Hippo Pathway Activation in Aged Mesenchymal Stem Cells Contributes to the Dysregulation of Hepatic Inflammation in Aged Mice

Abstract Aging is always accompanied by chronic diseases which probably attribute to long‐term chronic inflammation in the aging body. Whereas, the mechanism of chronic inflammation in aging body is still obscure. Mesenchymal stem cells (MSCs) are capable of local chemotaxis to sites of inflammation and play a powerful role in immune regulation. Whether degeneration of MSCs in the aging body is associated with unbalanced inflammation is still not clear. In this study, immunosuppressive properties of aged MSCs are found to be repressed. The impaired immunosuppressive function of aged MSCs is associated with lower expression of the Hippo effector Yes‐associated protein 1 (YAP1) and its target gene signal transducer and activator of transcription 1 (STAT1). YAP1 regulates the transcription of STAT1 through binding with its promoter. In conclusion, a novel YAP1/STAT1 axis maintaining immunosuppressive function of MSCs is revealed and impairment of this signal pathway in aged MSCs probably resulted in higher inflammation in aged mice liver.


Introduction
[3][4][5][6][7] These age-related chronic diseases are always caused by longterm aberrant chronic inflammation, which has been named "inflamm-aging." [8]Normal levels of inflammation can help organisms to defend against microbial invasion and repair tissues, and the inflammatory response will fade down when it is no longer needed through various immunosuppressive mechanisms. [9,10]Unbalanced inflammation could lead to severe organ damage and perturbed homeostasis.In aging organisms, inflammation is difficult to control once triggered.Multiple mechanisms have been reported to contribute to augmentation of age-related inflammation, [11] including redox stress, glycation, dysfunction of mitochondria, deregulation of the immune system, and so on.Whereas, the negative regulation mechanism of inflammation in the "inflamm-aging" remains unclear and deserves more attention.
Inflamm-aging is also associated with higher incidence and poor outcomes of liver diseases in the elderly. [12][15] In this study, we aimed at studying why aged liver is prone to a higher level of inflammation when exposed to injury.Mesenchymal stem cells (MSCs) are a population of adult stem cells that are defined by their ability of selfrenewal and multipotent differentiation. [16]MSCs have also been focused on for their nonstem cell properties. [17]They are widely considered to be able to migrate into tissue injury sites and inhibit inflammation locally. [18]However, aged MSCs have been proved to exhibit an functional decline. [19]It is unclear whether immune regulatory dysfunction in aging MSCs results in enhanced age-related hepatic inflammation and liver diseases.In this study, we will investigate changes in immune regulatory function of aged MSCs and their impact on enhanced age-related hepatic inflammation.[23] Thus, this work may also help to build new strategies for improving the clinical efficacy of MSCs.
[29] The active Hippo kinase cascade (which involves the kinases MST1, MST2, LATS1, and LATS2) phosphorylates the core effector Yes-associated protein 1 (YAP1), which results in its cytoplasmic retention or degradation. [30]When the Hippo pathway is shut down, YAP1 translocates into the nucleus and promotes the transcription of target genes.YAP1 has been demonstrated to play a key role in both innate and adaptive immune responses.The role of YAP1 in adaptive immune response was mostly studied in T-cells of the tumor microenvironment.For instance, YAP1-induced tumor necrosis factor  (TNF-) expression promotes myeloid-derived suppressor cells and impedes cytotoxic T-cell infiltration in ovarian carcinomas. [31]On the other hand, several studies showed that YAP1 inhibited effector T-cell differentiation by acting as an internal factor.Gene expression analyses of tumor-infiltrating T-cells following Yap1 deletion implicates YAP1 as a negative mediator of global T-cell responses in the tumor microenvironment and patient survival in diverse human cancers. [32]Therefore, YAP1 can be considered as a negative regulator of T cell response.Overall, the immunoregulatory roles of YAP1 in parenchymal cells, immune cells, and tumor cells have started to be demonstrated. [33,34]However, it is still unclear how YAP1 affects the regulation of stromal cells on the adaptive immune response.
In this study, we found that the immunosuppression ability of MSCs declines with aging and this phenomenon is accompanied with YAP1 downregulation.Furthermore, we investigated the un-derlying mechanisms by which YAP1 regulates immunoregulation of MSCs.This study will help us understand why inflammation in the liver increases with aging and what we can do to improve the clinical efficacy of MSCs in the treatment of immunerelated diseases.

Liver Inflammation Response Increases with Aging
For observing the inflammation level in aged liver, we reanalyzed the data from a previous scRNA-seq study of liver CD45 + immune cells in young (3-6 months) and aged (17-24 months) C57BL/6J mice.Based on the cell markers presented in Figure S2A, Supporting Information, we identified eleven clusters of immune cells (Figure S2B, Supporting Information and Figure 1A) and found that there were more T cells infiltration (CD8 + effector T-cells, naive T-cells, and NKT cells) in aged liver (23.3% in aged liver &12.5% in young liver) (Figure 1B and Figure S2C).Among the T cells, we can see that CD8 + effector T cells were most obviously different between young and aged liver.To further study the alteration of inflammation level, especially CD8 + T cell infiltration, in the aged liver with injury, we established a classical autoimmune hepatitis model [35][36][37] by using 15 mg kg −1 concanavalin A (conA) in young and aged mice, in which CD8 + T cells were reported to play main roles at the stage of hepatocyte apoptosis and liver injury. [38]Eight hours after conA injection, we quantified total T-cells by detecting CD3positive cells using flow cytometry and immunohistochemistry.The data showed an increased number of infiltrated T-cells in aged liver of both normal and injured mice (Figure S2D,E, Supporting Information).Then, we analyzed distinct T cell subtypes and several other immune cell types by flow cytometry, including CD8 + T cells, CD4 + T cells, NK cells, and neutrophils.As shown in Figure 1C, Figure S2F,G, Supporting Information, CD8 + Tcells, and neutrophils responded to conA in both young and aged mice at 8 h after treatment.No difference of neutrophils was observed between conA-treated young mice and conA-treated aged mice.However, compared to young mice treated with conA, there was a significantly higher proportion of CD8 + T-cells in conAtreated aged mice.Immunohistochemical evaluation of CD8 expression also showed that CD8 + T-cells were over-activated in the aged liver, which is consistent with the results of flow cytometry (Figure 1D,E).To test the inflammatory status of mice, we measured the levels of inflammatory factors in the serum, including TNF-, interferon- (IFN-), interleukin 5, interleukin 6, and interleukin 12.[41] As a result, the serum concentration of inflammatory factors was higher in aged mice (Figure 1F).General observation and H&E staining revealed more severe pathological structural alteration and more hepatic necrosis in aged liver (Figure 1G,H).Serum levels of alanine transaminase (ALT) and aspartate transaminase (AST) were also much higher in aged mice, consistent with liver injury (Figure 1I).These data demonstrated that there is an increased CD8 + T-cell infiltration and higher inflammation in aged mice with conA treatment.

Immunosuppression Capacity of Aged MSCs Decreases
MSCs have the ability to migrate to injury sites and regulate inflammation locally.To test if uncontrolled inflammation in the aged liver results from degenerative alterations in MSCs, we isolated liver MSCs from young and aged Lepr-Cre; loxP-tdTomato mice upon conA treatment.[45] TdTomato-labeled MSCs from young and aged liver adhered to the bottom of the plate and presented a spindle-shape morphology (Figure 2A).The liver MSCs were then co-cultured with CFSE-labeled splenocytes.Fluorescence activated cell sorter (FACS) analysis showed that aged liver MSCs inhibited the proliferation of splenocytes much less effectively than young liver MSCs (Figure 2B,C).
[48] Therefore, we acquired bone marrow MSCs from young and aged WT mice to study the mechanism of weakened immunosuppressive function of aged MSCs.We first characterized their phenotypes and gene expression properties, and aged MSCs expressed more P53 and P21 (Figure S3A,B, Supporting Information).Transcriptome sequencing analysis also showed that the gene expression profile was significantly different between young and aged MSCs (Figure S3C, Supporting Information).Enrichment analysis of the differentially expressed genes showed that aged MSCs expressed higher levels of genes associated with aging-associated biological processes (Figure S3D, Supporting Information).The inflammatory microenvironment plays an important role in promoting the immunosuppressive effects of MSCs.The combination of IFN- and TNF- (I+T) was reported to be able to activate MSCs to suppress inflammation. [49]e treated MSCs with I+T inflammatory factors and found that they induced changes of gene expression in both young and aged MSCs (Figure 2D).However, many of the genes which were dramatically upregulated in young MSCs with I+T stimulation were expressed at much lower levels in aged MSCs with I+T stimulation.GO analysis showed that the downregulated genes in I+Ttreated aged MSCs compared with I+T-treated young MSCs were enriched in processes related to negative regulation of inflammation (Figure 2E).Next, we co-cultured bone marrow young and aged MSCs with splenocytes to observe the direct immunosuppression effect of MSCs.As shown in Figure 2F,G, we can see that aged MSCs only weakly inhibited splenocyte proliferation.
To further assess the effect of young and aged MSCs on the regulation of inflammation in vivo, we injected young and aged MSCs into conA-treated young mice with untreated normal young mice as controls.CD8 staining in liver tissue showed that young MSCs greatly reduced the activation of CD8 + T cells induced by conA, whereas aged MSCs had a much weaker effect on CD8 + T-cell activation (Figure 2H,I).Detection of inflammatory factors in serum showed that transplantation of young MSCs greatly reduced the levels of inflammatory factors such as TNF- and IFN-, but aged MSCs had a weaker effect (Figure 2J).In this liver injury model, conA induced severe hepatocyte necrosis and high levels of the aminotransferases ALT and AST.Transplantation of young MSCs strongly reduced the necrosis and the aminotransferase levels, but the protective effect of aged MSCs was less significant (Figure 2K-M).
The ability of MSCs to suppress inflammation can be affected by various factors, including their survival in vivo, migration, and recruitment to the targeted sites.As YAP1 also plays an important role in regulating cell survival and apoptosis, [50,51] we assessed acquisition and apoptosis of young and aged MSCs.As shown in Figure S4A,B, , Supporting Information there were no differences in acquisition between young and aged MSCs on the third and fifth days after isolation.Moreover, the expression levels of apoptosis-related proteins BIM, BID, and cleaved CASPASE-3, were comparable between young and aged MSCs.In addition, YAP1 overexpression did not affect apoptosis of aged MSCs (Figure S4C).To verify the migration and homing ability of aged MSCs, we transfected both young and aged MSCs with EGFP-labeled adenovirus and then transplanted them into mice with liver injury.Recruitment of young and aged MSCs into injured liver was detected by immunofluorescence staining at 8 h posttransplantation.The recruitment of young and aged MSCs to liver injury sites did not exhibit any discernible differences (Figure S4D,E, Supporting Information).We also assessed the migration ability of young and aged MSCs in vitro by transwell assay.The migratory capacity of aged MSCs was comparable to that of young MSCs (Figure S4F,G).Thus, the repressed immunosuppression of aged MSCs is probably attributed to their weakened inhibitory effect on T-cells, rather than any alterations in survival and migration abilities.

YAP1 is Downregulated in Aged MSCs and Necessary for Immunosuppressive Function of MSCs
From GSEA analysis of transcriptome sequencing data from young and aged MSCs, we found that Hippo signaling pathway was upregulated in aged MSCs compared with young MSCs (Figure 3A).We also investigated the activation of MST and LATS kinases of classical Hippo signaling pathway.Hippo signaling was over-activated in aged MSCs, as indicated by higher expression of LATS1, p-LATS1/2, MST1/2, and p-MST1/2 (Figure S5, Supporting Information).The Hippo signaling pathway plays a key role in the immune response which is mainly dependent on the core effector YAP1, which is negatively regulated by Hippo signal activation. [25]We then examined the mRNA expression levels of Yap1 and its target genes (Areg, Ccnb1, Birc5 and Cyr61) by RT-PCR.The results showed that all of them were downregulated in aged MSCs (Figure 3B).YAP1 regulates the expression of target genes by translocating to the nucleus and interacting with transcription factors.We observed that nuclear levels of YAP1 were higher in young MSCs than in aged MSCs (Figure 3C-F).We also tested YAP1 expression in young and aged liver MSCs, we found that YAP1 protein levels were indeed downregulated in aged liver MSCs (Figure 3G).On the other hand, we are interested in whether conA could induce YAP1 alteration in liver MSCs and bone marrow MSCs in young and aged mice.We isolated liver and bone marrow MSCs from young and aged mice with and without conA treatment.Western blotting assay showed that conA did not affect YAP1 expression in young and aged MSCs from liver and bone marrow (Figure S6, Supporting Information).To further verify the role of YAP1 in the immunosuppressive effect of MSCs, we established mice with MSC-specific knockout (KO) of Yap1 (Yap1 MSCs-KO ) by crossing Yap1 FL/FL and Lepr-cre mice (Figure S1A,B, Supporting Information).LEPR can be used to trace MSCs in vivo, [52] and thus we used it here as a marker of MSCs.WT, Yap1 MSCs+/− (heterozygous Yap1 knockout in MSCs), and Yap1 MSCs-KO mice were subjected to liver injury by conA injec-tion through the tail vein.As shown in Figure 4A,B, Yap1 knockout in endogenous MSCs led to more CD8 + T cell infiltration.H&E staining and aminotransferase detection showed that Yap1 knockout in MSCs increased the level of liver injury compared with WT mice (Figure 4C-E).Heterozygous Yap1 MSCs+/− MSCs did not show any difference in CD8 + T cell infiltration and liver injury compared with WT mice.These data suggested that YAP1 played an essential role in the immunosuppression function of MSCs.

YAP1 Maintains the Immunosuppressive Function of MSCs by Promoting iNOS Expression
Next, we wanted to find the YAP1-regulated effector molecule which determines the decreased immunosuppressive function of aged MSCs.We analyzed the mRNA expression levels of a series of genes encoding immunosuppression-related factors in young and aged MSCs after treatment with IFN- and TNF- (I+T).The expression of Nos2 in aged MSCs after I+T treatment was much lower than that in I+T-treated young MSCs (Figure 5A).Expression of other immunosuppression-related genes, such as Tnfaip6, Tgfb1, and Il10, was similar between young and aged MSCs (Figure 5A).Analysis of iNOS protein expression and nitrate concentration in the medium of I+T-treated MSCs verified the above results (Figure 5B,C).Our previous study showed that immunosuppression of MSCs was dependent on iNOS [49] and iNOS knockout MSCs had a significantly reduced therapeutic effect on conA-induced liver injury. [53]Further, we verified the effect of iNOS in immunosuppression of young and aged MSCs by reverse intervention experiments in conA-treated mice.At first, we verified iNOS expression in Nos2 knockout young MSCs and Nos2 overexpressing aged MSCs (Figure S7C,D, Supporting Information).As shown in Figure 5D-G, knockout of iNOS weakened the CD8 + T-cell inhibition effect and liver injury repair of young MSCs.Conversely, iNOS overexpression enhanced CD8 + T cell inhibition and liver injury repair of aged MSCs.These data indicated that reduction in the level of the immunosuppression factor iNOS probably played a key role in the diminished immunosuppression function of aged MSCs.
Combining the above data, it appears that YAP1 expression is consistent with iNOS expression in young and aged MSCs.Next, we wanted to clarify whether YAP1 regulates iNOS expression.We knocked down Yap1 in young MSCs and then treated MSCs with IFN- and TNF-.iNOS expression was inhibited after Yap1 knockdown.Overexpression of Yap1 in aged MSCs enhanced iNOS expression (Figure 5H,I).These data suggested that YAP1 can control iNOS expression.

YAP1 Dominates iNOS Expression via STAT1
The pathways regulating iNOS expression seem to vary in different cells and different species.In general, activation of the transcription factors nuclear factor B (NF-B)) and signal transducer and activator of transcription 1(STAT1) seems to be an essential step in the regulation of iNOS expression in most cells. [54]e tested both STAT1 and NF-B signaling in young and aged MSCs.As shown in Figure S9A, after I+T stimulation, the expression of p-P65 in aged MSCs was higher than in young MSCs, and the expression of IB was lower in aged MSCs.There was also more P65 in the nucleus in aged MSCs with or without I+T treatment (Figure S9B,C, Supporting Information).All of these data indicated that the NF-B signal pathway was over-activated in aged MSCs, which was not consistent with the iNOS expression.We concluded that the diminished immunosuppressive function of aged MSCs was not dependent on NF-B signaling alteration.When we analyzed STAT1, we found that both unmodified STAT1 and phosphorylated STAT1 expression were downregulated in aged MSCs (Figure 6A).Furthermore, less p-STAT1 translocated into the nucleus in aged MSCs after I+T treatment for 60 min (Figure 6B,C).STAT1 is activated by IFN-.Thus, these results could explain why the biological process "response to IFN-" was downregulated in aged MSCs (Figure 2E).Next, we tested iNOS expression after manipulating Stat1 expression in young and aged MSCs.As shown in Figure S7E, Supporting Information, sh2-Stat1 was chosen to knock down the expression of Stat1.In young MSCs with Stat1 knockdown, p-STAT1 expression, and iNOS expression were both downregulated (Figure 6D).When Stat1 was overexpressed in aged MSCs, p-STAT1, and iNOS expression were both upregulated (Figure 6E).These data suggested that decreased expression of STAT1 plays a key role in the decreased iNOS expression in aged MSCs.In consideration of regulation effect of STAT1 on iNOS, we also verified the effect of STAT1 in immunosuppression of MSCs.From Figure 6F-I, we found that reverse intervention of STAT1 in young and aged MSCs reversed their T cell inhibition effect and liver injury repair, which is consistent with iNOS intervention effect as shown in Figure 5D-F.
To verify the effect of YAP1 on STAT1 expression, we used the Yap1 intervention adenoviruses to reverse the expression of Yap1 in young and aged MSCs.Knockdown of Yap1 in young MSCs suppressed total STAT1 expression and p-STAT1 expression after I+T treatment (Figure 7A).Overexpression of Yap1 in aged MSCs promoted STAT1 expression and p-STAT1 expression after I+T treatment (Figure 7B).To further verify the effect of YAP1/STAT1 signaling in regulating iNOS expression, we performed rescue experiments.Stat1 overexpression rescued the effect of Yap1 knockdown in young MSCs, and Stat1 knockdown reduced the effect of Yap1 overexpression in aged MSCs (Figure 7C,D).Thus, to conclude, YAP1 regulates iNOS expression through the regulation of STAT1 expression.

YAP1 Regulates STAT1 Expression by Binding the Stat1 Promoter through YAP1/TEADs Complex
To further elucidate the molecular mechanism underlying the regulation of STAT1 expression by YAP1, we investigated the effect of YAP1 on Stat1 mRNA expression by RT-PCR.Knockdown of Yap1 suppressed the mRNA expression of Stat1 in young MSCs, and Yap1 overexpression in aged MSCs promoted the mRNA expression of Stat1 (Figure 7E).We wanted to understand how YAP1 regulates Stat1 transcription.[57] Verteporfin (VP) inhibits the activity of YAP1 by blocking the interaction of YAP1 and TEADs.As shown in Figure 7F, STAT1 expression was suppressed in the presence of VP along with other YAP1 target genes.This result suggested that regulation of STAT1 expression by YAP1 is probably dependent on the binding of YAP1 with TEADs.Through JASPAR, several target sequences of TEADs were predicted in the mouse Stat1 promoter.To verify molecular mechanism of YAP1/TEADs complex regulating STAT1, we used the CUT&Tag assay upon YAP1/TEADs in both young and aged MSCs.We found that the sequence 5′-AGCTTACCTC-3′ in the Stat1 promoter was bound by YAP1/TEADs complex (Figure 7G).

Higher Expression of YAP1 in MSCs is Correlated with Lower Inflammation in Human Liver Disease
To further explore whether YAP1 expression in MSCs is correlated with the inflammation status in human liver diseases, we analyzed publicly available scRNA-seq datasets of human liver diseases.Of note, a dataset of human liver fibrosis (GSE186343) contained information related to the inflammation status of the patients.For this dataset, based on cell markers presented in Figure S10A, Supporting Information, we identified 12 cell types (Figure S10B, Supporting Information).[60][61][62][63] Then, we analyzed the corre-  lation between the expression of YAP1 in MSCs and the inflammation levels in liver.We divided the samples into two groups according to the inflammation levels shown by the authors (Figure S10C, Supporting Information) and we found that samples with higher expression of YAP1 in MSCs were associated with lower inflammation (Figure S10D, Supporting Information).We also analyzed the correlation of the expression of YAP1 in MSCs with the ratio of CD8 + T-cells in another four datasets (GSE129933, GSE124395, GSE189903, GSE151530).We integrated these four datasets together and identified thirteen cell types (Figure S11A-C, Supporting Information).Then, we divided the samples into two groups based on YAP1 expression in MSCs (Figure S11D, Supporting Information) and analyzed the ratio of immune cells.There was a decrease of CD8 + T cell proportion in the group with higher expression of YAP1 in MSCs (Figure S11E, Supporting Information).We also analyzed the correlation of YAP1 and STAT1 in human liver MSCs.As shown in Figure S11F, Supporting Information, a strong positive correlation between YAP1 and STAT1 was observed.
Altogether, our animal results combined with the scRNA-seq analysis in human samples suggested that the function of MSCs in suppressing inflammation in liver diseases was dependent on YAP1 expression.

Discussion
While aging itself is not generally considered as a pathological process, aged liver is more sensitive to injuries and poten- tially lead to liver diseases.[15] Understanding why the elderly liver is prone to high levels of inflammation is very important for preventing and treating age-related diseases.Here, we show that downregulation of the YAP1/STAT1 axis activity in aged MSCs resulted in impaired immunosuppression ability of MSCs, and thus contributed to higher inflammation in aged liver.
First, we analyzed publicly available single-cell RNA sequencing data carried out in young and aged mouse liver, and found that the proportion of T-cells, especially CD8 + effector T cells, increased significantly in aged mouse liver.[66] These effector T-cells also directly kill infected or damaged cells through releasing cytotoxic molecules, such as perforin and granzymes.This killing of cells can cause tissue damage and further exacerbate inflammatory responses. [67]To investigate the higher inflammation in aged liver and the immuneregulation role of aged MSCs, we used an inflammatory liver injury model induced by conA.This conA-based model has been reported to be suitable for investigating the involvement of T-cell infiltration. [68]Our study revealed that aged liver exhibited higher inflammation after injury, and this probably resulted from the repressed immunosuppressive function of aged MSCs targeting CD8 + T cells (Figure 8).We also identified a critical role of the YAP1/STAT1 axis in maintaining the immunosuppressive effect of MSCs and the downregulation of YAP1/STAT1 axis probably led to repressed immunosuppression of aged MSCs.
Nonetheless, we also observed a mild increase of other immune cell types in aged liver from the scRNA-seq analysis (Figure 1B, Supporting Information).Other immune cell types, such as CD4 + T cells, [69] NK cells, [70] and neutrophils [71,72] were shown to be involved in conA-induced liver injury.Therefore, we also tested these immune cells in our model.However, only a significant difference of CD8 + T cells between young and aged mice treated with conA was observed.Thus, besides the modulation of CD8 + T cells, whether the weakened immunosuppression capability of aged MSCs is also relevant to other immune cell subtypes should be further studied at different stages of conA induced liver injury model.
In previous studies, YAP1 was reported to maintain human MSCs in a younger state, and overexpression of YAP1 rejuvenated senescent MSCs in a replicative senescence model. [73]An anti-aging effect of YAP1 was also found in hepatocytes, hepatic stellate cells, and tumor cells. [26,74,75]There are studies demonstrating that YAP1 plays an important role in maintaining the function of stem cells. [76]In terms of immune regulation, most studies focused on the role of YAP1 in immune cells and tumor cells.YAP1 can regulate the differentiation of Treg and B cells and the polarity of macrophages. [74]There are studies reporting that stromal cells can regulate the tissue microenvironment.Stromal cells have been reported to promote cancer resistance and metastasis. [77,78]However, there are few researchers studying the role of YAP1 in immunoregulation in stromal cells.We conducted studies on YAP1-mediated immunoregulation in MSCs, which are very important stromal cells in the inflammatory microenvironment.
iNOS, one of the most important immunosuppressive factors of MSCs in rodents, was found to be downregulated in aged MSCs, which is due to downregulation of YAP1/STAT1 axis.YAP1 could regulate STAT1 transcription by binding with its promoter through YAP1/TEADs complex.A CUT&Tag experiment using an antibody against YAP1/TEADs helped us to find the exact motif of Stat1 promoter bound by YAP1/TEADs.We also found Hippo pathway was over-activated in aged MSCs.However, we are also interested in understanding how the Hippo pathway is regulated in aged MSCs.The Hippo pathway has been reported to be controlled by a variety of mechanisms, including intrinsic cell machineries such as cell-cell contact, cell polarity, and the actin cytoskeleton molecular cascades, as well as a wide range of extracellular signals, including cellular energy status, mechanical cues, and hormonal signals that act through G-protein-coupled receptors (GPCRs). [29]GPCR signaling can either activate or inhibit YAP1/TAZ, depending on which classes of downstream heterotrimeric G-protein are coupled with. [79]In addition, data from multiple systems suggested major roles for the GPCR pathways in regulating stem cell function both in vivo and in vitro. [79,80]nterestingly, the loss of complexity, sensitivity, and specificity of GPCR signaling over time is likely to be a potent driving factor of pathological aging worldwide. [80]Currently, 40% of drugs targeting GPCRs were approved by the FDA [81] and have been used in ongoing clinical trials. [82]Thus, it would be interesting to conduct further research on the GPCR pathway in aged MSCs.
From the clinical prospective, the role of YAP1 in immune regulation function of human MSCs should be studied further.Using publicly available scRNA-seq data, we observed a negative correlation between YAP1 expression in MSCs and inflammation level of patients with liver disease.We also saw a decreased CD8 + T-cells infiltration in patients with higher expression of YAP1 in MSCs.Our in vivo experiments complemented by the human liver scRNA-seq analyses suggested that the YAP1/STAT1 axis plays a crucial role in the immunosuppression by MSCs.Therefore, YAP1 overexpression or activation in MSCs could be an attractive strategy to improve MSCs-based therapies.Currently, several small molecules were identified as potential activators of YAP1 in vitro. [83,84]However, further research is needed to develop an efficient platform for YAP1 activation or overexpression in MSCs.
In conclusion, we show a reduced immunosuppressive ability of aged MSCs, which is associated with higher inflammation in aged liver.Mechanistically, we describe a YAP1/STAT1 axis which is important for maintaining immunosuppression properties of MSCs.Therefore, our study provides insight into the mechanism of aging-associated inflammation and opens new perspectives for improving existing MSCs-based therapies.

Figure 1 .
Figure 1.Liver inflammation response increases with aging.A) Previous scRNA-seq data from liver of young and aged mice were re-analyzed.Identification of different clusters of cell types in young and aged mouse liver.B) Proportion of different immune cell populations in young and aged mouse liver.C-H) Young and aged mice were intravenously injected with 15 mg kg -1 of conA, followed by sacrificed at 8 h.C, The CD8 + T-cells were subsequently analyzed by FACS and quantified.D, E, IHC staining and quantification of CD8 + T-cells in the liver.F, Concentrations of inflammatory factors in serum were detected by Bio-plex assay.G, H&E staining of liver sections.H, Quantification of the necrotic area in the liver.I) ALT, and AST were detected in the serum of young and aged mice with or without conA treatment.N.D., not detected.Data were analyzed using two-tailed unpaired Student t-test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.Graphs showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 2 .
Figure 2. The immunosuppression capacity of aged MSCs is reduced.A) MSCs with tdTomato fluorescence isolated from young and aged Lepr-cre; loxP-tdTomato mice were observed by a fluorescence microscopy.B,C) FACS analysis of the proliferation of splenocytes after co-culture with young and aged liver MSCs.Splenocytes without treatment were used as negative control.D,E) mRNA expression profiles of young and aged bone marrow MSCs after IFN- plus TNF- (I+T) treatment for 2 h were detected by transcriptome sequencing.Heatmap (D) and GO analysis (E) of the differentially expressed genes between young and aged bone marrow MSCs.F,G) FACS analysis and quantification of the proliferation of splenocytes after co-culture with young and aged bone marrow MSCs.Splenocytes without treatment were used as negative control.H-M) Young and aged bone marrow MSCs (1 × 10 6 per mouse) were transplanted into young mice at 30 min after injection with 15 mg kg −1 of conA.Mice were sacrificed at 8 h after conA injection.Normal young mice were used as control.H, I, IHC staining and quantification of CD8 + T-cells in the liver.J) Levels of inflammatory factors were quantified by Bio-plex assay in the serum of conA-treated mice after transplantation of young and aged MSCs.K,L) H&E staining of liver tissue was performed and areas of necrosis were quantified.M) ALT, and AST levels were measured in the serum of mice.Y-MSCs, young MSCs; A-MSCs, aged MSCs.N.D., not detected.Data were analyzed using one-way ANOVA test.*p<0.05,**p < 0.01, ***p < 0.001, ****p < 0.0001.Graphs showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 3 .
Figure 3. Downregulation of YAP1 in aged MSCs leads to reduced immunosuppressive function.A GSEA analysis of gene expression of young and aged MSCs shows that Hippo signaling is upregulated in aged MSCs.B) mRNA expression levels of Yap1 and YAP1 target genes were tested by RT-PCR.C,D) Cytosolic and nuclear expression of YAP1 was detected by western blotting and quantified relative to the internal controls -ACTIN and Histone H3.E,F) Translocation of YAP1 into the nucleus was detected in young and aged MSCs by immunofluorescence staining and corresponding quantification.G) YAP1 expression was tested in young and aged liver MSCs by western blotting.-ACTIN was used as the internal control.H,I) IHC staining and quantification of CD8 + T-cells in the liver.J,K) H&E staining of the liver after treatment with young and aged MSCs, with or without manipulation of Yap1 expression, and quantification of the necrotic areas.L) Measurement of serum ALT and AST levels in the indicated mice.Y-MSCs, young MSCs; A-MSCs, aged MSCs.N.D., not detected.Data (B, D, F) were analyzed using two-tailed unpaired Student t-test.Data (I, K, L) were analyzed using one-way ANOVA test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.Graphs showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 5 .
Figure 5. YAP1 maintains the immunosuppressive activity of MSCs by promoting iNOS expression.A) mRNA expression levels of genes encoding antiinflammation factors were detected in young and aged MSCs by RT-PCR after treatment with I+T for 0 h, 4 h, and 8 h.B) Expression of iNOS protein in young and aged MSCs was detected by western blotting assay at 0, 24, and 48 h after I+T treatment and corresponding quantification.-ACTIN was used as the internal control.C, Nitrate concentrations were measured by a Griess reagent kit in conditioned medium of young and aged MSCs after I+T treatment.D,E) IHC staining of CD8 positive cells in the liver after treatment with young and aged MSCs (with or without manipulation of Nos2 expression) and corresponding quantification.F,G) H&E staining of the liver after treatment with young and aged MSCs (with or without manipulation of No' expression) and quantification of the necrotic areas.H,I) Young MSCs were transfected by shYap1 adenovirus, and aged MSCs were transfected by Yap1 overexpression adenovirus.iNOS expression was detected by western blotting assay in young and aged MSCs treated with I+T for the indicated time.Y-MSCs, young MSCs; A-MSCs, aged MSCs.N.D., not detected.Data (A, B, C) were analyzed using two-tailed unpaired Student t-test.Data (E, G) were analyzed using one-way ANOVA test.***p < 0.001, ****p < 0.0001.Graphs showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 6 .
Figure 6.Reduction of iNOS expression in aged MSCs is dependent on STAT1 downregulation.A) Young and aged MSCs were treated with I+T for 0, 30, and 60 min.STAT1 and p-STAT1 (Ser727) expression were detected by western blotting, -actin was used as the internal control.B,C) Immunofluorescence staining and quantification of p-STAT1 translocation into the nucleus in MSCs after treatment with I+T for 60 min.D) Young MSCs were transfected by shStat1 adenovirus and then treated with I+T for 0, 30, and 60 min.STAT1 and p-STAT1 expression were detected by western blotting.-ACTIN was used as the internal control.iNOS expression in young MSCs was detected by western blotting after I+T treatment for 0, 24, and 48 h.E) Aged MSCs were transfected by Stat1 overexpression adenovirus and then treated with I+T for 0, 30, and 60 min.STAT1 and p-STAT1 expression were detected by western blotting.-ACTIN was used as the internal control.iNOS expression in aged MSCs was detected by western blotting after I+T treatment for 0, 24, and 48 h.F, G) IHC staining of CD8 positive cells in the liver after treatment with young and aged MSCs, with or without manipulation of Stat1 expression and corresponding quantification.H,I) H&E staining of the liver after treatment with young and aged MSCs, with or without manipulation of Stat1 expression, and quantification of the necrotic areas.Y-MSCs, young MSCs; A-MSCs, aged MSCs.N.D., not detected.Data (C) were analyzed using two-tailed unpaired Student t-test.Data (G, I) were analyzed using one-way ANOVA test.**p < 0.01, ***p < 0.001, ****p < 0.0001.Graphs showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 7 .
Figure 7. YAP1 promotes iNOS expression by binding the Stat1 promoter and enhancing Stat1 transcription.A, B, Young MSCs were transfected by shYap1 adenovirus and then treated with I+T for 0, 30, and 60 min.Aged MSCs were transfected by Yap1 overexpression adenovirus and then treated with I+T for 0, 30, and 60 min.Expression of YAP1, STAT1, and p-STAT1 (Ser727) was detected by western blotting.-ACTIN was used as the internal control.C,D) Detection of iNOS expression by western blotting assay in young and aged MSCs in the rescue experiment.-ACTIN was used as the internal control.E) The mRNA expression levels of Stat1 in young and aged MSCs were detected by RT-PCR after adenovirus-mediated manipulation of Yap1 expression.F) Young MSCs were treated with and without 10 μM VP for 24 h.The levels of STAT1, CTGF, and CYR61 were detected by western blotting assay.-ACTIN was used as the internal control.G) CUT&Tag assay was performed to verify the target motif bound by YAP1/TEADs in the Stat1 promoter.Anti-IgG was used as the negative control.Y-MSCs, young MSCs; A-MSCs, aged MSCs.Data were analyzed using two-tailed unpaired Student t-test.**p < 0.01, ***p < 0.001.The graph showed mean ± S.D. N = 3, the experiments were repeated for three times.

Figure 8 .
Figure 8. Schematic model to explain the higher level of liver inflammation in aged mouse liver.Higher inflammation in aged mice liver with injury is probably due to immunosuppression impairment of aged MSCs.In young mice, MSCs have an effective inhibitory effect on T-cells because of higher iNOS expression caused by YAP1/STAT1 axis activation.Whereas, in aged mouse, Hippo signal is over-activated in MSCs, and iNOS expression driven by YAP1/STAT1 axis is repressed, which leads to impaired inhibitory effect on T-cells.