 |
 |
| Back Issues |
|
|
|
|
ACS Members can sign up to receive C&EN e-mail newsletter.  |
|
|
 Join ACS |
|
|||||||||
|
DRUG DEVELOPMENT
A novel structural parameter can predict the oral bioavailability of a drug--that is, the degree to which it becomes systemically available when taken by mouth. A research team finds that compounds with good oral bioavailability tend to have reduced molecular flexibility, as measured by the number of freely rotatable bonds. The study was carried out by researchers Daniel F. Veber, Stephen R. Johnson, Hung-Yuan Cheng, Brian R. Smith, Keith W. Ward, and Kenneth D. Kopple at GlaxoSmithKline, King of Prussia, Pa. [J. Med. Chem., 45, 2615 (2002)]. The paper highlights "the importance of rotatable bonds in determining oral bioavailability, independent of molecular weight," Veber says. Low molecular weight has generally been considered a prerequisite for "druglike" properties. "Our analysis opens the possibility that new drugs can be found more easily in the high-molecular-weight range, provided the number of rotatable bonds is minimized." For the past few years, the "rule of five" developed by senior research fellow Christopher A. Lipinski (now retired) and colleagues at Pfizer in Groton, Conn., has been the standard approach for predicting oral bioavailability from structure data. "The number five or a multiple occurs in all of the cutoff parameters," Lipinski explains, "but there are only four rules--involving molecular weight, number of hydrogen bond donors and acceptors, and lipophilicity." The new rotatable-bond rule complements the rule of five. Veber and coworkers discovered the parameter by studying 1,100 compounds in GlaxoSmithKline's archive. About 65% of fairly rigid compounds in the collection (those with seven or fewer rotatable bonds) exhibited good-to-excellent oral bioavailability, independent of molecular weight. In contrast, more than 75% of floppy compounds (those with more than 10 rotatable bonds) had poor oral bioavailability. Compounds of intermediate rigidity fell somewhere in between. The scientists caution that data on oral bioavailability in rats were used in the study, so the results may not correspond exactly to human oral bioavailabilities. "It's very important work, and it's original," comments professor of bioorganic chemistry Ralph F. Hirschmann of the University of Pennsylvania. "It's a valuable addition to the armamentarium of how to end up with candidates for safety assessment that are orally bioavailable." In pharmaceutical research, "we used to improve potency all we could, and very often we ended up with compounds that were very insoluble and not available orally," Hirschmann says. "But in the past 15 years, the industry has learned that you had better worry about oral bioavailability early on in the game. The more rules you have to guide you in that endeavor, the better off you are." As soon as the paper by Veber and coworkers "popped up on the Journal of Medicinal Chemistry site, people here in Groton were circulating it," Lipinski says. "It's an intriguing finding, but it's also puzzling, because there's really no clear explanation of where the parameter is coming from. So when people look at this paper, they may scratch their heads, but it will prompt them--in their own data sets--to look and see if there's any relationship between rotatable bonds and oral bioavailability."
INTRIGUING FINDING Pharmaceutical researchers (from left) Ward, Smith, Veber, Kopple, Johnson, and Cheng discovered a new way to predict oral bioavailability. |
|||||||||
|
Chemical & Engineering News |
|||||||||
|
||||||||
|
||||||||
|
||||||||
|
||||||||
|
||||||||
|
||||||||
