Present address: The Sainsbury Laboratory, Norwich Research Park, Colney, NR4 7UH, Norwich, UK.
Phytophthora andina sp. nov., a newly identified heterothallic pathogen of solanaceous hosts in the Andean highlands
Article first published online: 13 APR 2010
No claim to original US government works
Volume 59, Issue 4, pages 613–625, August 2010
How to Cite
Oliva, R. F., Kroon, L. P. N. M., Chacón, G., Flier, W. G., Ristaino, J. B. and Forbes, G. A. (2010), Phytophthora andina sp. nov., a newly identified heterothallic pathogen of solanaceous hosts in the Andean highlands. Plant Pathology, 59: 613–625. doi: 10.1111/j.1365-3059.2010.02287.x
- Issue published online: 1 JUL 2010
- Article first published online: 13 APR 2010
- Phytophthora infestans;
- Solanum spp.;
- tree tomato
A blight disease on fruits and foliage of wild and cultivated Solanum spp. was found to be associated with a new species of Phytophthora. The proposed novel species is named Phytophthora andina Adler & Flier, sp. nov. based on morphological characteristics, pathogenicity assays, mitochondrial DNA haplotyping, AFLP fingerprinting and nuclear and mitochondrial DNA sequence analyses. Isolates of P. andina (n = 48) from the Andean highland tropics of Ecuador were collected from 1995 to 2006. Phytophothora andina is closely related to P. infestans and has semipapillate, ellipsoidal sporangia borne on sympodially branched sporangiophores. It is heterothallic and produces amphigynous antheridia. The species consists of several clonal lineages, including the EC-2 and EC-3 RFLP lineages, which were described previously as P. infestans. Approximately 75% of isolates react as compatibility type A2 when paired with an A1 compatibility type isolate of P. infestans. However, when A2 isolates from the Anarrhichomenum section of Solanum were paired in all combinations, viable oospores were obtained in several crosses, suggesting that there is a unique compatibility interaction in P. andina that is complementary to that described in P. infestans. Nuclear and mitochondrial sequence analysis supported the species designation of P. andina. This newly identified heterothallic pathogen shares a common ancestor with P. infestans and may have arisen from hybridization events with sister taxa in the Andes.
The Andean highlands are considered a centre of origin and diversity for the plant genus Solanum (Hawkes, 1990). In addition to potato (Solanum tuberosum) and tomato (S. lycopersicum), several solanaceous crops have been domesticated there, including S. betaceum, known as tomate de árbol or tree tomato, and as tamarillo outside the Andes, S. muricatum, known as pepino dulce, pepino or pear melon, and S. quitoense, known as naranjilla (Ecuador) or lulo (Colombia). The first two are globally important crops, whilst the latter are primarily restricted to the Andes (Popenoe et al., 1989). Cultivated Solanum crops are often grown in close proximity to wild Solanum spp. including S. caripense, S. hispidum and S. ochranthum and the poorly defined species in the Anarrhichomenum section (e.g. S. tetrapetalum, S. sodiroi, S. siphonobasis and S. brevifolium). Although the taxonomy of tuber-bearing Solanum spp. and solanaceous fruit crops remains complex, most species have been described and classified (Spooner et al., 2003). Much less is known, however, about the phylogenetic relationships and taxonomy of non-tuber-bearing wild Solanum spp.
Phytophthora populations associated with Solanum hosts in the Andes appear to be more complex than was originally believed. Until 1994, genetic diversity studies in Ecuador revealed only limited variation in the local populations of P. infestans on cultivated species. Late-blight epidemics on potato and tomato were exclusively caused by the EC-1 and US-1 clonal lineages of P. infestans, respectively (Forbes et al., 1997; Oyarzun et al., 1998). Both lineages have the A1 mating type and reproduce clonally. This simple appraisal of the genetic structure of the pathogen population began to change in 1995 when four Phytophthora isolates were collected from blighted leaves of a plant in Ecuador identified as S. brevifolium. The isolates from S. brevifolium were the A2 mating type when crossed with A1 tester isolates of P. infestans and were initially identified as P. infestans, although they were recognized as novel and characterized by a previously unknown mtDNA haplotype (Ic) and a new RFLP fingerprint (EC-2) (Ordoñez et al., 2000). A total of 53 isolates were collected between 1996 and 1999 from hosts identified at the time as S. brevifolium and S. tetrapetalum. Based on a unique RFLP fingerprint with limited polymorphism, a new mtDNA haplotype and A2 mating type, these isolates were classified in a new clonal lineage of P. infestans designated EC-2 (Ordoñez et al., 2000). Further study of the hosts revealed that the original 53 isolates came from one or more Solanum species in the Anarrhichomenum section of Solanum (Spooner et al., 2003). Ordoñez and co-workers did not rule out the possibility that the EC-2 lineage was indigenous to Ecuador. The identification of the clonal lineage as P. infestans remained unsatisfactory, so the isolates were classified as Phytophthora infestans sensu lato (Ordoñez et al., 2000).
In a subsequent study which included a systematic sampling of foliar Phytophthora pathogens from various wild and cultivated solanaceous hosts, Adler et al. (2004) reported the presence of another clonal lineage (EC-3) on S. betaceum. These authors also found numerous isolates that were not of the EC-2 lineage attacking plants in the Anarrhichomenum complex. These isolates were the A1 mating type and Ia mtDNA haplotype. In the study by Adler et al. (2004), amplified fragment length polymorphism (AFLP) fingerprinting confirmed the presence of host-specific groups of isolates, but the taxonomic status of these Phytophthora isolates attacking several non-tuber-bearing Solanum spp. remained unresolved (Adler et al., 2004).
The phylogeny of the genus Phytophthora was the subject of several studies in which molecular techniques were used to supplement morphological classification (Waterhouse, 1963) and facilitate phylogenetic inference (Cooke et al., 2000; Martin & Tooley, 2003; Kroon et al., 2004). Sequence variation for the ITS region of Phytophthora species was shown to be useful for resolving the general relationships within the genus (Cooke et al., 2000), but proved insufficient to discriminate among P. infestans, P. mirabilis, P. phaseoli and P. ipomoeae, which are all placed in group IV (sensu Waterhouse) and clade Ic (sensu Cooke) (Flier et al., 2002). This lack of resolution of the clade Ic species using ITS DNA has led to other studies with more variable DNA regions. Martin & Tooley (2003) sequenced part of the mitochondrial Cox I gene, whilst Kroon et al. (2004) used parts of the β-tubulin, translation elongation factor 1-alpha, NADH-4 and Cox I genes to construct a high-resolution molecular phylogeny for the genus Phytophthora. The latter study demonstrated that the EC-2 clonal lineage of P. infestans sensu lato was closely related to P. infestans, P. ipomoeae and P. mirabilis, yet not identical to them, thus stimulating discussion on the origin of the Phytophthora spp. in Ecuador. Although a new species had not been formally described, Kroon et al. (2004) referred to the one isolate from Ecuador they had examined as P. andina. That isolate had come from an Anarrhichomenum section host and corresponded to the EC-2 lineage of Ordoñez et al. (2000), and this was apparently the first reported use of the name. Wattier et al. (2003) also concluded that isolates from hosts of the Anarrhichomenum complex were distinct from P. infestans based on mitochondrial inter-gene sequences, although the species name P. andina was not used in their report. Four isolates of the EC-2 lineage from Anarrhichomenum section hosts were included in a coalescent analysis in which a South American origin was inferred for P. infestans, and these four isolates were referred to as P. andina (Gomez-Alpizar et al., 2007). The four isolates used in that study shared a common ancestor with P. infestans but were derived from a distinct mitochondrial lineage (Ic) that had undergone many mutations.
In an effort to clarify the taxonomy of this group of Phytophthora isolates from solanceous hosts in Ecuador, phylogenetic analysis of 11 EC-2 and eight EC-3 isolates was conducted using sequences of the Cox I gene and intron 1 of the Ras gene (Gomez-Alpizar et al., 2008). In that study EC-2 isolates with the Ic mtDNA haplotype were considered distinct and were designated P. andina, but no clear description of the new species was given (Gomez-Alpizar et al., 2008). Furthermore, the taxonomy of the EC-3 group of isolates from S. betaceum remained unclear since few EC-3 isolates were examined and some of these grouped more closely to P. infestans than P. andina in the mitochondrial genealogy. Using microsatellite (SSR) markers, Oliva et al. (2007) found that EC-2 and EC-3 lineages were clearly separated from the US-1 and EC-1 lineages of P. infestans. However, the authors referred to the whole population as P. infestans, only mentioning P. andina in reference to earlier papers. In a recent and comprehensive multilocus phylogeny of the genus Phytophthora, two Ecuadorian isolates corresponding to the EC-2 lineage were considered closely related to, but distinct from P. infestans (Blair et al., 2008). In reference to an uncertain taxonomic designation for the isolates, the authors referred to the taxon as P. sp “andina”.
All the previous studies examining the genetic relatedness among clade-Ic species of Phytophthora species and isolates from the putative P. andina, i.e. those originally described as EC-2 and EC-3 lineages of P. infestans (sensu lato), have suggested a close relationship between this novel species, P. infestans and P. mirabilis (Wattier et al., 2003; Kroon et al., 2004; Blair et al., 2008; Gomez-Alpizar et al., 2008), but a species description has not been given. Thus, as recognized by Blair et al. (2008) with their use of quotation marks, the taxon sometimes referred to as P. andina has not been defined. Phytophthora species attacking solanaceous hosts in the Andes cause severe damage on several regionally important crops and provide a very interesting model of host/pathogen coevolution (Oliva et al., 2007). Further studies of these pathogens are to be expected and taxonomic uncertainty of the species could hinder progress.
This papers formally describes the new species P. andina and clarifies its taxonomic status and relatedness to P. infestans. To achieve this, a representative sample of blight-causing Phytophthora isolates collected between 1995 and 2006 from the highlands of Ecuador were investigated. The new species is described on the basis of morphological characteristics, host range, mating behavior, allozyme genotype, RLFP genotype and mitochondrial DNA haplotype. Genetic relatedness to P. infestans is based on one mitochondrial gene (Cox II), ITS sequences and a portion of a single-copy nuclear gene (Ras). AFLP fingerprinting was used to explore genetic similarities between host- specific groups of P. andina and their relatedness to other Phytophthora species in the Ic clade that attack Solanum, Ipomoea and Mirabilis spp. Results are discussed in relation to previous studies, particularly for relatedness to P. infestans (Oliva et al., 2007) and for host range (Chacón, 2007).
Materials and methods
Pathogen collection and maintenance
Isolates of Phytophthora species used in this study were obtained from diseased host plants collected during sampling trips throughout the Ecuadorian Andes (Fig. 1, Table 1). Isolates were collected in almost all provinces of Ecuador between 1995 and 2006, ranging from 800 to 3800 m a.s.l. A map describing the main collection areas was published by Adler et al. (2004). Individual 1-cm2 pieces of diseased leaves and fruits were transferred to selective Rye B (RB) agar (Caten & Jinks, 1968) amended with vancomycin (100 mg), polymyxin-B (500 mg), ampicillin (200 mg), rifampicin (20 mg), pentachloronitrobenzene (67 mg) and benomyl (100 mg) in 1000 mL distilled water. Colonies growing on selective media were transferred to RB agar for initial identification as Phytophthora species. All isolates were maintained for short periods on Rye A (RA) or RB agar at 18°C in the dark (Ordoñez et al., 2000). Isolates were stored for longer periods on RA slants at 15°C with a 12-h photoperiod in Quito, Ecuador and in liquid nitrogen at Plant Research International in Wageningen, the Netherlands. Isolates were compared with reference isolates of P. infestans, P. mirabilis, P. ipomoeae and P. phaseoli from the Phytophthora culture collection at Plant Research International (Table 1).
|Isolate||Species||Origin||Host||Allozymea genotype||MtDNAb haplotype||RG57 RFLPc||Mating typee||Used ind|
|EC1836||P. andina||Ecuador||Anarrhichomenumf||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3186||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3189||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3190||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3229||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3231||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3232||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3233||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3261||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3262||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3323||P. andina||Ecuador||Solanum quitoense||76/100||86/100||Ia||nd||A1||d|
|EC3358||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3363||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||c|
|EC3364||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||c|
|EC3369||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b,e|
|EC3380||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||c, e|
|EC3394||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||c|
|EC3395||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||e|
|EC3399||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ia||EC-2||A1||a, b|
|EC3400||P. andina||Ecuador||Brugmansia spp.||76/100||100/100||Ia||EC-2||A1||e|
|EC3402||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||e|
|EC3414||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3421||P. andina||Ecuador||S. muricatum||76/100||100/100||Ic||EC-2||A2||c, e|
|EC3423||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||e|
|EC3424||P. andina||Ecuador||S. muricatum||76/100||100/100||Ic||EC-2||A2||c|
|EC3514||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||d|
|EC3542||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||e|
|EC3561||P. andina||Ecuador||S. quitoense||76/100||86/100||Ia||nd||A1||d|
|EC3644||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||d|
|EC3658||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ic||EC-2||A2||a, b|
|EC3783||P. andina||Ecuador||S. hispidum||76/100||100/100||Ic||EC-2||A2||d|
|EC3818||P. andina||Ecuador||Anarrhichomenum||76/100||100/100||Ia||EC-2||A1||a, d|
|EC3830||P. andina||Ecuador||S. hispidum||76/100||100/100||Ic||EC-2||A2||d|
|EC3860||P. andina||Ecuador||Solanum sp.||76/100||100/100||Ic||nd||A2||d|
|EC3875||P. andina||Ecuador||S. betaceum||76/100||86/100||Ia||EC-3||A1||d|
|EC3120||P. infestans||Ecuador||S. ochranthum||96/100||90/100||IIa||nd||A1||e|
|EC3241||P. infestans||Ecuador||S. ochranthum||92/100||86/100||Ib||US-1||A1||d, e|
|EC3322||P. infestans||Ecuador||S. quitoense||76/100||86/100||Ib||nd||A1||d|
|EC3338||P. infestans||Ecuador||S. paucijugum||96/100||90/100||IIa||EC-1||A1||e|
|EC3351||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||c|
|EC3355||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||c|
|EC3361||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||e|
|EC3378||P. infestans||Ecuador||S. lycopersicum||92/100||86/100||Ib||US-1||A1||c, e|
|EC3379||P. infestans||Ecuador||S. minutifoliolum||96/100||90/100||IIa||EC-1||A1||e|
|EC3381||P. infestans||Ecuador||S. lycopersicum||92/100||86/100||Ib||US-1||A1||c, d, e|
|EC3382||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||e|
|EC3383||P. infestans||Ecuador||S. phureja||96/100||90/100||IIa||EC-1||A1||c|
|EC3389||P. infestans||Ecuador||S. lycopersicum||92/100||86/100||Ib||US-1||A1||c, d|
|EC3390||P. infestans||Ecuador||S. solisii||96/100||90/100||IIa||EC-1||A1||e|
|EC3404||P. infestans||Ecuador||S. caripense||92/100||86/100||Ib||US-1||A1||c, e|
|EC3413||P. infestans||Ecuador||S. paucijugum||96/100||90/100||IIa||EC-1||A1||e|
|EC3415||P. infestans||Ecuador||S. minutifoliolum||96/100||90/100||IIa||EC-1||A1||e|
|EC3435||P. infestans||Ecuador||S. colombianum||96/100||90/100||IIa||EC-1||A1||e|
|EC3445||P. infestans||Ecuador||S. colombianum||96/100||90/100||IIa||EC-1||A1||e|
|EC3447||P. infestans||Ecuador||S. phureja||96/100||90/100||IIa||EC-1||A1||c|
|EC3528||P. infestans||Ecuador||S. andreanum||96/100||90/100||IIa||EC-1||A1||e|
|EC3529||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||d|
|EC3531||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||e|
|EC3532||P. infestans||Ecuador||S. tuberosum||96/100||90/100||IIa||EC-1||A1||e|
|EC3534||P. infestans||Ecuador||S. andreanum||96/100||90/100||IIa||EC-1||A1||e|
|EC3620||P. infestans||Ecuador||S. ochranthum||92/100||90/100||Ib||EC-1||A1||d|
|EC3774||P. infestans||Ecuador||S. ochranthum||96/100||86/100||Ib||US-1||A1||d|
|EC3808||P. infestans||Ecuador||S. muricatum||92/100||86/100||Ib||US-1||A1||d|
|EC3842||P. infestans||Ecuador||S. habrochaites||92/100||86/100||Ib||US-1||A1||d|
|PICWF021||P. mirabilis||Mexico||M. jalapaf||96/96||100/100||HaPmirI||nd||nd||c|
|PICWF031||P. mirabilis||Mexico||M. jalapa||96/96||100/111||HaPmirI||nd||nd||c|
|PICWF005||P. mirabilis||Mexico||M. jalapa||96/96||90/111||HaPmirI||nd||nd||c|
|PICWF053||P. mirabilis||Mexico||M. jalapa||96/96||100/111||HaPmirI||nd||nd||c|
|PIC991-2||P. ipomoeae||Mexico||I. longipedunculataf||78/78||108/108||HaPipoI||nd||sf||c|
|PIC991-1||P. ipomoeae||Mexico||I. longipedunculata||78/78||108/108||HaPipoI||nd||sf||c|
|PIC993-3||P. ipomoeae||Mexico||I. longipedunculata||78/78||108/108||HaPipoI||nd||sf||c|
|PIC993-4||P. ipomoeae||Mexico||I. longipedunculata||78/78||108/108||HaPipoI||nd||sf||c|
|CBS556.88||P. phaseoli||nd||Phaseolus sp.||nd||nd||HaPphaI||nd||sf||c|
|PiIag||P. infestans||AY894835||S. tuberosum||nd||nd||Ia||nd||nd||d|
|PiIIa||P. infestans||AY898627||S. tuberosum||nd||nd||IIa||nd||nd||d|
|PiIIb||P. infestans||AY898628||S. tuberosum||nd||nd||IIb||nd||nd||d|
|Pmi||P. mirabilis||AY129214||M. jalapa||nd||nd||nd||nd||nd||d|
|Pphas||P. phaseoli||AY129221||P. lunatus||nd||nd||nd||nd||nd||d|
Sixteen isolates were examined in morphological studies (Table 1). Sporangia were observed in an aqueous suspension prepared by flooding a 10-day-old culture grown on pea agar (PA), prepared by autoclaving 120 g frozen peas in 1 L water for 20 min at 121°C. The broth was filtered through cheesecloth to remove the peas and then autoclaved again with 17 g agar L−1. Measurements of 40 randomly selected sporangia for each isolate were made at ×400 magnification. Oospores of P. andina were measured from four in vitro crosses: EC3189 × EC3399, EC3233 × EC3414, EC3658 × EC3358 and EC3818 × EC3261. Diameters of 40 randomly selected oospores for each cross were measured.
Sexual compatibility and mating system
The P. infestans mating system is known to comprise two compatibility types, A1 and A2, and has been used as a phenotypic marker and for genetic studies for decades (Gallegly & Galindo, 1958). The following experiment was designed to learn more of the P. andina mating system. Ten isolates from P. andina (Anarrhicomenum complex; mtDNA Ic) were used for oospore production studies (Tables 1 and 2). These isolates, which reacted as A2 compatibility type when paired with A1 isolates of P. infestans, were mated in all possible combinations (Table 2). Parental isolates were transferred to RA plates and cultured for 10 days at 20°C. Agar discs (5-mm diameter) taken from margins of fast growing colonies of the parental isolates (one disc for each parent) were placed 30 mm apart in a plate (9-cm diameter) containing 10 mL RA, amended with 0·05 g β-sitosterol L−1 to stimulate oospore production (Flier et al., 2001). Two plates were prepared for each parental combination. Plates were incubated for 15 days at 20°C in the dark. The presence and density of oospores were determined microscopically at ×100 magnification.
Electrophoretic banding patterns for the allozymes glucose-6-phosphate isomerase (Gpi, EC 126.96.36.199) and peptidase (Pep, EC 188.8.131.52) were assessed on polyacrylamide gels following a published method (Ordoñez et al., 2000). Polyacrylamide gel electrophoresis (PAGE) was performed using 1-mm-thick 7·5% gels with 25 mm Tris, 0·19 m glycine, pH 8·8 as buffer. Bands were clearer when a 1-cm stacking gel (2·5% acrylamide, 0·06 m Tris-HCl, pH 6·7) was used. PAGE gels were run with a constant current of 5 mA for 1 h, and then increased to 10 mA. Voltage rose continuously throughout, from approximately 50 to 280 V. Allozyme genotypes (inferred from banding pattern phenotypes) were scored as described by Spielman et al. (1990).
Isolates were grown for 10–14 days at 20°C in pea broth. The mycelium grown on pea broth was harvested, lyophilized and stored at −80°C. Lyophilized mycelium (10–20 mg) was ground in microcentrifuge tubes with a pestle and sterile sand. Total DNA was extracted using the Puregene kit (Gentra/Biozym) according to the manufacturer’s instructions. DNA was dissolved in 100 μL TE (10 mm Tris-HCl pH 8·0 and 1 mm EDTA pH 8·0) and stored at −20°C.
Mitochondrial haplotypes were determined by amplification of DNA of each isolate using primers designed for four specific regions of the mitochondrial genome of P. infestans (Griffith & Shaw, 1998). PCR reactions were performed in a PTC200 thermocycler (MJ Research). Digestion of the amplified regions with restriction enzymes CfoI, MspI and EcoRI yielded banding patterns used for classification into mtDNA haplotype: Ia, Ib, IIa IIb (Griffith & Shaw, 1998) and Ic (Oliva et al., 2002).
All Phytophthora isolates from Ecuador and described in a previous study (Adler et al., 2004) were analysed with isolates of P. ipomoeae, P. mirabilis and P. phaseoli. AFLP fingerprinting was performed according to Flier et al. (2003). Selective PCR was performed using Cy5-labelled fluorescent Eco21 primer (5′-CTCGTAGACTGCGTACC-3′), and Mse16 primer (5′-GATGAGTCCTGAGTAACC-3′). A total of 172 distinct and reproducible AFLP bands were identified in 29 isolates. Bands were treated as putative single AFLP loci and a binary matrix for presence (1) or absence (0) of these reproducible bands was constructed and used for further analysis. A similarity matrix of AFLP genotypes was constructed using Rogers’ modified distance method in the statistical software package genstat version 6.1 (Lawes Agricultural Trust). Trees were constructed from the distance matrix using the neighbour-joining algorithm and visualized with treecon version 1.3b software (van de Peer & de Wachter, 1997).
Mitochondrial and nuclear sequences
The mitochondrial CoxII (cytochrome c oxidase subunit II) gene, the ITS regions and part of the single copy Ras gene (Chen & Roxby, 1996) were amplified and sequenced for a subset of 22 Phytophthora isolates collected in Ecuador (Table 1). The CoxII gene was amplified by primers FM35 (5′-CAGAACCTTGGCAATTAGG) and FMPh-10b (5′-GCAAAAGCACTAAAAATTAAATATAA) (Martin & Tooley, 2003; Martin et al., 2004). ITS DNA was amplified using ITS primers 4 and 6 (Cooke et al., 2000). Primers Iras F (5′-TTGCAGCACAACCCAAGACG) and Iras R (5′-TGCACGTACTATTCGGGGTTC) were used to amplify a 223-bp fragment of the intron 1 of the Ras gene (Gomez-Alpizar et al., 2007). Reactions were carried out in a 50-μL volume according to the authors’ description. DNA was sequenced in the forward and reverse directions using a Macrogen system (Macrogen Sequencing System). Sequences were aligned and chromatograms checked using bioedit (Hall, 1999) before alignment with other Phytophthora species from GenBank. Phylogenetic relationships were inferred using neighbour-joining analysis based on the Kimura 2-parameter distance method (Kimura, 1980) under mega 3 (Kumar et al., 2004). In addition, the maximum likelihood (ML) method described in the RAxML web interface (Stamatakis et al., 2005) was also performed.
Plant material of a range of species used for detached-leaf assays came from different sources: host plants collected directly from their natural environments, plants derived from true potato seed (from berries collected in the field) and seed tubers and plants growing near CIP station in Quito, Ecuador. Growing condition and plant accessions were used exactly as previously described by Chacón (2007). Fourteen different inoculation assays were conducted between 2001 and 2004 based on the availability of plant material. In potato, the first pair of fully expanded lateral leaflets of the upper third of the plant was used. In S. ochranthum, the first pair of fully expanded lateral leaflets of mature and well developed leaves was used. In tomato, S. caripense and Anarrhichomenum hosts, whole leaves were employed. In the case of S. betaceum, pieces of the central part of the leaves were cut out and used. For the sake of simplicity, all tissues are hereafter referred to as leaves. Prior to inoculation, leaves were washed with tap water, towel-dried and stored abaxial side up in the lids of inverted Petri dishes which contained water agar (4% w/v) in the base. Petri dishes were subsequently used as high-humidity chambers for inoculation and incubation.
Before inoculation, the isolates from tuber bearing species (such as potato and wild relatives), were multiplied on leaves of their original hosts to restore aggressiveness after cultivation on rye agar. Isolates from S. ochranthum and B. sanguinea were washed directly from pure culture plates (RA). Sporangia from either living tissue or RA medium were rinsed several times with distilled water over a 12-μm filter before use. Irregular sporulation among the isolates made it impossible to use a common inoculum concentration, and therefore concentrations ranged from 2·0 to 3·0 × 104 sporangia mL−1. One 10-μL−1 drop of sporangial suspension was placed on the abaxial side of each leaf close to the midrib.
Given the large number of host–pathogen interactions, difficulty in producing consistent inoculum and the further problem of erratic availability of tissue from such wide diversity of hosts, assays were run sequentially, as inoculum and leaf tissue became available. Overall, 79 host–pathogen interactions were assessed and the following applied to all cases. Each host-genotype × isolate interaction was assessed on six different leaves, distributed into three different Petri dishes. Petri dishes containing inoculated leaflets were incubated by replication (in blocks) at 15 ± 2°C with 14 h fluorescent light per day. Inoculated leaves were monitored 5, 6, 7 and 10 (sometimes 11) days after inoculation for disease development and were visually and qualitatively rated for the presence or absence of disease. Infection (compatible reaction) was defined as the presence of expanding, sporulating lesions. Any other type of reaction observed [no reaction at all; macroscopic hypersensitive response (nonsporulating, nonexpanding lesions); or expanding, nonsporulating lesions] was rated as incompatible. In total, 12 P. andina and 18 P. infestans isolates were tested (Tables 1 and 3).
|Host of origina||RFLPb||Nc||Inoculated hostd|
|P. infestans||S. caripense||US-1||1||0/+||+||+||−||0||+||−||−||−|
|P. andina||S. betaceum||EC-3||5||0||0||0||0/+||0||0||+||+||0/+|
All 16 isolates of P. andina (Table 1) assessed for morphology grew well on RA and PA with a radial colony expansion of 8–12 mm per day at 20°C. Neither hyphal swellings nor chlamydospores were produced in culture. Sporangia were semipapillate (Fig. 2) and although shape and dimension of sporangia varied considerably, no consistent differences among isolates were observed (Fig. 2). Length of sporangia ranged from 39·5 to 62·5 μm (average 44·6 μm) and width from 14·0 to 24·8 μm (average 17·8 μm), with a length:width ratio ranging from 2·4 to 2·7. Zoospores were released within 2 h from sporangia in water at 10°C, but zoospore release was also observed at 20°C on a few occasions. Oospores were formed following mating between compatible parental isolates on RA in Petri dishes (Fig. 3). No systematic differences in oospores size could be detected among the four crosses of P. andina isolates, which ranged overall in diameter from 26·5 to 37·5 μm.
Sexual compatibility in P. andina
Phytophthora andina is a heterothallic species. No oospores were observed in single cultures, and previous studies established that P. andina isolates interacted with P. infestans and could be classified as either A1 or A2 compatibility type (Ordoñez et al., 2000; Chacón, 2007). However, in the present study a complementary system was observed in P. andina, as oospores were produced in crosses between isolates that had been identified as A2 by a P. infestans tester. Oospores were observed in 29 out of 45 pairwise crosses within P. andina (Table 2) and oospore density at the interaction zone ranged from 1 to 160 oospores mm−2 with an average of 39·2 oospores mm−2 (Table 2). Oospores had a dark brown and thick outer wall, but appeared viable, with a large central vacuole and translucent cytoplasm (Fig. 3). Crosses of P. mirabilis × P. andina and P. infestans × P. andina also yielded oospores of similar diameter, but these were lighter in colour (not shown).
All isolates of P. andina had the Pep allele 76 that has not been reported previously for P. infestans. Phytophthora andina alleles for Gpi are common in P. infestans, although the 86 allele has to date only been found in the US-1 lineage of P. infestans (Table 1). Isolates sampled from hosts from the Anarrhichomenum section and S. hispidum share the Gpi 100/100 and Pep 76/100 allele combinations. The combination Gpi 86/100 and Pep 76/100 is typical for EC-3 isolates collected from S. betaceum (Table 1) and also occurs in some isolates from S. quitoense, although the latter had an RFLP genotype that differed from that of EC-3 (data not shown).
Two mtDNA haplotypes were found among isolates of P. andina in Ecuador. Haplotype Ia was associated with isolates from S. betaceum, some isolates from Anarrhichomenum hosts and S. quitoense. The haplotype Ic was associated with isolates from Anarrhichomenum hosts, S. muricatum and S. hispidum (Table 1). Haplotypes Ia, Ib and IIa were found in P. infestans from S. tuberosum and other Solanum hosts, which is consistent with previous reports from Ecuador (Adler et al., 2004). Neither the IIb haplotype, previously described for P. infestans, nor the mtDNA haplotypes typical for P. mirabilis and P. phaseoli (Flier et al., 2002) were found in Phytophthora isolates from Ecuador. However, the Ic haplotype associated with isolates from Anarrhichomenum hosts was found in isolates of P. ipomoeae from Mexico, possibly indicating a shared ancestry (Flier et al., 2002).
Phytophthora andina isolates had either EC-2 or EC-3 RFLP fingerprints. Limited polymorphism that has been described in more detail previously as EC-2.1 (Ordoñez et al., 2000; Adler et al., 2004) was found within the EC-2 group (data not shown). The EC-2.1 variant occurs with both Ic and Ia mtDNA haplotypes. RFLP fingerprints of P. infestans were either EC-1 or US-1, as previously reported (Forbes et al., 1997; Oyarzun et al., 1998; Adler et al., 2004).
Based on a neighbour-joining analysis using 172 AFLP marker loci, 12 isolates of P. andina formed a strongly supported clade (bootstrap support of 99%) clearly distinct from isolates of four other Phytophthora species included in the analysis (Fig. 4). Within the P. andina clade, three groups could be distinguished, one associated with the EC-3 lineage from S. betaceum, and two additional clusters associated with isolates of the EC-2 lineage that included both Ic and Ia mtDNA haplotypes.
CoxII sequence data grouped isolates of P. infestans in one clade with all isolates of P. andina that have the Ia mtDNA haplotype. Isolates of P. andina with the Ic mtDNA haplotype formed a distinct branch, as did individual isolates of P. mirabilis and P. phaseoli (Fig. 5). CoxII sequences generated in this study were submitted to GenBank (Accession Nos GQ260982 to GQ261004). ITS sequences were identical among isolates of P. andina and similar to those of P. infestans (GenBank Accession No. AY770741). The sequence of the Ras intron 1 gene was identical for all isolates of P. infestans, including GenBank Accession No. U30474. All isolates of P. andina had identical sequences as well, but this sequence differed from that of P infestans by 9 bp (data not shown). In addition to the previous report made by Gomez-Alpizar et al. (2008), who described six heterozygous sites, this study found three more sites within the Ras intron 1 sequence (GenBank Accession Nos. GQ261005 to GQ261027). The additional sites corresponded to positions 551, 610 and 688, according to Accession No. U30474. Within the 223-bp fragment, heterozygous sites were found in all isolates of P. andina, but these were absent in P. infestans.
Isolates of both P. infestans RFLP lineages, US-1 and EC-1, caused lesions on most of the solanaceous host species examined, whilst isolates of P. andina had a narrower host range (Table 3). Based on qualitative assessment of pathogenicity of the 79 host–pathogen interactions, there was a tendency for isolates to more consistently infect the host they were isolated from (Table 3). None of the P. andina isolates were pathogenic on potato cv. Cruza 148, whilst some caused weak infections on cv. Yungay, which is more susceptible. Other solanaceous hosts, such as S. caripense and S. ochranthum, were fully resistant to the P. andina isolates tested (Table 3). None of the isolates were pathogenic on S. muricatum.
Phytophthora andina Adler & Flier, sp. nov Coloniis mycelialibus in Secal bene crescentibus. Temperies ad crescendum neccessaria. Minima temperies super 5°C, maxima temperies infra 30, optima temperies 24°C. Hyphae eseptatae et copiose ramosae, 4–8 μm diam. Sporangiophori aerii in agaro ramis composito-sympodialibus et indeterminatis, cum tumoribus in loco sporangiis emergentes. Sporangia semipapillata, ellipsoidea, subovoidalibus, caduca cum pedicella brevi, valore medio 44·6 μm longa (variatione 39·5–62·5 μm), ratione longitudinis/latitudinis 2·4–2·7, germinantia directe tubo germinativo vel indirecte cum zoosporis 6–8. Antheridia amphigyna, valore medio 22·0 μm longa, ratione longitudinis/latitudinis 1·4. Oogonia laevitunicata, 34–41 μm diam., basi attenuata. Oosporae laevitunicatae colore luteo, 27–37·5 μm diam.. Segregatus heterothallicis (Figs 1 and 3).
Typus: Ecuador: Province of Tungurahua: City of Baños isol. Ex Solanum brevifolium along roadside, July 2001, N. A. Adler, CIP EC3189, deposited in World Phytophthora Collection (University of California at Davis). Mycobank number MB514427.
Mycelial colonies grow well on RA. Minimum growth at approximately 5°C, optimum at 24°C and maximum at 30°C. Hyphae nonseptate and freely branching, hyphal diameter 4–8 μm, mostly 5·6 μm. Sporangiophore aerial, on rye agar with compound-sympodial and intermediate branches, with swellings where sporangia emerge. Sporangium semipapillate, ellipsoid or semi-ovoid, caducous with short pedicel, on average 44·6 μm long (ranging from 39·5 to 62·5 μm), with a length/width ratio of 2·4 to 2·7, germinating directly with germ tubes or indirectly with 6–8 zoospores. Antheridia amphigynous, average length 22·0 μm, ratio of length/width 1·4. Oogonia smooth-walled, average diameter 34·0–41·0 μm, with tapered base. Oospores smooth-walled, tinted yellow-brown, almost filling the oogonial cavity, average diameter 31·0 μm. Isolates heterothallic.
This paper provides morphological and molecular evidence for the designation of P. andina as a distinct species attacking solanaceous plants in the Andean highlands. This designation is based on examination of a collection of isolates from different hosts made over a 10-year period. Based on AFLP, RFLP, allozyme genotypes and DNA sequence analyses, P. andina is closely related to P. infestans; however, these same markers indicate very little evidence of gene flow between the two species in nature, even though the potential for sexual reproduction has been established in vitro (Oliva et al., 2002). Molecular data presented elsewhere also (Wattier et al., 2003; Kroon et al., 2004; Oliva et al., 2007; Blair et al., 2008; Gomez-Alpizar et al., 2008) supports a new species designation for P. andina.
Analysis of genetic variation at the rRNA ITS region (Kroon & Flier, unpublished data) revealed that P. andina is very closely related to several other foliar pathogens classified into Phytophthora clade Ic (Cooke et al., 2000). The rRNA ITS sequence analysis identified P. andina as a putative sister group to P. infestans, P. mirabilis, P. ipomoeae and P. phaseoli, a result consistent with the AFLP marker data shown here (Fig. 4) and by Adler et al. (2004). Additional support for the classification of P. andina as a sister species of P. infestans and P. mirabilis was provided by the high-resolution molecular phylogeny for the genus Phytophthora published by Kroon et al. (2004) and the phylogenies proposed by Gomez-Alpizar et al. (2007, 2008) and Blair et al. (2008).
Phytophthora andina, P. mirabilis and P. infestans are morphologically similar in that they have a heterothallic mating system with amphigynous antheridia and produce large semipapillate sporangia on long sporangiophores. Unique features of P. andina (other than host range) are the EC-2 and EC-3 RFLP and AFLP fingerprints, heterozygosity in the ras intron 1 sequence, the Ic mtDNA haplotype and the Pep 76 allele.
The origin of P. andina is not directly addressed in this paper, but it is possible that P. infestans and P. andina may have evolved either from a common ancestor or as a consequence of interspecific hybridization. The presence of the Ia haplotype and the Gpi 86 allele in both P. infestans and P. andina suggests that hybridization may have occurred. This finding is consistent with the results of Gomez-Alpizar et al. (2008), who found the same sequence of the Cox I gene among isolates comprising both P. infestans and isolates of P. andina with the Ia mtDNA haplotype. A representative sample of blight-causing Phytophthora genotypes with different host preferences was selected in the present study, and it was found that some P. andina isolates have a unique Cox II sequence, but others share the same sequence with P. infestans. The occurrence of similar sequences in both Cox I and Cox II genes is also consistent with the hypothesis of previous hybridization, in which P. infestans must have been involved as one of the parents. Alternatively, both species could also have descended from a common mitochondrial ancestor (Gomez-Alpizar et al., 2007). The contribution of other Phytophthora species in the clade to the evolution of P. andina remains unclear and needs to be tested further using more rigorous statistical approaches.
Populations of P. andina appear to be clonal, with three predominant lineages. One lineage is characterized by the Ic mtDNA haplotype (Table 1) and the A2 mating type (using P. infestans testers), and has been isolated from hosts of the Anarrhichomenum section of Solanum and S. hispidum. The second lineage has also been isolated from hosts from the Anarrhichomenum section, but it has the Ia mtDNA haplotype and the A1 compatibility type with P. infestans tester isolates. Similar genotypes have also been isolated from S. quitoense. A third lineage is characterized by the EC-3 RFLP fingerprint and the Ia mtDNA haplotype, and to date has only been isolated from the cultivated tree tomato (S. betaceum) (Table 1). Two of the three lineages were distinguished in the cluster analysis of AFLP data (Fig. 4).
When 10 isolates of P. andina were crossed among themselves, oospores were produced in a number of the crosses. Since all 10 of these isolates react as A2 when tested against an A1 isolate of P. infestans, it would appear that whilst P. andina isolates can be classified for mating type with P. infestans testers, the mating system within P. andina is more complex. This highlights the limited utility of using mating-type testers from one species to classify isolates of another species. The incongruence of mating systems (complementary but not identical) is consistent with the designation of P. andina as a separate species. No attempt was made at this point to describe the mating system in P. andina as the sample was small (10 isolates) and not representative; all isolates came from hosts of the Anarrhichomenum section. Further work is needed to address mating behaviour in this species.
The complexity identified in the mating system of this species is not necessarily unique to P. andina with this Phytophthora clade. No mating system complementary to that of P. infestans has apparently been described for P. mirabilis.Goodwin & Fry (1994) were able to cross P. infestans with P. mirabilis and obtain hybrid oospores, but host specialization was suggested as a mechanism of reproductive isolation. In another study (R.F. Oliva, unpublished data) it was demonstrated that crosses between P. andina and P. infestans resulted in oospores that could survive up to 1 year in soil and hybrid progeny that could infect potato in the laboratory, but insufficient studies of viability of oospores resulting from crosses of P. andina by P. andina were done. Given the potential implications for genetic recombination, hybridization and disease initiation from oospores, further studies to characterize the mating systems of P. andina and oospore viability and epidemiology are needed.
In tree hosts (S. betaceum, S. quitoense and S. hispidum), P. andina causes similar but much larger leaf lesions to late blight on potato, with longer infectious periods. The pathogen also causes very serious stem infections (Fig. 1) on S. betaceum. Blight symptoms in general on the Andean solanaceous hosts appear to be defined more by the host than the pathogen. On the Anarrhichomenum hosts (Fig. 1), which produce vine–like growth in the underbrush, disease symptoms are very much like late blight of potato or tomato. Detached-leaf inoculations in the laboratory indicated that P. andina can also infect and sporulate on leaves of wild or cultivated potatoes. In nature, however, the pathogen appears to have strong host preferences, and through a decade of isolation studies P. andina lesions were not found on potato or tomato plants (Adler et al., 2004; Oliva et al., 2007). However, there are solanaceous hosts which P. andina and P. infestans may co-infect, and given the comments above on the potential for hybridization, this could be important. On two occasions, P. andina was isolated from S. muricatum, which is generally a host of US-1 genotypes of P. infestans (Adler et al., 2002). In each of those cases, P. andina appeared to cause disease of epidemic proportions; all isolates were P. andina. There are also other hosts from which both species have been isolated, including S. quitoense and S. ochranthum (R.F. Oliva, unpublished data).
Similar to P. infestans on potato or tomato, P. andina appears to be a pathogen of increasing and potentially great impact on native Andean fruit crops, some of which have or are being traded globally. Although epidemiological studies have not been done, a marked increase in severity of disease on S. betaceum can be noted. In the mid 1990s, P. andina infection on S. betaceum was difficult to find; now, for collection, one simply has to go to any of the major production areas and disease is always present. The dynamics of P. andina may also include S. quitoense in Ecuador and Colombia, which is produced in restricted regions at mid-level elevations (1000–2000 m a.s.l.) and occupies a different agroecological zone than does S. betaceum. Outside the genus Solanum, less is known about the occurrence of P. andina on other solanaceous crops. Late blight of Cape gooseberry (Physalis peruviana) was recently reported to be caused by P. infestans in Colombia (Vargas et al., 2009), although the markers used by the authors would not have distinguished P. infestans from P. andina.
Some hosts described in this study, which are commonly attacked by P. andina, are no longer restricted to the Andes. Perhaps the best example is S. betaceum, which is widely cultivated as small orchards around the world, but also considered economically important in some non-Andean countries, particularly New Zealand. Currently, there are no reports of the pathogen outside the Andean highlands.
Analyses of the impact of this pathogen on native fruit crops in the Andean highlands and studies on the evolutionary relationships of this species with other Phytophthora pathogens attacking solanaceous hosts are in progress.
We wish to express our thanks to the Instituto Nacional Autonomo De Investigaciónes Agropecuarias (INIAP) in Quito, Ecuador, for having granted CIP and PRI collection permits for investigating the diversity of P. andina and P. infestans in Ecuador. The Swiss Federal Institute of Technology (ETH-Z) is thanked for their participation in this project. Part of this work was done during a stay of N. Adler at CIP, financed by a grant from the Dutch Ministry of Agriculture, Nature Management and Food Safety. Thanks go to Trudy van den Bosch and Henry van Raaij for expert laboratory assistance, and to Chia-Hui (Julia) and Luis Gomez-Alpizar for providing sequence data from their studies and for review of the manuscript, and Marcelo Vinueza for culture work. This work was funded in part by the IPM CRSP of the USAID and the Swiss Center for International Agriculture (ZIL).
- 2002. The Andean fruit crop, pear melon (Solanum muricatum), is a common host for A1 and A2 strains of Phytophthora infestans in Ecuador. Plant Pathology 51, 802. , , , ,
- 2004. Genetic diversity of Phytophthora infestans sensu lato in Ecuador provides new insight into the origin of this important plant pathogen. Phytopathology 94, 154–62. , , et al. ,
- 2008. A multi-locus phylogeny for Phytophthora utilizing markers derived from complete genome sequence. Fungal Genetics and Biology 45, 266–77. , , , , ,
- 1968. Spontaneous variability of single isolates of Phytophthora infestans. I. Cultural variation. Canadian Journal of Botany 46, 329–48. , ,
- 2007. Variability of Phytophthora infestans sensu lato in the Highland Tropics of Ecuador. Zurich, Switzerland: Swiss Federal Institute of Technology Zurich, PhD thesis. ,
- 1996. Characterization of a Phytophthora infestans gene involved in vesicle transport. Gene 181, 89–94. , ,
- 2000. A molecular phylogeny of Phytophthora and related oomycetes. Fungal Genetics and Biology 30, 17–32. , , , , ,
- 2001. Formation, production and viability of oospores of Phytophthora infestans from potato and Solanum demissum in the Toluca Valley, central Mexico. Mycological Research 105, 998–1006. , , , ,
- 2002. Phytophthora ipomoeae sp. nov., a new homothallic species causing leaf blight on Ipomoea longipendunculata in the Toluca Valley of central Mexico. Mycological Research 106, 848–56. , , et al. ,
- 2003. The population structure of Phytophthora infestans from the Toluca Valley of Central Mexico suggests genetic differentiation between populations from cultivated potato and wild Solanum spp. Phytopathology 93, 382–90. , , et al. ,
- 1997. Population genetic structure of Phytophthora infestans in Ecuador. Phytopathology 87, 375–80. , , et al. ,
- 1958. Mating types and oospores of Phytophthora infestans in nature in Mexico. Phytopathology 48, 274–7. , ,
- 2007. An Andean origin of Phytophthora infestans inferred from mitochondrial and nuclear gene genealogies. Proceedings of the National Academy of Sciences, USA 104, 3306–11. , , ,
- 2008. Phylogenetic relationships of a new species, Phytophthora andina, from the highlands of Ecuador that is closely related to the Irish Potato Famine pathogen Phytophthora infestans. Mycologia 100, 590–602. , , , , ,
- 1994. Genetic analyses of interspecific hybrids between Phytophthora infestans and Phytophthora mirabilis. Experimental Mycology 18, 20–32. , ,
- 1992. Cloning and genetic analyses of two highly polymorphic, moderately repetitive nuclear DNAs from Phytophthora infestans. Current Genetics 22, 107–15. , , ,
- 1998. Polymorphisms in Phytophthora infestans: four mitochondrial haplotypes are detected after PCR amplification of DNA from pure cultures or from host lesions. Applied and Environmental Microbiology 64, 4007–14. , ,
- 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis for Windows 95/98/NT. Nucleic Acids Symposium Series 41, 95–8. ,
- 1990. The Potato: Evolution, Biodiversity and Genetic Resources. London, UK: Belhaven Press. ,
- 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16, 111–20. ,
- 2004. Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genetics and Biology 41, 766–82. , , , , ,
- 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics 5, 150–63. , , ,
- 2003. Phylogenetic relationships among Phytophthora species inferred from sequence analysis of mitochondrially encoded cytochrome oxidase I and II genes. Mycologia 95, 269–84. , ,
- 2004. Molecular detection of Phytophthora ramorum and a Phytophthora ilicis-like species associated with sudden oak death in California. Phytopathology 94, 621–31. , , ,
- 2002. Potential of sexual reproduction among host-adapted populations of Phytophthora infestans sensu lato in Ecuador. Plant Pathology 51, 710–9. , , , ,
- 2007. Is Phytophthora infestans a good taxonomist? Host recognition and co-evolution in the Phytophthora/Solanum interaction. Acta Horticulturae 745, 465–71. , , , , ,
- 2000. A novel population of Phytophthora, similar to P. infestans, attacks wild Solanum species in Ecuador. Phytopathology 90, 197–202. , , et al. ,
- 1998. Host specificity of Phytophthora infestans on tomato and potato in Ecuador. Phytopathology 88, 265–71. , , , , ,
- 1997. Construction of evolutionary distance trees with TREECON for Windows: accounting for variation in nucleotide substitution rate among sites. Computer Applications in the Biosciences 13, 227–30. , ,
- 1989. Lost Crops of the Incas: Little-Known Plants of the Andes with Promise for Worldwide Cultivation. Washington DC, USA: Office of International Affairs. , , , , , ,
- 1990. The genetics of Phytophthora infestans: segregation of allozyme markers in F2 and backcross progeny and the inheritance of virulence against potato resistance genes R2 and R4 in F1 progeny. Experimental Mycology 14, 57–69. , , , ,
- 2003. Species concepts and relationships in wild and cultivated potatoes. Acta Horticulturae 619, 63–75. , , , ,
- 2005. RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics 21, 456–63. , , ,
- 2009. Characterization of Phytophthora infestans population in Colombia: first report of the A2 mating type. Phytopathology 99, 82–8. , , et al. ,
- 1963. Key to the Species of Phytophthora de Bary. Mycological Papers, No. 92. Kew, UK: Common-wealth Mycological Institute. ,
- 2003. Sequence variation of intergenic mitochondrial DNA spacers (mtDNA-IGS) of Phytophthora infestans (oomycete) and related species. Molecular Ecology Notes 3, 136–8. , , et al. ,