Oral treatment with the d-enantiomeric peptide D3 improves the pathology and behavior of Alzheimer's Disease transgenic mice

Abstract

Several lines of evidence suggest that the amyloid-β-peptide (Aβ) plays a central role in the pathogenesis of Alzheimer's disease (AD). Not only Aβ fibrils but also small soluble Aβ oligomers in particular are suspected to be the major toxic species responsible for disease development and progression. The present study reports on in vitro and in vivo properties of the Aβ targeting d-enantiomeric amino acid peptide D3. We show that next to plaque load and inflammation reduction, oral application of the peptide improved the cognitive performance of AD transgenic mice. In addition, we provide in vitro data elucidating the potential mechanism underlying the observed in vivo activity of D3. These data suggest that D3 precipitates toxic Aβ species and converts them into nonamyloidogenic, nonfibrillar, and nontoxic aggregates without increasing the concentration of monomeric Aβ. Thus, D3 exerts an interesting and novel mechanism of action that abolishes toxic Aβ oligomers and thereby supports their decisive role in AD development and progression.

Keywords: Alzheimer’s disease; Mirror image phage display; d-enantiomeric peptide; drugs; oligomers.

Figure 1

Lewis structure of the d-enantiomeric peptide D3.

Figure 2

Results of the transgenic animal studies. (A) Time needed to find the hidden platform in the Morris water maze assay by the mice that were treated with D3 added to the drinking water (open circles), by direct brain infusion of D3 (gray filled circles), or not treated (black filled circles). For details see the Methods section. Error bars indicate standard deviations. (B) Photomicrographs of coronal sections of the hippocampus stained for Aβ (W0−2 antibody) from control mice (left column), orally D3 treated mice (middle column), and mice treated by brain infusion (right column). (C) Quantitative evaluation of Aβ plaque load. Bars indicate the Aβ plaque load (area covered by Aβ) in the dorsal hippocampus of D3 treated and untreated mice. * indicates significantly different, p < 0.05.

Figure 3

Aβ oligomer modulation by D3. (A) Particle size analysis by dynamic light scattering (DLS). Aβ and D3 as well as mixtures of both were prepared seedless in filtered sodium phosphate/NaCl buffer. DLS measurements were carried out at 20 °C, with a fixed angle (90°) and a cuvette path length of 3 mm. Data acquisition time was 1 s using a 655.6 nm (13 mW) laser. By using calculated autocorrelation functions, a regularization fit was performed in order to obtain size distribution profiles at t = 0 min (t0, upper row) and after 10 min (lower row). (B) Analysis of Aβ aggregation by density gradient ultracentrifugation. The size distributions of 125 μM Aβ42 and Aβ42-D3 mixtures (1:1) were determined by sedimentation analysis on a preformed gradient of iodixanol (Optiprep, AXIS-SHIELD, Oslo, Norway). One hundred microliters of aggregation assays containing 125 μM Aβ42 without or with 125 μM D3 was directly overlaid on a step gradient of 5−50% iodixanol. After centrifugation, 14 fractions from top to bottom of the centrifuge tube of 140 μL each were harvested. The 15th fraction represents the pellet. The fractions were analyzed by SDS−PAGE and silver staining.

Figure 4

Analysis of amyloid properties of Aβ in the absence and presence of D3. (A and B) Electron micrographs of Aβ samples with and without D3. Twenty-five micromolar Aβ (A) and (1:4) Aβ-D3 (B) were negatively stained by uranyl acetate (1%) and measured as described in the Methods section. Scale bars: 100 nm. (C) Aβ fibril formation with and without seeds monitored by ThT fluorescence. Aβ only fibrils and Aβ-D3 coaggregates were prepared by incubating Aβ42 with and without different concentrations of D3 for 3 days at 37 °C. After washing, the precipitated seeds (20% v/v) were added to the aggregation reactions consisting of freshly prepared Aβ42 solutions. The relative ThT fluorescence of freshly prepared Aβ with fibrillar Aβ seeds (red), Aβ-D3 coaggregates (1:1, light blue), Aβ-D3 coaggregates (1:10, green), and freshly prepared Aβ as a control (purple) is shown as a function of time.

Figure 5

Computational analysis of the Aβ-D3 interaction. The initial structure of the Aβ(15−42) nonamer (A,B) and the structures of the unliganded nonamer (C,D) and the Aβ-D3 complex (E,F) after 50 ns of molecular dynamics simulations are shown as the top and side views, respectively. These two views differ by a rotation of approximately 90° around the vertical axis. The top layers of the oligomer are colored in light gray, and a color gradient is used for the lower layers of the oligomer. The electrostatic surface is shown for the initial structure in Figure 3A, in which the positive and negative surface charges originate from Lys16 and Glu22, respectively. Glu22 is shown as red sticks in panels B to F. D3 is shown in stick representation (colored according to the atom types) in panels E and F. A dotted line represents a disordered region of the terminal Aβ-chain. (G) Changes of the twist angle over the simulation time for the unliganded and D3-bound Aβ-oligomer. (H) Schematic presentation of a twisted Aβ-oligomer. The yellow arrows represents the vector between the Cα-carbons of Val18 and Val24. The angle between two vectors of adjacent chains was used to calculate the twist angle as described in the Methods section. Note the significantly enhanced twist angle in the D3-bound form (F) compared to the starting structure (B), which reflects the properties of the fibril.