Background
Acute respiratory distress syndrome (ARDS) is a common and clinically complex acute pulmonary inflammatory syndrome with a mortality rate of up to 40% [
1]. Although after decades of exploration, there is still no special treatment for ARDS, treatment options are still confined to supportive care such as the protective mechanical ventilation and prone position ventilation and liquid conservative strategy [
2]. Immune cell-mediated aberrant inflammatory responses play a significant role in the development of ARDS [
3,
4], but glucocorticoid, surfactants, and other anti-inflammatory drugs, have been failed to show benefits [
5]. In order to alleviate aberrant inflammatory of ARDS and ameliorate lung injury, it is urgent to explore additional therapeutic measures.
Mesenchymal stem cells (MSCS), with their pluripotent and immunomodulatory functions, are expected to be a novel therapy for ARDS. Our previous studies showed that MSC can repair alveolar epithelial cells and vascular endothelial cells to ameliorate lung injury in ARDS [
6,
7]. MSCs also possess immunomodulatory properties to immune cells, such as T, B, NK cells, macrophages, neutrophils and dendritic cells (DCs) [
8‐
11]. DCs are the most important specialized antigen presenting cells and play an important role in the regulation of lung immune response network. Previous studies have shown that the maturity of classical dendritic cells (cDCs) in the lungs of ALI mice increased significantly, while inhibition of DCs maturation reduced the inflammatory response and pathological damage in the lungs [
12,
13]. MSC induced mature DCs (mDCs) differentiation into regulatory DC (DCreg) with low expression of CD80 and CD86 DCreg, which induced the expression of CD4, CD25, and Foxp3 in primary splenocytes isolated from mice [
14]. MSC also abrogated the capacity of mDC migration to CCL19, or for DC to display MHC class II peptide complexes recognized by specific antibody, or ovalbumin-pulsed DC to support antigen-specific CD4
+ T cell proliferation [
15]. Therefore, transplantation of MSC to regulate the function of mature DC may bring new light for ARDS treatment, but the regulatory mechanism remains to be elucidated.
Many studies suggest that the immunosuppressive effect of MSCs occurs through paracrine or cell-to-cell contact mechanisms [
16‐
18]. Cell–cell interactions have been demonstrated in the treatment of hematopoietic system-related diseases [
16,
18]. The interaction between the Notch receptor and Notch ligand plays a key role in dendritic cell differentiation or MSC protection against inflammation [
19,
20]. Cahill et al. [
19] reported that MSCs prevented DC antigen presentation, and knocking down Jagged-1 in MSCs partially reversed this effect. The cell-binding Notch ligand activates notch-1 and notch-2 receptors, leading to accumulation of immature DCs [
21]. Whether the Notch pathway is an important link in MSC-induced DCreg generation to alleviate acute lung injury remains unclear.
In the present study, we assessed changes in the lung cDC phenotype and their association with lung pathological injury in ALI mice, evaluated the effect of MSC transplantation on the lung cDC phenotype in ALI mice and the cellular mechanisms between them, and demonstrated that MSCs induce a shift in cDCs from a proinflammatory mature type to a tolerant DC to alleviate lung injury.
The aim of this study was to investigate whether the protective effect of MSCs on LPS-induced ALI was achieved by intercellular contact to induce mDCs to differentiate into DCregs and whether the mechanism was related to activation of the Notch pathway. We used coculture experiments to test the phenotype of dendritic cells, the stimulation of T cell proliferation, and the expression of PD-L1. To explore the mechanism of MSC-induced mDC conversion to DCregs to alleviate LPS-induced ALI, we used the Notch pathway inhibitor DAPT (a gamma secretase inhibitor) in vivo and in vitro and evaluated lung pathological injury, DC phenotype and function, and pulmonary edema. This study provides an immunological explanation for mesenchymal stem cell reduction of LPS-induced lung injury.
Materials and methods
Animals
Male wild-type 6- to 8-week-old BALB/c and C57BL/6 specific pathogen free (SPF) mice (Laboratory Animal Center of Yangzhou University, Yangzhou, China) were maintained under specific pathogen-free conditions. A SPF animal room and super-clean bench in the laboratory were used for animal feed and experimental manipulation respectively during the whole experiment. The Animal Care and Use Committee of Southeast University approved all of the experimental procedures (Ethics Committee Approval Number: 20170105007).
MSC culture
Mouse mesenchymal stem cells and dendritic cells were used in the present study. MSCs derived from the bone marrow of C57BL/6 mice were purchased from Cyagen Biosciences Inc. (Guangzhou, China). The supplier identified MSCs based on the cell surface phenotype and pluripotency. MSCs were cultured in DMEM/F12 containing 10% fetal bovine serum (Wisent, Nanjing, China) and grown in a humidified 5% CO2 sterile incubator at 37 °C.
Bone marrow isolation and dendritic cell culture
Bone marrow (BM)-derived DCs were generated as previously described with minor modifications [
22,
23]. BM cells were extracted from the medullary cavity of the femur and tibia in super-clean bench. The erythrocytes were lysed using lysing buffer (BD Pharm Lyse™, USA), washed three times in phosphate-buffered saline (PBS) and cultured in 100 mm dishes with 2 × 10
6 cells in fresh RPMI 1640 (Wisent, Nanjing, China), containing 10% FBS (Wisent, Nanjing, China), 40 ng/mL recombinant murine granulocyte–macrophage colony-stimulating factor (GM-CSF; NOVUS), and 40 ng/mL recombinant murine interleukin-4 (IL-4; NOVUS) in a humidified 5% CO
2 sterile incubator at 37 °C. For the isolation of immature DCs (imDCs), non-adherent cells were gently washed away on day 3, half of the culture supernatant was collected and centrifuged, and the cell pellet was resuspended in 5 mL fresh medium as described above and returned to the original plate at day 5. The remaining loosely adherent cell clusters were collected and purified by anti-CD11c micromagnetic beads (Miltenyi Biotec) on day 6. Purified imDCs were cultured for an additional 24 h under stimulation with 50 ng/mL bacterial lipopolysaccharide (LPS; Sigma-Aldrich) and were used as mDCs. The purity of CD11c
+ cells was greater than 90% (monoclonal antibodies against CD11c, Miltenyi Biotech, Bergisch Gladbach, Germany). Cytofluorimetric analysis was performed to evaluate the DC maturation phenotype (monoclonal antibodies against CD40, CD86, and MHC-II, Miltenyi Biotech, Bergisch Gladbach, Germany).
DC and MSC coculture
mDCs (5 × 105 cells/well) were seeded into 6-well plates 4 h after MSCs (5 × 106 cells/well) in RPMI 1640 containing 5% fetal bovine serum (FBS); in the presence or absence of DAPT (5 μM) at 37 °C in 5% CO2. After coculture for 72 h, the suspended dendritic cells were collected for examination or applied to subsequent experiments.
ALI model
In SPF animal room, the ALI model was induced as previously described with minor modifications [
7]. Briefly, mice were intraperitoneally injected with 50 mg/kg pentobarbital. LPS (5 mg/kg) (Sigma-Aldrich) was delivered to the lungs by transtracheal injection, and the incision was sewn up. The mice were returned to the cage until they were fully awake.
Reagent treatments
Previous studies have found that γ-secretase inhibitors impair MSC-induced DCreg production by inhibiting the Notch pathway [
18,
19]. To determine the inhibitory effect of DAPT on the production of MSC-induced tolerant DCs within 3 days, MSC + DC + DAPT group: DAPT (Absin, China) was added to MSC-DC or recombinant Jagged1 cultures at 1 mΜ on day 0. When the culture medium was changed, DAPT was added to the fresh medium at the same concentrations. Because DAPT was diluted in DMSO, the control MSC-DCs, Jagged1-DCs, mDC were cultured with DMSO. GM-CSF was added to the medium in each group to maintain the final concentration of 20 ng/mL.
Experimental groups and sample acquisition
To investigate the changes in lung DCs and pathological damage at different times after ALI modeling, the mice were randomly assigned to one of the following seven groups (n = 6 mice per group): 24 h, 12 h, 6 h, 4 h, 2 h, 1 h and 0 h. Mice in the 24, 12, 6, 4, 2, 1 and 0 h (control) groups were sacrificed after an intratracheal injection of 5 mg/kg lipopolysaccharide (LPS) (Additional file
1: Fig. S1A). Lung tissue was collected for single cell isolation and histological examination in accordance with slightly modified methods [
12].
After confirming the lung DC maturation time in ALI mice, MSCs were administered to the mice at 4 h after LPS, and their effects on lung mDCs and lung pathological damage were observed. The mice were randomly assigned to one of the following three groups (n = 6 mice per group): MSC + ALI group (MSC) mice received MSCs (500,000 cells in 150 μL PBS) via the tail vein 4 h after LPS; ALI group (ALI) mice received the same amount of phosphate-buffered saline (PBS) via the tail vein 4 h after LPS; and control group (Con) mice were given the same amount of PBS at the corresponding time (Additional file
1: Fig. S1B). To confirm the effect of DCregs on CD4
+ T cells in ALI lungs, we conducted another in vivo experiment. The control group (Con) and ALI group were treated as described above, and mice in the DCreg group (DCreg) received DCregs (500,000 cells in 150 μL PBS) via the tail vein 4 h after LPS (Additional file
1: Fig. S1C).
To confirm the effect of MSCs on DC migration in ALI mice, the following experiments were performed. ImDCs were labeled with 0.5 mM carboxyfluorescein diacetate succinimidyl ester (CFSE). A total of 1,000,000 labeled cells were injected via the tail vein into wild-type recipient mice before LPS or PBS. The mice were randomly assigned to one of the following three groups (n = 6 mice per group): MSC group (MSC) mice received MSCs (500,000 cells in 150 μl PBS) via the tail vein 4 h after LPS; ALI group (ALI) mice received the same amount of phosphate-buffered saline (PBS) via the tail vein 4 h after LPS; Control group (Con), mice were given the same amount of PBS at the corresponding time (Additional file
1: Fig. S1D).
To investigate the effect of MSC-mediated inhibition of DCreg production on the pathological damage in ALI mice, the mice were randomly assigned to one of the following two groups (n = 6 mice per group): MSC + DAPT group (MSC + DAPT) mice were injected intratracheally with DAPT (0.3 mg/kg) 0.5 h before LPS and received MSCs (500,000 cells in 150 μL PBS) via the tail vein 4 h after LPS [
24]; MSC + DMSO group (MSC + DMSO) mice were injected intratracheally with the same amount of DMSO 0.5 h before LPS and received MSCs (500,000 cells in 150 μL PBS) via the tail vein 4 h after LPS; DMSO group (DMSO) mice were injected with DMSO equivalent to DAPT at 0.5 h before modeling, and were injected with the same amount of PBS with LPS or MSCs at the corresponding time; DAPT group (DAPT) mice were injected with DAPT 0.5 h before modeling, and injected with the same amount of PBS as LPS or MSCs at the corresponding time (Additional file
1: Fig. S1E). Lymphocytes in monocyte suspensions of lung tissue and peripheral blood were isolated by lymphocyte separation medium density gradient centrifugation. Lung cell isolation and measurement of the accumulation and maturation of cDCs by flow cytometry were performed as previously described [
13].
Evaluation of lung edema
Lung wet weight to body weight (LWW/BW) ratios, which reflect the severity of lung vascular permeability and lung edema, were determined for the control, ALI and MSC groups.
Lung histopathology
The right upper lobe was embedded in paraffin and sagittally sliced into 5 μm thick sections. The sections were stained with hematoxylin and eosin. Edema, alveolar and interstitial inflammation and hemorrhage, atelectasis, necrosis, and hyaline membrane formation were each scored using a 0- to 4-point scale. The severity of lung injury was calculated as the sum of the scores as previously described [
25].
Flow cytometry
For phenotypic analysis of cell surface marker expression, cells were harvested, resuspended in PBS, incubated for 15 min with FcR blocking reagent, and then incubated for 15 min with PE-, APC-, PE-Cy7-, PerCP-, or FITC-conjugated monoclonal antibodies on ice. DCs were stained with antibodies against CD11c, CD40, CD86, CD11b, MHC-II, CD4, CD44, CD69 and PD-L1 (BD Pharmingen). Mouse IgG1 isotype control antibodies were used in parallel as negative controls. CFSE (BD Pharmingen) was used in DCs or lymphocytes. The stained cells were washed twice, resuspended in cold buffer and then analyzed by flow cytometry (FACSCalibur; NovoCyte), and the results were processed using NovoExpress software. The results are expressed as the percentage of positively stained cells relative to the total cell number.
Mitogen proliferation assay
The protocol was based on previous mitogen proliferation and mixed lymphocyte culture with minor modifications [
16]. CFSE-labeled splenocytes (5 × 10
5 cells/well) were cocultured with allogenic DCs (mDCs or MSC-DCs, 5 × 10
4 cells/well) in a total volume of 0.2 mL medium in 96-well U-bottom plates.
Western blot analysis
For the Notch assay, Jagged1, Jagged2, Notch1, Notch2 and Notch3 were measured by Western blot analysis. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were incubated with primary antibodies against Jagged1 (1:1000; Abcam), Notch3 (1:1000; Abcam), Jagged2 (1:1000; Cell Signaling), Notch1 (1:1000; Cell Signaling) and Notch2 (1:1000; Cell Signaling) at a 1:1000 dilution at 4 °C overnight. The membranes were incubated with secondary antibody for 1 h at room temperature. Immunoblots were visualized using enhanced chemiluminescence (ECL; Thermo Scientific). The expression levels from whole cell extracts were normalized to β-actin or tubulin (1:1000; Cell Signaling).
Statistical analysis
All statistical analyses were performed using SPSS 23.0 software and GraphPad Prism 7.0. One-way analysis of variance (ANOVA) or two-tailed Student’s t test was used to determine the significance between the groups. The data are expressed as the mean ± standard deviation (SD). A value of p < 0.05 was considered significant.
Discussion
In this study, we found significant accumulation and maturation of lung DCs 2 h after intratracheal injection of LPS, which were positively correlated with the lung pathological injury score. MSCs induce mDCs to differentiate into DCregs, which have low expression of MHCII, CD86 and CD40, by direct cell–cell contact. In addition, we also showed that MSCs activate the Notch pathway by Jagged1 binding and increasing the expression of Notch2 to induce the generation of DCregs. On the other hand, by inhibiting the Notch pathway in vivo and in vitro, the induction of DCregs by MSCs was weakened, and the pulmonary protective effect of MSCs on ALI was reduced. In conclusion, MSCs induce DCreg production by activating the Notch pathway to alleviate acute lung injury.
Previous studies showed that mature DCs increased significantly in the lung tissues of ALI mice, and Fms-like tyrosine kinase 3 pretreatment stimulates lung DC proliferation and maturation, exacerbating lung pathological damage. In contrast, Losartan pretreatment inhibited lung DC proliferation and maturation and alleviated lung damage [
12,
13]. These results indicated that mature DCs play a key role in the pathogenesis of ALI. We found that cDCs in the lung tissues of ALI mice matured and increased significantly as early as 2 h after LPS injection and continued to increase up to 12 h. Although we found a positive correlation between the percentage and maturity of cDCs and the lung pathological injury score, the contribution of intervention with mature DC function to lung protection in ALI should be interpreted cautiously. On the one hand, the evidence is that LPS can stimulate the mature activation of DC, and mature DCs express a large number of MHC II and CD80, CD86 and other co-stimulating molecules, inducing the activation of naive T cells and differentiating into effector T cells [
26], which aggravate the inflammatory response of ALI [
27]. Therefore, regulating the functions of mature DCs may be crucial in LPS-induced ALI treatment. On the other hand, respiratory DCs sample microorganisms and particulates from the respiratory tract and present them to T cells to initiate an adaptive immune response to clear bacteria in the lungs [
28]. In fact, the regulation of DC therapy for bacteria-induced ARDS requires finding a balance between the elimination of pathogens and the suppression of organ damage caused by excessive inflammatory responses. However, in this study, we observed the effect and mechanism of MSC on lung DCs in LPS-induced-ALI mice without considering the factor of bacteria.
MSCs show strong immunoregulatory functions by suppressing T cell proliferation and activation, inducing regulatory T cells, inhibiting B cell function, or promoting the phagocytic activity of macrophages [
8‐
10]. However, whether DCs are regulated by MSCs in the ALI microenvironment is not known. Our findings revealed that MSCs significantly downregulated the percentage and maturity of cDCs in ALI mice and alleviated lung injury. According to the findings of this study, the percentage of mDCs increased significantly in ALI mice 2 h after LPS administration but decreased after treatment with MSCs, suggesting that MSCs induce cDCs from mature phenotypes to differentiate into immature phenotypes, alleviating lung injury and reducing pulmonary edema. To support this conclusion, we performed in vitro experiments that showed that mDC expression of PD-L1, the antigen presenting molecule MHCII and costimulatory molecules CD86 and CD40 decreased significantly after 3 days of coculture with MSCs, and still maintained stable phenotypes in the LPS microenvironment, which was consistent with the previous 7 days of coculture [
16]. The expression levels of MHCII, CD86 and CD40 determine whether DCs can capture and process antigens and deliver them to T cells, which affects the proliferation and activation of T cells [
29,
30], and PD-1 is necessary for DC-mediated induction of regulatory T cells and tolerance [
31]. We confirmed that the proliferation and activation of T cells that were stimulated by DCs after coculture with MSCs were decreased, and we also found that treatment with MSC reduced the number and percentage activation of CD4
+T cells in the lung of ALI mice.
Studies have reported that after MSC culture with DCs, the expression of CCR7 was decreased after DC stimulation, and migration to CCL19 or CCL21 was also significantly reduced [
15,
32]. Analysis of respiratory dendritic cell subsets revealed significantly reduced lung DC infiltration in MSC-treated mice with acute lung injury induced by
Klebsiella pneumoniae [
33], and Chiesa also reported that MSCs inhibit DC migration to lymph nodes [
34]. Consistent with these results, we found that lung DCs were significantly reduced in ALI mice that were treated with MSCs, which may be due to MSC-mediated inhibition of DC migration. The results of in vivo experiments showed that CFSE-labeled DCs had increased retention times in ALI mouse blood, indicating that MSCs reduced the retention of CFSE-labeled DCs in ALI mouse blood, resulting in reduced migration of DCs to the lungs.
The Notch signaling pathway controls cell proliferation, apoptosis, survival and differentiation during cell development and homeostasis [
21,
35‐
38]. MSCs induced a semimature DC phenotype that required jagged1 to activate Notch signaling for the expansion of regulatory T cells, reducing the pathology in a mouse model of allergic airway inflammation [
19]. Consistent with these results, our study shows that under LPS stimulation, MSCs expressed more jagged1, and both MSCs and recombinant jagged1 induced the generation of DCregs. Jagged1/Notch2 signal activation is closely related to cell regeneration and immune cell regulation [
39,
40]. Previous studies have shown that promoting the expression of NOTCH2 reduces the efficiency of DC presentation of MHC class II-restricted antigens and limits the strength of CD4
+ T cell activation [
41]. This study similarly found that the expression of Notch2 receptor protein was significantly increased in MSC-treated DCs or recombinant jagged1-treated DCs. Therefore, these results suggest that the Notch pathway is involved in the mechanism by which MSCs induce mDC immune tolerance. In this study, the expression of costimulatory molecules in DCs and functional markers of T cells that were stimulated by DCs showed that MSCs induced DCreg production.
γ-Secretase inhibitors are a class of small molecular compounds that target the Notch pathway and have been used in preclinical and clinical trials to treat a variety of diseases [
42,
43]. Gamma-secretase inhibitors have been shown to attenuate neurogenic acute lung injury in rats [
44]. This study found that MSCs reduced the expression of DC costimulatory molecules and functional markers, which were blocked by the γ-secretase inhibitor DAPT. Therefore, we performed in vivo experiments and confirmed that induction of tolerant DCs is a key link in MSC-mediated alleviation of lung pathological damage by inhibiting the Notch pathway. These data suggest that MSCs induce tolerant DCs by activating the Notch pathway to alleviate acute lung injury.
The LPS-induced ALI model has some limitations compared with the bacteria-induced animal model. LPS is a potent agent that activates the innate immune response via the TLR4 pathway, and its use provides information about the effects of host inflammatory response, which occurs in bacterial infections [
45]. Many previous studies selected LPS-induced ALI model to study the function of immune cells in ARDS [
46,
47]. In addition to endotoxins, gram-negative bacteria (
E. coli) also continue to produce a variety of virulence factors such as adhesin, exotoxin and type III secretion system. The adhesin produced by bacteria mediates the adhesion between bacteria and epithelial cells [
48]. Bacteria produce potent cytotoxic exotoxins [
49] as well as utilize the type III secreting system to directly cause epithelial cell lysis [
50]. This is consistent with Wiener-Kronish’s findings that LPS treatment does not cause the severe endothelial injury that occurs in ARDS [
51]. LPS binds to specific LPS binding proteins to form a complex, which activates the CD14/TLR4 receptor structure mostly on the surface of immune cells, such as monocytes, macrophages, and dendritic cells, triggering the production of inflammatory mediators [
52]. This study focused on the effect of MSC treatment on DC function in ALI, so LPS-induced animal model was selected. In addition, LPS is easy to administer and has little direct toxicity to cultured cells in vitro [
45].
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.