Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Burn injuries are among the most severe types of trauma one can sustain. Nearly 100,000 people each year are admitted to hospitals due to burn injury (1). Approximately 50% of burn patients admitted to hospitals are reported to have been drinking alcohol (2). Numerous studies have indicated that alcohol exposure before thermal injury leads to increased morbidity and mortality (3).

In the past, postburn infection was a major cause of death in thermally injured patients (6). But in the last decade, the use of early excision and grafting protocols and the development of topicals have lessened the incidence of infection as a cause of morbidity and mortality (7, 8). This said, sepsis-related multiple organ failure, specifically pulmonary failure, is a major complication associated with burn injury and is the primary contributor to morbidity and mortality in burned patients (9). This postburn pulmonary failure is characterized by diffuse inflammatory cell infiltration and extracellular fluid accumulation into the lung interstitium. Recent studies have demonstrated that macrophages secrete higher levels of proinflammatory cytokines interleukin (IL)-6 and tumor necrosis factor (TNF)- after thermal injury (10). Additionally, IL-1 was found to be elevated in the lungs of rats after burn injury (11). These cytokines are known to be potent mediators in the onset of inflammation and pulmonary pathology.

Previous work from our laboratory and others revealed that both in vivo and in vitro exposure to alcohol alters the production of these and other cytokines (12, 13). Studies have demonstrated that chronic alcohol exposure adversely affects the immune system by altering gut integrity and neutrophil chemotaxis (14). Additional reports document the effects of ethanol exposure on monocyte/macrophage functions, including phagocytosis (17) and TNF- production (18). Recent studies have further shown that ethanol intoxication is associated with an increased susceptibility to infectious agents due to changes in lymphocyte number and macrophage function (19). This lack of ability to clear infectious agents may cause continuous stimulation of the inflammatory response, which would ultimately result in destruction of the host tissue. These observations suggest that multiple cell types can be adversely affected by injury in combination with alcohol consumption.

Macrophage inflammatory protein (MIP)-2 and KC are members of the  (CXC) chemokine family of inflammatory and immunoregulatory cytokines (20). These murine chemokines have been shown to be functionally homologous to human IL-8. They exhibit potent neutrophil chemotactic activity and are key mediators of neutrophil recruitment in response to tissue injury and infection (21). Additionally, increased MIP-2 and KC levels have been found to be associated with neutrophil influx in various inflammatory conditions (26).

The facts that the number of pulmonary neutrophils is elevated in the lungs of burn mice (31) and that alcohol alters the production of various proinflammatory cytokines suggest that alcohol exposure before thermal injury may exacerbate the postburn impairment in these animals. In this study, we determined that acute ethanol exposure before thermal injury increases pulmonary neutrophil sequestration relative to either insult alone. In addition, MIP-2 production is prolonged in mice subjected to burn injury with prior alcohol treatment as compared with mice given burn injury alone. This would suggest that MIP-2 plays a role in the enhanced pulmonary neutrophil sequestration seen in thermally injured animals with prior alcohol exposure. These studies may provide valuable information regarding acute alcohol-mediated pulmonary pathology in the burn patient. Studies such as these may provide for possible therapeutic interventions for burned individuals in the future.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Alcohol Exposure and Induction of Thermal Injury

These studies were performed using adult (8 to 10 wk of age) male B₆D₂F₁ mice (Jackson Laboratories, Bar Harbor, ME) weighing 25 to 30 g. Mice were randomly assigned to one of four treatment groups: sham + vehicle, sham + ethanol, burn + vehicle, and burn + ethanol, as previously described (32). At 30 min before scald (or sham) injury, animals received an intraperitoneal injection of either 0.4 ml vehicle (saline) or 0.4 ml of a 20% (vol/vol) ethanol solution, for a total dose of 2.4 g/kg. This dose has previously been shown by our lab to deliver a peak circulating ethanol level of 100 to 120 mg/dl at 30 min after injection (32). We have chosen to use this circulating blood ethanol concentration due to its clinical relevance, as this level of ethanol is comparable to the expected human circulating ethanol levels following 1 to 2 hard-liquor drinks (12, 32). This method of ethanol administration resulted in moderate impairments in balance and coordination. At no time after ethanol administration did the mice lose consciousness or appear to be completely inebriated.

At 30 min after ethanol (or vehicle) treatment, all mice were subjected to a dorsal scald (or sham) injury using a method previously described by Walker and Mason (33) as modified by Faunce and colleagues (32). In short, after anesthesia with Nembutal (40 mg/kg intraperitoneally) the dorsum of each mouse was shaved and the animal was placed into a plastic template that exposed 15 to 20% of its total body surface area (TBSA) as calculated according to the method of Spector (34). The mouse and template were then immersed into a 100°C water bath for 8 s. This method achieves a histologically proven, full-thickness scald injury (Faunce and Kovacs, unpublished observation). Sham-treated animals were anesthetized, shaved, and immersed in a room-temperature water bath for 8 s. Immediately following the scald (or sham) injury, all animals received intraperitoneal resuscitation with 1.5 ml of 0.9% normal saline. There was no premature mortality in mice exposed to saline or ethanol, regardless of thermal injury. All animal studies described herein were approved and performed in strict accordance with the guidelines set forth by the Loyola University Chicago Institutional Animal Care and Use Committee.

Tissue Collection

Mice were killed at 1, 2, 4, 8, 12, 24, and 48 h after scald (or sham) injury. These time points were chosen on the basis of work by others examining postburn pulmonary inflammation in this murine model (31). The left upper lobe of the lung was removed, formalin-inflated, and used for histologic evaluation. All other lung lobes were removed, flash-frozen in liquid nitrogen, and stored at 80°C for chemokine enzyme-linked immunosorbent assays (ELISAs). In addition to evaluating lungs from animals of each treatment group, we also took lungs from unmanipulated mice for comparative purposes. Unmanipulated mice were not administered any ethanol, saline, or nembutal, nor were they subjected to a burn or sham injury.

Histologic Evaluation via Hematoxylin and Eosin Staining

A lung lobe was formalin-inflated and -fixed. The tissue was subsequently paraffin-embedded and 5-µm sections were stained with hematoxylin and eosin (H&E). Sections were evaluated using light microscopy. Two nonadjacent sections were examined from each animal. The distinct morphologic nature of the neutrophil, granular cytoplasm with a multilobed nucleus, was used in their identification.

Immunohistochemistry for Murine Neutrophils

Paraffin sections of lung tissue from these same animals were analyzed immunohistochemically using the avidin- biotin complex method (35) modified for paraffin sections (Vector Laboratories, Burlingame, CA). In brief, sections were deparaffinized and incubated in 0.3% H₂O₂ to quench any endogenous peroxidase activity. Sections were then incubated with an unlabeled rat monoclonal antimouse neutrophil primary antibody supplied from Biosource International (Camarillo, CA) (36, 37), followed by a biotinylated secondary (antirat immunoglobulin G) antibody and then a preformed avidin and biotinylated horseradish peroxidase macromolecular complex. The pulmonary neutrophils were localized by incubation with 3-amino-9-ethylcarbazole chromogen substrate for horseradish peroxidase (Vector Laboratories), which produces a red reaction product. Sections were mounted with an aqueous mounting media and viewed with a light microscope at ×250 magnification. NIH Image Analysis software was used to count 10 random fields for positively stained neutrophils from two nonadjacent sections for each animal. The average number of neutrophils from one field was divided by the square area of the field to yield the number of neutrophils per unit area (mm²) as a measure of cellular density. In addition to evaluating lungs from animals of each treatment group, we also assessed lungs from unmanipulated mice for neutrophil content.

Chemokine ELISA

Wet weight of lung samples was determined before homogenization in 1.0 ml of cold protease inhibitor cocktail (Sigma Chemical Co., St. Louis, MO). Single-use aliquots of the homogenates were stored at 80°C before measurement of MIP-2 and KC content by ELISA (R&D Systems, Minneapolis, MN). The enzyme immunoassays were performed according to the manufacturer's specifications. The concentration of chemokine was normalized by the wet weight of the tissue before homogenization to yield the amount of chemokine per milligram of tissue. In addition to evaluating lungs from animals of each treatment group, we assessed lungs from unmanipulated mice for pulmonary MIP-2 and KC levels. The minimum detectable levels of MIP-2 and KC in these immunoassays were found to be typically less than 1.5 and 2.0 pg/ml, respectively (R&D Systems).

Statistics

Data were analyzed by two-way analysis of variance (ANOVA). In the analysis of changes within the treatment groups, Neuman-Keuls post hoc analysis was performed (38). Values are reported as means ± standard error of the mean (SEM). The 95% confidence limit was taken as significant (P < 0.05).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Histologic Assessment of Lung from Scalded Mice

The infrastructural changes in the lung after burn injury with and without prior alcohol exposure are depicted in Figure 1. The 15% TBSA dorsal scald injury resulted in engorgement of the lung interstitial space with neutrophils and emigration of neutrophils into the alveolar space. This increase in neutrophil emigration to the lungs was more prominent in the burn + ethanol-treated mice (Figure 1b) than in the burn + vehicle mice (Figure 1a). Moreover, the burn + ethanol lungs (Figure 1b) showed a greater amount of extravascular leakage into the alveolar space than did the burn + vehicle group (Figure 1a). In addition, there was little or no extravascular leakage in the lungs of sham-treated control animals, comparable to the unmanipulated controls (data not shown).

View larger version (133K): [in this window] [in a new window]   Figure 1.   Histologic assessment of lungs from mice at 24 h postburn. Paraffin sections of mouse lungs were stained for H&E. Representative H&E-stained sections are shown. (a) Burn + vehicle; (b) Burn + ethanol. Note the increase in neutrophils and extracellular fluid accumulation in the burn + ethanol lung as compared with the burn + vehicle lung. Horizontal bar = 10 µm.

Effect of Prior Alcohol Exposure on Pulmonary Edema in Thermally Injured Mice

Lung wet weight was used as a crude assessment of pulmonary edema after the burn injury. The remote dorsal scald injury resulted in extravascular fluid leakage into the lung interstitium (Figure 1) for both burn-treated groups, which led to differences in lung wet weight when compared with the sham-treated controls. This difference in pulmonary edema was most evident at 12 h postburn (Table 1). Regardless of ethanol status, burn-treated mice showed marked increases in lung wet weights as compared with the sham-treated control mice. The burn + vehicle group was significantly higher in lung wet weight than the sham + vehicle (P < 0.01) and sham + ethanol (P < 0.05) groups; whereas the pulmonary edema in the burn + ethanol group was significantly elevated as compared with both sham-treated controls (P < 0.01) and the burn + vehicle (P < 0.05) group. Interestingly, the most significant difference in pulmonary edema corresponded with the same time point postburn at which the highest pulmonary neutrophil sequestration was seen. The lung wet weight for the burn-treated groups was slightly elevated as compared with the sham-treated groups for the earlier time points postburn; however, this increase was not significant. By 48 h after injury, the differences in lung wet weight diminished (data not shown).

Neutrophil Sequestration in the Lung after Burn Injury, with and without Prior Alcohol Exposure

Sections of lung tissue were immunohistochemically stained using a rat antimouse neutrophil antibody. Representative stained sections at 12 h after burn or sham injury are shown in Figure 2. Mice subjected to scald injury (Figures 2c and 2d) showed a marked increase in pulmonary neutrophil sequestration as compared with those subjected to sham injury (Figures 2a and 2b). The ethanol-treated groups, sham + ethanol and burn + ethanol, displayed an even greater number of pulmonary neutrophils than did their respective vehicle controls.

View larger version (133K): [in this window] [in a new window]   Figure 2.   Immunohistochemical evaluation of pulmonary neutrophils at 12 h after burn (or sham) injury, with and without prior ethanol exposure. Paraffin sections of mouse lungs were immunohistochemically stained specifically for neutrophils utilizing a rat antimouse neutrophil antibody. Representative immunohistochemically stained sections are shown. (a) Sham + vehicle; (b) Sham + ethanol; (c) Burn + vehicle; (d) Burn + ethanol. There is a marked increase in pulmonary neutrophils in the burn-treated animals (c and d ) as compared with the sham-treated animals (a and b). Additionally, there is an increase in pulmonary neutrophils in the burn + ethanol group (d ) as compared with the burn + vehicle group (c). Arrows indicate immunohistochemically positive cells. Horizontal bar = 25 µm.

To determine whether neutrophil recruitment to the lung after thermal injury is more pronounced in animals given alcohol before burn injury, pulmonary neutrophil counts were performed on lung tissue at various time points postburn (or sham) injury (Figure 3). Baseline levels were defined as the number of neutrophils in the lungs of unmanipulated control animals. Pulmonary neutrophil counts were slightly above baseline levels (39.5 ± 5.1 neutrophils/mm²) at 1 h postburn for all treatment groups. By 2 h postburn, the pulmonary neutrophil accumulations for the burn-treated groups were significantly higher (P < 0.01) than those in the sham-treated groups. Peak neutrophil emigration to the lung was seen at 8 h postburn for both burn-treated groups (335.1 ± 37.4 neutrophils/mm² for burn + vehicle and 529.2 ± 61.0 neutrophils/mm² for burn + ethanol) and remained elevated at 12 h postburn (351.8 ± 22.5 neutrophils/mm² for burn + vehicle and 525.6 ± 52.4 neutrophils/mm² for burn + ethanol), as they were substantially higher (P < 0.01) than sham-treated controls at both time points. In addition, the burn + ethanol group showed significantly higher (P < 0.01) neutrophil counts than did the burn + vehicle group at 8 and 12 h postburn. Surprisingly, the sham + ethanol group showed significant elevation in pulmonary neutrophil counts at 8 (P < 0.01) and 12 (P < 0.05) h postburn as compared with the sham + vehicle group. Neutrophil counts had declined by 24 h postburn but remained significantly higher (P < 0.01) in the burn-treated groups as compared with the sham-treated groups. By 48 h after burn or sham injury, pulmonary neutrophil counts had returned to baseline levels for all groups.

View larger version (28K): [in this window] [in a new window]   Figure 3.   Neutrophil accumulation in the lung after burn injury. The number of neutrophils in lungs of mice at 1, 2, 8, 12, 24, and 48 h after burn or sham injury, with or without prior ethanol exposure, was determined by counting the distribution of cells in 10 high-powered (×250) fields of formalin-fixed, immunohistochemically stained specimens. Values are expressed as the number of neutrophils per unit area (mm2) ± SEM. n = 6 for all time points except 48 h, where n = 3. *P < 0.01 versus sham + vehicle and sham + ethanol. P < 0.01 versus all other groups as determined by ANOVA and Neuman-Keuls post hoc analysis.

Effect of Thermal Injury with and without Prior Alcohol Exposure on Pulmonary MIP-2 Levels

Given that pulmonary neutrophil sequestration was significantly elevated in the burn + vehicle animals as compared with the sham-treated animals, and that the burn + ethanol group showed an even further increase in pulmonary neutrophil accumulation, we chose to investigate whether neutrophil-specific chemotactic cytokines were differentially produced in the lungs. To accomplish this, lungs from all treatment groups were removed at 1, 2, 4, 8, 12, 24, and 48 h after burn (or sham) injury and assayed for MIP-2 by ELISA. Baseline levels were defined as the amount of chemokine in the lungs of unmanipulated control mice. Pulmonary MIP-2 levels were above baseline levels (0.28 ± 0.05 pg/mg lung) for all groups as early as 1 h after burn or sham injury (Figure 4). The elevation of MIP-2 levels in the sham + vehicle-treated mice at early time points was unexpected and may be a result of anesthesia.

View larger version (25K): [in this window] [in a new window]   Figure 4.   Pulmonary MIP-2 levels after burn or sham injury. Lung tissue was obtained from mice treated with ethanol (or vehicle) at various time points after burn or sham injury. The lower left lobe was homogenized in protease inhibitor cocktail and the homogenate assayed for MIP-2 by ELISA. Values are expressed as amount of MIP-2 in programs per milligram lung tissue ± SEM. n = 6 at each time point except 48 h, where n = 3. *P < 0.05 versus sham + vehicle; P < 0.01 versus sham + vehicle; §P < 0.01 versus all other groups; P < 0.05 versus all other groups; @P < 0.05 versus sham + vehicle and sham + ethanol as determined by ANOVA and Neuman-Keuls post hoc analysis.

Peak levels of MIP-2 were reached at 2 h postburn for both burn + vehicle (5.80 ± 1.43 pg/mg lung)- and burn + ethanol (7.10 ± 2.21 pg/mg lung)-treated groups. This represents a 7-fold increase over baseline levels (0.28 ± 0.05 pg/mg lung) and a significant elevation (P < 0.05) over the sham + vehicle (1.26 ± 0.21 pg/mg lung) group, but not over the sham + ethanol (2.78 ± 0.58 pg/mg lung) group. At 4 h postburn, the pulmonary MIP-2 levels decreased in both burn-treated groups, but the burn + ethanol (5.24 ± 1.32 pg/mg lung) group remained significantly elevated (P < 0.01) as compared with the sham + vehicle group (0.87 ± 0.32 pg/mg lung). Interestingly, the MIP-2 levels for the sham + ethanol (3.04 ± 0.89 pg/mg lung) group remained elevated and comparable to the burn + vehicle (3.08 ± 0.47 pg/mg lung) group. By 8 h after burn or sham injury, the sham-treated controls (0.39 ± 0.03 pg/mg lung for sham + vehicle, 1.12 ± 0.40 pg/mg lung for sham + ethanol) and the burn + vehicle (1.02 ± 0.15 pg/mg lung) group had returned toward baseline levels; but the burn + ethanol (3.70 ± 0.49 pg/mg lung) group remained significantly elevated (P < 0.01) as compared with all other groups. At 12 h postburn, MIP-2 levels for the burn + ethanol (1.78 ± 0.60 pg/mg lung)-treated group had decreased but remained significantly above (P < 0.05) that of the sham-treated control groups (0.36 ± 0.06 pg/mg lung for sham + vehicle, 0.62 ± 0.32 pg/mg lung for sham + ethanol) and the burn + vehicle (0.59 ± 0.13 pg/mg lung) group. MIP-2 levels continued to fall at 24 h postburn for the burn + ethanol (1.35 ± 0.38 pg/mg lung) group and remained significantly higher (P < 0.05) than those of the sham-treated control groups. However, the significance of the elevation in MIP-2 levels at this time point may be a function of the diminished level of the chemokine in the sham-treated groups. Finally, by 48 h after burn or sham injury, MIP-2 levels for all groups had returned to baseline levels.

In parallel, we elected to test for pulmonary KC levels at 8 h postburn. This time point was selected because the differences in MIP-2 levels were most evident with respect to the burn + ethanol group as compared with the other groups. If neutrophil sequestration were triggered by multiple chemokines, then we might expect other neutrophil chemokines to be elevated similarly to MIP-2 in this model. The relative levels of KC at 8 h postburn did not correspond to the levels of MIP-2 at the same time point (Figure 5). The burn + vehicle (21.29 ± 3.05 pg/mg lung), burn + ethanol (28.11 ± 8.18 pg/mg lung), and sham + ethanol (22.79 ± 7.55 pg/mg lung) groups showed comparable levels of pulmonary KC that were significantly higher (P < 0.05) than the sham + vehicle (1.90 ± 0.31 pg/mg lung) group. This would seem to indicate that not all of the neutrophil chemokines in the lung follow the same time course of production after burn injury with prior alcohol exposure. Interestingly, though, the amounts of KC per milligram of lung for the burn-treated groups and the sham + ethanol group were nearly three times higher than the peak levels of MIP-2 in the burn-treated groups. Meanwhile, KC levels for the sham + vehicle group were significantly lower and comparable to that of MIP-2 at this time point.

View larger version (23K): [in this window] [in a new window]   Figure 5.   Pulmonary KC levels 8 h after burn or sham injury. Lung tissue was obtained from mice treated with ethanol (or vehicle) at 8 h after burn or sham injury. The lower left lobe was homogenized in protease inhibitor cocktail and the homogenate assayed for KC by ELISA. Values are expressed as amount of KC in programs per milligram lung tissue ± SEM. n = 6. *P < 0.05 versus sham + vehicle as determined by ANOVA and Neuman- Keuls post hoc analysis.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The incidence of burn injuries in those with alcohol on board are quite high. These patients experience significantly more complications than burn patients not exposed to alcohol; complications include but are not limited to increased hospital stay, more surgical procedures, and greater likelihood of incurring serious infectious complications (5). To date, the study of burn injuries with prior alcohol exposure has not been thoroughly evaluated. Previous studies in our laboratory have shown that postburn effects include a moderate susceptibility to bacterial challenge and slight immune-cell dysfunction. These effects are greatly exacerbated when a low dose (100 mg/dl) of ethanol is given before the burn injury (32); thus suggesting that both inflammatory and immune functions are impaired to a greater extent in burn with acute alcohol exposure than burn alone.

In addition to the immunosuppression seen after burn injury alone, there was an influx of neutrophils into the lungs of these animals. A major complication associated with burn injury is sepsis-related multiple organ failure, primarily pulmonary failure, which appears to be worsened in those with alcohol on board. This important clinical problem has yet to be studied. Herein, we have used a murine dorsal scald injury model in combination with a low-dose, acute ethanol exposure to study the effects on pulmonary neutrophil sequestration. The present study demonstrates that neutrophil sequestration begins within 2 h after remote scald injury, peaks at 8 h, and remains elevated at 12 h postburn. The amount of neutrophil influx in the lungs of burn + ethanol-treated mice was significantly elevated at each of these time points as compared with that of burn + vehicle-treated mice. In addition to markedly higher pulmonary neutrophil counts in burn + ethanol-treated mice, there was a significant increase in pulmonary edema at 12 h postburn in the burn + ethanol group as compared with mice subjected to burn alone. The influx of neutrophils to the lungs after burn injury is preceded by increased levels of pulmonary MIP-2, a neutrophil chemoattractant. Elevated MIP-2 production was evident in mice subjected to burn injury with prior alcohol exposure relative to burn or alcohol alone, suggesting that the chemokine may play a key role in neutrophil emigration to the lungs in this model.

Previous studies have shown that pulmonary neutrophil sequestration occurs after remote insult such as a burn injury or intraperitoneal lipopolysaccharide (LPS) administration (31, 39). The observation that MIP-2 levels increase before neutrophil accumulation is in agreement with other models of inflammation (39, 40, 27). KC, like MIP-2, belongs to the  (CXC) chemokine family and has been shown to exhibit potent neutrophil chemotactic abilities (20). The role of KC in mediating inflammatory responses has been studied more extensively in the development of endotoxin-induced lung injury in rodents than in other systems. Similar to MIP-2, expression of KC messenger RNA is elevated after intratracheal injection of LPS and in alveolar macrophage cultured with LPS (41, 42). Unlike MIP-2, pulmonary KC levels are elevated at 8 h postburn in both burn-treated groups, as well as in the sham + ethanol group. This profile does not resemble that of MIP-2, nor does it parallel the pattern of neutrophil influx. This suggests that pulmonary sequestration of neutrophils after burn injury is not regulated by KC. Additionally, MIP-2 and KC appear to be differentially regulated, which suggests the complexity of cytokine cross-talk involved in mediating pulmonary neutrophil influx observed in this model. Although these studies do not conclusively demonstrate a role for either MIP-2 or KC as a mediator in this model, they do suggest that MIP-2 plays a more significant role. Studies using antichemokine antibodies to neutralize the individual mediators would aid in this determination. Such studies are beyond the scope of the present investigation.

The acute inflammatory response involves the removal of invading organisms through phagocytic clearance that occurs upon recruitment of neutrophils (43, 44). However, the normal inflammatory response may be as destructive as it is defensive. The release of lysosomal products by neutrophils can damage the local tissues through proteolysis by enzymes such as elastase and collagenase (45, 46). Neutrophil elastase is a major culprit in the pathogenesis of tissue destruction in the airway. This enzyme has been found to strip the bronchial epithelium, reduce ciliary beating, and stimulate excess mucus secretion, leading to mucus retention, bacterial proliferation, and recurrent infections. Neutrophil elastase also stimulates IL-8 secretion from epithelial cells, which, in turn, leads to further neutrophil recruitment (47). Free oxygen radicals are also released by neutrophils that destroy tissue (48). In addition, some of these compounds increase the vascular permeability (45). This is supported by our findings, in which increased pulmonary edema is evident at 12 h after injury in the burn + vehicle group and is more exaggerated when ethanol is present at the time of injury. Elevated neutrophil levels were seen as early as 8 h postburn and remained elevated at 12 h postburn. At both time points, even greater increases in pulmonary neutrophil counts were observed in the burn + ethanol-treated group than following either insult alone. This would suggest that in burn + vehicle- and burn + ethanol-treated mice, pulmonary neutrophils may be mediating the extravascular leakage via the mechanisms discussed previously.

Ethanol exposure can alter the inflammatory and immune functions after burn injury through several mechanisms. Ethanol administration has been shown to affect immune cells directly to inhibit chemotaxis and proliferation and to modulate the production of some cytokines (12, 15, 32, 49). Recent studies have further shown that ethanol intoxication suppresses pulmonary inflammation (50, 51) and impairs neutrophil chemotaxis, thereby exposing the host to increased subsequent infection (52, 53). This is consistent with our earlier observation that burn + ethanol-treated mice are less likely to survive a Pseudomonas aeruginosa infection than are burn + vehicle-treated mice (32). In addition, it is of interest to note that the studies showing depressed neutrophil chemotaxis after ethanol administration were performed in vitro and the same effects were not seen in vivo. This would explain the apparent contradiction of elevated pulmonary neutrophils in ethanol-treated animals as compared with their vehicle-treated counterparts. Alternatively, it may suggest that the pulmonary neutrophil counts that we observed were attenuated due to the prior alcohol exposure and, in response to chemokine signaling, we should have seen even greater neutrophil sequestration. Further, deficits in neutrophil chemotaxis that may have been caused by ethanol administration could have led to the prolonged production of pulmonary MIP-2. If neutrophil chemokines were being produced but neutrophil chemotaxis was not occurring, then one might suspect that the elevated production of chemokine would continue; whereas normally, upon neutrophil emigration to a site of concentrated chemokine levels, the production of the chemokine would be downregulated. Alternatively, the ethanol-treated animals remained under the effects of anesthesia longer than did the vehicle-treated controls. The sedated animals may have had a diminished ability to clear mucus or debris from their airways, which may have led to the elevated pulmonary MIP-2 levels and subsequently higher number of pulmonary neutrophils as also seen in the sham + ethanol group relative to the sham + vehicle group.

Ethanol exposure may also impart indirect effects in this model. It has been shown that glucocorticoids, which are known to be potent immunosuppressive agents, are increased in the circulation following stresslike conditions such as burn injury and ethanol exposure (54). Glucocorticoids have been shown to affect pulmonary neutrophil influx and inflammatory mediators differentially (55). These studies report that neutrophil emigration to the lungs after intratracheal LPS treatment is attenuated in rats pretreated with dexamethasone, a synthetic glucocorticoid. Interestingly, dexamethasone pretreatment resulted in a significant reduction of TNF- levels in bronchoalveolar lavage fluid, whereas MIP-2 levels remained unchanged. We have measured circulating corticosterone levels in this model (58). These studies showed elevated levels of circulating corticosterone at 8 h and 1 d after burn injury in the burn + vehicle group, but these levels were significantly depressed in the burn + ethanol group. This suggests that endogenous corticosterone may play a protective role by suppressing the production of key inflammatory mediators and limiting the number of neutrophils accumulating in the lungs of burn + vehicle-treated mice. Such a marked increase in circulating corticosterone is not observed in the burn + ethanol-treated mice; thus, cytokine levels are not attenuated. Additional reports document ethanol's suppression of endotoxin-induced TNF- production in the lung (18). Studies measuring the levels of TNF- in the lungs of our burn + ethanol-treated mice are currently in progress. Changes in TNF- levels may be attributed to differences in corticosterone levels and prior ethanol treatment. This further adds to the complex nature of cytokine cross-regulation in animals subjected to burn injury with prior alcohol exposure.

Because we know that TNF- upregulates MIP-2, but glucocorticoids and ethanol downregulate TNF-, it is possible that other mediators also play a role in the pulmonary cytokine cascade of burn mice with alcohol on board. We anticipate that these studies will provide valuable information regarding acute alcohol-mediated pulmonary pathology. Studies such as these may provide for possible therapeutic interventions for burn patients in the future.