High-Throughput Screening at the Membrane Interface Reveals Inhibitors of Amyloid-β

Abstract

Aggregation and the formation of oligomeric intermediates of amyloid-β (Aβ) at the membrane interface of neuronal cells are implicated in the cellular toxicity and pathology of Alzheimer's disease. Small molecule compounds have been shown to suppress amyloid aggregation and cellular toxicity, but often the presence of a lipid membrane negates their activity. A high-throughput screen of 1800 small molecules was performed to search for membrane active inhibitors, and 21 primary hits were discovered. Through the use of fluorescence-based assays, transmission electron microscopy, and dot blot assays, the initial 21 primary hits were narrowed down to five lead compounds. Nuclear magnetic resonance and circular dichroism experiments were used for further confirmation of amyloid inhibition at the membrane interface and to obtain insights into the secondary structure of amyloid-β, while size exclusion chromatography was used to characterize the size of Aβ species. Lastly, dye-leakage assays allowed us to understand how the addition of the five lead compounds affected amyloid-β's ability to permeate the lipid bilayer. These results provide insights into small molecules that stabilize small amyloid species in the presence of membranes for the development of tool compounds for deeper investigations of these transient species.

Figure 1.. Experimental screening for amyloid inhibitors.

(A) An illustration of the components of the screen: Aβ₄₀ monomer (2LFM), LUVs, small molecules and ThT. (B) A schematic of reading the assay plates before aggregation and then heating and shaking the plates before reading the final fluorescence intensity after 24 hours of incubation; time points for reading are indicated by arrows. (C) Percentage of inhibition as calculated from final fluorescent intensity of every compound screened for inhibition. Value of twin plates are averaged and normalized with respect to the positive (red dotted line) and negative (blue dotted line) controls.

Figure 2.. Small molecule amyloid inhibitors identified from screening.

21 primary small molecule hits chosen after initial screen and CRC (concentration response curve) testing. The names of the compounds are abbreviated based on the similarities between the functional groups; additional details on these compounds are given in Tables S1, S2 and S3.

Figure 3.. Effects of small molecules on Aβ aggregation kinetics.

ThT fluorescence kinetics for the 21 selected compounds from CRC testing. 10 μM Aβ₄₀ was mixed with varying concentrations of small molecule (10, 5, & 2 equivalents) and 500 μM LUVs in 20 mM phosphate buffer with 50 mM NaCl at 37 °C with slow shaking.

Figure 4.. Effect of small molecules on antibody reactivity for Aβ.

Signal intensities from the dot blot assay using the OC antibody. Samples were measured after ThT experiments with 5 equivalents of a compound in respect to Aβ₄₀. Red line indicates 50% signal.

Figure 5.. TEM images of small molecules induced Aβ species.

TEM images showing the morphologies of Aβ₄₀ aggregates containing the indicated compounds that produced less than 50% antibody reactivity from the dot blot assay. Samples from the ThT fluorescence measurements, with 5 equivalents of a compound with respect to Aβ₄₀, were used to obtain the TEM images after the completion of ThT based experiments.

Figure 6.. Effect of small molecules on the secondary structure of Aβ.

CD spectra of 25 μM Aβ₄₀ in the presence of 500 μM of compound loaded LUVs from 0 hours – 7 days.

Figure 7.. Effect of small molecule interaction with Aβ probed by NMR.

(Left) 2D SOFAST-HMQC NMR spectra of 25 μM ¹⁵N- Aβ₄₀ in the presence of 500 μM of compound loaded LUVs. (Right) Signal-to-noise ratios taken from spectra obtained at 0, 24 and 96 hours.

Figure 8.. Aβ species formed in the presence of small molecules.

Size exclusion chromatography of Aβ₄₀ NMR samples in the presence of the indicated compound loaded LUVs. Normalized area under each peak area (indicated as I, II and III) was measured for all the samples and summarized in the Table.

Figure 9.. Effect of small molecules on Aβ induced membrane disruption.

Dye-leakage experiments using compound loaded LUVs filled with 6-carboxyfluorscin. Dye-leakage was measured in a buffer of 20 mM phosphate and 50 mM NaCl at 37 °C under slow shaking over 40 hours. Percentage of dye released was determined after the lysis of LUVs with triton-X.