A New Class of Rat Glutathione s-Transferase Yrs-Yrs Inactivating Reactive Sulfate Esters as Metabolites of Carcinogenic Arylmethanols*

A glutathione (GSH) ,S-transferase (GST), catalyzing the inactivation of reactive sulfate esters as metabolites of carcinogenic arylmethanols, was isolated from the male Sprague-Dawley rat liver cytosol and purified to homogeneity in 12% yield with a purification factor of 901-fold. The purified GST was a home-dimeric enzyme protein with subunit Mr 26,000 and ~17.9 and designated as Yrs-Yrs because of its enzyme activity toward “reactive sulfate esters.” GST Yrs-Yrs could neither be retained on the ELhexylglutathione gel column nor showed any activity toward 1,2-dichloro-4nitrobenzene, 4-nitrobenzyl chloride, and 1,2-epoxy3-(4’-nitrophenoxy)propane. l-Chloro-2,4-dinitrobenzene was a very poor substrate for this GST. lMenaphthyl sulfate was the best substrate for GST Yrs-Yrs among the examined mutagenic arylmethyl sulfates. The enzyme had higher activities toward ethacrynic acid and cumene hydroperoxide. N-terminal amino acid sequence of subunit Yrs, analyzed up to the 25th amino acid, had no homology with any of the known class alpha, mu, and pi enzymes of the SpragueDawley rat. Anti-Yrs-IgG raised against GST Yrs-Yrs showed no cross-reactivity with any of subunits Ya, Yc, Ybl, Yb2, and Yp. Anti-IgGs raised against Ya, Yc, Ybl, Yb2, and Yp also showed no cross-reactivity with GST Yrs-Yrs. The purified enzyme proved to differ evidently from the 12 known cytosolic GSTs in various tissues of the rat in all respects. Immunoblot analysis of various tissue cytosols of the male rat indicated that apparent concentrations of the GST YrsYrs protein were in order of liver > testis > adrenal > kidney > lung > brain > skeletal muscle = heart = small intestine = spleen = skin = 0.

A glutathione (GSH) ,S-transferase (GST), catalyzing the inactivation of reactive sulfate esters as metabolites of carcinogenic arylmethanols, was isolated from the male Sprague-Dawley rat liver cytosol and purified to homogeneity in 12% yield with a purification factor of 901-fold.
l-Chloro-2,4-dinitrobenzene was a very poor substrate for this GST. l-Menaphthyl sulfate was the best substrate for GST Yrs-Yrs among the examined mutagenic arylmethyl sulfates.
The enzyme had higher activities toward ethacrynic acid and cumene hydroperoxide. N-terminal amino acid sequence of subunit Yrs, analyzed up to the 25th amino acid, had no homology with any of the known class alpha, mu, and pi enzymes of the Sprague-Dawley rat. Anti-Yrs-IgG raised against GST Yrs-Yrs showed no cross-reactivity with any of subunits Ya, Yc, Ybl, Yb2, and Yp. Anti-IgGs raised against Ya, Yc, Ybl, Yb2, and Yp also showed no cross-reactivity with GST Yrs-Yrs. The purified enzyme proved to differ evidently from the 12 known cytosolic GSTs in various tissues of the rat in all respects. Immunoblot analysis of various tissue cytosols of the male rat indicated that apparent concentrations of the GST Yrs-Yrs protein were in order of liver > testis > adrenal > kidney > lung > brain > skeletal muscle = heart = small intestine = spleen = skin = 0.
In spite of their very weak carcinogenicity, benz [u]anthracene (BA)' and chrysene (CR) turn into extremely potent * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
In the adult rat liver cytosol, the metabolically formed sulfate esters of 5-HCR, 7-HMBA, and DHBA are enzymatically scavenged so readily by glutathione (GSH) (17,22,23) that they cannot exert mutagenicity (17,22,23), nor bind covalently to calf thymus DNA (16,23). From the hepatic cytosolic incubation mixtures fortified with 3'-phosphoadenosine 5'-phosphosulfate and GSH, stable and nonmutagenic S-(aryl)methylglutathiones, Ar-CH*-SG, have been isolated and identified with authentic specimens (17,22,23). These facts, therefore, indicate that GST does play an important role in preventing rat liver, having the highest level of hydroxysteroid sulfotransferase activity among various tissues (24), from tumorigenesis caused by the carcinogens, methyland hydroxymethylarenes. The present paper deals with (i) 11973 This is an Open Access article under the CC BY license.
the existence of at least three GSTs in the rat liver cytosol, which catalyzed GSH conjugation of reactive sulfate esters of carcinogenic arylmethanols and were all unretainable on an S-hexyl-SG affinity column, (ii) the isolation and purification of the major one of these GSTs, which was designated as Yrs-Yrs (a homo-dimeric Y protein that catalyzes GSH conjugation of the "reactive sulfate esters") and had no homology in N-terminal amino acid sequence to any of the known classes of rat GST isozymes, (iii) the unique properties of GST Yrs-Yrs in substrate specificity and immunochemical reactivity, and (iv) the tissue distribution of GST Yrs-Yrs in the rat, estimated by immunoblot analysis. Sulfates-A preliminary study carried out by using rat liver cytosol prior to establishing a purification scheme for GST, tentatively designated as RS, catalyzing the GSH conjugation of 5HCR sulfate, the most stable compound among highly mutagenic arylmethyl sulfates, provided two important facts; one was that, unlike rat liver soluble GSTs with subunit proteins Ya, Ybl, YbZ, Ye, and Yp, GST RS had little affinity for and, consequently, passed through an S-hexyl-SG-labeled Sepharose 6B column, and the other that the enzyme had strong affinity for the sulfonated azo dye-labeled gel (blue Sepharose) column. On direct application of the rat liver cytosol to the S-hexyl-SGgel column, approximately 96% of the hepatic cytosolic activity for GSH conjugation of 5HCR sulfate passed through the column, whereas 86% of the cytosolic activity for GSH conjugation of I-chloro-2,4-dinitrobenzene (CDNB) was retained on this column. Thus, the S-hexyl-SG-labeled affinity column could be used for elimination of the known GST subunit proteins interacting with this column from the rat liver cytosol, and the sulfonated azo dye-labeled column for adsorption of the GST active toward arylmethyl sulfates during the course of purification.
For isolation of GST RS, the rat liver cytosol was applied to an anion-exchange gel (DE-52) column. Most of the cytosolic RS, surveyed by the assay of the 5HCR sulfate-GSHconjugating activity, was retained at pH 8.25 on this column and then eluted with NaCl ( Fig. 1 in the Miniprint). By this column chromatographic procedure, more than 90% of non-RS proteins were removed. From the RS fraction (Fraction 120-140) of the anion-exchange chromatogram, most of non-RS GSTs active toward CDNB were removed by adsorption on the S-hexyl-SG-gel column ( Fig. 2 in the Miniprint). The GST RS protein in the flow-through fraction of the S-hexyl-SG-gel column chromatogram was applied to a chromatofocusing gel column to remove non-RS GSTs active toward CDNB completely (Fig. 3 in the Miniprint). The major GST RS peak, corresponding to the chromatofocusing fractions (Fraction 54-63) eluted at pH 8.2-8.0, contained no detectable activity toward CDNB (Fig. 3 in the Miniprint). The major GST RS protein isolated by chromatofocusing was collected on and eluted from the blue Sepharose column (Fig. 4 in the Miniprint) and then subjected to gel filtration hplc carried out on a TSK gel G3000 SW column for further purification to homogeneity (Fig. 5 in the Miniprint).
The purified RS preparation, obtained in 12% yield with a purification fold of 901 (Table I) 6 in the Miniprint), estimated from the result of a comparative electrophoretic study carried out on the same gel plate by using not only various Mr marker proteins but the authentic GST subunit proteins, Ya (Mr 25,000), YbI and Yb2 (Mr 26,500), and Yc (28,000), isolated and purified from the rat liver cytosol. Therefore, based on these molecular data, the homo-dimeric enzyme GST RS was re-designated as GST Yrs-Yrs. GST Yrs-Yrs had a p1 value of 7.9, determined by the isoelectricfocusing method. The N-terminal amino acid sequence of this enzyme, determined from the N-terminal to 25th amino acids with an automatic amino acid sequencer based on the Edrnan degradation method, was as follows: Gly-Leu-Glu-Leu-Tyr-Leu-Asp-Leu-Leu-Ser-Gln-Pro-Ser-Arg-Ala-Val-Tyr-Ile-Phe-Ala-Lys-Lys-Asn-Gly-Ile-, A co-hplc study, carried out on an octadecylsilica column developed with aqueous acetonitrile containing 0.1% (v/v) trifluoroacetic acid in gradient manner, indicated that the subunit protein Yrs was eluted as a single peak at a completely different retention time from those of the rat liver cytosolic GST subunits which were collected on and eluted from the Shexyl-SG affinity column into a single fraction (Fig. 7). Under the hplc conditions, all the dimeric enzymes examined were dissociated into subunits. The subunit Yrs was less polar than GST subunits Ybl, Yb2, and Yc and more polar than Ya, the last one of which was eluted as a doublet peak as had been demonstrated by Ostland Farrants et al. (39). These subunit peaks were all identified with the corresponding pure homodimeric GSTs isolated from the rat liver cytosol.
Amino acid composition of an acid hydrolyzate of GST Yrs-Yrs indicated that the enzyme contained Glx (Glu and Gln) in the highest molar ratio and, next to this, Leu, Ala, Gly, and Asx (Asp and Asn) in decreasing order (Table II in the  Miniprint).
Substrate Specificities of GST Yrs-Yrs-GST Yrs-Yrs catalyzed GSH conjugation of three examined mutagenic, reactive sulfate esters of arylmethanols other than 5-HCR sulfate (Table III). Of the four sulfate esters, 1-menaphthyl sulfate was the best substrate for this GST. A kinetic study, carried out by the double reciprocal plot method based on apparent rates for the enzymatic GSH conjugate formation from 5HCR sulfate (lo-50 PM) at pH 7.4, indicated that purified GST Yrs-Yrs had a K,,, value of 17.7 PM and a !Z~~~ value of 0.123 s-l. A Km value for GSH was 1.34 mM in the 5-HCR sulfate conjugation reaction catalyzed by this enzyme at pH 7.4.
A comparative study carried out by using purified specimens of six major rat liver cytosolic GSTs with high affinity for the S-hexyl-SG-gel column and high activities toward CDNB indicated that only two GSTs bearing subunit protein Yc could catalyze the GSH conjugation reaction of 5-HCR sulfate  A, the center well contained anti-YrsIIgG raised against GST Yrs-Yrs (Abl). WeUs 1 (GST Yrs-Yrs), 2 (GST Ya-Yc), 3 (GST Ybl-Yb2), and 4 (GST Yu-YD) contained 6 ug each of respective enzyme proteins puritied from rat liver cytosol. g, well 1 contained 6 pg of GST Yrs-Yrs. Wells Abl (against GST Yrs-Yrs), Ab2 (against GST Ya-Yc), Ab3 (against GST Ybl-Yb2), and Ab4 (against GST Yp-Yp) contained sufficient amounts of IgG fractions of rabbit antisera. C, weU 5 contained a mixture of 6 pg each of GSTs Ya-Yc, Ybl-Yb2, and Yp-Yp. The welLs Abl-Ab4 contained the same amounts of the anti-IgG fractions as used for B.
with concomitant formation of GSSG from GSH.
Immunochemical Properties of GST Yrs-Yrs-An anti-IgG preparation from rabbit antisera raised against purified GST Yrs-Yrs caused significant precipitation against the antigen, but showed no cross-reactivity with GSTs Ya-Yc, Ybl-Yb2, and Yp-Yp, isolated as homogenous proteins from the rat liver cytosols, when examined by the Ouchterlony doubleimmunodiffusion test (Fig. 8A). Moreover, anti-IgG preparations from rabbit antisera raised against purified GSTs Ya-Yc, Ybl-Yb2, and Yp-Yp also showed no cross-reactivity with the enzyme GST Yrs-Yrs, whereas under the same conditions, the anti-GST Yrs-Yrs-IgG formed a significant precipitate against the antigen (Fig. 8B). These antibodies as well as that against GST Yrs-Yrs had all sufficient immunoreactivities with the corresponding purified GSTs as antigens (Fig. 8C).
Tissue Distribution of GST Yrs-Yrs in the Rut-Immunoblot analysis of various tissue cytosols of the male rats, carried out by using SDS-polyacrylamide gel electrophoresis and the anti-GST Yrs-Yrs-IgG preparation, suggested that testis contained a high concentration of GST Yrs-Yrs, comparable with that in the liver and also that the enzyme existed at somewhat lower concentrations in adrenal and kidney and at a much lower concentration in lung (Fig. 9). GST Yrs-Yrs, however, existed at extremely low concentrations in all the cytosols of the skin, heart, small intestine, and spleen, so that it could not be detected unless much larger amounts of these cytosolic proteins (at least more than five times of the liver protein) were applied. of rats. Protein samples were resolved by SDSpolyacrylamide gel electrophoresis (15% gel) and transferred electrophoretically to a nitrocellulose membrane. The membrane was sequentially incubated with 3% bovine serum albumin, IgG of rabbit antiserum (at dilution of l/1000) raised against the purified enzyme, goat anti-rabbit IgG, rabbit peroxidase anti-peroxidase, and finally with 50 mM Tris-I-El buffer, pH 7.5, containing hydrogen peroxide and 3.3'-diaminobenzidine as reoorted nreviouslv (38). Lanes 1,8,9,and 15, GST Yrs-Yrs (0.05 pg each); lanes 2-5 and 7~ cytosolic proteins (20 fig each) of adrenal, brain, heart, kidney, and lung, respectively; lanes 6 and 14, cytosolic proteins (5 pg each) of liver and testis; lunes 10-14, cytosolic proteins (20 pg each) of skeletal muscle, skin, small intestine, and spleen, respectively. Arrows indicate A4r markers used egg albumin (45,000), rabbit muscle glyceraldehyde-3-phosphate dehydrogenase (36,000), bovine erythrocyte carbonic anhydrase (29,000), bovine trypsinogen (24,000), and soybean trypsin inhibitor (20,100).

DISCUSSION
So far as concerned with rat liver cytosolic GSTs, little is known of such an isozyme, except GST E (5-5), that is neither retained on the S-hexyl-SG affinity column (40), nor detected by the most widely used standard substrate, CDNB,or/and DCNB (30,41). Extensive studies on the isozymes of GST in the rat liver have been made by using the affinity column and these standard substrates (41), and six major dimeric isozymes, GSTs Ya-Ya, Ya-Yc, Yc-Yc, Ybl-Ybl, Ybl-Yb2, and Yb2-Yb2, have been isolated and purified to homogeneity (28,42). An additional homo-dimeric isozyme, GST P (GST Yp-Yp), was also isolated as a major GST, retainable on the affinity column and detected by CDNB, from the liver cytosol of rats bearing hepatic hyperplastic nodules induced by hepatocarcinogens (29,43,44). These seven rat liver GSTs are well studied on their primary structures (45), immunochemical properties (46), substrate specificities, tissue distribution, and species difference (41). Classification of these GSTs has been made in relation to human GSTs a -6, p, and 7r mainly based on their immunochemical homology and N-terminal amino acid sequences (46). The rat liver GSTs with subunit proteins Ya and Yc belong to class alpha, GSTs with Ybl and Yb2 to class mu, and GST P to class pi (46) (Table VI). N-terminal amino acid sequences of GST subunit proteins in each class have a strong homology and can be readily differentiated from those of the other classes of GSTs. As to GST Yrs-Yrs, no homology was found in its N-terminal amino acid sequence and immunochemical property to any of the known rat GSTs. Furthermore, the Yrs subunit protein contained  higher molar compositions of leucine and histidine residues than that of any other GST (Table II in the Miniprint). GST Ya, isolated as a minor GST protein from the rat liver cytosol has been demonstrated by an immunochemical study not to belong to class alpha nor to class mu, although no further attempt has been made to classify the isozyme by using the class pi enzyme GST P or an antibody against GST P (47, 48). GST Ya differs from GST Yrs-Yrs in that it can be retained on the S-hexyl-SG affinity column and detected by CDNB with a low activity (much higher than for GST Yrs-Yrs) and has high activities toward EPNP and 4-nitrophenyl acetate. The N-terminal amino acid sequence has not been shown with this GST isozyme.
Several different isozymes have been also found as major GSTs in the extrahepatic tissues of the rat, all of which had high catalytic activities toward CDNB similarly to the aforementioned hepatic GSTs. They were, a class alpha enzyme, GST Yk-Yk, existing in the testis, kidney, and lung (60, 61), and four class mu enzymes, GSTs Ynl-Yn2 and Yn2-Yn2, both in the testis (55), GST Ynl-Ynl in the brain (54, 62), and GST YO-YO in the testis (40,56). Except GST Yk-Yk, these enzymes were retained on the S-hexyl-SG affinity column (40,54,55,63).
In the rat liver cytosol, GST Yk-Yk also existed as a very minor enzyme (41,61), and the subunit protein Yn as a component of very minor hetero-dimeric GSTs associating with the subunit proteins Ybl and Yb2 (51,64,65). The class pi enzyme, GST P, exists rather as a major isozyme in the normal rat kidney and small intestine (60,66,67). The subunit protein Yk reacted with the antibodies to the class alpha enzyme subunits Ya and Yc, but did not with those of the other classes of enzyme subunits (60). Similarly, the extra-hepatic class mu enzymes reacted only with the antibodies to the hepatic class mu enzyme subunits Ybl and Yb2 (46,60).
GST Yrs-Yrs isolated in the present study, however, was found to be completely different from any of these three classes of GSTs in all respects so far as estimated from their chromatographic behaviors, immunochemical properties, substrate specificities, and N-terminal amino acid sequences (Table VI).

GST Yrs-Yrs
was also completely different in substrate specificity and in subunit Mr and p1 values from GST E (5-5), although both of them appeared in the flow-through fraction of the S-hexyl-SG affinity column chromatogram of the rat liver cytosol without being retained. GST E is the only reported GST isozyme that can neither be retained on the affinity column (66), nor show any catalytic activity toward CDNB and DCNB (30, [68][69][70]. GST E has been demonstrated to be highly active toward the epoxide EPNP (30, 6%70), toward which GST Yrs-Yrs, however, had no activity, and to be completely inactive toward menaphthyl sulfate (69), the best substrate among examined sulfate esters for GST Yrs-Yrs (Table III). GST E has a subunit Mr value of 24,700 and p1 value of 7.3 with the preparation from Sprague-Dawley rat liver (70) whereas the Mr and p1 values of the hepatic subunit protein Yrs of the Sprague-Dawley rat were 26,000 and 7.9, respectively.
Little systematic study has been made on the immunochemical property of GST E, although antibodies raised against a few rat liver GSTs were demonstrated to show no cross-reactivity with this enzyme (30,71). In addition, no information has been available on the N-terminal amino acid sequence of the GST E protein.
A GST isozyme, designated as 5*-5* because of its similarity to GST 5-5, has very recently been isolated from the nuclei of rat liver cells (72). This enzyme can neither be retained on the S-hexyl-SG affinity column nor cross-reactive with the antibodies against GSTs 1-Z and 3-4. GST 5*-5* has high activities toward EPNP and DNA hydroperoxide, but shows no appreciable activity toward CDNB. The N-terminal amino acid sequence has not been shown with this GST isozyme.
Earlier than 20 years ago, an attempt was made by Gillham (73) to prove the existence of an enzyme catalyzing GSH conjugation of benzyl and menaphthyl sulfates as putative precursors of the benzyl-and menaphthyl-mercapturic acids excreted into the urine of rats given the corresponding arylmethanols (74,75). He found two GST isozymes active toward menaphthyl sulfate exist in the rat liver cytosol by an isoelectric focusing method and partially purified one of them to a purification fold of 76 (76). However, the partially purified GST preparation might be different from GST Yrs-Yrs, because the former was active toward DCNB and 4-nitrobenzyl chloride (69). GST Yrs-Yrs had no appreciable activity toward these chlorinated substrates (Table V). He might have isolated one of the two uncharacterized GSTs active toward 5 HCR sulfate other than GST Yrs-Yrs, which were shown in the chromatofocusing chromatogram (Fig. 3 in the Miniprint). Little was reported by Gillham about molecular data of his enzyme. Our preliminary study indicated that the enzyme eluted just behind the Yrs-Yrs peak at a lower pH range in chromatofocusing (Fig. 3 in the Miniprint) was separable from GST Yrs-Yrs on a blue Toyopearl column and had a crossreactivity with the anti-Yrs-IgG preparation (data not shown). GSTs play a key role in preventing a variety of reactive metabohtes of xenobiotics from their attacks on cellular biomacromolecules (41,77,78), which may induce necrosis and tumorigenesis of tissues. Either excessive formation or relatively low GST-mediated scavenging of the reactive metabolites may result in their covalent binding to DNA, leading to mutation or death of cells. A typical example is N-hydroxyacetylaminofluorene, a hepatocarcinogen, which has been demonstrated to be activated by phenol sulfotransferase IV (79,80) and to be merely scavenged in nonenzymatic manner by GSH (81,82). However, the active metabolites, sulfate esters of nonhepatocarcinogenic arylmethanols, are rapidly and completely scavenged by GSTs in the rat liver cytosol fortified with 22,23), so that the carcinogens can neither bind covalently to DNA (16, 23) nor induce mutation of cells (15,17,22,23) so far as examined in uitro.
GST Yrs-Yrs may play a central role in scavenging the reactive sulfate esters of carcinogenic arylmethanols in the nontarget organ, rat liver, which has the highest level of hydroxysteroid sulfotransferases activating the carcinogens among all the examined tissues of the rat (24). Rat skin, a well known target organ for 5-HCR (5), which has a sulfotransferase activity to activate the carcinogen,3 was found by the immunoblotting method to lack in GST Yrs-Yrs. Isolation and purification of the other rat liver GST isozymes active toward 5-HCR sulfate are in progress in our laboratory.