Laboratory diagnosis of Phytophthora ramorum from field samples

Cheryl Blomquist1 and Tom Kubisiak2

1Associate Plant Pathologist (Diagnostician), California Department of Food and Agriculture, Plant Pest Diagnostics Branch, 3294 Meadowview Road, Sacramento, California 95832-1448.
2Plant Research Geneticist, USDA Forest Service, Southern Institute of Forest Genetics, 23332 Highway 67, Saucier, Mississippi 39574.

Introduction
A plant disease should never be diagnosed on the basis of a single test. Using as much information as possible leads to the most informed diagnosis. The species of host plant, its symptoms, the location of the plant, the status of the county or state (known infested versus not infested with the pathogen), the culture results, and the results of DNA tests should all be used to make the diagnosis. In the case of Sudden Oak Death (SOD), caused by Phytophthora ramorum (Pr) (Phylum Oomycota), different kinds of proof are required depending upon whether a sample comes from an infested county versus a county or state not yet known to be infested. In an infested county, recognized host plants with characteristic symptoms and an unequivocal DNA test result are adequate to confirm the presence of Pr. To confirm Pr in a previously uninfested county or state, or infecting a new host species, the pathogen must be grown in culture and identified using morphological characteristics and DNA sequence analysis. To be unequivocally confirmed, the sequence of the intergenic transcribed spacer (ITS) region of the ribosomal DNA of the suspect organism must exactly match that of Pr.

Symptoms associated with Pr infection look different in nurseries than in the wildlands of California. Nursery infections are characterized by large necrotic spots on rhododendron (Rhododendron spp.), typical of infection by many Phytophthora spp. (Figure 1). Dieback symptoms on Viburnum spp. also occur; however, to date, Pr has been found only on Viburnum in Europe. In the wildlands of California, symptoms include leaf tip necrosis with angular spotting in California bay laurel (Umbellularia californica) (Figure 2) and bleeding in oaks (Quercus spp.) (Figure 3). Large necrotic spots are symptoms in California coffeeberry (Rhamnus californica) and toyon (Heteromeles arbutifolia), and edge necrosis is a symptom in bigleaf maple (Acer macrophyllum) and California buckeye (Aesculus californica). However, other plant pathogens including other Phytophthora species cause identical symptoms on these host plants. Therefore, laboratory tests are necessary to determine if Pr is present.

Figure 1. Rhododendron leaf infected with Phytophthora ramorum. Notice the large necrotic lesion with a diffuse margin. Other Phytophthora spp. cause identical symptoms on rhododendron.
Figure 1. Rhododendron leaf infected with Phytophthora ramorum. Notice the large necrotic lesion with a diffuse margin. Other Phytophthora spp. cause identical symptoms on rhododendron.


 

Figure 2. California bay leaves infected with P. ramorum. Notice the leaf tip necrosis bordered by an uneven margin and the scattered squarish spots. A small plant sample is assayed from several symptomatic leaves including the margin between the necrotic and healthy tissue. P. ramorum can also be detected in some of the scattered spots.
Figure 2. California bay leaves infected with P. ramorum. Notice the leaf tip necrosis bordered by an uneven margin and the scattered squarish spots. A small plant sample is assayed from several symptomatic leaves including the margin between the necrotic and healthy tissue. P. ramorum can also be detected in some of the scattered spots.


 

Figure 3. Coast live oak bleeding due to P. ramorum infection. If the outer bark is peeled away, a canker with a defined edge is exposed and assayed for P. ramorum.
Figure 3. Coast live oak bleeding due to P. ramorum infection. If the outer bark is peeled away, a canker with a defined edge is exposed and assayed for P. ramorum.


 

Culturing Pr
Plant pathologists are fortunate to have a number of different laboratory tools to help them diagnose plant diseases. Classically, pathologists have plated infected plant tissue on general or selective media and used morphological characteristics to identify the pathogenic oomycete or fungus that grew into the media. For Phytophthora spp, a selective medium called PARP is used. PARP includes two antibiotics and an antifungal agent to prevent growth of competitive species of saprophytic bacteria and fungi. Figure 4 shows colonies growing out of lesion margins excised from four symptomatic California bay laurel leaves that were plated onto PARP media 3 to 5 days earlier. Three to five days after colonies first appear, they are examined by light microscopy for Pr's characteristic hyphae and chlamydospores. Culturing Pr from infected host plants seems to be dependent on the environmental conditions where the samples were collected, on host response, and on the presence of competing organisms in the specific plant species (Rizzo et al. 2002). In some infected host plants such as Douglas - fir (Pseudotsuga menziesii), Pr can be cultured only from samples collected during a limited time in the winter and spring. Except on rare occasions, culturing samples from host plants like Douglas-fir during other times of the year yields negative results. With these "hard-to-culture from" plant species, DNA-based technologies to determine if Pr is present in a symptomatic plant are more convenient and plants can be tested during much of the year.

Figure 4. Symptomatic California bay leaf pieces plated on PARP selective media after 5 days.
Figure 4. Symptomatic California bay leaf pieces plated on PARP selective media after 5 days.


 

DNA-based diagnosis of P. ramorum

Recently, a DNA-based method for detecting Pr was developed (Garbelotto et al. 2002). This method takes advantage of the sequence divergence between Pr and all other Phytophthora spp. (for which sequence data were available) in the ITS region of the nuclear ribosomal DNA gene repeat. We recently tested the first-round primers (Phyto1 and Phyto 4) (Garbelotto et al. 2002) on pure DNAs obtained from a number of Phytophthora spp. including P. boehmeriae, P. cambivora, P. cinnamomi, P. cryptogea, P. erythroseptica, P. gonapodyides, P. lateralis, P. megasperma, P. palmivora, P. parasitica, Pr, and P. syringae. Using the polymerase chain reaction (PCR), this primer pair was known to amplify a band of 687 base pairs in size from Pr. At high DNA concentrations, some cross-reactivity was observed in the expected size range with P. cambivora, P. cinnamomi, P. lateralis, and P. syringae. Cross-reactivity is a real concern and can lead to false-positive identification for Pr. In general, the cross-reactive bands were fainter than the band observed for Pr, and additional diagnostic bands were often amplified (Figure 5). As the DNA concentration was reduced, the cross-reactivity of the primers was no longer an issue. Given the standard 35-cycle PCR protocol, there was no detectable cross-reactivity observed with any of the Phytophthora spp. tested below 10 pg of DNA/ul of reaction. We found the presence of Pr to still be faintly detectable with as little as 2.5 fg DNA/ul of reaction (4000 times less DNA). The cross-reactivity of the primers with certain isolates of Phytophthora has since been reduced by further optimization of the PCR technique (M. Garbelotto, personal communication). In general, the technique appears to be highly specific for Pr especially in the range of DNA concentrations one might expect from extractions performed on symptomatic host tissues collected in the field.

Figure 5. DNAs from eight different pure Phytophthora isolates (order on gel: P. ramorum 016; P. ramorum 013; P. syringae 442; P. cinnamomi 447; P. cambivora 444; P. cambivora 443; P. lateralis 452; and P. lateralis 440) were amplified using the standard 35 cycle PCR protocol and the first-round primers Phyto1 and Phyto 4 (Garbelotto et al. 2002) at six different DNA concentrations (from upper left to lower right: 1000 pg/Ál of reaction; 100 pg/Ál; 10 pg/Ál; 1 pg/Ál; and 0.1 pg/Ál). Genomic lambda digested with the restriction enzyme PstI was used as the size standard.
Figure 5. DNAs from eight different pure Phytophthora isolates (order on gel: P. ramorum 016; P. ramorum 013; P. syringae 442; P. cinnamomi 447; P. cambivora 444; P. cambivora 443; P. lateralis 452; and P. lateralis 440) were amplified using the standard 35 cycle PCR protocol and the first-round primers Phyto1 and Phyto 4 (Garbelotto et al. 2002) at six different DNA concentrations (from upper left to lower right: 1000 pg/Ál of reaction; 100 pg/Ál; 10 pg/Ál; 1 pg/Ál; and 0.1 pg/Ál). Genomic lambda digested with the restriction enzyme PstI was used as the size standard.


 

DNA extraction from host tissues
Reliable use of a DNA test to assay for Pr requires that pathogen DNA be extracted from symptomatic plant tissues. Foliar samples from infected host species that support abundant sporulation such as rhododendron and California bay laurel (Davidson et al. 2002) or bark tissues around the margin of bleeding oaks appear to be the most reliable source of tissue for diagnosis. However, Pr is known to infect many other host species (Rizzo et al. 2002). In general, plants produce polyphenolic compounds along with tannins and other natural-occurring compounds (especially polysaccharides) that make the extraction of quality DNA difficult. Plants and oomycetes have strong cell walls that must be broken before the DNA can be released. Cells are usually broken manually with a mortar and pestle, mechanically with a dental amalgamator, or with more high-throughput technologies. Most leaves are easy to grind, but most woody tissue is not.

Many protocols for DNA extraction have been published. We have tested several of the more common protocols on Pr- infected leaf disks from several host species including bay laurel, rhododendron, and mountain-laurel (Kalmia latifolia) and have found some to be superior to others. One method found to work well across the host species tested was a nonionic detergent cetyltrimethylammonium bromide (CTAB)-based extraction protocol (Figure 6). Polysaccharides, polyphenolic compounds, and other enzyme-inhibiting contaminants often found in plant cells are generally removed because most do not precipitate with CTAB during the extraction process (Ausubel et al 1987.). Some of the more practical benefits of the protocol are that large quantities of reagents and solutions can be prepared at a fraction of the cost of commercially available kits, the protocol is fairly simple, and it is easily scaled from milligrams to grams of tissue. Unfortunately, it is not very amenable to high-throughput techniques (~50 samples per technician day). Another method found to work well across the host species tested was a commercial column or membrane-based DNA extraction kit (Figure 6). This protocol allows for the adsorption of DNA to a special membrane, thus allowing for the optimal removal of polysaccharides, polyphenols, and other unwanted plant metabolites and constituents. The main benefit associated with this protocol is that it is highly amenable to high-throughput technologies (~400 samples per technician day). Despite this, the cost is significantly higher per sample than with the CTAB-based protocol. A final method tested was a commercial resin-based DNA extraction kit. This protocol employs a resin that specifically binds and precipitates unwanted polysaccharides. Using this technique, DNAs obtained from two of the three host species tested were not amenable to enzyme manipulation (Figure 6). This protocol appears to be less reliable for the detection of Pr in infected host tissues than either the CTAB or membrane-based methods.

Figure 6. DNAs extracted from three different host species (bay laurel, rhododendron, and mountain-laurel) infected by P. ramorum were extracted using three different DNA extraction techniques [CTAB; DNeasy« kit (Qiagen Inc. Valencia, California, USA); and PhytoPure. kit (Amersham International plc, Buckinghamshire, England)] and amplified using the standard 35 cycle PCR protocol and the first-round primers Phyto1 and Phyto4 (Garbelotto et al. 2002). Genomic lambda DNA digested with the restriction enzyme PstI (λ-PstI) was used as the size standard. Lanes are: λ-PstI; H20 control; infected bay laurel/CTAB; infected bay laurel/CTAB; infected rhododendron/CTAB; infected rhododendron/CTAB; infected mountain- laurel/CTAB; λ-PstI; H20 control; uninfected mountain- laurel/DNeasy«; uninfected mountain- laurel/DNeasy«; infected bay laurel/DNeasy«; infected bay laurel/DNeasy«; infected rhododendron/DNeasy«; infected rhododendron/DNeasy«; infected mountain laurel/DNeasy«; λ-PstI; H20 control; uninfected mountain- laurel/PhytoPure.; uninfected mountain- laurel/PhytoPure.; infected bay laurel/PhytoPure.; infected bay laurel/PhytoPure.; infected rhododendron/PhytoPure.; infected rhododendron/PhytoPure.; infected mountain- laurel/PhytoPure.; λ-PstI.
Figure 6. DNAs extracted from three different host species (bay laurel, rhododendron, and mountain-laurel) infected by P. ramorum were extracted using three different DNA extraction techniques [CTAB; DNeasy« kit (Qiagen Inc. Valencia, California, USA); and PhytoPure. kit (Amersham International plc, Buckinghamshire, England)] and amplified using the standard 35 cycle PCR protocol and the first-round primers Phyto1 and Phyto4 (Garbelotto et al. 2002). Genomic lambda DNA digested with the restriction enzyme PstI (λ-PstI) was used as the size standard. Lanes are: λ-PstI; H20 control; infected bay laurel/CTAB; infected bay laurel/CTAB; infected rhododendron/CTAB; infected rhododendron/CTAB; infected mountain- laurel/CTAB; λ-PstI; H20 control; uninfected mountain- laurel/DNeasy«; uninfected mountain- laurel/DNeasy«; infected bay laurel/DNeasy«; infected bay laurel/DNeasy«; infected rhododendron/DNeasy«; infected rhododendron/DNeasy«; infected mountain laurel/DNeasy«; λ-PstI; H20 control; uninfected mountain- laurel/PhytoPure.; uninfected mountain- laurel/PhytoPure.; infected bay laurel/PhytoPure.; infected bay laurel/PhytoPure.; infected rhododendron/PhytoPure.; infected rhododendron/PhytoPure.; infected mountain- laurel/PhytoPure.; λ-PstI.


 

Although CTAB has been the standard for DNA extraction from leaf tissues, to date we have been unable to obtain reliable first-round PCR amplification from infected oak bark using this method. We have, however, obtained good quality DNA from a membrane-based technique. Freeze-drying woody tissue before grinding makes the tissue brittle and thus facilitates cell disruption.

Conclusion
Detection of a plant pathogen with quarantine status can have severe economic consequences. Therefore, diagnoses should be made using all available information. Until Phytophthora spp. more closely related to Pr are found, morphological identification and current DNA tests seem to provide reliable means of detection without a high number of false-positive results. To minimize the number of false-negatives, a robust DNA extraction procedure is recommended, i.e., one that is applicable across the widest range of host species and tissues. The current ITS-based PCR technique (described above) is highly specific for Pr at low DNA concentrations and has proven useful for detecting it in infected host tissues. Other DNA-based methods are being developed for the detection of Pr; these include assays based on the sequences for other nuclear genes such as ▀-tubulin, or sequences for organellar or mitochondrial genes such as cox2. A promising new technique called single strand conformation polymorphism (SSCP) analysis may eventually allow for the detection and unambiguous identification of most, if not all, Phytophthora spp. In the future, we look forward to additional molecular assays for detecting and identifying Pr.

References

Ausubel, F. M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith J. A, and Stuhl, K. 1987. Current protocols in molecular biology. Wiley, New York.

Davidson, J.M., Rizzo, D.M., Garbelotto, M., Tjosvold, S., and Slaughter, G.W. 2002. Phytophthora ramorum and sudden oak death in California: II. Transmission and survival. USDA Forest Service Gen. Tech. Rep. PSW-GTR-184: 741-749.

Garbelotto, M., Rizzo, D.M., Hayden, K., Meija-Chang, M., Davidson, J.M., and Tjosvold, S. 2002. Phytophthora ramorum and sudden oak death in California: III. Preliminary studies in pathogen genetics. USDA Forest Service Gen. Tech. Rep. PSW-GTR-184: 765-774.

Rizzo, D.M., Garbelotto, M., Davidson, J.M., Slaughter, G.W., and Koike, S.T. 2002. Phytophthora ramorum and sudden oak death in California: I. Host relationships. USDA Forest Service Gen. Tech. Rep. PSW-GTR-184: 733-739.