INHIBITION OF CARBONIC-ANHYDRASE

We report free energy perturbation simulations on a series of sulfonamide (RS(O)2NH-) inhibitors of the zinc metalloenzyme human carbonic anhydrase II (HCAII). In order to carry out these simulations, we had to incorporate the zinc ion into thc AMBER force field. To do this, we have found that the following modifications are appropriate: (1) the charge on zinc was reduced from +2.0 to +0.8; (2) explicit covalent bonds and angles were incorporated between the zinc and its ligands (His 94, His 96, His 119). This model was determined by parametrizing the force field against the known structure of a HCAII-acetazolamide complex. The series of compounds examined include p-hexylbenzenesulfonamide (1), benzenesulfonamide (2), and p-hexylbenzenesulfonate (3). Two conversions were studied: the first involved the direct conversion of 1 into 2, while the second involved changing the sulfonamide group to a sulfonate (1 --> 3). The former simulation involved direct conversion of a hexyl group into a hydrogen atom, an ambitious calculation, which has provided insight into the capabilities of the free energy perturbation method. We find that we can reproduce experimental relative binding constants but that this ability to do so is very dependent on the molecular mechanical model used and on the simulation protocol. In order for us to compare our calculated results with experimental ones for the latter simulation, we have had to account for the pK(a) difference between the sulfonamide and a sulfonate groups. With the appropriate correction for the pK(a) difference between 1 and 3 we find that we are able to reproduce the experimental DELTA-DELTA-G(bind). We also find that the reason why sulfonamides are better inhibitors of HCAII than are sulfonates can be traced to a single hydrogen-bond interaction present in sulfonamides, but lacking in sulfonates.