The Mobilization of Splenic Reservoir Myeloid-Derived Suppressor Cells in Sepsis-Induced Myocardial Injury

Background Myeloid-derived suppressor cells (MDSCs) play key roles in sepsis, but whether bone marrow is considered the only source remains unclear. The current knowledge about the mechanism of MDSCs leading to myocardial injury in sepsis is poor. In sepsis patients with cardiac dysfunction, the circulating percentage of CD14-CD11b+ and serum concentrations of IL-6 and IL-1β were measured. A mouse sepsis model was established through caecum ligation and puncture (CLP). Animals were divided into four groups: control, sham, CLP and CLP+splenectomy (CLPS). Plasma concentrations of IL-6, IL-1β, TnI and NT-proBNP were measured. CD11b+Gr-1+ cells were detected by immunofluorescence staining and RT-PCR. Myocardial injury was detected by HE, Masson and TUNEL staining. The expression of mTOR, P53 and caspase-3 was measured by Western blot. Human blood samples were collected in accordance with the medical ethics committees of Yi Ji Shan Hospital, Wuhu No. 1 Peoples’ Hospital, Wuhu No. 2 Peoples’ Hospital and Wuhu Traditional Chinese Medicine Hospital and performed according to the Declaration of Helsinki. This study was approved by Yi Ji Shan Hospital Ethics Committee and have informed consent from all patients and healthy control. Samples from 6 age- and sex-matched healthy volunteers were collected as controls. Peripheral venous blood samples were collected into EDTA-coated tubes and a procoagulant tube. Plasma was then separated from the procoagulant tube, collected into EP tubes and stored at -40°C. Plasma concentrations of IL-6 and IL-1β were measured using a Human IL-6 ELISA kit and a Human IL-1β ELISA Kit (Boster, China), respectively. Mononuclear cells were isolated on hydroxypropyl methylcellulose (HaoYang, China) by centrifugation at 500 g for 20 min. Cells were fixed in 1% BSA in PBS. The cells were then incubated with PE-conjugated anti-Human CD14 (BD Biosciences, USA) and FITC-conjugated anti-Human CD11b (BD Biosciences, USA) for 1 h on ice. After washing three times with 1% BSA in PBS, cells were analysed on a FACScan flow cytometer (Becton Dickinson, USA) to detect CD14-/CD11b+ cells. Biosciences) (except for CLPS). Total nucleated cells in the peripheral blood were isolated after erythrocyte lysis. For the FACS analysis, cell suspensions were stained with PE-conjugated anti-mouse CD11b antibodies and FITC-conjugated anti-mouse Gr-1 for 1 h on ice, and then they were washed with 1% BSA in PBS. The percentage of CD11b+/Gr-1+ cells was evaluated by multi-colour flow cytometry using a flow cytometer (EPICS® ALTRA, Beckman Coulter).

mRNA expression of CD11b, Gr-1, IL-6 and IL-1β in myocardium of mice In mouse hearts, mRNA levels of various factors, including CD11b, Gr-1, IL-6 and IL-1β, were detected by RT-PCR. The mouse hearts were stored in Trizol at -80°C. Total RNA was extracted as previously described and cDNA was synthesized using a PrimeScriptTM RT reagent kit with gDNA Eraser (TAKARA, Japan). RT-PCR was performed using CD11b primers (Forward: and IL-1β mRNA expression was normalized to GAPDH and calculated as 2-△△Ct.

Cardiac histopathology
The myocardial tissue was fixed in 10% formaldehyde, and then embedded in paraffin. Tissue sections (2 μm thick) were stained with haematoxylin and eosin (HE) and Masson staining using an HE Staining Kit (Beyotime, China) and Masson Staining Kit (NanJing KeyGen Biotech, China), respectively, according to the manufacturer's protocol. Morphological changes of myocardial tissue were assessed using an optical microscope (Olympus, Tokyo, Japan). The morphological evaluations were performed in a blinded manner by 2 independent investigators.
Immunofluorescence staining A frozen section of heart was prepared. The myocardial tissue section (2 μm thick) was stained with PE-conjugated anti-mouse CD11b antibodies and FITC-conjugated anti-mouse Gr-1 for 1 h at room temperature to measure the expression of CD11b and Gr-1. The tissue was further stained with 4,6diamidino-2-phenylindole (DAPI, Roche). Fluorescence microscopy (Olympus, Tokyo, Japan) was used to detect the fluorescence.

TUNEL immunohistochemistry (IHC) staining
Terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling (TUNEL) was performed to detect apoptotic nuclei by TUNEL Staining Kit (Roche, Indianapolis, IN) according to the manufacturer's protocol. The number of apoptotic cells with TUNEL-positive nuclei was counted by 2 independent observers blinded to the treatment group and expressed as a percentage of the total myocyte population.

Western blot analysis
For the Western blot analysis, protein was extracted from mouse myocardial tissue using a Tissue Protein Extraction Kit (Beyotime, China) according to the protocol provided by manufacturer. Protein concentrations were measured using the BCA Assay Kit (Pierce, USA). Western blots of the protein samples were performed to detect the mammalian target of rapamycin (mTOR) and phosphorylated Ser2448-mTOR (P-Ser2448 mTOR) using rabbit monoclonal anti-mouse mTOR antibodies and the rabbit monoclonal anti-mouse phosphorylated Ser2448-mTOR antibody (Cell Signaling Technology, USA). P53 and caspase-3 expression were detected by rabbit monoclonal anti-mouse P53 (Cell Signaling Technology, USA) and rabbit anti-mouse caspase-3 (Abcam, UK). GAPDH was used as a loading control (KangCheng Bio-tech, China). After primary antibody incubation, the blots were incubated with the appropriate secondary horseradish peroxidase conjugate mouse monoclonal antirabbit antibody (Boster, China). Each membrane was washed and then developed using Super Signal chemiluminescent substrate (Pierce, USA).

Statistical analyses
For all experiments, the data were analyzed using either a Student's t-test or Bonferroni's test, and values are expressed as the means ± SEM. All statistical analyses were performed using SPSS software (SAS Institute Inc., USA); p values < 0.05 were considered to indicate significance.

Results
Increased CD14-CD11b+ cells and plasma concentration of inflammatory factors in peripheral circulation of sepsis patients with myocardial injury No differences in age or sex were found between the sepsis with myocardial injury group (n=10) and the control group (n=6). The percentage of CD14-CD11b+ MDSCs was higher in the sepsis group than in the control group (88.1±5.3% vs 53.4±9.1%, P<0.05, figure 1). Furthermore, the plasma concentrations of IL-6 and IL-1β in sepsis patients were significantly higher than in controls (24.0±1.4 pg/mL vs 16.7±1.0 pg/mL and 11.4±1.4 pg/mL vs 7.4±0.5 pg/mL, respectively, P<0.05, figure 2A, B).
The percentage of CD11b+Gr-1+ cells in blood, spleen and bone marrow of mouse sepsis models As shown in figure 3, compared with the control and sham groups, the percentage of CD11b+/Gr-1+ cells in the blood was significantly higher in the CLP group (mean: 33.6±2.4% vs 6.9±1.0% and 6.8±0.8%, respectively, P<0.001). Furthermore, in the CLPS group, the percentage of CD11b+/Gr-1+ cells in the blood was lower than that in the CLP group (mean: 19.3±1.3% vs 33.6±2.4%, P<0.001).

Cardiac function of mouse sepsis models
In the sepsis group, the EF% and FS% were significantly lower than in the control and sham groups  figure 4).

The plasma concentration of troponin I (TnI), NT-proBNP and the expression of inflammatory factors in mouse sepsis models
The plasma concentration of TnI was significantly higher in the CLP group than that in the control, sham and CLPS groups (6.23±0.24 ng/mL vs 0.03±0.01 ng/mL, 0.04±0.01 ng/mL and 3.68±0.17 ng/mL, respectively, P<0.05). The plasma NT-proBNP concentration was significantly higher in the CLP group than that in the control, sham and CLPS groups (2799±536 pg/mL vs 93±8.0 pg/mL, 100±4.0 pg/mL and 1011±148 pg/mL, respectively, P<0.05, figure 5A, B).
The plasma concentration of IL-6 was also higher in the CLP group than that in the control, sham and CLPS groups (19.5±0.3 pg/mL vs 9.8±0.7 pg/mL, 4.2±1.0 pg/mL and 4.1±0.3 pg/mL, respectively, P<0.05). Furthermore, the plasma concentration of IL-1β in the CLP group was higher than that in the RT-PCR showed that the expression of IL-6 and IL-1β was significantly higher in the CLP group than that in the control and sham groups. In the CLPS group, the expression of IL-6 and IL-1β was downregulated more than in the CLP group ( Figure 6C, D).
Expression of CD11b and Gr-1 in myocardium tissue and cell apoptosis analysis HE staining showed the number of inflammatory cells that gathered in the myocardium tissue in the CLP group compared with the control and sham groups. In the CLPS group, the number of inflammatory cells was decreased more than in the CLP group (Figure 7). Masson staining showed that collagen fibres were detected instead of muscle fibres in sepsis (Figure 7). TUNEL staining showed that apoptotic cardiomyocytes had no obvious change between the control and sham groups. In the CLP group, apoptotic cardiomyocytes increased markedly in the myocardium tissue. However, apoptotic cardiomyocytes decreased in the CLPS group compared with the CLP group (Figure 7).
Immunofluorescence indicated that CD11b+Gr-1+ cells in the myocardium tissue tended to be higher in the CLP group than in the control or sham groups. In the CLPS group, the number of CD11b+Gr-1+ cells was lower than in the CLP group ( Figure 8). RT-PCR further showed that the expression of CD11b and Gr-1 was significantly higher in the CLP group than in the control and sham groups. In the CLPS group, the expression of CD11b and Gr-1 was lower than in the CLP group ( Figure 6A, B).

mTOR/P53 signalling pathway participated in sepsis-induced myocardial injury
In the CLP group, mTOR was dephosphorylated and the expression of P53 and caspase-3 was upregulated compared to the control and sham groups. In the CLPS group, the phosphorylation of mTOR was higher than in the CLP group. The expression of P53 and caspase-3 was downregulated in the CLPS group compared to the CLP group ( Figure 9).

Discussion
In this study, we revealed that spleen-derived CD11b+Gr-1+ cells mobilize into circulation and myocardial tissue and further inhibit the phosphorylation of mTOR that leads to apoptosis of cardiac myocytes through the upregulated expression of P53 and caspase-3 in sepsis-induced myocardial injury.
Severe sepsis contributes to the main causes of mortality in ICU patients and increases the medical burden [14,15]. The inflammatory response is a key characteristic in the development of organ injury in sepsis. Myocardial dysfunction is an important characteristic in sepsis that is associated with poor prognoses such as cognitive impairment and functional disability [16]. Pro-inflammatory factors released by inflammatory cells are a major mechanism participating in the progress of sepsis-induced myocardial dysfunction [17,18]. CD11b+Gr-1+ cells are a group of mononuclear cells usually defined as myeloid-derived suppressor cells (MDSCs). In humans, MDSCs are defined as CD14-CD11b+ [8].
CD11b+Gr-1+ cells have numerous functions, including producing pro-inflammatory cytokines and immunoregulatory properties that migrate to injured or infected sites [19,20]. CD11b+Gr-1+ cells participate in the systemic inflammatory response through different pathways during early and late sepsis [21]. However, under different circumstances, CD11b+Gr-1+ cells have a paradoxical role in sepsis, trauma and ischaemic injury [5,22]. It has remained in doubt whether CD11b+Gr-1+ cells are protective or injury-promoting cells. A widely accepted viewpoint considered that in different disease phases, different groups of CD11b+Gr-1+ cells showed contrary effects [6]. In this study, we indicated that CD11b+Gr-1+ cells migrated to the circulation of sepsis patients and the myocardial tissue in animal sepsis models. The concentration of blood pro-inflammatory cytokines increased dramatically in both patients and animal models, which further influenced the signalling pathway associated with cell survival.
Bone marrow is rich in mononuclear cells containing CD11b+Gr-1+ cells. Various research has indicated that bone marrow is the major source of CD11b+Gr-1+ cells [23]. However, the spleen contains numerous monocytes or promonocytes. In myocardial infarction, spleen-derived CD11b+Gr-1+ cells contributed maximally to circulation and myocardial tissue [10,24]. In other pathological injuries, splenic reservoir CD11b+Gr-1+ cells were also the major source of circulating inflammatory cells [25]. Our research also indicated that in sepsis-induced myocardial injury, spleen-derived CD11b+Gr-1+ cells were the major source that rapidly mobilized into circulation and the target organ to trigger an inflammatory reaction and damage of the organ through the CLPS model. The spleen and its primary cells have the potential to become a probable new therapeutic target. mTOR is a key molecule that regulates CD11b+Gr-1+ differentiation and immunomodulation during an inflammatory reaction [26,27]. The relationship between mTOR and organ damage is a key point that attracts a number of researchers. Previous research suggested that the activation of mTOR protected against murine immunological hepatic injury through limiting the recruitment of CD11b+Gr-1+Ly6Chigh cells [28]. However, in different organs and microenvironments, the role of mTOR showed an obvious heterogeneity. The activation of mTOR associated with vasodilator and pro-inflammatory mediator formation contributes to LPS-induced hypotension and inflammation [29]. In acute kidney injury, CD11b+Gr-1+ cells are recruited to the injured kidney following mTOR inhibition, and further protect mouse kidneys against AKI in vivo [27]. The key role of mTOR may act to adjust the balance of pro-inflammatory and anti-inflammatory responses [26]. Our data indicated that in sepsis-induced myocardial injury, the expression of phosphorylated mTOR in myocardial tissue was lower than that in the control group. Nevertheless, phosphorylated mTOR was highly expressed in CLPS mice, similar to healthy controls, due to few CD11b+Gr-1+ cells that mobilized from the spleen and migrated to the myocardial tissue. These results suggest that the inflammatory reaction inhibits the activation of mTOR and further leads to myocardial tissue damage in sepsis.
The mTOR signalling pathway has been widely studied in the senescence of tumour cells and immortalized cell lines [30]. It had been proven that the mTOR signalling pathway plays a key role in cell apoptosis and senescence. Inhibition of mTOR activates the expression of P53 and further leads to cell senescence and organ dysfunction [31]. P53 is considered a key molecular that regulates the cell cycle and loss of cell function. A previous study suggested that P53 mediates the accelerated onset of senescence of endothelial progenitor cells in diabetes [32]. Nevertheless, activation of the p53 tumour suppressor can lead to cell cycle arrest [33]. However, the relationship between the mTOR pathway and cell and organ injury is still under debate. In LPS-induced rat damage models, activation of mTOR is the key mechanism involved in inflammation [29]. In sepsis, activation of the mTOR signalling pathway contributes to long-term neuronal loss [34]. Our research demonstrated that in sepsis-induced myocardial injury, expression of phosphorylated mTOR was significantly inhibited and further led to the upregulation of P53, and finally, to cell apoptosis, which resulted in impaired heart function.
These data may partly explain the role of mTOR/P53 in sepsis-induced myocardial injury.
There are several apparent limitations in this study. Whether the mTOR signalling pathway is involved in the differentiation of CD11b+Gr-1+ cells and the roles of different types of MDSCs need to be further studied.

Conclusion
In a mouse sepsis-induced myocardial injury model, splenic reservoir CD11b+Gr-1+ cells rapidly        Monocyte mobilized to myocardial tissue in sepsis and led to myocardial damage. A. The first column: representative HE staining showed that monocyte accumulated to myocardial tissue in CLP group. No obviously inflammatory cells occurred in control, sham and CLPS.
Myocardial dissolving and inflammatory cell infiltration was observed in sepsis group (400*).
The second column: representative Masson staining photograph. Collagen fiber was detected in sepsis instead of muscle fibers (100*). The third column: representative TUNEL staining showed that apoptotic cardiomyocytes were no obviously changed between control and sham (400*). Markedly increased apoptotic cardiomyocytes in the CLP group. In CLPS, the number of apoptotic cardiomyocytes is lower than CLP. B: Histogram showed the number of apoptotic cardiomyocytes among the four groups. (n=5 for control and sham, n=10 for CLP and CLPS, *P<0.05 vs control, sham and CLPS). CD11b+Gr-1+ cells mobilized to myocardial tissue in sepsis. Representative immunofluorescence staining showed that in CLP group, both CD11b and Gr-1 expressed in myocardial tissue. In control and sham, CD11b and Gr-1 negatively expressed in myocardial tissue. In CLPS, the expression of CD11b and Gr-1 was lower than CLP (n=5 for control and sham, n=10 for CLP and CLPS). The mTOR/P53 pathway in septic myocardial tissue. A: Representative Western blots normalized to GAPDH showed that in CLP group, the phosphorylation of mTOR was inhibited compared to control and sham. However, the phosphorylation of mTOR was higher in CLPS than CLP. Further, the expression of P53 and caspase-3 was down-regulated in CLPS compared to CLP. B: Densitometry quantitation of protein expression levels are shown as fold changes in histogram (n=5 for control and sham, n=10 for CLP and CLPS, * P<0.05 vs control, sham and CLPS).