00l3-7227/90/1262-1199$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 126, No. 2 Printed in U.S.A.
Effect of Ovarian Steroids on a Nocturnal Surge of Prolactin Secretion that Precedes Parturition in the Rat DAVID R. GRATTAN AND ROBERT L. W. AVERILL Animal Physiology Research Unit, School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand
ABSTRACT. PRL secretion in several physiological and experimental conditions, including early pregnancy, is linked to the daily photoperiod. The aim of this study was to examine the antepartum increase in PRL secretion for evidence of a circadian pattern of release, as seen during early pregnancy. During the last 3 days of pregnancy blood samples were taken six times daily by means of previously implanted jugular cannulae. Plasma PRL concentrations were then measured by RIA. PRL levels were less than 10 ng/ml in all animals on day 19 of pregnancy, but during the light period of day 20 there was an increase to an average of 30 ± 10 ng/ml, with no evidence of a peak related to the time of day. However, in the dark period between days 20 and 21 there was a large surge of PRL secretion which reached peak levels of 356 ± 39 ng/ml at 0500 h on day 21, then returned to 48 ± 20 ng/ml at 1200 h, around the time of parturition. The peak always occurred at 0500 h and was not related to the time of parturition, which ranged from 1000-2200 h on day 21. Bilateral ovariectomy (OVX) on day 19 advanced
D
the time of delivery by approximately 12 h. In seven of nine animals, no surge of PRL secretion was observed during the dark period preceding parturition. Estradiol treatment after OVX on day 19 (OVX+E) advanced the time of delivery by approx 18 h. An antepartum PRL surge was present and was advanced by 24 h in all OVX+E animals, peaking at 0300 h on day 20. Progesterone treatment from day 18 to 21 in intact pregnant animals delayed parturition by approximately 18 h and prevented PRL secretion during the period of treatment. After progesterone treatment was stopped, a nocturnal surge of PRL secretion occurred, peaking at 0500 h on day 22, 24 h after the surge in normal animals. The results suggest that the increased PRL secretion during late pregnancy is linked to the daily photoperiod and is characterized by a nocturnal surge in the dark period preceding parturition. This surge is inhibited by progesterone, and it can be advanced 24 h by estradiol treatment in the absence of the ovaries. (Endocrinology 126: 1199-1205, 1990)
URING early pregnancy in the rat, PRL is secreted in a pattern of twice daily surges (1). One surge, the diurnal surge, occurs at the end of the light period, while the other, the nocturnal surge, peaks immediately before the end of the dark period. These surges cease at midpregnancy with the onset of placental lactogen secretion (2, 3), and PRL levels remain low until a small elevation occurs on the last day of pregnancy before parturition (4-8). PRL release is also linked to the daily photoperiod in several other physiological and experimental conditions. Like early pregnancy, pseudopregnancy is characterized by the presence of twice daily PRL surges (1). There is a large surge at the time of the diurnal surge during the afternoon of proestrus (9) and daily in estrogen-treated ovariectomized rats (9). Periodic pharmacological removal of the tonic dopaminergic inhibition of PRL in ovariectomized rats results in varying release of PRL depending on the time of day, with amplified peaks
associated with the regular surge times (10). The nature and control of the increase in PRL secretion preceding parturition has not been well studied. It is thought that the high levels of progesterone existing during late pregnancy inhibit PRL release, and the fall in progesterone concentration 36 h before parturition allows the PRL secretion to occur (6, 11, 12). Ovariectomy, however, has been variously reported to increase (6), decrease (7), and have no effect (8) on PRL secretion during late pregnancy. These contradictory results may be due to differences in the numbers of blood samples taken, the methods of collection, or a failure to allow for possible diurnal variations in PRL secretion. The aim of the present study was to determine the pattern of antepartum PRL secretion and to reexamine the role of ovarian steroids in the control of PRL secretion during late pregnancy.
Received September 5, 1989. Address all correspondence and requests for reprints to: David Grattan, School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand.
Virgin female Wistar rats (250-300 g) from our colony were maintained at 22 ± 1 C, with free access to food and water, under a 12-h light, 12-h dark cycle (lights on at 0600 h). Stages
Materials and Methods Animals
1199
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 May 2015. at 10:03 For personal use only. No other uses without permission. . All rights reserved.
PRL DURING LATE PREGNANCY
1200
of the estrous cycle were monitored by daily vaginal smears. Animals having at least two 4-day cycles were mated on the day of proestrus by placing them in individual cages with a male. Mating was confirmed by the presence of sperm in the vaginal smear the next morning, and this day was designated day 0 of pregnancy. Animals in our colony usually gave birth during the light period of day 21. Rats were observed hourly around the expected time of parturition. An estimated duration of pregnancy was calculated based on the assumptions that mating occurred at midnight during the night before sperm was observed in the vaginal smear, and that parturition began at the observation time when at least one pup was found (13). Blood sampling
Blood samples were taken during late pregnancy from conscious unrestrained animals by implanting an indwelling atrial cannula on day 17 of pregnancy. The right atrium was cannulated under ether anaesthesia, by the method of Popovic and Popovic (14), except that silicon tubing (Silastic Medical Grade Tubing, Dow-Corning, Medfield, MA; id, 0.025 in.; od, 0.047 in.) was used instead of polyethylene tubing. This silicon tubing has been shown to be generally unreactive with body tissues and does not induce blood clot formation, thus extending the time the cannula remains patent (15). The cannula was filled with sterile 0.9% saline containing 2.5 IU heparin/ml, plugged with a 21-gauge metal pin, then exteriorized through the skin at the back of the neck. The animals were injected with 50,000 U penicillin and 50 mg dihydrostreptomycin (Penstrep L.A., A/S Rosko, Denmark), sc, and allowed to recover for between 24-48 h before blood was removed for PRL measurement. Animals were kept in individual cages from the time of surgery throughout the course of the experiment. Blood samples (0.4 ml) were taken from 1200 h on day 19 of pregnancy at 0300, 0500, 1200,1700, and 2300 h each day until the first pup was delivered. Blood samples were centrifuged, and the plasma stored at -20 C until required. The red blood cells were resuspended in 0.2 ml sterile 0.9% saline and replaced into the animal after the next sampling. Alterations to ovarian hormones Fifteen animals were bilaterally ovariectomized under ether anesthesia through two small flank incisions between 0700 and 1100 h on day 19 of pregnancy (OVX). Blood levels of ovarian steroids were maintained in some OVX animals using Silastic implants (Silastic medical grade tubing, Dow-Corning). Six animals were treated with estradiol, with a 25-mm implant (id, 0.062 in.; od, 0.125 in.) containing 300 fig/ml estradiol benzoate dissolved in oil placed sc at the time of ovariectomy (OVX + E). Such an implant has been shown to maintain plasma estradiol levels at approximately 100 pg/ml (16). Five intact animals were treated with progesterone, with two 50-mm implants (id, 0.078 in.; od, 0.125 in.) containing crystalline progesterone placed sc under light ether anaesthesia at 0900 h on day 18. These implants were removed at 0700 h on day 21, the expected day of parturition. All implants were incubated sc for 24 h in nonexperimental
Endo • 1990 Vol 126 • No 2
animals before they were placed in the experimental animals, to initialize consistent hormone release. Hormone assays PRL was measured in duplicate by a RIA method modified from that of Neill and Reichert, Jr. (17). The assay reagents were provided by the NIDDK through the National Hormone and Pituitary Program (University of Maryland School of Medicine). PRL (rPRL 1-5) was iodinated with 125I by the chloramine-T method, using less than 2 /xg chloramine-T//*g PRL. Reference PRL (rPRL RP-3) was used to produce a standard curve from 0.4/500 ng/ml. Rabbit antiserum to rat PRL (anti-PRL-S-9) was used at an initial dilution of 1:2000, a titer that bound 37.5% of the labeled PRL in the assay. After an initial incubation overnight, bound PRL was separated from free PRL by a dilution of sheep antirabbit immunoglobulin antiserum (WARC 36, provided by Wallaceville Animal Research Centre, Upper Hutt, New Zealand) together with 10% polyethylene glycol. The use of polyethylene glycol enabled us to avoid the further incubation period normally associated with double antibody RIA and also reduced the amount of second antibody required to maximally precipitate the bound PRL. Nonspecific binding was 5.3%, and the inter- and intraassay coefficients of variation were 13.5% and 2.7%, respectively. Plasma PRL was expressed in terms of the NIDDK rPRL RP3 reference preparation. Results are shown as the mean ± SE. Statistical analysis Statistical significance of the data was determined by t test and analysis of variance at the 95% confidence level using the Statview 512+ program (Brainpower, Inc., Calabasas, CA) for the Apple Macintosh computer.
Results Time of parturition Figure 1 shows the mean duration of pregnancy for four different experimental groups. All animals had litters of at least eight pups. The mean duration of pregnancy in normal animals was 518.3 ± 1.6 h (Fig. 1), which corresponds to a time of parturition of 1420 h on day 21. Five of seven normal animals littered during the light phase of day 21 of pregnancy, while the remaining two animals littered early in the following dark period (Fig. 2). Ovariectomy on day 19 significantly advanced parturition relative to that in the controls (P < 0.001). All OVX animals littered during the dark period between days 20-21 (Fig. 3). Estradiol treatment after ovariectomy on day 19 produced a further advance in the time of delivery compared with OVX alone (P < 0.01). Progesterone treatment from days 18-21 significantly delayed parturition compared with that in the controls (P < 0.001), with all animals littering an average of 27.4 ± 0.9 h after removal of the progesterone implants, during the light phase of day 22.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 May 2015. at 10:03 For personal use only. No other uses without permission. . All rights reserved.
PRL DURING LATE PREGNANCY 400 -i
1201
Normal (mean values,n=7) D
9YX('PdLYidyaiyaiues, n=9)
Vol. 126, No. 2 Printed in U.S.A.
Effect of Ovarian Steroids on a Nocturnal Surge of Prolactin Secretion that Precedes Parturition in the Rat DAVID R. GRATTAN AND ROBERT L. W. AVERILL Animal Physiology Research Unit, School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand
ABSTRACT. PRL secretion in several physiological and experimental conditions, including early pregnancy, is linked to the daily photoperiod. The aim of this study was to examine the antepartum increase in PRL secretion for evidence of a circadian pattern of release, as seen during early pregnancy. During the last 3 days of pregnancy blood samples were taken six times daily by means of previously implanted jugular cannulae. Plasma PRL concentrations were then measured by RIA. PRL levels were less than 10 ng/ml in all animals on day 19 of pregnancy, but during the light period of day 20 there was an increase to an average of 30 ± 10 ng/ml, with no evidence of a peak related to the time of day. However, in the dark period between days 20 and 21 there was a large surge of PRL secretion which reached peak levels of 356 ± 39 ng/ml at 0500 h on day 21, then returned to 48 ± 20 ng/ml at 1200 h, around the time of parturition. The peak always occurred at 0500 h and was not related to the time of parturition, which ranged from 1000-2200 h on day 21. Bilateral ovariectomy (OVX) on day 19 advanced
D
the time of delivery by approximately 12 h. In seven of nine animals, no surge of PRL secretion was observed during the dark period preceding parturition. Estradiol treatment after OVX on day 19 (OVX+E) advanced the time of delivery by approx 18 h. An antepartum PRL surge was present and was advanced by 24 h in all OVX+E animals, peaking at 0300 h on day 20. Progesterone treatment from day 18 to 21 in intact pregnant animals delayed parturition by approximately 18 h and prevented PRL secretion during the period of treatment. After progesterone treatment was stopped, a nocturnal surge of PRL secretion occurred, peaking at 0500 h on day 22, 24 h after the surge in normal animals. The results suggest that the increased PRL secretion during late pregnancy is linked to the daily photoperiod and is characterized by a nocturnal surge in the dark period preceding parturition. This surge is inhibited by progesterone, and it can be advanced 24 h by estradiol treatment in the absence of the ovaries. (Endocrinology 126: 1199-1205, 1990)
URING early pregnancy in the rat, PRL is secreted in a pattern of twice daily surges (1). One surge, the diurnal surge, occurs at the end of the light period, while the other, the nocturnal surge, peaks immediately before the end of the dark period. These surges cease at midpregnancy with the onset of placental lactogen secretion (2, 3), and PRL levels remain low until a small elevation occurs on the last day of pregnancy before parturition (4-8). PRL release is also linked to the daily photoperiod in several other physiological and experimental conditions. Like early pregnancy, pseudopregnancy is characterized by the presence of twice daily PRL surges (1). There is a large surge at the time of the diurnal surge during the afternoon of proestrus (9) and daily in estrogen-treated ovariectomized rats (9). Periodic pharmacological removal of the tonic dopaminergic inhibition of PRL in ovariectomized rats results in varying release of PRL depending on the time of day, with amplified peaks
associated with the regular surge times (10). The nature and control of the increase in PRL secretion preceding parturition has not been well studied. It is thought that the high levels of progesterone existing during late pregnancy inhibit PRL release, and the fall in progesterone concentration 36 h before parturition allows the PRL secretion to occur (6, 11, 12). Ovariectomy, however, has been variously reported to increase (6), decrease (7), and have no effect (8) on PRL secretion during late pregnancy. These contradictory results may be due to differences in the numbers of blood samples taken, the methods of collection, or a failure to allow for possible diurnal variations in PRL secretion. The aim of the present study was to determine the pattern of antepartum PRL secretion and to reexamine the role of ovarian steroids in the control of PRL secretion during late pregnancy.
Received September 5, 1989. Address all correspondence and requests for reprints to: David Grattan, School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand.
Virgin female Wistar rats (250-300 g) from our colony were maintained at 22 ± 1 C, with free access to food and water, under a 12-h light, 12-h dark cycle (lights on at 0600 h). Stages
Materials and Methods Animals
1199
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 May 2015. at 10:03 For personal use only. No other uses without permission. . All rights reserved.
PRL DURING LATE PREGNANCY
1200
of the estrous cycle were monitored by daily vaginal smears. Animals having at least two 4-day cycles were mated on the day of proestrus by placing them in individual cages with a male. Mating was confirmed by the presence of sperm in the vaginal smear the next morning, and this day was designated day 0 of pregnancy. Animals in our colony usually gave birth during the light period of day 21. Rats were observed hourly around the expected time of parturition. An estimated duration of pregnancy was calculated based on the assumptions that mating occurred at midnight during the night before sperm was observed in the vaginal smear, and that parturition began at the observation time when at least one pup was found (13). Blood sampling
Blood samples were taken during late pregnancy from conscious unrestrained animals by implanting an indwelling atrial cannula on day 17 of pregnancy. The right atrium was cannulated under ether anaesthesia, by the method of Popovic and Popovic (14), except that silicon tubing (Silastic Medical Grade Tubing, Dow-Corning, Medfield, MA; id, 0.025 in.; od, 0.047 in.) was used instead of polyethylene tubing. This silicon tubing has been shown to be generally unreactive with body tissues and does not induce blood clot formation, thus extending the time the cannula remains patent (15). The cannula was filled with sterile 0.9% saline containing 2.5 IU heparin/ml, plugged with a 21-gauge metal pin, then exteriorized through the skin at the back of the neck. The animals were injected with 50,000 U penicillin and 50 mg dihydrostreptomycin (Penstrep L.A., A/S Rosko, Denmark), sc, and allowed to recover for between 24-48 h before blood was removed for PRL measurement. Animals were kept in individual cages from the time of surgery throughout the course of the experiment. Blood samples (0.4 ml) were taken from 1200 h on day 19 of pregnancy at 0300, 0500, 1200,1700, and 2300 h each day until the first pup was delivered. Blood samples were centrifuged, and the plasma stored at -20 C until required. The red blood cells were resuspended in 0.2 ml sterile 0.9% saline and replaced into the animal after the next sampling. Alterations to ovarian hormones Fifteen animals were bilaterally ovariectomized under ether anesthesia through two small flank incisions between 0700 and 1100 h on day 19 of pregnancy (OVX). Blood levels of ovarian steroids were maintained in some OVX animals using Silastic implants (Silastic medical grade tubing, Dow-Corning). Six animals were treated with estradiol, with a 25-mm implant (id, 0.062 in.; od, 0.125 in.) containing 300 fig/ml estradiol benzoate dissolved in oil placed sc at the time of ovariectomy (OVX + E). Such an implant has been shown to maintain plasma estradiol levels at approximately 100 pg/ml (16). Five intact animals were treated with progesterone, with two 50-mm implants (id, 0.078 in.; od, 0.125 in.) containing crystalline progesterone placed sc under light ether anaesthesia at 0900 h on day 18. These implants were removed at 0700 h on day 21, the expected day of parturition. All implants were incubated sc for 24 h in nonexperimental
Endo • 1990 Vol 126 • No 2
animals before they were placed in the experimental animals, to initialize consistent hormone release. Hormone assays PRL was measured in duplicate by a RIA method modified from that of Neill and Reichert, Jr. (17). The assay reagents were provided by the NIDDK through the National Hormone and Pituitary Program (University of Maryland School of Medicine). PRL (rPRL 1-5) was iodinated with 125I by the chloramine-T method, using less than 2 /xg chloramine-T//*g PRL. Reference PRL (rPRL RP-3) was used to produce a standard curve from 0.4/500 ng/ml. Rabbit antiserum to rat PRL (anti-PRL-S-9) was used at an initial dilution of 1:2000, a titer that bound 37.5% of the labeled PRL in the assay. After an initial incubation overnight, bound PRL was separated from free PRL by a dilution of sheep antirabbit immunoglobulin antiserum (WARC 36, provided by Wallaceville Animal Research Centre, Upper Hutt, New Zealand) together with 10% polyethylene glycol. The use of polyethylene glycol enabled us to avoid the further incubation period normally associated with double antibody RIA and also reduced the amount of second antibody required to maximally precipitate the bound PRL. Nonspecific binding was 5.3%, and the inter- and intraassay coefficients of variation were 13.5% and 2.7%, respectively. Plasma PRL was expressed in terms of the NIDDK rPRL RP3 reference preparation. Results are shown as the mean ± SE. Statistical analysis Statistical significance of the data was determined by t test and analysis of variance at the 95% confidence level using the Statview 512+ program (Brainpower, Inc., Calabasas, CA) for the Apple Macintosh computer.
Results Time of parturition Figure 1 shows the mean duration of pregnancy for four different experimental groups. All animals had litters of at least eight pups. The mean duration of pregnancy in normal animals was 518.3 ± 1.6 h (Fig. 1), which corresponds to a time of parturition of 1420 h on day 21. Five of seven normal animals littered during the light phase of day 21 of pregnancy, while the remaining two animals littered early in the following dark period (Fig. 2). Ovariectomy on day 19 significantly advanced parturition relative to that in the controls (P < 0.001). All OVX animals littered during the dark period between days 20-21 (Fig. 3). Estradiol treatment after ovariectomy on day 19 produced a further advance in the time of delivery compared with OVX alone (P < 0.01). Progesterone treatment from days 18-21 significantly delayed parturition compared with that in the controls (P < 0.001), with all animals littering an average of 27.4 ± 0.9 h after removal of the progesterone implants, during the light phase of day 22.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 May 2015. at 10:03 For personal use only. No other uses without permission. . All rights reserved.
PRL DURING LATE PREGNANCY 400 -i
1201
Normal (mean values,n=7) D
9YX('PdLYidyaiyaiues, n=9)
