Two Genomic Regions Involved in Catechol Siderophore

2 downloads 143 Views 1MB Size Report
Pathology and Weed Science, Colorado State University,Fort Collins, Colorado 805232; and Horticultural Crops. Research Laboratory, Agricultural Research Service, U.S. Department ..... Colonies that made zones were tested for catechol.
APPLIED

AND

Vol. 60, No. 2

ENVIRONMENTAL MICROBIOLOGY, Feb. 1994, p. 662-669

0099-2240/94/$04.00+0 Copyright X 1994, American Society for Microbiology

Two Genomic Regions Involved in Catechol Siderophore Production by Erwinia carotovora CAROLEE T. BULL,'t CAROL A. ISHIMARU,2 AND JOYCE E. LOPER3* Department of Botany and Plant Pathology, Oregon State University, Corvallis, Oregon 973311; Department of Plant Pathology and Weed Science, Colorado State University, Fort Collins, Colorado 805232; and Horticultural Crops Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture, Corvallis, Oregon 973303 Received 2 September 1993/Accepted 19 November 1993

Two regions involved in catechol biosynthesis (cbs) of Erwinia carotovora W3C105 were cloned by functional complementation of Escherichia coli mutants that were deficient in the biosynthesis of the catechol siderophore enterobactin (ent). A 4.3-kb region of genomic DNA of E. carotovora complemented the entB402 mutation of E. coli. A second genomic region of 12.8 kb complemented entD, entC147, entE405, and entA403 mutations ofE. coli. Although functions encoded by catechol biosynthesis genes (cbsA, cbsB, cbsC, cbsD, and cbsE) of E. carotovora were interchangeable with those encoded by corresponding enterobactin biosynthesis genes (entA, entB, entC, entD, and entE), only cbsE hybridized to its functional counterpart (entE) in E. coli. The cbsEA region of E. carotovora W3C105 hybridized to genomic DNA of 21 diverse strains of E. carotovora but did not hybridize to that of a chrysobactin-producing strain of Erwinia chrysanthemi. Strains of E. carotovora fell into nine groups on the basis of sizes of restriction fragments that hybridized to the cbsEA region, indicating that catechol biosynthesis genes were highly polymorphic among strains of E. carotovora.

erwinia soft rot of potato. Biological control of potato seed piece decay is thought to be determined in part by siderophore-mediated iron competition between Pseudomonas spp. and E. carotovora (23, 24, 56). Fluorescent siderophores, termed pyoverdines (also called pyoverdins or pseudobactins), produced by Pseudomonas spp. presumably deplete the microenvironment surrounding the pathogen of available iron by sequestering ferric ions as ferric-pyoverdine complexes, which are not utilized by E. carotovora. The outcome of siderophoremediated iron competition between Pseudomonas spp. and E. carotovora depends theoretically upon exchange of Fe(III) between pyoverdines and Erwinia siderophores (30). Thus, identification of the siderophores utilized by E. carotovora is needed to evaluate the hypothesis that pyoverdines produced by Pseudomonas spp. limit the levels of iron available to E. carotovora. Nevertheless, the siderophore production and utilization systems of E. carotovora have not been thoroughly characterized. The importance of siderophores in the biological control of erwinia soft rot diseases and in the virulence of other phytopathogenic erwinias prompted our investigations of siderophore production by E. carotovora. E. carotovora subsp. carotovora W3C105 produces a catechol(s) and the hydroxamate siderophore aerobactin, which functions in iron acquisition by that strain (21). In this report, we describe the cloning and characterization of two genomic regions required for catechol siderophore production by E. carotovora W3C105. (Abstracts describing this research have been published [5, 6].)

Iron is an essential element needed in trace amounts for many cellular processes. Although iron is abundant, it is present in soil of neutral pH as insoluble colloidal hydroxides that are not available biologically (29). Most microorganisms overcome the problem of iron limitation by producing siderophores, which are low-molecular-weight molecules that form tight, soluble coordination complexes with iron (33, 34). Siderophores are produced by microorganisms under ironlimiting conditions and are excreted into the environment where they chelate iron(III); the ferric-siderophore complex is transported into the cell via specific outer membrane receptors (34). Iron released internally from the ferric-siderophore complex is available for cellular functions (12). Members of the family Enterobacteriaceae commonly produce catechol siderophores such as enterobactin (also called enterochelin) and hydroxamate siderophores such as aerobactin (10, 12). Erwinia carotovora subsp. carotovora Jones and E. carotovora subsp. atroseptica Jones are phytopathogens that cause soft rot diseases of potato (Solanum tuberosum L.) (37, 38). Soil-borne pathogens such as E. carotovora presumably experience iron limitation at the oxygen levels and pH ranges present in many agricultural soils (4). Strains of E. carotovora produce aerobactin (21, 22) and an uncharacterized catechol siderophore(s) (28). The related bacterium Erwinia chrysanthemi produces chrysobactin (39, 40), a catechol siderophore that contributes to the systemic virulence of this phytopathogen (13, 14). Similarly, an uncharacterized hydroxamate siderophore is a virulence factor of Erwinia amylovora (53), the causal agent of fire blight disease. Pseudomonas spp. applied to potato seed pieces prior to planting (23, 56) or to tubers prior to storage (8) suppress

MATERIALS AND METHODS Bacterial strains and culture conditions. The bacterial strains and plasmids used in these studies are listed in Table 1. Escherichia coli, E. chrysanthemi, and Salmonella typhimurium strains were cultured routinely on Luria-Bertani (LB) medium (17) at 37°C. E. carotovora was cultured on LB medium, crystal violet pectate medium (11), or pectate agar (3) at 27°C. LB agar was supplemented with 5-bromo-4-chloro-3-indolyl-P-D-

Corresponding author. Mailing address: Horticultural Crops Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture, 3420 N.W. Orchard Ave., Corvallis, OR 97330. Phone: (503) 750-8771. Fax: (503) 750-8764. Electronic mail address: LoperJ @bcc.orst.edu. t Present address: Laboratoire de Biologie Microbienne, Universite de Lausanne, CH-1015 Lausanne-Dorigny, Switzerland. *

662

TABLE 1. Bacterial strains and plasmids or Source reference

Relevant characteristics'

Strain or plasmid

Escherichia coli AN194 AN193 AN192 AN191 AN93 AN90 AN117 MT147

F- tonA23 proC14 leuCJ6 trpE38 thi-1 Smr Ent' Same as those for AN194 but entA403 Same as those for AN194 but entB402 Same as those for AN194 but entC401 Same as those for AN194 but entE405 Same as those for AN194 but entD Same as those for AN194 but entF entC147::Km

26 J. B. J. B. J. B. J. B. 9 41 35

Salmonella typhimurium enb-7 enb-1

Ent- DHBA- Smr Ent - Smr

43 43

Isolated from potato in North Dakota, serogroup XXXIX Isolated from potato in Oregon, serogroup III Isolated from potato in Montana, serogroup V Isolated from potato in North Dakota, serogroup XL Isolated from potato in Montana, serogroup XV Isolated from potato in Montana, serogroup XI Isolated from potato in Montana, serogroup XXXVI Isolated from potato in Oregon, serogroup XXXVIII Isolated from soil in Oregon, serogroup XXXVII Isolated from soil in Oregon, serogroup XXXIII Isolated from potato in Oregon, serogroup XXIX Isolated from potato in Oregon, serogroup XXVII Isolated from soil in Scotland Isolated from soil in Wisconsin, serogroup XXIX Isolated from water in Colorado Isolated from artichoke in California Isolated from onion in Oregon Isolated from broccoli in Oregon Isolated from lettuce in California Isolated from broccoli in Oregon

56 18 18 18 18 18 18 18 18 18 18 18 20 A. Kelman M. Powelson M. Powelson M. Powelson M. Powelson M. Powelson M. Powelson

Erwinia carotovora subsp. atroseptica W3C37 SCRI-1043

Isolated from potato in Washington Isolated from potato in Scotland

56 20

Erwinia chrysanthemi 3937

Isolated from African violet in France

13

Plasmids pRK2013 pLAFR3 pUC8 pUC19 pMS101 pCP410 pCP1492 pJS151 pITS47 pPC104 pJEL1594 pJEL1595 pJEL1596 pJEL1892 pJEL1893 pJEL1894 pJEL1895 pJEL1896 pJEL1597 pJEL1599 pJEL1600 pJEL1601 pJEL1602 pJEL1868 pJEL1874

Kmr Tra+ Mob' ColEl replicon Tcr Tra- Mob' pRK2 replicon Apr Apr 10.5-kb HindIII fragment; entD+ fepA+ fes+ entF+ in pBR322 6.7-kb EcoRI fragment; entCEBA+ in pACYC184 2.4-kb EcoRI-PvuII fragment; entE+ in pACYC184 entA+ on 0.85-kb AccI-EcoRV fragment in pGEMblue entCJ47+ in pBR322 entD+ fepA+ fes+ in pBR322 cbsB+ in pLAFR3 cbsB+ in pLAFR3 cbsB+ in pLAFR3 EcoRI deletion of pJEL1596; cbsB+ BamHI deletion of pJEL1596 HindIIl deletion of pJEL1594 EcoRI deletion of pJEL1594; cbsB+ BamHI deletion of pJEL1594; cbsB+ cbs(DC)EA + in pLAFR3 cbs(DC)EA+ in pLAFR3 cbsA+ in pLAFR3 cbs(DC)EA+ in pLAFR3 cbs(DC)EA+ in pLAFR3 5.6-kb KpnI fragment from pJEL1602 subcloned in pUC19; cbsEA+ BamHI deletion of pJEL1599; cbs(DC)+

15 52 54

Erwinia carotovora subsp. carotovora W3C105 cclOl cc102

cclO3 cc104 cc106 cc108 ccl10 cc303 cc306 ccSOl ccSOS SCRI-193 SR319 274-1-2 JL1128

JL1131 JL1132 JL1133 JL1134

Neilands Neilands Neilands Neilands

57

25 42 42 C. F. Earhart 35, 36 7 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

a Ent' and Ent-, enterobactin producer or nonproducer, respectively; DHBA-, does not produce dihydroxybenzoic acid, an intermediate in the enterobactin biosynthesis pathway; fes, ferric enterobactin esterase; fepA, gene encoding for ferric enterobactin outer membrane transport protein; cbs, catechol biosynthesis gene of Erwinia spp.; Smr, streptomycin resistant; Kmr, kanamycin resistant; Apr, ampicillin resistant; Tcr, tetracycline resistant; Mob', mobilizable plasmid; Tra+, self-transmissible plasmid.

663

664

BULL ET AL.

galactopyranoside (X-Gal; 40 ,ug/ml; International Biotechnologies, Inc., New Haven, Conn.) and isopropyl-,B-D-thiogalactopyranoside (IPTG; 100 ,g/ml) in some cloning experiments. Catechol production by E. carotovora and other bacteria was determined from three replicate cultures grown in Tris minimal salts (TMS) medium (51) supplemented with tryptophan (0.003%, wt/vol), thiamine (0.0002%, wt/vol), and deferrated Casamino Acids (0.3%, wt/vol). The iron availability of TMS medium was varied by adding FeCl3 (100 or 0.1 j.LM) or 2,2'-dipyridyl (75, 150, or 225 ,uM). Antibiotics were used at the following concentrations: ampicillin, 100 ,ug/ml; tetracycline, 20 pug/ml; spectinomycin, 50 ,ug/ml; streptomycin, 100 ,ug/ml; rifampin, 100 ,ug/ml; and kanamycin, 50 jig/ml. All chemicals were purchased from Sigma Chemical Co. (St. Louis, Mo.) unless specified otherwise. Spontaneous rifampin-resistant colonies of E. coli and S. typhimurium strains listed in Table 1 were isolated and purified as described previously (55). Generation times of the rifampinresistant mutants and parental strains in cultures grown with shaking in LB broth were determined by the change in optical density at 600 nm. Only rifampin-resistant mutants with generation times that did not differ significantly from those of parental strains were used in further experiments. Siderophore bioassay. A cross-feeding bioassay relied upon the inability of indicator strains of E. coli and S. typhimurium, which had mutations in the enterobactin biosynthesis pathway, to grow unless a utilizable siderophore was provided from an exogenous source (43). Molten TMS medium containing 2,2'dipyridyl (150 ,uM) was seeded with an indicator strain (106 CFU/ml) that had been grown overnight with shaking in TMS broth. Strains to be tested for enterobactin production also were grown overnight with shaking in TMS broth. Ten microliters of an overnight test culture was spotted on the surface of solidified, seeded TMS medium and incubated at 27°C. After 24 to 48 h, plates were observed for the growth of an indicator strain surrounding enterobactin-producing colonies. E. coli AN194, which produces enterobactin, and AN192, which produces neither enterobactin nor its precursor dihydroxybenzoic acid, were test strains included as controls in all experiments. All indicator strains were FepA+, indicating the presence of the ferric-enterobactin receptor protein in the bacterial outer membrane. Detection of siderophore production. Siderophore production by bacterial strains was detected by observation of orange halos surrounding bacterial colonies grown on an agar medium containing chrome azurol S (CAS). CAS agar, a universal siderophore detection medium, was prepared as described previously (50). Catechol was detected in supernatants of bacterial cultures grown for 24 to 48 h in TMS medium as described by Arnow (1) or Rioux et al. (45). Partial purification and chromatography of catechol produced by E. carotovora. For catechol production, 50 ml of a 48-h culture of E. carotovora W3C105, E. coli AN194, or E. chrysanthemi 3937 was inoculated into 1-liter double-sidearm Celstir flasks (Wheaton, Millville, N.J.) containing 500 ml of TMS broth medium. The medium was stirred vigorously on a magnetic stirplate at 26°C for 48 h, after which cells were removed by centrifugation at 6,000 x g. Culture supernatants were adjusted to pH 3.0 with 1 N HCl and extracted three times with an equal volume of ethylacetate as described by Young and Gibson (58). Ethylacetate extracts were combined and concentrated by rotary evaporation to 1/10 the original volume, and the residue was dissolved in H20. The methods described by Persmark et al. (39) were used to purify the catechol produced by E. carotovora and chrysobactin produced by E. chrysanthemi. Catechol in supernatants, ethylacetate

APPL. ENVIRON. MICROBIOL.

extracts, and fractions from purification steps was detected by the assay of Arnow (1). Arnow assay-positive fractions from extracts of cultures of E. carotovora and E. chrysanthemi were evaluated by thin-layer chromatography on GF254 silica gel plates (Analtech, Newark, Del.) developed in methanol-chloroform-acetic acid (90:5:5, vol/vol/vol). Recombinant DNA techniques. Plasmids and cosmids were isolated from E. coli and S. typhimurium strains by an alkalinelysis extraction procedure and further purified by cesium chloride density gradient centrifugation for Southern hybridization analyses and cloning experiments (47). Genomic DNA was isolated as described in published methods (2, 47). T4 DNA ligase and restriction endonucleases were used as recommended by the supplier (GIBCO-BRL Life Technologies). DNA was separated by gel electrophoresis at 50 V for 3 h in 0.7% agarose gels (SeaKem LE; FMC BioProducts, Rockland, Maine). Identification of catechol biosynthesis genes ofE. carotovora. An existing library of W3C105 genomic DNA (21), constructed in the cosmid vector pLAFR3, was used in this study. Mobilization of recombinant cosmids into rifampin-resistant derivative strains of E. coli and S. typhimurium was accomplished by triparental matings with the helper plasmid pRK2013. Transconjugants, selected for their resistance to rifampin and tetracycline, were screened for production of zones on CAS agar. Colonies that made zones were tested for catechol production and by the enterobactin bioassay, which is described above. Cosmids from approximately 1,500 members of the genomic library of W3C105 were mobilized into a rifampin-resistant derivative of E. coli AN192 to identify those that complemented the entB402 mutation of E. coli. The W3C105 genomic library was mobilized into rifampin-resistant derivatives of E. coli AN193 and AN93 to identify cosmids that complemented the entA403 or entE405 mutations, respectively. Cosmids that complemented entB402, entA403, or entE405 mutations were mobilized into rifampin-resistant derivatives of AN191, AN90, and AN1 17 to evaluate complementation of the entC401, entD, and entF mutations, respectively. E. coli MT147 was transformed with cosmids that complemented entB402, entA403, or entE405 mutations to evaluate complementation of the entC147 mutation. Cosmids from approximately 2,500 members of the W3C105 genomic library were mobilized into AN117 in an attempt to identify a cosmid complementing the entF mutation. Southern hybridizations. Probes for Southern blots were as follows: (i) for the entD-fepA-fes-entF region of E. coli, the 10.5-kb HindIll fragment of pMS101; (ii) for the entCEBA region of E. coli, the 6.7-kb EcoRI-EcoRV fragment of pCP410; (iii) for the entA gene of E. coli, the 0.85-kb AccI fragment of pJS151; (iv) for the entC147 gene of E. coli, the 0.65-kb EcoRI-HindIII fragment of pITS47; (v) for the entE gene of E. coli, the 2.4-kb EcoRI-PvuII fragment of pCP1492; (vi) for the cbsFA region (catechol biosynthesis) of E. carotovora, the 5.6-kb KpnI fragment of pJEL1868 (see Fig. 1C); and (vii) for the cbsB region of E. carotovora, the 9.1-kb HindIII fragment of pJEL1895 (see Fig. 2C). Fragments used as probes were purified from agarose gels (SeaKem GTG; FMC BioProducts) by adsorption and elution from NA-45 DEAE membranes as described in the manufacturer's recommendations (Schleicher & Schuell, Inc., Keene, N.H.). Probes were labeled by nick translation of the fragments with [32P]dCTP (New England Nuclear, Boston, Mass.) and used at 0.25 ,ug/ml. DNA to be probed was transferred from agarose gels to nylon membranes (Nytran; Schleicher & Schuell) by standard methods (47). Hybridization conditions were of moderate strin-

CATECHOL SIDEROPHORE PRODUCTION BY E. CAROTOVORA

VOL. 60, 1994

gency (55 to 65°C, 50% formamide, and 0.16 x SSC) or low stringency (44 to 55°C, 50% formamide, and 0.2 x SSC [1 x SSC is 0.15 M NaCl plus 0.015 M sodium citrate]) (47). RESULTS

Catechol siderophore production by E. carotovora. Twenty

strains of E. carotovora subsp. carotovora and two strains of E. carotovora subsp. atroseptica (Table 1) provided a functional siderophore to indicator strains E. coli AN93 and S. typhimurium enb-1. Each strain also produces a catechol(s) in TMS broth and a halo on CAS agar, indicating siderophore production (21). Because E. coli AN93 and S. typhimurium enb-1 have the ferric-enterobactin outer membrane receptor (FepA+) and mutations in enzymatic steps that are late in the enterobactin biosynthetic pathway, they utilize ferric-enterobactin as a source of iron but cannot synthesize enterobactin from exogenous sources of the precursor dihydroxybenzoic acid. Although strains of E. carotovora provided a functional siderophore(s) to strains E. coli AN93 and S. typhimurium enb-1, they did not appear as proficient as enterobactin-producing strains of E. coli in doing so. Cross-feeding zones surrounding colonies of E. carotovora were less turbid and smaller in diameter than those surrounding colonies of enterobactinproducing strains of E. coli. E. chrysanthemi 3937, which produces the catechol siderophore chrysobactin (39), did not provide a functional siderophore to E. coli AN93 and S. typhimurium enb-1. E. chrysanthemi and all strains of E. carotovora provided a functional siderophore(s) to E. coli AN193 and S. typhimurium enb-7, which grow on an ironlimited medium in the presence of ferric-enterobactin or its precursor dihydroxybenzoic acid. Characterization of the catechol produced by E. carotovora. Catechol production by E. carotovora W3C105, which also produces the hydroxamate siderophore aerobactin (21), was inversely related to levels of available iron in TMS medium, as is expected if the catechol functions as a siderophore. Although enterobactin in culture supernatants of E. coli was extractable into ethylacetate as described previously (58), neither the catechol(s) produced by strain W3C105 nor chrysobactin produced by E. chrysanthemi 3937 partitioned into ethylacetate. Thus, the catechol produced by strain W3C105 was not enterobactin. Both chrysobactin and the catechol produced by W3C105 migrated with an Rf of 0.45 to 0.50 on silica gel thin-layer chromatography plates developed in methanol-chloroform-acetic acid (90:5:5, vol/vol/vol). Therefore, although chrysobactin and the catechol produced by W3C105 differed in cross-feeding of enterobactin-utilizing indicator strains, we did not detect differences between these compounds with respect to extractability into ethylacetate or migration on silica gel. Identification of catechol biosynthesis genes of E. carotovora. Catechol biosynthesis genes were identified from a cosmid library of genomic DNA of strain W3C105 by functional complementation of enterobactin biosynthesis mutations of E. coli. Of ca. 1,500 recombinant cosmids mobilized into E. coli AN193, four cosmids (pJEL1597, pJEL1599, pJEL1600, and pJEL1601) complemented the entA403 mutation (Fig. 1). Of ca. 1,500 cosmids initially mobilized into E. coli AN93, only cosmid pJEL1602 complemented entE405. Four of the five cosmids (pJEL1597, pJEL1599, pJEL1601, and pJEL1602) complemented entA403, entC401 (which is thought to be identical to entA403 [35]), entC147, entD, and entE mutations of E. coli and enb-7 and enb-1 mutations of S. typhimurium. These cosmids were designated cbs(DC)EA (cbs, catechol biosynthesis) to conform to the nomenclature of chrysobactin

665

biosynthesis genes of E. chrysanthemi (16) and with letters corresponding to complemented enterobactin biosynthesis mutations of E. coli. Cosmid pJEL1600, designated cbsA, complemented only entA and enb-7. None of the cosmids identified as cbsA or cbs(DC)EA complemented the entB mutation of E. coli. The five cosmids complementing the entA mutation contained overlapping sequences of DNA from W3C105 (Fig. 1). A 5.6-kb KpnI fragment, subcloned on plasmid pJEL1868, was sufficient for complementation of entA and entE and was designated cbsEA. Cosmid pJEL1600, which did not complement the entE mutation, lacked 0.5 kb of the 5.6-kb KpnI fragment located to the left, as can be seen in Fig. 1. Because pJEL1600 complemented the entA mutation, the entA-complementing region was located to the right of the entE-complementing region, as can be seen in Fig. 1C. Plasmid pJEL1874, generated by deleting the internal BamHI fragments of pJEL1599, complemented entC147, entD, and enb-7 but did not complement entA or entE (Fig. 1D). Because pJEL1874 contained only a portion of the 5.6-kb KpnI fragment, these results confirm the role of the 5.6-kb KpnI fragment in complementation of entE and entA. Cosmid pJEL1597 complemented entC147 and entD, whereas cosmid pJEL1600, which was truncated on the left border, did not complement these mutations. Thus, the DNA responsible for the complementation of entD and entC147 was localized to an 8.1-kb region. The relative arrangement of cbsD and cbsC regions was not determined. Of the 1,500 cosmids mobilized into E. coli AN192 from a genomic library of W3C105, three cosmids (pJEL1594, pJEL1595, and pJEL1596) complemented the entB mutation (Fig. 2). The three cosmids, designated cbsB on the basis of complementation of entB, had a 6.2-kb region in common. Of a series of plasmids derived from the original three cbsB cosmids by deletion of restriction fragments, only those that contained a 4.3-kb region complemented the entB mutation. The cbsB cosmids did not complement entA403, entC401, entC147, entD, or entE mutations in the enterobactin biosynthesis pathway. None of the cosmids identified as cbsB, cbsA, or cbs(DC)EA complemented the entF mutation. In addition, none of 2,500 recombinant cosmids from the W3C105 genomic library that were mobilized into E. coli AN117 complemented the entF mutation. Hybridization of enterobactin genes from E. coli to catechol biosynthesis genes of E. carotovora. Under low-stringency conditions, the entE gene and entCEBA region of E. coli hybridized to the 2.0-kb EcoRI, 1.3-kb BamHI, and 5.6-kb fragments present in all cbs(DC)EA cosmids (Fig. 1A) KpnI but did not hybridize to the cbsB cosmids. Probes of other enterobactin biosynthesis genes (entA, entC147, and the entDfepA-fes-entF region of E. coli) did not hybridize to the cbs(DC)EA or cbsB cosmids. The cbsB gene probe from E. carotovora, which consisted of the 9.1-kb HindIII-EcoRI fragment of pJEL1895, did not hybridize to DNA from the cbs(DC)EA cosmid pJEL1599. Thus, the cbsB fragment was distinct from the cbs(DC)EA region of E. carotovora. Under low-stringency conditions, the cbsB gene probe did not hybridize to genomic DNA of E. coli AN194 or to pCP410, which contained cloned entCEBA genes of E. coli (data not shown). This result further indicated that there were differences between the entB gene of E. coli and the functionally similar cbsB gene of W3C105. Hybridization of the cbsEA region of strain W3C105 to genomic DNA of diverse strains of E. carotovora. The cbsEA gene probe, which consisted of the 5.6-kb KpnI fragment of

666

APPL. ENVIRON. MICROBIOL.

BULL ET AL.

I

A |

1

10.5

1 2.71

5.5

3.5

11 1

1

5.8

1

5.6

1* I

|| |

6.6

HindIII BamHI

11 I

14.4

6.3

I

11.0

136

KpnI EcoRI

11.0

|

Complementation of Mutations

Region of homology to entE

B

H

entC147

entD

entE

entA

+

+

+

+

+

+

+

+

+

+

+

+

pJEL1600

_

m

m

+

pJEL1868

_

m

+

+

pJEL1874

+

+

_

_

E

pJEL1 601 H

E I

pJEL1 602 H

E

H

pJEL1599

E

pJEL1597 H z

E

C

K

K

11 5.6 kb

cbsE4

D

E

B I

II

IB H

x

c8.1kb

cbs(DC) FIG. 1. (A) Restriction map of the cbs(DC)EA region (cbs, catechol biosynthesis) of E. carotovora subsp. carotovora W3C105 genomic DNA. Numbers refer to sizes (in kilobases) of indicated restriction fragments. The asterisk indicates a region containing two KpnI fragments of 0.9 and 1.4 kb. The region hybridizing to the entE gene of E. coli is indicated by a shaded bar. (B) Cosmids containing cloned DNA from the cbs(DC)EA region. Solid lines refer to DNA derived from W3C105. The EcoRI and HindIII sites at the ends of each cosmid were contributed by the vector pLAFR3. The columns on the right indicate complementation (+) or lack of complementation (-) of enterobactin biosynthesis mutations of E. coli. (C) Location of the cbsEA region. The shaded portion of the bar is necessary only for complementation of entE; the unshaded portion is required for complementation of entA and entE. (D) Location of the cbs(DC) region indicated by the open bar. The striped line in pJEL1874 represents the region deleted from pJEL1599. Restriction endonuclease cut site abbreviations: E, EcoRI; B, BamHI; H, HindIll; K, KpnI.

pJEL1868, hybridized to genomic DNA of all 22 strains of E. listed in Table 1. There was great variability, however, in the sizes of the EcoRI and HindIll fragments to which the cbsEA probe hybridized; the strains fell into nine different groups on the basis of restriction fragment length polymorphisms. The cbsEA probe did not hybridize to genomic DNA from E. chrysanthemi 3937, even under low-stringency conditions. Thus, although regions hybridizing to the cbsEA region of strain W3C105 were common among strains of E. carotovora, they were not detected in a chrysobactin-producing strain of E. chrysanthemi. carotovora

DISCUSSION E. carotovora W3C105 provided iron to ferric-enterobactinutilizing strains of E. coli and S. typhimurium, as was expected if the catechol produced by W3C105 is structurally related to enterobactin and is transported into the cell by FepA, the ferric-enterobactin outer membrane receptor. Nevertheless, preliminary chemical analysis demonstrated that certain chemical properties of the catechol differed from those of enterobactin. The apparent discrepancy between the cross-feeding and chemical data may be explained by the lack of specificity of the FepA protein. FepA does not transport ferric-enterobactin

exclusively; other catechol siderophores, such as myxochelin and amonabactin (44), are transported by FepA. FepA also binds linear trimers and dimers of dihydroxybenzoylserine (48), monomeric dihydroxybenzoylserine (19), and MECAM, a synthetic analog of enterobactin (32). Ferric-chrysobactin, however, is not transported by FepA (16). Because catecholproducing strains of E. carotovora cross-fed FepA+ strains of E. coli and chrysobactin-producing strains of E. chrysanthemi did not cross-feed these strains, we expected to observe differences between the chemical properties of chrysobactin and the uncharacterized catechol produced by E. carotovora. These compounds did not differ in the limited number of chemical properties that were evaluated here. Differences between them may be detected, however, when ongoing studies elucidating the chemical structure of the catechol produced by E. carotovora are completed. Two distinct genomic regions involved in catechol biosynthesis (cbs) were identified by functional complementation of enterobactin biosynthesis (ent) mutations of E. coli. One region was a 12.8-kb fragment with catechol biosynthesis genes arranged in the order cbs(DC)EA according to subcloning and complementation data. Also present was a distinct 4.3-kb region termed cbsB that complemented the entB mutation of

VOL. 60, 1994

CATECHOL SIDEROPHORE PRODUCTION BY E. CAROTOVORA

A

8.1

|

*

BarrHI EcoRI HindIII

9.5

| 5.0 |

13.01

14.7 26.9

I

667

I

Complementation of entB

I

B

E H

H

I

I

I

I

pJEL1596 E

I.

+

pJEL1 594

i

E

t

pJEL1595

I

C

H H

|B

I

B

~~I

E

pJEL1 896

E

pJEL1895 I

I E

B

I

H

I

pJEL1 894

IB H -AMZ-

EE

E

I Hi I IgggFH"

pJEL1 893

E

IE

~rIF'I

pJEL1 892

absB FIG. 2. (A) Restriction map of the cbsB region of E. carotovora subsp. carotovora W3C105 genomic DNA. Numbers refer to sizes (in kilobases) of indicated restriction fragments. The asterisk indicates a region containing four EcoRI fragments of 0.5, 3.0, 3.5, and 4.7 kb. (B and C) Cosmids and subclones, respectively, derived from this region. Solid lines refer to genomic DNA derived from W3C105; striped lines are deleted regions. The EcoRI and HindIll sites at the extreme ends of each cosmid were contributed by the vector pLAFR3. The column on the right indicates complementation (+) or lack of complementation (-) of the entB mutation of E. coli. The open bar at the bottom defines the location of the cbsB gene. Restriction endonuclease cut site abbreviations: E, EcoRI; B, BamHI; H, HindIll.

E. coli. The entF mutation of E. coli was not complemented by cosmids from the genomic library of W3C105. The EntF protein is responsible for the activation of L-serine (46), a process that occurs late in the enterobactin biosynthesis pathway. Whereas enzymes catalyzing early steps in the biosynthesis of enterobactin and the E. carotovora catechol were crossfunctional, those catalyzing late steps in the biosynthetic pathways of these compounds presumably differ. Although functions encoded by cbs genes were interchangeable with those encoded by corresponding ent genes, only the entE gene of E. coli hybridized to its functional counterpart (cbsE) in E. carotovora. Similarly, genes for chrysobactin production of E. chrysanthemi and for enterobactin and amonabactin production of Aeromonas hydrophila complement enterobactin biosynthesis mutants of E. coli but do not hybridize to the E. coli genes (16, 31). In E. coli (12) and Shigella fiexneri (49), genes involved in enterobactin biosynthesis, uptake, and utilization are located on a continuous 24-kb genomic region. The entD gene is separated from entCEBA by approximately 16.6 kb. The remaining genes (entCEBA) are clustered in a 7.0-kb region (12). Similarly, genes for the biosynthesis of chrysobactin (cbs) and a gene encoding the ferric-chrysobactin receptor (fct) are clustered in the genome of E. chrysanthemi, forming an operon (fct-cbsCEBA) spanning approximately 8 kb (16). In the genome of E. carotovora W3C105, however, the cbsB gene is not flanked by cbsE and cbsA but is distal to the cbs(DC)EA gene cluster. On the basis of the observation that the cbsB probe did not hybridize to cosmids containing the cbs(DC)EA region, we propose that at least 17 kb of DNA separates cbsB from the other catechol biosynthesis genes on the chromosome of E. carotovora.

The cbsEA region of E. carotovcora W3C105 hybridized to genomic DNA of 21 diverse strains of E. carotovora but did not hybridize, even under low-stringency conditions, to genomic DNA of a chrysobactin-producing strain of E. chrysanthemni. Although the genomic DNA of all strains of E. carotovora hybridized to the cbsEA region of strain W3C105, sizes of hybridizing fragments differed markedly even among strains of a common serotype. Thus, catechol biosynthesis genes were highly polymorphic among strains of E. carotovora. To cause soft rot diseases, E. carotovora must grow and establish a large population size intercellularly and on root and tuber surfaces, where it competes with resident microflora for iron and other essential nutrients. Siderophores produced by E. carotovora may be important factors enhancing the fitness of producing strains in the iron-deplete environment of the plant rhizosphere or agricultural soils. The ubiquity of iron uptake systems and the prevalence of siderophores in natural habitats provide compelling evidence for the importance of iron acquisition to the fitness of plant-associated microbes (30). Every plant pathogen that has been evaluated produces siderophores (27), but only chrysobactin produced by E. chrysanthemi (13) and a hydroxamate produced by E. amylovora (53) have been identified as virulence factors in plant disease. Definition of an ecological role for siderophores produced by E. carotovora, achieved by comparing the virulence and fitness of the parental strain with those of derivative strains deficient in siderophore production, is a subject of future experiments. To this end, we have cloned aerobactin (21 ) and catechol biosynthesis genes of E. carotovora and determined their structural and functional relationships to siderophore biosynthesis genes of E. coli. Introduction of specific mutations into cloned genes and subsequently into the genome of W3C105 will result in sid-

668

BULL ET AL.

erophore-deficient derivatives that can be evaluated in ecological studies. ACKNOWLEDGMENTS We thank H. Barnes, S. Carnegie, K. Hoffbuhr, and J. Kron for technical assistance; J. H. Crosa, L. M. Crosa, C. R. Earhardt, D. C. Gross, M. A. McIntosh, A. Kelman, J. B. Neilands, and M. L. Powelson for providing bacterial strains; M. Henkels for preparation of the figures; and E. Clark and J. Kraus for critical reviews of the manuscript.

REFERENCES 1. Arnow, L. E. 1937. Colorimetric determination of the components of 3,4-dihydroxyphenylalanine-tyrosine mixtures. J. Biol. Chem. 118:531-537. 2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1987. Current protocols in molecular biology. Wiley-Interscience, New York. 3. Beraha, L. 1968. A rapid method for the preparation of a semi-solid agar medium for detection of pectolytic enzyme activity in Erwinia carotovora. Plant Dis. Rep. 52:167. 4. Bossier, P., M. Hofte, and W. Verstraete. 1988. Ecological significance of siderophores in soil. Adv. Microb. Ecol. 10:385-414. 5. Bull, C. T., C. A. Ishimaru, and J. E. Loper. 1989. Evidence for catechol siderophore production by Erwinia carotovora subsp. carotovora. Phytopathology 79:1155-1156. (Abstract.) 6. Bull, C. T., and J. E. Loper. 1991. Genetic analysis of catechol siderophore production of Erwinia carotovora subsp. carotovora. Phytopathology 81:1187. (Abstract.) 7. Coderre, P. E., and C. F. Earhart. 1984. Characterization of a plasmid carrying the Escherichia coli K-12 entD, fepA, fes, and entF genes. FEMS Microbiol. Lett. 25:111-116. 8. Colyer, P. D., and M. S. Mount. 1984. Bacterization of potatoes with Pseudomonas putida and its influence on postharvest soft rot diseases. Plant Dis. 68:703-706. 9. Cox, G. B., G. Gibson, R. K. J. Luke, N. A. Newton, I. G. O'Brien, and H. Rosenberg. 1970. Mutations affecting iron transport in Escherichia coli. J. Bacteriol. 104:219-226. 10. Crosa, J. H. 1989. Genetics and molecular biology of siderophoremediated iron transport in bacteria. Microbiol. Rev. 53:517-530. 11. Cuppels, D., and A. Kelman. 1974. Evaluation of selective media for isolation of soft-rot bacteria from soil and plant tissue. Phytopathology 64:468-475. 12. Earhart, C. F. 1987. Ferrienterobactin transport in Escherichia coli, p. 67-84. In G. Winkelmann, D. Van der Helm, and J. B. Neilands (ed.), Iron transport in microbes, plants and animals. VCH Publishers, Weinheim, Federal Republic of Germany. 13. Enard, C., A. Diolez, and D. Expert. 1988. Systemic virulence of Erwinia chrysanthemi 3937 requires a functional iron assimilation system. J. Bacteriol. 170:2419-2426. 14. Enard, C., T. Franza, C. Neema, P. R. Gill, M. Persmark, J. B. Neilands, and D. Expert. 1991. The requirement of chrysobactin dependent iron transport for virulence incited by Erwinia chrysanthemi on Saintpaulia ionantha. Plant Soil 130:263-271. 15. 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. 16. Franza, T., and D. Expert. 1991. The virulence-associated chrysobactin iron uptake system of Erwinia chrysanthemi 3937 involves an operon encoding transport and biosynthetic functions. J. Bacteriol. 173:6874-6881. 17. Gerhardt, P., R. G. E. Murray, R. N. Costilow, E. W. Nester, W. A. Wood, N. R. Krieg, and G. B. Phillips. 1981. Manual of methods for general bacteriology. American Society for Microbiology, Washington, D.C. 18. Gross, D. C., M. L. Powelson, K. M. Regner, and G. K. Radamaker. 1991. A bacteriophage-typing system for surveying the diversity and distribution of strains of Erwinia carotovora in potato fields. Phytopathology 81:220-226.

APPL. ENVIRON. MICROBIOL. 19. Hantke, K. 1990. Dihydroxybenzoylserine-a siderophore for E. coli. FEMS Microbiol. Lett. 67:5-8. 20. Hinton, J. C. D., M. C. M. Perombelon, and G. P. C. Salmond. 1985. Efficient transformation of Erwinia carotovora subsp. carotovora and E. carotovora subsp. atroseptica. J. Bacteriol. 161:786788. 21. Ishimaru, C. A., and J. E. Loper. 1992. High-affinity iron uptake systems present in Erwinia carotovora subsp. carotovora include the hydroxamate siderophore aerobactin. J. Bacteriol. 174:29933003. 22. Ishimaru, C. A., and A. Van Buren. 1991. Prevalence of hydroxamate and catechol siderophore production in Erwinia carotovora subsp. carotovora and E. carotovora subsp. atroseptica. Phytopathology 81:1245. (Abstract.) 23. Kloepper, J. W. 1983. Effect of seed piece inoculation with plant growth-promoting rhizobacteria on populations of Erwinia carotovora on potato roots and in daughter tubers. Phytopathology 73:217-219. 24. Kloepper, J. W., J. Leong, M. Teintze, and M. N. Schroth. 1980. Enhanced plant growth by siderophores produced by plant growth-promoting rhizobacteria. Nature (London) 286:885-886. 25. Laird, A. J., D. W. Ribbons, G. C. Woodrow, and I. G. Young. 1980. Bacteriophage Mu-mediated gene transposition and in vitro cloning of the enterochelin gene cluster of Escherichia coli. Gene 11:347-357. 26. Langman, L., I. G. Young, G. E. Frost, H. Rosenberg, and F. Gibson. 1972. Enterochelin system of iron transport in Escherichia coli: mutations affecting ferric-enterochelin esterase. J. Bacteriol. 112:1142-1149. 27. Leong, S. A., and D. Expert. 1989. Siderophores in plant-pathogen interactions, p. 62-83. In T. Kosuge and E. Nester (ed.), Plantmicrobe interactions, vol. 3. Academic Press, Inc., New York. 28. Leong, S. A., and J. B. Neilands. 1982. Siderophore production by phytopathogenic microbial species. Arch. Biochem. Biophys. 218: 351-359. 29. Lindsay, W. L. 1979. Chemical equilibria in soils. John Wiley & Sons, Inc., New York. 30. Loper, J. E., and J. S. Buyer. 1991. Siderophores in microbial interactions on plant surfaces. Mol. Plant-Microbe Interact. 4:513. 31. Massad, G., J. E. L. Arceneaux, and B. R. Byers. Diversity of siderophore genes encoding biosynthesis of 2,3-dihydroxybenzoic acid in Aeromonas spp. BioMetals 7, in press. 32. Matzanke, B. F., D. J. Ecker, T.-S. Yang, B. H. Huynh, G. Muller, and K. Raymond. 1986. Escherichia coli iron enterobactin uptake monitored by Mossbauer spectroscopy. J. Bacteriol. 167:674-680. 33. Neilands, J. B. 1981. Microbial iron compounds. Annu. Rev. Biochem. 50:715-731. 34. Neilands, J. B. 1982. Microbial envelope proteins related to iron. Annu. Rev. Microbiol. 36:285-309. 35. Ozenberger, B. A., T. J. Brickman, and M. A. McIntosh. 1989. Nucleotide sequence of Escherichia coli isochorismate synthetase gene entC and evolutionary relationship of isochorismate synthetase and other chorismate-utilizing enzymes. J. Bacteriol. 171: 775-783. 36. Ozenberger, B. A., M. S. Nahlik, and M. A. McIntosh. 1987. Genetic organization of multiple fep genes encoding ferric enterobactin transport functions in Escherichia coli. J. Bacteriol. 169: 3638-3646. 37. Perombelon, M. C. M., and A. Kelman. 1980. Ecology of the soft rot Erwinias. Annu. Rev. Phytopathol. 18:361-381. 38. Perombelon, M. C. M., and A. Kelman. 1987. Blackleg and other potato diseases caused by soft rot Erwinias: proposal for revision of terminology. Plant Dis. 71:283-285. 39. Persmark, M., D. Expert, and J. B. Neilands. 1989. Isolation, characterization, and synthesis of chrysobactin, a compound with siderophore activity from Erwinia chrysanthemi. J. Biol. Chem. 264:3187-3193. 40. Persmark, M., D. Expert, and J. B. Neilands. 1992. Ferric iron uptake in Erwinia chrysanthemi mediated by chrysobactin and related catechol-type compounds. J. Bacteriol. 174:4783-4789. 41. Pettis, G. S., and M. A. McIntosh. 1987. Molecular characterization of the Escherichia coli enterobactin cistron entF and coupled

VOL. 60, 1994

CATECHOL SIDEROPHORE PRODUCTION BY E. CAROTOVORA

expression of entF and the fes gene. J. Bacteriol. 169:4154-4162. 42. Pickett, C. L., L. Hayes, and C. F. Earhart. 1984. Molecular cloning of the Escherichia coli K-12 entACGBE genes. FEMS Microbiol. Lett. 24:77-80. 43. Pollack, J. R., B. N. Ames, and J. B. Neilands. 1970. Iron transport in Salmonella typhimurium: mutants blocked in the biosynthesis of enterobactin. J. Bacteriol. 104:635-639. 44. Rabsch, W., and G. Winkelmann. 1991. The specificity of bacterial siderophore receptors probed by bioassays. Biol. Metals 4:244250. 45. Rioux, C., D. C. Jordan, and J. B. M. Rattray. 1983. Colorimetric determination of catechol siderophores in microbial cultures. Anal. Biochem. 133:163-169. 46. Rusnak, F., M. Sakaitani, D. Drueckhammer, J. Reichert, and C. T. Walsh. 1991. Biosynthesis of the Escherichia coli siderophore enterobactin: sequence of the entF gene, expression and purification of EntF, and analysis of covalent phosphopantetheine. Biochemistry 30:2916-2927. 47. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 48. Scarrow, R. C., D. J. Ecker, C. Ng, S. Lui, and K. N. Raymond. 1991. Iron(III) coordination chemistry of linear dihydroxyserine compounds derived from enterobactin. Inorg. Chem. 30:900-906. 49. Schmitt, M. P., and S. M. Payne. 1988. Genetics and regulation of enterobactin genes in Shigella flexneri. J. Bacteriol. 170:5579-5587. 50. Schwyn, B., and J. B. Neilands. 1987. Universal chemical assay for

51. 52.

53.

54. 55.

56. 57. 58.

669

the detection and determination of siderophores. Anal. Biochem. 160:47-56. Simon, E. H., and I. Tessman. 1963. Thymidine-requiring mutants of phage T4. Proc. Natl. Acad. Sci. USA 50:526-532. 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. Vanneste, J. L., J.-P. Paulin, and D. Expert. 1990. Bacteriophage Mu as a genetic tool to study Erwinia amylovora pathogenicity and hypersensitive reaction on tobacco. J. Bacteriol. 172:932-941. Vieira, J., and J. Messing. 1982. The pUC plasmids and M13mp7derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19:259-268. Weller, D. M., and A. W. Saettler. 1978. Rifampin-resistant Xanthomonas phaseoli var. fuscans and Xanthomonas phaseoli: tools for field study of bean blight bacteria. Phytopathology 68:778-781. Xu, G.-W., and D. C. Gross. 1986. Selection of fluorescent pseudomonads antagonistic to Erwinia carotovora and suppressive of potato seed piece decay. Phytopathology 76:414-422. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33:103-119. Young, I. G., and F. Gibson. 1979. Isolation of enterochelin from Escherichia coli. Methods Enzymol. 56:394-398.