- Email: [email protected]
The effect of hydrogen peroxide on the viability of tomato cells and of the fungal pathogen Cladosporium fulvum
Physiological and Molecular Plant Pathology (1999) 54, 131–143 Article No. pmpp.1998.0195, available online at http:\\www.idealibrary.com on
The effect of hydrogen peroxide on the viability of tomato cells and of the fungal pathogen Cladosporium fulvum H L and V J. H* Department of Botany, Uniersity of Toronto, Toronto, Ontario, Canada, M5S 3B2 (Accepted for publication November 1998)
An oxidative burst was previously demonstrated to be induced in tomato plants by race specific elicitors of the fungal pathogen Cladosporium fulum. The in planta levels of H O estimated to occur # # during elicitor treatment, were compared with the levels required to show toxicity to host cells and to the fungal pathogen. Injection of Cf-9 tomato leaves with 100 m H O caused an insignificant # # degree of necrosis and 1 H O was required to cause complete leaf necrosis comparable to that # # induced by the AVR9 elicitor. Assays with Cf-5 tomato cell suspensions confirmed the low toxicity of H O to tomato cells but, as expected, the addition of Fe#+ with H O (or with intercellular # # # # fluids containing AVR5 elicitor) enhanced cell death as determined by the Evans Blue assay. Germination and germ tube growth of conidia of C. fulum were significantly retarded by 4–5 m H O , and at higher concentrations, death of germ tubes was observed (ED l 22 m), as # # &! determined by the fluorescein diacetate assay. The addition of Fe#+ with H O had little effect on # # fungal growth or viability in itro. These results suggest that the amount of H O accumulating # # during an elicitor-induced response in leaves may be sufficient to affect fungal colonization but not to affect viability of host cells unless the Fe#+ status in the apoplast is in some way altered by the elicitor to facilitate OHd production via the Fenton reaction. # 1999 Academic Press
INTRODUCTION
Rapid generation of active oxygen species (AOS), called the oxidative burst, is believed to be one of the earliest responses of plants to an attack by a non-pathogen or by an avirulent race of a pathogen [4 ]. AOS generated from the oxidative burst have been associated with several aspects of defense responses in plants, including direct toxicity to invading pathogens [42, 47 ], strengthening of plant cell walls via crosslinking of wall polymers [31 ], activating transcription of defense related genes [8, 35 ], triggering programmed cell death [16, 31, 34, 35 ], and mediating a reiterative signal network underlying systemic resistance responses [2 ]. These reactions of AOS, combined with other defense responses of the host plant, are assumed to lead to incompatibility between the host and pathogen. In the interaction between tomato (Lycopersicon esculentum Mill) and its fungal pathogen Cladosporium fulum Cooke (syn. Fulia fula), an oxidative burst has been * To whom correspondence should be addressed. Abbreviations used in text : AOS, active oxygen species ; FDA, fluorescein diacetate ; HR, hypersensitive response. 0885–5765\99\050131j13 $30.00\0
# 1999 Academic Press
132
H. Lu and V. J. Higgins
demonstrated to be induced by race specific elicitors, the products of Ar genes in the pathogen, in undetached leaves [37 ], in cotyledons [38 ], and in cell suspension cultures [44 ]. This induction of the oxidative burst was observed within 1–2 h in leaves or in cotyledons and within a few minutes in suspension cells. Data obtained from in io experiments using inhibitors (unpublished data), combined with those using isolated cell membranes [48 ], suggest that the AOS result from the activation of a plasma membrane NADPH oxidase in a manner similar to that of phagocytic neutrophils [3, 15 ]. In neutrophils, the function of the oxidative burst in killing the invading pathogens is well established [24 ], but in the tomato-C. fulum interaction, the role of the oxidative burst has not been fully investigated. In the tomato-C. fulum interaction, incompatibility is generally characterized by hypersensitive cell death which occurs as the fungus penetrates and reaches the substomatal space [32, 33 ]. Although the asynchronous nature of this stomatal penetration makes it difficult to correlate the restriction of fungal growth to cell death, most histopathology studies [10, 32, 33 ] suggest such a correlation. The hypersensitive response (HR) in this system was demonstrated to be triggered by race specific elicitors [11 ]. Although the molecular events and the recognition of some of these elicitors, e.g., AVR4 and AVR9, by host plants have been extensively studied [11, 25, 28, 30 ], the mechanism of the elicitor-induced leaf cell death is still unknown. On the other hand, several possibilities exist for the causes of fungal death. Rapid synthesis of pathogenesisrelated (PR) proteins, including chitinases and β-1,3-glucanases, was observed in incompatible interactions [14, 26 ]. However, these enzymes were unable to inhibit the growth of C. fulum in itro [27 ]. Accumulation of phytoalexins was more rapid in incompatible interactions than in compatible ones and these compounds were shown to have antifungal activity [12 ], but the extent of their role in the restriction of the growth of C. fulum continues to be debated [5, 23 ]. Likewise, a rapid localized deposition of callose in host cell walls at penetration sites was demonstrated in incompatible interactions [33 ]. It was suggested that the accumulation of callose might prevent diffusion of nutrients to the apoplast, resulting in the starvation and death of the fungus growing in the apoplast [33 ]. Again, conclusive evidence is lacking. It has been reported that plant cell death is triggered by H O generated from an # # oxidative burst in soybean cells [35 ] or produced by a fungal glucose oxidase in transgenic tobacco plants [29 ]. In addition, spore germination of several fungi has been shown to be inhibited by H O at micromolar levels [42 ], although, the inhibitory # # effect of H O on fungal germ tubes or hyphae has not been examined. In the present # # study, we investigated the possible direct role of the products of oxidative burst on the interaction between tomato and C. fulum. The effects of AOS, especially H O , on the # # viability of both host cells and pathogen were investigated. Here, we present data supporting the hypothesis that H O may contribute to the inhibition of fungal growth # # but it is unlikely to be responsible for host cell death unless the fungus interferes in some way with iron metabolism. MATERIALS AND METHODS
Plant materials Seeds of tomato cv. Bonny Best, which carry no resistance genes to tomato leaf mould, and cv. Moneymaker, near isogenic for resistance gene Cf-5 or Cf-9 were obtained as
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
133
described before [36 ]. Seeds were germinated in Plant Cell Paks containing Pro-Mix BX (Premier Peat Moss Ltd., Quebec, Canada) and seedlings were transplanted individually into 15 cm pots containing the same mix. Cell suspension cultures derived from a line of tomato with the resistance gene Cf-5 were grown in 500 ml Erlenmeyer flasks containing 100 ml of MS medium [40 ], in the dark at 25 mC on a rotatory shaker at 120 rpm and subcultured weekly [44 ]. Preparation of intercellular fluid containing the elicitor AVR5 Tomato plants cv. Bonny Best were inoculated with race 4 or 5 of C. fulum for production of intercellular fluids containing, or lacking, the elicitor AVR5, respectively, and intercellular fluids were prepared according to the method of De Wit and Spikman [13 ]. Intercellular fluids from the combination BB\r4 (e.g. isolated from leaves of Bonny Best inoculated with race 4 of C. fulum) are referred to as elicitor, whereas the combination BB\r5 is referred to as control. The effect of H O on leaf necrosis and on the viability of suspension cells # # Leaves of tomato plants were injected with various concentrations of H O and injected # # plants were incubated for 18 h at the same conditions as prior to treatment. The degree of necrosis was assessed by a rating system (see caption Fig. 1). Plant cells were treated with various concentrations of H O and\or supplemented # # with various concentrations of Fe#+ (FeSO :7H O) and incubated for 18 h. Cell death % # was determined using Evans Blue, a nonpermeating pigment of low phytotoxicity, as described by Turner and Novacky [43 ]. The selective staining of dead cells with Evans Blue depends upon exclusion of this pigment from living cells by the intact plasmalemma, whereas it passes through the damaged plasmalemma of dead cells and accumulates as a blue protoplasmic stain. Treated and untreated Cf-5 tomato cells were incubated for 15 min with 0n05 % Evans Blue and then washed extensively to remove excess and unbound dye. Dye bound to dead cells was solubilized in 50 % methanol with 1 % SDS for 30 min at 50 mC and quantified from measurements of absorbance at 600 nm. The effect of H O on the germination of conidia of C. fulvum # # Conidial suspensions (10& spores ml−") of C. fulum were prepared as previously described [5 ]. A known quantity of H O was added to 200 µl samples of the spore # # suspension and 50 µl of these treated spores were smeared on sterile microscope slides which were incubated at 25 mC for 24 h in Petri dish moisture chambers at which time the percentage of germinated spores was determined microscopically. Spores were considered germinated when the length of the germ tube exceeded the diameter of the spore. The effect of H O on growth and viability of germ tubes of C. fulvum # # The effect of H O on germ tube growth of C. fulum was tested on slides with untreated # # spores previously germinated for 24 h as described above. A known quantity of H O # # in 50 µl H O was added to the 50 µl of spore suspension on the slides and was well # mixed with the germinating spores by gently rotating the slides. The spores were incubated for a further 18 h and the lengths of the germ tubes were measured
134
H. Lu and V. J. Higgins
4
Necrosis rating
3
2
1
0
100
250
500
750
1000
Concentration (mM) of H2O2 F. 1. Leaf necrosis induced by exogenously applied H O . Leaves of tomato Cf-9 plants were # # injected with varying concentrations of H O and the injected plants were maintained in a growth # # room for 18 h under the conditions as described before. Necrosis rating was based on a scale of 0–4 [36, 37 ] in which 0 refers to no necrosis and 1, 2, 3 and 4 refer to about 25 %, 50 %, 75 % and 100 % necrotic area respectively, occurring on injected leaf areas. Values are meanspSD for three experiments with three replicates each.
microscopically. The average length of the germ tubes measured just before the H O # # treatment was used to determine the germ tube length at the time of treatment. Viability of germ tubes after treatment with H O for 18 h was determined by the # # fluorescein diacetate (FDA) assay as described by Widholm [45 ]. A stock solution of FDA (5 mg ml−") in acetone was diluted 1 : 100 with distilled water and then added to the assay slides. After 5 min, the number of fluorescent (viable) germ tubes was determined for 200 germ tubes per replicate under a Zeiss epifluorescence microscope, using a ParaLens Microscope Adapter\Fiber Optics System (Becton Dickinson Inc.) with a FITC filter cluster (520 nm barrier filter, 510 nm dichroic beam splitter, and a 470–490 nm excitation filter) and a fiber optic illuminator (Chiu Technical Corporation FO-150–115) with a 150 W quartz halogen lamp. Statistical analysis Comparisons of differences obtained in all experiments were made using the t-test. P 0n01 was considered significant.
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
135
RESULTS
The H O concentration needed to cause tomato leaf necrosis # # Leaves of Cf-9 tomato plants were injected with different concentrations of H O and # # the levels of necrosis induced by H O were rated for visible necrosis. With this # # approach, a concentration of 1 H O was required to cause complete leaf necrosis # # (i.e., a rating of about 4) comparable to that induced by race specific elicitor, e.g., AVR9 (Fig. 1). Little visible necrosis was observed when the leaves were injected with 100 m H O , hence lower concentrations (e.g., 20 m) mimicking those estimated to # # be produced by elicitors were not tested. When the leaves were injected with 1 H O , # # the injected areas became darkened and shrunken within 1 h, and the injected aqueous solution, unlike that with lower concentrations of H O or with elicitor preparations, # # was not transpired from the leaf tissues. The appearance and timing (about 6–9 h) of visible necrosis induced by H O at concentrations of 250, 500 and 750 m were similar # # to that induced by race specific elicitors.
5
1 mM Fe2+
Absorbance at 600 nm (cell death)
0.1 mM Fe2+ 0.01 mM Fe2+ 4
Without Fe2+
3
2
1
0
0
1
10
50
100
Concentration (mM) of H2O2 F. 2. Effects of H O and Fe#+ on the viability of tomato Cf-5 cells. Cell cultures were treated # # with different concentrations of H O and FeSO :7H O and incubated for 18 h on a rotatory # # % # shaker in the dark. Cell death was measured by uptake of Evans Blue stain by dead cells and quantified by absorbance of methanol\SDS extracts of the cell at 600 nm. The values represent the meanpSD of three independent experiments with three replicates each.
136
H. Lu and V. J. Higgins
Without Fe2+
Absorbance at 600 nm (cell death)
3
1 mM Fe2+
2
1
0
Untreated
Elicitor
Control
10 mM H2O2
F. 3. The effect of Fe#+ supplement on elicitor- or H O -induced cell death as measured by
# #
Evans Blue staining. Cf-5 tomato cells were treated with either intercellular fluid containing the AVR5 elicitor (BB\r4) or with intercellular fluid lacking the AVR5 elicitor (BB\r5) (control), supplemented with 1 m of FeSO :7H O after 30 min. Untreated cells and cells treated with % # 10 m H O also served as controls. Evans Blue staining was assessed at 18 h from measurements # # of absorbance of methanol\SDS extracts of the cells at 600 nm. The values represent meanspSD of three independent experiments with two replicates each.
The H O concentration required to cause cell death with or without Fe#j # # The toxicity of H O to living cells is known to be variable and the variability has been # # accounted for both by the activity of H O -scavenging enzymes, and by the rate of # # conversion of O − and H O into more highly reactive radicals, e.g., OHd [22 ]. The # # # conversion of H O into OHd, the Fenton reaction, is mediated by reduced transition # # metals such as Fe#+ which frequently results from the reaction of Fe$+ and O −. Plus, # the availability and release of Fe#+ from storage proteins, transport proteins, or ironcontaining enzymes is considered to be important in regulating the toxicity of H O to # # plant cells [6 ]. On such consideration, the effect of H O , supplemented with an Fe#+ # # salt, on viability of tomato cells was examined. Cf-5 cells were treated with different concentrations of H O and FeSO :7H O solution and incubated in the dark with # # % # shaking overnight (about 18 h). Cell viability was determined by the amount of Evans Blue taken up by dead cells as measured by absorbance after solubilization from cells [43 ]. Calibration of the Evans Blue assay with the percentage dead cells via the FDA viability test showed that the maximum absorbance (" 5n0) equaled about 100 % dead cells. Treatment with 100 m H O and 1 m Fe#+ caused 100 % cell death whereas # # 0n1 m or less Fe#+ caused significantly less cell death (Fig. 2). Comparable effects of 1 m Fe#+ were also seen in treatments with 50 m and 10 m H O . Hydrogen # #
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
137
% inhibition of spore germination
100
80
60
40
20
0
0
2
4 6 Concentration (mM) of H2O2
8
10
F. 4. Effect of H O on germination of conidia of C. fulum race 4. Conidia were freshly # # harvested from 10 day-old culture plates and washed with distilled water. Fifty microliters of spore suspension (10&) mixed with different concentrations of H O were germinated on slides and # # incubated in a moisture chamber for 24 h at 25 mC. Germinated spores were counted microscopically and the values represent the meanpSD of three independent experiments with 200 conidia for each treatment.
peroxide alone, at concentrations of 100 or 50 m, but not 10 m or less, also caused increased cell death as compared to controls. No effect of 0n01 m Fe#+ was seen in any treatments and no toxic effect was observed when cells were treated with Fe#+ alone at these concentrations. The effect of Fe#+ on cell viability after treatment with elicitor was also examined. Cells were treated with intercellular fluid containing the AVR5 elicitor, at a dilution 1\16 of the original, for 30 min and, then, FeSO :7H O solution was added to the cell % # cultures to give a final concentration of 1 m. Elicitor-induced cell death (i.e., no Fe#+ supplement) was not detected in these assays (Fig. 3), consistent with the observations of Vera-Estrella et al. [44 ]. In contrast, when cells were treated with elicitor, or 10 m H O , and supplemented with 1 m Fe#+, a significant increase in dead cells occurred. # # Cells treated with control solution of intercellular fluids lacking the AVR5 elicitor were not affected by the Fe#+ supplement. The effect of H O on conidial germination # # The conidia of C. fulum germinate poorly in solution ; to overcome this problem, conidial suspensions, treated with different concentrations of H O or untreated, were # # smeared on sterile slides incubated in moisture chambers. Conidial germination started
138
H. Lu and V. J. Higgins 200
Length of germ tubes (mm)
160
120
80
40
0
0
5
10
20
40
80
Concentration (mM) of H2O2 F. 5. Effect of H O on germ tube growth of C. fulum race 4. Fungal conidia were first # # germinated on slides for 24 h and then H O solutions of varying concentrations were added onto # # the slides and the treatments were incubated for a further 18 h. Germ tube length was measured and the values represent the meanpSD of two independent experiments with 50 germ tubes measured for each experiment. Average germ tube length at the time of H O treatment (i.e., # # 24 h) was 23n58p10n69 mm long.
9 to 12 h after incubation and more than 95 % of conidia germinated under these conditions. The germinated conidia were counted at 18 h when the length of most germ tubes exceeded the length of the conidia. Significant inhibition of germination was evident at 4 m H O and was almost complete at 6 m H O (Fig. 4). # # # # Effect of H O on germ tube elongation # # Inhibition of conidial germination of C. fulum is not likely to be a factor in resistant tomato cultivars, as race specific elicitors are assumed to be produced by penetrating hyphae [11 ]. Inhibition of germ tube or mycelial growth was noted in histological studies [32 ]. Thus, the effect of H O on germ tube elongation of germinated conidia # # (24 h incubation) was tested by treatment with different concentrations of H O and # # an additional incubation for 18 h. Germ tube elongation was significantly inhibited by 5 m H O and was completely inhibited by 20 m (Fig. 5), as compared to controls # # measured at the time of H O application, i.e., 24 h. # # To determine if the effect of H O on term tubes was fungitoxic or fungistatic, the # # viability of germ tubes treated with H O was examined using FDA. Treated germ # # tubes were stained with FDA 18 h after treatment and then observed with a micro-
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
139
scope. Non-fluorescent germ tubes indicated loss of a functional plasma membrane or decreased esterase activity, i.e., death. Concentrations of 5 m H O or higher # # caused germ tube death. The calculated ED value for H O on C. fulum germ tube &! # # death was 22n4 m (Fig. 6). 100
% of germ tube death
80
60
40
20
0 0
20
60 40 Concentration (mM) of H2O2
80
100
F. 6. Effect of H O on germ tube viability of C. fulum race 4. Conidia were first germinated # # on slides for 24 h and then H O solutions of varying concentrations were added onto the slides # # and the treatments were incubated for a further 24 h. Germ tube viability was assayed using the FDA method and the values represent three experiments with 50 germ tubes for each treatment.
Unlike the results for tomato cell cultures, the addition of 1 m FeSO with H O % # # at 10, 20 and 40 m to germinated conidia, had no significant effect on germ tube viability as compared with H O alone. The treatment means fell within the range # # shown for each of these H O concentrations in Fig. 6. Similarly, there was no # # significant effect of 1 m FeSO on the degree to which term tube growth was affected % by H O at 5, 10 and 20 m (data not shown). # # DISCUSSION
It is commonly known that AOS, e.g., O −, H O and OHd, cause cell damage via lipid # # # peroxidation, denaturation of proteins, and mutation of DNA [22 ]. The oxidative burst in neutrophils has been demonstrated to kill invading bacteria, e.g., in persons with a genetic inability to mount an efficient oxidative burst, bacterial infection results in chronic granulomatous disease, agonizing bacterial infections ending in early death
140
H. Lu and V. J. Higgins
[3 ]. As some AOS (e.g., OHd) are extremely reactive and indiscriminate in the targets which they attack, the production of AOS in neutrophils is tightly regulated both in space and time, e.g., they are mainly produced within phagolysosomes and produced only when NADPH oxidase, the enzyme responsible for AOS production, is activated in response to specific signals. Plant cells, however, are unable to phagocytose and are capable of response to many different stimuli [46 ]. Thus, AOS produced via an oxidative burst may affect not only the invading pathogen, but also the host cells themselves, unless the host cells have a high antioxidant capacity. It has been reported that AOS, produced by plant cell interaction with a pathogen, caused host cell death [35 ] and inhibited fungal spore germination [42 ]. In the present study, the toxicity of H O to both host cells and to the pathogenic fungus, C. fulum, was examined in an # # attempt to partially determine the role(s) of AOS in the interaction between tomato and this fungus. High levels of H O were required to kill tomato cells in both leaf tissues and # # cultured cell suspensions. The H O concentrations which caused cell death in this # # study were significantly higher than the amounts estimated from the oxidative burst in this system [37 ] and in other reported plant–pathogen interactions [39 ]. Although the leaf necrosis and cell death assays are difficult to compare, the direct toxicity of H O # # to leaf tissues appeared to be less efficient than to cell cultures. This implies that the green tissues of plants are more tolerant of oxidants, probably due to the chloroplast which is one of the main sites in plant cells to generate AOS and which thus possesses a highly efficient antioxidant system [1 ]. While Levine et al. [34, 35 ] suggested that programmed cell death of soybean cells occurred on exposure to 8–10 m H O , such H O concentrations had minimal effect # # # # on tomato cells. The toxicity of H O , and of elicitor, to tomato cell cultures was # # increased by supplementation with iron salts. The increased toxicity was probably contributed by OHd which is produced by an iron-catalysed reaction, called the Fenton reaction. Although H O has been shown to inactivate some enzymes, it is # # generally believed that most damage to living cells cannot be attributed to direct actions of H O or O − [22 ], as they have relatively poor reactivity with most organic # # # molecules in biological systems [19 ]. It is thought, therefore, that most of the damage associated with AOS toxicity is due, not to O − or H O , but instead to the OHd, the # # # most reactive species known in biological systems [24 ]. Although only a small amount of iron is required for catalysing the generation of OHd [18 ], the availability of iron to stimulate OHd generation in living cells is very limited, as cells have a set of protectant systems to handle iron ions [22 ]. In plants, a very important component of antioxidant defenses is based on mechanisms to remove iron from solution [9, 18 ]. Molecules, such as phytoferritin and phytic acid which are very abundant in plant cells, are considered to prevent OHd generation by forming complexes with iron [7, 20 ]. Thus, H O , either # # generated from an elicitor-induced oxidative burst or applied exogenously at relatively low concentrations, might not be expected to directly affect the viability of tomato cells. Yet, elicitor, but not moderate amounts of H O , triggered cell death in leaf # # tissue. One possible explanation is that elicitor in some manner affects iron metabolism in green tissue, resulting in OHd production. Obviously, the use of one exogenous application of H O to simulate elicitor-induced # # H O fails to take into account differences in the length of the cell exposure to H O . # # # #
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
141
Levine et al. [36 ] showed that, for soybean cell cultures, exogenously added H O # # (10 m) was destroyed within 10 min yet cell death was still observed. Similar experiments (data not shown) confirmed the rapid destruction of H O by our tomato # # cell culture with the actual rates dependent on cell numbers and the concentration of H O . Tests for rapid destruction of injected H O in leaf tissue are not possible with # # # # apoplastic probes because of the time required for the liquid from the initial injection to be transpired before the probes can be injected. The H O was more toxic to C. fulum than to tomato cells. Peng and Kuc [42 ] # # reported that spore germination of a number of fungi was completely inhibited by 26n1 µ H O . Growth of the bacterial pathogen, Erwinia carotoora subsp. carotoora, # # and the fungal pathogen, Phytophthora infestans, was shown to be inhibited by 100 µ H O [47 ]. In our experiments, inhibition of the germination of C. fulum conidia was # # achieved by 6 m H O but not by treatment with micromolar levels. As the oxidative # # burst is known to be triggered by race specific elicitors which are produced post penetration, inhibition of conidial germination by H O is probably irrelevant. The # # AOS from the oxidative burst most probably react with the penetrating hyphae. The sensitivity of fungal hyphae to the toxicity of H O has not been demonstrated before # # either in itro or in io. The results of this study show that the levels of H O # # required to inhibit germ tube growth are similar to those required to inhibit conidial germination, but that a higher level of H O is required to kill germ tubes. Obviously, # # the toxic effect of H O on penetrating hyphae of the fungus in io needs to be # # confirmed. Our data suggest that a significant difference in antioxidant capacity exists between the tomato plant and the pathogenic fungus, C. fulum. Variable toxicity of H O to # # cells and organisms has been observed, e.g., some bacteria and animal cells are injured by H O at micromolar concentrations, whereas photosynthetic algae generate and # # release large amounts of it [21 ]. As plants conduct various reductive processes such as photosynthesis, and so are particularly vulnerable to oxidative damage, a ubiquitous and multi-layered antioxidant defense has evolved [9 ]. As fungi are at somewhat less risk of oxidative damage than green plants [9 ], it is reasonable to expect that the antioxidant defenses of fungi are simpler than those in plants. This more effective antioxidant capacity should favor the host plant in plant– pathogen interactions. Transgenic potato expressing a gene encoding the H O # # generating glucose oxidase showed increased resistance to E. c. carotoora and P. infestans [47 ]. While the growth of these pathogens was inhibited by the elevated H O levels # # in plant tissues, the plant itself grew normally, indicating that the plant is capable of tolerating fairly high levels of H O without developing visible growth abnormalities. # # Treatment of potato leaf tissue with digitonin, which is a steroid saponin and not fungitoxic by itself, induced protection against P. infestans [17 ]. As superoxide dismutase treatment abolished the protective effect of digitonin, this protection was suggested to be due to the enhanced O − generation induced by digitonin. The differential # antioxidant capacity of plants and pathogens is also used to advantage in the control of plant diseases by certain fungicides, e.g., probenazole, fthalide, and tricyclazole [41 ]. These fungicides stimulate the generation of AOS in the infected plants and\or inactivate the antioxidant system in pathogen, resulting in the control of the pathogens. In the tomato–C. fulum interaction, it is possible that upon stimulation by the
142
H. Lu and V. J. Higgins
penetrating hyphae, plant cells undergoing a HR produce a certain amount of AOS which, combined with other defense mechanisms, arrests the growth of the fungus but do not affect the cells beyond the penetration site, thereby limiting further spread of the fungus. REFERENCES 1. Alscher RG, Donahue JL, Cramer CL. 1997. Reactive oxygen species and antioxidants : Relationships in green cells. Physiologia Plantarum 100 : 224–233. 2. Alvarez ME, Pennell RI, Meijer PJ, Ishikawa A, Dixon RA, Lamb C. 1998. Reactive oxygen intermediates mediate a systemic signal network in the establishment of plant immunity. Cell 92 : 773–784. 3. Babior BM, Benna JE, Chanock SJ, Smith RM. 1997. The NADPH oxidase of leukocytes : The respiratory burst oxidase. In : Scandalios JG, ed. Oxidatie Stress and the Molecular Biology of Antioxidant Defenses. Plainsview : Cold Spring Harbor Laboratory Press, pp. 737–783. 4. Baker CJ, Orlandi EW. 1995. Active oxygen in plant pathogenesis. Annual Reiew of Phytopathology 33 : 299–321. 5. Batista UG, Higgins VJ. 1991. Accumulation of phytoalexins in the compatible interaction between Cladosporium fulum and tomato in relation to colonization. Canadian Journal of Botany 69 : 822–830. 6. Bestwick CS, Brown IR, Bennett MHR, Mansfield JW. 1997. Localization of hydrogen peroxide accumulation during the hypersensitive reaction of lettuce cells to Pseudomonas syringae pv phaseolicola. The Plant Cell 9 : 209–221. 7. Bienfait HF, der Mark V. 1983. Phytoferritin and its role in iron metabolism. In : Robb DA, Pierpoint WS, eds. Metals and Micronutrients : Uptake and Utilization. London : Academic Press, pp. 111–123. 8. Chamnongpol S, Willekens H, Moeder W, Langebartels C, Sandermann H, Van Montagu M, Inze! D, Van Camp W. 1998. Defense activation and enhanced pathogen tolerance induced by H O # # in transgenic tobacco. Proceedings of the National Academy of Sciences USA 95 : 5818–5823. 9. Dalton DA. 1995. Antioxidant defenses of plants and fungi. In : Ahamad S, ed. Oxidatie Stress and Antioxidant Defenses In Biology. Chapman & Hall, pp. 298–355. 10. De Wit PJGM. 1977. A light and scanning-electron microscopic study of infection of tomato plants by virulent and avirulent races of Cladosporium fulum. Netherlands Journal of Plant Pathology 83 : 109–122. 11. De Wit PJGM. 1992. Molecular characterization of gene-for-gene systems in plant–fungus interactions and the application of avirulence genes in control of plant pathogens. Annual Reiew of Phytopathology 30 : 391–418. 12. De Wit PJGM, Flach W. 1979. Differential accumulation of phytoalexins in tomato leaves, but not in fruits, after inoculation with virulent and avirulent races of Cladosporium fulum. Physiological Plant Pathology 15 : 257–267. 13. De Wit PJGM, Spikman G. 1982. Evidence for the occurrence of race and cultivar-specific elicitors of necrosis in intercellular fluids of compatible interactions of Cladosporium fulum and tomato. Physiological Plant Pathology 21 : 1–11. 14. De Wit PJGM, Van der Meer FE. 1986. Accumulation of the pathogenesis-related tomato leaf protein P14 as an early indicator of incompatibility in the interaction between Cladosporium fulum (Syn. Fulia fula) and tomato. Physiological and Molecular Plant Pathology 28 : 203–214. 15. DeLeo FR, Quinn MT. 1996. Assembly of the phagocyte NADPH oxidase : molecular interaction of oxidase proteins. Journal of Leukoyte Biology 60 : 677–691. 16. Doke N. 1997. The oxidative burst : Roles in signal transduction and plant stress. In : Scandalios JG, ed. Oxidatie Stress and the Molecular Biology of Antioxidant Defenses. Plainsview : Cold Spring Harbor Laboratory Press, pp. 785–813. 17. Doke N, Chai HB. 1985. Activation of superoxide generation and enhancement of resistance against compatible races of Phytophthora infestans in potato plants treated with digitonin. Physiological Plant Pathology 27 : 323–334. 18. Foyer CH, Lelandais M, Kunert KJ. 1994. Photooxidative stress in plants. Physiologia Plantarum 92 : 696–717. 19. Foyer CH, Lopez-Delgado H, Dat JF, Scott IM. 1997. Hydrogen peroxide- and glutathioneassociated mechanisms of acclimatory stress tolerance and signaling. Physiologia Plantarum 100 : 241–254. 20. Graf E, Eaton JW. 1990. Antioxidant functions of phytic acid. Free Radicals in Biology and Medicine 8 : 61–69. 21. Halliwell B, Gutteridge JMC. 1989. Free Radicals in Biology and Medicine, 2nd Ed., Oxford : Oxford University Press.
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
143
22. Halliwell B, Gutteridge JMC. 1990. Role of free radicals and catalytic metal ions in human disease : An overview. Methods in Enzymology 186 : 1–85. 23. Hammond-Kosack KE, Jones JDG. 1996. Resistance gene-dependent plant defense responses. The Plant Cell 8 : 1773–1791. 24. Jones OTG. 1994. The regulation of superoxide production by the NADPH oxidase of neutrophils and other mammalian cells. BioEssays 16 : 919–923. 25. Joosten MHAJ, Cozijnsen TJ, De Wit PJGM. 1994. Host resistance to a fungal tomato pathogen lost by a single base pair mutation in an avirulence gene. Nature 367 : 384. 26. Joosten MHAJ, De Wit PJGM. 1989. Identification of several pathogenesis-related proteins in tomato leaves inoculated with Cladosporium fulum (syn. Fulia fula) as 1,3-β-glucanases and chitinases. Plant Physiology 89 : 945–951. 27. Joosten MHAJ, Verbakel HM, Nettekoven ME, Van Leeuwen J, Van der Vossen RTM, De Wit PJGM. 1995. The phytopathogenic fungus Cladosporium fulum is not sensitive to the chitinase and β1,3-glucanase defence proteins of its host tomato. Physiological and Molecular Plant Pathology 46 : 45–59. 28. Joosten MHAJ, Vogelsang R, Cozijnsen TJ, Verberne MC, De Wit PJGM. 1997. The biotrophic fungus Cladosporium fulum circumvents Cf-4-mediated resistance by producing unstable AVR4 elicitors. The Plant Cell 9 : 367–379. 29. Kazan K, Murray FR, Goulter KC, Llewellyn DJ, Manners JM. 1998. Induction of cell death in transgenic plants expressing a fungal glucose oxidase. Molecular Plant–Microbe Interactions 11 : 555–562. 30. Kooman-Gersmann M, Hone! e G, Bonnema G, De Wit PJGM. 1996. A high affinity binding site for the AVR9 peptide elicitor of Cladosporium fulum is present on plasma membranes of tomato and other solanaceous plants. The Plant Cell 8 : 929–939. 31. Lamb C, Dixon RA. 1997. The oxidative burst in plant disease resistance. Annual Reiew of Plant Physiological and Plant Molecular Biology 48 : 251–275. 32. Lazarovits G, Higgins VJ. 1976. Histological comparison of Cladosporium fulum race 1 on immune, resistant and susceptible tomato varieties. Canadian Journal of Botany 54 : 224–234. 33. Lazarovits G, Higgins VJ. 1976. Ultrastructure of susceptible, resistant and immune reactions of tomato to races of Cladosporium fulum. Canadian Journal of Botany 54 : 235–249. 34. Levine A, Pennell RI, Alvarez MA, Palmer R, Lamb C. 1996. Calcium-mediated apoptosis in a plant hypersensitive resistance response. Current Biology 6 : 427–437. 35. Levine A, Tenhaken R, Dixon R, Lamb C. 1994. H O from the oxidative burst orchestrates the # # plant hypersensitive disease response. Cell 79 : 583–593. 36. Lu H, Higgins VJ. 1993. Partial characterization of a non-proteinaceous suppressor of non-specific elicitors from Cladosporium fulum (syn. Fulia fula). Physiological and Molecular Plant Pathology 42 : 427–439. 37. Lu H, Higgins VJ. 1998. Measurement of active oxygen species generated in planta in response to elicitor AVR9 of Cladosporium fulum. Physiological and Molecular Plant Pathology 52 : 35–51. 38. May MJ, Hammond-Kosack KE, Jones JDG. 1996. Involvement of reactive oxygen species, glutathione metabolism, and lipid peroxidation in the Cf-gene-dependent defense response of tomato cotyledons induced by race-specific elicitors of Cladosporium fulum. Plant Physiology 110 : 1367–1379. 39. Mehdy MC, Sharma YK, Sathasivan K, Bays NW. 1996. The role of activated oxygen species in plant disease resistance. Physiologia Plantarum 98 : 365–374. 40. Murashige T, Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15 : 473–495. 41. Nikolaev ON, Aver’yanov AA, Lapikova VP, Djavakhia VD. 1994. Possible involvement of reactive oxygen species in action of some anti-blast fungicides. Pesticide Biochemistry and Physiology 50 : 219–228. 42. Peng M, Kuc J. 1992. Peroxidase-generated hydrogen peroxide as a source of antifungal activity in itro and on tobacco leaf discs. Phytopathology 82 : 696–699. 43. Turner JG, Novacky A. 1974. The quantitative relationship between plant and bacterial cells involved in the hypersensitive reaction. Phytopathology 64 : 885–890. 44. Vera-Estrella R, Blumwald E, Higgins VJ. 1992. Effect of specific elicitors of Cladosporium fulum on tomato suspension cells. Plant Physiology 99 : 1208–1215. 45. Widholm JM. 1972. The use of fluorescein diacetate and phenosafranine for determining viability of cultured plant cells. Stain Technology 47 : 189–194. 46. Wojtaszek P. 1997. Oxidative burst : an early plant response to pathogen infection. Biochemical Journal 322 : 681–692. 47. Wu G, Shortt VJ, Lawrence EG, Levine EB, Fitzsimmons KC, Shah DM. 1995. Disease resistance conferred by expression of a gene encoding H O -generating glucose oxidase in transgenic potato # # plants. The Plant Cell 7 : 1357–1368. 48. Xing T, Higgins VJ, Blumwald E. 1997. Race-specific elicitors of Cladosporium fulum promote translocation of cytosolic components of NADPH oxidase to the plasma membrane of tomato cells. The Plant Cell 9 : 249–259.
The effect of hydrogen peroxide on the viability of tomato cells and of the fungal pathogen Cladosporium fulvum H L and V J. H* Department of Botany, Uniersity of Toronto, Toronto, Ontario, Canada, M5S 3B2 (Accepted for publication November 1998)
An oxidative burst was previously demonstrated to be induced in tomato plants by race specific elicitors of the fungal pathogen Cladosporium fulum. The in planta levels of H O estimated to occur # # during elicitor treatment, were compared with the levels required to show toxicity to host cells and to the fungal pathogen. Injection of Cf-9 tomato leaves with 100 m H O caused an insignificant # # degree of necrosis and 1 H O was required to cause complete leaf necrosis comparable to that # # induced by the AVR9 elicitor. Assays with Cf-5 tomato cell suspensions confirmed the low toxicity of H O to tomato cells but, as expected, the addition of Fe#+ with H O (or with intercellular # # # # fluids containing AVR5 elicitor) enhanced cell death as determined by the Evans Blue assay. Germination and germ tube growth of conidia of C. fulum were significantly retarded by 4–5 m H O , and at higher concentrations, death of germ tubes was observed (ED l 22 m), as # # &! determined by the fluorescein diacetate assay. The addition of Fe#+ with H O had little effect on # # fungal growth or viability in itro. These results suggest that the amount of H O accumulating # # during an elicitor-induced response in leaves may be sufficient to affect fungal colonization but not to affect viability of host cells unless the Fe#+ status in the apoplast is in some way altered by the elicitor to facilitate OHd production via the Fenton reaction. # 1999 Academic Press
INTRODUCTION
Rapid generation of active oxygen species (AOS), called the oxidative burst, is believed to be one of the earliest responses of plants to an attack by a non-pathogen or by an avirulent race of a pathogen [4 ]. AOS generated from the oxidative burst have been associated with several aspects of defense responses in plants, including direct toxicity to invading pathogens [42, 47 ], strengthening of plant cell walls via crosslinking of wall polymers [31 ], activating transcription of defense related genes [8, 35 ], triggering programmed cell death [16, 31, 34, 35 ], and mediating a reiterative signal network underlying systemic resistance responses [2 ]. These reactions of AOS, combined with other defense responses of the host plant, are assumed to lead to incompatibility between the host and pathogen. In the interaction between tomato (Lycopersicon esculentum Mill) and its fungal pathogen Cladosporium fulum Cooke (syn. Fulia fula), an oxidative burst has been * To whom correspondence should be addressed. Abbreviations used in text : AOS, active oxygen species ; FDA, fluorescein diacetate ; HR, hypersensitive response. 0885–5765\99\050131j13 $30.00\0
# 1999 Academic Press
132
H. Lu and V. J. Higgins
demonstrated to be induced by race specific elicitors, the products of Ar genes in the pathogen, in undetached leaves [37 ], in cotyledons [38 ], and in cell suspension cultures [44 ]. This induction of the oxidative burst was observed within 1–2 h in leaves or in cotyledons and within a few minutes in suspension cells. Data obtained from in io experiments using inhibitors (unpublished data), combined with those using isolated cell membranes [48 ], suggest that the AOS result from the activation of a plasma membrane NADPH oxidase in a manner similar to that of phagocytic neutrophils [3, 15 ]. In neutrophils, the function of the oxidative burst in killing the invading pathogens is well established [24 ], but in the tomato-C. fulum interaction, the role of the oxidative burst has not been fully investigated. In the tomato-C. fulum interaction, incompatibility is generally characterized by hypersensitive cell death which occurs as the fungus penetrates and reaches the substomatal space [32, 33 ]. Although the asynchronous nature of this stomatal penetration makes it difficult to correlate the restriction of fungal growth to cell death, most histopathology studies [10, 32, 33 ] suggest such a correlation. The hypersensitive response (HR) in this system was demonstrated to be triggered by race specific elicitors [11 ]. Although the molecular events and the recognition of some of these elicitors, e.g., AVR4 and AVR9, by host plants have been extensively studied [11, 25, 28, 30 ], the mechanism of the elicitor-induced leaf cell death is still unknown. On the other hand, several possibilities exist for the causes of fungal death. Rapid synthesis of pathogenesisrelated (PR) proteins, including chitinases and β-1,3-glucanases, was observed in incompatible interactions [14, 26 ]. However, these enzymes were unable to inhibit the growth of C. fulum in itro [27 ]. Accumulation of phytoalexins was more rapid in incompatible interactions than in compatible ones and these compounds were shown to have antifungal activity [12 ], but the extent of their role in the restriction of the growth of C. fulum continues to be debated [5, 23 ]. Likewise, a rapid localized deposition of callose in host cell walls at penetration sites was demonstrated in incompatible interactions [33 ]. It was suggested that the accumulation of callose might prevent diffusion of nutrients to the apoplast, resulting in the starvation and death of the fungus growing in the apoplast [33 ]. Again, conclusive evidence is lacking. It has been reported that plant cell death is triggered by H O generated from an # # oxidative burst in soybean cells [35 ] or produced by a fungal glucose oxidase in transgenic tobacco plants [29 ]. In addition, spore germination of several fungi has been shown to be inhibited by H O at micromolar levels [42 ], although, the inhibitory # # effect of H O on fungal germ tubes or hyphae has not been examined. In the present # # study, we investigated the possible direct role of the products of oxidative burst on the interaction between tomato and C. fulum. The effects of AOS, especially H O , on the # # viability of both host cells and pathogen were investigated. Here, we present data supporting the hypothesis that H O may contribute to the inhibition of fungal growth # # but it is unlikely to be responsible for host cell death unless the fungus interferes in some way with iron metabolism. MATERIALS AND METHODS
Plant materials Seeds of tomato cv. Bonny Best, which carry no resistance genes to tomato leaf mould, and cv. Moneymaker, near isogenic for resistance gene Cf-5 or Cf-9 were obtained as
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
133
described before [36 ]. Seeds were germinated in Plant Cell Paks containing Pro-Mix BX (Premier Peat Moss Ltd., Quebec, Canada) and seedlings were transplanted individually into 15 cm pots containing the same mix. Cell suspension cultures derived from a line of tomato with the resistance gene Cf-5 were grown in 500 ml Erlenmeyer flasks containing 100 ml of MS medium [40 ], in the dark at 25 mC on a rotatory shaker at 120 rpm and subcultured weekly [44 ]. Preparation of intercellular fluid containing the elicitor AVR5 Tomato plants cv. Bonny Best were inoculated with race 4 or 5 of C. fulum for production of intercellular fluids containing, or lacking, the elicitor AVR5, respectively, and intercellular fluids were prepared according to the method of De Wit and Spikman [13 ]. Intercellular fluids from the combination BB\r4 (e.g. isolated from leaves of Bonny Best inoculated with race 4 of C. fulum) are referred to as elicitor, whereas the combination BB\r5 is referred to as control. The effect of H O on leaf necrosis and on the viability of suspension cells # # Leaves of tomato plants were injected with various concentrations of H O and injected # # plants were incubated for 18 h at the same conditions as prior to treatment. The degree of necrosis was assessed by a rating system (see caption Fig. 1). Plant cells were treated with various concentrations of H O and\or supplemented # # with various concentrations of Fe#+ (FeSO :7H O) and incubated for 18 h. Cell death % # was determined using Evans Blue, a nonpermeating pigment of low phytotoxicity, as described by Turner and Novacky [43 ]. The selective staining of dead cells with Evans Blue depends upon exclusion of this pigment from living cells by the intact plasmalemma, whereas it passes through the damaged plasmalemma of dead cells and accumulates as a blue protoplasmic stain. Treated and untreated Cf-5 tomato cells were incubated for 15 min with 0n05 % Evans Blue and then washed extensively to remove excess and unbound dye. Dye bound to dead cells was solubilized in 50 % methanol with 1 % SDS for 30 min at 50 mC and quantified from measurements of absorbance at 600 nm. The effect of H O on the germination of conidia of C. fulvum # # Conidial suspensions (10& spores ml−") of C. fulum were prepared as previously described [5 ]. A known quantity of H O was added to 200 µl samples of the spore # # suspension and 50 µl of these treated spores were smeared on sterile microscope slides which were incubated at 25 mC for 24 h in Petri dish moisture chambers at which time the percentage of germinated spores was determined microscopically. Spores were considered germinated when the length of the germ tube exceeded the diameter of the spore. The effect of H O on growth and viability of germ tubes of C. fulvum # # The effect of H O on germ tube growth of C. fulum was tested on slides with untreated # # spores previously germinated for 24 h as described above. A known quantity of H O # # in 50 µl H O was added to the 50 µl of spore suspension on the slides and was well # mixed with the germinating spores by gently rotating the slides. The spores were incubated for a further 18 h and the lengths of the germ tubes were measured
134
H. Lu and V. J. Higgins
4
Necrosis rating
3
2
1
0
100
250
500
750
1000
Concentration (mM) of H2O2 F. 1. Leaf necrosis induced by exogenously applied H O . Leaves of tomato Cf-9 plants were # # injected with varying concentrations of H O and the injected plants were maintained in a growth # # room for 18 h under the conditions as described before. Necrosis rating was based on a scale of 0–4 [36, 37 ] in which 0 refers to no necrosis and 1, 2, 3 and 4 refer to about 25 %, 50 %, 75 % and 100 % necrotic area respectively, occurring on injected leaf areas. Values are meanspSD for three experiments with three replicates each.
microscopically. The average length of the germ tubes measured just before the H O # # treatment was used to determine the germ tube length at the time of treatment. Viability of germ tubes after treatment with H O for 18 h was determined by the # # fluorescein diacetate (FDA) assay as described by Widholm [45 ]. A stock solution of FDA (5 mg ml−") in acetone was diluted 1 : 100 with distilled water and then added to the assay slides. After 5 min, the number of fluorescent (viable) germ tubes was determined for 200 germ tubes per replicate under a Zeiss epifluorescence microscope, using a ParaLens Microscope Adapter\Fiber Optics System (Becton Dickinson Inc.) with a FITC filter cluster (520 nm barrier filter, 510 nm dichroic beam splitter, and a 470–490 nm excitation filter) and a fiber optic illuminator (Chiu Technical Corporation FO-150–115) with a 150 W quartz halogen lamp. Statistical analysis Comparisons of differences obtained in all experiments were made using the t-test. P 0n01 was considered significant.
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
135
RESULTS
The H O concentration needed to cause tomato leaf necrosis # # Leaves of Cf-9 tomato plants were injected with different concentrations of H O and # # the levels of necrosis induced by H O were rated for visible necrosis. With this # # approach, a concentration of 1 H O was required to cause complete leaf necrosis # # (i.e., a rating of about 4) comparable to that induced by race specific elicitor, e.g., AVR9 (Fig. 1). Little visible necrosis was observed when the leaves were injected with 100 m H O , hence lower concentrations (e.g., 20 m) mimicking those estimated to # # be produced by elicitors were not tested. When the leaves were injected with 1 H O , # # the injected areas became darkened and shrunken within 1 h, and the injected aqueous solution, unlike that with lower concentrations of H O or with elicitor preparations, # # was not transpired from the leaf tissues. The appearance and timing (about 6–9 h) of visible necrosis induced by H O at concentrations of 250, 500 and 750 m were similar # # to that induced by race specific elicitors.
5
1 mM Fe2+
Absorbance at 600 nm (cell death)
0.1 mM Fe2+ 0.01 mM Fe2+ 4
Without Fe2+
3
2
1
0
0
1
10
50
100
Concentration (mM) of H2O2 F. 2. Effects of H O and Fe#+ on the viability of tomato Cf-5 cells. Cell cultures were treated # # with different concentrations of H O and FeSO :7H O and incubated for 18 h on a rotatory # # % # shaker in the dark. Cell death was measured by uptake of Evans Blue stain by dead cells and quantified by absorbance of methanol\SDS extracts of the cell at 600 nm. The values represent the meanpSD of three independent experiments with three replicates each.
136
H. Lu and V. J. Higgins
Without Fe2+
Absorbance at 600 nm (cell death)
3
1 mM Fe2+
2
1
0
Untreated
Elicitor
Control
10 mM H2O2
F. 3. The effect of Fe#+ supplement on elicitor- or H O -induced cell death as measured by
# #
Evans Blue staining. Cf-5 tomato cells were treated with either intercellular fluid containing the AVR5 elicitor (BB\r4) or with intercellular fluid lacking the AVR5 elicitor (BB\r5) (control), supplemented with 1 m of FeSO :7H O after 30 min. Untreated cells and cells treated with % # 10 m H O also served as controls. Evans Blue staining was assessed at 18 h from measurements # # of absorbance of methanol\SDS extracts of the cells at 600 nm. The values represent meanspSD of three independent experiments with two replicates each.
The H O concentration required to cause cell death with or without Fe#j # # The toxicity of H O to living cells is known to be variable and the variability has been # # accounted for both by the activity of H O -scavenging enzymes, and by the rate of # # conversion of O − and H O into more highly reactive radicals, e.g., OHd [22 ]. The # # # conversion of H O into OHd, the Fenton reaction, is mediated by reduced transition # # metals such as Fe#+ which frequently results from the reaction of Fe$+ and O −. Plus, # the availability and release of Fe#+ from storage proteins, transport proteins, or ironcontaining enzymes is considered to be important in regulating the toxicity of H O to # # plant cells [6 ]. On such consideration, the effect of H O , supplemented with an Fe#+ # # salt, on viability of tomato cells was examined. Cf-5 cells were treated with different concentrations of H O and FeSO :7H O solution and incubated in the dark with # # % # shaking overnight (about 18 h). Cell viability was determined by the amount of Evans Blue taken up by dead cells as measured by absorbance after solubilization from cells [43 ]. Calibration of the Evans Blue assay with the percentage dead cells via the FDA viability test showed that the maximum absorbance (" 5n0) equaled about 100 % dead cells. Treatment with 100 m H O and 1 m Fe#+ caused 100 % cell death whereas # # 0n1 m or less Fe#+ caused significantly less cell death (Fig. 2). Comparable effects of 1 m Fe#+ were also seen in treatments with 50 m and 10 m H O . Hydrogen # #
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
137
% inhibition of spore germination
100
80
60
40
20
0
0
2
4 6 Concentration (mM) of H2O2
8
10
F. 4. Effect of H O on germination of conidia of C. fulum race 4. Conidia were freshly # # harvested from 10 day-old culture plates and washed with distilled water. Fifty microliters of spore suspension (10&) mixed with different concentrations of H O were germinated on slides and # # incubated in a moisture chamber for 24 h at 25 mC. Germinated spores were counted microscopically and the values represent the meanpSD of three independent experiments with 200 conidia for each treatment.
peroxide alone, at concentrations of 100 or 50 m, but not 10 m or less, also caused increased cell death as compared to controls. No effect of 0n01 m Fe#+ was seen in any treatments and no toxic effect was observed when cells were treated with Fe#+ alone at these concentrations. The effect of Fe#+ on cell viability after treatment with elicitor was also examined. Cells were treated with intercellular fluid containing the AVR5 elicitor, at a dilution 1\16 of the original, for 30 min and, then, FeSO :7H O solution was added to the cell % # cultures to give a final concentration of 1 m. Elicitor-induced cell death (i.e., no Fe#+ supplement) was not detected in these assays (Fig. 3), consistent with the observations of Vera-Estrella et al. [44 ]. In contrast, when cells were treated with elicitor, or 10 m H O , and supplemented with 1 m Fe#+, a significant increase in dead cells occurred. # # Cells treated with control solution of intercellular fluids lacking the AVR5 elicitor were not affected by the Fe#+ supplement. The effect of H O on conidial germination # # The conidia of C. fulum germinate poorly in solution ; to overcome this problem, conidial suspensions, treated with different concentrations of H O or untreated, were # # smeared on sterile slides incubated in moisture chambers. Conidial germination started
138
H. Lu and V. J. Higgins 200
Length of germ tubes (mm)
160
120
80
40
0
0
5
10
20
40
80
Concentration (mM) of H2O2 F. 5. Effect of H O on germ tube growth of C. fulum race 4. Fungal conidia were first # # germinated on slides for 24 h and then H O solutions of varying concentrations were added onto # # the slides and the treatments were incubated for a further 18 h. Germ tube length was measured and the values represent the meanpSD of two independent experiments with 50 germ tubes measured for each experiment. Average germ tube length at the time of H O treatment (i.e., # # 24 h) was 23n58p10n69 mm long.
9 to 12 h after incubation and more than 95 % of conidia germinated under these conditions. The germinated conidia were counted at 18 h when the length of most germ tubes exceeded the length of the conidia. Significant inhibition of germination was evident at 4 m H O and was almost complete at 6 m H O (Fig. 4). # # # # Effect of H O on germ tube elongation # # Inhibition of conidial germination of C. fulum is not likely to be a factor in resistant tomato cultivars, as race specific elicitors are assumed to be produced by penetrating hyphae [11 ]. Inhibition of germ tube or mycelial growth was noted in histological studies [32 ]. Thus, the effect of H O on germ tube elongation of germinated conidia # # (24 h incubation) was tested by treatment with different concentrations of H O and # # an additional incubation for 18 h. Germ tube elongation was significantly inhibited by 5 m H O and was completely inhibited by 20 m (Fig. 5), as compared to controls # # measured at the time of H O application, i.e., 24 h. # # To determine if the effect of H O on term tubes was fungitoxic or fungistatic, the # # viability of germ tubes treated with H O was examined using FDA. Treated germ # # tubes were stained with FDA 18 h after treatment and then observed with a micro-
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
139
scope. Non-fluorescent germ tubes indicated loss of a functional plasma membrane or decreased esterase activity, i.e., death. Concentrations of 5 m H O or higher # # caused germ tube death. The calculated ED value for H O on C. fulum germ tube &! # # death was 22n4 m (Fig. 6). 100
% of germ tube death
80
60
40
20
0 0
20
60 40 Concentration (mM) of H2O2
80
100
F. 6. Effect of H O on germ tube viability of C. fulum race 4. Conidia were first germinated # # on slides for 24 h and then H O solutions of varying concentrations were added onto the slides # # and the treatments were incubated for a further 24 h. Germ tube viability was assayed using the FDA method and the values represent three experiments with 50 germ tubes for each treatment.
Unlike the results for tomato cell cultures, the addition of 1 m FeSO with H O % # # at 10, 20 and 40 m to germinated conidia, had no significant effect on germ tube viability as compared with H O alone. The treatment means fell within the range # # shown for each of these H O concentrations in Fig. 6. Similarly, there was no # # significant effect of 1 m FeSO on the degree to which term tube growth was affected % by H O at 5, 10 and 20 m (data not shown). # # DISCUSSION
It is commonly known that AOS, e.g., O −, H O and OHd, cause cell damage via lipid # # # peroxidation, denaturation of proteins, and mutation of DNA [22 ]. The oxidative burst in neutrophils has been demonstrated to kill invading bacteria, e.g., in persons with a genetic inability to mount an efficient oxidative burst, bacterial infection results in chronic granulomatous disease, agonizing bacterial infections ending in early death
140
H. Lu and V. J. Higgins
[3 ]. As some AOS (e.g., OHd) are extremely reactive and indiscriminate in the targets which they attack, the production of AOS in neutrophils is tightly regulated both in space and time, e.g., they are mainly produced within phagolysosomes and produced only when NADPH oxidase, the enzyme responsible for AOS production, is activated in response to specific signals. Plant cells, however, are unable to phagocytose and are capable of response to many different stimuli [46 ]. Thus, AOS produced via an oxidative burst may affect not only the invading pathogen, but also the host cells themselves, unless the host cells have a high antioxidant capacity. It has been reported that AOS, produced by plant cell interaction with a pathogen, caused host cell death [35 ] and inhibited fungal spore germination [42 ]. In the present study, the toxicity of H O to both host cells and to the pathogenic fungus, C. fulum, was examined in an # # attempt to partially determine the role(s) of AOS in the interaction between tomato and this fungus. High levels of H O were required to kill tomato cells in both leaf tissues and # # cultured cell suspensions. The H O concentrations which caused cell death in this # # study were significantly higher than the amounts estimated from the oxidative burst in this system [37 ] and in other reported plant–pathogen interactions [39 ]. Although the leaf necrosis and cell death assays are difficult to compare, the direct toxicity of H O # # to leaf tissues appeared to be less efficient than to cell cultures. This implies that the green tissues of plants are more tolerant of oxidants, probably due to the chloroplast which is one of the main sites in plant cells to generate AOS and which thus possesses a highly efficient antioxidant system [1 ]. While Levine et al. [34, 35 ] suggested that programmed cell death of soybean cells occurred on exposure to 8–10 m H O , such H O concentrations had minimal effect # # # # on tomato cells. The toxicity of H O , and of elicitor, to tomato cell cultures was # # increased by supplementation with iron salts. The increased toxicity was probably contributed by OHd which is produced by an iron-catalysed reaction, called the Fenton reaction. Although H O has been shown to inactivate some enzymes, it is # # generally believed that most damage to living cells cannot be attributed to direct actions of H O or O − [22 ], as they have relatively poor reactivity with most organic # # # molecules in biological systems [19 ]. It is thought, therefore, that most of the damage associated with AOS toxicity is due, not to O − or H O , but instead to the OHd, the # # # most reactive species known in biological systems [24 ]. Although only a small amount of iron is required for catalysing the generation of OHd [18 ], the availability of iron to stimulate OHd generation in living cells is very limited, as cells have a set of protectant systems to handle iron ions [22 ]. In plants, a very important component of antioxidant defenses is based on mechanisms to remove iron from solution [9, 18 ]. Molecules, such as phytoferritin and phytic acid which are very abundant in plant cells, are considered to prevent OHd generation by forming complexes with iron [7, 20 ]. Thus, H O , either # # generated from an elicitor-induced oxidative burst or applied exogenously at relatively low concentrations, might not be expected to directly affect the viability of tomato cells. Yet, elicitor, but not moderate amounts of H O , triggered cell death in leaf # # tissue. One possible explanation is that elicitor in some manner affects iron metabolism in green tissue, resulting in OHd production. Obviously, the use of one exogenous application of H O to simulate elicitor-induced # # H O fails to take into account differences in the length of the cell exposure to H O . # # # #
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
141
Levine et al. [36 ] showed that, for soybean cell cultures, exogenously added H O # # (10 m) was destroyed within 10 min yet cell death was still observed. Similar experiments (data not shown) confirmed the rapid destruction of H O by our tomato # # cell culture with the actual rates dependent on cell numbers and the concentration of H O . Tests for rapid destruction of injected H O in leaf tissue are not possible with # # # # apoplastic probes because of the time required for the liquid from the initial injection to be transpired before the probes can be injected. The H O was more toxic to C. fulum than to tomato cells. Peng and Kuc [42 ] # # reported that spore germination of a number of fungi was completely inhibited by 26n1 µ H O . Growth of the bacterial pathogen, Erwinia carotoora subsp. carotoora, # # and the fungal pathogen, Phytophthora infestans, was shown to be inhibited by 100 µ H O [47 ]. In our experiments, inhibition of the germination of C. fulum conidia was # # achieved by 6 m H O but not by treatment with micromolar levels. As the oxidative # # burst is known to be triggered by race specific elicitors which are produced post penetration, inhibition of conidial germination by H O is probably irrelevant. The # # AOS from the oxidative burst most probably react with the penetrating hyphae. The sensitivity of fungal hyphae to the toxicity of H O has not been demonstrated before # # either in itro or in io. The results of this study show that the levels of H O # # required to inhibit germ tube growth are similar to those required to inhibit conidial germination, but that a higher level of H O is required to kill germ tubes. Obviously, # # the toxic effect of H O on penetrating hyphae of the fungus in io needs to be # # confirmed. Our data suggest that a significant difference in antioxidant capacity exists between the tomato plant and the pathogenic fungus, C. fulum. Variable toxicity of H O to # # cells and organisms has been observed, e.g., some bacteria and animal cells are injured by H O at micromolar concentrations, whereas photosynthetic algae generate and # # release large amounts of it [21 ]. As plants conduct various reductive processes such as photosynthesis, and so are particularly vulnerable to oxidative damage, a ubiquitous and multi-layered antioxidant defense has evolved [9 ]. As fungi are at somewhat less risk of oxidative damage than green plants [9 ], it is reasonable to expect that the antioxidant defenses of fungi are simpler than those in plants. This more effective antioxidant capacity should favor the host plant in plant– pathogen interactions. Transgenic potato expressing a gene encoding the H O # # generating glucose oxidase showed increased resistance to E. c. carotoora and P. infestans [47 ]. While the growth of these pathogens was inhibited by the elevated H O levels # # in plant tissues, the plant itself grew normally, indicating that the plant is capable of tolerating fairly high levels of H O without developing visible growth abnormalities. # # Treatment of potato leaf tissue with digitonin, which is a steroid saponin and not fungitoxic by itself, induced protection against P. infestans [17 ]. As superoxide dismutase treatment abolished the protective effect of digitonin, this protection was suggested to be due to the enhanced O − generation induced by digitonin. The differential # antioxidant capacity of plants and pathogens is also used to advantage in the control of plant diseases by certain fungicides, e.g., probenazole, fthalide, and tricyclazole [41 ]. These fungicides stimulate the generation of AOS in the infected plants and\or inactivate the antioxidant system in pathogen, resulting in the control of the pathogens. In the tomato–C. fulum interaction, it is possible that upon stimulation by the
142
H. Lu and V. J. Higgins
penetrating hyphae, plant cells undergoing a HR produce a certain amount of AOS which, combined with other defense mechanisms, arrests the growth of the fungus but do not affect the cells beyond the penetration site, thereby limiting further spread of the fungus. REFERENCES 1. Alscher RG, Donahue JL, Cramer CL. 1997. Reactive oxygen species and antioxidants : Relationships in green cells. Physiologia Plantarum 100 : 224–233. 2. Alvarez ME, Pennell RI, Meijer PJ, Ishikawa A, Dixon RA, Lamb C. 1998. Reactive oxygen intermediates mediate a systemic signal network in the establishment of plant immunity. Cell 92 : 773–784. 3. Babior BM, Benna JE, Chanock SJ, Smith RM. 1997. The NADPH oxidase of leukocytes : The respiratory burst oxidase. In : Scandalios JG, ed. Oxidatie Stress and the Molecular Biology of Antioxidant Defenses. Plainsview : Cold Spring Harbor Laboratory Press, pp. 737–783. 4. Baker CJ, Orlandi EW. 1995. Active oxygen in plant pathogenesis. Annual Reiew of Phytopathology 33 : 299–321. 5. Batista UG, Higgins VJ. 1991. Accumulation of phytoalexins in the compatible interaction between Cladosporium fulum and tomato in relation to colonization. Canadian Journal of Botany 69 : 822–830. 6. Bestwick CS, Brown IR, Bennett MHR, Mansfield JW. 1997. Localization of hydrogen peroxide accumulation during the hypersensitive reaction of lettuce cells to Pseudomonas syringae pv phaseolicola. The Plant Cell 9 : 209–221. 7. Bienfait HF, der Mark V. 1983. Phytoferritin and its role in iron metabolism. In : Robb DA, Pierpoint WS, eds. Metals and Micronutrients : Uptake and Utilization. London : Academic Press, pp. 111–123. 8. Chamnongpol S, Willekens H, Moeder W, Langebartels C, Sandermann H, Van Montagu M, Inze! D, Van Camp W. 1998. Defense activation and enhanced pathogen tolerance induced by H O # # in transgenic tobacco. Proceedings of the National Academy of Sciences USA 95 : 5818–5823. 9. Dalton DA. 1995. Antioxidant defenses of plants and fungi. In : Ahamad S, ed. Oxidatie Stress and Antioxidant Defenses In Biology. Chapman & Hall, pp. 298–355. 10. De Wit PJGM. 1977. A light and scanning-electron microscopic study of infection of tomato plants by virulent and avirulent races of Cladosporium fulum. Netherlands Journal of Plant Pathology 83 : 109–122. 11. De Wit PJGM. 1992. Molecular characterization of gene-for-gene systems in plant–fungus interactions and the application of avirulence genes in control of plant pathogens. Annual Reiew of Phytopathology 30 : 391–418. 12. De Wit PJGM, Flach W. 1979. Differential accumulation of phytoalexins in tomato leaves, but not in fruits, after inoculation with virulent and avirulent races of Cladosporium fulum. Physiological Plant Pathology 15 : 257–267. 13. De Wit PJGM, Spikman G. 1982. Evidence for the occurrence of race and cultivar-specific elicitors of necrosis in intercellular fluids of compatible interactions of Cladosporium fulum and tomato. Physiological Plant Pathology 21 : 1–11. 14. De Wit PJGM, Van der Meer FE. 1986. Accumulation of the pathogenesis-related tomato leaf protein P14 as an early indicator of incompatibility in the interaction between Cladosporium fulum (Syn. Fulia fula) and tomato. Physiological and Molecular Plant Pathology 28 : 203–214. 15. DeLeo FR, Quinn MT. 1996. Assembly of the phagocyte NADPH oxidase : molecular interaction of oxidase proteins. Journal of Leukoyte Biology 60 : 677–691. 16. Doke N. 1997. The oxidative burst : Roles in signal transduction and plant stress. In : Scandalios JG, ed. Oxidatie Stress and the Molecular Biology of Antioxidant Defenses. Plainsview : Cold Spring Harbor Laboratory Press, pp. 785–813. 17. Doke N, Chai HB. 1985. Activation of superoxide generation and enhancement of resistance against compatible races of Phytophthora infestans in potato plants treated with digitonin. Physiological Plant Pathology 27 : 323–334. 18. Foyer CH, Lelandais M, Kunert KJ. 1994. Photooxidative stress in plants. Physiologia Plantarum 92 : 696–717. 19. Foyer CH, Lopez-Delgado H, Dat JF, Scott IM. 1997. Hydrogen peroxide- and glutathioneassociated mechanisms of acclimatory stress tolerance and signaling. Physiologia Plantarum 100 : 241–254. 20. Graf E, Eaton JW. 1990. Antioxidant functions of phytic acid. Free Radicals in Biology and Medicine 8 : 61–69. 21. Halliwell B, Gutteridge JMC. 1989. Free Radicals in Biology and Medicine, 2nd Ed., Oxford : Oxford University Press.
The viability of tomato cells and the fungal pathogen Cladosporium fulvum
143
22. Halliwell B, Gutteridge JMC. 1990. Role of free radicals and catalytic metal ions in human disease : An overview. Methods in Enzymology 186 : 1–85. 23. Hammond-Kosack KE, Jones JDG. 1996. Resistance gene-dependent plant defense responses. The Plant Cell 8 : 1773–1791. 24. Jones OTG. 1994. The regulation of superoxide production by the NADPH oxidase of neutrophils and other mammalian cells. BioEssays 16 : 919–923. 25. Joosten MHAJ, Cozijnsen TJ, De Wit PJGM. 1994. Host resistance to a fungal tomato pathogen lost by a single base pair mutation in an avirulence gene. Nature 367 : 384. 26. Joosten MHAJ, De Wit PJGM. 1989. Identification of several pathogenesis-related proteins in tomato leaves inoculated with Cladosporium fulum (syn. Fulia fula) as 1,3-β-glucanases and chitinases. Plant Physiology 89 : 945–951. 27. Joosten MHAJ, Verbakel HM, Nettekoven ME, Van Leeuwen J, Van der Vossen RTM, De Wit PJGM. 1995. The phytopathogenic fungus Cladosporium fulum is not sensitive to the chitinase and β1,3-glucanase defence proteins of its host tomato. Physiological and Molecular Plant Pathology 46 : 45–59. 28. Joosten MHAJ, Vogelsang R, Cozijnsen TJ, Verberne MC, De Wit PJGM. 1997. The biotrophic fungus Cladosporium fulum circumvents Cf-4-mediated resistance by producing unstable AVR4 elicitors. The Plant Cell 9 : 367–379. 29. Kazan K, Murray FR, Goulter KC, Llewellyn DJ, Manners JM. 1998. Induction of cell death in transgenic plants expressing a fungal glucose oxidase. Molecular Plant–Microbe Interactions 11 : 555–562. 30. Kooman-Gersmann M, Hone! e G, Bonnema G, De Wit PJGM. 1996. A high affinity binding site for the AVR9 peptide elicitor of Cladosporium fulum is present on plasma membranes of tomato and other solanaceous plants. The Plant Cell 8 : 929–939. 31. Lamb C, Dixon RA. 1997. The oxidative burst in plant disease resistance. Annual Reiew of Plant Physiological and Plant Molecular Biology 48 : 251–275. 32. Lazarovits G, Higgins VJ. 1976. Histological comparison of Cladosporium fulum race 1 on immune, resistant and susceptible tomato varieties. Canadian Journal of Botany 54 : 224–234. 33. Lazarovits G, Higgins VJ. 1976. Ultrastructure of susceptible, resistant and immune reactions of tomato to races of Cladosporium fulum. Canadian Journal of Botany 54 : 235–249. 34. Levine A, Pennell RI, Alvarez MA, Palmer R, Lamb C. 1996. Calcium-mediated apoptosis in a plant hypersensitive resistance response. Current Biology 6 : 427–437. 35. Levine A, Tenhaken R, Dixon R, Lamb C. 1994. H O from the oxidative burst orchestrates the # # plant hypersensitive disease response. Cell 79 : 583–593. 36. Lu H, Higgins VJ. 1993. Partial characterization of a non-proteinaceous suppressor of non-specific elicitors from Cladosporium fulum (syn. Fulia fula). Physiological and Molecular Plant Pathology 42 : 427–439. 37. Lu H, Higgins VJ. 1998. Measurement of active oxygen species generated in planta in response to elicitor AVR9 of Cladosporium fulum. Physiological and Molecular Plant Pathology 52 : 35–51. 38. May MJ, Hammond-Kosack KE, Jones JDG. 1996. Involvement of reactive oxygen species, glutathione metabolism, and lipid peroxidation in the Cf-gene-dependent defense response of tomato cotyledons induced by race-specific elicitors of Cladosporium fulum. Plant Physiology 110 : 1367–1379. 39. Mehdy MC, Sharma YK, Sathasivan K, Bays NW. 1996. The role of activated oxygen species in plant disease resistance. Physiologia Plantarum 98 : 365–374. 40. Murashige T, Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15 : 473–495. 41. Nikolaev ON, Aver’yanov AA, Lapikova VP, Djavakhia VD. 1994. Possible involvement of reactive oxygen species in action of some anti-blast fungicides. Pesticide Biochemistry and Physiology 50 : 219–228. 42. Peng M, Kuc J. 1992. Peroxidase-generated hydrogen peroxide as a source of antifungal activity in itro and on tobacco leaf discs. Phytopathology 82 : 696–699. 43. Turner JG, Novacky A. 1974. The quantitative relationship between plant and bacterial cells involved in the hypersensitive reaction. Phytopathology 64 : 885–890. 44. Vera-Estrella R, Blumwald E, Higgins VJ. 1992. Effect of specific elicitors of Cladosporium fulum on tomato suspension cells. Plant Physiology 99 : 1208–1215. 45. Widholm JM. 1972. The use of fluorescein diacetate and phenosafranine for determining viability of cultured plant cells. Stain Technology 47 : 189–194. 46. Wojtaszek P. 1997. Oxidative burst : an early plant response to pathogen infection. Biochemical Journal 322 : 681–692. 47. Wu G, Shortt VJ, Lawrence EG, Levine EB, Fitzsimmons KC, Shah DM. 1995. Disease resistance conferred by expression of a gene encoding H O -generating glucose oxidase in transgenic potato # # plants. The Plant Cell 7 : 1357–1368. 48. Xing T, Higgins VJ, Blumwald E. 1997. Race-specific elicitors of Cladosporium fulum promote translocation of cytosolic components of NADPH oxidase to the plasma membrane of tomato cells. The Plant Cell 9 : 249–259.