Abstract

Populations of Leptosphaeria maculans, responsible for Blackleg disease of oilseed rape, are divided into two genetically distinct main populations. A-group isolates mainly induce stem canker even though B-group isolates induce weak symptoms on stem. It was previously established that Phospho-glucose isomerase (PGI) patterns allow to distinguish these two distinct populations from pure culture mycelium. We showed that this method is also efficient to distinguish A from B isolates directly from leaf lesions and from pseudothecia. The PGI patterns obtained from leaf lesion extracts of different varieties corresponded to those obtained from extracts of isolates recovered from the same leaf lesion on malt agar. The different PGI patterns allowed to link qualitative symptoms on leaves with each group of isolates. Occasionally, the two groups of isolates were present in a single leaf lesion. In other respects, the symptoms due to Pseudocercosporella capsellae are sometimes very close to those developed by B-group isolates but we showed that P. capsellae can be distinguished from all L. maculans isolates by a specific PGI pattern.

So, this rapid and efficient method affords the possibility to analyze large samples of leaf lesions and pseudothecia and opens interesting prospects for studying the epidemiology of the disease.

Introduction

Leptosphaeria maculans (anamorph Phoma lingam) is responsible for important yield losses in oilseed rape (Brassica napus) worldwide (Gugel & Petrie, 1992). Most cultivars are susceptible to leaf infections under field conditions, but some of them are moderately resistant to stem canker. Current research strategies to improve B. napus resistance and to develop efficient chemical control strategies require a good understanding of pathogen diversity.

Investigations on the variability of L. maculans populations showed the co-existence in field of two genetically distinct sub-populations, designated A- and B- groups by Johnson & Lewis (1990), and which may represent separate species (Williams, 1992). With axenic cultures of A- and B-group isolates of L. maculans, Sippel & Hall (1995) distinguished unequivocally the two L. maculans groups by phospho-gluco-isomerase (PGI) patterns. All A-group isolates were characterized by one electrophoretic type (ET1), and all B-group isolates by another (ET2). We showed that another allozyme, ET3 more rapid than ET1, was characteristic of a few A-group isolates (Brun et al., 1997).

Our study investigates the potential for using PGI analysis directly on leaf lesions and pseudothecia from stubbles to determine the relative frequency of each L. maculans group and the relationships between the isolates obtained from leaf lesions and pseudothecia observed at the autumn according to the oilseed rape variety used. Additionally, stubbles of B. juncea resistant variety were analyzed.

Material and method

Sampling procedure and description of leaf lesions

Leaves bearing typical and atypical L. maculans leaf lesions were sampled during winter from four oilseed rape cultivars in the field. Fresh leaves were either processed immediately or stored individually at -20°C for later studies. Two of these (cv. Samouraï from INRA-SERASEM, susceptible to crown canker; cvs. Darmor from INRA-SERASEM) were double low cultivars and one, cv. Cresor from SERASEM, characterized by a good level of resistance under Canadian conditions, was a spring single low cultivar. Stubbles bearing pseudothecia were sampled at the next autumn from cvs. Samouraï, Darmor, Crésor. Brassica juncea cv. Picra for which many single ascospore isolates were previously obtained from different plants was also assessed for direct PGI analysis.

Before PGI analysis, each lesion was assessed for chlorosis. Binocular microscopic observations (50x magnification) were used to check for the presence of pycnidia. Typical Pseudocercosporella capsellae leaf lesions were confirmed by microscopic identification of the characteristic hyaline conidia (average size of spores : 80µ x 2.5µ).

Single pseudothecia were sampled separately from stem basis residue and gently crushed between slides. Their maturity condition (empty, ascus differentiated, ascospore differentiated, ascus and ascospore damaged) was assessed under microscope just before adding extraction buffer.

Single ascospore isolates were obtained from cv. Picra (B. juncea) according to the method described by Somda et al., 1996.

Starch gel electrophoresis of phospho gluco isomerase allozymes

Protein extracts were prepared either directly from leaf lesions, pseudothecia or from fungal cultures. Direct extraction from a single lesion was performed either on necrotic tissue only, or on necrotic and surrounding living tissue. Samples consisted of small tissue pieces (ca 25 mm²), cut with a scalpel. Similar samples, cut from healthy parts of the leaves, served as controls. A small piece (about 2 mm2 ) of each leaf lesion was identified and stored so that the pathogen could be isolated as described below. Extracts from axenic fungal cultures were prepared from mycelium scraped from 1-2 cm² of the surface of colonies actively growing on malt agar. A mixture of three isolates characterized as belonging to A- (isolates 940 and P22c) and B- (isolate 945) groups on the basis of sirodesmin PL production and pigment diffusion through malt agar medium were used as controls in all electrophoretic gels.

Samples of lea f lesions and single pseudothecium were crushed in 100 μl and 3.5 μl of Tris HCl 0.1 M buffer (pH 7.5) containing 1 % reduced glutathione, respectively. PGI allozymes were separated by horizontal starch gel electrophoresis with the histidine / tris citrate pH 7.0 gel - electrode buffer and stained by the method described by Chèvre et al. (1995). Allozymes were characterized by Rf values, calculated as the migration distance from the cathodal origin relative to that of the most anodal Pgi-2 allozyme of cv Samouraï.

Results

Relationships between characteristics of leaf lesions and PGI allozymes.

The fastest allozyme ET4 was specific to P. capsellae. The majority (130) of typical leaf lesions gave an ET1, two gave an ET3 and five gave a mixture of ETs patterns (Table 1). Atypical lesions usually produced an ET2 (58 of 77 lesions), 17 gave an ET1, one gave an ET3 and one gave a mixture of ET1 and ET2 patterns.

Relationships between analysis of single ascospore isolates and direct analysis of pseudothecia obtained from the same plants of B. juncea cv. Picra .

Two plants (1 and 7) beard only ETI isolates following analysis of either single ascospore isolates (39 and 19) or direct electrophoresis of pseudothecia (3 and 9)(Table 2).

The two other plants (3 and 4) beard only ET2 isolates whatever the analysis of single ascospore isolates or pseudothecia.

Table 1. Relationships between characteristics of leaf lesions expected to be due to L. maculans and of P. capsellae leaf lesions recovered from different oilseed rape cultivars and PGI electrophoretic types (ET).

		PGI allozymes
Characteristics of symptoms	Number of leaf lesions analyzed	ET1 (A-group isolates)	ET2 (B-group isolates)	ET3 (A-group isolates)	ET4 (Pseudoc.)	ET1 + ET2	ET1 + ET3	ET1 + ET4
Tp	130	123	0	2	0	1	2	2
At	8	2	5	1	0	0	0	0
AtC	28	3	25	0	0	0	0	0
Atp	3	3	0	0	0	0	0	0
AtpC	38	9	28	0	0	1	0	0
∑	207	140	58	3	0	2	2	2
Pseudocerco..	22	0	0	0	21	0	0	1

Tp : Typical leaf lesions with pycnidia

At : Atypical leaf lesions, without pycnidia, without chlorosis

AtC : Atypical leaf lesion, without pycnidia, with chlorosis

Atp : Atypical leaf lesion, with pycnidia, without chlorosis

AtpC : Atypical leaf lesion, with pycnidia, with chlorosis

Pseudocerco. : Typical Pseudocercosporella capsellae leaf lesions with characteristic spores

(1) : mean and standard error.

Table 2. Frequency of the different PGI electrophoretic types (ET) recovered from the pseudothecia and from single ascospore isolates obtained from the same plants of Brassica juncea cv. Picra.

Number of the plant analyzed	PGI Allozymes
	ET1		ET2		ET1 + ET3		ET3
	Pseudo.	SAI	Pseudo.	SAI	Pseudo.	SAI	Pseudo.	SAI
1	3	39	0	5	0	0	0	0
7	9	19	0	0	0	0	0	0
3	0	0	0	0	0	0	21	66
4	0	0	0	0	0	0	18	28

Pseudo. : pseudothecia

SAI : single ascospore isolate

Relationships between isolates obtained from leaf lesions and from pseudothecia on the same variety.

Mainly A-group isolates were detected from leaf lesions (92.9; 90.4; 90.2 % for cvs. Darmor, Crésor and Samouraï, respectively). Among A-group the ET1 allozyme was predominant (98.4%) (Table 3). Some ET2 were recovered from all cultivars in low proportion : between 2.4% and 7.4% for cvs. Crésor and Samouraï, respectively.

Only A-group isolates were detected from pseudothecia on the three oilseed rape cultivars. ET1, ET3 and mixture of ET1 + ET3 allozymes were present. ET1 allozyme alone was predominant (72.2%). Some single pseudothecia exhibited the two allozymes ET1 and ET3 (7.9%).

Table 3. Frequency of the different PGI electrophoretic types (ET) recovered from leaf lesions and from the pseudothecia of different cultivars of oilseed rape.

	Number of leaf lesions and pseudothecia analyzed		PGI Allozymes
			ET1	ET3	ET1 + ET3	ET2
‘Darmor’	Leaf lesions	59	53	0	2	4
	Pseudothecia	102 (8)1	65	0	14	0
‘Crésor’	Leaf lesions	42	38	2	1	1
	Pseudothecia	170 (9)	118	3	10	0
‘Samouraï’	Leaf lesions	215	194	5	2	16
	Pseudothecia	229 (10)	179	0	16	0

¹( ): Number of stubbles analyzed

Discussion

Typical leaf lesions (greyish with very distinct pycnidia) were due to A-group isolates. Atypical leaf lesions, particularly those with chlorosis and very small pycnidia, were mainly associated with B-group isolates but some A-group isolates were also detected in that type of symptoms. The low frequency of B-group isolates in leaf lesions could mean that these isolates were not frequent in L. maculans populations or that the cultivars tested were not very susceptible to that kind of isolates.

Electrophoresis of PGI performed directly on pseudothecia allowed to distinguish A and B-group isolates on cv. Picra. All ETs (ET1, ET2, ET3) were obtained from pseudothecia. There was a good correspondence between electrophoretic analyses of single ascospore isolates and pseudothecia recovered from single plants of cv. Picra.

The possibility of crosses between ET1 and ET3 isolates was demonstrated by the fact that both types were recovered from a single pseudothecia from B. napus genotypes but any B-group pseudothecia were detected. This result could be due to the size of the sample or to the low frequency of these isolates in leaf lesions of oilseed rape cultivars.

This method, rapid and reliable, could be of interest to analyze directly large samples of leaf lesions and pseudothecia and to increase knowledge on epidemiology of the disease.

Acknoledgement

This work was supported by a grant from CETIOM (France).

References

1. Brun H., Levivier S., Eber F., Renard M., Chèvre A.M., 1997. Electrophoretic analysis of natural populations of Leptosphaeria maculans directly from leaf lesions. Plant Pathology 46, 147-154.

2. Chèvre AM, Delourme R, Eber F, Margale E, Quiros CF, Arus P, 1995. Genetic analysis and nomenclature for seven systems in Brassica nigra, B. oleracea and B. campestris. Plant Breeding 114, 473-572.

3. Gugel RK, Petrie GA, 1992. History, occurrence, impact, and control of blackleg of rapeseed. Canadian Journal of Plant Pathology 14, 36-45.

4. Johnson RD, Lewis BG, 1990. DNA polymorphism in Leptosphaeria maculans. Physiological and Molecular Plant Pathology 37, 417-424.

5. Sippel DW, Hall R, 1995. Glucose phosphate isomerase polymorphisms distinguish weakly virulent from highly virulent strains of Leptosphaeria maculans. Canadian Journal of Plant Pathology 17, 1-6.

6. Somda I; Renard M., Brun H. 1996. Morphology, pathogenicity and isozyme variation among French isolates of Leptosphaeria maculans recovered from Brassica juncea cv. Picra. Plant Pathology 45, 1090-1098.

7. Williams PH, 1992. Biology of Leptosphaeria maculans. Canadian Journal of Plant Pathology 14, 30-35.