MULTIMODAL MOLECULAR IMAGING FOR HIF-1ALPHA NON-INVASIVE ASSESSMENT IN A MURINE GLIOMA MODEL

Introduction: Molecular imaging is an emerging discipline in biomedical research that allows the visual representation, the early characterization and quantification of biological processes at the cellular and subcellular level directly in vivo. Imaging techniques are of great importance since they allow the translation of the data obtained in preclinical models directly to the clinic. In this research, molecular imaging has been applied to the study of a preclinical glioma model, to assess intratumoural hypoxia, as a negative prognostic factor, since it has been correlated to chemo and radio-resistance, tumour progression, increase of invasiveness and to a poor prognosis. For these reasons, hypoxia represents a promising therapeutic target, both in GBM but also in other types of cancer, and an interesting marker able to estimate the prognosis and to predict the therapeutic outcome. In this perspective, the molecular study of hypoxia presented herein was focused mainly on the evaluation of the transcriptional activity of HIF-1α , which is a key factor of the hypoxia process. Objectives: The main purpose of this project was to study, in a non invasive manner, HIF-1α transcriptional activity as a result of the development and progression of neoplastic disease and after treatment with selective inhibitors and drugs already used in the clinical setting. The first objectives were the characterization and validation of the proposed cell and animal models and the evaluation of HIF-1α activity after treatments of various inhibitors and drugs, to understand its potential role as a prognostic biomarker. The results obtained could be used in the evaluation of the possibility of performing personalized treatments as a result of stratification of individuals depending on HIF-1α activity. Materials and Methods: The human U251 glioma cell line was used after its appropriatey engineering in order to obtain three models: in the first, U251-HRE, the luciferase reporter gene is under the control of the HRE sequences; in the second, U251-pGL3, the luciferase gene is expressed constitutively, whereas in the third, U251-HRE-mCherry, the mCherry reporter gene is constitutively expressed while the luciferase reporter gene is again under the control of HRE sequences. These lines were first used in vitro for the study of different agents modulating the transcriptional activity of HIF- 1α, whose action has been evaluated primarily by biochemical luciferase assay. Subsequently, the same cells were inoculated orthotopically in nude mice and tumor growth-related events, transcriptional activity of HIF-α, intratumoural hypoxia and therapeutic response, were analyzed using different imaging techniques (BLI, FLI, PET and MRI). The data obtained in vitro and in vivo were also validated ex vivo by histological and immunohistochemical staining. Results: The in vitro studies have shown that in U251-HRE model,the luciferase activity can be modulated through hypoxia mimetic drugs, such DFX, and it was related to HIF-1α nuclear accumulation In addition, in vitro experiments have allowed to evaluate the contribution of different molecular pathways (involving PI3K/Akt and Ras/MEK/ERK) on HIF-1α activity. As expected, in U251-pGL3 model, the luciferase activity was not modulated. In vivo studies, performed after cell injection in murine models receiving U251-pGL3 cells, showed a linear increase of luciferase activity, proportionally to tumour growth. In contrast, in U251-HRE models a bimodal trend was observed, since it was dependent on hypoxia establishment HIF-1α mediated. These results suggested that this model can be used both in vitro and in vivo for the study of the modulation of HIF-1α activity. Although optical imaging cannot be used in humans, cross- validation of the results obtained with techniques routinely used in the clinic, such as PET and MRI (which gave results comparable and complementary to those provided by optical imaging), allowed to enhance the value of the results obtained in bioluminescence and fluorescence studies due to the possibility to translate these procedures to the clinical setting. The preliminary study conducted with the TMZ has shown, both in vitro and in vivo, that previously to the onset of the cytotoxic effect of the treatment, an early reduction of HIF-1α activity was detectable. The mechanism of action and the significance of this finding, however, have to be explored by further studies. Conclusions: The U251-HRE model, recapitulating GMB features, could be a reliable model for the study of HIF-1α role in tumour progression. In this context, the optical imaging could be considered a sensitive and versatile technique for the study of the processes related to tumour progression especially those mediated by HIF-1α, but it is limited since it is impossible to translate these data into the clinics. For these reasons, it is extremely important the cross-validation with imaging techniques routinely used in the clinical practice, such as PET and MRI. Moreover, HIF-1α activity could be considered as an early efficacy biomarker after treatment with TMZ, although this crucial process has to be studied in depth.