No abstract

The process of developing antibody drugs for Alzheimer's disease therapy has been both long and difficult; however, recent advances suggest that antibodies against neurotoxic Αβ42 can suppress the progression of AD, especially on its early stage. Here, we obtained and characterized a novel anti-oligomeric Aβ42 aggregate scFv antibody, HT7, which could induce the significant disaggregation of Aβ42 aggregates through the release of stable and non-cytotoxic hexameric complexes that were composed of three scFv HT7s and one Aβ42 trimer, the latter being found to serve as the assembled subunit within larger Aβ42 aggregates in addition to existing freely between the cells. The docking model of the scFv HT7-Aβ42 complex revealed that only the N-terminal peptide of the Aβ42 molecule was bound into the groove between the VH and VL domains of scFv HT7. Thus, it was suggested that the hydrophobic interaction between the C-terminal peptides of Aβ42 molecules maintained the stability of the Aβ42 trimers or the Aβ42 trimer subunits. The saturation of Aβ42 trimer subunits by scFv HT7 and the subsequent dissociation of the scFv HT7-saturated Aβ42 trimer subunits from larger Aβ42 aggregates constituted the primary mechanisms underlying the high efficacy of scFv HT7. Our findings revealed that it was not sufficient for an anti-oligomeric Aβ42 antibody to exhibit high specificity and high affinity to the oligomeric Aβ42 aggregates in order to promote Aβ42 aggregate clearance and neutralize their cytotoxic effects. Here, for the first time, we proposed a "post-saturation dissociation" mechanism of Aβ42 oligomeric subunits for effective anti-Aβ42 antibodies.

Amyloid β peptide (Aβ42) is a major determinant of Alzheimer's disease (AD). In this study, we studied a novel single-chain variable fragment (scFv), AS, generated from an antibody library of AD patients, which recognized and bound specifically to medium-size amyloid β peptide (Aβ42) oligomers and immature protofibrils (25-55 kDa) and, more importantly, reduced their level by blocking their formation or inducing their disassembly. Consequently, scFv AS ameliorated or prevented their cytotoxicity and protected SH-SY5Y cells and primary cultured neurons in vitro from their damage in a concentration-dependent manner. Comparison of its cytotoxicity-inhibiting and cytotoxicity-neutralizing activities indicated that scFv AS displayed its protective effect on target cells mainly due to its cytotoxicity-inhibitory activity though it could also neutralize the cytotoxicity. We also found that scFv AS could efficiently cross the in vitro BBB model with a delivery efficiency of over 70% after a 60-min post-administration. The scFv AS was a monovalent antibody with an affinity constant (KD) of 5.5 × 10(-6) M and a binding threshold of 6.25 × 10(-4) μM for Aβ42 oligomers. The molecular docking simulations of Aβ42 to scFv AS revealed that scFv AS tends to approached Aβ42 oligomers and immature protofibrils mainly by their hydrophobic interaction and then drew Aβ42 molecule into the gap between VL and VH domains of scFv AS by hydrophilic interaction between scFv AS and the N-terminal region (residues 1-15) of Aβ42 and the hydrophobic interactions between scFv AS and the middle region (residues 20-33) of Aβ42. The combination of scFv AS with Aβ42 was realized likely through an induced-fit process.

Figures 2c and 7c of the original articles unfortunately contain mistakes. The authors would like to correct these figures and are now elaborated as follows:(1) In Fig. 2c, the names of the two lanes should be swapped left-right. In the second line from the bottom of its caption, the word "left" should be "right." (2) Aβ42-Q26 in Fig. 7c should be Aβ42-N27 Correct Figs. 2 and 7 are hereby published.The online version of the original article can be found at http://dx