- Email: [email protected]
Utilization of hemin and hemoglobin as iron sources by Vibrio parahaemolyticus and identification of an iron-repressible hemin-binding protein
ELSMER
FEMS Microbiology
Letters 128 (1995) 195-200
Utilization of hemin and hemoglobin as iron sources by Vibrio parahaemolyticus and identification of an iron-repressible hemin-binding protein Shigeo Yamamoto
*, Yoshihiro Hara, Ken-ichi Tomochika,
Sumino Shinoda
Faculty of Pharmaceutical Sciences, Okayama University, Tsushima-naka I-I-1, Okayama 700, Japan Received 19 January
1995; revised 7 March 1995; accepted 20 March 1995
Abstract Several clinical isolates of Vibrio parahaemolyticus were examined for their ability to utilize either hemin or hemoglobin as a sole source of iron. Both compounds appeared to be equally good iron sources. Maximum growth was obtained at 5 PM hemin or 1.25 PM hemoglobin under the conditions tested. Using a hemin-agarose batch affinity method, the hemin-binding protein was isolated from crude total membranes of a hemin-utilizing strain, WPl, grown under iron-deficient but not under iron-sufficient conditions. This protein was identical to the 83 kDa outer membrane protein which was expressed in response to iron limitation. The protein was susceptible to proteinase K cleavage in whole cells, indicating its exposure at the cell surface. Hemin and hemoglobin, but not protoporphyrin IX, inhibited binding of the protein to hemin-agarose.
Keywords: Hemin utilization; Iron source; Hemin-binding
protein; Vibrio paruhaemolyticus
1. Introduction The ability of bacterial pathogens to effectively acquire iron within a mammalian host is an essential factor which influences their pathogenesis. Most successful pathogens quickly adapt to conditions of poor iron availability in the host through the expression of a variety of iron acquisition systems [l]. The best studied system is mediated by high-affinity iron chelators (siderophores) and cell-surface receptor proteins specific for iron-siderophore complexes. Other alternative mechanisms of iron acquisition, such as utilization of heme compounds, are also important for a number of pathogenic bacteria [1,2].
* Corresponding
author. Fax: ( + 81) 86 255 7456.
037%1097/95/$09.50 0 1995 Federation KSDI 0378-1097(95)00112-3
of European
Microbiological
Vibrio parahaemolyticus is an enteropathogenic estuarine bacterium which is often isolated as the causative agent of seafood-borne gastroenteritis. The thermostable direct hemolysin and its related hemolysin, both of which are also cytotoxic, have already been characterized as major virulence factors [3,4]. Recently, the hemolysin production by V. parahaemolyticus was found to be enhanced under conditions of iron limitation [51, suggesting one possibility that the effect of the hemolysins, such as cytotoxicity to intestinal epithelial cells with a concomitant lysis of erythrocytes, is likely to provide an additional source of iron. In addition, it has been shown that Vibrio species such as V. cholerae, V. vulnificus and V. anguillarum can effectively utilize hemin (Hm) and hemoglobin (Hb) for their growth Societies.
All rights reserved
196
S. Yamamotoet al./ FEMS MicrobiologyLetters 128 (1995) 195-200
in a siderophore-independent manner [2,6]. However, little is known about the utilization of such heme compounds by V. parahaemolyticus, although the iron acquisition system mediated by a polycarboxylate siderophore, vibrioferrin, has been reported [7,8]. In this study, the utilization of Hm and Hb was determined in several clinical isolates of V. parahaemolyticus and then the iron-repressible outer membrane protein of 83 kDa was identified as the hemin-binding protein (HmBP) which is probably involved in the initial binding in the acquisition of Hm-iron.
2. Materials and methods
2.1. Bacterial strains and chemicals V. parahaemolyticus WPl and AQ 3354, and type strain ATCC 17802, each producing the thermostable direct hemolysin and vibrioferrin, were obtained from the Research Institute for Microbial Diseases, Osaka University and the American Type Culture Collection, respectively. Strain MY-l was previously isolated from AQ 3354 as a spontaneously arising, vibrioferrin-deficient mutant [8]. Hm, Hb and protoporphyrin IX were purchased from Sigma and were solubilized as described [9]. All reagent solutions were filter sterilized. Glassware used was immersed in 6 M HCl overnight and rinsed several times with glass-distilled water prior to use, and all reagents and media were prepared with glass-distilled water. Hm-agarose with two-carbon atom spacer arms (1.6 pmol Hm ml-’ ) was prepared in our laboratory according to the procedure of Tsutsui [lo]. 2.2. Growth assays The utilization of Hm and Hb was tested in a liquid medium as follows. Late-log phase cells of the 1/ parahaemolyticus strains precultured in the irondeficient succinate minimal medium [8] were each diluted into 20 ml of the same medium containing Hm or Hb at an indicated concentration to an A,,, of 0.02. Cultures were shaken at 37” C, and cell density was measured at various time intervals. In
every experiment, growth curves in the presence and absence of 5 /.LM FeCl, were included as positive and negative controls, respectively. All experiments were performed in triplicate. 2.3. Membrane preparation Crude total membranes were prepared from cells grown in the minimal succinate medium under irondeficient (0.2 PM FeCl,) and iron-sufficient (20 PM FeCl,) conditions. Harvested cells were suspended in 50 mM Tris-HCl, pH 7.4, containing 10 mM benzamidine [8] and sonicated intermittently. Debris was removed by centrifugation at 4000 X g for 5 min and the crude total membranes were collected by additional centrifugation at 40000 X g for 30 min. Sodium N-lauroylsarcosinate (Sarkosyl, Sigma) cl%, w/ v )- insoluble outer membranes were isolated from crude total membranes by the procedure of Filip et al. ill]. 2.4. Ajjinity isolation of HmBPs HmBPs were isolated from crude total membranes by a batch affinity method using Hm-agarose essentially according to the procedure of Lee [9] except that Hm-agarose with two-carbon spacer arms was used. For each batch, 20-30 mg of protein was incubated with 1 ml of Hm-agarose. Bound proteins were finally eluted with SDS-PAGE sample loading buffer (0.5 M Tris-HCl, pH 6.8, containing 2% SDS, 0.05% bromphenol blue and 30% glycerol), and the resultant solution was concentrated with a Centricon10 (Amicon) at 4” C. 2.5. Exposure of outer membrane proteins in whole
cells to proteinase K Whole cells expressing the iron-repressible outer membrane proteins were suspended to ASh0 of 1.0 in 10 mM Tris-HCl, pH 8.0, containing 2% NaCl, 10 mM MgCl, and 0.5 mg ml-’ proteinase K (15-22 units mg-‘) and incubated at 37“ C for 1 h. After centrifugation at 4” C, the treated cells were extensively washed with the above-mentioned buffer (without proteinase K) containing 1 mM phenylmethylsulfonyl fluoride, and then subjected to sonication to obtain their crude total membranes.
S. Yamamoto et al. / FEMS Microbiology
2.6. SDS-PAGE
and protein determination
107
Letters 128 (199.5) 195-200
0.2
5
20
MW
Discontinuous SDS-PAGE was performed with 4% stacking gels and 8 or 10% separating gels. Each preparation (20-30 pg of total protein) was mixed with an equal volume of the sample loading buffer described above. Gels were stained with Coomassie blue. The protein concentration was determined by the Lowry method.
3. Results and discussion Since acquisition of iron from Hm or Hb by V. parahaemolyticus has not been reported formally, growth assays were performed for some strains in the presence of Hm or Hb as a sole iron source in the iron-deficient succinate medium. Both compounds appeared to be equally good iron sources for strain WPl, but the growth rate was less than that using inorganic iron as a positive control (Fig. 1). Maximum growth of the bacterium was obtained at 5 /_LM Hm or 1.25 PM Hb (iron content equivalent to Hm) under the conditions tested. Similar optimum concentrations for Hm and Hb have been reported for Plesiomonas shigelloides [ 121. However, no growth was supported by bovine cytochrome c in which the Hm moiety is covalently attached to the protein.
10 Time(h) Fig. 1. Growth of V. parahaemolyticus WPl in the presence of hemin or hemoglobin as the sole source of iron. Cells were grown in succinate minimal medium (01, succinate minimal medium with 5 PM FeCI, (0) and succinate minimal medium supplemented with 5 /.LM hemin (0) or 1.25 PM hemoglobin (A ). Growth was monitored by measuring the A,,,.
Fig. 2. Expression of outer membrane proteins by V. parahaemolyticus WPl as a function of iron concentration. Sarkosylinsoluble outer membrane fractions were prepared from cells grown at indicated iron concentrations (in PM at the top of each lane) and analyzed by SDS-PAGE. MW, molecular size markers (kDa).
These experiments were performed with other strains AQ 3354 and ATCC 17802 with similar results (data not shown). Furthermore, the siderophore-deficient mutant, MY-l, could grow at almost the same rate as the parental strain, AQ 3354, in the presence of 5 PM Hm or 1.25 PM Hb (data not shown), suggesting that utilization of these compounds as iron sources is independent of a siderophore-mediated iron uptake system. The OMP profiles of strain WPl grown to a late-log phase at different concentrations of FeCl, were compared by SDS-PAGE. Fig. 2 shows that WPl grown under iron-deficient conditions (0.2 ,uM added FeCl,) coordinately expressed two outer membrane proteins with apparent molecular masses of 78 and 83 kDa, which were lost at higher concentrations of added FeCl, (5 and 20 PM). In contrast, Hm at 5 I_LMstill induced expression of these ironrepressible outer membrane proteins, implying that when Hm is the only iron source, growth is more iron-limited for V. parahaemolyticus. Similarly, strains AQ 3354 and ATCC 17802 expressed two outer membrane proteins with the same molecular
S. Yamamoto et al. / FEMS Microbiology Letters 128 (1995) 195-200
198
masses during growth under iron-deficient conditions (data not shown). The crude total membranes prepared from V. parahaemolyticus WPl cells grown under iron-deficient (0.1 PM FeCl,) and iron-sufficient (20 PM FeCl,) conditions were subjected to the Hm-agarose affinity protocol to assess the presence of the HmBP(s) by SDS-PAGE. The 83 kDa protein seen in crude total membranes from iron-deficient cells was affinity isolated (Fig. 3, lanes 3 and 4). The protein of 50 kDa was affinity isolated from both the total membrane preparations (Fig. 3, lanes 2 and 41, at least suggesting that its expression was independent of the iron concentration in the growth medium (see below). The 83 kDa protein was absent when the agarose matrix alone, without bound Hm, was used instead of Hm-agarose. SDS-PAGE analysis demonstrated that the 83 kDa protein isolated by the affinity method comigrated with the 83 kDa protein present in a Sarkosyl-insoluble outer membrane fraction from iron-deficient cells (data not shown). Fur-
205. 115 97.e 58.
-03 -50
Fig. 3. SDS-PAGE of the HmBPs isolated from crude total membranes of the V. parahaemolyticus WPl cells by heminagarose batch affinity method. MW, molecular size markers (kDa); (+), cells grown under iron-sufficient conditions; ( - ), cells grown under iron-deficient conditions. Lanes 1 and 3, crude total membranes; lanes 2 and 4, HmBPs isolated by hemin-agarose batch affinity method. MW, molecular size markers (kDa).
7 2 / iv I kDa
^;
>’
Fig. 4. Susceptibility of the iron-repressible outer membrane proteins to proteinase K. Whole cells of strain WPl were incubated at 37” C for 1 h in the absence (lane 1) and presence (lane 2) of proteinase K (0.5 mg ml-’ ) prior to the preparation of crude total membranes.
thermore, an indication of the cell-surface exposure of this protein was obtained by incubation of whole cells with proteinase K prior to the preparation of crude total membranes. Both of the 78 and 83 kDa outer membrane proteins were susceptible to proteolytic cleavage (Fig. 4). The proteolytic digestion also resulted in the appearance of additional protein bands. Consistent with this result, the 83 kDa protein was not recovered from whole cells pretreated with proteinase K by the Hm-agarose affinity method. Solubilization of crude total membranes with SDS or extraction with Sarkosyl to exclude inner membrane prior to affinity isolation almost completely abolished isolation of the 83 kDa protein. This may be due to subsequent conformational changes in the HmBP. Hm-agarose with two-carbon spacer arms appeared to be more effective and specific in isolating the 83 kDa protein than that with longer spacer arms (Sigma, 12-atoms, H 6390) that was used at the initial stage of this study. To demonstrate the specificity of binding of the 83 kDa protein to Hm-agarose, preincubation of total membranes from iron-deficient cells with Hm, Hb or protoporphyrin IX was carried out prior to affinity isolation. The results are shown in Fig. 5. Binding of
S. Yamamoto et al. / FEMS Microbiology
-83 -50
Fig. 5. Binding specificity of the HmBPs to hemin-agarose. Total membranes (20 mg of protein) prepared from the V. parahaemolyticus WPl cells grown under iron-deficient conditions were preincubated at 37” C for 1 h with hemin (Hm), hemoglobin (Hb) or protoporphyrin IX (PP) at a final concentration of 100 PM and subjected to the hemin-agarose affinity protocol.
Letters 128 (199.5) 195-200
199
haemolyticus. The recent genetic demonstrations that Hm uptake in Yersinia enterocolitica [13] and V. cholerae [14] requires an Hm-specific iron-regulated receptor support this contention. However, it still remains unclear whether these Hm receptor proteins as well as the HmBP identified in this study also serve as the receptor for Hb. Proteolytic action by V. parahaemolyticus on Hb and/or spontaneous release of Hm from Hb [15] may be, at least in part, in order to secure Hm. Recently, we found that a second, iron-repressible outer membrane protein, the 78 kDa protein, functions as the ferric vibrioferrin receptor (unpublished data). Vibrioferrin can sequester iron from human transferrin (30% iron saturation) to support in vitro growth [8]. Therefore, subsequent genetic and virulence studies of these two iron acquisition systems may allow us to define the importance of iron uptake to the physiology of V. parahaemolyticus or the pathogenic process of V. parahaemolyticus infection.
Acknowledgements the protein with Hm-agarose was specific, since its retention was completely abolished by the preincubation with 100 FM Hm, causing saturation of the Hm-binding site in the 83 kDa protein with Hm. Binding was almost inhibited by Hb at the same concentration as Hm. In contrast, protoporphyrin IX (100 PM) was incapable of inhibiting binding, indicating that the presence of iron in the tetrapyrrole ring is a prerequisite for recognition by the 83 kDa HmBP. In spite of its binding at high affinity to Hm-agarose as shown in Fig. 3, the 50-kDa protein, unlike the 83 kDa protein, was virtually present even after preincubation with Hm. This may be due to a high level of nonspecific binding. The HmBP identical in molecular mass to that found in strain WPl was also isolated from two other strains AQ 3354 and ATCC 17802 grown under conditions of iron starvation (data not shown), suggesting that this protein may be shared by members of this species for a similar function. The identification of a cell-surface exposed, ironregulated HmBP in this study is consistent with the supposition that Hm-iron utilization may proceed via such a receptor-mediated process in V. para-
This work was supported by a General Grant-inAid for Scientific Research from the Ministry of Education, Science, and Culture of Japan.
References HI Wooldridge,
K.G. and Williams, P.H. (1993) Iron uptake mechanisms of pathogenic bacteria. FEMS Microbial. Rev. 12, 325-348. I21 Otto, B.R., Verweij-van Vught, A.M.J.J. and MacLaren, D.M. (1992) Transferrin and heme-compounds as iron sources for pathogenic bacteria. Crit. Rev. Microbial. 18, 217-233. [31Nishibuchi, M., Fasano, A., Russell, R.G. and Kaper, J.B. (1992) Enterotoxigenicity of Vibrio parahaemolyticus with and without genes encoding thermostable direct hemolysin. Infect. Immun. 60, 3539-3545. [41Honda, T., Ni, Y. and Miwatani, T. (1988) Purification and characterization of a hemolysin produced by a clinical isolate of Kanagawa phenomenon-negative Vibrio parahaemolyticus and related to the thermostable direct hemolysin. Infect. Immun. 56, 961-965. t51 Dai, J.-H., Lee, Y.-S. and Wong, H.-C. (1992) Effects of iron limitation on production of a siderophore, outer membrane proteins, and hemolysin and on hydrophobicity, cell adherence, and lethality for mice of V. parahaemolyticus. Infect, Immun. 60, 2952-2956.
200
S. Yamamoto et al. / FEMS Microbiology Letters I28 (1995) 195-200
[6] Mazoy, R. and Lemos, M.L. (1991) Iron-binding proteins and heme compounds as iron sources for Vibrio anguillarum. Curr. Microbial. 23, 221-226. [7] Yamamoto, S., Okujo, N., Yoshida, T., Matsuura, S. and Shinoda, S. (1994) Structure and iron transport activity of vibrioferrin, a new siderophore of Vibrio parahaemolyticus. J. Biochem. 115, 868-874. [8] Yamamoto, S., Okujo, N., Matsuura, S., Fujiwara, I., Fujita, Y. and Shinoda, S. (1994) Siderophore-mediated utilization of iron bound to transferrin by Vibrio parahaemolyticus. Microbial. Immunol. 38, 687-693. [9] Lee, B.C. (1992) Isolation of an outer membrane hemin-bind ing protein of Haemophilus influenzae type b. Infect. Immun. 60, 810-816. [lo] Tsutsui, K. (1986) Affinity chromatography of heme-binding proteins: synthesis of hemin-agarose. Methods Enzymol. 123, 331-338. [ll] Filip, C., Fletcher, G., Wufll, J.L. and Earhart, CF. (1973)
[12]
[13]
[14]
[15]
Solubilization of the cytoplasmic membrane of Escherichia coli by the ionic detergent sodium lauroyl sarcosinate. J. Bacterial. 115, 717-722. Daskaleros, P.A., Stoebner, J.A. and Payne, SM. (1991) Iron uptake in Plesiomonas shigelloides: cloning of the genes for the heme-iron uptake system. Infect. Immun. 59, 2706-2711. Stojilikovic, I. and Hantke, K. (1992) Hemin uptake system of Yersinia enrerocolitica: similarities with other TonB-dependent systems in gram-negative bacteria. EMBO J. 11, 4359-4367. Henderson, D.P. and Payne, S.M. (1993) Cloning and characterization of the Vibrio cholerae genes encoding the utilization of iron from hemin and hemoglobin. Mol. Microbial. 7, 461-469. Smith, M.L., Paul, J., Ohlsson, P.I., Hjortsberg, K. and Paul, K.G. (1991) Heme-protein fission under nondenaturing conditions. Proc. Natl. Acad. Sci. USA 88, 882-886.
FEMS Microbiology
Letters 128 (1995) 195-200
Utilization of hemin and hemoglobin as iron sources by Vibrio parahaemolyticus and identification of an iron-repressible hemin-binding protein Shigeo Yamamoto
*, Yoshihiro Hara, Ken-ichi Tomochika,
Sumino Shinoda
Faculty of Pharmaceutical Sciences, Okayama University, Tsushima-naka I-I-1, Okayama 700, Japan Received 19 January
1995; revised 7 March 1995; accepted 20 March 1995
Abstract Several clinical isolates of Vibrio parahaemolyticus were examined for their ability to utilize either hemin or hemoglobin as a sole source of iron. Both compounds appeared to be equally good iron sources. Maximum growth was obtained at 5 PM hemin or 1.25 PM hemoglobin under the conditions tested. Using a hemin-agarose batch affinity method, the hemin-binding protein was isolated from crude total membranes of a hemin-utilizing strain, WPl, grown under iron-deficient but not under iron-sufficient conditions. This protein was identical to the 83 kDa outer membrane protein which was expressed in response to iron limitation. The protein was susceptible to proteinase K cleavage in whole cells, indicating its exposure at the cell surface. Hemin and hemoglobin, but not protoporphyrin IX, inhibited binding of the protein to hemin-agarose.
Keywords: Hemin utilization; Iron source; Hemin-binding
protein; Vibrio paruhaemolyticus
1. Introduction The ability of bacterial pathogens to effectively acquire iron within a mammalian host is an essential factor which influences their pathogenesis. Most successful pathogens quickly adapt to conditions of poor iron availability in the host through the expression of a variety of iron acquisition systems [l]. The best studied system is mediated by high-affinity iron chelators (siderophores) and cell-surface receptor proteins specific for iron-siderophore complexes. Other alternative mechanisms of iron acquisition, such as utilization of heme compounds, are also important for a number of pathogenic bacteria [1,2].
* Corresponding
author. Fax: ( + 81) 86 255 7456.
037%1097/95/$09.50 0 1995 Federation KSDI 0378-1097(95)00112-3
of European
Microbiological
Vibrio parahaemolyticus is an enteropathogenic estuarine bacterium which is often isolated as the causative agent of seafood-borne gastroenteritis. The thermostable direct hemolysin and its related hemolysin, both of which are also cytotoxic, have already been characterized as major virulence factors [3,4]. Recently, the hemolysin production by V. parahaemolyticus was found to be enhanced under conditions of iron limitation [51, suggesting one possibility that the effect of the hemolysins, such as cytotoxicity to intestinal epithelial cells with a concomitant lysis of erythrocytes, is likely to provide an additional source of iron. In addition, it has been shown that Vibrio species such as V. cholerae, V. vulnificus and V. anguillarum can effectively utilize hemin (Hm) and hemoglobin (Hb) for their growth Societies.
All rights reserved
196
S. Yamamotoet al./ FEMS MicrobiologyLetters 128 (1995) 195-200
in a siderophore-independent manner [2,6]. However, little is known about the utilization of such heme compounds by V. parahaemolyticus, although the iron acquisition system mediated by a polycarboxylate siderophore, vibrioferrin, has been reported [7,8]. In this study, the utilization of Hm and Hb was determined in several clinical isolates of V. parahaemolyticus and then the iron-repressible outer membrane protein of 83 kDa was identified as the hemin-binding protein (HmBP) which is probably involved in the initial binding in the acquisition of Hm-iron.
2. Materials and methods
2.1. Bacterial strains and chemicals V. parahaemolyticus WPl and AQ 3354, and type strain ATCC 17802, each producing the thermostable direct hemolysin and vibrioferrin, were obtained from the Research Institute for Microbial Diseases, Osaka University and the American Type Culture Collection, respectively. Strain MY-l was previously isolated from AQ 3354 as a spontaneously arising, vibrioferrin-deficient mutant [8]. Hm, Hb and protoporphyrin IX were purchased from Sigma and were solubilized as described [9]. All reagent solutions were filter sterilized. Glassware used was immersed in 6 M HCl overnight and rinsed several times with glass-distilled water prior to use, and all reagents and media were prepared with glass-distilled water. Hm-agarose with two-carbon atom spacer arms (1.6 pmol Hm ml-’ ) was prepared in our laboratory according to the procedure of Tsutsui [lo]. 2.2. Growth assays The utilization of Hm and Hb was tested in a liquid medium as follows. Late-log phase cells of the 1/ parahaemolyticus strains precultured in the irondeficient succinate minimal medium [8] were each diluted into 20 ml of the same medium containing Hm or Hb at an indicated concentration to an A,,, of 0.02. Cultures were shaken at 37” C, and cell density was measured at various time intervals. In
every experiment, growth curves in the presence and absence of 5 /.LM FeCl, were included as positive and negative controls, respectively. All experiments were performed in triplicate. 2.3. Membrane preparation Crude total membranes were prepared from cells grown in the minimal succinate medium under irondeficient (0.2 PM FeCl,) and iron-sufficient (20 PM FeCl,) conditions. Harvested cells were suspended in 50 mM Tris-HCl, pH 7.4, containing 10 mM benzamidine [8] and sonicated intermittently. Debris was removed by centrifugation at 4000 X g for 5 min and the crude total membranes were collected by additional centrifugation at 40000 X g for 30 min. Sodium N-lauroylsarcosinate (Sarkosyl, Sigma) cl%, w/ v )- insoluble outer membranes were isolated from crude total membranes by the procedure of Filip et al. ill]. 2.4. Ajjinity isolation of HmBPs HmBPs were isolated from crude total membranes by a batch affinity method using Hm-agarose essentially according to the procedure of Lee [9] except that Hm-agarose with two-carbon spacer arms was used. For each batch, 20-30 mg of protein was incubated with 1 ml of Hm-agarose. Bound proteins were finally eluted with SDS-PAGE sample loading buffer (0.5 M Tris-HCl, pH 6.8, containing 2% SDS, 0.05% bromphenol blue and 30% glycerol), and the resultant solution was concentrated with a Centricon10 (Amicon) at 4” C. 2.5. Exposure of outer membrane proteins in whole
cells to proteinase K Whole cells expressing the iron-repressible outer membrane proteins were suspended to ASh0 of 1.0 in 10 mM Tris-HCl, pH 8.0, containing 2% NaCl, 10 mM MgCl, and 0.5 mg ml-’ proteinase K (15-22 units mg-‘) and incubated at 37“ C for 1 h. After centrifugation at 4” C, the treated cells were extensively washed with the above-mentioned buffer (without proteinase K) containing 1 mM phenylmethylsulfonyl fluoride, and then subjected to sonication to obtain their crude total membranes.
S. Yamamoto et al. / FEMS Microbiology
2.6. SDS-PAGE
and protein determination
107
Letters 128 (199.5) 195-200
0.2
5
20
MW
Discontinuous SDS-PAGE was performed with 4% stacking gels and 8 or 10% separating gels. Each preparation (20-30 pg of total protein) was mixed with an equal volume of the sample loading buffer described above. Gels were stained with Coomassie blue. The protein concentration was determined by the Lowry method.
3. Results and discussion Since acquisition of iron from Hm or Hb by V. parahaemolyticus has not been reported formally, growth assays were performed for some strains in the presence of Hm or Hb as a sole iron source in the iron-deficient succinate medium. Both compounds appeared to be equally good iron sources for strain WPl, but the growth rate was less than that using inorganic iron as a positive control (Fig. 1). Maximum growth of the bacterium was obtained at 5 /_LM Hm or 1.25 PM Hb (iron content equivalent to Hm) under the conditions tested. Similar optimum concentrations for Hm and Hb have been reported for Plesiomonas shigelloides [ 121. However, no growth was supported by bovine cytochrome c in which the Hm moiety is covalently attached to the protein.
10 Time(h) Fig. 1. Growth of V. parahaemolyticus WPl in the presence of hemin or hemoglobin as the sole source of iron. Cells were grown in succinate minimal medium (01, succinate minimal medium with 5 PM FeCI, (0) and succinate minimal medium supplemented with 5 /.LM hemin (0) or 1.25 PM hemoglobin (A ). Growth was monitored by measuring the A,,,.
Fig. 2. Expression of outer membrane proteins by V. parahaemolyticus WPl as a function of iron concentration. Sarkosylinsoluble outer membrane fractions were prepared from cells grown at indicated iron concentrations (in PM at the top of each lane) and analyzed by SDS-PAGE. MW, molecular size markers (kDa).
These experiments were performed with other strains AQ 3354 and ATCC 17802 with similar results (data not shown). Furthermore, the siderophore-deficient mutant, MY-l, could grow at almost the same rate as the parental strain, AQ 3354, in the presence of 5 PM Hm or 1.25 PM Hb (data not shown), suggesting that utilization of these compounds as iron sources is independent of a siderophore-mediated iron uptake system. The OMP profiles of strain WPl grown to a late-log phase at different concentrations of FeCl, were compared by SDS-PAGE. Fig. 2 shows that WPl grown under iron-deficient conditions (0.2 ,uM added FeCl,) coordinately expressed two outer membrane proteins with apparent molecular masses of 78 and 83 kDa, which were lost at higher concentrations of added FeCl, (5 and 20 PM). In contrast, Hm at 5 I_LMstill induced expression of these ironrepressible outer membrane proteins, implying that when Hm is the only iron source, growth is more iron-limited for V. parahaemolyticus. Similarly, strains AQ 3354 and ATCC 17802 expressed two outer membrane proteins with the same molecular
S. Yamamoto et al. / FEMS Microbiology Letters 128 (1995) 195-200
198
masses during growth under iron-deficient conditions (data not shown). The crude total membranes prepared from V. parahaemolyticus WPl cells grown under iron-deficient (0.1 PM FeCl,) and iron-sufficient (20 PM FeCl,) conditions were subjected to the Hm-agarose affinity protocol to assess the presence of the HmBP(s) by SDS-PAGE. The 83 kDa protein seen in crude total membranes from iron-deficient cells was affinity isolated (Fig. 3, lanes 3 and 4). The protein of 50 kDa was affinity isolated from both the total membrane preparations (Fig. 3, lanes 2 and 41, at least suggesting that its expression was independent of the iron concentration in the growth medium (see below). The 83 kDa protein was absent when the agarose matrix alone, without bound Hm, was used instead of Hm-agarose. SDS-PAGE analysis demonstrated that the 83 kDa protein isolated by the affinity method comigrated with the 83 kDa protein present in a Sarkosyl-insoluble outer membrane fraction from iron-deficient cells (data not shown). Fur-
205. 115 97.e 58.
-03 -50
Fig. 3. SDS-PAGE of the HmBPs isolated from crude total membranes of the V. parahaemolyticus WPl cells by heminagarose batch affinity method. MW, molecular size markers (kDa); (+), cells grown under iron-sufficient conditions; ( - ), cells grown under iron-deficient conditions. Lanes 1 and 3, crude total membranes; lanes 2 and 4, HmBPs isolated by hemin-agarose batch affinity method. MW, molecular size markers (kDa).
7 2 / iv I kDa
^;
>’
Fig. 4. Susceptibility of the iron-repressible outer membrane proteins to proteinase K. Whole cells of strain WPl were incubated at 37” C for 1 h in the absence (lane 1) and presence (lane 2) of proteinase K (0.5 mg ml-’ ) prior to the preparation of crude total membranes.
thermore, an indication of the cell-surface exposure of this protein was obtained by incubation of whole cells with proteinase K prior to the preparation of crude total membranes. Both of the 78 and 83 kDa outer membrane proteins were susceptible to proteolytic cleavage (Fig. 4). The proteolytic digestion also resulted in the appearance of additional protein bands. Consistent with this result, the 83 kDa protein was not recovered from whole cells pretreated with proteinase K by the Hm-agarose affinity method. Solubilization of crude total membranes with SDS or extraction with Sarkosyl to exclude inner membrane prior to affinity isolation almost completely abolished isolation of the 83 kDa protein. This may be due to subsequent conformational changes in the HmBP. Hm-agarose with two-carbon spacer arms appeared to be more effective and specific in isolating the 83 kDa protein than that with longer spacer arms (Sigma, 12-atoms, H 6390) that was used at the initial stage of this study. To demonstrate the specificity of binding of the 83 kDa protein to Hm-agarose, preincubation of total membranes from iron-deficient cells with Hm, Hb or protoporphyrin IX was carried out prior to affinity isolation. The results are shown in Fig. 5. Binding of
S. Yamamoto et al. / FEMS Microbiology
-83 -50
Fig. 5. Binding specificity of the HmBPs to hemin-agarose. Total membranes (20 mg of protein) prepared from the V. parahaemolyticus WPl cells grown under iron-deficient conditions were preincubated at 37” C for 1 h with hemin (Hm), hemoglobin (Hb) or protoporphyrin IX (PP) at a final concentration of 100 PM and subjected to the hemin-agarose affinity protocol.
Letters 128 (199.5) 195-200
199
haemolyticus. The recent genetic demonstrations that Hm uptake in Yersinia enterocolitica [13] and V. cholerae [14] requires an Hm-specific iron-regulated receptor support this contention. However, it still remains unclear whether these Hm receptor proteins as well as the HmBP identified in this study also serve as the receptor for Hb. Proteolytic action by V. parahaemolyticus on Hb and/or spontaneous release of Hm from Hb [15] may be, at least in part, in order to secure Hm. Recently, we found that a second, iron-repressible outer membrane protein, the 78 kDa protein, functions as the ferric vibrioferrin receptor (unpublished data). Vibrioferrin can sequester iron from human transferrin (30% iron saturation) to support in vitro growth [8]. Therefore, subsequent genetic and virulence studies of these two iron acquisition systems may allow us to define the importance of iron uptake to the physiology of V. parahaemolyticus or the pathogenic process of V. parahaemolyticus infection.
Acknowledgements the protein with Hm-agarose was specific, since its retention was completely abolished by the preincubation with 100 FM Hm, causing saturation of the Hm-binding site in the 83 kDa protein with Hm. Binding was almost inhibited by Hb at the same concentration as Hm. In contrast, protoporphyrin IX (100 PM) was incapable of inhibiting binding, indicating that the presence of iron in the tetrapyrrole ring is a prerequisite for recognition by the 83 kDa HmBP. In spite of its binding at high affinity to Hm-agarose as shown in Fig. 3, the 50-kDa protein, unlike the 83 kDa protein, was virtually present even after preincubation with Hm. This may be due to a high level of nonspecific binding. The HmBP identical in molecular mass to that found in strain WPl was also isolated from two other strains AQ 3354 and ATCC 17802 grown under conditions of iron starvation (data not shown), suggesting that this protein may be shared by members of this species for a similar function. The identification of a cell-surface exposed, ironregulated HmBP in this study is consistent with the supposition that Hm-iron utilization may proceed via such a receptor-mediated process in V. para-
This work was supported by a General Grant-inAid for Scientific Research from the Ministry of Education, Science, and Culture of Japan.
References HI Wooldridge,
K.G. and Williams, P.H. (1993) Iron uptake mechanisms of pathogenic bacteria. FEMS Microbial. Rev. 12, 325-348. I21 Otto, B.R., Verweij-van Vught, A.M.J.J. and MacLaren, D.M. (1992) Transferrin and heme-compounds as iron sources for pathogenic bacteria. Crit. Rev. Microbial. 18, 217-233. [31Nishibuchi, M., Fasano, A., Russell, R.G. and Kaper, J.B. (1992) Enterotoxigenicity of Vibrio parahaemolyticus with and without genes encoding thermostable direct hemolysin. Infect. Immun. 60, 3539-3545. [41Honda, T., Ni, Y. and Miwatani, T. (1988) Purification and characterization of a hemolysin produced by a clinical isolate of Kanagawa phenomenon-negative Vibrio parahaemolyticus and related to the thermostable direct hemolysin. Infect. Immun. 56, 961-965. t51 Dai, J.-H., Lee, Y.-S. and Wong, H.-C. (1992) Effects of iron limitation on production of a siderophore, outer membrane proteins, and hemolysin and on hydrophobicity, cell adherence, and lethality for mice of V. parahaemolyticus. Infect, Immun. 60, 2952-2956.
200
S. Yamamoto et al. / FEMS Microbiology Letters I28 (1995) 195-200
[6] Mazoy, R. and Lemos, M.L. (1991) Iron-binding proteins and heme compounds as iron sources for Vibrio anguillarum. Curr. Microbial. 23, 221-226. [7] Yamamoto, S., Okujo, N., Yoshida, T., Matsuura, S. and Shinoda, S. (1994) Structure and iron transport activity of vibrioferrin, a new siderophore of Vibrio parahaemolyticus. J. Biochem. 115, 868-874. [8] Yamamoto, S., Okujo, N., Matsuura, S., Fujiwara, I., Fujita, Y. and Shinoda, S. (1994) Siderophore-mediated utilization of iron bound to transferrin by Vibrio parahaemolyticus. Microbial. Immunol. 38, 687-693. [9] Lee, B.C. (1992) Isolation of an outer membrane hemin-bind ing protein of Haemophilus influenzae type b. Infect. Immun. 60, 810-816. [lo] Tsutsui, K. (1986) Affinity chromatography of heme-binding proteins: synthesis of hemin-agarose. Methods Enzymol. 123, 331-338. [ll] Filip, C., Fletcher, G., Wufll, J.L. and Earhart, CF. (1973)
[12]
[13]
[14]
[15]
Solubilization of the cytoplasmic membrane of Escherichia coli by the ionic detergent sodium lauroyl sarcosinate. J. Bacterial. 115, 717-722. Daskaleros, P.A., Stoebner, J.A. and Payne, SM. (1991) Iron uptake in Plesiomonas shigelloides: cloning of the genes for the heme-iron uptake system. Infect. Immun. 59, 2706-2711. Stojilikovic, I. and Hantke, K. (1992) Hemin uptake system of Yersinia enrerocolitica: similarities with other TonB-dependent systems in gram-negative bacteria. EMBO J. 11, 4359-4367. Henderson, D.P. and Payne, S.M. (1993) Cloning and characterization of the Vibrio cholerae genes encoding the utilization of iron from hemin and hemoglobin. Mol. Microbial. 7, 461-469. Smith, M.L., Paul, J., Ohlsson, P.I., Hjortsberg, K. and Paul, K.G. (1991) Heme-protein fission under nondenaturing conditions. Proc. Natl. Acad. Sci. USA 88, 882-886.