Cilostazol, a phosphodiesterase 3 inhibitor, activates proteasome-mediated proteolysis and attenuates tauopathy and cognitive decline

Abstract

Alzheimer’s disease and several variants of frontotemporal degeneration including progressive supranuclear palsy and corticobasal degeneration are characterized by the accumulation of abnormal tau protein into aggregates. Most proteins, including tau, are degraded via the ubiquitin proteasome system, but when abnormal tau accumulates, the function of 26S proteasomes is downregulated. The negative effect of tau aggregates on the function of the proteasome can have deleterious consequences on protein homeostasis and disease progression. Developing therapies aimed at clearing abnormal tau are thus of considerable interest. In the present study, we investigated the effect of cilostazol, an FDA-approved selective phosphodiesterase 3 inhibitor, on a mouse model of tauopathy (line rTg4510). Administration of cilostazol for 30 days enhanced proteasome function via the cyclic adenosine 3',5'-monophosphate/protein kinase A pathway and attenuated tauopathy and cognitive decline in rTg4510 mice. These results suggest that cilostazol, or other FDA-approved drugs acting via the same pathway, has the potential to be repurposed for the treatment of patients with early-stage tauopathy.

INTRODUCTION

The accumulation of neurofibrillary tangles in Alzheimer’s disease (AD) and other tauopathies suggests that tau degradation through the ubiquitin proteasome system (UPS) or autophagy may be critically impaired, contributing to disease pathogenesis. The UPS is the principal mechanism for protein turnover in eukaryotic cells.^1,2 Degradation of proteins via the UPS involves 2 successive steps, starting with initial tagging of the substrate protein by the covalent attachment of multiple ubiquitin proteins followed by the degradation of tagged protein by the 26S proteasome.³ Neurons in particular rely on functional clearance mechanisms to avoid protein aggregation as they cannot dilute toxic substances by means of cell division.⁴ As we age, proteasome-mediated clearance becomes less efficient^5-7 and in neurodegenerative diseases like AD, the function of proteasomes becomes profoundly impaired.^8-16 Recently, components of the UPS have been implicated in AD pathogenesis by proteomics pathway analysis¹⁷ and in other tauopathies by tau interactome mapping.¹⁸ The accumulation of ubiquitinated tau in neuronal lesions of AD patients^9,10,19-24 and mouse models^23,25-27 provides strong evidence that the UPS controls degradation of tau. Furthermore, we have demonstrated that the accumulation of insoluble, hyperphosphorylated forms of tau both in vitro and in vivo correlates with progressively less functional UPS.²⁵ This is most likely due to the direct physical interaction of tau aggregates and oligomers with the proteasome machinery, which in turn can reduce the ability of cells to remove aggregated proteins via any clearance mechanism.²⁸ Thus, in principle, agents that enhance or protect proteasome function could have clinical utility in combating various neurodegenerative diseases characterized by toxic accumulation of misfolded proteins. Recently it has been shown that the proteasome’s ability to degrade proteins can be stimulated by agents that cause cyclic adenosine 3′,5′-monophosphate (cAMP) accumulation or by treatment with protein kinase A (PKA).^29-34 We have shown that chronic administration of rolipram (a phosphodiesterase 4 (PDE4) inhibitor) to mice with tauopathy activated proteasome function through cAMP/PKA-mediated phosphorylation and increased proteolysis, which resulted in decreased insoluble tau levels, attenuated tauopathy, and improved cognitive performance.²⁵ Rolipram use in humans has been hampered due to gastrointestinal side effects, especially nausea, emesis, and diarrhea, but more advanced, safer PDE inhibitors (PDEI) have been developed and approved by the FDA, such as PDE4 inhibitors, to treat asthma,³⁵ inflammation,³⁶ and chronic obstructive pulmonary disease³⁷; PDE3 inhibitors, to treat intermittent claudication³⁸ and ischemic stroke^39,40; and PDE5 inhibitors, to treat erectile dysfunction.⁴¹ Accumulating evidence from several animal models indicates that PDE inhibitors show cognitive enhancing capacity.^25,42-49 In particular, cilostazol, an FDA-approved PDE3 inhibitor, has been shown to decrease amyloid beta accumulation and protect against beta-amyloid-induced cognitive deficits in animal models.^50,51 Moreover, cilostazol’s potential application as a therapeutic for AD comes from a retrospective study showing that cilostazol improved memory in patients with mild cognitive impairment (MCI).^52-55 A clinical trial is underway to determine the therapeutic potential of cilostazol for patients with cognitive decline symptoms (ClinicalTrials.gov # NCT02491268).

Discovering new uses for approved drugs can greatly benefit patients as it can bypass early-stage development and safety testing and accelerate testing in clinical trials. In the present study, we investigated the effect of cilostazol in a mouse model of tauopathy that overexpresses human mutant P301L tau linked to FTDP-17, specifically in the cortex, limbic system, and basal ganglia.⁵⁶ This model (the rTg4510 line) is characterized by formation of abnormally hyperphosphorylated tau and neurofibrillary tangles-like pathology, spatial memory deficits, and progressive neuronal loss by 5–6 months of age.^56-58 Our data demonstrate that similar to rolipram, cilostazol enhances proteasome function via PKA, promotes the clearance of abnormal tau, and attenuates tauopathy and cognitive decline in early-stage tauopathy in rTg4510 mice.

MATERIALS AND METHODS

Animals.

Double-transgenic rTg4510⁵⁶ mice express human mutant (P301L) tau with 4 microtubule-binding domain repeats (4R0N). Line FVB-Tg (tetO-MAPT*P301L) under the control of mouse calcium-calmodulin kinase II-driven tetracycline controlled transcriptional activator (tTa) (line Tg (Camk2a-tTA). Strain of mice used is FVB/B6 F1. For the in vivo trial, rTg4510, 3- to 4-month-old males were used, as female mice from this line have more variability in tau expression. To test whether cilostazol could prevent overt tauopathy progression, we used 21 animals (vehicle, n = 10 and cilostazol n = 11). Cilostazol 3 mg/kg concentration or vehicle (0.5% carboxymethyl cellulose solution) was administered intraperitoneally twice daily for 30 days. No animal or extracted sample was excluded from any of the analyses. For the in vivo study, mice from the same litters were randomly assigned to each experimental group (vehicle or cilostazol). Mice were housed in 12-hour light to 12-hour dark cycles with free access to food and water. Animal experiments were in full compliance with the US National Institutes of Health Institutional Animal Care and Use Committee guidelines and overseen by Columbia University Medical Center. Mice were tested for cognitive performance in the Morris water maze (MWM) and short and long-term probe tests. Six hours after the final administration, mice were sacrificed. Investigators were blinded to mouse treatment during cognitive testing.

Antibodies.

Monoclonal antibodies to total human tau (CP27), pS396 and pS404 (PHF1), and pS202 and pT205 (AT8) tau were gifts from P. Davies. Polyclonal rabbit anti-pS214 tau was from Life Technologies (#44-742G); phospho (Ser and Thr) PKA substrate (#9621, this antibody detects proteins containing a phospho-serine/threonine residue with arginine at the −3 position) was from Cell Signaling; anti-Actin was from Sigma (clone AC-74 #A2228); anti-Rpt6/S8 (clone EPR13565(B), #PW9265) was from BIOMOL; and monoclonal mouse anti-proteasome 20S (α1–α7) (clone MCP231, #BML-PW8195) and rabbit polyclonal anti Rpn6 (#BML-PW8370) were from Enzo. Anti-Rpt5 (clone N-12,#sc-107976) was from Santa Cruz. Secondary antibodies were from Jackson Immunoresearch, anti-mouse (115-035-003) and anti-rabbit (115-036-003).

Morris water maze tests.

Behavioral testing started on day 25 of the drug treatments and lasted for 6 days. On the first day, mice underwent nonspatial (visible platform) training (lasting for 3 hours), employing an elevated platform above the water level marked by a flag. Mice were allowed to swim freely for 60 seconds. If they failed to locate the marked platform within time, they were gently guided to it. After mice were placed on the platform they were allowed to remain 10 seconds on the platform before they were removed from the pool and placed in their home cage. The MWM test was carried out as previously described.⁵⁹ After nonspatial training, mice underwent place-discrimination testing for 5 days with 2 trials per day. In the hidden platform acquisition test, mice were allowed to swim freely to search for the escape platform within a maximum of 60 seconds. The platform location remained constant throughout the test. The time taken to reach the platform was recorded as the escape latency. The mouse was kept on the platform for 10 seconds after it found the hidden platform. If a mouse failed to find the platform within 60 seconds, it was guided to the platform and placed on the platform for 10 seconds; in this case, the escape latency was recorded as 60 seconds for this trial. The experiment was repeated with 6 trials per mouse each day for 5 consecutive days. The mean escape latency was measured to evaluate the spatial learning ability.

Three hours and 24 hours after the hidden platform acquisition test, probe trials were conducted by removing the platform. Mice were placed in the diagonal quadrant of the hidden platform originally located and were allowed to swim freely in the pool for 60 seconds. The number of entries into the area around the original hidden platform, that is, the number of crossings over the previous platform location, was used to indicate the short-term and long-term memory maintenance.

Statistical methods were not employed to predetermine sample size. However, our published results with another PDE inhibitor gave us robust treatment results with similar number of animals. The current study is a follow-up which is designed to test for a treatment effect of an FDA-approved drug (cilostazol). Male mice from the same litter (4–5 mice per litter) were randomly assigned to each experimental group (vehicle or cilostazol). Investigators carrying out tests were blinded to treatment group allocation. No animal was excluded from any of the analyses. We did not apply formal statistical tests for normality or equality of variances but assumed an approximation to the normal distribution when appropriate. Statistical analyses for the MWM test were done by repeated-measures analysis of variance with Bonferroni correction (GraphPad Prism).

Native PolyAcrylamide Gel Electrophoresis (PAGE) assay for proteasome activity.

Frozen brain tissues were processed for proteasome activity according to published methods.⁶⁰ Briefly, frozen tissue was homogenized in a buffer containing 50 mM Tris-HCl, pH 7.4, 5 mM MgCl2, 5 mM ATP, 1 mM Dithiothreitol (DTT) and 10% glycerol, which preserves 26S proteasome assembly, and centrifuged at 20,000 × g for 18 minutes at 4°C. The supernatant was normalized for protein concentration. Samples were loaded on a 4% nondenaturing gel and run for 200 minutes at 160 V. Activity of the 26S proteasome was measured by 200 μM Suc-LLVY-amc (BACHEM Bioscience) diluted in homogenizing buffer. 26S proteasome bands were detected by UV light (365 nm) and photographed by iPhone 7S camera. Quantification of 26S proteasome activity and levels was done using imageJ software. The activity of 2- and 1-cap 26S proteasome was divided by the corresponding proteasome levels.

Immunoblot analysis.

Samples (5–20 μg protein) were typically run on 4%–12% Bis-Tris gels (Life Technologies; WG1403BOX10) using 3-(N-morpholino)propanesulfonic acid buffer (NP0001). Proteins were analyzed after electrophoresis on sodium dodecyl sulphate-PAGE and transferred onto 0.2-μm nitrocellulose membranes (Whatman). Blots were blocked and incubated with primary and secondary antibodies at concentrations listed below. Membranes were developed with enhanced chemiluminescent reagent (Immobilon Western HRP substrate) and luminol reagent (WBKLS0500, Millipore) using a Fujifilm LAS3000 imaging system. ImageJ (http://rsb.info.nih.gov/ij) was used to quantify the signal. Relative intensity (fold change or fold increase, no units) is the ratio of the value for each protein to the value of the respective loading control. Dilutions for western blotting were as follows: mouse monoclonal human anti-tau (CP27, 1:5000); pS396 and pS404 (PHF1, 1:5000), and pS202 and pT205 (AT8, 1:5000); anti-pS214 tau (1:3000); anti-phospho-(Ser and Thr) PKA substrate (1:1000); anti-actin (1:5000); anti-proteasome, 19S regulatory particle: anti-Rpt6 (1:1000), Rpn6 (1:500), and Rpt5 (1:1000); and 20S proteasome α1–α7 (1:1000). Secondary antibodies were from Jackson ImmunoResearch: anti-mouse (1:3000) and anti-rabbit (1:3000) diluted in blocking buffer containing 5% milk.

Immunofluorescence.

Mouse brains were isolated after transcardial perfusion with PBS, drop-fixed in 4% PFA overnight, and then subjected to cryoprotection treatment in 30% sucrose in PBS for 24 hours. Free-floating brain sections (35 μm) from brains sectioned in the sagittal plane were used. The sections were incubated at 4°C overnight with primary antibody diluted in PBS containing 0.3% Triton X-100 and 5% normal goat serum blocking solution (Vector Laboratories, #S-1000). Antibodies were as follows: anti-mouse monoclonal human tau (CP27, 1:2000); pS396 and pS404 (PHF1, 1:1000). Following washes, sections were incubated with goat anti-mouse IgG Alexa 594 (ThermoFisher Scientific #A-11005, 1:500). Staining was visualized by Zeiss Axio Vision Imager Z1 microscope. Images were processed using AxioVision 4.8 image software. We used ImageJ program to quantify tau positive neurons per area. For generation of graphs and statistical analysis, we used GraphPad Prism.

Purification of 26S proteasomes.

Cortical pieces from all the mice were pooled each time proteasome purification was carried out, and 26S proteasomes were affinity purified using a UBL (ubiquitin-like) domain as the ligand.⁶¹ Briefly, brain homogenates were spun for 1 hour at 100,000 × g. The soluble extracts were incubated at 4°C with 2 mg/mL glutathione-S-transferase-ubiquitin-like domain and a corresponding amount of glutathione-Sepharose 4B (GSH-Sepharose) for 2 hours. The slurry containing 26S proteasomes bound to glutathione-S-transferase-ubiquitin-like domain was poured into an empty column and washed, then incubated with 2 mg/mL His10-ubiquitin-interacting motif (10× His-UIM). The eluate was collected and incubated with Ni2+-NTA-agarose for 20 minutes at 4°C. The Ni2+-NTA-bound 10× His-UIM was removed by filtration. The resulting flow-through (~0.6 mL) contained purified 26S proteasomes. The molarity of 26S proteasome particles was calculated assuming a molecular weight of 2.5 MDa. To assess peptidase capacity of 26S proteasome subunits, we incubated 10 nM of proteasome with 50 μM Suc-LLVY-amc fluorogenic peptide for chymotrypsin-like activity (β5 activity). Kinetic reactions were carried out for a period of 90 minutes. The rate of kinetic reaction (slope) over time was calculated. Fluorescence signal was captured at 380 nm excitation and at 460 nm emission by Infinite 200 PRO multimode reader (TECAN).

Tissue fractionation and protein extraction.

Mice were sacrificed by cervical dislocation, and the brains were immediately dissected on wet ice and stored on dry ice. Briefly, frozen hemispheres free of cerebellum and brainstem were weighed and homogenized without thawing in Radioimmunoprecipitation assay (RIPA) buffer (10 × volume/weight) (50 mM Tris-HCl, pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM Ethylenediaminetetraacetic acid, 1 mM phenylmethyl-sulfonyl, 1 mM sodium orthovanadate, 1 mM sodium fluoride (NaF), 1 μL/mL protease inhibitor mix (#P8340, Sigma-Aldrich). Homogenates were centrifuged for 10 minutes at 3000 × g at 4°C. Protein concentrations, determined by the bicinchoninic acid assay, were performed on the clear supernatants representing the total extract used for analysis of the total protein levels. Sample volumes were adjusted with RIPA buffer containing 100 mM DTT and NuPAGE lithium dodecyl sulfate (LDS) Sample Buffer 4× buffer (Life Technologies) and boiled for 5 minutes. The sarkosyl-insoluble extracts, which are highly enriched in aggregated tau species, were generated when 200 μg aliquots from the total protein extracts were normalized into 200 μl final volume containing 1% sarkosyl, followed by ultracentrifugation at 100,000 × g for 1 hour at 20°C. Without disturbing the pellet, the supernatant was transferred to new tubes. The pellet was resuspended in 100 μL RIPA buffer containing DTT and NuPAGE LDS Sample Buffer 4× buffer, followed by vortexing for 1 minute and 5 minutes of heating at 95°C. The heat-stable extract, which contains soluble tau, was obtained when the supernatant was further processed, first by heating for 5 minutes at 95°C followed by 30 minutes of centrifugation at 20,000 × g. Extracts were transferred to new tubes containing NuPAGE LDS Sample Buffer 4× buffer (4:1 (extracts/buffer) ratio).

Statistical analysis.

Statistical analyses were performed with Prism6 (Graphpad Software, San Diego, Calif.). For quantitative immunoblotting, semiquantitative immunofluorescence, proteasome assays, and probe testing, we used unpaired 2-tailed, Student’s t test between groups with unequal variance. For MWM testing we used repeated-measures analysis of variance followed by Bonferroni correction for multiple comparisons. Data represent mean ± s.e.m.

RESULTS

Cilostazol reduces accumulation of toxic tau aggregates.

Cilostazol has been approved in the United States since 1999 for the treatment of symptoms of intermittent claudication due to its vasodilatory actions and its effect on inhibition of platelet aggregation.⁶² Given the safety profile of cilostazol and its effect in ameliorating cognitive decline in patients with MCI,^63,64 we tested whether cilostazol can attenuate tauopathy and cognitive impairment in the rTg4510 mouse line. Cilostazol, at 3 mg/kg concentration or vehicle (0.5% carboxymethyl cellulose solution) were injected twice daily for 30 days to rTg4510 mice at 3–4 months of age, an age when they have early-stage tauopathy. After behavioral testing, cortices were isolated and immediately processed for biochemistry (quantitative immunoblotting and proteasome function testing) or immunohistochemistry. Extracts containing all tau species (total extract) or enriched for sarkosyl-insoluble (insoluble extracts) tau species were generated. Cilostazol administration significantly reduced the load of all tau forms (identified using antibody CP27) and disease-associated phosphorylated tau (epitopes pS396pS404 (PHF1) and pS202/pT205 (AT 8) in the total (Fig 1, A) and the insoluble extracts (Fig 1, B). The same significant decrease was observed for the total tau (Fig 1, C, D, and G) and pS396pS404 tau (Fig 1, E, F, and H) when assessed by semiquantitative immunohistochemistry.

Fig 1.

Cilostazol administration reduces the levels of total and aggregated tau species in rTg4510 mice. (A, B) Immunoblot analysis and corresponding densitometric quantification of total and insoluble extracts of tau (CP 27 antibody), pS396 and pS404 (PHF1 antibody), and pS202 and pThr205 (AT 8 antibody) tau epitopes from cortical tissue of rTg4510 mice treated with vehicle or cilostazol. Actin was used as a loading control. (C, D, E, F) Immunofluorescence labeling and (G and H) quantification of fluorescence signal for (C and D) total tau and for (E and F) pS396 and pS404. (G and H) quantification of fluorescence puncta in the frontal cortex of rTg4510 mice treated with vehicle or cilostazol. Scale bars, 200 μm. Bar graphs for (A) and (B) represent quantification of tau immunoreactivity normalized to actin, and immunoreactivity for PHF1 and AT8 was normalized to total tau. For statistical analyses, we performed unpaired 2-tailed Student’s t test by GraphPad Prism software. For quantification of immunofluorescence signal (C, D, E, and F), slices from 6 mice per treatment group were analyzed. Error bars represent, mean ± s.e.m.; *P < 0.05, **P < 0.01, ***P < 0.001. Vehicle treatment served as control for all experiments.

Cilostazol enhances proteolytic activity of the 26S proteasomes.

To assess whether cilostazol is a positive regulator of proteasome function, we first measured peptidase activity of the 26S proteasomes by native PAGE in-gel activity assay. This assay enables good resolution of functionally intact 2- and 1-cap 26S proteasomes so that the activity of both forms of 26S proteasomes can be assessed separately. Cortical brain homogenates were prepared in nondenaturing conditions, and chymotrypsin-like activity of 26S proteasomes was assessed by incubating the gel for 10 minutes with 200 μM Suc-LLVY-amc, a fluorogenic substrate for the chymotrypsin-like activity of the proteasome. Cilostazol markedly increased the 2- and 1-cap 26S proteasome function (Fig 2, A), whereas the levels of the 2 forms of proteasomes were not changed between treatment groups (Fig 2, B). Interestingly, the activity of 2-cap proteasomes, which are composed of 2 19S regulatory subunits, showed significantly more activity when compared with 1-cap (1 regulatory subunit) 26S proteasomes in the cilostazol-treated group (Fig 2, C). This suggests that the changes in the 19S regulatory subunits render the elevated degradation capacity of 26S proteasomes. Indeed, independent studies have confirmed that several kinases such as PKA,^30,32,33 CaMKIIα,^65,66 and DYRK2⁶⁷ (dual specificity tyrosine-phosphorylation-regulated kinase 2) preferentially phosphorylate subunits of the 19S regulatory particle of the 26S holoenzyme which can account for the hyperactivation of 2- vs 1-cap 26S proteasomes. Additionally, the total levels of proteasome subunits that comprise the 19S regulatory (Rpt5 and Rpt6) and the 20S subunits remained unchanged (Fig 2, D). The same degree of increased activity of proteasome forms was seen in primary cortical neurons from WT animals after 8 hours of treatment (data not shown). We next purified 26S proteasomes from the brains of cilostazol- and vehicle-treated mice to assess the degradation rate of proteasome fluorogenic substrate (Suc-LLVY-amc) by kinetic analysis. Purified 26S proteasomes from cilostazol-treated animals showed a significant increase in hydrolyzing activity as determined by the degradation curve (Fig 2, E) and by the slope of the degradation curve (Fig 2, F) compared with 26S proteasomes from vehicle-treated animals.

Fig 2.

Cilostazol administration increases proteolytic activity of 26S proteasomes in vivo. (A) Native PAGE ingel activity of 2 and 1 cap 26S proteasomes. The fluorescence signal emitted by cleaved Suc-LLVY-amc was visualized upon exposure to UV light (365 nm). The bands represent peptidase activity of 2 and 1 cap 26S proteasomes. (B) Following proteasome activity, native gel was transferred and probed for antibody against 20S proteasome subunits (α1–α7) to determine the levels of 2 and 1 cap 26S proteasomes. (C) Densitometric quantification of the 2 and 1 cap 26S proteasome activity, normalized to the levels of 2 and 1 cap 26S proteasomes in vehicle and cilostazol treated rTg4510 mice. (D) Immunoblot analysis of proteasome subunits in total lysate. Rpt5 and Rpt6 represent the 19S regulatory particle of 26S proteasomes and α1–α7 represent the 20S core particle of 26S proteasomes, from vehicle and cilostazol treated mice. (E) Kinetic assay for hydrolyzation of Suc-LLVY-amc substrate by purified 26S proteasomes from vehicle- and cilostazol-treated animals. Emitted fluorescence of cleaved amc was monitored over time by Infinite 200 PRO multimode. (F) The slope (rate) of hydrolysis by 26S proteasomes purified from vehicle- and cilostazol-treated mice. Statistical analyses for native in-gel activity assay, immunoblotting (cilostazol n = 7 and vehicle n = 6) and the rate of substrate degradation by 26S proteasomes were carried out using unpaired 2-tailed Student’s t test between groups. Purified proteasomes were pooled from all the animals within treatment groups and at least 3 independent experiments were performed. Error bars represent mean ± s.e.m.; *P < 0.05, **P < 0.01. PAGE, polyacrylamide gel electrophoresis.

Cilostazol enhances proteasome function via cAMP/PKA pathway.

PDE3 isoforms are expressed throughout the brain, including cortex and hippocampus where they regulate cAMP hydrolysis.^68,69 Several studies have shown elevated cAMP levels in the brain tissue upon administration of PDE3 inhibitors.^70,71 Our previous study has shown that cAMP/PKA-mediated proteasome activation is dependent on phosphorylation of several proteasome subunits by activated PKA.²⁵ In support of this finding, we found that purified 26S proteasomes from cilostazol-treated mice showed a marked increase in the phosphorylation of serine and threonine of multiple proteasome subunits when immunoprobed with PKA-specific phospho-(Ser/Thr) epitopes (Fig 3, A), but with no change in the levels of the subunits of the 26S proteasome (Fig 3, B). Further confirmation that cilostazol activated PKA was indicated by the significant increase in the level of phosphorylated tau at serine 214 (pS214) (Fig 3, C) in the total tau fraction. This tau epitope is a well-established PKA substrate^72,73; hence, it not a surprise that the S214 epitope is significantly phosphorylated in cilostazol-treated animals, whereas PHF1 and AT8 epitopes (Fig 1, A) are significantly reduced in the total extract fractions. In the insoluble fraction, pS214 epitope was reduced similarly (Fig 3, C) as were other phospho-epitopes shown in Fig 1, B due to overall reduction of aggregated tau forms. The negative correlation between the levels of pS214 epitope and the levels of total tau and other phosphorylated tau species suggests that at least in the early stages of tauopathy, pS214 does not appear to be a pathologic marker and does not lead to aggregation of tau. In later stages of the disease, the inverse relationship between pS214 with other phospho tau epitopes is lost,²⁵ possibly due to extensive hyperphosphorylation and aggregation of tau.

Fig 3.

Inhibition of PDE3 by cilostazol results in PKA-dependent phosphorylation of 26S proteasome subunits. (A) Immunoblot analysis of PKA-specific phosphorylation of serine and threonine residues (pSer and pThre), (B) proteasome subunits Rpn6 and α-subunits 1–7 (α1–α7) in purified 26S proteasomes from vehicle- and cilostazol-treated rTg4510 mice. (C) Immunoblot analysis and corresponding densitometric quantification of total and insoluble extracts of phospho S214 tau epitope from cortical tissue of rTg4510 mice treated with vehicle or cilostazol. Statistical analyses employed 2-tailed Student’s t test between groups. Error bars represent, mean ± s.e.m.; **P < 0.01. PDE3, phosphodiesterase 3; PKA, protein kinase A.

Cilostazol improves cognitive performance in tauopathy mice.

The effect of cilostazol on tauopathy-associated cognitive impairment was assessed using the MWM test which reflects spatial reference memory. The rTg4510 mice show progressive memory loss with a significant deficit being seen during early-stage tauopathy (4–5 months of age).^56,57 Our previous study with these mice showed that rolipram attenuated cognitive decline in mice with early-, but not late-stage tauopathy when the proteasome is already significantly impaired.²⁵ A similar improvement in cognitive performance was seen in cilostazol-treated mice, whereby the cilostazol group showed significant reduction in escape latency compared with the vehicle-treated group (Fig 4, A). Moreover, compared with vehicle-treated littermates, cilostazol-treated mice crossed the platform location significantly more times in the immediate probe trials at 3 and at 24 hours after platform removal (Fig 4, B and C, respectively). The probe trials (3 and 24 hours) measure short- and long-term memory and were carried out immediately after escape latency trials. Percent time spent in the target quadrant was unchanged between treatment groups (Fig 4, D and E).

Fig 4.

Cilostazol administration improves cognition in early-stage tauopathy mice. (A) Spatial learning and cognitive performance assessed by Morris water maze (MWM) hidden platform in rTg4510 mice treated with vehicle (n = 10) or cilostazol (n = 11). Probe trials and number of crossings from the previous platform location (B) 3 hours after completing MWM testing and (C) 24 hours after completing MWM testing. (D and E) Time spent in target quadrant during probe 1 (3 hours) and 2 (24 hours) trials. One-way repeated-measures ANOVAs with Bonferroni post-tests were used to compare the escape latencies in 5 days of continuous MWM hidden platform trials. Separate unpaired t tests were used to compare the number of platform crossings in both 3- and 24-hour probe trials. Data are expressed as mean ± s.e.m. n.s., not significant; *P < 0.05, **P < 0.01. ANOVA, analysis of variance.

DISCUSSION

In AD, the distribution of tau tangles and not the distribution of plaques is very consistent, and tangles correlate strongly with cognitive decline and with neuronal and synapse loss.^74-77 Therefore, discovering therapies that can prevent or reduce tau tangles could have a real clinical impact in ameliorating or delaying cognitive impairment in AD patients.

In the present study, we show that cilostazol can reduce tau aggregates and improve cognitive performance via cAMP/PKA-mediated proteasome activation. The potential relevance of proteasome activation for the treatment of neurodegenerative disease has been shown recently in preclinical studies in tauopathy by our group²⁵ and in HD mouse model by others.⁷⁸ Emerging research on the UPS shows that 26S proteasomes are not simply a passive docking and degradation complex of ubiquitin-tagged proteins, but rather, the 26S proteasome is a finely regulated machine that can initiate neurotoxic protein clearance by phosphorylation in response to drug treatment. Our present study confirms our findings that PDE inhibitors can act as proteasome activators in vivo, and it demonstrates that this newly identified mechanism is not restricted to PDE4, making it more likely that it may be common to all drugs that elevate cAMP/PKA in brain cells. Given that the UPS clears several types of misfolded proteins, the approach may have utility for other neurodegenerative diseases. In this respect, it will be of interest to monitor the outcome of clinical trials using ibudilast (a PDE4 inhibitor), which is being tested in ALS patients (ClinicalTrials.gov, # NCT02238626).

Developing a novel drug for a newly identified target is a complex process which can take 10–15 years and comes with substantial costs.⁷⁹ Thus, repurposing existing drugs for a new therapeutic application is a promising approach that may have rapid clinical impact for diseases with no available treatments. PDE inhibitors have been proposed to alleviate dementia in AD, as they have been reported to act as cognitive enhancers in animal models^25,42-44,48 and in studies in patients.^52-54 Of note, retrospective analysis showed that cilostazol, prescribed as an antiplatelet drug, can preserve cognitive function in patients with MCI, which has been attributed to its favorable effect on cerebral vascular circulation.⁵³ Unfortunately, the effect on tau pathology in these patients is unknown. MCI is associated with early-stage AD which has been defined in postmortem mapping studies by the degree and distribution of tauopathy.^80-82 Whether cilostazol (or proteasome activators in general) can attenuate the levels and impact of pathologic proteins such as tau at later disease stages is unknown, but given our data showing that administering rolipram to mice at a late stage of disease had no effect on proteasome activation, tauopathy, or cognition, it is less likely that proteasome activation will be an appropriate therapeutic approach for late-stage diseases. Rather, our study suggests that early-stage AD patients may benefit from treatment with cilostazol, which has a good safety profile for treatment of chronic disorders.

In conclusion, our data have uncovered a new mechanism of action of clinically relevant PDE inhibitors as proteasome activators that can attenuate proteotoxicity and thus are a promising pharmacologic targets for treatments of neurodegenerative diseases.

AT A GLANCE COMMENTARY.

Schaler AW and Myeku N

Background

• Alzheimer’s disease and several other tauopathy disorders are characterized by accumulation of abnormal tau protein into aggregates. Tau and most proteins are removed from cells via the ubiquitin proteasome system. Here we show that activation of proteasome function in a relevant mouse model leads to attenuated tauopathy and improved cognitive performance.

Translational Significance

• This study has uncovered a new therapeutic mechanism of action of phosphodiesterase 3 inhibitor as an enhancer of proteasome function via activated cyclic adenosine 3′,5′,-monophosphate/protein kinase pathway in vivo. The study suggests that, cilostazol, a clinically safe drug, has a potential to be repurposed for therapies against tauopathies for rapid clinical impact.

Abbreviations:

AD

Alzheimer’s disease

CBD

corticobasal degeneration

FTD

frontotemporal degeneration

FDA

Food and Drug Administration

GST-UBL

glutathione-S-transferase-ubiquitin-like domain

HD

Huntington’s disease

MWM

Morris water maze

MCI

mild cognitive impairment

PDEI

phosphodiesterase inhibitor

PKA

protein kinase A

PSP

progressive supranuclear palsy

UPS

ubiquitin proteasome system