Aβ42-binding peptoids as amyloid aggregation inhibitors and detection ligands

Abstract

Alzheimer's disease (AD) is the most common form of dementia and currently affects 5.4 million Americans. A number of anti-Aβ (beta amyloid) therapeutic agents have been developed for AD, but so far all of them failed in clinic. Here we used peptoid chemistry to develop ligands selective for Aβ42. Peptoids are N-substituted glycine oligomers, a class of peptidomimics. We synthesized an on-bead peptoid library consisting of 38,416 unique peptoids. The generated peptoid library was screened and arrays of Aβ42-selective peptoid ligands were identified. One of those peptoid ligands, IAM1 (inhibitor of amyloid), and the dimeric form (IAM1)2 were synthesized and evaluated in a variety of biochemical assays. We discovered that IAM1 selectively binds to Aβ42, while the dimeric derivative (IAM1)2 has a higher affinity for Aβ42. Furthermore, we demonstrated that IAM1 and (IAM1)2 were able to inhibit the aggregation of Aβ42 in a concentration-dependent manner, and that (IAM1)2 protected primary hippocampal neurons from the Aβ-induced toxicity in vitro. These results suggest that IAM1 and (IAM1)2 are specific Aβ42 ligands with antiaggregation and neuroprotective properties. IAM1, (IAM1)2, and their derivatives hold promise as Aβ42 detection agents and as lead compounds for the development of AD therapeutic agents.

Figure 1

Structure and composition of peptoid library. (A) General structure of the peptoid library containing two constant monomers and four variable monomers (R1–R4). (B) Primary amines used for the synthesis of monomers. The abbreviation in the parentheses under each amine refers to the corresponding monomer formed from that amine.

Figure 2

Quantitative binding affinity of peptoid IAM1 to Aβ42 and Aβ40. (A) Chemical structure of the peptoid IAM1 identified from the screen of the library. (B) Chemical structure of a random peptoid (RP) used as a negative control. (C) Biotin-Aβ42 and biotin-Aβ40 binding assay with IAM1 and RP peptoid beads. (D) Principle of solid phase binding assay with fluorescent readout. (E) Solid-state binding curve for IAM1 peptoid using synthetic Aβ42 and Aβ40. (F) Solid-state binding curve for RP peptoid using synthetic Aβ42 and Aβ40. In panels (E) and (F), the average fluorescence reading at each Aβ concentration is shown as mean ± SE (n = 3). The average fluorescence data were fitted with a nonlinear regression curve using one site binding equation.

Figure 3

Inhibitory ability of IAM1 toward the aggregation of Aβ42 and Aβ40 using the in situ kinetic thioflavin T (ThT) assay. (A, B) Time course of the fluorescence of aggregate-bound ThT in the presence of Aβ42 (A) or Aβ40 (B) and different compounds. The RP (100:1) was used as a negative control and anti-Aβ antibody 6E10 (20-fold dilution) as a positive control. The scyllo-Inositol (500:1) was used as a reference compound. (C, D) Time courses of the fluorescence of aggregate-bound ThT in the aggregation processes of Aβ42 (C) or Aβ40 (D) in the presence of IAM1 at different concentrations. Molar ratio of IAM1:Aβ in the range from 1:1 to 100:1 as indicated. (E) The normalized ThT_max values for the Aβ42 and Aβ40 aggregation processes are plotted as a function of IAM1 concentration. The data in each aggregation experiment were normalized to the ThT_max value obtained in the presence of DMSO, averaged and shown as mean ± SEM (n = 3).

Figure 4

Evaluation of the dimeric derivative (IAM1)2. (A) Chemical structure of the dimeric derivative (IAM1)2. (B) The binding curves of (IAM1)2 with Aβ42 and Aβ40 using fluorescence solid phase binding assay. The average fluorescence reading at each Aβ concentration is shown as mean ± SE (n = 3). The average fluorescence data were fitted with a nonlinear regression curve using one site binding equation. (C, D) Time courses of the fluorescence of aggregate-bound ThT in the aggregation processes of Aβ42 (C) or Aβ40 (D) in the presence of (IAM1)2 at different concentrations. Molar ratio of (IAM1)2:Aβ in the range from 1:1 to 10:1 as indicated. (E) The normalized ThT_max values for the aggregation processes of Aβ42 and Aβ40 is plotted as a function of (IAM1)2 concentration. The data in each aggregation experiment were normalized to ThT_max value obtained in the presence of DMSO, averaged and shown as mean ± SEM (n = 3).

Figure 5

Evaluation of ASR1. (A) Chemical structure of ASR1. (B) The binding curves of ASR1 with Aβ42 and Aβ40 using fluorescence solid phase binding assay. The average fluorescence reading at each Aβ concentration is shown as mean ± SE (n = 3). The average fluorescence data were fitted with a nonlinear regression curve using one site binding equation (C, D) Time courses of the fluorescence of aggregate-bound ThT in the aggregation processes of Aβ42 (C) or Aβ40 (D) in the presence of ASR1 (50:1).

Figure 6

Neuroprotective effects of (IAM1)2 in amyloid toxicity assay. (A) MAP2 staining of neurons incubated in the control medium (upper panel) and in Aβ-containing conditioned medium (lower panel). The images are shown for DMSO, 6E10 antibodies, 100 nM RP, and 100 nM (IAM1)2 as indicated. (B) The normalized MAP2 signals for hippocampal neurons treated with Aβ-containing conditioned medium in the presence of 6E10 antibodies, RP or (IAMI)2 at increasing concentrations. The MAP2 signals were normalized to the signals in the wells that received only control medium in sister cultures. The normalized and averaged data presented as mean ± SE (n = 3). *p < 0.05; **p < 0.01 when compared to DMSO control.