Enhancing the therapeutic effect of biological drugs with protein engineering : Focusing on pre-clinical Alzheimer’s disease therapy

Sammanfattning: Aggregation of the amyloid-β peptide (Aβ) is one of the main pathological hallmarks in Alzheimer’s disease (AD). The soluble Aβ aggregates (oligomers and protofibrils) have shown to be the most harmful species. Hence, targeting these aggregates can be of therapeutic potential.Protein therapy is one of the fastest growing fields in drug development with more than 100 FDA approved protein-drugs in the last decade. Despite that, protein-drugs (mainly antibodies) targeting Aβ displayed limited beneficial effects in AD clinical trials. This might be attributed to the presence of the blood-brain barrier (BBB) that hinders the entry of big molecules such as proteins into the brain. In paper I, we fused somatostatin peptide (SST) to the previously developed BBB transporter (scFv8D3). The new protein, SST-scFv8D3, exhibited a 120-times longer plasma half-life compared to SST, and reached the brain at high levels when intravenously administered. When tested in APPswe mouse model of AD, SST-scFv8D3 significantly enhanced neprilysin (NEP)-mediated degradation of hippocampal Aβ42 after only three injections. In paper II, treatment with SST-scFv8D3 displayed a wide-ranging effect on AD brain proteome. Mitochondrial and neuronal growth proteins were among the most altered protein-groups, where SST-scFv8D treatment shifted them towards wild-type levels.There is potential to increase the binding strength and selectivity of antibodies to small Aβ aggregates (oligomers), which are thought to be the most toxic Aβ species. In paper III, we developed a multivalent antibody format with additional binding sites having short distances between them. The new antibody format displayed a 40-fold reduction in the dissociation rate from Aβ protofibrils. Furthermore, the multivalent antibody could strongly bind small Aβ oligomers, which has been difficult to achieve with conventional IgG antibodies. In paper IV, we developed a bispecific version of the multivalent antibody capable of passing the BBB. A single intravenous injection of the new antibody format was enough to significantly clear soluble Aβ aggregates from the brain of tg-ArcSwe mice. In paper V, we developed recombinant proteins with NEP linked to an Fc-fragment to provide long half-life and to the above-mentioned BBB transporter. When applied at therapeutic doses, these proteins significantly degraded plasma Aβ, but displayed limited effects on brain Aβ concentration, probably due to their short retention times in the brain.In conclusion, we developed new protein-drugs with improved binding properties to Aβ, ability to cross the BBB, and therapeutic potential in pre-clinical mouse models of AD.