Defence Signalling Triggered by Flg22 and Harpin Is Integrated into a Different Stilbene Output in Vitis Cells
Citation: Chang X, Nick P (
Defence Signalling Triggered by Flg22 and Harpin Is Integrated into a Different Stilbene Output in Vitis Cells
Xiaoli Chang 0
Peter Nick 0
Ching-Hong Yang, University of Wisconsin-Milwaukee, United States of America
0 Molecular Cell Biology, Botanical Institute 1, Karlsruhe Institute of Technology , Karlsruhe , Germany
Plants can activate defence to pathogen attack by two layers of innate immunity: basal immunity triggered by pathogenassociated molecular pattern (PAMP) triggered immunity (PTI) and effector-triggered immunity (ETI) linked with programmed cell death. Flg22 and Harpin are evolutionary distinct bacterial PAMPs. We have previously shown that Harpin triggers hypersensitive cell death mimicking ETI in Vitis rupestris, but not in the Vitis vinifera cultivar 'Pinot Noir'. In contrast, the bacterial PAMP flg22 activating PTI does not trigger cell death. To get insight into the defence signalling triggered by flg22 and Harpin, we compared cellular responses upon flg22 and Harpin treatment in the two Vitis cell lines. We found that extracellular alkalinisation was blocked by inhibition of calcium influx, and modulated by pharmacological manipulation of the cytoskeleton and mitogen-activated protein kinase activity with quantitative differences between cell lines and type of PAMPs. In addition, an oxidative burst was detected that was much stronger and faster in response to Harpin as compared to flg22. In V. rupestris, both flg22 and Harpin induced transcripts of defence-related genes including stilbene synthase, microtubule disintegration and actin bundling in a similar way, whereas they differed in V. vinifera cv. 'Pinot Noir'. In contrast to Harpin, flg22 failed to trigger significant levels of the stilbene trans-resveratrol, and did not induce hypersensitive cell death even in the highly responsive V. rupestris. We discuss these data in a model, where flg22- and Harpin-triggered defence shares a part of early signal components, but differs in perception, oxidative burst, and integration into a qualitatively different stilbene output, such that for flg22 a basal PTI is elicited in both cell lines, while Harpin induces cell death mimicking an ETI-like pattern of defence.
Plants employ two distinct layers of immunity to encounter
pathogen invasion . The first, evolutionarily ancient, layer
involves the perception of conserved pathogen structures termed
pathogen-associated molecular patterns (PAMPs) at the plasma
membrane through conserved and ubiquitous receptors generally
defined as pattern recognition receptors (PRRs). Binding to these
receptors initiates an active defence response, so-called
PAMPtriggered immunity (PTI), in both host and non-host plants. In a
second round of host-pathogen warfare, several microbial
pathogens have already developed the ability to secrete effector
proteins into the cytoplasm using type-III secretion systems (T3SS)
in bacteria. These effectors suppress PTI and result in the
effectortriggered susceptibility (ETS) ,. In response to pathogen
effectors, plants have acquired additional receptors that specifically
recognise the effectors, establishing a second layer of immunity
known as effector-triggered immunity (ETI). ETI is often
associated with a hypersensitive response (HR), a plant-specific
form of programmed cell death at the infection sites, in many cases
followed by systemic acquired resistance (SAR) in the hosts. The
dynamic and continuous co-evolution between the two opponents
stimulates on side of the pathogen the formation of novel effectors
to suppress the ETI response ,. On the side of the host, new
plant resistance (R) proteins are developed to recognise the
effectors to reconsolidate ETI ,.
Typically, perception of PAMPs rapidly activates early defence
responses including depolarisation of the plasma membrane ,
opening of ion channels ,, activation of a mitogen-activated
protein kinase (MAPK) cascades , generation of reactive
oxygen species (ROS), reinforcement of the cell wall, transcription
of defence genes, and phytoalexin accumulation ,. The
best characterised PAMP is the peptide flg22, corresponding to the
highly conserved N-terminal part of eubacterial flagellin,
activating defence responses in most plant species . Recognition of
flg22 by the leucine-rich repeat (LRR) receptor kinase FLS2
, leads to increased intracellular Ca2+ concentration,
oxidative burst, activation of MAPKs, transcription of
defencerelated genes through the WRKY transcription factors WRKY22/
29 and WRKY25/33, and ethylene biosynthesis ,,.
While PAMPs are commonly considered to be essential for
general microbial fitness and survival, effectors, secreted or
injected into the plant cell , specifically contribute to pathogen
virulence by affecting specific targets of the host, such as receptor
kinases , ubiquitination , vesicle trafficking , cell-wall
reinforcement , secretion of toxic plant proteins , and
hypersensitive reaction (HR) . Harpin proteins, first described
in Erwinia amylovora, the causal agent for the fire-blight disease of
apple, pear and other members of the Rosaceae , belong to a
group of effector proteins exported by a bacterial T3SS and have
been intensively studied for their ability to initiate hypersensitive
cell death and to induce systemic acquired resistance .
When applied to non-host plants, Harpin triggers
immunityassociated responses, such as ROS production ,,
accumulation of defence-related transcripts and cell death ,,,
Thus, Harpin proteins can mimick certain aspects of ETI.
The conceptual discrimination between PTI and ETI has been
challenged by recent studies identifying transitions between
PAMPs and effectors . The activation of immune responses
in PTI and ETI through PAMPs and effectors, and through
different PAMPs appears to share common events . Thus, the
dichotomy might be not of qualitative, but of quantitative nature,
and it might depend merely on magnitude and duration of the
interactions among the components. Moreover, upon recognition
of different pathogenic virulence factors, plants can trigger
different responses, and, in addition, different disease-tolerant
plants respond to specific pathogens differently ,.
Therefore, the complex signalling resulting in PTI or ETI, and the
similarities and differences among PTI responses triggered by
different PAMPs warrants further investigation.
Grapevine, a widespread and important agricultural fruit crop,
is affected by several diseases such as Downy and Powdery Mildew
,. During the long co-evolution with these pathogens,
North American Vitis species have developed sophisticated and
robust defence mechanism probably based on recognition of
pathogen effectors by the products of host resistance (R) genes. In
contrast, European grapevines have evolved without contact to
these pathogens, and therefore represent naive hosts that lack
effective mechanisms to limit pathogenic infection. As an
important strategy to improve the resistance against pathogens
without the need for expensive and ecologically problematic
pesticides, the innate immunity of grapevine can be activated by
pathogen-derived elicitors ,,. The success of this
strategy depends on our understanding of the molecular and
cellular signal mechanism underlying grapevine-pathogen
We therefore conducted a comparative analysis of early defence
signalling triggered by two evolutionarily different bacterial
PAMPs, flg22 versus Harpin, in suspension cells derived from
the pathogen resistant North American species Vitis rupestris and
the susceptible grapevine Vitis vinifera cultivar Pinot Noir. We
investigated the dependence of apoplastic alkalinisation on
calcium influx, MAPK cascades, and cytoskeleton, oxidative
burst, expression of defence genes, biosynthesis of stilbenes, and
cytoskeletal reorganisation, and arrive at a model, where early
defence responses triggered by flg22 and Harpin partially overlap,
but differ in perception and oxidative burst, which are integrated
into a qualitatively different final output with respect to stilbene
patterns and cell death. Whereas flg22 triggers a basal PTI in both
cell lines, Harpin, although commonly accepted as a class of
PAMPs due to its widespread distribution among the bacterial
pathogens, triggers an ETI-like defence.
Flg22-induced extracellular alkalinisation differs in two
One of the earliest responses detected is a modification of
plasma membrane permeability, in particular, Ca2+, H+ and K+,
and anion fluxes that can be conveniently followed as changes of
extracellular pH ,. We therefore followed apoplastic
alkalinisation after treatment with the bacterial PAMP flg22 to
compare it with our previous data on the bacterial secreted protein
Extracellular pH increased rapidly from about 30 s after
addition of flg22, culminated in about 20 min, and subsequently
decreased slowly in V. rupestris (Figure 1A). In V. vinifera cv. Pinot
Noir, the increase of pH initiated later (from 5 min), and the
amplitude of the peak at 20 min was lower by a factor of 2
(Figure 1B). The magnitude of the peak was dependent on the
concentration of flg22 (Figures 1A, B). We therefore compared the
difference between the two cell lines on a quantitative level, and
recorded numerous time-courses over different concentrations of
flg22. The dependency of maximal DpH on the respective
concentration of flg22 (Figures 1C, D) could be fitted using a
Michaelis-Menten equation (R2 = 0.960 for V. rupestris; and
R2 = 0.962 for V. vinifera cv. Pinot Noir), where effective
concentrations (EC50, inducing 50% of the maximal response)
could be calculated to be 4.825 nM in V. rupestris and 876.86 nM
in V. vinifera cv. Pinot Noir respectively. This means that the
sensitivity of V. rupestris is roughly 200 times higher, compared with
V. vinifera cv. Pinot Noir. Corresponding to EC50, DpHmax was
approximately 1.251 in V. rupestris and 0.497 in V. vinifera cv. Pinot
Noir. To establish a situation, where the pH response as readout
for signal input was comparable between V. rupestris and V. vinifera
cv. Pinot Noir, a concentration of 1 mM flg22 was used in the
In our previous work, we had quantified the response to Harpin
, and observed a similar difference in the sensitivity of the two
cell lines. However, compared to elicitation with Harpin, the pH
response triggered by flg22 was faster (maximum reached at about
20 min) than for Harpin (maximum reached at 30 min), indicating
a more rapid signal transfer between binding of the elicitor and
proton flux for flg22 as compared to Harpin.
Flg22-induced extracellular alkalinisation is more
sensitive to Gd ions
Extracellular alkalinisation records the activity of a calcium
influx channel essential for the activation of early defence  and
should be blocked by GdCl3, an inhibitor of mechanosensitive
calcium channels . We therefore measured extracellular
alkalinisation evoked by flg22 and Harpin in presence of GdCl3
in V. rupestris (Figures 1E, F) and V. vinifera cv. Pinot Noir
(Figures 1G, H). In both cell lines, alkalinisation in response to
flg22 was significantly inhibited by addition of 20 mM GdCl3 as
compared to the solvent control (Figures 1E, G). In contrast to
flg22, Harpin-triggered alkalinisation was not significantly affected
by 20 mM GdCl3 (Figures 1F, H), indicating that Harpin-triggered
pH change is not sensitive to Ca2+, and even a concentration as
high as 1 mM GdCl3 inhibited Harpin-elicited alkalinisation only
to a low extent. This finding suggests that Ca2+ influx through the
plasma membrane is required for the alkalinisation induced by
flg22, but is only indirectly linked to Harpin-triggered
Negative feedback of MAPK signalling on alkalinisation
The mitogen-activated protein kinase (MAPK) cascades
represent one of the major signalling systems of eukaryotic cells. Several
MAPK cascades were shown to be associated with the induction of
plant defence responses ,. To understand, why
alkalinisation is transient, we probed for a possible feedback of MAPK
signalling using PD98059, a specific inhibitor of the MAPK
cascades. For flg22-triggered alkalinisation, we observed a
conspicuous pH-response which decreased gradually after a peak
at 20 min, and the inhibitor significantly reduced the slope of
decrease resulting in an almost stable alkalinisation in V. rupestris
(Figure 2A). For Harpin-triggered alkalinisation that was already
constitutive in V. rupestris, it was not possible to increase pH even
Figure 1. Apoplastic alkalinisation evoked by flg22 and Harpin in the two grapevine cell lines. A, B Dose response of extracellular
alkalinisation to flg22 over time in Vitis rupestris (A) and Vitis vinifera cv. Pinot Noir (B). C, D Analysis of the maximum change of extracellular pH in
response to the increasing concentration of flg22. Data were fitted using a Michaelis-Menten equation [f (x) = DpHmax6x/(EC50+x)], where
DpHmax = 1.251 (V. rupestris) or 0.497 (V. vinifera cv. Pinot Noir), and EC50 = approximately 4.825 nM (V. rupestris) or 876.86 nM (V. vinifera cv. Pinot
Noir), respectively. EH Role of Gd-sensitive calcium channels for apoplastic alkalinisation induced by 1 mM flg22 (E, G) or 9 mg.ml21 Harpin (F, H) in
combination with the solvent DMSO (open circles) or with 20 mM of GdCl3 (closed circles) either in V. rupestris (E, F) or V. vinifera cv. Pinot Noir (G, H),
respectively. Representative timelines from five independent series are shown.
further by treatment with PD98059 (Figure 2B). However, flg22
slightly enhanced the pH-response in V. vinifera cv. Pinot Noir,
although the amplitude remained very low (Figure 2C). In
contrast, here the transient Harpin-triggered alkalinisation could
be rendered constitutive by 100 mM of PD98059 in V. vinifera cv.
Pinot Noir (Figure 2D). This means that, in this case, inhibition of
the MAPK cascades in the less responsive V. vinifera cv. Pinot
Noir line almost phenocopied the constitutive pH response in the
responsive V. rupestris. These findings indicate that the transient
nature of elicitor-triggered alkalinisation is caused by a negative
feedback from (downstream) MAPK signalling. This negative
feedback is more pronounced in Harpin-triggered signalling, and
it is more relevant in V. vinifera cv. Pinot Noir.
The cytoskeleton modulates alkalinisation
In addition to its role in the machinery driving cell division and
expansion, the cytoskeleton acts as a sensor for environmental
stimuli through a mechanosensitive activity at the plasma
membrane . To investigate, whether the organisation of the
cytoskeleton modulates the alkalinisation induced by flg22 or
Harpin, Oryzalin, an inhibitor of microtubule polymerisation
specific for plants, and Latrunculin B, impeding the assembly of
actin filaments, were used in this study. In both cell lines,
application of Oryzalin significantly (up to ,0.4 pH units in V.
rupestris) decreased the amplitude of alkalinisation for both
flg22(Figures 2E, G) and Harpin-elicitation (Figures 2F, H). In a control
experiment, the same concentration of Oryzalin caused a small
alkalinisation of ,0.1 in V. rupestris, and of ,0.05 in V. vinifera cv.
Pinot Noir (Figure S1 in supporting information). In contrast,
Latrunculin B caused a small, but significant elevation (about ,0.1
pH units) of alkalinisation in V. rupestris for both elicitors
(Figures 2E, F). In V. vinifera cv. Pinot Noir, this elevation was
not observed (Figures 2G, H). In case of Harpin, Latrunculin B
even caused a significant suppression of alkalinisation (Figure 2H).
Here, a control with the same concentration of Latrunculin B in
the absence of elicitor caused a slight alkalinisation that remained
insignificant (Figure S1 in supporting information). These results
demonstrate that microtubules act as positive modulators of
alkalinisation, whereas actin constrains alkalinisation in the
responsive V. rupestris line (but not in the less responsive V. vinifera
cv. Pinot Noir).
Oxidative burst is induced differently by flg22 and Harpin
The rapid generation of reactive oxygen species (ROS), termed
oxidative burst, is induced early during pathogen invasion or
elicitor treatment . To test, to what extent oxidative burst is
triggered by flg22 or Harpin, we used the fluorescent dye
dihydrorhodamine 123 (DHR 123) to follow ROS production
after incubation with either flg22 (1 mM) or Harpin (9 mg.ml21) as
compared to a water control. No significant changes were
observed for the solvent control in the two cell lines (Figure 3).
However, fluorescence was pronouncedly elevated after both flg22
and Harpin treatments in both cell lines. In V. rupestris (Figure 3A),
the signal increased immediately to a transient peak of about 3.0
fold at 1015 min after Harpin elicitation and then dropped back
rapidly, whereas flg22-induced ROS production initiated with a
delay of about 15 min with a peak of about 2.5 fold signal at 25
30 min and a subsequent decrease (Figure 3A). In contrast to V.
rupestris, in V. vinifera cv. Pinot Noir oxidative burst although
occurring with similar time courses as for V. rupestris was much
weaker with only slight inductions of 1.4 fold for Harpin and 1.2
fold for flg22 application, respectively (Figure 3B). In summary, we
observed that both flg22 and Harpin induced only a transient
oxidative burst, indicating that these ROS act as signal rather than
as components of the machinery executing hypersensitive cell
death . This early oxidative burst happens significantly earlier
in case of Harpin elicitation as compared to flg22.
Flg22 and Harpin induce expression of defence genes in
a similar way
The synthesis of phytoalexins and other antimicrobial
compounds represents a central element of plant defence. We therefore
followed the transcript levels of key players in grapevine defence
by semi-quantitative RT-PCR using an elongation factor 1a gene
(EF 1a) as internal standard. The transcription activation of the
flavonoid pathway was monitored by probing for phenylalanine
ammonium lyase (PAL), chalcone synthase (CHS), and chalcone
isomerase (CHI), the stilbene pathway by stilbene synthase (StSy)
and resveratrol synthase (RS), and the activation of
pathogenesisrelated proteins by probing for PR5, and PR10, and the
polygalacturonase-inhibiting protein (PGIP) ,,.
Compared to the results obtained using the Harpin elicitor published
previously by Qiao et al. , the gene expression profile induced
by flg22 was similar. In control cells, no significant transcript
accumulation of genes was detected during the incubation period.
Similar to elicitation by Harpin , the flg22-response was faster
and stronger in V. rupestris than in V. vinifera cv. Pinot Noir
(Figures 4A, B). In V. rupestris, the transcripts of StSy and RS,
driving stilbene biosynthesis, accumulated from 30 min, peaked at
1 h, and had decreased at 3 h, whereas in V. vinifera cv. Pinot
Noir at 30 min any accumulation was hardly detectable.
Similarly, flg22 induced a higher expression of PAL, and PGIP,
whereas there was not significant up-regulation for CHS and CHI.
Expression of PR10 and PR5, important factors for
pathogensusceptibility and hypersensitive cell death , were induced
strongly and rapidly in V. rupestris, but only weakly, if at all, in V.
vinifera cv. Pinot Noir. The transcript patterns observed after
treatment with flg22 were very similar to those triggered by
Harpin . It was shown that both flg22 and Harpin induced
defence gene expression in a similar way. Thus, flg22 and Harpin,
although activating different levels of immunity, seem to activate
comparable patterns of defence-related genes.
MAPK activity is necessary for flg22, but not for
Harpininduced StSy transcription
The MAPK cascades have also been implied in the activation of
defence gene expression in several studies . To test, whether
this signalling pathway, in addition to its feedback regulation of
alkalinisation (Figures 2AD), is involved in the activation of
defence genes, we investigated the transcription of StSy as
representative example and used the MAPK cascades inhibitor
PD98059. Analysis of semi-quantitative RT-PCR showed that
PD98059 in both cell lines partially inhibited StSy expression
triggered by either flg22 or Harpin (Figures 4C, D). However, the
inhibition was much stronger for flg22-induced, much weaker for
Harpin-induced StSy transcription. A comparison of flg22-induced
transcript abundance between the cell lines showed that the
inhibition was more pronounced in V. rupestris over that observed
in V. vinifera cv. Pinot Noir. Thus, MAPK signalling is necessary
for flg22- triggered transcription of StSy, but not so essential for
Harpin-triggered transcription, especially in the disease-susceptible
V. vinifera cv. Pinot Noir.
Stilbenes accumulate differently for flg22- versus
To investigate the effect of flg22 on the enzymatic StSy activity,
the products, the stilbenes trans-resveratrol, its glucoside
transpiceid, and its oxidised dimer d-viniferin were quantified in both
cell lines by HPLC after 10 h incubation with 1 mM flg22 or with
9 mg.ml21 Harpin, respectively, as described in our previous study
. As shown in Figure 5A, flg22 failed to induce any detectable
trans-resveratrol in any of the cell lines (Figure 5A, up). The
biologically inactive glucoside of resveratrol, trans-piceid
(Figure 5A, middle), was detectable in low abundance
(3.5 mg.g21) in V. vinifera cv. Pinot Noir, but was virtually absent
in V. rupestris (1.17 mg.g21). The biologically active oxidative dimer
d-viniferin accumulated to modest 20.76 mg.g21 in V. rupestris,
while there was almost no d-viniferin detectable in V. vinifera cv.
Pinot Noir (Figure 5A, low). This weak stilbene accumulation in
response to flg22 contrasted with the strong accumulation
triggered by Harpin (Figure 5B): Here, V. rupestris produced high
Figure 4. Expression of defence-related genes induced by flg22. A, B Representative gels showing transcript abundance followed by
semiquantitative RT-PCR after elicitation with 1 mM flg22 (A), and quantification relative to elongation factor 1a (B) as reference. The data represent mean
values from three independent experimental series; error bars show standard errors. Genes of interest encode proteins including PAL, phenylalanine
ammonium lyase; CHS, chalcone synthase; StSy, stilbene synthase; RS, resveratrol synthase; and CHI, chalcone isomerase; pathogenesis-related
proteins: PR10 ad PR5, and PGIP: polygalacturonase-inhibiting protein. C, D Influence of MAPK signalling on the abundance of StSy transcripts. Cells
were challenged by 1 mM flg22, by 9 mg.ml21 Harpin (both in the solvent DMSO) alone or in combination with the MAPK cascades inhibitor PD98059
(PD). A representative agarose gel is shown in C, the quantification relative to elongation factor 1a from four independent experimental series in D,
error bars represent standard errors.
levels of trans-resveratrol and d-viniferin
(56.06 mg.g21), but again low levels of trans-piceid (1.06 mg.g21),
In contrast, V. vinifera cv. Pinot Noir accumulated small amounts
of trans-resveratrol (2.99 mg.g21) and d-viniferin (0.05 mg.g21), but
significant amounts of trans-piceid (18.5 mg.g21). Thus, flg22 and
Harpin differ qualitatively in their ability to induce stilbenic
compounds, although both can activate StSy transcripts to a
comparable extent (Figure 4).
Flg22 can trigger cytoskeletal responses similar to Harpin
Since cytoskeletal reorganisation is associated with the resistance
of plant cells to penetration by pathogens , and since
cytoskeletal drugs can modulate apoplastic alkalinisation
(Figures 2EH) and can induce defence genes in the absence of
elicitor , we investigated the cytoskeletal organisation after
treatment with flg22. The response to Harpin had been analysed
Figure 5. Stilbene accumulation in response to flg22 and Harpin. Cells of V. rupestris and V. vinifera cv. Pinot Noir were exposed to either
1 mM flg22 or 9 mg.ml21 Harpin for 0 (white bars) or 10 h (striped bars). Contents of trans-resveratrol, trans-piceid and d-viniferin were determined by
HPLC and quantified relative to their corresponding calibration curves based on reference standards, respectively. Mean values and standard errors
from at least three independent experimental series are shown.
We observed disintegration of microtubules in V. rupestris 1 h
after treatment with 1 mM flg22, whereas microtubules were only
slightly affected in V. vinifera cv. Pinot Noir (Figure 6A),
resembling the situation observed for Harpin . Actin filaments
that, in control cells, formed fine strands in the periphery of the
cells, became strongly bundled and had contracted towards the
nucleus 3 h after incubation with 1 mM flg22 (Figure 6B) again
similar to the pattern observed after treatment with Harpin
Since the degree of flg22-induced microtubule disintegration
varied between the two Vitis cell lines, we wanted to understand
whether this difference in the microtubular response was related to
a difference in microtulular dynamics reported by the abundance
of tyrosinylated a-tubulin probed by the monoclonal antibodies
ATT. When soluble proteins from control and flg22-triggered cells
were compared, the signal labeled by ATT antibody was strongly
increased 24 h after elicitation with flg22 (Figures 6C, D). This
response was especially pronounced in V. vinifera cv. Pinot Noir
indicating that here microtubules acquired a higher turnover after
treatment with flg22.
Harpin, but not flg22, can induce cell death
Activation of defence responses often results in a hypersensitive
response (HR) occurring at infection sites which is characteristic in
ETI, but rare in PTI ,. Therefore, we followed cell viability
after challenge by flg22 or Harpin using Evans Blue staining. We
observed that Harpin induced cell death in both V. rupestris and V.
vinifera cv. Pinot Noir. In V. rupestris, cell death increased strongly
from 48 h, and reached more than 60% at 72 h after elicitation
(Figure 7A), whereas in V. vinifera cv. Pinot Noir mortality was
much lower with only some 23% at 72 h (Figure 7B). In contrast
to Harpin, 1 mM of flg22 did not induce significant mortality in
any of the two lines (Figure 7B), although this concentration
activated the full repertory of defence responses.
Plants have developed sophisticated defence systems typically
grouped around two levels of immunity, PTI and ETI. A limited
set of signal components is organised and integrated to efficiently
overcome host or non-host pathogens. However, not all
pathogenic defence activators conform to the common signalling
mechanism. It therefore is interesting to investigate the immunities
triggered by different pathogenic inducers. Accumulating evidence
suggests that inducible immunities by PAMPs or effectors often
share common signal components. However, at what point these
different immunities converge or diverge, is still far from
understood. The current models of defence signalling have mainly
been driven by hallmark discoveries from the model plant
Arabidopsis thaliana. It is to be expected that specific aspects from
other models will enrich and modify our knowledge of plant
immunity. We have therefore employed cell cultures from the
disease-resistant grapevine V. rupestris and the susceptible grapevine
V. vinifera cv. Pinot Noir to study signal events triggered by the
bacterial PAMPs flg22 or Harpin that differ with respect to their
ability to trigger cell death. We compared the dependence of
apoplastic alkalinisation (as readout for early signalling) on calcium
channels, cytoskeleton, and MAPK cascades, and we observed
certain differences depending on the nature of the trigger and the
cell line. From our observations and previous publications on this
system ,, a (simplified) model on defence signalling can be
deduced (Figure 8).
Figure 6. Response of the cytoskeleton to flg22. A Disintegration of microtubules visualised by immunofluorescence1 h after addition of 1 mM
flg22 or water as negative control. Size bar 20 mm. B Reorganisation of actin filaments visualised by FITC-phalloidin upon flg22 treatment as
compared to the water control. Representative geometrical projections from Apotome Z-stacks collected from control (left) or after 3 h
(flg22induced, right) of treatment with 1 mM flg22 are shown. Size bar 20 mm. C Abundance of tyrosinylated a-tubulin in total extracts 24 h after additioin
of 1 mM flg22 visualised by Western blotting probing with specific monoclonal antibodies. The same amount of total protein was loaded in each lane,
verified by staining of a replicate by Coomassie Brilliant Blue. D Relative abundance of tyrosinylated a-tubulin quantified for the flg22 treatment
(flg22, grey bars) as compared to control (con, white bars).
Elicitor perception and apoplastic alkalinisation
Changes in ion fluxes across the plasma membrane, especially
increased influx of Ca2+ and H+, and efflux of K+, have been
proposed to be part of the signal transduction chain . These can
be conveniently measured using apoplastic alkalinisation as
readout , which allows deriving quantitative data on
perception of the respective elicitor. The fact that the apparent
affinity of the putative perception system for flg22 is orders of
magnitudes higher as compared for Harpin, and the finding that
the alkalinisation in response to Harpin is delayed by 510 min as
compared to flg22 (Figures 1A, B) leads to a model, where the link
between flg22 and alkalinisation is more direct, whereas the link
between Harpin and alkalinisation is indirect. In our previous
work , we have shown that the induction of gene expression by
Harpin requires apoplastic ROS suggesting that the effect of
Harpin on alkalinisation is transduced via an apoplastic oxidative
burst, for instance through a grapevine homologue of the
NADPH-dependent oxidoreductase Rboh (Figure 8). In Arabidopsis
thaliana, flg22 is directly recognised by the plasma membrane
receptor-like kinase FLS2 that acts together with another
receptorlike kinase, BRI-1-associated receptor kinase 1 (BAK1)  to
activate downstream signalling ,,. A putative grapevine
homologue of AtFLS2 has been identified . So far, there is no
direct evidence for a specific host receptor binding Harpin.
However, oligomerisation and formation of ionophore-like
membrane pores was shown for Hrp7 to depend on a 24-amino-acid
motif in the C-terminus, indicating a certain specificity of
Apoplastic alkalinisation is thought to record the activity of a
(mechanosensitive) calcium influx-channel . In fact, we
observed that alkalinisation was inhibited by GdCl3 (Figures 1E
H), but flg22-triggered alkalinisation was much more sensitive as
compared to the Harpin-triggered response. This indicates that
the flg22-receptor interacts more directly with the calcium influx
channels, whereas the ion fluxes triggered by Harpin must involve
pathways that do not utilise Gd-sensitive calcium channels. In fact,
Harpin has been shown to cause membrane pores that are
permeable for cations such as calcium and protons . The
signalling target for this calcium influx remains to be elucidated, at
least for Harpin it could be shown that it is dispensable for gene
activation in tobacco .
Cytoskeleton and early signalling
We tested the role of actin for the apoplastic alkalinisation using
the specific inhibitor Latrunculin B. We observed a slight, but
significant stimulation of both, flg22- and Harpin-triggered
alkalinisation in the responsive V. rupestris line (Figures 2E, F)
indicating that actin negatively modulates membrane
permeability. This finding is consistent with previous findings that actin
stabilises plant membranes, probably by releasing membrane
tensions through mobilisation of membrane material . In
contrast to Latrunculin B, Oryzalin produced a significant
reduction of elicitor-triggered alkalinisation (Figures 2EH)
indicating that microtubules are required to activate defence
related ion fluxes in response to the elicitors. Oryzalin can activate
alkalinisation in the absence of elicitors (followed by a partial
activation of defence-related transcription), which can be
explained by gating of mechanosensitive calcium channels through
microtubules . However, the reduction of flg22- or
Harpintriggered alkalinisation by Oryzalin cannot be explained by
removal of the microtubular gating function, but suggests that
microtubules somehow help to convey the information of elicitor
binding to the channel. Since Oryzalin was added simultaneously
with the elicitors and therefore acted only over a short time span,
these sensory microtubules must be endowed with high dynamics.
A similar transducer function of highly dynamic microtubules has
been also observed in other sensory processes such as cold or
gravity sensing . Similar to Harpin elicitation, flg22 caused
bundling of actin filaments and a fragmentation of microtubules.
This microtubular response was hardly detectable in V. vinifera cv.
Pinot Noir but pronounced in V. rupestris, and accompanied by an
increase of tyrosinylated a-tubulin indicative of a stimulated
microtubular turnover (Figure 6). The mechanism for this
stimulated microtubular turnover is not known, but it should be
mentioned in this context that the MAPK cascade regulates,
through its the NACK-PQR branch, the activity of MAP65, an
important regulator of microtubular dynamics . An alternative
mechanism might involve the microtubule-stabilising protein
spiral1 that is recruited for proteasome-mediated degradation in
response to osmotic stress .
Many stress signals that induce changes in extracellular and/or
intracellular pH also activate mitogen-activated protein kinase
(MAPK) cascades ,. Typically, MAPK cascades are
composed of three layers: a MAPKKK (MAPK kinase kinase), a
MAPKK (MAPK kinase), and a MAPK  that can convey
signals from upstream kinases to downstream targets including
activation of transcription factors, differentiation, cell division, and
environmental stresses . In fact, MAPK activity is activated by
Harpin in cells of Arabidopsis thaliana and tobacco ,, and
flg22 treatment triggers a rapid phosphorylation of proteins and a
transient activation of the MAPK cascade including MPK3/
MPK4/MPK6 ,,. To avoid constitutive
overstimulation of defence signalling, the primary signals have to be switched
off, once the signal has been transferred to intracellular acceptors.
For instance, the flg22 receptor FLS2 is internalised following
binding of the ligand . Alternatively, the activity of the
triggering ion channel could be downregulated by negative
feedback from downstream signals. In fact, we observe that
PD98059, an inhibitor of MAPK signalling can render a transient
alkalinisation (in V. vinifera cv. Pinot Noir) into a constitutive
signal (Figure 2C) suggesting that MAPK signalling produces such
a negative feedback avoiding overstimulation of defence. In
addition to this feedback, MAPK signalling is required for the
activation of StSy transcription, a central player of phytoalexin
synthesis (Figures 4C, D), but seems to be more essential for the
transduction of flg22, whereas the Harpin signal seems to be
transduced in parts independently of MAPK signalling. This
contrasts with findings in tobacco, where Harpin triggered the
PRgene HIN1 through calcium-independent MAPK signalling .
Thus, the exact link between calcium influx, activation of MAPK
signalling and gene activation warrants further investigation.
Activation of defence genes
We tested a panel of defence-related genes that are activated by
Harpin in both grapevine cell lines  for their response to flg22
elicitation (Figures 4A, B). Although we found differences between
the cell lines (a weaker response of V. vinifera cv. Pinot Noir), the
pattern was fairly similar to that obtained for Harpin elicitation. In
Arabidopsis thaliana, it was observed that the PAMP flg22 and the
effector Avr9 activated a substantially overlapping set of genes
Figure 8. A simplified model for defence triggered by flg22 and Harpin in grapevine cells. Details are explained in the discussion. Flg,
flg22; Hrp, bacterial protein Harpin; flgr, flg22 receptor (grapevine homologue of AtFLS2); msc, mechanosensitive ion channel; MTs, microtubules;
mAFs, membrane-associated actin filaments; Rboh, grapevine homologue of NADPH dependent oxidase responsible for apoplastic oxidative burst
(ROSex) that can permeate the plasma membrane (ROSint); MAPK, MAPK-signalling pathway; StSy, stilbene synthase gene; iAFs, intracellular actin
filaments; Res, trans-resveratrol; d-Vin, d-viniferin; Pic, trans-piceid.
Oxidative burst has a dual function in defence, either as early
stress signal or as part of the downstream machinery that attacks
invading pathogens . The rapid and transient production of
ROS production in response to elicitors is dependent on a
NADPH oxidase . In our grapevine system, we observed a
distinct difference in timing of oxidative burst induced by flg22
and Harpin (Figure 3). Whereas Harpin triggered an early
oxidative burst (preceding alkalinisation), the oxidative burst
triggered by flg22 was later (and follows alkalinisation and even
activation of defence-related transcripts). This means that the
oxidative burst in flg22 cannot act as an early signal, but rather
represents a downstream response. In contrast, Harpin signalling
seems to employ oxidative burst. In our previous work, we have
shown for the grapevine cell system that apoplastic ROS are
necessary for the induction of StSy by Harpin .
The products of stilbene synthase/resveratrol synthase (StSy/
RS), the stilbene resveratrol, are a class of phytoalexin produced
by plants as part of the defence response. In grapevine, resveratrol
efficiently blocks pathogens such as Downy and Powdery Mildew
,. In addition to resveratrol, its metabolic compounds are
endowed with high antimicrobial activity and accumulate in
grapevine as a result of infection or stress ,,,.
Among those metabolic compounds, oxidised d-viniferin is even
more toxic than resveratrol itself and capable of inhibiting
zoospore mobility of Plasmopara viticola, whereas the glucoside
piceid shows no or little toxicity and no antimicrobial activity
,. Although in the two cell lines both, flg22 and Harpin
induced the StSy transcript to a similar degree (Figures 4A, B), the
educts of stilbene synthesis, resveratrol, and its oxidised dimer
dviniferin accumulated to significant amounts only in response to
Harpin elicitation (Figure 5) in V. rupestris, whereas flg22 only
induced marginal levels of d-viniferin. The inactive glucoside
transpiceid was formed instead in V. vinifera cv. Pinot Noir, again, only
Harpin can induce significant levels, whereas flg22 was almost
inactive. The reason for this difference between the two elicitors
remains unknown. The substrate of StSy/RS is also used by
chalcone synthase (CHS), a key enzyme of flavonoid synthesis.
StSy/RS has originated from CHS via gene duplication and
mutation . Since CHS is also induced by flg22, it is
conceivable that it diverts the substrate from StSy however,
CHS is also induced by Harpin to a similar degree . This
indicates that the balance between StSy and CHS activity might
be regulated and partitioned on the posttranslational level.
Resveratrol acts as important amplifier of oxidative burst  in
these grapevine cell lines. This means that Harpin is expected to
produce a resveratrol-induced second wave of oxidative burst that
is absent in V. vinifera cv. Pinot Noir. Although resveratrol and
dviniferin are considered as the pivotal phytoalexins, it should be
kept in mind that additional stilbenes that have not been addressed
in the present study could be relevant as well. We therefore have
launched a metabolomics approach to identify additional potential
players in the stilbenoid pathway.
Defence responses, in many cases, are accompanied by HR-type
programmed cell death, especially in ETI. V. rupestris originates
from North America, and has evolved sympatrically with several of
the major grapevine diseases. Its disease resistance has been
intensively studied in the context of resistance breeding and linked
with a pronounced capacity for hypersensitive cell death 
correlated with the Rpv3 locus, probably encoding a receptor for
oomycete effectors . In fact, elicitation by Harpin can trigger
pronounced cell death in V. rupestris, and to a weaker extent, in V.
vinifera cv. Pinot Noir, whereas flg22 was completely ineffective
with respect to cell death (Figure 7). Preliminary assays using the
TdT-mediated dUTP nick end labeling (TUNEL) assay indicate
that the Harpin-triggered response classifies for a HR-type PCD
event. However, recent studies emphasise that other forms of cell
death, such as autophagy, need to be taken into consideration as
When the cellular responses triggered by flg22 and Harpin in
this and our previous studies are compared, apoplastic
alkalinisation, cytoskeletal responses, and calcium influx are triggered by
both flg22 and Harpin, although differing in amplitude between V.
rupestris and V. vinifera cv. Pinot Noir. However, there is evidence
for a stricter dependency of StSy transcriptional activation on
MAPK signalling in case of flg22 elicitation, whereas in case of
Harpin signalling, MAPK seems to be at least partially
dispensable, indicating a parallel signal pathway. The primary steps of
both pathways differ of course: flg22 triggered signalling involves
binding of the PAMP to a receptor protein (probably the
grapevine homologue of AtFLS2) with high affinity. Harpin
activates at much lower affinity and probably not through a
receptor protein . A second qualitative divergence of the
pathways becomes manifest in oxidative burst: Whereas Harpin
causes an early wave of ROS (preceding apoplastic alkalinisation)
that is later followed by a second wave of ROS triggered by the
stilbene resveratrol , flg22 triggers only a sluggish oxidative
burst (following apoplastic alkalinisation) and fails to induce
formation of resveratrol and thus the signal that produces the
second wave of ROS. Since the induction of StSy by Harpin seems
to be at least partially independent of MAPK signalling, a
straightforward hypothesis would assume that it is triggered by a
parallel ROS-dependent pathway (Figure 8). It has to be tested,
whether the same ROS-dependent pathway is also responsible for
the formation of resveratrol and thus for the second wave of
oxidative burst correlated with the induction of osmotin-type
pathogenesis-related protein 5 and cell death observed in
Harpinelicited V. rupestris .
The comparison between the cellular responses to flg22 and
Harpin provides new insight into defence signalling in
Vitispathogen interactions. Harpin induced cell death in V. rupestris,
and thus mimicks an ETI situation. On the other side, for several
reasons Harpin proteins rather qualify as PAMPs. Unlike
canonical effectors that have evolved from a specific interaction
between host and pathogen, Harpin proteins are widespread
among bacteria and contribute to virulence, probably by
contributing to the functionality of the pilus. A further argument
against a canonical effector function is the lack of any identified
host receptor for Harpins. In fact, the conceptual dichotomy
between PTI and ETI, has been softened recently, and certain
PAMPs seem to be, in fact, able to cause programmed cell death
We can conclude that some of the early defence responses are
shared by flg22 and Harpin signalling and only differ in amplitude,
not in quality. We could pinpoint essentially five aspects, where
flg22- and Harpin-triggered events differed qualitatively: (i)
perception by high-affinity binding to a specific receptor protein
in case of flg22, but by low-affinity membrane attachment in case
of Harpin, (ii) the early oxidative burst observed within 1015 min
after challenge with Harpin, was delayed by about 15 min in
response to flg22, (iii) the accumulation of StSy transcripts that
required functional MAPK signalling in response to flg22, was
mostly independent from MAPK signalling in response to Harpin,
(iv) although both elicitor activated StSy transcription to a similar
extent, the enzymatic products resveratrol and its oxidised
derivative d-viniferin accumulated only in response to Harpin,
not in response to flg22, (v) cell death was triggered by Harpin, but
not by flg22. These findings suggest that the early defence
responses triggered by the flg22 and Harpin, on the one hand,
share common signal elements, but differ in a chain of events
running in parallel with this shared signalling. This specific parallel
signalling is integrated differently at a later stage resulting in a
qualitatively different output of defence: basal immunity (bona fide
PTI) versus cell-death related immunity. To what extent the
Harpin-triggered cell-death related immunity overlaps with
canonical ETI will be the target of further investigations.
Materials and Methods
Cell culture and treatments
Suspension cell cultures of V. rupestris and V. vinifera cv. Pinot
Noir established from leaves  were maintained in liquid MS
medium containing 4.3 g.l21 Murashige and Skoog salts (Duchefa,
Haarlem, The Netherlands), 30 g.l21 sucrose, 200 m g.l21
KH2PO4, 100 mg.l21 inositol, 1 mg.l21 thiamine, and
0.2 mg.l21 2,4-dichlorophenoxy-acetic acid (2,4-D), pH 5.8. Cells
were sub-cultured weekly by transferring 10 ml stationary cells
into 30 ml fresh medium in 100 ml Erlenmeyer flasks and then
incubated at on an orbital shaker (KS250 basic, IKA150
Labortechnik, Staufen, Germany) at 150 rpm, 25uC, in the dark.
The bacterial peptide flg22 was synthesised by GenScript and
diluted in sterile H2O. A commercially available Harpin elicitor
[Messenger, EDEN Bioscience Corporation, Washington, USA,
active ingredient: 3% (w/w) Harpin protein] was prepared into
300 mg.ml21 stock solution.
Measurement of extracellular alkalinisation
Extracellular alkalinisation was measured by combining a pH
meter (Schott handylab, pH 12) with a pH electrode (Mettler
Toledo, LoT 403-M8-S7/120) as described in Qiao et al. .
Before addition of elicitors, cells were preequilibrated on an orbital
shaker for at least 1 h. The maxima of the pH response were
plotted against the flg22 concentration and fitted using a
Michaelis-Menten equation: f(x) = DpHmax6x/(EC50+x), with
DpHmax as Vmax, EC50 (the half-maximal pH response) as Km,
and the concentration of flg22 as [S].
To test for the impact of calcium influx on flg22- or
Harpindependent extracellular alkalinisation, an inhibitor of
mechanosensitive calcium channels, GdCl3, was used. Cells were
coincubated with 1 mM flg22, 9 mg.ml21 Harpin, either with or
without 20 mM of GdCl3, a concentration derived from our
previous work . To assess the effects of cytoskeletal drugs on
flg22- or Harpin-dependent extracellular alkalinisation,
microtubules were eliminated by 20 mM of Oryzalin, actin filaments by
2 mM of Latrunculin B added together with flg22 or Harpin as
compared to the same volume of DMSO as a solvent control. To
examine the influence of MAPK signalling, the inhibitor PD98059
targeted to the mitogen-activated protein kinase kinases
(MAPKKs)  was added to the cells in variable concentration
in DMSO with either flg22 or Harpin.
Quantification of oxidative burst
The production of ROS was determined by dihydrorhodamine
123 (DHR 123) as previously described . Aliquots of 200 ml
suspension were (at day 4 after sub-cultivation) diluted into 800 ml
of PBS, pre-equilibrated on a shaker for 1 h and then
supplemented with dihydrorhodamine 123 (DHR 123 in DMSO,
final concentration 10 mM). After incubation for 30 min, cells
were washed 3 times using pre-warmed PBS at 37uC and
resuspended in 1 ml PBS in combination with either with 1 mM
flg22, with 9 mg.ml21 Harpin, or with a corresponding
concentration of the solvent (water) as negative control. Changes of the
fluorescent signal were followed over time under an AxioImager
Z.1 microscope (Zeiss, Jena, Germany) equipped with an
ApoTome microscope slider for optical sectioning and a cooled
digital CCD camera (AxioCam MRm, Zeiss, Jena, Germany)
using the filter set 38 HE (excitation at 470 nm, beamsplitter at
495 nm, and emission at 525 nm), a 206 objective and a constant
exposure time of 100 ms. Production of ROS fluorescence was
quantified as mean pixel intensity of each image at indicated time
points in relation to the corresponding image recorded at 0 min
using an Image J software (http://rsbweb.nih.gov/ij/). Error bars
represent standard errors from three independent experiments.
The abundance of defence-related transcripts was measured by
semi-quantitative RT-PCR. Total RNA was extracted and cDNA
was synthesised after treatment with 1 mM flg22 for 0.5, 1, 3 h, or
with water as control as previously described ,. Transcripts
were amplified by RT-PCR using the primers listed in the Table
S1 probing for genes encoding phenylalanine ammonium lyase
(PAL), chalcone synthase (CHS), stilbene synthase (StSy),
resveratrol synthase (RS), chalcone isomerase (CHI),
pathogenesisrelated proteins (PR10 and PR5), and polygalacturonase-inhibiting
protein (PGIP) . Values for relative transcript abundance
were calculated using elongation factor 1a  as internal
standard. The data from the quantification represent the mean
from at least three independent experimental series.
To determine the influence of mitogen-activated protein kinase
(MAPK) cascades on the expression of the marker gene StSy, cells
were treated with either flg22 or Harpin alone or in combination
with the MAPK cascades inhibitor PD98059 (100 mM) for 1 h.
Experiments were performed in three biological replicates as
Quantification of stilbene biosynthesis
To test whether the transcript of stilbene synthase (StSy) was
accompanied by the final product generated by this enzyme, cells
were challenged either with 1 mM flg22, or 9 mg.ml21 Harpin for
10 h, a time point chosen according to time-course studies on
stilbene synthesis in the same system  as compared to a water
control. Cells were drained from culture medium by a vacuum of
800 pa (Vacuubrand CVC2, Brand, Germany), frozen in liquid
nitrogen, and then stored at 280uC until further analysis. Aliquots
of 3 g fresh weight of untreated control or treated cells were
homogenised with 20 ml of 80% (v/v) methanol in water by an
ultrasonic processor (UP100H, Hielscher, Germany) for 3 min.
The homogenate was incubated for 2 h in the dark at room
temperature in a rotatory shaker and filtered through filter paper
by vacuum with 500 pa. The filtrate was concentrated to a
residual volume of 5 ml in a glass tube at 40uC (Heating Bath
B490, BU CHI, Germany) at 280 rpm (Rotavapor R-205,
BU CHI, Germany), under a vacuum of 80 Pa (Vacuubrand
CVC2, Brand, Germany). Stilbenes were extracted from the
aqueous phase by adding 2 ml of 5% (w/v) NaHCO3, and three
aliquots of 5 ml ethyl acetate. The pooled ethyl acetate phase was
completely dried and the residue suspended in 2 ml of methanol
prior to injection into the HPLC.
Analysis of stilbenes was carried out on a high performance
liquid chromatograph, HPLC (Agilent, 1200 series, Waldbronn,
Germany) as described previously . Trans-resveratrol,
transpiceid, and d-viniferin were quantified using external standards on
the basis of retention time and UV-VIS spectra. The standards for
trans-resveratrol (Sigma-Aldrich, Deisenhofen, Germany),
transpiceid (Phytolab, Vestenbergsgreuth, Germany) and d-viniferin
(kind gift of Dr. Kassemeyer, State Institute of Viticulture,
Freiburg) were dissolved in methanol at a concentration of
100 mg.l21. Calibration curves for quantification of the samples
were determined using these standards and found to be linear
Visualisation of cytoskeleton
The response of the cytoskeleton to 1 mM flg22 was assessed
using fully expanded cells (day 10 after subcultivation) as
compared to a negative control with the corresponding
concentration of solvent (water). Microtubules were stained by indirect
immunofluorescene using a monoclonal antibody against
atubulin (DM1A; Sigma-Aldrich, Deisenhofen; Germany), and a
secondary anti-mouse IgG antibody conjugated to fluorescein
isothiocyanate (FITC; Sigma-Aldrich, Deisenhofen; Germany)
following the protocol published previously . Cells were
visualised using an AxioImager Z.1 microscope (Zeiss) equipped
with an ApoTome microscope slider through the filter sets 38 HE
(excitation at 470 nm, beamspliter at 495 nm, and emission at
Actin filaments were visualised with FITC-phalloidin as
described previously . Cells were fixed in 1.85% (w/v)
paraformaldhyde in 0.1 M PIPES, pH 7.0, supplemented with
5 mM MgCl2 and 10 mM EGTA) for 30 min at 25uC.
Subsequently, samples were stained with 0.66 mM
FITC-phalloidin (Sigma-Aldrich, Deisenhofen, Germany) for 30 min. Cells
were then washed three times for 5 min in PBS and observed
immediately by an ApoTome microscopy as described above.
Quantification of tyrosinated a-tubulin
10 ml of cells were treated at day 5 after subcultivation with
1 mM flg22 for 24 h and collected by centrifugation for 10 min,
3 000 rpm (Hettich Centrifuge Typ 1300, Tuttlingen, Germany).
Soluble proteins were extracted according to Qiao et al. .
Samples were dissolved in 20 ml of single-strength sample buffer,
vortexed, denatured for 15 min at 95uC, and analysed by
SDSPAGE to equalise the loading volume of samples.
Tyrosinated a-tubulin was detected by Western blotting using
the monoclonal antibody ATT (Sigma-Aldrich, Deisenhofen,
Germany). Signals were visualised by a goat secondary anti-mouse
IgA, conjugated with alkaline phosphatase (Sigma-Aldrich,
Deisenhofen, Germany) at a dilution of 1:2500 in TBST with
3% low fat milk powder. Alkaline phosphatase activity was
detected by 66 ml of NBT solution (75 mg.ml21
Nitrobluetetrazolium in 75% dimethylformamid) and 33 ml of BCIP
solution (50 mg.ml21 5-Bromo-4-chloro-3-
indoxylphosphate-pTuloidin in 100% dimethylformamid) in 5 ml staining buffer
(100 mM Tris-HCl, 100 mM NaCl, pH 9.7) with 1:10 (v/v) of
500 mM MgCl2. A parallel set of lanes loaded in exactly the same
manner was stained with Coomassie Brilliant Blue 250
(SigmaAldrich, Deisenhofen, Germany) as a loading control. Abundance
of tyrosinated a-tubulin after treatment with flg22 was quantified
relative to control with the mean value of each lane using Image J
Determination of cell viability
After cells of V. rupestris and V. vinifera cv. Pinot Noir cells had
been incubated with flg22 or Harpin for 24, 48 or 72 h,
respectively, aliquots (0.2 ml) of each sample were transferred
into custom-made staining chambers to remove the medium,
incubated in 2.5% (w/v) Evans Blue  for 3 min, and then
washed with water several times. The frequency of dead cells was
scored using a Fuchs-Rosenthal hematocytometer under
brightfield illumination. The mortality values were determined from
three independent experiments with at least 1500 cells scored for
each data point.
Technical support in the HPLC-analysis by Ernst Heene and advice in the
staining of actin filaments by Fei Qiao is gratefully acknowledged, as well as
the gift of d-viniferin standard by Dr. Hanns-Heinz Kassemeyer, State
Institute of Viticulture, Freiburg.
Conceived and designed the experiments: XC PN. Performed the
experiments: XC. Analyzed the data: XC PN. Contributed reagents/
materials/analysis tools: XC PN. Wrote the paper: XC PN.
1. Jones JD , Dangl JL ( 2006 ) The plant immune system . Nature 444 : 323 - 329 .
2. Cunnac S , Lindeberg M , Collmer A ( 2009 ) Pseudomonas syringae type III secretion system effectors: repertoires in search of functions . Curr Opin Microbiol 12 : 53 - 60 .
3. Tsuda K , Katagiri F ( 2010 ) Comparing signalling mechanisms engaged in pattern-triggered and effector-triggered immunity . Curr Opin Plant Biol 13 : 459 - 465 .
4. Block A , Li G , Fu ZQ , Alfano JR ( 2008 ) Phytopathogen type III effector weaponry and their plant targets . Curr Opin Plant Biol 11 : 396 - 403 .
5. Boller T , Felix G ( 2009 ) A renaissance of elicitors: perception of microbeassociated molecular patterns and danger signals by pattern-recognition receptors . Annu Rev Plant Biol 60 : 379 - 406 .
6. Boller T , He SY ( 2009 ) Innate immunity in plants: an arms race between pattern recognition receptors in plants and effectors in microbial pathogens . Science 324 : 742 - 744 .
7. Felix G , Duran JD , Volko S , Boller T ( 1999 ) Plants have a sensitive perception system for the most conserved domain of bacterial flagellin . Plant J 18 : 265 - 276 .
8. Lee J , Klusener B, Tsiamis G , Stevens C , Neyt C , et al. ( 2001 ). HrpZ(Psph) from the plant pathogen Pseudomonas syringae pv. phaseolicola binds to lipid bilayers and forms an ion-conducting pore in vitro . Proc Natl Acad Sci U S A 98 : 289 - 294 .
9. Jeworutzki E , Roeflsema MR , Anschutz U, Krol E , Elzenga JT , et al. ( 2010 ) Early signalling through the Arabidopsis pattern recognition receptor FLS2 and EFR involves Ca2+-associated opening of plasma membrane anion channels . Plant J 62 : 367 - 378 .
10. Gomez-Go mez L, Boller T ( 2000 ) FLS2: an LRR receptor-like kinase involved in the perception of the bacterial elicitor flagellin in Arabidopsis . Mol Cell 5 : 1003 - 1011 .
11. Nu rnberger T ( 1999 ) Signal perception in plant pathogen defense . Cell Mol Life Sci 55 : 167 - 182 .
12. Zipfel C , Kunze G , Chinchilla D , Caniard A , Jones JDG , et al. ( 2006 ) Perception of the bacterial PAMP EF-Tu by the receptor EFR restricts Agrobacterium-mediated transformation . Cell 125 : 749 - 760 .
13. Zipfel C , Robatzek S , Navarro L , Oakeley EJ , Jones JDG , et al. ( 2004 ) Bacterial disease resistance in Arabidopsis through Yagellin perception . Nature 428 : 764 - 767 .
14. Chinchilla D , Bauer Z , Regenass M , Boller T , Felix G ( 2006 ) The Arabidopsis receptor kinase FLS2 binds flg22 and determines the specificity of flagellin perception . Plant Cell 18 : 465 - 476 .
15. Chinchilla D , Zipfel C , Robatzek S , Kemmerling B , Nu rnberger T , et al. ( 2007 ) A flagellin-induced complex of the receptor FLS2 and BAK1 initiates plant defence . Nature 448 : 497 - 500 .
16. Nicaise V , Roux M , Zipfel C ( 2009 ) Recent advances in PAMP-triggered immunity against bacteria: pattern recognition receptors watch over and raise the alarm . Plant Physiology 150 : 1638 - 1647 .
17. He P , Shan L , Lin NC , Martin GB , Kemmerling B , et al. ( 2006 ) Specific bacterial suppressors of MAMP signalling upstream of MAPKKK in Arabidopsis innate immunity . Cell 125 : 563 - 575 .
18. Abramovitch RB , Janjusevic R , Stebbins CE , Martin GB ( 2006 ) Type III effector AvrPtoB requires intrinsic E3 ubiquitin ligase activity to suppress plant cell death and immunity . Proc Natl Acad Sci U S A 103 : 2851 - 2856 .
19. Kim JG , Taylor KW , Hotson A , Keegan M , Schmelz EA , et al. ( 2008 ) XopD SUMO protease affects host transcription, promotes pathogen growth, and delays symptom development in Xanthomonas-infected tomato leaves . Plant Cell 20 : 1915 - 1929 .
20. Hauck P , Thilmony R , He SY ( 2003 ) A Pseudomonas syringae type III effector suppresses cell wall-based extracellular defense in susceptible Arabidopsis plants . Proc Natl Acad Sci U S A 100 : 8577 - 8582 .
21. Chisholm ST , Coaker G , Day B , Staskawicz BJ ( 2006 ) Host-microbe interactions: shaping the evolution of the plant immune response . Cell 124 : 803 - 814 .
22. Nomura K , Melotto M , He SY ( 2005 ) Suppression of host defense in compatible plant-Pseudomonas syringae interactions . Curr Opin Plant Biol 8 : 361 - 368 .
23. Wei ZM , Laby RJ , Zumoff CH , Bauer DW , He SY , et al. ( 1992 ) Harpin, elicitor of the hypersensitive response produced by the plant pathogen Erwinia amylovora . Science 257 : 85 - 88 .
24. Baker CJ , Orlandi EW , Mock NM ( 1993 ) Harpin, an elicitor of the hypersensitive response in tobacco caused by Erwinia amylovora, elicits active oxygen production in suspension cells . Plant Physiology 102 : 134 - 344 .
25. Bauer DW , Wei ZM , Beer SV , Collmer A ( 1995 ) Erwinia chrysanthemi harpinEch: an elicitor of the hypersensitive response that contributes to soft-rot pathogenesis . Mol Plant Microbe In 8 : 84 - 91 .
26. Dong H , Delaney TP , Bauer DW , Beer SV ( 1999 ) Harpin induces disease resistance in Arabidopsis through the systemic acquired resistance pathway mediated by salicylic acid and the NIM1 gene . Plant J 20 : 207 - 215 .
27. Samuel MA , Hall H , Krzymowska M , Drewiecka K , Hennig J , et al. ( 2005 ) SIPK signalling controls multiple components of harpin-induced cell death in tobacco . Plant J 42 : 406 - 416 .
28. Tampakaki AP , Skandalis N , Gazi AD , Bastaki MN , Sarris PF , et al. ( 2010 ) Playing the ''Harp'': Evolution of our understanding of hrp/hrc genes . Annu Rev Phytopathol 48 : 347 - 370 .
29. Desikan R , Reynolds A , Hancock JT , Neill SJ ( 1998 ) Harpin and hydrogen peroxide both initiate programmed cell death but have differential effects on defence gene expression in Arabidopsis suspension cultures . Biochem J 330 : 115 - 120 .
30. Ichinose Y , Andi S , Doi R , Tanaka R , Taguchi F , et al. ( 2001 ) Generation of hydrogen peroxide is not required for harpin-induced apoptotic cell death in tobacco BY-2 cell suspension culture . Plant Physiol Bioch 39 : 771 - 776 .
31. Gopalan S , Wei W , He SY ( 1996 ) Hrp gene-dependent induction of hin1: a plant gene activated rapidly by both harpins and the avrPto gene-mediated signal . Plant J 10 ; 591 - 600 .
32. Lee J , Klessig DF , Nurnberger T (2001b) A harpin binding site in tobacco plasma membranes mediates activation of the pathogenesis-related gene HIN1 independent of extracellular calcium but dependent on mitogen-activated protein kinase activity . Plant Cell 13 : 1079 - 1093 .
33. Qiao F , Chang XL , Nick P ( 2010 ) The cytoskeleton enhances gene expression in the response to the Harpin elicitor in grapevine . J Exp Bot 61 : 4021 - 4031 .
34. Thomma BP , Nu rnberger T , Joosten MH ( 2011 ) Of PAMPs and effectors: the blurred PTI-ETI dichotomy . Plant Cell 23 : 4 - 15 .
35. Tsuda K , Sato M , Glazebrook J , Cohen JD , Katagiri F ( 2008 ) Interplay between MAMP-triggered and SA-mediated defense responses . Plant J 53 : 763 - 775 .
36. Aslam SN , Erbs G , Morrissey KL , Newman MA , Chinchilla D , et al. ( 2009 ) Microbe-associated molecular pattern (MAMP) signatures, synergy, size and charge: inYuences on perception or mobility and host defence responses . Mol Plant Pathol 10 : 375 - 387 .
37. Langcake P ( 1981 ) Disease resistance of Vitis spp. and the production of resveratrol, e-viniferin, a-viniferin and pterostilbene . Physiol Plant Pathol 18 : 213 - 226 .
38. Pezet R , Gindro K , Viret O , Richter H ( 2004 ) Effects of resveratrol, viniferins and pterostilbenes on Plasmopara viticola zoospore mobility and disease development . Vitis 43 : 145 - 148
39. Poinssot B , Vandelle E , Bentejac M , Adrian M , Levis C , et al. ( 2003 ) The endopolygalacturonase 1 from Botrytis cinerea activates grapevine defense reactions unrelated to its enzymatic activity . Mol Plant Microbe In 16 : 553 - 564 .
40. Aziz A , Heyraud A , Lambert B ( 2004 ) Oligogalacturonide signal transduction, induction of defense-related responses and protection of grapevine against Botrytis cinerea . Planta 218 : 767 - 774 .
41. Vandelle E , Poinssot B , Wendehenne D , Bentejac M , Alain P ( 2006 ) Integrated signalling network involving calcium, nitric oxide, and active oxygen species but not mitogen-activated protein kinases in BcPG1-elicited grapevine defenses . Mol Plant Microbe In 19 : 429 - 440 .
42. Nu rnberger T , Scheel D ( 2001 ) Signal transmission in the plant immune response . Trends Plant Sci 6 : 372 - 379 .
43. Jabs T , Tschope M , Colling C , Hahlbrock K , Scheel D ( 1997 ) Elicitor stimulated ion fluxes and O22 from the oxidative burst are essential components in triggering defense gene activation and phytoalexin synthesis in parsley . Proc Natl Acad Sci U S A 94 : 4800 - 4805 .
44. Ding JP , Pickard BG ( 1993 ) Mechanosensory calcium-selective cation channels in epidermal cells . Plant J 3 : 83 - 110 .
45. Zhang S , Klessig DF ( 2001 ) MAPK cascades in plant defence signalling . Trends Plant Sci 6 : 520 - 527 .
46. Jonak C , Okresz L, Bogre L , Hirt H ( 2002 ) Complexity, cross talk and integration of plant MAP kinase signalling . Curr Opin Plant Biol 5 : 415 - 424 .
47. Nick P ( 2011 ) Mechanics of the Cytoskeleton . In: Wojtaszek P, editor. Mechanical Integration of Plant Cells and Plants . Springer Berlin Heidelberg. pp. 53 - 90 .
48. Wojtaszek P ( 1997 ) Oxidative burst: an early plant response to pathogen infection . Biochem J 322 : 681 - 692 .
49. Lamb C , Dixon RA ( 1997 ) The oxidative burst in plant disease resistance . Annu Rev Plant Physiol Plant Mol Biol 48 : 251 - 275
50. Kortekamp A ( 2006 ) Expression analysis of defence-related genes in grapevine after inoculation with a host and a non-host pathogens . Plant Physiol Bioch 44 : 58 - 67 .
51. Reid KE , Olsson N , Schlosser J , Peng F , Lund ST ( 2006 ) An optimized grapevine RNA isolation procedure and statistical determination of reference genes for real-time RT-PCR during berry development . BMC Plant Biology 6 : 27 .
52. Belhadj A , Telef N , Saigne C , Cluzet S , Barrieu F , et al. ( 2008 ) Effect of methyl jasmonate in combination with carbohydrates on gene expression of PR proteins, stilbene and anthocyanin accumulation in grapevine cell cultures . Plant Physiol Bioch 46 : 493 - 499 .
53. Chang X , Heene E , Qiao F , Nick P ( 2011 ) The Phytoalexin Resveratrol Regulates the Initiation of Hypersensitive Cell Death in Vitis Cell. PLoS ONE 6: e26405 . Doi:10.1371/journal.pone.0026405.
54. Schmidt SM , Panstruga R ( 2007 ) Cytoskeleton functions in plant-microbe interactions . Physiol Mol Plant P 71 : 135 - 148 .
55. Felix G , Regenass M , Boller T ( 1993 ) Specific perception of subnanomolar concentrations of chitin fragments by tomato cells: induction of extracellular alkalinization, changes in protein phosphorylation, and establishment of a refractory state . Plant J 4 : 307 - 316 .
56. Di Gaspero G , Cipriani G ( 2003 ) Nucleotide biding site/leucine-rich repeats, Pto-like, receptor-like kinases related to disease resistance in grapevine . Mol Genet Genomics 269 : 612 - 623 .
57. Haapalainen M , Engelhardt S , K ufner I, Li Chun-Mei , Nu rnberger T , et al. ( 2011 ) Functional mapping of harpin HrpZ of Pseudomonas syringae reveals the sites responsible for protein oligomerization, lipid interactions and plant defence induction . Mol Plant Pathol 12 : 151 - 166 .
58. Hohenberger P , Eing C , Straessner R , Durst S , Frey W , et al. ( 2011 ) Plant actin controls membrane permeability . Biochim Biophys Acta 1808 : 2304 - 2312 .
59. Komis G , Illes P , Beck M , S amaj J ( 2011 ) Microtubules and mitogen-activated protein kinase signalling . Curr Opin Plant Biol 14 : 650 - 657 .
60. Wang S , Kurepa J , Hashimoto T , Smalle JA ( 2011 ) Salt Stress-Induced Disassembly of Arabidopsis Cortical Microtubule Arrays Involves 26S Proteasome-Dependent Degradation of SPIRAL1 . Plant Cell 23 : 3412 - 3427 .
61. Yalamanchili RD , Stratmann JW ( 2002 ) Ultraviolet-B activates components of the systemin signalling pathway in Lycopersicon peruvianum suspension-cultured cells . J Biol Chem 277 : 28424 - 28430
62. Holley SR , Yalamanchili RD , Moura SD , Ryan CA , Stratmann JW ( 2003 ) Convergence of signalling pathways induced by systemin, oligosaccharide elicitors, and ultraviolet-B radiation at the level of mitogen-activated protein kinases in Lycopersicon peruvianum suspension-cultured cells . Plant Physiology 132 : 1728 - 1738 .
63. Zhang A , Jiang M , Zhang J , Tan M , Hu X ( 2006 ) Mitogen-activated protein kinase is involved in abscisic acid-Induced antioxidant defense and acts downstream of reactive oxygen species production in leaves of maize plants . Plant Physiology 141 : 475 - 487 .
64. Desikan R , Hancock JT , Ichimura K , Shinozaki K , Neill SJ ( 2001 ) Harpin induces activation of the Arabidopsis mitogen-activated protein kinases AtMPK4 and AtMPK6 . Plant Physiology 126 : 1579 - 1587 .
65. Nu hse TS , Peck SC , Hirt H , Boller T ( 2000 ) Microbial elicitors induce activation and dual phosphorylation of the Arabidopsis thaliana MAPK6 . J Biol Chem 275 : 7521 - 7526
66. Meszaros T , Helfer A , Hatzimasoura E , Magyar Z , Serazetdinova L , et al. ( 2006 ) The Arabidopsis MAP kinase kinase MKK1 participates in defence responses to the bacterial elicitor flagellin . Plant J 48 : 485 - 498 .
67. Robatzek S , Chinchilla D , Boller T ( 2006 ) Ligand-induced endocytosis of the pattern recognition receptor FLS2 in Arabidopsis . Gene Dev 20 : 537 - 542 .
68. Navarro L , Zipfel C , Rowland O , Keller I , Robatzek S , et al. ( 2004 ) Interplay between flg22 innate immune response, Avr9-race-specific defense response and other bacterial interactions . Plant Physiology 135 : 1 - 16 .
69. Torres MA , Jones JDG , Dangl JL ( 2006 ) Reactive oxygen species signalling in response to pathogens . Plant Physiology 141 : 373 - 378 .
70. Zhang J , Shao F , Li Y , Cui H , Chen L , et al. ( 2007 ) A Pseudomonas syringae effector inactivates MAPKs to suppress PAMP-induced immunity in plants . Cell Host Microbe 1 : 175 - 185 .
71. Jeandet P , Douillet-Breuil AC , Bessis R , Debord S , Sbaghi M , et al. ( 2002 ) Phytoalexins from the Vitaceae: biosynthesis, phytoalexin gene expression in transgenic plants, antifungal activity, and metabolism . J Agr Food Chem 50 : 2731 - 2741 .
72. Adrian M , Jeandet P , Veneau J , Weston LA , Bessis R ( 1997 ) Biological activity of resveratrol, a stilbenic compound from grapevines, against Botrytis cinerea, the causal agent for gray mold . J Chem Ecol 23 : 1689 - 1702 .
73. Bruno G , Sparapano L ( 2006 ) Effects of three esca-associated fungi on Vitis vinifera L.: III . Enzymes produced by the pathogens and their role in fungus-toplant or in fungus-to-fungus interactions . Physiol Mol Plant P 69 : 182 - 194 .
74. Celimene CC , Smith DR , Young RA , Stanosz GR ( 2001 ) In vitro inhibition of Sphaeropsis sapinea by natural stilbenes . Phytochemistry 56 : 161 - 165 .
75. Pezet R , Gindro K , Viret O , Spring JL ( 2004 ) Glycosylation and oxidative dimerization of resveratrol are respectively associated to sensitivity and resistance of grapevine cultivars to downy mildew . Physiol Mol Plant P 65 : 297 - 303 .
76. Tropf S , Lanz T , Rensing SA , Schro der J , Schro der G ( 1994 ) Evidence that stilbene synthases have developed from chalcone synthases several times in the course of evolution . J Mol Evol 38 : 610 - 618 .
77. Bellin D , Peressotti E , Merdinoglu D , Wiedemann-Merdinoglu S , AdamBlondon AF , et al. ( 2009 ) Resistance to Plasmopara viticola in grapevine 'Bianca' is controlled by a major dominant gene causing localised necrosis at the infection site . Theor Appl Genet 120 : 163 - 176 .
78. Casagrande K , Falginella L , Castellarin SD , Testolin R , Di Gaspero G ( 2011 ) Defence responses in Rpv3-dependent resistance to grapevine downy mildew . Planta 234 : 1097 - 1099 .
79. Lai Z , Wang F , Zheng Z , Fan B , Chen Z ( 2011 ) A critical role of autophagy in plant resistance to necrotrophic fungal pathogens . Plant J 66 : 953 - 968 .
80. Maisch J , Nick P ( 2007 ) Actin in involved in auxin-dependent patterning . Plant Physiology 143 : 1696 - 1704 .
81. Gaff DF , Okong'O-Ogola O ( 1971 ) The use of non-permeating pigments for testing the survival of cells . J Exp Bot 22 : 756 - 758 .