Hazard Analysis for PhytophthoraSpecies in Container Nurseries: ThreeCase Studies

Jennifer L. Parke2, Neelam R. Redekar, Joyce L. Eberhart,

and Fumiaki Funahashi1

ADDITIONAL INDEX WORDS. disease management, irrigation water, oomycetes,production practices, Phytopythium, Pythium, systems approach

SUMMARY. Phytophthora species cause crop losses and reduce the quality of green-house and nursery plants. Phytophthora species can also be moved long distances bythe plant trade, potentially spreading diseases to new hosts and habitats. Phytosa-nitary approaches based on quarantines and endpoint inspections have reduced, butnot eliminated, the spread of Phytophthora species from nurseries. It is thereforeimportant for plant production facilities to identify potential sources of contami-nation and to take corrective measures to prevent disease. We applied a systemsapproach to identify sources of contamination in three container nurseries in Ore-gon, California, and South Carolina. Surface water sources and recaptured runoffwater were contaminated with plant pathogenic species at all three nurseries, butone nursery implemented an effective disinfestation treatment for recycled irriga-tion water. Other sources of contamination included cull piles and compost thatwere incorporated into pottingmedia, infested soil and gravel beds, used containers,and plant returns. Management recommendations include preventing contact be-tween containers and contaminated ground, improving drainage, pasteurizingpotting media ingredients, steaming used containers, and quarantine and testing ofincoming plants for Phytophthora species. These case studies illustrate how recycledirrigationwater can contribute to the spread ofwaterborne pathogens and highlightthe need to implement nursery management practices to reduce disease risk.

Diseases caused by Phytoph-thora species are among themost damaging to greenhouse

and nursery-grown horticultural crops(Jones and Benson, 2001; U.S. De-partment of Agriculture, 2009). These

pathogens cause damping-off diseases,root rot, stem cankers, shoot diebackand foliar blight of annuals, herbaceousperennials, and woody plants. WhilePhytophthora includes soilborne andaerial species (Garbelotto et al., 2018),they are water molds, meaning theyrequire water to complete their lifecycles. Sporangia are formed duringmoist conditions, releasing zoosporesthat swim throughwater to infect plantroots, stems, and leaves. Phytophthoraand other plant pathogenic oomycetesincluding Phytopythium and Pythiumare common contaminants of green-house and nursery irrigation systems(Ivors and Moorman, 2017).

In addition to causing crop lossesin the nursery and reducing plant

quality, Phytophthora species can alsobe spread long distances by the nurserytrade, and somepose risks for forests andother natural vegetation. For example,the suddenoak death pathogenPhytoph-thora ramorum was likely introduced toNorth America on nursery plants in themid-1990s (Goss et al., 2009). In Cal-ifornia and Oregon, sudden oak deathhas killed �35 million forest trees(Cobb, 2018). Despite quarantines im-posed on nurseries in California, Ore-gon, and Washington, P. ramorum wasdispersed across the country with thenursery trade (Goss et al., 2009). Otherexamples of Phytophthora species spreadtowildlands by the plant trade includeP.lateralis, which causes port-orford-cedarroot disease (Hansen et al., 2000), andP. tentaculata, a pathogen that spreadfromnativeplant nurseries to restorationsites (Garbelotto et al., 2018; Rooney-Latham et al., 2015; Sims andGarbelotto, 2018).

Nursery plant distribution sys-tems are effective at moving patho-gens (Jung et al., 2018; Liebholdet al., 2012). Infected plants maynot show symptoms (Parke andLewis, 2007). In addition, some ofthe most widely used oomycete-specific pesticides, such as mefenoxemand fosetyl-Al, are fungistatic ratherthan fungicidal. Application of thesematerials can delay the developmentof symptoms and prevent pathogendetection until after plants are ship-ped. Resistance to mefenoxam hasalso developed in many nurseries(Olson et al., 2013). Once nurserybeds are infested, it is difficult toeradicate Phytophthora species. Soilsteaming (Schweigkofler et al., 2014)and soil solarization (Funahashi andParke, 2016) are effective but requirelarge energy inputs or summer fallowperiods, respectively. It is far less ex-pensive to implement protective mea-sures preventing disease than it is toeradicate Phytophthora species oncethey have established.

UnitsTo convert U.S. to SI,multiply by U.S. unit SI unit

To convert SI to U.S.,multiply by

0.4047 acre(s) ha 2.471129,574 fl oz mL 3.3814 · 10–5

3.7854 gal L 0.26422.54 inch(es) cm 0.3937

16.3871 inch3 cm3 0.06101 micron(s) mm 11 ppm ng�mL–1 1

(�F – 32) O 1.8 �F �C (�C · 1.8) + 32

Received for publication 4 Feb. 2019. Accepted forpublication 12 Apr. 2019.

Published online 5 August 2019.

Department of Crop and Soil Science, Oregon StateUniversity, Corvallis, OR 97331

Our project was funded by the U.S. Department ofAgriculture National Institute of Food and Agricul-ture Specialty Crop Research Initiative (USDA-NIFA-SCRI) program (grant 2014-51181-22372).Mention of a trademark, proprietary product, orvendor does not constitute a guarantee or warrantyof the product by Oregon State University and doesnot imply its approval to the exclusion of otherproducts or vendors that also may be suitable. Weare grateful to the three anonymous nurseries for theircooperation during the course of this study.

This paper is based on information presented duringthe CleanWateR3 program sessions, held as part of theASHS Annual Conference, 30 July–3 Aug. 2018 inWashington, DC.

1Current address: Copine International Agricultureand Environment LLC, Gifu, Japan.

2Corresponding author. E-mail: Jennifer.Parke@oregonstate.edu.

This is an open access article distributed under the CCBY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/).

https://doi.org/10.21273/HORTTECH04304-19

• December 2019 29(6) 745

For several years we have applieda systems approach to identify sourcesof pathogen contamination withinnurseries (Parke and Gr€unwald,2012). The system is a modificationof the hazard analysis of critical controlpoints (HACCP) approach designedto ensure food safety in food process-ing facilities. Critical control pointsare defined as the best stages in a pro-duction process at which significanthazards of contamination can be pre-vented or reduced. In a nursery, crit-ical control points for contaminationby plant pathogens commonly in-clude plants brought in from otherproduction facilities, potting mediaor ingredients, cull piles that are usedfor making potting media, recycledcontainers, the soil or gravel underthe containers, and untreated irriga-tion water. Once the source(s) ofcontamination is known, nurserygrowers can eliminate or reduce therisk of disease by implementing aman-agement strategy that targets thecontamination source (Junker et al.,2016; Parke and Gr€unwald, 2012;Parke et al., 2014).

In this article, we describe howwe conducted a hazard analysis ofthree container nurseries to deter-mine sources of contamination forPhytophthora species and show howour findings can inform managementstrategies to eliminate or reduce thesources of contamination.

Material and methods

SAMPLE COLLECTION, LEAF

BAITING, AND FILTRATION. Hazardanalyses of critical control points wereperformed at three container nurser-ies located in Oregon (Nursery A),California (Nursery B), and SouthCarolina (Nursery C) in Oct. 2017,Dec. 2017, and June 2016, respec-tively. Each nursery was surveyed tolocate areas with high disease risks toallow for thorough sampling of theseareas and surroundings. The types ofsamples collected from these nurser-ies included 1) diseased plants show-ing symptoms such as dieback, rootrot, shoot blight, leaf lesions, defoli-ation (Fig. 1A–E); 2) soil, gravel, andleaf debris from underneath the potsfrom a symptomatic area (Fig. 2A–D); 3) media components such aspotting mix and compost (Fig. 3B);4) scrapings from used containers tobe recycled (Fig. 3C); 5) plant debris

in cull piles (Fig. 3E); and 6) irriga-tion water from main sources, reten-tion reservoirs, and runoff channels(Figs. 3A and 4A). At least 50 sampleswere collected in each nursery. Weused a combination of enzyme-linkedimmunosorbent assay–based methods,culture-based methods, and DNA se-quencing approaches to detect andidentify Phytophthora species.

Plants exhibiting Phytophthora-like symptoms on roots or foliagewere tested with Phytophthora on-sitedetection kits (Fig. 4C and D), and ifpositive, directly plated on Phytoph-thora-selective media (Parke et al.,2014) (Fig. 4E). For Nursery A andB, we used the Phytophthora Rapidtest kit (Pocket Diagnostic; Abing-don Health, Sand Hutton, UK); forNursery C we used the ImmunoStriptest (Agdia, Elkhart, IN). DNA frompure-isolate cultures was amplifiedusing internal transcribed spacerprimers ITS4 and ITSDC6 (detailslater in the article) and the nucleotide

sequence of the polymerase chainreaction (PCR) product was deter-mined with the Sanger sequencingmethod to identify pathogen species.

All sample types (including dis-eased plants) were baited using pesti-cide-free leaves of ‘Grandiflorum’catawba rhododendron (Rhododen-dron catawbiense) grown in theOregon State University researchgreenhouses (Fig. 4F). For diseasedplants, either pour-through water[collected by pouring tap water intothe pot and collecting the leachate(Swiecki et al., 2018)] or the rootballs placed in a plastic bag andflooded with deionized water (Fig.4G) were baited. For other sampletypes such as soil, gravel, scrapings ofused containers, media components,and plant debris, �200 cm3 of mate-rial was placed in 1-gal plastic bags.Deionized water (1 L) was added tothese bags for baiting at room tem-perature (19 to 21 �C). Irrigationwater (1 L) was directly used for

Fig. 1. Diseased plants found at container nurseries in Oregon, California, orSouth Carolina: (A) pine (Pinus sp.) mortality, (B) juniper (Juniperus sp.) root rot(right) compared with healthy plant (left), (C) foliar blight of holly (Ilex sp.), (D)azalea (Rhododendron sp.) dieback, and (E) bald cypress (Taxodium distichum)mortality.

746 • December 2019 29(6)

WORKSHOP

baiting. For baiting, one-half ‘Gran-diflorum’ catawba rhododendron leafwas floated in a sample bag for 3 d,followed by incubation for 7 d more

in damp paper towels. If lesionsformed, up to 10 leaf disks (6 mmdiameter) of lesioned tissue were re-moved with a hole punch and stored

in bags with silica gel (Ockels et al.,2007) for future DNA extraction andidentification of all Phytophthora spe-cies present through Illumina MiSeq(Illumina, San Diego, CA) sequencing.Illumina MiSeq is a high-throughputsequencing platform that allows foridentification of individual speciesfrom a mixed pool of DNA in envi-ronmental samples.

An additional 1 L of irrigationwater was filtered through 5-mmmixed cellulose ester membrane fil-ters (catalogue no. SSWP04700;EMD Millipore, Billerica, MA) (Fig.4B) to detect species that could notbe recovered by baiting or culturing;filters were stored at –20 �C in tubescontaining 950 mL cetyl trimethylammonium bromide (CTAB) bufferwith polyvinyl pyrrolidone (PVP) be-fore DNA extraction.

DNA E X T R A C T I O N A N D

ILLUMINA MISEQ SEQUENCING. DNAwas extracted from filters using amodified chloroform/phenol extrac-tion method (Burke et al., 2006), andleaf baits using the Synergy 2.0 PlantDNA Extraction Kit (OPS Diagnos-tics, Lebanon, NJ). The extractedDNA was diluted with Tris-EDTAbuffer when necessary to 25 ng�mL–1

and stored at –20 �C. For pure-isolatecultures, DNA was extracted fromhyphae grown on Phytophthora-selective media plates using Extract-N-AmpPlant PCRKit (Sigma-Aldrich,St. Louis, MO). DNA was amplifiedwith oomycete-specific primers forSanger or Illumina MiSeq sequencingthat allowed us to detect Phytophthoraspecies as well as species of Phyto-pythium and Pythium.

For Sanger sequencing, the in-ternal transcribed spacer 1 (ITS1) re-gion (>900 bp) was amplified fromthe DNA using ITSDC6 (5#-GAGG-GACTTTTGGGTAATCA-3#) andITS4 (5#-TCCTCCGCTTATTGA-TATGC-3#) primers (White et al.,1990). PCR was performed withPhusion High-Fidelity PCR MasterMix with HF Buffer (New EnglandBioLabs, Ipswich, MA). The reactionwas carried out as follows: initial de-naturation at 98 �C for 30 s, 35 cyclesof denaturation at 98 �C for 10 s,annealing at 60 �C for 15 s, andextension at 72 �C for 20 s, followedby a final extension at 72 �C for 5min.PCR products were visualized ona 2.5% agarose gel to confirm positivePCR amplification. Amplified PCR

Fig. 2. Contamination hazards associated with container nursery managementpractices: (A) pots on poorly drained surface, (B) puddling of runoff water in thegrowing beds, (C) accumulation of plant leaf debris on top of weed cloth, and (D)pattern of plant mortality may reflect splashing from puddle.

Fig. 3. Contamination hazards involved in container nursery plant productionprocesses: (A) untreated recycled water, (B) potting mix with standing water, (C)reused containers, and (D) diseased plants added to cull pile.

• December 2019 29(6) 747

products were cleaned with ExoSAP-IT PCR Product Cleanup Reagent(Thermo Fisher Scientific, Waltham,MA). The cleaned PCR products werequantified, diluted to the required spec-ifications, and submitted for Sangersequencing, along with 12 pM ofeither ITS4 or ITS6 (5#-GAAGG-TGAAGTCGTAACAAGG-3#) as thesequencing primer. Sanger sequenc-ing was performed at the Centerfor Genomic Research and Biocom-puting (CGRB) at Oregon StateUniversity.

For Illumina MiSeq sequencing,the ITS1 region was amplified usingmodified ITS6 (5#-GAAGGTGAA-GTCGTAACAAGG-3#) and ITS7(5#-AGCGTTCTTCATCGATGT-GC-3#) primers. TheNexteraUniversaladapter sequence (5#-TCGTCGGCA-GCGTCAGATGTGTATAAGAGA-CAG-3#) (Nextera DNA LibraryPreparation kit, Illumina) was in-corporated at the 5# ends of bothprimers. Up to 50 ng of templateDNA was used for the PCR reaction.The PCR master mix consisted of1· reaction buffer, 800 mM dNTP,3.5mMMgCl2, 0.4 mM of each primer,

and Platinum Taq DNA polymerase(Thermo Fisher Scientific) enzyme in25 mL total PCR reaction volume. Thereaction was carried out as follows:initial denaturation at 94 �C for 2min, 35 cycles of denaturation at94 �C for 45 s, annealing at 60 �Cfor 30 s, and extension at 72 �C for1 min, followed by a final extensionat 72 �C for 10 min. PCR productswere sent to CGRB for dual-indexingwith barcodes from the Nextera IndexKit (Illumina) and library prepara-tion for high-throughput 250 to300 paired-end run Illumina MiSeqsequencing.

SEQUENCING DATA ANALYSES. Allthe sequencing data generated for thethree nurseries were analyzed sepa-rately. Sanger sequencing data weretrimmed at ends to remove noisynucleotide bases, and every sequencewas separately queried against the ITSsequences available in Phytophthora-ID (Gr€unwald et al., 2011) or theNational Center for BiotechnologyInformation (NCBI, Bethesda, MD)nucleotide database for species iden-tification using Basic Local AlignmentSearch Tool (BLAST).

MiSeq sequencing data were fil-tered to remove sequences that werelow quality (Phred quality score <30)and shorter in length (<100 bp). Theremaining high-quality, longer se-quences were queried against a cus-tom oomycete reference databaseusing a megablast search for speciesidentification (Redekar et al., 2019).TheBLAST search allowed for a singlebest matching (>99% similarity) highscoring query-subject alignment thatwas at least 150 bp long. The BLASTresults were transformed into opera-tional taxonomic units (OTU) andassembled in a table where each OTUcorresponded to a unique sequencethat was ‡99% similar to known oomy-cete species.

The samples were grouped by 10categories such as main source ofirrigation water, retention water, run-off water, compost, potting mix, usedcontainers, cull pile, sand and gravelfrom greenhouses, soil and water un-der the pots, and plant-associated.Species described within each cate-gory comprised at least 1% of the totalpopulation.

The Illumina MiSeq approachbased on the ITS1 amplicon cannotdistinguish some closely related spe-cies. Groups of indistinguishable spe-cies were classified as ‘‘clusters’’ or‘‘complexes.’’ Species within a clusterhave identical ITS1 sequences be-tween the ITS6 and ITS7 primingsites. Species within a complex haveITS1 sequences that are identicalalong the full length of the ITS1sequence (Redekar et al., 2019).

Case studies

NURSERY A (OREGON). NurseryA is a >500-acre wholesale containernursery in westernOregon that growspremium woody ornamentals and pe-rennials for retail garden centers.Most nursery beds are sloped to im-prove drainage and covered with 1 to2 inches of crushed rock, but some ofthe older hoop houses are built onless-sloped beds that have becomeclogged with plant debris and are nolonger well drained. The source ofirrigation water is a year-round creek,supplemented by well or pond waterin summer. All water is filtered anddisinfested with sodium hypochloriteor calcium chlorite before use inirrigation. Runoff water is capturedin a series of canals, reservoirs, and the

Fig. 4. Hazard analysis for Phytophthora species in container nurseries: (A) waterbeing collected from the main source of irrigation water, (B) filtration apparatusused to filter irrigation water samples, (C) rhododendron plant showing leafblight symptoms, (D) rhododendron leaf lesions tested with Phytophthora Rapidtest kit (Pocket Diagnostic; AbingdonHealth, SandHutton, UK; test was positivefor Phytophthora), (E) cleaned boxwood roots before plating on Phytophthora-selective media, (F) baiting of potting mix with two half leaves of fungicide-freerhododendron, and (G) sampling of root ball for baiting.

748 • December 2019 29(6)

WORKSHOP

pond except for occasional overflowsthat occur during heavy rain events inwinter (Fig. 5). The nursery producesits own custom potting media, someof which incorporate compost madeat the nursery from culled plant ma-terials. Compost temperatures aremonitored throughout the compost-ing process. Incoming plants that arehosts for Phytophthora are tested forthe presence of Phytophthora uponarrival.

Plants that were symptomatic ofPhytophthora diseases included firs(Abies sp.), false cypress (Chamaecy-paris sp.), pine (Pinus sp.), boxwood(Buxus sp.), andromedas (Pieris sp.),and rhododendron (Rhododendronsp.). Symptoms included damping-off, root rot, crown rot, leaf blight,dieback, and mortality. A list of plantpathogen species detected within thenursery is shown (Table 1). Someroot balls were infested with the Phy-tophthora cryptogea-complex, the Phy-tophthora citricola-complex, andthe Phytophthora citrophthora-complex;Phytophthora lateralis was isolatedfrom false cypress with crown rot.Leaf blight and dieback caused by

Phytophthora plurivora (a member ofthe P. citricola-complex) were foundon rhododendron.

The main critical control pointsin Nursery A were 1) the soil/gravelbeds, which serve as a persistent res-ervoir of inoculum for infesting sub-sequent crops; 2) certain batches offinished potting media; and 3) usedpots where the P. cryptogea-complex,the P. citricola-complex, and thePhytophthora parsiana-complex werewidespread. Although high standardsof sanitation were otherwise main-tained in the cutting room and prop-agation house, propagation traysdestined for reuse were contaminatedwith the P. cryptogea-complex and theP. citricola-complex; this occurreddespite a policy of steaming trays usedfor propagation. The P. citrophthora-complex was found in associationwith false cypress, pine, and rhodo-dendron; and in soil and water underthe pots. Several plant patho-genic species (Phytopythium litorale,Pythium dissotocum-complex, and theP. cryptogea-complex) were present inthe creek and pond water and wereenriched in the runoff relative to

saprophytic species (Fig. 5). How-ever, the nursery is doing an effectivejob of disinfesting water before ap-plying it to plants, and so irrigationwater is not a critical control point atNursery A.

The nursery should ensure thatpotting media and containers, espe-cially those used for propagation, arefree of Phytophthora species. If com-post is incorporated in potting media,steps should be taken to ensure thattemperatures achieved during com-posting are sufficient to kill patho-gens. The compost mix must beturned so that the entire mix reachesthe critical temperature, otherwisesome compost will escape treatment.Alternatively, media ingredients maybe pasteurized [65 �C for 30 min(Baker and Cook, 1974)]. In addi-tion, the nursery should considerways to prevent contact between con-tainer plants and infested ground.Nursery beds should be sloped toprevent puddling, and a 3-inch layerof crushed rock could be added toprevent direct contact between soiland containers. Plant debris shouldbe removed between crops and

Fig. 5. Diagram of the recycling systems for water, media ingredients, and containers in Nursery A (Oregon). Incoming waterfrom the main source of water is delivered to a retention reservoir. From there it is chlorinated before being applied to plants.Runoff is collected in a runoff channel and pumped back to the retention reservoir. Not all species of waterborne oomycetes areplant pathogens. Some are saprophytes. The pie charts illustrate how the relative abundance of plant pathogenic species (green)was enriched within the nursery compared with the saprophytic species (blue). Used containers are disinfested and reused forplant production. Media ingredients and potting mixes are also recycled by composting or disinfesting culled materials.

• December 2019 29(6) 749

drainage should be improved in poorlydrained areas. The nursery could so-larize nursery beds and greenhouses todisinfest the soil/gravel substrate if it is

feasible to keep these areas free ofplants during a 4-week period duringthe summer (Funahashi and Parke,2016). Nursery A tests host plants

obtained from off-site for the pres-ence of Phytophthora species; how-ever, plants should be set asidefor several weeks and tested before

Table 1. Summary of Phytophthora, Phytopythium, and Pythium species detected at critical control points in three nurseries:Nursery A (Oregon), Nursery B (California), and Nursery C (South Carolina).

(Continued on next page)

750 • December 2019 29(6)

WORKSHOP

blending plants with existing stock.Methods for testing plant material forPhytophthora species are demon-strated in two online videos (Redekaret al., 2018a, 2018b). Nondestructivemethods for testing intact root ballsfor Phytophthora are available (Ver-cauteren et al., 2013). Quarantinedplants should not be treated withoomycete-specific fungicides so thatplants that are potentially infectedmay express symptoms. To reducesporulation during the monitoringperiod, quarantined plants can besprayed with film-forming polymersor surfactants (Peterson et al., 2019).

NURSERY B (CALIFORNIA).Nurs-ery B is a large, 450-acre nursery insouthern California that producesplants for distribution through retailgarden centers and national chainstores. Water conservation is a majorconcern at this nursery, which drawsfrom a reservoir fed by the Colorado

River. Runoff water is collected insettling ponds and then pumpedthrough a rapid sand filter to a re-tention basin for recycling and treat-ment with chlorine dioxide. Plantstock includes a wide variety of her-baceous and woody plants grown incontainers either outdoors, in green-houses, or under shade. The nurseryaccepts unsold plant material, espe-cially roses (Rosa sp.) grown in 2- to5-gal containers, from national chainstores at the end of the summer, andprunes, fertilizes, and grows this ma-terial until the next growing season.

Plants infested with Phytophthoraspecies included bougainvillea (Bou-gainvillea sp.), ‘Star of Madeira’echium (Echium fastuosum), garde-nia (Gardenia veitchii), ‘HeavenlyCloud’ sage (Leucophyllum frutes-cens), mexican cardinal flower (Lobe-lia laxiflora), red yucca (Hesperaloeparviflora), hydrangea (Hydrangea

macrophylla), crape myrtle (Lager-stroemia indica), fragrant olive(Osmanthus fragrans), new zealandflax (Phormium tenax), stone pine(Pinus pinea), rose, and rosemary(Rosmarinus officinalis). Symptomsincluded chlorosis, root rot, wilting,dieback, and leaf blight. A wide di-versity of oomycete species was asso-ciated with plants including thePhytophthora cryptogea-complex, thePhytophthora nicotianae-complex,the Phytophthora tropicalis-complex,Phytophthora palmivora, Phytophthoracinnamomi, Pythium dissotocum-complex, and Phytopythium helicoides.P. helicoides causes root rot on a widevariety of greenhouse-grown crops(Beaulieu et al., 2017; Kageyamaet al., 2002; Yang et al., 2013), stemand root rot of field-grown trees(Chen et al., 2016; Fichtner et al.,2016) and could be causing subclini-cal levels of disease in nursery-grown

Table 1. (Continued) Summary of Phytophthora, Phytopythium, and Pythium species detected at critical control points in threenurseries: Nursery A (Oregon), Nursery B (California), and Nursery C (South Carolina).

zDetections were based on sequencing of the DNA extracted from baits, cultures, and filters with either Sanger sequencing or high-throughput methods.yCompost was not available to test in Nurseries B and C.

• December 2019 29(6) 751

container plants. The P. tropicalis-complex was only found in the soilmatrix in greenhouses, and no sourceof P. palmivora was detected in thenursery other than on infestedplants, suggesting that it was intro-duced into the nursery on infestedplants (Table 1, Fig. 2).

Critical control points in thisnursery were the cull pile, which wasinfested with the P. cryptogea-complex, and potting media madewith ground material from the cullpile. Recycled containers were alsoinfested with the P. nicotianae-complex. Phytophthora species weredetected in the soil/gravel beds un-der containers, and in mud on top ofthe weed cloth. The reservoir water isinfested with high levels of Pythiumflevoense, a fish pathogen, and lowlevels of two plant pathogens, Phy-tophthora cryptogea-complex and P.litorale. This untreated reservoir wa-ter is often blended with recycledrunoff water that has been disinfestedwith chlorine dioxide before reuse.Runoff water is enriched with the P.citricola-complex, the P. dissotocum-complex, and P. litorale. Evidence forplant pathogenicity of P. litorale ismixed. Phytopythium litorale is com-monly isolated from irrigation pondsin Georgia where it was shown tocause seedling damping-off and fruitrot of squash (Cucurbita pepo)(Parkunan and Ji, 2013), but isolatesfrom greenhouse water tanks inPennsylvania were not pathogenic inassays with geranium (Pelargonium·hortorum) seedlings (Choudharyet al., 2016). It is not known if P.litorale is causing disease in Nursery B.

Changes to nursery managementpractices should include preventingcontamination of potting media fromthe cull pile. The cull pile should beproperly managed to eliminate plantpathogens by composting or pasteur-ization, or it should not be includedin potting media. Recycled containersshould be steamed before reuse. Plantdebris and potting media residuesshould not be allowed to accumulateon greenhouse benches, weed cloth,or gravel nursery beds. The nurseryalso puts itself at risk by allowingreturns of unsold plant material. Thelarge container plants may harborseveral pests and diseases from theirexposure in retail centers; allowingreturns to the nursery could intro-duce these pests and diseases to their

on-site planting stock. Finally, waterfrom the reservoir should be disin-fested before blending with treatedrecycled water.

NURSERY C (SOUTH CAROLINA).Nursery C is a <100-acre facility in thecoastal lowlands of South Carolina.Specializing in container-grown pe-rennials, shrubs, and trees, Nursery Cdistributes their planting material tolandscaping professionals and retailgarden centers throughout the mid-Atlantic and southeastern states. Irri-gation water is sourced from a seriesof linked retention ponds that receiverunoff from the nursery and somesurface water and groundwater froman adjacent canal during high flowevents. Irrigation water is filtered butotherwise untreated except for chlo-rination of water used in the propa-gation greenhouses.

Symptomatic plants infested withPhytophthora and related genera in-cluded boxwood (Buxus richardii),japanese plum yew (Cephalotaxus har-ringtonia), holly (Ilex sp.), ‘SoftCaress’ mahonia (Mahonia eurybrac-teata), fragrant olive (Osmanthusfragrans), indian hawthorne (Raphio-lepis indica), and rhododendron. Incontrast to Nurseries A and B, fewplant-associated Phytophthora specieswere found in Nursery C, which issmaller and has less diversity of plantmaterial than the other two nurseries.The most widespread plant pathogensin Nursery C were Phytopythium heli-coides, found in association with plantsand also in irrigation water, andthe Phytophthora parsiana-complex.Other plant-associated species in-cluded the Phytophthora nicotianae-complex, Phytophthora cinnamomi,Phytophthora cactorum, and the P.citricola-complex. Phytophthora later-alis was found in association with‘Soft Caress’ mahonia, a surprisingfinding because P. lateralis’ hostrange is known to include only falsecypress and yew (Taxus sp.). It ispossible that P. lateralis was recentlybrought into the nursery ‘‘hitchhik-ing’’ in infested potting media orpots. P. lateralis is well adapted tosurvival in soil and dispersal in waterin cooler climates but does not appearto have established in the nursery,perhaps because the warm summertemperatures are not favorable to itsgrowth.

The main critical control point inNursery C is irrigation water that

harbors P. helicoides. All water usedfor irrigation should be disinfestedbefore application. Several methodsfor water disinfestation are available(Majsztrik et al., 2017). In addition,any plants acquired from off siteshould be grown in isolation forseveral weeks and tested for Phytoph-thora species as described previouslyto reduce the risk of introducing thesepathogens into the nursery.

DiscussionHazard analysis at the three nurs-

eries revealed that the main source ofirrigation water for each nursery isinfested with species of Phytophthora,Phytopythium, and Pythium. Many ofthese species are known plant patho-gens, although some species are sap-rophytic or aquatic opportunists(Hansen et al., 2012). Our findingsare consistent with other studies in-dicating that surface sources of water(rivers, streams, ponds) are com-monly infested with Phytophthora spe-cies (Copes et al., 2015;Hansen et al.,2012; Hong and Moorman, 2005;Hong et al., 2009, 2012; Loyd et al.,2014; Olson et al., 2013; Parke et al.,2014; Redekar et al., 2019; Simset al., 2015). All three nurseriesrecaptured runoff water; this practiceconserves water but appears to en-rich for plant pathogenic species asshown with Nursery A and B, under-scoring the importance of disinfestingrecycled water.

While our hazard analysis tar-geted critical control points for Phy-tophthora contamination, it shouldbe noted that other water molds(Phytopythium, Pythium) were alsodetected with our baiting and se-quencing methods. For those specif-ically interested in detecting Pythiumspecies, improved methods, such asdilution plating onto hymexazol-freemedia and possibly baiting witha plant leaf different from rhododen-dron, should be considered (Alcalaet al., 2016; Weiland et al., 2015).Unfortunately, there are no com-mercial diagnostic kits for detectingPythium species on baits, so growerswould need to submit water or plantsamples to a plant disease diagnosticclinic to confirm the presence ofPythium species.

Our hazard analysis is modeledafterHACCP, but we did not attemptto establish critical limits for eachcritical control point. For example,

752 • December 2019 29(6)

WORKSHOP

we only tested for the presence ofPhytophthora rather than establishinga threshold for damaging levels ofPhytophthora in water. The relation-ship between inoculum dose anddisease response is still poorly under-stood. Foliar infection of nurseryplants from irrigation water infestedwith Phytophthora has been demon-strated but mainly with ‘acute’ in-oculum levels, much higher that istypically found in nursery irrigationsystems (Benson and Jones, 1980;Tjosvold et al., 2008; Werres et al.,2007). In one study (Loyd et al.,2014), little disease developed inplants that were exposed to ‘chronic’low levels of Phytophthora inoculumapplied in infested irrigation waterover several months. The disease risklikely differs among Phytophthora spe-cies, hosts, and environmental condi-tions, so a prudent managementapproach is to disinfest irrigation wa-ter if any Phytophthora is detected.There are many options for watertreatment (Majsztrik et al., 2017;Raudales et al., 2014; Zheng, 2018).Growers can test the effectiveness oftheir water treatment by baiting anduse a diagnostic kit to determine ifPhytophthora is present on the baits.Critical limits have been developedfor steaming (Schweigkofler et al.,2014) or pasteurization of soil andmedia ingredients and of containers(Baker, 1957; Linderman and Davis,2008).

It is unlikely that a hazard anal-ysis such as ours detected all sourcesof contamination. It is impossible tosample every plant, container, orbatch of potting media. Nurseriesare inherently dynamic productionsystems; the plants themselves areconstantly moved around the nurs-ery, and new plants are coming in asothers are sold. There are a limitednumber of samples that can be pro-cessed in a timely way, and the choiceof selective media and the time of yearsampled influence the outcome. Ina 4-year study in Oregon nurseries(Parke et al., 2014), we determinedthat fall was the best time of year torecover the greatest diversity of Phy-topththora species, but the optimalsampling period has not, to ourknowledge, been determined forsouthern California or South Caro-lina, where the seasonal pattern oftemperature and rainfall are differ-ent. Although we avoided time

periods with extremely hot or coldtemperatures, it is possible that wesampled Nurseries B and C at times ofthe year that were suboptimal. More-over, any single type of bait is unlikelyto capture all species of Phytophthora.By employing both DNA-based andculture-based approaches, we wereable to overcome some of the inher-ent biases in baiting or plating, andweimproved our capacity to detect Phy-tophthora species that occur inmixed populations. Though incom-plete, hazard analysis provides a ‘snap-shot’ of contamination sources at thetime of sampling.

Once critical control pointswithin the nursery are identified,management practices can be imple-mented to reduce the risk ofeconomic loss and disease spread. Re-sources are available to help growersassess contamination hazards in theirnurseries (Griesbach et al., 2012), andan online decision tool will soon beavailable on the Clean WateR3 website(Parke et al., 2018; University of Flor-ida, 2019). Growers will be able toanswer a few questions about theirgrowing facility and then receive a dis-ease risk score to help them prioritizechanges they can make to reduce theirrisk. Although we have focused onPhytophthora species, many of the rec-ommended best management prac-tices should be effective in reducingother waterborne and soilborne pestsand diseases in the nursery. For exam-ple, growers that steam their usedcontainers to eliminate Phytophthoracontamination report greatly reducedlevels of weed seed germination. Thereduced labor costs more than paid forthe steam treatment (J.L. Parke, un-published data).

The long-term goal of the CleanWateR3 program is to encouragerecycling of runoff water. The threecase studies illustrate the need todisinfest recycled runoff water to pre-vent waterborne dissemination ofPhytophthora, Pythium, and Phyto-pythium species. Although growersshould implement a system for effec-tive water treatment, this is best ac-complished as part of an overallhazard analysis to identify and theneliminate critical control points ofPhytophthora contamination. Tar-geted changes to nursery manage-ment practices will reduce the risk ofdisease and help protect the health oflandscapes and wildlands.

Literature citedAlcala, A.V.C., T.C. Paulitz, K.L.Schroeder, L.D. Porter, M.L. Derie, andL.J. Du Toit. 2016. Pythium species as-sociated with damping-off of pea in cer-tified organic fields in the Columbia Basinof central Washington. Plant Dis. 100:916–925.

Baker, K.F. 1957. The U. C. systemfor producing healthy container-grownplants: Through the use of clean soil, cleanstock, and sanitation. Univ. California,Agr. Expt. Sta., Ext. Serv., Berkeley.

Baker, K.F. and R.J. Cook. 1974. Bi-ological control of plant pathogens.Freeman, San Francisco, CA.

Beaulieu, J., B. Ford, and Y. Balci. 2017.Genotypic diversity of Phytophthora cin-namomi and P. plurivora in Maryland’snurseries and mid-Atlantic forests. Phy-topathology 107:769–776.

Benson, D.M. and R.K. Jones. 1980.Etiology of rhododendron dieback causedby four species of Phytophthora. Plant Dis.64:687–691.

Burke, D.J., A.M. Kretzer, P.T. Rygiewicz,and M.A. Topa. 2006. Soil bacterial diver-sity in a loblolly pine plantation: Influenceof ectomycorrhizas and fertilization. FEMSMicrobiol. Ecol. 57:409–419.

Chen, X.-R., B.-B. Liu, Y.-P. Xing, B.-P.Cheng, M.-L. Liu, Y.-H. Tong, and J.-Y.Xu. 2016. Identification and characteriza-tion ofPhytopythium helicoides causing stemrot of Shatangju mandarin seedlings inChina. Eur. J. Plant Pathol. 146:715–727.

Choudhary, C.E., M.L. Burgos-Garay,G.W. Moorman, and C. Hong. 2016.Pythium and Phytopythium species in twoPennsylvania greenhouse irrigation watertanks. Plant Dis. 100:926–932.

Cobb, R. 2018. California Oak MortalityTask Force (COMTF) November 2018report. 25 Feb. 2019. <http://www.suddenoakdeath.org/wp-content/uploads/2018/11/COMTF-Report-Nov-2018.pdf>.

Copes, W.E., X. Yang, and C. Hong.2015. Phytophthora species recoveredfrom irrigation reservoirs in Mississippiand Alabama nurseries and pathogenicityof three new species. Plant Dis. 99:1390–1395.

Fichtner, E.J., G.T. Browne, M. Mortaz,L. Ferguson, and C.L. Blomquist. 2016.First report of root rot caused by Phyto-pythium helicoides on pistachio rootstockin California. Plant Dis. 100:2337.

Funahashi, F. and J.L. Parke. 2016. Ef-fects of soil solarization and Trichodermaasperellum on soilborne inoculum of

• December 2019 29(6) 753

Phytophthora ramorum and Phytophthorapini in container nurseries. Plant Dis.100:438–443.

Garbelotto, M., S. Frankel, and B. Scanu.2018. Soil- and waterborne Phytophthoraspecies linked to recent outbreaks innorthern California restoration sites. Ca-lif. Agr. 72(4):208–216.

Goss, E.M., M. Larsen, G.A. Chastagner,D.R. Givens, and N.J. Gr€unwald. 2009.Population genetic analysis infersmigrationpathways of Phytophthora ramorum in USnurseries. PLoS Pathog. 5:e1000583.

Griesbach, J., J. Parke, G. Chastagner, N.Gr€unwald, and J. Aguirre. 2012. Safeprocurement and production manual. 19Feb. 2019. <https://www.oan.org/p a g e/8 6 1 ?& h h s e a r c h t e r m s5 %22safe+and+procurement%22>.

Gr€unwald, N.K., F.N. Martin, M. Larsen,C. Sullivan, C.M. Press, M.D. Coffey,E.M. Hansen, and J.L. Parke. 2011.Phytophthora-ID.org: A sequence basedPhytophthora identification tool. Plant Dis.95:337–342.

Hansen, E.M., D.J. Goheen, E.S. Jules,and B. Ullian. 2000. Managing port-orford-cedar and the introduced patho-gen Phytophthora lateralis. Plant Dis. 84:4–14.

Hansen, E.M., P.W. Reeser, and W.Sutton. 2012. Phytophthora beyond ag-riculture. Annu. Rev. Phytopathol. 50:359–378.

Hong, C., J.D. Lea-Cox, D.S. Ross, G.W.Moorman, P.A. Richardson, S.R. Ghimire,and P. Kong. 2009. Containment basinwater quality fluctuation and implicationsfor crop health management. Irr. Sci. 27:485–496.

Hong, C.X. and G.W. Moorman. 2005.Plant pathogens in irrigation water:Challenges and opportunities. Crit. Rev.Plant Sci. 24:189–208.

Hong, C., P.A. Richardson, W. Hao, S.R.Ghimire, P. Kong, G.W. Moorman, J.D.Lea-Cox, and D.S. Ross. 2012. Phytoph-thora aquimorbida sp. nov. and Phytoph-thora taxon ‘Aquatilis’ recovered fromirrigation reservoirs and a stream inVirginia, USA. Mycologia 104:1097–1108.

Ivors, K.L. and G.W. Moorman. 2017.Oomycete plant pathogens in irrigationwater, p. 57–64. In: C. Hong, G.W.Moorman, W. Wohanka, and C. B€uttner(eds.). Biology, detection, and manage-ment of plant pathogens in irrigationwater. APS Press, St. Paul, MN.

Jones, R.K. and D.M. Benson. 2001.Diseases of woody ornamentals and treesin nurseries. APS Press, St. Paul, MN.

Jung, T., A. Perez-Sierra, A. Duran, M.Horta Jung, Y. Balci, and B. Scanu.2018. Canker and decline diseases causedby soil- and airborne Phytophthora species inforests and woodlands. Persoonia 40:182–220.

Junker, C., P. Goff, S. Wagner, and S.Werres. 2016. Occurrence of Phytoph-thora species in commercial nursery pro-duction. Plant Health Prog. 17:64–75.

Kageyama, K., T. Aoyagi, R. Sunouchi,and H. Fukui. 2002. Root rot of minia-ture roses caused by Pythium helicoides. J.Gen. Plant Pathol. 68:15–20.

Liebhold, A.M., E.G. Brockerhoff, L.J.Garrett, J.L. Parke, and K.O. Britton.2012. Live plant imports: The majorpathway for forest insect and pathogeninvasions of the US. Front. Ecol. Environ.10:135–143.

Linderman, R.G. and E.A. Davis. 2008.Eradication of Phytophthora ramorum andother pathogens from potting medium orsoil by treatment with aerated steam orfumigation with metam sodium. Hort-Technology 18:106–110.

Loyd, A.L., D.M. Benson, and K.L. Ivors.2014. Phytophthora populations in nurs-ery irrigation water in relationship topathogenicity and infection frequency ofRhododendron and Pieris. Plant Dis.98:1213–1220.

Majsztrik, J.C., R.T. Fernandez, P.R.Fisher, D.R. Hitchcock, J. Lea-Cox, J.S.Owen, Jr., L.R. Oki, and S.A. White.2017. Water use and treatment in con-tainer-grown specialty crop production: Areview. Water Air Soil Pollut. 228:151,doi: 10.1007/s11270-017-3272-1.

Ockels, F.S., M.V. Dileo, and P. Bonello.2007. Desiccation at ambient tem-perature effectively preserves plant tis-sues infected with Phytophthoras. PlantHealth Prog., doi: 10.1094/PHP-2007-0302-01-RS.

Olson,H.A., S.N. Jeffers, K.L. Ivors, K.C.Steddom, J.L. Williams-Woodward, M.T.Mmbaga, D.M. Benson, and C.X. Hong.2013. Diversity and mefenoxam sensi-tivity of Phytophthora spp. associated withthe ornamental horticulture industry inthe southeastern United States. Plant Dis.97:86–92.

Parke, J., C. Swett, and J. Majsztrik.2018. Tools for growers to assess diseaserisk. HortScience 53(9S):70 (Abstr.).

Parke, J.L. and N.J. Gr€unwald. 2012. Asystems approach for management ofpests and pathogens of nursery crops.Plant Dis. 96:1236–1244.

Parke, J.L., B.J. Knaus, V.J. Fieland, C.Lewis, and N.J. Grunwald. 2014. Phy-

tophthora community structure analyses inOregon nurseries inform systems ap-proaches to disease management. Phyto-pathology 104:1052–1062.

Parke, J.L. and C. Lewis. 2007. Root andstem infection of rhododendron frompotting medium infested with Phytoph-thora ramorum. Plant Dis. 91:1265–1270.

Parkunan, V. and P. Ji. 2013. Isolation ofPythium litorale from irrigation pondsused for vegetable production and itspathogenicity on squash. Can. J. PlantPathol. 35:415–423.

Peterson, E.K., E. Larson, and J.L. Parke.2019. Film-forming polymers and sur-factants reduce infection and sporulationof Phytophthora ramorum on rhododen-dron. PlantDis. 103:1148–1155.<https://apsjournals.apsnet.org/doi/abs/10.1094/PDIS-05-18-0802-RE>.

Raudales, R.E., J.L. Parke, C.L. Guy, andP.R. Fisher. 2014. Control of waterbornemicrobes in irrigation: A review. Agr.Water Mgt. 143:9–28.

Redekar, N.R., J.L. Eberhart, and J.L.Parke. 2019. Diversity of Phytophthora,Pythium, and Phytopythium species inrecycled irrigation water in a containernursery. Phytobiomes J. 10 Apr. 2019.<https://apsjournals.apsnet.org/doi/abs/10.1094/PBIOMES-10-18-0043-R>.

Redekar, N.R., J.L. Parke, and J. Eberhart.2018a. Baiting: A method for early de-tection of Phytophthora. 28 Jan. 2019.<http://www.youtube.com/watch?reload59&v5SJx7gzXyXoM>.

Redekar, N.R., J.L. Parke, and J.Eberhart. 2018b. How to use rapid testkits to detect Phytophthora in plant sam-ples. 28 Jan. 2019. <http://www.youtube.com/watch?v55BoqflIdNwg>.

Rooney-Latham, S., C.L. Blomquist, T.Swiecki, E. Bernhardt, and S.J. Frankel.2015. First detection in the US: Newplant pathogen, Phytophthora tentaculata,in native plant nurseries and restorationsites in California. Native Plants J.16:23–27.

Schweigkofler, W., K. Kosta, V. Huffman,S. Sharma, K. Suslow, and S. Ghosh.2014. Steaming inactivates Phytophthoraramorum, causal agent of sudden oakdeath and ramorum blight, from infestednursery soils in California. Plant HealthProg. 15:43–47.

Sims, L.L. and M. Garbelotto. 2018.Susceptibility to the rare Phytophthoratentaculata and to the widespread Phy-tophthora cactorum is consistent with hostecology and history. For. Pathol. 48:12446doi: 10.1111/efp.12446.

754 • December 2019 29(6)

WORKSHOP

Sims, L.L., W. Sutton, P. Reeser, andE.M. Hansen. 2015. The Phytophthoraspecies assemblage and diversity in ripar-ian alder ecosystems of western Oregon,USA. Mycologia 107:889–902.

Swiecki, T.J., M. Quinn, L. Sims, E.Bernhardt, L. Oliver, T. Popenuck, andM.M. Garbelotto. 2018. Three new Phy-tophthora detection methods, includingtraining dogs to sniff out the pathogen,prove reliable. Calif. Agr. 72(4):217–225.

Tjosvold, S.A., D.L. Chambers, S.T.Koike, and S.R. Mori. 2008. Disease onnursery stock as affected by environmentalfactors and seasonal inoculum levels ofPhytophthora ramorum in stream waterused for irrigation. Plant Dis. 92:1566–1573.

University of Florida. 2019. CleanWateR3—Reduce, remediate, recycle.10 Ap r . 2019 . <h t t p s : //www .cleanwater3.org>.

U.S. Department of Agriculture. 2009.Pest management strategic plan for con-tainer and field-produced nursery crops.25 Feb. 2019. <https://ipmdata.ipmcenters.org/documents/pmsps/GA-KY-NC-SC-TNnurserycropsPMSP.pdf>.

Vercauteren, A., M. Riedel, M. Maes, S.Werres, and K. Heungens. 2013. Survivalof Phytophthora ramorum in rhododen-dron root balls and in rootless substrates.Plant Pathol. 62:166–176.

Weiland, J.E., P. Garrido, Z.N. Kamvar,A.S. Espindola, S.M. Marek, N.J.Grunwald, and C.D. Garzon. 2015.Population structure of Pythium irregu-lare, P. ultimum, and P. sylvaticum inforest nursery soils of Oregon and Wash-ington. Phytopathology 105:684–694.

Werres, S., S. Wagner, T. Brand, K.Kaminski, and D. Seipp. 2007. Survival ofPhytophthora ramorum in recirculating

irrigation water and subsequent infection

of Rhododendron and Viburnum. Plant

Dis. 91:1034–1044.

White, T.J., T. Bruns, S. Lee, and J.L.

Taylor. 1990. Amplification and direct

sequencing of fungal ribosomal RNA

genes for phylogenetics, p. 315–322. In:

PCR protocols: A guide to methods and

applications. Academic Press, San Diego,

CA.

Yang, X., W. Copes, and C. Hong. 2013.

Phytophthora mississippiae sp. nov., a newspecies recovered from irrigation reser-

voirs at a plant nursery in Mississippi. J.

Plant Pathol. Microbiol. 4:180.

Zheng, Y. 2018. Greenhouse and nursery

water treatment information system. 28

Jan. 2019. <http://www.ces.uoguelph.ca/water/pathogen.shtml>.

• December 2019 29(6) 755