SEARCH

SEARCH BY CITATION

Keywords:

  • hybrids;
  • maize;
  • Phragmites australis;
  • Pythium arrhenomanes;
  • Pythium phragmitis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A comparison of oomycete diversity in reed stands (Phragmites australis) of Lake Constance, Germany, and maize fields close by provided evidence for the occurrence of natural hybridization between Pythium phragmitis, a newly described reed pathogen, and an as-yet unknown Pythium species closely related to P. phragmitis and P. arrhenomanes. Internal transcribed spacer and β-tubulin sequences of a large set of Pythium isolates from reeds showed dimorphic signals at several positions, indicative of a mixture of characters of two parent species. Involvement of P. phragmitis in the putative hybrid species was confirmed after cloning and sequencing of ITS regions and β-tubulin genes of the hybrid isolates. Mitochondrially inherited coxII gene sequences did not show dimorphic sites and suggested that the hybridization event was relatively ancient, or that other species might be involved. Intermediate habitat preferences, morphological characters and aggressiveness towards reeds and other grasses confirmed the suggestion that these isolates comprise a natural hybrid between two Pythium species. Pythium arrhenomanes, likely to be involved in the putative hybrid's evolution, was repeatedly isolated from maize fields adjacent to P. phragmitis-infested reed stands. The interface between natural habitats with established oomycete communities and agricultural fields with potentially introduced pathogens might constitute an evolutionary hot-spot giving rise to new species with as-yet unknown host ranges. As indicated by inoculation tests, the hybrid was significantly more pathogenic towards reed rhizomes than P. phragmitis, which caused no damage to these organs. This is apparently the first report of the occurrence of natural hybridization in Pythium.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The oomycete genus Pythium consists of well over 120 species, some of which are important plant pathogens with worldwide distribution, causing root, collar or fruit rot, and damping-off of seedlings. Pythium spp. such as P. ultimum or P. aphanidermatum are notorious pathogens, mostly of young, succulent plant tissues, such as fresh fine roots or seedlings. Species such as P. arrhenomanes are serious pathogens of graminaceous crops (Hendrix & Campbell, 1973; van der Plaats-Niterink, 1981; Dick, 1990; Lévesque & De Cock, 2004). Recently, an increase in interest in Pythium communities of natural or semi-natural ecosystems has occurred, and several new species from non-agricultural sites have been described. Pythium phragmitis, a newly described species from littoral sites, is considered a pathogen of common reed (Phragmites australis), preferentially attacking submerged leaves and likely to be significantly involved in reed decline phenomena of European freshwater lakes (Nechwatal et al., 2005).

In fungal or oomycete plant pathogens, natural hybridization events are considered an important means of rapid species evolution, potentially resulting in the emergence of new pathogens with host ranges and/or pathogenicity levels differing from those of the hybrids’ parent species (Érsek et al., 1995; Brasier et al., 1999; Brasier, 2000, 2001; Newcombe et al., 2000; Konrad et al., 2002; May et al., 2003; Schardl & Craven, 2003; Man in't Veld et al., 2007). International plant trade accompanied by the introduction of foreign pathogens is likely to favour the occurrence of hybridization events between resident and introduced species (Desprez-Loustau et al., 2007). In the oomycete genus Phytophthora, closely related to Pythium, occurrence of natural or in vitro-generated hybridization is well documented (e.g. Brasier et al., 1999; English et al., 1999; Bonants et al., 2000; May et al., 2003; Man in't Veld et al., 2007). Hybridization may occur between heterothallic or even homothallic (inbreeding) species, as a result of sporadic outcrossing events (May et al., 2003). The best-studied example of natural hybridization in Phytophthora is the destructive alder pathogen P. alni and its subspecies, a newly described swarm of hybrid taxa, originally considered a hybrid between P. cambivora and a species close to P. fragariae (Brasier et al., 1999, 2004). Recently, an alternative hypothesis on its origin and evolution was proposed (Ioos et al., 2006; Bakonyi et al., 2007).

Currently, no studies are available dealing with the occurrence of natural hybridization within the genus Pythium, although it is highly likely that, as a consequence of global plant trade and agricultural activity, members of the genus are being distributed all over the world and species introduced into new ecosystems. During a study on Pythium communities in reed stands of Lake Constance, Germany, several isolates of a homothallic Pythium species resembling the reed pathogen P. phragmitis were obtained from rhizosphere soil samples and diseased reed leaves (Nechwatal et al., 2008). Subsequently, isolates of the same species were also found in reed stands of other freshwater lakes in Germany and Europe. Analysis of their rDNA internal transcribed spacer (ITS) regions suggested an intermediate position of the isolates between P. phragmitis and the closely related species P. arrhenomanes. The occurrence of several dimorphic positions (double bases) in the electropherograms, mostly being representative of both of these two species, was considered strongly indicative of hybridization events between two closely related Pythium spp.

The aim of this study was to confirm a possible hybrid status of these P. phragmitis-like isolates. Data on morphology, ecology, pathogenicity, and analyses of nuclear and mitochondrial DNA sequences of the proposed hybrid taxon, in comparison with those of the potential parent species, are presented. This is apparently the first report of natural hybridisation within the genus Pythium.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Oomycete isolates

An oomycete survey in littoral reed stands of Lake Constance and other German and European freshwater lakes, conducted between 2003 and 2007, revealed more than 20 isolates of a putative hybrid Pythium species. These isolates, as well as isolates of P. phragmitis, one of the proposed parent species, were isolated from reed rhizosphere soil samples (permanently or only occasionally flooded) by baiting techniques, or from inundated diseased reed leaves by direct plating, as described earlier (Nechwatal et al., 2005). Isolates of P. arrhenomanes, also likely to be involved in the evolution of the hybrid, were obtained by baiting, mostly from soils of several maize fields (Zea mays) situated close to Lake Constance littoral reed sites, or from culture collections. The identification of oomycete isolates was based on morphological features and ITS sequence data (see below). Isolation sites and data of the isolates used in this study are given in Table 1.

Table 1. Pythium species and isolates used in this study
SpeciesIsolate IDaLocation, country, year of isolationIsolated fromReed habitatGene regions analysedTests performedc
  • a

    All isolates from Universität Konstanz, Phytopathology culture collection, except: CBS (Centraalbureau voor Schimmelcultures, Utrecht, Netherlands), OPU (Osaka Prefecture University, Osaka, Japan). P. phragmitis P13 = ex-type culture, CBS 117104.

  • b

    Sequences submitted to GenBank (P13 ITS: AY594259; Tub: EU152854; cox: AJ890351; P11 ITS: AY594258; Tub: EU152855; cox: EU145740; P54 ITS: AY743661; Tub: EU152856; cox: EU152857).

  • c

    g, growth rate; m, morphology; prc, pathogenicity to reed canary grass; prh, pathogenicity to reed rhizomes; prl, pathogenicity to reed leaves.

P. phragmitisP13Egg, D, 2003Phragmites australis, soilLake Constance, floodedITSb, Tubb, coxbg, m, prc, prl, prh
P40Egg, D, 2003P. australis, soilLake Constance, floodedITS, Tub, coxg, m, prc, prl, prh
P42Egg, D, 2003P. australis, soilLake Constance, floodedITS, Tub, coxprl
P52Egg, D, 2003P. australis, soilLake Constance, floodedITS, Tub, coxprl
P55Egg, D, 2003P. australis, soilLake Constance, floodedITS, Tub, coxprl
P63Egg, D, 2004P. australis, leafLake Constance, floodedITS, coxg, m
V11aReichenau, D, 2005P. australis, soilLake Constance, floodedITS, Tubg, m, prc, prh
V17aKonstanz, D, 2006P. australis, soilUniversity campus, floodedITS 
V21aMainau, D, 2006P. australis, soilLake Constance, floodedITS, Tubg, m
HybridP11Egg, D, 2003P. australis, soilLake Constance, dryITSb, Tubb, coxbg, m, prc, prl, prh
P53Egg, D, 2003P. australis, soilLake Constance, dryITS, coxg, m, prc, prl, prh
P56Egg, D, 2003P. australis, soilLake Constance, floodedITS, coxprl
P60Egg, D, 2004P. australis, leafLake Constance, floodedITS, Tub, coxg, m, prl
P67Egg, D, 2003P. australis, soilLake Constance, dryITS, Tubprl
P68Egg, D, 2003P. australis, soilLake Constance, dryITS, Tub 
P70Reichenau, D, 2005P. australis, soilLake Constance, floodedITS, Tub 
P72Reichenau, D, 2005P. australis, soilLake Constance, floodedITS, Tub 
V1aLitzelstetten, D, 2005P. australis, soilLake Constance, floodedITS, Tub 
V2aFließhorn, D, 2005P. australis, soilLake Constance, floodedITS, Tub, cox 
V3aDingelsdorf, D, 2005P. australis, soilLake Constance, floodedITS, Tub 
V4bKlausenhorn, D, 2005P. australis, soilLake Constance, floodedITS, Tub 
V7aErmatingen, CH, 2005P. australis, soilLake Constance, dryITS, Tub 
V10bAllensbach, D, 2005P. australis, soilLake Constance, floodedITS 
V16aWaschsee, D, 2005P. australis, soilLake Waschsee, dryITS, Tub, coxg, m, prc, prh
V18aMarkelfingen, D, 2005P. australis, soilLake Constance, dryITS, Tub 
V19aAugsburg, D, 2005P. australis, soilLake Kuhsee, floodedITS, Tub 
V22aEriskirch, D, 2005P. australis, soilLake Constance, floodedITS, Tubg, m
M1Magadino, CH, 2007P. australis, soilLago Maggiore, dryITS 
M2Magadino, CH, 2007P. australis, soilLago Maggiore, dryITS 
P. arrhenomanesP54Egg, D, 2003P. australis, soilLake Constance, dryITSb, Tubb, coxbg, m, prc, prl, prh
CBS324·62NL, 1962Zea maysITS, coxm, prl
CBS430·86NL, 1986Z. maysITSprl
OPU480Japan, 2001Oryza sp.ITSg, m, prl
ME1bHegne, D, 2006Z. mays, soil(close to reed stand)ITS, Tubg, m, prc, prl, prh
ME8aWahlwies, D, 2006Z. mays, soil(close to reed stand)ITS, Tubg, m, prc, prh
ME9dWahlwies, D, 2006Z. mays, soil(close to reed stand)ITS, Tubg, m

Morphology

Single hyphal tip cultures from water agar were prepared from all isolates used in the study. Isolates were kept on V8 agar (V8A; Ribeiro, 1978) plates at 6°C in the dark for long-term storage. For morphological studies, isolates were grown on fresh V8A plates for at least 6 days at 20°C, and 40 diagnostic structures (oospores, oogonia) were measured at ×400 magnification using a light microscope. Furthermore, the percentage of abortive oospores or unfertilized oogonia and the presence of sporangia produced aerially on agar cultures (i.e. without the addition of water) was recorded for each isolate. Zoospore production was also tested as described earlier, but did not yield countable numbers of released zoospores in all taxa (Nechwatal et al., 2005; data not shown). For the assessment of growth rates isolates were subcultured onto 135-mm V8A plates (50 mL), and incubated at 30°C after the onset of hyphal growth. Growth was recorded after 2 days along two straight axes drawn through the inoculation plug, and values averaged.

Sequence analysis

Aerial mycelium material scraped from fully colonized V8A plates was used for DNA extraction using the DNeasy Plant MiniKit (Qiagen). PCR amplification of complete ITS1, 5·8S and ITS2 rDNA regions was performed with primer pair ITS1 and ITS4 (White et al., 1990) according to described protocols. Amplification of a fragment of the β-tubulin gene was carried out using primers (TUBUF2, TUBUR1) and protocols described by Kroon et al. (2004). A subsequence of the mitochondrial coxII gene was amplified using the primers FM35 and FM58 as described by Martin (2000). Direct sequencing of the PCR products was carried out by MWG Biotech. Tubulin subsequences were trimmed to a length of 809 bp and cox II subsequences to a length of 563 bp to match the lengths of corresponding sequence entries of related species retrieved from GenBank. Representative sequences of the gene regions analysed were submitted to GenBank (for accession numbers see Table 1).

A PCR-RFLP assay to reveal polymorphic sequence positions in hybrid isolates was established for the β-tubulin gene. Amplification products generated as described above were digested using TaqI endonuclease (Fermentas), according to the supplier's protocols, and the resulting DNA banding patterns visualised in 3% agarose gels, after electrophoresis at 60 V for 2 h and ethidium bromide staining. Hybrid isolates were expected to show banding patterns consisting of the patterns of both P. phragmitis and P. arrhenomanes.

Purified amplification products of complete ITS regions and β-tubulin gene subsequences of putative hybrid isolates (ITS regions of P11 and P53 and β-tubulin subsequences of P11, V16a and V22a) were cloned using the Qiagen PCR Cloning Kit and Qiagen EZ Competent Cells. Ligation and transformation were carried out according to the manufacturer's instructions. For ITS, 24 (P11) and 22 (P53), and for β-tubulin 19 (P11, V16a and V22a) successfully transformed clones were randomly selected for sequence analysis. Amplification of plasmid DNA was performed using the primers used for the primary PCR reactions, and sequencing of selected clones (10 of each isolate) carried out by MWG Biotech. Additional clones were analysed using the above RFLP assay, with some clones expected to show P. phragmitis and some P. arrhenomanes patterns. All sequences were aligned and analysed using bioedit v. 7·0·1 (http://www.mbio.ncsu.edu/BioEdit/bioedit.html).

Pathogenicity

Reed leaves

Pathogenicity of the putative hybrid species towards reed leaves in comparison with the proposed parent species P. phragmitis and P. arrhenomanes was tested in a direct inoculation assay as described earlier (Nechwatal et al., 2005). In short, V8A plugs of five isolates of each species (see Table 1) were placed onto mature, greenhouse-grown reed leaves (seven per isolate) and incubated in the dark at 20°C in sealed Petri dishes on moist filter paper. Controls received plain agar discs. After 7 days leaf lesion length (longitudinal extension) was recorded. The test was repeated three times per isolate.

Reed rhizomes

Reed rhizome inoculation tests were performed six times using three isolates per Pythium species (see Table 1). Rhizome segments approx. 10–12 cm in length, 5 mm in diameter and comprising one to three nodes were harvested from mature, greenhouse-grown reed plants. Rhizomes were carefully washed and adhering rootlets and leaf sheaths removed. Only rhizome pieces without any visible damage or signs of disease were used for the test. For each isolate and replicate, five segments were placed in a glass Petri dish on moist filter paper and inoculated with a 5-mm-diameter V8A plug from the growing margin of the isolate tested. Agar plugs were fixed to the rhizomes with a sterilized needle (upside down), at the same time producing a wound to facilitate infection. Controls received plain agar plugs. Dishes were sealed with Parafilm and incubated in the dark for 8 days at 20°C. Symptoms developing on each segment were rated according to the scale: 0 = no necrosis or local discoloration (0–10% of rhizome affected), 1 = slight necrosis (11–25%) and 2 = heavy necrosis (26–100%).

Reed canary grass

Reed canary grass (Phalaris arundinacea), a species frequently occurring close to reed stands on similar sites, was selected for a pathogenicity test on grass species other than Phragmites australis. Three isolates of each oomycete species under investigation (see Table 1) were used for the test. Ten seeds of P. arundinacea (Jelitto Staudensamen) were placed in an approx. 80-mm circle in 90-mm Petri dishes on filter paper soaked with malt extract agar (1·5%), with three replicate dishes per isolate. Seeds were surface-sterilized before use (1 min 70% ethanol, 1 min 5% bleach, 1 min 70% ethanol). Infestation was acheived by placing a colonized agar disc (3 × 3 mm) of the respective isolate into the centre of the plate. Dishes were sealed with Parafilm and kept under natural light at 20°C. After 10 days the number of remaining live plantlets was recorded. The test was conducted four times, with six uninfested control dishes included in each test.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Oomycete isolates

While P. phragmitis was exclusively isolated from permanently flooded reed soils or inundated reed leaves, the putative hybrid isolates came from reed leaves and permanently as well as only occasionally flooded habitats (Table 1). Pythium arrhenomanes was consistently isolated from maize field soils in vicinity to reed sites, and a single isolate came from a drier reed site. It was never isolated from flooded reed soils or leaves.

Morphology

All isolates of P. phragmitis examined had low degrees of aborted or unfertilized, undeveloped oogonia, with the mean oospore diameter within the range reported earlier (Nechwatal et al., 2005) (Table 2). None of the isolates produced aerial sporangia on agar media. In contrast, P. arrhenomanes developed numerous aerial sporangia and had high rates of aborted oogonia. Some isolates completely failed to produce oogonia. In the hybrid isolates, rates of aborted or unfertilized oogonia were similarly high; some aerial sporangia were observed in some of the isolates examined. No significant differences were detected in oogonium/oospore diameter between the three species (Table 2). However, diameters of oogonia/oospores in P. arrhenomanes and the hybrid could often not be properly determined because of the distorted shapes of the abortive or unfertilized structures.

Table 2.  Morphological and growth features of selected Pythium isolates
SpeciesIsolateGrowth rate (mm day−1)aMean oospore diameter (µm)bOogonial abortion (%)Aerial sporangiac
  • a

    Means of two replicate growth tests at 30°C. Species means followed by the same letter are not significantly different (P < 0·05, anova, Tukey's multiple comparison test).

  • b

    Isolate means of 40 structures per isolate. Species means followed by the same letter are not significantly different (P < 0·05, anova, Tukey's multiple comparison test).

  • c

    Abundance of aerial sporangia: −, none; +, sporadic; ++, abundant.

P. arrhenomanesP5428·824·740++
OPU48021·429·282++
ME1b24·927·880++
ME8a27·9+
ME9d29·024·163+
Mean26·5a26·4a66 
HybridP1125·528·976+
P5325·129·578+
P6020·826·853
V16a25·326·868+
V22a25·429·1100
Mean24·4ab28·2a75 
P. phragmitisP1323·826·218
P4022·627·120
P6322·327·9 8
V11a22·425·5 3
V21a20·426·6 5
Mean22·3b26·7a11 

When considering isolate means, the growth rate of the hybrid was intermediate between those of P. arrhenomanes and P. phragmitis. While the latter two were significantly different from each other, the hybrid did not differ significantly from either of them (Table 2).

Molecular studies

blast searches with ITS sequences of the proposed hybrid isolates revealed that they were most closely related to P. phragmitis and P. arrhenomanes, with an identical length of 796 bp. Of a total of 15 variable positions, nine were identical to P. phragmitis in the hybrid isolates. The remaining six were ambiguous and proved to show distinct dimorphic peaks (double bases) after close examination of the electropherograms, representing sequence information of both P. arrhenomanes and P. phragmitis in four of the six cases. Two more ambiguous positions had both the parents’ information and bases unique to the hybrid. Cloning and sequence analysis of 46 ITS inserts confirmed the presence of six dimorphic positions in variable constellations within the hybrid isolates (Table 3). Both isolates analysed contained different ITS versions (types), five (A–E) in the one with strong dimorphic peaks (P11), three (A, B and E) in the other without (P53). In the prevalent combination (type A, 70% of all clones), of the 15 variable positions 10 were identical to P. phragmitis, three to P. arrhenomanes and two were unique to the hybrid. Fifteen percent of all clones analysed had the same base composition as P. phragmitis (type E), while none was identical to P. arrhenomanes.

Table 3.  Comparison of variable nucleotide positions of rDNA ITS sequences of Pythium arrhenomanes, P. phragmitis and a putative hybrid. The percentage of each sequence type (A–E) in the hybrid, as revealed by sequencing of single clones, out of a total of 46 clones, is given
 Positiona
493494501502526617669689699700730764768775776
  • a

    rDNA boundaries: 1–187, internal transcribed spacer 1; 188–346, 5·8S ribosomal RNA; 347–796, internal transcribed spacer 2.

P. arrhenomanesTCGGCGAGTGAGGTA
HybridYTARTCGTYRRKTAT
P. phragmitisCTAATCGTCGATTAT
Clone type A (70%)TTAGTCGTTAGTTAT
Clone type B (11%)TTAGTCGTTAGGTAT
Clone type C (2%)CTAATCGTTAGGTAT
Clone type D (2%)CTAATCGTTAGTTAT
Clone type E (15%)CTAATCGTCGATTAT

Subsequences (809 bp) of the β-tubulin gene of the putative hybrids were most similar to P. phragmitis and P. arrhenomanes. Of a total of 23 variable positions, 17 were like P. phragmitis and six proved dimorphic in most of the hybrid isolates after thorough examination of the electropherograms, showing both P. phragmitis and P. arrhenomanes signatures in three cases. Three more ambiguous positions had both the parents’ and unique bases (Table 4). The presence of double bases in β-tubulin fragments at particular positions was also confirmed in a PCR-RFLP assay using TaqI as a discriminatory enzyme cutting at 5′-TCGA-3′ (P. arrhenomanes: TTGA, P. phragmitis: TCGA, hybrid: TYGA at position 242–245 of the fragment analysed). The hybrid showed banding patterns consisting of both patterns of the proposed parent species (data not shown). Cloning of hybrid β-tubulin fragments and sequence analysis (via sequencing or RFLP assay) of 57 inserts revealed a c. 1:1 ratio of sequences identical to P. phragmitis (clone type B) and sequences showing three P. arrhenomanes signatures and three unique positions in the six dimorphic positions observed in the original sequence electropherograms (clone type A, Table 4).

Table 4.  Comparison of variable nucleotide positions of β-tubulin subsequences of Pythium arrhenomanes, P. phragmitis and a putative hybrid. The percentage of each sequence type in the hybrid, as revealed by sequencing or RFLP analysis of single clones, out of a total of 57 clones, is given
 Position
3651117135201243255303306315345369372375414459492564612663738744765
P. arrhenomanesTGCTCTCCACCTGCTGTTGTTCG
HybridCRGGTYYTCYTYAGGCCCAGCTR
P. phragmitisCAGGTCTTCCTTAGGCCCAGCTG
Clone type A (c. 50%)CGGGTTCTCTTCAGGCCCAGCTA
Clone type B (c. 50%)CAGGTCTTCCTTAGGCCCAGCTG

Subsequences of the coxII gene (563 bp) analysed in this study were again most closely related to P. arrhenomanes and P. phragmitis. While all P. phragmitis isolates were identical, a higher degree of sequence heterogeneity was observed among GenBank entries for P. arrhenomanes. When compared to the majority of entries, the putative hybrids were 0·5% different from the former and 2·8% different from the latter (3 or 16 bp, respectively). Hybrid isolates shared 14 positions with P. phragmitis, one with P. arrhenomanes and had two unique positions (Table 5). No overlapping peaks were observed in any of the isolates’coxII sequences.

Table 5.  Comparison of variable nucleotide positions of cytochrome oxidase II subsequences of Pythium arrhenomanes, P. phragmitis and a putative hybrid
 Position
152443117126171192198237300309351381396402432435
P. arrhenomanesTTCTGTTACCGAG/CATAA
HybridACTTTACTTTAGCTCTG
P. phragmitisACTAGACTTTAGCTCTA

Pathogenicity

Reed leaves

All isolates of P. phragmitis produced extensive necroses after inoculation onto mature reed leaves, while lesions caused by P. arrhenomanes were significantly smaller. Damage caused by the hybrid isolates was significantly different and intermediate between P. phragmitis and P. arrhenomanes. No necrosis developed on any of the control leaves (Table 6, Fig. 1).

Table 6.  Pathogenicity of Pythium phragmitis, P. arrhenomanes or the putative hybrid to reed (Phragmites australis) leaves, reed rhizomes and reed canary grass (Phalaris arundinacea) seedlings
 Reed leaf testReed rhizome testReed canary grass test
  • Mean and standard error of mean (SEM) are given for:

  • a

    three replicate tests, each consisting of five isolates per species, and seven leaves per isolate. Means followed by the same letter are not significantly different (P < 0·05, anova, Tukey's multiple comparison test).

  • b

    six replicate tests, each consisting of three isolates per species, and five rhizome pieces per isolate. Means followed by the same letter are not significantly different (P < 0·05, Kruskal-Wallis test, Dunn's multiple comparison test).

  • c

    four replicate tests, each consisting of three isolates per species, three replicate dishes per isolate, six replicates in the control, and 10 seeds per dish. Means followed by the same letter are not significantly different (P < 0·05, anova, Tukey's multiple comparison test).

 Lesion length (cm)aDisease index bNo. of surviving seedlingsc
Control0·0a0·53a (0·13)4·38a (0·47)
P. arrhenomanesa0·6b (0·10)1·22b (0·10)0·44b (1·14)
Hybridb1·6c (0·13)1·26b (0·06)1·78bc (0·25)
P. phragmitisc2·1d (0·08)0·81a (0·06)2·14ac (0·23)
image

Figure 1. Typical damage caused to reed (Phragmites australis) leaves after direct inoculation with (a) Pythium arrhenomanes (isolate P54), (b) the putative hybrid (isolate P67) and (c) P. phragmitis (isolate P13) after 7 days. Bar = 10 mm.

Download figure to PowerPoint

Reed rhizomes

Reed rhizomes infected with Pythium spp. developed necroses and discoloration around the point of inoculation, while control rhizomes remained healthy, except for some sporadic and localized damage by diverse moulds present in a non-sterile system. As indicated by a three-stage disease index, P. arrhenomanes and the hybrid isolates caused severe damage to the rhizomes. Necrosis for P. arrhenomanes and the hybrid isolates was significantly greater than in the controls or P. phragmitis (Table 6). Usually, the former two species completely colonized the rhizomes, leading to almost complete tissue disintegration, irrespective of the isolate used, while P. phragmitis caused only small localized discoloration, as did the control treatments (Fig. 2).

image

Figure 2. Damage caused to reed (Phragmites australis) rhizomes after direct inoculation with (a) the putative hybrid (isolate P11), (b) P. phragmitis (isolate P13) and (c) plain agar (control treatment) after 8 days. Arrows indicate sites of inoculation. Bar = 10 mm.

Download figure to PowerPoint

Reed canary grass

The rate of surviving P. arundinacea seedlings was approx. 40% in the uninfested controls. It was lowest in P. arrhenomanes-infested trials (below 5%), and ranged around 20% after P. phragmitis infection. The number of surviving seedlings in the dishes infested with the hybrid isolates was intermediate between the former two (Table 6). In contrast to seedling survival rates in the hybrids, those in the P. phragmitis-infested trials were not significantly different from the controls.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The results presented in this study provide strong evidence for the occurrence of natural hybridization between the reed pathogen P. phragmitis and another closely related, but as-yet unknown Pythium species. While in the past, several studies dealt with natural hybrids in the closely related oomycete genus Phytophthora (e.g. Brasier et al., 1999; Man in't Veld et al., 2007), this is apparently the first report of hybridization in Pythium.

Some of the morphological data collected would fit well to the hypothesis of both P. phragmitis and P. arrhenomanes being involved as direct progenitors of the hybrid. In particular, the production of aerial sporangia and growth rate in the hybrid isolates were intermediate between these two species. Similarly, aggressiveness of the hybrid towards reed leaves and P. arundinacea seedlings was intermediate. Confirming its relative specificity to P. australis leaves, P. phragmitis was most aggressive in the reed leaf assay, while P. arrhenomanes, a well-known pathogen of several grass species, was most aggressive to P. arundinacea. Similar results on intermediate traits in morphology or pathogenicity were reported for other hybrid fungi in the past (e.g. Newcombe et al., 2000; Brasier, 2001), or the artificially created oomycete hybrids Phytophthora sojae × vignae and P. capsici × nicotianae (English et al., 1999; May et al., 2003).

In contrast, molecular data were not consistent with a direct involvement of P. arrhenomanes sensu strictu in the hybridization process. In most of the putative hybrid isolates, the atypical feature of dimorphic double peaks in the sequence electropherograms of the ITS and the β-tubulin gene was observed, as also reported e.g. in the Phytophthora hybrids, P. alni (Brasier et al., 1999) and P. cactorum × hedraiandra (Man in't Veld et al., 2007). The sequence data of these two gene regions indicated that P. phragmitis is most likely to be directly involved in the evolution of the proposed hybrid, as high ratios of sequences identical to this species were detected after cloning and sequencing of the hybrids’ PCR products. In contrast, sequences identical to P. arrhenomanes could not be detected. Nevertheless, the occurrence of different sequence types with single positions containing information of either P. arrhenomanes or P. phragmitis in the hybrids’ ITS and β-tubulin sequences is indicative of the isolates being alloploid species hybrids (Brasier et al., 1999; Man in't Veld et al., 2007), involving P. phragmitis and an as-yet unknown species or ancestor, closely related to both P. phragmitis and P. arrhenomanes.

Although polymorphisms in the rDNA of single Pythium isolates have been reported (e.g. Martin, 1990, 1995), it is generally accepted that by means of concerted evolution, multi-copy ribosomal ITS sequences will usually be homogenous in well-defined Pythium species. Internal transcribed spacer polymorphisms in the hybrid isolates occurring in several different combinations among the clones (Table 3), and base ratios differing between the different isolates analysed, as observed in this study, were also found in the hybrid P. alni, and were considered indicative of homogenization being in progress between the ITS regions of the parent species involved (Brasier et al., 1999).

In the β-tubulin gene, the presence of both types of sequence information at certain positions in the hybrid isolates was confirmed by cloning, sequencing and a restriction enzyme assay, revealing the presence of banding patterns of both sequence types. One type was again shown to be identical to P. phragmitis. The β-tubulin gene is probably a single-copy gene in oomycetes [e.g. Phytophthora cinnamomi (Weerakoon et al., 1998) or Pythium ultimum (Mu et al., 1999)], so that the occurrence of double peaks in the sequences’ electropherograms can be considered as evidence for an alloploid hybrid organism, containing at least two different versions of the gene on the alleles present. Assuming that a putative Pythium hybrid has evolved after spontaneous outcrossing between two basically inbreeding species, it would normally be vegetatively diploid. Indeed, the presence of two versions of the gene in this study was confirmed by the 1:1 sequence ratio of clones in three hybrid isolates analysed. However, up to now no information was available on chromosome numbers and ploidy status of the particular Pythium species involved. As expected for an alloploid species hybrid, the present morphological studies proved that the hybrid isolates possessed significant numbers of unfertilized or abortive oogonia, indicative of problems in chromosome pairing during meiosis (Brasier et al., 1999). In this context, it is interesting to note that P. arrhenomanes is also known to exhibit high degrees of oogonial abortion (e.g. Dick, 1990) and might also be of hybridogeneous origin. However, in none of the P. arrhenomanes isolates studied here were dimorphic base positions indicative of this observed in the gene regions analysed.

Mitochondrial DNA is generally considered to be maternally inherited (Martin, 1989). Cytochrome oxidase II genes of species hybrids would thus be expected to contain sequences of one of the putative parent species, as reported by Man in't Veld et al. (2007). However, although closer to P. phragmitis, neither complete P. phragmitis nor P. arrhenomanes sequences were found in the putative hybrid isolates studied here, and the hybrids had two unique (‘private’) positions in the subsequence analysed. This might indicate that the hybridization event was relatively ancient, and mutations became fixed through ongoing inbreeding and propagation via single viable oospores in the hybrid, i.e. the hybrid being in a process of speciation. Sequence data from coxII genes might also indicate involvement of another parent or parents than those suggested, resulting in the particular hybrid progeny studied here. This was also observed in the hybrid P. alni, where coxI sequences of neither of the parent species originally thought to be involved could be found (Ioos et al., 2006).

In summary, molecular data indicate that the unknown isolates investigated here are probably the result of a relatively ancient hybridization event between P. phragmitis and an as-yet undiscovered or extinct taxon closely related to, but still distinct from both P. phragmitis and P. arrhenomanes. While the involvement of P. phragmitis could clearly be confirmed, the data showed that P. arrhenomanes is not likely to be among the parental species of the proposed hybrid. It might in fact itself be the product of an ancient hybridization event. However, several independent hybridization and backcrossing events that still remain to be unravelled might obscure the full range of processes and parental taxa involved and thus would require further investigation. A similar situation was proposed for the evolution of the alder pathogen P. alni and its subspecies that has long been presumed to be a hybrid between P. cambivora and an unknown taxon close to P. fragariae (Brasier et al., 1999). Currently the subspecies are considered to be interrelated via several hybridization events between each other (Ioos et al., 2006). Cytological data on ploidy status of the species involved, as well as forced in vitro crosses of P. phragmitis and P. arrhenomanes, might help to provide additional information on their roles in the generation of the proposed Pythium hybrid.

May et al. (2003) stated that a reduction in aggressiveness towards the parent species’ major hosts, as observed in the present study, is indicative of reduced fitness and competitiveness of hybridogeneous offspring in nature. This suggests that newly emerged hybrid species often will not be able to compete with their parent species in their respective niche (Brasier, 2001). Considering the results of the reed rhizome assay in the present study (Table 6), the putative Pythium hybrid seems to be an aggressive pathogen of reed rhizomes, a niche obviously not occupied by P. phragmitis. This might represent a significant selective advantage for the hybrid over its proposed reed-associated parent. The greater habitat plasticity of the hybrid – unlike both proposed parents consistently isolated from a range of both dry and permanently flooded littoral sites – is possibly connected to its ability to quickly and successfully colonize reed rhizomes, organs known to span large above- and below-ground areas at the land–water interface of littoral ecosystems. Indeed, the hybrid currently seems to be a highly successful and competitive evolutionary unit. During an oomycete survey conducted 2005–07 at Lake Constance and several other south German lakes, hybrid isolates were found at 14 of 26 sites studied, while P. phragmitis was found at only 5. Additional hybrid isolates from southern Switzerland (Table 1) show that it is not confined to limited geographical areas, and is possibly thriving throughout Europe. Surprisingly, P. arrhenomanes was also highly aggressive towards reed rhizomes. It might therefore be concluded that these plant organs constitute the sites of ancient or current hybridization events after infection by the parent species involved. The wide distribution of the hybrid across Europe, even in water catchment areas not connected to each other, implies that similar hybridization events might have occurred independently and more than once. Consistent differences between hybrid isolates of different geographic origins in the ras-related Ypt1-gene sequences (data not shown) also provide evidence that the hybrid isolates do not represent a single clonal lineage transmitted by human activity, but several independent hybridization events which have occurred or are still repeatedly occurring in the field.

Reed stands and agricultural land are in close vicinity in some situations at Lake Constance, with maize fields extending immediately to littoral stands covered with reeds. Although no data are available on the history of occurrence of P. arrhenomanes or other grass-associated species in European agriculture, the second species likely to be involved in the hybrid might have been introduced to and spread across Europe, or its spread might have been promoted, along with the advent of maize farming in the 1950s, and have been brought in contact with P. phragmitis populations through cultivation of this and other graminaceous crops close to littoral reed sites. Assuming that the reed-associated P. phragmitis has been coevolving with its host since ancient times, leading to a more or less balanced interplay of reed dieback and progression in undisturbed stands, the emergence of a new hybrid with high aggressiveness to rhizomes might have posed an additional threat to reed stands in relatively recent times.

Given the high pathogenic potential of the proposed hybrid towards reed rhizomes, in combination with its putative parent P. phragmitis, more adapted to leaf infection (Nechwatal et al., 2005), both species are likely to cause major damage to both above- and below-ground parts of reeds. Rhizome infection by a relatively newly emerged hybrid is probably more detrimental for the plant since it is independent from leaf inundation, and the interaction lacks a history of coevolution of host and pathogen. Furthermore, considering the hybrid's relatively wide distribution in Europe, it might play a more significant role in current reed decline events than previously recognized.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This study was funded by the Deutsche Forschungsgemeinschaft (DFG) as part of the programme SFB 454 (‘Littoral Zone of Lake Constance’). We thank Motoaki Tojo (Osaka Prefecture University, Japan) for providing isolate OPU480.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Bakonyi J, Nagy ZA, Érsek T, 2007. A novel hybrid with the nuclear background of Phytophthora alni subsp. alni exhibits a mitochondrial DNA profile characteristic of P. alni subsp. uniformis. Acta Phytopathologica et Entomologica Hungarica 42, 17.
  • Bonants PJM, Hagenaar-De Werdt M, Man in't Veld W, Baayen R, 2000. Molecular characterization of natural hybrids of Phytophthora nicotianae and P. cactorum. Phytopathology 90, 86774.
  • Brasier CM, 2000. The rise of the hybrid fungi. Nature 405, 1345.
  • Brasier CM, 2001. Rapid evolution of introduced plant pathogens via interspecific hybridization. BioScience 51, 12333.
  • Brasier CM, Cooke DEL, Duncan JM, 1999. Origin of a new Phytophthora pathogen through interspecific hybridization. Proceedings of the National Academy of Sciences, USA, 96, 587883.
  • Brasier CM, Kirk SA, Delcan J, Cooke DEL, Jung T, Man In't Veld WA, 2004. Phytophthora alni sp. nov. and its variants: designation of emerging heteroploid hybrid pathogens spreading on Alnus trees. Mycological Research 108, 117284.
  • Desprez-Loustau M-L, Robin C, Buée M et al ., 2007. The fungal dimension of biological invasions. Trends in Ecology and Evolution 22, 47280.
  • Dick MW, 1990. Keys to Pythium. Reading, UK: Published by the author.
  • English JT, Laday M, Bakonyi J, Schoelz JE, Érsek T, 1999. Phenotypic and molecular characterization of species hybrids derived from induced fusion of zoospores of Phytophthora capsici and Phytophthora nicotianae. Mycological Research 103, 10038.
  • Érsek T, English JT, Schoelz JE, 1995. Creation of species hybrids of Phytophthora with modified host ranges by zoospore fusion. Phytopathology 85, 13437.
  • Hendrix FF, Campbell WA, 1973. Pythiums as plant pathogens. Annual Review of Phytopathology 11, 7798.
  • Ioos R, Andrieux A, Marçais B, Frey P, 2006. Genetic characterisation of the natural hybrid species Phytophthora alni as inferred from nuclear and mitochondrial DNA analyses. Fungal Genetics and Biology 43, 51129.
  • Konrad H, Kirisits T, Riegler M, Halmschlager E, Stauffer C, 2002. Genetic evidence for natural hybridization between the Dutch elm disease pathogens Ophiostoma novo-ulmi ssp. novo-ulmi and O. novo-ulmi ssp. americana. Plant Pathology 51, 7884.
  • Kroon LPNM, Bakker FT, Van Den Bosch GBM, Bonants PJM, Flier WG, 2004. Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genetics and Biology 41, 76682.
  • Lévesque CA, De Cock AWAM, 2004. Molecular phylogeny and taxonomy of the genus Pythium. Mycological Research 108, 136383.
  • Man in't Veld WA, De Cock AWAM, Summerbell RC, 2007. Natural hybrids of resident and introduced Phytophthora species proliferating on multiple hosts. European Journal of Plant Pathology 117, 2533.
  • Martin F, 1995. Meiotic instability of Pythium sylvaticum as demonstrated by inheritance of nuclear markers and karyotype analysis. Genetics 139, 123346.
  • Martin FN, 1989. Maternal inheritance of mitochondrial DNA in sexual crosses of Pythium sylvaticum. Current Genetics 16, 3734.
  • Martin FN, 1990. Variation in the ribosomal DNA repeat unit within single-oospore isolates of the genus Pythium. Genome 33, 58591.
  • Martin FN, 2000. Phylogenetic relationships among some Pythium species inferred from sequence analysis of mitochondrially encoded cytochrome oxidase II gene. Mycologia 92, 71127.
  • May KJ, Drenth A, Irwin JAG, 2003. Interspecific hybrids between the homothallic Phytophthora sojae and Phytophthora vignae. Australasian Plant Pathology 32, 3539.
  • Mu JH, Bollon AP, Sidhu RS, 1999. Analysis of beta-tubulin cDNAs from taxol-resistant Pestalotiopsis microspora and taxol-sensitive Pythium ultimum and comparison of the taxol-binding properties of their products. Molecular and General Genetics 262, 85768.
  • Nechwatal J, Wielgoss A, Mendgen K, 2005. Pythium phragmitis sp. nov., a new species close to P. arrhenomanes as a pathogen of common reed (Phragmites australis). Mycological Research 109, 133746.
  • Nechwatal J, Wielgoss A, Mendgen K, 2008. Diversity, host, and habitat specificity of oomycete communities in declining reed stands (Phragmites australis) of a large freshwater lake. Mycological Research 112, 68996.
  • Newcombe G, Stirling B, McDonald S, Bradshaw Jr HD, 2000. Melampsora × columbiana, a natural hybrid of M. medusae and M. occidentalis. Mycological Research 104, 26174.
  • Van Der Plaats-Niterink AJ, 1981. Monograph of the genus Pythium. Studies in Mycology 21, 1242.
  • Ribeiro OK, 1978. A Source Book of the Genus Phytophthora. Vaduz, Liechtenstein: J. Cramer.
  • Schardl CL, Craven KD, 2003. Interspecific hybridisation in plant-associated fungi and oomycetes: a review. Molecular Ecology 12, 286173.
  • Weerakoon ND, Roberts JK, Lehnen Jr LP, Wilkinson JM, Marshall JS, Hardham AR, 1998. Isolation and characterization of the single β-tubulin gene in Phytophthora cinnamomi. Mycologia 90, 8595.
  • White TJ, Bruns T, Lee S, Taylor J, 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: InnisMA, GelfandDH, SninskyJJ, WhiteTJ, eds. PCR Protocols: A guide to Methods and Applications. San Diego, CA USA: Academic Press, 31522.