Mesenchymal Stem Cell Conditioned Medium Improves Non-alcoholic Fatty Liver Disease in Type 2 Diabetic Mice by Regulating SIRT1

patients with NAFLD clinically.


Introduction
Non-alcoholic fatty liver disease (NAFLD) is a series of liver diseases, de ned by the presence of steatosis in more than 5% of liver cells with little or no alcohol consumption [1]. NAFLD includes the benign nonalcoholic fatty liver (NAFL) and the more serious non-alcoholic steatohepatitis (NASH). Obesity, insulin resistance, and diabetes are closely related to the accumulation of excessive lipids in the liver parenchyma. NAFLD is characterized by excessive accumulation of lipids in liver cells and is the most common chronic liver disease in the world. In the past few decades, the prevalence of NAFLD has risen sharply. Studies have shown that up to 70% of diabetic patients suffer from NAFLD [2,3]. There is a strong association between NAFLD and diabetes risk [4,5]. When a patient is diagnosed with NAFLD, the risk of developing diabetes increases by approximately ve times [6]. Diabetes can also increase the risk of NAFLD progression [7][8][9]. Among 432 patients with NAFLD con rmed by biopsy, coexisting type 2 diabetes (T2DM) was found to be an independent risk factor for liver brosis [10]. Compared with any condition that exists alone, coexisting NAFLD is associated with serious adverse consequences of diabetes.
In the past 10 years, mesenchymal stem cells (MSCs) have become the most promising resource for cell transplantation due to their immunoregulatory function [11], paracrine function [12], and transdifferentiation regulatory function [13]. Extensive studies have shown that human umbilical cord mesenchymal stem cells (Hu-MSCs) reduce liver brosis in rats [14] and improve NAFLD in obese mice [15], highlighting their ability to reduce liver cell damage. However, the known treatments of MSCs have their own limitations, including residency in organs [16], a lack of effective utilization [17], a low survival rate in vivo [18], and the potential for tumorigenicity of implanted MSCs [19], which may hinder the development of stem cell therapy. However, the cell-free MSC conditioned medium (CM) is rich in nutrients, such as growth factors and cytokines, and is easily available, making the application of MSC-CM an alternative treatment option. Related studies based on MSC-CM have shown that MSC-CM can improve diabetic endothelial dysfunction [20] and reduce renal brosis [21]. However, the potential protective effects of MSC-CM on NAFLD in diabetes and the exact mechanism of action are still unclear.
Mitochondria are organelles with a double-membrane structure. They are the energy generators of cells and play a central role in the oxidation of nutrients to produce energy [22]. A large number of studies have shown that mitochondrial dysfunction plays an important role in the occurrence and progression of NAFLD [22,23]. Mitochondrial dysfunction (especially oxidative respiratory chain defects) plays a key role in the physiopathology of NASH. Oxidative stress is the key to the pathogenesis of NAFLD, and its main feature is the production of reactive oxygen species (ROS). In NAFLD, mitochondrial dysfunction is the main cause of ROS accumulation. Dysfunction of the oxidative respiratory chain and impaired lipid peroxidation (caused by mitochondrial dysfunction) further increase the production of ROS. The production of ROS will in turn aggravate lipid peroxidation and cause a variety of harmful effects to liver cells and other cells. The production of highly reactive aldehyde derivatives (such as malondialdehyde), further forming a vicious circle, may damage mitochondrial proteins and mitochondrial DNA (mtDNA) [9,24], further suggesting that mitochondrial dysfunction will aggravate the severity of NAFLD [25]. Therefore, improving mitochondrial function is a potential treatment strategy for reducing NAFLD [26].
Sirtuin 1 (SIRT1) is the rst member of the mammalian homologues of the class III histone deacetylases.
SIRT1 regulates lifespan and cell metabolism. SIRT1 can mediate the deacetylation of downstream target proteins such as peroxisome proliferator-activated receptor gamma coactivator 1α (PGC1α) to participate in fatty acid oxidation and mitochondrial biogenesis and function [27][28][29]. Upregulation of SIRT1 effectively reduces obesity and insulin resistance in NAFLD rodents [30], whereas downregulation or complete silencing of SIRT1 can exacerbate fatty liver and in ammation [31]. In obesity and NAFLD patients, the expression of SIRT1 was found to be reduced in plasma and the liver [32,33].
The purpose of the present study was to use in vitro palmitic acid (PA)-treated human primary hepatocytes (L-O2 cells) and a streptozotocin (STZ)-treated diabetic mouse model to explore the potential therapeutic effects of MSC-CM on diabetes with NAFLD. We further explored whether the observed effects depend on the improvement of mitochondrial function. We found that the SIRT1 signal transduction pathway plays an important role in the underlying molecular mechanisms. These ndings shed new light on the potential of cell-free CM to treat diabetes mellitus combined with NAFLD.

Materials And Methods
Cell culture and treatment With the approval of the ethics committee of Qilu Hospital of Shandong University, we obtained full-term human umbilical cords through cesarean section. All participants signed an informed consent form for the use of umbilical cord in this study. After washing the umbilical cord three times in saline, the arteries and veins were removed, and then the interstitial tissue of Wharton's jelly was exposed. Then, the tissue was cut into small pieces with sterile scissors and placed in a cell culture ask containing 20% fetal bovine serum (FBS) in α-MEM (Gibco, MD, USA), which was placed in an incubator at 37°C containing 5% CO 2 . We changed the medium every three days. The third to seventh passages of cells were used for experiments.

Animal experiments
A total of 45 male 6-week-old C57BL/6 mice (about 20 g each) were purchased from Synergy Pharmaceutical Bioengineering Co., Ltd. (Nanjing, China). After 2 weeks of adaptive feeding, the mice were randomly divided into the following two groups: normal chow diet (NCD, n = 15) and 45% high fat diet (HFD, n = 30). After 8 weeks of feeding the HFD and 12 h of fasting, these mice were injected with STZ (100 mg/kg, S0130; Sigma-Aldrich) intraperitoneally. If two consecutive fasting blood glucose levels were ≥16.7 mM, the T2DM model was considered successfully established. The T2DM mice were divided into two groups (n = 15 per group): the control (PBS) group and the MSC-CM group. Then, as mentioned above, after STZ injection, 200 μl PBS and MSC-CM were injected weekly through the tail vein, once every three days, for three consecutive months.

Preparation of MSC-CM from human umbilical cord mesenchymal stem cells (Hu-MSCs)
Hu-MSCs were inoculated into culture asks. When the MSCs were fused, the medium was changed to serum-free medium and cells were cultured for an additional 24 h. Then the supernatant was collected, and a centrifugal lter device (Ultracel-10K; Millipore, Billerica, MA) was used with a cut-off value of 10 kDa to perform ultra ltration on the MSC supernatant according to the manufacturer's instructions to produce MSC-CM ( nal concentration: 1 mg/mL), which was ltered with a 0.2 µm lter and stored at −80°C until use.

Metabolic parameter analysis
Body weight and fasting blood glucose were monitored weekly. Serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglyceride (TG) and total cholesterol (TC) were detected using ELISA kits (ColorfulGene Biological Technology, Wuhan, China). After fasting for 12-16 h, the intraperitoneal glucose tolerance test (IPGTT) was performed by injecting glucose (2 g/kg, Sigma-Aldrich). After fasting for 6 h, the intraperitoneal insulin resistance test (IPITT) was performed by injecting insulin (0.75 IU/kg, Jiangsu Wanbang Pharmaceutical, China). Then blood was collected from the tip of the tail vein at 0, 30, 60, 90, 120, and 180 min for glucose measurement. Before the anesthetized mice were euthanized, the body fat mass of the mice was measured by a dual-energy X-ray absorption method (DEXA, Ge Lunar Prodigy, USA) (n = 7 per group). Then, the liver tissue was harvested and stored in liquid nitrogen or xed with 4% phosphate buffered formaldehyde (PBF).

Western blot analysis
The cells were harvested and lysed in RIPA buffer. The protein concentration was measured with a BCA analysis kit (P0012S, Beyotime, Shanghai, China). After SDS-PAGE and transfer to a membrane, the membrane was incubated with the primary antibody overnight at 4°C and then with the horseradish peroxidase-labeled secondary antibody. Protein bands were imaged using Image Lab software (BioRad, USA). The intensity of protein bands was measured by ImageJ and normalized to β-actin. The liver tissue slices were processed according to standard hematoxylin and eosin (HE) staining techniques, and then the pathological changes of these tissues were observed under an optical microscope. The para n sections were depara nized and washed in PBS, and then incubated in preheated 10 mM sodium citrate buffer at 95°C for 15 min. The slides were washed, incubated with 3,3'diaminobenzidine (DAB) as a chromogen, and then examined under an optical microscope.
Glycogen periodic acid Schiff staining Terminal deoxynucleotidyl transferase dUTP nick-end labeling staining After washing with PBS, the L-O2 cell slides were xed in 4% paraformaldehyde (P1110; Solarbio) for 1 h and then permeabilized with 0.5% Triton X-100 for 10 min. A terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay was performed using the in situ cell death detection kit, POD (11684817910; Roche, USA). After depara nization of the liver para n section with xylene, the in situ cell death detection kit and TMR red (12156792910; Roche) were used to perform TUNEL staining according to standard methods.

Detection of intracellular ROS
The dichlorodihydro uorescein diacetate Reactive Oxygen Species Determination Kit (Beyotime, China) was used according to the manufacturer's instructions to measure intracellular ROS levels.

Statistical analysis
All experiments were repeated at least three times. The results are reported as the mean ± standard error of the mean (SEM). The statistical differences between the groups were determined by the paired Student t-test or by one-way analysis of variance (ANOVA) using GraphPad Prism 8 software (San Diego, California, USA). P < 0.05 was considered to indicate statistical signi cance.

MSC-CM improves glucose tolerance in T2DM mice and increases insulin sensitivity
To determine whether MSC-CM has therapeutic effects in T2DM mice, Hu-MSCs were rst isolated and identi ed. To identify Hu-MSCs, the surface marker levels and the multi-lineage differentiation ability of adherent cells were rst examined. The results of Oil Red O staining and Alizarin Red staining indicate that Hu-MSCs have the potential to differentiate into adipocytes (Fig.1a) and osteoblasts (Fig.1b). Then, ow cytometry analysis was used to characterize Hu-MSCs by determining whether they express surface markers such as CD105 and CD73. These markers were enriched in Hu-MSCs, while the negative markers HLA-DR and CD34 exhibited low expression (Fig.1c).
Next, we established a T2DM mouse model by feeding HFD and injecting the mice with STZ. MSC-CM was injected into mice through the tail vein, and PBS (200 μl) was used as a negative control. As shown in the gure, MSC-CM signi cantly reduced the body weight and body fat percent of T2DM mice (Fig.1d,e). The IPGTT and IPITT results revealed that MSC-CM signi cantly improved glucose metabolism and insulin sensitivity in T2DM mice. These data indicate that MSC-CM improves glucose tolerance and insulin sensitivity in T2DM mice (Fig.1f,g).

MSC-CM ameliorated the pathological changes and blood lipids in the livers of T2DM mice and alleviated liver dysfunction
We further observed the effects of MSC-CM on the liver structure of T2DM mice. HE staining showed that the liver structure of the T2DM group was disordered, and steatosis and vacuolar changes were increased, while in the livers of T2DM mice treated with MSC-CM, tissue structure improved, and steatosis and vacuolar changes were reduced. Glycogen staining revealed that glycogen deposition in the T2DM group was reduced, and glycogen accumulation was higher in the MSC-CM intervention group compared with the PBS group. Oil Red O staining revealed that lipid droplets in the T2DM group were signi cantly increased, and lipid droplets in the MSC-CM group were signi cantly reduced compared with the PBS group (Fig.2a). Then, we analyzed the effects of MSC-CM on liver dysfunction and the blood lipid pro le. Compared with the control group, the liver ALT (Fig.2b) and AST (Fig.2c) levels and the blood TC (Fig.2d) and TG (Fig.2e) levels were signi cantly increased in T2DM mice, while in T2DM mice treated with MSC-CM they were signi cantly lower compared with the PBS treatment group, indicating that MSC-CM alleviated liver dysfunction of T2DM mice and improved the lipid distribution.

MSC-CM enhances liver mitochondrial function and reduces in ammation and apoptosis
Mitochondrial function plays a key role in maintaining normal energy metabolism, and mitochondrial dysfunction is one of the main causes of NAFLD. SIRT1 mediates the deacetylation of PGC1α. The downstream transcription factors NRF1(nuclear respiratory factor 1), TFAM(mitochondrial transcription factor A), and UCP2(uncoupling protein 2) of PGC1α are all key to maintaining mitochondrial function. We found that the protein levels of SIRT1, PGC1α, NRF1, TFAM, and UCP2 in the liver cells of T2DM mice were lower compared with the normal control group, while the protein levels increased after MSC-CM treatment (Fig.3a,b). Compared with normal mice, the levels of pro-in ammatory cytokines TNF-α, IL-1β, and IL-6 in the livers of diabetic mice were signi cantly increased, while the expression levels of proin ammatory cytokines decreased after MSC-CM intervention (Fig.3c). Further evaluating the degree of liver cell damage, it was found that the ratio of BAX (a pro-apoptotic protein) to Bcl-2 (an anti-apoptotic protein) in liver cells of T2DM mice increased, and the protein levels of apoptotic markers (cleaved caspase-3/caspase-3 and cleaved PARP/PARP) were signi cantly increased, and the ratios of BAX/Bcl-2, cleaved PARP/PARP, and cleaved caspase-3/caspase-3 were lower after MSC-CM intervention than in the PBS group (Fig.3d). TUNEL staining indicated that the proportion of TUNEL-positive (apoptotic) cells in T2DM livers was signi cantly increased, and the proportion of TUNEL-positive (apoptotic) cells was reduced after MSC-CM treatment compared with the PBS group (Fig.3f). The total oxidative capacity of liver cells was measured using a commercial kit, and the results showed that the Trolox equivalent antioxidant capacity (TEAC) of T2DM liver cells decreased, while the total antioxidant capacity of the cells increased after MSC-CM intervention (Fig.3e).

The effects of MSC-CM on L-O2 cells treated with PA in vitro
In order to verify the speci c mechanism by which MSC-CM protects liver cells, PA and PA+MSC-CM were used to stimulate L-O2 cells. We found that the in vitro results were consistent with the in vivo experimental ndings: SIRT1 and PGC1α were down-regulated in L-O2 cells stimulated by PA (Fig.4a). The expression of downstream transcription factors NRF1, TFAM, and UCP2 decreased (Fig.4b). While, the expressions of SIRT1 and PGC1α in L-O2 cells treated with MSC-CM were higher than those in the PA treatment group (Fig.4a), the expressions of the downstream transcription factors NRF1, TFAM and UCP2 were also higher than those in the PA treatment group (Fig.4b). The PA-treated L-O2 cells pro-in ammatory cytokine TNF-α, IL-1β, and IL-6 are up-regulated, but down-regulated in the MSC-CM treatment group (Fig.4c). Apoptosis indicators BAX/Bcl-2, cleaved caspase 3/caspase 3 and cleaved PARP/PARP are increased in PA-treated L-O2 cells, while the expression was decreased after MSC-CM treatment (Fig.4d). TUNEL staining also proved that MSC-CM reduced the apoptosis of L-O2 cells treated with PA (Fig.4f). Finally, we tested the total antioxidant capacity and ROS production of the cells. The results showed that MSC-CM enhanced the total antioxidant capacity of the cells after PA treatment (Fig.4e) and reduced the production of ROS (Fig.5a). Overall, these data indicate that MSC-CM has positive therapeutic effects on L-O2 cells in vitro.
MSC-CM maintains the biological functions of mitochondria through the SIRT1 pathway, and after SIRT1 is silenced, the antioxidant, anti-in ammatory, and apoptotic effects of MSC-CM are offset To verify whether SIRT1 is essential for maintaining mitochondrial function, siRNA was used to silence SIRT1 expression to observe changes in L-O2 cell mitochondrial function. The results showed that after silencing SIRT1, the effects of MSC-CM on the SIRT1, PGC1α, NRF1, TFAM, and UCP2 proteins levels in L-O2 cells were signi cantly offset (Fig.5b,c), ROS production increased (Fig.6c), and the total antioxidant capacity decreased (Fig.6b). After SIRT1 siRNA was transfected into L-O2 cells, MSC-CM could not reduce the BAX/Bcl-2, cleaved PARP/PARP, and cleaved caspase 3/caspase 3 ratios (Fig.5e), TUNEL staining showed increased apoptosis (Fig.6a),, and the levels of the pro-in ammatory cytokines TNF-α, IL-1β, and IL-6 increased (Fig.5d),, con rming that SIRT1 mediates the protective effects of MSC-CM on liver cells.

Discussion
Globally, the incidence of NAFLD is increasing [34]. It is estimated that the global prevalence of NAFLD is as high as 1 billion [35]. NAFLD is caused by excessive lipid accumulation in the liver, and the presence of NAFLD is an indicator of insulin resistance [36]. Therefore, NAFLD is generally regarded as a liver manifestation of metabolic syndrome. In addition, it is a major risk factor for T2DM and is usually found as a comorbidity in patients with T2DM [37].
The molecular mechanisms underlying the development from steatosis and steatohepatitis to advanced liver disease and from steatosis to severe liver damage are not fully understood. Complex processes in the mitochondria play a central role, including energy and ROS production, calcium homeostasis, the regulation of apoptosis, and lipid metabolism [38]. Recent studies have shown that ROS are overproduced, thereby exacerbating mitochondrial dysfunction and increasing the vicious circle of liver cell oxidative damage being involved in the occurrence and progression of NAFLD [39]. In NAFLD, oxidative stress and lipid peroxidation induce the production of pro-in ammatory cytokines, including TNF-α, IL-1β, and IL-6 [40]. Excessive production of ROS and reduced antioxidant defenses can also damage mtDNA. Oxidative damage to nuclear DNA may amplify mitochondrial damage by endangering the transcription of key mitochondrial proteins. It has been reported that the expression levels of PGC1α, TFAM, and NRF1, the key regulatory factors involved in mitochondrial metabolism and organelle biogenesis in NAFLD, are reduced [41,42]. Their upstream target protein SIRT1 has been identi ed as a potential therapeutic target for NAFLD [43].
Recent research has indicated that MSC-CM therapy has several big advantages compared with MSC transplantation therapy. MSC-CM is rich in nutrients, is easy to obtain, has a high utilization rate, and does not increase potential tumorigenicity. Studies have shown that MSC-CM has positive effects in disease treatment, further highlighting the broad prospects of MSC-CM in the application of liver diseases.
Our research has shed light on the great potential of MSC-CM in the treatment of diabetes and NAFLD. MSC-CM can not only improve insulin resistance and increase insulin sensitivity in T2DM mice, but also relieve liver dysfunction and improve the blood lipid pro le. It is further found that MSC-CM can reduce in ammation and apoptosis and enhance the antioxidant capacity by maintaining mitochondrial function, thereby reducing the damage of NAFLD liver cells associated with diabetes. In vitro, MSC-CM had the same effect on PA-treated L-O2 hepatocytes. MSC-CM increased the expression of SIRT1, PGC1α, NRF1, TFAM, and UCP2 in PA-treated L-O2 cells, enhanced the total antioxidant capacity of the cells, improved the expression levels of pro-in ammatory cytokines, and reduced apoptosis. After silencing SIRT1, the positive effects of MSC-CM were partially offset. We conclude that MSC-CM can improve NAFLD in diabetic mice by regulating the expression of SIRT1.
Recent studies provided more insights into the role of SIRT1 in maintaining normal liver development and function [44] and the effects of SIRT1 activators on fatty liver disease [45,46]. SIRT1 enhances the antioxidant capacity by deacetylating FOXO and PGC1α, thereby protecting the liver against oxidative stress, and reducing local and circulatory conditions by deacetylating NF-κB in hepatic and adipose tissue. In our study, we explored the relationship between MSC-CM, NAFLD, and SIRT1, and found that MSC-CM can act as an agonist of SIRT1, signi cantly upregulate the expression level of SIRT1 in liver cells, and improve NAFLD. In summary, our ndings elucidate a new molecular mechanism underlying the effects of MSC-CM, which can prevent and/or treat diabetes and its related fatty liver disease.

Conclusions
In summary, our current results indicate that MSC-CM can effectively alleviate NAFLD by enhancing the biological functions of mitochondria, reducing in ammation, and inhibiting cell apoptosis in T2DM models both in vivo and in vitro and that these e cacies were associated with the upregulation of SIRT1.
Taken together, our ndings clarify a new molecular mechanism related to MSC-CM-based therapies that can prevent and/or treat diabetes and its related liver disease.

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding authors on reasonable request.
Ethics approval and consent to participate With the approval of the Ethics Committee at Qilu Hospital of Shandong University, all participants provided informed consent for the use of the umbilical cord in this experimental study. All animal experiments were conducted in accordance with the Animal Ethics Committee of Shandong University.

Consent for publication
Not applicable.