From basic science to blockbuster drug: the discovery of Lyrica.

Many great discoveries occur when they are least expected. These discoveries are most satisfying when they derive from studies of basic science. That is how the new blockbuster drug for the treatment of various neuropathic pains, epilepsy, and generalized anxiety disorder, Lyrica (pregabalin), was discovered. This essay describes the discovery and features of this new drug. One of the early projects I initiated when I started my independent career at Northwestern University was the design and mechanism of new inactivators of the pyridoxal 5’-phosphate (PLP)-dependent enzyme g-aminobutyric acid aminotransferase (GABA-AT). GABA-AT is the enzyme responsible for the degradation of the inhibitory neurotransmitter, GABA, leading to its conversion to the excitatory neurotransmitter l-glutamate. Compounds that inhibit this enzyme have anticonvulsant activity as well as exhibit activity against Huntington0s disease, Alzheimer0s disease, Parkinson0s disease, and drug addiction. Epilepsy, broadly defined as any disease characterized by recurring convulsive seizures, has been known for many millennia. There are numerous etiologies for epilepsy because it is not a single disease; consequently, 1–2% of the world population has some form of epilepsy. Of those afflicted with this disease, 30–40% do not respond to multiple anticonvulsant drugs. There are many causes for seizures, but one of them is an imbalance in the concentration of GABA relative to l-glutamate. When GABA levels in the brain diminish, seizures can result. Injection of GABA directly into the brain can terminate the seizure, but administration of GABA, either orally or intravenously, has no effect because GABA, a hydrophilic charged molecule, does not cross the blood–brain barrier (BBB), a membrane that protects the brain from chemicals in the blood while still allowing essential metabolic function. The BBB comprises very tightly packed endothelial cells, which provide the walls of the blood vessels perfusing the brain; this higher density of cells restricts passage of unwanted substances from the bloodstream into the brain. One approach to increase the GABA concentration in the brain is to design a compound that can cross the BBB and inhibit GABA-AT, the only enzyme that degrades brain GABA. This prevents the breakdown of GABA, and its concentration rises, resulting in an anticonvulsant effect. Because of the importance of increasing brain GABA levels in central nervous system (CNS) disorders, my group, in the years 1981–1988, designed a series of mechanism-based inactivators for GABA-AT. It became apparent, however, that to progress toward the design of a new anticonvulsant agent, it would be necessary to prepare compounds that were selective inhibitors of GABA-AT (to raise GABA levels) without inhibiting l-glutamic acid decarboxylase (GAD), the PLP-dependent enzyme that converts the excitatory neurotransmitter, l-glutamate, to the inhibitory neurotransmitter, GABA (Scheme 1). Inhibition of GAD would decrease the concentration of GABA, the opposite of the desired effect. Furthermore, for brain penetration, which would be required for an anticonvulsant drug, increased lipophilicity would be important. Consequently, in 1988 I asked Dr. Ryszard Andruszkiewicz, a visiting scholar from the Technical University of Gdansk, to synthesize a series of 3alkyl-GABA and 3-alkylglutamate analogues, then to measure their inhibition of GABA-ATand GAD to determine if we could identify more lipophilic analogues that selectively bound to the former and not the latter enzyme. Dr. Andruskiewicz proceeded to synthesize fourteen 3-alkyl-GABA analogues (including four stereoisomers) (Scheme 2), 4-methyl-GABA (and its two enantiomers) and seven 3-alkylglutamate analogues. All of the GABA analogues were substrates for GABAAT. As the substituent size increased, so did the Michaelis constant Km; the Vmax/Km value for the 3-methyl analogue was a little larger than that for GABA, but the Vmax/Km values for the remaining analogues were progressively smaller (Vmax is the maximum velocity of the enzymatic reaction). The unexpected surprise came when these compounds were tested as inhibitors for GAD. Not only was none of them an inhibitor, but all of them were found to activateGAD, that is, the addition of compound produced an increased rate of GABA formation (Figure 1)! This had not previously been observed. My immediate thought was to have these compounds tested as anticonvulsant agents because they might provide a new [*] R. B. Silverman John Evans Professor of Chemistry Department of Chemistry Department of Biochemistry, Molecular Biology, and Cell Biology Center for Drug Discovery and Chemical Biology, Northwestern University 2145 Sheridan Road, Evanston, Illinois 60208-3113 (USA) Fax: (+1)847-491-7713 E-mail: Agman@chem.northwestern.edu Essays