We report the synthesis of an original series of oxoazabenzo[de]anthracenes conjugated to an amino acid: Ala, Phe, Pro, Lys, or Gly (4a-e, respectively). The compounds, derived from 1,8-dihydroxyanthracene-9,10-dione, were studied for DNA binding and cytotoxicity. Melting temperature, fluorescence quenching, and surface plasmon resonance methods all indicated that the lysine derivative 4d binds to DNA much more strongly that the Pro, Ala, and Gly conjugates whereas the Phe analogue showed the lowest DNA binding capacity. These compounds form intercalation complexes with DNA, as judged from electric linear dichroism and topoisomerase I-based DNA unwinding experiments. Preferential binding of 4d to defined sequences such as 5'-CTAAAGG and 5'-ATGC was evidenced by DNase I footprinting. This Lys conjugate was found to be over 20 times more cytotoxic to CEM human leukemia cells than the other conjugates, with an IC50 in the submicromolar range. A high antiproliferative activity, likely attributable to the enhanced DNA binding capacity, is maintained despite the incapacity of the compound to stabilize topoisomerase-DNA covalent complexes. The cell cycle effects of 4d consisted in an S phase accumulation of cells coupled with a pro-apoptotic action (appearance of hypodiploid sub-G1 cells) which were confirmed by measuring the inhibition of BrdU incorporation into DNA and labeling of phosphatidylserine residues with annexin V-FITC by means of flow cytometry. Altogether, the work provides interesting structure-activity relationships in the oxoazabenzo[de]anthracene-amino acid conjugate series and identifies the lysine derivative 4d as a promising candidate for further in vivo evaluation and drug design.

Multiple myeloma (MM) is an incurable B-cell cancer characterised by the monoclonal proliferation of tumour cells in the bone marrow (BM). It has been described that matrix metalloproteinases (MMPs) and especially MMP-9 is secreted by MM cells. In this study, we investigated the possibility to exploit MMP-9 activity to activate prodrugs and to target MM cells as a new tumour-specific therapy. Cleavage of the prodrug EV1-FITC by MMP-9 resulted in release of fluorescence which can be used as a measure of prodrug activation. The 5T33MM mouse model was used in this proof-of-principle study. The prodrug was activated in a higher amount by addition to MMP-9-producing 5T33MMvv cells, homogenates from tumour-bearing organs (BM, spleen) and isolated 5T33MM-diseased BM and spleen cells compared to non-MMP-9-producing 5T33MMvt cells and homogenates/cells from nontumour-bearing organs/ mice, as measured by fluorescence release. This fluorescence release could be inhibited by the MMP-2/MMP-9-specific inhibitor, CTT. Activation of the prodrug in the 5T33MM spleen and BM homogenates was confirmed by chromatography. EV1-fluorescein isothiocyanate injection into 5T33MM-diseased animals resulted in a higher fluorescence release by the isolated BM and spleen cells compared to injection into healthy animals. In conclusion, MMP-9 activity can be used to activate prodrugs that target MM.