ARCHIVES

OF BIOCHEMISTRY

AND BIOPHYSICS

Vol. 296, No. 1, July, pp. 17-26, 1992

Purification and Characterization Synthase from Bovine Liver’

of Prostaglandin

F

Lan-Ying Chen, Kikuko Watanabe, and Osamu Hayaishi’ Department

of Enzymes and Metabolism,

Osaka Bioscience Institute,

Furuedai,

Suita, Osaka 565, Japan

Received October 31,199l

Prostaglandin Da 1 1-ketoreductase activity of bovine liver was purified 340-fold to apparent homogeneity. The purified enzyme was a monomeric protein with a molecular weight of about 36 kDa, and had a broad substrate specificity for prostaglandins D1, Da, Ds, and Hz, and various carbonyl compounds (e.g., phenanthrenequinone and nitrobenzaldehyde, etc.). Prostaglandin Dz was reduced to 9a, 1 la-prostaglandin Fz and prostaglandin Ha to prostaglandin Fz, with NADPH as a cofactor. Phenanthrenequinone competitively inhibited the reduction of prostaglandin Dz, while it did not inhibit that of prostaglandin Hz. Moreover, chloride ion stimulated the reduction of prostaglandin D2 and carbonyl compounds, while it had no effect on that of prostaglandin Hz. Besides, the enzyme was inhibited by flavonoids (e.g., quercetin) that inhibit carbonyl reductase, but was not inhibited by barbital and sorbinil, which are the inhibitors of aldehyde and aldose reductases, respectively. These results indicate that the bovine liver enzyme has two different active sites, i.e., one for prostaglandin D2 and carbonyl compounds and the other for prostaglandin Hz, and appears to be a kind of carbonyl reductase like bovine lung prostaglandin F synthase (Watanabe, K., Yoshida, R., Shimizu, T., and Hayaishi, O., 1985, J. Biol. Chem. 260, 7035-7041). However, the bovine liver enzyme was different from prostaglandin F synthase of bovine lung with regard to the K, value for prostaglandin Dz (10 pM for the liver enzyme and 120 PM for the lung enzyme), the sensitivity to chloride ion (threefold greater activation for the liver enzyme) and the inhibition by CuSO., and HgC& (two orders of magnitude more resistant in the case of the liver enzyme). These results suggest that the bovine liver enzyme is a subtype of bovine lung prostao 1982~~ad~rni~ press, IW. glandin F synthase.

Prostaglandin (PG)3 D2 11-ketoreductase, which catalyzes the conversion of PGDz to PGF2, was discovered independently in rat lung and rabbit liver by Watanabe et al. (1) and Wong (2), respectively. Liston and Roberts found that one of the urinary metabolites from PGD2 was Sa,llfi-PGF2, which is a stereoisomer of PGFza (3), and that the 100,OOOgsupernatant of human liver catalyzed the reduction of PGDz to Sa,ll@-PGF2 (4). Moreover, Sa,llo-PGF2 was synthesized from PGD2 by the 100,OOOg supernatant of human lung (5), and also was formed by the perfusion of PGD2 into rabbit liver (6) and rat lung (7). 9a,ll@-PGFz and PGF2, in some systems have been found to exert similar biological effects, including an increase in blood pressure in rats (4), contraction of human bronchial smooth muscle (5), constriction of human airway passages in vitro and in vivo (8) and natriuresis in rats (9). On the other hand, 9ar,ll&PGF2 has distinct biological characteristics, including coronary artery vasoconstrictor (4, 6, 10) and platelet antiaggregatory (4, 11) properties. Watanabe et al. purified PGD2 11-ketoreductase activity from bovine lung to apparent homogeneity (12). The purified enzyme, which is a kind of carbonyl reductase (EC 1.1.1.184), reduced not only PGDz and carbonyl compounds at one active site but also PGH2 at another. The product from PGD, was Sa,llfi-PGF2 and that from PGHz was PGF2, (13). Therefore, the enzyme was called PGF synthase (EC 1.1.1.188). The enzyme was a monomeric protein with a M, of 36,666 consisting of 323 amino acids, and its amino acid sequence showed high homology compared with those of aldo-keto reductases (14). Human liver aldehyde reductase (EC 1.1.1.2), which is a kind of aldo-keto reductase, also catalyzes the reductions of PGHz and other substrates (aldehyde compounds) at different active sites (15). Recently, PGDz 11-ketoreductase activity

r Portions of this paper (including the entire Materials and Methods section and part of the Results section, Figs. 1, 3, 4, and 9, and Tables 1, 3, and 5) are presented as a Miiprint Supplement at the end of this paper. ’ To whom correspondence should be addressed. Fax: 06-872-4818.

3 Abbreviations used PG, prostaglandin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; RP-HPLC, reversed-phase high-performance liquid chromatography; TLC, thin-layer chromatography.

0003.9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

17

18

CHEN,

WATANABE,

was reported in a 100,OOOgsupernatant fraction of bovine liver by the same group (16). The enzyme, which differs from the PGD2 11-ketoreductase of rabbit liver (2), rat lung (l), and PGF synthase of bovine lung (12), catalyzed the conversion of PGDp to 9a,11&PGF2 with a low K, value for PGD2 (6 PM) and showed a profile of immunoprecipitation with anti-bovine lung PGF synthase antibody different from that of bovine lung PGF synthase. In the present study, we describe the purification and characterization of bovine liver PGF synthase, and compare this enzyme with PGF synthase of bovine lung. Furthermore, the effects of various anions, especially chloride and metal ions, and inhibitors on the PGDz ll-ketoreductase, phenanthrenequinone, and nitrobenzaldehyde reductase activities of bovine liver PGF synthase were also compared. RESULTS

Purification of the PGD:, 11 Aetoreductase activity from bovine liver. Table I summarizes the results of a typical purification of PGDz 11-ketoreductase activity from bovine liver. The enzyme, which resides in the cytosol fraction of bovine liver, was purified to apparent homogeneity by sequential ammonium sulfate fractionation and chromatographies on DEAE-cellulose and red Sepharose (Fig. 1) columns. In a typical purification, 17% of the initial PGD2 11-ketoreductase activity was recovered, representing a 342-fold enrichment in specific activity (Table I). During the purification steps, phenanthrenequinone and nitrobenzaldehyde reductase activities were also measured. After red Sepharose chromatography, the enzyme was separated from the other carbonyl reductases and was purified to apparent homogeneity. The purified enzyme had an.approximate specific activity of 96 nmol/ min/mg of protein for PGDz. On sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE), the purified enzyme migrated as a single Coomassie bluestained band with a M, of about 36 kDa (Fig. 2), which is the same as that of the native enzyme determined by fast protein liquid chromatography with TSK gel G 2000 SW in an ultrapac column at a flow rate of 0.5 ml/min. These results indicate that the purified enzyme is a monomeric protein. In Western blot analysis with rabbit antiserum against PGF synthase from bovine lung, a single immunoreactive band was observed for the purified enzyme of bovine liver (Fig. 2). When the purified enzyme was stored at -80°C at a concentration of 1.5 mg/ml in 10 mM potassium phosphate buffer, pH 7.0, its activity was stable for several months. Approximately 80 to 90% of the initial activity remained after a l-month storage of the enzyme at the same concentration at 4°C or room temperature. However, the enzyme activity was reduced less than one-fifth after freezing and thawing several times, and no conversion of PGDz to ScU,llP-PGF2 was

AND

HAYAISHI

SDS-PAGE

Western blot

0.4 E 0.6r

4

:

0.6 v

-

front -

Y

.

1

2

. . .I....

.

I

4610 log Yr

FIG. 2. SDS-PAGE, Western blot, and iVf, of bovine liver PGF synthase. Five micrograms (for Coomassie blue dye) and 10 ng (for Western blot) of PGF synthase from bovine liver were electrophoresed in a 12.5% polyacrylamide gel in the presence of 0.1% SDS. Following the electrophoresis, the gel was separated into two parts. Half was stained for protein with Coomassie brilliant blue R-250, and the other half was used for immunostaining with bovine lung PGF synthase antibody (dilution 1300 times). The molecular weight of PGF synthase (0) was estimated from its mobility relative to those of the following standard proteins (0) from right to left: phosphorylase b (97,400), bovine serum albumin (66,200), ovalbumin (42,700), carbonic anhydrase (31,000), soybean trypsin inhibitor (21,500), and lysozyme (14,000).

observed in incubations using enzyme that had been boiled at 100°C for 5 min. Substrate specificity of the purified enzyme. Table II shows that the purified enzyme has a broad substrate specificity for carbonyl compounds. Quinones (e.g., 9,10phenanthrenequinone) were found to be the best substrates. Aromatic aldehydes (e.g., nitrobenzaldehyde) and ketones (e.g., hydrindantin, PGDi, PGDP, and PGD& were also readily reduced. The reaction rates for PGDz and PGHz were 31 and l%, respectively, of that for nitrobenzaldehyde. The K,,, values for PGDi , Dz , D3, Hz, phenanthrenequinone, and nitrobenzaldehyde were 7,10, 5.6, 25, 2, and 14.3 PM, respectively. DL-Glyceraldehyde, an aliphatic aldehyde, was also reduced, but its K,,, was relatively high (143 PM). D-Aldose, which is a substrate for aldose reductase (27), was reduced at a low rate with a K, of 135 mM, but D-glucose was not reduced. Except for PGJ2, which was reduced at a much slower rate, other PG compounds such as PGAi , PGB:!, and PGEz were not reduced. Testosterone, D-glucuronic acid, and 4-carboxybenzaldehyde, which are substrates for aromatic aldehyde-ketone reductase and aldehyde reductase derived from rabbit liver (28) and for aldehyde reductase from pig liver (29), respectively, were not reduced. 2,6-Dichloroindophenol and potassium ferricyanide, which are substrates for human brain carbonyl reductase (30) and bovine lung PGF synthase (12), were reduced neither by human liver carbonyl reductase (22) nor by the purified enzyme of bovine liver. These results suggest that PGDi , Dz, D3, and Hz are the best naturally occuring substrates.

PROSTAGLANDIN

F SYNTHASE TABLE

OF BOVINE

19

LIVER

II

Substrate Specificity of Bovine Liver PGF Synthase

Substrate 9,10-Phenanthrenequinone Daunorubicin Menadione Duroquinone Ubiquinone-10 p-Nitrobenzaldehyde Phenylglyoxal 4Carboxybenzaldehyde p-Nitroacetophenone Hydrindantin Cyclohexanone 5&Dihydrotestosterone PGD, PGD2 PGDB PGJs PGHz PGA,-GSH* PGA, PGBz PGEz Testosterone 2,6-Dichloroindophenol D,L-Glyceraldehyde n-Butyraldehyde D-Glucuronic acid D-Glucose D-XyIose Potassium ferricyanide

Concentration bM) 0.01 0.25 0.25 0.50 0.05 0.50 1 0.50 0.50 0.10 0.50 0.05 0.15 0.15 0.15 0.15 0.08 0.10 1 1 1 0.05 0.04 10 1 10 100 100 1

Specific activity (~mol/min/mg) 0.339 0.184 0.167 0.054 0.032 0.310 0.207 ND 0.174 0.167 0.159 0.130 0.046 0.095 0.089 0.008 0.003 ND ND ND ND ND ND 0.206 0.109 ND ND 0.065 ND

Relative activity (%) 109 59 54 17 10 100 67 ND 56 54 51 42 15 31 29 3 1 ND ND ND ND ND ND 67 35 ND ND 21 ND

KWI (FM) 2 16.7

14.3 20 11.9 3.1 53 17.5 7 10 5.6 25

143

135,000

Note. PGDr ll-ketoreductase and PGHp reductase activities were determined by radioisotope methods as described under Materials and Methods. Substrate specificity for other substrates was measured spectrophotometrically at 37°C as described under Materials and Methods. The 4-nitrobenzaldehyde reductase represents 100% activity. ’ Not detectable. * PGA, and 10 mM GSH were incubated in the assay mixture for 10 min before the purified enzyme was added, as reported by Wermuth (30).

Identification of the reactionproducts formed from PGDz and PGH,. In Fig. 3A, the enzymatic product formed from [3H]PGDz appeared as a single peak (the recovery of radioactivity was about 90%) on reversed-phase highperformance liquid chromatography (RP-HPLC) with a retention time of 14 min, which was identical to that of authentic Sa,llfi-PGFz. This indicates that the reduction product of PGDz by the purified enzyme is Sa,ll@-PGFz. On the other hand, there was a mixture consisting of about 50% [14C]9a,llP-PGF2 with a retention time of 20 min, identical to that of the authentic marker, and 50% [14C]PGF2, enzymatically formed from [14C]PGHz in the absence of quinone (Fig. 3B). In the presence of 100 PM phenanthrenequinone, a competitive substrate for PGDz, the enzymatic reduction of [14C]PGH2 yielded a mixture consisting of approximately 90% PGFr, and 10% 9a,ll@PGFz (Fig. 3C). This result suggests that the reaction product from PGHB by the purified enzyme is PGFP, and

that Sa,ll@-PGF2 from PGHz is formedvia PGD2, which is nonenzymatically produced from PGH2 during incubation. Therefore, we named the purified enzyme PGF synthase of bovine liver. Cofactor specificity. Table III summarizes the cofactor specificities of the purified enzyme with PGDB and other carbonyl compounds as substrates and with NADPH or NADH serving as a cofactor. The ratio of reduction rate for PGDz with 0.56 mM NADH to that with 0.56 mM NADPH was 3%. The K,,, for NADPH with PGDz as a substrate was around 3 PM. Although the reaction rates for reduction of other compounds as substrates for PGF synthase of bovine liver were high with 0.56 mM NADH, the K, values for NADH with carbonyl compounds as substrates were 70 to 580 times greater than those for NADPH. Therefore, the purified PGF synthase from bovine liver exhibits a high specificity for NADPH.

20

CHEN,

WATANABE,

AND

HAYAISHI

Enzyme properties of PGF synthase of bovine liver. Under our assay conditions, the enzyme activity

was found to be linear up to 1.8 pg of protein (Fig. 4A) and up to 60 min of incubation time at 37°C (Fig. 4B). The optimal pH of the reaction was around 6 to 7 (Fig. 4C). The purified enzyme from bovine liver had a high affinity for PGDz as substrate. As shown in Fig. 5, the K,,, for PGDz was estimated to be about 10 PM. Two different active sites. The purified enzyme of bovine liver catalyzed the conversion of PGHz to PGFP, as well as that of PGD2 to Sa,ll&PGF2. As shown in Table II, the specific activities for PGD2 and PGHp were 96 and 3 nmol/min/mg of protein, respectively. Phenanthrenequinone inhibited the conversion of PGD2 to 9a,llpPGFz, but did not inhibit and slightly stimulated that of PGH2 to PGFz, (Fig. 6A). The inhibition by phenanthrenequinone of PGDz reductase activity was competitive with a Ki of 0.9 PM (Fig. 6B), and that of nitrobenzaldehyde was also competitive, with a Ki of 23 PM (data not shown). These results suggest that PGDz, phenanthrenequinone, and nitrobenzaldehyde are reduced at the same active site of PGF synthase of bovine liver, while PGHz is reduced at a different active site. Modulation compounds.

of the enzyme activity by chloride and metal

A comparison of the effects of halides on PGF synthase of bovine liver with those on PGF synthase of bovine lung is presented in Fig. 7. NaF inhibited the reduction of PGDz to Sa,ll&PGFx by each enzyme, whereas the other halides had significantly different effects on the two enzymes. NaCl, NaBr, and NaI were found to stimulate PGDz 11-ketoreductase activity, as well as phenanthrenequinone and nitrobenzaldehyde reductase activities, of bovine liver PGF synthase in different ranges of concentration. The concentrations of NaCl, NaBr, and NaI that gave maximum velocity of enzyme

150

KmnlOfiM

100

-0.1

0.1

0.2

.

0.3

-1 Km Y

-0

I

I

I

50

100

150

PGD2101W FIG. 5. Kinetic profile of bovine liver PGF synthase with PGDs as substrate. Enzyme assays were performed as described under Materials and Methods in the presence of 0.02-0.37 pg of enzyme.

QoI

1

2 ” 20 40 60 60 Phenanthrenequinone (PM)

lad n

@I I

J

[PGDf’

(PM-‘)

FIG. 6. Effect of phenanthrenequinone on PGDr 11-ketoreductase and PGHz reductase activities of bovine liver PGF synthase. (A) PGDr 11-ketoreductase (0) and the PGHp reductase (0) activities determined as described under Materials and Methods with the addition of increasing amounts of phenanthrenequinone. (B) Lineweaver-Burk plots for the competitive inhibition of the PGDP 11-ketoreductase activity by phenanthrenequinone. The PGDs 11-ketoreductase activity was assayed as described under Materials and Methods in the presence of 37 ng of enzyme and phenanthrenequinone in the following concentrations: none (o), 0.05 PM (O), 0.2 PM (n), and 0.4 pM (m).

activity were approximately 0.5-1.0, 0.5, and 0.2 M, respectively. However, concentrations of NaBr and NaI beyond 0.5 and 0.2 M, respectively, were rather inhibitory. In contrast to PGF synthase of bovine liver, PGD2 llketoreductase activity of bovine lung PGF synthase was apparently not affected at all by concentrations of NaCl below 0.5 M, and was actually inhibited by concentrations of NaBr and NaI above 0.5 and 0.2 M, respectively. As shown in Fig. 8, NaCl activated the PGD2 ll-ketoreductase, and phenanthrenequinone and nitrobenzaldehyde reductase activities of bovine liver PGF synthase, but did not affect the PGHz reductase activity of the enzyme. This result indicates that PGF synthase of bovine liver has two different active sites: one for PGD2, phenanthrenequinone, and nitrobenzaldehyde and one for PGHz. When the effects of KCl, MgC&, and CaCl, on the PGDp 11-ketoreductase and phenanthrenequinone and nitrobenzaldehyde reductase activities were examined, all chloride compounds activated these enzyme activities like NaCl (Fig. 9). The kinetic mechanism of activation by chloride was further investigated. When phenanthrenequinone or nitrobenzaldehyde was used as substrate, the

PROSTAGLANDIN

F SYNTHASE

8 300 E 5 E ‘g 200 5 z 2g 100 N B 0

0

0.5

1.00

0.5

1.0

Sodium Halide (M) FIG. 7. Comparison of effects of NaF (O), NaCl (0), NaBr (m) and NaI (0) on PGDz 11-ketoreductase activity of bovine liver and lung PGF synthases. The bovine liver PGF synthase (A) and bovine lung PGF synthase (B) activities were determined as described under Materials and Methods with the addition of increasing amounts of sodium halide in the presence of 0.5 and 6 pg of enzyme, respectively.

enzyme was activated noncompetitively against each substrate, and uncompetitively against NADPH by NaCl (data not shown). In addition to halides, the effects of some other anions on the enzyme were studied. Compounds containing different anions such as Na2S04, NaHC03, NaAc, and Na2C03 at a concentration of 0.5 M inhibited the PGDz 11-ketoreductase activities of the purified enzyme from bovine liver by 22,48, 78, and 86%, respectively, whereas Na2S203 at the same concentration caused an approximately l&fold activation of the enzyme. Additionally, selenium-containing compounds like SeC& and NapSeOB did not affect enzyme activities at all, even in concentrations of up to 30 and 100 mM, respectively. As shown in Table IV, PGDz 11-ketoreductase and phenanthrenequinone and nitrobenzaldehyde reductase activities of PGF synthase derived from bovine liver were apparently inhibited by some metal compounds. The extent of inhibitory potency, as shown by increasing IC& for PGDz 11-ketoreductase activity, was AgNOB (0.005 mM) > HgClz (0.3) > CuS04 (0.4) > CdSO, (0.95) > ZnSOl (1.4), and their I&, values for the phenanthrenequinone and nitrobenzaldehyde reductase activities were about 2 orders of magnitude less than those obtained for PGDz 11-ketoreductase activity. In contrast to the above metal compounds, the PGDz 11-ketoreductase activity was similarly and slightly stimulated by 100 mM MnS04, Li2S04, and NiSOl. Moreover, MnS04 and L&SO4 at concentrations of 100 mM also activated the phenanthrenequinone and nitrobenzaldehyde reductase activities of the enzyme, while NiSO, at the same concentration inhibited them. These differences indicate that the effects of metal compounds on the enzyme may depend upon the substrate. The PGDz reductase activity of bovine lung PGF synthase was also inhibited by the above metal compounds. The

OF BOVINE

21

LIVER

order of inhibitory potency, as shown by increasing IC&, was CuS04 (0.003 mM) > HgClz (0.007) > AgNOB (0.009) > CdS04 (0.35) > ZnSOI (0.5) > NiS04 = MnS04 (160). However, bovine lung PGF synthase was much more susceptible to CuS04 and HgC& inhibition judging from the I(&, values, which were two orders of magnitude less than those for bovine liver PGF synthase. Numerous inhibitors Sensitivity to various inhibitors. have been used to identify the members of the aldo-keto reductase family and for the study of the carbonyl reductases (27). Table V shows the effects of various inhibitors on the PGDz 11-ketoreductase and phenanthrenequinone and nitrobenzaldehyde reductase activities of the purified enzyme from bovine liver. Among them, PGDz ll-ketoreductase activity was subjected to the greatest inhibition by quercetin, ethacrynic acid, and mefenamic acid. Flavonoids (e.g., quercetin), which are preferred for the identification of carbonyl reductases (27, 31), were the most potent inhibitors. The glycoside inhibitors (rutin and quercitrin) also inhibited the PGDz 11-ketoreductase and phenanthrenequinone and nitrobenzaldehyde reductase activities. Moreover, valproic acid, benzoic anhydride, and disulfam, which inhibited the carbonyl reductase of human liver (22), Japanese monkey liver (24), and rat ovary (32), respectively, also inhibited all three activities of bovine liver PGF synthase. The purified enzyme of bovine liver was also sensitive to the inhibition by febufen, mefenamic acid, clemastine, cyproheptadine, and flurbiprofen. On the other hand, piroxicam and warfarin stimulated the three enzyme activities. At the concentrations listed in Table V, the purified enzyme was not inhibited by barbital, sorbinil, and pyrarol, which are potent inhibitors of

300 ';; k )I .; 200 =x f 'G g 100

0

I

0.2

J

,

0.3

0.4

0.5

NaCl (M) FIG. 8. Effects of NaCl on PGD2 11-ketoreductase (O), phenanthrenequinone reductase (m), nitrobenzaldehyde reductase (Cl), and PGHz reductase (0) activities of bovine liver PGF synthase. Enzyme assays were carried out as described under Materials and Methods in the presence of increasing concentrations of NaCl. The PGDz 11-ketoreductase and PGHB reductase activities were determined by the radioisotope method, and phenanthrenequinone reductase and nitrobenzaldehyde reductase activities were measured spectrophotometrically.

22

CHEN,

WATANABE,

AND

TABLE

HAYAISHI

IV

Comparison of the Effects of Various Metal Compounds on Bovine Liver with Those on Bovine Lung PGF Synthase Bovine liver PGF synthase phenanthrenequinone

Concn. (t!n%)

(mM)

Rel. act. (%)

IC& mM

Concn. mM

Rel. act. (%)

WA AgNO3 CdS04 CUSOb ZnSO, NiSOl

0.3 0.005 0.95 0.4 1.4

1.0 0.05 5.0 5.0 5.0 10 100 10 100 10 100

6 11 7 7 29 102 146 120 164 99 154

0.003 0.001 0.007 0.05 0.04

0.05 0.05 0.2 1.0 2.0 10 100 10 100 10 100

6 3 6 6 0 66 44 335 315 129 177

LizSOI

Bovine lung PGF synthase

reductase

Compounds

MnSO,

Nitrobenzaldehyde

reductase

PGDz 11-ketoreductase

PGF Synthase

I&, bM) 0.001 0.001 0.002 0.004 0.09

PGDzreductase

Concn. (mM)

Rel. act. (%)

IQ,,, 6-d

Concn. bM)

Rel. act. (%)

0.1 0.05 0.1 0.1 0.5 10 100 10 100 10 100

0 0 0 4 0 60 40 153 297 135 182

0.007 0.009 0.35 0.003 0.5 160

0.1 1.0 2.0 0.5 5.0 10 100 10 100 10 loo

3 a 9 5 1 106 80 101 76 114 114

160

Note. Enzyme assays of bovine liver PGF synthase with PGD2, phenanthrenequinone, and nitrobenzaldehyde as substrates were carried out. in the reaction system containing 0.1 M Mes-Tris buffer, pH 6.5, and measured by the radioisotope method for PGDz and by the spectrophotometric method for phenanthrenequinone and nitrobenzaldehyde at 37’C as described under Materials and Methods. The PGDz reductase activities of bovine lung PGF synthase were estimated with an assay mixture containing 0.1 M Mes-Tris buffer, pH 6.5, and other conditions were the same as reported by Watanabe et al. (12). The enzyme activity for each substrate without various metal compounds is 100%.

aldehyde reductase, aldose reductase (EC 1.1.1.21), and alcohol dehydrogenase (EC l.l.l.l), respectively (27). These results suggest that PGF synthase of bovine liver is a kind of carbonyl reductase. DISCUSSION In 1985, we demonstrated that bovine lung PGF synthase is a dual function enzyme that catalyzes the reduction of PGD2 to Sa,ll&PGF2 and that of PGHz to PGF2, at different active sites (12). Recently, bovine liver PGD 11-ketoreductase, which reduced PGDz to Sa,llP-PGF2, was found by the same group (16). The Km value for PGDz of the crude hepatic PGD 11-ketoreductase was low (6 PM), and the immunoreactivity of its PGDp ll-ketoreductase activity to anti-PGF synthase of bovine lung antiserum was about one-fifth compared with that of bovine lung PGF synthase. In this study, the enzyme was purified to apparent homogeneity from bovine liver (Figs. 1 and 2, and Table I), and its properties were characterized. The purified enzyme also catalyzed the reduction of PGHz to PGFz, as well as that of PGDB to Sa,llP-PGF2 (Fig. 3) at different active sites (Figs. 6 and 8). Therefore, the liver enzyme belongs to the PGF synthase group. The specific activity for PGD2 of liver PGF synthase was over 30-fold greater than that for PGHz, and its Km values for PGDz and PGHz were 10 PM (Table II and Fig. 5) and 25 PM (Table II), respectively. On the other hand, the K,,, values for PGDz and PGHz of bovine lung PGF synthase

were 120 and 10 PM, respectively, and the V,,, value for PGHz was half of that for PGD2 (12). Thus the Km value for PGD2 of the liver enzyme is one order of magnitude less than that of lung PGF synthase, and is about the same as that of other PG-related enzymes (33-36). These results suggest that the liver PGF synthase prefers PGDp to PGH2 and that the lung enzyme prefers PGHz to PGDz as a substrate. Sa,ll@-PGF2 is also synthesized from PGDz by PGDz 11-ketoreductase of rabbit liver (2). PGDz ll-ketoreductase purified from the rabbit liver had a A& of 66 kDa, a K,,, value for PGDp of 200 PM, and a specific activity of 26 X 10e3 nmol/min/mg of protein. The liver PGF synthase had a M, of about 36 kDa, a Km value for PGD2 of 10 PM, and a specific activity of about 100 nmol/min/mg of protein. Thus, the bovine liver PGF synthase is quite different from the PGDz 11-ketoreductase from the rabbit liver in terms of both molecular and catalytic properties. Bovine liver and lung PGF synthases appear to be a kind of carbonyl reductase in view of A!, (Fig. Z), substrate specificity (Table II), preference for NADPH as a cofactor (Table III), and the sensitivity to inhibitors of carbonyl reductase (Table V). Moreover, both PGF synthases were inhibited by several heavy metal compounds (Table IV). Heavy metal compounds have inhibitory effects on carbony1 reductases (22, 24, 27, 32), but some metal ions have no effect on the carbonyl reductase activity of the mouse lung (37) and guinea pig liver microsomes (38).

PROSTAGLANDIN

F SYNTHASE

Metal ions and some non-heavy metal ions may take part in the metabolic regulation of carbonyl reductases in uivo. Terada et al. purified carbonyl reductase from the cytosol of bovine liver (39). This enzyme catalyzes the reduction of testosterone and nitrobenzaldehyde with K,,,s of 29 and 154 PM, respectively, in the presence of NADPH or NADH as a cofactor, and is inhibited by piroxicam, warfarin, cromoglycate, and ketoprofen. Bovine liver PGF synthase was specific for NADPH (Table III). The enzyme reduced nitrobenzaldehyde with a K,,, value of 14.3 PM and did not reduce testosterone (Table II). Moreover, piroxicam and warfarin stimulated the enzyme (Table V). In addition, the extent of inhibition of the enzyme by cromoglycate and ketoprofen was less than that of the 4benzalpyridine reductase. These differences suggest that bovine liver PGF synthase is distinct from the 4-benzalpyridine reductase. The purified enzyme, in both its substrate specificity and its sensitivity to inhibitors, was also clearly distinguishable from the carbonyl reductases purified from human (22), monkey (24), dog (25), and gerbil (26) liver. The liver PGF synthase was similar to the lung PGF synthase in terms of being a dual function enzyme and a kind of carbonyl reductase. However, in addition to the difference between K,,, values for PGD:! of liver and lung enzymes, the effect of chloride ion on bovine liver PGF synthase was different from that of bovine lung PGF synthase (Fig. 7). The effect of chloride ion on aldo-keto reductase, including carbonyl reductase, has not been examined. We found that sodium chloride activated the PGDz 11-ketoreductase, and phenanthrenequinone and nitrobenzaldehyde reductase activities of liver PGF synthase (Fig. 8) and that it had no effect on the PGD2 llketoreductase activity of lung PGF synthase (Fig. 7). In 1987, Steir et al. (9) found that Sa,ll/?-PGF2 increased sodium and water excretion in uivo in rats. Chloride ions may relate to the expression of PGDz 11-ketoreductase activity in liver, and the activation of PGD2 ll-ketoreductase activity of liver PGF synthase by chloride ions may be correlated with natriuresis. The liver and lung PGF synthases were inhibited by several metal compounds, but the liver enzyme was two orders of magnitude more resistant than the lung enzyme to inhibition by CuS04 and HgClz (Table IV). According to the immunochemical properties (16), kinetic characteristics, effects of chloride ion, and the inhibition by metal compounds, these two bovine tissue enzymes are clearly different. Although there was partial cross-reactivity and substantial homology between the mRNAs for the two enzymes (16), differences in primary structure between the two bovine tissue enzymes undoubtedly account for the above differences. Our comparative data thus suggest that bovine liver PGF synthase is an isozyme of bovine lung PGF synthase and, as such, may be dis-

OF BOVINE

23

LIVER

similar in physiological the same species.

role to its counterpart

in lungs of

ACKNOWLEDGMENTS We thank Dr. S. Ito for useful discussions, Drs. L. D. Frye and M. Connolly for critical reading of this manuscript, and M. Ueta for assistance with drawing the figures. This investigation was supported in part by research grants from the Ministry of Education, Science, and Culture of Japan, the Japan Foundation for Applied Enzymology, and the Japan International Science & Technology Exchange Center. Lan-Ying Chen is a recipient of a Science and Technology Agency, Japan fellowship.

REFERENCES 1. Watanabe, K., Shimizu, T., and Hayaishi, 0. (1981) Biochem. Int. 2,603-610. 2. Wong, P. Y.-K. (1981) Biochim. Biophys. Acta 669,169-178. 3. Liston, T. E., and Roberts, L. J., II (1985) J. Biol. Chem. 260, 13,173-13,180. 4. Liston, T. E., and Roberts, L. J., II (1985) Proc. Natl. Acad. Sci. USA 82,6030-6034. 5. Seibert, K., Sheller, J. R., and Roberts, L. J., II (1987) Proc. Nat{. Acad. Sci. USA 84,256-260. 6. Pugliese, G., Spokas, E. G., Marcinkiewicz, E., and Wong, P. Y.-K. (1985) J. Biol. Chem. 260, 14,621-14,625. 7. Hayashi, H., Ito, S., Watanabe, K., Negishi, M., Shintani, T., and Hayaishi, 0. (1987) Biochim. Biophys. Acta 917,356-364. 8. Beasley, C. R. W., Robinson, C., Feather&one, R. L., Varley, J. G., Hardy, C. C., Church, M. K., and Holgate, S. T. (1987) J. Clin. Inuest. 79,978-983. 9. Stier, J. R. C. T., Roberts, L. J., II, and Wong, P. Y.-K. (1987) J. Pharmacol. Exp. Ther. 243,487-491. 10. Robertson, R. M., Liston, T., Tantengco, M. V., and Roberts, L. J., II (1985) Clin. Res. 33, 221A. 11. Roberts, L. J., II, and Liston, T. E. (1985) Clin. Res. 33, 162A. 12. Watanabe, K., Yosbida, R., Shimizu, T., and Hayaishi, 0. (1985) J. Biol. Chem. 260,7035-7041. 13. Watanabe, K., Iguchi, Y., Iguchi, S., Arai, Y., Hayaishi, O., and Roberts, L. J., II (1986) Proc. Natl. Acad. Sci. USA 83, 158331587. 14. Watanabe, K., Fujii, Y., Nakayama, K., Ohkubo, H., Kuramitsu, S., Kagamiyama, H., Nakanishi, S., and Hayaishi, 0. (1985) Proc. Natl. Acad. Sci. USA 85, 11-15. 15. Hayashi, H., Fujii, Y., Watanabe, K., Urade, Y., and Hayaishi, 0. (1989) J. Biol. Chem. 264, 1036-1040. 16. Urade, Y., Watanabe, K., Eguchi, N., Fujii, Y., and Hayaishi, 0. (1990) J. Biol. Chem. 265, 12,029-12,035. 17. Yoshimoto, T., Yamamoto, S., Okuma, M., and Hayaishi, 0. (1977) J. Biol. Chem. 252,5871-5874. 18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 19. Laemmli, U. K. (1970) Nature 227,680-685. 20. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76,4350-4354. 21. Hsu, S. M., Rain, L., and Fanger, L. (1981) J. f&to&em. Cytochem. 29,577-580. 22. Nakayama, T., Hara, A., Yashiro, K., and Sawada, H. (1985) Biochem. Pharmacol. 34, 107-117. 23. Wermuth, B., Platts, K. L., Seidel, A., and Oesch, F. (1986) B&hem. Pharmacol. 35.1277-1282. 24. Sawada, T., Hara, A., Nakayama, T., Nakagawa, M., and Yashiro, K. (1984) Yakuguku Zasshi 104, 74-82.

24

CHEN,

WATANABE,

25. Hara, A., Nakayama, T., Deyashiki, Y., Kariya, K., and Sawada, H. (1986) Arch. Biochem. Biophys. 244, 238-247. 26. Molowa, P. T., Wrighton, S. A., and Guzelian, Biochem. Biophys. 26 1,487-494.

P. S. (1986) Arch.

27. Felsted, R. L., and Bachur, N. R. (1980) Drug Metab. Reu. 11, l-

60. 28. Sawada, H., Hara, A., Nakayama, T., and Kate, F. (1980) J. Biochem. 87.1153-1165. 29. Branlant, 611-621.

G., and Biellman,

J.-F. (1980) Eur. J. Biochem.

30. Wermuth,

B. (1981) J. Bid. Chem. 256, 1206-1213.

105,

564.

HAYAISHI

33. Ogino, N., Miyamoto, T., Yamomoto, S., and Hayaishi, 0. (1977) J. Biol. Chem. 252,890-895. 34. Shimizu, T., Yamamoto, S., and Hayaishi, 0. (1979) J. Biol. Chem.

264,5222-5228. 35. Urade, Y., Fujimoto,

N., and Hayaishi, 0. (1985) J. Biol. Chem. 260, 12,410-12,415. 36. Watanabe, K., Yamamoto, S., and Hayaishi, 0. (1979) B&hem. Biophys. Res. Commun. 87,192-199. 37. Nakayama, T., Yashiro, K., Inoue, Y., Matsuura, K., Ichikawa, H., Hara, A., and Sawada, H. (1986) Biochem. Biophys. Acta 882,220-

227. 38. Usui, S., Hara, A., Nakayama,

31. Wermuth, B. (1985) in Enzymology of Carbonyl Metabolism: Aldehyde Dehydrogenase, Aldo-Keto Reductase, and Alcohol Dehydrogenase, (Flynn, T. G., and Weiner, H., Eds.), Vol. 2, pp. 209230, A. R. Liss, New York. 32. Iwata, N., Inazu, N., and Satoh, T. (1989) J. Biochem.

AND

105, 556-

T., and Sawada, H. (1984) Biochem. J. 223,687-705. 39. Terada, T., Niwase, N., Shinagawa, K., Koyama, I., Hosomi, S., and Mizoguchi, T., (1989) in Enzymology and Molecular Biology of Carbony1 Metabolism: Aldehyde Dehydrogenase, Alcohol Dehydrogenase, and Aldo-Keto Reductase. (Weiner, H., and Flynn, T. G., Eds.), Vol. 2, pp. 293-305, A. R. Liss, New York.

PROSTAGLANDIN

F SYNTHASE MINIPRINT

Purification and Characterization of Prostaglandin F Synthase from Bovine Liver Lao-Ying Chen, Kikuko Watanabe, and Osamu Hayaishi MATERIALS AND METHODS Reagents [5,6,8,9,12,14,15-3H(N)] PGD2 (185 Ci/mmol) was obtained from Do Pant-New England Nuclear. [l-t“C] PGH2 was prepared as described previously (17), with acetone powder of sheep vesicular gland microsomes (Ram Biochemicals. Tel Aviv) used as a source of PG endooeroxide svnthase. PGF svnthase from bovine lung was purified by the method described’ previously (12). Authkntic PG’s were gener&s giits from.Ono Pharmaceutical Company (Osaka, Japan). Other materials and commercial sources were as follows: NADP (sodium salt), NADPH (tetrasodium salt, Type X), glucose-6-phosphate dehydrogenase from Baker’s yeast (Type XV), Trirma base, bovine serum albumin, Coomassie Brilliant Blue R-250. cromolvn. ethacrvnic acid, oiroxicam, warfarin, fenbofeo, florbiprofen, glutathione reduced form (GiH), D-gl;curonic acid (sodium salt), Cibacron blue 3Gk, and cyproheptadine, from Sigma; arachidonic acid, from P-L Biochemicals; phenylglyoxal monohydrate, from Aldrich Chemical Company, Inc.; Vectastain ABC kit, from Vector Laboratories. Inc. (USA); SDS-PAGE molecular weight standards, from Bio-Rad: DEAEcellulose, from Whatman; Red Sepharose CL-6B, from Pharmacia; thin-layer chromatography (TLC) plates precoated with silica gel 60 F254, from Merck: Millipore membrane, from Nihon Millipore Kogyo K. K. (Yonezawa, Japan); p-nitroacetophenone, DL-glyceraldehyde, nbutylaldehyde, rutin and Cosmosil 5Ct8 column, from Nacalai Tesque (Kyoto, Japan). Other chemicals were obtained from Wako Pure Chemical Industries. LTD (Osaka, Japan). Unless otherwise indicated, all chemicals were of analytical reagent grade. Enzyme PGD2 11.ketoreductase activity was measured at 37’C for 30 min as described previously (12). The standard assay mixture for bovine liver PGF synthase contained 0.15 mM [3H]PGD2 (0.10 pCi), 0.5 mM NADP. 5 mM glucose-6.phosphate. 1 unit of glucose6.ohosohate dehvdrosenase. 0.5 ue. of enzyme (about 20 % of substrate conversion) and 0.1 M p&ass&m phospiate\buffer, pH 6.3, in a t&al volume of 0.05 ml. The reaction was started by addition of enzvme. , and terminated bv addition of 0.25 ml of diethvl ether/methanol/O.2 M citric acid (30:4:1). PGD2 and PGF2 (8 erg each) were added to the solution as authentic markers. After the organic phase (50 ~1) had been subjected to TLC and the material scraped off in sections corresoondine to PGR. PGD? and others. the radioactivitv of each section was measured with a Packard 2200 CA Liquid Scintillation Analyzer. An assay mixture containing all the components except the enzyme served as a blank. The specific activity of the enzyme was expressed as the number of nmoles of 9~. 1lp-PGF2 formed/min/mg of protein at 37°C. The PGHz 9. 1I-endoperoxide reductase (PGH2 reductase) activity was determined as described previously (12), except for the addition of an enzyme amount of about 45 pg, incubation for 2 min. and the separation of product by boric acid-impregnated TLC (13). Carbonyl reductase actiiity was d&mined spectropho&&ically as the rate of oxidation of NADPH at 37°C (12). The reaction mixture in a 0.5.ml system contained 0.1 M potassium phosphate, pH 6.5, 80 pM NADPH, an appropriate amount of the enzyme, and substrates at various concentrations, as indicated in Tables 2-5. The reactions were initiated by the addition of substrate. and the decrease in absorbance at 340 nm was monitored. Control cuvettes contained all reagents except enzyme. The specific activity of the enzyme was expressed as the number of pmoles of NADPH oxidizedlminlmg of protein at 37°C. Protein concentration was measured by the method of Lowry er al.. with bovine serum albumin as a standard (18). For routine column monitoring, protein concentration was estimated by the “Ct,S”E”eltt Of &So. ran Products The products formed from PGD2 and PGH2 by PGF synthase from bovine liver were identified according to the method of Watanabe et al. (13). ‘Ihe reaction mixture for PGDz contained 0.1 M potassium phosphate buffer (pH 6.5), 15 pM [3H]PGD2 (1.5 pCi), 0.5 mM NADPH, 5 mM glucosed-phosphate,Blucosed-phosphate dehydrogenase (3 units), and enzyme (1.5 pg) in a total volume of 0.15 ml. Incubation was carried ottt at 37’C for 1 hr. The reaction was terminated by the addition of 0.8 ml a mixture of diethyl ether/methanol/O.2 M citric acid (30:4:1). The organic phase (0.15 ml) was subjected to TLC in a solvent system of diethyl ether/methanol/acetic acid (90:2:0.1). l?x major product with an Rf value of 0.06 was extracted from a silica gel plate with ethyl acetate/acetic acid (1oO:l). The extracts were evaporated and dissolved in 100 pl of 100 % ethanol. The sample was diluted IW times in the mobile phase (acetonitrilelwaterlacetic acid (30:70:0.1)) used for RP-HPLC. A 200.pl sample (20,000 dpm) was applied at a flow rate of I ml/min to an RPHPLC (LKB) apparatus equipped with a Cosmosil 5Ct8 column (4.6 x 150 mm). The eluate was fractionated at intervals of 30 s for 30 min after injection of the sample. The radioactivity of each fraction was measured with a liquid scintillation counter. The reaction mixture for PGH2 contained the same components as that for PGD2 except it had 10 pM [t4C] PGH2 (0.09 pCi) in place of the [3H]PGD2, 0.171 mg of enzyme, the absence or presence of phenanthrenequinone, and the incubation time of 2 min. The product formed from PGH2 (6$&l dpm) was applied to RP-HPLC. w Gel Electroohoresis SDS-PAGE was carried out by the method described by Laemmli (19). Samples were pretreated at 100°C for 2 min in the presence of 1% SDS and 2.5% 2-mercaptoethanol and run on separation gels containing 0.1 % SDS and 12.5 W polyacrylamide at a constant current of 35 mA. Protein bands were stained with Coomassie briiliani blue R-250 or silver. The molecular weight of the enzyme was estimated by calibration of the SDS-PAGE gel with a Bio-Rad low molecular weight calibration kit. lmmunoblottine Western blots of samples separated by SDS-PAEE were electrophoretically transferred to a Millipore membrane by the method of Towbin (20). The membrane was immunostained with anti-PGF synthase IgG, biotinylated anti-rabbit IgG antibody and a Vectastain ABC kit by the method of Hsu er al. (21). Diaminobenzidine was used as a substrate for the peroxidase reaction. Preparation of antiserum against PGF synthase from bovine lung was described mwiouslv (12). Purification of’Bovine iiver i’GF Svnrhaq All purification procedures were performed at 4°C. Bovine liver (100 g) was minced and homogenized in 3 volumes of 10 mM potassium phosphate buffer, pH 7.0 (buffer A) with a Ultra-turrax homogenizer (Janke and Kuokel, Germany) five times at top speed for 30 s each time. The homogenate was centrifuged at 10,080 x g for 10 min. and the supematant was again centrifuged at 100,OM) x g for 90 min. The supematant was recovered by decantation. Crude extracts were subjected to ammonium sulfate fractionation between 50.75 % saturation. The precipitated forms were suspended in about 7 ml of IO mM ootassium ohosohate buffer. oH 6.8 (buffer B). and dialvzed aeainst two changes of 2 liters of duffer B. ?& dialyzed samplk (1.2 g’of protein) was applied toa DEAEcellulose column (0.7 x 17 cm) previously equilibrated with buffer B. The PGD2 ll.

L

OF BOVINE

25

LIVER

SUPPLEMENT

ketoreductase activity was &ted with the same buffer at a flow rate of 20 ml/h (Fig. 1). The enzyme was concentrated by PM-IO membrane, dialyzed against two changes of 3 liters of buffer A, and applied a Red Sepharose column (0.7 x 12 cm) previously equilibrated with buffer A. ‘Ihe column was washed with buffer A, followed by a linear gradient of O-l.0 M KC1 in buffer A (150 ml), and then with 1 M KC1 in buffer A (250 ml). The enzyme was eluted with buffer A containing I M KCI and 1 mM NADP (74 ml)(Fig. I). The enzyme was concentrated and dialyzed against buffer A. The purified enzyme was stored at -8O’C. For and DEAE-cellulose and Red Sepharose column chromatographies, PGD2 1 1-ketoreductase phenantbmnequinone and nitorobenzaldehyde reductase activities were monitored. ‘Ihis process is simpler and more rapid than those methods previously used to purify other carbonyl reductases from liver (22-26).

Table I Purification of PGD2 Il.ketoreductase activity from bovine liver The PG& 11.ketoreductaseacrivity ws determinedas described under “Materials and Methods” step

Total protein mg

Total activity

nmollmin

20585 2406

I. crude extract

2. Ammonium sulfate 3. DFAE-cellulose

5810 2583

425 10.3

4. Red Sepharose

Spccifx activity nmoVmin!mg of protein

Yield 96

0.28 1.07

1837 985

loo 44

4.32 95.1

32 17

-

Table 3 Cofactor requirement for various substrates of twine

liver PGF syntbarc

The reactionmixturecontainedO.tM potassiumphosphatebuffer(pH6.5).enzyme.0.56 mM NADH or NADPHfor deteonining activities,andsubsaateat variouscommmtionr w indicati in a totatvolumeofO.5ml. Thenacttcmwasinidatedby theadditionof substrate. andtheenzymeactivitywasmeasured rpctmptxdomenica.tly at 37T as dwrtbed under “Materials and Methods”. Substrate

A&it-j

Gmcen-

Km

oation NADH NADPH NADH/NADPH NADH NADPH rnM PmO”Nlhllu’ w MM

NADWNADPH

p-Nitmbcnzakkhyde 9.l0-Phenambnnequinorle Pixnylglyoxal

0.5 7:;’

0.561 0.956 0.608

0.247 0.496 0.353

2.27 1.93 1.72

4wo ,E

6.9 2.9 6.7

580 276 I49

p-Nirmacctophenone Hydrindantin MCdiOil~ SD-Dihycbutestosrerone PGD2 a Nor detected

0.1 0.5 0.25 0.05 0.15

0.109 0.307 0.163 NW 0.003

0.255 0.272 0.233 0.140 0.099

0.40 1.20 0.70 oY3

2% IWO

5.7 5.4 6.7 8.0 3.3

::I I49

Table 5 Susceptibibty of bovine liver PGF synthase to various inhibnors Phenantbrenequincmc

Niwbenzaklehydc

CHEN, WATANABE,

26

&.I.

Elutiortw

pfprotein

and three different

reducrascsbyaBrpscDharoscm. Theenzyme purified from DEAE-cellulose was applied to a Red &.pharose column (0.7 x 12 cm). Protein concentration and NADPelution (No.s 51-67) ( . ) were monitored at 280 not. PGDz 11-ketoreductase (0). and phenanttuencquinone (0) and niuobenzaldehyde (m) reductase activities were determined as described under “Materials and Methods”.

AND HAYAISHI

&. 4. Deoendency de&,1 ffiD?onenzvmeamount(O)~~Qf~

1 B-PGF? svnthesis

from

!2yJhe~waImfnr~ti~)(b)s~ G!aKsspf%.l1~m2w d &endency ef 9a. 11 O-PCiFz svntheris from mkm & (Q. 0.1 M acetate buffer ( 0 ); 0.1 M Mes- aOH buffer (a); 0.1 M potaswm phosphate buffer (0).

ICI 1 IW Eie.e.i2Kemti~Q.,,mcr),~~) and CaCl2 U) QJI &t,!& [jH]PGDz; and-the middle one (B) and lower one (C) indicate the products from I*%T’GHz in the absence and pnsenc~ of phenanthr&eq&zmt~ respectively. Preoaration was made as described under “Materials a&Methods”.

1 I-ketorcductase

a),

ohenanthreneouinone-(B)d nmobenzal&&reductasc

fQ &y&j,&&&&

liver PGF&. Enzyme assays were carried out as described under “Materials and Methods” in the presence of increasing concentrations of chlorides.

Purification and characterization of prostaglandin F synthase from bovine liver.

Prostaglandin D2 11-ketoreductase activity of bovine liver was purified 340-fold to apparent homogeneity. The purified enzyme was a monomeric protein ...
1MB Sizes 0 Downloads 0 Views