Structure based Drug Design

Commondrugtargets The majority of available drugs have proteinmolecules as theirtargets. Although nucleic acids mayalso be considered, their use as drugtargets indrugdiscovery and structure baseddrug designhas been limiteddue to various effects like toxicity, difficultyinachievinghighspecificity, etc. In a survey of currently available marketeddrugs, it was shownthat out of the 21000 registeredinthe USA drugs, only 1357were unique. Of these1204were small-molecule drugs, while 166 were biological agents. The total number of proteindrugtargets, whichincluded bothhumanproteins andproteins from pathogenic organisms, wasfoundtobe 324. Of these 207were humanprotein drug targets (Overington, Al-Lazikani& Hopkins, 2006).Around 70% of the targets belonged to10proteinfamilies, while almost 50% of the drugs were shownto exert theiractivity via 4 families:G-protein coupledreceptors, nuclear receptors, ligand-gated andvoltage-gatedion channels. Lookingat the structural side, the authors usedthe CATH and SCOP databases, whichcontainclassificationof proteindomains, and identified130“druggable”domains. This is a nice exampleof how structuralbioinformatics resources are usedindrug discovery andstructure-based drugdesign.

Startinga structure-baseddrugdiscovery project- some generalconsiderations Most moderndrugdiscoveryprojects start withproteintarget identificationand verification toobtain a “verifieddrug target”.Forstructure-baseddrugdesignthe three-dimensionalstructure of the proteininquestionneeds tobe determinedby one of the available highresolutionexperimentalmethods:Protein crystallographyor NMR.When identifying a drug target, we first needtoanswersome general questions:

Does the target protein belongto a biochemicalpathway, which can be bypassedby the cell, if inhibited?Obviously, if the pathway canbe bypassed, inhibiting it will not make much difference. If our aim is toinhibit a protein, whichbelongs toa pathogen, an obvious questionwould be:Are there any related proteins inthe human host, which maybeaffectedby the drug? If the protein is not sowellstudied one couldalso ask if it is actually drugable, inthe sense that it has a small-molecule bindingsite forwhicha bindingcompoundcan be designed. Of course, inthe best case scenariothere will be someknowninhibitory compounds whichcanbe co-crystallizedwith theprotein.This willhelpinmapping available interactions within the active site, which inturn willhelpin thenext step when new compounds will be designed. If there is nothree-dimensional structureavailable forthe proteintarget one couldtry tofind a structure of a homologous protein, whichmay subsequently be usedfor homology modeling. There are of course just few examples, many more questions, often specific toeachtarget, need to be answeredbefore startinga structure-baseddrug designproject.

Actually, manyof the questions related to drug discoveryand structurebased drug design may be answered using the tools of structural bioinformatics. For example, a search of sequence databases followed by sequencealignmentand analysis mayeasily answer questions related to the specificity of a particular targetin a given organism. In the following page we will discuss in moredetails thequestions related to finding a useful compound (hit generation) and the use ofstructural information in optimization of newlyidentified compounds.