Enabling synthesis in fragment-based drug discovery (FBDD): microscale high-throughput optimisation of the medicinal chemist's toolbox reactions

Abstract

Miniaturised high-throughput experimentation (HTE) is widely employed in industrial and academic laboratories for rapid reaction optimisation using material-limited, multifactorial reaction condition screening. In fragment-based drug discovery (FBDD), common toolbox reactions such as the Suzuki-Miyaura and Buchwald-Hartwig cross couplings can be hampered by the fragment's intrinsic heteroatom-rich pharmacophore which is required for ligand-protein binding. At Astex, we are using microscale HTE to speed up reaction optimisation and prevent target down-prioritisation. By identifying catalyst/base/solvent combinations which tolerate unprotected heteroatoms we can rapidly optimise key cross-couplings and expedite route design by avoiding superfluous protecting group manipulations. However, HTE requires extensive upfront training, and this modern automated synthesis technique largely differs to the way organic chemists are traditionally trained. To make HTE accessible to all our synthetic chemists we have developed a semi-automated workflow enabled by pre-made 96-well screening kits, rapid analytical methods and in-house software development, which is empowering chemists at Astex to run HTE screens independently with minimal training.

Fig. 1. Fragment elaboration occurs along well-defined growth vectors (red arrows) and is often achieved from a pre-functionalised intermediate (e.g., halogenated analogue 1) using palladium-catalysed cross couplings. However, Lewis basic heteroatoms which are required to facilitate fragment–protein binding (blue) can deactivate the palladium-catalyst resulting in catalyst sequestration via unproductive off-cycle processes (a–c).

Fig. 2. Our recent analysis of 131 published fragment-to-lead (F2L) examples shows that most bonds formed in F2L originate on carbon-based vectors and involve C–C (53.4%), C–N (32.6%) and C–O (9.2%) bonds.

Fig. 3. The HTE optimisation workflow developed at Astex allows non-specialists to quickly screen 96 reaction conditions for a chosen bond-forming reaction, with minimal training. This semi-automated workflow is enabled by extensive upfront technology development from automation specialists, software developers and analytical chemists.

Fig. 4. Reaction conditions retrieved from Astex ELN data are shown on the left-hand side, these historical datasets include successful single-shot batch reactions and one-factor at a time optimisation (OFAT) reactions. They are limited to the top 90% to remove single entries or infrequently used conditions. When these results are compared to the Astex HTE screening plates on the right-hand side, the HTE plates offer the ability to screen more conditions in tandem whilst covering a larger expanse of optimisation space compared to traditional OFAT approaches.

Fig. 5. The Suzuki screen of brominated 7-azaindole analogue 1 and 2-fluoroboronic acid 2 was run on 10 μmol scale with multiple hits for desired compound 3 found via MISER analysis. The top conditions were repeated in batch (0.25 mmol scale) affording 89% yield of desired product 3. Literature conditions reported for close analogue 3-a were not directly transferable to generate 3via Suzuki coupling affording only 10% LCAP and 6% isolated yield, similar conditions were also captured in the plate screen (well B5) and were also poor yielding. The hit rate of wells from the HTE screen with a normalised response ≥66.7% = 7/96 = 8%.

Fig. 6. The primary amine Buchwald–Hartwig screen. Unhindered primary amine: hit conditions identified for product 4 correlate to 69% conversion (LCAP) affording 92% isolated yield on batch scale up (0.25 mmol), literature conditions reported for close analogue 4-a did not transfer to our desired compound and similar conditions contained in the plate screen (well C5) were also poor yielding. Hindered primary amine: hit conditions for product 5 correlate to 49% conversion (LCAP) resulting in 41% isolated yield on batch scale up (0.25 mmol), literature conditions reported for close analogue 5-a did not transfer to our desired compound and similar conditions contained in the plate screen (well B3) were also poor yielding. The hit rate of wells from the HTE screen with a normalised response ≥66.7% for unhindered amine screen = 3/96 = 3%, and hindered amine screen = 1/96 = 1%.

Fig. 7. The secondary amine and aniline Buchwald–Hartwig screen. Secondary amine: hit conditions for product 9 correlate to 24% conversion (LCAP)affording 61% isolated yield on batch scale up (0.25 mmol). The top three conditions from the plate screen (wells C3, C4, C5) were performed in batch and conditions C5 were found to be the highest yielding (see: ESI for more details). The literature conditions reported for close analogue 9-a also transferred well to furnish our desired compound and similar conditions contained in the plate screen (well C4) were also efficacious. Aniline: hit conditions for product 10 correlate to 76% conversion (LCAP) aff in 86% isolated yield on batch scale up (0.25 mmol), literature conditions reported for close analogue 10-a did transfer to our desired compound, although similar conditions contained in the plate screen (well D6) were poor yielding. The hit rate of wells from the HTE screen with a normalised response ≥66.7% for secondary amine screen = 9/96 = 9%. Hit rate for aniline screen = 4/96 = 4%.

Fig. 8. The third Buchwald–Hartwig screen contains conditions reported for weak nucleophiles and oxygen nucleophiles. Primary amide: hit conditions for product 11 correlate to 92% conversion (LCAP) affording 87% isolated yield on batch scale up (0.25 mmol). The literature conditions reported for close analogue 11-a also transferred well to furnish our desired compound although similar conditions contained in the plate screen (well A9) did not. Alcohol: hit conditions for product 12 correlate to 89% conversion (LCAP) affording 84% isolated yield on batch scale up (0.25 mmol), literature conditions were not performed on scale as the catalyst was not commercially available. The hit rate of wells from the HTE screen with a normalised response ≥66.7% for amide nucleophile screen = 3/96 = 3%, and oxygen nucleophile screen = 3/96 = 3%.

Fig. 9. The nucleophilic aromatic substitution (S_NAr) screen was tested using fragment-like indazole 13 and benzyl alcohol 14. To the best of our knowledge. There were no reported conditions in the literature for synthesising compound 15 by S_NAr, so this is an important example of where a HTE screen can help. Hit rate for oxygen nucleophile screen with normalised product ≥66.7% = 1/96 = 1%.