13. Experimental evidence for the action of M. fijiensis toxins on banana photosynthetic apparatus - Busogoro^[16], J.P., J.J. Etame^[17], G. Harelimana^[18], G. Lognay^[19], J. Messiaen^[20], P. Lepoivre^[21], and P. van Cutsem^[22]

Abstract

Fougamou and Grande Naine banana cultivars inoculated with a set of 4 M. fijiensis strains revealed partial resistance and susceptible reactions respectively whatever the fungal strain considered. To investigate a possible involvement of pathogen toxins in the pathogenesis, the fungal metabolites extracted with ethyl acetate (EaCE) from culture filtrates were injected into banana leaves. They induced necrosis and this toxicity of the fungal metabolites seemed to be always light-dependent. Moreover, a correlation between the susceptibility to EaCE and the sensitivity to M. fijiensis infection was observed for the two banana genotypes. Electron microscopy revealed chloroplast swelling in the susceptible cultivar tissues injected with EaCE. This apparent effect of M. fijiensis metabolites on banana chloroplasts was further investigated by quantifying the consequences of one purified metabolite (juglone) and 2 semi-purified fractions of the EaCE on banana isolated chloroplasts. An evident inhibiting effect on the electron exchange ability of banana chloroplasts was revealed. The cultivar most resistant to the pathogen infection presented the greatest tolerance to all the pathogen metabolites.

1. INTRODUCTION

The banana black Sigatoka (BS) disease, caused by Mycosphaerella fijiensis Morelet, is considered as the most economically important disease of banana [1]. Among the different strategies which can be used for the control of BS disease, genetic resistance is the most suitable for small growers in developing countries. Breeding banana for resistance to diseases through field evaluation under natural conditions of infection is time consuming and does not allow to screen large population of genotypes [2,3].

Pathogen toxins could constitute an alternative technique for rapid screening of resistant genotypes in banana allowing early selection at the level of in vitro plant tissues as well as of young plants. Based on the morphological aspect of banana BS symptoms, a possible involvement of phytotoxic compounds released by the pathogen was suggested [2] as indicated by the different metabolites identified in the culture filtrates of M. fijiensis [4,5]. These compounds cannot be successfully used for resistance selection without a preliminary understanding of their role in the development of BS disease and their mode of action.

Molina and Krausz [2] reported the toxicity of M. fijiensis crude extracts revealed by a water-soaked appearance of banana tissues. The same authors observed a correlation between the relative severity of phytotoxicity caused by extracts on a range of banana genotypes and the level of susceptibility to pathogen infection. A new bioassay based on measurements of chlorophyll fluorescence [6] revealed a direct effect of M. fijiensis ethyl acetate crude extracts (EaCE) by a decrease of the vitality index [7]. The light-dependency of this effect suggested the involvement of the photosynthetic apparatus.

An experimental approach to investigate physiological events prevailing in the photosynthetic apparatus relies on isolated chloroplasts [8]. Chloroplasts can be isolated either by mechanical disruption of plant tissues or by enzymatic digestion of protoplasts [9]. Mechanical preparation is more suitable because of its rapidity, economy and high yield of stable chloroplasts. After chloroplast isolation, their functioning ability can be measured by the Hill reaction [10], which is based on their ability to perform an electron exchange.

This paper describes the results acquired by analysing the reactions induced by M. fijiensis metabolites on 2 banana cultivars differing in their respective behaviour against the BS disease. A first analysis of the effects caused by the fungal metabolites on physiologically intact banana chloroplasts is presented.

2. MATERIALS AND METHODS

2.1. Plant and fungal materials

Grande Naine (Musa AAA group) and Fougamou (Musa ABB group) were micropropagated and grown in a greenhouse supplemented with a 16 h light photoperiod at a temperature of about 25°C. The fully expanded second leaf of 8 to 9 week-old acclimatized banana plants were either inoculated with M. fijiensis or used for toxin assays.

Monospore cultures of M. fijiensis (isolate 049 HND originating from Honduras, isolate 281 COOK from Cook Islands and isolates 282 TON and 283 TON from Tonga) were provided by Dr. X. Mourichon, Laboratory of Plant Pathology CIRAD-FLHOR (Montpellier, France). The fungus was maintained and inoculated according to the technique previously described [11].

2.2. Inoculations with the fungal pathogen

For the preparation of inoculum, the fungal strains were cultured on a V8 medium for 12 days. Conidia were then harvested and suspended in distilled water containing 1% of gelatine. The final concentration was adjusted at 2×10⁵ conidia per ml before spraying the suspension on the lower leaf surface until saturation. Incubation of the inoculated plants was performed according to the protocol previously described [11] until the final evaluation of symptoms on the 30^th day after inoculation.

2.3. Production of M.fijiensis toxins

A suspension of M. fijiensis spores (5 ml of 10⁵ conidia.ml^-1) harvested from 12 day-old colonies grown on the V8 modified medium was inoculated in 500 ml of the M1D medium [12] contained in 1-liter Erlenmeyer. During 4 weeks, the cultures were shaken (120 rpm) under a temperature of 25°C and with a 16h light photoperiod. Methanol was then added to the culture (1 vol/1 vol) which was kept at 4°C overnight and filtered through a cheese-cloth to discard the mycelium. M. fijiensis toxins were extracted from the culture filtrates with ethyl acetate according to the technique previously published [12]. The ethyl acetate extract was concentrated at 30°C and dissolved in 10% methanol. This preparation was defined as the "ethyl acetate crude extract" (EaCE) and used for direct bioassay and for further purification. EaCE obtained by several independent extractions were pulled together, dissolved with 10% methanol to a concentration of 5,000 ppm and stored at -70°C before toxin assay. EaCE obtained from uninoculated M1D medium or a solution of 10 % methanol was used as control.

2.4. Fractionation of the crude extracts

For the fractionation of the different constituents of the EaCE, preparative thin layer chromatography (PTLC) was carried out with silica gel plate (G60 Merck 7731, 20×20 cm, 0.5 mm of thickness). The plate was developed in chloroform: methanol (10: 1, v/v) and observed at 254 nm. The fluorescent spots were scraped from the plate, extracted with ethyl acetate and pooled according to their Rf. The pooled fractions were concentrated to dryness under reduced pressure at 30°C and the residue was dissolved in 10% methanol for the bio-assays to the initial volume EaCE. Being unable to weigh the dry residue obtained from the pooled fractions, we (a) made the assumption that the recovery from the scrapped silica gel was complete and (b) expressed fictitiously the concentration of the pooled fraction in methanol 10% by the virtual concentration of the total EaCE residue which would be similarly diluted.

2.5. Injection of toxins into banana leaves

For the toxin bioassays, the M. fijiensis EaCE, its semi-purified fractions and juglone (Sigma) [one metabolite previously identified by Stierle et al. [5] in M. fijiensis extracts and confirmed in our investigations (data not shown)] were diluted in a 10% methanol solution. They were then injected into the lower surface of the fully expanded second leaf with a micro-syringe at 2 to 4 sites per leaf (20 µl/injection site) and 4 plants per treatment. Before this injection bioassay, banana plants were previously incubated in a humid chamber at 25 ± 2°C with a 16 h light photoperiod during 2 days. Injected plants were incubated at 25 ± 2°C with a 16 h light photoperiod and a relative humidity of 100%. The evaluation of this injection operation was based on the appearance of necrosis at the injection site.

2.6. Electron microscopy

EaCE-treated leaves of the cultivars Grande Naine and Fougamou were cut into 1-mm² pieces after increasing incubation times (1, 2, 4 and 6 h) and were immersed in gluteraldehyde (3%, v/v in 0.1 M sodium cacodylate buffer pH 7.3) for 2 h 30 min at room temperature, rinsed with the same sodium cacodylate buffer and post fixed with osmium tetroxide (1%, w/v in the same buffer) for 2h at 20°C. The leaf pieces were then dehydrated in a series of ethanol solutions starting at 30% and increasing in 10% steps to 100%. The tissue pieces were embedded in epoxy, and ultrathin sections (50-80 nm) were cut with an ultratome III IKB. One part of the sections was contrasted with uranyl acetate and lead citrate while the other ones were stained with the PATAG tests in order to visualise polysaccharides. Specimen were examined with a Philips EM301 transmission electron microscope operating at 60 kV.

2.7. Bioassays on banana chloroplasts

A protocol to isolate physiologically intact banana chloroplasts was developed (unpublished results) on the base of differential centrifugations. The so obtained chloroplast suspension was adjusted at a final concentration of 14×10⁶ chloroplasts ml^-1 before mixing with either the M. fijiensis purified metabolite (juglone) or with 2 semi-purified fractions of the EaCE. Assessment of the possible impact of these fungal extracts was then performed by measuring the electron exchange ability of the isolated chloroplasts through the Hill reaction [9] in which DCPIP (dichlorophenoindolphenol) was used as the electron acceptor. By comparing the quantity of reduced DCPIP (absorbance data at 595 nm) in the presence of fungal metabolites with that of the control (chloroplasts mixed with a 10% methanol solution), the inhibitory effect was determined with the formula 100-[100×(d_m/d_c)] where d_m represents the absorbance decrease of DCPIP in suspension mixed with the fungal metabolites and d_c the absorbance decrease of the control suspension (mixed with only 10% methanol solution).

To investigate the specificity of juglone effect (5-hydroxy-1,4-naphthoquinone) on banana chloroplasts, a comparative analysis of its effect and the one of its isomer, the lawsone (2-hydroxy-1,4-naphthoquinone), was performed with chloroplasts isolated from the susceptible Grande Naine cultivar.

3. RESULTS AND DISCUSSION

3.1. Inoculation

Whatever the strain considered, the artificial inoculation in the greenhouse gave rise to typical symptoms of the black Sigatoka disease. The first period following the inoculation was characterised by spore germination, growth of the germinating tubes and penetration through stomata occurring 7 days after inoculation in the cultivar Grande Naine. The first macroscopic lesions (greenish yellow points) in the cultivar Grande Naine appeared about 20 days after inoculation, while dark brown foliar necrosis was observed around 30 days after inoculation. However, the cultivar Fougamou exhibited a slower lesion development resulting in dark brown spots 40 days after inoculation. For each strain inoculated, the cultivar Grande Naine exhibited the highest susceptibility in terms of rapidity of symptom development and severity of the disease (Table 1).

Table 1 Reactions of Grande Naine and Fougamou bananas 30 days after inoculation with 4 isolates of M. fijiensis

S = susceptibility
PR = partial resistance

Based on their difference in reaction to M. fijiensis infection, these 2 cultivars might be used as references for further analysis of the action dealing with the pathogen toxins.

3.2. Injection of M. fijiensis extracts into banana leaves

The EaCE prepared from uninoculated medium as well as a 10% methanol solution did not induce any necrosis on banana leaves after injection. However, injection of EaCE prepared from a liquid medium in which M.fijiensis was previously cultured caused necrosis (Table 2).

Table 2 Effects of M.fijiensis EaCE 48 hours after injection into leaves of two reference cultivars

Incubation of the plants inoculated was performed under a 16 h light photoperiod,
+: observation of necrotic lesions at the injection site,
-: no necrosis observed at the injection site.

Necrosis appeared 48 h after the injection of EaCE only for those plants incubated under a 16 h photoperiod while plants maintained in darkness did not show any necrosis after injection.

The cultivar Fougamou appeared to be more tolerant to the pathogen EaCE because a minimum of 1000 ppm was required to induce necrosis unlike with Grande Naine where a concentration of 250 ppm was sufficient. The correlation between sensitivity to juglone injection and susceptibility to fungal infection revealed through assays on cultivars Fougamou and Grande Naine constitutes an important step to understand the involvement of M. fijiensis toxins in Black Sigatoka disease.

3.3. Electron microscopy data

The effect of EaCE injected into the leaves was analyzed by transmission electron microscopy (Figure 1). After 6h incubation, the most striking abnormality was the swelling of the chloroplasts in the susceptible genotype (Grande Naine). The ratio of their length/width in the EaCE-treated tissues reached 46% of that observed with the control injected with 10% methanol. Moreover, a considerable decrease of the starch grana surface was observed in chloroplasts of the susceptible cultivar as revealed by the PATAG stain. This kind of morphological modifications appearing on chloroplasts of Grande Naine as a result of EaCE treatment did not occur in tissues of the partially resistant cultivar (Fougamou). Therefore, chloroplasts could be suspected to constitute target sites of M. fijiensis toxins in banana tissues.

Figure 1 Morphological modifications on Grande Naine chloroplasts 6 hours after infiltration of M. fijiensis EaCE into leaf tissues (magnification of the picture = 550 times)

starch grain: ®
chloroplast: ¯

Picture A: Injection with a 10% methanol solution,
Picture B: Injection with the M. fijiensis EaCE (50 ppm).

3.4. Effect of juglone on the electron exchange ability of banana chloroplasts

The Hill reaction was performed with banana chloroplasts isolated from Grande Naine leaves and mixed with juglone. The intensity of that reaction measured by a decrease of the absorbance showed an inhibiting effect of juglone on the ability of chloroplasts to transfer electrons to DCPIP (Figure 2).

Figure 2 Optical absorbance evolution at a wavelength of 595 nm in Grande naine chloroplast suspension mixed with DCPIP as electron acceptor

The absorbance decrease was significantly more important in the control than that observed in the chloroplast suspension treated with juglone. The absorbance decrease in the presence of juglone (0.263) represents only 39% of the absorbance decrease recorded with the control suspension (0.682) mixed with only a 10% methanol solution. That inhibiting effect of juglone on the quantity of reduced DCPIP constitutes the first observation of a direct effect of a M. fijiensis metabolite on banana chloroplasts.

3.5. Specificity of juglone for banana chloroplasts

The results of comparison of juglone and lawsone effect on Grande Naine chloroplasts are presented in the Figure 3 in terms of inhibitory effect on the chloroplast functioning ability.

Figure 3 Inhibiting effect of juglone and lawsone (60 ppm) on the physiological activity of Grande Naine chloroplasts as measured by the Hill reaction during 20 minutes

The inhibiting effect on Grande Naine isolated chloroplasts was very significantly higher with juglone than with lawsone. Although lawsone has the same crude chemical composition as juglone, these two molecules do not exhibit similar effects on the activity of banana chloroplasts. The small difference between the two molecules concerns only the position of one hydroxyl group and that seems to play a crucial role in the toxicity on chloroplasts. Hence, juglone can be considered to have a specific biological activity on banana chloroplasts.

3.6. Toxicity of semi-purified fractions

Fractionation of the M. fijiensis EaCE resulted in eight semi-purified fractions which were analysed for their biological activity on banana tissues after injection (Table 3).

Table 3 Toxicity of the semi-purified fractions A-L of the M. fijiensis EaCE (strain 049 HND) established 48 hours after injection into banana leaf tissues

¹ Fractions are characterised by their relative migration (Rf) in comparison to that of the solvent,

+ = induction of necrosis,
- = no necrosis.

Based on the results of injection bioassays with these fractions, four of them were found to be toxic. To study the activity of these EaCE semi-purified fractions on banana chloroplasts, the two most toxic fractions according to the chlorophyll fluorescence measurements (results not shown) were tested on Grande Naine isolated chloroplasts. Figure 4 illustrates the evolution of their inhibiting effect on the chloroplast activity according to data of the Hill reaction.

For each fraction tested, the inhibiting effect of the extracts on the electron transfer capacity of chloroplasts increased with their concentration. Moreover, a correlation with the results obtained with the injection bioassay was clearly demonstrated: fraction A, which was more toxic than the fraction L in the injection bioassay, showed the highest inhibiting effect in the chloroplast bioassay whatever the concentration considered. Most interestingly, juglone was found to be present in this most toxic fraction whatever the strain analysed (results not shown). Based on the similar global properties between EaCE semi-purified fractions and juglone on banana chloroplasts, investigations dealing with the involvement of toxins in the black Sigatoka disease were pursued with juglone.

Figure 4 Inhibiting effect of M. fijiensis EaCE (strain 049 HND) semi-purified fractions on the physiological activity of isolated chloroplasts of the cultivar Grande Naine

3.7. Comparative bioassay on chloroplasts of 2 reference cultivars

Suspensions of chloroplasts isolated from the 2 reference cultivars (Grande Naine as susceptible and Fougamou as partially resistant) were tested with juglone before comparing the impact of this fungal metabolite on the electron exchange ability. The inhibiting effect of juglone on the chloroplast function is illustrated in Figure 5.

Figure 5 Comparative illustration of the inhibiting effect of juglone on the physiological intactness of chloroplasts isolated from 2 reference cultivars 20 minutes after addition of the metabolite

Addition of juglone to a chloroplast suspension was followed by a decrease of the quantity of reduced DCPIP during 20 minutes whatever the cultivar considered. Inhibition by juglone was greater on the Grande Naine (susceptible cultivar) than on Fougamou chloroplasts. This observation is perfectly correlated with the susceptibility of these 2 banana genotypes to the pathogen infection as well as to their sensitivity to EaCE injection into leaf tissues.

4. DISCUSSION

Artificial inoculation bioassays performed on plants of the cultivars Grande Naine and Fougamou resulted in the development of BS specific lesions. The respective behaviour of these 2 banana genotypes to the pathogen infection confirmed previous observations [11]: whatever the M. fijiensis strain inoculated, the cultivar Grande Naine appeared susceptible and the cultivar Fougamou partially resistant. Hence, this group of 2 genotypes exhibiting differences in their respective behaviour to M. fijiensis infection are useful to study the possible roles of toxins in the development of the BS disease.

Bioassays to assess the toxicity of metabolites found in M. fijiensis culture filtrates were developed. The induction of necrosis after the injection of EaCE into the leaf tissues is easy to perform, but requires relatively high quantities of EaCE (at least 250 ppm) to initiate necrosis on banana leaves. Despite this limitation, the injection bioassay of EaCE gave rise to results similar to those of inoculations with the pathogen. This infiltration bioassay revealed a pronounced light-dependency of the toxicity of M. fijiensis extracts. From observations by electron microscopy a hypothesis was formulated about the sites of action of the M. fijiensis toxins. Indeed, chloroplast swelling represented a first reaction observed in toxin-treated leaves of the susceptible cultivar (Grande Naine) while chloroplasts of the partially resistant cultivar (Fougamou) remained unchanged. These microscopy data and the strict light-dependency of the toxicity due to M. fijiensis metabolites seem to indicate that the photosynthetic apparatus constitutes a potential target site for these toxins.

Purification of the EaCE revealed the presence of different fractions of which some exhibit similar properties as the crude extracts. Moreover, juglone, a purified metabolite previously described to be present in extracts of M. fijiensis culture filtrates [5] was identified in extracts of all the strains analysed and its injection into banana leaf tissues generated results similar to those of EaCE (ranking of the cultivars, light-dependency). Assessment of the eventual impact of M. fijiensis toxins on banana chloroplasts was based on an innovative bioassay performed on isolated chloroplasts. A protocol for mechanical isolation of chloroplasts [9] was adapted to banana leaves. Chloroplasts isolated with this protocol were able to perform the Hill reaction which indicates the ability to ensure electron transport from water to any electron acceptor [10]. Addition of juglone to banana chloroplast suspension inhibited the Hill reaction. This inhibiting effect induced by juglone was not observed for its isomer (lawsone) which differs from juglone only by the position a OH group in its chemical structure. The bioactive semi-purified fraction of EaCE showed similar global properties as juglone on banana chloroplasts. Since electron microscopy observations and chloroplast bioassays indicated a possible action of M. fijiensis extracts on banana chloroplasts, we compared the effects induced by juglone on isolated chloroplasts of the two cultivars. Chloroplasts of the cultivar Fougamou appeared to be less affected by juglone than those of Grande Naine, which correlates to the susceptibility of the two reference cultivars to pathogen infection. This indicates that the photosynthetic apparatus of banana is a potential site of action for juglone as well as the bio-active fractions of EaCE. This apparent correlation between juglone toxicity at the chloroplast level and BS susceptibility of banana cultivars is conforted by the fact that fungal naphthoquinones lead to chloroplast alteration like disruption of the organelle's membrane resulting in the release of nucleotides, amino acids, chlorophyll, proteins and mineral salts [13]. A deeper knowledge of mode of action of M. fijiensis toxins on chloroplasts would clearly help to further understand the mechanisms involved in resistance to these molecules.

REFERENCES

[1] CARLIER, J., et al., Black Leaf Streak. In: JONES, D.R. (Ed.). Diseases of banana, abaca and enset, CABI publishing, Wallingford, UK (2000) 37-79.

[2] MOLINA, G.C., KRAUSZ, J.P., A Phytotoxic Activity in Extracts of Broth Cultures of Mycosphaerella fijiensis var. difformis and its Use to Evaluate Host Resistance to Black Sigatoka, Plant Disease 73 (1989) 142-144.

[3] CARLIER, J., et al., DNA Restriction Fragment Length Polymorphisms in Mycosphaerella Species That Cause Banana Leaf Spot Diseases, Phytopathology 84 (1994) 751-756.

[4] UPADHYAY, R.K., et al., Some toxins of Mycosphaerella fijiensis. In: FULLERTON, R.A., STOVER, R.H. (Eds.). Sigatoka leaf spot diseases of bananas, INIBAP, France (1990) 231-236.

[5] STIERLE, A.A., et al., The phytotoxins of Mycosphaerella fijiensis, the causative agent of Black Sigatoka disease of bananas and plantains, Experientia 47 (1991) 853-859.

[6] HARELIMANA, G., et al., Use of Mycosphaerella fijiensis toxins for the selection of banana cultivars resistant to Black Leaf Streak, Euphytica 96 (1997) 125-128.

[7] LICHTENTHALER, H.K., et al., Application of chlorophyll fluorescence in ecophysiology, Radiation and Environmental Biophysics 25 (1986) 297-308.

[8] KETEL, D.H., et al., Prediction of a low dose herbicide effect from studies of binding of metribuzin to the chloroplasts of Chenopodium album L., Ann. Appl. Biol., 128 (1996) 519-531.

[9] LEEGOOD, R.C., MALKIN, R., Isolation of sub-cellular Photosynthetic Systems, In: HIPKINS, M.F., BAKER, N.R. (Eds.). Photosynthesis energy transduction, A practical approach, IRL PRESS, Washington, (1986) 9-26.

[10] ALLEN, J.F., HOLMES, N.G., Electron transport and Redox Titration, In: HIPKINS, M.F., BAKER, N.R. (Eds.). Photosynthesis energy transduction, A practical approach, IRL PRESS, Washington, (1986) 103-141.

[11] MOURICHON, X., et al., Inoculation expérimentale de M. fijiensis Morelet sur jeunes plantules de bananier issues de cultures in vitro, Fruits 42 (1987) 195-198.

[12] UPADHYAY, R.K., et al., Fijiensin, the first phytotoxin from Mycosphaerella fijiensis, the causative agent of Black Sigatoka disease, Experientia 46 (1990) 982-984.

[13] MEDENTSEV, A.G., AKIMENKO, V.K. Naphtoquinone metabolites of the fungi, Phytochemistry 47 (1998) 935-959.