A novel mammalian receptor for the evolutionarily conserved type II GnRH

Abstract

Mammalian gonadotropin-releasing hormone (GnRH I: pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2) stimulates pituitary gonadotropin secretion, which in turn stimulates the gonads. Whereas a hypothalamic form of GnRH of variable structure (designated type I) had been shown to regulate reproduction through a cognate type I receptor, it has recently become evident that most vertebrates have one or two other forms of GnRH. One of these, designated type II GnRH (GnRH II: pGlu-His-Ser-His-Gly-Trp-Tyr-Pro-Gly-NH2), is conserved from fish to man and is widely distributed in the brain, suggesting important neuromodulatory functions such as regulating K+ channels and stimulating sexual arousal. We now report the cloning of a type II GnRH receptor from marmoset cDNA. The receptor has only 41% identity with the type I receptor and, unlike the type I receptor, has a carboxyl-terminal tail. The receptor is highly selective for GnRH II. As with the type I receptor, it couples to G(alpha)q/11 and also activates extracellular signal-regulated kinase (ERK1/2) but differs in activating p38 mitogen activated protein (MAP) kinase. The type II receptor is more widely distributed than the type I receptor and is expressed throughout the brain, including areas associated with sexual arousal, and in diverse non-neural and reproductive tissues, suggesting a variety of functions. Surprisingly, the type II receptor is expressed in the majority of gonadotropes. The presence of two GnRH receptors in gonadotropes, together with the differences in their signaling, suggests different roles in gonadotrope functioning.

Figure 1

Expression of type II GnRH receptor in pituitary and brain. Staining is shown in human (a), mouse (b), and sheep (c–f) pituitary. Sheep pituitary staining was neutralized by type II receptor peptide (d) but not by type I receptor peptide (e). The specificity of the staining was also indicated by the demonstration of similar staining by antisera raised in three other rabbits to the type II receptor peptide and an absence of staining by preimmune serum from all of these rabbits. Moreover, the hemocyanin carrier protein failed to neutralize staining. No staining was present with preimmune serum (not shown). Dual staining with type II receptor antiserum (black) and LH antiserum (red) shows colocalization (f). Type II GnRH receptor expressing neurones in the non-human primate brain by immunocytochemical staining is shown in panels g–l. Type II GnRH receptor-positive neurones are seen in the basal nucleus of Meynert (g), in the medial preoptic area (j) of an adult cynomologus monkey (Macaca fascicularis), and the amygdala (k) and periventricular region of the hypothalamus (l) of a fetal rhesus monkey (Macaca mulatta) at embryonic day 70. Note that the confocal photomicrographs with immunofluorescent staining show that the type II GnRH receptor-positive neurones (g, red-Rhodamine) in the basal nucleus of Meynert were also positive to mammalian GnRH I ligand (h, green-FITC). Colocalization is seen as yellow (i). In contrast, type II GnRH receptor-positive neurones in the medial preoptic area (j, red-Rhodamine), amygdala (k), and periventricular area (l) were negative for mammalian GnRH I ligand (data not shown). In k and l, immunoreactive products were visualized with DAB (brown). Scale bar: 10 μm for a–i; 20 μm for j–l. These brain areas also stained positive with antiserum raised against the carboxyl-terminal tail of the type II GnRH receptor (data not shown). Staining of type II GnRH receptor was neutralized in all tissues by incubation with type II peptide immunogen but not by type I peptide. Type II GnRH receptor-positive neurones were also seen in extrahypothalamic regions, such as medial septum, bed nucleus of the stria terminalis, substantia innominata, claustrum, amygdala, and putamen, and in the hypothalamic regions, such as supraoptic nucleus, ventromedial nucleus, and dorsomedial nucleus (data not shown).

Figure 2

Alignment of the amino acid sequences of the marmoset type II and human type I GnRH receptors. Conserved residues are shaded. α-helical regions predicted by homology modeling with the rhodopsin crystal structure are indicated. These helices will encompass the membrane spanning regions. A putative glycosylation site (○) and disulphide bridge (●) are indicated. Ser or Thr residues occurring in putative protein kinase C (†), casein kinase II (‡), and cAMP-/cGMP-dependent kinase (§) phosphorylation sites are indicated. Numerical residue annotation refers to marmoset type II sequence.

Figure 3

Receptor binding (a) and inositol phosphate production (b) of mammalian GnRH I (○) and GnRH II (●) in COS-7 cells transfected with marmoset type II receptor (Left) and human type I receptor (Right). Stimulation of inositol phosphate by type I receptor antagonist 135-18 (□) at the type II receptor is also shown. Error bars represent SEM of three to six separate experiments.

Figure 4

Activation of ERK2 and p38α MAP kinases by type I (open bars) and type II (filled bars) GnRH receptors in COS-7 cells. Stimulation of type I (a) and II (b) GnRH receptors with mammalian GnRH I, GnRH II, and antagonist 135-18. Inset panels depict anti-phospho-ERK2 immunoblotting of anti-myc COS-7 cell immunoprecipitates. (c Left) selective and time-dependent activation of p38α by GnRH II stimulation (100 nM) of type II GnRH receptor. Each bar represents the mean ± SEM for fold p38α stimulation measured in the cells for the specified time of ligand stimulation. Open bars represent p38α stimulation by treatment of COS-7 transfected with type I receptor and stimulated with GnRH I. Filled bars represent p38α stimulation by GnRH II treatment of COS-7 cells transfected with type II GnRH receptor. The Right panel demonstrates analogous data, but each bar (open, type I receptor stimulated by GnRH I; filled, type II receptor stimulated by GnRH II) represents mean ± SEM for phosphorylation of ERK2.

Figure 5

Expression of type II GnRH receptor in marmoset and human tissues. (a) RT-PCR was carried out with specific primers on cDNA prepared from marmoset RNA isolated from various tissues. PCR products were fractionated by size on agarose gels. Type II GnRH receptor levels were normalized to actin RNA and represented as the log of the RNA expression relative to pituitary. Hatched bars indicate marmoset brain tissues, solid bars indicate marmoset reproductive tissues, whereas open bars indicate other marmoset tissues. (b and c) Expression of the type II GnRH receptor in human tissues was examined in Northern blots of mRNA (CLONTECH) by hybridization with ³²P-labeled human exon 1; (b) mRNA from human cerebellum (lane 1), cerebral cortex (lane 2), medulla (lane 3), spinal cord (lane 4), occipital pole (lane 5), frontal lobe (lane 6), temporal lobe (lane 7), and putamen (lane 8); (c) mRNA from heart (lane 1), whole brain (lane 2), placenta (lane 3), lung (lane 4), liver (lane 5), skeletal muscle (lane 6), kidney (lane 7), and pancreas (lane 8). Another blot showed moderate expression in the amygdala and low expression in caudate nucleus, corpus callosum, hippocampus, substantia nigra, subthalamic nucleus, and thalamus (data not shown). The exon I probe is specific for the type II GnRH receptor and will not detect the ribonucleoprotein transcribed on the opposite strand (17, 18) because its transcript terminates with polyadenylation before exon I.