Iron Uptake Mechanisms in the Fish Pathogen Tenacibaculum maritimum

We present here the first evidence of the presence of iron uptakemechanisms in the bacterial fish pathogen Tenacibaculum maritimum.Representative strains of this species, with different serotypesand origins, were examined. All of them were able to grow inthe presence of the chelating agent ethylenediamine-di- (o-hydroxyphenylacetic acid) (EDDHA) and also produced siderophores. Cross-feedingassays suggest that the siderophores produced are closely related.In addition, all T. maritimum strains utilized transferrin,hemin, hemoglobin, and ferric ammonic citrate as iron sourceswhen added to iron-deficient media. Whole cells of all T. maritimumstrains, grown under iron-supplemented or iron-restricted conditions,were able to bind hemin, indicating the existence of constitutivebinding components located at the T. maritimum cell surface.This was confirmed by the observation that isolated total andouter membrane proteins from all of the strains, regardlessof the iron levels of the media, were able to bind hemin, withthe outer membranes showing the strongest binding. proteinaseK treatment of whole cells did not affect the hemin binding,indicating that, in addition to proteins, some protease-resistantcomponents could also bind hemin. At least three outer membraneproteins were induced in iron-limiting conditions, and all strains,regardless of their serotype, showed a similar pattern of inducedproteins. The results of the present study suggest that T. maritimumpossesses at least two different systems of iron acquisition:one involving the synthesis of siderophores and another thatallows the utilization of heme groups as iron sources by directbinding.

INTRODUCTION

Top

Introduction

Tenacibaculum maritimum (formerly Flexibacter maritimus) (48)is the etiological agent of an ulcerative disease known as marineflexibacteriosis or tenacibaculosis, which affects a large numberof marine fish in the world and thus has considerable economicsignificance to aquaculture producers. The infection has diverseclinical manifestations depending on the species and age offish, the most significant symptom being the presence of grosslesions on the body surface (11, 16).

Although the phenotypic, antigenic, and molecular characteristicsof T. maritimum have been examined by several authors (for areview, see reference 49), the actual factors determining thevirulence of this pathogen have not yet been elucidated. Somesynergistic interactions of the toxins contained in extracellularproducts and a hemolysin might be involved in T. maritimum infections(8). Moreover, pathological properties of the bacterium, suchas a strong adherence to the skin mucus of different fish speciesand the capacity to resist its bactericidal activity (31), havebeen pointed out as possible virulence factors.

It is well known that the ability to take up iron during infectionis an essential factor in the pathogenicity of several bacteria,being necessary for the pathogen multiplication. However, thelevels of free iron in the biological fluids within the animalhost is often very limited because the element is strongly boundto a high-affinity iron-binding proteins. To obtain this unavailableiron, most pathogenic bacteria have developed iron uptake systemsthat usually involve two components: (i) low-molecular-weightsiderophores released by the bacteria that will chelate ironand subsequently transfer it to the pathogen and (ii) iron-regulatedouter membrane proteins (IROMPs) that function as receptorsof the iron-siderophores complexes (36, 43). These mechanismsof iron acquisition have been linked to the virulence of differentfish pathogens such as Vibrio anguillarum (Listonella anguillarum)(35, 53), Aeromonas salmonicida (21, 25), Photobacterium damselaesubsp. piscicida (17, 30), and Edwardsiella tarda (23, 24).However, other pathogenic bacteria have developed mechanismsto acquire iron relying on the interaction between specificmicrobial receptors and host transferrin or heme-containingcompounds (28, 39, 51).

Recently, in vitro studies have shown that iron seems to bean important factor for the resuscitation of a Tenacibaculumsp. strain from the viable but nonculturable state (32), suggestingthat, as in many bacteria, this micronutrient plays a key rolein the growth of T. maritimum. Therefore, we sought to get afirst insight into the mechanisms that T. maritimum possessfor iron assimilation from the host tissues.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains.
A total of 17 T. maritimum strains were included in the presentstudy (Table 1). This collection comprises 15 strains isolatedfrom six different marine fish species, which belongs to thedifferent serotypes and clonal lineages described within thispathogen (4, 5, 6), and two reference strains (NCIMB 2154^T and2158) from the National Collection of Industrial and MarineBacteria (Aberdeen, United Kingdom). These strains were confirmedas T. maritimum by PCR-based analysis as described by Toyamaet al. (50). For all experiments, T. maritimum cells were routinelycultivated on Flexibacter maritimus medium (FMM) agar or brothat 25°C for 48 h. FMM (0.5% peptone [Difco Laboratories,Madrid, Spain], 0.05% yeast extract [Oxoid, Ltd., Basingstoke,Hampshire, England], and 0.001% sodium acetate [Sigma AldrichQuímica, S.A., Madrid, Spain] supplemented with 1.5%agar [Cultimed Panreac Química S.A., Barcelona, Spain];pH 7.2 to 7.4) was prepared with seawater as diluent accordingto the original descriptions (41). Stock cultures were maintainedfrozen at –80°C in Cryo-bille tubes (AES Laboratories,France).

View this table:
[in this window]
[in a new window]
TABLE 1. T. maritimum strains used in this study and results of growth under iron-limiting conditions and siderophore production

Growth under iron-limiting conditions.
The ability of the T. maritimum strains to grow in the presenceof the nonassimilable iron chelator ethylenediamine-di-(o-hydroxyphenyl-aceticacid) (EDDHA; Sigma) was determined by using FMM broth as abasal medium due to the known strict halophilic nature of thisbacterium (7). MICs were obtained by adding increasing concentrationsof EDDHA and were defined as the lowest concentration at whichno bacterial growth was observed. All experiments were carriedout in triplicate with three different batches of media.

Production of siderophores.
Production of compounds with siderophore activity was testedby the universal chemical assays of Schwyn and Neilands (45)in solid and liquid media. The assays were performed by spotting10 µl of each bacterial culture grown in iron-limitingconditions onto a modified chromoazurol S (CAS; Sigma) agar,which was prepared with FMM as the basal medium. The siderophorelevels produced by the strains on plates were expressed as theratio of orange halo diameter to growth diameter after 72 hof incubation. For siderophore detection in liquid media, supernatantsfrom culture grown in FMM broth containing 80 µM EDDHAwere mixed with CAS supernatant solution, and the absorbanceof mixture was measured at 630 nm. As a positive control weused a V. anguillarum wild-type strain 775, and as a negativecontrol we used a non-siderophore-producing strain V. anguillarum775::Tn1-5 (30).

The presence of phenolic compounds and/or hydroxamic acids weredetected in cell-free supernatants obtained from iron-depletedcultures by the colorimetric assays of Arnow (3) and Csáky(14) as modified by Andrus et al. (1), respectively. The positivecontrols for phenolate-type siderophores consisted in the iron-limitingFMM broth containing 10 µM 2,3-dihydroxybenzoic acid,while the positive control for hydroxamate-type siderophoreswas the same media with 10 µM purified aerobactin.

Cross-feeding assays.
To test the ability of each T. maritimum strain to induce thegrowth of other strains of the same species subjected to ironstarvation, cross-feeding assays were used. Briefly, T. maritimumstrains were grown in FMM broth in iron-limiting and iron-repleteconditions (80 µM EDDHA and 20 µM FeCl₃, respectively)and centrifuged at 12,000 x g for 5 min. Cell-free supernatantsfrom the strains to be tested for the production of siderophore-likecompounds were pipetted onto sterile filter paper disks and,after being dried, were placed on the agar surface of FMM platesseeded with the strain to be used as indicator and containinga concentration of EDDHA higher than the MIC for that strain.Growth around the disks indicates the production by the testedstrain of a diffusible siderophore that can be utilized by theindicator strain. As a negative control a non-siderophore-producingstrain V. anguillarum 775::Tn1-5 was used (30).

Growth with different iron sources.
FMM broth supplemented with EDDHA at a concentration sufficientto achieve total growth inhibition of the strains tested wassupplemented by the utilization of various iron sources: transferrin(human), apotransferrin (human), hemin (bovine), hemoglobin(bovine), and ferric ammonium citrate (Panreac QuímicaS.A.). Transferrin and apotransferrin (Sigma) were dissolvedto 1 mM in 100 mM Tris, 150 mM NaCl, and 50 mM NaHCO₃ (pH 8.0).Hemin (Sigma) stock solution was prepared in 10 mM NaOH andhemoglobin (Sigma) in distilled water. All stock solutions weresterilized by filtration through 0.22-µm-pore-size membrane(Millipore) and maintained at –20°C until use.

Iron-containing compounds were used at concentrations rangingbetween 0.1 and 20 µM. FMM broth plus EDDHA tubes supplementedwith iron components were seeded with a 1:100 (vol/vol) inoculumfrom an overnight culture in FMM broth and were incubated at25°C. The absorbance of the mixture was determined after48 and, if necessary, 72 h of incubation at 600 nm. Growth withheme compounds as the only iron source was also examined onFMM solid medium plus EDDHA by a standard radial diffusion method.Sterile paper disks impregnated with 10 µl of either heminor hemoglobin at a concentration of 0.1, 5, 10, and 20 µMwere placed onto the previously inoculated medium. The plateswere observed after 72 h of incubation at 25°C to detecthalos of growth around disks. Moreover, in order to evaluatethe effect of hemin and hemoglobin on growth rates, bacterialgrowth curves were obtained. Inocula obtained as described abovewere added to iron-restricted medium containing these iron sourcesat a concentration of 5 µM and were incubated at 25°C.Samples were taken every 6 h and the optical density at 600nm was measured.

Hemin-binding assays.
The existence of possible membrane receptors for heme compoundswas determined by Congo red and hemin-binding assays in liquidand solid-phase according to the procedures described by Kayet al. (26) and Mazoy and Lemos (34). T. maritimum cells grownin FMM containing 20 µM FeCl₃ and FMM plus 80 µMEDDHA were harvested by centrifugation, washed in sterile 0.9%saline solution, and suspended in saline solution (20 ml) toan optical density of 1.5 at 620 nm. Congo red (Sigma) or heminwas added to a final concentration of 30 or 40 µg ml^–1,respectively. One-milliliter samples were immediately removedand centrifuged at 13,000 x g for 30 s, and the supernatantwas assayed spectrophotometrically for Congo red (A₄₈₈) or hemin(A₄₀₀). The remaining cells were shaken at 25°C and assayedat 30-min intervals for residual Congo red or hemin in the supernatantas described above. The experiment was performed in duplicate.

For the whole-cell binding assay in solid-phase, bacterial strainswere cultured under iron-limited and iron-supplemented conditionsand collected by centrifugation at 4,000 x g for 5 min. Thesecells were washed in saline solution and resuspended to an opticaldensity of 0.8 at 580 nm. A 30-µl volume of the cell suspensionwas placed onto nitrocellulose membranes (0.45-µm poresize) in a dot blot manifold (Schleicher & Schuell, Inc.,Dassel, Germany). After immobilization the membranes were airdried, blocked with 2% gelatin in Tris-buffered saline (TBS;50 mM Tris-HCl supplemented with 0.9% NaCl [pH 8.0]), incubatedfor 2 h with hemin (10 µM in TBS), immersed for 30 minin 12.5% trichloroacetic acid, washed in distilled water, andstained with 3,3'-dimethoxybenzidine (DMB; Sigma). The DMB solution(50 mg of DMB in 15 ml of distilled water) was freshly preparedjust before use and stirred for 15 min, and 5 ml of a 0.5 Msodium citrate buffer (pH 4.4) and 100 µl of 30% H₂O₂were added. After staining for 3 to 5 min, the membrane waswashed with water to clear the background. To determine whetherhemin binding was protein associated, bacterial suspensionswere incubated with proteinase K solution (500 µg ml^–1)in phosphate-buffered saline (pH 7.4) for 1 h at 37°C beforethe solid-binding assays. Escherichia coli HB101 and V. anguillarumstrain 775 were used as negative and positive controls, respectively(17).

Analysis of enzymatic activities.
The enzymatic activities of T. maritimum grown in iron-richand iron-restricted media were comparatively evaluated in orderto determine the possible role of iron in the enzymatic expression.The hydrolysis of the following substrates were examined byusing FMM as basal medium supplemented with 80 µM EDDHAor 20 µM FeCl₃: 2% (wt/vol) gelatin (Sigma), 1% (wt/vol)starch (Merck, Darmstadt, Germany), 3% (wt/vol) carboxymethylcellulose (Sigma) and 0.1% (wt/vol) esculin (Merck). These substratesare normally used for the biochemical characterization of T.maritimum (4). All of these activities were determined by standardradial diffusion method with filter paper disks impregnatedwith 10 µl of each live T. maritimum cells and with bacterial-cell-freesupernatants. All tests were read after 72 h at 25°C. Inaddition, the presence and levels of enzymatic activity wereexamined with the API ZYM (bioMèrieux) miniaturized systemof both live cells and supernatant of each T. maritimum straingrown in iron-supplemented and iron-deficient conditions accordingto the manufacturer's instructions with the exception of theincubation temperature that was fixed at 25°C.

Analysis of membrane proteins.
The presence of iron-regulated proteins was tested by growingeach T. maritimum strain in FMM broth containing 80 µMEDDHA and by comparing the patterns with those appearing whenthe strains were grown in FMM liquid plus 20 µM FeCl_3.Total and outer membrane proteins were obtained as previouslydescribed Avendaño-Herrera et al. (4) and were separatedby sodium dodecyl sulfate-polyacrylamide gel electrophoresis(27) with 12% (wt/vol) acrylamide in the resolving gel and 4%(wt/vol) acrylamide in the stacking gel using a Mini Protean3 cell apparatus (Bio-Rad). Both total membrane and outer membraneproteins were adjusted to a concentration of 40 ± 2 µgwith bovine serum albumin as the standard according to the methodsdescribed by Bradford (10). After electrophoresis (60V for 90min), the proteins were stained with 0.05% (wt/vol) Coomassieblue R (Sigma) in 25% (vol/vol) propan-2-ol-10% (vol/vol) aceticacid. Gels were destained with 10% acetic acid and 40% methanoland photographed. All experiments were carried out with proteinsobtained in two different extractions for each bacterial strain.

Hemin-binding by bacterial membrane protein extracts.
Binding of hemin by membrane protein extracts was assayed aspreviously described (17). Briefly, 20 µl of either totalor outer membrane proteins containing 80 or 40 µg of proteinin distilled water were immobilized on nitrocellulose membranes,which were then blocked with gelatin, incubated with hemin,immersed in 12.5% trichloroacetic acid, washed in distilledwater, and stained with DMB, as described above. The effectof protease treatment on hemin binding by protein extracts wasevaluated by incubating bacterial total membrane and outer membraneprotein with proteinase K solution (500 µg ml^–1)in phosphate-buffered saline for 1 h at 37°C. The sampleswere then loaded onto nitrocellulose membrane and subjectedto a dot blot assay to evaluate the binding of hemin.

RESULTS AND DISCUSSION

Top

Results and Discussion

The presence of efficient iron uptake systems has been consideredan important virulence determinant for several pathogens (40).These microorganisms must compete with the host for iron, andin many instances do so via the production of high-affinityiron chelators which, in conjunction with cell surface receptors,transport iron into cells (19, 43). Although tenacibaculosiscaused by T. maritimum is a disease affecting important marinefish (see reference 49 for a review), the studies about itsvirulence mechanisms are scarce (8). For this reason, in thepresent study we studied the presence of iron acquisition mechanismsin T. maritimum which could have an important role in the pathogenicityfor fish.

Growth in iron-limiting conditions and siderophore production.
The results obtained showed that the 17 T. maritimum isolatestested were able to grow under iron-limiting conditions withMICs for the iron-chelating agent EDDHA ranging from 130 to180 µM (Table 1). In fact, the siderophore productionassays revealed that all T. maritimum strains gave positivereaction on CAS agar plates and CAS liquid test, showing a goodcorrespondence between both assays. The T. maritimum isolatesshowed halo/growth diameter ratios ranging from 1.33 to 2.17and absorbance values ranging from –0.14 to –0.50(Table 1). However, although all strains produced siderophores,the chemical assays performed with cell-free supernatants showedthat they do not contain typical hydroxamate- or phenolate-typecompounds. The occurrence of a siderophore which is neithera phenolate nor a hydroxamate has been reported for other bacteria(22, 47). Further studies are needed to elucidate the chemicalstructure of the siderophore produced by T. maritimum.

Based on the biochemical homogeneity of the bacterium and theresults of siderophore production assays, three T. maritimumstrains isolated from sole (PC503.1 and ACC6.1) and turbot (PC424.1),representing the main serotypes described in this pathogen (4,6), were selected to be used in the subsequent studies.

The cross-feeding assays showed that the three strains testedwere able to produce iron-sequestering compounds only when theywere cultured under iron-limiting conditions (Table 2). Eachstrain was able to use the compounds secreted by itself, aswell as the compounds secreted by the other two strains. Thissuggests that all T. maritimum strains studied likely producehighly related siderophores.

View this table:
[in this window]
[in a new window]
TABLE 2. Results of cross-feeding experiments in T. maritimum grown in FMM broth under iron-replete and iron-restricted conditions

Growth with different iron sources.
Another finding in our study was that when high inhibitory concentrationof EDDHA in the culture medium was imposed, all T. maritimumisolates utilized transferrin, apotransferrin, hemin, hemoglobin,and ferric ammonium citrate as the only iron source. As observedin Table 3, all compounds could stimulate growth in all strainsat the lowest concentration tested (0.1 µM), with an optimumconcentration of 10 µM for hemin and hemoglobin. It isinteresting to denote that, with the exception of ferric ammoniumcitrate, higher concentration of the compounds tested did notresult in an increase of cells yield, being more evident whenthe assays with transferrin and apotransferrin were performed.Similar behavior has been reported for V. anguillarum (33).When the growth kinetics of all T. maritimum strains under heminor hemoglobin-conditions were tested, strains ACC6.1 and PC424.1exhibited very fast growth (Fig. 1). According to these observations,T. maritimum can efficiently use iron contained in hemin orhemoglobin in agreement with the observations reported for otherpathogenic microorganisms (for a review, see reference 38).

View this table:
[in this window]
[in a new window]
TABLE 3. Growth of the three T. maritimum strains with different iron sources in FMM broth supplemented with EDDHA at a concentration sufficient to achieve total growth inhibition of the strains tested

View larger version (24K):
[in this window]
[in a new window]
FIG. 1. Utilization of hemoglobin (Hb) at 5 µM (A) and hemin (He) at 5 µM (B) as iron sources by T. maritimum strains grown in FMM medium containing EDDHA.

Hemin-binding assays.
To determine whether the T. maritimum species contain iron-regulatedmembrane binding involved in iron acquisition from heme compounds,we grew all strains in the presence of 80 µM EDDHA andmeasured the Congo red-binding ability of these cells. The threeT. maritimum strains did not show an increase in Congo red absorptionin relation to the cells grown in FMM without EDDHA. The sameeffect was noted when the hemin-binding ability of T. maritimumwas measured (Fig. 2). Previous reports have shown that somepathogenic bacteria can absorb hemin and the structurally similararomatic dye Congo red and that this ability is strongly correlatedwith virulence (15, 42). The molecular basis for this absorptionis unclear. In our case no differences in the Congo red or heminacquisition was observed by all of the strains tested regardlessof the media used, indicating that this could be due to theexistence of constitutive binding-molecules located at the T.maritimum cell surface.

View larger version (14K):
[in this window]
[in a new window]
FIG. 2. Binding of Congo red (A) and hemin (B) by three T. maritimum isolates. Cells were cultured under either iron-restricted ( ${blacksquare}$ ) and iron-replete ( ${square}$ ) conditions. Negative control, Congo red, and hemin data from a cell-free flask ( ${blacktriangleup}$ ) are also shown. The uptake of Congo red and hemin are expressed here as depletion of the dye from solution as a function of time. Bars represent the standard deviation of the data from two different experiments.

This hypothesis was confirmed by using dot blot assays. It wasfound that whole cells of all T. maritimum strains tested, regardlessof the culture conditions, were able to bind hemin (Fig. 3).Interestingly, when iron-limited or iron-supplemented bacterialcells were pretreated with proteinase K, their ability to bindhemin was not reduced compared to that of cells not treatedwith the enzyme. It indicates that, in addition to protein binding,some protease-resistant components could also bind hemin. Similarresults have been reported for other gram-negative bacterialspecies (17, 18) in which some other components, such as capsularpolysaccharides and/or lipopolysaccharides, allows bacteriato bind hemin. In fact, the ability of T. maritimum purifiedlipopolysaccharides fraction to bind hemin, and thus its affinityfor the compound, was clearly detected (data not shown).

View larger version (38K):
[in this window]
[in a new window]
FIG. 3. Binding of hemin by T. maritimum. Lanes: A, strain PC503.1; B, strain PC424.1; C, strain ACC6.1. Rows: 1 and 2, whole cells cultured under iron-supplemented (FMM plus FeCl₃) (row 1) and iron-restricted (FMM with EDDHA) (row 2) conditions; 3 and 4, outer membrane proteins obtained under iron-supplemented (row 3) and iron-restricted (row 4) conditions; 5 and 6, effect of proteinase K on the binding activity cultured under iron-supplemented (row 5) or iron-restricted (row 6) conditions; 7, controls (positive control V. anguillarum 775 [lane A]; negative control E. coli HB101 [lane B]; proteinase K alone [lane C]).

Analysis of enzymatic activities.
Whole cells and supernatants of the three T. maritimum isolates,grown in iron-rich or iron-restricted conditions, exhibitedidentical enzymatic profiles in API ZYM gallery, matching thetypical profile described for T. maritimum species (positiveresults in the first 11 enzymatic reactions) (4, 9, 12). Moreover,none of the isolates were able to degrade starch, carboxymethylcellulose, and esculin when the enzymatic activity of the T.maritimum strains were compared by using iron-rich and iron-restrictedmedia. However, we found that the three strains failed to hydrolyzegelatin onto media supplemented with EDDHA (the incorporationof EDDHA to the medium did not affect the total growth of theT. maritimum strains tested), indicating that this activityis iron dependent because T. maritimum has been classicallyconsidered a proteolytic bacterium in normal culture conditions(4, 12, 52). This inability to produce gelatinase was due tothe reduction of levels of free iron present in FMM medium,strongly suggesting that iron is required for the synthesisand/or activity of this protease. In other fish pathogens suchas A. salmonicida it has been demonstrated the production ofmetalloproteases with gelatinolytic activity, which may havea broad specificity for degrading extracellular matrix component,muscle protein, and other substrates (2).

Interestingly, the pathogenicity of T. maritimum has been attributedto extracellular products and hemolysins (8), which could facilitatethe alteration and erosion of the host tissue contributing tothe colonization and invasion, as has been described in A. salmonicida(20), V. anguillarum (37), and Yersinia ruckeri (46). Thus,it would be reasonable to speculate that once the bacterialmultiplication begins, the production of lytic enzymes, suchas hemolysins, could make the heme groups or hemoglobin releasedfrom lysed erythrocytes readily available for utilization asan iron source, triggering the expression of virulence factors.However, preliminary in vitro studies about the possible regulationby iron of the hemolysins produced by T. maritimum showed nodifferences in the hemolytic activity when the cells were grownunder iron-restricted and iron-replete conditions (data notshown), suggesting that hemolysins are not iron regulated inthis bacterium.

Induction of outer membrane proteins.
When the T. maritimum strains were grown under iron-restrictedconditions, all of them, regardless of their serotype, showedthe induction of three IROMPs with molecular masses of ca. 128,85, and 66 kDa (Fig. 4). The induction of IROMPs has been reportedin other fish pathogens such as V. anguillarum (13, 29, 35),Y. ruckeri (44), P. damselae subsp. piscicida (30), and E. tarda(23). At present, the function of the T. maritimum IROMPs isunknown, although it is tempting to speculate that some or allof these proteins may serve as receptor(s) for siderophore-ironcomplexes or heme groups or in some way interact directly withhost iron-carrying compounds. Further experiments, includingan analysis of T. maritimum mutants lacking IROMPs, are necessaryto confirm the role of these proteins.

View larger version (69K):
[in this window]
[in a new window]
FIG. 4. SDS-PAGE analysis of outer membrane proteins of T. maritimum strains. Lanes: 1, molecular size markers; 2 and 3, strain PC503.1; 4 and 5, strain PC424.1; 6 and 7, strain ACC6.1. Lanes 2, 4, and 6 contained outer membrane proteins from cells cultured on FMM medium containing 20 µM FeCl_3. Lanes 3, 5, and 7 contained outer membrane proteins from cells grown on FMM medium containing 80 µM EDDHA. Arrows indicate the IROMPs. Numbers on the left indicate the positions of molecular size markers, and numbers on the right show the molecular masses of the IROMPs (in kilodaltons).

On the other hand, using the dot blot assay, it was found thatboth total and outer membrane proteins from iron-supplementedor from iron-restricted T. maritimum cells were able to bindhemin. The affinity for this compound was decreased by pretreatmentwith proteinase K. The residual binding observed was likelydue to lipopolysaccharide that could be stuck to protein componentsduring protein extraction. With the same amount of protein,stronger binding to outer membrane proteins than to total proteinswas detected, indicating that a higher density of hemin-bindingcomponents is present in the outer membrane fraction (Fig. 3).

Concluding remarks.
Our data clearly show for the first time that T. maritimum possessesat least two different systems of iron acquisition: one involvingthe synthesis of siderophores and another that allows the utilizationof heme groups as an iron source by direct binding. Furtherstudies will be developed in order to establish a relationshipbetween iron uptake ability and virulence in T. maritimum byin vivo assays.

ACKNOWLEDGMENTS

This study was supported in part by grant AGL2004-07037 fromthe Ministerio de Ciencia y Tecnología of Spain (A.E.T.)and grant PGIDIT04RMA261014PR-3 from Xunta de Galicia (M.L.L).R.A.-H. also acknowledges a research fellowship from the CONICYT-BIDprograms of Chile.

We thank O. Vidal for technical assistance.

FOOTNOTES

* Corresponding author. Mailing address: Departamento de Microbiología y Parasitología, Facultad de Biología and Instituto de Acuicultura, Universidad de Santiago, 15782 Santiago de Compostela, Spain. Phone: 34-981-563100, ext. 13255. Fax: 34-981-596904. E-mail: mpaetjlb{at}usc.es.

REFERENCES

Top