| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Division of Pulmonary and Critical Care Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea; and Department of Pathology, Ilsan Paik Hospital, Inje University College of Medicine, Koyang-Si, Korea.
Address correspondence and reprints requests to Chae-Man Lim, MD, Division of Pulmonary and Critical Care Medicine, Asan Medical Center, Songpa P.O. Box 145, Seoul, Korea 138-600. Address e-mail to cmlim{at}amc.seoul.kr.
Abstract
BACKGROUND: Hypothermia may be effective for endotoxin-induced acute lung injury. In most studies, hypothermia was induced before the development of neutrophilic inflammation, which would be clinically irrelevant. We investigated whether hypothermia induced after the onset of such neutrophilic inflammation reduces acute lung injury.
METHODS: In the first experiment, rats were allocated to one of four groups: intratracheal saline instillation/killed at 1 h (saline), intratracheal lipopolysaccharide (LPS) instillation/killed at 1 h (LPS-primed), intratracheal LPS instillation/killed at 6 h (LPS-NT), all under normothermia (NT) (37 ± 0.5°C) throughout study, and intratracheal LPS instillation/killed at 6 h with hypothermia (HT) (32 ± 0.5°C) for the last 5 h of study (LPS-HT). Lungs were lavaged for biochemical measurements. In the second experiment in 26 additional rats, we followed exactly the same protocol as described above for the saline group (n = 2), LPS-NT (n = 12), and LPS-HT (n = 12), and obtained a fresh pool of alveolar neutrophils to assess oxidative burst.
RESULTS: Compared with the LPS-primed group, the neutrophil count, protein level, and lactate dehydrogenase activity in the bronchoalveolar lavage fluid, and myeloperoxidase activity of the lung were all higher in the LPS-NT group. Compared with this LPS-NT group, the neutrophil count, protein level, and lactate dehydrogenase activity in the bronchoalveolar lavage fluid, and microscopic scores for alveolar neutrophilic infiltration were all lower in the LPS-HT group. The stimulated production of hydrogen peroxide in neutrophils was lower in the LPS-HT group than in the LPS-NT group.
CONCLUSION: Hypothermia, applied even after the onset of neutrophilic inflammation, was effective in reducing endotoxin-induced acute lung injury.
Neutrophils are regarded as the key effector cells in acute lung injury (ALI) associated with endotoxin (1–4). After endotoxin challenge via various routes, neutrophils are sequestered from the circulation into the airspace, where they release toxic intermediates, such as reactive oxygen species (ROS) (5,6), and proinflammatory cytokines, such as interleukin (IL)-1ß and tumor necrosis factor (TNF)-
(7–9), and consequently perpetuate ALI. No treatment is available that directly modulates this pathogenetic process.
Hypothermia (HT), whether accidental or induced, increases bacterial infections (10,11). The increased susceptibility to infection during HT is due to the inhibition of inflammatory cells, especially neutrophils (12–16). In human and animal studies, HT supresses neutrophil-mediated inflammation according by various means, including the neutrophil release from bone marrow stimulated by endotoxin, the increase in the number of circulating neutrophils (17), their interaction with the endothelium and subsequent migration into tissue (18,19), and their enzymatic activity (12,14). Because exuberant neutrophil-mediated inflammation is a salient pathophysiological feature of endotoxin-associated ALI, HT could be effective in the prevention and/or treatment of this disease.
On the basis of this assumption, we (20) have demonstrated the effects of HT (10°C below normal body temperature) on endotoxin-induced ALI (20). In that study, animals pretreated with HT showed significant attenuation of neutrophil recruitment and histological changes in the lung. Regarding a therapeutic effect, Chu et al. (21) reported a beneficial effect of HT on lipopolysaccharide (LPS)-induced ALI in rats in terms of lung edema, and neutrophilic infiltration of the lung. In addition to their investigation, we were further interested in the effect of therapeutic HT on the oxidative burst of sequestered neutrophils in the airspace. In addition, the HT adopted in our previous studies was deeper than the HT currently used in clinical practice, i.e. 32–35°C (22–25), even close to the range at which fatal ventricular arrhythmia may occur (26). Therefore, we thought it worthwhile to investigate if HT is effective even in lungs in which neutrophilic inflammation has already been initiated, and whether these effects can be achieved with a moderate degree of HT.
METHODS
Animal Preparation and Induction of HT
Male specific pathogen-free Sprague-Dawley rats (Laboratory Animal Section, Asan Institute for Life Sciences, Seoul, Korea) were used for this study. Care and handling of the animals complied with the guidelines of the National Institutes of Health, United States, and was approved by our institutional animal care committee. In the first series of experiments, rats were randomly assigned to one of four groups: Saline group (n = 8, 329 ± 25 g), LPS-primed group (n = 8, 311 ± 39 g) (baseline), LPS-normothermia (NT) group (n = 12, 305 ± 35 g), or LPS-HT group (n = 12, 308 ± 44 g). In the saline and LPS-primed groups, rats were intratracheally administered equivalent volumes of normal saline or LPS from Escherichia coli serotype 055:B5 in 10 mg/mL concentration (5 mg/kg; Sigma Chemical Co., St. Louis, MO), and subjected to thoracotomy at 1 h of the intratracheal treatment for bronchoalveolar lavage (BAL) and procurement of lung tissue. These 1-h groups were intended to confirm if the lung would be primed by neutrophilic inflammation before HT intervention. In the LPS-NT and LPS-HT groups, rats were intratracheally administered LPS in the same manner as that used in the LPS-primed group, and subjected to thoracotomy at 6 h of treatment. These 6-h groups were intended to show if HT posttreatment would decrease progression of ALI from the baseline as compared with the NT condition. In the 6-h experiment, we chose not to do saline-HT intervention, as there was no discernable ALI in our previous study (20). While core body temperature of all other groups as measured 5 cm deep in the rectum was maintained at NT (37 ± 0.5°C) either for 1 h (saline, LPS-primed) or 6 h (LPS-NT), that of the LPS-HT group was decreased to HT (32 ± 0.5°C) at 1 h of LPS administration, and maintained for the 5 ensuing hours.
In all animals, single intraperitoneal doses of xylazine (7 mg/kg) and ketamine (90 mg/kg) were administered to induce anesthesia. In the LPS-HT group, the time required to reduce rectal temperature to 32°C was 12–15 min. The rectal temperature in each group was then maintained within ±0.5°C of the target temperature by intermittent application of a blanket or ice bed until removal of the lungs. Throughout the study, rats were allowed to breathe spontaneously without mechanical ventilatory support or supplemental oxygen. The tracheal incision site was sutured immediately after intratracheal treatment of LPS or saline, and anesthesia was maintained until the thoracotomy using xylazine and ketamine at half the initial dose, given every 2 h.
In the second series of experiments, which was intended to compare the histology of the lung and the oxidative burst activity of sequestered neutrophils, another 26 rats were randomly assigned to normal (n = 2), the LPS-NT (n = 12), and LPS-HT (n = 12) groups, for which the experimental protocol was the same as in the LPS-NT and LPS-HT groups of the first series. All rats in both series of experiments survived the intended durations of experiment.
Lung Perfusion, BAL, and Lung Tissue Collection
At either 1 or 6 h after LPS or saline instillation, the rats were anesthetized again for surgical procedures with the same regimen of xylazine and ketamine, at half the initial dose. For the following procedure, rats' lungs were briefly ventilated through a tracheal cannula, using a Servo 300 ventilator (Siemens-Elema, Sonla, Sweden) at a controlled pressure of 8 cm H2O, a positive end-expiratory pressure of 2 cm H2O, and a fractional concentration of inspired oxygen of 0.21. In the LPS-NT and LPS-HT groups, arterial blood was obtained immediately before thoracotomy and analyzed using the Blood gas system 288 (Ciba-Corning, Medfield, MA). The thorax was opened using a midline thoracotomy, and held open with a rib spreader. After heparin (200 U) was administered into the right ventricle, a catheter was placed in the proximal pulmonary artery. With the pulmonary artery ligated, the lungs were then subjected to perfusion with phosphate-buffered saline (PBS) (pH 7.4) with gravity of 50 cm, until the eluent became clear. The lungs were removed en bloc, and a blunt-tipped needle was inserted into the right main bronchus for BAL. BAL was performed 3 times using 4 mL of cold (4°C) normal saline. After the left main bronchus had been tied, the left lung was cut off and frozen in liquid nitrogen for the later assay of myeloperoxidase (MPO) activity.
Because the two lungs in the first experiment were either lavaged or frozen, a second series of experiments was performed to obtain a fresh neutrophil pool from the bronchoalveolar space to assess the oxidative burst reaction and obtain independent lung specimens for microscopic assessment. BAL was performed in the right lung using PBS at a temperature of 37°C for the LPS-NT group or at 32°C for the LPS-HT group. A catheter was secured to the left main bronchus to inflate the lung with 10% neutral-buffered formalin (pH 7.0), until the pleural margins became sharp (1.5–2 mL). The lung was further fixed by immersion in formalin overnight until it was processed in paraffin wax.
First Set of Experiments
BAL Polymorphonuclear Leukocyte Counts
The lavage fluid retrieved from the rats' lungs was centrifuged at 400g for 10 min. The supernatant was collected and stored at –80°C for cytokine assays. Total leukocyte counts were determined by hemocytometer. The percentage of polymorphonuclear leukocyte (PMN) was determined in 300 cells, on a cytospin-prepared slide with modified Diff-Quik stain (Baxter Diagnostics, Düdingen, Switzerland). The total number of PMN was then calculated from the total leukocyte number and the percentage of PMN.
Tissue Extraction and Measurement of MPO Activity
Upon thawing, the rats' lung tissues were homogenized in phosphate buffer (20 mM, pH 7.4) and centrifuged at 30,000g for 30 min. The pellet was resuspended in another phosphate buffer (50 mM, pH 6.0) with 0.5% hexadecyltrimethylammonium bromide. MPO activity in the resuspended pellet was assayed by measuring absorbance at 460 nm with a Beckman DU spectrophotometer (Beckman Instruments, Fullerton, CA), using 0.167 mg/mL O-dianisidine hydrochloride and 0.0005% H2O2. The results were expressed as absorbance per gram of lung tissue (U/g).
Measurement of Cytokines in BAL Fluid and Serum
The concentrations of TNF- and IL-1ß of the rats' BAL fluid were measured by a solid-phase sandwich ELISA, using a commercial immunoassay Cytoscreen kit for rats (Biosource International, Inc., Camarillo, CA). Optical density was read at 450 nm using a ThermoMax Microplate Reader (Molecular Devices Corp., Menlo Park, CA).
Determination of Protein Concentrations and Lactate Dehydrogenase Levels in BAL Fluid
In the supernatant of the centrifuged rats' BAL fluid, the protein concentration was assayed as a measure of alveolo-capillary permeability (27), using a bicinchoninic acid protein assay kit (Pierce Chemicals Inc., Rockford, IL) (28). Lactate dehydrogenase (LDH) levels were assayed as a measure of epithelial injury in the alveoli (29), using a Cytotoxicity Detection Kit (Roche Molecular Biomedicals, Mannheim, Germany).
Second Set of Experiments
Assessment of the Oxidative Burst of Neutrophils Recovered from BAL Fluid
To determine the oxidative burst activity of neutrophils, a 2',7'-dichlorofluorescein (DCF) assay of H2O2 was performed with flow cytometry (30). This assay quantifies the intracellular oxidation of DCFH to the fluorescent compound DCF. The oxidation of DCFH occurs after the respiratory burst, after the treatment of neutrophils with a stimulating agent such as phorbol 12-myristate 13-acetate (PMA). Upon cell activation, H2O2 oxidizes the trapped DCFH to DCF, which is fluorescent, and the green fluorescence produced is proportional to the amount of H2O2 generated in the cell. In brief, the cells pelleted from the BAL fluid in the second experiment were immediately resuspended to 105–106 cells/mL. These were incubated in 20 mM DCFH- diacetate (Eastman Kodak, Rochester, IL) for 20 min. The neutrophil oxidative burst was determined by stimulation with PMA (100 ng/mL) of the cells resuspended in PBS solution. The samples were analyzed with a FACScan (Becton Dickinson, San Jose, CA) at 530 nm and Cytoquest software (Becton Dickinson). The DCF fluorescence histogram for the neutrophils was obtained by gating neutrophils on a histogram with forward scatter (x axis) and side scatter (y axis). Data were collected from reagent blanks and all resting and PMA-stimulated tubes. A total of 10,000 events was collected for each sample. At analysis, the scattergram of forward scatter versus side scatter was displayed, and the neutrophil population was gated by its typical location. The mean value of the fluorescence was regarded as the intensity of the intracellular H2O2 in the neutrophils (30).
Microscopic Scoring of ALI
Four sections were cut, from the lung base toward the apex, at 3 mm intervals and stained with hematoxylin and eosin to assess lung injury. Ten random fields per animal were read under 200x by two independent pathologists, who were unaware of the treatment and temperature regimes of the rats. Sections were inspected for abnormal cellular infiltration in alveolar, interstitial, or perivascular spaces, and for alveolar exudate, and these variables were semiquantitatively graded as described in Table 1. The scores of both pathologists for individual histological variables were averaged, and then all the variables were summed to give the histological score for ALI (total score: 0–11).
|
Statistical Analysis
Data are presented as medians (interquartile ranges). The statistical analysis was performed using SPSS 7.5 programs (SPSS Inc., Chicago, IL). The Kruskal–Wallis test was used to detect differences among the four groups. Post hoc multiple comparisons were performed with Tukey's method. Comparisons of two independent groups were made with the Mann–Whitney U-test. P values <0.05 were considered significant.
RESULTS
Characteristics of the LPS-Primed Group
Compared with the saline group, the neutrophil count (Fig. 1; P < 0.001), MPO activity [1.41 (1.22–2.28) U/g, 2.26 (1.10–4.02) U/g; P < 0.01], TNF- level [99.8 (73.5–365) pg/mL, 1880 (348–2220) pg/mL; P = 0.011], and IL-1ß level in the BAL fluid [1020 (576–1780) pg/mL, 2280 (1410–3550) pg/mL; P = 0.014] were all higher in the LPS-primed group.
|
Progression of ALI in the LPS-NT Group
Compared with this LPS-primed group (1 h after LPS injury), the neutrophil count (P = 0.02) (Fig. 1), protein content (P < 0.001) (Fig. 2), and LDH activity (P = 0.02) in the BAL fluid (Fig. 3) were all higher in the LPS-NT group (6 h after LPS injury).
|
|
Decreased Progression of ALI in the LPS-HT Group
Compared with the LPS-NT group, the neutrophil count (P = 0.011) (Fig. 1), protein content (P < 0.001) (Fig. 2), and LDH activity (P < 0.001) (Fig. 3) were all lower in the LPS-HT group. Compared with the LPS-NT group, the TNF- level was higher [363 (180–3550) pg/mL, 1330 (757–1980) pg/mL; P < 0.001] and the IL-1ß level was similar [2670 (2190–3210) pg/mL, 2740 (1990–3250) pg/mL; P = 0.928] in the LPS-HT group; MPO tended to be lower in the LPS-HT group [2.97 (2.01–3.66) U/g, 4.36 (2.64–6.08) U/g; P = 0.079]. The basal level of H2O2 was higher in the LPS-HT group (P = 0.015), whereas the intensity of the oxidative burst (increase in H2O2 induced by PMA stimulation) was lower in the LPS-HT group (P = 0.038) (Fig. 4). The stimulation index, or the ratio of H2O2 after PMA stimulation to the basal state, was 9.4 ± 2.3 in the LPS-NT group and 4.0 ± 1.6 in the LPS-HT group (P < 0.001). Representative DCF assays of intracellular H2O2 in the LPS-NT and LPS-HT groups are shown in Figure 5.
|
|
As expected, histology showed that the lungs of the normal rats were free of significant cellular infiltration in the perivascular, interstitial and alveolar spaces. In contrast, the lungs of the LPS-NT rats showed significant alveolar neutrophilic infiltration and perivascular cellular infiltration. The score for neutrophil infiltration (P = 0.001), the score for perivascular cellular infiltration (P = 0.005), and the total ALI score (P < 0.001) were all lower in the LPS-HT group than the corresponding values in the LPS-NT group (Fig. 6a). Appreciable alveolar exudates were not observed in both groups. Figure 6b shows representative microscopic views of rats' lungs from the LPS-NT and LPS-HT groups.
|
Results of analysis of rats' arterial blood obtained during mechanical ventilation prior to thoracotomy were not different between the LPS-NT and LPS-HT groups in pH 7.38 ± 0.08, 7.44 ± 0.06 (P = 0.083), PAco2 29.2 ± 9.3 mm Hg, 23.3 ± 5.8 mm Hg (P = 0.102), PAo2 107.9 ± 15.8 mm Hg, 107.3 ± 19.7 mm Hg (P = 0.936), base excess –7.0 ± 2.0 meq/L, –7.4 ± 2.4 meq/L (P = 0.689), HCO3– 16.4 ± 3.0 meq/L, 15.9 ± 3.0 meq/L (P = 0.632), respectively.
DISCUSSION
In the present study, hypothermic rats showed lower permeability of the lung, less cytotoxic damage to the alveolar epithelium, and fewer inflammatory cellular infiltrates associated with endotoxin than rats under normothermic conditions. Importantly, the effects of HT were obtained at a moderate degree of induced HT in lungs already primed with neutrophilic inflammation.
The hypothermic LPS-treated rats had fewer neutrophils in the airspace, in contrast to the normothermic LPS-treated rats, in which neutrophil sequestration had progressed further from that observed 1 h after LPS treatment. Indeed, neutrophil numbers in the LPS-HT group (6 h after LPS injury) did not differ from those in the LPS-primed group (1 h after LPS injury), suggesting that the sequestration of these key inflammatory cells of ALI was virtually arrested by induced HT. During HT, fewer neutrophils circulate in the blood and their migration into tissue is inhibited through the suppression of the endothelial expression of E-selectin (18) and the neutrophilic expression of ß-integrin (31). Our findings extend these previous observations to an animal model of acute neutrophilic inflammatory disease. In the initial stage of endotoxin-induced ALI, neutrophils are drawn into the airspaces by chemoattractants originating from macrophages. After the initial phase, neutrophils themselves are the principal sources of cytokines that attract other circulating neutrophils, and thus perpetuate neutrophilic inflammation in the lung (8,9). It has been shown repeatedly that neutropenia induced with chemical agents results in the attenuation of the indices of endotoxin-associated ALI (1,4,6,32). Considering that the number of neutrophils in the BAL fluid of the LPS-HT group was similar to that of the LPS-primed group in the present study, induced HT, if applied early, may halt the further sequestration of neutrophils into the airspace.
From a functional aspect, neutrophils of the LPS-HT group, when stimulated with PMA, showed a decrease in ROS compared with those in the LPS-NT group. In endotoxin-induced ALI, ROS not only injure lung tissue directly (33), but also act as a signal for nuclear factor-
B, a pivotal transcriptional molecule for cytokine genes (34–37). It has been shown that induced HT inhibits the production of oxygen radicals by neutrophils (38). To our knowledge, however, this has not been demonstrated in an endotoxin-associated disease model until now. In this regard, our present finding adds to the finding of Chu et al. (21), who showed a therapeutic effect of induced HT in a similar LPS-induced ALI model to ours.
At variance with the neutrophil numbers and histological scores for neutrophilic infiltration, MPO levels, an index of neutrophilic accumulation in tissue, did not differ between the LPS-HT and LPS-NT groups. Because MPO levels were assessed after BAL, this result may indicate that the overall burden of neutrophils in the lung compartments other than the airspace was similar in the two groups.
The histologically detected neutrophil infiltration in the LPS-HT group indicates that ALI was attenuated during HT compared with that under normothermic conditions. Supporting this finding, the airspace protein content was lower in the LPS-HT group than that in the LPS-NT group, which could reflect an attenuation of alveolo-capillary permeability. Furthermore, the lower level of LDH in the BAL fluid of the LPS-HT group suggests that epithelial necrosis was attenuated during HT.
Although the level of TNF- in the BAL fluid returned to baseline in the LPS-NT group, it was persistently elevated in the LPS-HT group. Conceivably, the decrease in the TNF- level in the LPS-NT group was related with the temporal dynamics of cytokines in endotoxin injury, in which the level of TNF- peaks earlier than most other cytokines and then abates rapidly (39). HT itself elicits TNF- release, as has been observed during cardiopulmonary bypass (40). Moreover, in one of our previous studies, induced HT caused an increase in TNF- in normal rats (20). TNF- exerts a multifaceted and time-dependent effect within the endotoxin-induced inflammatory cascade. For example, whereas excess TNF- causes relentless proinflammatory reactions, low concentrations of TNF- are necessary to induce a protective inflammatory response during infection or endotoxemia (41–43). At the present time, it is difficult to determine how the mild elevation in TNF- associated with HT affected the overall progression of ALI. Interestingly, the basal level of ROS was also higher in the LPS-HT group than in the LPS-NT group. The release of ROS from cells is increased by physical stresses, including HT (44). Again, the higher basal level of ROS in the LPS-HT group compared with that in the LPS-NT group, in conjunction with the mildly elevated level of TNF-, suggests that HT itself could induce a low-grade activation of innate immunity.
The present study is limited in several respects. Although the current investigation was performed in a well established animal model of endotoxin-induced ALI, the results cannot be extrapolated to human ALI, especially since the ALI in the present study was relatively mild with regard to histologic changes and oxygenation. In the present study, we did not include any measurement of respiratory mechanics, such as lung compliance, and could not compare them between HT and NT. The benefits of a longer-term application of HT beyond the acute phase of inflammation are not well known, and should be carefully evaluated against possible side effects, such as increased infection and bleeding, as well as the effects of associated interventions, such as sedation and muscle paralysis for the induction and maintenance of HT.
In conclusion, a moderate degree of HT applied even after the lung was primed with neutrophilic inflammation, decreased the progression of ALI compared with that under a NT. These results suggest a therapeutic role for HT, beyond its preventive effect, in ALI.
Footnotes
Accepted for publication January 15, 2007.
Supported by the Asan Institute for Life Sciences (2004-161), and the Korea Research Foundation (KRF-2002-003-E0070), Seoul, Korea.
REFERENCES
in endotoxin-challenged lung tissue. Am J Pathol 1993;143:1009–15.[Abstract], macrophage inflammatory protein-2, interleukin-1 ß, and interleukin-6 but not RANTES or transforming growth factor-ß 1 mRNA expression in acute lung inflammation. Am J Respir Cell Mol Biol 1994;10:148–53.[Abstract] B activation and neutrophilic lung inflammation. J Immunol 1996;157:1630–7.[Abstract] B. J Immunol 2004;172:2522–9.B activation and modulates endotoxin-induced cytokine production in alveolar macrophages. Shock 2005;23:186–93.[ISI][Medline] B activity in lipopolysaccharide-activated human primary monocytes. J Immunol 2005;175:5423–9. in murine salmonellosis. J Immunol 1990;144:1935–41.[Abstract]This article has been cited by other articles:
|
|
 |
|
  C.-M. Lim Hypothermia in Acute Lung Injury: Reduction of Barotrauma or Biotrauma? Anesth. Analg., November 1, 2007; 105(5): 1514 - 1514. [Full Text] [PDF]
|
|
|
|
|
|
N. P. Juffermans Hypothermia in Acute Lung Injury: Reduction of Barotrauma or Biotrauma? Anesth. Analg., November 1, 2007; 105(5): 1513 - 1514. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
