Telmisartan directly ameliorates the neuronal inflammatory response to IL-1β partly through the JNK/c-Jun and NADPH oxidase pathways

Background Blockade of angiotensin II type 1 (AT1) receptors ameliorates brain inflammation, and reduces excessive brain interleukin-1 beta (IL-1β) production and release from cortical microglia. The aim of this study was to determine whether, in addition, AT1 receptor blockade directly attenuates IL-1β-induced inflammatory responses in neuronal cultures. Methods SK-N-SH human neuroblasts and primary rat cortical neurons were pretreated with telmisartan followed by exposure to IL-1β. Gene expression was determined by reverse transcriptase (RT)-PCR, protein expression and kinase activation by western blotting, NADPH oxidase activity by the lucigenin method, prostaglandin E2 (PGE2) release by enzyme immunoassay, reactive oxygen species (ROS) generation by the dichlorodihydrofluorescein diacetate fluorescent probe assay, and peroxisome proliferator-activated receptor gamma (PPARγ) involvement was assessed with the antagonists GW9662 and T0070907, the agonist pioglitazone and the expression of PPARγ target genes ABCG1 and CD36. Results We found that SK-N-SH neuroblasts expressed AT1 but not AT2 receptor mRNA. Telmisartan reduced IL-1β-induced cyclooxygenase-2 (COX-2) expression and PGE2 release more potently than did candesartan and losartan. Telmisartan reduced the IL-1β-induced increase in IL-1R1 receptor and NADPH oxidase-4 (NOX-4) mRNA expression, NADPH oxidase activity, and ROS generation, and reduced hydrogen peroxide-induced COX-2 gene expression. Telmisartan did not modify IL-1β-induced ERK1/2 and p38 mitogen-activated protein kinase (MAPK) phosphorylation or nuclear factor-κB activation but significantly decreased IL-1β-induced c-Jun N-terminal kinase (JNK) and c-Jun activation. The JNK inhibitor SP600125 decreased IL-1β-induced PGE2 release with a potency similar to that of telmisartan. The PPARγ agonist pioglitazone reduced IL-1β-induced inflammatory reaction, whereas telmisartan did not activate PPARγ, as shown by its failure to enhance the expression of the PPARγ target genes ABCG1 and CD36, and the inability of the PPARγ antagonists GW9662 and T0070907 to modify the effect of telmisartan on COX-2 induction. The effect of telmisartan on IL-1β-stimulated COX-2 and IL-1R1 mRNA expression and ROS production was replicated in primary rat cortical neurons. Conclusions Telmisartan directly ameliorates IL-1β-induced neuronal inflammatory response by inhibition of oxidative stress and the JNK/c-Jun pathway. Our results support the hypothesis that AT1 receptor blockers are directly neuroprotective, and should be considered for the treatment of inflammatory conditions of the brain.

To further clarify the mechanisms of the direct antiinflammatory effects of ARBs in neuronal targets, we studied the effects of ARBs in a well-characterized human neuronal system widely used as an in vitro model of neuronal injury, the SK-N-SH neuroblastoma cell line [37,38]. In particular, we focused on telmisartan as an ARB prototype because of its reported pleiotropic antiinflammatory effects as an AT 1 receptor antagonist and a peroxisome proliferator-activated receptor gamma (PPARγ) agonist [23,32,[39][40][41]. We investigated whether telmisartan ameliorates the inflammatory response to IL-1β in SK-N-SH neuroblasts and what are the mechanisms involved in these effects, and we compared the effects of telmisartan in SK-N-SH neuroblasts with those in rat primary cortical neurons.

SK-N-SH neuroblast culture
Human SK-N-SH neuroblasts were obtained from the American Type Culture Collection (HTB-11, Rockville, MD, USA) and grown in MEM with Earle's salts and HEPES, supplemented with 10 % fetal bovine serum and 100 U/ml penicillin/streptomycin. Cells were cultured at 37°C in a humidified atmosphere of 5 % CO 2 /95 % air until they reached 80 % confluence, then confluent monolayers were passaged routinely by trypsinization. Cells between passages 3 and 10 were used in this study, and before each experiment, they were starved overnight in a serum-free medium.

Primary rat cortical neuron culture
All animal care and experimental procedures in the present study were approved by the National Institute of Mental Health Animal Care and Use Committee (Bethesda, MD, USA). All efforts were made to minimize the number of animals used and their suffering (National Institutes of Health Guide for the Care and Use of Laboratory Animals, Publication number 80- 23, received 1996). Primary cortical neuron cultures were obtained from fetal Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA USA) at embryonic day 18 (E18) [42]. Fetal cerebral cortices were collected and placed in ice-cold Hank's balanced salt solution. After removal of the meninges, the cortices were dispersed into the same buffer containing 0.25 % trypsin, and digested for 15 minutes at 37°C. Trypsin digestion was stopped by adding a two-fold volume of DMEM, supplemented with 10 % FBS and 0.1 mg/ml DNase I. After gentle trituration, digested tissues were separated by centrifugation at 200 × g for 5 minutes. The cell pellets were resuspended in complete Neurobasal culture medium supplemented with 2 % B27 and 0.5 mmol/l GlutaMax. After filtration through a 70 μm cell restrainer (BD Falcon, Vernon Hills, IL, USA), cells were plated at a density of 1 × 10 6 cells/ml onto poly-D-lysine coated plates (Becton Dickinson and Co., Franklin Lakes, NJ, USA). Cultures were incubated in a humidified atmosphere of 5 % CO 2 /95 % air at 37°C. Only mature cultures (10-14 days in vitro) were used in this study. Immunocytochemical validation with anti-microtubuleassociated protein 2 (MAP-2) antibody and 4',6-diamidino-2-phenylindole (DAPI) showed that more than 95 % of the cells in the culture system were neurons (data not shown).

Drug treatment
The cells were pre-incubated for 2 hours with telmisartan, candesartan, losartan, CGP 42112, PD 123319, DPI, SP600125, pioglitazone, T0070907, GW9662, or vehicle before exposure to IL-1β. Most of the experiments were performed with the maximum stimulatory concentration of 10 ng/ml IL-1β, and the exposure times were 2 hours for ROS determination, 3 hours for RT-PCR analysis, and 24 hours for COX-2 protein and PGE 2 determinations. The SK-N-SH neuroblasts were incubated with 100 μmol/l H 2 O 2 for 3 hours to determine the protective effect of telmisartan. Activation of MAPKs, c-Jun, and NF-κB was determined by western blotting at various time intervals up to 2 hours. All concentrations used and time intervals are indicated in the figure legend for each particular experiment. All drugs were initially prepared as 1000-fold concentrated stock solutions, and were added directly into the cell-culture medium. Telmisartan, DPI, SP600125, pioglitazone, T0070907, and GW9662 were dissolved in dimethyl sulfoxide (DMSO). The final concentration of DMSO in experimental conditions was 0.1 %. Candesartan was initially dissolved in 0.1 mol/l Na 2 CO 3 , and further diluted to stock concentration with isotonic saline, at a final pH of 7.5 to 8.0. All other drugs were dissolved in isotonic saline. Control cells were treated with the corresponding vehicle in all experiments.

Real-time PCR
Total RNA was isolated using TRIzol reagent followed by purification using an RNeasy Mini Kit (Qiagen, Valencia, CA, USA) in accordance with the manufacturer's instructions. Synthesis of complementary DNA (cDNA) was performed with 0.6 μg of total RNA and Super-Script III first-Strand Synthesis Kit (Invitrogen, Carlsbad, CA, USA). Quantitative real-time PCR was performed on DNA Engine Opticon™ (MJ Research, Waltham, MA) with SYBR Green PCR Master Mix. PCR was performed in a 20 μl reaction mixture containing 10 μl SYBR Green PCR Master Mix, 2 μl cDNA and 0.3 μmol/l of each primer for a specific target ( Table 1). The amplification conditions consisted of 1 denaturation/activation cycle at 95°C for 10 minutes, followed by 40 to 45 cycles at 95°C for 15 seconds and 60°C for 60 seconds. Serial dilutions of cDNA from the same source as samples were used to obtain a standard curve. The individual targets for each sample were quantified by determining the cycle threshold (Ct) and by comparison with the standard curve. The relative amount of the target mRNA was normalized to the level of GAPDH mRNA.

Western blotting
For the determination of NFκB-p65 nuclear translocation, nuclear protein extracts were prepared using Nuclear Extraction Kit (Pierce, Rockford, IL, USA) in accordance with the manufacturer's instructions. For other proteins, the whole-cell lysates were prepared in Tris-Glycine SDS Sample Buffer (Invitrogen). The protein extracts were separated by electrophoresis on 10 % SDS-PAGE gels and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were blocked for 1 hour and incubated overnight at 4°C with the primary antibodies, followed by washing and exposure to secondary antibodies for 1 hour at room temperature. The membranes were exposed to Super-Signal West Dura Substrate for chemiluminescent detection.

Measurement of reactive oxygen species
The levels of intracellular ROS were determined by the change in the fluorescence resulting from the oxidation of the fluorescent probe H 2 DCFDA using OxiSelect™ ROS Assay Kit (Cell Biolabs, San Diego, CA, USA) in accordance with the manufacturer's instructions. After preincubation with telmisartan or DPI, the cells were loaded with H 2 DCFDA for 30 minutes at 37°C and exposed to IL-1β for an additional 2 hours. The level of fluorescence, corresponding to intracellular ROS, was determined using a plate reader (VICTOR3; Perkin-Elmer, Torrance, CA, USA) with 485 nm excitation and 535 nm emission filters.
Prostaglandin E 2 measurement by enzyme immunoassay PGE 2 release was determined in cells culture medium by enzyme immunoassay (EIA) (PGE 2 EIA Kit; Cayman Chemical) in accordance with the manufacturer's instructions.

NADPH oxidase activity assay
The lucigenin method was used to determine NADPH oxidase activity in SK-N-SH cells. Cells were collected by scraping, and pelleted by centrifugation at 500 × g for 5 minutes. The pellets were resuspended and homogenized in ice-cold buffer containing 50 mmol/l Tris, pH 7.4, 1 mmol/l EDTA, 1 mmol/l DTT, 0.5 mmol/l phenylmethylsulfonyl fluoride (PMSF) and 1× protease inhibitor cocktail. The crude membrane fraction was pelleted by centrifugation at 16,000 × g for 90 minutes at 4°C, and the pellets were resuspended in 200 μl of assay buffer containing 8 mmol/l sodium phosphate, pH 7.4, 140 mmol/l NaCl, 10 mmol/l KCl, 2 mmol/l MgCl 2 , 50 mmol/l triethanolamine, 1 mmol/l DTT, and 1× protease inhibitor cocktail. The total protein concentration was determined by the Bradford assay and adjusted to 1 mg/ml. An aliquot (200 μl) of protein sample (100 μg of membrane proteins) were incubated in the presence of 5 μmol/l lucigenin and 100 μmol/l NADPH. The luminescence was monitored at 2-minute intervals using a plate reader (VICTOR3; Perkin-Elmer) to determine relative changes in NADPH oxidase activity.

Ang II measurement by enzyme immunoassay
Ang II concentration in the cell-culture medium was measured using a commercial kit (Ang II EIA Kit; Cayman Chemical) following the manufacturer's instructions. The limit of sensitivity of the assay was 1.5 pg/ml.

Statistical analysis
Statistical significance was determined using GraphPad Prism 5 Software (GraphPad Software, San Diego, CA, USA). Multiple group comparisons were performed by one-way ANOVA followed by Newman-Keuls Post test. Differences were considered significant at P < 0.05. Values are expressed as the mean ± SEM.

Dose response and time course of interleukin-1β-induced neuronal inflammatory response
Incubation of SK-N-SH neuroblasts in the presence of IL-1β induced COX-2 mRNA expression in a dosedependent and time-dependent manner ( Figure 1A,B). Maximum stimulation of COX-2 mRNA was obtained with 10 ng/ml IL-1β, and it reached a peak after 3 hours of exposure ( Figure 1A and 1B). Thus, this dose of IL-1β was selected for all subsequent experiments.
Angiotensin II receptor types in SK-N-SH neuroblasts and the effect of receptor blockade SK-N-SH neuroblasts expressed AT 1 receptor mRNA, and the receptor expression was not affected by IL-1β or telmisartan, either alone or in a combination (Figure 2A). AT 2 receptor mRNA was not detectable in our preparation of SK-N-SH neuroblasts. Incubation in the presence of the AT 2 receptor agonist CGP 42112 did not change IL-1β stimulation of COX-2 gene expression ( Figure 2B) or PGE 2 release ( Figure 2C). Similarly, incubation in the presence of the AT 2 receptor antagonist PD 123319 did not change IL-1β stimulation of PGE 2 release, and did not alter the inhibitory effect of telmisartan ( Figure 2C).
Telmisartan prevents interleukin-1β-induced NADPH oxidase activation, reactive oxygen species production and interleukin-1 receptor 1 gene expression High expression of the NADPH oxidase isoform NOX-4 and substantially lower expression of NOX-5 were found in SK-N-SH neuroblasts ( Figure 3A). Expression of NOX-1 and NOX-2 was not detected ( Figure 3A). Exposure to IL-1β significantly increased NOX-4 mRNA Telmisartan reduced IL-1β-induced reactive oxygen species (ROS) generation to a lesser extent than does diphenyleneiodonium (DPI). The cells were pretreated with 10 μmol/l Telm or 5 μmol/l DPI for 2 hours before 1 hours exposure to 10 ng/ml IL-1β to determine ROS generation. (E,F) DPI dose-dependently inhibited IL-1β-induced PGE 2 release with a potency similar to telmisartan. The cells pretreated with indicated concentrations of DPI or Telm for 2 hour were incubated with IL-1β for 24 hours to determine cumulative PGE 2 release. (G) Both telmisartan and DPI reduce hydrogen peroxide-induced COX-2 mRNA expression. The cells were pretreated with 10 μmol/l Telm or 5 μmol/l DPI for 2 hours before exposure for 3 hours to 100 μmol/l hydrogen peroxide (H 2 O 2 ) to determine COX-2 mRNA expression. (H) Both telmisartan and DPI reduce IL-1β-induced expression of IL-1β receptor IL-1R1 mRNA. The cells were pretreated with 10 μmol/l Telm or 5 μmol/l DPI for 2 hours before exposure for 3 hours to 10 ng/ml IL-1β to determine IL-1R1 mRNA expression. Results are presented as a percentage of Veh. All results are means ± SEM from at least three independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001 vs. IL-1β or H 2 O 2 ; ### P < 0.001 vs. Veh. expression, and this effect was reduced by telmisartan ( Figure 3B). IL-1β significantly increased NADPH oxidase activity, an effect also reduced by telmisartan ( Figure 3C). IL-1β enhanced ROS production, and this effect was decreased by both telmisartan and DPI ( Figure 3D). DPI dose-dependently inhibited IL-1βinduced PGE 2 release ( Figure 3E). The reduction in IL-1β-stimulated PGE 2 release was similar for both telmisartan and DPI ( Figure 3F).
Telmisartan reduced the enhanced COX-2 mRNA expression produced by H 2 O 2 to an extent similar to that resulting from exposure to DPI ( Figure 3G).
Exposure to IL-1β enhanced mRNA expression of its receptor, IL-1R1, and this change was reduced to a similar degree by telmisartan and DPI ( Figure 3H).
Telmisartan decreases interleukin-1β-induced c-Jun N-terminal kinase and c-Jun activation IL-1β time-dependently activated JNK in SK-N-SH neuroblasts, reaching maximum stimulation after 30 to 60 minutes of exposure, and this effect was significantly reduced by telmisartan ( Figure 4A). Exposure to IL-1β simultaneously and time-dependently enhanced c-Jun phosphorylation, a change significantly decreased by telmisartan ( Figure 4A). The effect of telmisartan was of similar magnitude to that of DPI ( Figure 4B). Incubation in the presence of the specific JNK inhibitor SP600125 abrogated the IL-1β-induced phosphorylation of JNK and c-Jun ( Figure 4B), COX-2 mRNA expression ( Figure 4C), and PGE 2 release, in a dose-dependent manner ( Figure 4D).

Peroxisome proliferator-activated receptor-γ is not involved in the neuroprotective effect of telmisartan
Incubation of SK-N-SH neuroblasts with the PPARγ agonist pioglitazone significantly reduced IL-1β-induced COX-2 mRNA expression ( Figure 7A), dose-dependently reduced PGE 2 release ( Figure 7B), and upregulated the mRNA expression of the PPARγ target genes ABCG1 and CD36, without affecting PPARγ mRNA expression ( Figure 7C). Conversely, telmisartan did not alter ABCG1 or CD36 mRNA expression ( Figure 7C). Incubation of SK-N-SH neuroblasts in the presence of the PPARγ antagonists T0070907 or GW9662 alone did not significantly alter IL-1β-induced COX-2 mRNA expression ( Figure 7D), and neither T0070907 nor GW9662 modified the inhibitory effect of telmisartan on IL-1βinduced COX-2 mRNA and protein expression ( Figure 7D,E).

Effect of angiotensin II on the telmisartan neuroprotection in SK-N-SH neuroblasts
Angiotensin II levels were undetectable in the cell-culture medium (results not shown). Exposure of SK-N-SH neuroblasts to 1 μmol/l Ang II for 24 hours did not alter PPARγ gene expression but strongly decreased gene expression of the PPARγ target genes ABCG1 and CD36 ( Figure 8A). Pretreatment of neuroblasts with Ang II for 24 hours did not change basal COX-2 mRNA expression or basal PGE 2 release. Ang II did not affect COX-2 mRNA expression induced by 10 ng/ml IL-1β, but did enhance IL-1β-induced PGE 2 release ( Figure 8B,C). Pretreatment with Ang II did not change the inhibitory effect of telmisartan on IL-1β-stimulated COX-2 gene expression and cumulative PGE 2 release ( Figure 8B,C).
(See figure on previous page.) Figure 4 Neuroprotective effect of telmisartan is partially mediated through inhibition of the c-Jun N-terminal kinase (JNK)/c-Jun pathway in SK-N-SH neuroblasts. (A) Telmisartan attenuated the time-dependent activation of JNK and c-Jun in response to interleukin-1 beta (IL-1β). The cells were pretreated for 2 hours with 10 μmol/l telmisartan (Telm) before exposure to 10 ng/ml IL-1β for the indicated time intervals. Phosphorylation of JNK and c-Jun was determined by western blotting. Representative blots are shown under the corresponding bar graphs. ### P < 0.001 vs. Veh; * P < 0.05, *** P < 0.001 vs. corresponding IL-1β group. (B) Telmisartan inhibited IL-1β-stimulated JNK and c-Jun activation with a potency similar to that of diphenyleneiodonium (DPI) but to a lesser extent than the JNK inhibitor SP600125. The cells were pretreated for 2 hours with 10 μmol/l Telm, 5 μmol/l DPI, or 10 μmol/l SP600125 (SP) before exposure for 30 minutes to 10 ng/ml IL-1β. The phosphorylation of JNK and c-Jun was detected as above. Results are shown as a percentage of the IL-1β-treated group. (C,D) The JNK inhibitor SP600125 abrogated the IL-1β-induced COX-2 mRNA expression and PGE 2 release. The cells were pretreated for (C) 2 hours with 10 μmol/l SP before exposure for 3 hours to 10 ng/ml IL-1β to determine COX-2 mRNA expression, or with (D) the indicated concentrations of SP600125 before exposure for 24 hours to 10 ng/ml IL-1β to determine cumulative PGE 2 release. All results are presented as means ± SEM from three independent experiments. *** P < 0.001 vs. IL-1β; ### P < 0.001 vs. Veh; $$$ P < 0.001 vs. IL-1β + Telm.
Telmisartan reduces interleukin-1β upregulation of reactive oxygen species formation, interleukin-1 receptor type 1 and cyclooxygenase-2 mRNA expression in primary rat cortical neurons Exposure of primary rat cortical neurons to IL-1β induced both COX-2 and IL-1R1 mRNA expression and ROS generation, and these effects were significantly reduced by telmisartan ( Figure 9A-C).

Discussion
This study was designed to test the hypothesis that direct neuronal exposure to ARBs is neuroprotective. IL-1β was selected based on its well-characterized participation in neuronal injury associated with inflammatory and neurodegenerative diseases of the brain [6][7][8][9][10]. The principal finding of our study is that ARBs, in particular telmisartan, directly and significantly ameliorate the IL-1βinduced neuronal inflammatory response.
Ang II stimulates two receptor types, the AT 1 and AT 2 receptors [43]. Excessive AT 1 receptor stimulation is associated with brain inflammation, whereas stimulation of AT 2 receptors has been proposed to exert balancing neuroprotective effects, particularly when AT 1 receptors are blocked by ARB administration [43][44][45]. SK-N-SH human neuroblasts expressed AT 1 receptor mRNA, whereas AT 2 receptor mRNA was undetectable in these cells. Furthermore, exposure of SK-N-SH neuroblasts to PD 123319 (an AT 2 receptor antagonist) or CGP 42112 (an AT 2 receptor agonist) did not change the effects of IL-1β, and PD 123319 did not modify the neuroprotective effect of telmisartan. These results indicate that the neuroprotective effect of telmisartan and other ARBs in SK-N-SH neuroblasts is dependent on AT 1 receptor blockade without involvement of AT 2 receptors.
The neurotoxic effects of IL-1β, confirmed in this study, have been well characterized. They depend on stimulation of the IL-1R1 receptor, and characteristically involve NADPH oxidase activation and ROS formation, COX-2 induction, and PGE 2 production and release, leading to neuronal toxicity and apoptosis [5,37,38,[46][47][48][49]. Our results support the hypothesis that IL-1β, when produced in excess by activated microglia, may directly  Figure 6 The nuclear factor-kappa B (NF-κB) pathway is not involved in the neuroprotective effect of telmisartan in SK-N-SH neuroblasts. (A) Telmisartan does not prevent time-dependent IκB-α protein degradation in cells in response to interleukin-1 beta (IL-1β). Cells were pretreated for 2 hours with 10 μmol/l telmisartan (Telm) before exposure to 10 ng/ml IL-1β for the indicated time intervals. IκB-α protein levels were determined in whole-cell extracts, and normalized to β-actin. (B) Neither telmisartan nor diphenyleneiodonium (DPI) modified IL-1βinduced expression of IκB-α mRNA. The cells were pretreated for 2 hours with 10 μmol/l Telm or 5 μmol/l DPI before exposure for 3 hours to 10 ng/ml IL-1β to determine IκB-α mRNA expression. (C) Neither telmisartan nor DPI affected IL-1β-induced nuclear translocation of the NF-κB p65 subunit. The cells were pretreated for 2 hours with 10 μmol/l Telm or 5 μmol/l DPI before exposure for 30 minutes to 10 ng/ml IL-1β. The NF-κB p65 subunit protein was determined in nuclear extracts and normalized to the level of the nuclear protein histone H4. Representative western blots are shown below the corresponding bar graphs. Results are presented as means ± SEM from three independent experiments. # P < 0.05, ### P < 0.001 vs. Veh.
generate further inflammatory cascades in neurons, contributing to their increased vulnerability to injury. Telmisartan, at concentrations similar to those found in clinical studies [50], significantly reduced the neuronal inflammatory response induced by IL-1β. Most of the downstream pathways activated by IL-1β in the present study, including IL-1R1 receptor upregulation, are associated with NADPH oxidase activation [51,52]. This (See figure on previous page.) Figure 7 Peroxisome proliferator-activated receptor gamma (PPARγ) activation is not involved in the neuroprotective effect of telmisartan in SK-N-SH neuroblasts. (A,B) The PPARγ agonist pioglitazone inhibited interleukin-1 beta (IL-1β)-induced cyclooxygenase-2 (COX-2) gene expression and prostaglandin E 2 (PGE 2 ) release. The cells were pretreated for (A) 2 hours with 10 μmol/l pioglitazone (PGZ) before exposure for 3 hours to 10 ng/ml IL-1β to determine COX-2 mRNA expression or (B) with indicated concentrations of PGZ before exposure for 24 hours to 10 ng/ml IL-1β to determine cumulative PGE 2 release. (C) Pioglitazone, but not telmisartan, induced gene expression of the PPARγ target genes ABCG1 and CD36. The cells were incubated for 3 hours with 10 μmol/l PGZ or 10 μmol/l Telm to determine gene expression of PPARγ and its target genes ABCG1 and CD36. Results are shown as fold change relative to the vehicle-treated group (Veh). (D,E) PPARγ antagonists did not change the inhibitory effect of telmisartan on IL-1β-induced COX-2 expression. The cells were pretreated for 1 hour with 1 μmol/l T0070907 (T007) or 20 μmol/l GW9662 (GW), followed by 10 μmol/l Telm for 2 hours before exposure for (D) 3 hours to 10 ng/ml IL-1β to determine COX-2 mRNA, or (E) 24 hours of IL-1β to determine COX-2 protein expression. The picture below is a representative western blot. All results are means ± SEM from at least three independent experiments. * P < 0.05, *** P < 0.001 vs. IL-1β; # P < 0.05, ### P < 0.001 vs. Veh.  Figure 8 Effect of Angiotensin II (Ang II) on SK-N-SH neuroblasts. (A) Ang II did not affect peroxisome proliferator-activated receptor gamma (PPARγ) gene expression, but strongly inhibited the expression of the PPARγ target genes ABCG1 and CD36. The cells were treated with 1 μmol/l Ang II for 24 hours to determine PPARγ, ABCG1 and CD36 mRNA expression. # P < 0.05 vs. Veh. (B) Ang II affected neither interleukin-1 beta (IL-1β)-induced cyclooxygenase-2 (COX-2) mRNA expression nor the inhibitory effects of telmisartan. Cells cultured for 24 hours in the presence of 1 μmol/l Ang II were pretreated for 2 hours with 10 μmol/l telmisartan (Telm) before exposure for 3 hours to 10 ng/ml IL-1β to determine COX-2 mRNA expression. Results are presented as a percentage of the IL-1β-treated group. (C) Ang II augmented IL-1β-induced PGE 2 release but did not modify the inhibitory effect of telmisartan. The cells, cultured for 24 hours in the presence of 1 μmol/l Ang II, were pretreated for 2 hours with 10 μmol/l Telm before exposure for 24 hours to 10 ng/ml IL-1β to determine cumulative PGE 2 release. Results are presented as a percentage of IL-1β. All results are means ± SEM from at least three independent experiments. *** P < 0.001.
indicates that inhibition of NADPH oxidase activity by telmisartan is a major neuroprotective mechanism. Telmisartan decreased not only IL-1β-induced ROS formation but also H 2 O 2 -induced COX-2 expression, suggesting that reduction of the intracellular ROS and ROS-related downstream pathway [10] may be important for the neuroprotective effects of telmisartan. The wide-ranging anti-oxidant effects described here were similar to those reported previously in non-neuronal cell lines [53] and were of a potency similar to that of the NADPH oxidase and NOS inhibitor DPI [54,55]. These results are in agreement with observations showing that ARBs decrease NADPH oxidase activation associated with oxidative stress and neuronal apoptosis [36,56,57]. The neuroprotective effects of telmisartan were replicated in rat primary cortical neurons, indicating that they were not limited to responses only in the neuroblast preparations.
The discovery that telmisartan significantly prevents the IL-1β-induced upregulation of its receptor IL-1R1 in both SK-N-SH neuroblasts and rat primary cortical neurons is of major interest. Most of the IL-1β effects are mediated by IL-1R1 receptor stimulation. Administration of IL-1R1 receptor inhibitors seems to lead to amelioration of brain inflammation, and protection from stroke and traumatic brain injury, thus the development of novel IL-1R1 receptor inhibitors is the subject of active research [12,13]. For these reasons, our finding that telmisartan significantly prevents IL-1β induction of its receptor indicates an additional anti-inflammatory mechanism that might be of clinical value.
In agreement with previous observations [58], we found that IL-1β significantly stimulates a number of kinases, including p38 MAPK, ERK1/2, and JNK/c-Jun, and produces a notable activation of NF-κB in human SK-N-SH neuroblasts. Incubation in the presence of telmisartan significantly reduced IL-1β-induced JNK/c-Jun activation, but had no effect on activation of p38 MAPK, ERK1/2, and NF-κB. Stimulation of inflammatory cascades is to a considerable extent the result of activation of the transcription factor NF-κB [10]. Our observations are therefore no unexpected and concur with those of previous studies showing that anti-inflammatory mechanisms are cell-specific, depending on the inflammatory component and on the anti-inflammatory agent studied. In monocytes, macrophages, and microglia, NF-κB activation seems to be a major factor leading to inflammation and COX-2/PGE 2 production [10,41,49]. However, in brain endothelial cell lines, several important components of the IL-1β-induced inflammatory response are independent of MAPK activity [9]. Moreover, glucocorticoids reduce IL-1β-induced inflammation in cells of neural origin by mechanisms independently of NF-κB [59]. These results and our present findings indicate that factors independent of NF-κB play a major role in the anti-inflammatory effect of ARBs in neurons.
All ARBs inhibit the Ang II-induced effects associated with stimulation of physiological AT 1 receptors, but some ARBs, particularly telmisartan, are also partial PPARγ agonists [39,40]. Surprisingly in SK-N-SH neuroblasts, telmisartan failed to activate PPARγ. Furthermore, addition of PPARγ antagonists did not modify the neuroprotective effects of telmisartan, indicating that in these cells, AT 1 receptor inhibition rather than PPARγ activation may be the primary mechanism for the direct anti-inflammatory effects of ARBs. These observations Results are means ± SEM from at least three independent experiments. * P < 0.05 vs. IL-1β; ## P < 0.01, ### P < 0.001 vs. Veh.
apparently contrast with the initial demonstration of PPARγ activation by telmisartan in cell culture [39,40], the PPARγ-associated anti-inflammatory effects of telmisartan in cultured human monocytes and THP-1 cells [41], and the PPARγ-activating neuroprotective effects of telmisartan shown in vivo [23,31,32]. It has been reported that although conventional PPARγ agonists can suppress expression of proinflammatory factors in primary microglia, they do not suppress expression of pro-inflammatory molecules in a microglial cell line expressing little or no PPARγ [60,61], and are not neuroprotective when applied to neurons [62]. In the SK-N-SH neuroblast preparations used in the present study, the PPARγ gene was expressed at relatively low levels compared with AT 1 receptors (data not shown). However, in spite of the low PPARγ gene expression, a conventional PPARγ full agonist, pioglitazone [63], significantly activated PPARγ in SK-N-SH neuroblasts. Conversely, under identical experimental conditions, telmisartan was ineffective, indicating that PPARγ activation is neuroprotective but is not mediating the effects of telmisartan in SK-N-SH neuroblasts.
Indeed, the PPARγ agonist properties of individual ARBs seem to depend on the cell type studied and on the conditions of the experiments. Reports from cellculture studies indicated that the PPARγ agonist effects of candesartan and losartan are not high [39,40]; however, losartan has been found to increase PPARγ activation in certain cell types [64,65], and long-term candesartan treatment upregulates PPARγ gene expression in vivo in adipose tissue [66]. Further studies are necessary to clarify the relative contribution of AT 1 receptor blockade and the PPARγ agonist activity of ARBs in specific cell populations. Whether the PPARγ agonist effect of ARBs may be dependent on the degree of PPARγ gene expression remains an open question.
It is known that Ang II strongly inhibits PPARγ activation, an effect dependent on AT 1 receptor stimulation, and the absence of Ang II may substantially stimulate PPARγ activity [45,67,68]. In accordance with this, addition of a high Ang II concentration decreased expression of PPARγ target genes in our study. However, in our studies, Ang II levels in the cell-culture media were below the 1.5 pg/ml (corresponding to 1.5 pmol/l) limit of detection. A concentration of 1 μmol/l of Ang II was required to produce a small increase in the IL-1β-induced PGE 2 release, whereas it did not change COX-2 induction, had no effect on NADPH oxidase expression or activity (data not shown), and did not influence the protective effects of telmisartan. For these reasons, it is very likely that in SK-N-SH neuroblasts, the neuroprotective effects of telmisartan are independent of Ang II-mediated stimulation of AT 1 receptors. Ligand-independent AT 1 receptor activation has been reported previously in cardiomyocytes as a consequence of mechanical stress [69].
Based on the present results, we propose that, in SK-N-SH neuroblasts, the AT 1 receptor may be constitutively active, and the neuroprotective effects of telmisartan and other ARBs may be the result of a decrease of such constitutive AT 1 receptor activity. Recently, the constitutive activity of AT 1 receptor has been reported under basal conditions in vivo even in the absence of Ang II [70]. However, there are no reports of ligand-independent activation or constitutive AT 1 receptor activity in neurons, and the hypothesis of constitutively active neuronal AT 1 receptors requires further confirmation.
Although it must be considered that neuronal cultures may not be representative of in vivo conditions, the SK-N-SH neuroblasts cultures are a good in vitro model to study the mechanisms of action responsible for direct ARB neuroprotection.
The present observations and those of the literature suggest that ARBs may exert neuroprotective effects by several associated mechanisms: decreasing inflammationinduced circulating IL-1β levels affecting the brain and activating microglia in brain parenchyma, by direct antiinflammatory effects in microglia as shown in isolated microglia in culture [15], and by direct effects in neurons, ameliorating the neuronal inflammatory responses produced by excess IL-1β, as reported here and illustrated in Figure 10.
Our results have important clinical implications. IL-1β is a strong stimulant of oxidative stress, COX-2 production, and PGE 2 release, and it has been clearly associated with both acute and chronic inflammatory conditions of the brain. Neuronal induction of COX-2, leading to increased release of its product PGE 2 , is strongly stimulated by IL-1β, and has been linked to neuroinflammatory aspects of neurodegenerative diseases such as AD and HIV-associated dementia [38,[71][72][73]. Furthermore, it was reported that maximal COX-2 expression predates maximal activation of astrocytes and microglia in the early stages of AD [74]. For this reason, the direct neuroprotective effects of ARBs reported here may be of major clinical significance.
Our present observations may explain the recent findings that ARB administration for the treatment of hypertension significantly protects cognition, and ameliorates the incidence and progression of AD, and that the neuroprotective effects of ARBs seem to be superior to those of similarly potent anti-hypertensive medications without direct effect on AT 1 receptors [75,76]. These clinical observations are supported by pre-clinical studies, showing that ARBs reduce NADPH oxidase activation and neuronal apoptosis and protect cognition in animal models of AD and PD [36].

Conclusions
Our observations highlight the pleiotropic neuroprotective effects of ARBs. As reported previously, these compounds reduce the inflammation-induced production of circulating inflammatory cytokines affecting the brain and inflammation-induced microglial activation, significantly diminishing inflammatory cascades. As we show here, ARBs directly decrease the pro-inflammatory effects of IL-1β in neurons, including reduction of IL-1β receptor upregulation, NADPH oxidase activation, ROS production, JNK and c-Jun activation, and proinflammatory COX-2/PGE 2 . We propose that ARBs may not only reduce production of excessive proinflammatory factors, but also decrease neuronal vulnerability to injury. These properties are of significant clinical value, and help to explain the increasing evidence that treatment with ARBs ameliorates the incidence and progression of acute and chronic neurodegenerative conditions such as AD and stroke, in which neuroinflammation plays an important role.