Introduction

The genus Phytophthora includes many agriculturally and ecologically important plant pathogens. It currently contains approximately 120 species [1]. These species were traditionally divided into six groups by morphological features [2]. They have been classified into 10 clades according to phylogenetic analyses of nuclear and mitochondrial sequences [3]–[8]. Members of this genus are capable of surviving in a variety of terrestrial and aquatic habitats [9]. However, species in certain clades or subclades are better adapted to specific ecosystems. For example, most clade 1 species such as P. infestans [9] and P. hedraiandra [10] appear as terrestrial pathogens which attack above-ground plant tissues, while many species in subclade 6b and clade 9 are often associated with aquatic environments such as irrigation reservoirs [11]–[17], rivers and riparian ecosystems [18], [19].

Even though Phytophthora species were among the earliest described plant pathogens, investigations into their interspecific hybridization were initiated only recently. One of the first studies describing this phenomenon was conducted in 1991 which revealed that some isolates initially assigned as P. meadii were actually polyploid and might be hybrids based on cytological evidence [20]. Thereafter, several artificial hybrids: P. infestans×P. mirabilis, P. nicotianae×P. capsici, P. sojae×P. vignae, and P. capsici×P. tropicalis have been produced by pairing in dual culture [21]–[23], zoospore fusion [24], [25], and nuclear transplantation [26]. In the meanwhile, eleven natural Phytophthora hybrids have been reported. These include P. ×pelgrandis (P. nicotianae×P. cactorum) [27]–[30], P. alni including three subspecies: P. alni subsp. alni, P. alni subsp. uniformis and P. alni subsp. multiformis [31]–[33], P. andina with P. infestans as one parent [34]–[36], P. ×serendipita (P. cactorum×P. hedraiandra) [29], [37], four hybrids in subclade 6b: P. amnicola×P. taxon PgChlamydo (A-PG), P. taxon PgChlamydo×P. amnicola (PG-A), P. thermophila×P. amnicola (T-A), and P. thermophila×P. taxon PgChlamydo (T-PG) [38], as well as three hybrids in subclade 8b: P. porri×P. taxon parsley, P. porri×a P. primulae-like species, and a third hybrid with two unknown species as parents [39]. It is interesting to note that parents of most individual hybrids belong to the same Phytophthora clade. The only inter-clade hybrid is P. nicotianae (clade 1)× P. capsici (clade 2), which was produced by zoospore fusion [24], [25] and nuclear transplantation [26].

A number of Phytophthora hybrids are emerging plant pathogens. By inheriting and recombining alleles or genes from both parents followed by rapid evolution [39]–[41], these hybrids have broader host ranges [39] and produce new virulence factors with higher aggressiveness, while overcoming weaknesses of their parental species. For example, P. alni and its variants are destructive pathogens that have killed more than 10,000 riparian Alnus trees in Europe in 1996 alone [31]. Phytophthora ×pelgrandis was found infecting plants in the genera of Cyclamen, Eriobotrya, Lavandula, Lewisia, Pelargonium, Primula, and Spathiphyllum in the Netherlands, Germany, Italy, Peru and Taiwan [27]–[30], [42], [43]. Phytophthora ×serendipita has been isolated from hosts in the genera of Idesia, Penstemon, Allium, Rhododendron, Kalmia, and Dicentra in Europe and the United States, while its parent P. hedraiandra only infects Rhododendron and Viburnum species, indicating this emerging hybrid pathogen has successfully utilized new habitats and adapted to novel hosts [29], [37]. Phytophthora porri×P. taxon parsley in subclade 8b has shown a similar expansion of host range including Allium victorialis, Allium grayi, Pastinaca sativa, Chrysanthemum species, and Parthenium argentatum, while its parents only infect leek and parsley [39]. Although their host ranges are unknown, the four subclade 6b hybrids A-PG, PG-A, T-A, and T-PG, which originated in Australia, have exploited new habitats in South Africa [38]. It must be noted that sexual reproduction of most Phytophthora hybrids is compromised due to their nature of allopolyploidy and resulting genetic incompatibility. Most Phytophthora hybrids are sterile, nonfunctional in meiosis, or produce numerous abortive oospores [29], [32], [38], [39].

Since 2005 we have obtained more than twenty isolates of a previously unknown Phytophthora species from irrigation systems. It has distinct morphology from all known species. Also, continual polymorphic sequences in the internal transcribed spacers (ITS) region of all isolates suggest that this is a Phytophthora hybrid. Here, we examine and describe its morphological, physiological and molecular characters and name this new hybrid as Phytophthora ×stagnum nothosp. nov.

Materials and Methods

Results

Sequence analysis of cox 1 gene

All four representative isolates of P. ×stagnum produced an identical 867-bp cox 1 sequence, which is distinct from those of all known Phytophthora species. This new species and P. mississippiae isolate 57J3 (GenBank Accession No. KF112860) differ by 18 bp in the alignment of cox 1 sequences. In the ML phylogenetic tree based on cox 1 sequences of P. ×stagnum and other selected species, P. ×stagnum isolates clustered in a distinct taxon which is closely related to P. mississippiae (Figure 1), indicating the maternal parent of P. ×stagnum is genetically close to P. mississippiae.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Maximum Likelihood phylogenetic tree based on mitochondrial cox 1 sequences of Phytophthora ×stagnum and representative species.

Alignment was conducted with MAFFT version 7. Phylogenetic tree was generated in MEGA5. GenBank accession numbers of sequences are given following the species names and isolate codes. Bootstrap values are shown on branches (1,000 replicates; values <50% are not shown).

https://doi.org/10.1371/journal.pone.0103450.g001

Sequence analysis of ITS clones

Among the 94 clones of the ITS region, 61 resulted in high-quality sequences. These included 16, 16, 16, and 13 from isolates 36H8, 36J7, 43F3, and 44F9, respectively. In the alignment of these 61 ITS sequences, 35 rare single-nucleotide polymorphism (SNP) sites occurred at low frequencies (∼1/61). These rare SNPs were mostly intraspecific polymorphisms of parent species of P. ×stagnum. We also observed six frequent SNPs (four in ITS1 region and two in ITS2 region) and three indels in the ITS1 region (Figure 2) at high frequencies (10–31/61).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Internal transcribed spacer (ITS) sequence alignment of Phytophthora ×stagnum, P. mississippiae and P. taxon PgChlamydo.

Position numbers are given based on the alignment. Yellow indicates sequences belong to P. taxon PgChlamydo authentic isolates. Blue indicates sequences belong to P. mississippiae type isolate 57J3. ^aPosition 15 is in the poly(A) region of ITS 1, which may contain sequencing errors. Thus, it is excluded from the analysis of hybridization. However, the indel of position 15 among three types of P. ×stagnum clones explains continual polymorphism and unreadable sequences of the ITS 1 regions amplified with the forward primer ITS6F in the initial sequencing before cloning. ^bType 1 and 2 occurred 21 and 10 times among 31 clones producing 816-bp sequences.

https://doi.org/10.1371/journal.pone.0103450.g002

These 61 clones can be generally grouped into three types by the six frequent SNPs and three indels in the ITS sequence. Thirty-one clones produced two types of 816-bp sequences while the other 30 clones produced an identical 819-bp sequence. According to the sequence alignment (Figure 2), the two 816-bp sequences are 99% identical to that of P. mississippiae, while the 819-bp sequence is ∼100% identical to those of P. taxon PgChlamydo in GenBank (www.ncbi.nlm.gov/genbank/). Clones of individual representative isolates produced all ITS sequence types.

Sequence analysis of beta-tubulin gene

Isolates 36H8, 36J7, and 44F9 resulted in an identical 1124-bp beta-tubulin sequence with 26 polymorphic positions (Figure 3). Isolate 43F3 also produced a 1124-bp sequence with 24 of the same 26 polymorphic positions. In spite of the polymorphic positions, beta-tubulin sequences of P. ×stagnum are identical to that of P. taxon PgChlamydo and 11 bp different from that of P. mississippiae (Figure 3). Sequences of P. taxon PgChlamydo and P. mississippiae are distinct and both occur at 18 of the 26 polymorphic positions of P. ×stagnum containing ambiguous sequences such as positions 99, 102, and 261 (Figure 3). At the other eight polymorphic positions such as positions 93, 306, and 450, both species share the same sequences which also occur as one of the ambiguous polymorphic sequences of P. ×stagnum (Figure 3). Putative sequences of the maternal parent of P. ×stagnum are shown in Figure 3. The maternal parent is approximately 21 bp different from P. mississippiae in beta-tubulin sequence.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Beta-tubulin sequence alignment of Phytophthora ×stagnum, P. mississippiae and P. taxon PgChlamydo.

Position numbers are given based on the alignment. Shading color: green indicates ambiguous sequences of polymorphic positions; yellow indicates sequences of the paternal parent P. taxon PgChlamydo; blue indicates sequences of the putative maternal parent. ^aY = T and C; S = G and C; M = A and C; R = G and A.

https://doi.org/10.1371/journal.pone.0103450.g003

Colony morphology

The four representative isolates had a similar growth pattern after 10-days incubation in the dark at 20°C (Figure 4). Colony pattern on CA and CV8A was stellate to radiate with a relatively smooth edge and abundant aerial mycelia at the center. Colony pattern on MEA and PDA was rosaceous except isolate 43F3, which produced a slightly cottony colony on PDA. Colony growth of all isolates was slowest on MEA among tested media.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Colony morphology of Phytophthora ×stagnum representative isolates on various media incubated at 20°C for 10 days in the dark.

CA = carrot agar, CV8A = 20% clarified V8 juice agar, MEA = malt extract agar, PDA = potato dextrose agar.

https://doi.org/10.1371/journal.pone.0103450.g004

Cardinal temperatures for vegetative growth

Radial growth rates were similar among four representative isolates (P = 0.71) and between two cardinal temperature tests (P = 0.74). Thus, data from the repeated tests were pooled and averages were plotted against temperature (Figure 5). The optimum temperature for the vegetative growth of P. ×stagnum in both media was 25°C. It also grew well at 30°C on both media. Limited growth occurred at 5 and 35°C. No growth was observed at 40°C.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 5. Average daily radial growth of Phytophthora ×stagnum representative isolates in carrot agar (CA) and 20% clarified V8 juice agar (CV8A) over an 8-days period.

https://doi.org/10.1371/journal.pone.0103450.g005

Taxonomy

Phytophthora ×stagnum.

X. Yang & C. X. Hong nothosp. nov. (Figure 6).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 6. Morphology of Phytophthora ×stagnum.

(a–g) Nonpapillate and noncaducous sporangia in various shapes; (a, b) Ovoid sporangia; (c, d) Ovoid to ellipsoid sporangia; (e, f) Obpyriform sporangia; (g) A germinated sporangium in distorted shape; (h) Direct germination of an obpyriform sporangium; (i) A sporangium releasing zoospores; (j) Nested internal proliferation; (k) Extended internal proliferation; (l) Hyphal swellings; (m) Catenulate hyphal swellings; (n) Thin-walled intercalary chlamydospores and hyphal swellings; (o) A thin-walled intercalary chlamydospore; (p) A thick-walled chlamydospore; (q–t) Ornamented, aborted oogonia produced by isolates 36H8 and 36J7; (q) An oogonium with an amphigynous antheridium; (r–t) Ornamented oogonia with distorted antheridia; (u, v) Relatively smooth-walled, darkly pigmented oogonia produced by isolates 43F3 and 44F9; (u) An oogonium containing a plerotic oospore with a globose antheridium; (v) Oogonia with abortive oospores. Bars = 10 µm.

https://doi.org/10.1371/journal.pone.0103450.g006

MycoBank: MB807978 [urn:lsid:mycobank.org: MB807978].

Sporangia were occasionally produced by aged cultures (>30 days) grown in CA and CV8A. Abundant sporangia were produced from fresh mycelial plugs submerged in 1.5% SWE within 10 hours. Sporangial shape varied from ovoid (Figures 6a, b) to ellipsoid (Figures 6c, d), obpyriform (Figures 6e, f) and distorted shapes (Figure 6g). Sporangia were terminal, nonpapillate and noncaducous. They ranged from 30.5 to 89.7 µm in length (average 54.3±11.0 µm) and 17.5 to 40.4 µm in width (average 30.3±3.9 µm). Direct germination of sporangia was frequently observed (Figures 6g, h). Nested and extended internal proliferations were common (Figures 6j, k). Hyphal swellings in irregular shapes were abundantly produced in both young and aged cultures (Figures 6l, m). Catenulate, globose hyphal swellings were frequently observed in aged cultures (Figure 6m). Intercalary chlamydospores were observed in aged cultures of all examined isolates (Figures 6n, o, p). They were mostly thin-walled (Figures 6n, o), rarely thick-walled (Figure 6p), and averaged 33.5±4.9 µm in diameter.

Phytophthora ×stagnum is heterothallic and all isolates examined are A1. They produced no sexual structure in single culture. Oogonia were produced in dual culture when each P. ×stagnum isolate was paired with an A2 tester of P. cinnamomi. In the polycarbonate test, a limited number of gametangia (∼40) were produced by the four isolates after being paired with an A2 mating type tester of P. nicotianae for more than 50 days. Two distinct groups of gametangia were observed. Isolates 36H8 and 36J7 mostly produced ornamented oogonia with characteristic protuberances (Figures 6q, r, s, t). These oogonia averaged 33.6±8.1 µm in diameter. Oogonial wall was pigmented golden at maturity. All observed ornamented oogonia aborted (Figures 6q, r, s, t). Antheridia were amphigynous, commonly distorted (Figures 6s, t). They averaged 19.4 µm in depth and 14.2 µm in width. Isolates 43F3 and 44F9 mostly produced oogonia with a relatively smooth surface (Figures 6u, v). These oogonia averaged 28.0±5.6 µm in diameter. The oogonial wall was darkly golden-brown. Plerotic oospores (Figure 6u) were also mostly aborted (Figure 6v). Antheridia were amphigynous, globose or distorted, and averaged 10.0 µm in depth and 12.3 µm in width (Figures 6u, v).

Discussion

Sequence analyses of the cox 1, ITS, and beta-tubulin genes have demonstrated that Phytophthora ×stagnum is a hybrid species with a species genetically close to P. mississippiae as its maternal and P. taxon PgChlamydo as its paternal parent. First, the mitochondrial cox 1 sequence of P. ×stagnum is mostly analogous to that of P. mississippiae (Figure 1), suggesting that its maternal parent is genetically close to P. mississippiae. Second, cloning of the ITS region of P. ×stagnum isolates consistently resulted in two types of 816-bp sequences and one type of 819-bp sequence. The 819-bp sequence is identical or only 1-bp different from those of authentic P. taxon PgChlamydo isolates [18]. The two types of 816-bp sequences only differ from that of the P. mississippiae type isolate [14] by 3 or 6 bp (Figure 2). Third, P. ×stagnum contains the beta-tubulin sequences of P. taxon PgChlamydo and P. mississippiae at 26 polymorphic positions (Figure 3). Its sequences at non-polymorphic positions are identical to that of P. taxon PgChlamydo and only ∼10 bp different from that of P. mississippiae. These results of ITS and beta-tubulin sequence analyses indicate that P. ×stagnum is a hybrid between P. taxon PgChlamydo and a species genetically close to P. mississippiae.

This hybrid species has diagnostic morphological and physiological characters of P. taxon PgChlamydo and P. mississippiae. For instance, P. ×stagnum is similar to P. taxon PgChlamydo in producing chlamydospores, which are not produced by P. mississippiae [14]. However, both P. ×stagnum and P. mississippiae produce abundant catenulate hyphal swellings (Figure 6m) in aged cultures, as well as nested or extended internal proliferations (Figures 6j, k). Also, both P. ×stagnum and P. mississippiae produce ornamented oogonia (Figures 6q–t). The relatively smooth-walled oogonia produced by P. ×stagnum (Figures 6u, v) may implicate the oogonial morphology of P. taxon PgChlamydo although it has not been reported. In addition, P. ×stagnum is similar to P. mississippiae in colony morphology and growth rate on CV8A [14]. Both species produce radiate to slightly petaloid colonies with a relatively smooth edge (Figure 4) and the fastest growth on CV8A occurs at 25°C (Figure 5). Phytophthora ×stagnum can be separated from both parents by its optimal growth temperature on CA at 25°C (Figure 5), while it occurs at 30°C for P. mississippiae [14] and about 28°C for P. taxon PgChlamydo [18].

Although we have identified the two parent species of P. ×stagnum by molecular and morphological evidences, the mechanism by which this subclade 6b hybrid was produced remains unknown. It seems likely that this new hybrid formed asexually. One major reason is that species in subclade 6b tend to be homothallic as exemplified by P. gibbosa, P. gregata, and P. megasperma [19], [53], or “sterile” with unknown sexual structures such as P. amnicola, P. thermophila and P. taxon PgChlamydo [18], [19], [54]. This tendency may be a result of their adaptation to aquatic habitats [18], [19], [38]. The four subclade 6b hybrid species reported in 2013, PG-A, A-PG, T-A, T-PG also produced no gametangia [38]. In this study, we only observed a limited number of sexual bodies of P. ×stagnum (∼40) in five polycarbonate-membrane tests. These results along with previous findings indicate that hybrids in subclade 6b were more than likely formed asexually via hyphal anastomosis or zoospore fusion. However, Nagel et al. [38] suggested that the conditions used in laboratory mating tests may be not conducive to the formation of sexual bodies of subclade 6b species, while suitable conditions may exist in natural environments [38]. The formation mechanism of sexual structures of these subclade 6b hybrid species warrants further investigations.

Aquatic environments are ideal for the development and survival of natural Phytophthora hybrids. Many known Phytophthora hybrids have close association with aquatic environments. Examples include the four subclade 6b hybrids recovered from river and riparian ecosystems [38]; Phytophthora alni and its variants associated with riparian Alnus trees [31], [32]; and Phytophthora ×pelgrandis initially recovered from horticultural plants grown in hydroponic systems [28]. The fact that most natural Phytophthora hybrid species were initially identified from aquatic environments is interesting. First, natural aquatic ecosystems such as rivers, streams, and riparian habitats provide ideal environments for many plant species to grow. Consequently, Phytophthora species from various plant hosts have greater chances to aggregate and subsequently form hybrids under suitable conditions, such as P. alni and subclade 6b hybrids described in 2013. Similarly, hundreds of ornamental plants are grown in nurseries using hydroponic or recycling irrigation systems which greatly increase the chance of close contact between species. Phytophthora ×pelgrandis, P. ×serendipita, and P. ×stagnum may have formed in these systems by mating or anastomosis [28]. Second, newly formed Phytophthora hybrids may have a better opportunity to survive and adapt to aquatic ecosystems that contain a diverse variety of plant species. Third, aquatic environments favor asexual reproduction via motile zoospores or chlamydospores. This may be important for species that are sterile or nonfunctional in sexual reproduction as are all known Phytophthora hybrids. Fourth, for the saprophytic Phytophthora species in subclade 6b including PG-A, A-PG, T-A, T-PG [18], [19], [38] as well as P. ×stagnum in this study, the abundant plant debris in aquatic environments provides ideal microhabitats and nutrient sources. Fifth, water also offers hybrids vehicles for mobility compared to terrestrial environments, which may allow them to migrate into new habitats. In summary, aquatic environments may provide favorable conditions for Phytophthora hybrids to form, survive and disseminate.

All four representative isolates of Phytophthora ×stagnum are genetically stable. They were routinely subcultured on artificial media during the experimental period (∼2 years) and did not revert to either parent type. Also, sequencing of the ITS region of representative isolates was conducted several times in three years (2008, 2012, and 2013), and all ITS sequences obtained displayed similar polymorphisms. In addition, isolates of P. ×stagnum have been continually recovered from the same irrigation reservoirs since 2005. These observations suggest that this new hybrid species is relatively stable in the laboratory and in nature, and may have adapted to the irrigation systems of the surveyed nurseries in eastern Virginia.

Similar to the other four Phytophthora hybrids in subclade 6b [38], the pathogenicity of P. ×stagnum is yet to be determined. No diseased plant samples associated with this novel hybrid species has been received in the Disease Clinic at Hampton Roads Agricultural Research and Extension Center in Virginia Beach, Virginia. Also, in a preliminary pathogenicity test, P. ×stagnum caused little if any dieback on rhododendron plants (data not shown). The low aggressiveness of P. ×stagnum may be inherited from its parent species. Phytophthora taxon PgChlamydo is considered as an opportunistic plant pathogen [18], although it has been found to cause leaf spot on nursery stocks in California [55]. The maternal parent species of P. ×stagnum is close to P. mississippiae, which has an unknown host range [14].

Origin of this novel hybrid is not known at this time. Although P. taxon PgChlamydo has been frequently recovered from the same irrigation reservoirs, the maternal parent of P. ×stagnum has never been isolated. This observation may suggest that the new hybrid had been introduced to these nurseries via incoming ornamental plant materials. Crop health risk posed by this new hybrid species has yet to be assessed.

Supporting Information

Acknowledgments

We thank Drs. Anton Baudoin and Erik Stromberg for providing experimental equipment and technical advice on the morphological examinations performed in their labs. We acknowledge Drs. Ping Kong and Giovanni Cafa for their advice on cloning the ITS regions in this study. We thank Dr. Brett Tyler for his advice on sequence analyses and critical review and Dr. Warren Copes for critical reading of this manuscript.