DAPK1 ‐Mediated Parkin Inactivation Enhances Neurotoxicity via MITOL ‐Dependent Degradation

FLAG‐tagged mammalian expression plasmids carrying the cDNA of human DAPK1 in its wild‐type form (FLAG‐DAPK1‐WT) or a kinase‐deficient mutant (FLAG‐DAPK1‐K42A, also termed FLAG‐DAPK1‐KD) were generously provided by T.H. Lee (Fujian Medical University, Fujian, China). A pcDNA3.1 construct expressing Myc‐tagged human parkin (pcDNA3.1‐Myc‐Parkin) was kindly obtained from K. Tanaka (Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan). Bacterial vectors encoding GST‐fused N‐terminal parkin, either wild type or carrying point mutations (S131A, S136A, or S198A), were generated as previously described [12]. The double‐mutant version (S136A/S198A) was generated by site‐directed mutagenesis of pcDNA3.1‐Myc‐Parkin. Truncated fragments of parkin (Myc‐parkin‐1–225 and Myc‐parkin‐226–465) were amplified with PrimeSTAR HS DNA polymerase (TAKARA, Shiga, Japan) and subsequently inserted into the pRK5‐Myc vector. Constructs expressing HA‐tagged ubiquitin mutants, in which either lysine 48 (pRK‐HA‐Ub‐K48) or lysine 63 (pRK‐HA‐Ub‐K63) was preserved while all remaining lysine residues were substituted with arginine, were obtained from Addgene (Cambridge, MA, USA). FLAG‐tagged human MITOL expression plasmids, including both the wild‐type protein (MITOL‐WT) and a catalytically inactive form (MITOL‐MT), were provided by S. Hirose (Tokyo Institute of Technology, Tokyo, Japan). siRNAs specific for DAPK1 and a scrambled negative control were obtained from IDT Korea (Hanam‐si, Gyeonggi‐do, Korea). The sequences of the DAPK1‐specific siRNA duplex were as follows: sense, 5′‐GGUGAGGCGUGACAGUUUAUCAUGA (dTdT)‐3′; antisense, 5′‐AACCACUCCGCACUGUCAAAUAGUACU (dTdT)‐3′.

Human embryonic kidney 293 (HEK293) cells were grown in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/mL penicillin–streptomycin. Mouse embryonic fibroblasts (MEFs) obtained from either wild‐type (DAPK1+/+) or knockout (DAPK1−/−) mice were generously provided by T.H. Lee (Fujian Medical University, Fujian, China) and cultured in Dulbecco's Medium containing 10% FBS and antibiotics (1% penicillin—streptomycin). Mouse neuroblastoma MN9D cells were maintained in high‐glucose DMEM with 10% FBS and 1% penicillin—streptomycin on dishes precoated with poly‐D‐lysine (25 μg/mL; Sigma‐Aldrich), under atmospheric conditions of 90% air and 10% CO₂. Unless specified otherwise, all cell lines were propagated at 37°C in a humidified incubator with 5% CO₂. Transient transfections were performed using either Lipofectamine 2000 or polyethyleneimine (PEI) following the manufacturers' standard instructions.

HEK293 cells ( Homo sapiens , female, embryonic kidney; RRID:CVCL_0045), MN9D cells ( Mus musculus , male, embryonic midbrain dopaminergic neurons; RRID:CVCL_M067) and wild‐type (DAPK1+/+) or DAPK1‐deficient (DAPK1−/−) mouse embryonic fibroblasts (MEFs, Mus musculus , embryonic fibroblasts) were used in this study. MEFs were generated in‐house from mouse embryos and therefore do not have an assigned RRID. All cell lines were obtained either from accredited repositories, such as ATCC and KCLB, or were kindly provided by collaborating laboratories acknowledged in this manuscript. Although in‐house authentication using STR or SNP profiling was not performed, all cell lines were obtained from institutions authorised to conduct such validations. According to the Cellosaurus and ICLAC databases, none of the cell lines used in this study have been reported as misidentified or cross‐contaminated. Routine mycoplasma testing was conducted using the e‐MYCO Mycoplasma Detection Kit ver. 2.0 (iNtRON Biotechnology, Seongnam‐si, Gyeonggi‐do, Korea), and all cell lines consistently tested negative throughout the experimental period. All experiments were conducted between October 2022 and June 2025, using cells at passages between 5 and 10.

Cells were washed with phosphate‐buffered saline, collected by scraping and lysed in buffer (50 mM Tris, 10% glycerol, 1% Nonidet P‐40, 150 mM NaCl, 1 mM EGTA, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 0.2 mM phenylmethylsulfonyl fluoride and 1 μg/mL aprotinin). The lysates were sonicated and then cleared by centrifugation at 13,000 × g for 15 min at 4°C. For immunoprecipitation, 500–1000 μg of protein was mixed with 1 μg of the appropriate antibody and incubated overnight at 4°C with constant rotation. Protein‐A Sepharose beads were subsequently added and allowed to bind for 2 h at 4°C. After several washes with lysis buffer, the immune complexes were released by heating in 2× SDS sample buffer. Samples were then subjected to SDS‐PAGE, and proteins were transferred onto nitrocellulose membranes (Whatman, GE Healthcare Life Sciences). Membranes were first incubated at room temperature for 1 h in TBST buffer containing 0.1% Tween 20, 137 mM NaCl and 20 mM Tris (pH 7.5) supplemented with 5% nonfat milk to block nonspecific binding. After blocking, membranes were incubated overnight at 4°C with the designated primary antibody. After the membranes were rinsed, they were incubated with horseradish peroxidase–conjugated secondary antibodies for 2 h, and protein expression was visualised via enhanced chemiluminescence, as recommended by the manufacturer.

MN9D neuroblastoma cells were plated onto coverslips pretreated with poly‐D‐lysine. Following two rinses with PBS, the cells were incubated in 3.7% formaldehyde at room temperature for 10 min, then permeabilised using 0.1% Triton X‐100 for an additional 10 min. Blocking was then performed for 1 h in TBST containing 1% BSA. The cells were then incubated with rabbit polyclonal anti‐FLAG and/or mouse monoclonal anti‐Myc antibodies, followed by Alexa Fluor 488– or 594–labelled secondary antibodies. Fluorescent microscopy images were collected on a Zeiss LSM 880 system (Carl Zeiss, Germany) and analysed afterward using the LSM Image Browser tool.

Brain tissues were isolated from 6‐week‐old male C57BL/6 mice (Orient Bio, Seongnam‐si, Gyeonggi‐do, Korea) and disrupted in lysis buffer (150 mM NaCl, 0.5% sodium deoxycholate, 50 mM Tris–HCl, 0.1% SDS, 1% Triton×‐100 and a protease inhibitor). The homogenates were sonicated and then centrifuged at 4°C for 20 min at 13,000 × g, after which the clarified supernatants were obtained and used for downstream analyses.

Cells were rinsed twice with ice‐cold PBS and lysed in a buffer containing 150 mM NaCl, 10 mM Tris–HCl (pH 7.4), 1% Triton X‐100, 10% glycerol, 20 mM N‐ethylmaleimide and protease inhibitors. The lysates were centrifuged at 4°C, 15,000 × g for 20 min and the supernatants were collected as the Triton X‐100–soluble fraction. The resulting pellets were rinsed, suspended in lysis buffer with 4% SDS and incubated at elevated temperature for 30 min to solubilise proteins. Both fractions were subjected to SDS‐PAGE and subsequently analysed by immunoblotting.

E. coli BL21 cells carrying GST–parkin plasmids were grown until the culture reached an OD₆₀₀ of approximately 0.7–0.8. Protein expression was initiated by adding 0.5 mM IPTG (Sigma‐Aldrich), and cultures were maintained at 37°C for 24 h. Bacterial pellets were collected and disrupted on ice by sonication in a buffer composed of 50 mM Tris–HCl (pH 7.4), 200 mM NaCl, 1 mM EDTA, 1 mM DTT, 0.1% Triton X‐100 and a protease inhibitor cocktail. The lysates were cleared by centrifugation at 12,000 × g for 20 min, and the supernatant fraction was incubated with glutathione Sepharose 4B beads (GE Healthcare Life Sciences) overnight at 4°C. After thorough washing of the beads, bound GST–parkin proteins were released using a buffer containing 10 mM reduced glutathione and 50 mM Tris–HCl (pH 7.4).

MN9D cells were transfected for 24 h with either FLAG‐DAPK1‐WT or FLAG‐DAPK1‐KD plasmids. Cell lysates prepared in NP‐40 buffer were subjected to immunoprecipitation using anti‐FLAG antibody overnight at 4°C, followed by capture with protein A–Sepharose beads for 2 h. After washing with lysis buffer, the beads were further washed using kinase buffer composed of 50 μM DTT, 20 mM MgCl₂ and 40 mM Tris–HCl (pH 7.5). The resulting immune complexes were incubated with 2 μg of recombinant GST–parkin in kinase buffer supplemented with 10 μM ATP. Kinase activity was initiated by the addition of 10 μCi [γ‐³²P] ATP, and the reactions were allowed to proceed at 30°C for 30 min. Reactions were stopped by adding SDS sample buffer, and phosphorylation of substrates was examined by SDS‐PAGE followed by autoradiography.

Cells transfected with DNA or siRNA were treated with or without 6‐hydroxydopamine (6‐OHDA) for 24 h. The CCK‐8 assay (Dojindo Laboratories, Kumamoto, Japan) was performed to evaluate cell viability. Culture medium was replaced with a 1:10 dilution of CCK‐8 reagent in complete medium and incubated for 30 min at 37°C. Using a microplate reader, the absorbance at 450 nm was recorded.

To explore structural changes in parkin, we employed the AlphaFold3 prediction server (https://alphafoldserver.com/about↗). The full‐length parkin sequence was submitted with serine 136 modified to phospho‐serine. The server generated five highly similar structural models, and the model considered most representative was subsequently examined using PyMOL (The PyMOL Molecular Graphics System, Version 3.0, Schrödinger LLC).

Data were analysed using one‐way analysis of variance (ANOVA). All quantitative data represent the mean ± standard deviation (SD) from three independent biological replicates (n = 3). Graphs were generated using GraphPad Prism 5.0 (GraphPad Software Inc.). Densitometric quantification of Western blot bands was performed using ImageJ software (version 1.53).

Recent studies identified DAPK1 as a regulator of synucleinopathy and dopaminergic neuron loss in a PD mouse model [17]. To explore its relationship with parkin, we compared their expression individually and in combination. While DAPK1 levels remained stable, parkin levels significantly decreased upon co‐expression, suggesting DAPK1 may promote parkin degradation (Figure 1A). This reduction was reversed by proteasome inhibition with MG132 (Figure 1B). Co‐IP in HEK293 cells confirmed a specific DAPK1‐parkin interaction with or without MG132 (Figure S1 and Figure 1B). Endogenous interactions were also detected in MN9D cells (Figure 1C) and mouse brain lysates (Figure 1D), with immunostaining showing colocalisation of DAPK1 and parkin in MN9D cells (Figure 1E). A GST pull‐down assay using purified recombinant proteins confirmed a direct biochemical interaction (Figure 1F). Together, these results demonstrate that DAPK1 physically interacts with parkin in mammalian cells and may regulate its stability through the proteasome pathway.

We next examined the functional relationship between DAPK1 and parkin. While parkin did not affect DAPK1 levels, DAPK1 reduced parkin expression in a dose‐dependent manner (Figure), suggesting DAPK1 promotes parkin degradation, possibly involving phosphorylation, given its kinase function. To assess the role of DAPK1's kinase activity, MN9D cells were transfected with Myc‐parkin and either wild‐type DAPK1 (DAPK1‐WT) or a kinase‐defective mutant (DAPK1‐KD). Parkin levels were significantly reduced by DAPK1‐WT but not by DAPK1‐KD, indicating that DAPK1's kinase activity is essential for parkin degradation (Figure). Consistent with this, endogenous parkin levels were elevated in DAPK1‐knockout MEFs, while reintroduction of DAPK1 restored the reduction in parkin expression (Figure). Cycloheximide (CHX) chase assays further revealed that DAPK1‐WT markedly accelerated the degradation of parkin protein compared to control, as evidenced by a significant reduction in parkin levels over time (Figure). In contrast, the kinase‐inactive DAPK1‐KD mutant failed to alter parkin turnover, and the half‐life of parkin remained comparable to that of the control group (Figure). These findings confirm that DAPK1 promotes parkin degradation in a kinase activity‐dependent manner by reducing its protein stability. S2A S2B S2C S2D S2E

Given that eukaryotic cells degrade proteins via the ubiquitin‐proteasome system (UPS) or lysosome‐mediated autophagy [19]. We investigated the pathway responsible for DAPK1‐mediated parkin degradation. As shown in Figure 1B, proteasome inhibition by MG132 reversed the DAPK1‐induced reduction in parkin. To confirm this, MN9D cells co‐expressing DAPK1 and parkin were treated with MG132, rapamycin or NH₄Cl. Only MG132 significantly restored parkin levels (Figure 2A), and similar effects were observed with epoxomicin, another proteasome inhibitor (Figure 2B), indicating UPS involvement.

We next examined whether DAPK1 promotes parkin ubiquitination. Cell‐based ubiquitination assays showed that DAPK1‐WT, but not the kinase‐dead DAPK1‐KD, enhanced parkin ubiquitination, confirming kinase‐dependent regulation (Figure 2C). Since K48‐linked polyubiquitin chains on target proteins are primarily associated with proteasomal degradation, whereas K63‐linked mono‐ or polyubiquitin chains do not promote degradation but instead alter the biochemical properties of proteins, leading to different cellular outcomes, we aimed to distinguish which type of ubiquitination is involved in this process. To identify the type of ubiquitin linkage, cells were transfected with ubiquitin mutants allowing only K48‐ or K63‐linked chains. Co‐IP revealed that DAPK1 specifically promoted K48‐linked, but not K63‐linked, polyubiquitination of parkin (Figure 2D), consistent with proteasomal targeting.

Collectively, these results demonstrate that DAPK1 facilitates parkin degradation by promoting its K48‐linked polyubiquitination and subsequent proteasomal degradation via a kinase‐dependent mechanism.

Given that DAPK1‐mediated parkin degradation requires its kinase activity, we investigated whether DAPK1 directly phosphorylates parkin. Phos‐tag immunoblot analysis revealed parkin phosphorylation in the presence of DAPK1‐WT, but not with the kinase‐defective DAPK1‐KD, indicating a kinase‐dependent modification (Figure 3A). An in vitro kinase assay using purified recombinant proteins confirmed that parkin is phosphorylated by DAPK1‐WT, but not DAPK1‐KD, supporting a direct phosphorylation event (Figure 3B).

To map the phosphorylation site(s), two parkin truncation mutants (1–225 and 226–465) were co‐expressed with DAPK1‐WT in MN9D cells. Phospho‐serine immunoblotting showed phosphorylation in parkin‐WT and the N‐terminal 1–225 mutant, but not in the C‐terminal 226–465 fragment, indicating the site resides within the N‐terminal region (Figure 3C).

To further narrow down the precise phosphorylation sites within the parkin‐1‐225 region, three well‐characterised serine residues on parkin were selected for mapping. These included two serine residues (S131 and S136), known targets of other kinases in parkin, as well as S198, a potential phosphorylation site inferred from analysis of DAPK1 substrates. In vitro kinase assays revealed that S136A and S198A mutants displayed significantly reduced phosphorylation, while S131A remained unaffected (Figure 3D). Notably, the double mutant S136A/S198A exhibited a complete loss of phosphorylation, while S136A alone showed minimal phosphorylation compared to the relatively higher levels in S198A (Figure 3E). These findings suggest that S136 is the primary phosphorylation site, with S198 as a secondary site.

Finally, Phos‐tag analysis confirmed robust phosphorylation in parkin‐WT and S131A, but not in the S136A/S198A mutant (Figure 3F), reinforcing that S136 and S198 are critical DAPK1‐targeted residues. Collectively, these results demonstrate that DAPK1 directly phosphorylates parkin at S136 and S198, providing mechanistic insight into how DAPK1 regulates parkin stability and function.

To determine whether DAPK1‐mediated parkin degradation is driven by phosphorylation at S136/S198, we used a double phosphorylation‐resistant mutant (parkin‐S136A/S198A; parkin‐2A). In MN9D cells, co‐expression of DAPK1‐WT significantly reduced wild‐type parkin levels, but not parkin‐2A levels, indicating that phosphorylation at these residues is essential for DAPK1‐induced degradation (Figure 4A). Similarly, DAPK1 enhanced polyubiquitination of parkin‐WT but not parkin‐2A, confirming the requirement of phosphorylation for ubiquitination (Figure 4B). A phosphorylation‐mimetic mutant (parkin‐S136E/S198E; parkin‐2E) exhibited increased polyubiquitination relative to wild type, further supporting that phosphorylation at these sites promotes ubiquitination and degradation (Figure 4C).

To assess whether this degradation depends on parkin's own E3 ligase activity, we examined a dominant‐negative mutant lacking the UBL domain (parkin‐ΔUBL), essential for its ubiquitination function [20]. DAPK1 still reduced parkin‐ΔUBL levels, indicating that degradation is independent of parkin's auto‐ubiquitination and instead driven by DAPK1‐mediated phosphorylation (Figure 4D,E).

Together, these findings demonstrate that DAPK1 phosphorylates parkin at S136/S198, facilitating its polyubiquitination and proteasomal degradation via a mechanism likely involving other UPS components.

Given that DAPK1‐mediated phosphorylation leads to parkin degradation, we hypothesised that DAPK1 may impair parkin's neuroprotective role under oxidative stress, contributing to neuronal vulnerability in PD [21]. To test this, MN9D cells were treated with various PD‐related neurotoxins. Among them, only 6‐OHDA significantly reduced parkin protein levels, suggesting selective sensitivity (Figure S3A). DAPK1 overexpression further enhanced 6‐OHDA‐induced parkin reduction, indicating a synergistic effect (Figure S3B). This effect was further examined using parkin‐2A and parkin‐2E mutants. Notably, parkin‐2A was resistant to 6‐OHDA‐induced degradation, while parkin‐2E showed enhanced reduction (Figure S3C), confirming the critical role of S136/S198 phosphorylation in this process.

Endogenous parkin levels were also decreased by 6‐OHDA treatment (Figure S3D). However, DAPK1‐knockdown using siRNA partially rescued this reduction, demonstrating that DAPK1 is necessary for 6‐OHDA‐induced parkin degradation (Figure S3E). Since 6‐OHDA induces dopaminergic cell death through oxidative stress and is widely used in PD models [22], these findings suggest that DAPK1 contributes to PD pathology by facilitating parkin degradation under oxidative stress. Thus, DAPK1 may act as a crucial mediator linking ROS‐induced damage to parkin inactivation in dopaminergic neurons.

To identify the UPS component responsible for DAPK1‐mediated parkin degradation, we focused on MITOL (also known as MARCH5), a mitochondrial E3 ligase previously reported to interact with parkin under mitophagic conditions [23]. We hypothesised that DAPK1‐induced phosphorylation may facilitate MITOL‐dependent parkin degradation. Co‐IP analysis revealed that under basal conditions, parkin and MITOL did not interact; however, DAPK1 overexpression significantly enhanced their binding (Figure 5A). To examine whether this interaction depends on parkin phosphorylation, we compared MITOL binding to wild‐type parkin, the phospho‐mimetic parkin‐2E and the phosphorylation‐resistant parkin‐2A mutant. Notably, MITOL interacted strongly with parkin‐2E but not with parkin‐2A or wild‐type parkin, indicating that DAPK1‐mediated phosphorylation enhances parkin‐MITOL association (Figure 5B).

Given MITOL's mitochondrial localisation, we tested whether phosphorylation promotes parkin translocation to mitochondria. Subcellular fractionation experiments demonstrated that parkin‐2E showed a 2.2‐fold increase in mitochondrial localisation compared to parkin‐WT and parkin‐2A (Figure 5C). This observation was further validated using immunostaining, where parkin localisation was analysed using an antibody against parkin and MitoTracker, a selective mitochondrial dye; notably, parkin‐2E displayed more than a six‐fold increase in colocalisation with mitochondria than the other variants (Figure 5D).

Collectively, these findings demonstrate that DAPK1‐catalysed phosphorylation of parkin facilitates its translocation to mitochondria, where it interacts with MITOL, leading to proteasomal degradation. This identifies MITOL as a likely mediator of DAPK1‐dependent parkin turnover and highlights a potential regulatory mechanism linking kinase activity, subcellular localisation and protein stability.

Parkin is a well‐established neuroprotective protein that defends dopaminergic neurons against a variety of toxic insults, including MPTP, 6‐OHDA and rotenone [23]. To assess whether phosphorylation by DAPK1 diminishes parkin's protective capacity, we investigated its involvement in 6‐OHDA‐induced cytotoxicity using MN9D cells. Treatment with 6‐OHDA progressively reduced cell viability, and exposure to 100 μM resulted in nearly 50% cell loss, as determined by CCK‐8 analysis (Figure 6A). Overexpression of Myc‐parkin‐WT significantly rescued cell viability, whereas co‐expression of FLAG‐DAPK1‐WT reversed this protective effect. In contrast, the kinase‐dead mutant DAPK1‐KD did not affect viability, suggesting that DAPK1's kinase activity is essential for suppressing parkin‐mediated protection (Figure 6A). Consistently, knockdown of DAPK1 enhanced parkin‐dependent cell survival under 6‐OHDA stress (Figure 6B).

To determine whether parkin phosphorylation is required for this effect, we compared the impact of DAPK1 on cells expressing either parkin‐WT or parkin‐2A mutant. Notably, the reduced neuroprotective effect of parkin observed upon DAPK1 overexpression was restored in cells expressing parkin‐2A, supporting a phosphorylation‐dependent mechanism (Figure 6C).

Finally, we evaluated the role of MITOL in this process. Overexpression of the phospho‐mimetic parkin‐2E markedly increased susceptibility to 6‐OHDA‐induced toxicity, indicating a loss of its neuroprotective function. This effect was further exacerbated by co‐expression of MITOL‐WT, consistent with its role in promoting parkin degradation. In contrast, the catalytically inactive MITOL mutant (MITOL‐MT), which carries C65S and C68S substitutions that abolish its E3 ligase activity, failed to enhance 6‐OHDA‐induced toxicity, highlighting the requirement of MITOL's enzymatic function for mediating parkin degradation and the resulting neurotoxicity (Figure 6D). These results suggest that DAPK1‐mediated phosphorylation of parkin promotes its MITOL‐dependent degradation, leading to loss of neuroprotection and increased vulnerability to neurotoxic stress.

Together, these findings reveal a novel mechanism by which DAPK1 impairs parkin's protective role, contributing to dopaminergic neurodegeneration in PD models.

The data that support the findings of this study are available from the corresponding author upon reasonable request.