Conformational analysis pharmacophore discovery

Conformational analysis. Pharmacophore discovery inherently involves the 3D structure of molecules and, by necessity, goes beyond merely the low-energy conformation. 

Constrained Search and the related approach of SCAMPI both integrate the conformational analysis closely with the pharmacophore discovery. This has the advantage that the sampling of conformational space can be more focused on key regions. With both Catalyst and DANTE, conformational analysis was explicitly kept separate, in the latter to allow one to take advantage of any innovations in conformational analysis tools. And, indeed, there continues to be a steady flow of new approaches in conformational analysis—pharmacophore discovery is critically dependent on high-quality exhaustive conformational analysis. Based on our experience thus far, we cannot conclude that either approach is superior (integrated vs. external). Furthermore, a consensus has not yet been reached on the optimal manner to perform conformational search as needed by pharmacophore discovery. This will continue to be a fruitful area of research. 

Figure 5.6. Overview of the Gridding and Partitioning (GaP) procedure as applied to monomers, exemplified using phenylalanine as a potential primary amine. This molecule thus contains two pharmacophoric groups (the aromatic ring and the carboxylic acid). During the conformational analysis the locations of these pharmacophoric groups are tracked within a regular grid. See color insert. [Reproduced from A. R. Leach and M. M. Hann, Drug Discovery Today, 5, 326-336 (2000),...

Since molecules are flexible and not static, a conformational analysis has to be carried out first to generate an ensemble of low-energy conformations. This is probably one of the most critical steps in the pharmacophore discovery process, since the goal is not only to consider the global miifima of a molecule, but also to include the bioactive conformation as part of an ensemble of low-energy conformations. 

After ALADDIN, this author and Martin parted ways, but each played a role in the rise of two distinct pharmacophore discovery methods. In 1990, this author joined, as a founding member and first scientist, a new company in Silicon Valley, BioCAD, dedicated to the development of software to assist drug discovery. Our software, Catalyst, contained three components conformational analysis, 3D database searching, and Hypothesis Generation —a novel approach to pharmacophore discovery. Catalyst represented the first fully automated method for pharmacophore discovery. 

Although Catalyst was not a commercial success (the company BioCAD folded in 1994, and the software was taken over by MSI), it stimulated much interest in fully automated pharmacophore discovery. Martin et al. [16] developed DISCO, borrowing code from ALADDIN to detect features and employing a clique detection algorithm mathematically similar to the MNMM method. DISCO also relies on separate, exhaustive conformational analysis, and, in general, produces many pharmacophores consistent with the SAR. 

The original work of Marshall et al. has been carried on, primarily by Beusen and Shands [17a] and Dammkoehler et al. [17b]. Their work has been innovative, primarily in that the pharmacophore discovery algorithm is now integrated with the conformational analysis, a step that should, in principle, significantly improve the quality of the results. The work of SCAMPI [18] is conceptually similar to the work of Beusen and Shands. 

After developing novel approaches to exhaustive conformational analysis, Crippen [21] at the University of Michigan took a novel approach to pharmacophore discovery, based on Voronoi polyhedra (using hyperplanes to partition space into regions encompassing active molecules). This line of investigation was ultimately abandoned, as Crippen was unable to find a satisfactory resolution to the problem of multiple solutions consistent with the SAR. 

An alternative metric to describe 3-D properties of molecules is discussed by Ashton et al.26 In their approach, a pharmacophore fingerprint is used in conjunction with conformational searching to determine possible 3-D shapes that molecules can adopt. Tools from Tripos and CDL are available to carry out this type of analysis. However as with other methods there are limitations, the completeness of conformational searching being one. Perhaps the most important limitation of the approach is that it has a tendency to pick the most flexible molecules (that set the most pharmacophore bits). In a lead discovery experiment, following up on flexible molecules can be a long and sometimes fruitless process. 

---