H19 Overexpression Improved Efficacy of Mesenchymal Stem Cells in Ulcerative Colitis by Modulating the miR-141/ICAM-1 and miR-139/CXCR4 Axes

Overexpression of C-X-C motif chemokine receptor 4 (CXCR4) and intercellular cell adhesion molecule-1 (ICAM-1) may promote homing of mesenchymal stem cells (MSC). In this study, we treated ulcerative colitis animals with MSC preconditioned with or without H19 and compared the therapeutic effect of MSC and MSC-H19. We evaluated the regulatory relationship of H19 vs. miR-141/miR-139 and miR-141/miR-139 vs. ICAM-1/CXCR4. We established an ulcerative colitis mouse model to assess the effect of MSC and MSC-H19. H19 was found to bind to miR-141 and miR-139. The activity of H19 was strongly decreased in cells c-transfected with miR-141/miR-139 and WT H19. ICAM-1 was confirmed to be targeted by miR-141 and CXCR4 was targeted by miR-139. The H19 expression showed a negative regulatory relationship with the miR-141 and miR-139 expression but a positive regulatory relationship with the ICAM-1 and CXCR4 expression. In summary, the overexpression of H19 in MSC downregulated miR-139 and miR-141, thus increasing the activity of their targets ICAM-1 and CXCR4, respectively, to exhibit therapeutic effects in ulcerative colitis.


Introduction
As a condition of the gastrointestinal tract system, inflammatory bowel disease (IBD) manifests as ulcerative colitis, Crohn's disease, and persistent inflammation in the stomach. The etiology of IBD remains unclear; although, studies on IBD pathogenesis have shown that the breakdown of the immune system in the digestive tract, including an imbalance in intestinal microbiota, is crucial for the onset of this condition [1,2].
Mesenchymal stem cells (MSC) are present in many body organs, such as dental pulp, bone marrow, fat, and muscles; furthermore, MSC are linked to the microvasculature in the entire body as pericytes. These cells may differentiate into different types of mesenchymal cells [3]. MSC produce cytokines and alter the microenvironment required for cell regrowth. MSC possess strong immunomodulatory properties by reducing the proliferation of inflammatory cells as well as the synthesis of cytokines [4]. MSC are presently utilized for the treatment of inflammatory conditions, such as myocarditis, sclerosis, and IBD [5,6]. While the functions of MSC remain elusive, MSC are used for IBD therapy and have displayed appealing outcomes in animals [7][8][9].
Noncoding RNA (ncRNA) has been shown to possess significant properties in the regulation of gene functions [10]. NcRNAs have been shown to become significant regulatory factors in the onset, progression, and prognosis of medical conditions [11]. NcRNAs include the extensively studied miRNAs, as well as the lncRNAs, which is one type of ncRNA recently recognized for their importance. However, although miRNAs have been intensively examined for their functionalities, lncRNAs also comprise a new and possibly crucial class of ncRNAs [12]. LncRNA H19 is located close to the telomeric region of chromosome 11p15.5 [13], and it was reported to be highly expressed in the developing embryo while being significantly downregulated in neonates [14]. Moreover, H19 has been described as an oncogene in several diseases such as bladder cancer, colorectal cancer, or breast cancer [15][16][17]. Apart from this, H19 silences target genes by upregulating the miR-139 expression. In one study, H19 was significantly upregulated under hypoxic conditions, indicating that H19 participates in cell injury induced by hypoxia [18]. And a previous study showed the that miR-139 expression was downregulated in nerve cells and rodent brains under hypoxia [19]. Moreover, H19 was demonstrated to compete with miR-141 and influence the expression of miR-141, as well as the ZEB1 which is the target gene of miR-141, in gastric cancer [20]. By making use of a bioinformatic approach, H19 was shown to bind to miR-139 and miR-141. Therefore, it was supposed that there may be interactions between miR-139/miR-141 and H19 [20,21].
Bioinformatic results and dual-luciferase assays have suggested that CXCR4 is a direct miR-139 target. Functionally, the suppressive effects of miR-139 on the proliferation of T-ALL cells were reversed by CXCR4 transfection, while the miR-139 overexpression substantially lowered the malignancy of T-ALL cells, and people with very high expression of miR-139 and reduced CXCR4 expression showed greater five-year survival than people with reduced miR-139 expression [22]. Also, the homing and migratory capacity of umbilical cord-derived MSC via regulating CXCR4 could lead to the boosting of protection against liver ischemia/ reperfusion injury by rapamycin-preconditioned umbilical cord-derived MSC [23], and by promoting the expression of CXCR4, HIV glycoprotein gp120 has also been found to enhance the migration of MSC [24]. Also, bioinformatic results and dual-luciferase assays have suggested ICAM-1 as a direct target of miR-141. ICAM-1 has been shown to interact with target cells during cytotoxic T cell and natural killer cell-mediated damage [25], and ICAM-1 has also been shown to prevent and reverse dextran sulfate sodiuminduced colitis in mice [26].
Moreover, It has been shown that the overexpression of CXCR4 and ICAM-1 may promote the homing of MSC in the treatment of IBD [27,28]. Furthermore, we suspected that the overexpression of H19 may enhance the expression of CXCR4 and ICAM-1, possibly by the findings that H19 could sponge the expression of miR-139 and miR-141 expression [21]. In this study, we treated ulcerative colitis animals with MSC preconditioned with or without H19 and compared the therapeutic effect of MSC and MSC-H19 in the treatment of ulcerative colitis.

Materials and Methods
2.1. Mice. All female C57BL/6 mice (approximately 8 weeks, body weight approximately 20 g) were obtained from the institutional animal center. Mice were fed a pellet-based diet and drinking water. The study protocol was approved by the institutional ethical committee.

Animal Model.
Colitis was triggered by administration of 36-50 kDa DSS (MP Biomedical, Santa Ana, CA). The mice were fed 5% DSS in sterile water for 1 week and then fed regular water for another 14 days (three cycles). The animals were evaluated every day to record food uptake and body weight (body weight change was calculated using the weight on day 0 as a reference). Bodyweight data were collected every 3 days to calculate body weight change (%) ðweight at day X/weight at day 0Þ × 100), and stool consistency and rectal bleeding occurrence were recorded. Thirty-two C57BL/6 mice were randomly divided into four groups (eight animals/group): a SHAM group (mice treated with PBS), animal model group (mice treated with DSS), and two treatment groups (mice treated with DSS + MSC and mice treated with DSS + MSC-H19). For the two treatment groups, mice were treated with MSC or MSC-H19 (3 × 10 6 cells/mL) via tail vein injection on days 4, 14, and 24. The animal model establishment was accomplished following previous published protocols [29]. Around 5% of the mice died during our experiment. At the end of the experiments (12 weeks in total), the animals in each group were killed to isolate colons for functional analysis.

Luciferase
Assay. According to previous publications, H19 was predicted to bind to miR-141 [20] and miR-139 [21]. To confirm the binding of H19 to miR-141 and miR-139, a luciferase assay was performed by cotransfecting cells with scrambled miRNA control + WT H19, miR-141/miR-139 + WT H19, scrambled miRNA control + MUT H19, and miR-141/miR-139 + MUT H19. In brief, fragments of miR-141 and miR-139 containing miR-141 and miR-139 binding sites were cloned into the pcDNA vector (Promega, Madison, WI) downstream of the luciferase gene to generate wild-type H19 plasmids for miR-141 and miR-139 binding, respectively. At the same time, site-directed mutagenesis was performed at the miR-141 and miR-139 binding sites of H19 to generate mutant H19 plasmids. In the next step, MSC were cotransfected with wild-type/mutant H19 and miR-141 or miR-139 using Lipofectamine 3000. The luciferase activity was tested 48 h later using a dual-Luciferase assay (Promega, Madison, WI). Moreover, by searching public databases (http:// mirtarbase.mbc.nctu.edu.tw, http://www.mirdb.org), ICAM1 and CXCR4 were shown to, respectively, bind to miR-141 and miR-139. In this study, fragments of ICAM-1 and CXCR4 containing binding sites for miR-141 and miR-139, respectively, were cloned into pcDNA vectors downstream of the luciferase gene to generate wild-type ICAM-1 and CXCR4 plasmids for miR-141 and miR-139 binding, respectively. At the same time, site-directed mutagenesis was performed at the miR-141 and miR-139 binding sites of ICAM-1 and CXCR4 to generate mutant-type ICAM-1 and CXCR4 plasmids for miR-141 and miR-139 binding, respectively. In the next step, MSC were cotransfected with wild-type/mutanttype ICAM-1 and CXCR4 plasmids and miR-141 or miR-139 using Lipofectamine 3000. Universally, transfection was performed following the general protocol: the concentration of plasmid-based constructs was 0.2 μg, and the concentration of miRNA mimics was 30 nM. The constructs and mimics were separately dissolved in PBS and added to the medium in a dish before shaking to mix. The luciferase activity of the cells was assayed 48 h posttransfection by using a Dual-Luciferase Reporter assay kit.
2.8. Western Blotting. Western blotting was done to assay the expression of ICAM-1 and CXCR4 proteins in samples using a conventional procedure. Anti-ICAM-1 and anti-CXCR4 primary antibodies were purchased from Abcam (Cambridge, MA).
2.9. MTT Assay. MSC proliferation was measured using an MTT assay kit (Beyotime Medical, Wuhan, China). Optical density was evaluated at 570 nm wavelength on a microplate reader (Bio-Rad 550, Bio-Rad Laboratories, Hercules, CA) in strict accordance with the manual provided by the machine manufacturer.
2.12. Histological Analysis. HE stained transverse sections from colon samples were sliced into 4 μm sections, fixed at 4°C overnight in 4% (v/v) formaldehyde, and embedded in paraffin. The extent of colonic inflammatory damage was assessed blindly according to previous publications [30].
2.13. Serum Analysis. The levels of peripheral blood biomarkers, such as C-reactive protein (CRP), were evaluated with a biochemistry analyzer (TBA-120FR; Toshiba Medical System Co., Tochigi, Japan) according to the manufacturer's instructions.

Statistical Analysis
One-way ANOVA (post hoc test: Tukey's test) and Student's t-tests were performed using SPSS 21.0 when appropriate to compare the results for different groups. A P value of 0.05 was taken to judge statistical significance.  Figure 1: Luciferase assay of the H19-miR-141-ICAM-1 and H19-miR-139-CXCR4 axis. (a) Binding region between H19 and miR-141 was predicted, and the luciferase activity was the lowest in cells cotransfected with miR-141 and wild-type H19 ( * P value <0.05 vs. control + WT H19). (b) Binding region between miR-141 and ICAM-1 was predicted, and the luciferase activity was the lowest in cells cotransfected with miR-141 and wild-type ICAM-1 ( * P value <0.05 vs. control + WT ICAM1). (c) Binding region between H19 and miR-139 was predicted, and the luciferase activity was the lowest in cells cotransfected with miR-139 and wild-type H19 ( * P value <0.05 vs. control + WT ICAM1). (d) Binding region between miR-139 and CRCX4 was predicted, and the luciferase activity was the lowest in cells cotransfected with miR-139 and wild-type CRCX4 ( * P value <0.05 vs. control + WT CXCR4). 4 Disease Markers

The Impacts of H19 on Cell Proliferation and Migration.
MSC were transfected with pcDNA, scramble control, pcDNA-H19, or H19 siRNA, and MTT and transwell assays were performed to detect cell proliferation and migration. MTT results showed that the proliferation rate of cells transfected with pcDNA-H19 was significantly higher than that of cells in other groups, while the proliferation rate of cells transfected with H19 siRNA was the lowest among all the cell groups (Figure 4(a)). Transwell assays revealed that higher expression of H19 had positive effects on cell migration, while downregulation of H19 inhibited cell migration (Figure 4(b)).

Impact of MSC and H19 on Ulcerative Colitis Mice.
Mice were randomly divided into 4 groups: SHAM, DSS, DSS + MSC ,and DSS + MSC-H19, and body weight was measured every three days. Accordingly, at day 90, mice treated with DSS showed significant weight loss ( Figure 5(a)) and shortened colon length ( Figure 5(b)), and the effects of DSS were attenuated by MSC treatment and further alleviated by MSC-H19 treatment (Figures 5(a) and 5(b)). Additionally, the level of the peripheral blood biomarker CRP was the lowest in the SHAM group and highest in the DSS group ( Figure 5(c)). As shown in Figure 5(d), HE staining indicated that the level of inflammation was evidently increased by DSS, and MSC treatment and MSC-H19 treatment partially restored the increased histological inflammatory score in DSS mice. Then, the levels of inflammatory cytokines (TNF-α, IL-1, IL-6, IL-8) were tested in the four groups. As expected, DSS  Disease Markers treatment induced an inflammatory response, manifested by increased expression of TNF-α (Figure 6(a)), IL-1 ( Figure 6(b)), IL-6 ( Figure 6(c)), and IL-8 ( Figure 6(d)). MSC injection reduced the effects of DSS, and MSC-H19 infusion further reduced the level of inflammatory responses.

Functions of MSC and H19 in Ulcerative Colitis Mice.
We further tested the targets of H19 (i.e., miR-141 and miR-139) and their downstream genes (i.e., ICAM-1 and CXCR4) in the four mouse groups. As shown in Figure 7, administration of DSS evidently increased miR-141 (Figure 7(a)) and miR-139 (Figure 7(c)) expression while decreasing ICAM-1 mRNA (Figure 7(b)) and CXCR4 mRNA (Figure 7(d)) expression, which was also confirmed by RT-qPCR. MSC injection weakened the effects of DSS, and MSC-H19 further abated the effects of DSS. ICAM-1 and CXCR4 protein expression was measured via Western blotting. As shown in Figures 7(e) and 7(f), the protein expression of ICAM-1 and CXCR4 was significantly decreased by DSS, while MSC infusion elevated the protein expression of CXCR4 and ICAM-1, and MSC-H19 had a more prominent effect in raising the ICAM-1 and CXCR4 expression (Figures 7(e) and 7(f)). The protein expression of CXCR4 (Figure 8(b)) and ICAM-1 (Figure 8(a)) was further examined via IHC, and the results were consistent with the WB results.

Discussion
IBD, such as Crohn's disease and ulcerative colitis, is persistent and relapsing conditions with chronic inflammation of the intestinal tract and injuries to the mucus [31]. Mucosal recovery is linked to improved and lasting medical efficacy. Thus, mucosal recovery has been a major target for IBD treatment [32]. The absence of mucosal recuperation generally leads to difficulties in patient recovery and complications, such as blood loss, fistulas, and perforation, which demand a hospital stay and surgery. Less than 50% of IBD patients can be managed with standard treatments, such as immunomodulators, corticosteroids, and biologic drugs [33]. Standard treatments used to treat IBD rely on suppression of the immune system; for that reason, new techniques are required to enhance mucosal regrowth [34]. While intravenous transplants of stem cells are frequently used in common treatments, the distribution of MSC to wounded tissues is still extremely complicated [35]. Initiatives are being made to promote the delivery of MSC, such as by using nanoparticles [36]. In addition, chemokine CXC receptor (CXCR), vascular cell adhesion molecule 1 (VCAM1), and E-selectin ligands have been assessed to cover MSC surfaces to aid their targeted delivery to wounded sites, improving the therapeutic results of IBD therapy [35,37]. In the present study, mice were randomly divided into 4 groups: SHAM, DSS, DSS + MSC, and DSS + MSC-H19. Mice treated with DSS showed significant weight loss and shortened colon length. The effects of DSS were obstructed by MSC treatment and were further alleviated by MSC-H19 treatment. In addition, the level of inflammatory cytokines was tested in the four groups. DSS treatment induced an inflammatory response, while MSC injection reduced the effects of DSS, and MSC-H19 infusion further reduced the inflammatory response. Furthermore, the levels of targets of H19 (miR-141 and miR-139) and their downstream genes (ICAM-1 and CXCR4) were tested in the four mouse groups. It was found that the expression of miR-141 and miR-139 was evidently increased while the expression of ICAM-1 and CXCR4 was decreased in DSS animals. MSC injection weakened the effects of DSS, and MSC-H19 injection further abated the effects of DSS.
Recently, the level of H19 was discovered to become substantially enhanced in osteoarthritis tissues, indicating a prospective role of H19 in the progression of inflammatory conditions [38]. A previous report on the role of H19 in protection of the intestinal barrier in sepsis seems to conflict with our results [39]. Such a discrepancy could be attributed to the different disease backgrounds. The basic disease investigated in this study is ulcerative colitis, and the basic disease examined in the reference is sepsis. Recently, scientists utilized bioinformatic and luciferase assay to verify the binding between miR-141 and H19. As assumed, it was found that miR-141 could bind complementarily to H19, inhibiting translation of an RLuc-H19 gene. It was likewise revealed that miR-141 acts as a repressor of ICAM-1 expression, and the miR-141 overexpression decreases the ICAM-1 expression induced by ischemia to alleviate reperfusion injury. Therefore, miR-141 might act as a beneficial factor in the treatment of heart disease [40]. Here, ICAM-1 is predicted to be a downstream gene of miR-141, and CXCR4 is a downstream target of miR-139. The activity of miR-141/miR-139 was decreased in cells cotransfected with miR-141/miR-139 and ICAM-1/CXCR4. In addition, mesenchymal stem cells (MSC) were transfected with pcDNA, pcDNA-H19, or H19 siRNA. The expression levels of miR-141 and miR-139 were decreased in pcDNA-H19 cells and increased in H19 siRNA cells, while the levels of ICAM-1 and CXCR4 were increased in pcDNA-H19 cells but decreased in H19 siRNA cells. Furthermore, MSC were transfected with pcDNA, scramble control, pcDNA-H19, or H19 siRNA. MTT and transwell assays revealed that the higher H19 expression had positive effects on cell proliferation and migration, while downregulation of H19 inhibited cell proliferation and migration. ICAM-1 has been shown to participate in the vital processes of immune reactions [9]. ICAM-1 can promote homing of numerous immune cells in secondary lymphoid organs. From a physiological standpoint, the ICAM-1 expression in MSC is very low, but ICAM-1 is dramatically enhanced in MSC located in an inflammatory environment [41,42]. In addition, ICAM-1 u-regulation enhances the immunosuppressive role of MSC [41][42][43].
The expression of ICAM-1 in IBD indicates a possible function of ICAM-1 in IBD pathophysiology. ICAM-1 was shown to be elevated in colon lysates collected from patients with ulcerative colitis [44,45].
The level of CXCR4 is an essential factor in promoting the proliferation and migration of stem cells. CXCR4 is mostly expressed in cell cytoplasm [46]. The level of CXCR4  Disease Markers is minimized with extended cell culture [47]. Various cytokines can promote the expression of CXCR4 on the cell membrane [48,49]. Transforming growth factor-β1 (TGF-β1) might upregulate the CXCR4 expression in basal cell carcinoma (BCC) cells [50]. TGF-β1 can also increase CXCR4 levels in MSC in acute I/R injury. Additionally, CXCR4 aids MSC migration to SDF-1.
The results of this study provided further understanding of the molecular mechanism underlying the therapeutic effect of stem cell in the treatment of IBD and identified the reinforcer, H19, to improve the therapeutic effect of stem cells. These findings may provide a basis for the future clinical use of stem cell as a modality to treat IBD.

Conclusion
In conclusion, collectively, the results of our study demonstrated that the overexpression of H19 in MSC downregulated the expression of miR-139 and miR-141, thus upregulating the expression of their target genes ICAM-1 and CXCR4, respectively. Since it has been verified that the overexpression of ICAM-1 and CXCR4 can promote the homing of MSC in the treatment of ulcerative colitis, the overexpression of H19 exhibited therapeutic effects in ulcerative colitis.

Data Availability
The data of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest
The authors declare that they have no conflicts of interest.