15d-PGJ2-loaded nanocapsules ameliorate experimental gout arthritis by reducing pain and inflammation in a PPAR-gamma-sensitive manner in mice

Gout arthritis (GA) is a painful inflammatory disease in response to monosodium urate (MSU) crystals in the joints. 15deoxy-Δ12,14-prostaglandin J2 (15d-PGJ2) is a natural activator of PPAR-γ with analgesic, anti-inflammatory, and pro-resolution properties. Thus, we aimed to evaluate the effect and mechanisms of action of 15d-PGJ2 nanocapsules (NC) in the model of GA in mice, since a reduction of 33-fold in the dose of 15d-PGJ2 has been reported. Mice were treated with 15d-PGJ2-loaded NC, inert NC, free 15d-PGJ2 (without NC), or 15d-PGJ2-loaded NC+ GW9662, a PPAR-γ inhibitor. We show that 15d-PGJ2-loaded NC provided analgesic effect in a dose that the free 15d-PGJ2 failed to inhibiting pain and inflammation. Hence, 15d-PGJ2-loaded NC reduced MSU-induced IL-1β, TNF-α, IL-6, IL-17, and IL-33 release and oxidative stress. Also, 15d-PGJ2-loaded NC decreased the maturation of IL-1β in LPS-primed BMDM triggered by MSU. Further, 15d-PGJ2-loaded NC decreased the expression of the components of the inflammasome Nlrp3, Asc, and Pro-caspase-1, as consequence of inhibiting NF-κB activation. All effects were PPAR-γ-sensitive. Therefore, we demonstrated that 15d-PGJ2-loaded NC present analgesic and anti-inflammatory properties in a PPAR-γ-dependent manner inhibiting IL-1β release and NF-κB activation in GA. Concluding, 15d-PGJ2-loaded NC ameliorates MSU-induced GA in a PPAR-γ-sensitive manner.

step in the pathogenesis of the disease and pain experienced by the patients (namely acute flares) 4,6 . Patients seek medical care due to acute flares 3 . In spite of self-resolving (about 10 days), gout acute flares are one of the most painful experiences to humans 3 . The management of pain in gout acute flares depends on steroidal and non-steroidal anti-inflammatory drugs (NSAIDs), colchicine and biological agents targeting IL-1 3 . However, the use of these drugs lack safety in patients with comorbidities (NSAIDs), often cause severe side effects (NSAIDs, colchicine, and corticoids), present high cost (biological agents), or possess non-satisfactory analgesic effects in some patients with gout 3,4 . Thus, novel analgesic drugs are still needed.
The cyclopentatone type-15deoxy-Δ 12,14 -prostaglandin J 2 (15d-PGJ 2 ) is formed as a consequence of dehydration of prostaglandin D 2 and is a natural ligand that activates the peroxisome proliferator-activated receptor gamma (PPAR-γ) 7,8 . 15d-PGJ 2 is reported to stimulate transcription of several target genes via PPAR-γ-dependent mechanisms and to promote the resolution of inflammatory process as an endogenous feedback regulator of the inflammatory process in vivo 7,9 . Moreover, 15d-PGJ 2 has been reported to reduce NF-κB activation, an effect intermediated by the activation of PPAR-γ (an indirect mechanism) 10 or by targeting IκB kinase (IKK) or p65 NF-κB subunit (a direct mechanism) 11,12 . In addition to resolve inflammation, activation of PPAR-γ by other molecules also induces analgesia 13,14 , which indicates that activating this nuclear receptor represents a promising analgesic approach. In fact, treatment with 15d-PGJ 2 reduces temporomandibular joint pain [15][16][17][18] . Interestingly, nanoencapsulated 15d-PGJ 2 requires a dose 33 times lower to inhibit inflammation 19 and pain compared to the effects of free 15d-PGJ 2 (30 μg/kg vs. 1000 μg/kg) 15 . Nanocapsules consist of polymeric involucres around a nucleus generally oily used to prolong pharmacological activity and decrease toxicity of molecules 20 . 15d-PGJ 2 was shown to be released slowly from nanocapsules, which avoided inactivation or reduction of its biological activity 21,22 by reactions that could include Michael's addition 23 . Another possibility is that cells could uptake nanoparticles loaded with 15d-PGJ 2 resulting in enhanced 15d-PGJ 2 levels inside the cells as well as increased pharmacological response. Thus, this enhancement of 15d-PGJ 2 activity demonstrates that the nanoencapsulation process worked properly to increase the efficacy of 15d-PGJ 2 18 . In fact, other alternatives, such as topical delivery of 15d-PGJ 2 further corroborates that changes in the formulation and delivery of this molecule increases its activity, i.e., it is required a lower dose to achieve the same effect compared to simply dissolved 15d-PGJ 2 16 . Thus, in this work, our aim was to investigate the efficacy and mechanisms of action of 15d-PGJ 2 -loaded nanocapsules (NC) in MSU-induced inflammation and pain in mice.

15d-PGJ 2 -loaded nanocapsules (NC) inhibit MSU-induced mechanical hyperalgesia and joint swelling via PPAR-γ.
First, it was addressed whether or not 15d-PGJ 2 -loaded NC (s.c., 30 min before MSU injection, 3-30 μg/kg) could reduce mechanical hyperalgesia and joint swelling induced by MSU crystals. MSU injection induced mechanical hyperalgesia (Fig. 1A) and joint swelling (Fig. 1B) at all measured time points. 15d-PGJ 2 -loaded NC at 10 μg/kg reduced MSU-induced mechanical hyperalgesia (Fig. 1A) and joint swelling ( Fig. 1B) (3-15 h). The dose of 30 µg/kg of 15d-PGJ 2 -loaded NC was able to inhibit these same parameters with statistical difference when compared to the dose of 10 μg/kg. Of note, all three doses of 15d-PGJ 2 -loaded NC reduced joint swelling at 1 h after MSU injection. However, the lower dose (3 μg/kg) lost effect over time, the mid dose (10 μg/kg) maintained its inhibitory effect not allowing the increase of joint swelling, and the higher dose (30 μg/kg) presented an increase of effect over time. It is possible that the lower dose of 15d-PGJ 2 -loaded NC was sufficient to inhibit MSU-induced joint swelling at 1 h, but since there was an increase of swelling as the inflammation developed, only higher doses were enough to limit swelleing progression. The inert nanocapsules (InNC), i.e. the nanocapsules without 15d-PGJ 2 , showed no effect. Thus, the dose of 30 µg/kg, s.c of 15d-PGJ 2 -loaded NC was select for the next experiments.
As 15d-PGJ 2 is an endogenous regulator of PPAR-γ 7 and activation of this ligand-activated transcription factor has been shown to have analgesic properties 15,24 , we next evaluated whether the effect of 15d-PGJ 2 -loaded NC occurred in a PPAR-γ-sensitive manner. To this end, it was used GW9662 (i.art., 30 min before MSU injection, 3-30 ng per joint), a selective and irreversible antagonist of PPAR-γ 25 . Treatment with 15d-PGJ 2 -loaded NC at 30 μg/kg reduced the pain-like behavior triggered by mechanical stimulus and joint swelling induced by MSU crystals. Both mechanical hyperalgesia (15 h, Fig. 1C) and joint swelling (15 h Fig. 1D) were reverted by the treatment with GW9662 at 10 and 30 ng without difference between these two doses. GW9662 at 3 ng was unable to reduce the analgesic effect of 15d-PGJ 2 -loaded NC at 30 μg/kg (Fig. 1C). Thus, dose 10 ng of GW9662 was select for the next experiments to further investigated the involvement of PPAR-γ in this model. It was also used free 15d-PGJ 2 , i.e., without nanocapsule, at 30 μg/kg (the same dose of 15d-PGJ 2 -loaded NC, s.c. 30 min before MSU injection) to investigate whether nanoencapsulation of 15d-PGJ 2 enhances the analgesic effect of this molecule. Neither free 15d-PGJ 2 (without nanocapsules) nor inert nanocapsules reduced MSU-induced pain-like behavior triggered by mechanical stimulus and joint swelling (Fig. 1C,D).
15d-PGJ 2 -loaded NC reduce MSU-induced pro-inflammatory cytokine production in a PPAR-γ-sensitive manner. The next step was to investigate the effect of 15d-PGJ 2 -loaded nanocapsules on cytokine production and whether this effect was PPAR-γ-sensitive. Treatment with 15d-PGJ 2 -loaded NC decreased the levels of IL-1β (

15d-PGJ 2 -loaded NC reduce MSU-induced IL-1β release in a PPAR-γ-sensitive manner.
The maturation of IL-1β is one of the main mechanisms of gout pathology 5,6 . To investigate whether 15d-PGJ 2 -loaded NC could reduce IL-1β release, an in vitro system with LPS-primed BMDM and stimulation with MSU crystals was used. IL-1β was measured in the supernatant of cells, indicating that it was released in its mature form 26 . First, it was performed concentration-response curve to select the optimal concentration of 15d-PGJ2-loaded NC. Only 15d-PGJ2-loaded nanocapsules at 3 μM reduced IL-1β maturation Fig. 5A, this concentration of 3 μM reduced 93% of IL-1β maturation. Next, a concentration-response curve of GW9662 was performed and we observed that the effect of 15d-PGJ 2 -loaded NC at 3 μM was inhibited only by GW9662 at 10 μM (Fig. 5B). The concentration of 10 μM reverted 80.3% of the protective effect of 15d-PGJ2-loaded NC. Neither free 15d-PGJ 2 (without nanocapsule, at 3 μM -the same concentration of 15d-PGJ 2 -loaded NC) nor inert nanocapsules reduced IL-1β maturation (Fig. 5A) . None of the concentration used in this study reduced cell viability as per LDH and Trypan blue assays (data not shown). SF3 shows control data in which BMDMs from caspase-1/11 deficient mouse did not release IL-1β in the culture supernatant upon MSU stimulus compared to WT BMDMs.   . 6A), Pro-caspase-1 (Fig. 6B), and NLRP3 (Fig. 6C) mRNA expression, and also the Pro-Il-1β mRNA expression (Fig. 6D). Furthermore, 15d-PGJ 2 -loaded NC inhibited NF-κB activation as observed by the reduction in total-p65/phosphorylated-p65 OD ratio (Fig. 6E). The decrease in the ratio is attributed to the increase in the p65 subunit phosphorylated, and therefore, indicating activation (phosphorylation) of the NF-κB signaling pathway. The 15d-PGJ 2 -loaded NC inhibition of NF-κB activation lines up with the inhibition of inflammasome  components mRNA expression. These effects produced by 15d-PGJ 2 -loaded NC were reverted by GW9662, as observed by an increase in the mRNA expression of the inflammasome components and NF-κB activation, indicating involvement of PPAR-γ in 15d-PGJ 2 -loaded NC effects. Neither free 15d-PGJ 2 (without nanocapsule, at 30 μg/kg -the same dose of 15d-PGJ 2 -loaded NC) nor inert nanocapsules showed effect in these parameters (Fig. 6) .

Discussion
In this work, we show that 15d-PGJ 2 -loaded nanocapsules (NC) ameliorate MSU-induced pain and inflammation by reducing disease parameters such as leukocyte recruitment to the knee joint, oxidative stress, and pro-inflammatory cytokine production. Treatment with GW9662, a selective and irreversible inhibitor of PPAR-γ, reverted the benefic effect of 15d-PGJ 2 -loaded NC. Thus, indicating that the effect of 15d-PGJ 2 -loaded NC occurs in a PPAR-γ-sensitive manner. Moreover, the nanoencapsulation of 15d-PGJ 2 improved its effect, given free 15d-PGJ 2 at same dose (30 μg/kg) failed at inhibiting all parameters.
Activation of PPAR-γ has been shown to induce analgesia and to drive active resolution of inflammation in varied experimental models 14,15,24,32,33 . Topical treatment with 15d-PGJ 2 reduces temporomandibular joint pain in rats 16 . Mechanistically, 15d-PGJ 2 activates PPAR-γ that induces the hyperpolarization of nociceptor neurons by activating the NO/cGMP/K + ATP channels signaling pathway through the release of β-endorphin 34,35 . In fact, activation of this signaling pathway is a mechanism by which morphine 36 , dypirone 37 , nitroxyl donor Angeli's salt 38 , and some flavonoids such as diosmin 39 and naringenin 40 produce analgesic effect. Thus, this mechanism contributes to the analgesic properties of 15d-PGJ 2 . In our work, 15d-PGJ 2 -loaded NC was used. The nanoencapsulation of 15d-PGJ 2 results in a reduction of up to 33 times in the active dose of 15d-PGJ 2 compared to its non-nanoencapusaled form (30 μg/kg vs. 1000 μg/kg) to reduce inflammation 20 and pain 41 . Herein, we show that free 15d-PGJ 2 at 30 μg/kg did not present analgesic effect. Considering nanoencapsulation of 15d-PGJ 2 promoted an analgesic effect with a dose that is inactive when not nanoencapsulated, it conceivable that the process of nanoencapsulation improves the efficacy of 15d-PGJ 2 .
Oxidative stress contributes to the development and maintenance of both peripheral and central sensitization 42 . Thus, molecules that target (directly or indirectly) oxidative stress are interesting therapeutic approaches as analgesic molecules 42 . Intrathecal administration of a ROS donor increases the excitability of nociceptors in the dorsal horn of the spinal cord 43 and intraplantar or intraperitoneal injection of ROS donors is sufficient to induce pain-like behaviors in naïve animals [44][45][46][47] . Concerning gout, promising analgesic and anti-inflammatory activities in MSU-induced arthritis were demonstrated for antioxidant molecules 26,48,49 . Furthermore, ROS also  51 . In turn, neutrophils produce ROS upon recognition of MSU crystals 52 . Herein, we show that 15d-PGJ 2 -loaded NC reduced MSU-induced oxidative stress. Part of the pro-resolving effects of 15d-PGJ 2 is related to its antioxidant activity 7 . In fact, 15d-PGJ 2 induces Nrf2 activation and increases the expression of the scavenger receptor CD36 in murine macrophages 9 . Activation of PPAR-γ by varied molecules up-regulates Nrf2/HO-1 signaling and thereby increase the endogenous antioxidant defenses [29][30][31] . Therefore, the increase of Nrf2/Ho-1 mRNA expression might account to the analgesic and anti-inflammatory activity of 15d-PGJ 2 -loaded NC.
The maturation of IL-1β is a crucial step in the pathogenesis of gout [4][5][6] . A study with 448 Chinese patients with recurrent gout shows that macrophages, peripheral blood mononuclear cells (PBMCs), and synoviocytes from these patients with a variant allele that reduce PPAR-γ activity present increased NLRP3 activity and IL-1β release 53 . A siRNA targeting this PPAR-γ allele variant also increases NLRP3 mRNA expression 53 , indicating that PPAR-γ limits MSU-induced inflammation. In MSU-induced inflammation and pain, NLRP3 inflammasome assembly is fundamental to IL-1β maturation 5,6 . In the present study, we show that 15d-PGJ 2 -loaded NC reduced MSU-induced IL-1β release in BMDM culture supernatant in a PPAR-γ-sensitive manner since the treatment with GW9662 reverted the 15d-PGJ 2 -loaded NC activity. Our data corroborate other studies showing that 15d-PGJ 2 54 or pioglitazone (other PPAR-γ activator) 55 reduce MSU-induced pro-inflammatory cytokine production. In comparison to the study of Akahoshi and collaborators 54 , we used a concentration 16 times lower (3 μM vs. 50 μM) indicating that nanoencapsulation increased 15d-PGJ 2 efficacy compared to the non-nanoencapsulated form. Treatment with 15d-PGJ 2 -loaded NC reduced the MSU-induced production of other pro-inflammatory cytokines such as TNF-α, IL-17, and IL-33. This is important given IL-1β 5 , TNF-α 56 , IL-17 57 , and IL-33 58,59 mediate neutrophil recruitment in rheumatic disease. In fact, using three different methodologies (leukocytes count in Neubauer chamber, HE staining, and MSU injection in LysM-eGFP mice), we show that PGJ 2 -loaded NC reduced neutrophil and mononuclear cell recruitment. Recruited neutrophil produce IL-1β, TNF-α, endothelin-1, and PGE 2 that altogether either sensitize nociceptor neurons or activate them 2,60,61 . Furthermore, MSU crystals induce the release of NETs by neutrophils 62 , which is increased by IL-1β 62,63 . Thus, inhibition of pro-inflammatory cytokine production, especially IL-1β (in vivo and in vitro), is fundamental to the analgesic effect of 15d-PGJ 2 -loaded NC. Moreover, we show 15d-PGJ 2 -loaded NC reduced NF-κB activation in vivo, corroborating other study that demonstrates 15d-PGJ 2 reduces NF-κB activation in a model of asthma 64 . Mechanistically, activation of PPAR-γ by 15d-PGJ 2 inhibits the transcriptional response mediated by AP-1 and NF-κB in macrophages 10 . This was observed through the reduction in the luciferase activity of TATA-containing promoter linked to these transcription factors 10 . Furthermore, 15d-PGJ 2 also reduces NF-κB activation by direct mechanism, i.e., that does not depend on PPAR-γ activation. This mechanism is related to covalently binding to IKK 12 or to the alkylation of the cysteine residue Cys38 of the NF-κB subunit p65 11 . Activation of NF-κB drives the production of pro-inflammatory mediators and components of inflammasome 65 . Thus, inhibition of this transcription factor certainly contributed to the analgesic and anti-inflammatory effects of 15d-PGJ 2 -loaded NC.
To conclude, in this work we show that nanoencapsulation of 15d-PGJ 2 increased the effect of this molecule, given free 15d-PGJ 2 (i.e. the same dose and non nanocapsulated) failed at inhibiting MSU-induced pain and inflammation. The protective effect of 15d-PGJ 2 -loaded NC is related to the inhibition of MSU-induced oxidative stress as observed by the increase of antioxidant defense and Nrf2/Ho-1 mRNA expression. In addition, treatment with 15d-PGJ 2 -loaded NC reduced MSU-induced pro-inflammatory cytokine production, especially IL-1β (both in vivo and in vitro) and neutrophil recruitment by a mechanism involving the reduction of NF-κB activation in the context of MSU inflammation. Moreover, the 15d-PGJ 2 -loaded NC effects were PPAR-γ-sensitive, given they were amenable by GW 9662 treatment. Thus, herein, we show that 15d-PGJ 2 -loaded NC possesses protective effect in MSU-induced pain and inflammation and this protective effect is PPAR-γ-sensitive. These results were summarized in Fig. 6F.

Materials and Methods
Experimental procedures. Mice were treated s.c. with 15d-PGJ 2 -loaded NC at the doses of 3, 10, or 30 μg/ kg 30 min before MSU injection. The 15d-PGJ 2 -loaded NC effect on pain-like behavior triggered by mechanical stimulus and joint swelling induced by MSU crystals were determined and the dose of 30 μg/kg was selected for the subsequent experiments. Next, mice were treated with GW9662 at the doses of 3, 10, or 30 ng (i.art., 30 min before MSU injection) to determine if this PPARγ receptor antagonist would inhibit the analgesic and anti-inflammatory effects of 15d-PGJ 2 -loaded NC. GW9662 at the dose of 10 ng was selected for the subsequent experiments. In all experiments, mice were treated with 15d-PGJ 2 -loaded NC, inert nanocapsules (InNC), free 15d-PGJ 2 (without nanocapsules), or 15d-PGJ 2 -loaded NC+ GW9662. All analyzes were conducted 15 h after MSU injection. Leukocyte recruitment was evaluated in the knee joint wash by Neubauer chamber and Rosenfelt stained slices (total leukocyte, neutrophil, and mononuclear cells count) and in MSU-stimulated LysM-eGFP+ mice by confocal microscopy. Histopathological analysis in the knee joint was assessed by HE staining. MSUinduced oxidative stress in the knee joint was evaluated by measuring GSH levels, total tissue antioxidant activity (FRAP and ABTS assays), and production of superoxide anion and nitrite. Further addressing oxidative stress, gp91phox, Nrf2, and Ho-1 mRNA expression were evaluated by RT-qPCR. MSU-induced pro-inflammatory cytokine production was assessed in vivo by ELISA in the knee joint. The mRNA expression of Nlrp3 inflammasome components in the knee joint was determined by RT-qPCR. The maturation of IL-1β was determined by ELISA in the supernatant of LPS-primed BMDM and stimulated with MSU crystals. NF-κB activation in the knee joint was also determined by ELISA. The 15d-PGJ 2 -loaded NC and InNC were prepared and characterized as previously described 20 . The 15d-PGJ 2 (non-nanocapsulated) and GW9662 were acquired from Sigma (Sigma Chemical Co., St. Louis, MO, USA). Supplemental Figures' (SF) experiments were conducted as mentioned above except that leukocyte recruitment was determined 7 h after MSU injection (SF1), a higher dose of 15d-PGJ 2 (300 μg/kg) was also tested in MSU arthritis (SF2), and the effect of caspase-1/11 deficiency over IL-1β release in BMDMs culture supernatant was verified (SF3). Preparation of the PLGA nanocapsules with 15d-PGJ 2 . As described by Alves et al. 20 , the poly (d,l-lactic-co-glycolic acid, 50:50) (PLGA) nanocapsules were prepared by a nanoprecipitation method 66 by mixing an organic phase into an aqueous phase. The organic phase consisted of PLGA polymer (100 mg), acetone (30 mL), 15d-PGJ 2 (100 mg), sorbitan monostereate (40 mg) and caprylic/capric acid triglyceride (200 mg). The aqueous phase consisted of polysorbate 80 (60 mg) and deionized water (30 mL). After disintegration of the components of both phases, the organic phase was gradually added to the aqueous phase, and the suspension agitated during 10 min. The suspension was concentrated to a volume of 10 mL under low pressure using a rotary evaporator to achieve a suspension of 15d-PGJ 2 with a final concentration of 10 µg/mL. A control formulation (without 15d-PGJ 2 ) was also prepared, following the same procedures. Zeta potential measurement and efficiency of association of 15d-PGJ 2 in the PLGA nanocapsules were used to ensure parameters such size and polydispersion measurements. Parameters of morphology and structure of PLGA nanocapsules with 15d-PGJ 2 were also assessed in a JEOL 1200EX II microscope, a Transmission electron microscopy (TEM) operating at 80 kV. A Nanosurf Easy Scan 2 Basic atomic force microscope (BT02217, Nanosurf, Switzerland) was used to verify the diameter of PLGA nanoparticles in suspension and size distribution by using the Nanosurf software 20,41 . Animals. Male Swiss mice (25-30 g) from the Universidade Estadual de Londrina, Londrina, Paraná, Brazil, and WT C57BL/6 background and caspase-1/11 deficient C57BL/6 background mice from Ribeirao Preto Medical School, University of São Paulo were used in this study. Mice were housed in standard clear plastic cages with free access to food and water with a light/dark cycle of 12/12 h at a constant temperature of 21° +/− 1 °C. All behavioral testing was performed in a temperature-controlled room (21° + /− 1 °C) between 9 a.m. and 5 p.m. Animal care and handling procedures were in accordance with the International Association for Study of Pain guidelines, and the Ethics Committee of the Universidade Estadual de Londrina approved all procedures of this study (process number 14600.2013.73). All efforts were made to minimize animal suffering and to reduce the number of animals used. Evaluation of knee joint hyperalgesia. Joint mechanical hyperalgesia was assessed using an electronic von Frey apparatus. Mice were accomodated in acrylic cages with a wire grid floor, and the stimulations were performed only when the animals were quiet (and with the four paws on the grid floor). This test consists of an electronic pressure-meter, with a force transducer fitted with polypropylene tip (Insight instruments, Ribeirao Preto, SP, Brazil). To assess the articular pain, a large tip (4.15 mm 2 ) was used to exclude the cutaneous nociception 67 . A progressive perpendicular pressure was applied to the central area of the plantar surface of the hind paw to induce flexion of the femur-tibial joint followed by the hind paw withdrawal. The intensity of the force applied (in g) at the moment of paw withdrawal was automatically recorded. The test was performed at the times indicated on figures. The investigators were blinded to the treatment.  ABTS and FRAP assays. The tissue antioxidant properties were determined by their free radical scavenging (ABTS [2,2′-Azinobis-3-ethylbenzothiazoline 6-sulfonic acid] assay, Sigma Chemical Co., St. Louis, MO, USA) and ferric reducing (FRAP assay, Sigma Chemical Co., St. Louis, MO, USA) properties. The tests were adapted to a 96-well microplate format as previously described 26 . Articular tissue samples were collected 15 h after MSU i.a injection (100 μg/10 µL) and homogenized immediately in ice-cold KCl buffer (500 μL, 1.15% w/v). The homogenates were centrifuged (200 g × 10 min × 4 °C), and the supernatants were used in both assays. ABTS solution (200 µL) and 10 μL of sample were added to each well and let to react during 6 min incubated at 25 °C followed by reading at 730 nm. For FRAP assay, the supernatants (10 μL) were mixed with the freshly prepared FRAP reagent (150 μL) and let to react during 30 min at 37 °C followed by reading at 595 nm (Multiskan GO Microplate Spectrophotometer, Thermo Scientific, Vantaa, Finland). A standard Trolox curve (Sigma Chemical Co., St. Louis, MO, USA) was used in the ABTS and FRAP assays.

Data analysis.
Data analyzes was performed using the Prism 6.0 statistical program (GraphPad software, Inc.). Comparison between groups and doses at all times when the parameters were measured at different time points after the stimulus injection was performed using two-way ANOVA. The analyzed factors were treatments, time, and time versus treatment interaction. For single time-point, we used one-way ANOVA and the post-test of Tukey. P < 0.05 was considered significant.

Data Availability
All data are presented in the manuscript.