Identification of a DtxR-regulated operon that is essential for siderophore-dependent iron uptake in Corynebacterium diphtheriae

Abstract

The diphtheria toxin repressor (DtxR) uses Fe(2+) as a corepressor and inhibits transcription from iron-regulated promoters (IRPs) in Corynebacterium diphtheriae. A new IRP, designated IRP6, was cloned from C. diphtheriae by a SELEX-like procedure. DtxR bound to IRP6 in vitro only in the presence of appropriate divalent metal ions, and repression of IRP6 by DtxR in an Escherichia coli system was iron dependent. The open reading frames (ORFs) downstream from IRP6 and previously described promoter IRP1 were found to encode proteins homologous to components of ATP-binding cassette (ABC) transport systems involved in high-affinity iron uptake in other bacteria. IRP1 and IRP6 were repressed under high-iron conditions in wild-type C. diphtheriae C7(beta), but they were expressed constitutively in C7(beta) mutant strains HC1, HC3, HC4, and HC5, which were shown previously to be defective in corynebactin-dependent iron uptake. A clone of the wild-type irp6 operon (pCM6ABC) complemented the constitutive corynebactin production phenotype of HC1, HC4, and HC5 but not of HC3, whereas a clone of the wild-type irp1 operon failed to complement any of these strains. Complementation by subclones of pCM6ABC demonstrated that mutant alleles of irp6A, irp6C, and irp6B were responsible for the phenotypes of HC1, HC4, and HC5, respectively. The irp6A allele in HC1 and the irp6B allele in HC5 encoded single amino acid substitutions in their predicted protein products, and the irp6C allele in HC4 caused premature chain termination of its predicted protein product. Strain HC3 was found to have a chain-terminating mutation in dtxR in addition to a missense mutation in its irp6B allele. These findings demonstrated that the irp6 operon in C. diphtheriae encodes a putative ABC transporter, that specific mutant alleles of irp6A, irp6B, and irp6C are associated with defects in corynebactin-dependent iron uptake, and that complementation of these mutant alleles restores repression of corynebactin production under high-iron growth conditions, most likely as a consequence of restoring siderophore-dependent iron uptake mediated by the irp6 operon.

FIG. 1.

(A) DtxR binding of IRP6 requires a divalent transitional metal ion, such as Co²⁺. (B) Effect of DtxR concentration on formation of DtxR-DNA complexes in gel mobility shift assays. Detectable shifts of IRP6 occurred at 100 nM DtxR, while detectable shifts of IRP1 started at 5 nM DtxR. Co²⁺ was present in all samples at 300 μM.

FIG. 2.

(A) Nucleotide sequence of IRP6. The primary DtxR binding site defined by DNase I footprinting assays with DtxR at 0.1 μM is indicated by boldface. The longer nucleotide sequence protected from DNase I by 0.5 μM DtxR is underlined and includes the primary DtxR binding site. (B and C) DNase I footprinting assays. All the fragments were 3′ end labeled with [α-³²P]dCTP on one strand and incubated in the presence of Co²⁺ (300 μM) and DtxR (0, 0.1, or 0.5 μM). Brackets indicate the sequences protected by DtxR from DNase I digestion. (B) With IRP6, a longer 60-bp sequence (b) was protected by 0.5 μM DtxR and a shorter 28-bp sequence (a) that is contained within sequence b was significantly protected by 0.1 μM DtxR. Although the total radioactivity loaded in the lane with 0.1 μM DtxR was greater than that in the other lanes, the intensity of the bands within the 28-bp (a) sequence was decreased significantly both in comparison withbands in other regions of the same lane and with bands in the same 28-bp region in the lane without DtxR. (C) In IRP1, two contiguous regions (a and b) were protected by 0.5 μM DtxR but only region a was protected by 0.1 μM DtxR.

FIG. 3.

Compilation of the 19-mer core sequences of DtxR binding sites and the consensus sequence. Dots above nucleotides indicate that the nucleotide matches the nucleotide of the consensus sequence. The identity score indicates the numbers of times that nucleotide was found among all the aligned DtxR binding sites. Arrows, inverted repeats within the core consensus sequence. The identity of each DtxR binding site with the consensus is indicated.

FIG. 4.

Organization of irp6 operon. (A) Restriction map of the 6.8-kb chromosomal HindIII fragment containing the irp6 operon. Long arrows, orientations of the three ORFs; short arrows, locations and orientations of the primers used to amplify the entire irp6 region, or regions containing only the first ORF (irp6A) or the first two ORFs (irp6AB) of the irp6 operon; black bar, location of IRP6 promoter fragment isolated by the SELEX-like procedure. (B) Restriction map of library clone pSK6a, containing the 1.9-kb EcoRI fragment that hybridized to an IRP6 probe, and library clone pSK6e, containing the 2.2-kb EcoRI fragment that is adjacent to pSK6a on the chromosome. Short arrows, locations of the primers (6C2 and 6D2) used in inverse PCR to amplify the flanking sequence on the 6.8-kb HindIII fragment. (C) Restriction map of the 3.6-kb SalI-BamHI insert containing the irp6 region in plasmid pCM6ABC. The asterisk indicates that the BamHI site was generated by primer 3Q2. Restriction enzyme abbreviations: B, BamHI; E, EcoRI; H, HindIII; S, SalI; P, PstI.

FIG. 5.

β-Galactosidase activities of C. diphtheriae strains C7(β), HC1, HC3, HC4, and HC5 carrying promoter fusion constructs pCM6 (A) and pCM1(B) cultured in HITW, HITW-Fe, and HITW with 100 μg of EDDA/ml. Overnight cultures were used for β-galactosidase activity assays, and the activity values were the averages of samples from cultures grown in triplicate. The standard deviations are shown by the error bars.

FIG. 6.

Gene organization of irp1 operon. Dashed lines indicate the location of the four ORFs (irp1A, irp1B, irp1C, and irp1D) in the 8-kb EcoRI insert in plasmid pWR382. Restriction enzyme abbreviations: B, BamHI; E, EcoRI; S, SalI.