J. Biochem. 89, (1981) Masako TANIGUCHI, Makoto AIKAWA, and Toshio SAKAGAMI. Received for publication, August 22, 1980

Size: px

Start display at page:

Download "J. Biochem. 89, (1981) Masako TANIGUCHI, Makoto AIKAWA, and Toshio SAKAGAMI. Received for publication, August 22, 1980"

Spencer Nichols
5 years ago
Views:

1 J. Biochem. 89, (1981) A Simple and Effective Method for Homolysis with a Hypoxanthine - Xanthine Oxidase System and Alteration of Erythrocyte Phospholipid Composition during the Hemolysis Masako TANIGUCHI, Makoto AIKAWA, and Toshio SAKAGAMI Department of Biochemistry, Sapporo Medical College Chuo-ku, Sapporo, Hokkaido 060, Received for publication, August 22, 1980 A very rapid hemolysis was found to be caused by active oxygen species produced by a hypoxanthine-xanthine oxidase system with very low concentrations of hypo xanthine. The addition of superoxide dismutase or catalase inhibited the hemolysis, indicating that 02- and H2O2 participate in this system. The extent of erythrocyte hemolysis was found to depend on the sex of the human donor. The change in phospholipid composition before and after hemolysis in human erythrocytes from donors of each sex was compared by thin layer chro matography. A significant decrease in phosphatidylethanolamine content and a concomitant increase in altered phospholipid fraction were observed in erythrocytes from male donors, suggesting that these erythrocytes were easily attacked by active oxygen species to produce modified phosphatidylethanolamine. Many reports have recently been published con cerning the structural changes of biological mem branes induced by active oxygen species. Human erythrocytes have also been used as a target for active oxygen, but the results obtained have never been clear-cut for several reasons, one of which is the formation of green pigments and precipitates of hemoglobin, which prevented the optical deter mination of hemolysis (1, 2). Therefore the hemolysis was barely observable with the xanthinexanthine oxidase system, except in the case of bovine erythrocytes (3). Moreover, very long incubation periods were required to cause hemoly sis, even when human erythrocytes were treated with thyroxin (4) or when rat erythrocytes were exposed to lead ions (5). In the present study, we investigated the hemolysis of normal human erythrocytes by hy poxanthine-xanthine oxidase. It was found that rapid and readily detectable hemolysis occurred when the level of active oxygen species produced was kept as low as possible. There are a few reports on the change of phospholipid composition in biological membranes caused by active oxygen (6, 7). However, no report has appeared on the changes in erythrocyte phospholipid compositions upon oxidative hemol ysis. In the present study, investigations on the change of phospholipid composition in erythro cytes in situ were carried out. The present work on the hemolysis induced by the hypoxanthine-xanthine oxidase system was undertaken in an effort to develop a simple, rapid and clear-cut method of hemolysis and to elucidate Vol. 89, No. 3,

796 M. TANIGUCHI, M. AIKAWA, and T, SAKAGAMI the effects of active oxygen species on the mem brane fragility as well as on the phospholipid com position.

2 796 M. TANIGUCHI, M. AIKAWA, and T, SAKAGAMI the effects of active oxygen species on the mem brane fragility as well as on the phospholipid com position. We also found that the effect of oxidative hemolysis induced by the oxidase system, depended on the sex of the erythrocyte donor. MATERIALS AND METHODS Reagents-Hypoxanthine, xanthine oxidase from butter milk, superoxide dismutase from bovine blood and catalase from bovine liver were all purchased from Sigma Chemical Co. (St. Louis). Xanthine oxidase was further purified by Sephadex G-200 and DEAF-cellulose column chromatog raphy according to the method of Waud et al. (8). Xanthine oxidase activity was assayed by the method of Lynch and Fridovich (9). One unit of enzyme activity was defined as the amount of enzyme that releases 1 ƒêmol of uric acid per min at 25 Ž. Catalase and superoxide dismutase were used without further purification. Enzymes were all dialyzed against 10 mm phosphate buffer (ph 7.0) before use. Erythrocytes-Blood was drawn by venous puncture into heparinized test tubes from 10 healthy male and 8 healthy female adults aged Erythrocytes were washed at least three times with ten volumes of ice-cold isotonic buffer (described below). Leucocytes were eliminated by aspirat ing the buffy coat. To remove leucocytes and platelets from whole blood, microcrystalline cellu lose and ƒ -cellulose (1 : 1, by weight) were also used according to the method of Beutler et al. (10). The erythrocytes separated by the two methods gave the same results in the following experiments. Assay System for Hemolysis-In a typical ex periment, a 1% suspension of washed erythrocytes were prepared in the standard buffer (145 mm NaCl, 5 mm KCl, 1 mm MgCl2, 1 mm CaCl2, 10 mm glu cose, and 5 mm sodium phosphate buffer, ph 7.0). Five ml of erythrocyte suspension was placed in a 50 ml centrifuge tube, and 0.25 unit of xanthine oxidase and 50 pl of hypoxanthine at a final con centration of mm were added to the suspen sion. The suspension was incubated at 37 Ž in an incubator oscillating at 200 rpm. After the incu bation, the extent of hemolysis was determined spectrophotometrically at 524 nm in the supernatant after centrifugation at 2,000 ~ g for 10 min. Addition of an aliquot of incubation mixture to distilled water gave 100% hemolysis. In the case of hypotonic hemolysis, the erythrocytes were centrifuged from the incubation mixture and a 0.05 ml portion of packed erythrocytes was added to a series of tubes containing 5 ml each of NaCl solution ranging from % to 0.475%. After standing for 30 min, the suspension was centri fuged and the percent hemolysis of each tube was assayed spectrophotometrically. Extraction of Lipids-Lipids were extracted from each incubation mixture with isobutanolchloroform (11 : 7, by volume) (11) or chloroformmethanol (2 : 1, by volume). Extracts were washed by the procedure described by Folch et al. (12). Thin layer chromatography was carried out with chloroform : methanol : acetic acid : water (100 : 60 :14 :4, by volume) (13). Each phospholipid was visualized by means of iodine vapor, ninhydrin spray and phosphomolybdic acid spray on thin layer chromatograms. Each spot was scraped off and lipid phosphorus was determined by the method of Bartlett (14). The thiobarbituric acid reaction was carried out by the method described in the report by Walls et al. (4). RESULTS Hemolysis was found to occur clearly at a hypoxanthine concentration of mm (Fig. 1 (A)), which is one-tenth of the hypoxanthine con centration used for the hemolysis of bovine erythrocytes by Bartosz et al. (3). Other investiga tors have failed to observe oxidative hemolysis of normal human erythrocytes with hypoxanthine or xanthine at a concentration of more than 1.67 mm in the xanthine oxidase system (4). In the present experiment, hemolysis could not be detected when erythrocytes were exposed to 0.68 mm hypo xanthine and xanthine oxidase, even though incu bations were carried out for 6 h or more (Fig. 1 (B)), but the erythrocytes underwent a rapid change in color and dark pigments and precipitates ap peared. The hypotonic fragility of the erythro cytes was greatly enhanced by incubation with 0.68 mm hypoxanthine and xanthine oxidase (Fig. 2). Optimal Conditions for Hemolysis-The opti mal conditions for hemolysis were studied. When erythrocytes were incubated at hypoxanthine con centrations in the range of mm to mm, J. Biochem.

3 OXIDATIVE HEMOLYSIS AND PHOSPHOLIPID IN ERYTHROCYTES 797 Fig. 1. Typical hemolytic curve. Erythrocytes from adult human males were incubated with xanthine oxidase (0.05 unit/ml) and hypoxanthine, mm (A) or 0.68 mm (B). Control incubation was carried out without the xanthine oxidase system (C). Other incubation conditions were as described in " MATERIALS AND METHODS." The variations are shown by vertical bars or are covered by the size of the symbol used. Fig. 3. Effects of hypoxanthine (A) and xanthine oxidase (B) concentrations. Incubations were performed for 60 min. Other conditions were as described in" MATERIALS AND METHODS." The variations are shown by the vertical bars or are covered by the size of the symbol used. not be determined. A considerable degree of hemolysis was observed during incubation in the absence of hypoxanthine, and this might be due to the presence of endogenous hypoxanthine in erythrocytes. Figure 3 (B) shows the effects of xanthine oxidase concentration on hemolysis; 0.05 unit of oxidase per ml of incubation mixture was found to be most effective for hemolysis. Fig. 2. Enhancement of osmotic fragility. Small ali quots of suspensions of erythrocytes from male donors were incubated with the oxidase system (A) (0.05 unit/ml xanthine oxidase and 0.68 mm hypoxanthine) or without the oxidase system (B), then added to 5 ml of NaCl solution of various concentrations. Percent hemolysis was observed as described in " MATERIALS AND METHODS." The variations are shown by the vertical bars or are covered by the size of the symbol used. clear hemolysis was found to occur (Fig. 3 (A)). On the other hand when the hypoxanthine con centration was above mat, the hemoglobin became dark green and apparent hemolysis could As shown in Fig. 4, hemolysis was significantly inhibited under these incubation conditions without CaCl2. On the other hand, the hemolysis seemed to be enhanced when MgC12 was removed from the complete mixture. Addition of glucose had no effect on this oxidative hemolysis. Protection by Superoxide Dismutase and Cata lase-as shown in Table I, the lytic effects of the hypoxanthine-xanthine oxidase reaction were re duced by superoxide dismutase or by catalase when they were added at such concentrations as 0.02 ƒêg of superoxide dismutase and 0.83 ƒêg of catalase per ml of incubation mixture. Addition of catalase seemed to be more effective than that of superoxide dismutase, suggesting that H2O2 participated to a Vol. 89, No. 3, 1981

4 798 M. TANIGUCHI, M. AIKAWA, and T. SAKAGAMI greater extent in this oxidative hemolysis than 02 did. Higher concentrations of protective enzymes resulted in enhancement of the hemolysis, as described for superoxide dismutase in the report by Fong et al. (15). Mannitol and ethanol as OH scavengers and histidine as a scavenger of singlet oxygen had no effect on hemolysis at con centrations of 5 mm. Dependence of Hemolysis and the Changes of Phospholipid Composition on the Sex of the Eryth rocyte Donor-The erythrocytes from male donors TABLE T. Effects of superoxide dismutase and cata lase on the oxidative hemolysis induced by the xanthine oxidase system. Erythrocytes were incubated for 60 min as described in " MATERIALS AND METHODS," except that superoxide dismutase or catalase was added. Fig. 4. Effect of divalent cations on hemolysis. Incu bations were carried out as described in " MATERIALS AND METHODS," except that the following com ponents were subtracted from the standard buffer. (A) minus MgCl2, (B) complete mixture, (C) minus KCl, (D) minus glucose, (E) minus CaCl2, (F) minus hypo xanthine and xanthine oxidase. Variations are shown by the vertical bars or are covered by the size of the symbol used. a Data are means of three experiments. TABLE U. Dependence of changes in phospholipid composition during incubation with the xanthine oxidase system on the sex of the erythrocyte donor. Erythrocytes were incubated and the lipids were extracted as described in " MATERIALS AND METHODS" at each incubation period. Values are expressed as % of total phospho lipids. Variations were within 5%. a Control incubation without the oxidase system. b Values are the averages of two experiments. c Including phos pholipids between sphingomyelin and the origin on the thin layer chromatogram. J. Biochem.

5 OXIDATIVE HEMOLYSIS AND PHOSPHOLIPID IN ERYTHROCYTES 799 Fig. 5. Sex difference in hemolysis. Erythrocyte sus pensions from male (A) and female (B) donors were incubated as described in " MATERIALS AND METHODS." The variations are shown by the vertical bars or are covered by the size of the symbol used. were found to be more easily hemolyzed than those from female donors (Fig. 5). Lipid extracts from the incubation mixtures at various incubation times showed that the phos pholipid composition had changed during incuba tion with the oxidase system. As shown in Table U, the proportion of phosphatidylethanolamine in erythrocytes from male donors was significantly decreased, and the phospholipid fraction (PX) with the same R f value as lysophosphatidylethanolamine (which was located just above phosphatidylcholine on a thin layer chromatogram and was ninhydrin positive) was observed to increase concomitantly. However, in erythrocytes from female donors the phospholipid fraction did not increase as markedly. On the other hand, the total phospholipid content in erythrocytes after incubation with the oxidase system was not changed. Figure 6 shows the ninhydrin-positive spot (PX), which was produced during incubation with the xanthine oxidase system. The amount of phosphatidylserine was also decreased by approximately 50% as compared to that in erythrocytes incubated without the oxidase system. A decrease in phosphatidylcholine was also found. The phospholipids observed be tween sphingomyelin and the origin on the thin layer chromatogram might be a mixture of lyso phosphatidylcholine, lysophosphatidylserine, and other phospholipids (if any) produced during the oxidative hemolysis. Fig. 6. Typical thin layer chromatogram of lipid ex tracts from erythrocytes of male donors. Erythrocytes from male adults were incubated with (O) or without (C) the xanthine oxidase system for 60 min and lipids were extracted as described in " MATERIALS AND METHODS." Each phospholipid was visualized by spraying with phosphomolybdic acid (A) or ninhydrin (B) reagent. PE, phosphatidylethanolamine; PS, phos phatidylserine; PI, phosphatidylinositol; PC, phos phatidylcholine; SPH, sphingomyelin; Or, origin. Under these experimental conditions, we could not detect lipid peroxidation by means of the TBA reaction, presumably at least partly because of the low concentration of erythrocytes in the incubation mixtures. DISCUSSION Goldberg and Stern (2) reported the toxic action towards erythrocytes of OZ induced by the auto - oxidation of dihydroxyfumaric acid. They ob served that human erythrocytes exposed to active oxygen species underwent a rapid breakdown of hemoglobin to methemoglobin and other green pigments, resulting in an increase of the osmotic fragility of the erythrocytes. We observed the same phenomenon with active oxygen species in duced by a high hypoxanthine concentration (0.68 mm) in the xanthine oxidase system, as shown in Fig. 2. Although some investigators have tried to hemolyze erythrocytes by exposure to the xanthine - xanthine oxidase reaction, addition of thyroxin to Vol. 89, No. 3, 1981

6 800 M. TANIGUCHI, M. AIKAWA, and T. SAKAGAMI the medium (4), or lead exposure (5) was required to obtain sufficient hemolysis. Moreover, Kellogg and Fridovich (1) noted that xanthine and urate prevented lipid peroxidation. Since the xanthine-xanthine oxidase system is a very simple system able to produce active oxygen species, we studied the experimental conditions and found that clear and rapid hemolysis occurred in experiments performed at a very low hypoxanthine concentration (16). In the present study a remarkable difference was observed in the extent of oxidative hemolysis between erythrocytes from male and female donors. Devenuto showed that the blood obtained from female donors was able to survive storage lesions better than the blood from male donors by the criterion of osmotic resistance (17). Erythrocytes from male donors seemed to be more labile and more susceptible to membrane changes caused by active oxygen, but further studies are required to clarify the mechanism of the difference in behavior. May and McCay (6) reported that NADPH oxidase brought about changes of microsomal membrane phospholipids. They showed that phospholipids altered by active oxygen species were considerably more polar than normal phospho lipids, but the ester value of the phospholipids remained constant. Further, they observed a de crease of polyunsaturated fatty acids in phospho lipid caused by active oxygen produced by NADPH oxidase and the microsomal system. Since the changes in erythrocyte membrane phospholipids exposed to active oxyten species have never been reported, experiments were carried out to see whether or not changes in the composition or structure of phospholipids are caused during incu bation with the xanthine oxidase system. As noted in " RESULTS," it was found that phos phatidylethanolamines rich in C20:4 and C22:6 fatty acids were attacked by active oxygen, resulting in an increment of more polar phospholipid with the same R f value as lysophosphatidylethanolamine. Phosphatidylserine is also known to be rich in polyunsaturated fatty acids. The amount of this phospholipid decreased during the incubation. These results are consistent with the report by May and McCay et al. that the polyunsaturated fatty acids were attacked more easily than other fatty acids by active oxygen species (6). Similar results were presented by Dodge and Phillips (18), who detected altered phospholipid in the lipid extract from human erythrocytes. At present, however, it is not certain that the spot PX is the same substance as that observed by May and McCay (6) and Dodge and Phillips (18). Experiments to determine the structure of PX are required. REFERENCES 1. Kellogg, E.W., III & Fridovich, I. (1977) J. Biol. Chem. 252, Goldberg, B. & Stern, A. (1977) Arch. Biochem. Biophys. 178, Bartosz, G., Fried, R., Grzelinska, E., & Leyko, W. (1977) Eur. J. Biochem. 73, Walls, R., Kumar, K.S., & Hochstein, P. (1976) Arch. Biochem. Biophys. 174, Gelman, B.B., Michaelson, I.A., & Bus, J.S. (1978) Toxicol. Appl. Pharmacol. 45, May, H. & McCay, P.B. (1968) J. Biol. Chem. 243, Tam, B.K. & McCay, P.B. (1970) J. Biol. Chem. 245, Waud, W.R., Brady, F.O., Wiley, R.D., & Rajazo palon, K.V. (1975) Arch. Biochem. Biophys. 169, Lynch, R.E. & Fridovich, I. (1978) J. Biol. Chem. 253, Beutler, E., West, C., & Blume, K.G. (1976) J. Lab. Clin. Med. 88, Rose, H.G. & Oklander, M. (1965) J. Lipid Res. 6, Folch, J., Lees, M., & Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, Kudo, N., Minagawa, K., Horiuchi, I., Taniguchi, M., & Sakagami, T. (1973) Seikagaku (in Japanese) 45, Bartlett, G.R. (1959) J. Biol. Chem. 234, Fong, K.L., McCay, P.B., Foyer, J.L., Keele, B.B., & Misra, H. (1973) J. Biol. Chem. 248, Taniguchi, M., Aikawa, M., & Sakagami, T. (1979) Proc. Jpn. Conf. Biochem. Lipid 21, Devenuto, F., Wilson, H.L., Wilson, S.S., & Ligon, D.F. (1971) Proc. Soc. Exp. Biol. Med. 136, Dodge, J.D. & Phillips, G.B. (1966) J. Lipid Res. 7, J. Biochem.

The Elution Behaviors of Acidic Phospholipids on. Chromatography*

J. Biochem., 69, 255-263 (1971) The Elution Behaviors of Acidic Phospholipids on Column Chromatography* Tei SHIMOJO, Hideo KANOH and Kokichi OHNO The Department of Biochemistry, Sapporo Medical College,