Human growth hormone receptor: cloning and expression of the full-length complementary DNA after site-directed inactivation of a cryptic bacterial promoter.

Growth hormone receptor is a cytokine-type receptor which is required for normal somatic growth and for numerous metabolic processes. Its complementary DNA (cDNA) has been isolated in various species leading to intensive studies to elucidate the mechanism of action of the growth hormone. However, serious difficulties have been reported in cloning in Escherichia coli, an intact full-length human cDNA. In this study, the cDNA is shown to contain a cryptic bacterial promoter driving inappropriate expression of a part of human growth hormone (hGH) receptor which is toxic for E. coli growth. Identification of this promoter and its inactivation by changing only one nucleotide led us to obtain stable bacterial clones containing a high copy number of full-length coding sequences. This molecular clone was used in a baculovirus/insect cell system to produce large amounts of glycosylated recombinant receptor. Binding studies with 125I-labelled hGH revealed an affinity constant of 2.8 x 10(9) M(-1), similar to that reported for the native liver receptor. This report described a general method of cloning which could be applied to similar unclonable cDNA fragments.

Evidence for N-glycosylation and ubiquitination of the prolactin receptor expressed in a baculovirus-insect cell system.

The molecular mass of the rabbit prolactin receptor (rbPRLR) deduced from cDNA cloning is 66 kDa. However, the molecular mass of the full-length receptor expressed in the insect Sf9 cells was found to be 94 kDa. In order to explain this discrepancy, we analyzed the possible post-translational modifications of the PRLR. Sf9 cells were infected with recombinant baculoviruses in the presence of tunicamycin, an inhibitor of N-glycosylation. Results showed that an additional approximately 9 kDa of the extracellular domain could be attributed to the N-glycosylation and another additional approximately 20 kDa covalent modification occurred in the cytoplasmic part of the receptor. Western blot analysis, using anti-ubiquitin antibodies, revealed that the rbPRLR was ubiquitinated in its cytoplasmic domain.

Purification and characterization of luteinizing hormone from the dromedary (Camelus dromedarius).

Luteinizing hormone (LH) has been purified from 150 dromedary pituitaries and its partial physicochemical, biological and immunological characterization has been achieved. Purification of the hormone was monitored by a porcine LH radioreceptor assay (RRA). In this system, the final camLH preparation exhibited an activity 0.6-fold that of highly purified porcine LH. The acid half-dissociation of camLH at equilibrium was observed at pH 4.2. A homologous camLH RRA was developed using the testicular plasma membrane fraction from prepubertal camels and radioiodinated, highly-purified camLH. Pituitary and chorionic gonadotropins (CG) from several mammalian species were compared to camLH in this system. The equine gonadotropins eLH and eCG were shown to be 6 times less potent in the camel RRA than in the porcine RRA, whereas the LH from other species exhibited similar activities in both systems. This particularity of camel LH receptors offers a new tool for the study of structural features of gonatropin interactions with their receptors.

Expression of the full-length rabbit prolactin receptor and its specific domains in baculovirus infected insect cells.

The prolactin receptor is a membrane protein mainly involved in the development of the mammary gland and in lactation in mammals. We used specific cDNA constructs and the insect/baculovirus expression system and produced independently and in large amounts several recombinant forms of the rabbit mammary gland prolactin receptor: the full-length receptor (L1, L2), a truncated membrane form (S), a secretable form of the extracellular domain (E) and two forms of the intracellular domain (I1, I2). Of these forms, the L1 and L2 are associated with the membrane fraction, the E is predominantly secreted into the medium and the I1 and I2 are expressed as soluble proteins and surprisingly, a great portion accumulates in the culture medium. The molecular mass (94 kDa) of the expressed full-length receptor corresponds to the translation product of the entire cDNA coding region. The receptor biochemically identified in the rabbit mammary gland is however much shorter. Thus, in the mammary gland, the receptor presumably undergoes post-translational modifications. The receptor forms L1, L2 and S bind prolactin with specificity and affinity similar to those reported for the native receptor. They also interact with two monoclonal antibodies, M110 and A917, specific for the native conformation of the hormone-binding site. The I1 and I2 forms do not bind prolactin, whereas the E form does. Thus, the hormone binding site is located in the extracellular domain which can function autonomously as a PRL-binding soluble protein. However, the E form binds prolactin with a higher affinity than the native receptor and it does not bind one of the two antireceptor monoclonal antibodies, known to be hormone binding-site specific. Thus, the conformation of the native receptor and that of the E form differ.

Expression of a recombinant human growth hormone binding protein in baculovirus/insect cell system.

An eucaryotic recombinant human growth hormone binding protein (rGHBP) was expressed in baculovirus-infected insect cells and purified by affinity chromatography from culture supernatant. This mannose-rich 34-kDa protein specifically bound human growth hormone (hGH) with the same affinity (kDa = 0.42 x 10(-9) M) than the 51.5 kDa GHBP we purified and characterised from human plasma (kDa = 1.1 x 10(-9) M). A high molecular form of the rGHBP was detected by silver-stained SDS-PAGE, Western blot (mAb 263), affinity cross-linking and Western ligand blot with 125I-hGH. Reduction experiments with beta-mercaptoethanol suggested that this form involved a disulfide bound between two rGHBPs.

Rapidly reversible binding of rabbit prolactin to the rabbit prolactin receptor accounts for the differences between homologous and heterologous binding.

The binding of radioiodinated rabbit (rb) prolactin (PRL) to rabbit mammary membranes is low and its affinity constant, 0.02 nM-1, calculated from heterologous inhibition assays, is about 300 times lower than that of ovine (o) PRL. Although the differences between homologous and heterologous binding are well documented in different species, the reasons for such differences are still unknown. Here we show that the low affinity of rbPRL for the native receptor does not affect its in vitro bioactivity compared with that of oPRL. We also show that rbPRL displays high specific binding to the baculovirus-expressed recombinant receptor and further establish that its lower affinity for binding to the homologous receptor is due to its faster and more complete dissociation compared with that of oPRL. Hormone binding affinity for full-length and carboxy-terminal truncated rbPRL receptor mutants expressed in mammalian or in baculovirus-infected cells was not affected by partial truncation of the cytoplasmic domain of the receptor, whereas the affinity for oPRL increased and that for rbPRL decreased upon truncation of both the cytoplasmic and membrane domains. The affinity of rbPRL for the native receptor is two orders of magnitude lower than that for the recombinant receptor. Affinity cross-linking and binding experiments showed that this difference in affinities is not related to selective cleavage of the native microsomal receptor during the binding reaction; however, this difference may be related to cell context-dependent differences in the oligomerization state of the receptor. Thus, obviously, the cloned receptor is alone sufficient for binding to rbPRL without requiring any receptor-associated protein. The lower affinity for rbPRL binding to its homologous receptor in comparison with higher affinity binding of oPRL to the same receptor is attributable to differences in their dissociation kinetics and in the conformational requirements of the receptor-hormone interaction site for binding to the two hormones.

[Origin of the FSH + LH double activity of equine chorionic gonadotropin (eCG/PMSG)]. / Origine de la double activité FSH + LH de la choriogonadotropine équine (eCG/PMSG).

The LH and FSH activities of equine choriogonadotropin (eCG) have been compared in several species with those of the highly purified homologous pituitary gonadotropins. The molar FSH/LH activity ratio of eCG determined by RRA is 0.20 in the pig, 0.25 in the rat and 0 in the horse. These data demonstrate the LH monospecificity of eCG in its own species as it is the case for hCG. We have also shown that equine LH exhibited a FSH-activity similar to that of eCG in the pig and in the rat but not in the horse. In the female rat, the binding activity to FSH receptors and the in vitro FSH activity of eCG are similar (about 0.005 X eFSH). In the male rat, the binding activity of eCG to FSH receptors is much higher (0.15 X eFSH) than its in vitro FSH activity (0.007 X eFSH). The behaviour of equine LH is similar to that of eCG in the four systems. Our data clearly indicate that the double activity of eCG is not unique to this molecule. Moreover, the double activity is not intrinsic to the eCG molecule since it depends on the species and the sex of the animal used as well as on the system used (RRA) or in vitro biological assay).

The intracellular domain of the rabbit prolactin receptor is able to promote the secretion of a passenger protein via an unusual secretory pathway in lepidopteran cells.

We have previously shown that the intracellular domain of the rabbit prolactin receptor (rbPRL-R), lacking typical signal sequences, was very efficiently secreted into the culture medium when expressed in the baculovirus-insect cell system. We have sought to take advantage of this characteristic for secreting cytoplasmic or nuclear proteins. We have constructed a series of recombinant viruses expressing a foreign gene product fused to the intracellular domain of rbPRL-R. Two passenger genes were used, one encoding a cytoplasmic protein (cyclin B) and the other a nuclear protein (cyclin A). The intracellular domain of rbPRL-R was able to promote the export of these two chimeric proteins with a very high efficiency. This new system should prove useful for secretion of proteins which do not require the post-translational modifications of the classical secretory pathway to be fully active.

Use of the immature rat uterotrophic assay for specific measurements of chorionic gonadotropins and follicle-stimulating hormones in vivo bioactivities.

The uterine weight growth stimulation by equine Chorionic Gonadotropin (eCG/PMSG) was found to occur at much lower eCG concentrations than ovarian growth. Human Chorionic Gonadotropin (hCG) which has only LH activity, was found to be as active as eCG in the uterotrophic assay whereas equine Luteinizing Hormone (eLH) which has dual LH+FSH activities like eCG, exhibited a much lower potency. In contrast to hCG, porcine and ovine LH as well as pFSH and oFSH exhibited no uterotrophic activity indicating that only gonadotropins with both LH activity and long half-lives are active alone in this assay. The FSH preparations were nevertheless found to trigger a dose-dependent response, but only in the presence of a subactive dose of hCG. The uterotrophic activity of hCG was found to be suppressed in ovariectomized immature rats and to be diminished after injection of GnRH antagonist suggesting an indirect pathway implicating the hypothalamo-pituitary complex. The data in this report together with the analysis of literature suggest that choriogonadotropins exert their stimulatory role on uterine growth by an indirect mechanism involving an increase in ovarian FSH receptors and FSH release by the pituitary. At the lowest concentrations of hCG, the increase in ovarian FSH receptors without endogenous FSH release is thought to be responsible for the sensitivity of the uterotrophic assay to exogenous FSHs. In conclusion, the immature rat uterotrophic assay is a sensitive and convenient assay for eCG and hCG as well as for FSHs in the presence of a sub-active dose of hCG.

Comparison of two reference preparations for horse chorionic gonadotrophin in four in-vivo and in-vitro assays.

A number of horse chorionic gonadotrophin (CG) preparations of different purities and from diverse sources have been compared in radioimmuno-, radioreceptor, in-vitro cell culture, and in-vivo assays. The relative activities of the great majority of the preparations tested were consistent in the 4 assay systems. Moreover, their relative activities in the 4 assays were consistent with those found for unfractionated plasmas. These preparations were therefore considered to represent the native form of hormone. The second International Reference Preparation (IRP2) was among the few preparations exhibiting discordant relative activities in the different assay systems. Its relative in-vivo activity was almost 50% lower than that found in the 3 other assays. This could be due to denaturation of the hormone during its preparation or to selection of isoform(s) not representative of the whole population of molecules. For standardization of horse CG preparations by in-vivo assay, IRP2 has proved to be a reliable standard. However, for the standardization of preparations by in-vitro methods, a standard giving consistent results in in-vivo and in-vitro assays must be used. The present report indicates that the NIH standard, but not IRP2, fulfils these requirements.

Inhibin activity in ram rete testis fluid: depression of plasma FSH and LH in the castrated and cryptorchid ram.

Ram "rete testis" fluid (RTF) routinely collected throughout the year has been used as a source of inhibin. The mean flow rate and mean concentration of spermatozoa in the fluid remained constant during the first 12 days of cannulation. More than 50 castrated or cryptorchid rams have been treated with low doses of steroid-free RTF over a 25-h blood sampling period. Human serum albumin was injected as a control. RTF depressed both FSH and LH plasma levels although the pattern was different for each hormone. There was no change in prolactin secretion. LH secretion was affected first while FSH remained unchanged in castrated and in cryptorchid rams. Thereafter, the maximum depression of FSH plasma levels occurred at a time when LH started to return or had returned to preinjection levels in the cryptorchid and castrated animals respectively. In the cryptorchid rams, RTF suppressed pulsatile LH secretion which was present before treatment but in the castrated animals, RTF lowered LH plasma levels which were constant and showed no pulsatile changes before treatment. Both FSH and LH inhibitory activities have been found in all active fractions obtained by purification of RTF. These activities are papain-sensitive and active fractions have a high apparent molecular weight (greater than or equal to 100 000) as shown by gel filtration and ultrafiltration. These and other results in the literature have lead to a re-definition of inhibin as a protein factor of gonadal origin able to depress plasma levels of FSH and LH, even at low doses.