Acute lung injury and the acute respiratory distress syndrome (ARDS) are significant causes of morbidity and mortality following sepsis and hemorrhage. Increased IL-1β production in the lung is important in the development of acute inflammatory lung injury. Although neutrophils are an important component of the inflammatory response that characterizes acute lung injury, there is little information to suggest that they are capable of initiating cytokine-mediated immune responses in the lung. To explore the role of neutrophils in the early stages of acute lung injury, we examined IL-1β production by mouse lung neutrophils after hemorrhage and endotoxemia. There was a significant increase in IL-1β expression among intraparenchymal pulmonary neutrophil/mononuclear cells (IPNMC) 1 h after hemorrhage or endotoxemia. IL-1β was detected only in a neutrophil-rich fraction of the IPNMC, but not in T and B lymphocytes positively selected from the IPNMC. Cyclophosphamide (CTX)-treated neutropenic mice expressed significantly less IL-1β in IPNMC after hemorrhage or endotoxemia compared with CTX-untreated controls. Immunohistochemical analysis of lung sections from mice after hemorrhage or endotoxemia revealed IL-1β expression in infiltrating neutrophils. These data indicate that IL-1β-producing neutrophils traffic to the lungs rapidly in response to hemorrhage or endotoxemia and support the concept that proinflammatory cytokine production by lung neutrophils may contribute to the development of lung injury after blood loss and sepsis.

Acute lung injury and the acute respiratory distress syndrome (ARDS)3 continue to be significant causes of morbidity and mortality following trauma, hemorrhage, and severe infection. The expression of proinflammatory cytokines, such as IL-1β, increases rapidly after hemorrhage, a model of blood loss and trauma, or endotoxin administration, a model of sepsis (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11). Clinically, enhanced proinflammatory cytokine expression, including IL-1β and TNF-α, is seen in the lungs of patients with ARDS (12, 13, 14, 15). These observations suggest that cytokines play a prominent role in the inflammatory responses that lead to acute lung injury and ARDS.

IL-1β is thought to be a major participant in the pulmonary inflammatory cascade in ARDS. IL-1β, rather than TNF-α, was found to be the major proinflammatory bioactive substance in the bronchoalveolar lavage fluid (BALF) of patients with ARDS (16). Intratracheal instillation of IL-1 produces acute inflammatory injury in the lungs (5, 17). In vitro, IL-1 induces IL-8 expression in alveolar macrophages, thus indirectly increasing neutrophil chemotaxis into the lung (18). IL-1 also appears to occupy an important role in sepsis-induced organ dysfunction and lung injury (2, 5, 15). IL-1-converting enzyme-deficient mice that cannot produce mature IL-1β are resistant to endotoxic shock (19). Additionally, IL-1 has multiple immunoregulatory and systemic effects that may contribute to the development of inflammatory injury in the lungs and other organ systems. Some of these effects include increased expression of adhesion molecules, such as ICAM-1; induction of lymphocyte NF-κB activity; increased fibroblast proliferation; and systemic effects, including hypotension and lactic acidosis (16, 20, 21, 22, 23).

The neutrophil is often considered to be the final effector cell responsible for lung injury, due to its ability to express multiple cytotoxic products, including reactive oxygen metabolites and proteinases. However, recent evidence suggests that the neutrophil may be a more active participant during the process of inflammation. In vitro data show that neutrophils can express IL-1Ra, IL-1β, IL-8, IL-10, TGF-β1, and TNF-α (3, 24). Human neutrophils produce IL-8 (25), and, in endotoxemic mice, neutrophils express TNF-α, IL-10, and IL-1α (3, 24). However, despite the potential for neutrophils to contribute to an acute inflammatory response through the expression of proinflammatory and immunoregulatory cytokines and chemokines, there is little in vivo information to suggest that neutrophils produce such cytokines in the initial stages of an inflammatory response or that such neutrophil-derived cytokines can initiate or potentiate inflammatory responses in the lung.

Neutrophils rapidly enter the pulmonary parenchyma after endotoxin administration (4, 5), ischemia/reperfusion (26), and hypovolemic shock (27). These neutrophils appear in advance of discernible lung injury, which occurs after 48 to 72 h in such models. Induced neutropenia followed by endotoxin challenge (4, 28) or complement activation (29) attenuates increases in lung vascular permeability and lung injury. These data suggest that there may be two distinct phases of the neutrophil response in acute lung injury: an early phase characterized by rapid neutrophil influx and a later phase characterized by neutrophilic alveolitis, interstitial infiltration, and a developing inflammatory response.

Because neutrophils are capable of expressing proinflammatory cytokines, such as IL-1β, we hypothesized that rapid trafficking of cytokine-producing neutrophils to the lungs could be important in initiating acute inflammatory lung injury. To investigate this issue, we examined IL-1β production among intraparenchymal lung cell populations after hemorrhage and endotoxemia.

Esherichia coli 0111:B4 endotoxin was obtained from Sigma Chemical Company (St. Louis, MO). Polyclonal rabbit anti-mouse IL-1β and control rabbit anti-mouse antiserum were obtained from PharMingen (San Diego, CA). Biotinylated goat anti-rabbit Abs were obtained from Southern Biotechnology Associates (Birmingham, AL). Streptavidin-horseradish peroxidase was obtained from Amersham Biologics (Arlington Heights, IL). Methoxyflurane was obtained from Pitman-Moore, Inc. (Mundelein, IL). The Bio-Rad DC protein assay was obtained from Bio-Rad (Hercules, CA). Percoll was obtained from Pharmacia (Uppsala, Sweden). Ab-coated magnetic beads were obtained from Dynal Corporation (New York, NY). Chemiluminescent detection on Western blots was performed using the Ultra Supersignal Detection system (Pierce, Rockford, IL). Immunohistochemistry was performed using the Vectrastain Quick Kit (Vector, Burlingame, CA). Low melting point agarose was obtained from SeaKem (FMC Bioproducts, Rockland, ME). Glucose oxidase was obtained from Boehringer Manheim (Indianapolis, IN). Gill’s hematoxylin was obtained from Fisher (Springfield, NJ). Male BALB/c mice, 8- to 12-wk-old, were obtained from Harlan Biologics (Indianapolis, IN).

All experiments were performed according to a protocol approved by the University of Colorado Health Sciences Institutional Animal Care Use Committee. A murine hemorrhage model was used as previously described by our laboratory (11, 30, 31). Briefly, male BALB/c mice, aged 8 to 12 wk, were used for this procedure. Mice were anesthetized, followed by cardiac puncture and removal of 30% of the calculated blood volume (∼0.55 ml for a 20-g mouse). Control mice were anesthetized and subjected to cardiac puncture without hemorrhage. The mortality rate with this hemorrhage protocol is ∼12% (32). Hemothorax, bleeding into the pericardial space, and lung or cardiac contusions do not occur in surviving mice (11). When blood pressure is monitored with femoral artery catheters in this model, removal of 30% blood volume decreases mean arterial blood pressure to ∼40 mm Hg, with restoration to normal levels over the 60 min following hemorrhage (6).

In those experiments in which the effects of endotoxin administration were examined, i.p. injections of 25 mg/kg E. coli 0111:B4 endotoxin in 0.2 ml PBS were used. This dose has been described as causing lung injury 72 h after administration (2). Control mice were given i.p. injections of 0.2 ml PBS.

IPNMC were obtained as previously described by our laboratory using collagenase/DNase digestion and Percoll gradient purification (30, 32, 33). Of note, collagenase, DNase, and Percoll are tested routinely for endotoxin in our laboratory and are used only if endotoxin levels of <1 U/ml are detected.

Neutrophil enrichment was performed according to the method of Sugawara et al. using a discontinuous Percoll gradient (34). Briefly, pulmonary parenchymal cells suspended in PBS and 0.1% BSA were layered over Percoll gradients with densities of 1.097 and 1.085. Isolated cells were then collected from the Percoll gradient interface, washed with PBS, and lysed in SDS sample buffer containing PMSF. Protein content was determined using the Bio-Rad DC protein assay. Aliquots of cells were stained and visualized using either modified Wright’s staining or in situ myeloperoxidase (MPO) staining and were found to be >90% neutrophils. The remaining cells were mononuclear cells.

In those experiments in which lymphocytes were isolated, anti-B220- and anti-Thy 1.1-coated magnetic beads were added, in succession, to the IPNMC using the manufacturer’s protocol (Dynal Corporation). Magnetic beads with attached lymphocytes were separated using a magnet. Cells were lysed in sample buffer and centrifuged after boiling to make postnuclear preparations and remove residual beads. Negatively selected cells were likewise lysed and postnuclear preparations were used for electrophoresis. These negatively selected cells were found to be 98% neutrophils as determined by Wright’s staining.

Mice were injected i.p. with 150 mg/kg CTX in 0.2 ml sterile water 4 days and 1 day before use in experiments, according to the protocol of Proietti et al. (35). Control mice were treated i.p. with 0.2 ml sterile water at the same time points. Whole blood smears were prepared at the time of hemorrhage and analyzed by Wright’s staining. Red cell-depleted samples of blood were used to measure the total number of white blood cells.

MPO was measured using a modification of the method described by Goldblum et al. (36). Briefly, lungs isolated from mice were washed, blotted dry, and frozen in liquid nitrogen. After weighing, frozen lungs were homogenized, centrifuged at 20,000 × g for 30 min, and resuspended in 50 mM potassium phosphate buffer, pH 6.0, with 0.5% hexadecyltrimethylammonium bromide. Samples were sonicated, incubated at 60°C for 2 h, and assayed for activity in a hydrogen peroxide/o-dianisidine buffer at 460 nm. Results are expressed as units of MPO activity per gram of lung tissue.

Cell samples were lysed in SDS sample buffer containing PMSF and boiled for 10 min. Cytoplasmic extracts were prepared by centrifugation. Protein content was measured using the Bio-Rad DC protein assay. β-mercaptoethanol was then added, and the samples were boiled again and run on 12% polyacrylamide gels in the amounts noted, following the method of Laemmli (37). Gels were blotted onto nitrocellulose and immunoblotted using anti-IL-1β Ab. Horseradish peroxidase second-step reagents were then used and blots were developed using the Pierce Ultra chemiluminescent system. Blots were finally exposed to Kodak X-OMAT AR film and developed.

Immunohistochemistry was performed as described previously (38, 39). Briefly, unmanipulated control mice, hemorrhaged mice, or endotoxin-treated mice were prepared as described above. After the right ventricle was perfused with 5 cc PBS (4°C), the lungs were gently infiltrated with 1% low melting point agarose (Seakem) at 42°C through the trachea. Lungs were then removed en bloc, and fixed in a 4% paraformaldehyde, 0.23 M sucrose solution overnight. Tissue was then embedded and 5-μm sections prepared as described previously (39). Briefly, embedded sections were treated with 0.2 M glucose, 1.5 U/ml glucose oxidase in PBS for 30 min followed by 10% hydrogen peroxide in PBS for 15 min. Immunohistochemistry was conducted using either anti-IL-1β or normal rabbit serum at a dilution of 1:1000 using the Vectrastain immunohistochemistry kit following the manufacturer’s protocol (Vector). Sections were then developed using a diaminobenzidine/peroxide-based development system (Vector) followed by counterstaining with Gill’s hematoxylin (Fisher).

All experimental groups consisted of three mice or more (where noted). Data are presented as the mean ± SEM. Densitometry was performed using an Ultraviolet Products (Cambridge, U.K.) charge-coupled device camera and image capture software together with Ultraviolet Products Gelblot image analysis software. Significance was determined using GraphPad Instat software (GraphPad Software, San Diego, CA) using either two tailed t testing or ANOVA analysis, as appropriate. A p value of <0.05 was considered significant.

To determine the cellular composition in the pulmonary parenchyma after hemorrhage or endotoxemia, mice were subjected to either anesthesia and cardiac puncture with hemorrhage of 30% blood volume, anesthesia and cardiac puncture without blood removal (i.e., sham hemorrhage), i.p. endotoxin administration, or i.p. PBS administration (i.e., sham endotoxin treatment), and killed 1 h later. There was a significant increase in neutrophil infiltration into the lungs 1 h after hemorrhage or endotoxin administration as measured by MPO content (Fig. 1). These findings were also evident by Wright’s staining, in situ MPO staining, and flow cytometric analysis of IPNMC.

FIGURE 1.

MPO content in lungs after hemorrhage or endotoxemia. Either normal BALB/c mice or BALB/c mice made neutropenic using CTX as described in Materials and Methods were used. Mice were anesthetized and subjected to cardiac puncture without hemorrhage (Control), cardiac puncture with hemorrhage of 30% blood volume (Hemorrhage), or treated with 25 mg/kg E. coli endotoxin i.p. (Endotoxin). After 1 h, lungs were isolated and MPO assays performed as described in Materials and Methods. Results represent the mean ± SEM of MPO activity per gram of lung tissue. *, Denotes p < 0.001 between experimental groups and control mice.

FIGURE 1.

MPO content in lungs after hemorrhage or endotoxemia. Either normal BALB/c mice or BALB/c mice made neutropenic using CTX as described in Materials and Methods were used. Mice were anesthetized and subjected to cardiac puncture without hemorrhage (Control), cardiac puncture with hemorrhage of 30% blood volume (Hemorrhage), or treated with 25 mg/kg E. coli endotoxin i.p. (Endotoxin). After 1 h, lungs were isolated and MPO assays performed as described in Materials and Methods. Results represent the mean ± SEM of MPO activity per gram of lung tissue. *, Denotes p < 0.001 between experimental groups and control mice.

Close modal

We sought to determine whether hemorrhage or endotoxemia induce expression of IL-1β in IPNMC. One hour after hemorrhage or endotoxemia, IPNMC were isolated and cytoplasmic extracts were immunoblotted for IL-1β.

There was a rapid increase in pro-IL-1β expression in IPNMC 1 h after hemorrhage or endotoxemia compared with control mice (Fig. 2). Because mature IL-1β is predominantly extracellular, only the precursor form of IL-1β, pro-IL-1β, was detected in isolated cells, although some processed pro-IL-1β consistent with IL-1β can be seen at m.w.s consistent with the various cleavage sites found on pro-IL-1β (Fig. 2) (40).

FIGURE 2.

Western blot analysis of IPNMC using anti-IL-1β Ab in control, hemorrhaged, or endotoxemic mice ± CTX treatment. IPNMC from control mice, mice hemorrhaged 30% of their blood volume 1 h previously (Hemorrhage), mice made neutropenic with CTX and then subjected to hemorrhage 1 h previously (CTX Hemorrhage), mice treated with 25 mg/kg endotoxin i.p. 1 h previously (Endotoxin), or mice made neutropenic with CTX and then treated with 25 mg/kg endotoxin i.p. 1 h previously (CTX Endotoxin) were isolated. Cytoplasmic preparations were made, 10 μg of protein was run on 12% polyacrylamide gels, and immunoblotting was performed using anti-IL-1β Ab. Note band of mature 17-kDa IL-1β in the first lane of the Endotoxin group. Approximate m.w.s are represented on the right of the figure.

FIGURE 2.

Western blot analysis of IPNMC using anti-IL-1β Ab in control, hemorrhaged, or endotoxemic mice ± CTX treatment. IPNMC from control mice, mice hemorrhaged 30% of their blood volume 1 h previously (Hemorrhage), mice made neutropenic with CTX and then subjected to hemorrhage 1 h previously (CTX Hemorrhage), mice treated with 25 mg/kg endotoxin i.p. 1 h previously (Endotoxin), or mice made neutropenic with CTX and then treated with 25 mg/kg endotoxin i.p. 1 h previously (CTX Endotoxin) were isolated. Cytoplasmic preparations were made, 10 μg of protein was run on 12% polyacrylamide gels, and immunoblotting was performed using anti-IL-1β Ab. Note band of mature 17-kDa IL-1β in the first lane of the Endotoxin group. Approximate m.w.s are represented on the right of the figure.

Close modal

To determine whether lymphocytes were the source of IL-1β production in the lung after hemorrhage, IPNMC were subjected to positive and negative selection using Ab-coated magnetic beads. IPNMC isolated from sham-hemorrhaged mice and mice hemorrhaged 1 h previously were incubated with anti-B220-coated magnetic beads to isolate B cells. T cells were then isolated by incubating the B cell-depleted preparation with anti-Thy-1-coated magnetic beads. The remaining nonadherent cells, depleted of T and B cells (Non-B, Non-T) were then collected. The latter cells were found to be 98% neutrophils according to Wright’s staining. Equivalent amounts of protein from purified T and B cells, as well as the remaining negatively selected cell population, were loaded onto polyacrylamide gels and the resulting immunoblots were probed for IL-1β.

Pro-IL-1β was only detected after hemorrhage in the lymphocyte-depleted lung cell population and not detected in B cell- or T cell- enriched samples (Fig. 3,A). Sham-hemorrhaged mice had no detectable pro-IL-1β in any sample (Fig. 3 A). Because macrophages and platelets are effectively excluded from the IPNMC by the Percoll gradient isolation step, these results suggest that either neutrophils or monocytes were producing the observed IL-1β. Furthermore, because monocytes are relatively few in number in these samples (<2%), these results indicate that neutrophils are the main source of IL-1β among the IPNMC following hemorrhage.

FIGURE 3.

Western blot analysis using anti-IL-1β Ab of T lymphocyte, B lymphocyte, or neutrophil-enriched fractions of the IPNMC obtained 1 h after hemorrhage or endotoxemia. A, IPNMC from three mice were pooled after either sham hemorrhage (C) or hemorrhage 1 h previously (H). Cells were then separated using anti-B220-coated magnetic beads (B cells) followed by anti-Thy-1-coated magnetic beads (T cells). Nonadherent cells, depleted of T and B cells, (98% neutrophils by Wright’s staining, Non-B and Non-T), were also isolated. Cytoplasmic preparations were made, 10 μg of protein was run on 12% polyacrylamide gels, and immunoblotting was performed using anti-IL-1β Ab. B, Mice (n = 3 in each group) were hemorrhaged 30% of their blood volume (H) or treated i.p. with 25 mg/kg endotoxin (E). One hour later, either total IPNMC (H-IPNMC, E-IPNMC) or the neutrophil enriched fraction of the IPNMC (H-PMN, E-PMN, 91% neutrophils) were isolated. IPNMC from control (sham endotoxin treated) mice were also isolated (C-IPNMC). Cytoplasmic preparations were made and immunoblotting was performed using anti-IL-1β Ab. Of note, 20 μg of total IPNMC and 2 μg of the neutrophil-enriched fraction of the IPNMC were loaded onto the gels.

FIGURE 3.

Western blot analysis using anti-IL-1β Ab of T lymphocyte, B lymphocyte, or neutrophil-enriched fractions of the IPNMC obtained 1 h after hemorrhage or endotoxemia. A, IPNMC from three mice were pooled after either sham hemorrhage (C) or hemorrhage 1 h previously (H). Cells were then separated using anti-B220-coated magnetic beads (B cells) followed by anti-Thy-1-coated magnetic beads (T cells). Nonadherent cells, depleted of T and B cells, (98% neutrophils by Wright’s staining, Non-B and Non-T), were also isolated. Cytoplasmic preparations were made, 10 μg of protein was run on 12% polyacrylamide gels, and immunoblotting was performed using anti-IL-1β Ab. B, Mice (n = 3 in each group) were hemorrhaged 30% of their blood volume (H) or treated i.p. with 25 mg/kg endotoxin (E). One hour later, either total IPNMC (H-IPNMC, E-IPNMC) or the neutrophil enriched fraction of the IPNMC (H-PMN, E-PMN, 91% neutrophils) were isolated. IPNMC from control (sham endotoxin treated) mice were also isolated (C-IPNMC). Cytoplasmic preparations were made and immunoblotting was performed using anti-IL-1β Ab. Of note, 20 μg of total IPNMC and 2 μg of the neutrophil-enriched fraction of the IPNMC were loaded onto the gels.

Close modal

To examine further the role of neutrophils in producing IL-1β in the lungs after endotoxemia and blood loss, we purified neutrophils from the IPNMC. To do so, pulmonary parenchymal cells were isolated and then subjected to discontinuous Percoll gradient density isolation using the method of Sugawara et al. (34). The resulting preparation was found to be >90% neutrophils according to Wright’s staining, with the remaining cells consisting of mononuclear cells. Neutrophil-enriched fractions from mice hemorrhaged or treated with endotoxin 1 h previously were lysed and subjected to immunoblotting for IL-1β.

Pro-IL-1β is detected in the neutrophil-enriched fraction of intraparenchymal lung cells from hemorrhaged or endotoxemic mice (Fig. 3,B). Because the neutrophil enrichment procedure results in the isolation of significantly less protein, neutrophil-enriched fractions consist of 2 μg protein/lane, while the total IPNMC contain 20 μg protein/lane. Twenty micrograms of total IPNMC are required to yield a detectable amount of IL-1β under the conditions used. Of note, the numbers of neutrophils among the IPNMC are too low in normal unmanipulated mice to permit isolation of enough of these cells to perform Western blots (Fig. 1).

We then determined whether IPNMC from neutropenic mice produced as much IL-1β as those from normal mice. To this end, mice were treated with CTX. By 4 days after treatment, there was a decrease in peripheral neutrophils from 9.5 ± 1.5 × 105 polymorphonuclear leukocytes (PMN)/ml at baseline to 1.8 ± 0.4 × 104 PMN/ml (p = 0.04). Concomitantly, there was a decrease in neutrophils in the pulmonary bed as assessed by MPO assays. CTX-treated mice had very little detectable MPO activity, even after hemorrhage or endotoxemia (Fig. 1). Control or CTX-treated mice were either hemorrhaged or treated with endotoxin, and the resulting IPNMC was isolated 1 h later. Lysates were probed for the presence of IL-1β. There is a significant decrease of pro-IL-1β expression in the IPNMC of CTX-treated mice compared with that seen in non-neutropenic control mice subjected to either hemorrhage or endotoxin administration (Fig. 2).

To localize IL-1β and determine the relative importance of neutrophils in the production of IL-1β after hemorrhage or endotoxemia, we performed immunohistochemical studies of IL-1β protein in the lung after hemorrhage and endotoxemia. The occasional neutrophils seen in the lungs of control mice expressed no detectable IL-1β (Fig. 4,A). By contrast, after either hemorrhage (Fig. 4,B) or endotoxemia (Fig. 4,D), IL-1β was detected in the neutrophils infiltrating the lung parenchyma. The predominant populations of neutrophils expressing IL-1β were those seen in the pulmonary vasculature and those migrating into the lung through the endothelium. Endothelial cells were also noted to express IL-1β (Fig. 4, B and D). Serial sections stained with control normal rabbit serum as the primary Ab showed no significant staining of any cell types (Fig. 4, C and E). Alveolar macrophages were not clearly positive for IL-1β staining in any sample.

FIGURE 4.

IL-1β is localized to neutrophils and endothelium in the lung after hemorrhage or endotoxemia. Mice were either sham hemorrhaged (A), hemorrhaged 30% of their blood volume (B and C), or treated with 25 mg/kg endotoxin i.p. (D and E). One hour later, lungs were infiltrated with 1% agarose, removed en bloc, sectioned, stained with either rabbit anti-IL-1β Ab (A, B, and D) or normal rabbit serum (C and E) followed by a diamidobenzidine/peroxide solution, and counterstained with Gill’s hematoxylin solution. Photographs represent ×1000 magnifications of the sections. Arrows represent neutrophils, with the open arrow indicating an intercalating neutrophil, and the arrowhead indicating the vascular endothelium.

FIGURE 4.

IL-1β is localized to neutrophils and endothelium in the lung after hemorrhage or endotoxemia. Mice were either sham hemorrhaged (A), hemorrhaged 30% of their blood volume (B and C), or treated with 25 mg/kg endotoxin i.p. (D and E). One hour later, lungs were infiltrated with 1% agarose, removed en bloc, sectioned, stained with either rabbit anti-IL-1β Ab (A, B, and D) or normal rabbit serum (C and E) followed by a diamidobenzidine/peroxide solution, and counterstained with Gill’s hematoxylin solution. Photographs represent ×1000 magnifications of the sections. Arrows represent neutrophils, with the open arrow indicating an intercalating neutrophil, and the arrowhead indicating the vascular endothelium.

Close modal

The present studies show that neutrophils that traffic to the lungs are a major source of IL-1β in the lungs after hemorrhage and endotoxemia. In particular, we found that IL-1β was detected in enriched pulmonary neutrophil populations isolated 1 h after hemorrhage or endotoxemia. Positive selection of lymphocytes failed to demonstrate significant IL-1β protein, thereby excluding them as a significant source of IL-1β. Because the predominant contaminating cells in our neutrophil enrichment protocol were lymphocytes, these two results together indicate that neutrophils are the major producers of IL-1β. While mononuclear cells are a potential source, the fact that they represent ∼2% of the cells in our samples makes this highly unlikely. The neutrophil-enriched fraction contained more pro-IL-1β relative to the amount of protein loaded per lane (Fig. 3) to yield the same intensity band on Western blots. This indicates that the enrichment of neutrophils also increases the IL-1β content of the samples, further supporting neutrophils as the source of IL-1β. Neutropenic mice, in contrast to non-neutropenic animals, failed to express significant IL-1β among lung mononuclear cell populations in response to either hemorrhage or endotoxin administration. Finally, IL-1β was visualized in lung neutrophils using immunohistochemistry after either hemorrhage or endotoxemia. Together, these data indicate that neutrophils express the majority of detectable IL-1β in IPNMC after hemorrhage or endotoxemia.

Our data do not exclude cell sources other than neutrophils as sources of IL-1β in the pulmonary parenchyma after hemorrhage or endotoxemia. Clearly, alveolar macrophages can produce IL-1β, and often represent the major cell type in BALF (5, 6, 41, 42, 43). Endotoxin induces IL-1β expression in alveolar macrophages in vitro (5). However, our previous studies did not show any increase in IL-1β mRNA levels in alveolar macrophages isolated over the 4 h after hemorrhage (6). Likewise, the present studies fail to demonstrate significant alveolar macrophage-derived IL-1β in the first hour after hemorrhage or endotoxemia. Because alveolar macrophages are poor Ag-presenting cells (44), our findings support the hypothesis that alveolar macrophages are relatively unimportant in the inflammatory response that occurs immediately after insults that may lead to lung injury.

We previously demonstrated that lung IL-1β protein levels increase within 1 h after hemorrhage (45). Pro-IL-1β contains cleavage sites for a variety of enzymes, including IL-1β-converting enzyme and neutrophil elastase, which can produce biologically active IL-1β (40, 46). Mature IL-1β is processed and excreted to the extracellular compartment (23). In our experiments, IPNMC are thoroughly washed during the steps required during processing. Therefore, most of the extracellular mature IL-1β is lost during processing, accounting for the detection of predominantly intracellular pro-IL-1β. Because neutrophil elastase is capable of producing biologically active IL-1β and we have seen the cleavage products of pro-IL-1β in our experiments (see Fig. 3), pro-IL-1β measurements should parallel IL-1β biologic activity.

Although neutrophil production of proteases, reactive oxygen species, and other cytotoxic mediators clearly plays a role in the effector phase of lung injury, our data suggest an expanded role for neutrophils in initiating this inflammatory process. The present results indicate that lung injury may be, at least in part, a “bystander phenomenon”, caused by IL-1β-producing neutrophils that traffic to the lungs, rather than by resident pulmonary cells. The role of IL-1β as a proinflammatory cytokine is well established, and it appears to have a central role in inducing acute lung injury (5, 16, 17). The decrease in IL-1β in neutropenic mice reported here may explain the observation that neutropenia decreases alveolar leak and acute lung injury in endotoxin-treated animals (4).

Because most in vivo studies of lung IL-1β expression are performed on whole lung homogenates (3, 5) or BALF (13, 14), the cellular source of IL-1β is uncertain. BALF used in measuring of IL-1 content may contain a significant number of neutrophils or other cells, particularly late in inflammation (5, 6). Therefore, the IL-1β detected in these studies may originate from neutrophils. Indeed, measures of neutrophil numbers in BALF often parallel the IL-1β content of the BALF (6, 47).

Our previous studies (6, 7, 9), showing different patterns of cytokine expression in alveolar macrophages and intraparenchymal lung cells, coupled with the present results, indicating a central role for neutrophil-derived IL-1β cytokine production in the lung after hemorrhage and endotoxemia, underscore the distinct contributory roles of intraparenchymal and alveolar cell populations in acute lung inflammation. These findings suggest that analysis of BALF samples from human or experimental models may incompletely reflect events in the lungs. For example, we previously found that IL-1β in BALF was increased only 72 h after hemorrhage, even though elevated protein levels were present in lung homogenates within 1 h of blood loss (6). Thus, reliance upon BALF analysis in human and experimental studies to determine the importance of various cytokines may underestimate the role of events in the pulmonary parenchyma, in which the inflammatory process is histologically most intense.

The present studies also suggest that endotoxin or hemorrhage induce IL-1β production by neutrophils before these cells enter the lung parenchyma. Neutrophils seen in the pulmonary vasculature on immunohistochemical sections were noted to express IL-1β. It is possible that interactions between the neutrophils and the vasculature induce the expression of IL-1β within the lumen of the blood vessels. Future studies will be directed at determining whether neutrophils from other sites, such as the peritoneum and the peripheral blood, also express IL-1β.

Both hemorrhage and endotoxemia cause an increase in neutrophil trafficking to the lung. Studies have implicated adhesion molecules (48, 49, 50, 51), catecholamines (52), oxygen radicals (27), cytokines (53, 54), and changes in neutrophil deformability (41, 55) as possible mechanisms for the accumulation of neutrophils in the lungs under these conditions. Our findings suggest that there may be a common pathway for lung injury after sepsis and blood loss that involves neutrophils that traffic to the lungs, produce proinflammatory cytokines such as IL-1β, and initiate an acute inflammatory response.

The authors thank Dr. Charles Dinarello for discussions and Debra Kaneko for her excellent technical support.

1

This work is supported in part by National Institutes of Health Grant HL50284-02. M.V.P. is supported in part by the Multidisciplinary Respiratory Diseases Research Training Grant HL07085-22.

3

Abbreviations used in this paper: ARDS, acute respiratory distress syndrome; BALF, bronchoalveolar lavage fluid; MPO, myeloperoxidase; IPNMC, intraparenchymal pulmonary neutrophil/mononuclear cells; CTX, cyclophosphamide; PMN, polymorphonuclear leukocytes.

1
Brigham, K. L., B. Meyrick.
1986
. Endotoxin and lung injury.
Am. Rev. Respir. Dis.
133
:
913
2
Denis, M., L. Guojian, M. Widmer, A. Cantin.
1994
. A mouse model of lung injury induced by microbial products: implication of tumor necrosis factor.
Am. J. Respir. Cell Mol. Biol.
10
:
658
3
Nill, M. R., T. M. Oberyszyn, M. S. Ross, A. S. Oberyszyn, F. M. Robertson.
1995
. Temporal sequence of pulmonary cytokine gene expression in response to endotoxin in C3H/HeN endotoxin-sensitive and C3H/HeJ endotoxin-resistant mice.
J. Leukocyte Biol.
58
:
563
4
Chang, S.-W., P. Feddersen, P. M. Henson, N. M. Voelkel.
1987
. Platelet-activating factor mediates hemodynamic changes and lung injury in endotoxin-treated rats.
J. Clin. Invest.
79
:
1498
5
Ulitch, T. R., L. R. Watson, Y. Songmei, K. Guo, P. Wang, H. Thang, J. D. Castillo.
1991
. The intratracheal administration of endotoxin and cytokines: characterization of LPS-induced IL-1 and TNF mRNA expression and LPS-, IL-1-, and TNF-inflammatory infiltrate.
Am. J. Pathol.
138
:
1485
6
Shenkar, R., W. F. Coulson, E. Abraham.
1994
. Hemorrhage and resuscitation induce alterations in cytokine expression and the development of acute lung injury.
Am. J. Respir. Cell Mol. Biol.
10
:
290
7
Shenkar, R., E. Abraham.
1993
. Effects of hemorrhage on cytokine gene transcription.
Lymphokine Cytokine Res.
12
:
237
8
Shenkar, R., Y.-H. Chang, E. Abraham.
1994
. Cytokine expression in Peyer’s patches following hemorrhage and resuscitation.
Shock
1
:
25
9
Abraham, E., S. Bursten, R. Shenkar, J. Allbee, R. Tuder, P. Woodson, D. M. Guidot, G. Rice, J. W. Singer, J. E. Repine.
1995
. Phosphatidic acid signaling mediates lung cytokine expression and lung inflammatory injury after hemorrhage in mice.
J. Exp. Med.
181
:
569
10
Abraham, E., N. Richmond, Y. H. Chang.
1988
. The effects of hemorrhage on interleukin 1 release.
Circ. Shock
25
:
33
11
Abraham, E., A. A. Freitas.
1990
. Hemorrhage produces abnormalities in lymphocyte function and lymphokine generation.
J. Immunol.
142
:
3743
12
Chollet-Martin, S., B. Jourdain, C. Gibert, C. Elbim, J. Chastre, M. A. Gougerot-Pocidalo.
1996
. Interactions between neutrophils and cytokines in blood and alveolar spaces during ARDS.
Am. J. Respir. Cell Mol. Biol.
154
:
594
13
Kurdowska, A., E. J. Miller, J. M. Noble, R. P. Baughman, M. A. Matthay, W. G. Brelsford, A. B. Cohen.
1996
. Anti-IL-8 autoantibodies in alveolar fluid from patients with the adult respiratory distress syndrome.
J. Immunol.
157
:
2699
14
Goodman, R. B., R. M. Streiter, D. P. Martin, K. P. Steinberg, J. A. Milberg, R. J. Maunder, S. L. Kunkel, A. Walz, L. D. Hundson, T. R. Martin.
1996
. Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome.
Am. J. Respir. Cell Mol. Biol.
154
:
602
15
Donnelly, T. J., P. Meade, M. Jagels, H. G. Cryer, M. M. Law, T. E. Hugli, W. C. Shoemaker, E. Abraham.
1994
. Cytokine, complement, and endotoxin profiles associated with the development of the adult respiratory distress syndrome.
Crit. Care Med.
22
:
768
16
Pugin, J., B. Ricou, K. P. Steiberg, P. M. Suter, T. R. Martin.
1996
. Proinflammatory activity in bronchoalveolar lavage fluids from patients with ARDS, a prominent role for interleukin-1.
Am. J. Respir. Cell Mol. Biol.
153
:
1850
17
Leff, J. A., J. W. Baer, J. M. Kirkman, M. E. Bodman, M. J. Ostro, J. E. Repine.
1992
. Post-injury treatment with liposome-encapsulated prostaglandin E1 (LIP-PGE1 or TLC C-53) but not free PGE1, decreases acute edematous lung injury (ARDS) in rats given interleukin-1 intratracheally.
Am. Rev. Respir. Dis.
145
:
A845
18
Streiter, R. M., S. W. Chensue, M. A. Basha, T. J. Standiford, J. P. Lynch, S. L. Kunkel.
1990
. Human alveolar macrophage gene expression of interleukin-8 by TNF-a, LPS, and IL-1β.
Am. J. Respir. Cell Mol. Biol.
2
:
321
-326.
19
Li, P., H. Allen, S. Banerjee, S. Franklin, L. Herzog, C. Johnston, J. McDowell, M. Paskind, L. Rodman, J. Salfeld, E. Towne, D. Tracey, S. Wardwell, F.-Y. Wey, W. Wong, R. Kamen, T. Seshadri.
1995
. Mice deficient in interleukin-1 converting enzyme (ICE) are defective in production of mature interleukin-1β and resistant to endotoxic shock.
Cell
80
:
401
20
Rosoff, P. M., N. Savage, C. A. Dinarello.
1988
. Interleukin-1 stimulates diacylglycerol production in T lymphocytes by a novel mechanism.
Cell
54
:
73
21
Stylianou, E., L. A. J. O’Neil, L. Rawlinson, M. R. Edbrooke, P. Woo, J. Saklatvala.
1992
. Interleukin-1 induces NF-κB through its type I but not type II receptor in lymphocytes.
J. Biol. Chem.
267
:
15836
22
Crown, J., A. Jakubowski, N. Kemeny, M. Gordon, C. Gasparetto, G. Wong, G. Toner, B. Meisenberg, J. Botet, J. Applewhite, S. Sinha, M. Moore, D. Kelsen, W. Buhles, J. Gabrilove.
1991
. A phase I trial of human interleukin-1β alone and in combination with myelosuppressive doses of 5-fluorouracil in patients with gastrointestinal cancer.
Blood
78
:
1420
23
Dinarello, C. A..
1996
. Biologic basis for interleukin-1 in disease.
Blood
87
:
2095
24
Cassatella, M. A..
1995
. The production of cytokines by polymorphonuclear neutrophils.
Immunol. Today
16
:
21
25
McCain, R. W., E. P. Holden, T. R. Blackwell, J. W. Christman.
1994
. Leukotriene B4 stimulates human polymorphonuclear leukocytes to synthesize and release interleukin-8 in vitro.
Am. J. Respir. Cell Mol. Biol.
10
:
651
26
Terada, L. S., J. J. Dormish, P. F. Shanley, J. A. Leff, B. O. Anderson, J. E. Repine.
1992
. Circulating xanthine oxidase mediates lung neutrophil sequestration after intestinal ischemia-reperfusion.
Am. J. Physiol.
263
:
L394
27
Anderson, B. O., E. E. Moore, F. A. Moore, J. A. Leff, L. S. Terada, A. H. Harken, J. E. Repine.
1991
. Hypovolemic shock promotes neutrophil sequestration in lungs by a xanthine oxidase-related mechanism.
J. Appl. Physiol.
71
:
1862
28
Heflin, A. C., Jr, K. L. Brigham..
1981
. Prevention by granulocyte depletion of increased vascular permeability of sheep lung following endotoxinemia.
J. Clin. Invest.
68
:
1253
29
Till, G. O., K. T. Johnson, R. Kunkel, P. A. Ward.
1982
. Intravascular activation of complement and acute lung injury: dependency on neutrophils and toxic oxygen metabolites.
J. Clin. Invest.
69
:
1126
30
Abraham, E., A. Robinson.
1992
. Oral immunization with bacterial polysaccharide containing liposomes enhances antigen-specific pulmonary secretory antibody response and resistance to pneumonia.
Vaccine
9
:
757
31
Robinson, A., E. Abraham.
1990
. Hemorrhage in mice produces alterations in pulmonary B cell repertoires.
J. Immunol.
145
:
3734
32
Robinson, A., E. Abraham.
1991
. Effects of hemorrhage and resuscitation on bacterial antigen specific pulmonary plasma cell function.
Crit. Care Med.
19
:
1285
33
Abraham, E..
1992
. Intranasal immunization with bacterial polysaccharide containing liposomes enhances antigen-specific pulmonary secretory antibody response.
Vaccine
10
:
461
34
Sugarawa, T., M. Miyamoto, S. Takayama, M. Kato.
1995
. Separation of neutrophils from blood in human and laboratory animals and comparison of the chemotaxis.
J. Pharmacol. Toxicol. Methods
33
:
91
35
Proietti, E., E. Tritarelli, L. Gabriele, U. Testa, G. Greco, E. Pelosi, M. Gabbianelli, F. Belardelli, C. Peschle.
1993
. Combined interleukin 1β/interleukin 2 treatment in mice: synergistic myelostimulatory activity and protection against cyclophosphamide-induced myelosuppression.
Cancer Res.
53
:
569
36
Goldblum, S. E., K.-M. Wu, M. Jay.
1985
. Lung myeloperoxidase as a measure of pulmonary leukostasis in rabbits.
J. Appl. Physiol.
59
:
1978
37
Laemmli, U. K..
1970
. Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature
227
:
680
38
Voelkel, N. F., R. M. Tuder, K. Wade, M. Höper, R. A. Lepley, J. L. Goulet, B. H. Koller.
1996
. Inhibition of 5-lipoxygenase-activating protein (FLAP) reduces pulmonary vascular reactivity and pulmonary hypertension in hypoxic rats.
J. Clin. Invest.
97
:
2491
39
Halbower, A. C., R. J. Mason, S. H. Abman, R. M. Tuder.
1994
. Agarose infiltration improves morphology of cryostat sections of lung.
Lab. Invest.
71
:
149
40
Fantuzzi, G., G. Ku, M. W. Harding, D. J. Livingston, J. D. Sipe, K. Kuida, R. A. Flavell, C. A. Dinarello.
1997
. Response to local inflammation of IL-1β-converting enzyme-deficient mice.
J. Immunol.
158
:
1818
41
Strieter, R. M., S. L. Kunkel.
1994
. Acute lung injury: the role of cytokines in the elicitation of neutrophils.
J. Investig. Med.
42
:
640
42
Becker, S., R. B. Devlin, J. S. Haskill.
1989
. Differential production of tumor necrosis factor, macrophage colony stimulating factor, and interleukin-1 by human alveolar macrophages.
J. Leukocyte Biol.
45
:
353
43
Jacobs, R. F., D. R. Tabor, A. W. Burks, G. D. Campbell.
1989
. Elevated interleukin-1 release by human alveolar macrophages during the adult respiratory distress syndrome.
Am. Rev. Respir. Dis.
140
:
1686
44
Chelen, C. J., Y. Fang, G. J. Freeman, H. Secrist, J. D. Marshall, P. T. Hwang, L. R. Frankel, R. H. DeKruyff, D. T. Umetsu.
1995
. Human alveolar macrophages present antigen ineffectively due to defective expression of B7 costimulatory cell surface molecules.
J. Clin. Invest.
95
:
1415
45
Le Tulzo, Y., R. Shenkar, D. Kaneko, P. Moine, G. Fantuzzi, C. A. Dinarello, E. Abraham.
1997
. Hemorrhage increases cytokine expression in lung mononuclear cells in mice: involvement of catecholamines in nuclear factor-κB regulation and cytokine expression.
J. Clin. Invest.
99
:
1516
46
Black, R. A., S. R. Kronheim, M. Cantrell, M. C. Deeley, C. J. March, K. S. Prickett, J. Wignall, P. J. Conlon, D. Cosman, T. P. Hopp, D. Y. Mochizuki.
1988
. Generation of biologically active interleukin-1 beta by proteolytic cleavage of the inactive precursor.
J. Biol. Chem.
263
:
9437
47
Suter, P. M., S. Suter, E. Girardin, P. Roux-Lombard, G. E. Grau, J.-M. Dayer.
1992
. High bronchoalveolar levels of tumor necrosis factor and its inhibitors, interleukin-1, interferon, and elastase, in patients with adult respiratory distress syndrome after trauma, shock, or sepsis.
Am. Rev. Respir. Dis.
145
:
1016
48
Rainger, G. E., A. Fisher, C. Shearman, G. B. Nash.
1995
. Adhesion of flowing neutrophils to cultured endothelial cells after hypoxia and reoxygenation in vitro.
Am. J. Physiol.
269
:
H1398
49
Thomas, D. D., S. R. Sharar, R. K. Winn, E. Y. Chi, E. D. Verrier, M. D. Allen, M. J. Bishop.
1995
. CD18-independent mechanism of neutrophil migration in the rabbit lung after ischemia-reperfusion.
Ann. Thorac. Surg.
60
:
1360
50
Smart, S. J., T. Casale.
1994
. TNF-α-induced transendothelial neutrophil migration is IL-8 dependent.
Am. J. Physiol.
266
:
L238
51
Johnson, R. C., T. N. Mayadas, P. S. Frenette, R. E. Mebius, M. Subramaniam, A. Lacasce, R. O. Hynes, D. D. Wagner.
1995
. Blood cell dynamics in P-selectin-deficient mice.
Blood
86
:
1106
52
Selby, C., W. MacNee.
1993
. Factors affecting neutrophil transit during acute pulmonary inflammation: mini review.
Exp. Lung Res.
19
:
407
53
Inoue, H., M. Hara, P. P. Massion, K. M. Grattan, J. A. Lausier, B. Chan, T. Kaneko, K. Isono, P. G. Jorens, I. F. Ueki, J. A. Nadel.
1995
. Role of recruited neutrophils in interleukin-8 production in dog trachea after stimulation with Pseudomonas in vivo.
Am. J. Respir. Cell Mol. Biol.
13
:
570
54
Blackwell, T. S., E. P. Holden, T. R. Blackwell, J. E. DeLarco, J. W. Christman.
1994
. Cytokine-induced neutrophil chemoattractant mediates neutrophilic alveolitis in rats: association with nuclear factor κ B activation.
Am. J. Respir. Cell Mol. Biol.
11
:
464
55
Brown, D. M., E. Drost, K. Donaldson, W. MacNee.
1995
. Deformability and CD11/CD18 expression of sequestered neutrophils in normal and inflamed lungs.
Am. J. Respir. Cell Mol. Biol.
13
:
531