Figure 1.
Phosphorylation of 4E-BP1 by LRRK2 in vitro and in mammalian cells.
(A) In vitro kinase assay with [32P]-γ-ATP, recombinant GST-tagged human LRRK2 (ΔN, residues 970–2527) and GST-tagged human 4E-BP1. Coomassie-stained SDS-PAGE gels indicate equal loading of 4E-BP1 and LRRK2 proteins in each condition. Autoradiographs indicate the phosphorylation of 4E-BP1 by WT LRRK2 compared to kinase-inactive D1994A LRRK2. Autophosphorylation of WT LRRK2 is also detected. (B) Western blot analysis of endogenous 4E-BP1 phosphorylation at Thr37/Thr46 or Ser65 in HEK-293T cells transiently expressing myc-tagged human LRRK2 variants (WT, G2019S and D1994A). LRRK2 overexpression fails to alter 4E-BP1 phosphorylation. Blots are representative of duplicate experiments. Molecular mass markers are indicated in kilodaltons (kDa).
Figure 2.
Effect of LRRK2 on 4E-BP1 subcellular localization and protein complex formation.
(A) Confocal fluorescence microscopy reveals minimal co-localization of FLAG-tagged human LRRK2 variants and endogenous 4E-BP1 in rat primary cortical neurons. Pathogenic mutations (R1441C or G2019S) do not alter the localization of LRRK2 with 4E-BP1 compared to WT LRRK2. Cytofluorograms and co-localization coefficients (Rcoloc; mean±SEM, n = 5–10 neurons) reveal the extent of co-localization between LRRK2 and 4E-BP1 fluorescent signals. Confocal images are taken from single z-plane at 0.1 µm thickness. Images are representative of at least five neurons taken from duplicate experiments. Scale bar: 10 µm. (B) Subcellular fractionation of cerebral cortex from WT and LRRK2 KO mice, or human G2019S LRRK2 transgenic (TG) and non-transgenic (NTG) mice. 4E-BP1 is enriched in soluble cytosolic (S1, S2 and S3) fractions, and at lower levels in synaptosomal (LS1) and synaptic vesicle (LS2) cytosolic fractions. 4E-BP1 subcellular localization is not altered by LRRK2 deletion or G2019S LRRK2 expression compared to control mice. Endogenous and human LRRK2 is enriched in the microsomal (P3) fraction and at lower levels in synaptosomal membrane (LP1) and soluble cytosolic (S1 and S2) fractions. The distribution of marker proteins demonstrates the enrichment of mitochondria/heavy membranes (TIM23; P2 and LP1), synaptosomal/synaptic vesicle membranes (synaptophysin 1; P2, P3, LP1 and LP2) and synaptosomal/synaptic vesicle cytosolic (α-synuclein; LS1 and LS2). (C) Size-exclusion chromatography on soluble whole brain extracts from WT and LRRK2 KO mice. Sequential fractions (0.5 ml) were analyzed by Western blotting with antibodies to total or phosphorylated (Thr37/46) 4E-BP1 and β-tubulin, whereas total homogenates were probed with antibodies to LRRK2 (c41-2/MJFF2). The elution profile of 4E-BP1 is similar in WT and KO brains, whereas the elution profile of individual protein standards is indicated. Blots are representative of duplicate experiments. Molecular mass markers are indicated in kilodaltons (kDa).
Figure 3.
Effect of LRRK2 on 4E-BP1 phosphorylation in mouse brain.
Total 4E-BP1 immunoprecipitates or input lysates from the (A) cerebral cortex or (B) striatum of WT and LRRK2 KO mice, or human LRRK2 (R1441C or G2019S) transgenic (TG) and non-transgenic (NTG) mice were analyzed by Western blot analysis with antibodies to phosphorylated (Thr37/46) and total 4E-BP1, or LRRK2 (total: c41-2/MJFF2; human-selective: c81-8/MJFF4). Densitometric analysis reveals unaltered 4E-BP1 phosphorylation by LRRK2 deletion or mutant human LRRK2 expression compared to littermate control mice. The levels of phosphorylated 4E-BP1 were normalized to total 4E-BP1 and expressed as a percent of control mice (mean±SEM, n = 3 mice/genotype). Molecular mass markers are indicated in kilodaltons (kDa).
Figure 4.
Effect of LRRK2 on 4E-BP1 post-translational modification in mammalian cells and brain.
(A) 2D SDS-PAGE (pH 3–10 and 8–16% SDS-PAGE) analysis of SH-SY5Y cell extracts expressing FLAG-tagged human LRRK2 variants (WT, G2019S or D1994A). 2D blots were probed with 4E-BP1 antibody or stained with Ponceau S red to reveal equivalent protein loading. The 2D migration profile of 4E-BP1 is not altered by LRRK2 kinase-inactive (D1994A) or kinase-hyperactive (G2019S) mutations relative to WT LRRK2. 1D blots were probed with anti-FLAG antibody to reveal equivalent human LRRK2 levels. Blots are representative of duplicate experiments. (B and C) 2D SDS-PAGE analysis of cerebral cortex and striatum extracts derived from WT or LRRK2 KO mice with 4E-BP1 antibody or Ponceau S red as a protein loading control. The 2D profile of 4E-BP1 is not altered by LRRK2 deletion. Blots are representative of three experiments using independent mice for each genotype. Molecular mass markers are indicated in kilodaltons (kDa).
Figure 5.
Phosphorylation of 4E-BP1 in brains of PD subjects.
Western blot analysis of (A) frontal cortex or (B) basal ganglia soluble fractions from human control, idiopathic PD (iPD) and G2019S LRRK2 PD subjects with antibodies to total or phosphorylated (Thr37/46) 4E-BP1, or β-actin as a protein loading control. Molecular mass markers are indicated in kilodaltons (kDa). Densitometric analysis of 4E-BP1 phosphorylation (upper protein band) or total 4E-BP1 levels in idiopathic or G2019S PD brains compared to control brains. The levels of phosphorylated 4E-BP1 were normalized to total 4E-BP1, whereas total 4E-BP1 levels were normalized to β-actin levels, and expressed as a percent of control subjects (mean±SEM, n = 4–5 brains/group). For basal ganglia, G2019S subject 3 was excluded from the densitometric analysis due to a lack of detectable 4E-BP1 expression. *P<0.05 or **P<0.01 by one-way ANOVA with Newman-Keuls post-hoc analysis. ns, non-significant.
Table 1.
Clinical details of human brain tissue.