Intravenous Injection of Umbilical Cord-Derived Mesenchymal Stem Cells Improved Regeneration of Rat Liver after 2AAF/CCl4-Induced Injury

Department of Anatomy, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Stem Cell Medical Technology Integrated Service Unit, Cipto Mangunkusumo Central Hospital, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Stem Cell and Tissue Engineering Cluster, Indonesian Medical Education and Research Institute, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Integrated Laboratory, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Master Program of Biomedical Sciences, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Department of Anatomy, Faculty of Medicine and Health Science, Universitas Kristen Krida Wacana, Jakarta, Indonesia Department of Anatomy, Faculty of Medicine, Universitas Yarsi, Jakarta, Indonesia Department of Anatomical Pathology, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia Animal Research Facilities, Indonesian Medical Education and Research Institute, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia


Introduction
Liver injury is still a neglected global health problem in its early stages and has a high potential to progress to liver fibrosis (scar tissue formation), ending in cirrhosis that leads to death (Marcellin and Kutala, 2018). Although the morbidity and mortality is increasing year to year, until now no effective therapy has been found. Globally, total cases of cirrhosis reached 41.4 million people in 2017, with the number of deaths reaching more than 1.32 million people or 2.4% of the world's population (GBD 2017Cirrhosis Collaborators, 2020. Cirrhosis conditions cannot be cured with conventional medicine. The most widely used treatment method for chronic liver damage to restore liver mass is a liver transplant; however, there are some shortages, such as limitation of the availability of the liver organ and immunological rejection. Therefore, nowadays, there is a consideration for using cell therapy. One of the candidate is stem cells therapy as an adjuvant to promote liver regeneration (Nicolas et al., 2016;Yun et al., 2016).
In chronic liver injury, hepatocytes are no longer able to regenerate the liver (decrease of their proliferation capacity). It is featured with the accumulation of collagen in Extracellular Matrix (ECM). This condition occurs in alcoholic and non-alcoholic cirrhosis, chronic viral hepatitis and primary biliary cirrhosis. If this condition proceed, although there are activation of liver progenitor cells population, recovery of epithelial cell mass, architecture and liver function will be exceeded by collagen deposition that lead the liver falls into the final state, liver cirrhosis (Riehle et al., 2011). In chronic liver injury, cytokine production by damaged parenchymal cells and disruption of ECM are stimuli for HSC (Lo and Kim, 2017). Responding to this stimuli, Hepatic Stellate Cells (HSC) are activated and their phenotypes turn into fibrogenic myofibroblast (Xu et al., 2014) and take part in the regulation of various forms of liver inflammation such as increase the expression of inflammatory receptors, including chemokine receptors, inflammatory mediators in the form of chemokines andleukocyte chemoactivity. This initiation phase changes the expression of the genes and phenotypes of liver cells; thereby, it makes them more responsive and perpetuates HSC activation to generate collagen accumulation in ECM and progressive fibrosis (Tsuchida and Friedman, 2017).
Recently, cell therapy has become an alternative treatment for chronic liver injury. It is developed to prevent worsening of the disease, support the liver regeneration and hopefully can reduce the mortality rate. Utilization of therapeutic cells can be carried out through stem cells that have the ability to modulate the endogenous regeneration process of the liver and inhibit fibrosis (Nicolas et al., 2016;Yun et al., 2016). There are several sources of stem cells to treat liver disease, especially in cirrhosis, including fetal hepatic stem cells or hepatoblasts (Habeeb et al., 2015), fetal biliary tree stem cells (Cardinale et al., 2014), Hepatic Progenitor Cells (HPC) (Best et al., 2015) and Induced Pluripotent Stem Cells (IPSC) (Zhu et al., 2014). Nowadays, Mesenchymal Stem Cells (MSC) are known and have been studied as the most promising and potential therapy for liver cirrhosis (Eom et al., 2015;Cao et al., 2020). MSC can originate from various tissues, one of them is the umbilical cord that has several benefits. Umbilical cord is known as a birth waste and a rich source of stem cells after bone marrow (Fazzina et al., 2015;Vangsness Jr et al., 2015). MSC isolated from the umbilical cord have the ability to secrete tropic factors such as cytokines, chemokines and growth factors to encourage the liver regeneration through reducing inflammation or fibrosis, immunomodulatory effects, also tissue repair (Troyer and Weiss, 2008;Ishige et al., 2009;Eom et al., 2015). It is also suggested that MSC have a capacity for multilineage differentiation, including differentiation into hepatocytes (Vangsness Jr et al., 2015;Haldar et al., 2016).
Although MSC application for liver fibrosis has became the attention of clinicians and scientists, the mode of action is still unclear. MSC may have direct antifibrotic effects; however, its possibility to have a profibrotic effects should be concerned as well (Haldar et al., 2016). Several basic and clinical studies on the administration of MSC for liver fibrosis showed positive improvement of liver functions (increase serum albumin level), hepatocyte integrity (decrease serum level of Alanine Transaminase/ALT and Aspartate Transaminase/AST) improvement of tissue morphology, decrease of fibrosis and decrease of hepatocyte apoptosis (Hwang et al., 2012;Volarevic et al., 2014;Haldar et al., 2016). For example, in a study done by Motawi et al. (2014) MSC administration in carbon tetrachloride (CCl4) inducedliver fibrosis rats could repair the fibrotic tissue better than simvastatin as seen on the improvement of serum liver function tests. MSC also decreased fibrosis score support by improvement of molecular finding, i.e., increased of Matrix Metalloproteinase-1 (MMP-1) gene expression, decreased liver procollagens I and III, Tissue Inhibitor of Metalloproteinase-1 (TIMP-1) and endoglin gene expressions, as well as decreased of Transforming Growth Factor-beta (TGF-ß1) immunostaining (Motawi et al., 2014). Nevertheless, a negative result, i.e., MSC did not reduce fibrosis, was ever been reported (Carvalho et al., 2008).
Dose, timing and route of the administration of MSC are important for therapeutic effect (Haldar et al., 2016;Cao et al., 2020). Administration doses in clinical studies were reported 0.510 6 cells/kg body weight (bw) for allogeneic infusion and 110 6 cells/kgbw to 3.410 8 cells/kgbw for autologous infusion. In animal studies using rat or mouse, the dose was ranged from 0.2510 6 cells to 1.610 7 cells, with the most widely used dosage was 110 6 cells (Haldar et al., 2016). Although it is not possible to determine the optimal dose of MSC to achieve therapeutic response, there was a trend of dose dependent reduction of fibrosis. The timing of administration, whether it is single dose or splitting dose, has not been proven yet which one is better (Haldar et al., 2016). In seriously ill patients, preferred route of administration is peripheral intravenous rather than direct to the target organ, such as portal vein or splenic vein, because it is safer, more comfortable and allows to repeat and to infuse large amount of administrations (Eggenhofer et al., 2012). In the intravenous route, most of the MSC will be trapped in the lung and the rest in the spleen, liver, bone marrow, thymus, kidney, skin and tumors (if any), which probably due to the bigger size of MSC (±20-30 µm) compare to the capillaries (<20 µm) and their strong adhesion properties of MSC (Crop et al., 2011;Shi et al., 2011). In vivo tracing of MSC showed that only a small number of these cells will home in the target organ and stay not detected 7-14 days after infusion. This suggests that MSC can produce clinical effects through paracrine signaling, such as Hepatocyte Growth Factor (HGF), Vascular Endothelial Growth Factor (VEGF), Prostaglandin E2 (PGE2) and Indoleamine 2,3-Dioxygenase (IDO) rather than transdifferentiation (Cao et al., 2020).
Several studies on human clinical trials and animal studies have provided new hope for the treatment of liver fibrosis. However, most of the study was exploring the potency of BM-MSC and did not explored enough the umbilical cord-derived MSC (UC-MSC). There are also conflicting research results; whether MSC could repair liver damage and reduce fibrosis (antifibrosis) or conversely. Liver fibrosis or cirrhosis is a later stage of liver injury or post-injury with decreasing ability of hepatocyte proliferation. The patients usually fall into severe illness and have weak physical condition that making them suffer more to obtain bone marrow aspiration procedure for autologous therapy. In addition, given the ability of MSC to produce soluble/paracrine factors, it is possible to administrate it intravenously, which is less invasive (Haldar et al., 2016;Cao et al., 2020). Therefore, this study has been carried out in an animal model of chronic liver injury, 2-Acetylaminofluorene (2AAF)/CCl4 rats, to find out whether intravenous injection of single dose, 110 6 UC-MSC can regenerate the liver tissue and reduce the fibrosis.

Animal Handling
Male Wistar rats (n = 18), aged 8 weeks with a body weight of 160-200g, were obtained from Animal Research Division of Research and Development Agency, Ministry of Health of the Republic Indonesia. Animals were housed in a constant temperature, under a 12h dark/light cycle and supplied with laboratory chow and water ad libitum. All animal experiments were approved by Ethics Committee of Faculty of Medicine Universitas Indonesia (No.1277/UN2.F1/ETIK/2018). Eighteen rats were divided into three groups of six rats. Group I was the control group (healthy rats). Group II was 2AAF/CCl4 induced liver injury rats and treated with Human Umbilical Cord Mesenchymal Stem Cells (hUC-MSC). Group III was 2AAF/CCl4-induced liver injury rats, but did not treat with hUC-MSCs. To determine the liver function, laboratory blood examination of serum ALT and Albumin were performed using the blood that was taken from orbital vein. All groups of rats were decapitated at two weeks after the induction and/or stem cells injection (week 14 th ), followed by collection of the liver.

2AFF/CCl4 Induction and hUC-MSC Injection
The 2-Acetylaminofluorene (2AAF; Sigma-Aldrich ® , Darmstadt, Germany) and carbon tetrachloride (CCL4; MERCK, Munchen, Germany) induction was done in 12 weeks. The stem cells injection was done one day after the last induction. Human Umbilical Cord blood-derived Mesenchymal Stem Cells (hUC-MSC; CD73 99.8%, CD105 95%, CD90 99.9%, negative marker 0.4%) for liver regeneration treatment was produced by Stem Cells and Tissue Engineering Cluster (SCTE), Indonesian Medical Education and Research Institute (IMERI), Faculty of Medicine Universitas Indonesia (FMUI). Group I (control group) received equal quantities of subcutaneous injection of olive oil and intragastric Polyethylene Glycol (PEG) for 12 weeks and received intravenous injection of 1000 µL physiological NaCl once, one day after. Group II (2-AAF/CCl4 group with stem cells) was induced with subcutaneous injection of 2 mL/kg CCl4 in olive oil twice a week for 12 week (week 1 th -8 th CCl4: olive oil = 1:1 and week 9 th -12 th CCl4: olive oil = 3:7) and intragastric 10 mg/kg 2AAF in PEG everyday for two weeks (week 9 th -12 th ) and treated with intravenous injection of 110 6 hUC-MSC in 1000 µL physiological NaCl once, one day after 2AAF/CCl4 induction. Group III (2AAF/CCl4 group without stem cells) was induced with CCl4 and 2AAF with the same regimen as group II and received intravenous injection of 1000 µL physiological NaCl. All groups were decapitated two week after the induction (week 14 th ).

Histopathological Staining
For histopathological analysis, liver specimens were fixed for 12-24 hours in 10% neutral formalin solution, embedded in paraffin, sectioned as thick as 3 μm and were stained with Hematoxylin Eosin (HE) to evaluate the liver morphology and pathological process, Masson's Trichrome (MT) to evaluate the degree of fibrosis, immunostaining against Caspase-3 (diluted 1:50, Abcam, ab13847) to observe the cells apoptosis. Histofine universal immuno-peroxidase polymer (NICHIREI, Tokyo, Japan) was used as secondary antibody and 3,3′-Diaminobenzidine Tetrahydrochloride (DAB) was used as a chromogen, yielding a brown signal. At the end of process, the tissue was cleared by immersing it in xylol solution, mounted in entelan and covered by a cover glass. Histopathological evaluation was performed under light microscope (LEICA type DM750, Wetzlar, Germany).

Data collection and Analysis
Morphology of the liver and Histopathology (HE and MT) were reported descriptively. Observation of the severity of liver damage in the HE staining's tissue was determined using French et al scores, includes fat degeneration, necrosis, cell swelling and inflammation, as follows: 0 = no visible liver cell damage, 1 = focal damage less than 25% of tissue, 2 = focal damage 25-50% of tissue, 3 = extensive but focal, 4 = total damage/necrosis (Eidi et al., 2013). In the MT staining's tissue, the degree of fibrosis was observed and determined by Non-Alcoholic Fatty Liver Disease (NAFLD) scores (Onyekwere et al., 2015). For SPSS statistical consideration, the score was adjusted for the division of staging/fibrosis as follows: 0 = no fibrosis, 1 = perisinusoidal fibrosis in zone 3 or portal or periportal, 2 = perisinusoidal fibrosis in zone 3 and fibrosis in portal/periportal, 3 = bridging fibrosis, 4 = cirrhosis. In addition, in each slice, the percentages of damaged and fibrosed areas was assessed in five microscopic fields and then averaged.
Data of ALT, Albumin and Caspase-3 were tabulated, tested the normality using Shapiro Wilk normality test and analyzed the significant different between groups using one-way ANOVA analysis of variance. The criteria for statistical significance was p<0.05. All values were expressed as mean values ± SD. Observation of each antibody expression was carried out in five fields of liver tissue. Cell counts and measurements of the surface area of the liver tissue are performed using the ImageJ application. Statistical analysis was carried out using SPSS 20.0 for Windows statistical package.

Results
Macroscopic observation and laboratory blood examination showed the trend of descriptive changes between group II and III (Fig. 1). Group II had similar appearance compared to group I, i.e. the liver had smooth surface whereas, group III had uneven surface. Group III had the highest level of the ALT, followed by group II and group I. ALT levels in group II was 1.2 times lower than group III. Group III had the lowest Albumin level, followed by the group II and group I. Nevertheless, one way ANOVA test resulted that both ALT and Albumin levels in the three groups were not significantly different (p = 0.302 and p = 0.103).
Observation of HE staining showed that group III (2-AAF/CCl4 group without hUC-MSC) had severe liver damage compared to the group II (2-AAF/CCl4 group with hUC-MSC) (Fig. 2, and Tables 1 and 2). Group I, the healthy control group, showed classic lobular with hepatocytes arranged radially to the central vein, which is separated by sinusoid. Group II had liver damage, includes fat degeneration (score 0-1), cell swelling (score 1-4) and inflammation (0-1). Group III showed all criteria of liver damage, includes fat degeneration (score 0-1), necrosis (score 0-1), cell swelling (score 0-4) and inflammation (score 0-1). Group III showed extensive fat degeneration (67%) than group II (30%). In all observed area of group II, cell swelling could be seen, but not the necrosis. Both group II and III showed irregular liver's lobules.
The MT staining showed decrease of collagen distribution in group II compared to group III ( Fig. 1 and Table 1 and 2). It could be seen that distribution of collagen fibers in zone 3 per sinusoidal or porta/ periportal in group II decreased. Moreover, there was no area of cirrhosis in group II compared to group III that showed one case of cirrhosis. In control group, the degree of fibrosis score was 0, which indicated normal collagen fiber distribution. The degree of fibrosis ranged 0-2 in group II, however it ranged 1-4 in group III. The fibrosis could be seen in all observed area of group III, but it was decreased to 87% in group II.
The expression of Caspase 3 between group showed a significant difference (p = 0.004; one way ANOVA; Fig. 3). The post hoc test, Tukey multiple comparison test, showed that group II significantly different with group III (p = 0.038) and did not differ with group I (p = 0.432). Caspase 3 expression of group III was more than 3 times higher compared to group II (Fig. 3E). In group I, the expression index was zero (Fig. 3B).

Discussion
Chronic severe liver injury caused by various factors including alcohol consumption, viral infections (hepatitis B and C), NAFLD and non-alcoholic steatohepatitis (NASH) (Zhou et al., 2014;Marcellin and Kutala, 2018) that has becoming a global health problem had challenged us to develop an animal model of it to understand the disease progression, regeneration and management. In this study, we used 2AAF/CCL4 in rat that cause liver damage mimicked human severe liver injury, i.e., extensive liver damage and suppression of hepatocyte proliferation. CCL4 can trigger necrosis in lobular zones of the liver leading to acute inflammation (Mao et al., 2014) due to its active metabolite, the free radicals of Trichloromethyl (CCl3 + ) (Omara et al., 2018). Meanwhile, 2AAF is known to trigger liver injury by changing the detoxification enzyme level (imbalanced antioxidant status), inflammation and apoptosis (Hasan and Sultana, 2015) that leads to the hepatocyte proliferation blocking (Abdellatif et al., 2017). This combination model has been used previously by Abdellatif et al (2017) to study the response of oval cells after human umbilical cord blood stem cells (mononuclear cells) transplantation. Our study demonstrated that the induction caused a liver tissue damage, includes the fat degeneration, cell swelling, necrosis and inflammation (Fig. 2E, Tables 1 and 2) and generation of extensive fibrosis to cirrhosis (shown by collagen deposition in MT staining; Fig. 2F, Tables 1 and 2). This model also shown a tendency of ALT increase and Albumin decrease (Fig.  1), which indicating a damage of liver parenchymal cells (Abdellatif et al., 2017).
The liver damage in this model is also supported by the increased of Caspase 3 expression that implied the increased of apoptosis of the liver cells (Hasan and Sultana, 2015). In the group III, Caspase 3 was expressed more than eight times compared to the control group (Fig. 3). Apoptosis in our model was one of the sign of liver injury that could be cause by both of 2AAF and CCl4. The mechanism of apoptosis was different but both caused oxidative stress due to the production of substantial Reactive Oxygen Species (ROS), which leads to the mitochondria damage (Lin et al., 2019). The chemical exposure might have an effect on DNA damage through the formation of free radicals. The DNA damage activates cytoplasmic sensors and results in decreased production of anti-apoptotic protein (Bcl-2) and an increase in the amount of pro-apoptotic protein (Bax/Bak). Consequently, leakage of cytochrome c protein that is usually produced in mitochondria. Cytochrome c protein leakage can activate various caspase series, starting with initiation by Caspase-9 and then is executed by Caspase 3 (Elmore, 2007;Lin et al., 2019).
Our study in injecting single dose, 110 6 hUC MSC intravenously into the model of severe chronic liver injury, 2AAF/CCl4 rat, has shown a positive result in supporting the liver regeneration. The group II (2-AAF/CCl4 with hUC-MSC) showed an improvement in the liver condition. The liver damage has decreased compared to the group III (2-AAF/CCl4 without hUC-MSC), especially in the degree of necrosis and coverage area of fat degeneration, even though the degree of inflammation and cell swelling has not changed. The fibrosis of the liver also improved, shown by decreased of the fibrosis level and coverage area. None of group II rat had fallen into cirrhosis as seen in group III. In line with our study, (Burra et al., 2012) reported that intravenous administration of hUC-MSC on CCL4-induced mice experienced liver recovery, i.e., decrease inflammatory cells and necrosis area and increase albumin synthesis. Moreover, some clinical trials in primary biliary cirrhosis, acute in chronic liver failure, decompensated liver cirrhosis and liver cirrhosis caused by autoimmune disease (as reviewed by Yang et al., 2020) have shown improvement of liver function. Although statistically insignificant, in this study, the liver tissue improvement appeared to be accompanied by a trend towards decrease in ALT and increase in albumin ( Fig. 1D and 1E), which implied a trend of increase in liver function. ALT and Albumin are common parameters used as indicator for hepatocyte damage. In liver disease, level of serum ALT increase and level of serum of albumin decrease (Giannini et al., 2005;Maulidia et al., 2020). In acute condition, the serum ALT usually increase significantly; however, in chronic condition or liver cirrhosis, the level depends on the stage of the disease. In compensated phase, it can stay in normal range or fluctuating around the upper reference value; however in the decompensated phase the level tend to increase (Giannini et al., 2005;Maulidia et al., 2020). In the alteration of Aminotransferase levels (ALT and/or Aspartate aminotransferase/AST), the changes of albumin level in the blood is prognostic for liver function since albumin is produced by hepatocytes (Maulidia et al., 2020).

Conclusion
Intravenous injection of single dose, 110 6 hUC-MSC could inhibit the progression of liver damage and reduce the liver fibrosis of 2AAF/CCl4 rats. The treatment could decrease the apoptosis, shown by significant reduction of Caspase 3 expression. This treatment regimen has not significantly improved the liver function; nevertheless, a tendency of ALT decrease and Albumin increase was observed. Since the MSC was injected intravenously and the cells are mostly trapped in the lung, it may be necessary to increase the MSC doses for further studies.