G M Ballance, L Lamari, R Kowatsch and C C Bernier.
Department of Plant Science, University of Manitoba, Winnipeg, MB R3T 2N2, Canada
G M Ballance, Department of Plant Science, University of Manitoba, Winnipeg, MB
R3T 2N2, Canada
telephone: 204 474 6086 fax: 204 261 5732 email: firstname.lastname@example.org
Accepted: 9 December 1996
Pyrenophora tritici-repentis differentially induces necrosis and chlorosis in its wheat host. Necrosis-inducing (nec+) isolates produce the Ptr necrosis toxin, a 14 kD protein, responsible for the induction of necrosis in necrosis-developing wheat genotypes. A cDNA expression library was constructed and screened with anti-Ptr necrosis toxin antiserum. A 900 nucleotide cDNA clone (PtrNEC), encoding a 19.7 kD protein precursor of the Ptr necrosis toxin was isolated. In addition to the antigenicity of its product, the identity of the clone was confirmed by i) the occurrence of a series of codons which exactly describe a 24 amino acid sequence from a Ptr necrosis toxin peptide, ii) the close similarity of the clone-derived amino acid composition to that of the Ptr necrosis toxin, and iii) the identical cultivar sensitivity-range for the E. coli-expressed cDNA product to that of pathogen-produced Ptr necrosis toxin. Genomic DNA was screened from a number of fungal isolates of nec+ and nec- pathotypes using the full length PtrNEC clone. A single hybridization band was detected in nec+ isolates, but was absent from nec- isolates.
Pyrenophora tritici-repentis is the causal agent of the leaf spot disease of wheat known as tan spot. This disease occurs in all the major wheat growing areas of the world and causes from 3 to 50% yield losses (Hosford, 1982). An increased incidence of tan spot has occurred in the past three decades. This trend is believed to have resulted from the adoption, for economic and soil conservation purposes, of minimum- and zero-tillage practices on a large scale. These practices leave large amounts of stubble on which the pathogen can overwinter. Tan spot of wheat is currently controlled by crop rotation and foliar applications of fungicides. However, the development of genetically resistant cultivars represents the single most economical and environmentally safe solution to this disease, but requires an in-depth understanding of host-pathogen relations.
Isolates of P. tritici-repentis have been classified into four pathotypes based on the ability of member isolates to induce, on a wheat differential set, necrosis and chlorosis [pathotype 1, (nec+nec+)], necrosis only [pathotype 2, (nec+chl-)], and chlorosis only [pathotype 3, nec-nec+)] (Lamari and Bernier, 1989b). Isolates from pathotype 4 do not induce tan necrosis nor chlorosis (nec-nec-) and are considered to be avirulent (Lamari et al., 1990). The recent identification of isolates in pathotype 3, capable of inducing chlorosis in wheat genotypes previously known to be resistant to isolates from this pathotype, led to the adoption of a race classification system based on the virulence of isolates on individual wheat differential genotypes, rather than on the type of symptom. To date, five races have been described; races 1-4 are represented by the isolates within pathotypes 1-4 respectively, and race 5 the newly identified race (Lamari et al., 1995b) is part of pathotype 3. Susceptible wheat genotypes are known to develop, differentially, tan necrosis and/or chlorosis, whereas resistant genotypes generally develop a small dark-brown spot at the site of infection, with little or no tan necrosis or chlorosis (Lamari and Bernier, 1989a).
Presently, P. tritici-repentis is known to produce at least two host-selective toxins (Ballance et al., 1989; Orolaza et al., 1995). The chlorotic symptom induced by race 5 has been shown to be associated with the production of a host-selective toxin, capable of inducing chlorosis in wheat genotypes which develop chlorosis to race 5 isolates only (Orolaza et al., 1995). Races 1 and 3 (pathotypes 1 and 3) also induce chlorosis in hexaploid wheat lines, other than those which are chlorotic to race 5. However, no chlorosis-inducing toxin has been recovered from races 1 and 3, in spite of the fact that the genetic and physiological studies we have conducted with these races suggested the involvement of a host-selective toxin (Lamari and Bernier, 1991, 1994).
The necrotic symptom observed in interactions between nec+ isolates and necrosis-developing wheat genotypes has now been shown conclusively to be due to a protein toxin. This toxin, designated Ptr necrosis toxin, is a host-selective protein toxin (Ballance et al., 1989) which is produced only by P. tritici-repentis isolates that induce the characteristic tan necrotic lesions (pathotypes/races 1 and 2) (Lamari and Bernier, 1989c, Lamari et al., 1995a). The proteinaceous nature and amino acid composition of Ptr necrosis toxin was revealed by Ballance et al. (1989) and subsequently confirmed by other groups (Tomas et al., 1990, Tuori et al., 1994). Conventional genetic analysis of a cross between isolates 86-124 (Ptr-necrosis toxin producer) and D308 (non-producer) suggested that the production of this toxin was controlled by a single locus in the pathogen (Otondo, 1994). Likewise, a single dominant gene in the host confers susceptibility to the Ptr necrosis toxin (Lamari and Bernier, 1989c, 1991; Duguid, 1995; Gamba, 1996). This gene was recently shown, in two independent studies, to be located on chromosome arm 5BL (Faris et al., 1996; Stock et al., 1996).
The necrotic and chlorotic symptoms induced by P. tritici-repentis in hexaploid wheat hosts have been conclusively shown to be distinct and independently inherited traits (Lamari and Bernier, 1991; Duguid, 1995; Gamba, 1996). The development of each symptom appears to follow the toxin model, whereby plant susceptibility results from a unique and specific interaction between a host receptor and a toxin produced by the pathogen. The disruption of this interaction would normally result in host resistance. Temperature studies with P. tritici-repentis clearly showed that the susceptible response, both for necrosis and chlorosis, was suppressed at temperatures above 27oC. The temperature-mediated shift from compatibility to incompatibility was attributed to a failure of the Ptr necrosis (Lamari and Bernier, 1994) and Ptr chlorosis toxins (Orolaza et al., 1995; Orolaza & Lamari, unpublished results) to interact with their putative receptors. A suppression of the compatible interaction was also obtained by antibody neutralization of the Ptr necrosis toxin (Lamari et al., 1995a).
To learn more about the Ptr necrosis toxin and its possible mode of action, we have cloned a cDNA from isolate 86-124 (race 2) which produces high levels of the toxin. The clone has been confirmed to encode the Ptr necrosis toxin by amino acid sequence identity with a selective cleavage-derived peptide, amino acid composition and demonstration of specific bioactivity in differential wheat lines of the expressed product.
Fungal isolates and plant material
Isolates used in this study included 94-2, 94-17, 91-76 and ASC1 from race 1 (nec+nec+); 94-83, 92-164, 90-68, 90-31, 92-106 and 86-124 from race 2 (nec+nec-); 94-116, 331-9, 94-25 and 94-8-2 from race 3 (nec-nec+); 90-2, 25JM and 49JA from race 4 (nec-nec-) and Alg3-24 representing race 5 (nec-nec+) and were from the University of Manitoba collection. With the exception of Alg3-24 (Lamari et al., 1995b) these isolates have been collected from different geographic locations in western Canada and classified into the designated pathotypes as described previously (Lamari and Bernier, 1989b). All isolates were tested for characteristic virulence patterns on a set of five differential wheat lines/cultivars (Table 1) to verify their classification prior to DNA extraction. For bioassays, 4-5 seeds of Ptr necrosis toxin sensitive (Glenlea and Katepwa) and insensitive (6B365, Erik and Salamouni) lines/cultivars were planted in clay pots containing a 2:1:1 (soil:sand:peat) soil mix (v/v/v). Seedlings were grown to the 2-leaf stage under controlled conditions at 22/18 oC (day/night) with a 16 h photoperiod (180 uEm-2s-1). Relative humidity was maintained at approximately 60%. Seedlings were watered and fertilized as needed.
Purified Ptr necrosis toxin protein was selectively cleaved at aspartic acid residues according to an adaptation of the methods described by Inglis (1983) and Tsugita (1982). The peptide fragments were separated by sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis (PAGE) using a linear gradient gel (6-20% acrylamide) prepared according to Fling and Gregerson (1986) and transferred to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Mississauga, ON) according to manufacturers protocol. Major peptides were visualized by Coomassie Blue staining and an 8 kD peptide was isolated. This peptide was N-terminally sequenced using an Applied Biosystems 473 pulsed liquid protein sequencer (Department of Biochemistry and Microbiology, University of Victoria, BC).
Nucleic acid isolations
Fungal isolate 86-124 was grown as previously described (Ballance et al., 1989) with the exception that the Fries medium contained only 2.5 mM phosphate. The mycelial mats were harvested 14 days post inoculation and total RNA was extracted using the procedure of Logemann et al. (1987). The polyadenylated RNA was isolated by oligo(dT) cellulose chromatography (Collaborative Research Incorporated, Bedford, MA).
For genomic DNA isolation, fungal isolates representing the five races (1-5) were grown under the same conditions as described above. DNA was isolated from each of these samples using the procedure outlined by White and Kaper (1989) with the following modifications. After the first precipitation, the DNA was washed with 70% ethanol, redissolved in 5 ml of extraction buffer (0.1 M glycine, pH 9.0, 50 mM NaCl, 10 mM EDTA, 2% SDS, 1% sodium lauryl sarcosine) and treated with 50 ul of RNase A (10 mg/ml in TE). DNA was re-extracted with an equal volume of phenol/chloroform, precipitated as above, washed with 70% ethanol, dried, redissolved in water and quantified using the Hoeffer DNA fluorometer (San Fransisco, model TKO 100).
In vitro translation
The fungal mRNA was analyzed by in vitro translation using a rabbit reticulocyte translation kit from Gibco BRL (Burlington, ON) with incorporation of [35S]-labelled methionine. Aliquots of the resulting translational products were complexed with three different polyclonal anti-Ptr necrosis toxin antisera (Lamari et al., 1995a) and precipitated with protein-A Sepharose as described by Sambrook et al. (1989) with the exception that the buffer used was 0.1 M Tris/HCl, pH 7.5, 0.15 M NaCl, 2% Triton X-100. Total translational products, immunoprecipitated products, and [14C] radiolabelled protein markers (Amersham, Oakville ON) were analyzed by SDS-PAGE (Fling and Gregerson, 1986) and fluorography using a modification of the procedure of Chamberlain (1979). The gel was fixed for 1 h in methanol:acetic acid:water (5:1:5) followed by two 20-min washes in 50% methanol. The gel was then soaked for 20 min in a solution comprised of 48 g sodium salicylate, 6 ml of glycerol and 60 ml of methanol made up to 300 ml with water. Following transfer to a piece of 3MM Whatman filter paper, the gel was dried in a gel drier (BioRad, model 583 Mississauga, ON) and exposed to X-ray film for 4 days at -70oC.
A cDNA expression library was created from 86-124-derived mRNA using the lambda ZAP-cDNA synthesis kit (Stratagene, La Jolla, CA). The library was screened with anti-Ptr necrosis toxin antiserum and the picoBLUE immunoscreening kit from Stratagene. From approximately 500,000 plaque forming units screened, 31 plaques yielded positive signals. Phages from eight of these plaques were purified and converted to phagemids. Plasmid DNA was prepared according to the procedure of Birnboim and Doily (1979). Insert sizes were determined following plasmid digestion with EcoRI and XhoI. The clone selected for sequencing and further characterization was designated PtrNEC (GenBank accession U79662 ).
Sequencing was carried out on a Perkin Elmer Cetus DNA Thermal Cycler using the Applied Biosystems Prism Ready Reaction Dyedeoxy Terminator Cycle Sequencing Kit (Plant Biotechnology Institute, Saskatoon, SK). Preliminary and confirmation sequencing was done using Sequenase (United States Biochemical Corporation, Cleveland, OH) following the manufacturers sequencing protocol.
Gene expression in E. coli and toxin bioassay
Bacterial cultures (strain XL1-Blue) carrying either pSK (plasmid vector from Strategene) or pPtrNEC (pSK with cloned insert) were initiated from single colonies and used to inoculate 30 ml LB cultures (50 ug/ml ampicillin). At the mid-log stage cultures were induced with isopropyl-beta-D-thiogalactopyranoside (IPTG) as described by Sambrook et al. (1989) and harvested 4 h later. Lysates of all cultures were prepared, subjected to three freeze/thaw cycles and followed by three 5-sec pulses on a Biosonik III sonicator, to reduce the viscosity. Samples were then diluted to 1/10 with sterile distilled water and aliquots (approx. 20 ul) were infiltrated into multiple leaves of the toxin-sensitive and toxin-insensitive cultivars, at the 2-leaf stage of development, using a Hagborg device (Hagborg, 1970). Plants were monitored daily for the presence or absence of necrosis in the infiltrated regions.
Northern and Southern analysis
Total RNA (15 ug) and mRNA (10 ug) samples were denatured and electrophoretically separated in a 1.5% agarose gel containing 2.2 M formaldehyde and transferred to Hybond-N nylon membrane (Amersham). Prehybridization (2 h) and hybridization (18 h) were carried out in 4xSSPE (0.6 M NaCl, 40 mM NaH2PO4, 4 mM EDTA), 5x Denhardts solution, 0.5% SDS at 62oC. Washing of the membrane was carried out at 62oC with the final washes in 2xSSPE, 0.1% SDS (20 min) and 0.1xSSPE, 0.1% SDS (10 min). The membrane was exposed to X-ray film with an enhancement screen for 3 days at -70oC.
For Southern analysis, genomic DNA (10 ug) of each isolate was digested to completion with XhoI and separated in 0.7% agarose gels in 1xTBE buffer. DNA from isolates ASC1 and 86-124 were also digested with HindIII, SalI and KpnI and separated. Ethidium bromide staining was used to confirm that similar amounts of DNA had been loaded in each lane. Separated DNA was transferred to Hybond-N nylon membrane (Amersham) according to manufacturers protocol. Blotting, prehybridization, hybridization and washing conditions were as described above.
For both Northern and Southern analysis, membranes were probed with the full length 32P-labelled insert of pPtrNEC. The insert was excised by digestion with XhoI and EcoRI, separated from the vector fragment in a 1% agarose gel and recovered using a Gene-Clean kit (BIO 101 Incorporated, La Jolla, CA). Probe DNA (25 ng) was labelled using random hexanucleotide primers according to Feinberg and Vogelstein (1983).
Primer preparation and PCR
The 19mer forward primer (GCCATGGGTTCTATCCTCG) was used in the polymerase chain reaction (PCR) (Saiki et al., 1988) in conjunction with the reverse primer (ATTTTCACGACCTGTATCA), corresponding to the complementary sequence of nucleotides 338 to 356. The forward primer differed from the nucleotide sequence 44 to 62 of PtrNEC at nucleotides 45 (changed to C) and 50 (changed to G) to create an NcoI site for other work. The PCR reaction mix (50 ul) consisted of Taq buffer, primers (20 pmoles of each), genomic DNA (50 ng) and Taq polymerase. An initial heat denaturation step at 95oC for 3 min was followed by 25 cycles of 92oC for 0.5 min, 49oC for 0.5 min and 72oC for 1 min. Samples were electrophoretically analysed on a 1.2% agarose gel in 1xTBE buffer.
In vitro translated products from mRNA extracted from fungal isolate 86-124 were treated with polyclonal antiserum raised against the purified Ptr necrosis toxin (Lamari et al., 1995a). The autoradiograph of the SDS-PAGE size-fractionated protein products (Figure 1) shows that a large proportion of the total translational products has a molecular size of 12-14 kD which is in the size range of the mature Ptr necrosis toxin (Ballance et al., 1989). Immunoprecipitation of the labelled proteins from the translated mRNA was assessed with three different antisera one of which was less effective than the others. The level of immunoprecipitated products was low but indicated the presence of two proteins of approximately 14 and 19 kD. The 19 kD protein band was observed in two of the three immunoselection reactions (lanes 2 and 4). The occurrence of an in vitro translated protein which is larger than the corresponding mature protein is suggestive that the mature protein may result from processing of a larger precursor.
cDNA isolation and identification
A cDNA expression library was prepared from mRNA of isolate 86-124 and screened with the anti-Ptr necrosis toxin antiserum. Of the putative toxin clones identified with the antiserum, several were selected for further analysis based on the plasmid insert size. Preliminary sequencing of several clones resulted in the selection of the clone, designated pPtrNEC, for complete sequencing. Excluding the polyA+ tail, the clone consists of 900 nucleotides. The nucleotide sequence of this clone and the deduced amino acid sequence are presented (Figure 2).
Beyond the antiserum recognition of the clone product, the identity of this clone as encoding the Ptr necrosis toxin was substantially confirmed from the partial sequence of an 8 kD cleavage peptide derived from the Ptr necrosis toxin. The N-terminal sequence of this peptide was found to be SVILGRPGAIGSWELNNFITIGLN, which corresponds exactly to translation of nucleotides 302-373 in the clone sequence. Furthermore, as expected from the method used for selective hydrolysis, an aspartic acid residue precedes the deduced amino acid sequence.
The coding region is likely to begin at nucleotide 47 (ATG, M-1) and runs to the TAG stop codon beginning at nucleotide 581. Several additional in-frame stop codons occur downstream of this first stop codon. Supporting evidence for the proposed translational start site comes from a sequence analysis indicating that amino acid residues from M-1 to A-16 constitute a putative signal peptide (von Heijne, 1986). The translational product of the proposed coding region corresponds to a protein of 177 amino acids with a molecular mass of 19,707. This is significantly larger than the mature Ptr necrosis toxin protein (14 kD) isolated from the culture filtrate of isolate 86-124 (Ballance et al., 1989) but is similar to the larger immunoprecipitated in vitro translation product (Figure 1). A second methionine codon (M-65) is considered an unlikely translation start point as it would yield a protein of only 12,837 kD (113 residues) which is smaller than the mature Ptr necrosis toxin protein.
To identify a likely processing site which would yield the mature toxin product, we have compared size and charge of predicted protein products from the identified C-terminus back to various N-termini. The mature Ptr necrosis toxin has an estimated mass of approximately 14,000 (Ballance et al., 1989) and a pI of near 9 (G M Ballance, unpublished result). The results of pI and size comparisons for potential mature proteins initiating from different residues in the region of residue 52 are presented in Table 2. We propose that the N-terminus of the mature protein is at L-51 or K-52 because processing from either of these residues would most closely fulfill the size and charge criteria. Further evidence comes from the good correspondence of the amino acid composition of the deduced proteins including these residues with that of the previously characterized Ptr necrosis toxin protein (Table 2; Ballance et al., 1989). Inclusion of K-52 is essential to account for the single lysine detected in the purified protein (Ballance et al., 1989). A single lysine is also predicted from the data of Tomas et al. (1990) and of Tuori et al. (1995) for the related toxin proteins that these groups have isolated. Although it has been suggested that these toxins are different from the Ptr necrosis toxin, the similarity in amino acid composition, protein properties, and functional activity suggests them to be very similar, if not identical.
While the sequence of the 3' end of the clone was confirmed as identical to several other clones, pPtrNEC contained the largest insert identified, the second largest starting with nucleotide 59. The 5' end was substantially confirmed by PCR using DNA from isolate 86-124 as a template, a forward primer which overlaps the translational start codon, and a reverse primer from near the 3' end of the coding region. A single product was obtained which was approximately 50 nucleotides longer than the amplified fragment from pPtrNEC (data not shown). The slight size differences of the amplification products are likely indicative of a small intron in the genomic sequence.
Final verification of the clone was achieved by testing the product of PtrNEC in bacterial cell lysates for host-selective toxin activity. Lysates from bacterial cells carrying either the pSK plasmid or pPtrNEC were prepared and infiltrated into wheat cultivars/lines Glenlea and Katepwa (toxin sensitive), Erik, 6B365, and Salamouni (toxin insensitive). Purified toxin from 86-124 culture filtrate was infiltrated separately for comparison. The only lysates which yielded necrosis were from bacteria carrying the pPtrNEC plasmid when infiltrated into Glenlea or Katepwa. Similarly, these were the only cultivars which developed necrosis with the purified Ptr necrosis toxin. Typical responses for these reactions are shown in Figure 3. All other lysate-cultivar combinations were symptomless, i.e. neither necrosis nor chlorosis occurred.
Western analysis of the pPtrNEC-bacterial lysate revealed two bands cross-reacting with the Ptr necrosis toxin antiserum (Figure 4). These bands represent proteins of estimated masses of 24 and 18.5 kD. Given that the cloned cDNA was inserted into an expressed lacZ gene fragment, two products could arise if translation was initiated from both the lacZ ATG start site in the vector and from the proposed ATG in the clone. The estimated masses of the antiserum-recognized bands are within 6% of the predicted size of these two translational products (25.5 and 19.7 kD, respectively). The 14 kD processed form of the toxin was not detected in the lysates. However, the observed toxicity indicates that one or both of the bacterial-produced forms of the protein have the same bioactivity as the purified form.
Northern analysis was run to verify that the size of the isolated clone is comparable with that of the transcripts. Total RNA and mRNA gave bands of similar size (Figure 5). The band in the mRNA sample was smaller which may be due to a loss of terminal nucleotides during purification. The tailing from this band also suggests that some degradation had occurred. The estimated sizes of the hybridizing RNA in the two preparations were 0.85 and 0.98 kb which are consistent with the size of the cloned cDNA PtrNEC.
To determine the occurrence of the Ptr necrosis toxin-encoding gene in different races of the pathogen, DNA from isolates of each of the five races was subjected to Southern analysis. Hybridizing fragments were observed with DNA from all isolates from races 1 and 2 but not of the isolates from races 3, 4 and 5 (Figure 6), even when less stringent hybridization conditions (1xSSPE) were used.
Additional Southern analyses were performed with DNA from four race 1 isolates and six race 2 isolates, collected from across Western Canada (Figure 7). In all races similar-sized restriction fragments indicated the presence of the gene. Furthermore, restriction analysis of DNA from isolates 86-124 and ASC1 using two enzymes, HindIII and SalI, which have internal restriction sites (Figure 2), yielded the same banding pattern (Figure 8). These findings suggest that the gene is very similar or identical in these isolates.
Using antisera raised against the Ptr necrosis toxin, a cDNA clone was isolated from an expression library of the toxin-producing isolate, 86-124. Based on toxin primary sequence information, amino acid composition data, and bacterial expression results, we are confident that the clone corresponds to the Ptr necrosis toxin gene. The translational start site is assumed to be at nucleotide 47 based on the observations that this is the only methionine position which could account for the size of the total protein, and that an immunoprecipitated in vitro translation product of similar size was found. In addition, the deduced amino acid sequence contains a putative transit peptide. The actual location of the N-terminus of the mature protein is uncertain but based on size, charge and amino acid composition of the deduced protein, it is predicted to be in the region of amino acid K-52.
For all the isolates so far examined the presence of necrosis inducing activity has coincided with the presence of a single hybridizing band in the restricted genomic DNA and the absence of such activity with the absence of the gene. The absence of hybridization in isolates from races 3-5 provides a clear explanation for why these isolates do not exhibit necrosis-producing activity nor produce protein products which are recognizable by the Ptr necrosis toxin antiserum (Lamari et al., 1995a). The higher level of toxin activity and toxin protein production which has been noted in earlier work for isolate 86-124 relative to ASC1 (Lamari et al., 1995a) does not appear to be due to expression from multiple copies of the gene in this isolate. Partial sequencing of several clones failed to detect any sequence variation. Also Southern analyses showed no evidence in 86-124 or in other isolates of the presence of more than one gene. A database search has failed to reveal any related protein sequence to suggest a similar gene has been identified previously. Such information could provide some clue as to how the toxin might produce its toxic effect.
Until recently tan necrosis and chlorosis were considered to be a single symptom induced by virulent isolates of P. tritici-repentis (Hosford, 1982). The failure to isolate a host-specific toxin(s) from race 3 isolates, to account for the development of chlorosis in some wheat genotypes, raised the question of whether the Ptr necrosis toxin was responsible for race 3-type chlorosis under some unknown conditions. The cloning of the Ptr necrosis toxin and its expression in E. coli presented the opportunity to further confirm our previous findings, that the Ptr necrosis toxin causes necrosis, but not chlorosis.
In this study, we have included in the bioassay host lines which cover the entire range of symptoms induced by all the virulent races of P. tritici-repentis described to date. The perfect match, with regard to necrosis, between the host range of nec+ isolates, sensitivity to the in vitro-produced purified Ptr necrosis toxin, and sensitivity to the E. coli-pPtrNEC produced toxin indicates that the cDNA we have cloned controls the production of the Ptr necrosis toxin only and has no effect on the development of chlorosis. This was evident in cultivar Katepwa [sensitive to both the Ptr necrosis (races 1,2) and Ptr chlorosis (race 5) toxins], which developed typical necrosis but no chlorosis when infiltrated with E. coli-pPtrNEC lysate. Similarly, line 6B365 did not develop any symptom, in spite of the fact that it develops extensive chlorosis to isolates from races 1 and 3 (Lamari and Bernier, 1989b). The fact that a single band was detected in isolates from pathotype 1 (race 1), which are capable of inducing necrosis and race 3-type chlorosis, further supports the hypothesis that the Ptr necrosis toxin is distinct from the predicted race 3 chlorosis-inducing toxin and is not associated with the induction of chlorosis in wheat.
The presence of a genomic sequence related to the cloned cDNA in all nec+ isolates (races 1 and 2) and its absence from the nec- isolates (races 3, 4 and 5) suggests that the gene encoding the Ptr necrosis toxin did not arise as a mutation in an originally avirulent isolate; some degree of hybridization would have been observed with nec- isolates if this was the case. A more likely scenario is that the Ptr necrosis toxin gene was acquired by P. tritici-repentis as a translocation from another species of Pyrenophora or an entirely different species and was maintained because of the selective advantage it confers in colonizing wheat and perhaps other grasses. P. tritici-repentis is known to have the widest host range of all species reported in the genus Pyrenophora (Shoemaker, 1962). Alternatively, nec- isolates could have originated from nec+ types by deletion of this gene; this case is however unlikely, as it would not yield a selective advantage. Additional studies on the phylogeny of P. tritici-repentis are needed to trace the origin of this gene. The results of this study conclusively confirm our initial classification of isolates into pathotypes, based on the ability of the isolates to induce necrosis and chlorosis (Lamari and Bernier, 1989b) and provide the strongest validation of the necrosis-chlorosis model for tan spot of wheat.
Financial support for this work by the Natural Sciences and Engineering Research Council (NSERC) of Canada is gratefully acknowledged. The authors also wish to thank R. Oree for technical assistance and B. McCallum for critical reading of the manuscript.
Ballance GM, Lamari L, Bernier CC, 1989. Purification and characterization of a host-selective necrosis toxin from Pyrenophora tritici-repentis. Physiological and Molecular Plant Pathology 35, 203-13.
Faris JD, Anderson JA, Francl LJ, Jordahl JG, 1996. Chromosomal location of a gene conditioning insensitivity in wheat to a necrosis-inducing culture filtrate from Pyrenophora tritici-repentis. Phytopathology 86, 459-63.
Lamari L, Bernier CC, 1989b. Virulence of isolates of Pyrenophora tritici-repentis on 11 wheat cultivars and cytology of the differential host reactions. Canadian Journal of Plant Pathology 11, 284-90.
Lamari L, Sayoud R, Boulif M, Bernier CC. 1995b. Identification of a new race in Pyrenophora tritici-repentis: Implications on the current pathotype classification system. Canadian Journal of Plant Pathology 17, 312-18.
Stock W, Brule-Babel AL, Penner GA, 1996. A gene for resistance to a necrosis inducing isolate of Pyrenophora tritici-repentis located on 5BL of Triticum aestivum, cv Chinese Spring. Genome 39, 598-604.
Tomas A, Feng GH, Reeck GR, Bockus WW, Leach JE, 1990. Purification of a cultivar-specific toxin from, Pyrenophora tritici-repentis, causal agent of tan spot of wheat. Molecular Plant-Microbe Interactions 3, 221-24.
Tuori RP, Wolpert TJ, Ciuffetti LM, 1995. Purification and immunological characterization of toxic components from culture of Pyrenophora tritici-repentis. Molecular Plant-Microbe Interactions 8, 41-8.White JL, Kaper JM, 1989. A simple method for detection of viral satellite RNAs in small plant tissue samples. Journal of Virological Methods 23, 83-94.
Table 1. Reaction of hexaploid wheat lines/cultivars to the Ptr necrosis and to five races of Pyrenophora tritici-repentis. ________________________________________________________________________________________________ Cultivars Ptr necrosis Race2 toxin1 _______________________________________________ ______________nec+__ _____________nec-_________ 1 2 3 4 5 Glenlea + S(N) S(N) R R R Katepwa + S(N) S(N) R R S(C) 6B365 - S(C) R S(C) R R Erik - R R R R R Salamouni - R R R R R ____________________________________________________________________________________________ 1reaction to the Ptnr necrosis toxin: + = sensitive, - = insensitive 2reaction to the fungus: S = susceptible, R = resistant, (N) = necrosis, (C)= chlorosisReturn to text
Table 2. Amnio acid composition, protein size and pl of purified Ptr necrosis toxin, and putative proteins from different N-terminal residues _______________________________________________________________________________________________ Acid Ptr necrosis Amnio toxin from #17(A) from #51(L) from #52(K) from #53(P) # of mol # of mol # of mol # of mol # of mol residues % residues % residues % residues % residues % Ala 4 3.1 7 4.4 3 2.4 3 2.4 3 2.4 Arg 13 10.2 15 9.4 14 11.0 14 11.1 14 11.2 Asn 15 9.4 14 11.0 14 11.1 14 11.2 Asp/Asx 20 15.7 10 6.3 7 5.5 7 5.5 7 5.6 Cys 2 1.6 2 1.3 2 1.6 2 1.6 2 1.6 Gin 6 3.8 6 4.7 6 4.8 6 4.8 Glu/Glx 10 7.9 9 5.6 5 3.9 5 3.9 5 4.0 Gly 13 10.2 15 9.4 13 10.2 13 10.3 13 10.4 His 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 lle 14 11.0 14 8.8 13 10.2 13 10.3 13 10.4 Leu 8 6.3 14 8.8 10 7.9 9 7.1 9 7.2 Lys 1 0.8 2 1.3 1 0.8 1 0.8 0 0.0 Met 1 0.8 2 1.3 2 1.6 2 1.6 2 1.6 Phe 3 2.4 4 2.5 3 2.4 3 2.4 3 2.4 Pro 4 3.1 7 4.4 4 3.2 4 3.2 4 3.2 Ser 9 7.1 11 6.9 8 6.3 8 6.3 8 6.4 Thr 10 7.9 11 6.9 9 7.1 9 7.1 9 7.2 Trp 3 2.4 4 2.5 3 2.4 3 2.4 3 2.4 Tyr 2 1.6 3 1.9 2 1.6 2 1.6 2 1.6 Val 10 7.9 9 5.6 8 6.3 8 6.3 8 6.4 Total Residues 127 160 127 126 125 M.Wt. 14,125 18,050 14,405 14,291 14,163 pl -- 5.21 9.37 9.37 8.9 ________________________________________________________________________________________________Return to text