Introduction

Uric acid (UA) is the end product of purine metabolism in humans and a major endogenous antioxidant in blood [1]. Moreover, systemic infusion of high UA doses was safe and increased serum antioxidant capacity in healthy humans [2]. As to acute ischemic stroke, the potential neuroprotective role of UA has been investigated in both preclinical and clinical scenarios. With regard to preclinical studies, mice heterozygous for a disrupted urate oxidase transgene (UOX+/−), with elevated serum UA levels, showed reduced brain damage and improved functional outcome after ischemic stroke [3]. Moreover, several studies have shown the neuroprotective effects of exogenous UA in murine models of acute ischemic stroke [4,5,6,7,8,9,10]; provided suitable doses are used [11].

In the clinical arena, a number of studies lend also support to the neuroprotective effect of UA in ischemic stroke patients. Serum UA level at stroke onset is a predictor biomarker of better prognosis, thus pointing to a protective effect on neurological outcome [12], confirmed by systematic reviews with meta-analyses [13, 14]. From an interventional perspective, a pilot study showed that the administration of UA was safe, decreased oxidative stress markers, and prevented an early fall of serum UA in stroke patients treated with recombinant tissue plasminogen activator (rt-PA) [15, 16]. The moderately sized Efficacy Study of Combined Treatment With Uric Acid and rtPA in Acute Ischemic Stroke (URICO-ICTUS) trial [17] showed that the addition of UA to thrombolytic therapy resulted in a 6% absolute increase in the rate of excellent outcome at 90 days compared with placebo, with no safety concerns [18], and decreased the risk of early ischemic worsening [19]. Moreover, prespecified subgroup analyses showed significant benefits [20,21,22,23].

Despite the promising neuroprotective effects of UA in ischemic stroke, the mechanisms underlying UA benefits seem pleiotropic and are not completely understood. Beyond UA antioxidant effects, recent evidence points to transcription factors as UA targets for neuroprotection and vasculoprotection in acute ischemic stroke rodent models [24]. UA conferred neuroprotection against transient focal cerebral ischemia-induced oxidative stress via Nrf2/HO-1 pathway activation as well as BDNF and NGF expression upregulations [25]. On the other hand, UA treatment after stroke modulated the Krüppel-like factor 2/VEGF-A axis to protect endothelial cell functions and thus reduce brain damage [26]. In humans, endogenous serum UA levels positively associated with IL-6 [27], and exogenous UA administration further increased IL-6 levels [28]. Interestingly, IL-6 played a neuroprotective role through the activation of STAT3 transcription factor in mice subjected to ischemic stroke [29, 30], thus setting out the rationale for the hypothesis that IL-6/STAT3 signaling pathway is involved in the neuroprotective effects of UA in ischemic stroke.

The main objective of the present study was to gain insight into the pleiotropic role of UA in acute ischemic stroke through IL-6/STAT3 signaling pathway, using a rat model of transient focal cerebral ischemia. For this purpose, we investigated the effects of UA administration on pertinent biomarkers for the most relevant features of ischemic stroke pathophysiology: (1) oxidative stress (Sod1, Sod2, and Cat gene expression and malondialdehyde); (2) neuroinflammation (Il6, Tnfa, Il1b, and Socs3 gene expression, STAT3, NF-κB p65, and reactive microglia); (3) brain swelling (Vegfa, Mmp9, and Timp1 gene expression); and (4) apoptosis (Bcl-2, Bax, caspase-3, and TUNEL-positive cells). Neurofunctional deficit, infarct, and edema were measured as the main outcomes to assess the neuroprotective effects of UA treatment.

Methods

Animals and Ethical Issues

A total of 40 male 12-week-old Wistar rats (300–350 g) from Charles River (Barcelona, Spain), housed under standard conditions with food and water ad libitum, were used in the study. Experiments were conducted in compliance with the legislation on protection of animals used for scientific purposes in Spain (RD 53/2013) and the EU (Directive 2010/63/EU). Protocols were approved by the Animal Experimentation Ethics Committee from IIS La Fe. The study was designed and conducted according to the STAIR/RIGOR guidelines [31, 32] regarding physiological monitoring, simple randomization, predefined exclusion criteria, allocation concealment, blinded assessment of several outcomes at different endpoints, and conflict of interest statement. The experimental timeline is summarized in Fig. 1.

Fig. 1
figure 1

Experimental timeline. Adult male Wistar rats underwent transient right middle cerebral artery occlusion (tMCAO, 60 min) by following the intraluminal suture procedure. The procedure included continuous monitoring of cerebrocortical laser-Doppler perfusion, mean arterial blood pressure (MABP) and core temperature, and discontinuous measurement of blood glucose at preischemia, ischemia, and reperfusion stages. Animals were randomly assigned to two experimental groups: vehicle-treated and uric acid (UA)–treated rats. UA was dissolved in vehicle and injected through the femoral vein (16 mg/kg) 20 min after the end of tMCAO. Plasmatic UA concentration (basal, 2, 4, and 6 h) was determined by spectrophotometry. Six or 24 h after the ischemic insult, the animals were euthanized to obtain the brain according to specific requirements for each determination. Neurofunctional deficit, infarct, and edema were the stroke outcome measures. Infarct volume, with edema correction, was assessed in 2-mm-thick coronal sections stained with 2,3,5-triphenyltetrazolium chloride (TTC). We investigated the effects of UA administration on cerebrocortical UA concentration and biomarkers for relevant features of ischemic stroke pathophysiology: oxidative stress, neuroinflammation, brain swelling, and apoptosis

Full size image

Transient Focal Cerebral Ischemia

Animals were anesthetized by intraperitoneal injection of 5 mg/kg diazepam, 100 mg/kg ketamine, and 0.3 mg/kg atropine. Inhalatory anesthesia was maintained with 0.5–1% sevoflurane in 80% medicinal air plus 20% O2. Transient right middle cerebral artery occlusion (tMCAO, 60 min) was performed by following the intraluminal suture procedure as originally described [33] and adapted to our experimental setup [34]. The procedure included continuous monitoring of cerebrocortical laser-Doppler flow (cortical perfusion, CP), arterial blood pressure and core temperature, and discontinuous measurement of blood glucose at preischemia (basal), ischemia, and reperfusion stages. Buprenorphine (s.c., 0.05 mg/kg) was used to provide analgesia. Six or 24 h after the ischemic insult, the animals were euthanized by intracardiac injection of KCl (200 mg/kg) under anesthesia to obtain the brain according to specific requirements for each determination.

Experimental Groups, Treatments, and Exclusion Criteria

Animals were randomly assigned to two experimental tMCAO groups: vehicle-treated and uric acid (UA)-treated rats. UA (Sigma-Aldrich, Madrid, Spain) was dissolved in Locke’s buffer (vehicle, pH 7.2) and injected through the femoral vein (16 mg/kg) 20 min after the end of tMCAO. This dose was chosen based upon previous studies showing a neuroprotective effect of UA in murine models of ischemic stroke [5,6,7,8,9,10,11, 25]. UA concentrations in plasma (basal, 2, 4, and 6 h) and in homogenized cerebrocortical samples (6 h) were determined by spectrophotometry using commercial Uric acid-LQ kit (Ref. 41,000, Spinreact, Girona, Spain). Fourteen rats were excluded from the study according to predefined criteria, as detailed in Table 1.

Table 1 Animals excluded from the study according to predefined criteria
Full size table

Neurofunctional Evaluation and Measurement of Infarct and Edema Volumes

Twenty-four hours after the ischemic insult, rats were subjected to neurofunctional evaluation before euthanization as reported [9]. The severity of functional deficits was scored with four tests assessing: a spontaneous activity (moving/exploring = 0, no moving or moving only when pushed = 1); b circling to the left (none = 0, circling when elevated by the tail and pushed = 1, circling without displacement, spinning = 2; c parachute reflex: protective abduction of forelimbs (symmetrical = 0, asymmetrical, contralateral forelimb retracted = 1); and d resistance to left forepaw stretching (not allowed = 0, allowed = 1). The total score could range from 0 (no neurological deficits) to 5 (highest neurological deficits) or 6 (death).

Brain infarct volume was determined by the 2,3,5-triphenyltetrazolium chloride (TTC, Sigma-Aldrich) vital staining method [35], followed by morphometric analysis [34]. Briefly, rats were euthanized and the brain was sliced in seven 2-mm-thick coronal sections, which were immersed in a 2% solution of TTC in saline solution at 37 °C for 10 min, fixed in 10% phosphate-buffered formalin (pH 7.4) overnight, and digitally photographed for image analysis. Infarct volume was calculated with edema correction [34], separately for cortical and subcortical regions. The subtraction between raw and corrected infarcted hemisphere equals to the edema volume.

RT-qPCR Analysis of Gene Expression

Rats were euthanized, and a 2-mm-thick brain coronal section (0.2 to − 1.8 mm from bregma) was obtained. Ipsilateral (ischemic) and contralateral (non-ischemic) cortices were dissected, flash-frozen in liquid N2, lyophilized, and grinded to obtain brain powder as reported [36]. For extraction of total RNA, brain powder samples were homogenized (T8.01 Ultra-Turrax, Ika, Staufen, Germany) in TRIzol reagent, centrifuged at 12,000×g for 10 min, the supernatant collected and mixed with chloroform, and centrifuged again. The upper aqueous phase was mixed with isopropanol for RNA precipitation. The isolated RNA (2 μg/lane) was size-fractionated by electrophoresis in a 1% agarose/formalin gel and stained with ethidium bromide to assess RNA quality. RNA concentration was measured in a NanoDrop Lite spectrophotometer (Thermo Scientific, Waltham, MA, USA), and purity was determined by the OD 260/280 ratio.

The cDNA used as template for amplification in the qPCR assay was obtained by the reverse transcription reaction using PrimeScript RT reagent Kit (TaKaRa Bio, Kusatsu, Japan) according to the manufacturer’s protocol, starting with equal amounts of RNA. The RNA level of the genes was analyzed by dsDNA binding dye SYBR Green PCR Master mix in an iQTM5 Multicolor Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Each reaction was run in triplicate, and the melting curves were constructed using Dissociation Curves Software (Bio-Rad) to ensure that only a single product was amplified. The specific primers used are shown in Table 2. The threshold cycle (CT) was determined, and the relative gene expression was expressed as follows: fold change = 2−Δ(ΔCT), where ΔCT = CTtarget − CThousekeeping, and Δ(ΔCT) = ΔCTtreated − ΔCTcontrol. Actb was used as housekeeping gene to normalize samples for the transcription analysis.

Table 2 Oligonucleotides used for RT-qPCR
Full size table

Western Blotting of Protein Expression

For protein extraction, cerebrocortical powder samples were homogenized in HEPES lysis buffer composed of 50 mM HEPES (pH 7.4), 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 10% (w/w) glycerol, 1 mM DTT, 50 mM NaF, and 10 mM Na-pyrophosphate. All products were obtained from Sigma-Aldrich. The lysate was centrifuged at 15,000×g for 15 min, and protein quantification was carried out in the supernatant using BCA protein assay (Pierce, Rockford, IL, USA).

For Western blotting, equal amounts of proteins were dissolved in loading sample buffer under reducing conditions, separated by 12% SDS-PAGE, and transferred to 0.2-μm nitrocellulose membranes. Immunolabeling was carried out with the following primary antibodies from Cell Signaling Technology (Beverly, MA, USA), diluted as 1/1000: anti-NF-κB p65 (ref. 8242), anti-phospho-NF-κB p65 (Ser 536) (ref. 3039), anti-STAT3 (ref. 4904), anti-phospho-STAT3 (Tyr 705) (ref. 9145), anti-Bax (ref. 2772), anti-Bcl-2 (ref. 2762S), and anti-caspase 3 (ref. 9662S). Anti-β-actin mouse monoclonal antibody (1/10000, Sigma-Aldrich, ref. A1978) was used as housekeeping protein. Membranes were incubated for 1 h at room temperature with the appropriate HRP-linked secondary antibody (1∶5000) (Cell Signaling Technology). Blots were visualized using a chemiluminescence detection kit ECL western blotting substrate (Fisher Scientific, Barcelona, Spain), and the signals were captured using a ChemiDoc™ XRS + Imaging System (Bio-Rad).

Determination of Malondialdehyde

Malondialdehyde (MDA), a lipid peroxidation end product, was determined in cerebral cortex according to the modified method based in the thiobarbituric acid reactive species (TBARS) [37]. Briefly, the supernatant (50 μl) of tissue homogenate was mixed with 500 μl of TBA (0.2%)-AcOH 2 M (pH 3.5). Then, samples were incubated for 60 min at 95 °C. After cooling, the precipitate was removed by centrifugation at 1000 rpm for 10 min. The reaction product (TBA2–MDA complex) was measured at 532 nm in a Multiskan Spectrum spectrophotometer (Fisher Scientific). The MDA concentration of the sample was calculated using an extinction coefficient of 1.56 × 105 M−1 cm−1 and expressed relative to total lipids.

Immunohistochemistry

Rats were euthanized, the brain removed, flash-frozen, and cut into sections (18 μm) in the coronal plane (0.2 to − 1.8 mm from bregma), which were fixed on glass slides. Endogenous peroxidase was inhibited for 30 min with 3% H2O2 in PBS, and tissue was permeabilized with sodium citrate buffer pH 6. Non-specific binding was blocked for 1 h with 10% normal goat serum (Sigma-Aldrich, ref. NS02L). The sections were washed with cold PBS containing 0.1% Tween-20 and bound with the Iba-1 primary antibody (ref. 019-19741; Wako, Richmond, VA, USA), incubated at 4 °C overnight. Biotinylated anti-rabbit IgG antibody (BA-1000, Vector Laboratories, Burlingame, CA, USA) was incubated for 1 h at room temperature. After secondary antibody incubation, sections were washed with PBS and then incubated for 30 min with avidin–biotin complex kit (PK6100, Vector Laboratories) and stained using SIGMAFAST 3,3′-diaminobenzidine tablets (Sigma-Aldrich). Sections were then rinsed in water, dried, and coverslipped with Entellan (Merck, Darmstadt, Germany). Sections were visualized through a Leica DM 4500B microscope (Leica Microsystems, Barcelona, Spain) and acquired with a Leica DFC300FX camera (Leica Microsystems).

TUNEL Detection of DNA Cleavage and Caspase Activity Assay

Apoptotic DNA fragmentation was assessed in the cortex of rat brain sections (18 μm) by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL), by following manufacturer’s instructions (In Situ Cell Death Detection Kit, Roche Molecular Biochemicals, Mannheim, Germany). Then, nuclei were stained using DAPI (1:5000, Molecular Probes, Eugene, OR, USA).

Caspase activity was assessed by the specific APO LOGIX™ FAM caspase detection kit (Cell Technology, Mountain View, CA, USA), as previously described [38]. In situ caspase-3 activity was assessed by labeling with FAM-DEVD-FMK, a relatively selective inhibitor of caspase-3 [39]. Sections were dry-mounted and coverslipped with ProLong (Molecular Probes) to reduce fluorescence quenching.

Images were viewed through a fluorescence microscope Leica DM 4500B (Leica Microsystems) and acquired with a Leica DFC 300 FX camera with Leica Application Suite V4. ImageJ NIH program was used for image analysis. The number of TUNEL-positive cells was expressed as a percentage of the corresponding DAPI-positive cells, and for quantitation of caspase-3 activity, the fluorescence was expressed relative to the field area, as usually done in our laboratory [38, 40, 41].

Statistical Analysis

Data are expressed as mean ± SEM, except for neurofunctional scores that are median with the first and third quartiles. Data analysis was performed using GraphPad Prism 6 software (GraphPad Software, CA, USA).

Statistical comparisons were carried out for body weight (unpaired Student’s t test); hemodynamic parameters and core temperature (two-way ANOVA, with treatment and tMCAO stage as independent factors); plasma UA levels (two-way ANOVA, with treatment and time after tMCAO as independent factors, followed by post hoc Holm-Sidak’s multiple comparisons test); cerebrocortical UA concentration (two-way ANOVA, with treatment and brain hemisphere as independent factors, followed by post hoc Holm-Sidak’s multiple comparisons test); neurofunctional score (Mann-Whitney test); total brain infarct and edema volumes (unpaired Student’s t test); cortical/subcortical infarct sizes (two-way ANOVA, with treatment and brain area as independent factors, followed by post hoc Fisher multiple comparisons test); mRNA expressions and MDA levels (two-way ANOVA, with treatment and brain hemisphere as independent factors, followed by post hoc Holm-Sidak’s multiple comparisons test); and apoptosis biomarkers (unpaired Student’s t test). Differences were considered significant at P < 0.05.

Results

Body Weight, Physiological Parameters, and Uric Acid Levels

Body weight values were similar in the two experimental groups: vehicle-treated and UA-treated rats. The reduction of CP during ischemia was comparable in the two groups, as was the extent of perfusion increase during reperfusion. Mean arterial blood pressure, core temperature, and serum glucose were in the physiologic range, remained stable along the ischemia-reperfusion procedure, and were similar in the two experimental groups (Table 3).

Table 3 Sample characteristics of the experimental groups
Full size table

Plasma UA levels in rats subjected to tMCAO were 0.49 ± 0.13 mg/dL at reperfusion and did not significantly change during the following 6 h. Intravenous UA injection (16 mg/kg), 20 min after the end of tMCAO, produced a significantly (P < 0.01) dramatic increase in plasma UA levels at 2 h, which decreased at 4 h, and returned to basal levels at 6 h of reperfusion (Fig. 2a). Cerebrocortical UA concentration was similar in the ipsilateral ischemic hemisphere (233 ± 47 ng/mg protein) and the contralateral hemisphere (174 ± 34 ng/mg protein) of vehicle-treated rats at 6 h after tMCAO. At this time point, UA concentration was not significantly modified in the contralateral hemisphere of UA-treated rats (203 ± 56 ng/mg protein). By contrast, UA concentration significantly increased in the ipsilateral ischemic hemisphere of UA-treated rats (727 ± 73 ng/mg protein), when compared with both the contralateral hemisphere of UA-treated rats (P < 0.01) and the ipsilateral ischemic hemisphere of vehicle-treated rats (P < 0.01).

Fig. 2
figure 2

Effects of uric acid (UA) administration on neurological function and brain infarct volume after ischemic stroke in rats subjected to tMCAO. a Time course of plasma UA levels after tMCAO. **Significantly different from vehicle, P < 0.01; ##significantly different from 0 h time, P < 0.01 (two-way ANOVA). b Severity of neurofunctional deficits, 24 h after tMCAO, scored with four tests assessing spontaneous activity, circling to the left, parachute reflex symmetry, and resistance to left forepaw stretching. **Significantly different from vehicle, P < 0.01 (Mann-Whitney test). c Representative images showing cortical and subcortical brain infarct (pale areas) in TTC-stained 2-mm-thick brain coronal sections (0.2 to − 1.8 mm from bregma), 24 h after tMCAO. d Brain infarct volume calculated with edema correction, separately for cortical and subcortical regions, as determined by morphometric analysis of seven 2-mm-thick TTC-stained coronal sections. *Significantly different from vehicle, P < 0.05 (two-way ANOVA). tMCAO transient middle cerebral artery occlusion. TTC 2,3,5-triphenyltetrazolium chloride. Plasma UA levels and infarct volume data are mean ± SEM, and neurofunctional scores are median with the first and third quartiles, from 4–6 vehicle-treated and 4–6 UA-treated rats

Full size image

Uric Acid Administration After tMCAO Improves Neurological Function and Reduces Brain Infarct Volume

Neurofunctional evaluation at 24 h after tMCAO showed significantly (P < 0.01) lower neurofunctional impairment scores in UA-treated rats, when compared with vehicle-treated rats (Fig. 2b). In accordance with improved neurofunctional outcome, UA treatment reduced the size of brain infarct (Fig. 2c). Total infarct volume, corrected for edema, significantly (P < 0.05) decreased from 29.53 ± 5.14% of the ipsilateral ischemic hemisphere in vehicle-treated rats to 17.89 ± 3.56% in UA-treated rats. There was a region-selective protective effect of UA, which significantly (P < 0.05) reduced infarct volume of the cortical region rather than the subcortical region (Fig. 2d).

Uric Acid Treatment Reduces tMCAO-Induced Early Vegfa Overexpression and Subsequent Brain Edema

In addition to the reduction of cortical infarct, UA administration significantly (P < 0.05) reduced brain edema volume at 24 h after tMCAO (Fig. 3a). To elucidate molecular mechanisms involved in the effects of UA on brain edema, we examined the time course of expression levels of Vegfa, Mmp9, and Timp1 genes in brain cortex.

Fig. 3
figure 3

Effects of uric acid (UA) administration on brain edema volume and expression of genes involved in brain swelling after ischemic stroke in rats subjected to tMCAO. a Brain edema volume, 24 h after tMCAO, determined as the subtraction between raw and edema-corrected infarcted hemisphere by morphometric analysis of seven 2-mm thick TTC-stained coronal sections. *Significantly different from vehicle, P < 0.05 (unpaired Student’s t test). b Time course of Vegfa gene expression in brain cortex. Vegfa mRNA levels were determined by RT-qPCR analysis, using Actb as housekeeping gene. **Significantly different from contralateral non-ischemic cortex, P < 0.01; #significantly different from vehicle, P < 0.05; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). c Time course of Mmp9/Timp1 gene expression ratio in ischemic brain cortex. Mmp9 and Timp1 mRNA levels were determined by RT-qPCR analysis, using Actb as housekeeping gene. &Significantly different from 6 h after tMCAO, P < 0.05; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). tMCAO transient middle cerebral artery occlusion. TTC 2,3,5-triphenyltetrazolium chloride. Data are mean ± SEM from 4–6 vehicle-treated and 4–6 UA-treated rats

Full size image

Vegfa mRNA levels significantly (P < 0.01) increased in the ipsilateral ischemic hemisphere at 6 h after tMCAO, when compared with the contralateral hemisphere of vehicle-treated rats. By contrast, this increase was not observed in UA-treated rats (Fig. 3b). On the other hand, expression levels of Vegfa were similar in the ipsilateral and contralateral hemispheres of vehicle-treated rats at 24 h after tMCAO. However, at this time point, UA treatment induced a significant (P < 0.01) Vegfa overexpression in the ipsilateral ischemic hemisphere (Fig. 3b).

Mmp9/Timp1 expression ratio was significantly (P < 0.05) increased in the ipsilateral ischemic hemisphere at 24 h after tMCAO, when compared with 6 h after the transient ischemic insult. Interestingly, UA treatment significantly (P < 0.01) decreased the rise in Mmp9/Timp1 expression ratio at 24 h after tMCAO, when compared with vehicle treatment (Fig. 3c).

Uric Acid Treatment Is Associated to IL-6/STAT3 Activation After tMCAO

In order to profile the inflammatory response in the cortex of ischemic brain, we first analyzed the mRNA expression levels of Il6, Tnfa and Il1b cytokine genes after tMCAO. Transient ischemia induced significant (P < 0.05) upregulation of Tnfa and Il1b mRNA expression, both at 6 h and 24 h, in the ipsilateral ischemic hemisphere, when compared with the contralateral hemisphere. By contrast, a significant (P < 0.01) increase in Il6 expression was only observed at 24 h post-tMCAO (Fig. 4a). Interestingly, UA-treated rats exhibited a significantly (P < 0.01) dramatic increase of Il6 expression in the ipsilateral ischemic hemisphere at 6 h after tMCAO, a significant (P < 0.01) decrease in Tnfa expression at 24 h, and no significant change in Il1b expression, when compared with vehicle-treated rats (Fig. 4a).

Fig. 4
figure 4

Effects of uric acid (UA) administration on brain expression of cytokine genes, transcription factor proteins, and cytokine transcription regulator gene, as well as microglia activation after ischemic stroke in rats subjected to tMCAO. a Time course of Il6, Tnfa, and Il1b gene expressions in brain cortex. Cytokine mRNA levels were determined by RT-qPCR analysis, using Actb as housekeeping gene. *Significantly different from contralateral non-ischemic cortex, P < 0.05; **significantly different from contralateral non-ischemic cortex, P < 0.01; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). b Representative images of p-STAT3 protein expression time course, relative to total STAT3, in brain cortex. Protein expression was shown by Western blotting, using β-actin as housekeeping protein for loading control. c Time course of Socs3 gene expression in brain cortex. Socs3 mRNA levels were determined by RT-qPCR analysis, using Actb as housekeeping gene. *Significantly different from contralateral non-ischemic cortex, P < 0.05; **significantly different from contralateral non-ischemic cortex, P < 0.01; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). d Representative images of p-p65 protein expression time course, relative to total p65, in brain cortex. Protein expression was shown by Western blotting, using β-actin as housekeeping protein for loading control. e Representative photomicrographs of microglia appearance in ischemic cortex, 24 h after tMCAO (Scale bar = 100 μm). Reactive amoeboid shaped and quiescent ramified microglia were shown by immunohistochemical staining of the microglia-specific calcium-binding protein Iba-1. tMCAO transient middle cerebral artery occlusion. Data are mean ± SEM from 4–6 vehicle-treated and 4–6 UA-treated rats

Full size image

Levels of phospho-STAT3, a transcription factor dependent on IL-6, were analyzed as a possible neuroprotective pathway. Although tMCAO elicited a slight increase of phospho-STAT3 levels in the ipsilateral ischemic hemisphere, when compared with the contralateral hemisphere, it is noteworthy that UA treatment triggered an intense upregulation of phospho-STAT3 protein at 6 h and 24 h after the transient ischemic insult (Fig. 4b). In close relation to IL-6/STAT3 signaling pathway, we studied Socs3 mRNA expression, a negative regulator of cytokine signaling. Again, although transient ischemia significantly (P < 0.05) increased Socs3 mRNA at both 6 h and 24 h, UA-treated rats exhibited a remarkable induction of Socs3 gene expression at 24 h after tMCAO in the ipsilateral ischemic hemisphere, significantly (P < 0.01) higher than in vehicle-treated rats (Fig. 4c).

Western blot analysis also showed changes in phosphorylation levels at the Ser536 site located in the transactivation domain of the p65 subunit of transcription factor NF-κB, a master regulator of the inflammatory response. There was a slight increase of phospho-p65 levels in the ipsilateral ischemic hemisphere at 6 h after tMCAO, when compared with the contralateral hemisphere. On the other hand, at this time point, there were markedly enhanced levels of phospho-p65 in the contralateral hemisphere and, to a higher extent, in the ipsilateral ischemic hemisphere of UA-treated rats (Fig. 4d). However, at 24 h after tMCAO, levels of phospho-p65 were higher in ipsilateral ischemic hemispheres, when compared with contralateral hemispheres, in both vehicle-treated and UA-treated rats (Fig. 4d).

Moreover, we evaluated the activation of microglia, given its central role in neuroinflammation. Immunohistochemical analysis of Iba-1, a microglia-specific calcium-binding protein, showed reactive amoeboid shaped microglia at 24 h after tMCAO in the ipsilateral ischemic hemisphere of vehicle-treated rats. By contrast, quiescent ramified microglia was observed in UA-treated rats (Fig. 4e).

Uric Acid Treatment Enhances Antioxidant Defense Gene Expression to Reduce Lipid Peroxidation After tMCAO

In order to assess whether UA-driven induction of IL-6/STAT3 pathway modulated transcriptional activity of antioxidant defense genes to control the levels of ROS associated with ischemic stroke, we evaluated Sod1, Sod2, and Cat mRNA expression. Neither the transient ischemic insult nor the UA treatment modified the cerebrocortical expression of these antioxidant defense genes at 6 h after tMCAO. UA treatment significantly (P < 0.05) enhanced expression of Sod2 and Cat genes in the ipsilateral ischemic hemisphere at 24 h post-tMCAO, when compared with vehicle treatment (Fig. 5a). Then, in order to assess the effects of UA-elicited overexpression of antioxidant defense genes on oxidative damage, malondialdehyde (MDA) was measured as a lipid peroxidation biomarker. In accordance with the time course of Sod2 and Cat transcriptional data, UA administration significantly (P < 0.01) restrained the elevated MDA levels induced by the transient ischemic insult in the ipsilateral hemisphere of vehicle-treated rats at 24 h after tMCAO (Fig. 5b).

Fig. 5
figure 5

Effects of uric acid (UA) administration on the brain expression of antioxidant defense genes and lipid peroxidation after ischemic stroke in rats subjected to tMCAO. a Time course of Sod1, Sod2 and Cat gene expressions in brain cortex. Antioxidant enzyme mRNA levels were determined by RT-qPCR analysis, using Actb as housekeeping gene. *Significantly different from contralateral non-ischemic cortex, P < 0.05; **significantly different from contralateral non-ischemic cortex, P < 0.01; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). b Time course of the lipid peroxidation biomarker malondialdehyde (MDA) content in brain cortex. MDA levels were determined by the method based in the thiobarbituric acid reactive species (TBARS). **Significantly different from contralateral non-ischemic cortex, P < 0.01; and ##significantly different from vehicle, P < 0.01 (two-way ANOVA). tMCAO transient middle cerebral artery occlusion. Data are mean ± SEM from 4–6 vehicle-treated and 4–6 UA-treated rats

Full size image

Uric Acid Treatment Diminishes tMCAO-Driven Apoptosis by Modulating the Expression of Key Regulatory and Executioner Proteins

Since apoptosis is involved in ischemic brain damage, we also investigated the modulatory effect of UA on key regulators/executioners of the cell death program and on resulting apoptosis in brain cortex. As expected, western blotting at 24 h after tMCAO revealed reduced levels of the anti-apoptotic protein Bcl-2, increased levels of the proapoptotic protein Bax, and increased levels of the apoptosis executioner protease cleaved caspase-3 in the ipsilateral ischemic hemisphere of vehicle-treated rats, when compared with the contralateral hemisphere (Fig. 6a). Carboxyfluorescein in situ labeling confirmed activation of capase-3 in the ipsilateral ischemic hemisphere of vehicle-treated rats at this time point (Fig. 6b). Interestingly, UA treatment induced the expression of Bcl-2 and reduced the levels of Bax, as well as of cleaved caspase-3, when compared with vehicle treatment (Fig. 6a). Again, a significant (P < 0.05) inhibitory effect of UA on caspase-3 activation was confirmed by in situ labeling (Fig. 6b). With regard to cell apoptosis execution, the presence of TUNEL-positive nuclei observed at 24 h after tMCAO in the ipsilateral ischemic hemisphere of UA-treated rats was significantly (P < 0.01) lower than in the corresponding hemisphere of vehicle-treated rats (Fig. 6c).

Fig. 6
figure 6

Effects of uric acid (UA) administration on brain apoptotic cell death after ischemic stroke in rats subjected to tMCAO. a Representative images of antiapoptotic protein Bcl-2, proapoptotic protein Bax, and apoptosis executioner protease cleaved caspase-3 (relative to total caspase-3) expression in brain cortex. Protein expression was shown by Western blotting, using β-actin as housekeeping protein for loading control. b Quantification (average of three selected fields, black circle) and representative photomicrographs of activated caspase-3 positive cells, 24 h after tMCAO, as shown by carboxyfluorescein in situ labeling (Scale bar = 100 μm). *Significantly different from vehicle, P < 0.05 (unpaired Student’s t test). c Quantification (average of three selected fields, black circle) and representative photomicrographs of apoptotic DNA fragmentation, 24 h after tMCAO, as shown by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) (Scale bar = 100 μm). **Significantly different from vehicle, P < 0.01 (unpaired Student’s t test). tMCAO transient middle cerebral artery occlusion. Data are mean ± SEM from 3 vehicle-treated and 3 UA-treated rats

Full size image

Discussion

In this study, male Wistar rats subjected to transient ischemic stroke benefited from i.v. uric acid (UA) treatment administered early after reperfusion. Rats receiving UA showed reduced infarct and edema volumes, less brain apoptosis, and improved neurofunctional status. Our group and others have previously demonstrated neuroprotective properties of exogenous UA in ischemic stroke using different rodent models [4,5,6,7,8,9,10,11]. However, the mechanisms responsible for this beneficial effect are not yet completely elucidated. We now show, for the first time, that UA favorably targets oxidative stress, neuroinflammation, brain swelling, and apoptotic cell death in the cortical region after tMCAO in association with IL-6/STAT3 neuroprotective signaling pathway.

How systemically administered UA reaches its brain targets in ischemic stroke models is a matter of debate. It has been suggested that UA is primarily a vasculoprotective compound and that this effect may be ultimately responsible for the overall neuroprotection and improved outcome shown in experimental studies [24]. Our results show that i.v. UA injection induced a transient increase in plasma UA levels, which peaked at 2 h and vanished at 6 h of reperfusion. Interestingly, systemic UA administration also induced a selective increase of UA concentration in the injured, but not in the healthy brain, 6 h after the ischemic insult. Therefore, UA indeed passed the functionally compromised blood–brain barrier (BBB), entered to the brain, and played its protective role in the ischemic area. A previous report also in rats subjected to tMCAO [10] showed that i.v. administered UA did not result in increased UA brain concentration, although measured at much earlier time points (10–30 min) when compared with the present study. Of note, methyl- and sulfur-containing analogs of UA with increased solubility were detected in the brain up to 120 min after i.v. administration to mice [42].

Our study analyzed the cerebrocortical mRNA expression levels of Tnfa, Il6, and Il1b cytokine genes after tMCAO and the effect of UA treatment. The ischemic insult increased the expression of Tnfa and Il1b, while UA reduced the expression of Tnfa but not of Il1b. Concomitant cortical expression of Tnfa and Il1b mRNAs has been reported as part of the neuroinflammatory response in transient focal ischemia [43], and downregulation of Tnfa by UA would therefore contribute to anti-inflammation. Most interestingly, in the present study, UA treatment induced marked upregulation of Il6 gene expression in rats subjected to transient focal ischemia. Previous reports showed that permanent MCAO in rats induced Il6 mRNA expression [44] and IL-6 bioactivity [45] in the ischemic hemisphere. Moreover, local i.c.v. IL-6 infusion reduced ischemic brain damage after permanent MCAO in rats [45] and had a trophic effect on hippocampal neurons in gerbils subjected to forebrain ischemia [46]. The source for the increase in levels of IL-6 early after stroke has long been debated. After experimental cerebral ischemia, local upregulation of IL-6 has been a consistent finding, mostly not only in neurons but also in glial cells and vascular endothelium [47, 48]. Based on these in vivo premises, Gertz and colleagues attempted to identify the exact cellular source of IL-6 in brain after in vitro ischemia, using pure and mixed cell cultures, and organotypic brain slices [49]. They concluded that the upregulation of IL-6 expression in the brain after ischemia takes the form of a self-amplifying network response of all resident cells. In our study, UA could increase this network response, rather than acting on a single cell type.

Our results show that, in parallel to Il6 mRNA upregulation, UA treatment also triggered an intense overexpression of cerebrocortical phospho-STAT3 transcription factor protein after the transient ischemic insult. Previous studies showed that neuroprotection by endogenous IL-6 is mediated by STAT3 activation after ischemic stroke in mice [29, 30]. Therefore, altogether, our results suggest a role for the IL-6/STAT3 signaling pathway in UA elicited reduction of ischemic brain damage after stroke in rats, which confirmation by using pathway inhibitors warrants further research. On the other hand, IL-6 receptor (IL-6R) is expressed in neurons and microglia [50, 51], but not in astrocytes [50]. Considering that IL-6 promotes STAT3 phosphorylation through its binding to IL-6R, STAT3 activation is likely to occur in neurons rather than in microglia. In fact, the blockade of IL-6 signaling exacerbated ischemic cerebral damage in mice through the reduction in p-STAT3 in peri-infarct cortical neurons [29]. By contrast, UA could reduce brain damage by activating the IL-6/STAT3 pathway in neurons, in line with neuronal STAT3 activation mediating neuroprotection induced by estradiol [52] and the PPAR-γ agonist pioglitazone [53] after ischemic stroke.

NF-κB is another well-characterized transcription factor with dual roles in cell death and survival in neurological disorders and implications in neuroprotective therapy [54]. In the present study, we measured levels of total p65 and phospho-p65 (Ser536) subunit of NF-κB in whole tissue extract. Phosphorylation of the Ser536 site—located in the p65 transactivation domain—is considered one of the most important inducible phosphorylations of NF-κB [55], and it has been associated with its nuclear translocation and with the transcriptional activation of its target genes [56,57,58,59]. Here, we report enhanced levels of phospho-p65 (Ser536) subunit of the transcription factor NF-κB in the ipsilateral ischemic hemisphere of UA-treated rats compared with vehicle-treated rats, at 6 h but not at 24 h after transient ischemia. These results support a selective transient activation of NF-κB in the injured brain of UA-treated rats after ischemic stroke. It is known that persistent neuronal NF-κB activation contributes to cerebral ischemic damage after permanent MCAO [60]. However, transient NF-κB signaling in cerebral ischemia is also responsible for beneficial effects [61] such as preconditioning development [62] and apoptosis inhibition [63]. Indeed, NF-κB activation enhances anti-apoptotic protein levels and thus promotes neuron survival [64]. On the other hand, diminished oxidative stress increases NF-κB related rapid defenses, such as immune response and anti-apoptotic factors, thus preventing brain damage after tMCAO [65]. We suggest that transient NF-kB activation elicited by UA may transcriptionally induce different immune factors such as the neuroprotective cytokine IL-6. In turn, high levels of IL-6 would induce STAT3 activation.

Interestingly, NF-κB and STAT3 cooperatively modulate the transcriptional regulation of a variety of anti-apoptotic genes [66]. On the other hand, nuclear translocation of STAT3 and NF-κB are independent of each other but NF-κB supports Il6-induced expression and activation of STAT3 [67]. Our data show that treatment with UA increased the cerebrocortical levels of both phospho-p65, subunit of NF-κB, and phospho-STAT3 after the transient ischemic insult. It has been reported that IL-32α exerts a neuroprotective role in mice subjected to tMCAO by cytokine signaling through both STAT3 and NF-κB pathways [68]. Similarly, our results support that the beneficial effects of UA associated with the IL-6/STAT3 signaling pathway also involve NF-κB activation.

Socs3 gene expression is activated by the IL-6/STAT3 signaling pathway [69], and SOCS3 protein is a transcription regulator with potential impact on cytokine production and microglia activation in neuropathological conditions including stroke [70]. The present study observed induction of cerebrocortical Socs3 expression by tMCAO, in line with gene expression analysis after a similar ischemic insult in SHR rats [71] and mice [72]. Moreover, our results show a dramatic Socs3 overexpression in UA-treated rats, thus supporting a role for the IL-6/STAT3/SOCS3 signaling pathway in neuroprotection by UA, in accordance with previous studies showing neuroprotective effects of SOCS3 induction in rodent ischemic stroke [71, 72].

Critical to the process of neuroinflammation is the role of microglia in pathological progression of ischemic tissue. In focal ischemia, the activated resident microglia proliferates and accumulates in the peri-infarct region, adapting an amoeboid morphology, instead of the characteristic ramified morphology of the inactive phenotype [73]. In our model, we have verified that UA administration maintains the ramified morphology of Iba-1-positive microglia in brain cortex, which suggests an anti-inflammatory role of UA during ischemia. However, beyond morphological transformation, reactive microglia display polarization functional phenotypes with beneficial or detrimental contributions to post-stroke recovery [74].

Restoration of blood flow in transient ischemic stroke contributes to overproduction of ROS, mainly by the mitochondria, and therefore to the presence of oxidative damage in brain tissue [75], setting the rationale for the use of antioxidants as neuroprotective therapy [76, 77]. Previous studies found that the neuroprotective capacity of UA in rat ischemic stroke was associated to reduced brain ROS production and decreased lipid, protein, and DNA oxidative stress biomarkers [7, 10, 11, 25]. Our results show that UA treatment enhances cerebrocortical Sod2 and Cat gene expression after induction of stroke. In mice subjected to ischemic stroke, STAT3 modulates brain Mn-SOD (i.e., mitochondrial SOD2) expression and superoxide accumulation [78], and IL-6-induced STAT3 activation promotes Mn-SOD expression [30], thus supporting a role for the IL-6/STAT3 pathway in UA-induced expression of the first line defense antioxidant enzymes. This STAT3 role adds to recently reported involvement of Nrf2 transcription factor in UA-elicited protection against reperfusion oxidative stress in ischemic stroke [25]. We also observed a decrease in the levels of the lipid peroxidation biomarker MDA in the cortex of UA-treated rats, in line with the link between SOD/catalase and MDA levels reported in rats subjected to forebrain ischemia [79].

Free radicals are triggers of both cytotoxic and vasogenic edema after stroke, with VEGF and matrix metalloproteinases as recognized mediators of vasogenic brain swelling [80]. In the acute phase of ischemic stroke, VEGF-A is involved in the development of cerebral edema due to endothelial cell–cell junction breakdown and subsequent vascular leak [81]. Our results show that UA represses the induction of cerebrocortical Vegfa expression in the acute stage (6 h) of stroke, and thus, this would explain the reduced edema observed in rats subjected to UA treatment after the ischemic insult. Noteworthy, a recent study showed that UA neuroprotective treatment normalized the increased circulating VEGF-A levels and BBB breakdown after ischemic stroke in SHR rats and reduced VEGF-A levels in hypertensive acute ischemic stroke patients from the URICO-ICTUS trial [26]. Moreover, VEGF-A has a Janus face in stroke pathophysiology: after its transient detrimental effects in the acute phase, late increased VEGF-A promotes neuronal protection as well as angiogenesis and functional recovery [82,83,84]. Importantly, at 24 h after stroke onset, we observed that Vegfa gene expression was upregulated by UA administration, which might further contribute to the beneficial effects of UA by promoting VEGF-A-induced neuroprotection. Delayed VEGF treatment also enhances angiogenesis and recovery after ischemic stroke in neonatal [83] and adult [84] rats. Whether UA treatment beyond 24 h from stroke onset would have additional beneficial effects by inducing angiogenesis through the IL-6/STAT3/VEGF-A signaling pathway [85] deserves further research.

VEGF-induced increase in BBB permeability in acute ischemic stroke is associated with upregulation of matrix metalloproteinases in rats [86, 87] and mice [88]. However, extracellular matrix homeostasis is maintained by a balance between proteolytic and anti-proteolytic factors [89], including MMP-9 and its natural inhibitor TIMP-1 [90, 91]. Because it is this balance that contributes to impairment of the BBB, the MMP-9/TIMP-1 ratio has been used to provide a closer assessment of the proteolytic potential of MMP-9 to degrade the extracellular matrix after ischemic stroke in rodents [92] and humans [93]. Interestingly, the increase of p-STAT3 triggered by neuroprotective leptin in cortical neurons after ischemic stroke is associated with an augmented expression of TIMP-1 [94]. Our data reveals that UA treatment decreases MMP-9/TIMP-1 ratio in brain cortex, thus suggesting that UA administration may contribute through early downregulation of the VEGF-A/MMP-9 pathway to restrict vascular permeability by improving BBB integrity in acute ischemic stroke.

Cell apoptosis plays a crucial role in the development of stroke brain damage in the ischemic penumbra, where the cell fate is determined by the expression of a number of proapoptotic and anti-apoptotic signaling proteins [95]. Among these proteins, upregulation of proapoptotic Bax and downregulation of anti-apoptotic Bcl-2 is a molecular landmark in neurons destined to apoptosis. On the other hand, caspase-3 is identified to act downstream Bax/Bcl-2 ratio and play a key role in apoptosis execution [96]. We have recently reported that blockade of caspase-3-induced apoptosis in involved in the beneficial effect of neuroprotective treatments for ischemic stroke [38, 40]. In the present study, the increased expression of Bax, decreased expression of Bcl-2, and activation of caspase-3 were detected in the cerebrocortical regions of animals subjected to ischemic stroke. However, UA treatment ameliorated the Bax/Bcl-2 ratio and consequently induced a reduction in cleaved caspase-3 and carboxyfluorescein in situ labeling of activated caspase-3, as well as a decrease in TUNEL-positive cell counts. Interestingly, STAT3 can promote cell survival by means of transcriptional regulation of apoptotic signaling proteins, such as the Bcl-2 family [97]. In rodent ischemic stroke, brain STAT3 activation by neuroprotective treatments has been associated with increased Bcl-2 expression and decreased caspase-3 activity [98], as well as with reduced number of TUNEL-positive cells [98, 99]. Therefore, the neuroprotective anti-apoptotic effect of UA could be due to activation of the IL-6/STAT3 signaling pathway.

In conclusion, the present study demonstrates for the first time that systemic (i.v.) UA treatment after ischemic stroke, which increases local UA concentration in the injured brain tissue, is associated to activation of the IL-6/STAT3 pathway, attenuation of VEGF-A/MMP-9 signaling, and modulation of relevant mediators of oxidative stress, neuroinflammation, and apoptotic cell death in brain cortex, thus leading to neuroprotection. Further research is warranted to identify the cell type/s targeted by UA. These findings extend the preclinical evidence supporting the neuroprotective effects of UA and point toward a multifaceted mechanism targeting ischemic stroke damage, with transcription factors such as STAT3 as key players. By targeting key steps of the ischemic cascade in a pleiotropic way, UA is a better adjunctive agent in endovascular treatments, compared with those based on monotargeted approaches. We are aware that these encouraging results should be confirmed in animals with stroke comorbidities, in females, and in old animals, to better model the human stroke population. Nonetheless, the present findings let us suggest that activation of the IL-6/STAT3 signaling pathway is a promising target in acute ischemic stroke therapy.