PROSTAGLANDINS
IN VIVO DESENSITIZATION OF A HIGH AFFINITY RECEPTOR IN THE OVINE CORPUS LUTEUM
PGFZos
J.C. Lamsa, R.A. Cu.&man, M.G. Nay and J.A. McCracken Worcester Foundation for Experimental Biology Shrewsbury, Massachusetts 01545
ABSTRACT
The corpus luteum (CL) of the sheep exhibits a differential sensitivity to PGF2a in vivo in terms of an increase in oxytocin (OT) secretion and a decrease in progesterone secretion, pointing to the presence in vivo of both high and low affinity receptors for PGF2a. The presence of the high affinity PGF2a receptor was assessed by monitoring the secretion rate of OT from the ovine CL in response to subluteolytic infusions of PGF2a. Rapid desensitization to PGFZCX occurred after only one hour of infusion, while a minimum rest period of six hours was required to restore sensitivity. The possibility that these findings could be explained by the depletion and resynthesis of OT was excluded by demonstrating an increase in OT secretion rate with supra-physiological levels of PGF2a two hours after desensitization. Collectively, these results indicate the presence of a high affinity receptor for PGF2a in the ovine CL which exhibits desensitization and recovery in vivo. The temporal nature of the desensitization and recovery of the high affinity PGF2a receptor controlling luteal OT secretion may contribute to the pulsatile nature of PGF2a release from the ovine uterus.
This work was supported by NIH grant HD-08129. Reprint requests to: Dr. John A. McCracken, Worcester Foundation Experimental Biology, 222 Maple Avenue, Shrewsbury, MA 01545.
FEBRUARY 1992 VOL. 43 NO. 2
for
165
PROSTAGLANDINS
INTRODUCTION Prostaglandin Fzo: (PGF2a). the luteolytic hormone in sheep (1). is released from the uterus as a series of hour long pulses at intervals of about six hours at the end on the luteal phase (2.3.4.5). Since the intraovarian administration of exogenous PGF2a as a series of discrete one hour long pulses given at six hour intervals is more effective in causing luteolysis than a constant infusion (61, a specific pulsatile pattern of PGF2a release from the uterus appears to be advantageous for luteolysis in this species. Thus, the mechanisms regulating the pulsatile release of PGF2a from the uterus are important in the process of luteolysis. During luteolysis, pulses of PGF2a and also its metabolite occur synchronously with elevations of oxytocin (OT) or OT neurophysin in the peripheral circulation (7.8.9). Exogenous OT causes the immediate release of PGF2a from the uterus when the oxytocin receptor (rOT) is present in the endometrium (10). In addition, PGF2a has been found to cause the release of OT from the corpus Thus, a positive feedback loop develops luteum (CL) (11,12,13). between the CL and the uterine endometrium, culminating in a high magnitude pulse of PGF2a. which initiates luteolysis (14). The factors regulating this positive feedback loop, and thus the duration of and intervals between luteolytic pulses of PGF2a. are controversial. It has been proposed that the OT in the CL is totally depleted by each hour long pulse of PGF2a, and until the supply of luteal OT is restored, no further pulses of PGF2a can occur (7,15). Alternatively, luteal OT may down-regulate its endometrial receptor, and thereby terminate PGF2a The interval required to regenerate rOT would thus secretion. regulate the timing of subsequent pulses of PGF2a from the uterus (14). However, an additional level of regulation may exist within the CL itself. The corpus luteum of the sheep is differentially sensitive to PGFZa, possessing a high sensitivity for increased OT secretion and and a low sensitivity for decreased progesterone secretion (13). This observation suggests the presence of both high and low affinity PGF2a receptors in luteal tissue. Recently, it has been reported that two classes of PGF2a receptor exist within the CL of the sheep (161. A high affinity receptor was found on the large luteal cell (the source of luteal OT: 17) and a low affinity receptor was found on both the small and large luteal cell (both of which secrete progesterone: 18). In the present study, we have explored the physiological role of the high affinity PGF2a receptor by monitoring the secretion rate of OT from the autotransplanted ovary in vivo in response to either the long term constant infusion or short term intermittent infusions of subluteolytic levels of PGF2a. The objectives of this study were to determine a) if desensitization of the ovine CL to PGF2a occurs in vivo; bl if such desensitization to PGF2a exists, how long does the CL remain refractory to subsequent infusions of PGFZa; and cl if such refractoriness occurs, can it be explained by desensitization of a high affinity receptor, or by the depletion and subsequent resynthesis of luteal OT?
166
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
MATERIAL!3 AND METHODS Exnerimental
animals
Merino sheep with the left ovary autotransplanted with vascular anastomoses into a jugulo-carotid loop and with the right ovary removed were used as experimental subjects (19.20). To perform in ~yi experiments in sheep with a CL of known age, estrus was induced by two intramuscular injections of 5 mg PGF2a (Lutalyse. Upjohn Co., Kalamazoo MI) administered at an interval of 4 hours. Estrus. which was determined by vasectomized rams, occurred 48 - 72 hours after PGF2a administration (estrus = cycle day 0). All infusion experiments, each in a separate animal, were performed 12 days after estrus in order to mimic conditions in cyclic sheep just prior to the beginning of luteolysis. Sheep were housed in metabolism cages with free access to food and water for the duration of each infusion experiment. Infusion of the ovarv in vivo with PGF2a Twenty-four hours prior to each experiment the left carotid artery contained in the isolated jugulo-carotid loop was cannulated as On the day of each experiment, described previously (13.20.21). stock solutions of PGF2a were prepared by dissolving PGFZa (PGFZa tromethamine salt: Cayman Chemical Co., Ann Arbor MI) in 100% ethanol at a concentration equivalent to 1 mg PGFZa/ml. The concentration was then adjusted for infusions with sterile pyrogen and preservative free aqueous 0.9% sodium chloride (Abbott Laboratories, cuff was North Chicago 11). For all infusions, a sphygmomanometer placed on the cranial side of the jugulo-carotid loop. Inflation of the cuff to 200 mm Hg insured the direct infusion of PGFZa into the ovary via the cannula in the carotid artery. Low concentrations of PGF2a were administered by means of an infusion pump (Harvard Apparatus, Cambridge MA) set to deliver 2.0 ml/hr (25 pg/min. 100 pg/min, or 250 pg/min PGFZa) . The infusion rates of PGFZa chosen for this study were far below the level necessary to cause the depression of progesterone secretion by direct arterial infusion of the ovary in this species (6.13). Collection of ovarian venous blood At the time of cannulation of the left carotid artery, a teflon catheter (1.5 mm i.d. x 40 cm) was inserted into the cranial portion of the left jugular vein as described previously (13.20). Upon completion of the cannulation procedure, each animal was given 5000 IU heparin as an anticoagulant i.v. every 4 hours throughout the course of each experiment. Timed samples of ovarian venous blood (5 ml) were collected into chilled tubes over ice by manual occlusion of the jugular vein on the cardiac side of the transplanted ovary. Blood flow was calculated by measuring the time required for each 5 ml blood sample to flow freely from the venous catheter (about 30 seconds). Immediately after collection, each blood sample was centrifuged at 40
FEBRUARY
1992 VOL. 43 NO. 2
167
PROSTAGLANDINS
C for 30 minutes at 2000 rpm, and the hematocrit and plasma volume of each sample was measured. This yielded the plasma flow (volume of plasma/minute), and hence allowed the calculation of the secretion rate of OT (mass/unit time) from the ovary (13). Aliquots of plasma were removed promptly and stored at -200 C until assayed for GT. Oxytocin radioimmunoassay OT radioimmunoassay (RIA) was performed as described previously (13). The mean extraction recovery of sH- OT (Amersham, Arlington Heights 11) from plasma was 75% and OT values were Intra- and interassay coefficients of corrected for extraction losses. variation were 10.8% and 14.3% respectively. Sensitivity of the RIA ranged between 10 and 20 pg/tube, as determined by the amount of GT that was two standard deviations above the value of the assay blank. ExDeriment
1
To determine basal levels of OT secretion, ovarian venous blood samples were collected every 15 minutes during a one hour period before the infusion of the vehicle control. Vehicle (sterile aqueous 0.9% NaCl free of pyrogens and preservatives: Abbott Laboratories, North Chicago 11) was infused at a rate of 2 ml/hr directly into the ovarian arterial blood supply for two hours (n = 3). Ovarian venous blood was collected at intervals ranging from 5 to 30 minutes during the infusion To determine if the CL became refractory to the infusion of low levels of PGFzu in terms of GT secretion, PGF2a was infused directly into the ovarian arterial blood supply for a period of 10 hours at a rate of either 25 (n = 4) or 250 pg/min (n = 3). Ovarian venous blood was collected 5, 10, 15 and 30 minutes after the start of the infusion, and at 30 minute intervals thereafter for the duration of the infusion. Five mg PGF2a was given i.m. one hour after stopping the low level PGF2a infusions to determine if any OT remained in the CL. ExDeriment
2
To determine the duration of refractoriness to PGF2a, brief incremental rest periods were introduced between the 2 hr PGF2a infusions. Since the CL was responsive to low levels of PGlQa for only about 1 hour (Expt. 1). PGF2a was infused at a rate of 25 pg/min for six 2 hour periods, each separated by a gradually increasing rest period (n = 2). Rest periods (no PGF2a infusion) of 10, 20. 30, 60 and 90 minutes were introduced between infusions of PGF2a. Ovarian venous blood was collected at 5, 10, 15 and 30 minute intervals during each Five mg infusion and at 5, 10 and 15 minutes between infusions. PGF2a was given i.m. one hour after stopping the low level PGF2o: infusions to determine if any OT remained in the CL.
168
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS EXDeriment
3
Since rest intervals of up to 90 minutes were insufficient to allow the return of OT secretion in response to low level PGF2a infusions, incremental rest periods of longer duration (3 to 9 hours) were introduced between the 2 hour infusions of PGF2a. PGFZCXwas infused for two separate two hour periods in nine sheep at a rate of 100 pg/min with rest periods between infusions of 3 hours (n = 3). 6 hours (n = 3). or 9 hours (n = 3). Ovarian venous blood was collected at 5, 10, 15 and 30 minute intervals during each infusion and at 5, 10, 15 and 30 minutes between infusions. Exneriment
4
To determine if releasable oxytocin remained in the corpus luteum one hour after desensitization to a low level of PGF2a. PGFZa was infused at a rate of 25 pg/min for a two hour period (n = 6). Ovarian venous blood was collected at 5. 10. 15 and 30 minute intervals during the infusion. Five mg PGF2a was given i.m. one hour after stopping the low level PGF2 a infusions to determine if any OT remained in the CL. Statistical analvsis The area under the secretion rate curve was calculated, yielding the mass of oxytocin secreted during each PGFZCI infusion. Significant secretion was defined as occurring when the mass of OT released into the ovarian vein during an infusion of PGF2o: was significantly (p c 0.05) greater than during an equivalent control period. Significance was determined by analysis of variance. Duncans Multiple Range Test was used to determine significance between treatment means (22). RESULTS Exneriment
1
OT secretion remained at the basal level throughout the duration of the infusion of vehicle (figure 1). The infusion of 25 pg/min PGF2a: for a ten hour period resulted in a rapid rise in OT secretion, which peaked within 15 minutes and dropped to control levels by one hour after the start of infusion (figure 2). The pattern of OT secretion in response to 250 pg/min PGF2a for 10 hours showed a similar pattern, although peak levels of OT were higher (figure 2). No further increase in OT secretion was seen during the remainder of the experiment at either dose level, despite the continued infusion of PGFzo. The absence of OT release in response to PGF% was not due to the lack of OT, as a supraphysiological dose of PGF2a (5 mg i.m.) stimulated the release of OT from the CL one hour after the completion of the low level PGF2a infusions.
FEBRUARY
1992 VOL. 43 NO. 2
169
PROSTAGLANDINS SALINE
NG’%Y
VEHICLE
INFUSION
8-
64-
, -1
,
.
.
.
.
,
1
0
TIME
.
.
;
.-.
2
, 3
IN HOURS
Figure 1: The effect of a 2 hour intra-arterial infusion of saline vehicle on oxytocin secretion rate from the ovarian autotransplant on day 12 after estrus (n = 3). NG/MIN
25 FG/MIN
FGF2a
5 MG PCF2a
INlWbAFZTERL4L
IM
4
g
2030z:;;
-1
0
1
2
3
4 TIME
Figure 2: The estrvs during a later by a single (n = 4). Bottom
Exoeriment
6
5 IN
7
8
9
10
11
12
13
14
HOURS
mean oxyfocin secretion rate from the ovarian autotransplant on day 72 after 10 hour intra-arterial infusion of subluteolytic levels of PGF2a followed one hour i.m. injection of a supraphysiological level of PGF2a (5 mg). Top panel: 25 pg!min panel: 250 pg/min (n = 3).
2
The initial two hour long infusion of PGF2a at 25 pg/min (figure 3) showed a pattern of OT release similar to that seen in Expt. 1. Rest intervals of 10, 20, 30, 60 or 90 minutes were insufficient to restore responsiveness to PGF2a. as no increase in OT secretion rate was seen during any of the subsequent 2 hour infusions of PGFza. Again, the
170
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
absence of response to PGFZa in terms of OT release was not due to the lack of luteal GT. as a supraphysiological dose of PGF2a (5 mg i.m.) stimulated the release of OT from the CL one hour after the completion of the experiment. 25 PG/MIN
NG/MIN
PGF2a
IN’TFU-ARTERIAL
5 MC PGF2a
IM
I
0
8
4
TIME
12
16
20
IN HOURS
Flgure 3: The mean oxyiocin secretion rate from the ovarian autotransplant on day 12 after estrus during 6 separate 2 hour intra-arterial infusions of subluteolytic levels of PGF2a (25 pg/min) followed one hour later by a single i.m. injection of a supraphysiological level of PGF2a (5 mg) (n = 2). Rest periods (no PGF2a infusion) of 10, 20, 30, 60 and 90 minutes respectively were introduced between each successive 2 hour infusions of PGF2a.
The infusion of PGF2a into the ovarian artery for a period of 2 hours resulted in a pattern of OT secretion similar to that seen in Expts. 1 and 2, that is the secretion rate of OT was elevated for only one hour. The pattern of OT secretion in response to PGF2a is shown in three representative experiments in figure 4. After a three hour rest interval, a second infusion of PGF2a at a rate of 100 pg/min was unable to induce a significant (p > 0.05) release of OT from the ovary (figure 5). When the rest interval was increased to six hours, PGFZa caused a significant (p < 0.05) release of luteal OT (figure 5). The mass of OT secreted during the second PGF2a infusion was When the significantly lower than during the first infusion (P < 0.05). interval between infusions was increased to 9 hours, PGF2a caused a significant (p < 0.05) release of OT. After the 9 hour interval, the mass of OT secreted during the second infusion was not significantly different (p > 0.05) from the total mass secreted during the first PGFZa infusion (figure 5). Since the results of Expt. 3 demonstrate that a difference between the mass of oxytocin secreted after desensitization depends on the interval between infusions, a pilot experiment was performed in which rest intervals (3, 6 and 9 hours) were introduced between PGFZa infusions (250 pg/min) in a single animal. Oxytocin levels remained at control levels after a 3 hour rest interval. However. after
FEBRUARY
1992 VOL. 43 NO. 2
171
PROSTAGLANDINS
hours PGF2a again induced the release of oxytocin, with OT secretion rate peaking at 25 ng/min. After a 9 hour rest interval, OT secretion was stimulated by PGF2a infusion, peaking at 37 ng/min (figure 6).
6
NG/MIN PGna
30
3hr vINIERVAL
20
b
2
50
3
40
E
30
$
20
6hr INIERVAL
cn g E 0
lo 0 FGF2a
0
Pow
30
v
9hr INTERVAL
V
D
20
10
0 -1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TIME IN HOURS Figure 4: The effect of 2 separate 2 hour intra-arterial infusions of subluteolytic levels of PGFZa (100 pg/min) given at intervals of 3 hours (top panel), 6 hours (middle panel) or 9 hours (bottom panel) onday 12 after estrus in three individual sheep .
172
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
a
T
a
n
CONI-ROL
0
PCF2a 1st Inf.
H
FGK!a 2nd Inf.
6hr
3hr REST
9hr
INTERVAL
Figure 5: The relative mass of oxytocin secreted in response to the second 2 hour infusion of PGF2a (100 pg/min) expressed as a percentage of the mass of oxytocin secreted during the first 2 hour infusion of PGF2a (100 pg/min). N = 3 for each rest interval. The designation “a” indicates a significant difference from control values for that rest interval, while b” indicates a significant difference from the first infusion.
6hr
9hr
Z-+-M
4
-2
0
2
4
6
8
10
12
14
16
18
20
22
24
26
TIME IN HOURS Flgore 6: Oxytocin secretion during repetitive 2 hour intra-artetial infusions of subluteolytic levels of PGF2a (250 pg/min) in a single sheep. The 2 hour infusions of PGF2a were given at intervals of 3, 6 and 9 hours on day 72 after estrus.
Exneriment 4 The infusion of PGF2a at 25 pg/min into the ovarian artery for a period of 2 hours resulted in a pattern of OT secretion similar to that seen in Expts. 1, 2 and 3, that is the secretion rate of OT was elevated for only one hour and then declined. However, the intramuscular
FEBRUARY
1992 VOL. 43 NO. 2
173
PROSTAGLANDINS
administration of 5 mg: PGF& secretion rate (figure 7): 25 PG/MIN
-1
0
i.m. resulted
PGFZa
1
in a rapid rise in OT
5 MC PGF2a IM
2
3
4
5
TIME IN HOURS Figure 7: The mean secretion rate of ovarian oxytocin in 6 sheep. Following desensitilation to PGFZU, a single i.m. injection of 5 mg PGF2a was administered one hour after completion of a 2 hour intra-arterial infusion of subluteolytic levels of PGF2a (25 pg/min) on day 72 after esters.
DISCUSSION It is well established that the CL of the sheep will release OT h m in response to exogenous PGF2a (11,121, and that small elevations of PGF2a in ovarian arterial blood elicit such a response (13). In the present study, a continuous infusion of subluteolytic levels of PGF2a into the ovarian artery initiated an immediate release of luteal OT followed by desensitization in the presence of continued low level infusion of PGFZa. The return of luteal sensitivity to PGFZa-induced OT secretion develops over time. While the mass of OT secreted following a 6 hour rest interval is significantly higher than control values, the amount of OT released was still significantly lower than that released during the first infusion. When the rest interval is increased to 9 hours, the mass of OT released did not differ significantly from the total amount secreted during the first PGF2a infusion, indicating that full recovery may occur between 6 and 9 hours following desensitization. It is worthy of note that the cessation of OT secretion observed in this study was not due to the depletion of luteal OT, as the i.m. administration of a supraphysiological dose of PGF2a stimulated OT secretion as soon as one hour after the completion of the low-level infusion of PGF2a. Paradoxically, others have been unable to demonstrate the release of OT by PGF2a from ovine luteal tissue ti vitro (23.24.25). One explanation for the failure of PGF2a to release OT from luteal tissue in vitro is that endogenous PGFzu, generated as a result of trauma associated with the handling of tissue (26). may desensitize the PGF2a receptors in luteal tissue and thus mask the response to exogenous PGF2a. A rest period of at least 6 hours is required before luteal responsiveness to low levels of PGF2a is
174
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
restored in viva. To mimic in vivo conditions, it may be necessary to observe a similar rest period in vitro, thus allowing the return of luteal sensitivity to PGF2a. It has been proposed that the release of OT from the ovine CL would form a positive feedback loop via PGF2a from the uterus, and that the secretion of OT would continue until all of the OT in the CL was exhausted (27). Thus, the interval between large pulses of PGFza from the uterus would be regulated by the time required for either & novo resynthesis of luteal OT or replenishment of OT from stored These possibilities are difficult to reconcile with prohormone (15). the findings of the present study, where the continuous infusion of low levels of PGF2a caused OT secretion for only a period of about one hour. In addition, the observed period of desensitization lasts 6 to 9 hours, similar to the intervals seen between luteolytic pulses of PGF2a in the cyclic animal (28). Ivell et al. (29) have shown that levels of OTmFWA are high only early in the ovine estrous cycle, and are very low towards the end of the luteal phase, when OT resynthesis has been suggested to occur. The absence of OT-mFWA at this time does not preclude replenishment of luteal OT from stored prohormone. However, as shown in the present study, luteal stores of OT are not depleted, since supraphysiological concentrations of PGFZCI will still release OT after a rest period of only one hour. This period is considerably shorter than the 6 to 12 hours observed between releases of OT seen during luteolysis in vivo (28). Lastly, a pharmacological dose of PGFZCX(5 mg i.m.1 will release a smaller mass of oxytocin after the infusion of 250 pg/min PGF2a compared with the 25 pg/min PGF2a infusion (figure 2). in keeping with the concept that the CL contains only a finite pool of OT (13). The high sensitivity of the CL to PGF2a is consistent with current knowledge of the cell types present in the CL. The CL is composed of two major lutein cell types: the large luteal cell and the small luteal cell. Small luteal cells possess a large number of LH receptors (30) and secrete most of the progesterone (18). Large luteal cells of the sheep have been identified as the source of ovarian OT (17). possess few LH receptors, but have the majority of PGFZa receptors (30). Specific receptors for PGF2a have been found in the plasma membrane from human, bovine (311, and ovine (32) corpora lutea. More recently, Balapure et al. (16) provided evidence that there are two classes of PGFZa receptors in the ovine CL, that is, a high affinity receptor on the large luteal cell (the source of luteal OT; 17). and a low affinity receptor on both large and small luteal cells. Very low infusion rates of PGF2a can selectively release OT, presumably by interacting with the high affinity receptor on the large luteal cell (13). while much higher infusion rates of PGFZa (in the order of 1000 pg/min into the ovarian artery) are required to cause the decline of progesterone secretion in vivo (6). It therefore seems likely that low levels of PGF2a. emanating from the uterus and acting on the high affinity receptors on the large luteal cell, triggers the release of luteal OT. The continued presence of PGF2 a will down-regulate the high affinity receptors on the large luteal cells, resulting in desensitization. In the naturally cycling animal at the time of luteolysis, the
FEBRUARY
1992 VOL. 43 NO. 2
175
PROSTAGLANDINS
supplemental release of OT from the CL, acting on rising levels of receptors for OT in the endometrium, appears to cause the major episodic secretions of PGFZa which are seen during luteolysis in the sheep. Such high levels of PGF2a initiates luteolysis, apparently by acting on low affinity PGF2a receptors on both the large and small luteal cells. Under physiological conditions, the higher levels of PGF2a initiated by luteal OT release may desensitize the both the high and low affinity receptor. Evidence supporting this concept is threefold. It has been shown that the pulsatile administration of luteolytic levels of PGF2a is more effective in causing luteolysis than a constant infusion (6). Furthermore, Watkins and Moore (12) demonstrated that the ovine CL becomes refractory to high levels of PGF2a administered into the peripheral circulation in vivo, as determined by the release of OT neurophysin. Finally, a very high, pharmacological dose of PGF2a will cause the release of luteal OT following desensitization to subluteolytic levels of PGF2a. This supraphysiological level of PGF2u may act through the low affinity. high capacity PGF2a receptor present on both small and large luteal cells. Alternatively, the very high levels of PGF2a may cause plasma membrane changes by direct intercalation into the membrane (331. The function of the desensitization mechanism may be to conserve the finite pool of OT, preventing depletion of luteal OT early in luteolysis, and thus insuring that sufficient OT remains for the initiation of several pulses of PGF2a. Recovery of the high affinity receptors, after a minimum rest period of about 6 hours, would then allow the initiation of a further discharge of OT from the CL and hence a luteolytic pulse of PGF2a from the uterus. In addition to desensitization of the PGF2a receptor on the large luteal cell, other factors may also play a role in regulating the duration and frequency of luteolytic pulses of PGF2a from the uterus,. For example, down-regulation of endometrial receptors for OT (14,341 would also abrogate uterine PGF2a secretion. Thus, the time required to regenerate receptors for OT in the endometrium may also influence the interval between uterine pulses of PGF2a. Since the amount of the finite pool of OT in the CL released during the first pulse is very large (- 50%; unpublished observation) and the number of receptors for OT in the endometrium is very low at this time (35.36). desensitization of the endometrial rOT is liable to occur. Desensitization of receptors for OT is most likely to occur between the early luteolytic pulses of PGIQa. as the pulse interval at this time is long (9 - 14 hours) and later decreases to about 6 hours (28). Thus, desensitization of the CL to PGF2a may play a more dominant role as luteolysis progresses. Finally, the initial release of OT from the CL may be brought about by changes in the frequency of a centrally located OT-pulse generator. An increase in OT pulse frequency from the posterior pituitary would be expected to stimulate low levels of uterine PGF2a secretion, and thus provide the initial trigger for the release of luteal OT (13.37). Further studies are needed to evaluate the contribution of these and other factors which may contribute to the initiation and frequency of the pulsatile release of PGF~cx from the endometrium during luteolysis in this species.
176
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
We thank Dr. James Lauderdale of the Upjohn Company for the gift of Lutalyse used in this study, and Dr. A.P.F. Flint for supplying the antibody to oxytocin. We wish to acknowledge the valuable technical assistance of William A. Silva and Christine A. F&&inin conducting this St@.
1. McCracken J.A. J.C. Carlson, M.E. Clew, J.R Goding, D.T. Baird, K. Green, and B. Samuelsson. Prostaglandin F2a identified as the luteolytic hormone in the sheep. Nature (New Biol) =:129. 1972. 2. Thorbum G.D.. RI. Cox. W.B. Currie, B.J. Restall. and W. Schneider. Prostaglandin F and progesterone concentrations in the utero-ovarian venous plasma of the ewe during the oestrous cycle and early pregnancy. J. Reprod. Fert. (supple.) u: 151. 1973. 3. Barcikowski B.. J.C. Carlson, L. Wilson, and J.A. McCracken. The effect of endogenous and exogenous estradiol-178 on the release of prostaglandin F2a from the ovine uterus. Endocrinology s: 1340. 1974.
4. Fairclough R.J., L.G. Moore, L.T. McGowan, AJ. Peterson, J.F. Smith, H.R. Tervit. and W.B. Watkins. Temporal relationship between plasma concentrations of 13,14dihydro-15keto prostaglandin F and neurophysin I/II around luteolysis in sheep. Prostaglandins a:199. 1980. 5. Zarco L., O.H. Stabenfeldt. S. Basu. G.E. Bradford. and H. Kindahl. Modification of prostaglandin F2a synthesis and release in the ewe during the initial establishment of pregnancy. J. Reprod. Fert. =:527. 1988. 6. Schramm W., L. Bovaird, M.E. Glew. G. Schramm, and J.A. McCracken. Corpus luteum regression induced by ultra-low pulses of prostaglandin F2a. Prostaglandins %:347. 1983. 7. Flint. A.P.F. and E.L. Sheldrick. Evidence for a systemic role for ovarian oxytocin in luteal regression in sheep. J. Reprod. Fert. =:215. 1983. 8. Hooper.S.B., W.B. Watkins. and G.D. Thorbum. Oxytocin. oxytocin-associated neurophysin and prostaglandin F2a concentrations in the utero-ovarian vein of pregnant and non-pregnant sheep. Endocrinology m:2590. 1986.
9. Moore, L.G.. V.J. Choy. RL. Elliot, and W.B. Watkins. Evidence for the pulsatile release of PGF2a inducing the release of ovarian oxytocin during luteolysis in the ewe. J. Reprod. Fert. a: 159. 1986. 10. McCracken. J.A.
the ovine 1980.
uterus.
FEBRUARY
Hormone receptor control of prostaglandin F2a secretion
Advances
in Prostaglandin
1992 VOL. 43 NO. 2
and Thromboxane
Research
by &1329.
177
PROSTAGLANDINS 11. Flint A.P.F. and E.L. Sheldrick. Ovarian oxytocin is stimulated by prostaglandins. Nature (London) =:587. 1982. 12. Watkins W.B. and L.G. Moore. Effect of the systemic infusion of PGF2a and 13.14-dihydro-15-keto PGF2a on the release of oxytocin-associated neurophysin from the ovary In the ewe. J. Reprod. Fert. &: 105. 1987. 13. Lamsa J.C., S.J. Kot, J.A. Elderling, M.G. Nay, and JA. McCracken. Prostaglandin F2a-stimulated release of ovarian oxytocin in the sheep In viva: threshold and dose dependency. Biol. Reprod. 4Q:1215. 1989. 14. McCracken J.A.. W. Schramm. and W.C. Okulicz. Hormone receptor control of pulsatile secretion of PGF2a from the ovine uterus during luteolysis and its abrogation during early pregnancy. Anim. Reprod. Sci. 2:3 1. 1984. 15. Flint A.P.F.. E.L. Sheldrick. T.J. McCann, and D.S.C. Jones. Luteal oxytocin: characteristics and control of synchronous episodes of oxytocin and PGF2a secretion at luteolysis in ruminants. Dom. Anim. Endocrinology 1: 111. 1990. 16. Balapure A.K.. I.C. Caicedo, K. Kawada. D.S. Watt, C.E. Rexroad Jr., and T.A. Fitz. Multiple classes of prostaglandin F2a binding sites in subpopulations of ovine luteal cells. Biol. Reprod&:385. 1989. 17. Rodgers R.J.. J.D. O’Shea, J.K. Findlay, A.P.F. Flint, and E.L. Sheldrick. Large luteal cells the source of luteal oxytocin in the sheep. Endocrinology m:2302. 1983. 18. Rodgers R.J.. J.D. O’Shea, and J.K. Findlay. Progesterone production in vitro by small and large luteal cells. J. Reprod. Fert. 69: 113. 1983. 19. Goding J.R. J.A. McCracken, and D.T. Baird. The study of ovarian function in the ewe by means of a vascular autotransplantation technique. J. Endocrinology =:37. 1967. 20. McCracken J.A.. D.T. Baird, and J.R Goding. Factors affecting the secretion of steroids from the transplanted ovary in the sheep. Recent Prog. Hor. Res. =:537. 1971. 21. McCracken J.A.. A. Uno. J.R. Goding. Y. Ichikawa, and D.T. Baird. The in vlvo effects of sheep pituitary gonadotrophins on the secretion of steroids by the autotransplanted ovary in the ewe. J. Endocrinology s:425. 1969. 22. Little T.M. and F.J. Hills. Agricultural Experimentation: Design and Analysis. John Wiley and Sons, NY NY. 1978. 23. Hirst J.J., G.E_ Rice, G. Jenkln, and G.D. Thorbum. Secretion of oxytocin and progesterone by ovine corpora lutea in vitro. Biol. Reprod. a: 1106. 1986. 24. Hirst J.J., G.E. Rice, G. Jenkin, G.D. Thorbum. Control of oxytocin secretion by ovine corpora lutea: effects of arachidonic acid, phospholipase and prostaglandins. Endocrinology =:774. 1988. 25. Wathes D.C., J. Barret, C.L. Gilbert, and 2. Yaron. Regulation of ovarian oxytocin In: Recent Progress in Posterior Pituitary secretion in domestic ruminants.
178
FEBRUARY 1992 VOL. 43 NO. 2
PROSTAGLANDINS Hormones. p. 227.
Eds. S Yoshida.
L Shore.
Elsevier Science Publishers
26. Green K. Determination of prostaglandins Obstet. Gynecol. Stand. (suppl.) =:15. 1979. 27. Flint A.P.F. and E.L. Sheldrtck. pregnancy. J. Reprod. Fert. =:831.
in body
fluids
BV, Holland,
1988.
and tissues.
Acta
Ovarian oxytocin and the maternal 1986.
recognition
of
28. Lamsa J.C. and JA. McCracken. Luteolytic pulses of PGF2a secretion from the ovine uterus: mechanisms regulating their frequency and triggering their release. Biol. Reprod. a (suppl. 1): 318 (Abstr.). 1986. 29. Ivell R, N. Hunt, N. Abend. B. Bra&man. McCracken. Structure and ovarian expression Reprod. Fert. Devel. 2:703. 1990.
D. Nolmeyer, J.C. Lamsa. and J.A. of the oxytocin gene in the sheep.
30. Fitz T.A.. M.H. Mayan, H.R. Sawyer, and G.D. Niswender. Characterization of two steroidogenic cell types in the ovine corpus luteum. Biol. Reprod. 27:703. 1982. 31. Powell W.S.. S. Hammarstrom. B. Samuelsson. and B. Sjoberg. Prostaglandin receptor in human corpora lutea. Lancet 1974. p. 1120. June 1, 1974. 32. Powell W.S., S. Hammarstrom. and B. Samuelsson B. Prostaglandin in ovine corpora lutea. Eur. J. Biochem. a: 103. 1974. 33. Riley J.C.M. and J.C. Carlson. Calcium-regulated membrane corpus luteum regression in the rat. Biol. Reprod. =:77. 1985.
F2a
F2a receptor
rigidification
during
34. Flint A.P.F. and E.L. Sheldrick. Continuous infusion of oxytocin prevents induction of uterine oxytocin receptor and blocks luteal regression in cyclic ewes. J. Reprod. Fert. =:623. 1985. 35. Roberts J.S., J.A. McCracken, J.E. Gavagan. and M.S. Soloff. Oxytocin-stimulated release of prostaglandin F2a from ovine endometrium in vitro: correlation with estrous cycle and olcytocin receptor binding. Endocrinology 99: 1107. 1976. 36. Htxon J.E. and A.P.F. Flint. Effects of a luteolytic dose of oestradiol benzoate on uterine oxytocin receptor concentrations, phosphoinositide turnover and prostaglandin F2o in sheep. J. Reprod. Fert. zQ:457. 1987. 37. McCracken J.A.. T.T. Smith, J.C. Lamsa. and A.G. Robinson. The effect of estradiol-17B (El and progesterone (PI on the oxytocin (OT) pulse generator in the ovariectomized sheep. Biol. Reprod. 44 (supple. 1): 137 (Abstr.). 1991. Editor:
H.
Behrman
Received:
FEBRUARY 1992 VOL. 43 NO. 2
9-13-91
Accepted:
1-24-91
179
IN VIVO DESENSITIZATION OF A HIGH AFFINITY RECEPTOR IN THE OVINE CORPUS LUTEUM
PGFZos
J.C. Lamsa, R.A. Cu.&man, M.G. Nay and J.A. McCracken Worcester Foundation for Experimental Biology Shrewsbury, Massachusetts 01545
ABSTRACT
The corpus luteum (CL) of the sheep exhibits a differential sensitivity to PGF2a in vivo in terms of an increase in oxytocin (OT) secretion and a decrease in progesterone secretion, pointing to the presence in vivo of both high and low affinity receptors for PGF2a. The presence of the high affinity PGF2a receptor was assessed by monitoring the secretion rate of OT from the ovine CL in response to subluteolytic infusions of PGF2a. Rapid desensitization to PGFZCX occurred after only one hour of infusion, while a minimum rest period of six hours was required to restore sensitivity. The possibility that these findings could be explained by the depletion and resynthesis of OT was excluded by demonstrating an increase in OT secretion rate with supra-physiological levels of PGF2a two hours after desensitization. Collectively, these results indicate the presence of a high affinity receptor for PGF2a in the ovine CL which exhibits desensitization and recovery in vivo. The temporal nature of the desensitization and recovery of the high affinity PGF2a receptor controlling luteal OT secretion may contribute to the pulsatile nature of PGF2a release from the ovine uterus.
This work was supported by NIH grant HD-08129. Reprint requests to: Dr. John A. McCracken, Worcester Foundation Experimental Biology, 222 Maple Avenue, Shrewsbury, MA 01545.
FEBRUARY 1992 VOL. 43 NO. 2
for
165
PROSTAGLANDINS
INTRODUCTION Prostaglandin Fzo: (PGF2a). the luteolytic hormone in sheep (1). is released from the uterus as a series of hour long pulses at intervals of about six hours at the end on the luteal phase (2.3.4.5). Since the intraovarian administration of exogenous PGF2a as a series of discrete one hour long pulses given at six hour intervals is more effective in causing luteolysis than a constant infusion (61, a specific pulsatile pattern of PGF2a release from the uterus appears to be advantageous for luteolysis in this species. Thus, the mechanisms regulating the pulsatile release of PGF2a from the uterus are important in the process of luteolysis. During luteolysis, pulses of PGF2a and also its metabolite occur synchronously with elevations of oxytocin (OT) or OT neurophysin in the peripheral circulation (7.8.9). Exogenous OT causes the immediate release of PGF2a from the uterus when the oxytocin receptor (rOT) is present in the endometrium (10). In addition, PGF2a has been found to cause the release of OT from the corpus Thus, a positive feedback loop develops luteum (CL) (11,12,13). between the CL and the uterine endometrium, culminating in a high magnitude pulse of PGF2a. which initiates luteolysis (14). The factors regulating this positive feedback loop, and thus the duration of and intervals between luteolytic pulses of PGF2a. are controversial. It has been proposed that the OT in the CL is totally depleted by each hour long pulse of PGF2a, and until the supply of luteal OT is restored, no further pulses of PGF2a can occur (7,15). Alternatively, luteal OT may down-regulate its endometrial receptor, and thereby terminate PGF2a The interval required to regenerate rOT would thus secretion. regulate the timing of subsequent pulses of PGF2a from the uterus (14). However, an additional level of regulation may exist within the CL itself. The corpus luteum of the sheep is differentially sensitive to PGFZa, possessing a high sensitivity for increased OT secretion and and a low sensitivity for decreased progesterone secretion (13). This observation suggests the presence of both high and low affinity PGF2a receptors in luteal tissue. Recently, it has been reported that two classes of PGF2a receptor exist within the CL of the sheep (161. A high affinity receptor was found on the large luteal cell (the source of luteal OT: 17) and a low affinity receptor was found on both the small and large luteal cell (both of which secrete progesterone: 18). In the present study, we have explored the physiological role of the high affinity PGF2a receptor by monitoring the secretion rate of OT from the autotransplanted ovary in vivo in response to either the long term constant infusion or short term intermittent infusions of subluteolytic levels of PGF2a. The objectives of this study were to determine a) if desensitization of the ovine CL to PGF2a occurs in vivo; bl if such desensitization to PGF2a exists, how long does the CL remain refractory to subsequent infusions of PGFZa; and cl if such refractoriness occurs, can it be explained by desensitization of a high affinity receptor, or by the depletion and subsequent resynthesis of luteal OT?
166
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
MATERIAL!3 AND METHODS Exnerimental
animals
Merino sheep with the left ovary autotransplanted with vascular anastomoses into a jugulo-carotid loop and with the right ovary removed were used as experimental subjects (19.20). To perform in ~yi experiments in sheep with a CL of known age, estrus was induced by two intramuscular injections of 5 mg PGF2a (Lutalyse. Upjohn Co., Kalamazoo MI) administered at an interval of 4 hours. Estrus. which was determined by vasectomized rams, occurred 48 - 72 hours after PGF2a administration (estrus = cycle day 0). All infusion experiments, each in a separate animal, were performed 12 days after estrus in order to mimic conditions in cyclic sheep just prior to the beginning of luteolysis. Sheep were housed in metabolism cages with free access to food and water for the duration of each infusion experiment. Infusion of the ovarv in vivo with PGF2a Twenty-four hours prior to each experiment the left carotid artery contained in the isolated jugulo-carotid loop was cannulated as On the day of each experiment, described previously (13.20.21). stock solutions of PGF2a were prepared by dissolving PGFZa (PGFZa tromethamine salt: Cayman Chemical Co., Ann Arbor MI) in 100% ethanol at a concentration equivalent to 1 mg PGFZa/ml. The concentration was then adjusted for infusions with sterile pyrogen and preservative free aqueous 0.9% sodium chloride (Abbott Laboratories, cuff was North Chicago 11). For all infusions, a sphygmomanometer placed on the cranial side of the jugulo-carotid loop. Inflation of the cuff to 200 mm Hg insured the direct infusion of PGFZa into the ovary via the cannula in the carotid artery. Low concentrations of PGF2a were administered by means of an infusion pump (Harvard Apparatus, Cambridge MA) set to deliver 2.0 ml/hr (25 pg/min. 100 pg/min, or 250 pg/min PGFZa) . The infusion rates of PGFZa chosen for this study were far below the level necessary to cause the depression of progesterone secretion by direct arterial infusion of the ovary in this species (6.13). Collection of ovarian venous blood At the time of cannulation of the left carotid artery, a teflon catheter (1.5 mm i.d. x 40 cm) was inserted into the cranial portion of the left jugular vein as described previously (13.20). Upon completion of the cannulation procedure, each animal was given 5000 IU heparin as an anticoagulant i.v. every 4 hours throughout the course of each experiment. Timed samples of ovarian venous blood (5 ml) were collected into chilled tubes over ice by manual occlusion of the jugular vein on the cardiac side of the transplanted ovary. Blood flow was calculated by measuring the time required for each 5 ml blood sample to flow freely from the venous catheter (about 30 seconds). Immediately after collection, each blood sample was centrifuged at 40
FEBRUARY
1992 VOL. 43 NO. 2
167
PROSTAGLANDINS
C for 30 minutes at 2000 rpm, and the hematocrit and plasma volume of each sample was measured. This yielded the plasma flow (volume of plasma/minute), and hence allowed the calculation of the secretion rate of OT (mass/unit time) from the ovary (13). Aliquots of plasma were removed promptly and stored at -200 C until assayed for GT. Oxytocin radioimmunoassay OT radioimmunoassay (RIA) was performed as described previously (13). The mean extraction recovery of sH- OT (Amersham, Arlington Heights 11) from plasma was 75% and OT values were Intra- and interassay coefficients of corrected for extraction losses. variation were 10.8% and 14.3% respectively. Sensitivity of the RIA ranged between 10 and 20 pg/tube, as determined by the amount of GT that was two standard deviations above the value of the assay blank. ExDeriment
1
To determine basal levels of OT secretion, ovarian venous blood samples were collected every 15 minutes during a one hour period before the infusion of the vehicle control. Vehicle (sterile aqueous 0.9% NaCl free of pyrogens and preservatives: Abbott Laboratories, North Chicago 11) was infused at a rate of 2 ml/hr directly into the ovarian arterial blood supply for two hours (n = 3). Ovarian venous blood was collected at intervals ranging from 5 to 30 minutes during the infusion To determine if the CL became refractory to the infusion of low levels of PGFzu in terms of GT secretion, PGF2a was infused directly into the ovarian arterial blood supply for a period of 10 hours at a rate of either 25 (n = 4) or 250 pg/min (n = 3). Ovarian venous blood was collected 5, 10, 15 and 30 minutes after the start of the infusion, and at 30 minute intervals thereafter for the duration of the infusion. Five mg PGF2a was given i.m. one hour after stopping the low level PGF2a infusions to determine if any OT remained in the CL. ExDeriment
2
To determine the duration of refractoriness to PGF2a, brief incremental rest periods were introduced between the 2 hr PGF2a infusions. Since the CL was responsive to low levels of PGlQa for only about 1 hour (Expt. 1). PGF2a was infused at a rate of 25 pg/min for six 2 hour periods, each separated by a gradually increasing rest period (n = 2). Rest periods (no PGF2a infusion) of 10, 20. 30, 60 and 90 minutes were introduced between infusions of PGF2a. Ovarian venous blood was collected at 5, 10, 15 and 30 minute intervals during each Five mg infusion and at 5, 10 and 15 minutes between infusions. PGF2a was given i.m. one hour after stopping the low level PGF2o: infusions to determine if any OT remained in the CL.
168
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS EXDeriment
3
Since rest intervals of up to 90 minutes were insufficient to allow the return of OT secretion in response to low level PGF2a infusions, incremental rest periods of longer duration (3 to 9 hours) were introduced between the 2 hour infusions of PGF2a. PGFZCXwas infused for two separate two hour periods in nine sheep at a rate of 100 pg/min with rest periods between infusions of 3 hours (n = 3). 6 hours (n = 3). or 9 hours (n = 3). Ovarian venous blood was collected at 5, 10, 15 and 30 minute intervals during each infusion and at 5, 10, 15 and 30 minutes between infusions. Exneriment
4
To determine if releasable oxytocin remained in the corpus luteum one hour after desensitization to a low level of PGF2a. PGFZa was infused at a rate of 25 pg/min for a two hour period (n = 6). Ovarian venous blood was collected at 5. 10. 15 and 30 minute intervals during the infusion. Five mg PGF2a was given i.m. one hour after stopping the low level PGF2 a infusions to determine if any OT remained in the CL. Statistical analvsis The area under the secretion rate curve was calculated, yielding the mass of oxytocin secreted during each PGFZCI infusion. Significant secretion was defined as occurring when the mass of OT released into the ovarian vein during an infusion of PGF2o: was significantly (p c 0.05) greater than during an equivalent control period. Significance was determined by analysis of variance. Duncans Multiple Range Test was used to determine significance between treatment means (22). RESULTS Exneriment
1
OT secretion remained at the basal level throughout the duration of the infusion of vehicle (figure 1). The infusion of 25 pg/min PGF2a: for a ten hour period resulted in a rapid rise in OT secretion, which peaked within 15 minutes and dropped to control levels by one hour after the start of infusion (figure 2). The pattern of OT secretion in response to 250 pg/min PGF2a for 10 hours showed a similar pattern, although peak levels of OT were higher (figure 2). No further increase in OT secretion was seen during the remainder of the experiment at either dose level, despite the continued infusion of PGFzo. The absence of OT release in response to PGF% was not due to the lack of OT, as a supraphysiological dose of PGF2a (5 mg i.m.) stimulated the release of OT from the CL one hour after the completion of the low level PGF2a infusions.
FEBRUARY
1992 VOL. 43 NO. 2
169
PROSTAGLANDINS SALINE
NG’%Y
VEHICLE
INFUSION
8-
64-
, -1
,
.
.
.
.
,
1
0
TIME
.
.
;
.-.
2
, 3
IN HOURS
Figure 1: The effect of a 2 hour intra-arterial infusion of saline vehicle on oxytocin secretion rate from the ovarian autotransplant on day 12 after estrus (n = 3). NG/MIN
25 FG/MIN
FGF2a
5 MG PCF2a
INlWbAFZTERL4L
IM
4
g
2030z:;;
-1
0
1
2
3
4 TIME
Figure 2: The estrvs during a later by a single (n = 4). Bottom
Exoeriment
6
5 IN
7
8
9
10
11
12
13
14
HOURS
mean oxyfocin secretion rate from the ovarian autotransplant on day 72 after 10 hour intra-arterial infusion of subluteolytic levels of PGF2a followed one hour i.m. injection of a supraphysiological level of PGF2a (5 mg). Top panel: 25 pg!min panel: 250 pg/min (n = 3).
2
The initial two hour long infusion of PGF2a at 25 pg/min (figure 3) showed a pattern of OT release similar to that seen in Expt. 1. Rest intervals of 10, 20, 30, 60 or 90 minutes were insufficient to restore responsiveness to PGF2a. as no increase in OT secretion rate was seen during any of the subsequent 2 hour infusions of PGFza. Again, the
170
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
absence of response to PGFZa in terms of OT release was not due to the lack of luteal GT. as a supraphysiological dose of PGF2a (5 mg i.m.) stimulated the release of OT from the CL one hour after the completion of the experiment. 25 PG/MIN
NG/MIN
PGF2a
IN’TFU-ARTERIAL
5 MC PGF2a
IM
I
0
8
4
TIME
12
16
20
IN HOURS
Flgure 3: The mean oxyiocin secretion rate from the ovarian autotransplant on day 12 after estrus during 6 separate 2 hour intra-arterial infusions of subluteolytic levels of PGF2a (25 pg/min) followed one hour later by a single i.m. injection of a supraphysiological level of PGF2a (5 mg) (n = 2). Rest periods (no PGF2a infusion) of 10, 20, 30, 60 and 90 minutes respectively were introduced between each successive 2 hour infusions of PGF2a.
The infusion of PGF2a into the ovarian artery for a period of 2 hours resulted in a pattern of OT secretion similar to that seen in Expts. 1 and 2, that is the secretion rate of OT was elevated for only one hour. The pattern of OT secretion in response to PGF2a is shown in three representative experiments in figure 4. After a three hour rest interval, a second infusion of PGF2a at a rate of 100 pg/min was unable to induce a significant (p > 0.05) release of OT from the ovary (figure 5). When the rest interval was increased to six hours, PGFZa caused a significant (p < 0.05) release of luteal OT (figure 5). The mass of OT secreted during the second PGF2a infusion was When the significantly lower than during the first infusion (P < 0.05). interval between infusions was increased to 9 hours, PGF2a caused a significant (p < 0.05) release of OT. After the 9 hour interval, the mass of OT secreted during the second infusion was not significantly different (p > 0.05) from the total mass secreted during the first PGFZa infusion (figure 5). Since the results of Expt. 3 demonstrate that a difference between the mass of oxytocin secreted after desensitization depends on the interval between infusions, a pilot experiment was performed in which rest intervals (3, 6 and 9 hours) were introduced between PGFZa infusions (250 pg/min) in a single animal. Oxytocin levels remained at control levels after a 3 hour rest interval. However. after
FEBRUARY
1992 VOL. 43 NO. 2
171
PROSTAGLANDINS
hours PGF2a again induced the release of oxytocin, with OT secretion rate peaking at 25 ng/min. After a 9 hour rest interval, OT secretion was stimulated by PGF2a infusion, peaking at 37 ng/min (figure 6).
6
NG/MIN PGna
30
3hr vINIERVAL
20
b
2
50
3
40
E
30
$
20
6hr INIERVAL
cn g E 0
lo 0 FGF2a
0
Pow
30
v
9hr INTERVAL
V
D
20
10
0 -1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TIME IN HOURS Figure 4: The effect of 2 separate 2 hour intra-arterial infusions of subluteolytic levels of PGFZa (100 pg/min) given at intervals of 3 hours (top panel), 6 hours (middle panel) or 9 hours (bottom panel) onday 12 after estrus in three individual sheep .
172
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
a
T
a
n
CONI-ROL
0
PCF2a 1st Inf.
H
FGK!a 2nd Inf.
6hr
3hr REST
9hr
INTERVAL
Figure 5: The relative mass of oxytocin secreted in response to the second 2 hour infusion of PGF2a (100 pg/min) expressed as a percentage of the mass of oxytocin secreted during the first 2 hour infusion of PGF2a (100 pg/min). N = 3 for each rest interval. The designation “a” indicates a significant difference from control values for that rest interval, while b” indicates a significant difference from the first infusion.
6hr
9hr
Z-+-M
4
-2
0
2
4
6
8
10
12
14
16
18
20
22
24
26
TIME IN HOURS Flgore 6: Oxytocin secretion during repetitive 2 hour intra-artetial infusions of subluteolytic levels of PGF2a (250 pg/min) in a single sheep. The 2 hour infusions of PGF2a were given at intervals of 3, 6 and 9 hours on day 72 after estrus.
Exneriment 4 The infusion of PGF2a at 25 pg/min into the ovarian artery for a period of 2 hours resulted in a pattern of OT secretion similar to that seen in Expts. 1, 2 and 3, that is the secretion rate of OT was elevated for only one hour and then declined. However, the intramuscular
FEBRUARY
1992 VOL. 43 NO. 2
173
PROSTAGLANDINS
administration of 5 mg: PGF& secretion rate (figure 7): 25 PG/MIN
-1
0
i.m. resulted
PGFZa
1
in a rapid rise in OT
5 MC PGF2a IM
2
3
4
5
TIME IN HOURS Figure 7: The mean secretion rate of ovarian oxytocin in 6 sheep. Following desensitilation to PGFZU, a single i.m. injection of 5 mg PGF2a was administered one hour after completion of a 2 hour intra-arterial infusion of subluteolytic levels of PGF2a (25 pg/min) on day 72 after esters.
DISCUSSION It is well established that the CL of the sheep will release OT h m in response to exogenous PGF2a (11,121, and that small elevations of PGF2a in ovarian arterial blood elicit such a response (13). In the present study, a continuous infusion of subluteolytic levels of PGF2a into the ovarian artery initiated an immediate release of luteal OT followed by desensitization in the presence of continued low level infusion of PGFZa. The return of luteal sensitivity to PGFZa-induced OT secretion develops over time. While the mass of OT secreted following a 6 hour rest interval is significantly higher than control values, the amount of OT released was still significantly lower than that released during the first infusion. When the rest interval is increased to 9 hours, the mass of OT released did not differ significantly from the total amount secreted during the first PGF2a infusion, indicating that full recovery may occur between 6 and 9 hours following desensitization. It is worthy of note that the cessation of OT secretion observed in this study was not due to the depletion of luteal OT, as the i.m. administration of a supraphysiological dose of PGF2a stimulated OT secretion as soon as one hour after the completion of the low-level infusion of PGF2a. Paradoxically, others have been unable to demonstrate the release of OT by PGF2a from ovine luteal tissue ti vitro (23.24.25). One explanation for the failure of PGF2a to release OT from luteal tissue in vitro is that endogenous PGFzu, generated as a result of trauma associated with the handling of tissue (26). may desensitize the PGF2a receptors in luteal tissue and thus mask the response to exogenous PGF2a. A rest period of at least 6 hours is required before luteal responsiveness to low levels of PGF2a is
174
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
restored in viva. To mimic in vivo conditions, it may be necessary to observe a similar rest period in vitro, thus allowing the return of luteal sensitivity to PGF2a. It has been proposed that the release of OT from the ovine CL would form a positive feedback loop via PGF2a from the uterus, and that the secretion of OT would continue until all of the OT in the CL was exhausted (27). Thus, the interval between large pulses of PGFza from the uterus would be regulated by the time required for either & novo resynthesis of luteal OT or replenishment of OT from stored These possibilities are difficult to reconcile with prohormone (15). the findings of the present study, where the continuous infusion of low levels of PGF2a caused OT secretion for only a period of about one hour. In addition, the observed period of desensitization lasts 6 to 9 hours, similar to the intervals seen between luteolytic pulses of PGF2a in the cyclic animal (28). Ivell et al. (29) have shown that levels of OTmFWA are high only early in the ovine estrous cycle, and are very low towards the end of the luteal phase, when OT resynthesis has been suggested to occur. The absence of OT-mFWA at this time does not preclude replenishment of luteal OT from stored prohormone. However, as shown in the present study, luteal stores of OT are not depleted, since supraphysiological concentrations of PGFZCI will still release OT after a rest period of only one hour. This period is considerably shorter than the 6 to 12 hours observed between releases of OT seen during luteolysis in vivo (28). Lastly, a pharmacological dose of PGFZCX(5 mg i.m.1 will release a smaller mass of oxytocin after the infusion of 250 pg/min PGF2a compared with the 25 pg/min PGF2a infusion (figure 2). in keeping with the concept that the CL contains only a finite pool of OT (13). The high sensitivity of the CL to PGF2a is consistent with current knowledge of the cell types present in the CL. The CL is composed of two major lutein cell types: the large luteal cell and the small luteal cell. Small luteal cells possess a large number of LH receptors (30) and secrete most of the progesterone (18). Large luteal cells of the sheep have been identified as the source of ovarian OT (17). possess few LH receptors, but have the majority of PGFZa receptors (30). Specific receptors for PGF2a have been found in the plasma membrane from human, bovine (311, and ovine (32) corpora lutea. More recently, Balapure et al. (16) provided evidence that there are two classes of PGFZa receptors in the ovine CL, that is, a high affinity receptor on the large luteal cell (the source of luteal OT; 17). and a low affinity receptor on both large and small luteal cells. Very low infusion rates of PGF2a can selectively release OT, presumably by interacting with the high affinity receptor on the large luteal cell (13). while much higher infusion rates of PGFZa (in the order of 1000 pg/min into the ovarian artery) are required to cause the decline of progesterone secretion in vivo (6). It therefore seems likely that low levels of PGF2a. emanating from the uterus and acting on the high affinity receptors on the large luteal cell, triggers the release of luteal OT. The continued presence of PGF2 a will down-regulate the high affinity receptors on the large luteal cells, resulting in desensitization. In the naturally cycling animal at the time of luteolysis, the
FEBRUARY
1992 VOL. 43 NO. 2
175
PROSTAGLANDINS
supplemental release of OT from the CL, acting on rising levels of receptors for OT in the endometrium, appears to cause the major episodic secretions of PGFZa which are seen during luteolysis in the sheep. Such high levels of PGF2a initiates luteolysis, apparently by acting on low affinity PGF2a receptors on both the large and small luteal cells. Under physiological conditions, the higher levels of PGF2a initiated by luteal OT release may desensitize the both the high and low affinity receptor. Evidence supporting this concept is threefold. It has been shown that the pulsatile administration of luteolytic levels of PGF2a is more effective in causing luteolysis than a constant infusion (6). Furthermore, Watkins and Moore (12) demonstrated that the ovine CL becomes refractory to high levels of PGF2a administered into the peripheral circulation in vivo, as determined by the release of OT neurophysin. Finally, a very high, pharmacological dose of PGF2a will cause the release of luteal OT following desensitization to subluteolytic levels of PGF2a. This supraphysiological level of PGF2u may act through the low affinity. high capacity PGF2a receptor present on both small and large luteal cells. Alternatively, the very high levels of PGF2a may cause plasma membrane changes by direct intercalation into the membrane (331. The function of the desensitization mechanism may be to conserve the finite pool of OT, preventing depletion of luteal OT early in luteolysis, and thus insuring that sufficient OT remains for the initiation of several pulses of PGF2a. Recovery of the high affinity receptors, after a minimum rest period of about 6 hours, would then allow the initiation of a further discharge of OT from the CL and hence a luteolytic pulse of PGF2a from the uterus. In addition to desensitization of the PGF2a receptor on the large luteal cell, other factors may also play a role in regulating the duration and frequency of luteolytic pulses of PGF2a from the uterus,. For example, down-regulation of endometrial receptors for OT (14,341 would also abrogate uterine PGF2a secretion. Thus, the time required to regenerate receptors for OT in the endometrium may also influence the interval between uterine pulses of PGF2a. Since the amount of the finite pool of OT in the CL released during the first pulse is very large (- 50%; unpublished observation) and the number of receptors for OT in the endometrium is very low at this time (35.36). desensitization of the endometrial rOT is liable to occur. Desensitization of receptors for OT is most likely to occur between the early luteolytic pulses of PGIQa. as the pulse interval at this time is long (9 - 14 hours) and later decreases to about 6 hours (28). Thus, desensitization of the CL to PGF2a may play a more dominant role as luteolysis progresses. Finally, the initial release of OT from the CL may be brought about by changes in the frequency of a centrally located OT-pulse generator. An increase in OT pulse frequency from the posterior pituitary would be expected to stimulate low levels of uterine PGF2a secretion, and thus provide the initial trigger for the release of luteal OT (13.37). Further studies are needed to evaluate the contribution of these and other factors which may contribute to the initiation and frequency of the pulsatile release of PGF~cx from the endometrium during luteolysis in this species.
176
FEBRUARY
1992 VOL. 43 NO. 2
PROSTAGLANDINS
We thank Dr. James Lauderdale of the Upjohn Company for the gift of Lutalyse used in this study, and Dr. A.P.F. Flint for supplying the antibody to oxytocin. We wish to acknowledge the valuable technical assistance of William A. Silva and Christine A. F&&inin conducting this St@.
1. McCracken J.A. J.C. Carlson, M.E. Clew, J.R Goding, D.T. Baird, K. Green, and B. Samuelsson. Prostaglandin F2a identified as the luteolytic hormone in the sheep. Nature (New Biol) =:129. 1972. 2. Thorbum G.D.. RI. Cox. W.B. Currie, B.J. Restall. and W. Schneider. Prostaglandin F and progesterone concentrations in the utero-ovarian venous plasma of the ewe during the oestrous cycle and early pregnancy. J. Reprod. Fert. (supple.) u: 151. 1973. 3. Barcikowski B.. J.C. Carlson, L. Wilson, and J.A. McCracken. The effect of endogenous and exogenous estradiol-178 on the release of prostaglandin F2a from the ovine uterus. Endocrinology s: 1340. 1974.
4. Fairclough R.J., L.G. Moore, L.T. McGowan, AJ. Peterson, J.F. Smith, H.R. Tervit. and W.B. Watkins. Temporal relationship between plasma concentrations of 13,14dihydro-15keto prostaglandin F and neurophysin I/II around luteolysis in sheep. Prostaglandins a:199. 1980. 5. Zarco L., O.H. Stabenfeldt. S. Basu. G.E. Bradford. and H. Kindahl. Modification of prostaglandin F2a synthesis and release in the ewe during the initial establishment of pregnancy. J. Reprod. Fert. =:527. 1988. 6. Schramm W., L. Bovaird, M.E. Glew. G. Schramm, and J.A. McCracken. Corpus luteum regression induced by ultra-low pulses of prostaglandin F2a. Prostaglandins %:347. 1983. 7. Flint. A.P.F. and E.L. Sheldrick. Evidence for a systemic role for ovarian oxytocin in luteal regression in sheep. J. Reprod. Fert. =:215. 1983. 8. Hooper.S.B., W.B. Watkins. and G.D. Thorbum. Oxytocin. oxytocin-associated neurophysin and prostaglandin F2a concentrations in the utero-ovarian vein of pregnant and non-pregnant sheep. Endocrinology m:2590. 1986.
9. Moore, L.G.. V.J. Choy. RL. Elliot, and W.B. Watkins. Evidence for the pulsatile release of PGF2a inducing the release of ovarian oxytocin during luteolysis in the ewe. J. Reprod. Fert. a: 159. 1986. 10. McCracken. J.A.
the ovine 1980.
uterus.
FEBRUARY
Hormone receptor control of prostaglandin F2a secretion
Advances
in Prostaglandin
1992 VOL. 43 NO. 2
and Thromboxane
Research
by &1329.
177
PROSTAGLANDINS 11. Flint A.P.F. and E.L. Sheldrick. Ovarian oxytocin is stimulated by prostaglandins. Nature (London) =:587. 1982. 12. Watkins W.B. and L.G. Moore. Effect of the systemic infusion of PGF2a and 13.14-dihydro-15-keto PGF2a on the release of oxytocin-associated neurophysin from the ovary In the ewe. J. Reprod. Fert. &: 105. 1987. 13. Lamsa J.C., S.J. Kot, J.A. Elderling, M.G. Nay, and JA. McCracken. Prostaglandin F2a-stimulated release of ovarian oxytocin in the sheep In viva: threshold and dose dependency. Biol. Reprod. 4Q:1215. 1989. 14. McCracken J.A.. W. Schramm. and W.C. Okulicz. Hormone receptor control of pulsatile secretion of PGF2a from the ovine uterus during luteolysis and its abrogation during early pregnancy. Anim. Reprod. Sci. 2:3 1. 1984. 15. Flint A.P.F.. E.L. Sheldrick. T.J. McCann, and D.S.C. Jones. Luteal oxytocin: characteristics and control of synchronous episodes of oxytocin and PGF2a secretion at luteolysis in ruminants. Dom. Anim. Endocrinology 1: 111. 1990. 16. Balapure A.K.. I.C. Caicedo, K. Kawada. D.S. Watt, C.E. Rexroad Jr., and T.A. Fitz. Multiple classes of prostaglandin F2a binding sites in subpopulations of ovine luteal cells. Biol. Reprod&:385. 1989. 17. Rodgers R.J.. J.D. O’Shea, J.K. Findlay, A.P.F. Flint, and E.L. Sheldrick. Large luteal cells the source of luteal oxytocin in the sheep. Endocrinology m:2302. 1983. 18. Rodgers R.J.. J.D. O’Shea, and J.K. Findlay. Progesterone production in vitro by small and large luteal cells. J. Reprod. Fert. 69: 113. 1983. 19. Goding J.R. J.A. McCracken, and D.T. Baird. The study of ovarian function in the ewe by means of a vascular autotransplantation technique. J. Endocrinology =:37. 1967. 20. McCracken J.A.. D.T. Baird, and J.R Goding. Factors affecting the secretion of steroids from the transplanted ovary in the sheep. Recent Prog. Hor. Res. =:537. 1971. 21. McCracken J.A.. A. Uno. J.R. Goding. Y. Ichikawa, and D.T. Baird. The in vlvo effects of sheep pituitary gonadotrophins on the secretion of steroids by the autotransplanted ovary in the ewe. J. Endocrinology s:425. 1969. 22. Little T.M. and F.J. Hills. Agricultural Experimentation: Design and Analysis. John Wiley and Sons, NY NY. 1978. 23. Hirst J.J., G.E_ Rice, G. Jenkln, and G.D. Thorbum. Secretion of oxytocin and progesterone by ovine corpora lutea in vitro. Biol. Reprod. a: 1106. 1986. 24. Hirst J.J., G.E. Rice, G. Jenkin, G.D. Thorbum. Control of oxytocin secretion by ovine corpora lutea: effects of arachidonic acid, phospholipase and prostaglandins. Endocrinology =:774. 1988. 25. Wathes D.C., J. Barret, C.L. Gilbert, and 2. Yaron. Regulation of ovarian oxytocin In: Recent Progress in Posterior Pituitary secretion in domestic ruminants.
178
FEBRUARY 1992 VOL. 43 NO. 2
PROSTAGLANDINS Hormones. p. 227.
Eds. S Yoshida.
L Shore.
Elsevier Science Publishers
26. Green K. Determination of prostaglandins Obstet. Gynecol. Stand. (suppl.) =:15. 1979. 27. Flint A.P.F. and E.L. Sheldrtck. pregnancy. J. Reprod. Fert. =:831.
in body
fluids
BV, Holland,
1988.
and tissues.
Acta
Ovarian oxytocin and the maternal 1986.
recognition
of
28. Lamsa J.C. and JA. McCracken. Luteolytic pulses of PGF2a secretion from the ovine uterus: mechanisms regulating their frequency and triggering their release. Biol. Reprod. a (suppl. 1): 318 (Abstr.). 1986. 29. Ivell R, N. Hunt, N. Abend. B. Bra&man. McCracken. Structure and ovarian expression Reprod. Fert. Devel. 2:703. 1990.
D. Nolmeyer, J.C. Lamsa. and J.A. of the oxytocin gene in the sheep.
30. Fitz T.A.. M.H. Mayan, H.R. Sawyer, and G.D. Niswender. Characterization of two steroidogenic cell types in the ovine corpus luteum. Biol. Reprod. 27:703. 1982. 31. Powell W.S.. S. Hammarstrom. B. Samuelsson. and B. Sjoberg. Prostaglandin receptor in human corpora lutea. Lancet 1974. p. 1120. June 1, 1974. 32. Powell W.S., S. Hammarstrom. and B. Samuelsson B. Prostaglandin in ovine corpora lutea. Eur. J. Biochem. a: 103. 1974. 33. Riley J.C.M. and J.C. Carlson. Calcium-regulated membrane corpus luteum regression in the rat. Biol. Reprod. =:77. 1985.
F2a
F2a receptor
rigidification
during
34. Flint A.P.F. and E.L. Sheldrick. Continuous infusion of oxytocin prevents induction of uterine oxytocin receptor and blocks luteal regression in cyclic ewes. J. Reprod. Fert. =:623. 1985. 35. Roberts J.S., J.A. McCracken, J.E. Gavagan. and M.S. Soloff. Oxytocin-stimulated release of prostaglandin F2a from ovine endometrium in vitro: correlation with estrous cycle and olcytocin receptor binding. Endocrinology 99: 1107. 1976. 36. Htxon J.E. and A.P.F. Flint. Effects of a luteolytic dose of oestradiol benzoate on uterine oxytocin receptor concentrations, phosphoinositide turnover and prostaglandin F2o in sheep. J. Reprod. Fert. zQ:457. 1987. 37. McCracken J.A.. T.T. Smith, J.C. Lamsa. and A.G. Robinson. The effect of estradiol-17B (El and progesterone (PI on the oxytocin (OT) pulse generator in the ovariectomized sheep. Biol. Reprod. 44 (supple. 1): 137 (Abstr.). 1991. Editor:
H.
Behrman
Received:
FEBRUARY 1992 VOL. 43 NO. 2
9-13-91
Accepted:
1-24-91
179
