- Email: [email protected]
A new in vivo model for luteolysis using systemic pulsatile infusions of PGF2α
Prostaglandins & other Lipid Mediators 97 (2012) 90–96
Contents lists available at SciVerse ScienceDirect
Prostaglandins and Other Lipid Mediators
A new in vivo model for luteolysis using systemic pulsatile infusions of PGF2␣ J.A. McCracken a,∗,1 , E.E. Custer b,1 , D.T. Schreiber a , P.C.W. Tsang c , C.S. Keator a,2 , J.A. Arosh d a
Department of Animal Science, University of Connecticut, United States Department of Microbiology and Physiological Systems, University of Massachusetts Medical School, United States c Department of Molecular, Cellular, and Biomedical Sciences, University of New Hampshire, United States d Department of Veterinary Integrative Biosciences, Texas A&M University, United States b
a r t i c l e
i n f o
Article history: Received 16 September 2011 Received in revised form 6 December 2011 Accepted 12 January 2012 Available online 25 January 2012 Keywords: Luteolysis Corpus luteum Progesterone Oxytocin PGF2␣
a b s t r a c t A new in vivo model for studying luteolysis was developed in sheep to provide a convenient method for collecting corpora lutea for molecular, biochemical, and histological analysis during a procedure that mimics natural luteolysis. It was found that the infusion of prostaglandin F2␣ (PGF2␣ ) at 20 g/min/h into the systemic circulation during the mid luteal phase of the cycle allowed sufficient PGF2␣ to escape across the lungs and thus mimic the transient 40% decline in the concentration of progesterone in peripheral plasma seen at the onset of natural luteolysis in sheep. Additional 1 h-long systemic infusions of PGF2␣ , given at physiological intervals, indicated that two infusions were not sufficient to induce luteolysis. However, an early onset of luteolysis and estrus was induced in one out of three sheep with three infusions, two out of three sheep with four infusions, and three out of three sheep with five infusions. Reducing the duration of each systemic infusion of PGF2␣ from 1 h to 30 min failed to induce luteolysis and estrus even after six systemic infusions indicating that, not only are the amplitude and frequency of PGF2␣ pulses essential for luteolysis, but the actual duration of each pulse is also critical. We conclude that a minimum of five systemic pulses of PGF2␣ , given in an appropriate amount and at a physiological frequency and duration, are required to mimic luteolysis consistently in all sheep. The five pulse regimen thus provides a new accurate in vivo model for studying molecular mechanisms of luteolysis. © 2012 Elsevier Inc. All rights reserved.
1. Introduction It is well established that the pulsatile release of PGF2␣ from the uterus induces luteolysis in ruminants as well as several other species of domestic animals [1]. In sheep, the pulsatile release of oxytocin (OT) from the posterior pituitary and the corpus luteum (CL) beginning on days 13–14 of the estrous cycle acts via receptors for OT in the luminal cells of the endometrium to induce luteolytic pulses of PGF2␣ [2]. Approximately 5–8 pulses of PGF2␣ , measured as its primary metabolite 13:14-dihydro-15-keto-PGF2␣ (PGFM), are secreted by the uterus during luteolysis in sheep [3]. It was found that a minimum of five 1 h-long pulses of PGF2␣ infused into the arterial supply of the ovary were required to cause luteolysis consistently in all treated sheep [4]. During natural luteolysis in sheep, approximately 1% of uterine PGF2␣ secreted into the uterine
∗ Corresponding author at: Department of Animal Science, White Building, Unit4040, Storrs, CT 06269-4040, United States. Tel.: +1 860 486 6197; fax: +1 860 486 4375. E-mail address: [email protected] (J.A. McCracken). 1 These authors contributed equally to this paper. 2 Present address: Department of Physiology, Ross University School of Medicine, P.O. Box 266, Portsmouth Campus, Picard, Dominica. 1098-8823/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.prostaglandins.2012.01.004
vein is transferred to the adjacent ovarian artery via a countercurrent transfer system in the utero-ovarian vascular plexus [5]. Local transport of PGF2␣ from the uterus to the ovary is required in sheep because of its high clearance rate in the pulmonary circulation [6]. The underlying mechanism of the countercurrent transfer of PGF2␣ in the utero-ovarian vascular plexus of the sheep was shown recently to depend on the expression of the prostaglandin transporter protein in the vasculature of the utero-ovarian plexus [7]. In the latter study, it was found that PGF2␣ levels in ovarian arterial plasma persisted for a duration of 1 h following an OT-stimulated pulse of uterine PGF2␣ suggesting that transfer of PGF2␣ from the utero-ovarian vein to the ovarian artery for a period of 1 h is most likely a prerequisite for successful luteolysis in sheep. In the present study, our goal was to establish a model for luteolysis that mimicked natural CL regression by infusing 1 h-long systemic infusions of PGF2␣ at a physiological frequency in amounts high enough to allow sufficient PGF2␣ to escape across the lungs into the arterial circulation and thus cause luteolysis. Such a model would provide a convenient means of harvesting luteal tissue for biochemical, molecular, and histological analysis during a process that simulates the natural progression of CL regression in sheep. Luteal tissue collected using this model would thus provide a means of determining protein levels of a variety of biochemical and molecular mediators implicated in the natural progression of functional
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
and structural luteolysis in sheep. To establish the amount of PGF2␣ required to be infused into the systemic circulation to simulate an endogenous luteolytic pulse of PGF2␣ , it was calculated that, based on GC/Mass Spectrometry measurements of 40 ng/ml of PGF2␣ in uterine venous blood during luteolysis [5], and a uterine blood flow from each uterine horn of 10 ml/min at the time of luteolysis [8], the whole uterus secretes approximately 1.0 g/min of PGF2␣ into the venous circulation during each endogenous 1 h-long pulse of PGF2␣ . However, most of the PGF2␣ secreted by the uterus is metabolized by the lungs [9,10] thus reducing PGF2␣ levels in systemic arterial plasma to levels insufficient to cause regression of the CL. Initially we tested the luteolytic activity of 10 times the endogenous uterine secretion rate of PGF2␣ of 1.0 g/min by infusing PGF2␣ at 10 g/min for 1 h into a jugular vein of sheep during the mid luteal phase of the cycle. This rate of infusion reduced the level of progesterone (P) in peripheral plasma by approximately 20% which did not equate with the 40% reduction of P in peripheral plasma seen during the first endogenous pulse of PGF2␣ in sheep [3]. We therefore increased the infusion rate of PGF2␣ to 20 times its endogenous uterine secretion rate by infusing PGF2␣ systemically at 20 g/min for 1 h during the mid luteal phase of the cycle. Such an infusion rate of 20 g/min/h induced a 40% drop in P levels identical to that seen during the first endogenous pulse of PGF2␣ . The objectives of the present study therefore were to: 1. Determine the minimum number of such effective systemic pulses of PGF2␣ , given at their endogenous frequency, required to cause luteolysis and the onset of estrus consistently in all treated sheep thus providing a convenient new model for harvesting luteal tissue during a process mimicking natural luteolysis. 2. Determine whether the actual 1 h duration of each pulse of PGF2␣ was also critical for the induction of luteolysis.
2. Materials and methods 2.1. Animals All experimental procedures in sheep were approved by the Institutional Animal Care and Use Committee at the University of Connecticut. Intact cross-bred Suffolk ewes were used in the present study during the fall and winter breeding season. Sheep were housed in small groups in outside sheep pens, fed twice daily, and given water ad libitum. A total of 30 sheep were used to determine the luteolytic effect of systemic pulses of PGF2␣ . Prior to use, all sheep exhibited at least two estrous cycles of 16–18 days (estrus = day 0), as determined by a vasectomized ram. To perform infusion experiments, sheep were brought into a metabolism unit and housed in individual metabolism cages. At least one companion animal was always present in the metabolism room during each experiment. Infusions were performed during the mid luteal phase of the cycle (day 10). At the end of each infusion experiment, sheep were returned to an outside pen and checked for estrous behavior using a vasectomized ram.
2.2. Infusion procedures After placing each sheep in a metabolism cage, a two and a half inch 16-gauge Teflon indwelling cannula (Beckton Dickinson, East Rutherford, NJ) was placed in both left and right jugular veins and secured with three sutures of #1 braided Dacron. A three way stopcock was inserted into the hub of each cannula and flushed with 2 ml heparin/saline (100 iu/ml). The left jugular cannula was used to withdraw blood samples (5 ml) during each experiment. Infusions were carried out using a Harvard Infusion Pump (Harvard Apparatus, Model #600-000) via a Teflon catheter connected to the cannula in the right jugular vein.
91
One hour-long infusions of sterile pyrogen free saline (Abbott Labs, Chicago), PGF2␣ (Pfizer, NY) or PGFM (Cayman Chemicals, MI) were administered systemically into the right jugular cannula. PGF2␣ was infused systemically at a rate of 10 g/min/h and/or at 20 g/min/h. To insure that the primary metabolite of PGF2␣ , 13:14-dihydro-15-keto-PGF2␣ (PGFM) did not contribute to the luteolytic effect of administered PGF2␣ , PGFM was infused at a rate of 20 g/min/h. Control 1 h-long infusions of saline vehicle, and PGF2␣ and PGFM in saline vehicle were given at the standard infusion rate of 0.1 ml/min/h (6 ml/h). A sample size of three sheep (n = 3) was used for all substances and doses used in the present study. The interval between sequential pulsed infusions of PGF2␣ was based on the reported intervals between endogenous pulses of PGFM in cycling sheep [3] which were 15 h between the first and second pulses and 7 h between subsequent pulses. To establish the importance of the duration of each systemic infusion of PGF2␣ , an additional three sheep during mid-luteal phase (day 10) were given infusions of PGF2␣ at 20 g/min for 30 min instead of 1 h at interpulse intervals of 7 h into the right jugular. Because our earlier results with 1 h-long infusions clearly indicated that PGF2␣ was most effective if given before there was any recovery of P levels (see Sections 3 and 4), the initial inter-pulse interval of 15 h was excluded in the 30 min infusions of PGF2␣ . 2.3. Blood sample collections All jugular blood samples were collected using syringes and tubes stored over ice. Immediately after collection, blood samples were centrifuged at 2000 rpm for 20 min at 4 ◦ C. The plasma was removed promptly, taking care not to include the buffy coat, and stored at −20 ◦ C. The start of each infusion experiment was designated as time zero. Control jugular blood samples were collected at −20 and −10 min. Additional blood samples were collected at 10, 20, 30, and 60 min during each systemic infusion, hourly between each infusion, and hourly for 24 h after the last infusion. Blood samples were then collected twice daily for 96 h or until estrus was observed. After each infusion of substances into the right jugular vein and after each collection of blood from the left jugular vein, cannulae were flushed with 2 ml heparin/saline (100 iu/ml). 2.4. Hormone assays The concentration of P in peripheral plasma was measured using a commercially available assay with antibody coated tubes and 125 Ilabeled P (Coat-A-Count, Diagnostic Productions Corporation; Los Angeles, CA). The concentration of OT in peripheral plasma was measured using a commercially available ELISA (Assay Designs; Ann Arbor, MI). We measured OT in peripheral plasma samples during the infusion of 1 h-long systemic infusions of PGF2␣ at 20 g/min to determine if the release of OT paralleled the pattern of release of OT observed during natural luteolysis [1,11]. The intraand inter-assay coefficients of variation for P measurement were 2.8% and 6.5% respectively. The intra- and inter-assay coefficients of variation for OT measurement were less than 5%. 2.5. Statistical analysis Plasma levels of P were analyzed by one-way ANOVA. Differences between means were determined by Tukey’s HSD test. Chi Square analysis was used to determine the relationship between the number of systemic pulses of PGF2␣ administered at mid cycle and the time of onset of estrus. A value of p < 0.05 was considered significant.
92
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
160
140
E
140 Jugular Plasma Progesterone ( % of Controls)
Jugular Plasma Progesterone (% of Controls)
120
100
80
60
40
20
120
100
80
60
40
20
0
0 0
2
4
6
8
10
12
14
16
0
24
48
72
96
120
144
168
Time In Hours
Time In Hours Fig. 1. The effect on peripheral plasma P (% of controls) of a single 1 h-long systemic infusion of PGF2␣ (20 g/min/h) given to sheep (n = 15) on day 10 of the cycle. This number of sheep also includes the data obtained after the first pulse of PGF2␣ (20 g/min/h) in all subsequent experiments. P fell by 40% (p < 0.05), reaching a nadir at 8 h and returning to control values by 16 h, identical to the changes in P levels seen after the first endogenous pulse of PGF2␣ in this species. Arrows indicate the infusion of PGF2␣ and asterisks indicate a significant difference in P levels from controls.
3. Results 3.1. Systemic administration of 1 h-long infusion of PGF2˛ , PGFM, or saline A single 1 h-long systemic infusion of PGF2␣ (n = 3 sheep) given into a jugular vein during the mid luteal phase (cycle day 10) at 10 g/min caused only a 20% drop in P levels (p < 0.05) reaching a nadir at 6 h and recovering to control levels by 12 h. However, since the fall in P did not match the reduction in P levels observed during the first endogenous pulse of PGF2␣ in sheep [3], the rate of PGF2␣ infusion was increased to 20 g/min for 1 h (n = 15 sheep). This number of sheep includes the data obtained after the first pulse of PGF2␣ (20 g/min/h) in all subsequent experiments. As shown in Fig. 1, there was a characteristic transient increase in P levels followed by a 40% decline (p < 0.05), reaching a nadir at 8 h and recovering by 16 h post-infusion, identical to the changes in P levels seen after the first endogenous pulse of PGF2␣ during natural luteolysis in this species [3]. Following recovery of P after a single 1-h pulse of PGF2␣ , there was a normal rise in P levels from days 10 to 13 as seen in natural cycles, followed by CL regression and estrus at the normal expected time. Because of the possibility that some of the luteolytic effect of a pulse of PGF2␣ infused at 20 g/min, might be caused by conversion into its metabolite PGFM, a 1 h-long systemic infusion of PGFM was given to three sheep at the same rate as PGF2␣ (20 g/min/h). This infusion rate of PGFM had no effect (p > 0.05) on P levels either during or for 24 h after the infusion. The systemic infusion of saline vehicle at 0.1 ml/min/h (n = 3 sheep) also
Fig. 2. The effect on peripheral plasma P (% of controls) of two 1 h-long systemic infusions of PGF2␣ (20 g/min/h) given to sheep (n = 3) on day 10 of the cycle at an inter-pulse interval of 15 h. P levels had recovered by 15 h but the second infusion of PGF2␣ caused a greater fall in P that never really fully recovered to controls. Premature regression did not occur with two systemic infusions of PGF2␣ and all the animals had normal cycle lengths of 16–17 days. E indicates the mean time of estrus.
did not reduce P levels (p > 0.05) either during or for 24 h after the infusion. 3.2. Systemic administration of multiple 1 h-long infusions of PGF2˛ A second 1 h-long systemic infusion of PGF2␣ given at 20 g/min/h after an inter-pulse interval of 15 h (n = 3 sheep), caused a second temporary but greater decline in P which lasted longer and did not completely recover to control values (see Fig. 2). Premature CL regression did not occur with two systemic infusions of PGF2␣ and estrus occurred at the expected time, even though P levels did not recover fully to normal mid luteal levels. A third 1 h-long systemic infusion of PGF2␣ given at the same rate as the first two pulses (20 g/min/h) after an inter-pulse interval of 7 h (n = 3 sheep), caused complete regression with the early onset of estrus in one sheep (Fig. 3A). In the remaining two sheep, P levels partially recovered after the third pulse and the animals did not exhibit an early onset of luteolysis and estrus (Fig. 3B). A forth 1 h-long systemic infusion of PGF2␣ given at 20 g/min/h after an inter-pulse interval of 7 h (n = 3 sheep), caused premature CL regression with the early onset of estrus in two sheep (Fig. 4A). In the remaining sheep, P levels showed an initial decline which partially recovered after the forth pulse but both CL regression and estrus occurred at the normal time (Fig. 4B). Lastly, as shown in Fig. 5, a fifth systemic infusion of PGF2␣ (20 g/min/h) given after a further inter-pulse interval of 7 h caused premature regression of the CL and the early onset of luteolysis and estrus in all treated sheep. Thus, as the number of PGF2␣ pulses increased, a greater proportion of sheep (p < 0.05) showed an earlier onset of luteolysis and estrus. This is illustrated in Fig. 6 which shows the rate of decline in plasma P and the relative time of estrus
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
160
160
(A)
E
140
Jugular Plasma Progesterone (% of Controls)
Jugular Plasma Progesterone (% of Controls)
100 80 60 40 20 0
E (B)
140 120 100 80 60
120
100
80
60
40
20
40
0 0
20 20
40
60
80
100
120
140
160
160 E
140
(A)
120 100
Jugular Plasma Progesterone (% of Controls)
40
60
80
100
120
140
180
80 60 40 20 0 E
140
(B)
Fig. 5. The effect on peripheral plasma P (% of controls) of five 1 h-long systemic infusions of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h, 7 h, 7 h, and 7 h given to sheep (n = 3) beginning in the mid luteal phase of the cycle (day 10). In all sheep, plasma P levels fell to <1.0 ng/ml and the sheep were mated by a vasectomized ram between 68 and 72 h after beginning the infusions of PGF2␣ . E indicates the mean time of estrus.
Jugular Plasma Progesterone (% of controls)
0
Fig. 3. The effect on peripheral plasma P (% of controls) of three 1 h-long systemic infusions of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h and 7 h given to sheep (n = 3) during the mid luteal phase of the cycle (day 10). One out of three sheep (A) showed a premature decline in P levels and an early onset of estrus. P levels recovered in the two remaining sheep (B) which did not exhibit premature CL regression or an early onset of estrus. E indicates the mean time of estrus.
160 E
140
E
E
120 100 80 60 40 20 0
0
20
40
60
80
100
120
140
Time In Hours Fig. 6. Comparison of the rate of decline of P levels in sheep that showed luteolysis and an early onset of estrus following three (blue triangles), four (black triangles), or five (red triangles) pulses of PGF2␣ . The non-responders in the 3 and 4 pulse groups are not included in the data. E indicates the average time of estrus following three (blue box), four (black box), or five (red box) pulses of PGF2␣ . It can be seen that the more rapid the decline in P levels, the earlier the animals exhibited estrus. Thus five pulses of PGF2␣ pulses ensures that estrus follows at an appropriate time after the initial onset of luteolysis which may be important for the timing of ovulation and hence for fertility rates in sheep. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.).
120 100 80 60 40 20 0
20
Time In Hours Time In Hours
0
E
140
120
0
93
0
20
40
60
80
100
120
140
160
180
Time In Hours Fig. 4. The effect on peripheral plasma P (% of controls) of four 1 h-long systemic infusion of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h, 7 h, and 7 h given to sheep (n = 3) on day 10 of the cycle. Two out of three sheep (A) showed a premature decline in P levels and an early onset of estrus. P levels recovered in the remaining sheep (B) which did not exhibit premature CL regression or an early onset of estrus. E indicates the mean time of estrus.
in sheep that showed premature luteolysis and estrus in groups receiving 3, 4, or 5 pulses of PGF2␣ . The animals receiving 3 or 4 pulses of PGF2␣ that did not exhibit an early onset of luteolysis and estrus are not included in the data shown in Fig. 6. As shown in Fig. 7, the concentration of OT in peripheral plasma increased during the first pulse of PGF2␣ (20 g/min/h) but declined progressively during each additional systemic pulse of PGF2␣ until it was no longer measurable during the fifth pulse of
94
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
200 300
Jugular Plasma Progesterone (% of Controls)
270
Jugular Plasma Oxytocin (pg/mL)
240 210 180 150 120 90 60 30 (15)
(12)
(9)
(6)
6
12
18
24
30
36
100
50
(3)
0 0
150
42
Time In Hours Fig. 7. The concentration of OT in peripheral plasma during successive systemic infusions of PGF2␣ (20 g/min/h) beginning on cycle day 10. The numbers in parentheses indicate the number of sheep receiving a given number of systemic pulses of PGF2␣ . OT declined progressively until it was no longer measurable during the fifth pulse of PGF2␣ (<15 pg/ml).
PGF2␣ (<15 pg/ml). This result is in agreement with the progressive decline in the levels of OT observed in peripheral plasma during natural luteolysis in sheep [1,11]. To investigate the importance of the duration of each pulse of PGF2␣ in causing luteolysis, PGF2␣ was infused at its effective dose of 20 g/min at inter-pulse intervals of 7 h but the duration of each infusion was reduced to 30 min instead of 1 h. As shown in Fig. 8, P levels showed a characteristic rise during most of the 30 min infusions of PGF2␣ , but the mean level of P at the end of six 30 min-long infusions of PGF2␣ was not significantly different from pre-infusion controls (p > 0.05). Premature luteolysis did not occur with this infusion regimen and all three sheep showed estrus at the normal expected time of 16–17 days. 4. Discussion The first objective of the present study was to confirm the amount of PGF2␣ infused for 1 h into the systemic circulation that would mimic the initial decline in levels of P reported to a occur at the onset of luteolysis during the natural cycle of sheep [3]. We first tested the luteolytic activity of 10 times the endogenous uterine secretion rate of PGF2␣ by infusing PGF2␣ at 10 g/min for 1 h into a jugular vein of three sheep during the mid luteal phase. This rate of infusion reduced the level of P by approximately 20% which did not equate with the 40% reduction of P levels seen during the first endogenous pulse of PGF2␣ [3]. When we increased the infusion rate of PGF2␣ to 20 g/min for 1 h during the mid luteal phase (n = 3 sheep), P levels were reduced by 40% which matched the fall observed after the first endogenous pulse of PGF2␣ during natural luteolysis in this species [3]. The question arose as to whether the amount of PGFM (the primary metabolite of PGF2␣ ) generated from the infused PGF2␣ might have some effect on P secretion by the CL. Accordingly, PGFM was infused at 20 g/min
0 0
8
16
24
32
40
48
Time In Hours Fig. 8. The effect on peripheral plasma P (% of controls) after six 30 min-long systemic infusion of PGF2␣ (20 g/min/30 min) at inter-pulse intervals of 7 h given to sheep (n = 3) on day 10 of the cycle. Although P levels showed a characteristic rise during most of the 30 min infusions of PGF2␣ , the mean level of P at the end of the infusions was not significantly different from controls (p > 0.05). All three sheep showed estrus at the normal expected time of 16–17 days.
for 1 h into the systemic circulation of three sheep during the mid luteal phase. The infusion of PGFM had no effect on P levels during or after the infusion (p > 0.05). This finding was not unexpected since PGFM has a low binding affinity for the PGF2␣ receptor [12]. Moreover, the observed lack of response to PGFM infusions is consistent with our previous finding that the infusion of PGFM directly into the arterial supply of the ovarian transplant model in sheep, given at the same rate as an effective luteolytic infusion of PGF2␣ , had no effect on the secretion rate of either P or OT [13]. Having established an effective systemic infusion rate of PGF2␣ that would cause a physiological decline in P, we then proceeded to determine the effect of additional systemic infusions of PGF2␣ given at time intervals that matched their physiological frequency [3]. When a second infusion of PGF2␣ was given after an inter-pulse interval of 15 h, P levels declined to a greater extent than after the first pulse, and then partially recovered again in all sheep and the onset of luteolysis and estrus occurred at the normal time of 16–17 days. After a third pulse given at an inter-pulse interval of 7 h, one sheep showed an early onset of luteolysis and estrus, while P levels recovered in two out of three treated sheep, both of which showed luteolysis and estrus at the normal time. After a fourth pulse of PGF2␣ , P levels recovered in one sheep while the other two sheep showed an early onset of luteolysis and estrus. Finally, the infusion of five 1 h-long systemic infusions of PGF2␣ given during the mid luteal phase (day 10) caused the early onset of luteolysis and estrus in all treated sheep. The finding that five 1 h-long infusions of PGF2␣ were required to cause luteolysis consistently in all animals is in agreement with our earlier studies in sheep where it was determined that a minimum of five episodic 1 h-long pulses of low levels of PGF2␣ infused directly into the arterial supply of the transplanted ovary were required to cause luteolysis in all treated
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
sheep [4,14]. In these studies, the total amount of PGF2␣ infused as pulses was 1/40th of the amount required to cause luteolysis when PGF2␣ was given as a constant infusion, suggesting that a pulsatile regimen of PGF2␣ is advantageous for luteolysis in sheep and most likely in other species. In a recent study in cows [15], four sequential intrauterine injections of PGF2␣ (0.25 mg or 1.0 mg every 12 h) resulted in a continuous steady decline in P levels with shortening of the estrous cycle, whereas a single intra-follicular injection of 0.9 mg PGF2␣ caused a rapid non-physiological drop in plasma P levels [16]. The above results by Ginther et al. [15] in the cow generally support our past and present studies in sheep that repetitive administration of a finite number of PGF2␣ pulses is required to mimic the rate of decline in plasma P seen during natural luteolysis. During sequential 1 h-long infusions of PGF2␣ at 20 g/min/h, we also measured OT in peripheral plasma. The concentration of OT increased to 250 pg/ml during the first systemic pulse of PGF2␣ but declined progressively by approximately 50% during each additional systemic pulse until it was no longer measurable during the fifth pulse (Fig. 7). These plasma levels of OT and their rate of decline during sequential pulses of PGF2␣ , are essentially identical to the peripheral plasma levels of OT and their rate of decline seen during the progression of natural luteolysis in sheep [1,11], thus further supporting the physiological nature of the model of luteolysis reported in the present study. It was noted that P levels recovered in all sheep treated with one pulse of PGF2␣ and also with two pulses of PGF2␣ , while after receiving three pulses of PGF2␣ , two of three sheep also showed recovery of P levels. Even after four pulses of PGF2␣ , one sheep in three, showed recovery of P levels. Thus the CL of these treated animals shows a remarkable ability to recover even after being subjected to up to four pulses of PGF2␣ . Indeed, in a previous report, we found that a rapid decline in protein levels of tissue inhibitors of metalloproteinases-1 and -2 occurred in CLs harvested at 1 h after beginning a single 1 h-long systemic infusion of PGF2␣ given at mid cycle in sheep. This was followed by a drop in plasma P levels at 8 h and a parallel rise in the concentration of protein levels of metalloproteinase-2 in CLs harvested at 8 h [17]. Most of these observed changes recovered by 16 h post-infusion confirming the ability of the CL to recover after the first pulse of PGF2␣ and indeed the same is largely true with P levels even after two pulses of PGF2␣ (Fig. 2). This initial resilience of the CL may be important during the recognition of pregnancy when the elongating blastocyst begins to secrete interferon-tau (INFT) in an effort to inhibit uterine luteolytic pulses of PGF2␣ [18]. Thus, if sufficient INFT is produced by the blastocyst before the third pulse of PGF2␣ is secreted from the uterus, it is very likely that, under these circumstances, the CL will survive and maintain the pregnancy. Also, the longer interval between the first and second endogenous pulses of PGF2␣ may provide an additional time advantage for survival of the blastocyst. The resilience of the CL to the early pulses of PGF2␣ in sheep is in agreement with a recent study in cows where it was observed that a transient rebound of P levels was observed only after the first few endogenous pulses of PGF2␣ [19]. Moreover, in an earlier study [4], it was observed that infusing an effective luteolytic dose of PGF2␣ for 1 h into the arterial supply of the transplanted ovary of sheep once daily for four days did not cause luteolysis. The ovarian secretion rate of P returned to control levels following each daily 1 h-long pulse of PGF2␣ , once again demonstrating the ability of the CL to recover from potentially luteolytic pulses of PGF2␣ . Our observation that only a subset of sheep receiving 3 pulses (1/3) or 4 pulses (2/3) of PGF2␣ showed CL regression and early onset of estrus suggests that some animals are more sensitive to sequential pulses of PGF2␣ than others. The reason for this variation in responsiveness is presently unclear, but it may be related to small variations in their natural cycle lengths that could increase or decrease their sensitivity to a finite number of PGF2␣ pulses [3]. It is
95
also possible that potential differences in pulmonary metabolism of PGF2␣ among individual sheep could explain the variation in responsiveness. This raises the question as to why the CL has to be subjected to a number of sequential pulses of physiological levels of PGF2␣ in order to cause permanent luteolysis which, as reported in the present study, can range from three to five sequential pulses of PGF2␣ . Fig. 6, which includes only animals that showed an early onset of luteolysis and estrus, illustrates that with increasing numbers of PGF2␣ pulses, there is a progressively earlier onset of estrus which is most likely due to the faster rate of decline in plasma P as the number of pulses are increased. Thus, it seems likely that the optimal number of five pulses insures that the onset of estrus follows at an appropriate time interval after the initial onset of luteolysis which in turn may determine that ovulation and fertilization occurs at an optimal time. Moreover, it is likely that the additional pulses of PGF2␣ are needed to increase potential intracellular mediators of PGF2␣ action to a critical level, such as the induction of luteal cyclooxygenase-2 [20], thus insuring that changes induced in the cellular and structural integrity of the CL become irreversible and in an appropriate time frame. The finding in the present report that reducing each sequential pulse of PGF2␣ from a duration of 1 h to a duration of 30 min fails to induce luteolysis and an early onset of estrus indicates that, not only are the number, amplitude, and frequency of PGF2␣ pulses necessary for luteolysis, but the absolute duration of each pulse of PGF2␣ is also critical for inducing permanent irreversible regression of the CL. This observation emphasizes that a pulse of PGF2␣ of 1 h duration is needed for each sequential pulse of PGF2␣ to exert its cumulative biochemical effect within luteal tissue in vivo thus causing complete functional and structural luteolysis. Why pulses of PGF2␣ must be approximately of 1 h duration to cause luteolysis is presently unclear. However, the observation suggests that for certain molecular mechanisms to become activated within luteal tissue, such as expression of the cyclooxygenase-2 gene, may require repetitive PGF2␣ signaling of 1 h duration. 5. Conclusions The finding that five systemic pulses of PGF2␣ given to sheep at intervals matching their endogenous frequency causes the early onset of CL regression and estrus in all treated animals, provides an important new model that permits the measurement of dynamic changes in mediators of functional and structural luteolysis occurring within luteal tissue in vivo. Thus, CLs can be harvested conveniently at selected times during a simulated physiological progression of luteolysis in sheep. The five 1 h-long systemic pulse model could also be useful to investigate the reported resistance of the ovine CL of early pregnancy to exogenous PGF2␣ [21–23]. Comparing P levels and molecular changes in CLs harvested at sequential time points during systemic pulses of PGF2␣ in non-pregnant and early pregnant sheep may assist in identifying putative intraluteal factors responsible for the observed resistance of the CL to PGF2␣ in early pregnancy. Acknowledgments We thank the Staff of the University of Connecticut Sheep Unit for their excellent care of the animals used in this project. We also wish to thank Dr. Tom Hoagland for advice regarding statistics and Mr. George Gagnon for help in collecting blood samples. We also thank Dr. Anup Kollanoor Johny for assistance with graphics. Funding for this study was provided by a NRI competitive USDA grant #2004-35203-14176 from the Cooperative State Research, Education and Extension Service awarded to JAMcC and in part by USDA grant #2008-35203-19101 awarded to JAA.
96
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
References [1] McCracken JA, Custer EE, Lamsa JC. Luteolysis: a neuroendocrine-mediated event. Physiol Rev 1999;79:263–323. [2] McCracken JA. Hormone receptor control of prostaglandin F2 alpha secretion by the ovine uterus. Adv Prostaglandin Thromboxane Res 1980;8:1329–44. [3] Zarco L, Stabenfeldt GH, Quirke JF, Kindahl H, Bradford GE. Release of prostaglandin F2 alpha and the timing of events associated with luteolysis in ewes with oestrous cycles of different lengths. J Reprod Fertil 1988;83:517–26. [4] Schramm W, Bovaird L, Glew ME, Schramm G, McCracken JA. Corpus luteum regression induced by ultra-low pulses of prostaglandin F2 alpha. Prostaglandins 1983;26:347–64. [5] McCracken JA, Carlson JC, Glew ME, et al. Prostaglandin F2 alpha identified as a luteolytic hormone in sheep. Nat N Biol 1972;238:129–34. [6] Davis AJ, Fleet IR, Harrison FA, Maule-Walker FM. Pulmonary metabolism of prostaglandin F2 alpha in the conscious non-pregnant ewe and sow. J Physiol (Lond) 1980;301:86P. [7] Lee J, McCracken JA, Banu SK, Rodriguez R, Nithy TK, Arosh JA. Transport of prostaglandin F2 alpha pulses from the uterus to the ovary at the time of luteolysis in ruminants is regulated by prostaglandin transporter-mediated mechanisms. Endocrinology 2010;151:3326–35. [8] Greiss Jr FC, Anderson SG. Uterine vascular changes during the ovarian cycle. Am J Obstet Gynecol 1969;103:629–40. [9] Ferreira SH, Vane JR. Prostaglandins: their disappearance from and release into the circulation. Nature 1967;216:868–73. [10] Piper PJ, Vane JR, Wyllie JH. Inactivation of prostaglandins by the lungs. Nature 1970;225:600–4. [11] McCracken JA, Custer EE, Lamsa JC, Robinson AG. The central oxytocin pulse generator: a pacemaker for luteolysis. Adv Exp Med Biol 1995;395:133–54. [12] Anderson LE, Schultz MK, Wiltbank MC. Prostaglandin moities that determine receptor binding in the bovine corpus luteum. J Reprod Fert 1999;116:133–41. [13] Custer EE, Lamsa JC, Eldering JA, McCracken JA. Identification of functional high and low affinity states of the prostaglandin F2 alpha receptor in the ovine
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
corpus luteum in vivo and their role in hormone pulsatility. Endocrine 1995;3: 761–4. McCracken JA, Schramm WS, Okulicz WC. Hormone receptor control of pulsatile secretion of PGF2 alpha from the ovine uterus during luteolysis and its abrogation in early pregnancy. Anim Reprod Sci 1984;7:31–55. Ginther OJ, Araujo RR, Palhao MP, Rodrigues BL, Beg MA. Necessity of sequential pulses of prostaglandin F2 alpha for complete physiologic luteolysis in cattle. Biol Reprod 2009;80:641–8. Fogwell RL, Weems CW, Lewis GS, Butcher RL, Inskeep EK. Secretion of steroids after induced luteal regression in beef heifers: effects of PGF2 alpha and removal of corpora lutea. J Anim Sci 1978;46:1718–23. Towle TA, Tsang PC, Milvae RA, Newbury MK, McCracken JA. Dynamic in vivo changes in tissue inhibitors of metalloproteinases-1 and -2, and matrix metalloproteinases-2 and -9, during prostaglandin F2 alpha-induced luteolysis in sheep. Biol Reprod 2002;66:1515–21. Spencer TE, Burghardt RC, Johnston BA, Bazer FW. Conceptus signals for establishment and maintenance of pregnancy. Anim Reprod Sci 2004;82/83: 537–50. Ginther OJ, Shrestha HK, Fuenzalida MJ, Shahiduzzaman AK, Beg MA. Characteristics of pulses of 13,14-dihydro-15-keto-prostaglandin F2 alpha before, during, and after spontaneous luteolysis and temporal intrapulse relationships with progesterone concentrations in cattle. Biol Reprod 2010;82:1049–56. Tsai SJ, Wiltbank M.C. Prostaglandin F2 alpha induces expression of prostaglandin G/H synthase-2 in the ovine corpus luteum: a potential positive feedback loop during luteolysis. Biol Reprod 1997;57:1016–22. Inskeep EK, Smutny WJ, Butcher RL, Pexton JE. Effects of intrafollicular injections of prostaglandins in non-pregnant and pregnant ewes. J Anim Sci 1975;41:1098–104. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of early pregnancy in the ewe. IV. Changes in luteal sensitivity to prostaglandin F2 alpha throughout early pregnancy. J Anim Sci 1986;63:1201–7. Weems CW, Weems YS, Randel RD. Prostaglandins and reproduction in female farm animals. Vet J 2006;171:206–28.
Contents lists available at SciVerse ScienceDirect
Prostaglandins and Other Lipid Mediators
A new in vivo model for luteolysis using systemic pulsatile infusions of PGF2␣ J.A. McCracken a,∗,1 , E.E. Custer b,1 , D.T. Schreiber a , P.C.W. Tsang c , C.S. Keator a,2 , J.A. Arosh d a
Department of Animal Science, University of Connecticut, United States Department of Microbiology and Physiological Systems, University of Massachusetts Medical School, United States c Department of Molecular, Cellular, and Biomedical Sciences, University of New Hampshire, United States d Department of Veterinary Integrative Biosciences, Texas A&M University, United States b
a r t i c l e
i n f o
Article history: Received 16 September 2011 Received in revised form 6 December 2011 Accepted 12 January 2012 Available online 25 January 2012 Keywords: Luteolysis Corpus luteum Progesterone Oxytocin PGF2␣
a b s t r a c t A new in vivo model for studying luteolysis was developed in sheep to provide a convenient method for collecting corpora lutea for molecular, biochemical, and histological analysis during a procedure that mimics natural luteolysis. It was found that the infusion of prostaglandin F2␣ (PGF2␣ ) at 20 g/min/h into the systemic circulation during the mid luteal phase of the cycle allowed sufficient PGF2␣ to escape across the lungs and thus mimic the transient 40% decline in the concentration of progesterone in peripheral plasma seen at the onset of natural luteolysis in sheep. Additional 1 h-long systemic infusions of PGF2␣ , given at physiological intervals, indicated that two infusions were not sufficient to induce luteolysis. However, an early onset of luteolysis and estrus was induced in one out of three sheep with three infusions, two out of three sheep with four infusions, and three out of three sheep with five infusions. Reducing the duration of each systemic infusion of PGF2␣ from 1 h to 30 min failed to induce luteolysis and estrus even after six systemic infusions indicating that, not only are the amplitude and frequency of PGF2␣ pulses essential for luteolysis, but the actual duration of each pulse is also critical. We conclude that a minimum of five systemic pulses of PGF2␣ , given in an appropriate amount and at a physiological frequency and duration, are required to mimic luteolysis consistently in all sheep. The five pulse regimen thus provides a new accurate in vivo model for studying molecular mechanisms of luteolysis. © 2012 Elsevier Inc. All rights reserved.
1. Introduction It is well established that the pulsatile release of PGF2␣ from the uterus induces luteolysis in ruminants as well as several other species of domestic animals [1]. In sheep, the pulsatile release of oxytocin (OT) from the posterior pituitary and the corpus luteum (CL) beginning on days 13–14 of the estrous cycle acts via receptors for OT in the luminal cells of the endometrium to induce luteolytic pulses of PGF2␣ [2]. Approximately 5–8 pulses of PGF2␣ , measured as its primary metabolite 13:14-dihydro-15-keto-PGF2␣ (PGFM), are secreted by the uterus during luteolysis in sheep [3]. It was found that a minimum of five 1 h-long pulses of PGF2␣ infused into the arterial supply of the ovary were required to cause luteolysis consistently in all treated sheep [4]. During natural luteolysis in sheep, approximately 1% of uterine PGF2␣ secreted into the uterine
∗ Corresponding author at: Department of Animal Science, White Building, Unit4040, Storrs, CT 06269-4040, United States. Tel.: +1 860 486 6197; fax: +1 860 486 4375. E-mail address: [email protected] (J.A. McCracken). 1 These authors contributed equally to this paper. 2 Present address: Department of Physiology, Ross University School of Medicine, P.O. Box 266, Portsmouth Campus, Picard, Dominica. 1098-8823/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.prostaglandins.2012.01.004
vein is transferred to the adjacent ovarian artery via a countercurrent transfer system in the utero-ovarian vascular plexus [5]. Local transport of PGF2␣ from the uterus to the ovary is required in sheep because of its high clearance rate in the pulmonary circulation [6]. The underlying mechanism of the countercurrent transfer of PGF2␣ in the utero-ovarian vascular plexus of the sheep was shown recently to depend on the expression of the prostaglandin transporter protein in the vasculature of the utero-ovarian plexus [7]. In the latter study, it was found that PGF2␣ levels in ovarian arterial plasma persisted for a duration of 1 h following an OT-stimulated pulse of uterine PGF2␣ suggesting that transfer of PGF2␣ from the utero-ovarian vein to the ovarian artery for a period of 1 h is most likely a prerequisite for successful luteolysis in sheep. In the present study, our goal was to establish a model for luteolysis that mimicked natural CL regression by infusing 1 h-long systemic infusions of PGF2␣ at a physiological frequency in amounts high enough to allow sufficient PGF2␣ to escape across the lungs into the arterial circulation and thus cause luteolysis. Such a model would provide a convenient means of harvesting luteal tissue for biochemical, molecular, and histological analysis during a process that simulates the natural progression of CL regression in sheep. Luteal tissue collected using this model would thus provide a means of determining protein levels of a variety of biochemical and molecular mediators implicated in the natural progression of functional
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
and structural luteolysis in sheep. To establish the amount of PGF2␣ required to be infused into the systemic circulation to simulate an endogenous luteolytic pulse of PGF2␣ , it was calculated that, based on GC/Mass Spectrometry measurements of 40 ng/ml of PGF2␣ in uterine venous blood during luteolysis [5], and a uterine blood flow from each uterine horn of 10 ml/min at the time of luteolysis [8], the whole uterus secretes approximately 1.0 g/min of PGF2␣ into the venous circulation during each endogenous 1 h-long pulse of PGF2␣ . However, most of the PGF2␣ secreted by the uterus is metabolized by the lungs [9,10] thus reducing PGF2␣ levels in systemic arterial plasma to levels insufficient to cause regression of the CL. Initially we tested the luteolytic activity of 10 times the endogenous uterine secretion rate of PGF2␣ of 1.0 g/min by infusing PGF2␣ at 10 g/min for 1 h into a jugular vein of sheep during the mid luteal phase of the cycle. This rate of infusion reduced the level of progesterone (P) in peripheral plasma by approximately 20% which did not equate with the 40% reduction of P in peripheral plasma seen during the first endogenous pulse of PGF2␣ in sheep [3]. We therefore increased the infusion rate of PGF2␣ to 20 times its endogenous uterine secretion rate by infusing PGF2␣ systemically at 20 g/min for 1 h during the mid luteal phase of the cycle. Such an infusion rate of 20 g/min/h induced a 40% drop in P levels identical to that seen during the first endogenous pulse of PGF2␣ . The objectives of the present study therefore were to: 1. Determine the minimum number of such effective systemic pulses of PGF2␣ , given at their endogenous frequency, required to cause luteolysis and the onset of estrus consistently in all treated sheep thus providing a convenient new model for harvesting luteal tissue during a process mimicking natural luteolysis. 2. Determine whether the actual 1 h duration of each pulse of PGF2␣ was also critical for the induction of luteolysis.
2. Materials and methods 2.1. Animals All experimental procedures in sheep were approved by the Institutional Animal Care and Use Committee at the University of Connecticut. Intact cross-bred Suffolk ewes were used in the present study during the fall and winter breeding season. Sheep were housed in small groups in outside sheep pens, fed twice daily, and given water ad libitum. A total of 30 sheep were used to determine the luteolytic effect of systemic pulses of PGF2␣ . Prior to use, all sheep exhibited at least two estrous cycles of 16–18 days (estrus = day 0), as determined by a vasectomized ram. To perform infusion experiments, sheep were brought into a metabolism unit and housed in individual metabolism cages. At least one companion animal was always present in the metabolism room during each experiment. Infusions were performed during the mid luteal phase of the cycle (day 10). At the end of each infusion experiment, sheep were returned to an outside pen and checked for estrous behavior using a vasectomized ram.
2.2. Infusion procedures After placing each sheep in a metabolism cage, a two and a half inch 16-gauge Teflon indwelling cannula (Beckton Dickinson, East Rutherford, NJ) was placed in both left and right jugular veins and secured with three sutures of #1 braided Dacron. A three way stopcock was inserted into the hub of each cannula and flushed with 2 ml heparin/saline (100 iu/ml). The left jugular cannula was used to withdraw blood samples (5 ml) during each experiment. Infusions were carried out using a Harvard Infusion Pump (Harvard Apparatus, Model #600-000) via a Teflon catheter connected to the cannula in the right jugular vein.
91
One hour-long infusions of sterile pyrogen free saline (Abbott Labs, Chicago), PGF2␣ (Pfizer, NY) or PGFM (Cayman Chemicals, MI) were administered systemically into the right jugular cannula. PGF2␣ was infused systemically at a rate of 10 g/min/h and/or at 20 g/min/h. To insure that the primary metabolite of PGF2␣ , 13:14-dihydro-15-keto-PGF2␣ (PGFM) did not contribute to the luteolytic effect of administered PGF2␣ , PGFM was infused at a rate of 20 g/min/h. Control 1 h-long infusions of saline vehicle, and PGF2␣ and PGFM in saline vehicle were given at the standard infusion rate of 0.1 ml/min/h (6 ml/h). A sample size of three sheep (n = 3) was used for all substances and doses used in the present study. The interval between sequential pulsed infusions of PGF2␣ was based on the reported intervals between endogenous pulses of PGFM in cycling sheep [3] which were 15 h between the first and second pulses and 7 h between subsequent pulses. To establish the importance of the duration of each systemic infusion of PGF2␣ , an additional three sheep during mid-luteal phase (day 10) were given infusions of PGF2␣ at 20 g/min for 30 min instead of 1 h at interpulse intervals of 7 h into the right jugular. Because our earlier results with 1 h-long infusions clearly indicated that PGF2␣ was most effective if given before there was any recovery of P levels (see Sections 3 and 4), the initial inter-pulse interval of 15 h was excluded in the 30 min infusions of PGF2␣ . 2.3. Blood sample collections All jugular blood samples were collected using syringes and tubes stored over ice. Immediately after collection, blood samples were centrifuged at 2000 rpm for 20 min at 4 ◦ C. The plasma was removed promptly, taking care not to include the buffy coat, and stored at −20 ◦ C. The start of each infusion experiment was designated as time zero. Control jugular blood samples were collected at −20 and −10 min. Additional blood samples were collected at 10, 20, 30, and 60 min during each systemic infusion, hourly between each infusion, and hourly for 24 h after the last infusion. Blood samples were then collected twice daily for 96 h or until estrus was observed. After each infusion of substances into the right jugular vein and after each collection of blood from the left jugular vein, cannulae were flushed with 2 ml heparin/saline (100 iu/ml). 2.4. Hormone assays The concentration of P in peripheral plasma was measured using a commercially available assay with antibody coated tubes and 125 Ilabeled P (Coat-A-Count, Diagnostic Productions Corporation; Los Angeles, CA). The concentration of OT in peripheral plasma was measured using a commercially available ELISA (Assay Designs; Ann Arbor, MI). We measured OT in peripheral plasma samples during the infusion of 1 h-long systemic infusions of PGF2␣ at 20 g/min to determine if the release of OT paralleled the pattern of release of OT observed during natural luteolysis [1,11]. The intraand inter-assay coefficients of variation for P measurement were 2.8% and 6.5% respectively. The intra- and inter-assay coefficients of variation for OT measurement were less than 5%. 2.5. Statistical analysis Plasma levels of P were analyzed by one-way ANOVA. Differences between means were determined by Tukey’s HSD test. Chi Square analysis was used to determine the relationship between the number of systemic pulses of PGF2␣ administered at mid cycle and the time of onset of estrus. A value of p < 0.05 was considered significant.
92
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
160
140
E
140 Jugular Plasma Progesterone ( % of Controls)
Jugular Plasma Progesterone (% of Controls)
120
100
80
60
40
20
120
100
80
60
40
20
0
0 0
2
4
6
8
10
12
14
16
0
24
48
72
96
120
144
168
Time In Hours
Time In Hours Fig. 1. The effect on peripheral plasma P (% of controls) of a single 1 h-long systemic infusion of PGF2␣ (20 g/min/h) given to sheep (n = 15) on day 10 of the cycle. This number of sheep also includes the data obtained after the first pulse of PGF2␣ (20 g/min/h) in all subsequent experiments. P fell by 40% (p < 0.05), reaching a nadir at 8 h and returning to control values by 16 h, identical to the changes in P levels seen after the first endogenous pulse of PGF2␣ in this species. Arrows indicate the infusion of PGF2␣ and asterisks indicate a significant difference in P levels from controls.
3. Results 3.1. Systemic administration of 1 h-long infusion of PGF2˛ , PGFM, or saline A single 1 h-long systemic infusion of PGF2␣ (n = 3 sheep) given into a jugular vein during the mid luteal phase (cycle day 10) at 10 g/min caused only a 20% drop in P levels (p < 0.05) reaching a nadir at 6 h and recovering to control levels by 12 h. However, since the fall in P did not match the reduction in P levels observed during the first endogenous pulse of PGF2␣ in sheep [3], the rate of PGF2␣ infusion was increased to 20 g/min for 1 h (n = 15 sheep). This number of sheep includes the data obtained after the first pulse of PGF2␣ (20 g/min/h) in all subsequent experiments. As shown in Fig. 1, there was a characteristic transient increase in P levels followed by a 40% decline (p < 0.05), reaching a nadir at 8 h and recovering by 16 h post-infusion, identical to the changes in P levels seen after the first endogenous pulse of PGF2␣ during natural luteolysis in this species [3]. Following recovery of P after a single 1-h pulse of PGF2␣ , there was a normal rise in P levels from days 10 to 13 as seen in natural cycles, followed by CL regression and estrus at the normal expected time. Because of the possibility that some of the luteolytic effect of a pulse of PGF2␣ infused at 20 g/min, might be caused by conversion into its metabolite PGFM, a 1 h-long systemic infusion of PGFM was given to three sheep at the same rate as PGF2␣ (20 g/min/h). This infusion rate of PGFM had no effect (p > 0.05) on P levels either during or for 24 h after the infusion. The systemic infusion of saline vehicle at 0.1 ml/min/h (n = 3 sheep) also
Fig. 2. The effect on peripheral plasma P (% of controls) of two 1 h-long systemic infusions of PGF2␣ (20 g/min/h) given to sheep (n = 3) on day 10 of the cycle at an inter-pulse interval of 15 h. P levels had recovered by 15 h but the second infusion of PGF2␣ caused a greater fall in P that never really fully recovered to controls. Premature regression did not occur with two systemic infusions of PGF2␣ and all the animals had normal cycle lengths of 16–17 days. E indicates the mean time of estrus.
did not reduce P levels (p > 0.05) either during or for 24 h after the infusion. 3.2. Systemic administration of multiple 1 h-long infusions of PGF2˛ A second 1 h-long systemic infusion of PGF2␣ given at 20 g/min/h after an inter-pulse interval of 15 h (n = 3 sheep), caused a second temporary but greater decline in P which lasted longer and did not completely recover to control values (see Fig. 2). Premature CL regression did not occur with two systemic infusions of PGF2␣ and estrus occurred at the expected time, even though P levels did not recover fully to normal mid luteal levels. A third 1 h-long systemic infusion of PGF2␣ given at the same rate as the first two pulses (20 g/min/h) after an inter-pulse interval of 7 h (n = 3 sheep), caused complete regression with the early onset of estrus in one sheep (Fig. 3A). In the remaining two sheep, P levels partially recovered after the third pulse and the animals did not exhibit an early onset of luteolysis and estrus (Fig. 3B). A forth 1 h-long systemic infusion of PGF2␣ given at 20 g/min/h after an inter-pulse interval of 7 h (n = 3 sheep), caused premature CL regression with the early onset of estrus in two sheep (Fig. 4A). In the remaining sheep, P levels showed an initial decline which partially recovered after the forth pulse but both CL regression and estrus occurred at the normal time (Fig. 4B). Lastly, as shown in Fig. 5, a fifth systemic infusion of PGF2␣ (20 g/min/h) given after a further inter-pulse interval of 7 h caused premature regression of the CL and the early onset of luteolysis and estrus in all treated sheep. Thus, as the number of PGF2␣ pulses increased, a greater proportion of sheep (p < 0.05) showed an earlier onset of luteolysis and estrus. This is illustrated in Fig. 6 which shows the rate of decline in plasma P and the relative time of estrus
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
160
160
(A)
E
140
Jugular Plasma Progesterone (% of Controls)
Jugular Plasma Progesterone (% of Controls)
100 80 60 40 20 0
E (B)
140 120 100 80 60
120
100
80
60
40
20
40
0 0
20 20
40
60
80
100
120
140
160
160 E
140
(A)
120 100
Jugular Plasma Progesterone (% of Controls)
40
60
80
100
120
140
180
80 60 40 20 0 E
140
(B)
Fig. 5. The effect on peripheral plasma P (% of controls) of five 1 h-long systemic infusions of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h, 7 h, 7 h, and 7 h given to sheep (n = 3) beginning in the mid luteal phase of the cycle (day 10). In all sheep, plasma P levels fell to <1.0 ng/ml and the sheep were mated by a vasectomized ram between 68 and 72 h after beginning the infusions of PGF2␣ . E indicates the mean time of estrus.
Jugular Plasma Progesterone (% of controls)
0
Fig. 3. The effect on peripheral plasma P (% of controls) of three 1 h-long systemic infusions of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h and 7 h given to sheep (n = 3) during the mid luteal phase of the cycle (day 10). One out of three sheep (A) showed a premature decline in P levels and an early onset of estrus. P levels recovered in the two remaining sheep (B) which did not exhibit premature CL regression or an early onset of estrus. E indicates the mean time of estrus.
160 E
140
E
E
120 100 80 60 40 20 0
0
20
40
60
80
100
120
140
Time In Hours Fig. 6. Comparison of the rate of decline of P levels in sheep that showed luteolysis and an early onset of estrus following three (blue triangles), four (black triangles), or five (red triangles) pulses of PGF2␣ . The non-responders in the 3 and 4 pulse groups are not included in the data. E indicates the average time of estrus following three (blue box), four (black box), or five (red box) pulses of PGF2␣ . It can be seen that the more rapid the decline in P levels, the earlier the animals exhibited estrus. Thus five pulses of PGF2␣ pulses ensures that estrus follows at an appropriate time after the initial onset of luteolysis which may be important for the timing of ovulation and hence for fertility rates in sheep. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.).
120 100 80 60 40 20 0
20
Time In Hours Time In Hours
0
E
140
120
0
93
0
20
40
60
80
100
120
140
160
180
Time In Hours Fig. 4. The effect on peripheral plasma P (% of controls) of four 1 h-long systemic infusion of PGF2␣ (20 g/min/h) at inter-pulse intervals of 15 h, 7 h, and 7 h given to sheep (n = 3) on day 10 of the cycle. Two out of three sheep (A) showed a premature decline in P levels and an early onset of estrus. P levels recovered in the remaining sheep (B) which did not exhibit premature CL regression or an early onset of estrus. E indicates the mean time of estrus.
in sheep that showed premature luteolysis and estrus in groups receiving 3, 4, or 5 pulses of PGF2␣ . The animals receiving 3 or 4 pulses of PGF2␣ that did not exhibit an early onset of luteolysis and estrus are not included in the data shown in Fig. 6. As shown in Fig. 7, the concentration of OT in peripheral plasma increased during the first pulse of PGF2␣ (20 g/min/h) but declined progressively during each additional systemic pulse of PGF2␣ until it was no longer measurable during the fifth pulse of
94
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
200 300
Jugular Plasma Progesterone (% of Controls)
270
Jugular Plasma Oxytocin (pg/mL)
240 210 180 150 120 90 60 30 (15)
(12)
(9)
(6)
6
12
18
24
30
36
100
50
(3)
0 0
150
42
Time In Hours Fig. 7. The concentration of OT in peripheral plasma during successive systemic infusions of PGF2␣ (20 g/min/h) beginning on cycle day 10. The numbers in parentheses indicate the number of sheep receiving a given number of systemic pulses of PGF2␣ . OT declined progressively until it was no longer measurable during the fifth pulse of PGF2␣ (<15 pg/ml).
PGF2␣ (<15 pg/ml). This result is in agreement with the progressive decline in the levels of OT observed in peripheral plasma during natural luteolysis in sheep [1,11]. To investigate the importance of the duration of each pulse of PGF2␣ in causing luteolysis, PGF2␣ was infused at its effective dose of 20 g/min at inter-pulse intervals of 7 h but the duration of each infusion was reduced to 30 min instead of 1 h. As shown in Fig. 8, P levels showed a characteristic rise during most of the 30 min infusions of PGF2␣ , but the mean level of P at the end of six 30 min-long infusions of PGF2␣ was not significantly different from pre-infusion controls (p > 0.05). Premature luteolysis did not occur with this infusion regimen and all three sheep showed estrus at the normal expected time of 16–17 days. 4. Discussion The first objective of the present study was to confirm the amount of PGF2␣ infused for 1 h into the systemic circulation that would mimic the initial decline in levels of P reported to a occur at the onset of luteolysis during the natural cycle of sheep [3]. We first tested the luteolytic activity of 10 times the endogenous uterine secretion rate of PGF2␣ by infusing PGF2␣ at 10 g/min for 1 h into a jugular vein of three sheep during the mid luteal phase. This rate of infusion reduced the level of P by approximately 20% which did not equate with the 40% reduction of P levels seen during the first endogenous pulse of PGF2␣ [3]. When we increased the infusion rate of PGF2␣ to 20 g/min for 1 h during the mid luteal phase (n = 3 sheep), P levels were reduced by 40% which matched the fall observed after the first endogenous pulse of PGF2␣ during natural luteolysis in this species [3]. The question arose as to whether the amount of PGFM (the primary metabolite of PGF2␣ ) generated from the infused PGF2␣ might have some effect on P secretion by the CL. Accordingly, PGFM was infused at 20 g/min
0 0
8
16
24
32
40
48
Time In Hours Fig. 8. The effect on peripheral plasma P (% of controls) after six 30 min-long systemic infusion of PGF2␣ (20 g/min/30 min) at inter-pulse intervals of 7 h given to sheep (n = 3) on day 10 of the cycle. Although P levels showed a characteristic rise during most of the 30 min infusions of PGF2␣ , the mean level of P at the end of the infusions was not significantly different from controls (p > 0.05). All three sheep showed estrus at the normal expected time of 16–17 days.
for 1 h into the systemic circulation of three sheep during the mid luteal phase. The infusion of PGFM had no effect on P levels during or after the infusion (p > 0.05). This finding was not unexpected since PGFM has a low binding affinity for the PGF2␣ receptor [12]. Moreover, the observed lack of response to PGFM infusions is consistent with our previous finding that the infusion of PGFM directly into the arterial supply of the ovarian transplant model in sheep, given at the same rate as an effective luteolytic infusion of PGF2␣ , had no effect on the secretion rate of either P or OT [13]. Having established an effective systemic infusion rate of PGF2␣ that would cause a physiological decline in P, we then proceeded to determine the effect of additional systemic infusions of PGF2␣ given at time intervals that matched their physiological frequency [3]. When a second infusion of PGF2␣ was given after an inter-pulse interval of 15 h, P levels declined to a greater extent than after the first pulse, and then partially recovered again in all sheep and the onset of luteolysis and estrus occurred at the normal time of 16–17 days. After a third pulse given at an inter-pulse interval of 7 h, one sheep showed an early onset of luteolysis and estrus, while P levels recovered in two out of three treated sheep, both of which showed luteolysis and estrus at the normal time. After a fourth pulse of PGF2␣ , P levels recovered in one sheep while the other two sheep showed an early onset of luteolysis and estrus. Finally, the infusion of five 1 h-long systemic infusions of PGF2␣ given during the mid luteal phase (day 10) caused the early onset of luteolysis and estrus in all treated sheep. The finding that five 1 h-long infusions of PGF2␣ were required to cause luteolysis consistently in all animals is in agreement with our earlier studies in sheep where it was determined that a minimum of five episodic 1 h-long pulses of low levels of PGF2␣ infused directly into the arterial supply of the transplanted ovary were required to cause luteolysis in all treated
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
sheep [4,14]. In these studies, the total amount of PGF2␣ infused as pulses was 1/40th of the amount required to cause luteolysis when PGF2␣ was given as a constant infusion, suggesting that a pulsatile regimen of PGF2␣ is advantageous for luteolysis in sheep and most likely in other species. In a recent study in cows [15], four sequential intrauterine injections of PGF2␣ (0.25 mg or 1.0 mg every 12 h) resulted in a continuous steady decline in P levels with shortening of the estrous cycle, whereas a single intra-follicular injection of 0.9 mg PGF2␣ caused a rapid non-physiological drop in plasma P levels [16]. The above results by Ginther et al. [15] in the cow generally support our past and present studies in sheep that repetitive administration of a finite number of PGF2␣ pulses is required to mimic the rate of decline in plasma P seen during natural luteolysis. During sequential 1 h-long infusions of PGF2␣ at 20 g/min/h, we also measured OT in peripheral plasma. The concentration of OT increased to 250 pg/ml during the first systemic pulse of PGF2␣ but declined progressively by approximately 50% during each additional systemic pulse until it was no longer measurable during the fifth pulse (Fig. 7). These plasma levels of OT and their rate of decline during sequential pulses of PGF2␣ , are essentially identical to the peripheral plasma levels of OT and their rate of decline seen during the progression of natural luteolysis in sheep [1,11], thus further supporting the physiological nature of the model of luteolysis reported in the present study. It was noted that P levels recovered in all sheep treated with one pulse of PGF2␣ and also with two pulses of PGF2␣ , while after receiving three pulses of PGF2␣ , two of three sheep also showed recovery of P levels. Even after four pulses of PGF2␣ , one sheep in three, showed recovery of P levels. Thus the CL of these treated animals shows a remarkable ability to recover even after being subjected to up to four pulses of PGF2␣ . Indeed, in a previous report, we found that a rapid decline in protein levels of tissue inhibitors of metalloproteinases-1 and -2 occurred in CLs harvested at 1 h after beginning a single 1 h-long systemic infusion of PGF2␣ given at mid cycle in sheep. This was followed by a drop in plasma P levels at 8 h and a parallel rise in the concentration of protein levels of metalloproteinase-2 in CLs harvested at 8 h [17]. Most of these observed changes recovered by 16 h post-infusion confirming the ability of the CL to recover after the first pulse of PGF2␣ and indeed the same is largely true with P levels even after two pulses of PGF2␣ (Fig. 2). This initial resilience of the CL may be important during the recognition of pregnancy when the elongating blastocyst begins to secrete interferon-tau (INFT) in an effort to inhibit uterine luteolytic pulses of PGF2␣ [18]. Thus, if sufficient INFT is produced by the blastocyst before the third pulse of PGF2␣ is secreted from the uterus, it is very likely that, under these circumstances, the CL will survive and maintain the pregnancy. Also, the longer interval between the first and second endogenous pulses of PGF2␣ may provide an additional time advantage for survival of the blastocyst. The resilience of the CL to the early pulses of PGF2␣ in sheep is in agreement with a recent study in cows where it was observed that a transient rebound of P levels was observed only after the first few endogenous pulses of PGF2␣ [19]. Moreover, in an earlier study [4], it was observed that infusing an effective luteolytic dose of PGF2␣ for 1 h into the arterial supply of the transplanted ovary of sheep once daily for four days did not cause luteolysis. The ovarian secretion rate of P returned to control levels following each daily 1 h-long pulse of PGF2␣ , once again demonstrating the ability of the CL to recover from potentially luteolytic pulses of PGF2␣ . Our observation that only a subset of sheep receiving 3 pulses (1/3) or 4 pulses (2/3) of PGF2␣ showed CL regression and early onset of estrus suggests that some animals are more sensitive to sequential pulses of PGF2␣ than others. The reason for this variation in responsiveness is presently unclear, but it may be related to small variations in their natural cycle lengths that could increase or decrease their sensitivity to a finite number of PGF2␣ pulses [3]. It is
95
also possible that potential differences in pulmonary metabolism of PGF2␣ among individual sheep could explain the variation in responsiveness. This raises the question as to why the CL has to be subjected to a number of sequential pulses of physiological levels of PGF2␣ in order to cause permanent luteolysis which, as reported in the present study, can range from three to five sequential pulses of PGF2␣ . Fig. 6, which includes only animals that showed an early onset of luteolysis and estrus, illustrates that with increasing numbers of PGF2␣ pulses, there is a progressively earlier onset of estrus which is most likely due to the faster rate of decline in plasma P as the number of pulses are increased. Thus, it seems likely that the optimal number of five pulses insures that the onset of estrus follows at an appropriate time interval after the initial onset of luteolysis which in turn may determine that ovulation and fertilization occurs at an optimal time. Moreover, it is likely that the additional pulses of PGF2␣ are needed to increase potential intracellular mediators of PGF2␣ action to a critical level, such as the induction of luteal cyclooxygenase-2 [20], thus insuring that changes induced in the cellular and structural integrity of the CL become irreversible and in an appropriate time frame. The finding in the present report that reducing each sequential pulse of PGF2␣ from a duration of 1 h to a duration of 30 min fails to induce luteolysis and an early onset of estrus indicates that, not only are the number, amplitude, and frequency of PGF2␣ pulses necessary for luteolysis, but the absolute duration of each pulse of PGF2␣ is also critical for inducing permanent irreversible regression of the CL. This observation emphasizes that a pulse of PGF2␣ of 1 h duration is needed for each sequential pulse of PGF2␣ to exert its cumulative biochemical effect within luteal tissue in vivo thus causing complete functional and structural luteolysis. Why pulses of PGF2␣ must be approximately of 1 h duration to cause luteolysis is presently unclear. However, the observation suggests that for certain molecular mechanisms to become activated within luteal tissue, such as expression of the cyclooxygenase-2 gene, may require repetitive PGF2␣ signaling of 1 h duration. 5. Conclusions The finding that five systemic pulses of PGF2␣ given to sheep at intervals matching their endogenous frequency causes the early onset of CL regression and estrus in all treated animals, provides an important new model that permits the measurement of dynamic changes in mediators of functional and structural luteolysis occurring within luteal tissue in vivo. Thus, CLs can be harvested conveniently at selected times during a simulated physiological progression of luteolysis in sheep. The five 1 h-long systemic pulse model could also be useful to investigate the reported resistance of the ovine CL of early pregnancy to exogenous PGF2␣ [21–23]. Comparing P levels and molecular changes in CLs harvested at sequential time points during systemic pulses of PGF2␣ in non-pregnant and early pregnant sheep may assist in identifying putative intraluteal factors responsible for the observed resistance of the CL to PGF2␣ in early pregnancy. Acknowledgments We thank the Staff of the University of Connecticut Sheep Unit for their excellent care of the animals used in this project. We also wish to thank Dr. Tom Hoagland for advice regarding statistics and Mr. George Gagnon for help in collecting blood samples. We also thank Dr. Anup Kollanoor Johny for assistance with graphics. Funding for this study was provided by a NRI competitive USDA grant #2004-35203-14176 from the Cooperative State Research, Education and Extension Service awarded to JAMcC and in part by USDA grant #2008-35203-19101 awarded to JAA.
96
J.A. McCracken et al. / Prostaglandins & other Lipid Mediators 97 (2012) 90–96
References [1] McCracken JA, Custer EE, Lamsa JC. Luteolysis: a neuroendocrine-mediated event. Physiol Rev 1999;79:263–323. [2] McCracken JA. Hormone receptor control of prostaglandin F2 alpha secretion by the ovine uterus. Adv Prostaglandin Thromboxane Res 1980;8:1329–44. [3] Zarco L, Stabenfeldt GH, Quirke JF, Kindahl H, Bradford GE. Release of prostaglandin F2 alpha and the timing of events associated with luteolysis in ewes with oestrous cycles of different lengths. J Reprod Fertil 1988;83:517–26. [4] Schramm W, Bovaird L, Glew ME, Schramm G, McCracken JA. Corpus luteum regression induced by ultra-low pulses of prostaglandin F2 alpha. Prostaglandins 1983;26:347–64. [5] McCracken JA, Carlson JC, Glew ME, et al. Prostaglandin F2 alpha identified as a luteolytic hormone in sheep. Nat N Biol 1972;238:129–34. [6] Davis AJ, Fleet IR, Harrison FA, Maule-Walker FM. Pulmonary metabolism of prostaglandin F2 alpha in the conscious non-pregnant ewe and sow. J Physiol (Lond) 1980;301:86P. [7] Lee J, McCracken JA, Banu SK, Rodriguez R, Nithy TK, Arosh JA. Transport of prostaglandin F2 alpha pulses from the uterus to the ovary at the time of luteolysis in ruminants is regulated by prostaglandin transporter-mediated mechanisms. Endocrinology 2010;151:3326–35. [8] Greiss Jr FC, Anderson SG. Uterine vascular changes during the ovarian cycle. Am J Obstet Gynecol 1969;103:629–40. [9] Ferreira SH, Vane JR. Prostaglandins: their disappearance from and release into the circulation. Nature 1967;216:868–73. [10] Piper PJ, Vane JR, Wyllie JH. Inactivation of prostaglandins by the lungs. Nature 1970;225:600–4. [11] McCracken JA, Custer EE, Lamsa JC, Robinson AG. The central oxytocin pulse generator: a pacemaker for luteolysis. Adv Exp Med Biol 1995;395:133–54. [12] Anderson LE, Schultz MK, Wiltbank MC. Prostaglandin moities that determine receptor binding in the bovine corpus luteum. J Reprod Fert 1999;116:133–41. [13] Custer EE, Lamsa JC, Eldering JA, McCracken JA. Identification of functional high and low affinity states of the prostaglandin F2 alpha receptor in the ovine
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
corpus luteum in vivo and their role in hormone pulsatility. Endocrine 1995;3: 761–4. McCracken JA, Schramm WS, Okulicz WC. Hormone receptor control of pulsatile secretion of PGF2 alpha from the ovine uterus during luteolysis and its abrogation in early pregnancy. Anim Reprod Sci 1984;7:31–55. Ginther OJ, Araujo RR, Palhao MP, Rodrigues BL, Beg MA. Necessity of sequential pulses of prostaglandin F2 alpha for complete physiologic luteolysis in cattle. Biol Reprod 2009;80:641–8. Fogwell RL, Weems CW, Lewis GS, Butcher RL, Inskeep EK. Secretion of steroids after induced luteal regression in beef heifers: effects of PGF2 alpha and removal of corpora lutea. J Anim Sci 1978;46:1718–23. Towle TA, Tsang PC, Milvae RA, Newbury MK, McCracken JA. Dynamic in vivo changes in tissue inhibitors of metalloproteinases-1 and -2, and matrix metalloproteinases-2 and -9, during prostaglandin F2 alpha-induced luteolysis in sheep. Biol Reprod 2002;66:1515–21. Spencer TE, Burghardt RC, Johnston BA, Bazer FW. Conceptus signals for establishment and maintenance of pregnancy. Anim Reprod Sci 2004;82/83: 537–50. Ginther OJ, Shrestha HK, Fuenzalida MJ, Shahiduzzaman AK, Beg MA. Characteristics of pulses of 13,14-dihydro-15-keto-prostaglandin F2 alpha before, during, and after spontaneous luteolysis and temporal intrapulse relationships with progesterone concentrations in cattle. Biol Reprod 2010;82:1049–56. Tsai SJ, Wiltbank M.C. Prostaglandin F2 alpha induces expression of prostaglandin G/H synthase-2 in the ovine corpus luteum: a potential positive feedback loop during luteolysis. Biol Reprod 1997;57:1016–22. Inskeep EK, Smutny WJ, Butcher RL, Pexton JE. Effects of intrafollicular injections of prostaglandins in non-pregnant and pregnant ewes. J Anim Sci 1975;41:1098–104. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of early pregnancy in the ewe. IV. Changes in luteal sensitivity to prostaglandin F2 alpha throughout early pregnancy. J Anim Sci 1986;63:1201–7. Weems CW, Weems YS, Randel RD. Prostaglandins and reproduction in female farm animals. Vet J 2006;171:206–28.