SEARCH

SEARCH BY CITATION

Keywords:

  • CD38;
  • ADP-ribosyl cyclase;
  • intracellular calcium concentration;
  • N9 microglial cells;
  • nitric oxide;
  • tumor necrosis factor α

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Lipopolysaccharide, the main component of the cell wall of Gram-negative bacteria, is known to activate microglial cells following its interaction with the CD14/Toll-like receptor complex (TLR-4). The activation pathway triggered by lipopolysaccharide in microglia involves enhanced basal levels of intracellular calcium ([Ca2+]i) and terminates with increased generation of cytokines/chemokines and nitric oxide. Here we demonstrate that in lipopolysaccharide-stimulated murine N9 microglial cells, cyclic ADP-ribose, a universal and potent Ca2+ mobiliser generated from NAD+ by ADP-ribosyl cyclases (ADPRC), behaves as a second messenger in the cell activation pathway. Lipopolysaccharide induced phosphorylation, mediated by multiple protein kinases, of the mammalian ADPRC CD38, which resulted in significantly enhanced ADPRC activity and in a 1.7-fold increase in the concentration of intracellular cyclic ADP-ribose. This event was paralleled by doubling of the basal [Ca2+]i levels, which was largely prevented by the cyclic ADP-ribose antagonists 8-Br-cyclic ADP-ribose and ryanodine (by 75% and 88%, respectively). Both antagonists inhibited, although incompletely, functional events downstream of the lipopolysaccharide-induced microglia-activating pathway, i.e. expression of inducible nitric oxide synthase, overproduction and release of nitric oxide and of tumor necrosis factor α. The identification of cyclic ADP-ribose as a key signal metabolite in the complex cascade of events triggered by lipopolysaccharide and eventually leading to enhanced generation of pro-inflammatory molecules may suggest a new therapeutic target for treatment of neurodegenerative diseases related to microglia activation.

Abbreviations used
[Ca2+]i

intracellular calcium concentration

ADPR

ADP-ribose

cADPR

cyclic ADP-ribose

ADPRC

ADP-ribosyl cyclase

PKA

cyclic AMP-activated protein kinase

PKC

protein kinase C

SDS-PAGE

sodium dodecyl sulfate-polyacrylamide gel electrophoresis

IMDM

Iscove's Modified Dulbecco's Medium

MoAb

monoclonal antibody

iNOS

inducible nitric oxide synthase

TLR-4

Toll-like receptor 4

TNF-α

tumor necrosis factor-α

Microglial cells, the monocyte/macrophage equivalent of the CNS, represent the first line of defence in the brain, capable of removing infectious agents and damaged cells (Farber and Kettenmann 2005). In response to pathological events, the normally resting microglia gradually transforms into motile, secretory and potentially cytotoxic phagocytes (Hoffmann et al. 2003). Activated microglial cells then participate in mechanisms of innate and immune defence, tissue repair and neuroprotection (Town et al. 2005). However, under conditions of chronic inflammation, excessive activation of microglia can contribute to the neurodegenerative process by producing and releasing a number of potentially cytotoxic substances, which include pro-inflammatory cytokines and nitric oxide (Aschner et al. 1999; Gonzalez-Scarano and Baltuch 1999). Recently, microglial cells have been implicated in the development of ischaemia-induced damage and neurodegenerative diseases such as multiple sclerosis, Alzheimer's disease and Parkinson's disease (Gebicke-Haerter 2001; Town et al. 2005). Accordingly, the identification of the molecular mechanisms of microglial activation might provide new therapeutic targets, representing an important step towards the successful treatment of these diseases.

Lipopolysaccharide, the main component of the cell wall of Gram-negative bacteria, is widely known to activate microglial cells ‘in vitro’ upon interacting with the CD14/Toll-like receptor (TLR-4) complex (Raetz and Whitfield 2001). Lipopolysaccharide activation of murine microglia leads to an elevation of the intracellular free calcium concentration ([Ca2+]i). This event is necessary to allow the overproduction and the release of inflammatory cytokines and nitric oxide to occur (Hoffmann et al. 2003). Despite its importance in physiology and pathology, the mechanisms and the chemical mediators of this lipopolysaccharide-induced signalling pathway seem to be multiple and are still incompletely known.

Cyclic ADP-ribose (cADPR) is a potent and universal Ca2+ mobiliser from ryanodine-sensitive stores/channels which is generated by a family of multifunctional enzymes designated ADP-ribosyl cyclases (Lee et al. 1989; Mehta and Malavasi 2000; Lee 2002; Schuber and Lund 2004). CD38, the most represented mammalian member of this family, is a multifunctional type II glycoprotein known to catalyse, at the surface of several mammalian cells but also intracellularly, the production of a number of signal molecules involved in the regulation of intracellular calcium concentration ([Ca2+]i) levels. Specifically, CD38 generates cADPR and ADP-ribose (ADPR) from the substrate NAD+ via ADP-ribosyl cyclase and cADPR hydrolase activities, respectively (Lee et al. 1989; Lee 2002; Guse 2005).

The lack of information about the CD38/cADPR system in microglial cells, and the recent finding that cADPR is a second messenger in the lipopolysaccharide-stimulated proliferation of human peripheral blood mononuclear cells (Bruzzone et al. 2003), prompted us to investigate whether cADPR plays a role in the basal [Ca2+]i elevation during the lipopolysaccharide-induced microglial activation process (Hoffmann et al. 2003). The results obtained in N9 cells, a murine microglial cell line, demonstrate that CD38 is constitutively expressed in these cells and that lipopolysaccharide stimulates its phosphorylation, resulting in increased ADP-ribosyl cyclase activity and enhanced generation of intracellular cADPR. This signal metabolite elicits a basal [Ca2+]i increase, which triggers production and release of nitric oxide and the pro-inflammatory cytokine tumor necrosis factor α (TNF-α). Therefore, cADPR behaves as a second messenger in the lipopolysaccharide-stimulated generation of nitric oxide and TNF-α in N9 microglial cells.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

Protein kinase inhibitors (staurosporine; protein kinase A (PKA) inhibitor peptide sequence 14–22, myristoylated; protein kinase C (PKC) inhibitor peptide sequence 20–28, myristoylated; tyrphostin (AG126)) were obtained from Calbiochem (Milan, Italy). The IB4 anti-CD38 monoclonal antibody (MoAb) was kindly provided by Prof. F. Malavasi, Turin, Italy. Rabbit anti-inducible nitric oxide synthase (NOS2) polyclonal antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-phosphoserine (clone PSR-45) MoAb was obtained from Sigma (Milan, Italy). Anti-phosphotyrosine (p-Tyr-100) MoAb was obtained from Cell Signalling Technology, Inc. (Danvers, MA, USA). Human recombinant CD38 was a generous gift from Prof. H. C. Lee (University of Minnesota, Minneapolis, MN, USA). TNF-α was obtained from ICN Biomedical Inc. (Milan, Italy). Iscove's Modified Dulbecco's Medium (IMDM) was purchased from Cambrex BioScience (Milan, Italy). All other chemicals were obtained from Sigma.

Cell cultures

Murine N9 microglial cells, originally developed by Prof. P. Ricciardi-Castagnoli (Righi et al. 1989) and kindly provided by Prof. M. Matteoli, Milan, Italy, were grown in IMDM with 25 mm HEPES and l-glutamine, supplemented with 5% foetal calf serum, 100 IU/mL penicillin, 100 μg/mL streptomycin and 50 nmβ-mercaptoethanol. Cells were cultured in a humidified 5% CO2 atmosphere at 37 °C.

The TNF-α sensitive murine fibroblast cell line L929 was obtained from the American Type Culture Collection (Rockville, MD, USA) and cultured at 37 °C in a 5% CO2 atmosphere in D-minimal essential medium containing 4 mm glutamine supplemented with 10% foetal calf serum, 100 U/mL penicillin and 100 μg/mL streptomycin.

Assays of CD38 enzymatic activities

N9 cells were pretreated by incubating flasks in the presence or absence of 200 ng/mL lipopolysaccharide for different times (from 1 to 24 h). To elucidate the effects of lipopolysaccharide on representative enzymatic activities of CD38, N9 cells were also incubated in the presence or absence of different kinase inhibitors. Staurosporine (50 nm) was supplemented for 2 h before lipopolysaccharide addition and cells were co-incubated for further 24 h. PKA-specific peptide inhibitors (1 μm), or 10 μm PKC-specific peptide inhibitors, or 50 μm tyrphostin AG126, or combinations thereof were supplemented for 2 h, and cells were co-incubated with lipopolysaccharide for another 2 h. The media were then removed and supplemented for a further 24 h with fresh media containing only the protein kinase inhibitors as above. The ectocellular cyclase activity of CD38 was measured using NGD+ instead of NAD+ as a surrogate substrate, because the product cyclic GDP-ribose (cGDPR) is not susceptible to enzymatic hydrolysis (Graeff et al. 1994). cADPR hydrolase was assayed under the same conditions by measuring the conversion of cADPR to ADPR. The NAD+-ase activity, reflecting the sequential reactions catalysed by the ADP-ribosyl cyclase and cADPR hydrolase activities, was measured by following the conversion of NAD+ to nicotinamide and ADPR (see below).

NAD+-ase, GDP-ribosyl cyclase and cADPR hydrolase activities were assayed on 0.2 mm NAD+, 0.2 mm NGD+, or 0.2 mm cADPR, respectively, by incubating intact N9 cells at 8 mg/mL in 0.4 mL of Krebs Ringers buffer (150 mm NaCl, 5 mm KCl, 1.2 mm MgSO4, 2 mm CaCl2, 10 mm glucose, 10 mm HEPES; pH 7.4) at 37 °C. At different times, 60-μL aliquots of the incubations were withdrawn and centrifuged for 30 s at 5000 × g. The corresponding supernatants were then deproteinised with trichloroacetic acid (5% final concentration), centrifuged and the excess trichloroacetic acid removed with diethyl ether. Samples were analysed by HPLC (Franco et al. 2001) to estimate the corresponding products generated by the three enzyme activities: ADPR for NAD+-ase, cGDPR for GDP-ribosyl cyclase and ADPR from cADPR for cADPR-hydrolase. These were uniformly expressed as pmol product/min/mg of protein. Protein concentration was measured according to Bradford (1976).

FACS analyses

N9 cells (106) were pretreated in the presence or absence of 200 ng/mL lipopolysaccharide for 24 h. Cells were then recovered by trypsin treatment, washed once in IMDM, resuspended in 200 μL IMDM and incubated with the IB4 anti-CD38 MoAb (15 μg/mL) for 30 min in ice. Cells were then washed and incubated for the same time with 10 μg/mL fluorescein isothiocyanate-conjugated anti-mouse IgG in IMDM. Fluorescence intensity was determined on washed samples by flow cytometry using a FACScan (Coulter, Epics XL, Milan, Italy). Control samples were incubated with the secondary antibody only.

Immunocytochemistry of CD38 and of inducible nitric oxide synthase (iNOS)

N9 cells, cultured in complete medium as already described, were incubated with 200 ng/mL lipopolysaccharide and then properly stained for analysis of CD38 and iNOS positivity by confocal microscopy. Cells were seeded in 4-well Laboratory-Teck chamber slides (Nalge Nunc Int., Naperville, IL, USA) at 105 cells/well. After 24 h, cells were incubated with or without lipopolysaccharide at 37 °C for 18 h, then washed three times with phosphate-buffered saline, pH 7.4, and fixed for 30 min with 4% paraformaldehyde in phosphate-buffered saline at 25 °C. The wells were then washed five times with phosphate-buffered saline and cells were permeabilised with 0.05% Triton X100 in phosphate-buffered saline for 30 min at 25 °C. Cells were again washed five times with phosphate-buffered saline and stained with the primary antibodies (goat anti-iNOS from Santa Cruz Biotechnology; IB4 anti-CD38 MoAb, see above), diluted up a final concentration of 1.0–1.5 μg/mL, followed by anti-mouse Alexa 488-conjugated or anti-goat Alexa 633-conjugated secondary antibodies (Molecular Probes, Invitrogen, Carlsbad, CA, USA), diluted to a final concentration of 1 μg/mL. Finally, cover glasses were mounted by using a 9 : 1 solution of glycerol-phosphate-buffered saline.

Images were obtained using a Leica TCS SL confocal microscope equipped with argon/He-Ne laser sources and a HCX PL APO CS 63.0 × 1.40 oil objective. An energy laser of 25%, for the 488 nm line, and of 50%, for the 633 nm line, were applied to the specimens during acquisition. Data were acquired in an emission range of 500–600 nm for Alexa-488-labelled secondary antibody used for CD38 staining, and an emission range of 645–750 nm for Alexa-633-labelled secondary antibody used for iNOS staining. Photomultiplier voltage gains were set to eliminate autofluorescence of cells in the same intervals of acquisition. Pinhole settings were adjusted to ensure the best confocality of the instrument.

Stacks of 50 sections with a Z-step of 122 nm for a total thickness of 5.5–6 μm were taken for each image (4× digital zoom). Finally, average projections were calculated using Leica LCS software (Wetzlar, Germany).

Immunochemical detection of native CD38 and iNOS

Adherent N9 cells (15 × 106 for each determination) were incubated in the presence or absence of 200 ng/mL of lipopolysaccharide for 24 h. Cells were recovered by scraping and washed twice with 5 mL Krebs Ringers buffer. Cell pellets were resuspended in 200 μL of de-ionised water containing protease inhibitors (10 μg/mL leupeptin; 10 μg/mL aprotinin; 20 μg/mL pepstatin and 10 μm aminoethyl-benzene sulphonyl fluoride hydrochloride), lysed by sonication (30 s at 3 W on ice) and the protein concentration measured as described (Bradford 1976). Aliquots of 200 μg of each sample were subjected to 10% SDS polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analyses. Immunochemical detection of CD38 was obtained with the IB4, anti-CD38 MoAb, using anti-mouse IgG (Amersham, Milan, Italy) as a secondary antibody and following the instructions in the Amersham ECL kit. Western blots incubated with secondary antibody alone failed to give any signal. Likewise, no signal whatsoever was observed when western blots were incubated with IB4 in the presence of excess human recombinant CD38. The immunoenzymatic detection of iNOS was performed using a rabbit anti-iNOS (NOS2 polyclonal antibody) and an anti-rabbit IgG (Santa Cruz Biotechnology) as secondary antibody, according to instructions in the Amersham ECL kit. No signal was observed in western blots incubated with the secondary antibody only, or when excess soluble i-NOS (murine recombinant, from Sigma) was present with the rabbit anti-iNOS polyclonal antibody.

Western blot analysis of phosphorylated immunopurified CD38

Samples were prepared by incubating N9 cells (5 × 107 cells for each assay) for 24 h in the presence or absence of lipopolysaccharide; before lipopolysaccharide addition, samples were also pre-incubated for 2 h with or without staurosporine or the other protein kinase inhibitors listed above. After 24 h incubation with lipopolysaccharide, cells were recovered by scraping, washed twice with 5 mL of Krebs Ringers buffer, and cell pellets resuspended in 300 μL of de-ionised water containing protease inhibitors (see above). Samples were lysed by sonication (30 s at 3 W on ice), protein concentration was determined and 2-mg aliquots of each sample, diluted in 250 μL of buffer A (10 mm Tris, pH 6.5, with 0.05% Triton) were loaded on 900 μL of a Sepharose resin previously coated with anti-CD38 (IB4) MoAb as described (Zocchi et al. 1993). To allow binding of CD38 to the immobilised MoAb, cells lysates were incubated for 18 h at 4 °C on a rotating plate. Samples were then centrifuged and Sepharose was washed three times with 2 mL of buffer A. Nonspecifically bound proteins were removed by two sequential pre-elution washings of the resin by centrifugation: the first was performed with 0.5 m KCl in buffer A and the second with 1 m KCl in buffer A. The resin was then washed four more times with 2 mL of buffer A. Elution of specifically bound proteins was then obtained by adding 50 μL of Laemmli sample buffer 4× (8% SDS, 0.4 g/mL sucrose, 80 μg/mL bromophenol blue, 62.5 nm Tris-HCl, pH 6.8), at 90 °C for 10 min. Samples were then centrifuged and the supernatants loaded on a 10% SDS-PAGE as described (Franco et al. 2001). The subsequent western blot analyses and immunoenzymatic detection with either an anti-phosphoserine, an anti-phosphotyrosine or the IB4 anti-CD38 MoAb were performed according to instructions in the Amersham ECL kit. Bioluminescence intensity was estimated by the Chemidoc System (Bio-Rad, Milan, Italy).

Fluorimetric determination of the [Ca2+]i

Adherent N9 cells were cultured on 20-mm diameter coverslips, in the presence or absence of 200 ng/mL lipopolysaccharide for 24 h, following a 1-h pre-incubation with or without 50 μm ryanodine or 100 μm 8-Br-cADPR. At the end of the lipopolysaccharide treatment, cells were incubated with Fura 2-AM for 45 min at 37 °C and the [Ca2+]i was measured as described (Zocchi et al. 1998).

Determination of intracellular cADPR levels

Adherent N9 cells (5 × 107) were incubated for 24 h in IMDM in the absence (control) or presence of 200 ng/mL lipopolysaccharide. Cells were recovered by trypsin treatment, washed once with Krebs Ringers buffer, and cell pellets were lysed in 300 μL of de-ionised water; 20-μL aliquots were withdrawn for protein assay (Bradford 1976), while 0.6 m perchloric acid was added at 4 °C to the rest of the sample volume. After centrifugation (1500 × g for 10 min), the cADPR content was measured on the neutralised cell extracts by means of a highly sensitive enzymatic cycling assay (Graeff et al. 1994) and expressed as pmol/mg of cell protein.

Determination of iNOS activity

N9 cells (0.25 × 106/mL) were incubated at 37 °C in 24-well plates in IMDM in the absence (control) or presence of 200 ng/mL lipopolysaccharide or 100 μm cADPR. Before addition of lipopolysaccharide, duplicate samples were pre-incubated with or without 100 μm 8-Br-cADPR, 10 mm nicotinamide or 50 μm ryanodine for 1 h. After 24 h, 500 μL of each culture supernatant was recovered and centrifuged at 5000 × g for 5 min. Three 100-μL aliquots of each sample were then incubated in triplicate in the presence of 0.02 mm FAD, 0.2 mm NADPH and 0.0001 U nitrate reductase from Aspergillus niger at 37 °C for 30 min to obtain a complete reduction of nitrates to nitrites. Samples were then diluted with an equal volume of Griess reagent (0.1% N-(1-naphthyl)ethylenediamine dihydrochloride, 1% sulfanylamide and 2.5% H3PO4) (Ignarro et al. 1987), and the absorbance at 545 nm was estimated after 10 min incubation at 25 °C. Nitrite production was interpolated from a standard curve generated in parallel with known amounts of NaNO2.

Assays of TNF-α

Two procedures of TNF-α assay were followed, using a mouse TNF-α (Mono/Poly) OptEIA kit (cat. no. 559732, BD Biosciences, Pharmingen, San Diego, CA, USA) and a cytotoxicity-based assay (Vercammen et al. 1997), respectively. The two procedures yielded quite comparable results, but the second one could not be performed to measure TNF-α release following pre-incubation of N9 cells with ryanodine because of cytotoxicity of this compound on the murine fibroblast cell line L929 used in this type of assay.

N9 cells (0.3 × 106/mL) were seeded in 24-well plates and incubated for 24 h at 37 °C in IMDM in the absence (control) or presence of 200 ng/mL lipopolysaccharide or 100 μm cADPR. Before addition of lipopolysaccharide, cells were pre-incubated for 1 h with or without each of the following compounds: 100 μm 8-Br-cADPR, 50 μm ryanodine, 10 mm nicotinamide, 1 μm I-PKA, 10 μm I-PKC, 50 μm tyrphostin AG126 or 50 nm staurosporine. The concentration of TNF-α secreted in the culture media after N9 stimulation was quantified by means of the EIA assay. Alternatively, TNF-α released by stimulated N9 cells was assayed by measuring its apoptotic effect on the murine fibroblast cell line L929 (Vercammen et al. 1997). For this purpose, target cells were seeded in 96-well plates at a concentration of 3 × 104 cells/well; after 18 h, their culture medium was removed and the N9 cell culture supernatants, diluted 10-fold in fresh medium, were added to the wells together with 1 μg/mL actinomycin D. A standard curve of TNF-α toxicity was obtained in parallel by adding recombinant murine TNF-α at concentrations ranging from 25 to 800 pg/mL. After 24 h culture, cell viability was determined with the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide]-reduction assay (Mossman 1983). Briefly, 20 μL of a 5 mg/mL solution of MTT were added to each well, the plates were further incubated for 90 min at 37 °C, the supernatant was then carefully removed from each well and the blue formazan crystals were solubilised with 200 μL/well of dimethylsulfoxide. The optical absorbance at 570 nm was determined with a Bio-Rad plate reader.

Statistical analyses

All parameters were tested by paired t-test. [Ca2+]i-values were analysed with one-way anova and two-sided Dunnett's t-test. p-values < 0.05 were considered significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Lipopolysaccharide induces an increase of CD38 enzymatic activities in intact N9 cells

To investigate whether CD38 plays a role in the reported increase of basal [Ca2+]i that follows lipopolysaccharide stimulation of microglial cells (Hoffmann et al. 2003), we first measured the levels of representative CD38 ectoenzyme activities in lipopolysaccharide-stimulated and in untreated N9 cells, respectively. As shown in Fig. 1, incubation of cells with lipopolysaccharide (200 ng/mL) induced a progressive increase of ectocellular GDP-ribosyl cyclase activity, with the highest levels being recorded after 24 h incubation (p < 0.005). Ectocellular NAD+-ase and cADPR-hydrolase activities were measured under the same conditions and showed a time-dependent increase comparable to that of the GDP-ribosyl cyclase activity. Levels of NAD+-ase rose from 32.76 ± 4.15 pmol ADPR/min/mg of protein (untreated cells) to 52.32 ± 8.36 (lipopolysaccharide-treated cells, 24 h; p < 0.01). cADPR hydrolase activity was 5.12 ± 0.44 pmol ADPR/min/mg of protein (untreated cells) and 9.11 ± 0.51 (lipopolysaccharide-treated cells, 24 h; p < 0.01).

image

Figure 1.  Effect of lipopolysaccharide on the ectocellular GDP-ribosyl cyclase activity of N9 cells. Adherent N9 cells (25 × 106) were exposed to lipopolysaccharide (200 ng/mL) for the times indicated. At the end of each incubation, cells were recovered from culture flasks by scraping and the ectocellular GDP-ribosyl cyclase activity was measured on intact cells, as described under Materials and methods. Results are expressed as cGDPR production relative to untreated controls (2.7 ± 0.6 pmol cGDPR/min/mg protein) and are the mean ± SD of seven experiments.

Download figure to PowerPoint

Next, we investigated whether the three ectoenzyme activities were featured by CD38. Confocal images of N9 cells, stained with the anti-human CD38 MoAb IB4, revealed diffuse immunoreactivity of all cells for CD38 (Fig. 2), very similar to patterns previously observed in rat hippocampal astrocyte cultures (Verderio et al. 2001). Levels of CD38 expression were comparable in control and lipopolysaccharide-treated (200 ng/mL for 24 h) N9 cells (Fig. 2 and b, respectively). Likewise, FACS analyses performed with IB4 MoAb showed CD38 immunoreactivity in N9 cells and comparable levels of expression of immunoreactive CD38 in control and lipopolysaccharide-treated N9 cells (not shown).

image

Figure 2.  Immunocytochemical localization of CD38 in control and lipopolysaccharide-treated N9 cells. Adherent N9 cells (2 × 105) were exposed to lipopolysaccharide (200 ng/mL) for 18 h. At the end of incubation, cells were fixed, permeabilised and stained for CD38 positivity using the IB4 MoAb followed by an Alexa-488 labelled secondary antibody. Images were obtained with a Leica TCS-SL confocal microscope and represent average projections of Z-stack files of 50 sections elaborated by using Leica LCS software. a, control, untreated cells. b, lipopolysaccharide-treated cells.

Download figure to PowerPoint

Lipopolysaccharide induces CD38 phosphorylation

These results prompted us to investigate whether the observed increase of CD38 enzymatic activities in the lipopolysaccharide-treated N9 cells was due to enzyme activation by post-translational modifications. For this purpose, ectocellular GDP-ribosyl cyclase activity of lipopolysaccharide-treated and control N9 cells was first determined in the presence or absence of 50 nm staurosporine, a nonspecific inhibitor of protein kinases. As shown in Fig. 3a, the stimulatory effect of lipopolysaccharide on GDP-ribosyl cyclase (i.e. doubling of activity) was almost completely abrogated by staurosporine. Cells incubated with staurosporine alone showed similar cyclase activity compared to untreated, control cells (not shown), indicating that (i) this broad protein kinase inhibitor did not affect basal CD38 enzymatic activity and (ii) the increase of cyclase activity was due to kinase-dependent cyclase activation.

image

Figure 3.  Effect of protein kinase inhibitors on the lipopolysaccharide-induced activation of GDP-ribosyl cyclase and either serine or tyrosine CD38 phosphorylation in N9 cells. a, N9 cells (25 × 106 per sample) were pre-incubated for 2 h with each of the protein kinase inhibitors indicated on the abscissa, before being stimulated (or not) with lipopolysaccharide (200 ng/mL) for 24 h, as described under Materials and methods. Cells were detached with trypsin, and ectocellular GDP-ribosyl cyclase activity was measured on intact cells as described under Materials and methods. Results are expressed as cGDPR production relative to control, not lipopolysaccharide-stimulated cells (see legend to Fig. 1) and are the mean ± SD of five experiments. Treatment with each of the protein kinase inhibitors did not significantly modify basal values of GDP-ribosyl cyclase activity in control, not lipopolysaccharide-stimulated, cells. b, For experimental details on CD38 immunopurification, see Materials and methods. Lanes i and vii: control, untreated N9 cells. Lanes ii and viii: cells incubated with lipopolysaccharide (200 ng/mL) for 24 h. Lane iii: cells pre-incubated with 50 nm staurosporine for 2 h and then incubated with lipopolysaccharide for 24 h. Lane iv: cells pre-incubated with 1 µm I-PKA for 2 h before lipopolysaccharide treatment. Lane v: cells pre-incubated with 10 µm I-PKC for 2 h before lipopolysaccharide treatment. Lane vi: cells pre-incubated with 1 µm I-PKA, 10 µm I-PKC and 50 µm AG126 for 2 h before lipopolysaccharide treatment. A representative experiment is shown of three giving comparable results. Lanes i-vi, immunodetection was performed with an anti-phosphoserine MoAb. Lanes vii and viii, immunodetection was with the IB4 anti-CD38 MoAb. The graph displayed below the western blots presents the bioluminescence intensity quantified using the Chemidoc System (see Materials and methods): values are expressed as optical density/mm2 and are the mean ± SD of the three experiments. c: lanes i and vi, control, untreated cells; lanes ii and vii, cells incubated with lipopolysaccharide (200 ng/mL) for 24 h; lane iii, cells pre-incubated with 50 nm staurosporine for 2 h and then incubated with lipopolysaccharide for 24 h; lane iv, cells pre-incubated with 50 µm AG126 for 2 h before lipopolysaccharide treatment; lane v, cells pre-incubated with 1 µm I-PKA, 10 µm I-PKC and 50 µm AG126 for 2 h before lipopolysaccharide treatment. A representative experiment is shown of three giving comparable results. Immunodetection was performed with an anti-phosphotyrosine MoAb (lanes i–v) or with the IB4 anti-CD38 MoAb (lanes vi and vii). The graph presents the bioluminescence intensity quantified using the Chemidoc System (see Panel b): values are the mean ± SD.

Download figure to PowerPoint

To investigate in more detail the mechanisms of CD38 phosphorylation leading to lipopolysaccharide-induced stimulation of its enzymatic activity, CD38 was immunopurified from N9 cells incubated for 24 h in the presence of lipopolysaccharide (200 ng/mL), or of staurosporine (50 nm) or of both compounds. Figure 3b shows a representative Western blot analysis of CD38, as developed with an anti-phosphoserine MoAb. Compared with control, untreated cells, lipopolysaccharide-incubated N9 cells showed a remarkably higher extent of CD38 phosphorylation, which was almost completely abolished by the presence of staurosporine during incubation of the cells with lipopolysaccharide (Fig. 3b, lanes i–iii). In an attempt to identify the protein kinase(s) responsible for the lipopolysaccharide-induced stimulation of GDP ribosyl cyclase activity, experiments were then carried out in the presence of different protein kinase-specific inhibitors. Indeed, the amino acid sequence of the intracellular domain of murine CD38 has three serine residues at positions 7, 10 and 19 and one tyrosine at position 4 (Harada et al. 1993). In these experiments, N9 cells were incubated in the presence or absence of the kinase inhibitors for 24 h; stimulation with lipopolysaccharide (200 ng/mL) occurred during the first 2 h only. Exposure of N9 cells to 1 μm I-PKA or 10 μm I-PKC, during the incubation with lipopolysaccharide and for the following 24 h, resulted in a partial inhibition of lipopolysaccharide-mediated cyclase activation (50% and 66%, respectively, Fig. 3a). The effect of the two protein kinase inhibitors together was not additive (not shown), indicating that PKA and PKC act sequentially, rather than synergistically, on CD38.

The partial extent of inhibition of CD38 activation observed with I-PKA and I-PKC suggested the likely involvement of other kinase(s), in addition to PKA and PKC, in lipopolysaccharide-triggered CD38 phosphorylation. This prompted us to explore the role of a protein tyrosine kinase sensitive to tyrphostin AG126, which has been recently demonstrated to be involved in the [Ca2+]i increase induced by lipopolysaccharide in microglial cells (Kann et al. 2004). Pre-incubation with AG126 reduced (approximately by 40%) the extent of lipopolysaccharide-induced GDP-ribosyl cyclase activation (not shown). Indeed, presence of I-PKA, I-PKC and AG126 (50 μm) together during lipopolysaccharide priming almost completely prevented CD38 cyclase activation (94% inhibition) (Fig. 3a), whereas each of the compounds alone achieved only a partial inhibition of CD38 activation. Thus, lipopolysaccharide-triggered CD38 phosphorylation appears to involve PKA, PKC and the AG126-inhibitable protein tyrosine kinase. Analysis of serine phosphorylation of CD38 in the presence of the various protein kinase inhibitors is shown in Fig. 3b. A more intense band (3-fold increase, as measured by the Chemidoc System) appeared in the lipopolysaccharide-treated sample, compared with the control from lipopolysaccharide-untreated N9 cells, at the Mr corresponding to immunoreactive CD38 band (46 kDa, lanes vii and viii). The phosphoserine signal was much weaker, and close to that of the control, in the samples from N9 cells treated with the various protein kinase inhibitors, added either individually or in combination (Fig. 3b). The inhibiting effect of AG126 on the lipopolysaccharide-induced stimulation of GDP-ribosyl cyclase activity and the presence of a tyrosine residue at position 4 in murine CD38 (Harada et al. 1993) prompted us to analyse the western blots using an anti-phosphotyrosine antibody. Inspection of Fig. 3c clearly shows that the basally low phosphotyrosine signal of CD38 was enhanced by lipopolysaccharide treatment of the N9 cells and returned to low levels, approaching those of the control cells, when cells were pre-incubated with AG126, staurosporine or a combination of the various protein kinase inhibitors, before lipopolysaccharide stimulation.

Role of intracellular cADPR in the lipopolysaccharide-induced [Ca2+]i increase in N9 cells

The enhancement of CD38 enzymatic activities induced by lipopolysaccharide was paralleled by an increase of the intracellular cADPR concentration ([cADPR]i) from 2.12 ± 0.25 pmol/mg in control cells up to 3.58 ± 0.41 pmol/mg in cells exposed to 200 ng/mL lipopolysaccharide for 24 h (p < 0.005). To establish whether the reported increase of the basal [Ca2+]i induced by lipopolysaccharide in microglial cells (Hoffmann et al. 2003) was related to the increase of the [cADPR]i, we measured the [Ca2+]i in cells incubated with lipopolysaccharide in the absence or presence of known cADPR antagonists. As shown in Table 1, both ryanodine (50 μm) and 8-Br-cADPR (100 μm) reduced the [Ca2+]i to levels only slightly higher than those measured in control, untreated cells, indicating a causal role of intracellular cADPR in the lipopolysaccharide-induced [Ca2+]i increase (p < 0.01).

Table 1.   Effect of cADPR antagonists on the [Ca2+]i increase induced by lipopolysaccharide in N9 cells
Addition[Ca2+]i nm
  • Results are the mean ± SD of five different experiments.

  • *

    p < 0.005, compared to control.

  • **

    p < 0.01, compared to lipopolysaccharide-treated.

None84 ± 11
Lipopolysaccharide 24 h171 ± 23*
8-Br-cADPR + lipopolysaccharide 24 h 106 ± 4**
Ryanodine + lipopolysaccharide 24 h95 ± 5**

Role of intracellular cADPR in the lipopolysaccharide-induced nitric oxide production by N9 cells

The lipopolysaccharide-induced increase of the [cADPR]i and of the [Ca2+]i levels in N9 cells prompted us to investigate whether cADPR by itself could stimulate functional effects of microglial cells known to be induced by lipopolysaccharide, downstream of the basal [Ca2+]i elevation. These include overproduction of nitric oxide and TNF-α and their release into the supernatants (Chang and Liu 1999; Hoffmann et al. 2003).

Extracellular cADPR can enter several cell types across multiple nucleoside transporters, both equilibrative and concentrative (ENT and CNT, respectively) (Guida et al. 2002; Guida et al. 2004; Podestàet al. 2005). Nitric oxide production and release from N9 cells was measured by assaying the accumulation of the end-products of nitric oxide reduction, nitrites and nitrates, in the culture medium (Lee et al. 2004). As shown in Fig. 4a, extracellular cADPR (100 μm) induced a time-dependent increase of the nitric oxide concentration in the cell supernatants as compared to untreated cultures. The putative role of cADPR in eliciting the functional effects observed in lipopolysaccharide-treated N9 microglial cells was further investigated by assessing whether known cADPR antagonists could reduce nitric oxide generation in lipopolysaccharide-stimulated N9 cells. Figure 4b shows that, indeed, the increase of nitric oxide production triggered by lipopolysaccharide over the values recorded in unstimulated N9 cells was significantly reduced by 100 μm 8-Br-cADPR (p < 0.005), 50 μm ryanodine (p < 0.01), and also by the ADP-ribosyl cyclase inhibitor nicotinamide at 10 mm (p < 0.01).

image

Figure 4.  Role of cADPR on nitric oxide release from N9 cells. a, Effect of extracellularly added cADPR on nitric oxide release from N9 cells. N9 cells (0.25 × 106 per sample) were incubated with 100 µm cADPR for the times indicated. Culture supernatants were recovered, centrifuged and the nitrite concentration was assayed as described under Materials and Methods. Results are expressed as NO2 relative to untreated, control cultures (2.49 ± 0.138 nm), and are the mean ± SD of five different experiments. b, Effect of cADPR antagonists on lipopolysaccharide-induced nitric oxide release from N9 cells after 24 h incubation. For experimental details see Materials and methods. Results are expressed as NO2 production relative to control, untreated cultures and are the mean ± SD of seven different experiments.

Download figure to PowerPoint

The above results demonstrate the involvement of cADPR in the widely reported lipopolysaccharide-induced stimulation of nitric oxide production by N9 cells. This is due to de novo expression of iNOS in the cytosol, following treatment of the cells with lipopolysaccharide (Fig. 5a). We next explored a possible role of cADPR in the lipopolysaccharide-induced, Ca2+-dependent enhancement of iNOS expression (Possel et al. 2000). N9 cells were stimulated with lipopolysaccharide following pretreatment with 100 μm 8-Br-cADPR or 50 μm ryanodine, and iNOS expression was then determined on cell lysates by western blot analysis. As shown in Fig. 5b, iNOS expression was higher in lipopolysaccharide-treated cells than in unstimulated cells, in line with the immunocytochemical patterns and also with the comparably higher nitric oxide production shown in Fig. 4 (p < 0.01). The presence of the cADPR antagonists 8-Br-cADPR or ryanodine markedly reduced iNOS expression, again in agreement with the decreased nitric oxide release observed with both compounds (Fig. 4; p < 0.05). These results demonstrate that cADPR is causally involved in the lipopolysaccharide-induced, Ca2+-dependent activation of iNOS expression leading to nitric oxide overproduction in N9 microglial cells.

image

Figure 5.  Immunocytochemistry and immunochemical detection of iNOS in control and lipopolysaccharide-treated N9 cells. a, Adherent N9 cells (2 × 105) were incubated with lipopolysaccharide (200 ng/mL) for 18 h. At the end of incubation, cells were fixed, permeabilised and stained for iNOS positivity using a goat polyclonal antibody followed by an Alexa-633 labelled secondary antibody. Images were obtained with a Leica TCS-SL confocal microscope and represent average projections of Z-stack files of 50 sections elaborated using Leica LCS software. a, control, untreated cells; b, lipopolysaccharide-treated cells. b, For experimental details see Materials and methods. Lane i: control, untreated cells. Lane ii: cells incubated with lipopolysaccharide. Lane iii: cells pre-incubated with 8-Br-cADPR and then lipopolysaccharide-treated. Lane iv: cells pre-incubated with ryanodine and then lipopolysaccharide-treated. A representative experiment is shown of four yielding comparable results. The graph displayed below the western blots presents the bioluminescence intensity quantified using the Chemidoc System (see Materials and methods): values are expressed as optical density/mm2 and are the mean ± SD of the four experiments.

Download figure to PowerPoint

Role of intracellular cADPR in TNF-α release from lipopolysaccharide-activated N9 cells

Along with nitric oxide, another cytokine responsible for the pro-inflammatory and anti-tumor activity of microglial cells is TNF-α. Activation of N9 cells by lipopolysaccharide also stimulates TNF-α release in the culture medium (Paris et al. 2000; Hoffmann et al. 2003). We first investigated the effect of exogenously added cADPR on TNF-α release. Figure 6a shows that incubation of N9 cells with 100 μm cADPR for 24 h induced an increase of TNF-α production three-fold that measured in control cells. Next, to assess the role of the endogenous cADPR overproduction induced by lipopolysaccharide in TNF-α release, N9 cells were pre-incubated in the presence of each of the cADPR antagonists or inhibitors down-regulating the lipopolysaccharide-induced stimulation of GDP-ribosyl cyclase and CD38 phosphorylation. After pre-incubation for 1 h, N9 cells were treated for 24 h with lipopolysaccharide (200 ng/mL). The results shown in Fig. 6bindicate partial inhibition of TNF-α release, measured with the EIA kit determined with each of the tested compounds (p < 0.01), whereas 50 nm staurosporine afforded near complete inhibition (p < 0.01). Superimposable results were obtained with the cytotoxicity-based assay (with the exception of experiments with ryanodine, see Materials and methods). Thus, [cADPR]i is causally involved in the lipopolysaccharide-induced stimulation of TNF-α release from N9 microglial cells.

image

Figure 6.  Role of cADPR on TNF-α release from N9 cells. a, Effect of extracellularly added cADPR on TNF-α release from N9 cells. N9 cells (0.15 × 106) were incubated with 100 µm cADPR. At the times indicated, culture supernatants were recovered, centrifuged and the TNF-α concentration was determined by the EIA kit as described under Materials and methods. Results are expressed as [TNF-α]e relative to untreated, control cultures (176 ± 16 pmol/mL) and are the mean ± SD of four different experiments. b, Involvement of cADPR in the lipopolysaccharide-induced TNF-α release from N9 cells. N9 cells (0.15 × 106) were incubated for 1 h in the absence (control and lipopolysaccharide) or presence of each of the compounds indicated on the abscissa, as reported in Materials and methods. Thereafter, 200 ng/mL lipopolysaccharide were added to all cultures (except for the control ones). After 24 h, culture supernatants were recovered, centrifuged and the TNF-α concentration was determined using the EIA kit. Results are expressed as the mean ± SD of four different experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Inflammatory events in the CNS are associated with infections and injuries, as well as with chronic degenerative diseases, such as multiple sclerosis, Parkinson's or Alzheimer's (Nakamura et al. 1999). For this reason, it is of great importance to obtain insights into the molecular mechanisms that underlie immune reactions in brain tissue. Given the particularly daunting technical difficulties of the in vivo study of the CNS, this task is best tackled by the in vitro study of the activation pathways of the major cell type responsible for brain inflammation – the microglia.

The results obtained in this study on the microglial murine cell line N9 show that cell activation by lipopolysaccharide, leading to nitric oxide and TNF-α release, occurs through the intracellular production of cADPR, which behaves as a second messenger in this signalling cascade. Lipopolysaccharide induces a doubling of the GDP-ribosyl cyclase activity of CD38 (Fig. 1) and a 70% increase of the intracellular concentration of cADPR (see Results). Lipopolysaccharide-induced cyclase activation occurs via protein kinase-mediated phosphorylation of CD38, as demonstrated by prevention of this activation by broad (staurosporine) and specific (I-PKA, I-PKC and AG126, Fig. 3) protein kinase inhibitors. In microglial cells, lipopolysaccharide is indeed known to bind to a TLR-4 and to induce activation of PKC (Kim et al. 2005) and a protein tyrosine kinase sensitive to AG126 (Kann et al. 2004). PKA-mediated activation of ADP-ribosyl cyclases has been also reported in several other cell types (Morita et al. 1997; Zocchi et al. 2001; Boittin et al. 2003; Xie et al. 2005; Bruzzone et al. 2006). The cytosolic amino-terminal region of murine CD38 harbours three serine residues, with Ser19 scoring the highest phosphorylation potential (0.992), and one tyrosine at position 4 scoring 0.55 (NetPhos 2.0 Server, Center for Biological Sequence Analysis, CBS, Technical University of Denmark, Copenhagen, Denmark). Our data strongly implicate Ser19 and Tyr4 as being involved in the lipopolysaccharide-triggered double phosphorylation of CD38 (by PKA + PKC and by the AG126-inhibitable protein tyrosine kinase, respectively) that results in stimulation of CD38 ADPRC activity and hence in overproduction of cADPR. The fact that several protein kinases are involved in the lipopolysaccharide-induced CD38 activation may be part of a redundant mechanism for ADP-ribosyl cyclase activation in microglia, possibly resulting in the fine tuning of the cyclase activity and of the [cADPR]i in response to extracellular stimuli.

It has already been reported that incubation with lipopolysaccharide induces an increase of the basal [Ca2+]i in primary cultures of microglial cells (Hoffmann et al. 2003). Here, we demonstrate that the elevation of the [cADPR]i in N9 cells, due to lipopolysaccharide-induced CD38 activation, is responsible for the sustained [Ca2+]i elevation. Indeed, pretreatment of N9 cells with either of the cADPR antagonists 8-Br-cADPR or ryanodine greatly down-regulated the lipopolysaccharide-induced [Ca2+]i increase (Table 1).

Elevation of the [Ca2+]i following lipopolysaccharide exposure is in turn fundamental to promote the microglial activation process, leading to induction of the pro-inflammatory activity typical of these cells, which includes nitric oxide production and TNF-α release (Hoffmann et al. 2003). Specifically, nitric oxide generation induced by lipopolysaccharide in microglial cultures was demonstrated to be significantly reduced in a concentration-dependent manner by the intracellular Ca2+ chelator BAPTA-AM (Hoffmann et al. 2003). Results obtained in the present study also demonstrate a causal role of cADPR in nitric oxide and TNF-α release from lipopolysaccharide-stimulated N9 cells. Exogenously added cADPR, which has been shown to be transported through the plasma membrane of different cell types by the nucleoside transporters ENT and CNT (Guida et al. 2002, 2004; Podestàet al. 2005), induced progressive release of both nitric oxide (Fig. 4a) and TNF-α (Fig. 6a). Pretreatment of N9 cells with either of the two cADPR antagonists 8-Br-cADPR or ryanodine or with the cyclase inhibitor nicotinamide resulted in the partial or total abrogation of the lipopolysaccharide-induced release of nitric oxide (Fig. 4b). A Ca2+-dependent, inducible nitric oxide synthase (iNOS) is known to be responsible for nitric oxide production in N9 microglial cells (Chang and Liu 1999; Possel et al. 2000). The results obtained in our study indicate that iNOS expression is up-regulated by the [cADPR]i (Figs 4b and 5b).

The signalling cascade that links lipopolysaccharide stimulation to enhanced TNF-α release also seems to involve cADPR, similar to what was observed for nitric oxide overproduction (Fig. 6b). This view is supported by the significant inhibition of TNF-α release elicited by the cADPR antagonist 8-Br-cADPR, the cyclase inhibitor nicotinamide and ryanodine, as well as by the various protein kinase inhibitors, with staurosporine being the most effective.

In conclusion, the present results indicate that lipopolysaccharide-induced stimulation of N9 microglial cells is dependent on CD38 phosphorylation, ADP-ribosyl cyclase activation and the consequent increase of the [cADPR]i, resulting in a sustained [Ca2+]i elevation and enhanced nitric oxide and TNF-α release. This signalling pathway has three main regulatory modules. The first one is related to activation of CD38, which takes place via cyclase phosphorylation involving several protein kinases (PKA, PKC and protein tyrosine kinase). The second regulatory module downstream of the cADPR overproduction is the increase of the [Ca2+]i. The substantial inhibition of the [Ca2+]i increase afforded by 8-Br-cADPR and ryanodine (Table 1) argues for a major role of cADPR in enhancing the [Ca2+]i levels in lipopolysaccharide-stimulated N9 cells. Failure of 8-Br-cADPR to completely abrogate this increase could imply a limited role for Ca2+-regulating signal molecules other than cADPR. Pointedly, ADPR has been recently shown to induce Ca2+ influx in lipopolysaccharide-stimulated microglial cells (Kraft et al. 2004). Indeed, cADPR seems to be a potent coregulator of Ca2+ entry across ADPR-gated TRPM2 channels (Kolisek et al. 2005; Gasser et al. 2006).

The third and last regulatory module in the lipopolysaccharide-induced activation of N9 cells ultimately leads to enhanced nitric oxide and TNF-α release and is dependent on the [Ca2+]i increase, although the molecular mechanisms leading to cytokine release are probably manifold. Indeed, 8-Br-cADPR, nicotinamide and ryanodine fail to afford comparable extents of inhibition on nitric oxide and TNF-α release: 50 μm ryanodine quenched nitric oxide release (Fig. 4B) and iNOS expression (Fig. 5B) almost completely, while inhibiting TNF-α release only by 30% (Fig. 6B). In addition, there are quantitatively poor correlations between the fractional extents of inhibition of CD38 double phosphorylation and those elicited by protein kinase inhibitors on TNF-α release, respectively (see Figs 3 and 6 for comparison). A relevant example is the sharp inhibiting effect on CD38 serine phosphorylation and a low inhibition of TNF-α release by the PKC inhibitor. These findings are consistent with the involvement of other molecular mechanisms or systems, which in fact have been implicated in lipopolysaccharide-induced cytokine release from glial cells: these include the (MAPK)/arachidonate/cyclooxygenase 2 cascade (Paris et al. 2000), the p38 MAPK subfamily, the AG126-inhibitable extracellular signal-regulated protein kinase (ERK) (Bhat et al. 1998; Hanisch et al. 2001; Kann et al. 2004) and nuclear factor-kB (Possel et al. 2000; Quin et al. 2005).

Demonstration of the role of cADPR as a second messenger responsible for the functional activation of microglial cells may suggest a new target for therapeutic strategies. cADPR antagonists (Guse 2005) hold promise for the prevention or reduction of microglial cell activation whenever the excessive pro-inflammatory effects induced by the prolonged activation of this cell population concur with brain tissue damage (Gebicke-Haerter et al. 1996; Tan et al. 1999).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported in part by grants from AIRC (Associazione Italiana per la Ricerca nel Cancro), the Italian Ministry of Education, University and Scientific Research (MIUR-PRIN 2003, MIUR-FIRB RBAU019A3C, MIUR-FIRB RBNE01ERXR, MIUR RBLA039LSF-002), the University of Genova and Fondazione Cassa di Risparmio di Genova e Imperia. We are indebted to Dr Paola Contini (Department of Internal Medicine, and Center of Excellence for Medical Research, University of Genova) for performing the FACS analyses.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Aschner M., Allen J. W., Kimelberg H. K., LoPachin R. M. and Streit W. J. (1999) Glial cells in neurotoxicity development. Annu. Rev. Pharmacol. Toxicol. 39, 151173.
  • Bhat N. R., Zhang P., Lee J. C. and Hogan E. L. (1998) Extracellular signal-regulated kinase and p38 subgroups of mitogen-activated protein kinases regulate inducible nitric oxide synthase and tumor necrosis factor-alpha gene expression in endotoxin-stimulated primary glial cultures. J. Neurosci. 18, 16331641.
  • Boittin F. X., Dipp M., Kinnear N. P., Galione A. and Evans A. M. (2003) Vasodilation by the calcium-mobilizing messenger cyclic ADP-ribose. J. Biol. Chem. 278, 96029608.
  • Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248252.
  • Bruzzone S., De Flora A., Usai C., Graeff R. and Lee H. C. (2003) Cyclic ADP-ribose is a second messenger in the lipopolysaccharide-stimulated proliferation of human peripheral blood mononuclear cells. Biochem. J. 375, 395403.
  • Bruzzone S., Moreschi I., Guida L., Usai C., Zocchi E. and De Flora A. (2006) Extracellular NAD+ regulates intracellular calcium levels and induces activation of human granulocytes. Biochem. J. 393, 697704.
  • Chang J. Y. and Liu L. Z. (1999) Manganese potentiates nitric oxide production by microglia. Brain Res. Mol. Brain Res. 68, 2228.
  • Farber K. and Kettenmann H. (2005) Physiology of microglial cells. Brain Res. Brain Res. Rev. 48, 133143.
  • Franco L., Bruzzone S., Song P., Guida L., Zocchi E., Walseth T. F., Crimi E., Usai C., De Flora A. and Brusasco V. (2001) Extracellular cyclic ADP-ribose potentiates ACh-induced contraction in bovine tracheal smooth muscle. Am. J. Physiol. Lung Cell. Mol. Physiol. 280, L98L106.
  • Gasser A., Glassmeier G., Fliegert R., et al. (2006) Activation of T cell calcium influx by the second messenger ADP-ribose. J. Biol. Chem. 281, 24892496.
  • Gebicke-Haerter P. J. (2001) Microglia in neurodegeneration: molecular aspects. Microsc. Res. Techn. 54, 4758.
  • Gebicke-Haerter P. J., Van Calker D., Norenberg W. and Illes P. (1996) Molecular mechanisms of microglial activation. A. Implications for regeneration and neurodegenerative diseases. Neurochem. Int. 29, 112.
  • Gonzalez-Scarano F. and Baltuch G. (1999) Microglia as mediators of inflammatory and degenerative diseases. Annu. Rev. Neurosci. 22, 219240.
  • Graeff R. M., Walseth T. F., Fryxell K., Branton W. D. and Lee H. C. (1994) Enzymatic synthesis and characterizations of cyclic GDP-ribose. A procedure for distinguishing enzymes with ADP-ribosyl cyclase activity. J. Biol. Chem. 269, 30 26030 267.
  • Guida L., Bruzzone S., Sturla L., Franco L., Zocchi E. and De Flora A. (2002) Equilibrative and concentrative nucleoside transporters mediate influx of extracellular cyclic ADP-ribose into 3T3 murine fibroblasts. J. Biol. Chem. 277, 47 09747 105.
  • Guida L., Franco L., Bruzzone S., Sturla L., Zocchi E., Basile G., Usai C. and De Flora A. (2004) Concentrative influx of functionally active cyclic ADP-ribose in dimethyl sulfoxide-differentiated HL-60 cells. J. Biol. Chem. 279, 22 06622 075.
  • Guse A. H. (2005) Second messenger function and the structure-activity relationship of cyclic adenosine diphosphoribose (cADPR). FEBS J. 272, 45904597.
  • Hanisch U.K., Prinz M., Angstwurm K., Hausler K. G., Kann O., Kettenmann H. and Weber J. R. (2001) The protein tyrosine kinase inhibitor AG126 prevents the massive microglial cytokine induction by pneumococcal cell walls. Eur. J. Immunol. 31, 21042115.
  • Harada N., Santos-Argumedo L., Chang R., et al. (1993) Expression cloning of a cDNA encoding a novel murine B cell activation marker. Homology to human CD38. J. Immunol. 151, 31113118.
  • Hoffmann A., Kann O., Ohlemeyer C., Hanisch U.K. and Kettenmann H. (2003) Elevation of basal intracellular calcium as a central element in the activation of brain macrophages (microglia): suppression of receptor-evoked calcium signaling and control of release function. J. Neurosci. 23, 44104419.
  • Ignarro L. J., Buga G. M., Wood K. S., Byrns R. E. and Chaudhuri G. (1987) Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. Natl. Acad. Sci. U S A 84, 92659269.
  • Kann O., Hoffmann A., Schumann R. R., Weber J. R., Kettenmann H. and Hanisch U. K. (2004) The tyrosine kinase inhibitor AG126 restores receptor signaling and blocks release functions in activated microglia (brain macrophages) by preventing a chronic rise in the intracellular calcium level. J. Neurochem. 90, 513525.
  • Kim D. C., Kim S. H., Jeong M. W., Baek N. I. and Kim K. T. (2005) Effect of rottlerin, a PKC-delta inhibitor, on TLR-4-dependent activation of murine microglia. Biochem. Biophys. Res. Commun. 337, 110115.
  • Kolisek M., Beck A., Fleig A. and Penner R. (2005) Cyclic ADP-ribose and hydrogen peroxide synergize with ADP-ribose in the activation of TRPM2 channels. Mol. Cell 18, 6169.
  • Kraft R., Grimm C., Grosse K., Hoffmann A., Sauerbruch S., Kettenmann H., Schultz G. and Harteneck C. (2004) Hydrogen peroxide and ADP-ribose induce TRPM2-mediated calcium influx and cation currents in microglia. Am. J. Physiol. Cell Physiol. 286, C129C137.
  • Lee H. C., ed. (2002) Cyclic ADP-ribose and NAADP: Structures, Metabolism and Functions, Kluwer, Dordrecht.
  • Lee H., Jeong J., Son E., et al. (2004) Ethanol selectively modulates inflammatory activation signaling of brain microglia. J. Neuroimmunol. 156, 8895.
  • Lee H. C., Walseth T. F., Bratt G. T., Hayes R. N. and Clapper D. L. (1989) Structural determination of a cyclic metabolite of NAD+ with intracellular Ca2+-mobilizing activity. J. Biol. Chem. 264, 16081615.
  • Mehta K. and Malavasi F. (2000) Human CD38 and related molecules. Chem. Immunol. 75, 256273.
  • Morita K., Kitayama S. and Dohi T. (1997) Stimulation of cyclic ADP-ribose synthesis by acetylcholine and its role in catecholamine release in bovine adrenal chromaffin cells. J. Biol. Chem. 272, 21 00221 009.
  • Mossman T. (1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxic assays. J. Immunol. Meth 65, 5563.
  • Nakamura Y., Si Q. S. and Kataoka K. (1999) Lipopolysaccharide-induced microglial activation in culture: temporal profiles of morphological change and release of cytokines and nitric oxide. Neurosci. Res. 35, 95100.
  • Paris D., Town T. and Mullan M. (2000) Novel strategies for opposing murine microglial activation. Neurosci. Lett. 278, 58.
  • Podestà M., Benvenuto F., Pitto A., et al. (2005) Concentrative uptake of cyclic ADP-ribose generated by BST-1+ stroma stimulates proliferation of human hematopoietic progenitors. J. Biol. Chem. 280, 53435349.
  • Possel H., Noack H., Putzke J., Wolf G. and Sies H. (2000) Selective upregulation of inducible nitric oxide synthase (iNOS) by lipopolysaccharide (LPS) and cytokines in microglia: in vitro and in vivo studies. Glia 32, 5159.
  • Quin H., Wilson C. A., Lee S. J., Zhao K. and Benveniste E. N. (2005) LPS induces CD40 gene expression through the activation of NF-kappaB and STAT-1-alpha in macrophages and microglia. Blood 106, 31143122.
  • Raetz C. R. and Whitfield C. (2001) Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71, 635700.
  • Righi M., Mori L., De Libero G., Sironi M., Biondi A., Mantovani A., Donini S. D. and Ricciardi-Castagnoli P. (1989) Monokine production by microglial cell clones. Eur. J. Immunol. 19, 14431448.
  • Schuber F. and Lund F. E. (2004) Structures and enzymology of ADP-ribosyl cyclases: conserved enzymes that produce multiple calcium mobilizing metabolites. Curr. Mol. Med. 4, 249261.
  • Tan J., Town T., Paris D., Placzek A., Parker T., Crawford F, Yu H., Humphrey J. and Mullan M. (1999) Activation of microglial cells by the CD40 pathway: relevance to multiple sclerosis. J. Neuroimmunol. 97, 7785.
  • Town T., Nikolic V. and Tan J. (2005) The microglial ‘activation’ continuum: from innate to adaptive responses. J. Neuroinflammation 2, 24.
  • Vercammen D., Vandenabeele P., Beyaert R., Declercq W. and Fiers W. (1997) Tumour necrosis factor-induced necrosis versus anti-Fas-induced apoptosis in L929 cells. Cytokine 9, 801808.
  • Verderio C., Bruzzone S., Zocchi E., Fedele E., Schenk U., De Flora A. and Matteoli M. (2001) Evidence of a role for cyclic ADP-ribose in calcium signalling and neurotransmitter release in cultured astrocytes. J. Neurochem. 78, 646657.
  • Xie G. H., Rah S. Y., Kim S. J., Nam T. S., Ha K. C., Chae S. W., Im M. J. and Kim U. H. (2005) ADP-ribosyl cyclase couples to cyclic AMP signaling in the cardiomyocytes. Biochem. Biophys. Res. Commun. 330, 12901298.
  • Zocchi E., Franco L., Guida L., Benatti U., Bargellesi A., Malavasi F., Lee H. C. and De Flora A. (1993) A single protein immunologically identified as CD38 displays NAD+ glycohydrolase, ADP-ribosyl cyclase and cyclic ADP-ribose hydrolase activities at the outer surface of human erythrocytes. Biochem. Biophys. Res. Commun. 196, 14591465.
  • Zocchi E., Daga A., Usai C., Franco L., Guida L., Bruzzone S., Costa A., Marchetti C. and De Flora A. (1998) Expression of CD38 increases intracellular calcium concentration and reduces doubling time in HeLa and 3T3 cells. J. Biol. Chem. 273, 80178024.
  • Zocchi E., Carpaneto A., Cerrano C., Bavestrello G., Giovine M., Bruzzone S., Guida L., Franco L. and Usai C. (2001) The temperature-signaling cascade in sponges involves a heat-gated cation channel, abscisic acid, and cyclic ADP-ribose. Proc. Natl. Acad. Sci. USA 98, 14 85914 864.