APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1999, p. 3335–3340 0099-2240/99/$04.00⫹0 Copyright © 1999, American Society for Microbiology. All Rights Reserved.
Vol. 65, No. 8
Genetic Diversity of Fusarium oxysporum Strains from Common Bean Fields in Spain FERNANDO M. ALVES-SANTOS, ERNESTO P. BENITO, ARTURO P. ESLAVA, ´ MARI´A DI´AZ-MI´NGUEZ* AND JOSE Area de Gene´tica, Departamento de Microbiologı´a y Gene´tica, Universidad de Salamanca, 37007 Salamanca, Spain Received 20 November 1998/Accepted 20 May 1999
Fusarium wilt is an endemic disease in El Barco de Avila (Castilla y Leo ´n, west-central Spain), where high-quality common bean cultivars have been cultured for the last century. We used intergenic spacer (IGS) region polymorphism of ribosomal DNA, electrophoretic karyotype patterns, and vegetative compatibility and pathogenicity analyses to assess the genetic diversity within Fusarium oxysporum isolates recovered from common bean plants growing in fields around El Barco de Avila. Ninety-six vegetative compatibility groups (VCGs) were found among 128 isolates analyzed; most of these VCGs contained only a single isolate. The strains belonging to pathogenic VCGs and the most abundant nonpathogenic VCGs were further examined for polymorphisms in the IGS region and electrophoretic karyotype patterns. Isolates belonging to the same VCG exhibited the same IGS haplotype and very similar electrophoretic karyotype patterns. These findings are consistent with the hypothesis that VCGs represent clonal lineages that rarely, if ever, reproduce sexually. The F. oxysporum f. sp. phaseoli strains recovered had the same IGS haplotype and similar electrophoretic karyotype patterns, different from those found for F. oxysporum f. sp. phaseoli from the Americas, and were assigned to three new VCGs (VCGs 0166, 0167, and 0168). Based on our results, we do not consider the strains belonging to F. oxysporum f. sp. phaseoli to be a monophyletic group within F. oxysporum, as there is no correlation between pathogenicity and VCG, IGS restriction fragment length polymorphism, or electrophoretic karyotype. F. oxysporum f. sp. phaseoli strains from El Barco de Avila with the F. oxysporum f. sp. phaseoli strains collected in the Americas. We assessed diversity by examining the isolates for pathogenicity, vegetative compatibility, restriction fragment length polymorphisms (RFLPs) in the intergenic spacer (IGS) region of the nuclear ribosomal DNA, and electrophoretic karyotypes (EKs). The genetic relationships within and between the nonpathogenic and pathogenic strains from El Barco de Avila and the pathogenic strains from the Americas may provide new insights into the evolution of F. oxysporum f. sp. phaseoli.
Fusarium oxysporum Schlechtend:Fr. is an anamorphic species with considerable morphological and physiological variation. Most of the interest in this fungus arises because of its ability to cause diseases of economically important plant hosts, but its near ubiquity in soils worldwide and its ecological activities indicate a much more diverse role in nature. Pathogenic isolates of F. oxysporum often display a high degree of host specificity and may be subdivided into formae speciales based on the species attacked and into races based on the host cultivars attacked (6). A description of the relationships between pathogenic and nonpathogenic forms might shed new light on the evolution of pathogenic F. oxysporum strains. Most research has focused on strains capable of causing disease; nonpathogenic strains largely have been neglected, except for some interest in their use as biocontrol agents (1, 28, 29). The limited studies available indicate that nonpathogenic forms occurring in individual fields show a high level of diversity (3, 14) and are different from pathogenic strains (16, 17, 20). However, given the close association of this fungus with plant roots, this diversity might be dependent on the cropping history of the field (13). Fusarium wilt is the most important disease affecting the common bean grown in El Barco de Avila (Castilla y Leo ´n, Spain). Symptoms closely matching those described as fusarium wilt or bean yellows (21) in this area were described as early as 1926 (7). We have isolated F. oxysporum f. sp. phaseoli strains capable of producing wilt and vascular disease when inoculated on common bean cultivars (11, 33). Our objectives in this study were (i) to analyze the diversity and genetic relationships of F. oxysporum isolates collected from common bean fields in the area of El Barco de Avila and (ii) to compare the
MATERIALS AND METHODS F. oxysporum isolates. One hundred twenty-eight isolates of F. oxysporum were recovered from roots and hypocotyls of common bean plants (Phaseolus vulgaris L.). The plants were sampled from fields with a long history of common bean monoculture cropping. The same local cultivars and land races have been grown for at least 100 years, and it is unlikely that foreign isolates of F. oxysporum have been introduced. Most of the plants sampled showed symptoms of fusarium wilt of the common bean (11, 21). The plants were collected during three growing seasons (May to October) from several plots in the area of El Barco de Avila (Spain). Mycelia suspected to be F. oxysporum were identified by microscopic examination of spores and conidiophores. The identification was confirmed by F. J. Tello Marquina. Pathogenic and nonpathogenic isolates were recovered from symptomatic plants, but pathogenic isolates were never recovered from asymptomatic plants. Single conidial isolates were obtained from each isolate, and only one isolate obtained from each plant was analyzed (Tables 1 and 2). Media and culture conditions. Potato dextrose agar (PDA) medium was made with 39 g of PDA (Difco, Detroit, Mich.) per liter of distilled water. It was acidified to pH 3.2 with 0.1 M HCl for fungal isolation to reduce bacterial growth. Potato dextrose broth (PDB) was made with 24 g of PDB (Difco) per liter of distilled water. PDC was made by adding 15 g of potassium chlorate per liter of PDA medium. Nitrate medium (MM), nitrite medium, hypoxanthine medium, and ammonium medium were used to test nit mutant phenotypes (9). All cultures were incubated at 22°C with continuous light. For long-term conservation, mycelia grown on PDA were scraped from the plates, resuspended in 25% glycerol, and stored at ⫺80°C. Isolation and characterization of nit mutants. Spontaneous chlorate-resistant sectors were recovered from most isolates of F. oxysporum when cultured on PDC. The recovered sectors were transferred to MM, and those that grew as thin colonies with no aerial mycelium were classified as nit mutants. The phenotypes of the nit mutants were determined by their colony morphology on media containing one of four different sources of nitrogen, as described by Correll et al. (9).
* Corresponding author. Mailing address: Area de Gene´tica, Departamento de Microbiologı´a y Gene´tica, Universidad de Salamanca, Plaza de los Doctores de la Reina s/n, 37007 Salamanca, Spain. Phone: 34-23-294663. Fax: 34-23-294663. E-mail:
[email protected]. 3335
3336
ALVES-SANTOS ET AL.
APPL. ENVIRON. MICROBIOL.
TABLE 1. Nonpathogenic isolates of F. oxysporum from Spain used in this study Isolate(s)a i
Geographic origin
i
AB82 , AB92 AB8d, AB26b, AB77d AB17c, AB19c, AB29c AB16e, AB25e, AB31e, AB59e, AB70e, AB73e, AB106–109e AB9, AB14, AB15, AB18, AB20–24, AB27, AB30, AB32f, AB54–61, AB62g, AB63h, AB64, AB65f, AB66–69, AB71, AB72, AB75, AB76, AB78, AB79f, AB81, AB83–85, AB89–91, AB93–97, AB99–102, AB103h, AB104, AB105, AB113–117, AB120–124 AB125, AB126 AB45a AB86e AB1, AB3, AB5, AB7, AB33, AB34, AB36–43, AB47–53 AN1b, AN4d, AN6a, AN7a, AN8b AN2, AN5, AN9, AN10, AN11g AS3 AS2a AS1, AS4
El El El El El
Barco Barco Barco Barco Barco
de de de de de
Avila Avila Avila Avila Avila
Bohoyo La Carrera La Carrera La Carrera Navalonguilla Navalonguilla Solana de Avila Solana de Avila Solana de Avila
IGS haplotype
IGS-A IGS-B1 IGS-B2 IGS-B3 NDb ND IGS-B1 IGS-B3 ND IGS-B1 ND IGS-A IGS-B1 ND
a Isolates with the same superscript are vegetatively compatible: a, VCG 1; b, VCG 3; c, VCG 4; d, VCG 5; e, VCG 6; f, VCG 14; g, VCG 18; h, VCG 21; i, VCG 0167. b ND, not determined.
Complementation tests. Complementation between different nit mutants was indicated by the development of dense aerial growth where the mycelia of the nit mutant colonies anastomosed to form a heterokaryon. Pairings were made by placing mycelia from each nit mutant on MM. In most cases, two different nit mutants of each isolate were tested for complementation. The number of nit1nit3 pairings was minimized, and all of these pairings were repeated at least once. None of the Spanish isolates tested were self-incompatible (10). Pathogenicity tests. A standard root dip inoculation technique was used to test the isolates for pathogenicity on common bean plants (27). Briefly, roots of 1-week-old seedlings were washed in running tap water; about 1 cm was cut from each tip, and the roots then were dipped for 5 min in a spore suspension at 106 spores/ml. Inoculated seedlings were transplanted into plastic pots with sterilized vermiculite. Plants were grown at 23 to 25°C, at a relative humidity of 60 to 80%, and fertilized once a week. Fusarium yellows severity was recorded at 1-week intervals after inoculation by using the Centro Internacional de Agricultura Tropical scale (27) of 1 (no external symptoms) to 9 (dead or severely infected plants with 100% of the foliage showing wilting, chlorosis, necrosis, and/or premature defoliation). All F. oxysporum isolates were tested for pathogenicity on two different Spanish cultivars, Blanca-Redonda and Blanca-Rin ˜o ´n (also known as Riojana), which are widely cultured in El Barco de Avila. DNA extraction. Fusarium cultures were started by inoculating a sample of frozen mycelium in 50 ml of PDB and grown at 22°C for 5 days on an orbital shaker (180 rpm). Total DNA was prepared from a maximum of 300 mg (wet weight) of mycelium by the method of Graham et al. (18). IGS RFLP analysis. The IGS region of the ribosomal DNA was amplified by using primers CNL12 (CTGAACGCCTCTAAGTCAG) and CNS1 (GAGACA AGCATATGACTACTG) (4). Amplification reactions were performed in a total volume of 50 l containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2.5 mM MgCl2, 200 M each deoxynucleoside triphosphate, 0.5 M each primer, 10 ng of genomic DNA, and 2.5 U of Taq DNA polymerase (AmpliTaq; Perkin Elmer,
TABLE 2. F. oxysporum f. sp. phaseoli isolates used in this study VCGa
Isolate(s)
Geographic origin
FOP-SP1 FOP-SP2 FOP-SP3 FOP-SP4, FOP-SP5, FOP-SP6, FOP-SP7 FOP-SP8 FOP-SP9 ATCC 42145/FOP-BR FOP-CL25 ATCC 90245/FOP-CO ATCC 18131/FOP-SC
La Carrera, Spain Barco de Avila, Spain Barco de Avila, Spain La Carrera, Spain
0167 0167 0168 0166
Barco de Avila, Spain Navalonguilla, Spain Brazil Colombia Colorado South Carolina
0166 0166 0162* 0164* 016–* 0161*
a VCGs are as determined in this study, except those indicated by an asterisk, which were previously described (34). VCGs which include pathogenic isolates are numbered according to Puhalla (30) and the generally accepted procedure for F. oxysporum (22).
Branchburg, N.J.). The amplification conditions included a hot start for 6 min at 80°C and a denaturation step for 5 min at 94°C, followed by 40 amplification cycles consisting of 1 min at 94°C, 1 min at 58°C, and 2 min at 72°C. A final extension step was performed for 10 min at 72°C. A sample (5 l) of each PCR product was digested with ClaI, XhoI, AvaI, AccI, and MspI (Roche Diagnostics, Mannheim, Germany) according to manufacturer’s instructions. Digested DNA was run on 1.3 or 3% agarose gels, depending on the size of the fragments analyzed. The gels were soaked in ethidium bromide solution (0.5 g/ml), and DNA was visualized under UV light. Preparation of intact chromosomal DNA. Protoplasts of selected isolates of F. oxysporum were formed by the procedure described by Boehm et al. (8). Protoplasts were purified by centrifugation at 1,500 ⫻ g for 30 min. The top layer containing the protoplasts was transferred to a fresh tube, and 1 volume of 1 M sorbitol–50 mM CaCl2–10 mM Tris (pH 7.4) (SCT) was added. The suspension was centrifuged at 1,000 ⫻ g for 5 min, and the supernatant was transferred to a fresh tube. Twenty milliliters of SCT was added and mixed with the protoplast suspension. Protoplasts were pelleted by centrifugation at 1,500 ⫻ g for 15 min and washed once with SCT. Finally, they were resuspended in 0.2 to 1 ml of SCT. The protoplast suspension was mixed with 1 volume of 1% low-melting-point agarose (Sea Plaque; FMC Bioproducts, Rockland, Maine) made up in SCT at a final concentration of 1 ⫻ 108 to 2 ⫻ 108 protoplasts/ml. The mixture was allowed to solidify in Bio-Rad (Hercules, Calif.) plug molds. Solidified agarose plugs were incubated in NDS buffer (0.5 M EDTA, 1% N-lauroylsarcosine, 10 mM Tris, pH 9.5) for 24 to 48 h at 50°C. Afterwards, plugs were washed in 50 mM EDTA and stored in the same buffer at 4°C. Pulsed-field gel electrophoresis. Reproducible EKs were generated for all of the isolates analyzed for IGS diversity. No differences were seen among sample preparations. Chromosomal DNA bands were separated in 0.7% (wt/vol) agarose (Fastlane; FMC Bioproducts) in 0.5⫻ Tris-borate-EDTA buffer (32). Electrophoresis was performed with a contour-clamped homogeneous electric field system (CHEF Mapper; Bio-Rad) at 10°C with continuous circulation of buffer. The Tris-borate-EDTA buffer was replaced every 2 days. Different electrophoretic conditions were used to resolve all or most of the chromosomal bands. As most of the variability was found among the smaller chromosomes, the electrophoretic conditions were selected to improve resolution in that range. Electrophoresis for 96 h, at 2.0 V/cm and with pulses ramped from 10 to 30 min, gave an optimal resolution in the range of 1.0 to 6.0 Mb. Gels were stained in ethidium bromide (0.5 g/ml) for 30 min, destained in distilled water overnight, and photographed under UV illumination. Chromosomal preparations of Saccharomyces cerevisiae, Hansenula wingei, and Schizosaccharomyces pombe (BioRad) were used as standards.
RESULTS VCGs. Chlorate-resistant sectors were recovered at a mean frequency of 2.7 sectors per colony on PDC medium (maximum, 5.4 sectors per colony; minimum, 0.7 sectors per colony). Of 722 resistant sectors, 388 (54%) were unable to utilize nitrate as the sole source of nitrogen and grew on MM with no aerial mycelium and thus were designated nit mutants. The
FUSARIUM OXYSPORUM FROM COMMON BEAN FIELDS IN SPAIN
VOL. 65, 1999
3337
TABLE 3. RFLP patterns of the IGS haplotypes IGS haplotype
IGS length (kb)
IGS-A IGS-B1 IGS-B2 IGS-B3 IGS-B4 IGS-B5
2.55 2.6 2.6 2.6 2.6 2.6
RFLP patterna ClaI
XhoI
AvaI
AccI
MspI
2.1, 0.45 2.6 2.1, 0.5 2.1, 0.5 2.1, 0.5 2.1, 0.5
2.55 1.3, 1.3 1.3, 1.3 1.3, 1.3 2.6 2.6
1.7, 0.75, 0.1 1.4, 0.95, 0.25 1.4, 0.95, 0.25 1.4, 0.95, 0.25 1.4, 0.95, 0.25 1.5, 0.95, 0.15
1.65, 0.45, 0.45 1.7, 0.45, 0.45 1.7, 0.45, 0.45 0.85, 0.85, 0.45, 0.45 1.7, 0.45, 0.45 1.7, 0.45, 0.45
0.6, 0.27, 0.2 0.42, 0.27, 0.2 0.55, 0.27, 0.2 0.55, 0.27, 0.2 0.6, 0.27, 0.2 0.55, 0.27, 0.2
a Lengths of the restriction fragments obtained are indicated in kilobase pairs. In the case of the restriction fragments obtained with MspI, only fragments larger than 0.2 kbp are indicated.
relative frequencies of the different nit mutants were as follows: nit1, 77%; nit3, 14%, and nitM, 8%. Ninety-six vegetative compatibility groups (VCGs) were found among 128 isolates analyzed for vegetative compatibility, but the distributions of VCGs among nonpathogenic and pathogenic isolates were very different (Tables 1 and 2). One hundred nineteen nonpathogenic isolates were placed in 93 VCGs, with the majority of them consisting of single-isolate VCGs. The nine F. oxysporum f. sp. phaseoli isolates recovered in the area of El Barco de Avila were placed in three VCGs: 0166, 0167, and 0168. Some VCGs include isolates recovered from nearby fields (10 to 100 m away) (VCGs 4, 14, 19, and 21), but many of them include isolates from distant fields (up to 11 km away) (VCGs 0166, 0167, 1, 3, 5, 6, and 18). Interestingly, VCG 0167 includes two pathogenic and two nonpathogenic isolates. The four isolates from the Americas are associated with different VCGs (34), and none of them were vegetatively compatible with any of the Spanish isolates, either pathogenic or nonpathogenic. ATCC 90245 was self-incompatible, confirming previous results (34). Pathogenicity. All of the Spanish isolates characterized as pathogenic reproduced the yellowing and wilting symptoms observed in the field when inoculated on cultivars Blanca Redonda and Blanca Rin ˜o ´n (11, 33). The F. oxysporum f. sp. phaseoli isolates used in this study reached ratings of higher than 3 in at least two tests with each cultivar. Vascular discoloration was typically observed in infected plants. The F. oxysporum f. sp. phaseoli strains from the Americas were pathogenic on the two cultivars used as testers. Nonpathogenic isolates failed to produce any external symptoms, and vascular discoloration was never observed.
IGS diversity. We examined all of the strains in the most abundant VCGs, a unique pathogenic member of VCG 0168, the nonpathogenic isolate AS3, and the four pathogenic isolates from the Americas. We found two types of polymorphisms in the IGS region. First, the size of the IGS fragment amplified with primers CNS1 and CNL12 was isolate dependent. The smaller IGS (IGS-A, 2.55 kb) was found in isolates AS3, AB82, AB92, and FOP-SP1 to FOP-SP9. The larger IGS (IGS-B, 2.60 kb) was found in the rest of the nonpathogenic isolates and in the isolates from the Americas. We identified a second level of polymorphism following analysis of restriction enzyme fragments. Five restriction endonucleases, ClaI, XhoI, AvaI, AccI, and MspI, generated polymorphic fragments (Table 3 and Fig. 1). Each isolate had only one IGS fragment after PCR amplification. When this fragment was digested with restriction enzymes, the sizes of the fragments generated were consistent with the size of the undigested IGS fragment amplified by PCR. Therefore, no IGS variation, in terms of size or restriction pattern, was detected within any of the isolates tested, and we interpret each combination of size and RFLP pattern as an IGS haplotype (Table 3). There were one IGS-A haplotype and five IGS-B haplotypes (Fig. 1). These haplotypes were not correlated with pathogenicity, vegetative compatibility, or geographic origin. All of the Spanish pathogenic isolates had the IGS-A haplotype, as did the nonpathogenic isolates AB82 and AB92, which are vegetatively compatible with FOP-SP1 and FOP-SP2, and AS3, which is not vegetatively compatible with any of the pathogenic isolates. The rest of the Spanish nonpathogenic isolates had IGS-B1, IGS-B2, and IGS-B3 haplotypes. Pathogenic strains
FIG. 1. Gel electrophoresis analysis showing the patterns obtained when the amplified IGS DNAs from FOP-SP5 (IGS-A), AN6 (IGS-B1), AB17 (IGS-B2), AB16 (IGS-B3), FOP-CL25 (IGS-B4), and ATCC 18131 (IGS-B5) were digested with restriction endonucleases AvaI (lanes 1), AccI (lanes 2), ClaI (lanes 3), and XhoI (lanes 4). Lane M, marker DNA fragments (lambda DNA digested with EcoRI and HindIII); lane L, 100-bp DNA ladder. Only fragments larger than 600 bp are represented. In IGS-B3 lane 3, the top band is partially digested DNA. Numbers on the left and right are kilobase pairs.
3338
ALVES-SANTOS ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 2. EKs of F. oxysporum f. sp. phaseoli and representative nonpathogenic F. oxysporum isolates separated by pulsed-field gel electrophoresis and visualized by ethidium bromide staining. Molecular size markers are chromosomes from H. wingei, S. pombe (Sp), and S. cerevisiae (Sc). Run conditions were as described in Materials and Methods.
from IGS haplotypes B2 (strain ATCC 42145), B4 (strains ATCC 90245 and FOP-CL25), and B5 (strain ATCC 18131) were identified from Brazil, Colorado, Colombia, and South Carolina, respectively. EKs. We observed numerous differences in both chromosome number and size (Fig. 2). In general, all members of the same VCG had the same EK, although all members of VCG 0166 had different karyotypes and had seven to nine resolvable chromosomes. Karyotype diversity among the nonpathogenic strains was much higher than that observed for the pathogenic isolates. The four strains in VCG 0167, two pathogenic and two nonpathogenic, all had seven chromosomes and differed only in the size of the smallest chromosome. The EK patterns of the four pathogenic strains from the Americas were very different from the ones found for the Spanish F. oxysporum strains. The chromosome number was lower than that of any Spanish F. oxysporum strain (five for FOP-CL25 and six for the other isolates). The EK patterns of the two strains from the United States (ATCC 90245 and ATCC 18131) were similar, but ATCC 90245 lacks one of the chromosomes present in ATCC 18131. DISCUSSION In the present work we analyzed genetic and physiological characters to assess the diversity within pathogenic and nonpathogenic populations of F. oxysporum. More diversity was found in the nonpathogenic strains, which is consistent with previous reports (4, 14). The number of pathogenic isolates collected in El Barco de Avila was small, probably because nonpathogenic opportunistic invaders outnumber the pathogenic strains once the bean plant is severely diseased. Despite this small number, the pathogenic strains collected in El Barco de Avila were placed in three new VCGs, although the analysis of IGS and EK diversity showed a high level of genetic similarity. The correlation between VCG and formae speciales and even races, as proposed by Puhalla (30), is based on the as-
sumption that vegetatively compatible isolates are clonally related, which in turn assumes the absence of the sexual cycle. This life cycle is thought to be the case for F. oxysporum, but high levels of VCG diversity, such as those we report for El Barco de Avila isolates, could be indicative of sexual exchange. The mechanisms thought to be responsible for concerted evolution (unequal crossing over and gene conversion) tend to homogenize the ribosomal DNA genotype within each individual, and fixation of this region occurs within populations of sexually reproducing species (12, 19). We found four different IGS haplotypes in El Barco de Avila and two more among the F. oxysporum f. sp. phaseoli isolates from the Americas. Previously, Appel and Gordon (4) identified 13 IGS haplotypes among a sample of 56 F. oxysporum isolates collected in Maryland and California. Thus, the diversity of IGS haplotypes within F. oxysporum suggests that sexual reproduction is infrequent or absent in this fungus. We also found high levels of EK variation among the nonpathogenic strains. Previous reports noted EK diversity for both pathogenic and nonpathogenic F. oxysporum isolates (24–26). As the extent of chromosomal polymorphism is inversely correlated to the frequency of meiosis, both EK and IGS polymorphism patterns support clonal propagation as the predominant mode of reproduction in F. oxysporum. The degree of diversity observed also could occur if the strains now identified as F. oxysporum belonged to different biological species. By using a combined approach to determine genetic relationships, we have shown that vegetative compatibility is an excellent indicator of a close genetic relationship but that vegetative incompatibility need not indicate a large genetic difference. Within each VCG analyzed, we found no IGS haplotype variation and relatively little variation in the EK patterns, which supports the hypothesis that vegetatively compatible strains are clonal. Several VCGs may be grouped on the basis of IGS haplotype and EK pattern. Such grouping suggests a close genetic relationship between the strains in VCGs 0166,
FUSARIUM OXYSPORUM FROM COMMON BEAN FIELDS IN SPAIN
VOL. 65, 1999
0167, and 0168 and strain AS3. This relationship does not correlate with pathogenicity, as the group includes three nonpathogenic strains. The finding of nonpathogenic strains that are vegetatively compatible with pathogenic strains is not new (3, 16, 17), but the pathogenic and nonpathogenic strains can be resolved by other techniques (5). In our case, the identity of the IGS haplotype and the similarity of the EK patterns suggest that FOP-SP1, FOP-SP2, AB82, and AB92 are very closely related. Identification of differences between AB82 and AB92 and the pathogens could shed light on the genetic basis of pathogenicity in F. oxysporum. Given the widespread occurrence of nonpathogenic F. oxysporum strains, it seems reasonable that pathogenic forms of this fungus may have evolved from nonpathogenic ancestors. If the mutations capable of inducing disease in a susceptible host are very rare events, then currently known pathogens within a forma specialis may trace their origins to a single ancestral genotype (monophyletic origin). Alternatively, the appearance of pathogenic genotypes may be more frequent, and then pathogens within a forma specialis may be affiliated with an array of distinct lineages (polyphyletic origin). If all of the pathogenic strains which belong to F. oxysporum f. sp. phaseoli have evolved from a common ancestor, then members of pathogenic VCGs should be more closely related to one another than to members of exclusively nonpathogenic VCGs. The pathogenic and nonpathogenic strains collected in El Barco de Avila differ from the pathogenic strains from the Americas in VCG, IGS haplotype, and EK pattern. These genetic differences suggest that pathogenicity to bean has developed more than once and that F. oxysporum f. sp. phaseoli has a polyphyletic origin. Similar explanations have been proposed for F. oxysporum f. sp. dianthi (2) and F. oxysporum f. sp. cubense (23). In contrast, the formae speciales causing wilt diseases on crops in the Cucurbitaceae seem to be closely related and thus monophyletic in origin (for a review, see reference 15). In a sexually reproducing population with a segregant pathogenicity locus, heterogeneity could also be maintained. The suggested independent origin of pathogenicity in F. oxysporum f. sp. phaseoli has practical implications for work on this disease. The most effective control of fusarium wilt of the common bean relies on the use of resistant cultivars, several of which carry single dominant resistance genes (31). This kind of resistance is easy to transfer to susceptible cultivars; however, its effectiveness depends on the race structure and genetic diversity of the pathogens. If F. oxysporum f. sp. phaseoli is as diverse as our results suggest, then simple forms of resistance may be of limited use. To increase the probability of developing cultivars resistant to genetically distinct populations of the pathogen, new hybrids should be screened against different F. oxysporum f. sp. phaseoli strains. ACKNOWLEDGMENTS This research was supported by grant SC95-002-C2-1 from the Spanish Instituto Nacional de Investigaciones Agrarias (MAPA) and grants SA01 and SA70/94 from Junta de Castilla y Leo ´n. F. M. Alves-Santos was the recipient of a fellowship from the Instituto Nacional de Investigaciones Agrarias (MAPA). We thank M. A. Pastor-Corrales and H. F. Schwartz for providing fungal isolates, F. J. Tello Marquina for assistance in fungal determinations, and E. A. Iturriaga for critically reading the manuscript. REFERENCES 1. Alabouvette, C. 1993. Recent advances in biological control of Fusarium wilts. Pestic. Sci. 37:365–373. 2. Aloi, C., and R. P. Baayen. 1993. Examination of the relationships between vegetative compatibility groups and races in Fusarium oxysporum f. sp. dianthi. Plant Pathol. 42:839–850.
3339
3. Appel, D. J., and T. R. Gordon. 1994. Local and regional variation in populations of Fusarium oxysporum from agricultural field soils. Phytopathology 84:786–791. 4. Appel, D. J., and T. R. Gordon. 1995. Intraspecific variation within populations of Fusarium oxysporum based on RFLP analysis of the intergenic spacer region of the rDNA. Exp. Mycol. 19:120–128. 5. Appel, D. J., and T. R. Gordon. 1996. Relationships among pathogenic and nonpathogenic isolates of Fusarium oxysporum based on the partial sequence of the intergenic spacer region of the ribosomal DNA. Mol. Plant-Microbe Interact. 9:125–138. 6. Armstrong, G. M., and J. K. Armstrong. 1981. Formae speciales and races of Fusarium oxysporum causing wilt diseases, p. 391–399. In P. E. Nelson, T. A. Toussoun, and R. Cook (ed.), Fusarium: diseases, biology and taxonomy. The Pennsylvania State University Press, University Park. 7. Benlloch, M., and J. D. Can ˜ izo. 1926. La enfermedad de las alubias en Barco de Avila. Bol. Estacio ´n Pathol. Vegetal 1:2–7. 8. Boehm, E. W. A., R. C. Ploetz, and H. C. Kistler. 1994. Statistical analysis of electrophoretic karyotype variation among vegetative compatibility groups of Fusarium oxysporum f sp cubense. Mol. Plant-Microbe Interact. 7:196–207. 9. Correll, J. C., C. J. R. Klittich, and J. F. Leslie. 1987. Nitrate nonutilizing mutants of Fusarium oxysporum and their use in vegetative compatibility tests. Phytopathology 77:1640–1646. 10. Correll, J. C., C. J. R. Klittich, and J. F. Leslie. 1989. Heterokaryon selfincompatibility in Gibberella fujikuroi (Fusarium moniliforme). Mycol. Res. 93:21–27. 11. Dı´az-Mı´nguez, J. M., F. M. Alves-Santos, E. P. Benito, and A. P. Eslava. 1996. Fusarium wilt of common bean in the Castilla y Leo ´n region of Spain. Plant Dis. 80:600. 12. Dover, G. A. 1986. Molecular drive in multigene families: how biological novelties arise, spread and are assimilated. Trends Genet. 2:159–165. 13. Edel, V., C. Steinberg, N. Gautheron, and C. Alabouvette. 1997. Populations of nonpathogenic Fusarium oxysporum associated with roots of four plant species compared to soilborne populations. Phytopathology 87:693–697. 14. Fiely, M. B., J. C. Correll, and T. E. Morelock. 1995. Vegetative compatibility, pathogenicity and virulence diversity of Fusarium oxysporum recovered from spinach. Plant Dis. 79:990–993. 15. Gordon, T. R., and R. D. Martyn. 1997. The evolutionary biology of Fusarium oxysporum. Annu. Rev. Phytopathol. 35:111–128. 16. Gordon, T. R., and D. Okamoto. 1992. Population structure and the relationship between pathogenic and nonpathogenic strains of Fusarium oxysporum. Phytopathology 82:73–77. 17. Gordon, T. R., and D. Okamoto. 1992. Variation within and between populations of Fusarium oxysporum based on vegetative compatibility and mitochondrial DNA. Can. J. Bot. 70:1211–1217. 18. Graham, G. C., P. Mayers, and R. J. Henry. 1994. A simplified method for the preparation of fungal genomic DNA for PCR and RAPD analysis. Biotechniques 16:48–50. 19. Hillis, D. M., and M. T. Dixon. 1991. Ribosomal DNA: molecular evolution and phylogenetic inference. Q. Rev. Biol. 66:411–453. 20. Katan, T., and J. Katan. 1988. Vegetative-compatibility grouping of Fusarium oxysporum f. sp. vasinfectum from tissue and the rhizosphere of cotton plants. Phytopathology 78:852–855. 21. Kendrick, J. B., and W. C. Snyder. 1942. Fusarium yellows of beans. Phytopathology 32:1010–1014. 22. Kistler, H. C., C. Alabouvette, R. P. Baayen, S. Bentley, D. Brayford, A. Coddington, J. Correll, M.-J. Daboussi, K. Elias, D. Ferna ´ndez, T. R. Gordon, T. Katan, H. G. Kim, J. F. Leslie, R. D. Martyn, Q. Migheli, N. Y. Moore, K. O’Donnell, R. C. Ploetz, M. A. Rutherford, B. Summerell, C. Waalwijk, and S. Woo. 1998. Systematic numbering of vegetative compatibility groups in the plant pathogenic fungus Fusarium oxysporum. Phytopathology 88:30–32. 23. Koenig, R. L., R. C. Ploetz, and H. C. Kistler. 1997. Fusarium oxysporum f. sp. cubense consists of a small number of divergent and globally distributed clonal lineages. Phytopathology 87:915–923. 24. Migheli, Q., T. Berio, and M. L. Gullino. 1993. Electrophoretic karyotypes of Fusarium spp. Exp. Mycol. 17:329–337. 25. Migheli, Q., T. Berio, M. L. Gullino, and A. Garibaldi. 1995. Electrophoretic karyotype variation among pathotypes of Fusarium oxysporum f. sp. dianthi. Plant. Pathol. 44:308–315. 26. Mills, D., and K. McCluskey. 1990. Electrophoretic karyotypes of fungi: the new cytology. Mol. Plant-Microbe Interact. 3:351–357. 27. Pastor-Corrales, M. A., and G. S. Abawi. 1987. Reactions of selected bean germplasms to infection by Fusarium oxysporum f. sp. phaseoli. Plant Dis. 71:990–993. 28. Paulitz, T. C., C. S. Park, and R. Baker. 1987. Biological control of Fusarium wilt of cucumber with nonpathogenic isolates of Fusarium oxysporum. Can. J. Microbiol. 33:349–353. 29. Postma, J., and H. Rattink. 1992. Biological control of Fusarium wilt of carnation with a nonpathogenic isolate of Fusarium oxysporum. Can. J. Bot. 70:1199–1205. 30. Puhalla, J. E. 1985. Classification of strains of Fusarium oxysporum on the basis of vegetative compatibility. Can. J. Bot. 63:179–183.
3340
ALVES-SANTOS ET AL.
31. Salgado, M. O., H. F. Schwartz, and M. A. Brick. 1995. Inheritance of resistance to a Colorado race of Fusarium oxysporum f. sp. phaseoli in common beans. Plant Dis. 79:279–281. 32. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
APPL. ENVIRON. MICROBIOL. 33. Vela ´squez-Valle, R., H. F. Schwartz, and J. M. Dı´az-Mı´nguez. 1997. Pathogenicity of Fusarium oxysporum f. sp. phaseoli isolates from Spain to Phaseolus vulgaris. Plant Dis. 81:312. 34. Woo, S. L., A. Zoina, G. del Sorbo, M. Lorito, B. Nanni, F. Scala, and C. Noviello. 1996. Characterization of Fusarium oxysporum f. sp. phaseoli by pathogenic races, VCGs, RFLPs, and RAPD. Phytopathology 86:966–973.
