We have previously generated an affibody molecule for the disease-associated amyloid beta (A beta) peptide, which has been shown to inhibit the formation of various A beta aggregates and revert the neurotoxicity of A beta in a fruit fly model of Alzheimer's disease. In this study, we have investigated a new bacterial display system for combinatorial protein engineering of the A beta-binder as a head-to-tail dimeric construct for future optimization efforts, e.g. affinity maturation. Using the bacterial display platform, we have: (i) demonstrated functional expression of the dimeric binder on the cell surface, (ii) determined the affinity and investigated the pH sensitivity of the interaction, (iii) demonstrated the importance of an intramolecular disulfide bond through selections from a cell-displayed combinatorial library, as well as (iv) investigated the effects from rational truncation of the N-terminal part of the affibody molecule on surface expression level and A beta binding. Overall, the detailed engineering and characterization of this promising A beta-specific affibody molecule have yielded valuable insights concerning its unusual binding mechanism. The results also demonstrated that our bacterial display system is a suitable technology for future protein engineering and characterization efforts of homo- or heterodimeric affinity proteins.

The amyloid hypothesis suggests that accumulation of amyloid β (Aβ) peptides in the brain is involved in development of Alzheimer's disease. We previously generated a small dimeric affinity protein that inhibited Aβ aggregation by sequestering the aggregation prone parts of the peptide. The affinity protein is originally based on the Affibody scaffold, but is evolved to a distinct interaction mechanism involving complex structural rearrangement in both the Aβ peptide and the affinity proteins upon binding. The aim of this study was to decrease the size of the dimeric affinity protein and significantly improve its affinity for the Aβ peptide to increase its potential as a future therapeutic agent. We combined a rational design approach with combinatorial protein engineering to generate two different affinity maturation libraries. The libraries were displayed on staphylococcal cells and high-affinity Aβ-binding molecules were isolated using flow-cytometric sorting. The best performing candidate binds Aβ with a K_D value of around 300 pM, corresponding to a 50-fold improvement in affinity relative to the first-generation binder. The new dimeric Affibody molecule was shown to capture Aβ_1-42 peptides from spiked E. coli lysate. Altogether, our results demonstrate successful engineering of this complex binder for increased affinity to the Aβ peptide.