Tohoku

J. Exp.

Specific

Med.,

1992, 168, 97-103

Expression

of Glutathione

S-transferase Pi Forms in (Pre)neoplastic Tissues : Their Properties and Functions KIYOMI SATO, KIMIHIKOSATOH, SHIGEKITSUCHIDA, ICHIRO HATAYAMA,HONGXIE SHEN, YOSHTHITO YOKOYAMA, YUICHIYAMADA and KATSUTO TAMAI* The Second Department of Biochemistryand Department of Dermatology*, Hirosaki University School of Medicine, Hirosaki O16 SATO,K., SATOH,K., TSUC$IDA,S., HATAYAMA, I., SHEN,H., YOKOYAMA, Y., YAMADA, '. and TAMAI,K. Specific Expression of Glutathione S-transferase Pi Forms in (Pre)neoplastic Tissues : Their Properties and Functions. Tohoku J. Exp. Med., 1992, 168 (2), 97-103 The detection of preneoplastic cells is very important for the analysis of carcinogenic processes and for developing strategies for prevention and treatment of cancer. We have been investigating enzyme alterations occurring during rat chemical hepatocarcinogenesis, especially to find more specific enzyme markers for preneoplastic hepatic lesions. We identified the placental form of glutathione S-transferese (GST-P ; GST 7-7) as a new marker enzyme for preneoplastic hepatocytes. We also found human placental form, GST-7c,to be a possible tumor marker for various human tissues except liver. In this article, their properties and possible functions are reviewed on basis of our recent investigations. A peroxisomal enzyme, enoyl CoA hydratese, in also described as a possible negative marker for rat preneoplastic hepatic foci/nodules and hepatomas induced by peroxisome proliferators. (pre)neoplastic marker ; glutathione S-transferase ; peroxisome proliferator ; enoyl CoA hydratase

Chemical carcinogenesis is thought to proceed through initiation and promotion stages. During the latter, in rat hepatocarcinogenesis, preneoplastic cells such as enzyme-altered foci and hyperplastic nodules appear, which express specific marker enzymes (see a review by Sato 1989). We thought chemical carcinogenesis depends on the balance between activation and inactivation of carcinogens and have focused our investigations on drug-metabolizing enzymes, and in particular on isoenzyme alteration of glutathione S-transferase ( GST ), one of the detoxicating enzymes, in rat preneoplastic hepatic lesions. GST has multi-functions performed by multi-molecular forms (isoenzymes) (Sato 1989; Sato and Tsuchida 1991). So far, a large number of isoenzymes have been identified from many species and grouped into Class Alpha, Mu, Pi, Theta Address

for reprints

: Zaifu-cho

5, Hirosaki 97

036, Japan.

98

K. Sato et al. TABLE 1.

Representative

rat, human

and

mouse

GSTs

and others (Sato and Tsuchida 1991). Representative forms of rat, humam and mouse GSTs are shown in Table 1. GSTs in the Alpha and Mu classes are abundantly present in normal livers, and basic forms in the former class act as binding proteins for various dyes, steroids and carcinogens, which were once known as ligandin. GSTs in the Mu class play main roles in detoxication of many carcinogens. Thus, GSTs in the Alpha and Mu classes, which are inducible by drugs in normal livers, are known to inhibit "initiation" of chemical carcinogenesis by detoxication of carcinogens and to act for chemoprevention or modification of carcinogenesis. GSTs in the Pi class have attracted attention of cancer researcher as (pre) neoplastic markers. GST forms including those in the Pi class, which are increased in preneoplastic cells, are thought to enhance "promotion" by participating in acquired resistance mechanisms of these cells to cytotoxic agents. Properties of GST P (7-7) as rat hepatic preneoplastic marker We identified a new GST form from rat placenta and named it the placental form, GST-P, since this form was almost only one form in this tissue (Sato et al. 1984). It is a neutral and homodimeric protein and smallest among rat cytosoldic GST forms. GST-P levels in normal rat tissues including fetal, adult and regenerating livers, are very low and even in the placenta. Among normal tissues, the kindney contains a highest amount of GST-P (287+ 76.9 j g/g tissue). However, livers bearing preneoplastic foci express it (1,160± 5201ag/g tissue) much higher than any normal tissues (Satoh et al. 1985). Immunohistochemically using the anti-GST-P antibody, GST-P was demonstmated to be expressed within foci (Sato et al. 1984). Furthermore, single GST-P-positive cells can be detected at an early stage ; e.g., within a few days after diethylnitrosamine administration (Satoh et al. 1989). Some of these putative initiated cells and related minifoci persist for a long time and develop to form large foci under growth promoting conditions such as hepatectomy (Satoh et al. 1989). These results may indicate clonal origin

Glutathione

S-transferase

Pi in (Pre)neoplastic

Tissues

99

of preneoplastic foci and hepatomas. GST-P has unique properties as marker useful for detection of preneoplastic hepatic foci (Sato 1988, 1989). It is not immunocrossreactive with other GST fomrs abundantly present in normal rat livers. It is not detectable in normal hepatocytes, but markedly increased and localized within foci induced by most of hepatocarcinogens. Contrary to other drug metabolizing enzymes, it is not inducible by drugs or carcinogens prior to appearance of foci, except for lead nitrate, which transiently induces GST-P in whole livers (Sato 1988, 1989). Exceptionally, however, it is not expressed in foci and hepatomas induced by nongenotoxic peroxisome proliferators such as clofibrate. GSTs in the Pi class in other species, such as human GST-n, are also possible tumor markers (Sato 1989). GST-n is several-fold increased in preneoplastic and neoplastic tissues of various organs such as esophagus, colon, uterine cervix, breast lung (non-small cell carcinoma), brain (glioma) and skin (melanoma), as compared with that in the normal counterpart (Sato 1989; Tsuchida et al. 1989). Enzymatic properties of GSTs in the Pi class GST-P and other GSTs in the Pi class have unique enzymatic properties ; low sensitivity to organic anion inhibitors (Satoh et al. 1991) and high sensitivity to sulfhydryl-modifiers and active oxygen species (Tamai et al. 1990, 1991; Shen et al. 1991). Rat GST-P, mouse type II and human GST-n, all in the Pi class, are timedependently inactivated with low concentrations (0.05-0.1 mM) of Nethylmaleimide (NEM), an SH-blocker, while other forms in other classes are not (Tamai et al. 1990). GST-P or human ~r is also inactivated with 1.0 mM cystamine and cystine, disulfides (Tamai et al. 1991), and similarly with 1.0 mM hydrogen peroxide (11202) (Shen et al. 1991), while other forms in other classes are not. However, in these cases, once inactivated GST-P (-n) is reactivated by addition of dithiothreitol (DTT) or mercaptoethanol (MCE ), reducing reagents (Tamai et al. 1991) or by thioltransferase (T. Tase) and reduced glutathione (GSH) (Shen et al. 1991). These results suggest that cysteine (Cys) residues may be involved in the inactivation. Therefore, the Cys residue(s) sensitive to these reagents were identified. We digested GST-P protein with trypsin after treatment with a colored analogue of NEM, the colored NEM-bound peptides were isolated by HPLC and their amino acid sequences were determined (Tamai et al. 1990). Among 4 Cys residues, the 47th, which is unique for all GST forms in the Pi class (Fig. 1), was found to be responsible for NEM inactivation (Tamai et al. 1990). These results were confirmed using mutants of a recombinant GST-P expressed in E. coli (JM109) by site-directed mutagenesis, in which all 4 Cys residues were independently substituted with alanine (Ala) (Tamai et al. 1991). The mutant A1a47 only became resistant to the irreversible inactivation with 0.1 mM

K. Sato

100

Fig. 1.

Position

of Cy s residues

et al.

in amino

acid

sequences

of GSTs.

NEM, whereas the other three mutants remained as sensitive as the nonmutant (wild) type. However, it was noted that all mutants have similar specific activities to those of the wild type and liver GST-P, indicating that none of the four Cys residues is essential for GST-P activity (Tamai et al. 1991). To the inactivation by 1.0 mM 11202, A1a101 was most resistant and A1a47 was less than A1a101 but still more resistant than the wild type, whereas A1a169 and A1a14 were more sensitive than the wild type to the same treatment. After treatment with 1.0 mM 11202 for 30 min, all mutants, the wild type and GST-P from rat livers were subjected to SDS-PAGE under nonreducing conditions without MCE, as described previously (Shen et al. 1991; Tamai et al. 1991). Extra bands smaller than native GST-P subunit band with Mr 23.5 kDa were observed. These bands were identified as inactive GST-P subunits with apparently reduced molecular weights produced in association with disulfide bond formation within a subunit, since they returned to the native band with accompanying restoration of the activity when treated with DTT, MCE or T. Tase

Fig. 2. Inactivation of GST-P (-7r) by NEM, disulfides and 11202, and reactivation with T. Tase, DTT or MCE. T. Tase, thioltransferase ; DTT, dithiothreitol ; MCE, mercaptoethanol.

GlutathioneS-transferase Pi in (Pre)neoplastic Tissues

101

(Shen et al. 1991). For example, the band with Mr 21.5kDa was lacked in A1a47 and 101, so this band was considered to be produced in association with disulfide bond formation between Cys47 and Cys101 within a subunit (intrasubunit disulfide bond formation). Extra bands larger than the native one are also observed. They seem to be a dimer or polymers associated with inteasubunit disulfide bond formation, in which mainly the 47th and 101st Cys residues also partipate. On the basis of these results, possible reactions are summarized in Fig. 2. NEM irreversibly inactivates GST-P or -n by covalent modification of the 47th Cys residue. Disulfides such as cystamine inactivate GST-P partly by protein-mixed disulfide formation with the 47th Cys residue and further by intrasubunit disulfide bond formation between the 47th and 101st Cys residues. 11202 inactivates GST-P or n by intrasubunit and intrasubunit disulfide bond formation between 47th and 101st residues. The inactive form thus produced can be reactivated enzymatically by T. Tase and GSH or by addition of DTT or MCE. Possible roles of GST P (-pr)in drug resistant cells According to previous results and others we propose possible detoxicating functions of GST-P or -Tr in drug resistant cells (Fig. 3). These forms may detoxify alkylating agents by GSH conjugation. Under oxidative stress conditions such as adriamycin and mitomycin C administration, however, reactive oxygen species are known to be produced. These compounds are usually inactivated by superoxide dismutase (SOD), catalase, GSH peroxidase (GSH Px), etc. Some of them may produce lipid peroxides, which, however, can be reduced

Fig. 3. Possible detoxicating functions of GST-P (-n) in drug-resistant cells and the irreversible inactivation by SR-blockers. (from Sato and Tsuchida 1991 with permission)

K. Sato

102

et al.

by GSH Px activity of GST-P or -2t, by their second function. Furthermore, reactive oxygen species can be also removed in association with inactivation of these forms, by the third function. These enzymatic properties of GST-P (or -n) might be helpful for drug-resistance mechanism(s) of resistant cells. GST-P or -7c, however, can be irreversibly inactivated by 811-blockers such as NEM, whereby the drug resistance may be overcome. Markers for foci/nodules and hepatomas induced by PPs As mentioned above, GST-P is expressed in preneoplastic foci and hapatomas induced with all genotoxic, mutagenic hepatocarcinogens so far examined and with some nongenotoxic ones such as ethionine (Sato 1989). However, peroxisome proliferators (PPs) such as clofibrate, which is used clinically as hypolipidemic agents, induce GST-P-negative nodules and carcinomas in rat livers. PPs are thought to perform carcinogenic activity by 11202 production (Reddy and Rao 1986). We have tried to find a marker for detection of these lesions and further to clarify the mechanism(s), by which GST-P is not expressed in them. We examined the expression of peroxisomal exzymes in foci/nodules and hepatomas and found several enzymes such as enoyl CoA hydratase and acyl CoA oxidase are either not or only weakly expressed in most of them, while being strongly induced in surrounding hepatocytes, suggesting that enoyl CoA hydratase and other peroxisomal enzymes may be negative markers for foci/nodules and hepatomas induced by PPs (Yokoyama et al. 1992). These foci/nodules and hepatomas are all negative for GST-P, as described previously (Reddy and Rao 1986; Sato 1988, 1989). In conclusion, GST-P is a positive marker for rat hepatic preneoplastic foci/ nodules and hepatomas induced by genotoxic and some nongenotoxic hepatocarcinogens, while enoyl CoA hydratase may be a negative marker for these lesions induced by nongenotoxic PPs. Acknowledgments Our Ministry

studies

were

of Education,

supported Science

in part and

by

Culture

Grants-in-Aid

for Cancer

Research

from

the

of Japan.

References 1) Reddy, J.K. & Rao, M.S. (1986) Peroxisome proliferators and cancer : Mechanisms and inplications. Trends Pharmacol. Sci., 7, 438-443. 2) Sato, K. (1988) Glutathione S-transferases and hepatocarcinogenesis. Jpn. J. Cancer Res., 79, 556-572. 3) Sate, K. (1989) Glutathione transferases as markers of preneoplasia and neoplasia. Adv. Cancer Res., 52, 205-255. 4) Sate, K. & Tsuchida, S. (1991) Glutathione transferases in normal, preneoplastic, and neoplastic tissues : Forms and functions. In : Biochemical and Molecular Aspects of Selected Cancers, edited by T.P. Pretlow & T.G. Pretlow, II, Vol. 1, Academic Press, San Diego, pp. 177-226.

Glutathione S-transferase Pi in (Pre)neoplastic Tissues

103

5) Sato, K., Kitahara, A., Satoh, K., Ishikawa, T., Tatematsu, M. & Ito, N. (1984) The placental form of glutathione S-transferase as a new marker protein for prenoeplasia in rat chemical hepatocarcinogenesis. Gann, 75, 199-202. 6) Satoh, K., Kitahara, A., Soma, Y., Inaba, Y., Hatayama, I. & Sato, K. (1985) Purification, induction, and distribution of placental glutathione transferase : A new marker enzyme for preneoplastic cells in the rat chemical hapatocarcinogenesis. Proc. Natl. Acad. Sci. USA, 82, 3964-3968. 7) Satoh, K., Hatayama, I., Tateoka, N., Tamai, K., Shimizu, T., Tatematsu, M., Ito, N. & Sato, K. (1989) Transient induction of single GST-P positive hepatocytes by DEN. Carcinogenesis,10, 2107-2111. 8) Satoh, K., Hatayama, I., Tsuchida, S. & Sate, K. (1991) Biochemical characteristics of a preneoplastic marker enzyme glutathione S-transferase P-form (7-7). Arch. Biochem. Biophys., 285, 312-316. 9) Shen, H., Tamai, K., Satoh, K., Hatayama, I., Tsuchida, S. & Sate, K. (1991) Modulation of class Pi glutathione transferase activity by sulfhydryl group modification. Arch. Biochem. Biophys., 286, 178-182. 10) Tamai, K., Sateh, K., Tsuchida, S., Hatayama, I., Maki, T. & Sate, K. (1990) Specific inactivation of glutathione S-transferases in class Pi by SH-modifiers. Biochem. Biophys. Res. Commun., 167, 331-338. 11) Tamai, K., Shen, H., Tsuchida, S., Hatayama, I., Satoh, K., Yasui, A., Oikawa. A. & Sato, K. (1991) Role of cysteine residues in the activity of rat glutathione transferase P (7-7). Biochem. Biophys. Res. Commun., 179, 790-797. 12) Tsuchida, S., Sekine, Y., Shineha, R., Nishihira, T. & Sate, K. (1989) Elevation of the placental glutathione S-transferase form (GST-n) in tumor tissues and the levels in sera of patients with cancer. Cancer Res., 49, 5225-5229. 13) Yokoyama, Y., Tsuchida, S., Hatayama, I., Sateh, K., Narita, T., Rao, MS., Reddy, J.K., Yamada, Y., Suga, T. & Sate, K. (1992) Loss of peroxisomal enzyme expression in preneoplastic and neoplastic lesions induced by peroxisome proliferators in rat livers. Carcinogenesis,13, 265-269.

Specific expression of glutathione S-transferase Pi forms in (pre)neoplastic tissues: their properties and functions.

The detection of preneoplastic cells is very important for the analysis of carcinogenic processes and for developing strategies for prevention and tre...
683KB Sizes 0 Downloads 0 Views