Elevated IL-33 promotes expression of MMP2 and MMP9 via activating STAT3 in alveolar macrophages during LPS-induced acute lung injury

Background Pulmonary inflammation and endothelial barrier permeability increase in acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) induced by pro-inflammatory cytokines and matrix metalloproteinases (MMPs). However, the relationship between pro-inflammatory cytokines and MMPs in ALI/ARDS remains poorly understood. Methods A lipopolysaccharide (LPS)-induced ALI rat model was established through intratracheal instillation. The wet/dry ratios of lung tissues were measured, and bronchoalveolar lavage fluid (BALF) was collected to test protein concentrations, total cell/macrophage numbers, and pro-inflammatory cytokine levels. LPS-treated alveolar macrophages were utilized in in vitro experiments. The expression and secretion of MMPs were respectively detected using quantitative PCR, Western blotting and ELISA assays. Results The levels of IL-33 and MMP2/9 in BALF increased in all the ALI rats with severe lung injury. LPS-induced IL-33 autocrine upregulated the expression of MMP2 and MMP9 through activating STAT3. Neutralizing IL-33 in culture medium with specific antibodies suppressed the expression and secretion of MMP2 and MMP9 in LPS-treated alveolar macrophages. Consistently, eliminating IL-33 decreased the levels of MMP2 and MMP9 in BALF and alleviated lung injury in ALI rats. Conclusion The IL-33/STAT3/MMP2/9 regulatory pathway is activated in alveolar macrophages during acute lung injury, which may exacerbate the pulmonary inflammation.

mortality rate of ALI patients is still as high as 20-50% [4]. Thus, it is urgent to develop more effective treatments for ALI.
The uncontrolled inflammatory response is a key step during the development of ALI/ARDS. Pro-inflammatory immune cells and the cytokines released by them can increase the permeability of the alveolar-capillary barrier, leading to pulmonary dysfunction [5]. Alveolar macrophages (AMs) are immune cells accounting for 95% of airspace leukocytes [6]. An essential process in ALI, activation of AMs leads to the oversecretion of various inflammatory cytokines, such as TNF-α, IFN-γ, and interleukins [7]. As a newly identified member of the interleukin (IL)-1 family, IL-33 can aggravate inflammatory reactions and increase microvascular permeability, two processes that decrease the survival rate in LPS-induced ALI mice [8]. In our previous study, the plasma level of IL-33 was found to be higher in ALI patients than in the healthy [9]. This evidence suggests a potential association between IL-33 and ALI.
Matrix metalloproteinases (MMPs) are involved in ALI. MMPs-mediated degradation of proteins in the alveolar epithelial-endothelial unit, including intercellular junction proteins, basement membrane (BM) and proteins anchoring cells to the BM, is a central process in ALI [10]. Among these MMPs, MMP2 and MMP9 are being intensely studied, and their levels are elevated in BALF from ALI patients [11]. The in vivo lung injury model demonstrated a rise of both MMP-2 and -9 in BALF that correlated with the alveolar-arterial oxygen gradient (PA-a,O 2 ) [12]. Also, in an immune complex deposition model, mice deficient in MMP-9 developed less severe lung injury than wild type mice [13]. However, the mechanisms underlying the rise of MMP2/9 during ALI remain elusive.
Here, we investigated the association between IL-33 and MMP2/9 in an LPS-induced ALI model in vitro and in vivo. Oversecretion of IL-33 promoted the expression of MMP2/9 by activating STAT3 in LPS-treated alveolar macrophages. The regulatory mechanism was further researched with LPS-induced ALI rats.

LPS-induced ALI rat model
All 6-week old male Wistar rats were obtained from Wenzhou Medical University and maintained in a pathogen-free environment. Nine rats were assigned to each treatment group and a control group. The ALI model was established following the well-established protocol. In brief, the rats were anesthetized with 3% sodium pentobarbital, followed by the instillation of 2 mg/kg LPS (Sigma-Aldrich, St. Louis, MO, USA) solution into the tracheas. Instead of LPS solution, an equal volume of normal saline was used in the control group. For neutralizing IL-33, the rats received an intratracheal instillation of anti-IL-33 antibodies (5 μg, ab187060, Abcam, Cambridge, UK) or isotype IgG (ab172730, Abcam) 1 h after the LPS treatment. Then the rats were allowed to recover and were euthanized 24 h later with pentobarbital. All animal procedures were approved by the Ethics Committee of the Second Affiliated Hospital of Wenzhou Medical University.

Lung wet/dry ratio analysis
After the rats were euthanized, their lungs were harvested and weighed immediately. Then the blood on the lung surface was washed away, and the lungs were dried at 60°L C for 72 h. The dried specimens were weighed again, and the wet/dry ratio was calculated accordingly.

BALF collection and inflammatory cell analysis
The BALF was collected at 24 h after LPS treatment. Using a tracheal cannula, the lung was washed with 2 ml of normal saline three times. All flushing fluid was collected. The lavaged sample was then centrifuged at 1500×g for 10 min at 4°C and the supernatant was collected for protein concentration analysis. Total protein concentration in the supernatant was measured. Each cell pellet was re-suspended in PBS and the total cell number determined in an automatic blood cell analyzer (Sysmex). Macrophages were marked with F4/80 antibodies or FITC-secondary antibodies and then analyzed and sorted with flow cytometry.

Quantitative real-time PCR
Extraction of total RNA from NR8383 cells was performed with RNAiso Plus reagent and further reverse-transcribed using a PrimeScript RT reagent kit (both from Takara, Tokyo, Japan). SYBR-Green mix (Roche) was used to carry out quantitative PCR according to the manufacturer's instructions. Target gene expression was normalized to β-actin levels in respective samples as an internal control and calculated using the 2 −ΔΔCq method, and the relative mRNA expression was further calculated through normalizing to the control group.

Statistical analysis
The software SPSS 13.0 and GraphPad Prism 5 were used in the statistical analyses. Group distributions were performed with Student's t-test or analysis of variance (ANOVA). P < 0.05 was considered statistically significant.

Establishment of acute lung injury (ALI) model
In order to establish the in vivo ALI model, LPS was used to induce ALI in the rats. The lung tissues of ALI rats were collected at 24 h after treatment with or without LPS. The results showed that LPS caused severe pulmonary edema, as indicated by an increased lung wet/dry ratio (Fig. 1a). This observation was further confirmed by the increase of protein concentration in BALF from LPS-treated rats (Fig. 1b). To investigate the level of pro-inflammatory cells and cytokines in ALI, infiltration of total cells and macrophages was evaluated after LPS treatment. The total cells and macrophages were evidently enriched in BALF from LPS-treated rats ( Fig. 1c and d). Additionally, the concentrations of IL-33, TNF-α, MMP2, MMP9 and TIMP1 in the BALF were significantly elevated ( Fig. 1e-i). Furthermore, we purified the alveolar macrophages (AMs) with positive F4/80 staining from BALF. Consistently, the AMs from LPS-treated rats showed a higher level of IL-33 mRNA than those from the control group (Fig. 1j). Notably, as an IL-33 receptor, the ST-2 level was increased in the AMs from LPS-treated rats but the IL-1RAP level remained unchanged (Fig. 1k). Considering the inflammatory response induced by LPS, the status of the inflammatory signaling, p-STAT3, was also examined in the purified AMs. As shown in Fig.  1l, the level of p-STAT3 was significantly increased in the AMs purified from LPS-treated rats. Altogether, our findings indicated that the acute pulmonary inflammatory response was triggered in LPS-induced ALI rats.

mary AMs
Primary AMs were obtained through purifying the alveolar macrophages (AMs) in BALF from normal rats. LPS induced secretion of IL-33, TNF-α, MMP2, and MMP9 in primary AMs (Fig. 2a-d). Meanwhile, ST-2 was upregulated but IL-1RAP showed no change (Fig. 2e). STAT3 was also activated by LPS in the primary AMs (Fig. 2f ). Therefore, the results suggest that primary AMs can be activated by LPS.
LPS induced secretion of IL-33, TNF-α, MMP2, MMP9, and TIMP1 in AM cell line NR8383 Considering the critical role of AMs in regulating pro-inflammatory events during ALI, we examined the effect of LPS on the secretion of IL-33, TNF-α, MMP2, and MMP9 in the AM cell line NR8383. Levels of TNF-α, MMP2, and MMP9 in culture medium increased in a time-dependent manner after LPS treatment, while the IL-33 level peaked at 12 h after LPS treatment (Fig. 3a-d). Although TIMP1 acts as a natural MMP inhibitor, the secretion of TIMP1 was also promoted by LPS (Fig. 3e). These results suggest that the levels of IL-33, TNF-α, MMP2, MMP9, and TIMP1 in BALF from the LPS-induced ALI rats may be raised by the alveolar macrophages which were activated by LPS. To determine whether the secretion of MMP2 and MMP9 is induced by IL-33, NR8383 cells were stimulated with recombinant IL-33 protein. It was found that the concentrations of MMP2 and MMP9 in culture medium were increased in an IL-33-dose-dependent manner ( Fig. 4a and b). Consistently, the mRNA and protein levels of MMP2 and MMP9 were upregulated in NR8383 cells treated with IL-33 ( Fig.  4c-e). As a powerful signal transducer, STAT3 is essential for the interleukin-mediated activation of macrophages. Here, we found that IL-33 induced the phosphorylated activation of STAT3 in primary AMs (Fig. 4f ). Blocking the activation of STAT3 with the specific inhibitor stattic attenuated the IL-33-induced expression and secretion of MMP2 and MMP9 in NR8383 cells, with no effect on activation of the MAPK or NFκb pathway (Fig. 4g-l). Additionally, we investigated whether other inflammatory cytokines were also produced through IL-33/STAT3 signaling in AMs. The results showed that IL-33 induced the secretion of TNF-α, IL-6, IL-10, and IFN-γ in NR8383 cells, which was not significantly changed after stattic addition (Fig. 4m). Similarly to stattic treatment, knocking down STAT3 with siRNA significantly decreased the mRNA levels of MMP2 and MMP9 in NR8383 cells as well as their concentrations in culture medium (Fig. 5a-f ). Therefore, these results demonstrate that IL-33 promotes expression of MMP2 and MMP9 in AMs through activating STAT3.

Neutralizing IL-33 inhibited the LPS-induced MMP2/9 expression and lung injury
To verify the role of IL-33 in LPS-induced ALI, IL-33 specific antibodies were administered into culture medium to neutralize the secreted IL-33 when AMs were being treated with LPS. It showed that IL-33 antibodies significantly lowered the mRNA and protein levels of MMP2 and MMP9 in LPS-treated NR8383 cells, as well as their medium concentrations (Fig. 6a-e). To further evaluate the protective effect of IL-33 antibody in the LPS-induced ALI model, the rats received an intratracheal instillation of IL-33 antibody after LPS administration to neutralize the secreted IL-33. The results showed that treatment with anti-IL-33 antibodies obviously curbed the edema development and inflammatory cell infiltration in the lung tissues from LPS-induced ALI rats (Fig. 7a-d). Moreover, the concentrations of MMP2 and MMP9 in BALF from the LPS-treated rats were substantially decreased after the addition of IL-33 antibodies ( Fig. 7e and f ). Taken together, IL-33 displays a central role in LPS-induced lung inflammation/injury and MMP2/9 secretion.

Discussion
In the present study, an in vivo LPS-induced ALI rat model and an in vitro model using AM NR8383 cells were established to investigate the link between IL-33 and ALI. BALF from the in vivo model showed an increased IL-33 level, which is consistent with the observation in LPS-induced ALI mice [14]. The expression of IL-33 was boosted by LPS in AMs, suggesting the ability of AMs to produce IL-33 during LPS-induced ALI. AMs, once activated by toll-like receptors (TLRs), serve as the first line of defense against invaders into the lung [15]. Wang et al. reported that the NF-κB pathway aberrantly activated in BALF-derived macrophages aroused ARDS in rats [16]. Therefore, we supposed that the LPS-induced IL-33 secretion requires activation of the TLR4/NF-κB pathway since NF-κB is the central mediator of inflammation. Several studies have demonstrated that IL-33 directly up-regulated expression of the LPS receptor TLR4 in macrophages, and this up-regulation in turn exaggerated the activation of NF-κB and increased the production of pro-inflammatory cytokines [17,18]. Furthermore, Fu et al. found that the up-regulation of IL-33 in LPS-induced ALI mice also involved HMGB1, a leading factor for the cascade amplification of inflammation [19]. Except for IL-33, LPS stimulation immediately activated MMP2 and MMP9 production in BALF from ALI rats, which is also validated by the results of in vitro studies using NR8383 cells. However, the level of IL-33 secreted by NR8383 cells peaked at 12 h after LPS treatment, while levels of MMP2 and MMP9 continued to increase, more evidently after 12 h. As the time-course study revealed that relatively early expression of IL-33 was followed by increased production of MMP2 and MMP9 after LPS challenge in AMs, we hypothesize that autocrine or paracrine IL-33 release after LPS stimulation further activates the downstream inflammatory pathways which up-regulate the expression of MMP2 and MMP9. Neutralizing the secreted IL-33 after LPS treatment not only decreased the production of MMP2 and MMP9 but also protected the rats from LPS-induced ALI, indicating that IL-33 may function as an inflammatory mediator in ALI development. A few studies have been conducted to investigate the mechanisms behind IL-33-mediated expression of MMP2 and MMP9 in macrophages. Ariyoshi et al. found that the transcription of MMP9 was induced by AP-1 activation, which is dependent on ERK1/2/CREB and NF-κB cascades in IL-33-stimulated macrophages [20]. However, in the present study, STAT3 was identified as a required mediator in IL-33-induced expression of MMP2 and MMP9. Although STAT3 directly activates the transcription of MMP2 and MMP9 in cancer cells [21,22], the regulatory mechanisms in alveolar macrophages should be further validated.
In conclusion, IL-33 production increases in LPS-induced ALI rats and LPS-treated alveolar macrophages. Then, the secreted IL-33 induces the expression of MMP2 and MMP9 in a STAT3-dependent manner. Disrupting the IL-33/STAT3/MMP2/9 pathway by neutralizing IL-33 relieves the pulmonary inflammation and injury in LPS-induced ALI rats, suggesting that IL-33 could be a potential therapeutic target for ALI.