JOURNAL OF BACrERIOLOGY, Mar. 1992, p. 1923-1931

Vol. 174, No. 6

0021-9193/92/061923-09$02.00/0

Copyright X 1992, American Society for Microbiology

Cloning and Characterization of Pathogenicity Genes from Xanthomonas campestris pv. glycines INGYU HWANG,' SUNG M. LIM,1'2 AND PAUL D. SHAW`* Department of Plant Pathology, University of Illinois at Urbana-Champaign,1 and Agricultural Research Service, U.S. Department ofAgriculture,2 Urbana, Illinois 61801 Received 13 September 1991/Accepted 13 January 1992

Nonpathogenic mutants ofXanthomonas campestris pv. glycines 8ra were generated with N-methyl-N-nitroN'-nitrosoguanidine to identify and characterize pathogenicity genes of the bacterium. A total of 16 nonpathogenic mutants were isolated from 2,000 colonies. One mutant, NP1, was chosen for further study. NP1 did not multiply in soybean cotyledons. A genomic library of strain 8ra was constructed in the cosmid pLAFR3, and the cosmids were tested for complementation in NP1. One cosmid clone, pIHl, which contained a 31-kb insert, complemented mutant NP1. A restriction map of plHl was constructed, and deletion analyses identified a 10-kb HindllI fragment that restored pathogenicity to NP1. Southern hybridization analysis indicated that DNA sequences in the 10-kb Hindlll fragment are conserved among other X. campestris pathovars tested. Three regions responsible for restoring pathogenicity have been identified by Tn3-HoHol mutagenesis. A 2.7-kb ClaI fragment was sequenced, and two possible open reading frames (ORF1 and ORF2) were found. Results indicated that ORF2 but not ORF1 may be expressed in Escherichia coli and in X. campestris pv. glycines. The carboxy terminus of the potential polypeptide encoded by ORF2 has an amino acid sequence similar to that of the gamma subunit of oxaloacetate decarboxylase, which is involved in sodium ion transport in Kiebsiella pneumoniae.

that are involved in pathogenicity. Preliminary reports of this work have appeared elsewhere (17, 18).

Xanthomonas campestris pv. glycines causes bacterial pustules on susceptible soybean cultivars. The disease symptoms are due primarily to hypertrophy of host mesophyll cells (20), which can result in premature defoliation of infected plants. Yields are lowered because of reduced seed size (47), and under certain environmental conditions, serious economic losses can occur. The mechanisms used by phytopathogenic bacteria to invade plants are very complex and poorly understood. Physiological studies to identify factors involved in diseases caused by X. campestris pathovars have given largely inconclusive or negative results. For example, work from Daniels' laboratory (8, 45) indicated that proteases, cellulases, and pectinases do not play a direct role in the diseases caused by X. campestris pv. campestris. The possible role of auxins in causing bacterial pustule has been investigated previously (12), but it appears unlikely that such hormones play a major role in the development of the disease. X. campestris pv. glycines is capable of producing indoleacetic acid from tryptophan; however, the hormone is produced in vitro by both pathogenic and nonpathogenic strains (12). Pathogenicity genes of X. campestris pv. campestris have been isolated by others (7, 46), and TnS mutagenesis and marker exchange allowed them to define a 10-kb region containing a cluster of genes involved in pathogenicity (46). It was suggested that the pathogenicity genes in the region may be involved in the secretion of proteins from X. campestns pv. campestris by controlling translocation across the outer membrane (9). Avirulence genes from X. campestris pv. vesicatoria (3), X. campestris pv. malvacearum (14), and X. campestris pv. oryzae (22) have been isolated. Genes that determine a host range have been cloned from X. campestris pv. translucens (28). This paper describes research to identify and characterize genes of X. campestris pv. glycines 8ra *

MATERIALS AND METHODS Bacterial strains and plasmids. The bacterial strains and plasmids used are listed in Table 1. All X. campestris pv. glycines strains are derivatives of strain 8ra. Media, antibiotics, and other chemicals. Strains of X. campestrs pv. glycines were cultured on YDC agar medium (37) or in L broth (26) at 30°C. All Escherichia coli strains were grown in L agar or L broth at 37°C. The concentrations of antibiotics were as follows: rifampin, 50 ,ug/ml; tetracycline, 15 ,g/ml for E. coli and 5 p,g/ml for X. campestris pv. glycines; and ampicillin, 50 p,g/ml. All antibiotics, N-methylN-nitro-N'-nitrosoguanidine (NTG), and Sephadex G-50 were purchased from Sigma Chemical Co. (St. Louis, Mo.). Restriction enzymes, T4 DNA ligase, bacterial alkaline phosphatase, 5-bromo4-chloro-3-indolyl-,B-D-galactoside (X-Gal), isopropylthio-p-galactoside (IPTG), and a nick translation kit were obtained from Bethesda Research Laboratories (Gaithersburg, Md.). [a-32P]dCTP (800 Ci/mmol) was purchased from Amersham International (Arlington Heights, Ill.). DNA sequencing kits, TAQuence and Sequenase, were purchased from United States Biochemical Corporation

(Cleveland, Ohio). Plant inoculations. Pathogenicity assays and growth studies in planta were done as described previously (19). Tobacco plants (Nicotiana tabacum cv. Havana 38) were used for hypersensitivity tests. Bacterial suspensions (108 cells per ml) of X. campestris pv. glycines 8ra and NP1 and Pseudomonas syringae pv. glycinea 61 were prepared and injected into fully expanded leaves of whole plants. The reactions of the plants were determined 24 h after inoculation. Chemical mutagenesis. One milliliter of an overnight culture of X. campestris pv. glycines 8ra was centrifuged at

Corresponding author. 1923

J. BACTERIOL.

HWANG ET AL.

1924

TABLE 1. Bacterial strains and plasmids used in this study Strain, plasmid, or

Relevant characteristic(s)a Rlvnchrceitcs

phage

E. coli HB101

DH5a

Reference or

sourceb

F- hsdS20(rB mB) recAl3 leuB6 ara-14proA2 lacYl galK2 rpsL20(Strr) xyl-S mtl-l supESS F- 480d lacZAM15 endA41 recAl hsdRJ7(rK- mK-) supE44 thi-I gyrA96 A(lacZYA-

26 BRL

argF)U169 C2110 HB101(pHoHol, pSShe) MV1304 X. campestris pv. glycines 8ra NP1, I-3, 1-8, I-9, I10, I-11, I-13, I-14, III-1, III-2, III-3,

Nalr poL4

Ampr Cmr A(srl-recA)306::TnlO thi rpsL endA sbcB15 hsdR4 A(lac-proAB) (F' traD36 proAB lacIq lacZAM15

42 42 IBI

Parent strain, Rifr Same as 8ra but nonpathogenic

E. J. Braun This study

Pathogenic Pathogenic

W. Fett This study

Pathogen of crucifers

D. Gabriel

Pathogen of rice, race 1, Rif'

Y. S. Cho

Pathogen of pepper, pepper race 2

D. Gabriel

Pathogen of cotton

D. Gabriel

Pathogen of soybean

S. M. Lim

Pathogen of soybean and tobacco

P. D. Shaw

Derivative of pRK290, Tetr Derivative of pRK290 with TrVx polylinker, Tetr Kmr, Tra+, ColEl replicon Ampr, promoterless cat cassette 31-kb DNA fragment from strain 8ra cloned into pLAFR3 HindIII deletion of pIHl BamHI deletion of pIHl EcoRI deletion of pIHl 10-kb HindIII fragment from pIH1 subcloned into pLAFR3 2.7-kb ClaI fragment from pIH7 subcloned into pWB5A 2.1-kb BamHI-XbaI fragment from pIH8 subcloned into pWB5A 2.7-kb EcoRI fragment from pIHlO cloned into pTZ18 Same as pSR2 but has different orientation 2.7-kb EcoRI fragment from pIHlO cloned into pTZ19 1.0-kb SmaI fragment deleted from pSR2 1.7-kb SmaI fragment deleted from pSR3 1.7-kb SmaI fragment deleted from pSR5 1.7-kb EcoRI-SalI fragment from pSR21 cloned into M13mpl8 1.0-kb EcoRI-SalI fragment from pSR31 cloned into M13mpl8 1.7-kb EcoRI-SalI fragment from pSR21 cloned into M13mpl9 1.0-kb EcoRI-SalI fragment from pSR31 cloned into M13mpl9 cat cassette inserted in BalI site of pSR21 Same as pSR21a but has different orientation cat cassette inserted in BglII site of pSR51 1.6-kb ClaI-XbaI fragment containing cat cassette from pSR5la cloned into pWB5A

43 T. Jacobs 13 40 27 27 This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study

Bacteriophage cloning vector used to isolate single-stranded DNA

29

III-5, III-6, IV-3, IV-6, V-5 B99 8601

Other bacteria X. campestris pv. campestris 528T X. campestris pv. oryzae 8206 X. campestris pv. vesicatoria 88-5 X campestris pv. malvacearum strain H P. syringae pv. glycinea 61 P. syringae pv. tabaci PT15 Plasmids pLAFR3 pWB5A pRK2013 pJS133 pTZ18 pTZ19 pIH1 pIH5 pIH6 pIH7 pIH8 pIH10 pIH11 pSR2 pSR3 pSR5 pSR21

pSR31 pSR51 pSR10 pSRll pSR12 pSR13 pSR21a pSR21b pSR5la pSR200a Phages M13mpl8 and M13mpl9

Amppr Ampr

a Abbreviations: Strr, streptomycin resistance. Nalr, nalidixic acid resistance; Ampr, ampicillin resistance; resistance; Tetr, tetracycline resistance; Kmr, kanamycin resistance. b BRL, Bethesda Research Laboratories. IBI, International Biotechnologies, Inc.

Cmr, chloramphenicol resistance; Rifr, rifampicin

VOL. 174, 1992

PATHOGENICITY GENES FROM X. CAMPESTRIS PV. GLYCINES

14,000 x g for 1 min, and the cells were washed with 0.1 M sodium citrate-citric acid buffer (pH 5.5) and resuspended in 1 ml of the same buffer. NTG was added to the bacterial suspension in the culture tube to a final concentration of 10 ,ug/ml, and the tube was incubated for 20 min at room temperature. These conditions caused 50 to 70% killing of the bacterial cells. The bacterial pellets were washed twice with 0.1 M sodium phosphate buffer (pH 7.0) and resuspended in 5 ml of L broth, and the cells were incubated for 24 h at 25°C to allow growth and segregation of the mutants. The cells were transferred to petri plates containing YDC agar medium, and individual colonies were isolated after incubating overnight at 30°C. These colonies were replica plated onto YDC agar medium and M9 minimal agar medium (30) to isolate prototrophic colonies. Colorless mutants were identified on YDC agar medium. Protease, cellulase, and polygalacturonate lyase assays. The protease, cellulase, and polygalacturonate lyase activities of each mutant and of the parent strain were determined as described by Daniels and coworkers (8, 45). DNA isolations. Total genomic DNA was isolated by the method of Chesney et al. (6), and plasmid DNA was isolated by a sodium dodecyl sulfate-alkaline lysis method (2). Recombinant DNA techniques. Restriction endonuclease digestions and DNA ligations were performed as described by the manufacturer. Competent cells of E. coli were prepared with calcium chloride and subsequently transformed as described by Maniatis et al. (26). For agarose-gel electrophoresis, either TBE (26) or glycine buffer (200 mM glycine, 15 mM NaOH, 4 mM Na2 EDTA [final pH 9.0]) was used. GeneScreen membranes were used as described by the manufacturer (Dupont, NEN Research Products, Boston, Mass.) for Southern hybridizations (26, 41). To construct a genomic library of parent strain 8ra, total genomic DNA of the strain was isolated, partially digested with Sau3A, and fractionated in sodium chloride gradients (26). Cosmid vector pLAFR3 was completely digested with BamHI, dephosphorylated with bacterial alkaline phosphatase as described by the manufacturer, and ligated to the pooled DNA fragments with T4 DNA ligase. In vitro packaging and transfection into E. coli HB101 were by the method described by the manufacturer (Promega Co., Madison, Wis.). After overnight incubation of the transductants, the pooled genomic library was stored at -70°C in 20%

glycerol. pJS133 (40) DNA was digested with appropriate restriction enzymes to isolate a promoterless cat (chloramphenicol acetyltransferase) cassette DNA by electroelution as described by Maniatis et al. (26). Mating. pLAFR3 was mobilized from E. coli into X. campestris pv. glycines strains as described by Daniels et al. (7), except that strains of X. campestris pv. glycines and E. coli were grown for 6 h at 30°C and for 2 h at 37°C, respectively, before mating. Tn3-HoHol mutagenesis of pIH8. Tn3-HoHol mutagenesis of pIH8 was done as described by Stachel et al. (42). Plasmids isolated from pooled transconjugants were transformed into E. coli DHc5x, and transformants were selected on L agar containing tetracycline and ampicillin. Tn3HoHol insertion sites were determined by digesting plasmid DNA with EcoRI and BamHI. The production of ,-galactosidase by E. coli DH5ao and by an X. campestris pv. glycines strain, NP1, was measured qualitatively on L agar medium and on M9 minimal agar medium containing X-Gal (0.02 ,ug/ml) with and without IPTG (0.005 ,g/ml). DNA sequencing. Both strands of a 2.7-kb ClaI fragment

A

B

C

1925

D

FIG. 1. Soybean cotyledon bioassay. Soybean cotyledons of the susceptible cultivar Pella (10 days old) were inoculated with water (A), X. campestris pv. glycines parent strain 8ra (B), nonpathogenic mutant NP1 (C), and NP1 containing cosmid clone pIH1 (D). Symptoms were detectable 3 to 5 days after inoculation. The cotyledons were photographed 5 days after inoculation.

were sequenced by the Sanger dideoxy method (36) with TAQence or Sequenase kits. DNA sequences were analyzed by the Mount and Conrad (32) program to find possible open reading frames (ORFs) and putative amino acid sequences of

polypeptides encoded by the ORFs. Predicted amino acid sequences were analyzed by using the SEQANAL program developed by A. Croft (6a). Hydropathy plots were obtained by the algorithm of Kyte and Doolittle (23). The GenBank data base (release 61.0) and the EMBL nucleotide data base (release 22) were searched to find significant homology with known DNA sequences. Both the Genbank data base and the Swiss-Prot protein sequence data base (release 13) were searched by using FASTA (34) and PATMAT (16) programs, respectively, to find whether the two putative polypeptides had any similarity to known proteins. Nucleotide sequence accession number. The sequence shown in Fig. 5 has been assigned GenBank accession no. M64094. RESULTS Generation of nonpathogenic mutants. Following treatment of X. campestns pv. glycines 8ra with NTG, 2,000 separate colonies were chosen at random and tested for pathogenicity on soybean cotyledons. Sixteen nonpathogenic mutants (0.8%) were isolated (Table 1). A soybean cotyledon inoculated with one of the nonpathogenic mutants (NP1) is shown in Fig. 1. Mutants that failed to produce the characteristic yellow pigment in culture and auxotrophic mutants were obtained at a frequency of about 1.0%. The parental strain (8ra) and all nonpathogenic mutants produced a yellow pigment, were prototrophic, and formed smooth colonies. Like strain 8ra, the 16 nonpathogenic mutants produced protease but not cellulase or pectatedegrading enzymes (data not shown). When NP1 was inoculated into a susceptible soybean cultivar, the number of cells remained constant for about 7 days and then declined to approximately 10% of that present at the time of inoculation. In contrast, the number of cells of the parent strain (8ra) increased about 10-fold during the first 6 days after inoculation and then remained nearly constant (Fig. 2). Neither 8ra nor NP1 was able to induce a hypersensitive reaction on tobacco; however, P. syringae pv. glycinea 61, a positive control, gave a typical hypersensitive reaction 16 to 24 h after being infiltrated into tobacco leaves (data not shown). Complementation. By transfer en masse, the library was able to complement the leu phenotype of E. coli HB101 (leu pro) at a frequency of 5 x 10-. When the genomic library was transferred en masse into a colorless mutant of X.

J. BAcrERIOL.

HWANG ET AL.

1926

9

S 8,

Days after inoculation FIG. 2. Growth of X. campestris pv. glycines parent strain 8ra, nonpathogenic mutant NP1, and NPl(pIH1) in soybean cotyledons. Bacterial numbers were determined daily after inoculation. Data are averages of three samples. Vertical bars indicate error ranges.

campestris pv. glycines by mating, 1 of 500 transconjugants of the mutant showed complementation of the mutant phenotype (data not shown). The library was transferred en masse into mutant NP1 by mating, and 500 separate transconjugants were chosen at random for pathogenicity assays. The nonpathogenic mutant phenotype was complemented in one of these transconjugants. The cosmid from the single transconjugant (pIHl) was isolated, transferred into E. coli HB101 by transformation, and mobilized back into NP1 to confirm the complementation. A total of 10 NP1 transconjugants, chosen at random and inoculated into soybean cotyledons, had all regained pathogenicity. The bioassay of one of these is shown in Fig. 1. pIH1 also restored the ability of NP1 to multiply in soybean cotyledons (Fig. 2). pIHl complemented 3 other nonpathogenic mutants (IV-3, IV-6, V-5) but failed to restore pathogenicity to the other 12 mutants. Deletion analysis and subcloning. A restriction map of pIH1 was constructed with 10 different restriction enzymes (Fig.

3). The derived physical map indicates that the size of the insert is approximately 31 kb. Southern hybridization results confirmed that the fragments in the insert were colinear with the X. campestris pv. glycines genome (data not shown). Deletions were made in pIH1 to define regions within the 31-kb insert that were responsible for the restoration of pathogenicity to NP1. pIH1 DNA was digested with HindIII, EcoRI, or BamHI and religated to yield pIH5, pIH6, and pIH7 (Fig. 3). The 10-kb HindIII fragment at the left end of the 31-kb pIH1 insert was subcloned into pLAFR3, and the recombinant plasmid was designated pIH8 (Fig. 3). The four plasmids pIH5, pIH6, pIH7, and pIH8 (Fig. 3) were each introduced into the four nonpathogenic mutants, NP1, IV-3, IV-6, and V-5, by mating. Transconjugants carrying the deletion plasmids were inoculated into soybean cotyledons. pIH5 did not restore pathogenicity to any of the mutants. Mutants NP1 and V-5, containing pIH6 or pIH7, occasionally showed very weak symptoms that appeared only 7 to 10 days after inoculation (Fig. 3). pIH8 was able to restore pathogenicity to NP1 and V-5 but not to IV-3 and IV-6 (Fig. 3). These results suggest that genes responsible for restoration of pathogenicity to mutants NP1 and V-5 are located on the 10-kb HindIII fragment contained in pIH8. Homology of the 10-kb HindmI fragment with DNA from other xanthomonads and pseudomonads. The 10-kb HindIII fragment from pIH8 was hybridized to EcoRI restriction fragments of DNA from X. campestris pathovars glycines, vesicatoria, malvacearum, oryzae, and campestris and P. syringae pathovars glycinea and tabaci DNA that had been separated by agarose gel electrophoresis. In addition to two predicted EcoRI fragments, of 9 kb and 6 kb, from X. campestris pv. glycines 8ra, the 10-kb fragment hybridized to a 22-kb fragment. These three bands were detected in digests of DNA from all X. campestris pv. glycines isolates. Hybridization to EcoRI fragments from X. campestis pathovars vesicatoria, malvacearum, oryzae, and campestris was also detected, although the fragment sizes were different. No hybridization to digested DNA of P. syringae pathovars glycinea and tabaci was detected (data not shown). Tn3-HoHol mutagenesis of p1H8. To define further the location of the pathogenicity genes present in the pIH8 insert and to begin preliminary experiments to determine the regulatory properties of the genes, pIH8 in E. coli C2110 was subjected to Tn3-HoHol mutagenesis. A total of 36 random Tn3-HoHol insertions were mapped within the

Resoation of

Ial plHl

a I

ill]

M I I

I

I

I I

It if

Ii

IU2 I

.

C02 Ca I

I

I

I

paIhogniciton NP1 IV3 IV6 V5

I

p115 p1I6

I

-

pIH

pnIH

U

3kb

46

-

4

4E

-

4/

+

.

.

+

FIG. 3. Deletion analysis of pIHl. pIHl DNA was digested with HindIII, BamHI, or EcoRI and religated to yield pIH5, pIH6, and pIH7, respectively. The 10-kb HindIII fragment from pIH1 was subcloned into pLAFR3 to produce pIH8. The four plasmids pIH5, pIH6, pIH7, and pIH8 were each introduced into the four nonpathogenic mutants, NP1, IV-3, IV-6, and V-5, by mating. Transconjugants carrying the deletion plasmids were inoculated into soybean cotyledons. The transconjugants produced (+) or did not produce (-) symptoms on soybean cotyledons or produced weak symptoms 7 to 10 days after inoculation (+/-).

PATHOGENICITY GENES FROM X. CAMPESTRIS PV. GLYCINES

VOL. 174, 1992 H A

CV

M

M

CV

lS SG C

- -

+ +

+

RT

S

SB

- - -

--

+ + +

-- + -

-

-

-

++

X

62B2

S8

S6

B-Galactosidase Complementation

1927

S

H

-

+ + .__

+

++

+

+

_

_-_

++ +I

+

+

1 kb FIG. 4. Mutagenesis of pIH8 with Tn3-HoHol. Tn3-HoHol insertion sites and their orientations were determined by restriction endonuclease digestion. ,3-Galactosidase activity in E. coli DH5a and X. campestris pv. glycines NP1 containing plasmids with transposons in the inserts was determined on complex and minimal media (M9). Each pIH8::Tn3-HoHol was introduced into the nonpathogenic mutant NP1 by mating. The transconjugants produced (+) or did not produce (-) symptoms on soybean cotyledons or ,B-galactosidase activity. The numbers are described in the text. H, HindIII; A, Sall; M, SmaI; S, BstEII; R, EcoRI; T, SstI; B, BamHI; X, XbaI; C, ClaI; G, BglII; V, EcoRV.

10-kb HindIII fragment, and 26 independent inserts were identified (Fig. 4). Plasmids containing these Tn3-HoHol insertions were introduced into mutant NP1 by mating, and the transconjugants were tested for pathogenicity (Fig. 4). Plasmids carrying insertions 56, 68, 17, 29, 38, and B35 had lost their ability to complement the nonpathogenic phenotype. These results indicate that three regions, two of which are in a 2.7-kb ClaI fragment and one of which is in a 2.1-kb BamHI-XbaI fragment, are important for pathogenicity. The 2.7-kb ClaI fragment, containing the regions defined by mutants 65 through 24, and the 2.1-kb BamHI-XbaI fragment, containing the sites of Tn3-HoHol insertions in mutants 7 through 16, were individually cloned into pWB5A, and the recombinant plasmids were designated pIH10 and pIH11, respectively. Plasmids pIH10 and pIH11 were mobilized into NP1, but neither was able to complement the mutant phenotype. The expression of ,-galactosidase activity by the 26 X. campestris pv. glycines transconjugants and by E. coli DH5Sc containing the same 26 plasmids was determined on complex medium or on M9 minimal agar medium containing glucose, glycerol, arabinose, lactose, or galactose as a carbon source. Only inserts B15, B13, B21, and B35, located in the right end of the 10-kb HindIII fragment, expressed 3-galactosidase activity (Fig. 4). ,-Galactosidase activity was enhanced in the presence of IPTG and was observed with both bacterial species, and expression was unaffected by the carbon source. Preliminary experiments failed to detect factors from soybean plants capable of inducing P-galactosidase activity in the insertion mutants. DNA sequence analysis. Both strands of the 2.7-kb ClaI fragment were completely sequenced by single- or doublestranded DNA sequencing. Analysis of sequence data revealed that there are two possible ORFs (ORFi and ORF2) within the fragment, both in the same strand but in different reading frames (Fig. 5). ORFi starts with an ATG at nucleotide 871 and terminates at nucleotide 1410 with TGA. It comprises 540 bases with a G+C content of 68.9% that could encode a polypeptide of 180 amino acids (18,681 Da) (Fig. 5). ORF2 comprises 642 bases and could encode a polypeptide of 214 amino acids (Fig. 5). The G+C content of ORF2 is 59.4%, and the molecular mass of the putative polypeptide is 23,474 Da. The G+C content of the entire 2.7-kb ClaI

fragment is 64%. Both ORFs have putative E. coli ribosomal binding sites (Fig. 5), but no E. coli consensus promoter sequences were found. Both polypeptides potentially encoded by ORFi and ORF2 had a high number of hydrophobic amino acids (Fig. 5). Hydropathy plots of the ORFi and ORF2 products were obtained by the algorithm of Kyte and Doolittle (23) (data not shown). The putative polypeptide encoded by ORFi has three hydrophobic -regions, and its hydrophobicity index was 1.94. The putative polypeptide encoded by ORF2 has four hydrophobic regions and a hydrophobicity index of 2.46. A search of the GenBank and EMBL nucleotide sequence data bases failed to locate known nucleotide sequences or polypeptides with significant homology with the two putative polypeptides encoded by ORFi and ORF2. However, an examination of the Swiss-Prot sequence data base indicated that the gamma subunit of oxaloacetate decarboxylase (83 amino acids) of KIlebsiella pneumoniae (24) showed considerable homology with the carboxy terminus (between amino acids 131 and 214) of the putative polypeptide encoded by ORF2 (Fig. 6). The two polypeptides have 23% identical and 35% conserved amino acids (Fig. 6). The predicted patterns of transmembrane a-helices of the region, calculated by Rao and Argos probability parameters (35), were similar but displaced by 3 to 5 residues (data not shown). Construction of pSR2la, pSR2lb, pSR5la, and pSR200a and expression of ORF1 and ORF2 with cat fusions in vitro. A 1.7-kb SmaI fragment was deleted from pSR5 to construct pSR51 (Table 1 and Fig. 7), which carries only ORF2, and a 1.0-kb SmaI fragment from pSR2 was deleted to construct pSR21 (Table 1 and Fig. 7), which carries only ORF1. Since a lacZ fusion in ORF2 (17 in Fig. 4) had an orientation opposite to the direction of transcription, the expression of ORF2 could not be determined. For expression studies with ORF2 and to confirm the expression of ORF1, a promoterless cat gene from pJS133 (40) was inserted into appropriate restriction sites of ORFi and ORF2. The cat cassette DNA, generated from SimaI digestions of pJS133, was inserted into the BalI site of pSR21 in both orientations, resulting in pSR21a and pSR21b (Table 1 and Fig. 7). The DNA fragment produced by BamiII digestion of pJS133 was inserted into the BglII site of pSR51 in only a single orientation to give pSRSla (Table 1 and Fig. 7). A 1.6-kb ClaI-XbaI fragment

1928

J. BACTERIOL.

HWANG ET AL.

FIG. 5. Nucleotide and deduced amino acid sequences of ORF1 and ORF2 in the 2.7-kb ClaI fragment. The stop codons of ORF1 and ORF2 are identified by asterisks below the coding region. Possible translation start and stop codons are indicated by boldface type. The predicted ribosome-binding sites are underlined.

CCGCCCCTCCGTCATGTGCGCATCCTGCCGGTCAGCGGCCCACTGCAACGGCCGGCGGC TCCTGACCACGCCGGCCCGGTCGGCGCAGCGCTCCAGCTCCTGCAACTGCGCCAGCGA CTGCGCAGCGCGCACTCGCGCTTCCCTGCATGCTGCTGCCGCCGCAGTGCGACGAGGA

CTCGACCGGAGCCCGATGCCGAGGACAGGTTCACGACCCCATGACGGCGCCCCCGCA CGACAATGGCGCGGTTGGGCGACATATCGCCACCGAGT CGTACGCACGCAGCGCGCCC

ACAACGGACCCGGCA TTACCTGTCG2TGCACGAA1TLCCAAGCACCACTCCAAGGCAC

AGATGGCGATCGACCACATCGCGCTGCGGGTGGCClGAGTTCTGCAACGCCACGCCAGTG

CGCAGCGCAGGGAGCTGGG AGCCTGCTGCCTATCGATCAAGGGCTCGTCGCACAGAC 3 GCCGGACCAGCTATCGCTTCGCTTCAATACCAGTTCGC GACGTTGTTTtTGC CGGCT( CTCCACTCAACTC GGTACTCT&CCCCCC AGCaGCC CGGACGCGCGCCC CACGCTGAGTAGCACGCTCCAGATCAGCATCGAGGTTGTCTAACGCGGCACTGGCGACA GAAAGCCATTQGGAGG CTCCATcTTGCTAaCCGAGCAGAGCCAGGCTCCACCGGCAC

GTGCACTCTCTCAGGCGTTGACGCGTTCGGCCCGAGCGCGCACAGCTGAGCAGGGTGTT CGGCGATCCACGCCGCGCATGCCAATGCGGCTC31CGACGCACTGCCGTCGCATCCCAC

CCGATGAGGCGGCCCGCCTGCGCCTGCAGTTCCg&ACCGGCCAGATG GAT GTG CGG met asp val arg

882

ATT GCC GCG CGC GAC GGC CTG GCG CTG CTA CTG AAC GAG GAC GAC ile ala ala arg asp gly leu ala leu leu leu asn glu asp asp

927

CAC GCA CTC CGC GTT TCG ATC GCG GGG ATA TTA CTG GCC GAT CGC his ala leu arg val ser ile ala gly ile leu leu ala asp arg

972

CTC GCG CGC TCT TGC GCC GCT GGG CTT GGC GCT GCC GAG GTC ATC 1017 leu ala arg ser cys ala ala gly leu gly ala ala glu val ile GCC TTC GAG CGC GAT GCA GAG CCA GAC GAC TGC CAC GGC ATC GGC 1062 ala phe glu arg asp ala glu pro asp asp cys his gly ile gly ATC ACA CTG GGC GAT CTC GAT GCG ATT GCG CTG ACG GCA AGC GCC 1107 ile thr leu gly asp leu asp ala ile ala leu thr ala ser ala

containing the cat cassette in ORF2 was cloned into pWB5A, resulting in pSR200a (Table 1 and Fig. 7). Transformants of E. coli DH5ao containing pSR5la were resistant to 50 ,ug of chloramphenicol per ml; however, transformants of E. coli DH5a containing pSR21a and pSR21b were sensitive to that concentration (Fig. 7). Transconjugants of X. campestris pv. glycines 8ra and NP1 containing pSR200a were resistant to 15 ,ug of chloramphenicol per ml, whereas parent strain 8ra was sensitive to 5 ,ug/ml (Fig. 7). A 2.3-kb fragment containing a cat cassette from pSR21a could not be cloned into pWB5A, because deletions occurred both in the vector and in the insert.

CCC CTG CTG GCC ACG CTG CGG ACC GCA GTT GCC GCC TTG GCG CGC 1152 pro leu leu ala thr leu arg thr ala val ala ala leu ala arg

CCG CGC CAC TGC CAG TCT GGC TGG CCG CAC TGC GCG TCA AGC ACG 1197 pro arg his cys gln ser gly trp pro his cys ala ser ser thr CGC CTG CGC ATC GGG CGA GCG CAC CGC ATC GGC GGC GCT GCT GCA 1242 arg leu arg ile gly arg ala his arg ile gly gly ala ala ala GTC GCT GCG GCC TGG CGA TGT CTT GCT GCA TTG CAC GGC CTC GGC 1287 val ala ala ala trp arg cys leu ala ala lou his gly leu gly TGC GGC CAC CAG CGG CGA GGT GCT CTG GGG TAT CGC CGG GGG CGC 1332 cys gly his gln arg arg gly ala leu gly tyr arg arg gly arg GGT CTT GCG CGC TCC GGT ACG CCT GAA CCT GCA ACA GAT GAT TTT 1377

gly leu ala arg ser gly thr pro glu pro ala thr asp asp phe GGA GGC CAC CCC CAC CAT GCA GCA CGA TAC GTT TA GCCCGAAGTCG 1424 gly gly his pro his his ala ala arg tyr val *

CACAGTCCGCCAGCAACGTTGCGGAACTGGAACTGCCGGTGCAGCTGGAAGTGGACCAA SiI CTTGCGCTGTCTCTATCGACGTTGTCGGGTTTGCAGCCCGGGCAGATCCTGGAGTTGTC GGTGCCGGTGGATCAAGCCGACATCCGCCTGGTCGTTTACGGCCAGACCATCGGCATCG GCAGGTTAGTGACAGTGGGCGAACATCTTGGCGTGCAGATCCTCAGCATGTCGGAGAC ACGCATG CAG ATG CCT GAC GTT CCC TCG CTG CTG CTG GTG GTC ATC met gln met pro asp val gly ser leu leu leu val val ile

1483 1542 1601 1660

1706

ATG CTG GGC CTG CTG CCC TTC GCA CCG ATG GTG GTC ACC TCC TAC 1751 met leu gly leu leu pro phe ala ala met val val thr ser tyr

ACC AAG ATC GTG GTG GTG CTC CGT CTG CTG CGC AAT CCG ATC GGG 1796 thr lys ile val val val leu gly leu leu arg asn ala ile gly GTA CAG CAG GTG CCG CCG AAC ATG GTC CTC AAT CGC GTT GCG TTG 1841 val gln gln val pro pro asn met val leu asn gly val ala leu CTG GTG TCG TGC TTT GTG ATG CCA CCG GTG CGT ATG GAG CCG TTC 1886 leu val ser cys phe val met ala pro val gly met glu ala phe

AAG GCC GCG CAG AAC TAC GGT GCC GGG TCC GAT AAC AGC CGT GTG 1931 lys ala ala gln asn tyr gly ala gly sor asp asn ser arg val GTG GTC CTG CTC GAC CCC TGC CCG GAG CCG TTT CCG CAA TTC CTC 1976 val val lou leu asp ala cys arg glu pro phe arg gln phe leu CTC AAA CAC ACA CGC GAA CGC GAA AAG GCC TTT TTC ATG CGT TCA 2021 leu lys his thr arg glu arg glu lys ala phe phe met arg ser

GCA CAA CAG ATC TGG CCC AAG GAC AAG ala gln gln ile trp pro lys asp lys GAC GAC CTG CTT GTC TTG CCG CCG CCC asp asp leu leu val leu ala pro ala

GCC CCC ACG CTC AAG TCG 2066 ala ala thr leu

lys

CCC

GTT CTG AGC TAT CGG val leu ser tyr arg

ACT

2403 2462

120

130

140

190

A. B.

150

160

170

180

QQIWPKDKAATLKSDDLLVLAPAlSELTEAFRIGFLLYLVFIVIDLVVA-NALMACGL ------------MTDNAVLLGEGFTLMCLQICFVLVFLLLLIFAIRGMSLAVNRLFPPP 40 30 20

10

*

110

------------------------------------------------------------

B.

2344

100

BA.

CCGAACTTCC CTC CCTCAACC TCTCCTTCTTCCTCTCC CCGGGCTGCTAGCGCCTTCATGGAGGCGGCGCTGTCGCTGCAGGACGCATCGATGATA

90

VSCFVMAPVGMEAFKAAQNYGAGSDNSRVVVLLDACREPFRQFlLKHTIREREKAFFMRSA

ile

TOA TGCCATGGACCACCACGATCTAGTGCGATTTACCT

80

A.

CTT 2201

CTG TTT GTC GCA ATG GAT GGC TGG TCA ATG CTG ATT CAT GGC CTG 2291 leu phe val ala mat asp gly trp ser met leu ile his gly leu

60

50

B- --------------------------------------------_________________

ATC 2156

TCG CAG GTC ACC CCG ACC AAT GTG GCG ATC CCA TTC AAG CTG CTG 2246 ser gln val thr pro thr asn val ala ile pro phe lys leu leu

40

MQPDVGSILLLVVILPFAA NVVTSYTKIVVVIGLtNAIGVQQVPPNIVLNGVALL 70

leu

30

20

10

A.

ser

TTC ACG CTC ACC GAA TTG 2111 phe thr leu ser glu lou

TTC CCC ATC CGC TTT CTG TTG TAT CTG GTG TTT thr glu ala phe arg ile gly phe leu leu tyr leu val phe GTC ATC GAC CTG CTC GTGCCC AAC CCC CTG ATG CCG ATG CGC val ile asp leu val val ala asn ala leu mat ala met gly

ACC GAA

DISCUSSION Transposon insertion mutagenesis has been used successfully to study genes involved in the interactions of P. syringae and some strains of X. campestris pathovars with plants (31). Tn4431 (39) appears to be useful with some strains of X. campestris pv. campestris, and more recently, Omegon-Km (11) has also been used with that pathovar. Attempts to generate insertion mutants of X. campestris pv. glycines with several transposons (TnS in pGS9 [38], TnlO in pBEE104 [10], and TnSOI in pMD100 [10]) were made, but this approach was unsuccessful. Therefore, NTG was used to generate nonpathogenic mutants. The development of a rapid cotyledon bioassay for pathogenicity allowed us to screen large numbers (up to 2,000 colonies per day) of X. campestris pv. glycines colonies from cultures that had been treated with NTG. By this method, 16 prototrophic nonpathogenic mutants were obtained. This was approximately the same frequency (1%) with which auxotrophic and pigmentless mutants were found. As none of the nonpathogenic mutants showed other phenotypic differences from the wildtype strain, NP1 was chosen at random for further study. The lack of growth of NP1 on soybean cotyledons suggested that its nonpathogenic phenotype might be related to its inability to replicate in soybean plants. This phenotype has also been observed for nonpathogenic insertion mutants

200

210

SQVTPTNVAIPFKLLLFVANDCWSILIHGLVLSYR AAPXPAPAAVAPADDFARIPAIVAIHHHEARLHP 80 50

60

70

FIG. 6. Comparison of amino acid sequences of the putative polypeptide encoded by ORF2 (A) and the oxaloacetate decarboxylase gamma subunit (B) sequences. Identical and similar amino acids are indicated by colons and periods, respectively.

VOL. 174, 1992

PATHOGENICITY GENES FROM X. CAMPESTRIS PV. GLYCINES

IORFI :]cI

ORF2

pSR2

Il m

-O

pSR21 ORFi

pSR2la cm s

|(CAT|

ORFI ||CAT >

0.5 kb

pSR2lb COS

1

0

pSR51

rS1 pSR5la cm

C

>I

ORF2

cr pSR200a Cm FIG. 7. Promoterless cat fusions in ORF1 and )RF2. A cat cassette was inserted into ORF1 in both orientations a]nd into ORF2 in one orientation. The arrows above the ORFs indic ate the directions of transcription. Cms, chloramphenicol sensitive ; Cm', chloramphenicol resistant.

of P. syningae pv. syringae (1) and P. syringae p)v. phaseolicola (25). Failure to grow in plants is, however, not always linked to nonpathogenicity. A group of nonpati hogenic mutants of X. campestris pv. campestris multip]lied well in turnip seedlings (7), whereas another nonpathogenic mutant of that organism, generated by Tn4431 insertion al mutagenesis, was able to grow in excised leaves of cau liflower but unable to grow in attached leaves (39). It isi not clear, therefore, whether the failure to grow in the ho;st is a cause or a result of the loss of pathogenicity. Extracellular enzymes that degrade plant cell[walls have been implicated in diseases caused by X. campe stris pathovars (8, 45). In these instances, however, there is always a generalized destruction of the plant tissue. Thes e symptoms are not observed in bacterial pustule of soybeain caused by X. campestris pv. glycines, suggesting that cell,%wall-degrading enzymes are not factors in the disease. This iis supported by our results, which indicate that neither str*ain 8ra nor mutant NP1 produced detectable extracellular p)ectolytic or cellulolytic activities. Both the parent and the miutant strains secrete enzymes with proteolytic activity. Allthough it is possible that a specific pathogenicity-related prc)tease might be produced by the parent but not the mutant str*ain, there is no evidence that extracellular proteases are p.athogenicity determinants of X. campestris pv. glycines. Sin iilar conclusions about the role of plant cell wall-degradiing enzymes were reached by Dow et al. (8) concerning the vwilt diseases of crucifers caused by X. campestris pv. cannpestris, although polygalacturonate lyase might enhance virulence in that pathogen (8). hrp mutants of P. syringae pathovars phaseolic:ola (25) and syringae (33) and Pseudomonas solanacearum (4I) have been generated and characterized, but neither the paroent strain of X. campestris pv. glycines nor the nonpathoge nic mutant, NP1, showed a hypersensitive reaction on tobEacco plants. This indicates that the parent strain does not havre the ability to induce hypersensitive reactions on tobacco p)lants. Similarly, Kearney and Staskawicz (21) reported that :X. campestris pv. glycines did not induce a hypersensitive reaction on tomato plants. Southern hybridization analysis indicated thaat the 10-kb HindIII fragment was conserved among the X. campestris pathovars tested but not in P. syringae pathov ars glycinea and tabaci. This indicates that the pathogeniicity genes

1929

located on the 10-kb HindIII fragment may be interchangeable among X. campestris pathovars. Although Pseudomonas and Xanthomonas spp. are closely related in terms of taxonomy (5), our results indicate that the DNA fragment carrying the pathogenicity genes of X. campestris pv. glycines may be unique for Xanthomonas species. At least three regions responsible for restoring the pathogenicity phenotype to NP1 were identified by Tn3-HoHol mutagenesis (Fig. 4). The genes in the 2.7-kb ClaI fragment probably are not organized as an operon, because the plasmid with Tn3-HoHol inserted between the two ORFs (insert 37) was able to complement the nonpathogenic phenotype. Neither the 2.7-kb ClaI nor the 2.1-kb BamHI-XbaI fragments alone was able to complement the mutant phenotype of NP1, and Tn3-HoHol insertions in ORF1 and ORF2 also failed to complement NP1. Three regions are therefore necessary for complementation, and there appear to be multiple mutations in the 10-kb fragment, at least two mutations in the 2.7-kb fragment, and another mutation in the 2.1-kb region. This was not unexpected, because NTG is known to produce multiple mutations in neighboring regions of DNA (15). Supporting this is the fact that pIHl complemented IV-3 and IV-6 but that neither pIH5 nor pIH8 did. This suggested that there might be additional mutations in the pIH5 insert. The expression of ,B-galactosidase activity was detected in cells containing Tn3-HoHol located at the right end of the 31-kb fragment (Fig. 4), but only when the transposon had inserted in an orientation such that transcription of lacZ is from right to left. The known map of the construct (Fig. 4) suggests that the most likely explanation for this is that the expression is under the control of the lacZ promoter in pLAFR3, which is located just to the right of the 31-kb insert. The fact that the expression of 0-galactosidase from inserts B15, B13, B21, and B35 was enhanced with IPTG in the lacP background of E. coli DHSa supports this explanation. In no clones, including those in which Tn3-HoHol was located in regions presumably involved in pathogenesis, was lacZ expression detected. Because there is a possibility that lacZ expression may be under catabolic repression, expression was tested on the minimal medium with various carbon sources. However, expression in both X. campestris pv. glycines and E. coli was unaffected by the culture media tested. Thus, either the transposon was not present in a site and in an orientation that allowed expression, or genes present on the 10-kb fragment are not expressed under the culture conditions used. Sequencing data showed that the directions of transcription of both ORFi and ORF2 are from left to right (Fig. 5). Tn3-HoHol insert 17 in ORF2 was therefore inserted in an orientation opposite to the direction of transcription, whereas Tn3-HoHol inserts 56 and 68 were in the same orientation as the transcription (Fig. 5). This indicates that the inability of insert 17 in ORF2 to produce 13-galactosidase may be due to the orientation of the Tn3HoHol insert. This is supported by the result of a cat fusion in ORF2 and indicates that the potential gene containing ORF2 is constitutively expressed in E. coli and X. campestris pv. glycines. However, lacZ and cat fusions in ORFi indicate that ORFi may not be expressed under our culture conditions by either E. coli or X. campestris pv. glycines. Cloning of the approximately 2.3-kb fragment containing a cat cassette from pSR21a into pWB5A could not be done because, for unknown reasons, deletions in both the vector and the insert occurred. Therefore, the expression of ORFi could not be determined. The G+C content of X. campestris is in the range of 63.5

1930

HWANG ET AL.

to 69.2% (5). The G+C contents of the 2.7-kb fragment (64%) and ORF1 (68.9%) were within the average range of X. campestris. In terms of codon usage of ORF1 and ORF2, G and C were used more frequently in position 3 than A and T, which is typical of microorganisms with high G+C content. The G+C content of ORF2 (59.4%) was relatively low by comparison, but not as low as that of avrB (46%) and avrC (47%) from P. syingae pv. glycinea race 0, which is also an organism with high G+C content (44). It was suggested that avrB and avrC might have been relatively recently introduced into P. syringae pv. glycinea from another source

(44).

The function of the putative polypeptide encoded by ORF1 is unknown. However, the putative polypeptide encoded by ORF2 showed some similarity to the gamma subunit of oxaloacetate decarboxylase from K pneumoniae. This protein is a trimer containing single alpha, beta, and gamma subunits. The alpha subunit is thought to be cytoplasmic, and the beta and gamma subunits are membrane associated and thought to be involved in sodium transport (24). The similarities between the hydropathy plots, the amino acid sequences, and the predicted tertiary structures of the gamma subunit of oxaloacetate decarboxylase and the putative polypeptide encoded by ORF2 suggest that the polypeptide might be membrane associated and possibly involved in ion transport. In addition, the putative transmembrane region contains two negatively charged (glutamic acid) amino acids in the region consisting of 144 through 166. Charged residues located in the intramembrane portions of the beta subunit of oxaloacetate decarboxylase are suspected of being functionally important for ion transport; however, no charged residues in the putative membranespanning helix of the gamma subunit were found (24). The functions and regulatory properties of pathogenicity genes are currently under investigation. ACKNOWLEDGMENTS We thank E. J. Braun, S. K. Farrand, T. W. Jacobs, and B. J. Staskawicz for generous gifts of plasmid vectors and strains. We acknowledge S. K. Farrand, A. G. Hepburn, and A. L. Kriz for their comments and suggestions in preparing the manuscript. REFERENCES 1. Bertoni, G., and D. Mills. 1987. A simple method to monitor growth of bacterial populations in leaf tissue. Phytopathology 77:832-835. 2. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1523. 3. 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. 4. Boucher, C. A., F. M. Gisegem, P. A. Barberis, M. Arlat, and C. ZischeL 1987. Pseudomonas solanacearum genes controlling both pathogenicity on tomato and hypersensitivity on tobacco are clustered. J. Bacteriol. 169:5626-5632. 5. Bradbury, J. F. 1984. Xanthomonas Dowson 1939, 187.Y, p. 199-210. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. The Williams & Wilkins Co., Baltimore. 6. Chesney, R. H., J. R. Scott, and 0. Vapnek. 1979. Integration of the plasmid prophages P1 and P7 into the chromosome of Escherichia coli. J. Mol. Biol. 130:161-173. 6a.Croft, A. 1989. Seqanal package version 2.00. University of

Illinois, Urbana, Ill.

7. Daniels, M. J., C. E. Barber, P. C. Turner, M. K. Sawczye, R. J. W. Byrde, and A. H. Fielding. 1984. Cloning of genes

J. BACTERIOL.

involved in pathogenicity of Xanthomonas campestris pv. campestris using the broad host range cosmid pLAFR1. EMBO J. 3:3323-3328. 8. Dow, J. M., D. E. Milligan, L. Jaison, C. E. Barber, and M. J. Daniels. 1989. Molecular cloning of a polygalacturonate lyase gene from Xanthomonas campestris pv. campestris and role of the gene product in pathogenicity. Physiol. Mol. Plant Pathol. 35:113-120. 9. Dow, J. M., G. Scofield, K. Trafford, P. C. Turner, and M. J. Daniels. 1987. A gene cluster in Xanthomonas campestris pv. campestris required for pathogenicity controls the excretion of polygalacturonate lyase and other enzymes. Physiol. Mol. Plant Pathol. 31:261-271. 10. Ely, B. 1985. Vectors for transposon mutagenesis of non-enteric bacteria. Mol. Gen. Genet. 200:302-304. 11. Fellay, R., H. M. Krisch, P. Prentki, and J. Frey. 1989. OmegonKm: a transposon element designed for in vitro insertional mutagenesis and cloning of genes in Gram-negative bacteria. Gene 76:215-226. 12. Fett, W. F., S. F. Osman, and M. F. Dunn. 1987. Auxin production by plant-pathogenic pseudomonads and xanthomonads. Appl. Environ. Microbiol. 53:1839-1845. 13. Figurski, D. H., 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. 14. Gabriel, D. W., A. Burges, and G. R. Lazo. 1986. Gene-for-gene interactions of five cloned avirulence genes from Xanthomonas campestris pv. malvacearum with specific resistance genes in cotton. Proc. Natl. Acad. Sci. USA 83:6415-6419. 15. Guerola, N., J. L. Ingraham, and E. Cerada-Olmedo. 1971. Induction of closely linked multiple mutations by nitrosoguanidine. Nature (London) 230:122. 16. Henikoff, S., J. C. Wallace, and J. P. Brown. 1990. Finding protein similarities with nucleotide sequence databases. Methods Enzymol. 183:111-132. 17. Hwang, I., S. M. Lim, and P. D. Shaw. 1988. Generation of nonpathogenic mutants and complementation in Xanthomonas campestris pv. glycines 8ra. J. Cell. Biochem. 12C:252 (Abstr.). 18. Hwang, I., S. M. Lim, and P. D. Shaw. 1989. Cloning and characterization of pathogenicity genes from Xanthomonas campestris pv. glycines 8ra. Phytopathology 79:1210 (Abstr.). 19. Hwang, I., S. M. Li.m, and P. D. Shaw. 1992. Use of detached soybean cotyledons for testing pathogenicity of Xanthomonas campestris pv. glycines. Plant Dis. 76:182-183. 20. Jones, S. B., and W. F. Fett. 1987. Bacterial pustule disease of soybean: microscopy of pustule development in a susceptible cultivar. Phytopathology 77:266-274. 21. Kearney, B., and B. J. Staskawicz. 1990. Widespread distribution and fitness contribution of Xanthomonas campestris avirulence gene avrBs2. Nature (London) 346:385-386. 22. Kelemu, S., and J. E. Leach. 1990. Cloning and characterization of an avirulence gene from Xanthononas campestris pv. oryzae. Mol. Plant-Microbe Interact. 3:59-65. 23. Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132. 24. LauBermair, E., E. Schwarz, D. Oesterhelt, H. Reinke, K. Beyreuther, and P. Dimroth. 1989. The sodium ion translocating oxaloacetate decarboxylase of Kiebsiella pneumoniae. J. Biol. Chem. 264:14710-14715. 25. Lindgren, P. B., R. C. Peet, and N. J. Panopoulos. 1986. Gene cluster of Pseudomonas syringae pv. "phaseolicola" controls pathogenicity of bean plants and hypersensitivity on nonhost plants. J. Bacteriol. 168:512-522. 26. Maniatis, T., E. F. Fritsch, and J. Sambrook 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 27. Mead, D. A., E. Szczesna-Skorupa, and B. Kemper. 1986. Single-stranded DNA 'blue' T7 promoter plasmids: a versatile tandem promoter system for cloning and protein engineering. Protein Eng. 1:67-74. 28. Mellano, V. J., and D. A. Cooksey. 1988. Development of host

VOL. 174, 1992

29. 30.

31. 32. 33.

34. 35. 36. 37. 38.

39.

PATHOGENICITY GENES FROM X. CAMPESTRIS PV. GLYCINES

range mutants of Xanthomonas campestris pv. translucens. Appl. Environ. Microbiol. 54:884-889. Messing, J., R. Crea, and P. H. Seeburg. 1983. A system for shotgun DNA sequencing. Nucleic Acids Res. 9:309-321. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mills, D. 1985. Transposon mutagenesis and its potential for studying virulence genes in plant pathogens. Annu. Rev. Phytopathol. 23:297-320. Mount, D. W., and B. Conrad. 1984. Microcomputer programs for graphic analysis of nucleic acid and protein sequences. Nucleic Acids Res. 12:811-817. Mukhopadhyay, P., J. Williams, and D. Mills. 1988. Molecular analysis of a pathogenicity locus in Pseudomonas syringae pv. syningae. J. Bacteriol. 170:5479-5488. Pearson, W. R. 1990. Rapid and sensitive sequence comparison with FASTP and FASTA. Methods Enzymol. 183:63-99. Rao, J. K. M., and P. Argos. 1986. A conformational preference parameter to predict helices in integral membrane proteins. Biochim. Biophys. Acta 869:197-214. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. Schaad, N. W., ed. 1988. Laboratory guide for identification of plant pathogenic bacteria, 2nd ed. American Phytopathological Society, St. Paul, Minn. Selvaraj, G., and V. N. Iyer. 1983. Suicide plasmid vehicles for insertion mutagenesis in Rhizobium meliloti and related bacteria. J. Bacteriol. 156:1292-1300. Shaw, J. J., P. Rogowsky, T. J. Close, and C. I. Kado. 1987.

1931

Working with bacterial bioluminescence. Plant Mol. Biol. Rep.

5:225-236. 40. Shiau, A., and J. M. Smith. 1988. Improved cat cassette for promoter analysis and genetic constructions. Gene 67:295-299. 41. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. 42. Stachel, S. E., G. An, C. Flores, and E. W. Nester. 1985. A Tn3lacZ transposon for the random generation of ,-galactosidase gene fusions: application to the analysis of gene expression in Agrobacterium. EMBO J. 4:891-898. 43. Staskawicz, B., D. Dahlbeck, N. Keen, and C. Napoli. 1987. Molecular characterization of cloned avirulence genes from race 0 and race 1 of Pseudomonas syringae pv. glycinea. J. Bacteriol. 169:5789-5794. 44. Tamaki, S., D. Dahlbeck, B. Staskawicz, and N. T. Keen. 1988. Characterization and expression of two avirulence genes cloned from Pseudomonas syringae pv. glycinea. J. Bacteriol. 170:

4846-4854. 45. Tang, J. L., C. L. Gough, C. E. Barber, J. M. Dow, and M. J. Daniels. 1987. Molecular cloning of protease gene(s) from Xanthomonas campestris pv. campestris: expression in Escherichia coli and role in pathogenicity. Mol. Gen. Genet. 210:443-448. 46. Turner, P., C. Barber, and M. Daniels. 1985. Evidence for clustered pathogenicity genes in Xanthomonas campestns pv. campestris. Mol. Gen. Genet. 199:338-343. 47. Weber, C. R., J. M. Dunleavy, and W. R. Fehr. 1966. Effect of bacterial pustule on closely related soybean lines. Agron. J. 58:544-545.