Consistent with the known tolerance
Consistent with the known tolerance of the Suzuki–Miyaura reaction to a wide range of aryl substituents, we subsequently found that the commercially available 3-hydroxyphenylboronic acetylcholine chloride pinacol ester () could be coupled to , to give directly and in comparable yield. Smooth conversion of into was readily achieved with the mild triflating agent -phenyltriflimide. Prior to undertaking library synthesis, a coupling reaction was conducted between triflate () and phenylboronic acid under previously optimised reaction conditions (Pd(PPh), KCO, dioxane, reflux). The resulting 8-biphenylchromenone () proved identical to that prepared previously by an alternative method. Analogous cross-coupling reactions were conducted with and 10 commercially available arylboronic acids in a GreenHouse™ reactor (Radleys) to furnish the target arylchromenones (–) (). Given the commercial availability of 1,3-phenylene-bis-boronic acid (), the possibility of undertaking a ‘one-pot’ double Suzuki–Miyaura coupling reaction with chromenone triflate () and the appropriate bromoheterocycle (ArBr) was also investigated. Again, a model reaction conducted with , the triflate and bromobenzene under microwave conditions (5min at 150°C) confirmed the viability of the reaction, with being obtained in good yield (60%). This approach was utilised for the preparation of compounds () as shown in . Replacement of the 8-phenyl substituent of by a substituted thienyl, thiazolyl or pyridyl group was achieved by coupling either the chromenone triflate () or the chromenone-8-boronate () with the appropriate heterocyclic boronic acid or dihaloheterocycle, respectively, as shown in . Subsequent Suzuki–Miyaura arylation of the intermediate with a range of arylboronic acids afforded the target heterobiaryl-3-ylchromenones (–). In all cases, reaction progress was monitored by LC–MS analysis and the products were purified by semi-preparative HPLC. The DNA-PK inhibitory activity of both compound libraries is summarised in , . Our ongoing programme to develop inhibitors of DNA damage-activated kinases as radio- and chemo-potentiators in cancer therapy has resulted in the identification of a number of potent and kinase-selective inhibitors of DNA-PK, most notably NU7441 (). That significant inhibitory activity also resides in the simple 8-biphenylchromenone () was surprising, in light of our previous structure–activity studies indicating that an extended planar aromatic system at the chromenone 8-position is a prerequisite for potent DNA-PK inhibition. Previous studies had also demonstrated that chromenones bearing a biphenyl-3-yl group (e.g., ) were more active than the corresponding biphenyl-2-yl and biphenyl-4-yl isomers. This is consistent with the likely disposition of the 3-phenyl group of relative to the terminal aryl ring of the dibenzothiophen-4-yl substituent in within the ATP-binding pocket. The fact that is approximately 10-fold more potent than the parent 8-phenylchromenone LY294002 (, IC=1.6 μM) strongly suggests that the 3-phenyl substituent of is making additional binding interactions within the ATP-binding domain of DNA-PK. The overall objective of this study was thus to probe this putative binding interaction further, with a view to delineating SARs and improving potency. For the small series of 4-substituted biphenyl-3-ylchromenones evaluated (–), it is evident that substitution did not improve activity, and with the exception of the 4-hydroxy derivative (), which proved to be equipotent with the biphenyl-3-ylchromenone (), a 5- to 10-fold reduction in potency was observed (). A modest improvement in activity over compounds – was observed for the 3,5-disubstituted derivatives ( and ), but both compounds were less potent than . By contrast, replacement of the 3-phenyl group of by an isosteric thiophen-2-yl () or a thiophen-3-yl () substituent improved DNA-PK inhibitory activity approximately 10-fold, with and exhibiting IC values of 18nM and 20nM, respectively. Interestingly, the introduction of a methyl group onto the thiophene ring (–) proved detrimental to inhibitory activity, with a 50-fold reduction in potency being observed for derivatives bearing a methyl group to the heteroatom ( and ). This is consistent with the evidence adduced previously indicating limited steric tolerance at this position. The high potency exhibited by the thiophene derivatives and implies that a small electron-rich aryl ring is a prerequisite for DNA-PK inhibitory activity, and this is supported by the activity of phenol . However, although the reduced potency of the corresponding 3-furanylchromenones ( and ), the thiazole () and the imidazole derivatives ( and ) is consistent with this proposal, the weak activity of the analogous pyrrole derivative () was unexpected. Perhaps not surprisingly, replacement of the pendant phenyl ring of by a pyridyl or pyrimidinyl heterocycle (–) resulted in a loss of potency.