Human Umbilical Cord Mesenchymal Stem Cells Grown in 3D Micro-environment Ameliorate Acute Liver Injury by Immunoregulation and Proliferation Promotion

Mesenchymal stem cells (MSCs) have been widely used for inammatory diseases like acute liver injury (ALI) and COVID-19 diseases. However, most MSCs instead of the cell-matrix are cultured in 2 D plates, resulting in degeneration of some key functions along with a limited understanding of the immune response in tissue repairing. Herein, we obtained the human umbilical cord MSCs grown on microcarriers in 3 D suspension culture (3 D hUCMSCs) that proved more appreciative than 2 D hUCMSCs for the ALI therapy. kinases (MAPKs) and STAT3 signaling pathways. Conclusively, the revelation of the superior moderating effect of 3 D hUCMSCs and possible regulatory mechanisms will improve the clinical ecacy of stem cell therapy for inammatory and immunological diseases.


Introduction
Acute liver injury (ALI) is one of the most important causes of mortality worldwide, which usually results in a rapid loss of hepatic function. ALI is associated with an adaptive in ammation response triggered by noxious stimuli, such as infection [1] and chemicals [2]. Macrophages, being the main in ammationmediated cells, release various in ammatory cytokines such as tumor necrosis factor α (TNF-α), interleukin-6 (IL-6), interleukin-1β (IL-1β) [3,4], and other in ammatory mediators upon activation by bacterial [5] or virus products [6]. Overexpression of these factors usually leads to a wide range of progressive diseases, including cancer, neurological disease, metabolic disorders and acute tissue injury [7][8][9][10][11]. Thus, suppressing the overexpression of in ammatory cytokines could be a robust tool to ameliorate ALI.
Mesenchymal stem cells (MSCs) are considered as the most important members of stem cell family and found in many tissues, such as bone marrow [12,13], fat [14,15], and umbilical cord [16]. Above all, one of the most intriguing biological characteristics of MSCs is their immunomodulatory property, which lays the therapeutic basis of MSCs for in ammatory diseases like autoimmune [17] and COVID-19 diseases [18,19]. The induction of tissue regeneration after MSCs administration was initially believed that the regeneration occurred upon MSCs-migration and differentiation into the damaged populations [20,21].
Although MSCs could transdifferentiate functional cells [22], increasing evidence gradually indicates that the speci c differentiation of MSCs in damaged tissues is only partly responsible for the disease treatments. Moreover, the paracrine effects act as an alternative strategy. However, most of the conclusion is based on the fact that the MSCs are cultured in 2 D plates in vitro, which disregard the complexity of interactions seen in vivo. Essentially, in 2D, cells have more surface area to contact with the culture plate and media than with other cells [2], forcing them into a polarization that fails to re ect physiological conditions. This probably results in the degeneration of some key functions and limits the understanding of the immune response in tissue repairing [23,24]. Thus, it is very necessary to explore a more realistic model like a three-dimensional (3D) culture system, where the cells are grown on a scaffold-based substrate that maximally mimics the extracellular matrix. Morphology, gene expression patterns, cell cycle and proliferation is much closer to physiological conditions.
To this end, herein, we obtained the human umbilical cord MSCs cultured in 3 D microcarriers (3 D hUCMSCs, Scheme 1) and provided the evidence that both the 3 D hUCMSCs and the conditioned medium were more favorable for the therapy of acute liver injury (ALI). Nitric oxide production, cell proliferation, in ammatory cytokines/mRNA expression were comprehensively evaluated. In addition, the change of phosphorylation level of extracellular signal-regulated kinases (ERK), c-Jun N-terminal kinases (JNK), the p38 and STAT3 were assessed to reveal the possible signaling pathways. Consequently, we demonstrated the 3 D hUCMSCs acts as a promising modulator of immune activation, cell proliferation and apoptosis, which suppresses ALI in mice by MAPKs and STAT3 signaling pathways. In short, we believe such a disclosure will bring new hope to patients to cure in ammatory and immunologically mediated diseases.

Materials And Methods
Macrophage In ammation Model in vitro RAW 264.7 murine macrophage-like cell line was purchased from the Stem Cell Bank, Chinese Academy of Sciences (Shanghai, China). Cells were cultured in petri dishes (Thermo Fisher Scienti c, LabServ, China) using Dulbecco's modi ed Eagle medium with high glucose (H-DMEM, Gibco, USA), supplemented with 10% heat-inactivated FBS and 100 U/mL penicillin/streptomycin in a humidi ed air with 5% CO 2 at 37 ℃. The medium was replaced after every 2 days until cells reached 70-80% con uency. An in ammatory model in vitro was set up by using lipopolysaccharide (LPS, Merck KGaA, Darmstadt, Germany) to induce RAW 264.7 cells at a concentration of 100 ng/mL (denoted as LPS group).

Culture of hUCMSCs
For 2 D culture, the hUCMSCs were seeded at a density of 5×10 4 cells / mL in a T-75 cell culture ask (Scheme 1) and the medium was replaced every three days. hUCMSCs on the third passage (P3) were adopted for the subsequent experiments [25]. All the numbers and diameter of cells were counted by the automatic counting cell analyzer (CountStar®Regel S3, Ruiyu, China).
For 3 D suspension culture, 1.5 L bioreactors (Guoqiang, China) was used in this study, equipped with dissolved oxygen (DO) probe (Mettler, Toledo) and a pH probe (Mettler, Toledo). The vessels were lled with α-MEM medium and 10% FBS ( 600 mL), hUCMSCs (1.35 × 10 5 cells/mL) and Cytodex3 microcarrier (1.8 g, 10 cells/bead). The cells were then inoculated at 37 ℃ with air and 40% DO by bottomspace gassing. The reactors was agitated at 55 rpm for 120 hours to provide the 3 D suspension cultured cells.

Phenotypic Analysis
The P3 hUCMSCs were stained with uorescent-labeled monoclonal antibodies, including positive markers (APC CD105, FITC CD73 and FITC CD90) and negative ones (FITC CD34, FITC CD45 and FITC HLA-DR). All the markers were purchased from BD Biosciences (USA). IgG was used as an isotype control and cultured under the same condition. The stained cells were sorted into de ned populations using uorescent-activated cell sorting (FACS, Beckman Coulter, CA, USA) to produce a pure population with the appropriate cell marker pro le.

Preparation of Conditioned Medium Derived from hUCMSCs
Six-well transwell co-culture system (Corning, USA) with 8 µm pore lters was used to co-culture RAW 264.7 cells and the hUCMSCs, where RAW-264.7 cells (1.0×10 6 ) and hUCMSCs (1.0×10 5 ) were seeded into the upper and lower units of the chamber, respectively. The control group was added only with RAW 264.7 cells (1.0×10 6 ) to the upper chamber. When the hUCMSCs reached 8090% con uency, they were washed three times with PBS and the medium was replaced with serum-free α-MEM medium. After further culturing for 48 h, the conditioned medium (2 D hUCMSCs©) was collected by centrifuge (2000 g for 10 min at 4 ℃) and stored at -80°C until usage. A similar method was used to obtain 3 D hUCMSCs©.

Cell Viability Assay
Cell viability was tested using MTT assay. Firstly, RAW 264.7 cells were seeded in a 96-well plate at a density of 1.0×10 4 cells per well. After incubation for 12 h to allow cells to attach on the wall, the growth medium was removed and the cells were washed 3 times with PBS. Then fresh growth medium and the conditioned medium were added. After further culturing for 24 h, MTT (5 mg/mL, 20 µL) was added and incubated for 4 h at 37 ℃ in a humidi ed atmosphere with 5% CO 2 . MTT was removed and dimethyl sulfoxide (150 µL) was added to dissolve the formazan crystals. The absorbance of colored solution was recorded at 570 nm to calculate the cell viability using a standard method described in the previous report [26].

Cell Apoptosis Assay
Cell apoptosis of RAW 264.7 cells was conducted by ow cytometry according to the manufacturer's instructions of Annexin V-FITC Apoptosis Detection Kit (Dojindo, Tokyo, Japan). RAW 264.7 cells were rstly seeded in a 6-well plate (5.0×10 5 cells per well). After 24 h incubation, the cells were harvested and washed twice with iced PBS. After removing supernatants, the cells were further treated with trypsin for the appropriate time and terminated with the culture medium. Cells were pelleted by centrifugation (1000 rpm, 3 min) and then washed twice. 10X diluted Annexin V Solution was added to re-suspend the cells (cell concentration of 1.0×10 6 cells/mL). Annexin V (5 µL), FITC conjugate and PI solution (5 µL) were respectively added to cell suspension (100 µL) and then the mixed solution was incubated for 15 min free of light. Further dilution using Annexin V solution (400 µL) was required before ow cytometry analysis (Accuri C6,BD Biosciences, USA).

Measurement of Nitric Oxide (NO) Production
RAW 264.7 cells were seeded in 96-well culture plates and incubated for 24 h, then the nitric oxide (NO) concentration in the supernatants of cultured RAW 264.7 cells was measured using a NO assay kit (Beyotime Institute of Biotechnology, China). The concentration of NO was obtained by reading the absorbance value at 540 nm, where sodium nitrite (NaNO 2 ) was used as an external standard.

Total RNA Extraction and Quantitative RT-PCR Analysis
Total RNA was extracted from cells using a Total RNA Kit (Invitrogen™ TRIzol™, Thermo, USA) according to the manufacturer's instructions. Then cDNA was synthesized with PrimeScript™ RT reagent Kit (Takara, Japan). Real-time PCR was performed with SYBR Premix Ex Taq™ II (Tli RNaseH Plus) Kit (TAKARA, Japan) in the CFX96 touch qPCR system (Bio-Rad, USA). The primer sense and antisense sequences are listed in Table 1. The 2 −ΔΔCq method was used to calculate the relative expression level for each gene [27] and normalized to the internal control Housekeeping β-actin gene.

Enzyme-linked Immunosorbent Assay (ELISA)
The concentration of IL-6 was measured via a mouse IL-6 ELISA Kit (R&D Systems, Minneapolis, MN, USA) in the supernatants obtained from RAW 264.7 cells. Absorbance at 450 nm was read on a microplate reader (BioTek, Winooski, Vermont, USA). All the experiments were performed in triplicate and the average value was used to minimize errors. Following the similar method, the concentrations of other cytokines were respectively evaluated.

Western Blotting Analysis
The cytoplasmic protein of RAW 264.7 cells was extracted using a Cytoplasmic Protein Extraction Kit (Beyotime, China). After the extraction, the protein concentrations were determined by Bicinchoninic acid Protein Assay. Then western blot analysis was carried out by applying polyclonal rabbit antibodies such as Erk MAPK, phospho-Erk MAPK, p38 MAPK, phospho-38 MAPK, JNK MAPK, phospho-JNK MAPK, STAT3, phospho-STAT3 and β-actin antibodies. All these antibodies were purchased from Cell Signaling Technology Inc (MA, USA). The secondary antibodies were bought from Signalway Antibody (Signalway Antibody LLC, USA). Protein bands were separated by SDS-polyacrylamide gel electrophoresis (SDS-Page), and the SDS-Page was electroblotted onto PVDF membranes (GE Healthcare, Chicago, IL, USA) and visualized by ECL detection kits (Beyotime, China). Western blot quanti cation was achieved with Image J software.

ALI Mice Model
Male BLAB-c mice with a bodyweight of ~ 20 g (x5 per group) were employed and the acute liver injury mice model was established by continuously CCl 4 injection via tail vein. The serum ALT and AST levels in the CCl 4 -induced mice were assessed to ensure the ALI mice model. After induction for 24 h, the hUCMSCs or the conditioned medium was transplanted into the ALI mice by tail vein. mRNA or cytokines in serum were respectively assessed according to the aforementioned methods.

Statistical Analysis
All data are representative experiments performed in three independent experiments. Data were expressed as mean value ± standard deviation (SD). The statistical signi cance of the mean values was compared by one-way ANOVA (analysis of variance) and Student's t-test using GraphPad Prism version 8.0.2 software (GraphPad Software Inc., San Diego, USA). *p < 0.05 vs. control group, **p < 0.01 vs. control group, ***p < 0.001 vs. control group; ##p < 0.01, ###p < 0.001 vs. control group.

D hUCMSCs Attenuates Acute Liver Injury
hUCMSCs were obtained from the human umbilical cord using the enzymatic digestion method [28]. The passage 3 (P3) hUCMSCs were selected for expansion culture due to the inheritance of characteristic stem cell morphology and excellent proliferation potential (Fig. S1). To better mimic the matrix environments in vivo, 3 D culture systems in vitro (Scheme 1) were explored to incubate the hUCMSCs on the commercialized microcarriers (Cytodex3, images of the scanning electron microscope, please see Fig.  S2) in 1.5 L bioreactors. As shown in Fig. 1, the P3 hUCMSCs cultured on the spherical cytodex3 (3 D hUCMSCs) exhibited fusiform or spindle shapes despite of the co-existence of spherical cells (day 1). On day 5, most of the expanded cells displayed homogeneous shapes and then radially arranged and aggregated in the microcarriers. Within the rst three days, the 3D hUCMSCs shared a similar proliferation rate with 2D hUCMSCs. For example, On the 3rd day, the cells proliferated to 8.23-fold in 2D and 9.35-fold in 3D Cytodex3 relative to the rst day. However, a signi cant difference was observed once extending to 5 days, when the 3 D hUCMSCs persisted proliferating on spherical Cytodex3 scaffolds to 25.89-fold increment while the cells cultured in 2D decreased to 7.86-fold (Fig. S3). Phenotypic analysis by standard ow cytometry (Fig. 1C) demonstrated that the ampli ed 3 D hUCMSCs were positive to the speci c markers (FITC CD90, APC CD105 and FITC CD73) [29] and negative for a series of surface antigens (FITC CD14, CD19, CD34, CD45 and HLA-DR). Multipotential differentiation evaluation showed that the cells could differentiate into osteocytes, adipocytes and cartilage. These results con rm that the 3 D hUCMSCs used in this study possessed stem cells' typical phenotypes and multipotential characteristics. ALI mice model [30] was successfully fabricated by consecutive injection with CCl 4 . As presented in Fig. 2, the ALT and AST levels were signi cantly boosted after 24 hours of incubation with CCL 4 . The time course of the hepatoprotective effect of the hUCMSCs against CCl 4 -induced acute liver injury (ALI) was assessed by evaluating the levels of serum ALT and AST in ALI mice. Compared with the control group, where equivalent PBS buffer as a vehicle, the ALI mice transplanted with the hUCMSCs displayed signi cantly attenuated serum ALT and AST levels. Especially, the mice transplanted with the 3D hUCMSCs had the lowest serum ALT and AST levels (176.54 and 833.88 U/mL). Additionally, histological examination revealed mild liver injury with cellular necrosis around the blood vessels as early as 12 h. More severe liver injury was observed at 24 h, seen as large areas of extensive cellular necrosis with loss of hepatic architecture and in ammatory cell in ltration around the blood vessels (Fig. 2B & Fig. S4). Interestingly, when transplanted with the 3 D hUCMSCs, a signi cantly-decreased liver injury were observed with a histological scores of 0.77, 2.52, 3.20, 2.02 and 1.79 respectively (0, 12, 24 36 and 48 h, all P < 0.05 or 0.01), superior than the control group (1.38, 4.59, 5.21, 4.43 and 3.51) and even the 2 D hUCMSCs (0.99, 3.20, 4.29, 3.38 and 2.60). The animal's body/liver weight showed time-dependent changes with the post-time of hUCMSCs treatment (Fig. 2E&2F). After 1 day post-transplantation, there was an apparent decrease in weight, while the 3 D hUCMSCs group exhibited a slight increase of body/liver weight compared to 2 D hUCMSCs groups.

3D hUCMSCs Altered Liver Responses to Damage through Paracrine Effects
Stem cell application in ALI treatment is attractive, and there are two possible mechanisms responsible for tissue repairing. One is that the liver was repaired through stem cell proliferation and transdifferentiated into hepatocytes. Recent studies further indicate a substantial role for paracrine effects in delivering overall bene ts. The activation of liver macrophages into the in ammatory (M1-like) and anti-in ammatory (M2-like) macrophages are usually associated with liver damage. To know the role of paracrine effects on liver repairing in ALI mice, a macrophage-like cell line RAW 264.7 cells was induced with LPS to provide the desired in ammation model in vitro. Further, it was then co-cultured with the hUCMSCs by transwell assay, where the hucMSCs were cultured in the lower compartment and the LPS induced RAW 264.7 cells in the upper compartment and were separated by a porous membrane. Conditioned medium derived from the coculture systems were collected by centrifugation, and the medium from 2 D hUCMSCs and 3 D hUCMSCs were denoted as 2 D hUCMSCs© and 3D hUCMSCs©, respectively. Interestingly, we found both 3D hUCMSCs© and 2D hUCMSCs© could relieve the liver injury with a histological score of 0.71, 2.81, 3.02, 2.23 and 1.59 vs. 1.25, 3.25, 4.15, 3.5 and 2.65, closer to that of the pertinent hUCMSCs ( Fig. 2D and Fig. S5). To learn more about the hUCMSCs on the in ammatory response, we rst evaluated the cytotoxicity of the medium in vitro. As shown in Fig. 3A, the viability of RAW 264.7 cells increased to 112% when treated with 3D hUCMSCs© (p < 0.001), indicating that the medium not only had no signi cant cytotoxicity to RAW 264.7 cells but also promoted the proliferation of macrophages. Furthermore, apoptosis was detected by ow cytometry (Fig. 3B&E), and the results showed that RAW 264.7 cells mainly displayed late-apoptosis. Surprisingly, the percentage of lateapoptosis cells in the 3D hUCMSCs© group dropped to 4.10%, lower than control group (7.39%) and even 2D hUCMSCs© (5.91%) group. In addition, the AnnexinV/PI double staining suggests that the 3D hUCMSCs© also increased the survival proportion of the RAW 264.7 cells, supporting the result of cell viability analysis. For example, the survival proportion increased to 91.4% from 89.3%. Such response elucidates that the 3D hUCMSCs© could help relieve the late apoptosis of macrophages and enhance the survival proportion, probably due to the cytoprotective effect [31]. Nitric oxide (NO) and inducible NO synthase (iNOS) is usually associated with in ammation, which generally participated in various diseases, including sepsis [32], cancer [33], neurodegeneration [34], and various types of pain [35]. In our experiments in vitro, the LPS stimulated RAW 264.7 cells strongly up-regulated the NO production (from 1.48 to 20.9 µM) [36] and signi cantly enhanced the iNOS mRNA expression level (rises to 7.42 times). However, the NO production and the iNOS mRNA expression were suppressed (Fig. 3C&D) once cocultured with the 3D hUCMSCs©.

D hUCMSCs Impeded the Expression of Pro-in ammatory Cytokines and Up-regulated the Anti-in ammatory Cytokines
To learn more about the therapeutic effect between hUCMSCs and recipient mice hosts, we measured the changes in the serum levels of cytokines and the related messenger RNA (mRNA) changes using cytokines arrays. Cytokine storms were suppressed in the both transplantation group at day 3. In the transplantation group, 9 cytokines (IL-1β, IL-6, IL-10, IL-17, IL-17F, IL-22, IL-28, TGFb1 and TNF-α) showed signi cant changes (Table 1), suggesting a disparity in the liver responses to damage between the two groups. From a functional perspective, most of the cytokines are related to immunoregulation, signal transduction and promotion/cell death. As a whole, pro-in ammatory cytokines (IL-6, IL-1β, IL-17, IL-22 and TNF-α) and apoptosis promoter IFNg also declined when transplanted 3D hUCMSCs compared to the control group (Fig. 3N). Differently, anti-in ammatory cytokines (IL-10) was up-regulated (Fig. 3M). For convenience, we also evaluate the cytokines change in the conditioned medium from the co-culture of RAW 264.7 cells and hUCMSCs. A similar cytokines change was observed, and the marker related with M 2 -like macrophages (Arg-1, CD206 and CD36) were up-regulated when transplanted with 3 D hUCMSCs ( Fig. 3F-3H). While the levels of IL-6, IL-1β and TNF-α, produced by the M 1 -like macrophages were decreased (Fig. 3I-3K). This result suggested that the 3 D hUCMSCs may be more favorable for M 1 /M 2 macrophage switch. Table 1 Changes of 18 cytokines levels in serum measured by cytokine arrays. Serum was collected on day 3 after the ALI mice were injected with PBS buffer or the hUCMSCs.
The value in brackets means cytokines level on day 5.   3D hUCMSCs Increase STAT3 Activation in RAW 264.7 Cells Stimulated with LPS IL-6 exerts its effects by the interaction with IL-6 receptors [37], which produces conformational changes in the gp130 subunit and activates the signal transducer and activator of transcription 3 (STAT3) via Janus kinases (JAKs). The dysregulation of STAT3 activation has been associated with the pathogenesis of in ammatory diseases [38]. Based on these facts, we hypothesize that hUCMSCs persistently regulated the expression of IL-6 in RAW 264.7 cells and probably stimulated the in ammatory responses of macrophages by inducing the phosphorylation of STAT3. In this study, we found decreased STAT3 protein kinases expression in the 3D hUCMSCs group, in which the declined IL-6 strongly inhibited the activation of STAT3 and the phosphorylation level (Fig. 4A).

D hUCMSCs Decrease ERK, p38, and JNK Activation in RAW264.7 Cells Stimulated with LPS
In the ALI model, the macrophages after interaction with the membrane receptors CD14 and Toll-like receptor (TLR)-4, lead to the generation of cytokines such as TNF-α, IL-1β and IL-6 [21]. The signaling pathways in the cell surface are usually mediated by the molecules of mitogen-activated protein kinase (MAPK) family, including ERK, p38 and JNK. To assess whether hUCMSCs interfere with the MAPK signaling cascade, we tested the phosphorylation level of ERK, p38 and JNK of the in ammatory macrophages. Results showed these phosphorylation levels were obviously lower in the presence of 3 D hUCMSCs than in the control sample. Moreover, a signi cant change was observed in p38 after introducing the hUCMSCs (Fig. 4B&C).

Triple Regulatory Roles of the 3 D hUCMSCs on Acute Liver Injury
Macrophages play pivotal roles in maintaining hepatic homeostasis and are closely associated with many liver diseases [21,22]. Differential expression is now used to identify the heterogeneous macrophage subsets responsible for ALI resolution [22][23][24][25]. Regulation of phenotypic switch from M 1 to M 2 -like macrophages provides a promising therapeutic intervention for treating the liver injury. Recently, mesenchymal stem cells (MSCs) have been considered the most important members of the stem cell family and found in many tissues, such as bone marrow, fat, and umbilical cord [16]. Above all, one of the most intriguing biological characteristics of MSCs is their immunomodulatory property, widely used for in ammatory diseases like ALI and COVID-19 diseases [18]. However, most MSCs are cultured in 2 D plates instead of the cell matrix-like 3 D environments, resulting in the degradation of some key functions and response sensitivity. Thus, the understanding of the immune response of MSCs in vivo was largely limited. Herein, we obtained the human umbilical cord mesenchymal stem cell cultured in 3 D microcarriers (3 D hUCMSCs) and provided the evidence that the 3 D hUCMSCs were more favorable to the MSCs therapy for ALI. Compared to 2 D hUCMSCs, upon liver injury, the ratio of M 1 /M 2 like macrophages were remarkably decreased after hUCMSCs transplantation. A similar result was achieved when injected with equal amounts of the conditioned medium [39]. All these indicated the secretions or exosomes of 3 D hUCMSCs would be more favorable for the effective promotion of M 1 /M 2 like macrophage switch with an ALI relieving cytokines pattern. Cytokines array or the Elisa data reveal the markedly-shrunk pro-in ammatory cytokines (such as IL-6, IL-1β, IL-17F, IL-22 and TNF-α) and enhanced IL-10 expression level once treated with 3 D hUCMSCs. Besides, in response to LPS stimulation, RAW264.7 produce NO by the enzyme iNOS [40]. The expression of iNOS is very important as a microbicide, antiviral, antiparasitic and antitumoral molecule. However, the aberrant upregulation of this enzyme is associated with various diseases, including autoimmune disease [41]. In our experiments, stimulation with LPS-stimulated upregulation of iNOS in RAW264.7 cells, and the conditioned medium prevented the upregulation of these transcripts. Reports have shown reciprocal interactions occur between hUCMSCs and macrophages after stimulation with LPS in vitro. Soluble factors derived from activated macrophages induce NO production by hUCMSCs in the presence of LPS. Moreover, 3 D hUCMSCs also express a lower level of IFNg to relieve the functional cell apoptosis in response to LPS/CCL 4 stimulation. The responsiveness of hUCMSCs to activating factors is expectable and unavoidable when using these cells for transplantation [42,43]. In short, the 3 D hUCMSCs demonstrate the advantages of immune-regulation and cell promotion/apoptosis administration, including the effective promotion of M 1 /M 2 macrophage switch by secreting an increased anti brogenic cytokines, ameliorating proliferation and downregulation of NO of macrophage (Fig. 4D). The scaffolds in 3 D culture probably provided an architectural skeletons in which cell-matrix, cell-cell, and growth factor interactions combine to create regenerative niches [44]. Such 3D systems have been shown to enhance the stem cell functionality and provide greater support for hepatocyte proliferation and functionality than routine 2D culture [45,46].

D hUCMSCs Ameliorate Acute Liver Injury by MAPKs and STATSignaling Pathways
MAPK activation such as p38, JNK, and ERK. MAPK controls the expression of proin ammatory cytokines such as IL-1β, IL-6 and TNF-α, among others [47]. Phosphorylation of p38, JNK, and ERK have also been reported upon stimulation of RAW264.7 cells with Corylin [48], and it could mediate the hyperphosphorylation of Tau molecules [49]. As expected, we could observe that stimulation of RAW264.7 cells with LPS stimulated the phosphorylation of ERK, p38 and JNK. Nevertheless, in the presence of 3 D hUCMSCs, the activation of these MAPKs was signi cantly prevented. In addition, binding of cytokines to their cognate receptors on the cell surface activates a family of cytoplasmic proteins, designated STATs (signal transducers and activators of transcription), through tyrosine phosphorylation by the receptor-associated Jak kinases. Also, as an important member of STAT family, STAT3 was activated by IL-6 family cytokines and subsequently facilitate tyrosine phosphorylated in response to a variety of stimulus [50]. In this study, we found that decreased STAT3 protein kinases expression in 3 D hUCMSCs, in which the activation of STAT3 was strongly inhibited. In addition, the STAT3 phosphorylation level was inhibited once treated with 3 D hUCMSCs in comparison with the control group (Fig. 4A). These ndings proved that the 3 D hUCMSCs and the conditioned medium could alleviate the ALI in mice by a combined strategy of immunoregulation, cell promotion and apoptosis.

Conclusion
In summary, we demonstrated that the 3 D hUCMSCs were more favorable for the ALI therapy compared to 2 D hUCMSCs. Triple regulatory roles of the 3 D hUCMSCs included the effective promotion of M 1 /M 2 macrophage switch on the secretion of increased anti brogenic cytokines, ameliorating proliferation and downregulation of NO of macrophage. The paracrine effect played a key role in ALI repairing. The conditioned medium of 3 D hUCMSCs acted as another potent secretory component containing speci c cytokines and small RNAs, which achieved a similar curative effect on ALI. Moreover, the effective suppression of the phosphorylation of extracellular signal-regulated kinases (ERK), c-Jun N-terminal kinases (JNK), the p38 and STAT3 indicated the mitogen-activated protein kinases (MAPKs) and STAT3 signaling pathways. The revelation of the superior moderating effect of 3 D hUCMSCs and the triple regulatory mechanism can open a new route for novel cell-free therapeutics to prevent immune system diseases, which will bring hope to patients with in ammatory and immunologically-mediated diseases.

Author Contributions
This study was designed and performed by JHZ &JQZ with help YP, YL, and AM. The paper was written and revised by JHZ, AM, MG and YZ. JHZ &JQZ contributed equally.

Data Availability Statement
The data that support the ndings of this study are available from the corresponding author upon reasonable request.
Ethics approval and consent to participate    cytokines, further damaging liver tissue. Implanted hUCMSCs respond to the signals and proliferate and transdifferentiate to repair the liver tissue. Through paracrine effects, the hUCMSCs also induce the expression of cytokines in the host. These cytokines are involved in immunoregulation and cells proliferation, and apoptosis. *p < 0.05 vs. control group, **p < 0.01 vs. control group, ***p < 0.001 vs. control group; ##p < 0.01, ###p < 0.001 vs. LPS group.

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