Involvement of puromycin-sensitive aminopeptidase in proteolysis of tau protein in cultured cells, and attenuated proteolysis of frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) mutant tau
Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
Associate Professor Toshihisa Tanaka MD, PhD, Department of Psychiatry, Osaka University Graduate School of Medicine, D3, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan. Email: firstname.lastname@example.org
In tauopathies, tau protein is hyperphosphorylated, ubiquitinated, and accumulated in the brain; however, the mechanisms underlying this accumulation remain unclear. To gain an understanding of the role of proteases in the metabolism of tau protein, in the present study we evaluated the effects of protease inhibitors in SH-SY5Y human neuroblastoma cells and COS-7 cells transfected with the tau gene. When cells were treated with 0.1–10 µmol/L of lactacystin and 1.0–20 µmol/L of MG-132 (inhibitors of proteasome), 0.1–10 µmol/L of CA-074Me (a cathepsin inhibitor), and 0.1–2 µmol/L of puromycin (a puromycin-sensitive aminopeptidase (PSA) inhibitor) for up to 24 h, there were no significant changes in tau protein levels. However, pulse-chase experiments demonstrated that the proteolysis of tau protein in SH-SY5Y cells was attenuated following treatment of cells with 200 nmol/L puromycin. Increased tau protein levels were also observed in SH-SY5Y cells treated with short interference (si) RNA to PSA to inhibit the expression of PSA. These data suggest that PSA is a protease that catalyses tau protein predominantly in SH-SY5Y cells. The protein metabolism of tau-containing mutations of frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) was also investigated using pulse-chase experiments. The results indicate attenuated proteolysis of tau in cells transfected with mutant tau genes after 48 h. Further immunocytochemical analysis and subcellular fractionation experiments revealed that the mutations did not alter the intracellular distribution of tau and suggested that impaired accessibility of tau to PSA is unlikely to account for the attenuated proteolysis of tau protein. Western blotting with phosphorylation-dependent antibodies revealed that phosphorylation levels of tau at Thr231, Ser396, and Ser409 were increased in cells transfected with V337M, R406W, and R406W mutant tau genes, respectively. Together, the data suggest that attenuated proteolysis of FTDP-17 mutant tau may be explained by increased phosphorylation levels, resulting in resistance to proteolysis.
The microtubule-associated protein tau is a major component of neurofibrillary tangles (NFTs) in the brains of patients with tauopathies, including Alzheimer disease (AD),1,2 frontotemporal dementia,3 corticobasal degeneration,4,5 and progressive supranuclear palsy. Furthermore, the hereditary disease frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) is caused by mutations of a gene coding tau protein.6 Therefore one part of neurodegenerative dementia is caused by tau. Tau protein is predominantly expressed in neurons, where it is believed to play major regulatory roles in microtubule assembly for the organization and integrity of the cytoskeletal network. Tau protein can promote microtubule assembly, an activity that is regulated by phosphorylation, which impairs microtubule assembly.7
In tauopathies, tau protein is hyperphosphorylated, ubiquitinated, and accumulated in the form of NFTs; however, the mechaisms underlying the accumulation of tau protein remain unclear. Previous studies have examined the expression of tau mRNA and found increased tau mRNA levels in the hippocampus of AD brain.8 In addition, the proteolysis of tau protein has been investigated and tau has been shown to be processed in vitro by various proteases, including calpains,9,10 cathepsins,11 caspases,12 thrombin,13 and proteasome, although the protein degradation systems responsible for tau metabolism are not well characterized.
The ubiquitin proteasome system is one of the major systems for protein quality control in eukaryotes14,15 and neurodegenerative diseases are characterized by aggregates and inclusions of aberrant proteins. Therefore, proteasome may be one of the most important factors in the degradation of aberrant proteins. Thus, the mechanisms underlying the degradation of tau protein have been investigated using several protease inhibitors, including proteasome inhibitors. SH-SY5Y cells have been treated with MG-132 and lactacystin and tau protein levels analyzed by western blotting. After treatment of cells with MG-132 or lactacystin, tau protein levels did not increase.16,17 These studies suggest that, in cultured cells, tau may be degraded by mechanisms other than proteasome.
The cathepsin family is one of the most prominent proteases for the degradation of aberrant proteins and works in organelles called lysosomes. Lysosomes are known to fuse with autophagic vesicles, in which aged organelles and aberrant proteins have been accumulated.18 A dysfunction of cathepsins and autophagy has been reported to cause neurodegeneration; however, its relevance to tauopathies is unclear.19–21
A recent genetic screening study using Drosophila identified puromycin-sensitive aminopeptidase (PSA) as a potent modifier of tau-induced pathology and suggested PSA as a possible tau-degrading enzyme.22 In addition, PSA was shown to digest recombinant human full-length tau in vitro, with this activity hindered by puromycin.23 These results suggest that PSA may be involved in the proteolysis of tau protein in living cells and thus the regulation of tau protein levels.
In FTDP-17, which is caused by mutations of the tau gene, tau is hyperphosphorylated, ubiquitinated, and accumulated in diseased brains. The microtubule assembly promoting activity of mutated tau is impaired and there is an increased tendency for aggregation. The effects of the FTDP-17 mutation of tau have been shown in cells transfected with tau genes and the degradation of some mutated tau proteins is impaired.24 However, the relevance of these mutations to proteolysis remains unclear. Therefore, in the present study, we investigated the involvement of several proteases in the proteolysis of tau protein and the effects of mutation of tau in cultured cells.
Preparation of gene constructs
The tau gene (the longest isoform: tau441), a generous gift from Dr M. Goedert (Medical Research Council, Cambridge, UK), was cut out with NdeI and EcoRI restriction enzymes and inserted with a linker into the pcDNA3.1 vector (Invitrogen, Carlsbad, CA, USA). The FTDP-17 mutated tau constructs (valine 337 to methionine (V337M) and arginine 406 to tryptophan (R406W)) were constructed using site-directed mutagenesis with a Quick Change kit (Stratagene, La Jolla, CA, USA) according to the manufacturer's instructions.
Cell lines, culture conditions, and transfections
The SH-SY5Y human neuroblastoma cells were kindly donated by Dr J. L. Biedler (Sloan Kettering Institute, New York, NY, USA) and were cultured in Dulbecco's modified Eagle's medium (DMEM)/F12 (Gibco BRL, Rockville, MD, USA) containing 5% fetal bovine serum (FBS; JRH Biosciences, Leneza, KS, USA). The COS-7 cells were cultured in DMEM (Gibco BRL) and 5% FBS. One day prior to transfection, the COS-7 cells were plated on dishes (10 cm diameter) and were grown overnight to obtain 90–95% confluence at the time of transfection. Cultured cells were transfected with the pcDNA3.1 vector containing wild-type tau gene (tau441) and mutant genes (V337M, and R406W) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Briefly, cells were incubated with a complex of each DNA construct and Lipofectamine 2000 in the absence of antibiotics for 5 h; after that time, the medium was exchanged for normal growth medium.
Treatment of cell cultures with protease inhibitors
Cell cultures were treated for up to 20 h with: (i) the proteasome inhibitors lactacystin (0–10 µmol/L; Calbiochem, San Diego, CA, USA) and MG-132 (0–20 µmol/L; Sigma-Aldrich Japan, Tokyo, Japan); (ii) the cathepsin inhibitor CA-074Me (0–10 µmol/L; Calbiochem); and (iii) the PSA inhibitor puromycin (0–2 µmol/L; Sigma-Aldrich Japan). Cell viability was evaluated using the lactate dehydrogenase (LDH) cytotoxicity assay kit (Medical and Biological Laboratory, Nagoya, Japan) to identify any non-specific toxic effects of each of the drugs tested.
Treatment of cell cultures with short interference RNA
SH-SY5Y cells were cultured in DMEM/F12 (Gibco BRL). One day prior to transfection, cells were plated into six-well plates with 2 mL growth medium in the absence of antibiotics and were cultured to 30–50% confluence at the time of transfection. For short interference (si) RNA treatment of cells, 200 pmol siRNA oligomer was incubated with 250 µL Opti-MEM I Reduced Serum Medium (Gibco BRL) at room temperature for 5 min before the oligomer was incubated for a further 20 min at room temperature with Lipofectamine 2000. The oligomer–Lipofectamine 2000 complexes were added to each of the wells containing cells and medium and cells were incubated in the presence of the oligomer for 36 h at 37°C in a CO2 incubator until harvesting.
Cells were lysed with RIPA buffer containing 50 mmol/L Tris-HCl, pH 8, 150 mmol/L NaCl, 20 mmol/L EDTA, 1% v/v Nonidet-P40, 50 mmol/L natrium fluoride, 20 mmol/L N-ethyl maleimide, and 100 mmol/L sodium orthovanadate (Sigma-Aldrich), supplemented with protease inhibitor cocktail (Calbiochem) and phenylmethylsulfonyl fluoride (PMSF; Sigma-Aldrich Japan), on ice for 1 h. The lysates were centrifuged for 30 min at 12 000 g to remove cellular debris, and protein concentrations in the supernatant were determined using the BCA Protein Assay Kit (Pierce, Cheshire, UK). Equal amounts of protein were applied to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. Blots were blocked with 5% (w/v) ECL blocking agent (GE Healthcare, Uppsala, Sweden) in TBS-T (50 mmol/L Tris HCl, pH 7.4, 150 mmol/L NaCl, 0.5% Tween-20). Membranes were probed overnight at 4°C with: (i) anti-tau phosphorylation-independent monoclonal antibody Tau-5 (Calbiochem; 1:1000 dilution); (ii) polyclonal antibody Tau(H-150) (Santa Cruz Biotechnology, Santa Cruz, CA, USA; 1:1000 dilution); (iii) anti-tau phosphorylation-dependent antibodies PT231, PS396, and PS409 (Biosource, Nivelles, Belgium), which recognize phosphorylated threonine 231, serine 396, and serine 409 on tau, respectively; and (iv) anti-PSA goat polyclonal antibody PSAP(N-20) (Santa Cruz). Then, peroxidase-labeled anti-mouse, anti-rabbit, and anti-goat IgG antibodies were applied as secondary antibodies. The reaction products were visualized using an ECL kit (Amersham Biosciences, Buckinghamshire, UK).
[35S]-Methionine pulse-chase experiments
Cells were rinsed in phosphate-buffered saline (PBS) and incubated for 1 h in 4 mL Met-free medium (Gibco) for starvation. Afterward, cells were rinsed with PBS and incubated for 3 h in 5 mL Met-free DMEM containing 1.5 MBq of [35S]-methionine-trans-labeled methionine (EXPRE35S 35S Protein Labeling Mix; Perkin Elmer Japan, Kanagawa, Japan). Cells were then rinsed twice with 2 mL PBS and incubated for chase intervals of 0 and 24 h in 5 mL non-radioactive growth medium supplemented with 5% FBS containing each chemical reagent. Cells lysates were harvested with RIPA buffer and precipitated with Tau-5 monoclonal antibody overnight at 4°C. The antibody–antigen complex was extracted from the lysate by incubating with Protein G–Sepharose (GE Healthcare) for 1 h at 4°C, followed by centrifugation for 1 min at 12 000 g. Protein G–Sepharose pellets were suspended in lysis buffer and washed six times with lysis buffer and once with washing buffer (50 mmol/L Tris, pH 8.0). The samples were then electrophoresed in 10% Tris-Glycine gel (Invitrogen) and vacuum dried. Radiolabeled tau bands were visualized and quantified densitometrically.
Subcellular fractionation was performed using the ProteoExtract Subcellular Proteome Extraction Kit (Calbiochem) according to the manufacturer's instructions. Briefly, cells were carefully washed twice with cold washing buffer before buffer I (supplied in the kit) with Protease Inhibitor Cocktail was added to the cells and the samples were rotated in 10 r.p.m. for 10 min at 4°C. Samples were centrifuged at 500 g for 10 min and the supernatant was collected as Fraction I, whereas the pellets were subjected to further fractionation. Similarly, the Fractions II and III were obtained by using buffers II and III (supplied in the kit), respectively. Fractions I, II, and III correspond to the cytosol, membrane/organelle, and nuclear fractions, respectively.
Tau-transfected COS-7 cells were plated onto Lab-Tek plates (Nunc, Roskilde, Denmark). Cells were rinsed with PBS and were fixed with ice-cold methanol for 5 min. After fixation, cells were rinsed with 2 mL PBS three times and were then treated with 5% bovine serum albumin (BSA)/TBS-T for 10 min before being probed with the Tau-5 (1:100 dilution) and PSAP(N-20) (1:100 dilution) antibodies overnight at 4°C. Cells were washed in 2% BSA/TBS-T three times before the secondary antibodies fluorescein isothiocyanate (FITC) anti-mouse IgG (Cappel, Philadelphia, PA, USA) and Alexa FluorR 633 Rabbit Anti-goat IgG (Molecular Probes, Eugene, OR, USA; 1:1000 dilution) were applied for 30 min at 37°C. After plates had been washed with PBS at least five times, cells were observed under confocal microscopy.
To evaluate the involvement of proteases in the metabolism of tau protein, different protease inhibitors were applied to SH-SY5Y human neuroblastoma cells and COS-7 cells transfected with the tau gene. First, cell viability was determined and essentially no cytotoxic effects were observed when SH-SY5Y human neuroblastoma cells were treated with the proteasome inhibitors lactacystin (0–10 µmol/L) and MG-132 (0–20 µmol/L), the cathepsin inhibitor CA-074Me (0–1 µmol/L), and the PSA inhibitor puromycin (0–2 µmol/L) for up to 20 h (data not shown). Although these conditions have been reported to be effective in reducing each of the protease activities,25–28 no significant changes were seen in tau protein levels (data not shown). The effects of these same protease inhibitors were also evaluated in COS-7 cells transfected with the tau gene (tau441). Essentially, no cytotoxic effects were observed when COS-7 cells were treated with lactacystin (0–10 µmol/L), CA-074Me (0–10 µmol/L), or puromycin (0–2 µmol/L) for up to 20 h (data not shown). Western blot analysis showed no significant changes in tau protein levels (n= 3; Fig. 1a–c). These data suggest that tau protein is not catalysed predominantly by one of those proteases and that it is likely to be catalysed by other proteases.
To investigate the involvement of those proteases in the catalysis of tau further, [35S]-methionine pulse-chase experiments were performed to detect subtle changes in tau protein levels. These pulse-chase experiments showed that tau protein in SH-SY5Y cells was degraded gradually and only 30–40% of tau protein remained after 24 h. When SH-SY5Y cells were treated with lactacystin (0–8 µmol/L), MG-132 (0–15 µmol/L), and CA-074Me (0–1 µmol/L), there were no significant changes in tau protein levels (n= 4) (Fig. 2a–c); however, significant increases were observed in tau protein levels when SH-SY5Y cells were treated with puromycin (200 nmol/L; n= 4; Fig. 2d). Increased tau protein level were not observed in cells treated with much higher concentrations of puromycin in this pulse-chase experiment; we assumed that this was due to an inhibitory effect of puromicin against ribosome-attenuated protein synthesis, which interfered with the effects of puromycin on tau protein levels.
On the basis of these results, experiments were performed investigating the inhibition of PSA expression using siRNA against PSA. In these experiments, increased tau protein levels were observed in SH-SY5Y cells treated with siRNA to PSA, but not in cells treated with siRNA against glyceroaldehyde 3-phosphate dehydrogenase (GAPDH) as an experimental control (Fig. 3). These data suggest that PSA is a protease that catalyses tau protein predominantly in SH-SY5Y cells.
Based on the results obtained in the siRNA experiments, we further investigated the protein metabolism of tau with FTDP-17 mutations using pulse-chase experiments. In these experiments, COS-7 cells were transfected with wild-type and mutant type (V337M and R406W) tau genes and tau protein levels were chased up to 48 h. No changes were observed in cells transfected with these tau genes after 24 h; however, attenuated proteolysis of tau was observed in cells transfected with the mutant tau genes after 48 h (n= 4; Fig. 4).
To clarify the effects of FTDP-17 mutations on the topological relevance of tau and PSA, immunocytochemistry and subcellular fractionation were performed to investigate the accessibility of tau to the PSA. In COS-7 cells transfected with tau genes, immunocytochemical studies showed that tau was located in the cytoplasm and nucleus, presumably because tau is overexpressed by the transfection (Fig. 5a), whereas PSA was located only in the cytoplasm (Fig. 5b). Tau protein and PSA were colocalized in the cytoplasm (Fig. 5c). No changes were observed in the localization of tau in cells transfected with the wild-type and mutant tau genes. Furthermore, subcellular fractionation revealed that tau was found predominantly in Fraction I (cytoplasm) and that faint tau was observed in Fraction III (nucleus; Fig. 5d). In contrast, PSA was observed only in Fraction I (cytoplasm; Fig. 5e). There were no changes in the intracellular distribution of tau and PSA among cells transfected with wild-type and mutant tau genes. These results suggest that attenuated proteolysis of FTDP-17 mutant tau is not explained by the accessibility of tau to PSA.
Finally, phosphorylation levels of tau protein were investigated in cells transfected with tau genes. First, tau expression levels were normalized using western blots stained with Tau-5, a phosphorylation-independent antibody, to ensure equal levels of protein loading. Three antibodies were used and, compared with results obtained for wild-type tau, significant increases were observed in the phosphorylation of tau at Thr231, Ser396, and Ser409 in cells transfected with V337M and R406W mutant tau genes (n= 4; Fig. 6a–c). These results suggest that attenuated proteolysis of FTDP-17 mutant tau may be explained by increased phosphorylation levels resulting in resistance to proteolysis.
In tauopathies, tau protein is hyperphosphorylated, ubiquitinated, and accumulated in the form of NFTs; however, the mechaisms underlying this accumulation remain unclear. In the present study, the involvement of several proteases in tau protein metabolism was investigated, and only PSA was found to be a protease that predominantly regulates tau protein levels in cultured cells. This is the first study to report that the effect of inhibition of PSA by puromycin in an increase of tau protein in cultured cells.
Previously, the involvement of proteasomes in tau degradation has been investigated. Feuillette et al. reported that, in SH-SY5Y cells treated with different proteasome inhibitors, including MG132 and lactacystin, instead of an increase in tau protein, an unexpected decrease in tau protein levels was observed.16 These authors also used mutant alleles of the 20S proteasome in Drosophila and found that genetic inactivation of the proteasome resulted in a decrease of tau levels in Drosophila. Delobel et al. have reported that inhibition of the proteasome leads to a decrease in tau levels via an increase in calpain levels, which result in accelerated degradation of tau protein.17 The results of the present study indicate no change in tau levels in cells treated with 0–15 µmol/L MG132; this discrepancy may be explained by the culture conditions or the concentration of the inhibitors used. Both studies used MG132 at low concentrations of 0–0.5 µmol/L and decreases in tau following inhibition of the proteasome may be observed under limited conditions. However, the interaction between several proteases should be taken into consideration. Cathepsin is a plauseible candidates for the degradation of tau protein; however, in the present study, no changes were observed in tau levels in cells treated with a cathepsin inhibitor. This enzyme is known to react with substrates in the lysosome and, in normal cell cultures, tau may not be degraded via incorporation into lysosome or autophagic vesicles.
Karsten et al. first reported PSA as a tau-degrading enzyme following genomic screening using Drosophila.22 In that study, the authors used a cross-species functional genomic approach to analyze gene expression in multiple brain regions in the mouse in parallel with validation in Drosophila to identify tau modifiers. They reported that PSA protected against tau-induced neurodegeneration in vivo. Furthermore, Senguputa et al. reported that tau is degraded by PSA in vitro;23 therefore, it is likely that PSA is involved in the metabolism of tau protein in living cells. However, the effects of PSA inhibition in cultured cells had not been examined. In the present study, we showed that tau levels were increased in cells treated with 200 nmol/L puromycin, but that higher concentrations of puromycin were not effective because of inhibition of protein synthesis. The effect of PSA inhibition on tau protein levels was also confirmed by siRNA leading to decreased expression of PSA. Therefore, our results suggest that PSA functions as the predominant regulator of tau protein levels in normally cultured cells.
Next, we tried to evaluate the delayed proteolysis of FTDP-17 mutated tau in COS-7 cells and studied the topological relevance of the mutated tau and PSA. We found attenuated tau proteolysis in cells transfected with the mutant tau (V337M and R406W) genes after 48 h compared with wild-type tau. However, the mutation did not result in differences in the intracellular distribution of tau and impaired accessibilities of tau to PSA is unlikely. Therefore, phosphorylation levels of tau protein were examined and both mutations resulted in increased phosphorylation of tau protein at at least one Ser/Thr site. Previously, it was repoted that phosphorylation induced resistance of tau to proteolysis29 and that FTDP-17 mutations induced hyperphosphorylation of tau in certain cells.30 These changes may explain the delayed degradation of tau protein in mutant tau-transfected cells. Hyperphosphorylation of tau could induce breakdown of microtubules,31 and this may be one of the causes of neuronal cell death in these neurodegenerative diseases. Further studies on the effects of phosphorylation and the FTDP-17 mutation on direct proteolysis of tau by PSA are required.
The protein metabolism of tau was investigated using several protease inhibitors and, in the present study, PSA was found to be the predominant regulator of tau protein levels in normally cultured cells. The FTDP-17 mutation delayed the proteolysis of tau and increased the phosphorylation of tau. Both tau and PSA are colocalized in the cytoplasm and PSA may be involved in the mechanism underlyng the delayed proteolysis of tau seen with the mutations.
This study was supported, in part, by frants from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (18591286 and 19390305).