Kinetic studies of DNA interstrand crosslink by nitrogen mustard and phenylalanine mustard

Phenylalanine mustard (PAM) and nitrogen mustard (HN2) are bifunctional alkylating agants which covalently crosslink DNA. Their crosslinking ability forms the basis of their usefulness as anti-cancer drugs since crosslinked DNA cannot replicate and thus the cancers cells cannot reproduce. Despite apparent similarities, the two drugs are known to have important differences. PAM is more effective against cancer and has fewer side effects. In reactions with cell cultures, PAM produces crosslinks more slowly, but the crosslinks are more persistant The experiments reported here offer kinetic explanations for the differences. In vitro time-dependent crosslinking reaction profiles for HN2 and PAM produced using an alkaline ethidium bromide assay for crosslinked DNA are virtually identical to what has been reported in vivo using an alkaline elution method. This suggests that the differences between the two drugs can be explained by purely chemical processes. In particular, the difference in the persistence of PAM and HN2 crosslinks in vivo does not require differences in the rate of enzymatic repair of the two types of lesions since loss of crosslinks in vitro follows the same schedule. A novel modification of the ethidium bromide assay reveals that PAM and HN2 have similar solution stabilities in neutral sodium phosphate buffer. This rules out the possibility that the pharmocological differences between PAM and HN2 are attributable to differential solution stabilities. The dependence of the rate of the crosslinking reaction on drug concentration is linear for both drugs. This confirms previous in vivo studies which suggested that the crosslinking reaction is pseudo first order for drug concentration. The dependence of the reaction rate on the temperature shows that PAM and HN2 have similar activation energies for the crosslinking reaction but PAM has a less favorable steric factor as predicted based on PAM's bulky phenylalanine portion. The crosslinked products for PAM and HN2 were isolated and their rates of decomposition were shown to be first order. PAM crosslinks have a longer half-life and a larger portion of PAM's crosslinks resist decay. These observations may be sufficient to explain the differences between PAM's and HN2's reaction profiles in vivo and in vitro and may be responsible for PAM's greater therapeutic value. A model is proposed to account for PAM's longer half-life and greater fraction of decay-resistant crosslinks.