Repeated or chronic biliary obstruction is a common result of choledocholithiasis, cholangitis or postoperative complications of cholecystectomy. Clinically, biliary obstruction can manifest as abdominal pain, fever, itching, dark urine and jaundice, but it can also give rise to more severe complications such as portal hypertension (PH), acute cholangitis, vitamin deficiencies and bacterial infections [1
]. However, the most severe consequence of chronic cholestasis is secondary biliary cirrhosis (SBC), a life-threatening disease with liver transplantation as the only curative treatment option [2
]. In cases of benign obstruction, cirrhosis can develop within 14–62 months, while malignant cases can quickly develop into SBC in as few as 9 months [1
Common treatments for cholestasis often begin with attempts at drainage, stent placement or, if choledocholithiasis is suspected, then endoscopic retrograde cholangiopancreatography and stone removal. In more severe cases of liver damage, the initial course of treatment usually involves management of symptoms. For example, ursodeoxycholic acid is often prescribed to improve liver function, while other drugs can be used to absorb circulating bile acids [8
]. However, in cases of SBC, the only curative treatment that exists to control disease spreading and liver failure is liver transplantation [2
]. Liver transplantation has been associated with good long-term outcomes for patients with SBC, increasing 5-year survival rates to 69–75% [4
]. Nevertheless, liver transplantation is only recommended in severe cases that exhibit functional changes in the liver. Furthermore, on top of the existing limitations surrounding liver transplantation, additional long-term complications associated with biliary obstruction can also hinder surgical efficacy. By the time patients are considered for liver transplantation, they have likely already undergone several, often failed, operations, which can serve as an obstacle for future surgical repair [11
]. Moreover, the high risk of PH in cirrhotic patients can lead to increased secondary bleeding during surgery, and the presence of PH can even increase 60-day postoperative mortality rates for patients undergoing any surgery for bile duct strictures from 2% to 23% [2
Given the severity of complications associated with liver transplantation for patients with chronic cholestasis that can develop into SBC, it is necessary to develop an alternative method that is still a curative, as opposed to symptomatic, treatment. One such option may arise from referring to treatment methods for other forms of cirrhosis. Specifically, the use of mesenchymal stromal cells (MSC) to cure liver cirrhosis has had positive results in both in vivo research and clinical trials. In mice, bone marrow-derived MSC were used to successfully ameliorate carbon tetrachloride (CCl4)-induced liver fibrosis by expressing a VEGF receptor that stimulated hepatic regeneration and collagen deposition [15
]. These results were expanded by Li et al., who found that MSC can promote liver regeneration after portal vein embolization in cirrhotic rats [16
]. Additional aspects about the treatment mechanism were elucidated by Higashiyama et al., who found that MSC express matrix metalloproteinase-9 with antifibrotic effects, and Oyagi et al., who discovered that culturing MSC with hepatocyte growth factor can improve their anti-fibrotic effects [17
]. Similar results were found in human clinical trials by Jang et al. and Amin et al., which showed that autologous bone marrow-derived MSC can treat alcoholic cirrhosis as well as post-hepatitis C virus cirrhosis [19
As such, although liver transplantation can be an effective treatment for SBC, the potential complications associated with surgery in severe cases inhibit its use. Thus, it is necessary to find a different method of treatment that still holds the same curative properties. MSC have the potential to treat biliary cirrhosis by both non-parenchymal cell response modulation as well as supporting and stimulating the regeneration of nascent hepatocytes [15
]. In this study, the efficacy of allogenic bone-marrow derived MSC on liver regeneration was analyzed in a novel porcine model of repeated biliary obstruction and partial liver resection.
The present study employed a novel porcine model of repeated biliary obstruction followed by left lobe resection in order to simulate clinical appearance and intervention of chronic cholestasis that can lead to SBC. After resection, MSC were transplanted via the portal vein, and their effects were estimated by volumetric, histological, biochemical and immunological analysis. The results show that portal vein transplantation of MSC stimulated growth of liver volume during regeneration (Figure 2
) and slightly modulated morphometric parameters of regenerated liver parenchyma (Figure 3
and Table 1
) over a 14-day period after liver resection. However, MSC transplantation did not affect liver biochemical parameters (Figure 4
) or pro-inflammatory cytokines and TGF-β (Figure 5
), which play a role in liver function and liver regeneration, respectively.
This study was the first of its kind to utilize a porcine model of intermittent biliary obstruction and benefits from its novelty on two fronts. First, even among studies that tested MSC as treatment for liver cirrhosis, none were found to use a large animal model, instead using rodents to replicate the disease [15
]. Using a porcine model of repeated biliary obstruction offers greater insight and translatability than small animal models for human disease. Second, due to commonly inducing continual biliary obstruction in rats, prior models have lacked true clinical relevance [21
]. Even the few studies found that used porcine models of biliary obstruction studied acute cholangitis and relied upon short-term (7 or 15 day) bile duct ligation at a single timepoint, similar to the previous rat models [24
While some studies have claimed benefit from successfully inducing SBC, they tend to lack translatability to humans, as very few patients suffer from such prolonged obstruction that develops into SBC. Instead, SBC is most commonly developed after repeated biliary obstruction [2
], which was achieved in this study by the successful use of Fogarty’s catheter to generate a cycle of obstruction each week. Histological analysis also demonstrated isolated areas of biliary obstruction visible in liver parenchyma (Figure 6
E,F), confirming morphological changes in the liver due to the repeated obstruction. Thus, this model better replicates clinically relevant cases of chronic human cholestasis and biliary obstruction, including those that eventually develop into SBC.
In its application, according to volumetric analysis, the model used in this study showed that MSC may serve as a viable treatment strategy to support liver regeneration. Initial results indicated that MSC significantly aided in increasing liver volume after partial resection in pigs experiencing repeated biliary obstruction. These results are in line with previous studies that have shown that MSC are capable of stimulating liver regeneration by increasing liver volume and/or weight. Adas et al. showed that MSC transplantation significantly increased liver volume, both with and without additional transfection of VEGF, after 70% resection in rats [27
]. With regards to cirrhosis, Li et al. showed that autologous MSC transplantation successfully resulted in a higher regeneration response of future liver remnant in CCL4-induced cirrhotic rats that experience portal vein embolization [16
]. Finally, in a small-scale clinical study, Mohamadnejad et al. showed that MSC successfully increased liver volume in patients with decompensated liver cirrhosis over a six-month timespan [28
However, MSC transplantation did not improve liver function based on plasma levels of liver enzymes. Similar results were obtained in studies utilizing severe liver failure models in which unchanged levels of transaminases and albumin were also detected in both MSC-treated and control groups [29
]. Plasma levels of pro-inflammatory cytokines IL-6, IL-8 and TNF-α and anti-inflammatory TGF-β also remained unaffected by MSC transplantation in our study, highlighting the questionable ability of MSC to modulate immune response upon injection into liver, where the cooperation of immune cells and MSC is still a matter for debate. Even though MSC has the potential to be delivered to the liver [31
], a possible explanation for this result is that MSC can be phagocytosed and eliminated by activated monocytes [32
]. In the liver, resident Kupffer cells can interact with injected MSC and switch into anti-inflammatory and anti-fibrotic stages, but this has only been shown in vitro or in rodent models [33
]. The anti- or pro-fibrotic behavior of MSC is also contested in current research. Although many studies showed that α-SMA expression, the volume of fibrotic tissue or hepatic stellate cell activity is ameliorated upon MSC application [34
], an increasing number of studies has found contrary results [36
The distribution of volume, nuclearity, and density within a healthy porcine liver was recently mapped [38
]. Despite extensive sampling, no differences were identified between porcine liver lobes. Therefore, we believe that the differences reported in the present study were caused by the experiment rather than by random biological variability between the various parts of the porcine liver. The mean volume of all hepatocytes in rat models of cadmium-induced hepatic damage [39
] was greater than that of the porcine mononuclear hepatocytes in our study, but similar to that of polynuclear hepatocytes. The volume of hepatocytes in our study also matches values obtained in an experiment testing the effects of cyclosporin A in rat livers [40
This study yielded interesting relationships between morphometric parameters in pigs who received MSC treatment. When compared with the control group, the MSC group showed different correlation patterns between the connective tissue fraction and volume of both mononuclear and polynuclear hepatocytes. The reduced volume fraction of hepatocytes within the liver on Day 14 after MSC transplantation was accompanied by insignificant increases in the volume fractions of the complementary fractions, i.e., the connective tissue fraction and the fraction of blood and biliary vessels. A possible explanation of this finding is that MSC transplantation can stimulate proliferation of connective tissue [36
], thus reducing the space for parenchymal hepatocytes.
Secondly, pigs in the MSC group exhibited a strong correlation between VV (hepatocytes, liver) and the fractions of mononuclear and polynuclear hepatocytes on Day 14 after resection. Specifically, the results indicate that regions containing more regenerated hepatocytes were mainly populated by mononuclear hepatocytes. A similar, but weaker, correlation was found in the control group as well. As an increase in polynuclear hepatocytes may be interpreted as a late consequence of oxidative stress [41
], these results may indicate an inclination towards a better microenvironment for hepatocyte regeneration with an increased fraction of mononuclear hepatocytes in these regions. However, further experiments would be necessary to test these relationships and to interpret the biological roles of mononuclear and polynuclear hepatocytes during liver regeneration.
These two morphological correlations present in the MSC group may also explain the differences between the volumetric and biochemical effects of MSC transplantation. While the increase in connective tissue and vessels in the liver and decrease in number of hepatocytes may have resulted in increased liver volume overall, the biochemical parameters that reflect hepatocyte function would not have changed. This may also suggest that MSC function through a paracrine effect on hepatic stellate cells, endothelial cells and cholangiocytes, a result that has been supported in previous studies as well [42
4. Materials and Methods
4.1. MSC Isolation and Culture
Porcine mesenchymal stromal cells (MSC) were isolated from healthy pigs (Sus Scrofa). Bone marrow from the tibia or femur bones was aspirated into 50 mL tubes (Techno Plastic Products, TPP, Trasadingen, Switzerland) containing heparin (B Braun, Melsungen, Germany) by puncture with a sterile needle. MSC were isolated from bone marrow by gradient centrifugation (440× g, 30 min) on Ficoll-Paque Plus (GE Healthcare, Chicago, IL, USA). The layer of mononucleated cells was washed with phosphate buffer saline and plated on a 75 cm2 culture flask (TPP) in culture media containing α-MEM cell culture media (Thermo Fisher Scientific, Waltham, MA, USA) supplemented by 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA), 1 mM L-glutamine (Biochrom, Cambridge, United Kingdom), 6.0 mg/mL penicillin, 10 mg/mL streptomycin (Biosera, Nuaille, France) and 0.25 mg/mL gentamicin (Biosera, Nuaille, France). Culture media was changed every second day. After 10 days, MSC were harvested by EDTA/Trypsin 1x (Biosera, Nuaille, France) and separated between three 75 cm2 culture flasks (TPP). Culture media was again changed every second day, and, after 10 days, MSC were harvested by EDTA/Trypsin (Biosera, Nuaille, France) and cryopreserved in liquid nitrogen (1 × 106 cells/cryotube). Four weeks before transplantation, MSC were thawed, plated on 150 cm2 flasks (TPP) in 20 mL of the culture media described above and cultured for 4 weeks to get about 50 million cells with one passage cycle, maintaining the stem cell properties of MSC. On the day of transplantation, MSC were harvested as described above, counted, re-suspended in 100 mL of physiological solution (B Braun, Melsungen, Germany) (37 °C, 106/kg of pig weight) per pig and immediately transplanted. Stem cell phenotype of transplanted MSC was identified by flow cytometry and differentiation ability.
4.2. Identification of MSC by Flow Cytometry
Before transplantation, stem cell phenotype of MSC was evaluated by flow cytometric detection of CD90, CD73 and CD44. First, 500,000 MSC were washed with PBS and stained with 5 µL of APC-CD90 (Biolegend, San Diego, CA, USA), PE-CD73 (Biolegend, San Diego, CA, USA), BV421-CD44 (Biolegend, San Diego, CA, USA) and FITC-CD45 (Bio-Rad, Hercules, CA, USA) for 15 min in the dark at room temperature. After staining, MSC were washed once, resuspended in 300 µL of PBS and analyzed by BD FACS Aria Fusion (Becton Dickinson, Franklin Lakes, NJ, USA) and BD FACS Diva 8.0.1 software (Becton Dickinson, Franklin Lakes, NJ, USA).
4.3. Adipogenic, Osteogenic and Chondrogenic Differentiation Ability of MSC
Differentiation ability of transplanted MSC was evaluated by their differentiation into adipo-, osteo- and chondro-lineage. MSC were seeded onto 12-well cultivation dishes (TPP) with a seeding density of 3.8 × 104 cells/well for adipogenic and chondrogenic differentiation and 1.9 × 104 cells/well for osteogenic differentiation in culture media. After a 24-h attachment period, the media was discarded and replaced with 3 mL of specific differentiation media: StemPro® Adipogenesis Differentiation Kit (Thermo Fisher Scientific, Waltham, MA, USA) for adipogenic differentiation, StemPro® Chondrogenesis Differentiation Kit (Thermo Fisher Scientific, Waltham, MA, USA) for chondrogenic differentiation and StemPro® Osteogenesis Differentiation Kit (Thermo Fisher Scientific, Waltham, MA, USA) for osteogenic differentiation. After the differentiation period of 21 days, oil red O (Sigma Aldrich, St. Louis, MO, USA) staining for lipid droplet visualization in adipogenesis, alcian blue (Sigma Aldrich, St. Louis, MO, USA) staining for glycoprotein visualization in chondrogenesis and alizarin red S (Sigma Aldrich, St. Louis, MO, USA) staining for calcium ion visualization in osteogenesis were performed.
4.4. Surgical Procedure and Development of Biliary Obstruction
A model for the development of biliary obstruction was prepared and performed in accordance with the laws of the Czech Republic, which are comparable with the legislature of the European Union. The number of approval was MŠMT 23097/2013-4. The experimental animal used was Sus Scrofa, about 2 months of age and approximately 20 kg. The pigs were acclimatized before each experiment.
4.4.1. Animal Preparation
Pigs (MSC group = 10 pigs; CONTROL group = 11 pigs) were starved for 12 h before surgical procedure and premedicated intramuscularly with 1.5 mg of atropine and 1.0 mg/kg of azaperone. Anesthesia was administered continuously through a peripheral venous catheter in the following average doses: 1.0 mg/kg/h of azaperone, 10 mg/kg/h of thiopental, 5–10 mg/kg/h of ketamine and 1–2 ug/kg/h of fentanyl. Muscle relaxation was provided by a bolus of 0.1–0.2 mg/kg of pancuronium at the beginning of surgery. Pigs were intubated and mechanically ventilated during surgical procedure, and electrocardiogram and oxygen saturation were monitored. The surgical procedure was performed under aseptic and antiseptic conditions. A total of 1.2 g of amoxicillin and clavulanic acid was administered as antibiotic prophylaxis in two doses (before surgery and 2 h after surgery). The abdomen was disinfected by an iodine solution, and the entire procedure was performed under sterile conditions.
4.4.2. Surgical Procedure
The surgical procedure began with collection of bone marrow in a 1:1 solution of saline and heparin from the tuberositas tibiae, which was used for MSC isolation, followed by middle laparotomy. The skin was disrupted by a scalpel and the muscle and fascia were disrupted by electrocautery. A classical retrograde cholecystectomy was performed, and the arteria hepatica and ductus choledochus were disrupted. The ductus choledochus was centrally ligated with two ligatures and fixed with Fogarty’s catheter number 6 using four plastic clips. The Fogarty’s catheter balloon was filled with 2 mL of saline. The proximal end of Fogarty’s catheter was subcutaneously passed on the right side of the pig and introduced into the implantofix. The balloon was adjusted and inflated. The laparotomy and the incision on the right side were then closed in the anatomical layers.
4.4.3. Postoperative Treatment
After the operation, the pigs were moved to a heated hutch and provided with ad libitum access to water and food. The pigs received a crushed pill of 20 mg pantoprazole each day to prevent gastroduodenal ulcus. The balloon initially remained filled for 13 days. On Day 13, the balloon was deflated under intramuscular anesthesia using 3–3.5 mL of Stresnil and 3–3.5 mL of Calypsol. Blood samples were measured for clinical and biochemical parameters (aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transferase (GGT), alkaline phosphatase (ALP), bilirubin, urea, creatinine, albumin and ammonia). The states of the bile duct and balloon were confirmed by ultrasonography (ultrasound machine Medison Sonoace 9900, convex probe with frequency 3.5 MHz). After Day 13, the Fogarty’s catheter balloon was alternately filled and emptied with saline each week for 9 weeks.
4.4.4. Liver Resection and MSC Transplantation
After 9 weeks repeatedly filling and emptying the balloon, the pigs underwent partial liver resection and MSC transplantation. General anesthesia for the procedure was used as described above. At the beginning of the procedure, an implantofix was introduced into the internal or external jugular vein and used for general anesthesia during liver resection and postoperative care and observation. Middle laparotomy was performed in the original scar after excision. The position of Fogarty’s catheter was checked, and the catheter was removed after relieving the obstruction. Liver resection was performed by removing the left lobe using the Pringle maneuver followed by reconstruction of the bile ducts by Roux-en-Y choledochojejunostomy. Then, 1.2 g of amoxicillin and clavulanic acid were administered as antibiotic prophylaxis divided into two doses (before surgery and 2 h after surgery). Blood samples were collected before the operation, during liver resection and 2 h after resection. After resection, the MSC group of pigs (n = 10) received a suspension of allogenic MSC (106 cells/kg in 100 mL of physiological solution) through the vena portae, while the control group (n = 11) received only 100 mL of physiological solution. A polypropylene mesh was used to close the laparotomy incision in the anatomical layer. Ultrasonographic controls were measured immediately after operation, and liver biopsy samples were taken for histological measurement.
4.4.5. Postoperative Observation
Experimental pigs recovered in a warm hutch. Experimental and control groups were kept under the same conditions with ad libitum access to food and water. Blood samples were collected on the 1st, 3rd, 7th, 10th and 14th day after resection under anesthesia via implantofix. Ultrasonographic controls of liver parenchyma and the bile duct were conducted at the same time points. The experiment was terminated on the 14th day after liver resection by sacrificing the pigs under general anesthesia with a concentrated solution of potassium chloride, administered via the central venous catheter. Samples of liver parenchyma were taken for histological examination.
Ultrasonographic examinations were performed at 0 h, 3 days, 7 days, 10 days and 14 days after hepatectomy. The diameters of the hypertrophic lobes were measured in B-mode in all three basic planes (axial, sagittal and coronal). The volume of the lobes was assessed by using the standard ultrasonographic formula used in human medicine (axial × sagittal × coronal/2).
4.6. Biochemical Analysis of Liver Enzymes
Blood samples were collected at 0 h, 3 days, 7 days, 10 days and 14 days after resection. Biochemical serum parameters were assessed focusing on liver function to detect the influence of the applied monoclonal antibody on the animals and to recognize possible differences between the experimental and control groups. Serum levels of albumin, bilirubin, urea, creatinine, AST, ALT, ALP and GGT were assessed by an Olympus 2700 biochemical analyzer (Olympus, Tokyo, Japan).
4.7. Quantification of Cytokines and Growth Factors in Plasma
Peripheral blood samples were collected at 0 h, 2 h, 1 day, 3 days and 14 days after resection and concentrations of IL 6, IL 8, TNF-α and TGF-β were determined by ELISA (e-Bioscience, Thermo Fisher Scientific, Waltham, MA, USA) or Luminex assay (e-Bioscience, Thermo Fisher Scientific, Waltham, MA, USA), according to manufacturer instructions.
4.8. Sampling and Histological Staining of Liver Tissue
For histological analysis, ten tissue block samples were collected from the livers of pigs who received MSC therapy, and eleven samples were collected from control pigs. All samples were taken at two time points, at the beginning of the experiment (resected liver on Day 0) and at the end (regenerated liver on Day 14). In total, 42 tissue blocks were collected and fixed in 10% phosphate-buffered formalin. The histological cutting plane of each sample was randomized using the orientator [44
]. The samples were then dehydrated, embedded onto paraffin blocks and cut into 5 µm-thick histological sections. Overall morphology was assessed using hematoxylin–eosin staining. Connective tissue was contrasted with aniline blue and nuclear fast red staining. The outlines of individual hepatocytes were stained using a combination of alcian blue and Periodic Acid Schiff (PAS) reaction. For quantitative analysis, 398 sections were used in total. From each section stained with alcian blue, one field of view with a random position was recorded using the PlanC N 4×/0.1 microscope objective in order to quantify the volume fractions of connective tissue and area of the lobules (Figure 6
A). In sections stained with PAS, systematic random uniform sampling [46
] of each fifteenth field of view was done using the PlanC N 40×/0.65 objective. This sampling resulted in 2113 micrographs used to determine the morphometry of hepatocytes. From each micrograph, hepatocytes were sampled using four unbiased counting frames (Figure 6
4.9. Quantitative Morphometric Analysis of Liver Tissue
Histological parameters listed in Table 2
were used for a quantitative analysis. The A(lobule), V(MH) and V(PH) [49
] were estimated using the nucleator stereological probe on two-dimensional images [50
]. Briefly, this method approximates the area or volume of spatial structures with geometric circles and spheres, thereby estimating their cross-sectional areas with multiple measurements of their radiuses (Figure 6
A,B). The measurement was done within the central regions of cells with clearly visible nuclei and nucleoli. Only cells selected by the unbiased counting frames were measured. From the measurements of these cross-sectional areas, the volume of hepatocytes was estimated. The volume fractions VV
(connective tissue, liver), VV
(hepatocytes, liver), VV
(MH, liver) and VV
(PH, liver) were determined using the point grid method [51
]. All quantitative analyses were done using Ellipse stereological software (ViDiTo, Kosice, Slovakia).
4.10. Statistical Analysis
The MSC and control groups contained 10 and 11 individual pigs, respectively. All experimental data are expressed as mean ± standard deviation. The Shapiro–Wilk test was used to ascertain data normality. The statistical significance of differences between the MSC and control groups was determined by the Wilcoxon matched pairs test (p < 0.05) for histological analysis and by the Mann–Whitney U test of cross-ranked values (p < 0.05) for the remaining analyses. All statistical analyses were performed in Statistica v12 software (Tibco Software, Palo Alto, CA, USA).
Our study established the first large animal (porcine) model of repeated biliary obstruction by using Fogarty’s catheter to generate cyclic obstruction. In this model, allogenic MSC transplantation via the portal vein improved liver regeneration by means of liver volume growth, accompanied by increased mononuclear hepatocyte content in regenerated liver tissue. Liver function, analyzed by serum levels of liver enzymes (AST, ALT, ALP and GGT) and liver metabolites (albumin, bilirubin, urea and creatinine), was not affected by MSC transplantation. Even plasma pro-inflammatory cytokines (IL-6, IL-8 and TNF-α) and TGF-β remained unchanged upon MSC transplantation. This study underscores the impact of diverse experimental setups, especially large animal models, prior to clinical trials, as well as the need for careful evaluation of MSC application in liver disease therapy.