Author:

Abstract:

Introduction: We developed a novel fetal mouse model for pulmonary gene therapy using adeno-associated viral vectors (AAV). With complex microsurgical protocol only a limited number of animals can be injected thus resulting in small cohorts for further follow-up. Non-invasive bioluminescence imaging (BLI) using firefly luciferase as reporter gene reduces the number of animals exposed to experimentation, but the absence of sensitive immunohistochemical detection and the limited spatial resolution limit the application of this technique. Therefore we investigated the combination of BLI with magnetic resonance imaging (MRI) to obtain an overlay image which combines the surface BLI signal with a visualisation of the deeper anatomical structures. Methods: Time mated pregnant NMRI mice underwent surgery at E17 (term=E19). Fetuses were injected intratracheally with adeno-associated viral vectors encoding the reporter genes firefly luciferase for BLI and β-galactosidase for immunohistochemistry. Operated pups were delivered by caesarean section and fostered. Pups were followed up by serial BLI once a week until the age of 1 month. At week 1 combined BLI – MRI images were acquired of all operated pups to determine the gene expression pattern. A second BLI – MRI image was obtained before sacrificing the animals. After collection of lung samples, X-gal staining was performed on frozen sections. As a control condition a few animals were injected in the liver. To combine BLI and MRI, a compatible animal holder was used to perform the imaging without repositioning the animals within the same anaesthesia session. For co-registration of BLI and MRI, the photographic images (from BLI) were overlaid with the coronal whole-body 2D RARE MRI (150 µm inplane resolution, 400 µm slice thickness). Limb position and size were used to overlay BLI and MRI. Results: BLI imaging overlaid with the standard photographic image of the scanned animal showed a signal emanating from the neck and the upper thoracic region. Co-registration of MRI with BLI showed that luciferase gene expression was situated in the pulmonary region. Final confirmation of gene expression in lung epithelium however was obtained after X-gal staining to visualise LacZ positive cells. Animals injected in the liver showed a different expression pattern with a BLI signal emanating from the abdominal region co-localizing with the liver on MRI. Conclusions: Together with the development of a novel fetal mouse model for pulmonary transduction using AAV vectors, we optimized a non-invasive follow-up method of injected animals combining BLI and MRI. BLI gives an indication of the site of gene expression, but by combining this with MRI more detailed anatomical information can be obtained. In the future, this model of imaging-guided gene therapy for the mouse lung can be used to further investigate the application of gene therapy for genetic disorders such as cystic fibrosis, surfactant deficiencies or α-antitrypsin deficiency.