Sphingosine kinase 1 regulates the expression of proinflammatory cytokines and nitric oxide in activated microglia

Neuroscience 166 (2010) 132–144

SPHINGOSINE KINASE 1 REGULATES THE EXPRESSION OF PROINFLAMMATORY CYTOKINES AND NITRIC OXIDE IN ACTIVATED MICROGLIA D. NAYAK,a Y. HUO,a W. X. T. KWANG,a P. N. PUSHPARAJ,b S. D. KUMAR,a E.-A. LINGa AND S. T. DHEENa*

Microglia are resident macrophages of the CNS involved in phagocytosis and general maintenance of neural environment (Dheen et al., 2007; Kaur et al., 2001). Chronic neuroinflammation activates microglia, which produce proinflammatory cytokines such as tumor necrosis factor-␣ (TNF-␣) and interleukin-1␤ (IL-1␤) and neurotoxic substances such as nitric oxide (NO), further contributing to inflammation, thus creating a vicious cycle of inflammation to microglial activation to inflammation (Dheen et al., 2005, 2007). Although it is still being debated whether they are “good” or “bad,” microglia have been implicated in neuroinflammation in various neurodegenerative diseases. Strategies to counter the harmful effects of microglial activation are studied widely, to improve and enhance potential treatment strategies for disease conditions linked to microglial activation, such as Alzheimer’s disease (Shie and Woltjer, 2007), amyotrophic lateral sclerosis (Dewil et al., 2007), multiple sclerosis, HIV associated dementia, and Parkinson’s disease (Rock and Peterson, 2006). Sphingomyelins are a class of membrane sphingolipids found largely in the brain and nervous tissue (Bryan et al., 2008). The sphingomyelin degradation pathway produces ceramide, which is broken down into sphingosine and ceramide-1-phosphate. Sphingosine is further phosphorylated into sphingosine-1-phosphate (S1P) by the action of isoenzymes, sphingosine kinases (SphK) 1 and 2. SphK1, localized in the cytosol, has been shown to be regulated in a number of forms such as phosphorylation, and protein–protein and protein–phospholipid interactions, which result in its translocation to cell membrane (Pyne et al., 2009). For example, various cytokines act as the agonists in many cells and activate SphK1 resulting in translocation of SphK1 from cytosol to plasma membrane where its substrate sphingosine resides, leading to generation and secretion of S1P (Jolly et al., 2005; Wattenberg et al., 2006). The regulation of SphK1 at the transcriptional level is associated with cell proliferation and survival, e.g. upregulation of SphK1 occurs in a number of tumors and in response to agonists (Pyne et al., 2009). S1P is known to be a modulator of cell proliferation, survival, apoptosis, migration, and Ca⫹2 hemostasis (Alemany et al., 2007). In the CNS, SphK1/S1P signaling has been shown to play a key role in neuron specific functions such as regulation of neurotransmitter release from neurons and in proliferation and survival of neurons and glia (Okada et al., 2009). S1P can act as a second messenger intracellularly and as a ligand for G protein coupled receptors (S1P1, S1P2, S1P3, S1P4, S1P5) extracellularly (Ozaki et al., 2003; Rosen and Goetzl, 2005) thereby

a Department of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore-117597 b

Medicine-Immunology, Infection and Inflammation, Glasgow Biomedical Research Centre, University of Glasgow, Glasgow, Scotland, UK G12 8TA

Abstract—Microglial activation has been implicated as one of the causative factors for neuroinflammation in various neurodegenerative diseases. The sphingolipid metabolic pathway plays an important role in inflammation, cell proliferation, survival, chemotaxis, and immunity in peripheral macrophages. In this study, we demonstrate that sphingosine kinase1 (SphK1), a key enzyme of the sphingolipid metabolic pathway, and its receptors are expressed in the mouse BV2 microglial cells and SphK1 alters the expression and production of proinflammatory cytokines and nitric oxide in microglia treated with lipopolysaccharide (LPS). LPS treatment increased the SphK1 mRNA and protein expression in microglia as revealed by the RT–PCR, Western blot and immunofluorescence. Suppression of SphK1 by its inhibitor, N, N Dimethylsphingosine (DMS), or siRNA resulted in decreased mRNA expression of TNF-␣, IL-1␤, and iNOS and release of TNF-␣ and nitric oxide (NO) in LPS-activated microglia. Moreover, addition of sphingosine 1 phosphate (S1P), a breakdown product of sphingolipid metabolism, increased the expression levels of TNF-␣, IL-1␤ and iNOS and production of TNF-␣ and NO in activated microglia. Hence to summarize, suppression of SphK1 in activated microglia inhibits the production of proinflammatory cytokines and NO and the addition of exogenous S1P to activated microglia enhances their inflammatory responses. Since the chronic proinflammatory cytokine production by microglia has been implicated in neuroinflammation, modulation of SphK1 and S1P in microglia could be looked upon as a future potential therapeutic method in the control of neuroinflammation in neurodegenerative diseases. © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: sphingosine kinase 1, sphingosine-1-phosphate, microglia, inflammation, cytokines, nitric oxide. *Corresponding author. Tel: ⫹65-6516-3217; fax: ⫹65-6778-7643. E-mail address: [email protected] (S. T. Dheen). Abbreviations: bp, basepairs; DAPI, 4’-6-diamidino-2-phenylindole; DMEM, dulbecco’s modified eagles medium; DMS-N, N Dimethylsphingosine; ELISA, enzyme linked immunosorbent assay; FBS, fetal bovine serum; IF, immunofluorescence; IFN␥, interferon␥; IL-1␤, interleukin 1␤; iNOS, inducible nitric oxide synthase; kD, kiloDalton; LPS, lipopolysaccharide; NO, nitric oxide.; PBS, phosphate buffered saline; RT–PCR, reverse-transcription polymerase chain reaction; SE, standard error; SphK1, sphingosine kinase 1; S1P, sphingosine 1 phosphate; TE, trypsin-EDTA; TNF-␣, tumor necrosis factor-␣

0306-4522/10 $ - see front matter © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.12.020

132

D. Nayak et al. / Neuroscience 166 (2010) 132–144

133

Table 1. Primer sequences used for RT–PCR

TNF-␣ IL-1␤ iNOS ␤-actin SphKI S1P1 S1P2 S1P3 S1P4 S1P5

Sense

Antisense

Size (bp)

CGTCAGCCGATTTGCTATCT GCCCATCCTCTGTGACTCAT GCTTGTCTCTGGGTCCTCTG TCACCCACACTGTGCCCATCTACGA CTTCTGGGCTGCGGCTCTATTCTG ACTATATTCTCTTCTGCACCAC TGTCACTCTGTCCTTAACTC CAACTTGGCTCTCTGCGACCT CTCTACTCCAAGGGCTATGT GTGTGTGCCTTCATTGTG

CGGACTCCGCAAAGTCTAAG AGGCCACAGGTATTTTGTCG CTCACTGGGACAGCACAGAA GGATGCCACAGGATTCCATACCCA GGAAAGCAACCACGGGCACA GCTTCGAGTCCTGACCCA GGCCACTTGTCTCTCGAT ACTGTTGGAGACAGACTGAACG TGGAGACTTCTGCCCATT CAGGTCCGACAAAGTGAG

205 229 217 314 507 80–120 80–120 80–120 80–120 80–120

modulating cellular processes (Alvarez et al., 2007) in peripheral macrophages and other immune cells (Gude et al., 2008; Hammad et al., 2008; Melendez, 2008). It has been shown that SphK1 is important for physiological responses such as chemotaxis, and NADPH oxidative burst of human immune cells and wound healing (Ibrahim et al., 2004; Melendez, 2008). SphK1 is also involved in inflammatory processes mediated by TNF-␣ in immune cells (Hammad et al., 2008; Melendez, 2008). Further, inhibition of SphK1 in neutrophils was found to reduce the secretion of TNF-␣ (Niwa et al., 2000). Since TNF-␣ secreted by activated microglia appears to be involved in neuroinflammation (Block and Hong, 2005; Dheen et al., 2007), the modulation of TNF-␣ levels and other neurotoxic substances in activated microglia could form a potential therapeutic basis for the treatment of neuroinflammatory diseases. In the present study, we demonstrate the key roles of expression of SphK1 and S1P receptors (1–5) in the BV2 microglia and their effects on proinflammatory cytokines such as TNF-␣ and IL-1␤ and neurotoxic modulators such as NO in activated microglia.

EXPERIMENTAL PROCEDURES Cell culture BV2, a murine microglial cell line, which is a suitable model for in vitro study of microglia (Bocchini et al., 1992) was used in this study. The cells were grown in a flask (75 cm2) and washed with phosphate buffered saline (PBS) solution twice and then treated with trypsin-EDTA (TE; Cat No. X0930, Biowest, Nuaille, France) in PBS for 3 min at 37 °C. The TE was inactivated by equal volume of 1⫻fetal bovine serum (FBS; Cat No. SV30160.03, HyClone, Utah, USA). The culture was centrifuged at 1000 rpm at 4 °C for 5 min and the pellet was resuspended in 10 ml of Dulbecco’s Modified Eagle’s Medium (DMEM, Cat No. D1152, Sigma, MO, USA) containing 10% FBS and 1% antibiotic antimycotic cocktail (Cat No. A5955, Sigma, MO, USA). The cells were counted using a hematocytometer and approximately 2⫻106 cells were plated into each flask containing 10 ml of 10% FBS in DMEM and grown at 37 °C and 5% CO2 in an incubator. The cells were subcultured every 2–3 days. For experiments, the BV2 cells were maintained in DMEM without antibiotics or FBS for the required periods of treatment (Basic medium). For RNA isolation and protein extraction, 2⫻106 cells were plated onto cell culture dishes. For siRNA treatment and immunofluorescence, 2⫻105 cells were used per well in a six well plate or 24 well plate, respectively.

Treatment of cell culture Cells were plated onto cell culture dishes and grown in 10% DMEM/FBS with antibiotics overnight. On the following day, the cells were washed twice with PBS, transferred to basic medium and treated with lipopolysaccharide (LPS; 1 ␮g/ml; Cat No. L6529, Sigma, USA), DMS (10 ␮M; Cat No. 310500, Calbiochem, Germany) and with S1P (10 nM; Cat No. S9666, Sigma, USA) in various experimental combinations for different time points (30 min, 1 h, 3 h, 6 h) in the incubator. The control was taken as cells grown in basic medium for the same time periods. The cells and supernatant collected were used for the isolation of RNA and protein, ELISA, immunofluorescence and calorimetric studies.

RNA isolation and reverse transcription Total RNA from BV2 microglial cells subjected to various treatments was extracted as per the instructions given by the Qiagen RNeasy Mini kit (Cat No. 74106, Qiagen, Germany). The quality of the RNA was determined spectrophometrically at 260 and 280 absorbance. The RNA samples were stored at ⫺80 °C until experiments. For cDNA synthesis, 2 ␮g of RNA was mixed with 2 ␮l of Oligo (dT) 15 primer (Cat No. c110A, Promega, Madison, USA) and incubated at 70 °C for 5 min. Each sample was then made to a final volume of 25 ␮l on ice with the following reagents: 1 ␮l of M-MLV reverse transcriptase (Cat No. M170A, Promega, Madison, USA), 5 ␮l of M-MLV RT 5⫻ buffer (Cat No. M531A, Promega, Madison, USA), 0.7 ␮l of RNasin, 0.5 ␮l of dNTP mix (Cat No. U1240, Promega, Madison, USA) and nuclease free water and incubated at 42 °C for 60 min and 70 °C for 10 min. The cDNA thus obtained was then diluted three times in sterile water and stored at ⫺20 °C.

Real time reverse–transcription polymerase chain reaction (RT–PCR) The real time RT–PCR was carried out using LightCycler Fast Start DNA master plus SYBR Green 1 kit as per the manufacturer’s instructions (Cat No. 03515885001, Roche, Mannheim, Germany). The oligonucleotide primer sequences used to amplify TNF-␣, IL-1␤, iNOS, SphK1 and S1P receptors (Kimura et al., 2008) are listed in Table 1. Aliquot (2.5 ␮l) of the each reverse transcription product was added to the 20 ␮l reaction mixture containing LightCyclerFastStart DNA Master SYBR Green I, 0.5 ␮M of each primer to amplify the genes in a LightCycler (Roche, Germany). Each sample was run with a corresponding internal control, ␤-actin. After pre-incubation at 95 °C for 10 min, the PCR was performed as follows: 35– 45 cycles of denaturation at 95 °C for 5 s, annealing at 60 °C for 5 s, and elongation at 72 °C for PCR product size per 25 s. The crossing points (the cycle number at which the Light cycler detected the upstroke of the exponential phase of PCR

134

D. Nayak et al. / Neuroscience 166 (2010) 132–144

Table 2. siRNA sequences used for silencing of SphKI siRNA

Sense

Antisense

1 2 3

GCAAGCAUAUGGAACUUGAtt GGUACGAGCAGGUGACUAtt UGAUACUCACCGAACGGAAtt

UCAAGUUCCAUAUGCUUGCcc UUAGUCACCUGCUCGUACCca UUCCGUUCGGUGAGUAUCAgt

product formation) were taken and normalized with ␤-actin for each sample. Statistical significance was estimated using Student’s t-test and the fold change was calculated using the 2⫺{⌬⌬Ct} method (Livak and Schmittgen, 2001).

Western immunoblot assay Proteins were extracted from BV2 cells subjected to different treatment conditions using protein extraction kit (Cat No. 78501, Pierce, IL, USA) and protease inhibitor cocktail kit (Cat No. 78410, HaltTM, IL, USA) and were quantified using protein assay kit (500 – 0007, Bio-Rad, California, USA). Each protein sample (20 ␮g) was separated on 10% SDS-polyacrylamide gel and transferred to polyvinylidene difluoride transfer membrane. The membranes were blocked with 5% non-fat dry milk for 1 h and incubated with primary antibodies according to manufacturers’ instructions (1:10,000 for mouse anti-actin, Cat No. ab18061, 1:1000 for rabbit polyclonal anti-IL-1␤, Cat No. 9722, Abcam, Cambridge, USA; 1:500 for rabbit polyclonal anti-SphK1, Cat No. AP7237c, Abgent, San Diego, USA) overnight. The membranes were washed with tris buffered saline with 0.1% Tween-20 (TBST) three times and incubated with horseradish peroxidase conjugated goat anti rabbit secondary antibody (Cat No. 7074, Cell Signaling Technology, Boston, MA, USA) for 1 h. The immunoblots were developed using enhanced chemiluminescence detection system (Cat No. 34095, Thermo scientific Pierce Protein Research Products, IL, USA). The membranes were stripped with stripping buffer (Cat No. 0021059, Pierce, IL, USA) and incubated with ␤-actin as internal control and developed. The optical density of the bands was analyzed with Quantity One Software (Bio-Rad, version 4.4.1., California, USA).

Immunofluorescence Cover slips were sterilized and placed into 24 well plates. The cover slips were treated with polylysine (Cat No. P4707, Sigma, St Louis, MO, USA) for 2 h at 4 °C. BV2 cells were subcultured and grown in 10% FBS medium overnight. The next day, the medium was discarded and the cells were washed with PBS twice. The cells were treated as per experimental requirements, then washed and fixed in 4% paraformaldehyde for 30 min at 4 °C. The cells were washed and blocked with normal goat serum at room temperature. The cells were then incubated with rabbit polyclonal anti-SphK1 (1:100, Abgent, Cat No. AP7237c) or rabbit polyclonal anti-TNF␣ (1:1000, Cat No. 3707, Cell Signaling Technology, Boston, MA, USA) primary antibodies overnight at 4 °C. On the following day, the cells were washed 3 times for 10 min each and incubated with the FITC-conjugated goat–anti-rabbit secondary antibody (1:200, Cat No. AQ132F Chemicon, Temecula, CA, USA) or CY3-conjugated goat–anti-rabbit secondary antibody (1: 200, Cat No. AP132C, Chemicon) and FITC-conjugated tomato (Lycopersicon esculentum) lectin (1:800; Sigma-Aldrich, St. Louis, MO, USA; Cat. No. L0401) at room temperature for 1 h in the dark. Subsequently, the cells were counterstained with DAPI (1:50,000, Invitrogen, USA; Cat. No. D1306) for 5 min at room temperature, washed, and mounted onto slides using fluorescent mounting medium. All steps were carried out in the dark. The slides were examined under a confocal microscope (Olympus Fluoview 1000, Tokyo, Japan).

Silencing of SphK1 with siRNA The BV2 cells were subcultured and plated onto 6-well plates at a density of 1⫻105 cells/ml. The total volume used per well was 2 ml. The cells were grown in DMEM with 10% FBS without antibiotics for 24 h to achieve 30%–50% confluency of the cells. The cells were incubated with predesigned siRNA (Table 2; Ambion, CA, USA) mix containing Optimem (Cat No. 31985, GIBCO, Invitrogen) and Oligofectamine (Cat No. 12252-011, Invitrogen) for 8.5 h. The transfected microglia were then treated with different combinations (S1P, LPS, S1P⫹LPS) for the time points followed previously (30 min, 1 h, 3 h, 6 h). The transfection conditions were optimized to obtain silencing efficiency of more than 70%– 80%. The silencing efficiency was calculated by comparison with cells treated with scrambled (negative control) siRNA (Silencer® Select Negative Control siRNA; Cat No. 4390843, Ambion, CA, USA).

Enzyme linked immunosorbent assay (ELISA) for TNF-␣ The TNF-␣ level was determined using a mouse TNF-␣ ELISA kit (Cat No. 88-7324, eBioscience, San Diego, CA, USA). ELISA plates (96 well; NUNC MaxiSorp, eBioscience) were coated with coating buffer (100 ␮l/well), sealed and incubated overnight at 4 °C. The wells were washed 5 times with wash buffer and blocked with 1⫻ assay diluent at room temperature for 1 h. The samples (100 ␮l) collected from BV2 cultures were added to each well and incubated overnight at 4 °C for maximal sensitivity. Subsequently each plate was incubated with 100 ␮l of antibody diluted in 1⫻ assay buffer for 1 h followed by avidin-HRP diluted in 1⫻assay diluent (100 ␮l) for 30 min at room temperature. Finally, each plate was incubated with tetramethylbenzidine (TMB) substrate solution (supplied with TNF-␣ ELISA kit) for 15 min and the reaction was stopped with 50 ␮l of 2 N H2SO4 (Cat No: 320501; Sigma–Aldrich, MO, USA) stop solution. The results were read using a microplate spectrophotometer at 450 nm. The standard curve using recombinant TNF-␣ standards was used to calculate the TNF-␣ production of the samples.

Nitrite assay The nitric oxide production was quantified colorimetrically using the nitric oxide colorimetric bioassay kit (Cat No. N2577-01, USBiological, MA, USA). NUNC Maxisorp 96 well plate was coated with 200 ␮l/well of diluted assay buffer. The supernatant of each (80 ␮l) collected from cultures of BV2 cells treated differentially (LPS, S1P, LPS⫹S1P) and transfected with siRNA was added to each well and mixed with 10 ␮l of the reconstituted enzyme cofactor mixture and 10 ␮l of the reconstituted nitrate reductase mixture sequentially. The plate was covered and incubated at room temperature for 1– 4 h. Griess Reagent R1 (50 ␮l/well) and Griess Reagent R2 (50 ␮l) were added to each well immediately. Color was allowed to develop for 10 min at room temperature and the absorbance was read using a microplate spectrophotometer at 540 nm. The standard curve of Nitrate standards was used to calculate the nitric oxide production of the samples.

D. Nayak et al. / Neuroscience 166 (2010) 132–144

135

Fig. 1. (A–C) Immunofluorescence analysis shows the expression of SphK1 (red) in the cytoplasm of BV2 cells, double labeled with lectin (green). (D, E) RT–PCR analysis of cDNA derived from BV2 cells shows the expression of SphK1 (507 bp) and S1P receptors, S1P1,2,3,4,5 (80 –120 bp). Scale bar, 50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

RESULTS Expression of SphK1 is increased in activated BV2 microglia The protein and mRNA expression of SphK1 was detected in BV2 microglial cells as revealed by the immunofluorescence (Fig. 1A–C) and PCR (Fig. 1D) analyses, respectively. The SphK1-positive cells were confirmed to be microglia as they were colocalized with lectin, the marker for microglia (Fig. 1C). Western blot analysis also revealed that SphK1 was expressed in the microglial cells (Fig. 2G). S1P, the breakdown product of sphingolipid metabolism acts intracellularly as a second messenger and extracellularly via G-protein coupled receptors, S1P1,2,3,4,5 (Ozaki et al., 2003; Rosen and Goetzl, 2005). PCR analysis showed that S1P receptors, S1P1,2,3,4,5 were expressed in BV2 cells (Fig. 1E). SphK1 expression was found to be increased in LPSactivated microglia. Immunofluorescence (IF) analysis showed the increase in SphK1 expression maximally at 1 h after LPS treatment with a return to baseline levels at 6 h post-treatment (Fig. 2A–F). This result was confirmed by the Western blot analysis which showed that the expression of SphK1 in activated microglia increased significantly by 40%, at 1 h post-treatment. However, the increase declined to base level by 6 h post-treatment (Fig. 2G, H). Suppression of SphK1 by DMS reduced the TNF-␣ expression in activated BV2 microglia Effects of suppression of SphK1 by its inhibitor, DMS on TNF-␣ production in LPS-activated BV2 microglia were studied at different time points by IF and RT–PCR. IF showed the increased expression of TNF-␣ in microglia treated with LPS, comparison to that in untreated cells

(Fig. 3A, B, E, F, I, J) and this increase was attenuated by the DMS treatment (Fig. 3D, H, L). Moreover, treatment of microglia with DMS alone also resulted in reduction of TNF-␣ expression compared to that of untreated samples (Fig. 3C, G, K). The real time RT–PCR analysis showed that LPS treatment increased the TNF-␣ mRNA expression in microglia but, upon concomitant suppression of SphK1 with DMS in LPS activated microglia, the level of TNF-␣ mRNA expression was significantly reduced by 51% at 1 h, 18% at 3 h post treatment and was at par with LPS alone treatment at 6 h (Fig. 3M, N). Moreover, in microglia treated with DMS alone, TNF-␣ expression level appears to be reduced or unaltered at different time points tested. Exogenous administration of S1P in BV2 microglia increased the TNF-␣ production The effects of administration of S1P on TNF-␣ expression in BV2 cells was studied by IF and RT–PCR. IF study showed increased expression of TNF-␣ in microglia upon treatment with LPS or S1P at 30 min, 1 h, and 6 h (Fig. 4A–L). When S1P was administered in LPS activated BV2 microglia, the TNF-␣ expression was markedly increased compared to LPS alone treated samples (Fig. 4D, H, L). The real time RT–PCR analysis showed a significant increase in the expression of TNF-␣ in microglia treated with S1P and LPS for 1 h,, 3 h and 6 h, compared to that of untreated microglia (Fig. 4M). Upon addition of S1P to LPS activated microglia, the TNF-␣ mRNA expression level was significantly increased by 126% at 1 h post-treatment, compared to cells treated with S1P alone. ELISA further confirmed that the exogenous administration of S1P increased the TNF-␣ release in untreated and LPS-activated microglia at 1 h, 3 h and 6 h (Fig. 4N). Moreover, the

136

D. Nayak et al. / Neuroscience 166 (2010) 132–144

Fig. 2. (A–F) Immunofluorescence images showing changes in SphK1 expression in microglia after treatment with LPS compared to untreated cells at 30 min, 1 h and 6 h. Western blot analysis (G, H) shows that the SphK1 expression peaks at 1 h post-treatment with LPS, but it returns to baseline levels at 6 h. The data represent the mean⫾SE of at least three independent experiments. Control vs. LPS-treated; * P⬍0.05. Scale bar (A), 50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

release of TNF-␣ was found to be reduced significantly in untreated microglia and LPS-activated microglia with the suppression of SphK1 by DMS (Fig. 4N). Suppression of SphK1 by siRNA reduced TNF-␣ production in activated BV2 microglia Immunofluorescence analysis showed that SphK1 immunoreactivity was markedly reduced in BV2 microglia trans-

fected with SphK1 specific siRNA (SphK1⫺) compared to cells transfected with negative control siRNA (Fig. 5A, B, E, F). The siRNA transfection efficiency in BV2 microglia was found to be 80%, as revealed by the real time RT– PCR (Fig. 5C, D). Administration of S1P increased the expression of TNF-␣ in SphK1⫺ cells treated with or without LPS (Fig. 5H, I). Treatment of LPS alone was unable to increase the expression of TNF-␣ in SphK1⫺ microglial

D. Nayak et al. / Neuroscience 166 (2010) 132–144

137

Fig. 3. (A–L) Immunofluorescence images show that the TNF-␣ expression (green) is reduced with the DMS-mediated suppression of SphK1 in LPS-treated BV2 cells at all time points (30 min, 1 h, 6 h) when compared to cells treated with LPS alone. Note that the marked increase in TNF-␣ expression at 6 h after LPS treatment in BV2 cells (J). (M, N) Real time RT–PCR analysis showed that TNF-␣ mRNA expression level was significantly increased in BV2 cells treated with LPS for 30 min, 1 h, 3 h, and 6 h. The maximum increase was found at 3 h after the treatment. Suppression of SphK1 with DMS in LPS-treated BV2 cells resulted in a significant reduction in TNF-␣ mRNA expression at 30 min, 1 h and 3 h, compared with controls. The data represent the mean⫾SE of at least three independent experiments. The PCR product (205 bp) for TNF-␣ is shown (M). Control vs. LPS-treated; LPS-treated vs. LPS⫹DMS-treated; * P⬍0.05. Scale bar (A–L), 50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

cells (Fig. 5G). The real time RT–PCR analysis showed that TNF-␣ mRNA expression was significantly silenced

(80%) in SphK1⫺ cells, compared to that in negative control (Fig. 5J). LPS treatment was unable to increase TNF-␣

138

D. Nayak et al. / Neuroscience 166 (2010) 132–144

Fig. 4. S1P treatment increased the TNF-␣ expression levels in BV2 cells. (A–L) Immunofluorescence images show the increased expression of TNF-␣ at 30 min, 1 h, and 6 h in BV2 cells treated with S1P or LPS alone or S1P together with LPS. (M) RT–PCR analysis shows the increased mRNA expression level of TNF-␣ in BV2 cells treated with S1P or LPS or S1P⫹LPS at all time points studied (30 min, 1 h, 3 h, 6 h), compared with untreated controls. (N) ELISA further confirms that the TNF-␣ release is increased in BV2 cells treated with S1P alone and S1P together with LPS and is reduced in cells treated with LPS⫹DMS. The data shown are the mean⫾SE of at least three independent experiments. Control vs. LPS/S1P-treated; LPS-treated vs. LPS⫹S1P-treated; LPS-treated vs. LPS⫹DMS-treated. * P⬍0.05. Scale bar (A), 50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

D. Nayak et al. / Neuroscience 166 (2010) 132–144

139

Fig. 5. Suppression of SphK1 (SphK1⫺) with siRNA reduces TNF-␣ production in BV2 microglial cells. (A, B, E, F) Immunofluorescence images show that SphK1 immunoreactivity is markedly reduced in SphK1⫺ cells compared to cells transfected with negative control siRNA. RT–PCR analysis shows that the efficiency of siRNA-mediated suppression of SphK1 is about 80% compared to negative controls (C). Crossing point of PCR cycles is also shown (D). (E–I) Immunofluorescence images further show that the TNF-␣ expression is reduced in untreated or LPS-treated SphK1⫺ cells (F, G) but is increased in SphK1⫺ cells when S1P was administered with or without LPS (H, I). RT–PCR (J) and ELISA (K) analyses showed that TNF-␣ mRNA expression and production were decreased in untreated or LPS-treated SphK1⫺ cells, respectively, compared with negative controls. Upon treatment of SphK1⫺ cells with S1P alone or S1P⫹LPS, the level of TNF-␣ mRNA expression level and release was moderately increased compared with that in untreated or LPS-treated SphK1⫺ cells (J, K). The data represent the mean⫾SE of at least three independent experiments. Negative control vs. treated SphK1⫺ cells; * P⬍0.05. ** P⬍0.01 Scale bar (A, B, E–I), 50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

140

D. Nayak et al. / Neuroscience 166 (2010) 132–144

Fig. 6. RT–PCR analysis showing time-course plot of IL-1␤ mRNA expression in BV2 cells. (A) IL-1␤ (229 bp) was amplified in BV2 microglial cells. (B) LPS treatment increased IL-1␤ expression at all time points studied but upon concomitant suppression of SphK1 with DMS in LPS activated cells, the IL-1␤ expression was significantly reduced until 3 h post treatment. (C) Addition of S1P further increased the IL-1␤ mRNA expression level significantly in LPS-treated cells, compared to cells treated with LPS alone. (D) RT–PCR analysis show that the siRNA-mediated suppression of SphK1 resulted in a reduction of IL-1␤ mRNA expression in untreated and LPS-treated BV 2 cells compared with negative controls. S1P treatment with or without LPS significantly increased the expression of IL-1␤ in SphK1⫺ cells compared with negative controls. The data represent the mean⫾SE of at least three independent experiments. Control vs. LPS-treated; LPS vs. LPS⫹DMS; Control vs. S1P; LPS vs. LPS⫹S1P; Negative control vs. treated. * P⬍0.05, ** P⬍0.01.

D. Nayak et al. / Neuroscience 166 (2010) 132–144

141

Fig. 7. RT PCR analysis showing suppression of SphK1 by DMS reduces iNOS mRNA expression in BV2 cells. (A) 217bp of iNOS PCR product was amplified. (B) LPS treatment increased the iNOS expression in BV2 cells at 6 h compared to control, but upon concomitant suppression of SphK1 with DMS in LPS activated cells, iNOS expression level was reduced significantly compared to cells treated with LPS alone. (C) RT–PCR analysis further shows that LPS-induced increase in iNOS mRNA expression in microglia was further augmented by the concomitant addition of S1P, compared to cells treated with LPS alone. (D) RT–PCR analysis shows that siRNA-mediated suppression of SphK1 (SphK1⫺) decreased iNOS mRNA expression significantly in BV2 cells,

142

D. Nayak et al. / Neuroscience 166 (2010) 132–144

mRNA expression in SphK1⫺ cells compared to negative controls. However, administration of S1P with or without LPS increased the expression of TNF-␣ by 20% in SphK1⫺ cells, compared to that in negative control (Fig. 5J). ELISA analysis confirmed that siRNA-mediated suppression of SphK1 inhibited the TNF-␣ production in untreated and LPS-treated microglia compared to that in negative control cells (Fig. 5K). However, exogenous administration of S1P with or without LPS increased TNF-␣ production in SphK1⫺ cells, compared to that in negative control cells (Fig. 5K). Suppression of SphK1 by DMS or siRNA and exogenous administration of S1P altered the mRNA expression level of IL-1␤ in activated BV2 microglia The real time RT–PCR analysis showed that the IL-1␤ mRNA expression was increased significantly in BV2 microglial cells treated with LPS for 30 min, 1 h, 3 h and 6 h, compared to that of control (Fig. 6A–C). Further, the LPSinduced expression level of IL-1␤ in activated microglia was reduced significantly by concomitant suppression of SphK1 with DMS at 30 min, 1 h and 3 h post-treatment (Fig. 6B). However, the reduction was not significant after 6 h post-treatment. On the other hand, the marked increase in the expression level of IL-1␤ in BV2 microglial cells treated with LPS was further augmented significantly by the concomitant addition of S1P at 1 h, 3 h and 6 h, compared to that of cells treated with LPS alone (Fig. 6C). The maximum increase in expression in cells treated with LPS and S1P was detected at 3 h post-treatment. Moreover, the significant increase of IL-1␤ mRNA expression level in microglia treated with S1P alone was detectable at 3 h post-treatment. Real time RT–PCR analysis also showed that siRNAmediated suppression of SphK1 (SphK1⫺) significantly reduced the IL-1␤ mRNA expression level (40%) in microglia compared to that of negative control (Fig. 6D). LPS treatment was unable to increase IL-1␤ mRNA expression in SphK1⫺ cells compared to negative controls. However, S1P treatment (for the same duration as that of LPS) with and without LPS increased the expression level of IL-1␤ markedly in SphK1⫺ cells, in comparison to that of negative control (Fig. 6D). SphK1 regulates the iNOS mRNA expression and NO production in activated BV2 microglia The real time RT–PCR analysis showed that LPS induced iNOS expression in BV2 cells at 6 h post-treatment (Fig. 7A, B). The LPS-induced increase in the iNOS expression in microglia was attenuated (20%) significantly by the con-

comitant suppression of SphK1 with DMS for 6 h (Fig. 7B). However, DMS was unable to alter the iNOS expression in activated microglia significantly in early time points studied (data not shown). LPS-induced iNOS mRNA expression in BV2 microglia was further augmented significantly by the concomitant addition of S1P (Fig. 7C). Further, the siRNA-mediated suppression of SphK1 (SphK1⫺) reduced the iNOS mRNA expression level (82%) in BV2 microglia, compared with negative controls (Fig. 7D). In addition, treatment of SphK1⫺ cells with LPS, and S1P with or without LPS increased the iNOS expression compared to that observed in untreated SphK1⫺ cells (Fig. 7D). However, this increase level was not above that of negative control. Nitric oxide assay showed the limited production of NO in SphK1⫺ microglia and the increase in production of NO upon administration of SIP (Fig. 7E). In SphK1⫺ cells treated with LPS, the NO levels remained at baseline control levels. However, exogenous administration of S1P with or without LPS increased NO production in SphK1⫺ cells significantly (Fig. 7E).

DISCUSSION Microglial activation is considered the hallmark of neuroinflammation. It is well established that activation of microglia in various neurodegenerative diseases and by exposing the cells to LPS, amyloid ␤-peptide (A␤), thrombin, and IFN-␥ experimentally enhances the release of large amounts of proinflammatory cytokines such as TNF-␣ and IL-1␤ and reactive oxygen intermediates such as ROS and NO, contributing to neuroinflammation and neurodegeneration (Block and Hong, 2005; Combs et al., 2001; Dheen et al., 2007; McCoy and Tansey, 2008; Vilhardt, 2005). Hence, determination of various mechanisms controlling microglial activation is believed to be an important step towards the suppression of neuroinflammation. SphK1, one of the enzymes in the sphingolipid metabolic pathway, is known to play a role in the activation of immune cells (Melendez, 2008), and is involved in TNF-␣ mediated inflammatory processes in immune cells (Hammad et al., 2008; Melendez, 2008). It has been reported that decreased levels of proinflammatory cytokines such as TNF-␣ and IL-1␤ are associated with inhibition of SphK1 (Maines et al., 2008; Niwa et al., 2000). These cytokines in CNS are largely produced by microglial cells in response to LPS, and are controlled by glucocorticoids (Allan and Rothwell, 2001; Nadeau and Rivest, 2002; Streit et al., 1999). The chronic microglial reactivity and uncontrolled production of TNF-␣ and IL-1␤ are the direct causes of the neurodegeneration (Nadeau and Rivest, 2003; Streit et al., 1999). The present results have shown that micro-

compared with negative controls. Treatment of SphK1⫺cells with LPS, and S1P with or without LPS increased the iNOS expression compared to that observed in control SphK1⫺cells (D). However, this increase was not above that of negative control. (E) Nitrite assay shows that in SphK1⫺ cells, the NO production was limited compared to negative control. In SphK1⫺ cells treated with LPS, the NO levels remained at baseline control levels. However, exogenous administration of S1P with or without LPS increased NO production significantly in SphK1⫺ cells. The data represent the mean⫾SE of at least three independent experiments. Negative control vs. SphK1⫺ cells; Control vs. LPS-treated; LPS vs. LPS⫹DMS; LPS vs. LPS⫹S1P. ** P⬍0.005.

D. Nayak et al. / Neuroscience 166 (2010) 132–144

glia, the resident immune cells of the CNS, express SphK1 which acts on sphingosine to form S1P, that in turn regulates diverse biological processes and functions by binding to its receptors (S1P1–5). The results in the present study have further revealed that SphK1 expression is upregulated in LPS-activated microglia. It is therefore plausible that the concomitant increased release of TNF-␣ and IL-1␤ by activated microglia is mediated via sphingosine-1-phosphate pathway as reported in peripheral immune cells (Melendez, 2008). In addition, the BV2 microglial cells in the present study and purified microglia from primary cultures have been shown to express all or some of the five S1P receptors (Tham et al., 2003). The results obtained clearly demonstrate that the sphingosine kinase signaling pathway is involved in the inflammatory response of activated microglia in an autocrine/paracrine signaling fashion in which the secreted S1P can regulate the release of proinflammatory cytokines and NO by microglia. A remarkable finding of this study was that the siRNAmediated or DMS-induced suppression of SphK1 activity in activated microglia inhibited the expression levels of TNF-␣, IL-1␤ and iNOS and release of TNF-␣ and NO and the addition of exogenous S1P to activated microglia enhanced their inflammatory responses, suggesting that S1P acts as an upstream factor which induces the production of proinflammatory cytokines and neurotoxic substances such as NO in activated microglia. However, exogenous addition of S1P could not restore the expression of cytokines completely in LPS-activated SphK1⫺ microglial cells although it enhanced the cytokine expression levels in untransfected LPS-treated BV2 cells. Since many of the biological responses of S1P are mediated via transactivation of S1P receptors (Spiegel and Milstien, 2003; Anliker and Chun, 2004), the present results suggest that the exogenous addition of S1P in SphK1⫺ microglial culture was not sufficient to induce the transactivation of S1P receptors and their downstream signaling pathways. Moreover, the elevated NO release in LPS-activated SphK1⫺ microglial cells in response to exogenous S1P did not correlate with mRNA expression level of iNOS, which is responsible for the rapid production of NO. It is suggested that this discrepancy could be attributed to the differential NO production which may be regulated at the level of protein translation. The finding that the sphingosine signaling pathway is upregulated in activated microglia has important implications not only in inflammatory responses of microglial cells, but for other physiological and pathological processes regulated by S1P in neurons as it is a pleiotropic lipid mediator that regulates many different biological responses, including growth, survival, differentiation, cytoskeleton rearrangements, angiogenesis, vascular maturation, and lymphocyte trafficking (Anliker and Chun, 2004; Olivera and Rivera, 2005; Olivera et al., 2006; Rosen and Goetzl, 2005; Saba and Hla, 2004; Spiegel and Milstien, 2003). Recently the accumulation of the A␤ has been shown to cause neuronal degeneration in Alzheimer’s disease (AD) brains through abnormal sphingolipid metabolism (Okada et al., 2009). It is well established that A␤ in AD brains

143

activates microglia which release large amounts of proinflammatory cytokines and reactive oxygen intermediates, contributing to the neuroinflammation and neurodegeneration (Dheen et al., 2005, 2007). It is possible that this inflammatory response of microglia in AD brains is mediated via S1P signaling pathway.

CONCLUSION In conclusion, it is suggested that suppression of SphK1 activity thereby reducing S1P, in activated microglia may be considered as a possible therapeutic mode for the control of production of proinflammatory cytokines and neurotoxic factors that contribute to the neuroinflammation. Acknowledgments—This research was supported by a research grant (NMRC/1113/2007) from National Medical Research Council, Singapore.

REFERENCES Alemany R, van Koppen CJ, Danneberg K, Ter Braak M, Zu Heringdorf DM (2007) Regulation and functional roles of sphingosine kinases. Naunyn Schmiedebergs Arch Pharmacol 374:413– 428. Allan SM, Rothwell NJ (2001) Cytokines and acute neurodegeneration. Nat Rev Neurosci 2:734 –744. Alvarez SE, Milstien S, Spiegel S (2007) Autocrine and paracrine roles of sphingosine-1-phosphate. Trends Endocrinol Metab 18:300 –307. Anliker B, Chun J (2004) Lysophospholipid G protein-coupled receptors. J Biol Chem 279:20555–20558. Block ML, Hong JS (2005) Microglia and inflammation-mediated neurodegeneration: multiple triggers with a common mechanism. Prog Neurobiol 76:77–98. Bocchini V, Mazzolla R, Barluzzi R, Blasi E, Sick P, Kettenmann H (1992) An immortalized cell line expresses properties of activated microglial cells. J Neurosci Res 31:616 – 621. Bryan L, Kordula T, Spiegel S, Milstien S (2008) Regulation and functions of sphingosine kinases in the brain. Biochim Biophys Acta 1781:459 – 466. Combs CK, Karlo JC, Kao SC, Landreth GE (2001) Beta-amyloid stimulation of microglia and monocytes results in TNF-alpha-dependent expression of inducible nitric oxide synthase and neuronal apoptosis. J Neurosci 21:1179 –1188. Dewil M, Van Den Bosch L, Robberecht W (2007) Microglia in amyotrophic lateral sclerosis. Acta Neurol Belg 107:63–70. Dheen ST, Jun Y, Yan Z, Tay SS, Ling EA (2005) Retinoic acid inhibits expression of TNF-alpha and iNOS in activated rat microglia. Glia 50:21–31. Dheen ST, Kaur C, Ling EA (2007) Microglial activation and its implications in the brain diseases. Curr Med Chem 14:1189 –1197. Gude DR, Alvarez SE, Paugh SW, Mitra P, Yu J, Griffiths R, Barbour SE, Milstien S, Spiegel S (2008) Apoptosis induces expression of sphingosine kinase 1 to release sphingosine-1-phosphate as a “come-and-get-me” signal. FASEB J 22:2629 –2638. Hammad SM, Crellin HG, Wu BX, Melton J, Anelli V, Obeid LM (2008) Dual and distinct roles for sphingosine kinase 1 and sphingosine 1 phosphate in the response to inflammatory stimuli in raw macrophages. Prostaglandins Other Lipid Mediat 85:107–114. Ibrahim FB, Pang SJ, Melendez AJ (2004) Anaphylatoxin signaling in human neutrophils. A key role for sphingosine kinase. J Biol Chem 279:44802– 44811. Jolly PS, Bektas M, Watterson KR, Sankala H, Payne SG, Milstien S, Spiegel S (2005) Expression of SphK1 impairs degranulation and motility of RBL-2H3 mast cells by desensitizing S1P receptors. Blood 105:4736 – 4742.

144

D. Nayak et al. / Neuroscience 166 (2010) 132–144

Kaur C, Hao AJ, Wu CH, Ling EA (2001) Origin of microglia. Microsc Res Tech 54:2–9. Kimura A, Ohmori T, Kashiwakura Y, Ohkawa R, Madoiwa S, Mimuro J, Shimazaki K, Hoshino Y, Yatomi Y, Sakata Y (2008) Antagonism of sphingosine 1-phosphate receptor-2 enhances migration of neural progenitor cells toward an area of brain. Stroke 39:3411–3417. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)) method. Methods 25:402– 408. Maines LW, Fitzpatrick LR, French KJ, Zhuang Y, Xia Z, Keller SN, Upson JJ, Smith CD (2008) Suppression of ulcerative colitis in mice by orally available inhibitors of sphingosine kinase. Dig Dis Sci 53:997–1012. McCoy MK, Tansey MG (2008) TNF signaling inhibition in the CNS: implications for normal brain function and neurodegenerative disease. J Neuroinflammation 5:45. Melendez AJ (2008) Sphingosine kinase signalling in immune cells: potential as novel therapeutic targets. Biochim Biophys Acta 1784:66 –75. Nadeau S, Rivest S (2002) Endotoxemia prevents the cerebral inflammatory wave induced by intraparenchymal lipopolysaccharide injection: role of glucocorticoids and Cd14. J Immunol 169:3370 –3381. Nadeau S, Rivest S (2003) Glucocorticoids play a fundamental role in protecting the brain during innate immune response. J Neurosci 23:5536 –5544. Niwa M, Kozawa O, Matsuno H, Kanamori Y, Hara A, Uematsu T (2000) Tumor necrosis factor-alpha-mediated signal transduction in human neutrophils: involvement of sphingomyelin metabolites in the priming effect of TNF-alpha on the Fmlp-stimulated superoxide production. Life Sci 66:245–256. Okada T, Kajimoto T, Jahangeer S, Nakamura S (2009) Sphingosine kinase/sphingosine 1-phosphate signalling in central nervous system. Cell Signal 21:7–13. Olivera A, Rivera J (2005) Sphingolipids and the balancing of immune cell function: lessons from the mast cell. J Immunol 174:1153–1158.

Olivera A, Urtz N, Mizugishi K, Yamashita Y, Gilfillan AM, Furumoto Y, Gu H, Proia RL, Baumruker T, Rivera J (2006) IgE-dependent activation of sphingosine kinases 1 and 2 and secretion of sphingosine 1-phosphate requires Fyn kinase and contributes to mast cell responses. J Biol Chem 281:2515–2525. Ozaki H, Hla T, Lee MJ (2003) Sphingosine-1-phosphate signaling in endothelial activation. J Atheroscler Thromb 10:125–131. Pyne S, Lee SC, Long J, Pyne NJ (2009) Role of sphingosine kinases and lipid phosphate phosphatases in regulating spatialsphingosine 1-phosphate signalling in health and disease. Cell Signal 21: 14 –21. Rock RB, Peterson PK (2006) Microglia as a pharmacological target in infectious and inflammatory diseases of the brain. J Neuroimmune Pharmacol 1:117–126. Rosen H, Goetzl EJ (2005) Sphingosine 1-phosphate and its receptors: an autocrine and paracrine network. Nat Rev Immunol 5: 560 –570. Saba JD, Hla T (2004) Point-counterpoint of sphingosine 1-phosphate metabolism. Circ Res 94:724 –734. Shie FS, Woltjer RL (2007) Manipulation of microglial activation as a therapeutic strategy in Alzheimer’s disease. Curr Med Chem 14: 2865–2871. Spiegel S, Milstien S (2003) Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol 4:397– 407. Streit WJ, Walter SA, Pennell NA (1999) Reactive microgliosis. Prog Neurobiol 57:563–581. Tham CS, Lin FF, Rao TS, Yu N, Webb M (2003) Microglial activation state and lysophospholipid acid receptor expression. Int J Dev Neurosci 21:431– 443. Vilhardt F (2005) Microglia: phagocyte and glia cell. Int J Biochem Cell Biol 37:17–21. Wattenberg BW, Pitson SM, Raben DM (2006) The sphingosine and diacylglycerol kinase superfamily of signaling kinases: localization as a key to signaling function. J Lipid Res 47:1128 –1139.

(Accepted 7 December 2009) (Available online 28 December 2009)