ABSTRACT

Over the last two years, a large sampling of isolates has been done to assess the French population structure of Leptosphaeria maculans. Isolates are collected in 6 regions and identified by PCR analysis to distinguish the aggressive Tox+ and the non aggressive Tox°. Tox° were characterized in sub-groups NA1, NA2 and NA3 by PCR-RFLP. Pathogenicity tests on cotyledons on a differential set of cultivars are performed in order to discriminate Tox+ isolates into pathogenicity groups PG2, PG3 and PG4.

Both Tox+ and Tox° types are present in France but their distribution depend on their geographical origin. In the south of France, isolates are almost Tox+ whereas isolates Tox° represent more than 60% of the population in the eastern area. In the other locations mixed populations occur with a ratio 1:1. The distribution of the Tox+ and Tox° isolates seems to be dependent on the timing of sampling. Thus, Tox+ are more prevalent in autumn rather than at the harvest time suggesting a competition between the both types during the crop season. Moreover co-infection by both types are noticed within the same plant or within the same leaf or stem. Furthermore all the Tox° isolates belong to NA1 sub group. Among Tox+ isolates, PG3 isolates represent about 80% and PG4 less than 20%. Two PG2 isolates were for the first time detected in 1997.

INTRODUCTION

Blackleg, caused by Leptosphaeria maculans (anamorph Phoma lingam) occurs on winter oilseed rape in most French regions where the crop is grown. The damage and yield loss may be important. In 1994, severe infections of the crown reduced yields by more 50% in some seriouly infested fields of Poitou region (Lagarde, 1995).

L. maculans currently considered as a species complex is divided into aggressive and nonaggressive groups of pathotypes which should be two distinct species (Rouxel et al, 1994). Because of more accurate identification tools, the non aggressive species have been separeted into three subgroups NA1, NA2 and NA3. Based on host interaction phenotypes on a set of differential cultivars, the aggressive species has been differentiated into three pathogenicity groups (PGs) : PG2, PG3 and PG4 (Koch et al, 1991).

A better understanding of the pathogen population structure will be of value in improving breeding for resistance and in developing efficient chemical control (Crespel et al, 1997).

The present paper deals with the results of a survey made in 1997 in collaboration with phytopathologists of INRA to provide information on the occurence and the distribution of the different pathotypes of L. maculans in the major rapegrowing areas of France.

MATERIALS AND METHODS

Sample collection :

Field isolates of L. maculans were sampled in six locations in France : Surgères, (southwestern), St Florent (central), Dijon (eastern), St Pathus (northcentral), Nancy (northeastern) and Béziers (southern), between autumn 1996 and harvest in 1997. Within each location, isolates were collected from four winter oilseed rape cultivars: Bristol (susceptible cv.), Goeland, Synergy (less susceptible cv.) and Vivol (resistant cv). For each one, isolations were made from fifteen plants collected randomly in the plots.

According to the timing of the collect, isolates were obtained from small pieces of leaves, stem and/or crown showing symptoms or apparently healthy. After superficial desinfection, they were plated on Malt agar medium and incubated at 20°C. Ten days later, isolates were purified on Malt agar plates to confirm on the basis of cultural characterics the occurence of L. maculans and on V8 juice agar medium to obtain high sporulating cultures.

Characterization of isolates :

Identification of isolates was performed directly on intact conidia. It was based on ITS amplification to differentiate aggressive and nonaggressive isolates, respectively termed Tox+ and Tox0 . It was followed by a digestion of PCR product with selected restriction enzymes (RFLP) according to the methods described previously (Balesdent et al, 1998).

Pathogenicity tests :

Pathogenicity tests on cotyledons on a set of 7 cultivars (Westar, Glacier, Quinta, Jet Neuf, Doublol, Capitol and Columbus) were performed in order to discriminate the aggressive isolates Tox+ into pathogenicity sub-groups PGs.

RESULTS

Occurence of Tox+ and Tox0 :

In 1996-97, 726 isolates were purified and identified. 57% of the isolates were Tox+, the remaining Tox0 were NA1 isolates (Tab I).

Groups	No. of isolates	% of Toxn isolates
Tox+	411	57 %
Tox0	315	43 %
Total	726

Table I : Characterization of the French isolates collected in 1997.

There were geographical differences in the occurence of the pathotypes with almost Tox+ type in the south of France (Béziers). In Saint-Florent Tox+ isolates were predominant whereas Tox0 isolates were more prevalent in Nancy. In the other locations, both Tox+ and Tox0 types were mixed in a ratio 1:1.

Locations	% of Tox+ isolates	% of Tox0 isolates
17-Surgères	58	42
18-Saint Florent	61	39
21-Dijon	51	49
34-Béziers	95	5
54-Nancy	38	62
77-Saint Pathus	49	51

Table II : Characterization of Tox+ and Tox0 isolates of L. maculans in different locations in France in 1997.

Whatever the cultivar, the Tox+ isolates were more frequent in Autumn whereas Tox0 isolates were more prevalent at the harvest time (Tab III). Morever, the frequency of Tox+ isolates seemed to be in agreement with the susceptibility of cultivars : Tox+ isolates from the most susceptible cv. Bristol represented 45 % but they were less 20% on the most resistant cv. Vivol .

	Autumn		Harvest period
Cultivar	Tox+	Tox0	Tox+	Tox0
Bristol	90	10	45	55
Goeland	95	5	25	75
Synergy	97	3	26	74
Vivol	84	16	17	83

Table III : Characterization of Tox+ and Tox0 isolates of L. maculans according to the cultivars and the date of isolation.

In some cases, both types Tox+ and Tox0 were identified within the same plant, and even within the same lesion (Tab IV).

Isolates	Group	Subgroup
within the same plant :
C21A10	Tox+	PG3
C21A11	Tox+	-
C21A13	Tox0	NA1
within the same plant and the same organ (Synergy / collar)
C17A103	Tox+	PG3
C17A104	Tox0	NA1
C17A105	Tox0	NA1

Table IV : Examples of co-infection by both Tox+ and Tox0 isolates.

Occurence of pathogenicity groups :

104 isolates Tox+ were further characterized in PGs subgroups. 81% of these Tox+ isolates were PG3 and 17% PG4 (Tab V).

Subgroups	No of isolates	% of PGs subgroups
Tox+ PG2	2	2 %
Tox+ PG3	84	81 %
Tox+ PG4	18	17 %
Total	104

Table V : Pathogenicity groups occured in France in 1997.

PG3 isolates were the most abundant in the six locations. PG4 isolates were only found in 4 sites (Tab VI). Two PG2 isolates were detected for the first time in France, respectively in the northeastern and southwertern areas.

Locations	No of PG3	No of PG4	No of PG2	Total of PGs
17-Surgères	14	5	1	20
18-Saint Florent	31	6	0	37
21-Dijon	3	0	0	3
34-Béziers	4	5	0	9
54-Nancy	12	2	1	15
77-Saint Pathus	16	0	0	16

Table VI : Characterization of subgroups PGs in different locations in France in 1997.

PG3 was also the most frequent subgroup recovered on each cultivar. However, more PG4 isolates were observed on Vivol cv. But Vivol is known to be resistant to PG3 isolates and a selection pressure would occur leading an increase in PG4 isolates (Tab VII).

	Number of isolates
Cultivar	PG3	PG4	PG2	Total of PGs
Bristol	22	0	1	23
Goeland	15	6	0	21
Synergy	22	2	1	25
Vivol	18	10	0	28

Table VII : Characterization of subgroups PGs recovered from different cultivars.

DISCUSSION - CONCLUSION

Our results show that both aggressive Tox+ and non aggressive Tox0 isolates are present in French fields. But geographical variation may occur in the incidence of Tox+ type in population of L. maculans. Tox+ type seems to be prevalent in all regions except in the eastern area. This result confirms previous report of dominance of Tox0 near Nancy (Rouxel and Balesdent, 1996). The winter conditions in this area would appear to be less favorable for Tox+ than in other locations and might explain the prevalence of Tox0.

During crop season, Tox+ type appears to be prevalent in autumn but declines before harvesting. Furthermore there is evidence of coinfection in plant by both Tox+ and Tox0 types. Tox+ type would establish in oilseed rape before the infection by Tox0 type and both would be in direct competition. More information is required on epidemiology of both isolates and their interactions.

Among French Tox+ isolates, PG3 forms the largest subgroup, followed by PG4. PG2 is currently rare, it has been detected for the first time in 1997. The predominance of PG3 in France is in constrast with other countries such as western Canada and Australia where PG2 and PG4 respectively are predominant.

The knowlegde of subgroups of aggressive populations of L. maculans will be usefull for resistance breeding. Nevertheless the introduction of a PG3 resistant cultivar such as Vivol could lead to increase a selection pressure of PG4.

Finally, molecular tools enable us to investigate variability of the pathogen. Improvement of the methods for in planta analysis may be of value as a rapid diagnostic technique for large-scale and more representative identification of L. maculans.

ACKNOWLEDGEMENTS

This study was partially supported by the FAIR CT96-1669 contract from the EU. We thank M. H. Balesdent and her team for providing molecular tools for characterization isolates of L. maculans.

REFERENCES

1. Balesdent M.H, Jedryczka M, Jain L, Mendes-Pereira E, Bertrandy J, Rouxel T. 1998. Conidia as a substrate for Internal Transcribed Spacer-based PCR identification of members of Leptosphaeria maculans species complex. Phytopathology 88 : 1210-1217.

2. Crespel L, Brun H, Cavelier N. 1997. Variabilité de populations naturelles françaises de Leptosphaeria maculans pour leur sensibilité à différents fongicides. 5ème Conférence Internationale sur les maladies des plantes ANPP Tours, II : 1109.

3. Koch E, Song K, Osborn T.C, Williams P.H. 1991. Relationship between pathogenicity and phylogeny based on Restriction Fragment Lengh Polymorphism in Leptosphaeria maculans. Mol. Plant-Microbe Interact. 4 : 341-349.

4. Lagarde F. 1995. Bilan phytosanitaire de la dernière campagne et point sur la campagne en cours pour le colza. Retour en force du phoma. Phytoma 469 : 21.23.

5. Rouxel T, Balesdent M.H. 1996. Un nom, plusieurs souches. Oléoscope 35 : 13-14.

6. Rouxel T, Gall C, Balesdent M.H. 1994. Du polymorphisme au complexe d'espèces: combien d'agents pathogènes sont impliqués dans la nécrose du collet du colza? Agronomie 14 : 413-432.