THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 261, No. 12, Issue of April 25, PP. 5363-5367,1986 Printed in U.S.A.

Covalent Labeling of the Nonsubstrate Ligand-binding Site of Glutathione &Transferases with Bilirubin-Woodward’s Reagent K*

(Received for publication, November 8,1985)

Thomas D. BoyerS From the Liver Studies Unit, Veterans Administration Medical Center, and the Department of Medicine and Liver Center, University of California, Sun Francisco, California 94121

The dimeric enzyme glutathione S-transferase B is composed of two dissimilar subunits, referred to as Ya and Y,. Transferase Yay, and the Yay, homodlmer were purified from rat liver cytosol. An enol ester derivative of bilirubin (bilirubin-Woodward’s reagent K) was prepared and used to label covalently the non- substrate ligand-binding site on these two proteins. There was a linear relationship between the amount of bilirubin-Woodward‘s reagent K added to the reaction mixture and the amount of labeling achieved up to a ratio of 2:1 (bilirubin-Woodward’s reagent K : protein- Yayc). A maximum of 0.87 mol of label bound per mol of transferase Yay,. At higher molar ratios, the label appeared to also be binding at a second site on the enzyme. The label blocked the nonsubstrate ligand- binding site of the two transferases but not the catalytic site. The divalent reagent was shown to label equally the Ya and Y, subunits of transferase Yay,, suggesting that the single high affinity bilirubin-binding site pres- ent on this protein is formed by an interaction between the subunits rather than residing on a specific subunit. At low ratios of label to protein, bilirubin-Woodward’s reagent K appears to label specifically the nonsubstrate ligand-binding site of two forms of glutathione S- transferase, and use of this label should allow for the localization of the nonsubstrate ligand-binding site in the primary amino acid sequence of the Y, and Y, subunits.

The glutathione S-transferases (EC 2.5.1.18) are a family of enzymes present in the soluble fraction of many tissues (1, 2). They are dimeric proteins of Mr = 50,000 which are formed by similar and dissimilar subunits. In the rat, several different subunits have been identified by NaDodSO4-po1yacrylamide’ gel electrophoresis, and two of these subunits termed YP (Mr = 25,000) and Y, (Mr = 28,000) form a family of three cationic enzymes: Yay,, Yay,, and Y,Y, (3-6).

In addition to catalyzing the reaction between GSH and

* This work was supported by National Institutes of Health Grant GM-31555 and the Medical Research Service of the Veterans Admin- istration Medical Center. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$To whom correspondence should be addressed Liver Studies Unit (151K), Veterans Administration Medical Center, 4150 Clement St., San Francisco, CA 94121. ’ The abbreviations used are: NaDodS04, sodium dodecyl sulfate; PBS, phosphate-buffered saline (10 mM sodium phosphate and 150 mM NaC1); BW, bilirubin-Woodward’s reagent K.

‘Recently, a new nomenclature has been suggested for the rat glutathione S-transferases (6). Using this nomenclature, Y, = 1 and Y, = 2.

electrophilic substrates, the transferases also bind a variety of compounds that they do not metabolize, and these nonsub- strate ligands can act as inhibitors of enzymatic activity (7, 8). Early studies found that the inhibition of transferase Yay, by a variety of nonsubstrate ligands, including bilirubin, was competitive, suggesting that the catalytic and nonsubstrate ligand-binding sites were similar (8). Subsequent studies, however, are more consistent with the concept that the cata- lytic site and bilirubin-binding site are different (9-11) al- though some nonsubstrate ligands may bind at the active site of the enzyme (10).

There is continued uncertainty about the role of the Y. and Y, subunits in the binding of bilirubin by the glutathione S- transferases. One group of investigators (12,13) reported that transferase Yaya bound, with high affinity, 2 mol of bilirubin/ mol of enzyme, and transferase Y,Y, lacked this high affinity binding site. The binding of bilirubin by transferase Y a y , which has a single high affinity site, therefore was ascribed to the presence of the Y, subunit (13). Similarly, the binding of lithocholate by transferases Yay, and Yay, was reported to be exclusively associated with the Y, subunit (14). Other studies, however, have found only a single high affinity bind- ing site for bilirubin on transferase Yay, (5,15,16), suggesting that the nonsubtrate ligand-binding site may be formed by an interaction between the subunits rather than residing on a specific subunit per se.

In the current study, we have studied the interaction of transferases Yay, and Yay, with the affinity label BW. This water-soluble enol ester derivative of bilirubin had been used previously to label covalently the primary bilirubin-binding site of human serum albumin. The covalent label is thought to be bilirubin bound to the protein via its two carboxyl groups (17). We questioned whether this same reagent might also label the primary bilirubin-binding site of the transfer- ases. In this report, the reaction between the affinity label and glutathione S-transferases is shown to be specific for the nonsubstrate ligand-binding site at low molar ratios of BW to enzyme. In addition, these studies support the idea that the Y, and Y, subunits are both involved in the formation of the nonsubstrate ligand-binding site.

EXPERIMENTAL PROCEDURES

Materials-Bilirubin, cyanogen bromide-activated Sepharose 4B, Sephadex G-25, Sephacryl S-200, and sulfobromophthalein-S-gluta- thione-agarose were obtained from Sigma. Woodward’s reagent K was purchased from Fluka Chemical Co., Hauppauge, NY, and 644- 14C]aminolevulinic acid hydrochloride was from New England Nu- clear.

Synthesis of BW-Recrystallized bilirubin IX” (18) was reacted with Woodward’s reagent K as described previously (17). This and all subsequent steps were protected from light. The deep yellow to orange in color zone was collected from the Sephadex G-25 column, lyophilized, and used in the subsequent experiments to label the

5363

5364 Labeling of Transferase Nonsubstrate Ligand-binding Site glutathione S-transferases. The absorption spectra of the synthesized BW was identical to that obtained by Kuenzle et al. (17). [l4CC] Bilirubin (a generous gift of Dr. John Gollan, Harvard Medical School, Boston, MA) had been purified from rat bile that was collected following the intravenous injection of 6-[4-’4C]aminolevulinic acid hydrochloride (19). [14C]Bilirubin was mixed with unlabeled bilirubin (final specific activity 1.86 X lo6 dpm/pmol) and used to prepare the radiolabeled affinity label as described above. The specific activity of the [14C]BW was 1.62 X lo6 dpm/pmol.

Purification of Glutathione S-Transferases and Enzymatic Assays- The Yay, and Yay, transferase isozymes were purified from rat liver cytosol as described previously (5). The purified enzymes were free of contamination by other proteins and other transferase isozymes based on NaDodS04-polyacrylamide gel electrophoresis (20) and isoelectric focusing in polyacrylamide gels (21). Glutathione S-trans- ferase activity was measured spectrophotometrically using GSH and 1-chloro-2,4-dinitrobenzene as substrates (22). Protein was deter- mined by the method of Lowry et al. (23) with bovine serum albumin as the standard.

Reaction between B W and Glutathione S-Transferases-The method described by Kuenzle et al. (17) was followed. The molar concentrations of the individual enzymes were calculated using mo- lecular weights of 50,000 (Yay,) and 53,000 (Yay,) (6). The individual enzymes (17-40 nmol) in PBS (pH 7.4) were mixed with the specified molar amounts of BW (also in PBS) and stirred at room temperature, under nitrogen and in the dark, for 2 h. EDTA was added to a final concentration of 1 mM, and the pH was increased to 9.4 by the addition of imidazole. The reaction was continued with stirring in the dark and under nitrogen for an additional 6-8 h. The reaction mixture was then applied to a Sephacryl S-200 column (1.5 X 45 cm) in PBS (pH 7.4, a t 4 “C). The covalent product was eluted in PBS (pH 7.4) a t a flow rate of 45 ml/h. Two-ml fractions were collected and assayed for enzymatic activity and absorbance at 447 nm. Frac- tions containing labeled enzyme were pooled and applied to an albumin affinity column (0.13 X 11 cm) in PBS (pH 7.4). The affinity column had been synthesized from cyanogen bromide-activated Seph- arose 4B following the manufacturer’s instructions. The labeled en- zyme was eluted from the column in PBS (pH 7.4), concentrated over a stirred membrane (Amicon, YM-lo), and stored at -80 “C until used. 25-60% of the starting protein was recovered as labeled enzyme. The EM of the covalently labeled enzymes were calculated by mea- suring the amount of label bound using radioactive BW.

Azo Coupling and Absorption and Fluorescent Spectra-The cova- lent products were reacted wtih the diazo reagent as described previ- ously (17). Absorption and fluorescent spectra were measured with a Perkin-Elmer dual-beam recording spectrophotometer and a Perkin- Elmer MPF-44B fluorescent spectrophotometer. Binding studies, us- ing protein fluorescence, were performed at 25 “C in PBS (pH 7.4) (5).

NaDodS04-Polyacrylamide Gel Ekctrophoresis and Autoradwgra- phy-NaDodS04-polyacrylamide gel electrophoresis was performed as described previously (20). The gels were stained with Coomassie Blue and treated with ENHANCE (New England Nuclear) preceding autoradiography. Radioactivity was also measured in individual 1- mm gel slices that had been solubilized with 30% Hz02 preceding addition of scintillation fluid.

Sulfobromophthalein Affinity Column-Covalently labeled trans- ferase Yay, was chromatographed on a sulfobromophthalein-S-glu- tathione column (1 X 11.5 cm) in PBS (pH 7.4) (24). In preliminary studies, unlabeled transferase Yay, adhered well to this column and was then eluted by increasing the buffer pH to 9.4.

RESULTS AND DISCUSSION

Labeling of Transferases with B W-In preliminary studies, we observed that covalent binding of BW to the transferases did not occur at a neutral pH. Increasing the pH of the reaction mixture with imidazole resulted in covalent labeling which increased during the first 6 h of the reaction but changed little between 6 and 8 h. Reactions were not run beyond 8 h because of concerns about the reported oxidative degradation of the label (17). After 6-8 h at pH 9.4, the reactants were chromatographed on a Sephacryl S-200 col- umn. A single peak of transferase activity was obtained, and this activity was associated with a yellow chromaphore (Fig. 1).

Fraction Number

FIG. 1. Elution profile of 5-200 column of the products of reaction with glutathione S-transferase Yay. and BW. The reaction was performed as described under “Experimental Proce- dures,’’ and the reactants were then applied to and eluted from the S-200 column in PBS (pH 7.4). Enzyme activity was measured at 340 nm using 0.5 mM l-chloro-2,4-dinitrobenzene and 1 mM GSH as substrates. The adherent label was identified by measuring absorb- ance at 447 nm.

The covalent nature of the label was established by precip- itating and washing the labeled enzyme (Yayc) with trichlo- roacetic acid and CHCls. The yellow color remained with the precipitated protein despite these treatments. The glutathione S-transferases are reversibly denatured by 8 M urea (14). We therefore dialyzed labeled transferase Yay, against 8 M urea and 0.2 M mercaptoethanol for 96 h. The yellow chromaphore remained with the denatured enzyme in the dialysis bag. In addition, the covalent label remained with the enzyme when the mixture was applied to an albumin affinity column. In contrast, when enzyme and BW were freshly mixed in PBS and then immediately applied to the albumin affinity column, all of the label adhered to the column and did not elute with the enzyme. Last, the covalent label remained bound to the protein following NaDodS04-polyacrylamide gel electropho- resis (see below).

The reaction between BW and human serum albumin is thought to occur via the activated carboxyl groups leaving the tetrapyrrole structure of bilirubin intact (17). If this were also true for the transferases, then the covalently bound label should remain susceptible to cleavage at its central methylene bridge by the diazo reagent and yet remain bound to the enzyme. To test for this, covalently labeled Yay, was treated with the diazo reagent, as described under “Experimental Procedures.” A pink derivative with an absorption peak at 520 nm was formed. The diazotized labeled enzyme was then dialyzed against 8 M urea for 56 h, and no movement of the pink derivative into the dialysate was detected. We believe, therefore, that both activated carboxyl groups of the label had bound to the enzyme.

Absorption and Fluorescent Spectra-The absorption max- imum of free bilirubin in PBS (pH 7.0) was 434 nm, which is similar to other reported values (25). The binding of bilirubin to its primary high affinity site on bovine serum albumin or human serum albumin causes a red shift in the absorption maximum of bilirubin (17, 25). A similar red shift has been reported for the binding of bilirubin to transferase Yayc (26). We observed a shift of the absorption maximum of bilirubin from 434 to 448 and 442 nm with transferases Yaya and Yayc, respectively. In contrast, when the Yaya and Yay, enzymes were labeled covalently with BW, the absorption maximum was 442 nm (Yaya) and 425 nm (Yay,) (Fig. 2). The EM for covalently labeled transferases Yaya and Yay, were, respec- tively, 34,600 and 36,500, in PBS (pH 7.0). The absorption

Labeling of Transferase Nonsubstrate Ligand-binding Site 5365

40-

A

.: 30-

7 I

c? - 0

W

2 20-

10-

1 : , 1 , 1 , , , 1 , .""

. ..

360 380 400 420 440 460 480 500 520 540

Wavelength (nm) FIG. 2. Absorption spectra of labeled transferases Y.Y. and

Yay,. The spectra were obtained in PBS (pH 7.0), and the respective molar ratios of bilirubin to protein were 1:l Y.Y. noncovalent ( . . . .), 1.1:1 Yay, noncovalent (-), 1.7:l Y.Y. covalent (- - -), and 1.6:l Yay, covalent (-). The XM for the covalently labeled enzymes was calculated by determining the amount of label bound using ["CC) BW.

Wavelength (nm)

FIG. 3. Fluorescent spectra of labeled transferase Yay,. The spectra were obtained in PBS (pH 7.0) with LXc = 280 nm. The cuvettes contained, respectively, transferase Yay, (0.13 p M ) alone (-), with 0.23 p~ free bilirubin (. . . .), or with bilirubin bound covalently (M).

maximum of bilirubin bound covalently to human serum albumin is also different from the maximum obtained when bilirubin is bound noncovalently. The difference observed with albumin was felt to be secondary to an alteration in protein conformation rather than to the label binding at an alternate site on the protein (17). Based on subsequent stud- ies, we also believe that, despite the difference in spectra between covalently and noncovalently bound bilirubin (Fig. 2), we have specifically labeled the nonsubstrate ligand-bind- ing site of transferases Yaya and Yayc.

The intrinsic protein fluorescence of the transferases is quenched by bilirubin, and if BW binds at the same site as bilirubin, then BW should also quench protein fluorescence. We observed that free BW quenches the fluorescence of transferase Yay, and that the KD values for BW and bilirubin were similar, being 0.62 and 0.77 pM, respectively. Similarly, when bound covalently, the label also reduced the intensity of fluorescence of transferase Yayc (Fig. 3).

NaDodS04-Polyacrylamide Gel Electrophoresis and Auto- radiography-BW will label other proteins that lack a biliru- bin-binding site, but the nature of the divalent reagent results in cross-linking of the proteins and only labeled polymers are formed (17). To determine whether the divalent label was causing polymerization of the transferases, the labeled en- zymes were subjected to NaDodS04-polyacrylamide gel elec- trophoresis. Although a small amount of enzyme subunits were cross-linked by the label such that they had an M , value of 40,000-66,000, most of the labeled enzymes migrated as the individual subunits (Fig. 4). Polymers composed of more than two subunits were minimally present. If the reagent were reacting nonspecifically with the transferases, polymers with M , values that were multiples of 25,000-28,000 would be expected to predominate.

Bhargava et al. (12) reported previously that sulfobromo- phthalein formed a covalent adduct only with the Y. subunit of transferase Yayc, which supports the idea that the Y. subunit contains the high affinity nonsubstrate ligand-bind- ing site. We were interested, therefore, in the distribution of the label on the individual subunits of transferase Yay, after the reaction with [14C]BW. Transferase Yay, was reacted with a 10-fold excess of [14C]BW to obtain sufficient labeling to allow for autoradiography. The molar ratio of covalently bound label to protein was 1.6:l. Lower molar ratios could not be used because of the relatively low specific activity of the label. Following NaDodS0,-polyacrylamide gel electro- phoresis, the gel was subjected to autoradiography. The Y.

1 M.W. % 2 ~ 1 0 ' ~

36

29 24

20.1

14.2 FIG. 4. NaDodS04-polyacrylamide gel electrophoresis of

transferase Yay, (lune I ) or Y.Y. (lune 2) that had been covalently labeled with BW. 30 pg of the enzymes was applied to the gels. The ratio of label to protein was 1.6-1.7:l. The gels were stained with Coomassie Blue. The M , values for the protein standards are shown in the center lane. Transferase Y.Y. was slightly contam- inated (<5%), by Yay, which accounts for the presence of the Y, subunits in lane 2.

5366 Labeling of Transferase Nonsubstrate Ligand-binding Site

and Y, subunits were found to be labeled equally, and a small amount of the label was associated with the cross-linked subunits (Fig. 5). The gels were also sliced, dissolved with 30% H202, and counted in a liquid scintillation counter. 84% of the radioactivity was associated with the Y, and Y, sub- units, and the distribution of the radioactivity between the two subunits was equal. Only 14% of the radioactivity was associated with the cross-linked subunits. In these exper- ments, the ratio of label to protein exceeded 1, and the labeling of the Y, subunit could have occurred after the binding site on the Y, subunit was filled. If the latter had occurred, then the distribution of the label between the two subunits should have been unequal (-40% more on the Y, subunit), and this was not observed. The equal distribution of the radiolabel between the two subunits therefore suggests that both sub- units are involved in the formation of the bilirubin-binding site on glutathione S-transferase Yay,.

Maximum Amount of Labeling-A number of investigators (8,9, 16) have demonstrated that transferase Yay, contains a single high affinity binding site for bilirubin. If the label is reacting only at the high affinity binding site, then a maxi- mum of 1 mol of label should be bound per mol of enzyme. Increasing amounts of [I4C]BW were reacted with a fixed amount (20 nmol) of transferase Yay,. A double reciprocal plot of the molar ratio of [I4C]BW to protein in the reaction mixture uersus the molar ratio of covalently bound label to protein was linear up to a ratio of 2:1, with a maximum of 0.87 mol bound per mol of enzyme. At a ratio of 5:1, there was about three times as much label bound as expected from the relationship found at the lower ratios (Fig. 6). At the lower ratios of BW to protein (52:1), the label appeared to be specifically reacting a t a single binding site on the protein. At

FIG. 5 . Autoradiogram of transferase Y.Y, labeled with [14C]BW. The ratio of label to protein was 1.6:l. The gel was prepared as described for Fig. 4 and treated with ENHANCE prior to autora- diography.

36 /

I I I 1 I I l 0 .5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

[male BW/mals protelnr'

FIG. 6. Double reciprocal plot of relationship between ratio of BW to protein in reaction mixture and amount of labeling achieved. Various amount of [14C]BW were mixed with a fixed amount (20 nmol) of transferase Yay,. Following chromatography on an S-200 and albumin affinity column, the amount of label bound per mole of enzyme was determined. The circled value was not used to calculate the regression line.

the higher ratios, labeling at more than a single site was observed. A second, low affinity, bilirubin-binding site has been reported to be present on transferase Yayc (9, 16), and it is possible that this is where the BW was reacting.

Specificity of the Label-Kuenzle et al. (17) had shown previously that bilirubin inhibited the labeling of human serum albumin by BW. Labeling was, however, reduced by a maximum of 50% (&fold excess of bilirubin) so that absolute specificity of the label was not established. We also attempted to inhibit labeling by adding bilirubin to the reaction mixture. The low yield of labeled protein that we obtained ( 4 0 % of starting material) from these experiments prevented us from investigating inhibition by bilirubin.

The nonsubstrate ligand-binding site of the transferases allows for their purification using a sulfobromophthalein af- finity column (24). In addition, sulfobromophthalein inhibits the binding of bilirubin by ligandin, which suggests that they bind at the same site on the protein (26). It seemed reasonable to expect, therefore, that covalently bound bilirubin should prevent the binding of the labeled enzyme to a sulfobromo- phthalein column if the label was at the sulfobromophthalein (bilirubin)-binding site. All of the labeled transferase Y.Y. failed to adhere to the affinity column, whereas unlabeled enzyme was retained and subsequently eluted by increasing the buffer pH (Fig. 7). Covalently labeled Yayc also did not adhere to the affinity column. The sulfobromophthalein-bind- ing site, and therefore the bilirubin-binding site, had been blocked by the covalently bound label. The catalytic site, in contrast, was not blocked by the covalently bound label as the enzyme retained catalytic activity. The specific activity of the labeled enzyme was, however, reduced by 40-50% (5- fold excess of BW).

In conclusion, we believe that BW has specifically labeled the nonsubstrate ligand-binding site of glutathione S-trans- ferases Yaya and Yayc. We base this conclusion on the follow- ing observations. 1) BW, whether free or bound covalently to transferase Yay,, quenched the protein fluorescence in a man- ner similar to bilirubin. 2) There was a linear relationship between the amount of BW added to the reaction mixture and amount of labeling achieved (up to a ratio of 2:l for BWY,Y,) with a maximum of 0.87 mol of label bound per

Labeling of Transferase Nons

Fraction Number

FIG. 7. Chromatography of labeled transferase Yay, on a sulfobromophthalein affinity column. Transferase Yay, labeled with BW was chromatographed on a sulfobromophthalein column. Following application of the enzyme, the column was eluted with PBS (pH 7.4) and then beginning at fraction 21 with 0.1 M NaP04 (pH 9.4). One-ml fractions were collected, and enzyme activity and location of the label were determined as described for Fig. 1.

mol transferase Yay,. 3) Labeling proteins that lack a specific bilirubin-binding site with BW leads to cross-linking of the protein monomers by the divalent reagent and the formation of polymers (17). There was a minimum of cross-linking of the subunits of transferases Y.Y. and Yayc when labeled with BW. 4) The binding site for sulfobromophthalein (bilirubin), but not the catalytic site, was blocked by the covalent level. In addition, at high molar ratios of BW to protein, both the Y, and Y, subunits were labeled equally. This observation, in concert with the finding of only a single high affinity binding site for bilirubin on transferases Y.Y. and Yayc (5,8, 15, 16)) suggests that the nonsubstrate ligand-binding site of trans- ferase Yayc is formed by an interaction between the Y, and Y, subunits. Subsequent studies using this label should allow for localization of the bilirubin-binding site in the primary amino acid sequence of the Y. and Y, subunits (27,28).

Acknowledgments-I wish to thank Jennifer Roberts and Neil Saley for their skilled technical assistance, Dr. Donald Vessey for his helpful comments about the manuscript, and Blanche Mays for typing the manuscript.

REFERENCES 1. Jakoby, W. B. (1978) Adv. Enzymol. 4 6 , 383-414 2. Boyer, T. D., and Kenney, W. C. (1985) in Biochemical Pharma-

cology and Toxicology (Zakim, D., and Vessey, D. V., eds) Vol.

rubstrate Ligand-binding Site 5367 1, pp. 297-364, John Wiley & Sons, New York

S. J. (1977) Biochim. Biophys. Acta 492 , 163-175

9912

Pharmol . 32,1843-1850

Pharmucol. 33,2539-2540

3. Bass, N. M., Kirsch, R. E., Tuff, S. A., Marks, I., and Saunders,

4. Mannervik, B., and Jensson, H. (1982) J. Biol. Chem. 257,9909-

5. Boyer, T. D., Kenney, W. C., and Zakirn, D. (1983) Biochem.

6. Jakoby, W. B., Ketterer, B., and Mannervik, B. (1984) Biochem.

7. Levi, A. J., Gatmaitan, Z., and Arias, I. M. (1969) J. Clin. Invest.

8. Ketley, J. N., Habig, W. H., and Jakoby, W. B. (1975) J. Biol.

9. Bhargava, M. M., Listowsky, I., and Arias, I. M. (1978) J. Biol.

10. Vander Jagt, D. L., Wilson, S. P., Dean, V. L., and Simons, P. C.

11. Boyer, T. D., Vessey, D. A., Holcomb, C., and Saley, N. (1984)

12. Bhargava, M. M., Ohmi, N., Listowsky, I., and Arias, I . M. (1980)

13. Bhargava, M. M., Ohmi, N., Listowsky, I., and Arias, I. M. (1980)

14. Hayes, J. D., Strange, R. C., and Percy-Robb, I. W. (1981)

15. Carne, T., Tipping, E., and Ketterer, B. (1979) Biochem. J. 177 ,

16. Maruyama, H., Arias, I. M., and Listowsky, I. (1984) J. Biol.

17. Kuenzle, C. C., Gitzelmann-Cumarasamy, N., and Wilson, K. J.

18. McDonagh, A. F., and Assisi, F. (1972) Biochem. J. 129, 797-

19. Ostrow, J. D., Hammaker, L., and Schmid, R. (1961) J. Clin.

20. Maizel, J. V. (1971) Methods Virol. 5 , 179-246 21. Kenney, W. C., and Boyer, T. D. (1981) Anal. Biochem. 116,

22. Habig, W. H., Pabst, M. J., and Jakoby, W. B. (1974) J. Biol. Chem. 249,7130-7139

23. Lowry, 0. H., Rosebrough, N. J., Fan, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275

24. Wolkoff, A. W., Bhargava, M. M., Chung, C., and Gatmaitan, Z. (1979) Proc. SOC. Exp. Biol. Med. 160,150-153

25. Reed, R. G. (1977) J. Biol. Chem. 252,7483-7487 26. Tipping, E., Ketterer, B., Christodoulides, L., and Enderby, G.

(1976) Biochem. J. 157,211-216 27. Tu, C.-P. D., Lai, H.-C. J., Li, N., Weiss, M. J., and Reddy, C . C.

(1984) J. Biol. Chem. 259,9434-9439 28. Pickett, C. B., Telakowski-Hopkins, C. A., Ding, G.J. F., Argen-

bright, L., and Lu, A. Y. H. (1984) J. Biol. Chem. 259 , 5182- 5188

48,2156-2167

Chem. 250,8670-8673

Chem. 253,4112-4115

(1982) J. Biol. Chem. 257 , 1997-2001

Biochem. J. 2 1 7 , 179-185

J. Biol. Chem. 255,718-723

J , Biol. Chem. 2 5 5 , 724-727

Biochem. J. 197,491-502

433-439

Chem. 259,12444-12448

(1976) J. Biol. Chem. 251,801-807

800

Invest. 4 0 , 1442-1452

344-348