This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Martin, F. N. Articles by Tooley, P. W. Search for Related Content PubMed Articles by Martin, F. N. Articles by Tooley, P. W. Agricola Articles by Martin, F. N. Articles by Tooley, P. W.

Phylogenetic relationships among Phytophthora species inferred from sequence analysis of mitochondrially encoded cytochrome oxidase I and II genes

     USDA-ARS, 1636 East Alisal St., Salinas, California 93905

Paul W. Tooley

     USDA-ARS, Foreign Disease-Weed Science Research Unit, 1301 Ditto Ave., Fort Detrick, Maryland 21702

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 

The phylogenetic relationships of 51 isolates representing 27 species of Phytophthora were assessed by sequence alignment of 568 bp of the mitochondrially encoded cytochrome oxidase II gene. A total of 1299 bp of the cytochrome oxidase I gene also were examined for a subset of 13 species. The cox II gene trees constructed by a heuristic search, based on maximum parsimony for a bootstrap 50% majority-rule consensus tree, revealed 18 species grouping into seven clades and nine species unaffiliated with a specific clade. The phylogenetic relationships among species observed on cox II gene trees did not exhibit consistent similarities in groupings for morphology, pathogenicity, host range or temperature optima. The topology of cox I gene trees, constructed by a heuristic search based on maximum parsimony for a bootstrap 50% majority-rule consensus tree for 13 species of Phytophthora, revealed 10 species grouping into three clades and three species unaffiliated with a specific clade. The groupings in general agreed with what was observed in the cox II tree. Species relationships observed for the cox II gene tree were in agreement with those based on ITS regions, with several notable exceptions. Some of these differences were noted in species in which the same isolates were used for both ITS and cox II analysis, suggesting either a differential rate of evolutionary divergence for these two regions or incorrect assumptions about alignment of ITS sequences. Analysis of combined data sets of ITS and cox II sequences generated a tree that did not differ substantially from analysis of ITS data alone, however, the results of a partition homogeneity test suggest that combining data sets may not be valid.

Key words: cox I, cox II, ITS, phylogeny

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Phytophthora is a complex genus containing about 60 described species that occupy a variety of terrestrial and aquatic habitats (Erwin and Ribeiro 1996). Although oomycetes share many morphological features in common with fungi, evolutionarily they are more closely related to chromophyte algae and plants than to eumycotan fungi (Förster et al 1990a, Knoll 1992, Baldauf and Palmer 1993, Wainright et al 1993, Bhattacharya and Stickel 1994, Leipe et al 1994, 1996, Silberman et al 1996, Weerakoon et al 1998).

Historically speaking, a range of morphological and physiological criteria has been used to classify members of this genus (Stamps et al 1990, Waterhouse 1963), including sporangial structure, antheridial form, host specificity and breeding system (homothallic or heterothallic). However, in light of growing evidence from molecular and ecological studies, it has become apparent that taxonomic groupings, based solely on morphological criteria, might be artificial (Brasier 1991, Brasier and Hansen 1992). The congruence between taxonomic inference derived from molecular data and that based on classical morphological taxonomy is a topic of interest in current studies. Molecular tools used in phylogenetic studies of oomycetes have included analysis of large and small subunit ribosomal RNA genes (rRNA; van de Peer et al 1996, Förster et al 1990a), mitochondrial DNA (Förster et al 1988, Förster and Coffey 1993), and sequence analysis of the internal transcribed spacer (ITS) regions of the rRNA genes (Förster et al 2000, Cooke and Duncan 1997, Crawford et al 1996). Recently, Cooke et al (2000) reported an ITS-based molecular phylogeny, which included 50 Phytophthora species, and is the most comprehensive to date. It was re-emphasized by Cooke et al (2000) that the six taxonomic groupings of Waterhouse (1963) do not represent natural assemblages because individual ITS clades contained taxa from multiple Waterhouse groups. Similarly, Förster et al (2000), based on ITS1 sequence analysis, concluded that the morphological characters used in Phytophthora taxonomy are of limited value in deducing phylogenetic relationships because they exhibit convergent evolution. Data using additional molecular markers are needed for comparison with ITS-based data, particularly in corroborating and/or clarifying the taxonomy of groups that cannot be resolved based on ITS data. For example, species such as P. infestans, P. mirabilis, and P. phaseoli are poorly resolved with ITS data while P. fragariae var. fragariae and P. fragariae var. rubi are not differentiated at all.

The cytochrome c oxidase subunit II (cox II) gene has been used previously in studies of Peronosporomycete phylogeny (Hudspeth et al 2000) and phylogenetic studies of 24 Pythium species (Martin 2000). The cox II gene is mitochondrially encoded and, in contrast to the ITS region, represents a coding region rather than intergenic region. Analysis of cox II grouped Pythium species into three major clades that were reflective of sporangial morphology (Martin 2000). The objective of this investigation was to characterize the evolutionary relationship among Phytophthora species representing different morphological groupings in the genus (Brasier 1983, 1991), with gene sequences from cox I and cox II genes. The relationship between specific morphological features and phylogeny will be examined in an attempt to identify features that can be used for deducing evolutionary groupings among species. Last, the cox gene phylogeny will be compared with the ITS phylogeny to clarify the relationships among species poorly resolved by ITS data and to evaluate congruence of phylogenetic results obtained with mitochondrial gene sequences compared to nuclear encoded spacer sequences.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Phytophthora cultures – The cultures used in this study are listed in Table I. Isolates of Phytophthora arecae, P. cryptogea, P. drechsleri, P. erythroseptica, P. lateralis, P. megasperma, and P. syringae included in this study (Table I) also were used in the ITS rDNA phylogenetic study of Cooke et al (2001). Cultures were grown on Rye A medium (Caten and Jinks 1968) at 20 C in darkness and maintained in liquid nitrogen for long-term storage (Tooley 1988).

Templates of the cox I and II gene cluster were amplified for sequencing by PCR using primers FM 75 (dCCTTGGCAATTAGGATTTCAAGAT) or FM 82 (dTTGGCAATTAGGTTTTCAAGATCC) and FM 77 (dCACCAATAAAGAATAACCAAAAATG) (Fig. 1). While these primers worked well for nearly all isolates examined, FM 77 did not work well for amplification of P. capsici, P. colocasiae or P. citricola (all clustered on Clade 2 in Fig. 2). However, primer FM 83 (dCTCCAATAAAAAATAACCAAAAATG) and FM 75 did amplify DNA from these species. Amplification reactions were done in 50 µL and contained about 50 ng of DNA, 0.5 µM final concentration of each primer, 5 µL of 10x buffer, 100 µM of each dNTP, 2 mM MgCl₂, and 3 units of Taq polymerase (Promega, Madison, Wisconsin). Amplifications were done with an Eppendorf (Westbury, NY) Mastercycler Gradient Thermalcycler with these run parameters: one cycle at 95 C for 2 min; 35 cycles of 1 min annealing at 56 C, 2 min extension at 72 C, and 1 min denaturation at 94 C; followed by one extension cycle at 72 C for 10 min. Before purifying the templates for sequencing, all amplifications were separated on a 1.5% agarose gel to check template concentration and purity.

View larger version (12K): [in this window] [in a new window]    FIG. 1. Organization of the mitochondrially encoded cox I and II gene cluster in Phytophthora and location of the PCR primers that were used for template amplification and sequencing reactions

View larger version (40K): [in this window] [in a new window]    FIG. 2. Phylogenetic relationships among Phytophthora species using cox II DNA sequence data, based on maximum parsimony inferred by a heuristic tree search. Fifty percent majority-rule consensus tree; the numbers above the nodes are the percentage of the trees from bootstrap analysis (1000 replications) that support the observed topography (values above 60% indicated). Values below the nodes and preceded by DI- are decay indices calculated by AutoDecay (v 4.0). Of the 568 total characters, 378 were constant, 64 were variable and parsimony uninformative, and 126 were parsimony informative. Tree length = 623, consistency index (CI) = 0.438, homoplasy index (HI) = 0.562, retention index = 0.682, rescaled consistency index = 0.299. Isolates marked with an (*) were the same cultures that were used by Cooke at al (2000) to infer rDNA ITS-based phylogenetic trees

All sequencing was done by the DNA Sequencing Laboratory of the Interdisciplinary Center for Biotechnological Research of the University of Florida at Gainesville, using ABI 373a automated sequencers (Applied Biosystems, Foster City, CA). Templates were sequenced in both directions with the primers used for amplification, as well as primers FM78 (dACAAATTTCACTACATTGTCC), FM79 (dGGACAATGTAGTGAAATTTGT) and FM 80 (dAATATCTTTATGATTTGTTGAAA) for cox II gene sequencing and FM 84 (dTTTAATTTTTAGTGCTTTTGC), FM 85 (dAACTTGACTAATAATACCAAA), and FM 50 (dGTTTACTGTTGGTTTAGATG) for sequencing the cox I gene (Fig. 1). Primers were synthesized by Invitrogen Corp. (Carlsbad, CA).

Data analysis – Overlapping sequences from each sequencing primer and the sequences from opposite strands were aligned with the computer program Omiga 1.1 (Accelrys Inc, Manassas, VA). If any regions of sequence ambiguity were observed, the original output files of complementary strands from the ABI 373a sequencer were compared for correction and the sequencing reactions were repeated if necessary. A total of 568 bp was used for phylogenetic analysis with the cox II gene, which included bases 94 to 661 of the gene. For cox I, 1299 bp was used, which represented bases 1 to 1299 of the gene. Sequence alignments and phylogenetic analyses were done with PAUP ver. 4.0b8. Decay indices were determined with AutoDecay (v 4.0) of Torsten Eriksson (1999; http:/www.bergianska.sepersonal/TorstenE/). Phylogenetic relationships among Phytophthora spp., using DNA sequence data, were inferred by maximum-parsimony (MP) analysis with a heuristic tree search and Neighbor-joining (NJ) tree reconstruction, using the Kimura 2-K parameter correction method. Heuristic searches were performed with MULPARS on, steepest-descent option off, simple addition of sequences and TBR branch swapping. Maximum-likelihood (ML) analysis with a heuristic tree search was done with the number of substitution types set at 2, transition/transversion ratio empirically determined from the data; a molecular clock was not enforced with the balance of the settings corresponding to the HKY85 model. To determine support for the various clades of the trees, the analysis was bootstrapped with the number of replicates indicated in the figure legends. The outgroup for the analysis was Saparomyces elongates; sequence data for Peronophythora litchii and Lagenidium giganteum were included as well (GenBank AF086698, AF086697, and AF086700, respectively; Hudspeth et al 2000). Data for an isolate of Pythium aphanidermatum from Mexico (1987-61) also was included in the analysis, using cox II data previously reported (Martin 2000; GenBank AF196579) and cox I data obtained in this study. Phylogenetic inferences based on ITS and 5.8S rDNA sequence data were based on the results of Cooke et al (2000) with the alignments retrieved from TreeBASE (accession M751). The same region for Pythium aphanidermatum (GenBank AF271227) was included in the analysis as an outgroup. The relative alignment for the Phytophthora sequences remained the same with a few modifications to improve alignment and with gaps introduced across all species of this genus to accommodate the alignment with sequences of Pythium aphanidermatum. DNA sequence data obtained in this study have been deposited in GenBank (Table I), and the results of this analysis have been deposited in TreeBASE (S774).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Sequence analysis – Gaps were not observed in the cox II or cox I sequence alignments, and sequences generally were conserved well within a species but were divergent among species. Sequence distance measurements within a species, using the Kimura 2-parameter correction method, ranged from 0 to 4.9%. Examination of multiple isolates of five species (P. citricola, P. colocasiae, P. fragariae, P. hibernalis and P. mirabilis) revealed no intraspecific variation in the gene sequence. However, intraspecific variation was observed in nine other species that were examined. In six of the species (P. cactorum, P. drechsleri, P. erythroseptica, P. ilicis, P. infestans and P. nicotianae), variation was low among isolates (<0.5%), but higher rates were observed among isolates of P. palmivora (0 to 1.1%), P. megakarya (3.4%) and P. megasperma (0 to 4.9%). Sequence divergence between Phytophthora spp. and Pythium aphanidermatum ranged from 9.6 to 16.5%. The estimated transition/transversion ratio for maximum-likelihood analysis of the cox II gene was determined to be 0.606, with nucleotide frequencies for A, C, G and T of 0.318, 0.110, 0.167 and 0.405, respectively. For cox I, this was determined to be 0.602, 0.280, 0.121, 0.174, and 0.424, respectively.

Phylogenetic relationships based on cox II DNA sequences – The topology of cox II gene trees, constructed by a heuristic search based on maximum parsimony for a bootstrap 50% majority-rule consensus tree, revealed 18 species grouping into seven clades and nine species unaffiliated with specific clades (Fig. 2). Species grouping did not always reflect morphological groupings (Table II). For example, Clade 2 contained species in morphological groups II (P. capsici), III (P. citricola) and IV (P. colocasiae). Likewise, species within a specific morphological group did not always cluster together. Clade 1 contained species in morphological Group VI (P. drechsleri, P. erythroseptica and P. cryptogea), but members of this morphological group also were present on Clade 5 (P. gonapodyides) and not associated with a specific clade (P. cinnamomi). Species of morphological Group II were found on clades 2 and 7 and not associated with specific clades (P. nicotianae, P. boehmeriae, P. megakarya and P. heveae). Species of morphological Group IV were found on clades 2, 4 and not associated with specific clades (P. hibernalis and P. ilicis). Species of morphological Group V were found on clades 5, 6 and not associated with specific clades (P. lateralis). The tree was identical when this analysis was rerun with cox II sequence data of Pythium aphanidermatum added, with P. aphanidermatum clustering with Lagenidium giganteum (data not shown). Another genus included for outgroup comparison (Peronophythora litchii) clustered with P. palmivora and P. arecae on Clade 7, although the bootstrap support was not strong (56%).

With a few minor differences, the relationships among species in the MP tree (Fig. 2) were the same as observed with maximum-likelihood or neighbor-joining analysis (data not shown). The only differences observed were the grouping of P. colocasiae, relative to other species in Clade 2 (this species grouped with P. capsici in the MP [58% bootstrap] and NJ tree, while in the ML tree it was on the same clade, but independent), and P. mirabilis, relative to other species on Clade 4 (this species grouped with P. phaseoli in the MP [58% bootstrap] and NJ tree, while in the ML tree it was on the same clade, but independent). The grouping of P. megakarya also was different for the NJ tree; it clustered as a separate branch with the isolates of Clade 7 (but with only 52% bootstrap support).

The topology of the shortest tree generated by a heuristic search based on maximum parsimony was similar to the majority-rule consensus tree, with the exception that more ancestral relationships among species were observed (Fig. 3). While many of the relationships among species observed on individual clades in the majority-rule consensus tree (Fig. 2) were maintained in the shortest tree (Fig. 3), rather than the clades unaffiliated with other species observed in Fig. 2, many of these clades were grouped together in Fig. 3. For example, the species groupings observed on clades 2, 5 and 6 in Fig. 2 were conserved in Fig. 3, but all three groupings were on the same clade on the shortest tree. Similar results were observed for clades 3 and 4 of Fig. 2 as well as P. megakarya and Clade 7. One difference, compared with the majority rule consensus tree, was that on the shortest tree (Fig. 3) Clade 1 was intermediate between L. giganteum and P. aphanidermatum and other species included in the analysis. In addition to the tree in Fig. 3, three other trees with a length of 599 were obtained and differed from Fig. 3 in that, in one tree, P. syringae grouped with species on Clade 1 (basal to the other species); P. nicotianae was intermediate between L. giganteum and P. aphanidermatum (one tree); the clade containing P. hibernalis and P. lateralis grouped with P. ilicis (two trees); and the positions of P. sojae and P. cinnamomi were reversed (one tree).

View larger version (40K): [in this window] [in a new window]    FIG. 3. Phylogenetic relationships among Phytophthora species using cox II DNA sequence data, based on maximum parsimony inferred by a heuristic tree search. This is one of four trees with the shortest length with the numbers at the nodes corresponding to the node labels in FIG. 2. Tree length = 599, consistency index (CI) = 0.467, homoplasy index (HI) = 0.533, retention index = 0.722. Isolates marked with an (*) were the same cultures that were used by Cooke at al (2000) to infer rDNA ITS-based phylogenetic trees

View larger version (20K): [in this window] [in a new window]    FIG. 4. Phylogenetic relationships among Phytophthora species using cox I DNA sequence data, based on maximum parsimony inferred by a heuristic tree search. The numbers above the nodes are the percentage of the trees from bootstrap analysis (100 replications) that support the observed topography (values above 60% indicated). Of the 1299 total characters, 1006 were constant, 115 were variable and parsimony uninformative and 178 were parsimony informative. Tree length = 607, consistency index (CI) = 0.634, homoplasy index (HI) = 0.366, retention index = 0.568

Comparison of phylogenetic relationships inferred from rDNA ITS sequences – ITS sequence data, for species included in this study and used by Cooke et al (2000), was retrieved from TreeBase and the analysis rerun with Pythium aphanidermatum as an outgroup. Tree reconstruction using maximum parsimony (MP) revealed four major clades that reflected the grouping of the neighbor-joining (NJ) tree reported by Cooke et al (2000), with the exception that some of the basal nodes of the NJ tree reflecting ancestral relationships were not observed in the MP tree (Fig. 5). For example, species on clades 2 and 3 of the MP tree were on the same clade along with P. ilicis in the NJ tree of Cooke et al (2000). Likewise, P. gonapodyides and P. megasperma were grouped with P. heveae on the NJ tree of Cooke et al (2000) but were grouped on Clade 4 of the MP tree (Fig. 5), although the bootstrap support was not strong (57%). The relationship between P. cinnamomi and P. sojae also was different; in the MP tree (Fig. 5) P. cinnamomi was ancestral to P. sojae, while in the NJ tree of Cooke et al (2000) P. sojae was ancestral.

View larger version (23K): [in this window] [in a new window]    FIG. 5. Phylogenetic relationships among Phytophthora species using ITS rDNA sequence alignments of Cooke et al (2000), based on maximum parsimony inferred by a heuristic tree search. The numbers above the nodes are the percentage of the trees from bootstrap analysis (1000 replications) that support the observed topography (values above 60% indicated). Of the 900 total characters, 465 were constant, 204 were variable and parsimony uninformative and 231 were parsimony informative. Tree length = 939, consistency index (CI) = 0.646, homoplasy index (HI) = 0.354, retention index = 0.675. Isolates marked with an (*) were the same cultures that were used in this submission to infer cox II-based phylogenetic trees

View larger version (39K): [in this window] [in a new window]    FIG. 6. Phylogenetic relationships among Phytophthora species using ITS rDNA sequence alignments of Cooke et al (2000) combined with cox II sequence data, based on maximum parsimony inferred by a heuristic tree search. The numbers above the nodes are the percentage of the trees from bootstrap analysis (1000 replications) that support the observed topography (values above 60% indicated). Of the 1468 total characters, 871 were constant, 223 were variable and parsimony uninformative and 374 were parsimony informative. Tree length = 1438, consistency index (CI) = 0.585, homoplasy index (HI) = 0.415, retention index = 0.765

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In terms of intraspecific conservation of sequences, no variation was observed for five of 14 species in which multiple isolates were examined, with less than 0.5% sequence divergence for an additional six species. In the case of P. cactorum, isolates from strawberry and fir were the same, while the isolate from apple was slightly different. There are reports of some level of host specialization and grouping of isolates into different nuclear DNA RFLP groups in this species (Hantula et al 2000), although a greater number of isolates from different geographic regions must be examined to determine whether this is important phylogenetically. It is interesting to note that there was no correlation for geographic origin or host in clustering of the four P. palmivora isolates examined in this study; an isolate from citrus in Florida clustered among isolates from Theobroma cacao in Nigeria, Costa Rica and Brazil.

The grouping of the isolate of P. arecae among isolates of P. palmivora in all cox II gene trees indicates that these two species might be conspecific. This observation is consistent with the conclusions of Oudemans and Coffey (1991) and Mchau and Coffey (1994) based on isozyme analysis, as well as Förster et al (1990b) based on mtDNA RFLP analysis. An affiliation between these two species also was observed in the ITS tree of Cooke et al (2000; their isolate of P. arecae also was used in this study), but only single isolates of each species were included in the analysis, so conclusions about conspecificity could not be drawn. Morphologically these two species share many features; the differences that are observed might reflect the range of variation inherent in the species.

There are several species for which additional investigations are needed to clarify species concepts. Four isolates of P. megasperma were assayed and, while they clustered together on Clade 5 (Fig. 2), they exhibited the greatest intraspecific variation observed in this study (sequence divergence as high as 4.9%). Although this sample size is not large enough to draw conclusions, no correlation between host and geographic origin of isolates was observed. Based on mtDNA RFLP analysis, the P. megasperma group is highly divergent and can be divided into six groups (Förster and Coffey 1993; nine groups were reported, but three subsequently have been renamed as distinct species (Erwin and Ribeiro 1996)). Additional studies are needed to determine if the groupings on the cox II tree reflect different species and how this relates to the groupings observed by Förster and Coffey (1993). Our analysis, as well as that of Cooke et al (2000), showed that P. gonapodyides and P. megasperma grouped together. Whether these two species are conspecific or if the relationship among these species on Clade 5 reflects the variation among isolates classified as P. megasperma remains to be determined. Morphologically, P. gonapodyides and P. megasperma share many features, however, they also exhibit distinct differences (P. gonapodyides is heterothallic with amphigynous antheridia while P. megasperma is homothallic with paragynous antheridia). Based on morphology and protein-banding patterns, P. gonapodyides has been linked to P. drechsleri and P. cryptogea (Brasier 1991, Hansen et al 1988), but it doesn't group with these species in the ITS-based tree (Cooke et al 2000) or in our analysis of cox II. While on the same clade, the isolates of P. megakarya from Nigeria and Cameroon exhibit a greater sequence divergence than in most other species examined (the only exception being P. megasperma). From mtDNA RFLP analysis Förster et al (1990b) also obtained variability among isolates from these two regions, with less than 60% similarity in banding pattern observed.

The grouping observed for the consensus cox II gene sequence tree is in general agreement with the conclusions of Cooke et al (2000) from their analysis of rDNA ITS, however, there were some notable exceptions. The isolates of P. syringae, P. lateralis, P. drechsleri (USA), P. erythroseptica (Ohio) and P. cryptogea (New Zealand) in this study were the same that were used for the ITS study of Cooke et al (2000). All five were on the same branch of Clade 4 in the ITS tree (Fig. 5) but were on different clades in the consensus cox II tree (Fig. 2); P. syringae and P. lateralis were unaffiliated with any specific clade and did not group with P. drechsleri, P. erythroseptica or P. cryptogea. Phytophthora cinnamomi grouped with P. fragariae on the ITS tree but was unaffiliated with a specific clade in the consensus cox II tree. Similar differences were observed for Clade 2 of the ITS tree (represented by clades 3 and 4 in the consensus cox II tree) and Clade 1 of the ITS tree (represented by Clade 7 of the consensus cox II tree, with P. megakarya clustering separately).

Comparison of the most-parsimonious cox II tree with the shortest length (Fig. 3) and the ITS tree (Fig. 5) revealed overall similarities in species groupings, but differences were observed. The species groupings observed on clades 2 and 3 of the ITS tree were the same for the cox II tree, but relationships among the clades exhibited some differences. For example, clades 2 and 5 on the cox II tree grouped together but were separated in the ITS tree. Furthermore, P. drechsleri, P. erythroseptica, and P. cryptogea were ancestral to other Phytophthora species in the shortest branch length cox II tree (which was not observed in the ITS tree) and these species did not group with P. syringae or P. lateralis (which was observed in the ITS tree). Thus, these differences in species groupings in the cox II and ITS sequence trees reflect differences in phylogenetic resolution of the two regions, either through differing rates of evolutionary divergence or incorrect assumptions about alignment of the ITS sequences.

Concatenation of the cox II sequences and the ITS data of Cooke et al (2000) and running phylogenetic analysis generated a tree that did not differ significantly from the tree derived from ITS data alone. However, because a partition-homogeneity test with 1000 replicates indicated there was significant heterogeneity between the two data sets (P = 0.002), analysis of combined data sets might not be valid. Possible reasons for conservation of the ITS tree topography after analysis of concatenated sequences might be the greater number of phylogenetically informative sites present in the ITS data set (231 vs 126 for the cox II data) and the greater level of sequence divergence observed for the ITS data (17.4% vs 12.2% for the cox II data).

The phylogenetic relationships among species on these cox II gene trees did not exhibit consistent similarities in groupings for morphology, sexual features or temperature optima (Table II). However, some associations were observed among species groupings on the cox II gene tree and Waterhouse groupings (Fig. 2). For example, all species in clades 1, 3, 4, 6, and 7 of Fig. 2 were members of a single Waterhouse group with similar sporangial papillation characteristics (groups VI, I, IV, V and II, respectively), but other members of Waterhouse groups II–VI were on separate branches and not included in the aforementioned clades. Förster et al (2000) used the ITS I region for phylogenetic analysis, which did not include the 5.8S and ITS 2 region that Cooke et al (2000) included. They found that species of groups V and VI, which are differentiated by the presence of amphigynous or paragynous antheridia, respectively, were interspersed in the ITS1 tree. Species with papillate and semipapillate sporangia (groups I–IV) clustered together, and this cluster was distinct from those of species with nonpapillate sporangia. There was no congruence among the mode of antheridial attachment, sporangial caduacity or homo- or heterothallic habit and the ITS1 grouping. Using a smaller number of species in the analysis, Crawford et al (1996) and Cooke and Duncan (1997) found that phylogenetic groupings based on ITS sequences generally agreed with morphological groupings, especially those based on sporangial morphology. In contrast, using a wider range of species, Cooke et al (2000) found that their ITS data did not support morphological subgroups. Briard et al (1995), who analyzed part of the D2 domain of the large subunit (28S) ribosomal rRNA, came to the same conclusion with their analysis of 15 Phytophthora species. In light of these results and our findings with cox gene phylogenies, it seems clear that at best only partial correlations between molecular and morphological characters might exist, and that morphology alone should not be used to infer phylogeny in Phytophthora sensu lato.

Cytochrome oxidase II gene sequences have been used for inferring phylogenetic relationships among members of the Peronosporomycetes (Hudspeth et al 2000) and the genus Pythium (Martin 2000). One advantage to using this gene in phylogenetic analysis at the genus level is that length mutations have not been observed, which simplifies sequence alignment. In contrast, alignment of ITS sequences can be more complicated due to the presence of length mutations in different regions of the ITS. For example, with the Phytophthora species examined in Fig. 5, the length of the ITS region ranged from 752 (P. capsici) to 832 bp (P. fragariae var. rubi). Following introduction of gaps for alignment purposes, there were 900 bp total, of which 465 characters were constant, 204 were variable and parsimony uninformative, and 231 were parsimony informative. One complicating factor in constructing alignments of sequences with this level of length variation is identifying alignments that accurately reflect phylogenetic divergence among species (e.g., determining correct boundaries of gaps and proper alignment of variable regions). Variation in alignment has been shown to affect subsequent phylogenetic clustering much more significantly than the particular algorithm (parsimony, distance, etc.) used in tree construction (Morrison and Ellis 1997). Another possible complication for using the ITS region for phylogenetic purposes is that multiple, nonorthologous ITS sequences have been found in some fungi (Fatehi and Bridge 1998, O'Donnell et al 1998), along with the occurrence of multiple and divergent ITS sequences within single spores of certain species (Pringle et al 2000), including the closely related genus Pythium (Martin 1990). Furthermore, ITS regions might not fully distinguish groups of species that are known to be biologically distinct. For example, P. infestans, P. mirabilis, and P. phaseoli (Cooke et al 2000, Goodwin et al 1999) and P. fragariae var. fragariae and P. fragariae var. rubi (Cooke et al 2000) are poorly resolved in ITS phylogenies (Fig. 5) yet more clearly differentiated in cox II phylogenies (Fig. 2). However, ITS has the advantage of being a nuclear region and sequence data will reflect genetic and evolutionary forces such as intra- and interspecific hybridization not reflected in patterns of evolution of cytoplasmic genes such as cox II. Given that interspecific hybridization has been reported in the genus Phytophthora (Brasier et al 1999, English et al 1999, Bonants et al 2000, Gu and Ko 2000, Delcan and Brasier 2001) this is an important consideration and a potential drawback in relying strictly on mitochondrial sequences (which are believed to be uniparentally inherited; Förster and Coffey 1990).

Clearly, additional analysis of cox II gene sequences with a greater number of species having a broader representation of the morphological variation present in the genus are needed to clarify the phylogeny of the genus. Likewise, inclusion of additional species that are more restricted in their host range, are nonpathogenic, or are representative of differing ecological habitats, is needed. Given the lack of congruence for some species between cox II and ITS data, comparison of these cox II gene trees with DNA sequence analysis from other regions of the nuclear genome using the same isolates for both analyses would be important as well for correct determinations of phylogenetic relationships in the genus.

	ACKNOWLEDGMENTS

 
The DNA sequencing was done by the DNA Sequencing Laboratory of the Interdisciplinary Center for Biotechnological Research of the University of Florida at Gainesville, Gainesville, Florida; the efforts of Savita Shanker and Ernesto Almira are gratefully acknowledged. Comments on this manuscript before submission by Dave Mitchell, Reid Frederick and Steve Rehner are gratefully acknowledged. Mention of trade names or commercial products in this manuscript is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.

	FOOTNOTES

 
¹ Corresponding author, fmartin{at}pw.ars.usda.gov

Accepted for publication August 27, 2002.

	LITERATURE CITED

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Baldauf SL, Palmer JD., 1993 Animals and fungi are each other's closest relatives: congruent evidence from multiple proteins. Proc Natl Acad Sci USA 90:11558-11562[Abstract/Free Full Text]

Bhattacharya D, Stickel SK., 1994 Sequence analysis of duplicated actin genes in Lagenidium giganteum and Pythium irregulare (Oomycota). J Mol Evol 39:56-61[Medline]

Bonants PJM, Hagenaar-de Weerdt M, Man in't Veld WA, Baayen RP., 2000 Molecular characterization of natural hybrids of Phytophthora nicotianae and P. cactorum. Phytopathology 90:867-874[Medline]

Brasier CM., 1983 Problems and prospects in Phytophthora research. In: Erwin DC, Bartnicki-Garcia S, Tsao P, eds. Phytophthora: its biology, taxonomy, ecology, and pathology. St. Paul, MN: American Phytopathological Society. p 351–364

———. 1991 Current questions in Phytophthora systematics: the role of the population approach. In: Lucas JA, Shattock RC, Shaw DS, Cooke LR, eds. Phytophthora. Cambridge, United Kingdom: Cambridge University Press. p 104–128

———, Hansen EM., 1992 Evolutionary biology of Phytophthora. Part II. Phylogeny, speciation, and population structure. Ann Rev Phytopathol 30:173-200

———, Cooke DEL, Duncan JM., 1999 Origin of a new Phytophthora pathogen through interspecific hybridization. Proc Natl Acad Sci USA 96:5878-5883[Abstract/Free Full Text]

Briard M, Dutertre M, Rouxel F, Brygoo Y., 1995 Ribosomal RNA sequence divergence within the Pythiaceae. Mycol Res 99:1119-27

Caten CE, Jinks JL., 1968 Spontaneous variability of single isolates of Phytophthora infestans I. Cultural variation. Can J Bot 46:329-348

Cooke DEL, Duncan JM., 1997 Phylogenetic analysis of Phytophthora species based on the ITS1 and ITS2 sequences of ribosomal DNA. Mycol Res 667–677

———, Drenth A, Duncan JM, Wagels G, Brasier CM., 2000 A molecular phylogeny of Phytophthora and related oomycetes. Fungal Genetics and Biology 30:17-32[Medline]

Crawford AR, Bassam BJ, Drenth A, Maclean DJ, Irwin JAG., 1996 Evolutionary relationships among Phytophthora species deduced from rDNA sequence analysis. Mycol Res 100:437-443

Delcan J, Brasier CM., 2001 Oospore viability and variation in zoospore and hyphal tip derivatives of the hybrid alder Phytophthora. Forest Pathology 31:65-83

English JT, Laday M, Bakonyi J, Schoelz JE, Ersek T., 1999 Phenotypic and molecular characterization of species hybrids derived from induced fusion of zoospores of Phytophthora capsici and Phytophthora nicotianae. Mycol Res 103:1003-1008

Eriksson T., 1999 AutoDecay ver. 4.0 (program distributed by the author). Stockholm, Bergius Foundation, Royal Swedish Acadamy of Sciences

Erwin DC, Ribeiro OK., 1996 Phytophthora Diseases Worldwide. St. Paul, APS Press, American Phytopathological Society

Fatehi J, Bridge P., 1998 Detection of multiple rRNA-ITS regions in isolates of Ascochyta. Mycol Res 102:762-766

Förster H, Coffey MD., 1990 Mating behavior of Phytophthora parasitica: evidence for sexual recombination in oospores using DNA restriction fragment length polymorphisms as genetic markers. Exp Mycol 14:351-359

———, ———. 1993 Molecular taxonomy of Phytophthora megasperma based on mitochondrial and nuclear DNA polymorphisms. Mycol Res 97:1101-1112

———, Cummings MP, Coffey MD., 2000 Phylogenetic relationships of Phytophthora species based on ribosomal ITS 1 DNA sequence analysis with an emphasis on Waterhouse groups V and VI. Mycol Res 104:1055-1061

———, Oudemans P, Coffey MD., 1990b Mitochondrial and nuclear DNA diversity within six species of Phytophthora. Exp Mycol 14:18-31

———, Coffey MD, Elwood H, Sogin ML., 1990a Sequence analysis of the small subunit ribosomal RNAs of three zoosporic fungi and implications for fungal evolution. Mycologia 82:306-312

———, Kinsherf TG, Leong SA, Maxwell DP., 1988 Estimation of relatedness between Phytophthora species by analysis of mitochondrial DNA. Mycologia 80:466-478

Goodwin SB, Drenth A, Fry WE., 1992 Cloning and genetic analysis of two highly polymorphic, moderately repetitive nuclear DNAs from Phytophthora infestans. Curr Genet 22:107-115[Medline]

———, Legard DE, Smart CD, Levy M, Fry WE., 1999 Gene flow analysis of molecular markers confirms that Phytophthora mirabilis and P. infestans are separate species. Mycologia 91:796-810

Gu YH, Ko WH., 2000 Segregation following interspecific transfer of isolated nuclei between Phytophthora parasitica and P. capsici. Can J Microbiol 46:410-416[Medline]

Hansen EM, Hamm PB, Shaw CGIII, Hennon PE., 1988 Phytophthora drechsleri in remote areas of southeast Alaska. Trans Br Mycol Soc 91:379-384

Hantula J, Lilja A, Nuorteva H, Parikka P, Werres S., 2000 Pathogenicity, morphology and genetic variation of Phytophthora cactorum from strawberry, apple, rhododendron, and silver birch. Mycol Res 104:1062-1068

Hudspeth DSS, Nadler SA, Hudspeth MES., 2000 A cox2 molecular phylogeny of the Peronosporomycetes. Mycologia 92:674-684

Klassen GR, McNabb SA, Dick MW., 1987 Comparison of physical maps of ribosomal DNA repeating units in Pythium, Phytophthora and Apodachlya. J Gen Microbiol 133:2953-2959

Knoll HA., 1992 The early evolution of eukaryotes: a geological perspective. Science 256:622-627[Abstract/Free Full Text]

Lévesque CA, Harlton CE, de Cock AWAM., 1998 Identification of some oomycetes by reverse dot blot hybridization. Phytopathology 88:214-222

Leipe DD, Tong SM, Goggin CL, Slemenda SB, Pieniazek NJ, Sogin ML., 1996 16S-like rDNA sequences from Developayella elegans, Labyrinthuloides haliotidis, and Proteromonas lacertae confirm that stramenopiles are a primarily heterotrophic group. Eur J Protistol 32:449-458

———, Wainright PO, Gunderson JH, Porter D, Patterson DJ, Valois F, Himmerich S, Sogin ML., 1994 The stramenopiles from a molecular perspective: the 16S-like rRNA sequences from Labyrinthuloides minuta and Cafeteria roenbergensis. Phycologia 33:369-377

Martin FN., 1990 Variation in the ribosomal DNA repeat unit within single oospore isolates of the genus Pythium. Genome 33:585-591

———. 2000 Phylogenetic relationships among some Pythium species inferred from sequence analysis of the mitochondrially encoded cytochrome oxidase II gene. Mycologia 92:711-727

———, Semer CR., 1997 Selection of drug-tolerant strains of Pythium sylvaticum using sublethal enrichment. Phytopathology 87:685-692[Medline]

Mchau GRA, Coffey MD., 1994 Isozyme diversity in Phytophthora palmivora: evidence for a southeast Asia center of origin. Mycol Res 98:1035-1043

Morrison DA, Ellis JT., 1997 Effects of nucleotide sequence alignment on phylogeny estimation: a case study of 18S rDNAs of Apicomplexa. Mol Biol Evol 14:428-44[Abstract]

O'Donnell KO, Cigelnik E, Nirenberg HI., 1998 Molecular systematics and phylogeography of the Gibberella fujiduroi species complex. Mycologia 90:465-493

Pringle A, Moncalvo J-M, Vilgalys R., 2000 High levels of variation in ribosomal DNA sequences within and among spores of a natural population of the arbuscular mycorrhizal fungus Acaulospora colossica. Mycologia 92:259-268

Sachay DJ, Hudspeth DSS, Nadler SA, Hudspeth MES., 1993 Oomycete mtDNA: Phytophthora genes for cytochrome c oxidase use an unmodified genetic code and encode proteins most similar to plants. Exp Mycol 17:7-23

Silberman JD, Sogin ML, Leipe DD, Clark CG., 1996 Human parasite finds taxonomic home. Nature (London) 380:398

Stamps DJ, Waterhouse GM, Newhook FJ, Hall GS., 1990 Revised tabular key to the species of Phytophthora. CABI Mycology Inst. Mycology Paper No. 162

Tooley PW., 1988 Use of uncontrolled freezing for liquid nitrogen storage of Phytophthora species. Plant Dis 72:680-682

Van de Peer Y, van der Auwera G, De Wachter R., 1996 The evolution of stramenopiles and alveolates as derived by "substitution rate calibration" of small ribosomal subunit RNA. J Mol Evol 42:201-210[Medline]

Wainright PO, Hinkle G, Sogin ML, Stickel SK., 1993 Monophyletic origins of the metazoa: an evolutionary link with fungi. Science 260:340-342[Abstract/Free Full Text]

Waterhouse GM., 1963 Key to the species of Phytophthora de Bary. Mycology Paper No. 92. Kew, United Kingdom: Commonwealth Mycological Institute

Weerakoon ND, Roberts JK, Lehnen LP, Wilkenson JM, Marshall JS, Hardham AR., 1998 Isolation and characterization of the single ß-tublin gene in Phytophthora cinnamomi. Mycologia 90:85-95

White TJ, Bruns T, Lee S, Taylor J., 1990 Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols: a guide to methods and applications. San Diego: Academic Press. p 315–322

This article has been cited by other articles:

				R. S. Donahoo and K. H. Lamour Interspecific hybridization and apomixis between Phytophthora capsici and Phytophthora tropicalis Mycologia, November 1, 2008; 100(6): 911 - 920. [Abstract] [Full Text] [PDF]

				L. Gomez-Alpizar, C.-H. Hu, R. Oliva, G. Forbes, and J. B. Ristaino Phylogenetic relationships of Phytophthora andina, a new species from the highlands of Ecuador that is closely related to the Irish potato famine pathogen Phytophthora infestans Mycologia, July 1, 2008; 100(4): 590 - 602. [Abstract] [Full Text] [PDF]

				E. Moralejo, A. Clemente, E. Descals, L. Belbahri, G. Calmin, F. Lefort, C. F.J. Spies, and A. McLeod Pythium recalcitrans sp. nov. revealed by multigene phylogenetic analysis Mycologia, March 1, 2008; 100(2): 310 - 319. [Abstract] [Full Text] [PDF]

				Z. G. Abad, J. A. Abad, M. D. Coffey, P. V. Oudemans, W. A. Man in 't Veld, H. de Gruyter, J. Cunnington, and F. J. Louws Phytophthora bisheria sp. nov., a new species identified in isolates from the Rosaceous raspberry, rose and strawberry in three continents Mycologia, January 1, 2008; 100(1): 99 - 110. [Abstract] [Full Text] [PDF]

				W. A. Man in 't Veld Gene flow analysis demonstrates that Phytophthora fragariae var. rubi constitutes a distinct species, Phytophthora rubi comb. nov. Mycologia, March 1, 2007; 99(2): 222 - 226. [Abstract] [Full Text] [PDF]

				N. O. Villa, K. Kageyama, T. Asano, and H. Suga Phylogenetic relationships of Pythium and Phytophthora species based on ITS rDNA, cytochrome oxidase II and beta-tubulin gene sequences. Mycologia, May 1, 2006; 98(3): 410 - 422. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Martin, F. N. Articles by Tooley, P. W. Search for Related Content PubMed Articles by Martin, F. N. Articles by Tooley, P. W. Agricola Articles by Martin, F. N. Articles by Tooley, P. W.