GABA (gamma-amino-butyric acid), the principal inhibitory neurotransmitter in the brain, signals through ionotropic (GABA(A)/ GABA(c)) and metabotropic (GABA(B)) receptor systems. Here we report the cloning of GABA(B) receptors. Photoaffinity labelling experiments suggest that the cloned receptors correspond to two highly conserved GABA(B) receptor forms present in the vertebrate nervous system. The cloned receptors negatively couple to adenylyl cyclase and show sequence similarity to the metabotropic receptors for the excitatory neurotransmitter L-glutamate.

The antispastic agent and muscle relaxant baclofen 1 is a potent and selective agonist for bicuculline-insensitive GABAB receptors. For many years efforts to obtain superior GABAB agonists were unsuccessful. We describe the syntheses and biological properties of two new series of GABAB agonists, the best compounds of which are more potent than baclofen in vitro and in vivo. They were obtained by replacing the carboxylic acid group of GABA or baclofen derivatives with either the phosphinic acid or the methylphosphinic acid residue. Surprisingly, ethyl- and higher alkylphosphinic acid derivatives of GABA yielded novel GABAB antagonists, which are described in part 2 of this series. Structure-activity relationships of the novel GABAB agonists are discussed with respect to their affinities to GABAB receptors as well as to their effects in many functional tests in vitro and in vivo providing new muscle relaxant drugs with significantly improved side effect profiles.

In 1987, 25 years after the synthesis of the potent and selective GABAB agonist baclofen (1), Kerr et al. described the first GABAB antagonist phaclofen 2. However, phaclofen and structurally similar derivatives 3-5 did not cross the blood-brain barrier and hence were inactive in vivo as central nervous system agents. As a consequence, the therapeutic potential of GABAB antagonists remained unclear. In exploring GABA and baclofen derivatives by replacing the carboxylic acid residue with various phosphinic acid groups, we discovered more potent and water soluble GABAB antagonists. Electrophysiological experiments in vivo demonstrated that some of the new compounds were capable of penetrating the blood-brain barrier after oral administration. Neurotransmitter release experiments showed that they interacted with several presynaptic GABAB receptor subtypes, enhancing the release of GABA, glutamate, aspartate, and somatostatin. The new GABAB antagonists interacted also with postsynaptic GABAB receptors, as they blocked late inhibitory postsynaptic potentials. They facilitated the induction of long-term potentiation in vitro and in vivo, suggesting potential cognition enhancing effects. Fifteen compounds were investigated in various memory and learning paradigms in rodents. Although several compounds were found to be active, only 10 reversed the age-related deficits of old rats in a multiple-trial one-way active avoidance test after chronic treatment. The cognition facilitating effects of 10 were confirmed in learning experiments in Rhesus monkeys. The novel GABAB antagonists showed also protective effects in various animal models of absence epilepsy.

The finale of the large-scale preparation of 60 g of the highly complex marine natural product, (+)-discodermolide (1), using a hybridized Novartis-Smith-Paterson synthetic route is presented. This contribution, which is the concluding part of a five-part series, highlights a reagent-controlled stereoselective boron enolate aldol reaction between 2 and 3 forming the C7 hydroxyl-bearing stereocenter, selective reduction of 4a to generate the 1,3-anti-diol 5, and a global deprotection and concomitant lactonization leading to (+)-discodermolide (1). A novel procedure for converting the minor epimeric aldol adduct 4b into discodermolide using a five-step sequence is also described. This large-scale synthesis of discodermolide involved 39 steps (26 steps in the longest linear sequence) and several chromatographic purifications and delivered sufficient material for early-stage human clinical trials. IntroductionAfter 36 chemical steps and a gallant effort by many dedicated scientists, we now describe the finale that resulted in the delivery of 60 g of (+)-discodermolide (1), attesting to the power of contemporary organic synthesis in making available sufficient quantities of a highly complex organic molecule, sourced from nature in submilligram quantities, for a thorough evaluation of its therapeutic potential.

The synthetic strategy for producing multigram quantities of (+)-discodermolide (1) using a hybridized Novartis−Smith−Paterson synthetic route via common precursor 3 is described. In the first part of this five-part series, we present a multikilogram preparation of α-methyl aldehyde 10 from Roche ester, its syn-aldol reaction with Evans boron enolate, removal of the chiral auxiliary, and the preparation of Weinreb amide 3 (Smith common precursor). The common precursor was produced without any chromatography.