Pest Management Science Pest Manag Sci 63:1178–1184 (2007)

Quantification of allele E198Ain beta-tubulin conferring benzimidazoleresistance in Monilinia fructicola usingreal-time PCRYong Luo,1∗ Zhonghua Ma2 and Themis J Michailides1

1Department of Plant Pathology, University of California, Davis, Kearney Agricultural Center, 9240 South Riverbend Avenue, Parlier, CA93648, USA2Institute of Biotechnology, College of Agriculture and Biotechnology, Zhejiang University, Hangzhou 320029, China

Abstract

BACKGROUND: Thiophanate-methyl, a member of the benzimidazole class of fungicides, is used in Californiato control brown rot of stone fruit caused by Monilinia fructicola (G. Wint.) Honey. The goal of this study was todevelop a real-time polymerase chain reaction (PCR) assay as an efficient method to quantify the E198A allele ofβ-tubulin that confers benzimidazole resistance.

RESULTS: Using the real-time PCR assay, the frequency of allele E198A (FEA) in a population was determinedfrom the quantities of DNA amplified with the E198A allele-specific primer pair HRF/HRR and the M.fructicola – specific primer pair MfF6/MfR6. The average proportions of highly resistant isolates determinedwith the conventional fungicide sensitivity method were within the range of average FEA values determined withthe real-time PCR assay. We also determined the FEAs of M. fructicola populations sampled from 21 stone fruitorchards in California. Only one orchard showed a high FEA over 0.20, seven orchards had values between 0.01and 0.1, and 13 orchards had values less than 0.01.

CONCLUSION: The real-time PCR assay developed in this study provides a potentially useful tool to efficientlyquantify benzimidazole resistance for large M. fructicola populations. 2007 Society of Chemical Industry

Keywords: fungicide resistance; mummified fruit; Prunus spp.; pseudosclerotia; real-time PCR

1 INTRODUCTIONBrown rot of stone fruit, caused by the fungalpathogen Monilinia fructicola (G. Wint.) Honey, isa devastating disease in California stone fruit crops.Blossom blight and fruit rot are two major phases ofthe disease in early and late season respectively.1–3

Fungicide application at bloom and mid-season isstill an important approach to reducing the risk ofinfection of blossoms and immature fruit.4 Dependingon crop, microclimatic condition and infection level,the number of fungicide sprays per season in Californiamay range from one or two in prune (dried plum)and plum orchards to four (two at bloom and two atpreharvest) in peach and nectarine orchards. Frequentuse of fungicides can lead to the selection of resistantstrains of pathogens and the loss of efficacy of specificchemicals. Thus, alternating fungicides with differentmodes of action becomes an important strategy toreduce resistance development and retain the efficacyof fungicides.

Benzimidazole fungicides have been used to controlbrown rot for more than 30 years. Although benomylwas withdrawn from the market in 2001, growerscan still use thiophanate-methyl, which is transformedto benzimidazole, as a replacement for benomyl.Resistance of M. fructicola to benomyl was firstreported in California in 1977, 5 years after thefungicide was introduced for control of brown rot.5

Resistant isolates have been found throughout thestate of California6–9 and elsewhere where benomylwas initially used to control the disease.5,10–12 A recentstudy13 demonstrated that resistance to thiophanate-methyl was found in 75% of historic isolates collectedfrom 1992 to 1998 and in 22% of more recent isolatescollected in 2002 in California. Based on inhibitionof mycelial growth in fungicide-amended medium,benzimidazole-sensitive (S) isolates had EC50 valuesless than 2.0 µg mL−1, low-resistance (LR) isolateshad EC50 values between 2.0 and 30.0 µg mL−1 and

∗ Correspondence to: Yong Luo, Department of Plant Pathology, University of California, Davis, Kearney Agricultural Center, 9240 South Riverbend Avenue,Parlier, CA 93648, USAE-mail: yluo@uckac.edu(Received 20 September 2006; revised version received 27 February 2007; accepted 7 March 2007)Published online 2 October 2007; DOI: 10.1002/ps.1425

2007 Society of Chemical Industry. Pest Manag Sci 1526–498X/2007/$30.00

Real-time PCR to quantify benzimidazole resistance

high-resistance (HR) isolates had EC50 values greaterthan 30.0 µg mL−1 of thiophanate-methyl.

In addition to other classes of fungicides,thiophanate-methyl is still effective as an alternativefungicide to control the disease8,14 in some stone fruitorchards in California. Detection of HR isolates of M.fructicola is critical in practical management of benz-imidazole resistance in California to help determinewhether this fungicide is still effective and how longit can be used in specific stone fruit orchards in theCentral Valley of California.

Resistance to benzimidazole fungicides has beendetected in many fungal species. The traditionalmethod for assessing benzimidazole resistance in M.fructicola requires the isolation of the pathogen inpure culture and subsequent plating of single-sporeisolates onto a medium amended with the fungicide.This procedure is time consuming if large numbers ofisolates need to be tested. Even for a small numberof samples, it still takes at least 5–7 days to obtainthe results.4 Using this method, the determination offrequencies of fungicide-resistant subpopulations inlarge areas does not seem feasible.

In most cases, benzimidazole resistance has beenassociated with point mutations in the β-tubulin genethat result in altered amino acid sequences at thebenzimidazole binding site.15 Based on the pointmutation E198A in the β-tubulin gene that confers ahigh level of resistance, allele-specific PCR assays havebeen developed for rapid detection of benzimidazole-resistant isolates of M. fructicola from stone fruit.4

Recently, real-time PCR has been demonstratedas a useful tool to study fungicide resistanceof pathogen populations.16 The most successfulexamples of the use of molecular techniques to monitorthe development of resistance involve resistance tothe Qo inhibitor fungicides (QoIs). For instance,recently, both fungicide manufacturing companies andpublic research laboratories have used the G143Amarker to monitor QoI resistance in several fungalpopulations.16–18

Based on availability of molecular markers fordetecting benzimidazole resistance, the conventionalfungicide sensitivity method could be significantlyimproved by application of molecular approaches. Theobjective of this study was to develop an efficient andreliable method to quantify the allele E198A thatconfers resistance to benzimidazoles in M. fructicolapopulations from California stone fruit orchards.

2 MATERIALS AND METHODS2.1 Sample collectionTwo peach orchards not sprayed with fungicides andwith severe fruit brown rot under normal weatherconditions at the University of California KearneyAgricultural Center (KAC) in Parlier were selected.In each orchard, about 1000 mummified fruit werearbitrarily collected from different trees (severalmummified fruit per tree) 2 months after harvest. The

mummified fruit were randomly separated into fivegroups each with 10, 25, 50, 75 and 100 fruit andused as one of three replicates for each group. Foreach replicate, the spores of M. fructicola from themummified fruit were washed off with water to makea spore suspension that was adjusted to 1 × 106 sporesmL−1, and 2 mL of each sample was used for DNAextraction and analysis by using the real-time PCRassay described below.

Additionally, 21 stone fruit orchards were arbitrarilyselected, one at KAC in Parlier and the others in thesurrounding towns of Reedley, Parlier, Sanger andDinuba in Fresno County. Mummified fruit on treesshowing sporulation of M. fructicola were randomlycollected (Table 1). Spores on the surface of thesemummified fruit were washed off into 7–10 mL ofwater using a brush. For each suspension, 1.5 mL wasdistributed into each of three 2 mL centrifuge tubes.The tubes were centrifuged at 14 000 × g for 5 min.After discarding the supernatant, the spores were usedto extract genomic DNA.

Spores of an HR isolate (MS9) and an S isolate(MS22) were obtained from separate PDA cultures.Spores of MS9 and MS22 were mixed to generatesuspensions with 0, 20, 40, 60, 80 and 100% ofMS9 spores. The spore suspensions were adjustedto ca 1 × 106 spores mL−1, with two replicates foreach mixture. Each spore mixture was centrifuged at

Table 1. Locations of stone fruit orchards, the numbers of mummified

fruit bearing sporulation of Monilinia fructicola collected in each

orchard and the frequency of allele E198A (FEA)

Orchardcode Location Crop

Number ofmummified

fruitMeanFEAa SD

A1 KACb Plum 22 0.0023 0.0013D1 Dinuba Nectarine 23 0.0000 0.0000D2 Reedley Nectarine 23 0.0358 0.0143K2 KAC Peach 25 0.0143 0.0098K3 KAC Peach 23 0.0167 0.0050K4 KAC Peach 25 0.0007 0.0003K5 KAC Plum 22 0.0003 0.0002P3 Parlier Nectarine 20 0.0002 0.0002P4 Parlier Nectarine 20 0.0331 0.0119P5 Reedley Peach 26 0.0060 0.0016R1 Parlier Peach 22 0.0013 0.0003R2 Parlier Peach 28 0.0080 0.0050R7 Reedley Nectarine 26 0.0315 0.0205R8 Reedley Peach 22 0.0023 0.0008R9 Reedley Peach 22 0.0002 0.0001R10 Reedley Nectarine 23 0.0017 0.0008R11 Reedley Nectarine 20 0.0081 0.0027R12 Parlier Nectarine 25 0.0141 0.0065S1 Sanger Peach 22 0.0702 0.0537S2 Sanger Nectarine 20 0.0006 0.0000TM KAC Plum 21 0.2545 0.0458

a Three replicates from the same spore suspension were used tocalculate the mean frequency of allele E198A (FEA) for each of the 21stone fruit orchards. The real-time PCR assay was used to determineFEA values.b University of California, Kearney Agricultural Center, Parlier.

Pest Manag Sci 63:1178–1184 (2007) 1179DOI: 10.1002/ps

Y Luo, Z Ma, TJ Michailides

14 000 × g for 5 min, the supernatant was discardedand 0.5 mL of the remaining sediment was used forDNA extraction.

2.2 DNA extractionThe spore samples from each orchard or from eachmixture of HR and S isolates of M. fructicola weretransferred into a FastDNA tube containing garnetmatrix and a 1/4 inch ceramic sphere (as shipped)and used for DNA extraction using the FastDNAextraction kit (QbioGene, Irvine, CA, USA). A 0.5 mLaliquot of Cell Lysis/DNA solubilizing solution forfungi was added to each FastDNA tube. The tubeswere shaken in the FastPrep cell disruptor (QbioGene,Irvine, CA, USA) twice for 40 s at 4.5 m s−1 with 2 mincooling on ice between the shaking periods. The finalDNA of each sample was dissolved in 50 µL of water.Each extracted DNA sample was diluted tenfold toreduce PCR inhibitors, and a 2 µL aliquot of thisdilution was used as template DNA for each real-timePCR amplification as described below.

2.3 Real-time PCR amplificationBased on the sequence of the β-tubulin genefrom M. fructicola (GenBank accession num-ber AY283 677), a pair of primers MfF6 (5′-GTCGAGCCATATAACGCTACCC-3′)/MfR6 (5′-GGGCTGTATGATGACCGAGAAG-3′) was de-signed. The reverse primer MfR6 is located in intron6 of the β-tubulin gene, and the forward primer MfF6in exon 6. This primer pair was expected to generate a475 bp PCR product only from M. fructicola (includ-ing both benzimidazole-sensitive and benzimidazole-resistant isolates) but not from other fungi. A numberof isolates of M. fructicola and the closely relatedspecies M. laxa and ten isolates each of Botrytis cinereaPers. ex Fr., Botryosphaeria dothidea (Moug. ex Fr.)Ces & de Not. and Alternaria alternata Keissl. wereused to determine the specificity of this primer pair.Additionally, the DNA extracted from standard iso-lates of M. fructicola stored at KAC was also used todetermine the specificity of the MfF6/MfR6 primerpair. The previously published primer pair19 HRF (5′-TAA CAA CTG GAA GGC TTC CGG-3′)/HRR(5′-CTCGTTATCGATACAGAAGGTTG-3′) wasdesigned on the basis of the E198A allele.

Real-time PCR amplifications were performed inthe DNA Engine Opticon 2 system (MJ Research,Waltham, MA, USA) using SYBR Green I fluorescentdye. Amplifications were conducted in 25 µL volumescontaining 12.5 µL iQ SYBR Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), 2 µLtemplate DNA extracted from spores and 2 µL ofboth forward and reverse primers (4 µM each). Thefollowing parameters were used for real-time PCRamplifications: an initial preheat for 5 min at 95 ◦C,followed by 40 cycles at 94 ◦C for 15 s, 64 ◦C forboth MfF6/MfR6 and HRF/HRR primer pairs for25 s, 72 ◦C for 30 s and 73 ◦C for 1 s in order todetect and quantify the fluorescence at a temperature

above the denaturation of primer-dimers. After theamplifications were completed, melting curves wereobtained and used to confirm the signal from the targetproduct without interference from primer-dimers.The real-time PCR products were also examinedby electrophoresis on 1.5% agarose gels in TAEbuffer.

To create a standard curve for the quantitativedetection of M. fructicola (including both S and HRalleles) with M. fructicola species-specific primer pairMfF6/MfR6, tenfold serial dilutions of DNA (rangingfrom 0.1 pg to 10 ng), which was extracted from theisolate MS22, were included in each experiment. Tocreate a standard curve for detection of the E198Aallele with the HRF/HRR primer pair, tenfold serialdilutions of genomic DNA (ranging from 0.1 pg to10 ng), which was extracted from the isolate MS9,were included. Additionally, each experiment alsocontained three controls: (1) water without templateDNA; (2) 1 ng of template DNA from the isolateMS22; and (3) 1 ng of template DNA from the isolateMS9. The controls of 1 ng of MS22 DNA and 1 ng ofMS9 DNA amplified with the MfF6/MfR6 primer pairand the control of 1 ng of MS9 DNA amplified withthe HRF/HRR primer pair were used as a control of1 ng DNA by the real-time PCR detection. The controlof 1 ng DNA from isolate MS22 in the amplificationwith the HRF/HRR primer pair was used to excludethe possibility of false positive reactions.

For each sample, two real-time PCR amplificationswere conducted in the same run, one for quantificationof allele E198A using the primer pair HRF/HRR, andthe second for quantification of both sensitive andresistant alleles using the M. fructicola-specific primerpair MfF6/MfR6. The frequency of allele E198A ineach sample (FEA) was calculated by the formulaFEA = QHR/QMf , where QHR is the quantity (pg) oftemplate DNA with allele E198A determined with theHRF/HRR primer pair, and QMf is the quantity ofM. fructicola DNA determined with the MfF6/MfR6primer pair.

2.4 Comparison between the real-time PRCassay and a conventional fungicide sensitivitymethodIn order to compare the conventional methodwith the real-time PCR assay in quantifying highbenzimidazole resistance in pathogen populations,about 300 mummified fruit were collected fromeach of two orchards at KAC (orchards 1 and2). These mummified fruit were arbitrarily splitinto three replicate samples. About 67–98 single-spore isolates were obtained from each replicatesample. Conidia were picked from the surface ofeach mummified fruit with a sterilized needle andstreaked onto water agar medium. The cultureswere incubated at room temperature (23–25 ◦C)for 16 h, and germinating conidia were picked andtransferred onto fresh acidified [2.5 mL of 25%(v/v) lactic acid L−1) potato dextrose agar (APDA;

1180 Pest Manag Sci 63:1178–1184 (2007)DOI: 10.1002/ps

Real-time PCR to quantify benzimidazole resistance

Microtech Scientific, Orange, CA, USA) medium,then incubated further.

Thiophanate-methyl 700 g kg−1 WP (Topsin M 70WP; Cerexagri, Inc., King of Prussia, PA, USA) wassuspended in sterile water, adjusted to a concentrationof 10 mg AI mL−1 and added to potato dextrose agar(PDA) medium to produce final concentrations of8.0 µg mL−1. A 5 mm diameter mycelial plug was cutfrom the edge of a four-day-old colony on APDA foreach isolate of M. fructicola tested and placed in thecenter of PDA dishes amended with the fungicide. Thegrowth of each isolate was recorded after incubationat 25 ◦C for 3 days, and each isolate was scoredas benzimidazole-resistant or benzimidazole-sensitive.The frequency of benzimidazole-resistant isolates wascalculated for each replicate.

A spore suspension was prepared by washing sporesfrom the surface of all the mummified fruit in eachorchard and split into three replicates using the methoddescribed above. The DNA of each replicate wasobtained, and the real-time PCR assay was used asdescribed above to obtain the FEA value for eachreplicate.

2.5 Data analysisFor the samples of proportional mixtures of MS22and MS9, the correlation between FEA values andthe corresponding percentages of MS9 in the mixtureof suspensions containing 0, 20, 40, 60, 80 or 100%spores of MS9 were analyzed by linear regressionusing the REG procedure of SAS (SAS Institute,Cary, NC, USA). For the samples collected from21 stone fruit orchards, the mean FEA and standarddeviation for each orchard were calculated from thethree replicates. For the study of the five groupscontaining different numbers of mummified fruit, theANOVA procedure of SAS was used to determine anysignificant differences in the mean FEA among thegroups for each of the two orchards.

3 RESULTS3.1 Specificity of the primer pair MfF6/MfR6When the DNA from eight isolates each of M. fructicolaand M. laxa and ten isolates each of B. dothidea, B.cinerea and A. alternata was used to test the specificityof species-specific primer pair MfF6/MfR6, only M.fructicola isolates showed an expected 475 bp PCRproduct using 1.5% agarose gel electrophoresis. Noproduct was generated by this primer pair from otherspecies, which demonstrated that this primer pair isspecific to M. fructicola.

3.2 Standard curvesIn each real-time PCR experiment, standard curvesfor the quantification of allele E198A and M.fructicola DNA with the primer pairs HRF/HRR andMfF6/MfR6 respectively were produced. An exampleof the standard curve generated with the primer pairMfF6/MfR6 and the corresponding electrophoresis

analysis for different concentrations of template DNAfrom isolate MS22 are provided in Fig. 1. Similarly,an example of the standard curve generated withthe primer pair HRF/HRR and the correspondingelectrophoresis analysis for different concentrations oftemplate DNA from isolate MS9 are presented inFig. 2.

3.3 Relationship between FEA value and thepercentage of benzimidazole-resistant isolate inspore mixturesResults from analysis of FEA values determinedwith the real-time PCR assay for the DNA samplesextracted from the spore mixtures of the sensitive andresistant isolates of M. fructicola showed that there is asignificant linear relationship between the proportionof the benzimidazole-resistant isolate MS9 in mixturesof spore suspensions and the corresponding FEAs(Fig. 3).

3.4 Comparison between real-time PCR and aconventional fungicide sensitivity methodIn orchard 1, only one benzimidazole-resistant isolatewas found out of 230 isolates. The mean frequency ofbenzimidazole-resistant isolates detected by using theconventional method was 0.005, with an SD of 0.009(Table 2). Results using the real-time PCR assay forfive groups of mummified fruit of different size showeddifferent FEAs with changes in sample size. The mean

-2

-1

0

1

2

3

4

5

14 16 18 20 22 24 26 28 30 32 34

Ct cycle number

Lo

g10

DN

A q

uan

tity

(p

g)

y = -0.28x + 8.36, r2 = 0.999

Figure 1. Standard curve for quantification of template DNA fromMonilinia fructicola isolate MS22 (benzimidazole sensitive), amplifiedwith the species-specific primer pair MfF6/MfR6 using real-time PCR.

-2

-1

0

1

2

3

4

5

14 16 18 20 22 24 26 28 30 32 34

Ct cycle number

Lo

g10

DN

A q

uan

tity

(p

g)

y = -0.26x + 7.80, r2 = 0.996

Figure 2. Standard curve for quantification of template DNA fromMonilinia fructicola isolate MS9 (benzimidazole resistant), amplifiedwith the primer pair HRF/HRR using real-time PCR.

Pest Manag Sci 63:1178–1184 (2007) 1181DOI: 10.1002/ps

Y Luo, Z Ma, TJ Michailides

0

20

40

60

80

100

0 0.2 0.4 0.6 0.8 1

FEA

Ben

zim

idaz

ole

-res

ista

nt

spo

res

(%)

y=6.396 + (90.21*x)

r2 = 0.94, P<0.0001

Figure 3. Linear relationship between the frequency of allele E198A(FEA) in Monilinia fructicola as determined with real-time PCR and thecorresponding percentage of spores of benzimidazole-resistantisolate MS9 in mixtures of spore suspension withbenzimidazole-sensitive isolate MS22. Two replicates for eachmixture of spore suspension were used.

FEA values ranged from 0.002 to 0.016 (Fig. 4). Themean FEA value for the group containing ten mum-mified fruit was significantly less than those for thefour larger groups and underestimated the benzim-idazole resistance (FEA = 0.002) as compared withthe conventional method. The groups of 50, 75 and100 mummified fruit showed a higher level of benz-imidazole resistance (FEA = 0.007–0.016) comparedwith the conventional method and overestimated thebenzimidazole resistance.

In contrast, in orchard 2, the mean frequencyof benzimidazole-resistant isolates detected with theconventional method was 0.025, with an SD of 0.023(Table 2). Among 281 isolates, seven showed highbenzimidazole resistance. The mean values of FEAranged from 0.012 to 0.028 (Fig. 4). There wasno significant difference (P > 0.05) in mean FEA

Table 2. Average frequencies of isolates of Monilinia fructicola

showing high resistance to benzimidazole from two peach orchards

determined using a conventional fungicide sensitivity method

Replicate Isolatesa

Number ofbenzimida-

zole resistantisolates

Frequency ofbenzimidazole-

resistantisolates

Orchard 1 1 78 0 0.0002 85 0 0.0003 67 1 0.015

Mean = 0.005SD = 0.009

Orchard 2 1 92 0 0.0002 98 3 0.0313 91 4 0.044

Mean = 0.025SD = 0.023

a Single spore isolates were obtained from mummified fruit collectedfrom trees in each of the orchards.

value among the groups containing 10, 25 and 50mummified fruit. However, the mean values of FEAfor groups containing 75 and 100 mummified fruitwere significantly (P > 0.05) greater than those ofthe other groups and were different from each other(Fig. 4). Compared with the mean frequency (0.025)of benzimidazole-resistant isolates detected with theconventional method, the FEA values for the groupscontaining 10, 25 and 50 mummified fruit were allunderestimates, that from the group containing 75mummified fruit was a slight underestimate (0.022),while that from the 100 mummified fruit group wasa slight overestimate (0.028) of the benzimidazoleresistance level.

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.0450.05

10 25 50 75 100

10 25 50 75 100

Orchard 1

C

BCAB

ABA

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.0450.05

Number of mummified fruit per replicate

Orchard 2

CC

B

C

A

FE

A

Figure 4. Average frequencies of the allele E198A (FEA) detected in mummified peach fruit samples of different sample sizes for orchards 1 and 2.The LSD method was used to compare the mean values at P = 0.05.

1182 Pest Manag Sci 63:1178–1184 (2007)DOI: 10.1002/ps

Real-time PCR to quantify benzimidazole resistance

3.5 FEA values from 21 stone fruit orchardsThe mean FEA values for 21 orchards in Californiaranged from 0 to 0.25, with SD values of 0–0.0537(Table 1). Only one orchard showed a high FEA(>0.20) value, for seven orchards values were between0.01 and 0.1 and those for 13 orchards were less than0.01.

4 DISCUSSIONThis study demonstrated a potentially useful tooleffectively to quantify the frequency of allele E198Aconferring benzimidazole resistance in M. fructicolapopulations in stone fruit orchards using real-timePCR. The average FEA values obtained from the real-time PCR assay for orchards 1 and 2 were withinthe range of mean frequencies of high benzimidazoleresistance determined with the conventional method.Thus, real-time PCR has the potential to substitutefor the conventional method in quantification ofbenzimidazole resistance in pathogen populations.

It is difficult clearly to define an exact sample size(number of mummified fruit) for the real-time PCRassay that might be suitable to represent a populationin an orchard. In this study, the benzimidazoleresistance level of orchard 1 was much lower thanthat of orchard 2. In orchard 2, the mean frequency ofhighly resistant isolates obtained with the conventionalmethod was 0.025, and the mean FEA values rangedfrom 0.013 to 0.028 (Fig. 4). Compared with the meanfrequency value of benzimidazole resistance detectedwith the conventional method, the FEA values forgroups containing 10, 25 and 50 mummified fruit wereall underestimates (0.012–0.014), and that from the75 mummified fruit group was a slight underestimate(0.022), while that from the 100 mummified fruitgroup was a slight overestimate (0.028) of theresistance level. These results imply that the minimumsample size might be between 75 and 100 mummifiedfruit for the situation in this orchard. Obviously,different sample sizes might be needed for orchardsshowing different levels of fungicide resistance. Thus,it might be optimal to collect 100 mummified fruit asa normal sample size for the real-time PCR assay. Ifit is not possible to collect this number of mummifiedfruit, as was the case in most of the orchards in thisstudy (since a number of mummies naturally dropto the ground or some growers intentionally removethem from the trees), collecting as many mummifiedfruit as possible is suggested. Most importantly, thereal-time PCR assay can detect a very low frequency ofthe E198A allele that the conventional method mightnot be able to detect.

In the present real-time PCR assay, by washing andcollecting the maximum possible amount of sporesfrom the sampled mummified fruit and mixing themtogether for analysis, it is possible to obtain a very largepopulation of the pathogen’s propagules per orchard.Thus, the method may partially overcome deficienciesfrom variations caused by small sample size, which are

a problem when using traditional methods for testingfungicide sensitivity. The conclusions from the real-time PCR assay should represent a situation closer tothe real field situation than that predicted from theconventional method, especially in cases where thelevels of resistance are low.

Since the primer pair used was specific to theallele E198A, the results from the real-time PCRassay would not detect resistant isolates caused byother resistance mutations.19 In a previous study,13

the present authors distinguished three groups of M.fructicola isolates: sensitive isolates and those with lowand high levels of resistance. Whereas allele E198Ais known to be responsible for high benzimidazoleresistance in M. fructicola,20 the molecular mechanismcausing low benzimidazole resistance is unknown.Thus, the real-time PCR assay used in this study canestimate only high benzimidazole resistance caused bythe E198A mutation.

Information on the development of fungicideresistance in pathogen populations is needed toguide growers in developing fungicide resistancemanagement programs. Using the technique describedin this study, it is possible to continue collecting datafrom the same orchards for several seasons in orderto determine changes in frequency of the E198Aallele conferring high benzimidazole resistance inpathogen populations under different fungicide sprayprograms. Obviously, more orchards should also beused in further studies to confirm the usefulness ofthe real-time PCR assay in quantifying benzimidazoleresistance. Ideally, these orchards should show a largerange of frequency of highly resistant isolates, so thatdifferent levels of benzimidazole resistance determinedwith the conventional method can be correlated betterwith those estimated from the real-time PCR assaythan was possible in the present study.

In the brown rot–stone fruit pathosystem, pseu-dosclerotia (mummified fruit or mummies) bearingsporulation are excellent sources for obtaining a largepathogen population from an orchard to study fungi-cide resistance. Under California conditions, the besttime period for collecting spores from the surface ofmummified fruit is in late fall from October to Novem-ber, especially after the majority of leaves have droppedand mummified fruit on trees become readily visibleand easy to collect. Thus, both a quick sampling ofmummified fruit and the subsequent detection of theE198A allele in a large number of samples are certainlyfeasible. Most importantly, results can be obtainedlong before the beginning of the next growing season.

The cost of the real-time PCR method per analyzedsample depends on how many samples are processedsimultaneously. Generally, at least 40 orchard samples(two tubes for each sample) can be processed in 1 daywith a real-time PCR machine that can analyze 96samples simultaneously (including 12 wells for thestandard curves). The results can be available thenext day. The average cost in using the methoddescribed in this study, including PCR mixtures and

Pest Manag Sci 63:1178–1184 (2007) 1183DOI: 10.1002/ps

Y Luo, Z Ma, TJ Michailides

DNA extraction kits, is estimated to about $3.5–4.0per orchard sample using two replicates, excludingthe labor cost for the collection of samples. Currently,the method requires expensive equipment and a highlyskilled technician and is still mainly used for laboratoryresearch.

ACKNOWLEDGMENTSThis study was supported by USDA CAR Grant No.2002-51100-01990. The authors thank Heraclio CReyes for his assistance in sample collection andprocessing. They also thank Dr Guido Schnabel,Department of Entomology, Soil, and Plant Sciences,Clemson University, for his critical comments andreview of the manuscript prior to submission.

REFERENCES1 Byrde RJW and Willetts HJ, The Brown Rot Fungi of Fruit:

Their Biology and Control. Pergamon Press, Oxford–New York(1977).

2 Luo Y, Michailides TJ, Morgan DP, Krueger WH and Buch-ner RP, Inoculum dynamics, fruit infection, and developmentof brown rot in prune orchards in California. Phytopathology95:1132–1136 (2005).

3 Luo Y, Morgan DP and Michailides TJ, Risk analysis of brownrot blossom blight of prune caused by Monilinia fructicola.Phytopathology 91:759–768 (2001).

4 Michailides TJ, Morgan DP, Ma Z, Luo Y, Felt D, Doster MA,et al, Conventional and molecular assays aid diagnosis of cropdiseases and fungicide resistance. Calif Agric 59:115–123(2005).

5 Szkolnik M, Ogawa JM, Manji BT, Frate CA and Bose EA,Impact of benomyl treatments on populations of benomyl-tolerant Monilinia fructicola. (Abstr.) Phytopathol. News 12:239(1978).

6 Michailides TJ, Ogawa JM and Opgenorth DC, Shift ofMonilinia spp. and distribution of isolates sensitive and resis-tant to benomyl in California prune and apricot orchards.Plant Dis. 71:893–896 (1987).

7 Ogawa JM, Manji BT, Bostock RM, Canez VM Jr and Bose EA,Detection and characterization of benomyl-resistant Monilinialaxa on apricots. Plant Dis. 68:29–31 (1984).

8 Sonoda RM, Ogawa JM, Manji BT, Shabi E and Rough D,Factors affecting control of blossom blight in a peach orchardwith low level benomyl-resistant Monilinia fructicola. Plant Dis67:681–684 (1983).

9 Szkolnik M and Gilpatrick JD, Tolerance of Monilinia fructicolato benomyl in western New York State orchards. Plant DisRep 61:654–657 (1977).

10 Jones AL and Ehret GR, Isolation and characterization ofbenomyl-tolerant strains of Monilinia fructicola. Plant Dis Rep60:765–769 (1976).

11 Whan JH, Tolerance of Sclerotinia fructicola to benomyl. PlantDis Rep 60:200–201 (1976).

12 Zehr EI, Toler JE and Luszcz LA, Spread and persistence ofbenomyl-resistant Monilinia fructicola in South Carolina peachorchards. Plant Dis 75:590–593 (1991).

13 Yoshimura MA, Luo Y, Ma Z and Michailides TJ, Sensitivityof Monilinia fructicola from stone fruit to thiophanate-methyl,iprodione, and tebuconazole. Plant Dis 88:373–378 (2004).

14 Osorio JM, Adaskaveg JE and Ogawa JM, Inhibition of mycelialgrowth of Monilinia species and suppression and control ofbrown rot blossom blight of almond with iprodione andE-0858. Plant Dis 78:712–716 (1994).

15 Koenraadt H, Somerville SC and Jones AL, Characterization ofmutations in the beta-tubulin gene of benomyl-resistant fieldstrains of Venturia inaequalis and other plant pathogenic fungi.Phytopathology 82:1348–1354 (1992).

16 Fraaije BA, Butters JA, Coelho JM, Jones DR and Hol-lomon DW, Following the dynamics of strobilurin resistancein Blumeria graminis f. sp. tritici using quantitative allele-specific real-time PCR measurements with the fluorescentdye SYBR green I. Plant Pathol 51:45–54 (2002).

17 Ma Z and Michailides TJ, A real-time PCR assay for the detec-tion of azoxystrobin-resistant Alternaria populations frompistachio orchards in California. Crop Prot 23:1259–1263(2004).

18 Russell PE, Sensitivity Baselines in Fungicide Resistance Researchand Management. Global Crop Protection Federation,Brussels (2004).

19 Albertini C, Gredt M and Leroux P, Mutations of the ß-tubulingene associated with different phenotypes of benzimidazoleresistance in the cereal eyespot fungi Tapesia yallundae andTapesia acuformis. Pestic Biochem Physiol 64:17–31 (1999).

20 Ma Z, Yoshimura MA and Michailides TJ, Identification andcharacterization of benzimidazole resistance in Moniliniafructicola from stone fruit orchards in California. Appl EnvironMicrobiol 69:7145–7152 (2003).

1184 Pest Manag Sci 63:1178–1184 (2007)DOI: 10.1002/ps