Aβ oligomer formation
Several experiments were executed to determine whether or not D-amino acid substitutions in Aβ40 and Aβ42 altered peptide behavior. Specifically, we sought to determine which substitutions would affect oligomerization, fibril formation kinetics, secondary structure formation, and morphology. From the PICUP and SDS-PAGE analysis of single D-amino acid substituted Aβ, individual amino acids were identified which alter oligomer formation in Aβ40 and in Aβ42 (Fig. 10). In Aβ40, the oligomerization of D-S26 and D-N27 differed most notably from WT when single D-amino acid substitutions were made (Fig. 10). S26 and N27 thus significantly affect Aβ40 oligomerization. These results are consistent with previous in hydro and in silico studies which postulated the importance of S26 in V24-K28 turn formation, a structure proposed to be important for the intramolecular nucleation of unfolded Aβ monomer to folded monomer 40, 41 . Other D-amino substituted variants such as D-H14, D-F20, D-A21, D-I31, and D-I32 also affected peptide oligomerization but in a more subtle way (Fig. 10). D-H14, D-F20, D-I31, and D-I32 thus may play more minor roles in controlling the oligomer formation of Aβ40. While we were able to identify single D-amino acid substitutions that changed the oligomer distribution of Aβ40, some of these single D-amino acid substituted variants did not differ in Aβ oligomer distribution while their di-D-amino acid substitution counterparts did, for Aβ40. We observed a smear of protein produced by Aβ40 D-[K16,L17] (Fig. 2a) – a smear that was not observed with Aβ40. However, the single D-K16 and D-L17 substitutions did not seem to produce oligomer distributions different than that of WT (Fig. 4b). This suggests that the interactions between amino acids and both K16 and L17 may be important in controlling the oligomerization of Aβ40, but also that there may be redundancy in the function of K16 and L17. Another interpretation would be that the D-amino acid substitutions in the di-D-amino acid substituted variants may be acting synergistically to alter the oligomer distribution of Aβ, explaining why changes in oligomer distribution were not observed with K16 and L17 were substituted individually.
While we observed several differences in the oligomer distribution of Aβ40 when di-D-amino acid substitutions were made, even fewer single D-amino acid substitutions altered the oligomer distributions of Aβ40. For instance, D-[K16,L17] shifted the oligomer distribution of Aβ40 toward higher order oligomers (Fig. 2a). However, when K16 and L17 were substituted with their D-enantiomers individually, we did not observe a substantial change in the oligomer distribution compared to Aβ40 (Fig. 5b).
Meanwhile, fourteen individual amino acids seemed to affect Aβ42 oligomerization: D01, H14, Q15, F20, A21, D24, A30, I31, I32, L34, M35, V39, V40, and I41 (Fig. 10). The D-amino acid substitutions at the C-terminus exhibited the greatest effect on Aβ oligomer formation, causing both the non-cross-linked and cross-linked Aβ42 variants at the C-terminus to display oligomer distributions more like Aβ40, with the absence of higher order oligomers (Fig. 10). These results are consistent with reports suggesting that a turn in the C-terminus plays an essential function in oligomerization and the assembly differences between Aβ40 and Aβ42 32, 41-43. In this study, we found that single amino acid substitutions of several amino acids (e.g., I31, I32, L34, M35, V39, V40, and I41) at the C-terminus of Aβ42 altered oligomer formation, thereby identifying amino acids in the C-terminus that control oligomerization. This suggests that the exact orientations of each of these amino acid side chains are essential in making Aβ42 behave as it does, and that altering any one of these can reduce the formation of higher order oligomers. This also suggests that the native sequence, with amino acids in their L-conformation, is uniquely suited to form high order oligomers. Furthermore, we identified distinct sets of amino acids that controlled the oligomer formation of Aβ40 and Aβ42, indicating that different residues govern the assembly of the two isoforms and confirming that the two isoforms form oligomers through distinct pathways 10. The densitometry analysis of the oligomer distribution for Aβ42 variants supports the qualitative analysis (Fig. 5d). Importantly, the cross-linking results are consistent with previous studies, where the C-terminus of Aβ was established as an important site in controlling oligomerization 32, 41, 43 .