N-Palmitoylethanolamide-Oxazoline Protects against Middle Cerebral Artery Occlusion Injury in Diabetic Rats by Regulating the SIRT1 Pathway

Diabetes causes various macrovascular and microvascular alterations, often culminating in major clinical complications (first of all, stroke) that lack an effective therapeutic intervention. N-palmitoylethanolamide-oxazoline (PEA-OXA) possesses anti-inflammatory and potent neuroprotective effects. Although recent studies have explained the neuroprotective properties of PEA-OXA, nothing is known about its effects in treating cerebral ischemia. Methods: Focal cerebral ischemia was induced by transient middle cerebral artery occlusion (MCAo) in the right hemisphere. Middle cerebral artery (MCA) occlusion was provided by introducing a 4–0 nylon monofilament (Ethilon; Johnson & Johnson, Somerville, NJ, USA) precoated with silicone via the external carotid artery into the internal carotid artery to occlude the MCA. Results: A neurological severity score and infarct volumes were carried out to assess the neuroprotective effects of PEA-OXA. Moreover, we observed PEA-OXA-mediated improvements in tissue histology shown by a reduction in lesion size and an improvement in apoptosis level (assessed by caspases, Bax, and Bcl-2 modulation and a TUNEL assay), which further supported the efficacy of PEA-OXA therapy. We also found that PEA-OXA treatment was able to reduce mast cell degranulation and reduce the MCAo-induced expression of NF-κB pathways, cytokines, and neurotrophic factors. Conclusions: based on these findings, we propose that PEA-OXA could be useful in decreasing the risk of impairment or improving function in ischemia/reperfusion brain injury-related disorders.


Introduction
Diabetes mellitus (DM) is a metabolic disorder connected with chronic hyperglycemia, which has been identified as enhancing systemic oxidative stress, inclining patients toward diabetic complications. World Health Organization (WHO) data show that around 386 million people in the world are now affected by diabetes, which is a main risk factor for atherosclerotic diseases, such as acute brain ischemia [1,2]. Additionally, it has been well demonstrated that diabetic patients have a higher This was monitored at baseline (10 min before), at 10 and 60 min during ischemia, and 5 min after ischemia (A). TTC staining of brain sections was performed 24 h after MCAo. Tissues from vehicle-treated animals displayed the presence of an unstained area (B,D) compared to the sham group (B,C). PEA-OXA administration (B,F) reduced this infarcted area more effectively than did the PEA treatment (B,E). A histological analysis of brain samples from vehicle-treated rats revealed the loss of neurons (H,M) compared to the sham group (G,M), which was ameliorated by PEA administration (I,M). PEA-OXA administration displayed more protective effects compared to the PEA treatment (L,M). For the histology, a 20× magnification is shown (50-µm scale bar). A p-value of less than 0.05 was considered significant: * p < 0.05 versus sham; ° p < 0.05 versus vehicle; ** p < 0.01 versus sham; °° p < 0.01 versus vehicle; *** p < 0.001 versus sham; °°° p < 0.001 versus vehicle; # p < 0.05 versus PEA; ## p < 0.01 versus PEA; ### p < 0.001 versus PEA. Figure 1. Efficacy of PEA-OXA on rCBF and the ischemic area induced by transient MCAo: rCBF was controlled to confirm the success of the induction of the transient ischemia. This was monitored at baseline (10 min before), at 10 and 60 min during ischemia, and 5 min after ischemia (A). TTC staining of brain sections was performed 24 h after MCAo. Tissues from vehicle-treated animals displayed the presence of an unstained area (B,D) compared to the sham group (B,C). PEA-OXA administration (B,F) reduced this infarcted area more effectively than did the PEA treatment (B,E). A histological analysis of brain samples from vehicle-treated rats revealed the loss of neurons (H,M) compared to the sham group (G,M), which was ameliorated by PEA administration (I,M). PEA-OXA administration displayed more protective effects compared to the PEA treatment (L,M). For the histology, a 20× magnification is shown (50-µm scale bar). A p-value of less than 0.05 was considered significant: •• p < 0.01 versus vehicle; *** p < 0.001 versus sham; ••• p < 0.001 versus vehicle; # p < 0.05 versus PEA; ### p < 0.001 versus PEA.

Effect of PEA-OXA Treatment on MCAo-Induced Mast Cell Degranulation
Toluidine blue staining was performed to investigate mast cell infiltration and degranulation in the injured tissue. In brain sections from vehicle-treated animals, an increased number of mast cells were identified ( Figure 2B,E) compared to the tissues from sham-operated rats (Figure 2A,E). In samples from animals administered PEA, fewer of these cells were observed ( Figure 2C,E). In addition, tissues collected from PEA-OXA-administered animals revealed fewer mast cells than did tissues from vehicle-treated animals ( Figure 2D,E). An immunohistochemical analysis of tryptase expression showed Toluidine blue staining was performed to investigate mast cell infiltration and degranulation in the injured tissue. In brain sections from vehicle-treated animals, an increased number of mast cells were identified ( Figure 2B,E) compared to the tissues from sham-operated rats (Figure 2A,E). In samples from animals administered PEA, fewer of these cells were observed ( Figure 2C,E). In addition, tissues collected from PEA-OXA-administered animals revealed fewer mast cells than did tissues from vehicle-treated animals ( Figure 2D,E). An immunohistochemical analysis of tryptase expression showed increased staining in tissue sections from vehicle-treated animals [46] ( Figure 2G, see densitometric analysis in Figure 2L), which was reduced in samples from PEA-OXAadministered animals ( Figure 2I, see densitometric analysis in Figure 2L). PEA administration was also able to reduce tryptase expression ( Figure 2H, see densitometric analysis in Figure 2L), but PEA-OXA treatment was more effective. The basal staining of tryptase ( Figure 2F, see densitometric analysis in Figure 2L) was detected in the sham-operated animals. MCAo. An increased number of mast cells were detected in tissues from vehicle-treated animals (B,E) compared to the control (A,E). PEA and PEA-OXA-treated ischemic rats showed fewer cells of this type (C-E). An increased expression of tryptase (G,L) was found in sections obtained from vehicletreated rats compared to the sham-operated animals (F,L). PEA-OXA administration reduced tryptase expression more effectively than did PEA (H,I,L). The number of mast cells was counted in three An increased number of mast cells were detected in tissues from vehicle-treated animals (B,E) compared to the control (A,E). PEA and PEA-OXA-treated ischemic rats showed fewer cells of this type (C-E). An increased expression of tryptase (G,L) was found in sections obtained from vehicle-treated rats compared to the sham-operated animals (F,L). PEA-OXA administration reduced tryptase expression more effectively than did PEA (H,I,L). The number of mast cells was counted in three sections per animal and is presented as the number of positive cells per high-power field. For the mast cells, a 100× magnification is shown (10-µm scale bar). For the immunohistochemistry, a 40× magnification is shown (75-µm scale bar). A p-value of less than 0.05 was considered significant: *** p < 0.001 versus sham; ••• p < 0.001 versus vehicle; # p < 0.05 versus PEA.

Effect of PEA-OXA Treatment on MCAo-Induced SIRT1 and UCP2 Expression and Redox Status
To understand the molecular mechanism by which PEA-OXA acts, we evaluated (by western blot) the modulation of SIRT1 (silent information regulator 1) and UCP2 (uncoupling protein 2) expression. PEA-OXA administration increased SIRT1 expression compared to the vehicle group ( Figure 3A and the densitometric analysis in Figure 3B), while PEA treatment displayed less of an increase ( Figure 3A and the densitometric analysis in Figure 3B). SIRT1 gene expression was increased in the PEA-OXA-administered animals compared to the vehicle group, while PEA-treated animals showed less upregulation ( Figure 3E). Treatment with PEA-OXA also diminished UCP2 expression compared to the vehicle-treated rats ( Figure 3C and the densitometric analysis in Figure 3D), while PEA treatment displayed less efficacy ( Figure 3C and the densitometric analysis in Figure 3D). To observe the effect of PEA-OXA administration on redox status, the NAD+/NADH ratio ( Figure 3F), the GSH level ( Figure 3G), and SOD activity ( Figure 3H) after MCAo were examined. The results showed that GSH and SOD were increased after PEA-OXA administration compared to the vehicle-treated animals, while PEA administration showed less efficacy ( Figure 3G,H). Compared to the sham-treated animals, the NAD+/NADH ratio was higher in the vehicle-treated rats, while PEA-OXA treatment restored its levels ( Figure 3F). PEA administration displayed less efficacy in the increase of NAD+/NADH ( Figure 3F).

Effects of PEA-OXA Treatment on MCAo-Induced Apoptosis
The apoptotic pathway was evaluated by western blot analysis of Bax and Bcl-2 and by TUNEL staining: 24 h after transient MCAo, an upregulated expression of Bax ( Figure 4A and the densitometric analysis in Figure 4B) and a reduced expression of Bcl-2 ( Figure 4C and the densitometric analysis in Figure 4D) were detected in samples from vehicle-treated animals The efficacy of PEA-OXA on SIRT1 and UCP2 expression, the NAD + /NADH ratio, GSH, and SOD activity (induced by transient MCAo). A western blot analysis of SIRT1 showed basal expression that was significantly increased after MCAo. After PEA-OXA treatment, this expression was significantly upregulated, while after PEA administration, it increased less (A,B). Real-time PCR analysis of the SIRT1 gene displayed the same trend (E). UCP2 expression in vehicle-treated animals was significantly increased compared to the sham-operated animals. PEA-OXA administration was able to significantly restore the basal levels (C,D). PEA displayed less protection. The NAD + /NADH ratio was reduced by PEA-OXA treatment compared to the vehicle-treated animals (E). MCAo reduced GSH (G) and SOD (H) activity in vehicle-treated animals, while PEA-OXA administration restored the basal levels. A p-value of less than 0.05 was considered significant: * p < 0.05 versus sham; • p < 0.05 versus vehicle; •• p < 0.01 versus vehicle; *** p < 0.001 versus sham; ••• p < 0.001 versus vehicle; # p < 0.05 versus PEA; ### p < 0.001 versus PEA.

Effects of PEA-OXA Treatment on MCAo-Induced Apoptosis
The apoptotic pathway was evaluated by western blot analysis of Bax and Bcl-2 and by TUNEL staining: 24 h after transient MCAo, an upregulated expression of Bax ( Figure 4A and the densitometric analysis in Figure 4B) and a reduced expression of Bcl-2 ( Figure 4C and the densitometric analysis in Figure 4D) were detected in samples from vehicle-treated animals compared to the control group. PEA-OXA administration both significantly decreased Bax and increased Bcl-2 levels; in contrast, treatment with PEA at the same dose was less beneficial ( Figure 4A,C and the densitometric analysis in Figure 4B,D). The same result was observed in TUNEL staining, where the number of positive cells was significantly diminished after treatment with PEA-OXA compared to the MCAo group ( Figure 4E-I). Moreover, since it is well known that caspases play pivotal roles in apoptosis, we also observed the expression of caspases by immunostaining. The results obtained showed significantly increased levels of caspase 3 in the MCAo group compared to the sham group ( Figure 5A and the densitometric analysis in Figure 5E). Treatment with PEA-OXA markedly reduced its expression ( Figure 5D and the densitometric analysis in Figure 5E). PEA treatment showed less efficacy ( Figure 5C and the densitometric analysis in Figure 5E). We also investigated the levels of LDH in serum. It was higher in the vehicle-treated animals compared to the sham, while PEA-OXA administration reduced its value ( Figure 5F). Treatment with PEA at the same dose was less beneficial ( Figure 5F). OXA compared to the MCAo group ( Figure 4E-I). Moreover, since it is well known that caspases play pivotal roles in apoptosis, we also observed the expression of caspases by immunostaining. The results obtained showed significantly increased levels of caspase 3 in the MCAo group compared to the sham group ( Figure 5A and the densitometric analysis in Figure 5E). Treatment with PEA-OXA markedly reduced its expression ( Figure 5D and the densitometric analysis in Figure 5E). PEA treatment showed less efficacy ( Figure 5C and the densitometric analysis in Figure 5E). We also investigated the levels of LDH in serum. It was higher in the vehicle-treated animals compared to the sham, while PEA-OXA administration reduced its value ( Figure 5F). Treatment with PEA at the same dose was less beneficial ( Figure 5F).   An immunohistochemical analysis of caspase 3 (E) was performed. A marked increase in its expression was detected in brain samples from animals administered vehicle (B) compared to the sham-treated animals (A). This expression was notably reduced by treatment with PEA-OXA (D), while PEA treatment displayed less efficacy (C). PEA-OXA administration also reduced LDH levels to normal levels (F). For the immunohistochemistry, a 20× magnification is shown (50-µm scale bar).

Effect of PEA-OXA Treatment on MCAo-Induced IκB-α Degradation, NF-κB Translocation, and TGFβ Expression
We found that IκB-α was significantly downregulated in vehicle-treated animals subjected to MCAo compared to the sham group. PEA-OXA administration partially restored this expression ( Figure 6A and the densitometric analysis in Figure 6B), while PEA treatment displayed less efficacy ( Figure 6A and the densitometric analysis in Figure 6B). Treatment with PEA-OXA also diminished NF-κB translocation into the nucleus compared to the vehicle-treated rats ( Figure 6C and the densitometric analysis in Figure 6D), while PEA administration reduced its activation with significantly less efficacy ( Figure 6C and the densitometric analysis in Figure 6D). NF-κB gene An immunohistochemical analysis of caspase 3 (E) was performed. A marked increase in its expression was detected in brain samples from animals administered vehicle (B) compared to the sham-treated animals (A). This expression was notably reduced by treatment with PEA-OXA (D), while PEA treatment displayed less efficacy (C). PEA-OXA administration also reduced LDH levels to normal levels (F). For the immunohistochemistry, a 20× magnification is shown (50-µm scale bar). A p-value of less than 0.05 was considered significant: • p < 0.05 versus vehicle; *** p < 0.001 versus sham; ••• p < 0.001 versus vehicle; # p < 0.05 versus PEA; ## p < 0.01 versus PEA.

Effect of PEA-OXA Treatment on MCAo-Induced IκB-α Degradation, NF-κB Translocation, and TGF-β Expression
We found that IκB-α was significantly downregulated in vehicle-treated animals subjected to MCAo compared to the sham group. PEA-OXA administration partially restored this expression ( Figure 6A and the densitometric analysis in Figure 6B), while PEA treatment displayed less efficacy ( Figure 6A and the densitometric analysis in Figure 6B). Treatment with PEA-OXA also diminished NF-κB translocation into the nucleus compared to the vehicle-treated rats ( Figure 6C and the densitometric analysis in Figure 6D), while PEA administration reduced its activation with significantly less efficacy ( Figure 6C and the densitometric analysis in Figure 6D). NF-κB gene expression was increased in the vehicle-administered animals compared to the sham group, while PEA-OXA-treated animals showed reduced expression. In addition, PEA administration reduced NF-κB gene expression but with less efficacy ( Figure 6E). Considering that recent evidence has demonstrated that TGF-β plays a key role in the formation of cerebral edema as well as in the neuronal processes that lead to necrosis in the acute stage of ischemia, we investigated the effect of PEA-OXA treatment on TGF-β expression after MCAo [47,48]. An immunohistochemical analysis of TGF-β expression showed increased staining in tissue sections from vehicle-treated animals ( Figure 6H and the densitometric analysis in Figure 6F), which was reduced in samples from PEA-OXA-administered animals ( Figure 6L and the densitometric analysis in Figure 6F). PEA administration was also able to reduce TGF-β expression ( Figure 6I and the densitometric analysis in Figure 6F), but PEA-OXA treatment was more effective. The basal staining of TGF-β expression ( Figure 6G and the densitometric analysis in Figure 6F) was detected in the sham-operated animals.
Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 9 of 23 expression was increased in the vehicle-administered animals compared to the sham group, while PEA-OXA-treated animals showed reduced expression. In addition, PEA administration reduced NF-κB gene expression but with less efficacy ( Figure 6E). Considering that recent evidence has demonstrated that TGF-β plays a key role in the formation of cerebral edema as well as in the neuronal processes that lead to necrosis in the acute stage of ischemia, we investigated the effect of PEA-OXA treatment on TGF-β expression after MCAo [47,48]. An immunohistochemical analysis of TGF-β expression showed increased staining in tissue sections from vehicle-treated animals ( Figure  6H and the densitometric analysis in Figure 6F), which was reduced in samples from PEA-OXAadministered animals ( Figure 6L and the densitometric analysis in Figure 6F). PEA administration was also able to reduce TGF-β expression ( Figure 6I and the densitometric analysis in Figure 6F), but PEA-OXA treatment was more effective. The basal staining of TGF-β expression ( Figure 6G and the densitometric analysis in Figure 6F) was detected in the sham-operated animals.  It is well known that TNF-α and IL-1 β have been implicated in the functional consequences of neuroinflammation. An immunohistochemical analysis of TNF-α and IL-1 β expression showed increased staining in tissue sections from vehicle-treated animals ( Figure 7B,F, see densitometric analysis in Figure 7I,L), which was reduced in samples from PEA-OXA-administered animals ( Figure 7D,H, see densitometric analysis in Figure 7I,L). PEA administration was also able to reduce TNF-α and IL-1β expression ( Figure 7C,G, see densitometric analysis in Figure 7I,L), but PEA-OXA treatment was more effective. The basal staining of TNF-α and IL-1β ( Figure 7A,E, see densitometric analysis in Figure 7I,L) was detected in sham-operated animals.

Effect of PEA-OXA Treatment on MCAo-Induced Reduced BDNF and GDNF Expression
Transient MCAo induced a marked reduction in BDNF and GDNF neurotrophic factors ( Figure 8B,F, see densitometric analysis in Figure 8I,L). On the other hand, an immunofluorescence analysis of brain samples from PEA-OXA-treated animals showed a significantly increased expression of these factors ( Figure 8D,H, see densitometric analysis in Figure 8I,L), while PEA displayed less protection ( Figure 8C,G, see densitometric analysis in Figure 8I,L). The basal expression of BDNF and GDNF was found in tissues from control animals ( Figure 8A,E, see densitometric analysis in Figure 8I,L).

Discussion
Cerebral ischemia is associated with a severe morbidity and high mortality rate, in particular in diabetic patients. Patients display related neurological disorders, such as cognitive and severe motor impairments [49,50]. Investigation of the mechanisms involved in worsening neuroinflammatory injury following cerebral ischemia in diabetes and linked hypoglycemia is very important. Suppressing potential candidates involved in enriching neuroinflammatory reactions may help to downgrade stroke severity and promote recovery in diabetic/hypoglycemic conditions [51]. It has been described that cerebral ischemia affects adaptive and innate immune systems [52]; leads to oxidative damage [53], excitotoxicity [54], blood-brain barrier dysfunction [55], microvascular injury [56], and post-ischemic inflammation [8,[57][58][59]; and induces neuronal apoptosis [60]. Further investigations to discover new therapeutic strategies remain a high priority. Palmitoylethanolamide (PEA) is an important endogenous molecule that controls tissue reactivity and associated inflammatory antalgic phenomena, both in the central nervous system (CNS) and in innervated peripheral tissues [61,62]. N-acylethanolamines (NAEs), the family of molecules to which PEA belongs, are selectively degraded by two intracellular enzymes: N-acylethanolamine-hydrolyzing acid amidase (NAAA) [63,64] and fatty-acid amide hydrolase (FAAH) [65]. In particular, the amidase NAAA is the most important for PEA degradation [66], which means that the inhibition of this enzyme should increase PEA tissue levels [38,41,67,68]. Interesting strategies to modulate substrate availability underline the ability of the oxazoline derivatives of fatty acids to inhibit NAAA and display inhibitory activity toward inflammatory processes [69]. Recent evidence has shown the capability of a new PEA derivate, PEA-oxazoline (PEA-OXA), in inhibiting NAAA and decreasing inflammation processes [43,[70][71][72]. In the present study, we focused our attention on the neuroprotective effects of PEA-OXA in a rat model of transient middle cerebral artery occlusion (MCAo) in diabetic rats. Our findings showed that oral administration of PEA-OXA at a dose of 10 mg/kg could significantly reduce lesion size, histological damage, and mast cell activation and degranulation associated with ischemia/reperfusion (I/R) injury. The inflammatory response to brain injury is characterized by a synchronized activation of various pathways that manage the expression of both anti-and proinflammatory mediators recruited from the blood and resident in tissue cells. In vitro studies have underlined a neuroprotective role for silent information regulator 1 (SIRT1) in ischemic injury [73]. Moreover, it is well known that PEA and SIRT1 share the same mechanism, expounding anti-inflammatory, analgesic, and neuroprotective effects [74][75][76][77]. In the present study, for the first time we show that PEA-OXA also has neuroprotective effects through the activation of SIRT1. During ischemia/reperfusion injury, the upregulation of SIRT1 is accompanied by the upregulation of mitochondrial uncoupling protein-2 (UCP2) [78,79]. PEA-OXA administration activated SIRT1 signaling and caused a reduced expression of UCP2. It is a proton channel of the inner mitochondrial membrane that regulates oxidative stress and energy supply [80] by modulating the oxidation of NADH, the level of ATP, and the redox balance [81,82]. An ischemic condition influenced the redox status, increasing the NAD + /NADH ratio and reducing SOD activity and the GSH level. SIRT1 activation induced by PEA-OXA administration caused a decrease in the NAD + /NADH ratio and an increase in SOD activity and the GSH level. It is well described that brain ischemia/reperfusion injury leads to the activation of the apoptotic pathway [83]. The activation of SIRT1 signaling through PEA-OXA administration increased the expression of B-cell lymphoma 2 (Bcl-2) and decreased the expression of Bax. It also decreased the expression of caspase 3, TUNEL staining, and LDH in serum. PEA-OXA, targeting SIRT1, mediated different effects that contributed to a protective effect of the nervous system. SIRT1 has been described as interacting with NF-κB and preventing its transcriptional activity [84]. It is one of the main pathways activated by ischemic brain injury [14,15]. Normally, it is sequestered into the cytoplasm by the inhibitor protein IkBα. Once IkBα is degraded, it allows for NF-κB translocation into the nucleus [85]. NF-κB, in turn, induces the expression of proinflammatory genes, including chemokines, cytokines, and adhesion molecules [86]. PEA-OXA administration reduced IκB-α degradation and consequently NF-κB translocation into the nucleus. Moreover, our results clearly demonstrate that PEA-OXA treatment downregulated the IL-1β, TGF-β, and TNF-α expression induced by cerebral ischemia [48]. In addition, neurotrophic family proteins exerted neuroprotective effects against ischemic stress, protecting against brain damage induced by ischemia and alleviating cognitive impairment [87]. Following MCAo, BDNF and GDNF expression was downregulated [88], while PEA-OXA administration could significantly restore their expression to sham levels.

Animals
Male Wistar rats (Envigo, Milan, Italy) weighing 200-250 g were employed. Animals were divided in groups of three with free access to water and food under standardized humidity and temperature. This study was approved by the University of Messina Review Board for the care of animals (Protocol number 8/U-apr16). Animal care was in conformity with current legislation for the protection of animals used for scientific purposes (Directive 2010/63/EU, 9 April 2016) and the ARRIVE guidelines.

Middle Cerebral Artery Occlusion
Focal cerebral ischemia was induced by transient MCAo in the right hemisphere. The animals were anesthetized with inhaled 1.0-2.0% isoflurane and 5.0% isoflurane (Baxter International) in air through the use of a mask. Body temperature was preserved at 37 • C with a heating pad and was supervised via an intrarectal type T thermocouple (Harvard, Kent, UK). The animals were located in a stereotaxic system (Kopf). Middle cerebral artery (MCA) occlusion was provided by introducing a 4-0 nylon monofilament (Ethilon; Johnson & Johnson, Somerville, NJ, USA), precoated with silicone (Xantopren; Heraeus Kulzer, Germany) via the external carotid artery into the internal carotid artery to occlude the MCA [89][90][91]. Sham animals were subjected to the same procedure, but the filament was introduced into the internal carotid artery and suddenly withdrawn. At the end of the procedure, anesthesia was discontinued, and the rats were returned to a prone position. Laser Doppler flowmetry (PeriFlux System 5000; Perimed AB, Stockholm, Sweden) with a flexible probe over the skull was used to monitor regional cerebral blood flow (rCBF), as previously described [92].

Synthesis of PEA and PEA-OXA
PEA and PEA-OXA were synthesized as previously described by Impellizzeri and colleagues [43].

Induction of Diabetes
Diabetes was induced by streptozotocin, as previously described by Di Paola et al. [93]. We confirmed a diabetic condition by evaluating glucose levels at 15 and 60 days through a blood glucose meter (Table 1) (Accu-Check Active ® ; Roche Diagnostic, Milan, Italy).

Experimental Groups
Sixty days after the induction of diabetes, rats were randomly assigned into different groups, as described below (n = 30): TZ-hyperglycemic rats (ischemia/reperfusion (I/R) + vehicle): Diabetic rats were subjected to MCAo (1 h) followed by 24 h of reperfusion [94]. One hour after ischemia and six hours after reperfusion, carboxymethyl cellulose (CMC) in saline (1.5 %, w/v) was administered (os), and the rats were sacrificed 24 h later.
STZ-hyperglycemic rats (I/R + PEA): Diabetic rats were subjected to the surgical procedures described above. PEA (10 mg/kg in 1.5% CMC) was administered (os) 1 h after ischemia and 6 h after reperfusion, and the rats were sacrificed 24 h later; STZ-hyperglycemic rats (I/R + PEA-OXA): Diabetic rats were subjected to the surgical procedures described above. PEA-OXA (10 mg/kg in 1.5% CMC) was administered (os) 1 h after ischemia and 6 h after reperfusion, and the rats were sacrificed 24 h later; Sham + vehicle: Rats were subjected to the same procedures, but the filament was introduced into the internal carotid artery and suddenly withdrawn, and the rats were kept under anesthesia for the duration of the experiment. The animals were administered (os) 1.5% (w/v) CMC in saline at the same time point as the MCAo group and were sacrificed 24 h later; Sham + PEA: The animals were the same as the sham-operated rats except for the administration of PEA (10 mg/kg in 1.5% CMC, os) 1 h after ischemia and 6 h after reperfusion, and the rats were sacrificed 24 h later (data not shown); Sham + PEA-OXA: The animals were the same as the sham-operated rats except for the administration of PEA-OXA (10 mg/kg in 1.5% CMC, os) 1 h after ischemia and 6 h after reperfusion, and they were sacrificed 24 h later (data not shown).
Rats were given buprenorphine (0.03 mg/kg, subcutaneous) and saline immediately after surgery. Randomization based on a single sequence of random assignments is known as simple randomization [95][96][97]. This technique maintains complete randomness of the assignment of a subject to a particular group. We used the most common and basic method of simple randomization: flipping a coin.
The doses (10 mg/kg) of PEA and PEA-OXA, the administration route (os), and the vehicle employed were chosen basing on our previous study [44,72,89].
At the conclusion of the experiment, animals were euthanized under anesthesia: the brain was removed and fixed in 10% neutral-buffered formalin and then embedded in paraffin for future histological analysis or stored at −80 • C for western blot or biochemical analyses (see graphical timeline).

Quantification of Infarct Volume
Tissues were incubated in a 2% solution of 2,3,5-triphenyltetrazolium chloride (TTC) for 30 min at 37 • C, processed, and quantified as previously described [96][97][98]. All analyses were carried out by two observers blinded to the treatment.

Histological Evaluation
A histological evaluation was made as previously described by Ahmad et al. [88,99]. For the histology, a 20× magnification is shown (50-µm scale bar). All analyses were carried out by two observers blinded to the treatment.

Toluidine Blue Staining
In order to evaluate mast cell numbers and degranulation, brain tissue sections were stained with toluidine blue. Sections were stained blue, while the mast cells were stained purple. The mast cell count was performed on each slide through an Axiovision Zeiss (Milan, Italy) microscope and is shown at a 100× magnification (10-µm scale bar). All analyses were carried out by two observers blinded to the treatment.

Immunohistochemical Localization of Tryptase, Caspase-3, Interleukin-1 Beta (IL-1β), Tumor Necrosis Factor-Alpha (TNF-α), and Transforming Growth Factor Beta (TGF-β)
An immunohistochemical localization was done as previously described by Cordaro et al. [45]. Slices were incubated at room temperature overnight with one of the following primary antibodies: anti-tryptase (Santa Cruz Biotechnology, Heidelberg, Germany, 1:100 in PBS, v/v), anti-caspase-3 (Cell Signalling, Danvers, MA, USA 1:300 in PBS, v/v), anti-IL-1β (Santa Cruz Biotechnology, Heidelberg, Germany, 1:50 in PBS, v/v), anti-TNF-α (Santa Cruz Biotechnology, Heidelberg, Germany, 1:100 in PBS, v/v), or anti-TGF-β (Santa Cruz Biotechnology, Heidelberg, Germany 1:50 in PBS, v/v). For a graphic display of the densitometric analyses, the % of positive staining (brown staining) was measured by computer-assisted color image analysis (Leica QWin V3, Newcastle, UK). The percentage area of immunoreactivity (determined by the number of positive pixels) was expressed as % of total tissue area (red staining) within five random fields at a 40× magnification. In particular, first the colors of the images that were stained on the molecule of interest were defined. Once these colors were defined, they were automatically detected in all samples. This is a semiquantitative analysis that measures areas and not intensities [100,101]. Replicates for each experimental condition and histochemical staining were acquired from each mouse in each experimental group. All immunohistochemical analyses were carried out by two observers blinded to the treatment.

TUNEL Staining
The TUNEL staining protocol was according to the Roche protocol previously described by Fusco et al. [102]. The tissue was then rinsed in PBS 3 times for 5 min and then observed using an excitation wavelength in the range of 520-560 nm (maximum 540; green) and in the range of 570-620 nm (maximum 580 nm; red). For TUNEL staining, a 100× magnification is shown (10-µm scale bar). All analyses were carried out by two observers blinded to the treatment.

NAD+/NADH Assay
NAD+ and NADH levels were analyzed by an assay kit according to the manufacturer's instructions (Suzhou Comin, China). The brain was homogenized in the lysis buffer and was then centrifuged at 4 • C. MTT was then reduced to formazan by the supernatant. The formazan was dissolved in the buffer, and the absorbance was read at 570 nm. The relative level of NAD+ was expressed as an NAD+/NADH ratio. All analyses were carried out by two observers blinded to the treatment.

GSH Levels Assay
GSH levels were analyzed using an assay kit according to the manufacturer's instructions (Nanjing Jiancheng, Jiangsu, China). The brain was homogenized in saline and was then centrifuged at 4 • C. The supernatant was set to react with 5,5 -dithiobis(2-nitrobenzoic acid) (DTNB). Then the absorbance was read at 405 nm. The level of GSH was expressed as µmol/g protein. All analyses were carried out by two observers blinded to the treatment.

SOD Activity Assay
SOD activity was determined using an activity assay kit according to the manufacturer's instructions. The brain was homogenized in saline and was then centrifuged at 4 • C. The absorbance was read at 450 nm. The SOD activity was expressed as U/mg protein. All analyses were carried out by two observers blinded to the treatment.

LDH Release Detection
Twenty-four hours after reperfusion, each rat was euthanized, the blood was collected, and the serum was separated. Lactate dehydrogenase (LDH) release was measured. The detection was conducted with an assay kit (Solarbio, Beijing, China). All analyses were carried out by two observers blinded to the treatment.

RT-PCR
Total RNA was extracted according to the manufacturer's protocol and reverse-transcribed using 2-µg oligo(dT)15 primer, 10 units of AMV reverse transcriptase, 40 units of RNase inhibitor (all from Promega, Southampton, UK), and 1.25 mM of dNTP (Bioline, London, UK) for 45 min at 42 • C. Real-time PCR was carried out using TaqMan Universal PCR master mix and fluorescent primers obtained from Quiagen (QuantiTect primers, Skelton House, UK). Cycling conditions were set according to the manufacturer's instructions. Sequence-specific fluorescent signals were detected by an ABI Prism 7700Sequence Detector System, and mRNA data were normalized relative to GADPH and then used to calculate expression levels.

Materials
STZ and the other compounds were acquired from Sigma-Aldrich Company Ltd. (Milan, Italy). All other chemicals were of the highest viable grade available. All stock solutions were arranged in nonpyrogenic saline (0.9% NaCl; Baxter, Milan, Italy). PEA and PEA-OXA were kindly provided by the Epitech Group (Saccolongo, Padua, Italy).

Statistical Analysis
All values are shown as the mean ± standard error of the mean (SEM) of N observations. In the experiments, including the immunohistochemistry, histology, and immunofluorescence, the figures shown are representative of at least three experiments performed on different days on tissue sections collected from all animals in each group. Data were analyzed by one-way ANOVA followed by a Bonferroni post hoc test for multiple comparisons. A p-value of less than 0.05 was considered significant: * p < 0.05 versus sham; • p < 0.05 versus vehicle; ** p < 0.01 versus sham; •• p < 0.01 versus vehicle; *** p < 0.001 versus sham; ••• p < 0.001 versus vehicle; # p < 0.05 versus PEA; ## p < 0.01 versus PEA; ### p < 0.001 versus PEA.

Conclusions
In summary, our data clearly demonstrated for the first time that PEA-OXA treatment (10 mg/kg) significantly improved neurological injuries induced by transient MCAo compared to a simple PEA treatment at the same dose.

Conflicts of Interest:
Salvatore Cuzzocrea is a coinventor on patent WO2013121449 A8 (Epitech Group Srl), which deals with methods and compositions for the modulation of amidases capable of hydrolyzing N-acylethanolamines employable in the treatment of inflammatory diseases. This invention is wholly unrelated to the present study. Moreover, Cuzzocrea is also, with Epitech Group, a coinventor on the patents EP 2 821 083, MI2014 A001495, and 102015000067344, which are unrelated to the study. The remaining authors report no conflicts of interest.