WO2008091799A2 - Cell-based methods for identifying inhibitors of parkinson's disease-associated lrrk2 mutants - Google Patents

Abstract

Mutations in leucine-rich repeat kinase-2 (LRRK2) are the most common genetic cause of Parkinson's disease (PD). PD-associated LRRK2 mutants induce the formation of intracellular filaments which can be visualized and quantified. The invention provides cell- based screening assays and methods for identifying inhibitors of PD-associated LRRK2 mutants. Inhibitors can be identified by assessing intracellular distribution of LRRK2 protein. Identified inhibitors are potential therapeutic agents which may be useful for treating PD or other neurodegenerative disorders. The methods of the invention were used to identify 14-3-3Θ and the inhibitor of HSP90 geldanamycin, as inhibitors of LRRK2 toxicity in cells. The invention provides methods for preventing or decreasing neurotoxicity in a subject by administering a LRRK2 inhibitor.

Description

CELL-BASED METHODS FOR IDENTIFYING INHIBITORS OF PARKINSON'S DISEASE-ASSOCIATED LRRK2 MUTANTS

[0001] The invention disclosed herein was made with U.S. Government support under

NIH Grant No. K02 NS045798-01 Al from the NINDS. Accordingly, the U.S. Government may have certain rights in this invention.

[0002] All patents, patent applications and publications cited herein are hereby incorporated by reference in their entirety. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described and claimed herein.

[0003] A portion of the disclosure of this patent document contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.

BACKGROUND OF THE INVENTION

[0004] Neurodegenerative illnesses such as Parkinson's disease (PD) and

Alzheimer's disease are a common and increasingly prevalent problem in aging societies. Yet the mechanisms that underlie cell death in these diseases are incompletely understood, and no therapies exist that retard or prevent neurodegeneration.

[0005] PD is characterized by motor and cognitive dysfunction reflecting widespread neurodegeneration, particularly of midbrain dopaminergic neurons . Although PD is typically a sporadic illness, there is growing recognition that genetic susceptibility plays an important role. Further, the discovery of mutations underlying rare inherited forms of PD has shed light onto the molecular mechanisms that contribute to the sporadic disease⁹.

[0006] Despite this progress, much remains unknown about the molecular mechanisms underlying PD. Dominant missense mutations in leucine-rich repeat kinase 2 (LRRK2) are the most common genetic cause of Parkinson's disease (PD)^1"7, and patients with LRRK2-related PD exhibit neuropathology that is indistinguishable from the sporadic disease. Thus, it is important to better define the biology of LRRK2 to uncover signaling networks of particular importance to neurons and impact an array of patients with neurodegenerative illnesses. SUMMARY OF THE INVENTION

[0007] In one aspect, the invention provides a method for determining whether a compound inhibits mutant leucine-rich repeat kinase-2 (LRRK2) kinase activity, the method comprising (a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell; (b) contacting the cell with a compound; and (c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound, wherein determination of a reduction of filamentous distribution in (c) indicates that the compound inhibits the LRRK2 mutant protein activity.

[0008] In another aspect, the invention provides a method for determining whether a compound inhibits mutant leucine-rich repeat kinase-2 (LRRK2) GTPase activity, the method comprising (a) contacting a cell with a compound; (b) expressing a Parkinson's Disease- associated LRRK2 mutant protein in the cell; and (c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound, wherein determination of a reduction of filamentous distribution in (c) indicates that the compound inhibits the LRRK2 mutant protein activity.

[0009] In another aspect, the invention provides a method for determining whether a compound inhibits LRRK2 toxicity in a cell, the method comprising (a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell; (b) contacting the cell with a compound; (c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound; and (d) correlating LRRK2 toxicity to filamentous distribution of the mutant protein, wherein determination of a reduction of filamentous distribution in (c) and a decrease in toxicity in (d) indicates that the compound inhibits LRRK2 toxicity in a cell.

[0010] In one embodiment, determining a reduction of filamentous distribution comprises detecting a change in diffuse distribution of the protein, punctuate distribution of the protein, filamentous distribution of the protein, or any combination thereof in the cell. In another embodiment, the determining comprises computer-assisted quantification of filamentous distribution of the protein. In another embodiment, the determining comprises computer-assisted quantification of diffuse distribution of the protein. In another embodiment, the determining comprises computer-assisted quantification of punctate distribution of the protein.

[0011] In one aspect, the invention provides a method for determining whether a compound enhances interaction between a Parkinson's Disease-associated LRRK2 mutant protein and a 14-3-3 peptide, the method comprising (a) co-expressing a Parkinson's Disease- associated LRRK2 mutant protein and a 14-3-3 peptide in a cell; (b) contacting the cell with a compound; and (c) determining whether interaction of the LRRK2 mutant protein and the 14- 3-3 peptide is enhanced in the cell compared to in a cell co-expressing the LR.RK2 mutant protein and the 14-3-3 peptide in the absence of the compound. In one embodiment, the 14- 3-3 peptide is 14-3-3Θ. In another embodiment, the 14-3-3 peptide is linked to a detectable tag. In another embodiment, the determining comprises fluorescent resonant energy transfer (FRET).

[0012] The embodiments of the invention encompass the use of all 14-3-3 isoforms.

In one embodiment, the 14-3-3 peptide is a mammalian isoform. In another embodiment, the 14-3-3 peptide is a murine isoform (SEQ ID NO:1). In another embodiment, the 14-3-3 peptide is a human isoform (SEQ ID NO:2).

[0013] In one embodiment of the methods of the invention, the cell comprises a nucleic acid vector capable of expressing a Parkinson's Disease-associated LRRK2 mutant protein. In other embodiments, the LRRK2 mutant protein comprises a R 1441C mutation, a R1441G mutation, a Y1699C mutation, a I2020T mutation, a G2019S mutation or any combination thereof. All LRRK2 mutations described herein are based on the amino acid sequence of human LRRK2 (SEQ ID NO:3, Figures 21 A-21B).

[0014] In another embodiment, the LRRK2 mutant protein is linked to a detectable tag. In one embodiment, the detectable tag is an epitope tag. In another embodiment, the detectable tag is a fluorescent protein. In another embodiment, the determining comprises detecting a detectable tag. hi another embodiment, the determining comprises detecting fluorescence. In another embodiment, fluorescence is detected directly. In another embodiment, fluorescence is detected indirectly.

[0015] In one embodiment, the cell is a primary neuron. In another embodiment, the cell is a catecholaminergic CAD cell. In another embodiment, the cell is a HeLa cell, a human embryonic kidney (HEK) cell, a baby hamster kidney (BHK) cell, a ShSy5y cell, or a PCl 2 cell.

[0016] In one embodiment, the compound is a small molecule, a polypeptide, a protein, a peptide, a peptidomimetic, a nucleic acid, an RNA, a DNA, an antisense RNA a small interfering RNA (siRNA), a double stranded RNA (dsRNA), a short hairpin RNA, a cDNA, or any combination thereof.

[0017] In another embodiment, the method is carried out in a high-throughput manner. In another embodiment, the method is carried out for more than 100 compounds. In another embodiment, the method is carried out in a multi-well plate.

[0018] In another aspect, the invention provides a method for preventing or decreasing neurotoxicity in a subject, the method comprising administering an effective amount of an inhibitor of a LRRK2 mutant protein, wherein the inhibitor decreases LRRK2 filament formation in the subject.

[0019] In one embodiment, the LRRK2 inhibitor comprises a 14-3-3 peptide, or a fragment thereof. In another embodiment, the 14-3-3 peptide is a 14-3-3Θ (for example, SEQ ID NO:1 or 2). In one embodiment, the 14-3-3 peptide fragment comprises an amino acid sequence that mediates binding of 14-3-3 to the phosphomotif on LRRK2. In another embodiment, the inhibitor comprises a nucleic acid capable of expressing a 14-3-3 peptide (see, for example, GenBank Accession Nos. X56468 and U57312), or a fragment thereof. In one embodiment, the inhibitor comprises an inhibitor of HSP90. In another embodiment, the HSP90 inhibitor is gledanamycin, radicicol, radamycin or novobiocin. In another embodiment, the HSP90 inhibitor is 17-allylamino-17-demethoxygeldanamyem (17 AAG), 17-dimethylaminoethylamino-17-demethoxygeldanamycin (17DM AG), a radici col-related oxime derivative, beta-zearalenol, PU24FC1, CCTOl 8159, GA dimer, GA-testosterone, GA- oestrogen, coumermycin, cisplatin, depsipeptide or suberoylanilide hydroxamic acid (SAHA).

BRIEF DESCRIPTION OF THE FIGURES

[0020] Figures IA - IE. LRRK2 kinase function is required for Parkinson's disease mutations to enhance neurodegeneration, but kinase activity and neurotoxicity are not correlated. Figure IA, Schematic of LRRK2 domain organization. GTP = Ras-like GTPase. Figure IB, Parkinson's disease mutant forms of LRRK2 cause apoptotic cell death of primary neurons. Figure 1C, The neurotoxicity of PD mutant forms of LRRK2 requires the kinase function of LRRK2. The double mutations (e.g., RC/KR) indicate doubly mutated molecules. Figure ID, Autoradiograms of in vitro autophosphorylation assays of WT and PD mutant forms of LRRK2 show that many LRRK2 PD mutations do not exhibit enhanced kinase activity. Figure IE, Quantification of LRRK2 in vitro autophosphorylation assays. Data are from 4 - 6 independent experiments. Error bars = S. E.; *p<0.05, ** p<0.01, *** pO.OOl.

[0021] Figures 2 A - 2E. LRRK2 filaments are enhanced by multiple Parkinson's disease mutations and require kinase activity. Figure 2 A, Patterns of WT-GFP-LRRK2 subcellular localization in the neuronal dopaminergic CAD cell line with bar graph depicting relative frequencies of each pattern. Figure 2B, The formation of LRRK2 aggregates is not affected by PD mutations or the loss of kinase activity. Figure 2C, The formation of LRRK2 filaments is enhanced by multiple PD mutations. Figure 2D, Immunofluorescent images of LRRK2 transfected into CAD and HeLa cells. The lower left image is reconstructed from a confocal Z-stack; green = LRRK2, red = cytosolic β-galactosidase. Figure 2E, Blocking LRRK2 kinase activity inhibits LRRK2 filament formation. Error bars = S.E.; ** p<0.01, *** pO.OOl.

[0022] Figures 3A - 3E. Multiple Parkinson's disease mutations enhance LRRK2 oligomerization. Figure 3A, LRRK2 homo-oligomerization is enhanced by PD mutations. Pull-downs of LRRK2 in CAD cells cotransfected with differentially tagged (V5 and GFP) forms of the same WT or PD mutant LRRK2 allele. Figure 3B, LRRK2 is able to hetero- oligomerize. Pull-downs of cotransfected differentially tagged WT and I2020T-LRRK2 in CAD cells. For each co-transfection, two different amounts (IX and 1.5X) of immunoprecipitate are shown. Figure 3C, Kinase deficient LRRK2 inhibits the formation of filaments by PD mutant LRRK2. Quantification of filament formation (using anti-GFP antisera) in cells transfected with different alleles of differentially-tagged LRRK2. Figure 3D, Representative images of cells selected for analysis using fluorescence correlational spectroscopy. Selected cells expressed the lowest visible levels of LRRK2 (direct fluorescence) and showed no evidence of filament formation. Figure 3E, The mobility of LRRK2 in live cells reflects its oligomeric state. Diffusion constant (τ) of wild type and mutant forms of fluorescently-tagged forms of LRRK2. Error bars = S. E.; ** p<0.01, *** pO.OOl. [0023] Figures 4 A - 4E. 14-3-3Θ blocks LRRK2 oligomerization. Figure 4A,

LRRK2 co-IPs with endogenous 14-3-3Θ. Figure 4B, Multiple LRRK2 PD mutants co-IP similar amounts of cotransfected 14-3-3Θ. Figure 4C, 14-3-3Θ inhibits the oligomerization of WT and mutant forms of LRRK2. WT-, Rl 441 C- and I2020T-GFP-LRRK2 were cotransfected with 14-3-3Θ or β-galactosidase. Figure 4D, The co-IP of 14-3-3Θ and LRRK2 involves a phosphate-dependent interaction. KE denotes the mutation of a key lysine residue (K49) involved in the binding of 14-3-3 proteins to phosphate. Figure 4E, The oligomerization-inhibiting activity of 14-3-3Θ involves a phosphate-dependent interaction. Error bars = S.E.; *** pO.001.

[0024] Figures 5A - 5C. LRRK2 oligomerization provokes neurodegeneration.

Figure 5A, The WD40 domain is required for LRRK2 filament formation. Figure 5B, The WD40 domain is required for LRRK2 autophosphorylation. Autoradiogram of in vitro autophosphorylation assay of I2020T-GFP-LRRK2 fragments. Figure 5C, The phosphorylation of LRRK2 occurs via autophosphorylation. Autoradiogram of in vitro kinase assay of the ROC-WD-IT LRRK2 fragment alone or together with kinase deficient Kl 906R- GFP-LRRK2. Error bars = S.E.; *** pO.001.

[0025] Figure 6. The Kl 906R mutation blocks LRRK2 kinase activity.

Autoradiogram of in vitro autophosphorylation assay of WT and K1906R GFP -tagged LRRK2. Equal amounts of immunoprecipitated LRRK2 were subjected to kinase assays, analyzed by SDS-PAGE, and exposed to films. The upper and the lower panels are autophosphorylation of LRRK2 and the corresponding silver-stained bands, respectively.

[0026] Figures 7A - 7B. LRRK2 filaments form in neurons and with a C-terminal V5 tag. Figure 7A, Neuron transfected with I2020T-GFP-LRRK2 exhibiting numerous discrete filamentous structures when immunostained with anti-GFP antisera. Green = I2020T-GFP- LRRK2; Blue = DAPI. Figure 7B, C-terminally tagged I2020T forms filaments in CAD cells similar to those observed with GFP-LRRK2 (as shown in Figure 2). Red = I2020T-V5- LRRK2; Blue = DAPI.

[0027] Figures 8A - 8B. LRRK2 expression levels are unaffected by PD or kinase- inactivating mutations. Figure 8A, Western immunoblot showing expression of N-terminal GFP -tagged WT or disease mutant LRRK2 in murine dopaminergic CAD cell lines using anti-GFP antisera. Equal protein loading was confirmed using anti-GAPDH antisera. Figure 8B, Western immunoblot showing expression of N-terminal GFP-tagged WT or disease mutant LRRK2, or disease mutant plus K.1906R (kinase inactive) double mutants in CAD cells using anti-GFP antisera. Equal protein loading was confirmed using anti-GAPDH antisera.

[0028] Figure 9. LRRK2 filaments do not colocalize filamentous cellular structures.

HeLa cells transiently transfected with I2020T-GFP-LRRK2 were double stained with anti- GFP (left panels) and Mito-Tracker, rhodamine-phalloidin, anti-α-tubulin, or anti-pan cytokeratin (intermediate filaments) (middle panels). The merged images (right panels) illustrate that there is no significant colocalization of I2020T-GFP-LRRK2 with any of these markers.

[0029] Figure 10. Autocorrelation curves of WT and I2020T-LRRK2. Fluorescently tagged WT- and I2020T-LRRK2 display differences in diffusion that parallel their ability to form filaments. The autocorrelation curves were generated as described in Example 7. Similar curves were generated for all other mutants shown in Figure 3 E.

[0030] Figures 1 IA - 11C. The inhibition of LRRK2 filament formation by 14-3-

3Θ is not due to reduced LRRK2 protein expression. Figure HA, Silver stained gels of anti- GFP immunoprecipitate from CAD cells transfected with GFP or GFP-LRRK2. Figure HB, Lysates of CAD cells cotransfected with GFP-LRRK2 (WT, R1441C or I2020T) and 14-3- 3Θ or a mock control (β-gal) were analyzed by immunoblotting with anti-GFP or anti-14-3- 3Θ antibodies. Co-expression of 14-3 -3 θ did not effect LRRK2 level compared to the mock control, β-actin blot was used to ensure equal loading. Figure HC, Immunoblots of CAD cells coexpressing GFP-LRRK2 (WT or I2020T) and 14-3-3Θ (WT or K49E) or the mock control (β-gal).

[0031] Figures 12A - 12C. The WD40 domain is necessary and sufficient for

LRRK2 oligomerization. Figure 12 A, Deletion of the WD40 domain does not significantly alter the kinase activity of LRRK2. Equal amounts of full-length (FL) and ΔWD WT-LRRK2 were immunoprecipitated for in vitro kinase assays using myelin basic protein (MBP) as a substrate. The reaction mixtures were separated by SDS-PAGE and the MBP bands were excised for liquid scintillation counting. Error bar = S.E.M. Figure 12B, Immunoblot of lysates from CAD cells expressing GFP-I2020T-LRRK2 fragments. Figure 12C, The WD40 domain is necessary and sufficient for LRRK2 filament formation. CAD cells expressing GFP-I2020T- LRRK2 fragments were immunostained with anti-GFP. ΔWD-IT exhibited a diffuse cytosolic pattern, while Roc- WD-IT and FL-IT show cytoplasmic filaments. [0032] Figure 13. Immunoprecipitation results indicate that LRRK2 is phosphorylated by another kinase.

[0033] Figure 14. Examples of embodiments of the invention. An RNAi screen or small molecule screen can be carried out to identify molecules that prevent or inhibit LRRK2 toxicity.

[0034] Figure 15. Amino acid sequence of human 14-3-3Θ (SEQ ID NO:1). The amino acid sequence has GenBank Accession No. P27348.

[0035] Figure 16. Amino acid sequence of mouse 14-3-3Θ (SEQ ID NO:2). The amino acid sequence has GenBank Accession No. AAC53257.

[0036] Figure 17. The HSP90 inhibitor geldanamycin decreases LRRK2 filament formation. Geldanamycin decreases the percentage of cells with filaments in a concentration- dependent manner (1 nM, 50 nM, 100 nM) in CAD cells expressing wildtype (WT) LRRK2 and mutant I2020T (IT) LRRK2.

[0037] Figure 18. Examples of HSP90 inhibitors which can be used in the methods of the invention.

[0038] Figures 19 A - 19C. Multiple sequence alignments of the WD40 domain of

LRRK2 proteins from human (see GenBank Accession No. NP_940980), chimpanzee (chimp) (GenBank Accession No. XP_001168494), dog (see GenBank Accession No. XP_545823 ), mouse (see GenBank Accession No. NP_080006), rat (see GenBank Accession No. XP 001057114), tetraodon (terra) (see GenBank Accession No. CAG05593), and drosophila (drome) (see GenBank Accession No. ABF29833) by three alignment methods: ClustalX {Nucleic Acids Research, 1997, 24:4876-4882) (Figure 19A), PROBCONS {Genome Research, 2005, 15:330-340) (Figure 19B), and SPEM (Figure 19C). SPEM was shown to be better than the other two methods in large benchmark tests {Bioinformatics, 2005, 21 :3615-3621). The first residue in the human sequence corresponds to residue 2016 in the full sequence of human LRRK2 (GenBank Accession No. NP_940980). Triangles indicate basic residues identified as conserved in all of the three homology models. Three characters ('*', ':', and '.') are used to mark positions conserved in sequence. '*' indicates positions which have a single, fully conserved residue. ':' indicates that one of the following 'strong' groups is fully conserved: STA, NEQK, NHQK, NDEQ, QHRK, MILV, MILF, HY, FYW. '.' Indicates that one of the following 'weaker' groups is fully conserved: CSA, ATV, SAG, STNK, STPA, SGND, SNDEQK, NDEQHK, NEQHRK, FVLIM, HFY. [0039] Figure 20. Three-dimensional model of a LRRK2 WD40 domain highlighting the conserved basic amino acid residues identified in the multiple sequence alignments shown in Figures 19A-19C.

[0040] Figures 21A - 21B. Amino acid sequence of human LRRK2 (SEQ ID NO:3).

The amino acid sequence has GenBank Accession No. NP_940980.

DETAILED DESCRIPTION OF THE INVENTION

[0041] The invention provides a discovery that expression of one or more Parkinson's disease (PD)-associated LRRK2 mutants in a cell results in the formation of filaments in the cell. The invention also provides a cell-based assay which comprises expression of one or more Parkinson's disease-associated LRRK2 mutants that can be used to identify molecules that treat or prevent neurotoxicity.

[0042] LRRK2 is a large protein that contains LRR (leucine-rich repeat) and WD40 protein-protein interaction domains, and an evolutionarily conserved multifunctional "ROCO" cassette defined by the presence of Ras-like GTPase and kinase domains⁶' ^{7> 10} (Figure IA), suggesting that it may integrate signaling information. LRRK2 mutations are also associated with neuropathology characteristic of other neurodegenerative diseases⁷. The kinase activity of LRRK2 may be critical in causing cell death^11"13, thus targeting this activity might be a potential therapeutic option. Insights into the biology of LRRK2 may uncover a signaling network of particular importance to neurons and which has implications for patients with diverse neurodegenerative illnesses.

[0043] LRRK2 mutations are the most common genetic cause of PD. PD caused by

LRRK2 mutations is clinically and pathologically indistinguishable from typical (sporadic) PD; this is not true for other mutated genes that have been linked to PD (e.g., α-synuclein, parkin, UCH-Ll, PINKl and DJ-I). Five dominantly inherited missense mutations in LRRK2 have been identified as being associated with PD⁶'⁷ (Figure IA): two mutations in the GTPase domain (R 144 IG, R 1441C), one mutation between the GTPase and kinase domains (Yl 699C), and two mutations in the kinase domain (G2019S and I2020T). LRRK2 mutations are the cause of at least 2-5% of all PD cases and up to 30% of PD cases in certain populations (e.g., Ashkenazie Jews). In addition to presently known PD-associated LRRK2 mutations, PD-associated LRRK2 mutations identified in the future are included in the invention. Any PD-associated LRRK2 mutation can be used in the methods provided by the invention.

[0044] The five PD-causing mutations in LRRK2 are similarly effective in causing cell death when expressed in neurons (see Example 1). Although kinase activity is required for all of these mutations to cause neurodegeneration, most mutants exhibit normal kinase activity. Thus, an effect other than direct kinase activation underlies the neurotoxicity of many LRRK2 PD mutations.

[0045] The invention provides cell-based methods to identify and functionally characterize molecules that interact with and alter the function of LRRK2 in a manner that will reduce its ability to mediate neurodegeneration.

Cell-based screening methods

[0046] Four PD-linked mutations with normal or modestly elevated kinase activity enhance LRRK2 oligomerization, which is visible as long filamentous structures when expressed in cell lines or primary neurons (see Example 2). The formation of filamentous oligomers is linked to neurotoxicity because blocking filament formation prevents cell death. For example, LRRK2 oligomerizes via its WD40-containing C-terminus, and deleting this domain blocks oligomerization and LRRK2-induced neurodegeneration.

[0047] The invention provides methods for determining whether a compound inhibits the formation of filaments. Using the methods of the invention, one can determine whether a compound prevents the formation of filaments or causes the dissolution of pre-formed filaments. One aspect of the invention provides a method for determining whether a compound inhibits mutant leucine-rich repeat kinase-2 (LRRK2) protein activity, the method comprising (a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell; (b) contacting the cell with a compound; and (c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound, wherein determination of a reduction of filamentous distribution in (c) indicates that the compound inhibits the LRRK2 mutant protein activity. In this aspect of the invention, one can determine whether a compound causes dissolution of filaments that are formed in the cell prior to contacting the cell with the compound. [0048] In another aspect of the invention, the step of contacting the cell with a compound can be performed prior to the step of expressing a Parkinson's Disease-associated LRRK2 mutant protein in the cell. In this aspect, one can determine whether a compound prevents the formation of filaments in the cell.

[0049] In another aspect, the invention provides a method for determining whether a compound inhibits LRRK2 toxicity in a cell, the method comprising (a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell; (b) contacting the cell with a compound; (c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound; and (d) correlating LRRK2 toxicity to filamentous distribution of the mutant protein, wherein determination of a reduction of filamentous distribution in (c) and a decrease in toxicity in (d) indicates that the compound inhibits LRRK2 toxicity in a cell.

[0050] The cell-based screening methods provided by the invention are based on the discovery that PD-associated LRRO mutants induce the formation of intracellular filaments. To determine if a compound inhibits the activity of a LRRK2 mutant, the methods of the invention provide for detecting a change in the percentage or portion of cells displaying various patterns of subcellular protein distribution. Patterns of protein distribution that can be quantified include, filamentous distribution of protein, punctuate or aggregate distribution of protein, diffuse distribution of protein, or any combination thereof. For example, a decrease in the percentage of cells displaying a filamentous distribution indicates that the compound inhibits the activity of a LRRK2 mutant. Protein distribution patterns can be quantified by computer-assisted calculations and/or software. A non-limiting example of a technique that can be used to assess subcellular distribution of protein is fluorescence correlational spectroscopy (FCS) {see Example 7).

[0051] Proteins used in the practice of the invention (for example, 14-3-3Θ, and PD- associated LRRK2 mutants) can be linked to a detectable tag which allows visualization of the protein in the cell. Tags that can be used within the context of the invention include epitope tags (for example, FLAG and V5) to which antibodies bind. Detection of a tag (and thus the protein of interest) can be carried out using direct detection (for example, if a tag comprises a detectable moiety such as a fluorescent protein) or indirect detection (for example, by detecting an antibody bound to an epitope tag, where the antibody is linked to a detectable moiety). The invention also encompasses the use of detectable antibodies that specifically bind to a protein to visualize the protein. For example, an antibody or antibody fragment that specifically binds to a PD-associated LRRK2 mutant can be used to detect the LRRK2 mutant in a cell. To detect protein-protein interactions, the proteins of interest can be linked to appropriate fluorescent tags, allowing detection of an interaction by fluorescent resonance energy transfer (FRET). Other techniques for visualizing proteins in cells would be apparent to one skilled in the art. Additionally, linking a protein of interest to a detectable tag can be accomplished using recombinant expression methods widely used in the art.

[0052] Any cell type can be used in the practice of the methods of the invention.

Non-limiting examples of cell types include, a neuronal cells, primary neurons, catechol aminergic CAD cells, HeLa cells (human cervical cancer cells), human embryonic kidney (HEK) cells, baby hamster kidney (BHK) cells, ShSy5y cells (human neuroblastoma cells), and PC 12 cells (rat adrenal pheochromocytoma cells). One skilled in the art would recognize how to obtain cells or cell lines that can be used in the practice of the invention. For example, most cell lines can be obtained from the American Type Culture Collection (ATCC).

Inhibitors of LRRK2 toxicity

[0053] The methods of the invention can be used to screen for compounds that prevent or inhibit LRRK2 toxicity by assessing the ability of a compound to make a filamentous distribution of protein become a more diffuse distribution of protein (Figure 14). Examples of compounds that may be identified by the screening methods as regulators of LRRK2 toxicity include, but are not limited to, molecules that regulate upstream and downstream kinases, molecules that regulate LRRK2 oligomerization, molecules that inhibit LRRK2 kinase, molecules that inhibit LRRK2 GTPase, molecules that inhibit other kinases that phosphorylate LRRK2, and molecules that alter the biology of proteins that modify LRRK2 oligomerization. To identify inhibitors of LRRK2, the methods of the invention can be carried out in a high-throughput manner, for example, to screen libraries of compounds such as kinase inhibitor libraries or RNAi libraries.

[0054] The methods of the invention have been used to identify the heat shock protein

90 (HSP90) inhibitor geldanamycin as an inhibitor of LRRK2 filament formation (see Example 7). Geldanamycin (GA) is a natural product inhibitor of HSP90 chaperone activity. Radicicol is structurally unrelated to geldanamycin, but inhibits HSP90 chaperone activity by the same mechanism as geldanamycin, by binding to the N-terminal ATP-binding pocket of HSP90. A number of semi-synthetic derivatives and synthetic compounds have also been developed which inhibit the activity of HSP90. Non-limiting examples of HSP90 inhibitors that may be used in the methods of the invention include, GA, 17-allylamino-17- demethoxygeldanamycin (17AAG), 17-dimethylaminoethylamino-17- demethoxygeldanamycin (17DMAG), radicicol (and related oxime derivatives), beta- zearalenol, PU24FC1, CCT018159, radamycin, GA dimer, GA-testosterone, GA-oestrogen, novobiocin, coumermycin, cisplatin, depsipeptide, and suberoylanilide hydroxamic acid (SAHA). For a review of HSP90 inhibitors, see Whitesell and Lindquist, Nat Rev Cancer 5:761-772 (2005).

[0055] The methods of the invention have also been used to identify 14-3-3Θ as an inhibitor of the formation of LRRK2 filaments {see Example 5). These findings show that the oligomeric state of LRRK2 is a regulated, and therefore therapeutically accessible, process. Example 5 shows that 14-3-3Θ (SEQ ID NO:2) inhibits the oligomerization of LRRK2, which provides insight into the regulation of LRRK2 function. These results show a link between LRRK2 oligomerization and neurotoxicity, and show that therapeutic agents that reduce LRRK2 self-association may block PD-related neurodegeneration.

[0056] Non-limiting examples of LRRK2 inhibitors include protein or peptide fragments, or small molecules. For example, the structure of an inhibitor of LRRK2 oligomerization can be based on the portion of 14-3-3Θ that mediates binding of 14-3-3Θ to the phosphomotif on LRRK2. The portion of 14-3-3Θ can comprise one or more conserved basic amino acid residues that bind one or more phosphate moieties in LRRK2. For example, an inhibitor of LRRK2 oligomerization can comprise one or more of the four basic residues that comprise the phosphate-binding pocket of 14-3-3 proteins {see Example 5).

[0057] Inhibitors of LRRK2 oligomerization can be based on the structure of the

LRRK2 WD40 domain. As discussed in Example 3, the WD40 domain of LRRK2 is necessary for oligomerization of LRRK2. Conserved basic amino acid residues in the WD40 domain bind phosphate residues (e.g., phosphate residues in the phoshomotif of LRRK2, resulting in homo-oligomerization of LRRK2). Figures 19A - 19C depict multiple sequence alignments of the WD40 domains of LRRK2 proteins from human, chimpanzee, dog, mouse, rat, tetraodon and Drosophila. The basic residues highlighted by a red triangle in Figures 19A-19C depict conserved residues in the LRRK2 WD40 domain. The structure of the WD40 domain can be used to develop peptides or small molecules that block the site in the WD40 domain that binds the LRRK2 phosphomotif. Inhibitors that block a LRRK2 WD40 domain, or a portion thereof, can be used in the assays provided by the invention or therapeutically to prevent or decrease neurotoxicity in a patient with a neurodegenerative disorder.

[0058] An example of a LRRK2 inhibitor is a phosphoantibody that binds to one or more phosphorylated sites on LRRK2. Methods for identifying sequence motifs preferred by kinases have been disclosed (for example, see Nat Biotechnol (2005) 23(l):94-101). Identified motifs can be used to generate phosphopeptides (for example, from about eight to about ten amino acids) against which antibodies can be made by methods known to one skilled in the art.

[0059] Proteins identified as inhibitors of LRRK2 oligomerization (for example, 14-

3-3Θ and HSP90) can be utilized in cell-based assays to screen for compounds that enhance the LRRK2 inhibitory effect of the proteins, for example, by regulating levels of the proteins in cells, by regulating the interaction of the proteins with LRRK2, or promoting the function effects of the proteins upon LRRK2.

[0060] The invention provides methods for determining whether a compound enhances the interaction between a Parkinson's Disease-associated LRRK2 mutant protein and a 14-3-3 peptide, the method comprising (a) co-expressing a Parkinson's Disease- associated LRRK2 mutant protein and a 14-3-3 peptide in a cell; (b) contacting the cell with a compound; and (c) determining whether interaction of the LRRK2 mutant protein and the 14- 3-3 peptide or is enhanced in the cell compared to in a cell co-expressing the LRRK2 mutant protein and the 14-3-3 peptide in the absence of the compound. In one embodiment, the cell comprises a nucleic acid capable of expressing a PD-associated LRRK2 mutant and a nucleic acid capable of expressing a 14-3-3 peptide. In another embodiment, the nucleic acids are located on separate plasmids within the cell. In another embodiment, the nucleic acids are contained within the same plasmid wherein each nucleic acid is regulated by its own promoter.

[0061] LRRK2 is an intracellular target, thus in one embodiment of the invention, the compounds can cross the cell membrane and inhibit the activity of LRRK2. Nonlimiting examples known in the art of methods by which compounds may enter a cell include transduction peptides, transmembrane carrier peptides, internalization factors and liposomes. U.S. Patent Nos. 5,652,122, 5,670,617, 6,589,503 and 6,841,535 describe membrane- permeable peptides that are useful as transfection agents to facilitate the efficient cellular internalization of a broad range and size of compounds including nucleic acids, oligonucleotides, proteins, antibodies, inorganic molecules and PNAs. U.S. Patent No. 5,922,859 describes a method for facilitating endocytosis of therapeutically active nucleic acids (i.e., antisense oligonucleotides, ribozymes or plasmid DNA) into cells using an internalizing factor such as transferrin. As described in U.S. Patent Nos. 5,135,736 and 5,169,933, covalently linked complexes (CLCs) comprising a targeting moiety, a therapeutically active compound (i.e., toxins, radionuclides or peptides) and a peptide facilitating translocation/internalization of the complex across the cell membrane and into the cytoplasm. Also see U.S. Publication No. 20050008617A1, describing compositions and methods for delivery of siRNAs and shRNAs and U.S. Patent No. 5,593,974 covering localized oligonucleotide therapy.

Methods of treatment

[0062] It is a discovery of the invention that LRRK2 oligomerization is linked to neurotoxicity. Therefore, compounds that reduce LRRK2 self-association may represent a novel therapeutic strategy to prevent or decrease PD-related neurodegeneration in patients. The invention provides methods that can be used to screen for potential therapeutic agents. Also provided by the invention are methods for treating neurotoxicity or neurodegeneration in a subject comprising administering an inhibitor of LRRK2. Using the methods of the invention, the HSP90 inhibitor geldanamycin and 14-3-3Θ have been identified as LRRK2 inhibitors. The invention provides for the use of an HSP90 inhibitor, or an isoform of 14-3-3 (for example, SEQ ID NO:1 or 2), or fragments thereof, or nucleic acids encoding an isoform 14-3-3 {see, for example, GenBank Accession Nos. X56468 and U57312), or fragments thereof, to prevent or treat neurodegeneration.

[0063] It is a discovery of the invention that PD-associated mutations utilize at least two mechanisms that generate LRRK2 -mediated neurodegeneration. One mutation acts by enhancing the intrinsic kinase activity of LRRK2. The other LRRK2 PD mutations enhance LRRK2 oligomerization without directly altering kinase function. While these two mechanisms could operate in a common molecular pathway, human postmortem studies of LRRK2-related PD identify two patterns of neuropathology that parallel the two pathogenic mechanisms described here, supporting the possibility that different mutations lead to distinct downstream signals. It is another discovery that 14-3-3Θ reverses LRRK2 oligomerization via a phospho-specific interaction, and this may be used to determine the signals that regulate LRRK2 function, and the consequences of LRRK2 oligomerization. These discoveries link LRRK2 oligomerization and neurotoxicity, and show that agents that reduce LRRK2 self- association represent a novel therapeutic strategy to block PD-related neurodegeneration. The methods of the invention can be used to determine the relationship between 14-3-3Θ, LRRK2 oligomerization and kinase function may ultimately allow for the rational design of novel therapies for PD.

[0064] It is to be understood and expected that variations in the principles of the invention herein disclosed in an exemplary embodiment can be made by one skilled in the art and it is intended that such modifications, changes and substitutions are included within the scope of the present invention.

[0065] The examples set forth below illustrate several embodiments of the invention.

These examples are for illustrative purposes only, and are not meant to be limiting.

EXAMPLES

EXAMPLE 1: CELL-BASED MODEL SYSTEM OF MUTANT LRRK2

NEUROTOXICITY

[0066] To determine molecular mechanisms of LRRK2 -related neurodegeneration, the invention provides a cell-based system to model the neurotoxicity of mutant forms of LRRK2. Primary cultures of cortical neurons were transfected with wild type (WT) or mutant forms of GFP-tagged LRRK2 and apoptotic cell death was measured. All five PD-causing mutants of LRRK2 caused significantly greater cell death than the wild type protein (Figure IB), showing that the neurotoxic effects of these PD mutations are recapitulated in the model system provided by the invention.

[0067] PD mutations increase LRRK2 kinase activity, potentially linking LRRK2 kinase function to cell death^11"13. The model system and methods of the invention were used to determine a relationship between LRRK2 kinase function and neurotoxicity. Autophosphorylation of wild type and mutant forms of LRRK2 was quantified, as well as phosphorylation of the model substrate myelin basic protein ("MBP"; Figures 1C - ID). The kinase activity of only one mutant (G2019S) was markedly elevated, which may account for its pathogenic effect. However, the others exhibited normal (R1441C, R1441G, Y1699C) or only modestly elevated (I2020T) catalytic activity. Thus, these data do not support a clear relationship between LRRK2 kinase activity and neurotoxicity. To further explore the role of kinase function in LRRK2-mediated cell death, the kinase activity of LRRK2 PD mutants was blocked by mutating a residue required for kinase function (K1906R; Figure 6), and assessed the effect on neurodegeneration. The K1906R mutation reduced the neurotoxic effects of all of the PD mutants to WT-LRRK2 levels (Figure IE). K1906R-LRRK2 caused significantly greater cell death than the GFP control. These results show that while kinase function is required to mediate the enhanced neurotoxicity of LRRK2 PD mutations, and elevated catalytic function may be important in a subset of mutations, some effect other than increased intrinsic kinase catalysis may underlie the toxicity of many known LRRK2 mutations. A similar lack of correlation between kinase-activating effects and tumorigenesis has been demonstrated for disease mutations of the kinase B-RAF²².

EXAMPLE 2: FORMATION OF FILAMENTS IN CELLS EXPRESSING PD- ASSOCIATED LRRK2 MUTANTS

[0068] The presence of abnormal protein aggregates is a common occurrence in neurodegenerative disease¹⁴'¹⁵. In this Example, experiments were designed to determine whether abnormal aggregation is a feature of LRRK2 PD mutations. The subcellular distribution of GFP -tagged WT and PD-causing forms of LRRK2 was assessed in the neuronal catecholaminergic CAD cell line. For WT-LRRK2, three distinct immunofluorescent patterns were observed (Figure 2A): diffuse (~60% of transfected cells), punctate ("aggregate"; -35%) and, least commonly occurring, a string-like filamentous distribution ("filaments"; ~5%). Despite the link shown for other proteins between aggregation and neurodegeneration, the percentage of cells with aggregates was not increased by any of the LRRK2 PD mutant alleles. Additionally, in contrast to the kinase requirement for neurodegeneration (Figure IE), blocking kinase function had no effect on LRRK2 aggregate formation (Figure 2B). Furthermore, unlike Lewy bodies in PD, these aggregates were not immunoreactive for ubiquitin or α-synuclein. Based on these results, aggregation of LRRK2 does not appear contribute to its neurotoxicity.

[0069] The four PD-causing alleles that showed normal (R1441 C, R1441 G, Yl 699C) or modestly elevated (I2020T) kinase activity (Figure 1C) significantly increased the percentage of cells bearing LRRK2 filaments (Figures 2C - 2D). Increased filament formation was not observed (Figures 2C - 2D) for the mutant with the highest kinase activity (G2019S; Figure 1C). These filaments also occurred in other cell lines (HeLa and HEK) and primary neurons, or when smaller (e.g., FLAG or V5) epitope tags were used on the N- or the C-terminus respectively (Figures 2D and 7). The ability to form filaments was not explained by differences in LRRK2 protein expression (Figure 8A). The fact that most disease-causing mutations produced a similar effect showed that filament formation was leading to cell death. Since kinase function is required to mediate the enhanced neurotoxicity of LRRK2 PD mutations, experiments were designed to test whether kinase activity was also required for filament formation. In contrast to its effect on aggregates, the kinase blocking mutation (K1906R; doubly mutated molecules) virtually abolished filament formation (Figures 2E and 8B). Based on these results, increased filament formation and enhanced kinase activity may be two mechanisms capable of provoking neurodegeneration.

[0070] The formation of filaments may represent either the binding of LRRK2 to existing filamentous structures (e.g., cytoskeleton) or LRRK2 self-association. Immunohistochemical studies show that LRRK2 filaments do not co-localize with actin, tubulin, intermediate filaments or mitochondria, decreasing the possibility that mutant LRRK2 is recruited to these filamentous structures (Figure 9). To investigate whether LRRK2 oligomerizes, differentially tagged LRRK2 molecules (V5- or GFP -tagged) were co- transfected and assessed to determine whether they co-immunoprecipitate (co-IP). It was found that WT-LRRK2 can self-associate (Figure 3A). The Rl 141C (GTPase) and I2020T (kinase) filament-forming mutants co-IP 'd to a greater extent than WT, whereas the mutant that does not exhibit abnormal filament formation (G2019S) behaved similarly to WT- LRRK2 (Figure 3A). Experiments were also performed to test whether WT and mutant LRRK2 hetero-oligomerize, as might occur in this dominantly inherited form of PD. WT- GFP-LRRK2 and I2020T-V5-LRRK2 co-IP, and do so more efficiently than WT-LRRK2 (Figure 3B). Co-transfection of kinase dead K1906R-V5-LRRK2 (which lacks the ability to form filaments) with I2020T-GFP-LRRK2 inhibited I2020T-GFP-LRRK2 filament formation in trans (Figure 3C), showing that a functional consequence results from the physical association. These data show that self-association may be a normal aspect of LRRK2 function and that the majority of PD mutations may act by enhancing this interaction.

EXAMPLE 3: THE WD40 DOMAIN IS REQUIRED FOR OLIGOMERIZATION OF PD-

ASSOCIATED LRRK2 MUTANTS

[0071] Because of the potential importance of oligomerization in mediating the neurotoxic effects of most LRRK2 PD mutants, studies were designed using the methods of the invention to identify the region responsible for this property. A series of GFP -tagged I2020T-LRRK2 fragments (Figure 3D) were generated and the ability of these molecules to form filaments was tested. The C-terminal region of LRRK2 was assessed first, since WD40 domains have been implicated in phosphorylation-dependent protein-protein interactions²⁰, including homo-oligomerization²¹. Deleting the WD40 domain did not disrupt the intrinsic kinase activity of LRRK2 (Figure 12A). Structure-function studies demonstrated that the WD40 domain was necessary and sufficient for filament formation (Figures 5 A, 12B and 12C). The fragment that lacked the WD40 domain and failed to form filaments was also unable to autophosphorylate (Figure 5B), but retained the ability to phosphorylate MBP. These data show a relationship between autophosphorylation and oligomerization. The phosphorylation of LRRK2 appears to occur via autophosphorylation rather than trans- phosphorylation, as the highly active ROCWD-IT fragment does not phosphorylate kinase dead LRRK2 (Figure 5C).

EXAMPLE 4: THE CELL-BASED MODEL SYSTEM MIRRORS IN VIVO LRRK2

BEHAVIOR

[0072] The methods of the invention were used to determine if the highly overexpressed LRRK2 protein assessed in the model system mirrors the behavior of more modest levels of soluble LRRK2, as may exist in vivo. The mobility of WT and mutant forms of fiuorescently-tagged LRRK2 were measured using fluorescence correlational spectroscopy (FCS)¹⁶. This technique allows one to measure the mobility of low levels of fluorescently tagged protein as they move in living cells, enabling an assessment of in vivo oligomerization, since oligomeric species diffuse more slowly than monomers¹⁶'¹⁷. The mobility of filament enhancing (I2020T) and non-filament enhancing (G2019S) LRRK2 PD mutants was compared. The lowest expressing cells (barely visible with direct fluorescence) that showed no evidence of filament formation (Figure 3D) were selected for further analysis. The diffusion of I2020T-LRRK2 was significantly slower than WT-LRRK2 (Figures 3E and 10). In contrast, no difference was found between WT-LRRK2 and G2019S-LRRK2 (Figure 3E), the only disease mutation that does not enhance filament formation. Thus, modest levels of LRRK2 exhibit mobility differences that correlate with their ability to form filaments.

[0073] The methods of the invention were used to determine whether blocking the kinase activity would return I2020T-LRRK2 mobility to normal, similar to its inhibitory effect on oligomerization and neurodegeneration. The diffusion of kinase dead 12020T- LRRK2 was similar to WT-LRRK2. These results show that the abnormal behavior observed with highly overexpressed LRRK2 mutants reflects the behavior of much lower levels of protein. EXAMPLE 5: IDENTIFICATION OF 14-3-3Θ AS AN INHIBITOR OF MUTANT LRRK2

OLIGOMERIZATION

[0074] To probe the physiological relevance of LRRK2 oligomerization, the methods of the invention can be used to ascertain interacting proteins that regulate LRRK2 oligomerization. The methods of the invention can be used to identify molecules that regulate the formation of LRRK2 filaments. To search for such molecules, dopaminergic CAD cells were transfected with WT-GFP-LRRK2 or GFP alone, then these proteins were immunoprecipitated, and mass spectrometry was performed on silver stained bands that selectively co-purified with LRRK2. This experiment identified 14-3-3Θ (SEQ ID NO:2) as a LRRK2 interacting protein (Figure 1 IA). The 14-3-3 class of molecules is well known to regulate protein-protein interactions, and to interact with and regulate protein kinases¹ '* .

[0075] To further explore this interaction, experiments were designed to determine whether 14-3-3Θ co-IPs with GFP-tagged forms of LRRK2. In cells transfected with WT- LRRK2, endogenous 14-3-3Θ and WT-LRRK2 efficiently co-IP (Figure 4A). Transfected 14- 3-3Θ similarly co-IPd similarly with WT and mutant forms of LRRK2, thus decreasing the possibility that PD mutations act by altering the affinity of 14-3-3Θ for LRRK2 (Figure 4B). To determine if a functional relationship exists between these proteins, the methods of the invention were used to test whether 14-3-3Θ altered the degree of filament formation. Co- transfection of 14-3-3Θ with WT-, R1441C- or I2020T-LRRK2 significantly reduced the proportion of transfected cells with filaments (Figure 4C). The inhibition of LRRK2 filament formation was not due to differences in LRRK2 protein expression (Figure 1 IB). To test whether the association of 14-3-3Θ with LRRK2 occurs through a phospho-motif, one of the four basic residues that comprise the phosphate-binding pocket of 14-3-3 proteins (K49E¹⁹ of SEQ ID NO:2) was mutated. This mutation both decreased the amount of 14-3-3Θ that co- immunoprecipitated with LRRK2 (Figure 4D), and significantly reduced the ability of 14-3- 3Θ to block filament formation of I2020T-LRRK2 (Figures 4E and HC). Taken together, these observations show that 14-3-3Θ regulates LRRK2 oligomerization by binding to a phosphorylated motif. Further, these data are consistent with the regulation of the oligomeric state of LRRK2 by a phosphorylation-dependent cellular signal.

EXAMPLE 6: IDENTIFICATION OF GELDANAMYCIN AS AN INHIBITOR OF MUTANT LRRK2 OLIGOMERIZATION [0076] Geldanamycin is a natural product that inhibits HSP90 chaperone activity by binding to the N-terminal ATP -binding pocket of HSP90. As described below and shown in Figure 18, geldanamycin abolishes LRRK2 filament formation. Geldanamycin analogs that can also be used in the methods of the invention include, but are not limited to, 17- allylamino-17-demethoxy geldanamycin (17AAG), 17-dimethylaminoethylamino-l 7- demethoxygeldanamycin (17DMAG), GA dimer, GA-testosterone and GA-oestrogen. Examples of other HSP90 inhibitors that can be used include novobiocin, coumermycin, cisplatin, depsipeptide, suberoylanilide hydroxamic acid (SAHA), radicicol (and related oxime derivatives), beta-zearalenol, PU24FC1, CCT018159, and radamycin.

[0077] CAD cells were transfected with WT or I2020T GFP-tagged LRRK2.

Following 4 hour incubation of DNA complexes, the media was removed and replaced with growth medium containing the indicated concentrations of geldanamycin (Time 0). Geldanamycin (Stressgen) stock was prepared in DMSO (1 mM). The concentrations used were as follows (in nM): 1, 10, 50, 100; or DMSO vehicle control. Forty-eight hours following transfection, the cells were fixed and processed for anti-GFP immunofluorescence (see Example 7). The percentage of GFP -positive cells containing filaments was determined in a blinded manner. The remaining cells were processed for determination of GFP-LRRK2 expression levels by western immunoblot (see Example 7). Figure 18 shows a geldanamycin concentration-dependent decrease in the percentage of cells containing filaments in cells expressing wildtype (WT) LRRK2 and in cells expressing I2020T (IT) mutant LRRK2.

EXAMPLE 7: EXPERIMENTAL METHODS

[0078] Cloning of Human LRRK2 cDNA: A human LRRK2 cDNA was amplified from HEK cell cDNA. The clone was fully sequenced and conformed to human LRRK2 cDNA in the NCBI database. All subsequent mutations were generated using site directed mutagenesis and all mutant clones were resequenced to confirm their accuracy.

[0079] Immunoprecipitation and in vitro kinase assay: CAD cells transiently expressing GFP-LRRK2 for 48h were Dounce-homogenized in ice-cold lysis buffer (20 mM HEPES, pH 7.4, 150 mM NaCk, 0.1% NP- 40, 2 mM EGTA, 1 mM dithiothreitol, 10% glycerol, 1 mM sodium orthovanadate, 10 mM NaF, 25 mM β -glycerophosphate, and protease inhibitors). After centrifugation at 20,000 x g for 15 min, the supernatants were precleared with protein-A agarose (Roche) for 30 min at 4 °C. Bradford protein assay was performed to normalize protein concentrations before immunoprecipitation. Lysates containing 2-4 mg protein were immunoprecipitated with rabbit anti-GFP antibody (Abeam) for 1 h followed by incubation with protein-A beads for 2 h. The immunoprecipitates were washed three times with lysis buffer without glycerol, and twice with kinase buffer (25 mM Tris-HCL, pH 7.4, 10 mM MgCl₂, 1 mM dithiothreitol, 0.1 mM sodium orthovanadate, and 5 mM β -glycerophosphate). Kinase assay was carried out by incubating the immunoprecipitates for 30 min at 30°C in kinase buffer containing 25 μM ATP and 5 μCi [γ-

P]ATP with or without 5 μg myelin basic protein (MBP) as a substrate. The reaction was terminated by boiling the mixture in IX SDS sample buffer for 5 min. Equivalent amounts of LRRK2 were resolved by SDS-PAGE and ³²P incorporation into LRRK2 or MBP was quantified by scintillation counting of the excised silver stained bands. The results represent the combined data from 4-8 experiments for all LRRK2 PD mutants.

[0080] Co-immunoprecipitation: CAD cells were co-transfected with GFP-LRRK2 and LRRK2-V5 or 14-3-3Θ 40 h post-transfection, cells were lysed in lysis buffer (20 mM HEPES, pH 7.4, 150 mM NaCh, 0.1% NP-40, 2 mM EGTA, 2 mM MgCk, 10% glycerol, 1 mM sodium orthovanadate, 10 mM NaF, 25 mM β-glycerophosphate, and protease inhibitors). After centrifugation and preclearing, lysates were incubated with rabbit anti-GFP antibody and protein-A agarose for 3 h to overnight at 4°C. The complexes were washed five times with lysis buffer and released from beads by boiling in SDS-sample buffer for immunoblots.

[0081] Immunolabelling: Immunofluorescence labeling was performed on CAD cells, HeLa cells or primary cortical neurons 48 h after transfection using methanol- or formaldehyde-fixed cells grown on coverslips. Coverslips were blocked for 30 min at RT in block solution (PBS, 0.25% Triton X-100 and 10% normal donkey serum), incubated overnight at 4⁰C in primary antibodies diluted in block solution. The next day coverslips were washed, incubated with FITC- or Texas red-conjugated secondary antibodies, and washed in PBS before mounting using Vectashield Mounting Media with DAPI (Vector Laboratories). For quantification, at least 100 cells were counted on 3 independent coverslips and the percentage of cells displaying each pattern was determined. Triplicate cultures were assessed in the same manner. Images were obtained using a IOOX oil-immersion objective with a Zeiss LSM510 2-photon confocal microscope.

[0082] Mass spectrometry: Immunoprecipitation of EGFP or GFP-LRRK2 was done as described below. The immunoprecipiates were analyzed on 7.5% or 12% Criterion Tris- HCl gel (Bio-Rad) and gels were silver stained using the SilverQuest kit as per the manufacturers instructions. Protein bands were excised from gels and digested with trypsin. The tryptic peptides were extracted with 2X 50ml 50% acetonitrile/2% TFA. The combined extracts were reduced in volume and subjected to LC/MS/MS analysis on a Micromass Q-tof mass spectrometer using in-line reversed-phase separation with a linear acetonitrile gradient at 200nL/min flow rate. Operating conditions were 1.8 kV capillary voltage, 32 V cone voltage. MS/MS spectra were acquired in a data-dependent manner for the entire digest, scanning ions in the mass range 350-1500 amu. Raw data were processed using the MassLynx MaxEnt 3 software from Micromass, and the resulting .pkl files were submitted to a Mascot search at www.matrixscience.com.

[0083] Primary Embryonic Cortical Neuron Culture: The cortex of El 6 mice was isolated aseptically in HBSS, enzymatically dissociated with 0.25% Trypsin followed by mild mechanical dissociation in Neurobasal medium (Invitrogen) containing 10% heat-inactivated horse serum with a flame-polished Pasteur pipette²⁴. Dissociated cells were plated on poly- D-lysine coated glass coverslips at a density of 200,000 cells/cm². After a period of 4-6 h, the media containing serum was removed and replaced with Neurobasal medium containing B-27 serum-free supplements (Invitrogen), L-glutamine, and penicillin/streptomycin. Three days following plating, the cells were transfected. The percentage of apoptotic cells was determined in a blinded fashion by counting at least 100 GFP -positive cells from three independent coverslsips from three to four individual litters. Apoptotic nuclei were defined as those cells having two or more condensed apoptotic nuclear bodies.

[0084] Cell Lines: Mouse catacholaminergic CAD cells were grown in

DMEM/F12+8% FBS with penicillin/streptomycin. HEK cells were grown in DMEM+10%FBS with penicillin/streptomycin. HeLa cells were grown in MEM+10%FBS with penicillin/streptomycin.

[0085] Transfection of Cell Lines and Primary Neuronal Cultures: CAD, HeLa, and

HEK cells were transfected using Lipofectamine/PLUS (Invitrogen) as per the manufacturers instructions. Primary cortical neurons were transfected using Lipofectamine 2000 (Invitrogen) as follows: for each coverslip, 0.8 μg of DNA was mixed with 1.3 μl Lipofectamine 2000 reagent for a period of Ih at room temperature before adding to the cells in pre-warmed Opti-MEM. Following 4h incubation at 37°C, the media containing DNA complexes was removed and replaced with normal growth medium (Neurobasal/B-27). [0086] Assessment of LRRK2 Subcellular Distribution: GFP-LRRK2 immunofluorescence was categorized based on the following criteria by an observer blinded to the genotype: A) diffuse: cells showing uniform GFP immunoreactivity throughout the cytoplasm with any evidence of punctate staining or aggregation; B) aggregates: cells showing numerous GFP-positive punctate structures distributed variously throughout the cytoplasm; and C) filaments: cells containing distinct continuous/filamentous GFP-positive structures within the cytoplasm without any additional aggregates.

[0087] Fluorescence Correlational Spectroscopy (FCS): HEK cells were transfected with N-terminal tagged GFP or YFP-LRRK2. Before performing measurements, the cells were washed with DMEM without phenol red to ensure that no background fluorescence due to phenol red in the medium would interfere with the analysis. FCS measurements were carried out at room temperature (20⁰C) with a FCS extension attached to a Leica TCS SP2 AOBS confocal laser-scanning microscope (Leica Microsystems, Manheim, Germany). The 514-nm line of Ar laser was used for YFP and the 488-nm line was used for GFP excitation. The excitation beam was focused onto the sample through a HCX PL Apo CS 63x1.2 water immersion objective. The YFP emission was collected through a 535-585 nm band pass filter and the GFP emission was collected at 500-550 nm. A single cell image expressing WT or mutant LRRK2 was taken prior to actual FCS measurement and then the laser beam was focused at a selected spot within the cytoplasm where FCS measurements were acquired during 100 sec.

[0088] FCS Data Analysis: The autocorrelation curves were acquired, processed and evaluated using the Leica/ISS FCS software. The autocorrelation function, G(τ), is calculated from the photon counts by,

where F(ή represents the detected photon counts at time t, and τ is the lag time. The angle brackets represent the time average. The normal three-dimension diffusion is fitted using following equation,

where N is the average number of fluorescent particles in the observation volume defined by radius coo and length 2zo, and S is the structure parameter representing the ratio, S = ZQ / coo ■

The diffusion time of free YFP in solution is calculated as approximately 50 μs. However, diffusion in the cell could be obstructed by obstacles such as membrane, organelles and other proteins. Therefore, diffusion in the cell could be expressed by the anomalous diffusion equation.

The anomaly of the diffusion is described by one parameter, the exponent, which has a value between 0 and 1. In experiments described in the previous Examples, the autocorrelation function of YFP expressed in the HEK cell was measured as a standard sample and the characteristic diffusion time is obtained as ca, 500 μs, which is about 10 times slower than in solution.

[0089] While the foregoing invention has been described in some detail for purposes of clarity and understanding, these particular embodiments are to be considered illustrative and not restrictive. It will be appreciated by one skilled in the art from a reading of this disclosure that various changes in form and detail can be made without departing from the true scope of the invention.

REFERENCES

1. Goldwurm, S. et al. The G6055A (G2019S) mutation in LRRK2 is frequent in both early and late onset Parkinson's disease and originates from a common ancestor. J Med Genet 42, e65 (2005).

2. Ozelius, L. J. et al. LRRK2 G2019S as a cause of Parkinson's disease in Ashkenazi Jews. N Engl J Med 354, 424-5 (2006).

3. Lesage, S. et al. LRRK2 G2019S as a cause of Parkinson's disease in North African Arabs. N Engl J Med 354, 422-3 (2006).

4. Gilks, W. P. et al. A common LRRK2 mutation in idiopathic Parkinson's disease. Lancet 365, 415-6 (2005).

5. Bras, J. M. et al. G2019S dardarin substitution is a common cause of Parkinson's disease in a Portuguese cohort. Mov Disord 20, 1653-5 (2005). 6. Paisan-Ruiz, C. et al. Cloning of the gene containing mutations that cause PARK8-linked Parkinson's disease. Neuron 44, 595-600 (2004).

7. Zimprich, A. et al. Mutations in LRRK2 cause autosomal-dominant parkinsonism with pleomorphic pathology. Neuron 44, 601-7 (2004).

8. Dauer, W. & Przedborski, S. Parkinson's disease: mechanisms and models. Neuron 39, 889- 909 (2003).

9. Cookson, M. R., Xiromerisiou, G. & Singleton, A. How genetics research in Parkinson's disease is enhancing understanding of the common idiopathic forms of the disease. Curr Opin Neurol 18, 706-11 (2005).

10. Bosgraaf, L. & Van Haastert, P. J. Roc, a Ras/GTPase domain in complex proteins. Biochim Biophys Acta 1643, 5-10 (2003).

11. Gloeckner, C. J. et al. The Parkinson disease causing LRRK2 mutation I2020T is associated with increased kinase activity. Hum MoI Genet 15, 223-32 (2006).

12. West, A. B. et al. Parkinson's disease-associated mutations in leucine-rich repeat kinase 2 augment kinase activity. Proc Natl Acad Sci U S A 102, 16842-7 (2005).

13. Greggio, E. et al. Kinase activity is required for the toxic effects of mutant LRRK2/dardarin. Neurobiol Dis (2006).

14. Sherman, M. Y. & Goldberg, A. L. Cellular defenses against unfolded proteins: a cell biologist thinks about neurodegenerative diseases. Neuron 29, 15-32. (2001).

15. Muchowski, P. J. Protein misfolding, amyloid formation, and neurodegeneration: a critical role for molecular chaperones?PG - 9-12. Neuron 35, 9-12. (2002).

16. Bacia, K., Kim, S. A. & Schwille, P. Fluorescence cross-correlation spectroscopy in living cells. Nat Methods 3, 83-9 (2006).

17. Purkayastha, P. et al. Alpha 1 -antitrypsin polymerization: a fluorescence correlation spectroscopic study. Biochemistry 44, 2642-9 (2005).

18. Tzivion, G. & Avruch, J. 14-3-3 proteins: active cofactors in cellular regulation by serine/threonine phosphorylation. J Biol Chem 277, 3061-4 (2002).

19. Bridges, D. & Moorhead, G. B. 14-3-3 proteins: a number of functions for a numbered protein. Sci STKE 2005, relO (2005).

20. Yaffe, M. B. & EHa, A. E. Phosphoserine/threonine-binding domains. Curr Opin Cell Biol 13, 131-8 (2001).

21. Thornton, C. et al. Spatial and temporal regulation of RACKl function and N-methyl-D- aspartate receptor activity through WD40 motif-mediated dimerization. J Biol Chem 279, 31357-64 (2004).

22. Wan, P. T. et al. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell 116, 855-67 (2004). 23. Taylor, J. P., Mata, I. F. & Fairer, M. J. LRRK2: a common pathway for parkinsonism, pathogenesis and prevention? Trends MoI Med 12, 76-82 (2006).

24. Rideout, H. J., Dietrich, P., Wang, Q., Dauer, W. T. & Stefanis, L. alpha-synuclein is required for the fibrillar nature of ubiquitinated inclusions induced by proteasomal inhibition in primary neurons. J Biol Chem 279, 46915-20 (2004).

Claims

What is claimed is:

1. A method for determining whether a compound inhibits mutant leucine-rich repeat kinase-2 (LRRK2) protein activity, the method comprising:

(a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell;

(b) contacting the cell with a compound; and

(c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound,

wherein determination of a reduction of filamentous distribution in (c) indicates that the compound inhibits the LRRK2 mutant protein activity.

2. A method for determining whether a compound inhibits mutant leucine-rich repeat kinase-2 (LRRK2) protein activity, the method comprising:

(a) contacting a cell with a compound;

(b) expressing a Parkinson's Disease-associated LRRK2 mutant protein in the cell; and

(c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound,

wherein determination of a reduction of filamentous distribution in (c) indicates that the compound inhibits the LRRK2 mutant protein activity.

3. A method for determining whether a compound inhibits toxicity of a LRRK2 mutant protien in a cell, the method comprising:

(a) expressing a Parkinson's Disease-associated LRRK2 mutant protein in a cell, wherein expression of the mutant protein results in filamentous distribution of the protein in the cell; (b) contacting the cell with a compound;

(c) determining whether filamentous distribution of the mutant protein in the cell is reduced compared to filamentous distribution of the mutant protein in a cell expressing the LRRK2 mutant in the absence of the compound; and

(d) correlating toxicity of the mutant protein to filamentous distribution of the mutant protein,

wherein determination of a reduction of filamentous distribution in (c) and a decrease in toxicity in (d) indicates that the compound inhibits toxicity of the LRRK2 mutant protein in a cell.

4. The method of claim 1, 2 or 3, wherein the cell comprises a nucleic acid vector capable of expressing a Parkinson's Disease- associated LRRK2 mutant protein.

5. The method of claim 4, wherein the LRRK2 mutant protein comprises a Rl 441 C mutation, a R1441G mutation, a Y1699C mutation, a I2020T mutation, a G2019S mutation, or any combination thereof.

6. The method of claim 1, 2 or 3, wherein the LRRK2 mutant protein is linked to a detectable tag.

7. The method of claim 6, wherein the detectable tag is an epitope tag.

8. The method of claim 6, wherein the detectable tag is a fluorescent protein.

9. The method of claim 1 , 2 or 3, wherein the cell is a primary neuron.

10. The method of claim 1, 2 or 3, wherein the cell is a catecholaminergic CAD cell.

11. The method of claim 1, 2 or 3, wherein the cell is a HeLa cell, a human embryonic kidney (HEK) cell, a baby hamster kidney (BHK) cell, a ShSy5y cell, or a PC 12 cell.

12. The method of claim 1, 2 or 3, wherein determining a reduction of filamentous distribution comprises detecting a change in diffuse distribution of the protein, punctuate distribution of the protein, filamentous distribution of the protein, or any combination thereof in the cell.

13. The method of claim 1, 2 or 3, wherein the determining comprises detecting a detectable tag.

14. The method of claim 1, 2 or 3, wherein the determining comprises detecting fluorescence.

15. The method of claim 14, wherein fluorescence is detected directly.

16. The method of claim 14, wherein fluorescence is detected indirectly.

17. The method of claim 1, 2 or 3, wherein the determining comprises computer-assisted quantification of filamentous distribution of the protein.

18. The method of claim 1, 2 or 3, wherein the determining comprises computer-assisted quantification of diffuse distribution of the protein.

19. The method of claim 1, 2 or 3, wherein the determining comprises computer-assisted quantification of punctate distribution of the protein.

20. The method of claim 1, 2 or 3, wherein the compound is a small molecule, a polypeptide, a protein, a peptide, a peptidomimetic, a nucleic acid, an RNA, a DNA, an antisense RNA a small interfering RNA (siRNA), a double stranded RNA (dsRNA), a short hairpin RNA, a cDNA, or any combination thereof.

21. The method of claim 1, 2 or 3, wherein the compound is an inhibitor of HSP90.

22. The method of claim 21, wherein the inhibitor of HSP90 is geldanamycin, radicicol, radamycin or novobiocin.

23. The method of claim 1, 2 or 3, wherein the method is carried out in a high-throughput manner.

24. The method of claim 1, 2 or 3, wherein the method is carried out for more than 100 compounds.

25. The method of claim 1, 2 or 3, wherein the method is carried out in a multi-well plate.

26. A method for determining whether a compound enhances interaction between a Parkinson's Disease-associated LRRK2 mutant protein and a 14-3-3 peptide, the method comprising: (a) co-expressing a Parkinson's Disease-associated LRRK2 mutant protein and a 14-3-3 peptide in a cell;

(b) contacting the cell with a compound; and

(c) determining whether interaction of the LRRK2 mutant protein and the 14-3-3 peptide is enhanced in the cell compared to in a cell co-expressing the LRRK2 mutant protein and the 14-3-3 peptide in the absence of the compound.

27. The method of claim 26, wherein the 14-3-3 peptide is a 14-3-3Θ.

28. The method of claim 26, wherein the 14-3-3 peptide is linked to a detectable tag.

29. The method of claim 26, wherein the determining comprises fluorescence resonance energy transfer (FRET).

30. The method of claim 26, wherein the LRRK2 mutant protein comprises a Rl 441 C mutation, a R1441G mutation, a Y1699C mutation, a I2020T mutation, a G2019S mutation or any combination thereof.

31. The method of claim 26, wherein the LRRK2 mutant protein is linked to a detectable tag.

32. The method of claim 31, wherein the detectable tag is an epitope tag.

33. The method of claim 31, wherein the detectable tag is a fluorescent protein.

34. The method of claim 26, wherein the cell is a primary neuron.

35. The method of claim 26, wherein the cell is a catecholaminergic CAD cell.

36. The method of claim 26, wherein the cell is a HeLa cell, a human embryonic kidney (HEK) cell, a baby hamster kidney (BHK) cell, a ShSy5y cell, or a PC 12 cell.

37. The method of claim 26, wherein determining a reduction of filamentous distribution comprises detecting a change in diffuse distribution of the protein, punctuate distribution of the protein, filamentous distribution of the protein, or any combination thereof in the cell.

38. The method of claim 26, wherein the determining comprises detecting a detectable tag.

39. The method of claim 26, wherein the determining comprises detecting fluorescence.

40. The method of claim 39, wherein fluorescence is detected directly.

41. The method of claim 39, wherein fluorescence is detected indirectly.

42. The method of claim 26, wherein the determining comprises computer-assisted quantification of filamentous distribution of the protein.

43. The method of claim 26, wherein the determining comprises computer-assisted quantification of diffuse distribution of the protein.

44. The method of claim 26, wherein the determining comprises computer-assisted quantification of punctate distribution of the protein.

45. The method of claim 26, wherein the compound is a small molecule, a polypeptide, a protein, a peptide, a peptidomimetic, a nucleic acid, an RNA, a DNA, an antisense RNA a small interfering RNA (siRNA), a double stranded RNA (dsRNA), a short hairpin RNA, a cDNA, or any combination thereof.

46. The method of claim 26, wherein the method is carried out in a high-throughput manner.

47. The method of claim 26, wherein the method is carried out for more than 100 compounds.

48. The method of claim 26, wherein the method is carried out in a multi-well plate.

49. A method for preventing or decreasing neurotoxicity in a subject, the method comprising administering an effective amount of an inhibitor of a LRRK2 mutant protein, wherein the inhibitor decreases LRRK2 filament formation in the subject.

50. The method of claim 49, wherein the inhibitor comprises a 14-3-3 peptide, or a fragment thereof.

51. The method of claim 49, wherein the inhibitor comprises a nucleic acid capable of expressing a 14-3-3 peptide, or a fragment thereof.

52. The method of claim 50 or 51, wherein the 14-3-3 peptide is a 14-3-3Θ.

53. The method of claim 49, wherein the inhibitor comprises an HSP90 inhibitor.

54. The method of claim 53, wherein the HSP90 inhibitor is geldanamycin, radicicol, radamycin or novobiocin.

55. The method of claim 49, wherein the inhibitor blocks a LRRK2 WD40 domain, or a portion thereof.