Role of antioxidant enzymes in brain tumours

Clinica Chimica Acta 296 (2000) 203–212 www.elsevier.com / locate / clinchim

Short communication

Role of antioxidant enzymes in brain tumours Gayathri M. Rao a , Ashalatha V. Rao a , Annaswamy Raja b , b c, Suryanarayana Rao , Anjali Rao * a Department of Biochemistry, Kasturba Medical College, Mangalore, India Department of Neurology, Kasturba Medical College and Hospital, Manipal 576 119, Karnataka, India c Department of Biochemistry, Kasturba Medical College and Hospital, Manipal 576 119, Karnataka, India b

Received 2 November 1999; received in revised form 31 January 2000; accepted 9 February 2000

Abstract Erythrocyte antioxidant enzymes were analysed in 100 patients with intracranial neoplasm and in 47 controls. There was a significant decrease in RBC glutathione reductase (GRx) and superoxide dismutase (SOD) activity in most types of brain tumor cases. Patients with acoustic neurinoma showed a significant reduction in selenium-dependent glutathione peroxidase (Se-GPx) activity. A decrease in catalase (CT) activity was seen in most of the brain tumor patients but remained statistically insignificant when compared to controls. A significant increase in plasma ceruloplasmin concentration was observed in patients with glioma. These enzymes were also studied in 27 post-treatment cases. GRx activity returned to normal levels in these patients. RBC SOD and plasma ceruloplasmin levels showed a tendency to return to normal. Hence, a marked decrease in the antioxidant enzymes may have a role in the genesis of considerable oxidative stress in patients with brain tumors.  2000 Published by Elsevier Science B.V. All rights reserved. Keywords: Antioxidant enzymes; Ceruloplasmin; Brain tumors

1. Introduction Superoxide dismutase (SOD) catalase (CT) and glutathione peroxidase (GPx) along with glutathione reductase (GRx) constitute a supportive team of enzymes *Corresponding author. Tel.: 1 91-082-527-1201, ext. 2326; fax: 1 91-082-527-0061. E-mail address: [email protected] (A. Rao) 0009-8981 / 00 / $ – see front matter  2000 Published by Elsevier Science B.V. All rights reserved. PII: S0009-8981( 00 )00219-9

7269

DTD410 18 2 APR 2 2000 16:55:18.62 v1.4.129

204

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

which provide defense against the reactive intermediates of dioxygen reduction. These enzymes are cooperative in several aspects. At the most obvious level, SOD converts superoxide radical (O 2 2 ) into hydrogen peroxide (H 2 O 2 ) and the latter must then be disposed of by CT and peroxidase. Somewhat more subtle operations involve protection by CT and peroxidase, of the SOD against inactivation by H 2 O 2 . Reciprocally, SOD protects the GPx against inhibition by O 2 [1]. Ceruloplasmin, which is an acute-phase reactive protein synthesized in the liver also acts as an antioxidant. It is proposed that oxygen-derived free radicals (ODFR) play a key role in human cancer development [2]. Subnormal activities of SOD have been reported in tumors of GIT [3], multiple myeloma [4], and endometrial cancer [5]. There are reports of elevated SOD activity in Alzheimer’s disease [6], colon tumor cell lines [7] and hepatocellular carcinoma [8]. A significant increase in CT has been reported in various cancers such as gastric cancer [9], carcinoma of bladder [10,11] and colon tumor cell lines [7]. In gynecological cancers [5], hepatocellular carcinoma [8] and lung cancers [12], this antioxidant enzyme showed a subnormal activity. Altered activity of GPx has been reported in various cancer patients [4,5,7,13–15]. A significant drop in GRx activity has been observed in carcinoma of the uterine cervix [5]. Increased ceruloplasmin levels in serum / plasma have been reported in various types of malignancies such as lung cancer [16] and gynecological tumors [17–20]. Since there are low activities of protective antioxidant enzymes accompanied with a high ratio of membrane surface area compared to cytoplasm and nonreplicability of neuronal cells in comparison with other organs of the body, the nervous system may be especially vulnerable to ODFR-mediated injury [2]. Very few reports are available on the involvement of free radicals in intracranial tumor development and on the vulnerability of the brain to free radical ill-effects. Therefore the present study was undertaken to assess the erythrocyte antioxidant enzymes and plasma ceruloplasmin levels in patients with brain tumors.

2. Materials and methods Blood samples were obtained from 100 patients with intracranial neoplasm between 18 and 80 years of age. All were histologically confirmed for their neurological status. Age- and sex-matched healthy controls were also studied during this period (n 5 47). The various types of tumors included in this study were glioma, meningioma, acoustic neurinoma and other types (secondary tumors, tuberculoma, lymphoma, ventricular tumor, craniopharyngioma, Table 1). Post-operative blood samples

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

205

Table 1 Clinical profile of patients with intracranial tumors Clinical condition

Total

Males

Females

Control Glioma Glioblastoma Astrocytoma grade I & II Astrocytoma grade III & IV Ependymoma Meningioma Acoustic neurinoma Other types Craniopharyngioma Secondary tumors Lymphoma Ventricular tumors Tuberculoma Follow-up patients: Glioma Meningioma Acoustic neurinoma

47 41 5 5 30 1 31 17 18 5 6 2 2 3

31 26 1 3 21 1 12 8 14 5 4 1 2 2

16 15 4 2 9 0 19 9 4 0 2 1 0 1

22 4 1

12 1 1

10 3 0

were obtained from the patients when they were re-examined at a follow-up study.

2.1. Sample collection Blood was collected into EDTA tubes. The erythrocyte suspension was prepared according to the method of Beutler et al. [21]. It was immediately centrifuged under refrigeration at 3000 3 g for 10 min. Plasma and buffy coat were carefully removed and the separated cells washed thrice with cold saline phosphate buffer, pH 7.4 (sodium phosphate buffer containing 0.15 mol / l NaCl). The erythrocytes were then suspended in an equal volume of physiological saline and stored as 50% cell suspension at 48C until use. Appropriately diluted hemolysates were then prepared from the erythrocyte suspension by the addition of distilled water, for the estimation of SOD, CT, GPx and GRx activity. SOD estimation was performed according to the method of Beauchamp and Fridovich [22]. The basis for the SOD assay was inhibition of the reduction of nitroblue tetrazolium by superoxide radicals generated by the illumination of riboflavin in the presence of oxygen and electron donor, methionine. The procedure adapted for CT was that of Brannen et al. [23] with modifications. The assay was based on the disappearance of H 2 O 2 in the presence of the

206

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

enzyme source at 258C. GPx was estimated by the method of Paglia and Valentine [24] as modified by Lawrence and Burk [25]. The oxidized glutathione (GSSG) formed by the action of GPx reacted with GRx. The decrease in absorbance at 340 nm due to the depletion of the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) for a period of 5 min was recorded. GRx activity was estimated by the procedure of Horn and Burns [26]. This enzyme catalyses the reduction of GSSG to reduced glutathione (GSH) in presence of NADPH. The decrease in the absorbance at 340 nm for a period of 5 min was recorded. Plasma ceruloplasmin levels were determined by the diamine oxidase method [27]. The method is based on the property of ceruloplasmin to catalyze the oxidation of colorless para-phenylene diamine to a blue- or violet-colored product which is estimated spectrophotometrically. The hemoglobin content of the erythrocytes was determined by the cyanmethhemoglobin method [28]. Suitable controls were used throughout the enzymatic assays. Statistical analysis used unpaired and paired t-tests and nonparametric tests.

3. Results RBC SOD activity was significantly decreased in most types of intracranial neoplasm. Though there was a decrease in mean CT activity, it remained statistically insignificant compared to the control. There was no significant difference in Se-GPx activity except in acoustic neurinoma (P , 0.01), where it has decreased in comparison with controls. GRx activity showed a highly significant decrease in all types of brain tumor cases. There was a marked increase in plasma ceruloplasmin concentrations in glioma patients (Table 2). A significant reduction in RBC GRx activity was observed both in benign and malignant brain tumor patients when compared to controls. A significant decrease in SOD activity was noted in RBC of patients with malignant brain tumor in comparison to the normal levels (Table 3). GRx activity returned back to normal levels in post-treatment cases and was significantly different from that in the corresponding pre-operative cases. Though the RBC SOD activity remained low even after treatment, it showed a tendency to revert back to normal as did the plasma ceruloplasmin levels in glioma patients (Table 4 and 5).

4. Discussion Many studies illustrate the well-known observation that cell transformation alters cell responsiveness to oxidative stress [29]. Changes in the antioxidant defense system, such as SOD, have been widely described in cancerous cells.

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

207

Table 2 Antioxidant enzyme levels in brain tumors (mean6S.D., range in parentheses; n, sample size) Control Superoxide dismutase

Glioma

Meningioma

518661604

427361183

c

e

440761244

(2434–8902)

(1882–7250)

(1943–6818

Acoustic neurinoma 391061373

c

(1272–5949)

Other types 471161093 (1150–6853)

n 5 47

n 5 40

n 5 28

n 5 17

n 5 18

165 3006132 200

124 400660 600

105 300637 700

93 700626 100

100 600641 500

(U/g Hb)

(14 000–422 000)

(52 800–257 600)

(3470–173500)

(51 500–146 700)

(28 000–186 800)

(Mann–Whitney U-test)

n 5 46

n 5 40

n 5 27

n 5 17

n 5 18

7.9363.34

6.5962.98

7.3562.95

5.5962.17 d

7.0862.27

peroxidase (Se)

(0.83–16.53)

(1.12–17.92)

(2.73–12.92)

(2.29–9.8)

(2.6–11.24)

(mmol NADPH

n 5 47

n 5 40

n 5 28

n 5 17

n 5 15

Glutathione

1.2960.40

0.3260.26 a

0.3760.32 a

0.4460.24 a

0.3360.31 a

reductase

(0.8–2.45)

(0.00–0.78)

(0.0–0.96)

(0.00–0.77)

(0.00–0.82)

(mmol NADPH

n 5 47

n 5 41

n 5 31

n 5 17

n 5 18

Ceruloplasmin

18.2267.85

24.3369.30 b

21.2768.30

20.7668.35

22.2067.57

(mg/dl)

(2.06–38.58)

(6.3–50.78)

(8.1–38.63)

(9.48–37.45)

(4.5–34.6)

n 5 47

n 5 38

n 5 27

n 5 17

n 5 17

(U/g Hb) Catalase

Glutathione

oxidized/min per g Hb)

oxidized/min per g Hb)

a

c

Entries with different superior letters have the following P values: P , 0.0001, b P , 0.001, P , 0.005, d P , 0.01, e P , 0.05 (unpaired t-test).

Table 3 Comparison of antioxidant enzyme levels in benign and malignant brain tumours (mean6S.D., range in parentheses; n, sample size)

Control

Malignant

Benign

Superoxide

Catalase

Glutathione

Glutathione

Ceruloplasmin

dismutase

(U/g Hb)

peroxidase (Se)

reductase

(mg/dl)

(U/g Hb)

(Mann–Whitney

(mmol NADPH

(mmol NADPH

U-test)

oxidized/min per g Hb)

oxidized/min per g Hb)

518661604

165 3006132 200

7.9363.34

1.2960.40

18.2267.85

(2434–8902)

(14 027–422 045)

(0.83–16.53)

(0.8–2.45)

(2.06–38.58)

n 5 47

n 5 46

n 5 47

n 5 47

n 5 47

435561304 d

118 000656 500

7.2862.64

0.3160.26 a

24.1568.89 c

(1150–7250)

(11 076–25 672)

(1.12–17.25)

(0.00–0.78)

(6.32–36.74)

n 5 48

n 5 46

n 5 46

n 5 48

n 5 45

445561132

102 100632 800

6.8863.43

0.3660.32 b

21.9868.49 e

(1943–6818)

(28 022–186 765)

(2.73–12.92)

(0.00–0.96)

(4.5–38.63)

n 5 38

n 5 38

n 5 40

n 5 43

n 5 39 a

Entries with different superior letters have the following P values: controls vs. malignant, P , 0.0001; b controls vs. benign, P , 0.0001; c controls vs. malignant, P , 0.001; d controls vs. malignant, P , 0.01, e controls vs. benign, P , 0.05. Benign vs. malignant, not significant (Student’s unpaired t-test).

208

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

Table 4 Antioxidant enzyme levels in post treatment cases (mean6S.D., range in parentheses; n, sample size)

Superoxide dismutase (U / gHb) Catalase a (U / gHb) Glutathione peroxidase (Se) (mmols NADPH oxidized / min per g Hb) Glutathione reductase (mmols NADPH oxidized / min per g Hb) Ceruloplasmin (mg / dl)

Glioma

Meningioma

Acoustic neurinoma

42346957 (1947–6160) n 5 22 88 350624 910 (56 538–133 453) n 5 22 6.5662.6 (1.36–10.71) n 5 22

428061088 (3316–5725) n54 90 500625 700 (68 850–125 467) n 5 24 7.262.93 (4.11–10.53) n54

3270

1.1860.31* (0.92–2.19) n 5 22

1.2660.64 (1.05–2.16) n54

1.43

21.32610.25 (8.3–46.69) n 5 22

25.6610.00 (11.85–35.55) n54

28.12

n51 79300 n51 8.04 n51

n51

n51

a

(Mann–Whitney U-test and Wilcoxon rank sum test. *P , 0.0001 with respect to pre-operative values (Student’s unpaired and paired t-test).

Development of ischemia in or around the tumor mass may be one of the reasons for the development of oxidative stress. The iron from brain cells can be easily released on injury. If the free iron content of the system were raised by cell injury, an ascorbate–iron salt mixture would be expected to promote lipid peroxidation and hydroxyl radical formation. This primary radical damage can result in cytotoxic edema, which can have secondary effect by causing vascular compression [30]. Some products of lipid peroxidation are diffusible and can spread the damage far beyond the site of the original free radical attack. The changes in the erythrocyte and plasma may reflect the oxidative stress in the brain due to the development of tumor mass. In the present study, a significant reduction in SOD activity was seen in glioma, meningioma and acoustic neurinoma patients. These changes in erythrocytes may reflect the oxidative stress in brain tumors. Levchenko [31] reported a reduction of SOD activity with increase in malondialdehyde (MDA) levels in meningioma patients, both in blood and tumor tissues. Pu et al. [32] reported a decrease in SOD activity in tumor tissue in descending order: meningioma, low-grade astrocytoma, highgrade astrocytoma and medulloblastomas. Rolando et al. [33] reported a 50% decrease in SOD activity in glioblastoma multiforme. In the present study, SOD activity tended to increase after surgery in the follow-up cases, although the

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

209

Table 5 RBC GRx activity in pre- and post-operative brain tumors Diagnosis S. no.

Sex

Samples collected after surgery (days)

GRx (mmols NADPH oxidized / min / g Hb) Pre-operative

Post-operative

34 42 40 40 69 120 53 15 37 46 34 334 26 36 39 92 32 51 51 26 40 48

0.75 0.82 0.49 0.28 0.48 0.00 0.78 0.00 0.43 0.47 0.73 0.24 0.60 0.00 0.00 0.00 0.74 0.50 0.00 0.28 0.33 0.26

1.06 1.25 1.04 1.07 2.19 1.60 1.14 1.07 1.29 1.37 1.15 1.56 1.15 0.92 1.12 1.02 1.15 1.02 1.14 1.10 0.57 0.96

45 38 39 46

0.00 0.00 0.23 0.23

1.05 2.16 1.67 1.15

Acoustic neurinoma 27 M 730

0.29

1.43

Glioma 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Fa M M M F F F F F M M M M M M F F F M F M F

Meningioma 23 M 24 F 25 F 26 F

Mean6S.D.

Pre-operative 0.3760.29

Post-operative 1.1860.31*

Pre-operative 0.1360.12 Post-operative 1.2660.64

a

M, male; F, female. P , 0.0001 when compared with corresponding pre-operative values.

increase was not significant compared to the corresponding pre-operative values. However, it indicates a tendency to return to the normal state. Mean glutathione peroxidase activity in pre-operative cases remained in the same range as the controls, except in acoustic neurinoma where there was a significant decrease in Se-GPx. This may be due to the persistent oxidative stress.

210

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

The decrease in RBC GRx may be due to the ill effects of free radicals on the enzyme. Again, ischemia at the center or around the tumor may lead to oxidative stress. Dzhandzhgava and Shakarishivilli [34] reported a decrease in GRx activity in blood, serum and cerebrospinal fluid in ischemic brain diseases. A decrease in GRx activity may lead to a decrease in reduced glutathione. The GSH / GSSG in normal cells is kept high, because of the reduction of GSSG back to GSH by GRx enzyme. Reduced glutathione is a co-factor for several enzymes in different metabolic pathways. Moreover, it acts as a scavenger of hydroxyl [OH] radical and singlet oxygen and it can reactivate some enzymes that have been inhibited by exposure to high oxygen concentration. Presumably the oxygen causes oxidation of essential –SH groups on the enzymes which are regenerated on incubation with GSH. GSSG inactivates a number of enzymes, probably by forming mixed disulfides. It has been shown to inhibit protein synthesis in animals cells [35]. If a tissue is exposed to a large flux of hydrogen peroxide and / or –OH, a point might be reached at which GSH / GSSG cannot be maintained at its normal ratio. This can aggravate the oxidative stress developed. Moreover, Buckmann and co-workers [36] have reported that cytotoxicity was potentiated by the inhibition of glutathione reductase. In the follow-up study, GRx activity returned to normal level. This may indicate a relief from the oxidative stress after the resection of the tumor mass. Increase in plasma ceruloplasmin levels observed in glioma patients may be due to its increased production of ceruloplasmin by liver, as observed in Hodgkin’s disease [37], in a counter action of the cells to the ill effects of free radicals. In the follow-up study, plasma ceruloplasmin levels showed a tendency to return to normal. Thus, the results of the present study support the concept of involvement of free radicals in intracranial neoplasms. The decrease in SOD and GRx observed, may be the major factors responsible for oxidative stress. However, after treatment a significant relief from the pre-operative condition facilitated by improvement in the antioxidant profile, indicates a tendency to revert back to normal status. These aspects probably denote a good prognosis in such patients suffering from brain tumors.

References [1] Blum J, Irvin Fridovich I. Inactivation of glutathione peroxidase by superoxide radical. Arch Biochem Biophys 1985;240:500. [2] Evans PH. Free radicals in brain metabolism and pathology. In: Cheeseman KH, Slater TF, editors. Free radicals in medicine. Brit Med Bull New Jersey: Churchill Livingston 1993;49(3):577. [3] Chevaris S, Andial T, Benke K, Strenger A. Free radical reactions and cancer. Vopr Med Chem 1992;38:4.

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

211

[4] Zima T, Spi C, Ka I, Stipek S. Lipid peroxidation and activity of antioxidative enzymes in patients with multiple myeloma. Cas Lek Ces 1996;135:14. [5] Punnonen R, Kudo R, Punnonen K. Activity of antioxidant enzyme and lipid peroxidation in endometrial cancer. Eur J Cancer 1993;29A:226. [6] Lovell MA, Ehmann WD, Butten SM, Markesbery WR. Elevated thiobarbituric acid reactive substances and antioxidant enzyme activity in the brain the Alzheimer’s disease. Neurology 1995;45:1594. [7] Bravard A, Beaumatin J, Dussaulx E, Lesffleur Y, Zweibaum A, Luccioni C. Modification of the antioxidant metabolism during proliferation and differentiation of colon tumor cell lines. Int J Cancer 1994;59:843. [8] Haung CC. Impaired SOD activity and decreased content of lipid peroxides in hepatocellular carcinoma tissue. Chung Hua Wai Ko Tsa Chin 1991;29:195. [9] Beno I, Volkovov AK, Staruchov AM. Gastric mucosal antioxidant activity in patients at increased risk of gastric cancer. Neoplasm 1993;40:315. [10] Durak I, Ormeci N, Akyol O, Canbolt O, Kavutcu M, Bulbul M. Adenosine deaminase, 5-nucleotidase, xanthine oxidase, superoxide dismutase and catalase activity in gastric juices from patients with gastric cancer, ulcer and atrophic gastritis. Dig Dis Sci 1994;39:721. [11] Singh SV, Xu BH, Tkalccvic GT, Gupta V, Roberts B, Quiz P. Glutathione linked detoxification pathway in normal and malignant human bladder tissue. Cancer Lett 1994;77:15. [12] Tatig ZP. Observation on the activity of SOD and CT of alveolar macrophage in patients with lung cancer. Chung Hua Chick Ho Ho Hu Hsi Tsa Chih 1991;14:213, 255. [13] Punnonen K, Ahutopa M, Ashishi K, Hyoty M, Kudo R. Antioxidant enzyme activities and oxidative stress in human Sˇ breast cancer. J Cancer Res Clin Oncol 1994;120:374. [14] Oberli SAE, Joncourt F, Stadler M et al. Parallel assessment of glutathione based detoxifying enzymes. O6-alkylguanine DNA alkyltransferase and P-glycoprotein as indicators of drug resistance in tumor and normal lung of patients with lung cancer. Int J Cancer 1994;59:629. [15] Wasowicz W, Gromadzinska J, Skodowska M, Popadiuk S. Selenium concentration and GRx activity in blood of children with cancer. J Trace Elem Electrolytes Health Dis 1994;8:53. [16] Knett P, Aroma A, Matela J et al. Serum ceruloplasmin and risk of cancer in Finland. Br J Cancer 1992;65:292. [17] Chan A, Wong F, Arumanayagam M. Serum ultra filterable copper, total copper and ceruloplasmin concentration in gynaecological carcinomas. Ann Clin Biochem 1993;30:545. [18] Kirchheimer JC, Kolbl H, Tatra G, Binder BR. Relationship between plasma levels of components of the system and acute-phase reactants in patients with uterine malignancies. Eur J Clin Invest 1990;20:79. [19] Schultek T, Tode B, Wood WG. Ceruloplasmin in diagnosis of gynaecological tumors. J Clin Chem Clin Biochem 1986;24:179. [20] Tatra B. Acute phase markers in gynaecological tumors. Stahlentherapie 1985;161:487. [21] Beutler E, West C, Blume KG. Removal of leucocytes and platelets from whole blood. J Lab Clin Med 1976;88:328. [22] Beauchamp C, Fridovich I. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem 1971;44:276. [23] Brannen TS, Marker HS, Raes IP. Regional distribution of catalase in the adult rat brain. J Neurochem 1981;36:307. [24] Paglia DE, Valentine WN. Studies on quantitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med 1967;70:158. [25] Lawrence RA, Burk RF. Glutathione peroxide activity in Se deficient rat liver. Biochem Biophys Res Commun 1976;711:952.

212

G.M. Rao et al. / Clinica Chimica Acta 296 (2000) 203 – 212

[26] Horn HD, Burns FH. In: Bergmeyer HV, editor, Methods of enzymatic analysis, New York: Academic Press, 1978, p. 875. [27] Ravin HA. An improved colorimetric enzymatic assay of ceruloplasmin. J Lab Clin Med 1961;58:161. [28] Gowenlock AH, editor. Varley’s practical clinical biochemistry, 6th edn London: Heinemann Medical Books. 1988:664. [29] Cross CE, Halliwell B, Edward TB et al. Oxygen radicals and human disease. Ann Int Med 1987;107:526. [30] Cooper AJL, Pulsinelli WA, Puffy TE. Glutathione and ascorbate during ischemia and postischemic reperfusion in rat brain. J Neurochem 1980;35:1242. [31] Levchenko LI, Demchuk ML. Lipid peroxidation and antioxidative activity in the tumor tissue and blood of patients with neurological disease. Zh Vopr Neirokher Im NN Burdenko 1991;4:23. [32] Pu PY, Lan J, Shan SB, Huang EQ, Bai Y, Jiang DH. Study of the antioxidant enzymes in human brain tumors. J Neuro Oncol 1996;29:121. [33] Rolando DM, Warren M, Robert A. SOD, catalase and glutathione peroxidase in experimental and human brain tumors. In: Robert AG, Gohen G, editors. Oxyradicals and their scavenger systems. vol. II, Cellular and medical aspects, New York: Elsevier Science 1983:28. [34] Dzhandzhgava TG, Shakarishivilli RR. Activity of antioxidant protection enzymes and content of lipid peroxidation production in blood, serum and cerebrospinal fluid in patients with ischemic heart disease. Vopr Med Khim 1992;38:33. [35] Halliwell B, Gutteridge JMC. Protection against oxidants in biological systems. In: Free radicals in biology and medicine, 2nd ed, Oxford: Clarendon Press, 1989, p. 26. [36] Buckman TD, Sutphin MS, Mitroric B. Oxidative stress in a clonal cell line of neuronal cell origin: effects of antioxidant enzyme modulation. J Neurochem 1993;60:2046. [37] Wolf PL, Ray G, Kaplan H. Evaluation of copper oxidase (Ceruloplasmin) and related tests of Hodgkin’s disease. Clin Biochem 1979;12:202.