Dying-arm disease in grapevines: Diagnosis of infection with Eutypa lata by metabolite analysis

Article abstract of DOI:10.1021/jf0510236

Dying-arm disease in grapevines, produced by infection with the ascomycete Eutypa lata, is responsible for major production losses in vineyards. Dieback of the shoots and cordon is believed to be due to acetylenic phenol metabolites produced by the fungus. To identify specific metabolites that could potentially be used for diagnosis of infection, eight E. lata isolates were grown in vitro on hot water extracts from grape varieties with various degrees of tolerance to the foliar symptoms of E. lata dieback. HPLC analysis showed that eutypinol was consistently produced in large amounts, together with smaller amounts of methyleutypinol and eulatachromene; eutypine, the putative toxin, was produced solely on Sauvignon Blanc extract and then in only barely detectable amounts. When E. lata isolates from Cabernet Sauvignon and Merlot were grown on identical media, the amounts of metabolites produced differed significantly between isolates but the pattern of metabolites was quite similar, with eutypinol again predominating. The consistent production of eutypinol indicated that this was the most suitable metabolite for which to analyze in order to diagnose the presence of E. lata. Extraction and analysis of grapevine tissues exhibiting symptoms of dieback failed to show the presence of any metabolites. However, when infected cordon sections were placed in water and cultured for 5 days, eutypinol was readily detected in the aqueous solution; metabolites were not produced from uninfected tissue. This provides a method for detection of infected tissue and indicates that the toxic metabolites react at the point of production, disrupting the vascular structure and inhibiting transport of nutrients, rather than being translocated to tissues that exhibit symptoms.

Full text of DOI:10.1021/jf0510236

8148 J. Agric. Food Chem. 2005, 53, 8148−8155
Dying-Arm Disease in Grapevines: Diagnosis of Infection with
Eutypa lata by Metabolite Analysis
†
,†
†,‡
NOREEN MAHONEY, RUSSELL J. MOLYNEUX,* LEVERETT R. SMITH,
THOMAS K. SCHOCH, PHILIPPE E. ROLSHAUSEN,§ AND W. DOUGLAS GUBLER§
†
Western Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture,
800 Buchanan Street, Albany, California 94710, and Department of Plant Pathology, University of
California, One Shields Avenue, Davis, California 95616
Dying-arm disease in grapevines, produced by infection with the ascomycete Eutypa lata, is
responsible for major production losses in vineyards. Dieback of the shoots and cordon is believed
to be due to acetylenic phenol metabolites produced by the fungus. To identify specific metabolites
that could potentially be used for diagnosis of infection, eight E. lata isolates were grown in vitro on
hot water extracts from grape varieties with various degrees of tolerance to the foliar symptoms of E.
lata dieback. HPLC analysis showed that eutypinol was consistently produced in large amounts,
together with smaller amounts of methyleutypinol and eulatachromene; eutypine, the putative toxin,
was produced solely on Sauvignon Blanc extract and then in only barely detectable amounts. When
E. lata isolates from Cabernet Sauvignon and Merlot were grown on identical media, the amounts of
metabolites produced differed significantly between isolates but the pattern of metabolites was quite
similar, with eutypinol again predominating. The consistent production of eutypinol indicated that this
was the most suitable metabolite for which to analyze in order to diagnose the presence of E. lata.
Extraction and analysis of grapevine tissues exhibiting symptoms of dieback failed to show the
presence of any metabolites. However, when infected cordon sections were placed in water and
cultured for 5 days, eutypinol was readily detected in the aqueous solution; metabolites were not
produced from uninfected tissue. This provides a method for detection of infected tissue and indicates
that the toxic metabolites react at the point of production, disrupting the vascular structure and inhibiting
transport of nutrients, rather than being translocated to tissues that exhibit symptoms.
KEYWORDS: Grapes; Vitis vinifera; Eutypa lata; dieback; eutyposis; dying-arm disease; eutypinol
INTRODUCTION
longevity of the grapevines. Yield decreases for five vineyards
in California growing either Chenin Blanc or French Columbard
grapes were estimated to range from 30 to >60%, whereas
vineyards over 20 years old had up to 83% yield reduction,
relative to their peak production at ∼10-12 years of age (2).
There are significant differences in tolerance to foliar symptoms
of infection, with some of the most valuable cultivars, including
Cabernet Sauvignon, being particularly sensitive (3). The cost
to wine grape production alone in California has been estimated
to be in excess of $260 million per annum, or ∼16% of the
gross producer revenue of $1.672 billion for 1999 (4).
Dying-arm disease or Eutypa dieback is an important peren-
nial canker disease that affects grapevines (Vitis Vinifera;
Vitaceae) worldwide, including the United States, with particular
economic impact in the primary wine-producing areas of
California. The causative agent, the ascomycete Eutypa lata (1),
enters the plant principally through pruning wounds, leading to
necrosis of woody tissues in the vicinity of the point of infection
and may subsequently expand into the trunk of the vine. The
disease is cryptic, becoming apparent only after bud break as
stunting of new shoots, formation of small, deformed, chlorotic
leaves, and development of small fruit clusters. The disease is
progressive over many years, and failure to control it leads to
severe economic losses, primarily as a consequence of decreased
yields, increased vineyard management costs, and reduced
Phytotoxicity has been attributed to one or more of a number
of phenolic metabolites (Figure 1), bearing an unusual pen-
tenyne side chain ortho to the hydroxyl group, that have been
isolated from laboratory cultures of the fungus, together with
structurally related compounds in which the aromatic ring has
been reduced and epoxidized and the side chain cyclized (5,
6). A recent study using the yeast Saccharomyces cereVisiae as
a model bioassay system to investigate gene targets of E. lata
metabolites has shown that eutypinol, 1, and eulatachromene,
2, inhibit mitochondrial respiration, suggesting a probable mode
*
Corresponding author [telephone (510) 559-5812; fax (510) 559-6129;
e-mail molyneux@pw.usda.gov].
†
Western Regional Research Center.
Present address: Department of Chemistry, Contra Costa College, 2600
‡
Mission Bell Dr., San Pablo, CA 94806.
§
University of California, Davis.
1
0.1021/jf0510236 CCC: $30.25
© 2005 American Chemical Society
Published on Web 09/16/2005

Diagnosis of E. lata Infection in Grapevines
J. Agric. Food Chem., Vol. 53, No. 21, 2005 8149
Table 1. Identification of E. lata Isolates Investigated, Grape Variety
from Which Obtained, and Source of Fungal Inoculum
E. lata
isolate ID
grape varietal/origin
Shiraz/Australia
Merlot/Australia
Chardonnay/Australia
source
B001
D010
E002
N04
E113
E190
E203
E206
University of Adelaide
University of Adelaide
University of Adelaide
University of Adelaide
E&J Gallo
grape/Australia
grape/Napa Co., California
grape/Mendicino Co., California
grape/Sonoma Co., California
grape/San Joaquin Co., California
Merlot/Napa Co., California
Merlot/Napa Co., California
E&J Gallo
E&J Gallo
E&J Gallo
8
2
7
M
1M
CS
University of California, Davis
University of California, Davis
University of California, Davis
Cabernet Sauvignon/Napa Co.,
California
24CS
Cabernet Sauvignon/Napa Co.,
California
University of California, Davis
experiments established that eutypine was not consistently
produced, and when present, the amounts were generally low,
indicating that it is not the sole constituent responsible for
phytotoxicity, in contrast to earlier reports (8, 9). This was
confirmed in a grapeleaf disk bioassay that showed that
eulatachromene, eulatinol, eutypine, and the benzofuran all
exhibited necrotic effects with consequent chlorophyll reduction
relative to controls (11, 12). Furthermore, eutypine was barely
detectable or absent in cultures from a strain of E. lata from
California (E120) known to be pathogenic to grapes, even
though it was present in an Italian strain (E125) (11).
The frequent and abundant occurrence in vitro of eutypinol
and eulatachromene from isolates of E. lata suggested that either
or both of these compounds could be used as diagnostic markers
for the presence of the fungus in infected grapevines. A study
was therefore undertaken to establish that they were consistently
produced in hot water extracts from canes of grapevines other
than Cabernet Sauvignon, including varieties known to be
susceptible and tolerant, respectively, to dying-arm disease.
Grapevine tissues from a vineyard known to be infected with
dieback and showing visible symptoms of the disease were then
collected to develop both the protocol for sampling and the
analytical procedures necessary to use metabolite detection as
a surrogate for diagnosis of infection.
Figure 1. Chemical structures of E. lata metabolites: eutypinol, 1;
eulatachromene, 2; eutypine, 3; 2-isopropenyl-5-formylbenzofuran, 4;
siccayne, 5; eulatinol, 6; methyleutypinol, 7; methyleutypine, 8.
of action in grapevine cells (7). These results conflict with
attribution of the phytotoxicity of E. lata to the aldehyde,
eutypine, 3 (8), with tolerance of some cultivars to the disease
being inferred from the ability of a reductase in mung bean
(Vigna radiata) to reduce eutypine to eutypinol, 1, which was
not found to be phytotoxic to grape callus tissue in vitro (9);
however, there is no evidence for the presence of a similar
reductase in grapevines and this lack of phytotoxicity has not
been confirmed in planta (10). It is noteworthy that eutypine is
particularly sensitive to pH, cyclizing to 2-isopropenyl-5-
formylbenzofuran, 4, even under mildly acidic conditions (5),
and the relative phytotoxicity of the latter compound has also
not been established.
MATERIALS AND METHODS
Reagents. Pyridine (Fisher Scientific) was dried by storage over
molecular sieves, type 4A, 8-12 mesh (Aldrich Chemical Co.).
N-Methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) in 1 mL am-
pules was obtained from Pierce (Rockford, IL).
Instrumentation. UV spectra were measured at 190-500 nm in
MeOH solution, using a Hewlett-Packard 8452A diode array spectro-
photometer, and IR spectra were determined with a Nicolet Magna-IR
In view of the uncertainty as to the specific metabolites of
the fungus responsible for its phytotoxicity we have previously
analyzed by high-performance liquid chromatography (HPLC)
the metabolite composition of 11 fungal isolates from California,
Australia, New Zealand, Italy, and France grown on four
artificial media. These profiles were also compared with those
of the same isolates grown on Cabernet Sauvignon grapevine
wood hot water extracts, with and without sucrose supplementa-
tion. Only six compounds were produced in detectable amounts,
namely, eutypinol, 1, and its chromene analogue, eulatach-
romene, 2 (11); eutypine, 3, and its cyclization product, the
benzofuran, 4, together with the quinol, siccayne (5), and its
monomethyl analogue, eulatinol (6) (Figure 1). The two most
widely distributed and abundant metabolites were eutypinol and
eulatachromene, which were present in eight of the isolates
grown on grapewood aqueous extract fortified with sucrose (11).
Metabolite production on grapevine extract was greatly increased
relative to the artificial media, indicating that this native substrate
provides enhanced conditions and a more representative profile
of the metabolites produced in the natural disease state. These
5
50 series II spectrometer. Melting points were taken on a Mel-Temp
apparatus and are uncorrected. NMR spectra were obtained at 298 K
from samples dissolved in CDCl with TMS as an internal standard on
a Bruker ARX400 spectrometer at frequencies of 400.13 MHz ( H)
3
1
13
and 100.62 MHz ( C). A 90° pulse at a 7-8 s repetition rate was
1
used for H experiments, and a 30° pulse at a 2.3 s repetition rate was
used for 13C experiments. The number of attached protons for C signals
13
was determined from DEPT90 and DEPT135 assays.
Fungal Isolates. The fungal isolates investigated (Table 1) were
obtained as pure cultures from the E & J Gallo Winery Culture
Collection (Modesto, CA), the collection of Dr. Richard Lardner
(University of Adelaide, Glen Osmond, South Australia, Australia),
and the University of California (Davis, CA). E. lata isolates were
grown on potato dextrose agar (Difco) from which 5 mm plugs were
cut after 4 weeks and stored in water at 4 °C for use as inoculum in all
tested media.

8150 J. Agric. Food Chem., Vol. 53, No. 21, 2005
Mahoney et al.
Preparation of Grape-Based Media. Tested media consisted of a
hot water extract of selected grape cane varieties fortified with 1%
sucrose. Dormant canes (Napa Valley, CA) collected during the 2002
season were ground in a Wiley mill with a 1 mm screen. Liquid cane-
based medium was prepared by sonication (Branson) of 100 g of ground
cane of each variety with 500 mL of hot water (100 °C). This extract
was clarified by filtration through the following series of filters:
Miracloth (CalBiochem), multigrade GMF 150 (Whatman), GF/F
(Whatman), and Supor-200 0.2 µm (Gelman). The cane extract was
supplemented with 1% sucrose for the varietal cane media and with
1
% sucrose or glucose for the time course Cabernet cane medium.
Inoculation and Incubation of Cultures. Varietal cane media (50
mL of cane hot water extract per 250 mL flask) and Cabernet cane
media used for the time course (250 mL per 1 L flask) were autoclaved
followed by the addition of three plugs of E. lata inoculum. All cultures
were incubated at 22 °C. Varietal cane cultures were incubated for 30
days before metabolite extraction. Aliquots (20 mL) were removed from
the time course cultures for metabolite extraction after 12, 20, 28, 34,
3
9, 42, 46, 50, and 57 days of incubation.
Quantitative Analysis of Metabolites in E. lata Growth Media.
E. lata liquid cultures (50 mL) were filtered through Whatman no. 4
paper, and the filtrate was extracted with Et
O (2 × 50 mL). Time
course aliquots (20 mL) were extracted without filtration with Et O (2
20 mL). The Et O extracts were combined and extracted with H
50 mL). The organic phase was collected and the Et O removed under
reduced pressure. The residue was dissolved in MeOH (1 mL) and
filtered through a 0.45 µm, 13 mm, syringe filter (Gelman). Samples
Figure 2. Synthetic route to methyleutypinol, 7, and methyleutypine, 8.
2
2
2
under an N atmosphere was treated with anhydrous THF (6.0 mL),
×
2
2
O
then triethylamine (10.0 mL), and, finally, after a clear yellow solution
was obtained, 2-methyl-1-buten-3-yne (3.0 mL, 31.5 mmol). The
mixture rapidly darkened (1-2 min), and stirring was continued at
ambient temperature for 25 h in the dark, after which it was diluted
(
2
(
20 µL) were analyzed by HPLC (Agilent 1100) using a 250 mm ×
.6 mm i.d., 5 µm, Microsorb C18 column (Varian) with gradient
elution at 1.0 mL/min of 100% H O containing 0.5% AcOH changing
to 100% CH CN over 30 min and held at 100% CH CN for 5 min.
with hexane/Et
cotton wool, rinsed through with 20 mL of additional solvent, then
washed successively with saturated aqueous NH
Cl (3 × 20 mL), 5%
aqueous NaHCO
(2 × 20 mL), and saturated aqueous NaCl (2 × 20
mL). The solution was dried over anhydrous Na SO and the solvent
removed by rotary evaporation to give an amber oil, which was
dissolved in Et O (30 mL), and then purified by passage through silica
gel (5.0 g) using additional Et O (100 mL) to complete elution.
2
O (9:1 v/v, 30 mL). The supernatant was filtered through
4
2
4
3
3
3
Detection was either by UV at 254 nm (Agilent 1100 VWD) or
photodiode array (Agilent 1100 DAD). Metabolite concentrations
were determined by reference to standard curves prepared for each
compound, which were linear over the range tested, 0.2-20 µg/20 µL
injection.
2
4
2
2
Reconcentration gave an oil from which 8 was isolated by preparative
HPLC. The product solidified, giving a total purified yield of 0.544 g
GC-MS Analysis of TMS Derivatives. TMS derivatives of indi-
vidual compounds or mixtures were prepared by suspension of the
sample (∼0.5 mg) in dry pyridine (100 µL) in a 1.0 mL Reacti-Vial
(
(
(
(
(
2.72 mmol; 71%): mp 45.5-48 °C [lit. 49-50 °C (5)]; GC-MS
+
underivatized by treatment with MSTFA) t
R
13.99 min; m/z 200 [M ]
(Pierce), to which was added MSTFA (100 µL). The reaction mixture
100), 185 (13), 159 (37), 128 (76), 115 (15); UV (MeOH) λmax nm
was then heated at ∼60 °C for 1 h with periodic shaking to ensure
complete dissolution of all reactants. Analyses were performed on a
Hewlett-Packard 5890 series II instrument equipped with a 5971 mass-
selective detector (MSD) and a 60 m × 0.32 mm i.d., 0.25 µm, SE-30
fused Si capillary column (J&W Scientific, Folsom, CA). The column
was held at an initial temperature of 105 °C for 0.2 min, ramped at 30
1
log ꢀ) 210 (4.11), 262 (4.56); IR and H NMR matched literature values
13); 13C NMR (CDCl
5.8 (C-2′), 110.7 (C-5), 113.7 (C-3), 122.6 (dCH
) δ 23.4 (-CH
), 56.3 (-OCH
), 126.7 (C-3′), 129.7
3
3
3
), 83.2 (C-1′),
9
2
(C-1), 131.5 (C-6), 135.5 (C-2), 164.4 (C-4), 190.2 (-CHO).
Methyleutypinol, 4-Methoxy-3-(3′-methylbut-3′-en-1′-ynyl)benzyl al-
cohol, 7. Methyleutypine, 8 (0.093 g; 0.465 mmol), was dissolved in
MeOH (3.0 mL) and treated with NaBH (0.10 g; 2.6 mmol). After 30
min of stirring at ambient temperature, Et O (15 mL) was added,
°
C/min for 0.5 min, programmed from 120 to 300 °C at 10 °C/min,
and held at the final temperature for 10 min. Helium was used as carrier
gas with a head pressure of 60 psi. Derivatized samples (0.1-0.2 µL)
were introduced through an SGE model OC1-3 on-column injector held
at ambient temperature. The MSD was operated at 70 eV in the EI
mode with an ion-source temperature of 180 °C and scanning from
m/z 75 to 600 at a sampling rate of 1.5 scans/s. A postinjection delay
of 7.0 min was set to avoid MS data acquisition during elution of the
solvent and derivatization reagent.
4
2
followed by 2% aqueous HCl (2.5 mL). The organic phase was washed
with saturated aqueous NaCl (2 × 5 mL) and dried over anhydrous
2 4
Na SO , and the solvent was removed (rotary evaporator, then <1 Torr)
to give 0.091 g (0.45 mmol; 97%) of an almost colorless oil: GC-MS
+
(mono-TMS derivative) t
R
15.38 min; m/z 274 [M ] (65), 259 (17),
1
2
85 (100), 141 (12), 115 (13); UV (MeOH) λmax nm (log ꢀ) 216 (4.33),
1
76 (4.12), 264 (4.11), 306 (3.91); IR and H NMR spectra consistent
Synthesis of Methyleutypinol, 7, and Methyleutypine, 8 (Figure
13
with literature data (13); C NMR (CDCl
3
) δ 23.5 (-CH
OH), 84.5 (C-1′), 94.8 (C-2′), 110.9 (C-5), 112.6
2
), 127.0 (C-3′), 128.6 (C-6), 132.5 (C-4), 133.0
3
), 56.0
2
). 3-Iodo-4-methoxybenzaldehyde. 3-Iodo-4-hydroxybenzaldehyde (2.14
g, 8.63 mmol) in Et O (40 mL) was methylated with diazomethane
prepared from Diazald (N-methyl-N-nitroso-p-toluenesulfonamide) (10.0
g, 46.7 mmol) in Et O (100 mL) using the manufacturer’s (Aldrich,
(
(
(
-OCH
3
), 64.6 (-CH
2
2
C-3), 121.9 (dCH
C-1), 159.5 (C-4).
2
Milwaukee, WI) apparatus and procedure. When the yellow color of
Detection and Analysis of Metabolites in Infected Grapevine
the solution disappeared (30 min), the volume was reduced to ∼50
Tissues. Grapevine tissues (leaves, stems, fruits, canes, spurs, cordons,
and trunks) were obtained from a Zinfandel vineyard in the Sacramento
Valley and from Merlot and Cabernet Sauvignon vineyards in the Napa
Valley, California; we thank the vineyard management and staff for
identification of infected and uninfected vines and assistance in
collecting the material.
2
mL under N with a warm water bath; rotary evaporation was then
used to remove remaining solvent. Subsequent higher vacuum (<1 Torr)
gave a white powder (2.26 g; 8.63 mmol; 100%): mp 96-101 °C [lit.
1
1
(
1
12 °C (13)]; IR and H NMR spectra consistent with literature values
1
3
13); C NMR (CDCl
31.5 (C-4), 132.1 (C-6), 141.1 (C-2), 162.8 (C-1), 189.3 (-CHO).
Methyleutypine, 4-Methoxy-3-(3′-methylbut-3′-en-1′-ynyl)benzalde-
hyde, 8. A mixture of 3-iodo-4-methoxybenzaldehyde (1.05 g, 3.84
mmol), CuI (0.21 g, 1.11 mmol), and Pd(Ph P) (0.24 g, 0.208 mmol)
3
) δ 56.8 (-OCH
3
), 86.5 (C-3), 110.6 (C-5),
Canes, spurs, cordons, and trunks were ground in a Wiley mill with
a 1 mm screen. Leaves, stems, and fruits were separated by hand. Plant
samples (50 g) were homogenized with a Polytron (Brinkmann
Instruments) in water, water adjusted to pH 4 with HCl, water adjusted
3
4

Diagnosis of E. lata Infection in Grapevines
J. Agric. Food Chem., Vol. 53, No. 21, 2005 8151
Table 2. Metabolite Production by E. lata Isolates Grown on Grape
Variety Cane Hot Water Extracts with 1% Added Sucrose at 30 Days
Growth Stage
metabolite yieldsa
(µg/mL) for fungal isolate
medium
metabolite
B001 D010 E002 N04 E113 E190 E203 E206
Susceptible Varieties
Cabernet Sauvignon
eutypinol, 1
eulatachromene, 2
methyleutypinol, 7
eutypine, 3
5.6 16.4
1.7
0.2
0.1
−
5.8
0.6
4.6
−
4.8 17.6
3.3 40.8
0.2
0.1
−
1.1
7.6
−
0.5
0.2
−
1.0
0.2
−
0.3
0.4
−
3.1
4.3
−
Sauvignon Blanc
eutypinol, 1
eulatachromene, 2
methyleutypinol, 7
eutypine, 3
7.0
0.3
0.2 11.1
7.8
0.9
3.2
0.2
0.3
−
0.1
−
−
6.2 19.3 23.6 10.5
0.2
1.5
tr
0.8
0.4
0.1
1.1
7.9
tr
1.1
3.3
tr
Figure 3. HPLC analysis of metabolites produced by E. lata isolate E206
−
tr
−
grown on Merlot cane hot water extract supplemented with 1% sucrose.
Syrah
eutypinol, 1
6.8
0.4
0.6
−
5.6
0.9
2.1
−
1.1
tr
0.2
−
0.2
0.2
0.8
−
5.2 11.3 11.2
9.0
1.5
4.1
−
to pH 10 with NaOH, methanol, acetone, and dioxane (all 500 mL)
followed by filtration (Whatman no. 4). The water-based extracts were
eulatachromene, 2
methyleutypinol, 7
eutypine, 3
0.2
3.8
−
0.8
0.4 11.0
−
1.0
−
extracted with Et
extracts under reduced pressure and the extracts resuspended in water
and extracted with Et O as above. The organic phases were collected,
and the Et O was removed under reduced pressure. The residues were
2
O (500 mL). Solvents were removed from the organic
Tolerant Varieties
2
Merlot
eutypinol, 1
eulatachromene, 2
methyleutypinol, 7
eutypine, 3
15.3 17.9
0.8 1.3
0.4 11.1
1.8
0.1
0.2
−
1.0
0.1
2.5
−
0.6 20.6
7.2 21.0
2
0.6
0.1
−
0.8
0.2
−
0.7
1.8
−
2.0
9.2
−
dissolved in MeOH (1 mL) and filtered through 13 mm, 0.45 µm, nylon
syringe filters (Gelman) followed by HPLC analysis.
−
−
Preparation, Culture, and Analysis of Cordon Sections. Highly
symptomatic E. lata-infected cordons (six samples) and nonsymptomatic
cordons (three samples) from Zinfandel grapevines, selected by visible
inspection for evidence of dieback, were sectioned into 1 cm slices
with a handsaw and debarked with a chisel. Cordon sections were
dipped in 70% EtOH, flame sterilized, and placed in 100 mm × 50
mm crystallizing dishes containing 30 mL of sterile water. The dishes
were covered with foil and incubated at 22 °C for 5 days. The water
was removed from the crystallizing dish and extracted with 50 mL of
Semillon
eutypinol, 1
eulatachromene, 2
methyleutypinol, 7
eutypine, 3
2.4 10.7
0.2 1.0
0.1 12.8
1.0
0.2
0.2
−
0.6 20.6
6.0
0.6
0.2
−
9.9 16.5
0.2
1.8
−
0.6
2.3
−
0.5
2.1
−
1.4
4.8
−
−
−
a
−, none detected; tr, above detection limit but <0.1 µg/mL.
whereas the other did not correspond to any metabolites
available to us; eutypine, the putative toxin, was produced solely
on Sauvignon Blanc and then only in trace amounts. The
unidentified metabolite eluted at a retention time ∼3 min later
than that of eutypinol but had a similar UV spectrum, suggesting
that it was a less polar derivative of eutypinol. Methyleutypinol,
2 2
Et O. The organic phase was collected and the Et O removed under
reduced pressure. The residue was dissolved in 1 mL of MeOH, filtered
through a 0.45 µm syringe filter (Gelman), and analyzed for metabolites
by HPLC.
RESULTS AND DISCUSSION
7
, has been isolated previously from cultures of E. lata (13),
In previous work we established that 11 isolates of E. lata
obtained from grapevine, apricot, and oak in California, France,
Italy, Australia, and New Zealand consistently produced higher
levels of metabolites when grown on a hot water extract of
Cabernet Sauvignon supplemented with 1% sucrose than on four
artificial media (potato dextrose broth, PDB; malt yeast broth,
MYB; Pezet’s; and Vogel’s) (11). The sucrose-fortified medium
was chosen to represent a nutrient source as similar as possible
to that present in planta at the time of the early-season flush of
growth in grapevines. Cabernet Sauvignon was selected because
it is a variety of great economic value, generally regarded as
being particularly susceptible to Eutypa dieback (3). The ability
to produce fungal metabolites on grapevine extract medium
provided an opportunity to compare metabolite profiles and
yields on extracts from varieties susceptible and tolerant to the
disease in order to determine whether there could be constituents
in the tolerant varieties that suppressed toxin production. In this
work, eight E. lata isolates, all isolated from grapevines, were
therefore grown on Cabernet Sauvignon, Sauvignon Blanc, and
Syrah (susceptible) extracts and on Merlot and Semillon
and this compound therefore seemed to be the most likely
candidate for the unidentified metabolite. The synthesis of
methyleutypinol via methyleutypine, 8, another known metabo-
lite, was reported by Defranq et al. (13), and both of these
compounds were prepared by using the same general approach,
but with some slight modifications of the route (Figure 2). IR
1
and H NMR spectra for methyleutypinol and methyleutypine
13
were consistent with literature data (13), and the C NMR data,
which had not previously been reported, also supported the
structures. HPLC comparison of synthetic methyleutypinol with
the unknown metabolite showed that these were identical, having
the same retention times and UV spectra.
Table 2 shows the levels of production of metabolites for
the eight E. lata isolates on five different grape variety cane
extracts. No obvious correlations could be made for metabolite
levels produced on either susceptible or tolerant grape varieties.
For example, the most productive isolate, E206, produced 48.2
µg/mL total metabolites on Cabernet Sauvignon extract but only
14.9 and 14.6 µg/mL, respectively, on Sauvignon Blanc and
Syrah extracts, all of which are regarded as susceptible grape
varieties. In contrast, the metabolite levels were 32.2 and 22.7
µg/mL, respectively, on the tolerant varieties Merlot and
Semillon. The only consistent feature was the production of
eutypinol as the principal metabolite, with methyleutypinol and
eulatachromene in lesser amounts. Eutypinol and methyleuty-
pinol, which are obviously structurally closely related, together
(tolerant) extracts, respectively. The culture filtrates were
analyzed by HPLC, and a representative analysis for isolate
E206 grown on Merlot extract is shown in Figure 3. In all
isolates and grapevine extracts, eutypinol was consistently
produced in largest amounts, together with smaller quantities
of two additional metabolites, one of which was eulatachromene,

8152 J. Agric. Food Chem., Vol. 53, No. 21, 2005
Mahoney et al.
Figure 4. Metabolite production on Cabernet Sauvignon cane hot water extract, supplemented with sucrose or glucose, by E. lata isolates isolated from
Merlot (8M, 21M) or Cabernet Sauvignon (7CS, 24CS) grapevines.
Figure 5. Time course for eutypinol, 1, production on Cabernet Sauvignon cane hot water extract, supplemented with sucrose or glucose.
comprised 83-100% of the total metabolites, with the exception
of isolate E113 grown on Merlot extract, which had a
comparable amounts of eutypinol and eulatachromene.
Sauvignon hot water extract supplemented with sucrose or
glucose. Although levels of metabolites varied with strain, the
general pattern of metabolites was similar, with eutypinol again
predominating (Figure 4); supplementation with sucrose stimu-
lated metabolite production somewhat more than glucose
supplementation. A time course study (Figure 5) showed that
there was very little eutypinol production during the first 20
days of the culture period but that production increased rapidly
thereafter, attaining levels as high as 27 µg/mL for one strain
isolated from Cabernet Sauvignon at the end of the experiment
on day 57, whereas another strain from the same grape variety
produced <1 µg/mL. The isolates from Merlot had intermediate
eutypinol production levels, ranging from 2 to 18 µg/mL. The
Because differences in metabolite production did not correlate
with the variety of grapevine from which cane extracts were
derived, the time course and profiles of metabolites produced
by E. lata isolates isolated from Cabernet Sauvignon and Merlot
grown on identical media were examined to determine the effect
of fungal source. Cabernet Sauvignon is generally accepted to
be the more susceptible cultivar, so it was hypothesized that
isolates from the latter would produce a different pattern of
metabolites than those from Merlot; this proved not to be the
case. Two sets of media were tested, consisting of Cabernet

Diagnosis of E. lata Infection in Grapevines
J. Agric. Food Chem., Vol. 53, No. 21, 2005 8153
long time period prior to biosynthesis of significant amounts
of the metabolite and subsequent rapid increase suggest that
either an unidentified precursor must first attain a particular level
before eutypinol production is initiated or that eutypinol
biosynthesis is stimulated by nutrient depletion. The latter would
appear to be more likely and may have important implications
for the situation in grapevines, where cordon wood is likely to
be limited in nutrients relative to more actively growing tissues.
These results indicate that although eutypinol is the primary
metabolite, its level is dependent on the specific isolate of E.
lata, suggesting that some isolates can be highly pathogenic,
whereas others may be relatively innocuous, assuming this
compound is a marker of pathogenicity. It is noteworthy that
only the isolates from Merlot produced any eutypine, also
supporting the hypothesis of variability in pathogenicity of E.
lata isolates. Nevertheless, correlation between quantitative and/
or qualitative production of secondary metabolites and phyto-
toxicity still has to be demonstrated.
Because the fungus is restricted to the perennial wood of the
host plant (14), it has been postulated that fruiting spurs and
stems are damaged by translocation of toxins (13), although no
evidence has been advanced to support this. We therefore
undertook analysis of grapevine tissues exhibiting clearly visible
signs of Eutypa dieback for typical metabolites as well as some
control samples showing no symptoms of the disease. HPLC
analysis of the methanol extract of ground plant material was
performed on cordon, spurs, grape canes, young shoots, inflo-
rescences, and fruit bunches from Cabernet Sauvignon, Merlot,
and Zindfandel varieties. No E. lata metabolites were detected
in a total of 39 samples analyzed (results not shown), even
though the detection limit (0.01 µg/mL) for these compounds
by the HPLC method was very low. To ensure that the absence
of metabolites was not due to an inability to extract them from
the plant matrix with methanol, other solvents were used. These
encompassed a range of polarities and included water, acetone,
and dioxane; again, no metabolites were detected. It is possible
that the metabolites could be converted into glycosides or other
derivatives in planta, compounds which would not be formed
by culturing the fungus alone. In the event of such transforma-
tions, such derivatives might not be extractable from plant
samples with organic solvents, and in order to deal with this
eventuality, the plant material was subjected to treatment with
either acid or base prior to extraction; once again no metabolites
were detectable.
Figure 6. Cordon sections cultured in water at 22
symptomatic cordon, 14 days; (B) nonsymptomatic cordon, 60 days.
°C: (A) highly
removed with a chisel, and the wood samples were briefly
surface-sterilized in a flame. The samples were partially
immersed in water and kept at 22 °C. After 5 days, fungal
growth was visible to a greater or lesser extent on all samples
that had shown signs of dying-arm disease, and even earlier on
some samples, but not on samples with no symptoms of the
disease. After 14 days, the symptomatic cordons showed
extensive fungal growth (Figure 6A), whereas the nonsymp-
tomatic cordons showed no fungal growth even after 60 days
These results indicate that translocation of metabolites from
the site of fungal biosynthesis, with direct effects on vegetative
tissues, does not occur. In retrospect this is not surprising,
because if the compounds are sufficiently reactive to damage
plant tissues, then such phytotoxicity should be just as likely in
the vicinity of fungal infection. Furthermore, the toxins probably
react in such a way that they are bound irreversibly, so that it
is unlikely that appreciable amounts of metabolites are present
at any particular time. As phenolic compounds, the metabolites
would be susceptible to oxidative polymerization by plant phenol
oxidases, possibly accounting for the dark, wedge-shaped areas
typical of E. lata infection. Such a scenario renders the detection
of metabolites in planta as a surrogate for identification of E.
lata infection highly unlikely.
(Figure 6B). Extraction of aliquots of the aqueous media with
diethyl ether after culturing for 5 days and analysis by HPLC
showed the presence of eutypinol, 1, and methyleutypinol, 7
(Figure 7). The presence of these metabolites correlated with
visible evidence of pathogenic E. lata in the tissues, whereas
the control samples showed no eutypinol or any other metabo-
lites of similar structure. In addition, a major peak was also
observed at 16.63 min in some samples. Although this constitu-
ent had a UV spectrum typical of other E. lata compounds, it
did not correspond to any of the metabolites produced in culture
and may therefore be an artifact produced by the action of plant
enzymes. Culturing of larger amounts of cordon tissue will be
A different approach was therefore adopted, in which cordon
tissue was cultured in water to initiate fungal growth and the
aqueous solution analyzed for the presence of metabolites. Cross
sections ∼1 cm thick were cut from cordons that had been
classified as infected or uninfected by experienced vineyard
managers on the basis of visible symptoms; the bark was

8154 J. Agric. Food Chem., Vol. 53, No. 21, 2005
Mahoney et al.
likely to provide a much more representative picture of the
metabolite composition in planta. A recent publication (22) has
described similar findings when metabolites produced by
Penicillium species responsible for storage rot in root vegetables
and flower bulbs were compared for artificial media and plant
material agars.
The techniques developed in this exploratory study can be
performed in any suitably equipped analytical chemistry labora-
tory and do not require expert knowledge of fungal taxonomy.
We plan to investigate the metabolite profiles of other Eutypa
species by HPLC analysis to determine whether they differ
significantly from that of E. lata. Moreover, Botryosphaeria
species produce symptoms remarkably similar in appearance
to those of Eutypa dieback in grapevines, almonds, pistachios,
and walnuts, including the wedge-shaped necrosis of trunks and
cordons (23, 24). The metabolites produced by Botryosphaeria
on grapevine media have not been identified; it will be of interest
to determine whether they are structurally related to those of
E. lata. Such information should permit discrimination of Eutypa
dieback from other grapevine trunk diseases, without the need
to rely upon symptomatic descriptors, and also could establish
the structural features essential for phytotoxicity.
Figure 7. HPLC analysis of highly symptomatic and nonsymptomatic
cordon culture medium (water) after 5 days.
required to isolate sufficient quantities of this compound for
structural identification.
The metabolite profiles produced by the limited number of
E. lata isolates examined in this study, their heterogeneity with
growth media and time, and potential for structural transforma-
tion within the grape plant matrix suggest that phytotoxicity is
probably a consequence of a suite of compounds rather than
eutypine alone. In fact, eutypine was a rarely produced
metabolite in all of our studies and notable by its absence from
most E. lata isolates grown on grape cane extracts. Identification
of the complete range of metabolites produced by analysis of a
larger number of E. lata isolates, representative of those parts
of the world where the disease is a problem for grape producers,
and assessment of the relative phytotoxicity of each using a
standardized grape-tissue bioassay should permit decisions to
be made as to the economic impact in any particular situation.
The cordon culturing procedure and HPLC analysis of the
medium provides a proof-of-concept that the technique could
be applied to experimental studies of transmission and spread
of the disease and to ensure that tissues used for grafting are
not infected. It could also be used to ensure that reworked trunks
to which tissues are being grafted are free of E. lata. Improved
cordon sampling methods, such as removal of smaller tissue
samples by coring or plug cutting would permit high-throughput
testing of numerous samples, without sacrificing the entire
cordon. Appropriate management techniques can then be devised
to limit the spread of the most virulent isolates and ensure the
continued productivity of individual vineyards.
The analytical techniques developed in the course of this
research should be applicable to the detection of specific
phytotoxic metabolites in grapevines infected by E. lata, possibly
before visible symptoms of dying-arm disease develop. Although
a PCR assay for the fungus in grapevine wood has been
developed (15), it does not distinguish between pathogenic and
nonpathogenic isolates, nor does it provide a measure of the
degree of pathogenicity. The primers designed for ITS sequences
used by Lecomte et al. (15) have now been shown not to be
specific for E. lata (16), and Lardner et al. (17) have recently
demonstrated that the SCAR primer pairs were also not specific
to Australian populations of E. lata. Furthermore, several
diatrypaceous species occur on grapevines. A Diatrypella
species was recently recovered from wood canker of grapevines
(unpublished observations), Rohlshausen et al. (18) confirmed
the presence of a Diatrype species on grapevines, and Trouillas
et al. (19) also found the teleomorphs of Diatrype and
Diatrypella species on dead wood of grapes. Moreover, Eutypa
leptoplaca (20) and CryptoValsa ampelina (21) have been
reported to be pathogenic on grapes. The presence of several
related taxa on grapevine indicates that particular care has to
be taken for positive identification of the causal agent respon-
sible for disease. Disease diagnosis is commonly done by
isolating the organism from wood canker on PDA medium, but
because of the morphological resemblance of this group of fungi
in culture, specialized mycological expertise is required and
misidentification can occur (16). Further research is required
to establish species delimitation within the Diatrypaceae to better
identify the organisms causing disease, and metabolite analysis
of cultured grapewood samples as developed herein should
enable the presence of E. lata to be distinguished from other
diatrypaceous species, because it has been shown that production
of such secondary metabolites is a genetically derived character
ACKNOWLEDGMENT
We gratefully acknowledge Rosalind Wong, Western Regional
Research Center, Albany, CA, for obtaining NMR spectra. We
thank Dr. Richard Lardner, University of Adelaide (Waite
Campus), Glen Osmond, South Australia, for providing selected
E. lata isolates. We greatly appeciate the cooperation of Andrew
C. Johnson, Beringer Blass Wine Estates, St. Helena, CA, in
providing grape cane material for preparation of hot water
extract nutrient solutions.
(18). This is supported by the fact that 2 of the 11 isolates in
our previous study, E178 and SS1#1, which did not produce
metabolites (11), have subsequently been classified on the basis
of PCR amplification of DNA as Diatrype and CryptoValsa
species, respectively (17). It is important to note that in our
previous study (11) and in the current work the metabolite yields
are much higher and profiles quite different when using grape
cane extracts from those obtained using artificial media (10).
Use of grape cane extract for in vitro experiments is therefore
LITERATURE CITED
(
1) Carter, M. V.; Bolay, A.; Rappaz, F. An annotated host list and
bibliography of Eutypa armeniacae. ReV. Plant Pathol. 1983,
62, 251-258.

Diagnosis of E. lata Infection in Grapevines
J. Agric. Food Chem., Vol. 53, No. 21, 2005 8155
(
2) Munkvold, G. P.; Duthie, J. A.; Marois, J. J. Reductions in yield
and vegetative growth of grapevines due to Eutypa dieback.
Phytopathology 1994, 84, 186-192.
(14) Munkvold, G. P. Eutypa dieback of grapevine and apricot. Plant
Health Prog. on-line doi:10.1094/PHP-2001-0219-01-DG.
(15) Lecomte, P.; P e´ ros, J.-P.; Blancard, D.; Bastien, N.; D e´ lye, C.
PCR assays that identify the grapevine dieback fungus Eutypa
lata. Appl. EnViron. Microbiol. 2000, 66, 4475-4480.
(
3) Munkvold, G. P.; Marois, J. J. Factors associated with variation
in susceptibility of grapevine pruning wounds to infection by
Eutypa lata. Phytopathology 1995, 85, 249-256.
(
16) Rolshausen, P. E.; Trouillas, F. P.; Gubler, W. D. Identification
of Eutypa lata by PCR-RFLP. Plant Dis. 2004, 88, 925-929.
17) Lardner, R.; Stummer, B. E.; Sosnowski, M. R.; Scott, E. S.
Molecular identification of Eutypa lata in grapevine. Mycol. Res.
(
4) Siebert, J. B. Eutypa: the economic toll on vineyards. Wines
Vines 2001, 50-56.
(
(
5) Renaud, J. M.; Tsoupras, G.; Tabacchi, R. Biologically active
natural acetylenic compounds from Eutypa lata (Pers: F.) TUL.
HelV. Chim. Acta 1989, 72, 929-932.
2
005, 109, 1-10.
(
18) Rolshausen, P. E.; Mahoney, N.; Molyneux R. J.; Gubler, W.
D. A re-assessment of the species concept in Eutypa lata, the
causal agent of Eutypa dieback of grapevine. Phytopathology
(
6) Renaud, J. M.; Tsoupras, G.; Stoecklievans, H.; Tabacchi, R. A
novel allenic epoxycyclohexane and related compounds from
Eutypa lata (Pers: F.) TUL. HelV. Chim. Acta 1989, 72, 1262-
2
005, in press.
1
267.
(
7) Kim, J. H.; Mahoney, N.; Chan, K. L.; Molyneux, R. J.;
Campbell, B. C. Secondary metabolites of the grapevine pathogen
Eutypa lata inhibit mitochondrial respiration, based on a model
bioassay using the yeast Saccharomyces cereVisiae. Curr.
Microbiol. 2004, 49, 282-287.
(19) Trouillas, F. P.; Rolshausen, P. E.; Gubler, W. D. Importance
of Eutypa lata and occurrence of other Diatrypaceous fungi in
northern Californian vineyards. Phytopathology 2001, 91, S89.
(20) Trouillas, F. P.; Gubler, W. D. Identification and characterization
of Eutypa leptoplaca, a new pathogen of grapevine in northern
California. Mycol. Res. 2004, 10, 1195-1204.
(
8) Tey-Rulh, P.; Philippe, I.; Renaud, J. M.; Tsoupras, G.; Dean-
gelis, P.; Fallot, J.; Tabacchi, R. Eutypine, a phytotoxin produced
by Eutypa lata the causal agent of dying-arm disease of
grapevine. Phytochemistry 1991, 30, 471-473.
(
(
(
21) Mostert, L.; Halleen, F.; Creaser, M. L.; Crous, P. W. CryptoV-
alsa ampelina, a forgotten shoot and cane pathogen of grape-
vines. Australas. Plant Pathol. 2004, 33, 295-299.
22) Overy, D. P.; Blunt, J. W. Corymbiferan lactones from Peni-
cillium hordei: stimulation of novel phenolic metabolites using
plant tissue material. J. Nat. Prod. 2004, 67, 1850-1853.
23) Michailides, T. J.; Morgan, D. P. Panicle and shoot blight of
pistachio: A major threat to the California pistachio industry.
APSnet: Plant Pathology Online 2004; www.apsnet.org/online/
feature/pistachio (accessed Feb 3, 2005).
(
9) Guillen, P.; Guis, M.; Martinez-Reina, G.; Colrat, S.;
Dalmayrac, S.; Deswarte, C.; Bouzayen, M.; Roustan, J. P.;
Fallot, J.; Pech, J. C.; Latche, A. A novel NADPH-dependent
aldehyde reductase gene from Vigna radiata confers resistance
to the grapevine fungal toxin eutypine. Plant J. 1998, 16, 335-
3
43.
(
(
(
(
10) Molyneux, R. J.; Mahoney, N.; Bayman, P.; Wong, R. Y.; Meyer,
K.; Irelan, N. Eutypa dieback in grapevines: differential produc-
tion of acetylenic phenol metabolites by strains of Eutypa lata.
J. Agric. Food Chem. 2002, 50, 1393-1399.
11) Mahoney, N.; Lardner, R.; Molyneux, R. J.; Scott, E. S.; Smith,
L. R.; Schoch, T. K. Phenolic and heterocyclic metabolite profiles
of the grapevine pathogen Eutypa lata. Phytochemistry 2003,
(24) van Niekerk, J. M.; Crous, P. W.; Groenewald, J. Z.; Fourie, P.
H.; Halleen, F. DNA phylogeny, morphology and pathogenicity
of Botryosphaeria species on grapevines. Mycologia 2004, 96,
781-798.
6
4, 475-484.
12) Smith, L R.; Mahoney, N.; Molyneux, R J. Synthesis and
structure-phytotoxicity relationships of acetylenic phenols and
chromene metabolites, and their analogs, from the grapevine
pathogen Eutypa lata. J. Nat. Prod. 2003, 66, 169-176.
13) Defranq, E.; Zesiger, T.; Tabacchi, R. The synthesis of natural
acetylenic compounds from Eutypa lata (Pers: T.) Tul. HelV.
Chim. Acta 1993, 76, 425-430.
Received for review May 3, 2005. Revised manuscript received July
28, 2005. Accepted August 11, 2005. This research was conducted under
a Cooperative Agreement between USDA-ARS (No. 58-5325-0-158) and
the American Vineyard Foundation (Project V200); we thank the AVF
for financial support.
JF0510236