Synergic Therapeutic Potential of PEA-Um Treatment and NAAA Enzyme Silencing In the Management of Neuroinflammation

Inflammation is a key element in the pathobiology of neurodegenerative diseases and sees the involvement of different neuronal and non-neuronal cells as players able to respond to inflammatory signals of immune origin. Palmitoylethanolamide (PEA) is an endogenous potent anti-inflammatory agent, in which activity is regulated by N-acylethanolamine acid amidase (NAAA), that hydrolyzes saturated or monounsaturated fatty acid ethanolamides, such as PEA. In this research, an in vitro study was performed on different neuronal (SH-SY5Y) and non-neuronal cell lines (C6, BV-2, and Mo3.13) subjected to NAAA enzyme silencing and treated with PEA ultra-micronized (PEA-um) (1, 3, and 10 μM) to increase the amount of endogenous PEA available for counteract neuroinflammation provoked by stimulation with lipopolysaccharide (LPS) (1 μg/mL) and interferon gamma (INF-γ )(100 U/mL). Cell viability was performed by MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) staining, suggesting a protective effect of PEA-um (3 and 10 μM) on all cell lines studied. Western Blot analysis for inflammatory markers (Inducible nitric oxide synthase (iNOS) and cyclooxygenase 2 (COX-2)) was carried out in control and NAAA-silenced cells, highlighting how the concomitant treatment of the neuronal and non-neuronal cells with PEA-um after NAAA genic downregulation is satisfactory to counteract neuroinflammation. These in vitro findings support the protective role of endogenous PEA availability in the neuronal field, bringing interesting information for a translational point of view.


Introduction
N-acylethanolamine acid amidase (NAAA) is a specific hydrolases that degrades saturated or monounsaturated fatty acid ethanolamides, such as palmitoylethanolamide (PEA) and oleoylethanolamide (OEA) [1]. However, OEA and, to a lesser extent, anandamide (AEA) are hydrolyzed by NAAA, N-acylethanol-amines (NAEs) catabolic enzymes are currently viewed as potential therapeutic targets in which inhibition may increase tissue levels of PEA [2]. The recent researches in strategies to increase the endogenous levels of N-acylethanol-amines (NAEs) has led to   The potential of PEA-um in mediating neuroinflammation is known [4], but the in vitro role of PEA-um in modulating the action of neuronal and non-neuronal cells involved in inflammatory processes, such as astrocytes and microglia, has never been investigated in detail. In this context, to induce inflammatory stimuli, all cell lines were stimulated with LPS/IFN γ and at the same time treated with different concentrations (1, 3, and 10 μM) of PEA-um. The results obtained showed a significant protective effect, against LPS/IFN γ induced inflammatory process, in cells treated with PEA-um 3 and 10 μM, but not at the lowest concentration 1 μM (Figure 2A-D).   Anti-inflammatory effects of PEA-um in both neuronal and non-neuronal cell lines following LPS/IFN γ-stimulation. Cell vitality was assessed following 24 h treatment with LPS (1 µg/mL) and INF-γ (100 U/mL) and different concentrations (1, 3, and 10 µM) of PEA-um. PEA-um at 3 µM and 10 µM significantly locked damage caused by LPS/INF-γ in all cell lines (A-D). Particularly, in BV-2 microglial cells, the effects of PEA-um 3 µM were almost comparable to those of treatments at 10 µM (C). PEA-um 1 µM is unable to protect against LPS/INF-γ induced damage, in all cells object of the study (A-D). Data are representative of at least three independent experiments. *** p < 0.001 versus CTR # p < 0.05, ## p < 0.01 and ### p < 0.001 versus LPS/ INF-γ.

Effects of NAAA Silencing in Neuronal and Non-Neuronal Cells
NAAA enzymes, in terms of substrates, when assessed in vitro, has a strong preference for saturated NAEs, with PEA being the preferred substrate [15]. About that, an increasing number of studies have described the synthesis and pharmacological characterization of NAAA inhibitors leading to the development of potent and stable inhibitors that enable to study the effects of NAAA inhibition in preclinical disease models, notably in the context of inflammation [16]. However, though pharmacological blockade of receptors constitutes a reliable and widely used method, silencing of NAAA constitutes a matter of interesting perspective to completely rule out their implication in the protective effects produced by the treatments. For this reason, quantitative measurements of NAAA mRNA expression was performed in all cell lines to evaluate the success of NAAA silencing. The results indicated that the relative mRNA levels of NAAA enzyme were significantly reduced in all cell lines subjects of study: C6, SHSY-5Y, BV-2, and Mo3.13 cell lines (respectively, Figure 3A-D), suggesting a significant reduction in the amount of endogenous PEA degraded and therefore an increase in available endogenous PEA.

Effects of NAAA Silencing in Neuronal and Non-Neuronal Cells
NAAA enzymes, in terms of substrates, when assessed in vitro, has a strong preference for saturated NAEs, with PEA being the preferred substrate [15]. About that, an increasing number of studies have described the synthesis and pharmacological characterization of NAAA inhibitors leading to the development of potent and stable inhibitors that enable to study the effects of NAAA inhibition in preclinical disease models, notably in the context of inflammation [16]. However, though pharmacological blockade of receptors constitutes a reliable and widely used method, silencing of NAAA constitutes a matter of interesting perspective to completely rule out their implication in the protective effects produced by the treatments. For this reason, quantitative measurements of NAAA mRNA expression was performed in all cell lines to evaluate the success of NAAA silencing. The results indicated that the relative mRNA levels of NAAA enzyme were significantly reduced in all cell lines subjects of study: C6, SHSY-5Y, BV-2, and Mo3.13 cell lines (respectively, Figure 3A-D), suggesting a significant reduction in the amount of endogenous PEA degraded and therefore an increase in available endogenous PEA.

Synergistic Role of PEA-um Treatment And NAAA Silencing in Countering In Vitro Neuroinflammation
NAEs inactivation (AEA, OEA, and PEA) occurs essentially by enzymatic hydrolysis by NAAA and it is known that PEA-um administration reduces this degradation process by substrate competition [17]. Moreover, it is also believable that the increased levels of endogenous PEA upon PEA-um administration have an impact on the expression of NAAA hydrolytic enzyme [18]. In light of this, to corroborate that the anti-neuroinflammatory effects of PEA-um are enhanced by NAAA

Synergistic Role of PEA-um Treatment And NAAA Silencing in Countering In Vitro Neuroinflammation
NAEs inactivation (AEA, OEA, and PEA) occurs essentially by enzymatic hydrolysis by NAAA and it is known that PEA-um administration reduces this degradation process by substrate competition [17]. Moreover, it is also believable that the increased levels of endogenous PEA upon PEA-um administration have an impact on the expression of NAAA hydrolytic enzyme [18]. In light of this, to corroborate that the anti-neuroinflammatory effects of PEA-um are enhanced by NAAA silencing, we evaluated the expression of inflammatory markers as iNOS and COX-2 on C6, SHSY-5Y, BV-2, and Mo3.13 cells following NAAA siRNA knockdown and treatment with PEA-um.
In rat C6 glioma cells, treatment with PEA-um (10 µM) significantly reduced iNOS and COX-2 expression following damage induced by LPS/IFN γ ( Figure 4A,B), but the protective effects of PEA-um were enhanced in LPS/IFN γ-stimulated cells in which NAAA enzyme was muted ( Figure 4A,B). The same condition was observed in human SH-SY5Y, with a remarkable PEA-um protection in LPS/IFN γ-stimulated cells ( Figure 5A,B), but, in this case, iNOS and COX-2 expression was even more reduced in treated cells in which NAAA was silenced, compared to not NAAA-silenced SHSY-5Y ( Figure 5A,B). Activated microglia is able to secrete a number of pro-inflammatory and neurotoxic factors which cause neuronal damage [19]; in this experiment, we confirmed the expression of inflammatory markers iNOS and COX-2 in microglia BV-2 cells following LPS/IFN γ-activation ( Figure 6A,B), with a notably reduction when the cells were treated with PEA-um (10 µM); an enhancement of the effect in the PEA-um treated cells was obtained with the NAAA-siRNA knockdown, compared to cells only treated or only NAAA-silenced ( Figure 6A,B). Furthermore, also in Mo3.13 oligodendrocytes cell lines, COX-2 expression, more than iNOS protein levels, was significantly reduced in PEA-um (10 µM)-treated cells, compared to only stimulated cells ( Figure 7A,B); the NAAA-siRNA knockdown significantly reduced the expression of both inflammatory markers, with an enhanced effect in PEA-um treated NAAA-silenced cells ( Figure 7A,B). silencing, we evaluated the expression of inflammatory markers as iNOS and COX-2 on C6, SHSY-5Y, BV-2, and Mo3.13 cells following NAAA siRNA knockdown and treatment with PEA-um. In rat C6 glioma cells, treatment with PEA-um (10 μM) significantly reduced iNOS and COX-2 expression following damage induced by LPS/IFN γ ( Figure 4A,B), but the protective effects of PEAum were enhanced in LPS/IFN γ-stimulated cells in which NAAA enzyme was muted ( Figure 4A,B). The same condition was observed in human SH-SY5Y, with a remarkable PEA-um protection in LPS/IFN γ-stimulated cells ( Figure 5A,B), but, in this case, iNOS and COX-2 expression was even more reduced in treated cells in which NAAA was silenced, compared to not NAAA-silenced SHSY-5Y ( Figure 5A,B). Activated microglia is able to secrete a number of pro-inflammatory and neurotoxic factors which cause neuronal damage [19]; in this experiment, we confirmed the expression of inflammatory markers iNOS and COX-2 in microglia BV-2 cells following LPS/IFN γ-activation ( Figure 6A,B), with a notably reduction when the cells were treated with PEA-um (10 μM); an enhancement of the effect in the PEA-um treated cells was obtained with the NAAA-siRNA knockdown, compared to cells only treated or only NAAA-silenced ( Figure 6A,B). Furthermore, also in Mo3.13 oligodendrocytes cell lines, COX-2 expression, more than iNOS protein levels, was significantly reduced in PEA-um (10 μM)-treated cells, compared to only stimulated cells ( Figure  7A,B); the NAAA-siRNA knockdown significantly reduced the expression of both inflammatory markers, with an enhanced effect in PEA-um treated NAAA-silenced cells ( Figure 7A,B).  silencing, we evaluated the expression of inflammatory markers as iNOS and COX-2 on C6, SHSY-5Y, BV-2, and Mo3.13 cells following NAAA siRNA knockdown and treatment with PEA-um. In rat C6 glioma cells, treatment with PEA-um (10 μM) significantly reduced iNOS and COX-2 expression following damage induced by LPS/IFN γ ( Figure 4A,B), but the protective effects of PEAum were enhanced in LPS/IFN γ-stimulated cells in which NAAA enzyme was muted ( Figure 4A,B). The same condition was observed in human SH-SY5Y, with a remarkable PEA-um protection in LPS/IFN γ-stimulated cells ( Figure 5A,B), but, in this case, iNOS and COX-2 expression was even more reduced in treated cells in which NAAA was silenced, compared to not NAAA-silenced SHSY-5Y ( Figure 5A,B). Activated microglia is able to secrete a number of pro-inflammatory and neurotoxic factors which cause neuronal damage [19]; in this experiment, we confirmed the expression of inflammatory markers iNOS and COX-2 in microglia BV-2 cells following LPS/IFN γ-activation ( Figure 6A,B), with a notably reduction when the cells were treated with PEA-um (10 μM); an enhancement of the effect in the PEA-um treated cells was obtained with the NAAA-siRNA knockdown, compared to cells only treated or only NAAA-silenced ( Figure 6A,B). Furthermore, also in Mo3.13 oligodendrocytes cell lines, COX-2 expression, more than iNOS protein levels, was significantly reduced in PEA-um (10 μM)-treated cells, compared to only stimulated cells ( Figure  7A,B); the NAAA-siRNA knockdown significantly reduced the expression of both inflammatory markers, with an enhanced effect in PEA-um treated NAAA-silenced cells ( Figure 7A,B).

Discussion
Inflammation is a key element in the pathobiology of chronic pain, neurodegenerative diseases, stroke, and neuropsychiatric disorders; glial cells, key players in nervous system disorders, respond to inflammatory signals of immune origin [4]. Communication between the nervous system and immune system represents a fundamental principle underlying neuroinflammation; particularly, immune cell-derived inflammatory molecules are critical for regulation of responses to inflammation; these mediators can originate from various sources.
Neuroinflammation responds to a program of resolution that involves lipid mediators endowed with the capacity to switch off inflammation process. These naturally occurring lipid signaling molecules include the NAEs, specifically PEA. PEA may play a role in maintaining cellular homeostasis, in mast cell-mediated models of neurogenic inflammation and neuropathic pain and it is neuroprotective in in vivo models of stroke, spinal cord injury, traumatic brain injury, and neurodegenerative diseases [20,21]. Moreover, PEA in ultra-micronized form, known as PEA-um, shows superior efficacy in inflammatory models when compared to naïve PEA. Marked properties in peripheral inflammation models emerge by PEA administration, correlated to high effectiveness in a number of neurodegenerative disorders, including AD, Parkinson's disease, and multiple sclerosis [22,23].
NAEs are hydrolyzed to the corresponding fatty acid and ethanolamine by specific hydrolase and amidase, respectively, FAAH and NAAA [24,25]. In contrast to FAAH, NAAA hydrolyzes NAEs having less than 18 carbon atoms, i.e., PEA [26]. Inhibiting the enzymatic degradation of PEA by targeting NAAA, in principle, represents another route in treatment of neuroinflammation; for this, a number of selective NAAA inhibitors have been described including systemically active compounds which are able to modulate responses induced by inflammatory stimuli in vivo and in vitro [2,27], although the validity of this approach remains to be demonstrated.
PEA is produced on demand and its catabolic enzymes are proposed to modulate substrate availability; so, a modulatory approach intended for inhibit PEA degradation and make the exogenous PEA more available through ultra-micronization would maximize accessibility of the NAE's component molecules in biological system.
Given the complex nature of non-neuronal cellular involvement in inflammation-associated pathologies across the CNS and PNS, viewing neuroinflammation in the context of microglia astrocyte or mast cell involvement allows us to appreciate the homotypic and heterotypic cell-cell interactions that are an integral part of the inflammation process [28]. On this basis, the aim of the research was to perform an in vitro study of neuroinflammation on different types of neuronal and non-neuronal cell lines: C6 glioma cells, SHSY-5Y neuroblastoma cell line, BV-2 microglia cells, and Mo3.13 oligodendrocytes, together silencing NAAA enzyme and treating with PEA-um, thus providing a greater amount of PEA needed to counteract inflammation triggered at cellular level.
Neuroinflammation is linked to synapse loss and cognitive decline both in humans and in pre-clinical models, but important questions remain about the cellular mechanisms that existing experimental systems cannot easily address [29]. Lipopolysaccharide (LPS), a cell-wall immunostimulatory component of gram-negative bacteria, once activated, produces proinflammatory cytokines that are key mediators of the neuro-inflammatory process [29][30][31] The immunoregulatory properties of PEA, as an anti-inflammatory agent, are better known in vivo studies [32]. In the first time, in this work, the protective activity of PEA-um, at three different concentrations, was evaluated in counteracting the in vitro inflammation induced in the different cell lines stimulated with LPS and IFN γ. This study demonstrated, for the first time, that PEA-um, at the highest concentrations of 3 and 10 µM, significantly counteracted LPS-induced inflammation, as well as increased cell viability in glioma and glial cells LPS/IFN γ-damaged.
PEA availability into the cells depends on its degradation and deactivation, that preferentially happens by a lysosomal enzyme NAAA [14]. NAAA hydrolyzes PEA to palmitic acid and ethanolamine, with much greater efficacy and selectivity than FAAH, so inhibition of NAAA is a good strategy to target inflammation by modulating the tissue levels of PEA. However, recently NAAA inhibitors have begun to emerge, and most of them suffer from the chemical and biological unstable properties, which restrict functions of NAAA inhibition [33]. For the first time, in this research, the inhibition of NAAA expression was performed through genetic silencing of the enzyme in C6, SHSY-5Y, microglia, and oligodendrocytes cells. Using this novel, more potent, and selective method, it was possible to unravel in detail the role of NAAA activity and PEA disposability in mediating the inflammatory process.
Although the inflammatory response is a defense mechanism against injury, sustained inflammation is a pathological condition, which determines an over-expression of several pro-inflammatory enzymes and reactive species; particularly, iNOS being up-regulated during the inflammatory process [34]. Other important mediators over-expressed during inflammation are prostaglandins (PGs), bioactive signaling molecules derived from cyclooxygenase (COX) [35].
The up-regulation of iNOS and COX-2 during inflammation is controlled by the pro-inflammatory transcription factor NF-κB [34]. PEA prevents NF-κB nuclear translocation, confirming the involvement of this transcriptional factor as interesting tools for inflammatory control [36]. For the first time, in this study, we demonstrated that treatment with PEA-um, at the highest concentration of 10 µM, significantly reduced inflammation as showed by a decreased expression in iNOS and COX-2 protein levels, in glioma and in neuroblastoma cells, as well as in microglia and oligodendrocytes, compared to cells stimulated but not treated. Main actors in inflammatory scenario, together with neuronal cells, are astroglia and microglia; upon activation, also called "gliosis," these cells get involved in the production of cytokines and chemokines, which maintain and enhance the inflammatory condition [37]. The biological meaning if these experiments was how PEA-um is able to modulate inflammatory markers in the same way in different cell types (neuronal and non-neuronal).
Moreover, to understand if the treatment with exogenous PEA-um could enhance the PEA endogenous availability through the silencing of the degrading enzyme NAAA, thus regulating the expression of inflammatory markers, all cell lines were subjected to a NAAA siRNA-knockdown. A significant reduction in iNOS and COX-2 expression was obtained in NAAA-silenced cells, constantly in the different cell lines, compared to control. The amazing discovery, however, was to observe a significant reduction in inflammatory markers in NAAA-silenced cells treated exogenously with PEA-um, suggesting as the lack of NAAA empowered PEA protective effect, in diversified cellular context: neuronal and non-neuronal.
The data obtained, highlighted as in vitro silencing of intracellular NAAA, associated to a supply of exogenous PEA, made more available in the ultra-micronized form (PEA-um), could represent a novel therapy to control cellular neuroinflammation.

SH-SY5Y Cell Line
SH-SY5Y cells, obtained from American Type Culture Collection (ATCC CLR-2266, ATCC), are a cloned subline of SK-N-SH cells originally established from a bone marrow biopsy of a neuroblastoma patient [38]. SH-SY5Y neuroblastoma cells can be differentiated into neuron-like cells displaying morphological and biochemical features of mature neurons. Neuronal differentiation was achieved using retinoic acid (RA) as previously described [39]; differentiation of cell lines into neuronal-like cell lines is required to mimic the intracellular environment of a neuronal cell. SH-SY5Y cells were grown to monolayer in 75-cm 2 flasks containing DMEM and Ham's F12, modified with 2 mmL-glutamine, 1.0 mM sodium pyruvate, and supplemented with FBS to 10%, streptomycin 50 µg/mL. SH-SY5Y cells were maintained at 37 • C and 5% CO 2 . Cells were passaged at confluence using a solution of 0.025% trypsin and 0.01% EDTA.

BV-2 Cell Line
BV-2 mouse cell line is a brain microglial cell line from C57BL/6, obtained from Cell bank IRCCS AOU San Martino IST (Genova, Italy) Interlab Cell Line Collection (ICLC) (ICLC ATL03001). This cell line exhibits morphological, phenotypical, and functional properties of microglial cells; secretory and effector functions, also ascribed to microglia, have been extensively characterized in this cell line [40]. BV-2 cells were maintained in medium consisting of equal volumes of minimum essential medium (MEM) and DMEM supplemented with 5% heat inactivated fetal calf serum (FCS), 100 IU/mL penicillin G, and 100 µg/mL streptomycin in a humidified 5% CO2 atmosphere at 37 • C. Cells were passaged at confluence using a solution of 0.025% trypsin and 0.01% EDTA.

Cell Viability Assay (MTT Assay)
The cellular viability of different cell lines was assessed using a mitochondria-dependent dye for live cells (tetrazolium dye; MTT) to formazan. Cultures were treated with three different concentrations (1, 3, 10 µM) of PEA-um for 24 h and then incubated at 37 • C with MTT (0.2 mg/mL) for 1 h. The medium was removed and the cells lysed with dimethyl sulfoxide (DMSO) (100 µL), as previously described [41]. The extent of reduction in MTT to formazan was quantified by measurement of optical density at 550 nm with a microplate reader.

Statistical Analysis
Data are presented as mean ± SEM values of three independent determinations. All experiments were done at least three times, each time with three or more independent observations. Statistical analysis was performed by One-way ANOVA test, followed by multiple comparisons performed by Bonferroni's test. A value of p < 0.05 was considered significant.