2.2.60
CHARACTERIZATION OF MEXICAN ISOLATES OF COLLETOTRICHUM LINDEMUTHIANUM USING DIFFERENTIAL CULTIVARS AND MOLECULAR MARKERS

M GONZALEZ1, R RODRIGUEZ1, ME ZAVALA 1, F HERNANDEZ1, J JACOBO2, J ACOSTA 2, O MARTINEZ1 and J SIMPSON 1

1Department of Genetic Engineering, CINVESTAV, Unidad Irapuato, Mexico; 2INIFAP, Durango, Mexico; 3INIFAP, Chapingo, Mexico

Background and objectives
Anthracnose disease on common bean (Phaseolus vulgaris) caused by the fungal pathogen Colletotrichum lindemuthianum can be devastating when climatic conditions favor the pathogen. In developing countries the best strategy to combat the pathogen is to produce resistant germplasm. However this is hampered by the diversity of pathotypes within a given region. Although differential cultivars are useful in determining races or pathotypes of plant pathogens, classification of disease severity is often subjective. A possible solution is to combine data from differential cultivars with DNA based molecular marker analysis. Our interest is to study the variability of C. lindemuthianum populations in Mexico by combining analyses involving differential cultivars with molecular marker analysis. The objectives of the study were to: (1) determine the variability of the pathogen in terms of races or pathotypes as defined by inoculation of differential cultivars; (2) compare the efficiency of RAPD and AFLP markers for classifying C. lindemuthianum isolates, (3) analyse the variability of the pathogen at the genotype, level and (4) compare the variability in pathotypes with that determined by molecular marker techniques.

Materials and methods
Fungal samples were obtained during the months of September to November from 1992-96 from leaves or pods of naturally infected bean plants found in the field. A colony produced from a single spore was maintained as a pure isolate for infection of differential cultivars and genotype analysis. Symptoms were evaluated on a five point scale . DNA extraction, RAPD and AFLP analyses were all carried out under standard conditions.Data was analysed using SAS (Statistical Analysis System) 6.08 for windows and S-plus 4.0 for windows.
Results and conclusions
Ten distinct pathotypes or races were determined in the 59 isolates examined. Three of these, races 8, 1472 and 1088 are reported for the first time in Mexico. Only one race was common to the two geographic regions sampled. Differences in pathogen populations between in the two regions probably reflect the differences in germplasm and the agricultural practices employed in each region. The preferential infection of certain cultivars probably reflects the adaptation of the pathogen within Mexico. To date, only two of 129 Mexcian isolates tested infected cultivars of Andean origin suggesting that races of C. lindemuthianum capable of infecting Andean germplasm are rare in Mexico. None of the 129 isolates were capable of infecting the differential cultivars Widusa, Tu or G2333. Suggesting that these three resistant cultivars may provide effective resistance to C. lindemuthianum in Mexico. This study and previous analyses of C. lindemuthianum isolates using RAPD1s have failed to show direct association between haplotype patterns and geographic origin, Sicard et al. [1]. However, all 59 isolates examined had unique AFLP and RAPD haplotypes. Statistical analysis strongly supports the division of the isolates into four haplotype groups which may represent distinct genetic lineages of the pathogen as has previously been suggested for Pyricularia oryzae [2]. A correlation between the defined lineages and both cultivar type and pathotype can be established. In this study all isolates from bush type cultivars grouped together as did those from climbing type cultivars. We are currently developing a method to predict pathotype groups based on haplotype data.

References
1. Sicard D, Michalakis Y, Dron M, Neema C, 1987. Phytopathology 87, 807-813.
2. Ziegler RS, Cuoc LX, Scott RP, Bernardo MA, Chen DH, Valent B, Nelson RJ, 1995. Phytopathology 85, 443-451.