Half-lives of tyrosinase isozymes from Harding-Passey mouse melanoma

Cancer Letters, 38 (1988) 339- 346 Elsevier Scientific Publishers Ireland Ltd.

HALF-LIVES OF TYROSINASE MOUSE MELANOMA

JOSE H. MARTINEZ, Departamento

FRANCISCO

de Bioquimica,

Facultad

(Received 21 April 1987) (Revised version received 28 September (Accepted 19 October 1987)

339

ISOZYMES

SOLANO,

RAFAEL

de Medicina,

FROM HARDING-PASSEY

PEfiAFIEL

Universidad

and JOSE A. LOZANO

de MILT&, SO001 Murcia (Spain)

198’7)

SUMMARY

The half-lives of tyrosinase isozymes, a key enzyme in melanogenesis, have been determined using two different approaches: (a) cycloheximide treatment of mice bearing growing tumors and measurement of the residual enzymatic activity. This approach detected two soluble forms of cytosolic tyrosinase with half-lives of % and 8 h, respectively. The melanosomal isozyme showed a t,,2 of 3V2 h. (b) Metabolic labelling of tyrosinase with [35S]Met and immunoprecipitation analysis using a monoclonal antityrosinase. This method gave values slightly longer, 5 h and 9’12 h for the melanosomal and cytosolic tyrosinase, respectively. The origin of soluble tyrosinase and its utility to employ that enzymatic activity in melanoma chemotherapy using catechols as tyrosinase-dependent precursors of cytotoxic quinones is discussed.

INTRODUCTION

Melanization is a melanocyte-specific biochemical pathway leading to the accumulation of melanin, the main pigment found in mammalian skin, hair and eyes [5]. Tyrosinase is the key multifunctional-enzyme in that pathway. It catalyzes three reactions: the hydroxylation of tyrosine to dopa (tyrosine hydroxylase activity), the oxidation of dopa to dopaquinone (dopa oxidase activity) and the conversion of 5,6-dihydroxyindole to indol-5,6-quinone [4]. The transformed melanocyte is called malignant melanoma, a highly metastatic skin melanin formation is restricted to cancer. In mammalian melanocytes, melanosomes, presumably due to the cytotoxicity of o-quinones and other intermediate products of the melanin pathway [7,11]. However, there are at least 3 tyrosinase isozymes in those cells. Melanosome tyrosinase is the mature enzyme, the only one expressing activity ‘in vivo’, to form melanin in that Address

correspondence

to: Francisco Salano.

0304.3835/88/%03.50 0 1988 Elsevier Published and Printed in Ireland

Scientific

Publishers

Ireland Ltd.

organelle. Microsomal tyrosinase is the ‘de novo’ biosynthesized enzyme, which undergoes post-translational modifications travelling through intracellular membranous systems such as ER, Golgi apparatus etc. Finally, soluble tryosinase is an isozyme found in the soluble cytosol of malignant melanocytes. It might play a role in the transfer of tyrosinase from microsomes to melanosomes, but might also be released from melanosomes by digestion with various proteinases [14], since malignant melanocytes accumulate melanosomes which can be degraded by lysosomal enzymes. Several catechol compounds, including dopa and its natural derivatives, have been shown to possess significant antitumor activity against melanoma [7-g]. The preferential cytotoxicity of catechols to melanocytes is due to the biotransformation of these compounds into quinones by non-melanosomal tyrosinase, producing self-destruction of the malignant melanocytes [18]. On the other hand, tyrosinase is known to be inactivated during the ‘in vivo’ enzymecatalyzed dopa oxidation [15,17]. That these facts become the half-lives of tyrosinase isozymes is an important point to take into account in order to improve the effectiveness of the dopa-derived antimelanoma agents. Studies on this point are very scarce. Synthesis and degradation of tyrosinase have been sometimes studied in cultured melanoma cells, being largely dependent on the pH and composition of the culture medium [lo] as well as the kind of cultured melanocytes [2]. Further, in solid tumors, processing and mainly degradation of tyrosinase isozymes might be much different than in cultured melanocytes due to the well known elevation in tumor-bearing animals of lysosomal cathepsin B [12,13] and other proteinases [3]. In turn, the intratumoral acidic pH possibly found in these malignant tissues as consequence of a high rate of anaerobic glycolysis and lactate production [l] could enhance the activity of acid proteases. Therefore, these hydrolytic enzymes can solubilize and/or inactivate melanosomal tyrosinase, affecting its half-life. To investigate half-lives of tyrosinase isozymes in transplantable solid tumors, Harding-Passey mouse melanoma has been used in the present study. We have used two different approaches of study: (al treatment of melanoma-bearing animals with cycloheximide to inhibit protein synthesis. labelling of tyrosinase with [35S]Met and Metabolic (bl radioimmunoprecipitation using a monoclonal antibody directed against tyrosinase. MATERIALS

AND METHODS

Animals and melanomas C57Bl mice were obtained from Panlab, Spain. Mice were used at 6 - 8 weeks of age. Harding-Passey melanomas were originally obtained from the Institute of Cancer Research, Royal Cancer Hospital, London. They have been propagated and maintained in our laboratories by serial transfer of approximately lo5 dissociated viable tumor cells in male mice, After 15-20 days, some developed tumors were used for new implantation and the rest for tyrosinase experiments.

341

Reagents Monoclonal antibody directed against tyrosinase (TMH-11, was generously given by Dr. V.J. Hearing, N.I.H., Bethesda, U.S.A. It was produced as described in Ref. 16. [%]Methionine (1102 Cilmmol) was purchased from New England Nuclear, U.S.A. Protein-A Sepharose was from Pharmacia Fine Chemicals, Sweden. Other biochemical reagents were purchased from Sigma Chemical Co., U.S.A. Cycloheximide treatment A 25 group of previously weighed mice bearing actively growing tumors (approx. 1 g of tumor wt.1 were i.p. injected with a cycloheximide solution (10 mg/kg body wt.). Mice were randomly selected in 5 groups of 5 animals each. These groups were sacrificed at 0 (controls), 45,90, 180 and 300 min after the injection. Tumors were promptly excised and homogenized and the melanosome, microsome and soluble tyrosinase isozymes were prepared to estimate the enzymatic activities of each. Metabolic lubelling of tyrosinase Ten mice bearing actively growing melanoma were starved for 1 h. Then, each animal was i.p. injected with 0.4 mCi of r5S]Met. One hour after the radioactive pulse, 0.5 ml of a saturated (approx. 60 mglml) cold methionine solution was injected into all mice to stop the incorporation of labelled methionine to proteins. Groups of two mice were sacrificed at l,1112,3.5Y4 and 9 h after the injection of the radioactive amino acid. Tumors were excised, and the melanosomal, microsomal and soluble tyrosinase isozymes prepared for immunoprecipitation analysis. Preparation and measurement of tyrosinase isozymes Melanosomal, microsomal and soluble fractions were separated by according to Ref. 6. Briefly, melanomas were differential centrifugation, homogenized in 10 mM sodium phosphate buffer (pH 6.81, containing 0.25 M sucrose (1:2 w/v). After centrifugation at 700 x g for 10 min, the post-nuclear fraction was separated and centrifuged at 11,000 x g for 30 min. The resulting melanosome pellet was resuspended in 10 mM sodium phosphate buffer (pH 6.81, and considered as a source of melanosome-bound tyrosinase. The supernatant was further centrifuged at 105,000 x g for 1 h in a Beckman 60 Ti rotor. The new supernatant was filtered through several layers of gauze and considered crude soluble tyrosinase. The pellet from this ultracentrifugation was resuspended in the same buffer that the melanosomal fraction and considered as a source of microsomal tyrosinase. Separately, the melanosomal and microsomal resuspensions were solubilized by incubation for 2 h at 4OC with 1.5% Brij 35 in the same phosphate buffer. Then, both suspensions were again centrifuged at 105,000 x g for 1 h and the supernatants used as crude melanosomal and microsomal tyrosinases. Enzymatic activities of tyrosinase isozymes were determined by measuring spectrophotometrically the rate of dopachrome formation from L-dopa at 475

342

nm [6]. Immunoprecipitation analysis was carried out according to [16] slightly modified: 200 d of tyrosinase isozyme preparations were incubated with 50 ~1 monoclonal antibody for 90 min at 37 OC,then 2.5 ~1 rabbit IgG antirat IgG were added, and incubated 60 min at 37OC. NaOH-neutralized Protein A Sepharose suspension (100 mglmlf was added, and the incubation continued for another 30 min. Then, samples were centrifuged in Eppendorf tubes, removed to Millipore SCWP 025 filter-paper discs, and washed three times with 1 ml 10 mM phosphate buffer (pH 6.81 and once more with 15 ml of the same buffer. Finally, the discs were dried and counted. Blanks were carried out in the absence of monoclonal antibody, and the non-specific precipitation was substracted from each sample to obtain net cpm due to immunoprecipitation of labelled tyrosine. RESULTS

AND DISCUSSION

Figure 1 shows the semilogarithmic plot of percentage of residual tyrosinase activity versus time for melanosomal and soluble fractions of tumors from cycloheximide-treated mice. Microsomal tyrosinase activity dropped to less than 20% of the control activity after only 45 min of cycloheximide treatment (not shown), and the levels of that isozyme were very low in the rest of times studied, indicating the rapid processing of the newly biosynthesized tyrosinase in Harding-Passey melanoma [14]. Melanosomal tyrosinase showed a monophasic behavior, and a half-life of approximately 3% h. However, soluble tyrosinase showed a biphasic behavior, with an abrupt change of slope at 90 min, suggesting the presence of two forms. Assuming that half-lives of both

Fig. 1. Residual tyrosinase activity versus time of cyeloheximide treatment. Mice bearing tumors were treated with 10 mglkg. Melanosomal tyrosinase (01, half-life 205 min. Soluble tyrosinase (0 ), half-lives 30 and 485 min. Melanosomal and soluble isozymes (0 and n l after 3 doses of cycloheximide at 0, 100 and 200 min. The indicated values represent the mean + SD. of three enzymatic determinations.

343

forms are different enough to be analyzed independently, their estimated values are 30 min and 8 h, respectively. Since all tyrosinase isozymes are the same basic polypeptide at different stages of post-translational modification, these apparent half-lives reflect not only degradation or inactivation processes, but also the transfer of isozymes from a pool to another. It could be argued that the cycloheximide treatment did not inhibit protein biosynthesis along the whole time tested. Thus, another experiment was carried out using another group of animals which were i.p. injected with 3 doses of cycloheximide at 0,100 and 200 min, in order to ensure a complete inhibition of protein synthesis. These animals were sacrificed at 300 min, the tumors excised and the activity of melanosomal and soluble tyrosinase determined. The obtained data (squares in Fig. 11showed similar results to those obtained before (using only one dose of cycloheximidel, indicating that the change of slope observed in the residual activity of tyrosinase was not due to a partial inhibition of protein synthesis. The incorporation of [35S]Met into tyrosinase isozymes was investigated by immunoprecipitation using a monoclonal antibody directed against tyrosinase (Fig. 2). Previously, a study of binding of this antityrosinase to Harding-Passey mouse melanoma tyrosinase was carried out since Tomita et al. [16] reported that the monoclonal antibody recognized the mature form of B16 tyrosinase, and did not recognize the precursor forms of the enzyme. Supporting these results, the antityrosinase was able to precipitate the total tyrosinase activity obtained from the melanosomal and soluble melanocyte fractions, and was not able to precipitate approximately 80% of the tyrosinase activity obtained from microsomes. Thus, processing and half-life of this isozyme cannot be determined by this technique. The time courses of incorporation of 35S-labelled methionine into tyrosinase isozymes showed a sharp maximum lV2 h after the radioactive pulse in the melanosomal fraction, and a curve with a smooth maximum after 3 h in the soluble fraction. These times of maximal labelling suggest a precursor-product relationship between melanosomal and soluble tyrosinases. In addition, recognition of the soluble isozyme by antityrosinase supports that isozyme is a mature form having the antigenic determinants recognized by the antibody. Apparent half-lives of tyrosinase isozymes could be estimated from the slopes of the semilogarithmic plots of percentage of specific radioactivity versus time (Fig. 2, inset). Time courses for these plots were initiated at the point of maximal labelling in Fig. 2. It must be taken into account that the accuracy of this method is lower than that one of the cycloheximide treatment. It is due to the strong dependence of all points in the inset to the 100% value taken at the point of maximal labelling of Fig. 2. Half-life determination of soluble tyrosinase could be particularly sensitive to errors since the incorporation of [35S]methionine into that enzyme show a plate curve with a poor defined maximum. However, the obtained values were 5 and 9% h for melanosomal and soluble tyrosinase, respectively. These t,,, are slightly longer than those obtained by cycloheximide treatment, but the differences are similar for both

344

6

9 hours

Fig. 2. Time courses of incorporation of [35S]Met into melanosomal and soluble tyrosinase isozymes (cpm/mg of protein). Mice bearing melanoma were injected first with p%]Met and 1 h later with a large dose of unlabelled Met, as detailed in the text. Groups of two animals were killed at the times indicated in the figure and tyrosinase isozymes separated for immunoprecipitation analysis. The indicated values represent the mean 5 S.D. of two immunoprecipitation experiments. Inset: semilogarithmic plot of decay of the mean tyrosinase labelling after the time of maximal radioactivity incorporated. Melanosomal tyrosinase (0 1.half-life 5 h. Soluble tyrosinase (01, half-life 9% h.

isozymes, approximately 1112 h. Therefore, these differences might also be attributed to the different property of the tyrosinase molecule estimated by both methods: enzymatic activity in the cycloheximide treatment and amount of protein in the metabolic labelling and immunoprecipitation analysis. Thus, such a period (about 1 l/z h for both isozymes) could somehow be an estimation of the time elapsed in the last step of tyrosinase degradation: from the stage when tyrosinase is already catalytically inactive until that molecule is degraded and not recognized by the monoclonal antibody. The steep decay in the soluble tyrosinase activity observed in experiments of protein biosynthesis inhibition during the initial 100 min of cycloheximide treatment (Fig. 1) is not observed in the immunoprecipitation analysis (Fig. 2). Although it is difficult to explain that discrepancy, the form of soluble tyrosinase responsible of that decay, with t,,, about 30 min, could possibly be an intermediate in the processing of tyrosinase from microsomes to melanosomes. During some stage of the transfer, tyrosinase could become soluble or loosely associated to membranous systems, thus appearing in the soluble fraction. Monoclonal antityrosinase would not recognize that immature form, and therefore it would not be detected in the type of experiment. Alternatively, the large dose of unlabelled methionine administered during the chase in the metabolic labelling experiments might also alter the transfer and hide that immature form from being detected.

345

According to this, the two half-lives obtained for soluble tyrosinase suggest a dual origin for the activity found in the cytosolic fraction of Harding-Passey malignant melanocytes. The form with short half-life (l/z h) would be tyrosinase in the process of transferring to the melanosomes, and the form with large halflife (8-9%hl would be tyrosinase coming back from melanosomes partially digested by hydrolytic enzymes. The antitumor action of dopa and related compounds should be favored by the soluble isozymatic form with longer halflife in order to produce the cytotoxic quinones in the cytosol of malignant melanocytes. However, the presence of these compounds could tend to shorten the half-life due to the inactivation undergone by tyrosinase during the catalytic action [15- 171. Further experiments to know the significance of this point are planned to be performed in our laboratory. ACKNOWLEDGMENTS

We are indebted to Dr. V.J. Hearing for the generous gift of monoclonal antibody against tyrosinase. F.S.M. also thanks U.S.-Spanish Joint Committee for Scientific & Technological Cooperation for providing funds to visit N.I.H. U.S.A. This investigation was partially supported by the CAICYT, Spain. REFERENCES 1

7 8

9 10 11 12 13

Delicado, E., Torres, M. and Miras-Portugal, M.T. (19861 Effects of insulin on glucose transporters and metabolic patterns in Harding-Passey melanoma cells. Cancer Res., 46.3762 - 3767. Halaban, R., Pomerantz, S.H., Marshall, S., Lambert, D.T. and Lerner, A.B. (19831 Regulation of tyrosinase in human melanocytes grown in culture. J. Cell Biol., 97.480-488. Hart, D.A. (19861 Evidence that the elevated levels of proteinase activity in the plasma of melanoma-bearing mice may be of host origin. Haemostasis, 16.34 - 42. Korner, A and Pawelek, J. (19821 Mammalian tyrosinase catalyzes three reactions in the biosynthesis of melanin. Science, 217,1163- 1165. Lerner. A.B. and Fitzpatrick, T.B. (19501 Biochemistry of melanin formation. Physiol. Rev., 30, 91- 126. Martinez, J.H., Solano, F., Arocas, A., Garcia-Borron, J.C., Iborra, J.L. and Lozano, J.A. (1987) The existence of apotyrosinase in the cytosol of Harding-Passey mouse melanoma melanocytes and characteristics of enzyme reconstitution by CuUIl. Biochim. Biophys. Acta, 923, 413-420. Pawelek, J. and Lerner, A.B. (19781 56Dihydroxyindole is a melanin precursor showing potent cytotoxicity. Nature, 276,627- 628. Prakash, N.J. Sunkara, P.S. and Sjoerdsma, A. (1985) Potentiation by adifluoromethylornithine of the activity of 3&dihydroxybenzylamine, a tyrosinase-dependent melanolytic agent, against B16 melanoma. Biochem. Pharmacol., 34,1887 - 1890. Riley, P.A. Sawyer, B. and Wolff, M.A. (19751 The melanocytotoxic action of 4-hydroxyanisole. J. Invest. Dermatol., 64.86 - 89. Saeki, H. and Oikawa, A. (1980) synthesis and degradation of tyrosinase in cultured melanoma cells. J. Cell. Physiol., 104, 171- 175. Seiji, M., Shimao, K., Birbeck, M.S.C. and Fitzpatrick, T.B. (19631 Subcellular localization of melanin biosynthesis. Ann. N.Y. Acad. Sci., 100,497 - 533. Sloane, B.F., Dunn, J.R. and Honn, K.V. (19811 Lysosomal cathepsin B. Correlation with metastatic potential, Science, 212,1151- 1153. Sloane, B.F., Honn, K.V. Sadler, J.G.. Turner, W.A., Kimpson, J.J. and Taylor, J.D. (1982)

346

14 15 16 17 18

Cathepsin B activity in B16 melanoma cells: a possible marker for metastatic potential. Cancer Res., 42,980986. Tomita. Y., Hariu, A., Kato, C. and Seiji, M. (1983) Transfer to tyrosinase to melanosomes in Harding-Passey mouse melanoma. Arch. Biochem. Biophys., 225,75-85. Tomita, Y., Hariu, A., Mizuno, C. and Seiji, M. (19801 Inactivation of tyrosinase by dopa. J. Invest. Dermatol., 75,379 - 382. Tomita, Y., Montague, P.M. and Hearing, V.J. (19851 Anti-T,-tyrosinase monoclonal antibodies Specific markers for pigmented melanocytes. J. Invest. Dermatol., 85,426-430. Tomita, Y. and Seiji, M. (19771 Inactivation mechanism of tyrosinase in mouse melanoma. J. Dermatol., 4,245 - 249. Wick, M.M. (19801 Levodopa and dopamine analogs as DNA polymerase inhibitors and antitumor agents in human melanoma. Cancer Res., 40,1414 - 1418.