Humanization of Nanobodies and identification of Nanobodies for Alzheimer's disease immunotherapy.

Almost a century ago, antibodies were envisioned as 'magic bullets' for the specific targeting of disease sites, but until recently they struggled to meet their expectations. Clinical success with monoclonal antibodies has now reinforced the role of these molecules as a major source of drugs for the treatments of human diseases. Furthermore, protein engineering led to the design of smaller recombinant antibody fragments, such as Fabs (~50 kDa) and single-chain Fv fragments (~25 kDa), which are emerging as credible alternatives. Even single immunoglobulin variable domains have been engineered and shown to retain their antigen-binding capacity and functionality. It is anticipated that these domain antibodies (dAbs) will expand the repertoire of antibody-based reagents against a vast range of novel biomarkers. Unfortunately, dAbs are often associated with poor stability and solubility and low levels of expression. Nanobodies, single-domain antigen-binding fragments of camelid-specific heavy-chain only antibodies offer in this respect special advantages over classical antibody fragments because of their smaller size, robustness, improved solubility and a high degree of specificity and affinity for their antigen without requiring VL-domain pairing. The beneficial properties of Nanobodies have been attributed to the unique presence of four hallmark amino acids in the framework-2 region (positions 42, 49, 50, 52) and a longer third antigen-binding loop (H3) folding over this area. Human dAbs have been engineered, i.e. camelized, by integrating in their sequences the Nanobody hallmark residues, in an attempt to develop well-behaved entities. Additional studies suggested that other mutations within a VH domain might contribute to the solubility of dAbs as well. We evaluated, in this study the substitution of a residue at position 118, abutting the CDR3 region as an alternative approach to generate an autonomous human dAb library. However, in many aspects, Nanobodies remain superior to camelized VH domains. Obviously, for therapeutic applications, Nanobodies require to be humanized, i.e. mutating camelid-specific amino acid sequences in the framework to their human VH equivalent. We performed this humanization exercise on Nanobodies from different VHH subfamilies and investigated the effects on their biochemical and biophysical properties. We demonstrated that the humanization of Nanobody-specific residues outside framework-2, are neutral to the Nanobody properties. Surprisingly, the Glu49Gly and Arg50Leu humanization of hallmark amino acids generates a single domain that is more stable though probably less soluble. The other framework-2 substitutions, Tyr/Phe42Val and Gly/Ala52Trp, are detrimental for antigen affinity, due to a repositioning of the H3 loop as shown by their crystal structures. In addition, imprinting the sequence signature of a human VH in a VHH restores the VL binding capacity of a Nanobody with a short CDR3 loop that does not cover the former VL binding site. These insights were employed to identify a soluble, stable, well expressed universal humanized Nanobody scaffold that allows grafts of antigen-binding loops from other Nanobodies for transfer of the antigen specificity and affinity. Nanobodies are expected to provide new binding specificities and are creating possibilities for the development of therapeutic compounds to treat amyloid disorders such as Alzheimer's disease, a neurodegenerative impairment evolving towards a major public health issue. The amyloid cascade hypothesis holds that generation and deposition of the amyloid ?-peptide (A?; 40-42 amino acids), obtained by cleavage of the amyloid prec