The Study on the Regulation of Th Cells by MSCs Through the JAK-STAT Signaling Pathway to Protect Naturally Aged Sepsis Model Rats

Background: Sepsis is the leading cause of death among patients, especially elderly patients, in intensive care units worldwide. However, there is no effective treatment for sepsis in the aging population. Therefore, we designed such a study to conrm the protective effect of MSCs against sepsis in the elderly and to explore its mechanisms. Methods: In this study, we established a sepsis model using naturally aged SD rats and injected 5×10 6 umbilical cord-derived MSCs via the tail vein. Each group of rats was analyzed for survival, examined for biochemical parameters, stained for organ histology, and analyzed for the Th cell subpopulation ratio and inammatory cytokine levels by ow cytometry. Western blotting was performed to detect the activity of the JAK-STAT signaling pathway. We designed in vitro experiments to conrm the regulatory role of MSCs, and veried the possible mechanism using JAK/STAT inhibitors. Results: The results revealed that the 72 h survival rate of sepsis rats treated with MSCs was signicantly increased, organ damage and inammatory inltration were reduced, the levels of organ damage indicators were decreased, the ratios of Th1/Th2 and Th17/Treg in peripheral blood and spleen were signicantly decreased, the levels of pro-inammatory cytokines such as IL-6 were decreased, the levels of anti-inammatory cytokines such as IL-10 were increased, and the levels of STAT1 and STAT3 phosphorylation were reduced. These results were validated in in vitro experiments. Conclusions: Therefore, this study conrms that MSCs can control the inammatory response induced by sepsis by regulating Th cells and inammatory factors, and that this leads to reduction of tissue damage, protection of organ functions and ultimately improvement of survival in aged sepsis model rats. Inhibition of the JAK-STAT signaling pathway may be an important mechanism for their action. divided into a sham-operated group sepsis model and sepsis+MSCs treatment group Equal number of rats in each 3 subgroups and samples were collected after euthanasia at 6 24 h or 72 h post-operation and/or treatment. The CLP model was used as a model of sepsis, and UC-MSCs were injected into the tail vein 1 h after CLP operation to induce sepsis. Five million (5×10 6 ) UC-MSCs in 1 ml saline were administered to each animal in the treatment group, and the sham-operated and model groups were injected with equal amounts of sterile saline at the same time. The 72 h survival rate of the rats in each group was statistically analyzed. cell multi-organ JAK-STAT signaling pathway anti-inammatory cytokines IL-13), which indicates that individual STAT family members differentially the balance of Th cell subsets. In the helper T-cell subpopulation, Th1 cell differentiation is mainly regulated by STAT1 and STAT4(48), while STAT6 and STAT3 regulate the differentiation of the immunoregulatory Th2 and Th17 cells, respectively(49). One study showed that STAT1 knock-out mice were signicantly resistant to LPS-induced endotoxemia(50) and CLP-induced septic shock(51). This was attributed to the altered balance of the Th1/Th2 immune response, which resulted in signicantly reduced organ damage in the liver and kidney. Another study showed that JAK2 inhibitors improved survival after CLP operation in mice and rats by reducing the expression of pro-inammatory mediators such as TNF-α, IL-6 and high mobility group box-1 (HMGB-1)(52). To explore the molecular mechanism of Th cell regulation by MSCs, we examined the relevant STAT signaling pathways in the spleen of aged sepsis model rats. We found that the phosphorylation levels of STAT1 and STAT3 were signicantly increased in the spleen of sepsis model rats, the phosphorylation levels of both molecules were signicantly reduced after MSC treatment, and the expression levels of STAT1 and STAT3 were comparable to those of normal rats. In the in vitro experiments, we administered inhibitors of the JAK, STAT1 and STAT3 signaling pathways to the LPS-Page

in ammatory response, decrease the level of organ damage (10), and improve the survival of sepsis model animals (11).
The JAK/STAT pathway is considered one of the major signaling pathways involved in sepsis (12) and is a part of many key cytokine signaling pathways in the pathogenesis of sepsis, such as IL-4, IL-6, IL-10, IL-12, and IFN-γ (13)(14)(15). Early in the pathogenesis of sepsis, the body responds to infection with a strong immune response, manifested by a massive mobilization of pro-in ammatory cells and pro-in ammatory factors, which signi cantly elevates levels of pro-in ammatory cytokines such as IL-1β and IL-6, and causes an imbalance in the Th1/Th2 and Th17/Treg ratios(16-18). A prospective observational study found that an imbalance in the Th17/Treg ratio correlated with the severity and prognosis of sepsis patients (19). The in ammatory factor storm in the early stages of sepsis causes massive depletion of lymphocytes and leads to long-term immune dysfunction (20), increased susceptibility to secondary infections, and reduced 5-year survival (21,22). Currently, MSCs are widely used in the treatment of autoimmune diseases and acute and chronic in ammation, and a large number of studies have found that MSCs can regulate the helper T cell subsets represented by Th1 and Th17 cells.
The experimental animals investigated in the current study were all adult laboratory animals. To our knowledge, no studies have reported on the application of MSCs to naturally aged animals, which is inconsistent with the clinical situation in which the majority of sepsis patients are elderly. In the present study, we established a sepsis model by the CLP method in naturally aged SD rats, observed the effect of MSCs on improving survival and vital organ functions in aged sepsis model rats, clari ed the inhibitory effect of MSCs on the early in ammatory factor storm and the regulatory effect of helper T cell subsets in sepsis of the aged population and explored the molecular mechanisms of such effects.

Study design
This study was approved by the Ethics Committee of the Chinese PLA General Hospital. Animal experiments were approved by the Experimental Animal Welfare Ethics Committee of Chinese PLA General Hospital. The experimental animals were ninety (90) 21-month-old male SD rats (SPF (Beijing) Biotechnology, Beijing, China) weighing 801.5±79.5 g. The rats were divided into a sham-operated group (n=18), sepsis model group (n=36) and sepsis+MSCs treatment group (n=36). Equal number of rats in each group were randomly assigned to 3 subgroups and samples were collected after euthanasia at 6 h, 24 h or 72 h post-operation and/or treatment. The CLP model was used as a model of sepsis, and UC-MSCs were injected into the tail vein 1 h after CLP operation to induce sepsis. Five million (5×10 6 ) UC-MSCs in 1 ml saline were administered to each animal in the treatment group, and the sham-operated and model groups were injected with equal amounts of sterile saline at the same time. The 72 h survival rate of the rats in each group was statistically analyzed.
In vitro experiments were performed using lymphocytes isolated from the spleen of adult SD rats. Cells were either left unstimulated (control), stimulated with LPS (LPS), or stimulated with LPS and co-cultured with MSC (LPS+MSC). In vitro experiments were performed to validate the modulation of the JAK-STAT signaling pathway by MSCs using the JAK inhibitor AZD1480 (Cat# M2044, AbMole, TX, USA), the STAT1  inhibitor udarabine (Cat# M2028, AbMole, TX, USA), and the STAT3 inhibitor cryptotanshinone (Cat#   M3982, AbMole, TX, USA).

Cecal ligation-and-perforation (CLP)
A sepsis model was established by the modi ed CLP method (24). The rats were anesthetized by intraperitoneal injection of 50 mg/kg sodium pentobarbital to ensure that the rats did not wake up until the surgery was completed. After the rats were anesthetized and the abdominal body hair was removed, the skin was disinfected, a 1.5-cm-long midline incision was made in the abdomen, the cecum was located and gently externalized, the cecum was ligated at 1/2 of its length with a sterile 4-gauge wire, and a small amount of intestinal contents was extruded by penetrating the cecum once at the distal end of the ligature with a 22G needle. Finally, the cecum was placed back into the abdominal cavity, and the abdominal muscles and skin were sutured layer by layer. The rats in the sham-operated group only underwent open cecum externalization without ligation or perforation.

Histology and tissue staining
The right lung, liver, kidney, ileum and spleen of each experimental group of rats were harvested for histological observation and staining. The organs were xed in 4% paraformaldehyde, dehydrated and para n embedded. Para n sections were analyzed by hematoxylin-eosin (HE) staining (Cat# G1120, Solarbio, Beijing, China) to examine gross histology. TdT-mediated dUTP nick-end labeling (TUNEL) staining (Cat# C1088, Beyotime, Shanghai, China) was used to label apoptotic cells in the rat spleen. The stained sections were viewed and scanned with a panoramic MIDI CaseViewer system (3DHISTECH, Hungary).

Western blot
The spleen was homogenized using a low-temperature high-speed grinder (Servicebio, Wuhan, China) to obtain the homogenate. Lymphocytes for in vitro experiments were collected at pre-determined time points, rinsed three times with pre-cooled PBS (Servicebio, Wuhan, China), and lysed on ice for 10 min with RIPA lysis buffer (Thermo Fisher Scienti c, MA, USA). A BCA quanti cation kit (Solarbio, Beijing, China) was used for sample protein quanti cation, followed by incubation in a metal bath at 95°C for 10 min. Protein samples were transferred onto nitrocellulose membranes after sodium dodecyl sulfatepolyacrylamide gel electrophoresis ( When the data of each group conformed to normal distribution (Kolmogorov-Smirnov test) and chisquare (Levene test), one-way ANOVA was used for comparison of multiple sample means. A nonparametric test (Mann-Whitney U test) was used for pathological damage scores. The Kaplan-Meier method was used for survival analysis of each group of rats, and the statistical method was the log-rank (Mantel-Cox) test. p < 0.05 was considered a statistically signi cant difference. Graphs were plotted using GraphPad Prism 8.0 software.

MSCs improved survival and biochemical parameters in aged sepsis model rats
The mean body weight of the rats in each group was not signi cantly different and met the weight criteria for aged rats ( Fig. 1.A). A sepsis model was successfully established by the CLP method ( Fig. 1.B). UC-MSCs possessed lipogenic, osteogenic and chondrogenic differentiation abilities, and the expression of surface markers met the criteria(25) (Sup. Fig. 1). The mortality rate of aged sepsis model rats was reduced after MSC treatment ( Fig. 1. C-D), and the 72 h survival rate of rats in the treatment group was signi cantly higher than that of rats in the control-treated group (χ 2 = 13.56, P<0.01, Fig. 1.E). The levels of ALT, AST, CREA, BUN and LDH were signi cantly higher in aged sepsis model rats than in the shamoperated group. After treatment with MSCs, the levels of these biochemical indices became signi cantly lower and the levels of ALP became signi cantly higher ( Fig. 1.F-K).

MSCs ameliorate organ damage in aged sepsis model rats
MSCs were able to reduce the histopathological damage to critical organs, including the lung, liver, kidney and intestine. The alveolar structure of rats in the control-treated group were markedly disturbed, and widened septa and a large number of in ammatory cell in ltrates were observed. In contrast, rats in the treatment group showed signi cantly less lung damage and tissue damage scores were lower than the control-treated group (Fig. 2.A-B). The rats in the control-treated group showed massive in ammatory cell in ltration in the liver at the early stage of the disease, which was followed by hepatocellular edema, structural disorganization of liver lobules, vacuolar degeneration and debris-like necrosis, but the degree of liver injury was signi cantly reduced after MSC treatment ( Fig. 2.C-D). Sepsis caused kidney injury in rats, which manifested as interstitial edema with massive in ammatory cell in ltration and glomerular edema. The degree of injury was signi cantly reduced in the MSC-treated rats ( Fig. 2.E-F). In addition, the intestinal tract of rats in the control-treated group was severely damaged, the epithelial layer on both sides of the villi was markedly separated from the lamina propria, the apical part was broken, and the lamina propria was heavily in ltrated with in ammatory cells, which was signi cantly reduced after MSC treatment ( Fig. 2.G-H). The rats in the sham-operated group showed no signi cant tissue damage in these organs.
MSCs modulate systemic in ammatory responses in aged sepsis model rats Pro-in ammatory cytokines (IL-1β, IL-6, IL-17A, TNF-α, IFN-γ) were elevated in the control-treated rats compared to the sham-operated group; the levels of IL-1β, IL-6 and IL-17A began to increase early in the disease onset. In comparison, circulating levels of pro-in ammatory cytokines were signi cantly reduced in MSCs-treated rats, and anti-in ammatory cytokines (IL-4, IL 10, IL-13, TGF-β) were elevated compared to the control-treated group (Fig. 3.A-I). In addition, the proportions of Th1 and Th17 cell subsets among helper T cells were elevated in the peripheral blood of aged sepsis model rats, and after treatment with MSCs, the proportions of Th1 and Th17 cells were signi cantly decreased, and the ratios of Th1/Th2 and Th17/Treg were also signi cantly reduced compared to the control-treated group (Fig. 3.J-P).

MSCs modulate splenic cytokines and helper T cells in aged sepsis model rats
Flow cytometry revealed that the levels of pro-in ammatory cytokines were signi cantly higher in the spleens of aged sepsis model rats than in the spleens of sham-operated rats, and after MSC treatment, the levels of pro-in ammatory cytokines were reduced and the levels of anti-in ammatory cytokines were signi cantly increased in the treated rats ( Fig. 4A-I). The proportions of Th1 and Th17 CD4 + T cells in the spleen of aged sepsis model rats were signi cantly increased at the early stage of disease onset and gradually decreased thereafter, the proportions of Th1 and Th17 cells decreased after MSC treatment, and the ratios of Th1/Th2 and Th17/Treg cells were also signi cantly decreased ( Fig. 4J-P).
MSCs reduce apoptosis in the spleen and regulate STAT signaling activation levels in aged sepsis model rats Compared with the sham-operated group, the number of apoptotic cells in the spleen of rats in the control-treated group was signi cantly increased; a large number of apoptotic cells began to appear from the early stage of disease and persisted. The number of apoptotic cells in the spleen of MSC-treated rats was signi cantly lower than that of the control-treated group and comparable to that of the shamoperated group, and no signi cant apoptosis was observed (Fig. 5.A-B). Western blot assays of STAT1 and STAT3 revealed that the phosphorylation levels of STAT1 and STAT3 in the spleens of aged sepsis model rats were signi cantly increased at all time points compared with those in the spleens of normal rats, and maximal phosphorylation was observed at 24 h. The phosphorylation levels of STAT1 and STAT3 in the MSC-treated rats were signi cantly lower than those in the control-treated rats at the same time points.

MSCs ameliorate LPS-induced in ammatory responses in lymphocytes in vitro
Compared to blank controls, LPS-stimulated lymphocytes produced a large number of cytokines, of which the levels of pro-in ammatory cytokines were signi cantly elevated. In contrast, the levels of proin ammatory cytokines and anti-in ammatory cytokines in the cell supernatant of LPS-stimulated lymphocytes co-cultured with MSCs were signi cantly reduced. In addition, ow cytometry assays revealed that the fractions of Th1 and Th2 cells were signi cantly higher in LPS-stimulated lymphocytes, and the ratios of Th1/Th2 and Th17/Treg cells were also signi cantly higher, while the ratios of Th1/Th2 and Th17/Treg cells in the co-culture group was signi cantly lower than that in the LPS-stimulated group, which was consistent with the results of in vivo experiments.
Inhibitors validate the role of the JAK-STAT signaling pathway The administration of JAK inhibitor (AZD1480) resulted in a signi cant decrease in the ratios of Th1 and Th17 in lymphocytes compared to the LPS-stimulated sample, as well as a signi cant decrease in the ratios of Th1/Th2 and Th17/Treg. The levels of IFN-γ (Th1-speci c cytokine) and IL-17A (Th17-speci c cytokine) in the supernatant of the inhibitor-treated cells were also decreased. Furthermore, the ratios of Th1 and Th17 and the levels of their speci c cytokines were reduced in the cells co-cultured with MSC after the addition of JAK inhibitors compared to the inhibitor-treated cells, and the difference was statistically signi cant (Fig. 7.A-H). The proportion of Th1 cells and the levels of IFN-γ in lymphocytes were signi cantly reduced after the administration of STAT1 inhibitor, consistent with the changes in the co-culture sample. STAT1 inhibitor reduced the levels of STAT1 in lymphocytes, and the levels of p-STAT1 were signi cantly lower than those in the LPS-stimulated cells (Fig. 7I-N). Similarly, the proportion of Th17 in lymphocytes and the level of IL-17A were signi cantly reduced after the administration of STAT3 inhibitor, consistent with the changes in the co-culture sample, and the level of STAT3 in lymphocytes was reduced after the addition of the inhibitor, and the levels of p-STAT1 were signi cantly lower than those in the LPS-stimulated cells (Fig. 7.O-T).

Discussion
In this study, we established a sepsis model in the elderly using naturally aged SD rats to investigate the role and possible mechanisms of MSCs in treating sepsis in the elderly, regulating the immune in ammatory status and protecting organ functions. The results showed that MSCs derived from the umbilical cord could improve the overall status and 72 h survival of aged rats with sepsis and protect vital organ functions and ameliorate histopathological damages caused by sepsis. In vivo and in vitro experiments con rmed that MSCs could inhibit the systemic in ammatory response, regulate the Th1/Th2 and Th17/Treg ratios and in ammatory cytokine levels in aged rats with sepsis, and modulation of the JAK-STAT signaling pathway may be one of the mechanisms of action.
Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection(1), and the intense early in ammatory response causes a massive depletion of immune cells and immune factors, which is an important cause of immune suppression in the later stages of sepsis(26). The presence of some degree of immune decline in the elderly itself affects the progressive decline in the function of all types of immune cells of the innate and adaptive immune systems, especially CD8 + T cells and CD4 + T cells, and this results in a higher susceptibility to sepsis and mortality in the elderly (27). Currently, many studies have shown that MSCs can improve mortality and the degree of organ damage in sepsis model animals(28, 29), and the results of the latest preclinical meta-analysis have shown that MSC treatment can signi cantly reduce mortality in sepsis model animals. Current experimental results support a potential therapeutic effect of MSCs in clinical trials (30). To date, all three completed phase I clinical trials have con rmed the safety of MSCs applied to the treatment of patients with sepsis or endotoxemia (31)(32)(33). However, there are still no published studies on the application of MSCs in aged animals with sepsis or in elderly sepsis patients. We injected MSCs of umbilical cord origin into naturally aged rats with sepsis and found that the mental and feeding status of MSC-treated sepsis model rats improved, and that the 24 h and 72 h mortality rates were signi cantly reduced. Survival analysis revealed that the death of rats in the control-treated group was concentrated within 24 h after the completion of the CLP operation, which might be due to organ dysfunction or even failure caused by the outbreak of in ammatory response. In contrast, the death of rats in the MSC-treatment group occurred more randomly and the survival time was longer compared with that of the control-treated group, which indicated the important protective role of MSCs at the early stage of pathogenesis. Current studies con rm that the application of MSCs for sepsis not only improves the survival rate of experimental animals but also improves several vital organ functions. Clinically, patients with sepsis have a high morbidity and mortality rate, and in animal experiments, CLP sepsis animals had a 7-day mortality rate of 70%(34) and a 28-day mortality rate of 80% (11). After treatment with MSCs, survival rates were signi cantly improved, critical organ functions such as lung, liver, and spleen were signi cantly protected, and pathological damages to vital organs were signi cantly reduced (35). However, some researchers found that MSCs did not improve mortality or systemic in ammation and stress in pigs with sepsis (36).
In our study, we found that MSCs could signi cantly improve the pathological damage to vital organs, reduce the serum levels of ALT, AST, BUN, Cr and LDH and increase the level of ALP in aged sepsis model rats. This indicates that MSCs possess good therapeutic effects in the aged sepsis model rats, which can effectively reduce organ damage and protect organ functions, thus MSCs improved the survival rate of aged sepsis model rats. By histological staining, we found that local in ltration of in ammatory cells was reduced after MSC treatment. Accordingly, we speculate that allaying the in ammatory factor storm in the early stage of sepsis may be an important mechanism by which MSCs exert their protective effects.
The cytokine storm is an excessive immune response induced by various stimuli, which is the most important pathophysiological feature of early sepsis and the main cause of multi-organ dysfunction and long-term immune suppression (37). Immune cells play a crucial role in the initiation phase of a cytokine storm. When the organism is stimulated by pathogenic microorganisms, activated CD4 + T cells differentiate into different subpopulations to play different roles; they are mainly divided into proin ammatory cell subpopulations represented by Th1 and Th17 and anti-in ammatory cell subpopulations represented by Treg and Th2. Sepsis is a dysregulation of the pro-in ammatory/antiin ammatory response caused by severe infection, as evidenced by an imbalance in the ratios of Th1/Th2 and Th17/Treg cells (38,39). In the early stages of infection, a variety of in ammatory cells are recruited and activated, and they release large amounts of in ammatory cytokines and chemokines, such as TNF-α and IL-1β, which are rapidly secreted and reach a peak within a few hours. Under normal conditions, the body regulates the degree of in ammatory response by secreting anti-in ammatory factors to remove harmful substances and maintain intracellular homeostasis (40,41). However, when this balance is disrupted, early reactive cytokines trigger a cascade of additional cytokines, which results in the activation and release of a large number of in ammatory factors, including IL-6, IL-12, and macrophage in ammatory protein (MCP)-1α, and induces an uncontrolled systemic in ammatory response (42,43). MSCs have been found to reduce serum levels of in ammatory factors in adult sepsis model animal studies (30,44), and this has been con rmed in aged sepsis model animals. In our study, we found that the ratios of Th1/Th2 and Th17/Treg cells were signi cantly elevated in the peripheral blood of aged sepsis model rats, which indicates that helper T-cell subsets play a role in the systemic in ammatory response. We also found that the ratios of Th1/Th2 and Th17/Treg cells were signi cantly reduced after treatment with MSCs and approached the levels comparable to the sham-operated group.
Furthermore, cytokines associated with helper T cells, such as IFN-γ, IL-17, IL-4 and IL-10, showed the same trend as the change in the ratio of immune cells, which also con rmed that MSCs regulated the secretory function of Th cells, and kept the excessive in ammatory response under control. In addition, MSCs were able to reduce the levels of pro-in ammatory cytokines such as IL-1β, IL-6 and TNF-α and elevate the levels of anti-in ammatory cytokines such as IL-10 and TGF-β in the circulation of aged sepsis model rats. This prevented the initiation and development of cytokine storms to a certain extent. Based on the in vivo experiments, we established an in vitro sepsis model by adding LPS to rat lymphocytes cultured in vitro. The results showed that, compared with the LPS-stimulated sample, the cells co-cultured with MSCs produced signi cantly less pro-in ammatory factors and signi cantly more anti-in ammatory factors at all time points, and that the ratios of Th1/Th2 and Th17/Treg cells were decreased in the culture. From these results, we can conclude that MSCs can regulate the in ammatory response in vitro and in vivo by regulating the number and function of Th cell subpopulations.
The spleen, as the largest peripheral immune organ, accumulates a large number of lymphocytes and is an important site for the speci c immune response of the organism. Sepsis induces increased apoptosis of B lymphocytes and effector T cells, which results in immune paralysis of the organism against subsequent infections. Current studies have con rmed that sepsis causes apoptosis of a large number of lymphocytes in the spleen, and the number of apoptotic splenocytes and lymphocytes is closely related to the severity of in ammation and patient prognosis (45). In patients with sepsis, the rate of lymphocyte apoptosis was signi cantly higher among those who died within 28 days after disease onset than those who were alive on day 28 (46). Using TUNEL staining of spleen tissues of aged sepsis model rats in each experimental group, we found that the proportion of apoptotic cells was signi cantly higher in the spleen of aged sepsis model rats, and that the number of apoptotic cells was signi cantly reduced after treatment with MSCs. This indicates that MSCs can inhibit lymphocyte apoptosis induced by sepsis and suggests that MSCs have a positive effect on preventing immune dysfunction and long-term immunosuppression. To further investigate the source of changes in circulating Th cells and in ammatory factors, we examined the levels of in ammatory factors and the ratios of Th cell subpopulations in the spleen and found that the overall trends of changes in in ammatory factors and Th cell subpopulations in the spleen of rats in each group were consistent with those in peripheral blood.
This indicates that MSCs can regulate the ratio of Th cells in the spleen and maintain the Th1/Th2 and Th17/Treg balance, which in turn regulates the levels of in ammation-related cytokines and controls the in ammatory response. After treatment with MSCs, the activation of Th1 and Th17 cells in the spleen of aged sepsis model rats was reduced, and the number of activated cells released into the peripheral blood was correspondingly reduced. This regulated the in ammatory response and exerted a protective effect on all organs of the body.
The JAK-STAT pathway is capable of mediating cell proliferation and apoptosis (47), which is thought to be involved in sepsis-induced multi-organ dysfunction (12).The JAK-STAT signaling pathway is activated by both pro-in ammatory cytokines (IFN-γ, IL-12, and IL-27) and anti-in ammatory cytokines (IL-4, IL-10, and IL-13), which indicates that individual STAT family members differentially regulate the balance of Th cell subsets. In the helper T-cell subpopulation, Th1 cell differentiation is mainly regulated by STAT1 and STAT4(48), while STAT6 and STAT3 regulate the differentiation of the immunoregulatory Th2 and Th17 cells, respectively (49). One study showed that STAT1 knock-out mice were signi cantly resistant to LPSinduced endotoxemia (50) and CLP-induced septic shock (51). This was attributed to the altered balance of the Th1/Th2 immune response, which resulted in signi cantly reduced organ damage in the liver and kidney. Another study showed that JAK2 inhibitors improved survival after CLP operation in mice and rats by reducing the expression of pro-in ammatory mediators such as TNF-α, IL-6 and high mobility group box-1 (HMGB-1) (52). To explore the molecular mechanism of Th cell regulation by MSCs, we examined the relevant STAT signaling pathways in the spleen of aged sepsis model rats. We found that the phosphorylation levels of STAT1 and STAT3 were signi cantly increased in the spleen of sepsis model rats, the phosphorylation levels of both molecules were signi cantly reduced after MSC treatment, and the expression levels of STAT1 and STAT3 were comparable to those of normal rats. In the in vitro experiments, we administered inhibitors of the JAK, STAT1 and STAT3 signaling pathways to the LPS-stimulated cells and LPS-stimulated cells co-cultured with MSC. The results showed that the proportion of Th cell subpopulations in the LPS-stimulated sample changed after treatment with individual inhibitor, and the overall trend of the changes was consistent with that in the LPS+MSC cells. This indicated that MSCs and JAK-STAT inhibitors exerted the same effect, while the results of LPS+MSCs treated with inhibitors showed that MSCs and inhibitors could function synergistically to some extent. In addition, we evaluated the levels of IFN-γ and IL-17, which represent Th1 and Th17 cells, respectively, in each sample.
We found that JAK inhibitors were able to reduce the levels of both IFN-γ and IL-17, while STAT1 inhibitor reduced the levels of IFN-γ and STAT3 inhibitor reduced the levels of IL-17. The above results con rm that MSCs regulate the quantity and function of the relevant Th cell subpopulations by inhibiting the JAK-STAT signaling pathway to control the in ammatory response. If this is the case, can JAK inhibitors be used directly to treat sepsis? The answer is uncertain. The MSCs used in this study have more diverse regulatory effects than the single effect of JAK-STAT pathway inhibitors. The mechanisms of immunomodulatory effects of MSCs are complex; they include the synthesis and secretion of multiple mediators, direct interaction with target cells, and regulation by certain antigen-presenting cells. In MSCs, no single pathway controls the entire process (53), and their regulatory effects are in uenced by the microenvironment and immune status. Thus, we con rmed that MSCs could protect aged sepsis model rats through the regulation of the JAK-STAT signaling pathway, but we also acknowledge that this is only the tip of the iceberg of the mechanism of action of MSCs, and that more complex mechanisms remain to be investigated in depth.
Although our study con rmed the protective effect of MSCs in aged sepsis model rats, there are still some limitations; our understanding of the molecular mechanism of Th cell regulation by MSCs is incomplete.
We also have not clari ed whether the immune function of aged sepsis model rats is different or similar to that of adult sepsis model rats. We will continue to investigate related issues in depth in our subsequent studies.

Conclusion
In conclusion, this study con rmed that treatment with MSCs could improve survival, protect vital organ functions, regulate sepsis-induced in ammatory factor storms and ameliorate immune dysfunction in aged sepsis model rats. We propose that inhibition of the JAK-STAT signaling pathway may be one of the mechanisms of MSC-mediated protection.     the ratio of The ratio of Th17/Treg in spleen of each group (*p<0.05, **p<0.01, ***p<0.001).

Figure 5
Tunel staining and STAT1 and STAT3 expression levels in the spleen of rats in each group. (A) Apoptosis of spleen lymphocytes at 6h, 24h and 72h in each group of rats (scale bar=100μm); (2) number of apoptotic cells per eld of view in each group of rats; (C-D) expression levels and semi-quantitative analysis of STAT1 and p-STAT1 in spleen tissue of each group of rats; (E-F) expression levels and semiquantitative analysis of STAT3 and p-STAT3 in spleen tissue of each group of rats. Results (**p<0.01, ***p<0.001).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.