Modulation of glutathione S-transferases and glutathione peroxidase ...

8 downloads 6 Views 6MB Size Report
'Current address: Biomedical Research Centre, Ninewells Hospital and. Medical School, University of Dundee, Dundee DD1 9SY, Tayside, UK. Induction of ...
Carcinogenesis vol.13 no.12 pp.2255-2261, 1992

Modulation of glutathione S-transferases and glutathione peroxidase by the anticarcinogen butylated hydroxyanisole in murine extrahepatic organs

Lesley LMcLellan1, David J.Hanison 2 and John D.Hayes1 Department of Clinical Biochemistry, University of Edinburgh, Royal Infirmary, Edinburgh EH3 9YW and 2Department of Pathology, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, UK 'Current address: Biomedical Research Centre, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, Tayside, UK

Induction of glutathione S-transferases (GST) by the anticarcinogen butylated hydroxyanisole (BHA) has been examined in lung, kidney and small intestine of male and female BALB/c mice. BHA produced maximal induction of GST hi the gut and although it increased GST levels in the kidney, it had little effect on pulmonary GST. Dietary BHA induced Alpha (Ya and Yk), Mu (Yb) and Pi (Yf) class GST subunits at least 10-fold in the small intestine but, by contrast, selenium-dependent glutathione peroxidase activity was reduced by ~ 4-fold hi this organ following BHA treatment. In the kidney, all of the GST subunits, apart from Yk hi males, showed modest levels of induction by BHA. However, a pronounced sex difference hi the expression of renal alpha class subunits hi both control and BHA-treated mice was observed, with female mice expressing - 4-fold greater levels of Ya and Yk than male mice. All renal GST were localized primarily hi the proximal tubules. Dietary BHA was found to have the least inductive effect hi the lung, where the GST were localized solely hi the bronchi. The pulmonary Mu class GST subunits were induced ~ 2-fold by BHA; the expression of other GST was marginally increased by this inducer. Alpha class GST was also subject to sexual differentiation in the lung with female mice possessing higher levels of Yc and Yk than males. The Ya-type subunit was not detected in the lung nor was it Induced by BHA.

Introduction The ability of an animal to resist chemical carcinogenesis can often be augmented by administering compounds known as chemoprotectors (or anticarcinogens) prior to, or during, the experimental protocol used to induce carcinogenesis (1,2). Several categories of anticarcinogenic agent have been described (reviewed in 3) which have a wide variety of biological effects. Of these, the phenolic antioxidant, butylated hydroxyanisole (BHA*), has attracted considerable attention because of its relatively low toxicity but profound anticarcinogenic effect. This effect is thought primarily to be mediated by alterations in levels of xenobiotic-metabolizing enzymes, particularly by induction of those involved in conjugation reactions, such as the glutathione S-transferases (GST, EC 2.5.1.18). •Abbreviations: BHA, butylated hydroxyanisole; GST, glutathione S-transferase; CDNB, l-chloro-2,4-dinitrobenzene; DCNB, 1,2-dichloro-4-nitrobenzene; fPBO, /ra/ir-4-phenylbut-3-en-2-one; EA, ethacrynic acid; CuOOH, cumene hydroperoxide.

© Oxford University Press

GST are a super-gene family of isoenzymes that are responsible for catalysing the conjugation of a wide range of structurally diverse electrophilic xenobiotics and endogenous substrates with glutathione, facilitating their excretion from the body. At least five separate gene families of GST exist; these include both cytosolic (named Alpha, Mu, Pi and Theta) and a microsomal GST (4-7). The cytosolic GST are dimeric enzymes with estimated subunit Mr values ranging from 24 800 to 27 500 (8) and they exist as either homo- or heterodimers. Control of expression of GST is multifactorial. In addition to tissue-specific expression (9), the expression of certain GST in rodents is subject to sex-specific control (10—12). Furthermore, several GST isoenzymes can be induced by xenobiotics like phenobarbital and 3-methylcholanthrene and GST activity towards l-chloro-2,4-dinitrobenzene has been shown to be elevated by many compounds known to inhibit chemical carcinogenesis (13). The effect of dietary BHA on levels of individual GST isoenzymes in mouse liver has been studied both at the protein and mRNA level. The extent of elevation of GST enzymes can be as much as 10- to 30-fold. The BHA-mediated induction of a Pi class GST (subunit Yf, Mr 24 800), two Alpha class GST (subunit Ya, Mr 25 600) and several Mu class GST (subunit Yb, Mr 26 200-26 500) has been observed (14-18). A constitutive hepatic murine Alpha class GST (subunit Yc, M r 25 800), which was previously named Ya3 but has recently been redesignated (17,19,20), is not perceptibly induced by BHA treatment, nor is the microsomal GST inducible. The liver is the major site in mammals for metabolism and detoxication of drugs and carcinogens. Increased hepatic expression of xenobiotic-metabolizing enzymes like GST is likely to make a major contribution to the anticarcinogenic effect of BHA. However, induction of these enzymes in extrahepatic organs may also affect the potency of certain carcinogens, particularly as the liver is rarely a target for the chemical carcinogens such as benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene that have been used to demonstrate the anticarcinogenic effect of BHA (21,22). It has been demonstrated that GST activity can be increased in extrahepatic organs after feeding mice with BHA (13,23—25) but less attention has been paid to identifying the individual isoenzymes that are induced, or to the cellular localization of the induced enzymes within a specific organ. Glutathione Stransferases have distinct substrate specificities not only towards the model substrates used to measure activity in vitro, but also towards carcinogen metabolites such as benzof.a]pyrene-7,8diol-9,10-epoxide and aflatoxin B,-9,10-epoxide (37,26,27) that could be encountered in vivo. Certain GST (i.e. ligandin, YaYa) are also implicated in the covalent binding of carcinogens (28,29). It is thus important to identify those extrahepatic GST isoenzymes that are induced by BHA. During the present study we have examined the effect of BHA on extrahepatic concentrations of GST isoenzymes, and the histological localization of the induced enzymes in the lung, kidney and small intestine from male and female mice. 2255

L.I.McLeUan, D J.Harrison and J.D.Hayes

Table I. Effect of dietary 1BHA on extrahepatic cytosolic GST activity in mice Organ

Sex

Treatment

Specific activity (nmol/min/mg) CDNB

Lung Lung Lung Lung Kidney Kidney Kidney Kidney Sm. intestine Sm. intestine Sm. intestine Sm. intestine

female female male male female female male male female female male male

none BHA none BHA none BHA none BHA none BHA none BHA

278.95 369.49 183.73 299.14 1176.17 2420.88 1203.30 1896.12 81.68 650.55 89.17 956.26

DCNB ± 16.17 ± 13.83 20.36 16.44 ± 34.66 ± 55.25 ± 62.04 ± 66.84 ± 18.18 ± 77.97 ± 19.73 99.97

7.42 12.40 4.93 8.23 13.71 27.32 11.10 19.24 4.82 40.49 5.83 56.61

tPBO ± ± ± ± ± ± ± ± ±

0.60 1.15 1.27 0.58 0.62 0.85 0.53 1.99 1.45 ± 4.62 ± 0.86 ± 9.85

0.35 0.42 0.24 0.28 0.74 1.00 0.48 0.59 0.45 0.91 0.58 0.88

± 0.11

Tissue and cytosol preparation The organs used for subcellular fractionation were stored at —80°C until required. Lungs from 30 BHA-fed mice (15 male and 15 female) were divided into six groups of five pairs of lungs (three groups from male mice and three groups from female mice—to provide triplicate values). Kidneys and small intestines were similarly divided. The same organs from untreated mice (12 male and 12 female) were divided into six groups of organs (or organ pairs) in the same way. Cytosols were prepared from the 36 organ pools as described previously for mouse liver (17). Tissue for immunohistochemistry was fixed immediately after excision from the animal in 10% buffered formalin. After 24 h, the blocks were embedded in paraffin. Analytical All enzyme assays were carried out at 37°C as described previously (15,17). Protein determination was by the method of Bradford (30) adapted for use on a Cobas Fara centrifugal analyser (Roche Diagnostics, Welwyn Garden City, Herts, UK). Antibodies Murine hepatic GST [Alpha class ( Y a , ^ and YcYc), Mu class (Yb,Yb,) and Pi class (YfYf)] and rat Alpha class YkYk and Yc, Yc, were purified as described in earlier publications from this laboratory (12,15,18,31-33). Antibodies were raised to the purified proteins in female New Zealand White rabbits as described previously (9). Electrophoresis and immunoblotring The methods described by Hayes and Mantle (8,9) were used. Immunohistochemistry Anlisera raised to the pure murine hepatic GST were incubated separately overnight at 4°C on non-trypsinized 3 jim sections after blocking endogenous peroxidase activity with methanol/^Oj for 15 min. Detection was by a streptavidin-biotin— peroxidase method using diaminobenzidine as the peroxidase substrate as described previously (34).

Results The effect of dietary BHA on GST and glutathione peroxidase activity in murine extrahepatic organs GST activity towards a variety of substrates was measured in the lung, kidney and small intestine of male and female mice 2256

6.23 6.18 2.66 3.48 8.51 15.20 7.02 11.12 27.70 76.38 28.99 65.06

± 0.04 ± 0.05 0.06 ± 0.06 ± 0.07 ± 0.03 ± 0.06 ± 0.10 ± 0.01 ± 0.15 ± 0 14

Materials and methods Chemicals Androst-5-enc-3 ,17-dione was a gift from Dr P.K.Stockman (Royal Infirmary Edinburgh). All other chemicals were of analytical grask: and readily available commercially. Animals Male and female BALB/c mice were purchased from Bantin and Kingman, Hull, UK. They were received at 7—9 weeks old and were acclimatized for — 1 week before being fed a powdered diet containing 0.75 % (by wt) of BHA as described by Pearson et al. (14). Control animals were age matched and were fed with the same diet (Rat and Mouse no. 3 Breeding Diet; SDS, Stetfield, Whham, Essex, UK) without the addition of BHA. The mice had free access to drinking water and were given either the control or the BHA-containing diet ad libitum for 14 days before they were lolled.

Mb

(b)

* ^ _

2

3

4

± 0.94 ±1.14 ± 0.40 ± 0.24 ± 1.06 ± 2.38 ± 0.56 ± 0.92 ± 4.65 ± 9.89 ± 4.96 ± 8.31

5

8

7

s

e

H 2 O, ± 6.73 ± 7.41 ± 4.82 5.69 ± 1.83 ± 9.86 ± 5.54 ± 7.61 ± 7.99 ± 2.63 ± 4.54 ± 3.12

82.99 88.24 73.33 79.13 232.97 252.90 173.59 177.38 82.82 41.34 85.89 41.02

KDNEY

LJLJNQ

t

(a)

CHP

EA

10

INTESTINE

LNER

11

14 13

12

13

86.00 83.64 94.54 91.51 218.49 232.22 169.70 161.65 108.63 20.94 112.48 29.16

± ± ± ± ± ± ± ± ± ± ± ±

8.12 8.68 3.51 7.30 4.49 8.73 6.19 6.20 6.91 3.57 9.99 2.97

i Anti-(mouse Ya.1'a,) igQ



. _

__

_

4Ml^P

Anti-(mouse YcYc) IgG

Anii-(ral YkYk) IgG

Anti-fmouse YbYb) IgG

Anti-(mouse YfYf) IgG

Fig. 1. The effect of dietary BHA on extrahepatic GST isoenzymes in male and female mice. The GST content of lung, kidney and small intestine of male and female BALB/c mice that had been fed a standard diet or the same diet containing 0.75% BHA was determined by Western blotting. Cytosolic protein (30 jig) was subjected to SDS-PAGE as follows: lane 1, the antigen against which the relevant antibody was raised; lanes 2, 6, 10 and 14, control cytosols from lung, kidney, intestine and liver respectively of female mice; lanes 3, 7, 11 and 15, cytosols from lung, kidney, intestine and liver respectively of female mice that had been fed a BHA-containing diet; lanes 4, 8 and 12, control cytosol from lung, kidney and intestine of male mice; lanes 5, 9 and 13, cytosol from lung, kidney and intestine of male mice that had been fed a BHA-containing diet. The resolved proteins were probed for cross-reactivity with antisera raised against (a) mouse GST Y a ^ a j , (b) mouse YcYc, (c) rat YkYk, (d) mouse YbYb and (e) mouse YfYf.

that had been fed a standard diet or one containing 0.75% BHA. Treatment of mice with BHA increased GST activity towards l-chloro-2,4-dinitrobenzene (CDNB) in all organs (Table I). Using this general substrate, activity was elevated by between 1.5- and 2-fold in both lung and kidney from mice administered BHA. However, the small intestine snowed the greatest rise in activity towards CDNB, with an 8- to 10-fold increase being

Modulation of GST by BHA

Table n. Effect of dietary BHA on extrahepatic cytosolic GST isoenzymes Antibodies used for immunodetection

Sex

AntHmouse

female female male male female female male male female female male male female female male male female female male male

Anti-(mciuse YcYc)

AntHrat YkYk)

Anti-{mouse YbYb)

AntHmouse YfYO

Treatment Relative level of cross-reactivity Lung

Kidney

Intestine

none BHA none BHA none BHA none BHA none BHA none BHA none BHA none BHA none BHA none BHA

••**H

i";

••*

ND, not detected; tr., trace. (+) denotes a marginal increase.

observed in mice fed BHA-containing diets. Activity towards 1,2-dichloro-4-nitrobenzene (DCNB) was increased to a similar extent as observed with CDNB in each of these extrahepatic organs from BHA-treated mice, with activity in kidney and lung being increased by ~ 1.5- to 2-fold, and in the small intestine by 8.5- to 10-fold. Conjugation between fra/ts-4-phenylbut-3-en2-one (fPBO) and glutathione was only marginally increased by BHA treatment in lung or kidney, but was elevated between 1.5and 2-fold in the small intestine. When ethacrynic acid (EA) was employed as the substrate there was little change in activity in lung following BHA treatment but activity towards this substrate was increased by — 1.5-fold in kidney and ~2.5-fold in the small intestine. GSH peroxidase activity towards cumene hydroperoxide (CuOOH) and H2O2 in lung and kidney showed little or no change after BHA treatment (Table I). By contrast, the small intestine showed a pronounced decrease in GSH-peroxidase activity following dietary BHA. The GSH peroxidase activity towards CuOOH was reduced by —2-fold, while activity towards H2O2 was reduced by 4-fold. Sex-specific variation in GST activity Differences in GST activity between the sexes towards certain substrates were evident in the extrahepatic organs of both control and BHA-treated mice. Lung cytosol from untreated female mice exhibited at least 50% greater activity than that of untreated male mice towards the substrates CDNB, DCNB, fPBO and EA. Following dietary BHA, this sex difference was maintained. No major sex differences were observed in GSH-peroxidase activity in either control or BHA-treated mouse lung. In kidney, sex differences in GST activity were less marked than observed in the lung. In control animals, small sex differences were observed in activity towards rPBO, DCNB, EA, CuOOH and H2O2 where female animals had ~ 2 5 % higher GST activities; there was no sex difference in activity towards CDNB. After BHA treatment, minor sex differences were seen in activity towards all substrates tested, with female mouse kidney

Fig. 2. Expression of YaYa-type GST in mouse kidney. (A) Kidney from untreated female mice; (B) kidney from BHA-treated female mice. GST is primarily localized in proximal tubules. Magnification X28O.

again having higher levels of activity than renal cytosol from male mice. The small intestine showed no apparent sex differences in the uninduced levels of GST activity towards any of the substrates tested. Following BHA treatment, a small sex difference was observed in levels of activity towards CDNB and DCNB, but in this instance the small intestine of male mice exhibited a higher specific activity towards these substrates (-40%) than female mice. GSH peroxidase activity towards H2O2 was also higher (40%) in male BHA-treated mice than females. Distribution of GST isoenzymes in marine extrahepatic organs following dietary BHA The effect of dietary BHA on the expression of individual murine GST isoenzymes in each group of lungs, kidneys and small intestines (described above) was examined by Western blotting and immunohistochemistry. Antisera raised to the constitutive and BHA-inducible mouse hepatic GST Ya1/Ya2 (subunit Mr 25 600), YcYc (subunit Mr 25 800), Y^Ybi (subunit M r 26 400) and YfYf (subunit Mr 24 800), and to rat Yc,Yc, (subunit Mr 27 500) and rat YkYk (subunit Mr 25 000) were used to probe cytosols and tissue sections. Figure 1 shows the Western blots of cytosols from lung, kidney and small intestine from control and BHA-treated mice of either sex. This figure also includes the cytosolic fractions from the livers of BHA-fed and untreated female mice which served as a positive control. Each of the triplicate groups described were analysed, and as all three gave identical results, data from one set only are presented; these are summarized in Table n. The 2257

L.I.McLeUan, DJ.Harrison and J.D.Hayes

-

>si.-. • • • "•••-• •'••

* "

Fig. 4. GST YbYb localization in lung from untreated male mice. The expression of YbYb in mouse lung was restricted to bronchial epithelium. The surrounding alveoli were not stained. Magnification X520.

B '^ • —

' f

Fig. 3. Expression of Pi class GST in mouse small intestine. There was no detectable cross-reactivity in untreated female intestine (A), but the epithelium of the villi expressed YfYf strongly after BHA treatment (B). Magnification xllO.

relative staining intensities from the immunohistochemistry studies were found to correlate well with that obtained from Western blotting. Renal GST Basal expression of Ya-type subunits in untreated mouse liver is very low and is essentially undetectable using standard Western blotting techniques. However, in the kidney a protein crossreacting with anti-(mouse GST Y a ^ a J IgG was found to be constitutively expressed. This protein was present in ~ 4-fold higher concentrations in female kidney than in male mouse kidney. BHA induced the levels of expression of the Ya-type subunit in the kidneys of both males and females, and the sex difference was still apparent after BHA treatment, with female kidney maintaining a 4-fold higher level of expression than males. Like the Ya-type subunit, the Yk subunit was subject to sexspecific control in mouse kidney, where control female mice expressed higher levels of this subunit than control males. The Yk-type subunit was induced by BHA in female mouse kidney, but little or no change in levels of this subunit was found in males. This class Alpha subunit was also found to be constitutively expressed in mouse liver and was induced by BHA treatment. Subunits with immunological identity to the mouse hepatic Yc subunit, a subunit having high amino acid sequence homology ( — 95%) with the ethoxyquin-inducible rat Yc2 subunit (37), were also found to be expressed in control mouse kidney, with a slight predominance in tissue obtained from female mice, rather 2258

than male. Dietary BHA caused a small induction of Yc (~ 1.5to 2-fold). Antibodies raised to the rat Alpha class GST subunit, Yc,, were also used to probe the extrahepatic fractions. No subunit that cross-reacted with anti-(rat GST Yc,Yci) IgG was detected in any of the extrahepatic cytosols from male or female untreated or BHA-treated mice (data not shown). Subunits that cross-reacted with antiserum raised to mouse Y^Yfy (class Mu) were detected in control mouse kidney. Sex differences were not obvious in untreated mouse kidney. Dietary BHA induced the renal Yb subunit by 1.5- to 2-fold, and kidney from BHA-treated female mice contained slightly higher levels than BHA-treated males. A Pi class GST subunit related or identical to mouse Yf was present at low levels in untreated mouse kidney, and BHA caused only minor induction of this subunit. For all the GST subunits examined, localization in the kidney was primarily in the proximal tubules. This specific concentration of GST subunits was maintained after BHA treatment (Figure 2). Intestinal GST The Ya-type subunits were not detectable in the small intestine of untreated mice. However, an immunoreactive subunit with an ~ Mr value of 24 500 was detected in the samples of small intestinal cytosol from both male and female mice treated with BHA. As the Ya-type subunits in murine liver have an Mr of 25 600 it is possible that the intestinal Ya subunit is genetically distinct from the hepatic enzymes but, more likely, that its lower Mr value is a consequence of partial proteolytic digestion that occurs in vivo or in vitro. It is interesting to note that if the faster electrophoretic mobility of the intestinal Ya subunit is due to proteolysis, then it is relatively specific to this subunit; intestinal GST Yf shows little evidence of altered mobility during SDS-PAGE (Figure 1). The Yc- and Yk-type subunits were barely detectable in the small intestine of control mice but dietary BHA caused induction of Yc to detectable levels in both males and females. The Yk-related subunit was induced to a greater extent, with a pronounced sex-difference becoming apparent following BHA treatment. The small intestinal cytosol of BHA-treated males contained — 3-fold higher levels of Yk than BHA-treated females. The class Mu subunit was barely detectable in control mouse small intestine but was induced considerably by BHA treatment, to levels similar to that found in control mouse liver. Although the sex difference is not striking, BHA-treated males were found

Modulation of GST by BHA

to contain slightly higher levels of the Mu class subunit than females. Pi class GST was not detected in small intestine from untreated female mice; trace amounts were observed in control male mouse small intestine. BHA had a dramatic effect on levels of YfYf in the small intestine (Figure 3), with the intestinal content rising to greater than half that of BHA-treated female mouse liver. The Pi class GST in BHA-treated mouse intestine was localized in the epithelium of the villi (Figure 3). This was also true for the other GST that were expressed in the small intestine. Sex differences in the expression of intestinal Yf were not apparent. Pulmonary GST In the lung, all of the GST were localized exclusively in the bronchi (Figure 4) both in control and BHA-treated mice. No alveolar staining was detected. Of the Alpha class subunits, the Ya-type subunit was not detectable in lung even after BHA treatment, but the constitutive expression of both Yc and Yk was observed. Lungs from BHA-treated mice contained — 1.5-fold higher levels of Yc than untreated mice, and females expressed higher levels of Yc and YK than males. BHA-mediated induction of YK was marginal in both sexes. The Mu class GST was expressed constitutively in lungs from male and female mice. Dietary BHA caused induction of the Yb subunit to a modest degree (a 1.5- to 2-fold increase). Sex differences in levels of Yb were not obvious in control mouse lungs, but lungs from BHA-treated female mice contained approximately twice that of BHA-treated male mice. Very low levels of expression of Pi class GST were detected in mouse lungs. Female mice were found to express slightly more of the Yf-type subunit than males and BHA-mediated induction of this subunit was ~ 1.5-fold in both sexes. Discussion Since induction of GST is thought to be critical in the inhibition of chemically mediated carcinogenesis in extrahepatic organs of BHA-treated mice, the effect of dietary BHA on levels of specific GST isoenzymes in mouse lung, kidney and small intestine was examined. Of the extrahepatic organs examined, the small intestine appeared to be the most responsive in terms of induction of GST activity by BHA, a finding that is in agreement with the results of Benson et al. (23) and DeLong et al. (25) who measured changes in GST activity towards CDNB and DCNB in murine extrahepatic organs in response to dietary BHA. GST activity towards the substrates CDNB, DCNB, fPBO and EA was found to be increased by the greatest amount in small intestine after BHA treatment, compared to lung and kidney. In the mouse, DCNB, fPBO and EA are specific for the Yb,, Yb^ and Yf subunits respectively (12,18,33). It is therefore likely that the BHA-mediated induction of Yb|, Yl>2 and Yf is responsible for a large proportion of the increase in specific activity towards these substrates. Western blotting experiments showed that Ya, Yb, Yf and Yk were induced by BHA treatment in the small intestine. The extent of induction of these subunits was found to be considerable ( - 1 0 - to 30-fold). Our data also show that the Yc subunit is the least responsive to BHA treatment, only experiencing a maximum induction of — 2-fold in any of the organs examined. The effect of dietary BHA on GST mRNA levels has been investigated in other laboratories (19,35). These investigators have shown that mRNAs encoding Ya2, Yc, Yb, and Yf are all present in untreated mouse small intestine. The amount of mRNA

was found to be readily increased following BHA treatment. Intestinal levels of Ya2 and Yb! mRNA increased 100- and 15-fold respectively, whereas Yc and Yf mRNAs were only increased by 6- and 2- to 5-fold respectively. The relative levels of induction of the mRNAs encoding Ya2, Yb] and Yf do not, however, correlate accurately with the levels of induction of GST subunits determined during the present study. Unlike the BHAmediated induction of the mRNAs, the extent of induction of Yaand Yb-type subunits appears to be of a similar order of magnitude, albeit that uninduced levels of Ya were undetectable by Western blotting in the small intestine, making an estimation of the extent of induction difficult. Furthermore, it is possible that the Ya-type subunits are readily degraded, as the subunit Mr of the cross-reactive band is lower than the Ya!/Ya2 hepatic standard by ~ 1 kDa. The lack of correlation between mRNA levels and protein in this instance is thus perhaps not unexpected. On the other hand, the Pi class subunit appears to be induced by at least 10-fold in the intestine of BHA-treated mice, and the extent of induction is much greater than that seen in either kidney or lung, and possibly greater than that seen in liver. The work of Pearson and his colleagues (35) has shown that Pi class mRNA is increased by 2- to 5-fold in lung as well as the small intestine. During the present study, we found only minor induction of Pi class GST in lung (1.5- to 2-fold) compared to that seen in the small intestine. This anomaly implies that factors other than transcriptional activation are participating in the BHA-mediated induction of Pi class GST in the small intestine. Selenium-dependent glutathione peroxidase (GPX 1) activity in the small intestine was particularly responsive to BHA treatment. However, unlike the GST activity, GPX 1 activity was depressed by ~4-fold. The small intestine was the only organ examined in the present study where there was a significant decrease in GPX 1 activity, but it has been previously shown that dietary BHA causes a modest decrease in GPX 1 activity in mouse liver (36). The chemoprotector ethoxyquin has also been shown to reduce GPX 1 activity in rat liver (37). It remains to be determined if BHA (or its metabolites) is affecting GPX 1 at a transcriptional or post-translational level. Glutathione S-transferase activity towards CDNB and DCNB was also increased by BHA treatment in lung although to a lesser degree than that seen in the small intestine. Again, these results agree with the Western blotting and immunohistochemical data, where a Mu class enzyme (subunit Mr 26 400) was induced by BHA. Activity towards the model substrate for the Ybj subunit, fPBO, was not, however, much changed by BHA treatment, nor was the specific activity towards EA; the Western blots demonstrated only a slight increase in Yf levels. Both the Yb2 and Yf subunits have been implicated in the detoxication of carcinogenic metabolites of benzo[a]pyrene (26,38). If induction of GST is responsible for the anticarcinogenic effect of BHA in the lung of benzo[a]pyrene-treated mice, it is possible that it is the considerable hepatic induction of Yf and Yr>2, rather than pulmonary induction of these subunits, that is responsible for the major portion of chemoprotection. It should be noted, however, that Dock et al. (39) have suggested that BHA-induced GST activity may be of limited importance for protection against the most mutagenic metabolite of benzo[a]pyrene, (±)-7/3,8adihydroxy-9a, 10a-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE). These workers found that though the hepatic GST activity towards BPDE was increased after treating mice with BHA, the Km for this substrate was also increased. Small increases of Yk and Yc were also detected in mouse lung after BHA treatment, and, like the Mu class enzyme, were 2259

L.I.McLeUan, DJ.Harrison and J.D.Hayes

found to be localized exclusively in the bronchi. No GST were detected in the alveoli of either control or BHA-treated animals, suggesting that the bronchi and alveoli respond differently to chemical injury. It is noteworthy that the Ya-type subunits were not detectable in mouse lung and there was no evident response to dietary BHA. This result is compatible with that of Buetler and Eaton (19) who report undetectable levels of Ya mRNA in mouse lung and no induction by BHA. Control of expression of mouse and rat Ya subunits has been extensively studied and it has been shown that the BHA-mediated induction is under the control of a 5' flanking element called the antioxidant (or electrophile) response element (ARE or EpRE) (40,41). It has recently been reported that in the mouse the 'EpRE' comprise two adjacent AP-1-like binding sites and that transactivation occurs upon Fos/Jun binding (41). The present study demonstrates a detectable induction of the Yb, Yc and Yk subunits in mouse lungs, showing that a signal for induction of GST is reaching the bronchial epithelial cells (whether by inhalation of food dust containing BHA or via the circulation). The complete absence of induction (or any basal expression) of the Ya subunits in the lung is thus intriguing. The GST distribution in mouse kidney proved to be particularly interesting, not only in that dietary BHA caused notable induction but also in the sex difference in expression of these enzymes. Previous workers have shown that the Pi class GST Yf subunit is expressed at ~ 15-fold greater levels in male mouse liver than in female mouse liver (11,12). However, very little is known about the sex-specific regulation of other GST families. During the present study, Western blotting showed that female mice express — 4 times more of a Ya-type subunit in kidney than male mice. This sex-difference was also evident in the immunohistochemical studies, where, in contrast to male mice, female mice were found to express Ya in the proximal tubules. Furthermore, the sex difference in expression of Ya was maintained after BHA treatment. Sex-specific expression of Alpha-class GST is not restricted to the Ya-type subunit as a similar sex difference in mouse kidney was observed in the expression of a polypeptide (Mr 25 000) which cross-reacted with anti-(rat GST YkYk) IgG. In this instance, female mouse kidney contained ~ 3 times more of the Yk subunit than males. It is possible that these sex differences are relevant to sexually differentiated nephrotoxicity and carcinogenicity of certain compounds in mice, especially as the GSTs are localized primarily in the proximal tubules. As the proximal tubules are responsible for trie majority of the reabsorption of the glomerular filtrate and also for the active secretion of drugs and their metabolites, this localization of the GST is likely to be important in the renal handling of certain xenobiotics. Differences in the sensitivity of male and female mouse kidney to the toxic effects of chloroform have been mainly attributed to sex-differences in expression of the cytochrome P450 enzymes thought to be responsible for the metabolism of chloroform to its toxic metabolite, phosgene (42—44). Male mouse microsomes metabolize chloroform to phosgene more rapidly than female (43) and Western and Northern blotting studies have demonstrated that mouse renal cytochrome P450 isoenzymes are sexually differentiated (44). It has been shown, however, that phosgene reacts with glutathione (45). The involvement of GST in this conjugation reaction was not investigated and it is, as yet, unknown which, if any, of the individual GST isoenzymes is responsible for this catalysis. Henderson et al. (44) investigated sex-differences in renal GST content using antibodies raised to the constitutive hepatic murine GSTs Yf, Yb and Yc and found 2260

no obvious differences. The present study demonstrates considerably higher levels of Ya and Yk in female mouse kidney than in male and it is feasible that this sex-difference in expression contributes to the relative sensitivity of male mouse kidney to the toxic effects of chloroform, as well as to the relative susceptibility of male mouse kidney to certain renal carcinogens. Acknowledgements We thank Linda May for expert technical assistance. The Medical Research Council is gratefully acknowledged for financial support (grant G8622978CA).

References 1. Wallenberg,L.W. (1978) Inhibitors of chemical carcinogenesis. Adv. Cancer Res., 26, 197-226. 2. Wattenberg,L.W. (1986) Protective effects of 2(3H«t-°utyl^-hydroxyanisole on chemical carcinogenesis. Food Chem. Toxicol., 24, 1099-1102. 3. Wattenberg.L.W. H985) Chemoprevention of cancer. Cancer Res., 45, 1 - 8 . 4. Mannervik.B., Alin.P., Guthenberg.C, Jensson.H., Tahir.M.K., Warholm.M. and J6rnvall,H. (1985) Identification of three classes of cytosolic glutathione transferase common to several mammalian species: correlation between structural data and enzymic properties. Proc. Natl. Acad. Sd. USA, 82, 7202-7206. 5. Morgenstem.R. and DePierreJ.W. (1983) Microsomal glutathione transferase. Purification in unactivated form and further characterization of the activation process, substrate specificity and amino acid composition. Eur. J. Biochem., 134, 591-597. 6. Hiratsuka.A., Sebata.N., Kawashima.K., Okuda.H., Ogura.K., Watabe.T., Satoh.K., Hatayama.I., Tsuchida.S., Ishikawa.T. and Sato,K. (1990) A new class of rat glutathione S-transferase Yrs-Yrs inactivating reactive sulfate esters as metabolites of carcinogenic arylmethanols. J. Biol. Chem., 265, 11973-11981. 7.Meyer,D.J., Coles,B., Pemble.S.E., Gilmore.K.S., Fraser,G.M. and Ketterer.B. (1991) Theta, a new class of glutathione transferases purified from rat and man. Biochem. J., 274, 409-414. 8. Hayes.J.D. and Mantle.T.J. (1986) Anomalous electrophoretic behaviour of the glutathione S-transferase Ya and Yk subunits isolated from man and rodents. A potential pitfall for nomenclature. Biochem. J., 237, 731-740. 9. HaycsJ.D. and Mantle,T.J. (1986) Use of immuno-blot techniques to discriminate between the glutathione S-transferase Yf, Yk, Ya, Yn/Yb and Yc subunits and to study their distribution in extrahepatic tissues. Evidence for three immunochemically distinct groups of transferase in the rat. Biochem. J., 233, 779-788. 10. Kapkwitz,N., KuhlenkampJ. and Clifton.G. (1975) Drug induction of hepatic glutathione S-transferase in male and female rats. Biochem. J., 146, 351—356. 11. HatayamaJ., Satoh.K. and Sato.K. (1986) Developmental and hormonal regulation of the major form of hepatic glutathione S-transferase in male mice. Biochem. Biophys. Res. Common.. 140, 581-588. 12. McLellan.L.I. and HayesJ.D. (1987) Sex-specific constitutive expression of the pre-neoplastic marker glutathione S-transferase, YfYf, in mouse liver. Biochem. J., 245, 399-406. 13. Spamins,V.L., Venegas.P.L. and Wattenberg.L.W. (1982) Glutathione Stransferase activity: enhancement by compounds inhibiting chemical carcinogenesis and by dietary constituents. J. Nad. Cancer btst., 68,493-496. 14. Pearson.W.R., WindteJJ., MorrowJ.F., Benson.A.M. and Talalay.P. (1983) Increased synthesis of glutathione S-transferases in response to anticarrinogeruc antioxidants. Cloning and measurement of messenger RNA. J. Biol. Chem., 258, 2052-2062. 15. McLellan.L.I. and Hayes.J.D. (1989) Differential induction of class alpha glutathione S-transferases in mouse liver by the anticarcinogenk antioxidant butylated hydroxyanisole: purification and characterization of glutathione Stransferase Ya,Ya,. Biochem. J., 263, 393-304. 16. Benson,A.M., Hunketer.M.J. and YorkJ.L. (1989) Mouse hepatic glutathione transferase isoenzymes and their differential induction by anticarcinogens. Specificities of butylated hydroxyanisole and bisethylxanthogen as inducers of glutathione transferases in male and female CD-I mice. Biochem. J., 261, 1023-1029. 17. McLellan.L.I., Ken\L.A., Cronshaw.A.D. and HayesJ.D. (1991) Regulation of mouse glutathione S-transferases by chemoprotectors: molecular evidence for the existence of three distinct Alpha-class glutathione S-transferase subunits, Ya,, Yaj and Ya3, in mouse liver. Biochem. J., 276, 461-469. 18. HayesJ.D., Kerr.L.A., Pcacock.S.D., Cronshaw.A.D. and McLellan.L.I. (1991) Hepatic glutathione S-transferases in mice fed on a diet containing the anticarcinogenic antioxidant butylated hydroxyanisole. Biochem. J., 277, 501-512.

Modulation of GST by BHA 19. Buetler.T.M. and Eaton,D.J. (1992) Complementary DNA cloning, messenger RNA expression, and induction of a class glutathionc S-transferases in mouse tissues. Cancer Res., 52, 314-318. 20. HayesJ.D., Judah.D.J., Neal.G.E. and Nguyen.T. (1992) Molecular cloning and heterologous expression of a cDNA encoding a mouse GST Yc subunit possessing high catalytic activity for aflatoxin Bl-8,9-epoxide. Biochem. J., 285, 173-180. 21. Wattenberg.L.W. (1973) Inhibition of chemical carcinogen-induced pulmonary neoplasia by butylated hydroxyanisole. J. Nail. Cancer but. ,50, 1541 —1544. 22. Wattenberg,L.W. (1972) Inhibition of carcinogenic and toxic effects of polycyclic hydrocarbons by phenolic antioxidants and ethoxyquin. /. Natl. Cancer Inst., 48, 1425-1430. 23.Benson,A.M., Cha.Y.-N., Bueding.E., Heine.H.S. and Talalay.P. (1979) Elevation of extrahepatk glutathione S-transferase and epoxide hydratase activities by 2(3)-(e"-t>utyl-4-hydroxyanisole. Cancer Res., 39, 2971-2977. 24. Spamins,V.L. and Wattenberg.L.W. (1981) Enhancement of glutathione Stransferase activity of the mouse forestomach by inhibitors of benzo(a)pyreneinduced neoplasia of the forestomach. J. Natl. Cancer Inst., 66, 769-771. 25. De Long,M.J., ProchaskaJl.J. and Talalay.P. (1985) Tissue-specific induction patterns of cancer-protective enzymes in mice by tert-butyl-4-hydroxyanisole and related substituted phenols. Cancer Res., 45, 546-551. 26. Jemstr8m,B., Martinez.M., Meyer.D.J. and Ketterer.B. (1985) Glutathione conjugation of the carcinogenic and mutagenic electrophile (±)-7/S,8a,dihydroxy-9a,10or-oxy-7,8,9,10-tetrahydrobenzo(a)pyrene catalysed by purified rat liver glutathione transferases. Carcinogenesis, 6, 85 — 89. 27.Coles,B., Meyer.D.J., Ketterer.B., Stanton,C.A. and Garner.R.C. (1985) Studies on the detoxication of microsomally-activated aflatoxin Bl by glutathione and glutathione transferases in vitro. Carcinogenesis, 6, 693-697. 28. Ketterer.B., Ross-Mansell,P. and WhiteheadJ.K. (1967) The isolation of carcinogen-binding protein from livers of rats given 4-dimethylaminoazobenzene. Biochem. J., 103, 316-324. 29. Singer.S. and Litwack,G. (1971) Identity of corticosteroid binder 1 with the macromolecule binding 3-methylcholanthrene in liver cytosol in vivo. Cancer Res., 31, 1364-1368. 30. Bradford,M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein -dye binding. Anal. Biochem., 72, 248-254. 31. Hayes,J.D. (1986) Purification and physical characterization of glutathione S-transferase K. Differential use of S-hexylglutathione and glutathione affinity matrices to isolate a novel glutathione S-transferase from rat liver. Biochem. J., 233, 789-798. 32. HayesJ.D. and Clarkson,G.H.D. (1982) Purification and characterization of three forms of glutathione S-transferase A. A comparative study of the major YaYa-, YbYb- and YcYc-containing glutathione S-transferases. Biochem. J., 207, 459-470. 33.Hayes,J.D., Coulthwaite.R.E., Stockman.P.K., Hussey.A.J., Mantle.T.J. and Wolfe.C.R. (1987) Glutathione S-transferase subunits in the mouse and their catalytic activities towards reactive electrophiles. Arch. ToxicoL, SuppL, 10, 136-146. 34. Harrison.DJ., May.L., HayesJ.D. and Neal.G.E. (1990) Glutathione Stransferase localization in aflatoxin Bl-treated rat livers. Carcinogenesis, 11, 927-932. 35. Pearson.W.R., ReinhartJ., Sisk.S.C, Anderson.K.S. and Adler.P.N. (1988) Tissue-specific induction of murine glutathione transferase mRNAs by butylated hydroxyanisole. J. Biol. Chem., 263, 13324-13332. 36. Di Simplicio.P., Jensson.H. and Mannervik,B. (1989) Effects of inducers of drug metabolism on basic hepatic forms of mouse glutathione transferase. Biochem. J., 263, 679-685. 37.Hayes,J.D., Judah.D.J., McLellan.L.I., Kerr.L.A., Peacock,S.D. and Neal.G.E. (1991) Ethoxyquin-induced resistance to aflatoxin Bl in the rat is associated with the expression of a novel Alpha-class ghflamione S-transferase subunit, Yc2, which possesses high catalytic activity for aflatoxin Bl-8,9-epoxide. Biochem. J., 119, 385-398. 38. Robertson,I.G.C., Jensson.H., Manncrvik.B. and JemsuOm.B. (1986) Glutathione transferases in rat lung: the presence of transferase 7-7, highly efficient in the conjugation of glutathione with the carcinogenic (±)7$,8adihydroxy-9a, 10a-oxy-7,8,9,10-tetrahydrobenzo{a]pyrene. Carcinogenesis, 7, 295-299. 39. Dock.L., Martinez.M. and Jemstrom.B. (1984) Induction of hepatic glutathione S-transferase activity by butylated hydroxyanisole and conjugation of benzo{a]pyrene diol-epoxide. Carcinogenesis, 5, 841—844. 40. Rushmore.T.H., Morton,M.R. and Pkkett.C.B. (1991) The antioxidant response element Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J. Biol. Chem,, 266, 11632-11639. 41.Friling,R.S., Bergelson.S. and Daniel.V. (1992) Two adjacent AP-1-like

binding sites form the electrophile-responsive element of the murine glutamionc S-transferase Ya subunit gene. Proc. NalL Acad. Sri. USA, 89, 668-672. 42. Pohl.L.R. and Krishna.G. (1978) Deuterium isotope effect in bioactivation and hepatotoxicity of chloroform. Life Sri., 23, 1067-1072. 43. Pohl.L.R., GeorgeJ.W. and Satoh.H. (1984) Strain and sex differences in chloroform-induced nephrotoxicity. Different rates of metabolism of chloroform to phosgene by the mouse kidney. Drug. Metab. Dispos., 12, 304-308. 44. Hcnderson,C.J., Scou,A.R., Yang.C.S. and Wolf.C.R. (1990) Testosteronemediated regulation of mouse renal cytochrome P-450 isoenzymes. Biochem. J., 266, 675-681. 45. Pohl.L.R., Branchflower.R.V., Highet.R.J., Martin.J.L., Nunn,D.S., Monks.T.J., GeorgeJ.W. and HinsonJ.A. (1981) The formation of diglutathionyl dithiocarbonate as a metabolite of chloroform, bromotrichloromethane, and carbon tetrachloridc. Drug Metab. Dispos., 9, 334—339. Received on June 30, 1992; revised on August 24, 1992; accepted on August 26, 1992

2261