A study on the properties and reactivity of naphthoquinone-cobalt(III) prototypes for bioreductive prodrugs.

Our group has recently initiated a study on the development of new prototypes for bioreductive prodrugs, based on Co(III) complexes with the ligand 2,2'-bis(3-hydroxy-1,4-naphthoquinone), H2bhnq. The focus of this work is to investigate the dissociation of bhnq(-2) from the complex upon reduction, and the effects of pH, redox potential, oxygen concentration and nature of the auxiliary ligands on this reaction. The bhnq(2-) ligand is a "non-cytotoxic" agent that was chosen as a probe for the reactivity studies due to its suitable chromophoric properties, at the same time that it resembles more cytotoxic naphthoquinones relevant for cancer therapy. In this way, two Co(III) complexes [Co(bhnq)(L1)]BF4·H2O (1) and [Co(bhnq)(L2)]BF4·H2O (2) (L1=N,N'-bis(pyridin-2-ylmethyl)ethylenediamine and L2=N,N'-dimethyl-N,N'-bis(pyridin-2-ylmethyl)ethylenediamine) were synthesized and fully characterized. The gallium analogs [Ga(bhnq)(L1)]NO3·3H2O (3) and [Ga(bhnq)(L2)]NO3·3H2O (4) were also prepared for helping with the assignments of the redox properties of the cobalt complexes and the structure of 2. Cyclic voltammetry analysis revealed a pH-independent quasi-reversible Co(III)/Co(II) process at -0.22 and -0.08V vs NHE for 1 and 2, respectively. An O2-dependent dissociation of bhnq(2-) was observed for the reaction of 1 with ascorbic acid. For 2, the dissociation of bhnq(2-) was found to be independent on the concentration of O2 and faster than in 1, with little influence of the pH on both complexes. The difference in reactivity between 1 and 2 and their redox properties, among other factors, suggests that 1 undergoes redox cycling, pointed out as a key feature for a prodrug to achieve hypoxic selectivity.

The effect of the molecular structures of dicyanomethylene compounds on their supramolecular assembly, photophysical and electrochemical properties.

Two series of flexible dicyanomethylene compounds, specifically, class 1 and class 2 compounds, have been designed and synthesised. In class 1 compounds, the dicyanomethylene groups are separated by glycol chain spacers of different lengths, whereas, in class 2 compounds, the spacers are alkyl linkers of different lengths. The notion underlying the design of these compounds is that in class 1 molecules, the spacers contain donor oxygen atoms that could not only form hydrogen bonds during the course of crystal packing but also promote withdrawing effects that modify the photophysical and electrochemical properties of these molecules in solution; in contrast, these effects would be absent for class 2 molecules. However, this study revealed that, with respect to crystal packing, the size of the spacers and their even and odd numbers of atoms are more important than their chemical nature. All of the synthesised compounds exhibited blue emission in the solid state and in CH2Cl2 solutions. The photophysical and electrochemical properties of these compounds in solution were not significantly affected by the type and length of the spacer that was used in each molecule. In the solid state, however, the compound with the shortest spacer showed the highest Stokes shift. The electronic transitions for the synthesized compounds in solution were explained by density functional theory and time-dependent density functional theory calculations, which indicated that the methylene moieties control the properties of both classes of compounds and that the spacers do not conjugate with the end groups. These two series of flexible dicyanomethylene compounds could be utilised as molecular building blocks for the development of new solids with novel properties.