Onan average, a drug spends about 12-14 years in a lab, and a company, about 800million or more, on it. To say drug discovery is a tedious process, isan understatement. It is a multi-disciplinary endeavour that begins with theselection of a target site or molecule, followed by the identification of alead molecule, which will bind to the target and elicit a physiologicalresponse.
The development process is marked by modifying the lead molecule toensure that the lead can suffice as a potential drug. Often lead molecules arealso selected on the basis of Lipinski’s rule of five (Molecular weight <500Daltons; No. of H bond donors< 5; No. of H bond acceptors<10; logP ? 5,where logP is a measure of hydrophobicity). The idea is to arrive at a moleculewhich can act as an ideal drug candidate. Its toxicity should be low but mustbind to the target with high affinity and form a stable complex. Within thebody, it should be easily absorbed and distributed.
There is an extensive listof criteria that a lead molecule must fulfil before it can be called a drug.But the story does not end here. Potential drug candidates are subjected topre-clinical and clinical trials before they are released into the market.Often, their use and side-effects are monitored post market-release to ensurethat long-term usage of the drug is safe; information that cannot be gatheredduring the different phases of trial. While this may sound verystraight-forward, it unfortunately isn’t. There is a lot of research, time andeffort that is spent on discovering targets, lead molecules and optimizing leadmolecules to function in the physiological environment.
About 90% of potentialdrug candidates that enter clinical trials conducted by the U.S. Food and DrugAdministration (FDA) fail. Not to mention, the expense borne is significant;most of which can be attributed to the failures during the development process.There is an urgent need to tackle these issues. And a solution has alreadypresented itself. Computer-aided drug designing (CADD) orin-silico drug designing is a novel approach employed to reduce the time andmoney expended in drug discovery and development. At present databasespertaining to protein structure (Ex.
PDB- Protein Data Bank), small molecules(Ex. ZINC), metabolic pathways (Ex. KEGG), etc. exist that store a sizeableamount of data and provide an interface that makes data mining convenient. Drugmolecules are usually discovered serendipitously or designed based on naturalligands. With the advent of CADD, molecules can also be designed de novo to bind to a particular target.Information about the target can be found from databases such as PDB, whichhouses structural information about proteins, nucleic acids, and differentprotein complexes. The importance of CADD is that a huge amount of data can beanalysed in-silico, reducing manual effort, time and costs of such a venture.
As a result, the process of drug discovery and development can be expeditedsubstantially. Use ofCADD in stages preceding clinical trialsSince we already have databases that containstructural information about various macromolecules, target structures can beeasily found. If the target is a protein and its structure has not been determined,then by using its sequence, computational methods can determine its 3Dstructure. Potential drug candidates can be virtually docked to this targetmolecule (using software such as AutoDock) and the interaction can be studiedby using a variety of computational tools and algorithms. Binding affinity,binding pocket site, flexibility of the ligand, ionisation states, etc. can allbe determined in-silico. Molecular Dynamics (MD) simulations can also helppredict the interaction in different physiological environments. This approachis mainly termed structure-based drug design (SBDD).
Small molecules can bescreened in-silico to identify potential lead molecules and in some cases, ifthe target structure is known, the lead molecule can be built from scratchusing algorithms like LigMerge. The latter is a form of de novo drug designing. Once a potential drug candidate has beenidentified, the lead can be optimized, i.e.
its pharmacokinetic properties can beenhanced and the ADME (absorption, digestion, metabolism and excretion)properties can be altered to suit particular needs. In the event that thetarget structure cannot be determined, an alternative approach is employed.Ligand-based drug design (LBDD) designs lead molecules for unknown targetstructures by comparing them to already known ligands and modelling theaccordingly. Pharmacophore and QSAR (quantitative structure-activityrelationships) modelling build models based on the essential features on knownligands.Post leadoptimizationCADD finds heavy usage in targetidentification, lead discovery and optimization; but the use of computationalbiology doesn’t end there. During pre-clinical and clinical stages of trials, alarge amount of data is generated, which handling manually is a daunting task. Computersin this case, make storage extremely convenient and the data can be accessed atany point during the trial. Such data can also be used in future trials or researchprojects, when larger samples of people have to be tested or the effect of adrug has to be studied.
Clearly the intervention of computers ishelping us in significantly reducing the time and cost associated with drugdesign and development. Not only can we virtually discover our drug, but we canalso model it as per our requirement. A major chunk of the cost associated withthis process is often expended due to failures. CADD significantly reduces thecost associated with failures if not the chances of failure. CADD is theshort-cut that we have desperately been looking for in a world where diseasesand disease-targets are being identified at an ever-increasing rate but thesimultaneous production of potential drug therapies is comparatively slow. Thesuccess of CADD is evident.
Many drugs, that have already been in the marketfor a while now, have been developed by methods such as LBDD. Norfloxacin,Losartan and Zolmitriptan are a few to name. CADD is an important milestone inour scientific career. It is a good example of methods that expedite the lab tolife journey our discoveries have to make.