The Regulation of Prolactin Receptor Messenger Ribonucleic Acid ...

0013-7227/99/$03.00/0 Endocrinology Copyright © 1999 by The Endocrine Society

Vol. 140, No. 5 Printed in U.S.A.

The Regulation of Prolactin Receptor Messenger Ribonucleic Acid Levels in the Sheep Liver before Birth: Relative Roles of the Fetal Hypothalamus, Cortisol, and the External Photoperiod* I. D. PHILLIPS, R. V. ANTHONY, D. C. HOUGHTON,

AND

I. C. MCMILLEN

Department of Physiology, University of Adelaide (I.D.P., D.C.H., I.C.M.), Adelaide, South Australia 5005, Australia; and Department of Physiology, Colorado State University (R.V.A.), Fort Collins, Colorado 80523 ABSTRACT We have investigated the separate actions of hypothalamo-pituitary disconnection (HPD), with or without cortisol administration, and changes in the external photoperiod on the regulation of the levels of messenger RNA (mRNA) encoding long (PRLR1) and short (PRLR2) forms of PRL receptor in the liver of the fetal lamb. In pregnant Merino ewes (n 5 20), the hypothalamus and pituitary were surgically disconnected in 13 fetuses (HPD group), and fetal vascular catheters were implanted in the HPD group and in an additional 7 fetuses (intact 1 saline group) between 104 –120 days gestation (d). Fetal sheep in the HPD group were infused with either cortisol (3.5 mg/4.8 ml saline/24 h; HPD 1 F; n 5 5) or saline for 5 days between 134 –141 d, and saline was also infused in the intact group within the same gestational age range. A second group of pregnant ewes (n 5 12) was kept in a 12-h light, 12-h dark cycle from 70 d until implantation of fetal vascular catheters between 106 –120 d, after which ewes were allocated to either a long photoperiod (16 h of light, 8 h of darkness; LL group; n 5 6) or a short photoperiod (8 h of light, 16 h of darkness; SL group; n 5 6) regimen. Circulating cortisol concentrations were higher (P , 0.05) in the intact fetal sheep (18.7 6 3.8 nmol/liter) than in the HPD 1 saline group (1.5 6 0.6 nmol/liter), and were further

I

T IS WELL established in several species, including the sheep, that PRL is present in the fetal circulation, and that it does not cross the placenta from the ewe to the fetus (1, 2). In the sheep, PRL synthesis and secretion are highly regulated before birth. PRL mRNA levels in the fetal sheep pituitary and PRL concentrations in the fetal circulation are maintained by a tonic stimulatory drive from the hypothalamus throughout late gestation and increase concomitantly around 135 days [term 5 147 6 3 days gestation (d)] (3, 4). Furthermore, as in the adult sheep, PRL concentrations are low in the sheep fetus during the winter or in short photoperiods and increase during summer or in long photoperiods (4 – 6). Although circulating PRL concentrations are regulated by the fetal hypothalamus and the external photoperiod, the role of fetal PRL in the growth and development of fetal tissues during late gestation or in different photoperiod con-

Received June 29, 1998. Address all correspondence and requests for reprints to: Dr. Ian Phillips, Department of Physiology, Monash University, Clayton, Victoria 3168, Australia. E-mail: [email protected]. * This work was supported by the Australian Research Council.

increased (P , 0.05) in the HPD 1 cortisol group (97.4 6 23.7 nmol/ liter). Fetal PRL concentrations were lower (P , 0.05) in the HPD 1 saline (10.6 6 4.3 ng/ml) and HPD 6 cortisol (5.6 6 2 ng/ml) groups compared with those in the intact group (38.9 6 6.8 ng/ml). The levels of hepatic PRLR mRNA were higher (P , 0.05) in the intact (PRLR1, 27.4 6 6.1; PRLR2, 17.7 6 2.5) and HPD 1 cortisol (PRLR1, 23.4 6 0.4; PRLR2, 15.3 6 3.0) groups than in the HPD 1 saline group (PRLR1, 10.6 6 1.8; PRLR2, 8.9 6 1.8) at 140/141 d. The mean plasma PRL concentration in the LL group (70 6 9 ng/ml) was higher (P , 0.05) than that in the SL group (34 6 15 ng/ml), whereas the levels of hepatic PRLR1 mRNA (LL group, 4.6 6 0.9; SL group, 4.3 6 0.8) and PRLR2 mRNA (LL group, 3.4 6 0.4; SL group, 3.0 6 0.5) at 140 –141 d were not different. These data indicate that cortisol acts directly or indirectly to maintain hepatic PRLR mRNA levels in the sheep fetus during late pregnancy. In contrast, changes in the external photoperiod and circulating PRL concentrations in the sheep fetus do not directly alter PRLR expression in the fetal liver. These studies provide further insight into the role that the PRL axis may play in the transduction of signals about the external environment to the fetus as it prepares for the transition to extrauterine life. (Endocrinology 140: 1966 –1971, 1999)

ditions remains poorly understood. It is clear from a range of studies that mRNA transcripts encoding long and short forms of PRL receptor (PRLR) are widely distributed in tissues derived from all three germ layers in the fetal rat (7) and mouse (8), and that immunoreactive PRLR is present in tissues derived from mesoderm and endoderm in the early human fetus (9). We have also recently demonstrated that the expression of mRNAs encoding the long (PRLR1) and short (PRLR2) forms of PRLR increase in the liver of the sheep fetus during the last 3 weeks of gestation (10). Anthony and colleagues (11) have shown that the sequence of the ovine PRLR2 complementary DNA (cDNA) is identical to that of ovine (o) PRLR1 cDNA until nucleotide 420, corresponding to E 261 in the oPRLR peptide-coding sequence (12). At this point, an insertion of 39 bases occurs between homology boxes 1 and 2 within the cytoplasmic domain of the receptor. The additional nucleotides in PRLR2 cDNA encode 11 amino acids followed by 2 stop codons (11, 12). Translation of PRLR2 mRNA is therefore predicted to encode a truncated or short form of PRLR. Although the long form of PRLR is able to transmit a full intracellular signal, the function of the short form of PRLR is not clear, although in vitro studies have

1966

PROLACTIN RECEPTOR IN THE FETAL LIVER

shown that the ratio of long PRLR to short PRLR is important in intracellular signaling (13). We also found that intrafetal administration of cortisol stimulated the expression of PRLR1 and PRLR2 mRNA in fetal sheep liver and therefore suggested that increases in hepatic PRLR expression before birth may be a result of the prepartum increase in fetal cortisol (10). PRLR abundance in adult rat liver has also been shown to be regulated by circulating PRL (14), but there is no information in other species, including the sheep, on the role of PRL in the control of PRLR gene expression. There is also no information concerning the role of the external photoperiod in the control of PRLR gene expression during development in seasonal species such as the sheep. In the present study we have therefore investigated the relative roles of the fetal hypothalamus, cortisol, and the external photoperiod in the regulation of PRLR expression in the liver of the sheep fetus. We used an animal model in which the fetal hypothalamus and pituitary are surgically disconnected [hypothalamo-pituitary-disconnection (HPD)] (4, 15). It has previously been demonstrated that surgical disconnection of the fetal pituitary from the hypothalamus does not alter the morphology, distribution, or proportion of lactotrophs in the fetal pituitary (15). Furthermore, the fetal PRL response to intrafetal administration of TRH and to alterations in the external photoperiod are maintained after HPD, providing evidence that the lactotrophs are fully functional after this procedure (16). It has also been demonstrated that HPD prevents the normal prepartum increase in fetal cortisol (16, 17). We compared the mRNA levels of PRLR1 and PRLR2 mRNA in the liver of the HPD fetal sheep (low PRL, low cortisol) with those in the HPD sheep fetus in which cortisol was infused (HPD 1 F; low PRL, high cortisol). Finally, we determined the effects of exposure of the ewe to different photoperiods on the expression of PRLR mRNA in the fetal liver. We measured hepatic PRLR1 and PRLR2 mRNA levels in late gestation fetuses from ewes maintained in either long (LL) or short (SL) photoperiods. Materials and Methods Animals and surgery All experimental procedures were approved by the University of Adelaide standing committee on ethics in animal experimentation. Surgery was performed on pregnant Merino ewes (n 5 32) between 104 –120 d under aseptic conditions. General anesthesia was induced by an iv injection of sodium thiopentone (1.25 mg; Boehringer Ingelheim, Sydney, New South Wales, Australia) and maintained with 3– 4% halothane in oxygen and N2O-O2 (50:50, vol/vol). Catheters were inserted in the fetal and maternal carotid arteries and jugular veins and in the amniotic sacs of all ewes. Fetal catheters were exteriorized through an incision in the ewe’s flank, and all catheters were filled with heparinized saline. Disconnection of the fetal hypothalamus and pituitary (HPD) was performed in 13 fetuses (15). A midline incision was made in the fetal nose, and the nasal bone was opened just left of the intranasal septum. The optic chiasm was located and exposed to allow access to the median eminence. The neural tissues of both internal and external laminae of the median eminence were then removed using gentle suction. A small piece of gelfoam soaked in thrombin (Thrombostat, Parke-Davis, Caringbah, Australia) and antibiotics (Intervet, Lane Cove, Australia) was introduced to separate the remaining hypothalamic tissue from the pituitary. All sheep received a 2-ml im injection of antibiotics (procaine penicillin, 250 mg/ml; dihydrostreptomycin, 250 mg/ml; procaine hydrochloride, 20 mg/ml) after the induction of anesthesia. All ewes were fed alfalfa chaff once daily between 0900 –1100 h with water ad libitum and housed

1967

in individual pens in animal holding rooms with a 12-h light, 12-h dark cycle.

Experimental protocols HPD study. Ewes carrying intact (n 5 7) or HPD fetuses (n 5 13) were housed under a 12-h light, 12-h dark cycle from the time of entering the animal house at 95–110 days until postmortem. At 134, 135, or 136 d, cortisol (3.5 mg/4.8 ml sterile saline/24 h; Solucortef, Upjohn, Kalamazoo, MI), was infused iv to five fetal sheep in the HPD group for 5 days (HPD 1 cortisol group). Sterile heparinized saline (5 ml/24 h) was administered iv to the remaining eight HPD fetal sheep (HPD 1 saline group) and to seven intact fetal sheep (intact group) for 5 days between 134/135 and 139/140 d. Photoperiod study. Pregnant ewes (n 5 12) were kept in a central holding facility between April and September (autumn and winter in the southern hemisphere) under a 12-h light, 12-h dark cycle (lights off at 1900 h) from at least 70 d, calculated from the time of mating, until surgery was performed between 106 –120 d to implant fetal and maternal vascular catheters. Immediately after surgery, the ewes were housed in either a long photoperiod (LL group; 16 h of light, 8 h of darkness; n 5 6) or a short photoperiod (SL group; 8 h of light, 16 h of darkness; n 5 6) until 140 –141 d. Two ewes in each lighting regimen carried twins. Fetal blood sampling. Fetal arterial (2-ml) blood samples were collected into heparinized tubes at least three times per week between 0900 –1200 h from 104 –141 d. Plasma samples were prepared from fetal arterial blood by centrifugation at 1500 3 g for 10 min at 4 C and stored at 220 C until required for measurement of PRL and cortisol. Fetal and maternal well-being were monitored by the collection of additional arterial blood (0.5 ml) from the fetus and ewe, and whole blood pH, PO2, PCO2, oxygen saturation, and hemoglobin content were measured using an ABL 550 acid base analyzer and an OSM2 hemoximeter (Radiometer, Copenhagen, Denmark). Fetal liver collection. At 139 –141 d, all ewes were killed by an iv overdose of sodium pentobarbitone. All fetal sheep (intact 1 saline group, n 5 7; HPD 1 saline group, n 5 8; HPD 1 F group, n 5 5; LL group, n 5 8; SL group, n 5 8) were delivered by hysterotomy and weighed. Curved crown-rump lengths were measured in the LL and SL groups. Fetuses were killed by decapitation, and fetal livers (n 5 33; intact 1 saline group, n 5 6; HPD 1 saline group, n 5 6; HPD 1 F group, n 5 5; LL group, n 5 8; SL group n 5 8) were collected, weighed, snap-frozen in liquid nitrogen, and stored at 270 C. PRL RIA. Plasma PRL concentrations were measured using rabbit antioPRL (anti-oPRL-2, batch AFP35810691, donated by the National Hormone and Pituitary Program, NIDDK, NIH, and obtained through Ogden Bioservices Corp., Rockville, MD) and a RIA previously described for use in fetal sheep plasma (3, 4). The sensitivity of the assay was 0.1 ng/tube, and the inter- and intraassay coefficients of variation were less than 20% and 10%, respectively. Cortisol RIA. Total cortisol concentrations were measured in fetal sheep plasma by RIA using an Orion Diagnostica kit (Orion Diagnostica, Turku, Finland). Before assay, cortisol was extracted from the plasma using dichloromethane as previously described (18). The efficiency of recovery of [125I]cortisol from fetal plasma using this procedure was at least 90%. The sensitivity of the assay was 0.39 nmol/liter, and the cross-reactivity of the rabbit anti-cortisol antibody was less than 1% with cortisone and 17-hydroxyprogesterone and less than 0.01% with pregnenolone, aldosterone, progesterone, and estradiol. The inter- and intraassay coefficients of variation were always less than 10%. Total RNA extraction from fetal liver. Total RNA was extracted from fetal liver samples (n 5 29; intact 1 saline, n 5 6; HPD 1 saline, n 5 6; HPD 1 F, n 5 5; LL, n 5 7; SL, n 5 5) by homogenization in 4 m guanidine hydrochloride and ultracentrifugation through a 5.7-m cesium chloride cushion. The purity and concentration of the nucleic acids were quantitated by spectrophotometric measurement as described previously (19).

1968


Measurement of mRNA encoding long and short forms of PRLR: synthesis of oPRLR-2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) riboprobes. Antisense complementary RNA (cRNA) oPRLR-2 and oGAPDH riboprobes were transcribed in vitro as previously described (11) by addition of the following reagents to a microfuge tube: 4 ml 5 3 transcription buffer [200 mm Tris-HCl (pH 7.5), 30 mm MgCl2, 10 mm spermidine, and 50 mm NaCl], 2 ml 100 mm dithiothreitol, 0.8 ml RNA guard (Pharmacia Biotech, Sydney, Australia), 4 ml nucleotide triphosphate mix (2.5 mm ATP, 2.5 mm GTP, and 2.5 mm CTP), 2.4 ml 100 mm UTP, 0.5 mg (1.0 ml) linearized pCRII oPRLR-2 (11), 5 ml (50 mCi) [a-32P]UTP (3000 Ci/mmol; Bresatec, Adelaide, South Australia), and 1 ml (10 U) T7 RNA polymerase (Progen, Darra, Queensland). The reaction was incubated at 37 C for 1 h, then DNA template was removed by digestion with 1 U deoxyribonuclease 1 (Boehringer Mannheim, Sydney, Australia) for 20 min at 37 C. Newly synthesized cRNA was separated from unincorporated nucleotides by column chromatography through Sephadex G-50 (Nick Column, Pharmacia Biotech). The riboprobe complementary to a portion of oGAPDH cDNA was synthesized by transcription from 50 ng linearized pCRII oGAPDH in a reaction like that described above but in which 2.5 ml (25 mCi) [a-32P]UTP, 2.4 ml 500 mm cold UTP, and 10 U (1 ml) T3 RNA polymerase (Progen) were added to the reaction mix. The specific activities of newly synthesized antisense PRLR-2 (660 bases) and oGAPDH (620 bases) riboprobes were greater than 1 3 109 cpm/mg RNA and 1 3 108 cpm/mg RNA, respectively. The integrity and size of the transcribed riboprobes were assessed by autoradiography after electrophoresis of 30,000 cpm of each riboprobe on a 5% polyacrylamide gel containing 8 m urea in 1 3 TBE buffer (90 mm Tris, 89 mm boric acid, and 20 mm EDTA, pH 8). Radiolabeled cRNA transcripts of defined length (Century Base Markers, Ambion, Inc., Austin, TX) were also synthesized as described above and loaded into lanes (20,000 cpm/lane) on the gel. The gel was run in 1 3 TBE buffer at 250 V for 4 h, then dried on a vacuum gel drier and exposed to a phosphorimager plate (BAS, Fuji Photo Film Co., Ltd., Tokyo, Japan) for 16 h. The image was visualized using a phosphorimager machine (Fuji Photo Film Co., Ltd., BAS 1000) and MacBAS software (version 2.2). Hybridization of total RNA and oPRLR and GAPDH riboprobes. Liver total RNA (10 mg) and two controls containing yeast transfer RNA (10 mg) were combined with antisense PRLR2 (3 3 105 cpm), GAPDH (1 3 105 cpm) riboprobes, and sterile water to a final volume of 20 ml. The RNA was precipitated by the addition of 2 ml 3 m sodium acetate (pH 5.2), 50 ml absolute ethanol, and incubation on dry ice for 15 min. The RNA was pelleted by centrifugation in a microfuge for 20 min at 4 C, and the supernatant was carefully removed. The RNA pellet was briefly airdried, then resuspended in 20 ml hybridization buffer (80% formamide, 40 mm PIPES (pH 6.4), 400 mm NaCl, and 0.1 mm EDTA) and denatured by heating at 85 C for 4 min. Hybridization of the total RNA and riboprobes was carried out by overnight incubation at 45 C. Ribonuclease (RNase) digestion. RNases (1 mg/ml RNase A and 12.5 U/ml RNase T1; Sigma Chemical Co., St. Louis, MO) in RNase buffer [300 mm NaCl, 10 mm Tris-HCl (pH 7.4), and 5 mm EDTA], was added (200 ml) to each hybridization, except for one control tube to which RNase buffer without RNases was added. Single stranded RNA was removed by incubation for 1 h at 30 C. The RNases were inactivated and RNA was precipitated by the addition of 300 ml solution Dx (Ambion, Inc.) and incubation at 220 C for 20 min. The RNA hybrids were pelleted by centrifugation in a microfuge for 30 min at 4 C, and all supernatant was carefully removed. The RNA pellet was briefly air-dried, and then resuspended in 8 ml gel loading buffer [80% formamide, 0.1% (wt/vol) xylene cyanol, 0.1% (wt/vol) bromophenol blue, and 2 mm EDTA]. The solution was heated at 85 C for 4 min, loaded onto a 5% polyacrylamide gel containing 8 m urea in 1 3 TBE, and run as described above. Detection and quantification of protected PRLR2 and GAPDH cRNA fragments. Autoradiographic images on the phosphorimage plate were visualized and quantified as described above. The antisense oPRLR2 cRNA probe (660 bases) was protected from RNase digestion by mRNA encoding a long and a short form of PRL receptor. The sizes of antisense PRLR2 cRNA fragments protected by mRNA encoding a long form of PRL receptor (PRLR1) were 440 and 73 bases. In this study, the relative amount of the 440-base probe fragment protected by PRLR-1 mRNA was quantified. mRNA encoding a short form of PRLR (PRLR2) protected 552 bases of PRLR2 antisense cRNA probe. GAPDH mRNA in fetal liver

Endo • 1999 Vol 140 • No 5

samples protected the antisense oGAPDH cRNA probe (650 bases) from RNase digestion. The signal output for PRLR2 and GAPDH riboprobes protected by PRLR1, PRLR2, and GAPDH mRNAs increased linearly with the input (10 –30 mg) of liver total RNA, as previously described (10).

Statistical analyses All data are presented as the mean 6 sem. The mean plasma cortisol and PRL concentrations were calculated for each fetus in the intact 1 saline, HPD 1 saline, and HPD 1 F groups during the 5-day infusion period of either saline or cortisol. The mean plasma PRL and cortisol concentrations for fetuses in the intact 1 saline, HPD 1 saline, and HPD 1 F groups were compared using one-way ANOVA. Where the ANOVA indicated differences between groups, Fisher’s test (a 5 0.05) was used post-hoc to identify which groups were different. The densitometric values for the 440- and 552-base fragments of PRLR2 cRNA probe in liver total RNA samples, protected by PRLR1 and PRLR2 mRNA from hydrolysis by RNases, were calculated as a ratio of the GAPDH cRNA signal in the same sample. Fetal body and liver weights, liver/body weight ratio, and PRLR1/GAPDH mRNA and PRLR2/GAPDH mRNA ratios in the intact 1 saline, HPD 1 saline, and HPD 1 cortisol groups were all compared using one-way ANOVA and Fisher’s post-hoc test. A mean plasma PRL concentration was also calculated for each fetus within the LL and SL groups for the period between 110 –141 d, and these values were compared using Student’s unpaired t test. Fetal body and liver weights, liver/body weight ratio, and PRLR1/GAPDH mRNA and PRLR2/GAPDH mRNA ratios in the LL and SL groups were compared using Student’s unpaired t test.

Results HPD and cortisol administration: fetal body and liver weights

There were no differences in fetal body or liver weights among the intact 1 saline group (4.11 6 0.16 kg and 94.2 6 10.3 g, respectively; n 5 6), the HPD 1 saline group (3.86 6 0.22 kg and 77.0 6 7.0 g, respectively; n 5 6), and the HPD 1 cortisol group (3.68 6 0.18 kg and 103.4 6 4.1 g, respectively; n 5 5). The fetal liver/body weight ratio was significantly lower (P , 0.05), however, in the HPD 1 saline group (19.8 6 1.2 g/kg) than in the intact 1 saline group (22.7 6 2.0 g/kg). Administration of cortisol for 5 days to HPD fetuses increased (P , 0.05) the liver/body weight ratio (HPD 1 cortisol group, 28.2 6 1.5 g/kg) above that in the intact 1 saline group. Fetal cortisol and PRL concentrations

The mean plasma cortisol concentrations between 134 –141 d were significantly higher (P , 0.05) in the HPD 1 cortisol group (97.4 6 23.7 nmol/liter; n 5 5 fetuses) than in the intact 1 saline group (18.7 6 3.8 nmol/liter; n 5 5 fetuses) or the HPD 1 saline group (1.5 6 0.6 nmol/liter; n 5 5 fetuses). The mean circulating PRL concentrations between 134 –141 d in the HPD 1 saline group (10.6 6 4.3 ng/ml; n 5 8 fetuses) and the HPD 1 cortisol group (5.6 6 2.0 ng/ml; n 5 5 fetuses) were lower (P , 0.05) than in the intact 1 saline group (37.1 6 6.8 ng/ml; n 5 7 fetuses; Fig. 1). Hepatic PRLR1 and PRLR2 mRNA

PRLR mRNA encoding a long and a short form of PRLR was present in livers of the intact, HPD 1 saline, and HPD 1 cortisol groups (Fig. 2). At 139 –141 d, the relative amounts of hepatic PRLR1/GAPDH mRNA and PRLR2/GAPDH


FIG. 1. Mean plasma PRL concentrations in fetal sheep with (intact) or without (HPD) an intact hypothalamo-pituitary connection during the administration of saline (intact 1 saline, HPD 1 saline) or cortisol (3.5 mg/24 h; HPD 1 F) to fetuses for 5 days between 134 –141 d. Different superscripts denote significant differences (P , 0.05) between treatment groups.

mRNA (Fig. 3) were lower (P , 0.05) in the HPD 1 saline group (10.6 6 1.8 and 8.9 6 1.8, respectively; n 5 6) compared with those in the intact 1 saline group (27.4 6 6.1 and 17.7 6 2.5 respectively; n 5 6). Cortisol replacement after HPD increased the relative amounts of PRLR1/GAPDH and PRLR2/GAPDH mRNA (23.4 6 4.0 and 15.3 6 3.0, respectively; n 5 5) to levels that were not different from those in the intact 1 saline group (Fig. 3). There were no differences among the three groups in the abundance of hepatic GAPDH mRNA, measured in arbitrary densitometric units (intact 1 saline, 792 6 184; HPD 1 saline, 811 6 34; HPD 1 F, 985 6 157). Effects of long and short photoperiods: fetal body and liver weights

There were no differences in the mean curved crown-rump length, fetal body weight, fetal liver weight, or fetal liver/ body weight ratios between the two photoperiod groups (Table 1). Plasma PRL concentrations and hepatic PRLR mRNA levels

The mean PRL concentration in fetal plasma between 109 – 141 d was higher (P , 0.05) in the LL group than in the SL group (Table 1). The mean ratios of hepatic PRLR1/GAPDH mRNA and PRLR2/GAPDH mRNA in the LL group [4.58 6 0.86 (n 5 7) and 3.36 6 0.36 (n 5 7), respectively] were not different, however, compared with those in the SL group [4.26 6 0.79 (n 5 4) and 3.02 6 0.45 (n 5 4), respectively; Fig. 4]. There was no difference in GAPDH mRNA abundance (arbitrary densitometric units) in fetal liver between the LL (170.1 6 16.8) and SL groups (188.0 6 26.6). Discussion

We have demonstrated that the expression of PRLR1 and -2 mRNA levels in fetal liver is reduced after HPD and

1969

restored after cortisol replacement. In addition, we have shown that PRLR1 and PRLR2 mRNA expression in fetal sheep liver is not regulated by changes in external photoperiod during late gestation. We previously reported that there was an increase in the expression of the two forms of PRLR in the fetal sheep liver after 130 days gestation, coincident with the timing of the prepartum cortisol surge in the sheep (10). Furthermore, we found that intrafetal infusion of cortisol before 130 days resulted in 2.4- and 1.8-fold stimulation of hepatic PRLR1 and PRLR2 mRNA levels, respectively (10). Previous studies have shown a positive association between PRLR mRNA levels and lactogenic hormone binding in the liver of the adult rat (20). The normal prepartum surge in fetal cortisol does not occur after the fetal hypothalamus and pituitary have been surgically disconnected earlier in gestation (16, 17), and in the present study, we found that hepatic expression of PRLR1 and PRLR2 mRNA was reduced by 67% and 50%, respectively, in HPD fetal sheep at 140 days gestation. Cortisol administration to the late gestation HPD fetus restored hepatic PRLR mRNA levels to levels comparable to those measured in intact fetal sheep at 140 days gestation. In contrast to cortisol, there was no relationship between circulating PRL concentrations and hepatic PRLR expression in the HPD or intact fetal sheep. It appears from our studies that the basal level of expression of PRLR in the fetal liver before 130 days gestation and after fetal HPD is relatively cortisol independent. There is, however, no increase in hepatic PRLR expression in the late gestation HPD fetus in the absence of cortisol replacement, and this suggests that the gestational increase is cortisol dependent. Although plasma cortisol concentrations were significantly higher in the HPD 1 cortisol group than in the intact group, the levels of hepatic PRLR1 and -2 mRNA were the same in each group. This may indicate that there is a specific range of circulating cortisol levels across which a hepatic PRLR response occurs. Alternatively, it has also been demonstrated that infusion of glucocorticoids stimulates the synthesis of corticosteroid-binding globulin in fetal sheep liver, and this could also act to limit the availability of cortisol within the fetal circulation (21). It is interesting that the expression of PRLR1 appeared marginally more sensitive than that of PRLR2 to the effects of HPD and subsequent cortisol replacement. In the ovine fetus, the levels of mRNA encoding long and short forms of hepatic PRLR are comparable throughout late gestation (10), whereas in the liver of the pregnant rat, mRNA encoding a short form of PRLR predominates (22). These differences could be species dependent and may arise from the divergence in the structural organization of the ovine and rat PRLR genes (12). Alternatively, there may be differences in the requirements for PRL activity in fetal liver during development and in adult liver during pregnancy. The different amounts of mRNA encoding long and short forms of hepatic PRLR may influence the cellular response elicited by lactogenic hormones binding to the different forms of PRLR in the liver of the late gestation sheep fetus compared with those in the liver of the pregnant rat. It is not yet known whether cortisol acts directly or indirectly to regulate hepatic PRLR mRNA levels in the fetal

1970


Endo • 1999 Vol 140 • No 5

FIG. 2. Autoradiograph of PRLR2 cRNA products after hybridization of PRLR2 cRNA probe with total RNA (10 mg/lane) from livers of fetal sheep with (intact) or without (HPD) an intact hypothalamo-pituitary connection during the administration of saline (intact 1 saline, HPD 1 saline) or cortisol (3.5 mg/24 h; HPD 1 F) to fetuses for 5 days between 134 –141 d. PRLR2 cRNAs protected by PRLR mRNA from RNase degradation were separated by electrophoresis on a 5% polyacrylamide gel containing 8 M urea. Radiolabeled cRNA size standards (bases) were included on the gel.

FIG. 4. The mean levels of PRLR1/GAPDH and PRLR2/GAPDH mRNA in livers of fetal sheep at 140 –141 d in ewes exposed to a 16-h light, 8-h dark cycle (LL group) or an 8-h light, 16-h dark cycle (SL group) between 109 –141 d. FIG. 3. Hepatic PRLR1/GAPDH and PRLR2/GAPDH mRNA levels at 139 –141 d in fetal sheep with (intact) or without (HPD) an intact hypothalamo-pituitary connection after the administration of saline (intact 1 saline, HPD 1 saline) or cortisol (3.5 mg/24 h; HPD 1 F) to fetuses for 5 days between 134 –141 d. Different superscripts denote significant differences (P , 0.05) between treatment groups. TABLE 1. Body weights, crown-rump lengths, and liver weights at 140 –141 days and the mean plasma PRL concentrations of fetal sheep in ewes exposed to a 16-h light, 8-h dark cycle (LL group) or an 8-h light, 16-h dark cycle (SL group) between 109 –141 days

BW (kg) Crown-rump length (cm) Liver wt (g) Liver:BW ratio (g/kg) Plasma PRL (ng/ml) a

LL group

SL group

4.34 6 0.18 52.6 6 0.7 101.6 6 6.5 23.6 6 1.1 70 6 9a

4.54 6 0.14 51.7 6 9.7 107.4 6 6.4 23.3 6 2.7 34 6 15

Significant difference (P , 0.05) between groups.

sheep. Although promoter regions of the rat PRLR gene have been identified, none contains a glucocorticoid response element (23), and there is no current information that defines the regulatory regions of the oPRLR gene. It is also possible

that glucocorticoids may regulate PRLR transcription indirectly by modulating the transcriptional activity of gene(s)encoding proteins, which, in turn, regulate PRLR gene transcription. Furthermore, although infusion of cortisol does not stimulate PRL synthesis and secretion, it is possible that cortisol acts in the intact fetus during early gestation or in the HPD fetus during late gestation to stimulate an increase in other hormones, such as thyroid or placental hormones (24), which, in turn, increase PRLR expression in the fetal liver. It has also recently been demonstrated, however, that expression of the GH receptor (GHR) and insulin-like growth factor I (IGF-I) mRNA increases between 130 –140 days gestation in the fetal sheep liver, and that cortisol infusion before 130 days also stimulates hepatic GHR and IGF-I gene expression (25). PRLR and GHR are both members of the cytokine receptor superfamily and share common structural features, in that they possess a large extracellular domain, a single transmembrane region, and a cytoplasmic domain (26). It is possible, therefore, that regulation of PRLR, GHR, and IGF-I gene expression by cortisol plays an important role in the changes that occur in hepatic metabolism and somatotropic


regulation either before birth or in response to intrauterine conditions such as chronic hypoxemia that result in an increase in circulating fetal cortisol concentrations (27). There was a decrease in the size of the fetal liver relative to fetal body weight in the HPD group, and it was interesting that cortisol replacement increased the fetal liver/body weight ratio. Administration of PRL to rats causes hepatic hypertrophy (28); however, the relative roles of cortisol, PRL, and GH in hepatic growth and development before birth remain to be established. Although PRL synthesis and secretion are regulated by the external photoperiod before and after birth in sheep (4 – 6), we found no evidence for a photoperiod effect on the expression of the mRNAs encoding either the long or the short form of the PRLR in fetal liver. Previous studies in the rat have found that several days of exposure to moderately elevated levels of PRL increase PRLR levels in the liver (26). It is possible that in sheep during late gestation, cortisol acts directly or indirectly to maximally stimulate PRLR levels, and therefore, PRLR expression in the fetal liver is relatively independent of the prevailing photoperiod and fetal PRL levels. The effects of the increased plasma concentrations of PRL in the sheep fetus during a summer gestation are unknown, although there are reports of seasonal differences in fetal growth profiles in sheep that are independent of differences in maternal nutrition (29). In summary, we have demonstrated that expression of PRLR1 and PRLR2 mRNA levels in fetal liver are reduced after disconnection of the fetal hypothalamus and pituitary and are restored after cortisol replacement. In contrast to cortisol, there was no relationship in HPD or intact fetal sheep between circulating PRL concentrations and hepatic PRLR expression. In addition, we have also shown that PRLR1 and PRLR2 mRNA expression in fetal sheep liver are not regulated by changes in external photoperiod. These studies provide further insight into the potential role that the PRL axis may play in the transduction of signals about the external environment to the developing fetus as it prepares for the transition to extrauterine life. Acknowledgments We thank Dr. Ross Young of the Physiology Department, Monash University (Clayton, Victoria), for his expert surgical skills. We also thank Anne Jurisevic for surgical assistance and for performing the hormone RIAs.

6. 7.

8. 9. 10. 11. 12.

13. 14. 15. 16.

17.

18. 19. 20. 21.

22.

23. 24.

References 1. Gluckman PD, Grumbach MM, Kaplan SL 1981 The neuroendocrine regulation and function of growth hormone and prolactin in the mammalian fetus. Endocr Rev 2:363–395 2. McMillen IC, Jenkin G, Thorburn GD, Robinson JS 1983 Concentrations of prolactin in the plasma of the fetal sheep and in amniotic fluid in late gestation and during dexamethasone induced parturition. J Endocrinol 99:107–114 3. Phillips ID, Fielke SL, Young IR, McMillen IC 1996 The relative roles of the hypothalamus and cortisol in the control of prolactin gene expression in the anterior pituitary of the sheep fetus. J Neuroendocrinol 8:929 –933 4. Houghton DC, Young IR, McMillen IC 1995 Response of prolactin to different photoperiods after surgical disconnection of the hypothalamus and pituitary in sheep fetuses. J Reprod Fertil Dev 104:199 –206 5. Bassett JM, Bomford J, Mott JC 1988 Photoperiod: an important regulator of

25.

26. 27. 28. 29.

1971

plasma prolactin concentrations in fetal lambs during later gestation. Q J Exp Physiol 73:241–244 Serron-Ferre M, Vergara M, Parraguez VH, Riquelme R, Llanos AJ 1989 The circadian variation of prolactin in fetal sheep is affected by the seasons. Endocrinology 125:1613–1616 Royster M, Driscoll P, Kelly PA, Freemark M 1995 The prolactin receptor in the fetal rat: cellular localisation of messenger ribonucleic acid, immunoreactive protein, and ligand-binding activity and induction of expression in late gestation. Endocrinology 136:3892–3900 Brown-Borg HM, Zhang FP, Huhtaniemi I, Bartke A 1996 Developmental aspects of prolactin receptor gene expression in fetal and neonatal mice. Eur J Endocrinol 134:751–757 Freemark M, Driscoll P, Maaskant R, Petryk A, Kelly PA 1997 Ontogenesis of prolactin receptors in the human fetus in early gestation: implications for tissue differentiation and development. J Clin Invest 99:1107–1117 Phillips ID, Anthony RV, Butler TG, Ross JT, McMillen IC 1997 Hepatic prolactin receptor gene expression increases in the sheep fetus before birth and after cortisol infusion. Endocrinology 138:1351–1354 Anthony RV, Smith GW, Duong A, Pratt SL, Smith MF 1995 Two forms of prolactin receptor messenger ribonucleic acid are present in ovine fetal liver and adult ovary. Endocrine 3:291–295 Bignon C, Binart N, Ormandy C, Schuler LA, Kelly PA, Djiane J 1997 Long and short forms of the ovine prolactin receptor: cDNA cloning and genomic analysis reveal that the two forms arise by different alternative splicing mechanisms in ruminants and in rodents. J Mol Endocrinol 19:109 –120 Berlanga JJ, Garcia-Ruiz JP, Perrot-Applanat M, Kelly PA, Edery M 1997 The short form of the prolactin (PRL) receptor silences PRL induction of the b-casein gene promoter. Mol Endocrinol 11:1449 –1457 Posner BI, Kelly PA, Friesen HG 1975 Prolactin receptors in the rat liver: possible induction by prolactin. Science 188:57–59 Antolovich GC, Clarke IJ, McMillen IC, Perry RA, Robinson PM, Silver M, Young IR 1990 Hypothalamo-pituitary disconnection in the fetal sheep. Neuroendocrinology 51:1–10 Antolovich GC, McMillen IC, Robinson PM, Silver M, Young IR, Perry RA 1991 The effect of hypothalamo-pituitary disconnection on the functional and morphologic development of the pituitary-adrenal axis in the fetal sheep in the last third of gestation. Neuroendocrinology 54:254 –261 Ozolins IZ, Young IR, McMillen IC 1992 Surgical disconnection of the hypothalamus from the fetal pituitary abolishes the corticotrophic response to intrauterine hypoglycaemia or hypoxaemia in the sheep during late gestation. Endocrinology 130:1438 –2445 Bocking AD, McMillen IC, Harding R, Thorburn GD 1986 Effect of reduced uterine blood flow on fetal and maternal cortisol. J Dev Physiol 8:237–245 Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ 1979 Isolation of biologically active ribonucleic acid from sources enriched in ribonucleases. Biochemistry 18:5294 –5299 Jolicoeur C, Boutin JM, Okamura H, Raguet S, Djiane J, Kelly PA 1989 Multiple regulation of prolactin receptor gene expression in the rat liver. Mol Endocrinol 3:895–900 Berdusco ET, Hammond GL, Jacobs RA, Grolla A, Akagi K, Langlois D, Challis JR 1993 Glucocorticoid-induced increase in plasma corticosteroidbinding globulin levels in fetal sheep is associated with increased biosynthesis and alterations in glycosylation. Endocrinology 132:2001–2008 Jahn GA, Daniel N, Jolivet G, Belair L, Bole-Feysot C, Kelly PA, Djiane J 1997 In vivo study of prolactin (PRL) intracellular signalling during lactogenesis in the rat: JAK/STAT pathway is activated by PRL in the mammary gland but not in the liver. Biol Reprod 57:894 –900 Hu Z, Zhuang L, Dufau ML 1996 Multiple and tissue specific promoter control of gonadal and non-gonadal prolactin receptor gene expression. J Biol Chem 271:10242–10246 Silver M 1990 Prenatal maturation, the timing of birth and how it may be regulated in domestic animals. Exp Physiol 75:285–307 Li J, Owens JA, Owens PC, Saunders JC, Fowden AL, Gilmour RS 1996 The ontogeny of hepatic growth hormone receptor and insulin-like growth factor I gene expression in the sheep fetus during late gestation: developmental regulation by cortisol. Endocrinology 137:1650 –1657 Kelly PA, Djiane J, Postel-Vinay MC, Edery M 1991 The prolactin/growth hormone receptor family. Endocr Rev 12:235–251 Phillips ID, Simonetta G, Owens JA, Robinson JS, Clarke IJ, McMillen IC 1996 Placental restriction alters the functional development of the pituitaryadrenal axis in the sheep fetus during late gestation. Pediatr Res 40:861– 866 Buckley AR, Putnam C, Russell DH 1985 Prolactin is a tumor promoter in the rat liver. Life Sci 37:2569 –2575 Jenkinson CMC, Peterson SW, Mackenzie DDS, McCutcheon SN 1994 The effects of season on placental development and fetal growth in the sheep. Proc NZ Soc Anim Prod 54:227–230