Mutations to LmIFRD affect cell wall integrity, development and pathogenicity of the ascomycete Leptosphaeria maculans

Fungal Genetics and Biology 46 (2009) 695–706

Contents lists available at ScienceDirect

Fungal Genetics and Biology journal homepage: www.elsevier.com/locate/yfgbi

Mutations to LmIFRD affect cell wall integrity, development and pathogenicity of the ascomycete Leptosphaeria maculans Angela P. Van de Wouw, Filomena A. Pettolino, Barbara J. Howlett, Candace E. Elliott * School of Botany, The University of Melbourne, Vic. 3010, Australia

a r t i c l e

i n f o

Article history: Received 1 April 2009 Accepted 10 June 2009 Available online 17 June 2009 Keywords: Fungi Cell wall Monosaccharide Chitin Leptosphaeria maculans

a b s t r a c t Maintaining cell wall integrity is essential for fungal growth and development. We describe two mutants with altered expression of a gene, LmIFRD, from the ascomycete Leptosphaeria maculans. Truncation of the LmIFRD transcript in a T-DNA insertional mutant led to slower germination, less sporulation and loss-ofpathogenicity towards Brassica napus, whereas silencing of the LmIFRD transcript led to increased germination, sporulation and earlier infection. The increased tolerance to cell wall lysing enzymes and cell wall-disrupting compounds of the T-DNA mutant contrasts with decreased tolerance of the silenced mutant and suggests altered cell wall integrity and accessibility to 1,3-linked glucan and chitin. Lectin binding experiments and monosaccharide analysis revealed altered polysaccharide content and structure within the cell wall of the LmIFRD mutants, notably increased 1,3-linked galactose and chitin within the cell wall of the T-DNA mutant. This is the first analysis of monosaccharide linkage composition of cell walls of spores and mycelia for any dothideomycete. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Cell walls provide structural support as well as mediating exchanges between the cell and its environment. Despite its importance, little is known about the composition and structure of fungal cell walls with the exception of a few euascomycetes including Aspergillus fumigatus and Penicillium janczewskii and the hemiascomycete Saccharomyces cerevisiae (Latgé, 2007; Pessoni et al., 2005). Although the fungal cell wall varies in composition and structure between different taxonomic groups, glucose comprises between 50% and 60% of the total monosaccharides (Guest and Momany, 2000; Pessoni et al., 2005). The stability or integrity of the cell wall is achieved via various cross-linkages between polysaccharides, including glucan, chitin and mannan, and cell wall associated glycoproteins, some of which contain a glycosylphosphatidylinositol (GPI) anchor (for review see Bowman and Free (2006), Latgé (2007)). Mutations in genes encoding polysaccharide synthases, particular glycoproteins, transcription factors, and mitogen activated protein (MAP) kinases can result in changes in cell wall integrity due to altered cell wall content or assembly of the various components. This can affect hyphal growth and development, production of asexual/sexual fruiting bodies and pathogenicity (Odenbach et al., 2007; Valiante et al., 2008; Werner et al., 2007).

* Corresponding author. Fax: +61 3 9347 5460. E-mail address: [email protected] (C.E. Elliott). 1087-1845/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.fgb.2009.06.004

Mutants with altered cell wall integrity can be identified by screening for changes in tolerance towards cell wall-disrupting compounds such as Congo Red (CR) and Calcofluor White (CFW) (Ram and Klis, 2006). These dyes have affinity for fibrillar molecules and they can interact specifically with polysaccharides that have b1-4 and b1-3 linkages such as chitin and cellulose (Kopecka and Gabriel, 1992; Ram et al., 1994; Wood, 1980). However, the precise molecular basis of the dye-glucan interaction remains unclear (Ram and Klis, 2006). Although changes in sensitivity to cell wall-disrupting compounds indicate altered cell wall integrity, a range of mutations can lead to this phenotype. For instance, a large scale study of S. cerevisiae mutants with altered sensitivity to CFW revealed genes related to cell wall assembly as well as genes not previously associated with cell wall integrity or ones with unknown function (Lussier et al., 1997). In the pathogenic ascomycete Fusarium oxysporum, isolates mutated in Rho1, encoding a Rhotype GTPase protein, displayed dramatically restricted growth in the presence of CR. The mutations to Rho1 led to increased and decreased activity of chitin and glucan polysaccharide synthases, respectively, which altered cell wall integrity and reduced virulence of this fungus towards tomato plants (Martinez-Rocha et al., 2008). Changes in cell wall integrity can affect the development of important penetration structures such as appressoria, which are essential for some fungi to invade plants. In Magnaporthe oryzae, mutations to a MAP kinase kinase kinase, MCK1, altered the integrity of the appressorial cell wall, presumably affecting the level of turgor pressure, thus reducing the ability of the pathogen to penetrate the cells of rice plants (Jeon et al., 2008). The dothideomycete,

696

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

Leptosphaeria maculans is a pathogen of oil seed rape, Brassica napus, causing significant yield losses annually (Fitt et al., 2006). Unlike pathogens such as M. oryzae, L. maculans does not use specialized infection structures such as appressoria. Instead, germ tubes of asexual and sexual spores gain entry to B. napus plant tissue via stomata or wounds (Hammond and Lewis, 1987). The role of cell wall integrity in virulence of L. maculans towards B. napus is unknown. In this paper we describe mutations of a particular gene, LmIFRD, that affect germination, production of asexual fruiting bodies and virulence of L. maculans towards B. napus. Additionally, mutations to LmIFRD affect monosaccharide content and linkage within the cell walls of germinating spores and mycelia, and are associated with changes in tolerance towards cell wall lysing enzymes and cell wall-disrupting compounds.

2. Materials and methods 2.1. Fungal isolates, crossing and culturing L. maculans isolates were maintained on 10% Campbell’s V8 juice (Australia) agar at 22 °C with a 12 h photoperiod. L. maculans wild type isolate IBCN18 was transformed with plasmid pPK2 (Covert et al., 2001) using Agrobacterium tumefaciens-mediated transformation (Gardiner and Howlett, 2004). Plasmid pPK2 contains a gene encoding hygromycin resistance under transcriptional control of an Aspergillus nidulans glyceraldehyde-3-phosphate dehydrogenase promoter flanked by the right and left border transfer DNA (T-DNA) sequences of A. tumefaciens (Covert et al., 2001). This plasmid was randomly integrated into the L. maculans genome resulting in a bank of insertional mutants. One mutant (A9, subsequently referred to as the T-DNA mutant) showed reduced virulence on cotyledons of B. napus (Fig. 4) and was characterized further. Crosses were performed between isolate Lm691 and the T-DNA mutant, which harbor different idiomorphs, MAT1-1 and MAT1-2, respectively, at the mating type (MAT) locus, (Cozijnsen and Howlett, 2003). Random ascospore progeny were recovered as described previously (Cozijnsen et al., 2000). 2.2. Pathogenicity assays Seventy L. maculans random T-DNA insertional mutants were screened for their ability to infect cotyledons of 14-day old

seedlings of B. napus cv. Monty (a moderately susceptible variety) that had been wounded with a 26-gauge needle, as described previously (Purwantara et al., 1998). Lesions were visually assessed using a disease severity scale of 0 (no darkening around wounds) to 9 (large gray-green lesions > 7 mm diameter, with profuse sporulation) (Koch et al., 1991). Plants were inoculated with either a single isolate (four inoculation points per plant, two per cotyledon) or two isolates (one isolate on the left inoculation point and the other isolate on the right inoculation point on each cotyledon). Mean pathogenicity scores were calculated from two or four inoculation points on the cotyledons of at least 10 replicate plants. Data were analyzed using one-way ANOVA comparisons. 2.3. Gene prediction and expression analysis Genomic DNA was prepared from mycelia of the T-DNA mutant and digested with an enzyme that cut once within the T-DNA and with an enzyme that did not cut in the T-DNA. Southern analysis showed that a single copy of the T-DNA was integrated in the genome but that part of the T-DNA was truncated. Thermal asymmetric interlaced (TAIL)-PCR identified sequences flanking the T-DNA insertion as previously described (Mullins et al., 2001) with the following modifications. An 840 bp PCR product was amplified from the T-DNA left border (LB) using primers LB1, LB2 and LB3 in combination with redundant primer AD2 (Specht et al., 1996) and cloned into plasmid pCR2.1 (Invitrogen, USA), which was then sequenced. However, TAIL-PCR failed to amplify sequence flanking the right border (RB) of the T-DNA, probably due to truncation of the T-DNA sequence. A 600 bp SacI/XhoI fragment of the left border PCR product was radiolabeled with [a-32P] dCTP and used to screen a cosmid library of isolate IBCN18 to obtain the wild type copy of the mutated gene. A 42 kb hybridizing cosmid was identified and 3.4 kb flanking the T-DNA insertion was sequenced. All primers used in this study are listed in Table 1. The coding region of a gene flanking the T-DNA insertion (subsequently referred to as LmIFRD) and a gene downstream of the T-DNA (subsequently referred to as LmcABH-like) were predicted using FGENESH software (http:// www.softberry.com/berry.phtml). A full length cDNA of LmIFRD was amplified with primers Exon1F and Exon3R and then sequenced. Intron positions were confirmed by comparison with the genomic sequence. Transcriptional start and stop sites in the wild type LmIFRD gene were determined via 50 - and 30 Rapid Amplification of cDNA Ends (RACE) (Invitrogen) using primers 50 RACE and 30 RACE (Accession number: GQ183869).

Table 1 Sequences of oligonucleotides used in this study. Primer name

Sequence (50 –30 )

Use

LB1 LB2 LB3 AD2 Exon1F Exon 2R Exon3F Exon3R Exon 2F LmIFRDqF LmIFRDqR 50 RACE 30 RACE attB1LmIFRD attB2LmIFRD cABH1F cABH1R ActinF ActinR

GTGTAAAGCCTGGGGTGCCTAATGAGTG AGCTAACTCACATTAATTGCGTTGCG CGGGGAGAGGCGGTTTG AGWGNAGWANCAWAGG GCTCACGCATTGTCTCAACT GCACCTCGATCTCGTCTTTC CAGCAGTAGAAGCCGACGAT TTCGTTGTCTTGACCCGTCT GGAAGGCTGAGCTCAATACG GGAAGGCTGAGCTCAATACG CACACTGGACTCGGCTGATA GCGCCTTGCGAGAAACTGTCTTGTG AGGAGGACGCTTCAAGGTGGCTTCA GGGGACAAGTTTGTACAAAAAAGCAGGCTCGCGATCTTCGTAGACAGG GGGGACCACTTTGTACAAGAAAGCTGGGTTTACGTTGTCTCCTGTTTGATGC AGCGGTCTCGTGTCTTCTTC CCAGCGTTTGCCATACTCA TTGGTCTTGAAAGCGGTGGTAT CATCACTGTCCCACGAATTG

Tail PCR LB1 Tail PCR LB2 Tail PCR LB3 Redundant primer for tail PCR RT-PCR, Probe 1, sequencing RT-PCR and Probe 2 RT-PCR qRT-PCR, sequencing 50 RACE 30 RACE Silencing vector Silencing vector RT-PCR and Probe 3 RT-PCR and qRT-PCR

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

Total RNA from L. maculans mycelia isolated after 5 days growth in 10% Campbell’s V8 juice was resolved by gel electrophoresis and blotted onto nylon membranes, as previously described (Sexton and Howlett, 2000). Total RNA was hybridised to Probe 1 (amplified from wild type cDNA using primers Exon1F and Exon2R – to detect the 50 end of the transcript), Probe 2 (amplified from wild type cDNA using primers Exon3F and Exon3R – to detect the 30 end of the transcript) and Probe 3 (amplified from wild type cDNA using primers cABH1F and cABH1R – to detect expression of the LmcABH-like gene). Transcription of LmIFRD and LmcABH_like in the wild type isolate and the T-DNA mutant was also examined using reverse transcriptase-PCR (RT-PCR) experiments where cDNA was synthesized from total RNA using a first strand cDNA synthesis kit and an oligo-dT primer (Invitrogen). Several genes with potential roles in cell wall biosynthesis, fungal nutrition or germination (based on their reported function in closely related fungi) were identified in the draft sequence of the L. maculans genome of isolate JN3 (Genoscope and Unité de Recherche Génomique Info, France) by blasting predicted protein sequences (Table S1). Upon identification of matching L. maculans sequences, the location of genes was predicted using FGENESH software with settings for Aspergillus spp. or Phaeosphaeria nodorum. Expression of these genes and LmIFRD in the wild type and mutant isolates after 2, 3, 4 and 5 days growth in 10% Campbell’s V8 juice was determined using semi-quantitative RT-PCR. To quantify cDNA concentrations, actin was amplified from serial dilutions of cDNA (undiluted, 1 in 2, 1 in 5, 1 in 25, 1 in 125, 1 in 625). Products were separated on a 2% agarose gel. Dilutions of cDNA that amplified actin to similar levels were then used as templates to amplify fungal genes of interest. Expression of LmIFRD was also examined using RT-PCR when the wild type isolate was grown under starvation conditions (growth in 10% Campbell’s V8 juice for 5 days and then in minimal medium (Newton and Caten, 1988) with 15% sucrose for 8 h), and during in planta growth 4, 7 and 11 days after inoculation of cotyledons of B. napus cv. Monty. 2.4. Gene silencing

697

the naked eye were counted from day four until day 10. The presence or absence of asexual fruiting bodies (pycnidia) was also scored. Percentage germination was determined as the number of colonies that germinated from 100 spores. Percentage of pycnidia-producing colonies was determined as the number of colonies producing pycnidia from the total number of colonies. Three biological replicates were used to determine average values and standard errors. 2.6. Cell wall lysis assays Conidia (107) were incubated in 10% Campbell’s V8 juice (5 ml) for 24 h. The suspension was then centrifuged at 10,000g for 1 min and conidia were resuspended in 1 ml of treatment solutions. Treatment solutions included a cell wall lysis enzyme mixture containing b-glucanase (linkage specificity is unknown), cellulase, protease and chitinase activities from Trichoderma harzianum (3.5 mg/ ml in water, Sigma L1412), an endo-b1-3 glucanase from Arthrobacter luteus (1 mg/ml in water, MP Biomedicals 150214), a lysozyme (endo-chitinase) from chicken egg white (1 mg/ml in water, Sigma L3790) and a chitinase (a mixture of endo- and exo-chitinase) from Trichoderma viride (1 mg/ml in phosphate buffer pH 6.0, Sigma C8241). The control solution consisted of either water or phosphate buffer (pH 6.0). An aliquot of the treatment or control spore suspensions (10 ll) was applied to cover slips precoated with the polycation, 0.1% polyethylimine (Sigma) and left for 15 min. Glass slides were then lowered onto cover slips and sealed using VALAP (vaseline, lanolin and paraffin 1:1:1). The slides were incubated at 22 °C for up to 16 h and observed using a compound microscope with an eyepiece grid. Co-ordinates of fields of interest were recorded to permit ongoing observation. Intact conidia were oval in shape and were counted at 0 and 16 h. Five replicate slides were examined to determine the mean percentage of spore survival. Data were analyzed using one-way ANOVA comparisons. 2.7. Growth of isolates in the presence of cell wall-disrupting compounds

Since the T-DNA mutant produced a truncated transcript of LmIFRD that may have retained function, a gene silencing hairpin construct was designed to create an isolate with severely reduced expression of LmIFRD. A 438 bp product was amplified from cDNA of the wild type isolate using primers attB1LmIFRD and attB2LmIFRD, which bind in Exons 1 and 2, respectively. The PCR product was subsequently cloned in both sense and antisense orientation into pHygGS (Fox et al., 2008) resulting in the gene-silencing construct, pIFRDRNAi. The L. maculans wild type isolate was transformed with pIFRDRNAi and four transformants, RNAi1.1, RNAi1.4, RNAi1.6 and RNAi1.8, were selected. Levels of gene expression were determined by quantitative RT-PCR using SensiMix (dT) SYBRÒ Green PCR kit (Quantace) in a Corbett Rotor-Gene 3000 machine. Transcript levels of the gene of interest were normalized to L. maculans actin as described previously (Gardiner et al., 2004). Primers LmIFRDqF and LmIFRDqR were used to amplify LmIFRD, and primers ActinF and ActinR were used to amplify actin.

Isolates were grown in the presence of the cell wall-disrupting dyes, Congo Red (CR) or Calcofluor White, (CFW) or the growth inhibitors Nikkomycin Z (a chitin synthesis inhibitor (Gaughran et al., 1994)) or Picoxystrobin (an inhibitor of fungal respiration (Bartlett et al., 2002)). Growth mortality curves were used to determine the appropriate concentration to use in the survival assay (data not shown). Conidia (104) were added to 190 ll of 10% Campbell’s V8 juice supplemented with CR (0.02 mg/ml, Selby), CFW (0.3 mg/ml, Megazyme), Nikkomycin Z (100 ng/ml, Sigma) or Picoxystrobin (0.1 lg/ml, Sigma) in eight replicate wells of a 96 well microtitre plate and incubated at 22 °C. Light scattering (OD595) was measured using a Milenia Kinetic analyzer plate reader after 0 and 100 h growth. The relative growth of each isolate was calculated as a ratio of the OD595 value on supplemented medium to the OD595 value in un-supplemented medium. Data were analyzed using one-way ANOVA comparisons.

2.5. Germination and sporulation tests

Cotyledons of infected B. napus cv. Monty at 14 days post-inoculation (dpi) were stained with lactophenol trypan-blue and cleared in saturated chloral hydrate (2.5 g/ml, Sigma) (Keogh et al., 1980). Tissues were observed using either an Olympus BH2 microscope or a Leica MZ FLIII dissecting microscope, and photographed. Germinating conidia were harvested from 10% Campbell’s V8 juice at 9, 16, 24, 32 and 96 h and photographed with a Leica camera mounted on an Olympus BH2 microscope. To assess lectin

Conidia were harvested by flooding 10-day-old-cultures of L. maculans on 10% Campbell’s V8 juice agar with sterile water. Conidia concentrations were adjusted to 1  105 spores/ml. Three independent dilutions were prepared for each isolate and 100 spores were spread on 10% Campbell’s V8 juice agar dishes (9 cm diameter) and incubated in the dark at 22 °C. Colonies visible to

2.8. Microscopy

698

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

binding, germinating conidia were harvested at the times indicated in individual figures and centrifuged at 6000g for 10 min to remove growth media. These conidia were washed twice in phosphate buffered saline (pH 7.4) (PBS) before incubation with wheat germ agglutinin (WGA)-fluorescein isothiocyanate (FITC) (100 lg/ml, C-Y labs) or concanavalin A (ConA)-tetramethyl rhodamine isothiocyanate (TRITC) (100 lg/ml, C-Y labs) for 1 h at room temperature with gentle agitation. Conditions were optimized to determine appropriate concentrations of lectins. Samples were washed twice with PBS before being applied to cover slips coated with 0.1% polyethylimine as described above. Duplicate samples were also mounted in the presence of CFW (35 lg/ml in 0.1 M Tris-HCl pH 9, 0.1% Triton X) (Megazyme) to compare localization of staining. Competitive inhibition assays were performed with N-acetylglucosamine (40 nM) for WGA and mannose (40 nM) for ConA to indicate specificity of lectin binding. Samples were examined using a Leica TCS-SP confocal laser scanning microscope with excitation by an Argon–Krypton laser. Optical sections (0.5 lm) were collected through the specimen and a maximum projected image was produced. 2.9. Monosaccharide and linkage analysis of cell walls Conidia (107) were incubated in 20 ml of 10% Campbell’s V8 juice. Germinating spores were harvested after 16 h for the silenced mutant, 24 h for the wild type isolate and 32 h for the TDNA mutant by centrifugation at 3000g for 15 min. Germinating spores were resuspended in 1 ml of TE buffer (10 mM Tris, 1 mM EDTA pH 7.5). Alternatively, mycelia were harvested after 96 h growth in 20 ml of 10% Campbell’s V8 juice via filtration through miracloth. Mycelia were resuspended in TE buffer (1 ml). Resuspended germinating spores and mycelia were sonicated on ice for 1 min at constant maximum amplitude (494 microns) using a Branson Sonifier 450 with a 1/8 in. tapered microtip (101-148062). Following sonication, acid-washed glass beads (425– 600 lm, Sigma G-8772) were added in a 1:1 ratio of sample to beads. Samples were vortexed for 1 min, placed on ice for 1 min and then vortexed again for 1 min. The supernatant was then removed to a new tube and microscopically examined for the presence of cell wall fragments. All samples were then centrifuged at 17,900g for 2 min and the pellet (containing cell wall debris) was analyzed for monosaccharide linkage composition by GCMS of permethylated alditol acetates. Wall polysaccharides were methylated with methyl iodide in DMSO and sodium hydroxide (Ciucanu and Kerek, 1984). Permethylated polysaccharides were hydrolyzed with 2 M trifluoroacetic acid (TFA) at 121 °C for 2 h then reduced and acetylated (Sims and Bacic, 1995). Permethylated alditol acetates were separated on a CPSil5 column (25 m  0.3 mm, Chrompack, Middelburg, The Netherlands) using a Hewlett–Packard HP 6890 gas chromatograph equipped with an autosampler. They were detected by electron impact ionization (70 eV) mass spectrometry on a Hewlett–Packard 5973 mass selective detector. Data were acquired in full scan mode (total ion chromatogram) to detect ions from m/z 100–700. Two technical replicates were performed for each of the three biological replicates. Data were analyzed using t-test analysis.

3. Results 3.1. Pathogenicity of L. maculans T-DNA mutant, A9 Of 70 random insertional mutants of L. maculans tested, only one (A9) displayed a loss-of-pathogenicity phenotype on wounded cotyledons of B. napus cv. Monty. This mutant failed to develop lesions after 21 days whereas large lesions were caused by the wild

type isolate (data not shown). Lactophenol trypan-blue staining, indicative of fungal structures and host cell death, showed that the wild type isolate germinated on the surface of the cotyledon and colonized the palisade and spongy mesophyll layers. The TDNA mutant germinated on the surface of the cotyledon, but did not penetrate the host (data not shown). Ten random ascospore progeny from a cross between the T-DNA mutant and Lm691 were analyzed for their ability to infect wounded cotyledons of B. napus cv. Monty and for hygromycin resistance. All five hygromycinresistant progeny showed a loss-of-pathogenicity phenotype indicating that the phenotype segregated with the T-DNA insertion (data not shown). 3.2. Characterisation of the insertion in the L. maculans T-DNA mutant Sequencing and gene prediction of DNA flanking the T-DNA revealed that the insertion was located within the third Exon of a gene with best match to the ‘IFRD domain-containing protein’, UPTRG02527, from the dothideomycete, Pyrenophora tritici-repentis (72% identity, 85% similarity, accession number XP_001932860). Matches were also identified in other ascomycetes including A. fumigatus (43% identity, 59% similarity) and Schizosaccharomyces pombe (26% identity, 46% similarity) but none were identified in S. cerevisiae or in any Basidiomycete. These proteins all contain a putative conserved interferon-related developmental regulator (IFRD) domain, which includes single and sets of amino acids spanning the entire protein, and has not been assigned any particular function (Buanne et al., 1998; Latif et al., 1997). This domain in LmIFRD is dispersed over the first 370 amino acids, spanning past the insertion site in the T-DNA mutant. Two other domains, a bi-partite nuclear localization signal from amino acid 5 to 21 and a coiledcoil domain, spanning amino acids 248 through 275 are predicted by PSORTII (http://psort.ims.u-tokyo.ac.jp/form2.html). These latter domains are present in the T-DNA mutant. To confirm intron and transcription start and stop positions for LmIFRD, cDNA of the wild type isolate was subjected to RT-PCR and RACE-PCR methods. These analyses confirmed that the LmIFRD open reading frame is 1362 bp and interrupted by two introns of 112 and 46 bp (Fig. 1A). The transcript produced in the wild type isolate contains a 50 untranslated region (UTR) that is 273 bp in length and a 190 bp 30 UTR. This total transcript length is 1983 bp. No alternate splicing or transcriptional start sites were observed in the wild type following RACE-PCR and RT-PCR analysis using seven different primer combinations (data not shown). The effect of the T-DNA insertion on expression of LmIFRD in the T-DNA mutant was examined by Northern analysis. Two probes were used, one 50 of the T-DNA insertion site (hybridizing to Exons 1 and 2) and one 30 of the insertion site (hybridizing to Exon 3) (Fig. 1B). Using the 50 probe, a broad 1.7 kb band was detected in the wild type isolate, which is consistent with RT-PCR and RACEPCR analysis, whilst a 1.1 kb transcript was detected in the TDNA mutant. The 30 probe detected a 1.7 kb transcript in the wild type isolate, but no transcript in the T-DNA mutant (Fig. 1C). This was confirmed by RT-PCR whereby amplified products were detected for both isolates using primers amplifying Exons 1 and 2, but not in the T-DNA mutant using primers 30 of the T-DNA insertion site (Fig. 1C). These findings indicate that a truncated transcript of LmIFRD is produced in the T-DNA mutant. Amplification of cDNA of the T-DNA mutant using primers Exon 2F and primers in the T-DNA LB sequence (LB1, LB2 and LB3), showed that the mutant LmIFRD transcript extends at least 240 bp into the T-DNA. The predicted amino acid sequence of LmIFRD in the T-DNA mutant terminates with a fusion of 11 amino acids resulting from in-frame translation of the T-DNA sequence (Fig. 1D). Using RT-PCR, expression of LmIFRD was detected in the wild type isolate under all conditions tested including during germina-

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

699

A

B C

D gaa gag cca tca gac gcc gat gat cag gat ata ttg tgg tgt aaa caa att gac gct tag E E P S D A D D Q D I L W C K Q I D A -

Fig. 1. Structure and expression of the LmIFRD gene in the wild type Leptosphaeria maculans isolate and a T-DNA insertion mutant. (A) Structure of the LmIFRD gene. The position of the T-DNA insertion in the mutant is depicted by a triangle. (B) Position of primers and probes used for expression analysis of LmIFRD. Probes were amplified from cDNA of the wild type isolate. (C) Northern analysis, using a probe amplified from the 50 end of the gene, revealed a broad 1.7 kb band in the wild type (WT) isolate and 1.1 kb in the T-DNA mutant (Mutant). When a Northern blot was hybridized with a probe amplified from Exon 3, downstream of the T-DNA insertion site, a transcript sized 1.7 kb was observed in the wild type but not in the T-DNA mutant. These findings were confirmed using reverse transcriptase-PCR (RT-PCR). Ribosomal RNA bands for wild type and mutant isolates from the northern blot indicate equal amounts of RNA in each lane. Amplification of actin was used as a positive control for RT-PCR. (D) Structure of the truncated LmIFRD transcript produced in the T-DNA mutant. Primers specific to Exon 2 (Exon 2F) of the LmIFRD gene and the left border of the T-DNA (LB1, LB2 and LB3) were used to amplify the transcript. The underlined sequence represents the end of the wild type section of the LmIFRD sequence before the in-frame extension of at least 240 base pairs into the T-DNA, resulting in a fusion of 11 additional amino acids.

tion, in planta at 4, 7 and 11 days post-inoculation and under starvation conditions (data not shown). Semi-quantitative RT-PCR analysis revealed no differences in the patterns or levels of LmIFRD expression in the wild type and T-DNA mutant after 2, 3, 4 and 5 days growth in vitro (data not shown). The expression of LmIFRD in the T-DNA mutant was not tested in planta because the T-DNA mutant is non-pathogenic. Another gene (LmcABH-like) was predicted 647 bp downstream of the T-DNA insertion, with best match to a predicted protein, which contains a cyclophilin_ABH-like domain and tetratricopeptide repeat domain. The first domain is found in proteins involved in protein folding and stabilization whilst the second domain is involved in protein–protein interactions. The expression of the LmcABH-like gene was examined in the wild type and T-DNA mutant after 5 days growth in liquid V8 juice using RT-PCR and northern analysis. No differences in expression were observed between the two isolates (data not shown). 3.3. RNA-mediated silencing of LmIFRD RNAi-mediated gene silencing of LmIFRD was employed to create a mutant with reduced LmIFRD expression. The wild type isolate was transformed with the silencing construct, pIFRDRNAi, and four resultant transformants (RNAi1.1, RNAi1.4, RNAi1.6, RNAi1.8) were assessed for reduction of LmIFRD expression. In the wild type isolate the expression level of LmIFRD relative to actin was 1.07  102. Varying degrees of LmIFRD silencing was

achieved in the four RNAi mutants with expression levels (relative to actin) ranging from 8.26  103 in isolate RNAi1.1 to 1.13  103 in isolate RNAi1.4 (Supplementary Fig. S1A). Isolate RNAi1.4 (subsequently referred to as the silenced mutant) showed the highest level of silencing (89% reduction relative to wild type) and was used in subsequent studies. 3.4. Growth phenotypes and pathogenicity of wild type, T-DNA and silenced mutants of L. maculans Germination of conidia of the wild type isolate in 10% Campbell’s V8 juice was preceded by isotropic swelling whereby most conidia had swollen to twice their initial size after 16 h incubation. Polar growth was initiated after 24 h and most conidia had developed elongated and branched hyphae after 32 h (Fig. 2A). After 5 days, 52% of conidia had germinated (Fig. 2B). In contrast, the T-DNA mutant germinated more slowly. Conidia began to swell isotropically after 24 h, with emergence of germ tubes and elongation of hyphae after 32 h (Fig. 2A). After 5 days, only 3% of conidia had begun to form colonies, however, after 8 days the germination frequency was similar to that of the wild type isolate (Fig. 2B). Germination began earlier in the silenced mutant with isotropic growth visible by 9 h and germ tube elongation beginning at 16 h. After 24 h, germ tubes had elongated with obvious septa, and long branched hyphae were evident after 32 h (Fig. 2A). By 4 days, 75% of conidia of the silenced mutant had begun to form colonies compared to only 22% for the wild type isolate. After

700

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

Fig. 2. Germination of conidia of the wild type, T-DNA and silenced mutants of Leptosphaeria maculans. (A) Isotropic swelling preceded germination of wild type spores, whereby they had swollen to twice their initial size by 16 h. After 24 h, germ tubes had emerged with septa evident (arrows). Most spores had developed elongated and branched hyphae after 32 h. Isotropic swelling was not evident until 16 h in T-DNA mutant spores, with germ tube elongation and septa formation apparent after 32 h. Conversely, conidia of the silenced mutant germinated earlier than wild type with germ tube elongation and septa formation occurring after only 16 h. (B) Germination rates of the wild type, T-DNA and silenced mutants. After 5 days, only 3% of conidia of the T-DNA mutant had germinated. However, after 8 days the germination frequency was similar to that of the wild type isolate. In contrast, by 4 days, 75% of conidia of the silenced mutant had begun to germinate compared to only 22% of the wild type. Data points represent average percentage germination of 100 conidia plus or minus the standard error of three replicate experiments. Scale bars: 25 lm.

9 days, a higher percentage of spores of the silenced mutant had germinated compared to the wild type or T-DNA mutant (Fig. 2B). The other silenced mutants, RNAi1.1, RNAi1.6 and RNAi1.8, also germinated more quickly than the wild type isolate, with increased rate of germination correlating with reduction of LmIFRD expression (Supplementary Fig. S1B). Only 2% of colonies of the T-DNA mutant produced pycnidia after 10 days growth compared to 89% of the wild type colonies (Fig. 3A). In addition, each individual sporulating colony of the TDNA mutant had very low density of pycnidia (Fig. 3B). Colonies of the silenced mutant produced pycnidia earlier than the wild type isolate: all colonies had formed pycnidia after 7 days compared to only 52% of the wild type. The wild type isolate only produced pycnidia in the oldest part of the colony whereas the silenced mutant produced them throughout (Fig. 3B). Isolates were inoculated onto wounded cotyledons of B. napus cv. Monty. The T-DNA mutant produced significantly fewer lesions than both the wild type isolate and silenced mutant at all time points (Fig. 4). Conversely, significantly higher mean pathogenicity scores were observed for the silenced mutant at 9 days post-inoculation (dpi) compared to the wild type isolate (p = 0.02). At later

time points, this difference became less pronounced. Lesions did not increase in size after 17 dpi on cotyledons inoculated with any of the isolates. 3.5. Gene expression in wild type, T-DNA and silenced mutants of L. maculans Using semi-quantitative RT-PCR, the expression of genes with potential roles in cell wall biosynthesis, fungal nutrition or germination was determined. These included seven chitin synthase genes, a transcription factor essential for germination and sporulation, seven genes expressed in some ascomycetes during germination and sporulation, and two genes (histone deacetylase and Sin3a) encoding proteins predicted to interact with the mouse IFRD protein (Supplementary Table S1). Expression of these genes was assessed at 2, 3, 4 and 5 days in 10% Campbell’s V8 juice. Although the expression of individual genes varied at particular times, expression levels were similar in the wild type isolate and the TDNA and silenced mutants at all time points assessed (data not shown). Expression levels of these genes in the wild type isolate are shown in Supplementary Table S1.

701

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

A

B

Fig. 3. Sporulation rates and colony morphology of the wild type, T-DNA and silenced mutants of Leptosphaeria maculans. (A) Percentage of colonies that produced pycnidia of the wild type, T-DNA and silenced mutants. (B) Colony morphologies of these isolates after 10 days growth on 10% Campbell’s V8 juice. The T-DNA mutant produced fewer colonies with pycnidia than both the wild type isolate and silenced mutant. Individual colonies of the T-DNA mutant produced pycnidia (arrows) at a very low density within the older regions of the colony. The silenced mutant produced pycnidia earlier than the wild type isolate and throughout the colony. Scale bar: 2 mm. Data points represent average percentage of pycnidia-producing colonies relative to total number of colonies for three biological replicates. Error bars represent standard errors.

Table 2 Percentage survival of conidia of Leptosphaeria maculans isolates in the presence of cell wall lysing enzymes. Enzyme (activity)

Cell wall lysis enzyme mixture (b-glucanase, cellulase, protease and chitinase) Endo-b1-3 glucanase Chitinase (endo and exo) Lysozyme (endo-chitinase)

Fig. 4. Mean pathogenicity scores of the wild type, T-DNA and silenced mutants of Leptosphaeria maculans on cotyledons of Brassica napus cv. Monty. The T-DNA mutant produced very few lesions and these did not increase in size after 17 days post-inoculation (dpi). In contrast, at nine dpi, the silenced mutant caused a greater number and larger lesions than the wild type isolate, resulting in a higher mean pathogenicity score. However, as infection progressed, this difference in mean pathogenicity scores became less pronounced. Mean pathogenicity scores at nine dpi were significantly different () using one-way ANOVA analysis (p = 0.02). Data points represent mean pathogenicity score of 28 inoculation sites from 14 replicate plants. Error bars represent standard errors.

3.6. Cell wall integrity of wild type, T-DNA and silenced mutants of L. maculans Germinating conidia were incubated with particular cell wall lysing enzymes and the percentage of conidia that had not lysed (survival rate) following 16 h incubation was determined. In the presence of a mixture of enzymes including b-glucanase, cellulase, protease and chitinase, conidia of the T-DNA mutant had a significantly higher survival rate than conidia of the wild type isolate (Table 2). Conversely, conidia of the silenced mutant had a significantly lower survival rate than wild type. A similar trend was observed when conidia were incubated with endo-b1-3 glucanase (Table 2). When incubated with the endo- and exo-chitinase mixture, conidia of the T-DNA mutant had a significantly lower survival rate than

Isolate Wild type

T-DNA mutant

Silenced mutant

74 ± 6

95 ± 2a

28 ± 9b

81 ± 12 73 ± 4 81 ± 6

96 ± 3a 63 ± 3b 82 ± 4

46 ± 3b 81 ± 3a 77 ± 7

Values are the percentage of conidia that did not lyse after incubation with the enzyme for 16 h. Five replicates were examined (plus or minus standard error). Data were analyzed using one-way ANOVA comparisons. a Significantly increased tolerance compared to wild type (p < 0.01). b Significantly decreased tolerance compared to wild type (p < 0.01).

the wild type isolate. However, the silenced mutant displayed a significantly higher survival rate than the wild type isolate (Table 2). When the isolates were incubated with an endo-chitinase or control treatments (data not shown), no significant differences in survival were observed for any of the isolates. The three isolates were tested for their ability to grow in the presence of cell wall-disrupting compounds, Congo Red (CR) and Calcofluor White (CFW). Rather than being inhibited, the T-DNA mutant grew significantly better in the presence of either CR or CFW (Table 3). In contrast, after 100 h the silenced mutant showed a significantly decreased tolerance to both CR and CFW relative to wild type (Table 3). Isolates were screened for their ability to grow in the presence of a chitin synthase inhibitor, Nikkomycin Z. This molecule was more effective at inhibiting the growth of all three isolates than CR and CFW. The T-DNA mutant was significantly more tolerant to Nikkomycin Z than the wild type isolate. However, the silenced mutant was significantly more sensitive than the wild type isolate (Table 3). The increase in Nikkomycin Z tolerance of the T-DNA mutant also segregated with hygromycin resistance in progeny from crossing the T-DNA mutant and Lm691 (data not shown).

702

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

Table 3 Growth of Leptosphaeria maculans isolates in the presence of cell wall-disrupting compounds, Congo Red and Calcofluor White, and a chitin synthase inhibitor, Nikkomycin Z. Supplement

Congo Red (0.02 mg/ml) Calcofluor White (0.3 mg/ml) Nikkomycin Z (1 lg/ml)

Isolate Wild type

T-DNA mutant

Silenced mutant

88.6 ± 2.0 80.1 ± 2.0 9.5 ± 1.2

105.6 ± 3.0b 131.9 ± 3.0b 12.9 ± 0.8b

77.6 ± 2.4a 64.2 ± 2.6a 6.1 ± 0.3a

Growth was measured using light scattering (OD595) after 0 and 100 h growth. Values are the percentage growth of each isolate on supplemented media relative to un-supplemented media after 100 h. Values are the average of eight replicates plus or minus standard error. Data were analyzed using one-way ANOVA comparisons. a Significantly decreased tolerance compared to wild type (p < 0.01). b Significantly increased tolerance compared to wild type (p < 0.01).

The T-DNA and silenced mutants both constitutively expressed the selectable marker that conferred hygromycin resistance. To demonstrate that the hygromycin resistance gene was not altering tolerance to Nikkomycin Z, two hygromycin-resistant mutants produced by targeted mutation of individual genes (sirP, a peptide synthetase (Gardiner et al., 2004) and sirA, an ABC transporter (Gardiner and Howlett, 2004)) were examined. These isolates had a similar tolerance to Nikkomyzin Z as the wild type isolate (data not shown). To determine whether the changes in tolerance of the T-DNA and silenced mutants were specific to cell wall-related disruption, growth of the three isolates was assessed in the presence of Picoxystrobin, which has a different mode of action (inhi-

bition of fungal respiration) to that of Nikkomycin Z. No significant differences in survival were observed (data not shown). 3.7. Lectin binding to the cell walls of wild type, T-DNA and silenced mutants of L. maculans When the three isolates were at similar growth stages, whereby the hyphae were approximately twice as long as the spore diameter, they were incubated with Wheat Germ Agglutinin-FITC (WGA), CFW or Concanavalin A-TRITC (ConA) and examined by fluorescence confocal microscopy. When incubated with WGA, the TDNA mutant fluoresced more brightly than the wild type isolate, whilst the silenced mutant showed very little fluorescence (Fig. 5A). In the wild type isolate, fluorescence was more prominent in the spore body and growing hyphal tip. In the T-DNA mutant, fluorescence was uniformly distributed along the entire germinating spore and elongating hyphae. In the silenced mutant, fluorescence was only visible in the growing hyphal tip. When conidia of the wild type isolate were incubated with N-acetyl glucosamine (40 nM) and WGA simultaneously, fluorescence in cell walls was abolished, confirming the specificity of this lectin for b1,4-Nacetylglucosamine. When incubated with CFW, germinating spores of the T-DNA mutant fluoresced much more brightly than the wild type or the silenced mutant (Fig. 5A). As seen with WGA, the distribution of CFW fluorescence was distributed along the entire surface of the germinating spore and also septa of the T-DNA mutant. Conversely, fluorescence was localized to the septa and walls of the hyphae of the wild type isolate and silenced mutant. Mycelia from 72 h cul-

Fig. 5. Binding of lectins to cell walls of the wild type, T-DNA and silenced mutants of Leptosphaeria maculans. (A) Isolates at the same developmental stage (wild type, 24 h; T-DNA mutant, 32 h; silenced mutant, 16 h) were incubated with Wheat Germ Agglutinin-FITC (WGA; 100 lg/ml) or Calcofluor White (CFW; 35 lg/ml). The T-DNA mutant fluoresced more brightly than wild type when incubated with WGA or CFW whereas the silenced mutant showed very little WGA fluorescence, but similar CFW fluorescence to the wild type. (B) Mycelia from 72 h cultures were incubated with CFW (35 lg/ml). Hyphal tips and walls of the T-DNA mutant fluoresced more brightly than those of the wild type isolate, whereas only the tips of the silenced mutant fluoresced. (C) Germinating spores of the three isolates were incubated with ConcanavalinA-TRITC (ConA; 100 lg/ml). The spore cell wall fluoresced brightly in all three isolates. These photos are representative of three experiments. Scale bar 10 lm.

703

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

tures were also incubated with CFW to determine whether the differences in fluorescence were specific to germinating spores. As in germinating spores, increased fluorescence was observed in mycelia of the T-DNA mutant compared to that of the wild type isolate (Fig. 5B). The fluorescence in mycelia of the wild type isolate was located predominantly in the hyphal tips with minor levels in the lateral cell walls. A lower level of fluorescence was observed in the hyphal tip in the mycelia of the silenced isolate compared to both the wild type and T-DNA mutant (Fig. 5B). A similar pattern was observed after incubation of hyphal fragments from 96 h cultures of all three isolates with WGA (data not shown). Similar levels and distribution of fluorescence was observed in all three isolates when incubated with ConA. Staining was brightest in ungerminated spores and in the spore body after germ tube emergence (Fig. 5C). When conidia of the wild type isolate were incubated with mannose (40 nM) and ConA simultaneously, fluorescence in cell walls was abolished, confirming the specificity of this lectin for mannose 3.8. Monosaccharide linkage analysis of cell walls of wild type, T-DNA and silenced mutants of Leptosphaeria maculans In cell walls of germinating spores of the wild type, the most abundant monosaccharide was glucose (84.1%), with 1,4-linkages representing 49.9% of the total monosaccharides (Table 4). Glucose

was also the most abundant monosaccharide in the cell walls of wild type mycelia, however, glucose in 1,4-linkages represented only 13.9% of the total monosaccharides at this growth stage. There were significant differences in abundance of particular monosaccharide linkages between wild type germinating spores and mycelia. These included increases in mannose with 1,2-linkages and glucose with 1,6-linkages and decreases in mannose with 1,2,3linkages and glucose with 1,4- and 1,4,6-linkages. Monosaccharide composition analysis by hydrolysis of the walls with H2SO4 revealed similar patterns in monosaccharide content as those deduced from the linkage analysis (data not shown). Germinating spores of the T-DNA mutant had significantly increased proportions of galactose in 1,3-linkages whilst mannose with 1,6-linkages and glucose with 1,4- and 1,6-linkages were less abundant, compared to germinating spores of the wild type. In fact, galactose in 1,3-linkages made up 49.7% of the total monosaccharide content in germinating spores of the T-DNA mutant compared to trace amounts in wild type. Increased abundance of galactose in 1,3-linkages were also observed in mycelia of the T-DNA mutant compared to wild type. Proportions of mannose with 1,6-linkages, galactose with 1,3-linkages and glucose with 1,4- and 1,6-linkages were significantly different between mycelia of the wild type isolate and the T-DNA mutant. Additionally, mycelia of the T-DNA mutant had significantly increased abundance of N-acetylglucosamine (7.7%) compared to mycelia of the wild type (1.0%) (Table 4).

Table 4 Monosaccharide linkage composition of cell walls from isolates of Leptosphaeria maculans (mol%). Deduced linkage

Wild type isolate Germinating sporesa

c

T-DNA mutant Myceliab

Germinating sporesa d

Silenced mutant Myceliab e

Germinating sporesa

Myceliab

t-Manp 1,2-Manp 1,6-Manp 1,2,3-Manp 1,2,6-Manp 1,3,6-Manp

3.1 ± 0.1 – 7.6 ± 0.0 2.3 ± 0.1 tr –

14.6 ± 3.2 2.6 ± 0.4d 9.4 ± 0.4 – 1.0 ± 0.2 –

3.2 ± 0.2 – 3.8 ± 0.4d 1.4 ± 0.1 tr –

4.4 ± 0.5 tr 4.5 ± 0.6e – tr tr

3.4 ± 0.1 – 7.5 ± 0.4 3.7 ± 0.8d tr –

10.3 ± 2.8 1.3 ± 0.4e 6.7 ± 0.4 – 1.8 ± 0.0e –

Total

13.5

27.6

9.0

10.0

15.5

20.2

t-Galp 1,3-Galp 1,4-Galp 1,2,3,4-Galp

tr tr tr –

– – – –

tr 49.7 ± 14d – –

– 1.7 ± 0.1e – tr

tr tr – –

– tr – tr

Total

1.0

0.0

50.4

1.9

0.8

0.5

t-Glcp 1,2-Glcp 1,3-Glcp 1,4-Glcp 1,6-Glcp 1,2,3-Glcp 1,3,4-Glcp 1,3,6-Glcp 1,4,6-Glcp 1,2,3,6-Glcp 1,3,4,6-Glcp

15.2 ± 2.4 tr 13.5 ± 0.9 49.9 ± 1.5 2.3 ± 0.3 – tr tr 2.2 ± 0.0 – –

28.5 ± 1.5d tr 13.6 ± 0.2 13.9 ± 5.9d 12.4 ± 0.6d 1.0 ± 0.1 – tr – – –

6.7± 0.7d tr 17.1 ± 1.4 11.9 ± 1.1d 1.7 ± 0.2d – – tr tr – –

9.9 ± 2.5e tr 52.5 ± 3.8e 4.7 ± 0.6e 3.5 ± 0.2e 4.9 ± 1.3 – 2.0 ± 0.4 tr tr –

14.3 ± 0.5 tr 19.2 ± 1.2 40.8 ± 1.7 3.2 ± 0.2d – tr tr 2.4 ± 0.0 – –

18.3 ± 5.4 tr 24.7 ± 9.8 8.9 ± 2.5 17.1 ± 2.1e 2.7 ± 1.8 – 1.6 ± 0.9 tr tr tr

Total

84.1

70.5

38.6

78.1

81.3

73.6

Terminal 1,4-GlcNAcpf

– tr

– 1.3 ± 0.1

– tr

tr 7.6 ± 0.3e

– tr

– 2.0 ± 0.7

Total

tr

1.3

tr

7.7

tr

2.0

Values are the average molar percentage of monosaccharides relative to the total monosaccharide content plus or minus standard deviation. Average values are of three biological replicates. Data were analyzed using students t-test. tr = Trace values < 1.0%. – = Not detected. Trace amounts of arabinose and xylose were detected in every sample tested. a Germinating spores were grown in 10% V8 juice and were harvested after 24 h for the wild type isolate, 32 h for T-DNA mutant and16 h for the silenced mutant. b Mycelia were harvested after 96 h for all isolates. c p – Pyranosyl. d Significant differences compared to wild type germinating spores (p < 0.05). e Significant differences compared to wild type mycelia (p < 0.05). f This value represents both GlcNAc and GlcN as both appear as the same derivative after methylation analysis.

704

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

Fewer differences in monosaccharide content and structure were observed between cell walls of the wild type isolate and the silenced mutant. These differences included significantly increased abundance of glucose in 1,6-linkages in germinating spores and mycelia and significantly decreased levels of mannose in 1,2linkages in mycelia compared to the wild type isolate (Table 4).

4. Discussion This paper describes two mutants of L. maculans with aberrant expression of the same gene but with differing developmental phenotypes. These phenotypes are accompanied by changes in sensitivity of conidia to cell wall lysing enzymes and cell walldisrupting compounds and also by changes in monosaccharide linkages, and hence polysaccharide structure, in the cell walls. Analysis of these mutants and the wild type has revealed new information about the composition of the cell wall of a filamentous ascomycete and is the first detailed analysis of the cell wall of a dothideomycete. In addition, the use of random insertional mutagenesis has identified a novel fungal gene that may have a role in maintaining cell wall integrity or in cell wall remodeling. 4.1. Phenotypes associated with mutations of LmIFRD are consistent with a role for this gene in cell wall integrity The slow aberrant germination and sporulation rates of the TDNA mutant are in stark contrast to the rapid germination and profuse sporulation of the silenced mutant. Both of these opposing phenotypes have been associated with aberrations in cell wall structure in other fungi. For instance, deletion mutants of the A. fumigatus RasA gene, encoding a monomeric GTPase, had reduced germination, radial growth and asexual spore formation associated with a loss of cell wall integrity (Fortwendel et al., 2008). Conversely, deletion of the A. fumigatus ECM33 gene encoding a glycosylphosphatidylinositol-anchored protein, resulted in increased germination, altered sensitivity to Congo Red and increased virulence in immunocompromised mice (Romano et al., 2006). Increased tolerance of the T-DNA mutant and decreased tolerance of the silenced mutant to three cell wall-disrupting compounds with differing modes of action suggests that the cell wall polysaccharides of the mutants have altered accessibility, compared to the wild type. The alterations in tolerance to cell wall-disrupting compounds of the mutants have several interpretations. For instance, increased tolerance of the T-DNA mutant to CFW, CR, and the chitin synthase inhibitor Nikkomycin Z could be explained by a deficiency or inaccessibility of b1-4 and b1-3 linkages and chitin synthases in the cell wall. Alternatively, if an excess of b1-4 and b1-3 linkages and chitin synthases were present in the cell wall of the T-DNA mutant compared to the wild type, more of the cell wall-disrupting compounds would be required to produce a deleterious effect. In contrast to the phenotype of the TDNA mutant, RNAi-mediated silencing of the LmIFRD gene led to increased sensitivity to CR, CFW and Nikkomycin Z. Since an increased sensitivity to CFW has been associated with both increased and decreased levels of chitin (a b1-4 linked glucan), (Lussier et al., 1997), no firm conclusion can be drawn regarding chitin content from the sensitivity to these compounds alone. Increased binding of fluorescently-tagged WGA and CFW to the T-DNA mutant and decreased binding of the same compounds to the silenced isolate supports the idea that the T-DNA mutant has more chitin than wild type, whilst the silenced mutant has less chitin than wild type. Although the pattern of CFW binding in germinating spores of the silenced mutant was similar to that of wild type, the binding of CFW in older hyphae of the silenced isolate was only visible in hyphal tips, again suggesting less chitin in the

mycelia of the silenced isolate. This slightly different staining pattern for CFW and WGA in germinating spores is not without precedent. Although both CFW and WGA bind chitin polymers, previous studies in Ustilago maydis and Botrytis cinerea showed the dyes do not completely co-localize; WGA is restricted to the growing tips whereas CFW is concentrated in lateral walls and septa (Soulié et al., 2006; Wedlich-Soldner et al., 2000). Consistent with the microscopic analyses, monosaccharide analysis of the cell walls revealed that mycelia of the T-DNA mutant had an increased chitin content compared to wild type mycelia. However, there were no significant differences in chitin content between mycelia of the silenced mutant and wild type isolate, despite differences in microscopic and enzymatic assays. Only trace amounts of chitin were detected in germinating spores. Chitin is extremely difficult to quantify accurately because of its insolubility and the difficulty in methylating and hydrolyzing it with acid. Only minor changes in abundance of monosaccharide linkages of the cell wall of the silenced isolate were observed despite phenotypes consistent with cell wall integrity defects. This result is similar to findings of Hill et al. (2006) where a collection of mutants in A. nidulans with increased sensitivity to CFW showed no variation in chitin levels and only minor variation in the proportions of glucose and galactose. Perhaps small alterations to the monosaccharide content can lead to major impacts on cell wall integrity. However, limitations of the techniques used for the monosaccharide analysis may mask some additional differences between the wild type and the silenced mutant. For instance, differences in acetylation would not be detected in GCMS analyses of alditol acetates, because the acid hydrolysis at the beginning of the procedure removes acetyl groups from the monosaccharides. If the degree of acetylation were indeed altered in the silenced mutant, this would be consistent with the reduced WGA-FITC binding observed. Another limitation of the technique used for the monosaccharide analysis is that this approach does not determine anomeric conformation (alpha versus beta linkages). 4.2. Sensitivity to enzymatic digestion implies that polysaccharide accessibility and linkage of monosaccharides is altered in LmIFRD mutants The decreased sensitivity of the T-DNA mutant to a mixture of cell wall lysing enzymes is in striking contrast to the increased sensitivity of the silenced mutant and suggests these two mutants have different cell wall architecture to that of the wild type isolate. This may be due to the linkage or the accessibility of polysaccharides in the cell walls. The finding that the isolates had similar survival rates in either the cell wall lysis enzyme mixture or an endob1-3 glucanase implies that a b1-3 linked glucan is accessible to degradation, particularly in the silenced mutant. Alternatively, since the linkage specificity of the glucanase in the cell wall lysing enzyme mixture is unknown, the presence of glucanases of unknown specificity may also contribute to the increased sensitivity of the silenced mutant to the lysing enzyme mixture. The decreased sensitivity of the T-DNA mutant to the lysing enzyme mixture and endo-b1-3 glucanase may be due to the low abundance of glucose and high abundance of galactose. Since glucans are often the major polysaccharide of the cell wall of ascomycetes, the high abundance of galactans is likely to have a profound effect on the phenotype associated with the cell wall. Furthermore, the fact that the L. maculans mutants and the wild type were equally sensitive to an endo-chitinase, but that the T-DNA mutant was more sensitive to a mixture of endo- and exo-chitinase implies that the exo-chitinase activity is responsible for the reduced survival of the T-DNA mutant. Since exo-chitinases act on terminal glucosamine residues whereas endo-chitinases cleave at internal residues, this finding suggests that the T-DNA mutant has more

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

terminal glucosamine residues than wild type and the silenced mutant. 4.3. Cell walls of L. maculans spores and hyphae differ in monosaccharide analysis to those of other ascomycetes This study highlights differences between cell wall composition of L. maculans and ascomycetes such as A. fumigatus and S. cerevisiae and. For instance, L. maculans has a lower abundance of glucose than S. cerevisiae. While glucans comprise the major component (60%) of the cell walls of L. maculans, glucose in 1,3linkages in particular represents only 13–14% of the total monosaccharide content. This low proportion of 1,3-linked glucose is surprising since in many fungi, b-1,3-glucans forms a backbone to which other cell wall components are covalently attached (Bowman and Free, 2006). The remainder of the glucan component comprises mainly terminal, 1,4- and 1,6-linkages. The proportion of glucose in 1,4- and 1,6-linkages may be over-estimated due to the presence of residual glycogen contained in unbroken spores that were carried through the analysis. Indeed, staining of the same germinating spores that were used for monosaccharide linkage analysis with iodine showed some intensely stained black spores indicative of the presence of glycogen (data not shown). Another implication of this potential overestimation of glucose 1,4- and 1,4,6-linkages is that this may lead to an underestimation of glucose in 1,3-linkages in the samples. Cell wall remodeling is necessary for the elongation of hyphae and branching during mycelium development. When L. maculans spores germinate and hyphae elongate, monosaccharide content and linkages change significantly. Similar to our study, variation in the proportion of sugars (an increase in mannose and chitin and a decrease in glucose) was observed in the cell walls of another filamentous ascomycete, P. janczewskii, with increasing culture age (Pessoni et al., 2005). In contrast, a less comprehensive study of Aspergillus niger showed that chitin was the major polysaccharide in walls of mature spores whilst the mycelium contained more glucose (Feofilova et al., 2006). The implications of the changes in monosaccharide content of the cell wall of L. maculans are difficult to predict because little is known about how the components including proteins are organized. 4.4. Cell wall integrity plays a role in the virulence of L. maculans towards B. napus The pathogenicity phenotypes of the three isolates are consistent with their respective spore germination rates. Since the TDNA mutant germinates more slowly than the wild type isolate, the host plant may have time to activate a defense response leading to the inhibition of growth of the mutant isolate. Conversely, the earlier germination of the silenced mutant may allow this isolate to invade before the plant can mount an initial defense response. However, after 15 days the defense responses of the plant would impede growth of the silenced mutant and the wild type to a similar extent. An alternative scenario is that the altered polysaccharide structure, including chitin, results in changes to detection of the pathogen by the plant. 4.5. Potential role of ‘interferon-related developmental regulator’ proteins in fungi The precise mechanism of how mutations in LmIFRD lead to changes in cell wall integrity remains unknown. This gene has highest similarity to hypothetical proteins from other fungi that have an interferon-related developmental regulator (IFRD) domain. This domain has been described in rat pheochromocytoma cell-4 (PC4), mouse TPA induced Sequence 7 (TIS7) and human

705

TIS7 genes (Buanne et al., 1998; Guardavaccaro et al., 1994; Latif et al., 1997). The PC4/TIS7 proteins play regulatory roles in cell proliferation, differentiation and responses to stress (Vadivelu et al., 2004; Vietor and Huber, 2007). The best studied of these is mouse TIS7, which like LmIFRD has a bi-partite nuclear localization signal near the N terminus, but lacks a coiled-coil domain, which plays a role in protein–protein interactions. Deletion analyses, reporter gene assays and subsequent pull-down experiments have defined regions including some in the C-terminus of this protein involved in movement of the TIS7 protein from the cytoplasm to the nucleus upon particular stimuli (Micheli et al., 2005; Vietor et al., 2002). Also mice with a deletion of TIS7 upregulate a set of genes involved in tissue differentiation, whereas mice that over express TIS7 down-regulate the same gene set (Vietor and Huber, 2007). Like the IFRD proteins from higher organisms, LmIFRD appears to be involved in development. The finding that increases in silencing in the four RNAi silenced mutants correlated with increases in germination rate, strongly supports the role of this protein in germination. This key developmental process involves regulation of production of cell wall polysaccharides, such as chitin. The LmIFRD protein is unlikely to regulate this directly through chitin synthases as wild type and LmIFRD mutants have similar expression levels of seven chitin synthase genes. An alternative scenario is that changes in monosaccharide content may be an indirect response due to the involvement of IFRD in regulating genes involved in the biosynthesis of other cell wall components. It is surprising why a truncation of a gene (T-DNA mutant) has such an opposite phenotype to a reduction in gene expression, and also that the T-DNA mutant has such a different phenotype to the wild type, given that both produce similar levels of transcripts. Since the T-DNA mutant does not have the same phenotype as the silenced isolate it suggests that the truncated transcript is translated into a stable protein. This truncation at the C-terminus may be responsible for the different phenotype as in mouse TIS7, the C terminal region is involved in movement of the protein from cytoplasm to nucleus, and subsequent repression of a specific set of genes involved in differentiation (Vietor and Huber, 2007). Although the truncated LmIFRD protein of the T-DNA mutant has an N-terminal nuclear localization signal and a coiled-coil domain, C-terminal domains required for nuclear localization and gene regulation may be missing In summary, the study of LmIFRD mutants has revealed new information about the structure of the cell wall of a plant pathogenic dothideomycete and has identified a novel gene involved in maintaining cell wall integrity. Understanding the role of LmIFRD is key to a thorough understanding of cell wall integrity and may lead to the discovery of new fungicide targets for the control of such plant pathogens. Acknowledgments This work was supported by the Grains Research and Development Corporation, Australia and the Australian Research Council. We thank Thierry Rouxel and Marie-Hélène Balesdent, INRA and Patrick Wincker, Genoscope, and Joelle Anselem, URGI, France for access to the L. maculans sequence. We also thank Dr. Sarah Wilson, University of Melbourne, for help with iodine staining of germinating spores. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.fgb.2009.06.004.

706

A.P. Van de Wouw et al. / Fungal Genetics and Biology 46 (2009) 695–706

References Bartlett, D.W., Clough, J.M., Godwin, J.R., Hall, A.A., Hamer, M., Parr-Dobrzanski, B., 2002. The strobilurin fungicides. Pest. Manag. Sci. 58, 649–662. Bowman, S.M., Free, S.J., 2006. The structure and synthesis of the fungal cell wall. Bioessays 28, 799–808. Buanne, P., Incerti, B., Guardavaccaro, D., Avvantaggiato, V., Simeone, A., Tirone, F., 1998. Cloning of the human interferon-related developmental regulator (IFRD1) gene coding for the PC4 protein, a member of a novel family of developmentally regulated genes. Genomics 51, 233–242. Ciucanu, I., Kerek, F., 1984. A simple and rapid method for the permethylation of carbohydrates. Carbohyd. Res. 131, 209–217. Covert, S.F., Kapoor, P., Lee, M.H., Briley, A., Nairn, C.J., 2001. Agrobacterium tumefaciens-mediated transformation of Fusarium circinatum. Mycol. Res. 105, 259–264. Cozijnsen, A.J., Howlett, B.J., 2003. Characterisation of the mating-type locus of the plant pathogenic ascomycete Leptosphaeria maculans. Curr. Genet. 43, 351–357. Cozijnsen, A.J., Popa, K.M., Purwantara, A., Rolls, B.D., Howlett, B.J., 2000. Genome analysis of the plant pathogenic ascomycete Leptosphaeria maculans; mapping mating type and host specificity loci. Mol. Plant Pathol. 1, 293–302. Feofilova, E.P., Nemtsev, D.V., Tereschina, V.M., Memorskaya, A.S., 2006. Developmental change of the composition and content of the chitin–glucan complex in the fungus Aspergillus niger. Appl. Biochem. Microbiol. 42, 545–549. Fitt, B.D.L., Brun, H., Barbetti, M.J., Rimmer, S.R., 2006. World-wide importance of phoma stem canker (Leptosphaeria maculans and L. Biglobosa) on oilseed rape (Brassica napus). Eur. J. Plant Pathol. 114, 3–15. Fortwendel, J.R., Fuller, K.K., Stephens, T.J., Bacon, W.C., Askew, D.S., Rhodes, J.C., 2008. Aspergillus fumigatus RasA regulates asexual development and cell wall integrity. Eukaryot. Cell 7, 1530–1539. Fox, E.M., Gardiner, D.M., Keller, N.P., Howlett, B.J., 2008. A Zn(II)2Cys6 DNA binding protein regulates the sirodesmin PL biosynthetic gene cluster in Leptosphaeria maculans. Fungal Genet. Biol. 45, 671–682. Gardiner, D.M., Howlett, B.J., 2004. Negative selection using thymidine kinase increases the efficiency of recovery of transformants with targeted genes in the filamentous fungus Leptosphaeria maculans. Curr. Genet. 45, 249–255. Gardiner, D.M., Cozijnsen, A.J., Wilson, L.M., Pedras, M.S., Howlett, B.J., 2004. The sirodesmin biosynthetic gene cluster of the plant pathogenic fungus Leptosphaeria maculans. Mol. Microbiol. 53, 1307–1318. Gaughran, J.P., Lai, M.H., Kirsch, D.R., Silverman, S.J., 1994. Nikkomycin Z is a specific inhibitor of Saccharomyces cerevisiae chitin synthase isozyme Chs3 in vitro and in vivo. J. Bacteriol. 176, 5857–5860. Guardavaccaro, D., Montagnoli, A., Ciotti, M.T., Gatti, A., Lotti, L., Di Lazzaro, C., Torrisi, M.R., Tirone, F., 1994. Nerve growth factor regulates the subcellular localization of the nerve growth factor-inducible protein PC4 in PC12 cells. J. Neurosci. Res. 37, 660–674. Guest, G.M., Momany, M., 2000. Analysis of cell wall sugars in the pathogen Aspergillus fumigatus and the saprophyte Aspergillus nidulans. Mycologia 92, 1047–1050. Hammond, K.E., Lewis, B.G., 1987. The establishment of systemic infection in leaves of oilseed rape by Leptosphaeria maculans. Plant Pathol. 36, 135–147. Hill, T.W., Loprete, D.M., Momany, M., Ha, Y., Harsch, L.M., Livesay, J.A., Mirchandani, A., Murdock, J.J., Vaughan, M.J., Watt, M.B., 2006. Isolation of cell wall mutants in Aspergillus nidulans by screening for hypersensitivity to Calcofluor White. Mycologia 98, 399–409. Jeon, J., Goh, J., Yoo, S., Chi, M.H., Choi, J., Rho, H.S., Park, J., Han, S.S., Kim, B.R., Park, S.Y., Kim, S., Lee, Y.H., 2008. A putative MAP kinase kinase kinase, MCK1, is required for cell wall integrity and pathogenicity of the rice blast fungus, Magnaporthe oryzae. Mol. Plant Microbe Interact. 21, 525–534. Keogh, R.C., Deverall, B.J., McLeod, S., 1980. Comparison of histological and physiological-responses to Phakopsora-pachyrhizi in resistant and susceptible soybean. Trans. Brit. Mycol. Soc. 74, 329–333. Koch, E., Song, K., Osborn, T.A., Williams, P.H., 1991. Relationship between pathogenicity and phylogeny based on restriction-fragment-length-polymorphism in Leptosphaeria maculans. Mol. Plant Microbe Interact. 4, 341–349. Kopecka, M., Gabriel, M., 1992. The influence of congo red on the cell wall and (1-3)beta-D-glucan microfibril biogenesis in Saccharomyces cerevisiae. Arch. Microbiol. 158, 115–126. Latgé, J.P., 2007. The cell wall: a carbohydrate armour for the fungal cell. Mol. Microbiol. 66, 279–290. Latif, F., Duh, F.M., Bader, S., Sekido, Y., Li, H., Geil, L., Zbar, B., Minna, J.D., Lerman, M.I., 1997. The human homolog of the rodent immediate early response genes, PC4 and TIS7, resides in the lung cancer tumor suppressor gene region on chromosome 3p21. Hum. Genet. 99, 334–341. Lussier, M., White, A.M., Sheraton, J., diPaolo, T., Treadwell, J., Southard, S.B., Horenstein, C.I., ChenWeiner, J., Ram, A.F.J., Kapteyn, J.C., Roemer, T.W., Vo, D.H.,

Bondoc, D.C., Hall, J., Zhong, W.W., Sdicu, A.M., Davies, J., Klis, F.M., Robbins, P.W., Bussey, H., 1997. Large scale identification of genes involved in cell surface biosynthesis and architecture in Saccharomyces cerevisiae. Genetics 147, 435– 450. Martinez-Rocha, A.L., Roncero, M.I., Lopez-Ramirez, A., Marine, M., Guarro, J., Martinez-Cadena, G., Di Pietro, A., 2008. Rho1 has distinct functions in morphogenesis, cell wall biosynthesis and virulence of Fusarium oxysporum. Cell. Microbiol. 10, 1339–1351. Micheli, L., Leonardi, L., Conti, F., Buanne, P., Canu, N., Caruso, M., Tirone, F., 2005. PC4 coactivates MyoD by relieving the histone deacetylase 4-mediated inhibition of myocyte enhancer factor 2C. Mol. Cell. Biol. 25, 2242–2259. Mullins, E.D., Chen, X., Romaine, P., Raina, R., Geiser, D.M., Kang, S., 2001. Agrobacterium-mediated transformation of Fusarium oxysporum: an efficient tool for insertional mutagenesis and gene transfer. Phytopathology 91, 173– 180. Newton, A.C., Caten, C.E., 1988. Auxotrophic mutants of Septoria nodorum isolated by direct screening and by selection for resistance to chlorate. Trans. Brit. Mycol. Soc. 90, 199–207. Odenbach, D., Breth, B., Thines, E., Weber, R.W., Anke, H., Foster, A.J., 2007. The transcription factor Con7p is a central regulator of infection-related morphogenesis in the rice blast fungus Magnaporthe grisea. Mol. Microbiol. 64, 293–307. Pessoni, R.A.B., Freshour, G., Figueiredo-Ribeiro, R.D.L., Hahn, M.G., Braga, M.R., 2005. Cell-wall structure and composition of Penicillium janczewskii as affected by inulin. Mycologia 97, 304–311. Purwantara, A., Burton, W.A., Salisbury, P.A., Howlett, B.J., 1998. Reaction of Brassica juncea (Indian mustard) lines to Australian isolates of Leptosphaeria maculans under glasshouse and field conditions. Eur. J. Plant Pathol. 104, 895–902. Ram, A.F.J., Klis, F.M., 2006. Identification of fungal cell wall mutants using susceptibility assays based on Calcofluor white and Congo red. Nat. Protoc. 1, 2253–2256. Ram, A.F., Wolters, A., Ten Hoopen, R., Klis, F.M., 1994. A new approach for isolating cell wall mutants in Saccharomyces cerevisiae by screening for hypersensitivity to calcofluor white. Yeast 10, 1019–1030. Romano, J., Nimrod, G., Ben-Tal, N., Shadkchan, Y., Baruch, K., Sharon, H., Osherov, N., 2006. Disruption of the Aspergillus fumigatus ECM33 homologue results in rapid conidial germination, antifungal resistance and hypervirulence. Microbiol.-SGM 152, 1919–1928. Sexton, A.C., Howlett, B.J., 2000. Characterisation of a cyanide hydratase gene in the phytopathogenic fungus Leptosphaeria maculans. Mol. Gen. Genet. 263, 463– 470. Sims, I.M., Bacic, A., 1995. Extracellular polysaccharides from suspension cultures of Nicotiana plumbaginifolia. Phytochemistry 38, 1397–1405. Soulié, M.C., Perino, C., Piffeteau, A., Choquer, M., Malfatti, P., Cimerman, A., Kunz, C., Boccara, M., Vidal-Cros, A., 2006. Botrytis cinerea virulence is drastically reduced after disruption of chitin synthase class III gene (Bcchs3a). Cell. Microbiol. 8, 1310–1321. Specht, C.A., Liu, Y., Robbins, P.W., Bulawa, C.E., Iartchouk, N., Winter, K.R., Riggle, P.J., Rhodes, J.C., Dodge, C.L., Culp, D.W., Borgia, P.T., 1996. The chsD and chsE genes of Aspergillus nidulans and their roles in chitin synthesis. Fungal Genet. Biol. 20, 153–167. Vadivelu, S.K., Kurzbauer, R., Dieplinger, B., Zweyer, M., Schafer, R., Wernig, A., Vietor, I., Huber, L.A., 2004. Muscle regeneration and myogenic differentiation defects in mice lacking TIS7. Mol. Cell. Biol. 24, 3514–3525. Valiante, V., Heinekamp, T., Jain, R., Hartl, A., Brakhage, A.A., 2008. The mitogen-activated protein kinase MpkA of Aspergillus fumigatus regulates cell wall signalling and oxidative stress response. Fungal Genet. Biol. 45, 618–627. Vietor, I., Huber, L.A., 2007. Role of TIS7 family of transcriptional regulators in differentiation and regeneration. Differentiation 75, 891–897. Vietor, I., Vadivelu, S.K., Wick, N., Hoffman, R., Cotten, M., Seiser, C., Fialka, I., Wunderlich, W., Haase, A., Korinkova, G., Brosch, G., Huber, L.A., 2002. TIS7 interacts with the mammalian SIN3 histone deacetylase complex in epithelial cells. EMBO J. 21, 4621–4631. Wedlich-Soldner, R., Bolker, M., Kahmann, R., Steinberg, G., 2000. A putative endosomal t-SNARE links exo- and endocytosis in the phytopathogenic fungus Ustilago maydis. EMBO J. 19, 1974–1986. Werner, S., Sugui, J.A., Steinberg, G., Deising, H.B., 2007. A chitin synthase with a myosin-like motor domain is essential for hyphal growth, appressorium differentiation, and pathogenicity of the maize anthracnose fungus Colletotrichum graminicola. Mol. Plant Microbe Interact. 20, 1555–1567. Wood, P.J., 1980. Specificity in the interaction of direct dyes with polysaccharides. Carbohyd. Res. 85, 271–287.