Pathogenicity and race characterization of Fusarium oxysporum f. sp. phaseoli isolates from Spain and Greece

Authors

  • F. M. Alves-Santos,

    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author
    • Present address: Plant Breeding Department, Misión Biológica de Galicia-CSIC, PO Box 28, 36080 Pontevedra, Spain.

  • L. Cordeiro-Rodrigues,

    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author
    • §

      Present address: Escola Superior Agraria, Ap. 172, 5300 Bragança, Portugal

  • J. M. Sayagués,

    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author
    • Present address: Area de Medicina, Departamento de Medicina, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca, Spain.

  • R. Martín-Domínguez,

    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author
  • P. García-Benavides,

    1. Departamento de Producción Vegetal, Universidad de Salamanca, Avda. Filiberto Villalobos 119, 37007 Salamanca; and
    Search for more papers by this author
  • M. C. Crespo,

    1. Centro de Investigación del Toro de Lidia, Ap. oficial, Junta de Castilla y León, Salamanca, Spain
    Search for more papers by this author
  • J. M. Díaz-Mínguez,

    Corresponding author
    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author
  • A. P. Eslava

    1. Area de Genética, Centro Hispano-Luso de Investigaciones Agrarias, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca;
    Search for more papers by this author

*To whom correspondence should be addressed. E-mail: josediaz@gugu.usal.es

Abstract

Virulence (≡ severity of disease) and physiological specialization of nine isolates of Fusarium oxysporum f. sp. phaseoli recovered in El Barco de Avila (Castilla y León, west-central Spain) and of two isolates from Chryssoupolis (Greece) were determined. The susceptibility/resistance response showed by a differential set of common bean cultivars (Phaseolus vulgaris) selected at the Centro Internacional de Agricultura Tropical (CIAT) delineated the isolates into two new races: races 6 and 7. The results of pathogenicity tests did not show any significant differences in virulence among the isolates. However, the reactions of several Spanish common bean cultivars indicated the presence of two groups of isolates, highly virulent and weakly virulent, among the Spanish isolates analysed. These results indicate that isolates classified in the same race are not homogeneous with respect to virulence, and suggests that race analysis using the CIAT differential cultivars is insufficient to describe the physiological specialization of F. oxysporum f. sp. phaseoli.

Introduction

Fusarium wilt, caused by Fusarium oxysporum f. sp. phaseoli, is a serious disease of common bean (Phaseolus vulgaris). Crop losses have been reported in South America and Africa (Abawi & Pastor-Corrales, 1990), and the disease has also been described in the United States (Kendrick & Snyder, 1942) and in some European countries (Aloj et al., 1983). In Spain, it is the most important disease that affects the common bean cultivars grown in El Barco de Avila (Castilla y León, west-central Spain) (Díaz-Mínguez et al., 1996). These cultivars have achieved an excellent reputation due to their gastronomic characteristics.

Pathogenic variability has been analysed in F. oxysporum f. sp. phaseoli by the specific pathogenic interaction of the fungus with a set of differential cultivars and, so far, five races have been described (Woo et al., 1996). There is a good relationship between races and geographic origin, as race 2 includes isolates from Brazil (de Ribeiro & Hagedorn, 1979b), race 3 includes isolates from Colombia (Salgado et al., 1995), race 4 includes one isolate from Colorado (USA) (Salgado & Schwartz, 1993) and race 5 includes isolates from Greece (Woo et al., 1996). The only exception is race 1, occurring in isolates both from South Carolina (USA) and Portici (Italy) (de Ribeiro & Hagedorn, 1979a; Aloj et al., 1987). However, in most cases these analyses have been developed using only one or two isolates as representatives of each race, so that intraspecific variability was difficult to demonstrate. Also, most of the cultivars of common bean used as a differential set in the inoculation experiments are small seeded, indicating a likely meso-American domestication origin. Two centres of origin have been described for Phaseolus vulgaris, one in the meso-American region and the other in the Andean region of America (Gepts & Debouck, 1991). Most of the cultivars and land races grown in Spain correspond to the Andean genetic pool (Gepts & Bliss, 1988; Lioi, 1989; Gil & de Ron, 1992).

Recently, nine isolates of F. oxysporum f. sp. phaseoli have been isolated from the area of El Barco de Avila in Spain (Díaz-Mínguez et al., 1996; Velásquez-Valle et al., 1997). Their genetic diversity has been analysed and clear differences have been found between the Spanish isolates and those isolated in the Americas, indicating that the Spanish isolates are a natural group (Alves-Santos et al., 1999; Alves-Santos et al., 2002). However, the relationship of this genetic divergence with the race structure of the Spanish populations of F. oxysporum f. sp. phaseoli remains unknown. This is an important subject, as resistance to F. oxysporum f. sp. phaseoli seems to be race-specific and different models of inheritance have been found. Resistance determined by a single, completely dominant gene has been found in the interaction of common bean with races 2 and 4 of the pathogen (de Ribeiro & Hagedorn, 1979a; Salgado et al., 1995), resistance against race 1 is determined by a single incompletely dominant gene (de Ribeiro & Hagedorn, 1979a), and recessive gene action, and even quantitative models have also been found controlling resistance against race 4 (Salgado et al., 1995; Cross et al., 2000).

The objectives of the present work were: (i) to analyse the range of virulence (≡ severity of disease) and races in a number of European isolates of F. oxysporum f. sp. phaseoli, and (ii) to compare the reactions of several Spanish cultivars of common bean to infection by European and American isolates of F. oxysporum f. sp. phaseoli. This information is valuable for the development of resistance breeding programmes to select for cultivars to reduce the devastating effects of fusarium wilt in the area of El Barco de Avila.

Materials and methods

Fungal and plant material

All F. oxysporum f. sp. phaseoli isolates used in this work are listed in Table 1. Fungal cultures were established from monoconidial stock cultures stored on 25% glycerol at −80°C and grown in potato dextrose agar (PDA), acid PDA or potato dextrose broth (PDB) media as previously described (Alves-Santos et al., 1999) and incubated at 22°C under continuous light. Conidial inoculum was produced by filtration of fungal cultures in PDB through several layers of cheesecloth. The selected Spanish common bean (P. vulgaris) cultivars were Blanca-Redonda, Blanca-Riñón, Plancheta and Pinta. These four cultivars are widely cultivated in Castilla y León and other regions of Spain and all are large-seeded cultivars, thus indicating the likely presence of ‘T’, ‘C’, ‘A’ or ‘H’ phaseolin types and an Andean origin (Gepts & Bliss, 1988). Five cultivars from the Centro Internacional de Agricultura Tropical (CIAT, Cali, Colombia) were used for race determinations: Diacol Calima (red mottled, large seed), BAT 477 (cream-beige, small seed), HF 465-63-1 (cream-beige, small seed), IPA 1 (cream-beige, small seed) and A211 (black, small seed). All these cultivars have been used in previous studies for race determination in F. oxysporum f. sp. phaseoli (Salgado & Schwartz, 1993; Salgado et al., 1995; Woo et al., 1996), so that any new races should be described in terms of new susceptible/reaction patterns against these cultivars.

Table 1. Fusarium oxysporum f. sp. phaseoli isolates used in this study
IsolateSourceaGeographical originRacebVCGc
  • VCG, vegetative compatibility group.

  • a

    Source of isolates: SA, collection of the Departamento de Microbiología y Genética; HFS, Dr H F Schwartz (Colorado State University, USA); MPC, Dr MA Pastor-Corrales (CIAT, Cali, Colombia); KE, Dr K Elena (Benaki Phytopathological Institute, Kifisia, Athens, Greece).

  • b

    Races are as determined in this study, except those indicated by an asterisk, which were previously described (Woo et al., 1996).

  • c

    VCGs have been previously described (Woo et al., 1996; Alves-Santos et al., 1999). The two Greek isolates are vegetatively incompatible with the other isolates used in this study.

FOP-SP1SAAvila, Spain60167
FOP-SP2SAAvila, Spain60167
FOP-SP3SAAvila, Spain60168
FOP-SP4SAAvila, Spain60166
FOP-SP5SAAvila, Spain60166
FOP-SP6SAAvila, Spain60166
FOP-SP7SAAvila, Spain60166
FOP-SP8SAAvila, Spain60166
FOP-SP9SAAvila, Spain60166
ATCC18131HFSSouth Carolina, USA1*0161
ATCC42145MPCRio de Janeiro, Brazil2*0162
FOP-CL25HFSColombia3*0164
ATCC90245HFSColorado, USA4*016-
Fo 10PhKEKastoria, Greece5*0165
F551KEChryssoupolis, Greece7
F566KEChryssoupolis, Greece7

Pathogenicity tests

A standard root-dip inoculation technique was used (Abawi & Pastor-Corrales, 1990). Briefly, roots of 8- to 10-day-old-seedlings (unifoliate leaves were at three-quarters of full expansion) were washed in running tap water. One centimetre was cut off each of the tips, and then the roots were dipped for 5 min in a conidial suspension of 106 conidia mL−1 of water. Inoculated seedlings were then transplanted into plastic pots with sterilized vermiculite. Plants were grown in a environment-controlled growth chamber at 23–25°C with a 12-h photoperiod of fluorescent light, relative humidity of 60–80% and were fertilized (5%-5%-5·5% NPK) once a week to achieve vigorous growth. Fusarium wilt severity was recorded at 1-week intervals (for a total of 4–6 weeks) after inoculation using the CIAT scale (Pastor-Corrales & Abawi, 1987) of 1 (no external symptoms) to 9 (dead or severely diseased plants with 100% of the foliage showing wilting, necrosis and/or premature defoliation). After 3–4 weeks, plants rated 1–3 were classified as resistant, 3·1–6 as intermediate, and 6·1–9 as susceptible (Salgado & Schwartz, 1993). For race determination, intermediate reactions were combined with susceptible reactions, as previously described (Salgado et al., 1995). Internal discoloration was also checked at 1-week intervals after inoculation by cutting the main stem at the primary node and noting the presence or absence of discoloration.

All the isolates were tested simultaneously in each inoculation experiment in order to reduce the variability in the cultivar response due to environmental conditions. Each inoculation was repeated three times and 10 replicate plants were included per treatment. The disease ratings of the 30 plants evaluated in the three experiments were used to calculate the mean disease rating and the standard deviation for each time-point. The differences between means were considered significant (P ≤ 0·05) according to a least significant difference (LSD) multiple range test or an analysis of variance (anova) test.

Isolation of nit mutants and complementation tests

Complementation between nit mutants was performed for vegetative compatibility group (VCG) verification after an inoculation test. At the end of each inoculation experiment, stem portions from the diseased plants were cut and placed on PDA medium to recover the pathogen and for identification. For each isolate recovered, a nit1 mutant was obtained which was then paired with the NitM mutant corresponding to the original isolate used in the inoculation test, and vegetative compatibility was checked (Correll et al., 1987; Alves-Santos et al., 1999).

Results

Reaction of selected Spanish common bean cultivars to infection by F. oxysporum f. sp. phaseoli

The reaction of the four common bean cultivars to the F. oxysporum f. sp. phaseoli isolates collected in Spain and Greece was compared in inoculation tests by following the progression of the disease. All the isolates tested were pathogenic to these cultivars, but the disease ratings indicated that they may be separated into two virulence groups (Fig. 1a), irrespective of the cultivar tested. Some isolates (highly virulent) induced very severe symptoms in susceptible plants 2–3 weeks after infection (most of the plants were dying by 3 weeks). Other isolates (weakly virulent) induced a less severe response in the cultivars tested, characterized by a slower progression of the disease. The group of highly virulent isolates included isolates from Spain (FOP-SP1, FOP-SP2 and FOP-SP3) and isolates from Greece (F551 and F566). The weakly virulent isolates were all of Spanish origin (FOP-SP4, FOP-SP5, FOP-SP6, FOP-SP7, FOP-SP8, FOP-SP9). Disease ratings recorded 4 weeks after infection were compared, as these were the top ratings obtained, and no significant progression of the disease was observed after this period (data not shown). A statistical analysis was performed to confirm the validity of the two suggested virulence groups (Table 2). The LSD multiple range test confirmed the statistical significance of the disease ratings achieved by the highly virulent isolates (no significant differences at P≤ 0·05) and separated them from the ratings induced by the weakly virulent isolates. The latter were all homogeneous (at P≤ 0·05), except the ratings induced by FOP-SP7 and FOP-SP8 in cultivar Riñón, which were lower than those induced by FOP-SP4-6 and FOP-SP9.

Figure 1.

Progress of fusarium wilt in four Spanish common bean cultivars inoculated with different Fusarium oxysporum f. sp. phaseoli isolates. (a) Inoculation with nine Spanish (FOP-SP1 to FOP-SP9) and two Greek (F551 and F566) isolates. (b) Inoculation with representative isolates of the American races of F. oxysporum f. sp. phaseoli: ATCC18131, race 1; ATCC42145, race 2; FOP-CL25, race 3; ATCC90245, race 4. Severity of the disease is expressed as Centro Internacional de Agricultura Tropical (CIAT) scale ratings recorded weekly after inoculation.

Table 2.  Reactions of Spanish common bean cultivars to infection with isolates of Fusarium oxysporum f. sp. phaseoli. Disease ratingsa, according to Centro Internacional de Agricultura Tropical (CIAT) scale, were recorded 4 weeks after inoculation of the plants
IsolatesCommon bean cultivars
Blanca-RiñónBlanca-RedondaPintaPlancheta
  • a

    The ratings are means of three independent inoculation experiments. Means are followed by the standard deviation calculated in the three experiments.

  • b

    Means followed by the same lower case letter (a–d) within each column are not significantly different at P= 0·05 (LSD multiple range test).

FOP-SP18·8 ± 0·3 ab8·4 ± 0·4 a8·5 ± 0·5 a8·8 ± 0·2 a
FOP-SP28·8 ± 0·2 a8·5 ± 0·4 a8·8 ± 0·2 a9·0 ± 0·0 a
FOP-SP38·8 ± 0·2 a9·0 ± 0·0 a8·8 ± 0·2 a9·0 ± 0·4 a
FOP-SP46·8 ± 1·1 b4·7 ± 2·9 b5·8 ± 3·5 b6·7 ± 1·7 b
FOP-SP57·3 ± 1·6 b5·6 ± 2·9 b6·3 ± 3·2 b6·0 ± 1·9 b
FOP-SP66·9 ± 1·4 b4·6 ± 2·8 b6·3 ± 1·9 b6·3 ± 1·9 b
FOP-SP75·2 ± 0·9 c5·7 ± 0·7 b5·9 ± 3·1 b7·4 ± 1·7 b
FOP-SP84·6 ± 2·5 c3·0 ± 2·0 b8·1 ± 0·7 b7·8 ± 0·8 b
FOP-SP96·4 ± 1·3 b4·6 ± 2·8 b5·9 ± 3·3 b6·7 ± 1·6 b
F5519·0 ± 0·0 a9·0 ± 0·0 a8·9 ± 0·1 a9·0 ± 0·0 a
F5669·0 ± 0·0 a8·4 ± 0·0 a9·0 ± 0·0 a8·9 ± 0·1 a
ATCC181318·5 ± 0·5 a6·4 ± 2·6 b9·0 ± 0·0 a9·0 ± 0·0 a
ATCC421457·5 ± 1·1 b5·5 ± 1·5 b5·3 ± 2·2 b6·5 ± 0·9 b
FOP-CL257·3 ± 0·9 b8·8 ± 0·3 a8·8 ± 0·2 a2·0 ± 0·5 c
ATCC902459·0 ± 0·0 a8·6 ± 0·4 a9·0 ± 0·0 a8·4 ± 0·6 a

The Spanish common bean cultivars were also tested in inoculation experiments with representative isolates of the four American races described to date (Woo et al., 1996), as shown in Fig. 1(b). The progression of the disease was dependent on both the cultivar analysed and the race of the pathogen, in most cases. The race 4 isolate behaved as highly virulent against the four cultivars, and the race 1 isolate was also highly virulent against Riñón, Plancheta and Pinta, but weakly virulent against Blanca-Redonda. The race 3 isolate was highly virulent against Blanca-Redonda and Pinta, weakly virulent against Riñón and nonpathogenic to Plancheta. The race 2 isolate was weakly virulent against all four cultivars. The statistical analysis of the disease ratings 4 weeks after inoculation (Table 2) confirmed the differences observed in the progression of the disease induced by the different American races of the pathogen. Also, there was no clear difference in virulence between the American and the Spanish and Greek isolates. Only the reactions of cultivar Blanca-Riñón to Spanish isolates FOP-SP7 and FOP-SP8 showed mean disease ratings lower than those obtained in the infection with the other Spanish and American isolates which behaved as weakly virulent against this cultivar.

VCG identification of pathogenic isolates

Pathogenic mycelia were recovered from all the diseased plants, except in nonpathogenic interactions such as that between FOP-CL25/Plancheta. FOP-SP isolates, classified as highly or weakly virulent, belonged to different VCGs (Alves-Santos et al., 1999). The isolates representative of the four American races all belonged to a different VCG (Woo et al., 1996), and the Greek isolates F551 and F566 had been determined to be vegetatively compatible, but not compatible with the other isolates used in this work (data not shown). Thus, verification of the vegetative compatibility of NitM mutants obtained from the isolate used for inoculation and nit1 mutants obtained from the mycelium recovered from each plant made it possible to confirm the VCG of each pathogen or group of pathogens.

Two new races of F. oxysporum f. sp. phaseoli

The response of the Spanish common bean cultivars to the Spanish F. oxysporum f. sp. phaseoli isolates tested showed the presence of two groups of pathogens with a clear difference in virulence. Such differences in virulence were also observed in the response of the Spanish cultivars to the American races of the pathogen. Therefore, the question remained as to whether each group of Spanish isolates could also belong to a different race.

Pathogenicity tests were conducted on five P. vulgaris cultivars from a differential cultivar set selected at CIAT that had previously been used for race differentiation of F. oxysporum f. sp. phaseoli. All the Spanish FOP-SP isolates, the two Greek isolates from Chryssoupolis, and a representative isolate of each of the five previously reported races were included in this analysis. Disease ratings used for race classification were those scored at 1-week intervals up to 5 weeks postinoculation, as thereafter no progression of the disease was observed (data not shown). Ratings were classified in disease reactions as described above. Scores from 3·1 up to 9 were rated as susceptible. Therefore the response of a cultivar was classified as resistant when almost no external symptoms of wilt were observed, and susceptible when external symptoms and/or death of the plant was observed. The disease ratings induced on each cultivar by all the FOP-SP isolates were very similar. Five cultivars were susceptible to FOP-SP isolates, but only cultivars Bat 477 and Calima were susceptible to isolates F551 and F566 from Greece (Table 3). The representatives of races 1–5 showed patterns of resistance–susceptibility in agreement with those previously described (Woo et al., 1996). The resistance–susceptibility patterns exhibited by the Spanish and Greek isolates were different from the patterns exhibited by the representative isolates of races 1–5. Therefore, these isolates have been classified as new races: race 6, which comprises all the FOP-SP isolates recovered in El Barco de Avila (Spain), and race 7, which comprises the two isolates from Chryssoupolis (Greece).

Table 3.  Reactions of five common bean cultivars (Centro Internacional de Agricultura Tropical [CIAT] differential cultivars) to infection with isolates of Fusarium oxysporum f. sp. phaseoli. Disease ratings, according to CIAT scale, were recorded 5 weeks after inoculation of the plants
IsolateaCommon bean cultivars
A211BAT477CalimaIPA1HF 465-63-1
RatingbReactioncRatingbReactioncRatingbReactioncRatingbReactioncRatingbReactionc
FOP-SP1 (6)7·8 ± 1·2S8·3 ± 1·0S7·9 ± 1·0S9·0 ± 0·0S7·1 ± 2·7S
FOP-SP3 (6)8·9 ± 0·1S7·7 ± 1·0S8·5 ± 0·7S8·3 ± 0·9S6·7 ± 1·0S
FOP-SP5 (6)9·0 ± 0·0S6·7 ± 0·9S6·6 ± 2·4S8·2 ± 0·5S7·0 ± 0·7S
FOP-SP9 (6)9·0 ± 0·0S8·3 ± 0·7S6·9 ± 1·9S6·7 ± 0·3S7·0 ± 0·7S
F551 (7)5·0 ± 0·0R8·2 ± 1·8S9·0 ± 0·0S2·2 ± 1·8R4·8 ± 0·5R
F566 (7)4·0 ± 0·0R7·7 ± 2·1S6·3 ± 2·6S2·8 ± 3·0R5·0 ± 1·2R
ATCC18131 (1)SSRRR
ATCC42145 (2)SRRSR
FOP-CL25 (3)RRSRR
ATCC90245 (4)SSSSR
Fo 10Ph (5)SSRSR

Interestingly, the FOP-SP isolates previously differentiated into highly and weakly virulent towards Spanish common bean cultivars showed no difference in virulence towards the CIAT differential cultivars. The means of disease severity in susceptible reactions, recorded 5 weeks after inoculation, were not significantly different (P ≤ 0·05) for any combination of pathogen–bean cultivar. The analysis of the disease ratings at different times after inoculation did not allow the separation of the FOP-SP isolates into two virulence groups (data not shown), as had been the case with the reactions to the Spanish common bean cultivars. On the contrary, in some cases, e.g. with cultivar A211, the disease ratings were higher when inoculated with FOP-SP5 and FOP-SP9 (low virulence to Spanish cultivars) than when inoculated with FOP-SP2 or FOP-SP3 (high virulence to Spanish cultivars).

Vascular discoloration could be seen in all plants that showed a susceptible reaction. Mycelia were recovered from all susceptible plants and VCG was verified.

Discussion

The Spanish and Greek isolates of F. oxysporum f. sp. phaseoli analysed have been shown to belong to two new physiological races. All the isolates collected in El Barco de Avila (Spain) belong to the same race (race 6) while the isolates collected in Chryssoupolis (Greece) belong to a different race (race 7). Preliminary evidence suggested a physiological specialization of isolate FOP-SP1 (Velásquez-Valle & Schwartz, 1998). However, in their report the reactions of cultivars Mortino and IPA 1 did not differentiate the race 4 isolate (ATCC 90245) from the new race 6 isolate. Results reported here show that IPA 1 is clearly susceptible to the FOP-SP isolates and that HF 465-63-1 was susceptible when inoculated with FOP-SP1, although it previously had been described as resistant (Velásquez-Valle & Schwartz, 1998). Moreover, the inoculation of HF 465-63-1 with the other FOP-SP isolates also resulted in susceptible reactions, further confirming the susceptibility of this cultivar to the race 6 isolates. All previous analyses showed that this cultivar was resistant to the race 4 isolate from Colorado (Salgado et al., 1995; Woo et al., 1996; Velásquez-Valle & Schwartz, 1998), thus emphasizing the difference between the race 4 and the race 6 isolates. Former inoculations with isolate FOP-SP1 were conducted in a glasshouse (Velásquez-Valle & Schwartz, 1998), while the inoculations described here were conducted in a controlled environment chamber. This difference is important as it has been observed that disease reactions induced in common bean by F. oxysporum f. sp. phaseoli are slower in the former than in the controlled environment chamber (data not shown).

The pathogenicity assays conducted on the four Spanish bean cultivars indicated that the isolates analysed could be classified into two virulence groups, where virulence is used to indicate the degree of pathogenicity. These groups did not match the two races identified using CIAT differential cultivars. The race structure of the European isolates was more related to geographic origin, as all the Spanish isolates (FOP-SP) belonged to race 6 and the Greek isolates (F551 and F566) belonged to race 7. This is in good agreement with previous findings (de Ribeiro & Hagedorn, 1979b; Salgado et al., 1995; Woo et al., 1996) and suggests a complex race structure for the populations of F. oxysporum f. sp. phaseoli in both Europe and America (seven races have been found among 21 isolates analysed). This physiological specialization is not surprising when the highly diverse structure of common bean cultivars is taken into account and with coevolutionary events such as demonstrated for other bean diseases, e.g. angular leaf spot (Guzmán et al., 1995), anthracnose (Geffroy et al., 1999).

The designation of the nine isolates collected in El Barco de Avila to a single race confirms and complements previous results, which showed the genetic similarity of all the FOP-SP isolates (Alves-Santos et al., 1999, 2002). Both the race structure and the genetic data suggest a single origin for the Spanish race 6 isolates. However, the presence of two virulence groups suggests the expression of different virulence factors in the two groups of isolates and that the virulence factors involved interact differently with the centro-American and Andean bean cultivars. This suggests a limitation of the CIAT differential set of cultivars used for the physiological characterization of F. oxysporum f. sp. phaseoli isolates.

The patterns of genetic diversity and race structure have a direct bearing on breeding for resistance. Unfortunately, none of the cultivars of the differential set examined is resistant to the race from El Barco de Avila. The search for resistance using other bean cultivars of the meso-American bean gene pool was also unsuccessful (F.M. Alves-Santos, unpublished data). Currently, a survey of common bean germplasm of the Andean gene pool is underway, in an attempt to find a source of resistance which could be used in breeding programmes of the Spanish common bean cultivars. This approach, together with the rapid diagnostic procedure that has been developed (Alves-Santos et al., 2002), may assist in reducing the threat of fusarium wilt on common bean.

Acknowledgements

This research was supported by grant 1FD97-0308-C03-01 (Ministerio de Educación y Cultura-Fondos FEDER), grant SC95-002-C2-1 (Ministerio de Agricultura, Pesca y Alimentación) and grant SA70/94 from Junta de Castilla y León. FMA-S was the recipient of a fellowship from the Instituto Nacional de Investigaciones Agrarias (MAPA). We thank K. Elena, M.A. Pastor-Corrales and H.F. Schwartz for providing fungal isolates, and E.P. Benito and M. Perlin for critical reading of the manuscript.

Ancillary