DROSHA-Dependent AIM2 Inflammasome Activation Contributes to Lung Inflammation during Idiopathic Pulmonary Fibrosis

Idiopathic pulmonary fibrosis (IPF) has been linked to chronic lung inflammation. Drosha ribonuclease III (DROSHA), a class 2 ribonuclease III enzyme, plays a key role in microRNA (miRNA) biogenesis. However, the mechanisms by which DROSHA affects the lung inflammation during idiopathic pulmonary fibrosis (IPF) remain unclear. Here, we demonstrate that DROSHA regulates the absent in melanoma 2 (AIM2) inflammasome activation during idiopathic pulmonary fibrosis (IPF). Both DROSHA and AIM2 protein expression were elevated in alveolar macrophages of patients with IPF. We also found that DROSHA and AIM2 protein expression were increased in alveolar macrophages of lung tissues in a mouse model of bleomycin-induced pulmonary fibrosis. DROSHA deficiency suppressed AIM2 inflammasome-dependent caspase-1 activation and interleukin (IL)-1β and IL-18 secretion in primary mouse alveolar macrophages and bone marrow-derived macrophages (BMDMs). Transduction of microRNA (miRNA) increased the formation of the adaptor apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC) specks, which is required for AIM2 inflammasome activation in BMDMs. Our results suggest that DROSHA promotes AIM2 inflammasome activation-dependent lung inflammation during IPF.


Immunohistochemistry and Immunofluorescence Analysis
For immunohistochemistry analysis, lung tissues were fixed overnight in buffered 10% formaldehyde, embedded in paraffin, and sectioned at a thickness of 4 mm. The sections were stained with hematoxylin and eosin (H&E) using the Hematoxylin and Eosin Stain Kit (H-3502, Vector Laboratories) according to the manufacturer's protocol. The sections were stained with antibody against specific targets. Secondary antibody was horseradish peroxidase (HRP)-conjugated anti-mouse (SC-516102, SantaCruz Biotechnology), HRP-conjugated anti-rabbit (SC-2357, SantaCruz Biotechnology), HRP-conjugated anti-rat (SC-2750, SantaCruz Biotechnology), Goat anti-Mouse IgG1 Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (A-21121, Thermo Fisher Scientific), and Goat anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 555 (A-21429, Thermo Fisher Scientific). Chromogenic detection was performed using DAB Peroxidase (HRP) Substrate Kit (SK-4100, Vector Laboratories) according to the manufacturer's protocol. Lung sections were counterstained with Hematoxylin Solution, Harris Modified (HHS32, Sigma). For lung fibrosis analysis, Masson's trichrome (M/T) staining was performed using trichrome stain kit (ab150686, Abcam) according to the manufacturer's instructions. Stained lung sections were analyzed by Olympus BX53M microscope and quantified by using Olympus Stream software and ImageJ software v1.52a (Bethesda, MD, USA). For immunofluorescence analysis, lung sections were stained with antibody against specific targets. Lung sections were analyzed using a Zeiss LSM880 laser scanning confocal microscope.

ASC Speck Formation Assay
The WT alveolar macrophages and BMDMs were seeded on chamber slides. After LPS and poly(dA:dT) stimulation, cells were fixed with 4% paraformaldehyde and then incubated with polyclonal ASC antibody (ADI-905-173-100, Enzo Lifesciences) for 16 h and FITC goat anti-rabbit (IgG) secondary antibody (ab6717, Abcam (Cambridge, MA, USA)) for 1h followed by DAPI (P36962, ThermoFisher Scientific) staining [41][42][43]. The ASC specks were analyzed using a Zeiss LSM880 laser scanning confocal microscope and quantified using ImageJ software v1.52a (Bethesda, MD, USA). The graph in figure represents the quantification of percent of ASC speck-positive cells for each mouse.

Statistical Analysis
All data are mean ± SD, combined from three independent experiments. All statistical tests were analyzed using a two-tailed Student's t-test for comparison of two groups, and analysis of variance (ANOVA) (with post hoc comparisons using Dunnett's test), using a statistical software package (GraphPad Prism version 4.0, GraphPad Software Inc. (San Diego, CA, USA)) for comparison of multiple groups.

The DROSHA and AIM2 Protein Levels were Elevated in Lung Tissues of Patients with IPF
To investigate the role of DROSHA during lung fibrosis in patients with IPF, we analyzed whether the DROSHA protein levels were elevated in lung tissues from patients with IPF (Table 1). We first measured lung fibrosis by Masson's trichrome staining (M/T) in lung tissues from patients with IPF (IPF) or non-IPF patients (Control) ( Figure 1A). Lung fibrosis is highly increased in patients with IPF (IPF) relative to non-IPF patients (Control) ( Figure 1A). We next measured the DROSHA protein levels in lung tissues from patients with IPF (IPF) or non-IPF patients (Control) using immunohistochemistry staining ( Figure 1B). Notably, immunohistochemistry staining revealed the DROSHA protein expression was significantly elevated in alveolar macrophages of patients with IPF (IPF) compared to non-IPF patients (Control) ( Figure 1B). In contrast to DROSHA, the DGCR8 protein expression was comparable ( Figure  S1). Consistent with immunohistochemistry staining, the DROSHA protein levels were significantly increased in lung tissues from patients with IPF (IPF) compared to non-IPF patients (Control) ( Figure 1C), whereas the DGCR8 protein levels were unchanged ( Figure 1C). Next, we investigated whether the AIM2 inflammasome was increased in lung tissues from patients with IPF. We measured the AIM2 protein levels in lung tissues from patients with IPF. Similar to DROSHA expression, the AIM2 protein levels were significantly increased in lung tissues from patients with IPF (IPF) compared to non-IPF patients (Control) ( Figure 1D). These results suggest that the DROSHA and AIM2 protein levels were elevated in patients with IPF.

The DROSHA and AIM2 Expression Levels were Elevated in Alveolar Macrophages of Patients with IPF
Next, we investigated the role of DROSHA in AIM2 inflammasome-dependent lung inflammation during IPF. We first analyzed whether the DROSHA expression levels were increased in alveolar macrophages of patients with IPF using immunofluorescence staining. We measured the changes of DROSHA and AIM2 expression levels in cluster of differentiation 68 (CD68)-positive alveolar macrophages [44] of patients with IPF (IPF) and non-IPF patients (Control). Immunofluorescence staining revealed that the intensity and number of DROSHA-positive staining in CD68-positive alveolar macrophages were increased in patients with IPF (IPF) relative to non-IPF patients (Control) ( Figure 2A). Next, we investigated whether the AIM2 expression levels were elevated in CD68-positive alveolar macrophages of patients with IPF using immunofluorescence staining ( Figure 2B). Consistent with DROSHA expression, the intensity and number of AIM2-positive staining in CD68-positive alveolar macrophages were increased in patients with IPF (IPF) compared to non-IPF patients (Control) ( Figure 2B). Moreover, we examined whether the positive subcellular co-localization between DROSHA and AIM2 is elevated in patients with IPF. DROSHA is co-localized with AIM2 in patients with IPF (IPF) ( Figure 2C). Notably, the intensity and number of cells which have the positive subcellular co-localization between DROSHA and AIM2 were significantly increased in patients with IPF (IPF) relative to non-IPF patients (Control) ( Figure 2C). These results suggest that the DROSHA and AIM2 expression levels in alveolar macrophages were elevated in patients with IPF. non-IPF patients (Control, n = 12). For immunoblots, α-tubulin or β-actin was used as loading control. Data are representative of three independent experiments. Data are mean ± SEM. *** p <0.001, * p <0.05; by Student's two-tailed t-test.  We investigated the role of DROSHA in alveolar macrophages in a mouse model of bleomycininduced pulmonary fibrosis. We analyzed whether the DROSHA expression levels were elevated in alveolar macrophages of lung tissues from a mouse model of bleomycin-induced pulmonary fibrosis using immunohistochemistry staining. Notably, the density and number of DROSHA-positive staining in alveolar macrophages were increased in mice treated with bleomycin (Bleomycin) compared to mice treated with vehicle control (PBS) ( Figure 3A). In contrast, the DGCR8 expression levels were comparable ( Figure S2). Consistently, the DROSHA protein levels were elevated in lung tissues of mice treated with bleomycin (Bleomycin) relative to mice treated with vehicle control (PBS) ( Figure 3B). In contrast, the DGCR8 protein levels were unchanged ( Figure 3B). Moreover, we examined whether the DROSHA expression levels in alveolar macrophages were increased in mice treated with bleomycin using immunofluorescence staining. We analyzed the DROSHA expression levels in F4/80-positive alveolar macrophages [45] of mice treated with bleomycin (Bleomycin) or vehicle control (PBS). The intensity and number of DROSHA-positive staining in F4/80-positive alveolar macrophages were increased in lung tissues of mice treated with bleomycin (Bleomycin) relative to mice treated with vehicle control (PBS) ( Figure 3C). These results suggest that the DROSHA protein levels in alveolar macrophages were elevated in bleomycin-induced pulmonary fibrosis. We investigated the role of DROSHA in alveolar macrophages in a mouse model of bleomycininduced pulmonary fibrosis. We analyzed whether the DROSHA expression levels were elevated in alveolar macrophages of lung tissues from a mouse model of bleomycin-induced pulmonary fibrosis using immunohistochemistry staining. Notably, the density and number of DROSHA-positive staining in alveolar macrophages were increased in mice treated with bleomycin (Bleomycin) compared to mice treated with vehicle control (PBS) ( Figure 3A). In contrast, the DGCR8 expression levels were comparable ( Figure S2). Consistently, the DROSHA protein levels were elevated in lung tissues of mice treated with bleomycin (Bleomycin) relative to mice treated with vehicle control (PBS) ( Figure 3B). In contrast, the DGCR8 protein levels were unchanged ( Figure 3B). Moreover, we examined whether the DROSHA expression levels in alveolar macrophages were increased in mice treated with bleomycin using immunofluorescence staining. We analyzed the DROSHA expression levels in F4/80-positive alveolar macrophages [45] of mice treated with bleomycin (Bleomycin) or vehicle control (PBS). The intensity and number of DROSHA-positive staining in F4/80-positive alveolar macrophages were increased in lung tissues of mice treated with bleomycin (Bleomycin) relative to mice treated with vehicle control (PBS) ( Figure 3C). These results suggest that the DROSHA protein levels in alveolar macrophages were elevated in bleomycin-induced pulmonary fibrosis.  wild-type (WT) mice were exposed to PBS or bleomycin via oropharyngeal aspiration. Positive area and cells are indicated by the black arrow. Scale bars, 200 µm. Results are representative of three independent experiments. (B) Representative immunoblot analysis for DROSHA and DGCR8 (left) and densitometry quantification of DROSHA and DGCR8 levels (normalized to levels of β-actin) (right) in lung tissues from WT mice exposed to PBS (n = 3) or bleomycin (n = 5) via oropharyngeal aspiration. For immunoblots, β-actin was used as loading control. Data are representative of three independent experiments. Data are mean ± SEM. * p <0.05; by Student's two-tailed t-test. (C) Representative immunofluorescence image of F4/80 (Red), DROSHA (Green), and DAPI (Blue) staining in lung tissues from WT mice exposed to PBS or bleomycin via oropharyngeal aspiration. Positive area and cells are indicated by white arrows. Scale bars, 200 µm. Quantification of co-localization positive cells between DROSHA and F4/80 (the percent of co-localization positive cells in total 100 cells in 10 individual images per group) (right) in lung tissues from WT mice exposed to bleomycin (n = 5) or PBS (n = 3) via oropharyngeal aspiration. Data are mean ± SEM. ** p <0.01, * p <0.05; by Student's two-tailed t-test.

Deficiency of DROSHA Suppresses the AIM2 Inflammasome Activation in Alveolar Macrophages
Since DROSHA and AIM2 expression levels were elevated in alveolar macrophages during lung fibrosis, we next investigated the function of DROSHA during AIM2 inflammasome activation in alveolar macrophages. We analyzed whether the deficiency of DROSHA could suppress the secretion of IL-1β and IL-18 in lipopolysaccharide (LPS)-primed WT alveolar macrophages in response to poly(dA:dT), an AIM2 inflammasome activator. We used DROSHA-targeted guide RNA (Drosha gRNA) to delete mouse DROSHA in primary mouse alveolar macrophages via clustered regularly interspaced short palindromic repeats (CRISPR) technology ( Figure 4A). The DROSHA-targeted gRNA suppressed the secretion of IL-1β and IL-18 compared to control plasmid (control), whereas the secretion of tumor necrosis factor (TNF)-α was unchanged ( Figure 4A). Consistently, the DROSHA-targeted gRNA inhibited the activation of caspase-1 and IL-1β cleavage relative to control plasmid ( Figure 4B). In contrast, the DROSHA-targeted gRNA did not change on the secretion of IL-1β and IL-18 in response to ATP (a NLRP3 inflammasome activator), flagellin (an NLRC4 inflammasome activator), or muramyldipeptide (MDP) (an NLRP1 inflammasome activator) ( Figure 4C). Moreover, we investigated whether DROSHA-targeted gRNA could suppress the formation of ASC specks, which is required for AIM2 inflammasome activation [41,42]. Notably, the DROSHA-targeted gRNA significantly reduced the formation of ASC specks by LPS and poly(dA:dT) stimulation compared to the control plasmid ( Figure 4D). In contrast to alveolar macrophages, LPS and poly(dA:dT) stimulation did not changes the activation of primary mouse lung fibroblasts ( Figure S3). These results suggest that deficiency of DROSHA suppresses the AIM2 inflammasome activation in alveolar macrophages.

Deficiency of DROSHA Suppresses the AIM2 Inflammasome Activation in Macrophages
We next examined whether the deficiency of DROSHA can suppress the AIM2 inflammasome activation in bone marrow-derived macrophages (BMDMs). We used DROSHA-targeted small interfering RNA (Drosha siRNA) to delete mouse DROSHA in primary mouse BMDMs ( Figure 5A). Consistent with deficiency of DROSHA in alveolar macrophages (Figure 4), the Drosha siRNA significantly suppressed the secretion of IL-1β and IL-18 relative to control siRNA (Control siRNA), whereas the secretion of TNF-α was unchanged ( Figure 5A). Moreover, the Drosha siRNA reduced the activation of caspase-1 and IL-1β cleavage relative to control siRNA ( Figure 5B). In contrast, the Drosha siRNA did not change on the secretion of IL-1β and IL-18 in response to ATP, flagellin, or MDP compared to control siRNA ( Figure 5C). Furthermore, the Drosha-siRNA significantly inhibited the formation of ASC specks by LPS and poly(dA:dT) stimulation relative to the control siRNA ( Figure 5D). These results suggest that deficiency of DROSHA suppresses the AIM2 inflammasome activation in macrophages.  the activation of caspase-1 and IL-1β cleavage relative to control siRNA ( Figure 5B). In contrast, the Drosha siRNA did not change on the secretion of IL-1β and IL-18 in response to ATP, flagellin, or MDP compared to control siRNA ( Figure 5C). Furthermore, the Drosha-siRNA significantly inhibited the formation of ASC specks by LPS and poly(dA:dT) stimulation relative to the control siRNA ( Figure 5D). These results suggest that deficiency of DROSHA suppresses the AIM2 inflammasome activation in macrophages.

Transduction of miRNA Promotes the ASC Speck Formation for AIM2 Inflammasome Activation
We next investigated the underlying molecular mechanism by which DROSHA regulates AIM2 inflammasome activation in macrophages. Since DROSHA-induced miRNAs have double-stranded RNA (dsRNA) structure of the hairpins in a pri-miRNA and can bind to dsDNA to form hetero-triplex structures at specific target sequences in DNA [46], we examined whether the transduction of miRNA can directly promote the AIM2 inflammasome activation in response to poly(dA:dT). We transduced the five independent synthetic miRNAs (miRNA #1, #2, #3, #4, and #5) to increase the amount of miRNA during AIM2 inflammasome activation in macrophages. The transduction of miRNAs significantly elevated the secretion of IL-1β and IL-18 compared to vehicle control, whereas the secretion of TNF-α was unchanged ( Figure 6A). Consistently, the transduction of miRNAs increased the activation of caspase-1 and IL-1β cleavage relative to vehicle control ( Figure 6B). We next analyzed whether the transduction of miRNA could promote the formation of ASC specks for AIM2 inflammasome activation. Notably, the transduction of miRNA enhanced the formation of ASC specks by LPS and poly(dA:dT) stimulation relative to the vehicle control ( Figure 6C). These results suggest that the tranduction of miRNA promotes the complex formation of AIM2 inflammasome for AIM2 inflammasome activation.

Discussion
Here we demonstrate that DROSHA contributes to AIM2 inflammasome activation-dependent lung inflammation during idiopathic pulmonary fibrosis. Our results show that both DROSHA and AIM2 protein levels are significantly elevated in alveolar macrophages of patients with IPF and bleomycin-treated mice. We showed that the genetic inhibition of DROSHA suppresses AIM2 inflammasome-dependent caspase-1 activation and IL-1β and IL-18 secretion by inhibition of ASC speck formation. Furthermore, the transduction of miRNA promotes the AIM2 inflammasome activation. To the best of our knowledge, this is the first reported link between DROSHA-dependent miRNA biosynthesis and AIM2 inflammasome activation in pulmonary fibrosis. Our findings suggest DROSHA-dependent AIM2 inflammasome activation contributes to pulmonary fibrosis.
Previous studies have shown that alveolar macrophages are critical regulators of lung fibrosis [47]. Alveolar macrophages-derived cytokines are associated with the proliferation of fibroblast [48,49]. Also, bleomycin-induced lung fibrosis is linked to the activation of inflammation [50]. Among the various cytokines, IL-1β can directly induce collagen secretion by fibroblasts [48,49]. Transient and chronic expression of IL-1β mediates fibrosis [48,49]. Since inflammasome is a critical regulator of IL-1β production and secretion in alveolar macrophages, we investigated to discover the upstream novel molecule for AIM2 inflammasome activation in alveolar macrophages during IPF. The DROSHA-DGCR8 complex is required for the first step of miRNA biogenesis [51]. DROSHA is a double-strand RNA (dsRNA)-specific endoribonuclease that is involved in the initial step of miRNA biogenesis [40]. Although the role of miRNA has been demonstrated in the immune response [31], the function of DROSHA in AIM2 inflammasome activation has not previously been reported. Our results suggest that DROSHA could be a critical molecule for AIM2 inflammasome activation-dependent IL-1β production in alveolar macrophages during IPF.
The changes in miRNA are important for the immune response such as maturation, proliferation, differentiation, and activation of immune cells [51,52]. The total amount of miRNA was elevated in bronchoalveolar lavage (BAL) from patients with IPF compared to controls [29]. These previous studies indicated that miRNA could have a role as a DAMP in lung inflammation during IPF. Consistent with these previous studies, we demonstrate that excessive accumulation of miRNAs promotes the production of IL-1β and IL-18 via AIM2 inflammasome activation in macrophages. Moreover, we show that the DROSHA levels were elevated in alveolar macrophages in patients with IPF. These results suggest that the DROSHA-dependent miRNA production and the excessive accumulation of miRNA can regulate AIM2 inflammasome activation in alveolar macrophages during IPF. Although we showed the elevated expression of DROSHA in patients with IPF, there is limitation of small cohort number in our human study with patients with IPF. To investigate the role of DROSHA in the pathogenesis of IPF, the correlation between DROSHA expression and the progression of IPF needs to be clarified by further studies.
Although the function of individual miRNA during IPF was reported in the previous papers, the role of miRNA on AIM2 inflammasome activation has not been discovered the mechanism for the progression of IPF. As an upstream molecule of AIM2 inflammasome activation in alveolar macrophages of lung during IPF, we demonstrate that miRNA could be a critical activator for AIM2 inflammasome-mediated lung inflammation. We show that inhibition of DROSHA suppressed AIM2 inflammasome-dependent caspase-1 activation and ASC speck formation in response to poly(dA:dT) as dsDNA which is required for AIM2 inflammasome activation. Consistent with these results, the high levels of miRNA by transduction increased AIM2 inflammasome-dependent caspase-1 activation and AIM2 inflammasome complex formation via ASC speck formation by stimulation with poly(dA:dT). Our results suggest that DROSHA-mediated miRNA production or secretion may be a critical signature for the AIM2 inflammasome activation during IPF.
In summary, we found that (1) DROSHA is elevated in alveolar macrophages in both human IPF and a mouse model of pulmonary fibrosis, and (2) DROSHA promotes AIM2 inflammasome activation. Our results support that DROSHA-driven AIM2 inflammation activation could be a critical molecular pathway in lung inflammation during IPF.
Although not yet fully completed, our study provides the role of DROSHA in AIM2 inflammasome-dependent lung inflammation in the pathogenesis of pulmonary fibrosis.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4409/8/8/938/s1, Figure S1: DGCR8 expression is comparable in lung tissues from patients with IPF and a healthy subject. Figure  S2: DGCR8 expression is comparable in lung tissues during bleomycin-induced lung injury. Figure