**5. Alpha-synuclein-related synaptic pathology at the synapse: Biochemical characteristics and functional consequences**

Since many recent reports have highlighted that α-synuclein plays a crucial role in the regulation of synaptic functions by modulating lipid synaptic membrane composition and fluidity as well as the trafficking and function of several key synaptic proteins, it has been hypothesized that α-synuclein loss of function at the synapse may be one of the pathophysiological mechanisms of PD. Thus, many post-mortem studies on the brain of PD patients as well as investigations on "in vitro" and "in vivo" models of PD showing αsynuclein accumulation have been focused on the evaluation of the biochemical and functional consequences of α-synuclein accumulation at synaptic sites.

Numerous findings indicate that transgenic overexpression of wild type, A53T, A30P and truncated α-synuclein results in the degeneration of TH-positive terminals with a reduction of striatal dopamine levels (Rockenstein et al., 2002;Fernagut et al., 2007;Ono et al., 2009;Gao et al., 2008;Gomez-Isla et al., 2003;Nieto et al., 2006;Chen et al., 2006). Interestingly, a couple of these reports described specific alterations of crucial synaptic proteins for dopaminergic neuronal functions such as the DAT (Chen et al., 2006;Magen and Chesselet, 2010). However, results of these studies had different conclusions, with the first one showing a

Targeting α-Synuclein-Related

diseases.

**dysfunctions** 

Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy 149

required for fusion, thus leading to neurotransmitter release (Sudhof and Rothman, 2009). Authors found a co-localization between truncated α-synuclein-immunopositive aggregates and SNARE proteins and they showed that both the target-membrane localized SNAREs SNAP-25 and syntaxin-1 were accumulated in the striatum of PD patients, thus indicating that the observations in the SYN120 transgenic mice are of significance for human disease. In line with these findings, Burrè and coauthors (2010) found that triple-knockout mice lacking synucleins develop age dependent neurological impairments, exhibit decreased SNARE-complex assembly, and die prematurely, suggesting that synucleins may function to sustain normal SNARE-complex assembly in presynaptic terminals during aging. However, these results are in conflict with other findings by Darios and colleagues (2010) which showed an inhibitory action of α-synuclein on SNAREs which is mediated by its sequestration of arachidonic acid. Indeed, arachidonic acid–stimulated glutamate release from isolated synaptosomes is enhanced in α-synuclein C57BL/6S null mice. Thus, these latter two findings display controversial conclusions, although it is clear that, since hsyn120 overexpression results in a redistribution of SNAREs, the formation of these species of the protein may contribute to the induction of synaptic hypofunctions in α-synuclein-related

**5.1 Putative pharmacological interventions for** α**-synuclein-related synaptic** 

It has to be taken into account that the identification of a disease modifying therapy to slow, delay or stop the progression of pathology is one of the unmet needs in the management of PD. Multiple candidate targets and neuroprotective drugs have been proposed in recent years as disease-modifying agents by the analysis of their potential functions in experimental models of PD (Olanow and Kieburtz, 2010). Among these agents, molecules acting at synaptic sites have been investigated. In recent years, much research has been dedicated to dopaminergic agonists that have been hypothesized to possess diseasemodifying actions. In particular, several studies in experimental models of PD have shown that, beside their dopamine-like effect on striatal neurons, they may display neuroprotective actions on dopaminergic neurons (Sethy et al., 1997;Le and Jankovic, 2001;Schapira, 2002), and they are also able to block α-synuclein aggregation "in vitro" (Ono et al., 2007;Bellucci et al., 2008). In particular, we recently showed that dopamine D2/D3 receptor agonists can reduce the formation of α-synuclein+DAT-immunopositive inclusions in experimental models of PD (Bellucci et al., 2008) an effect which may be due to the fact that D2 receptors are able to stimulate DAT membrane translocation (Bolan et al., 2007). Thus, it emerges that the critical interaction of α-synuclein with synaptic proteins in dopaminergic terminals may underlie the specificity of α-synuclein toxicity in PD. On this line, in the above cited research report we found that cocaine is also able to prevent the formation of α-synuclein+DATimmunopositive inclusions likely acting through three possible mechanisms: 1) the block of dopamine uptake and the consequent formation of dopamine-toxic metabolites that can exacerbate α-synuclein aggregation, 2) the increase of extracellular dopamine that may result in an higher stimulation of D2/D3 receptors or 3) the direct stimulation of the membrane translocation of DAT and α-synuclein. Indeed, increased DAT and α-synuclein membrane translocation occur in the striatum of cocaine abusers, thus indicating that DAT blockers can modulate the trafficking of both proteins and confirming that α-synuclein plays a key role in the control of dopamine synaptic tone (Qin et al., 2005). Interestingly, DAT inhibitors have been shown to be able to alleviate specific parkinsonian motor deficits in 1-

reduction of DAT levels in the striatum and nucleus accumbens while the second one demonstrating an increase in DAT density. More recently, it was shown that DAT coimmunoprecipitates with both full length and C-terminally truncated α-synuclein in dopaminergic differentiated SH-SY5Y cells and its membrane levels are decreased in parallel with an increase of α-synuclein levels and aggregation (Bellucci et al., 2008). In this same research report, authors described that the DAT is retained into α-synucleinimmunopositive inclusions following α-synuclein aggregation. In addition, the DAT is specifically redistributed in striatal synapses in the brain of a human (1-120) C-terminallytruncated transgenic mouse model of PD (Garcia-Reitbock et al., 2010), thus suggesting that alterations in its subcellular distribution, mediated by the pathological aggregation of αsynuclein within striatal dopaminergic synapses, may underlie the beginning of the nigrostriatal dopaminergic loss of function. Hence, a decrease of DAT membrane content due to the fact that the protein is retained into α-synuclein intracellular inclusions, may be one of the first central pathologic events leading to dopaminergic neuron loss of function and degeneration. Most importantly, a reduction of the membrane content of the DAT may underlie the onset of PD-related motor symptoms. The critical relevance of α-synuclein in modulating DAT levels is also confirmed by recent findings describing that DAT expression is specifically decreased in fetal embryonic nigral grafts showing LB-like pathology, while TH and VMAT2 expression are unaffected (Kordower et al., 2008a). Thus, it emerges that the DAT is one of the protein member of the α-synuclein-related synaptic proteome which could be pivotally involved in the onset of PD-related synaptic dysfunctions.

However, alterations in other neurotransmitter systems relevant for PD, such as reductions in striatal serotonin (5-HT) or cortical noradrenalin (NE) levels, as well as increased cortical 5-HT and NE levels have also been detected in several α-synuclein transgenic mouse models (Magen and Chesselet, 2010). This is a crucial point, as cell loss in PD occurs not only in basal ganglia circuits. Indeed, changes in both 5-HT and NE levels are thought to underlie the manifestation of early symptoms of PD (Simola et al., 2010;Barone, 2010). Furthermore, α-synuclein knock out has been demonstrated to impair mobilization of glutamate from the reserve pool, thus confirming that the protein may play a role in the regulation of synaptic exocytotic mechanisms (Gureviciene et al., 2007). However, transgenic overexpression of either wild type or A30P mutated α-synuclein does not perturbe the functions of endogenous mouse α-synuclein in glutamate mobilization (Gureviciene et al., 2007), thus indicating that the effect of α-synuclein-mediated synaptic control is specific for each neurotransmitter pathway, and that monoaminergic neurons are significantly more sensitive to its pathological changes and accumulation. This notwithstandings, it emerges that α-synuclein is critically involved in the induction of synaptic dysfunctions not necessarily implicating a direct correlation with the dopaminergic system, because it could affect the basal molecular machinery involved in synaptic release. Remarkably, Garcia-Reitböck and colleagues (2010) have shown that overexpression of C-terminally truncated (1-120) human α-synuclein (hsyn120) impaired synaptic vesicle release "in vitro". In parallel, they found that transgenic overexpression of hsyn120 resulted in the redistribution of SNARE proteins at striatal dopaminergic terminals and coincided with a reduction of dopamine release (Garcia-Reitböck et al., 2010) further supporting that α-synuclein can regulate SNARE complex assembly. SNAREs are pivotally involved in synaptic vesicles release. In particular, vesicular (synaptobrevin-2 and synaptotagmin) SNAREs (V-SNAREs) and target membrane-localized (SNAP-25 and syntaxin-1) SNAREs (T-SNAREs) zipper up into an α-helical bundle that pulls the two membranes tightly together to exert the force

reduction of DAT levels in the striatum and nucleus accumbens while the second one demonstrating an increase in DAT density. More recently, it was shown that DAT coimmunoprecipitates with both full length and C-terminally truncated α-synuclein in dopaminergic differentiated SH-SY5Y cells and its membrane levels are decreased in parallel with an increase of α-synuclein levels and aggregation (Bellucci et al., 2008). In this same research report, authors described that the DAT is retained into α-synucleinimmunopositive inclusions following α-synuclein aggregation. In addition, the DAT is specifically redistributed in striatal synapses in the brain of a human (1-120) C-terminallytruncated transgenic mouse model of PD (Garcia-Reitbock et al., 2010), thus suggesting that alterations in its subcellular distribution, mediated by the pathological aggregation of αsynuclein within striatal dopaminergic synapses, may underlie the beginning of the nigrostriatal dopaminergic loss of function. Hence, a decrease of DAT membrane content due to the fact that the protein is retained into α-synuclein intracellular inclusions, may be one of the first central pathologic events leading to dopaminergic neuron loss of function and degeneration. Most importantly, a reduction of the membrane content of the DAT may underlie the onset of PD-related motor symptoms. The critical relevance of α-synuclein in modulating DAT levels is also confirmed by recent findings describing that DAT expression is specifically decreased in fetal embryonic nigral grafts showing LB-like pathology, while TH and VMAT2 expression are unaffected (Kordower et al., 2008a). Thus, it emerges that the DAT is one of the protein member of the α-synuclein-related synaptic proteome which

could be pivotally involved in the onset of PD-related synaptic dysfunctions.

However, alterations in other neurotransmitter systems relevant for PD, such as reductions in striatal serotonin (5-HT) or cortical noradrenalin (NE) levels, as well as increased cortical 5-HT and NE levels have also been detected in several α-synuclein transgenic mouse models (Magen and Chesselet, 2010). This is a crucial point, as cell loss in PD occurs not only in basal ganglia circuits. Indeed, changes in both 5-HT and NE levels are thought to underlie the manifestation of early symptoms of PD (Simola et al., 2010;Barone, 2010). Furthermore, α-synuclein knock out has been demonstrated to impair mobilization of glutamate from the reserve pool, thus confirming that the protein may play a role in the regulation of synaptic exocytotic mechanisms (Gureviciene et al., 2007). However, transgenic overexpression of either wild type or A30P mutated α-synuclein does not perturbe the functions of endogenous mouse α-synuclein in glutamate mobilization (Gureviciene et al., 2007), thus indicating that the effect of α-synuclein-mediated synaptic control is specific for each neurotransmitter pathway, and that monoaminergic neurons are significantly more sensitive to its pathological changes and accumulation. This notwithstandings, it emerges that α-synuclein is critically involved in the induction of synaptic dysfunctions not necessarily implicating a direct correlation with the dopaminergic system, because it could affect the basal molecular machinery involved in synaptic release. Remarkably, Garcia-Reitböck and colleagues (2010) have shown that overexpression of C-terminally truncated (1-120) human α-synuclein (hsyn120) impaired synaptic vesicle release "in vitro". In parallel, they found that transgenic overexpression of hsyn120 resulted in the redistribution of SNARE proteins at striatal dopaminergic terminals and coincided with a reduction of dopamine release (Garcia-Reitböck et al., 2010) further supporting that α-synuclein can regulate SNARE complex assembly. SNAREs are pivotally involved in synaptic vesicles release. In particular, vesicular (synaptobrevin-2 and synaptotagmin) SNAREs (V-SNAREs) and target membrane-localized (SNAP-25 and syntaxin-1) SNAREs (T-SNAREs) zipper up into an α-helical bundle that pulls the two membranes tightly together to exert the force

required for fusion, thus leading to neurotransmitter release (Sudhof and Rothman, 2009). Authors found a co-localization between truncated α-synuclein-immunopositive aggregates and SNARE proteins and they showed that both the target-membrane localized SNAREs SNAP-25 and syntaxin-1 were accumulated in the striatum of PD patients, thus indicating that the observations in the SYN120 transgenic mice are of significance for human disease. In line with these findings, Burrè and coauthors (2010) found that triple-knockout mice lacking synucleins develop age dependent neurological impairments, exhibit decreased SNARE-complex assembly, and die prematurely, suggesting that synucleins may function to sustain normal SNARE-complex assembly in presynaptic terminals during aging. However, these results are in conflict with other findings by Darios and colleagues (2010) which showed an inhibitory action of α-synuclein on SNAREs which is mediated by its sequestration of arachidonic acid. Indeed, arachidonic acid–stimulated glutamate release from isolated synaptosomes is enhanced in α-synuclein C57BL/6S null mice. Thus, these latter two findings display controversial conclusions, although it is clear that, since hsyn120 overexpression results in a redistribution of SNAREs, the formation of these species of the protein may contribute to the induction of synaptic hypofunctions in α-synuclein-related diseases.

#### **5.1 Putative pharmacological interventions for** α**-synuclein-related synaptic dysfunctions**

It has to be taken into account that the identification of a disease modifying therapy to slow, delay or stop the progression of pathology is one of the unmet needs in the management of PD. Multiple candidate targets and neuroprotective drugs have been proposed in recent years as disease-modifying agents by the analysis of their potential functions in experimental models of PD (Olanow and Kieburtz, 2010). Among these agents, molecules acting at synaptic sites have been investigated. In recent years, much research has been dedicated to dopaminergic agonists that have been hypothesized to possess diseasemodifying actions. In particular, several studies in experimental models of PD have shown that, beside their dopamine-like effect on striatal neurons, they may display neuroprotective actions on dopaminergic neurons (Sethy et al., 1997;Le and Jankovic, 2001;Schapira, 2002), and they are also able to block α-synuclein aggregation "in vitro" (Ono et al., 2007;Bellucci et al., 2008). In particular, we recently showed that dopamine D2/D3 receptor agonists can reduce the formation of α-synuclein+DAT-immunopositive inclusions in experimental models of PD (Bellucci et al., 2008) an effect which may be due to the fact that D2 receptors are able to stimulate DAT membrane translocation (Bolan et al., 2007). Thus, it emerges that the critical interaction of α-synuclein with synaptic proteins in dopaminergic terminals may underlie the specificity of α-synuclein toxicity in PD. On this line, in the above cited research report we found that cocaine is also able to prevent the formation of α-synuclein+DATimmunopositive inclusions likely acting through three possible mechanisms: 1) the block of dopamine uptake and the consequent formation of dopamine-toxic metabolites that can exacerbate α-synuclein aggregation, 2) the increase of extracellular dopamine that may result in an higher stimulation of D2/D3 receptors or 3) the direct stimulation of the membrane translocation of DAT and α-synuclein. Indeed, increased DAT and α-synuclein membrane translocation occur in the striatum of cocaine abusers, thus indicating that DAT blockers can modulate the trafficking of both proteins and confirming that α-synuclein plays a key role in the control of dopamine synaptic tone (Qin et al., 2005). Interestingly, DAT inhibitors have been shown to be able to alleviate specific parkinsonian motor deficits in 1-

Targeting α-Synuclein-Related

Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy 151

Since dopaminergic neuronal dysfunctions are central to the pathogenesis of PD, it has been recently hypothesized that a restoration of the correct function of dopamine neurons may be achieved by the employment of gene based strategies. In particular, a multicistronic lentiviral vector encoding three crucial genes involved in dopamine synthesis: TH, aromatic L-amino acid decarboxylase and guanosine 5'-triphosphate cyclohydrolase has been produced and tested in experimental models of PD, including MPTP-treated primates. Results of these studies indicate that this gene-based dopamine replacement may be able to correct motor deficits. However, the models used to test these compounds do not exactly reproduce all the features of PD. Indeed, toxin-based models don't show α-synuclein inclusions, or at least significant pathological modifications of α-synuclein has been reported only by one group in MPTP-treated squirrel monkeys (McCormack et al 2008). This is not surprising, as mitochondrial dysfunctions have been linked to the induction of α-synuclein aggregation, but these features doesn't necessarily resemble the disease. Indeed, dopaminergic degeneration in toxin-based models is achieved through methods that rapidly and drastically lead to neuronal death and thus they don't reproduce the exact pathological series of event that leads to this phenomenon in PD patients. Therefore, whether PD onset and progression are dependent on the pathological accumulation of α-synuclein at striatal terminals, the gene-based dopamine replacement approach may reproduce the effect of L-DOPA therapy without affectively modifying the pathophysiological aspects of the disease. Indeed, in light of the relevant role played by α-synuclein in the control of synaptic activity, a putative disease modifying intervention for PD should block α-synuclein-related synaptic dysfunctions. Indeed, its pathological accumulation impairs synaptic release through multiple mechanisms (Figure 2), thus implying that even in the presence of dopamine, the

**5.2 Gene-based strategies to block or delay synaptic dysfunctions in PD** 

synaptic terminals containing α-synuclein aggregates still don't act.

A concept that is reinforced by the fact that α-synuclein can be transmitted from cell to cell, thus implying that correction of dopamine deficiency may be uneffective also because of the spreading of α-synuclein through adjacent synapses. Thus, again, targeting the α-synucleinrelated synaptic proteome may represent a powerful tool to block or delay disease progression. As previously described, α-synuclein is able to interact with key synaptic proteins, such as DAT and synapsin 1. The synapsins are a family of neuron-associated proteins involved in the regulation of synaptic vesicle release (Cesca et al., 2010;Kile et al., 2010). Synapsin 3 is likely the only isoform which is present in striatal dopaminergic terminals. Indeed, synapsin 1 and 2 have been found to miscolocalize with VMAT2 in the neostriatum of mice (Bogen et al., 2006) although the mRNA for either forms is present in the striatum (Kile et al., 2010). Remarkably, synapsin 3 is a negative regulator of striatal dopamine release (Kile et al., 2010). Although studies reporting a possible involvement of synapsin 3 in the pathogenesis of α-synuclein-related disease are lacking, previous studies (Galvin et al., 1999) have shown that synapsin 1 appears to accumulate in α-synucleinimmunopositive lesions in axon terminals of the hippocampus of PD patients. Furthermore, synapsin 1 and α-synuclein share numerous functional and biochemical similarities. Indeed, besides being both part of the DAT (Maiya et al., 2007) and synaptic proteome (Burre and Volknandt, 2007), they are co-transported from the cell body to the axon by the slow component b as multiprotein complexes without being affected by actin depletion (Roy et al., 2007;Roy et al., 2008) and they can bind to synaptic vesicles and interact with presynaptic membranes and actin cytoskeleton, modulating the dynamics of the actin-based network during the exo-endocytotic cycle (Cesca et al., 2010;Tofaris and Spillantini, 2007).

methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated monkeys (Madras et al., 2006). Thus, drugs acting on protein members of the α-synuclein synaptic proteome may be effective for PD. However, the risk for abuse liability strongly unwarrant the establishment of clinical trials with DAT inhibitors. Furthermore, positive trial results with dopaminergic agonists using surrogate neuroimaging biomarkers as primary endpoints could not determinate with certainty if reduced rate of decline of the biomarker was due to a drug-mediated protective effect on dopamine neurons or pharmacologic effects on the biomarker that had nothing to do with their integrity (Rascol and Montastruc, 2000). In addition, a recent clinical report, has shown that dopamine agonist treatment of PD carries a substantial risk of pathological behaviors (Hassan et al., 2011;Ceravolo et al., 2009). These evidences are complicated by the fact that these drugs are usually associated with a wide panel of unwanted side effects such as cardiovascular and gastrointestinal troubles, fibrosis and sudden sleep attacks (Simola et al., 2010).

Among the PD-related therapeutic agents monoamine oxidases-B (MAO-B) inhibitors are widely used to attenuate dopamine catabolism and increase dopaminergic tone (Magyar et al., 2004;Simola et al., 2010). Selegiline is the most used MAO-B inhibitor since it can ameliorate motor impairments in early and moderate PD although its efficacy is weaker than L-DOPA (Caslake et al., 2009;Simola et al., 2010) and it has been recently shown to prevent α-synuclein aggregation (Ono et al., 2007;Braga et al., 2011), thus opening the way to the possibility that this agent may effectively display disease modifying properties by slowing α-synuclein-aggregation and the related synaptic pathology.

Entacapone and tolcapone, two Catechol-O-Methyl Transferase (COMT) inhibitors that also counteract dopamine metabolism, have shown to be effective as L-DOPA adjuvant for the control of motor symptoms and of the wearing off on fluctuations phases of PD. Remarkably, both these drugs are able to block fibril formation of α-synuclein (Di Giovanni et al., 2010) and therefore could display some beneficial effects by slowing αsynuclein-related pathology. However, a strong limitation for the employment of these drugs derives from the fact that they display a wide panel of untoward effects such as vomiting, nausea and hypotension, or in the sole case of tolcapone, hepatotoxicity (Simola et al., 2010).

Finally, in line with previously cited evidences indicating that cholesterol metabolites can accelerate α-synuclein fibril formation, cholesterol lowering agents have been shown to reduce α-synuclein accumulation and the associated neuronal pathology "in vitro" and "in vivo" (Koob et al., 2010). Thus it is likely that these compounds may be able to exert their effects by lowering the α-synuclein-related pathology at the synapse, but further investigation is needed in order to clarify these aspects.

Nonetheless, these considerations leave us with the simple conclusion that none of the compounds which has been shown to be effective in reducing PD symptoms could slow or delay disease progression. In particular, alternative therapeutic strategies aimed at counteracting the α-synuclein-related derangement of dopaminergic synapses may represent a powerful tool to slow or delay PD progression and correct the related simptomathology.

Finally, numerous agents have been shown to counteract α-synuclein aggregation, ranging from antioxidants, sex hormones or receptor agonists such as nicortine. Thus, it can be hypothesized that these compounds may affect the related synaptic pathology in PD. However, at present, none of these molecules has been demonstrated to be effective in modifying or slowing disease progression.

methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated monkeys (Madras et al., 2006). Thus, drugs acting on protein members of the α-synuclein synaptic proteome may be effective for PD. However, the risk for abuse liability strongly unwarrant the establishment of clinical trials with DAT inhibitors. Furthermore, positive trial results with dopaminergic agonists using surrogate neuroimaging biomarkers as primary endpoints could not determinate with certainty if reduced rate of decline of the biomarker was due to a drug-mediated protective effect on dopamine neurons or pharmacologic effects on the biomarker that had nothing to do with their integrity (Rascol and Montastruc, 2000). In addition, a recent clinical report, has shown that dopamine agonist treatment of PD carries a substantial risk of pathological behaviors (Hassan et al., 2011;Ceravolo et al., 2009). These evidences are complicated by the fact that these drugs are usually associated with a wide panel of unwanted side effects such as cardiovascular and gastrointestinal troubles, fibrosis and sudden sleep attacks (Simola et

Among the PD-related therapeutic agents monoamine oxidases-B (MAO-B) inhibitors are widely used to attenuate dopamine catabolism and increase dopaminergic tone (Magyar et al., 2004;Simola et al., 2010). Selegiline is the most used MAO-B inhibitor since it can ameliorate motor impairments in early and moderate PD although its efficacy is weaker than L-DOPA (Caslake et al., 2009;Simola et al., 2010) and it has been recently shown to prevent α-synuclein aggregation (Ono et al., 2007;Braga et al., 2011), thus opening the way to the possibility that this agent may effectively display disease modifying properties by

Entacapone and tolcapone, two Catechol-O-Methyl Transferase (COMT) inhibitors that also counteract dopamine metabolism, have shown to be effective as L-DOPA adjuvant for the control of motor symptoms and of the wearing off on fluctuations phases of PD. Remarkably, both these drugs are able to block fibril formation of α-synuclein (Di Giovanni et al., 2010) and therefore could display some beneficial effects by slowing αsynuclein-related pathology. However, a strong limitation for the employment of these drugs derives from the fact that they display a wide panel of untoward effects such as vomiting, nausea and hypotension, or in the sole case of tolcapone, hepatotoxicity (Simola

Finally, in line with previously cited evidences indicating that cholesterol metabolites can accelerate α-synuclein fibril formation, cholesterol lowering agents have been shown to reduce α-synuclein accumulation and the associated neuronal pathology "in vitro" and "in vivo" (Koob et al., 2010). Thus it is likely that these compounds may be able to exert their effects by lowering the α-synuclein-related pathology at the synapse, but further

Nonetheless, these considerations leave us with the simple conclusion that none of the compounds which has been shown to be effective in reducing PD symptoms could slow or delay disease progression. In particular, alternative therapeutic strategies aimed at counteracting the α-synuclein-related derangement of dopaminergic synapses may represent a powerful tool to slow or delay PD progression and correct the related

Finally, numerous agents have been shown to counteract α-synuclein aggregation, ranging from antioxidants, sex hormones or receptor agonists such as nicortine. Thus, it can be hypothesized that these compounds may affect the related synaptic pathology in PD. However, at present, none of these molecules has been demonstrated to be effective in

slowing α-synuclein-aggregation and the related synaptic pathology.

investigation is needed in order to clarify these aspects.

modifying or slowing disease progression.

al., 2010).

et al., 2010).

simptomathology.

### **5.2 Gene-based strategies to block or delay synaptic dysfunctions in PD**

Since dopaminergic neuronal dysfunctions are central to the pathogenesis of PD, it has been recently hypothesized that a restoration of the correct function of dopamine neurons may be achieved by the employment of gene based strategies. In particular, a multicistronic lentiviral vector encoding three crucial genes involved in dopamine synthesis: TH, aromatic L-amino acid decarboxylase and guanosine 5'-triphosphate cyclohydrolase has been produced and tested in experimental models of PD, including MPTP-treated primates. Results of these studies indicate that this gene-based dopamine replacement may be able to correct motor deficits. However, the models used to test these compounds do not exactly reproduce all the features of PD. Indeed, toxin-based models don't show α-synuclein inclusions, or at least significant pathological modifications of α-synuclein has been reported only by one group in MPTP-treated squirrel monkeys (McCormack et al 2008). This is not surprising, as mitochondrial dysfunctions have been linked to the induction of α-synuclein aggregation, but these features doesn't necessarily resemble the disease. Indeed, dopaminergic degeneration in toxin-based models is achieved through methods that rapidly and drastically lead to neuronal death and thus they don't reproduce the exact pathological series of event that leads to this phenomenon in PD patients. Therefore, whether PD onset and progression are dependent on the pathological accumulation of α-synuclein at striatal terminals, the gene-based dopamine replacement approach may reproduce the effect of L-DOPA therapy without affectively modifying the pathophysiological aspects of the disease. Indeed, in light of the relevant role played by α-synuclein in the control of synaptic activity, a putative disease modifying intervention for PD should block α-synuclein-related synaptic dysfunctions. Indeed, its pathological accumulation impairs synaptic release through multiple mechanisms (Figure 2), thus implying that even in the presence of dopamine, the synaptic terminals containing α-synuclein aggregates still don't act.

A concept that is reinforced by the fact that α-synuclein can be transmitted from cell to cell, thus implying that correction of dopamine deficiency may be uneffective also because of the spreading of α-synuclein through adjacent synapses. Thus, again, targeting the α-synucleinrelated synaptic proteome may represent a powerful tool to block or delay disease progression. As previously described, α-synuclein is able to interact with key synaptic proteins, such as DAT and synapsin 1. The synapsins are a family of neuron-associated proteins involved in the regulation of synaptic vesicle release (Cesca et al., 2010;Kile et al., 2010). Synapsin 3 is likely the only isoform which is present in striatal dopaminergic terminals. Indeed, synapsin 1 and 2 have been found to miscolocalize with VMAT2 in the neostriatum of mice (Bogen et al., 2006) although the mRNA for either forms is present in the striatum (Kile et al., 2010). Remarkably, synapsin 3 is a negative regulator of striatal dopamine release (Kile et al., 2010). Although studies reporting a possible involvement of synapsin 3 in the pathogenesis of α-synuclein-related disease are lacking, previous studies (Galvin et al., 1999) have shown that synapsin 1 appears to accumulate in α-synucleinimmunopositive lesions in axon terminals of the hippocampus of PD patients. Furthermore, synapsin 1 and α-synuclein share numerous functional and biochemical similarities. Indeed, besides being both part of the DAT (Maiya et al., 2007) and synaptic proteome (Burre and Volknandt, 2007), they are co-transported from the cell body to the axon by the slow component b as multiprotein complexes without being affected by actin depletion (Roy et al., 2007;Roy et al., 2008) and they can bind to synaptic vesicles and interact with presynaptic membranes and actin cytoskeleton, modulating the dynamics of the actin-based network during the exo-endocytotic cycle (Cesca et al., 2010;Tofaris and Spillantini, 2007).

Targeting α-Synuclein-Related

suggested by recent findings.

inclusions.

**6. Conclusions** 

PD and other α-synucleinopathies.

dopamine system. Neuron 25:239-252.

**7. Acknowledgements** 

insightful discussion.

**8. References** 

Synaptic Pathology: Novel Clues for Parkinson's Disease Therapy 153

synapsin 3 increase and redistribution may mechanistically induce the misregulation of other α-synuclein/synapsin 3 protein partners such as DAT and synaptobrevin-2 as

Therefore, in the scenario of the dopaminergic synapse, deregualtion of synapsin 3 by the pathological aggregation of α-synuclein may be a key event compromising synaptic

Another way to improve dopaminergic neuronal functions in relation to α-synuclein toxicity at the synapse could be the gene-based delivery of the DAT. Nonetheless, it should be taken into account that, since the DAT can precipitate with α-synuclein within intracellular inclusions, delivery of this protein may also foster the pathological aggregation of αsynuclein, an event which may be controlled by the administration of dopaminergic agonists which are known to prevent the formation of DAT/α-synuclein-immunopositive

A large body of evidence strikingly indicate that α-synuclein pathology at the synapse may be crucially involved in PD pathophysiology. Indeed, α-synuclein accumulation, pathological modifications and aggregation are able to significantly impair synaptic functions in experimental models of PD. In particular, it appears that α-synuclein aggregation at dopaminergic synaptic sites may damage neuronal functions by affecting the correct subcellular distribution of several molecular members of its lipidome and proteome, thus leading to synaptic vesicle stall and to the consequent block of synaptic release. This event may underlie the onset of a perilous axonal degeneration which may retrogradely compromise neuronal resilience, thus leading to dopaminergic cell dysfunctions and death. This view opens the way to the discovery and characterization of novel possible targets to develop drug- and gene-based therapeutical strategies to cure the disease, possibly by modifying and/or slowing disease progression. This is a crucial point, as current therapeutical approaches for PD are aimed at ameliorating disease simptomathology, but none of the present known agent is able to block the progression of the disease. Targeting the α-synuclein-related synaptic proteome by drug-based or gene-based therapeutical interventions may thus represent a new frontier to develop disease modifying strategies for

We are grateful to Dr. Laura Navarria and Michela Zaltieri for technical assistance and

Abeliovich A, Schmitz Y, Farinas I, Choi-Lundberg D, Ho WH, Castillo PE, Shinsky N,

Verdugo JM, Armanini M, Ryan A, Hynes M, Phillips H, Sulzer D, Rosenthal A (2000) Mice lacking alpha-synuclein display functional deficits in the nigrostriatal

functions as well as the correct tethering and fusion of synaptic vesicles.

Noteworthy, despite their different localization in subregions of the adult brain (Pieribone et al., 2002;Cesca et al., 2010) synapsin 1 and 3 share a strong structural homology in their functional domains (Hosaka and Sudhof, 1998), suggesting that they may likely interact with similar proteins and cytoskeletal components, although they differentially regulate neurotransmitter release (Kile et al., 2010;Feng et al., 2002;Cesca et al., 2010).

Fig. 2. Schematic representation of the physiological and pathological localization of αsynuclein within dopaminergic synapses. Alpha-synuclein aggregation may compromise dopaminergic synaptic release by altering the subcellular localization of key proteins involved in this process.

Thus, it is likely that α-synuclein can also interact with- and modulate synapsin 3. As a support to this notion, previous reports have shown that synapsin 1 levels in the striatum of PD patients are comparable to those of control subjects (Girault et al., 1989), thus suggesting that synapsin 1 is not specifically implicated in the pathophysiology of nigrostriatal dopamine deficiency in PD. Thus, α-synuclein and synapsin 3 likely share common regulatory roles in nigrostriatal dopaminergic synapses suggesting that α-synuclein accumulation may critically affect the function of synapsin 3 by altering its levels and distribution. In particular, an increase and redistribution of α-synuclein and synapsin 3 complexes in striatal dopaminergic synapses could be associated to deregulation or loss of dopamine release as both proteins are negative regulators of this event. Furthermore, synapsin 3 increase and redistribution may mechanistically induce the misregulation of other α-synuclein/synapsin 3 protein partners such as DAT and synaptobrevin-2 as suggested by recent findings.

Therefore, in the scenario of the dopaminergic synapse, deregualtion of synapsin 3 by the pathological aggregation of α-synuclein may be a key event compromising synaptic functions as well as the correct tethering and fusion of synaptic vesicles.

Another way to improve dopaminergic neuronal functions in relation to α-synuclein toxicity at the synapse could be the gene-based delivery of the DAT. Nonetheless, it should be taken into account that, since the DAT can precipitate with α-synuclein within intracellular inclusions, delivery of this protein may also foster the pathological aggregation of αsynuclein, an event which may be controlled by the administration of dopaminergic agonists which are known to prevent the formation of DAT/α-synuclein-immunopositive inclusions.
