Inhibition of Heme-promoted Enzymatic Lipid Peroxidation by Desferrioxamine and EDTA

Oxyhemoglobin, methemoglobin and hematin were found to catalyze xanthineoxidase-induced peroxidation of phospholipid liposomes, while oxyand metmyoglobin were inactive in this respect. The peroxidation was inhibited by desferrioxamine and by EDTA. Peroxidation catalyzed by 0.4 pM oxyhemoglobin was decreased by 50% by approximately 2 pM desferrioxamine or 20 pM EDTA and completely inhibited by 10 pM desferrioxamine or 100 pM EDTA. Inhibition of hemoglobin-catalyzed peroxidation was not accompanied by any changes in the absorbance spectra of hemoglobin, indicating that the heme iron was not withdrawn by the inhibitor. Inhibition of hematin-catalyzed peroxidation by desferrioxamine may have been due to iron chelation and removal, as judged from changes in absorbance spectra. The peroxidation was apparently not dependent on hydrogen peroxide since catalase did not inhibit peroxidation but on the contrary promoted it in some cases. INTRODUCTION Lipid peroxidation is an important component of the progressive tissue destruction observed after hemorrhage in, f o r example, brain and spinal cord tissue. This peroxidation is believed to be promoted by hemoglobin or hemoglobin degradation products (1.4.15) . Nonenzymatic peroxidation of lipids by heme compounds has been extensively studied (3,10,12,13), whereas heme-catalyzed peroxidation initiated by enzymatically generated oxygen radicals has not received similar attention so far (5,ll). The present study was an attempt to characterize further the heme-catalyzed phospholipid peroxidation initiated by xanthine oxidase. It was found that desferrioxamine and EDTA inhibited the peroxidation. It was also observed that catalase did not inhibit the peroxidation in the experimental system used.


INTRODUCTION
Lipid peroxidation is an important component of the progressive tissue destruction observed after hemorrhage in, f o r example, brain and spinal cord tissue. This peroxidation is believed to be promoted by hemoglobin or hemoglobin degradation products (1.4.15) . Nonenzymatic peroxidation of lipids by heme compounds has been extensively studied (3,10,12,13), whereas heme-catalyzed peroxidation initiated by enzymatically generated oxygen radicals has not received similar attention so far (5,ll).
The present study was an attempt to characterize further the heme-catalyzed phospholipid peroxidation initiated by xanthine oxidase. It was found that desferrioxamine and EDTA inhibited the peroxidation. It was also observed that catalase did not inhibit the peroxidation in the experimental system used.   X -x complete system with test compound and xanthine oxidase p r e s e n t d u r i n g t h e experiment.
x-----x test compound added a f t e r t h e   ( 8 ) . Furthermore, e a r l i e r studies have shown that hemoglobin iron is not withdrawn by desferrioxamine ( 8 ) .
It is possible that the lipid preparation used in the present investigation contained traces of peroxides. Gutteridge (6) found that preformed lipid peroxides reacted with hemoglobin to release iron which could be complexed to e . g .
desferrioxamine. The released ironnot the heme-ironwas suggested to be the catalyst of lipid peroxidation in hemoglobin-containing systems, since it was unlikely that heme-iron could act as a catalyst for the Fenton reaction.
However, lipid peroxidation does not require hydroxyl radicals (14) , which are the products of the Fenton reaction, and we found no effect of the Fenton reaction-inhibitor catalase in the study. It is thus possible that heme-iron, like ADP-iron (14) , promotes hydroxyl radical-independent lipid peroxidation.
The ability of desferrioxamine to block hemoglobin-promoted lipid peroxidation therefore suggests an interaction with hemoglobin, but further studies are required to elucidate the nature of this possible interaction. A similar case of complex formation was described by Winterbourn (16), who found that an EDTAlactoferrin complex, in contrast to lactoferrin alone, catalyzed the production of hydroxyl radicals from superoxide and hydrogen peroxide. In the present study EDTA was found to be less effective than desferrioxamine as an inhibitor of oxyhemoglobin promoted peroxidation, but at high concentration it inhibited all the heme compounds studied.
Nonenzymatic lipid peroxidation promoted by hemoglobin has been studied by Winterbourn and coworkers. Szebeni et a l . (12) found that hemoglobin entrapped in an emulsion of unsaturated phospholipids was rapidly oxidized under the formation of lipid peroxides, which suggested that the peroxidation was initiated by oxidation of oxyhemoglobin, leading to H202 production, o r by a reaction between oxyhemoglobin and peroxide contaminants of the lipid preparation. Catalase inhibited both lipid peroxidation and hemoglobin oxidation.
Carrel1 et a l . (3) discussed the concept that oxyhemoglobin may be analogous to a ferric superoxide and a potential producer of superoxide ions for oxidative stress reactions according to the formula: (Hb)Fe2+ + O2 <=> (Hb)Fe3+..0s -> (Hb)Fe 3+ + 02 -. Grisham (5) found that arachidonic acid was peroxidized by xanthine-oxidasegenerated radicals in the presence of myoglobin and hemoglobin. However, like the nonenzymatic peroxidation discussed above, and in contrast to the present findings, this peroxidation was sensitive to catalase inhibition and was enhanced by superoxide dismutase. The mechanism of myoglobin catalysis of arachidonic acid therefore seems to be quite different from the peroxidation observed in the present study, and is more similar to the phospholipid peroxidation induced by H 0 -activated metmyoglobin and methemoglobin described by Kanner and Hare1 (7).
Our results are in good accordance with those of Miura and Ogiso (ll), who found that oxyhemoglobin promoted xanthine-oxidase-induced peroxidation of erythrocyte ghosts, and that this peroxidation was inhibited by superoxide dismutase but not by catalase. Superoxide dismutase was included in our preliminary experiments and inhibited peroxidation completely (results not shown), but was considered to be of little interest since it would prevent the reaction between superoxide and heme compound. It is generally thought that in the absence of an appropriate complex of iron ( o r of certain other transition metals), superoxide is not able to bring about peroxidation, and the results of the present study are consistent with the notion that the lipid peroxidation is not caused by superoxide separated from the heme environment, since no lipid peroxidation was seen in the absence of heme compounds despite the presence of xanthine-oxidase-generated superoxide. It seems more likely that the peroxidation is initiated by a complex between heme and active oxygen. The inability of catalase to inhibit the reaction also supports the view that the heme compounds may be similar to ADP-iron complexes, which are also catalase-insensitive initiators of phospholipid peroxidation (14).