Bcr spring 2011 lo-res

COMPUTATIONAL DRUG DESIGN:
New Tricks for Old Drugs
When cheap drugs are needed fast, resistance to all known therapies, making SOIPPA (Sequence Order Independent
which was developed by Bourne and Lei
short-cut: repurposing existing drugs for Xie, PhD, research scientist at UCSD. For
actions in the body, the same drug may be searched for TB proteins with structurally similar binding sites. Then Bourne’s team “There’s a lot of evidence now that the issue of PLoS Computational Biology. virtually docked the drug into these TB pro- so-called ‘magic bullet’—one-drug to specif- ically bind to one receptor to treat one con- had been co-crystallized with at least one “There’s a computer cost associated with dition—is the exception rather than the rule,” says Philip E. Bourne, PhD, a profes-
identify new drugs (see “Dock This” in ly “dock” drugs into receptors’ 3-D ing old drugs, these same in silico explain a drug’s side effects. The algo- Virtual Docking:
Focusing on Proteins
In this network representation of the TB-drugome, red nodes are FDA approved drugs and blue nodes are
binding sites on TB protein receptors. An edge is drawn between drug and receptor if the drug is believed to
interact with the protein. Bourne's group hypothesizes that the highly connected drugs present the most
opportunity for disrupting the normal functioning of TB. Courtesy of Philip Bourne. Reprinted from Kinnings,
SL, et al., 2010, The Mycobacterium tuberculosis Drugome and Its Polypharmacological Implications, PLoS
Computational Biology 6(11): e1000976. doi:10.1371/journal.pcbi.1000976.
Published by Simbios, the NIH National Center for Physics-Based Simulation of Biological Structures
cluster in a couple of weeks,” Bourne says.
Guilt by Association:
Focusing on Ligands
interesting associations experimentally.
the need for 3-D protein structures, says we’re very excited about,” Keiser says.
Michael J. Keiser, PhD. Keiser is President
them,” Bourne says. For example, two drugs experiments, it bound to this receptor 10 for Parkinson’s disease were unexpectedly found to bind to an important TB enzyme. goals). His team builds structure-free pro- files of the receptors. “We forget every- ent TB proteins, which is a potential boon the antidepressants Prozac and Paxil (which for preventing drug resistance. If a drug act on neurons) also act as weak beta block- develop a mutation that escapes the drug, receptor, located in heart muscle and blood Bourne says. “So what you really want is vessels. The finding may explain why some people who stop taking these drugs experi- ence changes in heart rate and blood pres- In a 2009 paper in Nature, his team sure (“SSRI discontinuation syndrome”). drug-target space for TB—previously only nine TB proteins had been investigated as drugs based on their phenotypic similarities, potential therapeutic targets, Bourne says. human protein targets looking for novel for example gene expression or side effect profiles. In a 2008 paper in Science, cover potential drugs for other diseases as well as to predict drug side effects. For 746 marketed drugs solely based on the side effects listed on their inserts. They found entire set [of ligands] rather than on a one- more than 1000 pairs of side-effect related sites for the breast cancer drug tamoxifen by-one basis,” Keiser says. They identified drugs, including a couple hundred pairs that that help explain why it can cause cardiac 6928 drug-target pairs that were statistical- were otherwise unalike. In tests of 20 of ly likely to bind. When they tested 30 pre- these, they verified 13 novel drug-target interactions. For example, rabeprazole, a proton pump inhibitor used to treat ulcers, was found to bind neurologic targets, includ- Going Forward:
Complementary Approaches
based approaches don’t require 3-D struc- tures, they are limited to protein targets that are already known to bind to drugs. only the first step in drug repurposing.
of new drug-target pairs to explore. If even just a few prove effective, these could pro- vide life-saving alternatives for devastat- ing diseases such as drug-resistant TB. ■ Ignoring structures and focusing only on ligand
binding, Keiser and his colleagues identified
new targets (blue) to existing drugs (gold) The
drugs' known targets (violet) connected to the
drugs by gray lines. Node sizes increase with
number of incident edges. Reprinted with per-
mission from MacMillan Publishers Ltd, Keiser,
M, et al., Predicting new molecular targets for
known drugs, Nature (2009).

Source: http://biomedicalcomputationreview.org/sites/default/files/sp2011-computational-drug-design.pdf