**Targeting** α**-Synuclein-Related Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy**

Arianna Bellucci1 and PierFranco Spano1,2

*1Division of Pharmacology, Department of Biomedical Sciences and Biotechnologies and National Institute of Neuroscience –Italy, School of Medicine, University of Brescia, Brescia (BS), 2IRCCS San Camillo Hospital, Venice, Italy* 

#### **1. Introduction**

136 Etiology and Pathophysiology of Parkinson's Disease

Tatton NA. (2000). Increased caspase 3 and Bax immunoreactivity accompany nuclear

Van der Walt JM, Martin ER, Scott WK et al. (2003). Genetic polymorphisms of the N-

Vekrellis K, Rideout HJ & Stefanis L. Neurobiology of alpha-synuclein. *Molecular* 

Webb JL, Ravikumar B, Atkins J, Skepper JN & Rubinsztein DC. (2003). Alpha-Synuclein is

Xilouri M, Vogiatzi T, Vekrellis K, Park D & Stefanis L. (2009). Aberrant alpha-synuclein

autophagy. *PLoS One,* Vol.4, No.5, (May 2009), pp. e5515, ISSN 1932-6203 Yang Q, She H, Gearing M, Colla E, Lee M, Shacka JJ & Mao Z. (2009). Regulation of

Yoritaka A, Hattori N, Uchida K, Tanaka M, Stadtman ER & Mizuno Y. (1996).

Zhang J, Perry G, Smith MA, Robertson D, Olson SJ, Graham DG & Montine TJ. (1999).

*Neurobiology,*Vol.30, No.1, (August 2004), pp. 1-21, ISSN 0893-7648

Vol.278, No.27, (July 2003), pp. 25009-13, ISSN 0021-9258

Vol.323, No.5910, (January 2009), pp. 124-7, ISSN 0193-4511

Vol.93, No.7, (April 1996), pp. 2696-701, ISSN 0027-8424

(May 1999), pp. 1423-9, ISSN 0002-9440

*Neurology*, Vol. 166, No.1, (November 2000), pp. 29-43, ISSN 0014-4886 Tokuda T, Qureshi MM, Ardah MT, Varghese S, Shehab SA, Kasai T, Ishigami N, Tamaoka

No.1, (February 2000), pp. 67-75, ISSN 0006-8993

(November 2010), pp. 1766-72, ISSN 0028-3878

(April 2003), pp. 1189-91, ISSN 0028-3878

SY5Y cells via impairment of cellular energy metabolism. *Brain Research*, Vol.855,

GAPDH translocation and neuronal apoptosis in Parkinson's disease. *Experimental* 

A, Nakagawa M & El-Agnaf OM. (2010). Detection of elevated levels of α-synuclein oligomers in CSF from patients with Parkinson disease. *Neurology*, Vol.75, No.20,

acetyltransferase genes and risk of Parkinson's disease. *Neurology*, Vol.60, No.7,

degraded by both autophagy and the proteasome. *The Journal of Biological Chemistry*,

confers toxicity to neurons in part through inhibition of chaperone-mediated

neuronal survival factor MEF2D by chaperone-mediated autophagy. *Science,*

Immunohistochemical detection of 4-hydroxynonenal protein adducts in Parkinson disease. *Proceedings of the National Academy of Sciences of the United States of America*,

Parkinson's disease is associated with oxidative damage to cytoplasmic DNA and RNA in substantia nigra neurons. *The American Journal of Pathology,* Vol.154, No.5, Parkinson's disease (PD) is the most diffuse movement disorder, affecting approximately 6 milion individuals worldwide and presenting tremor, rigidity and bradykinesia as the main clinical features. The onset of PD typically occur in patients over the age of 50 years and its incidence slowly progresses with increasing age. Neuropathologically, it is characterized by loss of striatal-projecting dopaminergic neurons of the substantia nigra pars compacta, and by the presence of Lewy bodies (LB) and Lewy neurites (LN) (Forno, 1996). To date, LB and LN are the characteristic neuropathological alterations of another neurodegenerative disease: Dementia with Lewy bodies (DLB), a form of late-life dementia which eventually overlaps with Alzheimer's disease.

It is noteworthy that, despite the staggering impact of PD on society, the available therapeutic armamentarium for this disease is still limited. Indeed, the gold-standard treatment for PD from the 60'- the dopamine precursor L-DOPA - induces severe motor side effects and its efficacy declines with the progression of the disease. The rest of the pharmacological treatments for PD mainly include drugs that are usually employed in association with L-DOPA such as dopamine agonists, which however fail to match L-DOPA's efficacy. Best results are currently achieved with invasive strategies via subcutaneous or intraduodenal delivery of apomorphine or L-DOPA, or deep brain stimulation of the subthalamic nucleus. Nonetheless, usually after 15 years of pharmacological and therapeutical interventions, most of the patients start to present several motor complications, thus requesting multiple adjuvant strategies such as physiotherapy, hospitalization or social assistance.

For these reasons, in recent years, much effort has been made in order to outline novel therapeutical approaches to cure PD. In particular, interventions such as stem-cell based approaches have been tested as potential treatments for PD. Although results indicate that patients may gain a long-term clinical benefit from the intrastriatal transplantation of human embryonic mesencephalic tissue (Piccini et al., 1999), part of the graft-derived dopaminergic cells develop LB after 11-16 years (Li et al., 2008;Kordower et al., 2008a;Kordower et al., 2008b),

Targeting α-Synuclein-Related

resilience of dopaminergic neurons?

Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy 139

(Braak et al., 2003;Jellinger, 2009;Halliday and McCann, 2010), thus indicating that LB may be somehow related to nerve cell loss. Later on, it was shown that the number of LB in patients with mild to moderate loss of neurons in the substantia nigra is higher than in patients with severe neuronal depletion, thus indicating that LB-containing neurons may be the dying neurons (Wakabayashi et al., 2007). Furthermore, it seems that the presence of LB is not always related to nerve cell degeneration as every dying nerve cell does not necessarely form LB. Indeed, LB-containing neurons of the substantia nigra don't undergo apoptotic cell death to a greater degree than the general population and most neurons that undergo cell death do not contain LB (Schulz-Schaeffer, 2010;Tompkins and Hill, 1997). Finally, substantia nigra neurons, whether they contain LB or not, are similarly affected by morphological dendritic abnormalities or biochemical changes, thus indicating that the neurons in general are involved in the disease process (Bergeron et al., 1996;Hill, 1996;Javoy-

In addition, recent findings indicate that in DLB 90% or even more α-synuclein aggregates are located at presynapses in the form of very small deposits, pointing out that not cell death but rather α-synuclein aggregate-related synaptic dysfunctions may cause the neurodegeneration (Schulz-Schaeffer, 2010;Kramer et al., 2008). It may thus be feasible that PD may be characterized by dying back mechanisms that begin at the synapse and lead to axonal degeneration in the striatum as a consequence of the pathological accumulation of α-synuclein at the synapse. These events may lately drive dopaminergic cells of the substantia nigra toward neurodegeneration. Therefore, α-synuclein is central to the pathogenesis of PD. This view opens the way to a novel question: is it the αsynuclein deposition-associated synaptic pathology which may affect the function and

Alpha-synuclein is a 140 amino acids protein, a fragment of which has been identified as a non A-β component (NAC) in amyloid preparations of Alzheimer's disease (Ueda et al., 1993). It was firstly described as a brain specific protein (Nakajo et al., 1993;George et al., 1995) and since the initial findings indicated that the protein localizes at presynaptic site and portions of the nucleus in brain neuronal cells (Jakes et al., 1994;Iwai et al., 1995) it was denominated with the acronym "synuclein". Recombinant α-synuclein does not assume a uniform consistent secondary structure in aqueous solution, thus the protein is said to be natively unfolded (Weinreb et al., 1996). Several observations have shown that α-synuclein is implicated in the control of synaptic membrane processes and biogenesis (Jenco et al., 1998;Davidson et al., 1998). From these studies, it was hypothesized that α-synuclein could play a key role in the modulation of synaptic activity. Consequently, it has been demonstrated that α-synuclein interacts with- and modulates the expression, subcellular distribution and activity of numerous synaptic proteins as well as cytoskeletal components (Engelender et al., 1999;Bonini and Giasson, 2005;Chandra et al., 2005;Sousa et al., 2009;Scott et al., 2010;Garcia-Reitbock et al., 2010;Darios et al., 2010;Burre et al., 2010). Alpha-synuclein also shares numerous biochemical and functional similarities with synaptic proteins such as synapsins. Furthermore, it has been found to be implicated in the control of neurotransmitter release (Abeliovich et al., 2000;Gureviciene et al., 2007;Nemani et al., 2010) and since the protein is specifically upregulated in a discrete population of presynaptic terminals of the songbird during a period of song acquisition (George et al., 1995), it has been hypothesized that it may be implicated in synaptic plasticity. Thus, it emerges that αsynuclein has a critical role in the control of synaptic activity in specific neuronal populations, rendering it possible to outline a whole new scenario to single out novel

Agid et al., 1990;Patt et al., 1991;Kramer et al., 2008;Schulz-Schaeffer, 2010).

probably because α-synuclein can be transmitted from cell to cell (Desplats et al., 2009), and what's more, the availability of human mesencephalic tissue is limited. Thus, it became clear that lentiviral vectors are ideal candidates to design novel gene-based disease modifying strategies for neurodegenerative diseases such as PD. Indeed, surgical infusion is currently required for viral vector delivery into the brain and PD is the only neurodegenerative disorder routinely treated with neurosurgery (Bjorklund et al., 2009;Bjorklund and Kirik, 2009;Kaplitt, 2010). Furthermore, several animal models of PD are available to test new therapies. Actually, promising preclinical gene therapy studies focusing on the correction of dopamine deficiency are currently in progress (Azzouz et al., 2002;Jarraya et al., 2009). However, a limitation with this level of analysis is that it sidesteps the main goal that a current therapeutical approach for PD should afford: to block and/or counteract dopamine neuron derangement. Indeed, a disease modifying gene therapy approach is missing, as the mechanisms underlying dopamine neuron derangement in PD are not yet completely clear. For these reasons, unravelling the molecular mechanisms underlying PD onset is essential to discover new therapeutic targets to cure the disease.

In recent years, the idea that α-synuclein is a causative agent of PD has spread upon the scientific community, the link being that α-synuclein is deposited in the pathological hallmark of PD, the LB and that α-synuclein mutations are responsible for the onset of familial form of PD (Cookson and van der, 2008;Cookson, 2005). These considerations leave us with the simplest possible sketch of the pathogenesis of PD: α-synuclein is an initiator of damage and the final outputs, after many years, are cell loss and LB formation. However, it has to be taken into account that the majority of PD cases are idiopatic, thus: what are the causes underlying the pathological accumulation of α-synuclein? Many epidemiological studies have been conducted to verify whether environmental or genetic factors may predispose to the development of PD (Elbaz and Moisan, 2008). From these latter, it became evident that a correlation exists between pesticide or chemicals exposure and PD onset. Furthermore, besides α-synuclein, 12 other gene mutations, some of which with unknown function (Belin and Westerlund, 2008) have been found to be responsible for the development of early onset PD. Some of these genes encode mitochondrial associated proteins or members of the ubiquitin-proteasome system. Remarkably, mitochondrial and proteasomal dysfunctions have been linked to the onset of PD (Tritschler et al., 1994). Thus, the overall conclusions of epidemiological studies indicate that PD is a complex, multifactoriated disorder and that α-synuclein aggregation, mitochondrial and proteasomal dysfunctions play central pathogenic roles. Noteworthy, the observation that both mitochondrial and proteasomal inhibitors induce α-synuclein accumulation as well as dysfunction and degeneration of nigrostriatal dopaminergic neurons "in vivo" (Xiong et al., 2009;Xie et al., 2010) remind to the hypothesis that α-synuclein deposition is a central step event during the pathogenesis of PD. This notion has been reinforced by findings showing that the area that degenerate in PD (substantia nigra, striatum and ventral tegmental area) express low levels of α-synuclein in physiological conditions (Wersinger et al., 2004), and that α-synuclein levels in the substantia nigra decrease with age (Mak et al., 2009) supporting the hypothesis that these regions may be more vulnerable to a pathological increase of α-synuclein levels, especially during the aging process. At present, the neuropathological diagnosis of PD and DLB is based on the detection and quantification of LB (Beach et al., 2009b;Beach et al., 2009a;McKeith et al., 1996;McKeith et al., 2005;McKeith, 2006). Indeed, the spreading of LB pathology correlates with the progression of the disease In PD, LB are mainly found at predilection sites of neuronal loss such as the substantia nigra

probably because α-synuclein can be transmitted from cell to cell (Desplats et al., 2009), and what's more, the availability of human mesencephalic tissue is limited. Thus, it became clear that lentiviral vectors are ideal candidates to design novel gene-based disease modifying strategies for neurodegenerative diseases such as PD. Indeed, surgical infusion is currently required for viral vector delivery into the brain and PD is the only neurodegenerative disorder routinely treated with neurosurgery (Bjorklund et al., 2009;Bjorklund and Kirik, 2009;Kaplitt, 2010). Furthermore, several animal models of PD are available to test new therapies. Actually, promising preclinical gene therapy studies focusing on the correction of dopamine deficiency are currently in progress (Azzouz et al., 2002;Jarraya et al., 2009). However, a limitation with this level of analysis is that it sidesteps the main goal that a current therapeutical approach for PD should afford: to block and/or counteract dopamine neuron derangement. Indeed, a disease modifying gene therapy approach is missing, as the mechanisms underlying dopamine neuron derangement in PD are not yet completely clear. For these reasons, unravelling the molecular mechanisms underlying PD onset is essential to discover new

In recent years, the idea that α-synuclein is a causative agent of PD has spread upon the scientific community, the link being that α-synuclein is deposited in the pathological hallmark of PD, the LB and that α-synuclein mutations are responsible for the onset of familial form of PD (Cookson and van der, 2008;Cookson, 2005). These considerations leave us with the simplest possible sketch of the pathogenesis of PD: α-synuclein is an initiator of damage and the final outputs, after many years, are cell loss and LB formation. However, it has to be taken into account that the majority of PD cases are idiopatic, thus: what are the causes underlying the pathological accumulation of α-synuclein? Many epidemiological studies have been conducted to verify whether environmental or genetic factors may predispose to the development of PD (Elbaz and Moisan, 2008). From these latter, it became evident that a correlation exists between pesticide or chemicals exposure and PD onset. Furthermore, besides α-synuclein, 12 other gene mutations, some of which with unknown function (Belin and Westerlund, 2008) have been found to be responsible for the development of early onset PD. Some of these genes encode mitochondrial associated proteins or members of the ubiquitin-proteasome system. Remarkably, mitochondrial and proteasomal dysfunctions have been linked to the onset of PD (Tritschler et al., 1994). Thus, the overall conclusions of epidemiological studies indicate that PD is a complex, multifactoriated disorder and that α-synuclein aggregation, mitochondrial and proteasomal dysfunctions play central pathogenic roles. Noteworthy, the observation that both mitochondrial and proteasomal inhibitors induce α-synuclein accumulation as well as dysfunction and degeneration of nigrostriatal dopaminergic neurons "in vivo" (Xiong et al., 2009;Xie et al., 2010) remind to the hypothesis that α-synuclein deposition is a central step event during the pathogenesis of PD. This notion has been reinforced by findings showing that the area that degenerate in PD (substantia nigra, striatum and ventral tegmental area) express low levels of α-synuclein in physiological conditions (Wersinger et al., 2004), and that α-synuclein levels in the substantia nigra decrease with age (Mak et al., 2009) supporting the hypothesis that these regions may be more vulnerable to a pathological increase of α-synuclein levels, especially during the aging process. At present, the neuropathological diagnosis of PD and DLB is based on the detection and quantification of LB (Beach et al., 2009b;Beach et al., 2009a;McKeith et al., 1996;McKeith et al., 2005;McKeith, 2006). Indeed, the spreading of LB pathology correlates with the progression of the disease In PD, LB are mainly found at predilection sites of neuronal loss such as the substantia nigra

therapeutic targets to cure the disease.

(Braak et al., 2003;Jellinger, 2009;Halliday and McCann, 2010), thus indicating that LB may be somehow related to nerve cell loss. Later on, it was shown that the number of LB in patients with mild to moderate loss of neurons in the substantia nigra is higher than in patients with severe neuronal depletion, thus indicating that LB-containing neurons may be the dying neurons (Wakabayashi et al., 2007). Furthermore, it seems that the presence of LB is not always related to nerve cell degeneration as every dying nerve cell does not necessarely form LB. Indeed, LB-containing neurons of the substantia nigra don't undergo apoptotic cell death to a greater degree than the general population and most neurons that undergo cell death do not contain LB (Schulz-Schaeffer, 2010;Tompkins and Hill, 1997). Finally, substantia nigra neurons, whether they contain LB or not, are similarly affected by morphological dendritic abnormalities or biochemical changes, thus indicating that the neurons in general are involved in the disease process (Bergeron et al., 1996;Hill, 1996;Javoy-Agid et al., 1990;Patt et al., 1991;Kramer et al., 2008;Schulz-Schaeffer, 2010).

In addition, recent findings indicate that in DLB 90% or even more α-synuclein aggregates are located at presynapses in the form of very small deposits, pointing out that not cell death but rather α-synuclein aggregate-related synaptic dysfunctions may cause the neurodegeneration (Schulz-Schaeffer, 2010;Kramer et al., 2008). It may thus be feasible that PD may be characterized by dying back mechanisms that begin at the synapse and lead to axonal degeneration in the striatum as a consequence of the pathological accumulation of α-synuclein at the synapse. These events may lately drive dopaminergic cells of the substantia nigra toward neurodegeneration. Therefore, α-synuclein is central to the pathogenesis of PD. This view opens the way to a novel question: is it the αsynuclein deposition-associated synaptic pathology which may affect the function and resilience of dopaminergic neurons?

Alpha-synuclein is a 140 amino acids protein, a fragment of which has been identified as a non A-β component (NAC) in amyloid preparations of Alzheimer's disease (Ueda et al., 1993). It was firstly described as a brain specific protein (Nakajo et al., 1993;George et al., 1995) and since the initial findings indicated that the protein localizes at presynaptic site and portions of the nucleus in brain neuronal cells (Jakes et al., 1994;Iwai et al., 1995) it was denominated with the acronym "synuclein". Recombinant α-synuclein does not assume a uniform consistent secondary structure in aqueous solution, thus the protein is said to be natively unfolded (Weinreb et al., 1996). Several observations have shown that α-synuclein is implicated in the control of synaptic membrane processes and biogenesis (Jenco et al., 1998;Davidson et al., 1998). From these studies, it was hypothesized that α-synuclein could play a key role in the modulation of synaptic activity. Consequently, it has been demonstrated that α-synuclein interacts with- and modulates the expression, subcellular distribution and activity of numerous synaptic proteins as well as cytoskeletal components (Engelender et al., 1999;Bonini and Giasson, 2005;Chandra et al., 2005;Sousa et al., 2009;Scott et al., 2010;Garcia-Reitbock et al., 2010;Darios et al., 2010;Burre et al., 2010). Alpha-synuclein also shares numerous biochemical and functional similarities with synaptic proteins such as synapsins. Furthermore, it has been found to be implicated in the control of neurotransmitter release (Abeliovich et al., 2000;Gureviciene et al., 2007;Nemani et al., 2010) and since the protein is specifically upregulated in a discrete population of presynaptic terminals of the songbird during a period of song acquisition (George et al., 1995), it has been hypothesized that it may be implicated in synaptic plasticity. Thus, it emerges that αsynuclein has a critical role in the control of synaptic activity in specific neuronal populations, rendering it possible to outline a whole new scenario to single out novel

Targeting α-Synuclein-Related

expression of some o its protein partners.

neurotoxic mechanisms (Albani et al., 2004;Batelli et al., 2008).

Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy 141

The amino acid sequence and subcellular localization of α-synuclein indicate that it can interact with lipid membranes. Indeed, the repeat region, mediates reversible binding to acidic phospholipids by making up a conserved apolipoprotein-like class-A2 helix which is associated with a large shift in protein secondary structure from around 3% to about 80% α-helical (Davidson et al., 1998). Consistently, it has been observed that α-synuclein can inhibit a protein which is localized to plasma membrane and submembraneous vesicles: phospholipase D2, whose activity produces phosphatidic acid, which is implicated in vesicle budding (Jenco et al., 1998). Membrane-bound α-synuclein has an high aggregation propensity and seeds the aggregation of the cytosolic form (Lee and Lee, 2002). Lipids can facilitate the incorporation of α-synuclein into membranes and influence α-synuclein fibril elongation (Gai et al., 2000) suggesting that α-synuclein may be strongly implicated in synaptic membrane biogenesis. Furthermore, the key role of α-synuclein in membrane-associated processes is supported by findings indicating that α-synuclein knock-out mice have enhanced dopamine release at nigrostriatal terminals in response to electrical stimulation, indicating that α-synuclein is an activity-dependent negative regulator of dopaminergic transmission (Abeliovich et al., 2000). On this line, it has been shown that depletion of α-synuclein results in a decrease in the distal pool of presynaptic vesicles in cultured hippocampal neurons (Murphy et al., 2000). Furthermore, its overexpression reduces neurotransmitter release by inhibiting synaptic vesicle reclustering after endocytosis (Nemani et al., 2010). Cytosolic proteins regulate αsynuclein membrane interactions, thus suggesting that cytosolic cofactors may be implicated in disease pathogenesis (Wislet-Gendebien et al., 2006;Wislet-Gendebien et al., 2008). In particular, protein-protein interactions may affect α-synuclein ability to bind to synaptic membranes, thus affecting its biochemical properties and modulating its tendency toward aggregation, suggesting that, in the α-synuclein synaptic proteome, it could be possible to modulate α-synuclein aggregation by selectively regulating the

The relevance of the formation of protein complexes in the modulation of α-synuclein function is supported by several findings which suggest that the protein likely behave like a molecular chaperone. Indeed, some studies have shown that α-synuclein shares functional and physical homologies with a family of ubiquitous cytoplasmic chaperones: the 14-3-3 proteins (Ostrerova et al., 1999). Alpha-synuclein can bind 14-3-3 proteins as well as other protein members of its family proteome such as protein kinase C. Thus, it was hypothesized that an increase of α-synuclein levels could be harmful for the cells as it could affect signal transduction pathways involved in cell differentiation and survival. On this line, in 2001, Iwata and colleagues showed that α-synuclein inhibited mitogen-activated protein (MAP) kinase signaling and accelerated cell death following serum reduction in neuro2a cells. However, results from later studies indicated that overexpression of wild type α-synuclein can protect neuronal cells from apoptotic stimuli and delay cell death induced by serum withdrawal (Alves da Costa et al., 2000;Lee et al., 2001b), thus leading to controversial conclusions. Indeed, it has also been reported that α-synuclein protects against oxidative stress by inactivating c-jun-N-terminal kinase, which is strongly implicated in stress responses (Hashimoto et al., 2002). However, it is likely that the neuroprotective/toxic effects of α-synuclein may be ascribed to a different sensitivity of cells to α-synuclein levels. Indeed, it has been shown that at a nanomolar scale TAT-α-synuclein is neuroprotective against oxidative stress, while at a micromolar scale it is able to aggregate and trigger

therapeutical targets among the α-synuclein synaptic partners. Modulation of these proteins may open new ways toward the development of disease modifying strategies to cure PDrelated synaptic dysfunctions.

By performing a critical review of the PD-related α-synuclein proteome, this article will outline the most relevant findings defining the specific modulatory effects exerted by αsynuclein in the control of synaptic functions in physiological and pathological conditions. The overall conclusions of these studies will spot novel potential therapeutic targets for the development of pharmacological and gene-based strategies aimed at straightening αsynuclein-related synaptic dysfunctions as new clues to cure PD.
