Activation of the IL-1β/KLF2/HSPH1 pathway promotes STAT3 phosphorylation in alveolar macrophages during LPS-induced acute lung injury

Abstract Acute lung injury (ALI) is a lethal disease with diffuse lung inflammation, in which JAK/STAT3 signaling has been well recognized for its role in initiating and amplifying inflammatory processes. However, the mechanism for the enhancement and maintenance of signal transducer and activator of transcription 3 (STAT3) activation has not yet been clearly demonstrated in ALI. In the present work, we established a lipopolysaccharide (LPS)-induced ALI rat model through intratracheal instillation and isolated the alveolar macrophages (AMs) from the rats in the model. We demonstrated that the expression of Kruppel-like factor 2 (KLF2) significantly decreased in the AMs from LPS-induced ALI rats (LPS-AMs) as compared with the AMs from control rats (NC-AMs). Overexpressing KLF2 in LPS-AMs inhibited the phosphorylation of STAT3 and reduced the levels of STAT3 target genes, including matrix metalloproteinase (MMP)-2/9 (MMP-2/9). Further investigation indicated that KLF2 trans-inhibited heat shock protein H1 (HSPH1), which interacted with STAT3 and enhanced its phosphorylation. As a crucial inflammatory mediator in ALI, interleukin-1β (IL-1β) induced the down-regulation of KLF2 in LPS-AMs, as interrupting IL-1β signaling in LPS-AMs by antibody neutralization or IL1R1 knockdown rescued the expression of KLF2. Consistently, stimulating NC-AMs with IL-1β decreased KLF2 and increased HSPH1, while overexpression of KLF2 suppressed IL-1β-induced HSPH1. Additionally, in vivo studies showed that treatment with an IL-1β antibody or HSPH1 inhibitor alleviated lung injury in ALI rats, as well as decreased the levels of p-STAT3 and MMP-2/9. In conclusion, activation of the IL-1β/KLF2/HSPH1 pathway facilitated STAT3 phosphorylation in AMs, which exacerbated pulmonary inflammation in ALI.


Introduction
Acute lung injury (ALI) is a severe disease with diffuse alveolar damage, which has high morbidity and mortality in intensive care patients [1]. An uncontrolled acute inflammatory response, pulmonary endothelial, epithelial barrier dysfunction, loss of alveolar-capillary membrane integrity, and excessive secretion of pro-inflammatory cytokines are considered to be the main pathological factors leading to ALI [2]. The inflammatory microenvironment composed of immune cells, structure cells, and inflammatory factors plays a key role in the pathogenesis of ALI. Previous studies have reported that macrophages, neutrophils, lymphocytes, pulmonary epithelial fibroblasts, and platelets are involved in the occurrence and development of ALI. Additionally, the inflammatory cytokines produced by these cells form a complex signaling network that regulates all stages of the inflammatory response in ALI [3,4]. Therefore, it is of great significance to elucidate the key factors and their interactions within the inflammatory microenvironment of ALI.
Previous studies have shown that resident macrophages in the alveoli or recruited macrophages within the blood play a critical role in the pathogenesis of ALI [5]. During the acute phase of ALI, alternatively activated macrophages (M2-type) transform into classically activated macrophages (M1-type), producing a large amount of inflammatory mediators. During the late phase of ALI, M1-type macrophages transform back into M2-type macrophages, thus playing a role in the clearance of apoptotic cells and promotion of fibrosis [6]. It was found that eliminating alveolar macrophages (AMs) effectively reduced the extent of lung injury in the early stage of ALI [7,8]. At this stage, a large number of M1-type macrophages are produced, and their removal helps to reduce the levels of inflammatory cytokines in the alveoli, thereby inhibiting excessive and harmful inflammatory responses. However, understanding of the activation process of AMs within the inflammatory response of ALI is limited.
Signal transducer and activator of transcription 3 (STAT3) is a major signal transduction protein utilized by multiple inflammatory cytokines [9]. Previous studies have shown that the activation of STAT3 occurred before severe lung injury and was closely related to the secretion of various inflammatory factors, suggesting that STAT3 plays an essential role in the initial stage of inflammation during ALI [10,11]. In an ALI animal model, activation of STAT3 aggravated the inflammatory response and degree of lung injury. On the contrary, inhibiting STAT3 activation reduced the severity of ALI and the levels of inflammatory cytokines [12,13]. Recently, our work revealed that activation of STAT3 in AMs promoted the expression and secretion of matrix metalloproteinases (MMPs), resulting in the destruction of the alveolar epithelial-endothelial barrier in ALI [14]. Although the function of STAT3 in the inflammatory response of ALI has been well described, the regulatory mechanism for STAT3 signaling in AMs and its role in the development of ALI still needs to be further explored. In the present study, we found a novel regulatory pathway, interleukin-1β (IL-1β)/Kruppel-like factor 2 (KLF2)/heat shock protein H1 (HSPH1), which promoted the phosphorylation and activation of STAT3 protein in AMs.

Lipopolysaccharide-induced ALI rat model
The ALI model was constructed as described in our previous work [14]. All male Wistar rats were obtained from the Laboratory Animal Centre of Wenzhou Medical University. All animal procedures were approved by the Ethics Committee of the Second Affiliated Hospital of Wenzhou Medical University (ID number: wydw 2019-0753) and performed at the Central Laboratory of the Second Affiliated Hospital of Wenzhou Medical University. In brief, the rats were anesthetized by intraperitoneal injection with 3% sodium pentobarbital, followed by the instillation of 2 mg/kg lipopolysaccharide (LPS) (Sigma-Aldrich, St. Louis, MO, U.S.A.) solution into the tracheas. An equal volume of normal saline was used in the control group instead of LPS. For neutralizing IL-1β or inhibiting HSPH1, the rats received an intratracheal instillation of an anti-IL-1β antibody (5 μg; ab9722, Abcam, Cambridge, U.K.) or HSPH1 inhibitor (100 mg/kg; KNK437, Selleck, Texas, U.S.A.) 1 h after the LPS treatment. The rats were then allowed to recover and were killed 24 h later by intraperitoneal injection with 100 mg/kg pentobarbital.

Bronchoalveolar lavage fluid collection and inflammatory cell analysis
The bronchoalveolar lavage fluid (BALF) was collected at 24 h post-LPS treatment. Using a tracheal cannula, the lung was washed with 2 ml of normal saline three times. All the 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. Each cell pellet was re-suspended in PBS, and the total cell number was determined in an automatic blood cell analyzer (Sysmex, Kobe, Japan). M1-type macrophages (labeled with a CD86 antibody, ab213044, Abcam) and total macrophages (labeled with a F4/80 antibody, ab100790, Abcam) were sorted via flow cytometry (BD FACSAria III, BD Biosciences, NJ, U.S.A.).

Quantitative real-time PCR
RNA extraction, reverse-transcription, and quantitative PCR were performed as described in our previous work [14]. Target gene expression was normalized to β-actin levels and calculated using the 2 − C t method. The relative mRNA expression was further calculated through normalizing to the control group.

Luciferase assay
The promoter of the HSPH1 gene was amplified by PCR and inserted into the pGL3-basic. The reporter constructs with different lengths or mutated KLF2 binding sites (KLF BSs) were generated by subsequent PCR-based cloning. LPS-AMs or NC-AMs were plated on to 24-well plates. After 24 h, cells were co-transfected with pGL3 constructs and pRL-SV40 plasmids (Promega, Madison, WI, U.S.A.). The pRL-SV40 plasmid was used to normalize the transfection efficiency. Luciferase activity was measured using a dual-luciferase reporter assay system (Promega) and a luminometer (LB 9507, Berthold, BadWildbad, Germany).

Co-immunoprecipitation
The LPS-AMs and NC-AMs were lysed in 500 μl of RIPA lysis buffer (Millipore, Billerica, MA). The samples were centrifuged to remove insoluble debris after cell lysing, and the supernatant was split into two equal aliquots (20 μl lysate remained as input). Anti-STAT3 and rabbit IgG antibodies (Abcam) were added. Immunoprecipitation was performed using Pierce™ Protein A/G Magnetic Beads (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions.

Chromatin immunoprecipitation
Chromatin was cross-linked with 1% formaldehyde and sonicated to obtain 200-500 bp DNA fragments. Following centrifugation, the supernatants were subjected to immunoprecipitation for 12 h at 4 • C with antibodies against KLF2 or with normal goat IgG (Abcam). Chromatin-antibody complexes were isolated using Protein A/G PLUS Agarose (Santa Cruz). The cross-linking was reversed and genomic DNA fragments were purified and analyzed by PCR using the following primer pair for the HSPH1 promoter: CAAAATTGCTATGTCTCCATGG (forward) and GTTATTTTTGTTATGCATCCC (reverse).

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

Lung histology and immunohistochemistry
The lung tissues in all groups were fixed by tracheal perfusion and then soaked with 4% paraformaldehyde for 24 h. The tissues were embedded in paraffin and cut into 5-μm-sections. Hematoxylin and Eosin (H&E) staining was performed using standard protocol. For immunostainings, samples were deparaffinized, rehydrated in an alcohol concentration gradient, and blocked by incubating in 0.3% H 2 O 2 for 30 min. Antigen retrieval was performed by treating the slides in citrate buffer in a microwave oven for 10 min. The slides were incubated for 1 h with normal goat serum and then incubated in a moist chamber with an anti-p-STAT3 or anti-STAT3 antibody at 4 • C for 12 h. After a complete wash in PBS, the tissues were incubated in peroxidase-conjugated secondary antibody for 30 min at 37 • C, rinsed with PBS. The signal was detected using diaminobenzidine (DAB, Solarbio, Beijing, China).

Statistical analysis
Statistical analysis was performed using SPSS 22.0 and GraphPad Prism 6.0. Analysis of differences was performed using a two-tailed Student's t test and analysis of variance (one-way with Tukey's post-hoc test; two-way with Sidak's post-hoc test). The data are presented as the mean + − S.E of three separate experiments. A P-value <0.05 was considered statistically .

Establishment of ALI model
We established an in vivo ALI rat model according to our previous work [14]. H&E staining showed evident pathological changes, including alveolar wall thickening, alveolar collapse, and inflammatory cell infiltration, in the lung tissues of ALI rats ( Figure 1A). Flow cytometry indicated that the number of CD86-positive, M1-type macrophages in the BALF from ALI rats was significantly increased as compared with control rats ( Figure 1B). Additionally, the concentrations of TNF-α, IL-1β, IL-6, and IL-33 in the BALF from ALI rats were remarkably elevated ( Figure 1C).

KLF2 inhibits the phosphorylation of STAT3 protein in AMs
Based on the published gene expression profiles of the total cells in the BALF from LPS-induced ALI or control rats (GSE111241 [15]), we found that the mRNA level of KLF2 was decreased in ALI rats (Figure 2A). Consistently, we cultured the AMs from LPS-AMs and NC-AMs in vitro and found that both the mRNA and protein levels of KLF2 were lower in LPS-AMs than in NC-AMs ( Figure 2B,C). Furthermore, we noticed that the STAT3 phosphorylation induced in LPS-AMs was inhibited by the overexpression of KLF2 ( Figure 2D). KLF2 overexpression also suppressed the production of MMP-2 and MMP-9, which were up-regulated by STAT3 signaling in LPS-AMs ( Figure 2E). These results suggest that down-regulated KLF2 in the AMs from ALI rats may promote STAT3 phosphorylation.

Down-regulation of KLF2 promotes STAT3 phosphorylation by inducing HSPH1 expression in the AMs from ALI rats
Through analyzing the expression profiles (GSE111241), we found that the cells from ALI rats showed a high expression of HSPH1 as compared with the control rats ( Figure 2A). Consistently, both the mRNA and protein levels of HSPH1 were higher in LPS-AMs than in NC-AMs ( Figure 3A,B). Further investigations revealed that overexpression of HSPH1 had no effect on the phosphorylation of STAT3 in NC-AMs, but increased the phosphorylation level of  Figure 3C,D). On the contrary, knockdown of HSPH1 inhibited STAT3 phosphorylation in LPS-AMs ( Figure 3E). Moreover, we found that highly expressed HSPH1 interacted with STAT3 in LPS-AMs ( Figure 3F). Considering that both KLF2 and HSPH1 affected STAT3 phosphorylation, the relationship between KLF2 and HSPH1 was studied. The results showed that overexpression of KLF2 significantly reduced the mRNA and protein levels of HSPH1 in LPS-AMs, while knockdown of KLF2 exhibited the opposite effect in NC-AMs ( Figure 3G-J). These data suggest that the KLF2 decrease in AMs from ALI rats induces the expression of HSPH1, which contributes to STAT3 phosphorylation.

KLF2 trans-inhibits HSPH1 in AMs
In order to explore the mechanism of how KLF2 regulates HSPH1 expression, we analyzed the promoter sequence of the HSPH1 gene and found a potential KLF2 BS ( Figure 4A). Based on this prediction, we constructed truncated or mutated reporters of the HSPH1 promoter and examined their activities in NC-AMs and LPS-AMs. Our results showed that compared with LPS-AMs, the activities of the reporters containing KLF2 BS were suppressed in NC-AMs, while the reporter lacking this site was not ( Figure 4A). Consistently, the reporter with the wild-type KLF2 BS had a much lower activity than with the mutant in NC-AMs, but no significant difference was observed in LPS-AMs ( Figure 4B). In accordance with HSPH1 expression, overexpression of KLF2 decreased the reporter activity of the HSPH1 promoter in LPS-AMs, and knockdown of KLF2 increased the activity in NC-AMs ( Figure 4C,D). Notably, the binding of KLF2 to the HSPH1 promoter around the KLF2 BS was stronger in NC-AMs than in LPS-AMs ( Figure  4E), indicating that KLF2 represses the transcription of HSPH1 gene by directly binding to its promoter.

IL-1β signaling is required for the KLF2 inhibition in the AMs from ALI rats
To further explore the upstream signaling regulating KLF2 expression in the AMs from ALI rats, we examined the levels of KLF2 in NC-AMs treated with different inflammatory factors, which were up-regulated in the BALF of ALI rats. After IL-1β or TNF-α stimulation, KLF2 expression decreased, among which IL-1β had the most obvious effect ( Figure 5A,B). It was found that KLF2 expression was rescued by blocking the IL-1β signaling in LPS-AMs by using an IL-1β antibody or by knocking down IL1R1, the IL-1β receptor ( Figure 5C-E). Additionally, exogenous IL-1β induced HSPH1 expression in NC-AMs, which was interrupted by overexpression of KLF2 ( Figure 5F). These findings suggest that IL-1β signaling decreases the level of KLF2, thus inducing HSPH1 expression in the AMs from ALI rats.

Blocking the IL-1β/KLF2/HSPH1 axis relieves lung injury in ALI rats
To verify the role of the IL-1β/KLF2/HSPH1 axis in ALI, ALI rats received an intratracheal instillation of an IL-1β antibody or HSPH1 inhibitor to block the IL-1β/KLF2/HSPH1 axis. It was shown that treatment with an IL-1β antibody or HSPH1 inhibitor curbed the development of edema and infiltration of inflammatory cells in the lung tissues of ALI rats ( Figure 6A-D). Moreover, the concentration of MMP-2 and MMP-9 in the BALF from ALI rats was substantially decreased after the addition of an IL-1β antibody or HSPH1 inhibitor ( Figures 6E,F). Immunohistochemical analysis confirmed that blocking the IL-1β/KLF2/HSPH1 axis reduced the phosphorylation level of STAT3 and alleviated pathological changes in the lung tissues of ALI rats ( Figure 6G).

Discussion
KLF family proteins are a group of DNA-binding proteins with a zinc finger domain that participates in many biological processes, including inflammation regulation, by controlling the transcription of downstream effector genes [16]. As a member of the KLF family, KLF2 is mainly expressed in lung tissues and is crucial in the development of normal lung tissues [17]. Therefore, abnormal expression of KLF2 is closely related to pulmonary diseases, such as pulmonary hypertension, chronic obstructive pulmonary disease (COPD), and ALI [18][19][20]. It was found that the expression level of KLF2 was significantly decreased in patients with pulmonary cystic fibrosis, active pulmonary tuberculosis, or COPD [21,22]. Although KLF2 has not been reported in ALI patients, it was identified to be an early marker of ALI in an animal model, and its serum concentration and expression level in lung tissue were negatively correlated with the severity of ALI [20]. In an inflammatory response, high expression of KLF2 inhibited macrophage activation, while silencing KLF2 promoted macrophage activation and cytokine secretion [23,24]. In the present work, we found that KLF2 was significantly down-regulated in LPS-AMs as compared with NC-AMs. Overexpressing KLF2 in LPS-AMs decreased the phosphorylation level of STAT3. Early studies demonstrated the ability of KLF2 to influence transcription through direct DNA-binding and indirect cofactor sequestration mechanisms [15]. KLF2 resisted inflammation within macrophages and inhibited M1-type macrophage polarization by recruiting NF-κB cofactors  away from the promoters of inflammatory genes [25]. Here, we discovered a novel mechanism involving the inhibition of HSPH1 gene transcription by KLF2 directly binding to its promoter, thus hindering the phosphorylation of STAT3 induced by LPS in AMs.
HSPH1 is a molecular chaperone that prevents protein aggregation by participating in the correct folding of newly produced or misfolded proteins by binding to Hsp70 [26]. HSPH1 can be induced by a variety of stress factors in order to maintain cell survival. In tumors, HSPH1 regulates several inflammation-related signaling pathways. In large B-cell lymphoma, HSPH1 conferred chronic activation of NF-κB signaling by stabilizing MyD88 [27]. In colorectal cancer, HSPH1 promoted the phosphorylation and activation of STAT3 through a direct interaction with STAT3 [28]. Based on the published expression profiles of cells from the BALF of ALI rats (GSE111241), we found that HSPH1 was highly expressed in ALI rats as compared with the control. Consistent with the findings in tumor, HSPH1 was required for STAT3 phosphorylation in LPS-AMs. Importantly, we noticed that overexpressing HSPH1 increased the  phosphorylation level of STAT3 in IL-6-treated NC-AMs or LPS-AMs, but not in non-treated NC-AMs, suggesting that HSPH1-mediated STAT3 phosphorylation is also dependent on inflammatory signals, such as IL-6, in AMs.
Further investigation showed that IL-1β was responsible for the low expression of KLF2 in LPS-AMs. Exogenous IL-1β stimulation decreased KLF2 and increased HSPH1 in NC-AMs, indicating that the regulation of KLF2 and HSPH1 by IL-1β in AMs does not rely on other signals. IL-1 is a signal initiator and amplifier in immune and inflammatory cascades, and its abnormal regulation is associated with a variety of inflammatory diseases [29]. Furthermore, IL-1β is known as an important molecular marker for the development of ALI. The level of IL-1β was up-regulated in the lung tissues of the patients with early ALI [30]. Further work revealed that IL-1β increased pulmonary vascular permeability and inhibited fluid transport across pulmonary epithelial cells, thus accelerating the progression of ALI [31,32]. In ALI animal models, the severity of ALI was reduced by blocking IL-1β signals [33][34][35]. Clearance of neutrophils and macrophages in the alveoli of ALI animals effectively reduced IL-1β levels, suggesting that these cells may be the primary source of IL-1β [36]. In the present work, the concentration of IL-1β in the culture medium of LPS-AMs was significantly higher than that of NC-AMs, and the expression of KLF2 in LPS-AMs was rescued when knocking down the IL-1β receptor IL1R1 or adding an IL-1β antibody into the culture medium. The above results suggest that autocrine IL-1β may maintain the low expression of KLF2 in LPS-AMs.
Altogether, these findings reveal that the abnormally elevated level of IL-1β in ALI patients acts on AMs and relieves KLF2-meidated transcriptional inhibition of the HSPH1 gene by down-regulating KLF2 expression. Increased HSPH1 further interacts with STAT3 to promote the phosphorylation of STAT3 protein, thus contributing to activation of AMs (Figure 7).