1. PGC-1 Family: The Masters of Mitochondrial Biogenesis

The family of peroxisome proliferator-activated receptor γ coactivator 1 comprises three members, PGC-1α, PGC-1β, and PGC-related coactivator (PRC). PGC-1s act as ‘molecular switches’ in many metabolic pathways, coordinating transcriptional programs involved in the control of cellular metabolism and overall energy homeostasis as well as antioxidant defence ^[1]. Their versatile actions are achieved by interacting with different transcription factors and nuclear receptors in a tissue-specific manner.

The family of peroxisome proliferator-activated receptor γ coactivator 1 comprises three members, PGC-1α, PGC-1β, and PGC-related coactivator (PRC). PGC-1s act as ‘molecular switches’ in many metabolic pathways, coordinating transcriptional programs involved in the control of cellular metabolism and overall energy homeostasis as well as antioxidant defence [103]. Their versatile actions are achieved by interacting with different transcription factors and nuclear receptors in a tissue-specific manner.

Notably, in analysing the

PGC-1α gene, it has emerged that different variants originating from diverse transcription start sites exist ^[2]. This has led to the identification of a so-called “brain variant” that is more abundant in the human brain than the canonical isoform ^[3]. Moreover, an alternative splicing event, which introduces a premature stop codon, yields a shortened version of the coactivator, named

gene, it has emerged that different variants originating from diverse transcription start sites exist [104]. This has led to the identification of a so-called “brain variant” that is more abundant in the human brain than the canonical isoform [105]. Moreover, an alternative splicing event, which introduces a premature stop codon, yields a shortened version of the coactivator, named

NT-PGC-1α that is highly abundant in the mouse brain ^[4].

that is highly abundant in the mouse brain [106].

1.1. PGC-1s’ Architecture

The modular structure of PGC-1s is highly conserved among all three members of the family. The N-terminus of PGC-1 contains a strong transcriptional activation domain, whereas the C-terminal region holds a serine/arginine rich domain and an RNA binding domain that couples pre-mRNA splicing with transcription ^[5]. At its N-terminus domain, PGC-1 interacts with several histone acetyltransferase (HAT) complexes, including cAMP response element-binding protein (CREB)-binding protein, p300, and steroid receptor coactivator-1 (SRC-1) ^[6]. On the other side, in the C-terminal region, other activation complexes dock PGC-1, including the TRAP/DRIP (thyroid receptor-associated protein/vitamin D receptor-interacting protein), also known as Mediator complex, which facilitates direct interaction with the transcription initiation machinery ^[7], and the SWI/SNF (switch/sucrose non-fermentable) that acts as a chromatin-remodelling complex, through its interaction with BAF60a ^[8]. This peculiarity of PGC-1s to function as a protein docking platform for the recruitment and assembly of various chromatin remodelling and histone-modifying enzymes, which easily allow the access of the transcription machinery to DNA by altering the local chromatin state, contributes to the remarkably powerful PGC-1s coactivation capacity.

Furthermore, an alternative mechanism to increase gene expression relies on the capability of the PGC-1α transcriptional activator complex to displace repressor proteins, such as histone deacetylase and small heterodimer partner (SHP), on its target promoters ^[9].

PGC-1α and PGC-1β share a common similar domain in the internal region, which functions as a repression domain ^[6], and several clusters of conserved amino acids, such as the LXXLL motif that is recognized by nuclear receptors and host cell factor 1 interacting motif ^[10][11]. Although it contains the same activation domain and RNA-binding domain as the other members of the family, PRC shows poor sequence similarity to PGC-1α and PGC-1β ^[12]. Few studies have been conducted until now to elucidate the functions of PRC, and to our knowledge, none of them focus primarily on its activity in the brain. Therefore, this member of the PGC-1 family will be not considered in this review.

1.2. PGC-1s’ Activity: Boosting Mitochondrial Functions

Although PGC-1s display an extremely powerful autonomous transcriptional activity, the mechanism through which PGC-1s activate gene expression is to date poorly understood. The spatial and temporal assemblance of the several activation complexes to PGC-1 is still unknown. The major current hypothesis is that PGC-1 binds to a specific transcription factor in the promoter region, followed by the recruitment of P300 and TRAP/DRIP complexes which opens the chromatin through histone acetylation activity, thus allowing the initiation of the transcription via RNA polymerase II (RNApolII). Moreover, the involvement of additional proteins in RNA elongation and processing as part of the PGC-1 complexes suggests that it might move along the elongating RNA and take part in the mRNA maturation. To terminate gene expression, the acetyltransferase GCN5 acetylates PGC-1 at several lysine residues, inducing re-localization of PGC-1 from the promoter region to subnuclear foci where its transcriptional activity is inhibited ^[13][14]. On the contrary, sirtuin 1 (SIRT1) activates PGC-1 by deacetylating lysine residues, thus inducing the expression of PGC-1 target genes ^[15].

Another level of complexity is introduced by other post-translational modifications of PGC-1s, such as phosphorylation and methylation, as well as by the interaction with co-repressors which alters PGC-1s stability and activity. PGC-1α could be directly phosphorylated by three different kinases. p38 MAP kinase and AMP-activated protein kinase (AMPK) phosphorylate PGC-1α, stabilizing the protein and leading to an increase in gene expression activity ^{[16][17][18][19]}. Differently, protein kinase B (AKT) produces a more unstable PGC-1α protein, with consequently decreased expression of target genes ^[20]. Furthermore, PRMT1 (protein arginine N-methyltransferase 1) methylates PGC-1α on three arginine residues in the C-terminus, hence promoting its activation ^[21].

It is plausible that PGC-1α acts in multiple transcriptional complexes whose composition might depend on the specific target genes as well as on different metabolic signals ^[22]. Indeed, PGC-1s may interact with different transcription factors, activating diverse biological programs in a tissue-specific manner. PGC-1α was originally described through its functional interaction with peroxisome proliferator-activated receptor γ (PPARγ) in brown adipose tissue (BAT), a mitochondria rich tissue, where it regulates adaptive thermogenesis in response to cold ^[23]. Further studies revealed that the PGC-1s carry out a plethora of biological responses finalized to manage situations of energy shortage. One of the best characterized functions of PGC-1s resides in their ability to promote mitochondrial biogenesis by coactivating different transcription factors, such as the oestrogen-related receptor α (ERRα), the Nuclear Respiratory Factors 1 and 2 (NRF1 and NRF2, respectively), and the transcriptional repressor protein yin yang 1 (YY1) ^[24][25][26]. These transcription factors, in turn, regulate the expression of mitochondrial transcription factor A (

TFAM), which plays an essential role in mtDNA replication, maintenance, and transcription ^[27]. Moreover, both NRF1 and NRF2 guarantee mitochondrial homeostasis by modulating the expression of nuclear genes for components of the OXPHOS system, such as ATP synthase, cytochrome c, and cytochrome c oxidase ^[28][29]. In addition to the powerful activity as master regulators of mitochondrial biogenesis, PGC-1s positively control the expression of genes involved in antioxidant response ^[30]. Finally, PGC-1s can also regulate several other metabolic pathways in different tissues, including gluconeogenesis, fatty acid β-oxidation, thermogenesis, and

de novo lipogenesis ^[1]. The gluconeogenic pathway is mainly initiated by PGC-1α, rather than PGC-1β ^[31]. Particularly, PGC-1α activates gluconeogenesis interacting with forkhead box protein O1 (FOXO1), hepatocyte nuclear factor 4α (HNF4α), and glucocorticoid receptor (GR) ^[32][33][34]. The capacity to sustain fatty acid catabolism is particularly evident in the fasted liver, where PGC-1s act as powerful coactivators of peroxisome proliferator-activated receptor α (PPARα) and promote the synthesis of genes involved in fatty acid oxidation, such as medium-chain acyl-CoA dehydrogenase (

MCAD

CPT1A) ^[35][36]. In BAT, PGC-1s are able to drive the production of heat by inducing the expression of uncoupling protein 1 (

UCP1) ^[37][38]. Finally, PGC-1β alone is able to regulate

de novo

FASN

SCD1 ^[39][40][41]. It is important to note that in many tissues the functions of PGC-1α and PGC-1β may overlap. However, in some organs like the liver, these two coactivators exert opposite functions, with PGC-1α mainly regulating fatty acid β-oxidation during fasting and PGC-1β activating

de novo synthesis of fatty acids after the intake of a meal enriched in lipids ^[42]. Although still not well investigated, the occurrence of the same phenomenon in other organs cannot be excluded.

2. PGC-1s in Parkinson’s Disease

The association between PD and alteration of mitochondrial homeostasis has been extensively reported. However, only recently has it been highlighted that disruptions of mitochondrial biogenesis and dynamics, rather than mitophagy, are closely associated with the disease onset. This has resulted in a detailed evaluation of the role of PGC-1s on the onset and progression of PD.

The first evidence of the involvement of PGC-1s in neurodegenerative diseases came from two independent studies on whole body PGC-1α knock out (PGC-1αKO) mice ^[35][43]. Morphologically, the absence of PGC-1α in the brain results in a well-preserved cerebral cortex with the presence of vacuoles in neurons containing aggregates of membranous material ^[35]. Nevertheless, these mice display behavioural abnormalities peculiar to neurological disorders, indicative of lesions in the striatum ^[43]. A direct demonstration of the link between PGC-1α and PD was the higher vulnerability of PGC-1αKO mice to the neurodegenerative effects of MPTP and kainic acid, due to the lack of the PGC1α-dependent induction of the antioxidant response ^[44]. Furthermore, cultured PGC1α-

The first evidence of the involvement of PGC-1s in neurodegenerative diseases came from two independent studies on whole body PGC-1α knock out (PGC-1αKO) mice [137,145]. Morphologically, the absence of PGC-1α in the brain results in a well-preserved cerebral cortex with the presence of vacuoles in neurons containing aggregates of membranous material [137]. Nevertheless, these mice display behavioural abnormalities peculiar to neurological disorders, indicative of lesions in the striatum [145]. A direct demonstration of the link between PGC-1α and PD was the higher vulnerability of PGC-1αKO mice to the neurodegenerative effects of MPTP and kainic acid, due to the lack of the PGC1α-dependent induction of the antioxidant response [146]. Furthermore, cultured PGC1α-

null nigral neurons were more sensitive to the accumulation of the overexpressed human α-synuclein ^[45] and conditional PGC1α-KO in the

nigral neurons were more sensitive to the accumulation of the overexpressed human α-synuclein [147] and conditional PGC1α-KO in the

substantia nigra of adult mice caused DA neuron loss associated with a marked decline of mitochondrial biogenesis protein markers ^[46]. Indeed, while the expression of this coactivator protects neuronal cells from ROS-induced cell death via induction of several detoxifying enzymes (superoxide dismutase 2, SOD2; glutathione peroxidase 1, GPX1), its ablation results in the accumulation of nitrotyrosylated proteins and loss of DA neurons ^[44]. Furthermore, transgenic mice in which the expression of the PGC-1s target gene TFAM has been disrupted display lower mtDNA expression and respiratory chain deficiency that result in the adult onset of Parkinsonism ^[47]. All these insights provide the impetus to deepen the knowledge of PGC-1s in PD.

of adult mice caused DA neuron loss associated with a marked decline of mitochondrial biogenesis protein markers [148]. Indeed, while the expression of this coactivator protects neuronal cells from ROS-induced cell death via induction of several detoxifying enzymes (superoxide dismutase 2, SOD2; glutathione peroxidase 1, GPX1), its ablation results in the accumulation of nitrotyrosylated proteins and loss of DA neurons [146]. Furthermore, transgenic mice in which the expression of the PGC-1s target gene TFAM has been disrupted display lower mtDNA expression and respiratory chain deficiency that result in the adult onset of Parkinsonism [149]. All these insights provide the impetus to deepen the knowledge of PGC-1s in PD.

Despite PGC-1α and PGC-1β are both powerful coactivators of mitochondrial biogenesis and antioxidant response, PGC-1α has been far more investigated as compared to PGC-1β, especially in the context of PD. Although PGC-1α and PGC-1β control mitochondrial capacity in an additive and independent manner in different subtypes of neurons, the overexpression of PGC-1α significantly reduces the PGC-1β level ^[45][48]. Furthermore, while PGC-1α can compensate PGC-1β loss and restore the induction of antioxidant genes, PGC-1β fails to cope with the absence of PGC-1α, being only slightly induced in PGC-1α

Despite PGC-1α and PGC-1β are both powerful coactivators of mitochondrial biogenesis and antioxidant response, PGC-1α has been far more investigated as compared to PGC-1β, especially in the context of PD. Although PGC-1α and PGC-1β control mitochondrial capacity in an additive and independent manner in different subtypes of neurons, the overexpression of PGC-1α significantly reduces the PGC-1β level [147,150]. Furthermore, while PGC-1α can compensate PGC-1β loss and restore the induction of antioxidant genes, PGC-1β fails to cope with the absence of PGC-1α, being only slightly induced in PGC-1α

null mice ^[44][45]. However, since in other tissues PGC-1β is able to drive genes involved in

mice [146,147]. However, since in other tissues PGC-1β is able to drive genes involved in

de novo

lipogenesis, including

SCD1 ^[39][40], and given that

[141,142], and given that

SCD1

expression has proved to be deleterious for PD, it would be intriguing to evaluate if PGC-1β retains this ability also in DA neurons and whether the activation of this pathway may have detrimental effects.

2.1. Deepening the Role of PGC-1α in Parkinson’s Disease

Numerous studies have been carried out to fully elucidate the role played by PGC-1α in PD. A comparative analysis of a large cohort of PD patients and age-matched controls has revealed that two PGC-1α variants are associated with the risk of PD onset (rs6821591 CC and rs2970848 GG) ^[49]. Further studies have led to the identification of different PGC-1α isoforms in the brain, including a truncated 17 kDa protein that lacks the LXXLL site of interaction with several transcription factors ^[46][50]. This 17 kDa isoform has been found upregulated in the

substantia nigra of PD patients, where it inhibits the coactivation of several transcription factors by the full-length PGC-1α ^[50]. The selective knockout of different

PGC-1α isoforms in mice may lead to a decrease of dopamine content in the striatum and to an associated loss of DA neurons ^[46]. Accordingly, in human PD tissues, the levels of PGC-1α and of mitochondrial markers are reduced compared to control patients and are negatively correlated with the severity of the disease ^{[46][51][52][53][54][55]}. Conversely, primary fibroblasts from PD patients display upregulation of PGC-1α, even if its target genes involved in mitochondrial biogenesis and fatty acid β-oxidation processes are unchanged or downregulated and mitochondria display significant morphometric changes ^[56][57]. This may suggest that a post-translational mechanism may also occur, thus jeopardizing the elevated quantity of the coactivator and interfering with its activity.

The low expression of PGC-1α observed in the PD brain is probably due to the high level of gene methylation that has been found in PD patients ^[53]. Dense DNA methylation is usually associated with gene repression, and the

PGC-1α promoter is proximal to a non-canonical cytosine methylation site that is epigenetically modified in the brain of sporadic PD patients ^[51]. Notably, free fatty acids can induce the hypermethylation of the

PGC-1α

in vitro

in vivo causes promoter hypermethylation, thus lowering the level of PGC-1α and mitochondria-associated genes as well as the concomitant induction of pro-inflammatory genes ^[51]. Curiously, both a population-based case-control study and a prospective study indicate that a higher caloric intake, due to elevated consumption of animal-derived saturated fatty acids, tends to be associated with a greater risk of PD ^[58][59]. Moreover, rats subjected to high-fat diet feeding and to infusion of 6-hydroxydopamine displayed DA neurons depletion in the striatum, despite the absence of differences in locomotor activity ^[60]. Altogether this suggests that a high dietary introit of fatty acids, especially saturated ones, may repress the expression of

PGC-1α

Drosophila melanogaster

PGC-1α

D. melanogaster

Spargel and the lack of gene redundancy make this organism an ideal model system to determine the role of PGC-1α in PD. The inhibition of Spargel via RNAi in DA neurons causes an altered climbing activity, with an unexpected increase of mean lifespan probably ascribable to the mitochondrial unfolded protein response and/or a ROS-dependent mitohormesis ^[61]. Further studies have demonstrated that silencing Spargel in flies increases the PD related phenotypes, including climbing defects, decreased mitochondrial mass, and lower dopamine levels ^[62]. However, the genetic or pharmacological activation of Spargel is sufficient to rescue the disease phenotype ^[62].

null mice display abnormal mitochondria in neurons and are more prone to oxidative stress that may be eventually related to neurodegeneration. Nonetheless, re-expression of PGC-1α in these mice restores mitochondrial functions and oxidative stress detoxification ^[45].

in vitro

in vivo models results in an overall protection against neurodegeneration. Treatment of the SH-SY5Y neuroblastoma human cell line with the neurotoxin N-methyl-4-phenylpyridinium leads to serious mitochondrial damage that can be functionally reversed by the overexpression of PGC-1α ^[63]. Moreover, resveratrol treatment in Parkin-mutated fibroblasts promotes PGC-1α activity via SIRT-1, thus resulting in increased mitochondrial biogenesis together with lower ROS accumulation that together ameliorate the phenotypic impact of the mitochondrial dysfunctions caused by the Parkin mutation ^[64]. Accordingly, transgenic mice overexpressing PGC-1α in MPTP-treated DA neurons display induction of the antioxidant genes

Sod2

Trx2) as well as increased OXPHOS, which collectively promote neuronal viability and prevent striatal dopamine loss ^[65]. Conversely, the adenovirus-mediated overexpression of PGC-1α in the

substantia nigra of mice increases their susceptibility to MPTP, as indicated by the loss of DA neurons ^[66]. This deleterious effect may be ascribable to the high level of PGC-1α activity caused by the viral vector microinjection and may shed light on the importance of a balanced regulation of this coactivator to achieve beneficial effects.

2.2. PGC-1α and Parkinson’s Mutated Genes: Defining The Network Implicated in Parkinson’s Disease

Besides the roles of PGC-1α in different PD scenarios based on its capacity to boost mitochondrial biogenesis and the antioxidant response discussed above, it is now clear that PGC-1α protects against neurodegeneration as a player of a more intricated mechanism whose failure may drive the onset and progression of PD (

Figure 2

).

Figure 2.

The role of peroxisome proliferator-activated receptor gamma coactivator 1 (PGC-1α) in the onset of Parkinson’s disease. (

A

) In healthy subjects, damaged mitochondria are promptly replaced with new functional organelles. Mitochondrial depolarization induces PTEN-induced putative kinase 1 (PINK1) to phosphorylate serine 65 residue of ubiquitin (pSer65Ub) and of Parkin, which in turn interact together to increase the PINK1 phosphorylation rate of Parkin. The activation of the PINK1/Parkin pathway promotes the degradation of dysfunctional mitochondria via mitophagy and concomitantly induces the expression of PGC-1α by the ubiquitination of PARIS and α-synuclein. By coactivating nuclear receptors (NRs) or transcription factors (TFs), PGC-1α promotes the expression of genes involved in mitochondrial biogenesis (mitochondrial transcription factor A,

TFAM

) and antioxidant response (superoxide dismutase 2,

SOD2

; glutathione peroxidase 1,

GPX1

). (

B

) In Parkinson’s disease individuals, PINK1 and Parkin usually display loss-of-function mutations. Therefore, the PINK1/Parkin axis cannot prevent the accumulation of dysfunctional mitochondria due to its inability to sustain mitophagy. At the same time, altered Parkin fails to promote the degradation of both PARIS and α-synuclein, which start to accumulate in the nucleus inhibiting

PGC-1α transcription. The low levels of PGC-1α observed in PD patients are not sufficient to induce the expression of genes involved in the renewal of mitochondria and in the antioxidant response. Thereby, reactive oxygen species (ROS) start to accumulate, finally leading to the damage and the death of dopaminergic neurons. Dashed lines and soft colours represent inhibited actions/pathways.

transcription. The low levels of PGC-1α observed in PD patients are not sufficient to induce the expression of genes involved in the renewal of mitochondria and in the antioxidant response. Thereby, reactive oxygen species (ROS) start to accumulate, finally leading to the damage and the death of dopaminergic neurons. Dashed lines and soft colours represent inhibited actions/pathways.

The expression of

PGC-1α is finely regulated by PARIS, a KRAB and zinc finger protein that accumulates in the human PD brain ^[67]. PARIS acts as a physiological transcriptional repressor of

is finely regulated by PARIS, a KRAB and zinc finger protein that accumulates in the human PD brain [169]. PARIS acts as a physiological transcriptional repressor of

PGC-1α, downregulating the coactivator and its target genes, ^[67]. Generally, the amount of PARIS is tightly controlled by the PINK1/Parkin axis, which mediates PARIS degradation via ubiquitination ^[68]. However, modifications to either PINK1 or Parkin that alter this regulatory pathway allow PARIS to accumulate inside the neurons ^[69][70][71]. The overexpression of PARIS negatively affects mitochondrial biogenesis causing progressive DA neuron degeneration and loss ^[70]. In flies, the ubiquitous expression of PARIS results in shortened lifespan and climbing defects that are promptly reversed by PINK1, Parkin, or PGC-1α overexpression ^[70]. Noteworthy, the loss-of-function of Parkin in mice and in human-derived cells leads to mitochondrial respiratory function decline coupled with a decrease of mitochondrial mass and of the antioxidant response, a phenotype that closely reminds those of PARIS-overexpressing cells ^[69][71]. Accordingly, the reduction of PARIS level in Parkin knockout cells and mice is sufficient to restore mitochondrial biogenesis and cellular respiration ^[69][71]. By contrast, the effect of Parkin on mitochondrial density is abolished in PGC-1α

, downregulating the coactivator and its target genes, [169]. Generally, the amount of PARIS is tightly controlled by the PINK1/Parkin axis, which mediates PARIS degradation via ubiquitination [170]. However, modifications to either PINK1 or Parkin that alter this regulatory pathway allow PARIS to accumulate inside the neurons [171,172,173]. The overexpression of PARIS negatively affects mitochondrial biogenesis causing progressive DA neuron degeneration and loss [172]. In flies, the ubiquitous expression of PARIS results in shortened lifespan and climbing defects that are promptly reversed by PINK1, Parkin, or PGC-1α overexpression [172]. Noteworthy, the loss-of-function of Parkin in mice and in human-derived cells leads to mitochondrial respiratory function decline coupled with a decrease of mitochondrial mass and of the antioxidant response, a phenotype that closely reminds those of PARIS-overexpressing cells [171,173]. Accordingly, the reduction of PARIS level in Parkin knockout cells and mice is sufficient to restore mitochondrial biogenesis and cellular respiration [171,173]. By contrast, the effect of Parkin on mitochondrial density is abolished in PGC-1α

null

cortical neurons

in vitro and the synergic action on mitochondrial functions given by the co-expression of both Parkin and PGC-1α provides significant neuroprotection ^[72]. This indicates that Parkin is fundamental to ensure the proper action of PGC-1α to stimulate mitochondrial biogenesis, by shutting down PARIS.

and the synergic action on mitochondrial functions given by the co-expression of both Parkin and PGC-1α provides significant neuroprotection [174]. This indicates that Parkin is fundamental to ensure the proper action of PGC-1α to stimulate mitochondrial biogenesis, by shutting down PARIS.

PINK1/Parkin axis plays a double role in controlling both the genesis and the destruction of mitochondria. Therefore, it is easy to wonder how the PGC-1α action is finely tuned to keep mitochondrial homeostasis and if the activation of mitophagy in response to damage can start a signal that activates PGC-1α in order to restore the mitochondrial pool. Although this is still an open question, it is clear that the rescue of mitophagy in Parkin knockout mice does not increase

PGC-1α expression and activity ^[69]. However, PGC-1α regulates the abundance of mitofusin2 (Mfn2), a protein with a central role in mitochondrial fusion, whose loss in nigrostriatal DA neurons in mice leads to a neurodegenerative phenotype ^[72][73]. Once more, this observation provides clues for the existence of a precise regulatory loop that underlies mitochondrial homeostasis through PGC-1α.

expression and activity [171]. However, PGC-1α regulates the abundance of mitofusin2 (Mfn2), a protein with a central role in mitochondrial fusion, whose loss in nigrostriatal DA neurons in mice leads to a neurodegenerative phenotype [174,175]. Once more, this observation provides clues for the existence of a precise regulatory loop that underlies mitochondrial homeostasis through PGC-1α.

Although no direct interaction has been observed between PGC-1α and DJ-1, since DJ-1 can compensate for PINK1 loss, it will be intriguing to evaluate if PGC-1α functions can be rescued by DJ-1 expression in PD cells in terms of mitochondrial biogenesis and antioxidant response ^[74][75].

Although no direct interaction has been observed between PGC-1α and DJ-1, since DJ-1 can compensate for PINK1 loss, it will be intriguing to evaluate if PGC-1α functions can be rescued by DJ-1 expression in PD cells in terms of mitochondrial biogenesis and antioxidant response [90,176].

In addition to PINK1 and Parkin, several studies have demonstrated a strong interference with other genes frequently mutated in PD. One of them is α-synuclein, whose overexpression and oligomerization negatively correlates with PGC-1α level in the human PD brain as well as in murine and cell culture models of α-synuclein oligomerization ^[54]. Particularly, under oxidative stress, α-synuclein may localize in the nucleus where it specifically binds to the

In addition to PINK1 and Parkin, several studies have demonstrated a strong interference with other genes frequently mutated in PD. One of them is α-synuclein, whose overexpression and oligomerization negatively correlates with PGC-1α level in the human PD brain as well as in murine and cell culture models of α-synuclein oligomerization [156]. Particularly, under oxidative stress, α-synuclein may localize in the nucleus where it specifically binds to the

PGC-1α promoter, decreasing its activity. By reducing the expression of the coactivator and related target genes, α-synuclein impairs the mitochondrial functions ^[76]. Intriguingly, the ablation of PGC-1α renders both mice and human neurons more prone to α-synuclein accumulation and toxicity ^[45][77]. By contrast, the pharmacological activation or genetic overexpression of PGC-1α reduces α-synuclein oligomerization and attenuates neurotoxicity

promoter, decreasing its activity. By reducing the expression of the coactivator and related target genes, α-synuclein impairs the mitochondrial functions [177]. Intriguingly, the ablation of PGC-1α renders both mice and human neurons more prone to α-synuclein accumulation and toxicity [147,178]. By contrast, the pharmacological activation or genetic overexpression of PGC-1α reduces α-synuclein oligomerization and attenuates neurotoxicity

in vitro ^[54][76]. The reciprocal influence of PGC-1α and α-synuclein generates a vicious cycle that may play an important role in the disease progression. Since α-synuclein can be ubiquitinated by Parkin ^[78][79], it would also be interesting to understand how the different actors involved in PD pathogenesis may act in concert to protect from neurodegeneration and neuronal loss.

[156,177]. The reciprocal influence of PGC-1α and α-synuclein generates a vicious cycle that may play an important role in the disease progression. Since α-synuclein can be ubiquitinated by Parkin [179,180], it would also be interesting to understand how the different actors involved in PD pathogenesis may act in concert to protect from neurodegeneration and neuronal loss.