Parkinson's disease (PD) is a progressive neurodegenerative condition of complex etiology exhibiting a range of movement syndromes caused by the selective degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNc). The second most prevalent neurodegenerative disease after Alzheimer's disease, PD is largely sporadic - ~90-95% of cases, the remaining 5-10% being familial; age is a major risk factor. As life span is expected to increase, the incidence of PD is also likely to i
ncrease, further adding to the economic and health burden it imposes. Several genes have been implicated in the pathogenicity of PD. The early-onset, autosomal recessive form of PD (AR-PD) is linked to mutations in PINK1 and Parkin (PARK2) proteins, important players in the mitochondrial autophagy and also impacting on mitochondria dynamics. Associated with rare cases of an early-onset autosomal recessive form of PD is DJ-1 (PARK7), a possibly multifunctional protein. Leucine-rich repeat kinase 2 (LRRK2) protein is involved in late-onset, autosomal dominant PD; it accounts for the more common familial type and also represents a risk factor for the idiopathic type. But the key component of PD pathogenesis is alpha-synuclein (SNCA) protein, primarily associated with sporadic PD while a few mutations and gene multiplication are linked to rarer, dominant familial forms. Lewis Bodies (LB) and Lewis Neurites (LN) - pathological hallmarks of PD, are largely composed of aggregated fibrils of beta-sheet rich alpha-synuclein. Unlike PINK1 and Parkin, the function of SNCA, LRRK2 or PARK7, are not well understood. SNCA is thought to play a role in synaptic vesicle homeostasis and mitochondrial function; LRRK2 is thought to play roles in immune and cytoskeletal systems as well as vesicular trafficking and mitochondrial function; PARK7 is thought to have antioxidant and transcriptional modulation properties and also be important for mitochondrial function. Several other genes have been identified with roles in intracellular trafficking/lysosomal/synaptic vesicle functions and PD. Mitochondria dysfunction has long been associated with aging and neurodegeneration; however, this does not explain the selective loss of dopaminergic neurons in SNc. Distinctive features of these neurons include a complex architecture and specific autonomous pacemaker activity which could render them particularly vulnerable to alterations in energy and calcium homeostasis. Differential expression of microRNAs, distinct epigenetic marks and neuronal pigment levels in PD are also documented.
Alpha-synuclein (SNCA), the major contributor to PD, is one of the three members of the synuclein family but the only one associated with this and a few other neurodegenerative conditions. SNCA aggregates and the LB of which they are the main components, are also found in dementia with Lewy Bodies (DLB), multiple system atrophy (MSA) and Alzheimer's disease (AD), but with a pattern of distribution different from PD. SNCA is a small protein whose gene comprises six exons of which five encode the protein. Several transcripts are known, but the full length 140 amino acid-encoding transcript is the most common. The N-terminal unstructured domain, upon interaction with membranes, forms an amphipathic helix with several conformations; the middle domain, alpha-helical and hydrophobic, can acquire a beta-sheet conformation and plays a role in the oligomerization and aggregation of the protein; the acidic C-terminal end is unfolded. The protein exists as a monomer and oligomer, e.g. tetramer, but can transition to the fibrillar aggregates in LB. SNCA is found predominantly at presynaptic terminals, synaptic vesicles in particular, with which it interacts. It interacts with other lipid membranes such as lipid rafts, Golgi and mitochondria, the latter enhanced by cardiolipin (CL) and appears to have a higher affinity for curved membranes. The actual function of the protein is rather elusive. Due to its enrichment at and interaction with synaptic vesicles it is believed to have a role in synaptic vesicle homeostasis where it is thought to act as a chaperone for the SNARE complex, involved in the synaptic vesicle fusion step. SNCA interacts with one of the SNARE components and also with several other proteins but the significance of these interactions, particularly for PD, remains to be established. Of note however, are the Rab proteins with a role in vesicular trafficking, or those involved in DA metabolism. Interestingly, the two Rab partners have recently been shown to be substrates of LRKK2 and, at least for one of them, phophorylation affects the interaction with regulators. SNCA is subject to several types of post translational modification that include phosphorylation, nitration, ubiquitination, sumoylation, possibly acetylation, methionine oxidation and tyrosine nitrosylation. Aggregated SNCA induces ER stress by inhibiting the unfolded protein response (UPR) pathway, leading to cell death. Overexpression of SNCA inhibits the activity of respiratory complex I. It has also been shown that SNCA can inhibit mitochondria fusion and promote fission with cardiolipin (CL) playing a role. SNCA is a substrate of chaperone-mediated autophagy (CMA) and mutant protein is toxic to this pathway. Monomeric and aggregated SNCA are subject to exocytosis and their secretion is elevated when the proteasomal and mitochondrial function are impaired. Spreading of aggregated SNCA in a prion-like fashion has been observed.
PINK1 and Parkin (PARK2)
Pink1 and Park2 are the essential players of mitochondria quality-control (QC), prompting mitochondrial autophagy (mitophagy) in response to stress. Pink1, a serine/threonine kinase and downstream of it, Park2, an E3 ubiquitin ligase, act in tandem to assure that mitochondria with compromised membrane potential are delivered to the autophagosome and targeted for degradation. Substrates of Park2 include ubiquitin-binding autophagy receptors/adapters. Other ubiquitinated substrates are targeted for proteosomal degradation; they include the mitofusins and Miro GTPase proteins involved in mitochondria fusion and transport, respectively. Miro is also a substrate for Pink1. Thus, Pink1 and Park2 also modulate mitochondria dynamics. Inhibition of fusion helps fragment mitochondria networks and isolates damaged mitochondria via fission; inhibition of transport prevents delivery of damaged mitochondria at nerve terminals where they are necessary for energy supply, while promoting a retrograde transport for degradation. Inactivating mutations in Pink1 and Park2 alter mitophagy and also impair the important regulatory role these proteins play in mitochondria dynamics. Together, alterations in mitochondrial autophagy and accompanying defects in mitochondria dynamics can seriously damage mitochondria function and prompt cell death.
Leucine-rich repeat kinase 2 (LRRK2)
The leucine-rich repeat kinase 2 (LRRK2) is a large protein of 51 exons. It has a GTPase Ras-of-Complex (ROC) domain adjacent to C-terminal-of-ROC (COR) linker region followed by the serine/threonine kinase domain. Flanking the ROC-COR- kinase central region there are putative protein-protein interaction domains such as the ankyrin (ANK) and the leucine-rich repeat (LRR) domains, and WD40 repeats at the N- and C-termini, respectively. The kinase activity appears to be dependent upon the GTPase function, nucleotide binding rather than hydrolysis. The functional protein is thought to be a dimer with possible involvements in the cytoskeletal and immune systems, vesicular trafficking and mitochondrial function, and signaling. LRRK2 phosphorylates itself and a number of substrates. Mutations are found in both the kinase and GTPase domains, generally activating the kinase and diminishing the GTPase function. Several interacting partners and substrates are identified but the significance of these interactions and modifications, particularly for PD, remains to be established. Interestingly, among the interacting partners are the three dishevelled proteins and the LRP6 co-receptor of the Wnt signaling pathway. Wnt signaling, particularly the better understood canonical pathway, plays central roles in development, including neurogenesis and is also important for normal adult cell function, including the midbrain dopaminergic neurons, lost in PD. While the connection between altered Wnt signaling and PD needs further investigation, data suggest possible links between deregulation in the Wnt pathway and familial PD. A possible role for LRRK2 as a scaffold in the Wnt pathway has been suggested. Co-transfection of LRRK2 and the three disheveled proteins increases canonical Wnt activity and this effect is weakened by LRRK2 mutations. These mutations also weaken the interaction with LRP6 co-receptor. Intriguingly, silencing of LRRK2 also enhances canonical Wnt signaling. Possibly, LRRK2 exerts inhibitory effects under basal conditions - co-immunoprecipitations experiments point to the presence of Lrrk2 in complex with components of the beta-catenin 'destruction complex'. Other partners include Rab GTPases, involved in vesicular trafficking, of which Rab29 known as Rab7l1 is a risk factor for PD; Gak, also a risk factor; and VPS35, a component of the retromer, and associated with PD. Several MAP2 kinases are also partners and substrates of LRRK2. LRRK2 is a substrate for chaperone-mediated autophagy and mutant protein is toxic to this pathway.Studies reveal that a subset of Rab GTPases, and different from those identified as interacting partners, are substrate for the kinase function. Pathogenic mutations in LRRK2, augment phosphorylation of substrates, as expected. Interestingly, LRRK2-mediated phosphorylation affects the interaction of Rab GTPases with regulators such as guanine exchange factor (GEF), guanine activating protein (GAP), GDP dissociation inhibitor (GDI), or members of the Rab geranyltransferase complex, all of which preferentially bind a non-phosphorylated Rab. Note that two of the Rab substrates, are interacting partners for SNCA.
Parkinson protein 7 (PARK7)
Park7 (known as DJ-1) is thought to have anti-oxidant and transcriptional modulating properties. It localizes to the cytosol and nucleus and to a lesser extent to the mitochondria. However, it may have a role in mitochondria function by regulating the expression of genes relevant to it. Loss or knockdown of Park7 affects mitochondria morphology and dynamics. Interestingly, changes in mitochondria morphology after loss of Park7 are prevented by Park2 and Pink1. The possibility of Park2, Pink1 and Park7 participating in a common pathway has been suggested. Park7 can interact with Pink1, and all three proteins - Park7, Pink1 and Park2 associate with Hspa4 and Hspa9 chaperones. However, the significance of these interactions, and their possible role in PD, remain to be established. Through its transcriptional modulatory function, Park7 can also promote antioxidant responses. Of note, Park7 is known as an oncogene - its expression is elevated in several cancer types and the increase in both mRNA and proteins levels is restricted to tumor tissues.
Other PD related proteins
In addition to the more representative, better documented proteins above, several others have been identified, such as VPS35 in a rare autosomal dominant form, FBXO7 in a rare recessive form, DNAJC6, SYNJ1 and PLA2G6, associated with the recessive, juvenile-onset form of PD, and ATP13A2 and GBA, associated with other conditions but also linking to PD. Interestingly, they can either establish interactions with some of the players described above and/or are implicated in intracellular trafficking/lysosomal/synaptic vesicle functions and SNCA accumulation. VPS35 is a component of the retromer complex, a key element of endosomal sorting/trafficking. It mediates the selective sorting of cargo for traffic to the trans-Golgi network (TGN) or plasma membrane in the retrograde or recycling pathway, respectively, of endosome export pathways of endosomal sorting. The dominant VPS35 mutation affects cathepsin D (CTSD) trafficking, a lysosomal protease that degrades SNCA. This and other lysosomal hydrolases are modified with mannose-6-phosphate (M6P) and recognized by specific receptors (MPR) that deliver them from the TGN to lysosomes, the ligands dissociate and the receptors are returned to the TGN for another round of delivery. CTSD is recognized by the cation-independent receptor (CI-MPR/IGF2R), a known cargo of the retromer. In the context of VPS53 mutation, abnormal levels of SNCA in late endosomes/lysosomes are observed, likely due to impaired CTSD trafficking. FBXO7, a component of E3 ubiquitin ligases, can interact with PARK2 and could be involved in mitophagy. DNAJC6, known as auxilin, belongs to the DNAJ/HSP40 family of proteins and plays a role in clathrin-mediated endocytosis. Once the clathrin-coated vesicles are internalized, the clathrin coat is disassembled by HSPA8, a heat shock protein which functions as an ATPase in this process, with auxilin as its cofactor. DNAJC6 is selectively expressed in neurons and its mutations could perturb synaptic vesicle endocytosis and recycling. Changes in phosphoinositide composition are important for shedding the adaptors of the clathrin coat, a requirement met by phosphoinositide phosphatase synaptojanins. SYNJ1, highly expressed in nerve terminals, is one of the two main proteins in the family and is required for synaptic vesicle endocytosis. The SYNJ1 PD mutations are in one of the two phosphatase domains. Clathrin-mediated endocytosis is a main pathway of synaptic vesicle internalization. PLA2G6 phospholipase is a member of the calcium-independent group of the A2 family of phospholipases, with roles in fatty acid metabolism and phospholipid remodeling, among others. ATP13A2 is largely found in lysosomes and may have a role in divalent metal cation movement; it is associated with the Kufor-Rakeb syndrome but mutations are also associated with early onset PD. Mutations in the lysosomal glucocerebrosidase GBA are implicated in Gaucher disease, and they are also implicated in PD. GBA interacts with membrane-bound SNCA which inhibits the enzyme function. Like SNCA, GAK, RAB7L1 or LRRK2, GBA is a risk factor for idiopathic PD and with the highest odds ratio from meta-analysis of genome-wide association (GWAS) data sets. Of note is that the MAPT locus is also identified as a risk factor whose product, the TAU protein, plays important roles in cytoskeletal network and axonal transport and whose fibrillary tangles are associated with Alzheimer's and other neurodegenerative disorders. Also of note is DNAJC13, known as Rme8, also a component of the retromer pathway and whose recently identified mutations have been linked to PD.
Dopaminergic neurons in SNc
Dopaminergic neurons in the SNc have massive axonal arbors, and the axons are long (~4.5m in humans) and unmyelinated. Collectively, these features pose heightened energy demand and require adequate mitochondria supply, transport and function. A higher energy supply implies higher levels of reactive oxygen species (ROS) produced in the electron transfer chain (ETC) pathway, potentially stressing the cell antioxidant machinery. ROS can also be generated by the metabolism of dopamine during its degradation (DA is largely stored in vesicles). Defective mitochondria could further increase ROS production. Mitochondria are also sites of calcium storage. Calcium is a quintessential element for life processes, so an imbalance in its homeostasis can have devastating cellular consequences. Intracellular Ca2+ concentration is maintained by a steep gradient, and proper storage and handling, all carefully orchestrated by a large retinue of channels and pumps, transporters, buffers and sensors. SNc DA neurons have low Ca2+ buffering capacity and lower levels of calcium-binding protein calbindin. The 28k Calb1 is present in surviving PD neurons but not in those dying. And SNc DA neurons have autonomous pacemaker activity which is conferred by the presence of L-type calcium channel Cacna1d (CaV1.3). Other pacemaking neurons rely on sodium rather than calcium channels, including neighboring DA neurons such as those in the ventral tegmental area (VTA), which are PD resistant. Neuronal pigments such as lipofuscin and neuromelanin increase with age. Neuromelanin (NM) is thought to confer protection by chelating metals, in particular iron, by binding mitochondrial toxins and by eliminating non-vesicular, cytosolic DA. The levels of NM in PD brains are lower, and lower in dying, versus surviving neurons. The superoxide generated during respiration, if not promptly handled, can give rise to the toxic hydroxyl radical, in the iron-dependent Fenton reaction. Iron can also increase the rate of DA autoxidation. The iron content in this brain area increases with age, particularly Fe(III), which NM coordinates. The distinctive features of SNc DA neurons can render them more vulnerable to stresses. Impaired homeostasis, primarily mitochondria, but also Ca2+, due to age-dependent diminished functional efficiency but augmented by alterations in the pathways within, can have profound detrimental effects.
The extent of DNA promoter methylation affects gene transcription, methylation being generally repressive. The promoter of the SNCA gene is found hypomethylated in PD brains, compared to controls. The Dnmt1 methylase is abundantly expressed in neurons but its levels are decreased in PD patients. Interestingly, SNCA can interact with Dnmt1 and sequesters it from the nucleus, promoting global hypomethylation. Reduced methylation and increased expression/activity of several genes is observed in PD brains, including Cyp2E1 which promotes formation of toxic metabolites and TNF-alpha which promotes increased inflammatory responses. Histone modification is another epigenetic route for regulating gene expression. Among the many modification types, acetylation leads to a more relaxed chromatin structure conducive to transcription whereas deacetylation promotes a tighter chromatin structure, inhibiting transcription. Differential acetylation of histone 3 or 4 is observed in PD patients. In addition to DNA methylation and histone modification, chromatin modulators affect the differential expression of genes. Both SNCA and Pink1, through their interactions have been proposed to alter histone modification and modulator function with probably toxic effects.
micro RNAs (miRNAs) bind largely to the 3'-untranslated (3'-UTR) of target mRNA and repress their expression by inhibiting their translation or targeting them for degradation. Both up and down-regulated miRNAs are found in PD tissues, but the extent of down-regulated miRNA genes appears to be higher. Epigenetic changes, differential DNA methylation and chromatin modification and differential expression of non-coding RNAs, are also a function of normal aging. For instance, both hypo and hypermethylated CpG sites are observed in relation to aging. These patterns may be different in diseases; overall, DNA hypermethylation in cancer, DNA hypomethylation in neurodegenerative diseases. Strategies for epigenetic-based diagnostics and treatment are being considered.