Development of Contamination-Free Restriction Fragment Length Polymorphism Probes for the Obligate Biotroph Peronospora tabacina, an Oomycete Causing Blue Mold of Tobacco

ABSTRACT Peronospora tabacina is an obligately parasitic oomycete that causes blue mold, a devastating disease of tobacco. Genetic studies of this pathogen have been hampered by the lack of molecular markers. We generated a set of molecular markers for P. tabacina by collecting sporangiospores from infected tobacco leaves, extracting spore DNA, and cloning it in a plasmid vector. The resulting clones were then used to probe DNA from a collection of P. tabacina isolates to survey for polymorphisms. Most probes gave unexpected hybridization patterns with signal intensities that varied significantly from one DNA sample to another or between different DNA preparations of the same isolate. These results indicated that certain DNA preparations contained DNA from a source other than P. tabacina, which in turn suggested that some probes might have been derived from contaminating organisms present in the spore suspensions. Therefore, we characterized the inserts of several recombinant plasmids to determine their origins. Sequence analysis revealed that several of the inserts encoded peptides with similarity to bacterial proteins, suggesting that they were derived from bacterial contaminants. Of the remaining clones, five exhibited similarity to retroelements, one resembled eukaryotic helicase genes, and nine had no similarity to sequences in the databases. These were postulated to be true P. tabacina DNA clones. Verification of the origin of each probe was achieved by filtering a spore suspension, extracting DNA from the retentate and filtrate, and probing Southern blots of these DNA samples. These experiments confirmed the probe origins predicted by sequence analysis, resulting in the generation of 20 different restriction fragment length polymorphism probes that are specific for P. tabacina DNA. These probes should enable identification of reliable genetic markers for population studies of the blue mold organism.     

Genetic and pathogenic variation of Xanthomonas axonopodis pv. manihotis in Venezuela

DNA polymorphism and variation in virulence of Xanthomonas axonopodis pv . manihotis (Xam), the causal agent of cassava bacterial blight, were studied within a pathogen population from Venezuela. Collections were made in several fields at different sites within an edaphoclimatic zone where cassava is a major crop. DNA polymorphism was assessed by RFLP analysis, using an Xam plasmidic DNA sequence ( pth B) as a probe to determine the relatedness of 91 Venezuelan isolates. A high degree of polymorphism existed among the isolates, whether collected from the same or different fields. Based on a multiple correspondence analysis, the Xam population was distributed into eight clusters and no correlation was observed between genetic diversity and geographic origin. One set of haplotype strains representing the range of variability detected in Venezuela was further characterized by another RFLP analysis using two repetitive genomic probes (pBS6 and pBS8) to establish the usefulness of these probes and their complementarity with the pth B probe. Variation for virulence was observed in the Xam Venezuelan collection by inoculating a set of cassava cultivars with 28 isolates of the pathogen, each representing a haplotype. Understanding the genetic and pathogenic variation in the pathogen population is useful for designing cassava bacterial blight management strategies.     

Diversity of Virulence in Phytophthora parasitica on Tobacco, as Reflected by Nuclear RFLPs

ABSTRACT A worldwide collection of P. parasitica isolates was investigated for the ability to infect tobacco and tomato, as related to elicitin production. Elicitin was produced by all nontobacco isolates, and nonproducing strains all were isolated from tobacco. In addition, producing strains were isolated from tobacco and coexisted with nonproducing (TE ) strains. Elicitin production generally was associated with low virulence on tobacco and frequent pathogenicity on tomato, whereas TE isolates generally were highly virulent and specialized to tobacco. Analysis of both mitochondrial and nuclear DNA restriction fragment length polymorphisms indicated, for the first time, that black shank isolates can be distinguished from other P. parasitica isolates on the basis of genetic criteria. Our results suggest that severe black shank is caused by a limited number of TE strains that have been disseminated by clonal evolution. Mutations in the TE phenotype seem to have arisen independently in several genetic backgrounds and distinct geographic areas. The fortuitous absence of elicitin production has precluded population replacements in areas of intensive tobacco cultivation. Thus, monitoring the loss of elicitin production in developing tobacco areas should be considered in disease management.     

Evidence of parasexual exchange of DNA in the rice blast fungus challenges its exclusive clonality

ABSTRACT We applied DNA markers to determine whether parasexual recombination may contribute to the extreme genetic diversity and variability observed in Magnaporthe grisea, the causal agent of rice blast disease. Dispersed repetitive elements and mapped, low-copy restriction fragment length polymorphism (RFLP) probes were used to detect transfers of DNA between cultured isolates of M. grisea. Low-copy RFLP probes also were used to detect putative recombinants among isolates from well-characterized field populations of the pathogen. Microscopic examination of tufted mycelium between cocultured isolates revealed frequent hyphal fusions. Hyphal tips and conidia were recovered without selection from tufted zones in two separate vegetative pairings involving isolates with dissimilar haplotypes, based on the repetitive element MGR586. Haplotypic changes were observed at a higher frequency in tuft derivatives than in subcultures of each isolate alone. From 136 tuft derivatives analyzed, 5 putative recombinant haplotypes were identified. Introgression was demonstrated with two independent repetitive elements, fosbury and MGR586, as probes on DNA digested with several restriction enzymes. Introgressions were characterized by addition of 1 to 10 MGR586 bands, and 1 to 3 fosbury bands from one parent into the background of the other. Polymorphic single-copy probes were used to analyze putative recombinants. One probe detected an introgression event as predicted by analysis with MGR586. To assess the possible role of parasexual recombination in field populations of the pathogen, isolates in the Philippines previously grouped based on DNA fingerprinting were analyzed with low-copy RFLP markers. Polymorphism in single-copy loci typically was seen between, but not within, putative pathogen lineages. One lineage (designated lineage 4), however, was polymorphic for several probes. For some isolates, alleles at these loci comigrated with alleles characteristic of other lineages, suggesting the transfer of DNA fragments between lineages. One isolate was apparently a merodiploid, carrying an allele typical of lineage 4 plus another allele characteristic of a different lineage. In a survey of isolates from the Indian Himalayas, a merodiploid also was found with single- or low-copy probes. Examination of MGR586 profiles of the putative recombinant and its putative donor strains showed the expected introgression of MGR586 bands. The detection of parasexual DNA exchanges in wild-type strains under unselected conditions and the existence of merodiploids in nature suggest that parasexual recombination occurs in field populations of M. grisea. This raises questions concerning exclusive clonality in the blast fungus.     

Genetic Diversity and the Presence of Two Distinct Groups in Ophiostoma clavigerum Associated with Dendroctonus ponderosae in British Columbia and the Northern Rocky Mountains

ABSTRACT The sapstaining fungal pathogen Ophiostoma clavigerum is associated with the mountain pine beetle (Dendroctonus ponderosae), which is currently the most destructive forest pest in North America. The genetic diversity of O. clavigerum populations collected from five sites in Canada and two sites in the United States was estimated with amplified fragment length polymorphism (AFLP) analysis. Genomic DNA from 170 O. clavigerum isolates was digested with EcoRI and PstI and amplified with six primer sets. A total of 469 AFLP markers consisting of 243 monomorphic and 226 polymorphic loci were scored. The overall genetic diversity of the O. clavigerum population was low (Hs = 0.0531) and the differentiation of the seven O. clavigerum populations was moderate (Phi = 0.143). Genetic distances among the populations were not significantly correlated with geographic distance (r = 0.3235, P = 0.074). Two genetically distinct groups in the O. clavigerum populations were shown by unique AFLP profiles and the unweighted pair group method with arithmetic averages. Further work to characterize biological differences between the two groups will be needed to confirm whether cryptic species are present in the O. clavigerum population.     

Pineapple heart rot isolates from Ecuador reveal a new genotype of Phytophthora nicotianae

Pineapple heart rot disease, caused by Phytophthora nicotianae (syn. P. parasitica), is responsible for significant annual reductions in crop yield due to plant mortality. In Ecuador, new infections arise during the rainy season and increase production costs due to the need for biocontrol and fungicide applications. Studies of P. nicotianae population structure suggest that certain genetic groups are associated with host genera; however, it is not clear how many host‐specific lineages of the pathogen exist or how they are related. The objectives of this study were to determine the level of genetic variation in the P. nicotianae population causing heart rot disease of pineapple in Ecuador and compare the genotypes found on pineapple to those previously reported from citrus, tobacco and ornamentals. Thirty P. nicotianae isolates collected from infected pineapple leaves from four farms were genotyped using nine simple sequence repeat loci. In addition, the DNA sequences of mitochondrial loci cox2 + spacer and trnG‐rns were analysed. Together, these loci supported a single clonal lineage with two multilocus genotypes differing in a single allele and low mitochondrial diversity. This lineage was distinct but closely related to isolates collected from vegetables and ornamentals in Italy. The results support the hypothesis of host specialization of P. nicotianae in intensive cropping systems and contribute to the understanding of population structure of this important pathogen.     

Genetic Relatedness of African and United States Populations of Cercospora zeae-maydis

Two taxonomically identical but genetically distinct sibling species, designated groups I and II, of Cercospora zeae-maydis cause gray leaf spot of maize in the United States. Isolates of the gray leaf spot pathogen from Africa were compared with isolates from the United States by amplified fragment length polymorphism (AFLP) analysis and restriction digests of internal transcribed spacer (ITS) regions and 5.8S ribosomal DNA (rDNA), as well as by morphological and cultural characteristics. The isolates from Africa were morphologically indistinguishable from the U.S. isolates in both groups, but like isolates of group II, they grew more slowly and failed to produce detectable amounts of cercosporin in culture. Analysis of restriction fragments from the ITS and rDNA regions digested with five endonucleases indicated that all of the African isolates shared the profile of the C. zeae-maydis group II population from the eastern United States and, thus, are distinct from the group I population, which is more prevalent in the United States and other parts of the world. Cluster analysis of 85 AFLP loci confirmed that the African and U.S. group II populations were conspecific (greater than 97% average similarity) with limited variability. Among all group II isolates, only 8 of 57 AFLP loci were polymorphic, and none was specific to either population. Thus, although gray leaf spot was reported in the United States several decades prior to the first record in Africa, the relative age of the two populations on their respective continents could not be ascertained with confidence. The absence of C. zeae-maydis group I in our samples from four countries in the major maize-producing region of Africa as well as the greater AFLP haplotype diversity found in the African group II population, however, suggest that Africa was the source of C. zeae-maydis group II in the United States. The overall paucity of AFLP variation in this sibling species further suggests that its origin is recent or that the ancestral population experienced a severe bottleneck prior to secondary migration.     

Single Cystosorus Isolate Production and Restriction Fragment Length Polymorphism Characterization of the Obligate Biotroph Spongospora subterranea f. sp. subterranea

ABSTRACT Spongospora subterranea f. sp. subterranea causes powdery scab in potatoes and is distributed worldwide. Genetic studies of this pathogen have been hampered due, in part, to its obligate parasitism and the lack of molecular markers for this pathogen. In this investigation, a single cystosorus inoculation technique was developed to produce large amounts of S. subterranea f. sp. subterranea plasmodia or zoosporangia in eastern black nightshade (Solanum ptycanthum) roots from which DNA was extracted. Cryopreservation of zoosporangia was used for long-term storage of the isolates. S. subterranea f. sp. subterranea-specific restriction fragment length polymorphism (RFLP) markers were developed from randomly amplified polymorphic DNA (RAPD) fragments. Cystosori of S. subterranea f. sp. subterranea were used for RAPD assays and putative pathogen-specific RAPD fragments were cloned and sequenced. The fragments were screened for specificity by Southern hybridization and subsequent DNA sequence BLAST search. Four polymorphic S. subterranea f. sp. subterranea-specific probes containing repetitive elements, and one containing single copy DNA were identified. These RFLP probes were then used to analyze 24 single cystosorus isolates derived from eight geographic locations in the United States and Canada. Genetic variation was recorded among, but not within, geographic locations. Cluster analysis separated the isolates into two major groups: group I included isolates originating from western North America, with the exception of those from Colorado, and group II included isolates originating from eastern North America and from Colorado. The techniques developed in this study, i.e., production of single cystosorus isolates of S. subterranea f. sp. subterranea and development of RFLP markers for this pathogen, provide methods to further study the genetic structure of S. subterranea f. sp. subterranea.     

Investigating the genetic structure of Phytophthora capsici populations

Phytophthora capsici Leonian is a destructive soilborne pathogen that infects economically important solanaceous, cucurbitaceous, fabaceous, and other crops in the United States and worldwide. The objective of this study was to investigate the genetic structure of 255 P. capsici isolates assigned to predefined host, geographical, mefenoxam-sensitivity, and mating-type categories. Isolates from six continents, 21 countries, 19 U.S. states, and 26 host species were genotyped for four mitochondrial and six nuclear loci. Bayesian clustering revealed some population structure by host, geographic origin, and mefenoxam sensitivity, with some clusters occurring more or less frequently in particular categories. Bayesian clustering, split networks, and statistical parsimony genealogies also detected the presence of non-P. capsici individuals in our sample corresponding to P. tropicalis (n = 9) and isolates of a distinct cluster closely related to P. capsici and P. tropicalis (n = 10). Our findings of genetic structuring in P. capsici populations highlight the importance of including isolates from all detected clusters that represent the genetic variation in P. capsici for development of diagnostic tools, fungicides, and host resistance. The population structure detected will also impact the design and interpretation of association studies in P. capsici. This study provides an initial map of global population structure of P. capsici but continued genotyping of isolates will be necessary to expand our knowledge of genetic variation in this important plant pathogen.     

Sibling species of cercospora associated with gray leaf spot of maize

ABSTRACT Monoconidial isolates of the fungus causing gray leaf spot of maize were obtained from diseased leaves collected throughout the United States and analyzed for genetic variability at 111 amplified fragment length polymorphism (AFLP) loci. Cluster analysis revealed two very distinct groups of Cercospora zeae-maydis isolates. Both groups were found to be relatively uniform internally with an average genetic similarity among isolates of approximately 93 and 94%, respectively. The groups were separated from each other by a genetic distance of approximately 80%, a distance greater than that separating each group from the sorghum pathogen, C. sorghi (67 to 70%). Characteristics and dimensions of conidia and conid-iophores produced on infected plants or nutrient media were unreliable criteria for taxonomic differentiation of isolates composing the two groups of C. zeae-maydis. Nucleotide sequences of 5.8S ribosomal DNA (rDNA) and the internal transcribed spacer (ITS) regions were identical within each group but different between the two groups and different from C. sorghi. Restriction fragment length polymorphisms generated by digestion of the 5.8S rDNA and ITS regions with TaqI readily distinguished each group and C. sorghi. Isolates in one group were generally distributed throughout maize-producing regions of the United States; isolates in the other group were localized in the eastern third of the country. Both types were present in the same fields at some locations. The genetic distance based on AFLP profiles and different ITS nucleotide sequences between the two morphologically indistinguishable groups indicate that they are sibling species. Although it is unlikely that breeding for resistance to gray leaf spot will be confounded by local or regional variation in the pathogen, a vigilant approach is warranted, because two pathogenic species exist with unknown abilities to evolve new pathotypes.     

Genetic variation among populations of Pythium irregulare in southern Australia

Isolates of Pythium irregulare were sampled from seven cereal crops throughout South Australia to determine the extent of genetic diversity within this pathogen and the scale of genetic differentiation among populations. Data derived from 29 individual restriction fragment length polymorphism (RFLP) loci differentiated 54 DNA fingerprints among the 92 isolates analysed. Some isolates had two alleles at several RFLP loci and were scored as heterozygous. One such isolate was selfed in vitro and segregation ratios in the progeny were not significantly different from those expected for allelic variation in a diploid. These data provided evidence that outcrossing occurs within P. irregulare and may contribute to the high level of genetic variation within the species ( D _T = 0·502). Allelic frequencies were significantly different among all seven populations and G _ST values showed significant genetic differentiation between populations. The average genetic identity among populations was low and hierarchical cluster analysis provided no clear evidence that populations formed geographically related groups. These analyses indicate low levels of interpopulation gene flow within P. irregulare and imply that population differentiation results from genetic drift.     

Genetic Structure and Temporal Dynamics of a Colletotrichum graminicola Population in a Sorghum Disease Nursery

ABSTRACT Restriction fragment length polymorphisms (RFLPs) were used to study the population genetics of Colletotrichum graminicola (= C. sublineolum), the causal agent of sorghum anthracnose. Screening of 80 anonymous probes from a genomic library detected polymorphisms in 81% of 299 probe-enzyme combinations among nine international isolates. Seven single- or low-copy probes were used to study a collection of 411 isolates sampled during 1991 to 1993 from a sorghum disease nursery in Georgia. Nei's gene diversity was moderately high, with = 0.215 on average, while genotypic diversity was extremely low with an average genotypic diversity value of G = 1.513. Only nine multilocus haplotypes were identified, with one haplotype being present at a frequency of approximately 80% each year. Two other haplotypes were found at significant frequencies (4 to 10%). Allele and haplotype frequencies did not differ over the 3 years, indicating that this population was stable. Our findings suggest that genetic drift and gene flow were not major contributors to genetic structure, while asexual reproduction had a significant effect.     

Genetic structure of Phaeosphaeria nodorum populations in the north-central and midwestern United States

Stagonospora nodorum blotch, caused by Phaeosphaeria nodorum, is considered one of the most destructive foliar diseases of wheat in the United States. However, relatively little is known about the population biology of this fungus in the major wheat-growing regions of the central United States. To rectify this situation, 308 single-spore isolates of P. nodorum were analyzed from 12 populations, five from hard red spring wheat cultivars in Minnesota and North Dakota and seven from soft red winter wheat in Indiana and Ohio. The genetic structure of the sampled populations was determined by analyzing polymorphisms at five microsatellite or simple-sequence repeat (SSR) loci and the mating type locus. Although a few clones were identified, most P. nodorum populations had high levels of gene (H(S) = 0.175 to 0.519) and genotype (D = 0.600 to 0.972) diversity. Gene diversity was higher among isolates collected from spring wheat cultivars in North Dakota and Minnesota (mean H(S) = 0.503) than in those from winter wheat cultivars in Indiana and Ohio (H(S) = 0.269). Analyses of clone-corrected data sets showed equal frequencies of both mating types in both regional and local populations, indicating that sexual recombination may occur regularly. However, significant gametic disequilibrium occurred in three of the four populations from North Dakota, and there was genetic differentiation both within and among locations. Genetic differentiation between the hard red spring and soft red winter wheat production regions was moderate (F(ST) = 0.168), but whether this is due to differences in wheat production or to geographical variation cannot be determined. These results suggest that sexual reproduction occurs in P. nodorum populations in the major wheat-growing regions of the central United States, and that geographically separated populations can be genetically differentiated, reflecting either restrictions on gene flow or selection.     

Population Genetic Analysis of a Global Collection of Pyrenophora tritici-repentis, Causal Agent of Tan Spot of Wheat

ABSTRACT The work presented here is the first major study to analyze the genetic diversity within the worldwide population of the economically important wheat pathogen Pyrenophora tritici-repentis. The genetic structure of field populations of P. tritici-repentis was determined using amplified fragment length polymorphism markers along with sequence data from the internal transcribed spacer region of the ribosomal DNA. Ninetyseven fungal isolates were collected from naturally infected wheat and wild grass species. The collection of 97 P. tritici-repentis isolates included races 1, 2, 3, 4, 5, ND7, and ND8 and was collected from North America, South America, and Europe. Results show no genetic grouping of fungal races nor do results indicate grouping based on geographic location indicating that the population is preferentially outcrossing in nature and that the introduction and spread of this population is either relatively recent or that there has been a constant worldwide flow of this fungus possibly by seed movement between continents.     

Cocultures of Peronospora tabacina and Nicotiana Species to Study Host-Pathogen Interactions

ABSTRACT Long-term cocultures of the tobacco blue mold pathogen, Peronospora tabacina, with Nicotiana tabacum and N. repanda callus were derived from infected host plant tissue. In this apparently contaminant-free system, sporulation occurred under similar conditions as in intact plants. Sporangia were collected from cocultures and used to complete Koch's postulates. The cocultures were grown under two light regimes. One consisted of 23 h of light followed by 1 h of darkness and the second comprised total darkness. Sporulation occurred frequently in the 23 h light-grown cocultures but resulted in production of abnormal sporangiophores and sporangia. Production of normal sporangiophores and sporangia was achieved by transferring light-grown cocultures to overnight darkness and resulted in necrosis of the callus. Cocultures of Peronospora tabacina with either host species, grown in total darkness, frequently sporulated with minimal necrosis over the course of 1 year. Thus, cocultures should prove useful as a source of Peronospora tabacina over extended periods of time at low risk of pathogen release, for studying the physiology of Peronospora tabacina- Nicotiana interactions, maintaining Peronospora tabacina lines for genetic studies, and providing a reliable source of axenic inoculum for research.     

Genetic variation within and between southern Ontario populations of Sclerotinia homoeocarpa

Restriction fragment length polymorphisms (RFLP) of the intergenic spacer region (IGS) of rDNA and random amplified polymorphic DNA (RAPD) markers were used to survey genetic variability among 181 isolates of Sclerotinia homoeocarpa from Ontario and 10 isolates from Japan. RAPD and IGS-RFLP analyses revealed polymorphisms within and between populations of S . homoeocarpa , distinguishing 151 genotypes. Both types of markers gave similar results in phenetic analysis of genetic distances between populations. Cluster analysis showed that Japanese isolates of S. homoeocarpa were genetically distinct from Ontario isolates, demonstrating significant intraspecific differentiation. An average genetic similarity of 0.66 was found between Japanese isolates. Among Ontario isolates, average genetic similarity was 0.86, and genotypic diversity analysis showed that 49.3% of the total genetic variation observed within Ontario populations occurred among individuals within populations compared to 50.7% between populations. Gametic linkage disequilibrium analysis within Ontario populations revealed an average 15.6% significant nonrandom associations between putative RAPD loci, and that half of the populations showed signs of significant linkage disequilibrium. These results suggest that both clonal propagation and recombination events occurred in local populations of S. homoeocarpa . The high level of genetic similarity between populations and the low levels of intraspecific genetic variation may reflect a small founding population for southern Ontario isolates of S. homoeocarpa .     

Genetic Variation and Virulence on Lr26 in Puccinia triticina

ABSTRACT The genetic relationships between isolates of Puccinia triticina virulent on wheat with the Lr26 resistance gene were studied. The diversity within and between isolates of P. triticina from Israel, Europe, and the United States was determined by virulence on near-isogenic Thatcher lines and by random amplified polymorphic DNA. According to the molecular markers, isolates that were virulent on Lr26 had diversity levels similar to those of Lr26 nonpathogenic isolates. Distances between subpopulations of isolates virulent and avirulent on Lr26 varied and were unrelated to the Lr26 virulence phenotype. Cluster analysis suggested four groups, three of which were closely associated with the geographical origin of the isolates-Israel, the United States, and Europe. All four groups included both Lr26 virulent and avirulent pathotypes. The results showed that Lr26 virulent rust pathotypes are as genetically dissimilar as the rest of the population. The cluster analysis showed that the rust population in Israel includes at least two different subpopulations, both of which contain Lr26 virulent and Lr26 avirulent isolates.     

Population structure of dogwood anthracnose fungus

ABSTRACT Dogwood anthracnose, caused by Discula destructiva, affects several native dogwood species in North America, especially flowering dogwood in the east and Pacific dogwood in the west. The fungus behaves as a recently introduced plant pathogen under episodic selection. Two distinct disjunct groups of fungal isolates corresponding to eastern and western groups were detected by amplified fragment length polymorphisms and sequences of the intergenic spacer (IGS) of the nuclear ribosomal DNA, translation elongation factor-1alpha, and beta-tubulin genes. Of 20 genotypes identified among 72 isolates, 17 genotypes were from the eastern United States (n = 50), but only three were present among the western isolates (n = 22), indicating that the eastern population may be more diverse. Most eastern and western isolates belonged to a few widespread clones, and the genetic variability of this apparently asexual fungus was remarkably low compared with that of many other asexual fungi. We conclude that D. destructiva is still under intense selection pressure and that episodic selection may still be in effect. The New York City area, a possible epidemic center in the east, had relatively higher genetic variability than samples from other areas.     

Genetic structure of <Emphasis Type="Italic">Fusarium verticillioides</Emphasis> populations isolated from maize in Argentina

Fusarium verticillioides (sexual stage Gibberella moniliformis) is a common fungal pathogen of maize worldwide that also produces fumonisin mycotoxins. Populations of this fungus can be diverse with respect to neutral and selectable genetic markers. We used vegetative compatibility groups (VCGs) and amplified fragment length polymorphisms (AFLPs) to evaluate the genetic structure of three F. verticillioides populations from commercial maize fields in Argentina. Based on work with similar populations from outside South America, we expected individuals within the populations to be genetically diverse, that genotypic variation would be distributed in a manner consistent with random mating, and that populations from different locations would be genetically indistinguishable from one another. We analysed 62 AFLP loci for 133 fungal isolates. All three populations were genotypically diverse but genetically similar and potentially part of a larger, randomly mating population, with significant genetic exchange occurring between the three subpopulations. There was no evidence for linkage disequilibrium at P = 0.05. The low values of G _ST , the lack of frequent private alleles, and the lack of a systemic pattern of linkage disequilibrium all suggest that sexual reproduction is sufficiently common in F. verticillioides and that the dispersal of strains is sufficiently efficient for the population of F. verticillioides in the main maize growing region to be a single randomly mating population with no detectable genetic subdivision. Thus differences in disease and/or toxin production observed in this region are best attributed to differences other than the genetic composition of the population.     

Fusarium oxysporum f. sp. cubense Consists of a Small Number of Divergent and Globally Distributed Clonal Lineages

ABSTRACT A worldwide collection of Fusarium oxysporum f. sp. cubense was analyzed using anonymous, single-copy, restriction fragment length polymorphism (RFLP) loci. Several lines of evidence indicated that this pathogen has a clonal population structure. Of the 165 isolates examined, only 72 RFLP haplotypes were identified, and nearly half the isolates were represented by the five most common haplotypes. Individuals with identical haplotypes were geographically dispersed, and clone-corrected tests of gametic disequilibrium indicated significant nonrandom association among pairs of alleles for 34 of 36 loci tested. Parsimony analysis divided haplotypes into two major branches (bootstrap value = 99%) that together contained eight clades supported by significant bootstrap values. With the exception of two isolates, all isolates within a vegetative compatibility group were in the same clade and clonal lineage. Clonal lineages were defined by isolates having coefficients of similarity between 0.94 and 1.00. Ten clonal lineages were identified, and the two largest lineages had pantropical distribution. Minor lineages were found only in limited geographical regions. Isolates composing one lineage (FOC VII) may represent either an ancient genetic exchange between individuals in the two largest lineages or an ancestral group. The two largest lineages (FOC I and FOC II) and a lineage from East Africa (FOC V) are genetically distinct; each may have acquired the ability to be pathogenic on banana independently.