TEXT

Top Abstract Text References

Vibrios are gram-negative marine bacteria which often cause disease in humans. Infections by Vibrio cholerae (7), Vibrio fluvialis (4, 15), and Vibrio parahaemolyticus (12) are acquired through consumption of contaminated water or seafood, and they lead to excessive watery diarrhea (V. cholerae and V. fluvialis) or to acute gastroenteritis (V. parahaemolyticus). Vibrio alginolyticus (5, 20) causes extraintestinal infections, such as wound infections.

Bacterial pathogens must acquire iron inside the host to multiply to numbers sufficient to cause disease (for reviews, see references 14 and 19). V. cholerae and V. parahaemolyticus acquire iron by at least two methods. Under low-iron conditions, they produce the siderophores vibriobactin (9) and vibrioferrin (25), respectively. These low-molecular-weight compounds bind iron with high affinity and are transported back into the cell. Both Vibrio species also acquire iron from heme or hemoglobin (22, 23, 26). V. fluvialis and V. alginolyticus produce siderophores (2), but neither has been tested for the ability to utilize heme or hemoglobin as an iron source.

V. cholerae heme iron utilization involves the following genes: hutA, which encodes the heme receptor (10, 11); tonB1, which encodes an inner membrane protein required for the transport of heme into the periplasm; exbB1D1, which encodes the inner membrane proteins required for TonB function; and hutB and hutCD, which encode a periplasmic binding protein and a cytoplasmic membrane permease, respectively, which are involved in the transport of heme to the cytoplasm (18). A second V. cholerae TonB system encoded by tonB2 and exbB2D2, which also may be involved in heme transport, recently has been identified (18). To date, no other gram-negative bacterium has been found to contain two TonB systems.

Heme iron utilization systems have been studied for V. cholerae and Vibrio vulnificus (13), but heme iron utilization systems in other vibrios have not been well characterized. The goals of this study were to (i) identify vibrios that can acquire iron from heme or hemoglobin, (ii) determine if these species have heme utilization and tonB-like genes similar to those in V. cholerae, and (iii) determine if heme utilization proteins of V. cholerae and other heme-utilizing vibrios have similar amino acid sequences and whether they are functionally interchangeable.

Testing Vibrio species for heme and hemoglobin iron utilization. The ability of various strains to use several iron-containing compounds was tested. In the assay, cultures were seeded into Luria (L) agar containing the iron chelator ethylenediamine-di(o-hydroxyphenylacetic acid) (EDDA), and 5 µl of each iron-containing compound was spotted onto the media. All of the wild-type vibrios tested, including both clinical and environmental isolates, exhibited substantial zones of growth around the heme and hemoglobin spots (Table 1). V. cholerae DHH-11, a heme iron utilization deletion mutant (Table 2), exhibited no detectable growth around the heme or hemoglobin spots, and it served as the negative control (Table 1). All of the strains tested could utilize FeSO₄ as an iron source.

Determining if the Vibrio species contain DNA similar to V. cholerae heme iron utilization genes. Chromosomal preparations (17) were digested with HindIII, electrophoresed on agarose gels, and subjected to Southern blotting on charged nylon membranes (Boehringer Mannheim, Indianapolis, Ind.) under low-stringency conditions (21). Chromosomal DNA from the noncholera strains was loaded onto the gels at a threefold-higher concentration than that of the V. cholerae chromosomal DNA. The Genius DIG DNA labeling and detection system with CDP Star (Boehringer Mannheim) was used to label the probes and detect hybridization. The hutA probe contained a 1.8-kb EcoRI fragment from pHUT3 (10). The probes for tonB1, exbB1, and hutC and for tonB2 were generated by PCR from pHUT7 and pOUT11 (18), respectively, and were internal fragments of each gene. All of the strains tested contained DNA sequences similar to tonB2 (Fig. 1; Table 3), exbB1, and hutC (Table 3), whereas all the strains except V. fluvialis contained DNA sequences similar to hutA (Table 3) and tonB1 (Fig. 1; Table 3). The hybridization signal generated with the tonB2 probe in V. fluvialis DNA (Fig. 1) was barely detectable, but upon prolonged exposure of the blot to X-ray film, a distinct signal at 5.2 kb was detected (Table 3). The tonB1 probe hybridized to different-sized HindIII fragments than the tonB2 probe in V. alginolyticus and the three strains of V. parahaemolyticus, suggesting that these strains contained two distinct tonB-like genes (Fig. 1; Table 3). The hutA probe hybridized to two different fragments in all the strains in which a signal was generated (Table 3). This may reflect the presence of an internal HindIII site in each gene. In all the Southern blots, the hybridization signal of each probe which hybridized to DNA from the V. cholerae chromosome was significantly more intense than those obtained with the other Vibrio species.

View larger version (62K): [in this window] [in a new window]   FIG. 1.   Autoradiograms of Southern blots of Vibrio strains probed with V. cholerae tonB1 (A) and tonB2 (B). The following strains were tested: V. cholerae CA401 (lane 1); V. parahaemolyticus 474801, M47314, and 115 (lanes 2 to 4, respectively); V. alginolyticus BE2-542 (lane 5); and V. fluvialis BE2-819 (lane 6). Size markers are indicated on the left.

Isolation of a heme iron utilization mutant of V. alginolyticus. To isolate a heme iron utilization mutant of V. alginolyticus BE2-542, diethylsulfate mutagenesis and nalidixic acid enrichment were performed as previously described (10). Colonies were screened for growth on L-EDDA-hemin agar, and the heme utilization mutant, DTH-1, was isolated. When assayed as described above, DTH-1 exhibited no detectable growth around the spots containing heme or hemoglobin (Table 1). To determine if DTH-1 produced and/or utilized the V. alginolyticus siderophore, fully grown cultures of the mutant and its parent strain were spotted onto L-EDDA agar seeded with either bacterial strain. Significant zones of growth occurred around the spots of the mutant and the wild type on both plates, indicating that DTH-1 both produces and utilizes its siderophore. These data suggest that the heme utilization defect did not affect the siderophore synthesis or transport system.

Complementation of V. alginolyticus DTH-1 with the V. cholerae TonB1 system. To determine if the defect in DTH-1 was in a TonB system, the mutant was transformed by electroporation (18) with pHUT11, which contains V. cholerae tonB1 and exbB1D1, or with the vector pACYC184. DTH-1/pHUT11 grew as well as the parent strain, BE2-542, in L-EDDA-hemin broth (Table 4), suggesting that DTH-1 has a defective TonB, ExbB, or ExbD protein and that the comparable V. cholerae protein is functionally interchangeable. Combined with the data in Table 1 showing that the mutation in DTH-1 had no effect on siderophore production or transport, these data indicate that the V. alginolyticus siderophore transport system may use a second TonB system. Occhino et al. (18) recently determined that mutations in either the TonB1 system or the TonB2 system in V. cholerae had no impact on heme iron or vibriobactin uptake in V. cholerae. However, when both TonB systems in V. cholerae were defective, both heme iron and vibriobactin utilization were disrupted. Thus, either TonB system in V. cholerae can function in both siderophore and heme uptake. This does not appear to be the case in V. alginolyticus, where apparent disruption of one of the TonB systems leads to the loss of only heme iron utilization, not siderophore uptake. Thus, V. alginolyticus may contain one TonB system that plays a role in heme uptake and another that plays a role in siderophore uptake.

Isolation of a V. parahaemolyticus gene that is functionally interchangeable with V. cholerae hutA. A cosmid library of V. parahaemolyticus 474801 DNA was constructed (10) and transferred by triparental mating to Escherichia coli 1017 containing pHUT10 (10), which contains all the V. cholerae heme iron utilization genes except hutA. Bacteria were plated on L-EDDA-hemin agar, and a heme utilization-positive isolate was identified. The cosmid was named pPHU1 (parahaemolyticus heme utilization), and E. coli 1017/pHUT10/pPHU1 grew to a density similar to that of E. coli 1017/pHUT10/pHUT2 (pHUT2 contains the V. cholerae hutA gene) when tested for growth in L-EDDA-hemin broth (Table 4). These data suggested that pPHU1 contains the V. parahaemolyticus hutA equivalent and that it is functionally interchangeable with V. cholerae hutA. E. coli 1017 transformed with pPHU1 alone grew poorly in L-EDDA-hemin medium, indicating that it needs tonB1 and accessory genes present on pHUT10. To confirm that pPHU1 contained the V. parahaemolyticus hutA equivalent, pPHU1 was digested with HindIII, electrophoresed on a gel, and probed with the hutA probe described above. The probe annealed to cosmid clone fragments of 2.9 and 0.6 kb, which are the same size as the fragments observed in Southern blots of genomic DNA (Table 3 and data not shown).

Cloning and sequencing of a portion of the tonB1 locus from V. parahaemolyticus. Additional work was performed to confirm that V. parahaemolyticus has a TonB1 system similar to that in V. cholerae. As indicated in Table 3, both the tonB1 and exbB1 probes hybridized to a 2.8-kb HindIII fragment from V. parahaemolyticus, suggesting that these two genes are linked, as they are in V. cholerae (18). To clone the tonB1 and exbB1 genes, 2.5- to 3.4-kb HindIII fragments from V. parahaemolyticus 474801 chromosomal preparations were isolated, ligated into pACYC184, and transformed into E. coli DH5. Tetracycline-sensitive colonies were pooled into groups of 25, and plasmids were screened by Southern hybridization with the V. cholerae tonB1 probe. A clone was identified (pPHU2) to which the V. cholerae tonB1 and exbB1 probes, but not the tonB2 or hutC probes, hybridized. This indicated that pPHU2 contained tonB1 and exbB1, but not hutC or tonB2.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Genetic map of cloned DNA from pPHU2. The putative Fur box is indicated by an open box. The arrows labeled P1 and P2 indicate the locations of the proposed promoters for phuW and for tonB1 and exbB1, respectively. The arrows beneath the filled boxes denote the direction of transcription. Below the map is the DNA sequence containing the predicted divergent promoters and the proposed Fur box, which is marked with a thick black line above the DNA.

Nucleotide sequence accession number. The nucleotide and amino acid sequences corresponding to this region can be found under GenBank/EMBL accession no. AF119047.