General Anesthesia Delays the Inflammatory Response and Increases Survival for Mice with Endotoxic Shock

Animal model of anesthesia and endotoxic shock.

Eight-week-old, male B6 mice (20 to 30 g; Jackson Laboratory, Bar Harbor, ME) were housed in a Helicobacter species-free environment with ad libitum access to standard chow and water. The animal housing environment was maintained at a temperature of 22°C and with a 12-h light/dark cycle. Animals were acclimatized to their environment for 5 to 6 days following arrival and were then fasted for 16 h prior to any procedure. All animal experiments were performed in accordance with the Animal Care and Use Committee of the Johns Hopkins University School of Medicine and complied with NIH guidelines for animal experimentation. Mice were anesthetized using 2 to 2.5% vaporized inhaled isoflurane (IsoFlo; Abbott Laboratories, Chicago, IL). LPS from Escherichia coli serotype O26:B6 was obtained by trichloroacetic acid precipitation (Sigma-Aldrich, St. Louis, MO) and given via intraperitoneal (i.p.) injection at the following doses: 20 mg/kg of body weight for mortality studies and 15 mg/kg for all other studies. In one group of mice, anesthesia was immediately induced and then maintained for 1 h with isoflurane inhalation (simultaneous). A second group was anesthetized for 1 h after a 30-min interval following LPS injection (post-LPS injection). The third group received no anesthesia following LPS injection (LPS control). For another experiment, 8-week-old male mice from different strains (A/J, DBA/1J, BALB/CJ, 129S1/SvImJ, and C3H/HeN [20 to 30 g]; Jackson Laboratory, Bar Harbor, ME) all arrived the same week. Mice were randomized into either a simultaneous group or an LPS control group as described above. In a separate experiment, mice were anesthetized for 1 h and then injected with LPS (pretreatment group) or anesthetized for 1 h, allowed to recover for 30 min, and then injected with LPS (pretreatment/recovery group). Simultaneous and LPS control groups identical to those described above were also included in this experiment. For all animals, blood samples were collected via cardiac puncture 1.5 h after LPS (or saline) injection. Plasma was isolated via centrifugation and stored at −80°C. TNF-α, IL-6, and IL-10 plasma levels were determined by an enzyme-linked immunosorbent assay (Biosource, Camarillo, CA). For survival studies, animals were given ad libitum access to standard chow and water postprocedurally.

Other anesthetic agents.

Mice were injected with LPS intraperitoneally and simultaneously anesthetized for 1.5 h with pentobarbital (70 mg/kg i.p., Nembutal; Abbott Laboratories, Chicago, IL); ketamine (150 mg/kg i.p., Ketaject; Phoenix Pharmaceutical Inc., St. Joseph, MO) plus xylazine (15 mg/kg i.p., Xylaject; Phoenix Pharmaceutical Inc.); medetomidine (1 mg/kg i.p., Domitor; Orion Corp., Finland) plus ketamine (75 mg/kg i.p.); halothane (2% inhaled; Halocarbon Laboratories, River Edge, NJ); or ether (inhaled from gauze-soaked nose cone; J.T. Baker, Phillipsburg, NJ). Blood samples were collected via cardiac puncture immediately following the anesthetic period. Plasma was isolated via centrifugation and stored at −80°C.

Atropine treatment.

B6 mice were injected with atropine (Sigma-Aldrich, St. Louis, MO) prepared in sterile saline. The atropine was injected intraperitoneally at a dose of 1 mg/kg 10 min before the administration of LPS or the simultaneous administration of LPS and anesthesia (for 1 h). Two controls were used: mice injected with saline only (rather than atropine) 10 min prior to LPS, and mice injected with atropine without any further intervention (no LPS). Control animals received a comparable volume of saline.

Electrophoretic mobility shift assay.

Nuclear extracts were isolated using a modification of the protocol described by Dignam et al. (19). Briefly, 100 mg of liver was resuspended in 1 ml of hypotonic buffer (HEPES, 10 mM [pH 7.9]; KCl, 10 mM; EDTA, 0.1 mM; EGTA, 0.1 mM; dithiothreitol, 1 mM; phenylmethylsulfonyl fluoride, 0.5 mM; and 10 μg/ml each of leupeptin, aprotinin, and pepstatin), homogenized, and incubated on ice for 10 min. Samples were centrifuged at 850 × g at 4°C for 10 min, and the pellet was resuspended in 1 ml of buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride) and incubated on ice for 10 min. Nonidet P-40 at a final concentration of 0.6% was added, and lysates were vortexed for 1 min. Nuclei were isolated by centrifugation at 2,800 × g for 10 min at 4°C, and the pellet was resuspended in 200 μl of buffer C (HEPES, 20 mM [pH 7.9]; 1 M NaCl; 5% glycerol; 1 mM EDTA; 1 mM EGTA; 1 mM dithiothreitol; 0.5 mM phenylmethylsulfonyl fluoride; and 10 μg/ml each of leupeptin, aprotinin, and pepstatin) and gently agitated on an orbital shaker at 4°C for 30 min. The suspension was centrifuged at 14,000 × g for 10 min at 4°C, and the supernatant was aliquoted and stored at −80°C. An NF-κB DNA binding assay was performed using a double-stranded oligonucleotide containing the consensus binding site for NF-κB (Promega, Madison, WI), and the fragment was end labeled with [γ-³²P]ATP (7,000 Ci/mmol; NEN Life Science Products, Boston, MA) by using T4 polynucleotide kinase (Promega, Madison, WI) as described previously (23).

Endotoxin detection.

B6 mice (n = 3) were injected with LPS and simultaneously anesthetized with isoflurane or allowed to remain nonanesthetized (LPS control). Blood samples were obtained via cardiac puncture 0, 5, 10, 20, and 30 min after LPS injection, and samples were stored in sterile, pyrogen-free K⁺EDTA tubes. Plasma was isolated via centrifugation and stored at −80°C. Endotoxin levels were measured in plasma by use of an automated turbidimetric Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD). Before testing, serum was diluted 1:15 with endotoxin-free water and heated at 70°C for 7 min to remove inhibitors in plasma.

Statistical analysis.

Survival was analyzed via Kaplan-Meier analysis using the log rank test. Cytokine data shown are means ± standard errors of the means (SEM). General differences in cytokine levels among all groups were determined using one-way analysis of variance, with multiple pairwise comparisons by the Student-Newman-Keuls method being used to elucidate specific significances in these parameters between groups. All tests were two-tailed, and differences were considered significant when P was <0.05. Analysis was performed using Microsoft Excel (Microsoft Corporation, Seattle, WA) and SigmaStat (SPSS Incorporated, Chicago, IL) software.

Isoflurane anesthesia increases survival after LPS challenge.

The effects of anesthetics on the outcome of B6 mice after a lethal dose of LPS injection were compared between mice in the absence (control group) or presence of isoflurane for 1 h. Mice that were challenged with LPS and anesthetized developed fewer distress symptoms (e.g., decreased feeding, conjunctivitis, diarrhea, hypoactivity, piloerection, and trembling) than mice that were injected with LPS in the absence of isoflurane anesthesia. An 85% survival rate was observed for mice injected with LPS in the presence of isoflurane, compared to 23% survival in the control group (P < 0.01), within 72 h of the LPS injection (Fig. 1A). Furthermore, mice that survived the LPS challenge in the simultaneous isoflurane group were fully recovered by 48 h, as determined by locomotive activity and feeding, as opposed to 80 h for the nonanesthetized group.

Protection from endotoxic shock by isoflurane anesthesia correlates with attenuation of the inflammatory process.

The reduced mortality of anesthetized mice after LPS injection may be due to attenuation of the systemic inflammatory response induced by LPS. To test this idea, mice were injected with LPS and exposed to isoflurane for 1 h immediately (simultaneous group) or 30 min after the LPS injection (post-LPS injection group). These two groups were compared with nonanesthetized mice injected with LPS (control). Plasma samples were collected 1.5 h after LPS injection for each group. A decrease in TNF-α plasma levels was observed for the simultaneous group in comparison with levels for the other two groups (Fig. 1B). IL-6 and IL-10 levels were also reduced in the simultaneous and post-LPS injection groups compared to levels for the control group (Fig. 1B). To further characterize these findings, the effect of isoflurane pretreatment was investigated. Compared to levels in the LPS control group, TNF-α plasma levels 1.5 h after LPS injection were reduced in both the pretreatment/recovery and the pretreatment group. However, TNF-α levels for the two pretreatment groups were still significantly greater than the level for the simultaneous group (Fig. 1C). In contrast, IL-6 plasma levels were reduced only with the pretreatment and simultaneous groups, compared to levels for the control and pretreatment/recovery groups (Fig. 1C). Finally, IL-10 plasma levels in both pretreatment groups were greater than levels in the simultaneous and control groups (Fig. 1C). These data suggest that the suppressive effect of isoflurane on LPS-induced cytokine production increases survival in anesthetized animals. We also evaluated the effect of isoflurane anesthesia duration on the inflammatory response induced by LPS. Mice were injected with LPS and then administered continuous anesthesia for 0, 10, 30, or 60 min. Cytokine plasma levels were measured 1.5 h after injection. Decreasing levels of TNF-α and IL-6 were observed with increasing duration of anesthesia, whereas IL-10 levels decreased only with the 60-min anesthesia group (Fig. 2). Therefore, the timing and duration of isoflurane administration are important in modulating the LPS-induced inflammatory response.

Different classes of anesthetics also blunt the inflammatory response induced by LPS.

Our preceding observations prompted us to investigate whether other anesthetics would influence the inflammatory response similarly to isoflurane. Mice were injected with LPS and simultaneously anesthetized for 1.5 h with isoflurane, pentobarbital, or ketamine-xylazine. Plasma samples were collected upon completion of the anesthetic period. Both injectable and volatile anesthetics produced a significant reduction in serum TNF-α, IL-6, and IL-10 compared to levels for the LPS control group (Table 1). Halothane, ether, and ketamine-medetomidine were also found to attenuate cytokine production (data not shown). These results suggest that there is a common pathway of anesthetic attenuation of the inflammatory response.

Attenuation of LPS-induced cytokine levels is similar among genetically dissimilar mouse strains.

Different mouse strains have been found to display different sensitivities to anesthetics (42). Moreover, the inflammatory responses induced by LPS also differ among various mouse strains (17). Therefore, we compared the effects of isoflurane anesthesia on cytokine production induced by LPS in B6, A/J, DBA/1J, BALB/CJ, 129S1/SvImJ, and C3H/HeN mice. Among all strains, simultaneous administration of anesthesia and LPS attenuated TNF-α, IL-6, and IL-10 compared to use of LPS alone (Table 2), indicating that the effect of isoflurane is broad with respect to host genetic background.

Effect of isoflurane in decreasing LPS-induced cytokine production is not mediated through the cholinergic pathway.

Current hypotheses regarding the molecular mechanism underlying the anesthetic action of isoflurane involve alterations in neuronal synaptic transmission (26), with nicotinic acetylcholine receptors potentially being the most modulating targets (24). Therefore, we tested the hypothesis that inhaled isoflurane anesthesia stimulates the recently described cholinergic anti-inflammatory pathway (nicotinic receptors) (45), thus decreasing cytokine production in LPS-stimulated mice. Compared to saline control 1.5 h after LPS injection, atropine-treated mice displayed reduced plasma TNF-α levels (P < 0.01) and increased IL-10 levels (P < 0.001), with no change in IL-6 levels (Fig. 3). Most notably, isoflurane still significantly attenuated serum production of all three plasma cytokines (to levels comparable to atropine without LPS) in atropine-treated animals (10 min pre-LPS). These observations suggest that the effect of isoflurane on the inflammatory process may be independent of the cholinergic pathway and, thus, is unlikely to be mediated by the vagus nerve or nicotinic receptors.

Isoflurane delays the activation of NF-κB, cytokine production, and levels of LPS in circulation.

We also investigated the activation of NF-κB, a central component of the LPS response. B6 mice were injected with LPS alone or with LPS plus simultaneous isoflurane anesthesia for 0, 5, 10, 20, 30, or 60 min. Liver samples were harvested at the end of the anesthetic period, and NF-κB binding to DNA was determined by a gel shift assay. NF-κB normally resides in the cytoplasm bound to the inhibitor IκB complex. After stimulation with LPS, IκB is phosphorylated and degraded, allowing NF-κB to translocate into the nucleus, resulting in the transcription of many inflammatory genes (4, 16). NF-κB DNA binding was retarded in samples from isoflurane-anesthetized mice in comparison with samples from the nonanesthetized group (Fig. 4A). Thus, anesthesia appears to cause a delay in the activation of NF-κB.

The peak of plasma TNF-α after LPS injection also shifted to a later time point in anesthetized mice than in nonanesthetized controls (Fig. 4B). IL-6 levels were elevated overall and peaked earlier in the nonanesthetized group than in mice treated with isoflurane (Fig. 4B). Finally, IL-10 levels rose later and peaked higher in the isoflurane-treated group than in the nonanesthetized controls (Fig. 4B). The concentrations of LPS in systemic circulation following its intraperitoneal injection in both anesthetized and nonanesthetized mice were compared. The appearance of LPS in circulation was slower in anesthetized mice than in the control group (Fig. 5), which correlates with the delayed activation of NF-κB and cytokine kinetics.