**5. Neuropathologic outcome of molecular physiopathology: More than nigral degeneration**

In the previous paragraphs we have reported evidences regarding the occurrence of specific deleterious processes in degeneration of dopaminergic neurons; furthermore, we have highlighted the functional correlations that have been demonstrated among these processes. A wide body of data confirms that each of these mechanisms can be responsible for degeneration of dopaminergic neurons and involutional modifications of the substantia nigra. These modifications are responsible for the motor symptoms that represent the major criteria for diagnosis of Parkinson's disease. The reasons of the selective damage of nigral neurons is not currently known; particular biological conditions might favor the onset of pathogenetic mechanisms in these neurons or could be responsible for an extreme vulnerability to systemic biochemical alterations.

However, the selectiveness of nigral involvement in Parkinson's disease is not absolute. In fact, the cardinal symptoms of Parkinson's disease are often preceded by a series of early manifestations, such as depression, sleep disturbances, eye movement disorders, hyposmia, constipation and loss of cardiac and vasomotor reflexes, which can start several years before the motor disturbances. The neuropathologic substrate of these aspecific manifestations does not seem to be ascribable to the degeneration of substantia nigra, rather they would derive from the involvement of other monoaminergic neurons in central and peripheral nervous system. Postmortem studies in patients with Parkinson's disease have provided evidence of a typical temporal pattern of pathology progression from peripheral neurons, spinal cord and brainstem to basal ganglia and cortical regions. These studies have revealed the precocious degeneration of autonomic ganglia of mesenteric plexus, dorsal motor nucleus of vagus, rafe nuclei, locus ceruleus and ventral tegmental area; the involvement of these centers is consistent with the clinical symptoms preceding parkinsonian signs. Based on these premises, the degeneration of substantia nigra appears as a late manifestation of the neuropathologic process; moreover, the onset of the characteristic motor manifestations is usually insidious and can be appreciated only when about 50% to 60% of dopaminergic neurons have been lost. These remarks indicate that the diagnosis of Parkinson's disease is always made when the underlying neuropathologic process has already reached an advanced phase; this limits the benefits that can be expected from putative neuroprotective therapies. Hence, it would be desirable to be endowed of diagnostic tools useful to make diagnosis of Parkinson's disease during early or even pre-clinical stages. This assumption has promoted the pre-clinical and clinical research aimed at identifying biomarkers of disease.

#### **5.1 Peripheral biomarkers**

Different rationales justify the use of biomarkers for Parkinson's disease:

Alpha-Synuclein, Oxidative Stress and

patients with Parkinson's disease.

markers of disease.

subsequent experiments (Brighina et al 2010).

Autophagy Failure: Dangerous Liaisons in Dopaminergic Neurodegeneration 127

Actually, among the putative peripheral biochemical markers which have been investigated,

Alpha-synuclein levels in peripheral blood cells from patients with Parkinson's disease do not seem to differ from controls (Brighina et al 2010), even if an increased expression of the alpha-synuclein gene has been detected in fibroblasts from patients (Hoepken et al 2008). Data on total plasma level of the protein in patients compared to controls are contrasting but new alpha-synuclein assays able to detect oligomeric protein have demonstrated changes in plasma and cerebrospinal fluid from diseased patients. Recently, the concentration of alphasynuclein oligomers and the ratio of oligomers on total alpha-synuclein have been found significantly increased in cerebrospinal fluid from diseased patients compared to controls (Tokuda et al 2010). The oligomeric soluble fraction of alpha-synuclein has been reported to

A reduction of glutamate uptake has been observed in platelets from patients and the decrease has showed a positive correlation with clinical severity of the disease (Ferrarese et al. 1999, 2001). This finding indicates that a systemic defect in glutamate transport occurs in

Some studies have reported a decrease in proteasome function in lymphocytes from diseased patients (Blandini et al 2006), but this finding has not been confirmed by

Oxidative stress is induced in lymphocytes from patients with Parkinson's disease and inversely correlates with L-dopa daily doses (Prigione et al 2006, 2009). Increased nitrotyrosine modifications of alpha-synuclein have been found in peripheral blood mononuclear cells obtained from individuals with idiopathic Parkinson's disease compared to controls; moreover, the amount of nitrotyrosine-modified alpha-synuclein has been demonstrated to positively correlate with intracellular levels of reactive oxygen species; in the same ex vivo model a significant increase in LC3II level have been demonstrated in diseased patients (Prigione et al 2010). Therefore, in patients with sporadic Parkinson's disease, a systemic dysregulation of autophagy and oxidative stress seem to exist and correlate with alpha-synuclein post-translational modifications. This confirms the chance of identifying peripheral biomarkers able to mirror the activation of pathogenetic processes in the central nervous system: biochemical parameters of autophagy and oxidative stress, together with qualitative and quantitative alterations of alpha-synuclein, could be useful for diagnostic or prognostic purposes and to monitor the biochemical effects of putative neuroprotective therapies. Moreover, the physiopathologic correlation among these mechanisms could offer the possibility of a combined use of these parameters to increase the accuracy of single biomarkers. In particular, the specificity of autophagy dysregulation in peripheral models, not confirmed in other neurodegenerative diseases, qualifies autophagy machinery as a new and promising target of research in the field of peripheral biochemical

It is conceivable that neuroprotective drugs would have more chance of success if they are administered during early stages of disease, when the molecular pathogenetic processes occurring in dopaminergic neurons could be more easily blocked; therefore, the availability of tools useful to make early diagnosis would increase the potential neuroprotective effects of old and new putative therapies. Therefore, development of disease biomarkers is strongly related to the identification of drugs able to delay the progression of Parkinson's disease or

even to prevent its onset if diagnosis is made in preclinical stages.

no one has a reliability which could justify its use in clinical practice.

be increased in plasma samples obtained from patients.


Neuroimaging biomarkers include detection of dopamine transporter (DAT) levels in basal ganglia with radiolabeled tracers; 123I-FP-CIT SPECT, commonly known as DATscan, is actually used in clinical practice to assist in the differential diagnosis of movement disorders. Anyway, DAT depletion follows the death of nigral neurons; furthermore, the exposure to radiations and the high costs limit the use of this test in screening and follow up.

Peripheral biochemical markers would offer the advantage of being easily accessible for repeated measurements, feature that makes them particularly useful not only for diagnosis but even for monitoring the course of the disease. Various peripheral cell types share with neurons the molecular machinery for several biochemical processes; this provides the chance of testing peripheral ex-vivo models to seek for disease-specific biochemical alterations. The ideal biomarker should reflect a pathogenetic mechanism operative at central level (core biomarker) and playing a major role in the patophysiology of Parkinson's disease, such as alpha-synuclein aggregation, mitochondrial dysfunction, oxidative stress and impairment of the clearance machinery. Therefore, it is essential to identify peripheral models reproducing the molecular alterations detected in the affected neurons. Up to now, a wide range of putative peripheral models of disease have been proposed: platelets, peripheral blood mononuclear cells and fibroblasts represent ex vivo tissues which are actually under investigation with the aim of disclosing differences in biochemical parameters between patients with Parkinson's disease and controls; even plasma is currently studied with the same purpose. The accessibility of peripheral tissues represent the clearest advantage of their use, significantly compensating the fact that they are not the main target of the disease.

Cerebrospinal fluid (CSF) cannot be considered a peripheral sample, because it is not easily accessible; anyway, CSF represents a more direct window on central nervous system than blood, so that it can be analysed in pilot studies to determine biochemical alterations which could be more extensively assessed in peripheral models.

The development of therapies demonstrating a pre-clinical neuroprotective effect has raised the need of peripheral biomarkers more and more pressing. In fact, new drugs tested to delay the progression of Parkinson's disease are targeted to specific biochemical alterations occurring in the central nervous system; in this context, a great help would derive from the availability of reliable biochemical parameters linked to the targeted pathway, useful to monitor therapies in clinical studies. Biomarkers expected to predict the effect of a therapeutic intervention and assessed as substitutes for a clinically relevant endpoint are named surrogate markers; their use would allow to overcome some difficulties that often occur in clinical trials of neuroprotective therapies: a biomarker could give short-term information on drug biochemical efficacy, whereas clinical outcome can be evaluated only after years of treatment; a biomarker would not be influenced by a symptomatic effect that could complicate the demonstration of the neuroprotective effect of the drug.

• To establish drug mechanisms, assess drug dosing and secondary outcomes in clinical

Neuroimaging biomarkers include detection of dopamine transporter (DAT) levels in basal ganglia with radiolabeled tracers; 123I-FP-CIT SPECT, commonly known as DATscan, is actually used in clinical practice to assist in the differential diagnosis of movement disorders. Anyway, DAT depletion follows the death of nigral neurons; furthermore, the exposure to radiations and the high costs limit the use of this test in screening and follow

Peripheral biochemical markers would offer the advantage of being easily accessible for repeated measurements, feature that makes them particularly useful not only for diagnosis but even for monitoring the course of the disease. Various peripheral cell types share with neurons the molecular machinery for several biochemical processes; this provides the chance of testing peripheral ex-vivo models to seek for disease-specific biochemical alterations. The ideal biomarker should reflect a pathogenetic mechanism operative at central level (core biomarker) and playing a major role in the patophysiology of Parkinson's disease, such as alpha-synuclein aggregation, mitochondrial dysfunction, oxidative stress and impairment of the clearance machinery. Therefore, it is essential to identify peripheral models reproducing the molecular alterations detected in the affected neurons. Up to now, a wide range of putative peripheral models of disease have been proposed: platelets, peripheral blood mononuclear cells and fibroblasts represent ex vivo tissues which are actually under investigation with the aim of disclosing differences in biochemical parameters between patients with Parkinson's disease and controls; even plasma is currently studied with the same purpose. The accessibility of peripheral tissues represent the clearest advantage of their use, significantly compensating the fact that they are not the

Cerebrospinal fluid (CSF) cannot be considered a peripheral sample, because it is not easily accessible; anyway, CSF represents a more direct window on central nervous system than blood, so that it can be analysed in pilot studies to determine biochemical alterations which

The development of therapies demonstrating a pre-clinical neuroprotective effect has raised the need of peripheral biomarkers more and more pressing. In fact, new drugs tested to delay the progression of Parkinson's disease are targeted to specific biochemical alterations occurring in the central nervous system; in this context, a great help would derive from the availability of reliable biochemical parameters linked to the targeted pathway, useful to monitor therapies in clinical studies. Biomarkers expected to predict the effect of a therapeutic intervention and assessed as substitutes for a clinically relevant endpoint are named surrogate markers; their use would allow to overcome some difficulties that often occur in clinical trials of neuroprotective therapies: a biomarker could give short-term information on drug biochemical efficacy, whereas clinical outcome can be evaluated only after years of treatment; a biomarker would not be influenced by a symptomatic effect that could complicate the demonstration of the neuroprotective effect

• To study disease mechanisms.

studies.

main target of the disease.

of the drug.

up.

• To identify individuals with an increased risk of disease.

• To monitor disease progression throughout its course.

could be more extensively assessed in peripheral models.

• To diagnose patients before or at the threshold of motor symptoms.

Actually, among the putative peripheral biochemical markers which have been investigated, no one has a reliability which could justify its use in clinical practice.

Alpha-synuclein levels in peripheral blood cells from patients with Parkinson's disease do not seem to differ from controls (Brighina et al 2010), even if an increased expression of the alpha-synuclein gene has been detected in fibroblasts from patients (Hoepken et al 2008). Data on total plasma level of the protein in patients compared to controls are contrasting but new alpha-synuclein assays able to detect oligomeric protein have demonstrated changes in plasma and cerebrospinal fluid from diseased patients. Recently, the concentration of alphasynuclein oligomers and the ratio of oligomers on total alpha-synuclein have been found significantly increased in cerebrospinal fluid from diseased patients compared to controls (Tokuda et al 2010). The oligomeric soluble fraction of alpha-synuclein has been reported to be increased in plasma samples obtained from patients.

A reduction of glutamate uptake has been observed in platelets from patients and the decrease has showed a positive correlation with clinical severity of the disease (Ferrarese et al. 1999, 2001). This finding indicates that a systemic defect in glutamate transport occurs in patients with Parkinson's disease.

Some studies have reported a decrease in proteasome function in lymphocytes from diseased patients (Blandini et al 2006), but this finding has not been confirmed by subsequent experiments (Brighina et al 2010).

Oxidative stress is induced in lymphocytes from patients with Parkinson's disease and inversely correlates with L-dopa daily doses (Prigione et al 2006, 2009). Increased nitrotyrosine modifications of alpha-synuclein have been found in peripheral blood mononuclear cells obtained from individuals with idiopathic Parkinson's disease compared to controls; moreover, the amount of nitrotyrosine-modified alpha-synuclein has been demonstrated to positively correlate with intracellular levels of reactive oxygen species; in the same ex vivo model a significant increase in LC3II level have been demonstrated in diseased patients (Prigione et al 2010). Therefore, in patients with sporadic Parkinson's disease, a systemic dysregulation of autophagy and oxidative stress seem to exist and correlate with alpha-synuclein post-translational modifications. This confirms the chance of identifying peripheral biomarkers able to mirror the activation of pathogenetic processes in the central nervous system: biochemical parameters of autophagy and oxidative stress, together with qualitative and quantitative alterations of alpha-synuclein, could be useful for diagnostic or prognostic purposes and to monitor the biochemical effects of putative neuroprotective therapies. Moreover, the physiopathologic correlation among these mechanisms could offer the possibility of a combined use of these parameters to increase the accuracy of single biomarkers. In particular, the specificity of autophagy dysregulation in peripheral models, not confirmed in other neurodegenerative diseases, qualifies autophagy machinery as a new and promising target of research in the field of peripheral biochemical markers of disease.

It is conceivable that neuroprotective drugs would have more chance of success if they are administered during early stages of disease, when the molecular pathogenetic processes occurring in dopaminergic neurons could be more easily blocked; therefore, the availability of tools useful to make early diagnosis would increase the potential neuroprotective effects of old and new putative therapies. Therefore, development of disease biomarkers is strongly related to the identification of drugs able to delay the progression of Parkinson's disease or even to prevent its onset if diagnosis is made in preclinical stages.

Alpha-Synuclein, Oxidative Stress and

inhibition of SIRT2.

a mouse model of disease.

oxide in dopaminergic neurons.

Autophagy Failure: Dangerous Liaisons in Dopaminergic Neurodegeneration 129

targets among SIRTs. Promising pharmacologic approaches include stimulation of SIRT1, which imitates the protective effects produced by caloric restriction in animals, and

Stimulation of mitochondrial function might represent another target of putative neuroprotective drugs for Parkinson's disease. Resveratrol is a molecule that has been demonstrated to induce genes involved in mitochondrial biogenesis and aerobic

Several studies have evaluated the putative neuroprotective role of treatments aimed at reducing oxidative stress. An evidence of the potential benefit of antioxidant treatment in Parkinson's disease has been the finding that induction of glutathione peroxidase expression determines a protective effect against 6-hydroxydopa toxicity in dopaminergic neurons of mice. Similarly, overexpression of glutathione peroxidase, as well as Cu, Zn-superoxide

Several foods and natural molecules are supposed to exert a neuroprotective effect thanks to their antioxidant properties. Soy component genistein increases cellular reduced glutathione. Green tea contains polyphenols and flavonoids that prevent lipid peroxidation; another substance contained in green tea, epigallocatechin 3-gallate, has been demonstrated to stimulate nigrostriatal transmission and prevent dopaminergic neuronal loss through suppression of nitric oxide levels in a MPTP mouse model; this finding has qualified nitrosative stress as a putative target of neuroprotective therapies: 7 nitroindazole, a synthetic inhibitor of nitric oxide synthase, has been shown to prevent neurodegeneration in MPTP animal models. Administration of supplements of coenzyme Q10, an endogenous antioxidant that acts as an electron transporter for mitochondrial complexes I and II, has been found to increase the activity of mitochondrial complex I in patients with Parkinson's disease and to prevent dopaminergic neuronal degeneration in

Different plant extracts have antioxidant properties and produce positive effects on experimental models of Parkinson's disease. Withania somnifera belongs to the solanaceae family, extensively used in Indian Ayurvedic medicine. This plant contains various alkaloids which might mediate its putative therapeutic effects. Recent studies have shown that administration of extract from w. somnifera increases the levels of antioxidant molecules in the midbrain and corpus striatum of mouse models of Parkinson's disease; furthermore, this treatment has produced an improvement in dopaminergic transmission

Use of non-steroidal anti-inflammatory medications seems to reduce the risk of Parkinson's disease, probably as a consequence of a decreased generation of free radicals and nitric

Antioxidant therapies have demonstrated to be effective at preventing alpha-synuclein aggregation and neuronal death in many cellular and animal models of PD but clinical studies have failed to demonstrate any significant effect on disease progression in patients. Propalgylamines (selegiline and rasagiline) are drugs actually used in clinical practice for the treatment of Parkinson's disease. These molecules seem to positively modify the natural history of the disease thanks to their antioxidant effect, partially related to the inhibition of MAO-B-mediated dopamine catabolism (Magyar et al 2010). The neuroprotective effect of propalgylamines has been demonstrated in various models of disease but it has not been definitely proved in patients. In fact, MAO-B inhibition is also responsible for an

and motor function in the same models (Rajasankar et al 2009).

metabolism; this drug rescues dopaminergic neurons in MPTP mouse models.

dismutase, protects murine brain against paraquat-induced neurodegeneration.

### **5.2 Therapeutic perspectives**

Several pharmacological approaches have been tested in order to prevent dopaminergic neurodegeneration. It is obvious that the most promising targets of neuroprotective strategy for Parkinson's disease are represented by the molecular pathogenetic mechanisms known to contribute to nigral degeneration.

An ideal therapeutic approach would consist in preventing alpha-synuclein misfolding, accumulation and aggregation in dopaminergic neurons. Means to achieve this aim include the suppression of alpha-synuclein expression through siRNA technology, the development of agents able to rescue the formation of toxic oligomeric intermediates, the induction of the neuronal pathways involved in alpha-synuclein degradation. The first two strategies will be tested in the near future, whereas the stimulation of clearance pathways is actually subjected to intense research.

Different drugs have been successfully used to induce macroautophagy in cell and animal models of disease. Anyway, the use of these interventions to prevent neurodegeneration is limited by the lack of specificity; indeed, most of the tested drugs affect major intracellular pathways, thus explaining a series of undesired side effects. Molecules that activate the transcriptional factor EB, responsible for the induction of lysosomal biogenesis, have proved a neuroprotective effect in cell and animal models of disease (Dehay et al 2010). In particular, rapamycin and trehalose, two largely used macroautophagy inducers, have demonstrated to rescue neurons from the toxicity of various stimuli related to Parkinson's disease. The effectiveness of these two molecules seems to derive from their property to modulate the two most critical steps of macroautophagy: they induce the formation of autophagosomes through an mTOR-dependent (rapamycin) or mTOR-independent (trehalose) mechanism; furthermore, they activate the transcription factor EB, determining an increase of lysosomal biogenesis, so that the increased sequestration of cargoes within the autophagosomes is balanced by an increase of lysosomal function, thus preventing the intracellular accumulation of autophagic vacuoles. This might represent a reasonable model to develop therapeutic tools aimed at activating macroautophagy. Ongoing screenings to identify novel specific macroautophagy enhancers should provide, in coming years, novel agents useful to modulate macroautophagy for therapeutic purposes.

No effective strategies are available to activate the ubiquitin-proteasome system; pharmacologic research has focused on increasing the proteolytic activity of proteasome subunits, but ubiquitin targeting and substrate deubiquitinization are all possible target of future therapies which could be tested.

The importance of chaperone mediated autophagy in the degradation of alpha-synuclein makes it a priority to identify compounds able to stimulate this pathway. In this regard, the pharmacologic increase of lamp2a levels on lysosomal membrane appears as the most reasonable approach. Genetic overexpression of lamp2A has been demonstrated to upregulate chaperone mediated autophagy in animal models but pharmacologic tools able to obtain the same effect are currently not available; cholesterol-depleting agents has shown to enhance the activity of chaperone mediated autophagy through disruption of lysosome membrane lipid microdomains, where inactive lamp2A is stored.

As previously reported, sirtuins have been identified as mediators of modifications induced by starvation in dopaminergic neurons of animals; in particular, SIRT2 seems to enhance the toxic effect of alpha-synuclein, whereas SIRT1 has demonstrated to exert a protective role in dopaminergic neurons. These discoveries have led to the search for potential therapeutic

Several pharmacological approaches have been tested in order to prevent dopaminergic neurodegeneration. It is obvious that the most promising targets of neuroprotective strategy for Parkinson's disease are represented by the molecular pathogenetic mechanisms known

An ideal therapeutic approach would consist in preventing alpha-synuclein misfolding, accumulation and aggregation in dopaminergic neurons. Means to achieve this aim include the suppression of alpha-synuclein expression through siRNA technology, the development of agents able to rescue the formation of toxic oligomeric intermediates, the induction of the neuronal pathways involved in alpha-synuclein degradation. The first two strategies will be tested in the near future, whereas the stimulation of clearance pathways is actually subjected

Different drugs have been successfully used to induce macroautophagy in cell and animal models of disease. Anyway, the use of these interventions to prevent neurodegeneration is limited by the lack of specificity; indeed, most of the tested drugs affect major intracellular pathways, thus explaining a series of undesired side effects. Molecules that activate the transcriptional factor EB, responsible for the induction of lysosomal biogenesis, have proved a neuroprotective effect in cell and animal models of disease (Dehay et al 2010). In particular, rapamycin and trehalose, two largely used macroautophagy inducers, have demonstrated to rescue neurons from the toxicity of various stimuli related to Parkinson's disease. The effectiveness of these two molecules seems to derive from their property to modulate the two most critical steps of macroautophagy: they induce the formation of autophagosomes through an mTOR-dependent (rapamycin) or mTOR-independent (trehalose) mechanism; furthermore, they activate the transcription factor EB, determining an increase of lysosomal biogenesis, so that the increased sequestration of cargoes within the autophagosomes is balanced by an increase of lysosomal function, thus preventing the intracellular accumulation of autophagic vacuoles. This might represent a reasonable model to develop therapeutic tools aimed at activating macroautophagy. Ongoing screenings to identify novel specific macroautophagy enhancers should provide, in coming years, novel

No effective strategies are available to activate the ubiquitin-proteasome system; pharmacologic research has focused on increasing the proteolytic activity of proteasome subunits, but ubiquitin targeting and substrate deubiquitinization are all possible target of

The importance of chaperone mediated autophagy in the degradation of alpha-synuclein makes it a priority to identify compounds able to stimulate this pathway. In this regard, the pharmacologic increase of lamp2a levels on lysosomal membrane appears as the most reasonable approach. Genetic overexpression of lamp2A has been demonstrated to upregulate chaperone mediated autophagy in animal models but pharmacologic tools able to obtain the same effect are currently not available; cholesterol-depleting agents has shown to enhance the activity of chaperone mediated autophagy through disruption of lysosome

As previously reported, sirtuins have been identified as mediators of modifications induced by starvation in dopaminergic neurons of animals; in particular, SIRT2 seems to enhance the toxic effect of alpha-synuclein, whereas SIRT1 has demonstrated to exert a protective role in dopaminergic neurons. These discoveries have led to the search for potential therapeutic

agents useful to modulate macroautophagy for therapeutic purposes.

membrane lipid microdomains, where inactive lamp2A is stored.

future therapies which could be tested.

**5.2 Therapeutic perspectives** 

to intense research.

to contribute to nigral degeneration.

targets among SIRTs. Promising pharmacologic approaches include stimulation of SIRT1, which imitates the protective effects produced by caloric restriction in animals, and inhibition of SIRT2.

Stimulation of mitochondrial function might represent another target of putative neuroprotective drugs for Parkinson's disease. Resveratrol is a molecule that has been demonstrated to induce genes involved in mitochondrial biogenesis and aerobic metabolism; this drug rescues dopaminergic neurons in MPTP mouse models.

Several studies have evaluated the putative neuroprotective role of treatments aimed at reducing oxidative stress. An evidence of the potential benefit of antioxidant treatment in Parkinson's disease has been the finding that induction of glutathione peroxidase expression determines a protective effect against 6-hydroxydopa toxicity in dopaminergic neurons of mice. Similarly, overexpression of glutathione peroxidase, as well as Cu, Zn-superoxide dismutase, protects murine brain against paraquat-induced neurodegeneration.

Several foods and natural molecules are supposed to exert a neuroprotective effect thanks to their antioxidant properties. Soy component genistein increases cellular reduced glutathione. Green tea contains polyphenols and flavonoids that prevent lipid peroxidation; another substance contained in green tea, epigallocatechin 3-gallate, has been demonstrated to stimulate nigrostriatal transmission and prevent dopaminergic neuronal loss through suppression of nitric oxide levels in a MPTP mouse model; this finding has qualified nitrosative stress as a putative target of neuroprotective therapies: 7 nitroindazole, a synthetic inhibitor of nitric oxide synthase, has been shown to prevent neurodegeneration in MPTP animal models. Administration of supplements of coenzyme Q10, an endogenous antioxidant that acts as an electron transporter for mitochondrial complexes I and II, has been found to increase the activity of mitochondrial complex I in patients with Parkinson's disease and to prevent dopaminergic neuronal degeneration in a mouse model of disease.

Different plant extracts have antioxidant properties and produce positive effects on experimental models of Parkinson's disease. Withania somnifera belongs to the solanaceae family, extensively used in Indian Ayurvedic medicine. This plant contains various alkaloids which might mediate its putative therapeutic effects. Recent studies have shown that administration of extract from w. somnifera increases the levels of antioxidant molecules in the midbrain and corpus striatum of mouse models of Parkinson's disease; furthermore, this treatment has produced an improvement in dopaminergic transmission and motor function in the same models (Rajasankar et al 2009).

Use of non-steroidal anti-inflammatory medications seems to reduce the risk of Parkinson's disease, probably as a consequence of a decreased generation of free radicals and nitric oxide in dopaminergic neurons.

Antioxidant therapies have demonstrated to be effective at preventing alpha-synuclein aggregation and neuronal death in many cellular and animal models of PD but clinical studies have failed to demonstrate any significant effect on disease progression in patients. Propalgylamines (selegiline and rasagiline) are drugs actually used in clinical practice for the treatment of Parkinson's disease. These molecules seem to positively modify the natural history of the disease thanks to their antioxidant effect, partially related to the inhibition of MAO-B-mediated dopamine catabolism (Magyar et al 2010). The neuroprotective effect of propalgylamines has been demonstrated in various models of disease but it has not been definitely proved in patients. In fact, MAO-B inhibition is also responsible for an

Alpha-Synuclein, Oxidative Stress and

Autophagy Failure: Dangerous Liaisons in Dopaminergic Neurodegeneration 131

: genetic factors; : environmental factors; : therapeutic strategies.

Fig. 1. Scheme illustrating genetic and enviromental factors involved in alpha-synuclein (α-SYN ) toxicity and possible therapeutic targets. CMA: chaperone mediated autophagy

Therapeutic tools able to remove alpha-synuclein oligomers or to favor the protein refolding are not currently available. Anyway, several data support a critical role for the efficiency of the intracellular clearance machinery in the maintenance of neuronal integrity; in fact, strong links have been established between abnormal functioning of surveillance pathways and the pathogenesis of Parkinson's disease. Ubiquitin-proteasome system represents the first degradative pathway which has been investigated, however its importance has been downsized by the finding of its secondary role in the degradation of alpha-synuclein, which remains the most disease-specific pathogenetic factor. In this context, the scientific attention has switched to autophagy. Alpha-synuclein, UCH-L1, ATP13A2, DJ1 and PINK1 have all been demonstrated to regulate the clearance of proteins and mitochondria through autophagic-lysosomal pathway. Furthermore, the loss of a single allele of the glucocerebrosidase gene seems to be sufficient to increase the risk of developing the disease.

improvement in motor symptoms, based on the increase of dopaminergic nigrostriatal transmission, so that it is difficult to distinguish symptomatic from neuroprotective effect. this confirms the need of a peripheral, pathology-related biomarker, useful as a surrogate endpoint in clinical trials of neuroprotective drugs. Anyway, molecular mechanisms different from MAO-B inhibition are supposed to contribute to the putative neuroprotective effect of selegiline and rasagiline: propargylamines have demonstrated to prevent mitochondrial permeabilization, cytochrome c release, caspase activation and nuclear translocation of glyceraldehyde 3-phosphate dehydrogenase; moreover, rasagiline induces the expression of anti-apoptotic proteins, such as Bcl-2 and glial cell-line derived neurotrophic factor (GDNF).

Drugs interfering with excitotoxic insult, such as the NMDA antagonist amantadine, produce little improvement in motor parkinsonian symptoms and significantly reduce motor side effects of L-dopa treatment. Although NMDA antagonists have been found to prevent nigral degeneration in animal models of Parkinson's disease (Blandini et al 2001), no neuroprotective effect has been demonstrated in patients.

Dopaminergic drugs, such as L-dopa and dopamine agonists, as well as deep brain stimulation, might determine neuroprotective effects based on the hypothesis that an increase of dopaminergic transmission should inhibit the glutammatergic projection from subthalamus to the substantia nigra.
