Emerging role of S100B protein implication in Parkinson’s disease pathogenesis

Abstract

The exact etiology of Parkinson’s disease (PD) remains obscure, lacking effective diagnostic and prognostic biomarkers. In search of novel molecular factors that may contribute to PD pathogenesis, emerging evidence highlights the multifunctional role of the calcium-binding protein S100B that is widely expressed in the brain and predominantly in astrocytes. Preclinical evidence points towards the possible time-specific contributing role of S100B in the pathogenesis of neurodegenerative disorders including PD, mainly by regulating neuroinflammation and dopamine metabolism. Although existing clinical evidence presents some contradictions, estimation of S100B in the serum and cerebrospinal fluid seems to hold a great promise as a potential PD biomarker, particularly regarding the severity of motor and non-motor PD symptoms. Furthermore, given the recent development of S100B inhibitors that are able to cross the blood brain barrier, novel opportunities are arising in the research field of PD therapeutics. In this review, we provide an update on recent advances in the implication of S100B protein in the pathogenesis of PD and discuss relevant studies investigating the biomarker potential of S100B in PD, aiming to shed more light on clinical targeting approaches related to this incurable disorder.

Introduction

The exact etiology of Parkinson’s disease (PD), the most common neurodegenerative movement disorder, remains obscure, with limited diagnostic biomarkers and only symptomatic therapy till date [1]. PD is a clinically heterogeneous disorder, and prognostic biomarkers reflecting disease severity are lacking.

The pathological hallmarks of PD involve the progressive death of dopaminergic neurons in the substantia nigra pars compacta (SNpc), as well as the accumulation of Lewy bodies and Lewy neurites containing α-synuclein in the degenerative neurons [2, 3]. Emerging evidence highlights the multifunctional role of astrocytes in PD pathogenesis [4]. Following various brain insults, such as intoxication with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) or 6-hydroxydopamine (6-OHDA) (the main toxins used for mimicking PD in vivo), astrocytes become functionally activated [5]. Reactive astrocytes can phagocytose dead neurons and endocytose α-synuclein in synapses [6]. Τhey can also release neurotrophic factors, such as glial-derived neurotrophic factor (GDNF) which significantly contributes to the survival of dopaminergic neurons [7], but also potentially neurotoxic factors, such as glutamate, cytokines and S100B, thereby mediating neuroinflammatory responses [8–11]. Astrocytes play also a fundamental role in blood–brain barrier (BBB) formation that is often disrupted in PD [12].

S100B is a calcium-binding protein, widely expressed in the brain predominantly by astrocytes, but also by maturing oligodendrocytes, neural progenitor cells, dendritic cells and specific lymphocyte subpopulations [13, 14]. S100B is an acidic 21 kDa protein, and probably the most well studied member of S100 proteins [15]. It is a homodimer, and each subunit contains two EF-hand calcium-binding domains [15]. Intracellularly, S100B can interact with more than 20 different proteins in a Ca²⁺ sensitive manner, and is implicated in calcium homeostasis, energy metabolism, cell proliferation, mobility and cytoskeletal regulation [13]. S100B can be released extracellularly and is considered as a marker of brain damage, acting as a damage-associated molecular pattern (DAMP) protein through its interaction with the receptor for advanced glycation end products (RAGE) [13, 16]. RAGE is a multi-ligand receptor of the immunoglobulin superfamily mainly expressed by neurons and microglia that mediates inflammatory responses by activating multiple signaling pathways [17, 18]. Apart from S100B, AGEs, high mobility group box 1 (HMGB1), amyloid fibrils and other S100 proteins can interact with RAGE [19]. High extracellular Ca²⁺ levels have been associated with increased formation of S100B multimers, which induce stabilization of RAGE oligomers or RAGE dimerization, an important process in RAGE-mediated signal transduction [14]. S100B/RAGE interaction can activate small GTPases, including Ras, Rac1, and Cdc42, leading to the activation of the transcription factors NF-κB and activating protein-1 (AP-1), resulting in the increased expression of cyclooxygenase-2 (COX-2), interleukin (IL)-1β, and tumor necrosis factor (TNF)-α [20, 21]. Additionally, RAGE activation via S100B has been shown to induce microglia activation and migration via the activation of Ras/Rac1-Cdc42/NF-κB, Ras/MEK/extracellular signal-regulated kinase 1/2 (ERK1/2)/NF-κB, Ras/Rac1-Cdc42/c-Jun NH2 terminal protein kinase (JNK)/AP-1, and Src/Ras/PI3K/RhoA/ROCK pathway. ROCK phosphorylates LIM kinase (LIMK) that is involved in F-actin stabilization and cell motility [20]. At pro-inflammatory concentrations, S100B can stimulate the RAGE-dependent activation and release of CCL3, CCL5 and CXCL12 chemokines by microglia (Fig. 1) [20]. At nanomolar concentrations, S100B can upregulate the anti-apoptotic factor Bcl-2, whereas at micromolar concentrations, S100B triggers apoptosis, in a RAGE-dependent manner in neurons [22]. On the other hand, trophic and detrimental effects of low and high S100B levels, respectively, may be independent of RAGE in other cell types, such as myoblasts [23–25].

Fig. 1.

S100B/RAGE signaling. S100B/RAGE interaction activates small GTPases, such as Ras and Rac1, thereby activating the transcription factors NF-κB and activating protein-1 (AP-1), leading to the increased expression of cyclooxygenase-2 (COX-2), IL-1β, and TNF-α. S100B/RAGE interaction induces microglia activation and migration via the activation of Ras/Rac1-Cdc42/NF-κB, Ras/MEK/extracellular signal-regulated kinase 1/2 (ERK1/2)/NF-κB, Ras/Rac1-Cdc42/c-Jun NH2 terminal protein kinase (JNK)/AP-1, and Src/Ras/PI3K/RhoA/ROCK pathway. ROCK phosphorylates LIMK implicated in F-actin stabilization and cell motility. At pro-inflammatory concentrations, S100B has been also reported to stimulate the RAGE-dependent activation and release of CCL3, CCL5 and CXCL12 chemokines [14, 20]

At nanomolar concentrations, S100B has been shown to exert gliotrophic and neurotrophic roles by enhancing glial cell proliferation, neuronal survival, neurite outgrowth, synaptogenesis and neurogenesis, as well as by protecting neurons against toxic stimuli, delayed cellular damage after trauma and glutamate excitotoxicity [14, 26, 27]; however, at micromolar concentrations, it has been demonstrated to promote intracellular calcium overload, apoptosis, oxidative damage and excessive neuroinflammation by triggering the production of reactive oxygen species (ROS) and release of pro-inflammatory cytokines [14, 26, 27]. The specific role of S100B in the central nervous system (CNS) is still largely unknown. S100B expression has been detected elevated in the aging brain [14], and S100B dysregulation has been increasingly associated with various neurodegenerative diseases including Alzheimer’s disease (AD) and Parkinson’s disease (PD) [16]. For instance, at low doses, S100B protects neuroblastoma cells against Aβ-amyloid-induced neurotoxicity via RAGE, whereas at high doses, it enhances Aβ-induced neurotoxicity [27].

Although the role of S100B in neurological disorders has been already discussed [28], there is no recent review focusing on the role of S100B in PD. Given the controversial results of relevant studies and the increasing interest in the role of S100B in neurodegeneration, we critically discuss the accumulating preclinical and clinical evidence on the role of S100B in PD pathogenesis, its biomarker potential, as well as the therapeutic opportunities arising from the recent advances in the field.

The role of S100B in PD pathogenesis

Post-mortem human studies have revealed that S100B levels are increased in the substantia nigra of PD patients [8], as well as in the dorsomedial prefrontal cortex of patients with PD and multiple system atrophy (MSA, another synucleinopathy) at a transcriptional level, in comparison to controls [29], implying that S100B could represent a novel mediator of neurodegeneration in PD and paving the way for further exploration of its role in PD pathogenesis.

In vitro evidence has shown that MPTP enhanced the expression and release of S100B in C6 rat glioma cells co-cultured with neuronal PC12 cells, in a concentration- and time-dependent manner [30]. Elevation of S100B was associated with increased glial cell proliferation and reduced neuronal survival, while the pre-treatment of PC12 cells with S100B-antibodies prevented their death [30]. Interestingly, the administration of arachidonyl-2-chloroethylamide, a selective cannabinoid CB1 receptor agonist, inhibited the MPTP-induced glial cell proliferation, increased S100B expression and intracellular Ca²⁺ levels, as well as protected PC12 cells against apoptosis [30]. S100B and CB1 seem to exert opposing cellular effects. S100B results in Gs activation, as well as increased intracellular cyclic adenosine monophosphate (cAMP) and Ca²⁺ levels in neurons and glial cells [31], whereas CB1 receptor activation leads to Gi protein activation, reduced intracellular cAMP and decreased entry of Ca²⁺ into the cells [32]. CB1 receptor is increasingly gaining attention in PD research via yet unclear molecular mechanisms [33]. Another study has indicated that S100B is able to interact with the dopamine receptor D₂ [15]. In addition, the co-expression of S100B and the D2 receptor increased D2 receptor-mediated activation of extracellular signal-regulated kinases (ERKs) and inhibition of adenylate cyclase in HEK293 cells [15]. S100B expression was detected inside the cultured neostriatal neurons expressing the D2 receptor, and their interaction (S100B/D2) was mediated via the third intracellular receptor loop (D2-IC3) [15], implying that intracellular S100B was responsible for these effects. However, in vivo evidence revealed that S100B is absent in dopaminergic neurons [8]. These discrepancies could be attributed to the fact that the results from the firstly mentioned study may pertain to cell lines, since many cell types in culture express S100B. Interestingly, however, forced expression of S100B in PC12 neuronal cells (a surrogate of noradrenergic neurons) has been demonstrated to result in enhanced proliferation and reduced responsiveness to nerve growth factor (NGF) via intracellular activation of Akt [34]. ERK1/2 signaling is highly implicated in various cellular processes underlying PD pathogenesis, including mitochondrial dysfunction, oxidative stress, neuronal apoptosis and inflammation [35]. Therefore, S100B seems to be a key player in endocannabinoid system and D2-mediated intracellular pathways in PD, although further research is needed to elucidate the exact molecular mechanisms involved.

In vivo evidence has shown that S100B is primarily localized in astrocytes and to a lesser extent in microglia, while it is absent in dopaminergic neurons [8]. Increased S100B secretion levels have been observed in the striatum of systematically MPTP and probenecid-treated mice [36], in the substantia nigra (SN) of MPTP-treated mice at both mRNA and protein levels [8], as well as in the striatum and SN of 6-OHDA-treated rats [37]. However, decreased striatal S100B levels have been also reported in intracerebroventricularly MPP⁺-treated mice [38], without evidence of significant alterations in the intracellular S100B levels in the striatum and substantia nigra of 6-OHDA-treated rats [39]. However, it is not completely clear whether the described effects of overexpressed S100B are due to the intracellular or extracellular S100B in the last two mentioned studies. Although S100B levels were initially increased in the striatum and SN upon MPTP treatment, their concentration tends to decrease the following days after the insult [8, 40, 41]. In accordance, an initial transitory increase of CSF S100B levels has been observed in 6-OHDA-treated rats, with 6-OHDA-treated astrocytes shown to exhibit increased S100B secretion in vitro [39]. Also, a significant reduction of S100B was demonstrated to follow the initial increase in the SN of MPTP-treated mice [42]. After 6-OHDA injection, striatal dopaminergic denervation has been accompanied by astrogliosis and S100B upregulation [10]. Inhibitory RAGE isoforms have been also observed in S100B-overexpressing astrocytes at an early stage after MPTP intoxication in mice [19], while striatal S100B levels were found elevated in pre-symptomatic MPTP-treated mice [43]. It has been proposed that the mechanical insult in the basal ganglia accompanying the intrastriatal intoxication of 6-OHDA could possibly explain the initial increase of S100B in this case [39]; however, the intraperitoneal MPTP injection has been associated with a transient S100B elevation in some of the abovementioned studies, thus weakening this hypothesis. The observed S100B reduction following the initial S100B increase could be potentially attributed to a compensatory process involving a self-inhibitory feedback loop [42], to protect the surviving neurons. Indeed, S100B gene expression can be suppressed by various negative regulatory elements in all cell types [14], suggesting that S100B expression is triggered by specific factors that counteract the function of the negative regulatory elements. When the neurotoxic effects of inducing factors such as MPTP or 6-OHDA are no longer active, S100B may, therefore, return to its baseline levels. S100B might also induce astrocytic apoptosis in an autocrine manner, thus leading to a decreased number of astrocytes as neuroinflammation progresses, lowering S100B levels [14, 44].

S100B has been demonstrated to significantly contribute to neurodegeneration in S100B-overexpressed or S100B-ablated animal models of PD. In particular, MPTP-induced dopaminergic degeneration has been shown to be partially inhibited in S100B-ablated mice [8]. S100B-mediated MPTP neurotoxicity involved RAGE upregulation, TNF-α release and enhanced microgliosis [8]. In accordance, S100B overexpression in transgenic mice has been shown to induce motor deficits similar to PD phenotype [45, 46] by suppressing the expression of G protein-coupled receptor kinase 2 (GRK2) and dopamine D2 receptor, thus affecting dopamine metabolism and promoting oxidative stress [45, 46]. S100B transgenic mice have been indicated to display impaired metabolism of several PD-related neurotransmitters, including Glu, GABA, Phe, His and Trp [47]. Hence, these findings suggest that constantly increased S100B levels may exert detrimental effects on neurons, whereas baseline S100B levels may be essential for normal neuronal function.

The loss of beneficial physiological functions such as release of neurotrophic and anti-inflammatory factors, and the gain of detrimental pro-inflammatory properties accompanying aberrant microglia activation contribute to PD pathogenesis [48]. RAGE activation via S100B can induce microglia activation and migration via the activation of Ras/Rac1-Cdc42/NF-κB, Ras/MEK/extracellular signal-regulated kinase 1/2 (ERK1/2)/NF-κB, Ras/Rac1-Cdc42/c-Jun NH₂ terminal protein kinase (JNK)/AP-1, and Src/Ras/PI3K/RhoA/ROCK pathways [20, 21, 49, 50]. S100B has been shown to promote M1 and inhibit M2 polarization of microglia, respectively, via NF-κB, thus promoting cerebral ischemia [51]. However, the exact effects of extracellular S100B in microglia activation in PD still remains to be elucidated.

Collectively, the immediate astrocytic S100B upregulation after acute damage could reflect the initial reactive astrocytosis, which may trigger the activation of downstream factors, leading to the amplification of neuroinflammation, oxidative damage and dysregulation of neurotransmitter metabolism underlying PD pathogenesis (Fig. 2). Therefore, the role of S100B in PD pathogenesis seems to crucially depend not only on its concentration and specific animal model used, but also on the exact time of the disease course, highlighting the need for further studies to clarify the time-specific effects of S100B on PD.

Fig. 2.

Role of S100B in Parkinson’s disease (PD). S100B is mainly expressed by the astrocytes in the CNS. S100B-mediated MPTP neurotoxicity has been demonstrated to involve RAGE upregulation and secretion of pro-inflammatory cytokines, thereby enhancing neuroinflammation and oxidative damage. S100B has been shown to reduce the expression of dopamine D2 receptor, thereby affecting dopamine metabolism. S100B transgenic mice have been indicated to display impaired metabolism of several PD-related neurotransmitters, including Glu, GABA, Phe, His and Trp. Increased co-localization of S100B and A30P α-synuclein in astrocytes has been also observed in vivo, further supporting the implication of S100B in PD pathophysiology [8, 42, 43, 77]

S100B as a potential PD biomarker

Serum S100B levels have been elevated in patients with melanoma, acutely after brain damage, heart ischemia and intense physical exercise, potentially reflecting its release from damaged cells and entering into the peripheral circulation [14].

Clinical evidence points to the potential role of serum and CSF S100B as a PD biomarker. Although some studies have reported no differences in serum S100B levels between PD patients and controls [8, 52], serum S100B levels have been positively correlated with the severity of motor symptoms of PD patients, as assessed by Hoehn and Yahr (H&Y) and Activities of Daily Living (ADL) scales [52]. Overnight elevation of serum S100B levels has been further associated with increased disease severity and worse subjective sleep symptoms of patients with moderately advanced PD [53], implying that PD severity and non-restorative sleep may be related to sleep-related exaggerated brain inflammation or inadequate removal of potentially neurotoxic waste products. Deep brain stimulation (DBS) that is currently used for the treatment of advanced PD has been shown not to alter S100B levels in the serum of PD patients, suggesting that this surgical procedure may not affect serum S100B concentration [54]. Therefore, serum S100B might be a useful biomarker for disease severity or rate of progression of PD, potentially reflecting an “active”, pro-inflammatory molecular background of the disorder. However, given the elevation of S100B levels in other acute inflammatory conditions, caution is required for its specificity.

Additionally, serum autoantibodies against S100B (S100B-Abs) have been found four times higher in PD patients, 5 years after the diagnosis of the disease compared to controls, while they reached baseline levels at 10 years after PD diagnosis [55]. This biphasic pattern agrees with some of the abovementioned preclinical studies, reporting an initial transient increase of S100B in PD animal models and highlights the potential utility of S100B-Abs as a diagnostic tool at the initial stages of the disorder, which are diagnostically challenging. In accordance, increased CSF S100B levels have been detected in the early but not in late stages of Alzheimer’s disease [56]. However, the factor of age should be also carefully considered, given the fact that S100B levels are increased during normal aging [14]. Another study has demonstrated that serum S100B-Abs were higher in PD patients compared to controls [57], as well as in patients with PD dementia (PDD) and dementia with Lewy bodies (DLB) compared to AD, FTD, vascular dementia patients and controls [58], pointing towards a potential specificity of S100B for α-synucleinopathies, including PD. On the contrary, no significant differences in these autoantibodies have been detected between PD patients and controls, as well as in PD patients with and without dementia in another study [59].

In a large study of PD patients, S100B levels in the CSF were elevated compared to controls [8]. However, although increased CSF S100B levels have been demonstrated in patients with early PD (up to 36 months after clinical diagnosis) compared to controls, these results were not confirmed in an independent additional cohort [60]. In accordance, no significant differences have been reported in S100B levels of ventricular CSF derived from PD patients in a postmortem study [61]. Moreover, CSF S100B levels were higher in patients with PDD and DLB compared to controls [62]. While a gradual reduction of CSF S100B levels was also identified from DLB to PD, with PDD patients displaying intermediate values, no correlation was observed between S100B levels and the degree of cognitive impairment [62].

In summary, although existing evidence on the role of S100B as a PD biomarker are rather conflicting and the relevant results are characterized by relatively low reproducibility, S100B levels seem to be possibly associated with the severity of motor and non-motor PD symptoms, suggesting its potential efficacy as a marker of disease progression. Furthermore, S100B could facilitate the detection of PD endophenotypes, given the altered S100B levels in PD patients with dementia or sleep disorders. Since S100B levels are shown to differ significantly between PDD or DLB and other types of dementia, further studies are needed to validate this observed S100B specificity. In addition, given the fact that striatal S100B levels were found elevated in pre-symptomatic MPTP-treated mice [43], the potential serum or CSF S100B elevation in pre-symptomatic patients at risk for PD should be also further investigated. Finally, these controversial findings highlight the importance of the exact stage and severity of the disease that should be taken into consideration in further clinical studies investigating the role of S100B as a potential PD biomarker, since these factors could remarkably contribute to the observed differences.

S100B gene polymorphisms and PD risk

S100B gene is located on chromosome 21q22.2-q22.3 and consists of two exons [63]. Since the S100B rs3788266 variant has been associated with schizophrenia and psychosis in bipolar disorder [64, 65], it has been proposed that S100B gene variants may be implicated in the dopaminergic dysfunction accompanying PD.

A study screening a Chinese PD population for S100B gene polymorphisms associated with PD risk, detected two S100B variants, rs187503470 and rs1051169 [63] which, however, did not affect the amino acid sequence or splice site, suggesting that they may not be pathogenic mutations [63]. Moreover, no association was reported between the single-nucleotide polymorphism (SNP) rs3788266 of the S100B gene and PD occurrence [8]. Another recent study of a Swedish population has shown that the SNPs rs9722, rs881827, rs2239574, rs1051169 and rs9984765 of the S100B gene were not associated with PD development [66]. However, variants of rs9722, rs9984765, rs881827 and rs1051169 were indicated to be significantly more frequent in early-onset PD compared to late-onset, suggesting their potential role in determining the age of onset of the disease [66]. In particular, the higher frequency of the T allele of rs9722 in early-onset PD was confirmed in two independent Swedish cohorts in this study [66]. Importantly, healthy individuals carrying the T allele of rs9722, located in the 3′ untranslated region, displayed higher serum and frontal cortex S100B levels [67], implying a functional role of this variant, further supporting the detrimental role of S100B overexpression in PD development. These results suggest that S100B dysfunction may rather not be the initial cause of PD, but rather affect the rate of neurodegeneration and the “threshold” for the development of PD clinical symptoms. Early-onset and late-onset PD subtypes are characterized by clinical differences, potentially reflecting distinct pathophysiological backgrounds. Environmental factors, such as tobacco smoking and coffee consumption, have been shown to diversely affect the risk of PD development between these subgroups [68]. In this context, although existing evidence on the effects of smoking on S100B is limited [69], further studies could also examine these associations.

S100B as a potential therapeutic target against PD

Given the implication of S100B in PD pathogenesis, it has been proposed that it may represent a novel target for the development of treatment approaches. S100B neutralizing antibodies may represent a potentially relevant strategy, although there are no data regarding their use in clinical settings yet. At the transcriptional level, duloxetine, an antidepressant acting as a serotonin-norepinephrine reuptake inhibitor, has been identified as a transcriptional inhibitor of S100B [70]. Interestingly, duloxetine, which is able to cross BBB, has been demonstrated to exert antitumor properties in gliomas in vivo by downregulating S100B [70]. However, the selective serotonin reuptake inhibitor, fluoxetine, has been shown to stimulate S100B secretion from serotoninergic neurons, thus downregulating miR-16 in noradrenergic neurons, which consequently acquired properties of serotoninergic neurons [71]. Due to existing uncertainty of these results [72], the use of selective serotonin reuptake inhibitors to repress S100B expression should be viewed with caution. Intravenous administration of arundic acid, an agent indicated to inhibit astrocytic S100B biosynthesis at a transcriptional level, has shown promising results in cerebral ischemic brain damage in vivo [22, 73]. Oral administration of arundic acid to Tg2576 mouse models of AD has been demonstrated to ameliorate cerebral amyloidosis, reduce β-amyloid deposits and gliosis [74]. Importantly, in MPTP-treated mice, intraperitoneal administration of S100B has been indicated to protect dopaminergic neurons and ameliorate behavioral deficits [75]. Three distinct sites of S100B have been shown to bind to specific small molecule inhibitors, including SEN205A, compounds that covalently modify Cys84, and chlorpromazine that binds to site 1, 2 and 3 of S100B, respectively [76]. Currently, research efforts aim to improve the selectivity and bioavailability of these S100B inhibitors [76].

Pentamidine, an anti-protozoal agent that has been investigated in patients with melanoma in clinical trials, is an effective S100B inhibitor but with limited ability to cross BBB [77]. In a recent study of MPTP-treated mice, the intranasal delivery of chitosan-coated niosomes with entrapped pentamidine (inPentasomes) showed BBB penetration and ameliorated motor deficits as well as inhibited dopaminergic neuronal loss, RAGE/NF-κB pathway activation and subsequent neuroinflammation in the striatum and SN [77], paving the way for future studies investigating the potential S100B-targeted therapies against PD.

Given the fact that arachidonyl-2-chloroethylamide, a selective cannabinoid CB1 receptor agonist, inhibited the MPTP-induced increased S100B expression and neuronal apoptosis [30], non-psychotropic inhibitors of endocannabinoid metabolism, including AA-5-HT and VDM11 [30], could represent, novel tools against PD-related neurotoxicity.

Endurance exercise training has been associated with lower S100B expression in the striatum of PD mouse models [52]. Treatment with palmitoylethanolamide, a compound with anti-inflammatory effects mediated by PPAR-α receptor, has been shown to protect against dopaminergic loss and microglia activation in MPTP-treated mice, as well as inhibit S100B overexpression [78]. Oral treatment with chrysin, a natural flavonoid, has been indicated to protect against 6-OHDA-induced neurotoxicity and increase of S100B in the striatum of rats [79]. However, intraperitoneal treatment for 14 days with zonisamide, an antiepileptic agent that was shown to improve PD symptoms in clinical trials [80], has been demonstrated to be accompanied by increased astrocytic S100B levels in the basal ganglia of 6-OHDA-treated mice [81]. The combined treatment of the antioxidant N-acetylcysteine and the anti-inflammatory agent HA-1077 has been associated with enhanced dopaminergic death and astrocytic activation accompanied by elevated S100B expression in MPTP-treated mice [82]. Therefore, treatment approaches against PD may diversely affect S100B levels, probably depending on the specific molecular processes underlying each therapeutic strategy.

Final comments

Although the exact role of S100B in neurodegeneration still remains obscure, increasing evidence highlights the concentration-, time- and context-specific contribution of S100B to PD pathogenesis, possibly by regulating neuroinflammation, neurotransmitter metabolism and oxidative damage. S100B may not be a pathogenic factor specific to PD, since several studies have demonstrated its role in other neurodegenerative disorders such as AD, possibly suggesting that elevated levels of extracellular S100B in the brain might simply reflect ongoing inflammation and astrocyte activation. Although existing clinical evidence is conflicting, S100B seems to hold a promising potential as a PD biomarker, especially in terms of disease progression, severity and subtyping (early-onset versus late-onset PD, PD endophenotypes), as well as diagnosis at initial stages. However, concerns are raised in regard to its specificity and its altered levels during the course of the disease among others. Future studies with stratification of patients into subgroups (early-onset versus late-onset, early stages versus late stages of the disease, sporadic versus familial) may elucidate these unclear issues. Given the recent development of S100B inhibitors able to cross BBB, novel opportunities are also arising in the research field of PD therapeutics.

Another important aspect to be considered is the effects of levodopa or dopaminergic agonists administration on S100B levels, as PD patients at later disease stages probably use higher doses of these pharmaceutical agents [52] and levodopa usage has been indicated to enhance oxidative damage and inflammation [83].

Apart from RAGE, S100B can also bind to Toll-like receptors (TLRs) in the gut [84] and inflamed bronchi [85]. TLR-2 and TLR-4 have been shown upregulated in animal models of PD, and α-synuclein can act as a ligand for TLRs [86]. Although in general S100B/TLRs interaction has not been described in the brain, the potential role of the S100B/RAGE/TLRs axis in PD could be also explored.

Although MPTP and 6-OHDA have been widely used for mimicking PD in animal models, these neurotoxins induce rapid dopaminergic cell death, compared to the slower and progressive neuronal loss in humans [8]. Therefore, preclinical evidence obtaining from these experiments should be interpreted with caution. Increased co-localization of S100B and A30P α-synuclein in astrocytes has been also reported in transgenic mice expressing mutant A30P α-synuclein [87], further supporting the implication of S100B in PD pathogenesis. Nevertheless, the role of S100B should be further examined in genetic models of PD (SNCA, LRRK2), to better clarify its effects in PD pathogenesis.

Interestingly, S100B has been demonstrated to activate enteric glial cells and S100B overexpression has been associated with the initiation and maintenance of inflammatory response in the gut too [88]. Given the increasing evidence of the crucial involvement of the enteric glial cells in PD pathogenesis [89], the role of S100B as one of the molecules linking brain-gut communication should be further investigated.

Abbreviations

PD

Parkinson’s disease

SNpc

Substantia nigra pars compacta

MPTP

1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine

6-OHDA

6-Hydroxydopamine

GDNF

Glial-derived neurotrophic factor

BBB

Blood–brain barrier

DAMP

Damage-associated molecular pattern

RAGE

Receptor for advanced glycation end products

HMGB1

High mobility group box 1

LIMK

LIM kinase

AP-1

Activating protein-1

COX-2

Cyclooxygenase-2

TNF-α

Tumor necrosis factor- α

CNS

Central nervous system

ROS

Reactive oxygen species

cAMP

Cyclic adenosine monophosphate

ERKs

Extracellular signal-regulated kinases

GRK2

G protein-coupled receptor kinase 2

DBS

Deep brain stimulation

SNP

Single-nucleotide polymorphism

Author contributions

EA carried out the literature review, conceptualized, and prepared the initial draft. YNP edited and contributed in the final manuscript. CP provided critical inputs, edited and contributed to the final version of the manuscript. All authors read and approved the final manuscript.

Funding

This research has not received any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Compliance with ethical standards

Conflict of interest

The authors have no conflict of interest to declare.