We identified the cDNAs of three functional rat H3 receptor isoforms (H3A, H3B, and H3C) and one nonfunctional truncated H3 receptor (H3T). The H3A, H3B, and H3C receptor isoforms vary in the length of their third intracellular loop; the H3B and H3C receptor lack 32 and 48 amino acids, respectively. Transient expression of the H3A, H3B, and H3C receptors in COS-7 cells results in high affinity binding for the H3 antagonist [125I]iodophenpropit, which is displaced by selective H3 agonists and antagonists. The three isoforms differentially couple to the Gi protein-dependent inhibition of adenylate cyclase or stimulation of p44/p42 mitogen activated protein kinase (MAPK), a new signaling pathway for the H3 receptor. Whereas the H3A receptor was less effective in inhibiting forskolin-induced cAMP production compared with the H3B or H3C receptor, this isoform was more effective in the stimulation of p44/p42 MAPK. The H3 receptor isoforms also displayed differential CNS expression in key areas involved in regulation of sensory, endocrine, and cognitive functions. A differential H3 receptor isoform expression was seen in, for example, hippocampus, where a characteristic dorsoventral distribution was revealed. Differential H3 receptor expression was also characteristic for the cerebellum, indicating possible histaminergic regulation of motor functions. The identification of these new H3 receptor isoforms and their specific signaling properties adds a new level of complexity to our understanding of the role of histamine, and the H3 receptor in brain function. The heterogeneous distribution of the isoforms suggests that H3 receptor isoform-specific regulation is important in several brain functions.

To investigate the molecular mechanism for stereospecific binding of agonists to g32-adrenergic receptors we used receptor models to identify potential binding sites for the (3-OH-group of the ligand, which defines the chiral center.Ser-165, located in transmembrane helix IV, and situated (, are often studied as a model system for the large superfamily of G-protein-coupled receptors. These receptors most likely contain seven transmembrane a-helices, and their topography has been verified using biochemical and immunological techniques to identify extra-and intracellular domains (1, 2). The binding of agonists to these receptors has been studied in much detail to understand the molecular mechanisms of ligand docking and receptor activation (3). A series of mutagenesis experiments plus the analysis of large numbers of ligands has allowed the identification of several of the amino acids involved in agonist binding to ,82-AR (reviewed in refs. 4 and 5). The current concept of agonist binding proposes that the positively charged nitrogen interacts with Asp-113 in transmembrane domain III (6, 7), and that the two catechol-OH-groups form hydrogen bonds with Ser-204 and Ser-207 in transmembrane domain V (8).Despite these advances, one of the key properties of agonists at these receptors has not been clarified: their stereospecific binding. Stereospecificity of 3-adrenergic agonists is defined by their (3-OH-group, which is located at the chiral center.(32-AR bind their agonists such as isoproterenol in a stereospecific manner, with the (-)isomer being about 40-times more potent than the (+)isomer, both in the high-affinity state (equivalent to the receptor/G-protein complex) and in the low-affinity state (9). In addition to its potential role in stereospecificity, the ,B-OH-group might also be involved in the agonistic properties of these compounds (3). The amino acids that interact with this (3-OH-group and may therefore be responsible for stereospecificity have not been identified.Apart from relatively low-resolution electron diffraction results obtained with rhodopsin (10), there are no biophysical data on G-protein-coupled receptors. Several authors have therefore developed computer models of G-protein-coupled receptors to predict their structure and modes of interaction with their ligands. The first group of such models uses the known structure of bacteriorhodopsin, which also contains seven transmembrane a-helices (11), to develop a backbone of the transmembrane helices of G-protein-coupled receptors (12, 13). The second group of models avoids the use of bacteriorhodopsin as a template because it is a developmentally probably unrelated archaebacterial proton pump and not a receptor (14). The latter models are based essentially on sequence alignments of many G-protein-coupled receptors and their hydrophobicity profiles or their polarity conserved positions (15, 16). Finally, both sets of information have often been used in attempts to obtain models of G-protein-coupled receptors (reviewed in refs. 17 an...

No abstract