Rapid block of gonadotropin uptake by corpora lutea in vivo induced by prostaglandin F2α

PROSTAGLANDINS

RAPID ~ CORPORA ~

OF C43NADOTRfPIN UPTAKE BY IN VIVO INDUCED BY PROSTAGLANDIN F2u

HABfX~ P~ ~

AND M~R~IN HICH~qS

Department of C~stetrics and Gynecology and Pharmacology Yale University School of Medicine New Haven, Connecticut 06510 and Department of Reproductive Biology Merck Institute for Therapeutic Research Rahway, New Jersey 07065 ABST~CT Intravenous adninistration of 125i_h(~ to 7-8 day pseudcpregnant rats resulted in maximun uptake of radioactivity to corpora lutea 2 hours after treatment. At this time tissue/plasma radioactivity ratios oa an equal weight basis w~re: corpora lutea, 70.2 + 12.8; ovarian interstiti~n, 4.6 _+ 0.2; kidney, 2.2 + 0. i. No appreciable uptake was seen by adrenals cr liver. Radioactivity in corpora lutea was associated primarily with membranes WPich sedimented at 2000g and when released by heat it was more than 63% bound to luteal LH receptor preparation in vitro. Radioactivity in renal tissue was associated primarily with the i00,000g supernatant fraction and was board less than 1% to luteal LH receptcrs in vitro. PGF2e significantly reduced uptake (p<.001) of 125I-hCG by corpora lutea within 30 minutes (-63%) as well as at 1 (-64%), 2 (-75%), 4 (-68%) and 24 hours (-85%). No clear effect of PGFza on uptake of 125I-hCg by ovar/an interstitial tissue was seen. Plasma progesterone was significantly decreased (p<.001) within 30 minutes (-47%; p<.01) after PGF2a treatment and also at 1 (-65%), 2 (-82%), 4 (-68%) and 24 hours (-92%). Two hours after PGF2~ treatment the oontent of progesterone in corpora lutea was depressed (-46%; p<.001). It is suggested that the rapid inhibition of luteal progesterane productic~ ~ u c e d by PGF2= in v i ~ occurs through a block in gcnadotropin uptake by corpora lutea. A C K N ~ The authors a~knowledge the skillful techDical assistanoe of Jobanna Wittreich, Charles Carroll and Ruth Gu~nann.

JULY 1976

VOL. 12 NO. 1

83

PROSTAGLANDINS

INTRODUCTION

There is general agreement that prostaglandins cause loss of function of the corpus luteum in many species. The mecb~nisn by which luteolysis occurs is not known but available data supports the conclusion that the action of PGF2~ is directly on the corpus lute~n (i). It has been suggested that PGF2a causes luteolysis by interfering with blood flow to the ovary (2) or corpus luteum (3,4) but this is difficult to reconcile with data showing no change in ovarian (5) or luteal (6) blood flow when progesterone secretion was depressed and other data where PGF2e was shown to decrease progesterone output by luteal tissue in vitro (7,8,9,10). Recent reports frcm this laboratory have shown that PGF2e causes a rapid drop in seman progesterone in systemic blood of pseudopregnant rats within t~D hours of treatment (ii) and ovarian progesterone secretion as early as 30 minutes (5). Although a marked drop in LH receptor content of rat corpora lutea was seen 24 hours following PGF2~ treatment in vivo (12,13), ser~n progesterone was significantly depressed ~ t ~ two hours and this early effect of PGF2e was not acampanied with any change in apparent quantity of luteal LH receptors (ii). The purpose of the present study was first to validate an in vivo assay for corpus luteum uptake of labelled human chorionic gonad-otropin (hCG) and, second, to examine uptake of gonadotropin at various times following a luteolytic dose of PGF2e to determine if the early fall in progesterone output produced by PGF2e may be related to an early loss in goD~dotropin uptake in vivo.

MATER/ALS AND METHODS

Animals Immature, 28 day old, female rats (Charles River, CD strain) were treated with pregnant mare's ser~n (4 i.u.; s.c.; Gestyl; Organon) and cervically stimulated 72 hours later to induce pseudopregnancy. Seven days following cervical stimulation the animals were randcmly allotted to treatments. 12SI-hCG was administered by tail vein injection (0.2 cc) to animals under ether anaesthesia which were later sacrificed at times indicated in the text. At sacrifice the animals were anaesthetized with ether, 4 ml blood was drawn from the inferior vena cava into a heparinized syringe, various organs as shown in the text were removed, and the animals allowed to expire after opening the thoracic cavity. The radioactivity in trimmed and weighed tissues was determined in a gamma spectrometer (Packard I n s e t s ) . Corpora lutea were carefully dissected frc[n the ovary and the remainJmg ovarian tissue was designated as interstiti~n. Occasional contamination of the interstiti~n by r~nnants of luteal tissue may have occurred.

84

JULY 1976

VOL. 12 NO. I

PROSTAGLANDINS

Preparation of 12 Si_hCG As described earlier (12), hCG was purified from Pregnyl (Organon, East Orange, New Jersey) and iodinated using lactcperoxidase. The material useH in the present experiments had a specific activity of 30 ~Ci/~g, determined by ccmparing binding data for labelled hCG in the presence of increasing quantities of unlabelled hCG and for increasing quantities of labelled hCG alone, using a crude receptor preparation from superovulated rat ovaries (12). Characterization of Bound 12 Sl_hCG To verify that the radioactivity bound to tissue was hCG, tissues were homogenized in buffer (40 mM tris HCI, 5 ram MgSO4, pH 7.5), and centrifuged as indicated in the text. To r~uove bound ]2SI-hCG, the tissue was incubated in 3 ml buffer at 60°C for 5 minutes, centrifuged (2000 g for i0 minutes), and aliquots of the supernatant fraction incubated with a washed 2000 g pellet of homogenized luteal tissue prepared from day 8 pseudopregnant rats (crude receptor) as described earlier (ii). Binding of 125I-hCG was determined in duplicate at tissue equivalent weights of 0.25, 0.5, 1.0 and 2 mg of crude receptor. Non-specific binding was detenained in vitro by incubating 12SI-hCG in the presence of 0.5 ~g of unlabelled hCG at each concentration of crude receptor. Non-specific binding was determined in vivo by simultaneous injection (i.v.) of 125I-hCG and 1000 i.~. o--{---hOGto pseudopregnant rats 2 hours before sacrifice. Quantitation of Progesterone Progesterone content of plasma and corpora lutea was determined by radioinmunoassay as described earlier (14). Statistical Analysis Students' t-test w~s used to co,pare treatments to determine the probability level of differences. RESULTS In Figure 1 the t//re course is shc~a for uptake of radioactivity by corpora lutea, ovarian interstitium, and plasma. The specific radioactivity data shows clearly that accxm~lation of 12SI-hCG in corpora lutea was markedly greater than in any other tissue exauined. Maximum uptake to corpora lutea was observed 2 hours following administration of ]2SI-hCG and although the specific radioactivity declined at 4 hours, the ratio of binding to corpora lutea and plasma remained nearly constant between 2 and 4 hours (Table i). Ovarian inte/stititwa bound more radioactivity than plasma on an equal weight

JULY 1976

VOL. 12 NO. 1

85

PROSTAGLANDINS basis with a tissue-to-plasma ratio of about 5, 2 hours after treatment with zzSI-hCG. Surprisingly, the kidney bound appreciable radioactivity with a tissue-to-plasma ratio of about 2.2 at 2 hours. The content of radioactivity in all tissues was decreased 24 hours after treatment with 12SI-hCG, but corpora lutea, ovarian interstitiun and kidneys still contained 24, 3 and 5 times, respectively, more radioactivity than plasma at this time. Adrenals and liver bound little radioactivity.

50:

_

I

I

I

_

A N

5_ 40 x

E E

.o

30

>

0

~

20

0 u u lu

"

\ k

I0

5

30

120

240 Time

FIGURE i.

86

1440

(Minutes)

Time course for uptake of z2SI-hOG to rat corpora lutea (o), ovarian interstitial tissue (~) and plasma (&). (Mean -+ S.E.)

J U L Y 1976

V O L . 12 1NO. 1

co

o

©

~D 0%

4

4

4

4

4

5

30

120

240

1440

2 ± 1

29 ± 2

46 ± 2

100 ± 6

133 ± 3

Plasma d~/]~

5.7

0.3

23.5 ±' 5.'8

65.0 ± 11.3

70.2 ± 12.8

13.2 ±

2.4 ±

~rporaLu~a

3.2 ± 0.3

5 . 1 ± 1.2

4.6 ± 0.2

1.6 ± 0.4

0.4 ± 0.1

~ i ~ In~rstiti~

1.0 ± 0.7

0.2 ~ 0.i

0.2 ± 0.1

0 . 1 ± 0.1

0 . 1 ± 0.1

A~al

0.6 ± 0.1

0.2 ± 0.1

0.2 ± 0.1

0 . 1 ± 0.1

0 . 1 ± 0.1

~ r

RATIO CF SPECIFIC R A D I O A C f M T Y (TISSUE/PIASMA) IN VARIOUS ORGANS FOIIOWING TREATM~qT OF RATS WITH 12SI-hCG (MEAN ± S.E.)

Animals were administered 0.59 ~Ci 125I-hCG (19.7 ng). Maximtln binding of 125I-hCG was 47% and this material was used for all studies in the present report.

N

Time Min.

TABLE i.

4.9±0.6

2.2±0.1

2.2±0.1

1.2±0.1

0.5±0.1

~d~y

>

©

PROSTAGLANDINS

To determine the character of the radioactivity bound to luteal end renal tissue, corpora lutea and kidneys were homogenized and subjected to differential centrifugation. Approximately 65% of the radioactivity in the corpora lutea was associated with the fraction which sedimented at 2000 g (Table 2). Following treatment of the crude membrane fraction with heat, which in earlier studies was shown to effectively release hCG frcm luteal receptors, it was found that maxirman binding of the released radioactivity was 63%, and this binding was blocked by addition of unlabelled hCG in the receptor binding assay described earlier (12). In fact, 12SI-hCG r ~ frQm luteal tissue was bound to a greater extent than the original material used to treat the animals (e.g. 47% vs. 63%), confirming a similar observation made by other invest-£gators (21). No difference in the distribution or character of the radioactivity bound by luteal tissue was produced by treatment of animals with PGF2~. Altb~ugh the specific radioactivity was much lower in kidneys than in corpora lutea, a considerable quantity of radioactivity was present in this tissue cc~pared to liver or adrenals. As shown in Table i, renal radioactivity was about twice as great as plasma on an equivalent weight basis, 2 and 4 hours follo~ing treatment of the animals with 12SI-hCG. Fractionation of homogenized renal tissue demonstrated that the radioactivity present in the kidney probably was not 12SI-hCG since little of the radioactivity could be bound by the luteal LH receptor. PGF2~ had no measurable effect on the distribution of radioactivity present within the kidney.

TABLE2.

CHARAcI'~tIZATIC~OF R A D I O ~ T I V I T Y T A K I ~ U P BY ~ ~ D RE~LAL TISSUE ~ V l V O

Hcmogenate

2,000 g Pellet

Tissue Fraction (%) I00,000 g i00,000 g Superngtant Pellet

Corpora Lutea

100.0

66.6*

23.2

10.2

Kidney

100.0

21.7

65.0**

13.3

*The luteal pellet was suspended in 3 ml Tris buffer and incubated at 60° for 5 minutes. The reaction mixture was centrifuged at 2,000 g (10 minutes) and aliquots of the supernatant fraction tested in the receptor binding assay; 63% of the radioactivity was bound at maxim~n receptor concentrations. 125I-hCG was 47% bound in an identical assay at ~ receptor concentrations. * * T ~ kidney s~pernatant fraction was incubated at 60° for 5 minutes and treated as above. In the receptor binding assay less than 1% of tb~ radioactivity was bound at maxirm~ receptor concentrations. 12 sI_ hCG, treated identically, was 47% bound at maxirman reoeptor concentrations.

88

JULY 1976

VOL. 12 NO. 1

00

o

h~

©

4

44 ± 47 ±

52 -+ 38 ±

2 3

9 4

1126 ± 101 *'412 ± 123

0.5

5

29 ± 40 ±

79 ± 112 ±

1 2

30 26

1354 ± 92 **484 -+ 147

1

5

21 + 23 ±

155 ± *84 ±

1 1

35 7

1829 ± 177 **456 + 131

Time (hours) 2

125I-hOG UPTAKE IN VIVO BY CORPORA ~ INTERSTITII~ FOLLOWING P ~ 2 ~ T ~ n X M ~ T

5

37 ± 44 -+

184 ± 149 ±

3 3

23 ii

2771 +- 297 **885 ± 260

4

AND OVARIAN (MEAN ± S.E.)

4

36 ± **47 -+

177 ± 180 +

i 1

17 21

2684 ± 847 **396 ± 19

24

The 4 and 24 hour studies were run sira/itaneously and the animals received 0.41~Ci (13.6ng)125 I-hCG 2 hours before sacrifice. Animals in the 2 hour study reoeived 0.30vCi(9.1ng) 12SI-hCG 2 hours before sacrifice. Animals in the 0.5 and 1 hour study received 0.21~Ci 125I-hOG (6.4ng) 0.5 and 1 hour before sacrifice, respectively.

**p<. 001; *p<.05

Animals/Group

Control P(~2 ~

Pia.~ (diem9)

Ccntrol PGF2~

Inter stitium ( d p m ~ )

Ccntrol PGF2 ~

Corpora Lutea (dpm/mg)

TABLE 3.

>

>

©

PROSTAGLANDINS

TO detex~line the effect of PGF2~ on uptake of 125I-hCG, day 7 pseu~opregnant rats were ackainistered PGF2~ (3mg/kg; s.c. ) 0.5, 1 2, 4 or 24 hours before sacrifice. Animals in the 2, 4 and 24 hour groups received 125I-hCG 2 hours before sacrifice and simultaneously with PGF2~ in the 0.5 and 1 hour groups. Specific radioactivity data in corpora lutea, ovarian interstiti~n and plasma is shc~n in Table 3. Although maximum uptake of 125i_h0 G was not seen in corpora lutea until 2 hours after administration (Figure i), PGF2a significantly inhibited 125I-hOG uptake by corpora lutea within 30 minutes (-63%; p<.001) and 1 hour (-64%; p<.001). Indeed, a sJ~tilar and equally marked effect of PGF2~ was seen at 2 (-75%; p<.001), 4 (-68; p<.001) and 24 hours (-85%; p<.001). Ovarian interstitium had considerably less radioactivity and the responsese to PGF2~ was more variable with no clear inhibition of 12~I-hCG uptake at all time periods. An effect of PGF2a on 125I-hCG uptake by interstitial tissue cannot be concluded since contaninating remnants of lutea tissue may have ccntributed to the apparent drop in radioactivity, such as that seen 2 hours after PGF2a treatment. Radioactivity in plasna appeared to be 10% to 30% greater in PGF2e-treated animals and this effect was significant (p<.001) in the 24 hour group. It is possible that elevated radioactivity in plasma may have oocurred because of reduced uptake by corpora lutea from plasna. Non-specific binding in vivo in control animals was detezmined by administration of 1 2 5 I - ~ s~ultaneous with 1000 i.u. of unlabelled hCG; non-specific binding by corpora lutea was 1.1% and for interstitial tissue 8%, detezmined 2 hours after injection (i.v.) of hCG. The effect of PQ?2 a on progesterone ccntent in plasma and luteal tissue is shown in Table 4. Plasma progesterone fell at all times examined after PGF2o~treatment. This response occurred as early as 30 minutes with a decrease in plasm~ progesterone of 48% (p<.01). At 1 hour a 65% (p<.01) decrease in plasma progesterone was seen and at 2 hours a decrease of 82% (p<.001) occurred. Four hours after PGF2a treatment, a 68% decrease (p<.01) in plasma progesterQne was seen and at 24 hours little progesterone remained in plasaa. Luteal tissue obtained from animals sacrificed 2 hours after PGF2a treatment ccntained 46% less progesterone (p<.01) than ccntrol luteal tissue which indicates that IK~2a probably did not block t_he luteal secretory process. A significant difference in plamna progesterone was also evident between the control animals in the 0.5, 1 and 2 hour groups aD~ the control animals in the 4 and 24 hour groups; those used in the 0.5, 1 and 2 hour study r e d e d to the estrous synchronizing dose of pregnant

90

JULY 1976

VOL. 12 NO. 1

©

©

%o o%

Control PGF2~

Progesterone

(ng/mg)

(ng/ml)

4

114.6 -+ 13.6 *59.6 -+ 10.6

5

114.6 +- 13.6 *40.3 +- 15.2

1

5

30.6 -+ 3.1 "16.4 + 1.8

i18.8 +- 9.7 "21.7 + 7.6

2

Time (Hours) 4

5

57.9 + 4.9 19.9 +- 5.0

EFFECT OF P ~ 2 e ON PROGESTERONE CONTENT OF PLASMA AND LUTEAL TISSUE (MEAN +- SoE. )

4

57.9 -+ 4.9 "4.1 + 1.9

24

The above data is a ccmposite of 3 independent experiments comprised of a 0.5 and 1 hour study, a 2 hour study, and a 4 and 24 hour study. The same dose of PGF2~ (3 mg/kg; s.c.) was used in each experiment.

* ~ . 01

Animals/Group

Luteal

Control PGF2~

Plasma Progesterone

0.5

TA~,v 4.

~n

©

PROSTAGLANDINS

mare's ser~n with 12.5 + 0.7 corpora lutea (mean -+ S.E.) and the animals in the 4 add 24 hour study with 9.3 -+ 0.4 (mean + S.E.) corpora lutea. This difference was statistically significant (i~ .01) and probably accounts for the difference in plasma progesterone seen in the control animals between the two groups.

DISCUSSION

PGF2a caused a marked decrease in uptake of 12SI-hCG to corpora lutea as early as 30 minutes after administration and this response was mirrored at i, 2, 4, and 24 hours after PGF2~ treatment. It was interesting to note that the block in luteal uptake of 125i_hCG was closely related to the drop in p l a ~ a progesterone at each time period. These data, while not unequivocally establishing a basis for ug~erstanding the mechanism of PGF2~-induced luteolysis, support the hypothesis that luteolysis may occur as a consequence of a rapid loss of gonadotropin binding by corpora lutea. Earlier studies (5) demonstrated that an intravenous injection of PGF2~ reduced progesterone secretion into the ovarian vein of pseudopregnant rats within 30 minutes and both the effects of PGF2a and hypophysectcray on progesterone secretion were reversed by administration of LH. It was suggested on the basis of these early data that PGF2~ may block gonadotropin action on corpora lutea. It was surprising to observe the early drop in corpus lutetml uptake of gonadetropin when 12SI-hCG was administered in vivo since earlier studies showed that no loss of LH receptors was seen 2 hours after PGF2~ treatment (ii). Thus either gonadotropin binding capacity of corpora lutea quantitated in vitro represent receptors unavailable for binding in vivo when animals are treated with PGF2~, or exposure of receptors to gonadotropin is in same unknown manner blocked by PGF2~ in vivo. Studies have shown that PGF2~ does not block binding of TTSI-hCG to LH receptors in vitro in isolated membranes (ii) which indicates no direct i n h i b ~ i ~ by PGF2~ of the LH receptor. A decrease in blood supply to the corpus lute~n would result in a decrease in 12SI-hCG available for binding to the luteal receptor but previous experiments in the rat demonstrated that PGF2e did not cause a change in total ovarian blood flow 0.5, 6 or 12 hours after treatment, yet progesterone secretion was significantly depressed at each time period (5,15). In studies with the rabbit, PGF2e was reported to decrease blood flow to the corpus lute~n (4) and these data support the hypothesis that this may be the mecb~nnism of PGF2~induced luteolysis (3). However, it was recently concluded that corpus lute~n blood flow depression is probably not the mechanism of PGF2~-induced luteolysis in the rabbit since the drop in blood to corpora lutea (-29% at 24 hours) occurred several b~urs after the first significant fall in progesterone secretion (16). Other workers have reported a rapid decrease (-36%) in ovarian venous flow rate

92

JULY 1976

VOL. 12 NO. 1

PROSTAGLANDINS

following P(~2cx infusion, an effect directly related to a parallel decrease (-35%) in blood pressure, but found no change in blood flow to the corpus luteun (6). The effect of PGF2e an blood pressure is dependent upon species (17, 18). For example, in the rabbit PGF2~ causes a decrease in blood pressure and in the rat causes an increase in blood pressure. If blood flow to the ovary is directly related to arterial pressure, and PGF2u does not affect luteal blood supply prior to a fall in progesterone secretion, it is difficult to reconcile a PGFz~-induced increase in blood pressure in the rat with a drop in progesterone output. However, until further information is obtained, no ccnclusion can be made for or against an action of PGF2~ an ovarian hemcdynamics as the cause for loss of 125I-hCG uptak~ and progesterone production seen in the present studies.

PGF2a has been shown to decrease progesterone synthesis when added to organ cultures of rat corpora lutea (7), in incubations of rabbit corpora lutea (9) and in cultured human granulosa cells (i0). Moreover, with hanster corpora lutea placed in organ culture, PGF2m was shown to bloQk stimulation of progesterone output by either IB or the stin~/laticn produced by co-incubaticn of corpora lutea and hemipituitaries (8). In fact, in a recent report it was shown that PGF2a blocked I/~ stirmllation of cyclic AMP prcducticn in rat luteal tissue in vitro (23). These data, in additiou to the earlier cbservaticn an the similarity in response of acute hypcphysectcmy and PGFzu treatment cn progestercee secretion (5), lend suplx)rt to tl~ ccnclusion of a direct and rapid action of PGF2~ in blocking gonadotrapin acticn in corpora lutea independent of a vascular effect. This action of PGF2e may be due to a block in gcnadotrcpin binding, an effect which occurs only in intact cells since inhibition of the LH receptor in isolated membranes was not seen (ii). Binding of 125I-hCG to ovarian receptors has been described by others using the pseudcpregnant and superovulated rat (19, 20, 21 and 22). We confirmed the time ccurse previously described (19) and the acom~lation of radioactivity in the kidney, which is sa~d to occur principally in the proximal tubule (20). As in the present work,, heat-eluted radioactivity from lutea tissue has been found to bind to a receptor in vitro to a greater extent than the original hOG (21), presumably due to prior select/on by the tissue in vivo of biologically active material.

J U L Y 1976

VOL. 12 NO. 1

93

PROSTAGLANDINS

REFERm~F~

i.

Behrman, H. R. and B. V. Caldwell. In Greep, R. O. (ed.), MTP International Review of Science. Reproductive Physiology, volt,he 8, Butterworth and Co. Ltd., London, 1974, p. 63.

2.

Pharriss, B. B. and L. J. Wyngarden. Med. 130:92, 1969.

3.

Goding, J. R., D. T. Baird, I. A. C ~ A~ta Endocrin. (Suppl.) 158:169, 1972.

4.

Novy, M. J. and M. J. Cook.

5.

Behrman, H. R., K. Yoshinaga and R. O. Greep. Sci. 180:426, 1970.

6.

Janson, P. O., I. Albrecht a n d K . Ahren. 79:337, 1975.

7.

Demers, L. M., H. R. ~ Biosciences 9:701, 1972.

8.

Behrman, H. R., T. S. Ng and G. P. Orczyk. I_nnMoudgal, N. R. (ed.), Gonadotropins and Gonadal Function, Academic Press, New York, 1974, p. 332.

9.

O'Grady, J. P., E. I. Kohorn, R. H. Glass, B. V. Caldwell, W. A. Brock a n d L . Speroff. J. Reprod. Fert. 30:153, 1972.

Proc. Soc. Exptl. Biol.

and J. A. McCraeken.

Am. J. Obstet. Gynec. 117:381 , 1973. Ann. N. Y. ~P_ad.

Acta Endocrin.

and R. O. Greep.

Advanoes in the

i0.

McNatty, K. P., K. M. HeD~erson and R. S. Sawers. 67:231, 1975.

ii.

Grinwich, D. L., M. Hichens and H. R. Behrman. Reprod. 14:212, 1976.

Biol. of

12.

Hichens, M., D. L. G r i n w i c h a n d H . 7:449, 1974.

Prostaglandins

13.

Grinwich, D. L., E. A. Ham, M. Hichens and H. R. Behrman. Endocrinology 98:146, 1976.

14.

Orczyk, G. P., M. Hichens, G. E. Arth and H. R. Behrman. In Jaffe, B. M. and H. R. Behrman (eds.), Methods of Hormone Radioirmaunoassay, Academic Press, New York, 1974, p. 347.

15.

Behrman. H. R., K. Yoshinaga, H. Wyman and R. O. Greep. Am. J. Physiol. 221:189, 1971.

94

R. Behrman.

JULY 1976

J. Endocrin.

VOL. 12 NO. 1

PROSTAGLANDINS

16.

Bruce, N. W. and K. Hillier.

17.

Karim, S. M. M. and K. Scmers. In Karim, S. M. M° (ed.), The Prostaglandins, Progress in Research, John Wiley and Sons, Inc., New York, 1972.

18.

Ducharme, D. W. and J. R. Weeks. In Bergstrom, S. and B. Samuelsson (eds.), Nobel S y m p o s ~ 2, Prostaglandins, John Wiley and Sons, Inc., New York, 1967, p. 175.

19.

Ashitaka, Y. and S. S. Koide.

20.

Rajaniemi, H. J. and T. Vanha-Perttula.

21.

Han, S. S., H. J° Rajaniemi, M. J. Cb~, A. N. Hirshfield and A. R. Midgley. Endocrinology 95: 589, 1974,

22.

Rajaniemi, H. J., A. N. Hirshfield and A. R. Midgley. Endocrinology 95: 579, 1974.

23.

Lahav, M., A. Freud and H. R. Lindner. Ccrnm. 68:1294, 1976.

JULY 1976

VOL. 12 NO. 1

Nature 24.9:176, 1974.

Fert. Ster. 25:177, 1974. Endocrinology 90:1, 1972.

Biochem. Biophys. Res.

95