Abstract
The d(GGCGCC)(2) palindrome is encountered in several oncogenic and retroviral sequences. In order to target it, we previously designed several oligopeptide derivatives of the mitoxantrone and ametantrone anticancer intercalators. These have two arms with a cationic side-chain in the major groove, each destined to bind along each strand O(6)/N(7) of the two successive guanine bases (G1-G2/G1'-G2') upstream from the central anthraquinone intercalation site. We retained from a previous study (El Hage et al., 2022) a tris-intercalating molecule with two outer 9-aminoacridine (9-AA) intercalators, denoted as III. We sought enhancements in both affinity and selectivity by simultaneously targeting the minor groove of the extracyclic -NH(2) groups of these bases and G4-G4' of the intercalation site. We considered derivatives of distamycin, having each pyrrole ring replaced by an imidazole to act as an in-register electron acceptor from the -NH(2) group of a target guanine. We substituted the C(6) and C(7) carbons of anthraquinone, or the C(8) and C(9) ones of anthracycline, by an (imidazole-amide)3 chain. Four different derivatives of III were designed with different connectors to the anthraquinone/anthracycline and 9-AA. Polarizable molecular dynamics simulations of their complexes with a double-stranded DNA 18-mer with a central d(C GGGC GCCC G)(2) palindrome sequence showed in-register minor groove binding to -NH(2) of G1-G(2)/G1'-G2' to coexist with major groove recognition of O(6)/N(7). Up to 12 H-bonds could be stabilized in the minor groove coexisting with four bidentate interactions of the alkyl diammonium moieties in the major groove. Since there is no mutual interference, the binding enthalpies, ΔH, contributed by each groove could add up and enable significant enhancements of the affinity constants. As was the case for their Lys precursor, these derivatives are amenable to chemical syntheses and in vitro and in vivo tests, for which the present results provide an incentive. The construction of derivatives III-A-III-D is modular. For in vitro experiments, this should enable unraveling the most important structural elements to further optimize both ΔH and TΔS and sequence selectivity and how this could translate to in vivo tests.