New Paradigm for Nano-Bio Interactions: Multimolecular Assembly of a Prototypical Disordered Protein with Ultrasmall Nanoparticles

Abstract

Understanding the interactions between nanoparticles (NPs) and proteins is crucial for the successful application of NPs in biological contexts. Protein adsorption is dependent on particle size, and protein binding to ultrasmall (1-3 nm) NPs is considered to be generally weak. However, most studies have involved structured biomacromolecules, while the interactions of ultrasmall NPs with intrinsically disordered proteins (IDPs) have remained elusive. IDPs are abundant in eukaryotes and found to associate with NPs intracellularly. As a model system, we focused on ultrasmall gold nanoparticles (usGNPs) and tau, a cytosolic IDP associated with Alzheimer's disease. Using site-resolved NMR, steady-state fluorescence, calorimetry, and circular dichroism, we reveal that tau and usGNPs form stable multimolecular assemblies, representing a new type of nano-bio interaction. Specifically, the observed interaction hot spots explain the influence of usGNPs on tau conformational transitions, with implications for the intracellular targeting of aberrant IDP aggregation.

Keywords: NMR spectroscopy; intrinsically disordered proteins; protein aggregation; protein−nanoparticle interaction; ultrasmall nanoparticles.

Figure 1

Gel electrophoresis and photophysical measurements. (A) Agarose native gel electrophoresis: left, unstained; right, Coomassie staining. Lanes 1–7 were loaded with 6.5 μM usGNPs; lanes 2–8 contained tau^4RD at concentrations 3.25, 6.5, 13, 19.5, 43.5, and 65 μM; all components were dissolved in 0.5% TBE, pH 9 (1% agarose, 50 V, 40 min). (B) Tyrosine fluorescence emission spectra (λ_ex = 270 nm) measured on 6 μM tau^4RD in the presence of usGNPs at varying concentration. The spectrum of the free protein is shown in black (λ_max = 304 nm). (C) Relative Tyr fluorescence intensity as a function of usGNP concentration (Stern–Volmer plot). The solid line corresponds to an exponential function fit. (D) Colloidal solutions of 0.5 μM usGNPs in the presence of varying concentration of protein, visualized under UV lamp. (E) Fluorescence emission spectra (λ_ex = 530 nm) of 0.25 μM usGNP in the presence of tau^4RD at varying concentration. The dashed line indicates the shift of the peak maximum. (F) Relative usGNP fluorescence intensity as a function of protein concentration (logarithmic scale). The solid line corresponds to the best-fit curve (Hill function).

Figure 2

Energetics of the interaction between tau^4RD and usGNPs. Isothermal titration calorimetry data obtained on titrating tau^4RD into usGNPs, in the presence of 0 mM (A) and 100 mM (B) NaCl (see Figure S3 for additional NaCl concentration points). Left panels display corrected heat transfer rates; right panels display integrated heat plots. Orange lines are best-fit curves based on a two-sets-of-sites binding model; data displayed as empty circles were excluded from fitting.

Figure 3

Mapping of contact sites and conformational dynamics. (A) Selected portions of HN-HSQC spectra (600 MHz) of 50 μM [¹⁵N]tau^4RD (black) in the absence (top left) or presence of usGNPs (colored maps overlaid on black) at the reported molar ratios. Red and green maps are shown in scale with black spectrum, the cyan map is displayed with increased intensity for better visualization of the peak position changes; the rightmost panel displays an enlarged view for better appreciation of peak details; and protein (P):usGNP molar ratios are indicated on top. (B) Residue-specific HSQC-peak intensity versus residue number. Peak intensities were measured on tau^4RD in the absence (I₀) or presence (I) of usGNPs at a molar ratio P:usGNPs = 0.02; only isolated peaks were included in the analysis; the protein domain organization is schematized with bars of different gray shading for the four repeat motifs (R1–R4); and the hexapeptide motifs are indicated in orange. (C) Representative ¹⁵N-CPMG relaxation dispersion curves at 700 MHz spectrometer frequency observed for 200 μM [¹⁵N]tau^4RD in the absence (empty circles) or presence (filled circles) of 2 μM usGNPs (from left to right: Lys281, Leu282, Gln288) and 10 mM NaCl. Uncertainties were estimated from duplicate measurements; solid lines are the best-fit curves obtained by fitting the relaxation data to a two-state exchange model, the fitted exchange rate constants are reported; and dashed lines indicate the average R₂^obs for no exchange. (D) Exchange contributions to relaxation rates obtained from relaxation dispersion experiments on samples containing tau^4RD and usGNPs. Gray bars are errors propagated from the uncertainties of fitted parameters.

Figure 4

Protein conformational transitions and aggregation. (A) Aggregation kinetics monitored by ThT fluorescence. Measurements were performed on tau^4RD in the absence (gray dots) or presence (light-to-dark red dots) of usGNPs (P:usGNP molar ratios 1:0, 1:0.002, 1:0.004, 1:0.03, 1:0.1); solid lines correspond to the best-fit curves determined using an empirical sigmoid function; data represent the mean of five replicate measurements; and vertical gray lines indicate transition midpoints. (B) Far-UV CD spectra acquired on 6 μM tau^4RD after 0 h (black) and 24 h (brown) incubation, in the absence or presence of usGNPs at the indicated molar ratios. (C) Representative TEM images of tau^4RD samples after 48 h incubation in aggregating conditions, in the absence (left panel) or presence (remaining panels) of usGNPs (see Figure S9 for additional images). Scale bar is 100 nm.

Figure 5

(A) Schematic representation of the multimolecular assembly formed by ultrasmall gold NPs and a prototypical intrinsically disordered protein, characterized by two distinct exchange regimes. (B) Schematic representation of the influence of usGNPs on disease-related amyloid deposition of tau^4RD.