(Received for publication, August 9, 1996, and in revised form, October 24, 1996)
From the Department of Botany, University of Toronto, Toronto, Ontario M5S 3B2, Canada
Two novel elicitor peptides that are produced by the race 1 of the cowpea rust fungus Uromyces vignae and that specifically induce a hypersensitive response (a putative form of programmed cell death) in a resistant cultivar of cowpea (Vigna unguiculata L. Walp) have been purified to homogeneity. Purification steps included gel filtration, anion-exchange chromatography, continuous elution electrophoresis, and reversed-phase C18 high performance liquid chromatography. The relative molecular masses of the peptide elicitors as deduced from Tricine sodium dodecyl sulfate-polyacrylamide gel electrophoresis were 5600 Da (major) and 5800 Da (minor), respectively. Peptide 1 (major) and the minor copurifying peptide (peptide 2) resolved at the final C18 high performance liquid chromatography step. The NH2 terminus of peptide 1 was deblocked with anhydrous trifluoroacetic acid prior to sequencing. However, the NH2 terminus of peptide 2 was free. The acidic and hydrophobic peptides show some homology between themselves but do not show any significant similarity with known proteins. The two specific elicitors may be products of two avirulence genes corresponding to the two genes for resistance in the resistant cultivar.
The hypersensitive response (HR),1 a putative form of programmed cell death (1), is a nearly invariant marker of disease resistance in plants and is characterized by rapid localized pathogen-induced cell death and restriction of further pathogen growth. This phenomenon was first described by Stakman (2) in the wheat-wheat rust fungus interaction and is exemplified by the cowpea-cowpea rust fungus interaction used in this study.
The interaction between race 1 of the cowpea rust fungus and the resistant cultivar, Dixie Cream, is an incompatible one resulting in a hypersensitive response (3). On the other hand, the susceptible host (California Blackeye) shows very little or no cell death. For most rust fungi, cultivar resistance is controlled by a gene-for-gene interaction in which a specific avirulence gene in the pathogen has to be "matched" by a specific resistance gene in the host as described for the first time by Flor (4) from studies on the flax rust fungus. The current molecular interpretation of the gene-for-gene interaction is that the avirulence gene produces a "specific elicitor" that is recognized by the product of the resistance gene (5), eventually resulting in a plant defense response. However, the exact biochemical basis for gene-for-gene recognition of microbes by plants is largely unknown.
To date, although several "nonspecific" elicitors that trigger various plant defense responses have been identified and purified (6-8), cultivar-specific elicitors that mimic the race-cultivar specificity of the pathogen have been demonstrated only in a few plant-pathogen interactions (9, 10). The avirulence genes, avr9 and avr4 of the tomato fungal pathogen, Cladosporium fulvum, were shown to encode precursors of elicitor peptides that trigger a hypersensitive response in resistant plants carrying the cf9 and cf4 resistance genes, respectively (11-13). Other examples of specific elicitors include the NIP1 peptide of the barley pathogen Rhynchosporium secalis (14), the coat protein of tobacco mosaic virus (15), and the syringolide of the bacterial soybean pathogen, Pseudomonas syringae pv. glycinea (16). However, no specific elicitor has previously been characterized from a rust fungus, although evidence for their existence in the cowpea rust fungus-cowpea system was shown in earlier studies (17). Unlike the pathogens mentioned above, the cowpea rust fungus is unique in that it is an obligate biotroph that invades cells of susceptible plants without killing them and elicits a HR only after cell penetration in the resistant cultivar. Of the two parasitic stages produced by rust fungi, the monokaryotic stage of the cowpea rust fungus seems to be the only one that secretes the specific elicitors outside of the host cell (18). The fact that the monokaryotic stage is the most difficult to study and that rust fungi generally cannot be grown in culture explain why the isolation of specific elicitors from these fungi has not previously been accomplished. We report here the purification and characterization of two novel HR-inducing race-cultivar specific elicitors produced by the cowpea rust fungus.
The host-pathogen system used for the study was the cowpea-cowpea rust fungus. IWFs were obtained from the susceptible cultivar (California Blackeye) of cowpea plants (Vigna unguiculata L. Walp.) inoculated with basidiospores (monokaryotic stage) of the race 1 of the cowpea rust fungus (Uromyces vignae Barclay) as described earlier (17). Aqueous exudates from in vitro differentiated basidiospore germlings were collected as described previously (18). Media for column chromatography (Sephadex G-25F, CM-Sephadex C-25, DEAE-Sephacel) were from Pharmacia Biotech Inc. The Delta-Pak C18 reversed-phase HPLC column and C18 Sep-Pak light cartridges were from Waters. Equipment for electrophoresis, electroblotting, and continuous elution electrophoresis was from Bio-Rad. Mark 12 wide range (2.5-200 kDa) standard markers were from Novex. Immobilon-PSQ transfer membrane for blotting and microsequencing was from Millipore. All organic solvents for reversed-phase HPLC were of HPLC grade. Other reagents were of standard reagent grade.
Purification of HR-inducing Specific ElicitorsSpecific elicitors that elicit a HR were purified from IWFs of fungus-infected leaves. Identical purification procedures were carried out in the case of aqueous exudates of in vitro differentiated basidiospore germlings to determine whether the specific elicitors were fungus-derived or plant-derived.
One hundred ml of IWF was lyophilized. The lyophilized IWF was redissolved in 1 ml of 0.05 M ammonium bicarbonate, pH 7.8. After centrifugation at 14,000 × g for 15 min at 4 °C, the supernatant was loaded onto a Sephadex G-25F column (1.6 cm × 25 cm) that was pre-equilibrated with 0.05 M ammonium bicarbonate. Fractions of 1 ml were collected at a flow rate of 30 ml/h. The fractions showing HR-inducing activity were pooled. The pooled fractions were then loaded either onto a cation-exchange CM-Sephadex C-25 column (1 cm × 15 cm) or onto an anion-exchange DEAE-Sephacel column (1 cm × 15 cm) that was pre-equilibrated with 0.05 M ammonium bicarbonate, pH 7.8. Unbound material was washed out. The bound proteins were eluted with a linear ionic gradient of 0.05-0.5 M ammonium bicarbonate at a flow rate of 1 ml/min/fraction. The eluted HR-active fractions were pooled, lyophilized, and resuspended in 80 µl of gel loading buffer for continuous elution electrophoresis. The HR-active sample (80 µl) was run on a 10% native polyacrylamide gel (19) prepared in a 7-mm diameter tube of the Bio-Rad Mini Prep Cell apparatus, the elution chamber of which had an Mr 3500 cutoff precut dialysis membrane. A 5-cm length of 10% acrylamide separating gel and a 1-cm length of 4% acrylamide stacking gel were used. The sample was electrophoresed at 1 watt of constant power. Fractions were collected at 100 µl/min. 2-min fractions were collected 2.5 h after loading and immediately prior to complete exit of the tracking dye from the gel. Elution was monitored by absorbance at 215 nm. All the above steps were carried out at 4 °C. The purity of the HR-inducing specific elicitors was determined on a 10% native polyacrylamide gel and on a 16.5% Tricine SDS-polyacrylamide gel (20). The peptides were detected by silver staining (21). The electrophoretic masses of the peptides were determined by comparing the electrophoretic mobilities with those of molecular mass markers. The gels were also stained for detection of glycopeptides (22). Protein concentrations were determined according to the Lowry method (23) using bovine serum albumin as a standard.
The HR-active fractions obtained from continuous elution electrophoresis were pooled and desalted on a preconditioned reversed-phase C18 light cartridge. The cartridge was preconditioned with 3 ml of acetonitrile:isopropyl alcohol:water (6:3:1) and washed with 3 ml of 0.1% trifluoroacetic acid. Following application of the sample in 0.1% trifluoroacetic acid, the unbound material was washed out with 3 ml of 0.1% trifluoroacetic acid. The bound peptides were eluted with 3 ml of 0.1% trifluoroacetic acid in acetonitrile:isopropyl alcohol:water (6:3:1). The eluted peptides were vacuum-concentrated in a Savant Speed Vac concentrator.
High resolution purification of the elicitors was carried out on a Delta-Pak C18 reversed-phase HPLC column (3.9 × 150 mm) run on a Hewlett-Packard 1090 system. The column was equilibrated for 10 min with 0.1% trifluoroacetic acid. The concentrated Sep-Pak eluates were reconstituted in 0.1 ml of 0.1% trifluoroacetic acid. Injections of 20 µl were made into the column. The peptides were eluted with a linear gradient of 0-60% acetonitrile:isopropyl alcohol:water (6:3:1) containing 0.1% trifluoroacetic acid at a flow rate of 1 ml/min over 50 min. The peptides were detected by their UV absorption at 215 nm. Individual peaks with HR-inducing activity were collected separately.
MicrosequencingThe reversed-phase HPLC-purified peptides were run on a 16.5% Tricine SDS-polyacrylamide gel and electroblotted onto a 0.1-µ pore size Immobilon-PSQ transfer membrane in 10 mM CAPS, pH 11. The bands corresponding to the two peptides were excised and subjected to Edman gas-phase microsequencing on an Applied Biosystems Model 476 gas-phase sequencer (Protein Sequencing Service, University of British Columbia, Vancouver, Canada). The amino acid sequences obtained were compared to known sequences in the data base using the BLAST program (24).
BioassayThe HR-inducing activity of the peptides was assayed on the susceptible and resistant cultivar as described previously (17) at every step of the purification. The samples (100 µl) were injected into primary leaves of 9-day-old plants. Twenty-four hours later, 10 infiltrated areas (approximately 1 cm2) were harvested and prepared for observation of autofluorescent dead cells. Active samples had a total of 500-1000 autofluorescent cells per harvested tissue.
Effect of pH and Heat on the Specific ElicitorsThe
purified peptides were incubated overnight at pH 2 (0.01 M
HCl) and pH 12 (0.01 M NaOH), desalted on Sep-Pak
C18 light cartridges, and assayed for HR-inducing activity
following evaporation of the organic solvents. The HR-inducing activity
of the purified peptides was tested after heating the peptides for 10 min at 100 °C, after storing the peptides at 20 °C for over a
month, and after repeated freezing and thawing.
HR-inducing
specific elicitors were purified to homogeneity from IWFs of infected
plants and from exudates of in vitro differentiated basidiospore germlings. A typical purification commenced with 100 ml of
IWF as described under "Experimental Procedures." A summary of the
purification steps and the overall purification from the total protein
in the starting material is shown in Table I. The poor
yield of HR-inducing peptides obtained at the end of the reversed-phase
HPLC step and the difficulty in growing sufficient teliospores (from
which basidiospores are formed) were limiting factors in pursuing
complete characterization of the peptides. The HR-inducing activity
eluted from the Sephadex G-25F column after the void volume (Fig.
1), indicating that the elicitors under investigation
were rather small. The behavior of the HR-inducing activity was
evaluated on a CM-Sephadex C-25 (cation-exchange) and a DEAE-Sephacel
(anion-exchange) column. The acidic nature of the HR-inducing activity
was evident as it bound only to the DEAE Sephacel column and was eluted
with 0.25-0.35 M ammonium bicarbonate. The anion-exchange
chromatographic profile is shown in Fig. 2. Ammonium
bicarbonate-eluted fractions were tested for HR-inducing activity
following lyophilization. Twenty minutes after elution of the tracking
dye from continuous elution electrophoresis, twelve 2-min fractions
that were pooled and analyzed on a 10% native polyacrylamide gel
showed a single band. However, analysis on a silver-stained 16.5%
Tricine SDS-polyacrylamide gel revealed two closely moving bands, a
major band with a relative molecular mass of 5600 Da (peptide 1) and a
minor band with a relative molecular mass of 5800 Da (peptide 2) as
determined by comparison with protein standard markers. Tricine
SDS-PAGE with or without treatment of the sample with the reducing
agent -mercaptoethanol did not affect the migration of the peptides,
indicating that they were not covalently bound to each other by
disulfide bridges. Lack of staining by the periodic-silver reagent
indicated that the peptides were most likely not glycosylated.
|
Peptide 1 and peptide 2 resolved into individual peaks at the final
C18 reversed-phase HPLC step. Fig. 3
illustrates the reversed-phase HPLC chromatogram. The second and third
peaks in the chromatogram represented peptide 2 and peptide 1, respectively, as analyzed on a 16.5% Tricine SDS-polyacrylamide gel.
Peptide 2 eluted at approximately 25% acetonitrile:isopropyl
alcohol:water (6:3:1) and had a retention time of 20 min, while peptide
1 eluted at 35% acetonitrile:isopropyl alcohol:water (6:3:1) with a
retention time of 29 min in the 0-60% linear gradient. Both peptides
were biologically active. The HR-inducing activity was much stronger when a combination of the two peptides was injected into resistant plants rather than when individual peptides were injected. The small
peak that eluted at the 14th minute was not HR-active and had no effect
on the HR-inducing activity when injected along with the HR-active
peptides.
Sequence Analysis of HR-inducing Specific Elicitors
The individual HR-inducing peaks from a number of reversed-phase HPLC cycles were combined for microsequencing. Due to the hydrophobic nature of the peptides, binding to the polyvinylidene difluoride membrane was poor, resulting in considerable loss of the peptides during transfer. Approximately 50 pmol each of blotted peptide 1 and peptide 2 were subjected to NH2-terminal Edman microsequencing. This quantity, although small, allowed about 82% of the amino acid sequences of each of the peptides to be determined. The NH2-terminal sequences are shown in Table II. The average repetitive yield for each of the peptides was approximately 90%. Each amino acid was the major or only amino acid called by the automated sequencer at each cycle. The NH2 terminus of peptide 1 was blocked, while that of peptide 2 was free. The acetylated NH2 terminus of peptide 1 was deblocked with anhydrous trifluoroacetic acid (Protein Sequencing Service, University of British Columbia, Vancouver, Canada). A search for sequence similarity using the BLAST program did not reveal any significant sequence similarity to any known proteins. However, the peptides have a few stretches of amino acid residues between themselves that are similar.
|
The same purification procedure used for IWFs revealed identical peptide bands from exudates of in vitro differentiated germinated basidiospores, indicating that the HR-inducing elicitors were derived from the fungus and not from the plant and that they were secreted by the fungus both in vitro and in vivo. The amount of elicitors obtained from the exudates, however, was very low due to the limited growth of the differentiated basidiospore germlings in vitro.
Properties of HR-inducing Specific ElicitorsThe peptides
were highly soluble in water as well as in organic solvents,
viz. methanol, ethanol, acetonitrile, and isopropyl alcohol.
The peptides were heat-stable. There was no loss of the HR-inducing
activity on storage at 20 °C or on repeated freezing and thawing.
Incubation of the peptides at acidic or alkaline pH did not alter their
activity. The purified elicitors were active and specific. They induced
the HR only in the resistant cultivar and not in the susceptible
cultivar.
Purification and characterization of race-cultivar specific
elicitors is the first step toward understanding the biochemical and
molecular basis of gene-for-gene interactions between plants and their
parasites. The hypersensitive response, which is the common result of
such interactions in host cultivars resistant to the parasite, has been
suggested to be a form of programmed cell death evolved specifically as
a defense against microbial attack. In support of such a hypothesis,
the hypersensitive response in the cowpea-cowpea rust system involves
the same type of DNA cleavage as seen in animal apoptosis (1).
Preliminary data suggest the involvement of interleukin-1 converting
enzyme-like proteases2 as is also typical
of animal programmed cell death (25).
We have demonstrated here for the first time the existence of two novel race-cultivar specific HR-inducing elicitors from the cowpea-cowpea rust fungus interaction. The successful isolation of these elicitors was possible due to the application of powerful techniques such as continuous elution electrophoresis and high resolution reversed-phase HPLC. Our results have shown that a combination of both of the peptides elicits a stronger HR than the individual peptides; correspondingly, in previous studies of the resistant cultivar used to bioassay the peptides (26) the presence of both resistance genes in the heterozygous condition were shown to result in a faster and more effective HR in response to the fungus than the presence of either gene alone. Since specific elicitors are effective only in the presence of specific resistance genes (27), and since this resistant cultivar contains two resistance genes (3), it is possible that the two specific elicitors we have purified may be the products of two corresponding avirulence genes in U. vignae race 1. However, we could not test this because only non-isogenic plants containing single resistance genes are available and we could not isolate sufficient individual peptides to test them on the large number of plants needed to clearly show an association between sensitivity to each peptide and the presence of each gene.
The fact that the specific elicitors are hydrophobic in nature and are present in the exudates of in vitro differentiated basidiospore germlings as well as in the IWFs suggests that they could be the processed mature forms of larger precursor molecules, as is the case of the processed avr9 gene product of C. fulvum (28). However, the lack of cysteines emphasizes that the elicitors would be totally structurally different from Avr9, the three-dimensional structure of which has recently been resolved (29).3
Unlike other specific elicitors so far identified, the HR-inducing specific elicitors from the cowpea rust fungus are unique in that they are rich in the acidic residue aspartic acid and in the hydrophobic residues valine, proline, and leucine. An interesting feature of both peptides is the presence of proline-rich regions, which in other proteins have been shown to not only have an important influence on the conformation and folding of the peptide backbones but which also explicitly define a very rapid and remarkably strong protein-binding capacity (30). Rapid and strong binding of the elicitors to the receptors would be essential to quickly activate the cell death pathway and thus prevent further growth of the fungus. Indeed, the products of the cowpea plant resistance genes could be envisioned as receptors with binding sites rich in flat hydrophobic surfaces (such as aromatic amino acid residues). The stability and the water solubility of the specific elicitor peptides can be explained by the conformational constraint that proline residues induce and by their strong hydrogen-bond accepting nature, respectively.
Because differentiated basidiospore germlings fail to grow in vitro beyond the germling stage, and the teliospores from which basidiospores form cannot be obtained in large quantities, it was not possible to purify sufficient quantities of specific elicitors to obtain complete amino acid sequences. With the partial sequences of the peptides now known, the genes can be cloned and an expression system developed to produce sufficient quantities for further study of elicitor structure, mode of action, and relationship to fungal avirulence genes.
We thank Pui Tam for the excellent technical assistance provided during this study, Dr. Verna Higgins for providing accessibility to laboratory equipment, and Dr. Frank Dicosmo for the use of the HPLC system during the final stages of purification of the HR-inducing specific elicitors.