Fragment-to-Lead
Fragment-to-lead optimization may proceed by multiple mechanisms. In general, it is very helpful to have access to structural data; ideally experimentally derived although for some well characterized protein families, molecular modeling and computational chemistry can provide guidance. Zenobia scientists have experience optimizing fragments to early-stage lead compounds for over 20 targets including multiple diverse protein families. The most common methods used by our scientists are fragment growing and fragment merging.
Fragment growing
Using the crystal structure of a fragment hit as a guide, the fragment can be grown into other binding pockets of the protein. To keep molecular weight low, we often build into each putative pocket individually to determine the effect on potency and for large protein classes, selectivity. Ligand efficiency and chemical properties are monitored closely throughout the process. Existing inhibitors either from high-throughput screening or the literature and crystal structures may also be used as a guide during this process.
After each sub-site is explored, results are combined into a starting point usually of nM potency and ligand efficiency of > 0.4. Not all sites may be incorporated into the final lead molecule as they may not contribute effectively to potency or selectivity. Using structure-directed design, the lead molecule is further fine tuned to improve in vitro and in vivo ADME PK parameters towards the goal of delivering a lead compound appropriate for in vivo proof of concept studies and late stage lead optimization. Most often, this lead compound is delivered to our partner for additional optimization with Zenobia continuing to provide structural biology support.
Using the crystal structure of a fragment hit as a guide, the fragment can be grown into other binding pockets of the protein. To keep molecular weight low, we often build into each putative pocket individually to determine the effect on potency and for large protein classes, selectivity. Ligand efficiency and chemical properties are monitored closely throughout the process. Existing inhibitors either from high-throughput screening or the literature and crystal structures may also be used as a guide during this process.
After each sub-site is explored, results are combined into a starting point usually of nM potency and ligand efficiency of > 0.4. Not all sites may be incorporated into the final lead molecule as they may not contribute effectively to potency or selectivity. Using structure-directed design, the lead molecule is further fine tuned to improve in vitro and in vivo ADME PK parameters towards the goal of delivering a lead compound appropriate for in vivo proof of concept studies and late stage lead optimization. Most often, this lead compound is delivered to our partner for additional optimization with Zenobia continuing to provide structural biology support.
Fragment merging (or scaffold hopping)
In many cases, overlay of the assembly of all fragment hits gives considerable insight into favorable ligand-protein interactions. This may suggest "merging" fragment functionality together thereby modifying the core (fragment hopping) to improve potency or overall chemical properties. In addition, R-group functionality may be interchanged. In all cases, the goal is to provide a better starting point for follow-on synthesis or to improve intellectual property protection and freedom to operate.
An overlay of the x-ray structures for all of the fragment we identified for PDE10 are shown below.
In many cases, overlay of the assembly of all fragment hits gives considerable insight into favorable ligand-protein interactions. This may suggest "merging" fragment functionality together thereby modifying the core (fragment hopping) to improve potency or overall chemical properties. In addition, R-group functionality may be interchanged. In all cases, the goal is to provide a better starting point for follow-on synthesis or to improve intellectual property protection and freedom to operate.
An overlay of the x-ray structures for all of the fragment we identified for PDE10 are shown below.
