JOURNAL OF BACTERIOLOGY, Mar. 2002, p. 1340–1348 0021-9193/02/$04.00⫹0 DOI: 10.1128/JB.184.5.1340–1348.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 184, No. 5

Two Novel Type III-Secreted Proteins of Xanthomonas campestris pv. vesicatoria Are Encoded within the hrp Pathogenicity Island Laurent Noël, Frank Thieme, Dirk Nennstiel,† and Ulla Bonas* Institut für Genetik, Martin-Luther-Universität Halle-Wittenberg, D-06099 Halle (Saale), Germany Received 24 September 2001/Accepted 4 December 2001

The Hrp type III protein secretion system (TTSS) is essential for pathogenicity of gram-negative plant pathogen Xanthomonas campestris pv. vesicatoria. cDNA-amplified fragment length polymorphism and reverse transcription-PCR analyses identified new genes, regulated by key hrp regulator HrpG, in the regions flanking the hrp gene cluster. Sequence analysis revealed genes encoding HpaG, a predicted leucine-rich repeatcontaining protein, the lysozyme-like HpaH protein, and XopA and XopD, which are similar in sequence to Hpa1 from Xanthomonas oryzae pv. oryzae and PsvA from Pseudomonas syringae, respectively. XopA and XopD (Xanthomonas outer proteins) are secreted by the Xanthomonas Hrp TTSS and thus represent putative effector proteins. Mutations in xopA, but not in xopD, resulted in reduced bacterial growth in planta and delayed plant reactions in susceptible and resistant host plants. Since the xopD promoter contains a putative hrp box, which is characteristic of hrpL-regulated genes in P. syringae and Erwinia spp., the gene was probably acquired by horizontal gene transfer. Interestingly, the regions flanking the hrp gene cluster also contain insertion sequences and genes for a putative transposase and a tRNAArg. These features suggest that the hrp gene cluster of X. campestris pv. vesicatoria is part of a pathogenicity island. interaction: bacteria expressing a given avirulence gene (avr) are recognized by plants carrying the corresponding disease resistance gene (46). Activity of all avirulence genes tested so far depends on a functional Hrp TTSS system, and more than 10 avr genes elicit the HR when they are expressed within the resistant-plant cell, indicating that Avr proteins are translocated effector proteins (11). Xanthomonas campestris pv. vesicatoria is the causal agent of bacterial spot disease of peppers and tomatoes. The chromosomal 23-kb hrp gene cluster contains six operons, hrpA to hrpF (9, 15, 16, 24; U. Bonas, unpublished data). Among the more than 20 proteins encoded by the hrp region, nine are highly conserved in plant and animal pathogens. These genes were renamed hrc (hrp conserved) and are likely to form the core of the TTSS (8). Analysis of nonpolar mutants in the hrp operons also identified hpa (hrp-associated) genes, which contribute to, but are not essential for, the interaction with the plant (24; U. Bonas, unpublished data). We recently demonstrated Hrpdependent secretion of HrpB2 and HrpF, which are encoded within the hrp gene cluster and which are essential for pathogenicity (35). Other secreted proteins that are not essential for pathogenicity are Xanthomonas outer proteins (Xop) such as XopB (34) and avirulence proteins AvrBs3, AvrBs4, AvrRxv, AvrBsT, and AvrBs1 (5, 14, 34, 36). Expression of hrp genes is induced in plant leaves and in minimal medium XVM2 (38, 43) and is controlled by regulatory genes hrpG and hrpX, located outside of the hrp gene cluster. The HrpG protein belongs to the OmpR family of two-component regulatory systems and regulates the expression of a large regulon, which was recently identified using cDNA-amplified fragment length polymorphism (cDNAAFLP) analysis. The expression profiles of two isogenic X. campestris pv. vesicatoria strains, 85-10 and 85*, which differ in their hrp gene expression status, were compared (34). Strain

Gram-negative bacterial pathogens of plants and animals have evolved specialized export systems to target virulence factors to their hosts. In Agrobacterium tumefaciens, the causal agent of crown gall, the type IV secretion system mediates transfer of the transferred DNA (T-DNA) and associated proteins into plant cells, where proteins encoded by the T-DNA manipulate the plant’s hormone balance and metabolism (20). The type III secretion system (TTSS) allows vectorial secretion of so-called effector proteins across the bacterial envelope into the host, where they modulate host defense responses and physiology (11, 18). In general, mutations in type III or type IV secretion systems strongly affect pathogenicity, highlighting the collective contribution of the translocated effectors to virulence. Relatively few effectors transported by the TTSS have been identified in plant pathogens (11). Examples are PopB and PopC from Ralstonia solanacearum (21), DspA from Erwinia amylovora (19), and HopPsyA from Pseudomonas syringae pv. syringae (1), which are encoded in the regions flanking the respective hrp (hypersensitive reaction and pathogenicity) gene clusters. hrp genes encode the TTSS in plant pathogens and are essential for bacterial growth in planta, causing disease in susceptible hosts and eliciting a hypersensitive reaction (HR) in resistant plants (2). The HR is a rapid, localized cell death that is part of the plant’s innate defense responses and that halts pathogen ingress (27). In most cases, specific recognition of the pathogen by the resistant plants is based on a gene-for-gene * Corresponding author. Mailing address: Institut für Genetik, Martin-Luther-Universität Halle-Wittenberg, D-06099 Halle (Saale), Germany. Phone: (49) 345 5526290. Fax: (49) 345 5527277. E-mail: [email protected]. † Present address: Bayer-AG Zentrale Forschung, D-51368 Leverkusen, Germany. 1340

hrp PATHOGENICITY ISLAND

VOL. 184, 2002

85* expresses hrpG*, a mutated form of key regulatory gene hrpG, which leads to the constitutive expression of hrp genes (45). Expression of the second known regulator, hrpX, is dependent on hrpG. HrpX, an AraC-type transcriptional activator, controls expression of operons hrpB to hrpF (43) and is essential for expression of most new members of the hrpG regulon. Among the 35 known genes belonging to the hrpG regulon, seven hrpG-induced (hgi) cDNA fragments map to the large hrp gene cluster (34) (Fig. 1A). Three of these hgi genes map to the hrp flanking regions, which had not previously been characterized. Here, we report the detailed analysis of both regions flanking the X. campestris pv. vesicatoria hrp gene cluster. The entire region has features of a pathogenicity island (PAI) and encodes novel proteins secreted by the Hrp TTSS. MATERIALS AND METHODS Bacterial strains, growth conditions, and plasmids. Bacterial strains and plasmids used in this study are described in Table 1. Escherichia coli cells were cultivated at 37°C in Luria-Bertani medium, and X. campestris pv. vesicatoria strains were cultivated at 30°C in NYG (12) or in hrp-inducing medium XVM2 (43). Plasmids were introduced into E. coli by electroporation and into X. campestris pv. vesicatoria by conjugation, using pRK2013 as a helper plasmid in triparental matings as described previously (13, 17). Antibiotics were added to the media at the following final concentrations: ampicillin, 100 ␮g/ml; kanamycin, 25 ␮g/ml; tetracycline, 10 ␮g/ml; rifampin, 100 ␮g/ml; spectinomycin, 100 ␮g/ml. Plant material and plant inoculations. Inoculation of the nearly isogenic pepper cultivars Early Cal Wonder (ECW) and ECW-10R, ECW-20R, and ECW-30R, which carry the Bs1, Bs2, and Bs3 resistance genes, respectively, was

1341

performed as described previously (9). Unless stated differently, bacterial suspensions were infiltrated into the leaf at 108 CFU/ml in 1 mM MgCl2. In planta growth in ECW was determined as described previously (9). Experiments were repeated at least three times. DNA and RNA analyses. To sequence the region downstream of hrpA, the sequence of pK2 was completed using an ABI 377 Prism DNA sequencer (Applied Biosystems Inc., Foster City, Calif.). Deletions in both directions of the pBLB1 insert were generated using the Exo mung bean deletion kit (Stratagene, La Jolla, Calif.). A 3-kb HindIII fragment overlapping with pK2 and pBLB1 was cloned from pXV9 (9) into pBluescript II KS (pB-KS), giving pBLH1. To sequence the 3.1-kb region flanking hrpF, the sequence of pBF (5.4-kb EcoRV fragment encompassing hrpF) was completed. The adjacent 2.3-kb EcoRV fragment was subcloned from pXV4 into pB-KS, giving pBLr5, and partially sequenced. Sequences were analyzed using the Sequencher program (Gene Codes Corp., Ann Arbor, Mich.) and the DNASTAR package (DNASTAR Inc., Madison, Wis.). The tRNA gene was identified using the tRNAscan-SE algorithm (31). RNA extraction, cDNA synthesis, and reverse transcription-PCR (RT-PCR) experiments were performed as described previously (34). Primer sequences are available upon request. Molecular biology experiments were performed according to standard procedures (4). Generation of mutations in the genes downstream of hrpA. To introduce the ⌬L deletion into strain 85*, suicide construct pOLB (34) was used, creating 85*⌬L (24). A 1.0-kb deletion of ORF1, causing an out-of-frame (oof) mutation at codon 96, was achieved by BstEII digestion and religation of pULB1 (34), giving pU⌬ORF1. A 4.4-kb BamHI/ApaI fragment was cloned into suicide plasmid pOK (24), giving pOORF1, which was introduced in strain 85-10, creating 85-10⌬ORF1. In hpaH, a frameshift mutation was introduced at codon 38 by digestion of pULB1 by Csp45I, filling in, and religation, giving pUhpaH. The 5.6-kb BamHI/ ApaI fragment of pUhpaH was subcloned into pOK, giving pOhpaH, which was introduced into strain 85-10, creating 85-10 hpaH-oof. To delete xopA, the 2.7-kb NheI/ApaI fragment of pULB1 was cloned into pB-SK digested by SpeI/ApaI, creating pBxopA. pBxopA was partially digested

FIG. 1. Map of the large hrp gene cluster. (A) Schematic overview of the hrp operons hrpA to hrpF plus flanking regions. Gray circles, locations of seven hgi fragments; black dots, PIP boxes. (B) Restriction map of the region downstream of hrpA. Arrows indicate the extent of deletions generated. ⴱ, introduced frameshift mutations. (C) ORFs in the region downstream of hrpA. Single-headed arrows, ORFs with high coding probability (DNAstar; Genequest, Pseudomonas aeruginosa codon preference matrix) and the direction of transcription; double-headed arrow, insertion sequence ISXc 7; ⴱ, frameshifts; black dots and open dots, PIP and hrp boxes, respectively; gray circles, hgi fragments. The most distal part of the sequence does not contain ORFs with high coding probability (see also Table 2). (D) Comparison of the region shown in panel C to the orthologous sequence in X. oryzae pv. oryzae (Xoo; GenBank accession no. AF026197 and AF232058) (48). Gray areas, noncolinear DNA regions between X. oryzae pv. oryzae and X. campestris pv. vesicatoria. Conserved sequences are more than 80% identical on the DNA level.

1342

NOËL ET AL.

J. BACTERIOL.

TABLE 1. Strains and plasmids used in this study Strain or plasmid

X. campestris pv. vesicatoria 85-10 85-10⌬hrpG 85-10⌬R 85-10⌬L 85-10⌬ORF1 85-10⌬hpaH-oof 85-10⌬xopA 85-10⌬xopD 85* 82* 85*⌬L 85*⌬hrpX 85*⌬hrcV and 82*⌬hrcV 85*::pICxopA 85*::pICxopD 85-10::pICxopA 85-10::pICxopD E. coli DH10B DH5␣ DH5␣ ␭pir Plasmids pUC118 pULB1 pB-KS, PB-SK pK2 pBLB1 pBLH1 pBF pBL5r pRK2013 pDS300F pLAFR3 pL3hpaFC pXV331 pXV4 pXV4::462 pXV9 pOK pOG pOLB pOR pOORF1 pOhpaH pOxopA pOxopD pIC1 pICxopA pICxopD a

Relevant characteristicsa

Reference or source

Pepper race 2; wild type; expresses avrBs1 and avrBs2; Rifr 85-10 derivative; carries a nonpolar hrpG deletion; Rifr 85-10 derivative; carries a deletion from hpaF to hpaG; Rifr 85-10 derivative; carries a deletion from ORF1 to xopA; Rifr 85-10 derivative; carries a deletion in ORF1; Rifr 85-10 derivative; carries a frameshift mutation in hpaH; Rifr 85-10 derivative; carries a deletion in xopA; Rifr 85-10 derivative; carries a deletion and a frameshift mutation in xopD; Rifr 85-10 derivative containing the hrpG* mutation; constitutive hrp gene expression; Rifr 82-8 derivative containing the hrpG* mutation; constitutive hrp gene expression; pepper race 1; Rifr 85* derivative; carries a deletion from ORF1 to xopA; Rifr 85* derivative with a nonpolar hrpX deletion; Rifr 85* and 82* derivatives with a nonpolar hrcV deletion; nonfunctional TTSS; Rifr 85* derivative containing pICxopA; expresses c-myc-tagged XopA; Rifr Tcr 85* derivative containing pICxopD; expresses c-myc-tagged XopD; Rifr Tcr 85*::pICxopA derivative; wild-type hrpG restored using pOG; Rifr Tcr 85*::pICxopD derivative; wild-type hrpG restored using pOG; Rifr Tcr

10 45 34 34 This This This This 45 45

F⫺ mcrA (mrr-hsdRMS-mcrBC) ␾ 80d lacZ⌬M15 ⌬lacX74 endA1 recA1 deoR D(ara, leu)7697 ara⌬139 galU galK ␭⫺nupG rpsL F⫺ ␾80d lacZ⌬M15 ⌬(lacZYA-argF)U169 endA1 recA1 deoR hdsR17 (rk⫺ mk⫹) phoA supE44 ␭⫺ thi-1 gyrA96 relA1 DH5␣ derivative, ␭pir; for replication of pIC1 and pOK derivatives

BRL, Bethesda, Md.

ColE1 replicon; Apr pUC118 derivative with 7.5-kb BamHI insert from pXV331; contains ORF1 to xopD Phagemid, pUC derivative; Apr pB-KS derivative containing a 4.5-kb insert of pXV9; carries hrpA and 1.5-kb downstream region pB-KS derivative containing a 7.5-kb BamHI insert from pXV331; carries ORF1 to xopD pB-KS derivative containing a 3.0-kb HindIII insert from pXV331 (xopD) pB-KS derivative containing a 5.3-kb EcoRV insert from pXV2 containing hrpF pB-KS derivative containing a 2.3-kb EcoRV insert from pXV4 containing hpaF and hpaG Helper plasmid; ColE1 replicon TraRK⫹ Mob⫹; Kmr pDSK602 derivative carrying avrBs3 tagged with FLAG epitope at the C terminus; broad host range; Smr RK2 replicon Mob⫹ Tra⫺; contains plac; Tcr pLAFR3 derivative carrying hpaF and hpaG; HpaF is c-myc tagged at amino acid position 66 pLAFR3 derivative; contains a region from downstream of ORF1 to hrpA from X. campestris pv. vesicatoria strain 75-3 pLAFR3 derivative; contains a region from hrpC2 to hpaG from X. campestris pv. vesicatoria strain 75-3 Tn 3-gus insertion derivative of pXV4; Tcr Kmr pLAFR3 derivative; contains region from ISX c7 to hrpE from X. campestris pv. vesicatoria strain 75-3 Suicide vector; pKNG101 derivative; Smr/Sucs pOK derivative; carries wild-type hrpG pOK derivative; carries a 3.7-kb insert with a 3.8-kb deletion from ORF1 to xopA pOK derivative; carries a 5.7-kb insert with a 2.5-kb deletion from hpaF to hpaG pOK derivative; carries a 4.4-kb insert with a 1.0-kb deletion in ORF1 pOK derivative; carries a 5.6-kb insert with an out-of-frame mutation at codon position 38 of hpaH pOK derivative; carries a 2.4-kb insert with a 0.3-kb deletion in xopA pOK derivative; carries a 1.85-kb insert with a 0.15-kb deletion in xopD causing an out-of-frame mutation after codon position 311 of xopD Integrative vector; carries a triple-c-myc epitope followed by a promoterless uidA gene; Tcr pIC1 derivative; carries the entire xopA ORF fused in frame to the c-myc epitope pIC1 derivative; carries the entire xopD ORF fused in frame to the c-myc epitope

42 34 Stratagene 44

Ap, ampicillin; Km, kanamycin; Rif, rifampin; Sm, spectinomycin; Suc, sucrose; Tc, tetracycline; r, resistant; s, sensitive.

study study study study

This study 34 35, 36 This study This study This study This study

BRL 32

34 34 23 23 17 41 39 This study 34 43 43 9 24 34 34 34 This study This study This study This study 34 This study This study

hrp PATHOGENICITY ISLAND

VOL. 184, 2002

1343

TABLE 2. Description of genes in the regions flanking the hrp gene cluster of X. campestris pv. vesicatoria Gene

Downstream of hrpA ORF1 hpaH xopA ORF4 xopD ORF6 Downstream of hrpF hpaF hpaG

Closest protein homolog (organism; accession no.)a

Identity/similarityb (%)

Hpa2 (X. oryzae pv. oryzae; AAF61278) Hpa1 (X. oryzae pv. oryzae; AAC95121) Putative transposase (Y. pestis; CAB54971) PsvA (P. syringae pv. eriobotryae; BAA87062) Putative transposase (X. campestris pv. mangiferaeindicae; AF249895)

96/98 46/59 50/78 65/74c 74/78

340 aaf Imperfect PIP box; hgi8Ig; 157 aa PIP box; hgi27g; 50% G⫹C content; 117 aa Conserved frameshifting site (CAAAAAAA) hrp box; 54% G⫹C content; 612 aa; NLSh Frameshift; part of ISXc 7

HpaF (X. oryzae pv. oryzae; AB045312) HpaF (X. oryzae pv. oryzae; AB045312)

81/87d 87/93e

PIP box; hgi203g; 197 aa; NLS 432 aa; LRRs

Comments

a

Closest homologous protein using the BlastX algorithm (37) in nonredundant databases. Identities and similarities between the respective proteins were determined by using the Blast 2 Sequences algorithm (40). Refers to the last 300 amino acids of XopD and PsvA. d Refers to the first 180 amino acids of HpaF from X. oryzae pv. oryzae. e Refers to the last 430 amino acids of HpaF from X. oryzae pv. oryzae. f aa, amino acids; refers to the predicted size of the encoded protein. g hrpG-induced cDNA fragment; gene identified by cDNA-AFLP (34). h Putative NLS. b c

by ClaI and SalI, filled in, and religated, resulting in pB⌬xopA. In this construct, xopA (354 bp) contains a 312-bp deletion leaving only the first 11 amino acids of the protein unchanged. The 2.4-kb ApaI/XbaI insert was subcloned into pOK, giving pOxopA, which was introduced into 85-10, creating 85-10⌬xopA. To mutate xopD, the 2.0-kb ApaI/BamHI fragment of pULB1 was subcloned into pB-SK, giving pBxopD. pBxopD was digested by EcoRI, filled in, and religated, giving pB⌬xopD. The 1.85-kb ApaI/BamHI insert of pB⌬xopD was cloned into pOK, giving pOxopD, which was introduced into 85-10, resulting in 85-10⌬xopD, which carries an oof mutation after amino acid 311 of XopD. GUS assays. ␤-Glucuronidase (GUS) assays were performed with exponentially growing X. campestris pv. vesicatoria as described previously (36). One GUS unit is defined as 1 nmol of 4-methylumbelliferone released per min per bacterium. Epitope tagging of XopA, XopD, and HpaF. A 0.4-kb fragment containing xopA and a 2.0-kb fragment encompassing xopD were amplified by PCR from pXV331 and cloned into pIC1 in frame with a triple-c-myc epitope, giving pICxopA and pICxopD, respectively. Primer sequences are available upon request. To tag hpaF, a 2.1-kb fragment starting 100 bp before the putative translation start codon of hpaF was cloned into pB-KS, giving pBhpaF. A PCR-amplified triple-c-myc tag was cloned in frame into the SalI site, which is situated 165 bp after the putative translation start codon of hpaF, giving pBhpaFC. The pBhpaFC insert was cloned into pLAFR3 under the control of the lac promoter, giving pL3hpaFC. Further cloning details are available upon request. Protein analysis and secretion experiments. Secretion experiments and Western blot analyses were performed as described previously (36). The following primary antibodies were used: a polyclonal anti-AvrBs3 antibody (28), a monoclonal anti-c-myc antibody (Roche, Mannheim, Germany; dilution 1:10,000), polyclonal anti-HrcN antiserum (36), and polyclonal anti-HrpF antiserum (D. Büttner and U. Bonas, unpublished data). Horseradish peroxidase-labeled goat anti-mouse or goat anti-rabbit antibodies were used as secondary antibodies. Reactions were visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, N.J.). Nucleotide sequence accession numbers. The sequences of the 9.1-kb fragment downstream of hrpA and the 3.1-kb sequence downstream of hrpF from X. campestris pv. vesicatoria have been submitted to GenBank (U33548 and AF056246, respectively). The predicted HpaF protein sequence from Xanthomonas oryzae pv. oryzae has been added to the original submission (AB045312).

RESULTS Sequence analysis of the region flanking hrpA. We recently identified two hrpG- and hrpX-regulated genes (hgi27 and hgi81), which hybridized to a 7.5-kb BamHI fragment in the region downstream of hrpA (34) (Fig. 1B). Sequence analysis of 9.1 kb flanking hrpA (see Materials and Methods) revealed

six open reading frames (ORFs) with high coding probability (Fig. 1C; Table 2). The sequence containing hpaH (corresponds to hgi81) and xopA (corresponds to hgi27) revealed more than 80% homology to the orthologous sequence of X. oryzae pv. oryzae (Fig. 1C and D). The predicted HpaH protein contains an N-terminal signal peptide of 34 amino acids and is almost identical to Hpa2 from X. oryzae pv. oryzae (96% identity, 98% similarity) (48). Both proteins belong to the lysozyme-like protein family (33). The predicted XopA protein (Xanthomonas outer protein) shows similarity to Hpa1 from X. oryzae pv. oryzae (46% identity, 59% similarity) (48). Perfect (TTCGC-N15-TTCGC) and imperfect (TTCGC-N15-TTCGT) PIP (plant-inducible promoter) boxes are located 214 and 130 bp upstream of the predicted translation start codons of xopA and hpaH, respectively. PIP boxes are found in promoter regions of X. campestris pv. vesicatoria hrp operons (16, 43). The three genes located between xopA and hrpA do not have

FIG. 2. Genetic organization of the region downstream of hrpF. (A) Open arrows, ORFs; black arrow, tRNAArg gene; gray circles, hgi fragments; triangle, position and orientation of Tn3-gus insertion no. 462 (43); double-headed arrow, ⌬R deletion. (B) Schematic drawing of the orthologous sequence in X. oryzae pv. oryzae (GenBank accession no. AB045312). Black dots, PIP boxes (TTCGC-N16-TTCGC in X. campestris pv. vesicatoria and TTCGC-N16-TTCAC in X. oryzae pv. oryzae); gray areas, DNA sequences with more than 90% identity to the X. campestris pv. vesicatoria region.

1344

NOËL ET AL.

orthologs in X. oryzae pv. oryzae (Fig. 1; Table 2). Interestingly, the C-terminal portion of XopD contains a putative nuclear localization signal (NLS; KKKK) at amino acid position 602 and shows 74% similarity to the last 300 amino acids of virulence factor PsvA from P. syringae pv. eriobotryae (26). The xopD promoter does not contain a PIP box, but rather a hrp box, which is found in all hrpL-dependent promoters in P. syringae and Erwinia spp. (GGAACTNA-N13-CGACNNA; consensus: GGAACcNa-N13/14-cCACNNA; lowercase letters indicate less-well-conserved bases [25]). ORF4 is predicted to encode a protein with similarity to transposases from X. campestris and Yersinia pestis. However, the sequence conservation is restricted to the N-terminal 84 amino acids and then continues in the product of another reading frame. Finally, ORF6, which is interrupted by a stop codon at codon position 121, encodes a protein highly similar to the putative transposase encoded by insertion sequence (IS) IS1595 of X. campestris pv. mangiferaeindicae. Since ORF6 is flanked by 23-bp inverted repeats and 2-bp direct repeats, it has features typical of an IS element and was designated ISXc 7. Sequence analysis of the region flanking hrpF. Sequence analysis of the 3.1-kb region downstream of hrpF, which contains hgi203, revealed two ORFs, hpaF and hpaG, and a putative tRNAArg gene (anticodon CCG; Fig. 2). hrpF and the region downstream up to the tRNA gene are highly similar to the orthologous region of X. oryzae pv. oryzae. Curiously, in X. oryzae pv. oryzae, there is only one ORF, which we designated hpaF. X. campestris pv. vesicatoria cDNA-AFLP fragment hgi203 was derived from hpaF, which encodes a putative protein of 197 amino acids containing a putative NLS (RPRRR) at amino acid 43. hpaG encodes a putative protein of 432 amino acids, which contains six leucine-rich repeats (LRRs; consensus sequence, LXXLXXLXXLXLXXXXXLXXLPX X). HpaF is 87% similar to the N terminus of HpaF from X. oryzae pv. oryzae, and HpaG is 93% similar to its C terminus. The DNA sequences of hpaF and hpaG from X. campestris pv. vesicatoria and hpaF from X. oryzae pv. oryzae mainly differ by a 7-bp deletion in the 3⬘ region of hpaF in X. campestris pv. vesicatoria. This deletion leads to a frameshift in hpaF which is present in all three different X. campestris pv. vesicatoria strains tested (data not shown). An imperfect PIP box (TTCGC-N16-TTCGC) is located 82 bp upstream of the predicted translation start codon of hpaF in X. campestris pv. vesicatoria. xopA is necessary for growth in planta and full avirulence of X. campestris pv. vesicatoria. We previously found that X. campestris pv. vesicatoria strain 85-10⌬L was reduced in virulence compared to wild-type strain 85-10, i.e., delayed symptom formation and reduced growth in planta (34). Here, we introduced deletions or frameshift mutations in different genes flanking the hrp gene cluster into strain 85-10 to study their role individually. In strains 85-10⌬L and 85-10⌬R, the regions encompassing ORF1 to ORF4 and hpaF to hpaG, respectively, were deleted (34) (Fig. 1B and 2A). In addition, ORF1 (in 85-10⌬ORF1), xopA (in 85-10⌬xopA), and xopD (in 8510⌬xopD) were deleted, and a frameshift mutation was generated in hpaH (in 85-10 hpaH-oof; Fig. 1B). The mutant derivatives of strain 85-10 were tested for HR induction in resistant pepper line ECW-10R (contains the Bs1 resistance gene, which recognizes avrBs1) and the induction of water-soaking symptoms in susceptible pepper line ECW (Fig. 3). Strains 85-10⌬L

J. BACTERIOL.

FIG. 3. Effects of deletions in the hrp gene cluster flanking regions on the plant interaction. (A) HR induction in resistant ECW-10R pepper plants. X. campestris pv. vesicatoria strains were inoculated at 108 CFU/ml in 1 mM MgCl2 into the intercellular spaces of a fully expanded leaf of an 8-week-old ECW-10R plant. The lower side of the leaf was photographed 48 h after inoculation. See text for details. (B) Bacterial growth in the susceptible ECW pepper plants. X. campestris pv. vesicatoria strains were inoculated at 104 CFU/ml in 1 mM MgCl2 into the intercellular spaces of fully expanded leaves of 8-weekold ECW plants. Growth of strains 85-10, 85-10⌬hrpG, 85-10⌬ORF1, 85-10 hpaH-oof, 85-10⌬xopA, 85-10⌬xopD, and 85-10⌬L was monitored as described earlier (9). Values represent the means of four samples from four different plants. Error bars, standard deviations. For the sake of clarity, growth of strains 85-10⌬ORF1 and 85-10⌬xopD, which was identical to that of wild-type strain 85-10, is not shown and error bars for strains 85-10⌬xopA and 85-10 hpaH-oof were omitted. Results (A and B) are from one representative experiment.

hrp PATHOGENICITY ISLAND

VOL. 184, 2002

1345

TABLE 3. Promoter activities of xopA, xopD, and hpaF in X. campestris pv. vesicatoria strains 85-10 and 85*a Reporter construct (gene)

FIG. 4. The ⌬L deletion has no effect on type III secretion in vitro. X. campestris pv. vesicatoria strains expressing AvrBs3 from pDS300F were incubated in secretion medium. Total protein extracts (lanes 1 to 3) and supernatants (lanes 4 to 6) of strains 85* (lanes 1 and 4), 85*⌬hrcV (lanes 2 and 5), and 85*⌬L (lanes 3 and 6) were separated by sodium dodecyl sulfate–10% polyacrylamide gel electrophoresis and analyzed by immunoblotting using AvrBs3- and HrpF-specific antibodies. Total protein extracts and supernatants were concentrated 10- and 100-fold, respectively. Membranes were reprobed with the antiserum directed against cytoplasmic protein HrcN to assure that no bacterial lysis had occurred during the experiment (data not shown).

and 85*⌬L (85-10⌬L derivative carrying the hrpG* mutation) both caused clearly delayed water-soaking symptoms in ECW pepper plants and delayed HR in ECW-10R pepper plants (12 h to 24 h later than strains 85-10 and 85*, respectively) (Fig. 3A; data not shown). While strains 85-10⌬R, 85-10⌬ORF1, and 85-10⌬xopD behaved like wild-type strain 85-10, strain 85-10⌬xopA displayed an intermediate phenotype compared to strains 85-10⌬L and 85-10, i.e., slightly delayed water-soaking symptoms and HR. Strain 85-10 hpaH-oof induced sometimes a slightly delayed HR. The mutant phenotypes of strain 8510⌬L in ECW and ECW-10R plants could be complemented by pXV331 (Fig. 3A; data not shown). The reduced avirulence of strain 85-10⌬L is not specifically associated with the avrBs1-Bs1 interaction and was also observed in association with avrBs2 recognition in ECW-20R plants (contains the Bs2 resistance gene; data not shown). Similarly, strain 85-10⌬L(pDS300F), which expresses the avrBs3 gene, induced a delayed and partial HR in the corresponding resistant ECW-30R plants (contains the Bs3 resis-

FIG. 5. Expression profiles of ORF1, hpaH, xopA, xopD, hpaF, and hpaG. Shown is RT-PCR analysis of X. campestris pv. vesicatoria strains 85-10, 85*, and 85*⌬hrpX, grown in NYG, and of strains 85-10 and 85-10⌬hrpG, grown in hrp gene induction medium XVM2. 16S rDNA was used as a standard. The DNA samples were separated on a 1.5% agarose gel and stained with ethidium bromide.

pICxopA (xopA) pICxopD (xopD) pXV4::462 (hpaF)

Activity (GUS units) ⫾ SDb in strain: c

85-10

85*

1.2 ⫾ 0.2 0.24 ⫾ 0.06 0.06 ⫾ 0.02

24 ⫾ 3 1.9 ⫾ 0.4 2.4 ⫾ 0.5

Fold induction (compared to 85-10)

20 8 40

a

Bacteria were grown in NYG medium. One unit corresponds to 1 nmol of 4-methylumbelliferone released per min per bacterium. c Carries the hrpG* mutation, which leads to constitutive hrp gene expression (45). b

tance gene; data not shown). In addition, the necrosis normally induced by strain 85* in tobacco was weaker with 85*⌬L (data not shown) (45). To determine which gene is responsible for the reduced growth of strain 85-10⌬L in susceptible pepper line ECW, we determined the growth of mutants with mutations in individual genes. Strain 85-10⌬ORF1 behaved like the wild type, whereas strain 85-10⌬xopA always showed a reduction in growth compared to strain 85-10 with levels intermediate between those of wild-type strain 85-10 and strain 85-10⌬L (Fig. 3B). The growth of strain 85-10 hpaH-oof was not affected or was weakly affected. The mutation in xopD, which was not included in the ⌬L deletion, did not affect growth in planta (data not shown). In summary, these results indicate that the reduced avirulence and growth in planta of strain 85-10⌬L are essentially due to the loss of xopA but partially also due to the loss of hpaH. AvrBs3 and HrpF secretion is not affected in strain 85*⌬L. To test if the reduced-virulence and avirulence phenotypes of the deletion mutants are due to reduced secretion of effector proteins by the Hrp system, we analyzed the secretion of two TTSS substrates: AvrBs3, which is secreted by the X. campestris pv. vesicatoria TTSS (36), and HrpF, which is an essential, secreted pathogenicity protein that, it has been suggested, plays a role in translocation (35). Levels of in vitro secretion of both AvrBs3 and HrpF in strain 85*⌬L were comparable to the wild-type levels (Fig. 4). Thus, the genes deleted in strain 85*⌬L do not encode components of the type III secretion machinery. xopD belongs to the hrpG regulon. We investigated the regulation of gene expression in the two regions flanking the hrp gene cluster using different strains grown in complex medium NYG or in hrp-inducing medium XVM2 (Fig. 5). The expression of xopA (hgi27), hpaH (hgi81), and hpaF (hgi203) is strictly dependent on hrpG and hrpX and is induced in XVM2 (34); these findings were confirmed here by RT-PCR experiments (Fig. 5). The expression of ORF1 appears to be constitutive and independent of hrpG and hrpX. Interestingly, xopD expression is induced in a hrpG- and hrpX-dependent manner, indicating that xopD also belongs to the hrpG regulon (Fig. 5). Finally, primers that amplify a 520-bp fragment spanning hpaF and hpaG were used to detect a cDNA fragment of the expected size. This shows that hpaG is regulated in the same way as hpaF and indicates that the two genes are probably in the same operon (Fig. 5).

1346

NOËL ET AL.

FIG. 6. XopA and XopD are secreted by the Hrp TTSS. (A) Expression of XopA and XopD is hrpG dependent. pICxopA and pICxopD were conjugated into 85-10 and 85* to study the expression of XopA–c-myc and XopD–c-myc, respectively. Total protein extracts of X. campestris pv. vesicatoria grown in NYG were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 15 and 8% polyacrylamide for XopA and XopD, respectively) and analyzed by immunoblotting using an anti-c-myc antibody. The approximate sizes of the proteins are shown at the right. (B) Total protein extracts (10-fold concentrated) and supernatants (200-fold concentrated) of strains 85*⌬hrcV and 85*, containing pICxopA or pICxopD, were separated by SDS-PAGE (15 or 8% polyacrylamide for XopA and XopD, respectively) and analyzed by immunoblotting using the c-myc antibody. Membranes were reprobed with antiserum directed against cytoplasmic protein HrcN to assure that no bacterial lysis had occurred during the experiment (data not shown).

The findings on transcriptional regulation were confirmed by using GUS reporter constructs to monitor the promoter activity of xopA, xopD, and hpaF (Table 3). To study hpaF promoter activity, we used a Tn3-gus insertion (no. 462) in the 3⬘ region of hpaF, which was obtained in pXV4 (pXV4::462 [43]) and which was conjugated in 85-10 and 85*. For xopA and xopD, we used pIC1, a suicide vector, to simultaneously tag the C termini of the proteins with a triple-c-myc tag (see below) and generate a transcriptional fusion with a promoterless uidA reporter gene (34). The resulting constructs, pICxopA and pICxopD, were conjugated into 85*. To exclude the possibility that lack of expression is due to an aberrant insertion of the pIC constructs, 85-10::pICxopA and 85-10::pICxopD were generated by restoring the wild-type hrpG allele in 85*::pICxopA and 85*::pICxopD, respectively, using pOG (34). Analysis of GUS activities showed that xopA, xopD, and hpaF expression was 20, 8, and 40 times higher, respectively, in strains expressing hrpG* than in the hrpG background (Table 3). These results confirm the hrpG-dependent regulation of xopA, xopD, and hpaF (Fig. 5). Secretion of XopA and XopD by the Xanthomonas Hrp TTSS. Because it was proposed that Hpa1, the XopA homolog in X. oryzae pv. oryzae, acts as a harpin (48) (see Discussion) and because both XopD and HpaF contain putative NLSs, all three proteins are possible substrates for the TTSS. To allow detection, the proteins were tagged with a triple-c-myc epitope. As described above, integrative vector pIC1 allowed C-terminal tagging of XopA and XopD. To detect HpaF, a triple-c-myc tag was inserted after amino acid 55 and the gene was cloned into the pLAFR3 vector under the control of the lac promoter, which is constitutive in X. campestris pv. vesicatoria, giving pL3hpaFC. XopA and XopD expression was analyzed in

J. BACTERIOL.

strains 85* and 85*⌬hrcV, a strain, which carries a deletion in a conserved component of the TTSS. For technical reasons, expression studies of HpaF were carried out in 82* and derivatives (Table 1). Western blot analysis of bacterial total protein extracts revealed that, as expected from the promoter studies, XopA–cmyc and XopD–c-myc were not detectable in strain 85-10 grown in NYG (Fig. 6A). However, a protein of 18 kDa, corresponding to the expected size for XopA–c-myc (13 kDa plus a 5-kDa epitope), was detected in extracts of 85*::pICxopA. XopD–c-myc (predicted size: 70 plus 5 kDa) was detected at approximately 110 kDa in 85*::pICxopD. HpaF–c-myc (predicted size: 21 plus 5 kDa) could be visualized at approximately 33 kDa in total bacterial protein extracts (data not shown). To test for hrp-dependent secretion, bacteria were incubated in secretion medium. While HpaF could not be detected in culture supernatants (data not shown), XopA and XopD were present in culture supernatants of 85* strains (Fig. 6B). Since XopA and XopD were undetectable in culture supernatants of TTSS mutant 85*⌬hrcV, we conclude that XopA and XopD are secreted by the Hrp TTSS. DISCUSSION In this study we investigated the role of genes induced by hrpG and hrpX in the regions flanking the hrp gene cluster of X. campestris pv. vesicatoria and identified new Xop proteins and virulence factors. Nonpolar mutations in four genes revealed that xopA and hpaH contribute to bacterial growth and induction of plant reactions. The predicted HpaH protein belongs to the family of lysozyme-like proteins (33). The most interesting features of members of this protein family, besides an N-terminal signal peptide, are conserved lytic transglycosylase motifs and association with type III or type IV transport machinery. It is conceivable that the establishment of the type III transport machinery requires remodeling of the periplasmic peptidoglycans and that HpaH plays a role similar to that proposed for VirB1 from A. tumefaciens and related proteins (33). While analyses of mutants revealed reduced tumorigenesis by the A. tumefaciens virB1 mutant (7), mutations in hpaH and hpa2 (48) from Xanthomonas had only a weak effect (hpaH) or no effect (hpa2) on the interaction with the host. The second virulence factor, XopA, from X. campestris pv. vesicatoria is similar only to the predicted Hpa1 protein from X. oryzae pv. oryzae, which also plays a role in disease in susceptible plants and HR induction in resistant plants (48). Here we show that xopA encodes a protein that is secreted by the Hrp TTSS. This finding, together with the fact that xopA deletion mutants are reduced in virulence without affecting type III-dependent in vitro secretion, suggests that XopA could be involved in effector protein translocation into the host cell. Whether XopA plays a role before or after its secretion and whether it is itself translocated into the host cell remain to be elucidated. Interestingly, both the predicted Hpa1 protein from X. oryzae pv. oryzae and PopA, a secreted harpin from R. solanacearum (3), are very glycine rich (more than 20% of the amino acid residues), a feature that is typical for harpins from plant pathogenic bacteria (48). Harpins are secreted by the Hrp TTSS and trigger an HR-like response when infiltrated into nonhost plants (11), and recent data suggest an interaction

VOL. 184, 2002

of the P. syringae harpin with the plant plasma membrane (29, 30). Although XopA (8% glycine) is not homologous to harpins, it shares a lack of cysteine residues and an acidic isoelectric point with PopA and Hpa1 and is secreted by the Hrp system, which was also shown for PopA (3). Preliminary results indicate that infiltration of an XopA–glutathione Stransferase fusion protein into tobacco leaves did not induce an HR-like response. Another Xop identified in this study is XopD, which is similar to PsvA from P. syringae pv. eriobotryae but not to predicted proteins in the sequenced genomes of P. syringae pv. tomato DC3000 and R. solanacearum (http://www.tigr.org/cgi -bin/BlastSearch/blast.cgi?organism⫽p_syringae and http: //sequence.toulouse.inra.fr/ralsto/public/doc/RalstoForm .html). PsvA plays a role in virulence on loquat trees (26), whereas the pathogenicity of the X. campestris pv. vesicatoria xopD mutant was indistinguishable from that of the wild type under the conditions used. Interestingly, both XopD and PsvA contain a putative NLS, suggesting that they might be translocated to the plant nucleus to meet their virulence targets. Surprisingly, the xopD promoter contains an hrp box, which is typical for hrpL-regulated promoters in pathovars of P. syringae and Erwinia spp. (25, 47) and which is also found in the predicted promoter of the psvA gene from P. syringae pv. eriobotryae (GGAACCNA-N13-CCTACTA) 240 bp upstream of the predicted translation start codon. This is the first description of an hrp box in the promoter of an X. campestris pv. vesicatoria gene and is indicative of horizontal transfer of xopD from Pseudomonas or Erwinia spp. It would be interesting to know whether xanthomonads contain hrpL homologs and whether xopD could be regulated by hrpL in P. syringae. On the other side of the hrp gene cluster, downstream of hrpF, we identified ORFs hpaF and hpaG, which belong to the hrpG regulon. Since deletion mutations have no effect, their role for the bacterium-plant interaction is unknown. Both genes are homologous to the single hpaF gene in X. oryzae pv. oryzae, which is predicted to encode a protein with nine LRRs, a feature found in many eukaryotic proteins but also in a number of type III effector proteins, e.g., PopC from R. solanacearum (21). The presence of LRRs in the predicted X. oryzae pv. oryzae HpaF protein is suggestive of secretion by the Hrp system. However our attempts to detect HpaG-His6 in X. campestris pv. vesicatoria were unsuccessful, suggesting that the hpaG RNA is not translated under the conditions used (F. Thieme and U. Bonas, unpublished data). DNA sequence comparison of the X. campestris pv. vesicatoria and X. oryzae pv. oryzae hrp gene clusters and flanking regions revealed almost identical sequences in the core hrp gene clusters and downstream of hrpF. While the regions flanking hrpA in X. campestris pv. campestris and X. oryzae pv. oryzae show very similar organizations (M. Arlat, personal communication), the orthologous region in X. campestris pv. vesicatoria has a scrambled organization, probably due to deletions and insertions. The G⫹C contents of xopA (50%) and xopD (54%) are significantly lower than that of the hrp gene cluster (64%), suggesting that xopA and xopD might have been acquired by horizontal transfer. In addition, the Xanthomonas regions flanking the hrp gene cluster contain genes involved in virulence (xopA and hpaH), a tRNA gene, and mobile genetic elements (ISXc 7, IS1114, transposase coding sequence). Tak-

hrp PATHOGENICITY ISLAND

1347

ing into account that mobility of the hrp gene cluster has been observed in X. campestris pv. vesicatoria (6), the region containing the hrp genes fulfills the criteria of PAIs (22). Similar conclusions were drawn for P. syringae (1). Sequencing orthologous regions from different P. syringae pathovars revealed an exchangeable effector locus (EEL) differing in G⫹C and gene content and containing sequences related to mobile genetic elements, whereas the region on the other side of the hrp gene cluster is highly conserved (conserved effector locus; CEL) (1). Genome sequencing of different xanthomonads will show whether the EEL/CEL concept is also valid for Xanthomonas spp. and will address questions of diversity and dynamics of the PAIs. ACKNOWLEDGMENTS We are grateful to J. Mansfield, D. Büttner, R. Koebnik, and T. Lahaye for critically reading the manuscript and to M. Arlat for communicating unpublished sequences. We thank C. Kretschmer and A. Landgraf for excellent technical assistance and E. Huguet for identifying hpaG and hpaI. L.N. was supported in part by the École Normale Supérieure de Lyon. This work was funded by a grant from the Deutsche Forschungsgemeinschaft (SFB 363) to U.B. REFERENCES 1. Alfano, J. R., A. O. Charkowski, W. L. Deng, J. L. Badel, T. Petnicki-Ocwieja, K. van Dijk, and A. Collmer. 2000. The Pseudomonas syringae Hrp pathogenicity island has a tripartite mosaic structure composed of a cluster of type III secretion genes bounded by exchangeable effector and conserved effector loci that contribute to parasitic fitness and pathogenicity in plants. Proc. Natl. Acad. Sci. USA 97:4856–4861. 2. Alfano, J. R., and A. Collmer. 1997. The type III (Hrp) secretion pathway of plant pathogenic bacteria: trafficking harpins, Avr proteins, and death. J. Bacteriol. 179:5655–5662. 3. Arlat, M., F. Van Gijsegem, J. C. Huet, J. C. Pernollet, and C. A. Boucher. 1994. PopA1, a protein which induces a hypersensitivity-like response on specific Petunia genotypes, is secreted via the Hrp pathway of Pseudomonas solanacearum. EMBO J. 13:543–553. 4. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1996. Current protocols in molecular biology. John Wiley & Sons, Inc., New York, N.Y. 5. Ballvora, A., M. Pierre, G. van den Ackerveken, S. Schornack, O. Rossier, M. Ganal, T. Lahaye, and U. Bonas. 2001. Genetic mapping and functional analysis of the tomato Bs4 locus, governing recognition of the Xanthomonas campestris pv. vesicatoria AvrBs4 protein. Mol. Plant-Microbe Interact. 14: 629–638. 6. Basim, H., R. E. Stall, G. V. Minsavage, and J. B. Jones. 1999. Chromosomal gene transfer by conjugation in the plant pathogen Xanthomonas axonopodis pv. vesicatoria. Phytopathology 89:1044–1049. 7. Berger, B. R., and P. J. Christie. 1994. Genetic complementation analysis of the Agrobacterium tumefaciens virB operon: virB2 through virB11 are essential virulence genes. J. Bacteriol. 176:3646–3660. 8. Bogdanove, A., S. V. Beer, U. Bonas, C. A. Boucher, A. Collmer, D. L. Coplin, G. R. Cornelis, H.-C. Huang, S. W. Hutcheson, N. J. Panopoulos, and F. Van Gijsegem. 1996. Unified nomenclature for broadly conserved hrp genes of phytopathogenic bacteria. Mol. Microbiol. 20:681–683. 9. Bonas, U., R. Schulte, S. Fenselau, G. V. Minsavage, B. J. Staskawicz, and R. E. Stall. 1991. Isolation of a gene-cluster from Xanthomonas campestris pv. vesicatoria that determines pathogenicity and the hypersensitive response on pepper and tomato. Mol. Plant-Microbe Interact. 4:81–88. 10. Bonas, U., R. E. Stall, and B. Staskawicz. 1989. Genetic and structural characterization of the avirulence gene avrBs3 from Xanthomonas campestris pv. vesicatoria. Mol. Gen. Genet. 218:127–136. 11. Cornelis, G. R., and F. Van Gijsegem. 2000. Assembly and function of type III secretory systems. Annu. Rev. Microbiol. 54:735–774. 12. Daniels, M. J., C. E. Barber, P. C. Turner, M. K. Sawczyc, R. J. W. Byrde, and A. H. Fielding. 1984. Cloning of genes involved in pathogenicity of Xanthomonas campestris pv. campestris using the broad host range cosmid pLAFR1. EMBO J. 3:3323–3328. 13. Ditta, G., S. Stanfield, D. Corbin, and D. Helinski. 1980. Broad host range DNA cloning system for gram-negative bacteria: construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 77:7347–7351. 14. Escolar, L., G. Van den Ackerveken, S. Pieplow, O. Rossier, and U. Bonas. 2001. Type III secretion and in planta recognition of the Xanthomonas avirulence proteins AvrBs1 and AvrBsT. Mol. Plant Pathol. 2:287–296.

1348

NOËL ET AL.

15. Fenselau, S., I. Balbo, and U. Bonas. 1992. Determinants of pathogenicity in Xanthomonas campestris pv. vesicatoria are related to proteins involved in secretion in bacterial pathogens of animals. Mol. Plant-Microbe Interact. 5:390–396. 16. Fenselau, S., and U. Bonas. 1995. Sequence and expression analysis of the hrpB pathogenicity operon of Xanthomonas campestris pv. vesicatoria which encodes eight proteins with similarity to components of the Hrp, Ysc, Spa, and Fli secretion systems. Mol. Plant-Microbe Interact. 8:845–854. 17. Figurski, D., and D. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76:1648–1652. 18. Galán, J. E., and A. Collmer. 1999. Type III secretion machines: bacterial devices for protein delivery into host cells. Science 284:1322–1328. 19. Gaudriault, S., L. Malandrin, J. P. Paulin, and M. A. Barny. 1997. DspA, an essential pathogenicity factor of Erwinia amylovora showing homology with AvrE of Pseudomonas syringae, is secreted via the Hrp secretion pathway in a DspB-dependent way. Mol. Microbiol. 26:1057–1069. 20. Gelvin, S. B. 2000. Agrobacterium and plant genes involved in T-DNA transfer and integration. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51:223–256. 21. Guénéron, M., A. C. Timmers, C. Boucher, and M. Arlat. 2000. Two novel proteins, PopB, which has functional nuclear localization signals, and PopC, which has a large leucine-rich repeat domain, are secreted through the Hrp-secretion apparatus of Ralstonia solanacearum. Mol. Microbiol. 36:261– 277. 22. Hacker, J., and J. B. Kaper. 2000. Pathogenicity islands and the evolution of microbes. Annu. Rev. Microbiol. 54:641–679. 23. Huguet, E., and U. Bonas. 1997. hrpF of Xanthomonas campestris pv. vesicatoria encodes an 87-kDa protein with homology to NolX of Rhizobium fredii. Mol. Plant-Microbe Interact. 10:488–498. 24. Huguet, E., K. Hahn, K. Wengelnik, and U. Bonas. 1998. hpaA mutants of Xanthomonas campestris pv. vesicatoria are affected in pathogenicity but retain the ability to induce host-specific hypersensitive reaction. Mol. Microbiol. 29:1379–1390. 25. Innes, R. W., A. F. Bent, B. N. Kunkel, S. R. Bisgrove, and B. J. Staskawicz. 1993. Molecular analysis of avirulence gene avrRpt2 and identification of a putative regulatory sequence common to all known Pseudomonas syringae avirulence genes. J. Bacteriol. 175:4859–4869. 26. Kamiunten, H. 1999. Isolation and characterization of virulence gene psvA on a plasmid of Pseudomonas syringae pv. eriobotryae. Ann. Phytopathol. Soc. Jpn. 65:501–509. 27. Klement, Z. 1982. Hypersensitivity, p. 149–177. In M. S. Mount and G. H. Lacy (ed.), Phytopathogenic prokaryotes, vol. 2. Academic Press, New York, N.Y. 28. Knoop, V., B. Staskawicz, and U. Bonas. 1991. Expression of the avirulence gene avrBs3 from Xanthomonas campestris pv. vesicatoria is not under the control of hrp genes and is independent of plant factors. J. Bacteriol. 173: 7142–7150. 29. Lee, J., D. F. Klessig, and T. Nürnberger. 2001. A harpin binding site in tobacco plasma membranes mediates activation of the pathogenesis-related gene hin1 independent of extracellular calcium but dependent on mitogenactivated protein kinase activity. Plant Cell 13:1079–1093. 30. Lee, J., B. Klu ¨sener, G. Tsiamis, C. Stevens, C. Neyt, A. P. Tampakaki, N. J. Panopoulos, J. Noller, E. W. Weiler, G. R. Cornelis, J. W. Mansfield, and T. Nürnberger. 2001. HrpZPsph from the plant pathogen Pseudomonas syringae pv. phaseolicola binds to lipid bilayers and forms an ion-conducting pore in vitro. Proc. Natl. Acad. Sci. USA 98:289–294. 31. Lowe, T. M., and S. R. Eddy. 1997. tRNAscan-SE: a program for improved

J. BACTERIOL.

32.

33. 34. 35. 36. 37.

38.

39. 40. 41. 42. 43. 44.

45. 46. 47.

48.

detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25:955–964. Meinhardt, L. W., H. B. Krishnan, P. A. Balatti, and S. G. Pueppke. 1993. Molecular cloning and characterization of a sym plasmid locus that regulates cultivar-specific nodulation of soybean by Rhizobium fredii USDA257. Mol. Microbiol. 9:17–29. Mushegian, A. R., K. J. Fullner, E. V. Koonin, and E. W. Nester. 1996. A family of lysozyme-like virulence factors in bacterial pathogens of plants and animals. Proc. Natl. Acad. Sci. USA 93:7321–7326. Noël, L., F. Thieme, D. Nennstiel, and U. Bonas. 2001. cDNA-AFLP analysis unravels a genome-wide hrpG-regulon in the plant pathogen Xanthomonas campestris pv. vesicatoria. Mol. Microbiol. 41:1271–1281. Rossier, O., G. Van den Ackerveken, and U. Bonas. 2000. HrpB2 and HrpF from Xanthomonas are type III-secreted proteins and essential for pathogenicity and recognition by the host plant. Mol. Microbiol. 38:828–838. Rossier, O., K. Wengelnik, K. Hahn, and U. Bonas. 1999. The Xanthomonas Hrp type III system secretes proteins from plant and mammalian pathogens. Proc. Natl. Acad. Sci. USA 96:9368–9373. Schaffer, A. A., L. Aravind, T. L. Madden, S. Shavirin, J. L. Spouge, Y. I. Wolf, E. V. Koonin, and S. F. Altschul. 2001. Improving the accuracy of PSI-BLAST protein database searches with composition-based statistics and other refinements. Nucleic Acids Res. 29:2994–3005. Schulte, R., and U. Bonas. 1992. Expression of the Xanthomonas campestris pv. vesicatoria hrp gene cluster, which determines pathogenicity and hypersensitivity on pepper and tomato, is plant inducible. J. Bacteriol. 174:815– 823. Staskawicz, B. J., D. Dahlbeck, N. Keen, and C. Napoli. 1987. Molecular characterization of cloned avirulence genes from race0 and race1 of Pseudomonas syringae pv. glycinea. J. Bacteriol. 169:5789–5794. Tatusova, T. A., and T. L. Madden. 1999. Blast 2 sequences–-a new tool for comparing protein and nucleotide sequences. FEMS Microbiol. Lett. 174: 247–250. Van den Ackerveken, G., E. Marois, and U. Bonas. 1996. Recognition of the bacterial avirulence protein AvrBs3 occurs inside the host plant cell. Cell 87:1307–1316. Vieira, J., and J. Messing. 1987. Production of single-stranded plasmid DNA. Methods Enzymol. 153:3–11. Wengelnik, K., and U. Bonas. 1996. HrpXv, an AraC-type regulator, activates expression of five of the six loci in the hrp cluster of Xanthomonas campestris pv. vesicatoria. J. Bacteriol. 178:3462–3469. Wengelnik, K., C. Marie, M. Russel, and U. Bonas. 1996. Expression and localization of HrpA1, a protein of Xanthomonas campestris pv. vesicatoria essential for pathogenicity and induction of the hypersensitive reaction. J. Bacteriol. 178:1061–1069. Wengelnik, K., O. Rossier, and U. Bonas. 1999. Mutations in the regulatory gene hrpG of Xanthomonas campestris pv. vesicatoria result in constitutive expression of all hrp genes. J. Bacteriol. 181:6828–6831. White, F. F., B. Yang, and L. B. Johnson. 2000. Prospects for understanding avirulence gene function. Curr. Opin. Plant Biol. 3:291–298. Xiao, Y., and S. W. Hutcheson. 1994. A single promoter sequence recognized by a newly identified alternate sigma factor directs expression of pathogenicity and host range determinants in Pseudomonas syringae. J. Bacteriol. 176:3089–3091. (Erratum, 176:6158.) Zhu, W. G., M. M. Magbanua, and F. F. White. 2000. Identification of two novel hrp-associated genes in the hrp gene cluster of Xanthomonas oryzae pv. oryzae. J. Bacteriol. 182:1844–1853.