No abstract

The process development and large scale synthesis for the β-hydroxyamino amide 1 is described. The route evolved from a multistep sequence utilizing a classical resolution to a two-step enantioselective process involving an enzyme-catalyzed aldol reaction and a direct amidation of a carboxylic acid. By utilizing a siliconmediated direct amidation strategy, the route was devoid of protecting and deprotecting steps while retaining the stereochemical integrity of a highly sensitive β-hydroxyamino acid. The two-step strategy employed herein significantly improved the yield, process greenness, cycle time and estimated cost in the production of 1.

The chiral β-hydroxy-α-amino acid, (2R,3S)-2-amino-3-hydroxy-3-(pyridin-4-yl)-propanoic acid, is a key intermediate in the synthesis of the API (2R,3S)-2-amino-3-hydroxy-3-(pyridin-4-yl)-1-(pyrrolidin-1-yl)propan-1-one, a developmental drug candidate. Two D-threonine aldolase enzymes were identified to catalyze the aldol addition of glycine and pyridine 4-carboxaldehyde for the synthesis of the β-hydroxy-α-amino acid. The two D-threonine aldolase enzymes have similar properties. Efficient recombinant E. coli fermentation processes were developed for producing the enzymes. The stabilities of the enzymes were significantly improved by addition of divalent cations. An unexpected and beneficial finding was that the β-hydroxy-α-amino acid aldol addition product directly crystallized out from the reaction mixture in high purity and high diastereoand enantioselectivity, contributing also to high yield and allowing easy isolation, processing, and downstream utilization. The temperature, pH, and amounts of reactants and enzyme were optimized to minimize reaction time and enzyme and raw material usage and maximize amino acid formation. Efficient D-threonine aldolase-catalyzed synthesis and recovery of the β-hydroxy-α-amino acid at the 100 L scale was demonstrated leading to a highly efficient and environmentally friendly process for the production of the API.

A novel bicyclic prostaglandin analogue, [1S-[1 alpha, 2 alpha (Z), 3 alpha, 4 alpha]]-7-[3-[[[[(1- Oxoheptyl)amino]acetyl]amino]-methyl]-7-oxabicyclo[2.2.1]hept-2- yl]-5-heptenoic acid [-)-7) was found to be a potent and selective thromboxane A2 (TxA2) receptor antagonist. Unlike the related series of omega-chain allylic alcohols, amide 7 and its congeners were uniformly free of direct contractile activity in vitro (bovine coronary) and in vivo (anesthetized guinea pig). Amide 7 was effective in the inhibition of (a) arachidonic acid induced platelet aggregation of human platelet-rich plasma (I50 = 0.18 +/- 0.006 microM), (b) 11,9-epoxymethano-PGH2 induced platelet aggregation of human platelet-rich plasma (I50 = 0.24 microM), (c) 11,9-epoxymethano-PGH2 induced contraction of guinea pig trachea (Kb = 3.0 +/- 0.3 nM) or rat aorta (Kb = 8.8 +/- 1.1 nM), and (d) arachidonic acid induced bronchoconstriction in the anesthetized guinea pig (0.1-1.0 mg/kg iv). Amide 7 inhibited the binding of [5,6-3H2]-[1S- (1 alpha, 2 alpha (Z), 3 alpha, 4 alpha)]-7-[3-[[2-[(Phenyl- amino)carbonyl]hydrazino]methyl]-7-oxabicyclo[2.2.1]hept-2-yl]-5- heptenoic acid to human platelet membranes in a specific and saturable manner with a Kd = 49.6 +/- 1.4 nM.

No abstract