Microsatellite markers for population studies of the ascomycete Phyllosticta ampelicida, the pathogen causing grape black rot

Abstract

Grape black rot, caused by the homothallic ascomycete Phyllosticta ampelicida, is a disease originating from North America and is widespread in Europe. To investigate population structure and epidemics of this pathogen, we developed 11 microsatellite markers. A multiplex PCR assay was used to amplify genomic DNA from environmental samples including mummified berries and foliar lesions, and from fungal cultures. Environmental samples were collected from five countries (Switzerland, France, Germany, Luxembourg and the USA), and consisted of 64 different genotypes. Five additional genotypes were identified from pure cultures isolated in Switzerland and Germany. The allele rarefaction approach indicated that French vineyards in the region of Bordeaux displayed the greatest mean allelic richness, probably related to the fact that France is the country where the disease was first reported outside North America, in 1885. Our results also suggest the absence of links between the species/cultivar of Vitis hosts and the infecting P. ampelicida genotypes. This is the first report of development of microsatellite markers and their deployment for population studies of P. ampelicida.

Full text

470
ISSN (print): 0031-9465 www.fupress.com/pm
ISSN (online): 1593-2095 © Firenze University Press
Corresponding author: B. Narduzzi-Wicht
E-mail: barbarawicht@yahoo.it
Phytopathologia Mediterranea (2014) 53, 3, 470−479
DOI: 10.14601/Phytopathol_Mediterr-14481
RESEARCH PAPER
Microsatellite markers for population studies of the ascomycete
Phyllosticta ampelicida, the pathogen causing grape black rot
BarBara NarDUZZI-WICHT
1,2
, MaUro JErMINI
1
, CEsarE GEssLEr
2
and GIovaNNI aNToNIo LoDovICo BroGGINI
2,3
1
Agroscope, Institute for Plant Production Sciences, A Ramél 18, 6593 Cadenazzo, Switzerland
2
Plant Pathology, Institute of Integrative Biology, Swiss Federal Institute of Technology (ETH), Universitätstrasse 2, 8092 Zürich,
Switzerland
3
Current address: Agroscope, Institute for Plant Production Sciences, Schloss 1, 8820 Wädenswil, Switzerland
Summary. Grape black rot, caused by the homothallic ascomycete Phyllosticta ampelicida, is a disease originat-
ing from North America and is widespread in Europe. To investigate population structure and epidemics of this
pathogen, we developed 11 microsatellite markers. A multiplex PCR assay was used to amplify genomic DNA
from environmental samples including mummied berries and foliar lesions, and from fungal cultures. Environ-
mental samples were collected from ve countries (Switzerland, France, Germany, Luxembourg and the USA),
and consisted of 64 dierent genotypes. Five additional genotypes were identied from pure cultures isolated in
Switzerland and Germany. The allele rarefaction approach indicated that French vineyards in the region of Bor-
deaux displayed the greatest mean allelic richness, probably related to the fact that France is the country where the
disease was rst reported outside North America, in 1885. Our results also suggest the absence of links between
the species/cultivar of Vitis hosts and the infecting P. ampelicida genotypes. This is the rst report of development
of microsatellite markers and their deployment for population studies of P. ampelicida.
Key words: Guignardia bidwellii, SSR, population genetics, population structure.
Introduction
Grape black rot is caused by the ascomycete fun-
gus Phyllosticta ampelicida (Engelm.) Aa (anamorph
name; see Wikee et al., 2013; teleomorph: Guignar-
dia bidwellii), that originated in North America.
The pathogen was introduced into Europe in the
early 1800s (Reddick, 1911) and is currently wide-
spread in several European wine growing areas in
France (Ramsdell and Milholland, 1988), Germany
(Maixner and Holz, 2003), Northern Italy, Southern
Switzerland (Pezet and Jermini, 1989) and Lux-
embourg (Anonymous, 2008), especially where
summer precipitation is high. If not properly man-
aged, grape black rot can lead to substantial yield
losses and is therefore considered among the most
economically destructive diseases in viticulture
(Harms et al., 2005). Primary hosts of P. ampelicida
include most varieties of bunch grapes (Vitis vinife-
ra, Vitis labrusca and Vitis arizonica) and muscadine
grape (Vitis rotundifolia), while secondary hosts in-
clude ornamental Vitaceae such as Ampelopsis sp.,
Cissus sp. and Parthenocissus sp. (van der Aa, 1973).
Luttrell (1946) proposed the teleomorph G. bidwellii
to consist of three distinct pathovars (var. euvitis,
parthenocissi and muscadinii), and recent molecu-
lar studies conrmed at least the dierentiation in
two distinct clades of isolates from Vitis and from
Parthenocissus (Wicht et al., 2012). Identication of
the pathogen, which is usually based only on black
rot symptoms and mycelium macromorphology
(Kong, 2009), is therefore not suitable to reveal this
dierentiation, which should be conrmed by mo-
lecular analyses.

471
Vol. 53, No. 3, December, 2014
Microsatellite markers for Phyllostica ampelicida
Table 1. Cultures of Phyllosticta analyzed in this study. CH, Switzerland; D, Germany. *Lyophilized mycelium.
ID Species Year Country Vineyard Host
9390 P. ampelicida 2010 D Geilweilerhof V. vinifera
9449 P. ampelicida 2010 D Diedesfeld V. vinifera
9450 P. ampelicida 2010 D Deidesheim V. vinifera
9494 P. ampelicida 2010 D Mühlheim-Andel V. vinifera
9495 P. ampelicida 2010 D – Vitis sp.
170 P. ampelicida 1996 CH St. Prex Vitis sp.
1003 P. ampelicida 2003 CH – Vitis sp.
559* P. ampelicida 1981 CH Belmont V. labrusca
264 Phyllosticta sp. 1996 CH – Parthenocissus sp.
1002 Phyllosticta sp. 2003 CH – Parthenocissus sp.
The life cycle of P. ampelicida has been described
by several authors (e.g. Wilcox, 2003), but not all as-
pects of the disease have been elucidated (Ullrich
et al., 2009). The fungus overwinters in mummied
berries within host plants and on the ground, but
also in canes or spurs. Spring rains trigger the release
of ascospores from mummied berries, and conidia
from these and canes within the vine trellises. The
contributions of sexual and asexual spores in the ini-
tiation and spread of black rot have not been clearly
quantied. Moreover, the homothallic reproductive
nature of the fungus does not allow distinguishing
conidial infection from ascosporic infection by as-
cospores resulting from self-fertilization (Jailloux,
1992). Both spore types are able to infect green tis-
sues, especially young leaves. Later in the growing
season, they can infect berries and mummify them.
Despite the economic importance of grape black rot,
very few studies have been carried out to investigate
the genetic structure of P. ampelicida populations.
Such information may become crucial in estimating
the evolutionary potential of the pathogen while de-
veloping new black rot control strategies.
Microsatellite markers are commonly used for
population studies of grape pathogens (Gobbin et al.,
2005, 2006; Matasci et al., 2010; Frenkel et al., 2012), as
well as for many other organisms. This is because the
microsatellites have high length polymorphism, and
the possibility of automated high-throughput geno-
typing using a capillary sequencer can be used for
their indentication. However, their development
involves laborious processes of isolation, starting
from puried genomic material of the targeted spe-
cies (Zane et al., 2002). Such an approach is currently
relieved by cheaper next-generation sequencing (Za-
lapa et al., 2012).
The aim of the present study was to develop
simple sequence repeat (SSR) markers suitable for
genotyping P. ampelicida isolates, to investigate the
genetic variability of pathogen populations from dif-
ferent European wine growing areas, by comparing
their polymorphism.
Materials and methods
Fungal isolates and environmental samples
A subset of monohyphal cultures of P. ampelicida
used by Wicht et al. (2012), provided by the fungal
public collections of Mycoscope (Agroscope, Swit-
zerland) and the DLR Rheinpfalz (Germany), were
grown on PDA at 20°C. The isolates are listed in Ta-
ble 1. DNA was extracted using the QIAamp DNA
Mini Kit (QIAGEN) according to the manufacturer’s
instructions. Lyophilized mycelium of isolate 559
was used, as no living culture of this isolate was
available.
Environmental samples (n = 1354) consisted of
berries (n = 662) and leaves (n = 692) aected by
grape black rot, collected from 17 European vine-
yards located in Switzerland, France, Germany and

Phytopathologia Mediterranea
472
B. Narduzzi-Wicht et al.
Table 2. Environmental samples investigated in this study. CH: Switzerland; F: France; D: Germany; LUX: Luxembourg
and USA: United States of America.
Year Country Vineyard
Plants
(n)
Collected
berries (n)
Collected
foliar
lesions (n)
Complete
SSR
proles (n)
Genotypes
(n)
Latitude Longitude
2007 CH Cugnasco 24 192 - 149 5 46.17499 8.918889
2010 CH Castelrotto 16 62 47 18 2 45.98972 8.842783
2010 CH Astano 16 - 46 0 0 46.012298 8.817383
2010 CH Sessa 48 - 51 0 0 46.001225 8.816993
2010 CH Rivera 55 - 55 0 0 46.09999 8.933333
2010 CH Verscio 3 26 38 4 2 46.18333 8.733331
2010 CH Minusio 1 4 - 1 1 46.18333 8.816672
2010 CH Grancia 1 6 15 2 1 45.97027 8.927778
2011 CH Rudolngen 13 7 23 7 1 47.64275 8.67254
2011 CH Trüllikon - 7 (from
soil)
- 2 1 47.63868 8.69103
2007 F Salleboeuf 14 116 - 104 33 44.84333 -0.39707
2007 F Blanquefort 10 80 - 68 19 44.91547 -0.64033
2010 F Blanquefort 24 24 123 10 4 44.91547 -0.64033
2010 F Colmar 15 46 51 17 1 48.08588 7.358379
2010 D Geisenheim 16 48 80 18 2 49.98294 7.954738
2010 D Wolf 24 23 96 9 3 49.97237 7.089218
2010 LUX Grevenmacher 24 21 56 3 3 49.68507 6.43558
2010 USA Geneva (NY) 1 - 11 2 2 42.86822 -77.04586
Luxembourg, and from one single plant located in
a North American vineyard (Table 2). DNA was ex-
tracted from infected berries following the CTAB
method (Aldrich and Cullis, 1993) and from leaves
following a “Quick and Dirty” procedure with the
extraction and dilution buers of the Extract-N-Amp
TM Plant PCR Kit (Fluka; see Frey et al., 2004).
Construction of microsatellite library and primer
design
Microsatellite markers were developed accord-
ing to the protocol described by Brunner and Frey
(2004), with the following modications: genomic
DNA was extracted from P. ampelicida cultures 170
and 1003 (reference cultures provided by Mycoscope
Agroscope, see Table 1) following the CTAB proce-
dure (Aldrich and Cullis, 1993) and then digested
with restriction enzyme RsaI (New England Bio-
lab). Super SNX24 linkers (see Glenn and Schable,
2005) were ligated onto DNA fragments using T4
DNA ligase (Fermentas). Polymerase chain reac-
tion (PCR) using one strand of the linker as primer
was performed using Dream Taq PCR Mastermix kit
(Fermentas). Fragments from 500 to 1000 bp were
excised from agarose gels following electrophore-
sis. Linker-ligated DNA was denatured for 5 min at
95°C and hybridized separately to 0.1 mM 5´ biotin-
labeled oligonucleotide probes (AT)
10
, (AG)
10
and
(AGC)
8
with hybridization buer (12 × SSC, 0.2%
SDS) by Touch Down PCR in 50 μL total volume.
The cycle of amplication was the following: 95°C

473
Vol. 53, No. 3, December, 2014
Microsatellite markers for Phyllostica ampelicida
for 5 min, 72°C to 50°C (∆ = 0.2°C) for 5 s, 50°C for
10 min, 50°C to 40°C (∆ = 0.5 °C) for 5 s, and con-
servation at 15°C. Hybridized DNA fragments were
captured with streptavidin-coated magnetic beads
(Dynabeads M-280; Invitrogen), eluted overnight in
25 μL Tris-Low-EDTA (TLE) and recovered by PCR
with the following cycle: 95°C for 2 min, 25 × (95°C
for 20 s, 60°C for 20 s, 72°C for 90 s), 72°C for 30 min,
conservation at 15°C. Enriched DNA fragments were
ligated into plasmid vectors using the CloneJet PCR
Cloning Kit, following the “Sticky-End cloning” pro-
tocol (Fermentas). Transformation into chemically
competent Escherichia coli TOP10 cells was carried
out using a TOPO TA Cloning Kit, according to the
“OneShot TOP10” protocol (Invitrogen). For each
probe, 48 to 96 positive colonies were re-plated on LB
agar containing 50 μg mL
-1
ampicillin, then screened
by PCR for microsatellite inserts using pJET1.2 prim-
ers (Fermentas). Amplicons were visualized by aga-
rose gel electrophoresis. For each probe, fragments
of dierent sizes were puried with NucleoFast 96
PCR plates (Macherey-Nagel) and sequenced using
the BigDye Terminator v3.1 Cycle Sequencing Kit
(Applied Biosystems) in an automated ABI PRISM
3130xl Genetic Analyzer (Applied Biosystems).
Primers were designed from the sequences con-
taining microsatellites with the software Primer3 v.
0.4.0 (Rozen and Skaletsky, 1999), in order to obtain
amplicons between 100 and 350 bp with a minimum
dierence in size of 20 nucleotides, allowing the vi-
sualization of clearly distinct peaks in the fragment
analysis proles. The melting temperature was set
with an optimum of 60°C to enable a single-step am-
plication by multiplex PCR. Primer specicity for
P. ampelicida was checked by BLAST search (Altschul
et al., 1990). All sequences virtually binding to non-
target DNA, including Vitis spp., all Phyllosticta/
Guignardia spp., and other grape pathogens, were
not considered.
A specicity test was carried out using the SSR
markers on two pure cultures of Phyllosticta isolated
from Parthenocissus, provided by Mycoscope Agro-
scope (cultures 264 and 1002, see Wicht et al., 2012),
because we consider this to be the closest taxon to
P. ampelicida derived from Vitis in terms of genic ho-
mology (Wicht et al., 2012). Moreover, nine microsat-
ellite markers developed for Phyllosticta cultures 264
and 1002 in the course of a previous study (data not
shown) were tested on all P. ampelicida monohyphal
cultures derived from Vitis.
SSR amplication
Microsatellites were amplied by multiplex PCR,
using the Multiplex PCR kit (Qiagen) following the
“Amplication of microsatellite loci using multiplex
PCR” protocol. The cycle was set as follows: 95°C for
15 min, 35 × (94°C for 30 s, 55°C for 90 s, 72°C for 1
min), and 60°C for 30 min. Fragment analysis was
performed loading 1 μL PCR product with 8.8 μL
formamide and 0.2 μL size standard GeneScan 500
LIZ (Applied Biosystems) in an automated sequenc-
er (ABI PRISM 3130xl; Applied Biosystems). Allelic
proles were analyzed and corrected using the soft-
ware GeneMapper v. 3.2 (Applied Biosystems).
Data analyses
In a rst step, only isolates were considered for
which the allele content at all microsatellite loci
could be determined, and that therefore had com-
plete SSR proles. This set was rst used to count the
number of dierent genotypes present in each popu-
lation. After genotype correction, haploid genetic di-
versity was calculated using GENALEX (Peakall and
Smouse, 2006) and the clustering of individuals was
performed using STRUCTURE v. 2.3.3 (Pritchard et
al., 2000), using the no admixture model, with Al-
lele frequencies correlated, Burning length: 100,000,
Number of MCMC Reps after Burning: 10,000, ve
replicates per run, varying K from two to 14. Allelic
richness was calculated using ADZE (Szpiech et al.,
2008), considering rst only populations with com-
plete SSR proles (prior genotype correction) and
showing population size n > 15. Neighbour-joining
genetic trees of individuals were generated using
POPULATIONS (Langella, 2002), calculating dis-
tances according to Nei et al. (1983), with a 500 boot-
strap replicas. Samples were grouped in populations
according to the region of sampling.
Results
Microsatellite marker identication and
polymorphism
Forty-nine colonies were sequenced for inserts,
23 of which contained repeat motifs of appropriate
length and suitable for designing specic primers.
Sets of 23 primer pairs were designed, 11 of which
were polymorphic in the reference cultures, and
were therefore selected for the amplication of P.

Phytopathologia Mediterranea
474
B. Narduzzi-Wicht et al.
Table 3. Primers amplifying microsatellite markers developed for the characterization of Phyllosticta ampelicida.
Locus ID
GenBank
Accession
Number
Primer sequence (5´–3´)
5´
modication
Repeat
motif
Allele
range (bp)
Alleles
(n)
Mean
haploid
genetic
diversity
GBMS01 - KF730254
F: CCTTTGAGACCCCTCAACAT
R: GCCTTCCTCCATGTGTAACG
FAM (GCT)
8
131-143 2 0.145
GBMS02 - KF730255
F: GCCAGTAACCAATCGTTCG
R: CTGGTTCATGCGTTGGAAG
AT550 (GA)
9
173-177 2 0.139
GBMS03 - KF730256
F: GGCTTCTGCGAATAGCAAAC
R: CTTCCTCAATCCTTCCGATG
YYE (AG)
16
189-203 6 0.195
GBMS04 - KF730257
F: GTGGACGAAGACTCCCATGT
R: GCAATTTGGCAATAGGTGGT
AT565 (GA)
10
211-227 3 0.174
GBMS05 - KF730258
F: CAGCGGAACTGTAGTCGTCA
R: TGGATTCGAGATTTGAAGCA
FAM (TC)
8
244-246 2 0.072
GBMS06 - KF730259
F: GAATGAGCGCATGACGAGTA
R: ATTCAACGCACCATCTCCTC
AT550 (GAC)
14
257-278 4 0.373
GBMS07 - KF730260
F: AAGCTTTGCAGGGACTTGAA
R: TGCTGCTGTCTATCTTGGCTA
YYE (CAG)
10
280-298 3 0.283
GBMS08 - KF730261
F: CTCAATTGCCTGGCTTTCAC
R: CCGACTCACCGTCTTTTTGT
AT565 (GA)
12
321-335 4 0.312
GBMS09 - KF730262
F: TTGGACCAAGGTTGAAGGAC
R: CGTTTCGTTGTAGCGTTCAG
FAM (GA)
10
342-360 2 0.056
GBMS10 - KF730263
F: GGATATCGTTCGGTTTGTGG
R: GTCTGCATCTAGGCCAGCTC
AT550 (GAC)
10
378-390 3 0.243
GBMS11 - KF730264
F: ACTAAGCCGCATTCTGCAAT
R: GGGGAGATTTGGTGTTTTGA
AT565 (CAT)
9
387-396 2 0.247
ampelicida microsatellite markers. The sequence of
the primer pairs, 5´- modication and allele size are
shown in Table 3.
The specicity test carried out on monohyphal
cultures of Phyllosticta isolated from Parthenocissus
was negative, i.e. the primers designed for P. ampeli-
cida did not amplify any microsatellites in strains
264 and 1002. Negative amplication was neither
observed by testing nine SSR markers specically
developed for these Phyllosticta strains (data not
shown) on P. ampelicida isolates derived from Vitis.
The 1354 environmental samples collected were
screened with the 11 microsatellites in multiplex
PCR, yielding 1454 genotypes (including complete
and partial proles), as multiple alleles per SSR lo-
cus were observed in 88 samples (two alleles in one
locus in 73 cases, two alleles in two loci in 11 cases,
two alleles in three loci in two cases, and three alleles
in one locus in two cases). Investigating the geno-
types with a maximum of two missing values, the
most polymorphic marker was GBMS03 (see Table
3), with six alleles. Marker GBMS08 also showed six
alleles, but closer examination of their sizes showed
the presence of two groups of alleles diering by
a single base (320‒321 bp and 330‒331 bp, respec-
tively). For further analysis, alleles 320 and 321 were
therefore considered to be the same (called “321”), as
well as alleles 330 and 331 (called “331”).
Pathogen populations
Four hundred and fourteen environmental sam-
ples (407 of mummied berries and seven of foliar le-
sions) showed a complete SSR prole, among which

475
Vol. 53, No. 3, December, 2014
Microsatellite markers for Phyllostica ampelicida
64 distinct genotypes were retrieved. Samples were
divided by country, region and year of collection.
Samples collected in 2010 in Northern Switzerland
were also grouped together (CH-DE-10, collected in
Rudolngen and Trüllikon). All complete genotypes
from Northern Switzerland (CH-DE-10, n = 9) rep-
resented a single SSR prole. Samples from South-
ern Switzerland (Canton Ticino) collected in 2010
(Castelrotto-2010, n = 18; Grancia-2010, n = 2; Minu-
sio-2010, n = 1; Verscio-2010, n = 4) were grouped
into a population (CH-TI-10, n = 25) consisting of ve
genotypes. The population from Cugnasco (South-
ern Switzerland) collected in 2007 (n = 149) consisted
of ve genotypes, one of which was found 115 times,
while the second most frequent genotype was found
27 times. In France, the population collected in Blan-
quefort in 2007 (F-BLQ-07, n = 68) consisted of 19
genotypes, while the population collected in 2010 (F-
BLQ-10, n = 10) was only of eight. The population
from Salleboeuf collected in 2007 (n = 104) consisted
of 33 genotypes, and that from Colmar (n = 17) was
found to be clonal. In Germany, the population from
Wolf consisted of three genotypes; two were found
only once and a third was found seven times. The
population from Geisenheim consisted of two geno-
types, one found 17 times, and one only once. The
few samples from Luxembourg (Grevenmacher:
LUX-GRE, n = 3) and the USA (New York: USA-NY,
n = 2) consisted of dierent genotypes.
Twelve genotypes out of the 64 identied were
found in more than a single location, usually only
two or three, but a specic genotype (134-175-193-
215-246-269-280-321-342-390-390) was observed in
seven dierent populations. Genotype-corrected
data (genotype-specic per population) have been
used to generate the phylogenetic tree shown in Fig-
ure 1. Several clades were identied, but only the
one consisting of seven isolates from Canton Ticino
showed a correlation with the geographic origin of
the samples. However, the remaining genotypes
Figure 1. Neighbour-joining genetic tree of Phyllosticta ampelicida individuals gener-
ated using the POPULATIONS software, calculating distances according to Nei et al.
(1983), with bootstrap values from 500 repetitions. Samples were grouped in popula-
tions according to region of sampling. Environmental samples are coded by country,
region, collection year and a number assigned to the genotype. Isolates are coded by
country and collection year, the isolate codes corresponding to those in Table 1. Coun-
tries: CH, Switzerland; F, France; DE, Germany; LUX, Luxembourg; USA, United
States. Regions, TI, Southern Switzerland; DE, Northern Switzerland; BLQ, Blanque-
fort; COL, Colmar; SB, Salleboeuf; GRE, Grevenmacher; WOL, Wolf; NY, New York.

Phytopathologia Mediterranea
476
B. Narduzzi-Wicht et al.
sampled in this region were spread over several dif-
ferent clusters. The same genotype-corrected popu-
lation data were used for population clustering, and
data resulting from the dierent numbers of cluster
(K) were estimated by the statistic ΔK (Evanno et al.,
2005). This indicated that the samples could not be
separated into subpopulations. Allelic richness was
calculated for a sampling size up to 16 (Figure 2),
considering complete SSR proles from only ve
populations with n > 15. This indicated that for the
two populations in France (2007) the allelic richness
increased steadily with increasing sampling size. For
the other populations this value increased much less.
The population from Colmar could not be analyzed
as this showed a single genotype, identical to a geno-
type identied from the USA. However, it was quite
dierent from any other genotype identied in the
other populations investigated.
Complete SSR proles were retrieved for all
monohyphal cultures and the lyophilized mycelium.
Isolate 559 was identical to 1003, while isolate 9494
was found to be identical to two genotypes identied
among the environmental samples (DE-GEI-10-2,
DE-WOLF-10-2). Isolates 9450 and 9449 showed all
common alleles, except one (marker GBMS06, alleles
268 and 277 respectively). Isolate 9390 was identical
to F-SLB-07-26. In total, ve new genotypes were
identied from the reference strains of P. ampelicida
provided by the Mycoscope (Agroscope).
Genotype distribution in vineyards from more than
50 collected berries
In order to study the distribution within a single
vineyard, three locations were sampled extensively
in 2007, collecting eight berries per bunch for each
infected plant.
The vineyard of Cugnasco (Southern Switzer-
land) was made up of eight rows of 80 plants, and
each row consisted of a dierent grape cultivar
(Chambourcin, Gamaret, Bianca, Solaris, Müller-
Thurgau, Isabella, Merlot and Regent). One hundred
forty-nine berries from 24 dierent bunches were
collected and a total of ve genotypes (CH-TI-2007-1
to CH-TI-2007-5) were identied. The distribution of
the genotypes (renamed A-E for simplicity) over the
whole vineyard is shown in Figure 3. The number
of genotypes identied from a single bunch ranged
from one to four, with an average of 1.6 ± 0.8 geno-
types/bunch. No particular subdivision of the geno-
types was linked to the cultivar from which they had
been sampled.
From the sampling performed in 2007 in Blanque-
fort (France), 19 genotypes were observed from 68
complete proles of mummied berries out of ten
bunches, with an average of 4.1 ± 1.5 genotypes/
bunch. A similar situation was retrieved in Salle-
boeuf (France) in the same year: 33 genotypes were
identied from 104 complete proles from mum-
Figure 2. ADZE rarefaction results from 11 SSR loci of Phyllosticta ampelicida populations. The rareed mean allelic richness
is shown in relation to increasing sample size (g). Maximum g sample sizes are limited to 16.

477
Vol. 53, No. 3, December, 2014
Microsatellite markers for Phyllostica ampelicida
mied berries out of 14 bunches, with two to seven
genotypes per bunch (average 4.4 ± 1.5 genotypes/
bunch).
Discussion
We present for the rst time the development of
P. ampelicida specic SSR markers. They were used
in multiplex PCR to provide knowledge of the biol-
ogy of the causal agent of grape black rot, by using
pure cultures as well as environmental samples from
France, Germany, Switzerland, Luxembourg and
few samples collected in New York state, USA. The
11 SSR primer pairs were designed in order to am-
plify regions diering by about 20 bp. This strategy,
in combination with the use of dierent 5´-labelling
uorescent dyes, allowed the amplication of all
samples in a single PCR reaction. However, the suc-
cessful amplication of all loci was dependent on the
quality of the template: DNA extracted from pure
cultures using the Qiagen kit mostly amplied all
loci successfully, while the CTAB-extraction method
used for the mummied berries led in some cases to
amplication failure of single loci. The “Quick and
dirty” method used for the leaves instead led to a
low number of complete SSR proles. This prob-
lem can be ascribed to the fact that the DNA was
not isolated from leaves but just “eluted”, and still
contained PCR-inhibiting substances. For two thirds
of the samples three or more loci failed to amplify,
which can also be ascribed to the large number of
SSR combined in a single multiplex amplication,
as well as the poor quality of the DNA. A kit-based
DNA extraction protocol could reduce the amount
of inhibitors, allowing better DNA amplication of
all SSR loci in all samples. This tool, combined with
the splitting of the 11 SSR markers into two separate
multiplex PCRs, may increase the number of com-
plete proles.
The 1354 collected samples were subjected to mul-
tiplex SSR amplication, and 6.5% of these showed
“mixed genotypes” (i.e. more than a single allele per
SSR locus), indicating the presence of two dierent
genotypes on individual berries. Seven monohyphal
cultures and one lyophilized mycelium from public
collections revealed that two Swiss isolates, i.e. iso-
late 1003 collected in 2003 on Vitis sp. (unknown loca-
tion) and isolate 559 collected in 1981 (Belmont) from
V. labrusca, showed an identical SSR prole. This in-
dicates that this genotype has survived for over 22
years without undergoing any outcrossing, or that
these two genotypes are identical in the SSR prole
Figure 3. Graphical representation of the Phyllosticta ampelicida genotypes identied in the vineyard in Cugnasco (2007);
Southern Switzerland; by investigating eight berries from a single bunch per plant showing black rot symptoms. The plot
consisted of eight rows of 80 plants each; each row consisting of a single grape cultivar. The groups of eight circles indicate
the positions where the bunches were sampled; and each colour or pattern corresponds to a genotype (A–E). Empty circles
(m) indicate incomplete genotypes that were excluded from the analyses.

Phytopathologia Mediterranea
478
B. Narduzzi-Wicht et al.
by chance. Isolate 9494, collected in 2010 in Mulheim
(Mosel, Germany) was also found to be identical
with isolates collected near Colmar and Geisenheim,
at distances of 60 to 200 km from each other. This
could mean that a single genotype has spread clon-
ally (asexually or sexually via self-mating) over this
region, or that here identical genotypes have arisen
by chance. Similar to the mentioned monohyphal
cultures, we identied an identical clone from sev-
en dierent regions over two collection years, even
from overseas (the USA).
Due to the very limited number of foliar lesions
providing a complete SSR prole (only seven com-
pared to the 407 from mummied berries), it was not
possible to draw a clear conclusion on whether iso-
lates identied on berries also occur on leaves, and on
which kind of tissue more genotypes occur. In Blan-
quefort (France, 2010) four genotypes were found
from six berries, which were dierent from the four
genotypes retrieved from foliar lesions. We tried to
investigate and more generally compare samples
showing up to two missing data, and to determine
if the number of genotypes present on leaves was
greater than that on berries. However, this was not
possible due to the limited number of samples, and
therefore we focused our research on mummied
berries.
SSR data suggest the absence of relationships be-
tween the species/cultivar of Vitis host plants and
the infecting genotype of P. ampelicida. In the vine-
yard of Colmar, only one genotype was found on
14 dierent cultivars of two Vitis species (V. vinifera:
Chambourcin, Villard blanc, Regent, Merlot, Johan-
niter, Solaris, Bronner and six experimental varieties;
and Vitis amurensis), while in Blanquefort and Salle-
boeuf 68 genotypes were isolated from a single vari-
ety (Merlot).
In general, the French populations collected in
2007 showed the greatest genetic diversity, as shown
by the mean allelic richness plot (Figure 2), and the
number of genotypes identied in these vineyards.
France was the rst country in which black rot was
reported (already occurring in 1885), probably due to
introduction of infected host material from the Unit-
ed States. From there, the disease probably spread
undergoing both sexual reproduction (either by self-
fertilization or by mating with another genotype), as
well as overwintering as asexual spores.
The likelihood of this kind of mixed propagation
is supported by the fact that identical clones can be
found to be spread all over a vineyard (Figure 3), as
well as at several hundreds of kilometers of distance
(as result of self-fertilization or asexual reproduc-
tion), while the large numbers of genotypes found in
some vineyards possibly result from sexual recombi-
nation of two or three genotypes (as in Blanquefort
where all loci showed one to three alleles). As dem-
onstrated for other plant pathogens (see Boehm et al.,
2003), it is possible that climatic conditions inuence
the sexual/asexual form of overwintering, while the
proportion of self-fertilization to outcrossing must
depend on the availability of dierent genotypes
within a vineyard. These aspects should be further
studied in P. ampelicida.
More information about other modes of overwin-
tering of P. ampelicida could be obtained by investi-
gating the primary lesions emerging in dierent lo-
cations under dierent climatic conditions, and fol-
lowing the progression of the disease. The number of
clones occurring in spring could be estimated with
assessment of how many are able to prevail and suc-
cessfully disseminate by secondary infections. Simi-
lar research has already been performed studying
the pathogen causing grape downy mildew (Gobbin
et al., 2005; Matasci et al., 2010). All information from
this research may allow the evolutionary potential of
P. ampelicida to be estimated.
Acknowledgements
This study was nancially supported by the Di-
partimento Educazione, Cultura e Sport (DECS; Bell-
inzona, Switzerland). Samples were kindly provid-
ed by Alex Angst (ETHZ, Switzerland), the sta of
Mycoscope (Agroscope), Andreas Kortekamp (DLR
Rheinpfalz, Neustadt a.d. Weinstraße, Germany),
Daniel Molitor (Centre de Recherche Public Gabriel
Lippmann, Belvaux, Luxembourg), Nicole Siebert
(Forschungsanstalt Geisenheim, Germany), Marc
Raynal and Marc Vergnes (IFV, Blanquefort, France),
Sabine Merdinoglu (UMR/INRA, Colmar, France),
Andreas Naef (Agroscope, Wädenswil, Switzer-
land), Matteo Bernasconi (Ucio della Consulenza
Agricola, Bellinzona), Michelle Moyer (Cornell Uni-
versity, NY, USA), and several private wine growers
in Canton Ticino. We are grateful to Renzo Lucchini
and all the collaborators of the Cantonal Institute of
Microbiology (Bellinzona), and the Plant Pathology
group (ETHZ) for technical assistance.

479
Vol. 53, No. 3, December, 2014
Microsatellite markers for Phyllostica ampelicida
Literature cited
Aldrich J. and C. Cullis, 1993. RAPD analysis in ax: optimi-
zation of yield and reproducibility using klenTaq 1 DNA
polymerase, chelex 100, and gel purication of genomic
DNA. Plant Molecular Biology Reporter 11, 128–141.
Altschul S.F., W. Gish, W. Miller, E.W. Myers and D.J. Lipman,
1990. Basic Local Alignment Search Tool. Journal of Molecu-
lar Biology 215, 403–410.
Anonymous, 2008. Das Weinjahr 2008. Institute Viti-vinicole in
Remich, Luxembourg, 27 pp.
Boehm E.W.A., S. Freeman, E. Shabi and T.J. Michailides,
2003. Microsatellite primers indicate the presence of asex-
ual populations of Venturia inaequalis in coastal apple or-
chards. Phytoparasitica 31, 236–251.
Brunner P.C. and J.E. Frey, 2004. Isolation and characterization
of six polymorphic microsatellite loci in the western ow-
er thrips Frankliniella occidentalis (Insecta, Thysanoptera).
Molecular Ecology Notes 4, 599–601.
Evanno G., S. Regnaut and J. Goudet, 2005. Detecting the num-
ber of clusters of individuals using the software STRUC-
TURE: a simulation study. Molecular Ecology 14, 2611–2620.
Frenkel O., I. Portillo, M. Brewer, J.-P. Peros, L. Cadle-David-
son and M. Milgroom, 2012. Development of microsatel-
lite markers from the transcriptome of Erysiphe necator for
analysing population structure in North America and Eu-
rope. Plant Pathology 61, 106–119.
Frey J.E., B. Frey, C. Sauer and M. Kellerhals, 2004. Ecient
low-cost DNA extraction and multiplex uorescent PCR
method for marker-assisted selection in breeding. Plant
Breeding 123, 554–557.
Glenn T.C. and N.A. Schable, 2005. Isolating microsatellite
DNA loci. Methods in Enzymology 395, 202–222.
Gobbin D., M. Jermini, B. Loskill, I. Pertot, M. Raynal and C.
Gessler, 2005. Importance of secondary inoculum of Plas-
mopara viticola to epidemics of grapevine downy mildew.
Plant Pathology 54, 522–534.
Gobbin D., A. Rumbou, C.C. Linde and C. Gessler, 2006. Pop-
ulation genetic structure of Plasmopara viticola after 125
years of colonization in European vineyards. Molecular
Plant Pathology 7, 519–531.
Harms M., B. Holz, C. Homann, H. Lipps and W. Silvanus,
2005. Occurrence of Guignardia bidwellii, the causal fun-
gus of black rot on grapevine, in the vine growing areas
of Rhineland-Palatinate, Germany. In: Plant Protection and
Plant Health in Europe: Introduction and Spread of Invasive
Species (British Crop Production Council, ed.). Humboldt
University, Berlin, Germany, 9–11 June. Symposium Pro-
ceedings No. 81, pp. 127–132.
Jailloux F., 1992. In vitro production of the teleomorph of Guig-
nardia bidwellii, causal agent of black rot of grapevine. Ca-
nadian Journal of Botany 70, 254–257.
Kong G., 2009. Diagnostic Methods for Black Rot of Grapes. PaDIL
Plant Biosecurity Toolbox, 28 pp.
Langella O., 2002. POPULATIONS 1.2.32, Population Genetic
Software (Individuals or Population Distances, Phylogenetic
Trees), http://bioinformatics.org/~tryphon/populations.
Luttrell E., 1946. Black rot of muscadine grapes. Phytopathology
36, 905–924.
Maixner M. and B. Holz, 2003. Risiken durch invasive gebiets-
fremde Arten für den Weinbau. Angewandte Wissenschaft
498, 154–164.
Matasci C.L., M. Jermini, D. Gobbin and C. Gessler, 2010. Mi-
crosatellite based population structure of Plasmopara viti-
cola at single vine scale. European Journal of Plant Pathology
127, 501–508.
Nei M., F. Tajima and Y. Tateno, 1983. Accuracy of estimated
phylogenetic trees from molecular-data .2. Gene-Frequen-
cy Data. Journal of Molecular Evolution 19, 153–170.
Peakall R. and P.E. Smouse, 2006. GENALEX 6: genetic analy-
sis in Excel. Population genetic software for teaching and
research. Molecular Ecology Notes 6, 288–295.
Pezet R. and M. Jermini, 1989. Le Black-Rot de la vigne: symp-
tômes, épidémiologie et lutte. Revue Suisse de Viticulture,
Arboriculture, Horticulture 21, 27–34.
Pritchard J.K., M. Stephens and P. Donnelly, 2000. Inference
of population structure using multilocus genotype data.
Genetics 155, 945–959.
Ramsdell D. and R. Milholland, 1988. Black rot. Compendium
of Grape Diseases. The American Phytopathological Society.
St. Paul, MN, USA, 15–17.
Reddick D., 1911. The black-rot disease of grapes. Cornell Agri-
cultural Experiment Station Bulletin 293, 287–364.
Rozen S. and H. Skaletsky, 1999. Primer3 on the WWW for
general users and for biologist programmers. Bioinformat-
ics Methods and Protocols, 365–386.
Szpiech Z.A., M. Jakobsson and N.A. Rosenberg, 2008. ADZE:
a rarefaction approach for counting alleles private to com-
binations of populations. Bioinformatics 24, 2498–2504.
Ullrich C.I., R.G. Kleespies, M. Enders and E. Koch, 2009. Biol-
ogy of the black rot pathogen, Guignardia bidwellii, its de-
velopment in susceptible leaves of grapevine Vitis vinifera.
Journal für Kulturpanzen 61, 82–90.
van der Aa H., 1973. Studies in Phyllosticta I. Studies in Myco-
logy 5, 1–110.
Wicht B., O. Petrini, M. Jermini, C. Gessler and G.A.L. Broggi-
ni, 2012. Molecular, proteomic and morphological charac-
terization of the ascomycete Guignardia bidwellii, agent of
grape black rot: a polyphasic approach to fungal identi-
cation. Mycologia 104, 1036–1045.
Wikee S., L. Lombard, C. Nakashima, K. Motohashi, E. Chuke-
atirote, R. Cheewangkoon, E.H. McKenzie, K.D. Hyde and
P.W. Crous, 2013. A phylogenetic re-evaluation of Phyllo-
sticta (Botryosphaeriales). Studies in Mycology 76, 1–29.
Wilcox W.F., 2003. Black rot. Cornell Cooperative Extension,
Disease identication sheet No. 102GFSG-D4.
Zalapa J.E., H. Cuevas, H.Y. Zhu, S. Stean, D. Senalik, E.
Zeldin, B. McCown, R. Harbut and P. Simon, 2012. Using
next-generation sequencing approaches to isolate simple
sequence repeat (ssr) loci in the plant sciences. American
Journal of Botany 99, 193–208.
Zane L., L. Bargelloni and T. Patarnello, 2002. Strategies for
microsatellite isolation: a review. Molecular Ecology 11,
1–16.
Accepted for publication: 5 June 2014
Published online: December 22, 2014