Pharmacology & Toxicology 1992,71, 297-301.

Lipid Peroxidation of Erythrocyte Membrane Induced by Lipoamide Dehydrogenate in the Presence of ADP=Fe3+ Toshiaki Miura, Sanae Muraoka and Taketo Ogiso Hokkaido Institute of Pharmaceutical Sciences, 7-1 Katsuraokacho, Otaru, 047-02, Japan (Received January 8, 1992; Accepted May 13, 1992) Absrruct: Lipid peroxidation of rat erythrocyte membranes was induced by lipoamide dehydrogenase (LADH) (EC 1.8.1.4) in the presence of ADP-Fe’+. Superoxide dismutase (SOD) (EC 1.15.1.1) strongly inhibited the peroxidation reaction but catalase did not. Hydroxyl radical scavengers, mannitol and dimethylsulfoxide did not inhibit the lipid peroxidation. These results indicated that the lipid peroxidation was a superoxide (0;)-dependent reaction, but the hydroxyl radical was not involved. ADP-Fe’+, in the presence of LADH, was reduced more rapidly under aerobic than anaerobic conditions and SOD under aerobic conditions strongly inhibited the iron reduction, indicating that 0; plays a predominant role in iron reduction. Hydrogen peroxide enhanced 0; generation by LADH, but the peroxidation reaction was not affected. In the presence of lipoamide, lipid peroxidation was also induced but the reactions were not inhibited by SOD. Evidently, the lipid peroxidation induced in the presence of lipoamide was 0;-independent. Dihydrolipoamide may be involved in the peroxidation reaction. Abbreviurions: BPS - bathophenanthroline sulfonate; DMSO - dimethylsulfoxide; FAD - Flavin adenine dinucleotide; LADH - lipoamide dehydrogenase; MDA - malondialdehyde; 0; - superoxide; SOD - superoxide dismutase

Flavoenzymes may play a predominant role in the production of oxidative stress in vivo. Some of these flavoenzymes reduce O2directly (Fridovich 1970), while others, including microsomal N A D ( P ) H dehydrogenase (Iyanagi & Yamazaki 1969 & 1970), N A D P H cytochrome P-450 reductase, NADH cytochrome b5 reductase (Bachur et af. 1979; Forman & Boveries 1982) and dihydroorotate dehydrogenase (Forman & Boveries 1982), do so indirectly by reducing some electron receptors, which in turn react with 02.One of the flavoenzymes, lipoamide dehydrogenase ( L A D H ) [EC 1.8.I .4], reacts with O2 by way of electron transfer to produce superoxide (0,) (Massey et al. 1969; Sreider et d. 1990; Grinblat et af. 1991). However, the reaction rate is very slow and it is unclear whether or not oxidative damages are caused by this enzyme. Recently it has been reported that LADH releases iron from ferritin, which is a major iron storage protein in cells (Bando & Aki 1990). Furthermore, it has been found that hydroxyl radicals may be generated via metal-catalyzed Haber-Weiss reaction during the release of iron from ferritin by LADH (Bando & Aki 1991). These findings suggest that oxygen damage may be caused by LADH. In the present study, we demonstrated that lipid peroxidation o f erythrocyte membranes was induced by LADH in the presence of ADP-Fe3+. Materials and Methods Lipoamide dehydrogenase (porcine heart), lipoamide, superoxide dismutase (bovine erythrocytes), catalase (bovine live, thymol free) and cytochrome c (horse spleen) were obtained from Sigma Chemical Co. St. Louis, Mo., U.S.A. Thiobarbituric acid was from Merck, Japan. NADH and ADP were purchased from Oriental Yeast Co. Ltd. Tokyo, Japan. Other chemicals were analytical grade from commercial suppliers. The native LADH suspension was dialyzed against 10 mM Tris-HCI buffer pH 7.4 containing 0.15 M NaCI.

Lipid peroxidation assays. Rat erythrocyte membranes were prepared by the method previously reported (Miura & Ogiso 1982). Reaction mixtures contained erythrocyte membranes (200 pg protein), 50 nM LADH, 0.1 mM NADH, 10 pM ADP-Fe’+ (ADP:Fe’+=l.7:0.1) and 0.15 M NaCl in 3.0 ml of 10 mM TrisHCl buffer pH 7.4, unless otherwise noted. Lipid peroxidation reactions were performed by incubating the reaction mixture at 37”. Reactions were initiated by the addition of LADH and terminated by 0.3 ml of 30% trichloroacetic acid. After centrifugation at 3,000 r.p.m. for 10 min., 1.0 ml of 0.6% TBA was added to the supernatant and then heated for 30 min. at 100”. To prevent the formation of TBA reactive substances during heating, 0.1 ml of 1% butylated hydroxytoluene dissolved in dimethylsulfoxide was added to the reaction mixture prior to heating. Formation of malondialdehyde was determined by measuring the absorbance at 535 nm as described previously (Buege & Aust 1977). Determination of 0;production. 0; production was measured by the cytochrome c method (McCord & Fridovich 1969), using a Hitachi-U2000 spectrophotometer. The reaction was started by adding LADH, and other conditions were as described in the figure legends. The 0; dependent reduction of the ADP-Fe’+ was measured in a reaction mixture containing LADH (50 nM), 0.1 mM NADH, 100 pM ADP-Fe’+ complex (molar ratio = 1.7 : 0. I), I .O mM bathophenanthroline sulfonate (BPS), 0.15 M NaCl in 3.0 ml of 10 mM Tris-HCI pH 7.4. Formation of the stable Fe2+-BPS complex was measured at 530 nm. The amounts of Fez+-BPSwere calculated from ~ = 2 2 . 1 4mM-’ (Carter 1971). Enzyme assays. LADH activity was measured by the rate of NADH oxidation using lipoamide as an electron acceptor. The reaction mixture contained 50 nM LADH, 0.1 mM NADH, 1.O mM lipoamide, 0.15 M NaCl in 10 mM Tris-HCI buffer pH 7.4. Diaphorase activity was determined by measuring the rate of dichrolophenolindophenol reduction at 600 nm (Sreider et al. 1990). The reaction mixture contained 50 nM LADH, 0.1 mM NADH, 40 pM dichlorophenol-indophenol in 10 mM Tris-HCI buffer pH 7.4. Statistical analysis. Data were evaluated by one-way analysis of variance followed, where appropriate, by a Student’s t-test.

298

TOSHIAKI MIURA ET AL.

60

i

n I

8

40

W

c

@

I

/

d , i t

20

0 * 50

0

Tire (rin) Fig. I. Time course of lipid peroxidation induced by LADH in the presence of ADP-Fe’+. Conditions were described in Materials and Methods. Each point represents the mean of three experiments. (o), complete reaction mixture; (o)?minus LADH and (A), minus ADPFe’ . +

Results LAD H- induced lipid per oxidut ion. As shown in fig. 1, lipid peroxidation of erythrocyte membranes measured by the formation of MDA was induced by LADH in the presence of ADP-Fe3+.Omission of LADH or replacement of native LADH with heat-denatured LADH resulted in very slight formation of MDA. In the absence of ADP-Fe’+, however, no detectable lipid peroxidation was found. Replacement of ADP-Fe3+ with free Fe3+, Fe3+EDTa (molar ratio= 1 : 1) or Fe3+-citrate (molar ratio= 1 : 1) also induced lipid peroxidation by LADH (data not shown). It can be seen from fig. 2 that the rate of lipid peroxidation was dependent upon the concentration of the LADH up to 100 nM. Fig. 3 shows that 0, generation depends on the concentrations of LADH. As summarized in table 1, SOD strongly inhibited the lipid peroxidation and inhibition was related to the SOD concentration, 1.0

0.1

1

10

100200

LAOH ( n H ) Fig. 3. Superoxide generation induced by LADH. The reaction mixture contained 30 pM cytochrome c, 0.1 mM NADH, 0.15 M NaCl and various concentrations of LADH in 3.0 ml of 10 mM Tris-HCI buffer pH 7.4. Each point represents the mean of three experiments. (o), complete reaction mixture and ( 0 ) plus SOD (5 WmI).

pg/ml of the enzyme did so about 90%. It has often been observed that catalase promotes the xanthine oxidase-induced lipid peroxidations (Miura et al. 1984; Thomas et ul. 1985). In LADH-induced lipid peroxidation, however, catalase has no significant effects. It is unlikely that hydroxyl radicals are involved in the peroxidation reaction because the lipid peroxidation was caused in Tris-HCI, which is a strong hydroxyl radical scavenger (Hicks & Gebicki 1986). Moreover, hydroxyl radical scavengers, mannitol and DMSO did not inhibit the peroxidation reaction. As summarized in table 2, ADP-Fe3+was reduced more rapidly under aerobic than anaerobic conditions. SOD under aerobic conditions inhibited the iron reduction about 60%. Evidently, two-thirds of ADP-Fe3+ reduction was due to 0, generated by LADH, and the remainder may be directly reduced by the enzyme. The addition of catalase had no significant effect on the ADP-Fe3+ reduction. These results suggest that 0,-dependent iron reduction was responsible for the lipid peroxidation.

Table I Effect of oxygen radical scavengers on lipid peroxidation. Scavengers I. 2. 3. 4.

5. 6. 7. 8.

LAOH (nfl) Fig. 2. Effect of varying LADH concentrations on lipid peroxidation. The incubations were performed for 1 hr at 37‘. Other conditions were the same as those of fig. I , except for concentrations of LADH.

None SOD (0.1 pg/ml) SOD (1.0 pg/ml) Catalase (1.0 pg/ml) Catalase (10.0 pg/ml) SOD (1 .O pg/ml) and catalase (10.0 pg/ml) Mannitol (100 mM) DMSO (100 mM)

MDA (nmol/ hr)

Inhibition

60.5+ 12.8 20.8 +2.6* 8.7 f 3.7* 57.2k 14.4 63.4k 12.7

0 65.6 85.6 5.4

8.0*2.1* 65.6f 1 1 . I 62.9 8.5

86.8

(1%))

-

-

Scavengers were added to the reaction mixture before the start of the reaction. Incubation was performed for I hr. Other conditions were the same as those of fig. 1. Each value represents the mean+ S.D. of five experiments. *Significantly different from experiment 1 (P< 0.001).

299

LIPID PEROXIDATION O F ERYTHROCYTE MEMBRANES Table 2. Effect of SOD and catalase on ADP-Fe2+ formation induced by LADH.

Effect of SOD on the reductions of cytochrome c, ADP-Fe3+ and lipid peroxidation induced by LADH in the presence of lipoamide. Cytochrome ADP-Fe'+ c reduced reduced

ADP-Fe'+ reduction (nmolimin. iml)

Conditions I. 2. 3. 4.

Table 4.

5.87 k0.29 (5) 3.84k 1 . 1 1 (4)"' 2.64k0.39 (5)* 5.76k0.12 (5)

Aerobic Anaerobic I +SOD (1 .O pg/ml) I +catalase (10 pgiml)

Reaction mixture contained 100 pM ADP-Fe3+,1.0 mM BPS, 0.15 M NaCI, 0.1 mM NADH, and 50 nM LADH in 3.0 ml of 10 mM Tris-HCI buffer pH 7.4. Increase in absorption at 530 nm was monitored continuously. Each value represents the mean fS.D. The number in parenthesis is the number of experiments. *Significantly different from experiment 1 (P