Gene therapy for cystic fibrosis in a mouse model using adeno-associated viral vectors

Cystic fibrosis (CF) is the most common monogenic life-threatening disease in the Caucasian population, caused by mutations in CFTR, a chloride/bicarbonate channel that regulates fluid transport across epithelium of different organs (airways, pancreas, intestine, sweat glands and vas deferens). CF affects multiple organs, but lung pathology is the major clinical manifestation. Mutations in the CFTR gene lead to an imbalance in ion and water transport followed by the formation of viscous mucus which lines the lung epithelium and contributes to vicious cycle of airway obstruction, persistent infection and inflammation resulting in irreversible decline of lung function. For the majority of CF patients the treatment is symptomatic and there is no cure available.

Gene therapy holds promise to cure a wide range of genetic and acquired diseases. Recently, rAAV-based gene therapy made remarkable success showing efficacy for several hereditary disorders and progressing to the market with a first gene-based product approved by the European Medicines Agency. This encouraged us to re-explore rAAV-based gene therapeutic approach for cystic fibrosis. Furthermore, the conceptual advantage of an early treatment gains more and more support in the gene therapy field since several studies showed beneficial effect of an early treatment on the clinical outcome. In that light, early gene therapy, prior to disease onset, may even prevent disease, rather than having to cure it.

In parallel with the immune system disorders, CF was at the forefront of the gene therapy field since its inception in early 1990`s. Since then, more than 20 clinical trials for CF have been conducted, but none have led to a persistent clinical benefit. Hence the question on how to further improve pulmonary gene transfer remains unanswered. In that perspective, the work presented in this thesis aimed at developing a preclinical strategy for viral vector-based gene therapy, as a first step towards a cure (curative treatment) for CF airway disease.

In a first part of my thesis, I developed and characterized a model for pulmonary gene transfer using adeno-associated viral vector with airway tropism (rAAV2/5) carrying reporter genes in fetal, neonatal and adult mice to answer generic questions on stability and efficacy of gene transfer. Combining non-invasive bioluminescent imaging and histological analysis, efficient transduction of both the upper (nose) and the lower (lung) mouse airways was observed. In order to cure an inherited disease, like CF, lifelong expression of the therapeutic gene is required. rAAV2/5-mediated gene transfer in the fetal or neonatal mouse airways, followed by a single re-administration later in adult animals resulted in sustained gene expression up to 7 months, without a marked decrease. In addition, we demonstrated that a single dose of rAAV2/5 administered to adult mice also resulted in reporter gene expression at least up to 15 months, which corresponds to the expected life-span of a mouse (1.5-2 years). This generic model highlights the clinical potential of rAAV2/5 vector for treatment of inherited and acquired pulmonary disorders for which long-term gene correction is required and no effective therapy is available.

In a last part of this thesis, I translated the above-mentioned generic rAAV2/5-based technology for pulmonary gene transfer to a therapeutic model for CF gene therapy. Compared to previous unsuccessful rAAV-based clinical trials, we designed an improved vector based on an airway-tropic serotype (rAAV2/5) and carrying a truncated, but functional transgene (CFTRΔR) that allows incorporation of an external promoter for optimal gene expression. Combining different functional tests (iodide efflux assay, patch-clamp and forskolin-induced swelling), we demonstrated that CFTRΔR, a CFTR version that lacks a portion of the regulatory R-domain, retains ion channel activity and it is regulated by cAMP/PKA pathway in cultured cells. Finally we assessed the therapeutic potential of rAAV-CFTRΔR vector in two complementary models: intestinal organoids derived from CF patients and in vivo across the nasal mucosa of CF mice homozygous for ΔF508 mutation. A single dose of the therapeutic vector rAAV-CFTRΔR restored chloride secretion in both models proffering a rAAV-based gene therapy option for CF, a small but promising step forward opening avenues towards a curative treatment for CF.