JOURNAL OF BACTERIOLOGY, Oct. 2002, p. 5723–5732 0021-9193/02/$04.00⫹0 DOI: 10.1128/JB.184.20.5723–5732.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 184, No. 20
Construction and Characterization of Transposon Insertion Mutations in Corynebacterium diphtheriae That Affect Expression of the Diphtheria Toxin Repressor (DtxR) Diana Marra Oram, Ana Avdalovic, and Randall K. Holmes* Department of Microbiology, University of Colorado Health Sciences Center, Denver, Colorado 80262 Received 2 April 2002/Accepted 22 July 2002
Transcription of the bacteriophage-borne diphtheria toxin gene tox is negatively regulated, in response to intracellular Fe2ⴙ concentration, by the chromosomally encoded diphtheria toxin repressor (DtxR). Due to a scarcity of tools, genetic analysis of Corynebacterium diphtheriae has primarily relied on analysis of chemically induced and spontaneously occurring mutants and on the results of experiments with C. diphtheriae genes cloned in Escherichia coli or analyzed in vitro. We modified a Tn5-based mutagenesis technique for use with C. diphtheriae, and we used it to construct the first transposon insertion libraries in the chromosome of this gram-positive pathogen. We isolated two insertions that affected expression of DtxR, one 121 bp upstream of dtxR and the other within an essential region of the dtxR coding sequence, indicating for the first time that dtxR is a dispensable gene in C. diphtheriae. Both mutant strains secrete diphtheria toxin when grown in medium containing sufficient iron to repress secretion of diphtheria toxin by wild-type C. diphtheriae. The upstream insertion mutant still produces DtxR in decreased amounts and regulates siderophore secretion in response to iron in a manner similar to its wild-type parent. The mutant containing the transposon insertion within dtxR does not produce DtxR and overproduces siderophore in the presence of iron. Differences in the ability of the two mutant strains to survive oxidative stress also indicated that the upstream insertion retained slight DtxR activity, whereas the insertion within dtxR abolished DtxR activity. This is the first evidence that DtxR plays a role in protecting the cell from oxidative stress. The Corynebacterium diphtheriae diphtheria toxin (DT) repressor (DtxR) is the prototype for a group of bacterial metaldependent regulator proteins (13). Examples of this DtxRdefined regulator protein group are found in a variety of bacterial species, including the AT-rich gram-positive staphylococcal species in which it is called SirR (20), acid-fast mycobacterial and rhodococcal species (4, 9), and the commercially relevant Brevibacterium and Streptomyces species (19, 36). Many members of this metal-dependent regulator family bind and are activated by Fe2⫹ in vivo, but TroR from Treponema pallidum and MntR from Bacillus subtilis are activated by Mn2⫹ (38, 42). The ability of a bacterial pathogen to scavenge iron is vital to its survival in the host and to the establishment of a productive infection (32). DtxR directly regulates genes involved in iron acquisition, including the irp1 and irp6 operons that encode products with homology to proteins involved in siderophoremediated uptake systems (27, 48) and hmuO, the first heme oxygenase identified in bacteria, which is required for utilization of iron from heme (44). In addition, DtxR regulates the transcription of the bacteriophage-borne tox gene whose product is the major virulence factor DT (5, 37, 49). It is likely, based on these observations, that DtxR is required for virulence of C. diphtheriae. The report that expression of a hyperactive variant of DtxR in Mycobacterium tuberculosis attenu-
ates its virulence in BALB/c mice (30) supports the putative role of DtxR and other iron-dependent regulators in virulence. In addition to their function in iron-dependent gene regulation, a role has been suggested for DtxR-like metal-dependent regulators in protecting the bacterial cell from oxidative stress. Mycobacterium smegmatis cells containing the wild-type DtxR homologue IdeR were better able to survive in the presence of hydrogen peroxide than those containing a null mutation in ideR (11). Further characterization demonstrated that IdeR-deficient cells produce fewer RNA messages for katG, encoding catalase and peroxidase, and sodA, encoding manganese superoxide dismutase, suggesting a positive role for IdeR in regulating these genes (12). Similarly, in gram-negative bacteria, the ferric uptake regulatory protein Fur, which is the prototype for a family of metal-dependent regulators that is distinct from DtxR, regulates genes in response to both iron and oxidative stress (10). The DNA binding, metal specificity, and physical structure of DtxR have been studied extensively in vitro and in Escherichia coli, leading to several important observations. DtxR binds a 19-bp operator sequence that overlaps the promoters of the genes that it regulates (46, 57). In both E. coli and C. diphtheriae, gene regulation by DtxR is observed in response to changing iron concentrations, but in vitro, DtxR can be activated by other divalent transition metals, including cadmium, cobalt, manganese, nickel, and zinc (46, 48, 56). Crystallographic studies have shown that DtxR has three domains (41, 43). Amino acids 1 to 73 form the DNA binding domain 1, which contains a classical helix-turn-helix motif, amino acids 74 to 140 form domain 2, which is required for dimerization and metal binding, and amino acids 140 to 226 form domain 3,
* Corresponding author. Mailing address: Department of Microbiology, Campus Box B-175, University of Colorado Health Sciences Center, Denver, CO 80262. Phone: (303) 315-7903. Fax: (303) 3156785. E-mail:
[email protected]. 5723
5724
ORAM ET AL.
which has the same topology as the SH3 domains found in signal transduction proteins (40). Analysis of gene function in C. diphtheriae has been greatly hampered by the lack of sufficient genetic tools. Only one plasmid replication origin derived from pNG2 has been shown to function in C. diphtheriae, and methods for DNA transformation are inefficient (49, 52, 53). Recently, construction of one C. diphtheriae strain containing the directed insertion of a nonreplicating plasmid into its chromosome was reported (45). While this was an exciting advance in the field, the techniques used in the construction were laborious and not easily applicable to screening large numbers of mutants. Most reported genetic investigations in C. diphtheriae have relied on the insights provided by analyses of bacteriophage crosses and chemically induced mutations (8, 21, 23). Therefore, the development of an efficient system to construct and isolate insertion mutations in the chromosome of C. diphtheriae would greatly facilitate genetic analysis of this toxin-producing bacterial pathogen. Transposon mutagenesis is a powerful tool used in a wide variety of bacterial species to determine the function of gene products, discover new genes, and study gene regulation (6). The complexity of the transposition reaction utilized by many transposons has made them difficult to adapt for use in bacterial species other than their native hosts. One exception is Tn5, which, following electroporation, has been shown to insert via transposition in the chromosome not only of its native host E. coli but also those of Salmonella enterica serovar Typhimurium, Proteus vulgaris, and the yeast Saccharomyces cerevisiae (17). The transposition of Tn5 requires only the DNA ends of the transposon, the cognate transposase protein, and a DNA target both in vitro and in vivo (18). We adapted a Tn5-based mutagenesis system marketed by Epicentre, Madison, Wis., for use in C. diphtheriae. We demonstrated that Tn5 transposition in C. diphtheriae C7() has no detectable target site specificity, and we isolated two different variants with altered DtxR activity. In C7()18.5 the Tn5 insertion occurred within dtxR and no DtxR production was detected, indicating that dtxR is not an essential gene in C. diphtheriae. The Tn5 insertion in C7()3B11 occurred upstream of the wild-type DtxR coding region, and detectable but decreased amounts of DtxR were still produced. We investigated both variants for their ability to repress DT and siderophore production in the presence of iron concentrations sufficient to repress production of both gene products in the wild-type parent C7(). Finally, we assessed the ability of both variants to survive oxidative stress caused by exposure to H2O2. MATERIALS AND METHODS Bacterial strains and growth media. C. diphtheriae strain C7(⫺) was originally isolated from clinical specimens received by the California State Department of Public Health (16), and C7() was isolated after  phage infection of C7(⫺) (2). Both strains have been used extensively for experimental work since the 1950s. E. coli strain DH5␣ (Bethesda Research Laboratories, Gaithersburg, Md.) was used for cloning and DNA isolation. C. diphtheriae strains were routinely cultivated in heart infusion broth (Difco, Detroit, Mich.) plus 0.2% Tween 80 (HITW), and E. coli strains were cultivated in Luria-Bertani broth (33). Kanamycin was added at a concentration of 5 g/ml for C. diphtheriae. Spectinomycin was added at a concentration of 100 g/ml for both C. diphtheriae and E. coli. DNA isolation, sequencing, PCR, and Southern blot. DNA was isolated from E. coli by alkaline lysis (28). DNA was isolated from C. diphtheriae with a mini-beadbeater-8 (Biospec Products, Inc., Bartlesville, Okla.). C. diphtheriae
J. BACTERIOL. cells were concentrated 20-fold in solution 1 (100 mM Tris [pH 8], 10 mM EDTA, 50 mM glucose). An equal volume of 0.1-mm-diameter glass beads was added, and samples were processed in the beadbeater twice for 2 min at full speed. After centrifugation, the supernatant was extracted two or three times with an equal volume of water-saturated phenol and once with an equal volume of chloroform. Finally, the samples were precipitated with ethanol, and the DNA pellets were resuspended in water. PCR was performed with Taq polymerase (MBI Fermentas, Amherst, N.Y.) as per the manufacturer’s instructions. The sequences of the primers mentioned in Results are as follows: dtxR-H3, GAG CAGGTAAACAAGCTTTCTCG; dtxR-A3, CAGACACTTCCTACGTATCC GGC; KAN-FP-2, ACCTACAACAAAGCTCTCATCAACC; and KAN-RP-2, GCAATGTAACATCAGAGATTTTGAG. The University of Colorado Cancer Center DNA Sequencing and Analysis Core Facility, Denver (supported by National Institutes of Health-National Cancer Institute Cancer Support grant CA 46934), performed the DNA sequencing. Vector NTI software (Infomax, Golden, Colo.) was used to analyze the sequence data. Southern blots were performed with the DIG DNA labeling and detection kit (Roche Molecular Biochemicals, Indianapolis, Ind.) with a probe homologous to the aphA gene contained on the EZTnKan2 transposon, as per the manufacturer’s instructions. Electroporation of C. diphtheriae. A fresh colony of C. diphtheriae was inoculated into 10 ml of HITW, and the culture was incubated overnight at 37°C with shaking. Five to 10 ml of the overnight culture was used to inoculate 500 ml of HITW plus 2% glycine plus 15% sucrose to give an absorbance at 600 nm (A600) of ⬃0.1. The culture was incubated at 37°C with shaking until the A600 was between 0.4 and 0.6, and then it was put on ice for 15 min. The cells were collected by centrifugation and washed with 500 ml of sterile 15% glycerol and again with 50 ml of sterile 15% glycerol. Finally, the pelleted cells were resuspended in an equal volume of sterile 15% glycerol, DNA was added, and 100 l was electroporated in 0.2-cm-gap cuvettes with a Bio-Rad gene pulser set at 2.5 kV, 25 F, and 200 ⍀. The samples were immediately transferred to 1 ml of heart infusion broth containing 1% glucose (without Tween 80), incubated with shaking for 1 h at 37°C, and plated onto selective medium. Isolation of protein from C. diphtheriae and Western blotting. C. diphtheriae cells from cultures grown in HITW were concentrated 10-fold in solution 1 and lysed with the beadbeater as described above. Following lysis, samples were spun in a microcentrifuge for 5 min and the supernatant (extract) was transferred to a new tube. An equal volume of extract was added to 2⫻ sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis loading buffer (2% SDS, 125 mM Tris, 20% glycerol, and 0.1% bromophenol blue) and then run on an SDS–12% polyacrylamide gel. After separation on a polyacrylamide gel, the proteins were either stained with Coomassie blue or transferred to nitrocellulose and probed in a Western blot (22) with polyclonal anti-DtxR antiserum (50). Densitometry. Western blots developed with SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, Ill.) were exposed to X-ray film. The developed film was scanned and digitized with the Fluor-S imager (Bio-Rad, Hercules, Calif.). Analysis was performed with Quantity One software from Bio-Rad. ELISA to detect DT. A detailed description of the enzyme-linked immunosorbent assays (ELISAs) will be given elsewhere (Y. Qian, C. B. Erbe, and R. K. Holmes, unpublished data). Briefly, a 96-well plate was coated with affinitypurified goat anti-DT antiserum and then washed with phosphate-buffered saline. The plate was blocked with phosphate-buffered saline plus 10% horse serum. Next, 100-l samples of serially diluted C. diphtheriae supernatants were added to individual wells of the plate. The plate was washed, and rabbit anti-DT antiserum was added. The plate was washed again, and a goat anti-rabbit horseradish peroxidase-conjugated antibody was added. After a final wash, Sigmafast OPD reagent (Sigma, St. Louis, Mo.) was added and the A450 of each well in the plate was recorded. A standard curve was created by using purified DT in concentrations from 1 to 40 ng/ml. Siderophore assays. C. diphtheriae strains were grown overnight at 37°C with shaking in the deferrated casein hydrolysate medium of Mueller and Miller (34) as modified by Barksdale and Pappenheimer (2) (PGT) with or without 8 M FeCl3. Stocks of deferrated PGT were prepared by Chelex-100 treatment for 2 h as described by Tai et al. (55) followed by filter sterilization. Chrome azurol S (CAS) assays for corynebacterial siderophores were performed by a modification of a previously described method (51). Supernatants of the overnight cultures were serially diluted in deferrated PGT, and 0.5 ml of the diluted supernatant was added to 0.5 ml of CAS reagent. The samples were incubated at room temperature for 2 h, and then the A630 of each sample was measured. A standard curve was constructed by performing assays with ethylenediamine-N,N-diacetic acid at concentrations from 1 to 20 M. One siderophore unit was defined as the A630 of a control assay performed with a 0.5-ml sample of 1 M ethylenediamineN,N-diacetic acid.
TRANSPOSON MUTAGENESIS OF C. DIPHTHERIAE
VOL. 184, 2002
5725
Plasmid construction. Primers SacIIFwd, TATTTATAGATTTCCGCGGCT TCTAAATTT, and SacIIRev, AAATTTAGAAGCCGCGGAAATCTATAAA TA, were used in the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, Calif.) to amplify pJRS525 (31), resulting in plasmid pJRS525SacII. Next, primers LZSpFwd, CCAATTAGCATGCATATTTCCC, and LZSpRev, CTGA TGGGCCCCCATGTAATG were used to amplify a 1,648-bp region of pJRS525SacII that included the spectinomycin resistance gene. The PCR product was blunted with T4 polymerase (MBI Fermentas) and ligated to a blunted 3,200-bp EcoRI restriction fragment of pNGA2 (Qian et al., unpublished) to construct the vector pJKS1. A 4,770-bp SacI/EcoRV fragment of pJKS1 was ligated to a 1,541-bp HincII/SacI fragment of pMS297 (47) containing dtxR to construct pJKS-dtxR.
RESULTS Transposon mutagenesis of C. diphtheriae. In order to isolate transposon insertions in C. diphtheriae, we adapted the EZTnKan2 transposome system from Epicentre. In this system, a hyperactive variant of Tn5 transposase is bound in vitro to a short linear fragment of DNA containing the ends of Tn5 and the Tn903 kanamycin resistance gene aphA. The transposome (the transposase bound to EZTn DNA) is then electroporated into a host cell, and transformants are selected with kanamycin. Before attempting transposon mutagenesis in C. diphtheriae, we optimized the electroporation conditions and improved the efficiency of DNA transformation. In both grampositive and gram-negative bacteria, the addition of glycine to the growth media can increase the efficiency of DNA transformation by electroporation (1, 7). We included glycine in the medium of C. diphtheriae being prepared for electroporation (see Materials and Methods) and increased the transformation efficiency of the replication-competent plasmid pCM2.6 (49) approximately 100-fold (from ⬃102 to ⬃104 transformants per g of DNA). Electroporation of C7(⫺) and C7() strains of C. diphtheriae with 1 l of the transposome, containing ⬃20 ng of transposon DNA, resulted in approximately 2 ⫻ 105 kanamycin-resistant (Kmr) transformants per g of DNA. Transformation of C. diphtheriae with the EZTn transposome was higher than that reported in the Epicentre literature for M. smegmatis, (⬃1 ⫻ 103 transformants per g of DNA) but lower than that reported for E. coli (⬃1 ⫻ 106 transformants per g of DNA). Southern blot analysis of DNA isolated from 15 different Kmr colonies of C. diphtheriae with a probe specific for aphA demonstrated that all 15 contained a single copy of EZTn and that the insertion site in each was unique (data not shown). Since target site selection of EZTn in C. diphtheriae was varied and transposition frequency was high, we used the transposome system to generate libraries of insertion mutants in both C7(⫺) and C7(). Construction and PCR screening of a C7(ⴚ) transposon insertion library. We electroporated the EZTn transposome into C. diphtheriae strain C7(⫺) and selected Kmr colonies. Sterile sticks were used to pick and inoculate individual Kmr transformants into 200 l of HITW plus kanamycin in separate wells of 96-well microtiter plates. The inoculated plates were incubated overnight at 37°C with shaking. Sterile glycerol was added to each well at a final concentration of 16%, and the library was stored at ⫺70°C. Approximately 3,500 individual Kmr transformants were frozen in this manner. Next we screened the C7(⫺) EZTn insertion library by using a PCR-based method for transformants containing insertions
FIG. 1. Diagrams of primer binding sites and location of transposon insertions. Primer binding sites are represented by arrows, and the distances between sites are indicated in kilobases on the lines connecting the primers. A map of the wild-type C7 gene arrangement is shown in panel A along with a simple map of EZTn. Arrows: 1, primer dtxR-H3; 2, primer dtxR-A3; 3, primer KAN2-FP-1; 4, primer KAN2RP-1. The gene arrangements of the transposon insertion strains 3B11 and 18.5 are represented in panels B and C, respectively. Panel D shows an ethidium bromide-stained 1% agarose gel showing the products of PCRs performed with the primers diagramed in panels A, B, and C, and the template chromosomal DNA isolated from the strain is indicated above each lane. Lanes 1 to 3, primers 1 and 2; lanes 4 to 6, primers 2 and 3; lanes 7 to 9, primers 1 and 4. Sizes of molecular markers are indicated on the left in kilobases.
in or near the dtxR coding sequence. One of two primers with binding sites that flank dtxR, dtxR-H3 (Fig. 1), which binds upstream of the dtxR start codon, and dtxR-A3 (Fig. 1), which binds downstream of the dtxR stop codon, was used in combination with two EZTn-specific primers KAN2-FP-1 and KAN2-RP-1 (Fig. 1) to amplify DNA isolated from Kmr transformants. Since DNA products generated from the EZTnspecific primers will extend in opposite directions, only template DNA containing an insertion located between the
5726
ORAM ET AL.
dtxR-specific primers 1 and 2 will result in a PCR product (Fig. 1A). As predicted, no products were detected when wild-type C7(⫺) DNA was used as a template (Fig. 1D). DNA was isolated from pools of 12 transformants in the EZTn library and screened by this PCR method until a transformant with an insertion between the dtxR primer binding sites was identified and subsequently isolated in pure culture. We showed that this positive transformant could also be detected in a DNA pool containing 95 other transformants, so the pool size for subsequent screening was increased accordingly. Screening of the entire library did not reveal any additional insertions located between the binding sites for primers 1 and 2. The mutant isolated from this library was designated C7(⫺)3B11 (Fig. 1B and D). When chromosomal DNA from C7(⫺)3B11 was analyzed by Southern blotting, a single EZTn insertion was identified (data not shown). The orientation and exact location of the EZTn insertion in C7(⫺)3B11 was determined by sequencing the PCR products used to identify the mutant (data not shown). The insertion occurred 121 bp upstream of the start codon of dtxR (Fig. 1B) (see Fig. 6). Additional sequencing confirmed that the dtxR allele in C7(⫺)3B11 is identical to the wild-type dtxR allele in the parental strain C7(⫺). In order to determine the effect of this insertion on the ability of DtxR to repress DT production, C7(⫺)3B11 was lysogenized with the DT-encoding bacteriophage , resulting in the strain C7()3B11. Use of an antitoxin halo assay to screen for transposon insertions that result in deregulation of DT secretion. When wild-type C. diphtheriae C7() is plated on iron-depleted M4 agar containing DT antitoxin, the DT secreted by the bacteria interacts with the antitoxin in the medium and results in a halo of white precipitate surrounding each individual bacterial colony (8) (Fig. 2A). On M4 agar containing 10 M FeCl3, DtxR inhibits transcription of the tox gene and no halos are formed by C7() (Fig. 2A). In contrast, halos are formed under these high-iron conditions by C. diphtheriae C7()hm723, which encodes the R47H variant of DtxR (23), and C7()HC3 (8) which has a mutation affecting siderophore-dependent iron uptake as well as a chain-terminating mutation in the codon for residue W104 of DtxR (39). With this in mind, we electroporated C7() with EZTn transposomes and selected the transformants on M4 agar containing DT antitoxin, 10 M FeCl3, and kanamycin. Colonies that grow on this medium contain EZTn insertions, and those that are surrounded by halos must be able to secrete DT under high-iron conditions. After screening approximately 3,500 colonies from an EZTn library of C7(), we identified one insertion mutant that secreted DT in the presence of 10 M FeCl3 (Fig. 2A). This strain was designated C7()18.5, and Southern blot analysis of this mutant demonstrated the presence of a single EZTn insertion (data not shown). PCR analysis of DNA isolated from C7()18.5 with the primers shown in Fig. 1 located the EZTn insertion within dtxR. DNA sequencing pinpointed the insertion 119 bp downstream of the dtxR start site, placing it within the codon for threonine-40 in the helix-turn-helix DNA binding motif in domain 1 of the DtxR polypeptide (Fig. 1C and D and data not shown). Based on the position of this insertion, a complete lack of DtxR function was the most likely explanation for the iron-independent expression of DT by mutant C7()18.5.
J. BACTERIOL.
FIG. 2. Halo assay plates. (A) C7() colonies containing the EZTn transposon at various chromosomal locations were grown on low-iron (0 M FeCl3) and high-iron (10 M FeCl3) M4 agar plates. The DT-overproducing variant C7()18.5 was isolated from the colony indicated by the arrow on the high-iron plate. (B) Complementation of C7()18.5 with the dtxR-expressing plasmid pJKS-dtxR is demonstrated. C7()18.5 containing the vector plasmid pJKS1 is shown as a control. Strains were streaked onto M4 agar plates and incubated for 48 h at 37°C.
In all, we isolated approximately 7,000 independent mutants of C. diphtheriae C7(⫺) or C7() containing insertions of an EZTn transposon. The size of the C. diphtheriae chromosome is approximately 2.49 Mb as per the Sanger Centre sequencing project (www.sanger.ac.uk/Projects/C_diphtheriae), so our present library is not saturating. However, the high frequency of transposition of EZTn insertion in C. diphtheriae will allow us to expand our library easily in the future to saturate the genome of C. diphtheraie. Production of DtxR by transposon insertion mutants. Protein samples from C7(), C7()3B11, C7()18.5, C7()hm723, and C7()HC3 were analyzed for the presence of DtxR in Western blots. Immunoreactive DtxR was not found in protein extracts isolated from C7()18.5 but was detected in protein extracts from all other strains (Fig. 3A). The lack of detectable DtxR production in C7()18.5 was expected since the EZTn insertion in this strain occurs within the coding region for the amino-terminal domain of DtxR. In C7()HC3, the immunoreactive protein detected in the Western blot migrated faster than wild-type DtxR and correlated with the expected size for DtxR truncated at amino acid 104. Previous work indicated that the level of mRNA encoding dtxR did not change significantly in response to iron (47), and our Western blots demonstrated no detectable difference in the amount of DtxR in extracts from wild-type C7() grown
VOL. 184, 2002
FIG. 3. Western blots for DtxR in wild-type and mutant strains of C. diphtheriae. (A) Molecular mass markers are indicated in kilodaltons on the left. Sources of extracts were as follows: lane 1, C7(); lane 2, C7()hm723; lane 3, C7()3B11; lane 4, C7()18.5; and lane 5, C7()HC3. The arrow indicates the truncated DtxR produced by C7()HC3. (B) Lanes 1 to 4, protein extracted from C7(); lanes 5 to 8, protein extracted from C7()3B11. The total protein applied to each well was 100 g in lanes 1 and 5, 50 g in lanes 2 and 6, 20 g in lanes 3 and 7, and 10 g in lanes 4 and 8.
under high- and low-iron conditions (data not shown). To determine if there was a small difference in the expression of dtxR between C7() and C7()3B11, we performed semiquantitative Western blots with several dilutions of extracts containing equal amounts of protein as determined with the bicinchoninic acid protein assay kit (Pierce) (Fig. 3B). Densitometry on
TRANSPOSON MUTAGENESIS OF C. DIPHTHERIAE
5727
the DtxR-specific bands demonstrated that C7()3B11 produced approximately threefold-lower levels of DtxR than C7() under both high-iron (data not shown) and low-iron (Fig. 3B) conditions and that the band intensities were within the linear range for detection of purified DtxR by Western blots in control experiments. Regulation of DT and siderophore production in C7()3B11 and C7()18.5. The ability to repress transcription of tox in the presence of iron and thereby prevent the production of DT is the hallmark of DtxR. The amounts of DT in the media from overnight cultures of C7()3B11 and C7()18.5 under highand low-iron conditions were determined by using an ELISA and were compared with the amounts of DT produced under similar conditions by wild-type C7() and the dtxR mutant C7()hm723. We controlled the effects of batch to batch variation in the medium by using the same batch of deferrated PGT for both the high-iron (with added FeCl3) and low-iron (without added FeCl3) cultures of all strains and by determining the repression ratio, defined as the amount of DT produced under low-iron conditions divided by the amount of DT produced under high-iron conditions. The observed repression ratios did not vary significantly from one batch of medium to another, but the absolute amounts of DT produced did vary from one batch to another (data not shown). In the presence of 10 M FeCl3, C7() produced 200 times less DT than it did in medium containing no added FeCl3 (Fig. 4A), but C7()hm723 culture supernatants contained nearly equal amounts of DT in the presence and absence of FeCl3 (repression ratio of 1.1), as had been reported previously (8). C7()3B11 and C7()18.5 culture supernatants grown in the absence or presence of 10 M FeCl3 also contained nearly equal amounts of DT (repression ratios of 1.9 and 2.0, respec-
FIG. 4. DT production (low versus high iron). (A) The total amount of DT detected in the culture supernatant of the strain indicated under low-iron conditions was divided by the amount detected under high-iron conditions to determine the repression ratio. The average ratio for each strain was determined by using data from at least three sets of low-iron and high-iron cultures. Note that the y axis is a log scale. (B) DT production was measured in the presence or absence of a DtxR-expressing plasmid. The error bars indicate the standard deviations, with data from at least three cultures under each set of conditions.
5728
ORAM ET AL.
tively). These data indicate that neither C7()3B11 nor C7()18.5 was able to repress production of DT in the presence of iron. This result was expected for C7()18.5, since this strain was isolated by its toxin-producing phenotype under high-iron conditions and Western blots had failed to detect DtxR in this strain. It was not expected, however, for C7()3B11, since this strain retained the ability to produce wild-type DtxR, albeit in decreased amounts. The failure of iron to repress DT production by strain C7()3B11 suggests, therefore, that relatively small decreases in the concentration of DtxR in the intracellular milieu can interfere dramatically with DtxR-dependent repression of gene expression by iron. In order to confirm that the phenotypes of C7()3B11 and C7()18.5 were the result of a lack of DtxR, we introduced a plasmid carrying a copy of wild-type dtxR into each strain. Recent independent work in our lab resulted in the construction of a plasmid designated pNGA2 that is stably maintained in C. diphtheriae (Qian et al., unpublished). We replaced the Kmr gene carried on pNGA2 with a spectinomycin resistance (Spcr) gene, aad9 (25), so that we could easily select for its presence in the transposon-containing strains. Next, we added dtxR to the Spcr plasmid, resulting in pJKS-dtxR. Addition of pJKS-dtxR to wild-type C7() did not significantly change the dramatic repression of DT production that occurs under highiron conditions (Fig. 4B). In contrast, the addition of pJKSdtxR to C7()3B11 and C7()18.5 resulted in substantial, but incomplete, restoration of repression of DT production in the presence of iron (Fig. 4B). We tested the stability of pJKSdtxR in all strains and determined that, after overnight growth in both high- and low-iron media, ⬎99% of the cells maintained the plasmid. This result made it unlikely that the slightly greater DT production observed with C7()3B11 and C7()18.5 versus C7() in liquid medium under high-iron conditions in the presence of pJKS-dtxR was primarily from plasmid-cured cells that lacked dtxR and therefore could not repress DT production. We also assayed the complemented strains by Western blot and were able to detect DtxR production in all complemented strains at levels at least 10 times higher than is observed in wild-type C7() containing only the vector plasmid (data not shown). Finally, we confirmed complementation by streaking C7()18.5/pJKS-dtxR onto the antitoxin halo plates used to isolate C7()18.5 (Fig. 2B) and confirmed no halos were produced under high-iron conditions. Therefore, the incomplete complementation by pJKS-dtxR for repression of DT production by iron in mutant strains C7()3B11 and C7()18.5 could not be explained by poor expression of the plasmid-borne dtxR gene. The possibility remains, however, that the transposon insertions in C7()3B11 and C7()18.5 may have polar effects on transcription of a gene homologous to galE that is located immediately downstream of dtxR and is predicted to encode a UDP-galactose 4-epimerase (www.sanger.ac.uk/Projects/C_diphtheriae). In wild-type C. diphtheriae C7(), siderophore and DT production are coregulated, but previous studies suggested that production of siderophores might be more stringently repressed by iron than the production of DT (55). We therefore tested the mutants C7()3B11 and C7()18.5 for their ability to repress siderophore production in the presence of iron and compared them to C7() and C7()hm723. In addition, we added strain C7()HC3 to our analysis, since previous obser-
J. BACTERIOL.
vations indicated that it produced more siderophores under high-iron conditions than any of the other strains assayed (39). In PGT medium, 8 M FeCl3 fully repressed production of DT and siderophores from C7() (data not shown), so we performed the siderophore assays in the presence or absence of 8 M FeCl3. After overnight growth in low-iron PGT medium, supernatants of both mutant and wild-type strains of C. diphtheriae contained 85 to 200 units of siderophore per ml (Fig. 5A). In the presence of 8 M FeCl3, siderophore production decreased to less than 15 units per ml in C7(⫺), C7(), and C7()3B11 (Fig. 5A), while C7()hm723 produced more siderophore (25 units/ml) than wild-type strains but was not as derepressed as either C7()HC3 (61 units/ml) or C7()18.5 (43 units/ml). Our results with wild-type C7() and C7()hm723, which encodes the R47H variant of DtxR, are consistent with previous analyses (55). Since the phenotype of C7()HC3 was comparable to that of C7()18.5, which produces no immunoreactive DtxR, it is likely that the truncation of DtxR in C7()HC3 abolishes DtxR activity and accounts for the inability of that strain to regulate siderophore production in response to iron in the medium. In contrast, C7()3B11, which produces less DtxR than wild-type C7(), retains the ability to repress siderophore production, but not DT production, under high-iron conditions. This finding provides further support for the hypothesis that siderophore production is more stringently repressed by DtxR and iron than is DT production. The supernatants from high-iron cultures of strains C7()HC3 and C7()18.5, both of which fail to produce functional DtxR, contained less detectable siderophore than supernatants from low-iron cultures. However, the CAS assay only detects iron-free siderophores and does not detect ferric siderophore complexes. In supernatants from high-iron cultures we expected some of the siderophore produced by bacterial growth to become complexed with excess iron in the medium, thereby causing an apparent decrease in siderophore production in comparison with supernatants from cultures grown under low-iron conditions. To test this hypothesis, we adjusted the pH of supernatants from low-iron overnight cultures of C7() to values between pH 6.5 and 8.5, added 10 M FeCl3, and then assayed for siderophore (Fig. 5B). When FeCl3 was added to siderophore-containing supernatants at pH 7.5 or less, the amount of siderophore detected by the CAS assay decreased. Adjusting the pH to comparable values without adding FeCl3 did not affect the results of the siderophore assays (data not shown). Interaction of siderophore with iron in the medium is a plausible explanation for the decrease in measurable siderophore produced under high-iron conditions by strains like C7()18.5 and C7()HC3 that lack DtxR, but we cannot rule out completely the possibility that DtxR-independent mechanisms may contribute to regulation of siderophore production by iron in C. diphtheriae. Role for DtxR in protecting C. diphtheriae from oxidative stress. The dtxR mutant C7()18.5 grew poorly in PGT medium containing 10 M FeCl3. After overnight growth in PGT plus 10 M FeCl3, the A600 of a culture of C7()18.5 was often less than 1.0, whereas the A600 of C7() was ⬎5. Furthermore, the bacteria that survived after overnight growth of C7()18.5 under these conditions grew as well as wild-type C7() when they were reinoculated into PGT plus 10 M FeCl3 (data not
VOL. 184, 2002
TRANSPOSON MUTAGENESIS OF C. DIPHTHERIAE
5729
FIG. 6. Growth curves. Cultures were grown at 37°C with shaking, and the A600 was monitored. Plus signs indicate C7(), black squares indicate C7()3B11, and black diamonds indicate C7()18.5.
FIG. 5. Siderophore assays. (A) Total siderophore units (the average of at least three experiments) produced during overnight growth in high (8 M FeCl3)- and low (0 M FeCl3)-iron PGT medium divided by the A600 values of the overnight cultures are shown on the y axis. Error bars indicate the standard deviations. (B) Total siderophore units detected when supernatants from overnight cultures of C7() grown in PGT medium (0 M added FeCl3) were buffered to the pH indicated and 10 M FeCl3 was added before CAS assays for siderophore were performed.
shown). The C7()18.5 cells that survived the high-iron conditions maintained Kmr, and PCR analysis indicated that EZTn was still inserted in dtxR (data not shown). We concluded that these survivors most likely contained suppressor mutations that allowed them to overcome the growth defect of the original C7()18.5 strain that occurred in high-iron medium.
To analyze further the effects of iron on the growth of isogenic wild-type and dtxR variants, we compared the growth curves of C7(), C7()3B11, and C7()18.5 in PGT and PGT plus 10 M FeCl3. The growth of C7() and C7()3B11 was similar in both media, with doubling times of 0.9 to 1.1 h (Fig. 6). In contrast, C7()18.5 exhibited an extended lag phase and slower exponential growth with a doubling time about 1.5 h in PGT and about 4 h in PGT plus 10 M FeCl3. The final growth yield for C7()18.5 in PGT plus 10 M FeCl3 was also much less than that for wild-type C7() or the C7()3B11 mutant strain. Growth of the wild-type and mutant strains, in both high- and low-iron media, caused the pH of the medium to increase from 6.8 to 8.2 over the course of the growth curve. It is unlikely, therefore, that the observed differences in growth were caused by differences in the pHs of the medium. IdeR, the DtxR homologue encoded by M. smegmatis, is required to protect the bacterium from oxidative stress (11). Therefore, we tested the ability of C7()18.5 and the other mutants described above to survive exposure to hydrogen peroxide. We determined the sizes of the zones of inhibition produced when 20 l of various concentrations of H2O2 were applied to 0.6-cm-diameter paper disks in the centers of plates containing heart infusion agar and lawns of various strains of C. diphtheriae. In this assay, C7() was completely resistant to concentrations of H2O2 between 500 M and 1 mM (Table 1). Increasing zones of inhibition for C7() were observed around the disks as concentrations of H2O2 increased. The sizes of the zones of inhibition observed for C7()3B11 were comparable to those observed for C7(), and for C7()hm723, the zones
5730
ORAM ET AL.
J. BACTERIOL.
TABLE 1. Resistance of C. diphtheriae strains to hydrogen peroxide Strain
Inhibition zone diameter (cm) with H2O2 concentration of: 500 M 1 mM 100 mM 1 M 8.8 M
C7() C7()hm723 C7()3B11 C7()18.5
0 0.7 0 1.2
0 0.8 0 1.6
2 2.3 2 2.8
3.1 3.4 2.9 4.5
4.5 4.6 4.5 6.9
% Killing in exposure assay with plasmida: None
pJKS-dtxR
43 ⫾ 13 38 ⫾ 16 32 ⫾ 15 97 ⫾ 1.2
40 ⫾ 5 NT 28 ⫾ 6 27 ⫾ 7
a Values are given as averages ⫾ standard deviations for at least three samples. NT, not tested.
were slightly larger. The largest zones of inhibition were observed with C7()18.5, and at high concentrations of H2O2, the zones were approximately 1.5 times larger than those observed with C7(). These data suggest that DtxR is required for protection against oxidative stress induced by H2O2. In order to confirm these observations, we performed a more-sensitive assay for killing by H2O2 adapted from Keyer and Imlay (24). Cultures were grown in PGT until the A600 was between 0.4 and 0.6, at which time H2O2 was added to the growth medium at a final concentration of 50 mM and the cultures were then incubated for an additional 30 min. Viable counts from each culture were then determined by plating dilutions of the culture onto heart infusion agar. The last two columns of Table 1 show the results of these experiments. C7()18.5, which expressed no detectable DtxR, was much more sensitive to killing by H2O2 than was the wild-type parent or any of the other mutant strains. The presence of pJKS-dtxR in C7()18.5 completely complemented the increased susceptibility to oxidative stress and resulted in wild-type levels of resistance to killing by H2O2. These findings indicate that the lack of DtxR in this mutant strain is wholly responsible for its increased sensitivity to H2O2. DISCUSSION The repertoire of available genetic tools in C. diphtheriae was broadened significantly by our introduction of a system for constructing transposon insertion mutants, a method for increasing transformation efficiency, and identification of spectinomycin resistance as an additional, useful, selectable marker. Recently, use of the Tn5 transposome was described in Rhodococcus (14), a mycolic acid-containing genus related to Corynebacterium, and it seems very likely that this versatile system will be valuable in a variety of organisms for which genetic analysis has thus far proven to be very difficult. We used these new tools to characterize further the role of the multigene regulator DtxR in the survival of and gene regulation in C. diphtheriae. Since DtxR regulates the transcription of multiple genes, and null mutations in dtxR had not previously been characterized, it seemed possible that dtxR was essential for the survival of C. diphtheriae. An insertion mutation in ideR, the gene encoding the DtxR homologue IdeR, was constructed in M. smegmatis, indicating that in this species IdeR is not essential. In other species with DtxR homologues, including M. tuberculosis, null mutations in genes homologous to dtxR have not yet been reported. Fur, an iron-dependent regulator with phenotypic properties resembling those of DtxR, is dispensable for in vitro growth of some species, in-
cluding E. coli and Yersinia pestis (29, 54), but it is an essential gene in others, including Pseudomonas aeruginosa and Neisseria gonorrhoeae (3, 60). Our isolation of the transposon insertion mutant C7()18.5, which contains an EZTn5 insertion within dtxR and produces no DtxR, conclusively demonstrates that DtxR is not required for the growth and survival of C. diphtheriae. Interestingly, the dtxR mutant C7()18.5 is more-easily killed by exposure to high iron conditions or H2O2 than is its wild-type parent. This observation is consistent with the theory that DtxR plays an important role in protecting the cell from damage caused by oxidative stress. Similarly, fur mutants in E. coli are more sensitive to oxidative stress than wild-type strains (59). It seems likely that DtxR regulates genes involved in protecting the cell from oxidative damage, as was observed for IdeR from M. smegmatis (11) and for Fur in several gramnegative bacterial species (58, 60). A role for DtxR in protecting the cell from the damage caused by H2O2 is the first evidence that DtxR may regulate gene expression in response to a signal other than iron starvation, namely oxidative stress. It is also possible that DtxR has a direct role in binding and sequestering free iron in the cell and thereby preventing it from damaging cellular components. Interestingly, in gramnegative bacteria, Fur, in addition to its function as a transcriptional regulator in response to both iron and oxidative stress, also regulates genes required for the acid shock response (15). The functional similarities between Fur and DtxR suggest that there may be a role for DtxR in the regulation of gene expression in response to acid shock as well. We demonstrated that in the transposon insertion mutant C7()3B11, DtxR is produced at a lower level than it is in wild-type C7(). Based on sequence analysis, other authors suggested that dtxR is transcribed from a promoter positioned approximately 60 bp upstream of the dtxR start codon (Fig. 7) (5). The EZTn insertion in C7()3B11 occurred 121 bp upstream of the dtxR ATG initiation codon, indicating that an insertion that far upstream can and does affect dtxR expression. The open reading frame upstream of dtxR is predicted to encode a sigma factor (5) that is transcribed in the same direction as dtxR. Our data are consistent with the idea that there may be a single transcript that encodes the sigma factor and DtxR. In addition, we noticed a possible ⫺10 promoter sequence located 135 bp upstream of the dtxR start codon that could be part of a promoter used to transcribe a message containing only dtxR (Fig. 6). Amplification of RNA isolated from wild-type C7() by reverse transcriptase PCR (RT-PCR) with a primer positioned within the 5⬘ end of dtxR and RT followed by PCR with primers within the 3⬘ end of the upstream proposed sigma gene and the 5⬘ end of dtxR indicated that there is a transcript that spans the intergenic region (data not shown). Thus far, attempts to identify the 5⬘ end of the transcript(s) carrying dtxR by primer extension analysis have been unsuccessful. Further characterization of the transcripts carrying dtxR and its promoter region will be required to determine the precise molecular basis for the effect of the EZTn insertion on gene expression in C7()3B11. At least one transcript encoding DtxR is also likely to encode the GalE homologue located immediately downstream of dtxR in the C. diphtheriae chromosome (www.sanger.ac.uk/ Projects/C_diphtheriae). In Brevibacterium lactofermentum,
VOL. 184, 2002
TRANSPOSON MUTAGENESIS OF C. DIPHTHERIAE
5731
FIG. 7. Sequence upstream of dtxR. The possible ⫺10 and ⫺35 promoter sequences identified by Boyd et al. (5) are underlined once. The ribosome-binding site proposed by Boyd et al. is underlined and labeled rbs. The EZTn insertion site in C7()3B11 is indicated by the plus sign. The possible ⫺10 sequence upstream from the EZTn insertion site in C7()3B11 is underlined twice.
dtxR and galE are cotranscribed, and in addition, there is at least one other transcript that contains dtxR but does not include galE (35). In wild-type C. diphtheriae C7(), RT-PCR with a reverse primer positioned within the 5⬘ end of galE and a forward primer within dtxR indicated that the two genes are cotranscribed (data not shown). Our transposon insertion mutants are likely, therefore, to have polar effects on the transcription of galE. Misregulation or lack of GalE may account for the lack of complete complementation for the regulation of DtxR by iron that was observed when DtxR was provided in trans from a plasmid. The possible mechanism by which GalE may affect iron-dependent DtxR gene regulation is unknown and will require additional investigation. Finally, the phenotype of C7()3B11 also suggests that a small decrease in the amount of DtxR in the cell can dramatically affect gene regulation. Our assays indicated that DtxR was produced by C7()3B11, but at an approximately threefold-lower level than in wild-type C7(), resulting in deregulation of DT but maintaining normal regulation of siderophore production. These data are consistent with observations made previously about C7()hm723. C7()hm723 produces a DT repressor protein that contains an amino acid change at position 47 from an R to an H. This mutant form of DtxR retains some ability to regulate siderophore production but is unable to repress DT production in the presence of iron. Both our observations with C7()3B11 and the previous observations with C7()hm723 support the hypothesis that gene regulation inside the C. diphtheriae cell is very responsive to small changes in the activity or amount of DtxR (47). Gene regulation by DtxR in C. diphtheriae is a complex and delicate process that we are only just beginning to characterize and understand. The tools we have developed in this work have helped shed light on several interesting facets of regulation by DtxR and will greatly aid further characterization of gene function and regulation in C. diphtheriae. We have demonstrated that although dtxR is not an essential gene in C. diphtheriae, it is required for survival under conditions of oxidative stress. In addition, our observation that a transposon insertion located 121 bp upstream of the coding region has an effect on the concentration of DtxR in the cell indicates that the regulation of dtxR expression is more complex than first proposed and may involve more than one RNA transcript. Finally, we have also shown that regulation of DT production by DtxR is extremely sensitive to the intracellular concentration of DtxR. Aberrations in DT production appear as the result of a decrease in DtxR concentration as small as threefold, whereas
the regulation of siderophore production is affected little by a change of this magnitude. These findings are consistent with previous observations that the affinity of DtxR-regulated promoters and operators for metal-activated DtxR may vary over a considerable range (26, 39). ACKNOWLEDGMENTS This work was supported by grant number AI14107 from NIAID, NIH. We thank Jennifer Spinler for constructing and providing the pJKS1 plasmid from which pJKS-dtxR was derived. REFERENCES 1. Akhtar, M. K., N. Kaderbhai, and M. A. Kaderbhai. 2000. Growth of Escherichia coli on medium containing glycine increases transformation efficiency. Anal. Biochem. 277:273–276. 2. Barksdale, L. W., and A. M. J. Pappenheimer. 1954. Phage-host relationships in nontoxigenic and toxigenic diphtheria bacilli. J. Bacteriol. 67:220– 232. 3. Berish, S. A., S. Subbarao, C. Y. Chen, D. L. Trees, and S. A. Morse. 1993. Identification and cloning of a fur homolog from Neisseria gonorrhoeae. Infect. Immun. 61:4599–4606. 4. Boland, C. A., and W. G. Meijer. 2000. The iron dependent regulatory protein IdeR (DtxR) of Rhodococcus equi. FEMS Microbiol. Lett. 191:1–5. 5. Boyd, J., M. N. Oza, and J. R. Murphy. 1990. Molecular cloning and DNA sequence analysis of a diphtheria tox iron-dependent regulatory element (dtxR) from Corynebacterium diphtheriae. Proc. Natl. Acad. Sci. USA 87: 5968–5972. 6. Craig, N. L. 1997. Target site selection in transposition. Annu. Rev. Biochem. 66:437–474. 7. Cruz-Rodz, A. L., and M. S. Gilmore. 1990. High efficiency introduction of plasmid DNA into glycine treated Enterococcus faecalis by electroporation. Mol. Gen. Genet. 224:152–154. 8. Cryz, S. J., L. M. Russell, and R. K. Holmes. 1983. Regulation of toxinogenesis in Corynebacterium diphtheriae: mutations in the bacterial genome that alter the effects of iron on toxin production. J. Bacteriol. 154:245–252. 9. Doukhan, L., M. Predich, G. Nair, O. Dussurget, I. Mandic-Mulec, S. T. Cole, D. R. Smith, and I. Smith. 1995. Genomic organization of the mycobacterial sigma gene cluster. Gene 165:67–70. 10. Dubrac, S., and D. Touati. 2000. Fur positive regulation of iron superoxide dismutase in Escherichia coli: functional analysis of the sodB promoter. J. Bacteriol. 182:3802–3808. 11. Dussurget, O., M. Rodriguez, and I. Smith. 1996. An ideR mutant of Mycobacterium smegmatis has derepressed siderophore production and an altered oxidative-stress response. Mol. Microbiol. 22:535–544. 12. Dussurget, O., M. Rodriguez, and I. Smith. 1998. Protective role of the Mycobacterium smegmatis IdeR against reactive oxygen species and isoniazid toxicity. Tuber. Lung Dis. 79:99–106. 13. Feese, M. D., E. Pohl, R. K. Holmes, and W. G. J. Hol. 2001. Iron-dependent regulators, p. 850–863. In A. Messerschmidt, R. Huber, R. Poulos, and K. Wieghardt (ed.), Handbook of metalloproteins. John Wiley & Sons, Ltd., Chichester, United Kingdom. 14. Fernandes, P. J., J. A. Powell, and J. A. Archer. 2001. Construction of Rhodococcus random mutagenesis libraries using Tn5 transposition complexes. Microbiology 147:2529–2536. 15. Foster, J. W., and M. Moreno. 1999. Inducible acid tolerance mechanisms in enteric bacteria. Novartis Found. Symp. 221:55–69, 70–74. 16. Freeman, V. J. 1951. Studies on the virulence of bacteriophage-infected strains of Corynebacterium diphtheriae. J. Bacteriol. 61:675–688.
5732
ORAM ET AL.
17. Goryshin, I. Y., J. Jendrisak, L. M. Hoffman, R. Meis, and W. S. Reznikoff. 2000. Insertional transposon mutagenesis by electroporation of released Tn5 transposition complexes. Nat. Biotechnol. 18:97–100. 18. Goryshin, I. Y., and W. S. Reznikoff. 1998. Tn5 in vitro transposition. J. Biol. Chem. 273:7367–7374. 19. Gunter-Seeboth, K., and T. Schupp. 1995. Cloning and sequence analysis of the Corynebacterium diphtheriae dtxR homologue from Streptomyces lividans and S. pilosus encoding a putative iron repressor protein. Gene 166:117–119. 20. Hill, P. J., A. Cockayne, P. Landers, J. A. Morrissey, C. M. Sims, and P. Williams. 1998. SirR, a novel iron-dependent repressor in Staphylococcus epidermidis. Infect. Immun. 66:4123–4129. 21. Holmes, R. K., and L. Barksdale. 1969. Genetic analysis of tox⫹ and tox⫺ bacteriophages of Corynebacterium diphtheriae. J. Virol. 3:586–598. 22. Holmes, R. K., and E. M. Twiddy. 1983. Characterization of monoclonal antibodies that react with unique and cross-reacting determinants of cholera enterotoxin and its subunits. Infect. Immun. 42:914–923. 23. Kanei, C., T. Uchida, and M. Yoneda. 1977. Isolation from Corynebacterium diphtheriae C7() of bacterial mutants that produce toxin in medium with excess iron. Infect. Immun. 18:203–209. 24. Keyer, K., and J. A. Imlay. 1996. Superoxide accelerates DNA damage by elevating free-iron levels. Proc. Natl. Acad. Sci. USA 93:13635–13640. 25. LeBlanc, D. J., L. N. Lee, and J. M. Inamine. 1991. Cloning and nucleotide base sequence analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob. Agents Chemother. 35:1804–1810. 26. Lee, J. H., and R. K. Holmes. 2000. Characterization of specific nucleotide substitutions in DtxR-specific operators of Corynebacterium diphtheriae that dramatically affect DtxR binding, operator function, and promoter strength. J. Bacteriol. 182:432–438. 27. Lee, J. H., T. Wang, K. Ault, J. Liu, M. P. Schmitt, and R. K. Holmes. 1997. Identification and characterization of three new promoter/operators from Corynebacterium diphtheriae that are regulated by the diphtheria toxin repressor (DtxR) and iron. Infect. Immun. 65:4273–4280. 28. Lee, S. Y., and S. Rasheed. 1990. A simple procedure for maximum yield of high-quality plasmid DNA. BioTechniques 9:676–679. 29. Litwin, C. M., and S. B. Calderwood. 1993. Role of iron in regulation of virulence genes. Clin. Microbiol. Rev. 6:137–149. 30. Manabe, Y. C., B. J. Saviola, L. Sun, J. R. Murphy, and W. R. Bishai. 1999. Attenuation of virulence in Mycobacterium tuberculosis expressing a constitutively active iron repressor. Proc. Natl. Acad. Sci. USA 96:12844–12848. 31. McIver, K. S., and J. R. Scott. 1997. Role of mga in growth phase regulation of virulence genes of the group A streptococcus. J. Bacteriol. 179:5178–5187. 32. Mietzner, T. A., and S. A. Morse. 1994. The role of iron-binding proteins in the survival of pathogenic bacteria. Annu. Rev. Nutr. 14:471–493. 33. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 34. Mueller, J. H., and P. A. Miller. 1941. Production of diphtheric toxin of high potency (100 Lf) on a reproducible medium. J. Immunol. 40:21–32. 35. Oguiza, J. A., A. T. Marcos, M. Malumbres, and J. F. Martin. 1996. The galE gene encoding the UDP-galactose 4-epimerase of Brevibacterium lactofermentum is coupled transcriptionally to the dmdR gene. Gene 177:103–107. 36. Oguiza, J. A., X. Tao, A. T. Marcos, J. F. Martin, and J. R. Murphy. 1995. Molecular cloning, DNA sequence analysis, and characterization of the Corynebacterium diphtheriae dtxR homolog from Brevibacterium lactofermentum. J. Bacteriol. 177:465–467. 37. Pappenheimer, A. M., Jr. 1977. Diphtheria toxin. Annu. Rev. Biochem. 46:69–94. 38. Posey, J. E., J. M. Hardham, S. J. Norris, and F. C. Gherardini. 1999. Characterization of a manganese-dependent regulatory protein, TroR, from Treponema pallidum. Proc. Natl. Acad. Sci. USA 96:10887–10892. 39. Qian, Y., J. H. Lee, and R. K. Holmes. 2002. Identification of a DtxRregulated operon that is essential for siderophore-dependent iron uptake in Corynebacterium diphtheriae. J. Bacteriol. 184:4846–4856. 40. Qiu, X., E. Pohl, R. K. Holmes, and W. G. Hol. 1996. High-resolution structure of the diphtheria toxin repressor complexed with cobalt and manganese reveals an SH3-like third domain and suggests a possible role of phosphate as co-corepressor. Biochemistry 35:12292–12302.
J. BACTERIOL. 41. Qiu, X., C. L. Verlinde, S. Zhang, M. P. Schmitt, R. K. Holmes, and W. G. Hol. 1995. Three-dimensional structure of the diphtheria toxin repressor in complex with divalent cation co-repressors. Structure 3:87–100. 42. Que, Q., and J. D. Helmann. 2000. Manganese homeostasis in Bacillus subtilis is regulated by MntR, a bifunctional regulator related to the diphtheria toxin repressor family of proteins. Mol. Microbiol. 35:1454–1468. 43. Schiering, N., X. Tao, H. Zeng, J. R. Murphy, G. A. Petsko, and D. Ringe. 1995. Structures of the apo- and the metal ion-activated forms of the diphtheria tox repressor from Corynebacterium diphtheriae. Proc. Natl. Acad. Sci. USA 92:9843–9850. 44. Schmitt, M. P. 1997. Transcription of the Corynebacterium diphtheriae hmuO gene is regulated by iron and heme. Infect. Immun. 65:4634–4641. 45. Schmitt, M. P., and E. S. Drazek. 2001. Construction and consequences of directed mutations affecting the hemin receptor in pathogenic Corynebacterium species. J. Bacteriol. 183:1476–1481. 46. Schmitt, M. P., and R. K. Holmes. 1993. Analysis of diphtheria toxin repressor-operator interactions and characterization of a mutant repressor with decreased binding activity for divalent metals. Mol. Microbiol. 9:173–181. 47. Schmitt, M. P., and R. K. Holmes. 1991. Characterization of a defective diphtheria toxin repressor (dtxR) allele and analysis of dtxR transcription in wild-type and mutant strains of Corynebacterium diphtheriae. Infect. Immun. 59:3903–3908. 48. Schmitt, M. P., and R. K. Holmes. 1994. Cloning, sequence, and footprint analysis of two promoter/operators from Corynebacterium diphtheriae that are regulated by the diphtheria toxin repressor (DtxR) and iron. J. Bacteriol. 176:1141–1149. 49. Schmitt, M. P., and R. K. Holmes. 1991. Iron-dependent regulation of diphtheria toxin and siderophore expression by the cloned Corynebacterium diphtheriae repressor gene dtxR in C. diphtheriae C7 strains. Infect. Immun. 59:1899–1904. 50. Schmitt, M. P., E. M. Twiddy, and R. K. Holmes. 1992. Purification and characterization of the diphtheria toxin repressor. Proc. Natl. Acad. Sci. USA 89:7576–7580. 51. Schwyn, B., and J. B. Neilands. 1987. Universal chemical assay for the detection and determination of siderophores. Anal. Biochem. 160:47–56. 52. Serwold-Davis, T. M., N. Groman, and M. Rabin. 1987. Transformation of Corynebacterium diphtheriae, Corynebacterium ulcerans, Corynebacterium glutamicum, and Escherichia coli with the C. diphtheriae plasmid pNG2. Proc. Natl. Acad. Sci. USA 84:4964–4968. 53. Serwold-Davis, T. M., and N. B. Groman. 1986. Mapping and cloning of Corynebacterium diphtheriae plasmid pNG2 and characterization of its relatedness to plasmids from skin coryneforms. Antimicrob. Agents Chemother. 30:69–72. 54. Staggs, T. M., J. D. Fetherston, and R. D. Perry. 1994. Pleiotropic effects of a Yersinia pestis fur mutation. J. Bacteriol. 176:7614–7624. 55. Tai, S. P., A. E. Krafft, P. Nootheti, and R. K. Holmes. 1990. Coordinate regulation of siderophore and diphtheria toxin production by iron in Corynebacterium diphtheriae. Microb. Pathog. 9:267–273. 56. Tao, X., J. Boyd, and J. R. Murphy. 1992. Specific binding of the diphtheria tox regulatory element DtxR to the tox operator requires divalent heavy metal ions and a 9-base-pair interrupted palindromic sequence. Proc. Natl. Acad. Sci. USA 89:5897–5901. 57. Tao, X., and J. R. Murphy. 1992. Binding of the metalloregulatory protein DtxR to the diphtheria tox operator requires a divalent heavy metal ion and protects the palindromic sequence from DNase I digestion. J. Biol. Chem. 267:21761–21764. 58. Touati, D. 2000. Iron and oxidative stress in bacteria. Arch. Biochem. Biophys. 373:1–6. 59. Touati, D., M. Jacques, B. Tardat, L. Bouchard, and S. Despied. 1995. Lethal oxidative damage and mutagenesis are generated by iron in delta fur mutants of Escherichia coli: protective role of superoxide dismutase. J. Bacteriol. 177:2305–2314. 60. Vasil, M. L., and U. A. Ochsner. 1999. The response of Pseudomonas aeruginosa to iron: genetics, biochemistry and virulence. Mol. Microbiol. 34:399– 413.
