Growth Hormone Pretranslationally Regulates the Sexually Dimorphic Expression of the Prolactin Receptor Gene in Rat Liver
John A. Robertson*, Lars-Arne Haldosen, Timothy J. J. Wood, Maureen K. Steed, and Jan-Ake Gustafsson Department of Medical Nutrition Huddinge University Hospital S-141 86 Huddinge, Sweden
The female-specific expression of the rat liver PRL receptor (PRL-R) gene was investigated by Northern analysis of hypophysectomized rats after two alternative human GH treatments that were to mimic either 1) the continuous female-specific or 2) the discontinuous male-specific serum GH patterns. The former (female-specific) pattern was shown to result in a dramatic increase in PRL-R mRNA in both males and females, while the latter (male-specific) pattern failed to evoke this response. A similar inductive effect in hypophysectomized females was shown after continuous administration of bovine GH and was found to constitute an approximately 60-fold increase in PRL-R mRNA levels. This effect by bovine GH, which, unlike the human isoform, is devoid of lactogenic properties, thus indicates the somatogenic origin of the signal resulting in this inductive response. These observations in conjunction with previous data obtained for other GH-regulated nonreceptor genes are interpreted to support the proposal of GH serum patterns being an early signal in a more general mechanism for pretranslational regulation of sex-specific gene expression. In contrast to GH, only a slight elevation of PRL-R mRNA was evoked by the ligand ovine PRL, while coadministration of ovine PRL with bovine GH failed to enhance the mRNA level found with bovine GH alone. The detection of previously unreported PRL-R mRNAs in liver of approximately 3.0, 3.8, and 5 kilobases in addition to the major 2.2-kilobase form was also evident after continuous GH administration. (Molecular Endocrinology 4: 1235-1239, 1990)
Comparisons of the amino acid sequences predicted from the recently available cDNA sequences for the corresponding receptors from various species have shown regions of evolutionary conservation within both the extracellular domain thought to be responsible for ligand binding and the internal domain via which an uncharacterized signal is transmitted to hormone-responsive processes (2-4). This evolutionary conservation indicates common functional domains and may again reflect divergence from an ancestral prototypic gene or exon shuffling (5) between distinct progenitor genes with common functional peptide domains within their corresponding proteins. Both receptors have been shown to constitute subfamilies of isoforms at both the protein level (6-10) and, more recently, the mRNA level. Indeed, multiple cDNAs have recently been isolated for the rat GH receptor (GH-R) (11) and the PRL receptor (PRL-R) (3, 4,12). In addition, use of the 1.6-kilobase (kb) rat liver PRL-R cDNA (to a 2.2-kb message) has detected PRLR mRNAs of different sizes (2.8, 4.0, 6.5, and 10 kb) in rabbit mammary gland (3, 4) as well as certain 0.6-, 2.2-, and 4.0-kb messages occurring in rat ovary, prostate, mammary gland, adrenal, kidney, and testis. However, the presence of only one mRNA (2.2 kb) is reported in liver (3). The PRL-R gene in rat liver is clearly expressed in a sexually dimorphic manner, with males having negligible ligand-specific binding (13-15). Feminization via estrogen treatment of male (or gonadectomized female) rats results in feminine (i.e. elevated) levels of PRL-R protein (7, 15) and mRNA (Ref. 3 and 16, and these authors, unpublished). Analysis of the sexually dimorphic expression of certain genes (for review, see Ref. 17) has been consistent with this essentially being due to the distinct and sexdependent serum GH levels (18). Indeed, mimicking the male secretory pattern via intermittent administration of GH to hypophysectomized female rats caused a complete masculinization of (male-specific) cytochrome P450 16a (P450IIC11) expression (19, 20), while continuous administration of GH to males to mimick the
INTRODUCTION
The genes for GH and PRL are currently thought to have originated from a common ancestral gene (1). 0888-8809/90/1235-1239$02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society
1235
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 8
MOL ENDO-1990 1236
female secretory pattern induced the female-specific P450 15/3 (P450I1C12) (21). In the present study we sought to investigate the female-specific expression of PRL-R in rat liver via Northern analysis of RNA from rats with different GH regimens. Data are presented to further support the proposal of GH serum patterns being a more general mechanism by which sexually dimorphic gene expression is regulated at a pretranslational level.
C\2
O C\2
CD LTD
RESULTS AND DISCUSSION
Preliminary Northern analyses with total RNA and prolonged autoradiography indicated a major effect of GH administration patterns on the levels of PRL-R message (not shown). This effect was confirmed via Northern analysis using poly(A)+ RNA and clearly shows a major effect of GH treatment on expression of the PRL-R gene in rat liver (Fig. 1). The continuous administration of human (h) GH in a female-specific pattern to hypophysectomized male or female rats resulted in a dramatic elevation in the mRNA corresponding to the rat liver isoforms of PRL-R (Fig. 1, compare lanes 1 and 2 with 4, and 5 and 6 with 8). In contradistinction, the intermittent administration of hGH (to mimick the malespecific secretory pattern) to female or male hypophysectomized rats failed to elicit this response (Fig. 1, lanes 3 and 7). Quantification of the signal intensities from this Northern blot after normalization against 0actin is shown in Fig. 2. Certain additional observations regarding these data
7
8
Fig. 1. Northern Analysis of Hepatic PRL-R mRNA after Various hGH Treatments Individual lanes represent 2.5 nQ poly(A)+ RNA (see Materials and Methods) obtained from rats with the following treatments: 1) control males, 2) hypophysectomized (HX) males, 3) HX males with intermittent hGH administration, 4) HX males with continuous hGH, 5) control females, 6) HX females, 7) HX females with intermittent hGH, and 8) HX females with continuous hGH treatment. A, The blot probed with PRL-R (exposed for 7 days); B, the same blot reprobed with 0-actin (see Materials and Methods). Intensifier screens were not used for these autoradiograms.
•i—i
CD Cti 1—4
CD
2 3 4 Fig. 2. Quantification of Northern Signal Intensities from Data in Fig. 1 Signals from male P ) and female (•) mRNA were normalized against /3-actin. Individual treatments are: 1) control, 2) hypophysectomized (HX), 3) HX plus intermittent GH, and 4) HX plus continuous GH. No signals were obtained for lane 1 (male) or lanes 2 and 3 (both males and females); see Results and Discussion.
deserve comment. Firstly, at these autoradiographic exposures, the absence of signals indicates the low levels of PRL-R message in male controls and male or female hypophysectomized animals with or without intermittent GH administration and is in accord with the low levels of PRL-R protein reported (13-15). The lower levels of PRL-R mRNA detected in females after hypophysectomy (Fig. 1, compare lanes 5 and 6) are also in agreement with ligand binding data (22-24) and would indicate that the down-regulated expression of this gene after such treatment also occurs at the pretranslational level. Second, Fig. 2 indicates a lower inductive response (~60%) by continuous GH administration to male hypophysectomized animals compared with the response of their female equivalents. Although a rationale for this is obscure, a similar sex difference in response to a continuous GH pattern has been identified for cytochrome P450 f (P450 11C7; Zaphiropoulos, P., personal communication). Finally, the level of PRL-R
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GH Regulation of PRL-R mRNA
1237
message in superfeminized females (Fig. 1, lane 8, and Fig. 2, lane 4) is almost twice that in untreated female controls, which indicates that this hepatic gene is not maximally expressed in normal females. As no major differences were detected in relative signal intensities for the individuals used in the preliminary study (not shown) compared with data subsequently obtained with RNA from pooled animals (Figs. 1 and 2), poly(A)+ RNA from individual rat livers was used to investigate if an elevation of PRL-R mRNA levels was evoked by identical continuous treatment of hypophysectomized female rats with the bovine (b) isoform of GH which, unlike the human, is devoid of lactogenic effects (25). The positive effect obtained (see Figs. 3 and 4, lanes 1 and 2) thus indicates a somatogenic origin of the signal resulting in this elevation of steady state levels of PRL-R mRNA. These figures also illustrate the barely detectable levels of PRL-R mRNA occurring in the hypophysectomized female rat liver (Fig. 3, lane 1, and see above) and that the elevation via continuous GH treatment is about 60-fold (Fig. 4, lanes 1 and 2). Up-regulation of receptor gene expression via their ligands has been reported for GH, epidermal growth factor, insulin, LH, GnRH, angiotensin-ll, and PRL (26). The data for ovine (o) PRL in Figs. 3 and 4 (compare lanes 1 and 3) may support a weak up-regulatory effect by oPRL and that this occurs, at least partly, pretranslationally. Although a weak inductive effect has also been shown with oPRL in primary hepatocytes (26), this must be interpreted with caution, as oPRL has
1 2
3
4
B Fig. 3. Northern Analysis of Hepatic PRL-R mRNA from Hypophysectomized Female Rats after Continuous Hormonal Treatments Individual lanes represent 7 ^g poly(A)+ RNA from animals (see Materials and Methods) receiving the following treatments: 1) control, 2) bGH, 3) oPRL, and 4) bGH plus oPRL. A, The blot probed with PRL-R (exposed for 2 days); B, the same blot reprobed with /3-actin (see Materials and Methods).
Fig. 4. Quantification of Signal Intensities from Data Corresponding to Fig. 3 Normalization was against /?-actin (see Materials and Methods). Lane numbers correspond to those in Fig. 3. Intensifier screens were not used for these autoradingrams.
been shown to have somatogenic in addition to its classically lactogenic properties (27). In view of evidence for the somatogenic origin of this inductive signal (see above), it seems reasonable to consider the weak inductive effect of oPRL as being a manifestation of its somatogenic properties. It is in this context, therefore, that the noninductive effect of rat PRL (nonsomatogenic) in normal and hypophysectomized male and female rats (24, 28) might provide a more accurate indication of the inductive capacity of this ligand. The coadministration, however, of oPRL and bGH did not appear to enhance PRL-R mRNA levels compared to those found with bGH alone (see Figs. 3 and 4, lanes 2 and 4) and thus does not support the almost doubled specific binding previously reported (26). This discrepancy might implicate posttranslational events that would not be detected by the present analysis. In addition to detection of the 2.2-kb hepatic isoform previously reported (3), the pres^ ice of larger PRL-R mRNA homologs (provisionally 3.0, 3.8, and 5 kb) was apparent after continuous GH treatment (see Fig. 3, lanes 2 and 4). The presence of these low abundance messages in liver has not previously been re^ These in conjunction with the tissue-specific mRNAs
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 8
MOL ENDO-1990 1238
previously reported (see above) may reflect products from several distinct PRL-R genes or, alternatively, may result from the differential (and tissue-specific) premRNA splicing of a single gene for this receptor. These data are, therefore, in support of previous studies concerning the female-specific GH up-regulation of cytochrome P450 15/3 mRNA (21) and 4-androstene-3,17-dione 5a-reductase enzyme activity (29) as well as the opposite, though complimentary, effect of rat urinary a2-microglobulin (30) or cytochrome P450 16a (21), where constant GH levels extinguished the production of this male-specific protein or mRNA. During the course of this study, a decreased expression of the male-specific cytochrome P450 g (P450 11C13) gene via continuous GH administration was also reported (31). Accordingly, as this receptor and these five distinct (and nonreceptor) genes have now been shown to be similar in their sensitivity to the female-specific pattern of serum GH, the possibility is further strengthened that this pattern is a more general mechanism for regulating female-specific gene expression and occurs at least in part at the pretranslational level.
MATERIALS AND METHODS Materials [a-32P]dCTP (3000 Ci/mmol), [a-35S]dATP (1000 Ci/mmol), and Multiprime kits were obtained from Amersham (Arlington Heights, IL). Polymerase chain reaction components were purchased from Cetus/Perkin Elmer (Emeryville, CA), Nytran hybridization membranes from Schleicher and Schuell (Keene, NH), and oligo(dT)-cellulose (type 7) from Pharmacia (Piscataway, NJ). Agaroses were obtained from FMC BioProducts (Rockland, ME), cloning vectors and restriction enzymes from Promega (Madison, Wl), T7 DNA polymerase (Sequenase)from U.S. Biochemicals (Cleveland, OH), and poly(A)* RNA size standards from Bethesda Research Laboratories (Gaithersburg, MD). Osmotic minipumps were obtained from Alza Corp. (Palo Alto, CA). Biosynthetic methionyl-somatotropin (hGH, Somatonorm) was a generous gift of AB Kabi (Stockholm, Sweden), and oPRL (NIDDK oPRL 16) and bGH (NIH GH B-14) were obtained from the Hormone Distribution Program, NIDDK, NIH (Bethesda, MD). Animals Male and female Sprague-Dawley rats, hypophysectomized at 42 days of age, and age-matched controls were purchased from Mollegaards Avelslaboratorium (Skensved, Denmark). The completeness of hypophysectomy was determined by monitoring weight gain for 1 week before the initiation of hormone treatment. The animals were maintained under a standard light regimen (0600-1700 h). Hormone Treatments Human GH was administered to hypophysectomized rats either at daily doses of 0.36 fig hormone/g w t d a y (61 /xg/day for females or 70 ^g/day for males) by sc injections every 24 h for 7 days or at the same daily dose by continuous infusion via minipumps for 7 days. Bovine GH and/or oPRL (each at 0.73 fig/g wtday) was administered by minipumps for a total duration of 7 days.
PRL-R cDNA Probe The 5'-most 290 basepairs of the rat liver cDNA sequence (3) were synthesized via the polymerase chain reaction (32) using the two opposed primers: 5' GGT CAA GCG AGC TGG ATT CTA 3' and 5' ATC TCA GGT TTC CCT GGT GGT 3'. This fragment shows no significant homology to the nucleotide sequence reported for rat GH-R (11). The polymerase chain reaction-generated product was excized from a 5% Nuseive GTG agarose gel, subcloned at the Sma\ site in pGEM-7, and introduced into the E. coli strain JM109 via electroporation (33), and authenticity was confirmed by sequencing. For use as a PRL-R-specific radiolabeled probe, this was liberated with EcoRI and AY/ndlll, isolated via the gel system described above, and "in gel" (34) radiolabeled via random priming (35) to a specific activity in excess of 109 cpm/iig DNA. Northern Analysis Preliminary studies (data not shown) with total RNA isolated from individual rat livers and their Northern analysis (see below) used equivalent fluorescent intensities of ribosomal RNA bands per gel lane before transfer to indicate similar amounts of mRNA loaded and to approximate normalization. For subsequent initial studies with hGH, total cellular RNA from rat livers (three animals per group) was isolated as previously described (36), and the poly(A)+ RNA-enriched fraction was prepared by standard methods (37) from bulked and equivalent amounts of total RNA from each animal per group. The corresponding poly(A)+ fraction was electrophoresed in 0.66 M formaldehyde and 1.2% Seakem GTG agarose containing 0.5 fig ethidium bromide/ml and transferred to Nytran by capillary blotting. Hybridizations were performed at high stringency (50% formamide at 42 C), and washes were performed twice in 2 x SSC (20 x SSC = 3 M NaCI, 0.3 M sodium citrate, pH 7.0)-1% sodium dodecyl sulfate (SDS) at 65 C, followed by twice in 0.1% SSC-0.1% SDS at room temperature. Intensifier screens were used unless otherwise indicated. PRL-R signals were normalized against /3-actin by reprobing with /3-actin cDNA after removal of the PRL-R probe in 0.1 % SSC-1 % SDS after removal of the PRL-R probe in 0.1 % SSC-1 % SDS (slow cooling to 60 C from boiling) and confirmation of removed signals by autoradiography before reuse. Northern signals were quantified using a model 301 densitometer (X-rite, Inc., Grandville, Ml). Subsequent studies with oPRL and bGH used poly(A)+ RNA from individual animals for Northern analysis by the method described above. Acknowledgments Received November 16,1989. Revision received May 7,1990. Accepted May 11,1990. Address requests for reprints to: Dr. John Robertson, Department of Medical Nutrition, F60, NOVUM Huddinge University Hospital, S-141 86 Huddinge, Sweden. This work was supported by a grant from the Swedish Medical Research Council (B89-03X-06807-06C). * Recipient of a grant from the Swedish Medical Research Council.
REFERENCES
Nicoll CS, Baldocchi RA 1988 Evolutionary origin of prolactins, growth hormones and placental lactogens. In: Imura H, Shizume K, Yoshida S (eds) Progress in Endocrinology 1988. Elsevier, Amsterdam, pp 1379-1384 Leung DW, Spencer SA, Cachianes G, Hammonds RG, Collins C, Henzel WJ, Barnard R, Waters MJ, Wood Wl 1987 Growth hormone receptor and serum binding pro-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GH Regulation of PRL-R mRNA
3.
4.
5. 6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
tein: purification, cloning and expression. Nature 330:537-543 Boutin J-M, Jolicoeur C, Okamura H, Gagnon J, Edery M, Shirota M, Banville D, Dusanter-Fourt I, Djiane J, Kelly PA 1988 Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family. Cell 53:69-77 Edery M, Jolicoeur C, Levi-Meyrueis C, Dusanter-Fourt I, Petridou B, Boutin J-M, Lesueur L, Kelly PA, Djiane J 1989 Identification and sequence analysis of a second form of prolactin receptor by molecular cloning of complementary DNA from rabbit mammary gland. Proc Natl Acad Sci USA 86:2112-2116 Rogers J 1985 Exon shuffling and intron insertion in serine protease genes. Nature 315:458-459 Gorin E, Goodman HM 1984 Covalent binding of growth hormone to surface receptors on rat adipocytes. Endocrinology 114:1279-1286 Haldosen L-A, Gustafsson J-A 1987 Characterization of hepatic lactogen receptor. Subunit composition and hydrodynamic properties. J Biol Chem 262:7404-7411 Smith WC, Talamantes F 1987 Identification and characterization of a heterogeneous population of growth hormone receptors in mouse membranes. J Biol Chem 262:2213-2219 Yamada K, Lipson KE, Donner DB 1987 Structure and proteolysis of the growth hormone receptor on rat hepatocytes. Biochemistry 26:4438-4443 Husman B, Haldosen L-A, Andersson G, Gustafsson J-A 1988 Characterization of the somatogenic receptor in rat liver. Hydrodynamic properties and affinity cross-linking. J Biol Chem 263:3963-3970 Mathews LS, Enberg B, Norstedt G 1989 Regulation of rat growth hormone receptor gene expression. J Biol Chem 264:9905-9910 Boutin J-M, Edery M, Shirota M, Jolicoeur C, Lesueur L, Ali S, Gould D, Djiane J, Kelly PA 1989 Identification of a cDNA encoding a long form of prolactin receptor in human hepatoma and breast cancer cells. Mol Endocrinol 3:1455-1461 Kelly PA, Posner Bl, Tsushima T, Friesen HG 1974 Studies of insulin, growth hormone and prolactin binding: ontogenesis, effects of sex and pregnancy. Endocrinology 95:532-539 Norstedt G, Eneroth P, Gustafsson J-A, Hokfelt T, Skett P 1980 Hypothalamo-pituitary regulation of hepatic prolactin receptors in the rat. Brain Res 192:77-88 Tsim KWK, Pak RCK, Cheng CHK1985 Prolactin receptor in rat liver: sex difference in estrogenic stimulation and imprinting of the responsiveness to estrogen by neonatal androgen in male rats. Mol Cell Endocrinol 40:99-105 Jolicoeur C, Boutin J-M, Okamura H, Raguet S, Djiane J, Kelly PA 1989 Multiple regulation of prolactin receptor gene expression in rat liver. Mol Endocrinol 3:895-900 Zaphiropoulos P, Mode A, Norstedt G, Gustafsson J-A 1989 Regulation of sexual differentiation in drug and steroid metabolism. Trends Pharm Sci 10:149-153 Eden S 1979 Age and sex-related differences in episodic growth hormone secretion in the rat. Endocrinology 105:555-560 Mode A, Wiersma-Larsson E, Gustafsson J-A 1989a Transcriptional and posttranscriptional regulation of sexually differentiated rat liver cytochrome P-450 by growth hormone. Mol Endocrinol 3:1142-1147
1239
20. Mode A, Wiersma-Larsson E, Strom A, Zaphiropoulos P, Gustafsson, J-A 1989b A dual regulation of growth hormone as a feminizing factor in the control of sex-specific cytochrome P-450 isozymes in rat liver. J Endocrinol 120:311-317 21. MacGeoch C, Morgan ET, Cordell B, Gustafsson J-A 1987 Growth hormone regulates expression of the rat liver cytochrome P-450 15/3 at a pretranslational level. Biochem Biophys Res Commun 143:782-788 22. Posner Bl, Kelly PA, Friesen HG 1974 Induction of a lactogenic receptor in rat liver: influence of estrogen and the pituitary. Proc Natl Acad Sci USA 71:2407-2410 23. Posner Bl, Kelly PA, Friesen HG 1975 Prolactin receptors in rat liver: possible induction by prolactin. Science 188:57-59 24. Baxter RC, Zaltsman Z 1984 Induction of hepatic receptors for growth hormone (GH) and prolactin by GH infusion is sex independent. Endocrinology 115:2009-2014 25. Ranke MB, Stanley CA, Tenore A, Rodbard D, Bongiovanni AM, Parks JS 1976 Characterization of somatogenic and lactogenic sites in isolated rat hepatocytes. Endocrinology 99:1033-1045 26. Barash I, Cromlish W, Posner Bl 1988 Prolactin (PRL) receptor induction in cultured rat hepatocytes: dual regulation by prolactin and growth hormone. Endocrinolgy 122:1151-1158 27. Murphy LJ, Tachibana K, Friesen HG 1988 Stimulation of hepatic insulin-like growth factor-l gene expression by ovine prolactin: evidence for intrinsic somatogenic activity in the rat. Endocrinology 122:2027-2033 28. Baxter RC, Zaltsman Z, Turtle JR 1984 Rat growth hormone (GH) but not prolactin (PRL) induces both GH and PRL receptors in female rat liver. Endocrinology 114:1893-1901 29. Mode A, Norstedt G, Eneroth P, Gustafsson J-A 1983 Purification of liver feminizing factor from rat pituitaries and demonstration of its identity with growth hormone. Endocrinology 113:1250-1260 30. Husman B, Norstedt G, Mode A, Gustafsson J-A 1985 The mode of growth hormone administration is of major importance for the excretion of the major male rat urinary proteins. Mol Cell Endocrinol 40:205-?10 31. Zaphiropoulos P, Strom A, Robertson JA, Gustafsson JA 1990 Structural and regulatory analysis of the malespecific rat liver cytochrome P450g: repression by continuous growth hormone administration. Mol Endocrinol 4:53-58 32. Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, Arnheim N 1985 Enzymic amplification of /3-globin genomic sequence and restriction site analysis for diagnosis of sickle cell anaemia. Science 230:1350-1353 33. Dower WJ, Miller JF, Ragsdale CW 1988 High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145 34. Struhl K 1985 A rapid method for creating recombinant DNA molecules. Biotechniques 3:452-453 35. Feinberg AP, Vogelstein B 1983 A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13 36. Chomczynski P, Saachi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Anal Biochem 162:156-159 37. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning-A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
John A. Robertson*, Lars-Arne Haldosen, Timothy J. J. Wood, Maureen K. Steed, and Jan-Ake Gustafsson Department of Medical Nutrition Huddinge University Hospital S-141 86 Huddinge, Sweden
The female-specific expression of the rat liver PRL receptor (PRL-R) gene was investigated by Northern analysis of hypophysectomized rats after two alternative human GH treatments that were to mimic either 1) the continuous female-specific or 2) the discontinuous male-specific serum GH patterns. The former (female-specific) pattern was shown to result in a dramatic increase in PRL-R mRNA in both males and females, while the latter (male-specific) pattern failed to evoke this response. A similar inductive effect in hypophysectomized females was shown after continuous administration of bovine GH and was found to constitute an approximately 60-fold increase in PRL-R mRNA levels. This effect by bovine GH, which, unlike the human isoform, is devoid of lactogenic properties, thus indicates the somatogenic origin of the signal resulting in this inductive response. These observations in conjunction with previous data obtained for other GH-regulated nonreceptor genes are interpreted to support the proposal of GH serum patterns being an early signal in a more general mechanism for pretranslational regulation of sex-specific gene expression. In contrast to GH, only a slight elevation of PRL-R mRNA was evoked by the ligand ovine PRL, while coadministration of ovine PRL with bovine GH failed to enhance the mRNA level found with bovine GH alone. The detection of previously unreported PRL-R mRNAs in liver of approximately 3.0, 3.8, and 5 kilobases in addition to the major 2.2-kilobase form was also evident after continuous GH administration. (Molecular Endocrinology 4: 1235-1239, 1990)
Comparisons of the amino acid sequences predicted from the recently available cDNA sequences for the corresponding receptors from various species have shown regions of evolutionary conservation within both the extracellular domain thought to be responsible for ligand binding and the internal domain via which an uncharacterized signal is transmitted to hormone-responsive processes (2-4). This evolutionary conservation indicates common functional domains and may again reflect divergence from an ancestral prototypic gene or exon shuffling (5) between distinct progenitor genes with common functional peptide domains within their corresponding proteins. Both receptors have been shown to constitute subfamilies of isoforms at both the protein level (6-10) and, more recently, the mRNA level. Indeed, multiple cDNAs have recently been isolated for the rat GH receptor (GH-R) (11) and the PRL receptor (PRL-R) (3, 4,12). In addition, use of the 1.6-kilobase (kb) rat liver PRL-R cDNA (to a 2.2-kb message) has detected PRLR mRNAs of different sizes (2.8, 4.0, 6.5, and 10 kb) in rabbit mammary gland (3, 4) as well as certain 0.6-, 2.2-, and 4.0-kb messages occurring in rat ovary, prostate, mammary gland, adrenal, kidney, and testis. However, the presence of only one mRNA (2.2 kb) is reported in liver (3). The PRL-R gene in rat liver is clearly expressed in a sexually dimorphic manner, with males having negligible ligand-specific binding (13-15). Feminization via estrogen treatment of male (or gonadectomized female) rats results in feminine (i.e. elevated) levels of PRL-R protein (7, 15) and mRNA (Ref. 3 and 16, and these authors, unpublished). Analysis of the sexually dimorphic expression of certain genes (for review, see Ref. 17) has been consistent with this essentially being due to the distinct and sexdependent serum GH levels (18). Indeed, mimicking the male secretory pattern via intermittent administration of GH to hypophysectomized female rats caused a complete masculinization of (male-specific) cytochrome P450 16a (P450IIC11) expression (19, 20), while continuous administration of GH to males to mimick the
INTRODUCTION
The genes for GH and PRL are currently thought to have originated from a common ancestral gene (1). 0888-8809/90/1235-1239$02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society
1235
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 8
MOL ENDO-1990 1236
female secretory pattern induced the female-specific P450 15/3 (P450I1C12) (21). In the present study we sought to investigate the female-specific expression of PRL-R in rat liver via Northern analysis of RNA from rats with different GH regimens. Data are presented to further support the proposal of GH serum patterns being a more general mechanism by which sexually dimorphic gene expression is regulated at a pretranslational level.
C\2
O C\2
CD LTD
RESULTS AND DISCUSSION
Preliminary Northern analyses with total RNA and prolonged autoradiography indicated a major effect of GH administration patterns on the levels of PRL-R message (not shown). This effect was confirmed via Northern analysis using poly(A)+ RNA and clearly shows a major effect of GH treatment on expression of the PRL-R gene in rat liver (Fig. 1). The continuous administration of human (h) GH in a female-specific pattern to hypophysectomized male or female rats resulted in a dramatic elevation in the mRNA corresponding to the rat liver isoforms of PRL-R (Fig. 1, compare lanes 1 and 2 with 4, and 5 and 6 with 8). In contradistinction, the intermittent administration of hGH (to mimick the malespecific secretory pattern) to female or male hypophysectomized rats failed to elicit this response (Fig. 1, lanes 3 and 7). Quantification of the signal intensities from this Northern blot after normalization against 0actin is shown in Fig. 2. Certain additional observations regarding these data
7
8
Fig. 1. Northern Analysis of Hepatic PRL-R mRNA after Various hGH Treatments Individual lanes represent 2.5 nQ poly(A)+ RNA (see Materials and Methods) obtained from rats with the following treatments: 1) control males, 2) hypophysectomized (HX) males, 3) HX males with intermittent hGH administration, 4) HX males with continuous hGH, 5) control females, 6) HX females, 7) HX females with intermittent hGH, and 8) HX females with continuous hGH treatment. A, The blot probed with PRL-R (exposed for 7 days); B, the same blot reprobed with 0-actin (see Materials and Methods). Intensifier screens were not used for these autoradiograms.
•i—i
CD Cti 1—4
CD
2 3 4 Fig. 2. Quantification of Northern Signal Intensities from Data in Fig. 1 Signals from male P ) and female (•) mRNA were normalized against /3-actin. Individual treatments are: 1) control, 2) hypophysectomized (HX), 3) HX plus intermittent GH, and 4) HX plus continuous GH. No signals were obtained for lane 1 (male) or lanes 2 and 3 (both males and females); see Results and Discussion.
deserve comment. Firstly, at these autoradiographic exposures, the absence of signals indicates the low levels of PRL-R message in male controls and male or female hypophysectomized animals with or without intermittent GH administration and is in accord with the low levels of PRL-R protein reported (13-15). The lower levels of PRL-R mRNA detected in females after hypophysectomy (Fig. 1, compare lanes 5 and 6) are also in agreement with ligand binding data (22-24) and would indicate that the down-regulated expression of this gene after such treatment also occurs at the pretranslational level. Second, Fig. 2 indicates a lower inductive response (~60%) by continuous GH administration to male hypophysectomized animals compared with the response of their female equivalents. Although a rationale for this is obscure, a similar sex difference in response to a continuous GH pattern has been identified for cytochrome P450 f (P450 11C7; Zaphiropoulos, P., personal communication). Finally, the level of PRL-R
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GH Regulation of PRL-R mRNA
1237
message in superfeminized females (Fig. 1, lane 8, and Fig. 2, lane 4) is almost twice that in untreated female controls, which indicates that this hepatic gene is not maximally expressed in normal females. As no major differences were detected in relative signal intensities for the individuals used in the preliminary study (not shown) compared with data subsequently obtained with RNA from pooled animals (Figs. 1 and 2), poly(A)+ RNA from individual rat livers was used to investigate if an elevation of PRL-R mRNA levels was evoked by identical continuous treatment of hypophysectomized female rats with the bovine (b) isoform of GH which, unlike the human, is devoid of lactogenic effects (25). The positive effect obtained (see Figs. 3 and 4, lanes 1 and 2) thus indicates a somatogenic origin of the signal resulting in this elevation of steady state levels of PRL-R mRNA. These figures also illustrate the barely detectable levels of PRL-R mRNA occurring in the hypophysectomized female rat liver (Fig. 3, lane 1, and see above) and that the elevation via continuous GH treatment is about 60-fold (Fig. 4, lanes 1 and 2). Up-regulation of receptor gene expression via their ligands has been reported for GH, epidermal growth factor, insulin, LH, GnRH, angiotensin-ll, and PRL (26). The data for ovine (o) PRL in Figs. 3 and 4 (compare lanes 1 and 3) may support a weak up-regulatory effect by oPRL and that this occurs, at least partly, pretranslationally. Although a weak inductive effect has also been shown with oPRL in primary hepatocytes (26), this must be interpreted with caution, as oPRL has
1 2
3
4
B Fig. 3. Northern Analysis of Hepatic PRL-R mRNA from Hypophysectomized Female Rats after Continuous Hormonal Treatments Individual lanes represent 7 ^g poly(A)+ RNA from animals (see Materials and Methods) receiving the following treatments: 1) control, 2) bGH, 3) oPRL, and 4) bGH plus oPRL. A, The blot probed with PRL-R (exposed for 2 days); B, the same blot reprobed with /3-actin (see Materials and Methods).
Fig. 4. Quantification of Signal Intensities from Data Corresponding to Fig. 3 Normalization was against /?-actin (see Materials and Methods). Lane numbers correspond to those in Fig. 3. Intensifier screens were not used for these autoradingrams.
been shown to have somatogenic in addition to its classically lactogenic properties (27). In view of evidence for the somatogenic origin of this inductive signal (see above), it seems reasonable to consider the weak inductive effect of oPRL as being a manifestation of its somatogenic properties. It is in this context, therefore, that the noninductive effect of rat PRL (nonsomatogenic) in normal and hypophysectomized male and female rats (24, 28) might provide a more accurate indication of the inductive capacity of this ligand. The coadministration, however, of oPRL and bGH did not appear to enhance PRL-R mRNA levels compared to those found with bGH alone (see Figs. 3 and 4, lanes 2 and 4) and thus does not support the almost doubled specific binding previously reported (26). This discrepancy might implicate posttranslational events that would not be detected by the present analysis. In addition to detection of the 2.2-kb hepatic isoform previously reported (3), the pres^ ice of larger PRL-R mRNA homologs (provisionally 3.0, 3.8, and 5 kb) was apparent after continuous GH treatment (see Fig. 3, lanes 2 and 4). The presence of these low abundance messages in liver has not previously been re^ These in conjunction with the tissue-specific mRNAs
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 8
MOL ENDO-1990 1238
previously reported (see above) may reflect products from several distinct PRL-R genes or, alternatively, may result from the differential (and tissue-specific) premRNA splicing of a single gene for this receptor. These data are, therefore, in support of previous studies concerning the female-specific GH up-regulation of cytochrome P450 15/3 mRNA (21) and 4-androstene-3,17-dione 5a-reductase enzyme activity (29) as well as the opposite, though complimentary, effect of rat urinary a2-microglobulin (30) or cytochrome P450 16a (21), where constant GH levels extinguished the production of this male-specific protein or mRNA. During the course of this study, a decreased expression of the male-specific cytochrome P450 g (P450 11C13) gene via continuous GH administration was also reported (31). Accordingly, as this receptor and these five distinct (and nonreceptor) genes have now been shown to be similar in their sensitivity to the female-specific pattern of serum GH, the possibility is further strengthened that this pattern is a more general mechanism for regulating female-specific gene expression and occurs at least in part at the pretranslational level.
MATERIALS AND METHODS Materials [a-32P]dCTP (3000 Ci/mmol), [a-35S]dATP (1000 Ci/mmol), and Multiprime kits were obtained from Amersham (Arlington Heights, IL). Polymerase chain reaction components were purchased from Cetus/Perkin Elmer (Emeryville, CA), Nytran hybridization membranes from Schleicher and Schuell (Keene, NH), and oligo(dT)-cellulose (type 7) from Pharmacia (Piscataway, NJ). Agaroses were obtained from FMC BioProducts (Rockland, ME), cloning vectors and restriction enzymes from Promega (Madison, Wl), T7 DNA polymerase (Sequenase)from U.S. Biochemicals (Cleveland, OH), and poly(A)* RNA size standards from Bethesda Research Laboratories (Gaithersburg, MD). Osmotic minipumps were obtained from Alza Corp. (Palo Alto, CA). Biosynthetic methionyl-somatotropin (hGH, Somatonorm) was a generous gift of AB Kabi (Stockholm, Sweden), and oPRL (NIDDK oPRL 16) and bGH (NIH GH B-14) were obtained from the Hormone Distribution Program, NIDDK, NIH (Bethesda, MD). Animals Male and female Sprague-Dawley rats, hypophysectomized at 42 days of age, and age-matched controls were purchased from Mollegaards Avelslaboratorium (Skensved, Denmark). The completeness of hypophysectomy was determined by monitoring weight gain for 1 week before the initiation of hormone treatment. The animals were maintained under a standard light regimen (0600-1700 h). Hormone Treatments Human GH was administered to hypophysectomized rats either at daily doses of 0.36 fig hormone/g w t d a y (61 /xg/day for females or 70 ^g/day for males) by sc injections every 24 h for 7 days or at the same daily dose by continuous infusion via minipumps for 7 days. Bovine GH and/or oPRL (each at 0.73 fig/g wtday) was administered by minipumps for a total duration of 7 days.
PRL-R cDNA Probe The 5'-most 290 basepairs of the rat liver cDNA sequence (3) were synthesized via the polymerase chain reaction (32) using the two opposed primers: 5' GGT CAA GCG AGC TGG ATT CTA 3' and 5' ATC TCA GGT TTC CCT GGT GGT 3'. This fragment shows no significant homology to the nucleotide sequence reported for rat GH-R (11). The polymerase chain reaction-generated product was excized from a 5% Nuseive GTG agarose gel, subcloned at the Sma\ site in pGEM-7, and introduced into the E. coli strain JM109 via electroporation (33), and authenticity was confirmed by sequencing. For use as a PRL-R-specific radiolabeled probe, this was liberated with EcoRI and AY/ndlll, isolated via the gel system described above, and "in gel" (34) radiolabeled via random priming (35) to a specific activity in excess of 109 cpm/iig DNA. Northern Analysis Preliminary studies (data not shown) with total RNA isolated from individual rat livers and their Northern analysis (see below) used equivalent fluorescent intensities of ribosomal RNA bands per gel lane before transfer to indicate similar amounts of mRNA loaded and to approximate normalization. For subsequent initial studies with hGH, total cellular RNA from rat livers (three animals per group) was isolated as previously described (36), and the poly(A)+ RNA-enriched fraction was prepared by standard methods (37) from bulked and equivalent amounts of total RNA from each animal per group. The corresponding poly(A)+ fraction was electrophoresed in 0.66 M formaldehyde and 1.2% Seakem GTG agarose containing 0.5 fig ethidium bromide/ml and transferred to Nytran by capillary blotting. Hybridizations were performed at high stringency (50% formamide at 42 C), and washes were performed twice in 2 x SSC (20 x SSC = 3 M NaCI, 0.3 M sodium citrate, pH 7.0)-1% sodium dodecyl sulfate (SDS) at 65 C, followed by twice in 0.1% SSC-0.1% SDS at room temperature. Intensifier screens were used unless otherwise indicated. PRL-R signals were normalized against /3-actin by reprobing with /3-actin cDNA after removal of the PRL-R probe in 0.1 % SSC-1 % SDS after removal of the PRL-R probe in 0.1 % SSC-1 % SDS (slow cooling to 60 C from boiling) and confirmation of removed signals by autoradiography before reuse. Northern signals were quantified using a model 301 densitometer (X-rite, Inc., Grandville, Ml). Subsequent studies with oPRL and bGH used poly(A)+ RNA from individual animals for Northern analysis by the method described above. Acknowledgments Received November 16,1989. Revision received May 7,1990. Accepted May 11,1990. Address requests for reprints to: Dr. John Robertson, Department of Medical Nutrition, F60, NOVUM Huddinge University Hospital, S-141 86 Huddinge, Sweden. This work was supported by a grant from the Swedish Medical Research Council (B89-03X-06807-06C). * Recipient of a grant from the Swedish Medical Research Council.
REFERENCES
Nicoll CS, Baldocchi RA 1988 Evolutionary origin of prolactins, growth hormones and placental lactogens. In: Imura H, Shizume K, Yoshida S (eds) Progress in Endocrinology 1988. Elsevier, Amsterdam, pp 1379-1384 Leung DW, Spencer SA, Cachianes G, Hammonds RG, Collins C, Henzel WJ, Barnard R, Waters MJ, Wood Wl 1987 Growth hormone receptor and serum binding pro-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GH Regulation of PRL-R mRNA
3.
4.
5. 6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
tein: purification, cloning and expression. Nature 330:537-543 Boutin J-M, Jolicoeur C, Okamura H, Gagnon J, Edery M, Shirota M, Banville D, Dusanter-Fourt I, Djiane J, Kelly PA 1988 Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family. Cell 53:69-77 Edery M, Jolicoeur C, Levi-Meyrueis C, Dusanter-Fourt I, Petridou B, Boutin J-M, Lesueur L, Kelly PA, Djiane J 1989 Identification and sequence analysis of a second form of prolactin receptor by molecular cloning of complementary DNA from rabbit mammary gland. Proc Natl Acad Sci USA 86:2112-2116 Rogers J 1985 Exon shuffling and intron insertion in serine protease genes. Nature 315:458-459 Gorin E, Goodman HM 1984 Covalent binding of growth hormone to surface receptors on rat adipocytes. Endocrinology 114:1279-1286 Haldosen L-A, Gustafsson J-A 1987 Characterization of hepatic lactogen receptor. Subunit composition and hydrodynamic properties. J Biol Chem 262:7404-7411 Smith WC, Talamantes F 1987 Identification and characterization of a heterogeneous population of growth hormone receptors in mouse membranes. J Biol Chem 262:2213-2219 Yamada K, Lipson KE, Donner DB 1987 Structure and proteolysis of the growth hormone receptor on rat hepatocytes. Biochemistry 26:4438-4443 Husman B, Haldosen L-A, Andersson G, Gustafsson J-A 1988 Characterization of the somatogenic receptor in rat liver. Hydrodynamic properties and affinity cross-linking. J Biol Chem 263:3963-3970 Mathews LS, Enberg B, Norstedt G 1989 Regulation of rat growth hormone receptor gene expression. J Biol Chem 264:9905-9910 Boutin J-M, Edery M, Shirota M, Jolicoeur C, Lesueur L, Ali S, Gould D, Djiane J, Kelly PA 1989 Identification of a cDNA encoding a long form of prolactin receptor in human hepatoma and breast cancer cells. Mol Endocrinol 3:1455-1461 Kelly PA, Posner Bl, Tsushima T, Friesen HG 1974 Studies of insulin, growth hormone and prolactin binding: ontogenesis, effects of sex and pregnancy. Endocrinology 95:532-539 Norstedt G, Eneroth P, Gustafsson J-A, Hokfelt T, Skett P 1980 Hypothalamo-pituitary regulation of hepatic prolactin receptors in the rat. Brain Res 192:77-88 Tsim KWK, Pak RCK, Cheng CHK1985 Prolactin receptor in rat liver: sex difference in estrogenic stimulation and imprinting of the responsiveness to estrogen by neonatal androgen in male rats. Mol Cell Endocrinol 40:99-105 Jolicoeur C, Boutin J-M, Okamura H, Raguet S, Djiane J, Kelly PA 1989 Multiple regulation of prolactin receptor gene expression in rat liver. Mol Endocrinol 3:895-900 Zaphiropoulos P, Mode A, Norstedt G, Gustafsson J-A 1989 Regulation of sexual differentiation in drug and steroid metabolism. Trends Pharm Sci 10:149-153 Eden S 1979 Age and sex-related differences in episodic growth hormone secretion in the rat. Endocrinology 105:555-560 Mode A, Wiersma-Larsson E, Gustafsson J-A 1989a Transcriptional and posttranscriptional regulation of sexually differentiated rat liver cytochrome P-450 by growth hormone. Mol Endocrinol 3:1142-1147
1239
20. Mode A, Wiersma-Larsson E, Strom A, Zaphiropoulos P, Gustafsson, J-A 1989b A dual regulation of growth hormone as a feminizing factor in the control of sex-specific cytochrome P-450 isozymes in rat liver. J Endocrinol 120:311-317 21. MacGeoch C, Morgan ET, Cordell B, Gustafsson J-A 1987 Growth hormone regulates expression of the rat liver cytochrome P-450 15/3 at a pretranslational level. Biochem Biophys Res Commun 143:782-788 22. Posner Bl, Kelly PA, Friesen HG 1974 Induction of a lactogenic receptor in rat liver: influence of estrogen and the pituitary. Proc Natl Acad Sci USA 71:2407-2410 23. Posner Bl, Kelly PA, Friesen HG 1975 Prolactin receptors in rat liver: possible induction by prolactin. Science 188:57-59 24. Baxter RC, Zaltsman Z 1984 Induction of hepatic receptors for growth hormone (GH) and prolactin by GH infusion is sex independent. Endocrinology 115:2009-2014 25. Ranke MB, Stanley CA, Tenore A, Rodbard D, Bongiovanni AM, Parks JS 1976 Characterization of somatogenic and lactogenic sites in isolated rat hepatocytes. Endocrinology 99:1033-1045 26. Barash I, Cromlish W, Posner Bl 1988 Prolactin (PRL) receptor induction in cultured rat hepatocytes: dual regulation by prolactin and growth hormone. Endocrinolgy 122:1151-1158 27. Murphy LJ, Tachibana K, Friesen HG 1988 Stimulation of hepatic insulin-like growth factor-l gene expression by ovine prolactin: evidence for intrinsic somatogenic activity in the rat. Endocrinology 122:2027-2033 28. Baxter RC, Zaltsman Z, Turtle JR 1984 Rat growth hormone (GH) but not prolactin (PRL) induces both GH and PRL receptors in female rat liver. Endocrinology 114:1893-1901 29. Mode A, Norstedt G, Eneroth P, Gustafsson J-A 1983 Purification of liver feminizing factor from rat pituitaries and demonstration of its identity with growth hormone. Endocrinology 113:1250-1260 30. Husman B, Norstedt G, Mode A, Gustafsson J-A 1985 The mode of growth hormone administration is of major importance for the excretion of the major male rat urinary proteins. Mol Cell Endocrinol 40:205-?10 31. Zaphiropoulos P, Strom A, Robertson JA, Gustafsson JA 1990 Structural and regulatory analysis of the malespecific rat liver cytochrome P450g: repression by continuous growth hormone administration. Mol Endocrinol 4:53-58 32. Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, Arnheim N 1985 Enzymic amplification of /3-globin genomic sequence and restriction site analysis for diagnosis of sickle cell anaemia. Science 230:1350-1353 33. Dower WJ, Miller JF, Ragsdale CW 1988 High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145 34. Struhl K 1985 A rapid method for creating recombinant DNA molecules. Biotechniques 3:452-453 35. Feinberg AP, Vogelstein B 1983 A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13 36. Chomczynski P, Saachi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Anal Biochem 162:156-159 37. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning-A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
