Amyloid-like aggregates of neuronal tau induced by formaldehyde promote apoptosis of neuronal cells

Abstract

Background: The microtubule associated protein tau is the principle component of neurofibrillar tangles, which are a characteristic marker in the pathology of Alzheimer's disease; similar lesions are also observed after chronic alcohol abuse. Formaldehyde is a common environmental contaminant and also a metabolite of methanol. Although many studies have been done on methanol and formaldehyde intoxication, none of these address the contribution of protein misfolding to the pathological mechanism, in particular the effect of formaldehyde on protein conformation and polymerization.

Results: We found that unlike the typical globular protein BSA, the natively-unfolded structure of human neuronal tau was induced to misfold and aggregate in the presence of ~0.01% formaldehyde, leading to formation of amyloid-like deposits that appeared as densely staining granules by electron microscopy and atomic force microscopy, and bound the amyloid-specific dyes thioflavin T and Congo Red. The amyloid-like aggregates of tau were found to induce apoptosis in the neurotypic cell line SH-SY5Y and in rat hippocampal cells, as observed by Hoechst 33258 staining, assay of caspase-3 activity, and flow cytometry using Annexin V and Propidium Iodide staining. Further experiments showed that Congo Red specifically attenuated the caspase-3 activity induced by amyloid-like deposits of tau.

Conclusion: The results suggest that low concentrations of formaldehyde can induce human tau protein to form neurotoxic aggregates, which could play a role in the induction of tauopathies.

Figure 1

Effect of formaldehyde on tau aggregation at different concentrations. (A) Recombinant htau-40 (20 μM final concentration) was incubated with formaldehyde at desired concentrations in 100 mM phosphate buffer (pH 7.2) at 37°C for 24 h and aliquots (10 μl) were loaded for 10% SDS-PAGE. Lane M contains molecular mass standards. (B) BSA was used as a control. (C) Gray densities of tau polymers on SDS-PAGE were measured (curve 1) and changes in the light scattering of tau with formaldehyde at different concentrations were detected (curve 2). (D) The gray densities of BSA monomers from panel B (curve 1) and the light scattering of BSA (curve 2). (E) Tau-40 (1.2 μM final concentration) was incubated with 0.1 % formaldehyde and changes in the light scattering at 480 nm were measured at different time intervals in the presence (curve 1) or absence (curve 2) of formaldehyde. BSA alone (curve 3) or BSA incubated with formaldehyde (curve 4) was used as controls. (F) The same data as shown in panel E is plotted on a semi-logarithmic scale [36].

Figure 2

Reaction of formaldehyde with amino groups of neuronal tau. (A) Neuronal tau (final concentration 0.1 μM) was resuspended in phosphate buffer containing OPT (20 times molar excess compared to protein) in the presence of formaldehyde at different concentrations at 37°C for 120 min. The fluorescence (Ex340 nm/Em455 nm) was then measured. (B) Under the same conditions, tau was incubated with (curve 1) or without (curve 2) 0.005% formaldehyde and 2 μM OPT. Aliquots were then taken to measure the fluorescence at different time intervals. The data were plotted on a semilogarithmic scale as described by Tsou et al. [59].

Figure 3

Changes in fluorescence spectra of thioflavin T and absorption spectra of Congo Red in the presence of tau deposits. (A) Thioflavin T (10 μM final concentration) was incubated with 0.1% aldehyde-treated tau (2 μM) in 100 mM potassium phosphate buffer pH 7.2 for 15 min before measurement. Emission spectra of ThT were recorded (excitation at 450 nm) in the presence of formaldehyde-treated tau (curve 1), self-aggregated tau (curve 2), acetaldehyde-treated tau (curve 3) and in the absence of protein (curve 4). (B) Kinetics of the increase in the fluorescence emission of ThT incubated with formaldehyde-treated tau (curve 1) or self-aggregated tau (curve 2). (C) Under the same conditions, Congo Red (5 μM final concentration) was incubated with formaldehyde-treated tau (curve 1), acetaldehyde-treated tau (curve 2) or native tau (curve 3) for 15 min before measurement. Congo Red alone is shown as a control (curve 4).

Figure 4

Tau deposits were imaged by transmission electron microscopy and atomic force microscopy. Tau deposits were formed by incubation of tau with formaldehyde in 100 mM sodium phosphate, pH 7.2, for one day. (A) Formaldehyde-treated tau deposits stained with uranyl acetate were observed under the electron microscope (Bar: 100 nm); (B) Tau (40 μM) was incubated with heparin (1 mg/ml) under the same conditions (Bar: 50 nm). (C) Formaldehyde-treated tau deposits observed by AFM (Bar: 125 nm). (D) Tau alone as a control was observed by AFM (Bar: 125 nm).

Figure 5

Conformational changes of polymerized tau in formaldehyde solution. (A) Circular dichroism spectra of 2 μM native tau (curve 2), and tau incubated with 0.1% formaldehyde (curve 1) or acetaldehyde (curve 3) at 37°C for 24 h. Scattering contributions of the aldehyde were subtracted from the spectra. (B) Changes in the fluorescence of tau in the presence of ANS. Tau (1.2 μM final concentration) was incubated in 100 mM phosphate buffer (pH 7.2) with or without 0.1 % aldehyde at 37°C overnight and then ANS (molar ratio: tau/ANS = 1/40) was added. Changes in the ANS fluorescence spectra at 480 nm for tau incubated with formaldehyde (curve 1), acetaldehyde (curve 3) or without aldehyde (self-aggregation, curve 2) were measured by excitation at 350 nm. Formaldehyde alone (curve 4) is shown as a control.

Figure 6

Contrast microscope image of cells treated with amyloid-like tau. The same SH-SY5Y cells were imaged after incubated with tau deposits (2 μM) for 0, 24, 48 or 72 h (A-D). Cells were visualized by inverted contrast microscopy. Bar = 25 μm.

Figure 7

Hoechst staining of cells in the presence of amyloid-like tau. Treatment with formaldehyde-treated tau resulted in apoptotic death. Note that formaldehyde was removed from the protein samples by ultrafiltration. SH-SY5Y cells were treated for 3 days with tau deposits (2 μM) induced by pretreatment with 0.1% formaldehyde. The cells were collected and stained with Hoechst 33258. Nuclei were visualized by fluorescence microscopy. Bar: 25 μm. (A) Control culture in DMEM without serum. (B) Cells treated with self-aggregated tau. (C) Cells treated with 0.1 %-acetaldehyde-treated tau. (D) Cells treated for 72 h with the mock solution without formaldehyde were collected and stained with Hoechst 33258. (E) Cells treated with 0.1 %-formaldehyde-treated tau. The arrows designate the presence of apoptotic nuclear profiles. Data are expressed as a percentage of the control (cells treated with vehicle alone) and presented as the mean ± SEM (n = 6). (F) Statistics of apoptotic cells from A-E.

Figure 8

Cell viability and Caspase-3 activity measurements over time. (A) Cell viability was measured by the MTT assay as described in Materials and methods. (B) After SY5Y cells were treated with different tau samples (2 μM) for 24 h, cell lysates were collected at the times indicated and used to measure caspase-3 activity. (C) SH-SY5Y cells were treated with tau deposits (2 μM) in the presence or absence of 10 μM Congo red. Cell lysates were collected at the times indicated and caspase-3 activity was measured. Data are expressed as a percent of the control (cells treated with vehicle alone) and presented as the mean ± SEM (n = 6).

Figure 9

Amyloid-tau treatment induces rat hippocampal cell apoptosis. Flow cytometric analysis of primary hippocampal cells after treatment with 2 μM formaldehyde-treated tau. The percentage of apoptotic cells were characterized as those that stained with Annexin-V and excluded PI (see Materials and Methods). (A) Control cultured in DMEM without serum. (B), (C) and (D) represent the results of cells exposed to 0.1%-formaldehyde treated amyloid-tau for 24, 48, and 72 h, respectively. (E) and (F) represent cells incubated for 72 h with 0.05%- or 0.01%- formaldehyde-treated tau, respectively. Data represent the mean values of three independent experiments.