CELL-SPECIFIC EFFECTS OF PENTOXIFYLLINE ON NITRIC OXIDE PRODUCTION AND INDUCIBLE NITRIC OXIDE SYNTHASE mRNA EXPRESSION

SUMMARY Cytokine-stimulated astrocytes and macrophages are potent producers of nitric oxide (NO), a free radical proposed to play an important role in organ-specific autoimmunity, including demyelinating diseases of the central nervous system. The aim of this study was to investigate effects of pentoxifylline (PTX), a phosphodiesterase inhibitor with immunomodulatory properties, on NO production and inducible NO synthase (iNOS) mRNA levels in rat astrocytes and macrophages. We have shown that PTX affects cytokine (IFN-g, IL-1, TNF-a)-induced NO production in both cell types, but in the opposite manner - enhancing in astrocytes and suppressive in macrophages. While PTX did not have any effect on enzymatic activity of iNOS in activated cells, expression of iNOS mRNA was elevated in astrocytes and decreased in macrophages treated with cytokines and PTX. Treatment with PTX alone did not affect neither NO production nor iNOS mRNA levels in astrocytes or macrophages. This study indicates involvement of different signaling pathways associated with iNOS induction in astrocytes and macrophages, thus emphasizing complexity of regulation of NO synthesis in different cell types. INTRODUCTION Nitric oxide (NO), a gaseous, short-lived free radical is an important signaling molecule with pleiotropic activities including vasodilatation, neurotransmission, microbial and tumor cell killing1 and target-tissue damage in organ-specific autoimmune disorders.2 NO has also been postulated to act in regulation of immune response, suppressing proliferation of T cells3 and Th1 clones4, as well as enhancing Th2-cytokine production.5 Excessive production of NO by inducible NO synthase (iNOS)-mediated oxidation of L-arginine in cytokine-activated astrocytes, microglia and infiltrating macrophages is associated with demyelinating diseases such as multiple sclerosis6 and its animal model experimental allergic encephalomyelitis (EAE).7 Interference with NO production in EAE by using iNOS inhibitors showed conflicting results - atenuation8 or agravation9 of disease - suggesting both effectory (destruction of myelin sheet) and regulatory (suppression of T-cell proliferation and Th1 response) role of NO in EAE. Phosphodiesterase inhibitor pentoxifylline is a vasodilatatory drug with a variety of immunosuppressive properties including inhibition of TNF-a,10-13 IL-1b, IL-6, IL-8,11 IFN-g13 and IL-212,13 release, suppression of T-cell proliferation,12 ICAM expression14 and attenuation of EAE12 and experimental autoimmune neuritis.15 PTX is also proposed as a supportive drug in multiple sclerosis.16 While recently PTX was found to inhibit NO production in murine macrophages,17 its effect on NO production in astrocytes was not investigated so far. In this study we wanted to examine whether PTX affects NO synthesis in rat astrocytes in the same way as it does in macrophages. MATERIALS AND METHODS Reagents Recombinant rat IFN-g was obtained from Holland Biotechnology (Leiden, Netherlands). Rat recombinant IL-1b and TNF-a were from Genzyme (Cambridge, MA). Fetal calf serum (FCS), RPMI 1640 and phosphate buffered saline (PBS) were from Flow Laboratories (Irvine, Scotland). Cycloheximide (CHX) was obtained from US Biochemical Corporation (Cleveland, OH). Pentoxifylline (PTX) was from Hemofarm (Vrsac, Yugoslavia). Sulfanilamide and naphtylethylenediamine dichydrochloride were from Sigma (St Louis, CA). Moloney leukemia virus reverse transcriptase was obtained from Eurogentec (Seraing, Belgium). Random primers were from Pharmacia (Uppsala, Sweden). Isolation of astrocytes and macrophages Astrocytes were isolated from mixed glial cell cultures prepared from brains of newborn DA rats as previously was described.18 Cells were maintained in HEPES-buffered RPMI 1640 medium supplemented with 5% FCS, 2 mM glutamine, antibiotics, sodium pyruvate and 6 g/l glucose in 75 cm2 tissue culture flasks (Flow). After ten days of cultivation at 37C in humidified atmosphere with 5% CO2, the cultures were mechanically shaken to dislodge microglia and oligodendrocytes and astrocytes were further purified by repetition of trypsinization (0.25% trypsin and 0.02% EDTA) and replating. The cells used in the experiments described here were obtained after third to fourth passage when they were routinely more than 97% positive for glial fibrilary acidic protein, an astrocyte-specific intermediate filament component. Resident macrophages were obtained from female 10-18-weeks-old DA rats by peritoneal lavage with PBS containing 20 IU/ml heparin. Following adherence to FCS coated plastic dishes for 1 h on 37C, nonadherent cells were removed by washing with cold PBS. Adherent cells were harvested by treatment with cold 0.02 % EDTA in PBS, washed and resuspended in RPMI 1640 medium prepared as for astrocytes. Cell cultures A total of 2 x 105 astrocytes or 105 macrophages was seeded in each well of the 24-well plate (Flow) along with different combinations of IFN-g (100 U/ml), IL-1 (10 ng/ml) and TNF-a (200 U/ml) (astrocytes) or with IFN-g and IL-1 (macrophages) in the presence or absence of PTX (200 mg/ml). Each culture was performed in triplicate wells, except for RNA isolation, in final volume of 1 ml. Cells were incubated for 6 h (RNA isolation) or 48 h (nitrite measurement) at 37C in humidified 5% CO2 atmosphere. To asses influence of PTX on iNOS activity, after 24 h of cultivation with IFN-g and IL-1, cells were washed and incubated for further 24 h in 1 ml of fresh medium containing CHX (10 mg/ml) or CHX and PTX (200 mg/ml). Cell viability was determined by trypan blue exclusion. Nitrite measurement Nitrite accumulation, an indicator of NO production, was measured using the Griess reagent.19 Briefly, 50 ml aliquots of culture supernatants were mixed with an equal volume of Griess reagent - mixture at 1:1 of 0.1 % naphtylethylenediamine dihydrochloride and 1 % sulphanilamide in 5 % H3PO4 - and incubated at room temperature for 10 min. The absorbance at 570 nm was measured in an automated Titertek multiplate reader (Flow). Nitrite concentration was calculated from a NaNO2 standard curve. Results are presented as mean + SEM of triplicate observations. Determination of iNOS mRNA by RT-PCR Total RNA was isolated by acid guanidinium thiocyanate-phenol-chloroform extraction20 after 6 h of cultivation, when iNOS message peaked.4 RNA was reverse transcribed, using Moloney leukemia virus reverse transcriptase and random primers. PCR amplification of cDNA with primers specific for iNOS and GAPDH was carried out in the same tube and the products visualized by electrophoresis through 2.5% agarose gel containing ethidium bromide. Number of cycles (for macrophages-35 for both iNOS and GAPDH and for astrocytes-40 and 30 for iNOS and GAPDH, respectively) ensuring nonsaturating PCR conditions was established in preliminary experiments. For the iNOS, primers were: sense, 5’-AGAGAGATCCGGTTCACA-3’; antisense, 5’-CACAGAACTGAGGGTACA-3’ and the PCR product was 377 bp. The primers for GAPDH used to show equal loading of RNA were: sense, 5’-GAGGGTGGGGCCAAAAG-3’; antisense, 5’-GGATGCAGGGATGATGTTCT-3’, and the PCR was 295 bp. Statistical analysis Results were analyzed using the Student’s t-test. A p value less than 0.05 was considered significant. RESULTS PTX enhances cytokine-induced NO production in rat astrocytes Proinflamatory cytokines IFN-g, IL-1 and TNF-a have been shown to induce NO production in rodent astrocytes.21,22 We observed a similar up-regulation of NO production, measured as a rise of nitrite concentrations in culture supernatants, in rat astrocytes cultivated for 48 h with different combinations of IFN-g (100 U/ml), IL-1 (10 ng/ml) and TNF-a (200 U/ml) (Fig. 1). Protein synthesis inhibitor CHX (10 mg/ml) and NOS inhibitor nitro-L-arginine methyl ester (3 mM) both prevented cytokine-induced nitrite accumulation (not shown), indicating that NO in astrocytes was generated via iNOS-dependent L-arginine-NO pathway.23 In the presence of PTX (200 mg/ml), NO production in cytokine-stimulated astrocytes was increased aproximately by 60-100% (Fig. 1). Non-stimulated astrocytes did not produce detectable levels of nitrites and PTX by itself did not have any stimulatory effect on NO synthesis in astrocytes (Fig. 1). PTX inhibits IFN-g + IL-1-induced NO production in rat peritoneal macrophages Although stimulation with IFN-g was sufficient to induce maximal NO production in rat peritoneal macrophages (not shown), we used combination of IFN-g (100 U/ml) and IL-1 (10 ng/ml) in order to compare data from astrocyte and macrophage cultures. Contrary to results obtained with astrocytes, PTX (200 mg/ml) treatment reduced NO production in IFN-g + IL-1-stimulated macrophages (Fig. 2), not affecting their viability. A relatively high spontaneous production of NO by rat macrophages (Fig. 2) was reported also by others24 and remained unchanged by PTX treatment (Fig. 2). Suppressive effect of PTX on IFN-g + IL-1-induced NO production in rat macrophages, observed in our experiments, is consistent with findings of other investigators describing PTX-mediated inhibition of NO production in murine macrophages treated with LPS.17 PTX does not affect activity of iNOS in astrocytes or macrophages Considering PTX-mediated enhancement of NO synthesis in astrocytes, we wanted to examine whether PTX can modulate catalytic properties of astrocyte iNOS. Therefore, PTX (200 mg/ml) and CHX (10 mg/ml) were added to cells pretreated with IFN-g (100 U/ml) and IL-1 (10 ng/ml) for 24 h, and nitrite production was measured after a further 24 h. The finding that in cells in which iNOS had already been induced by cytokines and any further induction blocked by CHX, PTX did not change nitrite production (Fig. 3a), strongly argues against ability of PTX to modulate iNOS catalytic activity in astrocytes. Similar results, using the same cultivating conditions were obtained with macrophages as well (Fig. 3b). These data suggested that in astrocytes, as well as in macrophages, PTX affected NO production through modulation of iNOS induction rather than its activity. Expression of iNOS mRNA is enhanced by PTX in astrocytes and reduced in macrophages In order to confirm that PTX-mediated modulation of NO production is a pretranslational event, we investigated expression of iNOS mRNA in astrocytes and macrophages treated with cytokines and/or PTX. After 6 h of cultivation with IFN-g (100 U/ml) and IL-1 (10 ng/ml) in the presence or absence of PTX (200 mg/ml), RNA was isolated from astrocytes and macrophages and RT-PCR for iNOS mRNA was carried out. Again, as with NO production, results concerning PTX treatment of astrocytes were in contrast to those obtained from macrophage cultures: while elevating the expression of iNOS mRNA in activated astrocytes (Fig. 4a), PTX reduced it in macrophages (Fig. 4b). In both astrocytes and macrophages, the expression of iNOS mRNA was unaffected by treatment with PTX alone (Fig. 4). This finding that the observed effects of PTX on NO production in astrocytes and macrophages are reflection of changes affecting iNOS mRNA levels suggests transcriptional regulation of cytokine-induced iNOS activation by PTX. DISCUSSION In this study we showed that phosphodiesterase inhibitor PTX has opposite effects on NO production in different rat cell types - inhibitory in macrophages and enhancing in astrocytes. While suppressive action of PTX on NO production was previously described in LPS-stimulated murine macrophages,17 and confirmed here in rat, using IFN-g + IL-1, a potentiation of cytokine (IFN-g, IL-1, TNF-a)-induced NO production in astrocytes by PTX, to our knowledge, is the first report of this kind. In addition, we showed here that the observed effects of PTX on NO production in astrocytes and macrophages were not due to the changes of iNOS enzymatic activity, but to the alterations of iNOS mRNA expression which was up-regulated in astrocytes and down-regulated in macrophages. The finding that PTX did not change spontaneous NO production also argues against ability of PTX to affect already induced iNOS. Therefore, PTX most probably exerts its influence on NO production through modulation of cytokine-triggered iNOS gene transcription, although changes in iNOS mRNA stability can not be excluded. Recent studies have provided evidence for different role of cAMP in the expression of macrophage and astrocyte iNOS. The increase of intracellular cAMP levels enhanced NO production in activated murine and rat macrophages,25-27 but reduced it in rat astrocytes,27 probably by differently affecting activation of iNOS transcription factor NF-kb.27,28 These and our study, describing differential regulation of iNOS activation in two different cell types by the same intracellular messenger (cAMP) or pharmacological agent (PTX) suggest existence of different signaling pathways in the induction of iNOS. Moreover, it seems that cAMP and PTX are involved in different intracellular events related to iNOS activation. Although PTX increases intracellular cAMP10 through inhibition of phosphodiesterase activity, its effects on NO production in astrocytes and macrophages, observed in this study, are in contrast to those of cAMP.25-27 This finding suggests that in these cells, besides cAMP rise, PTX could trigger other intracellular event(s) able to overcome effects of cAMP on iNOS induction. With regard to data describing that cAMP analogues modulated NO production only at very high concentrations,25 it is also possible that PTX-mediated increase of cAMP level was insuficient to cause similar effect, not disturbing some other intracellular event(s) triggered by PTX to interfere with activation of iNOS. Reports from Greten et al.26 and Pahan et al.27 showing that specific phosphodiesterase type IV inhibitor rolipram acted similar to cAMP analogues, but contrary to PTX in our study, to enhance NO production in macrophages and reduce it in astrocytes, argue in favor of hypothesis that effects of PTX on NO production are independent of cAMP rise. The ability of PTX to inhibit macrophage NO production by mechanisms unrelated to cAMP increase is not unique among the phosphodiesterase inhibitors - Matsumori at al. had observed similar effects with amrinone, pimobendan and vesnarinone.25 This interesting cAMP-independent feature of some phosphodiesterase inhibitors, including PTX, to suppress NO production in macrophages probably is not a consequence of TNF-a transcription-block, because cAMP analogues also inhibit TNF-a synthesis,26 and is still to be explained, as well as mechanism responsible for PTX-mediated enhancement of NO production in astrocytes. Up-regulation of NO synthesis is a well-known marker of activation of different cell types, including astrocytes. Considering a myriad of immunosuppressive activities of PTX previously described,10-17 a proactivating effect of PTX on astrocytes reported in this study is particularly interesting in terms of initiation of future studies to explore effects of PTX on other parameters of astrocyte activation as well. Findings from such approach could contribute to better understanding of EAE-modulating properties of PTX and still undefined effectory/regulatory role of astrocytes in this disease. However, results presented here indicate that signaling events involved in regulation of iNOS induction in astrocytes and macrophages are different. These data could be relevant for therapy of demyelinating diseases, considering great importance and complexity of the role that NO from astrocytes and macrophages plays in both effectory and regulatory stages of immune processes affecting the brain.