Umbilical cord mesenchymal stem cells limit poststroke infection

Brain ischemia leads to excessive inltration of clusters of CD8 + T and natural killer (NK) cells in the brain, which aggravate ischemic brain injury. Acute ischemic stroke also has a negative impact on the antibacterial immune response, leading to stroke-induced immunodepression and infection. Umbilical cord mesenchymal stem cell (ucMSC) have an immunosuppressive function. Therefore, we aimed to determine whether ucMSC treatment alleviates the excessive inltration of CD8 + T and NK cells. We also investigated signicant concerns that ucMSC treatment might suppress antimicrobial immunity, leading to an increased risk of infection. of and blood biochemical indexes also detected.We assessed autophagy and apoptosis of platelets, as well as mitochondrial membrane potential (MMP) and ATP levels.In vitro ucMSC was co-cultured with platelet and Escherichia followed by detection of the coli growth for the benecial effect of ucMSC transplantation in patients with post-stroke


Introduction
Infectious complications-primarily pneumonia and urinary tract infection-are a leading cause of death in ischemic stroke patients [1,2].The impairment of immune responses after brain ischemia increases susceptibility to infections [3,4]. Excessive in ltration of cluster of differentiation (CD)8 + T cells or natural killer (NK) cells in the brain aggravates ischemic brain injury, while brain ischemia compromises NK cellmediated immune defence in the periphery and can result in post-stroke infection [5][6][7].In addition,the grade of immunoin ammatory activation could be related to pathogenesis of neuronal damage in ischemic stroke [8,9].For example,TNF-and IL-6 express a higher level in plasma of acute ischemic stroke patients,which play a pivotal role in in ammatory processes that aggravate ischemic neural damage [10].While mesenchymal stem cells (MSCs) have been shown to exert therapeutic effects following stroke [11], they inhibit the proliferation and effector functions of various immune cells, including T and B lymphocytes and NK cells [12][13][14][15][16][17]. It is currently unknown, however, whether ucMSCs also suppress antimicrobial immunity and thereby increase the risk of infection.
In our previous study, we detected no obvious signs of infection in mice with stroke with or without umbilical cord (uc)MSC treatment. In another study, however, we established a post-stroke infection model using the gram-positive intracellular bacteria Listeria monocytogenes [6] or Escherichia coli, and this decreased the number of platelets in mice, while ucMSC treatment reversed this change. Symptoms of infection were also alleviated when these mice were injected with ucMSC, suggesting their therapeutic potential against post-stroke infection. This potential was investigated further in the present study by examining the effects of ucMSC on platelets and post-stroke infection, as well as determining the underlying mechanisms for these effects.

Animals
C57BL6 mice were purchased from the Laboratory Animal Center of Southern Medical University and were maintained under standard laboratory conditions, with the temperature controlled at 24°C and with free access to a standard diet and sterile water. All animal procedures were performed in accordance with the guidelines and approval of the Animal Ethics Committee of Southern Medical University.

Middle cerebral artery occlusion (MCAO) and post-stroke infection model
Mice weighing 20-22 g (aged 7-8 weeks) were allowed free access to water but were fasted for 12 h to standardize glycaemic state. MCAO was performed under anaesthesia induced by intraperitoneal injection of pentobarbital (100 mg/kg). Body temperature was maintained at 37°C ± 0.5°C using a heating pad (RWD Life Science, Shenzhen, China). To induce MCAO, a 6 − 0 nylon suture (Covidien, Mans eld, MA, USA) with a round tip and silicon coating was inserted from the left external carotid artery into the middle cerebral artery. The success of the surgery was veri ed by monitoring surface cerebral blood ow using a laser Doppler owmeter (Moor Instruments, Devon, UK). After 1 h, the occluding lament was gently withdrawn back into the common carotid artery to allow reperfusion. Mice in the sham group underwent a sham operation without suture insertion.
E. coli were cultured as previously described [18] and stored in 30% glycerol at − 80°C until use. E. coli were grown in Luria-Bertani (LB) medium (10 g/l tryptone, 5 g/l yeast extract, and 171.1 mM NaCl). Growth was determined by measuring the optical density at 620 nm (OD 620 ) or by plating the cells on LB plates and counting viable cells. For infection, age-matched male mice were intravenously injected with 10 7 colony forming units (CFU) of E. coli resuspended in 500 µl phosphate-buffered saline (PBS) immediately after sham or MCAO operation.
To determine the degree of infection, the mouse liver, lung, and brain were removed and homogenized in distilled water with 0.01% Triton X-100. The number of viable E. coli cells was counted after plating serial dilutions of organ homogenates and blood on LB plates and culturing overnight at 37°C.

Assessment of neurological function and measurement of cerebral infarct area
Neurological function was determined based on the Modi ed Neurological Severity Score (mNSS). The test was carried out by a blinded investigator before and 3 days after MCAO, as previously described [19]. The infarct areas of different experimental groups were measured in photomicrographs of methylthioninium chloride-stained tissue sections (5 sections/animal). Experiments were repeated ve times.
Hematoxylin and eosin (HE) staining, immunohistochemistry, immuno uorescence analysis and ow cytometry analysis At 24 h after MCAO or post-stroke infection, mice were anesthetized and transcardially perfused with 20 ml cold PBS and 20 ml of 4% paraformaldehyde in 0.1 M PBS. The brain, lung, liver, and spleen were removed, post-xed, and embedded in para n. The tissue blocks were cut into 5-mm sections that were depara nized and stained with HE according to standard protocols.
CD8 + T cells and NK cells in the brain and spleen were identi ed by immuno uorescence analysis and immunohistochemistry, as previously described. For the latter, brain and spleen tissue sections were incubated overnight at 4°C with primary antibodies against CD8 (ab25117) and natural cytotoxicity receptor (NCR) (ab199128),Iba1(ab5076),CD68(ab125212)(Abcam,Cambridge,MA,USA) respectively followed by processing with avidin-biotin-peroxidase (BosterBio, Wuhan, China). The sections were stained with diaminobenzidine, and nuclei were counterstained with hematoxylin.

Blood biochemical analysis
Mouse blood was collected via the angular vein under anaesthesia into an anticoagulant-containing tube.
Biochemical analyses were performed at Southern Medical University Huayin Laboratory.

Enzyme-linked immunosorbent assay (ELISA)
Plasma was isolated by centrifugation of blood samples at 1500 rpm for 20 min. TNF-α, IL-6, IL-10 in the plasma were detected with ELISA kits (Cusabio, Wuhan, China) according to the manufacturer's instructions. Brie y, 100 µl of plasma was added to each well of a 96-well plate. After incubation for 2 h at 37°C, the plasma was removed, and the plates were sequentially incubated with biotin-conjugated primary antibody followed by horseradish peroxidase-conjugated secondary antibody for 1 h at 37°C, with three washes between each step. After adding the chromogenic substrate, the plates were incubated in the dark for 30 min at 37°C. The reaction was terminated, and the OD 450 was measured using an iMark microplate reader (Bio-Rad, Hercules, CA, USA).

Co-culture of bacteria and ucMSC
Platelets alone or in combination with ucMSC were inoculated with E. coli for 1, 2, 4, or 6 h. Bacterial growth was determined by measuring the OD 620 .

Statistical analysis
Statistical analysis was performed using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Data are presented as the mean ± SD. The signi cance of differences between means was examined by Student's t-test or oneway analysis of variance. Results with P < 0.05 were considered signi cant.

Results
ucMSC decrease brain lesion size and improve neurological function after stroke We evaluated the effect of ucMSC on stroke based on measurement of the lesion area and the mNSS in an MCAO mouse stroke model. mNSS scores (P < 0.05; Fig. 1A) as well as the lesion area (P < 0.05; ucMSC inhibit immunological function after stroke We next examined the immunomodulatory effects of ucMSC treatment on the post-stroke brain by examining the abundance of CD8 + T cells and NK cells by immunohistochemistry and immuno uorescence analysis,as well as ow cytometric analysis. Both cell populations were diminished in mice treated with ucMSC as compared to that in the MCAO group ( Fig. 2A-C). We detected the activated microglia by staining Iba1and CD68 (activated microglia marker) after induction of MCAO. Iba1-and CD68-positive cells were increased after MCAO,while ucMSC treatment decreased the Iba1-positive and CD68-positive cells (Fig. 2D).Additionally, the proportions of CD8 + T cells and NK cells were decreased in the spleen and peripheral blood after MCAO, but these changes were not abrogated in the MCAO + ucMSC group (Fig. 2D-G). Meanwhile, plasma levels of the pro-in ammatory cytokines interleukin (IL)-6 and tumour necrosis factor (TNF)-α were lower, whereas that of the anti-in ammatory cytokine IL-10 was higher, in ucMSC-treated mice as compared to levels in untreated MCAO mice, as determined by ELISA (Fig. 3A-C).
ucMSC treatment mitigates infection after stroke and prevents organ damage HE staining of lung, liver, and spleen tissue sections as well as blood routine revealed no signs of infection in MCAO mice with or without ucMSC treatment (Fig. 4). To assess the effect of systemically administered ucMSC on post-stroke infection, MCAO mice with or without ucMSC treatment and E. coli infection were examined for the presence of bacteria in the brain, lung, liver, and spleen. Compared to the MCAO group, MCAO + ucMSC mice showed a lower bacterial burden in these organs, including a reduction in the size of the germinal centre of the spleen (Fig. 5E). During post-stroke infection, in ammatory cells in ltrated the lung, brain, and liver and caused cell and tissue damage; these effects were alleviated by ucMSC treatment. We also measured aspartate aminotransferase (AST), alanine aminotransferase (ALT), creatine kinase(CK), and lactate dehydrogenase (LDH) levels in plasma and found that AST, ALT, CK and LDH were downregulated in the ucMSC treatment group as compared to levels in the MCAO group during the course of infection. Plasma TNF-α and IL-6 levels were also reduced, whereas IL-10 was upregulated by ucMSC treatment.

ucMSC and platelets have synergistic antibacterial effect
To investigate whether ucMSC have the ability of platelets to kill bacteria, ucMSC was co-cultured with E. coli. Finally,we found the growth of E. coli was inhibited in the presence of ucMSC (Fig. 9).

Discussion
Infection in the lungs and other organs are relatively common during the subacute stage of stroke and are associated with unfavourable outcomes [20]. Preventative antibiotic therapy does not in uence functional outcomes in the overall population [21,22]. Ischemic stroke negatively impacts the antibacterial immune response, leading to stroke-induced immunosuppression and infection [7,23]. For example, brain ischemia can cause a reduction in NK cell numbers and response in the periphery via activation of the catecholaminergic system and hypothalamic-pituitary-adrenal axis, which can result in infectious complications [6]. In accordance with previous studies [24], we observed a decrease in the numbers of CD8 + T cells and NK cells in the spleen and peripheral blood after stroke, whereas more of these cells in ltrated the brain tissue, which could aggravate brain injury. MSCs(mesenchymal stem cells) have immunomodulatory activity and are therefore promising agents for cell-based therapies. MSCs regulate a variety of immune cells-for example, they inhibit the activation and proliferation of T cells and induce T cell apoptosis while suppressing the differentiation and maturation of dendritic cells [25][26][27][28][29]. MSCs have also been shown to block NK cell activity [30][31][32]. In our study, we found that ucMSC reduced the number of CD8 + T cells and NK cells in brain tissue but not in the spleen or peripheral blood of mice following stroke, suggesting that ucMSC can prevent brain injury. Furthermore, plasma levels of the proin ammatory cytokines IL-6 and TNF-α were reduced, whereas that of the anti-in ammatory cytokine IL-10 was increased by ucMSC treatment, con rming that ucMSC induce immunosuppression [33].
One point of concern is whether ucMSC can increase the risk of infection after stroke by suppressing antimicrobial immunity. However, symptoms of post-stroke infection were alleviated in mice following ucMSC treatment, which not only inhibited the growth of bacteria in certain organs but also prevented tissue damage caused by bacteria and in ammatory factors. Post-stroke pneumonia is a major cause of death after stroke [34]. In our study, ucMSC treatment reduced haemorrhage, oedema, and cellularity in injured lung lobes caused by E. coli. So our results show that ucMSC play a protective role against poststroke infection,but the underlying mechanisms were not completely clear. Some studies demonstrsted MSC have anti-infection effect of is mediated in part from secretion of the antimicrobial peptide [35,36],which may be one reason of ucMSC inhibiting the post-stroke infection .However,we also found ucMSC inhibit the apoptosis of platelets,as well as maintain the count of platelets after post-stroke infection.
In previous studies, platelets have been shown to inhibit bacterial growth by surrounding bacteria and secreting a high concentration of antimicrobial substances [37]. Platelets also activate some immune cell types to ght bacteria and work with Kupffer cells to eradicate blood-borne bacterial infection caused by Bacillus cereus and methicillin-resistant Staphylococcus aureus [38]. Moreover, they interact with neutrophils to form a neutrophil extracellular trap that sequesters bacteria [39]. However, platelets invariably show diminished function and numbers after severe infection. For example, patients with sepsis often exhibit thrombocytopenia, which is associated with poor prognosis [40][41][42]. The mitochondrial dysfunction in platelets observed in sepsis and bacterial infection can lead to apoptosis: Bcl-xL-an essential regulator of platelet survival-is upregulated in the platelets of sepsis patients [43,44]. Autophagy is important for platelet functions, including haemostasis and thrombosis [45]. In our study, ucMSC treatment reversed the decrease in the autophagy marker LC3-II caused by MCAO and E. coli infection.Mitochondria are the main target of the intrinsic apoptosis pathway, and mitochondrial membrane depolarization serves as a marker of apoptosis [46]. ATP provided by mitochondria plays an important role in normal cellular functioning, including the response to physiological stress [47]. Thus, a decrease in ATP levels re ects platelet damage. In the present study, ucMSC treatment increased the expression of the anti-apoptotic proteins Bcl-2 and Bcl-xL, while restoring MMP and ATP production in platelets. In vitro, ucMSC and platelets can synergistically inhibit the proliferation of Escherichia coli.So we conclude that ucMSC may play a protective role against post-stroke infection by restoring the count and the function of platelet.

Conclusions
These results suggest that ucMSC have the ability to modulate the function of CD8 + T cells, NK cells. Our study serves as the basis for future studies and offers new insights into the mechanisms responsible for the bene cial effect of ucMSC transplantation in patients with stroke and post-stroke infection.

Availability of data and materials
All the data and informations used and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. Author's contributions JH, ZF, YX and XJ designed the experiments; HT S, FL, and Q OUYANG performed the experiments; ZZ performed data collection and analysis; X Z, YC, YZ and YT wrote the manuscript together; XJ guided this study, revised the manuscript and provided nancial support. All authors performed the nal approval of the manuscript. Figure 1 The therapeutic effect of ucMSC in MCAO mice. MCAO mice were treated with or without ucMSC. (A) mNSS and (B) lesion area in each group. The data are plotted as the means ± SD. *P < 0.05, n=5 or 3.v Figure 2 ucMSC reduced CD8+ T/NK cells in MCAO mouse brain but not spleen or blood. (A-C) Immunohistochemistry and immuno uorescence of CD8+ T cells and NK cells in the brain. D Immuno uorescence of the activated state of microglia.(E-H) Immuno uorescence and ow cytometry of CD8+ T cells and NK cells in the spleen and blood. Scale bar: 100 μm. The data are plotted as the means ± SD. *P < 0.05, n=5.

Figure 3
Expression of pro-in ammatory cytokines IL-6 and TNF-α in mouse plasma. The data are plotted as the means ± SD. *P < 0.05, n=3. No infective signs were found in the MCAO or ucMSC groups. MCAO mice were treated with or without ucMSC, followed by analysis of (A) white blood cells, (B) lymphocytes, (C) neutrophils, and (D) platelets, as well as (E) HE staining of the brain, lung, liver, and spleen. Scale bar: 100 μm. The data are plotted as the means ± SD. NS, not signi cant. *P < 0.05, n=5.