Yamada-kami, Suita, Osaka 565, Japan. type 3) was cultured without shaking in Todd-Hewitt. giloo ml), and the resulting precipitate was applied

Transcription

1 INFECTION AND IMMUNITY, Aug. 977, p Copyright 977 American Society for Microbiology Vol. 7, No. 2 Printed in U.S.A. Effect of Zinc Ion on the Hemolytic Activity of Thermostable Direct Hemolysin from Vibrio parahaemolyticus, Streptolysin, and Triton X- YOSHIFUMI TAKEDA,* YUKO OGISO, AND TOSHIO MIWATANI Department of Bacteriology and Serology, Research Institute for Microbial Diseases, Osaka University, Yamada-kami, Suita, Osaka 6, Japan Received for publication 3 March 977 The effect of Zn2+ on hemolysis induced by the thermostable direct hemolysin from Vibrio parahaemolyticus (vibriolysin), streptolysin, and Triton X- was studied. We found that in certain buffers, such as tris(hydroxymethyl)- aminomethane-hydrochloride, boric acid-borax, and N-hydroxyethyl piperazine-n'-2-ethanesulfonic acid-sodium hydroxide, hemoglobins released from erythrocytes were easily precipitated by addition of Zn2+, thus resulting in a false inhibition of hemolysin by Zn2+ when hemolysis was assayed by measuring absorbance at 4 nm of released hemoglobins. Under experimental conditions in which hemoglobin was not precipitated, hemolysis induced by streptolysin was inhibited by Zn2+, whereas that induced by vibriolysin and Triton X- was not. Thus, we concluded that the mode of inhibitory action of Zn2+ on hemolysis was not due to a reversible alteration in the state of the lipid bilayer of erythrocytes membranes as proposed by Avigad and Bernheimer (Infect. Immun. 3:378-38, 976). Avigad and Bernheimer () recently reported that hemolysis induced by various bacterial toxins, such as streptolysin and aerolysin, and other lytic agents, such as Triton X- and saponin, was inhibited by Zn2+. They proposed that the inhibition by Zn2+ was due to a reversible alteration in the state of the lipid bilayer of erythrocytes (). During the course of our study to confirm the results of Avigad and Bernheimer (), in which we used purified thermostable direct hemolysin produced by Vibrioparahaemolyticus (vibriolysin), we found that the inhibitory effect of Zn2+ on hemolysis induced by bacterial toxins and other lytic agents was not universal as reported. In this paper we study the effect of zinc ion on two bacterial hemolysins and one nonbacterial hemolytic agent. MATERLALS AND METHODS Preparation of purified vibriolysin. Vibriolysin was isolated from culture filtrates of V. parahaemolyticus WP- (RIMD 2286) and purified by ammonium sulfate precipitation and successive column chromatographies on diethylaminoethylcellulose, hydroxylapatite, and Sephadex G-2; the detailed methods have been reported previously (4). The purified vibriolysin was demonstrated to be a single protein by sodium dodecyl sulfate-polyacrylamide gel disc electrophoresis, analytical ultracentrifugation, and immunoelectrophoresis. Preparation of partially purified streptolysin. Streptococcus pyogenes RIMD 3264 (group A, type 3) was cultured without shaking in Todd-Hewitt medium at 37 C for 6 h. The culture filtrate was precipitated by adding solid ammonium sulfate (39. giloo ml), and the resulting precipitate was applied to a hydroxylapatite column (2.2 by 3 cm) previously equilibrated with. M phosphate buffer (Na2HPO4-KH2PO4, ph 7.) containing 3 mg of cysteine hydrochloride per ml. After the column was washed with the above-mentioned buffer, it was eluted with a linear gradient of. to.3 M phosphate buffer containing 3 mg of cysteine hydrochloride per ml. Fractions containing hemolytic activity were used as partially purified streptolysin. The hemolytic activity ofthese fractions was lost at room temperature when cysteine hydrochloride was not added, and it was recovered after these fractions were incubated with cysteine hydrochloride at 37 C for 3 min. In this study, the amount of streptolysin that lyses % of human erythrocytes (about 8.3 x 7 cells/ml) in. ml after incubation at 37 C for 3 min was defined as hemolytic unit. Assay of hemolytic activity. Hemolytic activity was assayed essentially as described previously (7). The standard reaction mixture contained.9% NaCl, about 8.3 x 7 human erythrocytes per ml, and indicated amounts ofhemolytic agent in a specified buffer. The reaction mixture was incubated at 37 C for 3 min and then centrifuged at 3, rpm for min. The hemolytic activity was determined by measuring the absorbance of the resulting supernatant fluid at 4 nm. 239 Downloaded from on September 24, 28 by guest

2 24 TAKEDA, OGISO, AND MIWATANI Buffer. Tris(hydroxymethyl)aminomethane (Tris)-hydrochloride buffer (ph 7.2), phosphate buffer (ph 7.), glycine-sodium hydroxide buffer (ph 8.7), boric acid-borax buffer (ph 7.), acetate buffer (ph.), cacodylate buffer (ph.), and barbital buffer (ph 7.3) were prepared as described by Gomori (2). N-hydroxyethylpiperazine-N'-2-ethanesulfonic acid-sodium hydroxide (HEPES) buffer was prepared as described by Good et al. (3). RESULTS Inhibition of hemolysis by Zn2+ and Cu2+ induced by vibriolysin, streptolysin, and Triton X-. Table shows the effect of various divalent cations on hemolysis induced by vibriolysin, streptolysin, and Triton X-. The reactions were carried out in isotonic Trishydrochloride-buffered saline, ph 7.2. Confirming the results reported by Avigad and Bernheimer (), hemolysis by streptolysin and Triton X- was inhibited by Zn2+ and Cu2+. Hemolysis by vibriolysin was also inhibited by these cations. In addition, Hg2+ was inhibitory to hemolysis by these hemolytic agents. Ni2+, Co2+, and Cd2+ were slightly inhibitory to hemolysis by the bacterial hemolysins, but did not affect that by Triton X-. The effect of other divalent cations was rather complicated, some being stimulatory to some hemolytic agents and the others being inhibitory. Hemolysis by vibriolysin, streptolysin, and Triton X- in isotonic Tris-hydrochloridebuffered saline was completely inhibited by Zn2+ at concentrations of. mm or more (Fig. ). The concentration of Zn2+ inhibiting 9% of the hemolysis was around.2 to.3 mm, which confirned the results of Avigad and Bernheimer (). At lower Zn2+ concentrations the hemolysis was somewhat stimulated. Kinetics of inhibition of hemolysis by Zn2+. The results of Figure 2 show the kinetics of the inhibition of hemolysis by Zn2+ induced by vibriolysin, streptolysin, and Triton X- in isotonic Tris-hydrochloride-buffered saline buffer. Under the experimental conditions used, hemolysis by vibriolysin reached a plateau at about 4 to 6 min (Fig. 2A). When mm Zn2+ was added at either,, or 2 min after incubation was started, the absorbance at 4 nm immediately decreased to an almost negligible level. Further incubation did not result in any significant increases in absorbance at 4 nm. Similar results were obtained when streptolysin (Fig. 2B) and Triton X- (Fig. 2C) were used as the hemolytic agents. These results suggest that the inhibition of hemolysis by Zn2+ induced by these hemolytic agents is not a real inhibition of the hemolytic reaction. For comparison, inhibition by ethylenedia- TABLE. Effect of various divalent cations on hemolysis by vibriolysin, streptolysin, and Triton X- Hemolysis (A4)b Cationa Vibriolysin Streptolysin Triton X- None Zn2+ c..2 Cu Hg Ni C Cd Fe Mg Sr Ca Mn Cr aaallcatins cations wre aschloidexcep u2 andi were as chloride except CU2+ Fe2, which were sulfate. b Hemolysis was assayed in isotonic Tris-hydrochloride-buffered saline ( mm, ph 7.2) as described in the text. The amounts of hemolytic agents added in. ml of the reaction mixture were: vibriolysin, 3.7,ug; streptolysin, hemolytic unit; and Triton X-, 6 x -3%. Each value is an average of duplicate determinations. A4, Absorbance at 4 nm. c Millimolar concentration. INFECT. IMMUN. minetetraacetate of the hemolysis by vibriolysin and that byn-ethylmaleimide of the hemolysis by streptolysin are shown in Fig. 3A and B, respectively. These data represented the real inhibition of the hemolytic reactions. Effect of Zn2+ on osmotic lysis of erythrocytes. Avigad and Bernheimer () reported that Zn2+ does not inhibit osmotic lysis. This was Downloaded from on September 24, 28 by guest

3 VOL. 7, 977 confirmed as shown in Fig. 4A. Surprisingly, however, we found that osmotic lysis was inhibited by Zn2+ under certain conditions. Osmotic lysis in certain buffers, such as Trishydrochloride, boric acid-borax, HEPES-sodium hydroxide, and barbital, was inhibited by Zn2+ (Fig. 4). On the other hand, osmotic lysis in either phosphate buffer, glycine-hydroxide buffer, cacodylate buffer, or acetate buffer was not inhibited by Zn2. These results indicate that released hemoglobin could be precipitated only in certain media. This was further confirned by the results shown in Fig.. In this experiment, Zn2+ was rz Un Ez z 2 CA.... FIG.. Effect ofzn2+ on hemolysis by various hemolytic agents. Hemolysis was assayed in isotonic Tris-hydrochloride-buffered saline ( mm, ph 72) as described in the text. Zinc acetate was added as indicated. The amounts ofhemolytic agents added in. ml of reaction mixture were: vibriolysin, 3.7 pg (); streptolysin, hemolytic unit (O); and Triton X-, 6 x -3% (A). Each value is an average of duplicate determinations. C r. S z C INHIBITION OF HEMOLYSIS BY ZN2+ 24 added after erythrocytes were lysed in either distilled water, Tris-hydrochloride buffer, or phosphate buffer. We found that hemoglobins in Tris-hydrochloride buffer were precipitated by adding Zn2+, whereas those in either distilled water or phosphate buffer were not.. - (A). o TIME (MIN) TIME MIN) FIG. 3. Effect of ethylenediaminetetraccetate (EDTA) and N-ethylmaleimide on hemolysis by vibriolysin and streptolysin. (A) Inhibition by EDTA of hemolysis induced by vibriolysin. Hemolysis was assayed in isotonic Tris-hydrochloride-buffered saline ( mm, ph 7.2) as described in the text. EDTA (2 mm) was added where marked by arrows. The amount of vibriolysin added in. ml of the reaction mixture was 3.7 ug. Hemolysis was assayed as described in the text at the time indicated. Each value is an average of duplicate determinations. (B) Inhibition by N-ethylmaleimide of hemolysis induced by streptolysin. Experimental conditions were as described in (A) except that hemolysis was assayed in mm phosphate buffer (ph 7.) and that N-ethylmaleimide ( mm) was added. The amount of the streptolysin added was hemolytic unit. Symbols: () A o without adding the inhibitor; (O) A,.o when the inhibitor was added at either min (A) or min (B); (A) A_MO when the inhibitor was added at either min (A). or min (B); (A) A, when the inhibitor was added at 2 min. -O.8 r (B) 4 E/ v~~~~.6 - N2 E- a,~~. t,. 4 re In el az to U) go IC.6 (C).4 ( FE E. zu :co Downloaded from on September 24, 28 by guest (n UX O. I.: 2 4 TIME (MIN) TIME (MIN ) 2 4 FIG. 2. Kinetics of the inhibition by Zn2+ ofthe hemolysis by various hemolytic agents. The components of the reaction mixture were as described in the legend offig.. Zinc acetate ( mm) was added where marked by arrows. Hemolysis was assayed as described in the text at the time indicated. Each value is an average of duplicate determinations. (A) Vibriolysin, 3.7 pg; (B) streptolysin, hemolytic unit; and (C) Triton X-, 6 x -3%. Symbols: () absorbance at 4 nm (A,4o) without adding Zn2+; (-)A when Zn2+ was added at min; (A) A,4o when Zn2+ was added at min; (A) A w when Zn2+ was added at 2 min. TIME (MIN)

4 242 TAKEDA, OGISO, AND MIWATANI. n z B (A) (B).. FIG. 4. Effect of Zn2+ on osmotic lysis. About 8.3 x 7 erythrocytes per ml in the indicated buffers were mixed in the presence of the indicated concentrations ofzinc acetate. The hemolysis was measured as described in the text. Each value is an average of duplicate determinations. Symbols for the buffer used: (A) () distilled water; (a) mm Tris-hydrochloride buffer (ph 7.2); (A) mm phosphate buffer (ph 7.); (A) mm boric acid-borax buffer (ph 7.); (B) () mmglycine-sodium hydroxide buffer (ph 8.7); (M) mm HEPES-sodium hydroxide buffer (ph 7.2); (A) mm acetate buffer (ph.); (A) mm cacodylate buffer (ph.); (O) mm barbital buffer (ph 7.3). B F. U'I C E-) B B uz B) co.... FIG.. Precipitation of hemoglobin in Tris-hydrochloride buffer by Zn2+. After about 8.3 x 7 erythrocytes per ml were lysed in the indicated buffers, various concentrations of zinc acetate were added and the reaction mixture was centrifuged. Then the absorbance at 4 nm of the resulting supernatant was measured. Each value is an average of dulicate determinations. Symbols for the buffer used: () distilled water; () mm Tris-hydrochloride buffer; (A) mm phosphate buffer. Effect of Zn2+ on hemolysis by vibriolysin, streptolysin, and Triton X- in phosphate buffer. Hemoglobin released from erythrocytes was not precipitated in phosphate buffer (Fig. 4 INFECT. IMMUN. and ). Thus, the effect of Zn2+ on hemolysis by vibriolysin, streptolysin, and Triton X- was investigated in phosphate buffer. Hemolysis by streptolysin was inhibited by Zn2+, whereas that by vibriolysin and Triton X- was not (Fig. 6). DISCUSSION The results presented in this paper show that the mode of inhibitory action by Zn2+ ofhemolysis induced by various bacterial and other cytolytic agents is not universal. Avigad and Bemheimer () studied the effect of Zn2+ and proposed that inhibition of Zn2+ occurred by a reversible alteration in the state of the lipid bilayer of erythrocytes membranes. But this proposal was based on their finding that inhibition of hemolysis by Zn2+ is universal to various bacterial and other cytolytic hemolysis. The present study demonstrates that in certain buffers, such as Tris-hydrochloride, boric acid-borax, HEPES-sodium hydroxide, and barbital, hemoglobins released from erythrocytes are easily precipitated by the mere presence of Zn2+. In view of our finding that hemoglobin is precipitated in the Tris-hydrochloride buffer used by Avigad and Bernheimer (), additional experiments with the various hemolytic agents that they studied, using a buffer that does not cause hemoglobin precipitation, should be done to determine the possible effect of Zn2+ on the hemolytic activity of various bacterial hemolysins. Reports by Montgomery et al. () and Yamamoto and Takahashi (8), which describe the inhibition by Zn2+ of complement-mediated hemolysis, should also be rein- E r- I vr) - Bm m B... FIG. 6. Effect of Zn2+ on hemolysis by various hemolytic agents in phosphate buffer. Hemolysis was assayed in mm phosphate buffer (ph 7.) as described in the text. Zinc acetate was added as indicated. The amounts of hemolytic agents added in. ml of the reaction mixture were: vibriolysin, 3.7 mg (); streptolysin, hemolytic unit (-); and Triton X-, 6 x -3% (A). Each value is an average of duplicate determinations. Downloaded from on September 24, 28 by guest

5 VOL. 7, 977 vestigated, since these workers used barbital buffer to study the inhibition by Zn2+. Oberly and Duncan (6) studied the hemolytic reaction induced by streptolysin in phosphate buffer and reported that the hemolytic process of streptolysin consists of at least two steps: (i) adsorption of the toxin to erythrocyte membranes and (ii) the step(s) after adsorption. They demonstrated that the latter step was inhibited by divalent cations, such as Ca2+ and Mg2+. Our results did not show any significant inhibition by these cations of the hemolysis induced by streptolysin (Table ). The discrepancy is probably due to a difference in the concentration of cations used; we used to mm concentrations, whereas Oberly and Duncan (6) used much higher concentrations (2 to 2 mm). The results in Fig. 6 show that Zn2+ inhibited hemolysis by streptolysin. It is clear that this inhibition is not the inhibition due to precipitation of released hemoglobins. Preliminary experiments in this laboratory show that Zn2+ might inhibit the binding of streptolysin to erythrocytes membranes. INHIBITION OF HEMOLYSIS BY ZN LITERATURE CITED. Avigad, L. S., and A. W. Bernheimer Inhibition by zinc of hemolysis induced by bacterial and other cytolytic agents. Infect. Immun. 3: Gomori, G. 9. Preparation of buffers for use in enzyme studies. Methods Enzymol. : Good, N. E., G. D. Winget, W. Winter, T. N. Connolly, S. Izawa, and R. M. M. Singh Hydrogen ion buffers for biological research. Biochemistry : Honda, T., S. Taga, T. Takeda, M. A. Hasibuan, Y. Takeda, and T. Miwatani Identificatin oflethal toxin with the thermostable direct hemolysin produced by Vibrio parahaemolyticus, and some physicochemical properties of the purified toxin. Infect. Immun. 3: Montgomery, D. W., L. K. Don, C. F. Zukoski, and M. Chvapil The effect of zinc and other metals on complement hemolysis of sheep red blood cells in vitro. Proc. Soc. Exp. Biol. Med. 4: Oberly, T. D., and J. L. Duncan. 97. Characteristics of streptolysin action. Infect. Immun. 4: Takeda, Y., Y. Hori, and T. Miwatani Demonstration of a temperature-dependent inactivating factor of the thermostable direct hemolysin in Vibrio parahaemolyticus. Infect. Immun. : Yamamoto, K., and M. Takahashi. 97. Inhibition of the terminal stage of complement-mediated lysis (reactive lysis) by zinc and copper ions. Int. Arch. Allergy Appl. Immunol. 48: Downloaded from on September 24, 28 by guest