Chemical library aids in developing drug system for nerve damage

Apr 12 2005

A researcher studying drug design for nerve damage therapies has gotten her answer to questions by following some old advice: she used the library.

It's not the kind of library her mother or teacher suggested, but a combinatorial chemistry library of many different protein sequences that some day might help her and her colleagues develop a successful timed drug delivery system.

Shelly Sakiyama-Elbert, Ph.D., assistant professor of biomedical engineering at Washington University in St. Louis, has screened a large number of molecules to find which ones have varying affinity, or attraction, to a sugar that binds nerve repair drugs called heparin, as well as a nerve repair protein called nerve growth factor.

Sakiyama-Elbert ran a library of viruses called bacteriophage that contained small random portions, or sequences of their surface proteins — which could be used to attract or bind other proteins — through a column with the drug bound to it. She then made the playing field more difficult for the bacteriophage to bind so that eventually she could find bacteriophage peptides that bound to heparin or nerve growth factor. By repeating this process numerous times, she identified peptide sequences that have low, medium or high affinity for the heparin drug.

Sakiyama-Elbert and her colleagues are looking for protein sequences that bind to drugs to help a drug delivery vehicle provide timed release of a drug. Such drug delivery systems are called affinity-based, and it is hoped that eventually they will provide the signals necessary to stimulate tissue regeneration for conditions such as nerve damage on an appropriate time scale.

In conjunction with the sequence technique, Sakiyama-Elbert and her group developed a mathematical model that identifies the kind of drug release desired as a basis to narrow down the range of affinities they want to identify from the library. Between the modeling and future experimental studies, they hope to refine their drug delivery design to get the optimal rate of drug release.

"We started with a model I'd previously developed, then added in some features that allow us to model degradation of the delivery system by enzymes," Sakiyama-Elbert said. "Specifically, we added a component where we can model what would happen if there is a cell in part of the delivery system and how that would affect release throughout the delivery system.

"Before you only could address what would happen in a culture dish with no cells around it. We're really interested in what will happen in cell culture or an animal model where there will be active cell-mediated degradation. We are trying to get closer to the real situation."

The results were published in the January 2005 issue of Acta Biomateriali. The work was supported by a grant from the Whitaker Foundation.

Sakiyama-Elbert said lots of researchers are adopting the concept of affinity-based drug delivery systems, and the Washington University library screening technique and mathematical model together provide a good tool to expand the usefulness of these approaches.

"One interesting thing we've found in this work is that it appears the activities of the drugs that we're delivering vary with the affinity of the binding site," she said. "We're not sure if that's a function of the affinity controlling the rate of release or if there is actually some separate biological modulator that's being affected.

"The good thing, though, is that we've identified several different affinities of binding sites so we can now test whether it's the affinity or the rate of release and determine what's really going on."

This ability is important for researchers to get insight into the biological activity of different drugs and how they might be modulated for drug release.

"We have low, medium and high affinity binding proteins," Sakiyama-Elbert. "We can look at fast and slow release rate for all three of them, so we can control affinity and concentration to our advantage."