Cloning and characterisation of glutamine synthetase from ...

2 downloads 0 Views 994KB Size Report
Cloning and characterisation of glutamine synthetase from. Colletotrichum gloeosporioides and demonstration of elevated expression during pathogenesis on ...
Curr Genet (1997) 31: 447 – 454

© Springer-Verlag 1997

O R I G I N A L PA P E R

Sally-Anne Stephenson · Jonathan R. Green John M. Manners · Donald J. Maclean

Cloning and characterisation of glutamine synthetase from Colletotrichum gloeosporioides and demonstration of elevated expression during pathogenesis on Stylosanthes guianensis Received: 14 November 1996 / 22 January 1997

Abstract Experiments were designed to clone and identify genes of the fungal phytopathogen Colletotrichum gloeosporioides expressed at high levels during growth on the compatible host Stylosanthes guianensis when compared with expression in axenic culture. A cDNA clone (pCgGS) that hybridised preferentially to a cDNA probe prepared from infected leaves was isolated by the differential screening of a cDNA library from a nitrogen-starved axenic culture of C. gloeosporioides. The DNA sequence of pCgGS is highly homologous to genes for glutamine synthetase (GS) in other organisms. pCgGS contained all of the conserved regions assigned as catalytic domains in GS enzymes. Comparison with genomic sequences indicated that in C. gloeosporioides the GS gene is present as a single copy with three introns. To our knowledge this is the first report of the cloning of a GS from a filamentous fungus. A second clone (pCgRL1) was also isolated and represented a partial cDNA of the 25s rRNA of C. gloeosporioides. Because pCgRL1 did not hybridise to plant rRNA under high-stringency hybridisation conditions, it was used as a reference to quantify the expression of fungal GS mRNA during pathogenesis in S. guianensis compared to fungal growth in axenic culture. The results indicated that elevated expression of GS occurred during pathogenesis of C. gloeosporioides on S. guianensis, particularly at early stages of infection where expression was about six-times higher than during growth in rich culture media. This work also demonstrates that fungal-specific 25s rRNA fragments, such as pCgRL1, have considerable utility as a reference for quantifying pathogen gene expression in infected plants. S.-A. Stephenson · J. M. Manners (½) · D. J. Maclean Cooperative Research Centre for Tropical Plant Pathology, John Hines Building, The University of Queensland, Brisbane 4072, Australia J. R. Green School of Biological Sciences, The University of Birmingham, P.O. Box 363, Birmingham B15 2TT, UK Communicated by P. de Wit

Key words Expression quantification · Glutamine synthetase · Pathogenesis · Nitrogen metabolism

Introduction

The ability of a fungal pathogen to infect a host plant depends on its capacity to obtain adequate nutrients while avoiding or negating any defense responses. Genes that are either induced or highly expressed in the pathogen during the infection process may play a direct role in pathogenesis, and hence contribute to events conferring compatibility in host-pathogen interactions (Pieterse et al. 1993; Hawthorne et al. 1994; Schäfer 1994). Pathogen genes induced in planta may produce products required for pathogenesis processes that suppress the host defences, re-direct nutrient supplies to the pathogen, and control essential adaptations to the nutritional environment encountered by the pathogen. Identification of such fungal genes and the elucidation of the regulatory processes necessary for their expression during successful pathogenesis may reveal new information that can be targeted for the development of durable strategies of disease management (Valent and Chumley 1991; Schäfer 1994; Oliver and Osbourne 1995). Anthracnose caused by Colletotrichum gloeosporioides (Penz.) Penz. and Sacc. is an important disease of many tropical plants and restricts the utilisation of the tropical forage legume Stylosanthes guianensis and other Stylosanthes spp. in Australia, South-East Asia, Africa and South America (Manners et al. 1992; Lenne 1994). The infection processes of C. gloeosporioides on S. guianensis have been well studied histologically (Trevorrow et al. 1988; Ogle et al. 1990). In compatible interactions of biotype B of C. gloeosporioides on S. guianensis, conidia germinate within 12 h to give melanized appressoria, followed by penetration of the cuticle by means of a penetration peg within 24 h after inoculation. After infecting two to four adjacent epidermal cells, hyphae begin to spread intra- and inter-cellularly into the mesophyll. Collapse of the meso-

448

phyll is rapid, and fungal acervuli are present 96–120 h after inoculation. Hyphae also emerge from the epidermal cells to grow sub-cuticularly, or more often superficially, initiating new infections without the formation of appressoria. This rapid fungal growth associated with extensive blight of the host tissue represents the symptoms characteristic of Type-B anthracnose disease on S. guianensis (Trevorrow et al. 1988; Ogle et al. 1990). Little, however, is known about the physiological and molecular processes that occur in this fungus during various stages of infection. The aim of our current research is to isolate genes which are either induced, or show elevated expression, in biotype B of C. gloeosporioides during infection of S. guianensis, with the expectation that some of these genes will provide clues to the developmental, metabolic and pathogenic processes that are essential for infection to be successful. Several cDNA clones, that showed strong hybridisation to cDNA probes prepared from total RNA from infected leaves when compared to mycelium grown on rich medium, have been isolated. In this report we discuss the characterisation of one of these clones (pCgGS) which has been identified as coding for glutamine synthetase (GS). To our knowledge this is the first report of the cloning of a GS gene in a filamentous fungus. The availability of this clone permitted a direct analysis of the regulation of expression of the GS gene during pathogenesis. To compare gene expression of a fungal phytopathogen in planta relative to mycelia grown in axenic culture, it is necessary to find a common standard that can be quantified in both samples. Recently, for example, fungal DNA has been used to estimate fungal biomass in infected plants (Talbot et al. 1993), and DNA probes for constitutively expressed fungal genes encoding actin and EF-1α (Pieterse et al. 1993) have been used as a reference to measure fungus-specific gene expression in infected plants. Alternatively, ribosomal transcripts are commonly used in Northern-blot analysis as a reference to quantify relative levels of gene expression. Ribosomal DNA probes (rDNA) are convenient to use because they detect the highly abundant rRNA species. Because of the close homology between plant and fungal rRNA sequences, full-length clones of fungal rRNA genes are unsuitable to distinguish fungal from plant rRNA. In the present paper, we demonstrate that a small non-conserved region of the 25s rRNA of C. gloeosporioides can be used as a sensitive hybridisation probe to measure fungal material in infected leaves of the host plant S. guianensis. This type of fungal-specific rRNA probe should find wide application in assessing the expression of genes in fungal pathogens during infection of their hosts.

Materials and methods Fungal isolates and culture conditions. Isolates UQ62 and 21808 of biotype B of C. gloeosporioides (Manners et al. 1992; Maclean et al. 1993) from the culture collection of The University of Queensland

were used for this study. Cultures derived from single conidia were maintained on oatmeal agar supplemented with 30 µg ml–1 of tetracycline and 100 µg ml–1 of streptomycin under near UV light at 28oC. Conidia (5 × 107) were used to inoculate 200 ml of a rich complete medium termed DEF, defined by the following ingredients, Czapek’s minerals (Johnston and Booth 1983), 10 mM Na3Citrate, 1% (w/v) Casamino acids (Difco Ltd.) and 2% sucrose, and cultures were incubated at 28°C with shaking. The mycelium was harvested by filtration through Miracloth (Calbiochem), washed with sterile water and then either transferred to fresh media or stored under liquid nitrogen. A medium lacking organic and inorganic nitrogen compounds (prepared by omitting both nitrate and casamino acids) is referred to as D-N medium, and a medium lacking all carbon compounds is referred to as D-C medium. Infection experiments. Plants of S. guianensis cv Graham were grown in controlled environment chambers at 28°C with a 14-h light period. Infections were carried out using 6-week-old plants which were sprayed with freshly prepared spore suspensions (107 conidia ml–1 0.1% Tween 20) of C. gloeosporioides strain UQ62 which is virulent on cv Graham. Following inoculation, plants were incubated inside plastic bags at 25°C without light for 2 days and then transferred to normal growth conditions. At various times after inoculation, the leaves and stems of the infected plants were removed and stored under liquid nitrogen until required for RNA extraction. Northern- and Southern-blot analysis. DNA was isolated from C. gloeosporioides as described by Yoon et al. (1991). DNA was isolated from S. guianensis as described by Curtis et al. (1995). Southern blots were prepared using Hybond N+ (Amersham) and 10 µg of genomic DNA (Sambrook et al. 1989). Total RNA (20 µg) from infected plant material at various days after inoculation, and fungal mycelia grown under the conditions specified in the results section, was isolated and analysed by gel electrophoresis and Northern hybridisation as described by Higgins et al. (1985). Hybridisations were carried out at high stringency using cDNA inserts (prepared by PCR as described below) labelled with 32P-dCTP using the Amersham Megaprime labelling reaction. The relative intensities of bands were compared quantitatively using a Molecular Diagnostics Phosphorimager. Filters were hybridised again, following removal of the previous radioactive probe, with boiling 0.1% SDS for 5 mins. cDNA library construction and differential screening. Total RNA was extracted from isolate 21808 mycelia grown in rich DEF medium for 4 days followed by incubation in D-N medium for a further 4 days. Poly(A)+ RNA, prepared from total RNA using the Pharmacia mRNA purification kit according to the manufacturer’s instructions, was used to prepare a cDNA library in λgt10 using the Amersham cDNA Synthesis System Plus, cDNA Rapid Adaptor Ligation Module and cDNA Rapid Cloning Module for λgt10. The library was termed the D-N cDNA library and comprised 106 plaque-forming units. Duplicate colony blots of random clones from the D-N cDNA library were differentially screened, with either first-strand cDNA prepared using poly(A)+ RNA from S. guianensis leaves 2 days after inoculation with C. gloeosporioides isolate UQ62, or first-strand cDNA from poly(A)+ RNA isolated from UQ62 mycelia grown in DEF medium. The cDNA probe was prepared as follows: 1 µg of poly(A)+ RNA was annealed to random hexamers (30 µg ml–1) in 5 µl and added to 15 µl of reaction mix containing 20 units of AMV reverse transcriptase (Boehringer), 500 µM of dATP, dGTP and dTTP, RNase inhibitor (0.1 mg ml–1 ), BSA (100 µg ml–1) and 5 µCi of 32 P-dCTP (Amersham), and the mixture was incubated for 1 h at 37 °C. The labelled probes were purified through Bio-spin 30 chromatography columns (Biorad) and denatured at 100°C for 5 min. Duplicate filter-lifts of the cDNA library (approximately 5000 plaqueforming units) were pre-hybridised for 4 h at 65°C in 5 × SSPE (1 × SSPE contains 0.15 M NaCl, 10 mM sodium phosphate, 1 mM EDTA, pH 7.4), 0.5% SDS, 5×Denhardt’s Solution [100×Denhardt’s Solution is 2%(w/v) of each of Ficoll 400, polyvinylpyrollidone and Bovine Serum Albumin], 20 µg ml–1 of sheared denatured salmonsperm DNA (Sigma) and 10 µg ml–1 of polyadenylic acid (Sigma). Hybridisations with the labelled cDNA probes were carried out under

449 the same conditions as the pre-hybridisation for 16–24 h. The filters were then washed under high-stringency conditions (65°C for 15 mins in 0.1% SDS, 0.1%SSPE) and exposed to X-OMAT film (Kodak). Clones from the D-N cDNA library that hybridised preferentially to the cDNA probe from infected leaves, compared to the cDNA probe from DEF medium, were selected for further analysis. Subcloning and sequencing. cDNA inserts were isolated from λgt10 using PCR as described in the Amersham λgt10 cloning manual, purified using QIAquick PCR spin columns (Qiagen) and cloned into pGEM-T (Promega) according to the manufacturer’s instructions. Ligations were transformed by electroporation into E. coli strain DH5α and transformants selected by plating on MacKonkey’s Agar (DIFCO) supplemented with 100 µg ml-1 of ampicillin. Plasmid DNA was prepared using QIAprep spin columns (Qiagen) and was used as a template to sequence both strands of the inserts using an automated Applied Biosystems 373A DNA Sequencer and the ABI PRISM DyeDeoxy Terminator Cycle Sequencing Kit. Internal sequencing primers were synthesised on a Beckman Oligo 100 instrument, and M13 forward and reverse primers were used. Preparation of a genomic clone by PCR. Primers designed from the termini of the cDNA insert were used in a PCR reaction with genomic DNA as the template. Reaction conditions were as follows: 4 mM of MgCl2, 200 µM of each deoxyribonucleoside triphosphate (dNTP), 10 ng of each primer, 10 ng of genomic DNA and 1.3 U of Tth Plus DNA polymerase in 1 × Buffer (Biotech International, Australia). The amplification was performed in a PTC-100™ Thermal Cycler (MJ Research, Inc.) which, after an initial denaturation at 94°C for 5 min, included 30 cycles of reaction with 15-s denaturing at 94°C, 15-s annealing at 60°C and 1-min extension at 72°C. PCR products were cloned and sequenced as previously stated.

Genomic organisation of the glutamine synthetase gene Genomic DNA from biotype B of C. gloeosporioides was digested with either BamHI, EcoRI, EcoRV or HindIII and resolved by gel electrophoresis. A Southern blot of this gel was probed with the partial cDNA for glutamine synthetase, pINF2. The simple banding pattern observed is consistent with glutamine synthetase being encoded by a single-copy gene (Fig. 1). Similar experiments indicated that there was no cross hybridisation with DNA isolated from the host plant S. guianensis (data not shown). Pulsed-field gel electrophoresis of C. gloeosporioides chromosomes showed that pCgGS hybridised to a single large (> 6 Mb) chromosome band that was poorly resolved from other large chromosomes (data not shown). To assess the arrangement of introns within the GS gene, PCR primers were designed from the pCgGS sequence and used to amplify a fragment from genomic DNA of isolate UQ62 of biotype B of C. gloeosporioides. A genomic fragment of 2206 bp was subsequently subcloned as pCgGS2 and sequenced. This sequence perfectly matched that of pCgGS and also indicated the presence of three introns. Each intron (see sequence annotation in database for positions) contained sequences that fitted the consensus for 5′ and 3′ splice sites and lariat formation in fungal introns (Ballance 1986). The nucleotide sequence of the GS gene derived from the genomic and cDNA clones of C. gloeosporioides can be accessed as Accession number L78067 at the Genome Sequence Database, Santa Fe.

Results

cDNA cloning of glutamine synthetase by differential hybridization Recent research on fungal phytopathogens has demonstrated that genes induced in planta may also be induced during nutrient starvation in vitro (Talbot et al. 1993; Pieterse et al. 1994; Van den Ackerveken et al. 1994). Based on this premise, total RNA was extracted from mycelia of C. gloeosporioides after incubation in a medium lacking a nitrogen source. The RNA was enriched for poly(A)+ and used to construct a cDNA library. This library was differentially screened with cDNA probes derived from poly(A)+ RNA from leaves of S. guianensis at 2 days after inoculation, and poly(A)+ RNA derived from mycelium growing in the complete DEF media. Seven distinct clones that hybridised preferentially to the cDNA probe of infected leaves were selected, subcloned into pGEM-T (Promega) and partially sequenced using M13 forward and reverse primers. The nucleotide and deduced amino-acid sequences of one clone showed strong homology to glutamine synthetase (GS) sequences from other organisms and is presumed to encode GS in C. gloeosporioides. This cDNA clone, named pINF2 (581 bp), was fully sequenced in both directions, then used to re-screen the cDNA library and a 1402-bp clone, part of which perfectly matched the sequence of pINF2, was selected for further analysis. This clone, named pCgGS, was also sequenced fully in both directions.

Fig. 1 Southern-hybridisation analysis of GS sequences in genomic DNA from biotype B of C. gloeosporioides. The DNA was digested with BamHI (B), EcoRI (EI), EcoRV (EV) and HindIII (H) and hybridised with the GS insert. The sizes of HindIII-digested bacteriophage λ standards (λ) are shown in kbp

450 Fig. 2 Multiple alignment of the deduced amino-acid sequence of GS from C. gloeosporioides (Colletotrichum) compared to GS sequences from other organisms using the Clustal V program (Higgins et al. 1992). The following organisms (with their GSDB accession numbers) were used; yeast (M65157); Drosophila (X52759); human (Y00387); alfalfa (U15591), Streptomyces (X52842) and Rhizobium (X65929). The four motifs thought to comprise the catalytic domains are underlined

Sequence analysis of the GS gene We deduced a large open reading frame (orf) of 1080 nucleotides in pCgGS, and the sequence around the putative start codon (CAAAAUGGC) was identical to the consensus for filamentous fungi described by Ballance (1986). Translation from this site yielded a deduced amino-acid (aa) sequence that showed significant homology with GS amino-acid sequences from a variety of diverse prokaryotic and eukaryotic organisms (Fig. 2). The untranslated upstream sequence of 168 nucleotides was pyrimidine-rich with 63% being either C or T. There were 150 nucleotides of 3′ untranslated sequence including the potential polyadenylation sequences AAUAA and AAUUAAA at 73

and 14 nucleotides, respectively, before a 17-nucleotide poly-A tail. The deduced amino-acid sequence of the C. gloeosporioides glutamine synthetase derived from pCgGS was compared to 14 GS sequences previously reported for plants, animals, bacteria and yeast (Fig. 2 and data not shown). A very high overall homology of over 55% aa identity was shown to the yeast GS, with a mean of 44.2% identity to a group of animal GS sequences and 44% to a group of plant GS sequences. The deduced aa sequence of pCgGS includes all conserved and consensus aa residues present in four domains (Fig. 2) believed to contribute to the active site of the GS enzyme (Tateno 1994). On the basis of these sequence homologies, we conclude that the cDNA clone

451

Fig. 3 The DNA sequence of the clone pCgRL1, representing a partial sequence of the 25s rRNA gene from C. gloeosporioides. The figure shows a multiple alignment and dendrogram to indicate the homolgy of pCgRL1 to orthologous sequences of the 25s rRNA gene from other organisms. The following GSDB accession numbers were used for the comparison: Aspergillus nidulans (L21921); Saccharomyces cerevisiae (J01355); Fragaria ananassa (X58118); Sinapis alba (X66325); Oryza sativa (M11585); and Medicago sativa (Z11498). The alignment and dendrogram were made using the Clustal V program and the scale represents the percentage identity between sequences

pCgGS probably encodes a glutamine synthetase. The fluG gene of Aspergillus nidulans, believed to be involved in developmental signalling, has been reported to have weak (28%) homology to the prokaryotic glutamine synthetase-I group of enzymes (Lee and Adams 1994). However this gene product showed little homology to the pCgGS aa sequence and that of other eukaryotic glutamine synthetase-II enzymes. Lee and Adams (1994) suggested that the fluG gene is probably not involved in glutamine synthesis. Estimation of fungal 25s ribosomal RNA in planta A major aim of this investigation was to determine the expression of GS mRNA by the fungus growing in planta compared to growth in rich culture media. Ribosomal RNA is commonly used as a reference to compare the relative abundance of mRNA species in fungal cells grown under different conditions. To avoid interference from the host, we sought regions of rRNA genes that could distinguish C. gloeosporioides from S. guianensis. The full sequence of rRNA genes of these two organisms is not known. However, during this study, three fungal clones with homology to rDNA were isolated on the basis of preferential hybridisation to cDNA probes of fungal mycelia compared to that with a cDNA probe of infected leaves. These clones presumably were derived from ribosomal RNA contamination in the poly(A)+ sample used to construct the library. One of these clones, a 91-bp fragment termed pCgRL1, hybridised specifically to fungal, but not to plant, ribosomal RNA. The sequence of the pCgRL1 insert matched a region of the 25s rDNA gene of fungi that was quite divergent from that of higher plants (Fig. 3). Hybridisation of pCgRL1 to total RNA of S. guianensis and C. gloeosporioides RNA demonstrated specificity to the 25s rRNA of C. gloeosporioides and no hybridisation to RNA of S. guianensis

Fig. 4A–D Northern-hybridisation analysis of C. gloeosporioides grown in culture and in infected host leaves. Total RNA was obtained from C. gloeosporioides growing axenically in rich medium (DEF) and at 4 days after transfer to the same medium lacking either carbon-containing compounds (D-C) or all nitrogen sources (D-N). Total RNA was also obtained from leaves of S. guianensis cv Graham prior to inoculation (C) and at 1–7 days following inoculation with a virulent race of biotype B of C. gloeosporioides. The Northern blot was probed successively with the cDNA for GS (pCgGS), the fungal-specific 25s rRNA fragment of C. gloeosporioides (pCgRL1), a genomic fragment of the rDNA repeat unit from wheat (pTA71, Gerlach and Bedbrook 1979), and a genomic fragment of the rDNA repeat unit from A. nidulans (pAR1, Lockington et al. 1982), as indicated in the figure

452 Table 1 The abundance of fungal GS transcripts relative to fungal 25s rRNA in total RNA samples of C. gloeosporioides growing in culture or in leaves of S. guianensis. Total RNA samples (20 µg) were analysed by Northern hybridization using the inserts from pCgGS (detects fungal GS mRNA) or pCgRL1 (detects fungal 25s rRNA) as probes as described for Figure 6. The radioactivity on blots after hybridization and washing was estimated using a Phosphorimager. For both of the hybridization probes, the radioactivity in the detected band for each mycelial or leaf sample was normalised relative to the value obtained for the sample growing in complete medium (columns [a] and [b]). The abundance of GS mRNA in each sample relative to fungal 25s rRNA was then calculated as a ratio hybridisation to each probe, i.e. GS:25s rRNA (column [c]). The ratio of GS:25s rRNA for a second infection experiment is included in column [d]. Sample

Complete (defined) Defined-nitrogen Defined-carbon Control leaf Infected leaf day 1 Infected leaf day 2 Infected leaf day 3 Infected leaf day 4 Infected leaf day 5 Infected leaf day 6 Infected leaf day 7 a

a

b

c

d

pCgGS

pCgRL1

GS:25s rRNA 1

GS:25s rRNA 2

1 3.8 1.06 NDa NDa 0.396 0.710 0.834 0.678 0.624 0.577

1 0.75 1.01 NDa 0.04 0.061 0.227 0.311 0.324 0.310 0.298

1 5.07 1.05 – – 6.49 3.10 2.68 2.09 2.01 1.93

1 5.03 1.28 – – 6.60 – – 2.71 – –

No signal discernible above background

was detected (Fig. 4B). In contrast, total RNA from both C. gloeosporioides and S. guianensis showed characteristic hybridisation bands when probed with clones of either the entire rDNA repeat of wheat or Aspergillus nidulans (Fig. 4 C and 4D respectively). Using pCgRL1 as a probe to study the time-course of fungal rRNA gene expression during infection of S. guianensis by C. gloeosporioides, low levels of fungal 25s rRNA were detected 1 day after inoculation (Fig. 4 B, Table 1 column b), followed by a small increase at 2 days following inoculation and then a rapid rise over the next 24 h, reaching a plateau at about 4 days after inoculation. Quantitative values obtained by the phosphorimager (Table 1 columns a and b) are expressed relative to a value of 1.0 for the sample of RNA from fungus grown in DEF medium. Interestingly, there was very little change in the total abundance of the fungal 25s rRNA from 3 to 7 days after inoculation, during which time the blight symptoms and fungal reproductive structures develop.

bridisation of pCgGS to RNA was observed from plants at 1 day after inoculation (Fig. 4 A, Table 1 column a); however, measurable hybridisation to pCgGS was detected at 2 days after inoculation, and expression of the fungal GS appeared to peak at 4 days. Quantitative comparisons of hybridisation signals from both probes pCgGS and pCgRL1, using a phosphorimager, allowed us to calculate the relative abundance of GS transcripts compared with 25s rRNA (Table 1 column c). In turn, this allowed us to compare quantitatively the relative abundance of GS mRNA early in infection with that after fungal growth in rich culture medium, and hence determine whether the GS gene demonstrated altered expression during infection. The results indicated that fungal GS transcripts were sixtimes more abundant in fungus growing in planta at 2 days after inoculation than in mycelium growing in the rich DEF medium. Indeed, relative to 25s rRNA, fungal GS transcripts were more abundant than that observed in DEF medium at all time points sampled, although this apparent elevated expression of GS mRNA appeared to decline as infection progressed (Table 1 column c). This experiment was repeated for key time points with RNA samples obtained independently on another occasion, and yielded essentially identical results (Table 1 column d). The relative abundance of GS transcripts in mycelium subjected to nutrient starvation was also estimated (Fig. 4, Table 1). After 4 days of nitrogen starvation, GS expression was increased five-fold compared to the value observed for mycelium growing in the rich DEF medium. Deprivation of carbon appeared to make no appreciable difference to the abundance of GS transcripts compared to the rich DEF medium (Fig. 4, Table 1). A time-course experiment was undertaken to determine how rapidly fungal GS mRNA induction occurred following the removal of exogenous nitrogen compounds in axenic culture. The mycelium of a 4-day-old culture grown in DEF media was filtered and washed, then half-transferred back to DEF media with the other half transferred to DEF media lacking nitrogen compounds. The amount of GS transcript in these samples at different times after transfer was determined by quantitative Northern blots as described above. The results (data not shown) indicated that the amounts of GS mRNA in the nitrogen-starved mycelium were 0.9-, 0.9-, 1.2- and 2.0-fold those of the mycelium growing in rich medium at 4, 10, 48 and 90 h after transfer, respectively. These results suggest that, for mycelium grown under the conditions of this experiment, GS mRNA was not rapidly induced by depletion of exogenous nitrogen compounds but was slowly induced by prolonged nitrogen starvation.

Comparative abundance of fungal GS transcripts in planta and in culture Discussion

The abundance of GS transcripts of C. gloeosporioides was determined by Northern-blot analysis using total RNA samples from: (1) control and inoculated leaves, and (2) from the fungus growing in DEF medium and following incubation in D-N and D-C media. No detectable hy-

In this investigation we describe the cloning and expression of a gene encoding glutamine synthetase (GS) in an isolate of biotype B of the fungus C. gloeosporioides, a phytopathogen of the legume S. guianensis. The deduced

453

aa sequence of the GS cDNA clone of C. gloeosporioides was highly homologous to aa sequences of GS from other organisms, with 55% identity to yeast and about 44% identity to other eukaryotes, and included all four conserved domains believed to contribute to the active site of GS enzymes of both prokaryotes and eukaryotes (Tateno 1994). Despite intensive research on the regulation of activity of the enzyme glutamine synthetase in filamentous fungi (Dávila et al. 1978; Vichido et al. 1978; Mora et al. 1980; Dávila et al. 1983; Cornwell and MacDonald 1984; Muñoz and Agosin 1993), to our knowledge, this is the first report of the molecular cloning of glutamine synthetase in any filamentous fungus. Southern analysis suggested that GS is encoded by a single-copy gene within the C. gloeosporioides genome. Biochemical studies of GS in fungi, principally A. nidulans, Neurospora crassa and Saccharomyces cerevisiae, have suggested that only one isoform is present (Sanchez et al. 1978; Dávila et al. 1983; Cornwell and MacDonald 1984; Minehart and Magasanik 1992), consistent with our evidence for a single copy of GS in C. gloeosporioides. Because of its essential role in cellular metabolism the GS gene is thought to be of ancient origin and has been used for evolutionary comparisons (Tateno 1994). The only other published fungal GS sequence of which we are aware is from the yeast S. cerevisiae, which clustered more closely with the GS of C. gloeosporioides than any other eukaryotic sequence (Fig. 2). Additional sequence analysis of the GS gene may be useful to further our understanding of phylogenetic relationships among fungi, and may assist clarification of phylogenetically complex group species such as C. gloeosporioides. To characterise the specific pattern of GS expression during the development of C. gloeosporioides in leaves of its host S. guianensis, it was necessary to use a readily quantified fungal reference that is unaffected by the presence of host material. We chose a fungal-specific DNA sequence pCgRL1, originating from a phylogenetically variable region of the 25s rRNA gene of C. gloeosporioides, as a hybridisation probe to detect and quantify fungal 25s rRNA levels in the presence of plant rRNA in infected leaves. pCgRL1 hybridised specifically to 25s rRNA derived from the large subunit (LSU) of fungal ribosomes, but not to rRNA from the host LSU, in Northern-blot analyses. In turn, this reference allowed us to compare the relative abundance of specific mRNAs in the fungus growing in planta and in culture. Previous research has used estimations of either fungal DNA, or the levels of expression of fungal housekeeping genes such as actin, to provide references against which fungal gene expression can be measured in infected leaves (Pieterse et al. 1993; Talbot et al. 1993). Fungal-specific rRNA probes, such as the clone pCgRL1 described herein, are more sensitive than probes for fungal actin mRNA. During infection of leaves of S. guianensis by C. gloeosporioides, levels of fungal 25s rRNA showed a rapid increase during the first 3 days after inoculation, and reached a maximum at 4–5 days. This rapid increase in fungal 25s rRNA, which presumably parallels the increase in fungal

biomass, coincides with the ramification of fungal hyphae through the leaf tissue and the development of superficial and sub-cuticular hyphae during the first 3 days (Ogle et al. 1990). Host-cell necrosis appears to commence during the 3rd day after inoculation and spreads rapidly afterwards (Irwin et al. 1984). Relative to the fungal 25s rRNA reference, however, GS transcripts were most abundant 2 days after inoculation, followed by a gradual decrease in abundance as infection progressed. These changes in fungal GS transcripts may result from changes in the nutritional environment encountered in the host during infection. Transcripts of GS of C. gloeosporioides were induced by nitrogen starvation in culture. Recent studies of a range of plant-fungus interactions suggest that nutrient deprivation in axenic culture may mimic growth conditions at early stages of the infection process (St Leger et al. 1992; Talbot et al. 1993; Jellito et al. 1994; Pieterse et al. 1994; Van den Ackerveken et al. 1994). These studies imply that some pathogen genes induced by nutrient starvation may play an important role in pathogenesis. Certainly, it is technically easier to isolate transcripts induced by nutrient starvation in axenic culture than those induced in planta where fungal and plant gene expression must be distinguished. It was for this reason that we commenced our investigations by differential screening of a cDNA library derived from poly A+ RNA of mycelium starved of nitrogen compounds. One disadvantage of our strategy is that it would only recover cDNAs of transcripts that are induced during both nitrogen starvation and pathogenesis and would not select clones of transcripts induced only by pathogenesis. We are currently investigating the importance in pathogenicity of other cDNAs isolated by the differential screening procedure described herein. The enzyme GS is found in all organisms and plays a key role in amino-acid metabolism. It catalyses the condensation of ammonium ion and glutamate to yield glutamine in a reaction which requires ATP. This reaction is essential for the synthesis of glutamine and provides a mechanism for ammonia re-assimilation and de-toxification. To thoroughly assess the role of GS in infection, it would be desirable to disrupt the gene to see if the host can supply all the glutamine required by the fungus during pathogenesis. Acknowledgements Sally Stephenson was supported by an Australian Postgraduate Award and a Cooperative Research Centre for Tropical Plant Pathology Postgraduate Scholarship. Jon Green is grateful for sabbatical funding from the Royal Society, London, the Department of Botany of the University of Queensland and the C.S.I.R.O. Division of Tropical Crops and Pastures. We are grateful to Dr. A. Masel for advice and assistance with cDNA library construction, to C. Stephens for assistance with Northern blots, and to J. Nourse for assistance with DNA sequence manipulations. We thank Prof. George Stewart for his interest and for comments on the manuscript and Dr. R. Birch for access to the Phosphorimager.

References Ballance DJ (1986) Sequences important for gene expression in filamentous fungi. Yeast 2:229–236

454 Cornwell EV, MacDonald DW (1984) glnA mutations define the structural gene for glutamine synthetase in Aspergillus. Curr Genet 8:33–36 Curtis MD, Cameron DF, Manners JM (1995) Molecular evidence that diploid Stylosanthes humilis and diploid Stylosanthes hamata are progenitors of allotetraploid Stylosanthes hamata. Genome 38:344–348 Dávila G, Sanchez F, Palacios R, Mora J (1978) Genetics and physiology of Neurospora crassa glutamine auxotrophs. J Bacteriol 134:693–698 Dávila G, Brom S, Mora Y, Palacios R, Mora J (1983) Genetic and biochemical characterisation of glutamine synthetase from Neurospora crassa glutamine auxotrophs and their revertants. J Bacteriol 156:993–1000 Gerlach WL, Bedbrook JR (1979) Cloning and characterisation of ribosomal RNA genes from wheat and barley. Nucleic Acids Res 7:1869–1885 Hawthorne BT, Ball RD, Rees-George J (1994) Genetic analysis of variation of pathogenicity in Nectria haematococca (Fusarium solani) on Cucurbita sp. Mycol Res 98:1183–1191 Higgins C, Manners JM, Scott KJ (1985) Decrease in three messenger RNA species coding for chloroplast proteins in leaves of barley infected with Erysiphae graminis f.sp. hordeii. Plant Physiol 78:891–894 Higgins DG, Bleasby AJ, Fuchs R (1992) ClustalV: improved software for multiple sequence alignment. Comput Appl Biosci 8:189–191 Irwin JAG, Trevorrow PR, Cameron DF (1984) Histopathology of compatible interactions involving biotypes of Colletotrichum gloeosporioides that cause anthracnose of Stylosanthes spp. Aust J Bot 32:631–640 Jelitto TC, Page HA, Read ND (1994) Role of external signals in regulating the pre-penetration phase of infection by the rice blast fungus, Magnaporthe grisea. Planta 194:471–477 Johnston A, Booth C (1983) Plant pathologist’s pocketbook 2nd edn. The Cambrian News Ltd.,Wales Lee BN, Adams TH (1994) The Aspergillus nidulans fluG gene is required for production of an extracellular developmental signal and is related to prokaryotic glutamine synthetase I. Genes Dev 8:641–651 Lenne JM (1994) Diseases of Stylosanthes. In: Lenne JM, Trumann P (eds) Diseases of tropical plants. CAB International, UK, pp 21–42 Lockington RA, Taylor GG, Winther M, Scazzocchio C, Davies W (1982) A physical map of the ribosomal DNA repeat of Aspergillus nidulans. Gene 20:135–137 Maclean DJ, Braithwaite KS, Manners JM, Irwin JAG (1993) How do we identify and classify fungal plant pathogens in the era of DNA analysis? In: Andrews JH, Tommerup IC (eds) Adv Plant Pathol 10:207–244 Manners JM, Masel AM, Braithwaite KS, Irwin JAG (1992) Molecular analysis of Colletotrichum gloeosporioides pathogenic on the tropical pasture legume Stylosanthes. In: Bailey A, Jeger MJ (eds) Colletotrichum biology, pathology and control, CAB International, Oxford, UK, pp 250–268 Minehart PL, Magasanik B (1992) Sequence of the GLN1 gene of Saccharomyces cerevisiae: role of the upstream region in the

regulation of glutamine synthetase expression. J Bacteriol 174: 1828–1836 Mora Y, Chávez J, Mora J (1980) Regulation of Neurospora crassa glutamine synthetase by the carbon and nitrogen source. J Gen Microbiol 118:455–463 Muñoz GA, Agosin E (1993) Glutamine involvement in nitrogen control of gibberellic acid production in Gibberella fujikuroi. Appl Envir Microbiol 59:4317–4322 Ogle HJ, Gowanlock DH, Irwin JAG (1990) Infection of Stylosanthes guianensis and S. scabra by Colletotrichum gloeosporioides. Phytopathology 80:837–842 Oliver R, Osbourne A (1995) Molecular dissection of fungal phytopathogenicity. Microbiology 141:1–9 Pieterse CMJ, Riach MBR, Bleker T, van den Berg-Velthuis GCM, Govers F (1993) Isolation of putative pathogenicity genes of the potato late blight fungus Phytphthora infestans by differential hybridization of a genomic library. Physiol Mol Plant Pathol 43:69–79 Pieterse CMJ, Derksen ACE, Folders J, Govers F (1994) Expression of the Phytophthora infestans ipiB and ipiO genes in planta and in vitro. Mol Gen Genet 244:269–277 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual 2nd edn. Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York Sanchez F, Campomanes M, Quinto C, Hansberg W, Mora J, Palacios R (1978) Nitrogen source regulates glutamine synthetase mRNA levels in Neurospora crassa. J Bacteriol 136:880–885 Schäfer W (1994) Molecular mechanisms of fungal pathogenicity. Annu Rev Phytopathol 32:461–477 St Leger RJ, Staples RC, Roberts DW (1992) Cloning and regulatory analysis of starvation-stress gene, ssgA, encoding a hydrophobin-like protein from the entomopathogenic fungus, Metarhizium anisopliae. Gene 120:119–124 Talbot NJ, Ebbole DJ, Hamer JE (1993) Identification and characterisation of MPG1, a gene involved in pathogenicity from the rice blast fungus Magnaporthe grisea. Plant Cell 5:1575–1590 Tateno Y (1994) Evolution of glutamine synthetase genes is in accordance with the neutral theory of molecular evolution. Jpn J Genet 69:489–502 Trevorrow PR, Irwin JAG, Cameron DF (1988) Histopathology of compatible and incompatible interactions between Colletotrichum gloeosporioides and Stylosanthes scabra. Trans Br Mycol Soc 90:421–429 Valent B, Chumley FG (1991) Molecular genetic analysis of the rice blast fungus, Magnaporthe grisea. Annu Rev Phytopathol 29:443–467 Van den Ackerveken GFJM, Dunn RM, Cozijnsen TJ, Vossen P, Van den Broek HWJ, De Wit PJGM (1994) Nitrogen limitation induces the expression of the avirulence gene avr9 in the tomato pathogen Cladosporium fulvum. Mol Plant-Microbe Interact 6:210–215 Vichido I, Mora Y, Quinto C, Palacios R, Mora J (1978) Nitrogen regulation of glutamine synthetase in Neurospora crassa. J Gen Microbiol 106:251–259 Yoon CS, Glawe DA, Shaw PD (1991) A method for rapid smallscale preparation of fungal DNA. Mycology 83:835–838