Role of tissue antioxidant defence in thyroid cancers

CANCER LETTERS

ELSEVIER

Cancer Letters 109 (1996) 231-235

Role of tissue antioxidant defence in thyroid cancers Geeta R. Sadania, Ganeshsunder D. Nadkarnib3* “Radiation Medicine Centre (BARC), Tata Memorial Centre Annexe, Parel, Bombay, 400 012, India bFree Radical Biochemistry Section, Radiation Medicine Centre, Bhabha Atomic Research Centre, Tata Memorial Centre Annexe Parel, Bombay, 400 012, India Received 9 September 1996; accepted 18 September 1996

Abstract Reactive oxygen species (ROS), consisting mainly of superoxide, hydrogen peroxide and hydroxyl radical, have been implicated in many diseases including cancer. ROS have been known to play an important role in the initiation and promotion of multistage carcinogenesis. The cellular antioxidant defenceplays a crucial role in neoplastic disease.However, very little is

known about the tissue antioxidant defencein thyroid cancers.We therefore undertooka study to assessthe role of ROS in the pathogenesisof thyroid cancers.Our samplesconsistedof post-operatedthyroid tissues(normal, goiters, follicular adenomas, follicular carcinomas and papillary carcinomas). The parameters studied were lipid peroxidation (LP), antioxidant enzymes -

superoxidedismutase(SOD), catalase(CAT) and glutathione peroxidase(GPx) - and non-protein thiols (GSH). Comparedto normal thyroid no changeswere seenin goiters. LP was significantly higher in adenomas(16%) and carcinomas(60-69%). SOD was decreased by 15% in adenomas while in carcinomas it increased by 9-12%. GPx was raised in carcinomas by lo21%. Follicular carcinomas showed a 4% increase in CAT activity while GSH was raised in adenomasand papillary carcinomas by 17’%. Thus, in adenomas (initial stage) involvement of superoxide radicals and in carcinomas (later stage)

hydrogen peroxide: and, possibly, hydroxyl radical involvement cannot be ruled out. These ROS may be responsible for elevated LP observed in adenomas and carcinomas. Keywords:

Reactive oxygen species;Tissue antioxidant defence; Thyroid cancer

1. Introduction During the last 25 years, a large body of experimental evidence has accumulated which suggests an important role of reactive oxygen species (ROS) in numerous pathophysiological/pathobiological processeslike arteriosclerosis, reperfusion injury, Parkinson’s disease, ageing and cancer [l-3]. An alteration in the prooxidant-antioxidant balance in favour of

* Corresponding author. Fax: +91 22 4146937.

prooxidant has come to be known as oxidative stress [4,5]. Oxidant carcinogens are of particular relevance to human carcinogenesis because of their ubiquitous occurrence in our environment and their generation in tissues by metabolic reactions and inflammatory processes [6]. Oxidants can act at several stages of malignant transformation [7]. Among the several radical species which have been proved to be cytotoxic, particular attention has been focused in recent years on the partially reduced forms of dioxygen (superoxide O;- hydrogen peroxide H202, and singlet oxygen ‘OJ. The combination of these reactive substances, parti-

0304-3835/96/$12.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PII SO304-3835(96)04484-9

232

G.R. Sadani, G.D. Nadkarni I Cancer Letters 109 (1996) 231-235

cularly in the presence of Fe*+ ions, generates the very reactive hydroxyl radical (OH) or lipid alkoxyl (Lo) and peroxyl radicals (LOG), which are widely assumed to be critical in causing molecular damage. The most important cell targets of these free radicals (FRs) are nucleic acids, proteins (including oxygen radical-scavenging enzymes) and membrane lipids. Several lines of evidence strongly suggest the involvement of cellular oxidative stress in carcinogenic processes [8,9]. Despite many years of active research, the molecular mechanisms, by which prooxidant stress may affect the outcome of tumours, are not well characterised. Yet, according to the present understanding, the prooxidant states may play a role in tumor promotion [ 10,7]. Studies carried out both in human and experimental animals show that lipid peroxidation (LP) has a very important role in the initiation and promotion of cancer. The damage to the defence systems has been shown to induce LP in the course of carcinogenesis [ 1I]. Tumor cells have been shown to have abnormal levels of antioxidant enzyme activities when compared with normal cells. However, enzyme activities also differ among individual tumours [ 121.A fundamental difficulty is the inability to distinguish whether the altered pattern of antioxidant enzymes is the primary metabolic disturbance that induces cancer, or whether these are secondary responses to neoplastic changes. The thyroid gland contains mechanisms for superoxide production such as xanthine oxidase and NADPH oxidase [ 131. The dismutation of superoxide by superoxide dismutase (SOD) produces hydrogen peroxide which is used as a substrate by thyroid peroxidase for thyroid hormone (TH) synthesis or is degraded by catalase (CAT) and glutathione peroxidase (GPx) [ 141.It is conceivable that SOD deficiency may cause accumulation of superoxide and superoxide itself may disrupt the function of the thyroid cell. Paradoxically, excessive SOD also seems to be harmful because it leads to the formation of more hydrogen peroxide and its conversion product, hydroxyl radical, depending on the C4T-GPx enzyme system. The nature of thyroid damage by oxygen free radicals is not known. As the thyroid gland presents with many pathologies which range from rather harmless goiters to benign adenomas to indolent malignant carcinomas, it provided us with an ideal model to study the involvement of ROS (indirectly by measuring the

antioxidant status and LP) in the pathogenesis of a spectrum of thyroid abnormalities. Very little information is available about the activities of these enzymes along with LP in nodular goiters, follicular adenomas, follicular carcinomas and papillary carcinomas of the thyroid to explore the possible involvement of specific ROS in different thyroid pathologies. 2. Materials

and methods

2.1. Patients and tissue samples Eighty-eight thyroid tissue samples were obtained as surgical specimens from euthyroid patients with various thyroid disorders (normal thyroid, 20; nodular goiters, 124;follicular adenomas, 10; follicular carcinomas, 12 and papillary carcinomas, 22). Normal thyroids were obtained after laryngectomies while the other thyroid tissues were procured after thyroidectomies performed by the Department of Head and Neck Surgery at Tata Memorial Hospital, Bombay. The studies met the criteria of the local ethical committee. 2.2. Processing of tissues Each thyroid tissue (0.5-1.0 g) was rinsed in saline to remove blood and homogenised in 9 ~01s.of chilled phosphate buffer (pH S.O), which also contained 0.25 M sucrose and 5 mM EDTA, using a motor -driven Teflon-coated pestle with 3-4 strokes of lo-15 s while being kept on ice (0-4°C). One aliquot of the homogenate was kept aside for the estimation of LP, GSH and proteins. Remaining aliquot of the homogenate was treated with Triton X-100 (0.05% v/v) and centrifuged at 12 000 x g for 30 min at 4°C. The supematant so obtained was then used to estimate the antioxidant enzymes. 2.3. Estimation of LP, GSH and proteins in the homogenate Lipid peroxide levels in the homogenate were estimated by determining the thiobarbituric acid-reactive substances (TBARS) by the method of Uchiyama and Mihara [ 151. The non-protein thiols (reduced glutathione., GSH) in the homogenate were measured by the method of Beutler et al. [ 161. The protein con-

233

G.R. Sadani. G.D. Nadkarni I Cancer Letters 109 (1996) 231-235

tent of the homogenate was estimated by the Lowry method [ 171.

2.4. Estimation of antioxidant enzymes SOD was estimated by the modified pyrogallol autoxidation method. SOD activity was expressed as (a - b)/a, where a and b are the increments of absorbance of blank and sample at 3 min, respectively [ 181. CAT activity was measured by the reduction in absorbance of hydrogen peroxide with time at 240 nm as per the method of Gogun et al. [ 191. One unit of CAT activity was defined as the amount of CAT which showed a decrease of 0.05 in absorbance in 30 s at 25°C. GPx activity was analysed by the method of Hafeman et al. based on consumption of GSH. One unit of GPx activity was defined as the decrease in log(GSH) by O.OOl/min after the decrease in log(GSH)/min of the non-enzymatic reaction was subtracted [20]. Antioxidants and LP values were expressed per gram thyroid tissue.

2.5. Statistical analysis Statistical significance of the determined parameters between normal and diseased thyroid samples was analysed by the non-paired Student’s t-test. Differences at P < 0.05 were considered significant.

3. Results The data on LP, GSH, proteins and the activities of antioxidant enzymes in thyroid tissues obtained from different thyroid disorders and from normal control thyroid are summarised in Table 1. As compared to normal control thyroid, there were no changes in any

of the parameters in goiter tissues. LP was significantly higher in all adenomas (16%) and carcinomas (65-69%). As far as the antioxidant status is concerned, there were specific changes in adenomas and carcinomas. SOD was decreased by 15% in adenomas, while there was an increase in carcinomas by 9-12%. GPx was elevated only in carcinomas by lo-21%. Follicular carcinoma showed raised CAT activity by 4%, while GSH was raised by 17% in adenomas and papillary carcinomas. The proteins were reduced by 6-18% in all cases compared to nomials.

4. Discussion A variety of experimental tumours have been shown to express an elevated antioxidant capacity and a reduced potential for LP [21]. In line with this, the production of ROS together with altered antioxidant functions are suspected to favour the neoplastic growth of initiated cells [22]. It is not clear, however whether the elevation of antioxidant capacity is more generally true for other types of experimental tumours, not to mention the numerous types of human malignancies. In support of this, during the last few years several studies have been published regarding the role of prooxidant state in carcinogenesis. The interpretation of results is difficult as they seem to vary from one study to another [23-261. The inconsistency is likely to be due to the heterogenicity of tumor tissues and the studies published so far do not support a general hypothesis regarding the role of oxidative stress in the pathobiology of cancer Various studies in the literature have demonstrated that tumours from different organs have wide variations in their antioxidant enzyme status. In addition to

Table 1 The LP and antioxidant status in various thyroid tissues Parameter studied

Normal control

Nodular goiter

Follicular adenoma

Follicular carcinoma

Papillary carcinoma

LP (nM MDA/g T) CAT W/g V SOD (U/g T) GPx(U/min/g T x 103) GSH @M/g T) Proteins (m& T)

141 f 17 564f29 58 +9 18.6 f 3.2 2.9 + 0.5 173 If: 31

148 f 21 564f28 54* 10 20.9 f 5.6 2.8 f 0.5 151 + 13*

164 + 566 f 50 * 18.7 k 3.4 k 162 f

233 + 36* 586 f 35* 63 +I ll* 22.6 f 5.5* 2.8 f 0.1 147 * 21*

239 + 29* 559 Yk31 65 f lO* 20.4 + 3.9* 3.4 ?I 0.4* 142 4 24*

*P < 0.05.

13* 41 9* 9.3 0.6* 23

234

G.R. Sadani, G.D. Nadkarni I Cancer Letters 109 (1996) 231-235

this, the nature of the link between LP and ROS has also given rise to much controversy. On the one hand LP has been reported to be elevated in neoplastic tissues [ll], while on the other hand it has been pointed out that not only neoplastic cells and tissues but rapidly dividing cells also exhibit low lipid peroxides [27]. For example in some tumor types, like breast cancer [23] and colon cancer [25], increased LP products have been TepOrted, whereas in endometrial cancer, the antioxidant enzyme defence system is altered, but the content of LP products in the tumor tissues is unchanged [28]. We have observed an elevation in LP levels from thyroid adenomatous and cancerous tissues, while goiter tissues were comparable to normals. In this connection it is interesting to note that all patients with goiter were euthyroid. In hypothyroid goiters induced in rats (by feeding propylthiouracil) we have observed the possible involvement of superoxide radicals and LP products (unpublished observations). In thyroid adenomas a marginal decrease in SOD activity points to the possible involvement of accumulated superoxide radicals in the resultant higher LP levels observed. In contrast to the adenomas, an increase in SOD activity was found to occur in thyroid carcinomas. This m.ay lead to production of higher amounts of hydrogen peroxide, which has been shown to be true in many human tumor cells [29]. In spite of elevated CAT in follicular carcinomas as well as GPx in both types of carcinomas, an increase in LP has been observed by us. This increase in LP may be due to the inability of the peroxide-scavenging enzymes to keep pace with the rate of peroxides being produced by the thyroid tissues and indicates that hydrogen peroxide and its active product, the hydroxyl radical, too may be involved in this neoplastic change. Thus, in adenomas (initial stage) superoxide radicals and in carcinomas (later stages) hydrogen peroxide, and possibly hydroxyl radical, may be involved. This may be the explanation for an increase in LP. The rise in GSH in follicular adenomas and papillary carcinomas may be serving as a substrate for the enhanced GPx activities. In this context it may be interesting to note that reduced glutathione has been shown to cause regression of aflatoxin-induced liver tumours in rats [30]. From the data of this study, ROS CO;-, HZ02 and

possibly OH radicals) involvement in thyroid adenomas and carcinomas cannot be ruled out. The possible sequence of events may be visualised as follows: cellular malignancy may be initiated initially, followed by the change in antioxidant enzymes by ROS mechanisms, which might also be necessary for the maintenance of the malignant state. Further, as mentioned earlier, since all the three enzymes act in concert, any imbalance in them may result in the accumulation of specific toxic oxygen metabolites which may become the cause of further damage and perpetuation of malignant state. As far as euthyroid nodular goiters are concerned there appears to be no disturbance in this balance. However, during the transfomration of harmless goiters to benign adenomas to malignant carcinomas, there is graded but definite perturbation in this balance.

Acknowledgements The authors are grateful to the Department of Head and Neck Surgery, Tata Memorial Hospital, Bombay for providing the post-operative thyroid tissue samples and the Department of Atomic Energy for the fellowship granted to Miss Geeta Sadani.

References [l] Kehrer, J.P. (1993) Free radicals as mediators of tissue injury and disease, Crit. Rev. Toxicol., 23, 21-48. [2] Bonorden, W.R. and Pariza, M.W. (1994) Antioxidant mments ;md protection from free radicals. In: Nutrition Toxicology, pp. 19-48. Editors: F.N. Katsonis, M. Mackey and J. Hjelle. Raven Press, New York. [3] Knight, J.A. (1995) Diseases related to oxygen-derived free radicals, Ann. Clin. Lab. Sci., 25, 111-121. [4] Cerutti, P.A.(1985) Proxidant states and tumor promotion Science, 221, 375-381. [5] Halliwell, B., Gutteridge, J.M. and Cross, C.M. (1992) Free radicals, antioxidants, and human diseases: where are we now?, J. Lab. Clin. Med., 119, 598-620. [6] Cerutti, P.A. (1991) Oxidant stress and carcinogenesis, Eur. J. Clin. Invest., 21, l-5. [7] Kensler, T.W. and Taffe, B.G. (1986) Free radicals in tumor promotion, Adv. Free Rad. Biol. Med., 2, 347-387. [8] Demopoulos, H.B., Pietrongro, D.D., Flamms, E.S. and Seligman, M.L. (1980) The possible role of free radical reactions in carcinogenesis, J. Environ. Pathol. Toxicol., 3, 273303. [9] Perchellet, J.P. and Perchellet, E.M. (1989) Antioxidants and multistage carcinogenesis in mouse skin, Free Rad. Biol. Med., 7, 377-408.

G.R. Sadani, G.D. Nadkarni / Cancer Letters 109 (1996) 231-235 [lo] Copeland, E.S. (1983) A national institute of health workshop report: free radicals in promotion - a chemical pathology study (Section workshop), Cancer Res, 43, 5631-5637. [ 1l] Capel, I.D. and Thornley, A.C. (1983) Lipid peroxide levels, glutathione status and glutathione peroxidase activity in liver and tumors of mice bearing the Lewis lung carcinoma, Cancer Biochem. Biophys., 6, 167-172. [12] Sun, Y. (1990) Free radicals, antioxidant enzymes and carcinogenesis, Free Rad. Biol. Med., 8, 583-599. [13] Ficher, A.G. and Lee, H. (1973) Xanthine oxidase from bovine thyroid glands, Life Sci., 12, 267-269. [14] Nakamura, Y., Makino, R., Tanaka, T., Ishimura, Y. and Ohtaki, S. (199 1) Mechanism of Hz02 production in porcine thyroid cells: evidence for the intermediary formation of superoxide amon by NADPH-dependent H20r-generating machinery, Biochemistry, 30,4880-4886. [ 151 Uchiyama, M. and Mihara, M. (1978) Determination of MDA precursor in tissue by TBA test, Anal. B&hem., 86, 271278. [16] Beutler, E., Duron, 0. and Mikus-Kelly, B. (1963) Improved method for the determination of blood glutathione, J. Lab. Clin. Med., 61, 882-886. [17] Lowry. O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with Folin phenol reagent, J. Biol. Chem., 1!>3, 265-275. [18] Sugawam, M., Kita, T., Lee, E., Takamatsu, J., Hagen, G., Kuma, K. and Medeiros-Neto, G. (1988) Deficiency of SOD in endemic goiter tissue, J. Endocrinol. Metab., 67, 11561161. [ 191 Gogun. Y ., Sakurda, S., Kimura, Y. and Nagumo, M. (1990) Enhancement of SOD activity in cancer cell lines by anticancer drugs, J. Clin. B&hem. Nutr., 8, 85-92. [20] Hafeman, D.G., Sunde, R.A. and Hoekstra, W.G. (1974) Effect of dietary selenium on erythrocyte and liver glutathione peroxidase in the rat, J. Nutr., 104, 580-587.

235

[21] Slater, T.F. (1988) Lipid peroxidation and cell division in normal and tumor tissues. In: Eicosanoids, Lipid Peroxidation and Cancer, pp. 137-142. Editors: S.K. Nigam, D.C.H. McBrien and T.F. Slater. Springer, Berlin. [22] Cerutti, P.A., Krupitza, G., Larsson, R., Muchlematter, D., Crawford, D. and Amstad, D. (1988) Physiological and pathological effects of oxidants in mouse epidermal cells, Ann. N. Y. Acad. Sci., 551, 75-82. [23] Hietanen, E., Purmonen, K., Punnonen, R. and Auvinen, 0. (1986) Fatty acid composition and lipid peroxidation in human breast cancer and hpoma tissue, Carcinogenesis, 7, 1965-1969. [24] Cheeseman, K.H., Collins, M., Proudfoot, K., Slater, T.F., Burton, G.W., Webb, A.C. and Ingold, K.U. (1986) Studies on lipid peroxidation in normal and tumor tissue, Biochem. J., 235, 507-514. [25] Otamiri, T. and Sjodahl, R. (1989) Increased lipid peroxidation in malignant tissues of patients with colorectal cancer, Cancer, 64,422-425. [26] Petruzzelli, S., Hietanen, E., Bartsch, H., Camus, A.M., Mussi, A., Angletti, CA., Saracci, R. and Giontini, C. (1990) Pulmonary lipid peroxidation in cigarette smokers and lung cancer patients, Chest, 98, 930-935. [27] Masotti, L., Casali, E. and Galeotti, T. (1988) Lipid peroxidation in tumor cells, Free Rad. Biol. Med., 4, 377-386. [28] Punnonen, R., Kudo, R., Punnonen, K., Hietanen, E., Kuoppala, T., Kainolainen, H., Sato, K. and Ahutupa, M. (1993) Activities of antioxidant enzymes and lipid peroxidation in endometrial cancer. Eur. J. Cancer, 29A, 266-269. [29] Szatrowski, T.P. and Nathan, C.F. (1991) Production of large amounts of hydrogen peroxide by human tumor cells, Cancer Res, 5 1, 794-798. [30] Novi, A.M. (1981) Regression of aflatoxin B, induced hepatocellular carcinomas by reduced glutathione, Science, 212, 541-542.