Ginsenoside Rg1 Inhibits STAT3 Expression by miR-15b-5p to Attenuate Lung Injury in Mice with Type 2 Diabetes Mellitus-Associated Pulmonary Tuberculosis

Type 2 diabetes mellitus (T2DM) has been regarded as a critical risk factor for pulmonary tuberculosis (PTB). Ginsenoside Rg1 has been identified as a potential therapeutic agent for T2DM by suppressing the inflammatory response. However, the effect of Rg1 on T2DM-associated PTB has not been reported. In this study, we aimed to explore the function of Rg1 in the regulation of T2DM-associated PTB. We established a T2DM-associated PTB mouse model and found that the fibrosis of lung tissues was inhibited by Rg1 in T2DM-associated PTB mice. The lung injury of T2DM-associated PTB mice was repressed by Rg1. Moreover, the levels of IL-6, TNF-α, and IL-1β in the lung tissues and serum were decreased by Rg1 in T2DM-associated PTB mice. The treatment with Rg1 inhibited the levels of free fatty acid and enhanced the expression of miR-15b-5p in lung tissues of T2DM-associated PTB mice. MiR-15b-5p targeted and inhibited the STAT3 expression. The expression of STAT3 was downregulated by Rg1, while the inhibition of miR-15b-5p reversed the downregulation. The expression of miR-15b-5p was reduced, but the expression of STAT3 was upregulated in the lung tissues of T2DM-associated PTB mice. We validated that miR-15b-5p attenuated inflammation and lung injury in the T2DM-associated PTB mouse model. The overexpression of STAT3 or the suppression of miR-15b-5p restored lung fibrosis and injury inhibited by Rg1 in T2DM-associated PTB mice. Meanwhile, the Rg1-repressed levels of IL-6, TNF-α, and IL-1β were enhanced by the overexpression of STAT3 or the suppression of miR-15b-5p. In addition, the levels of free fatty acid repressed by Rg1 were reversed by STAT3 overexpression and miR-15b-5p inhibition. Thus, we conclude that ginsenoside Rg1 inhibits the STAT3 expression by miR-15b-5p to attenuate lung injury in mice with type 2 diabetes mellitus-associated pulmonary tuberculosis.


Introduction
Type 2 diabetes mellitus (T2DM) has been regarded as a critical risk factor for pulmonary tuberculosis (PTB) [1]. Individuals with T2DM show two to eight times the risk of developing PTB [2]. Nowadays, the emerging T2DM pandemic presents a major challenge in global public health, which poses T2DM and PTB as major health issues, especially in areas where PTB is rampant [3]. Pathologically, infection caused by mycobacterium tuberculosis (MTB) is the major factor of PTB [4,5]. e current first-line therapy for PTB includes isoniazid, rifampicin, ethambutol, and pyrazinamide and lasts for more than 6 months [6,7]. T2DM can not only increase the susceptibility to MTB but also affect the progression of PTB and the unresponsiveness to therapies, as well as the risk of death and relapse [8,9]. erefore, the development of novel therapeutic strategies for T2DM-associated PTB is imperative.
Traditional Chinese medicine has been widely applied in the therapy and prevention of various diseases [10,11]. Ginseng is one of the most widely used and studied traditional Chinese medicines worldwide, owing to its diverse pharmacological functions [11]. Among the bioactive components of ginseng, ginsenoside Rg1 (Rg1) is extensively studied [12][13][14]. Treatment with Rg1 increases the expression of VEGF and promotes the proliferation of endothelial cells and angiogenesis in vitro and in vivo in ischemic models [15]. In glucose-induced hepatic gluconeogenesis, Rg1 elevated phosphorylation of AKT and decreased the level of glucose [16]. A previous study indicated that Rg1 is a potential therapeutic agent for T2DM which works by suppressing inflammatory responses [17]. Nevertheless, the function of Rg1 in T2DM-associated PTB has not been reported.
Signal transducer and activator of transcription-3 (STAT3) is an important transcriptional factor of JAK signaling and plays critical roles in regulating proliferation, differentiation, apoptosis, inflammatory response, and immune reaction [18][19][20]. STAT3 participates in the expression and function of cytokine receptors, such as einterleukin-6 (IL-6), interferon-c (IFN-c), and platelet-derived growth factor (PDGF) [20]. Moreover, STAT3 is involved in the replication and pathogenesis of viruses and causes the adverse effects of viral infections [21]. However, studies have indicated the role of STAT3 as a mediator in MTB infection [22,23].
MicroRNAs (miRNAs) are a form of noncoding RNAs with lengths of 20 nucleotides, which modulate target miRNA stability and transcription and consequently participate in various cellular processes including inflammation and immune responses [24]. MiR-15b-5p is a recently reported regulator of disease progression. Studies have suggested that miR-15b-5p participates in the regulation of cell proliferation, metastasis, autophagy, and other cell processes during the development of cancers and cardiovascular diseases [25][26][27]. A recent work also pointed out that the abnormal level of miR-15b-5p in urine may be an indicator of kidney dysfunction [28]. Yet, no studies have currently highlighted the role of miR-15b-5p in T2DMassociated PTB.
In this study, we determined the therapeutic effects of Rg1 against T2DM-associated PTB using the in vivo mouse model and further explored the regulatory mechanisms involving the miR-15b-5p/STAT3 axis. Our work presented Rg1 as a novel and effective treatment agent for T2DMassociated PTB.

Mouse Model.
We first established the T2DM mouse model induced by streptozotocin (STZ). In short, C57BL/6J mice aged 4 weeks were purchased from Vital River Laboratory (Beijing, China). e STZ (60 mg/kg) was intraperitoneally injected for 5 consecutive days. en, the blood glucose level was checked once a week after the STZ injection. A value over 16.7 mM was regarded as the successful establishment of a T2DM mouse model.
One month after STZ injection, the T2DM mice were treated with MTB. H37Rv (ATCC, USA) for PTB induction [29,30]. e H37Rv was diluted in PBS to 2 × 10 6 colonyforming units (CFU)/mL, and then it was atomized by an atomizer. T2DM mice were exposed to the bacterium for 15 minutes. Rg1 was brought from Solarbio (China) and was dissolved in DMSO for stock. One month after H37Rv infection, Rg1 (2 mg/ml) and oligonucleotides (10 nmol) were administrated two times every 10 days via intraperitoneal injection within one month. PTB mice in the control group were treated with PBS. After treatment, the mice were sacrificed and lung tissues and blood were collected. To check CFU in the lung, the tissues were homogenized and incubated on agar plates at 37°C with 5% CO 2 for 20 days. All animal experiments complied with the ARRIVE guidelines and were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 8023, revised 1978). e animal experiments in this work were approved by the animal care and use committee at the Xi'an Chest Hospital.

Serum Analysis.
e blood was centrifuged so as to collect serum, and then the level of free fatty acid (FFA) and triglyceride (TG) were measured by a commercial detection kit (Solarbio, China) following the manufacturer's protocols.

Enzyme-Linked Immuno Sorbent Assay (ELISA).
e levels of IL-6, IL-1β, and TNF-α in lung tissues were measured by commercial ELISA kits and by microplate reader.

Histopathologic Analysis.
Lung tissues were fixed in 4% paraformaldehyde (PFA), embedded with paraffin, and made into 4 μm slices. e tissue damage and fibrosis were checked by haematoxylin and eosin (HE) staining and Masson's trichrome staining (Beyotime, China). For HE staining, samples were deparaffinated, rehydrated, and stained with haematoxylin and eosin. For Masson's trichrome staining, the samples were stained with haematoxylin, ferric oxide, and acid fuchsin following the commercial kit's instructions.

Cell Culture and Treatment.
e human kidney cell line HEK293T was brought from Shanghai Cell Bank of Chinese Science Academy, maintained in CMEM (Hyclone, USA) with 10% FBS (Gibco, USA) and 1% penicillin/streptomycin. e STAT3 overexpressing vectors, siSTAT3, miR-15b-5p mimics, and miR-15b-5p inhibitors, and corresponding negative controls were synthesized by RiboBio (China). Cell transfection was performed by mixing 50 nM of oligonucleotides using Lipofectamine 2000 (Invitrogen, USA) as per the manufacturer's protocol.

Real-Time PCR Analysis.
e lung tissues were homogenized, and total RNA was extracted by using a Trizol reagent ( ermo, USA). e cDNA was synthesized from RNA by using a PromeScript RT kit (Takara, Japan) and quantified by SYBR Green following the 2-ΔΔCt method. e RNA and miRNA expression was normalized to GAPDH and U6, respectively.

Western Blotting.
e tissues and cells were homogenized by using RIPA buffer that contains antiprotease mixture ( ermo, USA) to obtain total protein. e protein samples were quantified with a BCA kit (Beyotime, China), 2 Evidence-Based Complementary and Alternative Medicine resolved in SDS-PAGE, shifted into NC membranes, blocked with 5% skim milk at room temperature for 1 hour, and hatched with primary antibodies against STAT3, phosphorylated STAT3, and β-actin (Proteintech, China, 1 : 500) at 4°C overnight. e next day, the membranes were incubated with HRP-conjugated, antimouse, or antirabbit secondary antibodies, visualized by an ECL substrate (Millipore, USA) on a gel image system (Bio-Rad, USA).

Luciferase Reporter Gene Assay.
e STAT3 3′UTR promoter reporter vectors that contain either wild-type or mutated miR-25b-5p binding site were cotransfected with pRL-SV40 Renilla vectors into the cell and incubated for 48 hours. e cells were then lysed, and luciferase activity was checked by a dual-luciferase reporter assay system (Promega, USA).

2.9.
Statistics. Data were shown as mean ± standard deviation (SD) and analyzed by using SPSS 20.0 (USA). e student's t-test and one-way analysis of variance (ANOVA) were performed to determine a comparison between the two groups and among multiple groups. Values of p < 0.05 were regarded as statistically significant.

MiR-15b-5p Expression Is Reduced and STAT3 Expression Is Increased in the T2DM-Associated PTB Mouse Model.
We established a T2DM-associated PTB mouse model and assessed the expression of miR-15b-5p and STAT3. e fibrosis of lung tissues was observed in the T2DM-associated PTB mice (Figure 1(a)), and we observed the lung injury of T2DM-associated PTB mice (Figure 1(b)). Meanwhile, the enhanced levels of IL-6, TNF-α, and IL-1β in the lung tissues and the serum of T2DM-associated PTB mice compared with the control (PBS) group were validated (Figure 1(c)). Significantly, the expression of miR-15b-5p was reduced but STAT3 expression was upregulated in the lung tissues of T2DM-associated PTB mice, compared with the PBS group (Figures 1(d) and 1(e)).

Rg1 Attenuates Inflammation and Lung Injury in the T2DM-Associated PTB Mouse Model.
We were then interested in the effect of Rg1 on the inflammation and lung injury of T2DM-associated PTB mice. Interestingly, we observed that the fibrosis of lung tissues was inhibited by Rg1 treatment in T2DM-associated PTB mice, compared with the ones treated with PBS as control (Figure 2(a)). Consistently, the lung injury of T2DM-associated PTB mice was repressed by Rg1 (Figure 2(b)). Moreover, the levels of IL-6, TNF-α, and IL-1β in the lung tissues and serum were decreased by Rg1 treatment in T2DM-associated PTB mice, compared with the control group (Figures 2(c) and 2(d)). Moreover, the treatment with Rg1 inhibited the levels of free fatty acid (Figure 2(e)).

MiR-15b-5p Relieves Inflammation and Lung Injury in the T2DM-Associated PTB Mouse Model.
We then validated the function of miR-15b-5p in T2DM-associated PTB mice and confirmed the upregulation of miR-15b-5p by miR-15b-5p mimic in lung tissues of T2DM-associated PTB mice ( Figure  S1(a)). We found that the fibrosis of lung tissues was attenuated by miR-15b-5p in T2DM-associated PTB mice ( Figure S1(b)).

MiR-15b-5p Relieves Inflammation and Lung Injury in the T2DM-Associated PTB Mouse Model by Targeting STAT3.
We found that the fibrosis of lung tissues was promoted by miR-15b-5p inhibitor compared with the control group, while the depletion of STAT3 reversed the effect in T2DMassociated PTB mice ( Figure S2(a)). e lung injury of T2DM-associated PTB mice was induced by miR-15b-5p inhibitor and STAT3 knockdown inhibited the phenotype in T2DM-associated PTB mice ( Figure S2(b)). Moreover, the levels of IL-6, TNF-α, and IL-1β in the lung tissues and serum were enhanced by miR-15b-5p inhibitor, compared with the control group, but the STAT3 depletion reversed the effect in T2DM-associated PTB mice (Figures S2(c) and S2(d)). Besides, the treatment with miR-15b-5p inhibitor upregulated the levels of free fatty acid, whereas the silencing of STAT3 reversed the effect ( Figure S2(e)).

Rg1 Inhibits Inflammation and Lung Injury in the T2DM-Associated PTB Mouse Model by miR-15b-5p/STAT3 Axis.
Next, we observed that the overexpression of STAT3 or the suppression of miR-15b-5p restored lung fibrosis and injury inhibited by Rg1 in T2DM-associated PTB mice (Figures 4(a) and 4(b)). Meanwhile, the Rg1-repressed levels of IL-6, TNF-α, and IL-1β were enhanced by the overexpression of STAT3 or by the suppression of miR-15b-5p (Figures 4(c) and 4(d)). In addition, the levels of free fatty acid repressed by Rg1 were reversed by STAT3 overexpression and miR-15b-5p inhibition (Figure 4(e)).   (e) e expression of STAT3 was detected by Western blot in 293T cells treated with Rg1 and miR-15b-5p inhibitor. Cont represents the PBS treatment group; Rg1 represents the Rg1 treatment group; Rg1 + miR-15b-5p inhibitor represents the treatment with Rg1 and miR-15b-5p inhibitor; WTrepresents the wild-type vectors; and MUTrepresents the mutated vectors; * * p < 0.05 vs Cont group, ## p < 0.05 vs Rg1 group.

Discussion
T2DM has been regarded as a critical risk factor for PTB [31,32]. Ginsenoside Rg1 has been identified as a potential therapeutic agent for T2DM that works by suppressing inflammatory response [33]. But the effect of Rg1 on T2DMassociated PTB has not been reported. In the current world, we uncovered the function of Rg1 in the inhibition of T2DM-associated PTB, and we identified that the fibrosis of lung tissues was inhibited by Rg1 in T2DM-associated PTB mice. e lung injury of T2DM-associated PTB mice was repressed by Rg1. Rg1 has presented multiple biomedical activities in several diseases. It has been reported that ginsenoside Rg1 enhances cerebral angiogenesis in ischemic mice by the PI3K/Akt/mTOR signaling [15]. Ginsenoside Rg1 protects PTSD-like behaviors by inducing synaptic proteins, as well as by repressing Kir4.1 and TNF-α in the hippocampus [34]. Ginsenoside Rg1 attenuates ischemic/reperfusion-induced neuronal injury by the miR-144-Nrf2-ARE signaling [35]. Ginsenoside Rg1 inhibits cardiomyocyte inflammation and apoptosis by the TLR4/NF-kB/NLRP3 signaling [12].
Ginsenoside Rg1 suppresses glucagon-associated hepatic gluconeogenesis by the interaction of Akt-FoxO1 [16]. Consistent with the previous studies, we identified that the levels of inflammatory factors, IL-6, TNF-α, and IL-1β, in the lung tissues and serum were decreased by Rg1 in T2DMassociated PTB mice. Moreover, the treatment with Rg1 inhibited the levels of free fatty acid. ese data elucidate the critical impact of Rg1 on T2DM-associated PTB, enriching the understanding of the function of natural compounds in the treatment of T2DM-associated PTB.
Furthermore, it has been reported that the phosphorylation of STAT3 represses the microRNA-19b expression to promote lung injury in T2DM-associated PTB [29]. e enhancement of miR-196b-5p inhibits the uptake of BCG by regulating SOCS3 and promoting STAT3 in PTB [36]. ese studies indicated the participation of STAT3 in PTB. Our data revealed that the treatment with Rg1 enhanced the expression of miR-15b-5p and suppressed the expression of STAT3 in lung tissues of T2DM-associated PTB mice. Further mechanical research demonstrated that miR-15b-5p targeted and inhibited the STAT3 expression, and the inhibition of miR-15b-5p reversed the Rg1-downregulated  (c, d). e levels of inflammation factors were detected by ELISA in lung tissues and serum. (e) e levels of free fatty acid were analyzed. Cont represents PTB mice treated with PBS as control; Rg1 represents the Rg1 treatment group; Rg1 + miR-15b-5p inhibitor represents the treatment with Rg1 and miR-15b-5p inhibitor; Rg1 + STAT3 represents the treatment with Rg1 and the overexpression of STAT3; * * p < 0.05 vs Cont group, ## p < 0.05 vs Rg1 group. STAT3 level. Besides, the expression of miR-15b-5p was reduced but STAT3 expression was upregulated in the lung tissues of T2DM-associated PTB mice. ese findings suggested that Rg1 may upregulate the miR-15b-5p to suppress STAT3 expression and function in the PTB model. Furthermore, we validated that miR-15b-5p relieves inflammation and lung injury in the T2DM-associated PTB mouse model. e overexpression of STAT3 or the suppression of miR-15b-5p restored lung fibrosis and injury inhibited by Rg1 in T2DM-associated PTB mice. Meanwhile, the Rg1-repressed levels of IL-6, TNF-α, and IL-1β were enhanced by the overexpression of STAT3 or by the suppression of miR-15b-5p. In addition, the level of free fatty acid repressed by Rg1 was reversed by STAT3 overexpression and miR-15b-5p inhibition.
ese data imply that Rg1 may affect T2DM-associated PTB by targeting the miR-15b-5p/STAT3 signaling. However, the regulatory mechanisms underlying Rg1-upregulated miR-15b-5p expression still need further exploration and this regulatory axis should be verified in in vitro cell models. Other potential targets and mechanisms of Rg1 in the treatment of T2DMassociated PTB should be explored by more studies.
In conclusion, we can say that ginsenoside Rg1 inhibits STAT3 expression by miR-15b-5p to attenuate lung injury in mice with type 2 diabetes mellitus-associated pulmonary tuberculosis.

Data Availability
No data were used in this study.

Conflicts of Interest
e authors declare that they have no conflicts of interest. Figure S1. MiR-15b-5p relieves inflammation and lung injury in the T2DM-associated PTB mouse model. (A-E) e T2DM-associated PTB mice were treated with miR-15b-5p mimic. (A) e expression of miR-15b-5p in lung tissues was analyzed by qPCR. (B) e fibrosis of lung tissues was detected by Masson staining. (C) e lung injury was analyzed by H&E staining. (D) e levels of inflammation factors were detected by ELISA in lung tissues and serum. (E) e levels of free fatty acid were analyzed. Cont represents PTB mice treated with PBS as control and miR-15b-5p mimic represents the treatment with miR-15b-5p mimics; * * p < 0.05 vs Cont group. Figure S2. MiR-15b-5p relieves inflammation and lung injury in the T2DMassociated PTB mouse model by targeting STAT3. (A-E) e T2DM-associated PTB mice were treated with miR-15b-5p inhibitor and STAT3 siRNA. (A) e fibrosis of lung tissues was detected by Masson staining. (B) e lung injury was analyzed by H&E staining. (C, D) e levels of inflammation factors were detected by ELISA in lung tissues and serum. (E) e levels of free fatty acid were analyzed. Cont represents PTB mice treated with PBS as control; miR-15b-5p inhibitor represents the treatment with miR-15b-5p inhibitor; and miR-15b-5p inhibitor + siSTAT3 represents the treatment with miR-15b-5p inhibitor and the depletion of STAT3; * * p < 0.05 vs Cont group; ## p < 0.05 vs miR-15b-5p inhibitor group. (Supplementary Materials)