Genetic diversity, sensitivity to phenylamide fungicides and aggressiveness of Phytophthora ramorum on Camellia, Rhododendron and Viburnum plants in Spain

Authors

  • A. Pérez-Sierra,

    Corresponding author
    1. Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain
    Search for more papers by this author
  • L. A. Álvarez,

    1. Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain
    Search for more papers by this author
  • A. Vercauteren,

    1. Institute for Agricultural and Fisheries Research, Plant Sciences Unit – Crop Protection, Burgemeester Van Gansberghelaan 96 bus 2, 9820 Merelbeke, Belgium
    Search for more papers by this author
  • K. Heungens,

    1. Institute for Agricultural and Fisheries Research, Plant Sciences Unit – Crop Protection, Burgemeester Van Gansberghelaan 96 bus 2, 9820 Merelbeke, Belgium
    Search for more papers by this author
  • P. Abad-Campos

    1. Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain
    Search for more papers by this author

E-mail: aperesi@eaf.upv.es

Abstract

Phytophthora ramorum has been detected in official plant health surveys on Rhododendron, Viburnum and Camellia in ornamental nurseries in northern Spain since 2003. A collection of 94 isolates of P. ramorum was obtained from 2003 to 2008 from plants with symptoms at different geographical locations. Isolates were identified based on morphology and sequence of the rDNA ITS region. Mating type, genetic variation, sensitivity to phenylamide fungicides and aggressiveness of these isolates were determined. All isolates belonged to the A1 mating type, ruling out the possibility of genetic recombination. Seven microsatellite markers were used to study genetic diversity; three out of the seven microsatellite markers were polymorphic within the Spanish population of P. ramorum. This study confirms that all Spanish isolates of P. ramorum belonged to the EU1 lineage. Twelve intralineage genotypes were detected, five that are unique to Spain (EU1MG38, EU1MG41, EU1MG37, EU1MG39 and EU1MG40) and seven that are also present in at least one other European country (EU1MG1, EU1MG29, EU1MG22, EU1MG13, EU1MG2, EU1MG18 and EU1MG26). Genotypes EU1MG37, EU1MG39 and EU1MG40 were isolated from Rhododendron from one region; EU1MG38 and EU1MG41 were isolated from Camellia from two different regions. Isolates of genotype EU1MG38 were resistant to metalaxyl and mefenoxam. The level of genetic diversity within the Spanish population of P. ramorum is limited and indicates a relatively recent clonal expansion.

Introduction

Phytophthora ramorum was described in Europe as a new species associated with twig and foliar blight disease on Rhododendron and Viburnum plants in Germany and the Netherlands (Werres et al., 2001). In North America, P. ramorum is responsible for the mortality of Californian oak species (Quercus agrifolia and Q. kelloggii) and tanoak (Lithocarpus densiflorus). The disease is commonly known as sudden oak death (SOD) (Rizzo et al., 2002). Since the description of the species, the number of host species identified as susceptible has increased on both continents. In Spain, this pathogen was first reported in the Balearic Islands in 2002 on Rhododendron sp. (Moralejo & Werres, 2002) and subsequently it was found on the mainland on Camellia japonica (Pintos Varela et al., 2003) and Viburnum tinus (Pintos Varela et al., 2004).

Emergency measures were taken by European Union member states (directives 2002/757/EC, 2004/426/EC and 2007/201/EC) to avoid further introductions and to avoid the spread of P. ramorum. These measures included inspections of all nurseries where susceptible Rhododendron, Viburnum or Camellia plants were grown. In addition, several regulations were adopted to eliminate affected plants and to control the movement of plant material. The incidence of this pathogen on ornamental plants in Spain has been recorded annually and nurseries have been surveyed regularly since 2002.

Phytophthora ramorum is a heterothallic species that needs two mating types, A1 and A2, for sexual recombination. In situations where a heterothallic Phytophthora sp. is introduced in a new environment, usually only a single mating type can be identified within the initial population. Phytophthora ramorum mating type A1 is predominantly found in Europe and mating type A2 is found almost exclusively in North America. However, isolates belonging to the Al mating type were reported from commercial nurseries in Oregon (Hansen et al., 2003; Prospero et al., 2007) and Washington (Garbelotto et al., 2006). Since then, both mating types have been detected in these US states (Ivors et al., 2006) with the risk of possible recombination between Al and A2 (Garbelotto et al., 2006). In Europe three isolates of the A2 mating type were detected in Belgium (Werres & De Merlier, 2003; Vercauteren et al., 2011). There have been no studies to establish the mating type of P. ramorum detected in ornamental plants in nurseries in Spain to date.

Some studies have emphasized the genetic differences between EU and US isolates of P. ramorum (Ivors et al., 2004; Kroon et al., 2004) and supported the exotic nature of this pathogen on both continents. Three clonal lineages of P. ramorum have been described, with isolates of the EU1 lineage mainly present in Europe, and isolates of the NA1 and NA2 lineages only present in North America (Ivors et al., 2006; Martin, 2008; Goss et al., 2009; Grünwald et al., 2009). For the study of genetic diversity of P. ramorum, various techniques based on amplified fragment length polymorphisms (AFLP) and simple sequence repeats (SSR) (Ivors et al., 2004, 2006) have been used. However, these techniques showed low genetic variation within the population of P. ramorum. Prospero et al. (2004, 2007, 2009) and Mascheretti et al. (2009) used additional microsatellite markers to study the North American population of P. ramorum and also found low levels of variation. Recently Vercauteren et al. (2010), using new polymorphic SSR markers and previously described markers, were able to differentiate 30 genotypes within the Belgian population of P. ramorum.

Fungicide application is a major component in the management of Phytophthora diseases. However, curative fungicide applications or applications in the 10-m quarantine zone are not allowed under EU phytosanitary measures. Metalaxyl or one of its isomers, mefenoxam, are phenylamide fungicides (PAFs) used in nurseries to control Phytophthora. Many producers continue to rely on these compounds because of their effectiveness in the control of P. ramorum. However, the use of these fungicides can impose a selection for resistance on the population of Phytophthora (Grünwald & Flier, 2005; Grünwald et al., 2006; Vercauteren et al., 2010). Monitoring the pathogen population for resistance to a fungicide is essential for the development of management strategies that can delay or prevent development of resistance to fungicides and fungicide failure.

Globalization has led to an increase in trade and travel, and alien or immigrant species are now being exchanged at unprecedented rates between geographically isolated regions (Brasier, 2008; Moralejo et al., 2009). As a result of the ornamental plant trade among European countries or plant trade from USA/Canada, the probability of exchange of isolates of different lineages of this pathogen is high.

On the presumption that knowledge of the population genetics of fungal plant pathogens could eventually contribute to the development of more durable disease management strategies, the objective of this study was to generate information about the genetic variation and population structure of P. ramorum in Spain. Therefore, the aims of this paper were to determine for Spanish P. ramorum isolates: (i) the predominant mating type; (ii) the amount of genetic variation; (iii) the existence of metalaxyl/mefenoxam resistance and (iv) the degree of aggressiveness on Camellia, Rhododendron and Viburnum plants.

Materials and methods

Survey and isolation

From 2003 to 2008, Camellia, Viburnum and Rhododendron plants with symptoms were sampled from commercial nurseries to detect infections caused by P. ramorum. Surveys were carried out during late March to early June of each year, as this is considered to be the period with the most favourable weather conditions for development of this disease in Spain. Samples were collected from nine different geographical areas in Spain (numbered 1–9): 1–4 in the north, 5–7 central and 8 and 9 in the eastern area of the Iberian Peninsula.

The aerial symptoms displayed by affected plants were brown lesions on the leaves, foliar blight, shoot dieback and stem canker. Plants were kept in a cool box (10–15°C) to minimize tissue deterioration during the transportation of the samples from the survey site to the laboratory. Plant samples were washed under running tap water for at least an hour. Isolations were made from leaves, shoot or stem tissues on CMA-PARP selective medium (Jeffers & Martin, 1986). The plates were incubated at 20°C in the dark for 2–3 days. Hyphal tips from developing Phytophthora colonies were transferred to potato dextrose agar (PDA; Biokar-Diagnostics) or to V8 juice agar (2 g CaCO3, 200 mL V8 juice and 15 g agar in 800 mL distilled water) for identification. The isolates were identified as P. ramorum based on morphological features: colony morphology, growth rate, cardinal growth temperatures, and production, morphology and dimensions of sporangia. For long-term storage, cultures were preserved in the fungal culture collection maintained at the Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia (Spain): fifteen 5 mm-diameter agar plugs from the edge of a colony grown on PDA were placed into a 15 mL vial containing 10 mL sterile soil extract (100 g soil flooded with 1 L distilled water for 24 h at room temperature, 50 mL of this solution mixed with 950 mL of distilled water and then sterilized by autoclaving). The isolates were immediately stored at 14°C. One reference isolate of P. ramorum (BBA 9/95), kindly provided by S. Werres in 2003, was also included in this collection and used in subsequent studies.

Mating type

Heterothallism of P. ramorum isolates was tested by pairing them with known A1- and A2-type isolates (P. cryptogea CBS-308.62 and CBS-307.62) on carrot piece agar (CPA) as described by Werres et al. (2001). CPA was prepared using 50 g carrot pieces and 22 g agar L−1 distilled water (Werres et al., 2001; Werres & Kaminski, 2005). Disks (5 mm-diameter) were taken from the margin of growing colonies on V8 juice agar and placed 3 cm apart. Plates were incubated at 20°C in the dark for 2–3 weeks. The presence of oogonia indicated that the unknown isolate was the opposite mating type to the tester.

DNA extraction, ITS and microsatellite markers

Mycelium of each P. ramorum isolate grown on PDA medium for 2 weeks at 20°C was scraped and mechanically disrupted by grinding to a fine powder under liquid nitrogen using a mortar and pestle. Total DNA was extracted using the E.Z.N.A. Plant Miniprep kit (Omega Bio-Tek) following the manufacturer’s instructions. DNA was visualized under UV light on 0·7% agarose gels stained with ethidium bromide and was stored at −20°C.

The ITS region of the isolates (Table 1) was amplified and sequenced using the universal primers ITS-6 (5′-GAAGGTGAAGTCGTAACAAGG-3′) (Cooke et al., 2000) and ITS-4 (5′-TCCTCCGCTTATTGATATGC-3′) (White et al., 1990) in order to confirm their identification. The cycling program for the amplification of the ITS was an initial denaturing step of 94°C for 2 min; 35 cycles of 94°C for 30 s, annealing at 55°C for 30 s, an extension at 72°C for 45 s; and a final extension of 72°C for 10 min. The PCR products were visualized in 1·5% agarose gels (agarose D-1 Low EEO; Conda). A 100-bp DNA ladder was used as a molecular weight marker (Dominion MBL).

Table 1.   Multilocus genotypes and host/geographical origin of Phytophthora ramorum isolates obtained from ornamental plants in nurseries in Spain
Multilocus genotypeNo. of isolatesHost and geographical originaYear of isolation
RhododendronCamelliaViburnum
  1. aNumber of isolates per host and the area codes (in italics in brackets) where found.

  2. bIsolates resistant to both metalaxyl and mefenoxam.

EU1MG15333 (1,2,3,6,7,9)6 (2)14 (2,3,5,6,8,9)2003, 2004, 2005, 2006, 2007, 2008
EU1MG2202 (2)02007
EU1MG1355 (2,4)002004, 2005, 2007
EU1MG1822 (2,4)002006
EU1MG2283 (3)05 (2,3,8)2004, 2008
EU1MG2622 (2)002008
EU1MG291212 (2)002007
EU1MG3711 (2)002007
EU1MG38b404 (2)02007
EU1MG3911 (2)002007
EU1MG4011 (2)002007
EU1MG41303 (3)02008
Total94601519 

PCR products were purified using the High Pure PCR Product Purification kit (Roche Diagnostics) and sequenced in both directions by the DNA Sequencing Service of the Universidad Politécnica de Valencia-CSIC (Spain). Resultant sequences were compared with sequences deposited at the National Center for Biotechnology Information (NCBI) nucleic acid database GenBank.

Microsatellite loci 18, 64, 82a (Ivors et al., 2006), PrMS43 (Prospero et al., 2007), 82b, ILVOPrMS145a and ILVOPrMS145c (Vercauteren et al., 2010) were characterized to determine the microsatellite multilocus genotype of each P. ramorum isolate. The loci were amplified with universal fluorescent primers in a multiplex PCR reaction following the protocol of Vercauteren et al. (2010).

Metalaxyl and mefenoxam sensitivity

The reaction of all P. ramorum isolates to the phenylamide fungicides metalaxyl (Ridomil, 25% w/w, WP; Syngenta Agro) and mefenoxam (Ridomil Gold, 48%, w/v, SL; Syngenta Agro) was evaluated by transferring each isolate to PDA medium supplemented with 0, 0·1, 1, 5, 10 and 100 μg mL−1 active ingredient of each fungicide. Mycelial plugs (5-mm diameter) were cut from the margin of actively growing colonies of 7-day-old agar cultures. One disk was placed in the centre of a 9 cm Petri dish with mycelia in contact with the medium. Three replicates per fungicide concentration were tested for each isolate and each assay was performed twice. The growth of each isolate at each concentration was determined by measuring colony diameters in two perpendicular directions on each culture plate after 12 days of incubation in darkness at 20°C. Growth of an isolate on supplemented media was compared to growth on media lacking fungicide by subtracting the inoculation plug diameter (5 mm) from the diameter of each colony and dividing the average colony diameter on supplemented plates by the average colony diameter on control plates with no fungicide, with the result expressed as a percentage. An isolate was scored as sensitive (S) if colony growth on media with 5 μg mL−1 fungicide was <40% of the isolate growth on control plates. Intermediate isolates (I) exhibited growth on media with 5 μg mL−1 > 40% of that on control plates, but growth on media with 100 μg mL−1 fungicide <40% of that on control plates. Resistant isolates (R) exhibited growth on media with 100 μg mL−1 >40% of that on control plates (Parra & Ristaino, 2001; Taylor et al., 2002). Furthermore, the effective concentration (EC50) for each isolate was calculated by plotting the growth of each isolate (as the percentage of growth in the absence of fungicide) against log10 of the fungicide concentration. The point on the regression line at which growth of the isolate was inhibited 50% was the EC50 value. Analysis of variance (anova) was performed on the values of the EC50 to differentiate P. ramorum isolates according to their responses to concentrations of metalaxyl or mefenoxam. anova and the standard error of the EC50 values of each isolate were calculated using the software Statgraphics Plus 5·1 (Manugistics Inc.).

Aggressiveness of P. ramorum isolates

A cross-inoculation experiment was conducted to test the aggressiveness of isolates of P. ramorum on Camellia, Rhododendron and Viburnum. Eight isolates were studied: four fungicide-sensitive isolates (Pr-4, Pr-22, Pr-31 and Pr-67) and four fungicide-resistant isolates (Pr-47, Pr-61, Pr-64 and Pr-77). Fungicide-sensitive isolates were chosen randomly from each geographic origin and host. Aggressiveness of each isolate was assessed by inoculating detached leaves from these hosts using two different inoculation procedures.

The first inoculation method used a zoospore suspension. Ten to twelve 5 mm-diameter plugs were cut from the culture margin of each isolate grown on V8 juice agar and plugs were then placed into 10 mL sterile soil extract in a 9 cm-diameter Petri dish. Plates were incubated at 20°C for 3–4 days with a 12 h-light photoperiod. After abundant sporangia had formed, plates were incubated at 4°C for 30 min and then brought to room temperature for 30 min to induce zoospore release. The zoospore concentration was estimated using a haemocytometer and adjusted to 104 mL−1. A 40 μL drop of the suspension was placed on the midrib of each leaf; thus each drop produced a single lesion. Control leaves were inoculated as described above with a drop of sterile water. Inoculated leaves and controls were placed in transparent humid chambers and incubated at 20°C with a 12 h-light photoperiod for 6 days.

In the second inoculation procedure, the undersides of leaves of the different plant species were wound-inoculated with a needle (0·1 mm-diameter) to a depth of 0·2 mm. A colonized PDA plug (5 mm-diameter) was placed on the wound. Control leaves were wound-inoculated with sterile PDA plugs. Inoculated leaves were placed in humid chambers as described above and incubated for 4 days.

The experimental design consisted of three randomized blocks, with nine treatments (eight isolates plus control) and five detached leaves per treatment. Experiments with both inoculation procedures were repeated once.

The aggressiveness of the isolates was recorded on the basis of the lesion area developed on each treatment. The area of the lesion developed in each leaf was traced on transparent plastic sheets and subsequently digitized then quantified using Assess software (American Phytopathological Society).

Data from both inoculation procedures were analysed by one-and two-way anova for effects of trial, treatment, isolate and host. If trial effects were detected, subsequent analyses were also conducted for each trial separately. Pairwise comparisons for effect of treatment were conducted using Fisher’s least significant difference procedure (< 0·05). Analyses were performed using Statgraphics Plus 5·1 (Manugistics Inc.).

Results

Isolates and mating type

During the course of this study (2003–2008) a total of 94 P. ramorum isolates were obtained, predominantly from leaves, twigs and stem infections of Camellia (15 isolates), Rhododendron (60 isolates) and Viburnum (19 isolates) (Table 1).

All the isolates formed oogonia when paired with the P. cryptogea A2 tester isolate; none formed oogonia when paired with the A1 tester isolate. Therefore, all the isolates were classified as belonging to mating type A1.

ITS and microsatellite markers

The PCR amplification and DNA sequence of the ITS region confirmed the identification of all the isolates as P. ramorum when compared with the NCBI GenBank database.

Microsatellite analysis revealed that all isolates belonged to the EU1 lineage of P. ramorum. Four of the seven markers used (64, 82b, ILVOPrMS145a, ILVOPrMS145c) were polymorphic in the Spanish population of P. ramorum and 12 intralineage genotypes (MG groups) were identified (Table 1). Genotype EU1MG1 was dominant in the population (62%). In five of the regions studied only the EU1MG1 genotype was detected. Two of the regions had two genotypes (EU1MG1 and EU1MG22, or EU1MG13 and EU1MG18). One of the regions had three genotypes (EU1MG1, EU1MG22 and EU1MG41). One region (geographical area 2) had 11 of the 12 genotypes, including three of the new genotypes (EU1MG37, EU1MG39 and EU1MG40).

Metalaxyl and mefenoxam sensitivity

The addition of the fungicides, metalaxyl or mefenoxam, to the culture medium significantly reduced growth rates in most of the isolates tested. Of the 94 isolates of P. ramorum studied, the majority (95·7%) were classified as sensitive, because the percentage growth rates on media with 5 μg mL−1 metalaxyl relative to the control ranged from 0% to 9·5%. Only 4·3% of the isolates (four) were resistant to both metalaxyl and mefenoxam (Table 1); the growth on media with 100 μg mL−1 fungicide was >40% relative to the control. With the fungicide metalaxyl, two isolates were considered intermediate in sensitivity because the growth on media with 100 μg mL−1 metalaxyl was <40% compared to the control. All isolates with reduced sensitivity to the fungicide mefenoxam were categorized as intermediate in sensitivity (5·1%).

Analysis of variance showed significant differences in EC50 values with the fungicide metalaxyl (< 0·05) and mefenoxam (< 0·01) between P. ramorum isolates. For the fungicide metalaxyl, most of the isolates fell within the group with EC50 values <1·0 μg mL−1. Values for sensitive isolates ranged from 0·01 to 0·11 μg mL−1 (standard deviation (SE) 0·001–0·04 and 0·08–0·21; respectively). Isolates intermediate in sensitivity presented EC50 values that fluctuated from 3·6 to 4·2 μg mL−1 (SE 3·4–3·7 and 4·1–4·4 respectively), while the resistant isolates presented EC50 values from 7·7 to 9·5 μg mL−1. Mean EC50 values for sensitivity of P. ramorum isolates to the fungicide mefenoxam ranged from 0·01 to 0·06 μg mL−1 (SE 0·001–0·015 and 0·04–0·08). Isolates with intermediate sensitivity ranged from 2·9 to 4·4 μg mL−1 (SE 2·6–3·1 and 4·3–4·6).

Aggressiveness of P. ramorum isolates

Two-way anova indicated an effect of the procedure of inoculation (< 0·001), isolate (< 0·001), host (= 0·011) and trial (= 0·046) on lesion area. Interaction between isolate × host factors was not significant (> 0·05). Consequently, there was no evidence for pathogenic specialization among P. ramorum isolates for their respective hosts (> 0·05). Although an effect of trial was observed, differences between the two trials were minor and subsequent analyses on each trial separately produced results similar to those of the pooled data set. Therefore pooled trial results are presented (Table 2).

Table 2.   Aggressiveness of Phytophthora ramorum isolates with varying degrees of fungicide sensitivity to three different hosts, as determined by two inoculation methods
IsolateArea codeOriginal hostFungicide sensitivityaAggressivenessb
Zoospore inoculationcWound inoculationd
MetalaxylMefenoxamCamelliaRhododendronViburnumCamelliaRhododendronViburnum
Fungicide-sensitive isolates
 Pr-046ViburnumSS0·9 ± 0·19 b1·7 ± 0·15 b0·7 ± 0·11 b4·7 ± 0·45 b3·5 ± 0·53 b2·7 ± 0·61 b
 Pr-221RhododendronSS1·3 ± 0·13 b2·2 ± 0·29 b0·7 ± 0·17 b5·3 ± 0·38 b4·4 ± 0·45 b3·5 ± 0·40 b
 Pr-317ViburnumSS0·2 ± 0·09 a2·6 ± 0·11 b1·2 ± 0·14 b4·.4 ± 0·43 b3·0 ± 0·59 b2·7 ± 0·52 b
 Pr-672RhododendronSS0·8 ± 0·12 ab1·9 ± 0·19 b0·3 ± 0·08 a4·0 ± 0·53 b4·2 ± 0·48 b3·0 ± 0·38 b
 Mean0·82·10·74·63·83·0
Isolates with reduced fungicide sensitivity
 Pr-472CamelliaRI3·4 ± 0·13 c5·1 ± 0·35 c2·3 ± 0·20 c5·2 ± 0·55 b4·5 ± 0·61 b3·7 ± 0·43 b
 Pr-612CamelliaRI2·8 ± 0·18 c4·4 ± 0·16 c2·6 ± 0·10 c4·2 ± 0·44 b6·1 ± 0·43 bc3·1 ± 0·55 b
 Pr-772CamelliaRI3·0 ± 0·32 c3·8 ± 0·25 c2·0 ± 0·09 c4·8 ± 0·47 b5·9 ± 0·48 bc3·2 ± 0·43 b
 Pr-642CamelliaII3·6 ± 0·27 c4·9 ± 0·21 c2·5 ± 0·25 c5·6 ± 0·51 bc4·8 ± 0·59 b3·8 ± 0·51 b
 Control0·0 ± 0·00 a0·0 ± 0·00 a0·0 ± 0·00 a0·0 ± 0·01 a0·0 ± 0·01 a0·0 ± 0·01 a
 Mean3·24·62·45·05·33·5
 Total2·0 a3·3 b1·5 a4·8 b4·6 b3·2 a
 CamelliaRhododendronViburnumCamelliaRhododendronViburnum
P-valueMSP-valueMSP-valueMSP-valueMSP-valueMSP-valueMS
  1. aMetalaxyl and mefenoxam sensitivity: S, sensitive; I, intermediate; R, resistant.

  2. bMean area (cm2) of 30 replicate lesions ± standard deviation. Means within columns or rows followed by the same letter are not significantly different according to Fisher’s least significant difference test (< 0·05).

  3. cLesions 6 days after inoculation.

  4. dLesions 4 days after inoculation.

Main effects
 Treatment0·000040·23890·000058·25560·000020·00·043233·51090·020142·12340·027338·0328
 Block0·21340·06050·19520·08660·20690·03850·29140·05550·25900·06160·17340·0489
 Treatment × block interaction1·00000·00241·00000·00281·00000·00140·78870·00150·63540·01120·83370·0036

Using two procedures to assess the aggressiveness of P. ramorum isolates, zoospore suspension or wound-inoculation with colonized agar plugs, significant differences (< 0·001) were detected only in the experiments where a zoospore suspension was used (Table 2). Control treatments in all experiments did not develop lesions. A comparison of the different host species inoculated showed that, irrespective of the P. ramorum geographical or original host source, Camellia and Viburnum leaves developed lesion areas significantly smaller than Rhododendron in inoculations with a zoospore suspension, while Viburnum leaves had smaller lesion areas than the other hosts in the wound-inoculation procedure.

The use of a zoospore suspension as inoculum showed that isolates with varying levels (resistant or intermediate) of metalaxyl or mefenoxam resistance produced significantly larger lesion areas than sensitive isolates (Table 2). There was no significant difference in aggressiveness between the four isolates classed as intermediate or resistant to both metalaxyl and mefenoxam using this method.

Leaves wound-inoculated with P. ramorum isolates resistant to metalaxyl or mefenoxam developed larger lesions than those inoculated with sensitive isolates. The analysis of variance (Table 2) showed a significant difference between treatments (< 0·05). However, Fisher’s least significant difference test did not register differences between the treatments where P. ramorum isolates were used. In conclusion, despite some apparent correspondence between lesion area on the host and aggressiveness among P. ramorum isolates resistant to metalaxyl or mefenoxam, the relationship was weak and significant differences were not detected in aggressiveness among sensitive and resistant P. ramorum isolates.

Discussion

In Spain, P. ramorum has been found only in nurseries and garden centres, mainly in the north, and in all cases measures were taken to eradicate the pathogen. There is a potential risk of Mediterranean climates, especially mild and wet ones, being favourable for the establishment and spread of this pathogen, as reported in the Pest Risk Analysis for P. ramorum (http://rapra.csl.gov.uk/). Also, P. ramorum can survive as chlamydospores in Mediterranean climates with hot and dry summers, as demonstrated in California (Fichtner et al., 2007), and potentially also in colder climates with cold winters (Meentemeyer et al., 2004). However, it has never been found in natural ecosystems in Spain.

The symptoms observed on Rhododendron and Camellia plants were mainly leaf symptoms and in most cases only a few leaves per plant were affected. Stem lesions were observed on affected Viburnum plants. Most of the nurseries surveyed were trading nurseries and did not produce their own plants; plants had been imported from other European countries. After a positive case of P. ramorum was identified, the plants were destroyed and containment measures were applied, therefore a follow-up of the population of P. ramorum was not possible as it was eradicated from the affected nurseries.

All the isolates studied were mating type A1 ruling out the possibility of genetic recombination. Also, the fact that only one mating type has been found supports the theory of a newly introduced pathogen. Mating type A2 was not detected within the Spanish population of P. ramorum. As all NA1 and NA2 lineage isolates are of the A2 mating type, this intimates that the Spanish P. ramorum isolates are of the EU1 lineage. It also confirms that the presence of A2 mating type isolates in Europe is very rare and that such isolates may have been eradicated. Together with the absence of NA1 and NA2 isolates, this limits the risk of sexual reproduction in Europe (Vercauteren et al., 2011).

The results of the microsatellite analysis confirmed that all isolates belonged to the EU1 lineage of P. ramorum and 12 intralineage genotypes were detected, of which five were detected for the first time in Europe. This does not necessarily mean that these genotypes are unique to Spain. Studies of the genetic variation among European P. ramorum isolates are currently in progress and only a few populations have been studied in detail, including the Belgian (Vercauteren et al., 2010) and now the Spanish population. Therefore, it is possible that the five new genotypes detected in Spain will be detected in other European countries in the near future.

Most of the Spanish isolates belonged to EU1MG1, which is the most common genotype in Europe. All differences detected between genotypes could be explained by a stepwise mutation model from a single microsatellite founder genotype EU1MG1, which is still dominant (Vercauteren et al., 2010). The genotype EU1MG1 was detected in all the different geographical areas except one (geographical area 4), where only EU1MG13 and EU1MG18 were detected. Genotypes EU1MG13 and EU1MG18 were also found in Belgium and according to Vercauteren et al. (2010) these genotypes, together with EU1MG22, are also common in Europe and probably were derived originally from EU1MG1 and have been spreading through Europe via nursery shipments of Rhododendron and Viburnum. In Spain these two genotypes were only found on Rhododendron. This is consistent with this initial spread: genotypes EU1MG1, EU1MG13, EU1MG18 and EU1MG22 were also the earliest detected (2003–2006), while all other genotypes were detected in 2007 or 2008.

Genotype EU1MG29 was the second most common genotype detected in Spain and it probably increased its population within Spain, in contrast with genotypes EU1MG1, EU1MG13, EU1MG18 and EU1MG22, which seem to be introduced more frequently via the international trade. Genotype EU1MG29 was only found once more in Europe in Belgium in 2008 (Vercauteren et al., 2010). It has two marker mutations. It is not possible to determine whether these have happened independently or whether there was an international transfer of this genotype between Belgium and Spain (or from a third country), but in any case it is not frequent in the rest of Europe.

The five new genotypes detected in Spain were found in two different geographic areas in the north of Spain on Camellia and Rhododendron. This could just be due to the higher number of affected plants identified in those regions. The largest number of genotypes was detected on Rhododendron (nine), followed by Camellia (four) and Viburnum (two).

Only isolates of P. ramorum isolated from Camellia in 2007 were resistant to the fungicides metalaxyl and mefenoxam. Overall, there is very little metalaxyl resistance in the Spanish P. ramorum population and, in this case, it could be due to the abuse of these fungicides on Camellia. This is the first finding of resistance of P. ramorum isolates to phenylamide fungicides in Spain, in contrast with other regions in Europe where it has been detected at higher frequency (Vercauteren et al., 2010). The isolates that were resistant to both fungicides belonged to one of the newly detected genotypes (EU1MG38), which implies the resistance developed rather recently, after this genotype developed from EU1MG1. Inoculation tests showed that isolates that were resistant to the fungicides were more aggressive than sensitive isolates. These resistant isolates are genetically very uniform (all EU1MG38), were isolated from the same host species and from the same region and may therefore have originated clonally from an isolate that was more aggressive in the first place.

In conclusion, at present the number of positive cases of P. ramorum detected in Spain in ornamental nurseries is very low in comparison with the positive samples found in other European countries. Monitoring this pathogen will continue and all the efforts will be in place to avoid its establishment and spread to (semi-) natural ecosystems, as is for example occurring in the south-west of England.

Acknowledgements

The work of A. Vercauteren was supported in part by grant RT-05/04-PHYRAM1 of the Belgian Federal Public Service ‘Health, Food Chain Safety and Environment’. This research was financially supported by the Project AGL2007-64690/AGR (Ministerio de Educación y Ciencia, Spain).

Ancillary