I use computational chemistry every day. It’s a wonderful tool. As a synthetic organic chemist, I can use software suites installed on my regular desktop PC to predict properties of compounds that may only exist in my imagination. While the methods available are fairly powerful, and some good results can be obtained, an important drawback – something that I always have to keep in mind – is that most of the software available treats molecules as isolated, single compounds. It’s simply not practical (in terms of computational burden) to do otherwise. In order to simplify the number of calculations needed, most software assumes that the molecules I input exist in the gas phase, without any other molecules nearby.
The problem with this approach is that it doesn’t accurately reflect the “real world”. Most of the time, there are other molecules in close contact with the compound of interest; solvents, for example, usually surround a molecule. Take the “real world” example of pharmaceuticals: if you study the properties of a new drug molecule in most of the computational software available, you’re ignoring the fact that these drugs are going to be in a water environment. Our bodies are based on the use of water as a solvent to dissolve interesting compounds. All of these interactions between water and a molecule have to be taken into account when designing a new medication.
If a drug does not dissolve very well in water, it has a limited bioavailability and therefore the tissues that will benefit from the drug will be limited. It’s also important to know how potential new pharmaceuticals will interact with lipids, the fat-like layers that form barriers to molecular passage in many bodily systems. All of this is knowledge that is very difficult to calculate for an as-yet unknown, purely hypothetical drug candidate.
“Hydration free energy”, or HFE, is a mathematical term that quantifies how a molecule will behave in a certain solvent. It’s made up of various factors – the energy required to form a cavity in the solvent, the electrostatic interactions between solvent and the molecule, van der Walls interactions between the solvent and the molecule, and so forth. It’s incredibly complicated, but yet it’s also a vital piece of information for medicinal chemists. It’s no use designing a wonderful new drug “on paper”, only to discover it can’t be dissolved in the body or refuses to cross the blood-brain barrier, etc. Current computational software really struggles with calculations of HFE, as it’s simply too complex to address in any reasonable amount of time.
I’m well aware of this limitation of current software suites, and that’s why I was very excited to read a recent article in the Journal of Chemical Physics. A new theoretical model has been developed. This new model is extremely rapid (taking about 30 seconds) and can accurately predict the hydration free energy for most organic compounds, which covers 99% of existing and future pharmaceuticals. The secret behind the speed is that instead of calculating every single possible interaction between solvent molecules and the desired compound, the new algorithm uses several dozen experimentally-determined HFE values for similar molecules to provide a database. This database is then developed into a computer algorithm to calculate the hydration thermodynamics. These “semi-empirical” techniques have been known for some time, but they usually suffer from a loss of accuracy. This new algorithm hits the perfect balance between computational time and accuracy of results.
The results of this paper are important because chemists can’t afford to develop new pharmaceuticals in a vacuum (using the term literally, in this case). New drug candidates are going to be dissolved in water, most likely, as they’re probably going to enter the human bloodstream at some point. Accurate knowledge of how the molecules are going to interact with all of that water – and with the lipid membranes they encounter during their trip through the body – is critical to the rational development of new drug candidates.
While we’ll always need to actually produce potential drug candidates and test them in clinical trials, computational chemistry is finally advancing to the point where we’re moving beyond isolated single molecules. We can now place these theoretical new medicines in a virtual “real world” environment. It’ll probably take several years for this new HFE calculation method to become widely adopted, but it will no doubt eventually lead to remarkable advances in pharmaceuticals. As a Ph.D. chemist who’s been suffering under the limitations of this type of software for the past ten years, I’m extremely happy to see this development.
The source of this article can be found at:
Palmer, D.S.; Sergievskyi, V.P.; Jensen, F.; Fedorov, M.V. “Accurate calculations of the Hydration Free Energies of Drug-like Molecules using the Reference Interaction Site Model”. Journal of Chemical Physics, 2010.