Release of active oxygen radicals by leukocytes of Fanconi

Transcription

1 Release of active oxygen radicals by leukocytes of Fanconi anemia patients Ludmila G. Korkina, Elena V. Samochatova, Aleksei A. Maschan, Tatjana B. Suslova, Zinaida P Cheremisina, and lgor B. Afanas ev Russian Institute of Hematology for Children, Moscow, and Vitamin Research Institute, Moscow Abstract: The release of oxygen radicals by blood and bone marrow leukocytes of patients with Fanconi anemia (FA) has been studied. It was found that the nonstimulated FA leukocytes and those stimulated by concanavalin A, Si02, latex, and opsonized zymosan produced enhanced levels of luminol- and lucigenin-dependent chemiluminescence (CL) in comparison with normal leukocytes. At the same time, the ratio of the intensity of lucigenindependent CL to that of luminol-dependent CL was significantly smaller for FA leukocytes than for normal leukocytes. From these findings and from the effects of antioxidative enzymes and free radical scavengers on CL, it was concluded that FA leukocytes release enhanced amounts of oxygen radicals and that these free radicals contain enhanced amounts of hydroxyl or hydroxyl-like radicals more active than superoxide ion. It was proposed that elevated reactivity of the oxygen radicals released by FA leukocytes may be a major factor in the development of Fanconi anemia; this proposal is supported by the first positive results of treatment of FA patients with rutin (a nontoxic natural free radical scavenger and chelator). J. Leukoc. Biol. 52: ; Key Words: oxygen radicals. superoxide. leulcocytes. chemiluminescence Fanconi anemia Introduction CL, whereas those stimulated with concanavalin A (Con A) released superoxide ion and hydroxyl radicals simultancously because they produced both luminol- and lucigenindependent CL. One might expect the development of free radical pathologies to affect the structures and amounts of oxygen radicals released by phagocytes. A well-known example is the inability of leukocytes of patients with chronic granulomatous disease to produce oxygen radicals due to some defects in the structure and the activation mechanism of NADPH oxidase [4]. Another example of free radical pathologies is Fanconi anemia (FA). FA is an autosomal recessive disorder characterized by progressive pancytopenia, chromosomal instability, growth retardation, and congenital malfunctions [5]. The erythrocytes of FA patients exhibited decreased superoxide dismutase (SOD) activity [6-9] and enhanced superoxide production [10]. Therefore, it may be proposed that the development of FA is a consequence of overproduction of oxygen radicals. We have studied the production of oxygen radicals by leukocytcs of FA patients, using the luminol- and lucigenindependent CL technique. In contrast to normal leukocytes, which released mainly superoxide ion, leukocytes of FA patients released enhanced amounts of active hydroxyl or hydroxyl-like free radicals. Based on the results of in vitro experiments, the use of the free radical scavenger and chelator rutin was recommended for treatment of FA patients. Longterm rutin treatment resulted in some decrease in chromosomal aberrations and improvement of hematological characteristics and health of FA patients. Oxygen radical production by leukocytcs has been widely studied [i]. It was shown that the primary oxygen species produced by NADPH oxidase of leukocytes and other phagocytes is superoxide ion. Superoxide ion may be released by the cells or be converted to much more active oxygen species such as hydroxyl or hydroxyl-like free radicals via the supcroxidc-driven Fenton reaction (reactions I and 2): 02 + Fe3-0 + Fe2 Fe2 + H2O2 : Fe3 + HO #{247} H0 Indeed, release of hydroxyl radicals by stimulated neutrophils, monocytes, and macrophages has been shown [1]. The yields of superoxide ion and other oxygen species can be measured by means of luminol- and lucigenin-dependent chemiluminescence (CL); the latter is considered a specific test for superoxide production [2]. Both methods have been applied for measuring oxygen radical production by phagocytes. Thus, it has been shown [3] that human granulocytes stimulated with phytohemagglutin did not release superoxide ion because they produced only luminol-dependent MATERIALS AND METHODS Patients with Fanconi anemia The diagnosis of FA was confirmed clinically and cytogenetically for nine patients (one male and eight females). The patients ages ranged from 6.5 to 14 years. All major manifesta- (1) tions of the disease were due to pancytopenia (leukocytopenia, x i0 cells/l blood; erythrocytopenia, 33- (2) 70 g Hb/L, and platelctpenia, 1-3 x i0 cells/l). Bone marrow was hypocellular and contained no more than 2-3 x 1010 cells/l. In all cases, lymphocytosis (more than 30%), granulocytopenia, and megakariocytopenia were observed. Diagnosis was confirmed cytogenetically: the total Abbreviations: CL, chemiluminescence; Con A, concanavalin A; FA, Fanconi anemia; HBSS, Hanks balanced salt solution; NDGA, nordihydroguaretic acid; PMN, polymorphonuclear leukocyte; SOD, superoxide dismutase. Reprint requests: I. B. Afanas ev, Vitamin Research Institute, Nauchny pr. 14A, , GSP-7, Moscow, Russia. Received March 9, 1992; accepted April 14; Journal of Leukocyte Biology Volume 52, September

2 amount of chromosomal aberrations in blood lymphocytes was within the range 6-56% and strongly correlated with FA phenotype. All patients manifested physical and psychic malfunctions. Chemicals Luminol, lucigenin, acctylsalicylic acid (aspirin), colchicine, nordihydroguaretic acid (NDGA), indomethacin, and latex were from Sigma; Si02 (DQ 12, world standard) was from Germany; concanavalin A (Con A) was from Serva; dextran was from Farmacia, Sweden; phytohemagglutinin P was from Difco, USA; and rutin and taurin were of USSR production. Enzymes Superoxide dismutase and catalase from bovine erythrocytes were from Sigma, and horseradish peroxidase (HRP) was from Rcanal, Hungary. Preparation of leukocytes Leukocytes were isolated from peripheral blood of healthy donors (n = 4) and from peripheral blood and bone marrow of FA patients (n = 9). The blood samples (5 ml) were collected by venipuncture and anticoagulated with 1 ml of heparin (20 U) in Hanks balanced salt solution (HBSS). The bone marrow samples (i ml) were obtained by puncture of sternum. The samples were then scdimcnted with 1 ml of 6% dextran solution at 37#{176}Cfor 30 mm. The white cell-rich supernatant was centrifuged at lsog for 10 mm. Cell pellets were washed twice in the Ca2, Mg-frec HBSS. The final suspension of 2-3 x 106 leukocyte/ml was prepared in medium 199. The cells were counted with a Coulter counter, and their viability was assessed by exclusion of 0.2% trypan blue dye. Cell differential count was confirmed by Giemsa staining. It was found that the blood cell suspension isolated from normal donors contained 70-80% polymorphonuclear leukocytes (PMNs), 1-2% monocytes, and 20-30% lymphocytes; the bone marrow cell suspension had the same compo- TABLE 1. Luminol-Dependent CL of Whole Blood, Isolated Mononuclear Leukocytes, and Total Leukocyte Fraction of FA Patients CL intensi ty (mv) per 106 cells Without Opsonised stimulus Con A SiO, zymosan Latex Whole blood Mononuclear leukocytes N.A.b Total leukocyte fraction Average values for three FA patients. bnot analyzed. sition. Because of neuropenia and lymphocytosis, differential isolation of PMNs from the cell suspensions of FA patients was not performed. All measurements of the blood samples from FA patients and donors were carried out simultaneously. Chemiluminescence measurements Lucigenin- and luminol-dependent CL was measured on an LKB luminometer (model 1251, Sweden). Leukocyte suspension (50 jl, 2 x 106 cells/mi), luminol or lucigenin (50 jm, final concentration), and 0.85 ml of HBSS were mixed in the 1-ml polystyrene cell at 37#{176}C.After 5 mm, the 0.5% suspension of latex (50 gil), the suspension of Si02 (50 gil, 10 mg/mi), or Con A (50 gil, I mg/ml) in 0.9% NaC1 solution was added and the light emission was recorded continuously. The intensity of spontaneous CL and the difference between the maximal values of the cellular CL response to an amplifier (luminol or lucigenin) and of spontaneous CL were measured. When the effects of free radical scavengers and enzyme inhibitors were studied, 50 tl of the solutions of corresponding substances ( cm) was added before the beginning of experiments. Cytogenetic analysis The whole blood of FA patients was cultured according to the standard method [11]. The cultured mixture contained 10% blood, 15% fetal calf serum, and 75% medium 199. Lymphocyte mitosis was stimulated by the addition of phor- TABLE 2. Luminol-Dependent CL of a Total Fraction of Blood Leukocytes of FA Patients and Normal Donors (PMN) (mv ) per 106 cells Without Patients/donors stimulus Con A SiO, Latex FA patients D-va 17,300 (2,900) 4,300 (740) 15,600 (2,600) 19,000 (3,300) Sh-ov 1,100 (300) NA. 5,000 (1,400) 6,400 (1,800) M-ko NA. 1,200 (710) 4,300 (2,500) NA. O-ch 500 (220) 1,200 (560) 6,000 (2,700) NA. Mi-ov 1,400 (420) NA. 11,100 (3,300) 12,200 (3,700) Dz-ov 230 (90) NA. 2,800 (1,100) 3,400 (1,300) G-ov 590 (200) NA. 7,200 (2,500) 14,000 (4,900) M-ovaA. 1,300 (240) NA. 3,600 (690) 7,300 (1,400) M-ova E. 1,200 (290) NA. 6,300 (1,500) 15,000 (3,700) Average 3, ,900 ± 4,070 11,000 ± 5,500 Donorsb (190) 360 (250) 1,140 (800) 1,010 (710) (210) 310 (220) 760 (530) 990 (690).3 14 (10) 400 (280) 210 (150) NA (40) 600 (420) 1,160 (810) NA. Average ± ± 440 1,000 ± 15 (Icuk) values are given in parentheses.!(pmn) values of FA patients are calculated from equation 3. bnot analyzed. For donors I(PMN) values are equal to 358 Journal of Leukocyte Biology Volume 52, September 1992

3 TABLE 3. Luminol-Dependent CL of Bone Marrow Leukocytes of FA Patients and Normal Donors (PMN) (mv) per 106 cells Patients/donors % PMNs Without stimulus Con A SiO, Opsonized zymosan FA patients D-va M-ko Donors (100) 240 (100) 290 (200) 110 (80) 310 (60) 360 (150) 140 (100) 210 (150) 1000 (180) 620 (260) 540 (380) 170 (120) 1600 (290) 620 (260) 510 (360) 830 (580) (kuk) values are given in parentheses. bol myristate acetate at the beginning of cultivation. The total period of cultivation at 37#{176}Cwas 56 h. At 2.5 h before fixation of the cells, 0.1 ml of a colchicine solution of final concentration 0.25 mg/mi was added. Fixation was performed by a standard cytological procedure. The cells were exposed to M KC1 solution for mm and fixed in ethanol-acetic acid solution (3:1 v/v). RESULTS Luminol- and lucigenin-dependent CL of blood and bone marrow leukocytes The results of measurements of luminol- and lucigenindependent CL of nonstimulated and stimulated blood and bone marrow leukocytes of healthy donors and FA patients are given in Tables 1-4. It is known [12] that PMNs make a major contribution to the total CL response of whole blood or isolated cell suspensions. Therefore, the intensity of nonstimulated and stimulated CL depends linearly on the amount of PMNs in cell suspensions. Because of this, the CL intensity induced by PMNs can be calculated from the CL intensity of the total leukocyte fraction, using equation 3: -1(PMNs) = (leuk) 100 V vcn where v is the cell suspension volume, V the volume of the cuvette, C the leukocyte concentration, and n the PMN content (%). The data in Table 1 confirm the above proposal. In all cases the CL intensity of mononuclear leukocytes did not exceed 10% of the intensity of the total leukocyte fraction, indicating a predominant contribution of PMNs in the CL response. As could be expected, the CL intensity of the whole blood was much lower than that of the total leukocyte fraction because of inhibition by erythrocytes and other blood components. Opsonised zymosan (acting through the receptor of the C3 component of complement), Con A (acting through lectin-like receptors), latex, and SiO2 (stimulating the phagocytes without receptor activation) significantly enhanced the luminol-dependent CL induced by blood leukocytes of FA patients and normal donors (Tables I and 2). The same was true for bone marrow leukocytes (Table 3), although in this case the levels of stimulated and nonstimulated CL were much lower. The luminol-dependcnt CL of leukocytes of FA patients always exceeded that of normal donors. Similar findings were obtained for the lucigcnin-dependcnt CL of leukocytes; however, here the difference between the CL levels for leukocytes of FA patients and donors was significantly smaller (Table 4). Effects of antioxidative enzymes, free radical scavengers, and enzyme inhibitors on chemiluminescence of leukocytes The antioxidative enzyme SOD inhibited the luminoldependent CL of blood leukocytes of FA patients by 5-15% and that of blood leukocytes of normal donors by 60-70% (Table 5). Similarly, SOD strongly inhibited the lucigenindependent CL of leukocytes (data not shown). Catalase, another major antioxidative enzyme, had little effect on the luminol-dependent CL of leukocytes of Fanconi patients; in three cases it even slightly enhanced the CL intensity. The inhibitory effect of catalase on the luminol-dependent CL of lcukocytcs of normal donors was 40-65%. The bioflavonoid rutin, which is a free radical scavenger and an iron chclator simultaneously [13], strongly inhibited the luminol-dcpcndent CL of FA leukocytes in a concentration-dependent manner (Fig. 1) and was a less effective inhibitor in the case of normal donors (Table 5). Aspirin and indomethacin, inhibitors of cyclooxygenase, and taurin, an inhibitor of myeloperoxidase, inhibited the luminol-dcpendcnt CL of both FA patients and donors by only 10-30%, but NDGA, an inhibitor of lipoxygenase and (3) a free radical scavenger, was a very strong inhibitor of luminol-dependent CL of leukocytes of both FA patients (Fig. 2) and normal donors (Table 5). Therapeutic treatment Patients were given 10 mg rutin/kg body weight three times per day. Long-term treatment with rutin was possible only TABLE 4. Comp Latex-Stimulate Patients/donors FA patients arison of Luminol- and Lucigenin-Dependent CL of d Leukocytes of FA Patients and Normal Donors (i,minoi) (PMN) (mv) per 106 cells l,,rig,nm) kignin) (Iu,mnoi) D-va 20,600 1, M-ko 29, Dz-ov 3, Donors G-ov 15,700 4, , , , The luminol- and lucigenin-dependent CL data were obtained for samples other than those cited in Table 2. This explains the difference in the luminol-dependent CL data presented in Tables 2 and 4. Korkina et al. Oxygen radicals in Fanconi anemia 359

4 TABLE 5. Effects of Antioxidative Enzymes, Free Radical Scav and Enzyme Inhibitors on the Luminol-Dependent CL of Blood Leukocytes Concentration Inhibition (%) engers, Inhibitor (tim) FA patients Donors SOD 20 (ig/ml) Catalase 50 (ig/ml) Rutin NDGA Aspirin na. Indomethacin Taurin (or activation) Mannitol 10, In three cases there was activation by 5-15% TABLE 6. Hematological Parameters of FA Patients Before and After Patient D-va M-va M-va E. 5 A. Refractory to splenectomy and treatment with prednisolone and methandrostenolone. Twins. Before Time analysis 11 months 18 months Long-Term Rutin Treatment of treatment Before treatment Before treatment Hb Leukocytes (% neutrophils) Platelets 3.4 (17) 6.2 (65) 3.8 (42) 5.2 (59) 3.0 (30) 4.1 (56) 1.3 (26) 2.3 (50) Single 50 for three patients (for 3 years in the case of one female child and for 1 year in the case of two female children). No side effects were observed in patients during the long-term rutin treatment. Hematological parameters and CL data for FA patients subjected to rutin treatment are given in Tables 6 and 7. Chromosome aberrations in lymphocytes of peripheral blood were measured for three FA patients to whom rutin was administrated for. It was found that chromosomal aberrations diminished from 15.5 to 7.7% for one patient (D-va) and did not change essentially for two others (M-va A. and M-va E.). There were improvements in physical conditions of FA patients. For example, D-va s height increased by 8 cm and I(.L( 102) IHV Fig. 1. Effect of rutin on luminol-dependent CL by the latex-stimulated leukocytes of Fanconi patients. In a luminometer cell 50 sl of leukocyte suspension (2 x 106 cells/mi), 50 1sM luminol, and 0.85 ml of HBSS were mixed. After 5 mm, 50 tl of 0.5% latex suspension in 0.9% NaCI solution was added and the light emission was recorded continuously. (1) Without rutin; (2-4) 10, 20, and 100 sm rutin were added, respectively; (5) without latex. body weight by 7.5 kg in 1.5 years. The secondary sexual indications DISCUSSION appeared. The results of the study of luminol-dependent CL suggest that blood leukocytes of FA patients produce an enhanced level of active oxygen species in comparison with those of normal donors (Table 2). A similar but weaker tendency is observed for bone marrow leukocytes (Table 3). The stimulation of luminol-dependent CL by zymosan, latex, SiO2, and Con A indicates that an increase in oxygen radical production by FA leukocytes may be a consequence of binding C3 or lectin-like receptors or of nonreceptor stimulation. The level of lucigenin-dependent CL was also higher for blood leukocytes of FA patients than for normal donors, but the most significant difference was observed for a ratio of the intensities of lucigenin- and luminol-dependent CL (Table 4). It is thought [2] that lucigenin-dependent CL is a specific test for superoxide production, whereas luminol-dependent CL is apparently relevant to the production of all active oxygen species. Therefore, one may conclude not only that the leukocytes of FA patients produce an enhanced amount of oxygen species but also that these species contain much more active oxygen radicals that are different from superoxide ion. This proposal was confirmed in the study of inhibitory effects of SOD and hydroxyl radical scavengers on the production of oxygen radicals by Icukocytes of normal donors and FA patients (Table 5). SOD inhibited the luminoldependent CL of normal leukocytes by 60-70% and that of FA leukocytes by 5-15%. These data support the conclusion that superoxide production is predominant in normal leukocytes and relatively unimportant in FA leukocytes. On the other hand, mannitol, a classical hydroxyl radical scavenger, inhibited the luminol-dependent CL of FA leukocytes by 50% and did not affect that of normal leukocytes. This indicates that FA leukocytes produce a significant amount of hydroxyl radicals and normal leukocytes do not. Two other hydroxyl radical scavengers, rutin and NDGA, were also very efficient inhibitors of oxygen radical production by FA leukocytes (Table 5). However, they also inhibited 8 mi oxygen radical production by normal leukocytes, being simultaneously a chelator and superoxide scavenger (rutin) and a lipoxygenase inhibitor (NDGA). Catalase inhibited luminol-dependent CL of normal leukocytes (by 40-65%) and did not affect it or even enhanced it in the case of FA leukocytes (Table 5). The last is probably explained by the fact that exogenous catalase can protect the leukocyte NADPH 360 Journal of Leukocyte Biology Volume 52, September 1992

5 TABLE 7. Generation of Oxygen Radicals by Blood Cells and Antioxidant Parameters of the Blood..of FAPatients Luminol- dependent CL intensity (pe r 106 neutrophils) Time of SOD activity Catalase activity Patient analysis Spontaneous SiO, Latex (arbitrary units) (arbitrary units) D-va Before treatment M-va E. Before treatment M-va A. Before treatment 17, , , ,600 1,230 6,300 4,830 3,600 1,920 19,000 2,180 15,000 6,490 7,300 2, oxidase from inactivation by a large flux of hydrogen peroxide. The cyclooxygenase inhibitors aspirin and indomethacin only slightly affected luminol-dependent CL for both normal and FA leukocytes. The same is true for the myeloperoxidase inhibitor taurin. The overproduction by FA leukocytes of oxygen species containing enhanced amounts of active hydroxyl or hydroxyllike radicals may be a result of a change in the activation or activity of NADPH oxidase, a decrease in the cellular antioxidant defense system, and an increase in the level of cellular free iron ions catalyzing the superoxide-driven Fenton reaction (reactions 1 and 2). An increase in the level of free iron may be a consequence of microlysis of erythrocytcs and deficiency in the proteins binding heme iron - for example, haptoglobin [14] -or of iron release from iron-transporting 1CL( 102) fly proteins by oxygen radicals [1, p. 151]. The origin of these changes may be some inherited chromosome deformation in an FA patient. The deficiency in erythrocyte SOD in the blood of FA patients has already been shown [6-9] and is supported by our data (Table 6), which in general show 25-30% lower than normal SOD activity. However, this decrease in SOD activity apparently cannot be the only origin of the observed overproduction of oxygen radicals, and the other factors mentioned above - the change of activation and activity of NADPH oxidase and an increase in the free iron level - must be important. It should also be stressed that although the enhanced level of oxygen radical production by blood leukocytes was observed for all FA patients, there was no straight correlation between oxygen radical overproduction and FA phenotype because the rate of oxygen radical production changed during FA development. Thus, at the terminal stage of the disease the level of oxygen radical production sharply decreased, as it does in the terminal stages of other pathologies such as sepsis and burns. These results suggest that hydroxyl radical scavengers and chelators can be used for the treatment of FA patients. Because rutin, a nontoxic natural bioflavonoid (vitamin P), efficient hydroxyl radical scavenger, and chelator, was found to be an efficient inhibitor of luminol-dependent CL produced by FA leukocytes (Table 5), we administrated it to three FA patients. The long-term treatment of the patients with rutin substantially decreased oxygen radical production by leukocytcs, diminished to some degree the amount of chromosomal aberrations, and improved hematologic characteristics of patients. It is interesting that rutin did not change SOD activity, supporting the idea that although SOD deficiency may be a primary inherited defect, chromosome aberrations are at least in part induced by the overproduction of active oxygen radicals. The patients health was essentially improved. Thus rutin may be recommended as an effective drug for the treatment of FA patients. ACKNOWLEDGMENTS [ mi n Fig. 2. Effects of enzymes, enzyme inhibitors, and rutin on luminoldependent CL by SiO2-stimulated leukocytes of Fanconi patients. In a luminometer cell 50 il of leukocyte suspension (2 x 106 cells/mi), 50 sm luminol, 0.85 ml of HBSS, and 50 sl of inhibitor solution were mixed. After 5 mm, 50 sl of SiO2 suspension in 0.9% NaCI solution was added and the light emission was recorded continuously. (1) 100 M taurin was added; (2) control; (3) 50 sg/ml catalase was added; (4) 20 ag/ml SOD was added; (5) 10 tm NDGA was added; (6) 100 tm rutin was added. We are grateful to Dr. A. D. Durnev for performing cytogenetic analysis. REFERENCES 1. Afanas ev, LB. Superoxide Ion: Chemistry and Biological Implications, Vol. 2. Boca Raton, FL: CRC Press, Chapter 4, Gyllenhammar, H. Lucigenin chemiluminescence in the assessment of neutrophil superoxide production. j Immunol. Methods 97, 209, Meretcy, K., Antal, M., Rozsnyay, Z., Bohn, U., Elekes, E., and Geuti, G. PHA- and Con A-induced chemiluminescence of hu- Korkina et a!. Oxygen radicals in Fanconi anemia 361

6 man blood mononuclear cells and granulocytes in luminol or lucigenin. Inflammation 11, 417, Bellavite, P. The superoxide-forming enzymatic system of phagocytes. Free Radicals Biol. Med. 4, 225, Schroeder-Kurth, TM., Auerbach, AD., and Obe, G. (Eds.) Fanconi Anemia, Clinical, Cytogenetic and Experimental Aspects. Berlin: Springer-Verlag, Joenje, H., Eriksson, A.W., Frants, R.R., Arwert, F., and Houwen, B. Erythrocyte superoxide dismutase deficiency in Fanconi s anaemia. Lancet 1, 204, Mavelli, I., Ciriolo, MR., and Rotilio, G. Superoxide dismutase, glutathionc peroxidase and catalase in oxidative hemolysis. A study of Fanconi s anemia erythrocytes. Biochem. Biophys. Res. Common. 106, , Okahata, S., Kobayashi, Y., and Usui, T. Erythrocyte superoxide dismutase in Fanconi s anaemia. C/in. Sci. 58, 173, Yoshimitsu, K., Kobayashi, Y., and Usui, T. Decreased superoxide dismutase activity of erythrocytes and leukocytes in Fanconi s anemia. Acta HaemotoL 27, 208, Scarpa, M., Rmgo, A., Momo, E, Isacchi, G., Novelli, G., and Dallapiccola, B. Increased rate of superoxide ion generation in Fanconi anemia erythrocytes. Biochem. Biophys. Res. Commun. 130, 127, Korkina, L.G., Durnev, AD., Suslova, TB., Cheremisina, Z.P., Daugel-Dauge, NO., and Afanas ev, lb. Oxygen radical mediated mutagenic effect of asbestos on human lymphocytes: suppression by oxygen radical scavengers. Mutat. Ies. 265, 245, Lindena, J., and Purkhardt, H. Separation and chemiluminescence of human, canine and rat polymorphonuclcar cells, j ImmunoL Methods 115, 141, Afanas e lb., Dorozhko, Al., Brodskim, A.V., Kostyuk, V.A., and Potapovitch, A.!. Chelating and free radical scavenging mechanisms of inhibitory action of rutin and quercetin in lipid peroxidation. Biochein. Pharmacol. 38, 1763, Gutteridge, J.M.C., Westermarck, T., and Halliwell, B. Oxygen radical damage in biological systems. In Free Radicals, Aging and Degenerative Diseases. New York: Alan R. Liss, p. 99, Journal of Leukocyte Biology. Volume 52, September 1992