6.1. NOX2 in Parkinson’s Disease

The PD is a progressive ND characterized by progressive loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) of the brain, with a consequent reduction of dopamine (DA) levels. The hallmarks of PD are the aggregates of α-synuclein (α-syn) included in Lewy’s bodies in dopaminergic neurons in the SNpc and other areas of the brain [64]. The main disorders in PD are tremor, bradykinesia, rigidity, and balance disturbances and they influence the individuals’ quality of life [65]. Neuronal degeneration is promoted from several factors such as inflammation and oxidative stress. Most cases of PD (about 90%) are in the sporadic form. Instead, a small percentage (about 10%) represent a familial form, since genetic mutations such as Synuclein Alpha (SNCA), leucine-rich repeat kinase 2 (LRRK2), Parkin (PRKN), protein deglycase (DJ-1), phosphatase, and tensin homolog-induced putative kinase 1 (PINK1) are involved [66]. Recent studies have shown that PD patients may be more susceptible to SARS-CoV-2 infection as it compromises adaptive immunity. Furthermore, it has been hypothesized that this viral infection could induce the accumulation of α-synuclein, causing neurodegeneration. Furthermore, it has been hypothesized that this viral infection could induce the accumulation of α-synuclein, causing neurodegeneration. In this regard, it has been shown an increase of inflammation following the infiltration of SARS-CoV-2, which in turn can lead to the activation of endothelial cells in the pulmonary vascular system, as well as to the development of hypoxia. Consequently, respiratory hypoxia in COVID-19 patients could promote ROS formation and oxidative stress in the brain. Additionally, the increase of ROS levels could lead to an alteration of the redox balance in the mitochondria and thus promote apoptosis. Moreover, there is evidence that oxidative stress induced by increased ROS following hypoxia, may contribute to the death of dopaminergic neurons and thus promote PD [40].

The microglial activation and pro-inflammatory cytokines have been observed in the substantia nigra (SN) in patients with PD. Recent studies have shown that α-Syn is related to NOX2 activation, which in turn was able to induce the microglia activation, however, the mechanism is not yet known. Hou et al., investigated the role of integrins in NOX2 activation, especially in CD11b, the α chain of the αMβ2 integrin, expressed in immune cells as microglia. This experimental study was conducted on microglial cells treated with α-Syn (50, 100, and 200) nM useful for reproducing a PD model. Indeed, it was possible to observe the translocation of p47^PHOX (NOX2 subunit) in the membrane and consequently, an increase in ROS production which would represent a key step for NOX2 activation. Subsequently, α-Syn-induced microglia were treated with anti-CD11b antibodies showing neither p47^PHOX translocation nor ROS production. Likewise, as observed in vitro experiments, NOX2 did not activate in CD11b ⁻/⁻ compared to wild-type (WT) mice. Additionally, an increase in RhoA and ROCK levels was observed. To evaluate whether Rho-kinase was involved in NOX2 activation, α-Syn-induced microglia were transfected with RhoA siRNA. RhoA knockdown in microglia was able to reduce RhoA and ROCK expressions as well as inhibiting superoxide production and translocation of p47^PHOX. Thus, in this study, it was demonstrated that CD11b might play a role in α-Syn-induced NOX2 activation [67]. In compliance with these results, another study has focused on the involvement of integrins in NOX2 activation and its possible implication in PD. To reproduce a PD model, primary midbrain neuron-glia cultures were exposed to active metabolite 2,5-hexanedione (HD) (8 mM). Zhang et al. observed a significant loss of dopaminergic neurons caused by the microglia activation. Indeed, Microglia-depleted cultures subjected to the same treatment did not show significant alteration in cell viability. Furthermore, the authors observed increased ROS levels in BV2 microglial cells exposed to 4, 8, and 16 mM (HD). To clarify whether superoxide was produced by NOX2, microglia cells were co-treated with apocynin suggesting that NOX2 was responsible for inducing ROS production and also the translocation of p47^PHOX. Likewise, also in vivo was obtained an increase in p47^PHOX levels in the midbrain membranes of HD-treated mice compared to control. Instead, to assess whether αMβ2 integrin (also known as CD11b) was involved in NOX2 activation, HD-induced, microglia cells were co-treated with integrin inhibitors Arginylglycylaspartic acid (RGD) or anti-αM antibody. The findings showed that both the RGD and the anti-αM antibody were able to inhibit superoxide production and p47^PHOX translocation. Coherently, the same results were found in HD-treated mixed-glia cultures obtained from αMβ2-deficient (αM ⁻/⁻) mice Further demonstrating the implication of this receptor in NOX2 activation. Finally, the authors also assessed that the αMβ2-Src-ERK pathway would play a key role in NOX2 activation. Thus, HD-treated BV2 microglia were pre-treated with Src and ERK inhibitors, demonstrating that inhibition of the αMβ2-Src-ERK pathway had attenuated neurotoxicity in dopaminergic neurons [68]. Instead, Hou et al. also investigated NOX2 activation the potential role of complement receptor 3 (CR3) in NOX2 activation of microglia cells after exposure to pesticides. BV2 microglial cells were treated with a combination of paraquat and maneb (P+M), respectively 10 + 0.6 μM concentrations for inducing dopaminergic neurodegeneration. In this way, increased ROS levels and p47^PHOX translocation were observed, demonstrating the NOX2 activation. Conversely, pretreatment with NOX2 inhibitors such as apocynin and diphenyleneiodonium (DPI) in BV2 microglia demonstrating that ROS levels and p47^PHOX translocation were induced by NOX2. Subsequently, the CR3 receptor (integrins) role was studied, thus its inhibition obtained with RGD-pretreatment in BV2 microglia cells. The results decreased both superoxide production and translocation of p47^PHOX, supporting the hypothesis of integrins involvement in NOX2 activation. Consequently, integrin inactivation impacted the Src-ERK pathway. Indeed, both inhibitions of Src and ERK reduced the translocation of p47^PHOX. Instead, immunofluorescence analysis was performed to detect the microglia activation and its morphology using an anti-Iba-1 antibody, a microglia marker. The findings revealed greater cellular bodies, reduction in the number of dopaminergic neurons, and major reactivity to Iba-1 antibody after P+M-treatment in WT than CR3 ⁻/⁻ mice. Based on these results, it was possible to confirm that the C3-Src-ERK pathway could be strongly implicated in NOX2 activation [69]. In compliance with the previous study, P+M treatment (10 + 30 mg/kg, respectively) was also administered to male C57BL/6J mice for inducing dopaminergic neurotoxicity after 2, 4, and 6 weeks of exposure. The dopaminergic neurodegeneration was increased in a time-dependent manner. Moreover, P+M treatment was able to increase α-synuclein oligomer levels. Therefore, the authors administered taurine (150 mg/kg) to mice treated with P+M for 6 weeks, obtaining encouraging results on the improvement of both dopaminergic neuron viability and α-synuclein expression. Furthermore, taurine has contributed to attenuate microglial activation and p47^PHOX and gp91^PHOX levels after P+M exposure in the midbrain of mice. NOX2 could play a role not only in inflammation but would be able to induce microglia polarization through the activation of the NF-Κb pathway. However, taurine was able to mitigate dopaminergic neurodegeneration through the reduction of α-synuclein levels and NOX2 activation after exposure to P + M [70]. As widely known, the α-Syn aggregation was implicated in ND, including PD. Several studies have shown that α-synuclein mutations such as A30P and A53T resulted in increased activation of microglia, resulting in severe loss of dopaminergic neurons. For this reason, Wang et al. experimented with a neuron-glia culture exposed to A29-V40 peptide (α-synuclein mutations) or phorbol 12-myristate 13-acetate (PMA) to reproduce PD. These cultures were co-treated with apocynin to inhibit NOX2, and the co-treatment has contributed to restoring the viability of dopaminergic neurons and their ability DA uptake. Nevertheless, NOX2 inhibition is less efficient in neuronal cells LPS-treated, indicating that other factors could promote neurotoxicity besides NOX2. Further comparison was made between neurons-astrocytes and neuron-glia cultures. Initially, neuron-astrocytes culture treated with A29-V40 did not show any neuronal damage. However, the addition of microglia to the latter culture showed a significant loss of DA uptake. Instead, gp91^{PHOX −}/⁻ microglia were added to the neuron-astrocytes culture treated with A29-V40, demonstrating that this peptide did not compromise dopaminergic neurons. Based on the results obtained, NOX2 activity in microglia promoted A29-V40-induced neurotoxicity. Furthermore, both PMA and A29-V40 treatment had induced translocation of p47^PHOX and p67^PHOX in rat microglia-derived highly aggressively proliferating immortalized (HAPI) cells. A29-V40 peptide could bind gp91^PHOX on the surface and activate increasing superoxide levels. Hence, the superoxide could be converted in H₂O₂ by superoxide dismutase (SOD), then H₂O₂ could cross the membrane and subsequently promote phosphorylation. To assess whether treatment with PMA or A29-V40 could affect p47 ^PHOX and p67^PHOX phosphorylation, the microglia culture was pretreated with SOD and catalase or H₂O₂. In consideration of the results, H₂O₂-induced gp91^{PHOX −}/⁻ microglia promoted the phosphorylation of p47^PHOX and Erk1/2 demonstrating that several factors could affect NOX2 activation. Likewise, microglia activation after A29-V40 peptide injection was evaluated in vivo. As evidence, activated microglia and increased malondialdehyde (MDA) levels were observed in WT mice rather than gp91^{PHOX −}/⁻ mice. These results suggested that α-Syn could both activate NOX2 and promote microglial activation leading to neuronal damage [71]. The effects of simultaneous administration of LPS and α-Syn were evaluated in both in vivo and in vitro, to investigate the neurodegenerative mechanisms in the PD. Zhang et al. observed severe loss of dopaminergic neurons and decreased DA uptake in rat primary midbrain neuron-glia cultures treated with LPS (0.5 ng/mL) + α-Syn (20 nmol/L). It was interesting to observe that the co-treatment induced a less DA uptake in neuron-microglia culture compared to neuron-astroglia culture, thus demonstrating the role of microglia in promoting neurotoxicity. In addition, an increase in ROS levels was observed, therefore the role of NOX2 as a cause of increased ROS levels was investigated. The in vivo experiment was conducted in which NOX2⁺/⁺ and NOX2⁻/⁻ mice were compared to highlight different factors after administration of the LPS (0.2 ug/uL) + syn (0.0125 ug/uL) treatment in SN. Interestingly, a decrease in both dopaminergic levels and DA uptake was found in NOX2⁺/⁺ mice, conversely, both the number of microglia and the ROS levels increased. The same results were found in rat primary midbrain neuron-glia cultures obtained from NOX2⁺/⁺ and NOX2⁻/⁻ mice. However, p47^PHOX and gp91^PHOX were increased in LPS + syn treatment, demonstrating their involvement in NOX2 activation. Instead, DPI pre-treatment (NOX2 inhibitor) improved the viability of dopaminergic neurons, moreover, microglia and ROS were reduced. Furthermore, the combined treatment in microglia increased both mRNA expression and levels of Protein Kinase C (PKC-δ), p38, ERK, and NF-ΚB, thus ex-plaining that NOX2 promoted neuroinflammation [72]. Additionally, Jiang et al., evaluated the pharmacological proprieties of clozapine to reduce the inflammatory state induced by NOX2 activation. The neuroprotective effects of clozapine were investigated in neuron-glia mixed culture cells after induced LPS or 1-methyl-4-phenylpyridinium (MPP⁺) to explore their role in neurotoxic conditions. To evaluate the efficacy of pre-treatment with clozapine N-oxide (CNO) or N-Desmethylclozapine (NDC) for 30 min before LPS (15 ng/mL) or MPP⁺ (0.25 μM) treatment. CNO (0.01 μM) or NDC (0.1 μM) pretreatment demonstrated its neuroprotective effects as an increase in both DA uptake and viability of dopaminergic neurons. These compounds also reduced Iba-1 and CD11b levels in LPS-induced microglia. Of note, LPS treatment induced the release of inflammatory compounds such as superoxide, TNF-α, and NO. However, superoxide release was mitigated in pre-treatment with CNO or NDC, modulating neuroinflammation. To investigate the role of NOX2, dopaminergic neurons obtained in gp91^PHOX-deficient mice were compared with WT (gp91^{PHOX +}/⁺) mice and treated with LPS. Dopaminergic neurons of gp91^PHOX-deficient mice after LPS-treatment showed no difference in DA uptake. Further demonstration of NOX2 involvement after the LPS-treatment was given by the inhibition of p47^PHOX translocation caused by CNO and NDC in microglia HAPI cells. To prove the neuroprotective effects of these compounds, male C57BL/6J mice were treated with MPP⁺ (20 mg/kg) for 6 days and pretreated with CNO or clozapine (1 mg/kg) for 21 days. In line with previous findings, CNO reduced neuronal damage and ameliorated motor deficit MPP+-induced. Therefore, these pharmaceutical compounds have demonstrated their beneficial properties for protection against neurotoxicity useful for the management of NDs such as PD [73]. To illustrate the implications of both microglia and NOX activation in NDs such as PD, a group of researchers conducted an in vivo experiment. In this study, a PD model was induced by intraperitoneal (i.p) injection of LPS (5 mg/kg) in male C57BL/6J (NOX2⁺/⁺) and Cybb (NOX2⁻/⁻) mice which served to promote neurodegeneration and increase oxidative stress. Immunohistochemical analysis was performed through which the loss of dopaminergic neurons was observed in NOX2⁺/⁺ mice 10 months after treatment with LPS in SNpc. In addition, it was observed an upregulated expression of Iba-1 mRNA, a characteristic morphology of activated microglia, and a progressive increase in the number of activated microglia in LPS-treated NOX2⁺/⁺ mice after 10 months. It was seen that the mice of the control group treated with saline only showed an increase in activated microglia 2, 7, and 10 months later, demonstrating that microglia activation could depend on both aging and exposure to toxic compounds. Besides, LPS-treatment in C57BL/6J (NOX2⁺/⁺) mice promoted both mRNA NOX2 expression and release of ROS in a time-dependent manner. Consequently, the authors have investigated the cause of NOX2 induction. For this reason, they used anti-gp91^PHOX, anti-Iba, anti-TH (a marker of the dopaminergic neuron), and anti-GFAP (a marker of astrocyte) antibodies in the brain tissue of hydroethidine-treated mice (10 mg/kg, i.p.) 10 months after a single dose of LPS-injection. The results highlighted that NOX2 and ROS were prevalently co-localized in both microglia and dopaminergic neurons. To confirm the implication of NOX2 in neurodegeneration, DPI (3 mg/kg, ip), was administered to mice that have reduced both ROS and inflammatory cytokines such as TNFα, the monocyte protein chemoattractant-1 (MCP-1), and LPS-induced IL-1β. Hence, NOX2 activation was responsible for microglia activation and, subsequently, for the release of inflammatory factors, thus promoting neurodegeneration [74]. Gao et al. performed an investigation based on the anti-inflammatory proprieties of IL-10 to discover the mechanism for preventing neuroinflammation and neurodegeneration. The investigation focused on the activation mechanism of NLR family pyrin domain containing 3 (NLRP3) and its influence on both pro-caspase-1 and pro-IL-1β cleavage. As observed, pretreatment of IL-10 in both WT and IL10⁻/⁻ microglia culture reduced NLRP3, IL-1β, and caspase-1 mRNA expression. The same results were obtained also in vivo, demonstrating the ability of IL-10 to modulate NLRP3 activation. Recent investigations have shown that ROS were able to induce NLRP3 inflammasome activation. To demonstrate the involvement of ROS in this process, both WT and IL10⁻/⁻ microglia culture LPS-treated were compared. It was examined whether NOX2 could be a possible source of ROS. The translocation of p47^PHOX which is reflected in NOX2 activation was observed in the culture of IL10⁻/⁻ microglia treated with LPS. The neuro-degenerative effects of LPS treatment were attenuated through the administration of NOX2 inhibitors such as DPI or apocynin, which had reduced the cleavage of pro-caspase-1 and pro-IL-1β in LPS-induced IL10⁻/⁻ mice. Therefore, it was possible to deduce that NOX2 activation and the consequent ROS increase were responsible for the activation of the NLPR3 inflammasome. The current study allowed us to highlight that IL-10 was able to suppress LPS-induced ROS-NOX2 and to block the activation of NLPR3, preventing neuroinflammation. Given the above, it was possible to demonstrate that IL-10 exerts neuroprotective and anti-inflammatory actions [75]. Recently, the possible connection between NOX2 and Nucleotide-binding oligomerization domain-containing protein (NOD2) has been investigated aimed to observe their role in the pathogenesis of PD. In this context, male C57BL/6J and NOD2 knockout (NOD2⁻/⁻) mice were treated with 6-hydroxydopamine (6-OHDA) (2 μL) to reproduce the PD model. It was observed a progressive reduction of tyrosine hydroxylase (TH) was observed in both striatum and SN which is reflected in a loss of DA after treatment. Furthermore, the results also showed NOD2 upregulation in WT mice, especially in microglia. Cheng et al. investigated that NOD2 was related to the apoptotic process. Indeed, an increase in both the number of apoptotic cells and apoptotic proteins such as BAX, Bcl-2, caspase-3, and cytochrome C were observed after 6-OHDA or muramyl dipeptide (MDP) treatment in WT. In line with these results, it was observed that activation of both microglia and astroglia were more represented in WT compared to NOD2⁻/⁻ mice, thus explaining neuroinflammation. Likewise, it emerged from the increase in monocyte chemoattractant protein-1 (MCP-1), IL-6, TNF-α, and IL-12p70 in WT mice. In this context, gp91^PHOX expression was observed in NOX2 and ROS production in SN after 6-OHDA injection in a time-dependent manner. In vitro, the results showed an increased NOX2 and ROS levels in BV2 microglial cells in a time-dependent manner. In the same way, an increase of inflammatory compounds such as NOX2, NOD2, inducible nitric oxidase synthase (iNOS) was observed in 6-OHDA induced microglia. To explain the role of NOX2 in NOD2 mediated-neurotoxicity, a viability assay was performed, in which SH-SY5Y and BV2 microglia cells were co-incubated and subsequently treated with 6-OHDA or MDP. This treatment demonstrated that microglial activation induced neurotoxicity. Furthermore, the neuroprotective effects of apocynin have been demonstrated since the latter increased the viability of dopaminergic neurons after 6-OHDA or MDP treatment. Interestingly, this study suggested that both NOX2 and NOD2 promoted neurodegeneration, especially in the PD model [76]. In compliance with the previous study, also Hernandes et al. wanted to examine the effects of 6-OHDA treatment in mice (C57BL/6) and gp91^{PHOX −}/⁻ useful for reproducing the PD model. The first evaluation was made on NOX mRNA expression after the 6-OHDA treatment, in which an upregulated NOX2 expression was found both in SN and striatum. Although no translocation of NOX2 subunits was observed, it was detected Ras-related C3 botulinum toxin substrate 1(Rac1), an important protein involved in NOX activation. Instead, it was interesting to observe that 6-OHDA treatment in the striatum of gp91^PHOX−/⁻ mice significantly reduced the number of dopaminergic neurons, emphasizing NOX2-mediated neurotoxicity. To assess microglia-promoted neurotoxicity, 6-OHDA-induced mice were co-treated with minocycline, a microglial inhibitor with both anti-inflammatory and antimicrobial proprieties. Treatment with minocycline increased the viability of dopaminergic neurons in WT while it was able to promote neurodegeneration in gp91^{PHOX +}/⁺ mice. On other hand, more pro-inflammatory cytokines and chemokines were found in WT mice treated with 6-OHDA but were not increased after co-treatment with minocycline. Furthermore, neurotoxic treatment promoted a significant inflammatory response through a major re-lease of TNF-α in WT compared to gp91^PHOX−/⁻ mice, further supporting NOX2 involvement [77].

Experimental studies have shown that iron deposits have been observed in the SN of patients with PD. Indeed, Fe²⁺ would have the ability to induce neurotoxicity, so Zhang et al. wanted to investigate the implications of that in the pathogenesis of PD. The authors examined the mechanism of microglia activation in rat primary midbrain neuron–microglia–astroglia cultures after Fe²⁺-treatment (5, 25, and 100) μM. In this experiment, exposure to Fe²⁺ was responsible for the loss of dopaminergic neurons and has induced morphological changes in the midbrain neurons as they exhibited smaller cellular bodies. Likewise, Fe²⁺-treatment in rat primary microglia cultures produced an increase in superoxide anion supported by morphologic changes in microglia. Therefore, it was interesting to verify that the same results were observed in primary microglia culture obtained from NOX2⁺/⁺ and NOX2⁻/⁻ mice treated with Fe²⁺. Indeed, an increase in both superoxide levels and the number of activated microglia in NOX2⁺/⁺ was shown after Fe²⁺ exposure. Nevertheless, Fe²⁺-treated NOX2⁻/⁻ mice showed less dopaminergic neuron loss demonstrating their resistance to treatment. Consequently, to confirm the activation of NOX2, this culture was pre-treated with DPI showing a reduction in both the activated microglia and the loss of dopaminergic neurons. Furthermore, the evidence showed that Fe²⁺ had significantly increased both p47^PHOX and gp91^PHOX expression (NOX2 subunits) and their increased expressions suggested that Fe²⁺-induced NOX2 activation. Finally, it was observed an increase in mRNA expression and protein levels of p38, ERK 1/2, and JNK supporting that Fe²⁺ exposure could promote neuroinflammation. Therefore, iron has been shown to induce neurotoxicity by causing both the loss of dopaminergic neurons and the activation of microglia [78].

The substance P (SP) is widely studied for its implication in the pathophysiology of PD. Wang et al., examined the role of SP in both in vivo and in vitro, demonstrating its correlation in neurodegeneration. For this reason, male WT (C57BL/6J), SP-deficient (TAC1⁻/⁻), and SP receptor-deficient (NK1R⁻/⁻) mice were treated with LPS or MPTP to induce neurotoxicity. Immunostaining was performed 10 months after treatment with LPS or MPTP in WT and TAC1⁻/⁻ mice. This treatment highlighted a progressive loss of dopaminergic neurons in WT compared to TAC1⁻/⁻ mice. Furthermore, an increase of TNF-α, iNOS, and MCP-1 expressions were observed in WT mice. To uncover microglia activation in SN after LPS-treatment, Iba-1 and CD11b were found downregulated in the transgenic mouse, suggesting that SP stimulated the neuroinflammation process in WT mice. Instead, the role of NOX2 in enhancing the effects of combined treatment of subpicomolar SP with LPS or MPTP in WT or NOX2-deficient (gp91^{PHOX −}/⁻) glia-neuron cultures obtained from mice was demonstrated. In WT culture treated with SP + LPS or SP + MPP⁺ was observed a significant loss of dopaminergic neurons compared to gp91^{PHOX −}/⁻ culture; thus it was inferred that NOX2 may play a role in promoting neurotoxicity. Moreover, the treatment with LPS + SP was able to promote p47^PHOX and p67^PHOX translocation. In addition, it was also observed that gp91^PHOX was identified as a target of SP further demonstrating NOX2 activation. Obviously, NOX2 activation had generated ROS, which in turn stimulated the activation of MAPK and NF-Κb pathways. As evidence, subpicomolar SP was found to have the ability to promote MAPK and NF-Κb activation in microglia in a NOX2-dependent manner. Pretreatment with DPI reduced ROS production and, subsequently demonstrated that MAPK and NF-Κb pathways were activated following NOX2 activation in microglia after treatment. Therefore, it was possible to define that both SP and NOX2 modulate neuroinflammation, thus playing a role in the viability of dopaminergic neurons [79]. Recently, preclinical observations have focused on the study of glucose metabolism in PD patients. Metabolic dysfunction is caused by alterations in the pentose phosphate pathway, the main pathway used by neuronal cells to metabolize glucose. Several studies have shown an increase in glucose-6-phosphatase-dehydrogenase (G6PD) levels in PD models. For this reason, Tu et al. conducted both in vivo and in vitro experiments to investigate the involvement of G6PD in neurodegeneration. It was performed in an in vivo experiment using both C57BL/6 J mice and A53T mutant α-synuclein transgenic mice. C57BL/6 J mice were administered with a single i.p. or intranigral injection of LPS (5 mg/kg) or subcutaneous injection of MPTP (15 mg/kg) for 6 days. The treatment has shown increased levels in both gp91^PHOX and G6PD were observed in LPS-induced mice resulting in a reduction in both TH and Iba-1 levels. Moreover, upregulation of G6PD was observed mainly in microglia after treatment with LPS or MPTP. Instead, it was found in the midbrain of A53T mutant α-synuclein transgenic mice increased in both gp91^PHOX and G6PD levels compared to WT mice. Subsequently, the enzymatic activity of G6PD and the levels of NADPH in neuron-glia culture treated with saline or LPS were evaluated to demonstrate a possible implication of NOX2 activation. Indeed, neuron-glia culture treated with LPS demonstrated an increase of NADPH levels and G6PD activity after LPS-treatment, so it could be the promoter of NOX2 activation that induces neuroinflammation. To observe that G6PD could affect NOX2 activation, G6PD was inhibited. Indeed, pre-treatment with the G6PD inhibitor reduced ROS increases as well as oxidative stress. Likewise, a reduction in activation of the NF-Kb pathway demonstrating that G6PD inhibition could prevent neuroinflammation. Substantially, the increase of both G6PD and NOX2 in microglia implemented the oxidative stress, the NF-Kb translocation, and subsequent neurodegeneration. Therefore, the inhibition of G6PD limited both oxidative stress and inflammation in the microglia [63].

Exposure to environmental factors such as 6-OHDA, HD, MPTP, or genetic factors such as α-syn, A29-V40 induced the loss of dopaminergic neurons and the reduction of DA uptake, hallmarks of PD. In these experimental models, the studies focused mainly on the activation of microglia which activated NOX2 by promoting neuroinflammation through the release of ROS and pro-inflammatory cytokines (Table 1). To demonstrate the involvement of NOX2, pretreatments with apocynin, DPI were carried out demonstrating that their inhibitory action towards NOX2 would attenuate the release of ROS and reduce the translocation of NOX2 subunits such as p47^PHOX and p67^PHOX.

Table 1

Role of NOX2 in Parkinson’s disease.

Models	Treatments	Results	Ref
	In Vitro and Vivo Experimental Models
Primary microglial cells obtained from whole brains of wild type (WT) or CD11b−/− mice; microglia BV2 cell line	α-Syn (50, 100, and 200) nM; pre-treatment with anti-CD11b antibody (2.5 μg/mL) for 30 min.	The translocation of p47PHOX into the membrane and consequently, ROS production in microglia α-Syn-induced. BV2 microglia α-Syn-induced were pre-treated with anti-CD11b antibody to block CD11b activation for demonstrating its implication in NOX2 activation.	[67]
Adult male SD rats (9 weeks old); Primary mesencephalic neuron-glia, microglia-depleted, and mixed-glia cultures; BV2 microglial cell line;	Intraperitoneal injection of HD (400 mg/kg/die) five times in 1 week for consecutive 5 weeks in rats; HD (1, 4, 8, and 16) mM.	Increased levels of both ROS and p47PHOX levels were observed in HD-treated BV2 microglia, demonstrating NOX2 activation. Instead, microglia cells were pre-treated with apocynin (NOX2 inhibitor), further suggesting that NOX2 was responsible for inducing ROS production. Furthermore, integrin αMβ2 (also known as CD11b) has been shown to be involved in NOX2 activation. Indeed, its inhibition would reduce ROS and p47PHOX translocation.	[68]
BV2 microglial cells	Combination of paraquat and maneb (10 + 0.6) μM.	The co-treatment induced an increase in ROS levels and p47PHOX translocation, evidencing NOX2 activation. Furthermore, the inactivation of the C3 receptor (integrins) decreased the production of superoxide and translocation of p47PHOX, supporting the involvement of integrins in the activation of NOX2. The inactivation of integrins impacted the Src-ERK pathway.	[69]
Male C57BL/6J (NOX2+/+ and NOX2−/−) mice	Intraperitoneal injection of paraquat + maneb (10 + 30) mg/kg 2 times in one week for 6 weeks in mice; Pre-treatment with Taurine (150 mg/kg)	Exposure to paraquat + maneb induced both an increase in NOX2 subunits (p47PHOX and gp91PHOX), as well as α-Syn expression levels which demonstrate neurodegeneration. Moreover, NOX2 could induce microglial polarization. Instead, pretreatment with taurine counteracted neurodegeneration and reduced translocation of NOX2 subunits.	[70]
Male C57BL/6J (NOX2+/+ and NOX2−/−) mice; neuron-glia culture cells obtained from mouse or rat embryos; HAPI cells	Injection of A29-V40 (α-Syn) peptide (5 mg/kg) for 24 h in mice; Treatment with PMA (100 Nm) or LPS (150 EU/mL); Pre-treatment with apocynin (0.25 mM)	PMA and A29-V40 treatment-induced translocation of p47PHOX and p67PHOX and could bind gp91PHOX subunit in mice. The co-treatment with apocynin restored the viability of dopaminergic neurons and their ability to DA uptake. Instead, microglia culture pre-treated with H2O2 in gp91PHOX −/− microglia promoted the phosphorylation of p47PHOX and Erk1/2 demonstrating that several factors could affect NOX2 activation.	[71]
Primary midbrain neuron-glia cultures obtained from brains of SD rats; primary glia cultures obtained from NOX2+/+ and NOX2−/− mice; Male C57BL/6J (NOX2+/+ and NOX2−/−) mice;	Cultures were treated with LPS (0.5 ng/mL) and/or Syn (20 nmol/L) for 8 days LPS (0.2 ug/uL) + syn (0.0125 ug/uL) treatment was administered in mice. Pre-treatment with DPI (0.01 and 0.1) μM	ROS levels, p47PHOX, and gp91PHOX were increased after LPS + syn treatment. Interestingly, a decrease in both dopaminergic levels and DA uptake was found in NOX2+/+ mice. Conversely, both the number of microglia and the ROS levels increased. The same results were found in rat primary midbrain neuron-glia cultures obtained from NOX2+/+ and NOX2−/− mice. Instead, DPI improved the viability of dopaminergic neurons and reduced ROS.	[72]
neuron-glia mixed culture cells of Time-pregnant Fisher F344 rats Wild-type C57BL/6J (gp91phox+/+) and NOX2-deficient (gp91phox−/−) mice	LPS (15 ng/mL) or MPP+ (0.25 μM) treatment MPTP injections (20 mg/kg, s.c.) for 6 days. One day prior to MPTP injection, clozapine or CNO (1 mg/kg, s.c.) was administered twice daily for 21 consecutive days and pre-treatment with DPI (3 mg/kg, i.p)	LPS-treatment in NOX2+/+ mice promoted both mRNA NOX2 expression and release of ROS in a time-dependent manner. Co-treatment with CNO or NDC showed how NOX2 activation influenced DA uptake and release of pro-inflammatory compounds. DPI-Pre-treatment reduced both ROS and inflammatory cytokines such as TNFα, MCP-1, and LPS-induced IL-1β	[73]
B6.129S6-Cybbtm1Din (NOX2−/−) and C57BL/6J 000,664 (NOX2+/+) mice	Intraperitoneal injection of LPS (5 mg/kg).	LPS-treatment in C57BL/6J mice promoted both mRNA NOX2 expression and release of ROS in a time-dependent manner. Instead, pre-treatment with DPI reduced both ROS and inflammatory cytokines such as TNFα, MCP-1and IL-1β demonstrating that NOX2 promoting microglia activation, subsequently the release of inflammatory cytokines.	[74]
Mixed-glia cultures of B10.129P2(B6)-IL-10tm1Cgn/J (IL-10 knockout or IL-10−/−) mice and their WT or IL-10+/+ control mice (C57BL/10J), as well as B6N.129S2-Casp1tm1Flv/J (caspase-1 knockout or CASP-1−/−) mice and their WT (CASP-1+/+) control mice.	Intranigral injection of LPS (3 μg)	NOX2 activation and the consequent increase in ROS were responsible for the activation of NLPR3 inflammasome. The increased levels of IL-10 were able to suppress ROS-NOX2 induced and to block the NLPR3 activation, preventing neuroinflammation.	[75]
Microglial cultures of male C57BL/6J and NOD2 knockout (NOD2−/−) mice. BV2 microglial cells SH-SY5Y cells	Injection of 6-OHDA (2 μL) Injection of MDP (4 μg/μL) in the right striatum	Treatment with MDP or 6-OHDA induced a reduction in DA and an increase in both apoptotic proteins and inflammatory cytokines. Instead, increased NOX2, NOD2, and iNOS promoting neuroinflammation was observed in 6-OHDA-induced microglia.	[76]
Ten week-old male C57BL/6 (gp91PHOX−/−) and WT mice Microglial cultures C57BL/6) and gp91PHOX−/− mice	6-OHDA (10 µg/µL) was unilaterally injected into the right striatum Minocycline (40 mg/kg) was injected(i.p) 7 days before PD induction	Treatment with 6-OHDA in the striatum of gp91PHOX−/− mice reduced the dopaminergic neurons, explaining NOX2-mediated neurotoxicity. Furthermore, co-treatment with minocycline promoted the neurodegeneration and release of TNFα in gp91PHOX+/+ mice, supporting NOX2 activation.	[77]
Microglial cultures of C57 BL/6J (NOX2+/+ and NOX2−/−) mice.	Fe2+-treatment (5, 25, and 100) μM	Treatment with Fe2+ significantly increases both p47PHOX and gp91PHOX expression, suggesting Fe2+-induced NOX2 activation. Moreover, an increase in mRNA expression and protein levels of p38, ERK 1/2, and JNK was observed, therefore Fe2+ exposure could promote neuroinflammation.	[78]
Mesencephalic neuron-glia, microglia-depleted, and microglia-enriched cultures of C57BL/6J, SP-deficient (TAC1−/−), and SP receptor-deficient (NK1R−/−)mice	Treatment with LPS (15 × 106 EU/kg) or MPTP (15 mg/kg) for 6 days	Significant loss of dopaminergic neurons was observed in WT mice treated with SP + LPS or SP + MPP + compared to gp91PHOX−/− culture, thus it was inferred that NOX2 may play a role in promoting neurotoxicity. Moreover, it was observed an increase in the translocation of p47PHOX and p67PHOX as well as of several inflammatory factors such as TNFα, iNOS, and MCP-1, suggesting activation of NOX2. Furthermore, MAPK and NF-Κb pathways were activated by NOX2 in microglia after toxicity-induced.	[79]
Mesencephalic neuron-glia cultures from the ventral mesencephalon of embryonic Fischer 334 rats and also on A53T mutant α-synuclein transgenic mice.	The intranigral and intraperitoneal injection of LPS (5 mg/kg) and subcutaneous injection MPTP (15mg/kg)	Increased levels in both gp91PHOX and G6PD were observed after LPS or MPTP treatment in mice. Neuron-glia culture treated with LPS demonstrated an increase of NADPH levels and G6PD activity, so it could be the promoter of NOX2 activation that induces neuroinflammation. Moreover, an increase of both G6PD and NOX2 in microglia are responsible to implement oxidative stress, the NF-Kb translocation, and subsequent neurodegeneration.	[63]

Open in a separate window

WT: wild type; CD11b: cluster of differentiation molecule 11b; α-Syn: α-Synuclein; ROS: reactive oxygen species; NOX2: NADPH oxidase 2; SD: Sprague Dawley; HD: 2,5-hexanedione; Src: Proto-oncogene tyrosine-protein kinase; ERK: extracellular signal-regulated kinase; DA: dopamine; HAPI: highly aggressively proliferating immortalized; PMA: phorbol 12-myristate 13-acetate; LPS: lipopolysaccharide; H₂O₂: hydrogen peroxide; DPI: diphenyleneiodonium; MPP⁺: 1-methyl-4-phenylpyridinium; MPTP: 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; CNO: clozapine N-oxide; NDC: N-Desmethylclozapine; TNFα: tumor necrosis factor-α; MCP-1: monocyte chemoattractant protein-1; IL-1β: interleukin-1beta; IL-10: interleukin-10; KO: knockdown; casp-1: caspase-1; NLRP3: NLR family pyrin domain containing 3; 6-OHDA: 6-hyroxydopamine; MDP: muramyl dipeptide; NOD2: Nucleotide-binding oligomerization domain-containing protein; Fe²⁺: iron (II); p38: p38 mitogen-activated protein kinase; JNK: c-Jun N-terminal kinase; SP: substance P; iNOS: inducible nitric oxidase synthase; MAPK: mitogen-activated protein kinase; NF-Kb: nuclear factor-Κb; G6PD: glucose-6-phosphatase-dehydrogenase.

Furthermore, clozapine, IL-10, taurine reduced the inflammatory state promoted following exposure to neurotoxic compounds and by-products of oxidizing and inflammatory compounds released by NOX2 activation. Although further investigations are needed, these studies demonstrate that NOX2 in microglia is involved in the signaling pathways underlying the inflammatory response. NOX2 would be involved in the toxicity of dopaminergic neurons therefore its inhibition in microglia could provide neuroprotection in the context of PD.

6.2. NOX2 in Alzheimer’s Disease

AD is a neurological disease characterized by cognitive impairment, progressive loss of memory, and brain function that can lead to death [80]. The hallmarks of AD are the extracellular aggregates of beta-amyloid (Aβ) peptides leading to the formation of amyloid plaques and intracellular accumulation of hyperphosphorylated tau protein, which in turn leads to the formation of neurofibrillary tangles (NFTs). Aβ peptides originate by the enzyme proteolysis of the amyloid precursor protein (APP), this latter is cleaved by β and γ-secretase to generate a peptide of 40 amino acids and a peptide of 42, known as, Aβ_1–40 and Aβ_1–42 [81,82].

NDs such as AD are characterized by neuroinflammation and oxidative stress which can promote disease progression. Consequently, the protein aggregation and neuronal damage can lead to microglial activation that induces ROS formation and neuroinflammation [83].

It has been demonstrated that TGF-β in the hippocampus regulates the microglial activity, it also plays an important role in neuroinflammation by controlling the change of microglia from a protective to harmful state [84,85,86]. Moreover, with aging, it has been shown a further increase of TGF-β levels as well as the activation of pathways related to inflammation, including ERK)/MAPK, p38 MAPK, c-Jun N-terminal kinase (JNK), and PI3K [87], that a decay of mitochondrial and lysosomal activity, as well as an increase of ROS that induces an exacerbation of neuroinflammation [88,89]. Consequently, changes in the CNS related to aging can lead to inflammation, an increase of BBB permeability, and ROS production thus neuronal degeneration such as AD [90]. Noteworthy as microglia have been found in AD patients, both in amyloid plaques and neurofibrillary tangles, where once activated they produce neurotoxic factors including superoxide, nitric oxide, and TNFα [91,92]. The important role of Aβ in plaque formation was explained by the amyloid hypothesis, which reported that microglia-mediated neurotoxicity is a key mechanism in disease progression [92,93]. Based on this, it was demonstrated that Aβ recruits and activates microglia [94]. Indeed, microglia can be activated by fibrillar Aβ through the receptor complex formed both by TLR2, TLR4, and also CD14 co-receptor, to activate intracellular signaling as well as ROS production and phagocytosis [95], by increasing oxidative stress in the brain of AD patients [96]. Furthermore, it has been hypothesized that SARS-CoV-2 infection may stimulate interferons, which also play a role in the development of AD. In fact, the latter would stimulate the microglia by activating a storm of cytokines that would lead to the degeneration of the synapses. [40]. We hypothesize persistent systemic inflammation caused by SARS-CoV-2 may trigger microglial activation, mediated by the massive release of pro-inflammatory cytokines secreted in response to this viral infection, aggravating the neurodegeneration leading to AD.

Geng et al. demonstrated both in vitro and in vivo the role of NOX2 in ROS production, as well as the signaling involved in the regulation of the microglial activation. BV2 microglial cells treated with Aβ₄₂ (0.1–10 µM) for 24 h showed a significant increase of NOX2 and ROS. Moreover, activation of oxidative stress and related pathways including ERK1/2 and p38MAPK was evaluated by Western Blot and immunofluorescence analysis. Consequently, a significant increase in IL-1β and neuroinflammation has also been reported. While, treatment with NOX2 inhibitors including apocynin (20 μM) or NOX2tat (10 μM) significantly reduced ROS production, as well as confirming the important role of NOX2 in oxidative stress.

Additionally, ROS production was completely abolished using Tiron (a ROS scavenger) and also confirming that TNF-α induces microglial proliferation through its action on microglial NOX. To evaluate both the role of NOX2 and ROS in Aβ-induced microgliosis and the effects of inflammation in aging, the same authors performed the study on mesencephalic sections of both WT C57BL/6 and NOX2-Knockout (KO) mice, respectively of young age (3–4 months) and elderly (20–22 months). In vivo results showed a greater Aβ aggregation, as well as high levels of ROS produced by NOX2, and also an increase in microgliosis and IL-1β in elderly WT mice (20–22 months) compared to young WT mice (3–4 months). In the same way, the results of the study were also confirmed in post-mortem mesencephalic samples of young (25–38 years) and elderly (61–85 years) humans without NDs. Indeed, it has been shown that in elderly human brains, in addition to the increase in ROS, also an increase in microglia and expression of microglial NOX2, as well as an increase of IL-1β and activation of ERK1/2 signaling. Therefore, the study performed in all the different models has evaluated and confirmed the important role played by NOX2 in Aβ-mediated microglial activation, as well as the effects of ROS produced by NOX2 activation in aging. Similarly, inhibition or KO of NOX2, as well as the inhibition of ROS production NOX2-induced, has been shown to protect the brain from aging related to oxidative damage, inflammation, and microglial alterations [7].

Therefore, studying and realizing compounds capable of both avoiding the assembly of NOX2 and stabilizing free radicals, could be a useful strategy to counteract the oxidative stress as well as the neurodegeneration that characterizes the AD. Indeed, it has been shown that apocynin, a NOX2 inhibitor it was capable to prevent the translocation of the p47^PHOX and p67^PHOX subunits, as well as reducing the damage produced by oxidative stress. In addition, it was reported that apocynin can also inhibit enzymes such as ROCK, which is involved in the adhesion, morphology, and motility of cells [97].

ROCK is associated with NOX; indeed, it has been shown that the phosphorylation of ROCK-I through Rac, induces the assembly of NOX2. As they are involved in the growth and retraction of neurites, both the ROCK-I and NOX2 pathways are interesting for investigating neurological diseases including AD. Indeed, it has been shown that the overexpression of NOX2 and ROCK-I in microglia can contribute to the progression of neuronal damage, as well as to the development of the disease. Therefore, both inhibition of ROCK-I and NOX2 pathways may be a useful strategy for counteracting the neuroinflammation in neurological diseases [98].

Alokam et al. evaluated the biological activity of 18 new molecules potentially capable of inhibiting ROCK-I and NOX2. The authors with computational methods combined between molecular docking and pharmacophore have realized ROCK-I and NOX2 inhibitors through information obtained by ligand-bound protein. The pharmacophore model for ROCK-I was performed using fasudil at a dose approximately of (1µM) which, thanks to its crystalline structure, proves to be a potent inhibitor of ROCK in human neuroblastoma cells. The results of the study demonstrated that in human neuroblastoma cells the compound 3 was the most active compared to fasudil, indeed it has reduced the inflammation measured by the decrease of pro-inflammatory mediators including IL-6, IL-1β, and TNFα. Therefore, the study showed that the realizing of ROCK-I and NOX2 inhibitory compounds could be useful for the treatment of neurological diseases such as AD [99].

The importance of NOX2, as well as the confirmation of its role in oxidative stress, has also been demonstrated by Wilkinson et al., in B6-R1.40 transgenic mice and human THP-1 monocytes. The authors showed that chronic treatment for 9 months of ~5 g/day/animal of rodent food enriched with ibuprofen in a final dosage of 62.5 mg/kg/day has reduced Aβ aggregation in the brain in mice, evaluated by thioflavin S positivity. Moreover, the chronic administration of ibuprofen has led to reducing both microglial activation and oxidative stress measured by the decrease of 4-hydroxynonenal levels. Although there are several sources of microglial ROS including NOX, cyclooxygenase (COX), xanthine oxidase, and lipoxygenase, the primary source of ROS has been shown to be attributable to Aβ-induced NOX2 activation in the brains of AD patients. To evaluate the effects of ibuprofen on NOX2, the authors demonstrated that the pretreatment with racemic ibuprofen for 1 h before fibrillar treatment Aβ_22–35 (60 μM) in primary murine microglia reduced the intracellular production of superoxide Aβ-induced. Furthermore, Wilkinson et al., also performed the study on human THP-1 monocytes where they observed the same response obtained in primary microglia. As both function and assembly of NOX2 depend on Vav phosphorylation induced by Aβ aggregates, ibuprofen could inhibit Vav phosphorylation and hence NOX2, as demonstrated by ibuprofen pretreatment in human THP-1 monocytes. Therefore, these results suggest that ibuprofen could inhibit NOX activation and ROS production, thus leading to the reduction of damage oxidative stress-induced [100].

The cytotoxicity of Aβ aggregates, as well as the increase in the ROS production and the mechanism related to NOX2 activation, was evaluated by Carrano et al., in endothelial cells of the brain. The authors performed the study on brain samples from patients with capillary cerebral amyloid angiopathy, a disease that occurs in most AD patients characterized by Aβ aggregates in the cerebral arteries and capillaries. As for how the cerebral amyloid angiopathy is related to AD not yet fully known, the authors hypothesized that Aβ aggregation in capillaries may involve the integrity and functionality of BBB, dysfunction of which may represent a pathological event associated with NDs. Indeed, it has been suggested that ROS, especially those produced by NOX2 in response to Aβ aggregation, can modulate the integrity of BBB through the interruption of tight junctions. In this regard, the presence of Aβ aggregation in the capillaries was related both to an increase in the microglial NOX-2 activity and ROS production which in turn led to the loss of occludins and claudins, as well as to the loss of cellular adhesion. Carrano et al., performed an in vitro study using a human cerebral microvascular endothelial cell line (hCMEC/D3) which was administered synthetic Aβ_1–42 at different concentrations. Indeed, it was shown that the treatment with Aβ (10 μM) for 24 h induced cell death. While the treatment with Aβ_1–42 (1 μM) for 24 h led to a reduction in mitochondrial activity as well as an increase in H₂O₂ in a dose-dependent manner. Consequently, the presence of H₂O₂ led the authors to investigate the involvement of NOX2 and xanthine oxidase in the ROS production Aβ-mediated. Indeed, before treatment with Aβ_1–42, the hCMEC/D3 cells were pretreated at concentrations (10 nM and 100 nM) for 24 h with different NOX2 inhibitors including DPI and allopurinol. The results of the study showed that all antioxidants were able to preserve hCMEC/D3 cells from cytotoxicity Aβ_1–42-induced. Moreover, the allopurinol pretreatment 2 h before of inducing model restored both occludin and claudin expression levels, as well as confirming that ROS production Aβ-mediated is also involved in the alterations of tight junctions. Further confirmations of the cytotoxic effects of Aβ in capillaries were shown by upregulation of Receptor for advanced glycation end products (RAGE), which can transport Aβ across BBB. Additionally, to carrying Aβ into the brain, RAGE through NOX-2 activation can induce ROS production. Moreover, it was shown that RAGE blockade by anti-RAGE blocking antibody has prevented the reduction of cell viability Aβ-induced. Therefore, the study results demonstrated that NOX2 inhibitors, exogenous antioxidants, and RAGE inhibitors could be useful to reduce the cytotoxicity Aβ-induced and neuronal damage related to its [101].

Moreover, increased levels of C-X-C motif chemokine ligand 1 (CXCL1), the latter widely expressed in the CNS and activated when it binds to the CXCR2 receptor have also been demonstrated in AD patients. CXCL1 once bound to its receptor can activate PI3K/Akt and ERK/MAPK signaling, thus contributing to inflammatory responses and cell cycle. Shang et al., both in vivo and in vitro investigated the effects of CXCL1 on neural stem cells in the subventricular area. The authors showed elevated CXCL1 levels in CSF APP/PS1 mice of 4,8,12 months compared to the WT control mice of the same-aged. Likewise, a greater increase of CXCL1 levels was reported in 12-month-old APP/PS1 mice compared to 4-month-old APP/PS1 mice and also compared to wild-type mice. Therefore, the results obtained suggested that CXCL1 levels in CSF could be disease markers in these animals. CXCL1 is expressed in microglia, macrophages, oligodendrocytes, and neurons. To simulate and investigate the activation of macrophages and microglia during the neuroinflammation, the authors used respectively RAW264.7 and BV2 cells treated with LPS or Adenosine triphosphate (ATP) (1 mM or 5 mM) alone or combined. Indeed, it has been shown that in RAW264.7 cells the LPS + ATP treatment led to a significant increase in CXCL1 compared to RAW264.7 cells treated with LPS alone. Moreover, it was also observed that treatment with LPS and ATP (1 mM or 5 mM) in BV2 microglial cells had no significant effect on CXCL1 synthesis and secretion. Therefore, these results suggested that macrophages may be a major source of CXCL1 in APP/PS1 mice. Additionally, CXCL1 has also been reported to promote both proliferation and differentiation of neural stem cells. Indeed, it has been shown that infusion of CXCL1 (200 ng per mouse) into the lateral ventricle led to a significant increase in the cell proliferation assessed by nestin positivity. The same results were also confirmed in vitro, in neural stem cells treated with CXCL1 (100 ng/mL) for 24 h. Moreover, CXCL1 treatment has also been shown to inhibit the differentiation of neural stem cells into astrocytes, but not their differentiation into neurons. To understand how CXCL1 exerts these effects, Shang et al., demonstrated that CXCL1 promotes the proliferation of neuronal stem cells through the production of ROS as well as by the activation of PI3K/Akt signaling. Furthermore, it was shown that treatment with CXCL1 (100 ng/mL) for 24 h led to an increase in both ROS and NOX2. Confirmation of NOX2 involvement was performed in vivo, indeed treatment with apocynin (30 nmol per mouse) before the CXCL1 (200 ng per mouse) administration has inhibited CXCL1 in promoting the proliferation of neural stem cells. In conclusion, the study showed that in APP/PS1 mice CXCL1 originating from macrophages can favor the proliferation of neural stem cells in the subventricular area through the generation of ROS [102].

In vitro and in vivo studies have shown that Aβ peptides can promote NOX2 activation, therefore a direct correlation with the deposition of Aβ, dependent on both aging and NOX2 activity (Table 2). Preclinical studies demonstrated that exposure of microglial cultures with Aβ peptides resulted in NOX2 activation with a consequent increase in ROS, ERK 1/2, and p38 MAPK thus promoting oxidative stress and neuroinflammation. Furthermore, also ROCK-1 would be involved in NOX2 activation, therefore it could contribute to the progression of neuronal damage, as well as to the development of AD. In experimental AD models, the accumulation of Aβ activated RAGE, stimulating the NOX2 activation, hence the production of ROS. RAGE would facilitate both the transport of Aβ through the BBB and the pathological accumulation of Aβ in the brain parenchyma. Therefore, the activation of RAGE would induce alterations in BBB permeability and functionality as well as neurodegeneration and cognitive decline. Therefore, treatment with new molecules such as NOX2 and RAGE inhibitors could serve as a basis for the development of promising therapies for pathological conditions associated with NOX2 activation including dementia.

The authors would like to thank the Ministry of Health, Italy.