Dopamine has been implicated as the primary neurotransmitter associated with the psychomotor stimulant and reinforcing effects of many drugs of abuse, such as cocaine. These findings have resulted in intensive efforts to characterize and elucidate the roles of the various dopamine receptor subtypes in the pharmacology and addiction liability of these addictive drugs. In this pursuit, the dopamine D3R subtype has been intensively targeted. However, definitive behavioral investigations have been hampered by the lack of highly selective D3R agonists and antagonists. We initially used the classic D3R antagonist NGB 2904 as the template for our structural modifications to elucidate SAR and develop novel and selective D3R antagonists and partial agonists with drug-like physicochemical properties. NGB 2904 blocks cocaine-induced reinstatement of drug seeking behavior, an animal model of relapse. All NGB 2904 analogues included either a 2,3-dichloro- or 2-methoxy-substituted-phenylpiperazine, a four carbon linking chain with varying saturation (butyl, hydroxybutyl, and trans butenyl) and a terminal arylcarboxamide. Evaluation for in vitro binding in HEK 293 cells transfected with human D2, D3, or D4 receptor cDNAs resulted in D3R binding affinities ranging from Ki=0.3-100 nM. The most potent and selective analogs in this series initially demonstrated D3R/D2R selectivity of >100 and a D3R/D4 selectivity of >1000. Functional evaluation in vitro using a mitogenesis assay in D3R or D2R transfected CHO cells demonstrated that these compounds were either potent antagonists or partial agonists at D3Rs and were selective over D2Rs, in this functional assay. In binding studies, SAR demonstrated that the trans-butenyl linker provided additional D3R selectivity as compared to the saturated linking chain. Moreover, addition of a hydroxy (OH) group in the 2- or 3-position of the butyl linker also gave several highly selective and potent D3R antagonists or partial agonists. Further, replacement of the sterically bulky aryl ring system with various heteroaryl groups served to retain high affinity and selectivity for D3R, while decreasing lipophilicity. To this end we discovered very selective D3R antagonists and partial agonists with D3R/D2R-selectivites reaching 400-fold. In addition, several of these analogues have been further screened for binding in additional receptors and ion channels and did not show significant binding affinities at any of these other targets, highlighting that these agents are some of the most potent and selective D3R-antagonists and partial agonists reported to date. Further, the (+)- and (-)-enantiomers of one of these 3-OH analogues, PG648, were synthesized and demonstrated enantioselectivity at D3R (>10-fold), but not significantly (<2-fold) at D2Rs. This was the first demonstration of enantioselectivity of a D3R antagonist and further chimera studies, with these enantiomers, identified an extracellular loop (E1) region that appears to differ between D3R and D2R. The latter goal of reducing lipophilicity of the most potent agents was to improve physicochemical properties that would provide a more favorable pharmacokinetic/bioavailability profile than the currently existing D3R agents. Several of these analogues have been evaluated for pharmacokinetics, blood brain barrier penetration, and for potential metabolic pathways for degradation in vivo. In addition, several of the most potent and selective compounds of this series have been synthesized in multi-gram quantities and have been evaluated in numerous animal models of cocaine and methamphetamine abuse, in both rodents and non human primates, with PG01037, PG 619, PG 648 and several other analogues. Our newest series of analogues replaces the 3-OH group in the butyl linking chain with a F-group, and many of these compounds show favorable pharmacological profiles in vitro, with several compounds demonstrating D3R/D2R-selectivites >1000-fold. The enantiomers of our lead F-analogue, BAK2-66, were prepared, although robust enantioselectivity was not apparent. A chimera study showed the EL1 appears to play an important role in both binding and D3R selectivity. Recently, the D3R protein was crystallized and a computational model was created using the crystal coordinates. R-PG648 was docked in this D3R model and the homologous D2R model and significant binding domain differences have been identified that suggest the D3R has a secondary binding pocket that includes the extracellular loops EL1 and EL2. This distinction from D2Rs appears to be responsible for the binding selectivities of PG 648 and its F-analogues. In an effort to further explore the role of the pharmacophoric components of R-PG648 in D3R binding selectivity and efficacy, we systematically deconstructed the full-length molecules into components that we refer to as synthons. We hypothesized that the 2,3-diCl- or 2-OCH3-substituted-4-phenylpiperazine terminus (head group) defined as the primary pharmacophore, bind within the orthosteric binding site of both the D2R and D3Rs, while the indole amide terminus termed as the secondary pharmacophore, binds in a second binding pocket at the interface of transmembrane domains (TMs) 1, 2, and 7 and the EL1, EL2, that significantly differ from the D2R. Site-directed mutagenesis studies have identified a single amino acid (Gly94) in the EL1 that differs between D2 and D3 receptors and is critically important for subtype selectivity. These studies have provided a structural basis for the contribution of the each component in these molecules to the binding and functional efficacy at D3R, and to the relative orientation of the primary and secondary pharmacophores for optimal D3R binding affinity, selectivity and efficacy. More recently, modification of the primary pharmacophore or head group in these molecules with substituted phenylpiperazines has resulted in several new lead compounds that show improved metabolic stability over PG648 and efficacy in blocking the self-administration of heroin in a D3R-dependent manner, in mice. In addition, we have embarked on an SAR study of novel analogues of the D2R agonist sumanirole. Structural modification of this molecule, using clues from our D3R SAR work, coupled with analogous computational studies to the ones described above has revealed important structural features that impart high affinity for D2Rs and preference over D3R. This initial series of analogues has also provided clues to functional selectivity and synthesis and functional evaluation of these novel analogues is underway. These studies will undoubtedly provide new insight into novel drug design and provide important tools for mechanistic studies in vivo.
Showing the most recent 10 out of 56 publications
