Mesenchymal Stem Cells in Liver Fibrosis: A Dose-Dependent Recovery

Abstract

Mesenchymal stem cells (MSCs) are known to assist liver regeneration. In this study, we show a dose-dependent mode of recovery from liver fibrosis after intravenous injections of MSCs. Male Wistar rats experienced a 42-day-long modeling of liver fibrosis via CCl₄ poisoning and received either a single injection of 2.5 × 10⁶ MSCs on Day 3 after the last CCl₄ dose or two MSC injections on Days 3 and 10. We dynamically monitored levels of liver cytolysis markers and cytokines in the venous blood and performed a histological study of Mallory-stained liver sections. All experimental groups experienced a nearly complete recovery of biochemical markers up to 4 weeks after the end of CCl₄ administration, although we observed anti-inflammatory changes in the cytokine levels only in animals treated by two MSC injections. Histological study revealed minor signs of liver damage up to Day 90 in animals receiving two MSC doses with worse pathology in those who received a single MSC dose. Morphometric values stayed consistent with visual data, demonstrating a significantly larger number of binuclear hepatocytes, a smaller number of false lobules, and a lesser area of connective tissue proper in animals treated by two MSC injections. Our results reflect MSC grafting in applied doses to affect liver fibrosis in a dose-dependent mode. These findings provide a deeper understanding of MSC action in liver fibrosis, and the doses applied may serve as a milestone for further studies in humans.

1. Introduction

Liver fibrosis accompanies numerous cases of chronic liver disease, worsening the prognosis for survival. Epidemiologic studies show a broad prevalence of this pathology. For instance, advanced liver fibrosis accounts for 2.85% of adults in mainland China [1]. Despite the obvious severity of replacing normal liver tissue with connective tissue proper, liver fibrosis may regress, although this process is slow and challenging [2,3,4].

During the past ten years, a growing body of evidence has disclosed more components of the machinery for such a recovery from liver fibrosis. Briefly, both metabolic regulation and cellular immunity play important roles, especially in stimulation by bone marrow stem cells, although via different mechanisms for hematopoietic stem cells (HSCs) and mesenchymal stem cells (MSCs). For the hematopoietic factor, molecular pathways of regulation involve tumor necrosis factor (TNF)-related, apoptosis-induced ligand binding in apoptotic changes in HSCs, and natural killers affect the HSC effects via gamma interferon (IFN-γ) action, whereas HSC-derived macrophages govern the production of matrix metalloproteinases to prevent the excessive proliferation of fibroblasts [5].

MSCs are also well known for preventing further fibrosis development [6,7,8,9,10,11] and even enhancing the formation of new hepatocyte-like cells via various machinery involved [12]. MSCs affect the proliferation of perisinusoidal stellate cells, transforming them into myofibroblasts and reducing collagen production [13]. Furthermore, MSCs, located in the liver, regulate neighboring cells in a paracrine way by releasing exosomes [14]. Recently, it has been shown in studies of a murine model that MSC-derived exosomes reduce endoplasmic reticulum stress and increase autophagic flux [15].

Due to this multifaceted activity of MSCs in the liver, their grafting is a prominent frontier for medical research [16]. However, few clinical studies are accessible for MSCs in liver fibrosis, except for liver cirrhosis. In cirrhotic patients, eleven clinical studies have already been accomplished, revealing a beneficial impact on survival and quality of life [17]. Nevertheless, these reports do not focus on other reasons for liver fibrosis, especially relevant for non-alcohol fatty liver disease or autoimmune pathology, implying possible graft-versus-host reactions in MSC grafting.

To design clinical approaches in the future, there exists a need for a more comprehensive understanding of common animal models (e.g., rats) of MSC usage. In this study, we aimed to study which time- and dose-dependent biochemical and histological changes occur in liver fibrosis after MSC grafting at early terms and whether there generally exists a considerable dose dependency (Figure 1). For this, we focused on bone marrow-derived MSCs as the most appropriate for potential clinical extrapolation rather than amniotic or umbilical cord-derived stem cells and hypothesized over whether there is a bond between the dose regimen of bone marrow MSCs and the mode of recovery from liver fibrosis.

2. Materials and Methods

2.1. Animals

This experimental study involved 96 male Wistar rats of 6–8 months of age with body masses of 400–450 g. The rodents stayed in our vivarium at 18–22 °C with a mixed diet and free access to water. The experimental procedures with animals took place in daytime (9 to 12 am) to avoid fluctuations in mitotic activity in the liver. All the manipulations were carried out in accordance with international and local rules (European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes, Russian State Standards GOST 33215-2014 and GOST 33216-2014 [18,19]) and were approved by the local ethical committee of Federal State Budgetary Institution “Shumakov National Medical Research Center of Transplantology and Artificial Organs” of the Ministry of Health of the Russian Federation (protocol No. 050221-1/3e 05/02/21) for studies involving animals. Only healthy rats, after a 2-week-long quarantine and veterinarian check-up, were included in this study (a priori criteria); no exclusions were made after the study group was formed.

2.2. Mesenchymal Stem Cells

We obtained allogeneic MSCs from 24 intact male Wistar rat donors aged 6–8 months with body masses of 400–450 g. We avoided using MSCs from experimental rats due to their diminished adhesive and proliferative activity with excessive debris in primary cell cultures. After bone marrow aspiration, we centrifugated the samples to obtain stem cells and removed the remaining components. We cultivated the resulting cell fraction in Dulbecco’s modified Eagle medium with 10% fetal bovine serum, L-glutamine at 2 mmol/L, penicillin at 100 U/mL, and streptomycin at 100 U/mL. The cell cultivation proceeded at 37 °C in air containing 5% CO₂ up to a high-confluent state (90%). Two passages of MSC cultivation were provided with overall cultivation term of 15 days. To check the phenotype of the cells, we provided a flow cytometric study of key superficial positive and negative markers (CD14, CD19, CD34, CD45, and HLA-DR should be negative for MSCs, whereas CD73, CD90, and CD105 should be positive) [20]. The proliferative potential of the MSCs was further improved by differentiation for various lineages of MSCs (StemPro Adipogenesis/Osteogenesis Differentiation Kits; Gibco; Jenks, OK, USA), with only samples of >95% trypan blue-positive cells being considered viable. Then, the cells were washed twice in isotonic saline and grafted onto the tail veins of the rats with insulin syringes.

2.3. Liver Fibrosis Modeling

We modeled liver fibrosis by poisoning Wistar rats (n = 72, not including the donor animals) for 42 days using carbon tetrachloride (CCl₄). Briefly, we provided subcutaneous injections of 60% CCl₄ and peach oil solution twice a week (Monday and Thursday) at a dose of 0.3 mL per 100 g of animal body mass, although the first injection contained 0.5 mL of 60% CCl₄ per 100 g of animal body mass. Therefore, the cumulative dose of CCl₄ stayed equal to 3.5 mL per 100 g of animal body mass.

2.4. Study Groups

After the modeling of fibrosis, we randomized the rats (n = 72) into experimental groups by the method of random numbers. Animals of Group 1 (n = 24) received a single intravenous injection of 2.5 × 10⁶ MSCs in 1 mL of saline on Day 3 after the end of CCl₄ treatment. Dosage regimens were regarded with respect to doses tested in our pilot experiments, revealing a sharp increase in efficacy at the range of 2.5–5 × 10⁶ MSCs per animal. In Group 2 (n = 24), rats were injected twice with the same dose of MSCs on Days 3 and 10 after the last CCL₄ treatment (both doses of MSCs were obtained from the same donor in all cases). Group 3 (control, n = 24) received no MSCs but an intravenous injection of 1 mL saline, i.e., sham treatment. The endpoints for blood sampling were on Days 14, 21, and 28 (plus Day 35 for cytokines), whereas the histological endpoints were on Days 3, 7, 14, 21, 28, 60 (three animals from group per point), and 90 (the rest). We sacrificed the animals via perfusion fixation by neutral buffered formalin and anesthesia by Zoletil-100 (Virbac, Carros, France) followed by liver autopsy with further histological analysis.

2.5. Blood Tests

Every seven days, we performed blood tests to follow the dynamics of biochemical markers for cytolysis in the liver, as well as for pro- and anti-inflammatory cytokines. To obtain blood from tail veins, the rats were anesthetized, and their tail tips were clipped. In the peripheral blood, we measured levels of biochemical liver markers: alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP). The quantitative analysis was performed using a Reflotron biochemical analyzer and test strips from the same manufacturer (Reflotron; Roche, Switzerland; AST–107 45 120, ALT–107 45 138, ALP–116 22 773, respectively). Furthermore, we studied the blood levels of interleukins, pro-inflammatory IL-4 and anti-inflammatory IL-10, as well as such pro-inflammatory cytokines such as IFN-γ and TNF-α by enzyme-linked immunosorbent assay (ELISA) with appropriate test systems (Protein contour; Saint-Petersburg, Russia; Vector-Best; Novosibirsk, Russia) using a ChemWell analyzer (Awareness Technology; Palm City, FL, USA).

2.6. Histological Study

Immediately after the rats were euthanized, we fixed the samples of liver tissue in 10% neutral buffered formalin and, after a standard dehydration, embedded them into paraffin. We provided Mallory’s trichrome staining at sections of 5 µm for histological study with easy detection of collagen fibers. For these sections, we measured the degree of fibrosis with ImageScopeM 12.1 software (systems for microscopy and analysis; Moscow, Russia) using microphotographs obtained with a Leica DM1000 microscope and a Leica LTDCH9435 DFC 295 camera (Leica; Wetzlar, Germany). For this, we calculated the percentage of blue area on Mallory-stained slides per field of view at microscope magnification ×400. Also, we counted the number of false lobules and binuclear hepatocytes per field of view of the same size.

For immunofluorescent study, we employed anti-Ki-67 antibodies (clone SR00-02, lot No. HA721115, article HP0304; Huabio, PRC, Woburn, MA, USA) at 1:100 PBS dilution. After incubation, we washed the sections with PBS and then processed them with secondary polyclonal antibodies, conjugated with horseradish peroxidase (Anti-Rabbit-HRP, lot No. HA1119, article M05-22-P2; Huabio, PRC). DAB served as chromogen. Data analysis was performed with blinded data by contributors who did not work directly with the experimental animals.

2.7. Statistical Analysis

Sample size was calculated with respect to pilot experiments as the least number of animals to obtain statistical significance in morphometrical values with a power of more than 0.8. We processed the resulting numeric data with OriginPro 2024 software (OriginLab; Northampton, MA, USA) by distribution normality assessment with Shapiro–Wilk test and following one-factor analysis of variance (ANOVA) together with Bonferroni correction, as all the numeric data samples followed normal distribution. The differences between the group samples were considered significant for p < 0.05 and a statistical power of more than 0.8.

3. Results

3.1. MSCs Are the Principal Component of Cell Culture in This Study

We cultivated cells of the MSC line obtained from the bone marrow of healthy rat donors. We validated the lineage of MSCs by flow cytometry. First, we increased the signal-to-noise ratio by excluding debris and doublets (Figure 2A). We picked data for including cells into calculations and found out that cell culture corresponds to the aforementioned criteria of MSC phenotype (Figure 2B). Furthermore, we provided an increase in proliferative potential through differentiation stimulation for osteogenic and adipogenic lineages (Figure 3). Then, we modeled the liver fibrosis.

3.2. Biochemical Markers of Liver Cytolysis Recover Faster After MSC Grafting

Immediately after the last day of CCl₄ poisoning, all surviving rats had a sharp increase in blood markers for liver damage, i.e., ALT, AST, and ALP (Figure 4A). Further dynamics of levels were similar for these three markers but rather different for the study groups. The animals from Group 1 received a single dose of MSCs on Day 3, and Group 2, which experienced double intravenous grafting of MSCs on Days 3 and 10, showed a significant decline in ALP and ALT levels up to Day 14, almost to initial values. These levels stayed approximately the same until the end of the blood test timeline with means for Group 2 being slightly less than in Group 1, although not significantly. Control rats from Group 3 also tended to recover their levels of biochemical markers. However, this recovery was less remarkable than in Groups 1 and 2, and for AST no full recovery occurred.

3.3. Cytokine Levels Vary Greatly After MSC Grafting

The largest number of inconsistencies between the groups and timepoints were detected for dynamics of cytokine levels, especially pro-inflammatory ones (Figure 4B). Thus, IL-10, an anti-inflammatory cytokine, revealed an exponential growth in animals’ blood throughout the time of the study, although the peak of IL-10 by Day 35 was significantly higher in Group 2. Interestingly, almost the same pattern of time-dependent changes is typical for IL-4 levels, which is basically a pro-inflammatory cytokine.

Meanwhile, Group 2 is remarkable for a specific mode of level changes for IFN-γ and TNF-α. The content of both these markers in control animals tended to increase up to Day 21 and then declined nearly to the initial levels. In contrast, Groups 1 and 3 (MSC-treated with a single dose and not treated, respectively) had a slighter rise in IFN-γ and TNF-α. Nevertheless, the content of these markers for Groups 1 and 2 held an uptrend until the last test, with the graph passing significantly upwards from Group 3, as seen in Figure 4B.

3.4. MSC Grafting Ameliorates Histological Signs of Liver Damage

On the histological slides, stained by Mallory’s trichrome, animals of all groups demonstrated considerably impaired histoarchitectonic traits in the liver up to Day 14 after the end of liver fibrosis modeling (Figure 5A, the line above; arrows mark the changes mentioned further). Specifically, the microphotographs for Day 14 demonstrate that the structure of hepatic plates is poorly seen. Furthermore, there is evident blood congestion in the sinusoidal capillaries as they widen, which is more remarkable in the pericentral zones of hepatic lobules. False lobules and numerous adipocytes are also found. Collagen fibers of connective tissue proper, replacing normal liver tissue, are visualized well due to their blueish color. No infiltration by inflammatory cells is seen.

However, the morphological pattern changes differently in study groups up to three months after the grafting. The slide microphotograph for a rat from Group 1 (Figure 5A, the line below), which received a single MSC injection, reveals a notable improvement in liver histology. Thus, no adipocytes are found, and hepatic plates are visible, although strong blood congestion still exists when central veins and sinusoidal capillaries are enlarged. Group 2, treated by double MSC doses, has a better image of histological changes with almost complete recovery of the aforementioned quantitative parameters. In Group 3, including animals after the sham treatment with saline, morphological features on Day 90 do not differ greatly from those on Day 14.

In the histological study, the visual differences between the groups are consistent with calculated morphometric values. The number of false lobules (Figure 5B) increases in all animals with liver fibrosis, with nearly complete recovery to sporadic false lobules found in Group 2 (two MSC injections) and a slight decrease in Group 1 (single MSC injection), which is significantly different from Group 3 (sham treatment after CCl₄) with practically no recovery at all. The percentage of connective tissue area followed the same tendency (Figure 5C), except for the sharp turn downwards in Group 2. Therefore, at 90 days since fibrosis modeling, Group 2 experienced full recovery with excessive connective tissue disappearing, whereas Group 1 had partial recovery, and Group 3 had no recovery at all. The last morphometric parameter measured, the number of binuclear hepatocytes, followed the pattern of a gradual, exponential increase for all the groups (Figure 5D), although more pronouncedly for Group 2 after double MSC injection.

Furthermore, we have provided immunohistochemical/immunofluorescent study to observe whether MSC-treated groups have another rate of expression of markers for regeneration. For this, we employed anti-Ki67 immunofluorescent staining. Ki67-positive cells already appeared in Group 1 after being treated by a lesser amount of MSCs (Figure 6A), whereas their number increases in Group 2, treated by a larger dose of MSCs. Meanwhile, animals with fibrosis but no further treatment (Group 3), as well as healthy donors, experience almost no Ki67 expression at all. For the statistical calculations (Figure 6B), Groups 1 and 2 have a significantly larger number of Ki67-positive cells per field of view (p < 0.05) than Group 3. Therefore, the IHC-related data are consistent with the histochemical ones.

4. Discussion

In our study, we observed a dose-dependent recovery from liver fibrosis through histological and biochemical data after 6 weeks of CCl₄ poisoning. The dynamics of ALT, ALP, and AST revealed no unexpected findings, though they approved of the model and showed dose-dependent differences. The only point we should outline is the relatively rapid recovery of liver cytolysis markers even in the case of no actual treatment, which demonstrates the liver’s fascinating potential regeneration. Meanwhile, another cluster of blood testing data, the cytokine study, has provided us with more interesting outputs.

Animals from Group 2, receiving a double dose of MSC grafting, were remarkable for an initial increase in IFN-γ and TNF-α levels up to Day 21 and then a reversal almost to the baseline, a feature not observed in Groups 1 and 3 (one dose of MSCs and sham MSCs, correspondingly) with a slight but constant increase in IFN-γ and TNF-α content until the end of the experimental study. This pattern of IFN-γ and TNF-α is perhaps related to the recruitment of HSC-associated molecular pathways and the direct involvement of regulation by natural killers [5]. However, the role of HSCs should not be overestimated here. We consider that there exists a probable interaction between MSCs and T cells via exosomes in our experiments [21]. Furthermore, not only exosomes may mediate the effects but mitochondria too, if they are transported from MSCs into T cells [22]. This codependence between stem cells and T cells requires an elucidation in future studies by additional methods.

For interleukins, we observed two similar trends in both pro-inflammatory IL-4 and anti-inflammatory IL-10, although significantly less pronounced in MSC-treated Groups 1 and 2. While the dynamics of IL-4 are consistent with well-known data, the sharp peak of IL-10 up to Day 35 in animals with sham treatment (Group 3) demands discussion. We assume this CD10 peak is compensatory and beneficial but simply not required in the presence of MSC stimulation. In animals without treatment, CD10-related machinery may induce immune modulation via substance P-associated regulation [23], thereby supporting weak positive dynamics. Concerning the immune effects, we should also emphasize that all the changes seen were only biochemical, and no infiltration by leukocytes has been revealed on histological slides.

However, histological sections of the liver visualize the difference between single and double doses of MSCs (Figure 5), which is consistent with morphometric data from both Mallory’s and anti-Ki67 staining. Calculations showed that the fibrosis degree decreases gradually with a significant dose dependence, as seen from the reduced number of false lobules in MSC-treated groups and the diminished percentage of connective tissue proper found there. We also observed a larger number of binuclear hepatocytes and Ki67⁺ cells. Importantly, binuclear hepatocytes increase at many timepoints for Group 3 (not treated by MSCs) and even more than in Group 2 (treated by two doses). We believe this is due to a more rapid recovery of fibrosis in Group 2, requiring less healthy hepatocytes to experience a polyploid transformation. However, we avoided considering the increase in binucleated hepatocytes as an event of proliferation of hepatocytes, as it may just be a compensatory hypertrophy of the remaining functional portion of the liver. Indeed, we have not observed mitosis figures nor angiogenesis that could be related to de novo hyperplastic changes.

Furthermore, the remarkable regeneration with a normal histoarchitectonic structure formed de novo in double MSC grafting should be due not only to regressing fibrosis, survival of hepatocytes, or forming new hepatocyte-like cells but also to a recovery of hepatic plates and sinusoidal capillaries between them. It has been reported that MSCs may transform into endothelial cells [24] in vivo in cardiovascular pathology with a beneficial impact on its course. Thus, one may expect the same effect in enhancing the repair of hepatic plates, and that requires additional investigation in the future. Moreover, we may anticipate a direct link between the cytokine and histological changes observed, especially in the reconstruction of hepatic plates and sinusoidal capillaries.

Generally, our study showed MSCs’ well-known potential for liver regeneration but with novel data on the rigorous dose-dependent mode of such an impact observed both histochemically and immunohistochemically. Moreover, we observed that systemic administration is also beneficial for MSC activity in the liver, despite previously reported facts, although in mice and not in rats [8]. A similar dose-dependent mode of MSCs was revealed earlier in other pathologies, such as in acute kidney injury [25].

In this study, we intended to regard the impact of MSCs without any applicable support, like cell-engineered constructs involved to maintain their survival. Nevertheless, the dose-dependent effects of MSC usage in applied doses are well found. However, we do not assume a direct impact of MSCs on fibrosis in the liver. We consider the observed changes to be due to paracrine, exosomal, and immune-modulating regulation, as shown previously [26].

Generally, exosomal influence of grafted MSCs is considered the most prominent, with many RNAs transferred. For example, microRNAs such as miR-21-3p enhance endothelial growth and suppress fibroblastic proliferation through inhibition of phosphatase and tensin homologs, as well as sprouty homolog 1 (reported for umbilical MSCs) [27], while several RNAs act via the PI3K-AKT-mTOR axis (reported for amniotic MSCs) [28]. Moreover, both molecular machineries affect macrophage activity. Further, bone marrow-derived MSCs, relevant for our study, were also shown to include tumor growth factor beta and SMAD-protein-related mechanisms, thus preventing excessive connective tissue formation and favoring regeneration, and this is in skin wound healing [29]. Nevertheless, exosomal release alone is probably not sufficient to provide all the beneficial effects of bone marrow MSCs: in a model of renoprotection, bone marrow MSCs had a better impact on fibrosis than their separated exosomes [30].

Specially for liver fibrosis models of bone marrow MSC grafting, some supplementary machinery has been reported. First of all, not exosomes only but microvesicles of MSCs permit the release of active factors from MSCs [31]. Thus, it has been recently shown that the regenerative potential of bone marrow MSCs in the liver depends on protein tyrosine phosphatase type 4A activity. If it is strong in the mitochondria of MSCs, they can benefit from anaerobic metabolism [32]. Interestingly, the efficiency of bone marrow MSC grafting in the liver also correlates with erythropoietin levels [33]. Furthermore, matrix metalloproteinase 14 can serve as a marker of successful bone marrow MSC action in liver regeneration, as its levels increase due to excessive secretion by surviving MSCs [34]. Then, immune mechanisms are involved in the bone marrow MSC-associated effects in the liver. Indeed, NK cells increase in the liver after bone marrow MSC grafting, accelerating the removal of impaired cells [35].

Furthermore, research with stem cells raises an important issue of ensuring their viability and efficiency for successful grafting. For example, a lot of previous research, including ours, has reported possible approaches to improving the viability of MSCs via artificially designed gels [36,37,38]. However, the practical choice between MSCs and HSCs may be challenging in possible clinical implementation, as we showed the benefit of HSCs and their total RNA especially in chronic liver disease [39]. We consider this choice should be based on potential adverse effects in certain pathology and not on the direct impact of MSCs or HSCs.

Limitations and Perspectives

• The set of diagnostic tools involved is limited. For instance, a recent study with a similar design, although via 9 and not 6 weeks of CCl₄ poisoning, revealed antioxidant machinery involved in the MSC effects in liver fibrosis [40]. This research, parallel to ours, implies the need to enlarge the set of methods used, as the current study was obviously limited and may be even biased by the use of fixed tissues not able to reveal the molecular pathways of antioxidation. We propose that an aspiration biopsy via a known method [41] can be appropriate for the widening of research in this field. Hypoxic effects also deserve a detailed study because of their ambiguity in liver fibrosis. For example, a fascinating report by Kuo and colleagues showed MSCs from Wharton’s jelly were able to reverse fibrotic changes in the liver, and this impact has been more evident in hypoxic states [42]. Furthermore, future research should include additional methods of liver fibrosis visualization with higher resolution and/or accuracy (e.g., coherent anti-Stokes Raman scattering microscopy, fluorescence lifetime imaging, or second harmonic generation microscopy) to observe the dynamic recovery of hepatic plates.
• Our study does not consider the variety of subpopulations for MSCs [43]. Perhaps, a more refined selection of MSC subtypes may lead to a more remarkable beneficial impact. For example, Nishina and colleagues have recently demonstrated that Muse cells, a subpopulation of MSCs, exhibit the largest proliferative potential in liver fibrosis after twelve weeks of CCl₄ poisoning [44]. Furthermore, not only are MSC subpopulations not regarded in the current work but also immune- and apoptosis-related markers, as the study design was focused on revealing whether there exists a dose-dependent effect. However, further research should elucidate the issue of certain apoptotic and immune-mediated pathways involved in the liver’s recovery from fibrosis in this model.
• The choice of intervention mode for fibrosis modeling is challenging. We preferred CCl₄, as it is the most common agent for fibrosis modeling in animals [45], not only with rats but mice [46] and monkeys [47], too. Furthermore, we had tested it successfully in our previous studies [37,38]. However, other chemical models than CCl₄ have also been successfully tested in rats, for example, poisoning with thioacetamide [48], diethylnitrosamine, and acetaminophen [45]. Furthermore, surgical, organelle-targeting, immune-mediated, alcohol- or diet-provoked models exist, as well as liver fibrosis modeling via breeding transgenic animals with related genes involved [45]. All these modes require additional testing of our regimen for liver fibrosis recovery in our further research.
• Animals of the same sex were included to obtain consistent results; for ethical reasons, we avoided doubling animal number with both sexes observed. Thus, we preferred male rats, as comparable experiments with male rats have been conducted by other teams [49,50,51]. However, we assume female rats to be also appropriate for such studies, and further research should also account for testing MSCs in female animals.
• There exists a necessity to entirely and comprehensively describe if the MSC-induced changes permit survival of hepatocytes with their compensatory hypertrophy only or if certain hyperplasia can also be seen. In further studies, a detailed evaluation of cell growth is therefore required.
• An additional limitation raises the possibility of extrapolation between our studies and human disease. For sure, conventional models of liver fibrosis have already been regarded as partially available for translation to human pathology [52]. To bring these models closer to human liver structure, liver fibrosis models have been developed in primates in recent years [47,53]. However, not only do the dosage regimens between rodent/primate studies and clinical studies of MSCs in liver pathology differ, but also it is challenging in clinical studies to find a proper dosage (various clinical studies used doses differing in billions of ways) [54,55,56]; thus, the importance of our study is of high level.

5. Conclusions

In this study, we aimed to reveal if MSC grafting has an impact on liver fibrosis in a rat model at early terms. We employed easy-to-access biochemical and histological methods and therefore demonstrated a successful model of liver damage and further recovery consistent with parallel studies in the field. Importantly, we revealed a remarkable dose-dependent impact of MSC grafting on the course of liver fibrosis as well as on accompanying levels of liver enzymes and cytokines. Further studies with respect to the perspectives and limitations mentioned are required for a comprehensive understanding of fascinating MSC potential.