Double-hit mouse model of cigarette smoke priming for acute lung injury

Abstract

Epidemiological studies indicate that cigarette smoking (CS) increases the risk and severity of acute lung injury (ALI)/acute respiratory distress syndrome (ARDS). The mechanism is not understood, at least in part because of lack of animal models that reproduce the key features of the CS priming process. In this study, using two strains of mice, we characterized a double-hit mouse model of ALI induced by CS priming of injury caused by lipopolysaccharide (LPS). C57BL/6 and AKR mice were preexposed to CS briefly (3 h) or subacutely (3 wk) before intratracheal instillation of LPS and ALI was assessed 18 h after LPS administration by measuring lung static compliance, lung edema, vascular permeability, inflammation, and alveolar apoptosis. We found that as little as 3 h of exposure to CS enhanced LPS-induced ALI in both strains of mice. Similar exacerbating effects were observed after 3 wk of preexposure to CS. However, there was a strain difference in susceptibility to CS priming for ALI, with a greater effect in AKR mice. The key features we observed suggest that 3 wk of CS preexposure of AKR mice is a reproducible, clinically relevant animal model that is useful for studying mechanisms and treatment of CS priming for a second-hit-induced ALI. Our data also support the concept that increased susceptibility to ALI/ARDS is an important adverse health consequence of CS exposure that needs to be taken into consideration when treating critically ill individuals.

cigarette smoke (CS)-induced diseases remain a major cause of preventable premature deaths, resulting in a trillion dollars of economic loss each year worldwide (1). Smokers are more susceptible to infections by Mycobacterium tuberculosis (11), Streptococcus pneumoniae (6), and Influenza A (43). Current smokers with community-acquired pneumonia often develop severe sepsis (6). In fact, respiratory tract infections and complications are the most common causes of morbidity and mortality among smokers (1). CS also increases the frequency and severity of respiratory tract infections in experimental animals caused by bacteria (14, 16, 35) and viruses (30).

Acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) are life-threatening disorders (42) that can be triggered by sepsis, pneumonia, trauma, hemorrhage, acid aspiration, high-pressure ventilation, blood transfusion, pulmonary contusion, burn injury, acute pancreatitis, and hyperoxia. It remains unclear why only some individuals with known clinical risk factors progress to these syndromes. Early studies based on self-reported smoking status to address whether CS increases the risk for ALI/ARDS have yielded conflicting results (17, 18, 28). Using plasma cotinine to verify tobacco use, Calfee and colleagues (9) reported that both active and passive tobacco smoking were independently associated with development of ARDS in patients with severe trauma. Using urine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol, a validated biomarker for tobacco use, Hsieh et al. (26) demonstrated that patients with ARDS had a higher prevalence of cigarette smoking. These results support the notion that CS is a priming factor for development of ARDS in humans. A multicenter study reported that CS had dose-dependent adverse effects on morbidity and mortality in critically ill patients after adjusting for other confounders (24). In addition, tobacco smokers exhibited exaggerated lung inflammation and alveolar barrier disruption after inhalation of lipopolysaccharide (LPS) (37). Furthermore, donor lungs explanted from otherwise healthy smokers were heavier (suggesting edema) and had poorer outcomes after transplantation (50). Taken together, CS exposure appears, not only to increase the risk, but also to potentiate the severity of ALI/ARDS in humans. However, the mechanism by which CS primes the lungs for development of ALI/ARDS following otherwise only modest noxious secondary insult is unknown. To our knowledge, there is no clinically relevant double-hit animal model available to reproduce the key features/pathophysiology of this CS priming process. Such an animal model is crucial for understanding how CS primes/exacerbates ALI/ARDS and for establishing effective prevention and treatment for susceptible populations who are vulnerable to ALI/ARDS.

ALI/ARDS are inflammatory disorders characterized by activation of alveolar macrophages (AM), induction of cytokines and chemokines, and parenchymal infiltration by polymorphonuclear neutrophils (PMNs). As reviewed by Kopf and others (22, 31), AMs are one of the first lines of defense against external pathogens and particles. In response to foreign particles and respiratory tract infections, AMs are activated and produce a variety of early-response cytokines [e.g., tumor necrosis factor (TNF)-α and interleukin (IL)-6] and chemokines [e.g., macrophage inflammatory protein (MIP)-2, a mouse counterpart of human IL-8] to combat invading particles or pathogens. Macrophage-derived TNF-α potentiates production of keratinocyte-derived chemokines (KC) by Clara cells. In rodents, MIP-2 and KC are major chemokines for recruitment of PMN into lung tissue. These early response cytokines and chemokines also induce lung infiltration by circulating macrophages and lymphocytes. Resident AMs and interstitial and alveolar inflammatory cells can reduce lung injury by clearing external pathogens and particles. However, excessive activation and recruitment of inflammatory cells at the site of inflammation is thought to augment bystander tissue damage attributable to overproduction of cytotoxic proinflammatory cytokines (e.g., IL-1β), nitric oxide (NO), reactive oxygen species (ROS), matrix metalloproteinases (MMPs), and elastases. TNF-α and IL-1β are elevated in bronchoalveolar lavage (BAL) fluid and plasma of patients with ARDS (40). Increased numbers of PMN and IL-8 in the BAL fluid are correlated with severity of ARDS and poorer clinical outcomes (3, 12). Furthermore, severity of septic ALI is reduced by neutrophil depletion in mice (39). These results indicate that excessive AM activation and PMN infiltration are key features and important contributors to ALI/ARDS.

It is well known that genetic factors influence susceptibility to CS-induced chronic obstructive pulmonary disease (COPD) and lung cancer. However, it is unknown whether there is a genetic susceptibility to CS priming for development of ALI/ARDS. Mice of an inbred strain are genetically identical; thus different inbred strains of mice are useful models to investigate genetic susceptibility to CS-induced diseases. The most commonly used inbred strain, C57BL/6, is moderately deficient in serum α-1-proteinase inhibitor (10) and is mildly susceptible to CS-induced emphysema (19). AKR mice, which display high leukemia incidence but possess immunocompetent lymphocytes (21), are highly susceptible to CS-induced emphysema (19). In this study, we investigated the effects of both a brief and prolonged CS exposure on susceptibility to ALI induced by LPS in C57BL/6 and AKR mice. LPS, a lipid endotoxin expressed on the outer layer of cell wall of Gram-negative bacteria, has been widely used to induce experimental ALI (13). Using this mild ALI mouse model, we have previously shown that a brief (6 h) CS preexposure exacerbates LPS-induced lung edema in C57BL/6 mice (33). In the present study, we assessed lung static compliance, lung edema, vascular permeability, inflammation, and alveolar cell apoptosis in C57BL/6 and AKR mice preexposed to CS briefly (3 h) or subacutely (3 wk) followed by challenges with LPS. We demonstrated that AKR mice are more susceptible to CS priming for ALI. This study provides a clinically relevant mouse model of CS priming for a second-hit-induced ALI, which is critically needed for studying mechanisms of CS-induced disease processes. Our results also support the notion that increased susceptibility to ALI/ARDS is an important adverse health consequence of CS exposure that needs to be taken into consideration when treating critically ill individuals.

MATERIALS AND METHODS

Regents.

LPS (lot no. L5418) from Escherichia coli serotype 055:B5 (endotoxin level = 500,000 endotoxin units/mg) was purchased from Sigma-Aldrich (St. Louis, MO) and filtered with 0.2-μm filters before use. The same lot of LPS was used for the entire study. Mouse TNF-α, IL-6, IL-10, KC, and MIP-2 ELISA kits were purchased from BD Biosciences (San Diego, CA). Antibodies directed against CD68 and CD206 were purchased from Abcam (Cambridge, MA). Antibody directed against Ly6G was from Biolegend (San Diego, CA). Antibodies directed against caspases-3 and actin were purchased from Cell Signaling Technology (Danvers, MA) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively.

Mice.

Male 6-wk-old C57BL/6 and AKR mice were purchased from the Jackson Laboratory (Bar Harbor, ME). All experiments were performed according to the protocol approved by the Institutional Animal Care and Use Committee of the Providence Veterans Affairs Medical Center for the humane use of experimental animals. Animals were housed in standard conditions (12-h:12-h light/dark cycle, 68–72°F and a humidity of 30–70%) in ventilated racks fitted with automatic watering systems and fed with standard chow ad libitum. For smoking exposure, mice designated in the smoking group were temporally transported into standard shoe cages with bottles of water and standard chow ad libitum before entering the smoking chamber. Immediately after smoke exposure each day, mice were brought back to their original housing cages that were never directly exposed to CS.

CS exposure of mice.

Male 6-wk-old C57BL/6 and AKR mice were exposed to CS for 3 or 6 h or for 3 consecutive weeks at 6 h per day and 4 days per week (Monday, Tuesday, Thursday, and Friday), using a TE-10 mouse smoking machine (Teague Enterprises, Woodland, CA) and 3R4F reference cigarettes (University of Kentucky, Tobacco Research Institute, Lexington, KY), as we previously described (27, 33, 41). The room air control mice were housed on a shelf near the smoking machine, whereas smoke-exposed mice were kept inside the smoking chamber for 3 or 6 h per day. When mice were not exposed to CS, the smoke-exposed mice were brought back to the same shelf where the room air control mice were housed. Smoking chamber atmosphere was controlled and monitored for total suspended particles at a concentration of 120 mg/m³ (3 cigarettes were lit at a time). The smoke was a mixture of side-stream smoke (89%) and mainstream smoke (11%). Animals were observed, and body weights were recorded daily.

Intratracheal administration of LPS.

One hour after the final CS exposure, mice were anesthetized with 3% isofluorane and intratracheally instilled with 2.5 mg/kg of LPS (dissolved in saline) or an equal volume of saline (∼50 μl). Lung injury was assessed 18 h after LPS administration.

Measurement of lung mechanics.

As we described previously (34), lung mechanics was measured using the FlexiVent system (SCIREQ, Montreal, Canada). Mice were deeply anesthetized using 90 mg/kg pentobarbital via i.p. injection. While mice were under anesthesia, a calibrated 18-gauge blunt catheter was inserted into the trachea through an incision. The mice were then ventilated via the FlexiVent ventilator at a constant tidal volume of 10 ml/kg and respiratory rate of 150 breaths per min. Lung static compliance (C_st) was measured by using FlexiWare software per manufacturer recommendations.

Assessment of protein concentrations and inflammatory cell counts in BAL fluid.

As we described previously (33), lungs were lavaged for one time only with 600 μl of sterilized saline by using a 1-ml syringe via a tracheal catheter. The protein concentrations in the BAL fluid were measured by DC protein assay. The total number of inflammatory cells in the BAL fluid was counted using a phase-contrast light microscope at ×400 magnification.

Lung wet-to-dry weight ratio.

As we previously described (27, 33, 41), after mice were euthanized, lungs were harvested and weighed immediately as the wet weights. The dry weights of lungs were recorded after the lungs had been dried for 48 h at 90°C in an oven. The ratio of wet-to-dry weights was calculated.

Determination of lung vascular leakage.

The loss of lung vascular integrity was assessed by extravasation of albumin-conjugated Evans blue dye (EBD). The albumin-EBD solution (1% in saline) was intravenously administered at a dose of 4 ml/kg body wt via the tail vein 1 h before collection of lungs. For collection of lungs, mice were anesthetized with 3% isofluorane. Under anesthesia, mouse lungs were transcardially perfused with 10 ml of sterilized saline to completely flush intravascular albumin-EBD that may have remained in the vascular system. Albumin-EBD was extracted from lungs by incubation of the lung with formamide at 65°C overnight. Following centrifugation at 12,000 g for 30 min, the concentrations of EBD extracted in the supernatant was determined spectrophotometrically from the absorbance at 620 nm of each supernatant and corrected by absorbance at 740 nm and calculated by comparison with a standard curve of EBD prepared in the same assay.

Measurement of cytokines and chemokines in lung homogenates.

Lung tissue was collected and frozen immediately in liquid N₂ and stored at −80°C. Lung tissue was homogenized immediately before measurements. TNF-α, IL-6, IL-10, MIP-2, and KC were assayed by ELISA kits per manufacturer recommendations (Biosciences, San Diego, CA).

Lung histology and immunohistochemistry.

The left lungs were fixed with 10% formalin and embedded with paraffin. Sagittal sections (5 μm) were used for hematoxylin and eosin (H and E) staining and immunohistochemistry. Although this methodology has been widely used and validated, a limitation of this study is analysis of 5-μm sagittal sections, rather than stereometric analysis of three-dimensional lung tissue. Lung macrophages and neutrophils on 5-μm sagittal sections were identified based on their antigen expression. Immunohistochemistry for CD68, CD206, and Ly6G was performed after deparaffinization and rehydration in an ethanol series. Heat-induced epitope retrieval was performed using an antigen retrieval buffer for 20 min. For CD68 and CD206 staining, the slides were blocked with 10% goat serum for 60 min and then incubated with the primary antibodies for 60 min at room temperature. The sections were then incubated with secondary antibody (Vector Laboratories, Burlingame, CA) for 30 min at room temperature and developed with ImmPACT Vector Red (SK-5105). For Ly6G staining, following antibody binding, the sections were developed with ImmPACT DAB (SK-4105). All sections were counterstained with H and E, dehydrated in a sequence of graded alcohol/water mixtures and xylenes, and then covered with a cover slip. Images were captured at ×40 magnification by using Aperio Scanscope CS2 whole slide image system (Leica Biosystems, Nussloch, Germany), which captures an image of the entire lung, rather than a small portion of lung tissue. Ten high-power fields representing upper, middle, and lower regions of each lung were analyzed for CD68-, CD206-, and Ly6G-positive cells using ImageScope software by a blinded observer. Data are presented as means ± SE of CD68-, CD206-, or Ly6G-positive cells per high-power field.

Data analysis.

The number of mice used in this study is indicated in figure legends. Data are presented as means ± SE. ANOVA and Fisher's least significant difference post hoc test (StatView program; SAS Institute, Cary, NC) were used to determine significant differences among groups, with differences among means considered significant when P < 0.05.

RESULTS

Brief preexposure to CS exacerbated LPS-induced lung edema in C57BL/6 and AKR mice.

It is well known that only a fraction of smokers develop COPD or lung cancer, suggesting that genetic and environmental factors influence susceptibility to cigarette smoking-induced diseases. Similarly, AKR mice are more susceptible to CS-induced emphysema than C57BL/6 mice (19). We have previously shown that 6 h of exposure to CS exacerbates LPS-induced lung edema in C57BL/6 mice (33). In this study, we hypothesized that there is a strain-dependent susceptibility to CS priming for ALI, with AKR strain being more susceptible. To address this hypothesis, we examined the effects of 6 h of exposure to the same concentration of CS followed by administration of the same dose of LPS on AKR mice. We found that three of four (75%) AKR mice that were preexposed to CS for 6 h followed by LPS challenge died within 18 h after LPS challenge. The only surviving AKR mouse was so sick that we had to euthanize it according to our animal research regulations. No deaths were noted in C57BL/6 mice receiving the same treatments. The results suggest that AKR mice are highly susceptible to CS priming LPS-induced sickness. Because of the high mortality rate in AKR mice exposed to CS for 6 h followed by LPS challenge, we had to modify our protocol by shortening CS exposure time from 6 h and tested the effects of 3 h of CS preexposure on LPS-induced ALI in both strains of mice. As expected, compared with vehicle (saline) treatment, LPS significantly decreased lung static compliance (C_st), characteristic of lung edema, in both C57BL/6 and AKR mice that had been preexposed to either room air or CS (Fig. 1A). Exposure to CS alone for 3 h did not change lung C_st in either strain of mice (Fig. 1A). However, both C57BL/6 and AKR mice preexposed to CS for 3 h followed by LPS challenge had a lower C_st, compared with that of their counterparts challenged by LPS alone (Fig. 1A).

Fig. 1.

Effects of a brief cigarette smoke (CS) exposure on LPS-induced lung edema in two strains of mice. Male 6-wk-old C57BL/6 and AKR mice were exposed to room air (RA) or CS for 3 h as described in materials and methods. 1 h after CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control (ctrl). After 18 h, lung static compliance (C_st) was assessed using FlexiVent system (A). Bronchoalveolar lavage (BAL) fluid was collected for assessment of BAL protein content (B). Lung wet-to-dry weight ratio was assessed in an additional set of mice that were subjected to the same treatments (C). In A, 4–6 C57BL/6 mice per group and 6–7 AKR mice per group were used; In B and C, 4–7 C57BL/6 mice per group and 5–6 AKR mice per group were used. εP < 0.05 CS/ctrl vs. RA/ctrl; *P < 0.05 RA/LPS vs. RA/ctrl; δP < 0.05 CS/LPS vs. CS/ctrl; €P < 0.05 CS/LPS vs. RA/LPS.

We next assessed lung edema by measuring BAL protein content and lung wet-to-dry weight ratio. Not surprisingly, compared with vehicle controls, LPS significantly increased BAL protein content in both C57BL/6 and AKR mice preexposed to either room air or CS (Fig. 1B). Exposure to CS alone for 3 h did not change BAL protein content but significantly exacerbated LPS-induced increase in BAL protein content in both C57BL/6 and AKR mice (Fig. 1B). Similarly, compared with vehicle controls, LPS significantly increased lung wet-to-dry weight ratio in both strains of mice preexposed to either room air or CS (Fig. 1C). Preexposure to CS for 3 h, not only increased basal, but also exacerbated LPS-induced increase in lung wet-to-dry weight ratio in AKR mice (Fig. 1C).

Brief CS preexposure had a modest effect on proinflammatory response but inhibited anti-inflammatory response to LPS stimulation.

Lung infiltration of inflammatory cells is a characteristic of ALI. As expected, compared with vehicle controls, intratracheal administration of LPS significantly increased the total number of inflammatory cells in BAL fluid in both strains of mice preexposed to either room air or CS (Fig. 2A). Preexposure to CS for 3 h did not significantly increase baseline levels of BAL inflammatory cell counts in either strain of mice but significantly exaggerated LPS-induced increase in BAL inflammatory cells in AKR mice (Fig. 2A). Lung infiltration of inflammatory cells was also observed in AKR mice exposed to CS or LPS alone, with an additive effect by double-hit injuries (Fig. 2B). Taken together, these findings suggest that a brief CS exposure primes AKR mice for development of lung edema and inflammation.

Fig. 2.

Effects of a brief CS exposure on LPS-induced lung inflammation in 2 strains of mice. Male 6-wk-old C57BL/6 and AKR mice were exposed to RA or CS for 3 h as described in materials and methods. At 1 h after CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control. After 18 h, the number of the total inflammatory cells in BAL fluid was assessed (A). Lung tissue was collected for histology by hematoxylin and eosin staining (B). Arrows indicate inflammatory cells. Lung homogenates were prepared for assessments of levels of tumor necrosis factor α (TNF-α) (C), IL-6 (D), macrophage inflammatory protein 2 (MIP-2) (E), keratinocyte-derived chemokines (KC) (F), and IL-10 (G) by ELISA. 4–6 C57BL/6 mice per group and 5–6 AKR mice per group were used in each panel. εP < 0.05 CS/ctrl vs. RA/ctrl; *P < 0.05 RA/LPS vs. RA/ctrl; δP < 0.05 CS/LPS vs. CS/ctrl; €P < 0.05 CS/LPS vs. RA/LPS.

A functional innate immune response is essential for fighting against infections and for resolution of tissue injury. To understand the mechanism underlying the different sensitivity of different mouse strains to CS priming for ALI, we examined aspects of the innate immune response in our experimental models. As expected, intratracheal administration of LPS significantly increased the levels of proinflammatory cytokines, TNF-α and IL-6 (Fig. 2, C and D), and chemokines, MIP-2 and KC (Fig. 2, E and F), in both strains of mice exposed to either room air or CS. Preexposure to CS for 3 h did not alter basal but resulted in a trend toward exaggeration of LPS-induced production of these proinflammatory cytokines and chemokines in both strains of mice (Fig. 2, C–F).

IL-10 is an anti-inflammatory cytokine. As anticipated, LPS increased IL-10 levels in both strains of mice exposed to room air (Fig. 2G). Interestingly, 3 h of CS exposure alone significantly elevated lung IL-10 levels in C57BL/6 mice but not in AKR mice (Fig. 2G). Furthermore, compared with mice exposed to room air and challenged by vehicle control, there was no increase in IL-10 levels in mice preexposed to CS for 3 h followed by challenges of LPS in either strain of mice (Fig. 2G). These results suggest that blunting IL-10 production may play a role in acute CS priming for ALI.

Prolonged (3 wk) CS exposure caused an initial loss and a slower gain in body weight, particularly in AKR mice.

On the basis of visual observations, both strains of mice exhibited some levels of distress, e.g., reduced physical movement and tendency to stay together within the first 1 h following CS exposure. This behavior disappeared 1–2 h after removal from the smoking chamber in both strains of mice. During the 3 wk of room air exposure, C57BL/6 and AKR mice had a similar and steady body weight gain (Fig. 3). By day 21, the mean body weight gains in C57BL/6 and AKR mice exposed to room air were 117.73 ± 2.3% and 117.80 ± 2.75%, respectively. During the 3 wk of CS exposure, both strains of mice had a significant initial loss of body weight followed by a lower rate of gain than that of the respective room air-exposed strains of mice, particularly in AKR mice. These data suggest that prolonged CS exposure had a greater adverse effect on the overall health of AKR mice.

Fig. 3.

Effects of prolonged CS exposure on body weight of 2 strains of mice. Male 6-wk-old C57BL/6 and AKR mice were exposed to RA or CS for 3 wk as described in materials and methods. Mouse body weight was measured daily immediately before exposure to CS. 8–9 C57BL/6 mice per group and 9–11 AKR mice per group were used. £P < 0.05 CS vs. RA for C57BL/6 mice; *P < 0.05 CS vs. RA for AKR mice.

Prolonged CS preexposure exacerbated LPS-induced lung edema, particularly in AKR mice.

To determine whether prolonged exposure to CS exaggerates ALI, we examined the effects of 3 wk of CS exposure on LPS-induced ALI in both C57BL/6 and AKR mice. Consistent with our previous findings (34), after 3 wk of CS exposure, C57BL/6 mice did not show any significant changes in C_st (Fig. 4A). In contrast, AKR mice displayed a significant increase in C_st (Fig. 4A), suggesting that AKR mice had mild emphysema-like changes. Interestingly, preexposure to CS for 3 wk dramatically exaggerated LPS-induced decrease in C_st in AKR but not in C57BL/6 mice (Fig. 4A).

Fig. 4.

Effects of prolonged CS exposure on LPS-induced lung edema in 2 strains of mice. Male 6-wk-old C57BL/6 and AKR mice were exposed to RA or CS for 3 wk as described in materials and methods. 1 h after the last CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control. After 18 h, lung static compliance (C_st) was assessed using FlexiVent system (A). BAL fluid was collected for assessment of BAL protein content (B). Lung wet-to-dry weight ratio (C) and lung extravasation of albumin-conjugated Evans blue dye (EBD) (D) were assessed in additional sets of AKR mice that were subjected to the same treatments. In A, 9–10 C57BL/6 mice per group and 9–11 AKR mice per group were used; in B, 3–4 C57BL/6 mice per group and 4–6 AKR mice per group were used; in C, 3 AKR mice per group were used; in D, 4–5 AKR mice per group were used. εP < 0.05 CS/ctrl vs. RA/ctrl; *P < 0.05 RA/LPS vs. RA/ctrl; δP < 0.05 CS/LPS vs. CS/ctrl; €P < 0.05 CS/LPS vs. RA/LPS.

Compared with room air exposure, CS exposure for 3 wk slightly elevated BAL protein content in both strains of mice (Fig. 4B). Similar to the results shown in Fig. 1B, LPS significantly increased BAL protein content in both C57BL/6 and AKR mice preexposed to either room air or CS for 3 wk (Fig. 4B). Preexposure to CS for 3 wk exaggerated LPS-induced increase in BAL protein levels in both strains of mice, with a greater effect in AKR mice (Fig. 4B).

Similarly, LPS exposure significantly enhanced lung water content, as assessed by lung wet-to-dry weight ratio, and promoted lung vascular permeability, as assessed by extravasation of EBD into lung tissue in AKR mice preexposed to either room air or CS for 3 wk (Fig. 4, C and D). Preexposure to CS for 3 wk did not significantly alter lung water (Fig. 4C) or lung vascular permeability (Fig. 4D) in vehicle controls but dramatically exaggerated LPS-induced increase in lung water (Fig. 4C) and lung vascular permeability (Fig. 4D) in AKR mice. There was no additive effect of 3 wk of CS preexposure on LPS-induced changes in lung wet-to-dry weight ratio in C57BL/6 mice (data not shown). Together, these results indicate that AKR mice are more susceptible to development of lung edema after prolonged exposure to CS.

Prolonged CS preexposure induced divergent immune responses to LPS in C57BL/6 and AKR mice.

Similar to our observations in brief CS exposure model, LPS challenge significantly increased BAL total inflammatory cell count (Fig. 5A) and the levels of TNF-α (Fig. 5B), MIP-2 (Fig. 5C), and KC (Fig. 5D) in both strains of mice preexposed to either room air or CS for 3 wk.

Fig. 5.

Effects of prolonged CS exposure on LPS-induced lung inflammation in 2 strains of mice. Male 6-wk-old C57BL/6 and AKR mice were exposed to RA or CS for 3 wk as described in materials and methods. 1 h after the last CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control. After 18 h, the number of the total inflammatory cells in BAL fluid was assessed (A). Lung tissue was collected, and lung homogenates were prepared for assessments of levels of TNF-α (B), MIP-2 (C), KC (D), and IL-10 (E) by ELISA. In A, 3–4 C57BL/6 mice per group and 4–6 AKR mice per group were used; in B–E, 4 C57BL/6 mice per group and 4 AKR mice per group were used. εP < 0.05 CS/ctrl vs. RA/ctrl; *P < 0.05 RA/LPS vs. RA/ctrl; δP < 0.05 CS/LPS vs. CS/ctrl; €P < 0.05 CS/LPS vs. RA/LPS.

We noted that preexposure to CS for 3 wk did not significantly alter BAL total inflammatory cell count (Fig. 5A) or the levels of lung MIP-2 or KC (Fig. 5, C and D) in both strains of mice. Interestingly, the levels of TNF-α and IL-10 were increased in C57BL/6 mice but decreased in AKR mice after exposure to CS alone for 3 wk (Fig. 5, B and E).

Preexposure to CS for 3 wk promoted LPS-induced increase in BAL total cell count (Fig. 5A) and levels of MIP-2 and KC (Fig. 5, C and D) in AKR mice but not in C57BL/6 mice. On the other hand, preexposure to CS for 3 wk had an additive effect on LPS-induced increase in TNF-α (Fig. 5B) and IL-10 (Fig. 5E) in C57BL/6 mice but not in AKR mice. Taken together, these results indicate that prolonged (3 wk) CS exposure primes AKR mice for development of ALI. Our results also suggest an association of blunting of IL-10 production with increased susceptibility to ALI in the more susceptible AKR mice.

Prolonged CS preexposure exacerbated LPS-induced increase in AMs, PMNs, and M2 macrophages in AKR mice.

Because CS significantly elevated LPS-induced increase in BAL inflammatory cells and lung MIP-2 and KC levels in AKR mice, we examined CD68⁺ AMs and lung infiltration of Ly6G⁺ PMNs. Because anti-inflammatory M2 macrophages have been shown to be protective against lung injury (29), we also assessed lung infiltration of CD206⁺ M2 macrophages. As expected, LPS significantly increased the number of AMs, PMNs, and M2 macrophages (Fig. 6, A–C) in AKR mice preexposed to either room air or CS for 3 wk. Exposure of AKR mice to CS alone for 3 wk appears to elevate the number of AMs and M2 macrophages with less effect on PMNs (Fig. 6, A–C). Interestingly, LPS-induced increases in AMs, PMNs, and M2 macrophages were significantly potentiated by preexposure to CS for 3 wk (Fig. 6, A–C).

Fig. 6.

Effects of prolonged CS exposure on LPS-induced alveolar infiltration of inflammatory cells in AKR mice. Male 6-wk-old AKR mice were exposed to RA or CS for 3 wk as described in materials and methods. 1 h after the last CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control. After 18 h, lung tissue was collected, fixed, and paraffin imbedded. Alveolar macrophages (AMs), polymorphonuclear neutrophils (PMNs), and M2 macrophages were identified by immunohistochemistry using antibodies against CD68 (A and B), Ly6G (C and D), and CD206 (E and F), respectively. 3–4 mice per group were used in each panel. Red staining indicates CD68⁺ (A) or CD206⁺ (E) cells; brown staining indicates Ly6G⁺ cells (C). A, C, and E are representative images. Arrows indicate respective inflammatory cells. B, D, and F are expressed as means ± SE of number of positive cells per slides. εP < 0.05 CS/ctrl vs. RA/ctrl; *P < 0.05 RA/LPS vs. RA/ctrl; δP < 0.05 CS/LPS vs. CS/ctrl; €P < 0.05 CS/LPS vs. RA/LPS.

Alveolar cell apoptosis has been implicated in development of ALI. We noted that LPS instillation significantly increased lung cell apoptosis, as indicated by increased cleavage of caspase-3; this effect was not exaggerated by preexposure to CS for 3 wk (Fig. 7).

Fig. 7.

Effects of prolonged CS exposure on LPS-induced lung cell apoptosis in AKR mice. Male 6-wk-old AKR mice were exposed to RA or CS for 3 wk as described in materials and methods. 1 h after the last CS exposure, mice were intratracheally administered with 2.5 mg/kg of LPS or equal volume of saline as a control. After 18 h, lung tissue was collected for assessment of an apoptosis marker, caspase-3 cleavage. Actin was used for loading control. 3–4 mice per group were used. *P < 0.05 RA/LPS vs. RA/ctrl.

DISCUSSION

ALI/ARDS are characterized by injury to the alveolar-capillary barrier, accumulation of protein-rich edema fluid in airspace, and lung inflammation. The key pathological features of human ALI/ARDS are severe intra-alveolar infiltrates of PMN and lung edema. In this study, we show that both a brief (3 h) and prolonged (3 wk) CS exposure act as a priming factor for development of ALI after a subsequent endotoxin challenge. Importantly, using this approach, we are able to demonstrate that there is a strain difference in susceptibility to CS priming for ALI. Our data indicate that preexposure of AKR mice to CS for 3 wk is a reproducible clinically relevant animal model to study mechanism and treatment of CS priming for ALI. Our data also support the notion that increased susceptibility to ALI/ARDS is an important adverse health consequence of CS exposure in susceptible individuals.

It is well known that LPS directly increases lung endothelial cell permeability (47). LPS inhalation in humans also activates Toll-like receptor (TLR)-2 and TLR4 in AMs and triggers AM activation and production of proinflammatory cytokines (25), a mechanism critical to host defense against bacterial infections. As expected, compared with mice intratracheally challenged by vehicle control, intratracheal instillation of LPS decreased lung static compliance (C_st), increased BAL protein content and BAL inflammatory cells, enhanced lung wet-to-dry weight ratio and lung extravasation of albumin-conjugated EBD, elevated lung cytokines (TNF-α and IL-6) and chemokines (MIP-2 and KC), activated AMs (CD68⁺), and increased intra-alveolar infiltration of PMNs (Ly6G⁺). These effects of LPS are not strain dependent and occur in both room air- and CS-exposed mice. These results validate that we have established an LPS-induced mouse model of mild ALI.

Greater alveolar barrier permeability has been reported in human smokers (36). CS exposure also increases alveolar-capillary barrier permeability of guinea pigs (8, 38) and myocardial capillary permeability of rats (5). We have previously shown that a brief (6 h) exposure to CS, at a concentration similar to that of ambient air in a smoky bar, causes lung edema in C57BL/6 mice (33). In this study, we found that exposure to the same concentrations of CS for 3 h also increased lung wet-to-dry weight ratio in AKR mice. More prolonged (3 wk) exposure to CS increased BAL protein content in both C57BL/6 and AKR mice (see Table 1). These results suggest that CS exposure increases lung vascular permeability in vivo.

Table 1.

Effects of CS on lung injury in the absence of a second insult (CS vs. room air)

	C57BL/6		AKR
	3 h	3 wk	3 h	3 wk
Body weight		↓		↓↓
Cst	n.s	n.s	n.s	↑
BAL protein levels	n.s	↑	n.s	↑
Wet/dry	n.s		↑	n.s
Albumin-EBD				n.s
BAL cell counts	n.s	n.s	n.s	n.s
TNF-α	n.s	↑	n.s	↓
IL-10	↑	↑	n.s	↓
IL-6	n.s		n.s
MIP-2	n.s	n.s	n.s	n.s
KC	n.s	n.s	n.s	n.s
AM				↑
M2 macrophages				↑
PMN				n.s
Caspase-3 activation		n.s		n.s

CS, cigarette smoke; C_st, lung static compliance; BAL, bronchoalveolar lavage; EBD, Evans blue dye; TNF-α, tumor necrosis factor α; MIP-2, macrophage inflammatory protein 2; KC, keratinocyte-derived chemokine; AM, alveolar macrophages; PMN, polymorphonuclear neutrophils; n.s., not significant.

CS predisposes mice to bacterial pathogens by compromising innate and adaptive immunity (35). It has been reported that preexposure to CS for 2 wk augments mechanical ventilation-induced alveolar epithelial injury and airspace PMN influx in rats (23). Preexposure of hamsters to CS for 5 days significantly increases amiodarone-induced lung inflammation and parenchymal cell apoptosis (7). In contrast, CS has been shown to induce a temporary and reversible increase in clearance of lung bacterial pathogens (4). CS preexposure has also been reported to have a beneficial effect on lung inflammation and mortality in mice infected by H1N1 and H9N2 influenza viruses (20). These seemingly contradictory results may be due to different experimental models, different duration/concentrations of CS exposure, and/or different secondary insults. We found that LPS-induced increase in BAL protein content was significantly potentiated by either a brief (3 h) or prolonged (3 wk) CS preexposure in both strains of mice, with a greater effect of prolonged exposure in AKR mice (see Table 2). Prolonged (3 wk) CS preexposure also exacerbated the LPS-induced decrease in C_st, increase in lung water, lung vascular permeability (parenchymal extravasation of albumin-conjugated EBD), BAL total inflammatory cells, AM activation (CD68⁺), alveolar infiltration of PMNs (Ly6G⁺) and M2 macrophages (CD206⁺), and lung chemokines (MIP-2 and KC) in AKR mice (see Table 2). These results indicate that we have established and characterized a model of CS priming for LPS-induced ALI.

Table 2.

Effects of CS on lung injury after a second insult (CS/LPS vs. RA/LPS)

	C57BL/6		AKR
	3 h	3 wk	3 h	3 wk
Cst	n.s	n.s	n.s	↓
BAL protein levels	↑	↑	↑	↑↑
Wet/dry	n.s	n.s	↑	↑
Albumin-EBD				↑
BAL cell counts	n.s	n.s	↑	↑↑
TNF-α	n.s	↑	n.s	↓
IL-10	n.s	↑	n.s	n.s
IL-6	n.s		n.s
MIP-2	n.s	n.s	n.s	↑
KC	n.s	n.s	n.s	↑
AM				↑
M2 macrophages				↑
PMN				↑
Caspase-3 activation		n.s		n.s

In physiological conditions, C57BL/6 mice have a higher respiratory rate and a larger minute ventilation, despite a smaller tidal volume, than AKR mice (2). After acute exposure to CS, tidal volume was more suppressed in AKR mice compared with C57BL/6 mice, whereas respiratory rate was more suppressed in C57BL/6 mice (48). As a result, the plasma cotinine levels in C57BL/6 mice are higher than that of AKR mice when exposed to the same duration and concentrations of CS, suggesting that C57BL/6 mice inhale more CS than AKR mice (48). One would expect that C57BL/6 mice would be more sensitive to CS-induced lung edema and inflammation. In fact, in response to the same duration (6 h) and concentration of CS preexposure, we saw a substantial mortality in AKR mice, but not C57BL/6 mice, exposed to CS plus LPS. Our results indicate that AKR mice are more susceptible to CS-primed injury. Because of high mortality in AKR mice, we modified our protocol by decreasing exposure to CS to only 3 h instead of 6 h before LPS instillation. We found that 3 h of CS exposure did not significantly alter C_st, BAL protein levels, BAL inflammatory cells, TNF-α, IL-6, MIP-2, or KC in either strain of mice but significantly increased lung wet-to-dry weight ratio in AKR mice, suggesting that alveolar-capillary barrier of AKR mice is more susceptible to acute CS-induced disruption. Interestingly, acute (3 h) CS exposure elevated IL-10 levels in C57BL/6 mice but not in AKR mice. These results suggest that, although 3 h of CS exposure does not appear to cause significant lung inflammation in either strain of mice, it appears to enhance the anti-inflammatory response by elevating IL-10 expression in C57BL/6 mice. This anti-inflammatory response did not occur in susceptible AKR mice.

After a more prolonged period (3 wk) of CS exposure, AKR mice displayed a slower gain in body weight, along with a significant increase in C_st and activated AMs, as well as a decrease in TNF-α and IL-10. In contrast, C57BL/6 mice had a significant increase in TNF-α and IL-10 in response to these stimulations (see Tables 1 and 2). Our unpublished microarray data indicate that the expression of IL-10α and -β receptors was increased in C57BL/6 mice but decreased in AKR mice exposed to CS for 3 wk followed by challenge of LPS (data not shown). The balance between TNF-α and IL-10 is important for maintenance of immune homeostasis, as increase in TNF-α is counterbalanced by simultaneous synthesis of IL-10 (44). TNF-α was significantly reduced in AMs of guinea pigs and humans 20 h after CS exposure (15). Interestingly, TNF-α and IL-10 were also significantly reduced in lungs of AKR mice exposed to CS for 3 wk. Our data suggest an association of lower IL-10 expression with increased susceptibility to CS priming for ALI. Future studies are needed to clarify the mode of action through which the decrease in TNF-α and IL-10 may mediate CS-primed susceptibility of AKR mice to LPS-induced ALI. A variety of regulatory macrophages and lymphoid cell lineages, such as the T regulatory cells, have been reported to produce high levels of IL-10 and appear to contribute to the process of ALI/ARDS in other experimental insults (45, 46, 49). Future studies will be needed to clarify the mechanism and role of these select subpopulations in producing the prolonged CS-induced impairment of IL-10 release/signaling seen here.

Exposure to CS alone for 3 wk appears to increase lung AMs and M2 macrophages with less effect on lung MIP-2 and KC and intra-alveolar infiltration of PMNs in AKR mice, effects associated with increase in lung C_st. These data suggest that macrophages may be important players in development of an early emphysema-like state in AKR mice after exposure to CS alone for 3 wk. This notion is consistent with the report that AM-derived MMPs are key contributors to the development of emphysema (32).

In summary, we have characterized a brief and a prolonged exposure model of CS priming for lung edema and inflammation. Our data reveal an important difference in sensitivity to CS priming for ALI in mouse strains, where the AKR strain is more sensitive than the C57BL/6 strain. The key features we observed suggest that 3 wk of CS preexposure of AKR mice is a reproducible clinically relevant animal model that is useful for studying mechanisms and treatment of CS priming for a second hit-induced ALI. Our data also support the concept that increased susceptibility to ALI/ARDS is an important adverse health consequence of CS exposure that needs to be taken into consideration when treating critically ill individuals.

GRANTS

This study was supported with the use of facilities at the Providence VA Medical Center and by an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under grant number P20 GM103652 (S. Rounds; Project 1 to Q. Lu; Project 2 to J. Lomas-Neira) and other grants including T32 HL094300 (P. Sakhatskyy), RO1 HL130230 (Q. Lu), PULM-019-14F (S. Rounds), and P35 GM118097 (A. Ayala).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.S. and Q.L. conceived and designed research; P.S., Z.W., D.B., Y.C., and Q.L. performed experiments; P.S., Z.W., D.B., Y.C., and Q.L. analyzed data; P.S., Z.W., D.B., and Q.L. prepared figures; P.S., Z.W., D.B., J.L.-N., Y.C., A.A., S.R., and Q.L. approved final version of manuscript; J.L.-N., A.A., S.R., and Q.L. edited and revised manuscript; A.A., S.R., and Q.L. interpreted results of experiments; Q.L. drafted manuscript.