When flamingos are approximately 6 years of age, they are ready to start mating. The parallelism between this new way of approaching drug discovery and the courtship rituals of the flamingo is exemplified in Figure 1. However, potent ligands against a therapeutic target are abandoned along the drug discovery pathway if they do not show an acceptable spectrum of physicochemical, absorption, distribution, metabolism, and elimination (ADME) properties along with a minimal risk of toxic effects. In drug discovery, scientists are committed to adjusting several physicochemical and biological properties in the search for drugs. In nature, evolutionary improvement occurs via the continuous selection of well-established features enabling organisms to adapt, survive, and reproduce. The process of drug discovery can be directly paralleled to that of evolution, whose success depends on natural selection, among other driving forces. Thus, we suggest that the concept of evolution should be applied to drug discovery. In fact, most known natural drugs are likely to have been molded by the process of evolution indirectly via the enzymatic systems responsible for their synthesis, thus optimizing all the possible ‘facets’ to balance their on/off-target profile. Such an approach would also be more akin to that which occurs in nature. Instead of analyzing thousands of candidate compounds by using sequential filters, each accounting for one property at a time, we should attempt to optimize more properties simultaneously. To bridge drug discovery and biology, we should first acknowledge the multifaceted nature of drugs and then readdress the drug discovery approach. In this complex scenario, we should reconsider the way we search for new drugs and move beyond the reductionist ‘one-target fixation’ paradigm. Even drugs with relatively high target specificity are known to engage a multitude of proteins via a structured network of hydrogen, hydrophobic, and ionic interactions, thus inducing their 3D structures and modulating their functioning.
It is now widely accepted that drugs are inherently poly-pharmacological because they can act on multiple targets or disease pathways. However, approaching drug discovery in such an ‘inverted cone-shaped’ fashion constitutes a simplified procedural abstraction often detached from the intimate nature of drug biology encompassing the occurrence of simultaneous and multilevel complex interactions, that is, the mode of action of the drug.
The success rate along the drug discovery pipeline depends on the chance of crossing filters that are used to discard compounds whose features do not match those typical of drugs.
Flamingo dance series#
For years, the drug discovery pipeline has been outlined by a well-established series of rationally connected steps aimed at (i) defining a biological target (ii) screening large collections of compounds to identify hits (iii) hit-to-lead generation implying chemical modifications (iv) lead optimization for developing drug candidates and (v) performing preclinical trials validating a new potential drug, among others.