The Effect of miR122 and miR185 on Hepatic Differentiation of Human Umbilical Cord Mesenchymal Stem Cells

Background: Human umbilical cord mesenchymal stem cells (hUC-MSCs) have been considered as a promising cell source in liver diseases. miRNAs have been shown to play an important role in hepatic differentiation of hUC-MSCs. The study seeks to explore whether miR122 and miR185 could affect induction of hUC-MSCs into hepatic differentiation. Methods: miR122 and miR185 stable overexpression by hUC-MSCs were rstly created, then hUC-MSCs were cultured by hepatic differentiation conditional medium. After 28 days culture in hepatic inducing conditional medium, hepatic markers expressed by these cells were detected by qRT-PCR and western-blot. The cell functions were also evaluated by PAS staining and ICG phagocytosis. Results: Our results demonstrated that at the end of 28 days, hUC-MSCs overexpressing miR122 had increasing expression of hepatocyte markers including AFP, ALB, CK18, CK19 and HNF4α in both mRNA level and protein ecpression, while in the miR185 overexpression group, hUC-MSCs showed decreasing expression of hepatocyte markers. Moreover, there was also improvement of glycogen deposits as well as ICG phagocytosis ability in the hepatic inducing miR122 overexpression cells, while in the hepatic inducing miR185 overexpression group, hUC-MSCs showed decreasing glycogen deposits and ICG phagocytosis ability. Conclusions: to secrete Epstein-Barr virus interleukin-10 show enhanced immunosuppressive


Background
The evolution from hepatitis to cirrhosis and end-stage liver failure is a trilogy of developmental process which is di cult to control in clinical treatment. Liver transplantation and cell therapy is considered as a promising solution to end-stage liver failure treatment [1] , which can effectively improve the quality of life and prolong the life of patients. However, liver donor and seeding cell shortage is an obstacle to the above two techniques. In order to solve the problem, researchers are trying to explore more effective treatment methods, such as immunotherapy [2] , stem cell therapy [3] and arti cial liver support [4] . Induced hepatic differentiation of stem cells and the development of cell transplantation technology provide a potential prospect for liver replacement therapy.
MSC is one of the important members in the stem cell family. MSCs exist in almost all tissues and which can be isolated and successfully ampli ed in vitro from bone marrow [5] , adipose tissue [6] , umbilical cord [7] , placenta [8] , fetal liver, muscle tissue and lung tissue [9,10] . MSCs are a kind of cells with regenerative, secretory and immunomodulatory functions, which have the characteristics of self-replication, multidirectional differentiation potential, hematopoietic support and promotion of stem cell implantation, immune regulation. MSCs have the potential for cellular therapy and have been shown to be bene cial for the a variety of diseases [11] . Human umbilical cord mesenchymal stem cells (hUC-MSCs) have the following characteristics, self-renewal, replicate, low immunogenicity and multidirectional differentiation potential [12] . hUC-MSCs have been extensively studied in protective effect, immune regulation, disease prevention and disease treatment [13][14][15] . In recent years, hUC-MSCs have become an ideal source of cells for cell therapy and tissue engineering research [16] . miRNAs are involved in the regulation of multiple genes in embryonic development, such as differentiation, proliferation, and organ formation [17] . In addition, in almost all types of diseases, miRNAs expression levels are signi cantly different from those in normal tissues [18,19] . Changes in miRNAs network are closely related to liver diseases, including steatohepatitis [20] , viral hepatitis [21] , liver brosis [22] , cirrhosis [23] and hepatocellular carcinoma [24,25] . miRNAs play an important role in improving the e ciency of liver differentiation [26] . miR122 is the most common and highly abundant miRNA in adult liver [27] . miR122 has been shown to be an important host factor and antiviral target for hepatitis c virus infection [24] . Lou et al [28] . found that MSCs modi ed with miR122 were more conducive to the treatment of liver brosis. Reconstruction of miR122 expression is a promising therapeutic strategy in liver tumor diseases that simultaneously reduces tumor invasiveness and disease recurrence [29] . Another miRNA associated with liver disease is miR185. In the study of Zhou et al. [30] , miR185 inhibits the activation of hepatic stellate cells by inhibiting the Ras homolog enriched in brain (RHEB) and rapamycin-insensitive companion of mammalian target of rapamycin (RICTOR) pathways, thus preventing the occurrence of liver brosis. miR185 has been found to be involved in the differentiation of myelogenic inhibitory cells [31] . Inhibition of miR185 can promotes adipogenic differentiation [32] and osteogenic differentiation [33] , but no study has been conducted on hepatic differentiation.
The purpose of this study is to explore the effect of miR122 and miR185 overexpression on the hepatic inducing differentiation of hUC-MSCs.

hUC-MSCs culture
Umbilical cord were obtained from health puerperant who deliveried in our hospital and consented for the donation. The donation procedure and informed consent were approved by the ethics committee. The umbilical cords were rstly disinfected by immersion in 75% alcohol and rinsed in PBS to remove as much residual blood as possible. Then, the umbilical cords were cut into 3-5 cm segments and removed the arteries and veins. Wharton's jelly in tissues was cut into small pieces of 2 mm and evenly spread in a 6-well cell culture plates, and cultured in cell incubator (37℃, 5% CO 2) for one week in Dulbecco's modi ed Eagle's medium (DMEM). The tissue blocks were removed after one week and the medium was replaced every three days.

Flow cytometry
Cells were treated with 0.5% trypsin, then being pipetted repeatedly to separate them into single cell suspension, and then being transfered into the ow tubes. After wishing with PBS, cells were incubated respectively with antibodies as isoform control IgG-FITC/PE, CD34-FITC, CD45-FITC, CD90-PE, CD105-PE for 30 min. For Oct-4 analysis, cells were incubated and xed in 1% paraformaldehyde at 4℃ for 20 min after well suspension, Afterwards, they were centrifuged at 350 g for 10 min and washed twice with PBS. The cells were then re-suspended with PBS and labeled respectively with Oct-4-PE antibody and IgG-PE for 30 min. At the end of incubation, cells were washed thoroughly with PBS buffer and then tested by ow cytometry. 2.2 miRNA overexpression by transfection and their in uence on cell proliferation hUC-MSCs were plated onto 6-well cell culture plates at a number of 5 × 10 4 cells followed by the 50 nM transfection concentration (The experiment was conducted according to the protocol suggested by RNA transfection kit (Ribobio, Guangzhou)). The experiment was divided into ve groups: BC (blank control) group, NC (negative control) group (transfected with NC mimic), miR122 group (transfected with 50 nM miR122), miR185 group (transfected with 50 nM miR185), and miRAB group (transfected with both miR122 and miR185 50 nM each). After transfection of miRNA, hUC-MSCs were cultured in DMEM and their expression of miR122 and miR185 in all groups were detected at day 1, 7, 14, 21 and 28 by qRT-PCR. For qPCR, total RNA, including miRNAs, was rstly extracted using Eastep Super total RNA extraction kit (Promega). Then, miRNA expression was performed using the miRNA qRT-PCR Starter Kit (Ribobio, Guangzhou) according to the manufacturer's instructions. Finally, miRNA relative expression levels were detected and normalized by housekeeping gene U6. Cell proliferations after transfection were also analyzed by CCK-8 method. And all of the above experiments were repeated at least four times.
hUC-MSCs were divided into six groups: BC group which are cells cultured in normal DMEM, BCin group which are normal hUC-MSCs cultured with hepatic inducing procedure, NCin group which are hUC-MSCs transferred with non-speci c miRNA mimic and treated with hepatic inducing conditional medium, miR122in group which are hUC-MSCs overexpress miRNA122 and treated with hepatic inducing conditional medium, miR185in group which are hUC-MSCs overexpress miRNA185 and treated with hepatic inducing conditional medium, and miRABin group which are hUC-MSCs overexpress both miRNA122 and miRNA 185, and then treated with hepatic inducing conditional medium. Each experiment was repeated at least three times. After 28 days of culture, the expression level of ALB (albumin), AFP (alpha fetoprotein), CK18 (cytokeratin 18), CK19 (cytokeratin 19), and HNF4α (liver cell nucleus factor 4 α) in every group was detected by qRT-PCR for mRNA expression and by western-blot for protein expression. Cell function of glycogen storage and endocytosis were also tested by periodic acid-Schiff (PAS) staining and dilute indocyanine green (ICG).

RNA extraction and quantitative real time PCR (qRT-PCR)
Total RNA was extracted using Eastep Super total RNA extraction kit, Iscript cDNA sythesis kit (Bio-rad, USA) and PowerUp SYBR Green Masrer Mix (Thermo Fishers scienti c, USA) were used for reverse transcription of cDNA ampli cation. Three-step procedure PCR was performed with the conditions listed in Table 1 and the primers used in this experiment was listed in Table 2. GAPDH was used to normalize the target gene expression as a reference gene.  The cells were collected at 28th day of cell culture in each group, and the total proteins of cells in each group were extracted by RIPA lysate. Western-blot was used to detect the expression of the hepatocyte speci c proteins, ALB, AFP, CK18, CK19, and HNF4α (Proteintech, USA) in each group and the relative quanti cation was calculated referring to β-actin.

Cell function tests
Glycogen storage of hepatocyte like cells was evaluated with periodic acid-Schiff (PAS) staining. After 28 days of cell culture, the medium was washed with PBS twice, the cells were xed with 4% neutral formaldehyde for 30 min, then iodic acid reagent were added and incubated for 10 min, then xed cells were rinsed with ddH 2 O for 3 times. Afterwards, Schiff's reagent (Sigma Aldrich, USA) were added and incubated with cells for 15 min. Cells were rinsed with ddH 2 O again and stained in hematoxylin reagent for 90 seconds, rinsed once more with ddH2O, dried naturally at room temperature, and observed. 0.5 mg/mL dilute indocyanine green (ICG) solution was prepared according to the production protocol in serum-free DMEM. After 28 days of cell culture, cells in each group were washed with PBS twice, 0.5 mg/mL ICG solution was added to the cell culture plates and cultured in incubator at 37℃ for an hour. The ICG solution was removed, and cells were rinsed with PBS 3 times. Normal DMEM was added and ICG cells were observed the phagocytosis in each group. The culture medium was changed every 2 hours, and ICG in each group was observed after 6 hours.

Statistical analysis
Real-time PCR and western-blot data were expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) and post-hoc analysis were used to compare the differences between groups at the speci ed time points. P ≤ 0.05 was considered statistically signi cant. All statistical calculations were performed using the statistical package for social sciences (SPSS) for windows, version 16.

Identi cation of hUC-MSCs
Cells isolated from Wharton's jelly of umbilical cord showed broblast like growth pattern and they were observed a uniform shape of long spindles, closely arranged in irregular shape and showed a vortex growth under microscopy (Fig. 1). Flow cytometry results showed a high positive expression of CD90 (99.16% ± 0.85%) and CD105 (99.26% ± 0.78%), but no expression of CD34 (0.59% ± 0.16%) and CD45 (1.50% ± 2.04%) by these cells and an average of 66.57% ± 4.61% cells expressed positively Oct-4, an embryonic stem cell marker (Fig. 1). After 28 days of osteogenic induction, calcium salt deposition in hUC-MSCs were detected by Alizarin red staining which became dense red nodules (Fig. 1). After 28 days of adipogenic induction, small lipid droplets in hUC-MSCs were observed after red oil O staining (Fig. 1).
The cells obtained from umbilical cord expressing speci c MSC markers and had potential to be differentiated into osteoblast and adipocyte.

hUC-MSCs transfection with miRNA
hUC-MSCs were transfected by the 50 nM miRNA during 24 hours according to the protocol suggested by RNA transfection kit (Ribobio, Guangzhou). There was a plateau of miR122 and miR185 overexpression in all three miR122, miR185 and miRAB groups from day 1 to day 28 after transfection (Table 3, Table 4). Transfection of miRNA NC mimic had no effect on the expression level of miR122 and miR185 by hUC-MSC (Fig. 2). The number of cells were counted and used to draw cell growth curve which showed that the cells proliferated stably and grew well in all 5 groups (Fig. 2), which means the overexpression of miR122 and miR185 didn't have any impact on cell proliferation.

The effect of miR122 and miR185 overexpression on hepatic differentiation of hUC-MSCs
To study the effects of miR122 and miR185 on hUC-MSCs hepatic differentiation, hUC-MSCs rstly overexpressed miR122, miR185 or both by transfection, then were induced into hepatic differentiation using conditional culture medium. After 28 days of induction, q-RT-PCR results showed that the ALB expression level in miR122 group was the highest (1.72 ± 0.43) and the lowest in miR185 induction group (0.47 ± 0.17). The expression level of miR185 and the miRAB induction group (0.36 ± 0.13) was signi cantly lower than that in the BC induction group (1.10 ± 0.19) and NC induction group (1.41 ± 0.14). Although the ALB mRNA expression was the highest in miR122 group, there was no statistical signi cance between miR122 induction group and BC induction group (P 0.05). As for the AFP, the expression level is the lowest in miR185 induction group (0.64 ± 0.27), which was signi cantly lower than that in the BC induction group (2.29 ± 0.89). No statistical signi cance was observed among miR122 induction group (2.04 ± 0.96), BC induction group (2.29 ± 0.89) and NC induction group (2.39 ± 0.94) in AFP expression level (P 0.05). The expression of CK18 and CK19 mRNA were the highest in miR122 induction group (1.93 ± 0.21 and 1.50 ± 0.27). The same is true for HNF4α mRNA expression with an expression rate of 19.03 ± 2.13. The expressions of CK18 (0.21 ± 0.07) and CK19 (0.29 ± 0.28) were decreased in the miR185 induction group, which were signi cantly lower (P 0.05) than that in the BC induction group (0.84 ± 0.14 and 1.13 ± 0.25). And the expressions of CK19 (0.23 ± 0.15) and HNF4α(0.58 ± 0.33) were decreased in the miRAB induction group, which were signi cantly lower (P 0.05) than that in the BC induction group (1.13 ± 0.25 and 9.36 ± 1.81) (Fig. 3).
Cell function test was performed after 28 days of hepatic differentiation induction. The PAS staining showed no glycogen deposition in BC group. Comparing the ve induction groups, we found that the amount of glycogen deposition in BC induction groupand NC induction induction group is almost the same, while this amount in miR122 induction group is signi cantly higher than that of BCinduction group and NC induction induction group. The glycogen content in miR185 induction group was signi cantly lower than that of the BCinduction group and the NC induction group. The glycogen concentration of the miRAB induction group was similar to that of BCinduction group and NC induction group (Fig. 5).
In ICG phagocytosis and exocytosis experiment, after 1 hour incubation with ICG, no ICG was observed in BC group cells and the quantity of ICG in BC induction groupand NC induction group were similar, while a large number of ICG accumulation was found in the miR122 induction group. ICG accumulation was signi cantly lower in miR185 induction group cells than that of BC and NC induction group. ICG in miRAB induction group cells was slightly less than that of BC and NC induction group, but a little bit more than that in miR185 induction group. ICG was seen to be almost completely exocytosis after 6 hours of culture in the ICG-free medium (Fig. 6).
The result of hUC-MSCs miRNA transfection showed no effects on cell proliferation in all groups. qRT-PCR result showed that NC mimic had no effect on the expression of miR122 and miR185 of hUC-MSCs. The corresponding miRNA levels of miR122 group, miR185 group and miRAB group at days1, 7, 14, 21 and 28 were stable and sustained at a high level, which are satis ed with the requirement of subsequent experiments. The stem cells transfected in our experiment have a strong self-ampli cation ability and can withstand a long screening process.
miR122 is a hepatic speci c microRNA that plays an important role in liver self-regulation and hepatic differentiation [40] . Studies have shown that miR122 promotes hepatic differentiation while miR122 in MSCs gradually increases during the process of induction. miR122 has been shown to promote hepatic differentiation by down-regulating the expression of cellular interstitial marker genes SOX11 and VIM [26] . Down-regulation of miR122 transcription inhibits the hepatocyte differentiation potential of liver progenitor cells [41] . Chien et al. found that miR122 can shorten the culture time of iPSC to hepatocyte differentiation. miR122-iPSC-Heps may be a viable cell source, which providing an effective alternative strategy for liver regeneration in acute liver failure [40] . In our study, it was found that the expressions of ALB, AFP, CK18, CK19 and HNF4α in miR122 induction group were increased different degree compared with BC induction group. miR122 induction group also had stronger glycogen storage and ICG phagocytosis and exocytosis function than that of other groups. These two functions are two of the most important hepatocytic functions. Therefore, we speculate that miR122 had a certain promotion effect on hepatic differentiation of hUC-MSCs. We also transfected miR185 in hUC-MSCs because it is reported associating with cell induced differentiation and has been found to be involved in the differentiation of myelogenic inhibitory cells [31] . It has been shown that overexpression of miR185 can lead to a signi cant reduction in lipid accumulation, while the low level of miR185 can promote adipogenic differentiation [32] . Studies have demonstrated that the inhibition of miR185 could promote bone formation and osteogenic differentiation [33,42,43] . In this study, it was found that the expressions of hepatic relative genes and proteins by cells overexpression of miR185 induction group were decreased compared with normal hUC-MSCs after hepatic induction differentiation, which means, differentiation of hUC-MSCs was delayed by overexpression of miR185. Glycogen storage and ICG phagocytosis and exocytosis function were inhibited in miR185 overexpression cells comparing with normal hUC-MSCs differentiation induction group. Therefore, we hypothesize that miR185 could inhibit the hepatic differentiation of hUC-MSCs. There was no signi cant change in hepatic speci c protein expression and functions in cells overexpressed both miR122 and miR185 induction group compared with normal hUC-MSCs differentiation induction group. We thus surmise that the effect of miR122 and miR185 are contradicting and they can be neutralized by each other. The limitation of this study is that we have not been able to conduct animal experiments to further prove our ndings.

Conclusions
In conclusion of the above, overexpression of miR122 can promote hUC-MSCs differentiating into hepatocytes. While, overexpression of miR185 inhibites hepatic differentiation of hUC-MSCs. It seems that miR122 and miR185 have an antagonistic effect on hUC-MSCs hepatic differentiation.
Overexpression of certain kind of miRNA in cells by transfection or other gene modi cation skills could be an effective way to modulate stem cell fate. Availability of data and materials:

Abbreviations
The datasets generated and/or analysed during the current study are not publicly available due [REASON WHY DATA ARE NOT PUBLIC] but are available from the corresponding author on reasonable request.

Competing interests:
The Authors declared that they have no con ict of interests.  A and B represent the relative expression of miR122 and miR185, respectively, in NC group, miR122 group and miRAB group (n=3). C: The proliferation of hUC-MSCs after the over-expression of miRNA (n=3) Figure 3 mRNA expression of hepatocyte-speci c protein by hUC-MSC analyzed by qRT-PCR (n=6). A, B, C, D and E represent the relative mRNA expression of ALB, AFP, CK18, CK19 and HNF4α in each group, respectively (n=6). "*" represents P 0.05 compared with the BC group,, "**" and "***" represent P 0.025 and P 0.001 compared with BC group, respectively. The semiquantitative analysis was performed with help of ABI 7500 and statistic alanalysis was performed with GraphPad prism 6.0 Software.

Figure 4
Expression of hepatocyte-speci c proteins detected by western-blot (n=6). A, B, C, D and E represent the proteins expression of ALB, AFP, CK18, CK19 and HNF4α in each group, respectively. "*" represents P 0.05 compared with the BC group,, "**" and "***" represent P 0.025 and P 0.001 compared with BC group, respectively. F Chemiluminescence imaging of each experimental group. The imaging analysis software (Bio-Rad, USA) was used for analyzing the images and statistic alanalysis was performed with GraphPad prism 6.0 Software.