Biochimica et Biophysica Acta, 400 (1975) 451-460

© Elsevier ScientificPublishing Company, Amsterdam - - Printed in The Netherlands BBA 37114 T H E ESSENTIAL H I S T I D I N E RESIDUES OF BOVINE PLASMA A M I N E OXIDASE

SHIGEKO TSURUSHIIN, AKIRA HIRAMATSU, MELVIN INAMASU and KERRY T. YASUNOBU* Department of Biochemistry-Biophysics, University of Hawaii School of Medicine, Honolulu, Hawaii 96822 (U.S.A.)

(Received February 10th, 1975)

SUMMARY Ethoxyformic acid anhydride and photooxidation have been used to study the function of histidine residues in bovine plasma amine oxidase. Ethoxyformic acid anhydride at pH 6.1 reacted with nearly all of the histidine residues in the enzyme in 15 min but complete enzyme inactivation occurred in several minutes. The concentration of the reagent which caused 50% inhibition was 2.2.10 -5 M under the conditions of the experiment. The diamine oxidases, Aspergillus niger and pea seedling amine oxidases were also inhibited by ethoxyformic acid anhydride. The concentrations of reagent required for 50% inhibition were 6.6.10 -5 and 3.3.10 -4 M, respectively, for the two enzymes. NH2OH could not be used to regenerate the reacted histidine residues since NH2OH itself inhibited the enzyme. Photooxidation in the presence of 0.001% Rose Bengal at pH 7.0 also inactivated bovine plasma amine oxidase. Histidine was the only amino acid destroyed by photooxidation. About six histidine residues were destroyed but in the presence o f the substrate kynuramine, two less histidine residues were destroyed. Since lysine which is neither a substrate nor inhibitor of the enzyme did not protect the enzyme from photooxidation, it was concluded that two histidine residues, one in each subunit of the enzyme are essential for activity.

INTRODUCTION The plasma o f all animals contains an amine oxidase [1]. It has been shown that the level of the enzyme in females increases during pregnancy [2], in Laennac's and in postnecrotic cirrhosis, during massive neoplastic replacement of the liver, in hepatic sarcoidosis' and during chronic congestive heart failure [3]. Due to the possible medical importance of the enzyme, detailed investigations of the properties of the enzyme are warranted. The ready availability of bovine plasma at the local abattoir has prompted us to investigate the plasma amine oxidase from bovine plasma. The enzyme was initially * To whom requests for reprints should be addressed.

452 partially purified by Tabor et al. [4] and crystallized by Yamada and Yasunobu [5]. Some of the physicochemical properties of the enzyme have been previously reported [6-9]. In the present investigation, some group specific reagents were tested to determine essential amino acid residues in the bovine plasma amine oxidase. The results of these investigations are presented in this report. EXPERIMENTAL PROCEDURES

Materials Highly purified bovine plasma amine oxidase with a specific activity of about 1000 was used in most of the experiments. Highly purified pea seedling amine oxidase (specific activity of 3681) was kindly supplied by J. M. Hill. Highly purified Aspergillus niger amine oxidase, specific activity of 8219, was a gift of Dr O. Adachi. Ethoxyformic acid anhydride was purchased from the Aldrich Chemical Co. Rose Bengal was a product of the Eastman Kodak Co. In general, reagent grade chemicals were used throughout the present investigation.

Methods' The enzyme was assayed by the spectrophotometric procedure of Tabor et al. [4]. The oxidation of benzylamine is followed by the measurement of the benzaldehyde produced at 250 nm at 25 °C. Protein concentrations were determined spectrophotometrically and an -1 F 1°/° cm value at 280 nm of 19.2 was used in the present investigation [5]. Specific activity was defined as the change in absorbance per rain x 1000 per mg protein. Photooxidation of the enzyme was performed in the presence of 0.001-0.01 ~,~ Rose Bengal in 0.1 M phosphate buffer, pH 7.0 [10]. A 150 W-120 V projection lamp, operated at 82 V by use of a variable transformer, was located about 4 inches underneath the water bath set at 25 °C. Samples were removed from the reaction vessel and then passed through a 0.97 × 11.5 cm of Sephadex G-25 column which had been equilibrated with the phosphate buffer. The Rose Bengal adhered tenaciously to the Sephadex column and was removed from the enzyme solution. Amino acid analyses were performed in the Beckman-Spinco Model 120 automatic amino acid analyzer as described by Spackman et al. [11]. Normally, the samples were hydrolyzed in 5.7 M HC1 at 110 °C for 24 h. For tryptophan analyses, the samples were hydrolyzed at 100 °C for 24 h with p-toluenesulfonic acid-indoleethylamine as described by Liu [12]. For the determination of cysteine plus cystine residues, the samples were subjected to dimethylsulfoxide-HCl oxidation as described by Spencer and Wold [13] prior to amino acid analysis. The cysteine content of the enzyme was determined by the procedure described by Boyer [14]. Fo~ the analysis of methionine sulfoxide, alkaline hydrolysis (3.7 M NaOH) was performed on the enzyme prior to amino acid analysis. The reaction of the enzyme with ethoxyformic acid anhydride was carried out by the procedure described by Pradel and Kassab [15]. The enzyme, 0.067 mg, was also reacted with 1.7 mM (final concentration) of iodoacetamide or iodoacetic acid in 0.6 ml of 0.1 M phosphate buffer, pH 6.1, and 0.6 ml of 0.1 M phosphate buffer, pH 7.0, for up to 24 h and then assayed for activity.

453

100,

BO z =E tlJ e: 6 0

~, 4o • O

2o

6

5

4

-log (EFA)

Fig. ]. Effect of ethoxyformic acid anhydride concentration on the inactivation of the enzyme. For each point in the figure, 0.065 mg o£ enzyme in 0.55 ml of 0.! M phosphate buffer, pH 6.1, was incubated with 10 #! o£ varying concentrations o£ ethoxyfon~ic acid anhydride. After a l 0-min incubation period at 25 °C, samples were assayed at pH 7.4.

RESULTS

Ethoxyformic acid anhydride experiments Concentration dependence and time course of inactivation of amine oxidase by ethoxyformic acid anhydride. Fig. 1 shows the concentration dependence of ethoxyformic acid anhydride on the inactivation of the enzyme at p H 6.1. The concentration of enzyme required for 5 0 ~ inhibition was 2.2.10 -s M. The reaction of the

,ooJ0,o"

ioIo-

I/Y

~60=0.06ti I o~

o'

~

,

o

-

~

-

o

~

o

TIME (MINUTES)

Fig. 2. in the experiment, 0.33 mg of bovine plasma amine oxidase in 3.Oral of 0.] M phosphate buffer, pH 6.1, was reacted with 50/~l of a 0.0] M solution of cthoxyformic acid anhydride in acetonitrile. At the indicated time intervals, 0.6-ml aliquots were removed and assayed at pH 7.4 by the usual spectrophotomctric method. In the figure, ([~--[B) and ( A - - A ) represent the change in histidine content (from absorbancc measurements at 242 nm) and activity, respectively.

454 reagent and enzyme is also time dependent and requires a definite time for complete reaction as shown in Fig. 2. The results at p H 6.1 which are supposed to increase the specificity of ethoxyformic acid anhydride for histidine residues did not differ appreciably from the results at pH 7.2. It has been shown by Rosen and Fedorcsak [16] that when N-ethoxyformyl imidazole is formed in the reaction, there is an increase in the absorbance at 242 nm. The amino and mercapto derivatives do not absorb in the 242-290 nm region. Thus, the change in absorbance at 242 nm was monitored as well as the rate of inactivation as shown in Fig. 2. There is a rough parallelism between the loss in activity and the increase in absorbance at 242 nm. When the molar extinction coefficient for the N-ethoxyformyl-histidine of 3200 cm -~- M -1 [17] was used to calculate the number of histidine residues which had reacted, about 43-50 histidine residues had reacted in about I I min. However, the enzyme was nearly inactivated only 2 min after the addition of the reagent and about 32 histidine residues had reacted. Other aspects o['ethoxyformic acid anhydride inactivation of enzyme. Hydroxylamine has been reported to cleave N-ethoxyformyl-histidine residues in proteins and to reactivate enzymes inactivated by ethoxyformic acid anhydride [18]. When the ethoxyformic acid anhydride-inactivated bovine plasma amine oxidase was treated with 0.5 M NHEOH at pH 7.0 no reactivation was observed. This was true even after removal of the NH2OH by passage of the enzyme through the 0.9 >~ 11 cm Sephadex G-25 column. However, a control experiment in which the native enzyme was treated with 0.5 M N H z O H also resulted in an enzyme which was completely inactive. Thus, the inability of the reagent to reactivate the ethoxyformic acid anhydride derivative does not rule out the presence of N-ethoxyformyl-histidine groups in the enzyme. The spectrum of the enzyme after ethoxyformic acid anhydride treatment did show an increase in the absorbance around 242 nm indicating that formation of the N-ethoxyformyl-histidine derivative. In order to check the possibility that ethoxyformic acid anhydride is reacting with an unknown organic cofactor in the enzyme, 7.5 mg of the enzyme was treated with ethoxyformic acid anhydride to completely inactivate it. The visible spectrum of the ethoxyformic acid anhydride-inactivated enzyme was taken but the spectrum was identical to that of the native enzyme. Thus, it appears that the ethoxyformic acid anhydride is either not reacting with the organic cofactor or that if reaction of reagent with the cofactor had occurred, the substitution did not cause change in absorbance of the cofactor. Kinetics of inactivation by ethoxyformie acid anhydride. In these experiments enzyme (0.0065 mg) was incubated with the reagent for 10 min (1.6.10 -5 M final concentration) in 0.6 ml of 0.1 M potassium phosphate buffer and was then assayed at different concentrations of benzylamine (53.6-642 #tool) in 3.0 ml of buffer adjusted to pH 7.2. The amount of benzylamine at the lowest concentrations exceeded the concentration of ethoxyformic acid anhydride by at least 10-fold and therefore in the assay, even if the reagent did react with benzylamine, it did not have much o f an effect on the final benzylamine concentration. When the data was plotted according to the Lineweaver-Burk procedure, non-competitive inhibition was observed. Inactivation Of other diamine oxidases by ethoxyformic acid anhydride. If all diamine oxidases carry out the oxidation of benzylamine by a similar mechanism, one might expect the reagent to inactivate other diamine oxidases. Therefore, the effect of ethoxyformic acid anhydride on the pea seedling and A. niger amine oxidases

455 toe %

- ~ ' ~ ,

Pea Seedlin 9 Fnzyme

8C

_~ e o :2_ ~.. 40 o "

Niger Enzyme '

t

-log EFA conc. (M)

Fig. 3. The inactivation of other diamine oxidases by ethoxyformic acid anhydride. The experimental conditions are identical to those described in the legend for Fig. 1. In the experiment, 0.0182 mg of the pea seedling enzyme and 0.00949 mg of the .4. niger enzyme were used. and bovine aorta amine oxidase were checked. The effect of varying the ethoxyformic acid anhydride concentration on the activity of the pea and mold enzymes are shown in Fig. 3. Insufficient quantities of bovine aorta amine oxidase were on hand for detailed studies. However, 60 % inhibition of enzyme was observed when 0.07 mg of aorta amine oxidase (specific activity of 108 was incubated with 6.67.10 -s M ethoxyformic acid anhydride (final concentration) for 10 min at pH 6.1 and then assayed at pH 7.4 with the usual benzylamine assay. No aorta amine oxidase activity was detected when the final concentration of the reagent was increased to 3.3.10 -4 M. Thus, all the enzymes tested above were rather specifically inactivated by ethoxyformic acid anhydride.

Photooxidation with Rose Bengal Photooxidation of bovine plasma amine oxidase in the presence of Rose Bengal. Photooxidation appeared to offer a method of confirming that the histidine residues in the enzyme are essential for activity. This procedure allows one to quantitate the amino acid residues altered by performing amino acid analyses of the photooxidized enzyme samples. The initial experiments involved testing various light sources and it was found that the 150 W-120 V lamp operating a t 8 2 V provided the best results. The use of a 250 W-115 V lamp resulted in a very rapid rate of enzyme inactivation, while the 150 W-120 V lamp extended the period required for complete inactivation to about 50-60 min. The first-order inactivation of the bovine plasma amine oxidase by photooxidation in the presence of 0.001, and 0.01% Rose Bengal are shown in Fig. 4. Amino acid analyses of the photooxidized enzyme. Amino acid analyses of the photooxidized enzyme in the presence of 0.001 and 0.025 % Rose Bengal disclosed that histidine was the only amino acid destroyed. The rate of destruction of the histidine is shown in Fig. 5. The other amino acid residues which are generally susceptible to photooxidation were not oxidized. Methionine sulfoxide was not de-

456 lO0(

80 ~ 70 60 50 40.

.~ 20E

..3 l0 8 6

o

,~

,Co

\

,~o

TIME (MIN) Fig. 4. Photoinactivation of the enzyme in the presence of Rose Bengal. For each point in the figure, 0.067 mg of enzyme in ] .0 ml of 0.1 M phosphate buffer, pH 7.0, which contained varying amounts of Rose Bengal were photooxidized as described in the experimental section. In the figure, ([~--[~), ( O - - © ) , and ( A - - ~ ) represent 0.01, 0.0025 and 0.001 ~ dye, respectively. At the times indicated, the samples were passed through a Sephadex G-25 column which removed the Rose Bengal. The protein concentration was determined from the absorbance at 280 nm while the activity was determined by the standard spectrophotometric assay.

I00

~20 •~ 40

0 70

20

5

l0 PA0 Fig. 5. Number of histidine residues destroye,d by photooxidation. The samples which were obtained as described in the legend to Fig. 4, were hydrolyzcd in 5.7 M HC1 for 24 h and then analyzed in the amino acid analyzer. In the figure, ( © - - © ) and ( A - - A ) represent samples which were photooxidized in the presence of 0.0025 and 0,001% final concentration of dye, respectively. The times at which samples were removed are shown in the figure in minutes. Oxidized His per mole

457 tected when the photooxidized enzyme was analyzed in the automatic amino acid analyzer after alkaline hydrolysis for 24 h. No oxidation of cysteine residues in the amine oxidase was detected by Boyer's method [14]. For the tryptophan analyses, p-toluenesulfonic acid was used to hydrolyze the photooxidized enzyme by the method described by Liu [12]. Compounds whichprotect from photooxidation of enzyme. Different compounds were added to the photooxidation mixtures to see if they would protect the enzyme from photoinactivation. Two substrates, namely kynuramine and benzylamine and lysine which is neither a substrate nor an inhibitor of the amine oxidase were tested. The effect of varying the concentration of these compounds on the photoinactivation of the amine oxidase is shown in Fig. 6. The Km at p H 7.2 for benzylamine and

I00"

s~

= 4(1 "6 |

/

IJ 2

0

p,

"e

,o"

® - - - o Benzyl Amine Kynuramlne x - - - x Lyeine

tO

0.02

0.~:)4

oloe

0.'08

61o

Protector Concentration (M)

Fig. 6. Effect of the concentration of protectors on the photooxidation of enzyme. In the experiment, 0.11 mg of enzyme, 0.0025% (final concentration) of Rose Bengal irt 1.0 ml of 0.1 M potassium phosphate buffer, pH 7.4, was irradiated for 70 min. At the times indicated aliquots were removed and passed through a 0.97 × 11.3 cm column of Sephadex G-25 to remove the dye. The samples were assayed for activity by the spectrophotometric procedure of Tabor et al. [4].

kynuramine are 1.3.10 -4 and 6.5.10 -6 M, respectively. Control runs in which the enzyme was omitted during the photooxidation experiments showed that kynuramine was not destroyed and benzylamine was at least 95 ~o intact after the photooxidation experiments. Lysine did not protect the enzyme from photooxidation. Determinatioti of the number of histidine residues in the active site. Since kynuramine was the best protector of the enzyme and was also a substrate of the enzyme, it was used to determine the number of active site histidine residues. The results are summarized in Table I and from the one experiment in the presence of 0.001 ~o Rose Bengal and 2.10 -2 M kynuramine, four histidine residues were oxidized with retention of 86 ~o of the activity. In the control sample, which contained no kynuramine, six histidine residues were destroyed and only 16~o of the activity remained. Thus, there are about two active site histidine residues.

458 TABLE I DETERMINATION OF THE PHOTOSUSCEPTIBLE AMINO ACID RESIDUES A N D ACTIVITY AT VARIOUS TIME INTERVALS D U R I N G PHOTOOXIDATION OF ENZYME IN THE PRESENCE AND ABSENCE OF K Y N U R A M I N E Amino acid compositions of the protein samples were measured as described in the experimental section. Amino acid residue

Unoxidized

Photooxidized 0.01 ~ dye"

Photooxidized 0.0025~ dye*"

Photooxidized 0.001 ~ dye*'*

35 min

100 min

70 rain No amine

70 min Plus amine

70 rain No amine

70 rain Plus amine

Histidine 50 Cysteine* 4 Tyrosine 58 Tryptophan 26 Methionine 4 Methionine sulfoxide 0

45

41

58 26 4 0

58 26 4 0

43 4 58 26 4 0

46 4 58 26 4 0

44 4 58 26 4 0

46 4 58 26 4 0

Relative activity

56 ~

19 %

90%

16%

86%

100 To

7°'/o

* For Exp 1, 1.0 mg of the enzyme in 1.0 ml of 0.1 M phosphate buffer, pH 7.0, was used. Each sample was removed and then passed through a Sephadex G-25 column (0.9 x 11.0 cm) to remove the Rose Bengal. The fractions containing the enzyme were pooled and protein and activity determinations were made. ** In Exp 2, 0.213 mg of enzyme was used for each time interval taken, otherwise the conditions of the experiment were identical to those used for Exp I. The final concentration of kynuramine was 1.5.10 -2 M.

*** In Exp 3, 0.294 mg of enzyme was used for each time interval taken, otherwise the conditions of the experiments were identical to those used for Exp 1. The final concentration of kynurarnine was 2.10 -a M. * Determined by Boyer's [14] procedure.

DISCUSSION

The diamine type of amine oxidase is found in animals, plants, bacteria, and mold [1]. The enzyme from bovine plasma [5] has been obtained in crystalline form and some of the physicochemical properties of the enzyme have been determined [6-8]. Moreover, the enzyme has been shown to be a copper-protein [8, 9] and to contain an organic cofactor ol unknown structure which was erroneously reported to be pyridoxal phosphate [8]. The structure of the organic cofactor remains to be elucidated. In the present investigation, our goal has been to determine the role of the histidine residues in the enzyme. In the present investigation plasma amine oxidase was not inactivated by 0.1 M iodoacetic acid and iodoacetamide, even when incubated for 24 h at pH 7.4 (results not shown). However, bovine plasma is not the only enzyme whose essential histidine residues apparently do not react with these haloacetates. Therefore, investigations were made with the histidine reagent, ethoxyformic acid anhydride. The reagent has been shown to react with the histidine residues of arginine and creatinine kinases [t5]; ribonuclease [18]; tryptophan synthetase [19];

459 and lactate and 3-phosphoglyceraldehyde dehydrogenases [20]. However, as is the case with many group specific reagents, it was shown to react, in addition, with the amino group of the NH2-terminal amino acid in pepsin and the active site serine chymotrypsin [18] and a tryptophan residue in serum albumin [16]. Ethoxyformic acid anhydride reacted with all of the 50 histidine residues in the enzyme. However, even after 2 min of reaction 32 residues reacted with nearly complete inactivation of enzyme activity. However, the reagent had several drawbacks. (1) It was not possible to test the protective effect of the substrate since they all contain amino groups which also react with ethoxyformic acid anhydride. (2) Hydroxylamine cleavage of the Nethoxyformyl-histidine derivative could not be used to confirm the essential role of the histidine residues since the compound is a potent inhibitor of diamine oxidases. However kinetic studies showed that ethoxyformic acid anhydride was a noncompetitive inhibitor with respect to benzylamine. For an irreversible inhibitor such as ethoxyformic acid anhydride, the type of inhibition observed suggests but does not prove that the reagent was reacting with active center histidine residues. It was interesting to note that ethoxyformic acid anhydride not only inactivated bovine plasma amine oxidase but also the pea seedling amine oxidase, the A. niger amine oxidase and the aorta amine oxidase. Thus, it appears that all these enzymes may possibly contain essential histidine residues. It was desirable to confirm the presence of active site histidine residues in the plasma amine oxidase by a second procedure. Photooxidation was chosen since it was possible to quantitate the amino acid residues modified after photooxidation of the enzyme. Photooxidation has been reported to destroy the cysteine, methionine, tyrosine, tryptophan and histidine residues in proteins [21]. Photooxidation of bovine plasma amine oxidase in the presence of anionic dye Rose Bengal inactivated the enzyme at pH 7.0. Numerous amino acid analyses of acid hydrolyzates of the photooxidized enzyme showed that the tyrosine, tryptophan, cysteine and methionine residues in the enzyme were not altered. Thus, the report of Westhead [22] that photooxidation in the presence of Rose Bengal is more specific for histidine residues than other dyes is further confirmed by the present investigation. About six histidine residues were destroyed when the enzyme was completely inactivated but the value depended on the concentration of the Rose Bengal used. When benzylamine or kynuramine, which are substrates of the enzyme, were added considerable protection from inactivation occurred and with kynuramine as the protector, two less histidine were destroyed. Thus, it was concluded that one histidine residue is present in each of the active sites of the enzyme and these histidine residues are required for activity. Evidence has been previously presented that bovine plasma amine oxidase is an enzyme which contains two identical subunits held together covalently by cystine linkages [23]. Additional investigations are required to show the precise role of histidine residues in the oxidation of the substrate. Possible roles for the histidine residues include binding of the substrate by hydrogen bonding or the role as an acid-base catalyst during the oxidation of the substrate. Indeed, several theoretical schemes for the latter role have been published by McEwen et al. [24] for rabbit serum amine oxidase and by Buffoni et al. [25] for the pig plasma amine oxidase. However, these schemes must be altered since pyridoxal phosphate is not the coenzyme of the diamine oxidases [8].

460 ACKNOWLEDGMENTS This investigation was s u p p o r t e d in part by G r a n t No. G B 18739 f r o m the N a t i o n a l Science F o u n d a t i o n , G r a n t M H 21539 f r o m the N a t i o n a l Institutes o f Health, a g ran t f r o m the A m e r i c a n H e a r t A s s oci at i o n and the Hawaii H ear t Association. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Zeller, A. E. (1963) The Enzymes, 2nd ed., Vol. 8, p. 337, Academic Press, New York Ahlmark, A. (1944) Acta Physiol. Scand. 9, Suppl. 28 McEwen, Jr, C. M. (1971) Methods Enzymol. 17B, 692 Tabor, C. W., Tabor, H. and Rosenthal, S. M. (1954) J. Biol. Chem. 208, 645 Yamada, H. and Yasunobu, K. T. (1962) J. Biol. Chem. 237, 1511 Yasunobu, K. T. and Yamada, H. (1963) Proceedings of the symposium on chemical and biological aspects of pyridoxal catalysis, Rome, (1962) p. 453, Pergamon Press, New York Yasunobu, K. T., Achee, F. M., Chervenka, C. and Wang, T. M. (1968) Symposium on pyridoxal enzymes, p. 139, Nagoya, Japan, Maruzen Press Watanabe, K., Smith, R. A., Inamasu, M. and Yasunobu, K. T. (1972) Adv. Biochem. Psychopharmacol. 5, 107. Yasunobu, K. T. and Smith, R. A. (1971) Methods Enzymol. 17B, 698 Means, G. E. and Feeney, R. E. (1971) Chemical modification of proteins, Holden-Day, Inc., San Francisco Spackman, D. H., Moore, S. and Stein, W. H. (1958) Anal. Chem. 30, 1190 Liu, T. Y. (1972) Methods Enzymol. 25B, 44 Spencer, R. L. and Wold, F. (1969) Anal. Biochem. 32, 185 Boyer, P. (1954) J. Am. Chem. Soc. 76, 4321 Pradel, L. A. and Kassab, R. (1968) Biochim. Biophys. Acta 167, 317 Rosen, C. B. and Fedorcsak, I. (1966) Biochim. Biophys. Acta 130, 401 Muhlrad, A., Hegyi, G. and Horanyi, M. (1969) Biochim. Biophys. Acta 181, 184 Melchoir, W. B. and Fahrney, D. (1970) Biochemistry 9, 251 Miles, E. W. and Kummagai, H. (1973) Am. Soc. Exp. Biol. 32, 473 Abstr. d'Elodi, P., Thoai, N. V. and Roche, J. (1968) Biochemical evolution and homologous enzymes, p. 195, Gordon and Breach, New York Ray, Jr, W. J. (1967) Methods Enzymol. 11, 490 Westhead, E. W. (1967) Biochemistry 4, 2139 Achee, F. M., Smith, R. A., Yasunobu, K. T. and Chervenka, C. (1969) Biochemistry 1, 4329 McEwen, Jr, C. M., Cullen, K. T. and Sober, A. J. (1966) J. Biol. Chem. 241, 4544 Buffoni, F., Della Certe, A. and Ignesti, G. (1972) Pharmacol. Res. Commun. 4, 99