The Enzymatic Core of the Parkinson's Disease-Associated Protein LRRK2 Impairs Mitochondrial Biogenesis in Aging Yeast

Abstract

Mitochondrial dysfunction is a prominent trait of cellular decline during aging and intimately linked to neuronal degeneration during Parkinson's disease (PD). Various proteins associated with PD have been shown to differentially impact mitochondrial dynamics, quality control and function, including the leucine-rich repeat kinase 2 (LRRK2). Here, we demonstrate that high levels of the enzymatic core of human LRRK2, harboring GTPase as well as kinase activity, decreases mitochondrial mass via an impairment of mitochondrial biogenesis in aging yeast. We link mitochondrial depletion to a global downregulation of mitochondria-related gene transcripts and show that this catalytic core of LRRK2 localizes to mitochondria and selectively compromises respiratory chain complex IV formation. With progressing cellular age, this culminates in dissipation of mitochondrial transmembrane potential, decreased respiratory capacity, ATP depletion and generation of reactive oxygen species. Ultimately, the collapse of the mitochondrial network results in cell death. A point mutation in LRRK2 that increases the intrinsic GTPase activity diminishes mitochondrial impairment and consequently provides cytoprotection. In sum, we report that a downregulation of mitochondrial biogenesis rather than excessive degradation of mitochondria underlies the reduction of mitochondrial abundance induced by the enzymatic core of LRRK2 in aging yeast cells. Thus, our data provide a novel perspective for deciphering the causative mechanisms of LRRK2-associated PD pathology.

Keywords: LRRK2; Parkinson’s disease; aging; cell death; complex IV; mitochondria; neurodegeneration; yeast.

Figure 1

LRRK2^RCK triggers death of aging yeast cells. (A) Scheme of truncated leucine-rich repeat kinase 2 (LRRK2) constructs used in this study. The enzymatic core of human LRRK2 (amino acids 1300–2163) containing the Ras-of-complex (ROC) GTPase, the C-terminal-of-ROC (COR) and the protein kinase domain, was expressed in yeast cells during chronological aging under the control of a GAL1 promoter. The wild type form of truncated LRRK2 (hereinafter referred as LRRK2^RCK) and the point mutant R1398L^RCK with higher GTPase activity were used. The green star indicates GTPase activity, the star in magenta represents kinase activity. (B) Growth of control cells expressing LacZ compared to cells expressing LRRK2^RCK and the R1398L^RCK variant. OD₆₀₀ was measured in intervals of 2 h starting from the induction of galactose-driven expression. Means ± SEM; n = 4. (C) Flow cytometric quantification of loss of membrane integrity as indicated with propidium iodide (PI) staining of cells as described in (B). In addition, cells harboring the empty vector were analyzed to validate the suitability of LacZ expression as a control. Significances represent simple main effects between different expression types at each time point. Significances shown are valid for day 3–5. Means ± SEM; n = 4. (D) Clonogenic survival on day 3 of chronological aging determined by counting colony forming units (cfu) after plating 500 cells with indicated expression types on YEPD agar plates. Means ± SEM; n = 8. (E,F) AnnexinV/PI co-staining on day 3 of aging. Representative epifluorescence micrographs (E) and flow cytometric quantification (F) are shown. Scale bar represents 10 μm. Means ± SEM; n = 4. For AnnexinV/PI staining at earlier time points, please see Supplementary Figures S1D,E. (G) Immunoblot analysis of protein extracts from cells as described in (B). Blots were probed with antibodies directed against the V5-epitope to detect V5-tagged LacZ, LRRK2^RCK and R1398L^RCK, and against glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a loading control. ***p < 0.001, **p < 0.01, *p < 0.05, n.s. not significant.

Figure 2

LRRK2^RCK localizes in mitochondria and at the plasma membrane. (A) Representative confocal micrographs of strains harboring endogenously green fluorescent protein (GFP)-tagged Pma1 expressing mCherry alone or fused to LRRK2^RCK and R1398L^RCK on day 1 of aging. Z-projections of three-dimensional stacks as well as a representative section are shown. Scale bar represents 5 μm. For quantification of colocalization and confocal micrographs of strains harboring GFP-tagged Htb2, please see Supplementary Figures S2A,B. (B) Representative confocal micrographs of strains harboring endogenously GFP-tagged Om45, expressing the constructs as described in (A). Z-projections of three-dimensional stacks as well as a representative section are shown. Scale bar represents 5 μm. For quantification of colocalization please see Supplementary Figure S2B. (C) Representative immunoblots of subcellular fractionation of whole cell extracts on 12%–60% step sucrose gradients. Collected fractions were analyzed by immunoblotting. Blots were probed with antibodies directed against the V5-epitope to detect V5-tagged LRRK2^RCK, and against organelle-specific marker proteins as indicated. (D) Immunoblot analysis of purified mitochondria obtained from cells expressing LRRK2^RCK or LacZ on day 1. Prior to SDS-PAGE, samples were treated with depicted concentrations of proteinase K (ProtK). Blots were probed with antibodies directed against the V5-epitope, the outer mitochondrial membrane protein Tom70, the inner mitochondrial membrane proteins Tim54, Tim22 and Tim21, the mitochondrial matrix localized protease Mas1, as well as Mcr1, located in the outer membrane (OM) and intermembrane space (IMS).

Figure 3

LRRK2^RCK impairs mitochondrial function. (A) Oxygen consumption determined with a Fire-Sting optical oxygen sensor system in cells expressing LacZ, LRRK2^RCK or R1398L^RCK on day 1 and 2 of aging. Oxygen consumption was normalized to living cells and subsequently depicted as fold of control cells. Means ± SEM; n = 4. (B,C) Flow cytometric quantification of the reactive oxygen species (ROS)-driven conversion of non-fluorescent dihydroethidium (DHE) to fluorescent ethidium (Eth) in cells described in (A). Mean fluorescence intensities are shown as fold of control cells on day 1 (B). Dead cells, which accumulated Eth due to a loss of membrane integrity, were excluded from the analysis as shown in (C). Means ± SEM; n = 4. (D) Quantification of ATP content from cells described in (A) on day 2 of aging. Values were normalized to living cells and subsequently depicted as fold of control cells. Means ± SEM; n = 4. (E,F) Analysis of the mitochondrial transmembrane potential (ΔΨ_m) with Mitotracker CMXRos on day 2 of aging. Representative confocal micrographs (E) and flow cytometric quantification of the mean fluorescence intensity (F) are shown. Values for ΔΨ_m were normalized to control cells. Scale bar represents 5 μm. Means ± SEM; n = 4. (G) Flow cytometric quantification of loss of membrane integrity as indicated with PI staining of BY4741 and W303 strains expressing LacZ, LRRK2^RCK or R1398L^RCK on day 1 and day 2 of aging. Means ± SEM; n = 4. (H) Immunoblot analysis of protein extracts from cells as described in (G). Blots were probed with antibodies directed against the V5-epitope and against GAPDH as loading control. (I) Oxygen consumption determined with a Fire-Sting optical oxygen sensor system in cells as described in (G) on day 1 of aging experiments. Normalization was performed as described in (A). Means ± SEM; n = 4. ***p < 0.001, **p < 0.01, *p < 0.05, n.s. not significant.

Figure 4

Mitochondrial morphology and abundance is impaired by LRRK2^RCK. (A,B) Analysis of cells harboring endogenously GFP-tagged Om45 expressing LacZ, LRRK2^RCK or R1398L^RCK. Representative confocal micrographs on day 2 (A) and flow cytometric quantification of Om45-GFP fluorescence signal at day 1 and 2 (B) are shown. Dead cells were excluded via PI counterstaining. Values were normalized to control cells on day 1. Scale bar represents 5 μm. Means ± SEM; n = 4. For confocal micrographs of Om45-GFP strains on day 1, please see Supplementary Figure S3A. (C,D) Immunoblot analysis of extracts from cells as described in (A). Representative immunoblots (C) and densitometric quantification (D) are shown. Blots were probed with antibodies directed against the V5- and the GFP-epitopes, against the mitochondrial proteins Mdh1, Por1 and Tom22, and against glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as loading control. Values were normalized to the respective signals from control cells on day 1. Means ± SEM; n ≥ 6. For immunoblot analysis of Om45-GFP cells on day 1, and of Tim44-GFP strains, please see Supplementary Figures S3B–E. ***p < 0.001, **p < 0.01, *p < 0.05, n.s. not significant.

Figure 5

LRRK2^RCK compromises mitochondrial biogenesis and complex IV assembly. (A) Flow cytometric quantification of wild type and ATG11 deletion strains harboring Om45-GFP and expressing LacZ, LRRK2^RCK or R1398L^RCK on day 1 and 2 of aging. Dead cells were excluded via PI counterstaining. Values were normalized to control cells on day 1. Means ± SEM; n = 4. (B) Flow cytometric quantification of PI-stained wild type and ATG1, ATG11 and ATG32 deletion strains expressing LacZ or LRRK2^RCK on day 3 of aging. Means ± SEM; n = 4. (C) q-RT-PCR to determine mRNA levels of indicated mitochondria-related transcripts in cells expressing LacZ, LRRK2^RCK or R1398L^RCK. Normalization was performed using mRNA levels of UBC6. Means ± SEM; n = 4. (D) Immunoblot analysis of mitochondria isolated from cells expressing LacZ or LRRK2^RCK on day 1. Blots were probed with antibodies directed against the respiratory chain complex IV components Cox1, Cox2, Cox4 and Cox6, as well as Sdh1 (subunit of complex II), Tom70 (outer mitochondrial membrane) and Tim44 (inner mitochondrial membrane). (E) Immunoblots of Blue-Native-PAGE with samples described in (D). Blots were probed with antibodies against Cox6 (complex III/IV), Atp14 (complex V) and Tom22 (translocase of outer membrane (TOM) complex). (F) In-gel ATPase activity assay of samples described in (D). (G) Measurement of mitochondrial transmembrane potential (ΔΨ_m) in mitochondria isolated from cells described in (D). Values for ΔΨ_m were normalized to mitochondria from control cells. Means ± SEM; n = 3. ***p < 0.001, **p < 0.01, *p < 0.05, n.s. not significant.

Figure 6

Overview of LRRK2^RCK-induced mitochondrial dysfunction in yeast. In post-mitotic yeast cells, LRRK2^RCK (magenta) is targeted to mitochondria, leading to misassembly of respiratory chain complex IV (cytochrome c oxidase). Transcription of mitochondria (mt)-related gene products is reduced and a dissipation of the mitochondrial transmembrane potential (ΔΨ_m) can be observed. With increasing age, cells display a prominent decrease in mt mass and a concomitant drop in respiratory activity and ATP levels. In addition, mitochondrial dysfunction leads to enhanced production of ROS. This culminates in the collapse of the mt network, ultimately resulting in cell death.