The N-terminus of human prolactin modulates its biological properties
Introduction
Prolactin (PRL), growth hormone (GH) and placental lactogen (PL) are the main members of a family of homologous hormones identified more than three decades ago as having evolved from the duplication of a common ancestral gene (Miller and Eberhardt, 1983, Goffin et al., 1996b). These hormones are primarily secreted by the anterior pituitary (PRL, GH) or placenta (PL), but are also expressed in many tissues or cell types such as the mammary gland, the prostate and lymphocytes (Ben-Jonathan et al., 1996). All three hormones, but especially PRL, exert an extremely wide spectrum of functions including the regulation of reproduction, growth, morphogenesis, metabolism, immune response and behavior (Bole-Feysot et al., 1998). One of the most intriguing questions in the biology of these pleiotropic hormones is how their functional specificity is directed at the molecular level. Indeed, although they are homologous and present high sequence similarity (Sinha, 1995), their functional specificity is first attained by binding to membrane receptors. The only two receptors currently known for these ligands are the growth hormone receptor (GHR), cloned more than 15 years ago (Leung et al., 1987), and the prolactin receptor (PRLR) whose sequence was determined soon after in our laboratory (Boutin et al., 1988). The hormone–receptor interactions between GH or PRL and their cognate receptors have been extensively studied by the group of Wells and De Vos at Genentech (Cunningham et al., 1989, Cunningham et al., 1990, Cunningham et al., 1991, Cunningham and Wells, 1989, Cunningham and Wells, 1991, Lowman et al., 1991, Wells, 1994 Bass et al., 1991) and our group (Rozakis-Adcock and Kelly, 1991, Rozakis-Adcock and Kelly, 1992, Goffin et al., 1992, Goffin et al., 1993, Goffin et al., 1994, Goffin et al., 1995, Goffin et al., 1996b, Rhee et al., 1995, Kinet et al., 1996, Kinet et al., 2001), respectively. These studies led to the identification of two binding sites on the ligand, each of which interacts with one receptor molecule to achieve an active trimeric complex involving one ligand bound to two receptor chains. Like PRL, PL can activate the PRL receptor. The molecular bases of PL functions remain less well understood since, to our knowledge, systematic mutational studies of the two putative binding sites remain to be performed. However, results from various point mutations argue for hormone–receptor interactions similar to what is known for the two other ligands (Gertler and Djiane, 2002). It was proposed some years ago that certain PLs activate the GHR receptor, nevertheless, this observation has recently been shown to be physiologically untrue since binding does not occur in a homologous context (hormone and receptor from the same species) (Gertler and Djiane, 2002). Interestingly, PLs might be unique among this hormone family in their ability to induce functional heterodimerization of PRLR and GHR, which may in part underline some of their specific roles (Herman et al., 2000, Gertler and Djiane, 2002).
PRL, PL and GH fold into a four α-helix bundle (De Vos et al., 1992, Goffin et al., 1995, Elkins et al., 2000). The most divergent region between these hormones is the N-terminus (Fig. 1). With respect to the first helix, the N-terminus contains five amino acids in human (and primate) GHs and PLs, which evolved from the same ancestral gene (Miller and Eberhardt, 1983). In hPRL, the N-terminal is 14 residues and contains an internal disulfide bond between cysteines 4 and 11, which is highly conserved among all PRLs (Sinha, 1995). Finally, in non-primate PLs, which evolved from the PRL lineage, the N-terminus is still longer by a few amino acids, for example, it contains 17 residues in ovine (o)PL. Based on the fact that PRLs, PLs and primate GHs are all able to activate the PRLR despite of their divergent N-terminus, conventionally, this region has been considered to be functionally not important (Goffin et al., 1996b). However, two observations suggest that this assumption probably needs to be re-examined. First, extensive comparisons of structure–function studies of PRL/PL/GH hormone shows that binding determinants (i.e. residues involved in receptor binding) to the PRL receptor differ by their chemical nature and are located at topologically non-equivalent positions (Goffin et al., 1996b). This led us to propose that the mechanisms of hormone–receptor interactions may differ at the residue level between these ligands, and hence, that the different N-termini of PRL, GH and PL may be involved in receptor binding in a ligand-specific manner. Second, the recent determination of the three-dimensional (3D) structure of the trimeric complex between oPL and the extracellular domain of the rat PRLR (i.e. the soluble PRL binding protein (PRLBP)) has clearly shown that the first N-terminal amino acids of oPL play a critical role in receptor binding (Elkins et al., 2000). Obviously, the interactions involving the N-terminus of oPL do not occur in the closely-related hGH–PRLR complex since hGH lacks homologous N-terminal residues (Fig. 1). This experimental observation not only validates our hypothesis that hormone–PRLR interactions display ligand-specific features (Goffin et al., 1996b, Kinet et al., 2001), but it also prompted us to investigate the functional involvement of the N-terminus in hPRL biological properties, including receptor binding and activation. Therefore, we first engineered two N-terminal deletions in hPRL, involving removal of the 9 first residues (mutant Δ1-9-hPRL) or the 14 first residues (Δ1-14-hPRL; Fig. 1). Δ1-9-hPRL mimics the shorter N-terminal end of hGH (Goffin et al., 1996b), whereas Δ1-14-hPRL is assumed to have lost entire N-terminal loop (helix 1 is predicted to start at residue 15; (Goffin et al., 1995)). The results obtained with these N-terminal deleted analogs encouraged us to also engineer and characterize the intermediate analogs, lacking 10–13 N-terminal residues (Δ1-10-hPRL, Δ1-11-hPRL, Δ1-12-hPRL, Δ1-13-hPRL).
Section snippets
Reagents
Culture media, fetal calf serum (FCS), geneticin (G-418), trypsine and glutamine were purchased from Gibco-Invitrogen (Grand Island, NY, USA). Luciferin and cell lysis buffer were from Promega (Madison, WI) and luciferase activity was measured in relative light units (RLU) (Lumat LB 9501, Berthold, Nashua, NH). Iodogen was purchased from Sigma and carrier-free Na[] was obtained from Pharmacia-Amersham (Buckinghamshire, UK). Salts were high grade purified chemicals purchased from Sigma or
Production and characterization of hPRL analogs in E. coli
All hPRL mutants used in this study were produced in bacteria as inclusion bodies as previously reported (Goffin et al., 1992, Goffin et al., 1994, Goffin et al., 1996a, Kinet et al., 1999). N-terminal deletion mutants refolded correctly as assessed by the absence of precipitated proteins upon the renaturing dialysis and by their apparent molecular mass, similar to that of WT hPRL. Interestingly, the number of multimers/dimers observed after protein refolding, as monitored by gel filtration,
Discussion
Physiological ligands of the somatogen receptor (GHR) are restricted to growth hormones, since somatogenic binding of some PLs has been recently shown to occur only in heterologous, but not in homologous interactions (Gertler and Djiane, 2002). All GHs are similar in terms of the length of their N-terminus, and alanine-scanning mutagenesis suggested the involvement of this region in binding to the somatogen receptor, since mutation of isoleucine 4 and, to a lesser extent, of phenylananine 1
Acknowledgements
We are grateful to Dr. Dominique Madern for help with CD analysis. We also acknowledge Christine Kayser for technical help and Claudine Coridun for secretarial assistance. This work was supported in part by Inserm and the Comité de Paris de la Ligue Nationale contre le Cancer. Sophie Bernichtein was supported initially by a student fellowship from the Ministry of Research and technology of France, then by the fellowships from Fondation pour la Recherche Médicale et la Ligue Nationale contre le
References (49)
- et al.
Evaluation of secondary structure of proteins from UV circular dichroism spectra using an unsupervised learning neural network
Protein Eng.
(1993) - et al.
A systematic mutational analysis of hormone-binding determinants in the human growth hormone receptor
Proc. Natl. Acad. Sci. U.S.A.
(1991) - et al.
Extrapituitary prolactin: distribution, regulation, functions, and clinical aspects
Endocr. Rev.
(1996) - et al.
S179D-hPRL, a pseudo-phosphorylated human prolactin analog, is an agonist and not an antagonist
Endocrinology
(2001) - et al.
New homologous bioassays for human lactogens show that agonism or antagonism of various analogs is a function of assay sensitivity
Endocrine
(2003) - et al.
Prolactin and its receptor: actions, signal transduction pathways and phenotypes observed in prolactin receptor knockout mice
Endocr. Rev.
(1998) - et al.
Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family
Cell
(1988) - et al.
A specific, high affinity, saturable binding site for the 16-kilodalton fragment of prolactin on capillary endothelial cells
Endocrinology
(1992) - et al.
The 16-kilodalton N-terminal fragment of human prolactin is a potent inhibitor of angiogenesis
Endocrinology
(1993) - et al.
High-resolution epitope mapping in hGH–receptor interactions by alanine-scanning mutagenesis
Science
(1989)