ARTICLE IN PRESS

Physiological and Molecular Plant Pathology 69 (2006) 209–223

www.elsevier.com/locate/pmpp

Effects of three esca-associated fungi on Vitis vinifera L.: I.

Characterization of secondary metabolites in culture media and
host responses to the pathogens in calli
Giovanni Bruno, Lorenzo Sparapano
Dipartimento di Biologia e Patologia vegetale, University of Bari, Via G. Amendola 165/A, 70126 Bari, Italy
Accepted 30 April 2007

Abstract

Phaeomoniella chlamydospora (Pch) and Togninia minima (Tmi) produced scytalone, isosclerone and pullulans in liquid cultures, as
well as in calli. Secondary metabolites and host defense compounds were shown to occur in calli of Vitis vinifera cv. Italia and cv. Matilde
infected by the esca-associated fungi Pch, Tmi and Fomitiporia mediterranea (Fme). Calli of both cvs. were grown as dual cultures with
Pch, Tmi and Fme. The fungi grew well in the presence of calli of both cvs., but callus growth was reduced. Accumulation and changes of
total phenolics and recurring phenolics, and of two phytotoxic pentaketides and pullulans were analyzed by HPLC. On comparing
results for cv. Italia and cv. Matilde, it can be seen that concentrations of phenolics are strongly related to the cv. The paper discusses the
possible relationship between melanin biosynthesis in Pch and Tmi, which utilize pentaketide metabolites as intermediates and their
pathogenicity related to phytotoxity of scytalone and isosclerone.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Vitis vinifera; Phaeomoniella chlamydospora; Togninia minima; Fomitiporia mediterranea; Esca syndrome; Plant–pathogen interactions; Calli;
Phenolics; Phytoalexins; Pentaketides; Pullulans

1. Introduction possible combinations on the spurs, branches or trunks of

Esca of grapevine (Vitis vinifera L.) is a complex disease
whose symptoms may arise from the combined effect of
several pathogenic and physiological factors [1–5]. The
consistent isolation of the ascomycete Togninia minima
(Tul. and C. Tul.) Berl. (anamorph: Phaeoacremonium
aleophilum W. Gams, P.W. Crous, M.J. Wingfield and
L. Mugnai) (Tmi), the anamorphic ascomycete Phaeomo-
niella chlamydospora P.W. Crous and W. Gams (Pch) and
the basidiomycete Fomitiporia mediterranea M. Fisch.
(Fme) from discolored or decayed wood of esca-diseased
grapevines indicates a close relationship between those
fungi and particular stages of wood, leaf and berry
deterioration [2,4,6,7]. Pathogenicity of the three fungi
was confirmed by inoculating them singly and in all
Corresponding author. Tel.: +39 080 5443085; fax: +39 080 5442906.
E-mail address: sparlor@agr.uniba.it (L. Sparapano).
standing esca-free cv. Italia and cv. Matilde vines. All three
fungi were able to infect the vines through wounds, to
colonize and degrade the woody tissue and, to cause
disease symptoms very similar to those of esca [3,4,8]. The
results of recent studies provided new information on the
production of toxic metabolites by Pch and Tmi and host
response compounds in naturally infected vines, indicating
that the pathogens, their by-products and defense sub-
stances are translocated from the infected woody tissue of
the trunk to the aerial part of the affected vines [9–11].
In many fungal diseases, phytotoxins are important
virulence factors. Toxins can be host-specific or non-host
specific. Phytotoxins are secondary metabolites, which
have been isolated from artificial media rather than natural
substrates, usually as a consequence of their accumulation
in large and easily detected amounts. Pch and Tmi
produced two pentaketides (scytalone and isosclerone)
[12,13] and the a-glucan named pullulan [14]. These

0885-5765/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pmpp.2007.04.008
 ARTICLE IN PRESS
210 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
findings suggested that at least some of these metabolites
may be involved in the pathogenesis and symptom
expression of esca and related syndromes on grapevine. It
would be interesting to know more about the roles played
by those toxins, which are primary virulence factors, and
the advantages (if any) they confer on the fungus which
produces them.
It is now widely accepted that pullulan is a linear
polysaccharide with malthotriosyl repeating units joined by
a-(1 ) 6)-linkages. Alternatively, the structural formula of
pullulan may be presented as a regular sequence of panoses
bonded by a-(1 ) 4)-linkages [15]. The black yeast
Aureobasidium pullulans (de Bary) G. Arnaud is widespread
in all ecological niches including forest soils, fresh and sea
water, plant and animal tissues, and is the main producer
of pullulan [15]. A. pullulans is classified as a non-
pathogenic microorganism; however, more recently the
plant pathogenic fungi Cryphonectria parasitica (Murr.)
Barr, P. chlamydospora and T. minina have also proved to
be pullulan producers [14,16–19].
Resistance of plants to infection by plant pathogens is
the result of multiple defense reactions comprising both
constitutive and inducible barriers [20,21]. Active defense
mechanisms mainly involve the accumulation of phytoa-
lexins [22], rapid and localized cell death [23], synthesis of
pathogenesis-related proteins [24], and reactive oxygen
species (ROS) produced in the plant oxidative burst [25,26].
After infection, grapevines synthesize several compounds
such as phytoalexins (e.g., the viniferins), phenolics and
glycolic acid [9,27,28]. Synthesis and localization of
these phenolics appear to be primarily associated with
resistance against pathogens. Phytoalexins from the
Vitaceae seem to constitute a rather restricted group of
molecules belonging to the stilbene family, the skeleton of
which is based on the trans-3,5,40 -trihydroxystilbene
structure [24]. Flavonoids, phenolics with nuclei arranged
in a C6–C3–C6 configuration, are a diverse group of natural
plant products, which play important roles in growth and
development, and in defense against microorganisms and
pests [29,30]. They are one of the largest groups of
naturally occurring phenolics, which are likely to be
encountered in any plant extract. Most flavonoids are
glycosylated, and glycosylation makes them less reactive
and more water soluble. Because of their in vitro
antimicrobial activity, specific classes of flavonoid and
isoflavonoid compounds have long been thought to play a
role in plant–pathogen interactions as part of the host-
plant defense arsenal [31].
In the present paper, much attention has been paid to the
isolation of phytotoxins (pentaketides and a-glucans)
produced by Pch and Tmi in batch cultures and in the
infected calli of grapevine cv. Italia (more susceptible) and
cv. Matilde (less susceptible). Moreover, the paper
describes in vitro experiments on possible regulatory factors
of the accumulation of phenolics in calli. This paper infers
that the pentaketide pathway may be a wide-spread
mechanism for the formation of fungal melanin.
2. Materials and methods
2.1. Fungal strains and culture conditions
Stock cultures of P. chlamydospora (Pch) strain PVFi56
(CBS 229.95), T. minima (Tmi) strain PVFi69 (CBS 631.94)
and F. mediterranea (Fme) strain Fop1 isolated from
grapevines in Italy were maintained on slants of malt-agar
(MA) or potato-sucrose-agar at 4 1C. The three strains
have been used since the pathogenicity trials carried out in
1999 [3,5,8].
All the strains were grown in stationary cultures in 1 litre
Roux flasks containing 150 ml Czapek medium amended
with 0.1% yeast and 0.1% malt extract (pH 5.7). Each flask
was seeded with 5 ml of a suspension of three 10-day-old
cultures in 50 ml sterile water. The flasks were incubated in
the dark at 25 1C for 28 days. At harvest, the mycelial
mat was removed by filtration on Miracloth (Calbiochem,
La Jolla, CA, USA).
2.2. Dual cultures: fungus/callus
Explants of leaf petioles and young shoot internodes
from cv. Italia and cv. Matilde grapevines were excised and
cultivated both in Petri dishes and in Magenta vessels
containing respectively 20 ml or 40 ml of a medium (LSG1)
obtained by modifying Linsmaier and Skoog’s medium [32]
previously reported by Sparapano et al. [33,34]. Callus used
for the experiments was produced after two transfers of 60
days each.
Pch, Tmi and Fme strains were grown in the dark at
25 1C for 2 weeks in Petri dishes containing MA. Plugs
(3  3 mm) were aseptically removed from the margin of
actively growing colonies and were transferred singly to
Magenta vessels containing LSG1 medium with cv. Italia
or cv. Matilde calli [34]. The plugs were placed in Magenta
vessels parallel to the callus at a distance of 2 cm. All the
Magenta vessels were sealed with Parafilm M (American
National Can, Chicago, IL), and incubated in the dark at
25 1C. The experiment was carried out three times with
three replicates per fungus. The growth of the colonies was
measured every 4 days for 2 months. Calli were harvested
when completely colonized by each fungus, their fresh wt
was determined and then they were stored at 80 1C.
2.3. Extraction, purification and identification of phytotoxic
pentaketides
2.3.1. Batch cultures
The culture filtrates (3 litres per strain) were brought to
pH 4 with 1 N HCl and extracted four times with ethyl
acetate (1.5 litre each). The combined organic extracts were
dried on anhydrous sodium sulfate and evaporated under
reduced pressure to give a dark-brown oily residue.
Chromatographic separation was performed on silica gel
plates (Merk F254, 0.50 mm, 20  20 cm) using a mixture of
chloroform and ethyl acetate (3:1; v/v) as an eluent. Each
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
band was scraped off and dissolved in ethyl acetate. The
solution was first concentrated under reduced pressure and
then lyophilized. Scytalone crystallized as white prisms
from a benzene:diethyl ether (95:5; v/v) mixture. The crude
residue was dissolved in methanol:water (1:1; v/v), and
filtered through a 0.22 mm Millex-FG unit (Millipore,
Bedford, MA, USA). It was purified and analyzed by
HPLC following the procedures of Evidente et al. [12],
Malovaná et al. [35], and Rodrı́guez-Delgado et al. [36].
HPLC analysis of pentaketides extracted from batch
cultures was performed using a Waters liquid chromato-
graph (Milford, MA, USA) equipped with two pumps
(Model 510), an automated gradient controller (Model
680), an injector (Rheodine Model 7135), a tunable
absorbance detector (Model 481) and a data module
integrator (Model 740). The analytical column was a
Nova-Pak C18 150  3.9 mm i.d., 4 mm particle diameter
from Waters. A Nova-Pak pre-column was employed to
protect the analytical column, and a wavelength of 280 nm
was used for absorbance detector. Phytotoxins used as
standards were scytalone and isosclerone [12]. Chromato-
graphic peaks were identified by comparing retention times
of samples with those of standards, and quantification was
carried out by internal standardization.
2.3.2. Grapevine calli
Samples (100 g each) of frozen cv. Italia and cv. Matilde
calli previously grown with Pch, Tmi and Fme, were added
to a sterile distilled water in the ratio of 1 g callus fresh wt
and 2 ml of distilled water. Calli grown without any fungus
were used as controls. Callus suspension was homogenized
in a Waring Blendor and shaken for 30 min. The solid
material was discarded after centrifugation (3000g, 4 1C,
15 min) and then the resulting liquid was filtered through a
0.45 mm Millipore membrane and adjusted to pH 4 by
adding small amounts of 0.1 N HCl. Then it was extracted
with ethyl acetate (5  200 ml) following the procedures
suggested by Evidente et al. [12]. HPLC analysis was
carried out as described in Section 2.3.1. Chromatographic
peaks were identified by comparing retention times of
samples with those of standards, and quantification was
carried out by internal standardization.
2.4. Extraction, purification and chemical analysis of
pullulans
2.4.1. Batch cultures
The culture filtrates from Pch and Pal (2 litres per strain)
were treated with two volumes of cold absolute ethanol
following the procedure described in previous papers
[18,19]. The resulting precipitates were filtered through
Whatman GFC filters, dried and weighed.
HPLC analysis was performed using the same Waters
liquid chromatograph mentioned in Section 2.3.1 equipped
with a differential refractometer mod. 410. Chromato-
grams were recorded and analyzed on a Waters 740 Data
module. The column was a 1000 Å Nucleogel GFC
211
(300  7.7 mm I.D., Macherey-Nagel, Düren, Germany).
Elution was carried out at room temperature under
isocratic conditions with pure water which had been
filtered through Millipore membrane filter (0.45 mm), de-
aerated and protected from microbial contamination with
0.05% sodium azide, and finally delivered to the column at
a flow rate of 1 ml min1. Before injection, reference
compounds, crude EPS and purified samples of pullulan
were dissolved in water and filtered through a SPE tube
(LC-4,500 Å pores, Supelco, Bellefonte, PA, USA) and
then through a Millex FG (0.22 mm, Millipore). Chromato-
graphic peaks were identified by comparing retention times
of samples with those of standards. Quantification was
carried out by internal standardization and the IR spectra
were determined with a Perkin–Elmer mod. 1720 spectro-
meter, using the potassium bromide technique [37].
The mol. wt of purified samples of pullulan was estimated
by gel permeation chromatography (GPC) with the same
column used for their separation by HPLC. The column was
calibrated using pullulan standards (Macherey-
Nagel; Sigma-Aldrich Chemicals, Inc, St Louis, MO,
USA; and Hayashibara, Okayama, Japan) of known mol.
wt, ranging from 5.8 to 8.5  103 kDa, and dextran
standards (Sigma-Aldrich) from 9.3 to 2  103 kDa. The
reference pullulan (Sigma-Aldrich) was from A. pullulans.
2.4.2. Grapevine calli
Samples (100 g each) of frozen cv. Italia and cv. Matilde
calli previously grown as dual culture with Pch or Tmi or
Fme, were added to sterile distilled water in the ratio of 1 g
callus fresh wt and 2 ml of distilled water. Calli grown
without any fungus were used as controls. Callus suspen-
sion was homogenized in a Waring Blendor and shaken for
30 min. The solid material was discarded after centrifuga-
tion (3000g, 4 1C, 30 min) and the supernatant liquid was
filtered through a 0.45 mm Millipore membrane, then mixed
with three volumes of cold absolute ethanol and left
overnight at 20 1C [9]. The resulting precipitate was
collected by centrifugation (5000g, 4 1C, 30 min), dissolved
in ultrapure Milli-Q water (100 ml), and re-precipitated
with cold absolute ethanol (300 ml) as described above.
After 24 h, the resulting precipitate was collected by
centrifugation (5000g, 4 1C, 30 min) and the precipitate
was lyophilized to yield native a-glucans. Chemical analysis
was performed following the same procedures mentioned
above in Section 2.4.1.
2.5. Grapevine bioassay
Scytalone, isosclerone and pullulan obtained from liquid
cultures of each fungal species (Pch or Tmi) or from calli
inoculated with the same fungi were assayed on detached
leaves of three grapevine cvs.: Italia and Matilde (white
vines) and Sangiovese (red vine). The leaves with their
petioles were immersed in 3 ml toxic solution until
complete absorption, which usually took a few hours and
did not require more than 24 h even with the highest
 ARTICLE IN PRESS
212 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
molecular weight pullulan. After this they were transferred
to distilled water. During the assay, the leaves were kept in
a growth chamber at a relatively low temperature (23 1C),
RH (60%) and illumination (150 mE m2 s1). Toxicity
symptoms were recorded 48 h later.
Scytalone and isosclerone were tested at 50 and 100 mg ml1.
Purified pullulan was assayed at 20 and 100 mg ml1. Control
included distilled water.
2.6. Analysis of total phenolics
Total phenolics present in fungal batch cultures and in
each crude organic extract obtained from calli were
estimated according to the Folin-Ciocalteu method, using
gallic acid as the standard [38].
2.7. Content of recurring phenolics
Recurring phenolics are those which occur repeatedly in
fungal batch cultures and in grapevine calli, such as:
benzaldehyde derivatives (2,5-dihydroxy-benzaldehyde,
protocatechuicaldehyde, syringaldehyde), benzoic acid
derivatives (gallic acid, protocatechuic acid, syringic
acid, vanillic acid), cinnamic acid derivatives (caffeic acid,
p-coumaric acid, ferulic acid), flavonols (apigenin, kaemp-
ferol, myricetin, quercetin, rutin), flavan-3-ol derivatives
[(+)catechin, ()epicatechin], flavonol-3-O-glycosides
(quercetin-3-rhamnoside), and stibenes (trans-resveratrol)
supplied by Sigma-Aldrich.
2.7.1. In fungal batch cultures
Each fungus (Pch, Tmi and Fme) was grown in
stationary cultures following the same procedures reported
in a previous paper [33]. Each flask was seeded with 5 ml of
a suspension from three 10-day-old cultures on MA in
50 ml sterile water. Thirty Roux flasks were inoculated and
incubated in the dark at 25 1C for 28 days. At harvest, the
mycelial mat was first removed by filtration on a Miracloth
and then on a 0.45 mm Millipore membrane. The pooled
culture filtrate (4 litre) was brought to pH 2 with 1 N HCl
and was extracted with diethyl ether (4  2 litres). The
combined organic extracts were dried on anhydrous
sodium sulfate and evaporated under reduced pressure,
yielding a dark-brown oily residue. A suitable aliquot of
the crude residue was dissolved in methanol:water (1:1;
v/v), and filtered through a 0.22 mm Millipore Millex-FG
unit before HPLC analysis.
2.7.2. In grapevine calli
Samples (100–200 g each) of frozen cv. Italia or cv.
Matilde calli previously grown as dual culture with Pch or
Tmi or Fme were added to an acetone:water solution
(3:2; v/v) in the ratio of 1 g of callus fresh wt and 1.5 ml of
acetone:water (1:1; v/v) solution. Calli grown without any
fungus were used as controls. The callus suspension was
homogenized in a rotary blender and shaken for 30 min,
then the solid material was discarded after centrifugation
(3000g, 4 1C, 15 min). The resulting liquid was filtered
through a 0.45 mm Millipore membrane, the acetone was
removed under reduced pressure and the resulting liquid
mixture was adjusted to pH 2 with 0.1 N HCl. It was then
extracted with diethyl ether (5  100 ml) and thereafter with
ethyl acetate (5  100 ml) [35]. The combined organic
extracts were separately dried on anhydrous sodium sulfate
and evaporated under reduced pressure. Each crude
organic extract was solubilized with a small amount of
methanol:water (1:1; v/v) mixture and clarified by filtration
through a 0.22 mm Millipore Millex FG unit before HPLC
analysis.
2.8. HPLC-UV quantification of recurring phenolics
Analysis of recurring phenolics extracted from batch
cultures or grapevine calli was performed using the same
Waters liquid chromatograph mentioned in Section 2.3.1,
equipped with a tunable absorbance detector (Model 481).
The column, the pre-column and the analytical procedures
were those previously used by Malovaná et al. [35] or by
Rodrı́guez-Delgado et al. [36]. Phenolic standard solutions
were prepared just before starting chromatographic analy-
sis. Calibration graphs in the range of 0.02–50 mg ml1 for
every phenolic were prepared by replicate injection of
5–10 ml standard solutions. The dry residue of each organic
extract was dissolved in a 1 ml methanol:water (1:1; v/v)
mixture, clarified by filtration through a 0.22 mm Millex FG
filter and aliquots (up to 10 ml) were injected into the HPLC
system. Chromatographic peaks were identified by com-
paring retention times of samples with those of standards,
quantification was carried out by internal standardization.
3. Results
3.1. Pentaketide production in batch cultures
The oily residue extracts from culture filtrates of Pch or
Tmi analyzed by HPLC gave scytalone and isosclerone as
main secondary metabolites. The pattern of pentaketide
production by Pch and Tmi cultures is shown in Fig. 1A
and 1B. Scytalone began to accumulate after 7 days and
continued to accumulate rapidly for a further 21 days. Tmi
produced scytalone at a greater concentration than Pch,
while Pch produced isosclerone at a greater concentration
than Tmi. Both Tmi and Pch produced less isosclerone than
scytalone, while no pentaketides were produced by Fme.
3.2. Pullulan production in batch cultures
Pullulan was produced in large amount in liquid cultures
by Pch, in smaller amount by Tmi (Fig. 1C) and was not
produced at all by Fme. Regardless of its origin, pullulan
from both fungi was a mixture of a-glucans whose
molecular weights, as determined by GPC, are shown in
Table 1. The pullulan produced by Pch consisted of six
molecular species with a mol. wt ranging from 50 to
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223 213

25 Table 1
A Mean mol. wt (kDa) of pullulan families purified from culture filtrates of
OH O
Phaemoniella chlamydospora and Togninia minima stationary liquid
20 cultures and their concentration (mg ml1)
Concentration (µg ml-1)

HO OH
Molecular weight P. chlamydospora T. minima
15
2500 0.2 0.05
10 2000 0.1 n.d.a
900 0.2 n.d.
400 0.1 n.d.
5 150 0.2 n.d.
100 0.3 n.d.
80 n.d. 0.25
0
50 n.d. 0.25
5
B a
n.d.: no detected.
O OH

4
Concentration (µg ml-1)

becoming coalescent or diffused to large areas were

3 produced on the leaf lamina. Similarly, isosclerone assayed
HO
at 100 mg ml1 caused large yellowish spots, slowly becom-
2 ing coalescent and necrotic, followed by distortion and
withering of the leaf lamina. Finally, pullulan assayed at
1 50 mg ml1 caused the collapse of marginal and interveinal
tissue, which soon desiccated, eventually dried out. The
cvs. Italia and Sangiovese showed a greater sensitivity to
0
the phytotoxins than cv. Matilde.
5 C HO
O O 3.4. Dual cultures callus/fungus
4 HO
Concentration (mg ml-1)

OH O
HO O
3 HO
HO
OH O O
2 HO
OH
1
0 subculture preparation or after 4–5 sub-culturing passages.
0 7 14
Time (days)
Fig. 1. Average concentration of scytalone (A), isosclerone (B) and
pullulan (C) in liquid cultures of P. chlamydospora (K) or T. minima (m)
maintained in the dark at 25 1C for 28 days. Data are the means of two
independent experiments with three replicates each. Standard error bars
are shown.
2.3  103 kDa, whereas pullulan produced by Tmi consisted
of three molecular species ranging from 70 to
2.5  103 kDa.
3.3. Phytotoxic activity of secondary metabolites
Symptoms were produced on the foliar lamina when
detached leaves absorbed a toxic solution containing one of
the three secondary metabolites produced by Pch and Tmi
(Fig. 2). When scytalone was assayed at 50 mg ml1 on
leaves of the three grapevine cvs., light green to chlorotic,
round to irregular, interveinal or marginal spots eventually
Callus cultures were generated with cv. Italia and cv.
Matilde grapevines. The callus yield of cv. Italia was at
least 4-fold greater than that of cv. Matilde in the same
period of time (Table 2), showing that cv. Italia was better
O
adapted to the callus growing conditions than cv. Matilde.
Calli maintained in total darkness could be subcultured for
at least 18 months. Most browning originated on the
peripheral areas of the callus and occurred 2–3 days after
21 28
Table 2 presents a comparison of the effect of the fungi
on callus growth. When Pch and Tmi were tested against
cv. Italia calli, they caused almost 3-fold more growth
inhibition than Fme. However, in cv. Matilde, Tmi resulted
in significant callus fresh wt decrement of 47% less than
control callus, whereas Pch reduced callus fresh wt by 29%
and Fme reduced it by 34%.
Callus cultures were also generated with Italia and
Matilde cvs. grown as dual culture with Pch or Tmi
(Fig. 3). In both experiments (callus+Pch and callus+
Tmi), cv. Italia generally showed browning on the external
cells and dark areas in the inner cells, whereas cv. Matilde
inoculated with Pch or Tmi produced only browning in the
inner and outer cells. Callus cross sections from both
cultivars clearly showed the accumulation of dark pro-
ducts. Cell melanization occurred in callus cultures of cv.
Italia due to the phenolic oxidation by the enzymes
produced by the pathogens, while there was no occurrence
of melanization either in cv. Italia callus+Fme or cv.
Matilde callus+Fme.
 ARTICLE IN PRESS
214 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223

'Italia'

Matilde'

'Sangiovese'

Fig. 2. Effect of the absorption for a few hours of 3 ml each of: (A) cultures filtrate of 28-day-old P. chlamydospora liquid culture; (B) 50 mg ml1 scytalone;
(C) 100 mg ml1 isosclerone; (D) 50 mg ml1 pullulan purified from culture filtrate of 28-day-old P. chlamydospora liquid cultures on detached leaves of
grapevine cultivars: Italia, Matilde and Sangiovese. Esca symptoms on naturally infected leaves (E).

Table 2 organic extract yield compared with the control, and

Callus yield, percentage of growth inhibition on calli of cv. Italia and cv. callus+Tmi gave the lowest yield, whereas callus+Pch
Matilde grapevines caused by Phaeomoniella chlamydospora (Pch), yield was quite similar to the control.
Togninia minima (Tmi) and Fomitiporia mediterranea (Fme)a

Dual culture Cv. Italia Cv. Matilde 3.6. Average concentration and changes of two phytotoxic
pentaketides produced by Pch and Tmi in grapevine calli
Fresh wt Growth Fresh wt Growth
(g) inhibition (g) inhibition
(%) (%) Cv. Italia: Scytalone and isosclerone were isolated from
calli of cv. Italia grapevine grown with either Pch or Tmi
Callus+Pch 296723 30 97712 29 (Table 4). Scytalone concentration was 10-fold greater than
Callus+Tmi 299718 29 7278 47
the isosclerone concentration, while both pentaketides were
Callus+Fme 375721 11 8979 34
Callus (control) 420730 — 133714 — absent in the calli grown without any fungus or with Fme.
Cv. Matilde: Isosclerone was isolated from both
a
Calli were harvested two months after the inoculation with each fungal callus+Pch and callus+Tmi (Table 4). On the contrary,
species. The values for callus fresh wt are the means of 15 callus pieces
scytalone was never detected in any of the dual cultures
from 3 replicates 7SD.
callus/fungus. Isosclerone concentration in callus+Tmi
was almost twice the concentration in callus+Pch. More-
3.5. Extraction of calli with organic solvents over, isoclerone content for cv. Matilde calli grown in the
presence of Pch or Tmi was 20-fold greater than for the
Grapevine calli were extracted using diethyl ether and same dual culture with cv. Italia calli, while isosclerone
ethyl acetate. Diethyl ether gave high recovery of content for cv. Matilde calli grown in the presence of Tmi
phenolics, and ethyl acetate was also suitable for the was 50-fold greater than for the same dual culture with cv.
extraction of phytotoxic pentaketides. It can be seen Italia calli. Both pentaketides were absent in the calli
(Table 3) that when diethyl ether was used, the crude grown without any fungus or with Fme.
organic extract (mg g1 callus fresh wt) of each dual
combination with cv. Italia calli, resulted in a 3-fold 3.7. Pullulan recovery from grapevine calli
increase of organic extract yield in comparison with the
control value. The content of diethyl ether extracts of cv. Purification of the ethanolic precipitate of each callus
Matilde calli in dual cultures was greater than that of singly inoculated with Pch and Tmi led to the recovery of a
control for callus+Fme and lower than the control for very small amount of pullulan from each grapevine cultivar
callus+Tmi and callus+Pch. When ethyl acetate was used (Table 4). Pullulan was never found in non-inoculated calli
for extraction, the treatment of cv. Italia calli gave a 2-fold or in callus+Fme. When pullulan concentration was
increase of crude organic extract obtained from calli grown assessed for the dual cultures of cv. Italia calli it was
with each fungus in comparison to the control. When Pch, 1.2 mg g1 fresh wt in callus+Pch and 0.5 mg g1 fresh wt in
Tmi and Fme were grown singly with cv. Matilde calli, the callus+Tmi. For the dual cultures of cv. Matilde, pullulan
values of crude organic extracts after treatment with ethyl was found only in callus+Pch at a concentration of
acetate were different. Callus+Fme gave the highest 0.5 mg g1 fresh wt.
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223 215

Calli

Cross sections

Calli

Cross sections

Control Callus + Pch Callus +Tmi

Fig. 3. Callus melanization by P. chlamydospora (Pch) and T. minima (Tmi) grown in presence of calli obtained from grapevine cultivars: Italia (A) and
Matilde (B).

Table 3
Crude organic extract yield (mg g1 callus fresh wt) obtained from cv. Italia or cv. Matilde calli grown with Phaeomoniella chlamydospora (Pch), Togninia
minima (Tmi) or Fomitiporia mediterranea (Fme) and then extracted with organic solventsa

Dual culture Cv. Italia Cv. Matilde

Diethyl ether Ethyl acetate Combined extracts Diethyl ether Ethyl acetate Combined extracts

Callus+Pch 313.5718.5 156.7721.2 470.2730.2 252.6731.7 77.1715.2 329.7721.3

Callus+Tmi 295.1722.4 169.5718.4 464.6727.2 317.7728.7 55.3710.3 373.0725.8
Callus+Fme 296.7727.5 132.9720.5 429.6738.7 510.7735.8 115.479.8 626.1731.4
Callus (control) 91.7715.3 79.8712.7 171.5721.4 372.3722.5 83.277.6 455.5730.2
a
Calli were harvested two months after the inoculation with each fungal species. The values for organic extracts are the means of 15 callus pieces from 3
replicates 7SD.

Table 4 fresh wt for cv. Italia singly inoculated with Pch, Tmi or
Average concentration (mg g1 callus fresh wt) of scytalone (Scy), Fme, whereas it varied from 186 to 266 mg g1 callus fresh
isosclerone (Iso) and pullulan (Pul) in calli of cv. Italia and cv. Matilde wt for cv. Matilde (Fig. 4). In cv. Italia, the presence of
grapevines singly-grown in dual culture with one of the three fungal
species Phaeomoniella chlamydospora (Pch), Togninia minima (Tmi) and each fungus caused the decrease of total phenolic content
Fomitiporia mediterranea (Fme)a estimated in the control. In cv. Matilde, the total phenolics
concentration determined for each dual culture was higher
Dual culture Cv. Italia Cv. Matilde than that of the control. The trend of recurring phenolics in
Scy Iso Pul Scy Iso Pul both cvs. was very similar to that elicited by total
phenolics. In cv. Matilde calli, when recurring phenolic
Callus+Pch 76713 10.374 1.270.3 0 141716 0.570.1 concentration was compared to the control yield, it
Callus+Tmi 74710 9.872 0.570.1 0 267718 0
declined in callus+Pch, increased in callus+Tmi and
Callus+Fme 0 0 0 0 0 0
Callus (control) 0 0 0 0 0 0 decreased in callus+Fme. When the concentration of
recurring phenolics in cv. Italia calli was compared with
a
Data are the means of two independent experiments with three the control, it declined in callus+Pch, slightly increased in
replicates each 7SD.
callus+Tmi, and reached the control value in callus+Fme.

3.8. The trend of total phenolics and recurring phenolics in
grapevine calli
The content of total phenolics as determined by the
Folin-Ciocalteu method varied from 55 to 68 mg g1 callus
3.9. Accumulation of individual recurring phenolics
3.9.1. In batch cultures
The content of total phenolics as determined by the
Folin-Ciocalteu method was scant. The yield varied from

216 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
300
A
250
Concentration (g g-1callus fresh wt)
200
150
100
50
0 and vanillic acid] were absent for callus+Fme.
300
B control, p-coumaric acid and ferulic acid for callus+Pch,
250
Concentration (g g-1 callus fresh wt)
200
150
100
50
0
Callus Callus + Pch
Fig. 4. Concentrations of total (A) or recurring phenolics (B) extracted
from calli of grapevines cv. Italia (&) and cv. Matilde ( ) grown as dual
culture with P. chlamydospora (Pch), T. minima (Tmi) or F. mediterranea
(Fme). Data are the means of two independent experiments carried out in
2003 and 2004, with three replicates each. Standard error bars are shown.
0.25 to 0.49 mg ml1 in the ethyl ether extracts of culture
filtrates of each fungus. Pch was the highest producer
whereas Fme was the lowest. When the ethyl acetate
extracts were examined, Pch remained the highest producer
(1.33 mg ml1) and Fme the lowest producer (0.98 mg ml1).
No recurring phenolics were detected in the two organic
extracts (ethyl acetate or ethyl ether) of culture filtrates
obtained from the liquid cultures of each fungal species
(Pch, Tmi and Fme).
3.9.2. In dual cultures callus/fungus
Direct injection into the HPLC system of the organic
extracts of each dual culture, resulted in easily compre-
hensible chromatograms which made it possible to identify
and quantify peaks and made it easier to interpret them. As
ARTICLE IN PRESS
Fig. 5 shows for cv. Italia grapevine, greater efficiency was
obtained when methanol:water (1:1; v/v) mixture was used
as the injection solvent; this can be justified by the
differences in polarity between the extracted compounds.
Table 5 shows the concentration of each phenolic isolated
from cv. Italia or cv. Matilde calli.
Cv. Italia: In cv. Italia calli, out of the nineteen recurring
phenolics, eight [2,5-dihydroxybenzaldehyde, protocate-
chuicaldehyde, (+)catechin, p-coumaric acid, ferulic acid,
kaempferol, quercetin and vanillic acid] were absent for the
calli used as the control, while only two (caffeic acid and
apigenin) were absent for callus+Pch, five [protocatechui-
caldehyde, caffeic acid, p-coumaric acid, ferulic acid and
()epicatechin] were absent for callus+Tmi, and eigth
[protocatechuicaldehyde, syringaldehyde, caffeic acid,
p-coumaric acid, ferulic acid, ()epicatechin, apigenin
The lowest concentrations were of caffeic acid for the
and syringic acid for both callus+Tmi and callus+Fme.
Pch, Tmi and Fme stimulated the production of
2,5-dihydroxybenzaldehyde, (+)catechin, kaempferol and
quercetin which were absent in the control. In addition,
Tmi induced the production of vanillic acid, whereas Pch
induced the production of p-coumaric acid, ferulic acid,
protocathechuicaldehyde and vanillic acid. Myricetin was
the major flavonol found in the calli used as the control,
whereas quercetin was the major flavonol found for
both callus+Pch and callus+Tmi, and kaempferol for
callus+Fme. Rutin was detected at the highest concentra-
tion in the control sample and at the lowest concentration
in callus+Fme. Apigenin, present at a very low concentra-
tion in the control and in callus+Tmi, was not detected in
callus+Pch or in callus+Fme. When resveratrol content
Callus + Tmi Callus + Fme was compared with that of the control it decreased in all
dual cultures.
Cv. Matilde: When the concentrations of phenolics were
assessed for the dual cultures of cv. Matilde calli with each
fungus, a considerable change was observed in the amounts
of each phenolic when comparing their values with control
values (Table 5). The content of 2,5-dihydroxybenzalde-
hyde increased in callus+Pch and in callus+Tmi, and
decreased in callus+Fme. Protocathechuicaldehyde was
not detected in any of the dual cultures. Syringaldehyde,
the major benzoic aldehyde derivative in the control, was
present at low concentration in callus+Pch and in
callus+Tmi, and was absent in callus+Fme. Of the four
benzoic acid derivatives, two (gallic acid and syringic acid)
were found in the control sample, three (gallic acid,
protocatechuic acid and syringic acid) were found in
callus+Pch, three (gallic acid, protocatechuic acid and
vanillic acid) were found in callus+Fme, and only one
(gallic acid) was found in callus+Tmi. The cinnamic acid
derivatives were absent in the control sample, and were
also absent in all the dual cultures except for traces of
p-coumaric acid present in callus+Pch. Quercetin, the
major flavonol in the control sample, was still present in
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223 217

Fig. 5. Chromatograms of a mixture of phenolics used as standards (A) and of diethyl ether extracts of Vitis vinifera cv. Italia (left) or cv. Matilde (right)
calli grown with or without any esca-associated fungus: callus used as control (B), callus grown with P. chlamydosppora (C), callus grown with T. minima
(D), callus grown with F. mediterranea (E). Gallic acid (1), protocatechuic acid (2), protocatechuicaldehyde (3), (+) catechin (4), 2,5-dihydroxy-
benzaldehyde (5), vanillic acid (6), caffeic acid (7), syringic acid (8), () epicatechin (9), syringaldehyde (10), p-cumaric acid (11), ferulic acid (12), rutin
(13), trans-resveratrol (14), myricetin (15), quercetin-3-rhamnoside (16), quercetin (17), apigenin (18), kaempferol (19). For chromatographic conditions
see Section 2.8.

callus+Tmi and in callus+Pch but at a concentration The other two flavonols — kaempferol and myricetin —
6-fold lower than the control value in callus+Tmi, and were detected in all three callus+fungus combinations.
15-fold lower than the control value in callus+Pch. Kaempferol content, compared to the control, decreased in
Moreover, quercetin was not detected in callus+Fme. all callus cultures, whereas myricetin content was higher in
 ARTICLE IN PRESS
218 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223

Table 5
Average concentration (mg g1 callus fresh wt) of recurring phenolics in calli of cv. Italia and cv. Matilde grapevines grown alone or with one of the three
fungal species: Phaeomoniella chlamydospora (Pch), Togninia minima (Tmi) and Fomitiporia mediterranea (Fme)a

Phenolic compound Control (callus)b Callus+Pch Callus+Tmi Callus+Fme

cv. Italia cv. Matilde cv. Italia cv. Matilde cv. Italia cv. Matilde cv. Italia cv. Matilde

Benzaldehyde derivatives
2,5-dihydroxy-Benzaldehyde n.d.c 1.2570.15 0.4470.05 5.7370.25 1.6470.02 18.9470.82 0.6870.18 0.1970.02
Protocatechuicaldehyde n.d. n.d. 0.1770.03 n.d. n.d. n.d. n.d. Traces
Syringaldehyde 0.8070.02 3.5670.35 0.2570.02 0.0670.01 0.1770.01 0.0670.02 n.d. n.d.
Benzoic acid derivatives
Gallic acid 0.2170.01 0.3470.10 2.8270.44 0.4670.18 0.2670.02 0.0770.02 1.7470.33 0.6770.10
Protocatechuic acid 13.0470.06 n.d. 2.0370.02 0.2870.03 1.1970.22 n.d. 2.5870.02 0.0370.01
Syringic acid 0.0470.02 2.2070.19 Traces 0.5370.22 0.0270.04 n.d. 0.0470.01 n.d.
Vanillic acid n.d. n.d. 0.7370.06 n.d. 0.1870.05 n.d. n.d. 0.1070.01
Cinnamic acid derivatives
Caffeic acid Traces n.d. n.d. n.d. n.d. n.d. n.d. n.d.
p-Coumaric acid n.d. n.d. Traces Traces n.d. n.d. n.d. n.d.
Ferulic acid n.d. n.d. Traces n.d. n.d. n.d. n.d. n.d.
Flavonols
Apigenin 0.0270.01 n.d. n.d. Traces 0.0270.01 0.0870.1 n.d. Traces
Kaempferol n.d. 54.3271.24 1.9470.08 25.1670.16 8.5970.70 45.0470.25 14.0370.14 7.0670.08
Myricetin 13.2670.28 1.5070.07 0.3470.02 21.7370.26 3.8870.03 0.6370.03 1.0970.04 1.8670.03
Quercetin n.d. 64.5671.52 5.6570.10 19.3870.20 17.0870.39 46.9670.25 3.4670.18 n.d.
Rutin 0.9570.17 0.7870.11 0.6270.13 0.1170.03 0.2970.15 0.5870.12 0.1270.09 7.0670.25
Flavonol-3-O-glycosides
Quercetin-3-rhamnoside 0.8070.02 n.d. 3.8570.06 42.7171.70 14.5770.11 61.1070.70 14.0270.10 76.0970.03
Flavan-3-ol derivatives
(+)Catechin n.d. n.d. 0.2670.02 n.d. 0.1170.04 n.d. 0.1870.02 n.d.
()Epicatechin 4.0070.06 n.d. 0.2570.02 n.d. n.d. n.d. n.d. n.d.
Stilbenes
trans-Resveratrol 0.6170.04 0.0470.02 0.1470.06 Traces 0.1370.03 0.3070.10 0.1370.02 Traces
Total phenolics 33.75 128.55 19.51 116.12 48.13 173.69 38.07 85.99
a
The callus was extracted with diethyl ether and subsequently with ethyl acetate; then the organic extracts were combined. Values are the means of 3
replicates 7SD.
b
Traces quantities (p0.015 mg g1 callus fresh wt) of phenolics.
c
n.d.: no detected.

callus+Pch and in callus+Fme and lower in callus+Tmi.
Rutin concentration in both callus+Pch and callus+Tmi
was lower than that of the control, but was 10-fold greater
than callus+Fme. Apigenin was absent in the control, but
traces were present in callus+Pch and in callus+Fme, and
very low concentration in callus+Tmi. Quercetin-3-rham-
noside was absent in the control, but was the most
abundant phenolic in all dual cultures. Traces of resvera-
trol were detected in both callus+Pch and callus+Fme,
and it was present in callus+Tmi at a concentration 3-fold
lower than that of the control. Flavonols were the main
phenolics in cv. Matilde calli.
4. Discussion
Plant pathogens produce a variety of compounds in their
hosts as well as in soil or artificial media, and some of these
products have been characterized in terms of chemistry,
biology, plant/pathogen interactions, structure/biological
activity relationship [39–44]. Among these metabolites,
toxic compounds have attracted the interest of plant
pathologists as incitans of diseases and have been well
characterized as either pathogenicity or virulence factors.
The structure and the mode of action of host selective
toxins, e.g., victorin produced by Cochliobolus victoriae
R.R. Nelson, the causal agent of oat victoria blight,
represent the subject of frontier studies for the elucidation
of molecular mechanisms of pathogenesis [39,44]. Victorin
has been shown to bind to the P protein of the glycine
decarboxylase complex in mitochondria, and induce
defense-related responses such as phytoalexin synthesis,
extracellular alkalization and programmed cell death [44].
There is much more information available on the role
of non-selective toxins, e.g., seiridins, seiricardins and
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
cyclopaldic acid produced by three species of Seiridium,
causal agents of canker disease of cypress or sphaeropsi-
dins A–F (SA–SF) produced by Sphaeropsis sapinea (Fr.)
Dyko & B. Sutton f. sp. cupressi, and Diplodia pinea (Fr.)
Mont. than on the role of host-selective toxins [41–43].
When Cupressus macrocarpa Hartw. (susceptible to
Seiridium spp.), C. sempevirens L. and C. arizonica Green
(less susceptible to the same fungal species) were examined
for electrolyte loss during the infectious process, it was
found that different yields of electrolyte leakage were
obtained from each species. The experiments of toxin
treatment (50 or 100 mM seiridin) of the same cypress
species showed that shoot tissue of C. macrocarpa leached a
greater amount of electrolytes than C. sempervirens or C.
arizonica.
Biological results were obtained in terms of phytotoxi-
city of SA-SF against three species of cypress. SA, SB and
SC caused symptoms on host plants similar to those
observed on naturally infected plants (dieback of C.
macrocarpa, browning and necrosis on C. sempervirens,
and yellowing on C. arizonica. This finding shows that the
three cypress species have a different grade of tolerance to
the toxin action.
These data supported the idea that plants susceptible to
the pathogen were also sensitive to its toxins [42]. Non-
selective toxins play a significant role in pathogenesis and if
they are inactivated by either molecular modification or
degradation in or out of plant cells, infection will be
prevented [40].
The occurrence of scytalone, isosclerone and pullulan as
phytotoxic secondary metabolites produced by Pch and
Tmi in vitro and especially in vivo may contribute to a better
understanding of their involvement in pathogenesis and in
symptom development.
In grapevine, the most frequently observed and best
characterized active defense mechanisms are the accumula-
tion of phenolics and the synthesis of extracellular enzymes
[24]. Particular attention has been given in the present
article to: (i) phenolic production in calli, (ii) their
biological activity, (iii) two phytotoxic pentaketides and
a-glucans produced by Pch and Tmi in liquid cultures and
in calli, (iv) the relationship between the role of scytalone
and isosclerone in melanin biosynthesis and in pathogeni-
city. The grapevines studied in this investigation belong to
two cultivars: Italia (susceptible) and Matilde (intermediate
resistant) [5,34]. The results showed great differences in the
reactions of calli obtained from both cultivars, following
the infection process carried out by each fungus (Pch, Tmi
or Fme). The net accumulation of phenolics within the
plant tissues infected by the pathogens is probably
controlled by a balance which may result, on one hand,
from the ability of the host cells to resist colonization by
creating an inhibitory barrier to the parasite and, on the
other hand, from tolerance of the pathogen to antifungal
compounds produced by the plant and from its ability to
metabolize (or detoxify) the phenolics to which it is
exposed.
219
4.1. Dual culture callus/fungal pathogen
Callus fresh wt yield differed greatly with respect to the
susceptibility of each cultivar to esca-associated fungi.
Each fungus also reacted differently, with its growth rate
decreased (cv. Italia) or stimulated (cv. Matilde). The
major differences between cv. Italia and cv. Matilde were in
tissue browning, cell penetration and phenolic accumula-
tion or bioconversion. Incubation of the callus tissues
together with living mycelial plugs of Pch, Tmi and Fme led
to a rapid increase in the colony diameters of the fungi and
to a visible inhibition of callus cell growth of both cvs.,
with the exception of the dual culture cv. Matilde
callus+Pch. Callogenesis decreased progressively when
fungal colonies were near calli and stopped completely
when the callus was completely colonized by each fungus
and the color changed from brown to dark.
4.2. Callus extraction by organic solvents
Diethyl ether and ethyl acetate were tested for liquid–
liquid extraction of solutions obtained after callus homo-
genization. Reproducibility in recovery of phenolics with
diethyl ether was high. Differences in the content of several
phenolics can be observed between the cv. Italia calli and
cv. Matilde calli; this may be due to differences in
grapevine variety. Our results confirmed the analogous
data obtained by Malovaná et al. [35] for the extraction of
polyphenols from red wines. It could be seen that there was
greater efficiency when diethyl ether was used. Ethyl
acetate was suitable for phytotoxic pentaketide recovery
from calli.
4.3. Trend of total phenolics and recurring phenolics in calli
Calli from excised leaf petiole of both cv. Italia and cv.
Matilde have been used to study the possible effect of Pch,
Tmi and Fme on the endogenous pool size of phenolics.
Apparently, grapevine calli are critically unbalanced with
regard to storage and conversion of phenolics. The
metabolic changes associated with the reactions involving
phenolic bioconversion by each fungus may be basically
modified by their accumulation in the cells and vital
defense requirements. At the same time, fungal activity
resulted in an increase or decrease in the concentration of
phenolics, already present in uninfected calli. The highest
content in total or recurring phenolics was found in cv.
Matilde, while cv. Italia had a lower content. This
difference between the two cultivars could explain the
lower susceptibility of cv. Matilde to the esca-associated
fungi. Resveratrol was present at higher concentrations in
calli of both cvs. used as control, it decreased in
callus+Pch and callus+Fme. This pattern could be
explained by our previous observations that resveratrol
was easily degraded by all the three esca-associated fungi
[10]. Resveratrol is considered a phytoalexin, which
protects the host from fungal infection, especially against
 ARTICLE IN PRESS
220 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223

Botrytis cinerea Pers [22–24]. Thus, contrary to the current
theory, at least for the case of the esca-associated fungi,
resveratrol is degraded and it does not protect the host by
itself.
4.4. Production and bioconversion of recurring phenolics
The resistance of plants to infection depends partly on the
phytoalexin production/degradation balance following at-
tack by the pathogen. Among the plant substances showing
antifungal activity, phenolic hydroxyl groups—which have a
high affinity to proteins—may act as inhibitors to fungal
enzymes necessary to infect plants [45,46].
Phenolics are effective hydrogen donors, particularly
flavonols such as quercetin [29,46]. In a number of tree
crops, maximal flavonol concentrations occurred in the
outer protective layer of the fruits. In grapevine, it has been
demonstrated that flavonols are mainly deposited in the
exocarp and seed coat of grapes [47]. Results obtained by
several Authors [38,48,49] have shown that benzoic acid
derivatives and flavonols are the most significant phenolics
in grapevine and red vines. As expected, the amount of
flavonol in white wines is very low, since these compounds
are found in grape skin. Tylose formation and changes in
the level and composition of phenolics occurred in relation
to the severity of infection in grape roots infected by Pch
and several Phaeoacremonium sp. [50]. Day et al. [51]
showed that catechins are present only in the lower
epidermis of V. rotundifolia Michx. leaves before
infection. This layer of catechins in the lower epidermis
probably forms a chemical barrier to infection. Catechins
are key constituents of grape and mainly occurred in the
outer protective layers of both fruit skin and seed. Feucht
et al. [23] also demonstrated that in calli of the Dornfelder
and Riesling grapevine cvs., a change in flavanol deposition
occurred near Phomopsis viticola (Sacc.) Sacc. infections.
Advancing towards the infection site, cellular enlargement,
vacuolisation and formation of flavanol globules pro-
ceeded more intensely.
4.5. Scytalone as an intermediate of melanin pathway or
phytotoxic metabolite
Scytalone and isosclerone, as well as other closely related
naphthalene pentaketides, are known as fungal metabolites
[12]. The results of our experiments indicate that Pch and
Tmi could utilize pentaketide metabolites as intermediates
in melanin biosynthesis (Fig. 6). The pathway to melanin in
Pch and Tmi may be similar or identical to that proposed
for Verticillium dahliae Kleb. [52,53], Thielaviopsis basicola
(Berk. & Broome) Ferraris [54], Pyricularia oryzae Cavara
[55]. The highly unstable metabolite 1,3,8-trihydroxy-
naftalene (1,3,8-THN) is readily converted to 2-hydro-
xyjuglone by auto-oxidation and possibly by enzymatic
oxidation. The accumulation of the latter with 3,4,8-
trihydroxy-naftalenone (3,4,8-THN) and the conversion to
isosclerone lead to the deposition of brown pigment which
is concomitant with the inhibition of melanin biosynthesis
as an overflow into the 2-hydroxyjuglone branch pathway,
resulting from the inhibition of the reduction of 1,3,8-THN
to vermelone [53,55,56]. The final step of the melanin
pathway is the conjoining of 1,8-dihydroxynaftalene

Phytotoxic metabolite

Acetat e O OH O OH OH OH O OH
[H] [-H2O] [O]

HO HO
O HO HO HO
(Pentaketide 2-Hydroxyjuglone DTN 1,4,8-THN DDN (Isosclerone)
cyclization)

[O]

OH OH OH O OH OH O OH OH OH
[H] [-H2O] [H ] [-H2O] [O]
Melanin
HO OH HO OH HO HO
1,3,6,8-THN Scytalone 1,3,8-THN Vermelone 1,8-DHN

Phytotoxic metabolite

Fig. 6. Proposal of a scheme showing the metabolic steps leading to melanin pathway or pentaketide (scytalone and isosclerone) biosynthesis in P.
chlamydospora and T. minima. Pathway modelled after Tokousbalides and Sisler [52] and Wheeler [53]. DDN ¼ 3,4-dihydro-4,8-dihydroxy-1(2H)-
naphtalenone; DTN ¼ 3,4-dihydro-3,4,8-trihydroxy-1(2H)-naphtalenone; 1,3,8-THN ¼ 1,3,8-trihydroxynaphtalene; 1,3,6,8-THN ¼ 1,3,6,8-tetrahydrox-
ynaphtalene; 1,3,8-THN ¼ 1,3,8-trihydroxynaphtalene; 1,8-DHN ¼ 1,8-dihydroxynaphtalene.
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
(DHN) molecules to form the melanin polymer. There are
a number of candidate enzymes for this step, including
phenoloxidases such as tyrosinases, peroxidases, laccases,
and may be catalases [56].
Data obtained by pentaketide extraction from culture
filtrates of Pch and Tmi or from calli of both cvs. Italia and
Matilde singly inoculated with Pch and Tmi greatly
supported this hypothesis. The rate of accumulation and
concentration of the two pentaketides was due to their
environmental and nutritional conditions. Both fungi,
grown in stationary liquid cultures, produced a higher
concentration of scytalone than of isosclerone. Scytalone
accumulated at higher concentration in cv. Italia calli. On
the contrary, in cv. Matilde calli the fungi produced only
isosclerone, a metabolite less phytotoxic than scytalone.
4.6. Phytotoxic pullulan production
Phytotoxicity by polysaccharides was recognized since
the work of Hodgson et al. [57]. It has been demonstrated
that some plant-pathogenic fungi belonging to the genus
Phytophthora produce b-1,3-glucans. More recently has
been reported that C. parasitica, the causal agent of
chestnut blight, produced a large amount of pullulan
[16,17]. The main difference is that pullulan is an a-glucan
instead of b-glucan. Moreover, Pch and Tmi were reported
as pullulan producers in a mixture of exopolysaccharides
produced in stationary growth [14,33].
It is well known that the average molecular wt of
pullulan varies in a very broad range, from hundred to
thousands of kDa, depending on the culture strain, pH,
cultivation techniques, substrated used, and microorgan-
ism tested [15]. In our experiments, the ethanolic pre-
cipitates obtained from calli inoculated with Pch or Tmi
contained several molecular species of pullulan ranging
from 10 to 1000 kDa. This finding confirmed similar data
obtained by other Authors [15]. Certain computational
data indicated that the pullulan macromolecule might
exhibit slight helical twists [58,59]. The peculiar structure
and conformation of pullulan and its reological and
solution properties may account for the unusually easy
transport of these large-molecular metabolites in the plant
cells or in the xylem elements. It is possible that the toxic
effect of pullulan on the callus or leaf tissues is associated
with the formation of thin films on the plasmalemma,
which oxygen cannot permeate.
In conclusion, the findings indicated that the phytotoxic
metabolites (scytalone, isosclerone and pullulan) were
produced either in liquid culture or in callus tissues
infected by Pch and Tmi. These compounds are absorbed
by grapevine leaves (cvs. Italia, Matilde, Sangiovese) and
reproduce symptoms quite similar to those observed on the
leaves of the same cultivars naturally infected by the same
fungal species. Calli colonized by each fungus showed
reduced growth, gummosis, browning and necrosis as well
as several aggregates eliciting melanin-like deposits. More-
over, the results presented in this paper indicated a high
221
level of variability in the phenolic content of cv. Italia and
cv. Matilde, and in their susceptibility to the esca-
associated pathogens. In addition to producing significant
and typical esca syndrome symptoms, an in vitro culture
system (calli) should facilitate the evaluation of plant
reactions to the fungi. This method of culturing grapevine
with esca-associated fungi allows evaluation of the host/
pathogen interactions in controlled environmental condi-
tions. The primary advantage is in the possibility of
continuous observation without disrupting or interfering
with the host–parasite interactions, while at the same time
excluding secondary interactions with other biotic or
abiotic agents.
Acknowledgements
The Authors acknowledge the financial support given to
this work by the Italian Ministry of Instruction, University
and Research (MIUR), Rome, Italy. The technical
assistance of Mr. L. Scarola and Mrs. P. Basso is gratefully
acknowledged.
References
[1] Graniti A, Sparapano L, Bruno G. Alcuni progressi degli studi sulla
patogenesi del ‘‘mal dell’esca’’ e delle ‘‘venature brune del legno’’
della vite. Inf Fitopat 2001;51(5):13–21.
[2] Mugnai L, Graniti A, Surico G. Esca (black measles) and brown
wood-streaking: two old and elusive diseases of grapevines. Plant Dis
1999;83:404–18.
[3] Sparapano L, Bruno G, Ciccarone C, Graniti A. Infection of
grapevines by some fungi associated with esca. I. Fomitiporia punctata
as a wood-rot inducer. Phytopath Mediterr 2000;39:46–52.
[4] Sparapano L, Bruno G, Graniti A. Three-year observation of
grapevines cross-inoculated with esca-associated fungi. Phytopath
Mediterr 2001;40(Suppl):376–86.
[5] Sparapano L, Bruno G, Campanella A. Interactions between three
fungi associated with esca of grapevine, and their secondary
metabolites. Phytopath Mediterr 2001;40(Suppl):417–22.
[6] Larignon P, Dubos B. Fungi associated with esca disease in
grapevine. Eur J Plant Pathol 1997;103:147–57.
[7] Mugnai L, Surico G, Esposito A. Micoflora associata al mal dell’esca
della vite in Toscana. Inf Fitopat 1996;46(11):49–55.
[8] Sparapano L, Bruno G, Ciccarone C, Graniti A. Infection of
grapevines by some fungi associated with esca. II. Interaction among
Phaeoacremonium chlamydosporum, Pm. aleophilum, Fomitiporia
punctata. Phytopath Mediterr 2000;39:53–8.
[9] Bruno G, Sparapano L. Influenza di Phaeoacremonium aleophilum,
Phaeomoniella chlamydospora e Fomitiporia mediterranea sull’accu-
mulo e sulle variazioni della concentrazione di polifenoli e fitotossine
in calli e piante di Vitis vinifera. J Plant Pathol 2004;86:295–6.
[10] Bruno G, Sparapano L. Degradazione di resveratrolo e acido tannico
da parte di Phaeomoniella chlamydospora, Phaeocremonium aleophi-
lum e Fomitiporia mediterranea, funghi associati al mal dell’esca della
vite. Micol Ital 2005;33(1):34–52.
[11] Bruno G, Sparapano L, Amalfitano C, Evidente A, Graniti A.
Detection of phytotoxic fungal metabolites in xylem sap of esca-
affected grapevines. Proceedings of the fifth Congress of the
European Foundation for Plant Pathology: Biodiversity in Plant
Pathology, Italy, 2000. p. 204.
[12] Evidente A, Sparapano L, Andolfi A, Bruno G. Two naphthalenone
pentaketides from liquid cultures of Phaeoacremonium aleophilum, a
 ARTICLE IN PRESS
222 G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
fungus associated with esca of grapevine. Phytopath Mediterr 2000;
39:162–8.
[13] Tabacchi R, Fkyerat A, Poliart C, Dubin GM. Phytotoxins from
fungi of esca of grapevine. Phytopath Mediterr 2000;39:156–61.
[14] Sparapano L, Bruno G, Graniti A. Esopolisaccaridi fitotossici sono
prodotti in coltura da due specie di Phaeoacremonium associate al
complesso del ‘mal dell’esca’ della vite. Petria 1998;8:210–2.
[15] Shingel KI. Current knowledge on biosynthesis, biological activity,
and chemical modification of the exopolysaccharide, pullulan. Carb
Res 2004;339:447–60.
[16] Corsaro MM, De Castro C, Evidente A, Lanzetta R, Molinaro A,
Parrilli M, et al. Phytotoxic extracellular polysaccharide fractions
from Cryphonectria parasitica (Murr.) Barr strains. Carb Polym
1998;37:167–78.
[17] Forabosco A, Bruno G, Sparapano L, Liut G, Marino D, Delben F.
Pullulans produced by strains of Cryphonectria parasitica—I.
Production and characterisation of the exopolysaccharides. Carb
Polym 2006;63:535–44.
[18] Sparapano L, Bruno G. Production of pullulans by virulent and
hypovirulent strains of Cryphonectria parasitica. Proceedings of the
tenth Congress of the Mediterranean Phytopathological Union,
France, 1997. p. 489–93.
[19] Sparapano L, Bruno G. Effetti dell’azoto e del carbonio organico
sulla produzione di esopolisaccaridi in vitro e sulla morfologia dei
ceppi di Cryphonectria parasitica. Atti del Convegno Nazionale sul
Castagno, Italy, 1998. p. 455–74.
[20] Grayer RJ, Harborne JB. A survey of antifungal compounds from
higher plants, 1982–1993. Phytochemistry 1994;37:19–42.
[21] Prusky D, Keen NT. Involvement of preformed antifungal com-
pounds in the resistance of subtropical fruits to fungal decay. Plant
Dis 1993;77:114–8.
[22] Dercks W, Creasy L. The significance of stilbene phytoalexins in the
Plasmopara viticola-grapevine interaction. Physiol Mol Plant Pathol
1989;34:189–202.
[23] Feucht W, Treuter D, Christ E. Flavanols in grapevine: in vitro
accumulation and defence reactions in shoots. Vitis 1996;35(3):
113–8.
[24] Jeandet P, Douillet-Breuil AC, Bessis R, Debord S, Sbaghi M,
Adrian M. Phytoalexins from the Vitaceae: biosynthesis, phytoalexin
gene expression in transgenic plants, antifungal activity, and
metabolism. J Agric Food Chem 2002;50:2731–41.
[25] Low PS, Schroeder AT. Signal transduction patways of the plant
oxidative burst. In: Stacey G, Keen NT, editors. Plant–microbe
interactions, vol. 3. New York: International Thomson Publishing;
1997. p. 35–52.
[26] Mori IC, Schroeder JI. Reactive oxygen species activation of plant
Ca2+ channels. A signaling mechanism in polar growth, hormone
transduction, stress signaling, and hypothetically mechanotransduc-
tion. Plant Physiol 2004;135:702–8.
[27] Breuil AC, Jeandet P, Adrian M, Chopin F, Pirio N, Meunier P, et al.
Characterization of a pterostilbene dehydrodimer produced by
laccase of Botrytis cinerea. Phytopathology 1999;89:298–302.
[28] Pezet R. La pourriture grise de raisins. Le complexe plante-parasite.
Vigneron Champenois 1993;5:65–83.
[29] Larson RL. The antioxidants of higher plants. Phytochemistry
1988;4:969–78.
[30] McClure JW. Physiology and functions of flavonoids. In: Harborne
JB, Mabry TJ, Mabry H, editors. The flavonoids, vol. 2. New York:
Academic Press; 1975. p. 970–1055.
[31] Dixon RA, Steele LC. Flavonoids and isoflavoids—a gold mine for
metabolic engineering. Trends Plant Sci 1999;4:394–400.
[32] Linsmaier EM, Skoog F. Organic growth factors requirements of
tobacco callus tissue cultures. Physiol Plant 1965;18:100–27.
[33] Sparapano L, Bruno G, Graniti A. Effects on plants of metabolites
produced in culture by Phaeoacremonium chlamydosporum, P.
aleophilum and Fomitiporia punctata. Phytopath Mediterr 2000;39:
169–77.
[34] Sparapano L, De Leonardis S, Campanella A, Bruno G. Interaction
between esca-associated fungi, grapevine calli and micropropagated
shoot cultures of grapevine. Phytopath Mediterr 2001;40(Suppl):
423–8.
[35] Malovaná S, Garcı́a Montelongo FJ, Pérez JP, Rodrı́guez-Delgado
MA. Optimisation of sample preparation for the determination of
trans-reveratrol and other polyphenolic compounds in wines by high
performance liquid chromatography. Anal Chim Acta 2001;428:
245–53.
[36] Rodrı́guez-Delgado MA, Malovaná S Pérez JP, Borges T, Garcı́a
Montelongo FJ. Separation of phenolic compounds by high-
performance liquid chromatography with absorbance and fluori-
metric detection. J Chrom A 2001;912:249–57.
[37] Barker SA, Bourne EJ, Whifen DH. Use of infrared analysis in the
determination or carbohydrate structure. In: Glick D, editor.
Methods of biochemical analysis, vol. 3. New York: Interscience;
1956. p. 213–45.
[38] Simonetti P, Pietta P, Testolin G. Polyphenol content and total
antioxidant potential of selected Italian wines. J Agric Food Chem
1997;45:1152–5.
[39] Scheffer RP, Livingston RS. Host-selective toxins and their role in
plant diseases. Science 1984;223(4631):17–21.
[40] Graniti A, Durbin RD, Ballio A. Phytotoxins and plant pathogenesis.
NATO ASI Series H: Cell Biology, vol. 27. Berlin: Springer; 1989.
[41] Sparapano L, Evidente A. Biological activity of cyclopaldic acid, a
major toxin of Seiridium cupressi, its six derivates, and iso-cyclopaldic
acid. Nat Toxins 1995;3:156–65.
[42] Sparapano L, Evidente A. Studies on structure-activity relationship
of seiridins, phytotoxins produced by three species of Seiridium. Nat
Toxins 1995;3:166–73.
[43] Sparapano L, Bruno G, Fierro O, Evidente A. Studies on structure-
activity relationship of sphaeropsidins A-F, phytotoxins produced
by Sphaeropsis sapinea f. sp. cupressi. Phytochemistry 2004;65(2):
189–98.
[44] Tada Y, Kusaka K, Betsuyaku S, Shinogi T, Sakamoto M, Ohura Y,
et al. Victorin triggers programmed cell death and the defense
response via interaction with a cell surface mediator. Plant Cell
Physiol 2005;46(11):1787–98.
[45] Deverall BJ. Introduction. In: Bailey JA, Mansfield JW, editors.
Phytoalexins. Glasgow: Blackie & Sons Ltd.; 1982. p. 1–16.
[46] Rice-Evans CA, Miller NJ, Bolwell PG, Bramley PM, Pridham JB.
The relative antioxidant activity of plant-derived polyphenolic
flavonoid. Free Radical Res 1995;22:375–83.
[47] Escribano MT, Guerra MT, Rivas Gonzalo JC, Santos Buelga C.
Proanthocyanidin composition in skin and seeds from grapes.
Proceedings of the seventeenth international conference on poly-
phenols, Spain, 1994. p. 225–6.
[48] Frankel EN, Waterhouse AL, Tessendre PL. Principal phenolic
phytochemicals in selected California wines and their antioxidant
activity inhibiting oxidation of human low-density lipoproteins.
J Agric Food Chem 1995;43:890–4.
[49] Hertog MGL, Hollman PCH, van de Putte B. Content of potentially
anticancerogenic flavonoids of tea infusions, wines, and fruit juices.
J Agric Food Chem 1993;41:1242–6.
[50] Del Rio JA, Gonzalez A, Fuster MD, Botia JM, Gomez P, Frias V, et
al. Tylose formation and changes in phenolic compounds of grape
roots infected with Phaeomoniella chlamydospora and Phaeoacremo-
nium species. Phytopath Mediterr 2001;40(Suppl):394–9.
[51] Day GH, Andary C, Cosson L, Boubals D. Polyphenols and
resistance of grapevines (Vitis spp.) to downy mildew (Plasmopara
viticola). Acta Hortic 1994;381:763–6.
[52] Tokousbalides MC, Sisler HD. Site of inhibition by tricyclazole in the
melanin biosynthetic pathway of Verticillium dahliae. Pestic Biochem
Physiol 1979;11:64–73.
[53] Wheeler MH. Melanin biosynthesis in Verticillium dahliae: dehydra-
tion and reduction reactions in cell-free homogenates. Exp Mycol
1982;6:171–9.
 ARTICLE IN PRESS
G. Bruno, L. Sparapano / Physiological and Molecular Plant Pathology 69 (2006) 209–223
[54] Wheeler MH, Stipanovic RD. Melanin biosynthesis in Thielaviopsis
basicola. Exp Mycol 1979;3:340–50.
[55] Woloshuk CP, Sisler HD, Tokousbalides MC, Dutky SR. Melanin
byosynthesis in Pyricularia oryzae: Site of tricyclazole inhibition and
pathogenicity of melanin-deficient mutants. Pest Biochem Physiol
1980;14:256–64.
[56] Butler MJ, Day AW. Fungal melanins: a review. Can J Microbiol
1998;44:1115–36.
223
[57] Hodgson R, Peterson WH, Riker AJ. The toxicity of polysaccharides
and other large molecules to tomato cuttings. Phytopathology
1949;39:47–62.
[58] Buliga GS, Brant DA. Theoretical interpretation of the unperturbed
aqueous solution configuration of pullulan. Int J Biol Macromol
1987;9:77–86.
[59] Burton BA, Brant DA. Comparative flexibility, extension, and conforma-
tion of simple polysaccharide chains. Biopolymers 1983;22:1769–92.