TonB Energy Transduction Systems of Riemerella anatipestifer Are Required for Iron and Hemin Utilization

Riemerella anatipestifer (R. anatipestifer) is one of the most important pathogens in ducks. The bacteria causes acute or chronic septicemia characterized by fibrinous pericarditis and meningitis. The R. anatipestifer genome encodes multiple iron/hemin-uptake systems that facilitate adaptation to iron-limited host environments. These systems include several TonB-dependent transporters and three TonB proteins responsible for energy transduction. These three tonB genes are present in all the R. anatipestifer genomes sequenced so far. Two of these genes are contained within the exbB-exbD-tonB1 and exbB-exbD-exbD-tonB2 operons. The third, tonB3, forms a monocistronic transcription unit. The inability to recover derivatives deleted for this gene suggests its product is essential for R. anatipestifer growth. Here, we show that deletion of tonB1 had no effect on hemin uptake of R. anatipestifer, though disruption of tonB2 strongly decreases hemin uptake, and disruption of both tonB1 and tonB2 abolishes the transport of exogenously added hemin. The ability of R. anatipestifer to grow on iron-depleted medium is decreased by tonB2 but not tonB1 disruption. When expressed in an E. coli model strain, the TonB1 complex, TonB2 complex, and TonB3 protein from R. anatipestifer cannot energize heterologous hemin transporters. Further, only the TonB1 complex can energize a R. anatipestifer hemin transporter when co-expressed in an E. coli model strain.


Introduction
R. anatipestifer, belonging to the family Flavobacteriaceae and genus Riemerella, is a gramnegative bacterium. It is one of the most harmful pathogens to the duck, chicken, geese and turkey industries around the world, as it can quickly become endemic and is difficult to eradicate [1]. So far, at least 21 serotypes have been identified in different countries with no

Materials and Methods Bacterial Strains and Plasmids Used in This Study
The bacterial strains and plasmids used in this study are listed in Table 1.

E. coli Hemin-dependent Growth Assays and Iron-chelated Growth Assays
The tested E. coli strains, C600ΔhemA tonB::Tn10 pAM238::OMhemR RA , harboring plasmid pBAD24 and pBAD24 derivatives containing R. anatipestifer tonB genes were grown for 5 h in LB medium containing δ-ala, Spc and Amp. Bacteria were collected, re-suspended in 1 ml Phosphate Buffered Saline (PBS) and centrifuged for 5 min at 5,000 g. This procedure was repeated three times to wash the bacteria. The OD values at 600 nm were measured for bacterial suspensions, which were then adjusted to 10,000 bacteria/ml (1 OD 600 = 5×10 8 bacteria). Then, 20-μl samples (approximately 200 bacteria) of the bacterial suspension were used to inoculate LB plates, LB plates containing 50 μg/ml δ-ala, and LB plates containing 20 μM hemin. Growth was recorded after 1 day incubation at 37°C. All experiments were performed in triplicate.
E. coli strains, C600ΔhemA tonB::Tn10, harboring a pBAD24 derivative containing an R. anatipestifer tonB gene, were inoculated in LB medium containing 50 μg/ml δ-ala and Amp and were grown for approximately 5 h. Then, bacteria were harvested and numerated as mentioned above. Approximately 200 bacteria were inoculated onto LB plates or iron-depleted plates which contained Dip at 150 μM. The growth was recorded after 1 day incubation at 37°C. All experiments were performed in triplicate.
The Effect of R. anatipestifer tonB Knockouts on Hemin and Iron Uptake R. anatipestifer ATCC11845 and the tonB1, tonB2, and tonB1tonB2 mutant derivatives were inoculated onto LB plates containing 5% defibrinated sheep blood and incubated overnight at 37°C under 5% CO 2 atmosphere. Bacteria were collected, re-suspended in 1 ml PBS and centrifuged for 5 min at 5,000 g. This operation was repeated three times to wash the bacteria. The OD 600 values of the bacterial suspensions were measured and adjusted to 10,000 bacteria/ml (1 OD 600 = 6×10 8 bacteria). Then, 20 μl (approximately 200 bacteria) were inoculated onto LB plates, LB plates containing 20 μM hemin, LB plates containing 5% bovine serum or LB plates containing 5% bovine serum and 40 μM Dip. The growth was recorded after a 2-day-long incubation at 37°C under 5% CO 2 atmosphere. All experiments were performed in triplicate. Genetic Techniques E. coli cells were transformed by the calcium chloride method as described by Maniatis [36].

DNA Manipulation
Small-scale plasmid DNA preparations were performed using a QIAprep Spin Miniprep kit. Restriction, modification, and ligation were carried out according to the manufacturer's recommendations. DNA fragments were amplified in a Hybaid PCR thermocycler using Phusion DNA polymerase (NEB, Beijing, China). Purification of DNA fragments from the PCR and restriction-digest reactions was performed using the TianGEN Extract II kit (TIANGEN, Beijing, China). The validity of all the fragments amplified by PCR was determined by sequencing (BGI, Guangzhou, China).

Construction of R. anatipestifer tonB Deletion Mutants
The tonB genes were deleted by allelic exchange using a recombinant suicide vector, which replaced tonB genes with a 1140-bp SpcR cassette according to the methods described by Hu et al [37] or a 994-bp ErmR cassette from R. anatipestifer CH-1. Briefly, the approximately 800-bp left flanking sequence and 800-bp right flanking sequence of the R. anatipestifer ATCC11845 tonB1 and tonB2 genes were amplified by PCR using primers TonB1upP1 and TonB1upP2, TonB1downP1 and TonB1downP2, TonB2upP1 and TonB2upP2, TonB2downP1 and TonB2-downP2, respectively (S1 Table). The 1140-bp sequence containing the SpcR cassette was amplified from plasmid pAM238 using primers SpcRP1 and SpcRP2. The 994-bp sequence containing the ErmR cassette was amplified from the genome of R. anatipestifer CH-1 using primers ErmRP1 and ErmRP2. The PCR fragments (TonB1 upstream, TonB1 downstream, and ErmR cassette or TonB2 upstream, TonB2 downstream, and SpcR cassette) were ligated using the overlap PCR method [38]. The fused PCR fragments were purified and digested with KpnI and BamHI. Then, fragments were ligated into a suicide pEX18Gm plasmid digested with KpnI and BamHI. Ligation mixtures were transformed into CaCl 2 -competent E. coli strain XL1-Blue cells and transformants were selected on LB plates containing Gen (20 μg/ml final concentration). Clones were screened by PCR using ErmR cassette primers or SpcR cassette primers. Two PCR-positive clones were then sequenced. The donor strain E. coli S17-1, harboring the suicide vector pEX18Gm with the insert, and the recipient strain RA ATCC11845 were grown on LB or sheep blood plates, respectively, at 37°C overnight. Samples of the donor strain and the recipient strain were washed three times using 10 mM MgSO 4 . Then, the donor strain and the recipient strain were mixed at a ratio of 1:2 (2.5×10 8 : 5×10 8 ), and filtered through a Millipore membrane. The membrane was incubated on a blood agar plate at 30°C for 8 h. The transconjugants were selected on blood agar plates supplemented with Kan (20 μg/ml) and Erm (1 μg/ml) or Spc (80 μg/ml). The gene-deletion mutant strains were identified by PCR by amplifying the conserved 16S rRNA gene of R. anatipestifer using primers 16SrRNAP1 and 16SrRNAP1 and the deleted gene using the corresponding cloning primers (TonB1P1 and TonB1P2, TonB2P1 and TonB2P2). To exclude the possibility that gene inactivation had a polar effect on transcription of adjacent genes, total RNA was isolated from the mutant and wild-type (Qiagen), and RT-PCR (Qiagen) was performed to measure mRNA levels of the downstream gene (S1 Fig).

Construction of the Recombinant Vector for Complementation and Expression
Complete R. anatipestifer tonB1, tonB2, tonB3, exbBDtonB1, exbBDDtonB2 and OMhemR RA genes were amplified by PCR from R. anatipestifer chromosomal DNA using primers TonB1P1 and TonB1P2, TonB2P1 and TonB2P2, TonB3P1 and TonB3P2, ExbB1P1 and TonB1P2, ExbB2P1 and TonB2P2, OMhemRP1 and OMhemRP2, respectively (S1 Table), for complementation. Complete R. anatipestifer tonB1, tonB2, and tonB3 genes were amplified by PCR from R. anatipestifer chromosomal DNA using primers TonB1P1 and TonB1P2 his, TonB2P1 and TonB2P2 his, TonB3P1 and TonB3P2 his, respectively (S1 Table), for expression of TonB proteins. These PCR fragments were purified and digested with EcoRI and HindIII except for ExbBDDTonB2, which was digested with NheI and HindIII. Then, fragments, except for the OMhemR fragment, were ligated into the pBAD24 plasmid digested with corresponding restriction endonucleases. OMhemR fragment was ligated into the pAM238 plasmid digested with the corresponding restriction endonucleases. Ligation mixtures were introduced into the CaCl 2 -competent E. coli XL1-Blue strain and transformants were selected on LB plates containing Amp at 100 μg/ml. Clones were screened by the PCR method with corresponding primers. Validity of the sequences was determined by sequencing.

Expression and Purification of Recombinant His-tagged TonB proteins
Strains JP313 pBAD24::tonB1, JP313 pBAD24::tonB2 and JP313 pBAD24::tonB3 were grown overnight at 37°C in LB broth containing 100 μg/ml Amp. Then, 200 ml LB broth containing 100 μg/ml Amp were inoculated to an OD600 of 0.05 with the overnight culture and grown at 37°C. Expression was induced at an OD 600 of 0.6 for 2 h by adding 0.02% arabinose. Bacteria were harvested by centrifugation for 10 min at 5,000 g and 4°C, and the pellet was suspended in 20 ml binding buffer (50 mM Tris-HCl, 250 mM NaCl, 0.05% triton, pH 8.0) containing lysozyme (1 mg/ml final concentration) and DNaseI (1 U/ml final concentration). Lysis of bacteria was obtained by freezing and thawing at least three times. The suspension was then centrifuged at 12,000 rpm for 30 min at 4°C. The supernatant containing the soluble fraction was mixed with 200 μl of Ni-NTA-agarose beads according to the manufacturer's instructions. Purified protein was dialyzed twice against a buffer containing 50 mM Tris-HCl to eliminate any residual imidazole. The protein was stable for at least one month when kept at -80°C with 20% glycerol.

Antibody Preparation
200 μl of an emulsion containing each purified TonB (50 μg) and Freund's adjuvant (100 μl) were inoculated twice (at half-month intervals) into 3 four-week-old KunMing mice obtained from the Chengdu DaShuo Biological Technology Co., LTD. Two weeks after the second inoculation, 200-μl blood samples were collected every 3 weeks via retro-orbital bleeding when the mice were anesthetized from smelling ether. At the end of the study, the mice were euthanized with 150 mg/kg sodium pentobarbital. Blood samples were centrifuged twice (3,600 rpm for 5 min) to obtain serum, which was stored at -20°C. Before use, non-specific antibodies were removed by incubating the immune serum with E. coli cell extract for 1 h at 4°C followed by centrifugation for 10 min at 8,000 rpm. The supernatant was then used as serum.

Protein Analysis by Electrophoresis
Proteins were analyzed by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by Coomassie Blue staining.

Immunoblot Analysis
SDS-PAGE and immunoblotting, used to detect expression of TonB in R. anatipestifer or E. coli were performed as follows. The biomass of tested strains was collected, suspended in PBS buffer, and centrifuged. Bacterial pellets calculated to contain 20 μg of protein were suspended in loading buffer and heated for 5 min at 100°C. Proteins were separated by 12% SDS-PAGE and subsequently transferred to a nitrocellulose membrane. Non-specific binding sites were blocked with 5% skim milk in TBS-Tween 20 (0.05%). The blot was probed with polyclonal mouse sera raised against recombinant TonB (1:400), followed by a 1:2,000 dilution of a goat anti-mouse IgG alkaline phosphatase-conjugated secondary antibody (CST). The binding of antibodies to TonB protein was revealed using BCIP/NBT solution following the manufacturer's instructions (Sigma).

Protein Assay
The concentrations of the TonB proteins were determined using the BCA assay protein quantitation kit (Thermo Fisher).

Statistical Analysis
Statistical analysis was performed using GraphPad Prism 5 software for Windows. Statistical significance of the data was ascertained by use of Student's T test. A value of P<0.05 was considered significant.

Ethics Statement
Animals were handled in strict accordance with good animal practice as defined by the local animal welfare bodies. Animal work performed at the Sichuan Agriculture University was reviewed and approved by the Sichuan Agriculture University ethics committee in September 2014.

Sequence Analysis of the TonB Genes from R. anatipestifer
Genome analysis revealed the locations of three hypothetical tonB genes in the R. anatipestifer genome (Fig 1A). We termed them tonB1, tonB2 and tonB3, respectively. The tonB1 and tonB2 genes are organized into operons with their corresponding exbB and exbD genes, similar to the organization in Campylobacter jejuni, Vibrio anguillarum and Vibrio alginolyticus [39][40][41]. In contrast to tonB1, the tonB2 operon includes two corresponding exbD genes, as occurs in Flavobacterium psychrophilum, Acinetobacter baumannii and Xanthomonas campestris [34,42,43]. The tonB3 is present as a monocistronic copy, like in Acinetobacter baumannii [43], which does not correspond with expectations based on the exbB and exbD genes in R. anatipestifer.
The first hypothetical TonB1 system contains three genes: exbB, coding for a 24-kDa protein, exbD, coding for a 14.7-kDa protein, tonB, coding for a 30.6-kDa protein. This system is flanked by an upstream gene transcribed in the same direction that codes for a hypothetical protein and a downstream gene transcribed in the same direction that codes for a predicted hi0933 family protein (Fig 1B). It is worth noting that approximately 8 kb upstream of TonB1 system, there are three related genes transcribed in the opposite direction that code for a TonBdependent siderophore receptor, an ABC transporter related protein, and a transport-system permease protein. The second hypothetical TonB2 system contains four genes: exbB2, coding for a 30.8-kDa protein, exbD21, coding for a 21.1-kDa protein, exbD22, coding for a 20.4-kDa protein, and tonB2, coding for a 31-kDa protein. The TonB2 system, which is located 16 kb downstream of the TonB1 system, is flanked by an upstream gene transcribed in the same direction that codes for a leucyl-tRNA synthetase and a downstream gene transcribed in the same direction that codes for an phosphate ABC transporter substrate-binding protein ( Fig  1C). In addition, genes coding for an ABC transport related protein and fibronectin, which may have some connection with TonB [44], are closely located to the TonB2 system. The third tonB gene, coding for a 31-kDa protein, is located far apart from the other two tonB gene clusters and is flanked by two genes that code for a permease-family protein and a ATPase ( Fig  1D). The analysis of different R. anatipestifer strains, including ATCC, CH-1, and CH-2, showed that the TonB systems have similar organizational styles in all sequenced genomes and high nucleotide sequence similarity (more than 90%, data not shown) In A. baumannii, the three TonB proteins have high sequence identity to each other [43]. However, in R. anatipestifer, there is only approximately 10.00% identity between TonB1 and TonB2, 10.65% identity between TonB1 and TonB3, and 34.71% identity between TonB2 and TonB3. Their transmembrane domains also have a low identity, less than 17% (data not shown). A similar degree of divergence was observed in ExbB, less than 22% identity (data not shown). However, the transmembrane domains of ExbD1 and ExbD22 have 43.3% identity (data not shown). Sequence analysis showed that the closest homologs of these three TonB proteins are mainly distributed in the Flavobacteria, Cytophagia, Sphingobacteria and Bacteroidia classes and show greater than 25% identity.
The carboxy-terminal domain (CTD) of TonB proteins is believed to be essential for interactions with the TonB-dependent outer membrane receptor TonB-box [5,45]. Therefore, studies of the CTD of TonB are important for the further understanding for the mechanism of TonB energy support for hemin or iron acquisition. Based on the ClustalW-based multiple sequence alignment (MSA) (S2 Fig) for representative sequences from each of the nine CTD clusters of TonB proteins from gram-negative bacteria, as classified by Byron [27], and the three TonB proteins of R. anatipestifer, the neighbor-joining bootstrap tree (Fig 2) showed the evolutionary relationship between the CTDs of the three R. anatipestifer TonBs with other representative TonBs from previous studies [27]. The result indicates that the CTDs of TonB2 and TonB3 fall into the 3A cluster while that of TonB1 falls into the 1C cluster, where 3A and 1C refer to a classification system developed by Byron to distinguish CTDs from TonB proteins [27]. Similarly, the secondary structure of the TonB2 CTD is almost the same as that of the TonB3 CTD, while the secondary structure of the TonB1 CTD is clearly different (data not shown). These results indicate that TonB2 and TonB3 are similar in sequence and structure. However, the three TonB proteins of R. anatipestifer show the differences between the CTDs of each TonB protein (S2 Fig). All three TonBs lack the highly conserved YP motif (S2 Fig), which has been identified in close contact with the TonB box of TonB-dependent transporters [46,47]. Moreover, we were unable to identify the highly conserved SSG motif, which has been identified in many different TonB sequences and is believed to play a role in outer-membrane receptor recognition [27].

Both TonB1 and TonB2 Systems Are Required for Hemin Uptake in R. anatipestifer
To investigate the roles of the TonB systems of R. anatipestifer, we first tried to construct the isogenic derivatives of R. anatipestifer ATCC11845 deleted for each TonB system. We successfully built mutants deleted for tonB1 or tonB2, and a mutant deleted for both tonB1 and tonB2. Validation of the mutants was performed by PCR and western-bolt experiments using antibodies directed against TonB1 or TonB2 (S3 Fig). However, we did not successfully build a tonB3 mutant despite many attempts. A similar situation was also observed in A. baumannii [43], and we hypothesize that TonB3 is an essential gene for R. anatipestifer ATCC11845.
R. anatipestifer ATCC11845 is unable to grow on LB plates because of the lack of the essential hemin synthesis genes hemF, hemY, and hemG [4]. However, R. anatipestifer ATCC11845 can grow on LB plates containing serum (Fig 3A), which is hypothesized to provide hemin. To demonstrate that hemin is essential for R. anatipestifer ATCC11845, we first checked the effect of hemin addition on bacterial growth on LB medium. The results showed that R. anatipestifer ATCC11845 was able to grow on LB plates only in the presence of 20 μM hemin (Fig 3B  and 3C). Surprisingly, other commonly used hemin sources like bovine hemoglobin were insufficient to support the growth of R. anatipestifer ATCC11845 on LB medium (data not shown). To identify whether the TonB systems were involved in hemin uptake, we checked for the growth of R. anatipestifer ATCC11845 tonB1 and tonB2 mutants on LB plates in the absence and presence 20 μM hemin. Our results showed that the wild-type strain and tonB1 mutant strain formed small colonies after 24 h growth in the presence of hemin. In contrast, no growth was observed for the tonB2 mutant or tonB1tonB2 double mutant (data not shown). After 48 hours incubation, the tonB2 mutant strain gradually formed pinpoint colonies. In contrast, the tonB1tonB2 double mutant strain failed to exhibit any growth after 48 hours of incubation ( Fig 3C). All these strains could be grown on LB plates containing 5% bovine serum and had no growth on LB-only plates (Fig 3A and 3B).

The Fe 3+ Acquisition of R. anatipestifer is Mainly Supported by the TonB2 System
To identify the TonB systems of R. anatipestifer involved in iron uptake, we tested R. anatipestifer tonB mutants for the growth on iron-chelated medium. The results showed that all the TonB of R. anatipestifer Are Required for Iron and Hemin Utilization strains grew well on LB plates containing 5% bovine serum (Fig 3A). The wild-type strain and tonB1 mutant grew well on plates supplemented with 40 μM Dip. In contrast, the tonB2 mutant and tonB1tonB2 double mutant grew poorly on the same medium (Fig 3D). These results suggested that the TonB2 system of R. anatipestifer plays an important role in iron uptake.
Activity of TonB Systems of R. anatipestifer Expressed in E. coli As described above, TonB2 is involved in iron and hemin uptake processes in R. anatipestifer. TonB1 disruption strengthens the effect of the tonB2 mutation on hemin uptake. A heterologous host like E. coli is commonly used to characterize the activity of TonB proteins from various gram-negative bacteria, such as Bartonella birtlesii [48] and A. baumannii [43]. To characterize the function of the three TonB systems from R. anatipestifer, we assessed the ability of each system to complement a tonB mutation in E. coli. We thus compared the ability of the three TonB proteins and two TonB complexes from R. anatipestifer to provide energy to a heterologous heme transporter, HasR or HemR from S. marcescens [31], expressed in an E. coli tonB mutant. We introduced plasmids expressing TonB proteins from R. anatipestifer into the E. coli strain C600ΔhemA tonB::Tn10 pAM238::hasR/hemR and checked the growth on LB plates in the presence of hemin. All the C600ΔhemA tonB::Tn10 pAM238::hasR/hemR derivative strains, producing TonB1 complex, TonB2 complex or TonB3 protein from R. anatipestifer (S4 Fig), were unable to grow in the presence of hemin (data not shown).
The energizing activity of the three TonB proteins was also investigated in a E. coli C600ΔhemA tonB::Tn10 derivative expressing a putative TonB-dependent hemin transporter from R. anatipestifer (gene ID: 11995802) cloned into plasmid pAM238 (pAM238::OMhemR RA ). First, the plasmid pBAD24 and its derivatives expressing TonB1, TonB2,TonB3 or TonB2 complex were introduced in strain C600ΔhemA tonB::Tn10 pAM238::OMhemR RA and the resulting strains were tested for growth in the absence or in the presence of hemin. All the tested strains grew well on LB plate supplemented with 50 μM δ-ala (Fig 4A) while they showed limited growth on LB plates or LB plates containing 20 μM hemin (data not shown). In contrast, the C600ΔhemA tonB::Tn10 pAM238::OMhemR RA derivative strain expressing ExbB-ExbD-TonB1 of R. anatipestifer showed limited growth on LB plates ( Fig 4B) and grew well on LB plates containing 20 μM hemin (Fig 4C and 4D). These results indicate that TonB1 cannot interact with ExbB-ExbD from E. coli and that only the complete ExbB1-ExbD1-TonB1complex from R. anatipestifer can energize OMHemR RA for its hemin uptake activity.
The TonB1 System of R. anatipestifer can Restore the Growth of C600ΔhemA tonB::Tn10 under Iron-chelated Conditions In E. coli, all the iron Fe 3+ uptake systems require the energizing activity of TonB. An E. coli tonB mutant cannot grow on iron-depleted media. We assessed the ability of TonB1 or the ExbB1-ExbD1-TonB1 system of R. anatipestifer to energize E. coli iron transporters. The C600ΔhemA tonB::Tn10 strain derivatives trains expressing TonB1 or ExbB1-ExbD1-TonB1 from R. anatipestifer were tested for growth on LB plates with δ-ala in the absence or in the presence of 150 μM iron chelator Dip. The tested strains grew as well as the positive control C600ΔhemA pBAD24 on LB δ-ala plates in the absence of Dip (Fig 5A). Addition of Dip strongly inhibited the growth of strains C600ΔhemA tonB::Tn10 pBAD24 and C600ΔhemA tonB::Tn10 pBAD24::tonB1 (Fig 5B). In contrast, expression of the complete ExbB1-ExbD1-TonB1 system in the strain allowed normal growth of the C600ΔhemA tonB::Tn10 pBAD24:: exbB1-exbD1-tonB1 strain in the presence of Dip, compared with strain C600ΔhemA pBAD24 (Fig 5B and 5C).
We also checked for the ability of TonB2, ExbB2-ExbDD-TonB2 and TonB3 to restore the growth of C600ΔhemA tonB::Tn10 derivative strains in the presence of Dip. The results showed that the strain expressing TonB2, ExbB2-ExbDD-TonB2 and TonB3 had no increase in growth compared to the negative control C600ΔhemA tonB::Tn10 pBAD24 under iron-limited conditions (data not shown).

Discussion
R. anatipestifer does not encode the genes hemF, hemY, and hemG, which are essential for de novo synthesis of hemin. Hemin is not only an essential component as the prosthetic group of many proteins [4], but it is also a major iron source [8]. TbdR1 was shown to be a potential outer-membrane hemin receptor and hemin was shown to act as a potential iron source in R. anatipestifer [8], indicating that R. anatipestifer has the ability to express at least one hemin uptake system that includes a TonB system. Previous research revealed that the transport of siderophore-loaded Fe 3+ is also TonB dependent due to the energy transducing roles of TonB [8,9]. In this study, we proved that TonB proteins in R. anatipestifer are required for both iron and hemin transport. R. anatipestifer encodes two sets of TonB systems and one TonB family protein. The TonB system involved in hemin acquisition and/or iron uptake cannot be predicted in silico because analysis of the neighborhood of genes encoding these three TonB systems do not clearly suggest the individual roles of the three TonB proteins in hemin or iron uptake. For many gramnegative bacteria, such as P. damselae [49] and V. anguillarum [50], hemin/iron uptake-related genes are located in the same operon as the tonB gene in a cotranscribed direction. For other gram-negative bacteria, such as the Flavobacteriia sychrophilum [34], tonB genes are not located next to other genes encoding hemin/iron uptake systems. In still other types of gram-negative bacteria, such as Serratia marcescens, one tonB-encoding gene is located in a transcription unit expressing its cognate heme uptake system, and the other tonB neighborhood not contains genes encoding for hemin or iron uptake systems [31].
To explore the function of TonB systems in hemin and iron acquisition in R. anatipestifer, we sought to construct isogenic tonB mutant strains. We used spectinomycin-resistance genes to replace tonB genes as described by Hu et al [37]. However, we found that the single tonB mutants did not produce a strong growth defect on the hemin plates or iron-depleted plates. Therefore, we constructed the double mutant for the TonB1 and TonB2 systems of R. anatipestifer. Using sequence analysis, we found that the genome of R. anatipestifer CH-1 encodes for an erythromycin-resistance gene not present in the genome of R. anatipestifer ATCC11845. Therefore, we attempted to use this cassette to replace the tonB genes. Attempts were successful for the tonB1 and tonB2 genes, though we failed to construct a tonB3 mutant strain using the spectinomycin or erythromycin cassettes. This result suggests that the activity of TonB3 is absolutely required for growth in R. anatipestifer. A similar situation was also observed for the tonB genes in A. baumannii [43]. Therefore, we supposed that tonB3 may be an essential gene involved in fundamental nutrient acquisition.
Hemin transport is completely abolished only when both tonB1 and tonB2 are deleted. This result indicates that both TonB1 and TonB2 are involved in hemin transport in R. anatipestifer and that the proteins are able to compensate for one another. In addition, we demonstrated that TonB2 is more important than TonB1 for hemin uptake in R. anatipestifer.
To further investigate more precisely the role of TonB proteins of R. anatipestifer in hemin and iron acquisition, we assessed the capacity of the proteins to complement hemin utilization in E. coli model strain C600ΔhemA tonB::Tn10 pAM238::hasR/hemR. All the three tonB genes and two complete TonB system operons were insufficient to restore the growth of C600ΔhemA tonB::Tn10 pAM238::hasR/hemR under low-hemin conditions. This result reveals that the TonB systems of R. anatipestifer cannot work with the heterologous hemin transporter HasR/ HemR of S. marcescens to transport hemin. This result can be explained by the very low polypeptide sequence identity (lower than 10%) between the three TonB proteins of R. anatipestifer and the TonB and HasB proteins of S. marcescens.
Our results obtained with E. coli demonstrate that the TonB1 complex of R. anatipestifer can energize OMhemR (AFD55772) of R. anatipestifer. That the TonB1 complex can energize OMHemR instead of HasR/HemR can be explained by the facts that OMhemR shares homologies lower than 10% identity with HasR/HemR (data not shown) and some TonB-dependent receptors function only with TonB from their original host [51]. This latter result strengthens the previous results suggesting that OMhemR (AFD55772) of R. anatipestifer is energized only by TonB1. Similar specificity of TonB systems toward a hemin transporter was also described in S. marcescens [31] and in H. influenza [52].
According to our results, the Fe 3+ acquisition of R. anatipestifer is supported mainly by the TonB2 system. Surprisingly, in the E. coli model, only the TonB1 system from R. anatipestifer was capable of energizing the transport systems devoted to iron uptake. The question of why the TonB1 system can work with the Fe 3+ receptor of E. coli but appears to be unnecessary for iron uptake in R. anatipestifer is not understood at this time. To identify receptors of E. coli worked with TonB1 system of R. anatipestifer is an interesting project for further study.
In view of the function of the TonB systems in hemin and iron acquisition in R. anatipestifer, we hypothesize that TonB systems influence the virulence of R. anatipestifer, similar to the cases observed in S. dysenteriae [53], E. coli [35] and A. baumannii [43]. Understanding the relationships between the individual TonB systems and the virulence of the respective bacteria is important for understanding the pathogenic mechanisms and for developing new ways to prevent R. anatipestifer infection.
Supporting Information S1 Fig. The transcription of the downstream genes of tonB1 and tonB2 in wild type and mutant strains. Total RNA was isolated from the mutant and wild-type. Then, cDNA were obtained through reverse transcription and were acted as templates of RT-PCR. Total RNA were used as a control to exclude the contamination of DNA. Primers hi0933P1 and hi0933P2 (S1 Table) were used to amplify a 114 bp target in the downstream gene of tonB1. Primers ABCP1 and ABCP2 (S1 Table) were used to amplify a 154 bp target in the downstream gene of tonB2.