- Email: [email protected]
Prostaglandin-E2 9-ketoreductase from swine kidney. Production of antisera and application to development of a radioimmunoassay
Biochimica et Biophysica Acta,
1001 (1989) 9-15
9
Elsevier BBA 53006
Prostaglandin-E 2 9-ketoreductase from swine kidney. Production of anfisera and application to development of a radioimmunoassay S u s a r m e I(a'~ger, U l r i c h M e y e r , D i e t e r D a n i e l s a n d W e r n e r S c h | e g e l Universitiits- Frauenklinik, M~nster (E R. G.)
(Received2 August 1988)
Key words: Prostaglandin-E2 9-ketoreductase;Enzymepurification;Affinitypurifiedantibody; Radioimmunoassay;(Swinekidney) Prostaglandln.E 2 9.ketoreduc|ase 0~GE2-9-KR, EC 1.1.1.189), the enzyme which catalyzes the reaction from prostaglandin E z (PGE 2) to prostaglandiu F'~ (PGFz.), was purified 5804old from swine kidney. The molecular mass of the enzyme determined by SDS-gel electrnphoresls was 33 kDa. Antiserum against the purified enzyme was raised in three rabbits. The antiserum was able to precipitate PGE2-9-KR from swine kidney and to crossreact with PGE2-9-gR from several reproductive organ tissues, such as rabbit ovary, rabbit corpus httemn, rabbit endometrhma and human decldua vera. When swine kidney PGE2-9-KR was labelled with 12sl and incubated with affinity-purified antiserum in the presence of increasing amounts of uniabeiled enzyme, competitive binding of the uulabeiled enzyme to the antih~y was observed. A rad|oimmunoassay for the quantitatiun of the enzyme was develuped. The standard cur~e wa~ linear from 5 to 500 ng enzyme. The intra- and interassay coefficients of variation were 6.4 and 13.2%, respectively. T~e assay may be useful for the quantitatiun of PGE2-9-KR in several tissues under various physiological conditions.
Introduction
Prostaglandin-E 2 9-ketoreductase (PGE2-9-KR), the enzyme which catalyzes the reaction from prostaglandin E 2 (PGE2) to prostaglandin F2, (PGF2,0, has been identified and characterized in many mammalian species including chicken heart [1], swine and rabbit kidney [2-7], human erythrocytes and human placenta [8,9]. Recently, we reported the occurrence of the enzyme in human decidua vera [10,11] and in rabbit corpus luteum and ovary [12]. We suggested that, based on its PGF2~ production, the decidual enzyme might regulate the PGE2/F2, , ratio and therefore might be involved in the events leading to labour and delivery. The hiteal enzyme was believed to possibly play a role in the functional breakdown of the corpus hiteum. Recent findings showing that in the rabbit corpus luteum, PGE2-9-KR activ-
ity was significantly elevated during the initiation of luteolysis support our hypothesis [13]. In order to elucidate its physiological role in ovarian and luteal tissue, we tried to purify and to characterize the enzyme. We had to realize, however, that reproductive tissues would not provide enough material to isolate the large quantities of enzyme needed for in-vitro studies. Therefore, we decided to generate polyclonal antibodies against the enzyme which would allow the identification and quantitation of PGEz-9-KR in several tissues, even in low concentrations. As a source for enzyme purification, we used swine kidneys because they contain PGEz-9-KR in high concentrations [3] and are readily available in large quantities. In the following article, we describe the purification of swine kidney PGE2-9-KR to homogeneity, the characterization of antibodies, their cross-reactivity to PGE2-9-KR of ovarian, luteal and decidual tissue and a radioimmunoassay for quantitating the enzyme. Materials and Methods
Abbreviations: PGE2, prostaglandin E2; PGF2a, prostaglandin F2.; PGE2.9.KR, prostaglandin-E2 9.ketoreductase;SDS, sodiumdodeeyl sulfate; PAGE, polyacrylamidegel eleetrophoresis; FPLC, fast protein liquid chromatography;c.v., coefficientof variation. Correspondence: W. Schlegel, Universit~its-Frauenklinik, AlbertSehweitzer-Strasse33, 4400 Miinster,F.R.G.
Materials. [5,6,8,11,12,14,15-3H]Prostag landin E2 (6.3 TBq/mmol) and [5,6,8,11,12,14,15- 3H]prostaglandin F2~ (6.7 TBq/mmol) were purchased from Amersham Buchler (Braunschweig, F.R.G.), unlabelled prostaglandins, goat anti-rabbit IgG, anti-rabbit lgG-alk-
0005-2760/89/$03.50 © 1989 ElsevierSciencePublishers B.V. (BiomedicalDivision)
10 aline phosphatase conjugate, 5-bromo-4-chloro-3-indolyl phosphatase toluidine salt and p-Nitro blue tetrazolium chloride were obtained from Sigma (Miinchen, F.R.G.). NADPH was obtained from Boehringer (Mannheim, F.R.G.). DEAE-Trisacryl M, ultrogel AcA 44 and nitrocellulose membrane were obtained from LKil (lkomma, Sweden). Mono Q, phenyl-Superose, Superose 12 FPLC columns, CNBr-Sepharose 4B and Sephadex G.10 were obtained from Pharmacia (Up~ psala, Sweden). Ouchterlony double.immunodiffusion discs were obtained from ICN lmmunoBiochemicals (New York, U.S.A.), Enzyme purification. Swine kidneys (1000 g) were homogenized in 10 mM Tris/HCl buffer (pH 7.4)/1 mM dithiothreitol/0.02$ sodium azide/10$ (v/v) glycerol (buffer A) in a Waring Blendor for 2 rain (tissue to buffer ratio. 1:4). All purification steps were carried out at 0-4 e C. The homogenate was centrifuged at 20000 × g for 15 rain, The resulting supematant was adjusted to pH 5.2 with 1 M acetic acid as described by Moldave [14]. After 10 rain, the solution was centrifuged at 20000 × g for 15 rain in order to remove the precipitated protein. The supernatant was readjusted to pH 7.4 with 1 M NaOH and concentrated by 80% ammonium sulfate precipitation. The precipitate was redissolved and dialyzed against buffer A and applied to a 50 × 5 cm column of DEAE-Trisacryl M which had been equilibrated with buffer A. The effluent fractions were pooled and concentrated by 80~$ ammonium sulfate precipitation. The precipitate was dissolved in 10 mM Tris/HCI buffer (pH 7.4) containing 1 mM dithiothreitol, 0.02~$ sodium azide, 105 (v/v) glycerol and 0,1 M NaCI (buffer B) and applied to a 90 × 2,6 ¢m column of ultrogel AcA 44, Active fractions were pooled and concentrated by 805 ammonium sulfate precipitation. The precipitate was dissolved and dialyzed against 20 mM Tris/HC! buffer (pH 8,0)/1 mM dithiothxeitol/0,02~$ sodium azide/10$ (v/v) glycerol (buffer C) and applied to a $ × 0.5 cm FPLC Mono Q column ~qui. librated with buffer C, 20 m8 protein was injected into the column during each run. The column was washed with 5 ml of starting buffer, followed by a linear gradient of buffer C containing 0-0.15 M NaCI. Active fractions were pooled, concentrated by 80~$ ammonium sulfate precipitation and dialyzed against 100 raM potassium phosphate buffer (pH 7,4)/1 mM dithiothreitol/0,02~$ sodium azide/10$ (v/v) glycerol/1.2 M ammonium sulfate (buffer D), 9 mg protein was injected into the column during each run, The enzyme solution was applied to a 5 × 0.5 cm FPLC phenyl-Superose column equilibrated with buffer D. The column was washed with 5 ml of starting-buffer and eluted with a linear gradient of buffer D containing 1.2-0 M ammonium sulfate. Active fractions were pooled, concentrated by 80~ ammonium sulfate precipitation and dialyzed against buffer B. The enzyme solution was
applied to a 30 × 1 cm FPLC Superose 12 column equilibrated with buffer B. 0.5 mg protein was injected into the column during each run. Active fractions were pooled, dialyzed against buffer A and stored in aliquots at - 2 0 ° C . Affinity-purified antibodies. Antiserum was raised in female New Zealand white rabbits by injecting 50/tg of purified swine kidney PGEz-9-KR in complete Freund's adjuvant. The injection was repeated after 2 weeks and then every 2 weeks, but now with incomplete Freund's adjuvant. 9 days after the second injection, blood was taken weekly. The titer of antisera was followed by Ouchterlony double diffusion and by binding experiments to PGE2-9-KR previously labelled with 12Sl (see below). Affinity-purified PGE2-9-KR antibodies were obtained by chromatography on a PGEz-9-KR-Sepharose column (100/~g enzyme/ml gel). The gel (CNBr-Sepharose 4B) was prepared according to the manufactorer's advice [15] using CNBr-activated Sepharose 4B and 100 /~g of the purified enzyme. Affinity chromatography was performed by diluting 200/d of antiserum 5-fold in 0.1 M phosphate buffer (pH 7,0) and applying the mixture to the sepharose column. The column was washed with 10 ml of starting buffer until the absorption at 280 nm was zero. Ehtion was carried out using 0.1 M glycine/HCl buffer (pH 2,5). The purified antibodies were immediately dialyzed against phosphatebuffered saline (pH 7.4) and kept at 4°C.
Ouchterlony double diffusion and IZSl labelling of PGEz-9-KR. Ouchterlony double diffusion was performed in round (5 ram) agar gels (1.5~$ agar in phosphate-buffered saline). The central cavity contained 15 /d (15 Fg) of the purified enzyme and the peripheral wells contained 15 FI of the antiserum. The gels were incubated for 3 days at 4°C in a humid atmosphere until precipitation bands occurred. l~I labelling of purified PGE2-9-KR was performed a~ording to ReL 16 using the ehloramin T method (10 Fg of enzyme), t2SI-labelled protein was separated from free iodine by Sephadex G-10 chromatography and kept at 4°C,
Binding experiments and radioimmunoassay of PGEz. 9-KR. 10/,1 12Sl-labelled PGE2-9-KR (20000 cpm, 3.3 kBq) in 50 mM Tris/HCl buffer (pH 7.4)/0.1% gelatin was incubated with 100 Iti of antiserum (dilution 1:2000) or 50 ng of affinity-purified antibody in the presence or absence of unlabelled enzyme in a total volume of 200 ~ti. The mixture was incubated overnight at 4°C. Then 20/tl of goat anti-rabbit IgG (dilution 1 : 16) were added and the mixture was incubated overnight at 4°C. After the addition of 500 ttl of assay buffer, the incubation mixture was centrifuged at 10000 × g for 10 min. The supernatant was discarded and the bound radioactivity in the pellet was determined in a y-counter.
11
SDS-PAGE and immunoblotting. Discontinuous SDS-PAGE was performed according to Laemmli [17] (13 and 5% polyacrylamide, respectively). For immunoblotting, proteins were transferred electrophoretitally (100 mA, 1 h) from the gel to a nitrocellulose membrane. After blocking of protein binding sites with Tweea-phosphate-buffered saline containing 3% bovine serum albumin, the membrane was incubated first with the PGE2-9-KR antiserum (1:1000 diluted) for 2 h at room temperature and then for 2 h with goat anti-rabbit IgG-alkaline phosphatase conjugate (1:1000 diluted). The substrates used for staining were 5-bromo-4-chloro3-indolyl phosphate toluidine salt and p-Nitro blue tetrazolium chloride. Enzyme activity. PGE2-9-KR was measured as described previously [11] following the production of PGFzo by a radioimmunologicai method. The standard mixture contained 100 mM Tris/HCl buffer (pH 8.0) 100 pM NADPH, 71 /~M PGE 2 and ]0-100 ~1 of enzyme in a final volume of 1 ml. The incubation was carried out at 37°C for 60 rain and terminated by adjusting the pH to 3.5 with 1.5 M citric acid. The prostaglandins were extracted with 8 ml ethyl acetate and dried under a stream of nitrogen gas. The extracts were redissolved and chomatographed on silicic acid columns as described in Ref. 18. The PGF2ot fractions were analyzed with a specific radioimmunoassay [19]. One unit of enzyme activity is defined as the amount of
enzyme which catalyzes the production of 1 pmol PGF2,,/nfin under standard assay conditions. Other methods. Protein was determined according to the procedure of Bradford [20] using bovine serum albumin as standard. Staining of po~yacrylamide gels with silver was done according to Merril et al. [21]. Results PGE2-9-KR was purified from swine kidney, according to a procedure similar to that used for the decidual enzyme [11], but now FPLC chromatography was additionally used to improve purification. The ehition profile of the FPLC Mono Q step is shown in Fig 1. The isolation of PGE2-9-KR resulted in a 580-fold increase of its specific activity (Table I). SDS-polyacrylamide gel electrophoresis of the purified enzyme revealed a single protein band with an apparent molecular mass of 33 kDa (Fig. 2). The purified PGE2-9-KR was injected into three rabbits. The rabbits produced antisera which yielded a single immunoprecipitation line in Ouchterlony doublediffusion experiments when tested with th~ isolated enzyme (data not shown). When an excess amount of purified PGE2-9-KR was incubated with increasing amounts of antiserum and after 2 h with an excess of goat anti-rabbit IgG, up to 80% of the enzyme activity added was immunoprecipitated (data not shown). When the PGE2-9-KR separated by SDS-electrophoresis was
~ea-9-Km E280 {--)
( t J.cuC ) 0.20
NaCI {M)
4.0 ( - - - )
....... 2~
0,8 2(
0.15
0.6 15 0.t0 0.4 1@
i O.
. 0
.
~ ,
. . . . . . . . . . . . . . . . $ 10 IS T|mo
+
0.2
0 20
25
(rain)
Fig. l. FPLC Mono Q column chromatography of PGE2-9-KR. Pooled and dialyzed PGE2-9-KR was applied to a 5 ×0.5 cm FPLC Mono Q column that had been equilibrated with 20 raM T n s / H C I buffer ~pH 8.0)/1 mM dithiothreitol/0.02~ sodium azide/10~ ( v / v ) glycerol (buffer C). 20 nag protein was rejected into the column during each run. The column w~s washed with 5 ml of starting buffer and then eluted with 20 ml of a 0-0.15 M NaCi linear gradient at a flow rate of I ml/min. Fractions of 1 ml were collected. The aa:yme activity in each fraction was determined as described in Materials and Methods. - - - , A2BO;® @, PGE2-9-KR activity; - - - - - - , NaCI gradient.
12 TABLE !
Purification of PGE~-9.KRfrom swine kidney The enzyme activity was determined as described in Materials and Methods by measuring the production of PGF,=. One unit is defined as the amount of enzyme which catalyzes the production of 1 pmol PGF2= per min under standard assay conditions. Fraction
Volume (ml)
Protein (mR)
Activity (U)
20000 x g supematant
450
13486
61110
4.4
1
100
pH 5.2 supematant
380
6688
58786
8.8
2
96
FAuateof DEAE-Trbacryl M
700
45.~0
48650
10.7
2.4
79
Eluateof ultrogelAcA44
280
336
31164
92.8
21
51
Wuateof FPLC.MonoQ
S)
55
20775
377,7
86
34
Elutt~of FPLC.phenyI.Supetme
12
6
9776
1629.3
370
16
Eluateof FPLC.Superme 12
24
1.2
3055
2 545.8
580
5
transferred onto a nitrocellulose membrane which subsequently was incubated first with the antiserum and then with an alkaline phosphate labelled anti-rabbit IgG antibody, the enzyme could be detected immunochemically as a single band at 33 kDa (Fig. 2). The purified enzyme was labelled with lUl by the ¢hloramine T method in order to carry out binding experiments. The titration curve, where a constant amount of labelled enzyme was incubated with increasing dilutions of antiserum, demonstrated that 20~ of the total radioactivity was bound at a final antiserum dilution of 1:2000 (Fig. 3). When increasing amounts of unlabelled PGEa.9.KR were added to the incubation mixture, competitive binding of the unlabelled enzyme to the antibody was observed.
Specific activity (U/mR)
Enrichment
Yield (~)
The antibodies were purified using PGE2-9-KR CNBr-sepharose affinity chromatography. The tissue specificity of PGE2-9-KR antibody is shown in Fig. 4. hal-labelled PGE2-9-KR from swine kidney was incubated with affinity-purified antibody and increasing amounts of unlabelled enzyme protein from rabbit ovary, rabbit corpus luteum, rabbit endometrium and human decidua vera. The logit-log analysis yielded parallel lines for the different enzymes, indicating that the antibody was able to crossreact with PGE2-9-KR from the different sources. Rabbit corpus luteum and endometrium enzymes were 10000×g supernatants, rabbit ovary PGE2-9-KR had been partly purified (25fold, [12]), and PGE2-9-KR of human decidua vera had been purified to apparent homogeneity (232-fold [11]).
94 e7 48
14
1
23
56
Fig. 2, SDS-polyacrylamide gel eleetrophoi~esispatterns and immunoblot of swine kidney PGE2-9.KR. SDS-electrophoresis and immunobiot were carried out as described in Materials and Met,Nods, The protein was stained with silver. Amounts of protein analyzed were 0.2-2/tg per sample. From left to fight are shown the individual purification steps (2-6): (1) standards; (2) 20000× g supematants; (3) DEAE-Trisacryl M; (4) uitrogel AcA 44; (~) FPLC-Mono Q; (6) FPLC-Superose 12; (7) immunoblot of PGE2-9-KR.
13 so%B/T
in I
Fig. 3. Titration curve of PGE2-9-KRantiserum. 20000 cpm (3.3 kBq) of 12Si-labelled swine kidney PGE2-9-KR were incubated with antiserum of increasing dilution overnight at 4°C. 20 ~l of goat anti-rabbit [gG (dilution 1:16) were added and the mixture was incubated overnight at 4 o C. The incubation mixture was centrifuged at 10000× g for 10 rain, the supernatant was discarded and the bound radioactivity was determined in a y-counter. The x-axis represents the antiserum dilution and the y-~ds the ratio of bound radioactivity(B) versus total radioactivity (T). Each point represents the mean of three determinations + S.D.
The fact that binding of t25I P G E 2 - 9 - K R was competitively inhibited b y unlabelled PGE2-9-ketoreductase from swine kidney and from several reproductive tissues suggested the development of a radioimmunoassay for quantitation of the enzyme protein. The principle of the assay is that binding of radiolabelled enzyme to the
Io91t BlaO
o
°11 -4 ~
0,1
I
[ I IIIII
I
li
I
i
itllll
I
I
I IIllpt
10 I00 ProWIn (pg)
t
i
i
I
t
I
G
Dilution of Antlurum
iltttl
t
~)00
i
i
i ill
10000
Fig. 4. Tissuespecific of PGE2-9-KR antibody. 20000 cpm (3.3 kBq) of t251-1abelled swine kidney PGE2-9-KR were incubated with 50 ng of affinity.purified antibody in the p~esence of increasing 9mounts of PGE2-9-KRs from human decidua vera, rabbit ovary, rabbit corpus loteum and rabbit endometrinm overnight at 4°C. 20 pi of goat anti-rabbit [gG (dilution 1:16) were added and the mixture was incubated overnight at 4 ° C. The incubation mixture was centrifuged at 100ta3×g for 10 rain, the supernatant was discarded and the bound radioactivity was determined in a y-counter. The x-axis represents the log concentrations of added unlabelled enzyme and the y-axis the legit-log transformation of bound radioactivity (B) versus the value of B in the absence of unlabelled enzyme (Be). Each point represents the mean of three determinations+S.D. Enzymes: m m, human decidua vera PGE2-9oKR, purified to apparent homogeneity (232-fold) [Ill; * *, rabbit ovary PGE2-9-KR, partly purified (25-fold) [12]; A ~ A , rabbit corpus luteum PGE 29-KR, 10000xg superuatant; + - - - - + , rabbit endometrium PGE2-9-KR, 10000 × g supernatant.
i
i
ii
I
I
r
6O PGEe-O-KR 1~1
I
I
!
!
11
eOQ
Fig. 5. Standard curve of the radioimmunoassay for determining PGE2.9-KR concentration (legit-log transformation). 20000 cpm (3.3 kBq) of 12sI-labd|ed swine kidney PGEa-9-KR were incubated with 50 ng of affinity-purified antibody in the presence of increasing amounts of purified swine kidney PGE2-9-KR of known concentrations overnight at 4 ° C. 20 ~1 of goat anti-rabbit lgG (dilution 1:16) were added and the mixture was incubated overnight at 4°C. The incubation mixture was centrifuged at 10000×g for 10 m/n, the supernatant was discarded and the bound radioactivity in the pellet was determined in a 7-counter. The x-axis represents the log concentrations of added unlabelled PGEz-9-KR and the y-axis the log,it. log transformation of bound radioactivity (B) versus the value of B in the absence of unlabeUed enzyme (BO). Each point represents the mean of three determinations :t:S.D.
antibody is inversely proportional to the amount of unlabelled enzyme present, if the antibody concentration used is lower than that of the labelled enzyme. A constant amount of 125I-labelled swine kidney PGE2-9 K R was h, cubated with a constant a m o u n t of affinitypurified antibody in the presence of increasing amounts of purified swine kidney P G E e - 9 - K R of known concentration. As shown in Fig. 5, the l o g / t r o t transformation of the standard curve with unlabeled swine kidney enzyme was linear from 5 - 5 0 0 ng enzyme. Sensitivity is defined as the smallest a m o u n t of unl~belled enzyme protein causing 10% displacement compared with B e (value of B in the absence of unlabelled ligand). The calculated limit of detection was 5 n g ± 0% (mean :t: c.v.%) for swine kidney PGE2-9-KR. The upper detection limit was 600 ng protein. The precision of the assay was measured by adding a distinct a m o u n t of purified swine kidney P G E 2 - 9 - K R to the assay mixture. The intra- and interassay variation of n - - 1 0 assays were determined. The intraassay coefficient of variation was 6.4% aLd the interassay coefficient of variation was 13.2%, respectively. Discussion The purification of PGE2-9-KR from swine kidney resulted in a 580-fold increase in its specific activity, which is lfigher than the results obtained b y Chang et al. [4], who achieved a l l 5 - f o l d enrichment of the enzyme. The molecular mass of 33 kDa for the enzyme as
14
determined by SDS-gel electrophoresis fits well with the results obtained by other authors (29.5 kDa [22]) and with the molecular mass of PGE2-9-KR in other tissues. Tort and Hansen [23] found a molecular mass of 31 kDa for the enzyme iv .'abbit kidney, Lin and Jarabak [9] determined a molecular mass of 31-33 kDa for PGE2-9-KR in human placenta, and we found that the human decidual enzyme had a molecular mass of 31 kDa [11]. The antisera raised in rabbits by the injection of purified enzyme proved to react specifically with swine kidney PGE2-9-KR, as seen by a number of experimental results. In Ouchtedony diffusion experiments, the antisera yielded a single immunoprecipitation line when tested with PGE2-9-KR samples. The antisera were also able to precipitate the enzyme activity. Finally, immunoblotting of PGE2-9-KR revealed a single band which migrated to the same location as the purified enzyme. Production of antisera against swine kidney PGE2-9-KR has also been reported by Chang and Tai [22], These antisera, however, were derived from a partly purified enzyme (llS-fold enrichment). Furthermore, they were neither characterized nor used for further applications. Immunoassays for the quantitation of prostaglandinsynthesizing enzymes have been described by several authors, as for PGH synthase [24,25], PGI synthase [26] and thromboxane synthase [27], A method for the quantitation of PGE2-9-KR, however, has not been described up to now. The radioimmunoassay developed by us is a simple, specific assay and may be used for direct quantitation of immunoreactive PGE2-9-KR in swine kidney and several other tissues, as the antibody crossreacts with PGE2.9.KR from rabbit ovary, rabbit corpus luteum, rabbit endometrium and human decidua ,,'era, The assay is able to detect at least $ ng of the enzyme and is linear up to 500 n8 under the assay conditions described under Materials and Methods. The intra, and interassay coefficients of variation are below 15~, showing that the assay is quite reproducible and should be useful for the quantitation of enzyme levels under various physiological conditions. The availability of antibodies against PGEz-9.KR does not only provide a tool for quantitation of the enzyme, but also will enable us to purify the enzyme more rapidly using immunoaffinity chromatography. The use of antibodies against PGEz-9-KR will be helpful for the electron-micro.~opical localization of the enzyme in various tissues and for the investigation of the biosynthesis pathway of PGE2.9-KR using primary cell cultures, radiolabelling and immunoprecipitation methods. The physiological role of PGE2-9-KR is a subject of ~ontroversy. Weber et ai. [28] found a relation of the renal enzyme to the NaCI intake of the kidney, whereas Walkins and Jarabak [29] found that the enzyme obvi-
ously had no regulative function. Furthermore, the renal enzyme seems to have a broad substrate specifit.y for several aldehydes and ketones and Chang and Tai [30] suggested that the enzyme acts rather as general aldoketoreductase. The role of PGEa-9-KR in reproductive organs, however, seems to be more specialized. In tissues such as the ovary, endometrium and placenta, PGE 2 and PGF2+, often act in a different and sometimes contrary manner. At the end of human pregnancy, P G F . 2 c a u s e s ripening and dilatation of the cervix, while PGFaa causes u~erine contractions [31]. As PGEa-9-KR is supposed to regulate the PGE2/PGF2, ~ ratio, the enzyme may be involved in the events leading to labour and delivery [10]. In the corpus luteum of many species, P G E 2 s e e m s to act as a luteotrophic substance, while PGF2a is believed to act as luteolytic factor [32]. Recently, we found a possible correlation between PGE 29-KR activity and iuteolysis in the rabbit [13], indicating that this enzyme may be involved in this process by regulating the PGE2/PGF2, ratio. A quantitation and immunochemical detection of PGE2-9-KR in the rabbit corpus luteum at different stages of luteolysis and in human decidua at different stages of pregnancy, which will be possible with the help of our antibodies, will perhaps provide more insight into the physiological role of this enzyme, especially in re,roductive processes. Acknowledgements The technical assistance of R. Maucher is gratefully acknowledged. TF.is work was supported by the Deutsche Forschungsgemeinschaft (Schl 140/5-2). References
I Lee,S.C. and Levine.L. (1975) J. Biol. Chem. 250. 4549. 2 Lee, S.C., Pong, S.S., Katzen, D., Wu, $J. and Levine, L. (1975) Biochemistry14, 142-145. 3 Tai, H.H. and Chang, D.G. (1982)MethodsEnzymol.86, 142-147. 4 Chang, D.G., Sun, M. Tai, H.H. (1981) Biochem. Biophys. Res. Commun.99, 745-751. 5 Toh, B.S. and Hansen, H.S. (1979) Biochem.Biophys. Acta 574, 33-38. 6 Toh, B.S.and Hanscn, H.S. (1980) Prostaglandins20, 735-746. 7 Ko~ff,J. and Jarabak, J. (1980) Prostaglandins20, 111-125. 8 Kaplan, L., Lee, S.-C. and Levine, L. (1975) Arch. Biochem. Biophys. 167, 287-293. 9 Lin, Y.-M. and Jarabak, J. (1978) Biochem.Biophys. Res. Cornmen. 81, 1227-1234. 10 Schlegel,W., Kriiger, S. and Korte, K. (1984) FEBS Lett. 171, 141-144. 11 Krllger,S. and Schlegel,W. (1986) Eur. J. Biochem.157, 481-485. 12 Schlegel, W., Daniels, D. and Krfiger, S. (1987) Clin. Physiol. Biochem.5, 336-342. 13 Schlegel, W., Kriiger, S., Daniels, D., Fischer, B., Schneider, H.P.G. and Beier,H.M. (1988) .L Reprod. Fe~il. 83, 365-370. 14 Moldave,K. (1963) MethodsEnzymol.6, 757-761. 15 Pharmacia(1987)AffinityChromatography,Principlesand Methods, pp. 15-18.
15 16 Greenwood, F.C., Hunter, W.M~ and Glover, J.S. (1963) Biochem° istry 89, 114. 17 Laernmfi, U.K. (1970) Nature 227, 680-685. 18 3affe, B..I. and Behrman, H.R. (1974) in Methods of Hormone Radioimmunoassay (Jaffe, B.M. and Behrman, H.R., eds.), pp. 19-34, Academic Press, New York. 19 Schlegel, W., Wenk, K., Dolfinger, H.C. and Raptis, S. (1977) Clin. Sci. Mol. Med. 52, 255-258. 20 l~r~dford, M. (1976) Anal. Biochem. 72, 248-254. 21 Merril, C.R., Goldmann, D., Scdmann, $.A. and Ebert, M.H. (1981) Science 211, 1437-1438. 22 Cimng, D.C. and Tai, H.-H. (1982) Arch. Biochen~. Biophys. 214, 464-474.
23 Tort, B.S. and Hansen, H.S. (1979) Biochim. Biophys. Acta 574, 33-38.
24 Ro~h, G.J. (1982) Methods Enzymol. 86, 222-228. 25 DeWitt, D.L., Day, J.S., Gauger, J.A. and Smith, W.L (1982) Methods EnzymoL 86, 229-240. 26 DeWitt, D.L. and Smith, W.L. (1982) Methods Enzymol. 86, 240-2~. 27 Shen, R.-F. and Tai, H.-H. (1986) J. Biol. Chem. 261, 11585-11591. 28 Weber, P.C., Larsson. C. and Scherer, B. (1977) Nature 2(}6, 65-66. 29 Watkins, J.D. and .larabak, J. (1985) Prostaglan~ns 30, 335-349. 30 Chang, D.G.-B. and Tai, H.-H. (1981) Biochem. Biophys. Res. Commun. 101,898-904. 31 Keirse, MJ.N.C. (1978) Adv. Prostaglandin Thromboxane Res. 4, 87-102. 32 Horton, E.W. and Poyser, N.L. (1976) Physiol. Rev. 56, 595-651.
1001 (1989) 9-15
9
Elsevier BBA 53006
Prostaglandin-E 2 9-ketoreductase from swine kidney. Production of anfisera and application to development of a radioimmunoassay S u s a r m e I(a'~ger, U l r i c h M e y e r , D i e t e r D a n i e l s a n d W e r n e r S c h | e g e l Universitiits- Frauenklinik, M~nster (E R. G.)
(Received2 August 1988)
Key words: Prostaglandin-E2 9-ketoreductase;Enzymepurification;Affinitypurifiedantibody; Radioimmunoassay;(Swinekidney) Prostaglandln.E 2 9.ketoreduc|ase 0~GE2-9-KR, EC 1.1.1.189), the enzyme which catalyzes the reaction from prostaglandin E z (PGE 2) to prostaglandiu F'~ (PGFz.), was purified 5804old from swine kidney. The molecular mass of the enzyme determined by SDS-gel electrnphoresls was 33 kDa. Antiserum against the purified enzyme was raised in three rabbits. The antiserum was able to precipitate PGE2-9-KR from swine kidney and to crossreact with PGE2-9-gR from several reproductive organ tissues, such as rabbit ovary, rabbit corpus httemn, rabbit endometrhma and human decldua vera. When swine kidney PGE2-9-KR was labelled with 12sl and incubated with affinity-purified antiserum in the presence of increasing amounts of uniabeiled enzyme, competitive binding of the uulabeiled enzyme to the antih~y was observed. A rad|oimmunoassay for the quantitatiun of the enzyme was develuped. The standard cur~e wa~ linear from 5 to 500 ng enzyme. The intra- and interassay coefficients of variation were 6.4 and 13.2%, respectively. T~e assay may be useful for the quantitatiun of PGE2-9-KR in several tissues under various physiological conditions.
Introduction
Prostaglandin-E 2 9-ketoreductase (PGE2-9-KR), the enzyme which catalyzes the reaction from prostaglandin E 2 (PGE2) to prostaglandin F2, (PGF2,0, has been identified and characterized in many mammalian species including chicken heart [1], swine and rabbit kidney [2-7], human erythrocytes and human placenta [8,9]. Recently, we reported the occurrence of the enzyme in human decidua vera [10,11] and in rabbit corpus luteum and ovary [12]. We suggested that, based on its PGF2~ production, the decidual enzyme might regulate the PGE2/F2, , ratio and therefore might be involved in the events leading to labour and delivery. The hiteal enzyme was believed to possibly play a role in the functional breakdown of the corpus hiteum. Recent findings showing that in the rabbit corpus luteum, PGE2-9-KR activ-
ity was significantly elevated during the initiation of luteolysis support our hypothesis [13]. In order to elucidate its physiological role in ovarian and luteal tissue, we tried to purify and to characterize the enzyme. We had to realize, however, that reproductive tissues would not provide enough material to isolate the large quantities of enzyme needed for in-vitro studies. Therefore, we decided to generate polyclonal antibodies against the enzyme which would allow the identification and quantitation of PGEz-9-KR in several tissues, even in low concentrations. As a source for enzyme purification, we used swine kidneys because they contain PGEz-9-KR in high concentrations [3] and are readily available in large quantities. In the following article, we describe the purification of swine kidney PGE2-9-KR to homogeneity, the characterization of antibodies, their cross-reactivity to PGE2-9-KR of ovarian, luteal and decidual tissue and a radioimmunoassay for quantitating the enzyme. Materials and Methods
Abbreviations: PGE2, prostaglandin E2; PGF2a, prostaglandin F2.; PGE2.9.KR, prostaglandin-E2 9.ketoreductase;SDS, sodiumdodeeyl sulfate; PAGE, polyacrylamidegel eleetrophoresis; FPLC, fast protein liquid chromatography;c.v., coefficientof variation. Correspondence: W. Schlegel, Universit~its-Frauenklinik, AlbertSehweitzer-Strasse33, 4400 Miinster,F.R.G.
Materials. [5,6,8,11,12,14,15-3H]Prostag landin E2 (6.3 TBq/mmol) and [5,6,8,11,12,14,15- 3H]prostaglandin F2~ (6.7 TBq/mmol) were purchased from Amersham Buchler (Braunschweig, F.R.G.), unlabelled prostaglandins, goat anti-rabbit IgG, anti-rabbit lgG-alk-
0005-2760/89/$03.50 © 1989 ElsevierSciencePublishers B.V. (BiomedicalDivision)
10 aline phosphatase conjugate, 5-bromo-4-chloro-3-indolyl phosphatase toluidine salt and p-Nitro blue tetrazolium chloride were obtained from Sigma (Miinchen, F.R.G.). NADPH was obtained from Boehringer (Mannheim, F.R.G.). DEAE-Trisacryl M, ultrogel AcA 44 and nitrocellulose membrane were obtained from LKil (lkomma, Sweden). Mono Q, phenyl-Superose, Superose 12 FPLC columns, CNBr-Sepharose 4B and Sephadex G.10 were obtained from Pharmacia (Up~ psala, Sweden). Ouchterlony double.immunodiffusion discs were obtained from ICN lmmunoBiochemicals (New York, U.S.A.), Enzyme purification. Swine kidneys (1000 g) were homogenized in 10 mM Tris/HCl buffer (pH 7.4)/1 mM dithiothreitol/0.02$ sodium azide/10$ (v/v) glycerol (buffer A) in a Waring Blendor for 2 rain (tissue to buffer ratio. 1:4). All purification steps were carried out at 0-4 e C. The homogenate was centrifuged at 20000 × g for 15 rain, The resulting supematant was adjusted to pH 5.2 with 1 M acetic acid as described by Moldave [14]. After 10 rain, the solution was centrifuged at 20000 × g for 15 rain in order to remove the precipitated protein. The supernatant was readjusted to pH 7.4 with 1 M NaOH and concentrated by 80% ammonium sulfate precipitation. The precipitate was redissolved and dialyzed against buffer A and applied to a 50 × 5 cm column of DEAE-Trisacryl M which had been equilibrated with buffer A. The effluent fractions were pooled and concentrated by 80~$ ammonium sulfate precipitation. The precipitate was dissolved in 10 mM Tris/HCI buffer (pH 7.4) containing 1 mM dithiothreitol, 0.02~$ sodium azide, 105 (v/v) glycerol and 0,1 M NaCI (buffer B) and applied to a 90 × 2,6 ¢m column of ultrogel AcA 44, Active fractions were pooled and concentrated by 805 ammonium sulfate precipitation. The precipitate was dissolved and dialyzed against 20 mM Tris/HC! buffer (pH 8,0)/1 mM dithiothxeitol/0,02~$ sodium azide/10$ (v/v) glycerol (buffer C) and applied to a $ × 0.5 cm FPLC Mono Q column ~qui. librated with buffer C, 20 m8 protein was injected into the column during each run. The column was washed with 5 ml of starting buffer, followed by a linear gradient of buffer C containing 0-0.15 M NaCI. Active fractions were pooled, concentrated by 80~$ ammonium sulfate precipitation and dialyzed against 100 raM potassium phosphate buffer (pH 7,4)/1 mM dithiothreitol/0,02~$ sodium azide/10$ (v/v) glycerol/1.2 M ammonium sulfate (buffer D), 9 mg protein was injected into the column during each run, The enzyme solution was applied to a 5 × 0.5 cm FPLC phenyl-Superose column equilibrated with buffer D. The column was washed with 5 ml of starting-buffer and eluted with a linear gradient of buffer D containing 1.2-0 M ammonium sulfate. Active fractions were pooled, concentrated by 80~ ammonium sulfate precipitation and dialyzed against buffer B. The enzyme solution was
applied to a 30 × 1 cm FPLC Superose 12 column equilibrated with buffer B. 0.5 mg protein was injected into the column during each run. Active fractions were pooled, dialyzed against buffer A and stored in aliquots at - 2 0 ° C . Affinity-purified antibodies. Antiserum was raised in female New Zealand white rabbits by injecting 50/tg of purified swine kidney PGEz-9-KR in complete Freund's adjuvant. The injection was repeated after 2 weeks and then every 2 weeks, but now with incomplete Freund's adjuvant. 9 days after the second injection, blood was taken weekly. The titer of antisera was followed by Ouchterlony double diffusion and by binding experiments to PGE2-9-KR previously labelled with 12Sl (see below). Affinity-purified PGE2-9-KR antibodies were obtained by chromatography on a PGEz-9-KR-Sepharose column (100/~g enzyme/ml gel). The gel (CNBr-Sepharose 4B) was prepared according to the manufactorer's advice [15] using CNBr-activated Sepharose 4B and 100 /~g of the purified enzyme. Affinity chromatography was performed by diluting 200/d of antiserum 5-fold in 0.1 M phosphate buffer (pH 7,0) and applying the mixture to the sepharose column. The column was washed with 10 ml of starting buffer until the absorption at 280 nm was zero. Ehtion was carried out using 0.1 M glycine/HCl buffer (pH 2,5). The purified antibodies were immediately dialyzed against phosphatebuffered saline (pH 7.4) and kept at 4°C.
Ouchterlony double diffusion and IZSl labelling of PGEz-9-KR. Ouchterlony double diffusion was performed in round (5 ram) agar gels (1.5~$ agar in phosphate-buffered saline). The central cavity contained 15 /d (15 Fg) of the purified enzyme and the peripheral wells contained 15 FI of the antiserum. The gels were incubated for 3 days at 4°C in a humid atmosphere until precipitation bands occurred. l~I labelling of purified PGE2-9-KR was performed a~ording to ReL 16 using the ehloramin T method (10 Fg of enzyme), t2SI-labelled protein was separated from free iodine by Sephadex G-10 chromatography and kept at 4°C,
Binding experiments and radioimmunoassay of PGEz. 9-KR. 10/,1 12Sl-labelled PGE2-9-KR (20000 cpm, 3.3 kBq) in 50 mM Tris/HCl buffer (pH 7.4)/0.1% gelatin was incubated with 100 Iti of antiserum (dilution 1:2000) or 50 ng of affinity-purified antibody in the presence or absence of unlabelled enzyme in a total volume of 200 ~ti. The mixture was incubated overnight at 4°C. Then 20/tl of goat anti-rabbit IgG (dilution 1 : 16) were added and the mixture was incubated overnight at 4°C. After the addition of 500 ttl of assay buffer, the incubation mixture was centrifuged at 10000 × g for 10 min. The supernatant was discarded and the bound radioactivity in the pellet was determined in a y-counter.
11
SDS-PAGE and immunoblotting. Discontinuous SDS-PAGE was performed according to Laemmli [17] (13 and 5% polyacrylamide, respectively). For immunoblotting, proteins were transferred electrophoretitally (100 mA, 1 h) from the gel to a nitrocellulose membrane. After blocking of protein binding sites with Tweea-phosphate-buffered saline containing 3% bovine serum albumin, the membrane was incubated first with the PGE2-9-KR antiserum (1:1000 diluted) for 2 h at room temperature and then for 2 h with goat anti-rabbit IgG-alkaline phosphatase conjugate (1:1000 diluted). The substrates used for staining were 5-bromo-4-chloro3-indolyl phosphate toluidine salt and p-Nitro blue tetrazolium chloride. Enzyme activity. PGE2-9-KR was measured as described previously [11] following the production of PGFzo by a radioimmunologicai method. The standard mixture contained 100 mM Tris/HCl buffer (pH 8.0) 100 pM NADPH, 71 /~M PGE 2 and ]0-100 ~1 of enzyme in a final volume of 1 ml. The incubation was carried out at 37°C for 60 rain and terminated by adjusting the pH to 3.5 with 1.5 M citric acid. The prostaglandins were extracted with 8 ml ethyl acetate and dried under a stream of nitrogen gas. The extracts were redissolved and chomatographed on silicic acid columns as described in Ref. 18. The PGF2ot fractions were analyzed with a specific radioimmunoassay [19]. One unit of enzyme activity is defined as the amount of
enzyme which catalyzes the production of 1 pmol PGF2,,/nfin under standard assay conditions. Other methods. Protein was determined according to the procedure of Bradford [20] using bovine serum albumin as standard. Staining of po~yacrylamide gels with silver was done according to Merril et al. [21]. Results PGE2-9-KR was purified from swine kidney, according to a procedure similar to that used for the decidual enzyme [11], but now FPLC chromatography was additionally used to improve purification. The ehition profile of the FPLC Mono Q step is shown in Fig 1. The isolation of PGE2-9-KR resulted in a 580-fold increase of its specific activity (Table I). SDS-polyacrylamide gel electrophoresis of the purified enzyme revealed a single protein band with an apparent molecular mass of 33 kDa (Fig. 2). The purified PGE2-9-KR was injected into three rabbits. The rabbits produced antisera which yielded a single immunoprecipitation line in Ouchterlony doublediffusion experiments when tested with th~ isolated enzyme (data not shown). When an excess amount of purified PGE2-9-KR was incubated with increasing amounts of antiserum and after 2 h with an excess of goat anti-rabbit IgG, up to 80% of the enzyme activity added was immunoprecipitated (data not shown). When the PGE2-9-KR separated by SDS-electrophoresis was
~ea-9-Km E280 {--)
( t J.cuC ) 0.20
NaCI {M)
4.0 ( - - - )
....... 2~
0,8 2(
0.15
0.6 15 0.t0 0.4 1@
i O.
. 0
.
~ ,
. . . . . . . . . . . . . . . . $ 10 IS T|mo
+
0.2
0 20
25
(rain)
Fig. l. FPLC Mono Q column chromatography of PGE2-9-KR. Pooled and dialyzed PGE2-9-KR was applied to a 5 ×0.5 cm FPLC Mono Q column that had been equilibrated with 20 raM T n s / H C I buffer ~pH 8.0)/1 mM dithiothreitol/0.02~ sodium azide/10~ ( v / v ) glycerol (buffer C). 20 nag protein was rejected into the column during each run. The column w~s washed with 5 ml of starting buffer and then eluted with 20 ml of a 0-0.15 M NaCi linear gradient at a flow rate of I ml/min. Fractions of 1 ml were collected. The aa:yme activity in each fraction was determined as described in Materials and Methods. - - - , A2BO;® @, PGE2-9-KR activity; - - - - - - , NaCI gradient.
12 TABLE !
Purification of PGE~-9.KRfrom swine kidney The enzyme activity was determined as described in Materials and Methods by measuring the production of PGF,=. One unit is defined as the amount of enzyme which catalyzes the production of 1 pmol PGF2= per min under standard assay conditions. Fraction
Volume (ml)
Protein (mR)
Activity (U)
20000 x g supematant
450
13486
61110
4.4
1
100
pH 5.2 supematant
380
6688
58786
8.8
2
96
FAuateof DEAE-Trbacryl M
700
45.~0
48650
10.7
2.4
79
Eluateof ultrogelAcA44
280
336
31164
92.8
21
51
Wuateof FPLC.MonoQ
S)
55
20775
377,7
86
34
Elutt~of FPLC.phenyI.Supetme
12
6
9776
1629.3
370
16
Eluateof FPLC.Superme 12
24
1.2
3055
2 545.8
580
5
transferred onto a nitrocellulose membrane which subsequently was incubated first with the antiserum and then with an alkaline phosphate labelled anti-rabbit IgG antibody, the enzyme could be detected immunochemically as a single band at 33 kDa (Fig. 2). The purified enzyme was labelled with lUl by the ¢hloramine T method in order to carry out binding experiments. The titration curve, where a constant amount of labelled enzyme was incubated with increasing dilutions of antiserum, demonstrated that 20~ of the total radioactivity was bound at a final antiserum dilution of 1:2000 (Fig. 3). When increasing amounts of unlabelled PGEa.9.KR were added to the incubation mixture, competitive binding of the unlabelled enzyme to the antibody was observed.
Specific activity (U/mR)
Enrichment
Yield (~)
The antibodies were purified using PGE2-9-KR CNBr-sepharose affinity chromatography. The tissue specificity of PGE2-9-KR antibody is shown in Fig. 4. hal-labelled PGE2-9-KR from swine kidney was incubated with affinity-purified antibody and increasing amounts of unlabelled enzyme protein from rabbit ovary, rabbit corpus luteum, rabbit endometrium and human decidua vera. The logit-log analysis yielded parallel lines for the different enzymes, indicating that the antibody was able to crossreact with PGE2-9-KR from the different sources. Rabbit corpus luteum and endometrium enzymes were 10000×g supernatants, rabbit ovary PGE2-9-KR had been partly purified (25fold, [12]), and PGE2-9-KR of human decidua vera had been purified to apparent homogeneity (232-fold [11]).
94 e7 48
14
1
23
56
Fig. 2, SDS-polyacrylamide gel eleetrophoi~esispatterns and immunoblot of swine kidney PGE2-9.KR. SDS-electrophoresis and immunobiot were carried out as described in Materials and Met,Nods, The protein was stained with silver. Amounts of protein analyzed were 0.2-2/tg per sample. From left to fight are shown the individual purification steps (2-6): (1) standards; (2) 20000× g supematants; (3) DEAE-Trisacryl M; (4) uitrogel AcA 44; (~) FPLC-Mono Q; (6) FPLC-Superose 12; (7) immunoblot of PGE2-9-KR.
13 so%B/T
in I
Fig. 3. Titration curve of PGE2-9-KRantiserum. 20000 cpm (3.3 kBq) of 12Si-labelled swine kidney PGE2-9-KR were incubated with antiserum of increasing dilution overnight at 4°C. 20 ~l of goat anti-rabbit [gG (dilution 1:16) were added and the mixture was incubated overnight at 4 o C. The incubation mixture was centrifuged at 10000× g for 10 rain, the supernatant was discarded and the bound radioactivity was determined in a y-counter. The x-axis represents the antiserum dilution and the y-~ds the ratio of bound radioactivity(B) versus total radioactivity (T). Each point represents the mean of three determinations + S.D.
The fact that binding of t25I P G E 2 - 9 - K R was competitively inhibited b y unlabelled PGE2-9-ketoreductase from swine kidney and from several reproductive tissues suggested the development of a radioimmunoassay for quantitation of the enzyme protein. The principle of the assay is that binding of radiolabelled enzyme to the
Io91t BlaO
o
°11 -4 ~
0,1
I
[ I IIIII
I
li
I
i
itllll
I
I
I IIllpt
10 I00 ProWIn (pg)
t
i
i
I
t
I
G
Dilution of Antlurum
iltttl
t
~)00
i
i
i ill
10000
Fig. 4. Tissuespecific of PGE2-9-KR antibody. 20000 cpm (3.3 kBq) of t251-1abelled swine kidney PGE2-9-KR were incubated with 50 ng of affinity.purified antibody in the p~esence of increasing 9mounts of PGE2-9-KRs from human decidua vera, rabbit ovary, rabbit corpus loteum and rabbit endometrinm overnight at 4°C. 20 pi of goat anti-rabbit [gG (dilution 1:16) were added and the mixture was incubated overnight at 4 ° C. The incubation mixture was centrifuged at 100ta3×g for 10 rain, the supernatant was discarded and the bound radioactivity was determined in a y-counter. The x-axis represents the log concentrations of added unlabelled enzyme and the y-axis the legit-log transformation of bound radioactivity (B) versus the value of B in the absence of unlabelled enzyme (Be). Each point represents the mean of three determinations+S.D. Enzymes: m m, human decidua vera PGE2-9oKR, purified to apparent homogeneity (232-fold) [Ill; * *, rabbit ovary PGE2-9-KR, partly purified (25-fold) [12]; A ~ A , rabbit corpus luteum PGE 29-KR, 10000xg superuatant; + - - - - + , rabbit endometrium PGE2-9-KR, 10000 × g supernatant.
i
i
ii
I
I
r
6O PGEe-O-KR 1~1
I
I
!
!
11
eOQ
Fig. 5. Standard curve of the radioimmunoassay for determining PGE2.9-KR concentration (legit-log transformation). 20000 cpm (3.3 kBq) of 12sI-labd|ed swine kidney PGEa-9-KR were incubated with 50 ng of affinity-purified antibody in the presence of increasing amounts of purified swine kidney PGE2-9-KR of known concentrations overnight at 4 ° C. 20 ~1 of goat anti-rabbit lgG (dilution 1:16) were added and the mixture was incubated overnight at 4°C. The incubation mixture was centrifuged at 10000×g for 10 m/n, the supernatant was discarded and the bound radioactivity in the pellet was determined in a 7-counter. The x-axis represents the log concentrations of added unlabelled PGEz-9-KR and the y-axis the log,it. log transformation of bound radioactivity (B) versus the value of B in the absence of unlabeUed enzyme (BO). Each point represents the mean of three determinations :t:S.D.
antibody is inversely proportional to the amount of unlabelled enzyme present, if the antibody concentration used is lower than that of the labelled enzyme. A constant amount of 125I-labelled swine kidney PGE2-9 K R was h, cubated with a constant a m o u n t of affinitypurified antibody in the presence of increasing amounts of purified swine kidney P G E e - 9 - K R of known concentration. As shown in Fig. 5, the l o g / t r o t transformation of the standard curve with unlabeled swine kidney enzyme was linear from 5 - 5 0 0 ng enzyme. Sensitivity is defined as the smallest a m o u n t of unl~belled enzyme protein causing 10% displacement compared with B e (value of B in the absence of unlabelled ligand). The calculated limit of detection was 5 n g ± 0% (mean :t: c.v.%) for swine kidney PGE2-9-KR. The upper detection limit was 600 ng protein. The precision of the assay was measured by adding a distinct a m o u n t of purified swine kidney P G E 2 - 9 - K R to the assay mixture. The intra- and interassay variation of n - - 1 0 assays were determined. The intraassay coefficient of variation was 6.4% aLd the interassay coefficient of variation was 13.2%, respectively. Discussion The purification of PGE2-9-KR from swine kidney resulted in a 580-fold increase in its specific activity, which is lfigher than the results obtained b y Chang et al. [4], who achieved a l l 5 - f o l d enrichment of the enzyme. The molecular mass of 33 kDa for the enzyme as
14
determined by SDS-gel electrophoresis fits well with the results obtained by other authors (29.5 kDa [22]) and with the molecular mass of PGE2-9-KR in other tissues. Tort and Hansen [23] found a molecular mass of 31 kDa for the enzyme iv .'abbit kidney, Lin and Jarabak [9] determined a molecular mass of 31-33 kDa for PGE2-9-KR in human placenta, and we found that the human decidual enzyme had a molecular mass of 31 kDa [11]. The antisera raised in rabbits by the injection of purified enzyme proved to react specifically with swine kidney PGE2-9-KR, as seen by a number of experimental results. In Ouchtedony diffusion experiments, the antisera yielded a single immunoprecipitation line when tested with PGE2-9-KR samples. The antisera were also able to precipitate the enzyme activity. Finally, immunoblotting of PGE2-9-KR revealed a single band which migrated to the same location as the purified enzyme. Production of antisera against swine kidney PGE2-9-KR has also been reported by Chang and Tai [22], These antisera, however, were derived from a partly purified enzyme (llS-fold enrichment). Furthermore, they were neither characterized nor used for further applications. Immunoassays for the quantitation of prostaglandinsynthesizing enzymes have been described by several authors, as for PGH synthase [24,25], PGI synthase [26] and thromboxane synthase [27], A method for the quantitation of PGE2-9-KR, however, has not been described up to now. The radioimmunoassay developed by us is a simple, specific assay and may be used for direct quantitation of immunoreactive PGE2-9-KR in swine kidney and several other tissues, as the antibody crossreacts with PGE2.9.KR from rabbit ovary, rabbit corpus luteum, rabbit endometrium and human decidua ,,'era, The assay is able to detect at least $ ng of the enzyme and is linear up to 500 n8 under the assay conditions described under Materials and Methods. The intra, and interassay coefficients of variation are below 15~, showing that the assay is quite reproducible and should be useful for the quantitation of enzyme levels under various physiological conditions. The availability of antibodies against PGEz-9.KR does not only provide a tool for quantitation of the enzyme, but also will enable us to purify the enzyme more rapidly using immunoaffinity chromatography. The use of antibodies against PGEz-9-KR will be helpful for the electron-micro.~opical localization of the enzyme in various tissues and for the investigation of the biosynthesis pathway of PGE2.9-KR using primary cell cultures, radiolabelling and immunoprecipitation methods. The physiological role of PGE2-9-KR is a subject of ~ontroversy. Weber et ai. [28] found a relation of the renal enzyme to the NaCI intake of the kidney, whereas Walkins and Jarabak [29] found that the enzyme obvi-
ously had no regulative function. Furthermore, the renal enzyme seems to have a broad substrate specifit.y for several aldehydes and ketones and Chang and Tai [30] suggested that the enzyme acts rather as general aldoketoreductase. The role of PGEa-9-KR in reproductive organs, however, seems to be more specialized. In tissues such as the ovary, endometrium and placenta, PGE 2 and PGF2+, often act in a different and sometimes contrary manner. At the end of human pregnancy, P G F . 2 c a u s e s ripening and dilatation of the cervix, while PGFaa causes u~erine contractions [31]. As PGEa-9-KR is supposed to regulate the PGE2/PGF2, ~ ratio, the enzyme may be involved in the events leading to labour and delivery [10]. In the corpus luteum of many species, P G E 2 s e e m s to act as a luteotrophic substance, while PGF2a is believed to act as luteolytic factor [32]. Recently, we found a possible correlation between PGE 29-KR activity and iuteolysis in the rabbit [13], indicating that this enzyme may be involved in this process by regulating the PGE2/PGF2, ratio. A quantitation and immunochemical detection of PGE2-9-KR in the rabbit corpus luteum at different stages of luteolysis and in human decidua at different stages of pregnancy, which will be possible with the help of our antibodies, will perhaps provide more insight into the physiological role of this enzyme, especially in re,roductive processes. Acknowledgements The technical assistance of R. Maucher is gratefully acknowledged. TF.is work was supported by the Deutsche Forschungsgemeinschaft (Schl 140/5-2). References
I Lee,S.C. and Levine.L. (1975) J. Biol. Chem. 250. 4549. 2 Lee, S.C., Pong, S.S., Katzen, D., Wu, $J. and Levine, L. (1975) Biochemistry14, 142-145. 3 Tai, H.H. and Chang, D.G. (1982)MethodsEnzymol.86, 142-147. 4 Chang, D.G., Sun, M. Tai, H.H. (1981) Biochem. Biophys. Res. Commun.99, 745-751. 5 Toh, B.S. and Hansen, H.S. (1979) Biochem.Biophys. Acta 574, 33-38. 6 Toh, B.S.and Hanscn, H.S. (1980) Prostaglandins20, 735-746. 7 Ko~ff,J. and Jarabak, J. (1980) Prostaglandins20, 111-125. 8 Kaplan, L., Lee, S.-C. and Levine, L. (1975) Arch. Biochem. Biophys. 167, 287-293. 9 Lin, Y.-M. and Jarabak, J. (1978) Biochem.Biophys. Res. Cornmen. 81, 1227-1234. 10 Schlegel,W., Kriiger, S. and Korte, K. (1984) FEBS Lett. 171, 141-144. 11 Krllger,S. and Schlegel,W. (1986) Eur. J. Biochem.157, 481-485. 12 Schlegel, W., Daniels, D. and Krfiger, S. (1987) Clin. Physiol. Biochem.5, 336-342. 13 Schlegel, W., Kriiger, S., Daniels, D., Fischer, B., Schneider, H.P.G. and Beier,H.M. (1988) .L Reprod. Fe~il. 83, 365-370. 14 Moldave,K. (1963) MethodsEnzymol.6, 757-761. 15 Pharmacia(1987)AffinityChromatography,Principlesand Methods, pp. 15-18.
15 16 Greenwood, F.C., Hunter, W.M~ and Glover, J.S. (1963) Biochem° istry 89, 114. 17 Laernmfi, U.K. (1970) Nature 227, 680-685. 18 3affe, B..I. and Behrman, H.R. (1974) in Methods of Hormone Radioimmunoassay (Jaffe, B.M. and Behrman, H.R., eds.), pp. 19-34, Academic Press, New York. 19 Schlegel, W., Wenk, K., Dolfinger, H.C. and Raptis, S. (1977) Clin. Sci. Mol. Med. 52, 255-258. 20 l~r~dford, M. (1976) Anal. Biochem. 72, 248-254. 21 Merril, C.R., Goldmann, D., Scdmann, $.A. and Ebert, M.H. (1981) Science 211, 1437-1438. 22 Cimng, D.C. and Tai, H.-H. (1982) Arch. Biochen~. Biophys. 214, 464-474.
23 Tort, B.S. and Hansen, H.S. (1979) Biochim. Biophys. Acta 574, 33-38.
24 Ro~h, G.J. (1982) Methods Enzymol. 86, 222-228. 25 DeWitt, D.L., Day, J.S., Gauger, J.A. and Smith, W.L (1982) Methods EnzymoL 86, 229-240. 26 DeWitt, D.L. and Smith, W.L. (1982) Methods Enzymol. 86, 240-2~. 27 Shen, R.-F. and Tai, H.-H. (1986) J. Biol. Chem. 261, 11585-11591. 28 Weber, P.C., Larsson. C. and Scherer, B. (1977) Nature 2(}6, 65-66. 29 Watkins, J.D. and .larabak, J. (1985) Prostaglan~ns 30, 335-349. 30 Chang, D.G.-B. and Tai, H.-H. (1981) Biochem. Biophys. Res. Commun. 101,898-904. 31 Keirse, MJ.N.C. (1978) Adv. Prostaglandin Thromboxane Res. 4, 87-102. 32 Horton, E.W. and Poyser, N.L. (1976) Physiol. Rev. 56, 595-651.