Annuals and Herbaceous Perennials Tolerant or Resistant to Phytophthora Species in the Landscape¹

This research provides knowledge of 22 plant cultivars within 13 herbaceous plant species that can be successfully cultivated in Phytophthora-infested soils for ornamental plant producers, landscapers, and the home gardener. This study also identified ornamental plants that are not ideal candidates for infested beds due to their susceptibility to other common diseases. Future investigations should re-evaluate plants that did not perform well due to abiotic issues. Additional plant cultivars exist that also need to be evaluated in order to provide stakeholders with as many options as possible for areas infested with Phytophthora species.

Phytophthora (de Bary 1876) is a genus of soil-inhabiting oomycetes that attack a broad range of economically important crops including vegetables, small and tree fruits, herbaceous and woody ornamentals, forest trees, and field crops (Erwin and Ribeiro 1996, Patel et al. 2016). In general, Phytophthora species cause root, crown, or stem rots, but foliar blight can also occur on some plant species (Banko and Stefani 2000, Lamour et al. 2003). Symptoms in landscape plants can vary depending on the plant species, but generally include loss of older leaves, decline in plant vigor and growth, branch dieback, crown rot, root rot, wilting, chlorosis, plant collapse, and death (Banko and Stefani 2000). The disease is favored by excessive moisture (Banko and Stefani 2000) and can be particularly devastating and difficult to control if it becomes established.

In North Carolina, Phytopthora nicotianae Breda de Haan (formerly P. parasitica [Dastur] G.M. Waterh.), Phytophthora drechsleri Tucker, and Phytophthora tropicalis Aragakia and J.Y. Uchida are commonly found attacking ornamental plants (Hwang and Benson 2005), but the susceptibility of many popular bedding plants is still largely unknown and some are suspected to be resistant (Creswell et al. 2011). In this study, disease resistance is defined as the ability of the plant to suppress growth of the pathogen on or in the plant. Fungicide applications can provide some disease suppression (Hagan et al. 1996), but this practice involves frequent applications that are costly and not applicable to gardeners and landscapers who grow these plants on a small scale. Therefore, obtaining the latest research and knowledge of resistant plant species to Phytophthora species, and understanding the susceptibility of these plants to other common diseases, would make disease avoidance possible in landscape settings and provide an economically and environmentally sustainable approach. The objective of this study was to evaluate one or two cultivars of 16 annuals and 14 herbaceous perennials for resistance to Phytophthora root or crown rot and for susceptibility to other common diseases that may occur.

In spring 2018, bedding plant taxa were selected for evaluation based on availability, popularity, anecdotal or experimental evidence of resistance to Phytophthora species (Banko and Stefani 2000, Creswell et al. 2011, Hagan and Akridge 2001, M. Yelanich, Van Wingerden International, personal communication) and evidence of resistance to other diseases (Beckerman and Lerner 2009). Four replicates of 16 annual species and 14 herbaceous perennial species were selected (Table 1). An additional four plant species and six cultivars known to be susceptible to Phytophthora species were selected as susceptible controls: Barberton daisy (Gerbera jamesonii Bolus ex Hooker) (‘Crazy Daisy'), dusty miller (Jacobaea maritima [L.] Pelser and Meijden) (‘Silver Dust'), annual vinca (Catharanthus roseus [L.] G. Don) (‘Pacifica Raspberry'), and petunia (Petunia x hybrida [Hooker] Vilmorin) (‘Purple Wave', ‘Easy Wave White', ‘Easy Wave Red') (Olson and Benson 2011, Hao et al. 2010).

Three landscape beds [each approximately 18.6 m² (200 sq ft)] were established at three research stations (one bed per location) in western and central North Carolina [Mountain Horticultural Crops Research Station (MHCREC), Mills River, NC; Mountain Research Station (MRS), Waynesville, NC; and Piedmont Research Station (PRS), Salisbury, NC] in uncultivated areas that had no previous agricultural use. Beds were tilled before establishing the bed frame and then filled with a 1:1 blend of compost (blend of pine soil conditioner with mushroom, dairy, and leaf composts) and screened, enriched top soil (1:1 blend of compost and sandy-loam topsoil). Each bed was divided into four rectangular quadrants as if a “cross” was drawn over the bed. Each quadrant was a mirror image of the adjacent quadrants [each 4.65 m² (50 sq ft)] and plants were laid out in the same pattern in each quadrant. The taller plants were placed towards the center of the bed and shorter plants placed toward the edge of the bed. The bed at PRS was planted on May 15, 2018, the bed at MRS was planted on May 17, 2018, and the bed at MHCREC was planted on May 18, 2018. Plants were allowed to establish for two weeks before inoculation. Pendulum® 2G granular herbicide (BASF, Ludwigshafen, Germany) was applied as a pre-emergent after planting to each bed (PRS: May 25, 2018; MHCREC: May 31, 2018; MRS: June 1, 2018) at a rate of 112 kg·ha⁻¹ (100 lb·A⁻¹) with a handheld spreader to suppress weeds. Care was taken to pass a hand over each plant to dislodge any herbicide granules. One plant of each of 45 cultivars was planted in each quadrant of the bed so that there were four plants of each cultivar in each bed except for Impatiens hawkeri ‘Sunpatiens Compact Lilac' (n=9; 3 plants per bed) and Rudbeckia hirta ‘Indian Summer' (n=11; 3 or 4 plants per bed). Therefore, there were twelve plants of each cultivar, collectively, among the three beds, except for the two cultivars where there were only nine or eleven plants in total. Plants were spaced at 30 to 46 cm (12-18 in) between each cultivar and pine bark mulch was spread on the bed surfaces to ensure adequate soil moisture and to suppress weeds.

Two isolates each of P. nicotianae (17-008, 17-036), P. drechsleri (16-041, 17-025), and P. tropicalis (16-043, 17-072) were selected from a collection of isolates from bedding plants in North Carolina. If the mating type was unknown, isolates were grown on 5% clarified V8 agar (CV8) (V8® juice, Campbell Soup Company, Camden, NJ, USA; water; calcium bicarbonate, Sigma-Aldrich, St. Louis, MO, USA) (Ferguson and Jeffers 1999) and challenged with an isolate of P. nicotianae of known mating type A1 (18-020) for 7 to 14 days at 22 C (72 F) (Tooley et al. 1989). Isolates were characterized as mating type A1 if no oospores were formed or A2 if oospores were produced.

The identification of the selected isolates studied were confirmed before inoculation. Pure cultures of Phytophthora species were grown on 5% CV8 agar for 3 to 5 days and a single agar plug (5 mm diameter) was transferred to a petri plate containing 10% CV8 broth. Cultures were allowed to grow at room temperature for 3 to 5 days into thick mycelial mats and then collected by vacuum filtration and rinsed with distilled water through Fisherbrand plain filter paper (Fisher Scientific, Waltham, MA, USA) (Klassen et al. 1987). The mycelium was stored in 2 ml microcentrifuge tubes at -20 C (-4 F) until processed.

Frozen mycelial mats were disrupted with two sterile 3-mm glass beads and shaken with a BioSpec Mini Beadbeater (Bartlesville, OK, USA) at 42 rpm for 20 s. DNA was extracted using the Omega Bio-Tek Plant DNA Kit (Norcross, GA, USA) and subjected to PCR by amplifying a portion of the internal transcribed spacer regions using primer pairs ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) and ITS6 (5′-GAAGGTGAAGTCGTAACAAGG-3′) (Cooke and Duncan 1997, White et al. 1990, Cooke et al. 2000). In cases where ITS was non-informative, the β-tubulin region with primers β-tubF (5′- CGGTAACAACTGGGCCAAGG-3′) and β-tubR (5′- CCTGGTACTGCTGGTACTCAG-3′) (Kroon et al. 2004), and the cytochrome c oxidase subunit 1 region using cox1F (5′-TCCGTAGGTGAACCTGCGG-3′) and cox1R (5′-TCCTCCGCTTATTGATATGC-3′) (Martin et al. 2012) were used to differentiate between species. Amplifications were performed using a Bio-Rad iCycler Thermal Cycler (Version 4.006, Bio-Rad Laboratories, Inc, Hercules, CA, USA). Thermocycler conditions were as follows: (i) initial denaturation at 94 C (201 F) for 2 min (ii) 35 cycles at 94 C (201 F) for 30 sec, 52 C (126 F) for 30 sec, and 72 C (162 F) for 1 min (iii) and a final extension at 72 C (162 F) for 10 minutes (Kroon et al. 2004). Successful amplification was confirmed by gel electrophoresis. PCR products were purified using the Invitrogen Quick PureLink kit (Thermo Fisher Scientific, Waltham, MA, USA). Purified products were sent to McLab Molecular Cloning Laboratories (San Francisco, CA) for Sanger sequencing and consensus contigs were aligned and trimmed using Geneious Prime software (Auckland, New Zealand). Resulting sequences were compared to accessions in GenBank and Phytophthora-ID.org (Grünwald et al. 2011) for identification.

To prepare the inoculum, each isolate was grown on 5% CV8 agar at 22 C (72 F) for 5-7 days. Five plugs (5 mm diameter) were aseptically transferred into a sterile mixture of 10% clarified V8 juice broth and fine vermiculite (25% v:v) and maintained at 20 to 22 C (68 to 72 F) for 14 days in the dark (Ivors 2015). A sample (approximately 5 ml) of each container of infested vermiculite was grown on CV8 agar for 1-2 days at 22 C (72 F) just prior to inoculation to confirm that the organism had fully colonized the vermiculite and that no contamination had occurred. To infest the soil, five parallel, shallow trenches were dug 8-10 cm (3-4 in) into the soil and approximately 940 ml (32 fl oz) of inoculum was spread in each trench [4,700 ml (160 fl oz)] of inoculum per bed). Trenches were re-covered with soil and irrigation was initiated to provide adequate moisture for the inoculum. Beds were infested on June 1, 2018 and again on June 28, 2018 to ensure the inoculum was active and all plants were exposed to the inoculum.

Plants were rated every 10 to 21 days for disease incidence and severity based on the percentage of plants showing symptoms of infection by Phytophthora species including root rot, crown rot, or aerial blight or disease by another pathogen (Table 1) to avoid recommending plants that are susceptible to other diseases. The disease rating scale included: 0 = healthy, not exhibiting symptoms of infection by Phytophthora species, 0% wilting or chlorosis; 1 = slight wilting, ≤10% wilting or chlorosis; 2 = moderate wilting, 10-50% wilting or chlorosis; 3 = severe wilting, >50% wilting or chlorosis; 4 = 100% wilting or chlorosis, still alive; 5 = dead. Statistical analyses were not conducted on the data collected as this would require each cultivar to be compared to a healthy (non-infected) control. This was not achievable due to financial constraints.

As disease symptoms progressed, symptomatic plants were removed, transported to the laboratory, and pathogen isolations were attempted to determine the presence of Phytophthora on roots and crown. Plants also were observed for other diseases as they occurred and were diagnosed. Plants were either diagnosed in the field based on symptoms or were sent to the North Carolina State University Plant Disease and Insect Clinic for diagnosis. Resulting diagnoses were identified to the genus level or the common name of the disease was recorded. Roots and stems were washed with tap water and were cut into pieces approximately 1 cm (0.4 in) in length and embedded onto a semi-selective medium for Phytophthora species, PARPH-V8A (Jeffers and Martin 1986). After 3 to 5 days of incubation at 22 C (72 F), suspected colonies of Phytophthora species were sub-cultured onto clarified V8 agar (CV8). Species were identified based on sporangium morphology (Gallegly and Hong 2008) after 24 h in 1.5% non-sterile soil extract solution (NS-SES) (Jeffers and Aldwinkle 1987). DNA was extracted from species that could not be identified based on morphological features by methods described above.

A soil-baiting bioassay (Ferguson and Jeffers 1999) was used to confirm the species of Phytophthora in the soil of each landscape bed were active in August and October 2018. Five to six soil sub-samples were compiled and maintained at 22 C (72 F) for less than four days until processed. Three sub-samples of soil [50 cm³ (3.1 in³)] from each bed were each placed in a plastic cup and flooded with 100 ml (3.4 fl oz) of deionized water. Six floating leaf discs of Camellia japonica L. and Rhododendron catawbiense Michx. were placed in each cup and cups were maintained at 22 C (72 F) for 48 to 72 hr. Leaf discs were then embedded onto PARPH-V8 medium and maintained at 20 C (68 F) for 3 to 10 days. Suspect colonies of Phytophthora species were sub-cultured and identified by morphology and DNA sequencing as described above. All resulting cultures of Phytophthora species were placed in storage on CV8 discs in sterile water with two hemp seeds at 22 C (72 F) in the dark.

At the end of the growing season, plants from all three landscape beds were removed from the bed and washed to dislodge soil. Final plant evaluations were conducted visually and based on the following scale: Excellent = no disease symptoms, excellent floral quality, and plants survived the entire growing season; Good = minor disease symptoms (<25% leaf area affected), good floral quality, and most of the plants survived the entire growing season; Fair = moderate disease symptoms (∼50% leaf area affected) and fewer than half (<6) of the plants died before the end of growing season; and Poor = severe disease symptoms (>50% leaf area affected) and more than half (>6) of the plants died before the end of growing season. Other = more than half (>6) of the plants had abiotic or unknown issues that prevented a fair trial. On 15, 17, and 22 Oct 2018 at MRS, MHCREC, and PRS, respectively, two healthy plants of each species that remained in each bed were removed and assayed for the presence of Phytophthora on the roots or stem. Root pieces (1 to 3 cm length) from selected plants were embedded onto PARPH-V8 to determine the presence of Phytophthora species as described above. Suspected colonies of Phytophthora species were sub-cultured and identified based on methods outlined above.

At the end of the growing season, 17 of the suspected resistant plant cultivars did not exhibit symptoms of infection by Phytophthora species or other diseases and were rated as Excellent, five were rated as Good, seven were rated as Fair, eight were rated as Poor, and eight rated as Other (Tables 1 and 2). These ratings reflect overall disease susceptibility; however, not all plants were exposed to the same pathogens (other than species of Phytophthora used in this study) so conclusions drawn from susceptibility to other diseases must be considered with caution.

Of the susceptible plants, P. nicotianae and/or P. drechsleri were isolated from each of the plants known to be susceptible to Phytophthora species in each bed, which confirmed that the inoculum was active (Table 2). Soil bait assays also confirmed the presence of Phytophthora species in the beds; however, P. tropicalis was not detected in any of the baits or the susceptible plants. Interestingly, we detected P. cryptogea at MRS by soil baiting, although this species was not intentionally introduced into the bed. It is possible this species was introduced into the bed through one or more of the plants as Phytophthora species are known to be transported through ornamental plants (Bienapfl and Bolci 2014, Callaghan and Guest 2015, Goss et al. 2011). Plants were not assayed before planting since they appeared healthy, so it is unclear how this organism became established in the bed.

Of the suspected resistant cultivars, Phytophthora species were recovered from 11 plant species and 12 cultivars among all three beds: 12 cultivars at MRS, four cultivars at PRS and three cultivars at MHCREC. Numerically, a larger number of susceptible cultivars were infected by Phytophthora spp. at MRS than the other two beds. It is unclear why this occurred, but it may be that some underlying condition in the bed at MRS promoted disease, such as extended periods of saturated soil, a condition known to promote diseases caused by species of Phytophthora (Erwin and Ribeiro 1996), but soil moisture was not measured in this study. However, total rainfall for the period that each bed was planted was numerically greater at MHCREC and PRS than at MRS and was 72.0 cm (28.3 in), 70.4 cm (27.7 in), and 57.0 cm (22.5 in), respectively, and does not add evidence to suggest our theory of increased soil moisture. The cause of the numerical disparity in Phytophthora affected plants in each bed remains elusive. Of the evaluated plants, P. nicotianae was recovered most frequently (30/48 plants [62.5%]) and from twelve plants in all three beds of petunia ‘Yellow Madness' in all three beds, eleven plants of petunia ‘Violet Picotee' among all three beds, six plants of calibrachoa ‘SuperCal' among all three beds, and one plant of zinnia ‘Star White'. All plants had either died or were evaluated as Poor by the end of the season except for zinnia—it was evaluated as Excellent and not showing any above ground symptoms. The cultivars of petunia and calibrachoa had been suggested as potentially resistant to Phytophthora species, but this study confirms that this is not the case. This result is not surprising given the susceptibility of petunia and petunia hybrids, in general (Erwin and Ribeiro 1996, Hwang and Benson 2005, Olson and Benson 2011).

Of the evaluated plants, P. drechsleri was isolated from six plants at MRS: one plant of calibrachoa ‘SuperCal', one plant of mealy blue sage ‘Victoria Blue', one plant of fernleaf yarrow ‘Moonshine', one plant of African marigold ‘Inca Yellow', and two plants of blanket flower ‘Goblin'. Phytophthora drechsleri was not found on any of the evaluated plants at MHCREC or PRS. Only the African marigold, calibrachoa, and fernleaf yarrow were dead at the end of the season; all others from which P. drechsleri was recovered were not showing symptoms. Given the low incidence of this oomycete on plants evaluated in this study, it is difficult to draw any firm conclusions. The reported host range of P. drechsleri is over 400 plant species (Erwin and Ribeiro 1996, Farr and Rossman 2020) and, yet, this is the first report of P. drechsleri on these twelve plants. As mentioned above, there may have been an unknown, underlying condition with the bed at MRS that was different from the other two beds and that induced P. drechsleri to cause disease on these plants, such as extended periods of saturated soil, but this was not measured in the study. Therefore, the role of P. drechsleri on these plants is unknown and requires further study.

Phytophthora tropicalis was only isolated from one vinca ‘Cora Strawberry', but was not isolated from any of the susceptible plants. It is unknown why the low incidence of this pathogen occurred. The host range of P. tropicalis is less than 60 plant species (Farr and Rossman 2020), many of which are not considered ornamentals and the lack of preferred hosts is one possible explanation.

We only detected P. cryptogea from MRS from one plant of blanket flower ‘Mesa Bicolor', two plants of mealy blue sage ‘Victoria Blue', two plants of celosia ‘New Look Celosia', two plants of ornamental grass ‘Rotstrahlbusch', and two plants of bee balm ‘Jacob Cline'. This pathogen was not intentionally introduced to our landscape beds and, yet, it was recovered from multiple plant species. It is possible it was introduced either through one of plants, as mentioned above, or spread naturally from surrounding areas. Regardless, we will evaluate more plants against this species if more studies like this one are pursued in the future.

Other common diseases not attributed to Phytophthora species were also observed on several plants throughout the study (Table 2) to avoid recommending plants that are susceptible to other diseases that may occur in the landscape. These pathogens were identified to genus morphologically or the common name of the disease was recorded. The fungal genera identified and the number of plants affected by each were: Alternaria (57), Cercospora (8), Cladosporium (1), Entyloma (10), Fusarium (3), Phytopythium (1), Rhizoctonia (5), Septoria (10), and Verticillium (3). Powdery mildew (34) and leaf rust (4) fungi also were observed. The presence of other diseases presumably caused by the fungi mentioned above prevented those plants from being rated as Excellent to avoid recommending plants that are susceptible to other diseases, except for narrowleaf zinnia ‘Star White', which remained in Excellent condition. However, those plants in the Good rating category were not affected enough by the fungi listed to not recommend them as possible alternative plants in Phytophthora infested beds. It is not surprising that a number of other fungal genera that are known to cause disease were detected as the climate in North Carolina is generally humid and warm with frequent rainfall, which are conditions known to favor such diseases.

Plants that died prematurely or were rated as Poor due to abiotic or unknown issues included phlox ‘Phlox Star' and ‘Intensia Red Hot', French marigold ‘Disco Mix' and ‘Disco Yellow', garden phlox ‘Amethyst True Gal', yarrow ‘Desert Eve Red', and black-eyed Susan ‘Indian Summer' and ‘Prairie Sun'. Even though no species of Phytophthora were isolated from these plants, we could not yet recommend them until they can be re-evaluated to provide a more accurate and fair representation of their performance in Phytophthora-infested beds.

This study provides evidence of 22 plant cultivars within 13 herbaceous plant species that are resistant to Phytophthora nicotianae, Phytophthora drechsleri, and Phytophthora tropicalis and provides information of susceptibility to other diseases that may occur in the landscape. Continuation of this study to evaluate additional plant species and cultivars is needed to provide stakeholders with more options for landscape beds infested with species of Phytophthora.