Non-cross-linked polystyrene-supported triphenylphosphine-microencapsulated palladium: An efficient and recyclable catalyst for Suzuki-Miyaura reactions

Article abstract of DOI:10.1016/j.cell.2006.02.047

A recyclable, all-in-one, polystyrene-microencapsulated palladium catalyst has been developed and used as a heterogeneous catalyst in Suzuki-Miyaura coupling reactions. This catalyst can be recovered and reused with only minimal palladium leaching observed. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1016/j.cell.2006.02.047

Specific Bacterial Suppressors of MAMP
Signaling Upstream of MAPKKK
in Arabidopsis Innate Immunity
Ping He,1,4 Libo Shan,1,4 Nai-Chun Lin,2 Gregory B. Martin,2 Birgit Kemmerling,3 Thorsten Nu¨ rnberger,3
and Jen Sheen1,
*
1 Department of Molecular Biology, Massachusetts General Hospital, and Department of Genetics, Harvard Medical School,
Boston, MA 02114, USA
2 Boyce Thompson Institute for Plant Research and Department of Plant Pathology, Cornell University, Ithaca, NY 14853, USA
3
¨
Eberhard-Karls-Universitat Tubingen, Zentrum fur Molekularbiologie der Pflanzen (ZMBP), Auf der Morgenstelle 5, D-72076
¨
¨
Tu¨ bingen, Germany
4 These authors contributed equally to this work.
*Contact: sheen@molbio.mgh.harvard.edu
DOI 10.1016/j.cell.2006.02.047
SUMMARY
tide binding oligomerization domain (NOD) proteins with
leucine-rich repeats (LRR) are involved in the recognition
of evolutionarily conserved PAMPs or MAMPs (pathogen-
or microbe-associated molecular patterns) and activate
common signaling pathways involving MAPK cascades
and nuclear factor-kB (NF-kB) to induce antimicrobial
cytokine and peptide production (Barton and Medzhitov,
2003; Inohara et al., 2005). As in mammals, plants respond
to an array of MAMPs from both nonpathogenic and path-
ogenic microbes (Ausubel, 2005; Nu¨ rnberger et al., 2004).
However, because plants do not have specialized immune
cells, all plant cells appear to have the ability to respond to
MAMPs, including flagellin, harpin (HrpZ), lipopolysaccha-
ride (LPS), chitin, and necrosis-inducing proteins (NPP),
and activate defense gene transcription and MAPK signal-
ing (Asai et al., 2002; Fellbrich et al., 2002; Lee et al., 2001;
Navarro et al., 2004; Ramonell et al., 2005; Zeidler et al.,
2004). The best-characterized plant MAMP receptor is the
LRR receptor kinase FLS2 that perceives a conserved
22 amino acid peptide (flg22) from bacterial flagellin and
activates MAPK cascades and WRKY transcription fac-
tors in Arabidopsis (Asai et al., 2002; Gomez-Gomez and
Boller, 2002). Emerging evidence indicates the impor-
tance of MAMP and MAPK signaling in plant defense
against pathogenic bacteria and fungi (Ramonell et al.,
2005; Zeidler et al., 2004; Zipfel et al., 2004). For example,
the Arabidopsis fls2 mutant is more susceptible than wild-
type (wt) to infection by the virulent pathogen Pseudomo-
nas syringae pv. tomato DC3000, and the treatment of
wt but not the fls2 mutant plants with flg22 enhances
resistance to DC3000 (Zipfel et al., 2004). Activation of
the flg22-mediated MAPK cascade confers resistance to
both bacterial and fungal pathogens (Asai et al., 2002). A
very recent study shows that the flagellin mutant of a non-
host bacterium P.s. tabaci causes disease symptoms in
Arabidopsis (Li et al., 2005).
Plants and animals possess innate immune sys-
tems to prevent infections and are effectively
‘‘nonhosts’’ for most potential pathogens. The
molecular mechanisms underlying nonhost im-
munity in plants remain obscure. In Arabidop-
sis, nonhost/nonpathogenic Pseudomonas sy-
ringae sustains but pathogenic P. syringae
suppresses early MAMP (microbe-associated
molecular pattern) marker-gene activation. We
performed a cell-based genetic screen of viru-
lence factors and identified AvrPto and AvrPtoB
as potent and unique suppressors of early-
defense gene transcription and MAP kinase
(MAPK) signaling. Unlike effectors of mamma-
lian pathogens, AvrPto and AvrPtoB intercept
multiple MAMP-mediated signaling upstream
of MAPKKK at the plasma membrane linked to
the receptor. In transgenic Arabidopsis, AvrPto
blocks early MAMP signaling and enables non-
host P. syringae growth. Deletions of avrPto
and avrPtoB from pathogenic P. syringae reduce
its virulence. The studies reveal a fundamental
role of MAMP signaling in nonhost immunity,
and a novel action of type III effectors from
pathogenic bacteria.
INTRODUCTION
Molecular mechanisms that distinguish self and nonself
are fundamental in innate immunity to prevent potential
infections by microorganisms in plants and animals (Ausu-
bel, 2005; Barton and Medzhitov, 2003; Nu¨ rnberger et al.,
2004). In mammals, Toll-like receptors (TLRs) and nucleo-
Many gram-negative bacterial pathogens have evolved
type III secretion system (TTSS) to inject effector proteins
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 563

into plant and animal cells (Alfano and Collmer, 2004;
Galan and Cossart, 2005). A key function of type III effec-
tors in animal pathogens is to block immune responses
(Galan and Cossart, 2005). In the case of plant bacterial
pathogens, many type III effectors were originally identi-
fied as so-called avirulence (Avr) factors that turn virulent
strains into avirulent ones. For instance, in the presence
of disease resistance (R) genes in specific plant genotypes,
Avr factors trigger potent gene-for-gene resistance and
hypersensitive response (HR), a localized programmed
cell death (PCD; Dangl and Jones, 2001; Staskawicz
et al., 2001). Interestingly, one P. syringae type III effector,
AvrPto, triggers disease resistance in tomato plants carry-
ing the corresponding R gene Pto, that encodes a serine/
threonine kinase (Pedley and Martin, 2003). Pto could also
recognize another type III effector, AvrPtoB, which shares
little sequence similarity with AvrPto (Kim et al., 2002).
However, in the absence of Pto, AvrPto actually promotes
pathogen growth and virulence (Shan et al., 2000a).
Recently, more type III effectors of plant bacterial path-
ogens have been observed to promote pathogenicity as
the type III effectors of mammalian pathogens (Alfano and
Collmer, 2004; Mudgett, 2005). Genetic and functional
analyses have revealed that a large number of P. syringae
type III effectors, including AvrPtoB and a tyrosine protein
phosphatase HopPtoD2, suppress R gene-mediated PCD
in plants (Abramovitch et al., 2003; Espinosa et al., 2003;
Jamir et al., 2004). Moreover, many type III effectors,
such as AvrPto, AvrRpt2, AvrRpm1, and HopAI1, suppress
defense responses elicited either by TTSS-defective mu-
tant bacteria or flg22 (Hauck et al., 2003; Kim et al., 2005;
Li et al., 2005; Oh and Collmer, 2005). It remained un-
resolved how a large number of type III effectors shared
similar but unknown molecular actions.
with DC3000 and a DC3000 TTSS mutant, but the expres-
sion appeared to be subsequently suppressed only by vir-
ulent DC3000. Using a cell-based genetic screen of type III
effectors as virulence factors, we discovered that AvrPto
and AvrPtoB from DC3000 were specific suppressors
of the MAMP-mediated early-defense responses, includ-
ing transcription activation and MAPK signaling occurring
within minutes of elicitation. AvrPto and AvrPtoB specifi-
cally targeted the MAMP signaling pathways, which can
be uncoupled from some gene-for-gene-mediated tran-
scription and PCD in Arabidopsis. Distinct from type III
effectors of mammalian bacterial pathogens that directly
target MAPKK and MAPK (Galan and Cossart, 2005; Orth
et al., 1999), AvrPto and AvrPtoB acted upstream of
MAPKKK in MAMP signaling near the plasma membrane
receptor. Mutagenesis analysis of AvrPto and AvrPtoB
revealed that their virulence activity in Arabidopsis was
different from that in tomato and tobacco, in which AvrPto
and AvrPtoB can also activate gene-for-gene responses.
The new findings provide strong evidence for dynamic co-
evolution of type III effector actions in individual plant-bac-
terium warfare. Importantly, AvrPto blocks early MAMP
signaling and facilitates the growth of two nonhost P. syrin-
gae strains in plants, verifying the robustness of cell-based
genetic screens. Our results suggest that plant MAMP sig-
naling is essential in nonhost immunity and shed new light
on the molecular action of type III virulence effectors.
RESULTS
Type III Effector-Mediated Suppression of MAMP-
Specific Early-Defense Gene Induction
Purified MAMPs have been shown to activate specific
transcription programs in Arabidopsis, parsley, and to-
bacco (Asai et al., 2002; Lee et al., 2001; Navarro et al.,
2004; Ramonell et al., 2005; Zeidler et al., 2004). For ex-
ample, flg22 treatment of Arabidopsis protoplasts, leaves,
and seedlings leads to the rapid induction of FRK1 (Asai
et al., 2002; see Table S1 in the Supplemental Data avail-
able with this article online). To investigate the expression
of MAMP marker genes in natural plant-microbe interac-
tions, we compared the activation of FRK1 by various
P. syringae strains, including the nonhost Psp NPS3121,
the virulent strain DC3000, and a DC3000 TTSS mutant
hrcC in Arabidopsis leaves. Quantitative RT-PCR analysis
showed that all three bacterial strains strongly activated
FRK1 expression 2 hr postinoculation (hpi; Figure 1A). Sig-
nificantly, FRK1 activation was reduced by 6 hpi with the
virulent strain but was maintained or enhanced further
following infection with the nonhost or the TTSS mutant
strain. Simultaneous inoculation with the virulent and the
nonhost or the TTSS mutant strains did not diminish the
apparent suppression of FRK1 expression by DC3000
(Figure 1A). The data suggest that DC3000 likely secretes
type III effectors to shut down early-defense signaling.
The expression kinetics of FRK1 calculated from AtGen
Express microarray database (http://Arabidopsis.org/
info/expression/ATGenExpress.jsp) in Arabidopsis leaves
Nonhost immunity is the most prevalent form of plant
defense against a broad spectrum of potential pathogens.
Currently, the molecular mechanisms underlying nonhost
immunity and pathogenicity are obscure. It is also unclear
whether nonhost and gene-for-gene defense pathways
share the same regulatory mechanisms (Mysore and
Ryu, 2004; Thordal-Christensen, 2003). The conventional
phenotypic tests that have been used to characterize
the plant immune responses, including PCD, cell wall
modification, pathogenesis-related (PR) gene activation
and the inhibition of bacterial growth, measure relatively
late outcomes in plant defense. To distinguish different
defense pathways and to differentiate defense suppres-
sion mechanisms by type III effectors, we tested the idea
of using molecular markers and biochemical assays to
monitor specific and early-defense responses. We first
investigated the activation of early MAMP-specific marker
genes (Asai et al., 2002) in Arabidopsis leaves inoculated
with a naturally nonhost/nonpathogenic bacterial strain,
P. syringae pv. phaseolicola NPS3121 (Psp NPS3121).
Although not a pathogen in Arabidopsis, Psp NPS3121
rapidly activated FRK1 and other MAMP-specific early-
defense genes. Interestingly, the same marker-gene acti-
vation was also observed in Arabidopsis leaves inoculated
564 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

was very similar to that quantified in our experiments by
RT-PCR (Figure 1B). In addition, we collected data from
AtGenExpress for three other genes (At1g51890,
At2g17740, and At5g57220) that were highly activated
by flg22 in Arabidopsis protoplasts, leaves, and seedlings
(Table S1). All three genes showed similar expression pat-
terns as FRK1, including a strong diminution following in-
fection with DC3000 by 6 hpi (Figure 1B). It appears that
the diminution of FRK1, At1g51890, At2g17740, and
At5g57220 expression during early DC3000-Arabidopsis
interactions (2–12 hr) involves specific mechanisms for
the suppression of these genes because no bacterial pro-
liferation, cell death, or general repression of genes was
observed (He et al., 2004; AtGenExpress; Table S2).
Moreover, these early-defense marker genes were
strongly activated by multiple MAMPs, including the bac-
terial elicitors flg22 and HrpZ and the fungal/oomycete
elicitor NPP1 (Figure 1C), but not by other stress-related
signals based on a survey of available global gene-
expression profiles (AtGenExpress).
AvrPto and AvrPtoB Are Specific Suppressors of
Early-Defense Signaling
To screen for type III effectors in suppressing the early-
defense-related gene induction, we transiently expressed
individual type III effectors in Arabidopsis mesophyll pro-
toplasts and determined their effects on the activation of
the FRK1-LUC reporter gene by flg22 (Asai et al., 2002).
These bacterial effector proteins were well expressed in
plant cells (Figure 2A). We examined HopPtoD2, HopPtoE,
HopPtoK, AvrPto (DC), and AvrPtoB from DC3000, all of
which displayed host defense-suppression activities by
other assays (Abramovitch et al., 2003; Espinosa et al.,
2003; Hauck et al., 2003; He et al., 2004; Jamir et al.,
2004). Surprisingly, AvrPto (DC) and AvrPtoB, but not
the other effectors tested, suppressed flg22 activation of
FRK1-LUC (Figure 2B). The effect of AvrPto from DC3000
on suppressing MAMP signaling was as potent as AvrPto
from Pst JL1065 (which differs in four amino acids; Fig-
ure 2B). Here, we designate AvrPto from Pst JL1065 as
AvrPto and AvrPto from DC3000 as AvrPto (DC).
We extended our screen to several well-studied Avr
effectors that have also been shown to have virulence ac-
tivities, including AvrRpm1, AvrB, and AvrRpt2. Because
the expression of AvrRpm1 and AvrB induced rapid PCD
in wt Arabidopsis protoplasts (P.H. and L.S., unpublished
data), we expressed AvrRpm1 and AvrB in the rps3 mutant
protoplasts, which lack functional RPM1 protein. Interest-
ingly, although it has been reported that AvrRpm1 blocked
flg22-elicited callose deposition (Kim et al., 2005), AvrRpm1
and AvrB did not suppress flg22 activation of the FRK1
promoter (Figure 2C). Similarly, AvrRpt2 did not suppress
Figure 1. Virulent Pseudomonas Inhibits but Nonhost Pseu-
domonas Potentiates MAMP-Mediated Early Gene Expres-
sion
(A) Real-time RT-PCR analysis of FRK1 expression in Arabidopsis
leaves. Inoculation was performed with MgCl2 (Control), DC3000,
DC3000hrcC, Psp NPS3121 (Psp), and the combination of DC3000
with DC3000hrcC or Psp NPS3121 at 1 3 108 cfu/ml. The gene induc-
tion (fold change) by bacterial infiltration was compared to the expres-
sion level of MgCl2 infiltration. The data are shown as means stan-
dard errors from three independent biological replicates.
(B) Expression patterns of FRK1, At1g51890, At2g17740, and
At5g57220 in Arabidopsis leaves inoculated with different Pseudomo-
nas strains. The triplicated data were searched and presented as
means standard errors from AtGenExpress microarray results pub-
lished on TAIR (The Arabidopsis Information Resource) website.
(C) Convergent gene activation by flg22, HrpZ, and NPP1 in Arabidop-
sis leaves. GST protein was used as a control for the NPP1-GST fusion
protein. The peak induction level (means standard errors) from the
AtGenExpress triplicated data is shown.
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 565

Figure 2. AvrPto and AvrPtoB Specifi-
cally Suppress Early MAMP Marker-
Gene Activation by Flg22
(A) Protein expression of effectors in wt, rps2,
and rps3 mutant Arabidopsis protoplasts. Ex-
pression was detected by Western blot using
an anti-HA antibody.
(B) Suppression of flg22-induced FRK1 pro-
moter activity by AvrPto and AvrPtoB. Proto-
plasts were cotransfected with an effector and
a LUC reporter and incubated for 6 hr before
treated with 100 nM flg22 for 3 hr.
(C) AvrRpt2, AvrRpm1, and AvrB do not sup-
press flg22-induced FRK1 promoter activity in
rps2 or rps3 protoplasts.
(D) RT-PCR analysis of AvrPto and AvrPtoB
suppression. Transfected protoplasts were in-
cubated for 6 hr before treated with 1 mM flg22
for 1 hr. UBQ10 was used as an internal control.
(E) AvrPto does not affect endogenous
WRKY46 transcription activation by AvrRpm1,
AvrB, and AvrRpt2.
(F) AvrPto and AvrPtoB do not affect WRKY46
promoter activation by AvrRpm1, AvrB, and
AvrRpt2.
(G) AvrPto and AvrPtoB do not interfere with cell
death induced by AvrRpm1, AvrB, and AvrRpt2.
Cell death was observed by Evans blue staining
16 hr after transfection.
All data are the representatives of four indepen-
dent replicates and the pooled data are shown
as means standard errors.
FRK1-LUC in either wt or rps2 protoplasts, which lack
functional RPS2 (Figures 2B and 2C). Recently, nine effec-
tors were identified as the suppressors of flg22-induced
NONHOST1 (NHO1) expression (Li et al., 2005). However,
HopAI1, one of the most potent suppressors of NHO1, did
not affect flg22-induced early and specific MAMP reporter
gene FRK1-LUC (Figure 2B). As reported, HopAI1 sup-
pressed flg22-induced NHO1 expression in the protoplasts
(Figure S1). The results suggest that distinct mechanisms
are utilized by type III effectors to suppress different de-
fense responses occurring at different time points or steps
during infection.
NPS3121. Surprisingly, unlike AvrPtoB, VirPphA did not
suppress flg22 induction of FRK1-LUC (Figure 2B). Similar
to the reporter-gene assays, the activation of endogenous
FRK1 expression by flg22 was significantly blocked by
AvrPto, AvrPto (DC), and AvrPtoB as shown by RT-PCR
analysis (Figure 2D). Importantly, the flg22-induced ex-
pression of other MAMP marker genes, At1g51890,
At2g17740, and At5g57220, was also substantially sup-
pressed by AvrPto, AvrPto (DC), and AvrPtoB (Figure 2D).
AvrPto and AvrPtoB Do Not Interfere with Gene-for-
Gene Defense in Arabidopsis
Although avrPto-like sequences exist only in a small
subset of P. syringae strains, avrPtoB-like sequences are
more widely distributed in P. syringae strains, including
the nonhost Psp (Kim et al., 2002). Analysis of the whole-
genome sequence of Psp strain 1448A did not identify
avrPto-like sequences, and the avrPtoB-like sequence
only produced a truncated protein (Figure S2). We also
cloned and tested virPphA, an avrPtoB homolog from Psp
Many studies have suggested that several type III effec-
tors, including AvrPtoB, can interfere with PCD induced by
gene-for-gene interactions or other signals in Arabidopsis,
tobacco, andtomato (Abramovitch et al., 2003; Jamir et al.,
2004). We used the protoplast transient assay to test
whether AvrPto or AvrPtoB could suppress Avr-R-medi-
ated responses, including defense-gene activation and
elicitation of cell death by AvrRpm1, AvrB, and AvrRpt2.
566 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

Figure 3. AvrPto and AvrPtoB Block MAPK Activation by Flg22
(A) AvrPto and AvrPtoB block endogenous MAPK activation by flg22. Transfected protoplasts were incubated for 6 hr before 1 mM flg22 treatment for
10 min. The kinase activity was detected by an in-gel kinase assay (top). Effector protein expression was examined by Western blot using an anti-HA
antibody (bottom).
(B) AvrPto and AvrPtoB inhibit flg22 activation of MPK3 and MPK6. HA-tagged MPK3 or MPK6 was coexpressed with FLAG-tagged effectors. Trans-
fected protoplasts were incubated for 6 hr before 1 mM flg22 treatment for 10 min. An anti-HA antibody was used for immunoprecipitation of MPK3 or
MPK6. Kinase activity was detected by an in vitro kinase assay (top). Protein expression is shown for MAPKs (middle) and effectors (bottom).
(C) HopPtoD2 does not affect MPK3 and MPK6 activation by flg22.
(D) VirPphA and HopAI1 do not affect flg22 activation of MPK3.
(E) AvrBsT does not suppress flg22 activation of MPK3 and MPK6.
All of the above experiments were repeated three times with similar results.
Because these effectors do not appear to activate the
MAMP-specific genes in Arabidopsis, we selected another
marker gene, WRKY46, for these studies (P.H. and L.S.,
unpublished data). AvrPto or AvrPtoB was cotransfected
with AvrRpm1, AvrB, or AvrRpt2 in wt protoplasts.
AvrRpm1, AvrB, and AvrRpt2 induced endogenous
WRKY46 (Figure 2E), WRKY46-LUC (Figure 2F), and cell
death (Figure 2G) equally well in the presence or absence
of AvrPto or AvrPtoB, suggesting the occurrence of normal
gene-for-gene-mediated defense signaling. Our results
obtained using protoplast transient assays are consistent
with the finding obtained in AvrPto-expressing plants,
in which PCD (HR) occurred normally in response to
DC3000 (avrRpt2) (Hauck et al., 2003). In addition, conju-
gated DC3000, carrying functional avrPto and avrPtoB,
with plasmid expressing avrRpm1, avrB, or avrRpt2 could
trigger gene-for-gene responses in Arabidopsis ecotype
Col-0. Thus, AvrPto and AvrPtoB appear to specifically in-
hibit MAMP-mediated but not the Avr-R defense signaling
in Arabidopsis. Moreover, because AvrPto and AvrPtoB
did not affect the expression of the endogenous UBQ10
gene, the cotransfected UBQ10-GUS (Figures 2E and
2F), or a variety of other genes and promoters (data not
shown), it does not seem likely that these effectors non-
specifically killed the protoplasts resulting in an apparent
suppression of early-defense gene expression.
AvrPto and AvrPtoB Block MAPK Activation in Early
Flg22 Signaling
MAPK activation is a convergent and evolutionarily con-
served event in the earliest MAMP signaling in plants
and animals (Asai et al., 2002; Barton and Medzhitov,
2003; Nu¨ rnberger et al., 2004). To elucidate the molecular
mechanisms underlying AvrPto and AvrPtoB suppression
of early MAMP signaling, we investigated the activation of
endogenous MAPKs by flg22 in transfected protoplasts
expressing AvrPto or AvrPtoB. Both AvrPto and AvrPtoB
significantly blocked MAPK activation by flg22 (Figure 3A).
The remaining low level of MAPK activation is most likely
due to flg22 activation in untransfected cells since the pro-
toplast transfection efficiency is about 90%. It has been
shown that flg22 specifically activates MPK3 and MPK6
in Arabidopsis (Asai et al., 2002). To determine whether
AvrPto and AvrPtoB suppress the activation of these same
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 567

MAPKs by flg22, a construct expressing epitope-tagged
MPK3 or MPK6 was cotransfected into protoplasts. As
shown in Figure 3B, the activation of MPK3 and MPK6
by flg22 was completely blocked by AvrPto and AvrPtoB
but not by AvrRpt2.
HopPtoD2, a tyrosine protein phosphatase, blocks
MAPK-associated cell death in plants (Espinosa et al.,
2003). However, it was ineffective in blocking MPK3
and MPK6 activation by flg22 (Figure 3C). Consistent
with the analysis of early gene activation, VirPphA from
Psp NPS 3121 and HopAI1 from DC3000 did not block
MPK3 and MPK6 activation by flg22 (Figure 3D and data
not shown).
In mammalian cells, it has been shown that the Yersinia
type III effector YopJ, a cysteine protease, can bind
directly to MKKs and block both MAPK and IKK-NF-kB
signaling pathways (Orth et al., 1999). However, AvrBsT,
a YopJ-related protein from Xanthomonas campestris
pv. vesicatoria (Mudgett, 2005), did not block the activa-
tion of MPK3 and MPK6 induced by flg22 (Figure 3E). Sim-
ilarly, AvrBsT did not inhibit flg22 activation of FRK1-LUC
(Figure 2B). Expressing YopJ in Arabidopsis protoplasts
also did not affect MAPK activation by flg22 (data not
shown). It appears unlikely from these results that AvrBsT
targets plant MKKs, at least in Arabidopsis.
AvrPto and AvrPtoB Intercept the Early-Defense
Signaling Activated by Several MAMPs
Plant innate immune responses are triggered by a variety
of different MAMPs, and although different MAMPs are
probably perceived by distinct receptors, convergent
early-signaling events, including MAPK activation and
specific defense-gene induction, have been observed in
Arabidopsis plants and mesophyll protoplasts (Asai et al.,
2002; Fellbrich et al., 2002; Lee et al., 2001; Navarro
et al., 2004). To determine whether AvrPto interrupts the
immune responses activated by MAMPs other than
flg22, we treated AvrPto- or AvrPtoB-transfected proto-
plasts with HrpZ and NPP1. Similar to flg22, both HrpZ
and NPP1 activated FRK1 promoter in transfected Arabi-
dopsis protoplasts, and the activation of this promoter
was dramatically inhibited in the presence of AvrPto,
AvrPto (DC), or AvrPtoB (Figure 4A). Moreover, AvrPto
and AvrPtoB abolished MPK3 and MPK6 activation by
HrpZ and NPP1 (Figure 4B). These results indicate that
diverse MAMPs activate common Arabidopsis innate
immunity-signaling pathways.
Figure 4. AvrPto and AvrPtoB Block Early-Defense Gene and
MAPK Activation by HrpZ and NPP1
(A) AvrPto and AvrPtoB suppress FRK1 promoter activation by multiple
MAMPs. Transfected protoplasts were incubated for 6 hr and treated
with 100 nM flg22, 1mM HrpZ, 20 nM GST, or 20 nM NPP1-GST for 3 hr.
The data are shown as means standard errors from four repeats.
(B) AvrPto and AvrPtoB intercept activation of MPK3 and MPK6 by
HrpZ and NPP1. Transfected protoplasts were incubated for 6 hr be-
fore treatment with 1mM HrpZ, 20 nM GST, or 20 nM NPP1-GST for
10 min. The experiments were repeated three times with similar results.
upstream of WRKY transcription factors. Similar results
were obtained for AvrPto (DC) and AvrPtoB (data not
shown). Moreover, constitutively active MKK4/5 or MEKK1,
which activates MPK3/6 in the absence of flg22 (Asai
et al., 2002), overrode AvrPto and AvrPtoB suppression
(Figures 5B and 5C and data not shown), indicating that
AvrPto and AvrPtoB block MAMP signaling very early,
probably immediately after signal perception at or up-
stream of MAPKKK.
AvrPto and AvrPtoB Suppress MAMP Signaling
Upstream of MAPKKK
To further elucidate the molecular mechanism of AvrPto
and AvrPtoB action in suppressing MAMP signaling, we
carried out epistasis analysis of the flg22 signaling path-
way using gain-of-function components that activate ei-
ther MPK3/MPK6 or FRK1-LUC in the absence of flg22
signal (Asai et al., 2002). As shown in Figure 5A, ectopic
expression of WRKY22 or WRKY29 bypassed AvrPto sup-
pression, suggesting that AvrPto inhibits flg22 signaling
Mutational Analysis of AvrPto and AvrPtoB as Novel
Suppressors of MAMP Signaling
To further investigate the activity requirement of AvrPto as
a virulence effector in Arabidopsis cells, we analyzed a set
of well-defined AvrPto mutants (Table S3 and Figure 6B)
that structurally separate distinct avirulence and virulence
functions in tomato and tobacco (Shan et al., 2000a,
568 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

2000b). As in transgenic tobacco (Shan et al., 2000b),
AvrPto-GFP was properly targeted to the plasma mem-
brane in Arabidopsis protoplasts (Figure 6A). The G2A mu-
tation of the myristoylation site in AvrPto disrupted its
membrane localization (Figure 6A) as well as its functions
in suppressing FRK1-LUC (Figure 6C) and MPK3/6 (Fig-
ure 6D) activation by flg22. Three other point mutations,
S46P, S94P, and I96T, but not P146L and S147R, also
abolished the AvrPto suppressor functions in Arabidopsis
(Figures 6C and 6D). S94 and I96 were only required for
AvrPto interaction with Pto and the avirulence function in
tomato, but not the virulence function in tomato or the avir-
ulence function in tobacco. P146L and S147R mutants
specifically blocked avirulence function in tobacco but still
kept the interaction with tomato Pto (Shan et al., 2000b). It
is intriguing that residues required for the specific interac-
tion between AvrPto and tomato Pto were also important
for AvrPto in blocking MAMP-mediated defense in Arabi-
dopsis (Table S3). To further examine this observation, we
tested whether expression of tomato Pto in Arabidopsis
could inhibit the AvrPto suppressor function in MAMP sig-
naling. As shown in Figure 6E, expression of Pto partially
relieved the AvrPto-mediated suppression of FRK1-LUC
induction by flg22. However, it is likely that AvrPto is very
effective with multiple targets and cannot be completely
sequestered by the ectopic expression of tomato Pto in
Arabidopsis cells. The mutational analysis suggests that
plasma membrane localization and protein-protein inter-
action via S46, S94, and I96 are essential for the suppres-
sor activity of AvrPto.
Because AvrPto acts at the plasma membrane and
blocks MAPK activation upstream of MAPKKK within min-
utes of MAMP elicitation, and the flg22 receptor FLS2
shares a Pto-like kinase domain (Shiu and Bleecker,
2003), we tested the ability of FLS2 to interfere with the
AvrPto suppressor activity in early signaling initiated by
multiple MAMPs in Arabidopsis. Significantly, FLS2 could
partially interfere with the suppressor activity of AvrPto, al-
though expression of FLS2 alone did not enhance the
flg22 response (Figure 6E). The interference apparently
required the membrane localization of FLS2 because
overexpression of DFLS2 with only the kinase domain
did not affect the AvrPto function (Figure 6E). The steady-
state level of the FLS2 protein was too low to carry out
coimmunoprecipitation experiments with AvrPto. How-
ever, the low level of FLS2 expression was also sufficient
to partially relieve the suppression of HrpZ and NPP1 sig-
naling by AvrPto (Figure 6F).
Recent studies have discovered that AvrPtoB carries an
E3 ubiquitin ligase activity, and its conserved E2 binding
residues are required to suppress PCD activated by
a gene-for-gene response in tomato (Janjusevic et al.,
Figure 5. AvrPto and AvrPtoB Suppressors Act Upstream of
MAPKKK
(A) AvrPto functions upstream of WRKY transcription factors. WRKY
transcription factors were coexpressed with FRK1-LUC, UBQ10-
GUS, and AvrPto. The expression of WRKYs and AvrPto is shown.
The data are shown as means standard errors.
(C) AvrPto and AvrPtoB do not block MEKK1 activation of MPK3/6.
Protoplasts were transfected with HA-tagged MAPK, FLAG-tagged
constitutively active MEKK1 (DMEKK1), and FLAG-tagged effectors.
All of the above experiments were repeated four times with similar
results.
(B) AvrPto does not inhibit MKK4/5 activation of MPK3/6. Protoplasts
were transfected with HA-tagged MAPK, MYC-tagged constitutively
active MKK (MKKac), and FLAG-tagged AvrPto.
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 569

Figure 6. Mutational Analysis of AvrPto
and AvrPtoB Suppressor Function
(A) Membrane localization of AvrPto in proto-
plasts. The pictures were taken 12 hr after
transfection with AvrPto-GFP or AvrPtoG2A-
GFP mutant in the protoplasts using confocal
microscope.
(B) Expression of different AvrPto mutants.
(C) Effect of flg22-activated FRK1 promoter by
different AvrPto mutants.
(D) Effect of flg22-activated MPK3/6 by differ-
ent AvrPto mutants.
(E) Pto and FLS2 partially interfere with AvrPto
suppressor function in flg22 signaling. A con-
struct expressing the tomato Pto or the Arabi-
dopsis FLS2 was cotransfected with a plasmid
expressing AvrPto under the control of the
dexamethasone-inducible promoter. Trans-
fected protoplasts were incubated for 4 hr to
express Pto or FLS2 first before treated with
20 mM dexamethasone (DEX) to induce AvrPto
for 3 hr and then treated with 100 nM flg22 for
3 hr. Expression of Pto or FLS2 alone did not
affect FRK1-LUC activation.
(F) Pto and FLS2 partially interfere with
AvrPto suppressor function in HrpZ and NPP1
signaling.
(G) A mutation in E2 binding residue of AvrPtoB
does not affect its suppressor function in flg22
signaling.
The experiments were repeated at least three
times and the pooled data are shown as
means standard errors.
2006). Since AvrPto does not carry the E3 ligase se-
quence, the suppressor function shared by AvrPto and
AvrPtoB in Arabidopsis is likely uncoupled from the E3 li-
gase activity of AvrPtoB. Consistent with this hypothesis,
a key mutation in one of the E2 binding residues F525 abol-
ished AvrPtoB E3 ligase and anti-PCD activities in tomato
(Janjusevic et al., 2006) but did not affect its suppressor
function in flg22 signaling in Arabidopsis (Figure 6G).
and S4). Consistent with transient expression of AvrPto in
protoplasts, the expression of AvrPto in stable transgenic
plants suppressed flg22-activated endogenous MAPKs
(Figure 7B), and the induction of FRK1, At1g51890,
At2g17740, and At5g57220 by flg22, DC3000 hrcC, and
Psp NPS3121 (Figure 7C). In addition, expression of
AvrPto in transgenic plants also suppressed MPK3/6 acti-
vation by flg22 (Figure S5). The suppression appeared
to act upstream of MAPKKK because preexpression of
AvrPto at a high level could not block MEKK1 activity (Fig-
ure S6). The transgenic plant analyses validated the proto-
plast transient assays and demonstrated that expression
of a single type III effector in plant cells can suppress non-
host immunity.
AvrPto Impaired Nonhost Immunity and Enabled
Nonhost Bacteria Growth in Transgenic Plants
To determine the importance of MAMP signaling in non-
host defense, we tested whether inducible expression of
AvrPto could inhibit nonhost immunity in intact plants.
We generated avrPto transgenic plants under the control
of the dexamethasone-inducible promoter (McNellis et al.,
1998). As shown in protoplasts, AvrPto was localized to
the plasma membrane in transgenic plants (Figure S3). Sig-
nificantly, AvrPto expression enabled the growth of two
nonhost strains Psp NPS3121 and P.s. tabaci (Figures 7A
Both AvrPto and AvrPtoB Contribute to DC3000
Virulence in Arabidopsis
To determine the effect of avrPto and avrPtoB on DC3000
virulence in Arabidopsis, we dip-inoculated Col-0 plants
with the DC3000 avrPto or avrPtoB deletion mutant, or
570 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

Figure 7. Analysis of avrPto Transgenic
Plants and DC3000 Deletion Mutants
(A) AvrPto transgenic plants support nonhost
bacteria growth. Arabidopsis wt or avrPto
transgenic plant (L120 and L121) leaves were
inoculated with Psp NPS3121 (Psp), or P.s. ta-
baci (Psta) at 5 3 105 cfu/ml. The experiment
was repeated three times with similar results.
AvrPto protein expression after DEX treatment
is shown.
(B) Expression of AvrPto suppresses flg22 acti-
vation of endogenous MAPKs. Arabidopsis wt
or avrPto transgenic plant (L120) leaves were
collected 30 min after infiltration with 10 mM
flg22. The experiment was repeated twice
with similar results.
(C) Real-time RT-PCR analysis of MAMP
marker gene induction. Arabidopsis wt or
avrPto transgenic plant (L120 and L121) leaves
were collected 6 hpi with H2O, flg22 (10 mM), or
bacteria, DC3000hrcC or Psp NPS3121 at 1 3
108 cfu/ml. The gene induction (fold change) by
bacterial infiltration was compared to the ex-
pression level of control infiltration. The data
are the average of three independent repli-
cates.
(D) Bacterial growth assay of DC3000 deletion
mutants. Arabidopsis plants were dipped with
DC3000 or avrPto or/and avrPtoB mutant bac-
terial strains for 30 s at the density of 1 3 107
cfu/ml. (ev): empty vector; (avrPtoB): comple-
mentation with avrPtoB.
The data are shown as means
errors.
standard
DavrPtoDavrPtoB double mutant (Lin and Martin, 2005).
Two days after inoculation, populations of DavrPto and
DavrPtoB were 3- and 2-fold, respectively, lower than
that of DC3000, whereas the bacterial number of the dou-
ble mutant was reduced by more than 10-fold (Figure 7D).
Ectopic expression of avrPtoB partially restored bacterial
growth of DavrPtoDavrPtoB. Thus, both AvrPto and AvrP-
toB appear to contribute to the pathogenicity of DC3000
in Arabidopsis. The DavrPtoDavrPtoB double mutant par-
tially reduced the suppression activity of DC3000 on FRK1
expression (data not shown).
but not a nonhost strain or a TTSS-deficient mutant,
MAMP signaling was subsequently repressed. To investi-
gate the underlying molecular mechanisms, we estab-
lished a cell-based genetic screen focusing on the earliest
defense responses and identified AvrPto and AvrPtoB as
specific suppressors of early MAMP signaling. Distinct
from the Yersinia type III effector YopJ, which blocks
MKK and IKK signaling in mammalian innate immunity,
AvrPto and AvrPtoB most likely exert their suppressor ac-
tivities upstream of MAPKKK near the plasma membrane
MAMP receptors. Mutagenesis analysis of AvrPto reveals
the importance of plasma membrane localization and pro-
tein-protein interaction for its suppressor action. Impor-
tantly, AvrPto suppresses early MAMP signaling and
impairs nonhost immunity in plants, and the deletions of
avrPto and avrPtoB from DC3000 reduce its virulence.
Previous studies already indicate the importance of
MAMP and MAPK signaling in plant defense against infec-
tions (Asai et al., 2002; Zipfel et al., 2004). Together, our re-
sults provide molecular evidence that the convergent and
sustained MAMP signaling may prevent most microorgan-
isms from infection in most plants while successful virulent
bacteria inject specific type III effectors targeting the early
MAMP signaling pathways.
DISCUSSION
Although plant cells have long been known to respond to
diverse MAMPs, there has been limited molecular and
genetic evidence supporting a major role of the conserved
and convergent MAMP signaling pathways in natural
plant-microbe encounters (Gomez-Gomez and Boller,
2002; Nu¨ rnberger et al., 2004). Using MAMP-specific
early-defense marker genes, our data show that MAMP
signaling was rapidly activated by both pathogenic and
nonpathogenic Pseudomonas strains in Arabidopsis
leaves. However, in the interactions with a virulent strain,
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 571

Rapid Activation of MAMP Signaling in Plant
Nu¨ rnberger et al., 2004). As one of the key mechanisms
to protect plants from infection by a broad spectrum of po-
tential pathogens, it is not surprising that the conserved
MAMP signaling components, such as the MAPK cas-
cades and WRKY transcription factors, are encoded by
functionally redundant genes in Arabidopsis (Asai et al.,
2002; Zhang and Klessig, 2001). The discovery of potent
suppressors such as AvrPto and AvrPtoB offers a powerful
tool for dissecting the role of MAMP signaling in the plant
nonhost immunity.
Immunity
Direct exposure to diverse purified MAMPs has been
shown to rapidly activate MAPK signaling and alter tran-
scription programs in Arabidopsis, parsley, and tobacco
(Asai et al., 2002; Fellbrich et al., 2002; Lee et al., 2001;
Ligterink et al., 1997; Navarro et al., 2004; Ramonell
et al., 2005; Zeidler et al., 2004; Zhang and Klessig,
2001). Moreover, bacterial extracts suppress DC3000
growth in the absence of the flg22 receptor FLS2 in Arabi-
dopsis (Zipfel et al., 2004), indicating that other MAMPs
are likely recognized during plant-microbe interactions
to activate rapid changes in transcription profiles.
Type III Effectors as Suppressors of Plant Defense
System
Real-time RT-PCR analysis of the early-defense genes
(Figures 1 and 7) shows that specific and convergent
MAMP-mediated defense responses occur rapidly after
inoculation with different bacteria. The data are consistent
with the results from triplicated microarray experiments
from the AtGenExpress project (Figure 1B). Similar early
and robust FRK1 activation by DC3000, DC3000 hrpA,
and DC3000 (avrRpm1) has been observed in Arabidopsis
leaves (de Torres et al., 2003). However, infiltration with vir-
ulent bacteria carrying different effector genes in leaf cells
showed very complex gene-expression patterns, includ-
ing genes induced by wounding, flooding, and cell death
(de Torres et al., 2003; Tao et al., 2003). It has been sug-
gested that nonhost, basal, MAMP, and gene-for-gene
defense responses are very similar and only differ in quan-
titative nature and timing (Mysore and Ryu, 2004; Navarro
et al., 2004; Tao et al., 2003). Our data reveal distinct de-
fense gene regulation in early gene-for-gene and MAMP
signaling pathways (Figure 2; P.H., L.S., and J.S., unpub-
lished data). Further comparison of Arabidopsis gene-ex-
pression profiles induced by purified MAMPs, individual
type III effectors, and different bacterial strains using
higher stringency (e.g., low p value and high log2 ratio)
at different time points (earlier than 3 hr) will more precisely
define genes that are either specific or common to MAMP-
mediated and gene-for-gene defense pathways or in-
duced by abiotic stresses.
Animal bacterial pathogens secrete a few type III effectors
with well-characterized enzymatic activities (Galan and
Cossart, 2005). Plant bacterial pathogens, on the other
hand, produce a relatively large number of putative type
III effectors (Alfano and Collmer, 2004). Recently, several
Pseudomonas type III effectors have been shown to sup-
press different types of PCD in plants and in yeast (Abra-
movitch et al., 2003; Jamir et al., 2004). For example,
AvrPtoB blocks HopPsyA-mediated PCD in tobacco and
Arabidopsis Ws-0 (Jamir et al., 2004). However, AvrPto
and AvrPtoB do not interfere with PCD and transcription
activation triggered by AvrRpt2, AvrRpm1, or AvrB in Ara-
bidopsis (Figures 2F and 2G). It is possible that distinct
mechanisms are utilized by different R genes to trigger
PCD. Our results uncouple MAMP-mediated defense
from some specific gene-for-gene defense in Arabidopsis.
Although AvrPto, AvrRpt2, AvrRpm1, and HopAI1 all sup-
port the growth of DC3000 TTSS mutants or inhibit callose
formation in transgenic Arabidopsis (Hauck et al., 2003;
Kim et al., 2005; Li et al., 2005; Figure S4), our analyses
show that AvrPto supports nonhost bacteria growth in
plants (Figure 7A). It is clear from our studies that AvrPto
and AvrPtoB act uniquely at a very early stage immediately
after MAMP signal perception (Figures 5 and 6). In con-
trast, AvrRpt2, AvrRpm1, and HopAI1 do not interfere
with early MAMP-specific gene activation and MAPK sig-
naling (Figures 2 and 3). Thus, there appears to be diverse
mechanisms by which type III effectors block the plant de-
fense responses. Recent studies have uncovered MAPK-
independent pathways acting downstream of the FLS2 re-
ceptor in flg22 signaling (S. Ramu and J.S., unpublished
data; M. Willmann and J.S., unpublished data). AvrPto
and AvrPtoB could suppress the flg22-mediated induction
of PAL1 (Figure S7), which is activated by both MAPK-de-
pendent and MAPK-independent pathways in flg22 sig-
naling (J.S., unpublished data). It is likely that AvrPto and
AvrPtoB block multiple signaling pathways initiated from
the receptor complex. Future research will unravel the
detailed molecular mechanisms blocking the convergent
signaling pathways in response to diverse MAMPs.
Although AvrPto and AvrPtoB function similarly as sup-
pressors of MAMP-mediated defense responses in Arabi-
dopsis, they share very limited sequence homology (Kim
et al., 2002). It is intriguing that AvrPto and AvrPtoB were
originally identified as Avr factors displaying the same
Stability and robustness of inducible nonhost defense is
the likely consequence of activation of multiple signaling
pathways against a broad range of microorganisms (My-
sore and Ryu, 2004; Thordal-Christensen, 2003). Recent
new findings in the genetic dissection of nonhost immunity
include the isolation of the Arabidopsis nho1 and pen
(penetration) mutants that are more susceptible to Psp
NPS3121 and Blumeria graminis f. sp. Hordei, respectively
(Kang et al., 2003; Lipka et al., 2005). However, the regu-
lation and function of NHO1, encoding a glycerol kinase,
is complex (Eastmond, 2004). It is not only activated by
flg22 and other MAMPs but is also strongly activated by
leaf senescence and various stresses, such as heat,
cold, wounding, and UV-B, and different hormones, in-
cluding gibberellin and ABA (AtGenExpress, Eastmond,
2004). Few mutants in the convergent MAMP-mediated
defense signaling pathways downstream of the receptors
have been identified (Gomez-Gomez and Boller, 2002;
572 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

fugation, washed, and diluted to the desired density with 10 mM
MgCl2. Arabidopsis leaves were either infiltrated with bacteria using
a needleless syringe or inoculated by dipping for 30 s containing
0.025% silwet L-77. Plant leaves were collected at the indicated
time for RNA isolation and bacterial counting. To measure bacterial
growth, two leaf discs were ground in 100 ml H2O and serial dilutions
were plated on KB medium with appropriate antibiotics. Bacterial col-
ony forming units (cfu) were counted 2 days after incubation at 28ºC.
Each data point is shown as triplicates.
recognition specificity in their interactions with different
variants of Pto resistance protein (Kim et al., 2002). Sur-
prisingly, mutational analysis of AvrPto identified the same
residues essential for the avirulence function in tomato
and the virulence MAMP suppressor activity in Arabidop-
sis. Furthermore, expression of tomato Pto in Arabidopsis
protoplasts partially interfered with the AvrPto suppressor
function in MAMP signaling. It is possible that AvrPto and
AvrPtoB could target Pto-like kinases or Pto-like receptor
kinases involved in MAMP perception in Arabidopsis. Ap-
parently, expression of a small amount of the flg22 recep-
tor FLS2 can interfere with the AvrPto suppressor activity.
To further elucidate the molecular actions of AvrPto and
AvrPtoB in blocking early MAMP signaling, it will be of
great interest to identify more target(s) in Arabidopsis.
AvrPto transgenic plants were generated by Agrobacterium-medi-
ated transformation with the avrPto construct under the control of the
dexamethasone-inducible promoter with an HA epitope tag (McNellis
et al., 1998). To induce AvrPto expression, transgenic plants were
sprayed with 20 mM dexamethasone containing 0.025% silwet L-77 1
day before infiltration with flg22 or bacteria for RT-PCR and disease
assays.
Type III Effector and Reporter Constructs
All effector constructs were made by cloning PCR fragments into
a plant expression vector with an HA or FLAG epitope tag at the C ter-
minus (Asai et al., 2002). avrRpm1 was amplified from DC3000
(avrRpm1); avrB was amplified from DC3000 (avrB); avrRpt2 was am-
plified from P.s. pv. maculicola ES4326 (avrRpt2); avrPto was amplified
from DC3000 and Pst JL1065; avrPtoB, hopAI1, hopPtoD2, hopPtoE,
and hopPtoK were amplified from DC3000; virPphA was amplified
from Psp NPS3121; avrBsT was amplified from Xcv 75-3.
Dynamic Coevolution in Plant-Bacterium
Interactions
Although plants lack the elaborate adaptive immune sys-
tem found in mammals, they have expanded the innate
immune system through the evolution of a large set of pat-
tern-recognition receptors (PRRs; Meyers et al., 2003). It
appears that plant bacterial pathogens evolved numerous
type III effectors that can suppress plant immunity (Alfano
and Collmer, 2004; Mudgett, 2005). To survive, individual
plants further evolved highly specific resistance genes to
counteract specific virulence effectors in gene-for-gene
defense (Dangl and Jones, 2001; Pedley and Martin,
2003; Staskawicz et al., 2001). Thus, dynamic coevolution
in individual plant-bacterium interactions drives the resis-
tance/susceptibility or nonhost/host relationship.
The WRKY46 (At2g46400) promoter was amplified from Arabidopsis
Col-0 genomic DNA and fused with a luciferase reporter gene to gen-
erate WRKY46-LUC (Asai et al., 2002). The reporter-gene construct
FRK1-LUC, the MAPK constructs, MPK3 and MPK6, the active MKK4,
MKK5, and MEKK1 constructs, and the FLS2 construct were reported
previously (Asai et al., 2002). Pto was PCR amplified from pEG202::Pto
(Kim et al., 2002).
Arabidopsis Protoplast Transient Expression and MAPK Assays
Protoplast transient expression and MAPK in-gel and in vitro assays
were carried out as reported previously (Asai et al., 2002) and the de-
tailed procedures are described in Supplemental Experimental Proce-
dures. UBQ10-GUS was always cotransfected with FRK1-LUC as an
internal control, and the promoter activity was presented as LUC/
GUS ratio. Protoplasts were collected 6 hr after transfection for protein
expression, kinase activity, and promoter activity assays. Protoplasts
transfected with plasmid DNA without effectors were used as controls.
DC3000 appears to be a highly evolved virulent patho-
gen in plants and evolved or acquired AvrPto and AvrPtoB
as potent suppressors of MAMP-mediated immune re-
sponses. To survive infection by this sophisticated patho-
gen, tomato but not Arabidopsis evolved the unique Pto
and Prf gene products to recognize AvrPto and AvrPtoB
and trigger specific gene-for-gene defense (Pedley and
Martin, 2003). Analogous evolution also occurred in an un-
known R protein recognizing different part of the AvrPto
protein in tobacco (Shan et al., 2000b). Apparently, AvrPto
can either activate MAPK signaling in Pto-mediated gene-
for-gene defense in tomato (Pedley and Martin, 2004) or
suppress MAPK signaling in MAMP-mediated defense in
Arabidopsis. The manifestation of the distinct virulence
and avirulence actions of AvrPto from Pseudomonas in dif-
ferent plants provides a fascinating example for how type
III effectors and genes involved in plant defense continue
to dynamically evolve in the plant-bacterium battlefield.
RT-PCR Analysis
Total RNA was isolated from leaves or protoplasts after treatment by
using TRIzol Reagent (Invitrogen). Complementary DNA was synthe-
sized from 1 mg of total RNA using 0.1 mg oligo (dT) primer and reverse
transcriptase (Invitrogen). RT-PCR was run for 30 cycles. Real-time
RT-PCR analysis was carried out with an iCycler iQ real-time PCR de-
tection system with iQ SYBR green supermix (BIO-RAD). The expres-
sion of FRK1, At1g51890, At2g17740, or At5g57220 was normalized to
the expression of UBQ10. The primer sequences for different effectors
and RT-PCR are listed in Supplemental Experimental Procedures.
Supplemental Data
Supplemental Data include seven figures, three tables, and Supple-
mental Experimental Procedures and can be found with this article on-
line at http://www.cell.com/cgi/content/full/125/3/563/DC1/.
EXPERIMENTAL PROCEDURES
Plant Growth, Bacterial Inoculation, and Generation
of Transgenic Plants
ACKNOWLEDGMENTS
Wild-type (wt) Col-0 and mutant rps2-101C or rps3-3 Arabidopsis
plants were grown in a growth chamber at 23ºC with a 13 hr photope-
riod for 30 days before bacterial inoculation or protoplast isolation.
Different P. syringae strains were grown overnight at 28ºC in the KB
medium with appropriate antibiotics. Bacteria were pelleted by centri-
We thank F. Ausubel, X. Tang, and J. Zhou for providing different
bacterial strains, J. Mecsas and C. Lesser for the YopJ clone, K.
Chu for the analysis of microarray data, J. Bush for plant management,
G. Tena and S. Yoo for MAPK clones, B. Mueller for the assistance in
operating confocal microscope, R. Patharkar for sharing information,
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 573

and F. Ausubel and members of the Sheen lab for critical reading of the
manuscript. This work was supported by grants from the NSF (DBI
0077692, MCB 0446109) and NIH (R01 GM70567) to J.S., grants
from the Deutsche Forschungsgemeinschaft and KWS Einbeck,
Germany to T.N., and grants from USDA-NRI (2005-00972) and NSF
(IOB 0533970) to G.B.M.
Pseudomonas syringae type III effectors that can suppress pro-
grammed cell death in plants and yeast. Plant J. 37, 554–565.
Janjusevic, R., Abramovitch, R.B., Martin, G.B., and Stebbins, C.E.
(2006). A bacterial inhibitor of host programmed cell death defenses
is an E3 ubiquitin ligase. Science 311, 222–226.
Kang, L., Li, J., Zhao, T., Xiao, F., Tang, X., Thilmony, R., He, S., and
Zhou, J.M. (2003). Interplay of the Arabidopsis nonhost resistance
gene NHO1 with bacterial virulence. Proc. Natl. Acad. Sci. USA 100,
3519–3524.
Received: July 29, 2005
Revised: November 8, 2005
Accepted: February 8, 2006
Published: May 4, 2006
Kim, Y.J., Lin, N.C., and Martin, G.B. (2002). Two distinct Pseudomo-
nas effector proteins interact with the Pto kinase and activate plant
immunity. Cell 109, 589–598.
REFERENCES
Kim, M.G., da Cunha, L., McFall, A.J., Belkhadir, Y., Debroy, S., Dangl,
J.L., and Mackey, D. (2005). Two Pseudomonas syringae type III effec-
tors inhibit RIN4-regulated basal defense in Arabidopsis. Cell 121,
749–759.
Abramovitch, R.B., Kim, Y.J., Chen, S., Dickman, M.B., and Martin,
G.B. (2003). Pseudomonas type III effector AvrPtoB induces plant
disease susceptibility by inhibition of host programmed cell death.
EMBO J. 22, 60–69.
Lee, J., Klessig, D.F., and Nu¨ rnberger, T. (2001). 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.
Alfano, J.R., and Collmer, A. (2004). Type III secretion system effector
proteins: double agents in bacterial disease and plant defense. Annu.
Rev. Phytopathol. 42, 385–414.
Asai, T., Tena, G., Plotnikova, J., Willmann, M.R., Chiu, W.L., Gomez-
Gomez, L., Boller, T., Ausubel, F.M., and Sheen, J. (2002). MAP kinase
signalling cascade in Arabidopsis innate immunity. Nature 415, 977–
983.
Li, X., Lin, H., Zhang, W., Zou, Y., Zhang, J., Tang, X., and Zhou, J.M.
(2005). Flagellin induces innate immunity in nonhost interactions that is
suppressed by Pseudomonas syringae effectors. Proc. Natl. Acad.
Sci. USA 102, 12990–12995.
Ausubel, F.M. (2005). Are innate immune signaling pathways in plants
Ligterink, W., Kroj, T., zur Nieden, U., Hirt, H., and Scheel, D. (1997).
Receptor-mediated activation of a MAP kinase in pathogen defense
of plants. Science 276, 2054–2057.
and animals conserved? Nat. Rev. Immunol. 6, 973–979.
Barton, G.M., and Medzhitov, R. (2003). Toll-like receptor signaling
pathways. Science 300, 1524–1525.
Lin, N., and Martin, G.B. (2005). An avrPto/avrPtoB mutant of Pseudo-
monas syringae pv. tomato DC3000 does not elicit Pto-mediated resis-
tance and is less virulent on tomato. Mol. Plant Microbe Interact. 18,
43–51.
Dangl, J.L., and Jones, J.D. (2001). Plant pathogens and integrated
defence responses to infection. Nature 411, 826–833.
de Torres, M., Sanchez, P., Fernandez-Delmond, I., and Grant, M.
(2003). Expression profiling of the host response to bacterial infection:
the transition from basal to induced defence responses in RPM1-
mediated resistance. Plant J. 33, 665–676.
Lipka, V., Dittgen, J., Bednarek, P., Bhat, R., Wiermer, M., Stein, M.,
Landtag, J., Brandt, W., Rosahl, S., Scheel, D., et al. (2005). Pre-
and postinvasion defenses both contribute to nonhost resistance in
Arabidopsis. Science 310, 1180–1183.
Eastmond, P.J. (2004). Glycerol-insensitive Arabidopsis mutants: gli1
seedlings lack glycerol kinase, accumulate glycerol and are more
resistant to abiotic stress. Plant J. 37, 617–625.
McNellis, T.W., Mudgett, M.B., Li, K., Aoyama, T., Horvath, D., Chua,
N.H., and Staskawicz, B.J. (1998). Glucocorticoid-inducible expres-
sion of a bacterial avirulence gene in transgenic Arabidopsis induces
hypersensitive cell death. Plant J. 14, 247–257.
Espinosa, A., Guo, M., Tam, V.C., Fu, Z.Q., and Alfano, J.R. (2003). The
Pseudomonas syringae type III-secreted protein HopPtoD2 possesses
protein tyrosine phosphatase activity and suppresses programmed
cell death in plants. Mol. Microbiol. 49, 377–387.
Meyers, B.C., Kozik, A., Griego, A., Kuang, H., and Michelmore, R.W.
(2003). Genome-wide analysis of NBS-LRR-encoding genes in Arabi-
dopsis. Plant Cell 15, 809–834.
Fellbrich, G., Romanski, A., Varet, A., Blume, B., Brunner, F., Engel-
hardt, S., Felix, G., Kemmerling, B., Krzymowska, M., and Nu¨ rnberger,
T. (2002). NPP1, a Phytophthora-associated trigger of plant defense in
parsley and Arabidopsis. Plant J. 32, 375–390.
Mudgett, B.M. (2005). New insights to the function of phytopathogenic
bacterial type III effectors in plants. Annu. Rev. Plant Biol. 56, 509–531.
Galan, J.E., and Cossart, P. (2005). Host-pathogen interactions: a di-
versity of themes, a variety of molecular machines. Curr. Opin. Micro-
biol. 8, 1–3.
Mysore, K.S., and Ryu, C.M. (2004). Nonhost resistance: how much do
we know? Trends Plant Sci. 9, 97–104.
Navarro, L., Zipfel, C., Rowland, O., Keller, I., Robatzek, S., Boller, T.,
and Jones, J.D. (2004). The transcriptional innate immune response to
flg22. Interplay and overlap with Avr gene-dependent defense re-
sponses and bacterial pathogenesis. Plant Physiol. 135, 1113–1128.
Gomez-Gomez, L., and Boller, T. (2002). Flagellin perception: a para-
digm for innate immunity. Trends Plant Sci. 7, 251–256.
Hauck, P., Thilmony, R., and He, S.Y. (2003). A Pseudomonas syringae
type III effector suppresses cell wall-based extracellular defense in
susceptible Arabidopsis plants. Proc. Natl. Acad. Sci. USA 100,
8577–8582.
Nu¨ rnberger, T., Brunner, F., Kemmerling, B., and Piater, L. (2004). In-
nate immunity in plants and animals: striking similarities and obvious
differences. Immunol. Rev. 198, 249–266.
He, P., Chintamanani, S., Chen, Z., Zhu, L., Kunkel, B.N., Alfano, J.R.,
Tang, X., and Zhou, J.M. (2004). Activation of a COI1-dependent path-
way in Arabidopsis by Pseudomonas syringae type III effectors and
coronatine. Plant J. 37, 589–602.
Oh, H.S., and Collmer, A. (2005). Basal resistance against bacteria in
Nicotiana benthamiana leaves is accompanied by reduced vascular
staining and suppressed by multiple Pseudomonas syringae type III
secretion system effector proteins. Plant J. 44, 348–359.
Inohara, N., Chamaillard, M., McDonald, C., and Nunez, G. (2005).
NOD-LRR proteins: role in host-microbial interactions and inflamma-
tory disease. Annu. Rev. Biochem. 74, 355–383.
Orth, K., Palmer, L.E., Bao, Z.Q., Stewart, S., Rudolph, A.E., Bliska,
J.B., and Dixon, J.E. (1999). Inhibition of the mitogen-activated protein
kinase kinase superfamily by a Yersinia effector. Science 285, 1920–
1923.
Jamir, Y., Guo, M., Oh, H.S., Petnicki-Ocwieja, T., Chen, S., Tang, X.,
Dickman, M.B., Collmer, A., and Alfano, J.R. (2004). Identification of
574 Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc.

Pedley, K.F., and Martin, G.B. (2003). Molecular basis of Pto-mediated
resistance to bacterial speck disease in tomato. Annu. Rev. Phytopa-
thol. 41, 215–243.
Staskawicz, B.J., Mudgett, M.B., Dangl, J.L., and Galan, J.E. (2001).
Common and contrasting themes of plant and animal diseases. Sci-
ence 292, 2285–2289.
Pedley, K.F., and Martin, G.B. (2004). Identification of MAPKs and their
possible MAPK kinase activators involved in the Pto-mediated de-
fense response of tomato. J. Biol. Chem. 279, 49229–49235.
Tao, Y., Xie, Z., Chen, W., Glazebrook, J., Chang, H.S., Han, B., Zhu,
T., Zou, G., and Katagiri, F. (2003). Quantitative nature of Arabidopsis
responses during compatible and incompatible interactions with the
bacterial pathogen Pseudomonas syringae. Plant Cell 15, 317–330.
Ramonell, K., Berrocal-Lobo, M., Koh, S., Wan, J., Edwards, H., Sta-
cey, G., and Somerville, S. (2005). Loss-of-function mutations in chitin
responsive genes show increased susceptibility to the powdery
mildew pathogen Erysiphe cichoracearum. Plant Physiol. 138, 1027–
1036.
Thordal-Christensen, H. (2003). Fresh insights into processes of non-
host resistance. Curr. Opin. Plant Biol. 6, 351–357.
Zeidler, D., Zahringer, U., Gerber, I., Dubery, I., Hartung, T., Bors, W.,
Hutzler, P., and Durner, J. (2004). Innate immunity in Arabidopsis thali-
ana: lipopolysaccharides activate nitric oxide synthase (NOS) and in-
duce defense genes. Proc. Natl. Acad. Sci. USA 101, 15811–15816.
Shan, L., He, P., Zhou, J.M., and Tang, X. (2000a). A cluster of muta-
tions disrupt the avirulence but not the virulence function of AvrPto.
Mol. Plant Microbe Interact. 13, 592–598.
Shan, L., Thara, V.K., Martin, G.B., Zhou, J.M., and Tang, X. (2000b).
The Pseudomonas AvrPto protein is differentially recognized by to-
mato and tobacco and is localized to the plant plasma membrane.
Plant Cell 12, 2323–2338.
Zhang, S., and Klessig, D.F. (2001). MAPK cascades in plant defense
signaling. Trends Plant Sci. 6, 520–527.
Shiu, S.H., and Bleecker, A.B. (2003). Expansion of the receptor-like
kinase/Pelle gene family and receptor-like proteins in Arabidopsis.
Plant Physiol. 132, 530–543.
Zipfel, C., Robatzek, S., Navarro, L., Oakeley, E.J., Jones, J.D., Felix,
G., and Boller, T. (2004). Bacterial disease resistance in Arabidopsis
through flagellin perception. Nature 428, 764–767.
Cell 125, 563–575, May 5, 2006 ª2006 Elsevier Inc. 575