1.1.1 Aging, dopamine transporter (DAT), and the vulnerability to PD

In the presynaptic terminal, the reuptake of DA through the actions of the high-affinity DA transporter (DAT) represents a key step whereby DA is repackaged into the storage vesicles by the action of the vesicular monoamine transporter, VMAT. DAT is a sodium-coupled symporter protein belonging to the superfamily of SLC transporters, responsible for modulating the concentration of extraneuronal DA in the brain (Amara & Kuhar, 1993). Notably, association of a polymorphism in the DAT gene with Parkinson's disease (Le Couteur et al., 1997; Wang et al., 2000) underlines its potential role in PD vulnerability (Schmitt et al., 2013). Specifically, age-dependent changes in DAT and accumulation of nitrosylated tyrosine (3-nitrotyrosine, 3-NT) in rhesus monkey (Kanaan et al., 2008) and rodent mDAns (Marchetti et al., 2013) support dysfunctional DAT as a vulnerability factor for nigrostriatal degeneration. Particularly, recent studies of Illiano et al. (2020) showed that in rodents, the lack of DAT results in increased vulnerability and aberrant autonomic response to acute stress. In particular, DAT represents a preferential target for parkinsonian neurotoxins, as the active metabolite of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), MPP⁺, is specifically transported by DAT and concentrated within the nigral DA neurons where it inhibits complex I of the mitochondrial electron transport chain (METC), resulting in ATP depletion and subsequent neuronal cell death (Di Monte & Langston, 1995; Langston, 2017; Schildknecht et al., 2017) (Figure 3). The induction of oxidative stress results in the opening of mitochondrial permeability transition pore (mPTP), the release of cytochrome C, and the activation of caspases. It seems important to recall that mitochondria represent the primary energy-generating system, involved in multiple processes, including energy metabolism, reactive oxygen (ROS) generation, mitochondrial dynamics, and distribution (Blesa et al., 2015; Bose & Beal, 2016; Schildknecht et al., 2017). Of specific mention, mitochondrial damage due to Ca²⁺ overload-induced opening of mPTP is believed to play a key role in selective degeneration of nigrostriatal DAns in PD. Hence, endoplasmic reticulum (ER) acts as a reservoir of Ca²⁺ ions, and increased Ca²⁺ released from the ER further enhances mitochondrial oxidative stress of mDAns in SNpc (Blesa et al., 2015; Schildknecht et al., 2017). Reportedly, reduction in complex I activity in the SNpc of patients with sporadic PD has been well described, being considered as one of the primary sources of ROS in PD, and accounting for the majority of mDAn cell death (Hattori et al., 1991; Hattingen et al., 2009; Schapira et al., 1990). Of note, in Str, DA terminals actively degenerated proportionally to increased levels of DA oxidation following a single injection of DA into the striatum (Rabinovic et al., 2000).

Not only too little but also too much of DAT-mediated mechanisms may have harmful consequences, since increased uptake of DA through DAT in transgenic (Tg) mice overexpressing the DA transporter results in oxidative damage, neuronal loss, and motor deficits (Masoud et al., 2015). Here, the effects of increased DAT expression on DA homeostasis, neuronal survival, oxidative stress, and motor behavior of DAT-Tg mice were evaluated together with the nigrostriatal response to MPTP (Masoud et al., 2015). Hence, an almost 30%–36% loss of mDAns and fine motor deficits were associated with an increased vulnerability to MPTP-induced mDAn loss, indicating that overactivation of DAT-mediated uptake of dopamine leads to basal neurotoxicity and heightened sensitivity to exogenous insults (Masoud et al., 2015).

Together, the presynaptic transporter DAT in nigral dopaminergic neurons confers susceptibility and represents a principal age-dependent vulnerability factor for PD.

1.1.2 Aging, DA oxidative metabolism, and nigrostriatal neuron vulnerability in PD

Given the high metabolic activity that is required to support their extensive axonal arborization, mDAns are physiologically subjected to various levels of oxidative stress, and reciprocally, among a number of brain regions studied, the SNpc, where A9 DA cell bodies are located, is the more vulnerable region, as DA metabolism constantly generates ROS (Chinta & Andersen, 2008). Notably, the aging process, associated with a progressive mDAn dysfunction, may add a further oxidative load to the system, with harmful consequences for nigrostriatal neuron integrity (as summarized in Section 1.2). Two enzymes are primarily responsible for DA inactivation, monoamine oxidase isoforms (MAO-A and MAO-B) and catechol-O-methyl transferase (COMT), predominantly expressed by glial cells. MAO, a flavin-containing enzyme is located on the outer membrane of the mitochondria. This enzyme oxidatively deaminates catecholamines to their corresponding aldehydes; these can be in turn converted either by aldehyde dehydrogenase to acids or by aldehyde reductase to form glycols. Due to its intracellular localization, MAO has a strategic role in the inactivation of DA when the amine is not protected by the storage vesicles in presynaptic terminal. MAO breaks down DA to 3,4-dihydroxyphenylacetaldehyde (DOPAL), which, in turn, is degraded to form 3,4-dihydroxyphenylacetic acid (DOPAC) by the action of the enzyme aldehyde dehydrogenase. COMT converts DA to 3-methoxytyramine (3-MT), which is further reduced by MAO to homovanillic acid (HVA) and then eliminated in the urine. During DA metabolic steps, ROS and RNS can be produced (Afanas, 2005). These may include hydrogen peroxide (H₂O₂), singlet oxygen (¹O2), hydroxyl (OH), and superoxide (O2) radicals (Halliwell & Gutteridge, 1984; Kumar et al., 2012; Sies et al., 2017). RNS are produced in neuronal cells from arginine by the neuronal nitric oxide synthase (nNOS) and include nitric oxide (NO), nitrite (NO2), and S-nitrosothiols and peroxynitrite (OONO) (Adams et al., 2015). Additionally, DA metabolites and certain derivatives such as N-methyl-(R)-salsolinol (NMSAL) (Naoi et al., 2002) are prone to oxidation, generating reactive quinones, which may further engender a neurotoxic cycle able to readily modify proteins and potentially cause protein aggregation (Sulzer & Zecca, 2000; Zucca et al., 2014).

Overall, age- and PD-dependent chronic DA neuronal dysfunction, altered DA metabolism, and dysregulated reactive species production then have to face the harmful gene x environment interactions promoting a feedforward oxidative/inflammatory cycle, contributing to progressive neuronal deterioration and motor deficiency of PD.

1.2.1 Aging and the glial inflammatory network in PD

A critical hallmark of aging is the progressive decline in nigrostriatal DA neurons (Bezard & Gross, 1998; Boger et al., 2010; Collier et al., 2007; de la Fuente-Fernández et al., 2011; Hindle, 2010) associated with the failure of the adaptive/compensatory potential of mDAns, recognized to be implicated in the slow but progressive nigrostriatal degeneration of PD, with the late appearance of clinical signs (Bezard & Gross, 1998; Hornykiewicz, 1993; Kanaan et al., 2008). Here, a crucial causative role is represented by the exacerbation of the astroglial microenvironment, as a result of a dysfunctional gene–environment crosstalk. Reportedly, the major aging culprits, namely oxidative stress and low-grade inflammation, may further be exacerbated under basal ganglia injury, neurotoxin exposure, male gender, and PD genetic mutations (Gao & Hong, 2011; Gao et al., 2003, 2011; Hu et al., 2008; Marchetti & Abbracchio, 2005). Notably, with age, defective mitochondrial turnover by autophagy may trigger chronic inflammation and critically contribute to the impairment of immune defense, in as much as malfunctioning autophagy has been reported in several diseases NDs, including PD, with its consequent toxicity considered to be a main cause of the disease (Bektas et al., 2019; Cuervo, 2008; Cuervo & Macian, 2014; Scrivo et al., 2018). Recently, the pentose phosphate pathway (PPP, a metabolic pathway parallel to glycolysis), which converts glucose-6-phosphate into pentoses and generates ribose-5-phosphate and NADPH thereby governing anabolic biosynthesis and redox homeostasis, has gained a critical attention (Tu et al., 2019). Hence, expression and activity of G6PD were elevated in an in vitro model of PD (e.g., LPS-treated midbrain neuron–glial cultures) and the SN of vivo PD models, associating with microglial activation and mDAn neurodegeneration, whereas inhibition of G6PD elevation or knockdown of microglial G6PD attenuated LPS-elicited chronic mDAn neurodegeneration (Tu et al., 2019). Further, microglia with elevated G6PD activity/expression produced excessive NADPH and provided abundant substrate to overactivated NADPH oxidase (NOX2) resulting in exacerbated ROS, which suggests that G6PD and NOX2 are potential therapeutic targets for PD (Tu et al., 2019).

In glial cells, mutated genes then cooperate with environmental influences to impair mitochondrial homeostasis and the autophagy–lysosomal pathways, all implicated in mDAn dysfunction observed in PD (Ashley et al., 2016; Barodia et al., 2019; Booth et al., 2017; Choi et al., 2016; Cuervo & Macian, 2014; Dzamko et al., 2015; Gillardon et al., 2012; Kim et al., 2013; Kline et al., 2021; Lastres-Becker et al., 2012; Marchetti, 2020; Schmidt et al., 2011). While this topic is outside the scope of the present work, it seems important to underscore that all major human PD-linked mutations (i.e., SNCA, LRRK2, PINK1, and DJ1) induce complex I inhibition, and synergism with the age-dependent oxidative stress and inflammation further promotes increased generation of oxidative and nitrosative stress mediators, in turn exacerbating the proinflammatory microglial “M1” phenotype, then promoting progression of mDAn death (Lastres-Becker et al., 2012). Notably, with age, progressive acquisition by glial cells of the capacity to produce greater levels of a set of proinflammatory mediators both in physiological conditions, and more actively under immune or neurotoxic stimuli on the one hand, coupled to the failure of host surveillance systems, on the other, can translate into harmful consequences both at central and at peripheral levels (Boche et al., 2013; De Cecco et al., 2019; Perry & Teeling, 2013; Tansey & Romero-Ramos, 2019). This so-called “microglial cell shift” to the “harmful,” M1 phenotype promoting the release of an array of factors that are detrimental for the vulnerable mDAns depends upon inflammasome activation.

1.2.2 Aging, inflammasome activation, and mitochondrial dysfunction in PD

Significantly, nuclear factor kappa-light-chain-enhancer of activated B cells (NF-ĸB, a protein complex that controls cytokine production and cell survival) is the first signal for inflammasome induction and a key interactor of DA signaling (as detailed in the next section). Among the numerous inflammatory cytokines, interleukin-1β (IL-1β) produced by glial cell Nod-like receptor protein (NLRP) inflammasome exerts a central role in regulating neuroinflammation (Codolo et al., 2013; Haque et al., 2020; Heneka et al., 2014). Upon stimulation by adenosine triphosphate (ATP), reactive oxygen species, lysosomal contents, or other factors, NLRP3 recruits the adapter molecule apoptosis-related speck-like protein (ASC) and procaspase-1 to promote caspase-1 activation (Dinarello, 2007). This process leads to the maturation of the proinflammatory cytokines (IL-1β, IL-18). The secretion of IL-1β by glial cells contributes toward the destruction of mDAns in the brain of PD patients and the initiation of cell death (McGeer & McGeer, 2008). Hence, MPTP-driven NLRP3 inflammasome activation in microglia plays a central role in mDAns demise (Gordon et al., 2018; Lee, 2018), in as much as aging represents a synergic trigger directing microglia toward the M1 proinflammatory phenotype (L’Episcopo, Tirolo, Testa, Canigilia, Morale, Impagnatiello, et al., 2011). Additionally, mitochondrial impairment in microglia amplifies NLRP3 inflammasome proinflammatory signaling in cell culture and animal models of PD (Sarkar et al., 2017; Zhu et al., 2021), whereas the suppression of NLRP3 inflammasome-derived proinflammatory cytokines mitigates mDAn degeneration and may be beneficial to PD patients (Ahmed et al., 2021; Gordon et al., 2018; Haque et al., 2020; Zhu et al., 2018). Interestingly, the vicious crosstalk between the impaired mitochondrial signaling and NLRP3 machinery can contribute to amplify further the noxious mDAns outcome, as NLRP3/caspase-1 activation under toxic exposure is mediated by mitochondrial ROS generation (Afonina et al., 2017; Sarkar et al., 2017).

Altogether, glia acts as a common final pathway of gene x environment interactions in PD, playing critical roles in the exacerbation of age-dependent mDAn degeneration, and intersecting the harmful DA oxidative metabolism. As a result, the modulatory role of DA signaling in glial cell networks appears decisive, since they might either help the imperiled mDAns to combat oxidative stress and inflammation through a wide variety of mechanisms addressed in the following sections.

1.3.1 DA signaling intersects harmful microglial inflammatory networks in PD: DA/NF-ĸB /NLRP3 crosstalk

A most robust evidence linking DA to inflammation is the recognized notion that DA deficit within the nigrostriatal system, as observed in preclinical and clinical models of PD, strongly associates with exaggerated inflammation both at central and at peripheral levels. Studies conducted in the MPTP model of PD, including our own results, clearly showed an inverse relationship between microglial inflammatory activation and the sharp inhibition of DA, DRD2 and DAT in striatum of basal ganglia-injured mice (Serapide et al., 2020). The greatest effects were observed in aged mice, coincident with a robust activation of major proinflammatory transcripts, including NF-ĸB, IL-1β, TNF-α, and IL-6, as well as oxidative and nitrosative stress markers such as ROS, RNS, and 3-NT. Likewise, at the midbrain level, progressive decline in DA resulting from the aging process associates with increased reactivity of the microglial cell compartment, further amplified after basal ganglia injury, in the face of downregulation of major DA transcripts, including DRD2 and protein levels in the midbrain (L’Episcopo et al., 2018). Such a dramatic loss of DA inhibitory tonus onto nigrostriatal astrocytes and microglia likely contributes to the observed exacerbated neuroinflammation during aging and PD (Serapide et al., 2020).

Reportedly, microglia harbor DA receptor subfamilies (Pocock & Kettenmann, 2007). Studies of Mastroeni et al. (2009) showed that cultured human elderly microglia expressed mRNAs for DRD1-DRD4 but not DRD5 receptors (Table 1). In addition, PD microglia in situ were also immunoreactive for DRD1-DRD4 but not for DRD5 receptors, suggesting that activated PD microglia expressing DA receptors might play roles in the selective vulnerability of DA neurons in PD (Mastroeni et al., 2009). In PD rats, DRD1 activated by acetyl-L-carnitine attenuates microglial activation and the release of proinflammatory mediators, a phenomenon potentially linked to the amelioration of cognitive deficits and neurodegeneration (Singh, Mishra, Mohanbhai, et al., 2018). Here, acetyl-L-carnitine inhibited microglial activation-mediated inflammatory response and weakened TNF-α levels by increasing the production of the anti-inflammatory cytokine, IL-10, which led to improved neuronal survival (Singh, Mishra, Mohanbhai, et al., 2018), implicating DA regulation of inflammasome/NF-ĸB pathway (recently reviewed by Feng & Lu, 2021). Within this context, emerging evidence also indicates that microglial polarization and generation of ROS are tightly related to the DA-targeted brain intrinsic renin–angiotensin system (RAS), a local/paracrine modulatory mechanism playing an important role in inflammatory processes (Dang et al., 2021; Dominguez-Meijide et al., 2017; Gong et al., 2019; Mowry & Biancardi, 2019; Xia et al., 2019).

The roles of DA signaling in regulating these key inflammatory pathways stem from the transduction machinery of DRD1/DRD2-like receptor subtypes. In DRD1-like receptors, the elevated cAMP induced by DA directly binds to NLRP3 (Figure 4). Here, elevated cAMP activates PKA and phosphorylates cAMP-response element binding protein (CREB), thus disrupting NF-ĸB homeostasis and resulting in the inhibition of the inflammatory response (Neumann et al., 1995; Xia et al., 2019 and Refs herein). Also, via DRD5 signaling, DA can block NF-kB pathway, thus suppressing proinflammatory mediators (Wu et al., 2020; Zhang et al., 2015). DRD2-like receptor signaling may involve either a GPC-dependent or a β-arrestin-dependent GPC-independent pathway to modulate glial inflammatory activation, with the β-arrestin-dependent mechanism playing a critical role (Fan, 2014) (Figure 4).

Interestingly, DA receptor expression is induced by the activated microglial phenotype, as cerebral ischemia induced the expression of DRD2 on Iba1-immunoreactive inflammatory cells in the infarct core and penumbra (Huck et al., 2015). Similarly, DA has a differential role in influencing cellular functions of resting and activated microglia, such as phagocytosis and adhesion, depending on the activation states of microglia (Fan et al., 2018). Notably, while DRD3 were reported not to be expressed in microglial cells, DRD3 deficiency results in attenuated microglial activation upon systemic LPS treatment (Montoya et al., 2019). Hence, the role of DRD3 signaling in the acquisition of inflammatory phenotype by microglial cells was recently further studied by the determination of the M1 and M2 phenotypes acquired by microglia 24 h after LPS treatment in WT and DRD3-KO mice (Montoya et al., 2019). Interestingly, the percentage of M1 microglia was not affected by genetic deficiency or pharmacological antagonism of DRD3 signaling, but the percentage of M2 phenotype in microglial cells was significantly reduced upon DRD3 antagonism in LPS-treated WT mice (Montoya et al., 2019). On the bases of these and other results (Elgueta et al., 2017; Montoya et al., 2019), DRD3 has been indicated to be expressed selectively in astrocytes, but not in microglial cells, thereby implicating astrocyte intermediacy in M1-M2 microglial switch.

1.3.2 DA intersect astrocyte's harmful signaling in PD: DA/αβ-crystallin /STAT3 crosstalk

Astrocytic DA modulation carried out by DRD2 can suppress neuroinflammation through αB-crystallin-dependent mechanism (Figure 5). Hence, DRD2 agonist quinpirole increased resistance of the nigral dopaminergic neurons to MPTP through partial suppression of inflammation (Shao et al., 2013). Conversely, knockout mice lacking DRD2 showed robust inflammatory responses in different brain regions. Additionally, DRD2 knockout increased the vulnerability of mDAns to MPTP-induced neurotoxicity. Interestingly enough, DRD2-deficient astrocytes became hyper-responsive to immune stimuli in the face of a significant decrease in the level of αB-crystallin (Shao et al., 2013). Further evidence comes from experiments carried out after ablation of DRD2 in astrocytes resulting in a robust activation of astrocytes in SNpc (Shao et al., 2013). Using gain-of function or loss-of-function settings and pharmacological treatments with the selective DRD2 agonist, quinpirole, increased resistance of the SNpc DA neurons to MPTP, through a partial suppression of inflammation. Overall, these studies indicated that astrocytic DRD2 activation physiologically downregulates neuroinflammation in the studied model, via αB-crystallin-dependent mechanism, suggesting a potential novel approach aimed at targeting the astrocyte-mediated innate immune response (Shao et al., 2013). Likewise, in the study of Qiu et al. (2016), sinomenine was shown to activate astrocytic DRD2 receptors, thereby alleviating neuroinflammatory injury via the αβ-crystallin /STAT3 pathway after ischemic stroke in mice.

A novel interaction between DA and α-Syn was recently studied by Du et al. (2018). Here, the authors showed that the selective DRD2 agonist quinpirole can suppress inflammation in the midbrain of wild-type mice, but not in α-Syn-overexpressing mice. DRD2 agonists were also capable to significantly mitigate LPS-induced inflammatory response in astrocytes (Du et al., 2018).

Interestingly, such DRD2-mediated anti-inflammatory effect was dependent on β-arrestin-2-mediated signaling, but not on classical G protein pathway. Additionally, α-Syn reduced the expression of β-arrestin-2 in astrocytes, whereas it increased the β-arrestin-2 expression and restored the anti-inflammatory effect of DRD2 in α-Syn-induced inflammation. Such α-Syn-mediated disruption of DRD2 anti-inflammatory effect was carried out by inhibiting the association of β-arrestin-2 with transforming growth factor-beta-activated kinase 1 (TAK1)-binding protein 1 (TAB1) and promoting TAK1-TAB1 interaction in astrocytes, underscoring the ability α-Syn disrupts the function of β-arrestin-2 and inflammatory pathways in the pathogenesis of PD (Du et al., 2018). DRD2 agonists were also found to suppress the upregulation of caspase-1 and IL-1β expression in primary cultured mouse astrocytes in response to LPS plus ATP-induced NLRP3 inflammasome activation (Zhu et al., 2018). Furthermore, using the MPTP mouse model of PD, the authors found that DRD2 agonists inhibited NLRP3 inflammasome activation, evidenced by decreased caspase-1 expression and reduced IL-1β release in the midbrain of wild-type mice. Such anti-inflammasome effect of DRD2 was abolished in β-arrestin-2 knockout and β-arrestin-2 small interfering RNA-injected mice, suggesting a critical role of β-arrestin-2 in DRD2-regulated NLRP3 inflammasome activation (Zhu et al., 2018).

On the contrary, the studies of Elgueta et al. (2017) showed a selective DRD3 transcription in astrocytes but not in microglia. Interestingly, D3R selective antagonist PG01037 reduces the acquisition and activation of M1 phenotype microglia, contributing to an anti-inflammatory effect and displaying a significant therapeutic effect in PD mouse model (Elgueta et al., 2017).

Of note, the DRD3 immunoreactivity in astrocytes is associated with a clustered pattern, resembling the expression pattern observed for those proteins contained in lipid rafts (see Montoya et al., 2019).

Altogether, DA powerfully modulates glial inflammatory responses via both D1 and D2 receptor subtypes, via its intersection within the major proinflammatory circuits. Moreover, DA-mediated counter-regulation of immune response may change according to the activation stage and/or the severity of the proinflammatory glial phenotype, thus suggesting that DA-mediated immunomodulation not only varies according to the DA receptor subtype and operated transduction pathway but also depends on the severity of inflammation and the counter-modulatory effects elicited by DA crosstalk with key antioxidant/anti-inflammatory/cytoprotective Nrf2/Wnt pathways, as discussed below inthe next sections).

1.4.1 DA-Nrf2 crosstalk

Nrf2 activation-induced targeting of antioxidant response elements (AREs) in the promoter region of several hundred genes results in the promotion of a wide panel of cytoprotective, anti-inflammatory and phase 2 proteins, such as heme oxygenase (HO1), NAD(P)H quinone oxidoreductase (NQO1), superoxide dismutases (SOD1, SOD2), glutathione S-transferase (GST), glutathione peroxidase (GPx), glutathione reductase (GR), and catalase (CAT), which together are capable of regulating the cellular redox state by decreasing ROS. Specifically, Nrf2 has a multifaceted role in mitochondrial function and inflammatory networks (Blackwell et al., 2015; Dinkova-Kostova & Abramov, 2015; Holmström et al., 2016; Ryoo & Kwak, 2018). Of major importance, Nrf2 induction is primarily observed in non-neuronal cells. In astrocytes, this inducible mechanism coordinates expression of several cellular defense pathways including the following: detoxification of reactive oxygen/nitrogen species and xenobiotics, GSH synthesis, and generation of NADPH (see Vargas & Johnson, 2009). Notably, Nrf2 is an important player in the pathogenesis of cancer and common inflammatory, age-dependent, and most neurodegenerative diseases, and its multifunctional role has been emphasized in several earlier and more recent studies and reviews (Abdalkader et al., 2018; Cano et al., 2021; Cuadrado et al., 2019; Dinkova-Kostova & Abramov, 2015; Johnson et al., 2008; Lastres-Becker, 2021; Lastres-Becker et al., 2016; Marchetti, 2020; Strong et al., 2016; Vargas & Johnson, 2009). Activation of Nrf2 in astrocytes protects neurons from a wide array of insults in different in vitro and in vivo paradigms, including MPTP-induced mDAn neurotoxicity, whereas Nrf2 deficiency contributes to neuronal death, supporting the role of astrocytes in determining the vulnerability of neurons to noxious stimuli, in particular mDAns (Calkins et al., 2010; Chen et al., 2009; Copple et al., 2010; Gan et al., 2012; Vargas & Johnson, 2009) (Figure 5 and Table 2). Of note, loss of Nrf2 in the presence of α-syn expression cooperates to aggravate protein aggregation, neuronal death, and inflammation in early-stage PD (Lastres-Becker et al., 2012), further highlighting the critical role of gene–environment harmful interactions in PD.

Also, in aged MPTP mice, the old parenchymal astrocytes in VM loose both DRD2 and Nrf2 transcriptional activity, whereas grafting young astrocytes rejuvenates the microenvironment, resulting in a gain of Nrf2 function, as ARE transcriptional activity and mitochondrial beneficial effects are associated with mDAn neurorescue (Serapide et al., 2020). In particular, in vivo and ex vivo experiments carried out in astrocyte-grafted aged MPTP mice underscored the ability of “young” astrocyte's grafts to reprogram the aged parenchymal astrocyte metabolic activity, switching mitochondrial dysfunction, in turn resulting in mitigation of ROS, RNS, and inflammatory mediators, compared with aged MPTP control astrocytes transplanted with a non-specific cell type (Serapide et al., 2020).

Further, the activation of Nrf2 enables protection against 6-hydroxydopamine-(6-OHDA)-induced ferroptosis, a form of cell death involving the iron-dependent accumulation of GSH depletion and lipid peroxide in DA cells (Sun et al., 2020; Wei et al., 2020). By contrast, Nrf2 deficiency was associated with exaggerated mitochondrial dysfunction and blockade of Nrf2’s mitochondrial protective response, as recently reported by Cano et al. (2021) in Nrf2-deficient retinal pigmented epithelium. The pivotal function of Nrf2 stems from its modulatory role on key aspects of mitochondrial health (see Ammal Kaidery et al., 2019; Cano et al., 2021; Ryoo & Kwak, 2018). Interestingly, in Caenorhabditis elegans, where the Nrf proteins are represented by their ortholog SKN-1, recent studies implicate Nrf/SKN-1 in a wide range of homeostatic functions (Blackwell et al., 2015). Reportedly, as underscored by Blackwell et al. (2015), “SKN-1 plays a central role in diverse genetic and pharmacological interventions that promote C. elegans longevity, suggesting that mechanisms regulated by SKN-1 may be of conserved importance in aging” (Blackwell et al., 2015). Accordingly, a number of experimental approaches evaluating the potential regulation of the transcription factor Nrf2 to enhance the expression of genes that contrast oxidative stress and promote healthy aging have been provided, particularly with Nrf2 activators described to expand the life span, contrasting oxidative stress and inflammation (Liu et al., 2009; Nelson et al., 2006; Strong et al., 2016; Velmurugan et al., 2009).

Against this background, a direct DA-Nrf2 crosstalk may represent a further protective mechanism whereby DA activation triggers Nrf2-regulated pathways (Figure 5). Hence, in astrocytes, excessive extracellular DA itself likely served as an endogenous signal to activate Nrf2-dependent neuroprotective pathways (Shih et al., 2007). Indeed, the ability of Nrf2 activation in protecting cells from DA toxicity has long been recognized, and in part attributable to enhanced H₂O₂ scavenging by the GSH system, and detoxification of reactive quinones by NAD(P)H: NQO1 (Duffy et al., 2018). Particularly, physiological oxidative stressors or subthreshold concentrations of neurotoxins support DA neuron survival and neural stem progenitor cell (NSC) differentiation via activation of Nrf2/Wnt signaling in glial cells (Marchetti et al., 2020; Wang et al., 2020). Notably, DA activation of DRD1-like receptor, DRD5, was recognized as a necessary trigger for the normal expression of Nrf2 and inhibition of harmful oxidative cascades, as DRD5 deficiency causes an increase in NADPH oxidase activity and prevents the translocation of Nrf2 nuclear (Jiang et al., 2018). Then, DA activation of D1-like receptors in astrocytes might further contribute to an autoregulatory feedback triggered by endogenous DA and DA agonists (Figure 5, Table 2).

Together, DA-Nrf2 crosstalk appears a feasible counter-regulatory mechanism triggered by DA to prevent the deleterious effects of exacerbated oxidative stress and inflammation.

1.4.2 DA-Wnt/β-catenin crosstalk

Earlier studies on functional interactions between DA and Wnt/β-catenin signaling focused on DRD2 under long-term treatment with antipsychotic drugs, which are the blockers of D2-like receptors (Alimohamad et al., 2005; Freyberg et al., 2010), supporting a functional interaction between Wnt pathway and DRD2/DRD3. The chief role of Wnt signaling for neurogenesis in the adult and aged PD brain has been recently reviewed (Marchetti et al., 2020). During age and basal ganglia injury, the progressive decline in DA targeting glial cells via DRD2 in VM and Str of aged MPTP-treated PD mice was associated with decreased Wnt/β-catenin signaling genes and proteins, in turn affecting both glial cell reactivity and mDAn loss (Marchetti, 2018). Moreover, decreased D1 receptor expression, mitochondrial biogenesis, mitochondrial functions, and dopaminergic neuron differentiation were associated with downregulation of Wnt/β-catenin signaling in the hippocampus of rats lesioned with the PD neurotoxin, 6-OHDA (Mishra, Singh, Tiwari, Chaturvedi, et al., 2019; Mishra, Singh, Tiwari, Parul, et al., 2019). Conversely, pharmacological stimulation of D1 receptor enhanced mitochondrial biogenesis, mitochondrial functions, and DA neurogenesis that lead to improved motor functions in 6-OHDA-injured rats. The specificity of these effects was underscored using a D1 antagonist, whereas shRNA-mediated knockdown of Axin-2, a negative regulator of Wnt signaling, significantly abolished D1 antagonist-induced impairment in mitochondrial biogenesis and DA neurogenesis in 6-OHDA-lesioned rats (Mishra, Singh, Tiwari, Chaturvedi, et al., 2019; Mishra, Singh, Tiwari, Parul, et al., 2019).

A number of studies investigated the molecular mechanisms of DRD2-Wnt/β-catenin crosstalk (Han et al., 2019; Min et al., 2011). In the study of Min et al. (2011), among the five DA subtypes, DRD2 interacted with β-catenin through the second and third intracellular loops and inhibited the entry of β-catenin into the nucleus, leading to an inhibition of the LEF-1-dependent transcription (Min et al., 2011). In this work, the authors suggested that the functional regulation of Wnt signaling by DRD2 could occur through direct interaction with β-catenin independently of the upstream signaling components (Min et al., 2011). Notably, of the two DRD2 downstream intracellular pathways, the β-arrestin-dependent pathway appears to be the one targeting Wnt/β-catenin signaling (Bryja et al., 2007), with GSK-3β, being the critical intersector, and the contribution of serine/threonine kinase (AKT) counter-regulation (Figure 6).

In fact, in addition to AKT’s roles in β-arrestin-2-dependent DRD2 signaling, AKT regulates GSK-3β through phosphorylation. In its non-phosphorylated state, GSK-3β is constitutively active, whereas AKT-induced phosphorylation inactivates GSK-3β (Beaulieu et al., 2007) (Figure 6). Regarding the so-called “canonical Wnt/β-catenin” signaling, GSK-3β is part of a destruction complex, whereby GSK-3β-induced phosphorylation of β-catenin results in its proteasomal degradation, blockade of β-catenin nuclear translocation associated with inhibition of Wnt-dependent transcription of a panel of downstream target genes. Then, DA activation of DRD2-β-arrestin-2-dependent pathway may also modulate Wnt signaling via AKT-mediated phosphorylation of GSK-3β, thereby modulating β-catenin nuclear translocation (Figure 6).

Significantly, in the study of Han et al. (2019), DRD2-dependent crosstalk was shown to modulate Wnt3a expression via an evolutionarily conserved TCF/LEF site within the Wnt3 promoter. Moreover, DRD2 signaling also modulated cell proliferation and modifies the pathology in a renal ischemia/reperfusion injury disease model, via its effects on Wnt/β-catenin signaling, thus suggesting DRD2 as a transcriptional modulator of Wnt/β-catenin signal transduction, with broad implications for health and development of new therapeutics (Han et al., 2019).

Importantly, DRD2-mediated Wnt-β-catenin signaling also crosstalks with major immune signaling actors. Hence, if not phosphorylated by GSK-3β, β-catenin forms a complex with both the units of NF-κB, altering its DNA binding activity, and consequently inhibits the inflammatory cascade (Marchetti & Pluchino, 2013). However, when GSK-3β is activated, it phosphorylates β-catenin protein for proteasomal degradation that directly promotes the inflammatory events (Deng et al., 2002; Marchetti & Pluchino, 2013). Activated GSK-3 also modulates CREB-DNA activity, phosphorylating NF-κB, and degrades β-catenin, thus promoting systemic inflammation.

The ability of active GSK-3β to phosphorylate Nrf2 (Cuadrado et al., 2018; Hayes et al., 2015) then represents a further vulnerability factor, as its overexpression exacerbates inflammation, thus impairing neuron–glial and glial–NSC interactions leading to enhanced neuronal vulnerability and/or cell death, associated with reduced neurorepair (Marchetti, 2020). By contrast, DA-activated DRD2-β-arrestin-2-dependent signaling via AKT can boost the antioxidant, anti-inflammatory, prosurvival, and neurogenic downstream gene cascade.

As a whole, DA-mediated signaling at the astrocyte–microglial interface via DRDs appears as a pivotal counter-regulatory system contributing to limit both Nrf2 and β-catenin phosphorylation and subsequent degradation, thereby reinforcing the Nrf2-ARE/Wnt/β-catenin neuroprotective and immunomodulatory axis to combat aging and PD (Figure 7), and can be envisaged for the treatment of other CNS diseases.