A number of analogues of the low-efficacy partial GABA(A) agonist 5-(4-piperidyl)-3-isoxazolol (4-PIOL, 5), in which the 4-position of the 3-isoxazolol ring was substituted by different groups, were synthesized and tested as GABA(A) receptor ligands. Substituents of different size and structural flexibility such as alkyl, phenylalkyl, diphenylalkyl, and naphthylalkyl were explored. Pharmacological characterization of the synthesized compounds was carried out using receptor binding assays and by electrophysiological experiments using whole-cell patch-clamp techniques. Whereas none of these compounds significantly affected GABA(B) receptor sites or GABA uptake, they did show affinity for the GABA(A) receptor site. While alkyl or benzyl substitution, compounds 7a-h, provided receptor affinities comparable with that of 5 (K(i) = 9.1 microM), diphenylalkyl and naphthylalkyl substitution, as in compounds 7m-t, resulted in a dramatic increase in affinity relative to 5. The 3,3-diphenylpropyl and the 2-naphthylmethyl analogues, compounds 7s and 7m, respectively, showed the highest affinities of the series (K(i) = 0.074 microM and K(i) = 0.049 microM). In whole-cell patch-clamp recordings from cultured cerebral cortical neurons, all of the tested compounds were able to inhibit the effect of the specific GABA(A) agonist isoguvacine (1), compounds 7m and 7s showing antagonist potency (IC(50) = 0.37 microM and IC(50) = 0.02 microM) comparable with or markedly higher than that of the standard GABA(A) antagonist 4 (IC(50) = 0.24 microM). Highly potent convulsant activity was demonstrated in mice with compounds 7m (ED(50) = 0.024 micromol/kg) and 7s (ED(50) = 0.21 micromol/kg) after intracerebroventricular administration, whereas no effects were found after subcutaneous administration. According to a previously proposed pharmacophore model for GABA(A) receptor agonists, a receptor cavity in the vicinity of the 4-position of the 3-isoxazolol ring in 4-PIOL exists. A molecular modeling study, based on compounds 7o,m,l,q,s, was performed to explore the dimensions and other properties of the receptor cavity. This study demonstrates the importance of the arylalkyl substituents in 7m and 7s and the considerable dimensions of this proposed receptor cavity.

No abstract

The serotonin transporter (SERT) is one of the neurotransmitter transporters that plays a critical role in the regulation of endogenous amine concentrations and therefore is an important target for therapeutic agents affecting the central nervous system. The recently published, high resolution X-ray structure of the closely related amino acid transporter, Aquifex aeolicus leucine transporter (LeuT), provides an opportunity to develop a three-dimensional model of the structure of SERT. We present herein a homology model of SERT using LeuT as the template and containing escitalopram as a bound ligand. Our model explains selectivities known from mutational studies and varying ligand data, which are discussed and illustrated in the paper.

We have performed molecular dynamics simulations of a homology model of the human serotonin transporter (hSERT) in a membrane environment and in complex with either the natural substrate 5-HT or the selective serotonin reuptake inhibitor escitalopram. We have also included a transporter homologue, the Aquifex aeolicus leucine transporter (LeuT), in our study to evaluate the applicability of a simple and computationally attractive membrane system. Fluctuations in LeuT extracted from simulations are in good agreement with crystallographic B factors. Furthermore, key interactions identified in the X-ray structure of LeuT are maintained throughout the simulations indicating that our simple membrane system is suitable for studying the transmembrane protein hSERT in complex with 5-HT or escitalopram. For these transporter complexes, only relatively small fluctuations are observed in the ligand-binding cleft. Specific interactions responsible for ligand recognition, are identified in the hSERT-5HT and hSERT-escitalopram complexes. Our findings are in good agreement with predictions from mutagenesis studies.

No abstract