- Email: [email protected]
Effect of prostacyclin inhibition by tranylcypromine on uterine 6-keto-PGF1α levels during estrogen hyperemia in rats
PROSTAGLANDINS
EFFECT OF PROSTACYCLIN INHIBITION BY TRANYLCYPROMINE ON UTERINE 6 - K E T O - P G F . ~ LEVELS DURING ESTROGEN HYPEREMIA IN RA~S Farley, D.B. and Van Orden, D.E. Department of Obstetrics and Gynecology University of Iowa Iowa City, IA 52242 ABSTRACT A role for prostacyclin (PGIp) as a mediator of estrogeninduced increases in uterine blood-volume (UBV) was investigated by measuring uterine tissue levels of 6-keto-prostaglandin F (6-keto-PGF , ), and testing estrogen responses in rats 1 ~ pretreated with the PGI~, synthesis inhibitor, tranylcypromine (TCP). Uterine 6-keto-1#GFl~ content was determined by radioimmunoassay of tissue extracts purified through the use of high-performance liquid chromatography (HPLC). Estrogen treatment of castrate rats resulted in a significant increase of uterine 6-keto-PGFl~ as compared to saline treated controls (9.3 ng/uterine horn vs 6.7 ng/uterine horn, p=O.Ol). Pretreatment with TCP (20 mg/kg) markedly reduced the uterine content of 6-keto-PGF. (2.5 ng/uterine horn). The typical 50% Increase In UBV o~served after estrogen was unaffected by tranylcypromine pretreatment. It was concluded that the increased PGI 2 synthesis, as indicated by elevated levels of 6-keto-PGF. , may function as an amplifying mechanism for the uterlne vasodilation-induced by estrogen in castrate rats, but that production of this prostanoid is not essential for the estrogen response. o
•
.
IC~
INTRODUCTION Prostacyclin (PGIp) is a recently discovered vasoactive product of a r a c h i d o n i C acid metabolism, generated enzymatically from the endoperoxide, prostaglandin H~ (1,2). Because of its instability and short halflife in blological samples, endogenous PGI 2 can not be accurately measured. Therefore, to estimate PGI 2 in a biological system, many investigations have relied upon the measurement of 6-keto-PGF 1 , one of the stable but less active breakdown products (~). Using this approach, studies have focused on production of PGI^ in vari2 ous vascular beds (4). Synthesis has been demonstrated in large isolated blood vessel segments, vascular endothelial cells and, most recently, arterioles and capillaries (5-9). The uterus is a highly vascularized organ which has marked blood flow changes in response to hormonal stimuli
MAY 1982 VOL. 23 NO. 5
657
PROSTAGLANDINS (I0). The most dramatic vascular effects are observed following administration of estrogen (ii). The time delay between estrogen administration and onset of the vasodilatory response suggests that this estrogen action may involve a mediator. Several lines of evidence lead to the possibility that PGI 2 may play a role in estrogen-hyperemia. Earlier studies from our laboratories (12) which demonstrated that estrogen-hyperemia could be attenuated by cyclooxygenase inhibitors suggested that one of the products of arachidonic acid metabolism might be the mediator. Several other studies demonstrated that the major prostaglandin produced by the uterus was 6-keto-PGF~ , lq indicating that uterine tissue metabolizes arachidonic acld preferentially to PGIp (13,14). Since intra, arterial infusion of PGI 2 results in immediate increases in uterine blood flow, this prostanoid has been viewed as a potent uterine vasodilator (15,16). The present experiments were designed to test the involvement of PGI 2 in estrogen hyperemia using two separate approaches. TNe uterine content of 6 - k e t o - P G F during estroi gen hyperemia was measured using a specific ra~ioimmunoassay. Furthermore, the importance of prostacyclin synthesis to estrogen hyperemia was evaluated by testing uterine blood volume changes in response to estrogen in rats pretreated with tranylcypromine, a known inhibitor of PGI 2 formation in vitro (17). MATERIALS AND METHODS Solvents and Reagents: Tritiated 6-keto-PGF_~ was obtained from New England Nuclear, Corp. (Boston~ Mass.). Non-radioactive prostaglandins were kindly supplied by Dr. John Pike from the Upjohn Company (Kalamazoo, MI). Indomethaein was obtained from Merck, Sharp & Dohme Research Lab (Rahway, N.J.). Spectro-grade ethyl acetate and benzene, HPLC-grade acetonitrile, certified grade isopropanol, and reagent grade glacial acetic acid were obtained from Fischer Chem. Co. (Fairlawn, N.J.). HPLC-grade water was purchased from Alltech Assoc. (Arlington Heights, IL). Thyroglobulintype 1 and l-ethyl-3 (3 dimethyl amino-propyl) carbodiimideHCI (ethyl-CDl) were purchased from Sigma Chem. Co. (St. Louis, MO). Bovine gamme globulin, Fr. II (BGG) was obtained from Pel-Freez Biologicals (Rogers, Arkansas); polyethylene glycol 4000 was purchased from Union Carbide Co. (N.Y.,NY); Budget-Solve scintillation cocktail was obtained from Research Products International Corp. (Elk Grove Village, IL). For the radioimmunoassay, phosphate buffer (O.01 M, pH 7.4), containing 2 mg/ml bovine gamma globulin (PB-BGG) was used to re-dissolve tissue extracts after HPLC isolation of 6-keto-PGF 1 a n d for the dilution of antibody, cold standards and tracer. Drugs: 17B-estradiol-3-benzoate (E2B ; Sigma Chem. Co., St. Louis, MO) was dissolved in corn oil at a concentration of O.14 mg/ml. 17B-estradiol (E2; Sigma Chem. Co., St. Louis,
658
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS MO) was dissolved in 95% ethanol and diluted with phosphatebuffered saline (0.001 M phosphate buffer, pH 7.4, 0.15 M NaCI) to a concentration of 0.5 ug/ml. E 2 vehicle consisted of phosphate-buffered saline containing an amount of ethanol equal to that found in the estradiol solution. Tranylcypromine sulfate (Smith, Kline, & French Labs, Philadelphia, PA) was prepared at a concentration of 20 mg/ml in saline. To block prostacyclin synthesis during UBV studies, TCP was given (20 mg/kg BW, i.p.) 30 minutes prior to the injection of E 2 or E 2 vehicle. 6-keto-PGF I ~ radioimmunoassay: Antibody was produced to 6-keto-PGF _ -thyroglobulin conjugates prepared by the carbodiimide condensation method (18). Briefly, 5 mg 6 - k e t o - P G F 1 ~ dissolved in 1 ml acetonitrile was added dropwise to i ~ml distilled water containing 10 mg thyroglobulin and 7.5 mg ethyl-CDl. The reaction was gently stirred at 4 ° C overnight. After one hour of dialysis against cold water the conjugate was lyophilized. New Zealand white rabbits were immunized intradermally with various doses (i00 ug-5 mg) of c o n j u g a t e dissolved in 0.15 NaCI and emulsified with complete Freund's as previously described (19). Rabbits were boosted i.v. on a monthly basis with 1/10 the initial dose of conjugate dissolved in saline. Blood was collected into chilled, heparinized containers 1 week after each boost and centrifuged immediately to harvest the antiplasma. The RIA was carried out in 12 mm x 50 mm polypropylene test tubes (Sarstedt, Inc. d Princeton, NJ). Unlabeled 6-ketoPGF~ ~ was stored at -20- C in acetonitrile (i00 ug/ml) and di1~ted in PB-BGG to yield solutions of 10 ng/ml and 1 ng/ml at the time of assay. Standard curve tubes (2 pg-160 pg) were aliquoted in triplicate using an automatic pipetting station (Micromedic Systems, Inc., Horsham, PA). Tissue samples from the HPLC step were re-dissolved with 0.5 ml PB-BGG and delivered to assay tubes in three different volumes. The volume of each tube was simultaneously adjusted to 200 ul with PB-BGG by means of the rinse pump on the pipetting station. A dilution ~f antiplasma previously shown to bind 40% of 5000 cpm's of H - 6 - k e t o - P G F l ~ was prepared in PB-BGG and 50 ul was added to each tube; non-specific binding tubes received 50 ul of PB-BGG in place of antibody. Labeled 6 - k e t o - P G F 1 ~ (5000 cpm's in ~ ul PB-BGG) was added and the entire assay was incubated at C overnight. Precipitation of antibody-bound 6-keto-PGF1~ was accomplished by the addition of 0.2 ml PB-BGG and 0.5 ml polyethylene glycol 4000 (40% w/v) as previously described (18). After decanting the supernatant, the pellet was analyzed for bound-radioactivity by re-dissolving each with 0.i ml 0.I N acetic acid and then adding 3 cc of Budget-Solve. All test tubes were counted to i0,000 cpm's and the data calculated on a PRIME computer using Rodbard's program which provides weighted standard curves and assessment of potency and parallelism of unknown and standard (20). Uterine tissue sample data were corrected for procedural recovery previously determined to be
MAY 1982 VOL. 23 NO. 5
659
PROSTAGLANDINS 45
•
5°/~ w i t h 3 H - 6 - k e t o - P G F ° i ~ " Tissue extraction and HPLC
isolation
o~
6-keto-PGF
i~:
Uterine horns, previous l~ soaked in indomethacin solution (20 ug/ml) and stored at -20-- C, were weighed, immersed in liquid nitrogen, pulverized, and quickly transferred to glass extraction tubes (12 x i00 mm), held in an ice bath. Extraction of prostaglandins was carried out as previously described (19). Briefly, one milliliter 0.05 M citric acid containing 0.15 M NaCI per I00 mg of tissue was used as the tissue diluent; tissue suspensions were sonicated for 20 minutes at 4 ° C followed by extraction with two volumes of ethyl acetate: isopropanol (i:i, v:v). Three milliliters of 0.15 M NaCI and two milliliters of additional ethyl acetate were added• The resulting suspensions were centrifuged at 15OO x g for IO minutes• All procedures were carried out at 4 ° C. The entire organic phase was transferred to a 12 mm x 75 mm glass tube and evaporated under vacuum using a Speedvac concentrator (Savant Instruments, Inc., Hicksville, NY). The dried extracts were stored at -20 ° C until chromatographed. In the ~xperiments designed to assess tissue extraction efficiency, H-6-keto-PGF I ~ was added to the tissue suspension and the entire extract counted in Budget-Solve. Each tissue extract was re-dissolved in 200 ul HPLC solvent (water:acetonitrile:benzene:acetic acid, 76.7%:23.0%:0.2%: 0.1%), briefly vortexed and sonicated. S a m p l e s were freed of suspended matter via a microfiltration apparatus (Bioanalytical Systems, Inc., West Lafayette, IN) using a 1.0 u teflon membrane initially• After a second microfiltration using a 0.2 u regenerated cellulose membrane, a measured volume (150-200 ul) of the clear extract was resolved on a 30 cm x 3.9 mm-i.d, reversed phase HPLC column (Fatty Acid Analysis Column, Waters Assoc., Milford, Mass.) as previously described (21). The flow rate was 2 ml/min/tube; fifteen fractions were collected in minivials. Recovery of 6-keto-PGF I ~ from the HPLC was established in preliminary experiments utilizing labeled 6-keto-PGFl~ as well as trace-labeled tissue extracts• When uterine extracts were chromatographed to isolat~ the fraction containing 6keto-PGF 1 ~, a sample of pure H-6-keto-PGFl~ was injected after every two or three extracts to affirm that the elution profile of the 6 - k e t o - P G F ~ was unchanged• The column was stripped with filtered methanol between each injection to remove residual prostaglandins. The fraction containing 6keto-PGF 1 ~ was taken to dryness in a Speedvac concentrator and stored at -20 ° C until the time of radioimmunoassay. Hyperemia Model: Twenty-nine female Sprague-Dawley rats (Bio-Labs, Madison, WI) weighing between 175-200 gm were prepared for testing of uterine vascular responses to estrogen according to the "uterine hyperemia" protocol previously established in this laboratory (22). On day O, each rat was ovariectomized via the dorsal approach; on day 7, each animal received a subcutaneous injection (O.i ml/lO0 mg BW) of E2B
660
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS dissolved in corn oil. On day 14, fourteen of the animals were pretreated with TCP (20 mg/kg, i.p.). The remaining fifteen animals (TCP controls) received saline at the same time. Thirty minutes after administration of TCP or saline, half of the animals in each group were given E_ (0.5 ug/kg) into the femoral vein under light ether anesthesia~ the remaining animals in each group received E 2 vehicle. Two hours following injection of E 2 or E^ vehlcle, all animals were again briefly anesthetized with ether and the 13clontralateral femoral vein was injected with i0 uCi of I-human serum albumin (RISA) (E.R. Squibb & Sons, Inc., Princeton, NJ) in 0.i saline. After an equilibration time of five minutes the animals were sacrifificed, the abdomen was opened and blood was withdrawn from the inferior vena cava. The uterus was rapidly excised and trimmed of adherent fat and mesometrium. One horn was used for blood volume determination and the other horn was processed for 6 - k e t o - P G F ~ t~ssue content. A section of small intestine was a~so removed and used for control intestinal blood volume determination. Blood volumes were calculated on a dry weight basis from the radioactivity of the tissues relative to that of the blood. The uterine horns to be analyzed for 6-keto-PGF~ ~ were soaked in a solution of indomethacin (20 ug/ml) ~reshly prepared in phosphate buffer (0.i M, pH 7.4) by gently heating and stirring. After approximately five minutes, these tissues were transferred to an aluminum foil wrapper and stored at -20 ° C until pulverization and extraction. Blood Pressure Determinations: Blood pressure in conscious animals was determined by means of a PE iO cannula inserted into the descending aorta via the left carotid artery. Pressure changes were monitored with an Alltech (City of Industry, CA) pressure transducer which was coupled to a Beckman (Irvine, CA) R611 dynograph. Blood pressure and heart rate were recorded at 30 minute intervals and animals returned to housing cages between monitoring periods. Statistical Analysis: Blood volumes were compared using a one way analysis of variance with the level of significance being p < 0.02. Tissue contents of 6 - k e t o - P G F l ~ w e r e compared by the same method of analysis.
RESULTS Radioimmunoassay characteristics: All rabbits immunized with 6 - k e t o - P G F l ~ conjugates developed antibodies after the first boost. In general, higher titre antibodies were produced when initial immunization was with larger amounts (5 mg) of conjugate; low titre antibodies yielding assay sensitivity in the femtogram range were obtained when low dose (IO0 ug) immunization was employed. The antibody used in the present study was obtained after % months of immunization and bound 43% of 2.5 pg (5000 cpm) H-6-keto-PGF 1 ~ at a final anti-
MAY 1982 VOL. 23 NO. $
661
PROSTAGLANDINS
plasma dilution of 1:1200. A typical standard curve is illustrated in Fig. 1 (sensitivity=l.7pg; ED =24 2pg; slope=50 " -1.00667; correlation coefficient:-0.993; range=2-160 pg).
1°°f 80
60
=o 40
20
01
4
16
40
Picograms 6-Keto-Prostaglandin F 1~
Fig. i: 6-keto-PGF. standard curve with antiplasma R132-7/2/80 and 3H-6-PGFI~ . I~
662
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
TABLE
I:
SPECIFICITY 7/2/80
OF
THE
6-KETO-PGF l a
ANTIPLASMA
R132-
Relative crossreaction (%) at 50% 3displace ment of H-6-ketoPGF 1
Compound
6-keto-PGF l a PGE 1 PGE 2 PGF 1 a PGF~
i00.0 4.6 4.9 7.6 5.7
TxB~ a PGD 2
<0.1 <0.1
PGA. PGA~ PCB~ 13,~4-DIHYDRO-15-KETO 15-KETO PGE_ 13,14-DIHYD~O PGE^ 13,14-DIHYDRO-15-~ETO 13,14-DIHYDRO PGF 2 a
<0.i
PGE 1
PGF 2
The specificity of the assay is shown in Table I. Only the primary PGF's and PGE's showed any appreciable crossreactivity. Within-assay coefficient of variation was 7.5% at a concentration level of 1-5 ng/uterine horn (n=29). Interassay coefficient of variation was determined for eight uterine samples falling in a concentration range of 0.7-5.0 ng/ horn and was 8.7%. The RIA recovery of unlabeled 6 - k e t o - P G F l ~ (50-1000 pg) added to 2 ml of column solvent (concentrated and re-dissolved in protein buffer for aliquoting) averaged 82% with a correlation coefficient of 0.9982 (n=9). The solvent blank was 12.7 pg/sample. Qualitative validity of the 6-keto-PGF 1 ~ RIA was indicated by the markedly reduced tissue levels of 6-ket°-PGF 1 a measured in uteri obtained from rats treated with tranylcypromine, a known inhibitor of PGI 2 synthesis (3) (See Fig. 4 below). Ti@sue extraction and isolation efficiency: Recovery of added--JH-6-ket0~PGF, a from uterine tissue suspensions wasi J 59.4 ~ 1.6% (n=12). No loss in concentration of H-6-ketoPGF.~ was incurred during the filtration step prior to HPLC analysis. The --H-6-keto-PGF. ^ chromatographed as a single peak (Fig. 2) at a retention time of 9 minutes with a column
MAY 1982 VOL. 23 NO. 5
663
PROSTAGLANDINS recovery of 76.6 + 2.1% (n=13). The reproducibility of the rete~ntion time is ~ i s o shown in Fig. 2. The elution pattern of -H-6-keto-PGF 1 ~ was not altered either by presence of uterine tissue extracts or by stripping the HPLC column with methanol after each tissue extract run. Testing of the HPLC fractions with antibody to 6-keto-PGF 1 ~ showed the peak radioactivity fraction to be fully immunoreactive.
Conditions: Column:
25,000 E
Mobile Phase:
F a t t y Acid Analysis, 1 0 p.m 3 . 9 x 3 0 i~m 76.7%: 2 3 . 0 % : 0 . 2 % 0 . 1 % , water: acetonitrile: b e n z e n e : a c e t i c acid
Flow Rate:
2 ml/min
A
20,000 15,000
0 0
•~
10,000
o .£
5,000
~
0 0
I
I
I
I
I
I
I
5
10
15
0
5
10
15
RETENTION TIME IN MINUTES Fig. 2: Retention time in minutes of 3H-6-keto-PGF.~ detected by quantity of radioactivity: A) Initial resolution; B) Resolution after chromatography of three tissue extracts.
Effect of tranylcyPromine on systemic hemodynamics: Blood pressure and heart rate changes observed in four conscious rats following TCP administration (20 mg/kg, i.p.) are presented in Table II. Although blood pressure was elevated at 30 minutes after TCP, it was not statistically different from baseline at 2½ hours, the time after drug treatment at which animals were sacrificed in the uterine hyperemia experi-
664
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
ment. Heart rate remained elevated throughout the experiment. TCP-treated rats displayed hyperactivity, salivation and piloerection over the 150 minutes of observation. TABLE
II:
EFFECT OF TRANYLCYPROMINE (20 mg/kg) PRESSURE (BP) AND HEART RATE (HR) OF RATS
ON BLOOD CONSCIOUS
TIME AFTER TCP-TREATMENT ere~reatment
BP (mmHg)
HR (beats/ min)
30 min,
60 min.
90 min.
120 m i n
150 min.
107+6
135+6
127+16
118+15
115+15
114+13
356+15
428+30
420+20
404+20
423+23
438+14
Values
represent mean + SEM
Effect of tranylcypromine on uterine 6-keto-PGF and is-blood volume: As shown in Fig. 3, animals tested for uterine vascular responses to estrogen demonstrated the typical 50% increase in blood volume two hours after steroid administration (23). UBV of saline control animals was 217 + 18.1 ug/g, whereas that of estrogen treated controls was 3~6.8 + 25.5 ul/g. Estrogen treatment of these castrate rats also resulted in significant elevation of uterine 6-keto-PGF~ content as shown in Fig. 4. At two hours, content of 6-~eto-PGF. ~ in uteri of saline treated controls was 6.7 ~ 0.5 ng~horn; whereas, estrogen treated animals had 9.3 + 0.8 ng/horn. Pretreatment with the TCP 30 minutes before estrogen administration resulted in a markedly reduced tissue content of 6-keto-PGF~ (2.5 ! 0.3 ng/horn) as shown in Fig. 4. NevertheIn less, TCP did not prevent the estrogen-induced increase in UBV (376.6 ! 32.1 ul/g), nor did it alter saline UBV (265.1 ~ 22.6 ul/g) (Fig. 3). Although not shown, intestinal blood volumes were not altered either by estrogen administration or TCP pretreatment.
MAY 1982 VOL. 23 NO. 5
665
PROSTAGLANDINS
[~ Saline [ ] Estrogen (0.5p.g/kg) 400
e
(7)
m
(7)
300
(7)
=m
I-
(8)
t-
D
N N N
200
E
N
0
II1 ~
,
l
=m t,=
100
Control
Tranylcypromine
Fig. 3: Uterine blood volume (determined on a dry weight basis) reponse to estrogen or saline in control or t r a n y l c y p r o mine treated animals. Each bar represents the mean + SEM; the number of animals in each group is shown in parentheses. Asterisks indicate values that are s i g n i f i c a n t l y different from their respective saline controls at P <0.02.
666
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
t"
m
L_
(11)
o 10 "1" c L_
8 :3
LL 0
(13)
T
6
n
I
o
,4--1
4
t
(1)
(8)
I
©
{31 C
2
0
Saline
Estrogen
Tranylcypromine & Estrogen
Fig. 4: Uterine 6-keto-PGFl~ content two hours after saline, estrogen or tranylcypromine + estrogen treatment. Asterisk indicates value significantly different from saline treated animals at P < 0.02. Daggar indicates value significantly different from saline treated and estrogen treated animals at P < 0.001.
MAY 1982 VOL. 23 NO. 5
667
PROSTAGLANDINS DISCUSSION
A number of reports have appeared on the measurement of 6-keto-PGFl~ in various biological samples as an index of prostacyclln involvement (24-26). The radioimmunoassay described in the present study employing HPLC isolation of 6-keto-PGF.~ from tissue extracts fulfills the requirements of high sensitivity, specificity and accuracy. A sensitivity of less than two picograms observed with the antibody used for this assay compares favorably with that reported for several other RIA's for this PGI 2 metabolite (24,27,28). The use of low doses (iO0 ug) of immunogen may explain the production of specific and avid antisera by these animals within a few weeks of immunization. Vaitukaitis et al found similar results following a primary immunization with as little as 20 ug of HCG subunits and testosterone-BSA conjugates (29). In addition, we have found that the conjugation reaction is highly efficient (molar ratios greater than iO00:i) when thyroglobulin is used instead of albumin as the carrier protein (unpublished observation). Crossreactivity of this antibody compares well with that reported by Salmon (27). Other investigators have described production and use of antibodies that have limited cross-reaction (less than 1.0%) with primary prostaglandins (24,30,31). Nevertheless, through the use of HPLC purification of tissue extracts in the present study, interference by other arachidonic acid metabolites is minimized in the final assay of sample. In the course of defining the best method of isolating 6-keto-PGF 1 ~ from biological samples, various prostaglandin plasma extraction methods were explored (cyclohexane:ethyl acetate, I:i; repeated volumes of ethyl acetate; washing of organic extracts with water; extraction of acetone precipitates of plasma; preliminary lipo-protein depletion of plasma; extraction at various pHs). None of these specialized methods thought to yield better extraction efficiency of 6-keto-PGF. improved the 73% recovery obtained with the standard method ~ routinely use for other prostaglandins (acidification with citric acid and a single extraction with ethyl acetate after removal of neutral lipids with petroleum ether). Because of the above findings and the reproducible extraction of tissue homogenates using the method already established in this laboratory (18) and used by many others, we did not pursue alternate tissue extraction methods. Pulverization of frozen tissue, as opposed to homogenization using ground glass tissue grinders, was preferred because the rapidity of the method minimizes the possibility of generation of prostaglandins during tissue processing. The inclusion of a synthesis inhibitor at this step was an added assurance that levels reflected true endogenous content. In preliminary attempts to isolate 6-keto-PGFl~ via stan-
668
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
~ ard
silicic acid chromatography we were unable to isolate H-6-keto-PGF~ in any one fraction to an extent greater than 1 50%. Other authors have used either TLC (27) or bulk collections of all prostaglandin classes from silicic acid columns (30) in order to purify 6-keto-PGF_ from sample extracts. With applications of HPLC coupled ~i~h RIA for determination of many cyclooxygenase products (21), it is possible to isolate endogenous 6 - k e t o - P G F ~ in a relatively small volume. As reported in the present study, HPLC resolution of 6-keto-PGFl~ is highly reproducible, unaffected by presence of large quantities of prostaglandins as found in uterine tissue and requires only 10-15 minutes for elution of the 6-keto-PGFA peak. All other arachidonic acid metabolites are stripped ~r~m the column by the methanol injection permitting HPLC isolation of 6-keto-PGFl~ from approximately 20 samples per eight hour period. Although the procedural loss of 6-keto-PGF~ during is extraction of tissue and HPLC isolation is rather substantial (55%), the high concentration of endogenous 6-keto-PGF.~ (i00200 pg/mg) in this vascular organ eliminates any problem of assay sensitivity. Furthermore, by assaying samples freed of other prostaglandins, one may have greater confidence that inhibition of antibody-binding is due to 6-keto-PGFl~ in the sample rather than to high concentrations of crossreacting prostanoids. If assay of 6-keto-PGF l ~ is to be used as a measure of prostacyclin production, one must show reduced tissue levels of the stable metabolite in animals pretreated with specific PGI 2 synthesis inhibitors. Such was the case in the study presented here as TCP treatment reduced the tissue content to one-fifth basal levels. Although the original evaluation by Gryglewski, et al (17) suggested that TCP is a selective inhibitor of PGI 2 synthesis in vitro, some recent work has raised questions concerning its specificity (32-34). Nevertheless, the in vivo use of TCP in other biologic systems known to involve prostaglandins indicate that it does block PGI 2 formation. Jonsson, et al (35) used the same dose in mice as was used in the present study and found uterine levels of PGE and PGF to be unaffected. Unfortunately they did not measure 6-keto-PGF ~ . Likewise, the platelet aggregation study of Rosenblum land El Sabban (36) using i0 mg/kg TCP in mice strongly suggests that TCP acts to inhibit PGI^ synthesis. Both studies eliminated the monoamine oxidase (M~O) inhibitory properties of the drug as an explanation for their results by showing that other MAO inhibitors had no effect on the response under study. Furthermore, the ID50 " of TCP as a MAO inhibitor at a dose of 6 mg/kg in rats requlres 16 hours (32); hence, drug related changes occurring in the 2½ hour period of the present study would not be on that basis. Elevation of uterine 6-keto-PGFl~ following acute estrogen administration to castrate rats parallels our earlier •
MAY 1982 VOL. 23 NO. 5
669
PROSTAGLANDINS
observation of increased uterine PGE in estrogen hyperemia (22). The tissue levels of 6 - k e t o - P G F ~ in control animals of the present study are approximately ~.5 times greater than that found for the other potent uterine vasodilator prostaglandin, PGE. Although few studies exist concerning uterine 6-keto-PGF~ ~ concentrations in any species (28,37-39), that of Poyser I and Scott (28) reported very low tissue levels (15-38 pg/mg) of this metabolite in rat uterine homogenates. These authors also found a high PGI^ synthesizing capacity for 2 the rat uterus, with both concentration and production of 6-keto-PGF~ higher on the day of estrus than on diestrus. Although t~e relative amount of 6-keto-PGF~ measured does not compare well with the content reported in the present study, the positive influence of estrogen is similar. In contrast, Gimeno, et al, (40) looking at production of PGI 2like substance by uteri from estrous rats or ovariectomized rats treated in vivo with high doses of estradiol, associated estrogen with an inhibitory action on the uterine PGI^ synthe2 tase system. The time after estrogen administration (24 hours) chosen by these authors for study of PGI^k production is a time when estrogen hyperemia is known to have subsided. Therefore, if estrogen does indeed have a positive effect on PGI_ as we indicate by the increased quantity of 6-keto-PGF~ found in estrogen stimulated uteri, it is an early effecl. ~ Although the present study measures content and that of Gimeno, et al, measures production, one would expect the effect of estrogen to be similar if PGI 2 synthesis by the uterus is a consequence of estrogen action. The observation, in the present study, that UBV responses to estrogen persist after administration of TCP to rats, indicates that PGI^ synthesis is not essential for this estroz gen response. Although PGI 2 has repeatedly been shown to be a uterine vasodilator and uterine levels of 6-keto-PGF~ are 1 elevated at the time of maximal response, this prostanold can not be considered to fulfill a mediator role. Since UBV increased normally after estrogen administration to TCP treated rats, we must conclude that the increased PGI 2 synthesis during the response is not obligatory for this estrogen vascular action. It is more likely that this concomitant production of PGI Z functions as an amplifying mechanism for the vasodilation initiated by the primary estrogen mediator. REFERENCES i.
Moncada, S., Gryglewski, R., Bunting, S., and Vane, J.R. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 263: 663-665, 1976.
2.
Johnson, Mcguire,
670
R.A., Morton, D.R., Kinner, J.H., Gorman, R.R., J.C. , Sun, F.F. , Whittaker, N. , Bunting, S. ,
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
Salmon, J.A., Moncada, S., and Vane, J.R.: structure of prostaglandin x (prostacyclin). ins 12: 915-928, 1976.
The chemical Prostagland-
3.
Pace-Asciak, C. Isolation, structure and biosynthesis of 6-keto-PGF is in the rat stomach. J. Am. Chem. Soc 98: 2348-2349, 1976.
4.
Moncada, S., and Vane, J.R. vascular tissue. Fed. Proc.
5.
Moncada, S., Higgs, E.A., and Vane, J.R. Human arterial and venous tissues generate prostacyclin, a potent inhibitor of platelet aggregation. The Lancet i: 18-20, 1977.
6.
Botha, J.H., Leary, W.P., and Asmal, A.C. Enhanced release of PGI^-like substance in experimental hypertension. Prost. and Mefdicine 3: 251-252, 1979.
.
The role of prostacyclin 38: 66-71, 1979.
in
MacIntyre, D.E., Pearson, J.D., and Gordon, J.L. Localization and stimulation of prostacyclin production in vascular cells. Nature 271: 549-551, 1978.
8.
Levine, L., and Alam, I. Arachidonic acid metabolism by cells in culture: Analysis of culture fluids for cyclo-oxygenase products by radioimmunoassay before and after separation by high pressure liquid chromatography. Prost. and Medicine 3: 295-304, 1979.
9.
Maurer, P., Moskowitz, M.A., Levine, L., and Melamed, E. The synthesis of prostaglandins by bovine cerebral microvessels. Prost. and Medicine 4: 153-161, 1980.
iO. Reynolds, S.R.M. Physiology of the uterus, York, 1949. Paul B. Hoeber, Inc. 297-307.
ed.
2, New
ii. Greiss, F.C., Jr., and Anderson, S.G. Effect of ovarian hormones on the uterine vascular bed. Am. J. Obstet. Gynecol. 107: 829-836, 1970. 12. Ryan, M.J., Clark, K.E., Van Orden, D.E., Farley, D.B., Edvinsson, L., Sjoberg, N.O., VaN Orden, L.S. III and Brody, M.J. Role of prostaglandins in estrogen-induced uterine hyperemia. Prostaglandins 5: 257-268, 1974. 13. Fenwick, L., Jones, R.L., Naylor, B., Poyser, N.L., and Wilson, N.H. Production of prostglandins by the pseudopregnant rat uterus, in vitro, and the effect of tamoxifen with the identification of 6-keto-prostaglandin FI~ as a major product. Br. J. Pharmac. 59: 191-199, 1977.
MAY 1982 VOL. 23 NO. 5
671
PROSTAGLANDINS
14. Jones, R.L., Poyser, N.L., and Wilson, N.H. Production of 6-oxo-prostaglandin F ~ by rat, guinea-pig, and sheep uteri in vitro. Br. J~ Pharmac. 59: 436-437, 1977. 15. Resnick, R., and Brink, G.W. Uterine vascular response to prostacycli n in nonpregnant sheep. Am. J. Obstet. Gynecol. 137: 267-269, 1980. 16. Clark, K.E., Mills, E.G., Stys, S.J., and Seeds, A.E. Effects of vasoactive polypeptides on the uterine vasculature. Am. J. Obstet. Gynecol. 139: 182-188, 1981. 17. Gryglewski, R.J., Bunting, S., Moncada, S., Flower, R.J., and Vane, J.R. Arterial walls are protected against a substance (prostaglandin x) which they make from prostaglandin endoperoxides. Prostaglandins 12: 685-713, 1976. 18. Van Orden, D.E. and Farley, D.B. Prostaglandin F2~radioimmunoassay utilizing polyethylene glycol separation technique. Prostaglandins 4: 215-233, 1973. 19. Van Orden, D.E., Farley, D.B., and Clancey, C.J. Radioimmunoassay of PGE and an approach to the specific measurement of PGE I. Prostaglandins 13: 437-453, 1977. 20. Rodbard, D., and Faden, V.B. National Technical tion Service, Springfield, VA, 22151.
Informa-
21. Alam, I., Ohuchi, D., and Levine, L. Determination of cyclooxygenase products and prostaglandin metabolites using high-pressure liquid chromatography and radioimmunoassay. Anal. Biochem. 93: 339-345, 1979. 22. Clark, K.E., Baker, H.A., Bhatnagar, R.K., Van Orden, D.E., and Brody, M.J. Prevention of estrogen-induced uterine hyperemia by ~-adrenergic receptor-blocking agents. Endocrinol. 102: 903-909, 1978. 23. Van Orden, D.E., Clancey, C.J., and Farley, D.B. Uterine serotonin and receptor blockade during estrogen-induced uterine hyperemia. Proc. Soc. Exp. Biol. Med. 167: 469-474, 1981. 24.
Oliw, E. A radioimmunoassay for 6-keto-prostaglandin FI~ utilizing an antiserum against 6-methoxine-prostaglandln F1 ~ Application to study renal in vitro synthesis of prostacyclin. Prostaglandins 19: 271-284, 1980.
25.
Abdel-Halim, M.S., Lunden, I., Cseh, G. and Anggard, E. Prostaglandin profiles in nervous tissue and blood vessels of the brain of various animals. Prostaglandins 19:
672
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
249-258, 1980. 26.
Mitchell, M.D., Keirse, M.J.N.C., Brunt, J.D., Anderson, A.B.M. and Turnbull, A.C. Concentrations of the prostacyclin metabolite, 6-keto-prostaglandin F 1 a in amniotic fluid during late pregnancy and labour. Br. J. Obstet. and Gynecol. 86: 350-353, 1979.
27. Salmon, J.A. A radioimmunoassay for 6-keto-prostaglandin F1 ~ Prostaglandins 15: 383-397, 1978. 28. Poyser, N.L. and Scott, F.M. Prostaglandin and thromboxane production by the rat uterus and ovary in vitro during the oestrous cycle. J. Reprod. Fert. 60: 33-40, 1980. 29.
Vaitukaitis, J., Robbins, J.B., Nieschlag, E. and Ross, G.T. A method for producing specific antisera with small doses of immunogen. J. Clin. Endocr. 33: 988-991, 1971.
30. Mitchell, M.D. A sensitive radioimmunoassay for 6-ketoPGFI~ : Preliminary observations on circulating concentrations. Prost. and Medicine i: 13-21, 1978. 31.
Ylikorkala, O. and Viinikka, L. Measurement of 6-ketoprostaglandin F I~ in human plasma with radioimmunoassay: Effect of prostacyclin infusion. Prost. and Medicine 6: 427-436, 1981.
32.
Rajtar, G. and de Gaetano, G. Tranylcypromine is not a selective inhibitor of prostacyclin in rats. Thromb. Res. 14: 245-248, 1979.
33.
Chignard, M., Vargaftig, B.B., Sors, H. and Dray, F. Synthesis of thromboxane B_ in incubates of dog plateletZ rich plasma with arachidonic acid and its inhibition by different drugs. Biochim. Biophys. Res. Commun. 85: 16311639, 1978.
34. Hong, S.L., Carty, T. and Deykin, D. Tranylcypromine and 15-hydroperox~-arachidonate affect arachidonic acid release in addition to inhibition of prostacyclin synthesis in calf aortic endothelial cells. J. Biol. Chem. 255: 9538-9540, 1980. 35. Jonsson, H.T., Rankin, J.C., Ledford, B.E. and Baggett, B. Uterine prostaglandin levels following stimulation of the decidual cell reaction: Effect of indomethacin and tranylcypromine. Prostaglandins 18: 847-857, 1979. 36. Rosenblum, W.I. aggregation by
and Ei-Sabban, F. Enhancement of platelet tranylcypromine in mouse cerebral micro-
MAY 1982 VOL. 23 NO. 5
673
PROSTAGLANDINS
vessels.
Circ. Res. 43: 238-241,
1978.
37. Alwachi, S.N., Bland, K.P. and Poyser, N.L. Production of 6-oxo-prostaglandin F , PGF^ , and PGE_ by sheep uterine tissue. Prost. and M ~ i c i n e ~: 329-332, ~980. 38. Zamecnik, J., Kennedy, T.G. and Armstrong, D.T. Concentrations of thromboxane B^ and 6-oxo-prostaglandin F ~ in amniotic fluid, decidual2tissue, and placenta in the ~ater stages of pregnancy in the rat. Adv. Prostaglandin Thromb. Res. 8: 1423-1429, 1980. 39.
Phillips, C.A. and Poyser, N.L. Studies on the ment of prostaglandins in implantation in the Reprod. Fert. 62: 73-81, 1981.
involverat. J.
40. Gimeno, M.F., Borda, E°S., Lazzari, M.A. and Gimeno, A.L. Ovarian hormones inhibit the release of prostacyclin-like material from isolated rat uterus. Prostaglandins 20: 223-232, 1980.
Editor: Harold R. Behrman Received: 11-16-81 Accepted: 3-24-82
674
MAY 1982 VOL. 23 NO. 5
EFFECT OF PROSTACYCLIN INHIBITION BY TRANYLCYPROMINE ON UTERINE 6 - K E T O - P G F . ~ LEVELS DURING ESTROGEN HYPEREMIA IN RA~S Farley, D.B. and Van Orden, D.E. Department of Obstetrics and Gynecology University of Iowa Iowa City, IA 52242 ABSTRACT A role for prostacyclin (PGIp) as a mediator of estrogeninduced increases in uterine blood-volume (UBV) was investigated by measuring uterine tissue levels of 6-keto-prostaglandin F (6-keto-PGF , ), and testing estrogen responses in rats 1 ~ pretreated with the PGI~, synthesis inhibitor, tranylcypromine (TCP). Uterine 6-keto-1#GFl~ content was determined by radioimmunoassay of tissue extracts purified through the use of high-performance liquid chromatography (HPLC). Estrogen treatment of castrate rats resulted in a significant increase of uterine 6-keto-PGFl~ as compared to saline treated controls (9.3 ng/uterine horn vs 6.7 ng/uterine horn, p=O.Ol). Pretreatment with TCP (20 mg/kg) markedly reduced the uterine content of 6-keto-PGF. (2.5 ng/uterine horn). The typical 50% Increase In UBV o~served after estrogen was unaffected by tranylcypromine pretreatment. It was concluded that the increased PGI 2 synthesis, as indicated by elevated levels of 6-keto-PGF. , may function as an amplifying mechanism for the uterlne vasodilation-induced by estrogen in castrate rats, but that production of this prostanoid is not essential for the estrogen response. o
•
.
IC~
INTRODUCTION Prostacyclin (PGIp) is a recently discovered vasoactive product of a r a c h i d o n i C acid metabolism, generated enzymatically from the endoperoxide, prostaglandin H~ (1,2). Because of its instability and short halflife in blological samples, endogenous PGI 2 can not be accurately measured. Therefore, to estimate PGI 2 in a biological system, many investigations have relied upon the measurement of 6-keto-PGF 1 , one of the stable but less active breakdown products (~). Using this approach, studies have focused on production of PGI^ in vari2 ous vascular beds (4). Synthesis has been demonstrated in large isolated blood vessel segments, vascular endothelial cells and, most recently, arterioles and capillaries (5-9). The uterus is a highly vascularized organ which has marked blood flow changes in response to hormonal stimuli
MAY 1982 VOL. 23 NO. 5
657
PROSTAGLANDINS (I0). The most dramatic vascular effects are observed following administration of estrogen (ii). The time delay between estrogen administration and onset of the vasodilatory response suggests that this estrogen action may involve a mediator. Several lines of evidence lead to the possibility that PGI 2 may play a role in estrogen-hyperemia. Earlier studies from our laboratories (12) which demonstrated that estrogen-hyperemia could be attenuated by cyclooxygenase inhibitors suggested that one of the products of arachidonic acid metabolism might be the mediator. Several other studies demonstrated that the major prostaglandin produced by the uterus was 6-keto-PGF~ , lq indicating that uterine tissue metabolizes arachidonic acld preferentially to PGIp (13,14). Since intra, arterial infusion of PGI 2 results in immediate increases in uterine blood flow, this prostanoid has been viewed as a potent uterine vasodilator (15,16). The present experiments were designed to test the involvement of PGI 2 in estrogen hyperemia using two separate approaches. TNe uterine content of 6 - k e t o - P G F during estroi gen hyperemia was measured using a specific ra~ioimmunoassay. Furthermore, the importance of prostacyclin synthesis to estrogen hyperemia was evaluated by testing uterine blood volume changes in response to estrogen in rats pretreated with tranylcypromine, a known inhibitor of PGI 2 formation in vitro (17). MATERIALS AND METHODS Solvents and Reagents: Tritiated 6-keto-PGF_~ was obtained from New England Nuclear, Corp. (Boston~ Mass.). Non-radioactive prostaglandins were kindly supplied by Dr. John Pike from the Upjohn Company (Kalamazoo, MI). Indomethaein was obtained from Merck, Sharp & Dohme Research Lab (Rahway, N.J.). Spectro-grade ethyl acetate and benzene, HPLC-grade acetonitrile, certified grade isopropanol, and reagent grade glacial acetic acid were obtained from Fischer Chem. Co. (Fairlawn, N.J.). HPLC-grade water was purchased from Alltech Assoc. (Arlington Heights, IL). Thyroglobulintype 1 and l-ethyl-3 (3 dimethyl amino-propyl) carbodiimideHCI (ethyl-CDl) were purchased from Sigma Chem. Co. (St. Louis, MO). Bovine gamme globulin, Fr. II (BGG) was obtained from Pel-Freez Biologicals (Rogers, Arkansas); polyethylene glycol 4000 was purchased from Union Carbide Co. (N.Y.,NY); Budget-Solve scintillation cocktail was obtained from Research Products International Corp. (Elk Grove Village, IL). For the radioimmunoassay, phosphate buffer (O.01 M, pH 7.4), containing 2 mg/ml bovine gamma globulin (PB-BGG) was used to re-dissolve tissue extracts after HPLC isolation of 6-keto-PGF 1 a n d for the dilution of antibody, cold standards and tracer. Drugs: 17B-estradiol-3-benzoate (E2B ; Sigma Chem. Co., St. Louis, MO) was dissolved in corn oil at a concentration of O.14 mg/ml. 17B-estradiol (E2; Sigma Chem. Co., St. Louis,
658
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS MO) was dissolved in 95% ethanol and diluted with phosphatebuffered saline (0.001 M phosphate buffer, pH 7.4, 0.15 M NaCI) to a concentration of 0.5 ug/ml. E 2 vehicle consisted of phosphate-buffered saline containing an amount of ethanol equal to that found in the estradiol solution. Tranylcypromine sulfate (Smith, Kline, & French Labs, Philadelphia, PA) was prepared at a concentration of 20 mg/ml in saline. To block prostacyclin synthesis during UBV studies, TCP was given (20 mg/kg BW, i.p.) 30 minutes prior to the injection of E 2 or E 2 vehicle. 6-keto-PGF I ~ radioimmunoassay: Antibody was produced to 6-keto-PGF _ -thyroglobulin conjugates prepared by the carbodiimide condensation method (18). Briefly, 5 mg 6 - k e t o - P G F 1 ~ dissolved in 1 ml acetonitrile was added dropwise to i ~ml distilled water containing 10 mg thyroglobulin and 7.5 mg ethyl-CDl. The reaction was gently stirred at 4 ° C overnight. After one hour of dialysis against cold water the conjugate was lyophilized. New Zealand white rabbits were immunized intradermally with various doses (i00 ug-5 mg) of c o n j u g a t e dissolved in 0.15 NaCI and emulsified with complete Freund's as previously described (19). Rabbits were boosted i.v. on a monthly basis with 1/10 the initial dose of conjugate dissolved in saline. Blood was collected into chilled, heparinized containers 1 week after each boost and centrifuged immediately to harvest the antiplasma. The RIA was carried out in 12 mm x 50 mm polypropylene test tubes (Sarstedt, Inc. d Princeton, NJ). Unlabeled 6-ketoPGF~ ~ was stored at -20- C in acetonitrile (i00 ug/ml) and di1~ted in PB-BGG to yield solutions of 10 ng/ml and 1 ng/ml at the time of assay. Standard curve tubes (2 pg-160 pg) were aliquoted in triplicate using an automatic pipetting station (Micromedic Systems, Inc., Horsham, PA). Tissue samples from the HPLC step were re-dissolved with 0.5 ml PB-BGG and delivered to assay tubes in three different volumes. The volume of each tube was simultaneously adjusted to 200 ul with PB-BGG by means of the rinse pump on the pipetting station. A dilution ~f antiplasma previously shown to bind 40% of 5000 cpm's of H - 6 - k e t o - P G F l ~ was prepared in PB-BGG and 50 ul was added to each tube; non-specific binding tubes received 50 ul of PB-BGG in place of antibody. Labeled 6 - k e t o - P G F 1 ~ (5000 cpm's in ~ ul PB-BGG) was added and the entire assay was incubated at C overnight. Precipitation of antibody-bound 6-keto-PGF1~ was accomplished by the addition of 0.2 ml PB-BGG and 0.5 ml polyethylene glycol 4000 (40% w/v) as previously described (18). After decanting the supernatant, the pellet was analyzed for bound-radioactivity by re-dissolving each with 0.i ml 0.I N acetic acid and then adding 3 cc of Budget-Solve. All test tubes were counted to i0,000 cpm's and the data calculated on a PRIME computer using Rodbard's program which provides weighted standard curves and assessment of potency and parallelism of unknown and standard (20). Uterine tissue sample data were corrected for procedural recovery previously determined to be
MAY 1982 VOL. 23 NO. 5
659
PROSTAGLANDINS 45
•
5°/~ w i t h 3 H - 6 - k e t o - P G F ° i ~ " Tissue extraction and HPLC
isolation
o~
6-keto-PGF
i~:
Uterine horns, previous l~ soaked in indomethacin solution (20 ug/ml) and stored at -20-- C, were weighed, immersed in liquid nitrogen, pulverized, and quickly transferred to glass extraction tubes (12 x i00 mm), held in an ice bath. Extraction of prostaglandins was carried out as previously described (19). Briefly, one milliliter 0.05 M citric acid containing 0.15 M NaCI per I00 mg of tissue was used as the tissue diluent; tissue suspensions were sonicated for 20 minutes at 4 ° C followed by extraction with two volumes of ethyl acetate: isopropanol (i:i, v:v). Three milliliters of 0.15 M NaCI and two milliliters of additional ethyl acetate were added• The resulting suspensions were centrifuged at 15OO x g for IO minutes• All procedures were carried out at 4 ° C. The entire organic phase was transferred to a 12 mm x 75 mm glass tube and evaporated under vacuum using a Speedvac concentrator (Savant Instruments, Inc., Hicksville, NY). The dried extracts were stored at -20 ° C until chromatographed. In the ~xperiments designed to assess tissue extraction efficiency, H-6-keto-PGF I ~ was added to the tissue suspension and the entire extract counted in Budget-Solve. Each tissue extract was re-dissolved in 200 ul HPLC solvent (water:acetonitrile:benzene:acetic acid, 76.7%:23.0%:0.2%: 0.1%), briefly vortexed and sonicated. S a m p l e s were freed of suspended matter via a microfiltration apparatus (Bioanalytical Systems, Inc., West Lafayette, IN) using a 1.0 u teflon membrane initially• After a second microfiltration using a 0.2 u regenerated cellulose membrane, a measured volume (150-200 ul) of the clear extract was resolved on a 30 cm x 3.9 mm-i.d, reversed phase HPLC column (Fatty Acid Analysis Column, Waters Assoc., Milford, Mass.) as previously described (21). The flow rate was 2 ml/min/tube; fifteen fractions were collected in minivials. Recovery of 6-keto-PGF I ~ from the HPLC was established in preliminary experiments utilizing labeled 6-keto-PGFl~ as well as trace-labeled tissue extracts• When uterine extracts were chromatographed to isolat~ the fraction containing 6keto-PGF 1 ~, a sample of pure H-6-keto-PGFl~ was injected after every two or three extracts to affirm that the elution profile of the 6 - k e t o - P G F ~ was unchanged• The column was stripped with filtered methanol between each injection to remove residual prostaglandins. The fraction containing 6keto-PGF 1 ~ was taken to dryness in a Speedvac concentrator and stored at -20 ° C until the time of radioimmunoassay. Hyperemia Model: Twenty-nine female Sprague-Dawley rats (Bio-Labs, Madison, WI) weighing between 175-200 gm were prepared for testing of uterine vascular responses to estrogen according to the "uterine hyperemia" protocol previously established in this laboratory (22). On day O, each rat was ovariectomized via the dorsal approach; on day 7, each animal received a subcutaneous injection (O.i ml/lO0 mg BW) of E2B
660
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS dissolved in corn oil. On day 14, fourteen of the animals were pretreated with TCP (20 mg/kg, i.p.). The remaining fifteen animals (TCP controls) received saline at the same time. Thirty minutes after administration of TCP or saline, half of the animals in each group were given E_ (0.5 ug/kg) into the femoral vein under light ether anesthesia~ the remaining animals in each group received E 2 vehicle. Two hours following injection of E 2 or E^ vehlcle, all animals were again briefly anesthetized with ether and the 13clontralateral femoral vein was injected with i0 uCi of I-human serum albumin (RISA) (E.R. Squibb & Sons, Inc., Princeton, NJ) in 0.i saline. After an equilibration time of five minutes the animals were sacrifificed, the abdomen was opened and blood was withdrawn from the inferior vena cava. The uterus was rapidly excised and trimmed of adherent fat and mesometrium. One horn was used for blood volume determination and the other horn was processed for 6 - k e t o - P G F ~ t~ssue content. A section of small intestine was a~so removed and used for control intestinal blood volume determination. Blood volumes were calculated on a dry weight basis from the radioactivity of the tissues relative to that of the blood. The uterine horns to be analyzed for 6-keto-PGF~ ~ were soaked in a solution of indomethacin (20 ug/ml) ~reshly prepared in phosphate buffer (0.i M, pH 7.4) by gently heating and stirring. After approximately five minutes, these tissues were transferred to an aluminum foil wrapper and stored at -20 ° C until pulverization and extraction. Blood Pressure Determinations: Blood pressure in conscious animals was determined by means of a PE iO cannula inserted into the descending aorta via the left carotid artery. Pressure changes were monitored with an Alltech (City of Industry, CA) pressure transducer which was coupled to a Beckman (Irvine, CA) R611 dynograph. Blood pressure and heart rate were recorded at 30 minute intervals and animals returned to housing cages between monitoring periods. Statistical Analysis: Blood volumes were compared using a one way analysis of variance with the level of significance being p < 0.02. Tissue contents of 6 - k e t o - P G F l ~ w e r e compared by the same method of analysis.
RESULTS Radioimmunoassay characteristics: All rabbits immunized with 6 - k e t o - P G F l ~ conjugates developed antibodies after the first boost. In general, higher titre antibodies were produced when initial immunization was with larger amounts (5 mg) of conjugate; low titre antibodies yielding assay sensitivity in the femtogram range were obtained when low dose (IO0 ug) immunization was employed. The antibody used in the present study was obtained after % months of immunization and bound 43% of 2.5 pg (5000 cpm) H-6-keto-PGF 1 ~ at a final anti-
MAY 1982 VOL. 23 NO. $
661
PROSTAGLANDINS
plasma dilution of 1:1200. A typical standard curve is illustrated in Fig. 1 (sensitivity=l.7pg; ED =24 2pg; slope=50 " -1.00667; correlation coefficient:-0.993; range=2-160 pg).
1°°f 80
60
=o 40
20
01
4
16
40
Picograms 6-Keto-Prostaglandin F 1~
Fig. i: 6-keto-PGF. standard curve with antiplasma R132-7/2/80 and 3H-6-PGFI~ . I~
662
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
TABLE
I:
SPECIFICITY 7/2/80
OF
THE
6-KETO-PGF l a
ANTIPLASMA
R132-
Relative crossreaction (%) at 50% 3displace ment of H-6-ketoPGF 1
Compound
6-keto-PGF l a PGE 1 PGE 2 PGF 1 a PGF~
i00.0 4.6 4.9 7.6 5.7
TxB~ a PGD 2
<0.1 <0.1
PGA. PGA~ PCB~ 13,~4-DIHYDRO-15-KETO 15-KETO PGE_ 13,14-DIHYD~O PGE^ 13,14-DIHYDRO-15-~ETO 13,14-DIHYDRO PGF 2 a
<0.i
PGE 1
PGF 2
The specificity of the assay is shown in Table I. Only the primary PGF's and PGE's showed any appreciable crossreactivity. Within-assay coefficient of variation was 7.5% at a concentration level of 1-5 ng/uterine horn (n=29). Interassay coefficient of variation was determined for eight uterine samples falling in a concentration range of 0.7-5.0 ng/ horn and was 8.7%. The RIA recovery of unlabeled 6 - k e t o - P G F l ~ (50-1000 pg) added to 2 ml of column solvent (concentrated and re-dissolved in protein buffer for aliquoting) averaged 82% with a correlation coefficient of 0.9982 (n=9). The solvent blank was 12.7 pg/sample. Qualitative validity of the 6-keto-PGF 1 ~ RIA was indicated by the markedly reduced tissue levels of 6-ket°-PGF 1 a measured in uteri obtained from rats treated with tranylcypromine, a known inhibitor of PGI 2 synthesis (3) (See Fig. 4 below). Ti@sue extraction and isolation efficiency: Recovery of added--JH-6-ket0~PGF, a from uterine tissue suspensions wasi J 59.4 ~ 1.6% (n=12). No loss in concentration of H-6-ketoPGF.~ was incurred during the filtration step prior to HPLC analysis. The --H-6-keto-PGF. ^ chromatographed as a single peak (Fig. 2) at a retention time of 9 minutes with a column
MAY 1982 VOL. 23 NO. 5
663
PROSTAGLANDINS recovery of 76.6 + 2.1% (n=13). The reproducibility of the rete~ntion time is ~ i s o shown in Fig. 2. The elution pattern of -H-6-keto-PGF 1 ~ was not altered either by presence of uterine tissue extracts or by stripping the HPLC column with methanol after each tissue extract run. Testing of the HPLC fractions with antibody to 6-keto-PGF 1 ~ showed the peak radioactivity fraction to be fully immunoreactive.
Conditions: Column:
25,000 E
Mobile Phase:
F a t t y Acid Analysis, 1 0 p.m 3 . 9 x 3 0 i~m 76.7%: 2 3 . 0 % : 0 . 2 % 0 . 1 % , water: acetonitrile: b e n z e n e : a c e t i c acid
Flow Rate:
2 ml/min
A
20,000 15,000
0 0
•~
10,000
o .£
5,000
~
0 0
I
I
I
I
I
I
I
5
10
15
0
5
10
15
RETENTION TIME IN MINUTES Fig. 2: Retention time in minutes of 3H-6-keto-PGF.~ detected by quantity of radioactivity: A) Initial resolution; B) Resolution after chromatography of three tissue extracts.
Effect of tranylcyPromine on systemic hemodynamics: Blood pressure and heart rate changes observed in four conscious rats following TCP administration (20 mg/kg, i.p.) are presented in Table II. Although blood pressure was elevated at 30 minutes after TCP, it was not statistically different from baseline at 2½ hours, the time after drug treatment at which animals were sacrificed in the uterine hyperemia experi-
664
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
ment. Heart rate remained elevated throughout the experiment. TCP-treated rats displayed hyperactivity, salivation and piloerection over the 150 minutes of observation. TABLE
II:
EFFECT OF TRANYLCYPROMINE (20 mg/kg) PRESSURE (BP) AND HEART RATE (HR) OF RATS
ON BLOOD CONSCIOUS
TIME AFTER TCP-TREATMENT ere~reatment
BP (mmHg)
HR (beats/ min)
30 min,
60 min.
90 min.
120 m i n
150 min.
107+6
135+6
127+16
118+15
115+15
114+13
356+15
428+30
420+20
404+20
423+23
438+14
Values
represent mean + SEM
Effect of tranylcypromine on uterine 6-keto-PGF and is-blood volume: As shown in Fig. 3, animals tested for uterine vascular responses to estrogen demonstrated the typical 50% increase in blood volume two hours after steroid administration (23). UBV of saline control animals was 217 + 18.1 ug/g, whereas that of estrogen treated controls was 3~6.8 + 25.5 ul/g. Estrogen treatment of these castrate rats also resulted in significant elevation of uterine 6-keto-PGF~ content as shown in Fig. 4. At two hours, content of 6-~eto-PGF. ~ in uteri of saline treated controls was 6.7 ~ 0.5 ng~horn; whereas, estrogen treated animals had 9.3 + 0.8 ng/horn. Pretreatment with the TCP 30 minutes before estrogen administration resulted in a markedly reduced tissue content of 6-keto-PGF~ (2.5 ! 0.3 ng/horn) as shown in Fig. 4. NevertheIn less, TCP did not prevent the estrogen-induced increase in UBV (376.6 ! 32.1 ul/g), nor did it alter saline UBV (265.1 ~ 22.6 ul/g) (Fig. 3). Although not shown, intestinal blood volumes were not altered either by estrogen administration or TCP pretreatment.
MAY 1982 VOL. 23 NO. 5
665
PROSTAGLANDINS
[~ Saline [ ] Estrogen (0.5p.g/kg) 400
e
(7)
m
(7)
300
(7)
=m
I-
(8)
t-
D
N N N
200
E
N
0
II1 ~
,
l
=m t,=
100
Control
Tranylcypromine
Fig. 3: Uterine blood volume (determined on a dry weight basis) reponse to estrogen or saline in control or t r a n y l c y p r o mine treated animals. Each bar represents the mean + SEM; the number of animals in each group is shown in parentheses. Asterisks indicate values that are s i g n i f i c a n t l y different from their respective saline controls at P <0.02.
666
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
t"
m
L_
(11)
o 10 "1" c L_
8 :3
LL 0
(13)
T
6
n
I
o
,4--1
4
t
(1)
(8)
I
©
{31 C
2
0
Saline
Estrogen
Tranylcypromine & Estrogen
Fig. 4: Uterine 6-keto-PGFl~ content two hours after saline, estrogen or tranylcypromine + estrogen treatment. Asterisk indicates value significantly different from saline treated animals at P < 0.02. Daggar indicates value significantly different from saline treated and estrogen treated animals at P < 0.001.
MAY 1982 VOL. 23 NO. 5
667
PROSTAGLANDINS DISCUSSION
A number of reports have appeared on the measurement of 6-keto-PGFl~ in various biological samples as an index of prostacyclln involvement (24-26). The radioimmunoassay described in the present study employing HPLC isolation of 6-keto-PGF.~ from tissue extracts fulfills the requirements of high sensitivity, specificity and accuracy. A sensitivity of less than two picograms observed with the antibody used for this assay compares favorably with that reported for several other RIA's for this PGI 2 metabolite (24,27,28). The use of low doses (iO0 ug) of immunogen may explain the production of specific and avid antisera by these animals within a few weeks of immunization. Vaitukaitis et al found similar results following a primary immunization with as little as 20 ug of HCG subunits and testosterone-BSA conjugates (29). In addition, we have found that the conjugation reaction is highly efficient (molar ratios greater than iO00:i) when thyroglobulin is used instead of albumin as the carrier protein (unpublished observation). Crossreactivity of this antibody compares well with that reported by Salmon (27). Other investigators have described production and use of antibodies that have limited cross-reaction (less than 1.0%) with primary prostaglandins (24,30,31). Nevertheless, through the use of HPLC purification of tissue extracts in the present study, interference by other arachidonic acid metabolites is minimized in the final assay of sample. In the course of defining the best method of isolating 6-keto-PGF 1 ~ from biological samples, various prostaglandin plasma extraction methods were explored (cyclohexane:ethyl acetate, I:i; repeated volumes of ethyl acetate; washing of organic extracts with water; extraction of acetone precipitates of plasma; preliminary lipo-protein depletion of plasma; extraction at various pHs). None of these specialized methods thought to yield better extraction efficiency of 6-keto-PGF. improved the 73% recovery obtained with the standard method ~ routinely use for other prostaglandins (acidification with citric acid and a single extraction with ethyl acetate after removal of neutral lipids with petroleum ether). Because of the above findings and the reproducible extraction of tissue homogenates using the method already established in this laboratory (18) and used by many others, we did not pursue alternate tissue extraction methods. Pulverization of frozen tissue, as opposed to homogenization using ground glass tissue grinders, was preferred because the rapidity of the method minimizes the possibility of generation of prostaglandins during tissue processing. The inclusion of a synthesis inhibitor at this step was an added assurance that levels reflected true endogenous content. In preliminary attempts to isolate 6-keto-PGFl~ via stan-
668
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
~ ard
silicic acid chromatography we were unable to isolate H-6-keto-PGF~ in any one fraction to an extent greater than 1 50%. Other authors have used either TLC (27) or bulk collections of all prostaglandin classes from silicic acid columns (30) in order to purify 6-keto-PGF_ from sample extracts. With applications of HPLC coupled ~i~h RIA for determination of many cyclooxygenase products (21), it is possible to isolate endogenous 6 - k e t o - P G F ~ in a relatively small volume. As reported in the present study, HPLC resolution of 6-keto-PGFl~ is highly reproducible, unaffected by presence of large quantities of prostaglandins as found in uterine tissue and requires only 10-15 minutes for elution of the 6-keto-PGFA peak. All other arachidonic acid metabolites are stripped ~r~m the column by the methanol injection permitting HPLC isolation of 6-keto-PGFl~ from approximately 20 samples per eight hour period. Although the procedural loss of 6-keto-PGF~ during is extraction of tissue and HPLC isolation is rather substantial (55%), the high concentration of endogenous 6-keto-PGF.~ (i00200 pg/mg) in this vascular organ eliminates any problem of assay sensitivity. Furthermore, by assaying samples freed of other prostaglandins, one may have greater confidence that inhibition of antibody-binding is due to 6-keto-PGFl~ in the sample rather than to high concentrations of crossreacting prostanoids. If assay of 6-keto-PGF l ~ is to be used as a measure of prostacyclin production, one must show reduced tissue levels of the stable metabolite in animals pretreated with specific PGI 2 synthesis inhibitors. Such was the case in the study presented here as TCP treatment reduced the tissue content to one-fifth basal levels. Although the original evaluation by Gryglewski, et al (17) suggested that TCP is a selective inhibitor of PGI 2 synthesis in vitro, some recent work has raised questions concerning its specificity (32-34). Nevertheless, the in vivo use of TCP in other biologic systems known to involve prostaglandins indicate that it does block PGI 2 formation. Jonsson, et al (35) used the same dose in mice as was used in the present study and found uterine levels of PGE and PGF to be unaffected. Unfortunately they did not measure 6-keto-PGF ~ . Likewise, the platelet aggregation study of Rosenblum land El Sabban (36) using i0 mg/kg TCP in mice strongly suggests that TCP acts to inhibit PGI^ synthesis. Both studies eliminated the monoamine oxidase (M~O) inhibitory properties of the drug as an explanation for their results by showing that other MAO inhibitors had no effect on the response under study. Furthermore, the ID50 " of TCP as a MAO inhibitor at a dose of 6 mg/kg in rats requlres 16 hours (32); hence, drug related changes occurring in the 2½ hour period of the present study would not be on that basis. Elevation of uterine 6-keto-PGFl~ following acute estrogen administration to castrate rats parallels our earlier •
MAY 1982 VOL. 23 NO. 5
669
PROSTAGLANDINS
observation of increased uterine PGE in estrogen hyperemia (22). The tissue levels of 6 - k e t o - P G F ~ in control animals of the present study are approximately ~.5 times greater than that found for the other potent uterine vasodilator prostaglandin, PGE. Although few studies exist concerning uterine 6-keto-PGF~ ~ concentrations in any species (28,37-39), that of Poyser I and Scott (28) reported very low tissue levels (15-38 pg/mg) of this metabolite in rat uterine homogenates. These authors also found a high PGI^ synthesizing capacity for 2 the rat uterus, with both concentration and production of 6-keto-PGF~ higher on the day of estrus than on diestrus. Although t~e relative amount of 6-keto-PGF~ measured does not compare well with the content reported in the present study, the positive influence of estrogen is similar. In contrast, Gimeno, et al, (40) looking at production of PGI 2like substance by uteri from estrous rats or ovariectomized rats treated in vivo with high doses of estradiol, associated estrogen with an inhibitory action on the uterine PGI^ synthe2 tase system. The time after estrogen administration (24 hours) chosen by these authors for study of PGI^k production is a time when estrogen hyperemia is known to have subsided. Therefore, if estrogen does indeed have a positive effect on PGI_ as we indicate by the increased quantity of 6-keto-PGF~ found in estrogen stimulated uteri, it is an early effecl. ~ Although the present study measures content and that of Gimeno, et al, measures production, one would expect the effect of estrogen to be similar if PGI 2 synthesis by the uterus is a consequence of estrogen action. The observation, in the present study, that UBV responses to estrogen persist after administration of TCP to rats, indicates that PGI^ synthesis is not essential for this estroz gen response. Although PGI 2 has repeatedly been shown to be a uterine vasodilator and uterine levels of 6-keto-PGF~ are 1 elevated at the time of maximal response, this prostanold can not be considered to fulfill a mediator role. Since UBV increased normally after estrogen administration to TCP treated rats, we must conclude that the increased PGI 2 synthesis during the response is not obligatory for this estrogen vascular action. It is more likely that this concomitant production of PGI Z functions as an amplifying mechanism for the vasodilation initiated by the primary estrogen mediator. REFERENCES i.
Moncada, S., Gryglewski, R., Bunting, S., and Vane, J.R. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 263: 663-665, 1976.
2.
Johnson, Mcguire,
670
R.A., Morton, D.R., Kinner, J.H., Gorman, R.R., J.C. , Sun, F.F. , Whittaker, N. , Bunting, S. ,
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
Salmon, J.A., Moncada, S., and Vane, J.R.: structure of prostaglandin x (prostacyclin). ins 12: 915-928, 1976.
The chemical Prostagland-
3.
Pace-Asciak, C. Isolation, structure and biosynthesis of 6-keto-PGF is in the rat stomach. J. Am. Chem. Soc 98: 2348-2349, 1976.
4.
Moncada, S., and Vane, J.R. vascular tissue. Fed. Proc.
5.
Moncada, S., Higgs, E.A., and Vane, J.R. Human arterial and venous tissues generate prostacyclin, a potent inhibitor of platelet aggregation. The Lancet i: 18-20, 1977.
6.
Botha, J.H., Leary, W.P., and Asmal, A.C. Enhanced release of PGI^-like substance in experimental hypertension. Prost. and Mefdicine 3: 251-252, 1979.
.
The role of prostacyclin 38: 66-71, 1979.
in
MacIntyre, D.E., Pearson, J.D., and Gordon, J.L. Localization and stimulation of prostacyclin production in vascular cells. Nature 271: 549-551, 1978.
8.
Levine, L., and Alam, I. Arachidonic acid metabolism by cells in culture: Analysis of culture fluids for cyclo-oxygenase products by radioimmunoassay before and after separation by high pressure liquid chromatography. Prost. and Medicine 3: 295-304, 1979.
9.
Maurer, P., Moskowitz, M.A., Levine, L., and Melamed, E. The synthesis of prostaglandins by bovine cerebral microvessels. Prost. and Medicine 4: 153-161, 1980.
iO. Reynolds, S.R.M. Physiology of the uterus, York, 1949. Paul B. Hoeber, Inc. 297-307.
ed.
2, New
ii. Greiss, F.C., Jr., and Anderson, S.G. Effect of ovarian hormones on the uterine vascular bed. Am. J. Obstet. Gynecol. 107: 829-836, 1970. 12. Ryan, M.J., Clark, K.E., Van Orden, D.E., Farley, D.B., Edvinsson, L., Sjoberg, N.O., VaN Orden, L.S. III and Brody, M.J. Role of prostaglandins in estrogen-induced uterine hyperemia. Prostaglandins 5: 257-268, 1974. 13. Fenwick, L., Jones, R.L., Naylor, B., Poyser, N.L., and Wilson, N.H. Production of prostglandins by the pseudopregnant rat uterus, in vitro, and the effect of tamoxifen with the identification of 6-keto-prostaglandin FI~ as a major product. Br. J. Pharmac. 59: 191-199, 1977.
MAY 1982 VOL. 23 NO. 5
671
PROSTAGLANDINS
14. Jones, R.L., Poyser, N.L., and Wilson, N.H. Production of 6-oxo-prostaglandin F ~ by rat, guinea-pig, and sheep uteri in vitro. Br. J~ Pharmac. 59: 436-437, 1977. 15. Resnick, R., and Brink, G.W. Uterine vascular response to prostacycli n in nonpregnant sheep. Am. J. Obstet. Gynecol. 137: 267-269, 1980. 16. Clark, K.E., Mills, E.G., Stys, S.J., and Seeds, A.E. Effects of vasoactive polypeptides on the uterine vasculature. Am. J. Obstet. Gynecol. 139: 182-188, 1981. 17. Gryglewski, R.J., Bunting, S., Moncada, S., Flower, R.J., and Vane, J.R. Arterial walls are protected against a substance (prostaglandin x) which they make from prostaglandin endoperoxides. Prostaglandins 12: 685-713, 1976. 18. Van Orden, D.E. and Farley, D.B. Prostaglandin F2~radioimmunoassay utilizing polyethylene glycol separation technique. Prostaglandins 4: 215-233, 1973. 19. Van Orden, D.E., Farley, D.B., and Clancey, C.J. Radioimmunoassay of PGE and an approach to the specific measurement of PGE I. Prostaglandins 13: 437-453, 1977. 20. Rodbard, D., and Faden, V.B. National Technical tion Service, Springfield, VA, 22151.
Informa-
21. Alam, I., Ohuchi, D., and Levine, L. Determination of cyclooxygenase products and prostaglandin metabolites using high-pressure liquid chromatography and radioimmunoassay. Anal. Biochem. 93: 339-345, 1979. 22. Clark, K.E., Baker, H.A., Bhatnagar, R.K., Van Orden, D.E., and Brody, M.J. Prevention of estrogen-induced uterine hyperemia by ~-adrenergic receptor-blocking agents. Endocrinol. 102: 903-909, 1978. 23. Van Orden, D.E., Clancey, C.J., and Farley, D.B. Uterine serotonin and receptor blockade during estrogen-induced uterine hyperemia. Proc. Soc. Exp. Biol. Med. 167: 469-474, 1981. 24.
Oliw, E. A radioimmunoassay for 6-keto-prostaglandin FI~ utilizing an antiserum against 6-methoxine-prostaglandln F1 ~ Application to study renal in vitro synthesis of prostacyclin. Prostaglandins 19: 271-284, 1980.
25.
Abdel-Halim, M.S., Lunden, I., Cseh, G. and Anggard, E. Prostaglandin profiles in nervous tissue and blood vessels of the brain of various animals. Prostaglandins 19:
672
MAY 1982 VOL. 23 NO. 5
PROSTAGLANDINS
249-258, 1980. 26.
Mitchell, M.D., Keirse, M.J.N.C., Brunt, J.D., Anderson, A.B.M. and Turnbull, A.C. Concentrations of the prostacyclin metabolite, 6-keto-prostaglandin F 1 a in amniotic fluid during late pregnancy and labour. Br. J. Obstet. and Gynecol. 86: 350-353, 1979.
27. Salmon, J.A. A radioimmunoassay for 6-keto-prostaglandin F1 ~ Prostaglandins 15: 383-397, 1978. 28. Poyser, N.L. and Scott, F.M. Prostaglandin and thromboxane production by the rat uterus and ovary in vitro during the oestrous cycle. J. Reprod. Fert. 60: 33-40, 1980. 29.
Vaitukaitis, J., Robbins, J.B., Nieschlag, E. and Ross, G.T. A method for producing specific antisera with small doses of immunogen. J. Clin. Endocr. 33: 988-991, 1971.
30. Mitchell, M.D. A sensitive radioimmunoassay for 6-ketoPGFI~ : Preliminary observations on circulating concentrations. Prost. and Medicine i: 13-21, 1978. 31.
Ylikorkala, O. and Viinikka, L. Measurement of 6-ketoprostaglandin F I~ in human plasma with radioimmunoassay: Effect of prostacyclin infusion. Prost. and Medicine 6: 427-436, 1981.
32.
Rajtar, G. and de Gaetano, G. Tranylcypromine is not a selective inhibitor of prostacyclin in rats. Thromb. Res. 14: 245-248, 1979.
33.
Chignard, M., Vargaftig, B.B., Sors, H. and Dray, F. Synthesis of thromboxane B_ in incubates of dog plateletZ rich plasma with arachidonic acid and its inhibition by different drugs. Biochim. Biophys. Res. Commun. 85: 16311639, 1978.
34. Hong, S.L., Carty, T. and Deykin, D. Tranylcypromine and 15-hydroperox~-arachidonate affect arachidonic acid release in addition to inhibition of prostacyclin synthesis in calf aortic endothelial cells. J. Biol. Chem. 255: 9538-9540, 1980. 35. Jonsson, H.T., Rankin, J.C., Ledford, B.E. and Baggett, B. Uterine prostaglandin levels following stimulation of the decidual cell reaction: Effect of indomethacin and tranylcypromine. Prostaglandins 18: 847-857, 1979. 36. Rosenblum, W.I. aggregation by
and Ei-Sabban, F. Enhancement of platelet tranylcypromine in mouse cerebral micro-
MAY 1982 VOL. 23 NO. 5
673
PROSTAGLANDINS
vessels.
Circ. Res. 43: 238-241,
1978.
37. Alwachi, S.N., Bland, K.P. and Poyser, N.L. Production of 6-oxo-prostaglandin F , PGF^ , and PGE_ by sheep uterine tissue. Prost. and M ~ i c i n e ~: 329-332, ~980. 38. Zamecnik, J., Kennedy, T.G. and Armstrong, D.T. Concentrations of thromboxane B^ and 6-oxo-prostaglandin F ~ in amniotic fluid, decidual2tissue, and placenta in the ~ater stages of pregnancy in the rat. Adv. Prostaglandin Thromb. Res. 8: 1423-1429, 1980. 39.
Phillips, C.A. and Poyser, N.L. Studies on the ment of prostaglandins in implantation in the Reprod. Fert. 62: 73-81, 1981.
involverat. J.
40. Gimeno, M.F., Borda, E°S., Lazzari, M.A. and Gimeno, A.L. Ovarian hormones inhibit the release of prostacyclin-like material from isolated rat uterus. Prostaglandins 20: 223-232, 1980.
Editor: Harold R. Behrman Received: 11-16-81 Accepted: 3-24-82
674
MAY 1982 VOL. 23 NO. 5