Nitric oxide-mediated modulations of the tumor oxygenation : blood flow and radiation response

Oxygen is a key environmental factor in the development of tumors and their response to treatment. The pO2 plays important roles in the response of tumors to cytotoxic treatments such as chemotherapy, radiotherapy, and photodynamic therapy. Both oxygen diffusion and oxygen consumption by metabolism in tumor cells contribute to the occurence of hypoxia. Oxygen deficiency is caused by an insufficient oxygen supply as a result of inadequate tumor perfusion (chronic hypoxia) and fluctuation in red cell flux (acute hypoxia). An interesting approach to the problem consists in the manipulation of tumor blood flow and oxygen delivery in order to improve either radio- or chemotherapeutic response. The aim of this thesis was to evaluate the potential interst of nitric oxide-mediated treatments in terms of tumor hemodynamic parameters and tumor radiosensitivity. The nitric oxide dependent treatments under investigations were (1) i.p. injection of a NO donor (isosorbide dinitrate) ; (2) i.v. insulin infusion and (3) electrical stimulation of the host tissue. Two differnet tumor models were used in this study : a transplantable liver tumor model (TLT) and the syngeneic FSaII tumor model were implanted in the thight of NMRI and C3H mice. Tumor oxygenation was monitored using EPR (Electron Paramagnetic Resonance) oximetry and OxiLite™. Tumor blood flow was evaluated by OxyFlo™ and constrastenhanced MRI. Oxygen consumption was determined by EPR spectroscopy. Finally, the radiosensitizing properties of the treatments were studied in vivo via the determination of FSaII tumor regrowth delays. The three treatments induced a prolonged increase in tumor oxygenation on both tumor models. Isosorbide dinitrate administration induced an increase in tumor perfusion as a consequence of NO delivery that resulted in an increase in tumor pO2. Contrarily, the increase in tumor pO2 during and after insulin infusion was not due to an increase in flow (which was even decreased as shown using flash MRI), but to a decrease in tumor cells oxygen consumption. This mechanism turned out to be NO mediated (inhibition of complex I and IV of the mitochondrial respiratory chain) and was demonstrated by immunoblotting (eNOS phosphorylation), by the up-regulation of the tumor cGMP content and by inhibition of the effects using a NOS inhibitor. The electrical stimulation protocol induced NO production that resulted in a rapid increase in tumor oxygenation (flow effect) and that was relayed by a NO dependent decrease in tumor cells oxygen consumption after the end of the protocol. Isosorbide dinitrate injection and insulin infusion radiosensitized the FSaII tumor model with the same efficacy or even with more efficacy than carbogen breathing (reference treatment). Up to this point, we have classically attributed the radiosensitizing properties of those approaches to the single oxygen effect. Nevertheless, we could not explain all our observations with this unique concept. There was indeed no direct relationship between the efficacy of irradiation and the increase in tumor pO2 that was reached after a given treatment. Accordingly, as those three treatments were active via a nitric oxide pathway, we made the assumption that NO could be involved in tumor radiosensitivity. In order to validate or discredit this hypothesis, NO was detected ex-vivo using an EPR spin-trapping technique and its bioavailability was assessed by the tumor cGMP quantification after each treatment. We established a positive correlation between the relative level of tumor NO content and the factor of regrowth delay after irradiation. By this way, we showed the potential intrinsic radiosensitizing properties of nitric oxide in vivoThèse de doctorat en sciences pharmaceutiques -- UCL, 200