizosaccharomyces pombe)1.1e−380.00.40.8
7A7Elongation factor 1-gamma (Artemia sp.)2.5e−200.1*0.70.8
8A960S ribosomal protein L10 (Saccharomyces cervisiae)2.8e−830.10.00.5
5C260S ribosomal protein L8 (Xenopus laevis)2.0e−980.10.00.5
6C8WD-repeat GTPase CPC2 protein (Neurospora crassa)2.8e−1270.2*0.20.5
5A1Ribosomal protein (Arabidopsis thaliana)6.5e−320.0*0.00.5
EP1202B17RNA-dependent RNA polymerase (Helicobasidium mompa ds RNA mycovirus)d1.0e−460.01.30.5
7C2Hydrophobin HydPt-8 (Pisolithus tinctorius)6.4e−380.00.00.3
1E9Hypothetical protein0.00.00.3
EP2102N060S ribosomal protein L242.3e−720.00.20.3
6A1Choline-P-cytidyltransferase (Brassica napus)5.2e−200.03.10.3
9E6Protein kinase (Arabidopsis thaliana)5.1e−050.00.00.3
7A6Hypothetical protein0.0  
7E3Ubiquinol cytochrome C oxidoreductase (Saccharomyces cervisiae)2.0e−090.00.3 

In P. microcarpus mycelium, several of the most abundant transcripts coded for structural proteins such as the hydrophobin HydPt-3 (GenBank no. AF097516) and the secreted SnodProt1 protein. The latter displays a strong similarity to the hydrophobin-related cerato-platanin, a phytotoxin from the ascomycete Ceratocystis fimbriata (Pazzagli et al., 1999). Other abundant TCs showed strong similarity to metallothionein-related cysteine-rich proteins, which are likely involved in metal transport, cellular detoxification and stress response (Lanfranco et al., 2002). None of these transcripts were detected in L. bicolor so far. This seemed to be the main reason for the higher proportion of cell structure and cell/organism defense proteins in P. microcarpus compared to L. bicolor mycelium (5% vs 3%) when the proportion of ESTs assigned to different functional categories were compared (Fig. 1a). The proportion of transcripts coding for ribosomal proteins and other components of the gene/protein expression machinery was strikingly different (7%, P. microcarpus; 13%, L. bicolor), which might indicate that the mycelium of L. bicolor was more active at the sampling time.

Figure 1.

Functional classification of expressed sequence tags (ESTs) (a) from mycelial cultures of Laccaria bicolor S238N and Pisolithus microcarpus 441 and (b) from free-living mycelium and symbiotic tissues of Pisolithus microcarpus. For ectomycorrhizal ESTs, randomly collected clones of various mycorrhizal libraries (Tagu & Martin, 1995; Voiblet et al., 2001) were included in the analysis.

Transcript abundance in the mycelium and symbiotic tissues of P. microcarpus Sequence redundancy analysis of randomly sampled cDNA clones from the different developmental stages indicated that several genes were differentially expressed in the free-living mycelium and symbiotic tissues of P. microcarpus (Table 2; Fig. 1b). This is in agreement with our previous studies on gene expression in P. microcarpus mycelium and ectomycorrhizas using two-dimensional PAGE (Hilbert et al., 1991), as well as suppression subtractive hybridization (SSH) and macroarray analyses (Voiblet et al., 2001). Significant differential expression was detected for two (hydrophobin HydPt-2, SRAP17) of the five most abundant transcripts found in ectomycorrhizas, which were not detected in the free-living mycelium so far; both coded for known symbiosis-regulated cell wall proteins (Martin et al., 1999). By contrast, transcripts coding for a metallothionein-related cysteine-rich protein and SnodProt1 showed a significant lower expression in mycorrhizas compared to mycelium (TC_11A2: mycorrhizas, 1.3%, mycelium, 5.5%; TC_P062A07: mycorrhizas, 0.0%; mycelium, 1.5%). These results were mirrored in the proportions of ESTs in the different functional categories (Fig. 1b). They supported our previous findings that the symbiotic interaction alters protein synthesis (Hilbert et al., 1991; Voiblet et al., 2001) and the synthesis of cell wall and extracellular matrix components in the fungal partner (Laurent et al., 1999; Voiblet et al., 2001). The up-regulation of the protein synthesis machinery and of central metabolic activities (Hampp et al., 1999, Fig. 1b) in mycorrhizas may explain why in the symbiotic fungal tissues, the proportion of transcripts with GenBank homologues was higher (mycorrhizas, 61%; mycelium, 34%).

An interesting unique transcript (TC_EP1202B17) showed high similarity to a RNA-dependent RNA polymerase (Table 2). This transcript was identified as deriving from a double-stranded RNA mycovirus expressed in the root rot fungus Helicobasidium mompa (Osaki et al., 2002). The transcripts found in P. microcarpus are therefore likely to originate from a mycovirus, which seemed to be highly expressed in both mycelial and mycorrhizal tissues.

Digital transcript profiling vs microarray analyses To assess whether the transcription level of genes, which were identified as highly expressed in EST abundance analysis, could be confirmed by microarray analysis, total RNA of free-living mycelium of L. bicolor and P. microcarpus was extracted and hybridized separately to cDNA arrays of the respective species. We compared the 20 most highly expressed TCs revealed by the two methods (Table 3). For L. bicolor, EST abundances in the two cDNA libraries constructed from free-living mycelium under different growth conditions (library 1: high sugar, grown for 3 weeks; library 2: low sugar, grown for 10 weeks) were separately analysed. The comparison revealed that five of the 20 most abundant transcripts were detected as highly expressed in all three mycelial tissues of L. bicolor, irrespective of the analysis technique (Table 3a). Similarly, eight TCs were identified by both methods to appear among the 20 most abundant transcripts in P. microcarpus mycelium (Table 3b). However, several transcripts identified as highly expressed by microarray analysis were found in only one of the two L. bicolor libraries among the 20 most abundant TCs. In addition, some TCs were detected as highly expressed in only one of the fungal mycelia studied, irrespective of the analysis method used. Overall, these data indicated on the one hand that in silico profiling was robust to detect the most abundant transcripts (i.e. = 8 ESTs). On the other hand, they suggested that results of digital profiling should be carefully interpreted, since they are subjected to both, technical (e.g. due to limited sampling of ESTs) as well as biological variation. Therefore, unless verified by independent repetition using for example microarray analysis, RNA blot, or real-time PCR, the present results should provide an initial glimpse into the gene expression in the fungal tissues studied.

Table 3.  Comparison of the 20 most abundant transcripts in Laccaria bicolor (a) and Pisolithus microcarpus (b) mycelium detected by expressed sequence tag (EST) abundance in libraries and by cDNA microarray analysis
(a)
L. bicolorMycelium library 1 (high sugar)*L. bicolorMycelium library 2 (low sugar)**L. bicolorMycelium microarray (low sugar)I*
Lb01B20Hypothetical protein19/4Lb02A14Ras related protein11/5Lb02A04Hypothetical protein100 3/0
Lb05E05Hypothetical protein 9/1Lb26D11Arginine-rich, histone-like protein10/2Lb01B20Hypothetical protein 4119/4
Lb24H10Hypothetical protein 8/2Lb05D07Hypothetical protein 8/5Lb20A08Glycoprotein precursor 33 3/1
Lb02A14Ras related protein 5/11Lb11E01Hypothetical protein, mucin-like 8/3Lb03P16Hypothetical protein 32 1/0
Lb05D07Hypothetical protein 5/8Lb25B08Hypothetical protein 6/0Lb02B23Hypothetical protein 31 3/4
Lb10G01Proline-rich, LEA-like protein 5/0Lb21F10Hypothetical protein 6/0Lb12C09Hypothetical protein 25 4/0
Lb03F15Cytochrome C oxidase subunit 1 5/0Lb21G12CipC related protein 6/0Lb01O17Hypothetical protein 23 1/0
Lb01B1160S ribosomal protein 4/0Lb01F03Hypothetical protein 5/2Lb05E05Hypothetical protein 20 9/1
Lb12C09Hypothetical protein 4/0Lb18D021,4-Benzoquinone reductase 5/0Lb05D07Hypothetical protein 19 5/8
Lb11E01Hypothetical protein, mucin-like 3/8Lb01B20Hypothetical protein 4/19Lb26D11Arginine-rich protein 17 2/10
Lb02B23Hypothetical protein 3/4Lb02B23Hypothetical protein 4/3Lb25B08Hypothetical protein 17 0/6
Lb23D07Hypothetical protein 3/2Lb10F12Hypothetical protein 4/2Lb02A14Ras related protein 14 5/11
Lb05E1040S ribosomal protein S12 3/2Lb18H08Hypothetical protein 4/2Lb18E02Hypothetical protein 14 2/0
Lb03K04Glutathione S transferase 10 3/0Lb16B04Profilins Ia/Ib 4/0Lb11A02Hypothetical protein 14 1/2
Lb07A10Transcription inititation factor TFA2 3/0Lb19G03Elongation factor IA 4/0Lb24H10Hypothetical protein 13 8/2
Lb29A12Glutathione S transferase 3/0Lb17F10MAR-binding protein AHM1 3/2Lb02A07Hypothetical protein 13 1/0
Lb01L02Hypothetical protein 3/0Lb05A07CipC related protein 3/2Lb29A10Polyphenol oxidase 12 0/1
Lb03H14Hypothetical protein 3/0Lb20A08Glycoprotein precursor 3/1Lb02H20Hypothetical protein  9 1/1
Lb02A04Hypothetical protein 3/0Lb19D11Ubiquitin fusion protein 3/0Lb15C11Hypothetical protein  9 1/1
Lb12G05Hypothetical protein 3/0Lb24H1160S ribosomal protein L41 3/0Lb01F03Hypothetical protein  9 5/2
(b)
P. microcarpusMycelium library***P. microcarpusMycelium microarrayI***
  1. Transcripts are sorted by descending order of their abundance. Gray scale indicates shared transcripts among the 20 most abundant tentative consensus sequences (TCs). * = Number of ESTs in L. bicolor mycelium library 1/library 2; **= Number of ESTs in L. bicolor mycelium library 2/library 1; *** = Number of ESTs in P. microcarpus mycelium library; I = absolute spot intensity (highest value was set at 100). atranscripts likely originating from double-stranded RNA mycovirus (Osaki et al., 2002) expressed in P. microcarpus.

ud240Hypothetical protein48EP1202B17RNA-dependent RNA polymerasea100 7
11A2Metallothionein-related protein30EP1202P07Hypothetical protein 93 0
7A6Hypothetical protein17ud240Hypothetical protein 5948
7A3Hypothetical protein 9P061F11Hypothetical protein 57 1
P062A07SnodProt1 8EgPtdD43Hypothetical protein C24c9.13c 41 1
5C4Hydrophobin HydPt-3 85C4Hydrophobin HydPt-3 29 8
EP1202B17RNA-dependent RNA polymerasea 7P031D09Hypothetical protein 28 1
P012F05Hypothetical protein 77A6Hypothetical protein 2717
P072E09Hypothetical protein 5EP1201F05Farnesyl-pyrophosphate synthetase 27 0
7A7Elongation factor 1-gamma 48E5Hypothetical protein 23 0
P013F06Hypothetical protein 4EP402E10Hypothetical protein 22 0
P031A04Hypothetical protein 411A2Metallothionein-related protein 2230
P013H04Cysteine-rich protein 3P012D0450S ribosomal protein 18 1
P063A06Hypothetical protein 3EP1202E05Hypothetical protein 17 0
P061B07Hypothetical protein 3EP1202P08EF-hand calcium binding peflin 17 0
P063H10Hypothetical protein 37A3Hypothetical protein 16 9
P063G11Hypothetical protein 3P013H04Cysteine-rich protein 15 3
P014G02Hypothetical protein 3P062A07SnodProt1 14 0
P063G11Hypothetical protein 3P012E09Beta GTPase 14 8
10C5Hypothetical protein 3st54Transmembrane FUN34 protein 13 2

Analysis of shared transcripts between tissues, isolates and species

Only a low percentage of shared transcripts (between 11% and 26%; Table 4) was detected in free-living mycelium grown under different conditions (e.g. L. bicolor grown on high vs low sugar concentration) or in different tissues (e.g. free-living mycelium vs symbiotic tissues) of the same fungal isolate. Tissue-specific expression patterns, with 30–90% of all TCs to be differentially expressed, were reported in several in silico and conventional gene expression studies of fungal and plant species (Ewing et al., 1999; Ospina-Giraldo et al., 2000; Lacourt et al., 2002; Lee et al., 2002). The reason for such a low proportion of shared transcripts found in the present study may be explained by the fact that only between 600 and 1000 ESTs of each tissue were sampled. This might not be enough to represent the transcript populations in the tissues, since the resulting 400–600 TCs corresponded to only 5% of the total number of expected genes in basidiomycetes (Le Quéréet al., 2002). However, by sequencing 6000–20 000 ESTs of a single library (with EST redundancies of 80–95%), only a maximum 20% of the expected number of genes from the respective plant (e.g. Solanum tuberosum; Ronning et al., 2003) or fungal species (e.g. Neurospora crassa; Zhu et al., 2001) was found in other EST projects. In addition, whereas the number of detected unique transcripts increased by sampling more ESTs of a library, the overall proportion of the most abundant transcripts and functional categories were similar by sampling 1000 or 6000 ESTs of poplar roots (A. Kohler, personal communication). We therefore assume that augmenting the sequencing effort will increase the number of shared transcripts between tissues, but will still reveal differential gene expression between them.

Table 4.  Comparison of tentative consensus sequences (TCs) between the same fungal tissue grown under different conditions, between different fungal tissues of the same isolate, and between different fungal isolates
 No. of TCsNo. of TCs shared determined on
nt-sequence levelaAmino acid sequence level (tBlastX; E-value ≤ 1e−10)
No.%bNo.%b
  • a

    Within the same fungal isolate: nucleotide-sequence identity ≥ 80%; between different fungal isolates: identity ≥ 60%.

  • b

    (No of shared TCs)/(smaller No of TCs of the two respective tissues) × 100.

  • c

    G20, high sugar (20 g l−1 glucose, 5 g l−1 maltose) Pachlewski medium, mycelium grown for 3 weeks; G1, low sugar (1 g l−1 glucose) Pachlewski medium, mycelium grown for 10 weeks

  • d

    Lee et al. (2002).

  • e

    Ospina-Giraldo et al. (2000).

  • f

    ESTs of L. bicolor DR170: Podila et al. (2002) and G. Podila (unpublished data).

  • g

    g ESTs derived from GenBank, release 121302, no corresponding publication.

Laccaria bicolor S238N (mycelium, G20) vs L. bicolor S238N (mycelium, G1)c416 vs 4286115 6917
Pisolithus microcarpus (mycelium) vs P. microcarpus (mycorrhizas)336 vs 5396519 6620
Pleurotus ostreatus (mycelium) vs P. ostreatus (fruitbody)d650 vs 6526610 7111
Agaricus bisporus (primordia) vs A. bisporus (fruitbody)e245 vs 2895723 6326
Laccaria bicolor S238N (mycelium) vs L. bicolor DR170 (mycelium)f905 vs 2472811 4518
Laccaria bicolor S238N vs Pisolithus microcarpus905 vs 80637 5 8711
Laccaria bicolor S238N vs Pleurotus ostreatus905 vs 125674 812614
Laccaria bicolor S238N vs Agaricus bisporus905 vs 47726 5 27 6
Laccaria bicolor S238N vs Hebeloma cylindrosporumg905 vs 2682710 3212
Pisolithus microcarpus vs Pleurotus ostreatus806 vs 125654 712916
Pisolithus microcarpus vs Hebeloma cylindrosporum806 vs 26821 8 2911
Pisolithus microcarpus vs Agaricus bisporus806 vs 47728 6 35 7

At the interspecies level, the percentage of shared TCs between L. bicolor and P. microcarpus tissues amounted to 5% based on nucleotide sequence similarity and to 11% based on translated protein sequence similarity. The transcripts exhibiting the highest nucleotide sequence similarity between these two species coded for genes involved in gene/protein expression, such as ribosomal proteins (Table 5). Similar percentages of shared TCs were observed when ESTs of other basidiomycetous fungi, retrieved from GenBank, were compared to the current sets of ESTs. At the protein sequence level, 6% (L. bicolor vs Agaricus bisporus) to 16% (P. microcarpus vs Pleurotus ostreatus) of the transcripts were shared by two species. As mentioned above, the high percentage of ESTs without similarity to sequences of other fungal species might result from the limited number of ESTs available so far. Furthermore, since the libraries were constructed from various fungal tissues and from mycelium grown under different conditions, we assume that also the set of expressed transcripts differed at least partially. To obtain a more accurate idea of gene diversity in basidiomycetes, sequencing complete genomes of basidiomycetous fungi would be necessary. Comparing 2555 TCs of Aspergillus fumigatus (30% of the total number of predicted genes) to other ascomycetes whose genomes were completely or almost completely sequenced, such as Saccharomyces cerevisiae, Schizosaccharomyces pombe and Candida albicans,Kessler et al. (2002) found only 30–40% of the TCs to have a homologue in these fungi. By contrast, up to 70% of tomato and poplar ESTs showed a match with Arabidopsis genes (Van der Hoeven et al., 2002; Kohler et al., 2003). Tomato and Arabidopsis have diverged around 150 million years ago, which would be in the same range as the divergence of Pisolithus and Laccaria (Bruns et al., 1998). Whether the percentage of homologous genes between L. bicolor and P. microcarpus is similar to the one observed between plant species can only be answered with an additional sequencing effort.

Table 5.  Nucleotide sequence identity (≥ 60%) of 37 tentative consensus sequences (TCs) expressed in both basidiomycetous fungi Laccaria bicolor and Pisolithus microcarpus
Identity (%)nt-Seq overlap (bp)Protein homolog (GenBank)TC #Functional categorya
EMCSFNI
  • a

    E, gene/protein expression; M, metabolism; C, communication/signaling; S, cell structure; F, cell fate; NI, no identification.

7829040S ribosomal protein S3ae-aLb03E23, EgPtdB53x     
7842040S ribosomal protein S28Lb25A02, P064A05x     
7732060S ribosomal protein L10Lb13H09, 7A9x     
73410GTP-binding protein, beta subunitLb16E08, EgPtdA23  x   
7333140S ribosomal protein S11Lb18B07, 10C8x     
7243060S ribosomal protein L26Lb23F05, 9A5x     
7237060S ribosomal protein L27Lb21H04, P052D04x     
7235060S ribosomal protein L20Lb03O03, EP1202K01x     
72320Ubiquitin conjugating enzyme E2Lb02E18, EP2102F13x     
71380Zinc metalloproteaseLb27H04, EP402A10  x   
7140060S ribosomal protein L35Lb16G04, 7C3x     
7143040S ribosomal protein S9-bLb24C03, 9E3x     
71330Glutamine synthetaseLb23E06, P011H04 x    
70400ATP/ADP carrier proteinLb01B22, 10D4  x   
70 9060S ribosomal protein L2Lb01L23, EP1202M06x     
7029060S ribosomal protein L18Lb13G02, EgPtdD5x     
69400Peptidyl-prolyl cis/trans isomeraseLb11E03, P072G09x     
69300Glyceraldehyde 3-P dehydrogenaseLb22G04, 6D5 x    
68300Mitogen-activated protein kinaseLb16B01, P062C01  x   
6838060S ribosomal protein L44Lb08F01, EP1201H01x     
68180Adenylate kinase bLb19D05, P021A08 x    
67550Rho GDP-dissociation inhibitor 2Lb01A06, EP402N08  x   
67600Arp 2/3 complexLb31D04, EP2102D05   x  
67390Ubiquitin fusion proteinLb19D11, 12C3x     
67520Transcription initiation factor IIaLb06D07, P062G03x     
66550Symbiosis-related proteinLb29B12, P031F04    x 
6535040S ribosomal protein S30Lb22E02, EgPtdB47x     
64220Elongation factor 1-gammaLb15B12, P021F02x     
64348Transcriptional regulatorLb01H19, EP2102H20x     
64380Hypothetical proteinLb16D04, P063F06     x
63200Small GTPase RAC1Lb20F12, P021D07  x   
61300Mitochondrial acyl carrier proteinLb02C21, EgPtdD19x     
61420CipC related proteinLb21G12, P011G10     x
61420Ubiquitin-carboxy extension protein fusionLb08C08, 3C5x     

Conclusion

The present data provided a first global overview of the gene diversity expressed in the mycelium of two model ectomycorrhizal fungi, P. microcarpus and L. bicolor. EST collections obtained from L. bicolor and P. microcarpus by extraction of RNA from tissues exposed to different developmental and physiological conditions resulted in a considerable variability in the most highly expressed transcripts. Therefore, the analysis of ESTs not only of various ectomycorrhizal species, but also of different tissues and tissues subjected to different growth conditions is an efficient tool to discover novel genes of these ecologically and economically important fungi. The study confirmed that there is a lack of genomic information of basidiomycetous, in particular ectomycorrhizal fungi and revealed the necessity for an additional sequencing effort. To our knowledge, there are several ongoing EST projects of ectomycorrhizal basidiomycetes (e.g. Paxillus involutus, Hebeloma cylindrosporum), which will release large sets of ESTs (3000–5000) within the next months (Tunlid et al. and Sentenac et al. personal communications). Therefore, we are positive that in the near future, more exhaustive comparative analyses will help to complete the picture of the gene diversity expressed in these fungal species. In the present sets of ESTs, several abundant TCs, such as families of cell wall components, hydrophobins and SRAPs, were already characterized (Laurent et al., 1999; Martin et al., 1999), but many yet unknown proteins await their functional delineation. The EST databases of L. bicolor and P. microcarpus contain many genes, such as transporters, assimilating enzymes and transcriptional factors, which hold great promise to elucidate the nutrient uptake and assimilation pathways in both, the free-living and symbiotic phases of these ectomycorrhizal fungi.

Acknowledgements

Martina Peter was supported by a postdoctoral fellowship from the French Ministry of Foreign Affairs. Annegret Kohler was supported by postdoctoral fellowships from the INRA and the Région de Lorraine. The present investigation was supported by grants from the INRA (Programmes ‘Sequencing genomes of symbionts and pathogens’ and LIGNOME). The research utilized in part the DNA Sequencing Facilities at INRA-Nancy financed by the INRA, Région de Lorraine and the European Commission. We would like to thank two unknown referees for their constructive comments on the manuscript.

Ancillary