Substituted Aβ42 peptides produced six different classes of oligomer distributions (Fig. 2, Aβ42 panel, colored arrows), compared with two in the Aβ40 case. The largest class (red) was characterized by decreased amounts of pentamers and hexamers. The orange class (D-[H14,Q15] and in addition to displaying exceptionally faint pentamer and hexamer bands, produced a diffuse band with an most consistent with trimer, but with substantial areas of this band migrating at significantly lower Mr. These same characteristics were observed with the dimer band. D-[D7, S8] (purple arrow) was unique in that it showed increased intensity of tetramer, pentamer, and hexamer. The D-[I31,I32] peptide also was notable because it displayed the lowest amount of monomer relative to the other bands in the lane. D-[S26,N27] also was unique in that decreases in hexamer and trimer band intensities were observed in concert with increased tetramer intensity (green arrow). D-[L34,M35] and D-[M35,V36] peptides displayed exceptionally intense pentamer and hexamer bands, but less intense trimer bands (blue arrows). D-[V39,V40] and D-[I41,A42] showed exceptionally intense dimer and trimer bands, but tetramers were difficult to detect (black arrows).
To determine the effects of di-D-amino acid substitutions on the time dependence of β-sheet formation, a proxy for Aβ fibril formation 33-35 , we monitored T (ThT) fluorescence (Fig. 3). The average maximum intensity (FUmax) of Aβ40 was 200 fluorescence units (FU). Aβ40 variants D-[V12,H13], D-[K16,L17], D-[V18,F19], D-[F20,A21], D-[E22,D23], D-[N27,K28], D-[L34,M35], and D-[V39,V40] displayed substantially greater (>2×) intensities. D-[Glu3, Phe4], and D-[Asp7,Ser8] had very similar maximum intensities, all of which were ≈2× that of wild type Aβ40. The intensities of D-[Asp23,Val24] and <1/10, respectively, than that of wild type Aβ40. The time at which half maximal fluorescence intensity occurred (t1/2) was ~220 h for Aβ40. D-[D01,A02], D-[E03, F04], D-[R05,H06], D-[H14,Q15], D-[V18,F19], D-[E22,D23], and D-[S26,N27] displayed substantially shorter (<1/2×) half times (Fig. 3, Table 1). D-[V12,H13], D-[N27,K28], and D-[A30,I31], and D-[L34,M35] displayed longer half times, all of which were (<2×). We also calculated the change in fluorescence per unit time (dFU/dt) during the quasi-linear phases of growth for each peptide (Table 1). Cross-correlation of the dFU/dt, and FUmax determined by linear fitting to each of the respective scatter plots revealed a correlation (R=0.73) between dFU/dt and FUmax (data not shown). No correlations were observed dFU/dt and t1/2 or between FUmax and t1/2.
In experiments using di-D-amino acid substituted Aβ42 variants, the range of maximum fluorescence intensities was ~50-350, approximately an order of magnitude lower than that of the Aβ40 peptides. Aβ42 displayed a maximum fluorescence intensity of ≈130 FU, thus the relative differences in intensity between the substituted peptides and wild type Aβ42 often were much smaller than in the Aβ40 case. D-[M35,V36] and D-[H14,Q15] produced the highest intensities, ≈3× that of wild type Aβ42. The intensities of D-[I41,A42], D-[F20,A21], and D-[Ser26, Asn27] were ≦2× that of wild type Aβ42. D-[V12,H13], D-[Asn27,Lys28], and D-[Ala31, Ile32] had the lowest maximal intensities. The remaining peptides fluoresced with intensities less than or equal to that of wild type Aβ42. The t1/2 for WT Aβ42 was 90 h. With the exception of D-[Phe20, Ala21], which had a t1/2 of 190 h, all of the other substituted peptides displayed t1/2 values of ~10 h. Many di-D-amino acid substitutions resulted in somewhat slower emanation of β-sheet structure, a phenomenon also observed in the Aβ40 system, as well as producing increased half times. D-[Y10,E11], D-[H14,Q15], D-[S26,N27] and D-[A30,I31] assembled the slowest, with half times that all were > 200 h. Only D-[E22,D23] had a half time that was substantially lower than that of WT Aβ42 (68 vs. 90 h).