Figure 13 Examples of structure-based, computer-aided drug design.
Ki = 11,700 nM PD0178390
Ki = 0.028 nM >410,000 Fold Increase
Ki = 11,700 nM PD0178390
Ki = 0.028 nM >410,000 Fold Increase
Figure 15 Third generation, non-peptidic HIV protease inhibitors.
chemistry to modify the structure based on what we had learned from modeling, and looking at the requisite biochemistry and enzymology of the agents. This cycle was repeated to determine the appropriate binding of these molecules in the three-dimensional structure using structure-based drug design.
We were able to alter this relatively simple pyrone with a series of modifications, principally in the P2 and P2-prime sites, which very rapidly led to nanomolar level inhibitors and represented legitimate advances in non-peptide based HIV protease inhibitors. This has now been extended to an interesting mol ecule whose IC50 is 5 nm. It has a different structure, but nonetheless binds in the active site, does not use water within the binding site, and is quite different from the first- and second-generation inhibitors . This demonstrates the power of structure-based drug design, coming originally from a screening approach, to yield a drug candidate lead.
In the muscarinic area, quite a different approach was taken. We had determined that there was a three-point pharmacophore based on a series of agonists that to bind to the Ml receptor; however, the pharmacophore analysis had gotten bogged down at that point. As we modeled this within the GPCR model, we thought about where it was binding and looked at the other muscarinic receptors (M2 or M5) in terms of differences in the transmembrane region. We hypothesized that, not unlike retinal binding to rhodopsin, if this very small agonist was binding in the transmembrane region, we could elongate this simple pharmacophore (much like retinal is extended) to generate something that would have greater specificity. In fact, we were able to extend the basic pharmacophore to produce a molecule that has much greater specificity for the central versus the peripheral muscarinic receptors. This compound is now under investigation in clinical studies targeting the treatment of patients with Alzheimer's disease. Thus, this original idea of molecular modeling and pharmacophore analysis is in fact a validated approach to drug discovery.
How will we discover drugs in the future? While I believe that all five categories will be used, modification of known drugs (validated chemotypes) will probably be of decreasing interest over time. The integration of the other four approaches will become more important in the future application of drug discovery. The targets are manifold. Figure 4 is a highly simplistic representation of the source of molecular targets that we currently spend so much time identifying and finding antagonists or agonists for. They can be extracellular, membrane-based receptors, nuclear receptors, or cytosolic signaling receptors. There is a growing list of targets such as serine/threonine kinases, tyrosine kinases, and phosphatases, as well as a significant interest in the cell cycle and understanding the regulatory enzymes involved in the cycle. Probably the newest frontier is transcription factors, where many are working on ways of finding agents that would regulate gene expression and affect some particular pathological state.
The technological foci for drug discovery, shown in Figure 16, are organized for a specific reason. The technologies of high-throughput screening, and biomol-ecular structure determination, molecular modeling, and combinatorial chemistry are having a dramatic impact on drug discovery today and will continue to become even more important in the future. However, they are very dependent on finding the molecular targets. A veritable avalanche of new targets will be coming via genomics and proteomics. Concomitantly, there is the critical aspect of managing the plethora of data that emerge as a result of these data-rich approaches, and
understanding from those data which are the truths that are important in the drug discovery process.
Genomics is a simple word, but it envelops many components that are more than simply the genetic map, or the physical map, or even the gene sequence. The advances made in these three areas during the last 3 or 4 years alone have been striking, and much more rapid than any of us would have imagined. In many ways, gene sequencing, as well as the maps, will probably be well in hand for a number of genomes in the next couple of years. However, gene function and gene regulation represent a formidable challenge for drug discovery and the molecular sciences. Genetic functional analysis has a number of tools already available, and there will undoubtedly be more to come. Transgenics, knock-outs, and gene replacement are very powerful technologies in our understanding of gene function. Antisense is already available, and of course the two-hybrid technique is being exploited in many laboratories investigating gene function. Synteny, differential display, and single-nucleotide polymorphism (SNP) analyses are additional tools. Nonetheless, this is a bottleneck, and improvements are needed before we can move forward from sequence to function and understand regulation. The challenges in this field will include sequencing, informatics, multiple species, and the fact that it is not only the natural state that we are interested in but the pathological state as well. We need to understand function and mutations relevant to the pathological state. The output is the genetic footprint. Disease phenotype is what we are interested in. It has implications for diagnostics as well as for drug discovery, and it has implications down the road for preventive medicine and gene therapy.
Proteomics is also an important area. In fact, this is here and now, not in the future. Two-dimensional gel electrophoresis, together with matrix-assisted laser desorption/ionization (MALDI) mass spectrometry and image analysis, is used to determine the output of genomics. This is an area of intense investment in many laboratories that must be included in the bioinformatics database that is being generated. This database will be used to help determine which targets are the appropriate ones for drug discovery or for diagnostics.
Lastly, there are some important components for which tools are slowly evolving. We need to optimize a drug candidate, not only for selectivity but for bioavailability, toxicity, target organ selectivity, stability, and scalability. These are all legitimate and important components. We have drug class issues in drug discovery and development that act as guides for us, but nonetheless there are going to be discoveries unique to particular drug candidates. Small molecules continue to present difficulties with toxicity and bioavailability. Proteins have associated cost and delivery issues. Peptides have stability and bioavailability issues. Oligonucleotides often have the same class issues as proteins and peptides. Gene therapy certainly has to face the safety, delivery, and duration-of-effect issues.
What, then, are the future issues for drug discovery and development, given all of these technological foci? Determining the importance of the molecular targets is one such issue. For the next few years, we will continue to operate with a pretty tenuous linkage of molecular target to disease. In many cases, our molecular targets are hypothesized to be linked to a particular disease; we don't really determine this until we find inhibitors and take them to the clinic to see if they work. Also, the diseases we are facing today and tomorrow are more challenging than those that confronted us yesterday and the day before. Many are chronic diseases with multiple etiologic factors and will probably require combination ther-apy—another complication to drug development. Of course, we will have novel toxicities. Finally, as we move further into the area of gene regulation, nuclear targets will represent additional complexities. The opportunities and the challenges will yield new drugs, combination therapies, and patient-specific treatments during the next decade.
Was this article helpful?
The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.