Two novel proteins, TtpB2 and TtpD2, are essential for iron transport in the TonB2 system of Vibrio vulnificus

Abstract In gram‐negative bacteria, energy‐dependent active transport of iron‐bound substrates across the outer membrane is achieved through the TonB systems of proteins. Three TonB systems have been identified in the human pathogen Vibrio vulnificus. The TonB1 system contains three proteins: TonB1, ExbB1, and ExbD1. Both the TonB2 and TonB3 systems have been shown to also contain a fourth protein, TtpC2 and TtpC3, respectively. Here, we report and begin to characterize two additional proteins in the TonB2 and TonB3 systems: TtpB and TtpD. Both TtpB2 and TtpD2 are absolutely required for the function of the TonB2 system in V. vulnificus. However, although both TtpB3 and TtpD3 in the TonB3 system are related to the proteins in the TonB2 system, neither are active in iron transport. All six protein components of the TonB2 system—TonB2, ExbB2, ExbD2, TtpB2, TtpC2, and TtpD2—are essential for the uptake of both endogenously produced iron‐bound siderophores and exogenous siderophores produced from other organisms. Through complementation, we have shown that V. vulnificus is capable of using different TtpD2 proteins from other Vibrio species to bring in multiple siderophores. In contrast, we also demonstrate that TtpB2 must come from V. vulnificus, and not other species within the genus, to complement mutations in the TonB2 system.

Iron is an essential element required for survival by nearly every living organism. Iron can be highly versatile, acting as both an electron donor and an electron acceptor, as well as being involved in processes ranging from cell signaling to metabolism (Crosa, 1997).
Gram-negative bacteria, such as V. vulnificus, possess an outer membrane (OM) that contains a large assortment of protein receptors. These receptors bind nutrients, minerals, and iron-bound siderophore complexes. The inner membrane (IM) is where the energy for the cell is derived. This can be in the form of adenosine triphosphate (ATP) or the proton motive force (PMF), which provides the energy needed to power the uptake of iron-bound siderophores.
The periplasmic space between these two membranes presents a challenge to gram-negative bacteria. In order to breach that space, energy created by the PMF is transduced to the OM through a group of proteins known as the TonB system (Braun, 1995;Crosa et al., 2004;Postle, 2007).
The TonB system has been extensively studied in Escherichia coli, where it has been shown to contain three integral IM proteins: TonB, ExbB, and ExbD (Bassford, Bradbeer, Kadner, & Schnaitman, 1976;Bradbeer & Woodrow, 1976;Braun et al., 1996). Besides providing stabilization, both ExbB and ExbD have been shown to transduce the PMF energy to TonB, thus causing a conformational change in TonB. This new conformation allows TonB to span the periplasmic space and make contact with TonB-dependent OM receptors allowing the uptake of iron-bound siderophores (Ahmer, Thomas, Larsen, & Postle, 1995;Germon, Ray, Vianney, & Lazzaroni, 2001;Higgs, Myers, & Postle, 1998;Kuehl & Crosa, 2009;Larsen et al., 2007;Larsen, Wood, & Postle, 1993;Ollis, Manning, Held, & Postle, 2009;Swayne & Postle, 2011). Every member of the Vibrionaceae family contains two TonB systems: the TonB1 and TonB2 systems. The TonB1 system is very similar to E. coli in that it contains three proteins: ExbB, ExbD, and TonB. The TonB2 system has previously been shown to contain a fourth protein, TtpC2 (Kuehl & Crosa, 2009, 2010Kustusch, Kuehl, & Crosa, 2011Stork, Otto, & Crosa, 2007). When grown in iron-limiting conditions, both the TonB1 and TonB2 systems have been shown to be up-regulated, while iron-rich conditions down-regulate expression (Alice, Naka, & Crosa, 2008). In addition, some Vibrio species have been shown to contain a third TonB system. The TonB3 system looks very similar to the TonB2 system in that, along with the all of the proteins found in the TonB1 system, it also has previously been shown to contain a fourth protein, TtpC3 (Kuehl & Crosa, 2009, 2010Kustusch, Kuehl, & Crosa, 2011. Unlike the TonB1 and TonB2 systems, the TonB3 system does not appear to be regulated by iron-limiting conditions, but instead expression of this system has only been shown when the organism is grown in human serum (Alice et al., 2008).
We describe here the previously uncharacterized genes, ttpB2 and ttpD2, in the TonB2 system of V. vulnificus. The gene ttpB2 is located immediately upstream of ttpC2 and is the first gene in the tonB2 operon. In contrast, ttpD2 is the last gene in the tonB2 operon and is found downstream of tonB2. Homologs of both genes are found as part of tonB2-like systems in other Vibrio species examined.
Similar uncharacterized genes, ttpB3 and ttpD3, have also been observed in the operon of the TonB3 system of V. vulnificus. Previous reports have indicated that both ttpB2 and ttpD2 are up-regulated, along with the rest of the genes within the TonB2 system, in V. vulnificus during iron-limiting conditions (Alice et al., 2008). Together, this suggests that both genes play vital roles in the TonB2-mediated iron transport of V. vulnificus.
In this study, we demonstrate that both TtpB2 and TtpD2 are essential in the TonB2 system of V. vulnificus. We demonstrate that both proteins are necessary for growth in iron-deplete conditions. In addition, we show that both TtpB2 and TtpD2 are essential for the uptake of both V. vulnificus-produced endogenous siderophores and exogenous siderophores created by other microorganisms. We have determined that the TtpB3 and TtpD3 proteins from the TonB3 system cannot complement respective mutations in the TonB2 system.
Finally, we show that the TtpD2 proteins from other Vibrio species can complement a ΔttpD2 mutation and function as a part of the
Iron-rich conditions were obtained by the addition of ferric ammonium citrate (FAC) at indicated concentrations.

| Sequence identity
The percent identity matrices and protein alignments were generated through Clustal Omega (Sievers et al., 2011

| Construction and complementation of V. vulnificus mutants
In-frame deletions of the entire coding sequences of genes were generated using splicing by overlap extension (SOE) PCR (Higuchi, Krummel, & Saiki, 1988;Senanayake & Brian, 1995). Upstream and downstream regions (approximately 700 bp) flanking each gene were amplified with specific primers (Table A2, Primers 1/3 and 2/4). The two 700-bp fragments with overlapping primer ends were spliced using SOE PCR with the two outside primers (Primers 1/2). The amplified fragment was cloned into the pCR2.1 vector (Invitrogen), digested with appropriate restriction enzymes, and subcloned into the suicide vector pDM4, which had been previously digested with the same restriction enzymes. E. coli S17-1 λpir transformed with the pDM4 (Milton, O'Toole, Horstedt, & Wolf-Watz, 1996) derivatives was conjugated with V. vulnificus, and exconjugants were selected on TCBS agar with 2 μg/ml Cm. For the excision of the suicide vector, clones were incubated in the absence of Cm and plated on TSAS plates (TSBS with 1.5% agar) containing 15% sucrose. Colonies that grew on these plates were screened for Cm sensitivity, and deletions within the genes of interest were confirmed by PCR.
Complementation of deleted genes was achieved by amplifying each gene of interest by PCR with primers containing restriction sites (Table A2). The fragments were cloned into pCR2.1 vector (Invitrogen), sequenced, and then subcloned into pMMB208 (Morales, Bäckman, & Bagdasarian, 1991) under the control of the Ptac promoter. This promoter is under the control of the lacI gene harbored in the vector. The constructs were transformed into V. vulnificus strains by triparental conjugation with the plasmid helper pRK2013 (Figurski & Helinski, 1979). In order to induce transcription of the cloned genes, 1 mM isopropyl-d-thiogalactopyranoside (IPTG) was added to the solid and/or broth medium.

| Growth assays
Overnight cultures were diluted to an optical density at 600 nm (OD 600 ) of 0.02 into 25 ml of either TSBS or CM9 media and incubated at 37°C while shaking at 160 rpm. Readings were taken every 30 min for 300 min. For iron-limiting conditions, 75 µM 2,2′-dipyridyl was added to CM9 media. Complementation growth assays also contained 10 µg/ml chloramphenicol and 1 mM IPTG.

| CAS assays
Overnight cultures were diluted to an optical density at 600 nm (OD 600 ) of 0.02 into 25 ml of CM9 media and incubated at 37°C to an OD 600 of 0.5 (∼3 hr). Cells were normalized to an OD 600 of 0.5 in 1 ml of CM9 media. Cells were then centrifuged, and 0.5 ml of supernatant was used in the chrome azurol S (CAS) assay. A ratio of 1:1 of cell supernatant to CAS solution (600 μM hexadecyltrimethylammonium bromide [HDTMA], 150 μM FeCl 3 , 150 μM CAS) was mixed and placed into 0.2-cm cuvettes, and the OD 630 was read using a spectrophotometer at 20 min after mixing. The assay was normalized to the wild type (WT), and the final values were inverted to show a positive change for ease of reading.
Complemented mutants also had appropriate antibiotics added as well as 1 mM IPTG. In these experiments, bacterial cells were included in the agar and, upon solidification, plates were spot inoculated with different iron sources. After incubation of the plates at 37°C for 24 hr, halos of bacterial growth surrounding the locations of the spots indicated positive results. In these experiments, ferric ammonium citrate (FAC), which does not require active transport to obtain, was included to confirm that the strain inoculated in the agar was viable. The purified compounds spotted on top of the bioassay plates to determine the functionality of the different TonB cluster of genes in V. vulnificus were as follows: 1 mg/ml iron-free vibriobactin (EMC Microcollections), 1 mg/ml iron-free aerobactin, 1 mg/ ml iron-free enterobactin, 1 mg/ml iron-free ferrioxamine (all three from EMC Microcollections), and 500 μg/ml FAC (Sigma). Each bioassay plate was spotted with 2 μl of each indicated iron source. To test vibrioferrin, WT V. parahaemolyticus was streaked onto the plate.
Vulnibactin and the hydroxamate siderophore, both from V. vulnificus, were tested by streaking either WT or the ΔvenB strain onto the bioassay plate. Halo growth around these streaks indicates that the strain within the plate can use the siderophore being produced by the streaked strain.

| RNA extractions
Wild-type V. vulnificus CMCP6 was grown overnight in TSBS and diluted 1/400 in CM9 with the addition of 250 µg/ml FAC for iron-rich conditions or with the addition of 50 µM 2′2-dipyridyl for iron-limited conditions. Cultures were grown to an optical density at 600 nm (OD 600 ) of 0.3. Samples were centrifuged, and RNA was harvested using Qiagen's RNeasy kit with Qiagen's RNAprotect following manufacturer's instructions. TurboDNase (Thermofisher) was used, and manufacturer's protocols were followed to ensure complete degradation of genomic DNA. At least three biological replicates were created for each condition.

| Semi-quantitative RT-PCR
Purified RNA samples underwent reverse transcription to synthesize cDNA using the Qiagen QuantiTect Reverse Transcription kit following manufacturer's instructions. Control reactions without reverse transcriptase were also performed for each sample. Reverse transcriptase PCRs (RT-PCRs) were then performed using 2 μl of each reverse transcription (RT) reaction mix using gene-specific primers described in Table A2. A control without reverse transcriptase enzyme in the RT reaction mix was used for each PCR. Fragments were resolved by electrophoresis on agarose gels. At least three replicates from each biological replicate were created. Pictures of gels were taken under UV light with a Kodak Gel Logic 2200 imaging system at an exposure time of 0.1 s. Images were then analyzed using ImageJ software. Numerical values for each gene could then be normalized to our control gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for each band.

| vulnificus contains two novel proteins associated with the TonB2 and TonB3 systems
The bacterium V. vulnificus contains three TonB systems spread across its two chromosomes ( Figure 1a; Kustusch et al., 2011;Tagomori, Lida, & Honda, 2002). The TonB1 system, present in all members of the Vibrionaceae, consists of three genes: tonB1, exbB1, and exbD1. This system is always found associated with the genes needed for heme uptake-hutB, hutC, and hutD ( Figure 1a; Henderson & Payne, 1994;Kuehl & Crosa, 2010). The TonB1 system found in Vibrio is very similar to the well-studied TonB system of E. coli.
All bacteria in the Vibrio genus examined to date also contain a second TonB system, the TonB2 system. This second system has shown uptake specificity to certain iron-bound substrates in V. vulnificus, as well as V. cholerae and V. anguillarum (Kuehl & Crosa, 2010;Kustusch et al., 2012;Stork et al., 2007). Three proteins similar to those found in the TonB1 system-TonB2, ExbB2, and ExbD2-are found in this system. In addition, a fourth protein, TtpC2, has been shown to be essential for the function of the TonB2 system in multiple Vibrio species (Kuehl & Crosa, 2009;Kustusch et al., 2012;Stork et al., 2007). It has been alluded to, but never shown, that two additional genes may be a part of the TonB2 system (Kustusch et al., 2012). DNA sequence analysis of the tonB2 operon, located on V. vulnificus' second, larger chromosome, suggested that two open reading frames (ORFs) are present as the first and last ORFs in the tonB2 operon. Each appears to encode for hypothetical proteins of unknown function. We have designated the first ORF in this operon as ttpB2, and the last ORF as ttpD2 for, TonB2 complex-associated transport protein B and D ( Figure 1a).
The tonB2 operon therefore appears to contain six genes, beginning with ttpB2 ( Figure 1a). A detailed analysis of the nucleotide region immediately upstream of ttpB2 is shown in Figure 1b. A potential ferric uptake regulator (Fur) binding site (shown in bold) and promoter region (underlined) are clearly identified before the translational start site of TtpB2. The fur sequence identified is 78.95% similar to the fur consensus sequence 5′-gataatgataatcattatc-3′ (Lavrrar & McIntosh, 2003). Fur is a DNA-binding protein that recognizes and binds Fur binding sites. Together with Fe 2+ , as a corepressor, this protein complex can block transcription of F I G U R E 1 The three TonB systems of V. vulnificus and the promoter region of the TonB2 system. (a) The TonB1, TonB2, and TonB3 systems of V. vulnificus are depicted with the novel genes, ttpB and ttpD, noted in dark gray. (b) Nucleotide sequence analysis in the intergenic region immediately upstream of the TonB2 operon shows the putative Fur box sequence (bold) and the predicted −10 and −35 regions (underlined). The translational start site of ttpB2 is indicated by the ATG codon and the methionine (M) designation hutB hutC hutD TonB1 system TonB2 system

TonB3 system
TTCCGTTACAATGCGCCGCATTGATAATGCGAATAAATATCATTTATTGTTTTGGTAGTCACTTTAGGTAACAAAAACGTCT ATG -35 -10 M target genes (Troxell & Hassan, 2013). These results suggest that ttpB2 is the first gene in the tonB2 operon and suggests that the lack of iron would induce transcription of these genes.
To determine whether the TonB2 system, specifically the ttpB2 and ttpD2 genes, is transcribed at elevated levels during iron-limiting conditions, semi-quantitative reverse transcriptase (RT) PCR was performed. We saw a ~2.5-fold and ~8-fold elevated expression of ttpB2 and ttpD2, respectively, in iron-depleted conditions compared to iron-rich conditions ( Figure 2). We used tonB2 expression (~3-fold increase) as a positive control. The housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used for normalization. These results are very similar to what has been seen in another study (ttpB2 = ~5-fold, ttpD2 = ~8-fold, and tonB2 = ~4-fold increase) using microarray analysis (Alice et al., 2008). Together, these results clearly show that both ttpB2 and ttpD2 are induced by iron-depleted conditions and are a part of the TonB2 system.
The ttpB2 gene encodes for a hypothetical protein with a large domain of unknown function (DUF3450). This family of proteins, DUF3450, appears in both eukaryotes and bacteria and always contains about 260 amino acids (Altschul, Gish, Miller, Myers, & Lipman, 1990). This is very similar to the length of ttpB2, which encodes for 255 amino acids in V. vulnificus. The gene, ttpB2, also appears to be the first ORF in all members of the Vibrionaceae's TonB2 systems examined. A protein alignment ( Figure A1) as well as a percent identity matrix (  (Table 1). This similarity is comparable to the rest of the known proteins in the TonB2 system (TonB2 ~ 45%, ExbB2 ~ 63%, ExbD ~ 86%, TtpC2 ~ 59%) when one compares V. vulnificus to these same four species (Kustusch et al., 2012). Finally, a leucine zipper motif can be found near the N-terminus in all of these homologs ( Figure A1; Altschul et al., 1990). Leucine zippers are used for protein-protein interactions and are located in alpha helices where the pattern Leu-X6-Leu-X6-Leu-X6-Leu is found (Hirst, Vieth, Skolnick, & Brooks, 1996).
The ttpD2 gene encodes for a hypothetical 425 amino acid protein.
It contains a number of potential tetracopeptide repeat (TPR) motifs throughout the protein (Altschul et al., 1990). TPR motifs are structural motifs present in a wide range of proteins. They generally mediate protein-protein interactions and help assemble larger multiprotein complexes (Schultz, Milpetz, Bork, & Ponting, 1998). As with ttpB2, ttpD2 also appears in all members of the Vibrionaceae examined. A protein alignment ( Figure A2) and a percent identity matrix (  (Table 2). Unlike TtpB2, where large stretches of similarity were observed ( Figure A1), TtpD2 had only very short stretches of homology between the different TtpD2 proteins found in the other four Vibrio species ( Figure A2). A third TonB3 system has been identified in V. vulnificus as well as in a number of Vibrio species and species within the genera Alivibrio, Photobacterium, and Teredinibacter (Kustusch et al., 2011). This system appears to contain the six proteins found in the TonB2 system ( Figure 1a). In contrast to the TonB2 system, a gene encoding an outer membrane (OM) receptor protein can be found at the beginning of the tonB3 operon. TtpB2 and TtpD2 were compared to their counterparts in the TonB3 system. Only a 30.3% similarity was found between the TtpB2 and TtpB3 proteins and a 23% similarity between TtpD2 and TtpD3 (Tables 1 and 2). This is very similar to homologies in other known TonB2 system proteins when comparing V. vulnificus' TonB2 system against its TonB3 system: TonB 27%, ExbB 40%, ExbD 57%, and TtpC 37% (Kustusch et al., 2012).

| Both TtpB2 and TtpD2 are necessary components of the TonB2 system in mediating growth and the transport of endogenous siderophores
Growth assays were used to determine the necessity and essentiality of TtpB2 and TtpD2 in the TonB2 system. An assortment of F I G U R E 2 ttpB2 and ttpD2 are transcriptionally activated in iron-deplete conditions along with tonB2. Relative fold changes in expression of ttpB2, ttpD2, and tonB2 from the TonB2 operon are shown. Semi-quantitative reverse transcriptase PCR (RT-PCR) was performed in triplicate on at least three biological replicates. Gel images were analyzed by ImageJ software to determine total pixels in each band. Fold changes in expression are the average values of iron-limiting conditions compared to iron-rich conditions normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Significance (*p < .005) was determined through Student's t test mutants was grown under iron-rich and iron-deplete conditions.
When grown in iron-rich media, TSBS, or minimal media, CM9, no differences in growth could be seen (data not shown). In contrast, when grown in CM9 plus the iron chelator dipyridyl (iron-deplete conditions), a considerable difference in growth was noted in strains missing components of both the TonB1 and TonB2 systems ( Figure 3). Single mutants in any of the TonB systems grew as well as wild-type. Double mutants, ΔtonB1 ΔtonB2, ΔtonB1 ΔttpB2, and ΔtonB1 ΔttpD2, showed a considerable growth deficit, as did the triple mutants ΔtonB1 ΔtonB2 ΔtonB3, ΔtonB1 ΔttpB2 ΔtonB3, and ΔtonB1 ΔttpD2 ΔtonB3. In all of these strains, both the TonB1 and TonB2 systems were missing components of their respective system. The double and triple mutant strains deleted for ttpB2 or ttpD2 showed the same lack of growth as strains deleted for tonB2.
In addition, any double mutant strain that left either the TonB1 or the TonB2 system functional (ΔtonB1 ΔtonB3, ΔtonB2 ΔtonB3, ΔttpB2 ΔtonB3, and ΔttpD2 ΔtonB3) retained the ability to grow as well as the wild-type strain and single mutants. This shows the importance and redundancy of the TonB1 and TonB2 systems when grown in these conditions and the inability of the TonB3 system to be used to overcome the loss of either of these systems. These data demonstrate the essentiality of TtpB2 and TtpD2 in the TonB2 system for growth in iron-depleted conditions.
Bioassays were used to investigate the essentiality of TtpB2 and TtpD2 within the TonB2 system for the uptake of endogenous siderophores. V. vulnificus can produce two siderophores, vulnibactin, and the uncharacterized hydroxamate siderophore (Simpson & Oliver, 1983). An assortment of mutants was tested with various combinations of deletions in different TonB systems (Table 3). We have previously tested the role of TtpC2 in the TonB2 system, but not TtpB2 nor TtpD2 (Kustusch et al., 2012).
It was previously shown that both the TonB1 and TonB2 systems, but not the TonB3 system, could transport both of the endogenous siderophores of V. vulnificus (Kustusch et al., 2012). As shown in and TtpD3 were still available in these strains. Both of these siderophores were also unable to be used when all three TonB systems were deleted using a ΔtonB1 ΔtonB2 ΔtonB3 strain, ΔtonB1 ΔttpB2 ΔtonB3 strain, or a ΔtonB1 ΔttpD2 ΔtonB3 strain (Table 3). To determine that the deletions did not have any polar effects and that the mutations were truly in the targeted genes, assorted complemented strains were also utilized in this assay. Endogenous siderophores could once again be brought into the cell when either the TonB1 system was complemented (w/ptonB1), or the TonB2 system was complemented (w/pttpB2 in a ttpB2 deletion or w/pttpD2 in a ttpD2 deletion), further confirming the importance of these proteins to the uptake of endogenous siderophores and the ability to use either (Table 3).
We further analyzed these mutants through the use of chrome azurol S (CAS) assays. Briefly, these colorimetric assays can be used to detect secreted iron-bound complexes, like siderophores. When in iron-depleted conditions, siderophores are secreted from V. vulnificus into the surrounding media. If the strain has a functional TonB system that can mediate the uptake of these now iron-bound siderophores, they will be brought back into the cell. If the TonB systems are nonfunctional and the iron-bound siderophores cannot be brought back into the cell, an accumulation builds up in the medium.
By harvesting the supernatant of different TonB mutants and mixing it with CAS reagents, one can determine which strain is capable of siderophore uptake through a color change.
The results from these experiments can be seen in Figure 4.

| Both TtpB2 and TtpD2 are essential for the TonB2 system in mediating the transport of exogenous siderophores
Vibrio vulnificus can thrive in a variety of environments and hosts, because of the evolved ability to use a diverse assortment of iron sources. These include the use of exogenous siderophores, that is, siderophores that are produced by other bacterial and fungal species (Alice et al., 2008;Aso, Miyoshi, Nakao, Okamoto, & Yamamoto, 2002). In order to evaluate any potential roles for TtpB2 and TtpD2 in the uptake of these exogenous siderophores by V. vulnificus, we performed bioassays with an assortment of F I G U R E 3 TtpB2 and TtpD2 are necessary components of the TonB2 system for growth in iron-limited conditions. Overnight cultures were diluted to an optical density at 600 nm (600 nm ) of 0.02 into CM9 media plus 75 µM 2,2′-dipyridyl (iron-deplete media). Cultures were grown at 37°C and monitored for growth every 30 min. The average growth at 150 min is depicted for each strain. Strains were grown in triplicate, and significance (*p = <.005) was determined through Student's t test mutants and complemented strains. Quadruple mutants, where vulnibactin production is halted due to the deletion of venB, were also unable to utilize exogenous siderophores listed above. Complementation of these quadruple mutants with either tonB1, ttpB2, or ttpD2 restored the uptake of all of the above exogenous siderophores (Table 4). These quadruple strains were important to assay because they prove that vulnibactin is unable to steal the iron that is bound by the exogenous siderophore and that the exogenous siderophores are indeed the source of iron in this study. These results again demonstrate the essentiality of TtpB2 and TtpD2 in the TonB2 system.
In contrast, it has been shown that aerobactin, a commonly produced siderophore by enterobacteria, can only be brought in through the TonB2 system (Kustusch et al., 2012;de Lorenzo & Martinez, 1988). Consistent with these findings, double and triple mutants behaved as would be anticipated in a similar fashion to single mutants in the TonB2 system preventing aerobactin from being utilized (Table 4).

| Complementation of a V. vulnificus ΔttpB2 must come from V. vulnificus, while a ΔttpD2 mutant can be complemented by different members of the Vibrionaceae
It has previously been demonstrated that the TtpC2 protein from different Vibrio species can be substituted for a ΔttpC2 mutation in V. vulnificus to restore function of the TonB2 system (Kustusch et al., 2012). To better examine the importance and conservation of function of both TtpB2 and TtpD2, proteins from other Vibrio species were substituted into a V. vulnificus triple mutant (ΔtonB1 ΔttpB2 ΔtonB3 or ΔtonB1 ΔttpD2 ΔtonB3). Complementing plasmids were moved into appropriate strains, and growth assays were performed.
Consistent with the past findings related to ΔttpC2, the ΔtonB1 ΔttpD2 ΔtonB3 mutant growth was complemented using the majority of other Vibrio species' homologous genes ( Figure 5a). All TtpD2 proteins from the six Vibrio species tested were able to restore growth. Growth in these conditions was not restored to the strains with nonfunctional TonB2 systems, when TtpD3 from the TonB3 system was used for complementation, demonstrating a distinction in roles for TtpD2 and TtpD3.
In contrast to these TtpD2 findings and previous findings with TtpC2, growth was only restored in a TtpB2 mutant when a complementing TA B L E 3 TtpB2 and TtpD2 are essential components of the  The triple mutant strain ΔtonB1 ΔtonB2 ΔtonB3 contained an additional mutation in the lacZ gene (Strain AA-12 in Table A1). This mutation was necessary for previous studies and has been shown to not affect Fe-siderophore transport (Alice et al., 2008). plasmid restoring TonB1 or a TtpB2 protein from a V. vulnificus species (wild-type or a Biotype 2 strain) was utilized. Results using the ΔtonB1 ΔttpB2 ΔtonB3 strains can be seen in Figure 5b. Provision of any other TtpB2 protein from different Vibrio species or TonB3 was not able to restore growth. These results demonstrate that the TonB2 system has a specificity for its own TtpB2 protein but not TtpD2.
In addition to determining if growth could be restored through complementation with homologous copies of TtpB2 and TtpD2, bioassays were used to examine restoration of the ability to utilize siderophores. As would be expected, bioassays utilizing homologous TtpB2 from other Vibrio strains show parallel results as growth assays-homologous strains did not restore siderophore utilization (Table 5). Confirming the ability of the TonB2 system to utilize homologous TtpD2 from other Vibrio species, the TptD2 complemented strains saw restoration of their ability to uptake each siderophore tested (Table 6). These data, along with the data from the growth assays, clearly demonstrate the specific need of a native form of TtpB2, but not TtpD2, for growth and siderophore uptake.

| D ISCUSS I ON
Vibrio vulnificus is a gram-negative opportunistic pathogen that can infect humans as well as eels (Gulig et al., 2005;Hor, Chang, Chang, Lei, & Ou, 2000). In humans, fatal septicemia is often associated with large amounts of iron found in a person's blood due to several chronic or acute conditions (Wright & Morris, 1991). This high concentration of iron is a major risk factor for many bacterial infections, including by V. vulnificus (Crosa et al., 2004). Siderophores are small molecular chelators of iron that many microbes secrete in order to sequester iron from their environment (Neilands, 1995). V. vulnificus produces two such siderophores, the dominate siderophore, vulnibactin, and a hydroxamate-type compound (Okujo et al., 1994;Simpson & Oliver, 1983). Once bound to iron, these siderophores first interact with an OM receptor before being internalized. Energy, produced through the PMF in the IM, must be transduced to the OM to allow iron-bound siderophores into the periplasm, and consequently into the cytosol of the cell. The TonB systems are responsible for harnessing this energy and transferring it to the OM receptor (Braun, 1995;Crosa et al., 2004;. Unlike the well-studied, single TonB system of E. coli, V. vulnificus contains three TonB systems (Kustusch et al., 2011).
Previous in silico work identified the potential genes in V. vulnificus' three TonB systems. While the TonB1 system has the same three genes (tonB1, exbB1, and exbD1) as E. coli, the TonB2 and TonB3 systems of V. vulnificus appeared to have an additional gene, ttpC, as well as two ORFs found at the beginning and end of each operon (Kustusch et al., 2012). In this study, we have demonstrated that these two ORFs, now renamed ttpB2 and ttpD2, are essential and necessary for the TonB2 system. We have identified a potential Fur binding site and promoter region upstream of ttpB2 consistent with previous studies that have F I G U R E 4 TtpB2 and TtpD2 within the TonB2 system are essential for vulnibactin uptake. Liquid chrome azurol S (CAS) assays were used to determine the relative production of vulnibactin, one of V. vulnificus' native siderophores. Supernatant from log phase cultures grown in CM9 minimal media was mixed in a 1:1 ratio with CAS solution. OD 630 was determined after 20 min of incubation at room temperature and normalized to WT. Strains were grown in triplicate, and significance (*p = <.005) was determined through Student's t test alluded to the presence of a Fur binding site through gel shift analysis using a His-tagged Fur protein (Alice et al., 2008). In addition, our semi-quantitative RT-PCR analysis has shown the same, if not greater, up-regulation of ttpB2 and ttpD2 in iron-deplete versus iron-rich conditions as compared to tonB2. This up-regulation has also been demonstrated through microarray analysis yielding similar fold expression as well as promoter fusion assays (Alice et al., 2008).
While present in the TonB3 system, we have shown that ttpB3 and ttpD3 cannot substitute for their apparent homologs in the TonB2 system. While the orientation and arrangement of the tonB2 and tonB3 operons appear similar, they in fact are very different from each other.  (Alice et al., 2008;Kuehl & Crosa, 2009;Kustusch et al., 2012). It appears that the TonB3 system is only transcriptionally activated when grown in human serum or grown in minimal media with glycerol as a sole carbon source with low concentrations of casamino acids (Alice & Crosa, 2012;Alice et al., 2008).
In this study, the essentiality of both TtpB2 and TtpD2 for iron transport was evaluated using growth assays, bioassays, and CAS TA B L E 4 TtpB2 and TtpD2 are essential components of the Growth was determined by the presence of halos (+) or lack thereof (−) around each iron source indicated after 18 hr at 37°C. Two microliters of each iron source was spotted on the surface of the plates in the following concentrations: FAC, 500 µg/ml; vibriobactin, 1.0 mg/ml; ferrioxamine, 1.0 mg/ml; enterobactin, 1.0 mg/ml; and aerobactin, 1.0 mg/ml. Wild-type V. parahaemolyticus was streaked on the plate as a source of vibrioferrin, and the halo of growth was monitored around the streak. b The triple mutant strain ΔtonB1 ΔtonB2 ΔtonB3 contained an additional mutation in the lacZ gene (Strain AA-12 in Table A1). This mutation was necessary for previous studies and has been shown to not affect Fe-siderophore transport (Alice et al., 2008).
F I G U R E 5 TtpD2, and not TtpB2, can be complemented by other Vibrio species to restore function of the TonB2 system. Triple mutants, ΔtonB1 ΔttpD2 ΔtonB3 (a) or ΔtonB1 ΔttpB2 ΔtonB3 (b), were used with various complementing plasmids. Overnight cultures were diluted to an optical density at 600 nm (600 nm ) of 0.02 into CM9 media plus 75 µM 2,2′-dipyridyl (iron-deplete media), 1 mM IPTG, and 10 µg/ml chloramphenicol. Cultures were grown at 37°C and monitored for growth every 30 min. The average growth at 150 min is depicted for each strain. Strains were grown in triplicate, and significance (*p = <.005) was determined through Student's t test The embedded strain, VSRK615 (ΔtonB1 ΔttpB2 ΔtonB3 ΔvenB), contained the complementing plasmid pMMB208 expressing TtpB2 from the various Vibrio species listed at the top of the table. b Growth was determined by the presence of halos (+) or lack thereof (−) around each iron source indicated after 18 hr at 37°C. Two microliters of each iron source was spotted on the surface of the plates in the following concentrations: FAC, 500 µg/ml; vibriobactin, 1.0 mg/ml; ferrioxamine, 1.0 mg/ml; enterobactin, 1.0 mg/ml; and aerobactin, 1.0 mg/ml. Wild-type V. vulnificus and V. parahaemolyticus were streaked on the plate as a source of vulnibactin and vibrioferrin, respectively, and the halo of growth was monitored around the streak. A V. vulnificus ΔvenB mutant was streaked onto the plate to test for growth around the hydroxamate siderophore.
TA B L E 5 Complementation of ttpB2 from other non-Vibrio vulnificus species does not restore the TonB2 systemmediated uptake of various iron sources a assays. Our results clearly demonstrate that both proteins are necessary components of the TonB2 system and are required for that system to function. This study has also demonstrated that either the TonB1 or the TonB2 system, with all six proteins present, must be functional for the iron uptake necessary for robust growth in iron-limited conditions. In contrast, these systems do not need to be present for growth in iron-rich conditions where freely available iron is readily available. In addition, our CAS assays and bioassays have shown that TtpB2 and TtpB2 are essential for the uptake of both endogenous and exogenous siderophores through the TonB2 system. While most exogenous siderophores could be brought in through the TonB1 or TonB2 system, aerobactin is only brought in by the TonB2 system. Finally, this study has shown that TtpB3 and TtpD3 cannot complement the TonB2 system deletions for growth, in both their natively expressed levels or when artificially overexpressed, in the iron-limited conditions tested in this study or for the uptake of siderophores.
TtpB2 and TtpD2 are predicted to be 255 and 425 amino acids, respectively. TtpB2 contains a leucine zipper motif (Leu-X6-Leu-X6-Leu-X6-Leu) toward the N-terminus, while TtpD2 contains a number of TPR motifs. Both of these conserved motifs suggest that TtpB2 and TtpD2 function as part of multiprotein complexes or through protein-protein interactions (Hirst et al., 1996;Schultz et al., 1998).
Current work is underway to examine these motifs and the way in which these proteins interact within the TonB2 system.
Our RT-PCR analysis shows a ~2.5-fold increase in the expression of ttpB2 and an 8-fold increase in the expression of ttpD2 confirming an up-regulation of these genes alongside tonB2 in iron-depleted conditions relative to iron-rich.
Previously, both TonB2 and TtpC2 have been shown to be interchangeable among Vibrio species to create a functional TonB2 system (Kustusch et al., 2012;Stork et al., 2007). Similarly to other genes in Previous studies have shown the importance of V. vulnificus' TonB systems and its contribution to growth and virulence (Alice et al., 2008;Kustusch et al., 2012). Here, we demonstrate that two additional proteins, TtpB2 and TtpD2, are essential elements for a functional TonB2 system. In conclusion, TtpB2 and TtpD2 are required for V. vulnificus growth and the uptake of endogenous and exogenous siderophores in iron-limited conditions similar to those the organism experiences within a host. Both the specificity of TtpB2 and the ubiquitous nature of TtpD2 will be further analyzed in future studies.

ACK N OWLED G M ENTS
We would like to thank the University of Portland, which provided research space and funds to R.J.K., and summer stipend money to all authors. This work was supported by grants 2014204 and NS-2017310 from The M.J. Murdock Charitable Trust to R.J.K.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to disclose.  investigated the study; designed the methodology; provided the software; supervised, validated, and visualized the study; wrote the review; and edited the manuscript.

E TH I C A L A PPROVA L
None required.