Delayed Treatment with Isoflurane Attenuates Lipopolysaccharide and Interferon γ–induced Activation and Injury of Mouse Microglial Cells

C8-B4 cells (CRL-2540), a microglial clone isolated from 8-day-old mouse cerebellum, were purchased from the American Type Culture Collection (Manassas, VA). Lipopolysaccharide from Escherichia coli  0111:B4 and other chemicals except for those described below were obtained from Sigma-Aldrich (St. Louis, MO). Heat inactivated fetal bovine serum, Griess Reagent Kit (G7921), and recombinant rat interferon γ (IFNγ) produced from E. coli  were purchased from Invitrogen Corporation (Carlsbad, CA). Rabbit polyclonal anti-iNOS immunoglobulin G whose epitope is at the C-terminus of iNOS was obtained from Santa Cruz Biotechnology (Santa Cruz. CA; catalogue number: sc-650). The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)-based colorimetric assay kit was obtained from Chemicon International, Inc. (Temecula, CA). Isoflurane was purchased from Abbott Laboratories (North Chicago, IL). Chelerythrine chloride and bisindolylmaleimide IX (R8220) were obtained from Biomol (Plymouth Meeting, PA). Lactate dehydrogenase (LDH) activity assay kit was from Clontech Laboratory (La Jolla, CA).

As we described before,4C8-B4 cells were cultured in Dulbecco Modified Eagle’s Medium containing 4 mm l-glutamine, 4,500 mg/l glucose, 1 mm sodium pyruvate, and 1,500 mg/l sodium bicarbonate supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin, and 100 pg/ml streptomycin in a humidified atmosphere of 95% air–5% CO²at 37°C. The medium was changed every 3 days. The cells were plated at a density of 4-5 × 10⁴cells/ml on 96-well tissue culture plates for viability experiments by MTT assay and LDH release experiments and plated at a density of 2 × 10⁴cells/ml in 6-well plates for other experiments.

Three hours after the cells were plated, they were exposed to 10 ng/ml lipopolysaccharide and 10 U/ml IFNγ for 24 h. The amounts of lipopolysaccharide and IFNγ applied were based on our previous experiment.4Lipopolysaccharide concentration was increased to 1,000 ng/ml (IFNγ was still at 10 U/ml) in the LDH-release experiments because our preliminary study showed that this combination of lipopolysaccharide and IFNγ induced a significant release of LDH.

Isoflurane exposure was performed in an airtight chamber as we described previously.4,11The chamber was gassed with 95% air–5% CO²through or not through an isoflurane vaporizer for 10 min at 37°C. The isoflurane concentrations in the gases from outlet of the chambers were monitored with a Datex infrared analyzer (Capnomac, Helsinki, Finland) and reached the target concentrations at 3 min after the onset of gassing. Isoflurane was applied for 1 h immediately after the addition of lipopolysaccharide and IFNγ to the incubation medium except for the time window experiment that isoflurane was applied at 0, 2, 4, 8, 16, and 23 h after the addition of lipopolysaccharide and IFNγ to the incubation medium. Two percent isoflurane was used except for the isoflurane dose-response experiment in which 1, 2, or 3% isoflurane was applied. At the end of the 1-h incubation time, isoflurane concentrations in the gases from outlet of the chamber were confirmed to be at the target concentrations. The cells were then removed from the chamber and were kept under isofluorane-free conditions for the rest times of incubation with lipopolysaccharide and IFNγ. The cells and/or incubation solution were harvested for viability assay, LDH activity assay, or nitrite measurement at 24 h after the initiation of lipopolysaccharide and IFNγ stimulation. As we described before,7the aqueous isoflurane concentrations at 37°C were 209, 415, and 620 μm, respectively, as measured by gas chromatography when 1, 2, and 3% isoflurane was delivered, and the samples were taken for measurements at the end of the 1-h isoflurane incubation.

Aminoguanidine (10 μΜ), an iNOS inhibitor, was added to some cells for 24 h during lipopolysaccharide and IFNγ application. These cells were not treated with isoflurane. Chelerythrine chloride (2 μΜ) or bisindolylmaleimide IX (10 μΜ), protein kinase C (PKC) inhibitors, were added to the incubation medium during isoflurane application. In these cases, the incubation solutions were replaced with fresh medium not containing these reagents after isoflurane application and before the reexposure to lipopolysaccharide and IFNγ. Similar procedure (incubation medium changing) was applied to other experimental groups in this set of experiments.

Cells plated on 96-well plates were gently mixed with 10 μl of MTT solution and incubated for 3 h at 37°C. After formation of black fuzzy crystals, 100 μl of isopropanol with 0.4 N HCl was added and thoroughly mixed in each well. One hour later, the absorbance of samples was measured at 570 nm with the reference wavelength of 650 nm by using a microplate reader (Bio-Rad Laboratories, Hercules, CA). The results of MTT measurements from the controls without any treatments were set as 1. The values from various treatments were then calculated as percentages of the controls.

To detect the concentrations of nitrite, a stable oxidation product of nitric oxide, equal volumes of N -(1-naphthyl)ethylenediamine and sulfanilic acid were mixed to form Griess reagent immediately before experiments. The culture media collected from 6-well plates with the cells treated with various conditions were centrifuged. Griess reagent (20 μl), the culture media (150 μl), and deionized water (130 μl) were mixed in 96-well plates and incubated in the dark for 30 min at room temperature. The absorbance of samples was measured at 570 nm against a standard curve in a spectrophotometry (Bio-Rad Laboratories).

Two-month old C57Bl/6 male mice received intraperitoneal injection of 4 mg/kg lipopolysaccharide or saline (control mice). They were then placed immediately or at 2 h after the lipopolysaccharide injection in an airtight chamber that was continuously gassed with air containing 1.5% isoflurane for 30 min. This isoflurane exposure condition was chosen because our preliminary study showed that a higher concentration (2% isoflurane) or a longer incubation (1 h) significantly altered the arterial blood gases (most animals had a Paco²greater than 60 mmHg and a Pao²less than 50 mmHg at the end of the exposure). The animals were killed at 6 h after the lipopolysaccharide injection, and their brain cortices were harvested for Western blotting.

Cells or cerebral cortices of animals after various conditions were harvested and homogenized in 25 mm TRIS hydrochloride, pH 7.4, containing 1 mm EDTA, 1 mm EGTA, 0.1% (vol/vol) α-mercaptoethanol, 1 μm phenylmethylsulfonyl fluoride, 2 μm leupeptin, and 1 μm pepstatin A.2The homogenates were centrifuged at 14,000 g  for 10 min at 4°C. The supernatants were used for Western blotting, and 40 μg of proteins were loaded to each lane. After incubation with the anti-iNOS or anti–β-actin antibody, the protein bands were visualized by the enhanced chemiluminescence reaction. The protein bands were densitometrically analyzed by an ImageQuant 5.0 densitometer (Amersham Biosciences, Piscataway, NJ). The results of iNOS protein bands were normalized by the data of the corresponding β-actin. The results of cells treated with various conditions were then normalized by the data of cells exposed to lipopolysaccharide and IFNγ only. The results of animals treated with various conditions were then normalized by the mean value of control animals. These different normalization procedures were used for cell culture and animal studies because microglial cell cultures without exposure to lipopolysaccharide and IFNγ did not express iNOS.

Incubation solution of the microglial cells treated with various conditions was centrifuged at 1,000g  for 10 min, and the cell-free supernatant was transferred to 96-well plates. The 100 μl of supernatant was incubated with the same amount of reaction mixture from the LDH detection kit. LDH activity was determined by a colorimetric assay. The absorbance of samples was measured at 492 nm, with the reference wavelength of 655 nm in a spectrophotometry (Bio-Rad Laboratories). Background absorbance from the cell-free buffer solution was subtracted from all absorbance measurements. After removal of the incubation solution from cells, 1% triton X-100 lysing solution was applied to the cells. The percentage of LDH released to incubation solution in total LDH was calculated: 100 × LDH in the incubation solution/(LDH in the incubation solution + intracellular LDH released by triton X-100).

Each experimental condition was repeated multiple times (n for each condition is described in the figure legends) by using at least three different batches of cells. At least seven animals were used for each experimental condition in the in vivo  study. Data are expressed as mean ± SD. Statistical analyses were performed by one-way analysis of variance followed by the Tukey test for post hoc  analysis or by Student t  test for comparisons between lipopolysaccharide plus IFNγ alone and the corresponding control in the time course studies involving MTT or LDH release assay. P < 0.05 was considered as statistically significant.

Lipopolysaccharide and IFNγ application reduced the cell viability to 70 ± 6% of control as measured by MTT assay. Two percent isoflurane applied at 0 and 2 h after the lipopolysaccharide and IFNγ application enhanced the cell viability to 94 ± 4% and 79 ± 4%, respectively (both P < 0.05 compared with lipopolysaccharide and IFNγ alone). However, delayed application of isoflurane at 4 h or longer after the addition of lipopolysaccharide and IFNγ did not improve the cell viability (fig. 1A). Concentration-response study showed that isoflurane at 2%, but not at 1% or 3%, applied immediately after the initiation of lipopolysaccharide, and IFNγ stimulation improved cell viability (fig. 1B). Consistent with our previous study,4the lipopolysaccharide and IFNγ-induced reduction of cell viability was abolished by aminoguanidine (fig. 1C), an iNOS inhibitor, suggesting the role of iNOS in this lipopolysaccharide and IFNγ effect.

Although chelerythrine chloride and bisindolylmaleimide IX, two PKC inhibitors, did not affect cell viability in the presence of lipopolysaccharide and IFNγ, these PKC inhibitors abolished the protective effects of isoflurane on microglial viability after the application of lipopolysaccharide and IFNγ (fig. 2).

The microglial cell cultures did not express iNOS under normal culture conditions (fig. 3A). Lipopolysaccharide and IFNγ induced a large amount of iNOS expression. This iNOS expression was inhibited by isoflurane applied immediately or at 2 h, but not at 8 h, after the initiation of the lipopolysaccharide and IFNγ stimulation (fig. 3). Chelerythrine chloride and bisindolylmaleimide IX did not affect the iNOS expression in cells exposed to lipopolysaccharide and IFNγ. However, chelerythrine chloride and bisindolylmaleimide IX abolished the isoflurane-induced inhibition of iNOS expression stimulated by lipopolysaccharide and IFNγ (fig. 3). Consistent with the results/effects of isoflurane on iNOS protein expression, 2% isoflurane reduced the production of nitrite after lipopolysaccharide and IFNγ stimulation (fig. 4).

To determine the effects of isoflurane on lipopolysaccharide-induced iNOS expression in the mouse brain under in vivo  condition, we first identified an isoflurane exposure condition that would not severely alter the arterial blood gases in mice without tracheal intubation and mechanic ventilation. The mouse arterial blood pH, Paco²and Paco²were 7.26 ± 0.05, 56 ± 13, and 96 ± 27 (n = 5), respectively, at the end of the exposure to 1.5% isoflurane for 30 min. We chose this isoflurane exposure condition for further experiment.

Similar to our in vitro  experiments, lipopolysaccharide injection significantly increased iNOS expression in the cerebral cortex. This increased iNOS expression was also inhibited by the exposure to 1.5% isoflurane for 30 min applied immediately or 2 h after the lipopolysaccharide injection (fig. 5).

Lipopolysaccharide and IFNγ caused a significant LDH release from microglial cell cultures (fig. 6A), suggesting that lipopolysaccharide and IFNγ induced microglial injury. Interestingly, this increased LDH release was abolished by isoflurane applied at later time points (8, 16, and 23 h after the addition of lipopolysaccharide and IFNγ), but it was not affected by isoflurane applied at early time points (fig. 6A). Further study showed that lipopolysaccharide and IFNγ caused more LDH release than control condition only after application of lipopolysaccharide and IFNγ for longer than 12 h (fig. 6B). In addition, the lipopolysaccharide and IFNγ-induced LDH release was not affected by aminoguanidine (fig. 6C).

Similar to our previous study,4the incubation of microglia with lipopolysaccharide and IFNγ reduced cell viability. It has been well-established that lipopolysaccharide and IFNγ can stimulate/activate microglial cells to express iNOS.3,4,12We have shown that this increased iNOS expression contributes to microglial damage caused by lipopolysaccharide and IFNγ.4Consistent with our previous findings, we showed that aminoguanidine, an iNOS inhibitor, abolished the decreased viability of microglial cells after lipopolysaccharide and IFNγ application.

The lipopolysaccharide and IFNγ-reduced microglial viability was dose-dependently attenuated by isoflurane that was applied immediately after the application of lipopolysaccharide and IFNγ. This protection existed when isoflurane was applied even at 2 h after the application of lipopolysaccharide and IFNγ. These results suggest that delayed treatment with isoflurane can induce a protective effect in microglia. Isoflurane exposure immediately or at 2 h after the application of lipopolysaccharide and IFNγ also decreased the lipopolysaccharide and IFNγ-induced iNOS expression and nitrite production. These results, along with the knowledge that iNOS plays a critical role in damaging microglial cells after stimulation with lipopolysaccharide and IFNγ,4,12indicate that the mechanisms for protection provided by delayed treatment with isoflurane involve decreasing lipopolysaccharide and IFNγ-induced iNOS expression.

Consistent with these in vitro  cell culture results, our studies also showed that delayed isoflurane treatment also inhibited lipopolysaccharide-induced iNOS expression in the mouse brain. This finding is significant because volatile anesthetic effects on iNOS expression in the brain under in vivo  conditions have not been reported. Although it is not known from our study in which cell types isoflurane treatment decreased iNOS expression, it has been shown that microglial cells are the primary cells expressing iNOS in rodent brains after lipopolysaccharide stimulation.13,14This knowledge, along with our results, suggests that delayed isoflurane treatment can also decrease lipopolysaccharide-induced iNOS expression in the microglial cells of mouse brain.

PKC is a group of important intracellular signaling molecules that are involved in a broad range of biologic functions, such as cell survival.15,16It has been shown that many isoflurane pharmacological effects may be mediated by PKC.11,17,18In this study, we showed that the isoflurane protective effects were inhibited by chelerythrine and bisindolylmaleimide IX. The isoflurane-decreased iNOS expression was also inhibited by chelerythrine and bisindolylmaleimide IX. Chelerythrine and bisindolylmaleimide IX are structurally different. Chelerythrine is a peptide site competitive PKC inhibitor, and bisindolylmaleimide IX is an adenosine triphosphate site competitive PKC inhibitor.19Our results suggest that PKC mediates the isoflurane protective effects in the microglial cells. Consistent with our results, volatile anesthetics may directly stimulate PKC.20Also, although activation of PKC has been shown to enhance lipopolysaccharide- and cytokines-induced iNOS expression,21activation of certain PKC isozymes, such as PKCϵ, can inhibit lipopolysaccharide- and cytokines-induced iNOS expression in macrophages and astroglial cells.4,22–24Although the exact mechanisms for the regulation of lipopolysaccharide-induced iNOS expression by PKC are not fully understood, it has been shown that PKC regulates activation of signal transducer and activator of transcription 1, which can then modulate iNOS expression.21 

Isoflurane has been shown to inhibit lipopolysaccharide-induced iNOS expression in J774A.1 cells, a macrophage-like cell line, and this isoflurane effect was attenuated by the calcium ionophore ionomycine.25These results suggest the role of decreased intracellular free Ca²⁺in mediating this isoflurane effect. Consistent with this idea, increased intracellular Ca²⁺has been proposed to mediate lipopolysaccharide- or IFNγ-induced iNOS expression in mouse macrophages,26although contrary results (increased intracellular Ca²⁺leads to decreased iNOS expression in mouse macrophages) have also reported.27Similar to this confusing picture, many studies have reported a decrease of intracellular Ca²⁺by volatile anesthetics,28,29and other studies showed that volatile anesthetics induced a slight increase of intracellular Ca²⁺.30,31Thus, although lipopolysaccharide has been consistently shown to increase iNOS expression in various cells, including microglial cells, further studies are needed to clarify the role of intracellular Ca²⁺in iNOS expression and the regulation of volatile anesthetics on this process.

We observed a bell-shaped response: 2% isoflurane, but not 1% and 3% isoflurane, reduced lipopolysaccharide and IFNγ-induced decrease of cell viability and nitrite production. The mechanisms for the lost protective effects at a higher concentration of isoflurane are not known. However, activation of different PKC isozymes may increase or decrease iNOS expression in macrophages stimulated by lipopolysaccharide and IFNγ,4,21–243% isoflurane may activate PKC isozymes that can increase iNOS expression. This effect may cancel out the inhibitory effects of isoflurane on iNOS expression via  other PKC isozymes. Preferential activation of PKC isozymes by various concentrations of isoflurane has been hypothesized to explain the bell-shaped effects of isoflurane on contraction of skinned pulmonary arterial strips and isoflurane preconditioning-induced protection in macrophages.18,32Also, we exposed microglial cells to various concentrations of isoflurane for 1 h. It is not known whether this length of exposure is optimal for its protective effect, especially for 1% or 3% isoflurane. Finally, prolonged exposure to high concentrations of isoflurane can induce cell injury.33Exposure to 3% isoflurane for 1 h in the presence of lipopolysaccharide and IFNγ may be strong enough to cause a detrimental effect, which counteracts its protective effect.

In addition to MTT assay, we used LDH release to measure isoflurane protective effects. An increased LDH release may require plasma membrane damage and signify that cells are injured. Our results showed that isoflurane application at later time points, but not at early time points, abolished lipopolysaccharide and IFNγ-induced LDH release. This finding contrasts sharply with our results from MTT assay, where isoflurane application only at early time points decreased lipopolysaccharide and IFNγ-induced decrease of MTT metabolism. Decreased MTT metabolism suggests a decrease of mitochondrial reduction ability and, therefore, decreased cell viability. Our results showed that the lipopolysaccharide and IFNγ-induced decrease of MTT metabolism was iNOS-dependent and that isoflurane exposure at early time points, but not a later time point, after the application of lipopolysaccharide and IFNγ reduced the lipopolysaccharide and IFNγ-induced iNOS expression. Thus, the attenuation of lipopolysaccharide and IFNγ-induced decrease of MTT metabolism by isoflurane may be the results of its inhibition of iNOS expression. In contrast, the isoflurane-decreased cell injury of the cells exposed to lipopolysaccharide and IFNγ, as measured by LDH release, may not be the result of the effects of isoflurane on iNOS expression because isoflurane applied at early time points reduced lipopolysaccharide and IFNγ-induced iNOS expression but failed to decrease the lipopolysaccharide and IFNγ-induced LDH release. Consistent with this idea, aminoguanidine did not reduce lipopolysaccharide and IFNγ-induced LDH release. In our study, it appears that more LDH was released from cells incubated with lipopolysaccharide and IFNγ for longer than 12 h, and isoflurane application around and during this period significantly attenuated this lipopolysaccharide and IFNγ-induced LDH release. Future studies are needed to understand the mechanisms for isoflurane to reduce the LDH release under this experimental condition.

Microglial cells play a critical role in immune reactions and host defense. They perform active tissue scanning and respond to various endogenous and exogenous stimuli. We used lipopolysaccharide and IFNγ, a common in vitro  method to simulate endotoxemia and to induce inflammation, to activate microglial cells. Considering the fact that inflammatory process is involved in almost all of the human diseases acquired later in their lives, our findings that isoflurane applied after the lipopolysaccharide and IFNγ administration can reduce microglial activation and injury may have a broad implication.

In conclusion, we have shown that delayed treatment with isoflurane can induce a protective effect in microglial cells and reduce iNOS expression, a microglial cell activation indicator, during inflammatory process. PKC may mediate these isofluorane-protective effects.