Occurrence of a lipase in spores of Alternaria brassicicola with a crucial role in the infection of cauliflower leaves

Abstract

Alternaria brassicicola is a pathogen that penetrates directly through the host cuticle thanks to several serine esterases, according to our findings. Among these, an 80-kDa lipase (E.C 3.1.1.3) was detected by SDS-PAGE and immunoblotting in the water washings of ungerminated spores. The purified lipase cross-reacted with Botrytis cinerea anti-lipase antibodies, which were reported to inhibit the in vitro lipase activity. Anti-lipase antibodies were added to a conidial suspension of A. brassicicola prior to inoculation. As a result, blackspot lesions were reduced by 90% on intact cauliflower leaves, but not on leaves from which surface wax had been removed. Spore surface-bound lipase is thought to interact closely with epicuticular leaf waxes for adhesion and/or penetration of the fungal propagules during the early stages of host-parasite interactions.

Spore, Lipase, Leaf infection, Anti-lipase antibody, Alternaria brassicicola

1 Introduction

Many phytopathogenic fungi invade the aerial parts of plants by penetrating directly through the host cuticle, a combination of cutin (a hydroxy fatty acid polyester) and waxes. Adhesion of fungal spores to the host surface and the subsequent cutin penetration of fungal germlings were observed as the result of enzymatic modification of the host cuticle by hydrolytic enzymes such as cutinases or esterases [1].

The controversial role of cutinases in the penetration process has been reviewed [2,3]. In addition to cutinases, other serine-esterases have been observed in various fungi [4,5] and a lipase synthesized during the early phases of pathogenic growth of Botrytis cinerea was also found to play a key role in the infection of tomato leaves [6]. In order to determine the role of lipases in penetration of the host cuticle, another fungal pathogen Alternaria brassicicola, also known to be a lipase producer [7], was investigated.

The fungus A. brassicicola, causal agent of blackspot in several cruciferous vegetables, is a directly-penetrating pathogen of Brassica spp. Cutinolytic esterases of A. brassicicola have been intensively studied [5,8], but the role of the 80-kDa lipase excreted by A. brassicicola has not yet been elucidated [7].

In this paper, the 80-kDa lipase was shown to be present in the water washings from ungerminated spores of A. brassicicola. Subsequently, the effects of anti-lipase antibodies were assessed on the penetration of cauliflower leaves by fungal spore germlings.

2 Materials and methods

2.1 Fungus and spore collection

The A. brassicicola (Schw.) isolate (CBS 103-24) was grown on V8 juice agar slants at 24°C under a 16-h fluorescent daylight photoperiod to induce sporulation. Conidial suspensions were prepared by washing the surface of 16-day-old cultures with sterile water, filtering it through glass wool and adjusting the concentration according to desired use.

2.2 Enzyme production and purification

The spore surface bound proteins (SSBP) were included in the supernatant of a 20-ml aqueous conidial suspension (1×10⁷ spores ml⁻¹) which had previously been vortexed vigorously for 5 min and centrifuged at 30 000×g for 20 min. The 80-kDa lipase was produced and purified from culture fluids of A. brassicicola according to Berto et al. (A. brassicicola was grown with methyl oleate) [7]. The 21-kDa and 23-kDa cutinases identified as saprophytic cutinases were previously extracted from the culture broth of A. brassicicola and purified as described by Trail and Köller (A. brassicicola was grown on apple cutin as sole carbon source) [9].

2.3 Enzyme assays

Lipase and cutinase assays were performed according to published protocols [6]. Briefly, lipase activity of A. brassicicola enzymes was measured by spectrofluorimetry using umbelliferyl oleate as substrate [10]. Specific enzyme activity was defined as µmol of umbelliferone (UMB) released per min and per mg of protein at 25°C. When tested, phenylmethylsulfonylfluoride (PMSF) was added (1 mM) to the enzymatic mixture. For all assays, heated enzymatic extracts were considered as controls (100°C, 5 min). Cutinase activity was determined using tritiated apple cutin prepared according to Bonnen and Hammerschmidt [11]. Enzymes were assayed in equal amounts of proteins (50 µg) both with and without 1.5% (w/v) β-cyclodextrin. Radioactivity was measured by liquid scintillation counting (Beckman LS 2800 spectrometer). Cutinolytic activity, i.e. the degradation of labelled [³H]cutin, was expressed in dpm×1000 after a 16-h incubation at 25 °C. Experimental values were calculated after controls were subtracted, including reaction mixtures with denatured heated enzymes. Protein concentration was determined according to Bradford [12] using the Bio-Rad protein assay kit with bovine serum albumin as a standard.

2.4 Gel electrophoresis

SDS-PAGE was performed according to Laemmli [13], using a 4.5% (w/v) stacking gel and a 12% (w/v) running gel in a Mini-Protean II electrophoresis cell (Bio-Rad). The run was carried out at 200 V constant voltage. Proteins were detected in gels by silver staining [14] and their molecular masses were determined by referring to the mobility of known molecular mass standards (Sigma).

2.5 Anti-lipase antibody assays

Purified polyclonal antibodies against the lipase of B. cinerea[6] were the basis for all immunological assays described below. Immunoblotting procedure was achieved according to Comménil et al. [6]. IgG content was determined using Bovine IgG (Bio-Rad) as a reference and immunodetection was performed with the Bio-Rad alkaline phosphatase conjugated antibody kit.

To study the effect of anti-lipase antibodies on the in vitro esterase activity of the SSBP sample, the purified lipase or cutinases was assayed with different concentrations of anti-lipase antibodies incubated with each enzymatic solution (20 µg protein) in 50 mM Tris-maleate buffer at pH 7.2 for 30 min. Esterase activity (UMB) was measured by spectrofluorometry and expressed as percent of control value (100%) recorded in experiments with no antibodies.

A detached leaf assay was used to test the in vivo effects of anti-lipase antibodies on infection of cauliflower leaves by A. brassicicola spores. Cauliflower plants (Brassica oleracea cv. Whiterock) were cultivated in a greenhouse at 20°C for 4 weeks. The second and third excised leaves were divided into two lots of intact and chloroform-treated leaves. Epicuticular wax was removed from treated leaves by wiping their upper surface with chloroform-soaked cotton-wool [15]. Leaves were placed with their adaxial surface up in Petri-dishes containing 0.6% (w/v) water agar. For each lot of leaves, six experimental sets were designed according to the type of leaf inoculum. Inocula included: spores only, sterilized spores, a mixture of spores and inactivated antibodies (5 min at 100°C), spores to which were added two concentrations of anti-lipase antibody solutions (192 and 384 µg ml⁻¹), and spores with 1 mM of PMSF. Spores were obtained from an aqueous conidial suspension of A. brassicicola (4×10⁶ spores ml⁻¹). Leaves were drop-inoculated with nine droplets (10 µl) per leaf and incubated at 24°C for 4 days, with a 16-h photoperiod. Blackspot symptoms were observed on cauliflower leaves in situ with a direct light microscope and recorded as a percentage of leaf symptoms (computation was based on the number of necrotic lesions on leaves expressed 96 h after inoculation). Results were obtained from three separate experiments comprising three leaves per set and per experiment.

3 Results

3.1 Electrophoretic characteristics of SSBP fraction associated with esterase activity

Esterase activity (Table 1) was detected in spore surface-bound proteins (SSBP) recovered from the ungerminated A. brassicicola spores. A high esterase activity (4,6 µmol UMB min⁻¹ mg⁻¹) was measured spectrofluorometrically in the SSBP fraction. When PMSF (1 mM) was added to the spore suspension, esterase activity of the SSBP extract was reduced by an impressive 74%.

1

Esterase activity (UMB) in A. brassicicola ungerminated spores

Sample	Proteins (mg)	Esterase activity (µmol UMB min−1)	Specific activity (µmol UMB min−1 mg−1)
SSBP	0.21 (±0.04)	0.98 (±0.12)	4.6
SSBP+PMSF (1 mM)	0.21 (±0.04)	0.26 (±0.08)	1.2

SSBP were in the supernatant of 20 ml centrifuged aqueous spore suspension (1×10⁷ spores ml⁻¹).

Mean of six replicates.

We investigated the types of esterase in the SSBP extract by separating proteins by SDS-PAGE (Fig. 1). We then compared them with the purified 80-kDa A. brassicicola lipase [7] and two partially purified 21–23 kDa saprophytic cutinases [4]. An 80-kDa band was observed by silver staining in the SSBP extract of ungerminated spores. The esterase activity of the 80-kDa protein in SSBP was revealed in the in situ gel by means of activity staining with α- and β-naphthyl acetate (data not shown). Another protein band of about 66 kDa also appeared in the SSBP extract after silver- and esterase-activity staining.

1

SDS-PAGE of SSBP in A. brassicicola ungerminated spores with silver staining. Lanes 1 and 3: Molecular mass standards. Lane 2: Saprophytic cutinases. Lane 4: Purified lipase. Lane 5: SSBP.

SDS-PAGE and immunoblotting (Fig. 2) showed that the polyclonal antibodies raised against the B. cinerea lipase [6] cross-reacted with the 80-kDa lipase purified from A. brassicicola. In addition, an immunoreactive band of 80 kDa and a cross-reactive band of about 66 kDa were observed in the SSBP spore extract. These results showed that the 80-kDa lipase was present in the water washings of ungerminated spores of A. brassicicola. No immunological cross-reactivity was detected with the two 21–23-kDa cutinases excreted by A. brassicicola grown on apple cutin. Thus it can be affirmed that the lipase and the cutinases from A. brassicicola are serologically distinct and the lipase demonstrates epitopes in common with those of B. cinerea.

2

Immunodetection of lipase in SSBP from A. brassicicola ungerminated spores. Lane 1: Molecular mass standards. Lane 2: Saprophytic cutinases. Lane 3: Purified lipase. Lane 4: SSBP. Immunoblots were performed with purified antibodies raised against lipase from B. cinerea.

3.2 Cutinolytic activity of SSBP extract and lipase

The lipase purified from A. brassicicola was found to have high specificity for the hydrolysis of oleic acid ester [7], a structural constituent of plant cuticle. The capacity of both the purified lipase and the SSBP extract to degrade cutin was studied using tritiated apple cutin as substrate. It was then compared with cutinase-induced degradation (Table 2). Significant cutinolytic activity was detected in the SSBP extract, although the activity of purified lipase measured after 16 h of incubation at 25°C was twice as high. Nevertheless, saprophytic cutinases were found to be about 6–10 times more efficient in hydrolyzing cutin. In enzymology, β-cyclodextrin is known for enhancing the solubility of hydrophobic compounds in water [16]. We thus proceeded to add β-cyclodextrin to the assay mixture. As a result, the cutinolytic activity of A. brassicicola enzymes was increased by three- to five-fold.

2

Cutinolytic activity of SSBP, purified lipase and saprophytic cutinases in A. brassicicola after 16 h incubation with [³H]cutin

Sample (50 µg protein)	Cutinolytic activity (dpm×1000)
	β-Cyclodextrin−	β-Cyclodextrin+
SSBP	17 (±4)	65 (±8)
Purified lipase	30 (±2)	157 (±12)
Saprophytic cutinases	180 (±32)	595 (±48)

Mean of three replicates.

3.3 Effect of anti-lipase antibodies on in vitro esterase activities in A. brassicicola

Data showed that in vitro esterase activity of SSBP and purified lipase from A. brassicicola decreased significantly when Botrytis cinerea anti-lipase antibody solution was added (Table 3). The degree of enzyme inhibition was correlated with the antibody concentration. Ninety six µg ml⁻¹ and 192 µg ml⁻¹ of antibodies reduced lipase activity by respectively 32% and 12% of control esterase activity (100%). Control values were obtained by spectrofluorometric measurements of the enzyme solution both without antibodies and with heat-inactivated antibodies. The SSBP esterase activity was modified to a lesser extent by the antibody solution, as 192 µg ml⁻¹ of antibodies only reduced enzymatic activity by 43%. Anti-lipase antibodies were also tested with the purified saprophytic cutinases, but 192 µg ml⁻¹ antibodies did not inhibit their esterase activity.

3

Effect of different concentrations of anti-lipase antibodies on the in vitro esterase activity (UMB) of SSBP, purified lipase and saprophytic cutinases in A. brassicicola

Anti-lipase antibodies (µg ml−1)	Esterase activity %
	SSBP	Lipase	Cutinases
0	100	100	100
12	100	95 (±2)	100
24	96 (±1)	81 (±4)	100
48	89 (±4)	66 (±7)	100
96	69 (±6)	32 (±6)	100
192	43 (±8)	12 (±4)	100
192	100	100	100

Esterase activity was measured by spectrofluorometry (UMB) and expressed in percent of control value (100%) monitored without antibodies.

Esterase activity was measured by spectrofluorometry (UMB) and expressed in percent of control value (100%) monitored without antibodies.

Enzyme samples: 20 mg protein.

Heat denaturated antibodies.

Mean of four replicates.

3.4 Effect of anti-lipase antibodies on in vivo cauliflower leaf infection

As shown in Table 4, a level of infection near 100% was observed in both lots of cauliflower leaves which developed dark lesions 4 days after inoculation with conidia alone or a mixture of spores and heat-inactivated anti-lipase antibodies. However, after the same 4-day period, no blackspot was observed on leaves incubated with autoclaved spores or antibody solution alone without spores (data not shown). Symptoms were similarly reduced on about 80% of intact or dewaxed leaves when conidia received 1 mM PMSF.

4

Effect of anti-lipase antibodies on the infection of cauliflower leaves by A. brassicicola spores

Inoculum droplet (10 µl)	Leaf symptoms (%)
	Intact leaf	Dewaxed leaf
Sterilized spores	0	0
Spores	98 (±2)	100
Spores+PMSF (1 mM)	14 (±2)	19 (±2)
Spores+antibodies (192 µg ml−1)	35 (±5)	95 (±3)
Spores+antibodies (384 µg ml−1)	12 (±1)	80 (±2)
Spores+heated antibodies	95 (±2)	96 (±1)

Nine inoculum droplets were deposited on the adaxial surface of each leaf.

Conidial suspension (4×10⁶ spores ml⁻¹).

Control: autoclaved spore suspension.

Denaturated antibodies (384 µg ml⁻¹, 5 min, 100°C).

Percentage of necrotic leaf lesions recorded 96 h after inoculation.

Mean of three replicates.

Anti-lipase antibodies inhibited the symptoms of disease on intact leaves, the degree of inhibition being related to the concentration of antibodies applied. In fact, the percentage of leaf necrosis was lowered to 35% when 192 µg ml⁻¹ of antibody solution was added to spore inoculum, while higher antibody concentration (384 µg ml⁻¹) almost completely inhibited infection 96 h after inoculation. On the contrary, anti-lipase antibodies did not prevent fungal infection of dewaxed leaves: they demonstrated high levels of leaf symptoms (80%) despite the spore-antibody treatment.

Microscopic examination of the spore infected leaves showed that conidial germination developed normally in droplets with anti-lipase antibodies, although no symptoms were observed at the site of inoculation.

4 Discussion

In this study, SDS-PAGE and immunoblotting detected an 80-kDa lipase, purified previously [7], in water washings of A. brassicicola ungerminated spores. The 80-kDa lipase was reported to be a serine esterase demonstrating cutinolytic activity (it degraded [³H]apple cutin). The lipase was inhibited by PMSF and structurally bound to the surface of ungerminated conidia. Both its constitutive expression and localization in A. brassicicola suggest that the lipase may have an early pathogenic activity in spore adhesion (as soon as spores land on host leaf). The 80-kDa lipase was part of the mixture of cutinolytic esterases constitutively released from A. brassicicola and well described elsewhere [8], as opposed to a B. cinerea lipase, which was inducible only in spore germlings [6]. The purified 80-kDa lipase was reported to cross-react with polyclonal antibodies raised against a 60-kDa B. cinerea lipase [6], while no immunological cross reactivity was detected with the 21-kDa and 23-kDa cutinases from A. brassicicola. In fact, in vitro esterase activities of A. brassicicola spore washings (SSBP) and purified lipase were reduced by 50% and 90%, respectively, with the addition of anti-lipase antibodies to the assay mixture, unlike esterase activity of cutinases, which was not modified by antibody solutions. Correlatively, the ability of the fungus to cause leaf symptoms decreased drastically (by 88%) when anti-lipase antibodies (384 µg ml⁻¹) were added to the conidial suspension prior to inoculation of intact leaves. Meanwhile, the same antibody solution could not prevent infection in dewaxed cauliflower leaves. These results suggest that the lipase may interact mainly with the surface wax, i.e. the first layer encountered by conidia landing on the host leaves. In B. cinerea, a pathogenic lipase may be induced by components of grape berry waxes [17]. Subsequently, the cutinolytic properties of the lipase in A. brassicicola may provide cutin monomers as putative inducers of early pathogenic cutinases.

Some authors have suggested that several cutinases or esterases may be involved in the penetration process of A. brassicicola conidia [5], but although genes from A. brassicicola have been cloned [5,18], the precise function of cutinases in plant infection remains to be clarified. Reduced symptoms (by 80%) in intact and dewaxed cauliflower leaves after addition of PMSF in spore inoculum support the hypothesis that other serine esterases may be involved in cauliflower cuticle degradation.

In summary, the 80-kDa lipase is one of the pathogenic serine esterases. It is postulated that early interactions between cuticular wax and lipase area prerequisite to adhesion and subsequent penetration. These phenomena are essential to successful infection and occur before germination [19]. Similarly, a 60-kDa lipase was considered to be an essential factor in B. cinerea pathogenicity because of its early presence on the leaf infection site [17]. Moreover, the inhibition of the lipase activity prevented infection of tomato leaf tissues by B. cinerea[6].

The crucial role of lipase can differ with plant-fungal pathogen systems and it remains to be clearly elucidated. Specific antibodies against the 80-kDa lipase and enzyme gene cloning should lead the way to better knowledge of A. brassicicola pathogenicity.

Acknowledgements

The authors wish to thank Dr. L. Huffman-Touzet for improving language usage.

Author notes

1

Present address: Faculté Universitaire des Sciences Agronomiques de Gembloux, Laboratoire de Phytopathologie, B-5030 Gembloux, Belgium.