Phosphorylated STAT3 suppresses microRNA‐19b/1281 to aggravate lung injury in mice with type 2 diabetes mellitus‐associated pulmonary tuberculosis

Abstract Type 2 diabetes mellitus (T2DM) is a risk factor for pulmonary tuberculosis (PTB) and increased mortality. This work focused on the functions of phosphorylated STAT3 in lung injury in mouse with T2DM‐associated PTB and the molecules involved. A mouse model with T2DM‐PTB was induced by administrations of streptozotocin, nicotinamide and mycobacterium tuberculosis (Mtb). A pSTAT3‐specific inhibitor AG‐490 was given into mice and then the lung injury in mice was observed. The molecules involved in AG‐490‐mediated events were screened out. Altered expression of miR‐19b, miR‐1281 and NFAT5 was introduced to identify their involvements and roles in lung injury and PTB severity in the mouse model. Consequently, pSTAT3 expression in mice with T2DM‐associated PTB was increased. Down‐regulation of pSTAT3 by AG‐490 prolonged the lifetime of mice and improved the histopathologic conditions, and inhibited the fibrosis, inflammation, Mtb content and number of apoptotic epithelial cells in mouse lung tissues. pSTAT3 transcriptionally suppressed miR‐19b/1281 expression to up‐regulate NFAT5. Inhibition of miR‐19b/1281 or up‐regulation of NFAT5 blocked the protective roles of AG‐490 in mouse lung tissues. To conclude, this study evidenced that pSTAT3 promotes NFAT5 expression by suppressing miR‐19b/1281 transcription, leading to lung injury aggravation and severity in mice with T2DM‐associated PTB.


| INTRODUC TI ON
Type 2 diabetes mellitus (T2DM) is a major risk factor for tuberculosis (TB) bringing two to eight folds at risk for active TB development, and the nowadays explosive DM pandemic has posed a health priority regarding DM and TB especially in places where TB infection is rampant. [1][2][3] TB is a leading cause of death from a single infectious disease agent which infected approximately one-fourth population in the world with more than 10 million people developing active disease annually, predominately active pulmonary TB (PTB), through the infectious mycobacterium tuberculosis (Mtb). [4][5][6] The first-line treatment for drug-sensitive PTB comprises rifampicin, isoniazid, pyrazinamide and ethambutol for over 6 months treatments. 7 In addition to susceptibility to Mtb infection, T2DM also imposes greater severity of TB disease by influencing disease presentation and increasing unresponsiveness to TB therapy as well as the risk of failure in treatment, relapse and death. 8,9 Considering the large coverage and the death incidences, developing new understandings and potential therapeutic options for PTB, especially for T2DM-associated PTB is of great significance.
The signal transducer and activator of transcription (STAT) family members are transcription factors activated by Janus kinase (JAK) proteins, which play important functions in regulating cell proliferation, differentiation, apoptosis and the inflammatory and immune responses. 10 Particularly, the STAT3 pathway plays unique roles in inflammatory responses and bacterial infection. 11 It has also been summarized to participate in the replication and pathogenesis of many different human and animal viruses and lead to detrimental effects of viral infection such as oncogenic viruses and Hepatitis C virus. 12 Importantly, STAT3 and suppressor of cytokine signalling-3 (SOCS3) have also been noticed as major mediators of the outcome of Mtb infection. 13 Phosphorylated STAT3 (pSTAT3, activated) translocates as homo-or heterodimers into the nucleus, where it binds to promoter sequences of specific genes or cytokines and regulates gene transcription. 11 MicroRNAs (miRNAs), a type of non-coding RNAs with the major role in messenger RNA (mRNA) regulation, are involved in diverse important cellular processes and in inflammation and innate immune responses. 14,15 In this paper, integrated miRNA microarray analysis, expression examination and online prediction were performed to identify the major differentially expressed miRNAs influenced by pSTAT3, with miR-19b and miR-1281 identified as two main subjects. Both miR-19b and miR-1281 have been studied to play versatile roles in many human malignancies as well as inflammatory responses. [16][17][18] However, their roles in immune responses to bacterial infection remain largely unknown. Intriguingly, nuclear factor of activated T-cells 5 (NFAT5) was predicted as a shared target mRNA of miR-19b and miR-1281 according to the bioinformatics analysis. Also known as tonicity-responsive enhancer binding protein (TonEBP), NFAT5 is one of Rel homology domain (RHD) proteins and shows potentials in immune response and inflammation regulations. 19 Taken together, this study was performed to evaluate the roles of pSTAT3 in Mtb infection-induced lung injury and inflammation and the potential molecules involved using a mouse model with T2DM-PTB.

| Ethics statement
This research was conducted with the approval of the Ethics Committee of Center for Disease Control and Prevention of Changji Hui Autonomous Prefecture. All experimental procedures were in line with the ethical guidelines for the study of experimental pain in conscious animals. Great efforts were made to minimize the usage and suffering of animals.

| Experimental reagents
Streptozotocin (STZ) was purchased from Sigma-Aldrich Chemical Company while nicotinamide (NA) was from Beyotime Biotechnology Co., Ltd. Enzyme-linked immunosorbent assay (ELISA) kits of interleukin (IL)-6 (#ab100712), IL-1β (#ab197742) and tumour necrosis factor-α (TNF-α, #ab208348) were purchased from Abcam Inc. The antibody against NFAT5 (#ab6721) and horseradish peroxidase (HRP)-labelled immunoglobulin G (IgG, #ab6721) were purchased from Abcam as well, while the antibody against pSTAT3 (#10253-2-AP) was acquired from Proteintech Group, Inc. The agomiR and antagomiR of miR-19b and miR-1281 (Sequences listed in Table 1 A month later, the pSTAT3-specific inhibitor AG-490 (1 mg/mL/ kg), miR-19b/1281 agomiR (5 nmol), or the shRNA or LV of NFAT5 (5 × 10 8 PFU) was given into the mice twice every 10 days through i.p. Another month later, the mice were collected for further experiments. Ten mice in each group were used for survival test, and then the number of mice in each group was supplemented to 8. The mouse serum was collected for inflammatory cytokine detection. After mouse euthanasia, the lung tissues were collected, and the left tissues were cut into slices for histopathologic staining, while the right tissues were made into homogenate for RNA and protein expression detection.

| ELISA
Protein levels of IL-6, IL-1β and TNF-α in serum and tissue homogenate were determined using the ELISA kits in strict accordance with the manufacturer's protocols. The measurement was performed on 96-well plates, and the optical density (OD) value at the wavelength of 450 nm was determined using a spectrophotometer.

| Measurement of serum lipid levels
One month after AG-490, miR-19b/1281 agomiR/antagomiR, or the shRNA/LV of NFAT5 treatment, the level of free fatty acid (FFA) was determined using fluorometry, while the level of triglyceride was evaluated using colorimetry on a BioVision™ assay kit (Bioptics) as per the kit's instructions.

| Immunohistochemistry staining
The lung tissue slices were dewaxed, rehydrated, and subjected to immunohistochemical staining using a two-step assay kit (PV-9000; Zsbio Commerce Store, Inc.) according to the kit's instructions. The slices were co-incubated with primary antibody pSTAT3 (1:200; Proteintech) at 4°C overnight, and then with an HRP-labelled secondary antibody 37°C for 1 hour.

| Reverse transcription-qPCR
Total RNA from lung tissues was extracted using an RNApure Total RNA Fast Extraction Kit (BioTeke Corporation) and reversely transcribed into cDNA using a PromeScript RT Kit (Takara Biotechnology Ltd.) as per the kit's instructions. Relative gene expression was determined using a real-time qPCR Kit (Gene Copoeia) on a PCR System (75000; Applied Biosystems) using the 2 -ΔΔCt method with U6 as the internal reference for miRNAs while GAPDH for mRNA.

| Determination of differentially expressed miRNAs
The whole-transcriptome library was prepared using a Ribo-Zero

| Chromatin immunoprecipitation -qPCR
The binding relationships between pSTAT3 and the promoter sequence of miR-378, miR-19b or miR-1281 were first predicted on JASPAR (http://jaspar.gener eg.net/). 23 Then, a chromatin immunoprecipitation (ChIP) analysis was performed using a ChIP assay kit (26157; Thermo Fisher Scientific Inc.). In brief, approximately 1 × 10 7 cells were fixed in 1% methanol and quenched using glycine. Then, the cells were washed in PBS, soaked in 1% halt cocktail-supplemented cold PBS and lysed by MNase and detached. Next, the mixture was subjected to ultrasonic treatment to destruct the nuclear membranes. The lysates were co-incubated with anti-pSTAT3 and protein G magnetic beads at 4°C overnight with normal rabbit antibody to IgG for negative control. The beads were washed 4 times, and the DNA was eluted using ChIP elution buffer, and the buffer was warm-incubated at 65°C for 1.5 hours. Next, a DNA purification kit was used for DNA recovery, and the purified DNA was determined using qPCR. These procedures were performed in triplicate.

| Dual-luciferase reporter gene assay
HEK293T cells were purchased from ATCC and incubated in Roswell Park Memorial Institute-1640 (Life Technologies) supplemented with 10% foetal bovine serum (Gibco, Thermo Fisher (Invitrogen). Twenty-four hours after transfection, the relative luciferase activity was determined using a dual-luciferase reporter kit (Promega Corporation). The firefly luciferase activity was normalized to the ranilla luciferase activity.

| Western blot analysis
Protein level of NFAT5 was determined by Western blot analysis.
In brief, total protein from lung tissues was collected. Then, 40 μg protein sample was run on 12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. The membranes were blocked by 5% skimmed milk for 2 hours, and then incubated with the primary antibody at 4°C overnight. After PBS washes, the membranes were further incubated with HRP-conjugated secondary antibody at room temperature for 2 hours. The membranes were visualized by the enhanced chemiluminescence reagent (EMD; Millipore), and the protein bands were analysed using the Image J 1.41 software (National Institutes of Health). SPSS 21.0 was used for data analysis (IBM Corp. Armonk, NY, USA).

| Statistical analysis
Data were in normal distribution according to Kolmogorov-SmiRnov test and presented as mean ± standard deviation (SD). The t test was used for data comparison between two groups, while one-way or two-way analysis of variance (ANOVA) was used among multiple groups with Tukey's multiple comparisons test used for post hoc test. The survival curve was plotted using the Kaplan-Meier method and analysed using the log rank test. Enumeration data were compared using the Fisher's exact test. P was obtained from two-tailed tests, and P < .05 was considered statistically significant.

| Expression of pSTAT3 is increased in a mouse model with T2DM-associated PTB
One month after STZ and NA co-treatment, the T2DM mice were further infected with Mtb to acquire PTB, and a diagram for model establishment is presented in Figure 1A. Thereafter, Masson's trichrome staining was performed, and a notable increase was found in lung fibrosis in Mtb-infected mice compared to the PBS-treated ones ( Figure 1B). In addition, the HE staining presented a large scale of inflammatory cell infiltration with disordered lung markings in the lung tissues after Mtb infection ( Figure 1C). According to the ELISA results, the production of pro-inflammatory cytokines including IL-6, IL-1β and TNF-α in mouse lung tissues was notably increased following Mtb infection ( Figure 1D). On the 50th day after Mtb infection, pSTAT3 expression in mouse lung homogenate, according to the ELISA kit detection again, was notably enhanced ( Figure 1E).
Likewise, the immunohistochemical staining identified a high expression level of pSTAT3 in mouse lung tissues, while was mainly in nuclei ( Figure 1F).

| Inhibition of pSTAT3 prolongs the lifetime of mice with T2DM-associated PTB
To further identify the roles of pSTAT3 in lung injury in mice with T2DM-associated PTB, 1 month after Mtb infection, each mouse was given AG-490 (1 mg/kg) twice every 10 days through i.p for a total of 30 days (Figure 2A). Then, the ELISA and immunohistochemical staining results identified that the pSTAT3 expression was notably decreased following AG-490 administration ( Figure 2B-C).
ELISA results also identified that the levels of TNF-α, IL-6 and IL-1β in lung tissue homogenate were decreased when pSTAT3 was reduced ( Figure 2D). In addition, we found that the median lifetime of mice with AG-490 treatment was 8.4 months (median survival days 252.7), while that of the control mice (PBS treatment) was 6.7 months (median survival days 204.3) ( Figure 2E). Importantly, it was found that AG-490 treatment led to a notable decline in Mtb concentration in mouse lung tissues ( Figure 2F). Moreover, we also found that the levels of FFA and triglyceride in serum were decreased after AG-490 treatment ( Figure 2G). In addition, the Masson's trichrome staining suggested that when pSTAT3 was suppressed by AG-490, the collagen deposition in mouse lung tissues was notably decreased, namely the tissue fibrosis was reduced ( Figure 2H).

| pSTAT3 transcriptionally suppresses miR-19b/1281 to up-regulate NFAT5 expression
To identify the potentially involved molecules, a miRNA transcriptome analysis was performed to identified the differentially expressed miRNAs in mouse lung tissues following AG-490 treatment. A total of 341 miRNAs were significantly changed under the |Log 2 FC > 2| and P < .05 criteria, and a heat map for top 20 differentially expressed miRNAs is presented in Figure 3A. We then measured miR-126, miR-19b, miR-392, miR-1281 and miR-378 expression in lung tissues using reverse transcription-qPCR (RT-Qpcr), and found the expression of these miRNAs was initially decreased after Mtb infection but then increased following further AG-490 treatment ( Figure 3B). Since pSTAT3 can transcriptionally regulate gene expression, we then predicted the binding relationships between pSTAT3 and the promoter sequences of the 3 mostly changed miRNAs miR-19b, miR-1281 and miR-19a F I G U R E 1 Expression of pSTAT3 is increased in mice model with T2DM-associated PTB. A, 1 mo after diabetes induction, mice were further exposed to ~100 CFU of aerosolized Mtb; B, fibrosis in mouse lung tissues measured by Masson's trichrome staining; C, histopathologic changes in mouse lung tissues determined by HE staining; D, protein levels of inflammatory factors (IL-6, IL-1β and TNF-α) in lung tissue homogenate determined using ELISA kits; E, expression level of pSTAT3 in lung tissue homogenate determined by ELISA; F, pSTAT3 expression and distribution in tissue sections evaluated by immunohistochemistry staining. Each spot in the images indicates one sample. N = 8 in each group. Three independent experiments were performed. Data were expressed as mean ± SD as the outcome of the unpaired t test. *P < .05, **P < .01 on JASPAR bio-information system ( Figure 3C-D). The ChIP-qPCR was performed and identified that pSTAT3 could suppress miR-19b and miR-1281 transcription but excluding miR-19a ( Figure 3E).

| miR-19b/1281 agomiR or sh-NFAT5 reduces inflammation in lung tissues and lung epithelial cell apoptosis in T2DM-PTB mice
To further confirm the effects of above molecules in lung injury in model mice, 1 month after Mtb infection, miR-19b/1281 agomiR, or the shRNA of NFAT5 (5 nmol/100 μL) was given into the mice twice every 10 days through i.p. ( Figure 4A). Thereafter, the miR-19b/1281 expression was increased and NFAT5 expression was decreased according to RT-qPCR results ( Figure 4B). Next, we found that overexpression of miR-19b or miR-1281, or down-regulation of NFAT5 led to a decline in the levels of TNF-α, IL-6 and IL-1β in lung tissues ( Figure 4C), and the lifetime of model mice was prolonged as well ( Figure 4D). In addition, we also found that the CFU of Mtb in mouse lung was notably reduced after miR-19b/1281 up-regulation or NFAT5 down-regulation ( Figure 4E). In the meantime, the fibrosis in lung tissues was decreased following miR-19b/1281 agomiR, or NFAT5 shRNA administration ( Figure 4F). The apoptosis of lung epithelial cells was inhibited as well ( Figure 4G).

| Lv-NFAT5 partially blocks the effects of miR-19b/miR-1281 agomiR
Following the findings above, we further co-transfected LV-NFAT5 (overexpressing NFAT5) and miR-19b/miR-1281 agomiR in mouse 30 days after Mtb infection ( Figure 5A). Then, the NFAT5 expression in mouse lung tissue homogenate was increased according to the Western blot analysis ( Figure 5B). Then, it was

| miR-19b/1281 antagomiR inhibits the functions of AG-490 in mice
To further validate the involvement of miR-1281/19b in AG-490mediated events, 30 days after AG-490 treatment, cells were further treated with miR-19b/1281 antagomiR (5 nmol/100 μL) twice every 10 days through i.p for a total of 30 days ( Figure 6A). Then, it was found that the miR-19b/1281 expression in lung tissues was declined ( Figure 6B). ELISA results found that the IL-6, IL-1β and TNF-α levels in lung tissues were increased following miR-19b/1281 down-regulation ( Figure 6C). Accordingly, the Mtb content, lung fibrosis and apoptosis of lung epithelial cells inhibited by AG-490 were aggravated when miR-19b/1281 was inhibited ( Figure 6D-F).

| Overexpression of NFAT5 blocks the roles of AG-490 in mice
In addition, T2DM-PTB mice treated with AG-490 were further administrated with LV-NFAT5 ( Figure 7A). The Western blot analysis identified an increased level of NFAT5 in mouse tissue homogenate ( Figure 7B). Reasonably, it was found that the secretion of inflammatory cytokines and Mtb content as well as epithelial cell apoptosis in lung tissues inhibited by AG-490 was promoted after NFAT5 overexpression ( Figure 7C-D). In addition, the HE staining found that the inflammatory cell infiltration in lung tissues was increased and the Masson's trichrome staining identified increased fibrosis in lung tissues ( Figure 7E-F).

| D ISCUSS I ON
Given to the increasing incidence of Mtb reactivation and the especially high recurrence and mortality rate in TB patients associated Three independent experiments were performed. Data were expressed as mean ± SD. In panel B, data were analysed using the unpaired t test, while data in panels C, E, F and G were analysed by one-way ANOVA and Tukey's multiple comparison test. *P < .05, **P < .01 the proliferation of intracellular pathogens including Mtb. 29 This pro-survival role of STAT3 in Mtb was further evidenced by reduced CFU of Mtb following STAT3 silencing. 29 Likewise, STAT3 activation has also been documented to increase the risk of successful Mtb infection in mice. 24 As a transcription factor, pSTAT3 is capable of regulating gene expression transcriptionally. After identification of the nuclear abundance of the pSTAT3 sub-cellular localization, we determined the potential miRNAs regulated by pSTAT3 through integrated miRNA microarray analysis, RT-qPCR, online prediction and ChIP-qPCR assays. Consequently, miR-19b and miR-1281 found as two major miRNAs that were up-regulated following AG-490 treatment. miRNAs play versatile roles in innate immune response and bacterial infection in lung. 30,31 For instance, miR-27b was found to suppress the secretion of pro-inflammatory factors and the nuclear factor-kappa B (NF-κB) signalling pathway to avoid an excessive inflammation during Mtb infection. 32 Likewise, miR-20b was suggested to suppress Mtb-induced inflammation in lung of mice through directly binding to NLRP3 inflammasome. 33 In terms of miR-19, a recent study has focused on its role in immune and inflammatory responses in human cancers. 34 In addition, protective roles of miR-19b-3p in patients with sepsis have been identified as well by relieving the levels of pro-inflammatory factors IL-6 and TNF-α. 35 MiR-19b-3p was also reported to attenuate IL-1β-induced inflammatory injury in chondrocytes through targeting GRK6. 16 As for miR-1281, it has been identified as a tumour suppressor in many malignancies such as glioma, 17 osteosarcoma 36 and gastric cancer. 37 Interestingly, it also presented protective functions in macrophages against Mtb infection by reducing the programmed necrosis and apoptosis. 38 Importantly, in this paper, F I G U R E 5 Lv-NFAT5 partially blocks the effects of miR-19b/miR-1281 agomiR. A, 1 mo after diabetes induction, mice were exposed to ~100 CFU of aerosolized Mtb; another 30 d later, each mouse was further given miR-19b/1281 agomiR, or the LV overexpressing NFAT5 (5 nmol/100 μL) twice every 10 d through i.p for a total of 30 d; B, protein level of NFAT5 in mouse lung tissues determined by Western blot analysis; C, secretion of pro-inflammatory factors in mouse lung tissue homogenate determined by ELISA kits; D, Mtb content in lung tissues; E, lung fibrosis in mouse lung tissues determined by Masson's trichrome staining; F, apoptosis in lung tissues determined by TUNEL. Each spot in the images indicates one sample. N = 8 in each group Three independent experiments were performed. Data were expressed as mean ± SD compared by one-way ANOVA and Tukey's multiple comparison test. *P < .05, **P < .01 Based on the findings above, we further identified NFAT5 as a targeting mRNA of both miR-19b and miR-1281 through online prediction and dual-luciferase reporter genes assays. NFAT5 has been noted to play crucial functions in immune and inflammatory response regulation. 19,39 Here, the evidence that NFAT5 expression was increased following Mtb infection but reduced after AG-490 administration suggested the potential involvement of NFAT5 in Mtb infection. Thereafter, we found artificial silencing of NFAT5 reduced inflammatory response, fibrosis and normal epithelial cell apoptosis in mouse lung tissues. Similarly, a previous study also found that both gene and protein expression of NFAT5 was induced by Mtb infection, while NFAT5 down-regulation led to a decline in Mtbstimulated HIV-1 replication in co-infected macrophages by interacting with the toll-like receptor pathway. 40 Additionally, up-regulation of NFAT5 blocked the protective roles of miR-19b/miR-1281 agomiR in this study, which further evidenced its implication in miR-19b/ miR-1218-mediated events.
To sum up, our present study suggested that pSTAT3 was at least partially responsible for Mtb-induced inflammation and lung injury and Mtb survival in a mouse model with T2DM-associated PTB.
Silencing of pSTAT3 reduced the symptoms and inflammation as well as Mtb survival through the up-regulation of miR-19b/miR-1281 which suppressed NFAT5 expression. This study may offer novel insights into T2DM-PTB treatment. In addition, considering the possibility that AG-490 might affect other pathways such as JAK, EGFR and STAT1/5, and we would like to discuss the potential involvement of other target pathways in the protective events mediated by AG-490 in our future studies.

CO N FLI C T O F I NTE R E S T
The authors declare no potential conflicts of interest. were performed. Data were expressed as mean ± SD compared by the unpaired t test. *P < .05, **P < .01