To continue on the theme of understanding your compound of interest’s mechanism of action, let’s discuss target identification (ID). You have just completed your screen and found a hit, but now you want to learn more about why it is a hit. As mentioned before, understanding mechanism of action can greatly help with the development of biomarkers for preclinical studies, understanding why your compound may be toxic or if your compound will affect other targets, and also will help you further derivatize your compound to make it more specific or potent.
There are many approaches to be used for target ID as depicted in the chart below.
At Perlstein Lab we are starting with a high through-put phenotypic screen, an unbiased approach to drug discovery, meaning we do not know our drug target beforehand. Therefore we have to determine what the targets of our hits are following the screen.
One way to identify a target after a phenotypic screen is via pull-downs. That involves conjugating your compound to a solid support to identify protein binders from cell lysates. Conjugating to a solid support could alter the activity of the compound so other approaches can be used, such as conjugating the compound to a small tag and treating cells with the modified compound. This would then be followed by a pull-down. Or the compound can be conjugated to a photoreactive group such that the compound can be covalently linked to the target. All these techniques are followed by mass spec for protein target identification.
A more detailed explanation of this is on the Broad Institute’s site. Assuming there is decent SAR (structure-activity relationship) data, a compound can be tethered to a solid support at an end that is not important for activity. Then pull-downs are performed with SILAC (stable isotope labeling by amino acids in cell culture) labeled lysates. Once candidate targets are identified, it is to be followed up with biochemical assays such as RNAi knock-downs, microscopy and SPR.
There are also label free approaches to target identification such as DARTS (Drug Affinity Responsive Target Stability). The idea behind this is that the compound of interest is more susceptible to proteolysis when unbound and less susceptible when bound to the target, as depicted in the diagram below. This is an alternative approach if conjugation of the compound leads to inactivity.
Other techniques involve expression cloning, such as expressing cDNA libraries in cells, exposing those cells to tagged compound, and pulling down the compound-target complex. Another expression cloning tool is phage display, where each phage particle is expressing a unique protein on the surface. The phage library is then passed over immobilized compound to identify the target of interest. The binding phage pool is collected, amplified and further selected on the target.
The approaches listed above are all permutations of in vitro and in vivo techniques, but there are also in silico approaches, where compound structures with known targets can be compared to the hit structure in order to identify the unknown target or mechanism of action. One in silico tool available to researchers in academia and industry is Target Hunter. The researchers who developed this tool use a predictive algorithm to search your compound against those in the ChEMBL database. Additionally there is a geo tool to allow researchers to find collaborators who can help them identify the targets experimentally.
Many of these examples were pulled from a great review article on target identification when starting with phenotypic screens. Check it out!