The contribution of hydrogen peroxide to the radiosensitizing effect of gold nanoparticles.

Plasmid DNA in aerated aqueous solution is used as a probe to determine whose of the reactive oxygen species (ROS) generated after absorption of ultra-soft X-rays (USX) take part in biomolecule damage in the presence and in absence of Gold Nano-Particles (GNP) and specific scavengers. Citrate-coated GNPs with core sizes of 6, 10 and 25 nm are synthetized and characterized, especially in terms of plasmon band shift, ζ-potential and hydrodynamic radii (respectively 9, 21 and 30 nm). We confirm the radiosensitizing effect of GNP and show that the SSB number per plasmid increases when, for a same mass of gold element, the core size of the gold nanoparticles decreases. Hydroxyl radicals (OH) are scavenged using the positively-charged 2-amino-2-hydroxymethyl-1,3-propanediol (TRIS) and the neutral dimethyl sulfoxide (DMSO) molecules. Due to both negatively-charged environments of DNA and GNP, at identical scavenging capacity, TRIS is more effective at quenching OH than DMSO. The strong radiosensitizing effect of hydroxyl radicals is confirmed. Methanoate anions are then used to transform OH into hydrogen peroxide; the latter being known to be non-aggressive regarding DNA in the absence of easily oxidable metallic ions (Fenton reactions). Surprisingly, in the presence of GNP, high DNA damage yields are observed even though hydrogen peroxide might not be hold as responsible. Conversely, the radiosensitizing effect of GNP is not observed anymore when H2O2 is scavenged using pyruvate ions. We demonstrate that hydrogen peroxide constitutes quite unexpectedly a hidden stock of OH which are activated at the surface of the GNP by decomposition of H2O2 molecules.

DNA strand break dependence on Tris and arginine scavenger concentrations under ultra-soft X-ray irradiation: the contribution of secondary arginine radicals.

In this study, we used a bench-top cold-cathode ultra-soft X-ray (USX) generator to expose aqueous DNA plasmid solutions to low-LET radiation under various scavenging conditions. Single- and double-strand breaks were assessed using classic gel electrophoresis quantification of linear, circular and supercoiled plasmid DNA topologies. With their very low penetration range in water, USX can only interact with matter up to short distances, of the order of 50 µm. We validated a stirring procedure which makes it possible to expose 100 µL of aqueous samples (2 mm thick). The scavenging of OH radicals by Tris buffer was studied at ambient temperature under aerobic conditions and compared to data gathered in the literature. A very good agreement was found with the rare data dealing with DNA plasmid exposed to Al Kα photons at low temperature (T ≤ 277 K), which therefore validated the experimental procedure. The yields for DNA single-strand breaks determined during this study enabled the ratio of indirect to direct effects to be determined at 96.2%, in good agreement with the value of 97.7% stemming from a study based on γ-ray irradiation of frozen solutions of plasmid DNA. Then, arginine was used both to create a "biological-like" chemical environment around the DNA plasmids and as an OH radical scavenger, in vitro. Although arginine has a greater scavenging (protecting) power than Tris, surprisingly, it led to higher rates of strand breakage. Based on the specific binding modes of arginine to DNA, we suggest that the side effects observed are due to the presence of arginine near to, but also inside, the DNA double helix.