Abstract
Although prior studies have shown that smoke inhalation causes lung endothelial injury and formation of pulmonary edema, there is no information about the effect of smoke inhalation on the function of the alveolar epithelial barrier. Therefore, the primary objective of this study was to determine the effect of smoke-induced lung injury on the alveolar epithelial barrier in a rabbit experimental model. The second objective was to investigate whether pretreatment with a monoclonal anti–interleukin (IL)-8 antibody prevented alveolar epithelial barrier injury after smoke inhalation. Anesthetized rabbits were tracheotomized and were insufflated with cooled smoke generated from burning cotton cloth (75 breaths). In some experiments, anti–IL-8 antibody or an irrelevant antibody (2 mg/ kg) was given intravenously 5 min before insufflation of cotton smoke. Smoke inhalation caused a significant increase in the alveolar epithelial permeability to protein and a 40% reduction in the fluid transport capacity of the alveolar epithelium. Pretreatment with anti–IL-8 antibody, but not with an irrelevant-isotype antibody, significantly reduced the smoke-mediated increase in bidirectional transport of protein across the alveolar epithelium, and restored alveolar liquid clearance to a normal level. The results of the study show that smoke inhalation causes injury to both the alveolar epithelial barrier and the lung endothelium, and that IL-8 is an important mediator of this injury. Laffon M, Pittet J-F, Modelska K, Matthay MA, Young DM. Interleukin-8 mediates injury from smoke inhalation to both the lung endothelial and the alveolar epithelial barriers in rabbits.
Smoke inhalation is an important cause of acute lung injury (ALI) in humans and is associated with a high mortality rate (1). The effect of smoke inhalation on the pulmonary endothelial barrier has been studied extensively. In animal models, smoke inhalation causes a significant increase in lung lymph flow, extravascular lung water content, and pulmonary vascular permeability to protein (2). These physiologic alterations in the pulmonary microvasculature occur from 4 to 6 h after smoke inhalation (2). In contrast, the effect of smoke inhalation on the lung epithelium has been less well studied. Early after smoke exposure, bronchial blood flow increases and edema forms in the airways (3). Twenty-four hours after smoke inhalation there is morphologic damage to the alveolar epithelium, such as denudation of basal lamina and swelling and rupture of alveolar epithelial type I cells (4). However, there is little direct information about the ability of the alveolar epithelium to actively transport electrolytes and water from the air spaces to the insterstitium of the lung in the early fluid resuscitation phase after smoke-induced ALI.
Activated neutrophils play an important role in the inflammatory process caused by smoke exposure. Depletion of neutrophils with nitrogen mustard significantly decreased the smoke-induced increase in lung microvascular permeability to protein (5). Production of nitric oxide (NO)-derived oxidants as a result of the activation of neutrophils contributes to the oxidative injury to the lung that follows smoke inhalation (6). In addition, nebulized dimethylsulfoxide (7) or administration of a synthetic antiprotease (8) reduced lung injury caused by smoke inhalation. More recently, selectin blockade caused by the intravenous administration of Sulfo Lewis C attenuated lung injury associated with smoke exposure, supporting the hypothesis that neutrophils play a pivotal role in smoke inhalation-induced lung injury (9).
Since interleukin (IL)-8 is a major chemotactic factor for neutrophils, we hypothesized that IL-8 plays an important role in recruiting neutrophils into the distal air spaces of the lung after smoke inhalation, and that neutralization of IL-8 will not only prevent smoke-induced lung endothelial injury but will also preserve the function of the alveolar epithelial barrier after smoke inhalation. The first objective of our study was therefore to determine the effect of smoke inhalation on the barrier function of the alveolar epithelium. Since the results of this investigation indicated that there was a significant decrease in net alveolar fluid clearance associated with a significant increase in bidirectional protein permeability across the alveolar epithelium at 6 h after smoke inhalation, the second objective of our study was to determine whether neutralization of IL-8 would improve alveolar epithelial barrier function after smoke-induced lung injury.
New Zealand White rabbits (weight range: 2.5 to 3.5 kg) were surgically prepared as described earlier (10). The protocol for the study was approved by the University of California San Francisco Animal Research Committee. Briefly, the rabbits were anesthetized with 0.8% halothane in 100% oxygen. Pancuronium bromide (0.3 mg/h × kg body weight) (Organon Diagnostics, West Orange, NJ) was given intravenously for neuromuscular blockade.
A 22-gauge intravenous catheter was inserted in the marginal ear vein for administering fluid and drugs. A right carotid arterial line (PE-90; Clay Adams, Becton Dickinson, Parsippany, NJ) was inserted for continuous monitoring of blood pressure and to obtain blood samples. An endotracheal tube (4.0 mm I.D.; Portex) was inserted through a tracheotomy. The rabbits were maintained in the supine position and ventilated with a constant-volume pump (Harvard Apparatus, South Natick, MA) and an inspired oxygen fraction (Fi O2 ) of 1.0 and with a peak airway pressure of 13 to 17 cm H2O, associated with a positive end-expiratory pressure (PEEP) of 4 cm H2O. During the baseline period, the respiratory rate (RR) was adjusted to maintain PaCO2 at 35 to 40 mm Hg. Thereafter, the ventilator settings were kept constant throughout the experiment.
A 5% rabbit albumin solution was prepared with the use of Ringer's lactate, and was adjusted with NaCl to be isoosmolar with the rabbit's circulating plasma, as previously described (11). Anhydrous Evans blue dye (0.5 mg; Sigma Chemical, St. Louis, MO) was added to the albumin solution to confirm the location of the instillate at the end of the study, and 3 μCi of 125I-labeled human serum albumin (Frosst Laboratories, PQ, Canada) was also added to the albumin solution. The 125I-albumin served as the alveolar protein tracer in all experiments. A sample of the instilled solution was saved for total protein measurement, radioactivity counts, and water-to-dry-weight ratio measurements, so that the dry weight of the protein solution could be subtracted from the final lung water calculation.
The generation of the monoclonal antibody (mAb) to rabbit recombinant IL-8 (rIL-8) (ARIL8.2) that was used in our study has been described in detail (12, 13). ARIL8.2 was selected by virtue of its ability to recognize rabbit IL-8, to inhibit binding of 125I-labeled rabbit rIL-8 to its receptor, to block rabbit rIL-8–induced signal transduction via its receptor, and to inhibit rabbit rIL-8–induced chemotactic activity for rabbit neutrophils (13). ARIL8.2 had a high affinity for rabbit IL-8 (K d = 0.42 nM). ARIL8.2 did cross-react with human IL-8, but not with closely related cytokines (platelet factor-4, β-thromboglobulin), other human cytokines (IL-1β, tumor necrosis factor-α), or other chemotactic factors (formyl methionyl leucyl phenylalanine, C5a). The antibody preparation was filtered and was found to contain no endotoxin with the Limulus assay.
In all experiments, heart rate (HR) and systemic blood pressure were allowed to stabilize for 60 min after surgery (Figure 1). The animals were then ventilated with air for 30 min, and the level of anesthesia was deepened by increasing the concentration of halothane to 2% to insure loss of the cough reflex before administration of smoke. Following this, inhalation injury was induced with a modified bee smoker (Ghorybee, Eugene, OR) (14). The modified bee smoker was filled with 50 g of burning cotton cloth and was connected to the tracheostomy tube with a connector containing a thermistor, to monitor the temperature of the smoke. During the insufflation procedure, the temperature of the smoke did not exceed 39° C. The rabbits were insufflated with 75 breaths (20 ml/kg) of smoke. Blood carboxyhemoglobin (COHb) was measured immediately at the end of the smoke inhalation with a co-oximeter, and the animals were ventilated with an Fi O2 of 1.0. For the next 360 min of the experiment, the rabbits received 3 ml/kg/h of Ringer's lactate solution given intravenously.
A vascular tracer, 1 μCi of 131I-labeled human albumin, was injected into each rabbit's blood 270 min after the end of smoke inhalation, in order to calculate the flux of plasma protein into the alveolar and extravascular spaces of the lung, as previously described (15). The rabbit was then placed in a right lateral decubitus position in order to facilitate liquid instillation into the right lung. Thirty minutes after this, an alveolar tracer, 1 μCi of 125I-albumin in 3 ml/kg of the 5% rabbit albumin solution, was instilled into the right lower lobe to calculate the flux of protein from the air spaces into the circulating plasma. This tracer was instilled over a period of 20 min by using a 12-ml syringe and pediatric feeding tube.
At the end of the experiment (1 h after the beginning of the alveolar instillation), the abdomen was opened and the rabbit was exsanguinated by transecting the abdominal aorta. Urine was obtained for radioactivity counts. The lungs were removed through a median sternotomy. An alveolar fluid sample from the distal air spaces was obtained by gently passing the Silastic tubing used for tracer administration into a wedged position in the instilled area of the right lower lobe. After centrifugation, the total protein concentration and radioactivity of the alveolar liquid sample were measured. Right and left lungs were homogenized separately for water-to-dry weight ratio measurements, and radioactivity counts was measured.
The animals were allocated to four experimental groups. Three of these groups were insufflated with smoke.
Group 1. Positive control group. To determine the effect of smoke inhalation on alveolar epithelial barrier function, anesthetized rabbits (n = 13) were insufflated with cooled smoke from cotton toweling. The rabbits were then studied for 6 h.
Group 2. Negative control group. Rabbits (n = 7) underwent the same surgical preparation as did those in Group I but were insufflated with air instead of smoke. The rabbits were then studied for 6 h.
Group 3: Pretreatment with anti-IL-8 antibody. To determine whether pretreatment with mAb to rabbit IL-8 would prevent injury to the alveolar epithelium caused by smoke inhalation, ARIL8.2, a monoclonal antibody to rabbit IL-8 (2 mg/kg) was administered intravenously to rabbits (n = 5) 5 min before the onset of smoke inhalation. The rabbits were then studied for 6 h.
Group 4: Pretreatment with an isotype-irrelevant antibody. To demonstrate that the protective effect provided by pretreatment with mAb to rabbit IL-8 was caused by the inhibition of IL-8, an isotype- irrelevant antibody (2 mg/kg) was administered intravenously to rabbits (n = 3), 5 min before the onset of smoke inhalation. The rabbits were then studied for 6 h.
Hemodynamics, pulmonary gas exchange, and protein concentration. The animals' HR, systemic blood pressure, and airway pressures were continuously measured with calibrated pressure transducers (Pd23; internal diameter; Gould, Inc., Oxnard, CA) and recorded continuously with a polygraph (Model 7; Grass Instrument Co., Quincy, MA). Arterial blood gases and pH were measured at 1-h intervals. Samples of alveolar fluid from the instillation procedure, of the final distal air space fluid, and of the blood taken at the beginning and at the end of the study were collected to measure total protein concentration with an automated analyzer (AA2; Technicon, Tarrytown, NY).
Protein concentration and hemoglobin measurement. Protein concentration was measured with the biuret method. Hemoglobin was measured spectrophotometrically on the last blood samples and on the supernatant obtained after centrifugation of lung homogenate (14,000 × g for 10 min).
Alveolar liquid clearance. Alveolar liquid clearance (ALC) (percent clearance from the alveolar space of the volume of liquid instilled) was measured from the increase in the final unlabeled alveolar protein concentration as compared with the initial alveolar protein concentration, as in earlier studies (16). ALC was calculated as: ALC = (Vi × Fwi − Vf × Fwf)/(Vi × Fwi) × 100, where Fw is the water fraction of the initial (i) and the final (f) alveolar fluid. The water fraction is the volume of water per volume of solution, as measured with the gravimetric method. Vi is the volume of the initial (i) and Vf of the final (f) alveolar fluid. Vf (ml) was estimated as: Vf = (Vi × Tpi × Fr)/ TPf, where Tpi is the total protein concentration of the initial (i) and Tpf of final (f) alveolar fluid. Fr is the fraction of alveolar tracer (125I-albumin) protein that remains in the lung at the end of the experiment.
The concentration of alveolar 125I-albumin was also used to estimate alveolar liquid clearance. The Vf was then estimated as: Vf = (Vi × 125cpmi × Fr)/125cpmf, where 125cpm is the cpm per milliliter of initial (i) and cpmf of the final (f) alveolar fluid. We adopted this method in our studies to provide an additional means of calculating alveolar liquid clearance. Because we knew that some edema fluid was present in the alveoli in rabbits 6 h after smoke instillation, we used the dilution of the instilled 125I-albumin solution at 5 min after the beginning of the instillation to calculate alveolar liquid clearance, as we have done before (16). Therefore, we collected a sample from the distal air spaces at 5 min after the beginning of the instillation of the 5% bovine albumin solution containing 1 μCi of 125I-albumin. The initial alveolar 125I-albumin concentration was diluted (92% of the instilled concentration) in rabbits that were insufflated with smoke (see Results). Therefore, alveolar liquid clearance was calculated from the increase in the 125I-albumin concentration over a period of 1 h, with the assumption that the initial labeled albumin concentration was 92% of the instilled concentration. Because there was no change in the total unlabeled protein concentration (probably because protein-rich edema fluid was present in the alveoli), no adjustment was needed to calculate alveolar liquid clearance with the unlabeled protein method.
Albumin flux across endothelial and epithelial barriers. Two different methods were used to measure the flux of albumin across the lung endothelial and epithelial barriers, as previously done (16). The first method measures residual 125I-albumin (the air space protein tracer) in the lungs, as well as accumulation of 125I-albumin in plasma. The second method measures 131I-albumin (the vascular protein tracer) in the alveolar and extravascular spaces of the lungs.
The total quantity of 125I-albumin (the air space protein tracer) instilled into the lung was determined by measuring duplicate samples of the instilled solution for total radioactivity counts (cpm/g) and multiplying this value by the total volume instilled into the lung. To calculate the residual 125I-albumin in the lungs at the end of the study, the average radioactivity counts of two 0.5-g samples obtained from the lung homogenate was multiplied by the total weight of lung homogenate. The counts for albumin in the lung homogenate were added to the counts in the final aspirated distal air space fluid to calculate the quantity of instilled 125I-albumin that remained in the lungs at the end of the study. The 125I-albumin in the circulating plasma was measured from a sample of plasma obtained at the end of the experiment. The 125I-albumin in the aspirates was measured from a sample obtained 5 min after the start of alveolar fluid instillation and a second such sample obtained at the end of the experiment. The plasma fraction of 125I-albumin was calculated for by multiplying the counts/g plasma by the plasma volume (body weight × 0.07 [1-hematocrit]).
The second method for measuring albumin flux requires measurement of the vascular protein tracer 131I-albumin, in the alveolar and extravascular spaces of the lungs. To calculate the amount of 131I-albumin present in the extravascular spaces of the lung, we deducted the counts for the blood in the lung from the 131I-albumin counts in the entire lung. The clearance of plasma into the extravascular spaces of the lung was estimated with the following equation: Extravascular plasma equivalents = 131cpm lung − (131cpm Plf × Qb)/131cpm PL, where 131cpm lung is the total cpm in lung, 131cpm Plf is the cpm per milliliter of plasma in the final blood sample, 131cpm PL is the cpm per milliliter plasma averaged over 90 min (time from the intravenous injection of 1 μCi of 131I-albumin to the end of the study), and Qb is the blood volume in the lung, which is calculated as: Qb = 1.039 × (Qh × Fwh × Hbs)/(FWs × HbB), where 1.039 is the density of blood, Qh the weight of lung homogenate, Fwh the water content of lung homogenate, Hbs the hemoglobin concentration of the supernatant of lung homogenate, Fws the water content in supernatant of lung homogenate, and HbB the hemoglobin concentration of blood.
To express the 131I-albumin counts in the air spaces as a ratio to the plasma counts, 131I-albumin counts were measured in the final alveolar fluid sample, and 131I-albumin plasma counts were averaged over the course of the experiment. This ratio provides an index of equilibration of the vascular protein tracer into the alveolar compartment, as shown in earlier experimental studies of epithelial permeability (10).
Tracer binding measurement. To determine 125I and 131I binding to albumin, trichloracetic acid (20%) was added to all tubes, which were then centrifuged to obtain the supernatant for measurement of free 125I and 131I radioactivity. The results are expressed as the unbound 125I and 131I radioactivity as a percentage of the total amount of 125I- or 131I-albumin radioactivity instilled. These fluid samples always contained less than 1% unbound iodine.
All data are summarized as mean ± SE. One-way analysis of variance (ANOVA) with repeated measurements analysis was used to compare samples obtained at several time points from the same animal. One-way ANOVA and Scheffe's exact t test were used to compare experimental with control groups. A value of p < 0.05 was considered statistically significant.
The arterial COHb levels just after smoke exposure were 76.2 ± 2.3% in rabbits insufflated with smoke (positive control group), 74.5 ± 2.3% in rabbits insufflated with smoke but pretreated with an mAb directed against rabbit IL-8, and 73.2 ± 2.5% in rabbits insufflated with smoke but pretreated with an isotype-irrelevant antibody. These values were not statistically different from each other, reflecting the consistency of exposure to smoke in each group.
Smoke exposure caused a significant increase in the bidirectional alveolar epithelial permeability to protein (Table 1). There was a significant increase in protein flux (alveolar protein tracer, 125I-albumin) from the air spaces to the plasma in smoke-insufflated as compared with air-insufflated rabbits (Table 1). In addition, there was a significant increase in protein flux (vascular protein tracer, 131I-albumin) from the plasma to the air spaces, as indicated by an increase in the alveolar-to-plasma–concentration ratio of 131I-albumin, in smoke-insufflated as compared with air-insufflated rabbits (Table 1).
| Experimental Condition | n | Alveolar Protein Tracer 125I-albumin (% of instilled ) | 131I-albumin Alveolar/Plasma Ratio | |||||
|---|---|---|---|---|---|---|---|---|
| Lung | Plasma | |||||||
| Air (negative controls) | 7 | 99.6 ± 0.3 | 0.4 ± 0.3 | 0.02 ± 0.01 | ||||
| Smoke (positive controls) | 13 | 97.0 ± 0.5* | 3.0 ± 0.5* | 0.19 ± 0.03* | ||||
| Smoke + anti–IL-8 antibody | 5 | 98.4 ± 0.6† | 1.5 ± 0.6† | 0.12 ± 0.03† | ||||
| Smoke + irrelevant antibody | 3 | 95.9 ± 2.3* | 3.8 ± 0.4* | 0.26 ± 0.02* | ||||
Five minutes after the beginning of alveolar fluid instillation, alveolar fluid samples were obtained to measure the dilution of the instilled solution by the alveolar edema that had formed 5 h after smoke insufflation. In these samples the concentration of 125I-albumin were 8 ± 1% lower then initial concentrations of 125I-albumin in the instilled solution. Thus, the labeled 125I-albumin in these samples was used as the initial concentration.
There was a significant decrease in the final to initial unlabeled protein concentration ratio in the distal air spaces in smoke-insufflated as compared with air-insufflated rabbits (Table 2), corresponding to a 40% decrease in alveolar liquid clearance in smoke-insufflated versus air-insufflated rabbits (p < 0.05) (Figure 2). In addition, there was a significant decrease in the final to initial labeled protein concentration ratio (125I-albumin) in the distal air spaces in smoke-insufflated as compared with air-insufflated rabbits (1.14 ± 0.02 versus 1.28 ± 0.03, p < 0.05), corresponding also to a decrease in alveolar liquid clearance comparable with that measured from the unlabeled protein concentration (11 ± 2% versus 21 ± 2%, p < 0.05).
| Experimental Conditions | n | Ratio of Final versus Initial Alveolar Protein Concentration | ||
|---|---|---|---|---|
| Air (negative controls) | 7 | 1.31 ± 0.02 | ||
| Smoke (positive controls) | 13 | 1.17 ± 0.03* | ||
| Smoke + anti–IL-8 antibody | 5 | 1.26 ± 0.04† | ||
| Smoke + irrelevant antibody | 3 | 1.15 ± 0.01* |
Fig. 2. Alveolar liquid clearance (% of instilled liquid) (mean ± SEM) is shown for rabbits insufflated with air or smoke, and for rabbits insufflated with smoke and pretreated with an mAb to rabbit IL-8 or with an isotype-irrelevant antibody. Smoke inhalation caused a significant decrease in alveolar liquid clearance. Pretreatment with mAb to rabbit IL-8, but not with isotype-irrelevant antibody, prevented the decrease in alveolar liquid clearance caused by smoke inhalation (*p < 0.05 from rabbits insufflated with air).
[More] [Minimize]Smoke exposure was associated with a significant increase in the endothelial permeability to protein, as indicated by a significant increase in the accumulation of extravascular plasma equivalents in noninstilled lung in smoke-insufflated as compared with air-insufflated rabbits (p < 0.05) (Figure 3). In addition, the water-to-dry weight ratio of the noninstilled lung was significantly greater in rabbits exposed to smoke than in rabbits insufflated with air (p < 0.05) (Figure 4). This increase in the lung endothelial permeability to fluid and protein as a consequence of smoke exposure was associated with a significant decrease in the final PaO2 /Fi O2 ratio of rabbits exposed to smoke as compared with that of rabbits insufflated with air (Table 3).
Fig. 3. Extravascular plasma equivalents of the noninstilled lung (mean ± SEM) are shown for rabbits insufflated with air or smoke, and for rabbits insufflated with smoke and pretreated with an mAb to rabbit IL-8 or with an isotype-irrelevant antibody. Smoke inhalation caused a significant increase in extravascular plasma equivalents of the noninstilled lung. Pretreatment with mAb to rabbit IL-8, but not with isotype-irrelevant antibody, prevented the increase in extravascular plasma equivalents of the noninstilled lung caused by smoke inhalation (*p < 0.05 from rabbits insufflated with air).
[More] [Minimize]
Fig. 4. Extravascular water content of the noninstilled lung (g water/g dry lung tissue) (mean ± SEM) is shown for rabbits insufflated with air or smoke, and for rabbits insufflated with smoke and pretreated with an mAb to rabbit IL-8 or with an isotype-irrelevant antibody. Smoke inhalation caused a significant increase in extravascular water content of the noninstilled lung. Pretreatment with mAb to rabbit IL-8, but not with isotype-irrelevant antibody, prevented the increase in extravascular water content of the noninstilled lung caused by smoke inhalation (*p < 0.05 from rabbits insufflated with air).
[More] [Minimize]| Experimental Conditions | PaO2 /Fi O2 Ratio | Arterial pH | Mean Arterial Pressure | |||
|---|---|---|---|---|---|---|
| Air (negative controls) (n = 7) | ||||||
| Baseline | 384 ± 33 | 7.40 ± 0.03 | 72 ± 3 | |||
| 6 h after insufflation of air | 356 ± 29 | 7.38 ± 0.03 | 69 ± 3 | |||
| Smoke (positive controls) (n = 13) | ||||||
| Baseline | 390 ± 31 | 7.41 ± 0.02 | 71 ± 2 | |||
| 6 h after insufflation of smoke | 229 ± 41* | 7.39 ± 0.02 | 72 ± 3 | |||
| Smoke + anti–IL-8 antibody (n = 5) | ||||||
| Baseline | 380 ± 30 | 7.40 ± 0.04 | 70 ± 5 | |||
| 6 h after insufflation of smoke | 363 ± 20† | 7.37 ± 0.03 | 73 ± 3 | |||
| Smoke + irrelevant antibody (n = 3) | ||||||
| Baseline | 385 ± 36 | 7.41 ± 0.04 | 72 ± 3 | |||
| 6 h after insufflation of smoke | 160 ± 56* | 7.37 ± 0.05 | 67 ± 5 |
Pretreatment with a mAb against rabbit IL-8, but not with an isotype-irrelevant antibody, significantly reduced the bidirectional increase in protein permeability across the alveolar epithelial barrier caused by smoke exposure. There was a significant decrease in the percentage of instilled 125I-albumin measured in plasma in smoke-insufflated rabbits pretreated with mAb against rabbit IL-8 as compared with the values measured in rabbits insufflated with smoke (positive controls) and in rabbits insufflated with smoke and pretreated with isotype-irrelevant antibody (Table 1). Similarly, there was a significant reduction in the protein flux (vascular protein tracer, 131I-albumin) from the plasma to the air spaces, as indicated by a significant decrease in the alveolar-to-plasma–concentration ratio of 131I-albumin, in smoke-insufflated rabbits pretreated with mAb against rabbit IL-8 as compared with the two other smoke-insufflated rabbit groups (Table 1). Also, pretreatment with mAb against rabbit IL-8, but not with isotype-irrelevant antibody, prevented most of the decrease in alveolar fluid clearance observed in smoke-insufflated rabbits (p < 0.05) (Figure 2). These results corresponded to a final to initial unlabeled protein concentration ratio in the distal air space in smoke-insufflated rabbits pretreated with mAb against rabbit IL-8 that did not differ from the values in rabbits insufflated with air (Table 2). There was also no significant change in the final to initial labeled protein concentration ratio (125I-albumin) in the distal airspace in smoke-insufflated rabbits pretreated with mAb against rabbit IL-8 from that in air-insufflated rabbits (1.25 ± 0.03 versus 1.28 ± 0.03, p < 0.05), corresponding to similar values of alveolar liquid clearance in the two rabbit groups (21 ± 2% versus 20 ± 1%, p = NS). In contrast, the value of alveolar liquid clearance measured with labeled albumin in smoke-insufflated rabbits pretreated with isotype-irrelevant antibody was significantly lower than that measured in air-insufflated rabbits (12 ± 2% versus 21 ± 2%, p < 0.05).
Pretreatment with mAb against rabbit IL-8, but not with isotype-irrelevant antibody, also a caused significant reduction in lung endothelial permeability to fluid and protein. The accumulation of extravascular plasma equivalents was attenuated in anti–IL-8 antibody-pretreated versus nonpretreated rabbits (p < 0.05) (Figure 3). The extravascular lung water was also normal in the anti–IL-8 antibody-pretreated rabbits (p < 0.05) (Figure 4). Pretreatment with mAb against rabbit IL-8 was also associated with a normalization of thePaO2 /Fi O2 ratio as compared with the values measured in nonpretreated smoke-insufflated rabbits (Table 3).
No differences among experimental groups were observed in systemic blood pressure or HR (data not shown), PaCO2 (data not shown), or arterial pH at any time during the study (Table 3).
The overall objective of our study was to determine the mechanisms that regulate fluid transport across the alveolar epithelium after smoke inhalation, an important cause of ALI in humans, and one that has been associated with a high mortality rate (1). Because preservation of the capacity of the alveolar epithelium to remove fluid from the air spaces is critical for the survival of patients with ALI (17), we developed an experimental animal model to investigate the effect of smoke inhalation on the function of the alveolar epithelium. The increase in lung vascular permeability caused by smoke inhalation has been well documented (2). In contrast, the effect of smoke inhalation on the lung epithelium has been less well studied. Twenty-four hours after smoke inhalation, there is morphologic damage to the alveolar epithelium, such as denudation of basal lamina and swelling and rupture of alveolar epithelial type I cells (4). However, there is little direct information about the capacity of the alveolar epithelium to actively transport electrolytes and water from the air spaces to the insterstitium of the lung in the early fluid resuscitation phase after smoke-induced ALI.
Therefore, the first objective of our study was to determine the effect of smoke inhalation on the barrier function of the alveolar epithelium. The results indicate that smoke inhalation caused a moderate but significant increase in the bidirectional alveolar epithelial and lung endothelial permeability to protein. There was also a 40% decrease in alveolar liquid clearance at 6 h after smoke inhalation as compared with the value measured in air-insufflated rabbits.
Because the first objective of our study was to examine net alveolar epithelial fluid clearance after smoke inhalation, we used two approaches to be sure that the calculation of alveolar fluid clearance was accurate in the presence of an increase in protein permeability of the alveolar epithelium. The first method we used to estimate alveolar liquid clearance was to measure the increase in the unlabeled alveolar protein concentration over a period of 1 h after inhalation. The second method we used to estimate alveolar liquid clearance was to measure the increase in the alveolar 125I-albumin concentration 1 h after instillation of a solution of this radiolabeled protein. Because we found in pilot studies that some edema fluid was present in the alveoli of rabbits at 6 h after smoke inhalation, we used the dilution of the instilled 125I-albumin solution at 5 min after the beginning of the instillation to estimate the initial labeled albumin concentration. In these samples, the concentration of 125I-albumin was 8 ± 1% below the concentration of 125I-albumin in the instilled solution. We therefore calculated alveolar liquid clearance from the increase in the 125I-albumin concentration over the 1-h period with the assumption that the initial labeled albumin was 92% of the instilled concentration. Interestingly, the unlabeled protein concentration sampled at 5 min after instillation did not differ from the values measured in the instilled solution, probably because some protein-rich edema fluid was already present in the alveoli. In addition, estimation of alveolar liquid clearance was done over a short period (1 h) to minimize the magnitude of the bidirectional protein flux across the injured alveolar epithelium. Using these different approaches, we found a comparable decrease in net alveolar liquid clearance in smoke- versus air-insufflated rabbits.
Did smoke insufflation directly affect the ability of the alveolar epithelium to remove fluid from the air spaces, or was the decrease in alveolar liquid clearance a result of smoke- induced lung endothelial injury? Several lines of evidence suggest that smoke insufflation could directly affect the active removal of fluid from the air spaces by the alveolar epithelium. First, an increase in lung vascular permeability is not necessarily associated with a decrease in the ability of the alveolar epithelium to remove fluid from the air spaces. We have previously reported that alveolar liquid clearance may be upregulated by catecholamine-dependent (11, 18) or -independent (16) mechanisms in the presence of an increase in lung vascular permeability. For example, alveolar liquid clearance was upregulated in sheep that received a 4-h continuous intravenous infusion of Pseudomonas aeruginosa, despite a 4-fold increase in extravascular lung plasma equivalents (31 ml versus 8 ml)—an increase comparable with that found in the present study. Second, we have reported that elimination of pulmonary blood flow in rabbits instilled with acid did not affect the acid-mediated decrease in alveolar liquid clearance despite the absence of fluid accumulation in the interstitial space of the lung (19). Third, other investigators have reported that neutrophil activation and production of NO-derived oxidants contribute to the lung and plasma indices of oxidative injury in an experimental model of smoke inhalation in rats (6). Because we have recently found that neutrophil activation and NO- derived oxidants are responsible for a significant decrease in alveolar epithelial fluid transport in the early phase after prolonged hemorrhagic shock and fluid resuscitation in rats (20), it is possible that these reactive nitrogen species can affect active removal of fluid from the air spaces after smoke inhalation.
The second objective of our study was to determine whether neutralization of IL-8 would improve alveolar epithelial barrier function after smoke-induced lung injury. Because we have previously found that IL-8 plays an important role in recruiting neutrophils to the lung from the airspaces in other models of ALI, such as acid aspiration (12), we hypothesized that IL-8 plays an important role in recruiting neutrophils into the distal air spaces of the lung after smoke inhalation, and that neutralization of IL-8 would not only prevent smoke- induced lung endothelial injury, but would also preserve the barrier function of the alveolar epithelium in this pathologic condition. The results of the present study indicate that pretreatment with an mAb against rabbit IL-8, but not with an isotype-irrelevant antibody, prevents the decrease in alveolar fluid clearance and significantly reduces the bidirectional increase in protein permeability of the lung endothelial and alveolar epithelial barriers caused by smoke exposure.
Several lines of evidence confirm the hypothesis that lung endothelial and alveolar epithelial injury caused by smoke exposure are primarily mediated by neutrophils recruited to the lung by IL-8. First, activation of neutrophils plays a crucial role in the development of pulmonary edema following smoke inhalation (4, 6, 9). Second, depletion of neutrophils with nitrogen mustard significantly decreases the smoke-induced increase in lung microvascular permeability to protein (5). Third, the administration of an mAb to rabbit IL-8 resembling the antibody used in the present study (ARIL8.2) reduced neutrophil influx and lung endothelial injury after acid aspiration (12), and endotoxin-mediated influx of neutrophils into the pleural space (13). Fourth, in two experimental studies with rats and rabbits respectively, antibodies to human IL-8 reduced neutrophil influx and lung injury caused by IgG immune complexes (21) and reperfusion after ischemia (22). Fifth, our present data indicate that pretreatment with an mAb to rabbit IL-8, but not with an isotype irrelevant antibody, decreased lung endothelial and alveolar epithelial injury secondary to smoke inhalation, and normalized oxygenation. Also, we found comparable results in a companion study in which an mAb to rabbit IL-8 significantly decreased the alveolar epithelial injury caused by acid aspiration (19). In particular, pretreatment with anti–IL-8 antibody reduced alveolar epithelial injury caused by acid aspiration even in the absence of pulmonary blood flow, indicating a direct protective effect of neutralizing IL-8 on the function of the alveolar epithelial barrier. Moreover, a recent in vitro study confirmed that cigarette smoke induces the release of IL-8 from human bronchial epithelial cells (23).
In summary, the present data provide the first in vivo evidence for a significant reduction in alveolar epithelial fluid transport with a moderate alteration in epithelial paracellular permeability to protein after smoke inhalation. Pretreatment with an mAb to rabbit IL-8 prevented both lung endothelial and alveolar epithelial injury caused by smoke exposure, providing the first evidence for the critical role of IL-8 in mediating ALI after smoke inhalation.
The authors appreciate the assistance of Dr. Caroline Hebert and Genentech, Inc. for supplying both the monoclonal anti–IL-8 antibody and the isotype irrelevant antibody for this study.
Supported by grant HL 51854 from the National Heart, Lung, and Blood Institute, a research grant from the American Lung Association ( J.F.P.), and a REAC grant from the University of California, San Francisco (D.M.Y.).
| 1. | Ryan C. M., Schoenfield D. A., Thorpe W. P., Sheridan R. L., Cassem E. H., Thompkins R. G.Objective estimates of the probability of death from burn injuries. N. Engl. J. Med.3381998362366 |
| 2. | Isago T., Noshima S., Taber L. D., Herndon D. N., Taber D. L.Analysis of pulmonary microvascular permeability after smoke inhalation. J. Appl. Physiol.17199114031408 |
| 3. | Kramer G., Herndon D. N., Linares H. A., Traber D. L.Effects of inhalation injury on airway blood flow and edema formation. J. Burn Care Rehabil.1019894551 |
| 4. | Guha S. C., Herndon D. N., Evans M. J., Schmalstieg F. C., Isago T., Traber L. D., Linares H. A., Traber D. L.Is the CD18 adhesion complex of polymorphonuclear leukocytes involved in smoke-induced lung damage? A morphometric study. J. Burn Care Rehabil.141993503511 |
| 5. | Basadre J. O., Sugi K., Traber D. L., Traber L. D., Niehaus G. D., Herdon D. N.The effect of leukocyte depletion on smoke inhalation injury in sheep. Surgery1041988208215 |
| 6. | Ischiropoulos H., Mendiguren I., Fisher D., Fisher A. B., Thom S. R.Role of neutrophils and nitric oxide in lung alveolar injury from smoke inhalation. Am. J. Respir. Crit. Care Med.1501994337341 |
| 7. | Kimura R., Mlcak R., Richardson J., Desai M., Herdon D. N., Traber L. D., Traber D. L.Treatment of smoke-induced pulmonary injury with nebulized dimethylsulfoxide. Circ. Shock251988333341 |
| 8. | Niehaus G. D., Kimura R., Traber L. D., Herndon D. N., Flynn J. T., Traber D. L.Administration of a synthetic antiprotease reduces smoke-induced lung injury. J. Appl. Physiol.691990694699 |
| 9. | Tasaki O., Mozingo D. W., Ishihara S., Brinkley W. W., Johnson A. A., Smith R. H., Srivastava O., Mason A. D., Pruitt B. A., Cioffi W. G.Effect of Sulfo Lewis C on smoke inhalation injury in an ovine model. Crit. Care Med.26199812381243 |
| 10. | Smedira N., Gates L., Hastings R., Jayr C., Sakuma T., Pittet J. F., Matthay M. A.Alveolar and lung liquid clearance in anesthetized rabbits. J. Appl. Physiol.70199118271835 |
| 11. | Pittet J. F., Wiener-Kronish J. P., McElroy M. C., Folkesson H. G., Matthay M. A.Stimulation of alveolar epithelial liquid clearance by endogenous release of catecholamines in septic shock. J. Clin. Invest.941994663671 |
| 12. | Folkesson H. G., Matthay M. A., Hebert C. A., Broaddus V. C.Acid aspiration-induced lung injury in rabbits is mediated by interleukin-8-dependent mechanisms. J. Clin. Invest.961995107116 |
| 13. | Broaddus V. C., Boylan A. M., Hoeffel J. M., Kim K. J., Sadick M., Chuntharapai A., Hebert C. A.Neutralization of interleukin-8 inhibits neutrophil influx in a rabbit model of endotoxin-induced pleurisy. J. Immunol.152199429602967 |
| 14. | Kimura R., Traber L. D., Herndon D. N., Linares H. A., Lubbesmeyer H. J., Traber D. L.Increasing duration of smoke exposure induces more severe lung injury in sheep. J. Appl. Physiol.64198811071113 |
| 15. | Modelska K., Matthay M. A., McElroy M. C., Pittet J. F.Upregulation of alveolar liquid clearance after fluid resuscitation for hemorrhagic shock in rats. Am. J. Physiol.2731997L305L314 |
| 16. | Rezaiguia S., Garat C., Delclaux C., Meignan M., Fleury J., Legrand P., Matthay M. A., Jayr C.Acute bacterial pneumonia in rats increases alveolar epithelial fluid clearance by a tumor necrosis factor-alpha-dependent mechanism. J. Clin. Invest.991997325335 |
| 17. | Matthay M. A., Wiener-Kronish J. P.Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am. Rev. Respir. Dis.142199012501257 |
| 18. | Pittet J. F., Wiener-Kronish J. P., Serikov V. S., Matthay M. A.Resistance of the alveolar epithelium to injury from septic shock in sheep. Am. J. Respir. Crit. Care Med.151199510931100 |
| 19. | Modelska, K., J. F. Pittet, H. G. Folkesson, V. C. Broaddus, and M. A. Matthay. 1999. Acid-induced lung injury: protective effect of anti- interleukin-8 treatment on alveolar epithelial barrier function in rabbits. Am. J. Respir. Crit. Care Med. (In press) |
| 20. | Modelska K., Matthay M. A., Pittet J. F.Inhibition of inducible NO sunthase activity after prolonged hemorrhagic shock attenuates oxidant-mediated decrease in alveolar epithelial fluid transport in rats (abstract). FASEB J.121998A839 |
| 21. | Mulligan M. S., Jones M. L., Bolanowski M. A., Baganoff M. P., Deppeler C. L., Meyers D. M., Ryan U. S., Ward P. A.Inhibition of lung inflammatory reactions in rats by an anti-human IL-8 antibody. J. Immunol.150199355855595 |
| 22. | Sekeido N., Mukaida N., Harada A., Nakanishi I., Watanabe Y., Matsushima K.Prevention of lung reperfusion injury in rabbits by a monoclonal antibody against interleukin-8. Nature3651993654657 |
| 23. | Mio T., Romberger D. J., Thompson A. B., Robbons R. A., Heires A., Rennard S. I.Cigarette smoke induces interleukin-8 release from human bronchial epithelial cells. Am. J. Respir. Crit. Care Med.155199717701776 |
