Eur J Plant Pathol (2008) 122:197–206
DOI 10.1007/s10658-008-9334-x
Effects of garlic (Allium sativum) juice containing allicin
on Phytophthora infestans and downy mildew of cucumber
caused by Pseudoperonospora cubensis
Daniela Portz & Eckhard Koch & Alan J. Slusarenko
Received: 29 January 2008 / Accepted: 8 May 2008
# KNPV 2008
Abstract The volatile antimicrobial substance allicin
(diallylthiosulphinate) is produced in garlic when the
tissues are damaged and the substrate allicin (Sallyl-L-cysteine sulphoxide) mixes with the enzyme
alliin-lyase (E.C.4.4.1.4). Allicin undergoes thioldisulphide exchange reactions with free thiol groups
in proteins and it is thought that this is the basis of its
antimicrobial action. At 50 μg ml-1, allicin in garlic
juice inhibited the germination of sporangia and cysts
and subsequent germ tube growth by Phytophthora
infestans both in vitro and in vivo on the leaf surface.
Disease severity in P. infestans-infected tomato seedlings was also reduced by spraying leaves with garlic
juice containing allicin over the range tested (55–
110 μg ml−1) with an effectiveness ranging from
approximately 45–100%. Similarly, in growth room
experiments at concentrations from 50–1,000 μg ml−1,
allicin in garlic juice reduced the severity of cucumber
D. Portz : A. J. Slusarenko (*)
Department of Plant Physiology,
RWTH Aachen University,
52056 Aachen, Germany
e-mail: Alan.slusarenko@bio3.rwth-aachen.de
E. Koch
Federal Research Centre for Cultivated Crops (JKI),
Institute for Biological Control,
Heinrichstr. 243,
64287 Darmstadt, Germany
downy mildew caused by Pseudoperonospora cubensis
by approximately 50–100%. These results suggest a
potential for developing preparations from garlic for
use in specialised aspects of organic farming, e.g. for
reducing pathogen inoculum potential and perhaps
for use under glass in horticulture.
Keywords Natural fungicides . Tomato leaf blight .
Plant antibiotic . Antimicrobial . Phytoanticipin
Introduction
Downy mildews and diseases caused by oomycetes in
general are among the most destructive and economically important agricultural problems world-wide.
According to Gisi (2002) almost 17% of the world
fungicides market in 1996 was for agents used in
downy mildew control. Effective control by planting
resistant varieties is in many cases not possible and
disease management problems have been compounded
by the emergence of fungicide-resistant/tolerant variants of several oomycete pathogens (Gisi 2002; Urban
and Lebeda 2006, 2007; Urban et al. 2007). Furthermore, the increasing public demand for organicallygrown produce, and the intended phasing out by the
EU of the use of copper-containing formulations, has
precipitated an urgent need for alternative control
methods. In this regard resistance-inducing treatments
and substances conditioning systemic acquired resis-
198
Eur J Plant Pathol (2008) 122:197–206
tance (SAR) are considered an alternative (MauchMani 2002; Körösi et al. 2007) and there is increased
interest in developing treatment strategies based on
natural plant defence products (KonstantinidouDoltsinis and Schmitt 1998; Konstantinidou-Doltsinis
et al. 2006; Slusarenko et al. 2008).
We have reported previously that the natural
antimicrobial substance allicin, which is a volatile
phytoanticipin produced in garlic (Allium sativum)
upon wounding, is active against a broad range of
phytopathogenic organisms in vitro and in planta
(Curtis et al. 2004) and indeed there are several
reports of garlic preparations containing allicin being
used to treat plant disease (e.g. Ark and Thompson
1959; Russell and Mussa 1977). Allicin (diallylthiosulphinate) is produced in garlic when the substrate
alliin (S-allyl-L-cysteine sulphoxide) mixes with the
enzyme alliinase (alliin-lyase, E.C.4.4.1.4; see diagram below). The antimicrobial
2
O
H
S
C
O
alliinase
H2O
NH2
COOH
alliin
S
+ 2pyruvate + 2NH 3
S
allicin
activity of garlic juice had long been known and
Cavallito and Bailey (1944) showed that this activity
was due to allicin, which they reported to be as active
against test bacteria as penicillin. Allicin crosses the
cell membrane easily and undergoes thiol-disulphide
exchange reactions with free thiol groups in proteins
(see diagram below). It is thought that these properties
are the basis of its antimicrobial action (Miron et al.
2000; Rabinikov et al. 1998). Allicin thus has several
O
2R
SH +
thiol
S
S
allicin
2R
S
S
+ H2O
mixed disulphide
targets in the cell and this makes it difficult for
organisms to develop resistance to it.
The use of natural products in plant protection,
either directly or as starting points for targeted
enhancement of desirable qualities by industry, has
been reviewed recently (Slusarenko et al. 2008) and
the current paper presents results using garlic juice
containing allicin to combat diseases caused by the
important plant pathogenic oomycetes Phytophthora
infestans and Pseudoperonospoa cubensis. The effect
of allicin in garlic juice was tested on the germination
rate and subsequent germ tube growth of sporangia
and cysts of P. infestans in vitro and in vivo on the
surface of tomato leaves. The effectivity of allicin in
garlic juice was tested in reducing leaf infection of
tomato seedlings by P. infestans, and cucumber
seedlings by Pseudoperonospora cubensis was also
tested under growth room conditions.
Materials and methods
Cucumber/Pseudoperonospora cubensis
Plant cultivation
Plants were cultivated in 8×8 cm plastic pots filled with
a 1:2 parts mixture of sand: commercial potting substrate
(Fruhstorfer® Erde Typ T; Industrie-Erdenwerk Archut,
Lauterbach). Twelve seeds of Cucumis sativus cv.
Chinesische Schlange were sown in each pot. The pots
were watered carefully and kept in a growth room at
20°C (cycle of 16/8 h light/dark). After 1 week the
seedlings were transplanted to fresh pots (one plant per
pot).
Inoculation
P. cubensis was maintained on plants grown as
outlined above. Fresh inoculum was prepared from
plants 10 days after previously being inoculated with
P. cubensis. Plants were incubated overnight in a
moist chamber to encourage sporulation and sporangia were harvested by washing the lower leaf surface
with water. The resulting suspension was adjusted to
5×103 sporangia ml−1 using a haemocytometer.
Plants were harvested approximately 3 weeks after
transplanting when the second true leaf was expanded. The upper, non-expanded leaves were excised and
the first and second leaves sprayed with the treatment
solution on both sides using a chromatography
sprayer. Control plants were sprayed with water or
with 0.2% Cuprozin Flüssig ™ (460.6 g l−1 copper
hydroxide) (Spiess-Urania, Hamburg). After 24 h the
first and second leaves were inoculated using a
chromatography sprayer on both sides with a suspension of P. cubensis sporangia (5×103 ml−1). The pots
were then incubated overnight at 15°C in a moist
chamber and the following day returned to the growth
room. Disease was rated 2 weeks after inoculation by
Eur J Plant Pathol (2008) 122:197–206
estimating the percentage of the affected leaf area. The
effectivity of the treatment was calculated according to
Abbott (1925):
% Effectivity ¼
affected leaf area ðcontrolÞ affected leaf area ðtreatmentÞ
100
affected leaf area ðcontrolÞ
199
allowed to dry (approx. 2 h) before being sprayinoculated. As a soil drench a single application of
5 ml of the appropriate dilution of garlic juice was
applied per 7×7 cm pot containing a single plant.
Five to seven intact tomato seedlings were inoculated
per experiment and each experiment was repeated at
least three times. A representative set of results for
each experiment is shown.
Tomato/Phytophthora infestans
Preparation of garlic juice
Plant cultivation
Tomato seeds (cv. Hoffmanns Rentita®, Schmitz &
Laux GmbH, Hilden, Germany) were sown in
seedling trays for germination in moist potting
compost covered with fine moistened sand and
incubated at 22°C in a light/dark cycle of 16/8 h.
After germination, 1 week-old seedlings were transferred to individual 7×7 cm pots and grown on for a
further 2 weeks.
Inoculation
The P. infestans isolate used in this work was kindly
donated by Bayer CropScience AG, Monheim. The
virulence of the isolate was ensured by regular
passaging through potato tuber discs. Phytophthora
infestans was cultivated under sterile conditions on
tomato juice agar (TJA) at 18°C in the dark (TJA=3 g
CaCO3, 12 g PDB (Difco™), 20 g agar (AppliChem
GmbH), 200 ml tomato juice (Fa. Krings Fruchtsaft
GmbH, Mönchengladbach) made up to a volume of
1 l and autoclaved at 121°C for 15 min). Phytophthora infestans inoculum was prepared by washing
the surface of 8 day-old Petri plate cultures with
cold (10°C), sterile deionized water and sieving
through a plastic kitchen sieve. Sporangia were
adjusted to a concentration of 4–5×104 ml−1 with a
haemocytometer. Zoospores were released from sporangia after approximately 2 h at 10–12°C. After
spray-inoculation, plants were placed in a seedling
tray and covered with a transparent plastic lid in the
growth chamber at 20°C with a light/dark cycle of
16/8 h.
Treatment with garlic juice
Unless otherwise stated, 3 week-old tomato plants
were sprayed with diluted garlic juice and the leaves
Garlic bulbs were purchased from the supermarket
and stored at 4°C in the dark until required. Axillary
buds from the composite garlic bulb were peeled,
weighed and a domestic juicer (Turmix Fabr. Nr.
1068, Turmix AG, 8645 Jona, Switzerland) was used
to extract the juice. The juice was poured into a sterile
50 ml Falcon tube and centrifuged at 5,000 rpm
(3,000×g) for 10 min in order to separate the majority
of the pulp from the liquid (Megafuge 1.0R, Heraeus
Instruments, Osterode, Germany). Floating debris was
removed from the top of the liquid with a spatula and
discarded. Filtering under pressure separated the
remaining pulp from the pure extract (Diaphragm
Vacuum Pump, Vacuubrand GmbH + Co., Wertheim,
Germany). The filtrate was transferred into a second
sterile 50 ml Falcon tube and sealed. The average
yield was approximately 1 ml of extract from 3 g
fresh weight of garlic tissue and typically contained
approximately 5 mg ml−1 allicin (determined by
HPLC). The garlic extract was used either immediately after appropriate dilution or stored undiluted at
10°C for a maximum of 2 weeks. Dilutions were
carried out with de-ionized water. Appropriate
amounts of stock solution to give the required end
dilution in Petri plates were incorporated into agar
medium kept just molten at 45°C. Plates were poured
immediately after adding and mixing the stock.
Determination of allicin by HPLC
The method used was based on that of Krest and
Keusgen (2002). Garlic juice was diluted 1:10 with
HPLC-grade water and 1.5 ml of a 0.05 mg ml−1
solution (in methanol) of butyl-4-hydroxybenzoate
(internal standard). To protect the column, this
mixture was first filtered through a polyethersulfonmembrane (0.2 μm pore size, Steriflip, Millipore)
before 20 μl were injected into the HPLC (Kontron
200
system with diode array detector, Kontron Instruments GmbH, Neufahrn). Using the HPLC software
Geminyx (version 1.91) a mixed gradient elution
(solvent A, 30% (v/v) HPLC grade methanol adjusted
to pH 2.0 with 85% (v/v) orthophosphoric acid;
solvent B, 100% HPLC grade methanol) was performed. Spectra were recorded between 200–600 nm
during elution with detection at 254 nm for the
chromatogram.
Effect of garlic juice on P. infestans sporangium
and cyst germination in vitro
Droplets (20 μl) of inoculum suspension, prepared as
described above and containing sporangia and zoospores, were pipetted onto the surface of 1% agar
containing 50 μg ml −1 allicin. Control plates
contained no allicin. Plates were sealed with Micropore™-tape and incubated in a plastic container with
moistened tissue paper at 18°C in the dark for 4 h.
Germination rate and germ tube length were measured
using a microscope (Leica DM R) at 50- to 200-fold
magnification. At least 50 sporangia or encysted
zoospores were scored for germination per plate and
photographed using a JVC digital camera (KY-F75U)
and Discus software (Version 32, Hilgers Co.,
Königswinter, Germany). Germ tube lengths of at
least 15 germinated sporangia or cysts were measured
per plate.
Effect of garlic juice on P. infestans sporangium
and cyst germination in vivo on tomato leaves
After spraying 3-week-old tomato plants with diluted
garlic juice containing 50 μg ml−1 allicin and
allowing them to dry, leaves were excised and placed
in plastic boxes (12×12 cm) on moistened tissue
paper. Droplets (20 μl) of sporangial or cyst suspensions were then pipetted onto the leaves and the lids
placed on the boxes for incubation for 4 h in the dark
at 20°C. The leaf lamina under the droplets was then
excised and stained with acid fuchsin (modified after
Carmichael 1955). Excised leaf segments were fixed
and decolourised for 48 h at 60°C in aqueous chloral
hydrate (2.5 g ml−1). Leaf segments were then stained
for 1–2 h in 0.01% acid fuchsin-lactophenol solution
and de-stained in 50% (v/v) glycerol before viewing
using a confocal laser-scanning microscope (Leica
TCS SP, using Leica software TCS NT) at 630- to
Eur J Plant Pathol (2008) 122:197–206
1000-fold magnification (excitation 543 nm; emission
filter 575–640 nm, 63× PL APO w, and 100× PL
FLUOTAR oil objective lenses).
Statistical treatments
Raw data were first tested for normal distribution and
variance homogeneity using Sigmastat® 3.1 (SYSTAT
software 2004) to a limit of P≤0.05. If the data
showed normal distribution and variance homogeneity they were subjected to parametric statistic tests to
show significant differences (t-test or one-way
ANOVA) to a probability of P≤0.05. Non-normal
data were analysed with either the Mann–Whitney
Rank Sum Test for two groups or the Kruskal–Wallis
ANOVA on Ranks for more than two groups. If these
treatments pointed to a significant difference between
groups, a post hoc test (Dunn’s or Tukey’s) was used
to determine which groups differed significantly at the
P≤0.05 level.
Results
Pseudoperonospora cubensis/Cucumis sativus
pathosystem
Cucumber plants were sprayed with either dilutions of
garlic juice, water (untreated controls) or Cuprozin™,
and spray-inoculated the next day with a suspension
of sporangia of P. cubensis (Fig. 1A). Two weeks
after inoculation infected leaf areas were estimated
(Fig. 1B, Table 1). Dilutions of garlic juice over a
wide range of allicin concentrations (50–1,000 μg
ml−1) led to a reduction in disease severity which
compared favourably with the degree of disease
control achieved with a copper-containing commercial fungicide (Cuprozin™).
Phytophthora infestans/Lycopersicon esculentum
pathosystem
Effects of garlic juice on P. infestans germination
and growth in vitro
The effect of garlic juice on P. infestans in vitro was
assessed by investigating the effects on sporangial
and cyst germination and on germ tube growth. Garlic
juice (50 μg ml−1 allicin) caused a clear reduction in
Eur J Plant Pathol (2008) 122:197–206
201
100
Water
50 µg ml–1 allicin
Germination [%]
80
60
a
a
40
b
b
20
0
Sporangia
Cysts
Fig. 2 Influence of garlic juice in agar (50 μg ml−1 allicin) on
the germination of sporangia and encysted zoospores of P.
infestans (in vitro). Means of nine replicate Petri plates of
sporangia and cyst preparations. Columns which differ significantly from one another are marked with a different letter (ttest, P≤0.05)
Effects of garlic juice on P. infestans germination
and growth in vivo
Fig. 1 Leaf of cucumber showing A, the spray inoculation
procedure and B, symptoms 14 days after inoculation with P.
cubensis (5×103 sporangia ml−1)
the germination of encysted zoospores and of sporangia under conditions where they germinate directly
with a germ tube (i.e. behave like conidia) (Fig. 2).
Hyphal growth from germinated sporangia or cysts
was also reduced by the presence of garlic juice in the
medium (50 μg ml−1 allicin) (Fig. 3).
The behaviour of sporangia and cysts on the tomato leaf
surface after treatment with garlic juice is shown in
Fig. 4. It can be seen that the inhibitory in vitro effects
of garlic juice are mirrored in the in vivo behaviour of
sporangia and cysts on the tomato leaf surface.
Effects of garlic juice on disease severity in tomato
leaves inoculated with P. infestans
To assess whether the inhibitory effects of garlic juice
on P. infestans observed in vitro and in vivo on the
leaf surface translated into an effect on disease
development, a systematic investigation on tomato
Table 1 Effect on disease severity of spraying garlic juice containing allicin at the concentrations shown onto leaves of 40-day-old
cucumber plants 24 h prior to spray inoculation with 5×103 conidia ml−1 of P. cubensis
Treatment
Water control
Allicin 1000 μg ml−1
Allicin 500 μg ml−1
Allicin 200 μg ml−1
Allicin 100 μg ml−1
Allicin 50 μg ml−1
Cuprozin™ (0.2%)c
Average effectivitya (%)
Infected leaf area (%) ±SD
Experiment 1
Experiment 2
Experiment 3
73.3±17.1
N.T.b
N.T.
3.7
19.0
8.2
N.T.
33.8±8.9
0.2
1.0
2.8
N.T. ±
N.T. ±
20.0±9.8
81.8±9.9
0.4
1.0
2.0
N.T.
N.T.
20.0±11.9
Plants (four per experiment, eight leaves in total) were scored 2 weeks after inoculation.
a
According to Abbott (1925), see “Materials and methods” section
b
N.T. Not tested
c
Equivalent to 0.92 g Cu(OH)2 l−1
>99
96–98
84–94
55
52
41–76
202
A
Average germtube length [µm]
Fig. 3 Influence of garlic
juice on germ tube growth
from germinating sporangia
and encysted zoospores of
P. infestans (in vitro). A
Means of ∼75 measurements (sporangia) and ∼135
measurements (cysts). Columns which differ significantly from one another are
marked with a different letter (Mann–Whitney Test,
P≤0.05). B Untreated sporangia. C Untreated cysts. D
Sporangia on agar incorporating garlic juice to give a
final concentration of 50 μg
ml−1 allicin. E Cysts on agar
incorporating garlic juice to
give a final concentration of
50 μg ml−1 allicin. Bar=
50 μm
Eur J Plant Pathol (2008) 122:197–206
800
a
Water
50 µg ml–1 allicin
700
600
500
400
300
200
a
b
100
b
0
leaf infections was carried out. Firstly, potential
phytotoxic effects of garlic juice on leaves were
monitored. As shown in Table 2, spraying tomato
leaves of 3-week-old plants with dilutions of garlic
juice containing 200–800 μg ml−1 allicin led to leaf
damage in category 2 (<2.5% of the leaf area showing
chlorosis or necrosis), the least severe, and only at the
highest concentration tested.
The effect on disease development of spraying
tomato leaves with a single application of garlic juice
containing a range of allicin concentrations 2 h before
inoculation with P. infestans is shown in Fig. 5 (for a
photograph showing the appearance of control and
allicin-treated leaves see Fig. 6 in Slusarenko et al.
Sporangia
Cysts
B
D
C
E
2008). In the experiment shown in Fig. 5, control
tomato plants had lesions covering 77% of the leaf
area 4 days after inoculation (dai). As can be seen,
spraying with garlic juice very effectively reduced
disease development, with a 1:50 dilution (110 μg
ml−1 allicin) suppressing lesion development completely (Fig. 5).
The effectivity of a single pre-inoculation spray
with garlic juice containing a low concentration of
allicin (60 μg ml−1), which did not completely
suppress disease development, decreased with time
but was still apparent 10 dai (data not shown). Thus,
in plants treated with 60 μg ml−1 allicin the affected
leaf area increased from 16% at 4 days to 37% at 10
Eur J Plant Pathol (2008) 122:197–206
203
Fig. 4 Influence of garlic
juice on germ tube growth
from germinating sporangia
and encysted zoospores of
P. infestans on the tomato
leaf surface (in vivo) shown
after acid fuchsin staining
under a confocal laser scanning microscope excitation,
543 nm, emission, 575–
640 nm; Scale bars = 50 µm
(A & B), 25 µm (C & D). A
Untreated sporangia showing germination and healthy
germ tube growth. B Sporangia on a leaf sprayed
with garlic juice (50 μg
ml−1 allicin) approximately
2 h prior to inoculation,
germinated at a lower rate
and has formed abnormal
germ tubes with reduced
growth. C Untreated cyst
showing normal germ tube
growth. D Ungerminated
cysts on a leaf sprayed with
garlic juice (50 μg ml−1
allicin)
dai. In the untreated controls, however, the infected
leaf area was 60% after 4 days and increased to 63%
by 10 dai.
The effect of various garlic juice application times
in relation to the time of inoculation with P. infestans
was investigated and it was found that the nearer to
the inoculation time that allicin was sprayed, the more
effective a given dosage was in suppressing disease
development (Fig. 6). In contrast, spraying leaves
with garlic juice 24 h after inoculation had little
effect. Direct spraying onto leaves was also compared
with a single application as a soil drench. As can be
seen in Fig. 7, allicin was more effective when
sprayed on the leaves than when applied to the soil.
Discussion
Curtis et al. (2004) previously reported that dilutions
of garlic juice containing allicin were effective in
reducing the production of conidiophores and
Table 2 Phytotoxicity scores for individually potted 3-weekold tomato seedlings sprayed to run-off with dilutions of garlic
juice containing various concentrations of allicin
Concentration of allicin in garlic juice
(μg ml−1)
0
200
400
800
Phytotoxicity
categorya
1
1
1
2
a
Phytotoxicity categories (Gorog née Privitzer et al. 1988): 1=
no damage, 2=<2.5% leaf area damaged (showing chlorosis or
necrosis), 3=<5% leaf area damaged. The scale progresses to
9=100% leaf damage. Scores<2 are considered acceptable in
screens of potential candidates for plant protection substances.
Plants were allowed to dry, and then pots were placed under
plastic hoods for 4 days before the hoods were removed. Plants
were incubated in a growth chamber (22°C, cycles of 18 h light
6 h dark) and scored 6 days after spraying.
204
Eur J Plant Pathol (2008) 122:197–206
120
bc
Effectivity [%]
Effectivity [%]
100
80
60
ab
40
20
b
100
c
bc
80
b
ab
a
a
60
40
20
a
0
Control
1:100
1:85
1:65
1:50
dilution
dilution
dilution
dilution
55 µg ml–1 65 µg ml–1 85 µg ml–1 110 µg ml–1
allicin
allicin
allicin
allicin
Fig. 5 Dose-dependency of disease control by allicin in garlic
juice in the P. infestans/tomato leaf pathosystem. Three-weekold plants (cv. Hoffmans Rentita) were sprayed with garlic
juice, the leaves allowed to dry (approx. 2 h) and then sprayinoculated with 4–5×104 sporangia ml−1. The effectivity of
treatment (Abbot 1925) is shown at 4 dai. Columns which
differ significantly from one another are marked with a different
letter (Dunn’s Test, P≤0.05)
oospores in downy mildew of Arabidopsis caused by
Hyaloperonospora parasitica. In the present study
these observations are extended to show that macroscopic disease symptoms of cucumber downy mildew
can be markedly reduced by spraying the leaves with
garlic juice containing a range of allicin concentrations 24 h prior to inoculation. The disease reduction
compared very favourably with a commercial copperfungicide treatment and suggests that development of
garlic products for at least small-scale application
such as in glasshouse situations might be feasible and
100
b
Effectivity [%]
80
60
a
40
a
20
a
0
48 h before
inoculation
24 h before
inoculation
2 h before
inoculation
24 h after
inoculation
Fig. 6 Influence of time between treatment with garlic juice
(70 μg ml−1 allicin) and time of inoculation on effectivity in the
P. infestans/tomato pathosystem at 4 dai. Columns which differ
significantly from one another are marked with a different letter
(Dunn’s Test, P≤0.05)
0
250 µg ml–1 100 µg ml–1 150 µg ml–1 100 µg ml–1 50 µg ml–1
allicin
allicin
allicin
allicin
allicin
drench
drench
spray
spray
spray
Fig. 7 Comparison of the effectivity of garlic juice containing
allicin as a soil drench or a foliar spray in the P. infestans/
tomato pathosystem at 4 dai. Columns which differ significantly from one another are marked with a different letter (Tukey’s
Test, P≤0.05)
desirable as an alternative to standard treatments
(Fig. 1, Table 1). Resistance of P. cubensis against
conventional fungicide treatments is increasing (Urban and Lebeda 2006, 2007; Urban et al. 2007) and
because allicin appears to have a multi-site mode of
action (Portz et al. 2005; Slusarenko et al. 2008) it
will presumably be difficult for pathogens to mutate
to resistance against it, thus conferring a strong
advantage on allicin-based disease treatments.
The inhibitory effect of allicin on the vegetative
mycelial growth of P. infestans and the reduction of
potato tuber colonization by allicin in the gas phase
have been reported previously (Curtis et al. 2004).
Now, the inhibitory effects of garlic juice containing
allicin on the germination of sporangia and encysted
zoospores and subsequent reduction in germ tube
growth, both in vitro and on the tomato leaf surface
(Figs. 2, 3, 4) are reported. These effects presumably
contribute to the reduction in infection seen in
inoculated tomato seedlings (Fig. 5). The tomato
leaf/P. infestans pathosystem was used in preference
to potato/P. infestans because it is easier to work with
in the laboratory. Nevertheless, since it appears that
the effect of allicin is directly against the pathogen,
rather than via an induced resistance mechanism
(Curtis et al. 2004), it seems likely that a similar
degree of control might be expected in the potato/P.
infestans pathosystem, particularly in view of the
effects of allicin in reducing tuber colonisation at least
under controlled conditions (Curtis et al. 2004). The
effectivity of garlic juice in reducing disease in
tomato leaves was very high and approached 100%
Eur J Plant Pathol (2008) 122:197–206
at an allicin concentration of 110 μg ml−1 (Fig. 5). In
fungicide screening, substances are usually only
considered for further development if they do not
cause leaf damage above category 2 (<2.5% leaf area
affected) on a scale of 1–9 (Gorog née Privitzer et al.
1988) (see Table 2). Garlic juice was assessed at
various dilutions for phytotoxicity, and disease control was achieved at allicin concentrations well below
those where phytotoxicity was observed (Table 2,
Fig. 5). Thus, allicin in garlic juice would not be
excluded in a conventional screening programme
based on this criterion.
The effectivity of the allicin treatment in reducing
disease on tomato seedlings is more pronounced in
the early stages after treatment. If allicin is working
mainly via a reduction of successful infections by
killing a certain proportion of the spores and
subsequently by suppressing germ tube growth from
surviving propagules, then a time-lag in disease
development would be expected until inoculum levels
had reached those present before the sanitation
treatment. However, the dynamics of disease development in fungicide-treated plants are difficult to
model and disease development often deviates from
the ideal mathematical description (Jeger 1987). In
control plants not treated with allicin, the disease level
4 dai was already high and this increased only
marginally in subsequent days. In the allicin-treated
plants the affected leaf area increased from 16% at
4 days to 37% by 10 dai. Thus, even a single
treatment with allicin at a dose (60 μg ml−1) below
that necessary to completely eradicate disease
(∼110 μg ml−1, see Fig. 5) is already effective at
reducing the rate of disease progress over a substantial time period.
The data presented in Fig. 6 show the effectivity of
a single allicin treatment in relation to the time of
inoculation and support a low-persistence, contactfungicide type of effect for allicin. Thus, the effectivity of the treatment increases with decreasing time
before inoculation (from 48 to 24 h), is maximal when
inoculation takes place approximately 2 h after
treatment with garlic juice, and is least effective at
later times after inoculation (e.g. 24 h) when the
pathogen has already penetrated the leaf and is
perhaps less easily accessed by allicin. In this regard
the kinetics of allicin behaviour on the leaf surface,
and its uptake by the leaf, are aspects which need
further investigation.
205
In downy mildew of Arabidopsis it was shown that
treatment of the plant with garlic juice did not lead to
the accumulation of SAR markers (Curtis et al. 2004)
and the authors suggested that garlic juice was
exerting its antimicrobial effect directly on the
pathogen rather than via inducing SAR in the plant.
The data presented in Fig. 6 for tomato support this
conclusion and extend it to a further pathosystem.
Interestingly, in the tomato/P. infestans pathosystem, applying garlic juice as a soil drench was also
effective at reducing disease levels, although a better
degree of control was achieved with lower concentrations of allicin as a direct spray on the leaves
(Fig. 7). As stated earlier, allicin appears to act
directly against the pathogen and it is unclear whether
the disease reduction after applying garlic juice as a
soil drench is due to the action of allicin against the
pathogen via the gas phase, or whether allicin is also
taken up via the roots and transported systemically
within the plant. Allicin is readily membrane-permeable (Miron et al. 2000; Rabinikov et al. 1998;
Slusarenko et al. 2008) and could therefore enter the
symplast in the roots, but whether it is transported
within the plant is unknown at present. In this regard,
it is perhaps important to mention that it is difficult to
quantify allicin in the gas phase because the temperature of the injection port in the GC is too high and
leads to modifications producing other polysulphides
(Block 1992).
The potential for allicin in garlic juice to be used as
an effective control agent against diseases caused by
oomycetes is clear, although there is scope for
optimisation of treatment regimes, and field testing
is certainly necessary. Very clearly, transfer from the
laboratory to the field/glasshouse is a stumbling block
which many otherwise promising compounds fail to
negotiate successfully (Slusarenko et al. 2008). This
may also prove true for allicin in garlic juice. Also, it
will be necessary to carry out organoleptic assessment
of harvested plant parts to ensure the absence of
undesireable flavour notes in any development of
garlic products for plant protection. Neither garlic
juice nor allicin are named presently as plant
protection substances specifically permitted for organic farming in the EU (Directive 2092/91). However, it is not likely that these substances, which are a
common foodstuff or a component thereof, have
properties that would not allow them to be added to
the list in the future. Furthermore, chemical modifi-
206
cation of allicin, which has an activity comparable to
several conventional antibiotics (Cavallito and Bailey
1944; Curtis et al. 2004; Slusarenko et al. 2008), to
enhance its desirable properties and reduce its
undesirable ones, might even lead to a new multitarget plant protection compound useable in conventional agriculture and horticulture.
Acknowledgements RWTH Aachen University provided a
student assistantship (D.P.) and financial support. Technical
assistance by Ulrike Noll (Aachen) and Monika Eitzen-Ritter
(Darmstadt) is gratefully acknowledged. Ales Lebeda and
Nikolaus Schlaich are thanked for critical reading of the
manuscript.
References
Abbott, W. S. (1925). A method for computing the effectiveness
of an insecticide. Journal of Econonomic Entomology, 18,
265–267.
Ark, P. A., & Thompson, J. P. (1959). Control of certain
diseases of plants with antibiotics from garlic (Allium
sativum L.). Plant Disease Reporter, 43, 276–282.
Block, E. (1992). The organosulfur chemistry of the genus
Allium – implications for the organic chemistry of sulfur.
Angewandte Chemie International Edition in English, 31,
1135–1178.
Carmichael, J. W. (1955). Lacto-fuchsin: A new medium for
mounting fungi. Mycologia, 47, 611.
Cavallito, C. J., & Bailey, H. J. (1944). Allicin, the antibacterial
principle of Allium sativum I. Isolation, physical properties
and antibacterial action. Journal of the American Chemical
Society, 66, 1950–1951.
Curtis, H., Noll, U., Störmann, J., & Slusarenko, A. J. (2004).
Broad-spectrum activity of the volatile phytoanticipin
allicin in extracts of garlic (Allium sativum L.) against
plant pathogenic bacteria, fungi and Oomycetes. Physiological and Molecular Plant Pathology, 65, 79–89.
Gisi, U. (2002). Chemical control of downy mildews. In P. T. N.
Spencer-Phillips, U. Gisi, & A. Lebeda (Eds.), Advances in
downy mildew research (pp. 119–159). Dordrecht: Kluwer
Academic Publishers.
Gorog née Privitzer, K., Kocsis née Bagyi, M., Orosz, G.,
Torok, T., Laszlo, K., Toth, I., et al. (1988). Herbicide
compositions containing a nitrile derivative as antidote.
US Patent 4787933.
Jeger, M. J. (1987). Modelling the dynamics of pathogen
populations. In M. S. Wolfe, & C. E. Caten (Eds.),
Populations of plant pathogens: Their dynamics and
genetics (pp. 91–107). Oxford: Blackwell Scientific Publishers.
Konstantinidou-Doltsinis, S., Markellou, E., Fanouraki, M. N.,
Kasselaki, A.-M., Koumaki, C. M., Schmitt, A., et al.
(2006). Efficacy of MilsanaÒ, a formulated plant extract
from Reynoutria sachalinensis, against powdery mildew
Eur J Plant Pathol (2008) 122:197–206
of tomato (Leveillula taurica) (Lév.) Arn. Biocontrol, 51,
375–392.
Konstantinidou-Doltsinis, S., & Schmitt, A. (1998). Impact of
treatment with plant extracts from Reynoutria sachalinensis
(F. Schmidt) Nakai on disease severity of powdery mildew
and yield in cucumber under Greek conditions. Crop
Protection, 17, 649–656.
Körösi, K., Lazar, N., & Viranyi, F. (2007). Resistance response
to downy mildew (Plasmopara halstedii) in sunflower
(Helianthus annuus) activated by chemical inducers. In
A. Lebeda, & P. T. N. Spencer-Phillips (Eds.), Advances in
downy mildew research (vol. 3 (pp. 237–241). Czech
Republic: Olomouc, Palacky University (Publishers).
Krest, I., & Keusgen, M. (2002). Biosensoric flow-through
method for the determination of cysteine sulfoxides.
Analytica Chimica Acta, 469, 155–164.
Mauch-Mani, B. (2002). Host resistance to downy mildew
diseases. In P. T. N. Spencer-Phillips, U. Gisi, & A. Lebeda
(Eds.), Advances in downy mildew research (pp. 59–83).
Dordrecht: Kluwer Academic Publishers.
Miron, T., Rabinikov, A., Mirelman, D., Wilchek, M., &
Weiner, L. (2000). The mode of action of allicin: Its ready
permeability through phospholipid membranes may contribute to its biological activity. Biochimica et Biophysica
Acta, 1463, 20–30.
Portz, D., Noll, U., & Slusarenko, A. J. (2005) Allicin from
garlic (Allium sativum L.): A new look at an old story.
Proceedings of the 14th International Reinhardsbrunn
Symposium, “Modern Fungicides and Antifungal
Compounds IV”. Eds. H.-W. Dehne, U. Gisi, K. H. Kuck,
P. E. Russell & H. Lyr. British Crop Protection Council,
Alton, U.K. pp. 227–234.
Rabinikov, A., Miron, T., Konstantinovski, L., Wilchek, M.,
Mirelman, D., & Weiner, L. (1998). The mode of action of
allicin: Trapping of radicals and interaction with thiolcontaining proteins. Biochimica et Biophysica Acta, 1379,
233–244.
Russell, P. E., & Mussa, A. E. A. (1977). The use of garlic
(Allium sativum) extracts to control foot rot of Phaseolus
vulgaris caused by Fusarium solani f. sp. phaseoli. Annals
of Applied Biology, 86, 369–372.
Slusarenko, A. J., Patel, A., & Portz, D. (2008). Control of
plant diseases by natural products: Allicin from garlic as a
case study. European Journal of Plant Pathology, 121,
313–322.
Urban, J., & Lebeda, A. (2006). Fungicide resistance in
cucurbit downy mildew—methodological, biological
and population aspects. Annals of Applied Biology, 149,
63–75.
Urban, J., & Lebeda, A. (2007). Variation for fungicide
resistance in Czech populations of Pseudoperonospora
cubensis. Journal of Phytopathology, 155, 143–151.
Urban, J., Lebeda, A., & Pejchar, M. (2007). Differential
sensitivity to fungicides in Czech populations of cucumber
downy mildew (Psedoperonospora cubensis). In A.
Lebeda, & P. T. N. Spencer-Phillips (Eds.),Advances in
downy mildew research (vol. 3 (pp. 251–255). Czech
Republic: Olomouc, Palacky University (Publishers).