https://orcid.org/0000-0002-3956-276X
Abstract
Hyperphosphorylation of the microtubule-associated protein tau is associated with many neurodegenerative diseases, including Alzheimer’s disease. Microtubule affinity-regulating kinases (MARK) 1–4 and cyclin-dependent kinase 5 (Cdk5) are tau kinases under physiological and pathological conditions. However, their functional relationship remains elusive. Here, we report a novel mechanism by which Cdk5 activates MARK4 and augments tau phosphorylation, accumulation and toxicity. MARK4 is highly phosphorylated at multiple sites in the brain and in cultured neurons, and inhibition of Cdk5 activity reduces phosphorylation levels of MARK4. MARK4 is known to be activated by phosphorylation at its activation loop by liver kinase B1 (LKB1). In contrast, Cdk5 increased phosphorylation of MARK4 in the spacer domain, but not in the activation loop, and enhanced its kinase activity, suggesting a novel mechanism by which Cdk5 regulates MARK4 activity. We also demonstrated that co-expression of Cdk5 and MARK4 in mammalian cultured cells significantly increased the levels of tau phosphorylation at both Cdk5 target sites (SP/TP sites) and MARK target sites (Ser262), as well as the levels of total tau. Furthermore, using a Drosophila model of tau toxicity, we demonstrated that Cdk5 promoted tau accumulation and tau-induced neurodegeneration via increasing tau phosphorylation levels at Ser262 by a fly ortholog of MARK, Par-1. This study suggests a novel mechanism by which Cdk5 and MARK4 synergistically increase tau phosphorylation and accumulation, consequently promoting neurodegeneration in disease pathogenesis.
Introduction
The microtubule-associated protein tau is a phosphoprotein whose major function is to regulate microtubule stability in axons (1). Tau is also known to form paired helical filaments and deposits of neurofibrillary tangles in the brains of patients with neurodegenerative diseases referred to as tauopathies, including Alzheimer’s disease (2–7). In diseased brains, tau is highly phosphorylated at additional sites, and phosphorylation status correlates with severity of pathology (4–6,8). Tau kinases include proline-directed Ser/Thr kinases (SP/TP kinases) such as Cyclin-dependent kinase 5 (Cdk5), GSK3β (10) and MAPK (11), as well as non-SP/TP kinases including microtubule affinity-regulating kinase (MARK)/Par-1, AMPK, PKA, PKC, CK, Chk2, CaMKII, p70S6K1 and NuaK1 (9–19). Blocking tau hyperphosphorylation has been suggested as a therapeutic strategy (7), and elucidating the mechanisms underlying abnormal tau phosphorylation is crucial to understanding disease pathogenesis.
Hyperphosphorylation of tau is achieved in a sequential manner, with Cdk5 and MARK/Par-1 both known to play initiating roles (9,20). Cdk5 was initially identified as a tau protein kinase (21) and appeared to be a multifunctional kinase, involved in physiological functions such as development and learning and memory (22). Cdk5 is activated by binding to an activating subunit, p35, which can be cleaved by calpain to produce a constitutively active form, p25 (23). Levels of p25 are increased in Alzheimer’s disease brains, and expression of p25 is sufficient to cause tau hyperphosphorylation and enhance neurodegeneration in Drosophila and mouse models of tauopathies (24,25). Cdk5 phosphorylates tau at SP/TP sites in the flanking regions of the microtubule-binding repeats (mainly Ser202, Thr205, Ser235 and Ser404; 9) and acts as the priming kinase for GSK3β in tau phosphorylation at (Ser/Thr)-X3-(pSer/pThr) sites in vitro and in cultured neurons (9). Thus, abnormal activation of Cdk5 leads to hyperphosphorylation of tau and neurotoxicity (26).
MARK/Par-1 is a highly conserved AMPK-related kinase that forms a family of four members in mammals (MARK1–4; 2). MARK/Par-1 phosphorylates tau within the microtubule-binding repeats, Ser262 and Ser356, which play critical roles in the physiological functions and pathological changes in tau. It regulates the binding affinity of tau for microtubules (13), and in diseased brains, phosphorylation at Ser262 is an early pathological change that plays an initiating role in abnormal metabolism and accumulation of tau (12,19,27–31). Blocking tau phosphorylation at Ser262 effectively decreases tau levels and mitigates tau-induced neurodegeneration in cultured cells, Drosophila and mouse models (12,19,30–34). MARK4 colocalizes with early pathological changes in AD brains (35), and de novo mutations in MARK4 have been linked to an elevated risk of early-onset AD (36), suggesting that the deregulation of MARK4 plays a role in the onset of AD. MARK4 is known to be activated by phosphorylation at its activation loop by liver kinase B1 (LKB1; 37). However, the mechanism leading to abnormal activation of MARK4 in disease pathogenesis is not well understood.
In this study, we demonstrated that Cdk5 regulates MARK4 via a novel mechanism and synergistically augments hyperphosphorylation, accumulation and toxicity of tau in vitro and in vivo. Our findings suggest that aberrant activation of Cdk5 is one of the mechanisms leading to the activation of MARK4 and promotion of tau-mediated neurodegeneration.
Results
MARK4 is highly phosphorylated in mouse fetal brain and neurons
MARK4 activity is known to be regulated through phosphorylation at T214 and T568 by LKB1 and aPKC, respectively (37,38). In addition, MARK4 possesses several potential phosphorylation sites for proline-directed kinases including Cdk5 (Fig. 2A). We asked whether Cdk5 is involved in MARK4 phosphorylation using Phos-tag analysis, which allows comprehensive quantitative profiling of the phosphorylation status of the target protein (39). MARK4 proteins from mouse fetal cortices were separated into multiple bands (Fig. 1). Treatment with alkaline phosphatase shifted multiple bands lower, indicating that bands in Phos-tag gel reflect different phosphorylation status of MARK4 (Supplementary Material, Fig. S1). We also found that the MARK4 proteins in the brain extract migrated much slower than those exogenously expressed in HEK293 cells (Fig. 1, HEK293), suggesting that MARK4 is highly phosphorylated in the brain. Endogenous MARK4 proteins in primary cultured neurons prepared from embryonic mouse brains showed a migration pattern similar to those from mouse brains, suggesting that MARK4 is highly phosphorylated in neurons (Fig. 1). To test whether Cdk5 is involved in MARK4 phosphorylation, primary cultured neurons were treated with roscovitine, a specific inhibitor of Cdk5. Treatment with roscovitine reduced the slowly migrating bands, while increasing the faster migrating bands (arrowhead; Fig. 1), without any change in the levels of total MARK4 (Fig. 1, lower panel, MARK4 Laemmli’s SDS Polyacrylamide Gel Electrophoresis-PAGE). These results suggest that Cdk5 is involved in MARK4 phosphorylation in brain neurons.
MARK4 is highly phosphorylated in mouse brains and neurons. HEK293 cells transfected with MARK4 (HEK293), whole extracts of mouse fetal cortex (cortex) or primary neurons (neurons) were separated by Phos-tag SDS-PAGE. MARK4 exhibited multiple bands in the high molecular weight region. Treatment of primary neurons with roscovitine (Ros) induced an increase in lower bands (arrowheads).
Cdk5 is involved in phosphorylation of MARK4 in the spacer domain
There are several SP/TP sites in MARK4, and they are clustered in the spacer domain (Fig. 2A). To test whether Cdk5 affects MARK4 phosphorylation at these sites, we constructed a MARK4 mutant, MARK47A, carrying unphosphorylatable alanine at those sites (Fig. 2A). We then analyzed the changes in phosphorylation patterns of wild-type MARK4 and MARK47A by co-expression of Cdk5-p35 in non-neuronal HEK293 cells (Fig. 2A). Phos-tag SDS-PAGE separated wild-type MARK4 proteins expressed in HEK293 cells into multiple bands reflecting their phosphorylation status, including the faster-migrating band of unphosphorylated MARK4 (up) and multiple signals of phosphorylated MARK4 including distinct bands (arrowhead) and smear signals in the higher molecular weight region (Fig. 2B). With co-expression of Cdk5-p35, MARK4 proteins that migrate faster (up) reduced and those migrate slower increased, and additional distinct bands at the higher molecular weight region (asterisks) appeared, suggesting that MARK4 is further phosphorylated by Cdk5. By contrast, co-expression of Cdk5-p35 did not affect the Phos-tag band pattern of MARK47A (Fig. 2B).
Cdk5 phosphorylates MARK4 at the spacer domain. (A) Schematic structure of human MARK4 (upper panel) and MARK47A (lower panel). (B) Wild-type (WT) or MARK47A (7A) was separated by Phos-tag SDS-PAGE. Co-expression of Cdk5-p35 induced upward shifts (asterisks) in WT but not in MARK47A. Unphosphorylated form of MARK4 indicated as up. (C) Incorporation of 32P into kinase-dead MARK4 (KD MARK4) by Cdk5-p25 was detected by autoradiography (ARG) (Fig. 2B).
To further confirm that Cdk5 directly phosphorylates MARK4, in vitro kinase assay was carried out. Since autophosphorylation of MARK4 might compromise the result, we used a kinase-dead form of MARK4 (MARK4D199A, indicated as KD MARK4) as a substrate. KD MARK4 was transfected to HEK293 cells and immunoprecipitated, then incubated with purified Cdk5-p25 in the presence of [γ-32P]ATP. Incorporation of 32P into MARK4D199A was detected in the presence of Cdk5, but not observed in the sample without Cdk5 (Fig. 2C).
These results indicate that Cdk5 phosphorylates MARK4 at the spacer domain.
Cdk5 increases MARK4 kinase activity through phosphorylation in the spacer domain
We examined whether Cdk5 affects wild-type MARK4 activity using an in vitro kinase assay. Wild-type MARK4 was expressed in HEK293 cells with or without Cdk5-p35. MARK4 proteins were immunoprecipitated from the cell lysate, and MARK4 kinase activity was measured using CHKtide, a peptide substrate, which is derived from the sequence containing phosphorylation site of Cdc25C by MARK3 and widely used to assess the kinase activity of MARK family members. We found that co-expression of Cdk5-p35 significantly increased wild-type MARK4 activity (Fig. 3) without changing total levels of MARK4 (see Fig. 2B, MARK4). We further asked whether MARK4 phosphorylation in the spacer domain is required for Cdk5 to increase MARK4 activity. In contrast to wild-type MARK4, co-expression of Cdk5-p35 failed to increase the kinase activity of MARK47A (Fig. 3), suggesting that MARK4 phosphorylation in the spacer domain is involved in the enhancement of MARK4 activity by Cdk5.
Cdk5 activates MARK4 through its spacer domain. In vitro kinase assay for wild-type MARK4 or MARK47A with or without Cdk5-p35. Incorporation of 32P into the substrate of MARK4 is expressed as the mean ± SEM (n = 3; *P < 0.05; **P < 0.05, Student’s t-test).
Cdk5 and MARK4 synergistically increase total tau and phosphorylation levels of tau in mammalian culture cells
Cdk5 phosphorylates tau at SP/TP sites in the flanking regions of the microtubule-binding repeats (mainly Ser202, Thr205, Ser235 and Ser404; 40), while MARK/Par-1 phosphorylates tau at non-SP/TP sites within the microtubule-binding repeats, Ser262 and Ser356 (2). Since Cdk5 increases MARK4 activity, we asked whether Cdk5 also enhanced tau phosphorylation at Ser262 and Ser356 in the presence of MARK4. In HEK293 cells, exogenously expressed human tau was phosphorylated at Cdk5 sites (Ser235 and Ser404), and expression of Cdk5-p35 further increased the phosphorylation levels at these sites (Fig. 4A, Cdk5 sites, also see Fig. S2). In sharp contrast, exogenously expressed tau was not phosphorylated at MARK4 sites (Ser262 and Ser356) in HEK293 cells without exogenous expression of MARK4 (Fig. 4A, MARK sites). Co-expression of Cdk5-p35 and MARK4 dramatically increased the levels of tau phosphorylated at MARK sites Ser262 and Ser356 (Fig. 4A, MARK sites). It is reported previously that tau phosphorylation at Ser262 and Ser356 stabilizes tau and leads to an increase in total tau levels (19,30,31). Total tau levels were increased with co-expression of Cdk5-p35 and MARK4 (Fig. 4A, total tau, also see Supplementary Material, Fig. S2). These results support our hypothesis that Cdk5 synergistically promotes tau phosphorylation through the activation of MARK4.
Cdk5 and MARK4 synergistically phosphorylate tau. (A) Cdk5-p35, MARK4 or both were expressed in HEK293 cells together with human tau, and the phosphorylation of tau at Cdk5 sites (Ser235 and Ser404) or MARK sites (Ser262 and Ser356) was examined by western blotting using phospho-specific antibodies. A quantitation is shown (Mean ± SD; n = 3; *P < 0.05; **P < 0.01 between MARK4 and Cdk5+MARK4, Student’s t-test). (B) A similar experiment was carried out either p35 or p25 as Cdk5-activating subunit.
Activating subunits of Cdk5 include p35 and its proteolytic fragment p25. p25 is a constitutively active form, and its increase is associated with neurodegenerative conditions (24). Thus, we were motivated to compare the relative activity of p25 and p35 in MARK4 activation. When Cdk5 was activated by p25, the levels of tau phosphorylated at MARK sites as well as total tau levels were increased with higher degree than those of p35 (Fig. 4B).
Cdk5 and MARK4/Par-1 synergistically increase the levels of phosphorylated tau in Drosophila models
We next asked whether a synergistic increase in tau phosphorylation levels by Cdk5 and MARK/Par-1 can be observed in vivo using Drosophila tauopathy models (41). Both Cdk5 and MARKs are conserved in Drosophila, and Cdk5 and Par-1 are functional homologs of mammalian Cdk5 and MARKs (12,42). Human tau ectopically expressed in Drosophila retina is phosphorylated at many disease-associated sites including Ser235 and Ser262. In Drosophila retina, tau phosphorylation at a non-SP/TP site Ser262 is mediated by endogenous Par-1 (12). To examine whether Cdk5 affects tau phosphorylation at Ser262, human tau and Drosophila Cdk5 were co-expressed under the control of the pan-retinal gmr-Gal4 driver. Head homogenates were subjected to western blotting with anti-tau or anti-phospho-tau antibodies. Co-expressing Cdk5-p35 with tau increased the levels of tau phosphorylated at Ser235, the SP/TP site in tau that is known as a direct target for Cdk5. Furthermore, consistent with the results in cultured HEK293 cells co-expressing Cdk5-p35 and MARK4 (Fig. 4), the levels of tau phosphorylated at pSer262 sites were elevated by expression of Cdk5 (Fig. 5A). Tau phosphorylation at MARK/Par-1 target sites Ser262/356 is known to stabilize tau and increases total tau levels (19,30,31). As expected from elevated levels of tau phosphorylated at Ser262, total tau levels were also elevated by co-expression of Cdk5 (Fig. 5A, also see Supplementary Material, Fig. S4). Moreover, RNAi-mediated knockdown of Cdk5 in Drosophila retina (Supplementary Material, Fig. S3) not only reduced the levels of tau phosphorylated at Ser235 but also reduced the levels of tau phosphorylated at Ser262 and those of total tau (Fig. 5B). These results suggest that Cdk5 positively regulates the levels of tau phosphorylated at non-direct target site Ser262 in vivo.
Cdk5 enhances tau phosphorylation at Ser262 via Par-1 in a Drosophila model. Overexpression (A) or knockdown (B) of Cdk5 in Drosophila eyes expressing human tau. Phosphorylation of tau at a Cdk5 site (Ser235) and a Par-1 site (Ser262), and total tau was examined by western blotting. (C) Cdk5 was overexpressed under Par-1 knockdown conditions. Actin was used as a loading control. Representative blots (left panels) and quantification (right panels) are shown. Mean ± SD; n = 4. n.s., P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.005 (Student’s t-test).
To further confirm that the observed effects of Cdk5 are mediated through Par-1, the experiment shown in Figure 5A was carried out in a Par-1 knockdown background. If the increase in tau phosphorylation at Ser262 caused by Cdk5 is mediated by Par-1, Cdk5 would not increase the levels of tau phosphorylated at Ser262 in Par-1 knockdown background. Expression of Cdk5 increased the levels of tau phosphorylated at Ser235 under Par-1 knockdown conditions (Fig. 5C), suggesting that Cdk5 activity was not affected by Par-1. By contrast, expression of Cdk5 did not increase either the levels of tau phosphorylated at Ser262 or the levels of total tau under Par-1 knockdown conditions (Fig. 5C).
Taken together, these results suggest that, as observed in mammalian cultured neurons, Cdk5 increases the levels of tau phosphorylated tau at Ser262 through Par-1 in Drosophila.
Overexpression of Cdk5 enhances tau-mediated axon degeneration through tau phosphorylation at Ser262
We next examined whether Cdk5 enhances tau toxicity through tau phosphorylation at Ser262 in Drosophila models. Expression of human tau causes age-dependent and progressive neurodegeneration in the lamina, the first synaptic neuropil of the optic lobe containing photoreceptor axons (33). Degeneration in the lamina is undetectable or very mild in 3-day-old flies, while it is prominent in 10-day-old flies (33). Co-expression of Cdk5 significantly enhanced tau-mediated neurodegeneration in the lamina (Fig. 6A; compare tau and tau+Cdk5). By contrast, Cdk5 overexpression by itself did not cause neurodegeneration (Fig. 6A; compare control and Cdk5), suggesting that Cdk5 activity enhances neurodegeneration caused by tau toxicity.
Cdk5 enhances tau toxicity via tau phosphorylation at Ser262 in the fly retina. (A) Overexpression of Cdk5 enhanced tau toxicity in Drosophila retina. (Top) Lamina of flies with driver-alone control (control), flies expressing tau (tau) or flies co-expressing tau and cdk5 (tau+cdk5). Neurodegeneration is observed as vacuoles, indicated by arrows. (Bottom) Quantification of vacuole area. Mean ± SEM; n = 6–10. ***P < 0.005 between tau and tau+Cdk5 (the one-way ANOVA and Tukey post hoc test). (B) Augmentation of tau toxicity by Cdk5 was not observed in the expression of S2A (S262/356A) tau. (Bottom) Quantification of vacuole area. Mean ± SEM; n = 6–10. n.s., P > 0.05 (one-way ANOVA and Tukey post hoc test). Flies used for experiments were 10 days after eclosion.
We asked whether tau phosphorylation at MARK/Par-1 target site Ser262 is involved in the enhancement of tau toxicity caused by Cdk5. We found that Cdk5 failed to enhance neurodegeneration caused by tau with unphosphorylatable alanine substitutions at Ser262 and Ser356 (S2A) (Fig. 6B).
Taken together, these results suggest that Cdk5 enhances tau-mediated neurodegeneration by promoting tau phosphorylation at Ser262 in vivo.
Discussion
Cdk5 activates MARK4 in neurons and enhances tau toxicity
Dysregulation of MARK4 activity has been associated with Alzheimer’s disease; however, it was not fully understood how MARK4 activity increases in diseased brains. In this study, we report a novel mechanism by which tau hyperphosphorylation is induced via interaction between two tau kinases, Cdk5 and MARK4. Cdk5 enhances MARK4 activity via phosphorylation in the spacer domain (Figs 1–3), resulting in elevated levels of tau phosphorylated at direct Cdk5 target sites as well as those at the MARK target site, Ser262 (Figs 4 and 5). p25, whose increase is associated with neurodegenerative conditions, enhanced tau phosphorylation at both sites more than p35 (Fig. 4). Cdk5 exacerbated tau-induced retinal degeneration in Drosophila eyes, and this effect was dependent on tau phosphorylation at MARK target sites (Fig. 6). Our findings suggest a novel feed-forward mechanism of tau phosphorylation. Aberrant Cdk5 activity under pathological conditions not only directly phosphorylates tau but also causes activation of another tau kinase MARK/Par-1, promoting its toxicity (Fig. 7).
Working model of hyperphosphorylation of tau via activation of Cdk5 and MARK4. Under pathological conditions, Cdk5 is aberrantly activated, phosphorylates MARK4 in the spacer domain and increases its activity. MARK4 phosphorylates tau at Ser262, which act in the initial steps of tau mismetabolism and neurodegeneration.
MARK4 activity is regulated by Cdk5 phosphorylation via a novel mechanism
Phosphorylation of MARK4 at Thr214 in the activation loop (T-loop) of the catalytic domain by LKB1 is the best-known mechanism for activating MARK4 and its family members (37). In addition to Thr214, MARK4 has several potential phosphorylation sites. It is reported that phosphorylation by aPKC at Thr568 in the spacer domain activates MARK4 in vitro (43). However, the phosphorylation status of the spacer domain of MARK4 and their roles in MARK4 activity in neurons have not been elucidated. Our results demonstrate that MARK4 is phosphorylated at multiple sites in the brain and primary cultured neurons, and Cdk5 activity affects its phosphorylation profiles (Fig. 1). Since the putative phosphorylation sites in MARK4 are located in the (Ser/Thr)-Pro-X-(Lys/Arg) motif (44), which Cdk5 is known to preferentially phosphorylate, Cdk5 may directly phosphorylate MARK4. It is also possible that Cdk5 indirectly enhances phosphorylation at these sites via interaction with PKC or other yet-to-be-identified kinases (43). In any case, our results suggest that Cdk5 enhances MARK4 activity via a novel mechanism.
Interaction between Cdk5 and MARK4 in health and disease
In summary, our results indicate that Cdk5 triggers upregulation of MARK4 activity leading to tau pathology and neurodegeneration. In addition to the pathogenesis of neurodegenerative diseases, Cdk5 and MARK function in common paradigms such as cell polarity formation and cell migration (45,46). More recently, both Cdk5 and MARK have been suggested as therapeutic targets for cancer (47,48) and diabetes (49–51). Therefore, the Cdk5-MARK axis, which we demonstrate here, may function in various diseases in addition to the pathogenesis of tauopathy. Further studies on this axis will shed light on the regulatory mechanisms of MARK4 and its dysregulation in pathological contexts.
Materials and Methods
Chemicals and antibodies
An anti-MARK4 antibody (Cell Signaling Technology, Danvers, MA), anti-MYC antibody (4A6; Merck Millipore, Billerica, MA), anti-tau Tau5 antibody (Merck Millipore), Anti-tau (T46; Thermo, Chelmsford, MA), anti-phospho-Ser235 tau antibody (Abcam, Cambridge, England), anti-phospho-Ser404 tau antibody (Abcam), anti-phospho-Ser262 tau antibody (Abcam) and anti-phospho-Ser356 tau antibody (Abcam), anti-Cdk5 antibody (DC17; Santa Cruz Biotechnology, Santa Cruz, CA) and anti-p35 antibody (C19; Santa Cruz Biotechnology) and anti-actin antibody (SIGMA, St. Louis, MO) were purchased. Anti-tau antibody (TauC) was a kind gift from Dr A. Takashima (Gakushuin University). The Cdk5 inhibitor roscovitine was purchased from Merck Millipore. Alkaline phosphatase was purchased from Wako chemicals (Osaka, Japan).
Plasmid construction
Human MARK4 (752 amino acids) fused to MYC-tag at the C-terminal was subcloned into pcDNA3. Human MARK47A-myc was chemically synthesized by Eurofins Genomics (Ebersberg, Germany) and subcloned into pcDNA3. Kinase-dead MARK4 (Asp199Ala) was generated polymerase chain reaction by PCR-based site-directed mutagenesis.
Cell culture and transfection
HEK293 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Sigma) supplemented with 10% (v/v) fetal bovine serum, 100 U/ml penicillin and 0.1 mg/ml streptomycin. Plasmids encoding MARK4, mouse Cdk5, p35 Cdk5 activator or human 2N4R tau was transfected with Lipofectamine 2000 (Thermo Fisher) according to the manufacturer’s protocol. Primary neurons were prepared from the mouse brain cortex at embryonic day 15 (E15) and plated on poly-L-lysine-coated dishes in DMEM, and Ham’s F-12 (1:1) supplemented with 5% fetal bovine serum, 5% horse serum, 100 U/ml penicillin and 0.1 mg/ml streptomycin at a density of 1.0 × 106 cells/ml. The medium was changed to Neurobasal medium supplemented with 2% B-27 (Invitrogen, Carlsbad, CA), 0.5 mm L-glutamine, 100 U/ml penicillin and 0.1 mg/ml streptomycin after 4 h of plating. Primary neurons at 4DIV were treated with 50 μm roscovitine for 3 h.
All animal experiments were performed according to the guidelines for animal experimentation of Tokyo Metropolitan University. The study was approved by the Research Ethics and Safety Committee of Tokyo Metropolitan University (approval number, A30-1).
In vitro kinase assay of MARK4 and in vitro phosphorylation
MARK4 expressed in HEK293 cells was immunoprecipitated from the cell lysate with monoclonal anti-Myc antibody (4A6) and Dynabeads protein G (Thermo Fisher). Its kinase activity was measured using Chktide (SignalChem, Richmond, BC) and [γ-32P] ATP as substrates. Incorporation of 32P into Chktide was quantified using liquid scintillation counter (Beckman, Brea, CA).
For in vitro phosphorylation assay of MARK4 by Cdk5, Cdk5-25 was purified from Sf9 cells infected baculovirus encoding Cdk5 and p25. Incorporation of 32P into MARK4 was detected by autoradiogram using FLA 7000 image analyzer (GE Healthcare, Chicago, IL).
SDS-PAGE, Phos-tag SDS-PAGE and immunoblotting
Laemmli’s SDS-PAGE was performed using 12.5% (w/v) polyacrylamide gels for Cdk5, p35 and actin, 10% for MARK4 and 9% for tau. Phos-tag SDS-PAGE was performed using 7.5% (w/v) polyacrylamide gels containing 50 μm Phos-tag acrylamide (Wako Chemicals) for tau, 5% polyacrylamide gel and 25 μm Phos-tag acrylamide for MARK4 expressed in HEK293 cells and 5% polyacrylamide gel and 5 μm Phos-tag acrylamide for MARK4 in neurons. Proteins separated in the gel were transferred to a PVDF membrane (Merck Millipore) using a submerged blotting apparatus and then visualized using Immobilon Western Chemiluminescent HRP Substrate (Millipore). The chemiluminescent signal was detected by Fusion FX (Vilber, Collégien, France) and intensity was quantified using ImageJ (NIH). Western blots were repeated a minimum of three times with different animals, and representative blots are shown. Flies used for western blotting were 2-day old after eclosion.
Fly stocks and husbandry
Flies were maintained in standard cornmeal media at 25°C under light–dark cycles of 12:12 h. The transgenic fly line carrying the human 0N4R tau, UAS-S2Atau and UAS-luciferase RNAi were reported previously (30,31,33). UAS-Cdk5 is a kind gift from Dr Edward Giniger (National Institute of Neurological Disorders and Stroke, NIH). UAS-Par-1 RNAi was a kind gift from Dr Bingwei Lu (Stanford University). UAS-Cdk5 RNAi (8203R-1) was obtained from the NIG Drosophila Stock Center. Genotypes of the flies used in the experiments are described in Supplementary Material, Table S1.
Histological analysis
Neurodegeneration in the fly retina was analyzed as previously reported (52). Fly heads were fixed in Bouin's fixative for 48 h at room temperature, incubated for 24 h in 50 mm Tris/150 mm NaCl and embedded in paraffin. Serial sections (7 μm thickness) through the entire heads were stained with hematoxylin and eosin and examined by bright-field microscopy. Images of the sections that include the lamina were captured with Keyence microscope BZ-X700 (Keyence, Osaka, Japan), and vacuole area was measured using ImageJ (NIH). Heads from more than three flies (more than five hemispheres) were analyzed for each genotype.
Statistics
Statistics were done with Microsoft Excel (Microsoft) and R (R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/). Differences were assessed using the Student’s t-test or one-way analysis of variance (ANOVA) and Tukey's honestly significant difference post hoc test. P values <0.05 were considered statistically significant.
qRT-PCR
Quantitative reverse transcription PCR was carried out as previously reported (53). More than 30 flies for each genotype were collected and frozen. Heads were mechanically isolated, and total RNA was extracted using Isogen Reagent (NipponGene, Tokyo, Japan) according to the manufacturer’s protocol with an additional centrifugation step (11 000g for 5 min) to remove cuticle membranes prior to the addition of chloroform. Total RNA was reverse-transcribed using PrimeScript Master Mix (Takara Bio, Shiga, Japan). qRT-PCR was performed using TOYOBO THUNDERBIRD SYBR qPCR Mix (Osaka, Japan) on a Thermal Cycler Dice Real Time System (Takara Bio). The average threshold cycle value (CT) was calculated from at least three replicates per sample. Expression of genes of interest was standardized relative to rp49. Relative expression values were determined by the ΔΔCT method. Primers were designed using Primer-Blast (NIH).
The following primers were used for RT-PCR: Drosophila Cdk5, forward 5′-AAGATCTTCCGTGTGCTGGG -3′, reverse 5′- GAGGTGATGGCCGGAAAAGA -3′, Drosophila rp49, 5′-GCTAAGCTGTCGCACAAATG-3′, reverse 5′- GTTCGATCCGTAACCGATGT-3.
Acknowledgements
For fly stocks, we thank Drs Edward Ginier and Bingwei Lu; TRiP at Harvard Medical School (NIH/NIGMS R01-GM084947); the Bloomington Stock Center; and NIG Drosophila stock center. We thank Dr Akihiko Takashima for the TauC antibody. We thank T. Miyashita for technical assistance. We thank Dr S.-I. Hisanaga for critical comments.
Conflict of Interest statement. None declared.
Funding
JSPS KAKENHI Grant-in-Aid for Scientific Research on Innovative Areas (Brain Protein Aging and Dementia Control) (15H01564 and 17H05703 to K.A.); JSPS KAKENHI Grants-in-Aid for Scientific Research (15K06712 to K.A.); Hoan-sha Foundation (to K.A.); the Takeda Science Foundation (to K.A.); Research Funding for Longevity Sciences (28-26 and 19-7) from the National Center for Geriatrics and Gerontology (NCGG), Japan (to K.M.I.); JSPS KAKENHI Grants-in-Aid for Scientific Research (16K08637 to K.M.I.); Takeda Science Foundation, Japan (to K.M.I.)
