|
|
||||||||
Articles |
Does Not Increase Around the Time of Luteolysis in the Pregnant Goat1
a Department of Physiology, Monash University, Clayton, Victoria 3168, Australia
| ABSTRACT |
|---|
|
|
|---|
secreted from the uterus is the luteolysin of the estrous cycle and is also believed to be responsible for luteolysis in the pregnant doe at term. We have reported that basal progesterone concentrations decrease before basal PGF2
concentrations increase, which is inconsistent with this view. In this study we investigated whether luteolysis is associated with increased frequency or amplitude of pulsatile PGF2
secretion in does over the last 2 wk of gestation.
Progesterone concentrations decreased approximately 1 wk before parturition. There was no accompanying increase in PGF2
concentrations or pulse frequency, and those pulses that were observed were of lesser amplitude and duration than those that have been associated with luteolysis in cycling ewes. A small increase in PGF2
pulse frequency was identified during the 3 days before parturition, but this was not associated with any change in progesterone concentrations. The biological significance of these small changes in PGF2
pulse frequency is obscure, although the high concentration of this eicosanoid at labor may have been related to the final, precipitous decline in plasma progesterone concentrations. These findings do not support the notion that PGF2
is the principal luteolysin in the pregnant doe at term.
| INTRODUCTION |
|---|
|
|
|---|
, from the uterus, since this compound is the luteolysin of the estrous cycle and is secreted into the utero-ovarian vein (UOV) in large quantities during the last 24 h before parturition [37].
We recently reported the profiles of progesterone and prostaglandins in the plasma of pregnant does over the last month of gestation, with the unexpected finding that the first detectable decline in progesterone concentrations preceded any detectable increase in PGF2
concentrations by some 34 days [8]. This finding was not consistent with the proposed role of PGF2
as the luteolysin in pregnant does; however, the conclusions were subject to the caveat that they were based on once-daily samples. Since PGF2
is secreted in a pulsatile fashion and has a very short half-life in the circulation, the possibility therefore remained that pulsatile PGF2
secretion had preceded the first significant decrease in plasma progesterone and had not been detected due to the sampling regimen we used.
The aims of this study were to compare PGF2
concentrations in the UOV and progesterone concentrations in the systemic circulation in the later stages of pregnancy using a frequent-sampling regimen, to determine whether changes in the concentrations of PGF2
occur prior to functional luteolysis in the goat. In addition, we describe a method for frequent blood collection that optimizes the use of both time and materials.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The animal experiment was approved by the Standing Committee on Ethics in Animal Experimentation of Monash University as conforming to the requirements of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. For this study, 6 angora-cross does of known gestational age were used. Surgery was performed between Days 116 and 125 of gestation and involved cannulation of the maternal carotid artery and jugular vein, as well as the placement of uterine electromyographic (EMG) electrodes for the detection of labor. In addition, UOVs from both sides of the uterus were cannulated via minor branches on the surface of the uterus, and the cannula tips were advanced to a point approximately 10 cm downstream from the junction of the ovarian vein. Surgery was performed under halothane anesthesia using intermittent positive pressure ventilation after induction with thiopentone sodium (20 mg/kg). Cannulae were secured with silk ligatures, and the serosal surfaces were approximated with 4/0 chromic gut sutures. The midline laparotomy was repaired with a continuous, locked suture of 2 chromic gut, and the skin was closed with a continuous horizontal mattress suture of 3/0 polypropylene. At least 6 days of postsurgical recovery time was allowed prior to the experimental protocol.
Experimental Protocol
Blood was collected from the maternal carotid artery and both UOVs every 15 min for 8 h on Days 131, 134, 137, 140, 143, and 146 of gestation. After Day 146, EMG activity was assessed several times daily for signs of increased uterine activity, a reliable indicator of imminent labor [9]. In the event that increased uterine electrical activity was detected, blood sampling was resumed until delivery of the first fetus. Prior to and after a day of frequent sampling, 0.5 ml of maternal arterial blood was collected for the assessment of maternal hemoglobin saturation and content (SaO2 and Hb). Plasma for the analysis of progesterone was collected from the carotid artery, and that for the analysis of PGF2
concentrations was collected from the UOV. Plasma samples collected from each frequent-sampling day were processed in a single RIA.
Frequent Blood Sampling Procedure
For the analysis of progesterone, 0.03 ml of heparin (50 IU/ml of 0.9% saline) was added to each 1 ml tuberculin syringe (Terumo, Elkton, MD) before the collection of 1 ml of whole blood. For the analysis of prostaglandins, 0.03 ml of a 1:2 indomethacin:EDTA solution, containing 10 mM indomethacin and 18.8 mM EDTA, was added to each syringe prior to the collection of 1 ml of whole blood. Immediately after blood collection, syringes were capped with a female luer cap and plunged into ice-cold water until centrifuged. Owing to the combined length of the internal plunger and full syringe, the syringe plunger was shortened prior to centrifugation, leaving only the rubber seal inside the syringe. This shorter length allowed for a secure fit of the syringe within the centrifuge tube carrier. Blood retained in the capped syringes was centrifuged in the "cap-up" position at 2000 x g for 5 min at 4°C.
After centrifugation, the syringe cap was removed and replaced with a 27-gauge needle. By partially removing the protective sheath housing the needle and pulling downward, the needle could be bent into a "U" shape while maintaining sterility of the needle tip. This U shape kept the syringe in an upright position during the dispensing of plasma and thus prevented remixing of plasma and erythrocytes. Using the previously clipped internal plunger, plasma was dispensed into sample cups for storage at -20°C until further analysis.
Hormone Assays
Progesterone Concentrations of progesterone in maternal plasma were measured using procedures previously described by Rice et al. [10]. The interassay coefficient of variation (CV) estimated from 13 assays used for this study was 14.7%. The intraassay CV was 10%. The sensitivity of the assay was 2.9 nM. Known amounts of progesterone (0.110.0 pmol) added to maternal plasma showed a significant correlation (r2 = 0.93, p < 0.001, n = 22) between progesterone measured versus that added, with a slope of 1.12 ± 0.07 (SEM).
PGF2
PGF2
was measured using an unextracted assay as previously described by Burgess et al. [11]. The interassay CV, estimated from 17 PGF2
assays, was 11.64%. The intraassay CV was 8.28%. The sensitivity of the assay was 0.486 nM. Known amounts (0.022.0 pmol) of PGF2
added to maternal plasma showed a significant correlation (r2 = 0.99, p < 0.001, n = 25) between PG measured versus that added, with a slope of 1.00 ± 0.02 (SEM).
Pulse Identification
Pulses were identified using the PULSAR analysis program of Merriam and Wachter [12].
In this study, to be considered a significant pulse, the hormone concentration of a single sample had to exceed 4 times the SD of the assay (i.e., G(1) = 4). A point was identified as forming part of a 2-point peak if it belonged to a pair of adjacent points that were both greater than G(2) (i.e., 3.5 times SD). Similarly, G(3) = 2.8, G(4) = 2, and G(5) = 1.
Statistical Analysis
Prior to analysis, progesterone and PGF2
data from each animal were grouped into 3-day blocks; -15 to -13, -12 to -10, -9 to -7, -6 to -4, -3 to -1, and 0 days relative to labor. In cases in which PGF2
samples were collected from both sides of a uterus occupied by two fetuses, data from both UOV cannulae were analyzed as replicates of the same time point for each animal. Five of the six does used were pregnant with twins; the other had a single fetus. Data collected from the UOV draining the gravid horn of the doe bearing a singleton pregnancy were included in all appropriate analyses. Owing to the occlusion of both UOV cannulae in one twin-bearing animal, frequent samples for the analysis of PGF2
were collected from only 5 of the 6 animals. Arterial cannulae from all six animals remained patent for the collection of progesterone and blood gas samples throughout the experimental period.
The interval between sampling days for each animal was 3 days, and thus each animal contributed one data set to each 3-day block. For each gestational age block, the mean concentration, mean pulse frequency (number of pulses/h), and mean amplitude of pulses of progesterone and PGF2
identified during the 8-h sampling day were calculated. Data were tested for homogeneity of variance using Bartlett's Box F and Cochrane's C tests. Data found nonhomogeneous were rendered homogeneous by log10 or square root transformation. Hormone data relative to labor were evaluated by one-factor repeated measures ANOVA with time (gestational age block) the factor tested. Where an effect of time was detected, differences between time points were identified using the test of least significant difference (LSD) with
= 0.05 [13]. Changes in blood gas status over the experimental period were assessed using a two-factor repeated measures ANOVA with blood gas sample (pre- versus post-8-h sampling period) and gestational age relative to labor as factors.
| RESULTS |
|---|
|
|
|---|
Gestational age at delivery Does went into labor at 148.8 ± 1.6 days, which was within the normal range of labor onset.
Maternal blood parameters There was no significant difference between sampling time (pre- and postsampling periods within a given day), animal, or gestational age relative to labor over the experimental period for maternal arterial SaO2 or Hb values.
Mean Concentrations of Progesterone and PGF2
Progesterone concentrations decreased significantly (p < 0.001; Fig. 1) from a mean concentration of 23.0 ± 1.0 nM between -15 and -7 days relative to labor to 14.2 ± 1.0 n/m between -6 and -1 days relative to labor. A further significant decrease to a final concentration of 1.6 ± 0.5 nM occurred on the day of labor. PGF2
concentrations fluctuated around a baseline value of 0.9 ± 0.1 nm between -15 and -1 days relative to labor before increasing significantly (p < 0.003) on the day of labor to 8.1 ± 3.2 nm.
|
Pulse Frequency
Progesterone The mean number of arterial progesterone pulses identified during the experimental period tended to decrease (p < 0.08; Fig. 2) with increasing gestational age, from 0.22 ± 0.09 pulses/h between -15 and -13 days relative to labor to 0.09 ± 0.04 pulses/h between -3 and -1 days relative to labor (Fig. 2). No pulses were identified in any animal on the day of labor (Day 0).
|
PGF2
The mean number of pulses/h changed significantly (p < 0.01) over the experimental period, decreasing from 0.55 ± 0.23 pulses/h between 15 and 13 days before labor to 0.10 ± 0.05 pulses/h between 6 and 4 days before labor (Fig. 2). Between -3 and -1 days relative to labor, the pulse frequency of PGF2
identified in the UOV increased to 0.54 ± 0.16. On the day of labor (Day 0), no pulses were detected in any animal. While it was the aim of this study to collect samples simultaneously from both sides of the uterus, this was rarely possible.
Figure 3 provides an example of coincident PGF2
release from both sides of the uterus when blood was collected simultaneously.
|
Amplitude of Identified Pulses
Progesterone The amplitude of progesterone pulses identified within the systemic circulation did not change significantly between -15 and -1 day relative to labor, averaging 17.60 ± 3.06 nmol (Fig. 4).
|
PGF2
The amplitudes of identified pulses did not change significantly between 15 and 7 days before labor. The mean pulse amplitude was 1.56 ± 0.10 nmol. Between 6 and 4 days before labor, the amplitude of identified pulses was 0.31 ± 0.31 nmol; 31 days before labor, the mean amplitude was 1.21 ± 0.66 nmol.
| DISCUSSION |
|---|
|
|
|---|
During Days -9 to -4, PGF2
concentrations in the UOV remained low and unchanged while PGF2
pulse frequency decreased. Therefore the initial decrease in plasma progesterone cannot be attributed to an increase in PGF2
secretion or pulse amplitude or frequency. At this stage of gestation the fetal cortisol concentration, which is the signal for parturition, is rising [5, 8]. It has been suggested that cortisol of fetal origin may act on the placenta to promote production of estrogen or PGF2
[2, 4, 5, 14, 16, 17]. While estrogen has traditionally been thought to exert its luteolytic action via a stimulatory influence on PGF2
, neither PGF2
concentrations nor pulse frequency increased during this time. Thus, the cause of the early decline in plasma progesterone concentration is obscure.
Previous studies of parturition in goats have uniformly identified the marked decrease in maternal plasma progesterone concentrations that occurs on the day of labor, while the finding of an earlier, more gradual decline has been less constant. This may reflect the greater emphasis in earlier studies [5, 18] on events during the last few days of gestation as well as the relatively small magnitude of the early decline in progesterone. Whether progesterone withdrawal actually occurs in two discrete steps as suggested in this and a previous study [8] is open to question. The appearance of a two-stage process may result from the grouping and analysis of our data, and we do not discount the possibility that progesterone concentrations may fall continuously over the last week of gestation.
Over the last 3 days of gestation, immediately before the final, abrupt decrease in plasma progesterone, an increase in PGF2
pulse frequency was identified in the UOV. Identified pulses were of low amplitude and high frequency. These findings are similar to those previously reported for early pregnancy in the ewe in vivo [19]. The PGF2
pulse frequency at this time was not greater than it had been during the earlier stages of the study, and we do not conclude that the final demise of the corpus luteum can be attributed unequivocally to this cause. It is significant that at no time did we detect high-amplitude, long-duration pulses that characterize luteolysis in the cycling ewe [1921]. Interestingly, no pulses were detected in any animal on the day of labor, during which time basal concentrations of PGF2
were increasing rapidly and those of progesterone were falling precipitously. Our results are not inconsistent with a concentration-dependent luteolytic action of PGF2
at this time.
During the experimental period, there was no change in the amplitude of PGF2
pulses identified in the UOVs. Thus, late pregnant animals differ from cyclic animals, which show an increase in pulse frequency and amplitude prior to luteolysis. As the second stage of luteolysis was preceded by an increase in pulse frequency without any corresponding increase in pulse amplitude, it is possible that, as in the cyclic ewe [22], an increase in the frequency of PGF2
pulses may be important for this process. This proposal is supported by the studies of Hooper and Thorburn [21] and Zarco et al. [23], who found that, in cycling animals, the interval between pulses of PGF2
was significantly longer in sheep with persistent corpora lutea in comparison to normal cyclic ewes despite the similarities in amplitude of PGF2
pulses between the two groups. It was concluded that the defect resided in the control of PGF2
pulse frequency rather than in an inability to produce PGF2
[21].
Progesterone pulses in the systemic circulation were similar in frequency and amplitude relative to basal secretion to those reported from luteal tissue of the cow [24]. The similarity of progesterone profiles obtained in vivo and in vitro suggests that the small pulses we identified may result from regulatory mechanisms within the corpus luteum. In the present study, the frequency of progesterone pulses tended to decrease (p < 0.08) over the experimental period. This trend corresponded to the decrease in progesterone concentrations measured in the systemic circulation with increasing gestational age. These observations may reflect either an increase in the action of some unidentified luteolytic factor, withdrawal of luteotrophic support, or both.
The existence of a luteotrophic complex including pituitary LH, caprine placental lactogen, and PGE2 has been suggested [1, 2, 18], and parturition in the doe may result from withdrawal of one or more components of this proposed complex. Withdrawal of LH [1] or PGE2 [8] is not critical in regulating this process, and the potential role of placental lactogen has not been adequately investigated, although its concentrations appear to decline as fetal glucocorticoids increase [2, 18].
It has been demonstrated [5, 20, 25, 26] that progesterone secretion is labile, responding quickly to variations in PGF2
, both in vivo and in vitro. Throughout this study, however, progesterone secretion appeared unrelated to the pattern of PGF2
pulses in the maternal circulation. If PGF2
was acting in a regulatory capacity, then the changes seen in PGF2
would presumably have changed the progesterone pulse frequency or amplitude. The extent to which low-amplitude, high-frequency PGF2
pulse production influences pulsatile progesterone release is unknown. From this study, it does not appear that PGF2
directly influenced the pulsatile output of progesterone.
It is unclear at this time whether PGF2
release is controlled by the fetus or a systemic circulating agent(s). However, in the one doe that reliably yielded samples from both UOVs, data collected during all sampling days revealed an obvious coincidence in the occurrence of PGF2
pulses identified on each side of the uterus (see Fig. 3). The occurrence of coincident, bilateral PGF2
pulses in a twin-bearing doe suggests either that the stimulus for PGF2
release is influenced by a circulating agent derived from the doe or, less likely, that two independent fetuses are capable of influencing PGF2
release simultaneously.
While the results of this study have clarified the pulsatile profiles of PGF2
and progesterone in the UOV and systemic circulations, respectively, during the last 15 days of gestation, it is unclear whether the sampling interval of 15 min was frequent enough to identify all PGF2
pulses. The sampling interval used in this study was chosen on the basis of previous studies that, using a sampling interval of 1 h, were able to demonstrate PGF2
pulses in both cyclic and pregnant animals. Owing to the very short half-life of PGF2
in the body [27], it is possible that both this and previous studies have underestimated the optimal sampling frequency for PGF2
. In this study, however, each animal was subjected to an identical sampling regimen and identical pulse analysis criteria, and changes in pulse frequency with increasing gestational age were successfully identified. Thus, although some PGF2
pulses may have been missed, the long duration of the sampling periods and the balanced nature of the data suggest that any marked changes in the pulsatile secretion of PGF2
would have been detected.
Although these data question the role of PGF2
secreted into the UOV in luteolysis in the pregnant doe, this eicosanoid could reach the corpus luteum via an alternative route such as the lymphatic vessels. PGF2
can reach the ovary from the ovine uterus by this route [28], and its concentration in uterine lymph increases with the onset of labor in cows [29]. A further possibility is that PGF2
may be synthesized within the ovary itself by conversion from PGE2 via prostaglandin 9-ketoreductase. This enzyme occurs in most tissues including the human decidua, where a role in the onset of labor has been proposed [3032].
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: I.R. Young, Dept. of Physiology, Monash University, Wellington Rd., Clayton, Victoria, Australia 3168. FAX: 03 9905 2547; ross.young{at}med.monash.edu.au ![]()
Accepted: March 15, 1999.
Received: November 16, 1998.
| REFERENCES |
|---|
|
|
|---|
as signals for luteolysis. Biol Reprod 1998; 58:10651070.
concentrations in the utero-ovarian vein of pregnant and non-pregnant sheep. Endocrinology 1986; 119:25902597.
and oxytocin release during persistence of the corpus luteum in sheep. Acta Endocrinol 1987; 115:469477.
. Prostaglandins 1983; 26:347364.[CrossRef][Medline]
. Biol Reprod 1990; 42:239245.[Abstract]
is transferred from the uterus to the ovary in the sheep by lymphatic and blood vascular pathways. J Reprod Fertil 1985; 74:645656.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |