Non-peptidal MCH1-R antagonists have been the topic of several patents and publications in recent years, indicating the fervor with which research in this area has been pursued. Initial reviews in the area have documented these endeavors, which have laid a solid foundation for recent discoveries of MCH1-R antagonists which demonstrate in vivo efficacy [53-56].
The first non-peptide MCH1-R antagonist, T-226296 (5, Ki = 5.5 nM) showed good selectivity over other homologous receptors such as MCH2-R, somatostatin (sst1-sst5), opioid, and urotensin II . Oral administration (30 mpk) suppressed the orexigenic effect of exogenous MCH by > 90% in lean rats, consistent with the in vivo results of peptidal antagonist studies. Structural variations in which the tetrahydronaphthyl group has been replaced with a ^ara-substituted phenyl have been recently described .
A second small molecule antagonist, SNAP-7941 (6, Kb = 0.5 nM), demonstrated similar effects to those of T-226296 upon intraperitoneal injection . Specifically, i.p. administration to lean rats suppressed the orexigenic effect induced by i.c.v. administration of MCH. Chronic administration to diet induced obese (DIO) rats (10 mpk, b.i.d.) suppressed food intake, providing a 26% weight reduction over 28 days (relative to controls). This contrasted with D-fenfluramine treatment wherein a pronounced hyperphagia and weight loss over 7 days was followed by a rebound in both by day 14. Though neither T-226296 nor SNAP-7941 were tested in MCH null mice to confirm that the observed effects are MCH1-R specific, radiolabelled SNAP-7941 was shown to specifically bind to MCH1-R in several brain sections. In conjunction with the anorectic effects, SNAP-7941 also exhibited an-xiolytic and antidepressant properties in forced swim and Vogel Conflict tests . Derivatives of SNAP-7941 lacking chirality have recently been disclosed [61,62].
These seminal contributions were followed by more recent reports of in vivo efficacy demonstrated by GW-803430 (7) [63,64]. Derived from a lead structurally similar to T-226296, a homology model was used to highlight key pharmacophore interactions as shown below (Fig. 1).
Lipophilic binding -s Cl pocket
Fig. 1. Pharmacophore model for MCH1-R antagonists.
Lipophilic binding -s Cl pocket
Biaryl surrogates wherein an isosteric amide replacement is fused to a hetero-biaryl were explored, along with modifications at the distal amine. GW-803430 has shown oral efficacy in the AKR mouse, a model prone to diet induced obesity, causing a dose-dependent weight loss after 12 days of 7.4% and 13.3% (0.3 and 3 mpk, respectively) with no rebound observed upon prolonged dosing. In contrast, sibutramine induced a weight loss of 3.2% in the same study.
Indazoles such as 8 have also exhibited oral efficacy in diet induced obese mice (10 and 30 mpk, b.i.d.) over 14 days, providing an 8-15% dose dependent weight loss . Comparable effects were initially seen with D-fenfluramine treatment, however a slight rebound in body weight change was observed during the final week of treatment. Consistent with the phenotype exhibited by MCH1-R, subjects treated with 8 did not have altered food intake relative to control, suggesting an alteration of energy expenditure as the causative factor in weight loss. DEXA (dual-energy X-ray absorptiometry) scanning analysis of body composition indicated a significant reduction in fat mass of the treated animals while lean mass was unaffected.
Oral efficacy in rodent models on a high-fat diet was also achieved by ATC-0175 9, which provided a 10% weight reduction during a 4-day feeding cycle (45 mpk) relative to a sibutramine control [66,67]. Anxiolytic activity was also demonstrated in a number of rodent anxiety models. In this and related structural series, the aminoquinazoline and aryl amide can be linked by a variety of structures including piperidyl and cyclohexyl moieties of differing chain lengths. The czs-1,4-cyclohexyl derivatives confer improved selectivity over Y5 and a2a receptors.
Recently, 10 (K = 2.1 nM) was disclosed as an orally active MCH1-R antagonist [68,69]. Employing a bicycloalkane as an aryl surrogate to circumvent potential mutagenic liabilities, improvements in degree and duration of receptor occupancy were also observed via an ex-vivo binding assay. This assay facilitated medium throughput screening of drug occupancy at MCH1-R in rodent models, and could be measured at several timepoints post-dose. Correlation between receptor coverage and efficacy in a DIO mouse model served as an important screening tool. Oral dosing of DIO mice with 10 (30 mpk, p.o.) provided a 22% reduction in food intake over 24 h relative to controls. Similar bicycloheptyl derivatives exhibiting less extensive receptor occupancy such as 11 failed to demonstrate efficacy in the DIO mouse model.
Another urea-derived MCH1-R antagonist is compound 12, a diaryl imidazolone core appended with a sidechain containing a basic nitrogen atom [10-12]. Structural variations covered in this series of patent applications include benzimidazole, ben-zothiazole, benzofuran and indole derived ureas. IC50 values are reported between 1 nM and 1 mM, with the specified compound reducing milk consumption by 58% in a fasted mouse model (10 mpk, p.o.). In a related structural series, 13 was shown to reduce milk consumption by 64% in a similar model, though higher dosing was performed .
The biaryl urea motif has also been exploited as a structural feature in MCH1-R antagonists as demonstrated by 14 (IC50 = 8 nM) which resulted from a combination of MCH1 receptor modeling and structural input from D2 and D3 receptor ligands as well as other known MCH1-R antagonists such as T-226296 . The dopamine ligands were chosen due to the physicochemical similarity of the D2 and D3 binding sites to that of MCH1-R. Considerable structural tolerance was observed in the aliphatic amine region, with side chain homologation and steric congestion at the terminus improving affinity. Amides and oxadiazoles served as urea replacements, however disruption of planarity in the core was detrimental, as was methylation of the urea nitrogen atoms. In vivo activity of a related truncated amide 15 was demonstrated in rats (10 mpk, i.p.), with a reduction in cumulative food intake over 6h .
Another group of patents details an aminoquinoline series which has shown efficacy in rats [76,77]. Piperazinyl quinoline 16 reduced cumulative food intake by 20% over 6h in rats (50 mpk). Variations such as acyclic diamines in lieu of pip-erazines, and homologated phenoxy acetamides with electron poor para-substi-tuents were also efficacious.
Though no in vivo efficacy has been reported, differentially substituted amino-quinolines such as 17 have been recently discovered as MCH1-R antagonists . Compound 17 was the culmination of SAR studies in which the pyrrolidyl side-chains exhibit an optimal combination of functional activity and CNS penetration relative to the acyclic benzylamine or benzamide derivatives. Hydrophobic subs-tituents on the terminal aryl group imparted enhanced activity relative to derivatives such as acetamides. The (S)-enantiomer (IC50 = 0.9 nM) provided a 40-fold increase in binding affinity relative to the (R)-configuration. Importantly, the pharmacokinetic profile of 17 in DIO mice was shown to be excellent, with a brain AUC> 17 mMh (20-fold relative to plasma AUC) at 10mpk p.o.. This contrasts with 10-fold lower brain levels exhibited by 18, which is devoid of the geminal difluoro group.
Aminoquinoline 19 was discovered as the result of a virtual screening approach involving substructure, similarity and homology models based on a set of published MCH1-R antagonists . Hits obtained via screening of over 615,000 commercial entities were then narrowed to a subset based on assessments of druglikeness such as molecular weight, ClogP and polar surface area as well as synthetic facility. Upon assay of this subset, 19 was identified as having an IC50 = 55 nM along with favorable physicochemical properties. Further analysis of 19 in terms of proposed binding mode was performed using a homology model derived from the crystal structure of rhodopsin, which showed good similarity with the transmembrane helical region of MCH1-R (Fig. 2). The following three interactions between 19 and the postulated binding site are deemed crucial: (1) a salt bridge between the distal piperazine nitrogen atom and Asp172, (2) a hydrogen bond between the amide carbonyl and Gln325, and (3) an aromatic binding interaction between the
chlorophenyl moiety and several Phe residues from helices 5 and 6. Consistent with the model of key pharmacophore interactions indicated by studies using 7 (vide supra), the importance of these receptor binding interactions for other MCH1-R antagonists is evident. Subsequent lead optimization was performed via conventional synthesis-based SAR, including probes of electronic and steric requirements on the aryl ring (relatively large, electron withdrawing para substituents were preferred) . Piperazine replacements such as pyrrolidines and acyclic amines (20) improved potency (IC50 = 11 nM) and selectivity versus other GPCRs such as 5-HT subtypes, D2 and ala. Though in vivo data has yet to be reported, these results demonstrate the utility of ligand-based virtual screening as an efficient approach to hit generation for GCPR targets.
Ring contracted variants of the quinazoline heteroaryl derivatives containing benzimidazoles have also exhibited feeding effects [81,82]. In rats, a dose-dependent (10-30 mpk) decrease in MCH-stimulated food intake was observed upon administration of 21.
Isosteric replacements for the amide bond have been incorporated into simpler aryl amide compounds such as LY-2049255 (22) and 23. LY-2049255 (^ = 1.9 nM) uses the oxadiazole as a central core upon which linkers to a basic nitrogen atom are attached . Though no alteration of unstimulated food intake was seen, 22 reduced MCH-stimulated food intake up to 6 h post-dose (82nmol, i.c.v.). Aryl tetrazoles such as 23 were derived from a library synthesis, in which structural modifications indicated that the piperazine and tetrazole were both crucial for activity . Substitution at the meta- or ^ara-positions on the benzylic aryl group and absolute configuration were important for increased potency. Initial in vivo activity was demonstrated at 10 and 30 mpk (i.p.) in a fasted rat model at 1 h postdose, however efficacy was only observed at 30 mpk after 2 h.
Was this article helpful?