Quantification of Pythium populations in ginseng soils - Mount Saint ...

448applied soil ecology 40 (2008) 447–455Pythium populations in soils and plant roots have beenassessed using a variety of techniques. The simplest method isbased on the number of CFU formed after dilution plating onselective media (Jeffers and Martin, 1986). However, thismethod is labour intensive, does not allow for discriminationamong morphologically similar species, and can be hinderedby competition from other soil organisms. More sophisticatedimmunological methods, such as ELISA, have also beenemployed for Pythium quantification (Yuen et al., 1998), butthese methods are still culture dependent.More recently, real-time PCR (qPCR) has been used for thefast, accurate and culture independent quantification of avariety of pathogens (including Pythium spp.) from planttissues (Schaad and Frederick, 2002; Schena et al., 2004) andsoils (Okubara et al., 2005; Schroeder et al., 2006; Lievens et al.,2006; Kernaghan et al., 2007). Although a variety of chemistrieshave been developed for quantitative real-time PCR (Ginzinger,2002), SYBR green intercalating dyes represent the mostflexible approach to the analysis of species assemblages.Taxon specific PCR primers have been designed for a varietyof Pythium species, for use in both conventional (Kageyamaet al., 1997; Wang et al., 2003) and real-time PCR reactions(Schroeder et al., 2006). In the present study, we designed PCRprimers specific to P. ultimum Trow and P. irregulare Buismansensu stricto, the two most destructive Pythium species in southwesternOntario ginseng plantations (Reeleder and Brammall,1994). P. ultimum is fairly well circumscribed both morphologicallyand genetically (Van der Plaats-Niterink, 1981; Kageyamaet al., 2007). However, the species concept of P. irregulare iscurrently under revision (Matsumoto et al., 2000; Garzón etal.,2005, 2007) and the occurrence of morphologically similar, butgenetically divergent isolates (cryptic species) makes quantificationdifficult.Although real-time PCR quantification methods are moresensitive and specific than culture based methods (Atkinset al., 2003; Ippolito et al., 2004; Kernaghan et al., 2007), platingon selective media still gives valuable and complementarydata on soil pathogen populations. For example, relationshipsamong DNA concentration, microbial biomass and infectivitymay not be equivalent across species and pathogen DNA levelsmay not necessarily be a direct indication of inoculumpotential. The relationship between DNA concentration andinoculum potential should therefore be determined empiricallyfor each pathogen species.As the ability to accurately quantify populations of Pythiumspecies directly from soils prior to ginseng planting wouldallow for more accurate assessments of disease risk and areduction in the need for fumigation, we developed qPCRassays for P. irregulare and P. ultimum and used them in bothartificially infested soil and naturally infested ginsengcultivatedsoils. We also conducted concurrent estimates ofinoculum potential using counts of colony forming units onsemi-selective agar media.2. Materials and methods2.1. Conventional PCR, sequencing of isolates andquantification of reference DNAIsolates of target and related species from the Southern CropProtection and Food Research Centre (SCPFRC), London, ON,and the Canadian Collection of Fungal Cultures (CCFC),Ottawa, ON (Table 1) were cultured on V8 broth. DNA wasextracted using the DNeasy Plant Mini Kit (Qiagen, Mississauga,ON) and the ITS 1 and 2 regions amplified on anTable 1 – Species used in this study and the results of conventional PCR (using primers ITS1/ITS4) and testing of taxonspecific primersSpecies Accession # Source Conventional PCR and presence of restriction sites qPCRPiF/PiRPuF/PuRIsolates used for Pythium primer design and/or testingPythium heterothallicum 491 CCFC +Pythium irregulare 419 CCFC + F +P. irregulare 598 CCFC + F +P. irregulare (Group II) PID008 SCPFRC + F +Pythium paroecandrum 568 CCFC +P. irregulare (Group IV) PID018* SCPFRC +Pythium intermedium PID091* SCPFRC +Pythium splendens 435a CCFC +P. slpendens 225914 CCFC +Pythium sylvaticum 259 CCFC +Pythium ultimum 144 CCFC + B +P. ultimum 418 CCFC + B +P. ultimum PID095* SCPFRC + B +Pythium violae 400 CCFC +Isolates used for Pythium primer design onlyP. irregulare (Group II) PID003* SCPFRC + FP. irregulare (Group II) PID096* SCPFRC + FP. intermedium PID306* SCPFRC ++: Amplification of product with predicted size and melting temperature; : no amplification of target; *: isolates sequenced; F: presence of FspIrestriction site; B: presence of BccI restriction site.

applied soil ecology 40 (2008) 447–455 449Eppendorf Mastercycler (Brinkmann Instruments, Mississauga,ON) thermocycler in 50 mL reactions containing 1 UPlatinum 1 Taq DNA Polymerase, 1 PCR buffer, 200 mM dNTPmix (all from Invitrogen, Burlington, ON), 2.5 mM MgCl 2 , 0.8%BSA, 0.4 mM each of primers ITS 1 and ITS 4 (White et al., 1990)and 5 mL template DNA. Cycling parameters were 94 8C for2 min followed by 35 cycles of 94 8C for 1 min, 55 8C for 1 minand 72 8C for 2 min, followed by 72 8C for 10 min.PCR products generated from Pythium isolates from southwesternOntario ginseng soils (PID003, PID018, PID091,PID095, PID096, PID306) (Table 1) were purified with theMinElute PCR Purification Kit (Qiagen, Mississauga, ON) andsequenced on an ABI 377 Sequencer (Applied Biosystems,Foster City, CA) using BigDye 1 Terminator chemistry(Applied Biosystems); Genbank accession numbersDQ083527–DQ083532.Reference DNA was obtained from fresh mycelium of P.irregulare (PID 008) and P. ultimum (PID 095) (Table 1) byextracting with the DNeasy Plant Mini Kit (Qiagen) andquantifying with an Eppendorf BioPhotometer calibratedwith calf thymus DNA (Sigma) in Qiagen elution buffer(buffer AE).2.2. Design of taxon specific primers and identification ofspecific restriction sitesTaxon specific primer pairs targeting the internallytranscribed spacer regions of P. irregulare Groups I and II(sensu Matsumoto et al., 2000) (PiF 5 0 -GTAGCATGCGT-GTTTGCTTA-3 0 /PiR 5 0 -GCAAGCTGTGCATTCATTGC-3 0 ) andP. ultimum (PuF 5 0 -ATGATGGACTAGCTGATGAA-3 0 /PuR 5 0 -TTCCATTACACTTCATAGAA-3 0 )weredesignedonthebasisof an alignment of sequenced SCPFRC and CCFC Pythiumcultures and over 100 Pythium sequences from Genbankusing BioEdit ver. 7.0.0 (Hall, 1999). The predicted PCRproduct sizes were 508 and 407 bp, respectively. The extentof primer dimer formation was predicted using thePrimerSelect module of DNASTAR (Madison, WI). Specificityof real-time amplifications was tested on a Roche DiagnosticsLightCycler 1, using DNA extracted from a range ofrelated Pythium species, previously isolated from ginsengsoils (Table 1).In order to confirm target amplification from soil, thealignment of Pythium species used for taxon specific primerdesign was also used to identify restriction sites whichcould distinguish between the amplified DNA of our targetspecies and that of other Pythium species from naturallyinfested soils. Restriction digests of the ITS region withenzymes FspI and BccI were predicted to result in restrictiondigestion patterns characteristic of P. irregulare Groups I andII (sensu Matsumoto et al., 2000) andP. ultimum, respectively.To test for the presence of these restriction sites, digests ofPCR products generated from 17 Pythium isolates byconventional PCR (Table 1) were carried out by combining10 mL ofPCRproductwith5UofeitherFspIorBccI(NewEngland Biolabs, Pickering, Ontario), 2 mL of 10 reactionbuffer and 7 mL H 2 O. Reactions were incubated at 37 8C for3 h, separated electrophoretically on 2% agarose in TAEbuffer and analyzed with the Gel Doc EQ System (Bio-Rad,Hercules, CA).2.3. Collection and preparation of soilArtificially infested soil was prepared using Fox loamy sand(Brunosolic Gray Brown Luvisol; Typic Hapludalf; 0.9% OM)from the Agriculture and Agri-food Canada research farm atDelhi, Ontario (42847 0 N, 80838 0 W). Soil was steam pasteurizedat 74 8C for 30 min using a Lindig soil treatment system (LindigManufacturing, St. Paul, MN), air-dried, sieved through a 4 mmmesh, mixed and stored at room temperature until used. Soildilution (1 part soil in 4 parts 0.25% water agar) procedureswere carried out using P 5 ARP semi-selective media (Jeffers andMartin, 1986; Reeleder et al., 2006b) to confirm that pasteurizationhad eliminated viable propagules. P. irregulare (PID008)and P. ultimum (PID095) were grown for 2 weeks in clarified V8broth. Twenty mycelial mats were then washed with sterilewater and vacuum filtration, macerated using autoclavedmini-blenders and re-suspended in approx. 240 mL sterilewater. Macerates were mixed into pasteurized soil atapproximately 40 mL/kg soil in a large plastic bag and mixedby vigorously shaking, then allowing the mixed soil to air dry.The resulting inoculum was then further mixed withpasteurized field soil in different proportions. The resultingsoil dilutions were 100, 50, 10 and 1%. Each dilution was thendivided into six sub-samples, air dried in the greenhouse andstored at 4 8C until plating and DNA extraction.Samples of naturally infested soils (ranging from sands tosandy loams) were collected at seven sites in south-westernOntario. At each site, one 441 m 2 plot was constructed anddivided into 49.9 m 2 quadrats. Approx. 2 L of topsoil wasremoved from each quadrat. Sampling locations wererecorded by GPS. Soil samples from within each site werepooled, passively air dried at room temperature, mixed with acommercial cement mixer, sieved, and mixed again. Allequipment, including mixer and sieve were sterilized withhousehold bleach (diluted with water to 0.5% sodium hypochlorite)between soils. Soils were placed in plastic bags andstored at 4 8C until used. DNA was extracted and purified from5 g subsamples of each artificially and naturally infested soilas described in Kernaghan et al. (2007).2.4. qPCRqPCR was performed on extracts of samples of the artificiallyinfested soil using SYBR green chemistry on a RocheDiagnostics Light Cycler TM I. Twenty microlitres reactionmixtures were prepared in glass capillaries (Roche Diagnostics)and included 2 mL soil DNA extract, 9.8 mL Quanti-Tect SYBR Green PCR Master Mix (Qiagen), 0.1 U HotStarTaqDNA Polymerase (Qiagen) and 0.6 mM each of primers PiFand PiR or PuF and PuR. Reactions using PuF/PuR received1mM MgCl 2 in addition to the 2.5 mM present in theQuantiTect Master Mix.For each primer pair, PCR reactions were carried out onDNA extractions from each of the six subsamples taken fromeach soil dilution, as well as on a dilution series of referenceDNA (Quantified DNA from pure culture) and the negative(yeast) control extract. All reactions received a 15 min pretreatmentat 95 8C, then 60 cycles of 94 8C for 15 s, 64 8C for 15 s,72 8C for 25 s for the PiF/PiR primer set and 94 8C for 15 s, 54 8Cfor 30 s, 72 8C for 35 s for the PuF/PuR primer set. For all

450applied soil ecology 40 (2008) 447–455reactions, temperature transition rates were 20 8C/s and thefluorescent signal was acquired at 72 8C. After cycling, meltingcurves were generated by increasing the temperature from 65to 95 8C at a rate of 0.1 8C/s with continuous acquisition of thefluorescent signal.For the naturally infested soils, qPCR was carried out as forartificially infested soils, except that each crude DNA extract(one from each of four subsamples of each of the seven soils)was split into three parallel extraction replicates after the firstcentrifugation and prior to purification and amplification. Thisallowed for increased precision in situations with very lowlevels of target DNA.2.4.1. Post amplification characterization of qPCR productsqPCR products amplified from soil extracts were run onagarose gels and stained with ethidium bromide to confirm theproduct size. For each soil sample which successfullyamplified using either primer set, one of the amplicons fromthe three parallel extractions for that sample was restrictedwith either FspI or Bcc I (as above), to further confirm theidentity of the amplified DNA. One real-time ampliconproduced with each primer set, was also randomly selectedand sequenced.2.4.2. Analysis of qPCR dataqPCR data generated by the light cycler was analyzed usingLinRegPCR software (Ramakers et al., 2003). For each set of PCRreactions (each Light Cycler carousel) the logarithms of theinitial fluorescence (N 0 calculated by LinRegPCR) for the serialdilutions of quantified reference DNA were plotted against thelogarithms of their actual DNA concentrations. The equationof the resulting line was then used to estimate the initialconcentration of target DNA templates from the soil samples.This approach does not require (or assume) equal PCRamplification efficiencies and accounts for any variation inamplification efficiency among sets of PCR reactions(Ramakers et al., 2003).2.5. Dilution platingThe number of colony forming units per gram dry soil wasdetermined for both artificially and naturally infested soilsusing P 5 ARP soil dilution plates as described above, but usingsoil:water agar dilutions of 1:10 and 1:100. Three 10 g aliquotsof each soil were oven-dried at 104 8C for 3 days. Oven-dryweights were used to determine soil moisture; plate countdata were then adjusted to a CFU per g dry soil basis. Isolates ofPythium spp. from each field were identified morphologicallyon the basis of Middleton (1952), Waterhouse (1967) and Vander Plaats-Niterink (1981). Pearson correlations between theresulting CFU data and the qPCR data were calculated in SPSSrelease 13.0 (SPSS Inc. Chicago, IL) using non-transformeddata.2.6. Sequencing of dilution plate coloniesDue to the presence of cryptic species in the P. irregularecomplex, not all of the isolates morphologically identified asP. irregulare possessed the PiF/PiR primer sites and wouldtherefore not be detected by our assay. In order to correlateour qPCR and CFU data sets from naturally infested soils, itwas necessary to determine the proportion of colonyforming units that possessed PiF/PiR primer sites and adjustthe qPCR data for P. irregulare from naturally infested soilsaccordingly. Representative dilution plate colonies morphologicallyidentified as P. irregulare were sequenced using theprimer sets ITS1/ITS4 or ITS5/ITS4 (White et al., 1990) asdescribed above. The 83 resulting ITS sequences were thenaligned with reference sequences from Genbank usingClustal X (Thompson et al., 1997) and then manuallyadjusted using Bioedit (Ver. 5.0.6) (Hall, 1999). Maximumparsimony analysis was then performed using PAUP* version4beta 10 (Swafford, 1998) withPhytopthora cinnamomi Randsas an outgroup, TBR branch swapping and 100 bootstrapreplications.3. Results3.1. Specificity of qPCR primersqPCR using the taxon specific primer pairs PiF/PiR and PuF/PuRon P. irregulare (PID008) and P. ultimum (PID095) DNA from purecultures (at between 100 and 1 pg/mL extract) producedamplicons with melting points at 87.5 and 82.3 8C, respectively.The melting profiles of P. irregulare amplicons alsoincluded a secondary peak at 82 8C, likely due to a highly ATrich region in the center of the amplicon (see Li et al., 2003).This melting profile is also predicted by the analysis of P.irregulare ITS sequences using the Poland algorithm (Steger,1994). Post amplification gel electrophoresis also revealed thatall amplicons were of the predicted sizes. Average PCRefficiencies, calculated using LinRegPCR (Ramakers et al.,2003), were 1.51 and 1.68 (of a maximum possible value of 2.0)respectively.Testing of the two primer pairs against a range of relatedspecies indicated that they were highly specific under theamplification parameters used. Only DNA from P. irregulare(Groups I and II) and P. ultimum resulted in qPCR ampliconsusing PiF/PiR and PuF/PuR, respectively (Table 1). All DNAextracts tested produced conventional PCR products using thenon-specific primers ITS1/ITS4 (White et al., 1990), indicatingthe presence of amplifiable templates.3.2. Specificity of restriction digestsFspI and BccI digests of conventional PCR products generatedfrom 17 Pythium isolates using primers ITS1–ITS4 (Table 1)resulted in characteristic restriction fragment patterns for P.irregulare (Groups I and II) and P. ultimum, respectively. Therestriction patterns clearly distinguished the two targetamplicons from each other and from those of all other Pythiumspecies tested (Fig. 1).Although the taxon specific primers used for qPCRamplification of the target Pythium species (PiF/PiR and PuF/PuR) are internal to the ITS primers used for the conventionalPCR, one characteristic restriction site was still present in eachof the qPCR amplicons, allowing us to confirm targetamplification by digesting the qPCR products with FspI andBccI.

applied soil ecology 40 (2008) 447–455 451the same melting profiles as those from pure culture. Gelelectrophoresis of the qPCR amplicons produced with the twoprimer sets revealed that they were all of the predicted size.Restriction digests using BccI and FspI also indicated that thetarget DNA had been amplified. Restriction of the 508 bp P.irregulare (Groups I and II sensu Matsumoto et al., 2000)amplicons with FspI resulted in fragments of 450 and 50 bp,whereas restriction of the 407 bp P. ultimum amplicons withBccI resulted in fragments of 290 and 120 bp. Sequences ofrandomly selected real-time amplicons produced using thetwo primer sets also aligned well with sequences derived frompure culture.3.4. Colony forming unitsFig. 1 – Restriction digests of conventional PCR ampliconsfrom 17 Pythium isolates demonstrating the characteristicrestriction sites in P. irregulare (Groups I and II) and P.ultimum. (a) BccI digestion discriminating P. irregulare(Groups I and II). (b) FspI digestion discriminating P.ultimum. For both gels, lanes 1 and 12, 100 bp size marker;lanes 2, 5 and 17, P. ultimum; lanes 6, 10, 13, 14 and 18, P.irregulare (Groups I and II); lanes 7 and 11, P. splendens;lanes 16 and 19, P. intermedium; lane 3, P. sylvaticum; lane4, P. violae; lane 8, P. heterothallicum; lane 9, P.parocandrum; lane 15, P. irregulare (Group IV). Lanenumbers corresponding to profiles of qPCR targets areunderlined.3.3. qPCR of DNA extracted from soilsqPCR amplification from the soils artificially inoculated withthe two Pythium species resulted in amplicons of the same sizeand melting temperatures as those from pure culture. PCRefficiencies of reactions using templates derived from artificiallyinoculated soil were also very similar to those from pureculture. Average target DNA concentrations for P. irregulareDNA ranged from 13.7 pg/mL 2.42 in the soil inoculated at100% to .04 pg/mL .01 in the soil inoculated at 1% and P.ultimum ranged from 19.51 pg/mL .52 to .43 pg/mL .23.PCR efficiencies of amplifications from naturally infested soilextracts were also very similar to those from pure culture, andagain resulted in amplicons of the predicted size. However,amplification of extracts from the fourth set of subsamplesusing the PiF/PiR primer set exhibited very poor PCR efficiency,perhaps due to prolonged storage of the soil samples. Data fromthis replication were not included in the analyses.Average concentrations of P. irregulare DNA ranged from3.7 10 5 pg/mL soil extract 4.5 10 5 (SE) in soil G, to.04 pg/mL .03 in soil F. P. ultimum DNA ranged from .21 pg/mL .05 in soil C to 1.2 pg/mL .57 in soil G.3.3.1. Post-amplification characterization of qPCR productsfrom soilqPCR amplification of the two Pythium species pathogenspecies from the naturally infested soils again resulted inPlating of soils artificially inoculated with either P. irregulare orP. ultimum onto P 5 ARP media gave average CFU counts rangingfrom 42.5 3.6 propagules/100 mg dry soil in the 100%infested (non-diluted) soil to 8.7 2.1 in the most diluted soil(1% inoculum) and from 62.1 7.4 to 7.9 .55 propagules/100 mg dry soil, respectively.Plating of the naturally infested soils onto P 5 ARP mediaresulted in average P. irregulare CFU counts (based onmorphology) ranging from 23.4 3.7 (SE) propagules/100 mgdry soil in soil E, to 3.6 1.1 propagules/100 mg dry soil in soilA. P. ultimum CFU counts ranged from 14.9 3.1 in soil E to.62 .36 in soil D.3.5. Analysis of sequences derived from dilution platesThe maximum parsimony analysis of the P. irregulare ITSsequences obtained from the colonies growing on P 5 ARPdilution plates clearly divided the colonies into two groups;those which possessed the PiF/PiR primer sites and thosewhich did not (Fig. 2). Overall 64% of the P. irregulare isolatessequenced contained the qPCR primer sites and would havebeen detectable by our qPCR assay. Proportions of isolateswhich possessed the primer sites ranged from 10% in soil G to100% in soil A (Fig. 3).3.6. Comparison of qPCR and CFU measures of PythiumpopulationsIn the soils artificially infested with either P. irregulare or P.ultimum, both the DNA concentrations and CFU countsdecreased with increasing inoculum dilution. The changesin the two measures were positively correlated (Pearsoncorrelation) with r = .998, p = .011, n = 4 for P. irregulare andr = .974, p = .026, n = 4 for P. ultimum.However, the relationship between measured pathogenconcentration and soil dilution was much stronger for DNAmeasures than for CFU (Fig. 4). In the case of P. ultimum a linearregression of DNA concentration on inoculum dilution (notshown) gives r 2 = .980; p = .001, while CFU on inoculumdilution gives r 2 = .921; p = .040.Concentrations of Pythium DNA were also positivelycorrelated with CFU counts in the seven naturally infestedsoils; r = .761, p = .046, n = 7 for P. irregulare (after correction forpresence/absence of qPCR primer sites) and r = .931, p = .002,n = 7 for P. ultimum. However, the CFU data indicates that the

452applied soil ecology 40 (2008) 447–455Fig. 2 – One of the most parsimonious trees based on the ITS sequences from 83 colonies from seven ginseng soils (A–G),isolated on selective media and morphologically identified as Pythium irregulare (labeled with a PID number). Consistencyindex (CI) = 0.876, rescaled consistency index (RCI) = 0.852, retention index (RI) = 0.972. Isolates in the upper clade possessthe PiF/PiR priming sites and are therefore detectable by the real-time PCR assay. Reference sequences from Genbank andbootstrap values above 70% are included.

applied soil ecology 40 (2008) 447–455 453Fig. 3 – Percentages of P. irregulare colonies possessing PiF/PiR qPCR primer sites (black) in each of the seven naturallyinfested soils. Data based on ITS sequences of dilutionplate colonies obtained using universal primers. Thenumber of colonies sequenced per soil is given at the topof each bar.Fig. 5 – Target DNA concentrations (black) compared to CFUdata (grey) in each of the seven naturally infested soils. (a)P. irregulare DNA VS CFU/100 mg dry soil; CFU dataadjusted to reflect the proportion of isolates possessingqPCR primer sites. (b) P. ultimum DNA VS CFU/100 mg drysoil. Each bar represents an average of threemeasurements. Error bars represent the standard error ofthe mean.naturally occurring P. irregulare populations are somewhathigher than those of P. ultimum, while the average concentrationof P. irregulare DNA was much lower than that of P. ultimumDNA (Fig. 5).4. DiscussionFig. 4 – Target DNA concentrations compared to numbers ofcolony forming units in dilution series of artificiallyinfested soils. (a) Pythium irregulare DNA VS P. irregulareCFU/100 mg dry soil. (b) Pythium ultimum DNA VS P.ultimum CFU/100 mg dry soil. Error bars represent thestandard error of the mean.We have used both real-time PCR and dilution plating onselective media in order to estimate populations of P. irregulareand P. ultimum in ginseng soils in south-western Ontario. Ingeneral, there was good agreement between the data setsproduced using the two techniques. However, the directcomparison of qPCR and dilution plate data proved difficult forP. irregulare, due to the presence of cryptic species within the‘‘P. irregulare complex’’ (Garzón et al., 2007).Our qPCR primers for P. irregulare were designed to targetthe most common members of the P. irregulare complex;essentially those defined by Matsumoto et al. (2000) as ‘‘GroupI’’ and ‘‘Group II’’ and more recently by Garzón et al. (2005) as P.irregulare sensu stricto. However, sequences from P. irregulare

454applied soil ecology 40 (2008) 447–455colonies formed on P 5 ARP indicated the presence of another,morphologically similar but less common, strain of P.irregulare, which does not possess our primer sites (geneticallysimilar to ‘‘Group IV’’ of Matsumoto et al., 2000). This resultedin overestimations of colony forming units with respect to thestrains targeted by our qPCR primers. This problem wasovercome by further sequencing of isolates from the dilutionplates and performing maximum parsimony analysis (Fig. 3)todetermine the intraspecific placement of each isolate. Theproportion of isolates detectable by our qPCR assay was thencalculated for each soil, and the dilution plate data correctedaccordingly.The maximum parsimony tree clearly divides the isolatesrecovered from the dilution plates into two groups. Theupper clade, the members of which possess our qPCR primersites, is equivalent to ‘‘Group I’’ and ‘‘Group II’’ ofMatsumoto et al. (2000), whilethelowercladeisanalogousto their ‘‘Group IV’’. We did not detect any members of‘‘Group III’’ in our soils. The upper clade of Fig. 3 is alsoanalogous to the ‘‘P. irregulare sensu stricto’’ clade of Garzónet al. (2005), while the lower clade is analogous to their ‘‘P.irregulare sensu lato’’. Garzón et al. (2007) have also proposeda new species within the P. irregulare complex on the basis ofITS and cox II sequences as well as AFLP data. These isolatesform the uppermost subclade of our tree and representapproximately 6% of the dilution plate isolates sequenced.On the basis of the above phylogentic analyses, it seemslikely that other new species will soon be delineated withinthe P. irregulare complex.The proportions of the two P. irregulare types differedgreatly across the seven soils analyzed, with four beingdominated by isolates with our primer sites and threedominated by isolates without our primer sites (Fig. 4).Although there is no obvious reason for this variation, thesoils do vary somewhat in soil characteristics and croppinghistories (see Kernaghan et al., 2007).Although pathogen DNA concentrations and CFU countswere positively correlated in both the artificially and naturallyinfested soils, the range of DNA concentrations for P. ultimumin the naturally infested soils was much lower than that of P.irregulare, even though the range of CFU counts for the twospecies was similar (Fig. 5). Reasons for this discrepancy mayinclude differences in extraction efficiency, or a highergermination rate of propagules produced by P. irregularerelative to those of P. ultimum. This is supported by thedifferences in the DNA:CFU ratios for the artificially infestedsoils (Fig. 2), in which both the DNA and CFU values for P.ultimum infested soils decrease rapidly with soil dilution, whileCFU counts from P. irregulare infested soils remain fairly high,relative to DNA concentrations, over the dilution series. Also,P. irregulare grew more rapidly on the selective medium usedand may have masked the presence of P. ultimum colonies insome cases, resulting in an underestimate of P. ultimumcolonies.Another possible explanation for differences in theDNA:CFU ratios between the two pathogens may lie indifferences between the copy number of the ribosomal DNAgene clusters (which includes the targeted ITS region) pernucleus, which would result in differences among calculatedDNA concentration to biomass ratios across species (Atkinset al., 2003). In fungi, rDNA gene cluster copy numbers havebeen estimated at approximately 100 per cell in Cochliobolusheterostrophus (Drechsler) Drechsler (Tsuchiya and Taga, 2001),and found to vary up to 400% among isolates of Glomusintraradices Schenk & Smith (Corradi et al., 2006). Althoughnumbers of rDNA copies per nucleus have not been determinedfor Oomycetes, it is likely that variation in the numberof rDNA repeats per nucleus will affect the ratio betweencalculated DNA concentrations and the actual concentrationof the pathogen in the soil.The difference between the ratio of calculated DNAconcentration to the number of colony forming units in P.irregulare and P. ultimum exemplifies the need for caution whenattempting to infer disease risk directly from pathogen DNAconcentrations.Taken alone, data on soil pathogen DNA concentrations areuseful for relative comparisons across samples, but do notnecessarily reflect fungal biomass, viability or inoculumpotential. However, if a consistent relationship betweenDNA concentration and the number of colony forming unitson selective media can be established, then data on DNAconcentration can be used as a faster and more accurate proxyfor the more time consuming culture based assessments ofinoculum potential.Although the incidence of disease will also depend uponplant susceptibility (e.g. seedling age), as well as on localenvironmental conditions, estimations of soil inoculumpotentials by real-time PCR (after calibration by dilutionplating) will improve the speed and accuracy of disease riskassessment and help to reduce unnecessary pesticideapplication.AcknowledgementsFunding for this project was supplied in part by the MatchingInvestment Initiatives Program of Agriculture and Agri-FoodCanada, the Ontario Ginseng Growers Association, and theAssociated Ginseng Growers of British Columbia.B. Capell, J. Miller, G. Patriquin, M. Pelletier and Y. Zhangprovided technical assistance.referencesAgriculture and Agri-Food Canada, 2003. Ginseng Statistics.http://www.agr.gc.ca/misb/spcrops/sccs_e.php?section=stats&page=ginseng.Atkins, S.D., Clark, I.M., Sosnowska, D., Hirsch, P.R., Kerry, B.R.,2003. Detection and quantification of Plectosphaerellacucumerina, a potential biological control agent of potatocyst nematodes, by using conventional PCR, real-time PCR,selective media, and baiting. Appl. Environ. Microbiol. 69,4788–4793.Corradi, N., Croll, D., Colard, A., Kuhn, G., Ehinger, M., Sanders,I.R., 2006. Gene copy number polymorphisms in anarbuscular mycorrhizal fungal population. Appl. Environ.Microbiol. 73, 366–369.Garzón, C.D., Geiser, D.M., Moorman, G.W., 2005. Amplifiedfragment length polymorphism analysis and internaltranscribed spacer and coxII sequences reveal a species

applied soil ecology 40 (2008) 447–455 455boundary within Pythium irregulare. Phytopathology 95,1489–1498.Garzón, C.D., Yánezand, J.M., Moorman, G.W., 2007. Pythiumcryptoirregulare, a new species within the P. irregularecomplex. Mycologia 99, 291–301.Ginzinger, D.G., 2002. Gene quantification using real-timequantitative PCR: an emerging technology hits themainstream. Exp. Hematol. 30, 503–512.Hall, T.A., 1999. BioEdit: a user-friendly biological sequencealignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41, 95–98.Howard, R.J., Garland, J.A., Seaman, W.L., 1994. Diseases andPests of Vegetable Crops in Canada. CanadianPhytopathological Society and Entomological Society ofCanada, Ottawa, Ontario, 554 pp.Ippolito, A., Schena, L., Nigro, F., Ligorio, V.S., Yaseen, T., 2004.Real-time detection of Phytophthora nicotianae and P.citrophthorain citrus roots and soil. Eur. J. Plant Pathol. 110,833–843.Jeffers, S.N., Martin, S.B., 1986. Comparison of two mediaselective for Phytophthora and Pythium species. Plant Dis. 70,1038–1043.Kageyama, K., Ohyama, A., Hyakumachi, M., 1997. Detection ofPythium ultimum using polymerase chain reaction withspecies-specific primers. Plant Dis. 81, 1155–1160.Kageyama, K., Senda, M., Asano, T., Suga, H., Ishiguro, K., 2007.Intra-isolate heterogeneity of the ITS region of rDNA inPythium helicoides. Mycol. Res. 111, 416–423.Kernaghan, G., Hoke, S.M., Reeleder, R., 2007. Quantification ofCylindrocarpon destructans f.sp. panacis in soils by real-timePCR. Plant Pathol. 56, 508–516.Li, W., Xi, B., Yang, W., Hawkins, M., Schubart, U.K., 2003.Complex DNA melting profiles of small PCR productsrevealed using SYBR 1 Green I. BioTechniques 35, 702–706.Lievens, B., Brouwer, M., Vanachter, A.C.R.C., Cammue, B.P.A.,Thomma, B.P.H.J., 2006. Real-time PCR for detection andquantification of fungal and oomycete tomato pathogens inplant and soil samples. Plant Sci. 171, 155–165.Matsumoto, C., Kageyama, K., Suga, H., Hyakumachi, M., 2000.Intraspecific DNA polymorphisms of Pythium irregulare.Mycol. Res. 104, 1333–1341.Middleton, J.T., 1952. Generic concepts in the Pythiaceae.Netherlands J. Plant Pathol. 58, 226–235.Okubara, P.A., Schroeder, K.L., Paulitz, T.C., 2005. Real-timepolymerase chain reaction: applications to studies onsoilborne pathogens. Can. J. Plant Pathol. 27, 300–313.Ontario Ministry of Agriculture, Food and Rural Affairs, 2005.Publication 610, Production Recommendations for Ginseng.Queen’s Printer for Ontario, Toronto, ON, 88 pp.Predy, G.N., Goel, V., Lovlin, R., Donner, A., Stitt, L., Basu, T.K.,2005. Efficacy of an extract of North American ginsengcontaining poly-furanosyl-pyranosyl-saccharides forpreventing upper respiratory tract infections: a randomizedcontrolled trial. Can. Med. Assoc. J. 173, 1043–1048.Ramakers, C., Ruijter, J.M., Lekanne Deprez, R.H., Moorman,A.F.M., 2003. Assumption-free analysis of quantitative realtimepolymerase chain reaction (PCR) data. Neurosci. Lett.339, 62–66.Reeleder, R.D., Brammall, R.A., 1994. Pathogenicity of Pythiumspecies, Cylindrocarpon destructans, and Rhizoctonia solani toginseng seedlings in Ontario. Can. J. Plant Pathol. 16,311–316.Reeleder, R.D., Capell, B., Miller, J.J., Schooley, J., 2007.Mefenoxam sensitivity and the impact of fumigation onPythium species and Phytophthora cactorum in ginseng soils.Can. J. Plant Pathol. 29, 427–436.Reeleder, R.D., Hoke, S.M.T., Zhang, Y., 2006a. Rusted root ofginseng (Panax quinquefolius) is caused by a species ofRhexocercosporidium. Phytopathology 96, 1243–1254.Reeleder, R.D., Miller, J.J., Ball Coelho, B.R., Roy, R.C., 2006b.Impacts of tillage, cover crop, and nitrogen on populationsof earthworms, microarthropods, and soil fungi in acultivated fragile soil. Appl. Soil Ecol. 33, 243–257.Reeleder, R.D., Roy, R., Capell, B., 2002. Seed and root rots ofginseng (Panax quinquefolius) caused by Cylindrocarpondestructans and Fusarium spp. J. Ginseng Res. 26,151–158.Schaad, N.W., Frederick, R.D., 2002. Real-time PCR and itsapplication for rapid plant disease diagnostics. Can. J. PlantPathol. 24, 250–258.Schena, L., Nigro, F., Ippolito, A., Gallitelli, D., 2004. Real-timequantitative PCR: a new technology to detect and studyphytopathogenic and antagonistic fungi. Eur. J. Plant Pathol.110, 893–908.Schroeder, K.L., Okubara, P.A., Tambong, J.T., Lévesque, C.A.,Paulitz, T.C., 2006. Identification and quantification ofpathogenic Pythium spp. from soils in eastern Washingtonusing real-time polymerase chain reaction. Phytopathology96, 637–647.Steger, G., 1994. Thermal denaturation of double-strandednucleic acids: prediction of temperatures critical forgradient gel electrophoresis and polymerase chain reaction.Nucleic Acids Res. 22, 2760–2768.Swafford, D., 1998. PAUP*: Phylogenetic Analysis UsingParsimony (*and other methods), Beta Version 4.0. SinauerAssociates, Sunderland, Massachusetts.Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F.,Higgins, D.G., 1997. The Clustal X windows interface:flexible strategies for multiple sequence alignment aided byquality analysis tools. Nucleic Acids Res. 24, 4876–4882.Tsuchiya, D., Taga, M., 2001. Application of fibre-FISH(fluorescence in situ hybridization) to filamentous fungi:visualization of the rRNA gene cluster of the ascomyceteCochliobolus heterostrophus. Microbiology 147, 1183–1187.Van der Plaats-Niterink, A.J., 1981. Monograph of the GenusPythium. Stud. Mycol. 21, 1–242.Wang, P.H., Wang, Y.T., White, J.G., 2003. Species-specific PCRprimers for Pythium developed from ribosomal ITS1 region.Lett. Appl. Microbiol. 37, 127–132.Waterhouse, G.M., 1967. Key to Pythium Pringsheim, MycologicalPapers No. 109. Kew, Surrey, England: CommonwealthMycological Institute.White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification anddirect sequencing of fungal ribosomal RNA genes forphylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J.,White, T.J. (Eds.), PCR Protocols: A Guide to Methods andApplications. Academic Press, New York, pp. 315–322.Yuen, G.Y., Xia, J.Q., Sutula, C.L., 1998. A sensitive ELISA forPythium ultimum using polyclonal and species-specificmonoclonal antibodies. Plant Dis. 82, 1029–1032.