TY - JOUR
AU - Seol, Wongi
AB - Abstract Leucine-rich repeat kinase 2 (LRRK2) has been identified as a causative gene for Parkinson’s disease (PD). LRRK2 contains a kinase and a GTPase domain, both of which provide critical intracellular signal-transduction functions. We showed previously that Rab5b, a small GTPase protein that regulates the motility and fusion of early endosomes, interacts with LRRK2 and co-regulates synaptic vesicle endocytosis. Using recombinant proteins, we show here that LRRK2 phosphorylates Rab5b at its Thr6 residue in in vitro kinase assays with mass spectrophotometry analysis. Phosphorylation of Rab5b by LRRK2 on the threonine residue was confirmed by western analysis using cells stably expressing LRRK2 G2019S. The phosphomimetic T6D mutant exhibited stronger GTPase activity than that of the wild-type Rab5b. In addition, phosphorylation of Rab5b by LRRK2 also exhibited GTPase activity stronger than that of the unphosphorylated Rab5b protein. Two assays testing Rab5’s activity, neurite outgrowth analysis and epidermal growth factor receptor degradation assays, showed that Rab5b T6D exhibited phenotypes that were expected to be observed in the inactive Rab5b, including longer neurite length and less degradation of EGFR. These results suggest that LRRK2 kinase activity functions as a Rab5b GTPase activating protein and thus, negatively regulates Rab5b signalling. GTPase activity, kinase substrate, LRRK2, Parkinson’s disease, Rab5 Parkinson’s disease (PD) is the second most common neurodegenerative disorder. Approximately 10% of PD cases occur as inherited forms (dominant and recessive). Samples from these patients with familial PD have been used to map >18 PD-associated (PARK) loci (1). Leucine-rich repeat kinase 2 (LRRK2) is encoded by a gene corresponding to PARK8, a locus causally linked to an autosomal-dominant form of PD (2, 3). Several LRRK2 missense mutations (e.g. G2019S, R1441C/G/H and Y1699C) have been identified in familial PD cases (2–4). Interestingly, LRRK2 mutations have also been observed in 1% of sporadic PD cases (5), suggesting that understanding the normal and pathogenic functions of LRRK2 may be important for elucidating the pathogenesis of PD. LRRK2 is a unique protein as it contains two distinct functional domains such as a GTPase (Roc) domain and a kinase (mitogen-activated protein kinase kinase kinase-like) domain both of which play critical roles in signal transduction (2, 3, 6, 7). Over-expression of LRRK2 promotes protein aggregation and neuronal toxicity (7–9), increases the level of intracellular reactive oxygen species (10), shortens neurite length (11, 12) and affects autophagy (12, 13). Over-expression of G2019S, which exhibits elevated kinase activity and is the most prevalent LRRK2 pathogenic mutant, aggravates most of these phenotypes (7, 8, 11, 14), suggesting that LRRK2 kinase activity plays a major role causing these abnormal cellular conditions. Therefore, identifying LRRK2 kinase substrates and elucidating the physiological functions of these substrates should provide critical clues for understanding the aetiology of PD and developing PD therapeutics, although a recent report raised a question on the relationship between kinase activity and the pathogenicity of LRRK2 (15). To date, several reports have identified putative substrates of LRRK2 kinase activity, including moesin (14, 16), 4E-BP [the eukaryotic initiation factor 4E-binding protein (17)], β-tubulin (18), Snapin (19), ArfGAP1 (20, 21), tau (22), members of the mitogen-activated protein kinase kinase family (23), Akt1 (24) and ribosomal protein S15 (25), although most of these proteins were not clearly determined whether they are actual pathophysiological substrates (26, 27). The Roc domain of LRRK2 was originally reported to mediate dimerization of LRRK2 (28) and exhibit weak GTPase activity (29). However, a recent study showed that the LRRK2 Roc domain is maintained as a stable monomer in solution (30), making the dimer issue unclear. The PD-associated R1441C/G/H mutation, which maps to the GTPase Roc domain, reduces the GTPase activity of LRRK2 (30, 31). Rab5b is an isoform of Rab5, a member of the small G protein family. Rab5 regulates fusion and motility of early endosomes, and is a marker of the early endosome compartment (32). We reported previously that LRRK2 interacts with Rab5b and colocalizes with Rab5b in hippocampal synaptic vesicles (33). In addition, the misregulation of LRRK2 in primary cultured neurons causes a decrease in the endocytosis rate, which is restored by the coexpression of LRRK2 with wild-type (WT) or constitutively active Rab5b (Q79L), but not with an inactive Rab5b mutant (N133I) (33). Here, we investigated the functional implications of Rab5b and LRRK2 interactions, demonstrating that LRRK2 can phosphorylate Rab5b at its Thr6 residue. Importantly, the phosphomimetic mutant Rab5 protein, T6D, exhibited increased GTP hydrolysis activity and phenotypes similar to inactive N133I. Our data suggest that LRRK2 functions as a Rab5b GTPase activating protein (GAP), and negatively regulates Rab5b signalling by reducing the duration of the active Rab5b-GTP state. Materials and Methods Plasmids and other reagents Construction of human Rab5b WT, Q79L and N133I has been previously reported (33). YFP-C1 Rab5b Q79L fusion plasmids were obtained from Dr. W.D. Heo (KAIST, Korea) and YFP-Rab5b WT, T6A and T6D plasmids were constructed by in vitro site-directed mutagenesis using YFP-C1 Rab5b Q79L as the template. A gene encoding Rab5b WT has been cloned into pET28 (Novagen, Merck KGaA, Darmstadt, Germany) to express it as a recombinant fusion protein with a histidine-tag at its C-terminus and two additional amino acids in its N-terminus. The Rab5b mutants T2A, T6A and T2,6A in pET28 were prepared by in vitro site-directed mutagenesis using pET28-Rab5b WT as template. The proteins were expressed in the Escherichia coli BL21(DE3) strain and purified through affinity chromatography using a nickel resin (Invitrogen, Carlsbad, CA, USA). GST-tagged LRRK2 WT, G2019S, R1441C and D1994A recombinant proteins lacking the N-terminal 969 amino acids were purchased from Invitrogen. In vitro kinase assay GST-LRRK2 WT, G2019S, R1441C or D1994A recombinant proteins (30 ng, Invitrogen) were mixed with 2.5 µCi of [γ-32P]ATP (SBP-501, IZOTOP, Budapest, Hungary) in kinase assay buffer [20 mM Tris (pH 7.5), 0.02% Tween-20, 10 mM MgCl2, 1 mM EGTA, 5 mM β-glycerophosphate, 1 mM Na3VO4, 2 mM DTT] with histidine-tagged Rab5b WT or mutant proteins (1.5 µg) as indicated, and the final volume was adjusted to 15 µl. The reaction mixture was incubated at 30°C for 10 min and subjected to protein gel electrophoresis. The dried gels were analysed by autoradiography. Immunoprecipitation of Rab5b in lysates of HEK 293T cells stably expressing G2019S We used HEK293T cells stably expressing LRRK2 G2019S by doxycycline treatment (34) to investigate whether LRRK2 phosphorylates Rab5b in cells. The Flag-tagged LRRK2 G2019S protein was induced from HEK 293T stable cells (34) with 2 µM doxycylcine for 2 days. The cells were harvested and lysed in lysis buffer [20 mM Tris HCl (pH7.5), 100 mM NaCl, 1 mM EDTA, 0.1% Triton X-100, 5% glycerol, 1 mM DTT, 1× PIC, 1× PhosStop (Roche, Mannheim, Germany)]. The cell lysates were incubated on ice for 20 min, centrifuged at 14,000 × g for 20 min at 4°C, and the supernatant was used for further analysis. The lysates were incubated with anti-Rab5b (#sc-598, Santa-Cruz Biotechnology, Dallas, TX, USA) in the lysis buffer for 2 h at 4°C. The mixture was then incubated with protein A-agarose (Pierce, Rockford, IL, USA) for 2 h at 4°C. The purified complex was washed three times and subjected to sodium dodecyl sulphate-polyacrylamide gel electrophoresis and western blot analyses using the indicated antibodies [phospho-threonine-X-arginine (p-TXR): Cell Signaling Technology #2351S; LRRK2: Abcam #ab133474 and Rab5b: Santa Crutz # sc-598]. GTP binding and in vitro GTP hydrolysis assays For GTP binding assay, we used two different kinds of GTP-beads, GTP-agarose (G-9678, Sigma) and immobilized γ-amino-octyl-GTP-Sepharose (AC-106S, Jena Biosciences, Jena, Germany). The difference of two beads is where the beads were linked to. The former was linked to the ribose moiety and the latter to the γ-phosphate of GTP. This difference caused that, after GTP hydrolysis, Rab5 proteins bound to the former were remained to be bound to beads whereas the proteins bound to the latter were released from the beads. The beads were preincubated with BSA, washed and incubated with the purified recombinant Rab5b proteins in the G buffer [100 mM Tris/HCl (pH 7.5), 50 mM KCl, 0.1 mM DTT, 1 mM EDTA, 5 mM MgCl2, 1% Triton X-100 and 1× PIC]. Then, the beads were washed with the G buffer containing 0.1% Triton X-100, incubated with 1× Laemmeli sample buffer, and the GTP-bound proteins were subjected to polyacrylamide gel electrophoresis and western blot analysis with anti-Rab5b antibody. GTP hydrolysis activity was assayed in vitro by incubating bacterially purified Rab5b recombinant protein (200 ng) with 1.25 µCi of [α-32P]GTP (IZOTOP, FP-208) in 10 µl of GTPase assay buffer [20 mM HEPES (pH7.5), 2 mM MgCl2, 1 mM DTT, 0.005% BSA] at 25°C for 60 min. When indicated, 100 ng of GST-LRRK2 WT, G2019S, R1441C or D1994A (Invitrogen) was coincubated with 10 µM of cold ATP and 200 ng of Rab5b recombinant protein in the kinase assay buffer for the GTPase assay. To stop the reaction, 2 µl of 6× stop solution (0.6% SDS, 6 mM EDTA, 3 mM GTP, 3 mM GDP) was added. To separate the hydrolyzed radioactive GDP from GTP, 1 µl of each mixture was spotted onto the thin layer chromatography (TLC) plate (Sigma, St. Louis. MO. USA, #Z122882) and run in 0.5 M KH2PO4 (pH 3.4) for ∼60 min. The plates were dried and analysed by autoradiography using a Typhoon 9200 imager (GE Healthcare, Buckinghamshire, UK). The densities of the GTP and GDP spots on the radioimages were measured, and the amount of radioactive GDP was divided by the total radioactivity of GTP plus GDP to calculate each GTPase activity. The final value was presented as relative GTPase activity compared to the WT of Rab5b. Neurite assay The neurite assay was carried out using Neuro 2A cells treated with dibutyryl–cAMP (35). The Neuro2A cells were maintained in DMEM high glucose containing 10% foetal bovine serum (FBS) at 37°C with 5% CO2. To overexpress Rab5b proteins, plasmids expressing the indicated YFP-Rab5b fusion proteins or the control EGFP vector were transiently transfected using Lipofectamine LTX (Invitrogen). Dibutyryl–cAMP was added to the cells at 1 mM and the cells were cultured for 2 days. Then, the cells were fixed and stained with antibody against ß-tubulin III (Sigma), then Texas Red-conjugated goat anti-rabbit IgG (Cell Signaling Technology) to clearly visualize the neurites. The stained cells were incubated with DAPI and ProLong Gold Antifade Reagent (Invitrogen). More than five independent images of each sample were captured with a FLoid Cell Imaging Station (Invitrogen). We chose only well-separated, transfected (i.e. double-stained) cells whose neurites did not overlap with neurites of other cells for analysis. The longest neurite of each chosen cell was measured by a researcher who did not know the identity of the images using the i-Solution program (IMT i-Solution Inc., Vancouver, BC, Canada) and the data were analysed to quantify neurite outgrowth. Results are shown as mean ± standard error (SE) of relative values to the vector control. Epidermal growth factor receptor degradation assay To measure degradation of epidermal growth factor receptor (EGFR), we followed a protocol described previously (36) with modifications. HeLa cells were seeded at 1 × 105 cells/well in a 12-well plate and maintained with DMEM containing 10% FBS and 1 × Antibiotic–Antimycotic (Invitrogen) at 37°C with 5% CO2. To overexpress Rab5, plasmids containing the Flag-tagged Rab5b WT and mutant genes or the control pcDNA (Invitrogen) vector were transiently transfected using Lipofectamine LTX plus (Invitrogen). One day after the transfection, the media was changed to that without FBS and incubated for 2 h. Then, 100 ng/ml of EGF (10 µg/ml of EGF stock prepared by dissolving 0.2 mg of EGF (E9644 [Sigma]) with 0.1 ml of 10 mM acetic acid and 0.1% BSA and mixing with 19.9 ml of DMEM) was added to the medium. After incubating for the indicated times, the cells were washed with cold DMEM twice and cold PBS once and lysed with lysis buffer (2% Triton X-100, and 1 × PIC in PBS). Then, the cell lysates were centrifuged at 1,000 × g for 10 min and the supernatants were subjected to SDS–PAGE and western blot analysis with the indicated and EGFR (Thermo Scientific Pierce, Rockford, IL, USA, MA5‐13269) antibodies. Results LRRK2 phosphorylates Rab5b We reported previously that Rab5b interacts and colocalizes with LRRK2, and coregulates the endocytosis of synaptic vesicles (33). To test whether LRRK2 could phosphorylate Rab5b, we employed in vitro kinase assays using N-terminally truncated, GST-tagged forms of WT LRRK2 and various LRRK2 mutants and bacterially expressed the Rab5b recombinant proteins. LRRK2 WT, G2019S and R1441C, but not the kinase-dead D1994A, exhibited both transphosphorylation of Rab5b and autophosphorylation (Fig. 1A). As expected, LRRK2 G2019S showed more robust phosphorylation activity than that of LRRK2 WT or R1441C. The conserved LRRK2-phosphorylated motif, F/Y-X-T-X-R (the underlined T is the phosphorylation site), has been reported previously (37) and such sites in moesin (14), Snapin (19) and peroxiredoxin (38) were phosphorylated although LRRK2 has also been reported to phosphorylate serine or threonine other than the threonine in this motif (25, 39). Two threonines in Rab5b, one in the second position and the other in the sixth position, matched the motif. Therefore, we prepared Rab5b T2A, T6A and T2,6A mutant recombinant proteins in which either or both putative phosphorylation sites were mutated to alanine to prevent phosphorylation. The purified proteins were used in the in vitro kinase assay, and the results showed that phosphorylation of both T6A and T2,6A was almost abolished at similar levels, whereas T2A was still considerably phosphorylated (Fig. 1B). Fig. 1 View largeDownload slide (A) LRRK2 phosphorylates Rab5b in the in vitro kinase assay. Recombinant GST-LRRK2 WT (WT), G2019S (G), R1441C (R) or kinase dead D1994A (D) proteins were incubated with [γ-32P]ATP and the same amounts of recombinant Rab5b WT proteins which were expressed as histidine-tagged proteins in E. coli and purified. The mixture was subjected to SDS–PAGE and the resulting autoradiogram showed that all LRRK2 proteins except D1994A phosphorylated Rab5b. (B) GST-LRRK2 G2019S was subjected to the in vitro kinase assay with recombinant Rab5b WT, T2A, T6A or T2,6A proteins. An autoradiogram of autophosphorylated GST-LRRK2 and phosphorylated Rab5b is shown. Use of equal amounts of kinase assay substrates was confirmed by SDS–PAGE and Coomassie Blue staining (bottom panel). Phosphorylated Rab5b (p-Rab5b) and LRRK2 (p-LRRK2) are indicated by arrows. (C) LRRK2 phosphorylates Thr6 of human Rab5b. Top panel: mass spectrometric identification of the Rab5b phosphorylation site from the in vitro kinase reaction. Reacted Rab5b was subjected to in-solution tryptic digestion, and collision-induced dissociation (CID) spectra obtained from LC-tandem MS, respectively. Tandem MS of the precursor ions in the phosphorylated Rab5b peptide (amino acids 1–18; sequence MAMTSRSTARPNGQPQASK where ‘Tp’ indicates phosphorylated threonine and the first two amino acids were derived from the expression vector, pET28) at a m/z of 1,078.96179 Da. Black lines (top) indicate peptide cleavage. The b and y ions represent peptide N- and C-terminal fragment ions produced by CID in the mass spectrometer. Labels denote the neutral loss of phosphoric acid (H3PO4), H2O or NH4 after CID fragmentation. The 116 Da mass difference between the y12 and y12-H2O-P ions indicates a phosphorylation modification (the neutral loss of H3PO4 and H2O). Bottom panel: The human Rab5b amino acid sequence. The conserved phosphorylation candidate sites at the threonine second and sixth (bold) residues are indicated by numbers above the amino acids. Recovered peptides by mass spectrometry are indicated by a black box and the two first amino acids derived from the vector plasmid are shown as small letters. (D) LRRK2 phosphorylates Rab5b in HEK 293T cells stably expressing Flag-tagged LRRK2 G2019S at its threonine residue in the TXR motif. Top panel: Flag-tagged LRRK2 G2019S was induced by treating HEK 293T stable cells with 2 µM doxycycline and induction of G2019S (Flag-LRRK2) was confirmed. The induced and non-induced cell lysates were immunoprecipitated with anti-Rab5b antibody. The immunoprecipitates were subjected to western blot analysis using anti-p-TXR antibody, and 5% of the input proteins are shown. Bottom panel: Intensity of the band detected by the p-TXR antibody was analysed and displayed as a bar graph. The amount of Rab5b proteins with phosphorylated threonine was calculated as the amount of immunoprecipitated Rab5b containing phosphorylated threonine divided by the amount of total immunoprecipitated Rab5b. The experiment was performed four times. Quantitative analysis expressed as mean ± SEM and a representative result is shown. **P < 0.01, Student’s t test, two-tailed. Fig. 1 View largeDownload slide (A) LRRK2 phosphorylates Rab5b in the in vitro kinase assay. Recombinant GST-LRRK2 WT (WT), G2019S (G), R1441C (R) or kinase dead D1994A (D) proteins were incubated with [γ-32P]ATP and the same amounts of recombinant Rab5b WT proteins which were expressed as histidine-tagged proteins in E. coli and purified. The mixture was subjected to SDS–PAGE and the resulting autoradiogram showed that all LRRK2 proteins except D1994A phosphorylated Rab5b. (B) GST-LRRK2 G2019S was subjected to the in vitro kinase assay with recombinant Rab5b WT, T2A, T6A or T2,6A proteins. An autoradiogram of autophosphorylated GST-LRRK2 and phosphorylated Rab5b is shown. Use of equal amounts of kinase assay substrates was confirmed by SDS–PAGE and Coomassie Blue staining (bottom panel). Phosphorylated Rab5b (p-Rab5b) and LRRK2 (p-LRRK2) are indicated by arrows. (C) LRRK2 phosphorylates Thr6 of human Rab5b. Top panel: mass spectrometric identification of the Rab5b phosphorylation site from the in vitro kinase reaction. Reacted Rab5b was subjected to in-solution tryptic digestion, and collision-induced dissociation (CID) spectra obtained from LC-tandem MS, respectively. Tandem MS of the precursor ions in the phosphorylated Rab5b peptide (amino acids 1–18; sequence MAMTSRSTARPNGQPQASK where ‘Tp’ indicates phosphorylated threonine and the first two amino acids were derived from the expression vector, pET28) at a m/z of 1,078.96179 Da. Black lines (top) indicate peptide cleavage. The b and y ions represent peptide N- and C-terminal fragment ions produced by CID in the mass spectrometer. Labels denote the neutral loss of phosphoric acid (H3PO4), H2O or NH4 after CID fragmentation. The 116 Da mass difference between the y12 and y12-H2O-P ions indicates a phosphorylation modification (the neutral loss of H3PO4 and H2O). Bottom panel: The human Rab5b amino acid sequence. The conserved phosphorylation candidate sites at the threonine second and sixth (bold) residues are indicated by numbers above the amino acids. Recovered peptides by mass spectrometry are indicated by a black box and the two first amino acids derived from the vector plasmid are shown as small letters. (D) LRRK2 phosphorylates Rab5b in HEK 293T cells stably expressing Flag-tagged LRRK2 G2019S at its threonine residue in the TXR motif. Top panel: Flag-tagged LRRK2 G2019S was induced by treating HEK 293T stable cells with 2 µM doxycycline and induction of G2019S (Flag-LRRK2) was confirmed. The induced and non-induced cell lysates were immunoprecipitated with anti-Rab5b antibody. The immunoprecipitates were subjected to western blot analysis using anti-p-TXR antibody, and 5% of the input proteins are shown. Bottom panel: Intensity of the band detected by the p-TXR antibody was analysed and displayed as a bar graph. The amount of Rab5b proteins with phosphorylated threonine was calculated as the amount of immunoprecipitated Rab5b containing phosphorylated threonine divided by the amount of total immunoprecipitated Rab5b. The experiment was performed four times. Quantitative analysis expressed as mean ± SEM and a representative result is shown. **P < 0.01, Student’s t test, two-tailed. To clearly identify the Rab5b amino acid residue phosphorylated by LRRK2, we incubated Rab5b WT recombinant proteins with ATP in the presence or absence of the LRRK2 G2019S recombinant protein, and analysed both samples with LC-tandem mass spectrophotometry (MS). Both samples yielded exactly the same peptides covering 163 of the 215 amino acids, Thr6 was the only phosphorylated residue in the sample incubated with G2019S, and none of the residues in the sample incubated without G2019S was phosphorylated (Fig. 1C). To confirm intracellular phosphorylation of Rab5b by LRRK2, we utilized HEK 293T cells stably expressing LRRK2 (34). Total Rab5b proteins of stable cells were immunoprecipitated with Rab5b antibody. Because isoforms of Rab5 proteins showed considerable sequence homology among them, we wanted to confirm specificity of Rab5b antibody. Therefore, the Rab5b immunoprecipitates were subjected to western blot analysis with other Rab antibodies such as Rab5a, 5c, 7 and 1a. Supplementary Fig. S1 shows that other Rab antibodies that we tested detected no band in the Rab5 immunoprecipitates although they specifically detected the corresponding Rab protein bands in the cell lysates (input in Supplementary Fig. S1). This result suggested that the Rab5b antibody used in this study was specific enough to detect Rab5b protein only (Supplementary Fig. S1). To test whether G2019S expression changes phosphorylation status of Rab5b, lysates of 293T cells stably expressing LRRK2 G2019S were immunoprecipitated and phosphorylated threonine was detected by western blot analysis with anti-p-TXR antibody. The densities of p-TXR bands in the Rab5b immunoprecipitated samples were normalized to the densities of immunoprecipitated total Rab5b protein. The result showed that doxycycline-induced expression of G2019S resulted in about 1.7-fold increased phospho-threonine in immunoprecipitated Rab5b compared with the non-induced cells, suggesting that LRRK2 phosphorylated the threonine site in Rab5b in cells (Fig. 1D). Rab5b phosphomimetic mutant T6D proteins exhibit enhanced GTPase activity We investigated the functional consequences of LRRK2-induced Rab5b phosphorylation. Because Rab5 is a GTPase protein, we compared GTP binding and GTP hydrolysis activities of phosphorylated or phosphomimetic Rab5b with those of non-phosphorylated recombinant Rab5b proteins. We first compared the GTP binding activity of Rab5b WT and mutant proteins using GTP-agarose. Both phosphomimetic Rab5b T6D and non-phosphorylatable T6A mutant exhibited GTP binding activity similar to that of the WT (Fig. 2A). Two well-characterized Rab5b mutant proteins, constitutively active Q79L and dominant negative N133I (40–43), were used as controls. The Q79L mutant proteins showed GTP binding activity whereas the N133I mutant proteins exhibited little binding activity as previously reported (41) (Fig. 2A). Fig. 2 View largeDownload slide The phosphomimetic Rab5b mutant proteins exhibit GTP hydrolysis activity stronger than that of the WT protein. (A) The GTP binding activity of Rab5 WT and mutant proteins using GTP-agarose. The indicated Rab5b proteins were incubated with GTP-agarose and, after extensive washing, each sample and 2.5% of the input proteins were subjected to western blot analysis with the anti-Rab5b antibody. The experiment was performed three times and a representative result is shown. (B) Phosphorylation of Rab5b by LRRK2 kinase enhances Rab5b GTP hydrolysis activity. GST-tagged LRRK2 WT (WT), G2019S (G), R1441C (R) and D1994A (D) proteins were coincubated with Rab5b for 60 min in the absence or presence of 10 µM ATP. GTPase activity was analysed by spotting the reaction mixture onto TLC plates and calculated as the amount of radioactive GDP was divided by the total radioactivities of GTP plus GDP. The experiment was performed three times. A representative TLC result and data expressed as the mean ± SEM are shown. The GTPase activity of Rab5b alone (−) in each set (− and + ATP) was set to 100% and the GTPase activities of other samples were normalized with GTPase activity of the Rab5b alone in the corresponding set. The statistical significance was shown. n.s.: not significant. (C) The GTP hydrolysis activity of Rab5b T6D protein was stronger than that of the WT. Recombinant Rab5b proteins were incubated with [α-32P]GTP and radiolabelled GTP and GDP were separated by TLC. The experiment was performed three times. A representative autoradiogram and the ratio of GTP hydrolysis activity which are expressed as the mean ± SEM are shown (**P < 0.01, ***P < 0.001). Fig. 2 View largeDownload slide The phosphomimetic Rab5b mutant proteins exhibit GTP hydrolysis activity stronger than that of the WT protein. (A) The GTP binding activity of Rab5 WT and mutant proteins using GTP-agarose. The indicated Rab5b proteins were incubated with GTP-agarose and, after extensive washing, each sample and 2.5% of the input proteins were subjected to western blot analysis with the anti-Rab5b antibody. The experiment was performed three times and a representative result is shown. (B) Phosphorylation of Rab5b by LRRK2 kinase enhances Rab5b GTP hydrolysis activity. GST-tagged LRRK2 WT (WT), G2019S (G), R1441C (R) and D1994A (D) proteins were coincubated with Rab5b for 60 min in the absence or presence of 10 µM ATP. GTPase activity was analysed by spotting the reaction mixture onto TLC plates and calculated as the amount of radioactive GDP was divided by the total radioactivities of GTP plus GDP. The experiment was performed three times. A representative TLC result and data expressed as the mean ± SEM are shown. The GTPase activity of Rab5b alone (−) in each set (− and + ATP) was set to 100% and the GTPase activities of other samples were normalized with GTPase activity of the Rab5b alone in the corresponding set. The statistical significance was shown. n.s.: not significant. (C) The GTP hydrolysis activity of Rab5b T6D protein was stronger than that of the WT. Recombinant Rab5b proteins were incubated with [α-32P]GTP and radiolabelled GTP and GDP were separated by TLC. The experiment was performed three times. A representative autoradiogram and the ratio of GTP hydrolysis activity which are expressed as the mean ± SEM are shown (**P < 0.01, ***P < 0.001). Next, we tested GTPase activities of the WT and phosphorylated Rab5b proteins. A simple GTP hydrolysis assay with radioactive GTP showed that recombinant Rab5b WT proteins could exhibit considerable GTPase activity without any Rab5 GAP (Supplementary Fig. S2A). However, Both Q79L proteins which are unable to hydrolyze GTP and N133I proteins which are unable to bind to GTP exhibited almost no GTP hydrolysis activity as expected (44) (Supplementary Fig. S2A). Because we wanted to perform phosphorylation of Rab5b by LRRK2 and GTPase assays of the phosphorylated Rab5b in one reaction mixture we tested whether Rab5b GTPase is functional in the kinase assay buffer. The result showed that Rab5b in the kinase assay buffer exhibited GTPase activity similar to that in the GTPase assay buffer (Supplementary Fig. S2B). Therefore, to maximize the effect of Rab5b phosphorylation and simplify the assay, we performed Rab5b GTPase assays in kinase assay buffer in one reaction tube containing LRRK2 and Rab5b with ATP so that Rab5b phosphorylated by LRRK2 could exhibit GTPase activity. The GTPase activity of Rab5b coincubated with LRRK2 WT, G2019S, R1441C or D1994A in the absence of ATP was not statistically significantly different from that of Rab5b alone (Fig. 2B). These results indicate that the interaction of LRRK2 with Rab5b has little effect on the GTPase activity of Rab5b. In contrast, coincubation of Rab5b with various LRRK2 proteins in the presence of ATP clearly increased the GTPase activity of Rab5b (1.2–1.4-fold) compared with that of Rab5b alone (Fig. 2B) in a kinase activity-dependent manner. Coincubation with LRRK2 G2019S produced the highest increase in Rab5b GTPase activity where that of LRRK2 R1441C and WT promoted moderate increases in GTPase activity (Fig. 2B), effects that correlated with relative strength of the kinase activity of each LRRK2 variant (7, 8, 45). To exclude the possibility that the increase of GTPase activity was caused by GTPase activity of LRRK2 itself, we also tested those of various LRRK2 proteins without Rab5b but with ATP coincubation. The result suggested that GTPase activities of various GST-LRRK2 proteins were very weak and almost undetectable in our condition (Supplementary Fig. S3). We also measured GTPase activities of phosphomimetic T6D and non-phosphorylatable T6A mutant proteins. The T6D proteins exhibited a significant 1.4-fold increase in GTPase activity compared with that of the WT proteins, but the T6A proteins showed the activity similar to that of the WT proteins (Fig. 2C), supporting the result of the GTPase assay with Rab5b WT proteins which were phosphorylated by LRRK2 (Fig. 2B). This result was also supported by the GTP binding assay using γ-amino-octyl-GTP-Sepharose in which the beads were attached to the γ-phosphate; thus, only the GTP-bound proteins even after GTP hydrolysis could be detected in this assay. The phosphomimetic T6D mutant exhibited weaker GTP binding activity than that of the WT (Supplementary Fig. S4). This result suggests that the phosphomimetic mutant catalyzed GTP hydrolysis faster than the WT and the resulting GDP-bound mutant proteins could have been released from the beads. The data altogether suggested that Rab5b phosphorylation by LRRK2 increases GTPase activity of Rab5b and reduces active form of Rab5b. Rab5b T6D proteins exhibit weaker signals than the WT for neurite outgrowth and EGFR degradation assays Next, we investigated whether the phosphomimetic mutant Rab5b T6D exhibited physiological phenotypes related to its weaker signal. Two Rab5b dominant negative mutants, S34N and N133I, which could not bind GTP, have been reported (46). These two mutants were reported to decrease the endocytosis rate (47), resulting in slow EGFR degradation after EGF stimulation (48), and to exhibit enhanced neurite outgrowth (42). First, we investigated the effect of T6D mutants on neurite outgrowth using Neuro 2A cells because their transfection efficiency is higher than other cells generally used for neurite analyses such as PC12 or MN9D cells (35, 49, 50). Our previous neurite assay with PC12 cells showed that expression of Rab5b Q79L or N133I significantly decreased or increased the neurite length upon stimulation of cells with neuronal growth factor (NGF) (50). In this study, we measured the length of the longest neurite in each Neuro 2A cell expressing the indicated YFP-Rab5 proteins. The result showed that the T6D, but not T6A mutants, enhanced neurite outgrowth in a similar manner to the inactive N133I mutants whereas active Q79L mutants decreased neurite outgrowth as reported previously (42) (Fig. 3). Fig. 3 View largeDownload slide The expression of phosphomimetic Rab5b mutant proteins resulted in an increase of neurite length in Neuro 2A cells. The cells were transiently transfected with plasmids expressing YFP-Rab5b WT, mutants or the vector (EGFP). The cells were treated with dibutyryl–cAMP to induce neurite outgrowth for 2 days and stained with antibody against ß-tubulin III (red) to clearly visualize the neurite ends. The well-separated and transfected (i.e. YFP-expressing: green) cells whose neurites did not overlap with neurites of other cells were chosen to measure the longest neurite length. Data were analysed to quantify neurite outgrowth. Nuclei were stained with DAPI (blue). (A) The average length of the longest neurites. All values are relative to the EGFP control and are expressed as the mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). The number of analysed cells for each type was 52–81. (B) A representative cellular image of each type. The arrows indicate cells chosen for neurite analysis and scale bars correspond to 100 μM. Statistical analysis was performed using two-way ANOVA with Tukey’s post hoc test. Fig. 3 View largeDownload slide The expression of phosphomimetic Rab5b mutant proteins resulted in an increase of neurite length in Neuro 2A cells. The cells were transiently transfected with plasmids expressing YFP-Rab5b WT, mutants or the vector (EGFP). The cells were treated with dibutyryl–cAMP to induce neurite outgrowth for 2 days and stained with antibody against ß-tubulin III (red) to clearly visualize the neurite ends. The well-separated and transfected (i.e. YFP-expressing: green) cells whose neurites did not overlap with neurites of other cells were chosen to measure the longest neurite length. Data were analysed to quantify neurite outgrowth. Nuclei were stained with DAPI (blue). (A) The average length of the longest neurites. All values are relative to the EGFP control and are expressed as the mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). The number of analysed cells for each type was 52–81. (B) A representative cellular image of each type. The arrows indicate cells chosen for neurite analysis and scale bars correspond to 100 μM. Statistical analysis was performed using two-way ANOVA with Tukey’s post hoc test. In addition to the neurite analysis, the effect of the Rab5b T6D mutant proteins on EGFR degradation after EGF stimulation was also investigated. We also tested constitutive active Q79L and dominant negative N133I mutants as controls. Overall, Q79L and N133I showed faster and slower degradation of EGFR, respectively, than those of the vector or WT protein (Fig. 4). This result was similar to a previous report in which the S34N inactive mutant showed the least, and the Q79L active mutant the most EGFR degradation among the samples transfected with the two mutants, WT, or the vector (48). As expected, the T6D mutant showed slow EGFR degradation rate similar to that of the N133I (Fig. 4), suggesting that the phosphomimetic T6D mutant decreased the endocytosis rate similar to the inactive mutant. Fig. 4 View largeDownload slide Expression of phosphomimetic Rab5b mutant proteins exhibited a decrease of EGFR degradation in HeLa cells. The indicated Flag-tagged Rab5 genes were overexpressed in HeLa cells by transient transfection and stimulated with EGF so EGFR could be internalized by endocytosis and degraded after fusion to a lysosome. (A) Each sample was harvested at the indicated times after EGF stimulation. The cell lysates were subjected to western analysis with the indicated antibodies. Asterisk indicates endogenous Rab5 and the arrow indicates exogenous Flag-tagged Rab5b. (B) The remaining amount of internalized EGFR was plotted as [EGFR]/[β-actin]. In each transfected sample, the amount of EGFR at each time point was normalized with the amount of EGFR at time point 0 which was set to 1. The experiment was performed three times. Statistical analysis was done using two-way ANOVA with Tukey’s post hoc test. (Following statistical significances [at least P < 0.05] were observed: T6D versus vector, Q79L and T6A at 10 min; T6D versus Q79L and T6A at 30 min; T6A versus all the others at both 10 and 30 min; Q79L versus all the others at 30 min). Fig. 4 View largeDownload slide Expression of phosphomimetic Rab5b mutant proteins exhibited a decrease of EGFR degradation in HeLa cells. The indicated Flag-tagged Rab5 genes were overexpressed in HeLa cells by transient transfection and stimulated with EGF so EGFR could be internalized by endocytosis and degraded after fusion to a lysosome. (A) Each sample was harvested at the indicated times after EGF stimulation. The cell lysates were subjected to western analysis with the indicated antibodies. Asterisk indicates endogenous Rab5 and the arrow indicates exogenous Flag-tagged Rab5b. (B) The remaining amount of internalized EGFR was plotted as [EGFR]/[β-actin]. In each transfected sample, the amount of EGFR at each time point was normalized with the amount of EGFR at time point 0 which was set to 1. The experiment was performed three times. Statistical analysis was done using two-way ANOVA with Tukey’s post hoc test. (Following statistical significances [at least P < 0.05] were observed: T6D versus vector, Q79L and T6A at 10 min; T6D versus Q79L and T6A at 30 min; T6A versus all the others at both 10 and 30 min; Q79L versus all the others at 30 min). Discussion Identifying physiological LRRK2 substrates is important to understand PD pathogenesis because LRRK2 contains a GTPase and a kinase both of which play critical roles in signal transduction and the most common observed pathogenic G2019S mutant increases its kinase activity. Several LRRK2 kinase substrates have been reported (14, 17–19, 23, 25, 39), although the downstream targets of LRRK2 GTPase are not clear. We reported previously that Rab5b, a small GTPase protein, interacts with LRRK2 and that Rab5b and LRRK2 coregulates endocytosis of synaptic vesicles (33). Colocalization of Rab5 and LRRK2 has been confirmed, but their colocalization was not significantly different from the LRRK2 WT, R1441C or G2019S variants (51). This study led us to reinvestigate the meaning of the interaction between Rab5b and LRRK2. Here, we found using in vitro kinase assays in conjunction with a targeted mutagenesis strategy that LRRK2 phosphorylates recombinant Rab5b protein on its Thr6 residue. Our previous data clearly showed that Arf1, a GTP binding protein expressed from a backbone vector the same as that of Rab5b, was not phosphorylated in a similar in vitro kinase assay (19), strongly suggesting that Rab5b is a specific substrate for LRRK2 kinase. Further studies suggested that Thr6 is a site phosphorylated by LRRK2. Our data do not exclude the possibility that there are additional phosphorylation site(s) other than Thr6 in Rab5b particularly because the MS experiment recovered only 75% of the peptides (Fig. 1C), and the T6A mutant still showed weak residual phosphorylation, 10% of WT, in the in vitro kinase assay (Fig. 1B). In addition, we demonstrated that phosphomimetic Rab5b T6D and Rab5b phosphorylated by LRRK2 enhanced the GTP hydrolysis activity of Rab5b and functions like an inactive Rab5b mutant (Figs 2–4). Taken together, our data suggest that LRRK2-mediated Rab5b phosphorylation facilitates switching of Rab5b from its active GTP-bound state to its inactive state; thus, turning off Rab5b signalling. Only a few reports have described phosphorylation of Rab family members and suggested that phosphorylation of Rab proteins could be important for GTP-binding affinity and the regulation of their cytosolic localization. For example, protein kinase C (PKC)-mediated phosphorylation of Rab24 has been suggested to decrease intrinsic GTPase activity of Rab24 (52). Moreover, phosphorylation of Rab6c by PKC increases GTP-binding of Rab6c and enhances Rab6c translocation to the cytosol (53). Furthermore, the differential phosphorylation of Rab5b isoforms has also been reported, with Rab5b is preferentially phosphorylated on Ser123 by Cdc2 kinase (43). Our results suggest that LRRK2 functions to enhance Rab5 GTPase activity via phosphorylation (Fig. 2). It has been reported that RabGAP-5, a GAP for Rab5, increases GTP hydrolysis activity of Rab5 by ∼8-fold in in vitro GTP-hydrolysis assays (54). The downregulation of RabGAP-5 by small hairpin RNAs in PC12 cells caused an inhibition of NGF-induced neurite outgrowth, similar to that observed with overexpression of Rab5b WT or Q79L (42), suggesting that active Rab5b signals inhibit neurite outgrowth. Because our GTPase assay results suggested that LRRK2-induced phosphorylation of Rab5 inactivates Rab5 (Fig. 2B), we tested this hypothesis using Rab5 T6D in two well-known Rab5 functional assays, neurite outgrowth and EGFR degradation assays (Figs 3 and 4). Both assays confirmed that T6D behaves as an inactive Rab5. However, how the weak 1.4-fold increase in GTPase activity of phosphorylated Rab5b exhibited a phenotype similar to that of the dominant negative mutant remains to be answered (Figs 2B and C versus4). It might be because phosphorylated Rab5b on Thr6 additionally affects functions of its downstream targets. The other explanation might be that the longer experiment time required in the functional assay compared with that in the biochemical assay could accumulate slight differences observed in the biochemical assays. The latter is more plausible when one considers the observation that the LRRK2 pathogenic mutations cause late-onset PD, although this requires more study. PD-associated genes, particularly α-synuclein and LRRK2, have been suggested to play roles in synaptic dysfunction, including deficits in vesicular transport/trafficking (33, 55). LRRK2 has been consistently suggested to function in vesicle trafficking (27, 56). This role is supported by our data showing Rab5b phosphorylation by LRRK2 inactivated Rab5b signalling, a regulator of early endosomes. Overexpression of active Rab5 Q79L has been reported to positively regulate endocytosis (32). We showed previously that overexpression of either LRRK2 or inactive Rab5 N133I in primary neuronal cells negatively regulated the endocytosis rate of synaptic vesicles, whereas coexpression of active, but not inactive, Rab5b with LRRK2 WT abolished the LRRK2 effects on the endocytosis rate and restored the normal endocytosis rate (33). Our current data provide a potential mechanism for our previous data. In the case of overexpression of LRRK2 alone, phosphorylation of endogenous Rab5b by LRRK2 inactivates Rab5 signalling by enhancing Rab5 GTPase activity. In contrast, overexpression of active Rab5b with LRRK2 increases active Rab5 protein levels and restores an active Rab5b signal, thereby restoring a normal endocytosis rate. Our previous neurite assay with PC-12 cells treated with NGF (50) showed that coexpression of LRRK2 WT with Rab5 WT partially rescued neurite length shortening observed when Rab5 WT was expressed alone, which was similar to our neurite assay result with T6D expression (Fig. 3). LRRK2 kinase activity negatively regulates neurite outgrowth (11), which seems to contradict our results. However, it is possible that there are other LRRK2 kinase substrates that negatively regulate neurite outgrowth upon phosphorylation by LRRK2 and, therefore, compensate for the positive effect on neurite outgrowth of Rab5 phosphorylation. Inhibiting moesin phosphorylation by LRRK2 rescues shortening of neurite length, suggesting that LRRK2 negatively regulates neurite outgrowth via moesin phosphorylation (14, 16). Furthermore, a recent study reported that LRRK2 rescues Mfn1-mediated neurite shortening (51) in a kinase activity-dependent manner, which is a phenotype similar to our LRRK2-induced Rab5 phosphorylation. The EGFR degradation assay also supported our hypothesis (Fig. 4). However, non-phosphorylatable T6A exhibited a rapid EGFR degradation, similar to that of Q79L mutant, although T6A showed the GTPase activity similar to that of the WT. (Fig. 2C). It might be due to that the mutation caused conformation changes which lead to change of interacting proteins related to lysosomes or protein degradation machinery, although it needs more experiment to make it clear. A recent study reported that expression of LRRK2 G2019S delayed clearance of Alexa555-EGF in a kinase-dependent manner and caused a decrease of active GTP bound Rab7, a late endosomal regulator (56). In addition, they showed delay in early-to-late endosomal trafficking as well as in trafficking out of late endosomes. Because coexpression of Rab7 rescued the deficits in trafficking and degradation of EGFR, the authors concluded that Rab7 is a target regulated by LRRK2. However, our data suggest that a part of delay in EGFR degradation is caused by Rab5 via phosphorylation by LRRK2. Because both G2019S and R1441C exhibited similar deficits in their study (56), other mechanisms in addition to phosphorylation might be involved in the endosomal trafficking regulated by LRRK2. Our results suggest that LRRK2 decreases the active Rab5b duration, reduces the endocytosis rate and possibly interferes with vesicle trafficking by phosphorylation of Rab5b. Abbreviations Abbreviations 4E-BP eukaryotic initiation factor 4E (eIF4E)-binding protein ANOVA analysis of variance BSA bovine serum albumin CID collision-induced dissociation DAPI 4′,6-diamidino-2-phenylindole DMEM Dulbecco Modified Eagle Medium FBS foetal bovine serum EGFR epidermal growth factor receptor GAP GTPase activating protein LRRK2 Leucine-rich repeat kinase 2 MAPKK mitogen-activated protein kinase kinase MAPKKK Mitogen-activated protein kinase kinase kinase NGF neuronal growth factor PD Parkinson’s disease PIC protease inhibitor cocktail PKC protein kinase C p-Thr phospho-threonine p-TXR phospho-threonine-X-arginine ROS reactive oxygen species SE standard error WT wild type Acknowledgements We thank Dr. W.D. Heo (KAIST, Korea) for providing the YFP-Rab5b Q79L plasmid. Funding This work was supported by an intramural grant from InAm Neuroscience Research Center (Sanbon Hospital, Wonkwang University, Korea) and grants from the Basic Science Research Program [2012R1A1A3008447 to W.S.] and SRC [MSIP No. 2011‐0030049, 2011‐0030928, 2012‐003338 to H.S.] through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology. Conflict of Interest None declared. References 1 Lubbe S,  Morris HR.  Recent advances in Parkinson’s disease genetics,  J. Neurol. ,  2014, vol.  261 (pg.  259- 266) Google Scholar CrossRef Search ADS PubMed  2 Paisan-Ruiz C,  Jain S,  Evans EW,  Gilks WP,  Simon J,  van der Brug M,  Lopez de Munain A,  Aparicio S,  Gil AM,  Khan N,  Johnson J,  Martinez JR,  Nicholl D,  Carrera IM,  Pena AS,  de Silva R,  Lees A,  Marti-Masso JF,  Perez-Tur J,  Wood NW,  Singleton AB.  Cloning of the gene containing mutations that cause PARK8-linked Parkinson’s disease,  Neuron ,  2004, vol.  44 (pg.  595- 600) Google Scholar CrossRef Search ADS PubMed  3 Zimprich A,  Biskup S,  Leitner P,  Lichtner P,  Farrer M,  Lincoln S,  Kachergus J,  Hulihan M,  Uitti RJ,  Calne DB,  Stoessl AJ,  Pfeiffer RF,  Patenge N,  Carbajal IC,  Vieregge P,  Asmus F,  Muller-Myhsok B,  Dickson DW,  Meitinger T,  Strom TM,  Wszolek ZK,  Gasser T.  Mutations in LRRK2 cause autosomal-dominant parkinsonism with pleomorphic pathology,  Neuron ,  2004, vol.  44 (pg.  601- 607) Google Scholar CrossRef Search ADS PubMed  4 Zabetian CP,  Samii A,  Mosley AD,  Roberts JW,  Leis BC,  Yearout D,  Raskind WH,  Griffith A.  A clinic-based study of the LRRK2 gene in Parkinson disease yields new mutations,  Neurology ,  2005, vol.  65 (pg.  741- 744) Google Scholar CrossRef Search ADS PubMed  5 Healy DG,  Falchi M,  O’Sullivan SS,  Bonifati V,  Durr A,  Bressman S,  Brice A,  Aasly J,  Zabetian CP,  Goldwurm S,  Ferreira JJ,  Tolosa E,  Kay DM,  Klein C,  Williams DR,  Marras C,  Lang AE,  Wszolek ZK,  Berciano J,  Schapira AH,  Lynch T,  Bhatia KP,  Gasser T,  Lees AJ,  Wood NW.  Phenotype, genotype, and worldwide genetic penetrance of LRRK2-associated Parkinson’s disease: a case-control study,  Lancet Neurol. ,  2008, vol.  7 (pg.  583- 590) Google Scholar CrossRef Search ADS PubMed  6 Gloeckner CJ,  Kinkl N,  Schumacher A,  Braun RJ,  O’Neill E,  Meitinger T,  Kolch W,  Prokisch H,  Ueffing M.  The Parkinson disease causing LRRK2 mutation I2020T is associated with increased kinase activity,  Hum. Mol. Genet. ,  2006, vol.  15 (pg.  223- 232) Google Scholar CrossRef Search ADS PubMed  7 West AB,  Moore DJ,  Choi C,  Andrabi SA,  Li X,  Dikeman D,  Biskup S,  Zhang Z,  Lim KL,  Dawson VL,  Dawson TM.  Parkinson’s disease-associated mutations in LRRK2 link enhanced GTP-binding and kinase activities to neuronal toxicity,  Hum. Mol. Genet. ,  2007, vol.  16 (pg.  223- 232) Google Scholar CrossRef Search ADS PubMed  8 Smith WW,  Pei Z,  Jiang H,  Dawson VL,  Dawson TM,  Ross CA.  Kinase activity of mutant LRRK2 mediates neuronal toxicity,  Nat. Neurosci. ,  2006, vol.  9 (pg.  1231- 1233) Google Scholar CrossRef Search ADS PubMed  9 Greggio E,  Jain S,  Kingsbury A,  Bandopadhyay R,  Lewis P,  Kaganovich A,  van der Brug MP,  Beilina A,  Blackinton J,  Thomas KJ,  Ahmad R,  Miller DW,  Kesavapany S,  Singleton A,  Lees A,  Harvey RJ,  Harvey K,  Cookson MR.  Kinase activity is required for the toxic effects of mutant LRRK2/dardarin,  Neurobiol. Dis. ,  2006, vol.  23 (pg.  329- 341) Google Scholar CrossRef Search ADS PubMed  10 Heo HY,  Park JM,  Kim CH,  Han BS,  Kim KS,  Seol W.  LRRK2 enhances oxidative stress-induced neurotoxicity via its kinase activity,  Exp. Cell Res. ,  2010, vol.  316 (pg.  649- 656) Google Scholar CrossRef Search ADS PubMed  11 MacLeod D,  Dowman J,  Hammond R,  Leete T,  Inoue K,  Abeliovich A.  The familial Parkinsonism gene LRRK2 regulates neurite process morphology,  Neuron ,  2006, vol.  52 (pg.  587- 593) Google Scholar CrossRef Search ADS PubMed  12 Plowey ED,  Cherra SJ3rd,  Liu YJ,  Chu CT.  Role of autophagy in G2019S-LRRK2-associated neurite shortening in differentiated SH-SY5Y cells,  J. Neurochem. ,  2008, vol.  105 (pg.  1048- 1056) Google Scholar CrossRef Search ADS PubMed  13 Orenstein SJ,  Kuo SH,  Tasset I,  Arias E,  Koga H,  Fernandez-Carasa I,  Cortes E,  Honig LS,  Dauer W,  Consiglio A,  Raya A,  Sulzer D,  Cuervo AM.  Interplay of LRRK2 with chaperone-mediated autophagy,  Nat. Neurosci. ,  2013, vol.  16 (pg.  394- 406) Google Scholar CrossRef Search ADS PubMed  14 Jaleel M,  Nichols RJ,  Deak M,  Campbell DG,  Gillardon F,  Knebel A,  Alessi DR.  LRRK2 phosphorylates moesin at threonine-558: characterization of how Parkinson’s disease mutants affect kinase activity,  Biochem. J. ,  2007, vol.  405 (pg.  307- 317) Google Scholar CrossRef Search ADS PubMed  15 Skibinski G,  Nakamura K,  Cookson MR,  Finkbeiner S.  Mutant LRRK2 toxicity in neurons depends on LRRK2 levels and synuclein but not kinase activity or inclusion bodies,  J. Neurosci. ,  2014, vol.  34 (pg.  418- 433) Google Scholar CrossRef Search ADS PubMed  16 Parisiadou L,  Xie C,  Cho HJ,  Lin X,  Gu XL,  Long CX,  Lobbestael E,  Baekelandt V,  Taymans JM,  Sun L,  Cai H.  Phosphorylation of ezrin/radixin/moesin proteins by LRRK2 promotes the rearrangement of actin cytoskeleton in neuronal morphogenesis,  J. Neurosci. ,  2009, vol.  29 (pg.  13971- 13980) Google Scholar CrossRef Search ADS PubMed  17 Imai Y,  Gehrke S,  Wang HQ,  Takahashi R,  Hasegawa K,  Oota E,  Lu B.  Phosphorylation of 4E-BP by LRRK2 affects the maintenance of dopaminergic neurons in Drosophila,  EMBO. J. ,  2008, vol.  27 (pg.  2432- 2443) Google Scholar CrossRef Search ADS PubMed  18 Gillardon F.  Leucine-rich repeat kinase 2 phosphorylates brain tubulin-beta isoforms and modulates microtubule stability—a point of convergence in Parkinsonian neurodegeneration? J,  Neurochem. ,  2009, vol.  110 (pg.  1514- 1522) Google Scholar CrossRef Search ADS   19 Yun HJ,  Park J,  Ho DH,  Kim H,  Kim CH,  Oh H,  Ga I,  Seo H,  Chang S,  Son I,  Seol W.  LRRK2 phosphorylates Snapin and inhibits interaction of Snapin with SNAP-25,  Exp. Mol. Med. ,  2013, vol.  45 pg.  e36  Google Scholar CrossRef Search ADS PubMed  20 Xiong Y,  Yuan C,  Chen R,  Dawson TM,  Dawson VL.  ArfGAP1 is a GTPase activating protein for LRRK2: reciprocal regulation of ArfGAP1 by LRRK2,  J. Neurosci. ,  2012, vol.  32 (pg.  3877- 3886) Google Scholar CrossRef Search ADS PubMed  21 Stafa K,  Trancikova A,  Webber PJ,  Glauser L,  West AB,  Moore DJ.  GTPase activity and neuronal toxicity of Parkinson’s disease-associated LRRK2 is regulated by ArfGAP1,  PLoS Genet. ,  2012, vol.  8 pg.  e1002526  Google Scholar CrossRef Search ADS PubMed  22 Bailey RM,  Covy JP,  Melrose HL,  Rousseau L,  Watkinson R,  Knight J,  Miles S,  Farrer MJ,  Dickson DW,  Giasson BI,  Lewis J.  LRRK2 phosphorylates novel tau epitopes and promotes tauopathy,  Acta Neuropathol. ,  2013, vol.  126 (pg.  809- 827) Google Scholar CrossRef Search ADS PubMed  23 Gloeckner CJ,  Schumacher A,  Boldt K,  Ueffing M.  The Parkinson disease-associated protein kinase LRRK2 exhibits MAPKKK activity and phosphorylates MKK3/6 and MKK4/7, in vitro,  J. Neurochem. ,  2009, vol.  109 (pg.  959- 968) Google Scholar CrossRef Search ADS PubMed  24 Ohta E,  Kawakami F,  Kubo M,  Obata F.  LRRK2 directly phosphorylates Akt1 as a possible physiological substrate: impairment of the kinase activity by Parkinson’s disease-associated mutations,  FEBS Lett. ,  2011, vol.  585 (pg.  2165- 2170) Google Scholar CrossRef Search ADS PubMed  25 Martin I,  Kim JW,  Lee BD,  Kang HC,  Xu JC,  Jia H,  Stankowski J,  Kim MS,  Zhong J,  Kumar M,  Andrabi SA,  Xiong Y,  Dickson DW,  Wszolek ZK,  Pandey A,  Dawson TM,  Dawson VL.  Ribosomal protein s15 phosphorylation mediates LRRK2 neurodegeneration in Parkinson’s disease,  Cell ,  2014, vol.  157 (pg.  472- 485) Google Scholar CrossRef Search ADS PubMed  26 Kumar A,  Greggio E,  Beilina A,  Kaganovich A,  Chan D,  Taymans JM,  Wolozin B,  Cookson MR.  The Parkinson’s disease associated LRRK2 exhibits weaker in vitro phosphorylation of 4E-BP compared to autophosphorylation,  PLoS ONE ,  2010, vol.  5 pg.  e8730  Google Scholar CrossRef Search ADS PubMed  27 Rideout HJ,  Stefanis L.  The neurobiology of LRRK2 and its role in the pathogenesis of Parkinson’s disease,  Neurochem. Res. ,  2014, vol.  39 (pg.  576- 592) Google Scholar CrossRef Search ADS PubMed  28 Deng J,  Lewis PA,  Greggio E,  Sluch E,  Beilina A,  Cookson MR.  Structure of the ROC domain from the Parkinson’s disease-associated leucine-rich repeat kinase 2 reveals a dimeric GTPase,  Proc. Natl Acad. Sci. USA ,  2008, vol.  105 (pg.  1499- 1504) Google Scholar CrossRef Search ADS   29 Guo L,  Gandhi PN,  Wang W,  Petersen RB,  Wilson-Delfosse AL,  Chen SG.  The Parkinson’s disease-associated protein, leucine-rich repeat kinase 2 (LRRK2), is an authentic GTPase that stimulates kinase activity,  Exp. Cell Res. ,  2007, vol.  313 (pg.  3658- 3670) Google Scholar CrossRef Search ADS PubMed  30 Liao J,  Wu CX,  Burlak C,  Zhang S,  Sahm H,  Wang M,  Zhang ZY,  Vogel KW,  Federici M,  Riddle SM,  Nichols RJ,  Liu D,  Cookson MR,  Stone TA,  Hoang QQ.  Parkinson disease-associated mutation R1441H in LRRK2 prolongs the ‘active state’ of its GTPase domain,  Proc. Natl Acad. Sci. USA ,  2014, vol.  111 (pg.  4055- 4060) Google Scholar CrossRef Search ADS   31 Li X,  Tan YC,  Poulose S,  Olanow CW,  Huang XY,  Yue Z.  Leucine-rich repeat kinase 2 (LRRK2)/PARK8 possesses GTPase activity that is altered in familial Parkinson’s disease R1441C/G mutants,  J. Neurochem. ,  2007, vol.  103 (pg.  238- 247) Google Scholar CrossRef Search ADS PubMed  32 Stenmark H,  Parton RG,  Steele-Mortimer O,  Lutcke A,  Gruenberg J,  Zerial M.  Inhibition of rab5 GTPase activity stimulates membrane fusion in endocytosis,  EMBO J. ,  1994, vol.  13 (pg.  1287- 1296) Google Scholar PubMed  33 Shin N,  Jeong H,  Kwon J,  Heo HY,  Kwon JJ,  Yun HJ,  Kim CH,  Han BS,  Tong Y,  Shen J,  Hatano T,  Hattori N,  Kim K-S,  Chang S,  Seol W.  LRRK2 regulates synaptic vesicle endocytosis,  Exp. Cell Res. ,  2008, vol.  314 (pg.  2055- 2065) Google Scholar CrossRef Search ADS PubMed  34 Nichols RJ,  Dzamko N,  Hutti JE,  Cantley LC,  Deak M,  Moran J,  Bamborough P,  Reith AD,  Alessi DR.  Substrate specificity and inhibitors of LRRK2, a protein kinase mutated in Parkinson’s disease,  Biochem. J. ,  2009, vol.  424 (pg.  47- 60) Google Scholar CrossRef Search ADS PubMed  35 Tremblay RG,  Sikorska M,  Sandhu JK,  Lanthier P,  Ribecco-Lutkiewicz M,  Bani-Yaghoub M.  Differentiation of mouse Neuro 2A cells into dopamine neurons,  J. Neurosci. Methods ,  2010, vol.  186 (pg.  60- 67) Google Scholar CrossRef Search ADS PubMed  36 Kim HJ,  Zhong Q,  Sheng ZH,  Yoshimori T,  Liang C,  Jung JU.  Beclin-1-interacting autophagy protein Atg14L targets the SNARE-associated protein Snapin to coordinate endocytic trafficking,  J. Cell Sci. ,  2012, vol.  125 (pg.  4740- 4750) Google Scholar CrossRef Search ADS PubMed  37 Pungaliya PP,  Bai Y,  Lipinski K,  Anand VS,  Sen S,  Brown EL,  Bates B,  Reinhart PH,  West AB,  Hirst WD,  Braithwaite SP.  Identification and characterization of a leucine-rich repeat kinase 2 (LRRK2) consensus phosphorylation motif,  PLoS ONE ,  2010, vol.  5 pg.  e13672  Google Scholar CrossRef Search ADS PubMed  38 Angeles DC,  Gan BH,  Onstead L,  Zhao Y,  Lim KL,  Dachsel J,  Melrose H,  Farrer M,  Wszolek ZK,  Dickson DW,  Tan EK.  Mutations in LRRK2 increase phosphorylation of peroxiredoxin 3 exacerbating oxidative stress-induced neuronal death,  Hum. Mutat. ,  2012, vol.  32 (pg.  1390- 1397) Google Scholar CrossRef Search ADS   39 Lobbestael E,  Baekelandt V,  Taymans JM.  Phosphorylation of LRRK2: from kinase to substrate,  Biochem. Soc. Trans. ,  2012, vol.  40 (pg.  1102- 1110) Google Scholar CrossRef Search ADS PubMed  40 Hoffenberg S,  Sanford JC,  Liu S,  Daniel DS,  Tuvin M,  Knoll BJ,  Wessling-Resnick M,  Dickey BF.  Biochemical and functional characterization of a recombinant GTPase, Rab5, and two of its mutants,  J. Biol. Chem. ,  1995, vol.  270 (pg.  5048- 5056) Google Scholar CrossRef Search ADS PubMed  41 Li G,  Stahl PD.  Structure-function relationship of the small GTPase rab5,  J. Biol. Chem. ,  1993, vol.  268 (pg.  24475- 24480) Google Scholar PubMed  42 Liu J,  Lamb D,  Chou MM,  Liu YJ,  Li G.  Nerve growth factor-mediated neurite outgrowth via regulation of Rab5,  Mol. Biol. Cell ,  2007, vol.  18 (pg.  1375- 1384) Google Scholar CrossRef Search ADS PubMed  43 Chiariello M,  Bruni CB,  Bucci C.  The small GTPases Rab5a, Rab5b and Rab5c are differentially phosphorylated in vitro,  FEBS Lett. ,  1999, vol.  453 (pg.  20- 24) Google Scholar CrossRef Search ADS PubMed  44 Sun P,  Yamamoto H,  Suetsugu S,  Miki H,  Takenawa T,  Endo T.  Small GTPase Rah/Rab34 is associated with membrane ruffles and macropinosomes and promotes macropinosome formation,  J. Biol. Chem. ,  2003, vol.  278 (pg.  4063- 4071) Google Scholar CrossRef Search ADS PubMed  45 Greggio E,  Lewis PA,  van der Brug MP,  Ahmad R,  Kaganovich A,  Ding J,  Beilina A,  Baker AK,  Cookson MR.  Mutations in LRRK2/dardarin associated with Parkinson disease are more toxic than equivalent mutations in the homologous kinase LRRK1,  J. Neurochem. ,  2007, vol.  102 (pg.  93- 102) Google Scholar CrossRef Search ADS PubMed  46 Li G,  Barbieri MA,  Colombo MI,  Stahl PD.  Structural features of the GTP-binding defective Rab5 mutants required for their inhibitory activity on endocytosis,  J. Biol. Chem. ,  1994, vol.  269 (pg.  14631- 14635) Google Scholar PubMed  47 Bucci C,  Parton RG,  Mather IH,  Stunnenberg H,  Simons K,  Hoflack B,  Zerial M.  The small GTPase rab5 functions as a regulatory factor in the early endocytic pathway,  Cell ,  1992, vol.  70 (pg.  715- 728) Google Scholar CrossRef Search ADS PubMed  48 Chen X,  Wang Z.  Regulation of epidermal growth factor receptor endocytosis by wortmannin through activation of Rab5 rather than inhibition of phosphatidylinositol 3-kinase,  EMBO Rep. ,  2001, vol.  2 (pg.  842- 849) Google Scholar CrossRef Search ADS PubMed  49 Dong Y,  Heien ML,  Maxson MM,  Ewing AG.  Amperometric measurements of catecholamine release from single vesicles in MN9D cells,  J. Neurochem. ,  2008, vol.  107 (pg.  1589- 1595) Google Scholar CrossRef Search ADS PubMed  50 Heo HY,  Kim KS,  Seol W.  Coordinate regulation of neurite outgrowth by LRRK2 and its interactor, Rab5,  Exp. Neurobiol. ,  2010, vol.  19 (pg.  97- 105) Google Scholar CrossRef Search ADS PubMed  51 Stafa K,  Tsika E,  Moser R,  Musso A,  Glauser L,  Jones A,  Biskup S,  Xiong Y,  Bandopadhyay R,  Dawson VL,  Dawson TM,  Moore DJ.  Functional interaction of Parkinson’s disease-associated LRRK2 with members of the dynamin GTPase superfamily,  Hum. Mol. Genet. ,  2013, vol.  15 (pg.  2055- 2077) 52 Ding J,  Soule G,  Overmeyer JH,  Maltese WA.  Tyrosine phosphorylation of the Rab24 GTPase in cultured mammalian cells,  Biochem. Biophys. Res. Commun. ,  2003, vol.  312 (pg.  670- 675) Google Scholar CrossRef Search ADS PubMed  53 Fitzgerald ML,  Reed GL.  Rab6 is phosphorylated in thrombin-activated platelets by a protein kinase C-dependent mechanism: effects on GTP/GDP binding and cellular distribution,  Biochem. J. ,  1999, vol.  342 (pg.  353- 360) Google Scholar CrossRef Search ADS PubMed  54 Haas AK,  Fuchs E,  Kopajtich R,  Barr FA.  A GTPase-activating protein controls Rab5 function in endocytic trafficking,  Nat. Cell Biol. ,  2005, vol.  7 (pg.  887- 893) Google Scholar CrossRef Search ADS PubMed  55 Plowey ED,  Chu CT.  Synaptic dysfunction in genetic models of Parkinson’s disease: a role for autophagy? Neurobiol,  Dis. ,  2010, vol.  43 (pg.  60- 67) 56 Gomez-Suaga P,  Rivero-Rios P,  Fdez E,  Blanca Ramirez M,  Ferrer I,  Aiastui A,  Lopez De Munain A,  Hilfiker S.  LRRK2 delays degradative receptor trafficking by impeding late endosomal budding through decreasing Rab7 activity,  Hum. Mol. Genet. ,  2014, vol.  23 (pg.  6779- 6796) Google Scholar CrossRef Search ADS PubMed  © The Authors 2015. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
TI - An early endosome regulator, Rab5b, is an LRRK2 kinase substrate
JF - The Journal of Biochemistry
DO - 10.1093/jb/mvv005
DA - 2015-01-19
UR - https://www.deepdyve.com/lp/oxford-university-press/an-early-endosome-regulator-rab5b-is-an-lrrk2-kinase-substrate-1P7RoGxOGi
SP - 485
EP - 495
VL - 157
IS - 6
DP - DeepDyve
ER -