Expression profiling in vivo demonstrates rapid changes in lung microRNA levels following lipopolysaccharide-induced inflammation but not in the anti-inflammatory action of glucocorticoids

Abstract

Background: At present, nothing is known of the role of miRNAs in the immune response in vivo despite the fact that inflammation is thought to underlie multiple acute and chronic diseases. In these circumstances, patients are commonly treated with corticosteroids such as dexamethasone.

Results: To address this question, we have examined the differential expression of 104 miRNAs using real-time PCR during the innate immune response in mouse lung following exposure to aerosilised lipopolysaccharide (LPS). Following challenge, we observed rapid and transient increase in both the mean (4.3-fold) and individual levels of miRNA expression (46 miRNAs) which peaked at 3 hrs. Crucially, this increase was correlated with a reduction in the expression of tumour necrosis factor (TNF)-alpha, keratinocyte-derived chemokine (KC) and macrophage inflammatory protein (MIP)-2, suggesting a potential role for miRNAs in the regulation of inflammatory cytokine production. Examination of the individual miRNA expression profiles showed time dependent increases in miR-21, -25, -27b, -100, 140, -142-3p, -181c, 187, -194, -214, -223 and -224. Corticosteroid studies showed that pre-treatment with dexamethasone at concentrations that inhibited TNF-alpha production, had no effect either alone or upon the LPS-induced miRNA expression profile.

Conclusion: We have shown that the LPS-induced innate immune response is associated with widespread, rapid and transient increases in miRNA expression in the mouse lung and we speculate that these changes might be involved in the regulation of the inflammatory response. In contrast, the lack of effect of dexamethasone in either control or challenged animals implies that the actions of glucocorticoids per se are not mediated through changes in miRNAs expression and that LPS-induced increases in miRNA expression are not mediated via classical inflammatory transcription factors.

Figure 1

Time course of LPS-induced changes in BAL cytokines and cell infiltrates. Mice were exposed to aerosolised LPS and at the indicated time, the BAL levels of TNF-α (A), KC (B) MIP-2 (C), total leukocyte (D) and neutrophils (E) were determined. The values give are the mean ± SEM obtained from 6 mice.

Figure 2

Time-dependent changes in the overall miRNA expression levels. Mice pre-treated intraperitoneally for 1 hr with either saline or dexamethasone, were challenged with either aerosilised saline or LPS. At the indicated time, the mean ± SEM (n = 5) of the change in lung tissue levels for 104 mature miRNAs was determined relative to time-matched saline controls (equivalent to 1). Statistical significance was determined by ANOVA followed by Tukey's post-test where ** = p < 0.01 and *** = p < 0.001 versus saline and ### = p < 0.001 for dexamethasone-LPS versus LPS-only treated animals.

Figure 3

Hierarchically clustered (average linkage) heat map of the time course of LPS- and/or dexamethasone-induced changes in miRNA expression in mouse lung tissue. Mice were exposed to either aerosilised LPS, intraperitoneally administered dexamethasone or were pre-treated with dexamethasone (1 hr) prior to LPS challenge. At the indicated time points following LPS challenge, the levels of 104 mature miRNAs were determined using a TaqMan assay panel. The log₂ transformants of the fold-change in expression compared to time-matched saline controls are represented.

Figure 4

Time dependent, LPS-induced increases in mature miRNA expression levels. At the indicated times, the levels of individual mature miRNAs were determined by TaqMan and were expressed as the fold change ± SEM (n = 5) compared to time-matched saline controls (where saline controls are equivalent to 1). Statistical significance was determined by non-parametric ANOVA followed by Dunn's post-test where * p < 0.01 and ** p < 0.001.

Figure 5

LPS-induced changes in miRNA-223 expression in the mouse lung. Lungs were removed from saline-challenged (A-D) and LPS-challenged (E-H) mice at 3 hrs post-challenge and tissue slices were examined by in situ hybridisation using either a miRNA-223-specific (A, B, E, F) or scrambled (C, G) LNA probe, or histologically following cresyl violet staining (D, H). Bronchial (BE) and alveolar (AE) epithelia, as well as the vascular endothelium (VE) and leukocytes (L) are indicated.

Figure 6

Dexamethasone-mediated inhibition of LPS-induced changes in lung tissue TNF-α, BAL fluid TNF-α and BAL neutrophils. Mice were pre-treated either intraperitoneally (-1 hr – A) or orally (-0.5 hr – B/C) with dexamethasone, then exposed to aerosilised LPS and at the indicated time points the tissue levels of TNF-α (A) or BAL TNF-α (B) were determined by sandwich ELISA and inflammatory neutrophilia (C) was determined by differential cell counting. The values give are the means ± SEM obtained from 6 mice.