VRK3 promotes nuclear localization of HSP70 induced by glutamate

We next examined the relationship between HSP70 and VRK3. The interaction of these two proteins confirmed by mass spectrometry analysis on FLAG-VRK3- expressing 293FT cell lysates immunoprecipitated with an antibody against anti-FLAG (Fig. 2a). To identify the VRK3-binding region in HSP70, we expressed HA-tagged full length HSP70 or its fragments along with FLAG-VRK3 in SH-SY5Y cells (Fig. 2b). Although it has a different electrophoretic mobility than that seen in the input, HA-tagged full length HSP70 retrieved in the immunoprecipitation. The HSP70 F2 fragment, which contains the substrate binding domain, was found to interact with VRK3 (Fig. 2b). For the interaction to occur, nuclear translocation of HSP70 is necessary, because VRK3 is mainly localized in the nucleus due to its nuclear localization signal (NLS)³⁴. HSPs are induced by various stresses^6,7 and especially, in response to heat shock stress, HSP70 are rapidly expressed and transiently relocated from cytoplasm into the nucleus⁶. To verify whether treatment with excessive glutamate induces nuclear translocation of HSP70, we tested cytoplasmic and nucleoplasmic fractions of non-treated and glutamate-treated SH-SY5Y cells (Fig. 2c). Nuclear HSP70 was markedly increased in cells treated with glutamate compared with control. In addition, co-localization of HA-tagged VRK3 and endogenous HSP70 was detected in glutamate-treated SH-SY5Y cells (Fig. 2d). The amount of nuclear HSP70 was increased in cells overexpressing VRK3. To clarify whether VRK3 binding affects nuclear localization of HSP70, we analyzed cytoplasmic and nucleoplasmic fractions of glutamate-treated SH-SY5Y cells (Fig. 2e, Supplementary Fig. S1). VRK3 promoted nuclear localization of glutamate-induced HSP70 in dose-dependent manner (Fig. 2e), compared with control which showed no differences in levels of nuclear HSP70 between the groups (Supplementary Fig. S1). In contrast, cells reducing VRK3 expression using siRNAs showed markedly suppressed nuclear localization of glutamate-induced HSP70 (Fig. 2f,g). We also tested the effects of overexpression or knockdown of VHR on nuclear localization of HSP70 in glutamate-treated SH-SY5Y cells (Supplementary Fig. S2a,b). There were no significant differences in levels of nuclear HSP70 between the groups. Together, these findings reveal that VRK3 facilitates glutamate-induced nuclear localization of HSP70.

Open in a separate window

Figure 2

VRK3 promotes nuclear localization of HSP70 in glutamate-treated SH-SY5Y cells.

(a) HSP70 bound to FLAG-tagged VRK3 in HEK2FT cells. (b) FLAG-tagged VRK3 bound to the HA-tagged substrate binding domain of HSP70 with high affinity. (c) Nuclear localization of endogenous HSP70 was increased in glutamate-treated SH-SY5Y cells. (d) HA-tagged VRK3 and endogenous HSP70 were co-localized in glutamate-treated SH-SY5Y cells. The dotted lines represent the boundary between nucleus and cytoplasm; White arrowheads highlight cells expressed relatively low level of HA-VRK3. Scale bar, 20μm. (e) Overexpression of FLAG-tagged VRK3 facilitated nuclear localization of HSP70 in dose-dependent manner. (f) Inhibition of glutamate-induced nuclear localization of HSP70 in cells with VRK3 knockdown. (g) Quantification of the levels of nuclear HSP70 as in (f). Values are normalized to control and shown as mean±s.d., n=3. Student’s t-tests, **P<0.01. (h) Glutamate-induced nuclear localization of HSP70 was suppressed by pre-treatment with importazole, a nuclear transport receptor importin-β inhibitor. (i) Nuclear localization of HSP70 facilitated by FLAG-tagged VRK3 was also suppressed by importazole treatment.

Regulation of the nucleocytoplasmic protein transport is mediated by members of the importin β family³⁵. Importin β family carriers interact with their specific cargos and facilitates transfer to the nucleus through the nuclear pore complex (NPC) upon binding³⁵. To elucidate the mechanism of enhanced nuclear localization of HSP70 by VRK3, we pre-treated cells with importazole, an inhibitor of importin-β transport receptors, before glutamate treatment and then analyzed cytoplasmic and nucleoplasmic fractions of SH-SY5Y cells (Fig. 2h,i). Pre-treatment of the cells with importazole resulted in suppression of glutamate-induced HSP70 nuclear translocation, even though the expression of endogenous VRK3 was increased (Fig. 2h). Consistent with previous results, VRK3 overexpression promoted nuclear localization of HSP70, however, pre-treatment with importazole resulting in reduced nuclear HSP70 (Fig. 2i). Taken together, this data indicates that VRK3 supports nuclear retention of glutamate-induced HSP70 rather than assisting in its translocation to the nucleus.

Glutamate-induced active ERK stimulates the gene expression of both HSP70 and VRK3, and inhibits its sustained activity that triggers cell death

Although ERK is associated with cell survival, it is also linked to cell death. ERK 1/2 determines cell fate based on its abundance, duration of activation, and subcellular distribution^23,24. Several studies have focused on the detrimental role of persistent ERK activation under stress^36,37,38. Furthermore, prolonged ERK activation is implicated in glutamate-induced cell death³⁷.

Previous studies have shown that ERK activation induced by phorbol 12-myristate 13-acetate increases VRK3 expression³². Before investigating the physiological significance of both HSP70 and VRK3 in the regulation of ERK activation, we explored whether glutamate-induced ERK activity affects the expression of HSP70 and VRK3. In agreement with previous findings, treatment with glutamate stimulated ERK activation (Fig. 3a). Increased ERK phosphorylation reached a peak 1–2h after glutamate treatment and returned to basal levels within 6h. Moreover, we detected positive correlation between ERK activity and the protein levels of HSP70 and VRK3. To confirm an association between the activation status of ERK and the levels of HSP70 and VRK3 proteins, we pretreated SH-SY5Y cells with PD98059, a specific MEK/ERK inhibitor, before glutamate treatment (Fig. 3b). Blocking ERK phosphorylation resulted in a delayed increase in both HSP70 and VRK3 proteins upon glutamate exposure. To clarify whether the elevation of HSP70 and VRK3 protein levels followed by ERK activation is due to upregulation of endogenous mRNA levels, we performed quantitative real-time RT-PCR using SH-SY5Y cells pretreated with either vehicle or PD98059 before glutamate exposure (Fig. 3c,d). Pharmacological inhibition of ERK activation delayed mRNA expression of both HSP70 and VRK3 following glutamate treatment compared with control. After initial peak, there were no significant differences in gene expression of both HSP70 and VRK3 between the groups. These results indicate that glutamate-induced active ERK stimulates HSP70 and VRK3 expression at the transcriptional level.

Open in a separate window

Figure 3

Glutamate-induced ERK activation stimulates the expression of both HSP70 and VRK3 to suppress its sustained activity that causes cell death.

(a) The protein levels of HSP70 and VRK3 were positively correlated with ERK activity. (b) Blocking ERK phosphorylation using PD98059, a specific MEK/ERK inhibitor, delayed an increase in both HSP70 and VRK3 protein upon glutamate exposure. (c,d) Inhibition of ERK activation delayed mRNA expression of HSP70 (c) and VRK3 (d) following glutamate treatment. The mRNA levels of endogenous genes were detected by quantitative real-time PCR. Values are normalized to RPL32 and shown as mean±s.d., n=3. Student’s t-tests, **P<0.01, ***P<0.001. (e) Nuclear translocation of active ERK and increased levels of both endogenous HSP70 and VRK3 were observed in glutamate-treated SH-SY5Y cells. (f) HSP70-mediated downregulation of p-ERK prevented glutamate-induced apoptosis. White arrowheads highlight cleaved caspase-3-positive cells. Scale bar, 20μm. (g) Glutamate-induced apoptosis was suppressed in HSP70-overexpressing SH-SY5Y cells. White arrows show HSP70-overexpressing cells. White arrowheads highlight cleaved caspase-3-positive cells. Scale bar, 20μm.

Upon sustained activation, ERK 1/2 translocates to the nucleus and regulates several transcription factors, resulting in the expression of pro-apoptotic proteins²⁵. Our results showed that glutamate-induced HSP70 is relocated from cytoplasm into the nucleus. To examine whether translocated HSP70 regulates nuclear ERK through the enhancement of VHR phosphatase activity, we analyzed the nucleoplasmic fraction of glutamate-treated SH-SY5Y cells (Fig. 3e). In agreement with previous findings, excessive active ERK stimulated by glutamate translocated into the nucleus, and active nuclear ERK levels were increased for up to 2h. The levels of both endogenous HSP70 and VRK3 were also increased in the nucleus after glutamate treatment. Results from immunocytochemistry using glutamate-treated SH-SY5Y cells expressing HA-tagged HSP70 revealed that overexpression of HSP70 leads to decrease of glutamate-induced activation of ERK (Fig. 3f). Cells showed prolonged ERK activation followed by increased level of cleaved caspase 3, an active form of caspase 3 that triggers apoptosis by degrading cytoskeletal and nuclear proteins^39,40. The cleavage of caspase 3 was markedly suppressed in cells overexpressing HSP70 (Fig. 3g), indicating that HSP70 inhibits neuronal cell death via prevention of glutamate-induced prolonged ERK activation in the nucleus. Taken together, these data indicate that glutamate-induced active ERK transcriptionally upregulates the expression of HSP70 and VRK3, and suppresses its prolonged activity that causes cell death.

VRK3-mediated nuclear localization of HSP70 is important for suppression of glutamate-induced prolonged ERK activation and subsequent cell death

To ascertain the involvement of HSP70 and VRK3 in reducing glutamate-induced persistent ERK activation, we examined glutamate-treated SH-SY5Y cells after downregulation of either VRK3 or HSP70 expression by siRNA (Fig. 4a–c). ERK activation in SH-SY5Y cells with either HSP70 or VRK3 knockdown was sustained for at least 6h after glutamate exposure compared with control (Fig. 4a,b). Interestingly, sustained ERK activation led to increased expression of VRK3 in HSP70-downregulated SH-SY5Y cells, indicating the existence of a compensatory mechanism between HSP70 and VRK3 in the regulation of ERK activation. In addition, VRK3-downregulating cells which showed low levels of active ERK compared with cells with HSP70 knockdown were more susceptible to glutamate excitotoxicity-induced cell death, suggesting that VRK3 has another crucial role for survival.

Open in a separate window

Figure 4

VRK3-mediated nuclear localization of HSP70 is crucial for inhibition of glutamate-induced prolonged ERK activation and subsequent cell death.

(a) Glutamate-induced persistent ERK activation in SH-SY5Y cells after downregulation of either VRK3 or HSP70. (b) Quantification of the levels of p-ERK versus total ERK as in (a). Values are normalized to GAPDH and shown as mean±s.d., n=3. Student’s t-tests, **P<0.01. (c) Reducing expression of either HSP70 or VRK3 rendered cells vulnerable to glutamate-induced apoptosis. Representative images of TUNEL-positive cells. Scale bar, 20μm. (d) Suppression of glutamate-induced ERK activation depended on VRK3-mediated nuclear localization of HSP70. (e) HSP70 fused to NLS decreased glutamate-induced ERK activation more efficiently than HSP70 alone. (f) HSP70 nuclear localization was important for the downregulation of glutamate-induced ERK activation. Scale bar, 20μm. (g) Quantification of p-ERK intensity as in (f). Values are normalized to control and shown as mean±s.d., n≥30 for each sample. Student’s t-tests, **P<0.01. (h) HSP70 fused to NLS prevented glutamate-induced apoptosis more effectively that HSP70 alone. Representative images of TUNEL-positive cells. Scale bar, 20μm.

We conducted TUNEL assay to verify whether the suppression of prolonged ERK activation by HSP70 and VRK3 attenuates glutamate-induced cell death. Downregulation of either HSP70 or VRK3 increased the number of TUNEL-positive cells in glutamate-treated SH-SY5Y cells (Fig. 4c). These results indicate that HSP70 and VRK3 inhibit glutamate-induced prolonged ERK activation that triggers neuronal cell death.

To confirm whether nuclear localization of HSP70 is required to decrease glutamate-induced ERK activation, we examined cytoplasmic and nucleoplasmic fractions of SH-SY5Y cells (Fig. 4d). Levels of active ERK were markedly reduced after increased nuclear localization of HSP70 along with VRK3. We created a construct capable of redirecting HSP70 to the nucleus through the fusion of three tandem repeats of NLS peptides of SV40 large T antigen (PKKKRKV)⁴¹ to the C-terminus of HSP70 (Fig. 4e). HSP70 fused to NLS decreased glutamate-induced ERK activation more efficiently than HSP70 alone. These results were further supported by immunofluorescence analysis (Fig. 4f,g). Notably, when HSP70 was highly concentrated in the nucleus, activate ERK in both the cytoplasm and nucleus was decreased. Moreover, TUNEL staining showed that HSP70 fused to NLS inhibits glutamate-induced apoptosis more effectively than HSP70 alone (Fig. 4h). Taken together, these results indicate that VRK3-mediated nuclear localization of HSP70 is critical for suppressing glutamate-induced persistent ERK activation and excitotoxicity-evoked neuronal cell death.

To further determine whether the interaction between the HSP70 F2 fragment and VRK3 is sufficient for HSP70 nuclear localization and subsequent downregulation of glutamate-induced ERK activation, we examined the nucleoplasmic fraction of glutamate-treated SH-SY5Y cells (Supplementary Fig. S3). Nuclear localization of HSP70 F2 fragment was increased along with VRK3. However, the F2 fragment did not have an effect on glutamate-evoked ERK activation. This observation is consistent with our finding that the HSP70 F2 fragment had almost no binding affinity towards VHR.

Neuroprotective role of HSP70 and VRK3 against glutamate excitotoxicity in mouse cortical neurons

To determine whether HSP70 and VRK3 affect the regulation of nuclear ERK in mouse cortical neurons treated with glutamate, cortical neurons obtained at postnatal day 0.5 were purified and cultured for 5 days. Cultured cells were considered neurons based on the presence of MAP2- and NeuN-positive projections (Fig. 5a). In agreement with previous findings glutamate-evoked ERK activation was detected in mouse cortical neurons (Fig. 5b). In addition, glutamate treatment significantly increased the number of cleaved caspase-3-positive neurons (Fig. 5c,d), indicating that glutamate-induced excitotoxicity causes persistent ERK activation that triggers neuronal cell death.

Open in a separate window

Figure 5

Neuroprotective role of HSP70 and VRK3 against glutamate in mouse cortical neurons.

(a) Primary cultured mouse cortical neurons expressed MAP2 and NeuN. Scale bar, 20μm. (b) Glutamate induced ERK activation in mouse cortical neurons. (c) Glutamate triggered apoptosis of mouse cortical neurons. Representative images of cleaved caspase-3-positive cells and NeuN-stained cell nuclei (blue). Scale bar, 20μm. (d) Quantification of the percentage of cleaved caspase-3-positive neurons treated in (c). Values are normalized to untreated controls and shown as mean±s.d., n=3. Student’s t-test, **P<0.01. (e) Overexpression of HSP70 downregulated glutamate-induced persistent ERK activation. Scale bar, 20μm. (f) Quantification of p-ERK intensity in mouse cortical neurons expressing HA (mock) or HA-HSP70 as in (e). Values are normalized to HA-controls, and shown as mean±s.d., n≥30 for each sample. Student’s t-test, *P<0.05. (g) VRK3-deficient mice exhibited excessive ERK activation induced by glutamate treatment. (h) VRK3-deficient mice showed increased sensitivity to glutamate-induced apoptosis. Values are normalized to untreated controls and shown as mean±s.d., n=3. Student’s t-test, **P<0.01, ***P<0.001. (i) Nuclear localization of HSP70 by either VRK3 or fused NLS is critical for suppressing glutamate-induced persistent ERK activation. Scale bar, 20μm. (j) Quantification of p-ERK intensity in VRK3-deficient neurons expressing HA (mock), HA-VRK3, HA-HSP70, or HA-HSP70-3NLS as in (i). Values are normalized to HA-controls, and shown as mean±s.d., n≥30 for each sample. Student’s t-test, n.s., not significant, ***P<0.001.

Consistent with previous results, overexpression of HSP70 attenuated persistent ERK activation induced by glutamate in mouse cortical neurons (Fig. 5e,f). To directly examine the functional importance of VRK3 in vivo, we have used VRK3-deficient mice generated by targeted gene disruption in embryonic stem cells (Fig. 5g–j). Despite induction of HSP70, VRK3-deficient neurons showed sustained ERK activation (Fig. 5g) and markedly decreased cell viability relative to control (Fig. 5h), indicating that both HSP70 and VRK3 are essential for maintaining neuronal viability against glutamate-evoked persistent ERK activation. The importance of nuclear localization of HSP70 in the regulation of glutamate-induced ERK activation and the role of VRK3 as a facilitator of HSP70 nuclear localization were further supported by immunofluorescence analysis of VRK3-deficient neurons (Fig. 5i,j). Neurons overexpressing VRK3 exhibited markedly reduced ERK activation compared with control. Moreover, HSP70 fused to NLS decreased glutamate-induced ERK activation more efficiently than HSP70 alone, indicating that VRK3-mediated HSP70 nuclear localization and subsequent downregulation of ERK activation are required to protect neurons from glutamate-evoked excitotoxicity.

Protein extraction from human brain tissue

Snap-frozen post-mortem human brain tissue was ground and homogenized in Triton lysis buffer (20mM Tris, 150mM NaCl, 1mM EDTA, 1mM EGTA, 1% Triton-X, 2.5mM sodium pyrophosphate, 1mM β-glycerophosphate, 1mM Na₃VO₄) supplemented with protease inhibitors. Samples were sonicated and centrifuged at 15,000rpm for 30min at 4°C. The supernatant was removed, and the protein concentration was determined using Bradford Reagent (AMRESCO). Samples were denatured for 5min at 95°C with SDS sample buffer containing β-mercaptoethanol. Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. Western blots were quantified by densitometry using ImageJ software.

Immunohistochemistry analysis of human brain tissue

Prefrontal cortical brain sections (4μm thickness) were cut at regular intervals on glass slides. Paraffin-embedded tissue sections were deparaffinized in xylene and then rehydrated with decreasing concentrations of ethanol and placed in water. Sections underwent antigen retrieval using citrate buffer (10mM sodium citrate, pH 6.0) and then were washed and blocked with PBS containing 3% bovine serum albumin and 0.2% Triton X-100 for 1h at room temperature. Primary antibodies were diluted in blocking solution. After overnight incubation, sections were washed three times with PBS. Secondary antibodies coupled to Alexa fluorophores were diluted in blocking solution. After 1h of incubation in the dark, sections were washed three times with PBS. For counterstaining, sections were further incubated with Hoechst (2μg/ml) for 10min at room temperature and then washed three times with PBS. Slices were mounted and visualized by fluorescence microscopy.

Mice

Mice deficient for VRK3 was generated by targeted gene disruption in embryonic stem cells. Specific pathogen-free (SPF) VRK3 knockout mice were bred and maintained on a C57BL/6 background. All animals were maintained on food and water ad libitum with 12-hr light-dark cycle. Approval of the study protocol was obtained from the Pohang University of Science and Technology Institutional Animal Care and Use Committee (POSTECH IACUC). All animal experiments were carried out according to the provisions of the Animal Welfare Act, PHS Animal Welfare Policy, and the principles of the NIH Guide for the Care and Use of Laboratory Animals. All mice were maintained under conventional conditions at the POSTECH animal facility under institutional guidelines.

Primary neuron cultures and transfection

Cortices were dissected from P0.5 newborn pups of wildtype and VRK3 knockout mice. Cultured cortical neurons were seeded onto a poly-L-lysine-coated culture dishes and were maintained in Neurobasal Medium supplemented with GlutaMAX, B27 supplements, and penicillin-streptomycin. Neuronal transfections were performed with Lipofectamine 2000 at 3 days in vitro (DIV) according to the manufacturer’s instructions.

Antibodies

The following monoclonal antibodies were purchased: anti-HSP70, anti-p-ERK, and anti-α-synuclein (Abcam); anti-GAPDH (Santa Cruz Biotechnology); anti-VHR (BD Biosciences); anti-HA (Roche); anti-Aβ (BioLegend); anti-NeuN (Merck Millipore); anti-HSP70, anti-GST, anti-cleaved caspase-3, anti-Bak, and anti-Bax (Cell Signaling Technology). The following polyclonal antibodies were also obtained: anti-PSD95 (Abcam); anti-HSP70, anti-VRK3, anti-VHR, anti-p-ERK, anti-ERK, and anti-His (Cell Signaling Technology); anti-VRK3 (Atlas antibodies); anti-HA (YPYDVPDYA; Bethyl Laboratories); anti-FLAG (D-8), anti-Lamin B, and anti-MAP2 (I-18) (Santa Cruz Biotechnology).

Plasmid construction

The coding regions of VRK3, HSP70, and VHR were amplified by RT-PCR from the mRNA of NIH3T3, SH-SY5Y, and HEK293A cells and cloned into pFLAG-CMV2, pcDNA3.1-HA, pGEX4T-1, pGEX4T-3, pEGFP-N1, or pProEX-Hta vectors. Partial fragments of HSP70 F1 (amino acids 1-380), F2 (amino acids 381-537), and F3 (amino acids 538-642) were cloned into pcDNA3.1-HA and pGEX4T-3 vectors. For nuclear expression of HSP70, three tandem repeats of NLS peptides of SV40 large T antigen (PKKKRKV)⁴¹ were fused to the C-terminus.

Cell culture, transfection, and RNA interference

SH-SY5Y cells were grown in minimal essential medium (MEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 100U/ml each penicillin G and streptomycin. Transient transfections were performed by electroporation using a Microporator MP-100 (Digital Bio) or Neon-Transfection System (Invitrogen). Scrambled siRNA (5′-CCU ACG CCA CCA AUU UCG U(dTdT)-3′) was purchased from Upstate Biotechnology (GE Healthcare, Dharmacon). VRK3 siRNA (5′-GGA CAA AUU GCC UUC CCA A (dTdT)-3′). HSP70 siRNA (5′-GUU UGA GGG CAU CGA CUU CU (dTdT)-3′) and VHR siRNA (5′-GGU CCU UCA UGC ACG UCA A (dTdT)-3′) were purchased from Bioneer.

Inhibitors and glutamate treatment

The transport receptor importin-β inhibitor importazole (SML0341, Sigma-Aldrich) and the MEK/ERK inhibitor PD98059 (#9900, Cell Signaling Technology) were dissolved in dimethyl sulphoxide (DMSO) to allow addition to cell cultures at a final concentration of 0.1% (v/v) DMSO. For inhibition of importin-β, SH-SY5Y cells were pre-treated with 40μM importazole for 1h before glutamate treatment. For inhibition of ERK activation, SH-SY5Y cells were pre-treated with 50μM PD98059 for 1h. L-glutamic acid was dissolved in distilled water. SH-SY5Y cells were treated with 20mM glutamate for 6h. Cultures of mouse primary cortical neurons were treated with 40μM glutamate for 24h.

RNA isolation and cDNA synthesis

SH-SY5Y cells were lysed with TRI-Solution (Bio Science Technology) and the total RNA was extracted after addition of 0.2 volume of chloroform. Samples were mixed with vortex, incubated at room temperature for 10min, and centrifuged at 12,000g for 15min at 4°C. Recovery of total RNA was then done by precipitation with 0.5 volume of isopropanol. After 10min incubation at room temperature, samples were centrifuged at 12,000g for 8min at 4°C. The RNA pellet was washed with 75% (v/v) ethanol, briefly air-dried, and dissolved in diethylpyrocarbonate (DEPC)-treated water by incubating for 5min at 65°C. Yield and purity of RNA was determined by NanoDrop 2000 spectrophotometer (Thermo Scientific). Following quantification, 1μg of each total RNA sample was reverse transcribed using oligo-dT and the ImProm-II Reverse Transcription System (Promega) according to the manufacturer’s instructions. For qRT-PCR analysis, each cDNA sample was diluted 5 times with nuclease free water.

Design and optimization of RT-qPCR primers

Gene-specific qPCR primers were designed using Primer3 (version 0.4.0) and Primer-BLAST (NCBI). Primer specificities were confirmed with the melting-curve after amplification by RT-qPCR. Amplicon sizes were checked by 2% agarose gel electrophoresis and ethidium bromide staining. The specificity of the primers was further confirmed by amplicon sequencing. The sequences of the forward and reverse primers were as follows: endogenous hHSP70, 5′-TGGAGTCCTACGCCTTCAAC-3′ and 5′-GATGGGGTTACACACCTGCT-3′; hVRK3, 5′-CTATTGCCCAAGTGGCAAAC-3′ and 5′-CTCAGTGTTGGGAAGGCAAT-3′; human ribosomal protein L32 (hRPL32), 5′-AACCCAGAGGCATTGACAAC-3′ and 5′-GTTGCACATCAGCAGCACTT-3′.

Quantitative real-time PCR

The mRNA levels of endogenous genes were detected by quantitative real-time PCR using a StepOnePlus Real-Time PCR System (Applied Biosystems) with the FastStart Universal SYBR Green Master (Roche Applied Science). A 20μl of reaction cocktail was constituted of 3μl diluted cDNA, 10μl 2X SYBR Green Master Mix and 0.5μl each of the forward and reverse primers. The following amplification program was used: polymerase activation at 95°C for 10min; 40 repeated cycles of 95°C for 15s and 60°C for 1min. A comparative C_t method was used for quantification.

Phosphatase activity assay

For measuring the in vitro phosphatase activity of VHR using the phosphatase substrate p-nitrophenyl phosphate (pNPP), 5μg of GST-VHR and 10μg of His-HSP70 were incubated in reaction buffer (50mM Tris at pH 7.5, 5mM MgCl₂, 0.5mM DTT, and 150mM KCl) for 30min at 30°C. Purified proteins were further incubated with 5×30mM of pNPP (P4744, Sigma-Aldrich) dissolved in recommended buffer (1M diethanolamine at pH 10.4 with 0.5mM MgCl₂) for 2h at 37°C. The hydrolysis of the substrate was measured at 405nm using an ELISA reader with Infinite 200 Pro NanoQuant (TECAN). For measuring the in vitro phosphatase activity of VHR using recombinant p-ERK2, 1μg of GST-VHR and 2μg of His-HSP70 were incubated in reaction buffer (50mM Tris at pH 7.5, 5mM MgCl₂, 0.5mM DTT, and 150mM KCl) for 30min at 30°C. Purified proteins were further incubated with 50ng recombinant p-ERK2 (Merck Millipore) for 1h at 37°C.

Image processing

General image processing were performed by using Adobe Photoshop 7.0 (Adobe Systems Inc.).

We are grateful to Dr. C.H. Park and Mr. D.H. Lee who produced and maintained the VRK3 knockout mice, respectively. We also would like to thank Mr. H.J. Kim, Mr. H.G. Ryu, and Mr. K.H. Kwon for their technical assistance for conducting RT-qPCR. We acknowledge the Netherlands Brain Bank for providing post-mortem human brain tissue samples. This work was supported by the “Cooperative Research Program for Agriculture Science & Technology Development (Project No. PJ01121602)” from Rural Development Administration, BK21 Plus funded by the Ministry of Education [10Z20130012243], and POSCO (AP-TP, Green science), Korea.