Mechanisms Responsible for High Energy Radiation Induced Damage to Single-Stranded DNA Modified by Radiosensitizing 5-Halogenated Deoxyuridines.

Experimental studies showed that high energy radiation induced base release and DNA backbone breaks mainly occur at the neighboring 5' nucleotide when a single-stranded DNA is modified by radiosensitizing 5-halogenated deoxyuridines. However, no mechanism can be used to interpret these experimental observations. To better understand the radiosensitivity of 5-halogenated deoxyuridines, mechanisms involving hydrogen abstraction by the uracil-5-yl radical from the C2' and C3' positions of an adjacent nucleotide separately followed by the C3'-O3' or N-glycosidic bond rupture and the P-O3' bond breakage are investigated in the DNA sequence 5'-TU(•)-3' employing density functional theory calculations in the present study. It is found that hydrogen abstractions from both positions are comparable with the one from the C2' site slightly more favorable. The N-glycosidic bond cleavage in the neighboring 5' nucleotide following the internucleotide C2'-Ha abstraction is estimated to have the lowest activation free energies, indicating that the adjacent 5' base release dominates electron induced damage to single-stranded DNA incorporated by 5-halogenated deoxyuridines. Relative to the P-O3' bond breakage after the internucleotide C3'-H abstraction, the C3'-O3' bond rupture in the neighboring 5' nucleotide following the internucleotide C2'-Ha abstraction is predicted to have a lower activation free energy, implying that single-stranded DNA backbone breaks are prone to occur at the C3'-O3' bond site. The 5'-TU(•)-3' species has substantial electron affinity and can even capture a hydrated electron, forming the 5'-TU(-)-3' anion. However, the electron induced C3'-O3' bond rupture in 5'-TU(-)-3' anion via a pathway of internucleotide proton abstraction is only minor in both the gas phase and aqueous solution. The present theoretical predictions can interpret rationally experimental observations, thereby demonstrating that the mechanisms proposed here are responsible for high energy radiation induced damage to single-stranded DNA incorporated by radiosensitizing 5-halogenated deoxyuridines. By comparing with previous results, our work proves that the radiosensitizing action of 5-bromo-2-deoxyuridine is not weaker but stronger than its isomer 6-bromo-2-deoxyuridine on the basis of the available data.

In vitro and in vivo evaluation of novel antitumor prodrugs of 5-fluoro-2'-deoxyuridine activated by hypoxic irradiation.

PURPOSE: We previously developed a novel antitumor prodrug that has a 2-oxopropyl substituent at the N(1) position of 5-fluorouracil (5-FU) and releases 5-FU via one-electron reduction on hypoxic irradiation. Although the compound was effective in vivo, its activity against murine tumors was not high enough to warrant clinical studies. Therefore, we developed a similar family of radiation-activated prodrugs of 5-fluoro-2'-deoxyuridine (FdUrd), which is generally more potent than 5-FU, and investigated their radiation chemical reactivity and in vitro and in vivo effects. METHODS AND MATERIALS: Compounds bearing various 2-oxoalkyl substituents at the N(3) position of FdUrd were synthesized and investigated. After aerobic or hypoxic irradiation to the prodrugs dissolved in water or culture medium, release of FdUrd was measured using high-performance liquid chromatography. To investigate in vitro cytotoxicity, SCCVII and EMT6 cells in culture were irradiated in the presence of the prodrug under aerobic or hypoxic conditions, and then kept with the compound for 24 h. Cell survival was then measured using a colony assay. To investigate in vivo effects, the drug was injected intraperitoneally at a dose of 100 or 300 mg/kg into Balb/c mice bearing EMT6 tumors 30 min before irradiation. The tumor growth delay-time was then assessed. RESULTS: In vitro, the prodrugs released FdUrd at G-values (molar numbers of molecules produced by 1 J of radiation energy) of 1.6-2.0 x 10(-7) mol/J after hypoxic irradiation. The G-values for FdUrd release with hypoxic irradiation were about 100-fold greater than those with aerobic irradiation. Among the prodrugs tested, OFU106 bearing a 2-oxocyclopentyl substituent released the highest amount of FdUrd in the culture medium, and it was subjected to further in vitro and in vivo assays. Although OFU106 administered alone showed no cytotoxicity up to a concentration of 0.2 mM, it produced an enhanced cytotoxic effect when administered before hypoxic irradiation and kept with the cells for 24 h. The enhancement ratios calculated at the surviving fraction of 1% were 1.35-1.4 at 0.04 mM and 1.45-1.5 at 0.2 mM. In vivo, however, administration of OFU106 (100 or 300 mg/kg) before 20 Gy of irradiation did not produce marked growth delays compared with 20 Gy of radiation alone. CONCLUSION: On hypoxic irradiation in vitro, the prodrugs of FdUrd were activated as efficiently as were the prodrugs of 5-FU, but marked in vivo effects could not be detected. This strategy of prodrug design should be used in further development of radiation-activated prodrugs of more potent anticancer agents.

DNA duplexes containing photoactive derivatives of 2'-deoxyuridine as photocrosslinking probes for EcoRII DNA methyltransferase-substrate interaction.

EcoRII DNA methyltransferase (M.EcoRII) recognizes the DNA sequence 5'.CC*T/AGG.3' and catalyzes the transfer of the methyl group from S-adenosyl-L-methionine to the C5 position of the inner cytosine residue (C*). We obtained several DNA duplexes containing photoactive 5-iodo-2'-deoxyuridine (i(5)dU) or 5-[4-(3-(trifluoromethyl)-3H-diazirin-3-yl)phenyl]-2'-deoxyuridine (Tfmdp-dU) to characterize regions of M.EcoRII involved in DNA binding and to investigate the DNA double helix conformational changes that take place during methylation. The efficiencies of methylation, DNA binding affinities and M.EcoRII-DNA photocrosslinking yields strongly depend on the type of modification and its location within the EcoRII recognition site. The data obtained agree with the flipping of the target cytosine out of the DNA double helix for catalysis. To probe regions of M.EcoRII involved in DNA binding, covalent conjugates M.EcoRII-DNA were cleaved by cyanogen bromide followed by analysis of the oligonucleotide-peptides obtained. DNA duplexes containing i(5)dU or Tfmdp-dU at the central position of the recognition site, or instead of the target cytosine were crosslinked to the Gly(268)-Met(391) region of the EcoRII methylase. Amino acid residues from this region may take part both in substrate recognition and stabilization of the extrahelical target cytosine residue.

Direct radiation action of heavy ions on DNA as studied by ESR-spectroscopy.

The effects of heavy ions on the mechanisms of free radical formation in DNA-constituents as compared to low-LET irradiation are investigated by means of Electron Spin Resonance (ESR) spectroscopy. Dose-yield curves were measured at low (T < 100 K) and ambient temperatures in order to obtain the G-value, that is number of radicals formed per 100 eV absorbed energy. These G-values show a characteristic LET-dependence and are one to two orders of magnitude lower than for low-LET irradiation. Measurements on 2'Deoxycytidine at 300 K using combined heavy ion and X-ray irradiation methods suggested that this effect can be partially explained by a destruction of radicals during the irradiation.

Identification of the two cis-syn [2 + 2] cycloadducts resulting from the photoreaction of 3-carbethoxypsoralen with 2'-deoxycytidine and 2'-deoxyuridine.

Near-ultraviolet photolysis of 2'-deoxycytidine (dCyd) and 3-carbethoxypsoralen (3-CPs) in the dry state was found to generate two main stable photoadducts which were separated by thin-layer and high-performance liquid chromatography. Fast atom bombardment and plasma desorption mass spectrometry analyses suggested that the bound molecule to 3-CPs is dCyd. These two compounds were found to produce the corresponding 2'-deoxyuridine (dUrd) derivatives through a deamination process when left in aqueous solutions with a lifetime close to 24 h at 20 degrees C. The chemical structure of the deaminated photoadducts was confirmed by photochemical synthesis using dUrd as the substrate. UV and fluorescent measurements indicated that the furan moiety of 3-CPs is involved in the photobinding reaction. The cyclobutane type structure of the modified dUrd derivatives was established on the basis of its photoreversibility and detailed 1H NMR analysis. The cis-syn stereoconfiguration of the two photocycloadducts was inferred from coupling constant considerations and on the basis of the complete assignment of the cyclobutyl protons, requiring the synthesis of deuterated nucleosides at pyrimidine carbon C(6). Further confirmation of the diastereoisomeric relationship between the two cis-syn dUrd <54' 65'> 3-CPs was provided by circular dichroism measurements.

Heavy-ion-induced free radical formation in solid DNA-constituents: quantitative and structural aspects.

Electron spin resonance (ESR) spectroscopy was used to study free radical formation in solid, polycrystalline pellets of DNA-constituents. Dose-yield curves were obtained at low (ca 90 K) and ambient temperatures, and were analyzed for initial G-values (radicals per 100 eV absorbed energy). At both temperatures, values of one or more orders of magnitude below the corresponding x-ray values were found. Combination of x-ray and heavy-ion irradiation indicated an enhanced radical destruction as probable cause of reduced G-values. Structural aspects as analyzed from the ESR-powder spectra revealed differences in initial, low-temperature radical population and in decay reactions upon annealing between heavy-ion bombardment and low-LET irradiation.

Free radicals induced in solid DNA by heavy ion bombardment.

Free radical formation after heavy-ion bombardment was studied in solid, polycristalline pellets of DNA-constituents by analysing the dose-yield curves and the spectra obtained by ESR-spectroscopy at low (< 100 K) and ambient temperatures. The dose-yield curves were found to correlate with those found after X-irradiation but shifted to higher doses and lower saturation concentrations. The corresponding radical yields (per 100 eV) exhibit values which are one to two orders of magnitudes lower. The structural aspects as revealed from powder ESR-spectra gave a complex inter-relation between substance, LET, dose and irradiation temperature, which is discussed in terms of mechanistic models.