A phytotoxic protein-lipopolysaccharide complex produced by Verticillium dahliae

0031-9422(93)E0084-R

Pergamon

@

Phytochemistty,yol.35, No. 6, pp. 1449 1453, 1994 Copyright @ 1994 Elsevier Sciene Ltd Printed itr Great Britain. All rights reserved m3t -9422 194 $6.00 + 0.00

A PHYTOTOXIC PROTEIN-LIPOPOLYSACCHARIDE COMPLEX PRODUCED BY VERTICILLIUM DAHLIAE RrAAN

MEyER,*t VnnNoN SLATER*f and IAN A. Dunrnvt{

*Plant Biotechnology Section, Yegetable and Ornamental Plant Institute, Agriculture Research Council, Private Bag X293, Pretoria 0001, South Africa; tDepartment of Chemistry and Biochemistry, R.A.U.-University, P.O. Box 524 Aucklandpark 2006, South Africa (Receiued in reuised.form 5 October 1993)

Key Word Index-Gossypiumhirsutum; Malvaceae; pathogenesis; protein-lipopolysaccharide; phytotoxin; wilt disease; Verticillium dahliae.

Abstract-A phytotoxic protein-lipopolysaccharide complex (PLPC) was purified to electrophoretic homogenetiy from seven-day-old culture filtrates of Verticillium dahliae by means of acetone precipitation, gel exclusion chromatography, and preparative agarose electrophoresis with a yield of 4.5mgl-1 culture filtrate. The PLPC consists of 15.70/" protein, 13.0% lipid, 0.4% phosphate and 70oh carbohydrate. The isoelectric point is 3.8 and the

amino acid cornposition of the protein fraction was determined. The complex has a M, of 197000 and can be dissociated into five protein-containing components, with M,s of 78000,62000,48000,32000 and 28000. Polygalacturonase and cellulase errzyme activities were identified in the PLPC by means of enzyme staining techniques, and were found to be associated with the 28 000 and 48 000 M, complexes, respectively. 1,3-B-Glucanase activity was associated with the 32 000 and 28 000 M. complexes. Treatment of cotton seedlings with concentrations + from 2.5 pg ml - 1 PLPC resulted in symptoms of wilting and necrosis and was accompanied by inhibition of the H ATPase activity of plasma membranes and the rapid transient elicitation of phenylalanine ammonia-lyase activity. Significant differences were found between resistant and susceptible cultivars. When applied to disks of mature cotton PLPC elicited the accumulation of pathogenesis-related proteins in the intercellular spaces.

leaves, the

virulent V. dahliae isolate from cotton, is reported. The

INTRODUCTION

Verticillium dahliae is a destructive soil-borne fungus that penetrates the host through the roots and spreads systemically through the xylem, leading to the appearance of wilt disease symptoms. Verticillium spp. produce both low and high M, phytotoxic metabolites in culture [1,2].

It

characterization of enzyme activities associated with the complex is described, as well as some of the responses of the plant tissue towards the toxin. RESULTS AND DISCUSSION

was previously shown that pathogenic isolates of

V. dahliae and V. albo-atrum produce extracellular high

Purification

M" protein-lipopolysaccharide complexes (PLPC).

Verticillium dahliae was grown on a modified Czapek's medium [3] which was reported to result in a high yield of complex. Maximal levels were reached after six to nine days. Longer periods of incubation resulted in degradation of the complex [3]. The phytotoxic activity of the various purified fractions was determined as described in the Experimental. The purification method used was based on that of Nachmias et al. [8] and Harlinget al.l9l with some alterations. A11 of the phytotoxic activity, recovered from the redissolved acetone precipitate from 200 ml of culture filtrate, was chromatographed on a Sepharose CL-6B column (Fig. 1). Peak 1 eluting between 35-60m| was characterized as a 197000 M,PLPC but was found to be electrophoretically heterogeneous. These data are in good agreement with Nachmias et al. l8f where it was noted that anaiysis of the PLPC peaks by PAGE showed one major band and at least five to seven minor bands. The PLPC was purified to electrophoretic

These

complexes are phytotoxic and induce most of the symp-

toms associated with Verticillium wilt disease [3-7]. Keen et al. l4f reported that the sensitivity of cotton

cultivars to PLPC from V. albo-atrum corresponded with the susceptibility of these cultivars to the fungus. From these results they concluded that the PLPC may be implicated in the development of disease symptoms. An antigen reacting to antiserum prepared against PLPC has been detected h V. dahliae-infected potato plants but not in healthy plants, indicating that the PLPC is probably produced by the pathogen in uioo, and may be part of the pathogenesis process [8]. Both toxic and vein blocking effects have been ascribed to the PLPC complex [3]. In this paper the purification to electrophoretic homogeneity of a 197000 M,PLPC from culture filtrates of a fAuthor to whom correspondence should be

addressed.

R. Mrysn e, al.

1450

Peok

2

0.8

oo 0.7

^

0.6 0.5

C 0.4 '0)

o L

o_

0.3 o.2 0.1

0

Peok

1

+t\

VI

Vo

0 102030405060708090 lution volume ( mi

,l 100 1 10 120 130 )

Fig. 1. Gel exclusion chromatography of extracellular products fuom V. dahliae. Acetone precipitates were dissolved in distilled water and loaded on to a Sepharose CL-6B (2.5 x 60) cm column. The column was eluted (20 ml hr- 1) with water at room temperature and monitored for protein by ,4 at 280 nm.

homogeneity (Fig. 2A) by means of preparative agarose gel electrophoresis. The flnal yield was 4.5 mg PLPC purified from 1 I of culture filtrate (Table 1). The PLPC was found to be stable for up to four weeks at 4'. P

hy sic o - c he mi

c

al

c

har ac

te

ri z at i on

The PLPC was composed of 15.70/, protein, l3.9oh lipid, 70% polysaccharide and 0.4oh phosphate. These data are in good agreement with the results obtained by Keen and Long for a PLPC isolated from culture fluid of V. dahliae l3f.

The M, of the purified compiex was determined by analytical gel permeation chromatography [10] as described in the Experimental. The PLPC eluted with a V"-

value

of 0.41, corresponding to an estimated M, of

197 000. This value was conflrmed with PAGE-Ferguson plots (190000 M,) and analytical PAGE gradient gels (195000M"). M, values for PLPC complexes reported in the literature vary from 2000000 [3] to 126000 M, L97, probably due to the reported tendency of the PLPC to

Fig. 2. Native PAGE and SDS-PAGE of the purified PLPC ftom V. dahliae ct:Jlure filtrates. (A) 5% PAGE: 7.5 and l0 pg protein per well. (B) 10% SDS-PAGE of 50 pg PLPC (right). M, markers (left) are myosin (200000), phosphorylase B (97400), bovine serum albumin (69000), ovalbumin (46000), carbonic anhydrase (30000) and trypsin inhibitor (21 500).

aggregate or break down to lower M, fragments depending on the ionic strength [3].

Analysis of the purified complex by IEF in pH 3-10 gradients also indicated only one band and the PLPC was judged to be electrophoretically homogeneous following Coomassie staining. The isoelectric point of the

purified complex was determined as 3.8 by PAGIEF, using a pH 2.5-5.0 gradient. Nachmias er a/. [8] also reported a pI value of 3.8 but with several contaminating protein bands present. The acidic pI correlates with the

Table 1. Purification table of the PLPC from V. dahliae culture fluid

Protein

Total

Volume concentration protein

Step

Crude filtrate Acetone precipitate Sepharose CL-6-B (Peak 1) Agarose gel electrophoresis

(ml)

200 10 35 5

(pgml-')

11 180 36.5 180

(mg)

Yield

(mgl-')

2.2 1.8

t.27

0.90

4.5

1451

Phytotoxin from Verticillium dahliae

phospholipid content of the toxin as well as with the amino acid composition of the protein fraction, where a relatively small number of basic amino acids (6oh) werc found, compared with the content of hydroxy (33%), hydrophobic (34%) and acidic (l4o/,) amino acids (data

not shown). Harling et al.l9l, Russel [11], Mussel ll2l and Carder 3] reported cellulase and/or polygalacturonase activities in culture filtrates of V. dahliae. Following native PAGE, efizyme activity staining revealed polygalacturonase, cellulase and 1,3-p-glucanase activities associated with the PLPC. The purified PLPC was subjected to dissociating, but non-reducing SDS-PAGE, and flve complexes were identified with estimated M,s of 78 000, 62000, 48000, 32000 and 28000 (Fig. 2B). Most pectic [1

Ul, 12] and reducing conditions could lead to degradation thereof with consequent deactivation of enzyme activities. Activity enzymes occur as enzyme complexes

staining was performed after removal of the SDS to determine which of the five components was responsible for these enzyrne activities. Two 1,3-B-glucanase activities were observed with M,s of 28000 and 32000. Cellulase activity was associated with the 48 000 M, component while the 28 000 M, component exhibited polygalacturonase activity.

Plant pathogens proJuce

at

affay of extracellular

enzymes capable of attacking plant cell wall components. A convincing role in pathogenesis, however, has been

established for those enzymes that attack the pectic fraction of the plant cell wall [14]. The pectic fragments released by the enzymes may also have effects on the host:

pathogen interaction, including the elicitation of defence reactions in the host. The occurrence of these enzyme activities in the PLPC may thus explain some of the symptoms (chlorosis, wilting and interveinal flaccidity) induced by the PLPC on cotton leaves. Work done by Mussel [12] showed that an endopolygalacturonase but not exopolygalacturonase purified from V. albo-atrum was toxic to susceptible cotton leaves when assayed in a divalent salt solution. Several reports stated that the enzyme activities were not required for toxic activity but this does not exclude the possibility that the enzyme activities might be needed for the induction of secondary symptoms in disease development.

d - I for OR19 and 10 1rg m1- I for Acala. Cultivars such as Letaba, Deltapine and Sicala exhibited an intermediate sensitivity towards the toxin. A PLPC-toxin from culture flltrates of V. dahliae has similarly been reported to possess some degree of hostspecificity and to induce a differential reaction between wilt-susceptible and wilt-tolerant potato genotypes [6, 8]. The interface for the plant-pathogen interaction necessitates the involvement of the plasmalemma and associated membrane enzymes in the initial response. Several fungal toxins have as their main target the active transport systems oftheir hosts [15] and the induction ofthe seedlings were 2.5 pg

hypersensitive response appears to be a membrane-associated process [16]. The PLPC inhibited the H*-ATPase activity in both cotton cultivars in the range of 0-50 nmol

(Fig. 3). In Acala l5l7-70 the H+-ATPase activity was 1.6-times more sensitive towards inhibition by the PLPC compared with that of OR19. A.ny alteration in the functional rate of the enzyme would be expected to strongly influence the metabolic behaviour of the plant cell. The inhibition of the ATPase activity might therefore lead to depolarization of the plasma membrane and ultimately cell death, due to inhibition of the electrogenic pump.

During periods of pathogen stress, products of activated defence genes such as phenylaianine ammonia-lyase are synthesizedde nouo, as has been well-proved in several host-pathogen interactions" This represents a transient induction of phenylalanine ammonia-lyase gene(s) transcription and is accompanied by an increase in enzyme activity [17]. Cotton hypocotyls exhibited a rapid transient induction of phenylalanine ammonia-lyase when exposed to solutions of the PlPC-phytotoxin produced by 1 V. dahliae. A concentration of 15 pg ml- was determined as optimal; at higher concentrations rapid wilting and necrosis resulted. Induction of PAL activity was observed in both cotton cultivars where OR19 rapidly reached a peak at 6 hr and then declined to basal levels (Fig. 4). In

Acala l5l7*70 there was a smaller initial response, followed by a major induction during the period

-C

o

I 7

E 6

l

Biological properties When applied to seedlings in concentrations between 2.5 and 25 pgml- t the PLPC resulted in interveinal flaccidity and chlorosis, followed by interveinal and then general desiccation and necrosis. These symptoms are all characteristic of those on cotton plants naturally infected with severe strains of V. dahliae. The sensitivity of the seedlings to PLPC correlated with the susceptibility of cotton cultivars to I/. dahliae,in that the resistant cultivar

OR19 exhibited a hypersensitive-like response within 4hr of 10pgml-1 PLPC application. The same symptoms were noticed in the susceptible cultivar Acaia 1517-70, but only from 10-12 hr onwards. Concentrations ofthe PLPC that resulted in toxic effects in cotton

5

=

4

o 3 o a) 2 a o 1 o_ F 0

o 5 101520253035404550 PLPC (nmoles)

Fig. 3. The effect of increasing PLPC concentrations on the Hr-ATPase activity of cotton plasma membrane fractions. Assays were performed with 20 pg membrane protein from OR and Acala (+) at pH 6.5 and at 33" as described in the

(I)

Experimental.

R. Mpvnn et a/.

t452

eluted with HrO at room temp. The column eluate was monitored for protein by A at 280 nm. Peak 1 contained the majority of the phytotoxic activity (determined as described in this section) and was characterized as a

E

c

o O)

N

197000 M, protein-lipopolysaccharide complex and found to be electrophoretically heterogeneous. Peak 1 (+35 ml) was concd against powdered sucrose to a concn of 1 mgml- 1. The PLPC was purifled to homogeneity by prep. agarose gel electrophoresis in a horizontal sub-

= .: o o

)

marine apparatus using 1.5% gels and

o-

810

12 14 16

182022.24263036

Iime(h) Fig. 4. Induction of phenylalanine ammonia-lyase in cotton

seedlings

by the PlPc-phytotoxin purified

from

cultivars, OR (* ) and Acala treated with 15 pg ml- 1 of the PLPC.

were

V. dahliae. Seedlings of the

(I)

10*24hr. Rapid expression ofdefence responses, following early molecular recognition of the pathogen, is generally associated with an incompatible interaction, while in a compatible interaction where the pathogen eludes the plants's surveillance mechanisms, defence responses are delayed and disease ensues [18]. Apart from the induction of phenylalanine ammonialyase (one of the earlier defence responses), the PLPC is also active in eliciting the accumulation of pathogenesisrelated proteins in the intercellular fluid (results not shown).

The biochemical function of the various components present in the PLPC requires further investigation, as well as the possibility ofearly recognition ofthe pathogen via a receptor for the PLPC on cell membranes. This

might contribute to more insight into the role of the PLPC in Verticillium wilt disease.

EXPERIMENTAL

Plant material. Two cultivars of cotton,

Gossypium

hirsutum L. OR19 (relatively resistant towards V. dahliae) and Acala l5l7-70 (susceptible towards V. dahliae) were used in the study. Culture methoils. A virulent isolate of V. ilahliae was obtained from the Tobacco and Cotton Research Institute, Rustenburg, South Africa, and was maintained on potato-dextrose lactic actd (4'/o vlv) agar. For the pro-

duction of the PLPC the fungus was cultivated in 1-l conical flasks containing 200 ml of a synthetic medium [8]. Flasks were incubated on a reciprocal shaker in darkness at23" for 7 days before the culture filtrates were

20

mM

NH4HCO3 as electrophoresis buffer. The complex was detected on longitudinal guide strips with amido black, the corresponding bands were cut from the gel, crushed and eluted by centrifugation for 10 min at 18000 9 at 4'

uel. Analytical techniques. Total protein was determined using the dye-binding assay ofref. [20] standardized with BSA. Polysaccharide was determined by means of the phenol-HrSOn acid reagent of ref. [21], using n-glucose

as standard. Lipid was determined gravimetrically following hydrolysis of the PLPC in 1 M HCI for 60 min at 100" and extraction into EtrO. The total P, content ofthe PLPC was determined using the method of ref. 122f. PAGE and SDS -PAGE were performed as described [23]. Protein bands were stained with Coomassie Blue R250 or with Schiff's base reagent for glycoproteins [24]. M, standards for SDS-PAGE were myosin (200000), phosphorylase B (97 400), bovine serum albumin (69 000), ovalbumin (46000), carbonic anhydrase (30000) and trypsin inhibitor (21 500). The M, of the PLPC was determined in 5a/o,loh and lloh acrylamide [25] as well as in a gradient gel 4-20Yo of acrylamide [26]. Reference proteins for the M, determination were ferritin (450 000), catalase (240000) and aldolase (158000). After electrophoresis, enzyme activity staining was performed for polygalacturonase, cellulase and 1,3-fglucanase activities. Cellulase and polygalacturonase activities were determined by the overlay technique of ref. L277. 1,3-P-Glucanase activity was determined by the method described by ref. [28]. Enzyme activity staining was also performed after SDS-PAGE. The SDS was removed by washing the gels after electrophoresis for at least 30 min with cold 0.1 M succinate bufler pH 5.8,

containing

10

mM dithiothreitol [29].

Isoelectric focusing was performed on agarose gels [30] with ampholytes in the pH range 3 10 and 2.5-5.0. The gel was run for 1.5 hr at 4' (30 min at 200 V and 60 min at 600 V). After isoelectric focusing the gel was stained with Coomassie-R250.

The M, of the purified PLPC was determined by analytical gel permeation chromatography [10], on a Sepharose CL-6B column eluted with 0.05

M KPi-buffer,

harvested.

pH 7 containing 0.1 M NaCl. The following

Purification of the PLPC. The culture fluid was filtered through glass wool and then centrifuged (100009, 2', 20 min) to remove the spores. The supernatant (200 ml) was treated with 4 vol. of MerCO at - 18" and allowed to ppt. overnight. The ppt. was collected by centrifugation as above and redissolved in HrO. The soln was then loaded on to a Sepharose CL-6B column (2.5 x 60) cm which was

were used for calibration of the column: ferritin (M,

standards

450000), catalase (M, 240W0), aldolase (M, 158000), BSA (M, 67 000), and ovalbumin (M, 45 000). V. and V, were determined with blue dextran (2 000 000) and DNPalanine (255), respectively. The amino acid composition was determined with the PICO-TAG method as described in ref. [31] using a Beckman HPLC.

Phytotoxin ft om Verticillium ilahliae

Bioassay

of toxin actiuity.

Two-week-old cotton

3. Keen,

N. T. and Long, M.

1453

(1972) Physiol. Plant

seedlings were used as a bioassay system to test the toxicity of the purifiedcompiex. The roots of the seedlings

4. Keen, N. T., Long, M. and Erwin, D. C. (1972)

symptoms to 3: severe chlorosis and necrosis with leaf collapse). At least throe replicates were used per assay and

Physiol. Plant Pathol. 2,317. 5. Malysheva, K. M. and Zel'tser, S. S. A. (1968) Proc. Acad. Sci. U.S.S.R. l'79,231. 6. Nachmias, A., Buchner, V. and Krikun, J. (1982)

were immersed in sterile distilled HrO containing 2.5-25 pgml- 1 PLPC. Symptoms were scored (0:no

scores were averaged. Induction of phenylalanine ammonia-lyase. In hypocotyls induction was monitored by the following procedure: single hypocotyls were extracted in triplicate with 200 pl

M Na borate (pH 9) containing 20 mM mercaptoethanol,0.2 mM 2-mercaptobenzothiazole and 1 mM phen-

0.2

yl-methyl sulphonylfluoride, at 2 hr intervals during a 20 hr period of exposure to a toxin concentration of 15,agml-1, by using a mortar and pestle. The resulting suspensions were centrifuged at 10 000 g for 5 min and the

supernatant assayed for phenylalanine ammonialyase activity. Phenylalanine ammonia-lyase assay. Activity was meas-

ured by a modified spectrophotometric assay [32]. The reaction mixt. (3 ml) contained 2.7 ml 0.1 M Na borate (pH 9), 200 p10.1 M Na borate (pH 9.0) containing 0.1 M

l-phenylalanine and 100 pl of enzyme. The assay was carried out for t hr at 30', and the reaction was stopped by addition of 100 pl 6 M HCl. The cinnamate formtd was extracted into 3 ml of toluene and estimated by measurement of the ,4 at 290 nm using toluene as blank. The effect of the PLPC onplasmamembrane associated H+ -ATPase actiuity. Sealed plasma membrane vesicles were isolated as described in ref. [33]. The assay for

plasma membrane ATPase activity contained in 1 ml: 30 mM Tris Mes, pH 6.5, 3 mM ATP, 3 mM MgSOn, 50 mM KCl, 0"1 mM NarMoOn, and 5 pM gramicidin D. Following incubation for 15-20min, the released Pi was determined by the method of ref. [34]. The assays were performed at 33' and the reaction was started by addition of the membranes. When the effect of PLPC was 1) analysed, the membranes (20 pg membrane protein mlwere preincubated for t hr with 5-50 nmol PLPC in the above reaction buffer. ATPase activity was expressed as

pmol Pi released hr-

1.

niluction of p atho g ene si s-r el at ed pr ot ein s. Cotton leaf disks, 12 mm in diameter were placed with the adaxial 1 side in contact with a 15 pgml soln of PLPC in 10 ml Control disks were treated HrO in Petri dishes. sterile with distilled HrO only. After a period of 48 hr under constant light, the intercellular fluid (IF) was extracted by vacr\um infiltration and centrifugation [35]. The protein content of the IF extracts was analysed by native PAGE I

as described.

Pathol. 2, 307.

Physiol. Plant Pathol. 20, 213.

7. Stoddart, J. L. and Carr, A. H. J. (1966) Ann. Appl. Biol. 58, 81.

8. Nachmias, A., Buchner, V. and Burtstein, Y. (1985) Immunochemisty 2,235. 9. Harling, R., Scheffer, R. J. and Elgersma, D. M. (1986) J. Phytopathology 177, 198. 10. Andrews, P. (1970) Meth. Biochem. Anal. 18, l. 11. Russel, S. (1970) Phytopath.2.84,222. 12. Mussel, H. W. (1972) Phytopathology 63,62. 13. Carder, J. H. (1989) Mycol. Res. 3, 297. 14. Collmer, A. and Keen, N. T. (1986) Ann. Reu. Phytopathol.24,383. 15. Frederico, R., Scalia, S., Ballio, A., Cocucci, M., Denti, A. and Marre, D. (1980) Plant Sci. Letters 17, 129.

16. Mathieu, Y., Kurkdjian, A., Xia, H., Guern, J., Koller, A., Spiro, M. D., O'Neill, P., Albersheim, P. and Darvill, A. (1991) Plant J. 1, 333. 17. Lawton, M. A., Dixon, R. A., Hahlbrock, K. and Lamb, C. J. (1983) Eur. J. Biachem.l29,593. 18. Lamb, C. J., Lawton, M. A., Dron, M. and Dixon, R. A. (1989) Cell 56,215. 19. Alarez, L. and Nieto, A. (1989) Analyt. Letters 22, 2737.

20. Bradford, M. (1976) Analyt. Biochem.12,248. 21. Hirs, C. H. W. (1967) Meth. Enzymol.ll,4ll. 22. Fiske, C. H. and Subbarow, y. (1925) J. Biol. Chem. 66,375. 23. Laemmli, U. K. (1970) Nature 227,680. 24.

Zacharits, R. M., Zell, T. E., Morrison, J. H. and

Woddlock, J. J. (1969) Analyt. Biochem.30, 148. 25. Ferguson, K. A. (1964) Metabolism 13, 985. 26. Slater, G. G. (1969) Analyt. Chern. 41,1039. 27. Bertheau, Y., Madgidi-Hervan, 8., Kotoujansky, A., Nguyen-The, C., Andro, T. and Coleno, A. (1984) Analyt. Biochem. 139, 383. 28. Pan. S., Ye, X. and Kuc, J. (1989j Analyt. Biochem.

t82,136. 29. Schwartz, W. H., Bronnenmeier, K., Grabuitz, F. and Straudenbauer, W. L. (1987) Analyt. Biochem.164,72. 30. Robertson, E. F., Danelly, H. K., Malloy, P. J. and Reeves, H. C. (1987) Analyt. Biochem. 167,290. 31. Bindiingmeyer, B. A. (1984) J. Chrorn.336,93. 32. Khan, N. U., Kamath, A. V. and Vaidyanathan, C. S. \1987) Biochem. Internat. 14, 451.

REFERENCES

1. Green, R. J. (1954) Phytopathology 44,433.

2. Krassilnikov, N. A., Khodjibayeva, C. M. and Mirchink, T. G. (1969). J. Gen. Appl. Mioobiol.15,l.

33. Giannini, J. L., R.uiz-Cristin, J. and Briskin, D. P. (1988) Analyt. Biochem. 17 4, 561. 34. Ohnishi, T., Gall, R. S. and Mayer, M. L. (1975) Analyt. Biochem. 69, 261. 35. Parent, J. G. and Asselin, A. (1984) Can. J. Bot.62, s64.