Design, synthesis, and evaluation of a long-acting, potent analogue of gonadotropin-releasing hormone.

The design, synthesis, and biological evaluation of a gonadotropin-releasing hormone (GnRH) agonist, [D-Lys(6)(1,3,8-trihydroxy-6-carboxyanthraquinone)]GnRH ([D-Lys(6)(Emo)]GnRH), is described. Synthesis of this analogue was carried out in a homogeneous solution as well as on a polymer support. [D-Lys(6)(Emo)]GnRH was found to bind to rat pituitary GnRH receptors (IC(50) = 0.25 nM), to induce luteinizing hormone (LH) release (ED(50) = 27 pM), and to be devoid of any toxicity. This analogue also proved to be a very potent agonist in vivo and exhibited a prolonged bioactivity. Six hours after its administration to rats, LH levels were substantially higher than those of rats treated with a 10-fold higher dose of the parent peptide. Moreover, chronic treatment of adult male rats with [D-Lys(6)(Emo)]GnRH (0.1 nmol/rat) for one week resulted in a further decrease of the weight of the testes and prostate as compared to those of rats that were treated with a higher dose of [D-Lys(6)]GnRH (1 nmol/rat). The prolonged activity of [D-Lys(6)(Emo)]GnRH may be attributed to its emodic acid moiety, which enhances the binding affinity of the analogue to human serum albumin. Indeed, we found that emodic acid binds to human serum albumin almost completely at the examined range of concentrations.

Transcriptional regulation of the human GnRH II gene is mediated by a putative cAMP response element.

Human neuronal medulloblastoma cells (TE-671) were recently demonstrated to express the two forms of GnRH (GnRH-I and GnRH-II). We have used this cell line as a model system to demonstrate regulation of the human GnRH-II gene by cAMP. RT-PCR and Southern hybridization demonstrated that GnRH-II mRNA is strongly up-regulated ( approximately 6-fold) by (Bu)(2)cAMP. The concentration of GnRH-II that was released into the medium of TE-671 cells treated with the cAMP analog was significantly higher than that of the untreated cells. TE-671 cells that were stimulated by (Bu)(2)cAMP demonstrated morphological changes and strong immunoreactive GnRH-II staining in part of the cell population. After screening of the GnRH-II promoter sequence, we identified a putative cAMP response element consensus site. The GnRH-I and GnRH-II promoters were isolated by PCR using human genomic DNA and cloned into the luciferase reporter plasmid. By measuring the basal activity of the promoters that were transfected to TE-671 cells, we found a much stronger basal activity of the GnRH-II promoter compared with that of GnRH-I. Treatment of transfected TE-671 cells with (Bu)(2)cAMP resulted in a strong activation of the GnRH-II promoter compared with a modest activation of the GnRH-I promoter. To determine the functionality of this putative cAMP response element site, we mutated this site. TE-671 cells that were transfected with cAMP response element mutant constructs demonstrated a diminished basal activity of the GnRH-II promoter. Treatment of the transfected cells with the cAMP analog demonstrated a decrease to 0.03% of the activity of the mutated promoter compared with that of the wild type. These results clearly demonstrate the importance of the putative cAMP response element site for the basal activity as well as for induction of the GnRH-II promoter by cAMP.

Two isoforms of gonadotropin-releasing hormone are coexpressed in neuronal cell lines.

GnRH-I serves as the neuropeptide that regulates mammalian reproduction. Recently, several groups have identified in the brain of rodents, monkeys, and humans a second isoform of GnRH (GnRH-II) whose structure is 70% identical to that of GnRH-I. In this study we demonstrate for the first time human and mouse neuronal cell lines that express both GnRH-I and GnRH-II. Following the screening of several human neuronal cell lines by RT-PCR and Southern hybridization, we demonstrated that two cell lines, TE-671 medulloblastoma and LAN-1 neuroblastoma cells, coexpress messenger RNA encoding the two isoforms of GnRH. Nucleotide sequencing indicated that the complementary DNA fragments are identical to those of the known human GnRH-I and GnRH-II sequences. Extracts obtained from the TE-671 and LAN-1 cell lines as well as from the immortalized mouse hypothalamic GT1-7 neuronal cell line were found to contain the two isoforms of GnRH, which exhibited identical chromatographic properties as synthetic GnRH-I and GnRH-II, in HPLC followed by specific RIAs. Furthermore, double immunofluorescence studies demonstrated the two GnRH isoforms in LAN-1, TE-671, and GT1-7 cells. The identification of neuronal cell lines expressing both GnRH-I and GnRH-II provides tools for studying the differential regulation of gene expression and secretion and for studying the interaction between the two isoforms. Such studies may contribute to elucidation of the physiological functions of GnRH-II, which are still unknown.

Structure-activity studies of reduced-size gonadotropin-releasing hormone agonists derived from the sequence of an endothelin antagonist.

We have previously determined that Ac-D-Trp-Leu-Asp-Ile-Ile-Trp (peptide I), an endothelin antagonist, binds specifically (Ki = 1.9 microM) to the rat pituitary gonadotropin-releasing hormone (GnRH) receptor. Moreover, peptide I exhibits a GnRH agonistic activity, mediated directly by the GnRH receptor. We now report structure-activity studies of peptide I in respect to its interactions with the GnRH receptor. Our studies suggest that the bioactive conformation of peptide I, recognized by the GnRH receptor, is of a cyclic nature. Thus cyclic analogues of peptide I exhibit higher affinity to the GnRH receptor and increased agonistic potencies as compared to peptide I itself. A linear peptide, Ile-Ile-Trp-D-Trp-Leu-Asp, which presumably forms a similar cyclic conformation, was also shown to be a GnRH agonist. Intraperitoneal administration of Ac-Ile-Ile-Trp-D-Trp-Leu-Cys-OH (Ki = 0.32 microM), one of the cyclic hexapeptides that we have synthesized, to rats induces secretion of luteinizing hormone (LH) with a potency which is only 1 order of magnitude less than that of GnRH itself. Moreover, plasma levels of LH remained elevated for a longer period of time following the administration of the cyclic hexapeptide. This novel class of GnRH agonists may prove useful in the development of new therapeutics.

Design and synthesis of potent hexapeptide and heptapeptide gonadotropin-releasing hormone antagonists by truncation of a decapeptide analogue sequence.

A novel strategy for designing reduced-size analogues of the decapeptide gonadotropin-releasing hormone (GnRH) was developed. As opposed to previous attempts to delete residues from either of the peptide's termini, our approach is based upon the known importance of both C- and N-terminals of GnRH analogues for receptor recognition, whereas the central part of the molecule is replaced by a short spacer. The present truncation strategy was successful for generation of reduced-size hexapeptide and heptapeptide antagonists possessing potent antagonistic capacity. The same methodology was not suitable for the generation of reduced-size agonists, suggesting different conformational characteristics for GnRH agonists and antagonists. A heptapeptide antagonist designed by this method was shown to inhibit serum levels of luteinizing hormone in castrated rats in vivo. Structure-activity studies suggested that the structural preferences for GnRH receptor recognition are similar to those reported for decapeptide antagonists. Our studies resulted in a heptapeptide GnRH antagonist (Ac-D-Nal2-D-Cpa-D-Pal-Gly-Arg-Pro-D-Ala-NH2) with high receptor binding affinity (IC50 = 7 nM), as compared to that of GnRH itself (IC50 = 2 nM). The highest affinity of a hexapeptide antagonist that we have synthesized was somewhat lower (IC50 = 45 nM).

Hexapeptide and cyclic pentapeptide endothelin antagonists directly activate pituitary gonadotropin-releasing hormone receptors.

In the course of our studies toward the development of novel analogs of the decapeptide gonadotropin releasing hormone (GnRH), we have examined a hexapeptide that is an antagonist of endothelin (ET). It was found that this peptide, Ac-D-Trp-Leu-Asp-Ile-Ile-Trp (peptide 1), binds specifically to the pituitary GnRH receptor. Moreover, peptide 1 exhibits a GnRH agonistic activity (i.e., it induces luteinizing hormone release from rat pituitary). This activity is mediated directly by the GnRH receptor and is suppressed by a GnRH antagonist. Removal of the acetyl group of peptide 1 results in a hexapeptide (peptide 2) with binding properties similar to those of GnRH but with a diminished affinity toward the ET receptor. Several other ET antagonists were screened for a potential interaction with the GnRH receptor. Two of these, the hexapeptide PD145065 and the cyclic pentapeptide BQ-123, expressed GnRH agonistic activity at micromolar concentrations in vitro. BQ-123, previously approved for trials on humans as an ET antagonist, is demonstrated to act in vivo as a GnRH agonist, in a dose that was demonstrated previously as the minimal required dose for significant ET antagonism. The GnRH agonistic activity of ET antagonists may therefore result in interference with the physiological control of the reproductive system. Such effects may be most severe when ET antagonists are used chronically. Thus, the major practical message of this study is the need to circumvent the potential side effects of ET antagonist-based drugs.

The gonadotropin-releasing hormone family of neuropeptides in the brain of human, bovine and rat: identification of a third isoform.

The mammalian gonadotropin-releasing hormone (GnRH-I), which regulates reproduction, was the first isoform of GnRH that was identified in mammals. Recently, we and others have demonstrated the existence of a second isoform of GnRH in the brain of mammals. The presence of a third isoform of GnRH, GnRH-III, in the brain of mammals is reported herein. GnRH-III, extracted from the brain of bovine and human, was purified by high performance liquid chromatography, using two distinct elution programs. In both, GnRH-III was eluted at the same positions as synthetic salmon GnRH, as demonstrated by radioimmunoassay. The luteinizing hormone-releasing activity of purified GnRH-III, using dispersed rat pituitary cells, was found to be similar to that of synthetic salmon GnRH. The total amount of GnRH-III, determined by radioimmunoassay, in the hypothalamus and midbrain of humans and calves is similar to that of GnRH-I. Immunohistochemical studies demonstrated GnRH-III-containing neurons in the hypothalamus and midbrain of human and GnRH-III fibers in the median eminence of rats. The distribution of GnRH-III in the brain suggests that in addition to a putative function as a neurohormone at the hypothalamic-pituitary axis, GnRH-III may have other functions. Our present results suggest that multiple isoforms of GnRH are present in the brain of mammals, and further studies are required in order to elucidate their biological functions.

Synthesis and bioactivity of fatty acid-conjugated GnRH derivatives.

Transdermal delivery of peptidic drugs is usually inefficient, notably due to their hydrophilic character which makes it difficult to cross the hydrophobic layers of the skin. In order to obtain transdermally deliverable analogs of gonadotropin releasing hormone (GnRH), we have synthesized related hydrophobic derivatives by attaching various aliphatic acids to the N(epsilon)-amino side chain of [D-Lys]6GnRH, a superactive GnRH agonist. It was found that the affinity toward the GnRH receptor gradually decrease with increased hydrophobicity, i.e. increase in chain length of the attached aliphatic acid. Nevertheless, analogs with 12-carbon or shorter aliphatic acids were shown to be GnRH superagonists, with in vitro and in vivo potencies similar to that of [D-Lys]6GnRH. [D-Lys-lauryl]6GnRH was shown to have a longer duration of action in vivo, as compared to [D-Lys]6GnRH. The transdermal penetration of the peptides was evaluated by in vivo functional experiments in rats. According to these studies the efficiency of penetration is gradually lowered in increasingly hydrophobic analogs. These results are discussed with respect to the circular dichroism spectra of the peptides in trifluoroethanol. The spectra of the aliphatic acid-conjugated superagonists examined do not express a significant tendency towards a beta-turn conformation, typical of GnRH and its agonists. This finding contradict previous publications which suggested a correlation between the conformations of GnRH analogs in trifluoroethanol and their biological activities.

A second isoform of gonadotropin-releasing hormone is present in the brain of human and rodents.

Gonadotropin-releasing hormone-I (GnRH-I), present in the mammalian hypothalamus, regulates reproduction. In this study we demonstrate, for the first time, that an additional isoform of GnRH, [His5, Trp7, Tyr8] GnRH-I (GnRH-II) is present in the brain of the mouse, rat and human. Human and rat brain extracts contain two isoforms of GnRH, GnRH-I and GnRH-II, which exhibited identical chromatographic properties to the respective synthetic peptides, in high performance liquid chromatography. Using immunohistochemical techniques we have found that GnRH-II is present in neuronal cells that are localized mainly in the periaqueductal area as well as in the oculomotor and red nuclei of the midbrain. It is of interest to note that in the hypogonadal mouse, although the GnRH-I gene is deleted, GnRH-II is present. Substantial concentrations of GnRH-II are also present in the hypothalamus and stored in the human pituitary stalk or in the mouse median eminence. By using reverse transcription (RT)-PCR we have also found that while GnRH-II is not expressed in the cerebellum, it is expressed in all three structures of the brain stem: midbrain, pons and medulla oblongata.

Expression of the gene for the receptor of gonadotropin-releasing hormone in the rat mammary gland.

Recent findings have demonstrated that the GnRH gene is expressed in the mammary gland of pregnant and lactating rats but not of virgin rats. Indeed, significant concentrations of biologically active GnRH have been found in milk of human, cow, sheep and rat. We have, therefore, looked for expression of the GnRH receptor in the rat mammary gland. By reverse transcription (RT)-PCR amplification, we have demonstrated the presence of GnRH receptor mRNA in mammary gland samples derived from virgin, pregnant and lactating rats. The GnRH receptor transcript cloned from the mammary gland was sequenced and found to have an identical coding region to the one cloned from the pituitary gland. In addition, we have found that the mammary gland, as the pituitary gland, contains at least two transcripts having the same coding region but different 5' non-coding regions. Binding studies, however, could demonstrate only low-affinity binding sites. These results, therefore, suggest that the regulation of the GnRH receptor occurs posttranscriptionally rather than at the level of transcription.

The gene for the neuropeptide gonadotropin-releasing hormone is expressed in the mammary gland of lactating rats.

The high concentration of gonadotropin-releasing hormone (GnRH) in milk of several species implies that the mammary gland is either a site of synthesis for this neuropeptide or that it is efficiently concentrated from plasma by this organ. By PCR amplification of mammary gland cDNA, we have demonstrated expression of the mRNA for GnRH. The GnRH mRNA was present in the mammary gland of pregnant and lactating rats but not of virgin rats, implying that expression of the GnRH gene is activated during pregnancy, probably by prolactin. In contrast, actin mRNA was evident in all the preparations of mammary glands. Since GnRH is also known to be synthesized by the placenta, it is likely that the placenta and the mammary gland are complementary units by which the mother exercises control over the development and the metabolism of the infant during pregnancy as well as after parturition. In addition, GnRH synthesized by the mammary gland may also affect the mother by a paracrine and/or an endocrine mechanism.

A differential effect of trypsin on pituitary gonadotropin-releasing hormone receptors from intact and ovariectomized rats. Evidence for the existence of two distinct receptor populations.

The present report demonstrates that pituitary receptors for gonadotropin-releasing hormone generated following ovariectomy are susceptible to the action of trypsin while those present in intact metestrous females are not. The action of trypsin of gonadotropin-releasing hormone receptors in pituitary cells derived from ovariectomized rats was dependent on time and dose. Trypsin exerted a maximal effect of 40-50% reduction without causing full abolition of the binding activity, indicating that two populations of gonadotropin-releasing hormone receptors can be differentiated in pituitaries derived from ovariectomized rats: one which is sensitive to limited proteolysis by trypsin and another which is resistant to it. Scatchard analysis of gonadotropin-releasing hormone antagonist binding to homogenate of trypsin-treated hemipituitaries derived from ovariectomized rats, revealed that trypsin reduced receptor density while slightly increasing binding affinity. The in vitro responsiveness of hemipituitaries from ovariectomized rats to gonadotropin-releasing hormone and to K+-induced depolarization was reduced following trypsin treatment, while that of hemipituitaries derived from metestrous females was not affected or increased. Trypsin reduced the maximal response level by 30-50% but did not modify the median effective dose of gonadotropin-releasing hormone (1 nM). The reduced gonadotropin-releasing hormone receptor density and the lower responsiveness caused by trypsin in pituitary tissue from ovariectomized rats are not due to a selective loss of gonadotrophs, as trypsin treatment did not affect the lutropin content of the glands. These findings indicate that two populations of gonadotropin-releasing hormone receptors can be distinguished on the basis of their susceptibility or resistance to limited proteolysis by trypsin and suggest that these two gonadotropin-releasing hormone receptor populations are reciprocally affected by ovariectomy.

Release of thyrotropin releasing hormone into hypophysial portal blood is high relative to other neuropeptides and may be related to prolactin secretion.

The amount of immunoreactive TRH released into hypophysial portal blood of female rats was about 2 orders of magnitude greater than gonadotropin releasing hormone and somatostatin. The turnover of TRH, as high as 80% of the total hypothalamic content per hour, was also much greater than that of any other known peptide. TRH release increased during the expected proestrous surge of prolactin and also in some animals during suckling.