Dysregulation of lysosomal morphology by pathogenic LRRK2 is corrected by two-pore channel 2 inhibition

1Two-pore channels (TPCs) are ubiquitous endolysosomal ion channels that mediate Ca2+ signals in response to the Ca2+ mobilising messenger nicotinic acid adenine dinucleotide phosphate (NAADP). The human isoforms, TPC1 and TPC2, target to discrete populations of acidic vesicles that comprise the endolysosomal system. These highly dynamic organelles undergo continual homo- and hetero-typic fusion in a Ca2+-dependent manner. Fusion of lysosomes with endosomes or autophagosomes is crucial for endocytosis and autophagy. Proper functioning of lysosomes is also dictated by their number and position within the cell. Lysosomal morphology might therefore serve as a sensitive read-out of endocytic well-being. Because TPCs regulate trafficking events within the endolysosomal system there is the possibility that aberrant TPC activity could underlie endocytic dysfunction.

2Parkinson disease is a progressive neurodegenerative disorder involving a complex aetiopathogenesis that includes several genetic causes and risk factors. Mutations in LRRK2 (also known as PARK8) are a cause of autosomal dominant familial Parkinson disease that is indistinguishable from sporadic forms. LRRK2 is a large modular protein comprising both enzymatic domains (a ROC and kinase domain) and domains involved in protein–protein interactions. The function of LRRK2 is not clear, but LRRK2 localises, at least in part, to the endolysosomal system, and a number of studies (albeit using recombinant systems and animal models) implicate LRRK2 in endolysosomal trafficking and associated processes such as endocytosis and autophagy.

3Here, we examined endolysosomal morphology in fibroblasts from Parkinson disease patients with the common LRRK2 G2019S mutation. We identify pronounced lysosomal morphology defects, which were reversed by inhibition of TPC2 and associated regulators. Our data thus suggest that TPC2 acts downstream of pathogenic LRRK2 to regulate trafficking within the endolysosomal system in a pathway of potential relevance to the pathology of LRRK2-mediated Parkinson disease.

4We examined the morphology of lysosomes in primary cultured fibroblasts from LRRK2 G2019S Parkinson disease patients (LRRK2-PD cells) using three independent methods. In the first analyses, we stained live cells with the fluorescent acidotrope, Lysotracker®. This probe can be used to infer lysosome volume, which has recently been validated as a novel biomarker for lysosomal storage disorders. In healthy control fibroblasts, lysosomes were well resolved as puncta (Fig. 1A). In contrast, lysosomes appeared enlarged and clustered in age-matched LRRK2-PD fibroblasts (Fig. 1B).

5In a second approach, we assessed lysosomal morphology in fixed cells by immunocytochemistry using a primary antibody raised to the late endosome and lysosome marker LAMP1. Again, lysosomes were enlarged in the patient fibroblasts (Fig. 1D) compared to controls (Fig. 1C). This defect was manifest as an approximate doubling in intensity of LAMP1 labelling in LRRK2-PD cells (Fig. 1E). We also noted a propensity of lysosomes to cluster close to the nucleus in the patient fibroblasts (Fig. 1D,E). Similar defects were obtained using cultures derived from three other patients when compared to healthy controls (supplementary material Fig. S1) although western blot analysis did not reveal any consistent change in total LAMP1 levels (supplementary material Fig. S2A).

6In the final approach, we performed electron microscopy to resolve the morphology of individual lysosomes. An example of a lysosome from a healthy control fibroblast displaying the typical electron dense interior is shown in Fig. 1F. Lysosomes in fibroblasts from Parkinson disease patients were more heterogeneous, and were often swollen and characterised by large translucent areas (Fig. 1F). Quantification of lysosomes in random sections showed that the average area was increased approximately sixfold whereas density was decreased approximately threefold (Fig. 1G).

7The G2019S mutation in LRRK2 falls within its kinase domain and is associated with increased kinase activity. We therefore examined the effect of LRRK2-In1, a recently described potent LRRK2 kinase inhibitor. As shown in Fig. 1H, lysosomal morphology in LRRK2-PD fibroblasts reverted to a normal appearance following a 3-day treatment with LRRK2-In1 (100 nM). Similar results were obtained upon shorter treatments (supplementary material Fig. S3). We also tested a structurally distinct LRRK2 kinase inhibitor, GSK2578215A. GSK2578215A (32 nM) also normalised lysosomal morphology in LRRK2-PD cells (Fig. 1I). Pooled data are presented in Fig. 1J. Taken together, we identified pronounced changes in lysosomal morphology in fibroblasts from LRRK2-associated Parkinson disease patients that are dependent on LRRK2 kinase activity.

8The observed changes in lysosomal morphology in LRRK2-PD fibroblasts described here are reminiscent of those recently described upon overexpression of TPC2. To probe the role of TPCs in LRRK2 action, we used small interfering RNAs (siRNAs) to silence TPC expression in LRRK2-PD fibroblasts. As shown in Fig. 2A–C, lysosomal morphology was normalised in LRRK2-PD fibroblasts transfected with a siRNA against TPC2. Quantitative PCR confirmed selective knockdown of TPC2 transcripts in siRNA-treated cells (Fig. 2F). Intriguingly, we noted little effect of TPC1 silencing on lysosomal morphology in LRRK2-PD fibroblasts (Fig. 2D,E) despite demonstrable knockdown at both the transcript (Fig. 2F) and protein (supplementary material Fig. S2B) level.

9Exaggerated perinuclear clustering of lysosomes in LRRK2-PD cells (Fig. 1) is consistent with the actions of both TPC2 and Rab7. We therefore tested the effects of inhibiting Rab7 GTPase activity on lysosomal morphology and distribution in LRRK2-PD fibroblasts. Importantly, lysosomal defects in LRRK2-PD cells were corrected upon Rab7 inhibition (Fig. 2G–K; supplementary material Fig. S3). This reversal was concentration dependent (Fig. 2L). Levels of endogenous VPS35, a component of the retromer complex, were unchanged in our LRRK2-PD fibroblasts compared to controls (supplementary material Fig. S2A). Taken together, data presented here reveal a specific role for TPC2 and a newly identified interactor (Rab 7) in regulating lysosomal morphology in LRRK2-PD cells.

10Much evidence has accumulated identifying TPCs as the long-sought endolysosomal targets for NAADP. Notably, NAADP-induced Ca2+ release is inhibited when TPCs are silenced or genetically deleted. However, this view has been challenged by evidence suggesting that TPCs are not NAADP-sensitive channels but are instead Na+ channels gated by phosphatidylinositol 3,5-bisphosphate [PtdIns(3,5)P2]. We therefore examined the effects of antagonising the action of NAADP and PtdIns(3,5)P2 on lysosomal morphology in LRRK2-PD fibroblasts (Fig. 3A). Potential NAADP involvement was assessed using the NAADP antagonist Ned-19 and its novel analogue Ned-K (see Materials and Methods). As shown in Fig. 3B–F and supplementary material Fig. S3, lysosomal morphology in LRRK2-PD fibroblasts treated with either Ned-19 or Ned-K was similar to control cells. Reversal of defective lysosomal morphology by both compounds was dependent on concentration (Fig. 3G). Ned-19 and Ned-K thus mimicked the effects of TPC2 silencing (Fig. 2). We used the PIKfyve inhibitor, YM-201636 to deplete PtdIns(3,5)P2 levels. Acute treatment with the drug was also sufficient to reverse lysosomal morphology defects (Fig. 3H–J), again similar to the effects of silencing TPC2. Collectively, these molecular and pharmacological data suggest that TPCs are regulated by both NAADP and PtdIns(3,5)P2, findings consistent with recent reports. These data also highlight the utility of mechanistically distinct small molecule inhibitors in correcting lysosomal pathology.

11TPCs are ostensibly Ca2+-permeable, and constitutive Ca2+ release events within the endolysosomal system are known to regulate organelle fusion. To probe the role of Ca2+ in lysosomal disturbances, we buffered Ca2+ levels using cell-permeable forms of either BAPTA or EGTA. As shown in Fig. 4A–C and summarised in Fig. 4E, lysosomal morphology in LRRK2-PD fibroblasts acutely treated with BAPTA-AM were similar to control cells, suggesting that the morphological defects are dependent on Ca2+. However, treatment with EGTA-AM (a slower Ca2+ chelator) proved ineffectual (Fig. 4D,E). These data indicate that disrupted lysosomal morphology is likely due to dysregulated local Ca2+ signalling. This might promote Ca2+-dependent fusion of lysosomes and thus enlargement. Indeed, we often encountered large hourglass-shaped organelles delineated by a continuous membrane consistent with a fusion defect in Parkinson disease fibroblasts (supplementary material Fig. S4).

12To further probe the role of Ca2+ in pathogenic LRRK2 action, we measured global cytosolic Ca2+ levels in response to NAADP stimulation. As shown in Fig. 4F, microinjection of fibroblasts with NAADP, but not vehicle, evoked Ca2+ signals. NAADP responses were significantly larger in LRRK2-PD fibroblasts than in healthy controls (Fig. 4G,H). These signals are likely the global correlate of enhanced TPC activity that underlies the trafficking defect. Thus, both local (constitutive) and global (NAADP-regulated) Ca2+ signals are disrupted upon LRRK2 mutation.

13In summary, we identify trafficking defects in LRRK2-PD fibroblasts that result from interplay between TPC2, its regulatory interactors and ligands, and the associated Ca2+ fluxes. TPC2 and LRRK2 co-immunoprecipitate raising the possibility that TPC2 is phosphorylated by LRRK2. Given that similar morphological changes within the endo-lysosomal system have been observed during normal ageing and in other neurodegenerative conditions, we suggest that aberrant TPC signalling might have broader relevance to declining cellular function.

14Fibroblast cultures from four Parkinson disease patients carrying the G2019S mutation in LRRK2 and four healthy donors (each pair age-matched within 2 years; age range 48–71) were established as described previously. The study was approved by the Hampstead Research Ethics Committee. All individuals provided written informed consent for the provision of samples. Cells were maintained in DMEM supplemented with 10% (v/v) fetal bovine serum, 100 units/ml penicillin and 100 µg/ml streptomycin (all from Invitrogen) at 37°C in a humidified atmosphere with 5% CO2. Cells were passaged by scraping and plated onto glass coverslips (for confocal microscopy and Ca2+ imaging), thermanox coverslips (for electron microscopy) or directly onto tissue culture plates or flasks (for western blotting) before experimentation. Cultures were used between passage 6–15 and at 4–7 days post plating. Control and LRRK2-PD cultures were analysed in parallel and differed by no more than two passages.

15All drugs used in this study were dissolved in DMSO, diluted into culture medium and the medium sterile filtered prior to use. The LRRK2 kinase inhibitors LRRK2-In1 and GSK2578215A were from Merck and R&D Systems, respectively. The Rab7 GTPase inhibitor, CID 1067700 was from EMD Millipore. The NAADP antagonist trans-Ned-19 was synthesised as described previously (Naylor et al., 2009). Ned-K is an analogue of Ned-19 in which the fluoride has been replaced with a cyano group. Its synthesis will be described elsewhere. Both Ned-19 and Ned-K were kind gifts from A. Ganesan (School of Pharmacy, University of East Anglia, UK), Raj Gossain (School of Chemistry, University of Southampton, UK) and Sean M. Davidson (Hatter Institute, UCL, UK). The PIKfyve inhibitor, YM-201636 was from Cambridge Bioscience. BAPTA-AM and EGTA-AM were from Sigma.

16Fibroblasts were transfected with siRNAs using Lipofectamine® RNAiMAX for 24 h, re-transfected for an additional 24 h and cultured for a final 24 h in the absence of siRNA prior to experimentation. A control siRNA duplex (Allstars Negative Control siRNA) and duplexes targeting human TPC1 (5′-CGAGCTGTATTTCATCATGAA-3′) and TPC2 (5′-CAGGTGGGACCTCTGCATTGA-3′) were purchased from Qiagen.

17Fibroblasts were washed three times in HEPES-buffered saline (HBS) comprising (in mM) 1.25 KH2PO4, 2 CaCl2, 2 MgSO4, 3 KCl, 156 NaCl, 10 glucose and 10 HEPES (pH 7.4; all from Sigma) and were then incubated with 100 nM Lysotracker® red (Invitrogen) for 20 min. Cells were washed again three times in HBS, and mounted in an imaging chamber (Biosciences Tools) prior to confocal microscopy.

18Fibroblasts were fixed for 10 min with 4% (w/v) paraformaldehyde, washed three times in phosphate-buffered saline (PBS) and then permeabilised for 10 min with 40 µM β-escin. Cells were washed again (three times in PBS), and blocked for 1 h with PBS supplemented with 1% (w/v) BSA and 10% (v/v) FBS. Fibroblasts were sequentially incubated for 1 h at 37°C with a primary anti-LAMP1 antibody (mouse, Developmental Studies Hybridoma Bank H4A3 clone supernatant; 1:10 dilution) and a secondary antibody conjugated to Alexa Fluor 647 (mouse, Invitrogen, 1:100 dilution) in blocking solution. Nuclei were labelled with 1 µg/ml DAPI (5 min). Cells were washed three times in PBS containing 0.1% (v/v) Tween® 20 in between incubations and mounted onto microscope slides with DABCO.

19Confocal images were captured using an LSM510 confocal scanner (Zeiss) attached to a Zeiss Axiovert 200M inverted microscope fitted with a 63× Plan Apochromat water-immersion objective. DAPI, Lysotracker® Red and Alexa Fluor 647 were excited at 364 nm, 543 nm and 633 nm, and emitted fluorescence captured using 385 nm long pass, 585–615 nm band-pass or 655–719 nm band-pass filters, respectively. Images for control and LRRK2-PD cells together with the various treatments were captured under identical acquisition settings in order to allow comparison of fluorescent intensity. Electron microscopy was performed as described previously using a JEOL 1010 transmission electron microscope.

20Total RNA was isolated using TRIzol® (Invitrogen) according to the manufacturer's procedures. cDNA was synthesised using SuperScript® III reverse transcriptase (Invitrogen). Samples were denatured for 2 min at 94°C followed by 40 cycles of denaturation (15 s, 94°C), annealing (30 s, 60°C) and extension (30 s, 72°C) using SYBR® Green PCR mix (Invitrogen) and oligonucleotide primers designed for human TPC1 and TPC2 as previously described. Expression levels were normalised to the expression of GAPDH following parallel amplification.

21Fibroblasts were harvested by scraping and lysed in Ripa buffer containing 150 mM NaCl, 50 mM Tris-HCl (pH 7.4), 0.5% sodium deoxycholic acid, 0.1% sodium dodecyl sulphate and 1% Triton X-100 in the presence of EDTA-free protease inhibitor (Roche) and Halt™ phosphatase inhibitor cocktail (Thermo Scientific) for 30 min on ice. Samples were centrifuged at 15,000 g at 4°C for 15 min and the resulting supernatants stored at −20°C until required. Samples (10–30 µg) were reduced with dithiothreitol (100 mM), separated on NuPAGE® 4–12% Bis-Tris gels (Invitrogen) and transferred onto PVDF filters (Biorad) according to standard procedures. The filters were then blocked with 5% (w/v) dried skimmed milk in Tris-buffered saline (25 mM Tris-HCl, 137 mM NaCl and 2.7 mM KCl, pH 7.4) containing 0.1% (v/v) Tween® 20 (TBS-T) for either 1 h at room temperature or overnight at 4°C. Blots were sequentially incubated with primary and secondary antibodies in TBS-T supplemented with 2.5% (w/v) dried skimmed milk. After each step, the filters were washed with TBS-T (3×30 min). The resulting blots were developed using the ECL™ Prime Western Blot Detection System (GE Healthcare) according to the manufacturer's instructions. The primary antibodies used were anti-LAMP1 (mouse, Santa Cruz Biotechnology; 1:500, overnight 4°C), anti-TPC1 (rabbit, Abcam, 1:200, 1 h room temperature), anti-VPS35 (rabbit, Abcam, 1:1000, overnight 4°C) and anti-actin (goat, Invitrogen, 1:500, 1 h room temperature) antibodies. The secondary antibodies used were anti-mouse-IgG (Santa Cruz Biotechnology), anti-rabbit-IgG (Bio-Rad) or anti-goat-IgG (Santa Cruz Biotechnology) conjugated to horseradish peroxidase (1:2000, 1 h room temperature).

22Cytosolic Ca2+ concentration measurements using Fura-2 and microinjection were performed as described previously.

23Images were analysed using ImageJ software. For Lysotracker® red and LAMP1 intensity measurements, background was subtracted from the images and mean grey intensity per cell measured within user defined regions-of-interest (comprising the whole lysosome population). Statistical analyses were performed using IBM SPSS statistics 22 software. Independent Student's t-tests or one-way ANOVA followed by Games–Howell post hoc tests were applied to calculate statistical significance. Values are presented as mean±s.e.m. For ANOVA analysis, threshold of significance was maintained at P<0.016 to correct for multiple testing error.