User:David C. Thompson/Projects/MD

There is some nice, albeit unintended, symmetry in the recent publication of Prof. Isaiah Arkin and co-scientists from the D. E. Shaw Research group [1]. Thirty years ago, in the summer of 1977, researchers at Harvard University published the first detailed molecular dynamics (MD) simulations of a folded globular protein [2]. Their computational investigations lasted mere pico seconds ([math]\displaystyle{ 10^{-12} }[/math] s) but helped illuminate the gross behavior of this intermediate size system as a function of time. Fast forward to present day where scientists in the D. E. Shaw group have presented an atomically resolved mechanism of ion transport derived from micro second ([math]\displaystyle{ 10^{-6} }[/math] s) MD simulations of the transmembrane Na+/H+ antiporter NhaA of E. Coli. The proposed mechanism agrees with existing experimentally derived hypotheses for the ion transporting action and, in addition, Prof. Arkin and co-workers propose detailed structural basis for both the pH regulation and ion selectivity exhibited by this channel.

That biologically relevant time scales are now computationally accessible for systems of a physiologically meaningful size, where and how might this new technology impact? One area that could benefit immeasurably is the field of drug discovery and development - is structurebased drug design about to enter a renaissance?

Within the competitive pharmaceutical/biotechnology space, with its aggressive project timeline and ever-hungry pipeline, an emphasis is placed on ‘failing fast’. The computational assessment of prospective inhibitors of therapeutically well validated targets cannot be a limiting step for computational chemists or biologists to be seen to add value to the drug discovery process. Advanced computational treatments of protein flexibility, using MD for instance, are thus not typically employed – even though the importance of protein movement is well established; this is especially true if one is trying to correctly account for the ‘induced-fit’ effect of the protein moving to accommodate the binding event of a small molecule. A fast, highly parallelizable, MD approach such as that developed by the D. E. Shaw group opens the potential for a more sophisticated, structural, understanding of small molecule inhibition for a competitive computational cost. A more thorough rationalization between theory and experiment could result in better-understood compounds being successfully passed through the myriad stages of drug development.

The prospective utility of this methodology, while compelling from an industrial point of view, is also particularly intriguing from the viewpoint of those few not-for-profit organizations that are developing medications for orphan diseases in the developing world. With only ten percent of the global research and development funding being spent on ninety percent of the global disease burden, organizations such as the Institute for OneWorld Health and Medicines for Malaria Venture are working with limited budgets. It is here that an advanced technology could have a substantial impact, which in turn could aid in the efficient development of pharmaceutical agents that would positively impact the lives of millions of individuals.

While this methodology clearly holds much promise, challenges still remain – not least of which is the proper inclusion of electronic effects as an MD simulation is only as effective as its underlying force field. However, given the substantial advances made since 1977, the future looks promising.

Bibliogrpahy

1. Arkin, I.T., et al., Mechanism of Na+/H+ antiporting. Science, 2007. 317(5839): p. 799-803.

2. McCammon, J.A., B.R. Gelin, and M. Karplus, Dynamics of folded proteins. Nature, 1977. 267(5612): p. 585-90.