Unsaturated Fatty Acid-Induced Conformational Transitions and Aggregation of the Repeat Domain of Tau

Abstract

Background: The intrinsically disordered, amyloidogenic protein Tau associates with diverse classes of molecules, including proteins, nucleic acids, and lipids. Mounting evidence suggests that fatty acid molecules could play a role in the dysfunction of this protein, however, their interaction with Tau remains poorly characterized.

Methods: In a bid to elucidate the association of Tau with unsaturated fatty acids at the sub-molecular level, we carried out a variety of solution NMR experiments in combination with circular dichroism and fluorescence measurements. Our study shows that Tau^4RD, the highly basic four-repeat domain of Tau, associates strongly with arachidonic and oleic acid assemblies in a high lipid/protein ratio, perturbing their supramolecular states and itself undergoing time-dependent structural adaptation. The structural signatures of Tau^4RD/fatty acid aggregates appear similar for arachidonic acid and oleic acid, however, they are distinct from those of another prototypical intrinsically disordered protein, α-synuclein, when bound to these lipids, revealing protein-specific conformational adaptations. Both fatty acid molecules are found to invariably promote the self-aggregation of Tau^4RD and of α-synuclein.

Conclusions: This study describes the reciprocal influence that Tau^4RD and fatty acids exert on their conformational states, contributing to our understanding of fundamental aspects of Tau/lipid co-assembly.

Keywords: NMR spectroscopy; Tau; fatty acids; neurodegeneration; protein aggregation; protein–lipid interactions.

Scheme 1

Illustration of Tau domain organization and the molecular structures of fatty acids.

Figure 1

¹H-NMR profiles of Tau^4RD and fatty acids. High-field portion of ¹H-NMR spectra of 25 μM Tau^4RD (black), 75 μM fatty acid (FA, blue), and 25 μM Tau^4RD mixed with 75 μM FA (red). (A) Data obtained with FA = oleic acid (OLA). (B) Data obtained with FA = arachidonic acid (ARA). Measurements were performed at 37 °C.

Figure 2

¹H-NMR measurements of Tau^4RD monomer depletion and translational diffusion. High-field region of ¹H-NMR spectra (A) and translational diffusion experiments (B) recorded on Tau^4RD in the presence of OLA. High-field region of ¹H-NMR spectra (C) and translational diffusion experiments (D) recorded on Tau^4RD in the presence of ARA. Samples contained 25 μM Tau^4RD and 75 μM fatty acid. Spectra in (A,C) were collected on samples that were incubated for 0 h (gray), 24 h (dark gray), and 48 h (black). In (B,D), the normalized intensity of a selected signal is shown as a function of the square of the fraction of gradient used; errors in peak intensities were smaller than the size of the symbols, as estimated from duplicate experiments; samples were incubated for 0 h (empty circles/diamonds), 24 h (gray filled circles/diamonds), and 48 h (black filled circles/diamonds); asterisks denote data collected on Tau^4RD in the absence of fatty acid; lines are fitted exponential curves. Protein samples were aliquots taken from 100 μM protein solutions incubated with 300 μM fatty acid under magnetic stirring. Samples were kept at 37 °C during both incubation and measurement.

Figure 3

Perturbations of the chemical environment of fatty acids in the presence of Tau. (A) 1D ¹³C spectrum measured on 80 μM [¹³C₁]OLA in the absence (top) and presence (bottom) of 20 μM Tau. The intensity of the bottom spectrum is increased four-fold compared to the top spectrum for better visualization. Measurements were performed at 37 °C. (B) DAUDA fluorescence spectra recorded on samples containing 300 μM OLA in the absence (gray solid line) or presence of 100 μM Tau^4RD (black solid line). (C) DAUDA fluorescence spectra recorded on samples containing 300 μM ARA in the absence (gray solid line) or presence of 100 μM Tau^4RD (black solid line). Spectra of control samples containing buffer (dotted line) or Tau4RD (dashed line) without FA are displayed in both (B,C). DAUDA was 1 μM in all samples.

Figure 4

Fatty acid-induced NMR signal perturbations monitored at single residue/atomic resolution. (A, B) Overlaid, selected regions of HN-HSQC spectra recorded on [¹⁵N]Tau^4RD in the absence (black) or presence of OLA (A, red) or ARA (B, red); measurements were performed at 25 °C; protein concentration was 25 μM and fatty acid concentration was 75 μM. (C) Overlaid, selected regions of HC-HSQC spectra recorded on [¹⁵N,¹³C]Tau^4RD in the absence (black) or presence of ARA (red); the measurement temperature was 37 °C; protein concentration was 100 μM and fatty acid concentration was 50 μM. (D) Overlaid, selected regions of CON spectra recorded at 37 °C on samples as in panel C.

Figure 5

¹⁵N transverse relaxation rates. ¹⁵N-spin transverse relaxation rate values measured on 100 μM [¹⁵N]Tau^4RD in the absence (gray dots) and presence of 60 μM ARA (black triangles). Measurements were performed at 25 °C and a magnetic field of 14.1 T. Error bars are fitting errors. Dotted vertical lines indicate the position of separation between distinct repeats. Gray horizontal bars indicate the position of the hexapeptide motifs. The amino acid residue numbering is based on full-length human Tau.

Figure 6

Far-UV circular dichroism spectroscopy. (A) Tau^4RD in the presence of OLA immediately after mixing (solid line) and after 24 h of incubation (dashed) at 37 °C under magnetic stirring, at a molar ratio Tau^4RD:OLA = 1:3. (B) Tau^4RD in the presence of ARA immediately after mixing (solid line) and after 24 h (dashed) or 72 h (dotted) of incubation at 37 °C under magnetic stirring, Tau^4RD:ARA = 1:3. (C) Tau^4RD in the presence of OLA immediately after mixing, Tau^4RD:OLA = 1:3 (light gray), 1:10 (gray), 1:25 (black). (D) Same as B, with Tau^4RD:ARA = 1:25. (E) Tau^4RD (solid) and Tau^4RD mixed with heparin (HEP), after 24 h incubation (dashed), Tau^4RD:HEP = 4:1. (F) Tau^4RD (solid) and Tau^4RD immediately after mixing with SDS, Tau^4RD:SDS = 1:25 (black), 1:100 (gray). [θ]_MRW is the mean residue molar ellipticity.

Figure 7

Far-UV circular dichroism spectroscopy. (A) Tau^4RD freshly prepared (solid line) and after 24 h (dashed) or 72 h of incubation (dotted) at 37 °C under magnetic stirring. (B) αS freshly prepared (solid line) and after 24 h (dashed) or 72 h of incubation (dotted) at 37 °C under magnetic stirring. (C) αS in the presence of OLA immediately after mixing (solid line) and after 24 h (dashed) or 72 h (dotted) of incubation at 37 °C under magnetic stirring, αS:OLA = 1:3. (D) αS in the presence of ARA immediately after mixing (solid line) and after 24 h (dashed) or 72 h (dotted) of incubation at 37 °C under magnetic stirring, αS:ARA = 1:3. (E) Same as C, αS:OLA = 1:25. (F) Same as D, αS:ARA = 1:25. [θ]_MRW is the mean residue molar ellipticity.

Figure 8

Thioflavin T (ThT) fluorescence assays. ThT fluorescence intensity was monitored over time on samples containing: (A) 10 μM Tau^4RD and 30 μM fatty acid (FA); (B) 10 μM Tau^4RD and 300 μM FA; (C) 100 μM Tau^4RD and 300 μM FA. Protein and FAs were dissolved in phosphate buffer, pH 6.8, also containing excess dithiothreitol (DTT). Samples were incubated at 30 °C under intermittent agitation. Control measurements were performed on protein (C) and FAs (B). Aggregation kinetic curves for Tau^4RD in the presence of heparin (HEP) are displayed for comparison (grey circles). The displayed data are the mean ± SD of measurements performed in quadruplicate (no error bars are shown for control samples).

Figure 9

TEM analysis of protein/FA aggregates. TEM micrographs were taken on Tau^4RD/FA samples incubated for 24 h at 30 °C under intermittent shaking. Samples contained: (A,B) 10 μM protein and 30 μM FA; (C,D) 10 μM protein and 300 μM FA; (E,F) 100 μM protein and 300 μM FA. Scale bars are 100 nm.