A real‐time (TaqMan) PCR assay to differentiate Monilinia fructicola from other brown rot fungi of fruit crops

Abstract

To prevent the entry and spread of the brown rot fungus Monilinia fructicola in Europe, a fast and reliable method for detection of this organism is essential. In this study, an automated DNA extraction method combined with a multiplex real-time PCR based on TaqMan chemistry was developed for fast, convenient and reliable detection of both the EU quarantine organism Monilinia fructicola and the three other brown rot fungi M. fructigena, M. laxa and Monilia polystroma. Using the internal transcribed spacer (ITS) region of the nuclear ribosomal RNA gene repeat, a Monilinia genus-specific primer pair and two differently labelled fluorogenic probes specific for M. fructicola and the group M. fructigena/M. laxa/Monilia polystroma were developed. The analytical specificity of the assay was assessed by testing 33 isolates of the four brown rot fungi and 13 isolates of related fungal species or other fungal species that can be present on stone and pome fruit. No cross-reactions were observed. The assay was found to have a detection limit of 0·6 pg of DNA, corresponding to 27 haploid genomes or four conidia. Comparison of a manual DNA isolation followed by a conventional PCR with an automated DNA isolation combined with the presently developed real-time PCR showed that the latter method gave improved results when tested with 72 naturally infected stone fruit samples. The detection rate increased from 65 to 97%.

Full text

A real-time (TaqMan) PCR assay to differentiate Monilinia
fructicola from other brown rot fungi of fruit crops
I. R. van Brouwershaven, M. L. Bruil, G. C. M. van Leeuwen and L. F. F. Kox*
Plantenziektenkundige Dienst, P.O. Box 9102, 6700 HC Wageningen, the Netherlands
To prevent the entry and spread of the brown rot fungus Monilinia fructicola in Europe, a fast and reliable method for detec-
tion of this organism is essential. In this study, an automated DNA extraction method combined with a multiplex real-time
PCR based on TaqMan chemistry was developed for fast, convenient and reliable detection of both the EU quarantine
organism Monilinia fructicola and the three other brown rot fungi M. fructigena,M. laxa and Monilia polystroma.Using
the internal transcribed spacer (ITS) region of the nuclear ribosomal RNA gene repeat, a Monilinia genus-specific primer pair
and two differently labelled fluorogenic probes specific for M. fructicola and the group M. fructigena ⁄M. laxa ⁄Monilia
polystroma were developed. The analytical specificity of the assay was assessed by testing 33 isolates of the four brown rot
fungi and 13 isolates of related fungal species or other fungal species that can be present on stone and pome fruit. No cross-
reactions were observed. The assay was found to have a detection limit of 0Æ6 pg of DNA, corresponding to 27 haploid
genomes or four conidia. Comparison of a manual DNA isolation followed by a conventional PCR with an automated
DNA isolation combined with the presently developed real-time PCR showed that the latter method gave improved results
when tested with 72 naturally infected stone fruit samples. The detection rate increased from 65 to 97%.
Keywords: automated DNA extraction, conventional PCR, detection, duplex assay, manual DNA extraction,
TaqMan probe
Introduction
Monilinia fructicola,M. fructigena,M. laxa and the ana-
morph species Monilia polystroma cause brown rot, an
extremely destructive disease occurring on stone fruit
trees (Prunus spp.) and other rosaceous fruit trees (e.g.
Malus spp. and Pyrus spp.). The disease may seriously
reduce or even destroy a crop by affecting blossoms and
fruits, either on the tree or after harvest. Monilinia fructi-
gena and M. laxa are established in Europe, and Monilia
polystroma, a close relative of M. fructigena, is known
from Japan (van Leeuwen et al., 2002) and has been
reported to occur in Hungary (Petroc
´zy & Palkovics,
2009). Monilinia fructicola is listed in Annex IV of EU
Directive 2000 ⁄29 (EU, 2000) as an organism whose
introduction and spread within the EU member states is
prohibited. It was introduced into France in 2001 (EPPO,
2003), and is still present there (R. Ioos, LNPV, Malze
´-
ville, France, personal communication). There have been
reports on M. fructicola from Hungary and the Czech
Republic (Petroc
´zy & Palkovics, 2006; Duchoslavova
´
et al., 2007), but these findings have not been confirmed
yet. Accurate and rapid identification of Monilinia spp. is
the essential first step towards early and adequate
measures to prevent introduction and further spread of
M. fructicola within Europe.
Traditionally Monilinia spp. are differentiated based
on morphological and cultural traits, which requires up
to 10 days after initial isolation (van Leeuwen & van
Kesteren, 1998; van Leeuwen et al., 2002). These meth-
ods generally require skilled personnel with specialized
taxonomic expertise. Furthermore, visual identification
is not always unambiguous due to qualitative, partly
shared morphological characteristics, so that identifica-
tion has to be conducted under standardized conditions
and on pure cultures. Despite that, atypical isolates of
M. fructicola may be misidentified as M. laxa and vice
versa (van Leeuwen & van Kesteren, 1998). Conse-
quently, classical methods alone are not adequate for
phytosanitary diagnosis, as they lack speed and reliabil-
ity. In the last decade, several conventional PCR-based
assays were introduced for differentiation of M. fructico-
la from other Monilinia spp. (Fulton & Brown, 1997;
Snyder & Jones, 1999; Fo
¨rster & Adaskaveg, 2000;
Hughes et al., 2000; Ioos & Frey, 2000; Co
ˆte
´et al.,
2004; Gell et al., 2007). However, the PCR assay based
on a group I intron in the 18S rDNA (SSU) (Fulton &
Brown, 1997; Snyder & Jones, 1999) was not reliable, as
some isolates of M. fructicola lack this intron (Fulton et
al., 1999). The analytical sensitivity (detection limit) of
these conventional PCR assays is sufficiently low when
using pure cultures of Monilinia spp., but is too high for
*E-mail: l.f.f.kox@minlnv.nl
Published online 14 December 2009
548
ª2009 The Authors
Journal compilation ª2009 BSPP
Plant Pathology (2010) 59, 548–555 Doi: 10.1111/j.1365-3059.2009.02220.x

routine detection of M. fructicola on fruit samples,
presumably because of the presence of PCR inhibitors
in the samples. A nested PCR approach as developed by
Ma et al. (2003) increases the analytical sensitivity,
although nested PCR is prone to contamination with
amplification products, because of the high number of
molecules (>10
12
per reaction) produced in an amplifica-
tion reaction, and is therefore less suitable in a diagnostic
setting.
Compared to conventional PCR, real-time PCR, mea-
suring accumulation of PCR product using fluorescence,
generally offers increased sensitivity, and is performed in
a closed tube system, which is less prone to contamina-
tion with PCR amplicons (Hughes et al., 2006). Further-
more, because real-time PCR does not require
electrophoresis, it is less laborious than conventional
PCR, and is therefore suitable for automation and high
throughput testing. Recently a SYBR green chemistry-
based real-time PCR assay (Luo et al., 2007) has been
developed. A primary disadvantage of this type of assay is
that both specific and non-specific PCR products are
detected, unless additional analysis is done such as melt
curve analysis. Fluorogenic TaqMan probes detect only
specific PCR products and eliminate the need for post-
PCR processing to analyse the products. TaqMan probes
can be labelled with distinguishable reporter dyes, which
allow the detection of different sequences in one reaction
tube.
The objective of this study was to develop a TaqMan-
chemistry based duplex real-time PCR for differentiation
of M. fructicola from the other three brown rot fungi,
combined with an automated DNA isolation method
enabling quick and reliable diagnosis. A primer set was
developed for amplifying Monilinia spp. and two differ-
ent TaqMan probes for detection: a FAM-labelled probe
to detect M. fructicola, and a VIC-labelled probe for
detection of M. fructigena, M. laxa and Monilia polystro-
ma as a group.The probes have different reporter dyes
that can be used together in a duplex PCR. The analytical
specificity of this assay was evaluated using 33 isolates of
Monilinia spp. and Monilia polystroma, and 13 isolates
of related fungal species or other fungal species that can
be present on stone fruit. The analytical sensitivity was
assessed using DNA from M. fructicola and M. fructigena
diluted in DNA from plant extract. Finally, the developed
assay was compared with the assay routinely used in this
laboratory: manual DNA isolation and conventional
PCR, using 72 naturally infected cherries and plums
showing symptoms of brown rot.
Materials and methods
Fungal isolates
Isolates used in this study are listed in Table 1. To extract
fungal DNA, each isolate was grown on a plate with
cherry decoction agar (Gams et al., 1998). Subsequently,
emerging Monilinia spp. colonies were transferred to
potato dextrose agar (Oxoid) and incubated for 5 days at
22C.
Fruit samples
Infected fruits (25 cherries and 47 plums) with brown rot
symptoms were received as part of a survey held during
May–July 2005 to obtain pest state information of M.
fructicola in the Netherlands.
DNA preparation
Genomic DNA from pure fungal cultures was extracted
using the DNeasy Plant Kit (Qiagen) according to the
manufacturer’s instructions. The DNA was eluted in
50 lL of elution buffer. The concentration and purity of
the DNA were determined using a ND-1000 Spectropho-
tometer (NanoDrop Technologies). The DNA suspen-
sions were then diluted to a concentration of 2 ng lL
)1
.
To isolate fungal DNA from infected fruit, surface
growth consisting of mycelium and conidia was dissected
from the fruit, removing as much plant material as possi-
ble. The dissected fungal tissue was transferred to a
1Æ5 mL micro centrifuge tube with a secure fitting flattop
cap (Superlock tubes, BIOzym TC) containing 300 lLof
extraction buffer (0Æ02 M PBS, 0Æ05% Tween T25, 2%
polyvinylpyrrolidone, 0Æ2% bovine serum albumin) and
one stainless steel bead (3Æ97 mm in diameter). The tube
was then placed in a bead mill (Mixer Mill MM300,
Retsch) for 80 s at 1800 beats min
)1
. The mixture was
centrifuged for 5 s at maximum speed in a micro centri-
fuge (16 100 g) and 75 lL of the resulting supernatant,
i.e. plant extract, was used for manual or automated
DNA isolation. Manual DNA isolation was performed
using the Qiagen DNeasy Plant Kit, according to the
manufacturer’s instructions. The DNA was eluted in
50 lL of elution buffer and was further purified using
polyvinylpolypyrrolidone (PVPP, Sigma) columns. The
columns were prepared by filling Multi-Spin columns
(Axygen) with 0Æ5 cm of PVPP, and washing twice with
250 lL of DNase- and RNase-free water by centrifuging
for 5 min at 4000 g. The DNA suspension was applied to
a PVPP column and centrifuged for 5 min at 4000 g. The
flow through fraction was used as input for the PCR
assays.
Automated DNA isolation was performed with the
KingFisher 96 magnetic particle processor (Thermo Elec-
tron Corporation) using the QuickPick Plant DNA Kit
(BioNobile), following the protocol developed by the
manufacturer (K. Kontu, personal communication).
Briefly, 5 lL of proteinase K and 50 lL lysis buffer were
added to 75 lL of plant extract. After 30 min incubation
at 65C, 5 lL of MagaZorb Magnetic Particles and
125 lL of binding buffer were added. The particle-bound
DNA was washed twice with 200 lL of wash buffer and
DNA was eluted in 130 lL of elution buffer. A total geno-
mic DNA programme was used to transfer the magnetic
particles through each of the wells.
Monilinia fructicola real-time PCR 549
Plant Pathology (2010) 59, 548–555

Each series of DNA extractions included multiple con-
trols: a negative control (DNase- and RNase-free water,
one for every five samples) to monitor false positives
caused by cross-contamination during DNA isolation,
and a positive control to check efficiency of the DNA iso-
lation. The positive controls were aliquots of a batch of
extract from known M. fructigena-infected fruit tissue,
prepared in the same manner as the samples.
Table 1 Isolates used for assessment of analytical specificity of the brown fruit rot fungi real-time assay
Species Isolate code
a
Geographic origin Host or substrate
Monilinia spp. and Monilia polystroma
M. fructicola CBS 166Æ24, PD 0603202070
b,c
Unknown Prunus triflora
M. fructicola CBS 167Æ24
d
New Zealand Pr. persica
M. fructicola CBS 203Æ25 USA Malus sylvestris
M. fructicola CBS 204Æ25 USA Pr. persica
M. fructicola CBS 205Æ25 USA Pr. domestica
M. fructicola CBS 301Æ31 Australia Pr. domestica
M. fructicola CBS 329Æ35 USA M. sylvestris
M. fructicola CBS 350Æ49 USA Pr. avium
M. fructicola CBS 101511, dar 27029 Australia Pr. persica
M. fructicola LNPV 08-0445, PD 083485779 France Pr. avium
M. fructicola PD 0803485584 Unknown Pr. avium
M. fructicola PD 0803485592 Unknown Pr. persica
M. fructigena CBS 231Æ57 the Netherlands Pr. domestica
M. fructigena CBS 348Æ72 the Netherlands M. sylvestris
M. fructigena CBS 493Æ50 the Netherlands M. sylvestris
M. fructigena CBS 494Æ50 the Netherlands Pr. cerasus
M. fructigena CBS 495Æ50 the Netherlands Pyrus communis
M. fructigena CBS 578Æ77 the Netherlands Py. communis
M. fructigena CBS 101499, es-48 Spain Pr. domestica
M. fructigena CBS 101500, cc 752 Poland Pr. domestica
M. fructigena CBS 101502, PD 0603202062
b
the Netherlands M. pumila
M. fructigena PD 0703468039 Czech Republic Pr. persica
M. laxa CBS 132Æ21 UK Py. malus
M. laxa CBS 298Æ31 Ireland M. sylvestris
M. laxa CBS 299Æ31 UK Pr. domestica
M. laxa CBS 333Æ47 the Netherlands Pr. cerasus
M. laxa CBS 488Æ50 the Netherlands Pr. domestica
M. laxa Jap 2466, PD 0603202054
b
Japan Pr. mume
M. polystroma CBS 101504, Jap 2317 Japan M. pumila
M. polystroma CBS 102686, Jap 1815 Japan M. pumila
M. polystroma CBS 102687, Jap 2314 Japan M. pumila
M. polystroma CBS 102688, Jap 2316, Japan M. pumila
M. polystroma Jap 2315, PD 0603202046
b
Japan M. pumila
Other species
Alternaria mali CBS 106Æ24 USA M. sylvestris
Botrytis cinerea PD 89 ⁄1757 the Netherlands Unknown
Cladosporium spp. PD 97 ⁄4384 the Netherlands Lobelia sp.
Colletotrichum acutatum PD 79 ⁄407 the Netherlands Allium porrum
Coniothyrium fuckelii PD 95 ⁄567 the Netherlands Rosa sp.
Cylindrocarpon obtusiusculum CBS 101069, PD 98 ⁄8⁄1415 the Netherlands Ribes sp.
Penicillium expansum IPO 1324 Unknown Unknown
Pestalotia funerea PD 83 ⁄728 the Netherlands Thuja plicata
Phomopsis viticola CBS 267Æ80, PD 05 ⁄01586691 Italy Vitis vinifera
Peacilomyces variotii IPO 1063 Unknown Unknown
Sclerotinia sclerotiorum PD 93 ⁄879 the Netherlands Brassica oleracea
Stigmina carpophila PD 89 ⁄566 the Netherlands Prunus sp.
Trichoderma viride IPO 1260 Unknown Unknown
a
Isolate codes. CBS = Centraalbureau voor Schimmelcultures, Utrecht, the Netherlands; cc = Central Science Laboratory (Food and
Environment Research Agency, since 1 April 2009), York, UK; es = Department of Plant Protection, CIT-INIA, Madrid, Spain; IPO = Plant
Research International, Wageningen, the Netherlands; Jap = Faculty of Agriculture and Life Science, Hirosaki University, Japan; LNPV =
Laboratoire Nationale de la Protection des Ve
´ge
´taux, Malze
´ville, France; PD = Plant Protection Service, Wageningen, the Netherlands.
b
Isolates used for determination of analytical sensitivity (detection limit) and used as reference strains.
c
Until July 2007 preserved as M. laxa.
d
Until May 2008 preserved as M. laxa.
550 I. R. van Brouwershaven et al.
Plant Pathology (2010) 59, 548–555

Real-time PCR
For real-time (TaqMan) PCR of M. fructicola and M.
fructigena ⁄M. laxa ⁄Monilia polystroma, a genus-specific
primer pair and two minor groove-binding (MGB)
probes with 3’ non-fluorescent quencher (NFQ) were
designed with the Primer Express software (Applied Bio-
systems). Primers were obtained from Isogen Life Science,
and the MGB-NFQ probes were obtained from Applied
Biosystems. The TaqMan assays were based on sequences
in the internal transcribed spacer (ITS) region of the
nuclear multi-copy ribosomal RNA gene (Table 2). Based
on the differences between the sequence of M. fructicola
and related Monilinia species (Fig. 1), the potentially best
discriminating combination was selected: the primers
Mon139F and Mon139R, amplifying a 140 bp fragment,
were chosen together with 6-carboxyfluorescein (FAM)-
MGB probe P_fc for detection of M. fructicola, and VIC-
MGB probe P2_fgn ⁄lx ⁄ps for detection of the group con-
sisting of M. fructigena, M. laxa and Monilia polystroma
(Table 2).Five microlitres of genomic DNA were ampli-
fied in 25 lL volumes in 0Æ2 mL optical grade plates with
optical adhesive covers (Applied Biosystems). The 25 lL
reaction mixture contained: 1 ·TaqMan Universal PCR
Master Mix (Applied Biosystems), 200 nM of each of the
primers Mon139F and Mon139R, 200 nM of probe P_fc
and 200 nM of probe P2_fgn ⁄lx ⁄ps (Table 2). Real-time
PCR was performed in an ABI PRISM 7900 HT Sequence
Detection System (Applied Biosystems), using the follow-
ing conditions: 95C for 10 min, followed by 40 to 55
cycles at 95C for 15 s and 60C for 1 min. The emission
was measured at the annealing-extension step. The
threshold value was set at a fluorescence (DRn) of 0Æ1. A
cycle threshold (Ct) value below 40 was scored as a posi-
tive result. For the assessment of analytical sensitivity and
analytical specificity, 55 PCR cycles were performed to
monitor signals above the cut-off Ct of 40.
Similar to each series of DNA extractions, each series
of amplifications included controls: a negative control
(DNase- and RNase-free water) to test for contamination
with DNA as well as controls to monitor the performance
of the PCR for amplification of each of the four brown rot
fungi. The controls comprised DNA (100 pg, 10 pg and
1 pg) from the reference strains M. fructicola CBS
166Æ24, M. fructigena CBS 101502, M. laxa Jap 2466
and Monilia polystroma Jap 2315.
Conventional PCR
The conventional PCR assays were performed in 25 lL
reactions containing 1 ·PCR buffer with 1Æ5mM
MgCl
2
(Roche), 1Æ25 Units of Taq DNA polymerase
(Roche), 200 lM of each dNTP, 0Æ2lM of each primer
ITS1Mfc1 and ITS4Mfc1, ITS1Mfgn1 and ITS4Mfgn,
or ITS1Mlx1 and ITS4Mlx1 (Table 2) (Ioos & Frey,
2000), and 5 lL of template DNA. The PCR was per-
formed in a 96-well Peltier-type thermocycler (PTC-200,
MJ-Research). After amplification, 10 lL of the PCR
Table 2 Oligonucleotides used as primers and probes in this study
Oligonucleotide Sequence (5’ to 3’) Orientation Target Reference
ITS1Mfc1 TATGCTCGCCAGAGGATAATT Forward Monilinia fructicola Ioos & Frey, 2000
ITS4Mfc1 TGGGTTTTGGCAGAAGCACAC T Reverse M. fructicola Ioos &Frey, 2000
ITS1Mfgn1 CACGCTCGCCAGAGAATAACC Forward M. fructigena Ioos & Frey, 2000
ITS4Mfgn1 GGTGTTTTGCCAGAAGCACACT Reverse M. fructigena Ioos & Frey, 2000
ITS1Mlx1 TATGCTCGCCAGAGAATAATC Forward M. laxa Ioos & Frey, 2000
ITS4Mlx1 TGGGTTTTGGCAGAAGCACACC Reverse M. laxa Ioos & Frey, 2000
Mon139F CACCCTTGTGTATYATTACTTTGTTGCTT
a
Forward Monilinia spp. This study
Mon139R CAAGAGATCCGTTGTTGAAAGTTTTAA Reverse Monilinia spp. This study
P_fc FAM-TATGCTCGCCAGAGGATAATT-MGBNFQ M. fructicola This study
P2_fgn ⁄lx ⁄ps VIC-AGTTTGRTTATTCTCTGGCGA
b
-MGBNFQ M. fructigena,M. laxa,Monilia polystroma This study
a
Y=CorT
b
R=AorG
M. fructicola
(Z73777)
M. fructicola
(Z73778)
M. fructigena
(AF150677)
M. fructigena
(AF150680)
M. fructigena
(Z73779)
M. laxa
(Z73784)
M. laxa
(Z73785)
M. laxa
(Z73786)
M. polystrom
a
(AM937114)
Figure 1 Partial sequence alignment of the ITS region showing species-specific base substitutions in bold. The regions chosen for the design
of primers and the species-specific probes are underlined. The respective forward and reverse primers and the probes P_fc for Monilinia
fructicola and P2_fgn ⁄lx ⁄ps for M. fructigena ⁄M. laxa and Monilia polystroma are indicated in Table 2.
Monilinia fructicola real-time PCR 551
Plant Pathology (2010) 59, 548–555

product was electrophoresed on a 1Æ5% agarose gel
according to standard methods (Sambrook et al., 1989)
along with a 100 bp DNA ladder (GeneRuler 100 bp
DNA Ladder, Fermentas GmbH) to size fragments. The
PCR products were viewed and photographed under UV
light. Each series of amplifications contained DNA
(10 ng, 1 ng, 100 pg) from the reference strains as used
for the TaqMan assays.
Interpretation of PCR results
The results of each sample in a PCR series (conventional
or real-time PCR) were considered to be reliable if all con-
trols in the series gave the expected results. In case the
results of the controls were not as expected, the following
procedures were followed. If one of the positive amplifi-
cation controls gave a negative result, a technical failure
had occurred and for all samples with a negative result
the corresponding PCR was repeated. If the positive
DNA isolation control gave a negative result, there was a
failure in the DNA extraction procedure or inhibitors of
the PCR were present in the DNA and the PCR assay was
repeated with undiluted, 10- and 100-fold diluted DNA
extract. If one of the negative controls was positive a con-
tamination had occurred and the assay was repeated with
DNA extract from the duplicate sample. An extensive
review regarding the use of positive and negative controls
and their interpretation has been published previously
(Kox et al., 2005, 2007).
Data analysis
The data obtained from standard curves, plotting the Ct
of each reaction against the logarithmic values of DNA
concentration were analysed by linear regression. The
slope of the curves (the regression coefficient k) was used
to determine the average amplification efficiency
E=10
)1⁄k
, with E= 2 corresponding to 100% efficiency
(Rasmussen, 2001).
The diagnostic utility of each test was quantified by cal-
culating the detection rate with naturally infected fruit
samples. The detection rates of each method were
expressed as proportions and compared with Fisher’s
exact test (Kendall & Stuart, 1979). Differences in Ct val-
ues were tested using paired Student’s t-tests (Altman,
1991). Confidence intervals (CI) for means were calcu-
lated using standard methods (Altman, 1991).
Results
Selection of primers and probes for real-time PCR
Based on the alignment of sequences available from the
National Center of Biotechnology Information
(NCBI) DNA database (GenBank), generic primers for
M. fructicola, M. fructigena, M. laxa and Monilia
polystroma were selected and two specific probes, one
FAM-labelled (P_fc) to detect M. fructicola and one
VIC-labelled (P2_fgn ⁄lx ⁄ps) to detect the combination of
M. laxa, M. fructigena and Monilia polystroma (Fig. 1,
Table 2). The probes have different reporter dyes to be
used in a duplex PCR. To minimize the likelihood of non-
specific detection, the probe sequences were compared
with sequences in GenBank using the BLASTN database
search program (Altschul et al., 1997). Probes P_fc and
P2_fgn ⁄lx ⁄ps showed only sequence homologies with
M. fructicola and M. fructigena ⁄M. laxa ⁄Monilia polyst-
roma, respectively.
Performance characteristics of duplex real-time PCR
The analytical specificity of the real-time PCR assays was
tested by performing reactions using DNA (10 ng) from
M. fructicola (12 isolates), M. fructigena (10 isolates),
M. laxa (six isolates), Monilia polystroma (five isolates),
the closely related Botrytis cinerea and Sclerotinia sclero-
tiorum and 11 isolates of other fungi occurring on stone
and pome fruit (Table 1). FAM fluorescence of probe P_fc
could only be measured when the assay contained DNA of
M. fructicola, and VIC fluorescenceof probe P2_fgn ⁄lx ⁄ps
could only be measured when the assay contained DNA of
M. fructigena, M. laxa or Monilia polystroma. Amplifica-
tion of duplicate DNA samples spiked with 10 pg of M.
fructigena DNA (isolate CBS 101502) were used to verify
that negative results were not due to inhibition. All reac-
tions were positive showing that no inhibition occurred.
The analytical sensitivity was assessed by testing a dilu-
tion series starting with 10 ng of DNA from M. fructicola
(isolate CBS 166Æ24) and M. fructigena (isolate CBS
101502). To mimic the presence of fruit tissue in a myce-
lium sample, the dilutions contained 20 ng DNA from
healthy plum. The preparation of each series of dilutions
was replicated eight times. The detection limit of the
assay (mean of lowest detectable amount [0Æ4 pg] + three
standard deviations [0Æ2 pg]) was 0Æ6pg of Monilinia
spp. DNA (Table 3). Standard curves plotting the Ct of
each reaction against the logarithmic values of DNA con-
centration of M. fructicola (Fig. 2) and M. fructigena
were calculated.The assay showed a linear response from
10 ng down to 0Æ4 pg with amplification efficiencies of
1Æ96 and 1Æ93 for M. fructicola and M. fructigena, respec-
tively (Table 3).
To demonstrate that the multiplex PCR method is able
to detect M. fructicola in fruit that is co-infected with one
of the other brown rot fungi, DNA from M. fructicola
(isolate CBS 166Æ24) and M. fructigena (isolate CBS
101502) mixed in different ratios (ranging from 1:10 000
to 10 000:1) were amplified (data not shown). When both
Monilinia spp. were present in equal amounts, both tar-
gets were amplified equally well. Excess of either target
had a negative effect on the amplification of the other tar-
get, although it only abolished amplification at 1000-fold
excess.
Comparison of DNA extraction methods
To determine whether the DNA isolation method influ-
ences the results of the amplifications, the automated
552 I. R. van Brouwershaven et al.
Plant Pathology (2010) 59, 548–555

DNA extraction method on the KingFisher 96 magnetic
particle processor using the QuickPick Plant DNA kit
was compared with the manual isolation using the Qia-
gen DNeasy Plant Kit followed by PVPP purification. The
DNA isolation methods were tested on 50 fruit samples
infected with M. fructigena using the real-time PCR for
the three other brown rot species in this study.Results
obtained after isolation with both methods gave similar
results (P = 0Æ142, two-tailed paired Student’s t-test).
The mean Ct-values were 19Æ1 (95% CI 18Æ3–21Æ3) for
the manual method and 20Æ0 (95% CI 18Æ7–21Æ3) for the
automated method.
Comparison of conventional and real-time PCR
assays using infected fruit samples
Gel-based conventional PCR assays, using the primers
ITS1Mfc1 and ITS4Mfc1 for M. fructicola, ITS1Mfgn1
and ITS4Mfgn1 for M. fructigena and ITS1Mlx1 and
ITS4Mlx1 for M. laxa (Ioos & Frey, 2000) was used in a
comparison with the real-time TaqMan assay developed
in this study (Table 4). The conventional PCR assays
included a manual DNA isolation and the TaqMan assays
included automatic DNA isolation using the KingFisher
96 magnetic particle processor. Both assays, with differ-
ent but equivalent DNA isolation methods, as demon-
strated in the previous paragraph, were tested on
mycelium ⁄conidia dissected from 72 naturally infected
samples of Prunus spp. (plums and cherries). For 49
samples the results were consistent in both assays. This
results in 68% concordance between both assays. In two
of these samples no brown rot fungi could be detected.
Because it was suspected that inhibition was the cause of
these negative results, duplicate samples spiked with
M. fructigena DNA were tested (Table 4). Both reac-
tions were negative showing that the samples contained
inhibitors of the PCR. Twenty-three samples were
positive only in the real-time PCR; the difference was due
to inhibition in the conventional PCR as verified by
negative results of spiked duplicate samples. Also,
dilution of the DNA did not improve the result. The
detection rates of the conventional and real-time assays
were 65 and 97%, respectively (P<0Æ001, two-tailed
Fisher’s exact test).
Discussion
In this study an automated duplex real-time TaqMan
PCR assay has been developed and evaluated for rapid
and specific detection of the brown rot pathogens M. fruc-
ticola,M. fructigena, M. laxa and Monilia polystroma on
Table 3 Performance characteristics of the duplex real-time PCR assay for Monilinia fructicola and M. fructigena
Target DNA
a
Detection limit
b,c
(pg DNA)
Dynamic range (pg
DNA) Linear regression
b,d
From To kR
2
E
e
M. fructicola 0Æ610Æ000 0Æ43Æ42 1Æ00 1Æ96
M. fructigena 0Æ610Æ000 0Æ43Æ49 1Æ00 1Æ93
a
M. fructicola CBS 166Æ24 and M. fructigena CBS 101502 in the presence of 20 ng DNA from healthy plum.
b
Calculated from eight separately prepared series of dilutions.
c
Detection limit = mean of lowest detectable amount + three standard deviations.
d
Linear regression analysis: k= slope of linear regression between logarithmic values of DNA quantity and Ct values; R
2
= average squared
regression coefficient; E= efficiency of amplification.
e
E=10
)1⁄k
, with E= 2 corresponding to 100% efficiency (Rasmussen, 2001).
y = –3·42x + 37·07
R2 = 1·00
20
25
30
35
40
45
–1012345
log10 DNA M. fructicola (pg)
Cycle threshold (Ct) value
Figure 2 Standard curve of cycle threshold (Ct) values calculated
from amplifications of serial dilutions of DNA from Monilinia
fructicola isolate CBS 166Æ24 using the duplex real-time PCR assay.
The preparation of each series of dilutions was replicated eight
times. Ct values shown are mean values of eight reactions; error
bars represent standard deviations.
Table 4 Results of conventional PCRs and duplex real-time PCR on 72
naturally brown rot affected stone fruits
Number
of samples
Result
Conventional
PCR assay
TaqMan
PCR assay
24 Monilinia fructigena M. fructigena and ⁄or M. laxa
14 M. laxa M. fructigena and ⁄or M. laxa
9M. fructigena
and M. laxa
M. fructigena and ⁄or M. laxa
23 Inhibition
a
M. fructigena and ⁄or M. laxa
2 Inhibition
a
Inhibition
b
a
Inhibition tested by spiking a duplicate DNA sample with 1 ng of
DNA of M. fructigena CBS 101502.
b
Inhibition tested by spiking a duplicate DNA sample with 10 pg of
DNA of M. fructigena CBS 101502.
Monilinia fructicola real-time PCR 553
Plant Pathology (2010) 59, 548–555

pome and stone fruit with clear symptoms (myce-
lium ⁄conidia) of brown rot. The assay is a PCR using two
differently labelled TaqMan probes enabling differentia-
tion of the EU quarantine fungus M. fructicola from the
other three brown rot fungi. One generic specific primer
pair is used for amplification of part of the ITS region of
all four brown rot fungi. Even though the TaqMan assay
was developed for the identification of fungal structures
directly on the fruit, it can also be used to identify the cul-
tures obtained from the fruit, circumventing the need for
pure cultures for morphological ⁄cultural identification
under standardized conditions.
The analytical specificity of the assay was excellent;
moreover, it proved useful for a revised identification of
two isolates (CBS 166Æ24 and CBS 167Æ24, deposited by
E.E. Honey in 1924) that were preserved as M. laxa at the
Centraalbureau voor Schimmelcultures (CBS), Fungal
Biodiversity Centre. Both isolates gave a positive signal
for the M. fructicola probe P_fc. To verify this finding the
ITS regions were sequenced and found to be identical to
those of M. fructicola. This outcome demonstrates the
analytical specificity of the real-time TaqMan PCR assay.
In the case of a fruit co-infected with M. fructicola and
one of the other three brown rot fungi, it is possible that
M. fructicola will not be detected because of competition
of both targets for PCR reagents, primers in particular, as
this PCR assay uses generic primers. This is a common
phenomenon, as observed for other PCR assays using
generic primers (Kox et al., 2005). Experiments with mix-
tures of DNA from M. fructicola and M. fructigena show
that this competition only results in false negatives if one
of the targets is in more than 1000-fold excess of the
other. The assay should therefore be able to detect M.
fructicola in the presence of large quantities of the other
brown rot fungi.
The detection limit of the assay is 0Æ6 pg of DNA estab-
lished in the presence of purified plant DNA. With a mean
haploid genome size of 0Æ022 pg for the two Monilinia
spp. listed in the Fungal Genome Size Database (Kullman
et al., 2005; Gregory et al., 2007), 0Æ6 pg of DNA corre-
sponds to 27 fungal genomes. As a conidium of the brown
rot fungi contains on average 6Æ6 nuclei (with a range
from 4 to 10) (Hall, 1963; Hoffman, 1972), it is estimated
that the detection limit of the assay is four conidia. Com-
pared to the conventional assays designed by Ioos & Frey
(2000) that have been implemented in this laboratory
with detection limits of 10 pg of DNA (or 240 conidia),
this is a more than a 10-fold improvement of analytical
sensitivity. The higher detection limit of the conventional
PCR should not be a problem for detection when using
this assay on fruit with visible surface growth, which was
the scope of this assay, because sufficient fungal DNA
should be extracted to obtain a positive PCR result.
Therefore similar detection rates are expected for conven-
tional and real-time PCR. However, when testing 72
infected fruits with the conventional PCR, 25 (35%) gave
false-negative results, while the TaqMan assay only
resulted in two (3%) false-negatives. Similar inhibition
rates have been reported in other studies comparing con-
ventional and TaqMan assays (Baric et al., 2006; van
Gent-Pelzer et al., 2007). Spiking duplicate DNA samples
with M. fructigena DNA showed that inhibitors were
associated with the DNA isolated from the fruit surface.
Both PCRs were preceded by different DNA isolation
methods, but the study showed that there were no signifi-
cant differences in mean Ct-values between the manual
and automated DNA isolation methods used. Therefore,
the observed difference in detection rates for both assays,
65 and 97% for conventional and real-time PCR, respec-
tively, cannot be explained by difference in effectiveness
of both DNA isolation methods to remove inhibitors.
A likely cause for the lower detection limit of the TaqMan
PCR compared to the conventional PCR is that in real-
time PCR smaller amplicons are produced (139 vs
365 bp), that are known to be amplified more efficiently.
In conclusion, the real-time PCR has a lower detection
limit and is less susceptible to inhibition than conven-
tional PCR, and therefore is now the method of choice in
this laboratory.
Real-time TaqMan PCR has many advantages com-
pared to conventional PCR, but there is one disadvan-
tage: the high costs for reagents and consumables. Using
list prices valid in the Netherlands, these costs are calcu-
lated as four times higher than that of conventional
PCR. However, these material costs can be (partly) com-
pensated by the lower labour costs. Baric et al. (2006) cal-
culated that the total costs (material and labour) of real-
time PCR is 1Æ5 times higher than that of conventional
PCR. Automation of DNA extraction will reduce the
costs further.
Manual and automated DNA extraction methods have
been shown to be equally fit for amplification. The auto-
mated DNA extraction enables high-throughput testing
in the laboratory, while manual extraction is the method
of choice for on-site detection on a portable real-time
PCR platform. Real-time PCR has proven to be a suitable
method for testing samples for Phytophthora ramorum,
the causal agent of sudden oak death, at the point of
sampling (Tomlinson et al., 2005). A short-time diagno-
sis for detecting Monilinia spp. on fruits is essential and
therefore application of this new real-time PCR on a por-
table real-time platform is preferred in cases where the
time of transport of samples is unacceptably long.
Since 2006, the TaqMan assay has been used as a tool
in the annual surveys that are held to obtain information
about the pest status of M. fructicola in the Netherlands.
The samples taken for the surveys held in 2006, 2007 and
2008 were cherries and plums with visible symptoms of
brown rot. The DNA was isolated from the dissected fun-
gal tissue using the automated method. None of the sam-
ples was positive for M. fructicola. The possible
application of the assay for detection of quiescent infec-
tions or airborne spores needs further study.
Acknowledgements
The authors thank J. Hubert for supplying isolate LNPV
08-0445; K. Kontu for providing a QuickPick Plant DNA
554 I. R. van Brouwershaven et al.
Plant Pathology (2010) 59, 548–555

isolation protocol for the KingFisher 96; M.M.P J. van
Raak, C.H.M. Rosendahl-Peters and A.C.M. Tonk for
technical assistance; and J. de Gruyter and the anony-
mous reviewers for helpful suggestions.
References
Altman DG, 1991. Practical Statistics for Medical Research.
London, UK: Chapman & Hall.
Altschul SF, Madden T, Scha
¨ffer AA et al., 1997. Gapped BLAST
and PSI-BLAST: a new generation of protein database search
programs. Nucleic Acids Research 25, 3389–402.
Baric S, Kerschbamer C, Dalla Via J, 2006. TaqMan real-time PCR
versus four conventional PCR assays for detection of apple
proliferation phytoplasma. Plant Molecular Biology Reporter
24, 169–84.
Co
ˆte
´M-J, Tardif M-C, Meldrum AJ, 2004. Identification of
Monilinia fructigena,M. fructicola,M. laxa,andMonilia
polystroma on inoculated and naturally infected fruit using
multiplex PCR. Plant Disease 88, 1219–25.
Duchoslavova
´J, S
ˇiruc
ˇkova
´I, Zapletalova
´E, 2007. First report of
brown rot caused by Monilinia fructicola on various stone and
pome fruits in the Czech Republic. Plant Disease 91, 907.
EPPO, 2003. Diagnostic protocols for regulated pests. Monilinia
fructicola.PM7⁄18(1). EPPO Bulletin 33, 281–8.
EU, 2000. Council Directive 2000 ⁄29 ⁄EC of 8 May 2000 on
protective measures against the introduction into the
Community of organisms harmful to plants or plant products
andagainsttheirspreadwithintheCommunity.Official
Journal of the European Communities. Legislation Series
129, 1–112.
Fo
¨rster H, Adaskaveg JE, 2000. Early brown rot infections in sweet
cherry fruit are detected by Monilinia-specific DNA primers.
Phytopathology 90, 171–8.
Fulton CE, Brown AE, 1997. Use of SSU rDNA group-I intron to
distinguish Monilinia fructicola from M. laxa and M. fructigena.
FEMS Microbiology Letters 157, 307–12.
Fulton CE, van Leeuwen GCM, Brown AE, 1999. Genetic
variation among and within Monilinia species causing brown rot
of stone and pome fruits. European Journal of Plant
Pathology 105, 495–500.
Gams W, Hoekstra ES, Aptroot A, 1998. CBS Course of
Mycology. 4th edn. Baarn, the Netherlands: Centraalbureau
voor Schimmelcultures.
Gell I, Cubero J, Melgajero P, 2007. Two different approaches for
universal diagnosis of brown rot and identification of Monilinia
spp. in stone fruit trees. Journal of Applied Microbiology 103,
2629–37.
van Gent-Pelzer, van Brouwershaven IR, Kox LFF, Bonants PJM,
2007. A TaqMan PCR method for routine diagnosis of the
quarantine fungus Guignardia citricarpa on citrus fruit. Journal
of Phytopathology 155, 357–63.
Gregory TR, Nicol JA, Tamm H et al., 2007. Eukaryotic
genome size databases. Nucleic Acids Research 35,
D332–8.
Hall R, 1963. Cytology of the asexual stages of the Australian
brown rot fungus Monilinia fructicola (Wint.) Honey. Cytologia
28, 191–3.
Hoffman GM, 1972. Heterokaryose bei Wildsta
¨mmen von
Monilinia fructigena.Phytopathologische Zeitschrift 73,
326–40.
Hughes KJD, Fulton CE, McReynolds D, Lane CR, 2000.
Development of new PCR primers for identification of Monilinia
species. EPPO Bulletin 30, 507–11.
Hughes KJD, Griffin RL, Tomlinson JA, Boonham N, Inman AJ,
Lane CR, 2006. Development of a one step real-time PCR assay
for diagnosis of Phytophthora ramorum.Phytopathology 96,
975–81.
Ioos R, Frey P, 2000. Genomic variation within Monilinia laxa, M.
fructigena and M. fructicola, and applications to species
identification by PCR. European Journal of Plant Pathology
106, 373–8.
Kendall M, Stuart A, 1979. The Advanced Theory of Statistics,
Volume 2. 4th edn. London, UK: Griffins.
Kox LFF, Boxman ILA, Jansen CCC, Roenhorst JW, 2005.
Reliability of nucleic acid amplification techniques: modified
target RNA as exogenous internal standard for a real-time
RT-PCR for Potato spindle tuber viroid.EPPO Bulletin 35,
117–24.
Kox LFF, van Brouwershaven IR, van de Vossenberg BTLH, van
den Beld HE, Bonants PJM, de Gruyter J, 2007. Diagnostic
values and utility of immunological, morphological, and
molecular methods for in planta detection of Phytophthora
ramorum.Phytopathology 97, 1119–29.
Kullman B, Tamm H, Kullman K, 2005. Fungal Genome Size
Database. Http://www.zbi.ee/fungal-genomesize.
van Leeuwen GCM, Baayen RP, Holb IJ, Jeger MJ, 2002.
Distinction of the Asiatic brown rot fungus Monilia polystroma
sp. nov. from M. fructigena.Mycological Research 106,
444–51.
van Leeuwen GCM, van Kesteren HA, 1998. Delineation of the
three brown rot fungi of fruit crops (Monilinia spp.) on the basis
of quantitative characteristics. Canadian Journal of Botany 76,
2042–50.
Luo Y, Ma Z, Reyes HC, Morgan D, Michailides TJ, 2007.
Quantification of airborne spores of Monilinia fructicola in stone
fruit orchards of California using real-time PCR. European
Journal of Plant Pathology 118, 145–54.
Ma Z, Luo Y, Michailides TJ, 2003. Nested PCR assays for
detection of Monilinia fructicola in stone fruit orchards and
Botryosphaeria dothidea from pistachios in California. Journal
of Phytopathology 151, 312–22.
Petroc
´zy M, Palkovics L, 2006. First report of brown rot caused by
Monilinia fructicola on imported peach in Hungary. Plant
Disease 90, 375.
Petroc
´zy M, Palkovics L, 2009. First report of Monilia polystroma
on apple in Hungary. European Journal of Plant Pathology
125, 343–7.
Rasmussen R, 2001. Quantification on the LightCycler. In: Meuer
S, Wittwer C, Nakagawara K, eds. Rapid Cycle Real-time PCR,
Methods and Applications. Heidelberg, Germany: Springer Press,
21–34.
Sambrook J, Fritsch EF, Maniatis T, 1989. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor, NY, USA: Cold
Spring Harbor Laboratory Press.
Snyder CL, Jones AL, 1999. Genetic variation between strains of
Monilinia fructicola and Monilinia laxa isolated from cherries in
Michigan. Canadian Journal of Plant Pathology 21, 70–7.
Tomlinson JA, Boonham N, Hughes KJD, Griffin RL, Barker I,
2005. On-site DNA extraction and real-time PCR for detection
of Phytophthora ramorum in the field. Applied and
Environmental Microbiology 71, 6702–10.
Monilinia fructicola real-time PCR 555
Plant Pathology (2010) 59, 548–555