Carcinogenesis vol.12 no.8 pp.1471 - 1475, 1991

Glutathione transferase isoenzymes in normal and neoplastic human kidney tissue

Carmine Di Ilio, Antonio Aceto, Tonino Bueciarelli, Stefania Angelucci, Mario Felaco1, Alfredo Grilli1, Andrea Zezza2, Raffaele Tenaglia and Giorgio Federici3 Istituti di Scienze Biochimiche, 'Biologia e Genetica e 2Clinica Urologica, Facolta' di Medicina e Chirurgia, Universita' 'G. D'Annunzio', 66100 Chieti and 3Dipartimento di Biologia, Universita' di Roma 'Tor Vergata', 00173 Roma, Italia

Introduction Glutathione transferases (GSTs*; EC 2.5.1.18) are a family of dimeric enzymes that catalyse the conjugation of GSH to a wide variety of reactive electrophiles (1 - 4 ) . This reaction is considered to be the initial step in mercapturic acid formation, a pathway through which hydrophobic xenobiotics, including carcinogens, are inactivated and eliminated from the body. These enzymes also have the ability to act as intracellular binding proteins, participating in transport or storage of exogenous as well as endogenous compounds (1 - 4 ) . In addition to their detoxication function, these proteins have also been ascribed the capacity to participate in prostaglandin and leukotriene biosynthesis as well •Abbreviations: GST, glutathione transferase; GSH, glutathione. © Oxford University Press

Materials and methods Processing of tumour and normal kidney tissue Neoplastic and normal kidney tissues were obtained at operation from 23 patients with renal cell carcinoma, two patients with tumour of the renal pelvis and one patient with hemangioma. None of them had received prior chemotherapy. Patient data are given in Table I. The kidney samples were immediately transferred to cold saline solution, washed exhaustively and homogenized in 4 vol of 10 mM potassium phosphate buffer, pH 7.0, supplemented with 1 mM EDTA and 1 mM dithiothrehol. The homogenate was centrifuged at 105 000 g for 60 min to prepare the cytosolic fractions which were used for GST activity and immunological studies. Immunoblotting for GSTs The antisera used in the present study were prepared in rabbits against GST pi 8.5 of human skin (class alpha) (19), GST V of human uterus (class pi) (20) and GST III of human uterus (class mu) (20). Our antisera recognize GSTs belonging to the same class but do not recognize members of other classes and were the same as those used in previous studies (21-23). For analysis of samples by immunoblotting, aliquots of cytosolic extract containing 100 fig of protein (when blotted against pi and alpha GST antisera) or 130 /ig of protein (when blotted against mu class GST antisera) were resolved by 12% SDS-PAGE by the method of Laemmli (24) and transferred to nitrocellulose paper by standard procedures (25). The following biotinylated SDS-PAGE standards (BioRad) were used: phosphorylase b (97.4 kd), bovine serum albumin (66.2 kd), ovalbumin (42.7 kd), carbonic anhydrase (31 kd), trypsin inhibitor (21.5 kd), lysozyme (14.4 kd). Electroblotting was done for 16 h at 30 V in 25 mM Tris base/192 mM glycine, pH 8.3, containing 20% (v/v) methanol. All incubations were performed at 25°C with intermediate rinses in 50 mM Tris base buffer, pH 7.5,400 mM NaCl (buffer B) containing 0.05% Tween 20 (buffer Q . Non-specific binding was blocked

1471

Downloaded from http://carcin.oxfordjournals.org/ at Purdue University Libraries ADMN on June 6, 2015

Glutathione transferase (GST) activity in the cytosolic fractions of renal cortex tumour was found to be significantly lower (215 ± 156 raU/mg) than that present in the corresponding non-tumour (466 ± 278 mU/mg) tissues. Using the immunoblotting technique, glutathione transferase isoenzymes expression in both tumour and non-tumour kidney was investigated. Alpha and pi class glutathione transferases were the most abundant enzymes in non-tumour kidney and were expressed by all samples investigated. Immunofluorescence analysis indicated that the pi class enzymes are localized mainly in the distal convoluted tubules, whereas alpha class enzymes are localized in the proximal tubules. In the tumour moiety the alpha class GST appears to be absent or expressed at low level as compared with non-tumour samples. On the contrary, no significant differences in the expression of pi class GST were found in tumour as compared with non-tumour tissues. Mu class GST protein was detected in 12 of 26 samples tested. When present, mu class GST constitutes a few per cent of total GST protein. Immunofluorescence studies indicate that mu class GSTs are localized within the distal convoluted tubules. According to the electrophoretic mobility at least two different mu GST subunits (26.5 and 27.5 kd) were found. In one sample only the faster mu class GST subunit was present, two samples expressed both types of GST subunits, whereas nine samples expressed only the slower GST subunit. With the exception of one sample, a reduction of mu class GST expression was seen in tumour as compared with non-tumour tissues. The decrease of activity seen in the cytosolic fraction of tumour kidney must be ascribed mainly to a reduction or to a lack of expression of alpha class GST and to a lesser extent of mu class GST.

as in steroid isomerization (3,4). According to their structural, kinetic and immunological properties, the large number of cytosolic GSTs isolated from mammalian tissues can be conveniently grouped into at least diree classes, i.e. alpha, mu and pi (5). Several reports have indicated that GSTs play an important role in the cellular inactivation of anti-cancer drugs and in some cases the acquisition of enhanced resistance of cancer cells to these drugs has been related to their GST expression (6,7). For example, in human cancer cells, increased resistance to chlorambucil, cis-platinum and adriamycin has been ascribed to elevated levels of pi class GST (8). On the other hand, in the rat gliosarcoma cell line, resistance to nitrosoureas has been linked to the overexpression of mu class GSTs (9). Additional evidence for the relevance of GSTs in die resistance of tumours to chemotherapeutic drugs derives from the study of Lewis et al. (10). They reported that two ovarian adenocarcinoma cell lines derived from a patient before and after the onset of drug resistance to cis-platinum, chlorambucil and 5-fluorouracil had a significant difference in glutadiione (GSH)-dependent enzyme content, including GST. A number of studies have also shown that the amount of pi class GST is higher in human tumour relative to non-tumour tissues (11 — 18). As GSTs may be involved in drug resistance, it is important, from the chemotherapeutic point of view, to characterize the distribution of various GST isoenzymes between tumour and non-tumour tissues. In die present investigation, by using immunoblotting techniques a number of human renal carcinoma samples (a tumour particularly refractory to chemotherapy) were examined widi respect to their GST isoenzyme pattern in order to establish possible differences with the pattern of adjacent non-tumour tissues.

C.Di Dio et al. Table I. GST activity in tumour and non-tumour human kidney tissue Patient

M M F M M M F F M F M M M F F F M M M F F M M M M M

Age

46 63 _« 48 64 46 60 68 42 49 54 68 68 41 53 37 73 68 55 _a

49

Grade

II a

II i i II II

in i i i

i

in i i i II i II i i

1

36 61 a a

i

m _a I II

Mean ± SD (all samples) Mean ± SD (carcinoma samples)

GST activity (mU/mg) Tumour

Non-tumour

Mu class GST (subunit kd value)

400 450 5g0 330 175 312 490 230 161 240 50 120 20 110 100 130 320 40 100 85 65 330 36 500 270 180

1070

27.5

224 ± 182 215 ± 156

456 ± 261 466 ± 278

_a

S25 710 600 580 1040 700 250 250 380 350 770 330 160 220 240 290 210 480 270 460 305 315 350 260

27.5

26.5/27.5

27.5 27.5

26.5 26.5/27.5 27.5

27.5 27.5 27.5 27.5 P < 0.001 P < 0.001

All samples were from primary renal cell carcinoma with the exception of samples 22 and 24 (tumour of renal pelvis) and sample 11 (hemangioma). Mu class GST protein was determined as present or absent in non-tumour tissues by Western blot analysis. "Not available. by placing nitrocellulose papers in buffer B supplemented with 3% bovine serum albumin. The nitrocellulose papers were incubated with appropriately diluted antiserum in buffer B containing 3% bovine serum albumin at 25°C overnight. The nitrocellulose papers were washed with buffer C and then incubated for 1 h at 25°C, with gentle shaking, in the same buffer containing 1% (w/v) gelatin and a horseradish peroxidase-conjugated goat anti-rabbit IgG (BioRad) diluted 1:3000. After treatment with peroxidase-conjugated antibody, the nitrocellulose papers were washed three times with buffer C (5 min each) and twice with buffer B, then immersed in development solution (100 ml buffer B containing 60 mg 4-chloro-l-naphthol (BioRad) and 60 ml 30% hydrogen peroxide). The blot was then washed once with distilled water, air dried and photographed. The concentration of the isoenzyme was determined in a semi-quantitative manner by scanning the bands using a Hofner densitomer. Immunofluorescence study Immunofluoresence studies have been carried out as described by Huang a al. (26). Briefly, paraffin sections were dewaxed, treated with trypsin (0.01 %) for 10 min at 37 °C and then incubated with different primary antisera for at least 60 min at room temperature at 1:200 to 1:1000 dilution in PBS containing 1% normal rabbit serum and 0.02% Triton X-100 (antisera diluting solution). After several rinses with antisera diluting solution the slides were incubated for 60 min at room temperature with anti-rabbit IgG coupled to fluorescein isothkxyanate which was diluted from 1:50 to 1:200 in the antisera diluting solution. After rinsing several times with the antisera diluting solution the slides were rinsed with distilled water and mounted in a 1:1 solution of glycerol and distilled water. The slides were examined and immediately photographed with a Leitz Dialux microscope equipped with a fluorescence illuminator. GST activity GST activity was assayed srjectrophotometncally using l-chloro-2,4-dinitrobenzene (1 mM)andGSH(2mM)inO.l M potassium phosphate buffer (pH 6.5) according to the method of Habig et aL (27). Protein concentration was determined by the method of Bradford (28) with 7-globulin as standard.

1472

Results and Discussion Table I gives the single value ranges, means and their standard errors for GST activity measured in the supernatants prepared from both tumour and non-tumour samples of human kidney. Statistical analysis revealed significantly (P < 0.001) lower activity in tumour than in non-tumour samples. Similar results were obtained in human hepatoma (29,30). However, our results are in marked contrast to the results obtained for human lung (12,18,31,32), breast (33), colon (14,16-17), cervix (34) and gastric tumour (13,18). In these latter tissues a significant increase of GST activity in tumour as compared with the adjacent nontumour tissues was measured. This increase has been ascribed to a higher amount of pi class GST protein and mRNA as evidenced by biochemical and immunochemical studies and by RNA slot blot analysis (12-14,16,17,34). It has to be noted that pi class GST is by far the most abundant enzyme in the abovementioned normal tissues (2) and the reason and the significance of its overexpression in tumour is not yet clear. In order to establish whether the level of expression of specific GST subunits would explain the differences in the specific activity values reported in Table I, the cytosols prepared from both tumour and non-tumour samples were analysed by SDS-PAGE, blotted against human alpha, mu and pi class GSTs antisera and subsequently studied by densitometric analysis of this blot. A representative set of data are reported in Figure 1. The pi class GST, which appears to be mainly localized in the distal tubules (Figure 2A), is also a prominent GST form of human kidney.

Downloaded from http://carcin.oxfordjournals.org/ at Purdue University Libraries ADMN on June 6, 2015

1 2 3 4 5 6 7 g 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Sex

GST Isoenzymes in human kidney tissue

GST pi NT

NT

NT

NT 11

N

T

N

24

T

N

*

8

T 26

S N

T 8

N T 16

/--

27.5 kDa

—v-

26.5 kDa

^ - 27.5 kDa

T

N

N

12

T

11

" ^ - 26.5 kDa

N

T 17

GST alpha

N T N T N T N T N T 18 13 12 11 ~ T ~

N

T 20

N T 21

N 25

Fig. 1. Immunoblotting analysis of GST isoenzyme content of tumour and non-tumour human kidney samples. For pi and alpha classes blot 100 fig of soluble proteins were taken. For mu class blots 130 /ig of soluble proteins were taken.

Semiquantitation of bands in kidney samples indicates that pi class GST represents - 5 5 - 7 5 % of total GST protein. We also found that the level of expression of pi class GST is not significantly changed in tumour tissues. The mean of the tumour/non-tumour ratio for pi class GST calculated by densitometric analysis was 1.14 ± 0.12. These results, which are in agreement with those of Howie et al. (18), do not correlate with the previous results obtained analysing a small number of tumour kidney samples with the isoelectric focusing technique (35). On the other hand, our data are not consistent with those

1473

Downloaded from http://carcin.oxfordjournals.org/ at Purdue University Libraries ADMN on June 6, 2015

GSTmu

recently reported by Klone et al. (36). These authors, by using Northern blot and HPLC analysis found a decrease of pi class GST in tumour as compared with non-tumour tissues. The tumour/non-tumour ratio for the pi class GST found by Klone et al. was 0.50 ± 0.07 with Northern blot analysis and 0.36 ± 0.07 with the HPLC technique. Alpha class GST, which is prevalently localized in the proximal tubules (Figure 2C), although expressed by all samples tested, varied significantly in non-tumour samples, representing —20—45% of total GST protein. Antisera raised against alpha class GST failed to detect expression of this type of protein in most of the tumour samples tested, suggesting that it is either not or weakly expressed. Similar results were recently obtained by Klone et al. by using Northern blots and HPLC analyses (36). It has to be noted that Harrison et al. (37) have demonstrated by immunohistochemical studies that kidney samples from patients with Wilms' tumour failed to express alpha class GST. In tumours derived from liver and stomach a reduction of alpha class GST also occurred (13,18,30). The mu class GST, which is prevalently localized in the distant tubules (Figure 2D), was found to be low in comparison with both alpha and pi class enzymes. The mu class GST, however, is expressed by only 47% of individuals examined. The mean tumour/non-tumour ratio obtained by densitometric analysis was 0.79 ± 0.14. The frequency of mu class GST in human kidney is similar to that observed for normal liver (38), intestine (39) and leukocytes (40). In the previous work we failed to detect mu class GST in human kidney (22). The limited number of samples investigated may be one of the possible explanations for this discrepancy. Mu class GST may play an important role in mutagenesis and carcinogenesis as suggested by the fact that individuals deficient in mu class GST are particularly sensitive to the induction of cytogenetic damage (41) and lung cancer risk (42). At least two electrophoretically different mu class GST subunits (26.5 and 27.5 kd) (Figure 1) are present in human kidney. Sample 16 expressed only the 26.5 kd GST subunit, whereas samples 8 and 17 expressed the 27.5 kd in addition to the 26.5 kd subunit. Owing to its poor staining on the gel, the 26.5 kd subunit of sample 8 could not be well reproduced photographically. The GST subunit with lower mobility may be identical or related to the mu type GST /t/GST * isolated from liver (38) and uteri (20), since its physicochemical properties are rather similar (27.5 kd). The presence of two mu type GST subunits was also detected, with immunoblotting analysis, in the cytosols prepared from some samples of human duodenum, jejunum and ileum as well as from human mononuclear blood cells (39). As suggested by its electrophoretic mobility, the 26.5 kd GST subunit of human kidney may be related to GST^ (N2N2 in an alternative nomenclature) found in human heart, muscle and aorta (43-45). This GST subunit was detected in only 3 out of 26 samples of kidney tested, whereas GST-4 was found in all the 9 samples of human muscle tested (45). It is interesting to note that Carmichael et al. (33) found in all samples of human lung tested (10 samples) a cross-reactive protein on Western blot with antisera against mu class GST whose electrophoretic characteristics appear quite similar to our kidney faster mu class GST subunit. However, it is not clear at present whether the faster mu class GST subunit of human kidney is a novel enzyme or represents one of the extrahepatic mu class GSTs already described (33,38,43-45). It is evident from the results obtained that both alpha and mu classes GSTs are absent or present at very low level in many tumour samples relative to matched non-tumour tissues. Thus, all together our results suggest that the expression of alpha and to a lesser extent mu class GSTs

are down-regulated in kidney tumour. However, it is intriguing that in one patient (patient 16) the reversed situation for mu class GST was found. Here the mu class GST subunit (26.5 kd) was slightly overexpressed in tumour. The tumour/non-tumour ratio value for mu class GST in this patient was 1.1. It is worth noting that the expression of pi and mu classes of GST are differentially affected by transformation, although they are both prevalently localized in the same nephron region, i.e. the distant convoluted tubules. Thus, from the results obtained it may be concluded that the decrease of activity measured in the cytosol of tumour kidney must be ascribed to a decrease of alpha class GST and to a lesser extent to mu class GST that is not compensated by the slight increase of pi class GST. The alteration in the GST isoenzyme distribution that occurs between the tumour and non-tumour tissues belonging to the same patient is of particular importance and suggests that the two tissues may have different responses to chemotherapeutic treatment. Once the difference in the GST isoenzyme expression between tumour and non-tumour kidney has been established, specific drugs could be chosen to plan an individualized cancer chemotherapy.

Acknowledgements We would like thank Miss Graziella Soldato for excellent technical assistance in performing the immunofluorescence analysis.

1474

References 1. Chasseaud,L.F. (1979) The role of glutathione and glutathione S-transferases in the metabolism of chemical carcinogens and other electrophilic agents. Adv. Cancer Res., 29, 175-293. 2. Mannervik,B- (1985) The isoenzymes of glutathione transferases. Adv. Enzymol.. 57, 357-417. 3. Jakoby,W.B. and Habig,W.H. (1980) Glutathione transferases. In Jakoby.W.B. (ed.), Enzymatic Basis of Detoxkation. Academic Press, New York, Vol. 2, pp. 63-69. 4. Ketterer,B., Meyer,D.J. and Clark,A.G. (1988) Soluble glutathione transferase isoenzyme. In Sies,H. and Ketterer,B. (eds), Glutathione Conjugation, Mechanisms and Biological Significance. Academic Press, London, pp. 73-135. 5. Mannervik,B., Alin,P., Gurhenberg.C, Jensson.H., Tahir,M.K., Warholm.M. and Jornvall,H. (1985) Identification of three classes of cytosolic glutathione transferase common to several mammalian species: correlation between structural data and enzymatic properties. Proc. Nail. Acad. Sci. USA, 82, 7202-7206. 6. Hayes,J.D. and Wolf,C.R. (1988) Role of glutathione transferase in drug resistance. In Sies.H. and Ketterer.B. (eds), Glutathione Conjugation, Mechanisms and Biological Significance. Academic Press, London, pp. 315-355. 7. Wolf.C.R., Wareing,C.J., Black.S.M. and Hayes,J.D. (1990) Glutathione S-transferases in resistance to chemotherapeutic drugs. In Hayes,J.D., Pickett,C.B. and Mantle,T.J. (eds), Glutathione S-transferases and Drug Resistance. Taylor & Francis, London, pp. 295-307. 8. Tew.K.D., Schisselbauer.J.C, Clapper.M.L. and Kuzmich.S. (1990) Glutathione S-transferases and resistance to alkylating agents. In Hayes,J.D., Pickett,C.B. and Mantle,T.J. (eds), Glutathione S-transferases and Drug Resistance. Taylor & Francis, London, pp. 309-317.

Downloaded from http://carcin.oxfordjournals.org/ at Purdue University Libraries ADMN on June 6, 2015

Fig. 2. (A) Immunofluorescent staining with (A) pi class GST of a section of human non-tumour kidney (original magnification X250); (B) pi class GST of a section of human tumour kidney (original magnification X250); (C) alpha class GST of a section of human non-tumour kidney (original magnification X250); (D) mu class GST of a section of human non-tumour kidney (original magnification x250).

GST Isoenzymes in human kidney tissue Glutathione peroxidase, glutathione S-transferase and glutathione reductase activities in normal and neoplastic human breast tissue. Cancer Lett., 29, 37-42. 34. Shiratori,Y., Soma,Y., Maruyama,H., Sato.S., Takano.A. and SatoJC. (1987) Immunochemical detection of the placental form of glutathione S-transferasc in dysplasn'c and neoplastic human uterine cervix lesion. Cancer Res., 47, 6806-6809. 35. Di Dio,C, Del Boccio.G., Aceto.A. and Federici.G. (1987) Alteration of glutathione transferase isoenzyme concentration in human renal carcinoma. Cardnogenesis, 8, 861—864. 36. Klone.A., Weinder.U., HuBnatter.R., Harris^., Meyer.D., Peter.S., Ketterer.B. and Sies.H. (1990) Decreased expression of the glutathione Stransferases alpha and pi genes in human renal cell carcinoma. Cardnogenesis, 11, 2179-2183. 37. Harrison.D.J., Hallam.L. and Laudcr.J. (1990) Glutathione S-transferase expression in fetal kidney and Wilms' tumour. Br. J. Cancer, 61, 836 — 840. 38. Warholm.M., Guthenberg.C. and Mannervik.B. (1983) Molecular and catalytic properties of ghitafhione transferase n from human liver: an enzyme efficiently conjugating epoxides. Biochemistry, 22, 3610-3617. 39. Peters.W.H.M., Kock.L., Nagengast.F.M. and Roelofs.H.M.J. (1990) Immunodetection with monoclonal antibody of glutathione S-transferase mu in patients with and without carcinomas. Biochem. Pharmacol., 39, 591 -597. 40. SekfegardJ., Guthenberg.C., Pero.R.W. and Mannervik,B. (1987) The transstilbenc oxide-active glutathione transferase in human mononuclear leucocytes is identical with the hepatic glutathione transferase p. Biochem. J., 246, 783-785. 41.WienckeJ.K., Kelsey,K.T., Lamela.R.A. and Toscano,W.A., Jr (1990) Human glutathione S-transferase deficiency as a marker of susceptibility to epoxide-induced cytogenetic damage. Cancer Res., 50, 1585 — 1590. 42. SiedeganU., Pcro,R.W., Milkr.D.G. and Beattie.E.J. (1986) A glutathione transferase in human leukocytes as a marker for the susceptibility to lung cancer. Cardnogenesis, 7, 751-753. 43,Board,P.G., Suzuki.T. and Shaw.D.C. (1988) Human muscle glutathione S-transferase (GST-4) shows close homology to human liver GST-1. Biochim. Biophys. Ada, 953, 214-217. 44. Tsuchida.S., Maki,T. and Sato.K. (1990) Purification and characterization of glutathione transferases with an activity toward nitroglycerin from human aorta and heart. / . Biol. Chem., 265, 7150-7157. 45. Suzuki.T., Coggan.M., Shaw.D.C. and Board.P.G. (1987) Electophoretic and immunological analysis of human glutathione S-transferase isoenzyme. Ann. Hum. Genet., 51, 95-106. Received on February 25, 1991; revised on May 4, 1991; accepted on May 6, 1991

Cardnogenesis, 9, 1617-1621. 33. Di Dio.C, Sacchetta.P., Del Boccio.G., La Rovere.G. and Federici.G. (1985)

1475

Downloaded from http://carcin.oxfordjournals.org/ at Purdue University Libraries ADMN on June 6, 2015

9. Smith.M.T., Evans,C.G., Doeane-Setzer.P., Castro.V.M., Thair.M.K. and Mannervik.B. (1989) Denitrosation of l,3-bis(2-chloroethyl)-l-nitrosourea by class mu glutathione transferase and its role in cellular resistance in rat brain tumor cells. Cancer Res., 49, 2621-2625. 10. Lewis,A.D., HayesJ.D. and Wolf,? (1988) Glutathione and glutathionedependent enzymes in ovarian adenocarcinoma cell lines derived from a patient before and after the-onset of drug resistance: intrinsic differences and cell cycle effects. Cardnogenesis, 9, 1283-1287. 11. Shea.T., Kelley.S.C. and Henner.W.D. (1988) Identification of an anionic form of glutathione transferase present in many human tumors and human cell lines. Cancer Res., 48, 527-533. 12. Di Ilio,C., Del Boccio.G., Aceto.A., Mucilli.F. and Federici.G. (1988) Elevation of glutathione transferase activity in human lung tumor. Cardnogenesis, 9, 355-340. 13. Peters.W.H.M., Wormskamp,N.G.M. and Thies.E. (1990) Expression of glutathione S-transferases in normal gastric mucosa and in gastric tumors. Cardnogenesis, 11, 1593-1596. 14. Peters.W.H.M., Nagengast.F.M. and Wobbes.T. (1989) Glutathione Stransferases in normal and cancerous human colon tissue. Cardnogenesis, 10, 2371-2374. 15. Mannervik.B., Castro.V.M., Danielson.U.H., Tahir.M.K., HanssonJ. and Ringborg,U. (1987) Expression of class pi glutathione transferase in human malignant melanoma cells. Cardnogenesis, 8, 1929-1932. 16. MoscowJ.A., Fairchild.C.R., Madden.M.J., Ransom.D.T., Wicand.H.S., O'Brien.E.E., Poplack.D.G., CossmanJ., Myers.C.E. and Cowan.K.H. (1989) Expression of anionic glutathione S-transferase and P-glycoprotein genes in human tissues and tumors. Cancer Res., 49, 1422 — 1428. 17. Tsuchida.S., Sekine.Y., Shineha.R., Nishihira,T. and Sato.K. (1989) Elevation of the placenta! glutathione S-transferase form (GST T) in tumor tissues and the levels in sera of patients with cancer. Cancer Res., 49, 5225-5229. 18.Howie,A.F., Forrester.L.M., Glancey.MJ., SchlagerJ.J., Powis.G., Beckett,G.J., Hayes.J.D. and Wolf.C.R. (1990) Glutathione S-transferase and glutathione peroxidase expression in normal and tumour human tissues. Cardnogenesis, 11, 451-458. 19. Del Boccio,G,. Di Hio,C, Alin.P., Jornvall.H. and Manncrvik.B. (1987) Identification of a novel glutathione transferase in human skin homologous with class alpha glutathione transferase 2-2 in the rat. Biochem. J.,144,21-25. 20. Di flio.C, Aceto.A., Del Boccio.G., Casalone.E., PennelliA and Federici.G. (1988) Purification and characterization of five forms of glutathione transferase from human uterus. Eur. J. Biochem., 171, 491-496. 21. Di Ilio,C, Aceto.A., Piccolomini,R., Allocau.N., Faraone,A., Cellini,L., Ravagnan.G. and Federici,G. (1988) Purification and characterization of three forms of glutathione transferase from Proteus mirabilis. Biochem. J., 255, 971-975. 22. Di Hio,C, Aceto.A., Zezza.A., Ricci,G. and Federici.G. (1989) Electrophoretic and immunological analysis of glutathione transferase isoenzymes of human kidney. Biochem. Pharmacol., 38, 1045 — 1051. 23. Di Dk>,C, Aceto.A., Bucciarelli,T., Angelucci.S., Felaco.M., Grilli.A. and Federici,G. (1990) Glutathione transferase isoenzymes from human prostate. Biochem. J., 271, 481-485. 24. Laemmli.U.K. (1970) Cleavage of structural proteins during the assembly of the head bactenophage T4. Nature, 227, 680-685. 25. Towbin,H., Stachelin.T. and Gordon,L. (1979) Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sd. USA, 76, 4350-4354. 26. Huang,S.-N., Minassian.H. and MooreJ.D. (1976) Application of immunofluorescent staining on paraffin sections improved by trypsin digestion. Lab. Invest., 35, 383-390. 27. Habig,W.H., Pabst.M.J. and Habig.W.B. (1974) Glutathione transferases, the first step in mercapturic acid formation. J. BioL Chent, 249, 7130-7139. 28. Bradford,M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein-dye binding. Anal. Biochem., 72, 248-254. 29. El Mouelhi,M., Didolkar.M.S., Elias.E.G., Guengerich.F.P. and Kauffman.F.C. (1987) Hepatic drug-metabolising enzymes in primary and secondary tumors of human liver. Cancer Res., 47, 460—466. 30. Sherman.M., Campbell.J.A.H., Titmus,S.A., Kew.M.C. and Kirsch.R. (1983) Glutathione S-transferase in human hepatocellular carcinoma. Hepatology, 3, 170-176. 31. Singh.S.V., Haque.A.K., Ahmad.H., Medh.R.D. and Awasthi.Y.C. (1988) Glutathione S-transferase isoenzymes in human lung tumor. Carcwogenesis, 9, 1681-1685. 32.CarmichaeU., Forrester.L.M., Lewis.A.D., HahesJ.D., Hayes,P.C. and Wolf.C.R. (1988) Glutathione S-transferase isoenzymes and glutathione peroxidase activity in normal and tumour samples from human lung.