- Email: [email protected]
Phytotoxicity of solanapyrones A and B produced by the chickpea pathogen Ascochyta rabiei(Pass.) Labr. and the apparent metabolism of solanapyrone A by chickpea tissues
Physiological and Molecular Plant Pathology (2000) 56, 235±244 doi:10.1006/pmpp.2000.0272, available online at http://www.idealibrary.com on
Phytotoxicity of solanapyrones A and B produced by the chickpea pathogen Ascochyta rabiei (Pass.) Labr. and the apparent metabolism of solanapyrone A by chickpea tissues K H A L I D H A M I D { and R I C H A R D N . S T R A N G E * Department of Biology, University College London, Gower Street, London WC1E 6BT, U.K. (Accepted for publication March 2000) When chickpea shoots were placed in solanapyrone A, the compound could not be recovered from the plant and symptoms developed. These consisted of loss of turgor, shrivelling and breakage of stems and ¯ame-shaped, chlorotic zones in lea¯ets. In similar experiments with solanapyrone B, only 9.4 % (22 mg) of the compound taken up was recovered and stems remained turgid but their lea¯ets became twisted and chlorotic and some abscized. Cells isolated from lea¯ets of 12 chickpea cultivars diered by up to ®ve-fold in their sensitivity to solanapyrone A and this compound was 2.6±12.6 times more toxic than solanapyrone B, depending on cultivar. Glutathione reacted with solanapyrone A in vitro reducing its toxicity in a cell assay and forming a conjugate. Measurement of reduced glutathione concentration and glutathione-S-transferase (GST) activity among cultivars showed that the dierences of their means were highly signi®cant and both were negatively and signi®cantly correlated with their sensitivity to solanapyrone A. Treatment of shoots with solanapyrone A enhanced total, reduced and oxidized glutathione content as well as GST activity 1.26-, 1.23-, 1.50- and 1.94-fold, respectively. Similarly, treatment of shoots with the safener, dichlormid, also raised total, oxidized and reduced glutathione levels and GST activity 1.42-, 1.07-, 1.43-, 1.42-fold, respectively. Cells isolated from shoots treated with dichlormid at 150 and 300 mg per shoot were 2.45 and 2.66 times less sensitive to solanapyrone A, with LD50 values of 71.5 and 77.8 mg ml ÿ1, respectively, as c 2000 Academic Press * compared to 29.2 mg ml ÿ1 for controls. Keywords: Citer arietinum; Ascochyta rabiei; solanapyrone toxins; glutathione; glutathione-S-transferase.
INTRODUCTION Blight, caused by Ascochyta rabiei, is one of the most serious diseases of chickpea [17]. The fungus attacks all aerial parts of the plant causing epinasty of petioles and young branches, followed by water-soaking and necrosis. When stems and petioles are girdled, they usually break. These symptoms are consistent with toxin production by the pathogen causing dysfunction of the host's membranes. Such dysfunction may cause plasma membranes to lose their selective permeability, resulting in the loss of turgor necessary for support of plant organs and the development of water-soaked lesions as the contents of cells leak into the intercellular spaces. Isolates of A. rabiei produced the toxins, solanapyrones A and C when grown in Czapek Dox liquid medium (CDLM) supplemented with chickpea seed extract [1] { Present address: Ayub Agricultural Research Institute, Jhang Road, Faisalabad, Pakistan. * To whom all correspondence should be addressed. E-mail: [email protected]
0885-5765/00/060235+10 $35.00/00
and also solanapyrone B when grown on CDLM supplemented with metal cations, of which Zn, Ca, Cu, Co and Mn, were the most important, rather than chickpea seed extract [3]. The toxins were separated and quanti®ed by HPLC on an analytical ODS column with an isocratic mobile phase determined by solvent optimization [4]. In order to determine the relative toxicities of solanapyrones A, B and C and their individual eects on chickpea genotypes, it was necessary to separate them in quantity and this was achieved using solvent partitioning and ¯ash chromatography on silica [8]. Although A. rabiei produces solanapyrones A, B and C in culture, only solanapyrone C has been claimed to have been found in infected plants [18], other workers failing to ®nd any of the three compounds [10]. There are several precedents for metabolism of phytotoxins by plants. For example, the tree pathogen, Heterobasidion annosum, produces a series of compounds with phytotoxic activity in culture, in particular fomajorin, fomajorins S and D, fomannosin and fomannoxin but up to now only c 2000 Academic Press *
236
K. Hamid and R. N. Strange
fomannoxin has been detected in vivo and isolated from naturally infected wood [9]. Cell cultures of Pinus sylvestris metabolize fommanoxin to fomannoxin alcohol and subsequently to fomannoxin acid b-glucoside both of which are less toxic than fomannoxin [22]. Since there are precedents for the metabolism of toxins by host tissues which may explain why some of these compounds cannot be extracted from plants infected by toxigenic organisms, experiments were conducted in order to determine how chickpea might metabolize solanapyrone A, the most toxic of the solanapyrone toxins. The reduced glutathione (GSH)/glutathione-S-transferase enzyme (GST) system is one mechanism by which plants detoxify xenobiotics [5, 13]. Accordingly, the reaction of solanapyrone A with glutathione and the concentrations of GSH (as well as oxidized GSH) and activity of GST in a range of chickpea cultivars were determined in order to ascertain if there were any correlation between them and the sensitivity of the cultivars to the toxin. Some safeners, such as dichlormid, prevent herbicide damage in cereals and some dicotyledonous species by enhancing the levels of GSH and GST. It was therefore of interest to determine whether chickpea was similarly aected and whether plants treated with safener were less sensitive to solanapyrone A.
MATERIALS AND METHODS
Production and isolation of the solanapyrone toxins Isolate PUT 7 of A. rabiei [14] was cultured without agitation for 12 days at 208C on a medium consisting of Czapek±Dox nutrients supplemented with the metal cations ZnSO4.7H2O, 0.05 g l ÿ1; CaCl2.2H2O, 0.1 g l ÿ1; CuCl2.2H2O, 0.02 g l ÿ1; CoCl2.6H2O, 0.02 g l ÿ1 and MnCl2.4H2O, 0.02 g l ÿ1 (250 ml ¯asks; 30 ml per ¯ask) and the toxins were separated as reported previously [8]. In brief, cultures were ®ltered through cheesecloth to remove mycelium and centrifuged to remove spores. The supernatant was acidi®ed to pH 3.0 and partitioned against ethyl acetate. Fractionation of the ethyl acetate phase was by ¯ash chromatography on a silica gel column (40 g: Biotage UK Ltd., Hertford, U.K.) with dichloromethane, cyclohexane, ethyl acetate (3 : 3 : 1 v/v/v), dichloromethane, cyclohexane, ethyl acetate (1 : 1 : 1 v/v/ v) and ®nally ethyl acetate as mobile phases. Fractions containing pure solanapyrones were recognized by their characteristic u.v. spectra [12].
Growth of plants Seeds of ®ve kabuli chickpea cultivars, ILC 3279, ILC 482, ILC 249, INRAT 88 and Kasseb as well as seeds of seven desi cultivars, AUG 424, CM 88, CM 68, 6153, C
44, C 235 and CM 72 were soaked for 12 h in water and planted in John Innes No. 2 compost in plastic pots (13 cm diam.: eight seeds per pot). Plants were raised in a greenhouse at 25 + 28C.
HPLC of solanapyrones Toxin samples were separated on a Philips HPLC equipped with a diode array detector essentially according to the method of Chen et al. [4] except that the solvent system consisted of methanol 23.1 %, water 56.3 % and tetrahydrofuran 20.6 % (v/v/v) which was pumped at a ¯ow rate of 1 ml min ÿ1. The stationary phase was an ODS column (Spherisorb ODS 2; 150 4.6mm i.d.; Jones Chromatography, Glamorgan, U.K.) and was protected by a guard column (20 4.6 mm i.d.). The solanapyrones were recognized by their retention times and UV spectra which were compared by superimposition on those of authentic samples. For quanti®cation of the solanapyrones, chromatograms were abstracted from the three dimensional chromascans (absorption wavelength time) at lmax 327 nm, lmax 303 nm and lmax 320 nm for solanapyrones A, B and C, respectively and the areas under the peaks were compared with those of standard solutions of the authentic compounds.
Toxicity of solanapyrones A and B The toxicity of solanapyrones A and B was determined qualitatively by allowing triplicate samples of shoot cuttings to take up the compounds. Ethanolic solutions of solanapyrone A and solanapyrone B were placed in separate glass vials [58 17 mm (diam.)] and, after evaporation of the ethanol, distilled water (4 ml) was added to each vial and vortexed for 3 min to give 450 mM solutions of the toxins. Weighed shoots of chickpea (ILC 3279: 3 weeks old) were placed in each vial and incubated in a greenhouse at 25 + 28C for 24 or 72 h. Transpiration was aided by an electric fan. Shoots incubated in water served as controls. All vials were topped up with water as required to keep the cut ends below the surface. In some experiments, after incubation for 24 h, the shoots were weighed again and, in order to recover any solanapyrones remaining in the vials, their contents, together with three washings of methanol (0.5 ml each) were transferred to universal bottles where they were diluted with distilled water to 25 ml. Solanapyrones in these solutions were adsorbed onto end-capped Isolute cartridges (1 g: C18; International Sorbent Technology, Hengoed, Glamorgan, U.K.) and, after washing with water (2 ml), the cartridges were eluted in acetonitrile (2 ml). Solanapyrones in samples (20 ml) of the acetonitrile eluates were quanti®ed by HPLC.
Phototoxicity and metabolism of solanapyrones in the chickpea The toxicity of solanapyrones A and B was also determined quantitatively with suspensions of chickpea cells (density 0.2 A units at l 620 nm 2.25 105 cells ml ÿ1) isolated from lea¯ets of newly-opened leaves of chickpea plants (33±55 days old) and expressed as units essentially as previously described [14]. In brief, after dispensing a two-fold dilution series of toxin preparation in buered osmoticum (holding buer) among wells of a microtest plate (50 ml well ÿ1), a suspension of cells, freshly isolated from chickpea lea¯ets by enzymic digestion, was added to each well (50 ml well ÿ1) and the microtest plate incubated for 3 h at 258C. Fluorescein diacetate solution was added to each well (50 ml well ÿ1) and ¯uorescent (live) and non-¯uorescent (dead) cells scored under an inverted ¯uorescence microscope. Probit percent cell death, corrected for viability in control wells without toxin, was plotted against the log2 toxin concentration to give a straight line from which the LD50 value, arbitrarily de®ned as 1 unit of activity, was extracted. Since sensitivity varied among assays done on dierent days, results of tests on cultivars of chickpea were expressed relative to that of solanapyrone A acting on cells of a standard cultivar, ILC 3279, which was included in all assays. In some assays, test cells were isolated from lea¯ets of shoots treated with dichlormid as described below and tested for their sensitivity to solanapyrone A, cells isolated from lea¯ets of shoots incubated in water serving as controls.
Metabolism of solanapyrone A by chickpea shoots Shoots allowed to take up the compounds as described above were cut into small pieces with scissors and homogenized in a Sorvall Omni-mixer with 80 % ethanol (3 10 ml) for 5 min. The homogenate (30 ml) was centrifuged at 3000 g for 5 min and the supernatant evaporated to dryness on a rotary evaporator at 308C. Residues were dissolved in methanol (1.5 ml) and transferred to universal bottles. After dilution with distilled water to 25 ml, hydrophobic compounds were puri®ed by solid phase extraction as above and samples (20 ml) of the acetonitrile eluates examined by HPLC. The remaining acetonitrile eluates were concentrated on a rotary evaporator to 500 ml and samples were spotted on silica TLC plates (Silica gel 60 F254 , Merck, Germany). The plates were developed in cyclohexane/ dichloromethane/ethyl acetate (1 : 1 : 1 v/v/v) and, after evaporation of the solvent, were observed under short wavelength u.v. light. Spots at Rf 0.81 were scraped from the TLC plates, placed in Eppendorf tubes (1.5 ml) and vortexed in acetonitrile (3 1 ml). Supernatants were combined and evaporated to dryness on a rotary evaporator and the residues dissolved in the same solvent (50 ml). Samples (20 ml) were spotted on TLC plates (Silica gel 60 F254) and the plates were developed in
237
cyclohexane/dichloromethane/ethyl acetate (1 : 1 : 1 v/v/ v). Plates were viewed under short wavelength u.v. light and were also sprayed with 2,4-dinitrophenylhydrazine (50 mg in 10 ml ethanol). Samples (20 ml) were also subjected to HPLC (see above for details).
Reaction of glutathione with solanapyrone A Tris-HCl buer (10 mM, pH 8) containing 50 mM glutathione was dispensed in 24 glass vials [58 17 mm (diam.); 2 ml per vial]. Solanapyrone A (272 mg per vial in 9.3 ml ethanol) was added to each vial and vortexed for 30 s before sterilizing by ®ltration through 0.22 mm ®lters (MSi, Microseparations Inc, U.S.A.). In controls, solanapyrone A was incubated in Tris-HCl buer without glutathione. Both sets of vials were incubated in the dark at 258C for 0, 1, 2, 3, 4, 5 and 6 h. Samples were injected directly onto the HPLC. In order to determine the reaction product of glutathione and solanapyrone A, the toxin was incubated in Tris-HCl buer ( pH 8) containing 500 mM glutathione for 16 h at 258C. After incubation and freeze-drying, the product was analysed by fast atom bombardment mass spectrometry.
The eect of glutathione on the toxicity of solanapyrone A The eect of glutathione on the toxicity of solanapyrone A was measured in the cell assay either by determining the eect of a constant standard solution of glutathione (50 mM) on the dose-response curve of solanapyrone A or, conversely, by determining the eect of a dilution series of glutathione on the toxicity of a standard solution of solanapyrone A. In the ®rst assay, an ethanolic solution of solanapyrone A was added to well 1 of a microtest plate so that, on evaporation of the ethanol, the well contained 15.1 mg of the compound. Holding buer (100 ml) containing 50 mM glutathione was added to the ®rst well (giving a 500 mM solution of solanapyrone A) and a two-fold dilution series in the glutathione holding buer was made across the plate. Controls contained no solanapyrone A. Another plate with the same dilution series of solanapyrone A but without glutathione was also set up for comparison. Plates were incubated for 16 h at 258C before adding cells isolated from chickpea lea¯ets (cv. ILC-3279: 3 weeks old). Viability of the cells was scored after incubation for 3 h as usual. In the second assay, a two-fold dilution series of glutathione in holding buer with 50 mM as the highest concentration was made in a microtest plate (50 ml well ÿ1). Solanapyrone A dissolved in ethanol (3 ml) was added to each well to give a ®nal concentration of 250 mg ml ÿ1 (828 mM). Control wells contained only holding buer or glutathione (50 mM). Plates were incubated for 16 h at 258C before adding cell suspension from chickpea lea¯ets (cv. ILC
238
K. Hamid and R. N. Strange
3279: 3 weeks old: 50 ml well ÿ1), giving a ®nal concentration of solanapyrone A of 414 mM. Viability of cells was scored after incubation for 3 h.
Quantitative analysis of glutathione in chickpea lea¯ets Reduced glutathione (GSH) and oxidized glutathione (GSSG) were measured essentially according to the methods of Anderson [2] and Coleman et al. [6]. Lea¯ets (0.5 g) of chickpea cultivars (21±25 days old) were pulverized in liquid nitrogen with a pestle and mortar. The resulting powder was suspended in extraction buer (2 ml) consisting of 5 % (w/v) sulphosalicylic acid (SSA: Sigma) and 6.3 mM diethyltriaminepentaacetic acid (DETAPAC: Sigma), vortexed for 45 s, kept on ice for 10 min, centrifuged for 12 min at 10 000 g and further centrifuged at 10 000 g for 3 min at 48C to remove ¯oating particles. Supernatant (50 ml) was added to 750 ml of 0.143 M potassium phosphate buer containing 6.3 mM DETAPAC ( pH 7.5), 100 ml of 6 mM 5,5-dithiobis (2-nitrobenzoic acid) (DTNB) and 100 ml of 2.1 mM bnicotinamide adenine dinucleotide phosphate in the reduced form (b-NADPH). The reaction was initiated by the addition of 25 ml of glutathione reductase (20 U ml ÿ1: Sigma) and the change in absorption was measured at 412 nm for 6 min on a spectrophotometer. A standard curve with a concentration range of 0±45 mM GSH was plotted and used to determine the concentration of total glutathione (GSH GSSG) which was expressed as nmoles g ÿ1 fresh wt of lea¯ets [6]. For the measurement of oxidized glutathione (GSSG), supernatant (400 ml) was mixed with 8 ml of 2-vinylpyridine (2VP) and 40 ml of triethanolamine (TEA: Sigma). The mixture was vortexed for 15 s and incubated at 258C for 1 h before assaying 50 ml aliquots as for total glutathione. GSSG was determined as described for total glutathione (as before) from a standard curve with a concentration range of 0±40 mM GSSG. The amount of GSSG was calculated and expressed as nmoles g ÿ1 fresh wt of lea¯ets. Reduced glutathione (GSH) was calculated by subtracting the amount of GSSG from total glutathione (GSSG GSH) and also expressed as nmoles g ÿ1 fresh wt of lea¯ets.
Extraction and measurement of glutathione-S-transferase (GST) activity in chickpea lea¯ets GST was extracted by a modi®cation of the method of Hunaiti and Ali [11]. Lea¯ets (0.5 g) from each cultivar were ground in liquid nitrogen in a pestle and mortar until a ®ne powder was obtained. All further steps were carried out at 0±48C. The resulting powder was suspended in 0.1 M potassium phosphate buer ( pH 7.0: 0.5 ml), containing polyvinylpyrrolidone (PVP-40, 5 %: Sigma)
vortexed for 1 min and centrifuged at 15 000 g for 15 min. After diluting the supernatant to 1 g of tissue per ml with extraction buer, it was passed through a 0.45 mm ®lter (Gelman Sciences, U.S.A.) to remove ¯oating particles and the ®ltrate was tested for GST activity. GST activity was measured by the formation of the conjugate of GSH and 1-chloro-2,4-dinitrobenzene (CDNB) determined spectrophotometrically at 340 nm [19, 7]. The assay mixtures (3 ml) consisted of 0.1 M phosphate buer ( pH 6.5, 2865 ml), CDNB (30 ml in ethanol, ®nal concentration 0.1 mM), GSH (75 ml, ®nal concentration 2.5 mM) and the reaction was started by adding the enzyme solution (30 ml). A complete assay mixture without enzyme served as a control. The reaction was monitored spectrophotometrically by the increase in absorbance at 340 nm for 6 min {e340 of the conjugate 10 mM ÿ1 cm ÿ1; [15]}. Units of enzyme activity were calculated by subtracting the units of activity obtained in controls without enzyme and one unit of activity was de®ned as the amount of enzyme that catalysed the formation of 1 mmol of S-2,4-dinitrophenylglutathione min ÿ1 at room temperature. Glutathione and GST activity were also measured in shoots treated with solanapyrone A. Solanapyrone A (60.4 mg in 2 ml ethanol) was vortexed with 2 ml water in polypropylene centrifuge tubes [115 mm 30 mm (diam.)] to give a 100 mM solution. Shoots (0.75 g) of chickpea (cv. ILC 3279: 3 weeks old) were placed in the tubes and were allowed to take up 1.5 ml of the solution ( 45.3 mg solanapyrone A) under greenhouse conditions (23 + 28C: 5 to 6 h incubation period). After uptake of the toxin, shoots were transferred to tubes containing distilled water (25 ml) and incubated for a further 96 h under the same conditions. The water level was maintained throughout the incubation period. Shoots incubated in distilled water without solanapyrone A served as controls. After the incubation period, the parts of shoots covered by solanapyrone A solution or water were discarded and the entire remaining material approx. 0.5 g was ground in liquid nitrogen and used for the estimation of glutathione and GST activity. In similar experiments, shoots were allowed to take up 1.5 ml of solutions of dichlormid (100 and 200 mg ml ÿ1 prepared from a stock solution of 10 mg ml ÿ1 in acetone 150 or 300 mg per shoot). Glutathione and GST activity were extracted from test shoots and controls (incubated in water) and quanti®ed as already described.
RESULTS
The eects of incubating solanapyrones A and B with chickpea shoots When shoots were incubated in solanapyrone A (544 mg 4 ml ÿ1 H2O: 450 mM) for 72 h, the stems became
Phototoxicity and metabolism of solanapyrones in the chickpea
239
F I G . 1. Shoots of chickpea (cultivar ILC 3279) after taking up 45.3 mg of solanapyrone A and incubating for a further 96 h in water. Note breakage of stem just below the uppermost leaf.
shrivelled, brown and corky and the lea¯ets developed ¯ame-shaped chlorotic zones. In a dierent experiment when shoots were allowed to take up 45.3 mg of solanapyrone A and incubated for a further 96 h in water, bleaching at the base of the stems occurred and stems broke just below the uppermost leaf (Fig. 1). Controls incubated in water remained green and turgid. Incubation of shoots in solanapyrone B (547 mg 4 ml ÿ1 H2O: 450 mM) for 72 h produced dierent symptoms: lea¯ets became wilted and chlorotic and some of them abscized, whereas those of controls remained green and turgid. Both solanapyrone A and solanapyrone B were degraded when incubated in water at 25 + 28C. Losses of solanapyrone A were 152.3 from 544 mg contained in 4 ml water (450 mM) and 93.5 mg of solanapyrone B from 547 mg also contained in 4 ml water (450 mM) over a period of 24 h. When chickpea shoots were incubated in aqueous solutions of solanapyrone A (450 mM) or solanapyrone B (450 mM) for 24 h in the greenhouse at 25 + 28C, they lost weight. The loss was highly signi®cant, 16.3 % (P 5 0.005, 100 % 573 mg) in the case of solanapyrone A but non-signi®cant, 6.25 % (P 4 0.05, 100 % 426.7 mg) in the case of solanapyrone B. Losses of the solanapyrones from the solutions were greater when they were incubated with shoots and amounted to an extra 175.6 mg for solanapyrone A and an extra 234.0 mg for solanapyrone B, both losses being highly signi®cant (P 5 0.01). No solanapyrone A was recovered from shoots and only 22.0 + 12.6 mg of solanapyrone B.
A new compound with lmax of 238 and 288 nm was found in extracts of shoots treated with solanapyrone A. It had a retention time of 779 s on HPLC, an Rf of 0.81 on TLC and reacted with 2,4-dinitrophenylhydrazine to give a brown spot. However, when the area of silica corresponding to that occupied by this spot (and not sprayed with 2,4-dinitrophenylhydrazine) was scraped from the TLC plate, eluted in methanol and chromatographed on the HPLC again, two peaks occurred, one with a retention time which was close to that of the new compound originally chromatographed and with a UV spectrum that matched it better than 95 %. The other compound had a longer retention time (1047 s) and a u.v. spectrum that was a 97 % match with demethylated solanapyrone A, a degradation product in which the methyl moiety of the methoxy group was replaced by hydrogen. Bioassay of the demethylated compound compared with solanapyrone A gave LD50 values of 514.0 + 55.0 mg ml ÿ1 and 31.3 + 2.1 mg ml ÿ1, respectively. Production of the demethylated compound also occurred on incubation with either 0.5 M NH4OH for 1 h at 508C or a mineral salts medium (consisting of NaHPO4 7.0 g, KH2PO4 3.0 g, NaCl 0.5 g, NH4Cl 1.0 g, CaCl2 0.011 g and MgSO4 0.2 g dissolved in 1 l water and adjusted to a pH of 7.25) at 378C for 24 h or longer.
The relative sensitivity of chickpea cultivars to solanapyrones A and B Although the cell assay was a convenient method for measuring the sensitivity of chickpea cultivars to the
K. Hamid and R. N. Strange Sensitivity (ILC 3279 to Solanapyrone A = 1)
240 5 4.5
Solanapyrone A Solanpyrone B
4 3.5 3 2.5 2 1.5 1 0.5 0
AUG 424
ILC 249
6153
C 44
ILC 482 CM 88
CM 68
C 235
ILC 3279
INRAT 88
CM 72
Kasseb
Cultivar
F I G . 2. Sensitivity of chickpea cultivars to solanapyrones A and B relative to the sensitivity of cv. ILC 3279 to solanapyrone A ( 1).
Formation of an adduct of solanapyrone A with glutathione One reason for the dierential sensitivity of cultivars to solanapyrone A could lie in their ability to detoxify the compound, possibly through the glutathione/glutathioneS-transferase system. When a concentration range of
9
6153
AUG 424
8
Disease rating
solanapyrones, the day to day variation in the assay was considerable. For example, the minimum concentration of solanapyrone A required to kill 50 % of the cells of cultivar ILC 3279 was 10.1 + 1.0 mg ml ÿ1 and the maximum 93.3 + 13.8 mg ml ÿ1. For this reason ILC 3279 was included in each assay as an internal standard and the sensitivity of other cultivars measured relative to it. The dierences of means of relative sensitivity to solanapyrone A and solanapyrone B among the chickpea cultivars were highly signi®cant for both toxins (P 0.001). Cultivars ILC 249, AUG 424 and 6153 were the most sensitive to solanapyrone A and cultivars Kasseb and CM 72 the least. The range of sensitivity was 0.65±3.3 on a scale in which ILC 3279 was rated as 1. Cultivars were less sensitive to solanapyrone B, the range being 0.14±0.46 on the same scale (i.e. sensitivity of ILC 3279 to solanapyrone A 1). Solanapyrone A was 2.62±12.64 times more toxic than solanapyrone B depending upon the cultivar (Fig. 2). Comparison of the relative sensitivities of cultivars to solanapyrone A with their disease ratings to Ascochyta blight showed that those which were most sensitive, such as 6153 and AUG 424, were also the most susceptible to the disease, scoring 9 on a scale of 1±9 [20], while less sensitive cultivars such as Kasseb, CM 72 and INRAT 88 scored 4.5, 6 and 4, respectively (Fig. 3). Spearman's correlation coecient values between the susceptibility of the cultivars to A. rabiei and their relative sensitivity to solanapyrone A ( 0.5166) and to solanapyrone B ( 0.5229) were positive but non-signi®cant.
7 6
C 235
CM 72
C 44
5 Kasseb
4 3 0.25 0.5
0.75
INRAT 88 ILC 3279
1
1.25
ILC 482
1.5 1.75
2
2.25
2.5 2.75
3
3.25 3.5 3.75
Relative sensitivity to solanapyrone A
F I G . 3. Relationship of the susceptibility of nine cultivars of chickpea to Ascochyta blight with their sensitivity to solanapyrone A (cv. ILC 3279 1).
glutathione was included in an assay with cells from ILC 3279 and 414 mM solanapyrone A (a concentration which killed all the cells), only the highest concentration of glutathione tested (50 mM) reduced cell death to less than 50 %. Conversely, when this concentration of glutathione was included in the assay the amount of solanapyrone A required to reach the LD50 value was double that without the addition (215.3 mM compared with 100.7 mM). When solanapyrone A (272 mg) was incubated in 10 mM Tris-HCl buer (2 ml, 10 mM, pH 8) for 6 h, recovery of the compound was 72.1 + 6.9 % but only 11.9 + 1.5 % when the buer contained 50 mM glutathione, most of the loss (69.5 + 1.7 %) occurring within the ®rst hour, suggesting that solanapyrone A spontaneously formed a conjugate with glutathione. Further evidence for this was obtained when the two compounds were incubated together overnight at 258C in 10 mM Tris-HCl buer ( pH 8) and the product subjected to fast atom bombardment mass spectrometry. A compound was
Phototoxicity and metabolism of solanapyrones in the chickpea
241
obtained with a molecular ion of 606 which, when accurately measured, gave a mass of 606.214300 corresponding to a formula of C28H36N3O10S (calculated 606.212142) and suggesting the conjugate shown in Fig. 4. This, however, is 1 Dalton greater than expected from the mass spectrum. One possible interpretation of this result is that the protons of the two terminal carboxyl groups on the glutathione moiety of the molecule are lost and the molecular ion is for M 1 of this species as usually obtained for fast atom bombardment mass spectrometry.
Levels of total, oxidized (GSSG) and reduced glutathione (GSH) in chickpea cultivars
Glutathione (n moles g fresh wt leaflets−1)
Measurement of GSH in chickpea cultivars showed that the dierences of their means were highly signi®cant (P 0.001). Cultivar AUG 424 had the least and Kasseb the most (200.7 + 27.2 and 561.0 + 112.9 nmoles g ÿ1 fresh wt of lea¯ets, respectively: Fig. 5). Comparison of the means of reduced glutathione levels of the cultivars at the 95 % con®dence level separated the cultivars into six overlapping groups in which cultivars within a group did not dier signi®cantly from each other (Table 1). Sensitivity to solanapyrone A was inversely related to GSH levels, cultivars that were least sensitive such as Kasseb and CM 72 having GSH concentrations that were 1.7± 2.8 times greater than those of the most sensitive cultivars such as 6153 and AUG 424. When all 12 cultivars were analysed, a Spearman's correlation coecient value (rs) of ÿ0.7323: P 5 0.01 was obtained. Cultivars had low levels of oxidized glutathione which ranged from 11.9 + 5.2 to 67.2 + 19.3 nmoles g ÿ1 fresh wt of lea¯ets (Fig. 4).
F I G . 4. The proposed structure of the solanapyrone A-glutathione adduct.
Glutathione S-transferase (GST) activity in chickpea cultivars The dierences of the means of GST activity among 12 cultivars of chickpea measured as units were highly signi®cant (P 0.001). Cultivar AUG 424 had the least and INRAT-88 the most, 40.1 + 3.8 and 66.8 + 3.9 units of activity g ÿ1 fresh wt of lea¯ets, respectively. At the 95 % con®dence level, cultivars were separated into ®ve overlapping groups in which those within a group did not dier signi®cantly from each other (Table 2). Sensitivity to solanapyrone A was inversely related to GST activity, the activity of cultivars such as Kasseb and INRAT-88 which were least sensitive, being 1.40±1.66 times greater
900
Total Glutathione GSH GSSG
800 700 600 500 400 300 200 100 0
AUG 424
6153
CM 88 ILC 249
C 235 ILC 482
INRAT 88
ILC 3279
CM 68
C 44
CM 72
Cultivars
F I G . 5. Total, reduced and oxidized glutathione of 12 chickpea cultivars.
Kasseb
242
K. Hamid and R. N. Strange
T A B L E 1. Reduced glutathione (n moles g ÿ1 fresh wt of lea¯ets) of 12 chickpea cultivars
T A B L E 2. GST activity (units) of chickpea cultivars g ÿ1 fresh wt of lea¯ets
Cultivars
Cultivars
Kasseb CM 72 C-44 CM-68 ILC 3279 INRAT 88 ILC 482 C-235 ILC 249 CM 88 6153 AUG 424
GSH content 561.0 + 112.9 A 512.7 + 83.0 AB 489.9 + 28.0 AB 482.1 + 103.2 AB 462.5 + 18.1 AB 440.3 + 88.7 BC 422.3 + 35.9 BCD 411.6 + 42.7 BCD 354.7 + 31.6 CDE 331.4 + 37.2 DE 298.9 + 7.8 EF 200.7 + 27.2 F
INRAT-88 Kasseb ILC 3279 CM 72 CM 88 C-235 ILC 249 CM-68 C-44 6153 ILC 482 AUG 424
Units of activity g ÿ1 fresh wt of tissue 66.8 + 3.9 57.9 + 5.1 54.6 + 5.1 54.5 + 3.9 52.3 + 6.6 51.2 + 6.9 50.1 + 3.8 49.0 + 5.8 49.0 + 3.3 41.2 + 5.1 41.2 + 1.9 40.1 + 3.8
A B BC BC BC BC BC CD CD DE DE E
Analysis of variance showed P 0.001. Means followed by the same letters are not signi®cantly dierent at the 95 % con®dence level (LSD 103.9).
Analysis of variance showed P 0.001. Means followed by the same letters are not signi®cantly dierent at the 95 % con®dence level (LSD 8.093).
than that of the most sensitive cultivars such as 6153 and AUG 424. When all 12 cultivars were analysed, a Spearman's correlation coecient value (rs) of ÿ0.8094: P 5 0.01 was obtained.
prevalent. Two trivial but persistent reasons for this are the lack of an agreed scale for scoring the disease and the lack of a pictorial key to facilitate this. A third and more fundamental reason is the ability of the pathogen to sporulate copiously and therefore cause explosive epidemics [17]. A fourth and also fundamental reason is the lack of knowledge of the attributes which make the pathogen so virulent. One possible virulence factor is the production of toxins which kill the host. Initially, two toxins, solanapyrones A and C were isolated from culture ®ltrates of A. rabiei when grown on Czapek±Dox nutrients supplemented with chickpea seed extract [1]. Later, using a de®ned medium and a mobile phase for HPLC, determined by solvent optimization, a third toxin, solanapyrone B was detected [4]. All three toxins were originally isolated from culture ®ltrates of Alternaria solani, the cause of early blight of potatoes [12]. Evidence that these toxins play important roles in either disease is lacking but, in the case of Ascochyta blight of chickpea, two observations indicate that they might do so. First, all reliably identi®ed isolates of the fungus produce at least one of the toxins in culture, including 20 isolates from Turkey whose identity was con®rmed by the sequences of the ITS1 and ITS2 regions of their rDNA (Dolar and Strange, unpublished results). Since the fungus reproduces sexually on debris remaining on the ground after harvest [21], loss of toxigenicity would be likely to occur were it not necessary for pathogenicity or virulence. Secondly, symptoms of the disease, epinasty, chlorosis, necrosis and breakage of stems, are mimicked by the toxins. In this paper, further evidence for the role of the solanapyrone toxins in Ascochyta blight of chickpea is reported. The mimicking of symptoms by pure preparations of solanapyrones A and B has been con®rmed. In particular, the breakage of stems caused by solanapyrone
The eect of solanapyrone A and dichlormid on glutathione concentration and glutathione S-transferase activity in shoots treated with the compounds When shoots of cv. ILC 3279 were allowed to take up solanapyrone A (45.3 mg per shoot), there was a signi®cant increase in glutathione content (P 5 0.05). The levels of total, oxidized and reduced glutathione increased from 672 + 55, 60 + 7 and 612 + 99 nmoles g ÿ1 fresh wt of shoots, respectively to 848 + 92, 90 + 15 and 758 + 82 nmoles g ÿ1 fresh wt. Treatment of shoots with solanapyrone A (45.3 mg per shoot) also caused a signi®cant 1.9-fold increase in GST activity from 57.9 + 10.0 units g ÿ1 fresh wt in controls to 112.3 + 18.6 units g ÿ1 fresh wt in tests. When shoots were allowed to take up dichlormid (300 mg per shoot), their reduced glutathione levels rose signi®cantly (P 5 0.001) from 304.3 + 41.7 to 436.5 + 68.6 nmoles g ÿ1 fresh wt of shoots. Some enhancement of GST activity was also noted but it was not signi®cant. Cells isolated from lea¯ets of shoots treated with 150 and 300 mg per shoot were 2.45 and 2.66 times less sensitive to solanapyrone A, respectively, than those isolated from controls incubated in water only.
DISCUSSION Although blight of chickpea has received considerable attention from plant pathologists and plant breeders, devastating attacks by the pathogen, A. rabiei, are still
Phototoxicity and metabolism of solanapyrones in the chickpea A is illustrated in Fig. 1. This is a characteristic symptom of the disease and may be explained by the loss of turgor of parenchyma cells surrounding the stele as a consequence of the attack by the toxin on their plasma membranes. Such stems would have only the inadequate support of their steles and not the additional strength imparted by the turgor of the surrounding cells. That the integrity of the plasma membrane of chickpea cells is attacked by the toxins is demonstrated by the bioassay. This involves treating lea¯ets with enzymes and agitation, processes which result predominantly in the production of separated cells but also a few protoplasts. Cells isolated from dierent cultivars of the plant diered in their sensitivity and, disappointingly, variation from day to day in the assay was considerable. Nevertheless, when a single cultivar was included as an internal standard, it was possible to place the 12 cultivars in these experiments in order. Comparison of this order with the disease scores available for nine of the cultivars showed that there was a positive correlation between sensitivity to solanapyrones A and B and severity of disease symptoms although this was not signi®cant. Whether this relationship will eventually prove to be signi®cant will depend on the establishment of a reliable and standard method for scoring disease and extension of the experiment to other cultivars diering in resistance to the pathogen. In this paper the reason for dierences in sensitivity of the cultivars to solanapyrone A was sought. The glutathione/glutathione-S-transferase system is one mechanism by which plants detoxify xenobiotics and some evidence consistent with the detoxi®cation of solanapyrone A by this mechanism was obtained. Firstly, inclusion of glutathione in the assay reduced toxin titres. Secondly, solanapyrone A formed an adduct with glutathione spontaneously. Thirdly, there was an inverse and signi®cant correlation of solanapyrone A sensitivity with both glutathione content and glutathione-S-transferase activity. Fourthly, boosting the glutathione concentration and glutathione-S-transferase activity of chickpea by treatment of plants with the safener, dichlormid, decreased the sensitivity of their cells to solanapyrone A in the bioassay. Demonstration of the adduct in planta may prove dicult, however, since glutathione conjugates of xenobiotics are often rapidly metabolized [16]. It is now important to determine unequivocally whether or not the solanapyrone toxins have a role to play in Ascochyta blight of chickpea. The most persuasive evidence would come from the creation of toxin-minus mutants whose virulence could be restored by the addition of toxin. Ideally, aected loci of such mutants would only occur in genes close to the end of the biosynthetic sequence leading to the toxins. The acquisition of such data would validate the use of the solanapyrones as an aid to the selection of chickpea genotypes resistant to Ascochyta blight.
243
It is unfortunate that the cell bioassay showed such great variation from day to day as this may impair its use as a screen for resistance, although use of one cultivar as an internal standard in all tests, as described in the experiments reported in this paper, may mitigate this. Since the glutathione/glutathione-S-transferase system may be the mechanism by which the plant detoxi®es solanapyrone A, the most toxic of the three solanapyrones, another possibility is to select plants in which this system is highly expressed either constitutively or in response to challenge by the toxins or the pathogen. Finally, the fact that this system may be boosted by safeners, such as dichlormid, would also seem to be worth further investigation.
REFERENCES 1. Alam SS, Bilton JN, Slawin AMZ, Williams DJ, Sheppard RN, Strange RN. 1989. Chickpea blight: production of the phytotoxins solanapyrones A and C by Ascochyta rabiei. Phytochemistry 28: 2627±2630. 2. Anderson ME. 1985. Tissue glutathione. In: Greenwald RA, ed. CRC Handbook of Methods Oxygen Radical Research. Boca Raton, Florida: CRC Press Inc., 317±323. 3. Chen Y, Strange RN. 1991. Synthesis of the phytotoxins by Ascochyta rabiei in response to metal cations and development of a de®ned medium for toxin production. Plant Pathology 40: 401±407. 4. Chen YM, Peh EK, Strange RN. 1991. Application of solvent optimisation to the separation of the phytotoxins, solanapyrones A, B and C from culture ®ltrates of Ascochyta rabiei. Bioseparation 2: 107±113. 5. Coleman JOD, Blake-Kalff MMA, Davies TGE. 1997. Detoxi®cation of xenobiotics by plants: chemical modi®cation and vacuolar compartmentation. Trends in Plant Science 2: 144±151. 6. Coleman JOD, Randal R, Blake-Kalf MMA. 1997. Detoxi®cation of xenobiotics in plant cells by glutathione conjugation and vacuolar compartmentalization: a ¯uorescent assay using monochlorobimane. Plant Cell and Environment 20: 449±460. 7. Habig WH, Pabst MJ, Jakoby WB. 1974. Glutathione Stransferases. The ®rst enzymatic step in mercapturic acid formation. Journal of Biological Chemistry 249: 7130±7139. 8. Hamid K, Strange RN. 1997. An easy method for isolating the solanapyrone toxins from culture ®ltrates of Ascochyta rabiei. International Chickpea and Pigeonpea Newsletter 4: 20±22. 9. Heslin MC, Stuart MR, Murchu PO, Donnelly DMX. 1983. Fomannoxin, a phytotoxic metabolite of Fomes annosus: in vitro production, host toxicity and isolation from naturally infected Sitka spruce heartwood. European Journal Forest Pathology 13: 11±23. 10. HoÈhl B, Weidemann C, HoÈhl U, Barz W. 1991. Isolation of solanapyrones A, B and C from culture ®ltrates and spore germination ¯uids of Ascochyta rabiei and aspects of phytotoxin action. Journal of Phytopathology 132: 193±206. 11. Hunaiti AA, Ali BR. 1990. Glutathione S-transferase from oxidiazon treated chickpea. Phytochemistry 29: 2431±2435.
244
K. Hamid and R. N. Strange
12. Ichihara A, Tazaki H, Sakamura S. 1983. Solanapyrones A, B and C. Phytotoxic metabolites from the fungus Alternaria solani. Tetrahedron Letters 24: 5373±5376. 13. Lamoureux GL, Richard H, Shimabukuro D, Frear DS. 1991. Glutathione and glucoside conjugation in herbicide selectivity. In: Caseley JC, Cussans GW, Atkin RK, eds. Herbicide Resistance in Weed and Crop. Oxford: ButterworthHeinemann Ltd. 14. Latif Z, Strange RN, Bilton J, Riazuddin S. 1993. Production of the phytotoxins, solanapyrones A and C and cytochalasin D among nine isolates of Ascochyta rabiei. Plant Pathology 42: 172±180. 15. Mannervik B, Guthenberg C. 1981. Glutathione transferase (Human placenta). In: Jakoby WB, ed. Methods in Enzymology (Detoxi®cation and Drug Metabolism: Conjugation and Related Systems), Vol. 77. Academic Press, Inc., 231±235. 16. Marrs KA. 1996. The functions and regulation of glutathione S-transferases in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47: 127±158.
17. Nene YL. 1982. A review of Ascochyta blight of chickpea. Tropical Pest Management 28: 61±70. 18. Shahid AA, Riazuddin S. 1998. Presence of solanapyrone C in blight infected chickpea plants. International Chickpea and Pigeonpea Newsletter 5: 15±16. 19. Simons PC, Jagt DLV. 1977. Puri®cation of glutathione S-transferase from human liver by glutathione-anity chromatography. Analytical Biochemistry 82: 334±341. 20. Singh KB, Hawtin GC, Nene YL, Reddy MV. 1981. Resistance in chickpeas to Ascochyta rabiei. Plant Disease 65: 586±587. 21. Trapero-Casas A, Navas-Cortes A, Jimenez-Diaz RM. 1996. Airborne ascospores of Didymella rabiei as a major primary inoculum for Ascochyta blight epidemics in chickpea crops in southern Spain. European Journal of Plant Pathology 102: 237±245. 22. Zweimuller M, Antus S, Kovacs T, Sonnenbichler J. 1997. Biotransformation of the fungal toxin fomannoxin by conifer cell cultures. Biological Chemistry 378: 915±921.
Phytotoxicity of solanapyrones A and B produced by the chickpea pathogen Ascochyta rabiei (Pass.) Labr. and the apparent metabolism of solanapyrone A by chickpea tissues K H A L I D H A M I D { and R I C H A R D N . S T R A N G E * Department of Biology, University College London, Gower Street, London WC1E 6BT, U.K. (Accepted for publication March 2000) When chickpea shoots were placed in solanapyrone A, the compound could not be recovered from the plant and symptoms developed. These consisted of loss of turgor, shrivelling and breakage of stems and ¯ame-shaped, chlorotic zones in lea¯ets. In similar experiments with solanapyrone B, only 9.4 % (22 mg) of the compound taken up was recovered and stems remained turgid but their lea¯ets became twisted and chlorotic and some abscized. Cells isolated from lea¯ets of 12 chickpea cultivars diered by up to ®ve-fold in their sensitivity to solanapyrone A and this compound was 2.6±12.6 times more toxic than solanapyrone B, depending on cultivar. Glutathione reacted with solanapyrone A in vitro reducing its toxicity in a cell assay and forming a conjugate. Measurement of reduced glutathione concentration and glutathione-S-transferase (GST) activity among cultivars showed that the dierences of their means were highly signi®cant and both were negatively and signi®cantly correlated with their sensitivity to solanapyrone A. Treatment of shoots with solanapyrone A enhanced total, reduced and oxidized glutathione content as well as GST activity 1.26-, 1.23-, 1.50- and 1.94-fold, respectively. Similarly, treatment of shoots with the safener, dichlormid, also raised total, oxidized and reduced glutathione levels and GST activity 1.42-, 1.07-, 1.43-, 1.42-fold, respectively. Cells isolated from shoots treated with dichlormid at 150 and 300 mg per shoot were 2.45 and 2.66 times less sensitive to solanapyrone A, with LD50 values of 71.5 and 77.8 mg ml ÿ1, respectively, as c 2000 Academic Press * compared to 29.2 mg ml ÿ1 for controls. Keywords: Citer arietinum; Ascochyta rabiei; solanapyrone toxins; glutathione; glutathione-S-transferase.
INTRODUCTION Blight, caused by Ascochyta rabiei, is one of the most serious diseases of chickpea [17]. The fungus attacks all aerial parts of the plant causing epinasty of petioles and young branches, followed by water-soaking and necrosis. When stems and petioles are girdled, they usually break. These symptoms are consistent with toxin production by the pathogen causing dysfunction of the host's membranes. Such dysfunction may cause plasma membranes to lose their selective permeability, resulting in the loss of turgor necessary for support of plant organs and the development of water-soaked lesions as the contents of cells leak into the intercellular spaces. Isolates of A. rabiei produced the toxins, solanapyrones A and C when grown in Czapek Dox liquid medium (CDLM) supplemented with chickpea seed extract [1] { Present address: Ayub Agricultural Research Institute, Jhang Road, Faisalabad, Pakistan. * To whom all correspondence should be addressed. E-mail: [email protected]
0885-5765/00/060235+10 $35.00/00
and also solanapyrone B when grown on CDLM supplemented with metal cations, of which Zn, Ca, Cu, Co and Mn, were the most important, rather than chickpea seed extract [3]. The toxins were separated and quanti®ed by HPLC on an analytical ODS column with an isocratic mobile phase determined by solvent optimization [4]. In order to determine the relative toxicities of solanapyrones A, B and C and their individual eects on chickpea genotypes, it was necessary to separate them in quantity and this was achieved using solvent partitioning and ¯ash chromatography on silica [8]. Although A. rabiei produces solanapyrones A, B and C in culture, only solanapyrone C has been claimed to have been found in infected plants [18], other workers failing to ®nd any of the three compounds [10]. There are several precedents for metabolism of phytotoxins by plants. For example, the tree pathogen, Heterobasidion annosum, produces a series of compounds with phytotoxic activity in culture, in particular fomajorin, fomajorins S and D, fomannosin and fomannoxin but up to now only c 2000 Academic Press *
236
K. Hamid and R. N. Strange
fomannoxin has been detected in vivo and isolated from naturally infected wood [9]. Cell cultures of Pinus sylvestris metabolize fommanoxin to fomannoxin alcohol and subsequently to fomannoxin acid b-glucoside both of which are less toxic than fomannoxin [22]. Since there are precedents for the metabolism of toxins by host tissues which may explain why some of these compounds cannot be extracted from plants infected by toxigenic organisms, experiments were conducted in order to determine how chickpea might metabolize solanapyrone A, the most toxic of the solanapyrone toxins. The reduced glutathione (GSH)/glutathione-S-transferase enzyme (GST) system is one mechanism by which plants detoxify xenobiotics [5, 13]. Accordingly, the reaction of solanapyrone A with glutathione and the concentrations of GSH (as well as oxidized GSH) and activity of GST in a range of chickpea cultivars were determined in order to ascertain if there were any correlation between them and the sensitivity of the cultivars to the toxin. Some safeners, such as dichlormid, prevent herbicide damage in cereals and some dicotyledonous species by enhancing the levels of GSH and GST. It was therefore of interest to determine whether chickpea was similarly aected and whether plants treated with safener were less sensitive to solanapyrone A.
MATERIALS AND METHODS
Production and isolation of the solanapyrone toxins Isolate PUT 7 of A. rabiei [14] was cultured without agitation for 12 days at 208C on a medium consisting of Czapek±Dox nutrients supplemented with the metal cations ZnSO4.7H2O, 0.05 g l ÿ1; CaCl2.2H2O, 0.1 g l ÿ1; CuCl2.2H2O, 0.02 g l ÿ1; CoCl2.6H2O, 0.02 g l ÿ1 and MnCl2.4H2O, 0.02 g l ÿ1 (250 ml ¯asks; 30 ml per ¯ask) and the toxins were separated as reported previously [8]. In brief, cultures were ®ltered through cheesecloth to remove mycelium and centrifuged to remove spores. The supernatant was acidi®ed to pH 3.0 and partitioned against ethyl acetate. Fractionation of the ethyl acetate phase was by ¯ash chromatography on a silica gel column (40 g: Biotage UK Ltd., Hertford, U.K.) with dichloromethane, cyclohexane, ethyl acetate (3 : 3 : 1 v/v/v), dichloromethane, cyclohexane, ethyl acetate (1 : 1 : 1 v/v/ v) and ®nally ethyl acetate as mobile phases. Fractions containing pure solanapyrones were recognized by their characteristic u.v. spectra [12].
Growth of plants Seeds of ®ve kabuli chickpea cultivars, ILC 3279, ILC 482, ILC 249, INRAT 88 and Kasseb as well as seeds of seven desi cultivars, AUG 424, CM 88, CM 68, 6153, C
44, C 235 and CM 72 were soaked for 12 h in water and planted in John Innes No. 2 compost in plastic pots (13 cm diam.: eight seeds per pot). Plants were raised in a greenhouse at 25 + 28C.
HPLC of solanapyrones Toxin samples were separated on a Philips HPLC equipped with a diode array detector essentially according to the method of Chen et al. [4] except that the solvent system consisted of methanol 23.1 %, water 56.3 % and tetrahydrofuran 20.6 % (v/v/v) which was pumped at a ¯ow rate of 1 ml min ÿ1. The stationary phase was an ODS column (Spherisorb ODS 2; 150 4.6mm i.d.; Jones Chromatography, Glamorgan, U.K.) and was protected by a guard column (20 4.6 mm i.d.). The solanapyrones were recognized by their retention times and UV spectra which were compared by superimposition on those of authentic samples. For quanti®cation of the solanapyrones, chromatograms were abstracted from the three dimensional chromascans (absorption wavelength time) at lmax 327 nm, lmax 303 nm and lmax 320 nm for solanapyrones A, B and C, respectively and the areas under the peaks were compared with those of standard solutions of the authentic compounds.
Toxicity of solanapyrones A and B The toxicity of solanapyrones A and B was determined qualitatively by allowing triplicate samples of shoot cuttings to take up the compounds. Ethanolic solutions of solanapyrone A and solanapyrone B were placed in separate glass vials [58 17 mm (diam.)] and, after evaporation of the ethanol, distilled water (4 ml) was added to each vial and vortexed for 3 min to give 450 mM solutions of the toxins. Weighed shoots of chickpea (ILC 3279: 3 weeks old) were placed in each vial and incubated in a greenhouse at 25 + 28C for 24 or 72 h. Transpiration was aided by an electric fan. Shoots incubated in water served as controls. All vials were topped up with water as required to keep the cut ends below the surface. In some experiments, after incubation for 24 h, the shoots were weighed again and, in order to recover any solanapyrones remaining in the vials, their contents, together with three washings of methanol (0.5 ml each) were transferred to universal bottles where they were diluted with distilled water to 25 ml. Solanapyrones in these solutions were adsorbed onto end-capped Isolute cartridges (1 g: C18; International Sorbent Technology, Hengoed, Glamorgan, U.K.) and, after washing with water (2 ml), the cartridges were eluted in acetonitrile (2 ml). Solanapyrones in samples (20 ml) of the acetonitrile eluates were quanti®ed by HPLC.
Phototoxicity and metabolism of solanapyrones in the chickpea The toxicity of solanapyrones A and B was also determined quantitatively with suspensions of chickpea cells (density 0.2 A units at l 620 nm 2.25 105 cells ml ÿ1) isolated from lea¯ets of newly-opened leaves of chickpea plants (33±55 days old) and expressed as units essentially as previously described [14]. In brief, after dispensing a two-fold dilution series of toxin preparation in buered osmoticum (holding buer) among wells of a microtest plate (50 ml well ÿ1), a suspension of cells, freshly isolated from chickpea lea¯ets by enzymic digestion, was added to each well (50 ml well ÿ1) and the microtest plate incubated for 3 h at 258C. Fluorescein diacetate solution was added to each well (50 ml well ÿ1) and ¯uorescent (live) and non-¯uorescent (dead) cells scored under an inverted ¯uorescence microscope. Probit percent cell death, corrected for viability in control wells without toxin, was plotted against the log2 toxin concentration to give a straight line from which the LD50 value, arbitrarily de®ned as 1 unit of activity, was extracted. Since sensitivity varied among assays done on dierent days, results of tests on cultivars of chickpea were expressed relative to that of solanapyrone A acting on cells of a standard cultivar, ILC 3279, which was included in all assays. In some assays, test cells were isolated from lea¯ets of shoots treated with dichlormid as described below and tested for their sensitivity to solanapyrone A, cells isolated from lea¯ets of shoots incubated in water serving as controls.
Metabolism of solanapyrone A by chickpea shoots Shoots allowed to take up the compounds as described above were cut into small pieces with scissors and homogenized in a Sorvall Omni-mixer with 80 % ethanol (3 10 ml) for 5 min. The homogenate (30 ml) was centrifuged at 3000 g for 5 min and the supernatant evaporated to dryness on a rotary evaporator at 308C. Residues were dissolved in methanol (1.5 ml) and transferred to universal bottles. After dilution with distilled water to 25 ml, hydrophobic compounds were puri®ed by solid phase extraction as above and samples (20 ml) of the acetonitrile eluates examined by HPLC. The remaining acetonitrile eluates were concentrated on a rotary evaporator to 500 ml and samples were spotted on silica TLC plates (Silica gel 60 F254 , Merck, Germany). The plates were developed in cyclohexane/ dichloromethane/ethyl acetate (1 : 1 : 1 v/v/v) and, after evaporation of the solvent, were observed under short wavelength u.v. light. Spots at Rf 0.81 were scraped from the TLC plates, placed in Eppendorf tubes (1.5 ml) and vortexed in acetonitrile (3 1 ml). Supernatants were combined and evaporated to dryness on a rotary evaporator and the residues dissolved in the same solvent (50 ml). Samples (20 ml) were spotted on TLC plates (Silica gel 60 F254) and the plates were developed in
237
cyclohexane/dichloromethane/ethyl acetate (1 : 1 : 1 v/v/ v). Plates were viewed under short wavelength u.v. light and were also sprayed with 2,4-dinitrophenylhydrazine (50 mg in 10 ml ethanol). Samples (20 ml) were also subjected to HPLC (see above for details).
Reaction of glutathione with solanapyrone A Tris-HCl buer (10 mM, pH 8) containing 50 mM glutathione was dispensed in 24 glass vials [58 17 mm (diam.); 2 ml per vial]. Solanapyrone A (272 mg per vial in 9.3 ml ethanol) was added to each vial and vortexed for 30 s before sterilizing by ®ltration through 0.22 mm ®lters (MSi, Microseparations Inc, U.S.A.). In controls, solanapyrone A was incubated in Tris-HCl buer without glutathione. Both sets of vials were incubated in the dark at 258C for 0, 1, 2, 3, 4, 5 and 6 h. Samples were injected directly onto the HPLC. In order to determine the reaction product of glutathione and solanapyrone A, the toxin was incubated in Tris-HCl buer ( pH 8) containing 500 mM glutathione for 16 h at 258C. After incubation and freeze-drying, the product was analysed by fast atom bombardment mass spectrometry.
The eect of glutathione on the toxicity of solanapyrone A The eect of glutathione on the toxicity of solanapyrone A was measured in the cell assay either by determining the eect of a constant standard solution of glutathione (50 mM) on the dose-response curve of solanapyrone A or, conversely, by determining the eect of a dilution series of glutathione on the toxicity of a standard solution of solanapyrone A. In the ®rst assay, an ethanolic solution of solanapyrone A was added to well 1 of a microtest plate so that, on evaporation of the ethanol, the well contained 15.1 mg of the compound. Holding buer (100 ml) containing 50 mM glutathione was added to the ®rst well (giving a 500 mM solution of solanapyrone A) and a two-fold dilution series in the glutathione holding buer was made across the plate. Controls contained no solanapyrone A. Another plate with the same dilution series of solanapyrone A but without glutathione was also set up for comparison. Plates were incubated for 16 h at 258C before adding cells isolated from chickpea lea¯ets (cv. ILC-3279: 3 weeks old). Viability of the cells was scored after incubation for 3 h as usual. In the second assay, a two-fold dilution series of glutathione in holding buer with 50 mM as the highest concentration was made in a microtest plate (50 ml well ÿ1). Solanapyrone A dissolved in ethanol (3 ml) was added to each well to give a ®nal concentration of 250 mg ml ÿ1 (828 mM). Control wells contained only holding buer or glutathione (50 mM). Plates were incubated for 16 h at 258C before adding cell suspension from chickpea lea¯ets (cv. ILC
238
K. Hamid and R. N. Strange
3279: 3 weeks old: 50 ml well ÿ1), giving a ®nal concentration of solanapyrone A of 414 mM. Viability of cells was scored after incubation for 3 h.
Quantitative analysis of glutathione in chickpea lea¯ets Reduced glutathione (GSH) and oxidized glutathione (GSSG) were measured essentially according to the methods of Anderson [2] and Coleman et al. [6]. Lea¯ets (0.5 g) of chickpea cultivars (21±25 days old) were pulverized in liquid nitrogen with a pestle and mortar. The resulting powder was suspended in extraction buer (2 ml) consisting of 5 % (w/v) sulphosalicylic acid (SSA: Sigma) and 6.3 mM diethyltriaminepentaacetic acid (DETAPAC: Sigma), vortexed for 45 s, kept on ice for 10 min, centrifuged for 12 min at 10 000 g and further centrifuged at 10 000 g for 3 min at 48C to remove ¯oating particles. Supernatant (50 ml) was added to 750 ml of 0.143 M potassium phosphate buer containing 6.3 mM DETAPAC ( pH 7.5), 100 ml of 6 mM 5,5-dithiobis (2-nitrobenzoic acid) (DTNB) and 100 ml of 2.1 mM bnicotinamide adenine dinucleotide phosphate in the reduced form (b-NADPH). The reaction was initiated by the addition of 25 ml of glutathione reductase (20 U ml ÿ1: Sigma) and the change in absorption was measured at 412 nm for 6 min on a spectrophotometer. A standard curve with a concentration range of 0±45 mM GSH was plotted and used to determine the concentration of total glutathione (GSH GSSG) which was expressed as nmoles g ÿ1 fresh wt of lea¯ets [6]. For the measurement of oxidized glutathione (GSSG), supernatant (400 ml) was mixed with 8 ml of 2-vinylpyridine (2VP) and 40 ml of triethanolamine (TEA: Sigma). The mixture was vortexed for 15 s and incubated at 258C for 1 h before assaying 50 ml aliquots as for total glutathione. GSSG was determined as described for total glutathione (as before) from a standard curve with a concentration range of 0±40 mM GSSG. The amount of GSSG was calculated and expressed as nmoles g ÿ1 fresh wt of lea¯ets. Reduced glutathione (GSH) was calculated by subtracting the amount of GSSG from total glutathione (GSSG GSH) and also expressed as nmoles g ÿ1 fresh wt of lea¯ets.
Extraction and measurement of glutathione-S-transferase (GST) activity in chickpea lea¯ets GST was extracted by a modi®cation of the method of Hunaiti and Ali [11]. Lea¯ets (0.5 g) from each cultivar were ground in liquid nitrogen in a pestle and mortar until a ®ne powder was obtained. All further steps were carried out at 0±48C. The resulting powder was suspended in 0.1 M potassium phosphate buer ( pH 7.0: 0.5 ml), containing polyvinylpyrrolidone (PVP-40, 5 %: Sigma)
vortexed for 1 min and centrifuged at 15 000 g for 15 min. After diluting the supernatant to 1 g of tissue per ml with extraction buer, it was passed through a 0.45 mm ®lter (Gelman Sciences, U.S.A.) to remove ¯oating particles and the ®ltrate was tested for GST activity. GST activity was measured by the formation of the conjugate of GSH and 1-chloro-2,4-dinitrobenzene (CDNB) determined spectrophotometrically at 340 nm [19, 7]. The assay mixtures (3 ml) consisted of 0.1 M phosphate buer ( pH 6.5, 2865 ml), CDNB (30 ml in ethanol, ®nal concentration 0.1 mM), GSH (75 ml, ®nal concentration 2.5 mM) and the reaction was started by adding the enzyme solution (30 ml). A complete assay mixture without enzyme served as a control. The reaction was monitored spectrophotometrically by the increase in absorbance at 340 nm for 6 min {e340 of the conjugate 10 mM ÿ1 cm ÿ1; [15]}. Units of enzyme activity were calculated by subtracting the units of activity obtained in controls without enzyme and one unit of activity was de®ned as the amount of enzyme that catalysed the formation of 1 mmol of S-2,4-dinitrophenylglutathione min ÿ1 at room temperature. Glutathione and GST activity were also measured in shoots treated with solanapyrone A. Solanapyrone A (60.4 mg in 2 ml ethanol) was vortexed with 2 ml water in polypropylene centrifuge tubes [115 mm 30 mm (diam.)] to give a 100 mM solution. Shoots (0.75 g) of chickpea (cv. ILC 3279: 3 weeks old) were placed in the tubes and were allowed to take up 1.5 ml of the solution ( 45.3 mg solanapyrone A) under greenhouse conditions (23 + 28C: 5 to 6 h incubation period). After uptake of the toxin, shoots were transferred to tubes containing distilled water (25 ml) and incubated for a further 96 h under the same conditions. The water level was maintained throughout the incubation period. Shoots incubated in distilled water without solanapyrone A served as controls. After the incubation period, the parts of shoots covered by solanapyrone A solution or water were discarded and the entire remaining material approx. 0.5 g was ground in liquid nitrogen and used for the estimation of glutathione and GST activity. In similar experiments, shoots were allowed to take up 1.5 ml of solutions of dichlormid (100 and 200 mg ml ÿ1 prepared from a stock solution of 10 mg ml ÿ1 in acetone 150 or 300 mg per shoot). Glutathione and GST activity were extracted from test shoots and controls (incubated in water) and quanti®ed as already described.
RESULTS
The eects of incubating solanapyrones A and B with chickpea shoots When shoots were incubated in solanapyrone A (544 mg 4 ml ÿ1 H2O: 450 mM) for 72 h, the stems became
Phototoxicity and metabolism of solanapyrones in the chickpea
239
F I G . 1. Shoots of chickpea (cultivar ILC 3279) after taking up 45.3 mg of solanapyrone A and incubating for a further 96 h in water. Note breakage of stem just below the uppermost leaf.
shrivelled, brown and corky and the lea¯ets developed ¯ame-shaped chlorotic zones. In a dierent experiment when shoots were allowed to take up 45.3 mg of solanapyrone A and incubated for a further 96 h in water, bleaching at the base of the stems occurred and stems broke just below the uppermost leaf (Fig. 1). Controls incubated in water remained green and turgid. Incubation of shoots in solanapyrone B (547 mg 4 ml ÿ1 H2O: 450 mM) for 72 h produced dierent symptoms: lea¯ets became wilted and chlorotic and some of them abscized, whereas those of controls remained green and turgid. Both solanapyrone A and solanapyrone B were degraded when incubated in water at 25 + 28C. Losses of solanapyrone A were 152.3 from 544 mg contained in 4 ml water (450 mM) and 93.5 mg of solanapyrone B from 547 mg also contained in 4 ml water (450 mM) over a period of 24 h. When chickpea shoots were incubated in aqueous solutions of solanapyrone A (450 mM) or solanapyrone B (450 mM) for 24 h in the greenhouse at 25 + 28C, they lost weight. The loss was highly signi®cant, 16.3 % (P 5 0.005, 100 % 573 mg) in the case of solanapyrone A but non-signi®cant, 6.25 % (P 4 0.05, 100 % 426.7 mg) in the case of solanapyrone B. Losses of the solanapyrones from the solutions were greater when they were incubated with shoots and amounted to an extra 175.6 mg for solanapyrone A and an extra 234.0 mg for solanapyrone B, both losses being highly signi®cant (P 5 0.01). No solanapyrone A was recovered from shoots and only 22.0 + 12.6 mg of solanapyrone B.
A new compound with lmax of 238 and 288 nm was found in extracts of shoots treated with solanapyrone A. It had a retention time of 779 s on HPLC, an Rf of 0.81 on TLC and reacted with 2,4-dinitrophenylhydrazine to give a brown spot. However, when the area of silica corresponding to that occupied by this spot (and not sprayed with 2,4-dinitrophenylhydrazine) was scraped from the TLC plate, eluted in methanol and chromatographed on the HPLC again, two peaks occurred, one with a retention time which was close to that of the new compound originally chromatographed and with a UV spectrum that matched it better than 95 %. The other compound had a longer retention time (1047 s) and a u.v. spectrum that was a 97 % match with demethylated solanapyrone A, a degradation product in which the methyl moiety of the methoxy group was replaced by hydrogen. Bioassay of the demethylated compound compared with solanapyrone A gave LD50 values of 514.0 + 55.0 mg ml ÿ1 and 31.3 + 2.1 mg ml ÿ1, respectively. Production of the demethylated compound also occurred on incubation with either 0.5 M NH4OH for 1 h at 508C or a mineral salts medium (consisting of NaHPO4 7.0 g, KH2PO4 3.0 g, NaCl 0.5 g, NH4Cl 1.0 g, CaCl2 0.011 g and MgSO4 0.2 g dissolved in 1 l water and adjusted to a pH of 7.25) at 378C for 24 h or longer.
The relative sensitivity of chickpea cultivars to solanapyrones A and B Although the cell assay was a convenient method for measuring the sensitivity of chickpea cultivars to the
K. Hamid and R. N. Strange Sensitivity (ILC 3279 to Solanapyrone A = 1)
240 5 4.5
Solanapyrone A Solanpyrone B
4 3.5 3 2.5 2 1.5 1 0.5 0
AUG 424
ILC 249
6153
C 44
ILC 482 CM 88
CM 68
C 235
ILC 3279
INRAT 88
CM 72
Kasseb
Cultivar
F I G . 2. Sensitivity of chickpea cultivars to solanapyrones A and B relative to the sensitivity of cv. ILC 3279 to solanapyrone A ( 1).
Formation of an adduct of solanapyrone A with glutathione One reason for the dierential sensitivity of cultivars to solanapyrone A could lie in their ability to detoxify the compound, possibly through the glutathione/glutathioneS-transferase system. When a concentration range of
9
6153
AUG 424
8
Disease rating
solanapyrones, the day to day variation in the assay was considerable. For example, the minimum concentration of solanapyrone A required to kill 50 % of the cells of cultivar ILC 3279 was 10.1 + 1.0 mg ml ÿ1 and the maximum 93.3 + 13.8 mg ml ÿ1. For this reason ILC 3279 was included in each assay as an internal standard and the sensitivity of other cultivars measured relative to it. The dierences of means of relative sensitivity to solanapyrone A and solanapyrone B among the chickpea cultivars were highly signi®cant for both toxins (P 0.001). Cultivars ILC 249, AUG 424 and 6153 were the most sensitive to solanapyrone A and cultivars Kasseb and CM 72 the least. The range of sensitivity was 0.65±3.3 on a scale in which ILC 3279 was rated as 1. Cultivars were less sensitive to solanapyrone B, the range being 0.14±0.46 on the same scale (i.e. sensitivity of ILC 3279 to solanapyrone A 1). Solanapyrone A was 2.62±12.64 times more toxic than solanapyrone B depending upon the cultivar (Fig. 2). Comparison of the relative sensitivities of cultivars to solanapyrone A with their disease ratings to Ascochyta blight showed that those which were most sensitive, such as 6153 and AUG 424, were also the most susceptible to the disease, scoring 9 on a scale of 1±9 [20], while less sensitive cultivars such as Kasseb, CM 72 and INRAT 88 scored 4.5, 6 and 4, respectively (Fig. 3). Spearman's correlation coecient values between the susceptibility of the cultivars to A. rabiei and their relative sensitivity to solanapyrone A ( 0.5166) and to solanapyrone B ( 0.5229) were positive but non-signi®cant.
7 6
C 235
CM 72
C 44
5 Kasseb
4 3 0.25 0.5
0.75
INRAT 88 ILC 3279
1
1.25
ILC 482
1.5 1.75
2
2.25
2.5 2.75
3
3.25 3.5 3.75
Relative sensitivity to solanapyrone A
F I G . 3. Relationship of the susceptibility of nine cultivars of chickpea to Ascochyta blight with their sensitivity to solanapyrone A (cv. ILC 3279 1).
glutathione was included in an assay with cells from ILC 3279 and 414 mM solanapyrone A (a concentration which killed all the cells), only the highest concentration of glutathione tested (50 mM) reduced cell death to less than 50 %. Conversely, when this concentration of glutathione was included in the assay the amount of solanapyrone A required to reach the LD50 value was double that without the addition (215.3 mM compared with 100.7 mM). When solanapyrone A (272 mg) was incubated in 10 mM Tris-HCl buer (2 ml, 10 mM, pH 8) for 6 h, recovery of the compound was 72.1 + 6.9 % but only 11.9 + 1.5 % when the buer contained 50 mM glutathione, most of the loss (69.5 + 1.7 %) occurring within the ®rst hour, suggesting that solanapyrone A spontaneously formed a conjugate with glutathione. Further evidence for this was obtained when the two compounds were incubated together overnight at 258C in 10 mM Tris-HCl buer ( pH 8) and the product subjected to fast atom bombardment mass spectrometry. A compound was
Phototoxicity and metabolism of solanapyrones in the chickpea
241
obtained with a molecular ion of 606 which, when accurately measured, gave a mass of 606.214300 corresponding to a formula of C28H36N3O10S (calculated 606.212142) and suggesting the conjugate shown in Fig. 4. This, however, is 1 Dalton greater than expected from the mass spectrum. One possible interpretation of this result is that the protons of the two terminal carboxyl groups on the glutathione moiety of the molecule are lost and the molecular ion is for M 1 of this species as usually obtained for fast atom bombardment mass spectrometry.
Levels of total, oxidized (GSSG) and reduced glutathione (GSH) in chickpea cultivars
Glutathione (n moles g fresh wt leaflets−1)
Measurement of GSH in chickpea cultivars showed that the dierences of their means were highly signi®cant (P 0.001). Cultivar AUG 424 had the least and Kasseb the most (200.7 + 27.2 and 561.0 + 112.9 nmoles g ÿ1 fresh wt of lea¯ets, respectively: Fig. 5). Comparison of the means of reduced glutathione levels of the cultivars at the 95 % con®dence level separated the cultivars into six overlapping groups in which cultivars within a group did not dier signi®cantly from each other (Table 1). Sensitivity to solanapyrone A was inversely related to GSH levels, cultivars that were least sensitive such as Kasseb and CM 72 having GSH concentrations that were 1.7± 2.8 times greater than those of the most sensitive cultivars such as 6153 and AUG 424. When all 12 cultivars were analysed, a Spearman's correlation coecient value (rs) of ÿ0.7323: P 5 0.01 was obtained. Cultivars had low levels of oxidized glutathione which ranged from 11.9 + 5.2 to 67.2 + 19.3 nmoles g ÿ1 fresh wt of lea¯ets (Fig. 4).
F I G . 4. The proposed structure of the solanapyrone A-glutathione adduct.
Glutathione S-transferase (GST) activity in chickpea cultivars The dierences of the means of GST activity among 12 cultivars of chickpea measured as units were highly signi®cant (P 0.001). Cultivar AUG 424 had the least and INRAT-88 the most, 40.1 + 3.8 and 66.8 + 3.9 units of activity g ÿ1 fresh wt of lea¯ets, respectively. At the 95 % con®dence level, cultivars were separated into ®ve overlapping groups in which those within a group did not dier signi®cantly from each other (Table 2). Sensitivity to solanapyrone A was inversely related to GST activity, the activity of cultivars such as Kasseb and INRAT-88 which were least sensitive, being 1.40±1.66 times greater
900
Total Glutathione GSH GSSG
800 700 600 500 400 300 200 100 0
AUG 424
6153
CM 88 ILC 249
C 235 ILC 482
INRAT 88
ILC 3279
CM 68
C 44
CM 72
Cultivars
F I G . 5. Total, reduced and oxidized glutathione of 12 chickpea cultivars.
Kasseb
242
K. Hamid and R. N. Strange
T A B L E 1. Reduced glutathione (n moles g ÿ1 fresh wt of lea¯ets) of 12 chickpea cultivars
T A B L E 2. GST activity (units) of chickpea cultivars g ÿ1 fresh wt of lea¯ets
Cultivars
Cultivars
Kasseb CM 72 C-44 CM-68 ILC 3279 INRAT 88 ILC 482 C-235 ILC 249 CM 88 6153 AUG 424
GSH content 561.0 + 112.9 A 512.7 + 83.0 AB 489.9 + 28.0 AB 482.1 + 103.2 AB 462.5 + 18.1 AB 440.3 + 88.7 BC 422.3 + 35.9 BCD 411.6 + 42.7 BCD 354.7 + 31.6 CDE 331.4 + 37.2 DE 298.9 + 7.8 EF 200.7 + 27.2 F
INRAT-88 Kasseb ILC 3279 CM 72 CM 88 C-235 ILC 249 CM-68 C-44 6153 ILC 482 AUG 424
Units of activity g ÿ1 fresh wt of tissue 66.8 + 3.9 57.9 + 5.1 54.6 + 5.1 54.5 + 3.9 52.3 + 6.6 51.2 + 6.9 50.1 + 3.8 49.0 + 5.8 49.0 + 3.3 41.2 + 5.1 41.2 + 1.9 40.1 + 3.8
A B BC BC BC BC BC CD CD DE DE E
Analysis of variance showed P 0.001. Means followed by the same letters are not signi®cantly dierent at the 95 % con®dence level (LSD 103.9).
Analysis of variance showed P 0.001. Means followed by the same letters are not signi®cantly dierent at the 95 % con®dence level (LSD 8.093).
than that of the most sensitive cultivars such as 6153 and AUG 424. When all 12 cultivars were analysed, a Spearman's correlation coecient value (rs) of ÿ0.8094: P 5 0.01 was obtained.
prevalent. Two trivial but persistent reasons for this are the lack of an agreed scale for scoring the disease and the lack of a pictorial key to facilitate this. A third and more fundamental reason is the ability of the pathogen to sporulate copiously and therefore cause explosive epidemics [17]. A fourth and also fundamental reason is the lack of knowledge of the attributes which make the pathogen so virulent. One possible virulence factor is the production of toxins which kill the host. Initially, two toxins, solanapyrones A and C were isolated from culture ®ltrates of A. rabiei when grown on Czapek±Dox nutrients supplemented with chickpea seed extract [1]. Later, using a de®ned medium and a mobile phase for HPLC, determined by solvent optimization, a third toxin, solanapyrone B was detected [4]. All three toxins were originally isolated from culture ®ltrates of Alternaria solani, the cause of early blight of potatoes [12]. Evidence that these toxins play important roles in either disease is lacking but, in the case of Ascochyta blight of chickpea, two observations indicate that they might do so. First, all reliably identi®ed isolates of the fungus produce at least one of the toxins in culture, including 20 isolates from Turkey whose identity was con®rmed by the sequences of the ITS1 and ITS2 regions of their rDNA (Dolar and Strange, unpublished results). Since the fungus reproduces sexually on debris remaining on the ground after harvest [21], loss of toxigenicity would be likely to occur were it not necessary for pathogenicity or virulence. Secondly, symptoms of the disease, epinasty, chlorosis, necrosis and breakage of stems, are mimicked by the toxins. In this paper, further evidence for the role of the solanapyrone toxins in Ascochyta blight of chickpea is reported. The mimicking of symptoms by pure preparations of solanapyrones A and B has been con®rmed. In particular, the breakage of stems caused by solanapyrone
The eect of solanapyrone A and dichlormid on glutathione concentration and glutathione S-transferase activity in shoots treated with the compounds When shoots of cv. ILC 3279 were allowed to take up solanapyrone A (45.3 mg per shoot), there was a signi®cant increase in glutathione content (P 5 0.05). The levels of total, oxidized and reduced glutathione increased from 672 + 55, 60 + 7 and 612 + 99 nmoles g ÿ1 fresh wt of shoots, respectively to 848 + 92, 90 + 15 and 758 + 82 nmoles g ÿ1 fresh wt. Treatment of shoots with solanapyrone A (45.3 mg per shoot) also caused a signi®cant 1.9-fold increase in GST activity from 57.9 + 10.0 units g ÿ1 fresh wt in controls to 112.3 + 18.6 units g ÿ1 fresh wt in tests. When shoots were allowed to take up dichlormid (300 mg per shoot), their reduced glutathione levels rose signi®cantly (P 5 0.001) from 304.3 + 41.7 to 436.5 + 68.6 nmoles g ÿ1 fresh wt of shoots. Some enhancement of GST activity was also noted but it was not signi®cant. Cells isolated from lea¯ets of shoots treated with 150 and 300 mg per shoot were 2.45 and 2.66 times less sensitive to solanapyrone A, respectively, than those isolated from controls incubated in water only.
DISCUSSION Although blight of chickpea has received considerable attention from plant pathologists and plant breeders, devastating attacks by the pathogen, A. rabiei, are still
Phototoxicity and metabolism of solanapyrones in the chickpea A is illustrated in Fig. 1. This is a characteristic symptom of the disease and may be explained by the loss of turgor of parenchyma cells surrounding the stele as a consequence of the attack by the toxin on their plasma membranes. Such stems would have only the inadequate support of their steles and not the additional strength imparted by the turgor of the surrounding cells. That the integrity of the plasma membrane of chickpea cells is attacked by the toxins is demonstrated by the bioassay. This involves treating lea¯ets with enzymes and agitation, processes which result predominantly in the production of separated cells but also a few protoplasts. Cells isolated from dierent cultivars of the plant diered in their sensitivity and, disappointingly, variation from day to day in the assay was considerable. Nevertheless, when a single cultivar was included as an internal standard, it was possible to place the 12 cultivars in these experiments in order. Comparison of this order with the disease scores available for nine of the cultivars showed that there was a positive correlation between sensitivity to solanapyrones A and B and severity of disease symptoms although this was not signi®cant. Whether this relationship will eventually prove to be signi®cant will depend on the establishment of a reliable and standard method for scoring disease and extension of the experiment to other cultivars diering in resistance to the pathogen. In this paper the reason for dierences in sensitivity of the cultivars to solanapyrone A was sought. The glutathione/glutathione-S-transferase system is one mechanism by which plants detoxify xenobiotics and some evidence consistent with the detoxi®cation of solanapyrone A by this mechanism was obtained. Firstly, inclusion of glutathione in the assay reduced toxin titres. Secondly, solanapyrone A formed an adduct with glutathione spontaneously. Thirdly, there was an inverse and signi®cant correlation of solanapyrone A sensitivity with both glutathione content and glutathione-S-transferase activity. Fourthly, boosting the glutathione concentration and glutathione-S-transferase activity of chickpea by treatment of plants with the safener, dichlormid, decreased the sensitivity of their cells to solanapyrone A in the bioassay. Demonstration of the adduct in planta may prove dicult, however, since glutathione conjugates of xenobiotics are often rapidly metabolized [16]. It is now important to determine unequivocally whether or not the solanapyrone toxins have a role to play in Ascochyta blight of chickpea. The most persuasive evidence would come from the creation of toxin-minus mutants whose virulence could be restored by the addition of toxin. Ideally, aected loci of such mutants would only occur in genes close to the end of the biosynthetic sequence leading to the toxins. The acquisition of such data would validate the use of the solanapyrones as an aid to the selection of chickpea genotypes resistant to Ascochyta blight.
243
It is unfortunate that the cell bioassay showed such great variation from day to day as this may impair its use as a screen for resistance, although use of one cultivar as an internal standard in all tests, as described in the experiments reported in this paper, may mitigate this. Since the glutathione/glutathione-S-transferase system may be the mechanism by which the plant detoxi®es solanapyrone A, the most toxic of the three solanapyrones, another possibility is to select plants in which this system is highly expressed either constitutively or in response to challenge by the toxins or the pathogen. Finally, the fact that this system may be boosted by safeners, such as dichlormid, would also seem to be worth further investigation.
REFERENCES 1. Alam SS, Bilton JN, Slawin AMZ, Williams DJ, Sheppard RN, Strange RN. 1989. Chickpea blight: production of the phytotoxins solanapyrones A and C by Ascochyta rabiei. Phytochemistry 28: 2627±2630. 2. Anderson ME. 1985. Tissue glutathione. In: Greenwald RA, ed. CRC Handbook of Methods Oxygen Radical Research. Boca Raton, Florida: CRC Press Inc., 317±323. 3. Chen Y, Strange RN. 1991. Synthesis of the phytotoxins by Ascochyta rabiei in response to metal cations and development of a de®ned medium for toxin production. Plant Pathology 40: 401±407. 4. Chen YM, Peh EK, Strange RN. 1991. Application of solvent optimisation to the separation of the phytotoxins, solanapyrones A, B and C from culture ®ltrates of Ascochyta rabiei. Bioseparation 2: 107±113. 5. Coleman JOD, Blake-Kalff MMA, Davies TGE. 1997. Detoxi®cation of xenobiotics by plants: chemical modi®cation and vacuolar compartmentation. Trends in Plant Science 2: 144±151. 6. Coleman JOD, Randal R, Blake-Kalf MMA. 1997. Detoxi®cation of xenobiotics in plant cells by glutathione conjugation and vacuolar compartmentalization: a ¯uorescent assay using monochlorobimane. Plant Cell and Environment 20: 449±460. 7. Habig WH, Pabst MJ, Jakoby WB. 1974. Glutathione Stransferases. The ®rst enzymatic step in mercapturic acid formation. Journal of Biological Chemistry 249: 7130±7139. 8. Hamid K, Strange RN. 1997. An easy method for isolating the solanapyrone toxins from culture ®ltrates of Ascochyta rabiei. International Chickpea and Pigeonpea Newsletter 4: 20±22. 9. Heslin MC, Stuart MR, Murchu PO, Donnelly DMX. 1983. Fomannoxin, a phytotoxic metabolite of Fomes annosus: in vitro production, host toxicity and isolation from naturally infected Sitka spruce heartwood. European Journal Forest Pathology 13: 11±23. 10. HoÈhl B, Weidemann C, HoÈhl U, Barz W. 1991. Isolation of solanapyrones A, B and C from culture ®ltrates and spore germination ¯uids of Ascochyta rabiei and aspects of phytotoxin action. Journal of Phytopathology 132: 193±206. 11. Hunaiti AA, Ali BR. 1990. Glutathione S-transferase from oxidiazon treated chickpea. Phytochemistry 29: 2431±2435.
244
K. Hamid and R. N. Strange
12. Ichihara A, Tazaki H, Sakamura S. 1983. Solanapyrones A, B and C. Phytotoxic metabolites from the fungus Alternaria solani. Tetrahedron Letters 24: 5373±5376. 13. Lamoureux GL, Richard H, Shimabukuro D, Frear DS. 1991. Glutathione and glucoside conjugation in herbicide selectivity. In: Caseley JC, Cussans GW, Atkin RK, eds. Herbicide Resistance in Weed and Crop. Oxford: ButterworthHeinemann Ltd. 14. Latif Z, Strange RN, Bilton J, Riazuddin S. 1993. Production of the phytotoxins, solanapyrones A and C and cytochalasin D among nine isolates of Ascochyta rabiei. Plant Pathology 42: 172±180. 15. Mannervik B, Guthenberg C. 1981. Glutathione transferase (Human placenta). In: Jakoby WB, ed. Methods in Enzymology (Detoxi®cation and Drug Metabolism: Conjugation and Related Systems), Vol. 77. Academic Press, Inc., 231±235. 16. Marrs KA. 1996. The functions and regulation of glutathione S-transferases in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47: 127±158.
17. Nene YL. 1982. A review of Ascochyta blight of chickpea. Tropical Pest Management 28: 61±70. 18. Shahid AA, Riazuddin S. 1998. Presence of solanapyrone C in blight infected chickpea plants. International Chickpea and Pigeonpea Newsletter 5: 15±16. 19. Simons PC, Jagt DLV. 1977. Puri®cation of glutathione S-transferase from human liver by glutathione-anity chromatography. Analytical Biochemistry 82: 334±341. 20. Singh KB, Hawtin GC, Nene YL, Reddy MV. 1981. Resistance in chickpeas to Ascochyta rabiei. Plant Disease 65: 586±587. 21. Trapero-Casas A, Navas-Cortes A, Jimenez-Diaz RM. 1996. Airborne ascospores of Didymella rabiei as a major primary inoculum for Ascochyta blight epidemics in chickpea crops in southern Spain. European Journal of Plant Pathology 102: 237±245. 22. Zweimuller M, Antus S, Kovacs T, Sonnenbichler J. 1997. Biotransformation of the fungal toxin fomannoxin by conifer cell cultures. Biological Chemistry 378: 915±921.