Clinical Biochemistry, Vol. 33, No.

4, 279 284, 2000 Copyright 2000 The Canadian Society of Clinical Chemists Printed in the USA. All rights reserved 0009-9120/00/$see front matter

PII S0009-9120(00)00067-9

Coenzyme Q10 Concentrations and Antioxidant Status in Tissues of Breast Cancer Patients
ZAY O ZKAYA,2 MINE ERDEN I BERRA KOS OYTUN PORTAKAL,1 O NAL,1 BERRIN BOZAN,3 MU AN,3 4 and ISKENDER SAYEK
2

Department of Biochemistry, The Medical School of Osmangazi University, Eskis ehir, Turkey; Department of Pathology, Faculty of Medicine, Hacettepe University, Ankara, Turkey; 3Medical Plant and Auromatic Research Center, The Pharmacy School of Anadolu University, Eskis ehir, Turkey; 4 Department of General Surgery, Faculty of Medicine, Hacettepe University, Turkey
Introduction

Objectives: An increasing amount of experimental and epidemiological evidence implicates the involvement of oxygen derived radicals in the pathogenesis of cancer development. Oxygen derived radicals are able to cause damage to membranes, mitochondria, and macromolecules including proteins, lipids and DNA. Accumulation of DNA damages has been suggested to contribute to carcinogenesis. It would, therefore, be advantageous to pinpoint the effects of oxygen derived radicals in cancer development. Design and methods: In the present study, we investigated the relationship between oxidative stress and breast cancer development in tissue level. Breast cancer is the most common malignant disease in Western women. Twenty-one breast cancer patients, who underwent radical mastectomy and diagnosed with infiltrative ductal carcinoma, were used in the study. We determined coenzyme Q10 (Q) concentrations, antioxidant enzyme activities (mitochondrial and total superoxide dismutase (SOD), glutathione peroxidase (GSH-Px), catalase), and malondialdehyde (MDA) levels in tumor and surrounding tumor-free tissues. Results: Q concentrations in tumor tissues significantly decreased as compared to the surrounding normal tissues (p 0.001). Higher MDA levels were observed in tumor tissues than noncancerous tissues (p 0.001). The activities of MnSOD, total SOD, GSH-Px and catalase in tumor tissues significantly increased (p 0.001) compared to the controls. Conclusions: These findings may support that reactive oxygen species increased in malignant cells, and may cause overexpression of antioxidant enzymes and the consumption of coenzyme Q10. Increased antioxidant enzyme activities may be related with the susceptibility of cells to carcinogenic agents and the response of tumor cells to the chemotherapeutic agents. Administration of coenzyme Q10 by nutrition may induce the protective effect of coenzyme Q10 on breast tissue. Copyright 2000 The Canadian Society of Clinical Chemists

KEY WORDS: coenzyme Q10; infiltrative ductal carcinoma; MnSOD.

Correspondence: Oytun Portakal, M.D., Dept. of Biochemistry, Hacettepe University, Medical School, Clinical Pathology Laboratory, Pediatric Biochemistry, Ankara, Turkey. E-mail: portakal@ada.net.tr Manuscript received July 13, 1999; revised February 25, 2000; accepted February 29, 2000.
CLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000

erobic metabolism produces reactive oxygen species which have an unpaired electron under physiological conditions (1,2). Especially in the presence of metal ions, oxygen derived radicals may cause an oxidative damage by reacting with macromolecules including proteins, lipids and DNA in the cell (2,3). The accumulation of DNA damages has been suggested to contribute to carcinogenesis (3). Carcinogenesis is a formation, proceeding through at least three different stages including initiation, promotion and malignant conversion, and is significant in human mortality (4,5). Oxygen derived radicals have been implicated in the etiology of cancer development perhaps since the demonstration that ionizing irradiation caused cancer (3,6). It has been shown that free radicals have a mutagenic capacity as a result of the interaction between highly reactive chemical molecules and DNA (4). Upon reaction with DNA, oxygen derived radicals produces base adducts and strand breaks. The former can cause mispairing lesions during DNA replication. Those of different types of chemical changes in DNA molecule could be mutagenic lesions involved in the initiation and progression processes of cancer. It was also reported that oxyradicals can also cause cytotoxcity and stimulate changes in gene expression (4,5,7). The cells protect themselves against oxidative damage by enzymatic and nonenzymatic antioxidant systems. Superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase form primary enzymatic defense system (1). SOD catalyses dismutation of superoxide radicals to hydrogen peroxide. Manganase containing form, MnSOD, is mainly found in mitochondria. Hydrogen peroxide is metabolized by catalase and GSH-Px via reducing into water and molecular oxygen. Another endojen antioxidant is coenzyme Q10 (Q). Q is an essential
279

PORTAKAL

ET AL.

component and a redox carrier of electrons in the transfer chain in mitochondria (8). It is situated between the flavoprotein complexes and the cytocrome bc1 complex and transfers electrons through protonmotive Q cycle (9,10). It was shown that the reduced form of Q (QH2) has a powerful antioxidative capacity (8 10). The purpose of this study was to examine the relationship between oxidative stress and breast cancer development in tissue level, to analyze antioxidant enzymes and to evaluate the consideration of coenzyme Q10 as chemopreventive agent. Breast cancer is the most common malignant tumor in western women. For this aim, peroxidative damage and antioxidant status in tissue levels were determined in invasive or infiltrative ductal carcinoma (IDC) which represents the most common histological type of breast cancer. Levels of Q in tumor tissues have not been investigated in any breast cancer. So in the present study, Q concentrations and primary defense enzyme activities including total SOD and MnS0D, GSH-Px and catalase were measured. Peroxidative damage was determined by assaying malondialdehyde levels. Methods Patients who applied to the Surgical Department at the Medical School of Hacettepe University in the first part of 1996, and underwent a mastectomy with full dissection of axillary lymph nodes, were enrolled in this study. All patients were operated on in the Surgery Department at Hacettepes Hospital. Twenty-one women, mean age 44.23 6.13 years, who were histologically diagnosed with primary IDC were included. Patients with a previous carcinoma of the breast, or multiple mammary carcinomas at presentation were excluded. The study was approved by the ethical committee of the university and all women gave their informed consents. For biochemical determinations, mastectomy materials were not treated with formalin. About 2.0 g fresh tumor tissue and surrounding tumor-free tissue were carefully extracted from mastectomy materials by a pathologist. The study materials were washed with saline three times, kept in the vials, and stored 70 C until the analysis. Then, mastectomy materials were fixed in 10% buffered formalin (pH 7.0). Paraffin embedded tissue blocks were sectioned at a thickness at 5 6 m. Original hematoxylin-eosin staining was performed for light microscopy. Bloom and Richardson System, as modified by Elston, was used for grading each IDC. The grade was obtained by evaluating tubule formation, nuclear pleomorphism and mitotic rate. Patients with histologically approved infiltrative ductal carcinoma were included in the study. For enzyme measurements, tissues were homogenized with 0.05 M phosphate buffer (1 g tissue per 10 mL phosphate buffer), and tissue protein was measured by the Lowry method. SOD activity was assayed
280

according to the method of Oberley and Spitz (11). In this method, xanthinexanthine oxidase system was used to generate a superoxide flux, and nitroblue tetrazolium (NBT) is used as an indicator of superoxide production. To prepare the blanks, xanthine oxidase was diluted in phosphate buffer until the tubes gave an absorbtion rate between 0.015 0.025/ min. MnSOD activity was determined by the same method, but 0.33 M sodium cyanide was added into the working solution. GSH-Px activity was determined by the method of Paglia and Valentina (12). In this method, the conversion of NADPH to NADP was followed by observing the change in absorbtion at 340 nm. Catalase activity was assayed according to the method of Beutler (13). Briefly, 1 M Tris-HCl 5 mM EDTA and 10 mM H2O2 were mixed. After incubation at 37 C for 10 min, homogenate was added to the mixture. The decrease of absorbance at 230 nm was measured. MDA concentration was assayed by the method of Ohkawa et al. (14). In this reaction, the red pigment, which was obtained by the colorimetric reaction of thiobarbituric acid (TBA) with malondialdehyde, was determined at 532 nm. For coenzyme Q10 measurement, tissues were homogenized with 0.05 M phosphate buffer including 0.25 M sucrose, 10 mM Tris-HCl, 0.5 mM EDTA and 0.1% pH 7.4 (1 g tissue per 10 mL buffer). Isolation of mitochondria was performed by the method of differential centrifugation. Inner membrane was isolated according to the method of Schneider and Hogeboom (15). Extraction of coenzyme Q10 from mitochondrial suspension was performed by the method of Ikenoya et al. (16). After three ethanol/n-hexane extractions, the upper nhexane layer of those was removed by nitrogen flux and injected into the HPLC column. In this determination, external standard method was used. Coenzyme Q10, from bovine heart 10 mg, was purchased from Sigma Chemical Co. (St. Louis, MO, USA). About 5.3 mg Q was dissolved 50 mL ethanol (0.106 mg/mL). This solution was used as a stock solution. Five dilutions in the range of 0.4 103 mg/mL to 20 103 mg/mL (1/2500, 1/1000, 1/500, 1/250, 1/125, 1/50 mL, respectively) were prepared from the stock solution to obtain the calibration curve. Ten microliters were injected from each dilution three times. Peak areas were obtained by an integrator. The calibration curve was plotted area versus concentration, and sample Q levels were calculated using this calibration curve. Recovery studies were also carried out. We prepared Q supplemented samples by adding a known concentration of exogenous coenzyme Q10 (36.8 49.0 mol/L) to the aliquots of the two tissue samples. The recoveries were 105% and 72%, respectively. HPLC
CONDITIONS

Apparatus: Model; Varion 2010 (USA), Integrator; Shimadzu C-R4A (Japan). Detector; Shimadzu SPD-6A UV, 275 nm (Japan), Column; ODS, C18 reverse phase column, 250 mm 4.6 mm, Beckman,
CLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000

COENZYME Q10 CONCENTRATIONS AND ANTIOXIDANT STATUS IN BREAST CANCER TISSUES

TABLE 1 Patient and Treatment Characteristics Number of patients Age at diagnosis, y 40 4060 60 Mode of detection Mammography Signs symptoms Hormonal status Premenopausal Perimenopausal Postmenopausal Tumor size 2 cm 25 cm 5 cm Location Original tumor site Elsewhere in breast Multisentric Diffuse Invasion of fascia Skin involved Lenfovascular invasion Histology IDC Other Histologic grade Grade I Grade II Grade III 3 17 1 1 20 11 4 5 3 12 4 21 3 1 1 1 Axillary node metastatis 0 13 49 10 Distant metastasis No Yes Stage 01 II III IV Surgical therapy MRM Simple mastectomy Adjuvant therapy None Preop. radiotherapy Postop. radiotherapy Preop. chemotherapy Postop chemotherapy Postop. hormone therapy Postop. radiotherapy chemotherapy Postop. radiotherapy hormone therapy Combined Outcome Disease free Relapsed Survival Alive Dead Number of patients 11 5 2 3 20 1 3 10 7 1 20 1 2 8 7 9 6 3 1 4 16 4 20 1

21 1 2 18

USA, Mobil phase; 3.5 g NaCIO4 50 mL MeOH 450 mL EtOH: 0.5 mL 70% HCIO4, Flow rate; 1 mL/min, Temperature; Ambient, Loop volume: 10 L, Sensitivity; 0.005. Chemicals: NaClO4, 70% HClO4 were purchased from Sigma. MeOH (HPLC grade) and EtOH (HPLC grade) were obtained from Merck. STATISTICAL
ANALYSIS

The paired-samples t test was used to compare the tumor tissue and surrounding tissue of the same patient. Correlation of the variables was established by Pearson method. Values of p 0.05 was considered significant. All results were expressed as mean SEM. Results All carcinomas were classified as infiltrative ductal carcinoma. There were 1 grade I, 4 grade II, and 16 grade III. Axillary lymph node status was known in 10 patients, and metastatic carcinoma was present in one patient. There were 3 multisentric cancer. Vascular invasion was present in one tumor and skin involveCLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000

ment was present in one tumor. Table 1 shows the overall properties of the breast cancer patients. Endogenous Q levels, superoxide dismutase, MnSOD, catalase and glutathione peroxidase activities and malondialdehyde values were investigated in tumor tissues of the breast cancer patients. Surrounding tissues of the tumor of the same patient were used as the control. Figure 1 presents antioxidant enzyme activities from paired control and infiltrative ductal carcinoma (IDC) groups. SOD activity significantly increased, ranging from 77.31 7.78 U/mg protein for noncancerous tissues to 207.82 21.94 U/mg protein for IDC tissues (p 0.001). The increase was 170%. Also, MnSOD activity in tumor tissues increased as compared to the control group (47.89 5.54 vs. 17.76 1.60 U/mg protein; p 0.001) and the increase was 169%. There was statistically significant correlation between both SOD activities (r 0.79; p 0.001). Figure 1 also shows catalase activity that increased significantly in IDC group compared to the controls (444.45 51.97 vs 137.18 15.84 U/mg protein; p 0.001). A 224% increase was observed.
281

PORTAKAL

ET AL.

Figure 1 The comparison of antioxidant enzymes including SOD, MnSOD, catalase and GSH-Px activities in infiltrative ductal carcinoma (IDC) group against to the those in noncancerous ones. Each bar represents the mean values standard error (SE) of enzyme activities. *Significant at p 0.001.

Figure 3 Coenzyme Q10 (Q) concentrations in infiltrative ductal carcinoma (IDC) group significantly decreased when compared to the controls. Each bar represents the mean values standard error (SE) of coenzyme Q10 levels. *Significant at p 0.001.

We found statistically linear correlations between MnSOD and catalase activities in both the normal and cancerous tissues (r 0.57, p 0.05). Glutathione peroxidase activity in tumor tissues was higher than in the corresponding normal tissues (Figure 1). The mean GSH-Px activity in cancer-free tissues was 19.78 2.69 U/mg protein and in cancerous tissues was 47.35 6.58 U/mg protein. The increase was 141%. The difference between two different tissues was statistically significant (p 0.001).

In the examined group of breast cancer patients, extremely high tissue malondialdehyde levels was found (Figure 2). In cancerous tissues, the mean MDA concentration was 1.14 0.20 nmol/mg protein and in noncancerous tissues was 4.07 0.69 nmol/mg protein. The difference was statistically significant (p 0.001). The values of Q in noncancerous and cancerous tissues could be measured in 11 out of 21 patients (Patients 1, 3, 7, 8, 11, 13, 15, 16, 17, 20, and 21, respectively). Q levels of four patients were below detection limit and also there were insufficient tissues of six patients after enzyme determinations, therefore, these could not be measured (Figure 3). The mean Q concentration in cancer-free tissues was 48.79 12.03 ng/g and in tumor tissues was 107.58 16.32 ng/g. Q levels in cancerous tissues decreased proportionally 120% (p 0.001). Discussion To our knowledge, this is the first study of mitochondrial Q concentrations in human breast carcinoma tissues. We have observed that Q levels in tumor tissues of the breast were lower than corresponding noncancerous tissues. This could reflect consumption of Q against peroxidative damage in tumor tissues. Previous in vivo and in vitro studies have reported that the reduced form of Q, QH2, is an important antioxidant for unsaturated lipids of mitochondrial membranes against free radical damage. This protective effect is related to direct interaction with lipid peroxidation products (8 10). Since, QH2 is present in the central plain of the membrane, it may react with perferryl, peroxyl and lipid radicals, and may inhibit initiating and promotion stages of lipid peroxidation
CLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000

Figure 2 Malondialdehyde (MDA) concentrations in tumor tissues was significantly higher than control group. Each bar represents the mean values standard error (SE) of MDA levels. *Significant at p 0.001.
282

COENZYME Q10 CONCENTRATIONS AND ANTIOXIDANT STATUS IN BREAST CANCER TISSUES

(9,10). QH2, also prevents lipid peroxidation by regeneration of alpha-tocopherol indirectly (8,9). Reduced Q plays a role not only against lipid peroxidation but also against oxidative damage of proteins and DNA (9). It was shown that QH2 may also directly react with superoxide radicals (10). A previous study reported that plasma Q concentration was lower than 0.5 g/mL in breast cancer (17). According to the ability of Q to prevent lipid peroxidation in the cell, it could reduce the susceptibility of the cell to the cancer development. Exogenous Q administration via nutrition may help increase the protective effect of Q in breast tissue especially in women in high risk group. Lipid peroxidation is a chain reaction that involves the oxidation of polyunsaturated fatty acids in membranes induced by free radicals and is an indicator of oxidative cell damage (3). Malondialdehyde, a secondary product of lipid peroxidation is used as an indicator of the rate of tissue chain reactions (14). It has been reported that products of lipid peroxidation may cause DNA damage (3,18,19). Lipid hydroperoxides may directly induce DNA chain breaking (20), and lipid peroxyl and alkoxyl radicals may cause base oxidation in DNA (21). Peroxide and hydroperoxides have also demonstrated tumor promoting activity in vivo (4). It was shown that the epoxy derivative of 4-hydroxynonenal (HNE) has a tumor causing activity (2224). Lipid peroxidation may also play an indirect role in the conversion of procarcinogens to the last carcinogens (7,25). In the present study, we have found that MDA content in cancerous tissues of the breast was higher than in the corresponding noncancerous tissues. Our finding was contrary to previous studies suggesting lipid peroxidation was lower in rapidly proliferating tissues and there was a negative correlation between lipid peroxidation and the degree of differentiation of the tumor (26). The increase we have found may be related to the presence of necrosis in invasive ductal carcinoma resulted from insufficient vascularity. Some previous observation reported higher MDA levels in solid tumor and human cell lines (2729). It was also shown that elevated MDA concentrations decreased with tamoxifen therapy in postmenopausal women with breast cancer (30). Oxidative DNA damage, including mutagenic and cytotoxic lesions, is implicated in the initiation phase the of cancer process. Oxygen derived radicals interaction with DNA produce base adducts, deletions, frameshifts, strand breaks, and DNA-protein crosslinks. Defective repair of DNA lesions leads to mutations, blocking of replication, transcription and chromosomal aberrations resulting in permanent genetic alterations (4,5,23). Oxygen derived radicals can participate in tumor promoters and stimulate expanding of initiated cell population. Free radicals generating compounds such as peroxides can mimic tumor promoting activity (4,5). It was shown that antioxidant enzymes and oxidant detoxifiers have the ability to inhibit tumor promotion and initiation in vivo and in vitro (4,7). Superoxide dismutase is the major primary deCLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000

fense enzyme in the cell. We have found MnSOD and total SOD activities in tumor tissues of the breast were significantly higher than in the corresponding cancer-free tissues. The increase we have found may be related to the higher enzyme expression in tumor cells. Oberley et al. (31,32) have reported that in tumor tissues total and mitochondrial SOD activities increased compared to noncancerous tissues and this increase was dependent upon overexpression of the enzymes. In a previous study, it was observed that the expression of mRNA in tumor tissues was higher than noncancerous tissues in breast and gastric carcinoma (33). It was suggested that there was a negative correlation between overexpression of MnSOD in human breast cell lines (MCF-7 cells) and local expansion of the tumor (34). Higher total and mitochondrial SOD activities could lead to an increase in the cure rate of breast cancer patients. Glutathione peroxidase is the first step defense enzyme against hydrogen peroxides and other hydroperoxides. In the present study, we have found that GSH-Px activity in tumor cells significantly increased as compared with cancer-free tissues. Previous studies have reported higher GSH-Px activity in breast cancer tissues which was due to overexpression of the enzyme (35,36). Doroshow have reported that GSH-Px activity was higher 25 times in human breast cell lines (MCF-7 cells) and was due to the increase in expression of genomic DNA (37). In another previous study, an increase was observed in GSH-Px activity in breast cancer tissues, especially early mammary carcinoma (38). Glutathione and glutathione-utilizing enzymes seem to be important in modulating the sensitivity of breast carcinomas to chemotherapeutic agents. Our finding may explain the increased expression of GSH-Px in tumor cells, which may be responsible lower effectiveness of anticancer drugs. In high levels of hydrogen peroxide, catalase is more effective than glutathione peroxidase (2). Catalase was also significantly higher in cancerous tissues than corresponding normal tissues. We also found a significant correlation between MnSOD and catalase activities. The increase may be due to an increase in enzyme expression in tumor cells. Some investigators have reported higher catalase activity in different tumor cell lines compared to the controls (39). Offner et al. (40) observed reducing endothelial cell damage, which was important in metastasis by extravasation of tumor cells, after adding catalase to the medium. Our finding may provide a better prognosis in breast cancer patients. In conclusion, the present work provides evidence for decreased levels of Q concentrations, increased levels of MDA, and primary antioxidant enzyme activities in breast cancer tissues compared to the controls. The results of this study may suggest that increased antioxidant enzyme activities may be effective in tumor cell response to the chemotherapeutic agents because of their antineoplastic affects via free radical mechanism. Tumor cells with higher antioxidant enzyme activities will be more protected, and will, therefore, be resistant. However, surrounding normal cells
283

PORTAKAL

ET AL.

will be less susceptible to their toxicity. So it may contribute to the development of anticancer drug resistance. Dietary or exogenously coenzyme Q10 intake may be effective as a chemopreventive agent. Acknowledgement
We thank Osmangazi University Research Foundation for supporting our study.

20. 21. 22.

References
1. Galleotti T, Masotti L, Borrello S. Oxy-radical metabolism and control of tumour growth. Xenobiotica 1991; 21: 104151. 2. Halliwell B. Reactive oxygen species in living systems: source, biochemistry, and role in human disease. Am J Med 1991; 91: 14 21. 3. Halliwell B, Gutterage JMC. Role of free radicals and catalytic metal ions in human disease: an overview. Methods Enzymol 1989; 186: 1 85. 4. Guyton KZ, Kensler TW. Oxidative mechanism in carcinogenesis. Br Med Bull 1993; 49: 523 44. 5. Hochstein P, Atallah A. The nature of oxidants and antioxidant systems in the inhibition of mutation and cancer. Mut Res 1988; 202: 36375. 6. McCord JM. Human disease, free radicals and the oxidant/antioxidant balance. Clin Biochem 1993; 26: 3517. 7. OBrien PJ. Antioxidants and cancer. Molecular mechanism. In: Armstrong D, Ed. Free radicals in diagnostic medicine. Pp. 215239. New York: Plenum Press, 1994. 8. Crane FI. Determination of ubiquinones. Methods Enzymol 1990; 176: 137181. 9. Beyer RE. An analysis of the role of coenzyme Q in free radical generation and as an antioxidant. Biochem Cell Biol 1992; 70: 390. 10. Ernster L, Dallner G. Biochemical, physiological and medical aspects of ubiquinone function. Biochim Biophys Acta 1995; 1271: 195204. 11. Oberley LW, Spitz DR. Assay of superoxide dismutase activity in tumor tissue. Methods Enzymol 1984; 105: 457 65. 12. Paglia DE, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med 1967; 70: 158 69. 13. Beutler E. Red cell metabolism. In: A manual of biochemical methods. Pp. 70. New York: Grune And Stratton, 1973. 14. Ohkawa H, Ohishi N, Yagi K. Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 95: 351 8. 15. Schneider WC, Hogeboom GH. Isolation of mitochondria. J Biol Chem 1950; 183: 1237. 16. Ikenoya S, Takada M, Yuzuriha T. Studies on reduced and oxidized ubiquinones. I. Simultaneous determination of reduced and oxidized ubiquinones in tissues and mitochondria by HPLC. Chem Pharm Bull 1981; 29(1): 158 64. 17. Folkers K, Osterborg A, Nilander M. Activities of vitamin Q10 in animal models and a serious deficiency in patients with cancer. Biochem Biophy Res Com 1997; 234: 296 9. 18. Esterbauer H. Possible mutagens derived from lipids and lipid procursors. Mutat Res 1990; 238: 22333. 19. Marinari UM, Ferro M, Sciaba L. DNA-damaging
284

23. 24.

25. 26. 27.

28. 29. 30. 31.

32. 33.

34.

35. 36. 37. 38.

39. 40.

activity of biotic and xenobiotic aldehydes in Chinese hamster ovary cells. Cell Biochem Funct 1984; 2: 243 8. Cochrane CG. Cellular injury by oxidants. Am J Med 1991; 91: 239. Park D. Peroxyl and alxocyl radicals cause DNA base modifications. Cancer Lett 1992; 28: 1235 40. Chung FL, Chen H-J.C, Guttenplan JB. 2,3-Epoxy4hydroxynonanal as a potential tumor-initiating agent of lipid peroxidaion. Carcinogenesis 1993; 14: 20737. Sodum RS. Steroselective formation of in vitro nucleic acid adducts by 2,3 epoxy-4-hidroxynonenal. Cancer Res 1991; 51: 137 43. Wang M, Dhingra K, Hittelman WN. Lipid peroxidation-induced putative malondialdehyde-DNA adducts in human breast cancer tissue. Cancer Epi Biom Prev 1996; 5(9): 70510. Floyd RA, Soong LM, Walker RN. Lipid hydroperoxide of N-hydroxyy-N-acetyl aminofluorane via free radical route. Cancer Res 1976; 36: 2761. Dianzoni MU. Lipid peroxidation and cancer. Crit Rev Oncol Hematol 1993; 15: 125 47. Diplock AT, Rice-Evans K, Burdon RH. Is there a significant role for lipid peroxidation in the causation of malignancy and for antioxidants. Cancer Prevent 1993; 54: 1952 6. Rossi MA. Lipid peroxidation of hepatomas of different degrees of deviation. Cell Biochem Funct 1983; 1: 49 54. Dianzoni MU. Further experiments on lipid peroxidation. Transplant Exper Hepatomas 1984; 12: 189 99. Thangaraju M. Effect of tamoxifen on lipid peroxide and antioxidative system in postmenopousal women with breast cancer. Cancer 1994; 74: 78 82. Liu R, Oberley LW. Transfection and expression of MnSOD cDNA decreases tumor malignancy of human oral squamouse carcinoma SCC-25 cells. Hum Gene Ther 1997; 8: 58595. Zhong W, Oberley LW. Suppression of the malignant phenotype of human glioma cells by overexpression of MnSOD. Oncogene 1997; 14(4): 48190. Izutani R, Katoh M, Asano S. Enhanced expression of manganese superoxide dismutase mRNA and increased tNF and mRNA expression by gastric mucosa in gastric cancer. World J Surg 1996; 20: 228 33. Li JJ, Colburn NH, Oberley LW. Maspin gene expression in tumor suppressor induced by overexpressing manganase containing SOD cDNA in human breast cancer. Carcinogenesis 1998; 19: 8339. Hartman Z. Glutathione peroxidase activity in tumorigenesis. Environ Med Mutagen 1987; 10: 315. Gromodniska J. Glutathione and glutathione metabolizing enzymes in tissue and blood of breast cancer patients. Neoplasma 1997; 44: 4551. Doroshow JH. Glutathione peroxidase and oxidative stress. Toxicol Lett 1995; 82/83: 395 8. Buser K, Jencourt F. Breast cancer pretreatment drug resistance parameters in tumor tissue and their correlation with clinical and prognostic charcteristics. Ann Oncol 1997; 8: 335 41. Ripple MO, Henry WF. Prooxidant antioxidant shift induced by androgen treatment of human prostate carcinoma cells. J Natl Cancer Inst 1997; 89: 40 8. Offner FA, Schiefer J, Wirtz HC. Tumor cell endothelial interactions. Free radicals are mediators of melanoma-induced endothelial cell damage. Virchows Arch 1996; 428: 99 106.
CLINICAL BIOCHEMISTRY, VOLUME 33, JUNE 2000