br Materials and methods br Results br

Materials and methods
Results
Discussion Several histamine H3 receptor antagonists have progressed into clinical development for indications including Alzheimer's disease, schizophrenia, pain and narcolepsy. Each of the compounds studied herein has been extensively characterized in its own right (Barbier et al., 2004; Esbenshade et al., 2005; Fox et al., 2005; Giannoni et al., 2010; Sors et al., 2017; Medhurst et al., 2008), but few studies have compared their pharmacology in detail to determine common or discriminating properties that might be associated with in vivo efficacy. Using functional cAMP and GTPγS assays we confirmed that all of the investigational and approved histamine H3 receptor antagonists are inverse agonists of the receptor, displaying a broad range of affinities. There were no major discrepancies when comparing the affinities for the full length, H3-445 receptor and the major truncated isoform, H3-365; although all the showed the same trend to have lower affinity for the truncated isoform. This would be consistent with a higher level of constitutive receptor activation associated with this isoform in recombinant Ciclopirox (Bongers et al., 2007) and the confirmed inverse agonist activity of the compounds (Fig. 2), assuming a two-state model of receptor activation (viz. Canals et al., 2012). Kinetic analysis of non-equilibrium Ca2+ mobilization assays revealed two major classes of H3 receptor antagonist: slowly dissociating (viz. GSK189254, PF-3654746) and fast dissociating (viz. ABT-239, pitolisant, S 38093-2). JNJ-5207852 displayed an interesting profile, appearing pseudo-irreversible at the full-length H3-445 receptor, but having faster dissociation kinetics at the shorter H3-365 isoform. It was noteworthy that dissociation rates were not well correlated with equilibrium dissociation constants; pitolisant and PF-3654746 had similar pKB values but very different dissociation rates (Table 2). This discrepancy is at odds with long-held dogma that dissociation rate drives affinity, although our data are consistent with other recent studies of GPCR ligand kinetics that reveal that association rate may also vary markedly and contribute to variance in observed equilibrium affinity (Riddy et al., 2015; Mould et al., 2014). Whilst long receptor residence is often regarded as beneficial for antagonists (potentially improving duration of action and/or efficacy; Sykes et al., 2016; Seow et al., 2016; Strasser et al., 2017; Schuetz et al., 2017), a short receptor half-life apparently does not prevent pitolisant from exerting clinical benefit and of the compounds evaluated it is the only one to be successful in Phase III and progress on to the market (as Wakix, for narcolepsy; Syed, 2016; Calik, 2017). Likewise, rapidly dissociating ABT-239 (rapidly dissociating) appears to have better wakefulness-promoting efficacy for the same degree of histamine H3 receptor occupancy than slowly dissociating GSK189254 (Le et al., 2008). However, the overall pattern of histamine H3 receptor pharmacology reported here is broadly consistent with the reported in vivo efficacy of the same compounds (Bitner et al., 2011; Medhurst et al., 2007; Barbier et al., 2004), with one significant exception. Unlike the other agents that display affinity in the nanomolar or sub-nanomolar range (Table 2, Table 3), S 38093-2 is a rapidly dissociating, low-affinity histamine H3 receptor antagonist. Although this is consistent with previous reports (Sors et al., 2017), it is at odds with its in vivo pharmacology. S 38093-2 displays potent activity in rodent models of learning and memory, as well as neuropathic pain (Chaumette et al., 2018), with effective doses as low as 0.1–0.3 mg/kg (PO), whereas for markers more classically linked to histamine H3 receptor blockade (such as pre-frontal cortical histamine or acetylcholine release and wakefulness in rats), effects are only seen at doses of 10 mg/kg (IP; Panayi et al., 2017). Although H3 antagonist-induced wakefulness has been suggested to require a higher degree of receptor occupancy than pro-cognitive effects (Le et al., 2008), the 100-fold difference in effective doses led us to investigate whether other targets may be involved in the learning and memory effects. Pan-receptor profiling revealed that S 38093-2 has micromolar antagonist activity in a guinea-pig vas deferens assay of sigma receptor function, although when tested in an assay specifically targeted at the central sigma-1 receptor (mouse brain [3H]-(+)-pentazocine binding), S 38093-2 displayed significantly (30-fold) higher affinity for this target than for the histamine H3 receptor (Table 3). Indeed, when tested in an ex vivo mouse brain receptor occupancy study, S 38093-2 preferentially occupied the sigma-1 receptor population compared to histamine H3 receptors.