Indoxyl Sulfate Affects Glial Function Increasing Oxidative Stress and Neuroinflammation in Chronic Kidney Disease: Interaction between Astrocytes and Microglia

1Neurodegenerative diseases have become a growing health burden and, in our aging population, are often linked to other comorbidities. Oxidative stress and neuroinflammation contribute to the pathogenesis of neuronal degeneration and can cause cell membrane damage from lipid peroxidation, changes in protein structure and function, due to protein oxidation, and structural DNA damage, hallmarks of several neurodegenerative diseases. The central nervous system (CNS) is particularly sensitive to oxidative stress, probably because of its high oxygen demand and the presence of polyunsaturated fatty acids and low levels of glutathione (GSH;; ). Increasing reactive oxygen species (ROS) production can exacerbate the expression of inflammatory mediators as detected in patients with neurodegenerative diseases.

2Chronic kidney disease (CKD) is characterized by a progressive loss of renal function that, in its terminal phase, shows signs and symptoms of uremic syndrome. Patients with CKD have many comorbidities such as immune disorders, with the coexistence of immunodeficiency and immune activation, and neurological complications that largely contribute to the morbidity and mortality of this disease. CKD is frequently associated with cognitive impairment and, among patients with terminal CKD receiving haemodialysis, more than 85% have cognitive deficits. Cognitive impairment in CKD is also associated with a poorer clinical outcomes. Patients with CKD are also at higher risk of cognitive decline and even dementia. Causes of cognitive impairment in CKD are multifactorial and they include cerebrovascular disease, renal anemia, secondary hyperparathyroidism, dialysis disequilibrium, and uremic toxins accumulation. Plasmatic levels of uremic toxins increase as CKD progresses, and they are believed to be the main cause of cognitive impairment. However, the exact role or mechanism of uremic toxins in cognitive disorders has not been determined yet. One of the most important uremic toxins is indoxyl sulfate (IS), a protein-bound uremic toxin, which is not effectively eliminated by dialysis. IS is a nephro-vascular toxin that causes nephrotoxicity especially on tubular cells, inhibits proliferation of endothelial cells and is an inducer of free radicals. Moreover, it has been reported that IS enhances inflammatory response and ROS in LPS-stimulated macrophages. Among various uremic toxins, IS is a likely candidate capable to trigger cerebral dysfunction in kidney disease. Therefore, we chose to investigate the effects of IS on glial cells and the impact on neuronal survival, all primary aspects involved in CNS homeostasis.

3All reagents and compounds, unless stated otherwise were purchased from Sigma Chemicals Company (Sigma, Milan, Italy).

4C6 glioma cell line was obtained from American Type Culture Collection (ATCC; Manassas, VA, United States). C6 were cultured in DMEM, 10% FBS (mL/L), penicillin/streptomycin (100 units/0.1 mg/mL) and 2 mML-glutamine at 37°C in 5% CO2 atmosphere and passaged at confluence using a solution of 0.025% trypsin and 0.01% EDTA. This cell line was originally derived from rat brain tumors and have oligodendrocytic, astrocytic and neuronal properties. C6 cells are widely used as an astrocyte-like cell line.

5Cultures of mixed glial cell from cortex were prepared from postnatal days 1–2 mouse pups (Female C57BL/6 mice; Harlan Laboratories, Udine, Italy). Mice were housed under specific pathogen-free conditions and fed with standard chow diet at the University of Messina, Department of Chemical, Biological, Pharmaceutical and Environmental Sciences. The animal experiments were performed according protocols following the Italian and European Community Council for Animal Care (DL. 26/2014). Cerebral cortices were excised, meninges, olfactory bulb and thalami removed, and the hemispheres were transferred to petri dishes containing HBSS and were cut into four small pieces. Brains were centrifuged for 1 min at 200–300 g. The supernatant was removed and the pellet was incubated with HBSS/10 mM HEPES buffer, 0.5 mg/ml Papain, 10 μg DNAse solution for 25 min at 37°C. The extracted cells were centrifuged for 5 min at 200–300 g and the pellet was resuspend in BME medium (10% FBS and 0.5% penicillin/streptomycin). The cell suspension was filtered through a 70-μm cell strainer to remove debris. The extracted cells were suspended in BME medium (10% FBS and 0.5% penicillin/streptomycin) in 75 cm3 flasks. The medium was changed after 48 h and then twice per week. After 20 days, in some flasks, to obtain only astrocytes in the culture, microglia were dislodged using an orbital shaker (200 rpm for 1 h, 37°C). Moreover, in order to further remove residual microglia from the remaining cell monolayers, it was used a 60-min exposure (50 mM) to the lysosomotropic agent Leu-Leu-OMe (<5% microglia, referred to some microglial cells not dethached from the treatments, was deteced by flow cytometry using anti-Iba1 as antibody; ).

6Dissociated cell cultures of mouse hippocampus and cortex were established from day 16 C57B/6J mouse embryos, as previously described. Hippocampal and cortical neurons were plated in 35, 60, or 100-mm diameter polyethylenimine-coated plastic dishes. Primary neurons were maintained in Neurobasal medium containing 25 mM of glucose, B-27 supplement (Invitrogen), 0.001% gentamycin sulfate, 2 mML-glutamine, and 1 mM HEPES (pH 7.2) at in 5% CO2 atmosphere 37°C. Approximately 95% of the cells in such cultures were neurons and the remaining cells were astrocytes.

7C6 cells and primary astrocytes and mixed glial cell cultures were plated 24 h before the experiments. The cellular medium was then replaced with fresh medium and cells were treated with IS (15–60 μM) for 24 h in all experiments, except for NF-kB and Nrf2 evaluation and AhR activation, where IS was added to cells for 20 min and 1 h, respectively.

8Primary hippocampal and cortical neuronal cultures were plated for 2 weeks before the experiments. Then the cells were treated with IS (15–60 μM) for 24 h. For the experiments, we considered the list of uremic toxins provided by the European Uremic Toxin Work group and thus used the IS concentration range found in the cerebrospinal fluid of CKD patients.

9Reactive oxygen species production was evaluated by the probe H2DCF-DA as previously reported. H2DCF, in presence of ROS, is rapidly oxidized to the highly fluorescent DCF. C6 (3.0 × 105 cells/well) and primary astrocytes and mixed glial cell cultures (1.5 × 105 cells/well) were plated into 24-well plates and then IS (15–60 μM) was added. After 24 h cells were collected, washed with PBS and incubated in PBS containing H2DCF-DA (10 μM) at 37°C. Cellular fluorescence was evaluated using fluorescence-activated cell sorting analysis (FACSscan; Becton Dickinson) and elaborated with Cell Quest software. In some experiments, in C6 cells, either DPI (10 μM), that has frequently been used to inhibit ROS production mediated by flavoenzymes, or NAC (2 mM), a free radicals scavenger as well as a major contributor to maintenance of the cellular GSH, were added 1 h before IS. In other experiments, in C6 cells, PDTC (200 μM) or CH-223191 (1 μM), a ligand-selective antagonist of the AhR, were added 1 h before IS.

10For immunofluorescence assay, C6 cells (3.0 × 105/well), primary astrocytes and mixed glial cells (2.0 × 105/well) were seeded on coverslips in 12-well plate and treated for 1 h with IS (30 μM). In some experiments with C6 cells, DPI (10 μM) and NAC (2 mM) were added 1 h before IS. In other experiments, CH-223191 (1 μM), was added 1 h before IS to C6 cells. Then cells were fixed with 4% paraformaldehyde in PBS and permeabilized with 0.1% Triton X-100 in PBS. After blocking with BSA and PBS, cells were incubated with rabbit anti-Nrf2 antibody (Santa Cruz Biotechnologies; sc-722; used at diluition 1:250), with mouse anti-AhR antibody (Abcam; ab2769; used at diluition 1:250) or with rabbit anti-p65 antibody (Santa Cruz Biotechnologies; sc-372; used at diluition 1:250). The slides were then washed with PBS for three times and fluorescein-conjugated secondary antibody (Immuno Reagents; used at diluition 1:2000) was added for 1 h. DAPI was used for counterstaining of nuclei. Coverslips were finally mounted in mounting medium and fluorescence images were caught using the Laser Confocal Microscope (Leica TCS SP5) as previously reported.

11C6 cells (5.0 × 104/well), primary astrocytes and mixed glial cells (3.5 × 104/well) were seeded on 96-well plate and treated, after 24 h, with IS (15–60 μM). After 24 h cells were collected, washed with PBS and then incubated in Fixing Solution for 20 min and then incubated in Fix Perm Solution for 30 min, at 4°C. Anti-HO-1 antibody (Santa Cruz Biotechnologies; sc-10789; 1:100), NQO1 antibody (Santa Cruz Biotechnologies; sc-376023; 1:100), superoxide dismutase (SOD) antibody (Santa Cruz Biotechnologies; sc-11407; 1:100), anti-iNOS (BD Transducion Laboratories; 610431; 1:100) antibody, anti-COX-2 (BD Transducion Laboratories; 610203; 1:100) antibody and anti-nitrotyrosine antibody (Millipore; 06-284; 1:100) were added to C6 cells, primary astrocytes and mixed glial cells. The secondary antibody (Immuno Reagents; used at diluition 1:100) was added in Fix Perm Solution and cells were evaluated using a fluorescence-activated cell sorting (FACSscan; Becton Dickinson) and elaborated with Cell Quest software as previously reported.

12Tumor necrosis factor-α and IL-6 concentration in the supernatant of cultured primary astrocytes and mixed glial cells stimulated for 24 h with IS (15–60 μM) were performed by an ELISA assay. For this we used commercially available kits for murine TNF-α and IL-6 (e-Biosciences, San Jose, CA, United States) as previously reported.

13The cytotoxic potential of IS (15–60 μM) on primary neuronal cultures after 3 h of treatment was performed using the Cytotoxicity Detection KitPLUS LDH (Roche) according to the manufacturer’s instructions. This assay was based on the evaluation of LDH activity. In the evaluation three controls are included: the first was the background control (assay medium), the second was low control (untreated cells), and the last was the high control (maximum LDH release). To determine the experimental absorbance values, the average absorbance values of the samples and controls were calculated and subtracted from the absorbance values of the background control. The percentage of cytotoxicity was determined using the equation:

14Female C57BL/6 mice (6–8 weeks; Harlan Laboratories, Udine, Italy) were fed a standard chow diet and housed under specific pathogen-free conditions at the University of Messina Animal Care Review Board approved the study. The animal experiments were performed following the regulations in Italy (D.M. 116192), Europe (O.J. of E.C. L 358/1 12/18/1986), United States (Animal Welfare Assurance No. A5594-01, Department of Health and Human Services, United States).

15IS was dissolved in PBS and it was injected into mice (800 mg/kg, i.p. given once). After 3 h of treatment, animals were sacrified and kidneys, brains and serum were collected and stored for the analysis.

16IS levels in mice serum were evaluated according the methods of as previously reported.

17Nitrite/nitrate, TNF-α, IL-6 release was evaluated on serum samples of mice treated with IS (800 mg/kg) for 3 h. Serum nitrite/nitrate (NOx) concentration is a marker of NO levels. For the evaluation, serum samples were incubated with FAD (50 μm), NADPH (1 mm), and nitrate reductase (0.1 U/mL). The samples were then incubated with sodium pyruvate (10 mm) and LDH (100 U/mL) for 5 min. The total NOx concentration was measured by Griess reaction adding 100 μL of Griess reagent (0.1% naphthylethylenediamine dihydrochloride in H2O and 1% sulfanilamide in 5% conc. H2PO4; 1:1 v/v) to 100 μL of serum treated samples, each in triplicate. The optical density at 550 nm (OD550) was measured at 540 nm in a microplate reader Titertek (Dasit, Cornaredo, Milan, Italy) and the NOx concentrations (μM) in the samples were calculated from a standard curve of sodium nitrite.

18TNF-α and IL-6 concentration in serum mice was assessed by an ELISA (e-Biosciences, San Jose, CA, United States).

19For the histological examination, kidney and brain from sacrificed mice were immediately incised and fixed in 10% formalin. For the morphological evaluation paraffin-embedded 4 μm sections were stained with haematoxylin and eosin (H&E). For the immunohistochemistry analysis, 4-μm-thick sections of the brain and kidney tissue were collected on silane-coated glass slides (Bio-Optica, Milan, Italy). Immunohistochemical stain was performed using HRP conjugated antibodies. Antigen retrieval pretreatments were performed using a HIER citrate buffer pH 6.0 (Bio-Optica, Milan, Italy) for 20 min at 98°C. EP activity was quenched with 3% H2O2 in methanol and sections were treated with a blocking solution (MACH1, Biocare Medical LLC, Concord, CA, United States) for 30 min each. Slides were then incubated overnight at 4°C with primary antibody diluted in PBS (0.01 M PBS, pH 7,2).

20The primary antibodies used were: a mouse anti-COX-2 [BD Transduction Laboratories used at dilution 1:250; a rabbit anti-nitrotyrosine purchased from Millipore (Temecula, CA) used at dilution 1:100].

21Antigen-antibody binding was detected by a HRP polymer detection kit (MACH1, Biocare Medical LLC, Concord, CA, United States). Antibody deposition was visualized using the DAB chromogen diluted in DAB substrate buffer and the slides were counterstained with haematoxylin. Between all incubation steps, slides were washed two times (5 min each) in PBS. For each tissue section, a negative control was performed using an irrelevant mouse or rabbit Ab.

22Data are presented as standard error of the mean (SEM) showing the combined data of at least three independent experiments each in triplicate. Statistical analysis was performed by analysis of variance test, and multiple comparisons were made by Bonferroni’s test. A P-value lower than 0.05 was considered significant.

23In order to assess the effect of IS on oxidative stress in C6 cells, we evaluated intracellular ROS production. Our results indicated that IS at all tested concentrations (15–60 μM), induced a significant and concentration-dependent increase in ROS production (P < 0.001 vs. control; Figure 1A). We examined ROS production also in presence of DPI (10 μM) and NAC (2 mM). As shown in Figure 1A, DPI and NAC significantly inhibited ROS release induced by IS (P < 0.001 vs. IS alone, Figure 1A).

24Following its activation, Nrf2 translocates into the nucleus and regulates cell protective gene expression. We labeled Nrf2 with a green fluorescence to track the influence of IS (30 μM) added for 1 h. In presence of IS, we observed a reduction in Nrf2 nuclear translocation (Figure 1B). To study the mechanisms of IS-induced reduction, we also examined Nrf2 nuclear translocation after treatment with IS (15–60 μM) in the presence of DPI and NAC. As shown in Figure 1B, NAC more than DPI increased Nrf2 nuclear translocation compared to IS alone (P < 0.01 vs. IS, Figure 1B).

25Enzymes dealing with oxygen radicals are HO-1, NQO1, and SOD. In oder to assess their expression profile in the presence of IS, we treated C6 cells with IS (15–60 μM). After 24 h, we observed a decrease in HO-1 and NQO1 expression (P < 0.05 vs. control for HO-1, P < 0.01 vs. control for NQO1; Figures 1C,D). A weak inhibition by IS was observed on SOD expression (Figure 1E).

26Aryl hydrocarbon Receptor (AhR) is the believed binding partner of IS. Therefore, we investigated AhR activation, through a green fluorescent labeling, in the presence of IS (30 μM) and DPI. After 1 h nuclear presence of AhR was increased after IS treatment and the IS effect could partially be blocked by DPI (Figure 1F).

27To evaluate the possible involvement of AhR in ROS release induced by IS, we analyzed ROS production in presence of the AhR inhibitor: CH-223191 (1 μM). CH-223191 significantly reduced IS induced ROS production (P < 0.001 vs. IS; Figure 1G).

28Nuclear factor-kB p65 was labeled with a green fluorescence to track the effect of IS (30 μM) on NF-kB activation. p65 nuclear traslocation resulted increased after IS treatment (Figure 2A).

29The IS-induced p65 NF-kB nuclear translocation was inhibited by DPI and NAC and to a lesser extent by CH-223192 (Figure 2A). To evaluate the possible involvement of NF-kB in ROS release induced by IS, we analyzed ROS production in presence of a NF-kB inihibitor: PDTC (200 μM; Figure 2B). PDTC significantly reduced IS induced ROS production (P < 0.05 vs. IS alone; Figure 2B).

30Primary astrocytes and mixed glial cell cultures are a less artificial cell culture system. In these cells IS (15–60 μM) also induced a significant ROS production (P < 0.001 vs. control; P < 0.05 vs. astrocytes alone; Figure 3A).

31IS led to an increase of nitrotyrosine formation (P < 0.05 vs. control, P < 0.05 vs. astrocytes alone; Figure 3B) and to a reduction of Nrf2 translocation (Figure 3C) and HO-1 expression (P < 0.05 vs. control; Figure 3D). In primary mixed glial cell cultures the response was much more prominent indicating the contribution of the microglial cells. Comparable to the C6 cell lines, IS (30 μM) added for 1 h enhanced nulcear translocation of AhR and p65 (Figures 4A,B).

32Moreover under the same experimental conditions, we observed a significant increase in iNOS and COX-2 expression in astrocytes and mixed glial cell cultures treated with IS (15–60 μM; P < 0.001 vs. control and P < 0.01 vs. astrocytes alone; Figures 5A,B). IS treatment also induced a significant production of TNF-α in astrocytes and mixed glial cell cultures and IL-6 in mixed glial cell cultures (P < 0.05 vs. control and P < 0.05 vs. astrocytes alone; Figures 5C,D).

33In order to investigate the effect of IS on neuronal death we used cortical and hippocampal neuron cultures. Our results showed that both, cortical and hippocampal neurons, are susceptible to IS-induced neuronal cell death in a dose-dependent fashion (P < 0.05 vs. control, P < 0.005 vs. hippocampal neurons; Figure 6).

34To match our in vitro findings with the in vivo situation, we injected mice with IS (800 mg/Kg) which resulted in a significantly higher IS serum concentration (79.66 ± 1.67 μM vs. 0.55 ± 0.00 μM, P < 0.001 vs. control). Total nitrite serum increased significantly in IS-treated mice compared to control mice (89.76 ± 9.98 vs. 55.99 ± 9.69 μM, P < 0.05). TNF-α and IL-6 evaluation indicated that IS induced a weak increase in TNF-α serum levels (18.91 ± 1.77 vs. 17.43 ± 2.02 of control group; P = NS) but a significant increase if IL-6 serum levels (23.99 ± 3.38 vs. 12.93 ± 2.48 of control group; P < 0.05).

35Our results indicate a COX-2 immunoreactivity in a subset of neurons in the brain tissue in normal as well as treated mice. In the treated mice, more cells showed an immunoreactivity which extended to degenerating neurons and blood vessels. Also in the kidney, we observed a strong COX-2 staining primarily in the glomeruli (Figure 7). Similarily, the anti-nitrotyrosine antibody stained neurons of the treated mice, while we saw only weak staining in the control group. Also in the kidney the immunostaining of the glomeruli was stronger in the treated mice compared to control (Figure 7).

36According to previous observations, we found atrophic glomeruli with thickening of the Bowman’s capsule and mesangial matrix and aspects of segmental solidification after IS treatment (Figure 7). The tubular epithelial cells showed granule-fatty degeneration and sometimes vacuoles and were arranged around amorphous and hypereosinophilic protein aggregates (“casts”; Figure 7). We observed interstitial edema, dilatation of renal arterioles and small hemorrhagic areas (Figure 7).

37We could also observed IS effects in the brain. Histological evaluation showed some neurons showing cytoplasm angular margins, with eosinophilic cytoplasm and pyknotic nuclei (neuronal necrosis). Around the necrotic neurons were slightly hyperplastic glial cells (satellitosis).

38In this study, we can provide evidence that IS can directly influence glial function and can cause neuronal damage, implicating IS directly in the pathways by which CKD influences cognitive functions. Cognitive impairment of CKD patients is one of the main complications despite pharmacological and dialytic treatment. We were able to show that IS induces oxidative stress and inflammatory mediators in glial cells. Oxidative stress and inflammation are essential for defense against injuries, but, if not properly regulated, they are capable of initiating various deleterious effects. Oxidative stress increases together with the progression of CKD and it correlates with the level of renal function and, therefore, also with IS levels. In addition, the antioxidant systems are also compromised in CKD patients and worsen with the progression of renal failure. Thus, the control of inflammation and oxidative stress is of particular importance in uremic syndrome.

39Our observations point to specific pathways underlying the oxidative stress and inflammation induced by IS in glial cells: (i) NADPH oxidase and glutathione levels, (ii) AhR and NF-kB activation, (iii) a reduced antioxidant response Nrf2-mediated, (iv) activation of pro-inflammatory mediators, and (v) alteration in glial proliferation/cell cycle. Moreover, we find a direct link between IS and neuronal damage linking IS to neurotoxicity. We found that IS induced a significant and concentration-related ROS release from cultured C6 astrocytes, primary astrocytes and to an even greater extent in mixed glial cell culture. Mechanistic studies revealed that both NADPH oxidase, as evaluated by the presence of DPI, and GSH homeostasis, as evaluated by NAC addition, are involved in IS-induced ROS release. These results are in accordance with previous studies reporting that IS interfered both with pro- and anti-oxidant factors in endothelial cells, vascular smooth muscle cells, kidney cells, and macrophages. It has been also reported that NAD(P)H oxidase levels increased in CKD patients and in experimental models of renal insufficiency. IS is a potent AhR ligand. In the brain, AhR is ubiquitously expressed including the cerebral cortex, hippocampus, and cerebellum. It has been implicated in sensorimotor and cognitive dysfunctions caused by oxidative stress or excitotoxicity. Our results indicated that IS activates AhR in astrocytes which likely promoted further oxidative stress. This result fits with data reporting an AhR-mediated oxidative stress pathway in human vascular endothelial cells. Interestingly, our data gives further insight in IS-induced AhR-ROS pathway in astrocytes since treatment with IS in presence of DPI is able to reduce AhR activation. Here we report that IS also activated NF-κB and previous studies indicated a reciprocal interaction between NF-κB and Nrf2. IS-induced ROS is able to induce NF-kB activation. This activation could, in turn, be responsible for the Nrf2 downregulation, because the interaction of p65 with Keap1 promotes reduction of Nrf2 protein level through Nrf2 ubiquitination. Moreover, the upregulation of p53 expression induced by IS-induced NF-κB activation is involved in the suppression of Nrf2 mRNA expression. Our data indicate a cross-talk between ROS and NF-kB because, DPI and NAC treatment were able to reduce NF-kB activation and NF-kB inhibition was able to interfere with ROS release in astrocytes. Nrf-2 is a transcription factor responsible for the regulation of the cellular redox balance and protective antioxidant and phase II enzymes. Nrf2 binding to the antioxidant response element (ARE) induced the regulation of some anti-oxidant proteins such as HO-1 and NQO1. We found that IS also reduced HO-1 and NQO1 expression, thus further contributing to a decrease of antioxidant defenses and to oxidative stress-induced damage in CNS cells.

40Astrocytes are the most abundant glial cells in the CNS and they have a number of important physiological properties related to the homeostatic control of the extracellular environment. Astrocyte cells provide structural, trophic, and metabolic support to neurons, modulate synaptic activity and are involved in multiple brain functions contributing to neuronal development. Moreover, astrocytes actively participate in processes triggered by brain injuries, aimed at repairing brain damage. It has been recently reported that astrocytes contribute actively to various forms of dementia and disturbances in the complex neuron–glia interaction are increasingly recognized as an important pathophysiological mechanism in a wide variety of neurological disorders including neurodegeneration.

41In response to a variety of stimuli and pathological events, astrocytes and microglia become activated. Microglia, activated earlier than astrocytes, promotes astrocytic activation by releasing inflammatory mediators and ROS. On the other hand, activated astrocytes facilitate the activation of distant microglia, and in some cases also inhibit microglial activities. Thus, atrocytes–microglia interactions are important in regulating both physiological and pathological conditions. We could demonstrate that in mixed glial cell cultures stimulation with IS resulted in higher levels of ROS and proinflammatory mediators.

42Activated microglia and astrocytes also release a variety of cytokines, chemokines, and toxic factors, such as TNF-α, IL-6, and NO, all of which may lead to neuronal toxicity and result in the aggressive neuronal apoptosis, that has been reported as the most crucial event in neuronal loss of neurological diseases. In this study, we observed that IS significant increases in astrocytes and mixed glial cells cytokines production and of pro-inflammatory enzymes as iNOS and COX-2 that, together with oxidative stress conditions can influence neuronal death, involved in neurodegeneration. Moreover, we observed that IS increased neuronal cell death in cortical and hippocampal neurons thus supporting its effect in neuronal loss.

43Taken together, we can show that the AhR is important for the IS induced activation of NF-kB, ROS and pro-inflammatory cytokine production, and downregulation of cell protective factors such as Nrf2, HO-1 or NQO1 in glial cells. Some of these pathways can be specifically blocked. Evidences of IS-induced effects on CNS are here supported also by in vivo experiments. IS induced histological brain alterations and the expression of oxidative stress and inflammatory markers, such as nitrotyrosine and COX-2. Until now, there was little information about the potential of IS on SNC cells. Taken together, our results highlight the effect of IS on CNS homeostasis. This study adds to hypothesis that IS significantly contributes to neurological complications observed in CKD, and that its levels could be not only a marker of disease progression but also a pharmacological target in cognitive dysfunction observed in CKD.

44SA drafted the manuscript, participated in research design and carried out the experiments; TM contributed to conceive and design the experiments and to the writing of the manuscript; SC and BR contributed to the writing of the manuscript and designing the experiments; MC participated to the in vivo experiments; OP performed the histological and immunohistochemical analysis; GA contributed to analyze the data; AP contributed to data analysis and to the writing of the manuscript; SM conceived and designed the research, the experiments and contributed to the writing of the manuscript. All authors read and approved the final manuscript.

45The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer DC and handling Editor declared their shared affiliation, and the handling Editor states that the process nevertheless met the standards of a fair and objective review.