Blackwell Science, LtdOxford, UKMMIMolecular Microbiology 1365-2958Blackwell Publishing Ltd, 2003492389400Original ArticleJ. R. Bretz et al.HopPtoD2 modulates defence responses
Molecular Microbiology (2003) 49(2), 389 400 doi:10.1046/j.1365-2958.2003.03616.x
A translocated protein tyrosine phosphatase of
Pseudomonas syringae pv. tomato DC3000 modulates
plant defence response to infection
James R. Bretz,1 Norton M. Mock,2 James C. Charity,1 delayed the development of several defence-associ-
Syed Zeyad,1 C. Jacyn Baker2 and ated responses including programmed cell death,
Steven W. Hutcheson1* active oxygen production and transcription of the
1
Department of Cell Biology and Molecular Genetics, pathogenesis-related gene PR1. The results indicate
University of Maryland, College Park, MD 20742,USA. that HopPtoD2 is a translocated effector with protein
2
Beltsville Agricultural Research Center, Agricultural tyrosine phosphatase activity that modulates plant
Research Service, United States Department of defence responses.
Agriculture, Beltsville, MD 20705, USA.
Introduction
Summary
Pseudomonas syringae is the causative agent of leaf
Pseudomonas syringae strains translocate effector blights and related diseases in many economically impor-
proteins into host cells via the hrp-encoded type III tant plant species. Individual P. syringae strains usually
protein secretion system (TTSS) to facilitate patho- only cause disease in a limited set of plants. The host
genesis in susceptible plants. However, the mecha- range of P. syringae strains appears to be controlled, in
nisms by which pathogenesis is favoured by these part, by the ability of the host to mount a cellular defence
effectors are not well understood. Individual strains response (Hutcheson, 1998; Dangl and Jones, 2001). In
express multiple effectors with apparently distinct resistant plants, primary defence responses to P. syringae
activities that are co-ordinately regulated by the alter- strains are usually rapid, commonly detectable within the
native sigma factor HrpL. Genes for several effectors first few hours of the interaction, and frequently culminate
were identified in the P. syringae pv. tomato DC3000 in programmed cell death (PCD) observable in the
genome using a promoter trap assay to identify HrpL- responding cells 6 h after inoculation. When artificially
dependent promoters. In addition to orthologues of high inocula are used, this defence-associated pro-
avrPphE and hrpW, an unusual allele of avrPphD was grammed cell death leads to a localized necrotic response
detected that carried an IS52 insertion. Using this in inoculated tissue known as the hypersensitive response
avrPphD::IS52 allele as a probe, a wild-type allele of (HR). Primary responding cells undergoing PCD release
avrPphD, hopPtoD1, and a chimeric homologue were active oxygen (Baker and Orlandi, 1995) and nitric oxide
identified in the DC3000 genome. This chimeric homo- (Delledonne et al., 1998) that stimulate additional
logue, identified as HopPtoD2 in the annotated responses in adjacent cells (Baker and Orlandi, 1999).
DC3000 genome, consisted of an amino terminal These additional responses can involve enhanced sec-
secretion domain similar to that of AvrPphD fused to ondary metabolism, accumulation of antimicrobial com-
a potential protein tyrosine phosphatase domain. Cul- pounds, and induction of pathogenesis-related proteins,
ture filtrates of strains expressing HopPtoD2 were such as PR1, in the adjacent cells. Mitogen-activated pro-
able to dephosphorylate pNPP and two phospho- tein kinases (MAPKs) have been implicated in the early
tyrosine peptides. HopPtoD2 was shown to be phases of these defence responses (Nurnberger and
translocated into Arabidopsis thaliana cells via the Schell, 2001; Zhang and Klessig, 2001). The combined
D
hrp-encoded TTSS. A DhopPtoD2 mutant of DC3000 activities of these defence responses are thought to limit
D
D
exhibited strongly reduced virulence in Arabidopsis the spread of the pathogen. In contrast, host defence
thaliana. Ectopic expression of hopPtoD2 in P. syrin- responses to P. syringae infection in susceptible plants are
gae Psy61 that lacks a native hopPtoD2 orthologue slow, thereby allowing the infection to progress before
host cells respond.
The induction of the cellular defence responses in resis-
tant plants and pathogenesis in susceptible plants has
Accepted 7 May, 2003. *For correspondence. E-mail sh53@
umail.umd.edu; Tel. [phone]; Fax [phone]. been linked to the activities of the P. syringae hrp patho-
© 2003 Blackwell Publishing Ltd
390 J. R. Bretz et al.
genicity island (PAI) (Hutcheson, 1999; Alfano et al., Pto (Scofield et al., 1996; Tang et al., 1996; Kim et al.,
2000). Within the hrp PAI is a central conserved region 2002) and to suppress defence responses (Abramovitch
that encodes the components of a type III secretion sys- et al., 2003). A family of proteins that includes AvrPphB
tem (TTSS). The primary function of the hrp TTSS has been shown to be cysteine proteases (Shao et al.,
appears to be the translocation of effector proteins into 2002). AvrB, AvrRpm1 and AvrRpt2 have been shown to
the cytosol of host cells through the needle-like HrpA pilus interact with RIN4 which then appears to target the com-
(Jin and He, 2002). A complex regulatory system has plex for degradation (Mackey et al., 2002; 2003; Axtell
been partially characterized in P. syringae strains that and Staskawicz, 2003). The mechanisms by which most
utilizes HrpL, a member of the ECF family of alternative other effectors function in host cells have not been
sigma factors (Xiao et al., 1994), to direct expression of established.
the TTSS as well as most of the known TTSS-dependent Recently we developed a method to identify candidate
effectors during pathogenesis (Xiao and Hutcheson, effectors genes by screening genomic libraries of P. syrin-
1994; Hutcheson et al., 2001; Bretz et al., 2002). In cells gae strains for promoters dependent upon HrpL for
of resistant plants, effectors translocated by the hrp- expression (Hutcheson et al., 2003). During preliminary
encoded TTSS, such as those encoded by avr genes, can trials of this HrpL-dependent promoter trap assay, the P.
be recognized by cytosolic receptors either encoded by syringae pv. tomato DC3000 genome was partially sur-
resistance genes or associated with resistance gene veyed for candidate effector genes. This assay led to the
products to initiate PCD in the responding cell identification of a modular effector with an amino terminal
(Hutcheson, 1998; Dangl and Jones, 2001). In susceptible TTSS-dependent secretion domain similar to that of Avr-
plants, the effectors are thought to facilitate parasitism of PphD and a carboxyl terminal domain that functions as a
the host cells by opening pores in the plasma membrane protein tyrosine phosphatase (PTP). This novel effector
(Lee et al., 2001), or by functioning in an unknown manner appears to have a specific role in DC3000 virulence by
to suppress other types of cellular defence responses suppressing activation of host defence responses.
(Jakobek et al., 1993; Jackson et al., 1999; Chen et al.,
2000; Abramovitch et al., 2003).
Because the effectors translocated into host cells by P. Results
syringae strains appear to control host range, recent
Identification of an unusual avrPphD homologue in P.
efforts have been directed at identifying the effectors pro-
syringae pv. tomato DC3000 by a HrpL-dependent
duced by individual strains. Assays screening for altered
promoter trap assay
host range of transformants carrying a genomic library of
another strain (Staskawicz et al., 1984), surveys of To evaluate the effectiveness of the HrpL-dependent pro-
secreted proteins found in culture filtrates (van Dijk et al., moter trap assay in identifying promoters for effector
1999) and screens for translocated fusions to  AvrRpt2 genes, a partial screen of the P. syringae pv. tomato
(Guttman et al., 2002) have been used previously to DC3000 genome was attempted. Escherichia coli trans-
identify translocated effectors. In the strains in which the formants carrying an arabinose-inducible  hrpL construct
hrp PAI has been sequenced, some putative effectors (pSHL4K) and a DC3000 genomic library fused to the
have been identified by their inclusion in the conserved promoterless  lacZYA cassette of pRG970 were screened
and exchangeable effector loci associated with the hrp for arabinose-dependent Lac phenotypes. In five of the
PAI (Alfano et al., 2000; Charity et al., 2003; Deng et al., seven colonies exhibiting arabinose-dependent Lac+ phe-
2003). Potential effectors have also been identified in the notypes, sequence analysis of the insert revealed HrpL-
P. syringae pv. tomato DC3000 and P. syringae pv. dependent promoter sequences (Xiao and Hutcheson,
phaseolicola B728a genomes by searching for HrpL- 1994). For two of these inserts, the HrpL-dependent pro-
dependent promoter sequences (Fouts et al., 2002) or moter appeared to control the DC3000 orthologues of the
potential type III secretion signals (Guttman et al., 2002; P. syringae pv. syringae (Psy) B728A hrpW and P. syrin-
Petnicki-Ocweija et al., 2002). These results indicate that gae pv. phaseolicola (Pph) 1302 A avrPphE. The three
individual P. syringae strains may express as many as 51 remaining clones carried a strong candidate HrpL-
genes for effector proteins that contribute to the pathoge- dependent promoter sequence that subsequently was
nicity of each strain (Hutcheson, 1998; Dangl and Jones, found to be 5ó to an IS52-inactivated homologue of avrP-
2001; Collmer et al., 2002). Some of these effectors phD that consisted of a 711 codon open reading frame
appear to be widely distributed among P. syringae (ORF) with a 1210 bp IS52-like insertion at codon 169
strains, whereas others are found in only a few strains. (Supplementary material, Figs S1 and S2). This previ-
Recently, biochemical activities for some P. syringae ously unreported avrPphD homologue was located in an
effectors have been identified. AvrPto and AvrPtoB have unannotated portion of the DC3000 genome, and exclud-
been shown to interact with the IRAK-like Ser-Thr kinase, ing the IS52 sequence, exhibited 86% identity (I) and 90%
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
HopPtoD2 modulates defence responses 391
similarity (S) to AvrPphD (Arnold et al., 2001). This locus
was designated here as hopPtoD3.
Identification of a candidate protein tyrosine phosphatase
The identification of an insertionally inactivated ortho-
logue of the widely distributed AvrPphD (Arnold et al.,
2001) was unusual. Using the amino terminus of
hopPtoD3 as a probe, the genome of DC3000 was found
to carry two other homologues of avrPphD. The first,
hopPtoD1, was an apparent wild-type allele of avrPphD
(705 aa) that was 89% I / 91% S to the Pph AvrPphD
(Fouts et al., 2002) (Supplementary material, Fig S1). The
second homologue was equivalent to hopPtoD2 (Petnicki-
Ocweija et al., 2002) and was located 1577 bp 5ó to
hopPtoD3. The amino terminal 142 aa of this homologue
retained 61% I / 70% S to the amino terminal 142 aa of
AvrPphD (Supplementary material, Fig. S1), whereas the
remaining 326 aa carboxyl terminal domain exhibited no
detectable similarity to AvrPphD. However, a protein
tyrosine phosphatase (PTP) domain was identified in the
carboxyl terminal domain by the conserved domain finder
of the BLAST algorithm. The PTP active site signature
sequence [LIVMF]HCxAGxxR[STC][STAG] (Fauman and
Saper, 1996) at position 376 386 contained the critically
Fig. 1. Protein tyrosine phosphatase activity of (A) DC3000 and (B)
spaced cysteine and arginine residues that are essential
Psy61 culture filtrates. DC3000 derivative or Psy61 transformants
for PTP activity (Supplementary material, Fig. S3). In
carrying the indicated plasmids were incubated under hrp-inducing
addition, the general acid motif typical of other PTPs was
conditions to an OD600 of approximately 1.2. Culture supernatant was
collected, concentrated 50-fold and normalized for protein content.
also detected at position 351.
Phosphatase activity was assayed using pNPP as described in the
Experimental procedures. Product accumulation was monitored
spectrophotometrically and the relative rate of PTPase activity calcu-
HopPtoD2 has protein tyrosine phosphatase activity
lated from the slope of the reaction (DA410 min-1) averaged over the
75 min assay period. Reaction was near linear during the entirety of
Whereas translocated PTPs have been identified as
the assay period. One unit of LAR (leukocyte antigen related protein),
TTSS-dependent effectors of Yersinia and Salmonella
a known PTP, was used as a control. Grey bars represent activity of
the enzyme alone. White bars represent the activity observed in the
(Guan and Dixon, 1990; Bliska et al., 1991; Kaniga et al.,
presence of 1 mM sodium orthovanadate. Error bars represent the
1996; Murli et al., 2001), similar effectors had not been
standard deviation obtained in at least three separate experiments.
reported previously for a plant pathogenic bacterium. To
determine if HopPtoD2 has PTP activity, culture filtrates
carrying secreted proteins from wild-type DC3000 and
JB4, a DhopPtoD2 mutant of DC3000, were analysed for were capable of hydrolysing the insulin receptor substrate
their ability to hydrolyse an artificial PTP substrate, paran- at rates two to threefold faster than the control extracts
itrophenyl phosphate (pNPP) (Zhang, 1995). Culture fil- (Table 1). Similar results were obtained using the EGF
trates from DC3000 were able to hydrolyse pNPP at a rate receptor peptide as the substrate (data not shown). The
similar to the activity of LAR, a known PTP (Streuli et al., gain of PTP activity by E. coli transformants specifically
1989). Culture filtrates from JB4 hydrolysed pNPP at a expressing hopPtoD2 combined with the strong conser-
much lower rate (Fig. 1A). Addition of the PTP inhibitor, vation of the critical PTP active site motifs indicate that
sodium orthovanadate (Gamper et al., 1996), eliminated HopPtoD2 is responsible for the PTP activity detected in
detectable phosphatase activity in culture filtrates of DC3000 culture filtrates.
DC3000. To confirm that HopPtoD2 was capable of
hydrolysing a phosphorylated protein substrate, extracts
HopPtoD2 is translocated into host cells by the hrp TTSS
of E. coli transformants expressing HopPtoD2 were sur-
veyed for activity against commercially available phospho- The high similarity of the amino terminus of HopPtoD2 to
tyrosine peptides derived from the insulin receptor or the the amino terminus of AvrPphD is consistent with translo-
EGF receptor. Extracts of cells expressing HopPtoD2 cation of HopPtoD2 into plant cells via the hrp TTSS. In
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
392 J. R. Bretz et al.
Table 1. Protein tyrosine phosphatase activity detected in lysates of
HopPtoD2 modulates host defence responses to infection
E. coli expressing hopPtoD2.
The role of HopPtoD2 in suppressing host defence
Expressed genea PTP activityb
responses was examined. The HR elicited by the
DhopPtoD2 mutant JB4 in Nicotiana tabacum L. leaves
None 88 Ä… 13
HopPtoD2 259 Ä… 4
was indistinguishable from that of DC3000 under all tested
LAR 210 Ä… 11
conditions. In contrast, a phenotype was detected when
HopPtoD2 was ectopically expressed in Psy61, which
a. None, E. coli DH5a (pDSK600); HopPtoD2, E. coli DH5a
(pJBHopPtoD2); LAR, purified leukocyte antigen related protein, a
does not appear to carry a native allele of HopPtoD2.
known PTP.
When N. tabacum leaves were inoculated with the lowest
b. Whole cell extracts of the indicated strains were monitored for PTP
concentration of bacteria capable of producing a visible
activity using residues 1142 1153 of the insulin receptor as the
phosphotyrosine substrate. PTP activity measured as the amount of
HR (106 CFU ml-1), the HR elicited by Psy61
phosphate released per minute (pmole PO42 min-1). The value
(pJBHopPtoD2) was observed 2 3 h after the response
reported represents the mean of duplicate samples and the experi-
elicited by the control Psy61 strain or a Psy61 transfor-
ment was repeated three times with similar results.
mant [Psy61 (pJBHopPtoD2DC)] expressing the amino
terminal 321 aa of HopPtoD2 from the same promoter as
P. syringae, the secretion signal required for hrp-depen-
the full length construct. The HopPtoD2DC fragment car-
dent TTSS is carried by the first 50 amino acids of the
ries the secretion signals for TTSS as shown above but
secreted effector whereas the effector activity localizes to
lacks PTP activity (Fig. 1B). Since ectopic expression of
the carboxyl terminal domain (Mudgett and Staskawicz,
the truncated HopPtoD2DC construct did not cause a
1999; Guttman and Greenberg, 2001). The carboxyl ter-
delay in the plant response, it is unlikely that the expres-
minal effector domain of AvrRpt2 has been shown to func-
sion system for HopPtoD2 was impeding translocation of
tion independently as a reporter for translocation into host
other effectors.
cells (Mudgett and Staskawicz, 1999; Guttman et al.,
An increase in active oxygen production commonly
2002). Fusion proteins carrying the effector domain of
accompanies the HR of tobacco cells and is one of the
AvrRpt2 elicit the HR in the RPS2 ecotypes of A. thaliana
earliest known responses of responding plant cells (Baker
after translocation into plant cells via the hrp TTSS. When
and Orlandi, 1995). Consistent with its HR phenotype in
DC3000 expressing a fusion between the amino terminal
leaves, inactivation of hopPtoD2 in DC3000 had little, if
171 aa of HopPtoD2 and the óAvrRpt2 effector domain
any, effect on active oxygen production from inoculated N.
was tested using this assay, the strain carrying the
tabacum cells (data not shown). A delayed active oxygen
HopPtoD2ó:óAvrRpt2 fusion elicited the HR in the reactive
response, however, was detected when cells were inocu-
RPS2 line of A. thaliana but not in the non-responsive
rps2 derivative (Table 2). Expression of the
HopPtoD2ó:óAvrRpt2 fusion in A9, a DhrpA mutant of Table 2. Translocation of HopPtoD2':  AvrRpt2 fusions into plant cells.
DC3000 that lacks a functional TTSS (Wei et al., 2000),
Responsea
produced a null phenotype. These results indicate that
HopPtoD2 is translocated into plant cells via the hrp
Strain Expressed geneb RPS2c rps2d
TTSS.
DC3000 Nonee Df D
avrRpt2 HRg D
hopPtoD2 D D
DC3000 DhopPtoD2 mutants exhibit reduced virulence in
hopPtoD2ó:óavrRpt2 HR D
A. thaliana A9h None null null
hopPtoD2ó:óavrRpt2 null null
To evaluate whether HopPtoD2 functions during patho-
a. Each strain was inoculated into individual leaves of the same plants
genesis, virulence of DC3000 and JB4 were compared in
by syringe infiltration and monitored for induction of the HR or devel-
the susceptible A. thaliana ecotype Columbia (Whalen
opment of disease. A total of nine leaves were infiltrated with each
et al., 1991). Although typical disease symptoms strain. All inoculated leaves responded in the indicated manner.
b. Indicated gene or construct expressed from lacUV5 promoter of
appeared 24 48 h post infection for both strains, popula-
pDSK600.
tions of JB4 were reduced relative to the wild-type
c. A. thaliana ecotype Columbia; recognizes the  AvrRpt2 peptide and
DC3000. In experiments in which initial populations were is resistant to P. syringae strains expressing AvrRpt2.
d. Susceptible A. thaliana derivative of ecotype Columbia that does
indistinguishable, populations of the DhopPtoD2 mutant
not recognize  AvrRpt2 peptide.
were nearly two orders of magnitude lower than those of
e. The unmodified vector pDSK600 alone.
DC3000 by 72 h after innoculation (Fig. 2) and popula- f. Classic disease (D) symptoms for DC3000 in A. thaliana developed
after 48 h.
tions of JB4 remained below those of DC3000 for the
g. Tissue collapse and subsequent necrosis typical of the HR
subsequent 48 h. No differences in growth of the two
observed 12 24 h post inoculation.
strains were detected in broth culture (data not shown). h. DC3000 DhrpA mutant lacking a functional TTSS.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
HopPtoD2 modulates defence responses 393
observed during pathogenesis by DC3000 in A. thaliana
(Fig. 3). The DhopPtoD2 mutant JB4 elicited a twofold
higher level of GUS expression. Complementary re-
sponses were observed in the tissue inoculated with the
Psy61 transformants ectopically expressing hopPtoD2.
PR1 expression was induced by the control Psy61 strain,
but ectopic expression of HopPtoD2 caused a 65%
reduction in expression of the PR1-GUS fusion. Pss61
derivatives expressing HopPtoD2DC elicited a response
indistinguishable from that elicited by Psy61. Thus, during
interactions with both resistant (N. tabacum) and suscep-
tible plants (A. thaliana), HopPtoD2 is associated with
suppression of plant defence responses.
Fig. 2. HopPtoD2 facilitates P. syringae DC3000 growth during patho-
genesis in A. thaliana. Arabidopsis thaliana ecotype Columbia (Col-
0) leaves were infiltrated with 105 CFU ml-1 DC3000 or JB4. Bacterial
Distribution of HopPtoD2 among P. syringae strains
populations were monitored over 5 days.
To determine what hopPtoD alleles are carried by other
P. syringae strains, P. syringae strains of diverse host
lated with Psy61 (pJBHopPtoD2). Plant cells ordinarily
ranges were surveyed for the presence of hopPtoD alle-
express several peroxidases and other enzymes that
les. Primer pairs were designed to amplify regions of DNA
degrade hydrogen pyroxide (Baker et al., 2002) (cells
specific to each hopPtoD allele. Similar to a previous
only; Fig. 3); thereby requiring the addition of basal levels
report (Arnold et al., 2001), a fragment indicative of the
of hydrogen peroxide in the assay medium to overcome
presence of hopPtoD1 could be amplified from 32 of the
this antioxidant activity. Hydrogen peroxide production
44 strains surveyed (71%). Indicative fragments for
from N. tabacum cells inoculated with Psy61 (pDSK600)
hopPtoD2 or hopPtoD3 could only be amplified from
could be detected between 180 and 240 min after inocu-
DC3000, P. syringae pv. maculicola 10 and P. syringae pv.
lation of the culture as indicated by the increase in hydro-
tomato 2844. Similar results were obtained by hybridiza-
gen peroxide levels in the medium relative to the  cells
tion using a probe unique to the PTP domain of hopPtoD2.
only control. Thereafter, the basal rate of hydrogen per-
Only the three strains that the PCR studies indicated
oxide degradation was restored. Hydrogen peroxide pro-
carried the hopPtoD2 and hopPtoD3 alleles hybridized to
duction from N. tabacum cells inoculated with Psy61
the probe (Supplementary material; Table S1). Thus,
(pJBHopPtoD2DC) initiated at 210 min but was also com-
although hopPtoD2 has a significant role in the virulence
plete by 240 min. The hydrogen peroxide production elic-
of DC3000, it appears that only a few strains carry an
ited by Psy61 (pJBHopPtoD2), however, was first
orthologue to hopPtoD2.
detected at 240 min and continued until 300 min. Thus,
the active oxygen response elicited by Psy61
(pJBHopPtoD2) occurred at least one hour later than that
elicited by Psy61 transformants not expressing
HopPtoD2. At least a 3 hour delay was observed in older
cultures of N. tabacum cells (data not shown). Thus, trans-
location of HopPtoD2 delays the active oxygen response
of tobacco cell consistent with the observed delay in HR-
associated PCD.
To determine if HopPtoD2 modulates other plant
responses, expression of PR1 was monitored during
DC3000 pathogenesis and during elicitation of the HR by
Psy61. PR1 is a pathogenesis-related gene associated
with multiple defence responses (Uknes et al., 1992) and
is induced during pathogen infection through a MAPK- Fig. 3. Ectopic expression of HopPtoD2 in Psy61 delays the active
oxygen response of cultured tobacco cells. Suspensions of cultured
linked pathway (Xing et al., 2001). To estimate PR1
tobacco cells 2 days after transfer were inoculated with 5 Ä„ 107
activity, expression of a PR1-GUS fusion created in A.
CFU ml-1 of the indicated strains and production of H2O2 was moni-
thaliana (Shapiro and Zhang, 2001) was monitored in tored using luminol. Data reported is the mean of four concurrent
assays. The following strains were tested: filled diamond,
tissue inoculated with DC3000, JB4, Psy61 (pDSK600),
Psy61(pDSK600); filled box, Psy61(pJBHopPtoD2); filled triangles,
Psy61 (JBHopPtoD2) and Psy61 (pJBHopPtoD2DC).
Psy61 (pJBHopPtoD2DC); open boxes, tobacco cells only. Similar
Modest levels of PR1-driven GUS expression were results were obtained in at least three separate experiments.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
394 J. R. Bretz et al.
HopPtoD2 acquired the ability to degrade PTP-specific
substrates. The presence of diagnostic structural motifs,
the ability to dephosphorylate pNPP, and the sensitivity of
the phosphatase activity to orthovanadate are considered
to be defining features of a PTP (Fauman and Saper,
1996). Among the four general classes of PTPs that have
been characterized, HopPtoD2 seems to be most similar
to the non-receptor-like PTP family that includes the Yers-
inia YopH. The PTP active site domain was nearly identical
to that of other non-receptor-like PTPs and HopPtoD2
contained no obvious P-loop typical of dual-specificity
PTPs, such as the VH1 PTP.
Fig. 4. Expression of HopPtoD2 suppresses PR1 expression. The
Several TTSS-dependent effectors have been shown to
indicated strains were infiltrated into leaves of A. thaliana Col0/PR1-
contribute to the pathogenic fitness of P. syringae strains.
GUS. GUS activity was assayed two days post inoculation. The bars
represent the mean units of GUS activity detected in five leaf discs For example, the effectors AvrE and AvrRpm1 are
corrected for the activity detected in water-inoculated tissue.
required for full virulence of the source P. syringae strain
(Lorang et al., 1994; Ritter and Dangl, 1995; Reuber and
Ausubel, 1996). AvrPphC, AvrPphF and VirPphA
Discussion
enhanced the virulence of a P. syringae pv. phaseolicola
Pathogenesis by P. syringae strains is dependent upon strain by suppressing R-gene-dependent PCD (Jackson
translocated effectors that facilitate parasitism of the host et al., 1999; Tsiamis et al., 2000). Similarly, the broadly
cells and suppress innate immunity responses (Dangl conserved AvrPtoB, a homologue of VirPphA (Jackson
and Jones, 2001; Collmer et al., 2002). Using a regulated et al., 1999), inhibited the Pto-dependent PCD elicited by
HrpL-dependent promoter trap assay to survey the P. AvrPto in N. benthamiana (Abramovitch et al., 2003). Like-
syringae DC3000 genome for effector genes, several wise, HopPtoD2 was also found to contribute to patho-
homologues of avrPphD were identified. One of the genic fitness of DC3000 in susceptible hosts. Populations
homologues, HopPtoD2, was shown to be a TTSS- of a DC3000 DhopPtoD2 mutant were typically reduced
dependent effector with a carboxyl terminal PTP domain. by 97% relative to the parent strain after three days
Like other TTSS-dependent effectors of P. syringae growth. This is among the largest reported reductions in
(Mudgett and Staskawicz, 1999; Guttman and Green- virulence of a P. syringae strain that can be attributed to
berg, 2001; Petnicki-Ocweija et al., 2002) and several inactivation of a single TTSS-dependent effector. Thus, a
mammalian pathogens (e.g. Sory et al., 1995; Kaniga translocated protein tyrosine phosphatase can be added
et al., 1996), HopPtoD2 was modular. The amino termi- to the growing list of TTSS-dependent effectors that are
nus of HopPtoD2 retained structural features typical of shared in common between plant and mammalian patho-
TTSS-dependent effectors, and consistent with the detec- gens and are required for virulence (Staskawicz et al.,
tion of HopPtoD2 in culture filtrates of DC3000 (Petnicki- 2001). Analogues, and in some cases, partial homologues
Ocweija et al., 2002), HopPtoD2 was demonstrated to be of the Yersinia YopJ (Orth et al., 2000), YopT (Shao et al.,
translocated into plant cells by the hrp TTSS. The car- 2002) and YopH (Bliska et al., 1991; this work) have now
boxyl-terminus of HopPtoD2, in contrast, included charac- been detected in plant pathogenic bacteria.
teristic motifs indicative of PTP activity. This region Unlike several P. syringae effectors that have been
included a general acid motif positioned 24 residues from shown to be epistatic to other effectors (Jackson et al.,
a consensus PTP active site domain in which the critical 1999; Abramovitch et al., 2003), HopPtoD2 appears to act
residues are conserved. As the 321 aa amino terminal more subtly. Ectopic expression of hopPtoD2 delayed
domain was ineffective in eliciting or affecting plant plant defence responses but did not fully suppress these
defence responses, the carboxyl terminal domain of responses. The RPS2-dependent recognition of AvrRpt2
HopPtoD2 appears to be responsible for the observed did not appear to be altered by the co-expression of
activities of this effector. hopPtoD2 resident in the wild-type DC3000 genome. Vir-
Consistent with the retention of diagnostic PTP struc- ulence enhancement caused by the effectors translocated
tural motifs, HopPtoD2 was shown to be a PTP. Culture by the hrp-encoded TTSS has been attributed to interfer-
filtrates of DC3000 expressing HopPtoD2 dephosphory- ence with effector receptor interactions, reduced detec-
lated three PTP-specific substrates. This phosphatase tion by host receptors, or suppression of the signal
activity was sensitive to orthovanadate and was substan- transduction pathway(s) leading to a defence response
tially reduced in culture filtrates obtained from a DC3000 (Tsiamis et al., 2000). HopPtoD2 most likely falls into the
DhopPtoD2 mutant. E. coli and Psy61 strains expressing later category.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
HopPtoD2 modulates defence responses 395
Plant cells have multiple signal transduction pathways to a growing list of virulence-enhancing effectors produced
that can be differentially activated in response to distinct by plant pathogenic bacteria. Virulence-enhancing effec-
elicitors and may function differently in distinct plant spe- tors provide a molecular explanation for the original obser-
cies (Cardinale et al., 2000). For several mammalian vation that hrp genes are required for pathogenicity of P.
pathogens that utilize TTSS in pathogenesis, translocated syringae strains (Lindgren et al., 1986). These virulence-
PTPs, such as the Yersinia YopH (Bliska et al., 1991) and enhancing effectors are likely to be strain-specific by affect-
the Salmonella SptP (Kaniga et al., 1996), affect signal ing the activity of one or a few effectors but could also be
transduction pathways (Black and Bliska, 1997; DeVinney host species-specific like other effectors, such as avr
et al., 2000; Cornelis, 2002). One possible target for the genes. With the identification of biochemical activities
PTP activity of HopPtoD2 is one or more of the MAPK- associated with specific effectors, such as HopPtoD2, the
dependent signal transduction pathways controlling mechanism by which these effectors function to suppress
defence responses (Nurnberger and Schell, 2001; Zhang at least a portion of the innate immune response of sus-
and Klessig, 2001). Upwards of 20 distinct MAPKs have ceptible plants can now be explored.
been identified in the Arabidopsis genome (Jonak et al.,
2001) whose roles in signal transduction are only begin-
Experimental procedures
ning to be established (Innes, 2001; Nurnberger and
Schell, 2001). For example, HR-associated programmed
Bacterial strains and plasmids
cell death (Yang et al., 2001), active oxygen production
Strains used in this study were: E. coli DH5a (Invitrogen,
(Ren et al., 2002), and induction of secondary metabolism
Carlsbad, CA), E. coli SLR400, Dlac Dara (a gift of S. Benson,
and pathogenesis-related proteins, such as PR1 (Xing
University of Maryland), P. syringae pv. tomato DC3000
et al., 2001), have been linked to MAPK signal cascades.
(Whalen et al., 1991), P. syringae pv. tomato A9, DhrpA (Wei
Thus, the apparent HopPtoD2-mediated suppression of
et al., 2000) and P. syringae pv. syringae Psy61 (Huang
the HR-associated programmed cell death, active oxygen
et al., 1988). Other P. syringae strains reported in the Sup-
production, and PR1 expression are consistent with inter-
plementary material are described in Charity et al. (2003).
ference with one or more signal transduction pathways, Pseudomonas syringae strains were routinely grown at 25"C
in KB broth or in M63 minimal salts medium containing 1 mM
perhaps linked to MAPKós. Experiments are underway to
MgSO4 and 0.2% fructose. E. coli strains were grown at 37"C
test this hypothesis, but other mechanisms are possible
in KB media. Antibiotics were added as indicated to the
as well. Irrespective of the mechanism by which
media at the following concentrations [mg ml-1]: kanamycin
HopPtoD2 acts, the inability to suppress some host
(Kan), 50; spectinomycin (Spc), 100; rifampicin (Rif), 200;
defence responses might account for the reduced viru-
and ampicillin (Amp), 100.
lence of the DhopPtoD2 mutant.
Whereas HopPtoD2 is important for DC3000 virulence,
General DNA manipulations
not all strains of P. syringae carry an orthologue of
HopPtoD2. Although many strains appear to have alleles
Genomic DNA was extracted using the cetyltrimethylammo-
of hopPtoD1 (Arnold et al., 2001), a survey of 44 P. syrin-
nium bromide (CTAB, Sigma) method (Ausubel et al., 1987).
gae strains of diverse host ranges revealed that only two Plasmid DNA isolations and gel extractions were performed
using kits manufactured by Bio-Rad (Hercules, CA). Restric-
other strains carried an allele of DC3000 hopPtoD2. The
tion enzymes and related reagents were purchased from
strains that lack an apparent hopPtoD orthologue are fully
Invitrogen (Carlsbad, CA) and used according to the manu-
virulent on their respective hosts, indicating that that PTP
facturer's recommendations. Ligations were performed using
activity is not essential to the virulence of all P. syringae
T4 DNA Ligase (New England Biolabs, Beverly, MA). Plasmid
strains or that other effectors have a similar activity. Inter-
DNA was transformed into electrocompetent cells as
estingly, all three strains shown to carry the hopPtoD2
described previously (Bretz et al., 2002). Polymerase chain
allele also appear to contain the hopPtoD3 allele as well. reactions (PCR) were performed using a Hybaid PCRSprint
Thermal Cycler and employed either Taq (Invitrogen) or
This suggests that these genes may be part of a mobile
ProofPro (Continental Laboratory Products, San Diego, CA)
gene cassette like those reported for some effector genes
polymerases.
of P. syringae strains (Jackson et al., 2000; Charity et al.,
2003).
In summary, this report and the companion report
DNA sequencing and analysis
(Espinosa et al., 2003) demonstrate that HopPtoD2 is a
Plasmid DNA or gel-purified PCR products were sequenced
translocated PTP that suppresses host defence re-
at the University of Maryland Biotechnology Institute. Raw
sponses. The observed delays in programmed cell death,
sequence data was assembled using MACDNASIS PRO
active oxygen production, and PR1 expression coupled
v3.0 and analysed using BLAST algorithms (Altschul et al.,
with reduced virulence of the DhopPtoD2 mutant are con-
1997). Sequences were aligned using CLUSTALW (http://
sistent with this hypothesis. Thus, HopPtoD2 can be added www.searchlauncher.bcm.tmc.edu). Preliminary sequence
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
396 J. R. Bretz et al.
data for P. syringae pv. tomato DC3000 was obtained from at room temperature in pNPP assay buffer, 20 mM pNPP, and
The Institute for Genomic Research website at http:// where indicated, 1 mM sodium orthovanadate, as described
www.tigr.org. previously (Gamper et al., 1996). The hydrolysis of pNPP was
detected by the increase in absorbance at 410 nm. To mea-
sure tyrosine-specific dephosphorylation of phosphorylated
HrpL-dependent promoter trap assay
peptides, whole cell extracts of E. coli strains expressing
pDSK600 or pJBHopPtoD2 were assayed for activity using a
A 986 bp fragment carrying a promoterless  hrpL cassette
commercial PTP Assay Kit according to the manufacturer s
was ligated into pMPM-K6 (Mayer, 1995) to create pSHL4K.
instructions to monitor the release of phosphate (Sigma-
DC3000 genomic DNA was digested with Sau3A, fraction-
Aldrich, St Louis, MO). For both assays, lymphocyte antigen-
ated into 2 4 kb fragments, and ligated into BamHI digested
related protein (LAR) PTP was used as a control.
pRG970. The genomic library was electroporated into E. coli
SLR400 (pSHL4K) and plated onto MacConkey agar contain-
ing 1% lactose and 0.02% arabinose. Following overnight
AvrRpt2-linked translocation assay
incubation at 37"C, Lac+ colonies were purified on KB and
screened for arabinose-dependent Lac phenotypes on Mac- A HopPtoD2ó:óAvrRpt2 fusion was created using the proce-
Conkey lactose agar. Fragments from those colonies exhib- dures of Guttman and Greenberg (2001). The region encod-
ing for the first 171 codons of hopPtoD2 was amplified by
iting an arabinose-dependent phenotype were amplified from
PCR using the primers DC37 and DC598R (GATGCTC
pRG970 using primers 431 8 (5ó-ACGCCAGGGTTTTCC
TTCACCCTAAGAGGACACGATTCATACC), while  avrRpt2
CAGTCA-3ó) and 431 9 (5ó-ATTGCCCGGCTTTCTTGTAA
was amplified using primers Rpt409 (GATGCTCTTCAGG
CG-3ó) and the nucleotide sequence was obtained.
GAAGCACGAGACGGGCGGT) and Rpt1028R (CCCAAG
CTTTAGGGACCAAAAAGCCAG AC). The amplified frag-
ments were digested with SapI, ligated and desired fusion
Construction of pJBHopPtoD2 and pJBHopPtoD2DC
amplified using primers DC37 and Rpt1028R. The resulting
hopPtoD2 was amplified via PCR using primers DC37
fusion was cloned as an EcoRI/HindIII fragment into
(CGGAATTCCCAATGCCTTTCGTCAGCC) and DC1522R
pDSK600 to create pJBHopPtoD2ó:óAvrRpt2. The fusion was
(CCCAAGCTTAGCGCGAGAAACACTAAAGG) and cloned
confirmed by sequencing. The positive control plasmid,
into pDSK600 (Murillo et al., 1994) as an EcoRI/HindIII frag-
pJBAvrRpt2, carrying the full length avrRpt2 was created
ment. The region encoding for the first 321 codons of
using primers Rpt131 (CGGAATTCAACCACCAACGGAC
hopPtoD2 was amplified by PCR using the primers DC37 and
GACTTA) and Rpt1028R. Arabidopsis thaliana RPS2 or A.
DC1048R and ligated into pDSK600 as an EcoRI/HindIII
thaliana rps2 (A gift of B. Staskawicz) leaves were inoculated
fragment to create pJBHopPtoD2DC.
in triplicate with 108 CFU ml-1 of the indicated strains using
syringe-mediated infiltration and monitored for development
of disease symptoms and/or HR production. The experiment
Plant assays
was repeated three times with identical results.
Overnight cultures grown at 25"C were harvested, washed
and diluted in sterile distilled water. To assay for the HR, N.
Construction of DhopPtoD2 mutant JB4
tabacum L. cultivar Samsun leaves were infiltrated in parallel
with bacterial suspensions of 106-109 CFU ml-1 using a
The regions flanking the hopPtoD2 locus of DC3000 were
syringe and incubated at 25"C. Infiltrated leaf panels were
sequentially cloned into pBlueScriptSK + (Strategene; La
scored hourly for water soaking/tissue collapse beginning 2 h
Jolla, CA). The primers DC974 (CGGAATTCGCAGCCT
after inoculation (Bretz et al., 2002). Virulence of DC3000
CATTTCCAACATC) and DC2037R (TCGCCCCGGGGG
was determined in A. thaliana ecotype Columbia (Col-0)
CTGACGAAAGGCATTGG) were used to amplify the
leaves infiltrated with 105 cells ml-1. Bacterial populations
upstream region, and the primers DC3497 (TCGCCCCGG
were monitored using the leaf disk assay (Bertoni and Mills,
GCCCAGCCCTTTAGTGTTTCT) and DC4631R (GCTCTA
1987).
GAGATTGCTGCCCATACACTGC) were employed to amplify
the downstream region. A kan cassette was amplified using
primers K12 (TCGCCCCGGGACA CATCTCAACCAT
Phosphatase assays
CATCG) and K1140R (TCGCCCCGGGCGCCACGGTTGAT
GAGAG) from EZTN < Kan2 > (Epicentre Technologies, Mad-
The indicated P. syringae strains were grown in KB medium
ison, WI) and ligated to the two flanking regions as a XmaI
overnight. Cells were diluted into fresh KB medium and
fragment. The resulting plasmid, pJBDhopPtoD2, was trans-
grown to an OD600 of 1.0. Cells were harvested, suspended
formed into DC3000 and kanamycin resistant mutants were
in M63 medium containing fructose and incubated at 25"C
created by marker exchange. Insertional replacement of
with shaking for an additional 4 h. The cells were removed
hopPtoD2 in JB4 was verified by PCR employing primers
by centrifugation and the supernatant concentrated 50-fold
DC974 and DC4631R.
using Millipore Ultra-free centrifugal filter devices with 10 kDa
exclusion limits. Protein concentration in culture filtrates was
determined using Pierce BCA Protein Assay kit (Pierce,
Active oxygen assay
Rockford, IL) and samples equivalent to 100 mg total protein
were used in the pNPP assay. The PTP activity was assayed Tobacco suspension cells (N. tabacum cv. Hicks) were pre-
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
HopPtoD2 modulates defence responses 397
pared (Baker et al., 2002) and resuspended in 25 ml assay Fig. S1. Pseudomonas syringae pv. tomato DC3000 AvrP-
buffer (0.5 mM MES, pH 6.0) to a final concentration of phD alleles. The deduced protein sequence from the uninter-
0.5 mg f.w. ml-1. Exogenous H2O2 was added to a final con- rupted ORF of hopPtoD3, the apparently wid-type allele
centration of 50 mM to overcome the antioxidant activities of HopPtoD1, and the amino terminal domain of HopPtoD2
the suspension cultured cells. Overnight cultures of P. syrin- were aligned against the original AvrPphD sequence from P.
gae grown at 25"C were harvested, washed with sterile dH2O syringae pv. phaseolicola. The carboxyl-terminal domain of
and added to the suspension cells to a final OD600 of 0.05. HopPtoD2 contains the PTP active site and does not align
Pseudomonas syringae-inoculated cell cultures were incu- with any of the other DC3000 AvrPphD alleles. The IS52
bated at 25"C and 0.4 ml samples were assayed in triplicate insertion site in HopPtoD3 is marked by  <> . Proteins were
every 30 min to determine hydrogen peroxide levels. Hydro- aligned using CLUSTALW. Identical residues are framed in
gen peroxide levels were determined using a Berthold 953 black while similar residues are marked in grey.
Luminometer (Bad Wildbad, Germany) as described previ- Fig. S2. The IS52-like element of hopPtoD3. The nucleotide
ously (Baker et al., 2002). Each reaction mixture contained sequence of the IS52-like insertion element from hopPtoD3
0.72 U peroxidase and 77.6 mM luminol. was aligned against an IS52 insertion element from P. syrin-
gae pv. savastanoi and the JR1 insertion element from
Pseudomonas sp. JR1 using CLUSTALW. The largest ORF in
PR1 gene expression assay the IS52-like element from hopPtoD3 encodes a 330 residue
putative transposase that is 91% identical to the JR1 trans-
Arabidopsis thaliana Col-O:PR1/GUS leaves were infiltrated
posase (data not shown). Identical residues are marked in
with 1 Ä„ 106 CFU ml-1 of the indicated strain and leaf samples
black. The direct repeat sequences from the insertion site in
were removed after 48 h (Shapiro and Zhang, 2001). b-
the DC3000 genome are marked by asterisks.
Glucuronidase (GUS) activity was assayed using the fluoro-
Fig. S3. Conservation of the protein tyrosine phosphatase
metric substrate 4-methylumbelliferyl-b-D-glucuronide (ICN
signature sequence in HopPtoD2. The HopPtoD2 apparent
Biochemicals) and a TKO 100 Fluorometer (Hoefer Scientific)
active site was aligned with the active site of other PTPs
and quantified using a standard curve.
(Fauman and Saper, 1996) using CLUSTALW. The catalytically
required cysteine and arginine residues are marked by *.
Identical residues are framed in black while similar residues
Detection of hopPtoD alleles in P. syringae strains
are marked in grey.
Table S1. Distribution of avrPphD homologues among
The genomes of P. syringae strains were initially surveyed for
Pseudomonas syringae strains.
presence of each of the three identified hopPtoD homologues
by intact cell PCR. The ability to amplify an 867 bp fragment
from the 3ó end of hopPtoD1 using primers DC1183
References
(CGGCAGCATCGGGGTCAG) and DC2050R (CACATGG
CGCAGGAACTCG) was used to indicate the presence of
Abramovitch, R., Kim, Y.-J., Chen, S., Dickman, M., and
hopPtoD1. A 763 bp fragment amplified by primers DC285
Martin, G. (2003) Pseudomonas type III effector AvrPtoB
(CATCACTGGTACGCAACGCT) and DC1048R (TTCTG
induces plant disease susceptibility by inhibition of host
TCGTTTGAGGATTACTT) was used to indicate the presence
programmed cell death. EMBO J 22: 60 69.
of hopPtoD2. Amplification of a 630 bp fragment spanning
Alfano, J.R., Charkowski, A.O., Deng, W.-L., Badel, J.L.,
the region between the IS52-like element and the 3ó region
Petnicki-Ocwieja, T., van Dijk, K., and Collmer, A. (2000)
of hopPtoD3 using primers DC971 (TACCGTGGG
The Pseudomonas syringae Hrp pathogenicity island has
CTGGAAAAGAA) and DC929R (AACGGGATATGGATGTC
a tripartite mosaic structure composed of a cluster of type
TGC) was indicative of the presence of hopPtoD3. To verify
III secretion genes bound by exchangeable effector and
the presence of the hopPtoD2, the 763 bp fragment amplified
conserved effector loci that contribute to parasitic fitness
by primers DC285 and DC1048R was labelled with [g-32P]-
and pathogenicity in plants. Proc Natl Acad Sci USA 97:
dCTP to generate a probe specific to hopPtoD2. Strains were
4856 4861.
surveyed for homologues by colony hybridization (Ausubel
Altschul, S., Madden, T., Schäffer, A., Zhang, J., Zhang, Z.,
et al., 1987).
Miller, W., and Lipman, D. (1997) Gapped BLAST AND PSI-
BLAST: a new generation of protein database search pro-
grams. Nucleic Acids Res 25: 3389 3402.
Acknowledgements
Arnold, D., Gibbon, M., Jackson, R., Wood, J., Brown, J.,
Mansfield, J., Taylor, J., and Vivian, A. (2001) Molecular
This work was supported by grants from the National Science
characterization of avrPphD, a widely distributed gene from
Foundation (MCB 9729524 and MCB0215417). We thank M.
Pseudomonas syringae pv. phaseolicola involved in non-
Howard and L. Losada for their critical comments on the
host recognition by pea (Pisum sativum). Physiol Mol Plant
manuscript.
Pathol 58: 55 62.
Ausubel, F.M., Brent, R., Kingston, R., Moore, D.D.,
Seidman, J.G., Smith, J.A., and Struhl, K. (1987) Current
Supplementary material
Protocols in Molecular Biology. New York: John Wiley and
The following material is available from: Sons.
http://www.blackwellpublishing.com/products/journals/ Axtell, M., and Staskawicz, B. (2003) Initiation of RPS2-
suppmat/mmi/mmi3616/mmi3616.sm.htm specified disease resistance in Arabidopsis is coupled to
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
398 J. R. Bretz et al.
the AvrRpt2-directed elimination of RIN4. Cell 112: 369 van Dijk, K., Fouts, D.E., Rehm, A.H., Hill, A.R., Collmer, A.,
377. and Alfano, J.R. (1999) The Avr (Effector) proteins HrmA
Baker, J.C., and Orlandi, E.W. (1995) Active oxygen in plant (HopPsyA) and AvrPto are secreted in culture from
pathogenesis. Annu Rev Phytopathol 33: 299 321. Pseudomonas syringae pathovars via the Hrp (Type III)
Baker, C., and Orlandi, E. (1999) Active oxygen and patho- protein secretion system in a temperature and pH-sensitive
genesis in plants. In Plant Microbe Interactions, Vol. 5. manner. J Bacteriol 181: 4790 4797.
Keen, N., and Stacey, G., (eds). New York: Chapman & Espinosa, A., Guo, M., Tam, V.C., Fu, Z.-G., and Alfano, J.R.
Hall, pp. 81 119. (2003) The Pseudomonas syringae type III-secreted pro-
Baker, J., O Neill, N., Deahl, K., and Lydon, J. (2002) tein HopPtoD2 possesses protein tyrosine phosphatase
Continuous production of extracellular antioxidants in sus- activity and suppresses programmed cell death in plants.
pension cells attenuates the oxidative burst detected in Mol Microbiol doi: 10.1046/j.1365-2958.2003.03588.x
plant microbe interactions. Plant Physiol Biochem 40: Fauman, E., and Saper, M. (1996) Structure and function of
641 644. the protein tyrosine phosphatases. Trends Biochem Sci 21:
Bertoni, G., and Mills, D. (1987) A simple method to monitor 413 417.
growth of bacterial populations in leaf tissue. Phytopathol Fouts, D., Abramovitch, R., Alfano, J., Baldo, A., Buell, C.,
77: 832 835. Cartinhour, S., et al. (2002) Genomewide identification of
Black, D., and Bliska, J. (1997) Identification of p130Cas as a Pseudomonas syringae pv tomato DC3000 promoters con-
substrate of Yersinia YopH (Yop51), a bacterial protein trolled by the HrpL alternative sigma factor. Proc Natl Acad
tyrosine phosphatase that translocates into mammalian Sci USA 99: 2275 2280.
cells and targets focal adhesions. EMBO J 16: 2730 2744. Gamper, M., Howard, P., Hunter, T., and Firtel, R. (1996)
Bliska, J., Guan, K., Dixon, J., and Falkow, S. (1991) Tyrosine Multiple roles of the novel protein tyrosine phosphatase
phosphate hydrolysis of host proteins by an essential Yers- PTP3 during Dictyostelium growth and development. Mol
inia virulence determinant. Proc Natl Acad Sci USA 88: Cell Biol 16: 2431 2444.
1187 1191. Guan, K., and Dixon, J.E. (1990) Protein tyrosine phos-
Bretz, J., Losada, L., Lisboa, K., Thareja, N., and Hutcheson, phatase activity of an essential virulence determinant in
S.W. (2002) Lon protease functions as a negative regulator Yersinia. Science 249: 553 556.
of type III protein secretion in Pseudomonas syringae. Mol Guttman, D., and Greenberg, J. (2001) Functional analysis
Microbiol 45: 397 409. of the type III effectors AvrRpt2 and AvrRpm1 of Pseudo-
Cardinale, F., Jonak, C., Ligternik, W., Niehaus, K., Boller, monas syringae with the use of a single copy genomic inte-
T., and Hirt, H. (2000) Differential activation of four specific gration system. Mol Plant-Microbe Interact 14: 145 155.
MAPK pathways by distinct elicitors. J Bio Chem 275: Guttman, D., Vinatzer, B., Sarkar, S., Ranall, M., Kettler, G.,
36734 36740. and Greenberg, J. (2002) A functional screen for the type
Charity, J., Pak, K., and Hutcheson, S. (2003) Novel III secretome of the plant pathogen Pseudomonas syrin-
exchangeable effector loci associated with the hrp patho- gae. Science 295: 1722 1726.
genicity island of P. syringae: evidence for integron-like Huang, H.C., Schuurink, R., Denny, T.P., Atkinson, M.M.,
assembly from transposed gene cassettes. Mol Plant- Baker, C.J., Yucel, I., Hutcheson, S.W., and Collmer, A.
Microbe Interact 16: 495 507. (1988) Molecular cloning of a Pseudomonas syringae pv.
Chen, Z., Kloek, A.P., Boch, J., Katagiri, F., and Kunkel, B.N. syringae gene cluster that enables Pseudomonas fluore-
(2000) The Pseudomonas syringae avrRpt2 gene product scens to elicit the hypersensitive response in tobacco. J
promotes pathogen virulence from inside plant cells. Mol Bacteriol 170: 4748 4756.
Plant-Microbe Interact 13: 1312 1321. Hutcheson, S.W. (1998) Currents concepts of active defense
Collmer, A., Lindegerg, M., Petnicki-Ocwieja, T., Schneider, in plants. Annu Rev Phytopathol 36: 59 90.
D., and Alfano, J. (2002) Genomic mining type III secretion Hutcheson, S.W. (1999) The hrp cluster of Pseudomonas
system effectors in Pseudomonas syringae yields new picks syringae. A pathogenicity island encoding a type III protein
for all TTSS prospectors. Trends Microbiol 10: 462 469. translocation complex? In Pathogenicity Islands and Other
Cornelis, G. (2002) Yersinia type III secretion: send in the Mobile Virulence Elements. Kaper, J.B., and Harker, J.,
effectors. J Cell Biol 158: 401 408. (eds). Washington, D.C.: American Society for Microbiol-
Dangl, J., and Jones, J. (2001) Plant pathogens and inte- ogy, pp. 309 329.
grated defence responses to infection. Nature 411: 826 Hutcheson, S.W., Bretz, J., Sussan, T., Jin, S., and Pak, K.
833. (2001) The enhancer binding protein HrpR and HrpS inter-
Delledonne, M., Xia, Y., Dixon, R., and Lamb, C. (1998) Nitric act to regulate hrp-encoded type III protein secretion in
oxide functions as a signal in plant disease resistance. Pseudomonas syringae strains. J Bacteriol 183: 5589
Nature 394: 585 588. 5598.
Deng, W.L., Rehm, A., Charkowski, A., Rojas, C., and Coll- Hutcheson, S., Bretz, J., Losada, L., Charity, J., and Sussan,
mer, A. (2003) Pseudomonas syringae exchangeable T. (2003) Regulation and detection of effectors translo-
effector loci: sequence diversity in representative patho- cated by Pseudomonas syringae. In Developments in Plant
vars and virulence function in P. syringae pv. syringae Pathology: Pseudomonas Syringae Pathovars and Related
B728a. J Bacteriol 185: 2592 2602. Pathogens, Vol., 2. Lacobelis, N., (ed.). Dordrecht: Kluwer
DeVinney, R., Steele-Mortimer, O., and Finlay, B. (2000) Academic Publishers, (in press).
Phosphatases and kinases delivered to the host cell by Innes, R. (2001) Mapping out the roles of MAP kinases in
bacterial pathogens. Trends Microbiol 8: 29 33. plant defense. Trends Pl Sci 6: 392 394.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
HopPtoD2 modulates defence responses 399
Jackson, R.W., Athanassopoulos, E., Tsiamis, G., Mansfield, tion of Salmonella with host cells. Cell Microbiol 3: 795
J.W., Sema, A., Arnold, D.L., et al. (1999) Identification of 810.
a pathogenicity island, which contains genes for virulence Nurnberger, T., and Schell, D. (2001) Signal transmission in
and avirulence, on a large native plasmid in the bean the plant immune response. Trends Pl Sci 6: 372 379.
pathogen Pseudomonas syringae pv. phaseolicola. Proc Orth, K., Xu, Z., Mudgett, M., Bao, Z., Palmer, L., Bliska, J.,
Natl Acad Sci USA 96: 10875 10880. et al. (2000) Disruption of signaling by Yersinia effector
Jackson, R., Mansfield, J., Arnold, D., Sesma, A., Paynter, YopJ, a ubiquitin-like protein protease. Science 290: 1594
C., Murillo, J., Taylor, J., and Vivian, A. (2000) Excision 1597.
from tRNA genes of a large chromosomal region, carrying Petnicki-Ocweija, T., Schneider, D., Tam, V., Chancey, S.,
avrPphB, associated with race change in the bean patho- Shan, L., Jamir, Y., et al. (2002) Genomewide identification
gen, Pseudomonas syringae pv. phaseolicola. Mol Micro- of proteins secreted by the Hrp type III protein secretion
biol 38: 186 197. system of Pseudomonas syringae pv. tomato DC3000.
Jakobek, L., Smith, J.A., and Lindgren, P.B. (1993) Suppres- Proc Natl Acad Sci USA 99: 7652 7657.
sion of bean defense responses by Pseudomonas syrin- Ren, D., Yang, H., and Zhang, S. (2002) Cell death mediated
gae. Plant Cell 5: 57 63. by MAPK is associated with hydrogen peroxide production
Jin, Q., and He, S. (2002) Role of the Hrp pilus in type III in Arabidopsis. J Bio Chem 277: 559 656.
protein secretion in Pseudomonas syringae. Science 294: Reuber, T.L., and Ausubel, F.M. (1996) Isolation of Arabidop-
2556 2558. sis genes that differentiate between resistance responses
Jonak, C., Olresz, L., Bogre, L., and Hirt, H. (2001) Complex- mediated by RPS2 and RPM1 resistance genes. Pl Cell 8:
ity, cross talk, and integration of plant MAP kinase signal- 241 249.
ing. Curr Opin Plant Biol 15: 415 424. Ritter, C., and Dangl, J.L. (1995) The avrRpm1 gene of
Kaniga, K., Uralil, J., Bliska, J., and Galan, J.E. (1996) A Pseudomonas syringae pv. maculicola is required for viru-
secreted protein tyrosine phosphatase with modular effec- lence on Arabidopsis. Mol Plant-Microbe Interact 8: 444
tor domains in the bacterial pathogen Salmonella typhimu- 453.
rium. Mol Microbiol 21: 633 641. Scofield, S.R., Tobias, C.M., Rathjen, J.P., Chang, J.H.,
Kim, Y.-J., Lin, N.-C., and Martin, G. (2002) Two distinct Lavelle, D.T., Michelmore, R.W., and Staskawicz, B.J.
Pseudomonas effector proteins interact with the Pto kinase (1996) Molecular basis of gene-for-gene specificity in bac-
and activate plant immunity. Cell 109: 589 598. terial speck disease of tomato. Science 2074: 2063 2065.
Lee, J., Klusener, B., Tsiamis, G., Stevens, C., Neyt, C., Shao, F., Merritt, P., Bao, Z., Innes, R., and Dixon, J. (2002)
Tampakaki, A.P., et al. (2001) HrpZPsph from the plant A Yersinia effector and a Pseudomonas avirulence protein
pathogen Pseudomonas syringae pv. phaseolicola binds define a family of cysteine proteases functioning in bacte-
to lipid bilayers and forms an ion-conducting pore in vitro. rial pathogenesis. Cell 109: 575 588.
Proc Natl Acad Sci USA 98: 289 294. Shapiro, A., and Zhang, C. (2001) The role of NDR1 in
Lindgren, P.B., Peet, R.C., and Panopoulos, N.J. (1986) avirulence gene-directed signaling and control of pro-
Gene cluster of Pseudomonas syringae pv. phaseolicola grammed cell death in Arabidopsis. Plant Physiol 127:
controls pathogenicity on bean plants and hypersensitivity 1089 1101.
on nonhost plants. J Bacteriol 168: 512 522. Sory, M.-P., Boland, A., Lambermont, I., and Cornelis, G.R.
Lorang, J.M., Shen, H., Kobayashi, D., Cooksey, D., and (1995) Identification of the YopE and YopH domains
Keen, N.T. (1994) avrA and avrE in Pseudomonas syrin- required for secretion and internalization into the cytosol of
gae pv. tomato PT23 play a role in virulence on tomato macrophages, using the cyaA gene fusion approach. Proc
plants. Mol Plant-Microbe Interact 7: 508 515. Natl Acad Sci USA 92: 11990 12002.
Mackey, D., Holt, B., Wiig, A., and Dangl, J. (2002) RIN4 Staskawicz, B.J., Dahlbeck, D., and Keen, N.T. (1984)
interacts with Pseudomonas syringae type III effector mol- Cloned avirulence gene of Pseudomonas syringae pv. gly-
ecules and is required for RPM1-mediated resistance in cinea determines race-specific incompatibility on Glycine
Arabidopsis. Cell 108: 743 754. max (L.) Merr. Proc Natl Acad Sci USA 81: 6024 6028.
Mackey, D., Belkhadir, Y., Alonso, J., Ecker, J., and Dangl, Staskawicz, B., Mudgett, M., Dangl, J., and Galan, J. (2001)
J. (2003) Arabidopsis RIN4 is a target of the type III viru- Common and contrasting themes of plant and animal dis-
lence effector AvrRpt2 and modulates RPS2-mediated eases. Science 274: 2063 2065.
resistance. Cell 112: 379 389. Streuli, M., Krueger, N., Tsai, A., and Saito, H. (1989) A
Mayer, M.P. (1995) A new set of useful cloning and expres- family of receptor-linked protein tyrosine phosphatases in
sion vectos derived from pBluescript. Gene 163: 41 46. humans and Drosophila. Proc Natl Acad Sci USA 86:
Mudgett, M.B., and Staskawicz, B.J. (1999) Characterization 8698 8702.
of the Pseudomonas syringae pv. tomato AvrRpt2 protein: Tang, X., Frederick, R.D., Zhou, J., Halterman, D.A., Jia, Y.,
demonstration of secretion and processing during bacterial and Martin, G.B. (1996) Initiation of plant disease resis-
pathogenesis. Mol Microbiol 32: 927 941. tance by physical interaction of AvrPto and Pto kinase.
Murillo, J., Shen, H., Gerhold, D., Sharma, A., Cooksey, D.A., Science 274: 2060 2063.
and Keen, N.T. (1994) Characterization of pPT23B, the Tsiamis, G., Mansfield, J., Hockenhull, R., Jackson, R.,
plasmid involved in syringolide production by Pseudomo- Sesma, A., Athanassopoulos, E., et al. (2000) Cultivar-
nas syringae pv. tomato PT23 Plasmid 31: 275 287. specific avirulence and virulence functions assigned to avr-
Murli, S., Watson, R., and Galan, J.E. (2001) Role of tyrosine PphF in Pseudomonas syringae pv. phaseolicola, the
kinases and the tyrosine phosphatase SptP in the interac- cause of bean halo-blight disease. EMBO J 19: 3204 3214.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400
400 J. R. Bretz et al.
Uknes, S., Mauch-Mani, B., Moyer, M., Potter, S., Williams, Xiao, Y., Heu, S., Yi, J., Lu, Y., and Hutcheson, S.W. (1994)
S., Dincher, S., et al. (1992) Acquired resistance in Arabi- Identification of a putative alternate sigma factor and char-
dopsis. Pl Cell 4: 645 656. acterization of a multicomponent regulatory cascade con-
Wei, W., Plovanich-Jones, A., Deng, W.-L., Jin, Q.-L., Coll- trolling the expression of Pseudomonas syringae pv.
mer, A., Huang, H.C., and He, S.Y. (2000) The gene cod- syringae Pss61 hrp and hrmA genes. J Bacteriol 176:
ing for the Hrp pilus structural protein is required for type 1025 1036.
III secretion of Hrp and Avr proteins in Pseudomonas syrin- Xing, T., Malik, K., Martin, T., and Miki, B. (2001) Activation
gae pv. tomato. Proc Natl Acad Sci USA 97: 2247 2252. of tomato PR and wound-related genes by a mutagenized
Whalen, M.C., Innes, R.W., Bent, A.F., and Staskawicz, tomato MAP kinase kinase through divergent pathways.
B.J. (1991) Identification of Pseudomonas syringae Plant Molec Biol 46: 109 120.
pathogens of Arabidopsis and a bacterial locus determin- Yang, K.-Y., Liu, Y., and Zhang, S. (2001) Activation of a
ing avirulence on both Arabidopsis and soybean. Pl Cell mitogen-activated protein kinase pathway is involved in
3: 49 59. disease resistance in tobacco. Proc Natl Acad Sci USA 98:
Xiao, Y., and Hutcheson, S.W. (1994) A single promoter 741 746.
sequence recognized by a newly identified alternate sigma Zhang, Z.-Y. (1995) Are protein-tyrosine phosphatases spe-
factor directs expression of pathogenicity and host range cific for phosphotyrosine? J Bio Chem 270: 16052 16055.
determinants in Pseudomonas syringae. J Bacteriol 176: Zhang, S., and Klessig, D. (2001) MAPK cascades in plant
3089 3091. defense signaling. Trends Pl Sci 6: 520 527.
© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 49, 389 400