Progression of Acute Lung Injury in Intratracheal LPS Rat Model: Efficacy of Fluticasone, Dexamethasone, and Pirfenidone

Acute lung injury (ALI) is a clinical disease marked by respiratory failure due to the disruption of the epithelial and endothelial cell barrier, flooding of the alveolar compartment with protein-rich fluid, and recruitment of neutrophils into the alveolar space [1]. ALI affects over 10% of patients hospitalized receiving critical care, with acute respiratory distress syndrome being the most severe form of ALI with a 40% mortality rate [2].

LPS, also known as endotoxin, is the main component of the outer membrane of Gram-negative bacteria [3]. LPS binds to a specific LPS-binding protein (LBP), forming a complex that activates the CD14/TLR4 receptor structure on monocytes and macrophages. TLR4 activation triggers the inflammatory reaction cascade by the means of production of pro-inflammatory cytokines and amplifies lung injury [4]. LPS induces features of ALI such as pulmonary inflammation, especially neutrophilic inflammation, endothelial and alveolar epithelial cell damage, pulmonary edema, and increased inflammatory mediators [5]. Therefore, the LPS-induced ALI model is extensively used for preclinical evaluations of anti-inflammatory drug candidates in pharmaceutical research. Investigators have used various routes/methods in rats to induce ALI with following LPS concentrations: 0.1 mg/mL for 40 min by aerosol [6], 1 mg/mL for 40 min by aerosol [7], 2.6 mg/kg thoracic dose by aerosol [8], 7.5 mg/kg intravenous [9], 3–30 mg/kg intraperitoneal [4,10‒12], and 1–8 mg/kg intratracheal (i.t.) [5, 8, 13, 14]. However, the effect of lower doses of LPS (<1 mg/kg) on ALI induction in rats has not been reported.

In this paper, we looked at the effects of i.t. LPS in the rat over time at multiple doses. We followed the kinetics of ALI progression at an optimized LPS concentration to better understand disease development and used pharmacological agents to verify the model and their efficacy.

Pirfenidone, a small-molecule drug with combined anti-inflammatory, antioxidant, and antifibrotic effects, has been shown to inhibit the progression of fibrosis in animal models and in patients with idiopathic pulmonary fibrosis [15‒19] and is known for anti-fibrotic, antioxidant, and anti-inflammatory effects [16, 20]. When administered perorally (p.o.), it was effective in reducing LPS-induced pulmonary inflammation based on neutrophil counts, TNFa level, and IL6 level [21]. However, to our knowledge, there are no previous reports on testing pirfenidone i.t. in a LPS rat model. Therefore, we wanted to investigate the potential of local administration of pirfenidone i.t. within the lung to improve ALI. We hypothesized that providing maximum availability of pirfenidone within the lung could improve efficacy. The corticosteroids fluticasone administered i.t. and dexamethasone given p.o. effectively ameliorate LPS-induced inflammation in rodent models [22‒25].

Reproducible preclinical rodent models with defined phenotypes that mimic human diseases are vital to understand the pharmacological effects of new therapeutics and to decide on future development for clinical trials. We report that 10–300 µg/rat LPS i.t. produced a rat ALI phenotype that is characterized by profound recruitment of neutrophils in the lung, pulmonary edema, and the involvement of inflammatory mediators. Using this model, we show that standard glucocorticoids p.o. and i.t. greatly ameliorate ALI at the 4 h timepoint in contrast to the antifibrotic pirfenidone i.t. Furthermore, we determined effects of LPS on various features of ALI at various timepoints (0–24 h) post-administration and showed a time-dependent progression of lung injury.

Animal experiments were approved by the Institutional Animal Care and Use Committee of PRISM. Female Sprague-Dawley rats weighing 200–220 g (Envigo, USA) were used after 7 days of acclimatization under pathogen-free conditions. Food and water were available ad libitum.

LPS was obtained from Sigma-Aldrich, USA, and dexamethasone, fluticasone propionate, and pirfenidone (=98.0% purity) were purchased from VWR, USA. LPS solutions were prepared in 1X PBS (pH 7.4). Dexamethasone was triturated with 0.1% Tween-80, and the volume for oral administration was adjusted in 0.5% methylcellulose solution. Fluticasone propionate and pirfenidone solutions for i.t. administration were prepared with 1X PBS (pH 7.4) containing 0.1% Tween using sonication. Rat Quantikine ELISA kits for IL6, IL1ß, TNFa, CINC-1, TGFß1, and TIMP-1 were purchased from R&D Systems (Minneapolis, MN, USA).

The rats were anesthetized with isoflurane (3–5%), the trachea was located using a laryngoscope, and ALI was induced by single i.t. challenge with 10, 30, 100, and 300 µg of LPS in 300-µL PBS using a needle (18 gauge) attached to Tuberculin syringe (n = 5–7/each). Control rats (NC) received 300 µL of PBS (n = 5). Animals were sacrificed at 4 h to obtain endpoint measurements. To assess the kinetics of ALI induced by LPS (10 µg/rat i.t.), animals were sacrificed at 2, 4, 6, 8, 10, and 24 h (n = 6/each). NC animals were sacrificed at 0 h.

For pharmacological interventions, animals were divided into following groups: NC, LPS 10 µg/rat i.t. challenge, LPS + fluticasone 100 µg i.t., and LPS + fluticasone 300 µg i.t. (n = 5 per each) OR NC, LPS 10 µg/rat treated, LPS + dexamethasone 1 mg/kg, p.o. OR NC, LPS 10 µg/rat treated, LPS + pirfenidone 4,000 µg i.t. (n = 5/each). The animals received respective treatments of fluticasone and pirfenidone or 200 µL of vehicle i.t. at -18 h and -1 h prior to LPS challenge. Dexamethasone at 1 mg/kg p.o. was given -1 h prior to LPS challenge. Animals were euthenized 4-h post-LPS challenge to obtain endpoint measurements.

At the indicated timepoints, animals were euthanized with Euthasol (100–120 mg/kg i.p.) and the trachea was exposed. Bronchoalveolar lavage (BAL) was performed three times with a plastic cannula using 2 mL 1X PBS. BAL aspirate was pooled, and equal volumes of BAL fluid (BALF) and Turk’s solution were mixed for total leukocyte count using a hemacytometer (Hausser Scientific Horsham, PA, USA). The remaining BALF was centrifuged at 4000 RPM for 5 min at 4°C, and the supernatant was stored at -80°C for cytokine and biomarker detection. Cell pellets were mixed with rat serum, and smears for differential cell counts were prepared on frosted glass slides. Total differential cell counts were evaluated manually at a magnification of ×100 using light microscopy (BX2, Olympus, Tokyo, Japan) by counting and categorizing 500 cells stained with Leishman stain according to their morphology into macrophages, lymphocytes, and neutrophils.

After BAL, whole lungs were harvested from rats, washed in 1X PBS, and stored at -80°C. Pulmonary edema was assessed by recording lung weights, lung weight to body weight ratio, and estimating BALF protein content. Briefly, the lungs were cleaned in 1X PBS, debris was removed, blotted with tissue papers, and the weight of the lungs was recorded. Protein content in BALF supernatants was measured using the bicinchoninic acid (BCA) assay and expressed in µg/mL of BALF.

The levels of cytokine/chemokines levels in BALF were determined using commercially available ELISA assays according to manufacturer’s instructions. The level of each biomarker is expressed in pg/mL of BALF.

All data were presented as the mean ± standard error of the mean (SEM). The data were analyzed by using one-way ANOVA followed by Dunnett’s test for multiple group comparisons or unpaired Student’s t test to compare two groups. For dose and kinetic studies, all groups were compared to control animals which received PBS. In the pharmacological intervention studies, all groups were compared to LPS (10 µg/rat i.t.)-treated controls. p value <0.05 was set as statistically significant.

We tested the effect of 10, 30, 100, and 300 µg LPS i.t./rat on pulmonary inflammation, edema, and inflammatory mediators at 4 h. All LPS doses induced pulmonary inflammation, as reflected in the significant (p < 0.001) increase in leukocyte and neutrophil count compared to NC (Fig. 1a, b), whereas no change in lymphocytes and macrophages was seen (data not shown). A significant increase (p < 0.001) in BALF protein content, lung weight, and lung weight to body weight ratio was observed at all doses of LPS-treated rats compared to NC (Fig. 1c, d; online suppl. Fig. 1A; for all online suppl. material, see https://doi.org/10.1159/000534329). Concomitantly, levels of IL6, IL1ß, TNFa, and CINC-1 significantly increased in the BALF of LPS-treated rats compared to NC (Fig. 1e–h). Generally, a dose of 300 µg/rat is used for mechanistic and pharmacological studies. But, since all tested doses of LPS induced robust and well-accepted features of ALI in rats, we chose a concentration of 10 µg/rat for further experiments.

LPS 10 µg, i.t., caused a significant (p < 0.001) increase of leukocytes, neutrophils, BALF protein content, lung weight, and the lung weight to body weight ratio compared with the control group. The BALF levels of IL6, IL1ß, TNFa, and CINC-1 were found to have significantly (p < 0.001) increased in the LPS group compared with the NC (Fig. 2a–h; online suppl Fig. 1B). These results indicate the reproducibility of our previous finding, and therefore 10 µg LPS was used for all subsequent pharmacological interventions and time-dependent study of ALI progression.

Fluticasone is a well-known, potent anti-inflammatory agent. In an exploratory study to select fluticasone i.t. doses, we tested 30 and 100 µg/rat (-1 h). Though we chose reportedly efficacious doses of fluticasone, we saw no nonsignificant effects on leucocytes and neutrophils compared to LPS controls (online suppl. Fig. 2). We then tested the effect of fluticasone 100 and 300 µg/rat (-1 h and -18 h) on the experimental features of ALI induced by 10 µg/rat LPS at 4 h post-challenge. Pretreatment with fluticasone at both concentrations significantly (p < 0.001) reduced leukocytes by 70–78% and neutrophils by 60–63% compared to LPS control (Fig. 3a,b, 4a). As shown in Figure 3c, d and online supplementary Figure 1C, fluticasone-attenuated LPS-induced increase in protein content, lung weight, and the lung weight to body weight ratio =90% (p < 0.001). Fluticasone treatment additionally inhibited LPS-induced IL6, IL1ß, and TNFa levels by 70–90% (p < 0.001) and CINC-1 levels by 26–35% compared to LPS control (Fig. 3e–h, 4a).

We tested the effect of dexamethasone p.o. treatment on the experimental features of ALI induced by 10 µg/rat LPS at 4 h post-challenge. Pretreatment with dexamethasone significantly (p < 0.001) decreased pulmonary inflammation and pulmonary edema by = 65%. Furthermore, pretreatment with dexamethasone significantly (p < 0.001) reduced BALF levels of IL6, IL1ß, and TNFa by = 75% and levels of TIMP-1 by 57% compared with the LPS group (Fig. 5a–h, 4b).

The results of pharmacological intervention using fluticasone and dexamethasone confirm that the experimental conditions were well controlled and validate this model. We used these controlled experimental conditions to further test the effects of pirfenidone on ALI disease progression.

We selected a staring dose of 3,000 µg i.t. of pirfenidone for pharmacological intervention which resulted in some, though non-significant, efficacy in leukocyte and neutrophil inhibition (online suppl. Fig. 2). We then tested the effect of pirfenidone (4,000 µg i.t./rat, -1 h, -18 h) on the experimental features of ALI induced by 10 µg/rat LPS at 4 h post-challenge. In the pirfenidone pretreatment group, pulmonary inflammation, edema, BALF levels of IL1ß readouts were not significantly different compared to LPS controls. The levels of BALF IL6, TNFa, and CINC-1 significantly (p < 0.05) increased due to pirfenidone treatment compared to LPS (Fig. 6a–h, 4c).

To understand the kinetics of LPS-induced ALI, we determined the effect of LPS (10 µg i.t./rat) on experimental features of ALI at 2, 4, 6, 8, 10, and 24 h post-challenge and compared with the 0 h timepoint. Leukocyte counts significantly (p < 0.001) increased after 4 h–24 h (Fig. 7a). The majorities of cells involved in pulmonary inflammation were neutrophils (90–95%) (Fig. 7b, c), and significant (p < 0.001) neutrophil increase was observed at all timepoints (Fig. 7b). Similarly, the BALF protein content, lung weight, and index readouts also significantly (p < 0.001) increased between 2 and 24 h during the study duration (Fig. 7d–f). The levels of IL6 in BALF significantly (p < 0.001) increased between 2 and 10 h post-challenge and returned to control levels at 24 h (Fig. 8a). The IL1ß and TNFa levels showed significant (p < 0.05) increase in BALF due to LPS starting at 2 h and lasting until 24 h (Fig. 8b, c). Furthermore, a significant (p < 0.05, p < 0.001) increase of BALF CINC-1 and TIMP1 was noticed starting at 2 h (Fig. 8d, f). TGFß-1 in BALF significantly (p < 0.05) increased only during the 4–8 h timepoints (Fig. 8e). These changes in experimental features of ALI over time post-LPS challenge indicate a progression of i.t. LPS-induced ALI.

ALI is a disorder of acute inflammation characterized by the loss of alveolar-capillary membrane integrity, excessive neutrophil infiltration, pulmonary edema, and the release of proinflammatory mediators [26]. Controlled experimental conditions that produce reliable disease model phenotypes are important for meaningful preclinical evaluations of any novel therapeutics under investigation. Here, we have established dose effect of i.t. LPS 10–300 µg/rat on features of ALI, validated and characterized ALI phenotype and progression induced by i.t. LPS 10 µg/rat, and refined the LPS model for accurate ALI phenotype assessment as well as preclinical efficacy evaluations.

The American Thoracic Society has identified features in rodent models of ALI that are widely accepted as relevant readouts for exploring human ALI, includes neutrophilic lung inflammation, pulmonary edema formation, and changes in biochemical markers of lung injury [27, 28]. Neutrophil infiltration and inflammation are major contributors to tissue damage in ALI [29]. They are the dominant leukocyte type found both in BALF and in histological specimens from patients with ALI [30]. We clearly demonstrated that LPS at doses ranging from 10 to 300 µg per rat induced significant lung inflammation, predominantly due to neutrophil infiltration in the lung. One hallmark of ALI is apparent increase in barrier permeability and edema formation [31]. In animal models, increased lung weight/index, lung wet-to-dry ratio, and BALF protein content are used as indicators of edema [28]. We found significant formation of pulmonary edema in LPS 10–300 µg challenged animals with increased lung weight/index and increased BALF protein content.

Proinflammatory cytokines IL6, IL1ß, TNFa, and CINC-1 have been implicated as mediators of LPS-induced airway inflammation in rodents [32, 33]. IL6 drives signaling via STAT3 and possibly through the upregulation of TGF-ß1 and MIP1a [34]. IL1ß is involved in ALI by inducing neutrophil recruitment and activation and increasing vascular permeability via the integrin pathway [35]. TNFa plays a central role in the development of ALI by acting locally to stimulate chemotaxis, recruitment, and activation of neutrophils. TNFa induces apoptosis in alveolar epithelial lung microvascular endothelial cells and promotes edema [36]. CINC-1, a member of the CXC chemokine family, plays a pivotal role in neutrophil migration in rats, and in vivo inhibition of CINC-1 reduced inflammation in a LPS rat model [14]. These cytokines are potentially important mediators of LPS-induced lung inflammation. In line with these findings, we showed that 10–300 µg LPS significantly and to a similar degree increased levels of IL6, IL1ß, CINC-1 in BALF. An increase of TNFa was seen at 100–300 µg LPS which was greater than at 10–30 µg LPS. Our kinetic studies show that the release of TNFa after LPS induction is transient (2 h). The dose-effect study suggests that the release of TNFa is also partly LPS concentration dependent compared to the release of IL6, IL1ß, and CINC-1 mediators under similar experimental conditions.

Notably, all doses of LPS (10–300 µg) exhibited profound induction of experimental features of ALI, but no LPS-dose-dependent effect was observed. This suggests that the maximum response at 4 h is already reached at all LPS concentrations we tested. This does not exclude the possibility of differences at later timepoints. Because the 10 µg/rat dose of LPS induced robust and reproducible ALI in rats at 4 h, we used it in our drug treatment studies and time-dependent ALI investigations.

Glucocorticoids are well-known, potent anti-inflammatory agents [37]. Fluticasone has been reported to inhibit neutrophil influx in a lung injury model at doses ranging from 10 to 1,000 µg/rat i.t. with an ED50 value of 30 µg/rat [38]. In our data, i.t. fluticasone (100 and 300 µg/rat) inhibited 70–78% of the LPS-induced inflammation by blocking 60–63% influx of neutrophils, when given prophylactically. Moreover, fluticasone showed profound inhibition of pulmonary edema with improved lung weight-to-body weight ratio and reduction in BALF protein content of =90%, suggesting complete absence of pulmonary edema formation. Inhibition of cytokine production by glucocorticoids is believed to participate in the anti-inflammatory effect [39]. Pretreatment of rats with fluticasone inhibited 70–90% of IL6, IL1ß, and TNFa as well as 25–35% of CINC-1 induced by LPS. We observed a correlation between the inhibition of cytokines and the inhibition of inflammation and pulmonary edema. In our observations, fluticasone showed similar efficacy in the range of 60–100% on endpoints except CINC-1 evaluated in this study, suggesting lower doses of fluticasone can be effective in this model. Similarly, dexamethasone showed 60–80% inhibition of inflammation and edema, a 70–90% inhibition of IL6, IL1ß, and TNFa, and a moderate 57% inhibition of TIMP-1. Surprisingly, the antifibrotic compound pirfenidone exhibited no efficacy on inflammation and edema. Also, pirfenidone increased IL6, TNFa, and CINC-1 levels. This may be due to local availability of pirfenidone. Although it has beneficial effects in lung fibrosis and exerts anti-inflammatory effects which might affect ALI, our data suggest no effect in the i.t. ALI rat model when pirfenidone is given i.t. We might need to optimize doses and suitable formulation to achieve efficacy. Our data of pharmacological interventions suggest that IL6, IL1ß, TNFa, CINC-1, and TIMP-1 are involved in the inflammatory response of ALI which is consistent with previous reports [28, 40].

Lastly, we performed a kinetic study of ALI features induced 0, 2, 4, 6, 8, 10, and 24 h post-LPS (10 µg/rat) challenge. In our study, within 4 h we saw a substantial increase in pulmonary inflammatory cells (predominantly neutrophils) with a sustained peak at 6–10 h. This pulmonary neutrophilia remained up to 24 h after. Additionally, we found significantly increased BALF protein content, lung weight, and lung weight to body weight ratio suggesting pulmonary edema which peaked at 2–4 h and remained elevated for 24 h suggesting that LPS induces progressive pulmonary inflammation and sustained pulmonary edema until 24 h. When we evaluated a set of BALF cytokines that are well-established inflammatory markers, the levels of IL6 in BALF gradually increased starting from 2 h, peaked between 6 and 8 h post-challenge, and then gradually declined to baseline levels seen at 24 h. BALF CINC-1 levels peaked at 2 h, remained elevated up to 6 h, and then dramatically declined to remain at same levels until 24 h post-LPS challenge. IL1ß showed similar kinetics as CINC-1. TNFa also dramatically increased to peak at 2 h and then gradual decreased up to 8 h, after which the TNFa levels returned to baseline. Our data suggest that BALF IL6, IL1ß, CINC-1, and TNFa contribute at various time points to LPS-induced ALI progression. In addition, we also estimated levels of TGFß and TIMP1. TGFß mediates neutrophil recruitment and edema development through the TGFß receptor/integrin pathway in the LPS ALI model [41, 42]. TIMP1 is also implicated in LPS-induced ALI, and anti-IL17 has been reported to inhibit recruitment of neutrophils by reducing the expression of TIMP1 in lung parenchyma [43]. Similar to the other cytokines, BALF TGFß levels gradually increased and peaked at 8 h, decreased thereafter, and recovered at 24 h post-LPS challenge. Our data suggest TGFß alone has a transient role in the LPS induced ALI rat model at a 10 µg/rat dose. BALF TIMP1 levels gradually increased from 2 h, peaked at 8–10 h, and remained elevated up to 24 h post-LPS challenge. BALF IL6, CINC-1, IL1ß, TNFa, and TGFß showed involvement in progression of injury mostly between 2 and 8 h post-LPS challenge, whereas BALF TIMP1 showed a completely different pattern. TIMP-1 levels neither declined after the peak nor returned to baseline at 24 h which infers that it may have a persistent role during LPS-induced ALI progression in rats. Overall, LPS induces progressive pulmonary inflammation and pulmonary edema which can be attributed to changes in inflammatory cytokines and chemokines at specific and dynamic time points during ALI progression.

Overall, we achieved a robust, profound, and reproducible phenotype with features of experimental ALI at all LPS concentrations tested including 10 µg/rat (0.05 mg/kg) which is 20–160 times lower than commonly used (1–8 mg/kg). The results of our pharmacological studies with fluticasone and dexamethasone validate our model. Also, differential pharmacological effects are confirmed in our report based on null efficacy of i.t. pirfenidone in comparison with efficacious fluticasone and dexamethasone. Our kinetic study provides insights on the progressive nature of inflammation and sustainable pulmonary edema up to 24 h, whereas the contribution of mediators is dynamic during ALI progression post-LPS (i.t.) challenge. Our model characterization was limited to predetermined ALI features and warrants future histopathology and lung function tests.

The results of this study provided insights on a 10–300 µg i.t. LPS-induced ALI model in rats validated by testing the effects of corticosteroids fluticasone and dexamethasone on ALI features. Furthermore, we have described the progression of ALI during the first 24 h after LPS induction using generally recognized outcome measures. We believe this model will be useful for designing future experiments to develop and test therapeutics for ALI.

This study protocol was reviewed and approved by PRISM IACUC, approval numbers 14-01 and 21-03.

The authors have no conflicts of interest to declare.

This study was supported by the National Institutes of Health (https://www.nih.gov/grants-funding) through grants awarded to JES (P01HL119165 and R01HL169760). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conceptualization, writing – original draft, and methodology: Anil H. Kadam and Jan E. Schnitzer. Funding acquisition: Jan E. Schnitzer. Investigation: Anil H. Kadam.

All relevant data are within the manuscript. Further inquiries can be directed to the corresponding author.