The Parkinson's disease-linked Leucine-rich repeat kinase 2 (LRRK2) is required for insulin-stimulated translocation of GLUT4

Abstract

Mutations within Leucine-rich repeat kinase 2 (LRRK2) are associated with late-onset Parkinson's disease. The physiological function of LRRK2 and molecular mechanism underlying the pathogenic role of LRRK2 mutations remain uncertain. Here, we investigated the role of LRRK2 in intracellular signal transduction. We find that deficiency of Lrrk2 in rodents affects insulin-dependent translocation of glucose transporter type 4 (GLUT4). This deficit is restored during aging by prolonged insulin-dependent activation of protein kinase B (PKB, Akt) and Akt substrate of 160 kDa (AS160), and is compensated by elevated basal expression of GLUT4 on the cell surface. Furthermore, we find a crucial role of Rab10 phosphorylation by LRRK2 for efficient insulin signal transduction. Translating our findings into human cell lines, we find comparable molecular alterations in fibroblasts from Parkinson's patients with the known pathogenic G2019S LRRK2 mutation. Our results highlight the role of LRRK2 in insulin-dependent signalling with potential therapeutic implications.

Figure 1

Animal models. (a) Schematic representation of 7 bp deletion (blue) in genome of Lrrk2 deficient rat line. A part of exon 2 (red) and intron II (black) are depicted for orientation. (b) Generation strategy for Lrrk2 knock-down mouse line carrying a shLrrk2 transgene inserted into the Rosa locus. (c) Western blot analysis of Lrrk2 expression in brain-, spleen- and kidney- protein extracts of Lrrk2 deficient rats and (d) Western blot analysis of Lrrk2 expression in different tissues (brain, spleen and kidney) of Lrrk2 knock-down mouse in comparison to wild-type using anti-LRRK2 MJFF#2 antibody. f: female; m: male.

Figure 2

Effect of different growth factors on neurite outgrowth and cell survival. (a) Neurite length of primary hippocampal neurons from wild-type and Lrrk2 deficient mice lines Lrrk2 knock-out and Lrrk2 knock-down at DIV3 in presence of different growth factors (hFGF, BDNF, CNTF, Epo and without any growth factors (w/o); mean and SEM). (b) Lrrk2 expression in mouse wild-type primary hippocampal neurons at DIV 1 to 7 (lanes 1 to 4) and Lrrk2 deficient neurons at DIV 7 (lane 5). Lrrk2 expression was detected using the LRRK2 specific antibody MJFF#2. (c) Survival of Lrrk2 deficient hippocampal neurons (Lrrk2 knock-out and Lrrk2 knock-down) in comparison to wild-type neurons at DIV 1, 3 and 7 in presence of hFGF and without addition of any growth factors (w/o) as negative control (mean and SEM). (d) Percentage of growing late erythrocyte progenitor colonies (BFU-Es) and erythrocyte progenitor containing mix colonies (CFU-GEMMs) in CFC-assay media with different Epo concentrations (0, 0.5, 1, 2.5, 5 and 10 ng/ml; mean and SEM).

Figure 3

Induced phosphorylation of Akt in Lrrk2 deficient cells. (a) Western blot analysis of P-Akt Thr 308 and Ser473 in protein extracts from monocytes of 1 year old Lrrk2 deficient and wild-type mice stimulated with insulin at different time points (0 to 40 min) after stimulation. (b) Quantification of P-Akt Thr308 (b, top) and P-Akt Ser473 (b, button) intensity in protein extracts from monocytes of 1 year old Lrrk2 deficient and wild-type mice at different time-points after stimulation with insulin. N = 8 (for Thr308) and 6 (for Ser 473) independent experiments, normalized to Akt, mean ± SEM. (c) Western blot analysis of protein lysates from fibroblasts of 22 months old Lrrk2 deficient and wild-type rats at different time points after stimulation with insulin and (d) the corresponding quantification of P-Akt Thr308 (top) and P-Akt S473 (button) signal intensity. N = 10 independent experiments, normalized to Akt, mean ± SEM. (e) Western blot analysis of protein lysates from fibroblasts of 6 months old rats (wild-type and Lrrk2-deficient) after stimulation with insulin and corresponding quantification of P-Akt Thr308 (f) and S473 (g) signal intensity. N = 7 independent experiments, normalized to Akt, mean and SEM.

Figure 4

GLUT4 translocation in Lrrk2 deficient fibroblasts. (a) GLUT4 immunostaining on the cell surface of fibroblasts from 6 months old Lrrk2 deficient and wild-type rats without (w/o) insulin and 10 and 30 min after insulin addition and (c) the corresponding quantification of the GLUT4 signal intensity at different time-points (0, 10 and 30 min) after stimulation (mean and SEM). (b) GLUT4 immunostaining on plasma membrane of fibroblasts from 22 months old Lrrk2 deficient and wild-type rats without insulin and 10 and 30 min after stimulation and (d) the corresponding quantification of the GLUT4 signal intensity (mean and SEM). Red: GLUT4 immunostaining; green: wheat germ agglutinin (=plasma membrane) staining; blue: DAPI.

Figure 5

Insulin signalling in Lrrk2 deficient animals. (a) Insulin-triggered phosphorylation of IRβ and AS160 (Thr642) in fibroblasts from 22 months old Lrrk2 deficient rats at different time-points after stimulation and the corresponding quantification of P-IRβ (b, n = 7, normalized to IR-β) and P-AS160 Thr642 (c, n = 10, normalized to AS160) signal intensity (mean ± SEM). (d) Western blot analysis (fibroblasts from 22 months old rats as example) and quantification of total GLUT4, AS160 and Rab10 expression in fibroblasts from 6 months (e) and 22 months old (f) Lrrk2 deficient and wild-type rats (normalized to tubulin, mean ± SEM). # are numbers of animals/cell lines. The difference in GLUT4 and AS160 signal intensity between 6 months und 22 months old sample-groups results from differences in experimental procedure and does not reflect the absolute quantity of GLUT4 or rather AS160 in these age groups. (g) Investigation of Rab10 phosphorylation by Mn²⁺ Phos-tag SDS-PAGE in fibroblasts from 6 months old Lrrk2 deficient and wild-type rats at different time points (0-10-30-40 min) after insulin addition and (h) corresponding quantification of P-Rab10 signal intensity in wild-type cells at different time points after stimulation (normalized to Rab10, n = 7, mean and SEM).

Figure 6

Insulin signalling in human fibroblasts from Parkinson’s patients with G2019S mutated LRRK2. (a) Example: Western blot analysis of protein extracts from fibroblasts of one PD patient with G2019S mutation (top) and one healthy control individual (button) and (b) the corresponding quantification of P-Akt Thr308 (top) and P-Akt Ser473 (button) signal intensity (normalized to Akt, n = 7 independent experiments, fibroblasts from 3 different healthy controls and 3 PD patients with G2019S mutation in LRRK2; mean ± SEM). (c) GLUT4 immunostaining on the plasma membrane of human fibroblasts derived from PD patients with G2019S mutation and healthy control individuals (red: GLUT4 immunostaining; green: wheat germ agglutinin (=plasma membrane) staining; blue: DAPI) and (d) the corresponding quantification of GLUT4 signal intensity (mean ± SEM). (e) Quantification of total GLUT4, AS160 and Rab10 protein expression in human fibroblasts (normalized to tubulin) analysed by Western blot (mean and SEM).

Figure 7

The role of LRRK2 for insulin signal transduction. (a) In wild-type, GDP-bound Rab10 is tightly bound by guanine dissociation inhibitor (GDI) in the cytosol. LRRK2 promotes the insertion of Rab10 in GLUT4 storage vesicles by phosphorylation. Insulin addition activates an insulin-dependent Phosphatase X with following dephosphorylation of Rab10 (accompanied by GDP-GTP exchange), necessary for efficient GLUT4 translocation. (b) In LRRK2 deficient situation, the phosphorylation of Rab10 by LRRK2 and the following insertion in GLUT4 storage vesicles fail. As consequence, GDP-bound Rab10 – GDI complexes accumulate in the cytosol. By insulin addition activated PI3K/Akt/AS160 signalling cascade did not reach the GLUT4 storage vesicles and the GLUT4 translocation fails.