Microbiol. Res. (2001) 156, 59–63 http://www.urbanfischer.de/journals/microbiolres
HCf-6, a novel class II hydrophobin from Cladosporium fulvum Peter S. Nielsen1, Anthony J. Clark2, Richard P. Oliver2, Martina Huber3, Pietro D. Spanu4 Carlsberg Laboratory, 1Department of Yeast Genetics, 2Department of Physiology, Gamle Carlsberg Vej 10, DK-2500 Copenhagen Valby, Denmark 3 Department of Plant Sciences, University of Oxford, South Parks Rd, Oxford, OXI 3RB, UK 4 Department Biology, Imperial College of Science Technology and Medicine, Imperial College Road, London, SW7 2AZ, UK Accepted: September 26, 2000
Abstract C. fulvum, a fungal tomato pathogen, has previously been shown to express a complex family of hydrophobin genes including four class I hydrophobins and one class II hydrophobin. Here we describe a gene for HCf-6, a sixth member of the hydrophobin family and the second class II gene. The protein is predicted to consist of a signal sequence, an N-terminus rich in glycine and asparagine and a C-terminal hydrophobic domain which bears the hall-marks of hydrophobins. In contrast to the previously described class II hydrophobin HCf-5, HCf-6 is expressed in mycelium growing in pure culture and mRNA levels do not increase during sporulation. It is downregulated by carbon starvation but not by depletion of nitrogen in the growth medium. Key words: Hydrophobins – Cladosporium fulvum – sporulation – glycine/asparagine-rich – tomato leaf mould
Introduction Hydrophobins are small amphipathic proteins which have been found in many Ascomycetes, Basidiomycetes and Zygomycetes. They have been found in many species of filamentous fungi, but are absent in yeast. A functionally similar class of proteins, called repellents, have been found in U. maydis (Wösten et al. 1996). Hydrophobins are relatively small proteins (around 100 amino acids long), with eight cysteines in a conserved array. Two distinct classes have been identified and, although size and the cysteine arrays are similar, there is little or no other sequence similarity between the clasCorresponding author: P. D. Spanu e-mail:
[email protected] 0944-5013/01/156/01-59
$15.00/0
ses. The hydropathy plots are distinct between classes but highly conserved within each class. No class II hydrophobin has yet been found in Basidiomycetes, but both classes are found in Ascomycetes. Hydrophobins are thought to be cell-wall proteins that form a layer on the outermost side of the hyphae and are therefore likely to play important roles in environmental recognition (Wessels 1996–1997). In line with this, Sc3, one of the best studied hydrophobins of Schizophyllum commune, is needed for formation of aerial hyphae and for emergence of hyphae through water-air interfaces (Wösten et al. 1999). Another classic example of a hydrophobin is Mpg1 of Magnaporthe grisea, the causal agent of rice blast. Appropriate formation of the appressoria and penetration requires surface perception mediated via Mpg1 ( Talbot et al. 1996). We have recently described five hydrophobins, named HCf-1 to HCf-5, from Cladosporium fulvum, the causal agent of tomato leaf mould. HCf-1, -2, -3 and -4 are similar to each other and are typical class I hydrophobins. HCf-5 is distinct and is a class II hydrophobin. The expression of all hydrophobins appears to be differentially regulated during development and growth (Spanu 1997; Segers et al. 1999). Initial experiments showed that HCf-1, the most abundant hydrophobin in this fungus, is not necessary for pathogenicity on tomato (Spanu 1998); however recent results appear to indicate that this protein could play an important role in dispersal mediated by water droplets (Spanu and Whiteford 2000). In this paper we report the discovery of HCf-6, a novel class II hydrophobin gene in C. fulvum, and describe its structure and expression in comparison to HCf-5. Microbiol. Res. 156 (2001) 1
59
Materials and methods Fungal isolates and culture conditions.C. fulvum race 4 was maintained on V8 or Potato Dextrose agar, as described (Harling et al. 1988 ; Spanu 1997). Mycelium for DNA and RNA extractions was grown in liquid B5 medium supplemented with 20 g/l sucrose, at 20°C on a rotary shaker (130 rpm). RNA was extracted from C. fulvum grown in different nutritional conditions as described in Segers et al. (1999). The yeast strain used for complementation M4271 (MATa leu2-3 leu2-112 ura3-1 trp1-1 his3-11 his3-15 ade2-1 met6 ::kanMX can1-100 GAL SUC2 mal0) was grown on synthetic complete, SC, medium (ammoniumbased medium containing 14 different amino acids, adenine, uracil and 2% glucose). For transformation (see below) the following media were used : SC-his-ura or SC-met-his-ura is SC without histidine and uracil or without methionine, histidine and uracil, respectively; 5 × Ade and 5 × Leu indicates 5 times normal SC concentration of adenine and leucine, respectively. Yeast transformation and plasmid extraction. Yeast transformation was performed according to Becker and Guarente (1991). For each electroporation, 100 ng of C. fulvum cDNA library plasmid DNA was mixed with 40 µl of competent yeast cells, after electroporation 500 µl of 1 M sorbitol was added. Approximately 100 µl of this mixture was plated on one SC-his-ura 5 × ade5 × leu plate and incubated at 30 °C. On day two, Ura+ colonies (1.3 × 105) were replica plated to SC-met-his-ura 5 × ade 5 × leu + 10 mM betaine. Plasmid DNA isolated from growing yeast colonies essentially as described by Hoffman and Winston (1987) was introduced into electro-competent E. coli (DH5α) cells according to standard procedures. In the first round of a functional screen, we isolated a number of C. fulvum cDNAs that complemented a met6 deletion in yeast. The isolated plasmid, however, did not confer the ability to grow on plates without methionine when it was reintroduced into the yeast strain M4271. The nature of the apparent complementation has not been explained. The complemented yeast strain showed no atypical morphology and an attempt to show aggregation between the cells proved negative. Nucleic acid manipulations. DNA manipulations were carried out using standard procedures (Sambrook et al. 1987) unless stated otherwise. C. fulvum cDNA inserts were sequenced using terminator cycle sequencing (Perkin-Elmer) and analysed on a 310 or a 377 ABI Prism DNA sequencer following the manufacturer’s instructions. Genomic clones containing HCf-5 and HCf-6 were isolated from a cosmid library of C. fulvum DNA described previously (Spanu 1997) using gene-specific PCR probes. The PCR probes were generated on cDNA 60
Microbiol. Res. 156 (2001) 1
templates using the following primer pairs: HCf-5 (sense) CGG TGT CCT TGA CTT GAC T; HCf-5 (antisense) GGT GTA TGT ATA TGC ACC TG; HCf-6 (sense) AGC CAA CCG CGT TCC TCA; HCf-6 (antisense) CAA GGC ATA CAC TAT GAC GC. PCR was performed on 10 ng template DNA in a final volume of 100 µl using 2.5 U TaqDNA polymerase (Life Technologies) containing 0.2 mM of each dNTP (Pharmacia) and 0.3 µM of each primer, following the manufacturer’s instructions, using a PTC-100 (MJ Research, Inc) thermal cycler. The cycling conditions were: 45’’ at 94°C, 30’’ at 50°C, 1’30’’ at 72°C and 24 cycles. The PCR products were purified by gel electrophoresis in a 1.5% TBE/agarose gel and extracted using QIAEX (Qiagen). RNA was isolated from mycelium and spores by grinding the tissues in liquid nitrogen, extracting in guanidinium thiocyanate/phenol (Chomczynsky and Sacchi 1987) followed by purification using the RNeasy kit (QIAGEN). RNA hybridisation analysis was carried out as described (Spanu 1997). cDNA library construction. Total RNA was prepared from C. fulvum grown in Erlenmeyer flasks with full B5 media for 5 days and then transferred for 16 h to B5 media containing individual amino acids as the sole nitrogen source. PolyA+ RNA was purified and 5 µg used to synthesize cDNA (Stratagene, La Jolla USA, catalog no. 200401) using an oligo-dT primer including a XhoI site and a reverse primer including an EcoR1 site. The cDNA was directionally cloned into the EcoR I–XhoI sites of the yeast expression vector pYPGE15 (Brunelli and Pall 1993) in which the constitutive PGK promoter is used to drive expression of the inserted sequence. The ligated DNA was used to transform E. coli strain XLI-Blue MRF'. 350,000 individual colonies were obtained. Plasmid DNA was prepared by washing the colonies from the plate in terrific broth and inoculating 4 l terrific broth containing 50 mg/l carbenicillin. Cultures were shaken for 8 h at 37°C, 4 mg of plasmid DNA was recovered and purified by CsCl equilibrium gradient centrifugation. Analysis of predicted protein sequences. Prediction of signal peptide cleavage sites used the programme SignalP (www.cbs.dtu.dk/services). Sequences were aligned using CLUSTAL_X (Thompson et al. 1997) and then adjusted by eye to align the cysteine residues. Phylogenetic analysis of the hydrophobin sequences was carried out using maximum parsimony and neighbour-joining programmes in the PHYLIP package 3.5c (Felsenstein 1993). Hydropathy plots were generated using the GCG package using the Kyte and Doolittle (1982) algorithm. Sequences. The genomic and cDNA sequences of HCf-5 and HCf-6 have been deposited in the databases (accession numbers: AJ251294 and AJ251295).
Results Several C. fulvum cDNAs complemented a met6 deletion in yeast. Sequencing the inserts revealed that one of them encoded a predicted protein with the features of a hydrophobin. The predicted product of translation is a 184-amino-acid polypeptide (Fig. 1). The protein has an N-terminus which is suggestive of a signal sequence and is predicted to be cleaved after Ala16. This would produce a protein with two evident domains. In the N-terminal domain there is 33-amino-acid region with 7 prolines followed by a 67-amino-acid region which is rich (92.5%) in glycine and asparagine (G-N rich). The C-terminal domain contains 8 cysteines in a C-X7-CCX11-C-X16-C-X8-CC-X10-C-X3 array and is 57% identical to the equivalent region of HCf-5, the previously described class II hydrophobin from C. fulvum (Segers et al. 1999). A cosmid genomic library of C. fulvum was screened using a PCR probe generated from the 3’ end of the cDNA. Sequence analysis of this clone and comparison with the cDNA of HCf-6 and HCf-5 and with the genomic sequences of HCf-6 and HCf-5 identified 3 and 2 introns, respectively, in the two genes. Two of the introns (II and III) are in conserved positions and are positioned in codons 146 and 175 of HCf-6. HCf-6 also has an intron (I) at codon 99 i.e. in the G-N rich domain. Computer-generated hydropathy plots (Fig. 2) show that HCf-6 and HCf-5 have C-termini which have broadly similar profiles, typical of class II hydrophobins, and are distinct from the class I hydrophobins (HCf-1 is shown here as an example). In particular the cysteine doublets are both followed by a stretch of hydrophobic amino acids. The N-terminal domain of the predicted mature polypeptide is predominantly hydrophilic. The sequences of the “hydrophobin core” (defined by the sequence following the first cysteine residue) were
Fig. 1. Protein sequence alignment of HCf-5 and HCf-6. Conserved amino acids are enclosed in boxes. The predicted cleavage sites of the signal sequences are shown by the black arrowhead and the position of the introns are shown by the white arrowheads.
compared to those from other hydrophobins in the database and the results were plotted as a phylogram to reflect sequence relatedness (Fig. 3). This analysis shows that the hydrophobins of C. fulvum within each class are more similar to each other than they are to those of any other organisms.
Fig. 2. Hydropathy plots of HCf-1, HCf-5 and HCf-6. The plots were calculated according to Kyte and Doolittle (1982). Hydrophobic regions are shown above the X-axis and hydrophilic regions below. The positions of the cysteine residues are indicated by the vertical bars.
Fig. 3. A non-rooted phylogenetic tree showing relationships between the hydrophobins from different fungi. Accessions numbers hcf-1, CAA67187; hcf-2,3,4,5; AJ133700-3; Hcf6, AJ251294, eas, X67339; rodA, M61113; rodaaf, U06121; cu, Z80081; cryp, L09559; hfb1, Z68124; hfb2, Y11894; srh1, Y11841; qid3, X71913; sc1, JU0321; sc6, AJ7504; hyp1pt, JC4608; hypb, Y15941; poh1,2,3, Y14656,7,8; hyp1, L25258; hyp2, AAC49308; dewA, U7935; mpg1, L20865; ssgA, M85281; hyp2pt, JC4608; coh1,2, CAA71652,3; sc3, P16933; Microbiol. Res. 156 (2001) 1
61
Fig. 4. RNA blot analysis of hydrophobin transcript levels in C. fulvum. A) RNA was extracted from mycelium grown under different nutritional regimes. The fungi were grown for 48 hours in full B5 medium. The mycelia were then harvested and transferred to: full B5 medium (+C+N), medium lacking carbon (-C+N), medium lacking nitrogen (+C-N) and medium lacking both carbon and nitrogen (-C-N) and incubated for a further 16 hours. B) RNA was extracted from non-sporulating mycelium and sporulating mycelium grown on solid, complete medium and from spores. Three µg of total RNA was loaded.
The expression of HCf-6 mRNA in response to carbon and/or nitrogen starvation was investigated by RNA blot analysis and compared to that of HCf-1 and HCf-5 (Fig. 4A). HCf-6 transcript is abundant in mycelium growing in complete medium. Removal of carbon results in a dramatic reduction in the steady-state levels of HCf-6 mRNA by 18 hours after transfer to carbon-less medium. Removal of nitrogen does not greatly affect HCf-6 mRNA accumulation nor does it interfere with the effect of carbon starvation. In contrast, and as found earlier (Segers et al. 1999), HCf-5 transcript levels are very low during growth in full medium and increase greatly following nitrogen starvation. HCf-1 mRNA levels are not greatly affected by carbon or nitrogen starvation. HCf-6 mRNA appears to decrease during sporulation (Fig. 4B) and there is very little, if any, in the RNA extracted from conidia whereas HCf-5 mRNA increases noticeably upon sporulation and is present at appreciable levels in the conidia.
Discussion In this study we report on the serendipitous discovery of a hydrophobin from C. fulvum. We have named the hydrophobin HCf-6 as it is the sixth hydrophobin de62
Microbiol. Res. 156 (2001) 1
scribed in this fungus thus far. HCf-6 shares a number of common features with other hydrophobins: it has a predicted signal sequence at the N-terminus and a C-terminal domain with 8-cysteines arranged in an array that is typical of fungal hydrophobins (Wessels 1994). The C-terminal domain shows distinct similarities to other class II hydrophobins and shares the greatest identity with HCf-5 (Segers et al. 1995), the other class II hydrophobin of C. fulvum. The N-terminus of the predicted mature polypeptide is extremely rich in glycine and asparagine (G-N-rich domain). There are reports of two other fungal hydrophobins which have such a G-N-rich domain: Qid3 of Trichoderma harzianum (Lora et al. 1995) and CFTH1, the trihydrophobin of Claviceps fusiformis (de Vries et al. 1999), both of which are class II hydrophobins. The function of this domain is still unclear, but computer algorithms predict that the G-Nrich domain of CFTH1 may be arranged in an amphipathic helix (de Vries et al. 1999), and glycine-rich proteins have been observed in the cell walls of plants (Cassab 1998). It is therefore conceivable that these domains play a role in the cell wall localisation. Indeed CFTH1 appears to be an extrinsic cell wall protein (de Vries et al. 1999). So far we have no indication of the localisation of HCf-6. The simple hydrophobins that do not possess the G-N-rich domain are thought to assemble on cell wall surfaces (Wösten et al. 1999). It is tempting to speculate that the G-N hydrophobins may play a role in mediating the appropriate anchoring of hydrophobin complexes to the surface of the cell wall carbohydrate matrix. Future experiments may elucidate if this is indeed the case. The analysis of the sequences of HCf-5 and -6 and the comparison between the cDNAs and genomic sequences show the presence of two and three introns, respectively, in the two genes. Two of these introns are in exactly the same position in both genes. This suggests that HCf-5 and -6 are likely to be evolutionarily related, probably arising from a common ancestral class II hydrophobin. Studies on the expression of the C. fulvum hydrophobins have shown that they are differentially expressed during development and under diverse nutritional conditions. Here we find that HCf-6 mRNA is abundantly present in mycelium growing in full liquid medium, unlike HCf-5 mRNA. When carbon is removed from the medium there is a very strong reduction in HCf-6 mRNA abundance, whereas nitrogen starvation has little or no effect. The levels of HCf-6 mRNA during sporulation are also greatly reduced in marked contrast to those of HCf-5, which accumulates noticeably in sporulating mycelium and in the spores. It therefore seems that HCf-6 expression is under control of very different regulatory elements and has contrasting developmental roles compared to HCf-5.
In conclusion, we have identified a novel class II bimodular hydrophobin of C. fulvum, which possesses a G-N-rich domain of as yet unknown function. The expression of HCf-6 suggests that its function is likely to be markedly different from that of its most similar homologue, HCf-5. The discovery of HCf-6 increases still further the complexity of the C. fulvum hydrophobin gene family. Future experiments will aim at elucidating the localisation of the hydrophobins and at analysing the phenotypes of mutants which lack the class II hydrophobins in C. fulvum in the hope that some light will be shed on the function of these proteins.
Acknowledgements The support of the Royal Society (to PDS) and of Føtek II (to PSN) is gratefully acknowledged.
References Becker, D. M., Guarente, L. (1991): High efficiency transformation of yeast by electroporation. Methods Enzymol. 194, 182–187. Brunelli, J. P., Pall, M. P. (1993): A series of yeast shuttle vectors for expression of cDNAs and other DNA sequences. Yeast 9, 1299–1308. Cassab, G. I. (1998): Plant cell wall proteins. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 281–309 Chomczynsky, P., Sacchi, N. (1987): Single-step method of RNA isolation by acid guanidinium isothiocyanate phenol chloroform extraction. Anal. Biochem. 162, 156–159. de Vries, O. M. H., Moore S., Arntz, C., Wessels, J. G. H., Tudzynski, P. (1999): Indentification and characterisation of a tri-partite hydrophobin from Claviceps fusiformis: a novel type of class II hydrophobins. Eur. J. Biochem. 262, 377–385. Felsenstein J. (1993) PHYLIP (Phylogeny Inference Package) version 3.5c. Department of Genetics, University of Washington, Seattle, WA. Harling, R., Kenyon, L., Lewis, B. G., Oliver, R. P., Turner, J. G., Coddington, A. (1988): Conditions for efficient isolation and regeneration of protoplasts from Fulvia fulva. J. Phytopathol. 122, 143–146 Hoffmann, C. S., Winston, F. (1987): A ten-minute DNA preparation from yeast efficiently releases autonomous
plasmids for transformation of Escherichia coli. Gene 57, 267–272. Kershaw, M. J., Talbot, N. J. (1998): Hydrophobins and repellents: proteins with fundamental roles in fungal morphogenesis. Fungal Genet. Biol. 23, 18–33. Kyte, J., Doolittle, R. F. (1982): A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132. Lora, J. M., Pintor-Toro, J. A., Benitez, T., Romero, L. C. (1995): Qid3 protein links plant bimodular proteins with fungal hydrophobins. Mol. Microbiol. 18, 377–382. Sambrook, J., Fritsch, E. F., Maniatis, T. (1987): Molecular cloning. Cold Spring Harbour Laboratory Press, New York. Segers, G. C., Hamada, W., Oliver, R. P., Spanu, P. D. (1999) Isolation and characterisation of five different hydrophobinencoding cDNAs from the fungal tomato pathogen Cladosporium fulvum. Mol. Gen. Genet. 261, 644–652. Spanu, P. (1997): HCf-1, a hydrophobin from the tomato pathogen Cladosporium fulvum. Gene 193, 89–96. Spanu, P. (1998): Deletion of HCf-1, a hydrophobin gene of Cladosporium fulvum, does not affect pathogenicity in tomato. Physiol. Mol. Plant Pathol. 52, 323–334. Spanu, P. D. Whiteford, J. (2000) Fungal Raincoats as Dispersal Aids. Abstract book, p140, ECFG 5, eds. Felenbook, B., Turcq, B. Talbot, N. J., Kershaw, M. J. Wakley, G. E., de Vries, O. M. H., Wessels, J. G. H., Hamer, J. E. (1996): MPG1 encodes a fungal hydrophobin involved in surface interactions during infection-related development of Magnaporthe grisea. Plant Cell 8, 985–999. Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F., Higgins, D. G. (1997) The ClustalX Windows interface; flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. Wessels J. G. H. (1994): Developmental regulation of fungal cell wall formation. Annu. Rev. Phytopathol. 32, 413–437. Wessels J. G. H. (1996): Fungal hydrophobins: proteins that function at an interface. Trends Plant Sci. 1, 9–15. Wessels J. G. H. (1997): Hydrophobins: proteins that change the nature of the fungal surface. Adv. Microbiol. Physiol. 38, 10–45. Wösten, H. A. B., Bohlmann, R., Eckerskorn, C., Lottspeich, F., Bölker, M., Kahmann, R. (1996): A novel class of small amphipathic peptides affect aerial hyphal growth an surface hydrophobicity in Ustilago maydis. EMBO J. 15, 4274–4281 Wösten, H. A. B., van Wetter, M. A., Lugones, L. G., van der Mei, H. C., Busscher, H. J., Wessels, J. G. H. (1999): How a fungus escapes the water to grow into the air. Curr. Biol. 9, 85–88.
Microbiol. Res. 156 (2001) 1
63
