- Email: [email protected]
Prolactin receptors and embryonic development: Gene expression and knockout studies
Trophoblast Research 11:13-21, 1998
PROLACTIN RECEPTORS AND EMBRYONIC GENE EXPRESSION AND KNOCKOUT - A Review -
DEVELOPMENT: STUDIES
Paul A. Kelly, Christopher J. Ormandy, Christine B61e-Feysot, Ronda Maaskant, Michael Freemark 1and Nadine Binart INSERM Unit6 344-Endocrinologie Mol6culaire Facult6 de M6decine Necker 75015 Paris, France 1Departments of Pediatrics and Cell Biology Duke University Medical Center Durham, North Carolina 27710 USA INTRODUCTION The anterior pituitary hormone prolactin (PRL) and placental lactogen (PL) both bind to the prolactin receptor (PRLR), which has a very broad distribution in known target tissues. In addition, with the advent of more sensitive techniques, many cells and tissues have been shown to express varying levels of PRL receptor mRNA or protein. PRL and PL have diverse effects on growth, metabolism, reproduction, mammary function and immune function (Nicoll and Bern, 1972; Ogren and Talamantes, 1988). Prolactin binding sites or receptors have been identified in a number of cells and tissues of adult mammals. In rats and mice, two forms of receptor, generated by alternative splicing of a single gene (Boutin et al., 1988; Shirota et al., 1990; Davis and Linzer, 1989; Clarke et al., 1993) have been identified: a short form of -290 amino acids and a long form of -600 amino acids. The expression of short and long forms of receptor have been shown to vary as a function of the stage of the estrous cycle, pregnancy and lactation (Nagano and Kelly, 1994; Ouhtit et al., 1993a; Ouhtit et al., 1993b). Although PRL and GH receptors show only limited overall sequence identity (~30%), their tertiary structures are highly related. Their extracellular domains (N200 aa) can be divided into two subdomains of ~100 aa, each of which displays homology with the type III module of fibronectin. The N-terminal subdomain contains four conserved disulfide-linked cysteines (De Vos et al., 1992; Somerset al., 1994) which border regions of up to 70% identity between PRLR and GHR (Kelly et al., 1991). The C-terminal subdomain of PRLR contains another typical feature, termed the WS motif composed of the Trp-Ser-X-Trp-Ser penta-peptide. Although this WS motif is not exactly replicated in the GHR, conservative substitutions are observed (Tyr-Gly-X-Phe-Ser). The threedimensional structures of the hPRLbp and hGHbp have been recently solved by crystallographic analysis (De Vos et al., 1992; Somerset al., 1994). Both receptors fold into two antiparallel D-sheets, each composed of seven B-strands. Despite their low degree of sequence similarity, the 3D structures of the ligand binding domains of GHR and PRLR appear to be almost superimposable (Kossiakoff et al., 1994), suggesting that interactions with their ligands, ligands that also display structural analogies, occur
13
9
of Rochester
14
Kelly et al.
through very similar mechanisms (Goffin et al., 1996). The intracellular domains of PRLR and GHR are more divergent than the extracellular domains and sequence consensus is restricted to two regions, called Box 1 and Box 2 (Kelly et al., 1991; Murakami et al., 1991). Box 1 is a membrane-proximal region composed of eight amino acids highly enriched in prolines and hydrophobic residues, whereas Box 2 is even less conserved and consists of a succession of hydrophobic, negatively-charged then positively-charged residues. Based on their structural features, PRLR and GHR were initially believed to form a distinct receptor family (Kelly et aL, 1991). However, further sequence comparison of newly-identified membrane receptors led to the identification of a new family, termed Class-1 cytokine receptors, that contains over 20 members (Figure 1).
I The Cytokine Receptor Superfamily I Class I
Class II
I
I
r,o~
...... 'c
..............
I
I
AI
F
IL-2R [3 EPOR IL-2R y GHR IL-4R PRLR IL-9R
13c TPOR
IFNABR G-CSFR gp 130
IFNGR c~
OBR
Figure 1. Schematic representation of various members of cytokine receptors. The structural features of Class I and Class II families are shown in the left and right insets respectively. *fibronectin type III module; **immunoglobulin-like region. IL, interleukin, EPO. Erythropoietin; GH, growth hormone; PRL, prolactinh, GM-CSF, granulocytemacrophage colony stimulating factor; TPO, thrombopoietin; G-CSF, granulocyte colony stimulating factor; gp130, glycoprotein 130 kDda; OBR, obese factor (leptin); LIF, leukemia inhibitory factor; CNTF, ciliary neurotropic factor; IFN, interferon; WSxWS, tryptophane, serine, any amino acid, tryptophane, serine.
PRL Receptor Expression and Knockout
15
PROLACTIN RECEPTOR EXPRESSION
As mentioned earlier, we have established a widespread distribution of the PRLR in almost all adult rat tissues (Nagano and Kelly, 1994; Ouhtit et al., 1993a; Ouhtit et al., 1993b; Ouhtit et al., 1994). There was, however, very little information on the expression of this receptor during fetal development. To that end, we have recently determined the cellular distribution and developmental expression of the PRLR in the late gestational fetal rat by in situ hybridization, immunocytochemistry and radioligand binding (Royster et al., 1995). Sense and antisense strand probes were prepared encoding the long and short isoforms of the rat PRLR and hybridized to various fetal tissues obtained at the end of pregnancy (days 17.5 to 20.5). These original studies showed that the mRNA expressing the short and long isoforms was widely expressed in tissues from all three germ layers: in addition to the classical target organs of PRL, tissues not known previously to contain PRL receptors, such as olfactory neuronal epithelium and bulb, trigeminal and dorsal root ganglia, cochlear duct, brown adipose tissue, submandibular glands, whisker follicles, tooth primordia, and proliferative and maturing chondrocytes of developing bone. There was also as high level of expression of receptor mRNA in the fetal adrenal cortex, gastrointestinal and bronchial mucosae, renal tubular epithelia, choroid plexus, thymus, liver pancreas, and epidermis. To complement the in situ studies, immunohistochemical studies using monoclonal anti-PRLR antibodies clearly demonstrated the distribution of PRLR immunoreactivity was similar to that of the mRNA, strongly suggesting that the receptor protein is expressed in the developing fetus. The functionality of the PRL receptors was established by the demonstration of specific rat placental lactogen II binding sites in fetal adrenal cortex, renal tubules, small intestinal villi, pancreatic ductules and islets, hepatic parenchyma cells, choroid plexus ependymal cells, fetal lung and thymus. The level of PRL receptor mRNA and protein actually increased between days 17.5 and 20.5 of pregnancy in a number of tissues, including the adrenal, pancreas, small intestine, pituitary, thymus, liver, and submandibular gland. These results suggest that lactogenic hormones such as prolactin and placental lactogens may play important roles in fetal and neonatal development (Royster et al., 1995). Since the PRLR is expressed at relatively low levels in the olfactory bulb of the adult rat, but is easily detected in late pregnancy in the fetal rat, we decided to investigate the ontogenesis of PRLR expression in the olfactory system, again using in situ hybridization and immunohistochemistry (Freemark et aI., 1996). At embryonic day 12.5 (e12.5), mRNAs encoding the long and short isoforms of the PRLR were detected in the medial and lateral nasal processes, the epithelial lining of the olfactory pit, and the neuroepithelium lining of the cerebral ventricles, in the region of the rhinencephalon. PRLR mRNA was also highly expressed in the frontonasal mesenchyme and the mesenchymal tissue m~derlying the developing brain and in the interpeduncular fossa. Once again, the PRLR immunoreactivity was similar to that of mRNA, suggesting that the PRL receptor gene was translated in lactogen binding sites or receptors in the developing embryo. As pregnancy advanced, the receptor was expressed intensely, albeit discontinuously, in the olfactory system. Receptor expression was also seen in the cartilage primordia of the ethmoid, sphenoid, temporal, and mandibular bones. Although the PRLR was expressed in the vomeronasal organ, it was limited to the luminal epithelial surface.
16
Kelly et al.
It was not until embryonic day 18 that PRLR m R N A and protein was detected in the olfactory bulb. The highest level of expression was seen in the periventricular neuroepithelium. Thereafter, strong staining was observed in the mitral and tufted cell neurons and the sensory neuronal cell bodies of the olfactory epithelium. The high level of expression continued until neonatal day 5. Interestingly, PRLR expression was also found in the mitral cells of the olfactory bulb of the lactating rat, although the levels appear to be much lower than those seen in the fetal and neonatal rat. These studies suggest novel roles for lactogenic hormones in olfactory differentiation and development and m a y provide new mechanisms by which lactogenic hormones m a y regulate neonatal behavior and maternal-infant interactions (Freemark et al., 1996). PHENOTYPES ASSOCIATED RECEPTOR IN MICE
WITH
KNOCKOUT
OF
THE
PROLACTIN
A 129/Sv mouse genomic DNA library constructed using lambda DASHII (Stratagene) was screened using the mouse PRLRS3 cDNA (Davis and Linzer, 1989). The coding region of the PRLR gene was isolated as a series of clones spanning approximately 80 kb of DNA. Exons 4 and 5 were each found to contain a pair of extracellular cysteine residues. Loss of just one of these cysteines results in complete lack of hormone binding activity (Rozakis-Adcock and Kelly, 1991). A targeting construct was prepared with 7.5 kb of overall homology in which a 1o5 kb fragment containing exon 5 was replaced with the similarly sized Tk-NEO cassette, which resulted in a mutation creating an in-frame stop codon. Thus, if a mR_NA was transcribed from this mutated gene, it would encode a very short protein of 44 amino acids, without any of the functional domains required for ligand binding, membrane insertion or signal transduction. After electroporation into E14.1 embryonic stem cells and neomycin selection, 3 out 214 G418 resistant clones were found to have undergone 3" and 5' homologous recombination, using Southern blot analysis with the appropriate restriction enzymes and probes. Two of these clones, E8 and H2, were used for microinjection into 3.5 day old C57BL/6 blastocysts and were able to generate germline chimeras. The presence of the mutated allele in the F1 progeny of the chimeras was revealed by Southern analysis. F1 intercrosses revealed a normal genotype distribution. The expression of the mutated PRLR gene was confirmed using RT-PCR and northern blotting, and the absence of the PRLR protein was demonstrated by western blot (Ormandy et al., 1997).
Impaired Mammary Gland Development and Lactation in Heterozygous Females When 6 to 8 week old PRLR heterozygote (PRLR*/)F1 females were mated with either + / + or + / - males, most of their first litter died within 24 hours, and virtually the entire litter had perished by 48 hours. All pups were observed to attach to the nipple and suckle, however dead pups were dehydrated, with loose skin and loss of weight, and examination of their stomach contents showed air bubbles but no milk present, indicating that PRLR +/- females were unable to lactate. When assessed by pup survival, this phenotype was not apparent following the second pregnancy, where all F1 PRLR */females produced surviving pups. Histological examination of the m a m m a r y glands from the F1 animals after 48 hours of their first lactation showed that lactational performance was correlated with the degree of m a m m a r y gland development. M a m m a r y glands from nonlactating mothers showed very little development, while those from mothers with some surviving pups
PRL Receptor Expression and Knockout
17
showed modest development when compared to the highly developed state of the mammary glands from PRLR*/+ mothers. Combined, these results demonstrate that two functional alleles of the PRLR are required for correct lactation and that this phenotype in heterozygotes is primarily due to a deficit in the degree of mammary gland development (Ormandy et al., 1997).
Homozygous Females Are Sterile PRLR-/- females when mated either with PRLR-/ or PRLR§247males were infertile~ Even after a two month period, pregnancy was never observed. The presence of vaginal seminal plugs was checked daily in a series of 11 PRLR-/ females housed with PRLR ~/~ males of established fertility: approximately half of the PRLR/ females mated every four days, while one animal mated every three days and the other PRLR/- females mated irregularly, every three to four days. In contrast, PRLR*/* females mated only every twelve days with vasectomized males. These results are indicative of the presence of a number of reproductive deficiencies in PRLR / females. Firstly, all PRLR4- females were sterile despite regular mating. Secondly, mating did not produce a pseudopregnancy h~ PRLR-/ females, as it did in the PRLR+/+ females for the twelve days following mating. This was confirmed by examination of estrogen levels which showed a large increase on day 3 after the vaginal plug, as the animals again entered estrus~ Thirdly, the irregular mating patterns of half of the females indicates an alteration of the estrous cycle. To investigate the cause of sterility of the PRLR-/- females, the preimplantation development of embryos in PRLR*/* and PRLR-/- females was compared after mating to PRLR*/+ males of established fertility. The embryos were flushed from the oviduct at various times, and their development were determined. Multiple abnormalities were observed in the PRLR-/- females: fewer eggs were fertilized, oocytes at the germinal vesicle stage were released from the ovary and fragmented embryos were found. The number of eggs ovulated was reduced in the PRLR/ females as compared to controls, and this observation was supported by histological investigation which showed fewer primary follicles in PRLR/- ovaries. Most importantly, fertilized eggs develop poorly to the blastocyst stage in PRLR4- animals. Only 19% of blastocysts were recovered at day 3.5 in the uterus of PRLR-/- against 85% in wild type animals. Interestingly, single cell fertilized eggs were recovered at all of the stages studied, suggesting for most oocytes that an arrest of development occurred immediately after fertilization. In order to test if the absence of development of the fertilized eggs in the PRLR-/- females was due to the lack of receptor in the oocytes or to the oviduct environment, transplantation experiments were performed. PRLR+/* and PRLR-/- females were mated to PRLR*/* males and fertilized single or 2-cell stage embryos were flushed from their oviducts and reimplanted into the oviducts of pseudopregnant F1 C57BL/6xCBA foster mothers at day 1. Of the 25 embryos recovered from PRLR*/* females, 13 produced normal day 12 embryos, while of the 22 embryos recovered from PRLR4- females, 15 also produced normal embryos. When this experiment was repeated using fertile PRLR-/- males (to exclude the possible paternal contribution of PRLR to the embryo) a similar rate of normal embryos was recovered, demonstrating that the eggs are viable, and thus that the environment of the embryo in the oviduct is deficient. Despite the fact that occasionally we found blastocysts in the uterus of PRLR/~ females at day 3.5, pregnancy was never observed. To examine whether the uterus of PRLR4 females was able to accept the implantation of wild type embryos, 3 PRLR*/§ or
18
Kelly et al.
PRLR -/- females were mated with vasectomized males, and 2.5 days after a vaginal seminal plug was observed, 7 to 8 + / + blastocysts were reimplanted into the same uterine horn. Although 15 of 21 blastocysts reimplanted into PRLR */+ females produced normal fetuses, none of 24 blastocysts were able to implant in PRLR -/- females, indicating that the uterus of these animals is refractory to implantation. Thus the absence of PRLR in female mice results in reduced ovulation, reduced fertilization, and almost complete arrest of preimplantation development. The small number of embryos which progress to the blastocyst stage are released into an environment refractory to implantation. The outcome is complete sterility (Ormandy et al., 1997).
Homozygous Male Fertility The fertility of a group of 15 12-18 week old PRLR ~/- males was examined by housing each separately for 50 days with a 12-14 week old PRLR §247female of proven fertility. Vaginal plugs were checked daily. Seven males produced a pregnancy following detection of the first vaginal plug, and all these females became pregnant again immediately following delivery, for a total of three pregnancies. When these males were later each housed separately with four females, initial mating produced a pregnancy in all cases, even when two of the females were simultaneously in estrus. These seven males were thus judged to be fully fertile. Five of the 15 males were, however, judged to be partially fertile, as multiple vaginal plugs were required for successful fertilization and pregnancy. Three of the 15 males (20%) were completely infertile, and never produced a litter~ Testes were of normal size and histological examination of testes from fertile, semifertile and infertile animals showed no obvious differences or abnormalities, with clearly defined germinal cell layers and spermatocytes present in the seminiferous tubules. The generation of a PRLR/- mouse has provided a means of studying the role of the PRL receptor system in mice. Short and long forms of PRL receptor exist in the mouse. It is probable that most of the phenotypes observed are related to the absence of the long form of the receptor, since this is the major form in all ceils involved in reproductive system. Almost every aspect of reproduction is altered in these animals: mating frequency, oocyte maturation, ovulation rates, pseudopregnancy, fertilization rates, control of preimplantation development, uterine receptivity to embryo implantation, m a m m a r y gland development, lactation, maternal behavior and male fertility, unambiguously demonstrating that the PRLR is a key regulator of reproduction. The ability of this new model to provide novel insights into the function of lactogenic hormones and their receptor illustrates the power of the knockout approach to discover unknown roles for well- investigated molecules (Ormandy et al., 1997). I D E N T I F I C A T I O N OF PROLACTIN-REGULATED GENES Although m a n y components of the signal transduction process of prolactin have been identified, and a number of prolactin-induced genes have been determined, there are a number of genes regulated by prolactin that remain unidentified. We are currently using two complementary techniques to identify and characterize prolactin-induced genes. The ideal models for such approaches include cell lines that proliferate in response to prolactin, such as the Nb2 lymphoma, and tissues and organs from normal versus prolactin receptor knockout mice, where a clear effect of prolactin can be discerned.
PRL Receptor Expression and Knockout
19
The first approach is representational difference analysis (RDA), which is PCRcoupled subtractive process to identify differentially regulated genes (Hubank and Schatz, 1994). We are in the process of analyzing over 20 different cDNAs isolated by RDA from prolactin stimulated ceils, some of which code for new proteins (B61e-Feysot, et al., paper in preparation). The second technique is serial analysis of gene expression (SAGE) that allows the quantitative and simultaneous analysis of a large number of transcripts from a single tissue (Velculescu et al., 1995). In fact, SAGE represents an ideal means of quantitatively cataloging and comparing expressed genes from tissues and cells from prolactin receptor ka~ockout compared to wild type mice, with the hope of detecting new prolactin-dependent genes. SUMMARY
Prolactin (PRL), secreted by the pituitary, decidua and lymphoid cells, has been shown to have a regulatory role in reproduction, immune function and cell growth in mammals including stimulation of mammary gland development, lactation and mammary tumor growth. The effects of PRL are mediated by a membrane-bound receptor which is a member of the superfalnily of cytokine receptors. In the rat, the PRL receptor has a broad distribution in fetat tissues, beginning at embryonic day 12.5 (e12.5) through late gestation (e20.5). To better define the specific and multiple roles of prolactin and its various receptor isoforms, we have produced mice by gene targeting in embryonic stern cells carrying a germline null mutation of the prolactin receptor gene. Heterozygous (+/-) females show almost complete failure to lactate, following their first, but not subsequent pregnancies. The lactational deficiency of +/- females has been confirmed in inbred 129 mice. Homozygous (-/-) females are infertile due to multiple reproductive abnormalities, including ovulation of premeiotic oocytes, reduced fertilization of oocytes, reduced preimplantation oocyte development, and lack of embryo implantation, and the absence of pseudopregnancy. Half of the homozygous males are infertile or show reduced fertility. In view of the widespread distribution of PRL receptors, other phenotypes are currently being evaluated in -/- animals. This study establishes the prolactin receptor as a key regulator of mammalian reproduction, and provides the first total ablation model to further study the role of the prolactin receptor and its ligands. We are currently using approaches such as representational difference analysis (RDA) and serial analysis of gene expression (SAGE) to identify new genes either induced by prolactin or involved in the signal transduction pathway of the hormone. REFERENCES
Boutin, J.M., Jolicoeur, C., Okamura, H., Gagnon, J., Edery, M., Shirota, M., Banville, D., Dusanter-Fourt, I., Djiane, J. and Kelly, P.A. (1988) Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family. Ceil 53, 69-77. Clarke, D.L., Arey, B.J. and Linzer, D.I.H. (1993) Prolactin receptor messenger ribonucleic acid expression in the ovary during rat estrous cycle. Endocrinology 133, 2594-2603. Davis, J.A. and Linzer, D.I.H. (1989) Expression of multiple forms of the prolactin receptor. Mol. EndocrinoI. 3, 674-680.
20
Kelly et al.
De Vos, A.M., Ultsch, M and Kossiakoff, A.A. (1992) Human growth hormone and extracellular domain of its receptor: Crystal structure of the complex. Science 255, 306-312. Freemark, M., Driscoll, P., Andrews, J., Kelly, P.A. and Royster, M. (1996) Ontogenesis of prolactin receptor gene expression in the rat olfactory system: potential roles for lactogenic hormones in olfactory development. Endocrinology 137, 934-942. Goffin, V., Shiverick, K.T., Kelly, P.A. and Martial, J.A. (1996) Sequence-function relationships within the expanding family of prolactin, growth hormone, placental lactogen and related proteins in mammals. Endocr Rev. 17, 385-410. Hubank, M. and Schatz, D.G. (1994) Identifying differences in mRNA expression by representational difference analysis of cDNA. Nucleic Acids Res. 22, 5640-5648. Kelly, P.A., Djiane, J., Postel-Vinay, M.C. and Edery, M. (1991) The prolactin/growth hormone receptor family. Endocr. Rev. 12, 235-251. Kossiakoff, A.A., Somers, W., Ultsch, M., Andow, K., Muller, Y.A. and De Vos, A.M. (1994) Comparison of the intermediate complexes of human growth hormone bound to the human growth hormone and prolactin receptors. Protein Sci. 3, 16971705. Murakami, M., Narazaki, M., Hibi, M., Yawata, H., Yazukawa, K., Hamaguchi, M., Taga, T. and Kishimoto, T. (1991) Critical cytoplasmic region of the IL-6 signal transducer, gp 130, is conserved in the cytokine receptor family. Proc. Natl. Acad. Sci. USA 88, 11349-11353. Nagano, M. and Kelly, P.A. (1994) Tissue distribution and regulation of rat prolactin receptor gene expression: Quantitative analysis by polymerase chain reaction~ J. Biol. Chem. ,~o . . . . v ~,,~,.
Nicoll, C.S. and Bern, H. (1972) On the actions of PRL among the vertebrates: is there a common denominator? In: Lactogenic Hormones, (ed.) G.E.W. Wotstenhohme and j. Knight, London, Churchill Livhqgstone, pp. 299=337~ Ogren, L. and Talamantes, F. (1988) Prolactins of pregnancy and their cellular source. InL LTy~UL1! 2 i -g'5 Oi-mandy, C.j., Camus, A., Barra, j., Damotte, D., Lucas, B.K., Bureau, H., Edery, M., Brousse, N., Babinet, C., Binart, N. and Kelly, P.A. (1997) Null mutation of the prolactin receptor gene produces multiple reproductive defects in tile mouse_ Genes Dev, 11, 16_7178. Ouhtit, A., Kelly, P.A. and Morel, G. (1994) Visualization of gene expression of short and long forms of proiactm receptor in rat digestive tissues. Am. J. PhysioL 266, G807G815. Ouhtit, A., Morel, G. and Kelly, P.A. (1993a) Visualization of gene expression of short and long forms of prolactin receptor in the rat. Endocrinoios 133, 135q44.
PRL Receptor Expression and Knockout
21
Ouhtit, A., Morel G. and Kelly, P.A. (1993b) Visualization of gene expression of short and long forms of prolactin receptor in rat reproductive tissues. Biol. Reprod. 49, 528-536. Royster, M., Driscoll, P., Kelly, P.A. and Freemark, M. (1995) The prolactin receptor in the fetal rat: Cellular localization of messenger RNA, immunoreactive protein, and ligand binding activity and induction of expression in late gestation. Endocrinology 136, 3892-3900. Rozakis-Adcock, M. and Kelly, P.A. (1991) Mutational analysis of the ligand binding domain of the prolactin receptor. J. Biol. Chem. 266, 16472-16477. Shirota, M., Banville, D., Ali, S., Jolicoeur, C., Boutin, J-M., Edery, M., Djiane, J. and Kelly, P.A. (1990) Expression of two forms of prolactin receptor in rat ovary and liver~ Mol. Endocrinol. 4, 1136-1142. Somers, W., Ultsch, M., De Vos, A.M. and Kossiakoff, A.A. (1994) The X~ray structure of the growth hormone-prolactin receptor complex: Receptor binding specificity developed through conformational variability. Nature 372, 478-481. Velculescu, V.E., Zhang, L., Vogelstein, B~ and Kinzler, K.W. (1995) Serial analysis of gene expression. Science 270, 484-487.
PROLACTIN RECEPTORS AND EMBRYONIC GENE EXPRESSION AND KNOCKOUT - A Review -
DEVELOPMENT: STUDIES
Paul A. Kelly, Christopher J. Ormandy, Christine B61e-Feysot, Ronda Maaskant, Michael Freemark 1and Nadine Binart INSERM Unit6 344-Endocrinologie Mol6culaire Facult6 de M6decine Necker 75015 Paris, France 1Departments of Pediatrics and Cell Biology Duke University Medical Center Durham, North Carolina 27710 USA INTRODUCTION The anterior pituitary hormone prolactin (PRL) and placental lactogen (PL) both bind to the prolactin receptor (PRLR), which has a very broad distribution in known target tissues. In addition, with the advent of more sensitive techniques, many cells and tissues have been shown to express varying levels of PRL receptor mRNA or protein. PRL and PL have diverse effects on growth, metabolism, reproduction, mammary function and immune function (Nicoll and Bern, 1972; Ogren and Talamantes, 1988). Prolactin binding sites or receptors have been identified in a number of cells and tissues of adult mammals. In rats and mice, two forms of receptor, generated by alternative splicing of a single gene (Boutin et al., 1988; Shirota et al., 1990; Davis and Linzer, 1989; Clarke et al., 1993) have been identified: a short form of -290 amino acids and a long form of -600 amino acids. The expression of short and long forms of receptor have been shown to vary as a function of the stage of the estrous cycle, pregnancy and lactation (Nagano and Kelly, 1994; Ouhtit et al., 1993a; Ouhtit et al., 1993b). Although PRL and GH receptors show only limited overall sequence identity (~30%), their tertiary structures are highly related. Their extracellular domains (N200 aa) can be divided into two subdomains of ~100 aa, each of which displays homology with the type III module of fibronectin. The N-terminal subdomain contains four conserved disulfide-linked cysteines (De Vos et al., 1992; Somerset al., 1994) which border regions of up to 70% identity between PRLR and GHR (Kelly et al., 1991). The C-terminal subdomain of PRLR contains another typical feature, termed the WS motif composed of the Trp-Ser-X-Trp-Ser penta-peptide. Although this WS motif is not exactly replicated in the GHR, conservative substitutions are observed (Tyr-Gly-X-Phe-Ser). The threedimensional structures of the hPRLbp and hGHbp have been recently solved by crystallographic analysis (De Vos et al., 1992; Somerset al., 1994). Both receptors fold into two antiparallel D-sheets, each composed of seven B-strands. Despite their low degree of sequence similarity, the 3D structures of the ligand binding domains of GHR and PRLR appear to be almost superimposable (Kossiakoff et al., 1994), suggesting that interactions with their ligands, ligands that also display structural analogies, occur
13
9
of Rochester
14
Kelly et al.
through very similar mechanisms (Goffin et al., 1996). The intracellular domains of PRLR and GHR are more divergent than the extracellular domains and sequence consensus is restricted to two regions, called Box 1 and Box 2 (Kelly et al., 1991; Murakami et al., 1991). Box 1 is a membrane-proximal region composed of eight amino acids highly enriched in prolines and hydrophobic residues, whereas Box 2 is even less conserved and consists of a succession of hydrophobic, negatively-charged then positively-charged residues. Based on their structural features, PRLR and GHR were initially believed to form a distinct receptor family (Kelly et aL, 1991). However, further sequence comparison of newly-identified membrane receptors led to the identification of a new family, termed Class-1 cytokine receptors, that contains over 20 members (Figure 1).
I The Cytokine Receptor Superfamily I Class I
Class II
I
I
r,o~
...... 'c
..............
I
I
AI
F
IL-2R [3 EPOR IL-2R y GHR IL-4R PRLR IL-9R
13c TPOR
IFNABR G-CSFR gp 130
IFNGR c~
OBR
Figure 1. Schematic representation of various members of cytokine receptors. The structural features of Class I and Class II families are shown in the left and right insets respectively. *fibronectin type III module; **immunoglobulin-like region. IL, interleukin, EPO. Erythropoietin; GH, growth hormone; PRL, prolactinh, GM-CSF, granulocytemacrophage colony stimulating factor; TPO, thrombopoietin; G-CSF, granulocyte colony stimulating factor; gp130, glycoprotein 130 kDda; OBR, obese factor (leptin); LIF, leukemia inhibitory factor; CNTF, ciliary neurotropic factor; IFN, interferon; WSxWS, tryptophane, serine, any amino acid, tryptophane, serine.
PRL Receptor Expression and Knockout
15
PROLACTIN RECEPTOR EXPRESSION
As mentioned earlier, we have established a widespread distribution of the PRLR in almost all adult rat tissues (Nagano and Kelly, 1994; Ouhtit et al., 1993a; Ouhtit et al., 1993b; Ouhtit et al., 1994). There was, however, very little information on the expression of this receptor during fetal development. To that end, we have recently determined the cellular distribution and developmental expression of the PRLR in the late gestational fetal rat by in situ hybridization, immunocytochemistry and radioligand binding (Royster et al., 1995). Sense and antisense strand probes were prepared encoding the long and short isoforms of the rat PRLR and hybridized to various fetal tissues obtained at the end of pregnancy (days 17.5 to 20.5). These original studies showed that the mRNA expressing the short and long isoforms was widely expressed in tissues from all three germ layers: in addition to the classical target organs of PRL, tissues not known previously to contain PRL receptors, such as olfactory neuronal epithelium and bulb, trigeminal and dorsal root ganglia, cochlear duct, brown adipose tissue, submandibular glands, whisker follicles, tooth primordia, and proliferative and maturing chondrocytes of developing bone. There was also as high level of expression of receptor mRNA in the fetal adrenal cortex, gastrointestinal and bronchial mucosae, renal tubular epithelia, choroid plexus, thymus, liver pancreas, and epidermis. To complement the in situ studies, immunohistochemical studies using monoclonal anti-PRLR antibodies clearly demonstrated the distribution of PRLR immunoreactivity was similar to that of the mRNA, strongly suggesting that the receptor protein is expressed in the developing fetus. The functionality of the PRL receptors was established by the demonstration of specific rat placental lactogen II binding sites in fetal adrenal cortex, renal tubules, small intestinal villi, pancreatic ductules and islets, hepatic parenchyma cells, choroid plexus ependymal cells, fetal lung and thymus. The level of PRL receptor mRNA and protein actually increased between days 17.5 and 20.5 of pregnancy in a number of tissues, including the adrenal, pancreas, small intestine, pituitary, thymus, liver, and submandibular gland. These results suggest that lactogenic hormones such as prolactin and placental lactogens may play important roles in fetal and neonatal development (Royster et al., 1995). Since the PRLR is expressed at relatively low levels in the olfactory bulb of the adult rat, but is easily detected in late pregnancy in the fetal rat, we decided to investigate the ontogenesis of PRLR expression in the olfactory system, again using in situ hybridization and immunohistochemistry (Freemark et aI., 1996). At embryonic day 12.5 (e12.5), mRNAs encoding the long and short isoforms of the PRLR were detected in the medial and lateral nasal processes, the epithelial lining of the olfactory pit, and the neuroepithelium lining of the cerebral ventricles, in the region of the rhinencephalon. PRLR mRNA was also highly expressed in the frontonasal mesenchyme and the mesenchymal tissue m~derlying the developing brain and in the interpeduncular fossa. Once again, the PRLR immunoreactivity was similar to that of mRNA, suggesting that the PRL receptor gene was translated in lactogen binding sites or receptors in the developing embryo. As pregnancy advanced, the receptor was expressed intensely, albeit discontinuously, in the olfactory system. Receptor expression was also seen in the cartilage primordia of the ethmoid, sphenoid, temporal, and mandibular bones. Although the PRLR was expressed in the vomeronasal organ, it was limited to the luminal epithelial surface.
16
Kelly et al.
It was not until embryonic day 18 that PRLR m R N A and protein was detected in the olfactory bulb. The highest level of expression was seen in the periventricular neuroepithelium. Thereafter, strong staining was observed in the mitral and tufted cell neurons and the sensory neuronal cell bodies of the olfactory epithelium. The high level of expression continued until neonatal day 5. Interestingly, PRLR expression was also found in the mitral cells of the olfactory bulb of the lactating rat, although the levels appear to be much lower than those seen in the fetal and neonatal rat. These studies suggest novel roles for lactogenic hormones in olfactory differentiation and development and m a y provide new mechanisms by which lactogenic hormones m a y regulate neonatal behavior and maternal-infant interactions (Freemark et al., 1996). PHENOTYPES ASSOCIATED RECEPTOR IN MICE
WITH
KNOCKOUT
OF
THE
PROLACTIN
A 129/Sv mouse genomic DNA library constructed using lambda DASHII (Stratagene) was screened using the mouse PRLRS3 cDNA (Davis and Linzer, 1989). The coding region of the PRLR gene was isolated as a series of clones spanning approximately 80 kb of DNA. Exons 4 and 5 were each found to contain a pair of extracellular cysteine residues. Loss of just one of these cysteines results in complete lack of hormone binding activity (Rozakis-Adcock and Kelly, 1991). A targeting construct was prepared with 7.5 kb of overall homology in which a 1o5 kb fragment containing exon 5 was replaced with the similarly sized Tk-NEO cassette, which resulted in a mutation creating an in-frame stop codon. Thus, if a mR_NA was transcribed from this mutated gene, it would encode a very short protein of 44 amino acids, without any of the functional domains required for ligand binding, membrane insertion or signal transduction. After electroporation into E14.1 embryonic stem cells and neomycin selection, 3 out 214 G418 resistant clones were found to have undergone 3" and 5' homologous recombination, using Southern blot analysis with the appropriate restriction enzymes and probes. Two of these clones, E8 and H2, were used for microinjection into 3.5 day old C57BL/6 blastocysts and were able to generate germline chimeras. The presence of the mutated allele in the F1 progeny of the chimeras was revealed by Southern analysis. F1 intercrosses revealed a normal genotype distribution. The expression of the mutated PRLR gene was confirmed using RT-PCR and northern blotting, and the absence of the PRLR protein was demonstrated by western blot (Ormandy et al., 1997).
Impaired Mammary Gland Development and Lactation in Heterozygous Females When 6 to 8 week old PRLR heterozygote (PRLR*/)F1 females were mated with either + / + or + / - males, most of their first litter died within 24 hours, and virtually the entire litter had perished by 48 hours. All pups were observed to attach to the nipple and suckle, however dead pups were dehydrated, with loose skin and loss of weight, and examination of their stomach contents showed air bubbles but no milk present, indicating that PRLR +/- females were unable to lactate. When assessed by pup survival, this phenotype was not apparent following the second pregnancy, where all F1 PRLR */females produced surviving pups. Histological examination of the m a m m a r y glands from the F1 animals after 48 hours of their first lactation showed that lactational performance was correlated with the degree of m a m m a r y gland development. M a m m a r y glands from nonlactating mothers showed very little development, while those from mothers with some surviving pups
PRL Receptor Expression and Knockout
17
showed modest development when compared to the highly developed state of the mammary glands from PRLR*/+ mothers. Combined, these results demonstrate that two functional alleles of the PRLR are required for correct lactation and that this phenotype in heterozygotes is primarily due to a deficit in the degree of mammary gland development (Ormandy et al., 1997).
Homozygous Females Are Sterile PRLR-/- females when mated either with PRLR-/ or PRLR§247males were infertile~ Even after a two month period, pregnancy was never observed. The presence of vaginal seminal plugs was checked daily in a series of 11 PRLR-/ females housed with PRLR ~/~ males of established fertility: approximately half of the PRLR/ females mated every four days, while one animal mated every three days and the other PRLR/- females mated irregularly, every three to four days. In contrast, PRLR*/* females mated only every twelve days with vasectomized males. These results are indicative of the presence of a number of reproductive deficiencies in PRLR / females. Firstly, all PRLR4- females were sterile despite regular mating. Secondly, mating did not produce a pseudopregnancy h~ PRLR-/ females, as it did in the PRLR+/+ females for the twelve days following mating. This was confirmed by examination of estrogen levels which showed a large increase on day 3 after the vaginal plug, as the animals again entered estrus~ Thirdly, the irregular mating patterns of half of the females indicates an alteration of the estrous cycle. To investigate the cause of sterility of the PRLR-/- females, the preimplantation development of embryos in PRLR*/* and PRLR-/- females was compared after mating to PRLR*/+ males of established fertility. The embryos were flushed from the oviduct at various times, and their development were determined. Multiple abnormalities were observed in the PRLR-/- females: fewer eggs were fertilized, oocytes at the germinal vesicle stage were released from the ovary and fragmented embryos were found. The number of eggs ovulated was reduced in the PRLR/ females as compared to controls, and this observation was supported by histological investigation which showed fewer primary follicles in PRLR/- ovaries. Most importantly, fertilized eggs develop poorly to the blastocyst stage in PRLR4- animals. Only 19% of blastocysts were recovered at day 3.5 in the uterus of PRLR-/- against 85% in wild type animals. Interestingly, single cell fertilized eggs were recovered at all of the stages studied, suggesting for most oocytes that an arrest of development occurred immediately after fertilization. In order to test if the absence of development of the fertilized eggs in the PRLR-/- females was due to the lack of receptor in the oocytes or to the oviduct environment, transplantation experiments were performed. PRLR+/* and PRLR-/- females were mated to PRLR*/* males and fertilized single or 2-cell stage embryos were flushed from their oviducts and reimplanted into the oviducts of pseudopregnant F1 C57BL/6xCBA foster mothers at day 1. Of the 25 embryos recovered from PRLR*/* females, 13 produced normal day 12 embryos, while of the 22 embryos recovered from PRLR4- females, 15 also produced normal embryos. When this experiment was repeated using fertile PRLR-/- males (to exclude the possible paternal contribution of PRLR to the embryo) a similar rate of normal embryos was recovered, demonstrating that the eggs are viable, and thus that the environment of the embryo in the oviduct is deficient. Despite the fact that occasionally we found blastocysts in the uterus of PRLR/~ females at day 3.5, pregnancy was never observed. To examine whether the uterus of PRLR4 females was able to accept the implantation of wild type embryos, 3 PRLR*/§ or
18
Kelly et al.
PRLR -/- females were mated with vasectomized males, and 2.5 days after a vaginal seminal plug was observed, 7 to 8 + / + blastocysts were reimplanted into the same uterine horn. Although 15 of 21 blastocysts reimplanted into PRLR */+ females produced normal fetuses, none of 24 blastocysts were able to implant in PRLR -/- females, indicating that the uterus of these animals is refractory to implantation. Thus the absence of PRLR in female mice results in reduced ovulation, reduced fertilization, and almost complete arrest of preimplantation development. The small number of embryos which progress to the blastocyst stage are released into an environment refractory to implantation. The outcome is complete sterility (Ormandy et al., 1997).
Homozygous Male Fertility The fertility of a group of 15 12-18 week old PRLR ~/- males was examined by housing each separately for 50 days with a 12-14 week old PRLR §247female of proven fertility. Vaginal plugs were checked daily. Seven males produced a pregnancy following detection of the first vaginal plug, and all these females became pregnant again immediately following delivery, for a total of three pregnancies. When these males were later each housed separately with four females, initial mating produced a pregnancy in all cases, even when two of the females were simultaneously in estrus. These seven males were thus judged to be fully fertile. Five of the 15 males were, however, judged to be partially fertile, as multiple vaginal plugs were required for successful fertilization and pregnancy. Three of the 15 males (20%) were completely infertile, and never produced a litter~ Testes were of normal size and histological examination of testes from fertile, semifertile and infertile animals showed no obvious differences or abnormalities, with clearly defined germinal cell layers and spermatocytes present in the seminiferous tubules. The generation of a PRLR/- mouse has provided a means of studying the role of the PRL receptor system in mice. Short and long forms of PRL receptor exist in the mouse. It is probable that most of the phenotypes observed are related to the absence of the long form of the receptor, since this is the major form in all ceils involved in reproductive system. Almost every aspect of reproduction is altered in these animals: mating frequency, oocyte maturation, ovulation rates, pseudopregnancy, fertilization rates, control of preimplantation development, uterine receptivity to embryo implantation, m a m m a r y gland development, lactation, maternal behavior and male fertility, unambiguously demonstrating that the PRLR is a key regulator of reproduction. The ability of this new model to provide novel insights into the function of lactogenic hormones and their receptor illustrates the power of the knockout approach to discover unknown roles for well- investigated molecules (Ormandy et al., 1997). I D E N T I F I C A T I O N OF PROLACTIN-REGULATED GENES Although m a n y components of the signal transduction process of prolactin have been identified, and a number of prolactin-induced genes have been determined, there are a number of genes regulated by prolactin that remain unidentified. We are currently using two complementary techniques to identify and characterize prolactin-induced genes. The ideal models for such approaches include cell lines that proliferate in response to prolactin, such as the Nb2 lymphoma, and tissues and organs from normal versus prolactin receptor knockout mice, where a clear effect of prolactin can be discerned.
PRL Receptor Expression and Knockout
19
The first approach is representational difference analysis (RDA), which is PCRcoupled subtractive process to identify differentially regulated genes (Hubank and Schatz, 1994). We are in the process of analyzing over 20 different cDNAs isolated by RDA from prolactin stimulated ceils, some of which code for new proteins (B61e-Feysot, et al., paper in preparation). The second technique is serial analysis of gene expression (SAGE) that allows the quantitative and simultaneous analysis of a large number of transcripts from a single tissue (Velculescu et al., 1995). In fact, SAGE represents an ideal means of quantitatively cataloging and comparing expressed genes from tissues and cells from prolactin receptor ka~ockout compared to wild type mice, with the hope of detecting new prolactin-dependent genes. SUMMARY
Prolactin (PRL), secreted by the pituitary, decidua and lymphoid cells, has been shown to have a regulatory role in reproduction, immune function and cell growth in mammals including stimulation of mammary gland development, lactation and mammary tumor growth. The effects of PRL are mediated by a membrane-bound receptor which is a member of the superfalnily of cytokine receptors. In the rat, the PRL receptor has a broad distribution in fetat tissues, beginning at embryonic day 12.5 (e12.5) through late gestation (e20.5). To better define the specific and multiple roles of prolactin and its various receptor isoforms, we have produced mice by gene targeting in embryonic stern cells carrying a germline null mutation of the prolactin receptor gene. Heterozygous (+/-) females show almost complete failure to lactate, following their first, but not subsequent pregnancies. The lactational deficiency of +/- females has been confirmed in inbred 129 mice. Homozygous (-/-) females are infertile due to multiple reproductive abnormalities, including ovulation of premeiotic oocytes, reduced fertilization of oocytes, reduced preimplantation oocyte development, and lack of embryo implantation, and the absence of pseudopregnancy. Half of the homozygous males are infertile or show reduced fertility. In view of the widespread distribution of PRL receptors, other phenotypes are currently being evaluated in -/- animals. This study establishes the prolactin receptor as a key regulator of mammalian reproduction, and provides the first total ablation model to further study the role of the prolactin receptor and its ligands. We are currently using approaches such as representational difference analysis (RDA) and serial analysis of gene expression (SAGE) to identify new genes either induced by prolactin or involved in the signal transduction pathway of the hormone. REFERENCES
Boutin, J.M., Jolicoeur, C., Okamura, H., Gagnon, J., Edery, M., Shirota, M., Banville, D., Dusanter-Fourt, I., Djiane, J. and Kelly, P.A. (1988) Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family. Ceil 53, 69-77. Clarke, D.L., Arey, B.J. and Linzer, D.I.H. (1993) Prolactin receptor messenger ribonucleic acid expression in the ovary during rat estrous cycle. Endocrinology 133, 2594-2603. Davis, J.A. and Linzer, D.I.H. (1989) Expression of multiple forms of the prolactin receptor. Mol. EndocrinoI. 3, 674-680.
20
Kelly et al.
De Vos, A.M., Ultsch, M and Kossiakoff, A.A. (1992) Human growth hormone and extracellular domain of its receptor: Crystal structure of the complex. Science 255, 306-312. Freemark, M., Driscoll, P., Andrews, J., Kelly, P.A. and Royster, M. (1996) Ontogenesis of prolactin receptor gene expression in the rat olfactory system: potential roles for lactogenic hormones in olfactory development. Endocrinology 137, 934-942. Goffin, V., Shiverick, K.T., Kelly, P.A. and Martial, J.A. (1996) Sequence-function relationships within the expanding family of prolactin, growth hormone, placental lactogen and related proteins in mammals. Endocr Rev. 17, 385-410. Hubank, M. and Schatz, D.G. (1994) Identifying differences in mRNA expression by representational difference analysis of cDNA. Nucleic Acids Res. 22, 5640-5648. Kelly, P.A., Djiane, J., Postel-Vinay, M.C. and Edery, M. (1991) The prolactin/growth hormone receptor family. Endocr. Rev. 12, 235-251. Kossiakoff, A.A., Somers, W., Ultsch, M., Andow, K., Muller, Y.A. and De Vos, A.M. (1994) Comparison of the intermediate complexes of human growth hormone bound to the human growth hormone and prolactin receptors. Protein Sci. 3, 16971705. Murakami, M., Narazaki, M., Hibi, M., Yawata, H., Yazukawa, K., Hamaguchi, M., Taga, T. and Kishimoto, T. (1991) Critical cytoplasmic region of the IL-6 signal transducer, gp 130, is conserved in the cytokine receptor family. Proc. Natl. Acad. Sci. USA 88, 11349-11353. Nagano, M. and Kelly, P.A. (1994) Tissue distribution and regulation of rat prolactin receptor gene expression: Quantitative analysis by polymerase chain reaction~ J. Biol. Chem. ,~o . . . . v ~,,~,.
Nicoll, C.S. and Bern, H. (1972) On the actions of PRL among the vertebrates: is there a common denominator? In: Lactogenic Hormones, (ed.) G.E.W. Wotstenhohme and j. Knight, London, Churchill Livhqgstone, pp. 299=337~ Ogren, L. and Talamantes, F. (1988) Prolactins of pregnancy and their cellular source. InL LTy~UL1! 2 i -g'5 Oi-mandy, C.j., Camus, A., Barra, j., Damotte, D., Lucas, B.K., Bureau, H., Edery, M., Brousse, N., Babinet, C., Binart, N. and Kelly, P.A. (1997) Null mutation of the prolactin receptor gene produces multiple reproductive defects in tile mouse_ Genes Dev, 11, 16_7178. Ouhtit, A., Kelly, P.A. and Morel, G. (1994) Visualization of gene expression of short and long forms of proiactm receptor in rat digestive tissues. Am. J. PhysioL 266, G807G815. Ouhtit, A., Morel, G. and Kelly, P.A. (1993a) Visualization of gene expression of short and long forms of prolactin receptor in the rat. Endocrinoios 133, 135q44.
PRL Receptor Expression and Knockout
21
Ouhtit, A., Morel G. and Kelly, P.A. (1993b) Visualization of gene expression of short and long forms of prolactin receptor in rat reproductive tissues. Biol. Reprod. 49, 528-536. Royster, M., Driscoll, P., Kelly, P.A. and Freemark, M. (1995) The prolactin receptor in the fetal rat: Cellular localization of messenger RNA, immunoreactive protein, and ligand binding activity and induction of expression in late gestation. Endocrinology 136, 3892-3900. Rozakis-Adcock, M. and Kelly, P.A. (1991) Mutational analysis of the ligand binding domain of the prolactin receptor. J. Biol. Chem. 266, 16472-16477. Shirota, M., Banville, D., Ali, S., Jolicoeur, C., Boutin, J-M., Edery, M., Djiane, J. and Kelly, P.A. (1990) Expression of two forms of prolactin receptor in rat ovary and liver~ Mol. Endocrinol. 4, 1136-1142. Somers, W., Ultsch, M., De Vos, A.M. and Kossiakoff, A.A. (1994) The X~ray structure of the growth hormone-prolactin receptor complex: Receptor binding specificity developed through conformational variability. Nature 372, 478-481. Velculescu, V.E., Zhang, L., Vogelstein, B~ and Kinzler, K.W. (1995) Serial analysis of gene expression. Science 270, 484-487.