A multistep scalable synthesis of the clinically important hepatitis C virus (HCV) protease inhibitor BILN 2061 (1) is described. The synthesis is highly convergent and consists of two amide bond formations, one etherification, and one ring-closing metathesis (RCM) step, using readily available building blocks 2-5. The optimization of each step is described at length. The main focus of the paper is the study of the RCM step and the description of the main problems faced when scaling up to pilot scale this highly powerful but very challenging synthetic operation. Eventually, the RCM reaction was smoothly scaled up to produce >400 kg of cyclized product.

Novel pyrido[2,3-b][1,4]benzodiazepinones (I), pyrido[2,3-b][1,5]benzodiazepinones (II), and dipyrido[3,2-b:2',3'-e][1,4]diazepinones (III) were found to inhibit human immunodeficiency virus type 1 (HIV-1) reverse transcriptase in vitro at concentrations as low as 35 nM. In all three series, small substituents (e.g., methyl, ethyl, acetyl) are preferred at the lactam nitrogen, whereas slightly larger alkyl moieties (e.g., ethyl, cyclopropyl) are favored at the other (N-11) diazepinone nitrogen. In general, lipophilic substituents are preferred on the A ring, whereas substitution on the C ring generally reduces potency relative to the corresponding compounds with no substituents on the aromatic rings. Maximum potency is achieved with methyl substitution at the position ortho to the lactam nitrogen atom; however, in this case an unsubstituted lactam nitrogen is preferred. Additional substituents on the A ring can be readily tolerated. The dipyridodiazepinone derivative 11-cyclopropyl-5,11-dihydro-4-methyl-6H-dipyrido[3,2-b:2',3'-e] [1,4]diazepin-6-one (96, nevirapine) is a potent (IC50 = 84 nM) and and selective non-nucleoside inhibitor of HIV-1 reverse transcriptase, and has been chosen for clinical evaluation.

No abstract

The development of a large scale synthesis of the glucocorticoid agonist BI 653048 BS H3PO4 (1·H3PO4) is presented. A key trifluoromethyl ketone intermediate 22 containing an N-(4-methoxyphenyl)ethyl amide was prepared by an enolization/bromine-magnesium exchange/electrophile trapping reaction. A nonselective propargylation of trifluoromethyl ketone 22 gave the desired diastereomer in 32% yield and with dr = 98:2 from a 1:1 diastereomeric mixture after crystallization. Subsequently, an asymmetric propargylation was developed which provided the desired diastereomer in 4:1 diastereoselectivity and 75% yield with dr = 99:1 after crystallization. The azaindole moiety was efficiently installed by a one-pot cross coupling/indolization reaction. An efficient deprotection of the 4-methoxyphenethyl group was developed using H3PO4/anisole to produce the anisole solvate of the API in high yield and purity. The final form, a phosphoric acid cocrystal, was produced in high yield and purity and with consistent control of particle size.

The development of zinc-mediated and -catalyzed asymmetric propargylations of trifluoromethyl ketones with a propargyl borolane and the N-isopropyl-l-proline ligand is presented. The methodology provided moderate to high stereoselectivity and was successfully applied on a multikilogram scale for the synthesis of the Glucocorticoid agonist BI 653048. A mechanism for the boron-zinc exchange with a propargyl borolane is proposed and supported by modeling at the density functional level of theory. A water acceleration effect on the zinc-catalyzed propargylation was discovered, which enabled a catalytic process to be achieved. Reaction progress analysis supports a predominately rate limiting exchange for the zinc-catalyzed propargylation. A catalytic amount of water is proposed to generate an intermediate that catalyzes the exchange, thereby facilitating the reaction with trifluoromethyl ketones.