The rationale follows (with a story!). Certain drugs show excellent specificity for killing fungal cells and not mammal cells. While this is easily understood upon cursory investigation (after all, fungi are markedly different to mammals in appearence), it does not hold up under a more detailed examination: essentially, mammal and fungal cells are almost identical in composition! However, there is one major component of each cell type that differs from the other: the type of sterol included in the bilayer membrane. Mammal cell membranes (as you probably know) have cholesterol, with an abundance of around 30-50% mol (they are a major component!). Fungal cells have a similar abundance of sterol, but of a different kind: they include ergosterol rather than cholesterol.
Punchline time: it is hypothesised that the small differences between cholesterol and ergosterol are enough so that anti-fungal drugs kill the tinea on your feet (if you're unfortunate to have tinea) but not you. What makes this even more unlikely is the fact that the bit that is most different, the methyl group on the tail (those extra double bonds don't really do much different between the two sterols) is buried in the middle of the bilayer. See the picture below; ergosterol is the one with red bits and the differences from cholesterol are shown in red. So how does it affect the interactions of drug molecules with the surface of the membrane bilayer?
 |
| Figure 1: Comparison of cholesterol (left) with ergosterol (right). Differences are shown in red on ergosterol. Note that formation of additional double bonds means that a hydrogen atom has been removed from the involved carbon atoms. |
This question is hard to answer with traditional experimental techniques, such as NMR or x-ray crystallography, because these techniques either cannot show an atomic level of detail (NMR just takes an average of everything in the sample) or cannot reliably be used to image flexible biological membranes (lipid bilayers are notoriously hard to crystalise, let alone a whole biological membrane!). This is where my project comes in: we can use molecular dynamics (MD) simulations to atomistically model complex lipid bilayers and get the atomic-scale picture we need to discern the actual effects of the different sterols on the bilayer.
MD simulations (Martin and Wayne have both done them!) essentially use Newtonian mechanics (and a bunch of other assumptions, like mean field approximations) to model bonds and non-bonded interactions. This method requires many other approximations to ensure the observed nature of chemical species is seen in the simulations; the absence of quantum mechanical constraints limits their accuracy in this way. Nonetheless, if the simulations work, they are an excellent method for examining the physical properties of a system, and can provide much information that cannot be obtained from physical model systems.
In my project, I simulated a bilayer composed of a 1:1 mixture 2-palmitoyl-1-oleoyl-sn-3-glycerol-phosphocholine (POPC) and cholesterol, representing animal cells (my supervisor actually performed this simulation, but I analysed it) and a bilayer composed of 1:1 POPC and ergosterol, representing fungal cells. Unfortunately, if one has an incorrect topology for a molecule, the simulations crash (usually due to all of the atoms in the molecule flying off in opposite directions and increasing the energies to very silly heights), and this is what happened with my ergosterol simulation. This means that Professor Mark's group has some work to do to improve the model for ergosterol; it turns out that it is a lot less similar to cholesterol in terms of topology than we would think. Cholesterol was seen to have a big effect on the bilayer structure, though, including thickening it, allowing more access to the methyl chains of POPC
So, although my original aim remains unfulfilled, I was able to participate in active research!
Josh Harbort
No comments:
Post a Comment