Study of Recellularized Human Acellular Arterial Matrix Repairs Porcine Biliary Segmental Defects

Abstract

Background:

With the popularity of laparoscopic cholecystectomy, common bile duct injury has been reported more frequently. There is no perfect method for repairing porcine biliary segmental defects.

Methods:

After the decellularization of human arterial blood vessels, the cells were cultured with GFP⁺ (carry green fluorescent protein) porcine bile duct epithelial cells. The growth and proliferation of porcine bile duct epithelial cells on the human acellular arterial matrix (HAAM) were observed by hematoxylin–eosin (HE) staining, electron microscopy, and immunofluorescence. Then, the recellularized human acellular arterial matrix (RHAAM) was used to repair biliary segmental defects in the pig. The feasibility of it was detected by magnetic resonance cholangiopancreatography, liver function and blood routine changes, HE staining, immunofluorescence, real-time quantitative PCR (RT-qPCR), and western blot.

Results:

After 4 weeks (w) of co-culture of HAAM and GFP⁺ porcine bile duct epithelial cells, GFP⁺ porcine bile duct epithelial cells grew stably, proliferated, and fused on HAAM. Bile was successfully drained into the duodenum without bile leakage or biliary obstruction. Immunofluorescence detection showed that GFP-positive bile duct cells could still be detected after GFP-containing bile duct cells were implanted into the acellular arterial matrix for 8 w. The implanted bile duct cells can successfully resist bile invasion and protect the acellular arterial matrix until the newborn bile duct is formed.

Conclusion:

The RHAAM can be used to repair biliary segmental defects in pigs, which provides a new idea for the clinical treatment of common bile duct injury.

Introduction

With the popularization of the laparoscopic technique, laparoscopic cholecystectomy (LC) has been popularized in hospitals all over the country because of its advantages such as less trauma, faster recovery and shorter hospitalization time. However, due to the spread of this technology, common bile duct defect has increased. The incidence of common bile duct defect in LC is 0.1–0.5% [1–4]. The total mortality rate of LC was 0.45%, but if the common bile duct defect occurred, the total mortality rate could be as high as 9% [1, 5, 6]. In the case of common bile duct defect, the patient’s long-term survival rate [7] and quality of life [8, 9] will decline. Although the Choledochojejunostomy can solve the problem of the stricture and obstruction of the bile duct in some patients, due to the loss of the function of the Oddi sphincter, the biliary tract will be narrowed again, and long-term medication should be taken after the operation. Complications such as reflux cholangitis, gallstones, and cholangiocarcinoma also appear [10]. Currently, 70% of congenital biliary atresia is treated by liver transplantation [11], while primary sclerosing cholangitis accounted for only 5% [12] of all liver transplants in the United States. Conversely, biliary tract complications are the main cause of liver transplantation failure [13, 14]. Biliary tube replacement is a potentially effective treatment but is currently hampered by the lack of suitable healthy tissues. Fortunately, tissue engineering can solve the historical problem [15]. In clinical work, we found that some arterial tissues such as the splenic artery, and part of the abdominal aorta, can be obtained from the cadaver. After satisfying the needs of vascular reconstruction, the remaining parts are frozen and preserved. Determining the use of these remaining tissues after liver transplantation has become the focus and hotspot of tissue engineering research.

Materials and methods

Preparation of HAAM

The human artery used in this experiment was received from donors after cardiac death who volunteered to donate organs between March 2017 and December 2018 in the Ganmei Hospital affiliated to Kunming Medical University. Informed consent was confirmed by the Ethics Committee of Ganmei Hospital affiliated to Kunming Medical University. Briefly, the obtained human splenic artery tissues were stored in University of Wiscon-sin (UW) solution (Bristol-Myers Squibb, Shanghai, China) under aseptic conditions and placed in a − 80 refrigerator (Aucma, Qingdao, Shandong, China). Before the experiment, it was reheated and cut into a segment of approximately 2 cm and the extra adventitia tissue around the arterial tissue was removed passively under sterile conditions. After that, a 5 ml 0.25% trypsin solution (Gibco, Grand Island, CA, USA) was added and placed in the orbital shaker (Thermo, Waltham, MA, USA) at 37 °C for 4 h (h) with a speed of 100 rpm. Then, 10 ml 1% Triton X-100 solution (Biosun, Shanghai, China) was added and shook with a frequency of 100 rpm for 24 h. Finally, 10 ml 1% sodium dodecyl sulfate (SDS) solution (Sunshinebio, Nanjing, Jiangsu, China) was added and shook at 100 rpm for 24 h.

Lentivirus-mediated GFP gene transfection into porcine bile duct epithelial cells

In order to detect the survival of porcine bile duct epithelial cells, a lentivirus-mediated GFP gene was used to label the cells. Porcine bile duct epithelial cells (CELLBIO, Shanghai, China) were briefly placed in HG-DMEM (Hyclone, Los Angeles, CA, USA) supplemented with 10% FBS (Gibco, Grand Island, CA, USA), 100u/ml penicillin (Lukang, Jining, Shandong, China) and 0.1 g/ml streptomycin (Hyclone, Los Angeles, CA, USA). Cells were cultured in a 37 °C moist incubator (Thermo, Waltham, MA, USA) containing 5% CO₂. The culture medium was changed every other day and the adherent cells were diluted at 1:4 every 5 or 7 days. The multiplicity of infection (MOI) value of the lentivirus vector (Hanbio, Shanghai, China) with the GFP reporter gene was determined to be 10.5%. Porcine bile duct cells with a density of 5 × 10⁵/ml were prepared in complete medium (Hanbio, Shanghai, China). After 24 h culture, 1 × 10⁸ virus solution (Hanbio, Shanghai, China) was added into the cells and then polybrene (Hanbio, Shanghai, China) was added to the cell to make the final concentration reach 5 μg/ml. After 4 h of small volume infection at 37 °C, the medium was supplemented to normal volume 4 h later. Finally, the virus-containing culture medium (Hanbio, Shanghai, China) was sucked out and replaced with fresh complete culture medium (Hanbio, Shanghai, China), which was further cultured at 37 °C.

Co-culture of HAAM and Porcine Bile duct epithelial cells

The human acellular arterial matrix (HAAM) was placed in a culture dish (Corning, Corning, NY, USA), and the whole layer of HAAM was cut along its long axis and spread over a culture dish (Corning, Corning, NY, USA). It was then trimmed into a 2 cm x 2 cm square. After that, a 5 ml DMEM/high glucose cell culture medium (Hyclone, Los Angeles, CA, USA) was added and placed in a 37 °C incubator (Hanbio, Shanghai, China) for 24 h. 200 μl of porcine bile duct epithelial cell suspension (about 1 × 10⁶ marked cells) was obtained and slowly dropped onto the surface of the acellular artery matrix. The plate (Corning, Corning, NY, USA) was covered and incubated at 37 °C, and the fluid was changed every 48 h. The growth and proliferation of porcine bile duct epithelial cells on the human acellular arterial matrix were observed at 1 w, 2 w, 3 w, and 4 w.

CCK8

In order to detect the survival rate of porcine bile duct epithelial cells transfected with lentivirus-mediated GFP, we performed a CCK8 assay. In briefly, porcine bile duct cells were divided into two groups. Group A was lentivirus-mediated GFP-transfected bile duct cells and group B was normal bile duct cells. At passages 1, 3, 5, 7, 9, 11, 13, and 15 some cell suspensions were taken from the cells of the A and B group respectively and the cell concentration was adjusted to l × l0⁴/ml to inoculate the cells on the 96-well plate (Corning, Corning, NY, USA) at 5 × l0² thickness. Each generation of cells was inoculated into 7 groups with 3 holes in each group. The culture plates (Corning, Corning, NY, USA) were taken out at 0, 1, 2, 3, 4, 5, and 6 days after inoculation. 10 ul of CCK8 reagent (Hanbio, Shanghai, China) was added, and mixed fully. The orifice plate (Corning, Corning, NY, USA) was covered and placed in 37 °C, 5% CO₂ incubator (Hanbio, Shanghai, China), and cultured for 4 h. The orifice plate (Corning, Corning, NY, USA) was taken out and placed into the enzyme marker (Corning, Corning, NY, USA) with an adjusted reference wavelength of 600 nm. The absorbance value was detected at 450 nm wavelength and the average value was calculated.

Reconstruction of porcine biliary segmental defects with RHAAM

Twenty-four southern Yunnan small ear pigs were purchased from the laboratory animal center of Kunming Medical University which was approved by the Ethics Committee of Ganmei Hospital affiliated to Kunming Medical University (2015-01). The pigs were randomly divided into the RHAAM group and Control group (n = 12). The Control group was repaired with HAAM, and the PHAAM group was repaired with recellularized human acellular arterial matrix (RHAAM). The pre-observation was performed at 1, 2, 4, and 8 weeks. Three animals were observed at each time point. The procedure is as follows: After general anesthesia with 3% pentobarbital solution (Solarbio, Shanghai, China) by a slow intravenous drip of 20 mg/kg from the dorsal auricular vein, the animals were fixed to the operating table and sterilized. After the midline incision of the abdomen was taken and the position of the common bile duct was determined, 1.5 cm of common bile duct was cut off. Then, the HAAM or RHAAM was anastomosed to the common bile duct by end-to-end with a 6 0 prolene line (Johnson & Johnson, Shanghai, China). After the operation, animals were raised in a single cage to observe their activity and diet for convenience, cefminox (Zhejiang Apeloa Tospo Pharmaceutical Co., Ltd, Dongyang, Zhejiang, China) were given at the first 3 days and cyclosporine soft capsule (10 mg/kg, Livzon, Zhuhai, Guanddong, China) were given daily. Magnetic resonance cholangiopancreatography (MRCP) was performed on the 1st, 2nd, 4th, and 8th w after the experiment. Blood samples were collected from the inferior vena cava for blood analysis and liver function tests. Finally, the animals were euthanized and the specimens were obtained.

MRCP

In order to clarify the effect of bile duct repair, after the 1st, 2nd, 4th and 8th w of general anesthesia, magnetic resonance cholangiopancreatography (MRCP) was performed and the results of the bile duct repair were observed.

Liver function and blood routine test

In order to investigate the changes of the liver after bile duct repair, we performed the liver function and blood routine examination. Before the animals were euthanized, the blood of the auricular vein was taken and immediately sent to Ganmei Hospital. The changes of glutamic pyruvic transaminase (ALT), glutamic oxaloacetic transaminase (AST), alkaline phosphatase (ALP), glutamyl transaminase (GGT), total bilirubin (TBIL), direct bilirubin (DBIL), albumin (ALB), total bile acid (TBA), hemoglobin content (HB) and white blood cell count (WBC) were analysed by YZ200705-963 automatic biochemical analyzer (Roche, Basle, Switzerland).

Observation of histomorphological changes

To examine the changes of common bile duct after bile duct repair, we observed the histological changes of the common bile duct. The tissues were cut into small pieces of 1.0 cm × 1.0 cm, fixed in 10% formaldehyde solution (Jiacheng group, Shenyang, Liaoning, China) for 48 h, and embedded in paraffin wax (Shanghai Huayong Olefin Co., Ltd, Shanghai, China). 5um continuous tissue slices were used for HE staining. For Immunofluorescence analysis, paraffin sections were dewaxed and hydrated and placed in a box filled with EDTA antigen repair buffer (PH8.0, Servicebio, Wuhan, Hubei, China) to repair the antigen in a microwave oven (Galanz, Foshan, Guangdong, China). After that, autofluorescence quenching agent (Servicebio, Wuhan, Hubei, China) was added, and the sample was sealed with 5% BSA (Servicebio, Wuhan, Hubei, China). An anti-incubation (Servicebio, Wuhan, Hubei, China) was added and kept overnight at 4 °C. In order to stain the nucleus, the second antibody (Servicebio, Wuhan, Hubei, China) was added and incubated at room temperature for 50 min, then incubated overnight with the antibody diluent (Servicebio, Wuhan, Hubei, China) and 4',6-diamidino-2-phenylindole (DAPI) (Servicebio, Wuhan, Hubei, China). After washing, the samples were incubated at room temperature for 50 min, DAPI dye solution was dropped in and incubated at room temperature for 10 min. Antibody (Servicebio, Wuhan, Hubei, China) was added as recommended by the manufacturer. Finally, the slices were observed under a positive fluorescence microscope (Nikon, Tokyo, Japan) and the images were collected.

Real-time quantitative polymerase chain reaction (RT-qPCR)

PCR was used to investigate the effect of bile duct repair. Total RNA was extracted from 100 mg fresh tissue and cDNA was synthesized by reverse transcription kit (Servicebio, Wuhan, Hubei, China). The PCR reaction was carried out with a synergy brands (SYBR) green real-time fluorescent quantitative PCR kit (Servicebio, Wuhan, Hubei, China). During the process, three holes were set up in each sample. The reaction conditions of RT-qPCR are as follows: pre-denaturation for 10 min at 95 °C, denaturation for 15 s (s) at 95 °C, and annealing at 60 °C for 60 s for a total of 40 cycles. It was heated at 0.3 °C every 15 s until the temperature reached 95 °C, the experiment was repeated 4 times. Taking the expression of actin mRNA as the intel RNA reference, the results were expressed as a 2^−ΔΔCt value, and the relative expression of TNFAIP8 mRNA in the cells of each group was analyzed by Rotor-Gene 6000 1.7 (Corbett, Sydney, NSW, Australia). The primer sequence (5′–3′) is shown in Table 1.

Table 1.

Primer sequence of RT-QPCR

pig-CK7(rz)-S	CGTGGTGCTGAAGAAGGATGTG
pig-CK7(rz)-A	GAGGGTCTCAAACTTGGTCTGGTA
pig-CK19-S	AAGTTCGAGACAGAGCAGGC
pig-CK19-A	CGCTCAGGATCTTGGCTAGG
pig-GGT1(rz)-S	TCTACAATGGGAGCCTCACCG
pig-GGT1(rz)-A	ACAATGCGGTGGTAGGTCAGG
pig-ACTIN-S	TGGTTCTGGGCTCTGTAAGGC
pig-ACTIN-A	TGATGCCGTGTTCTATTGGGTA

Western blot

WB was also used to detect the effect of bile duct repair. Firstly, the tissue mass was collected and cleaned 2–3 times with cold PBS (Hyclone, Los Angeles, CA, USA). When the blood was removed, the pieces were cut into smaller pieces and placed in the homogenized pipe (Servicebio, Wuhan, Hubei, China). Secondly, 2 small magnetic beads (Servicebio, Wuhan, Hubei, China) of 2 mm and the radio immunoprecipitation assay (RIPA) reagent (Servicebio, Wuhan, Hubei, China) were added and put in the homogenizer (Servicebio, Wuhan, Hubei, China), for 60 s. The homogenized sample tube (Servicebio, Wuhan, Hubei, China) was removed, frozen for 30 min, and vibrated every 5 min to ensure that the tissue is completely cracked. The sample was centrifuged at 2000 rpm for 10 min, supernatant was then collected and stored at − 80 °C for reserve. Next, the concentration of protein was detected by the bicinchoninic acid (BCA) method. The protein was obtained by 10% SDS-PAGE (Sunshinebio, Nanjing, Jiangsu, China), then the separated protein was transferred onto the polyvinylidene fluoride (PVDF) membrane (Servicebio, Wuhan, Hubei, China) by electroporation. The protein was sealed with 5% skimmed milk (Servicebio, Wuhan, Hubei, China) powder for 1 h and diluted with the primary antibody (Servicebio, Wuhan, Hubei, China) to incubate overnight at 4 °C. On the next day, after the PVDF membrane (Servicebio, Wuhan, Hubei, China) was washed fully with PBST buffer (Servicebio, Wuhan, Hubei, China), the goat anti-rabbit IgG labeled with horseradish peroxidase (Servicebio, Wuhan, Hubei, China, volume dilution ratio was 1:3000) was added and reacted at room temperature for 1 h. Finally, the electrochemiluminescence (ECL) kit (Servicebio, Wuhan, Hubei, China) was used to color the membrane. This experiment was repeated three times.

Statistical analysis

We performed data analysis using SPSS version 21.0 (SPSS Inc, Chicago, IL, USA). The results were expressed as mean ± standard deviation. One-way ANOVA was used to compare liver enzymes, bilirubin and blood routine differences between the Control group and the RHAAM group, and the student’s t test was used to compare the variables between the Control group and RHAAM group in western blot (WB) and PCR. P < 0.05 value was considered to be statistically significant. *P < 0.05, **P < 0.01.

Results

Successful preparation of HAAM

Before treatment the human splenic artery was slightly reddish, the wall was elastic, the lumen did not collapse, and the surface of the intima was smooth (Fig. 1). By decellularization, the splenic artery gradually changed from light red to porcelain white translucent jelly-like tubular structure (Fig. 1C). HE staining showed that there was no cellular component in HAAM compared with the splenic artery (Fig. 1B, D). It is indicated that the human acellular arterial matrix has been successfully prepared.

Fig. 1.

Preparation of HAAM. A The picture of bright field of normal human splenic artery matrix. B HE staining of normal human splenic artery matrix. C The pictures of the bright field of HAAM. D HE staining of HAAM. HAAM: human acellular arterial matrix; HE staining: hematoxylin–eosin staining. (scale bar = 200 μm)

Lentivirus-mediated GFP gene transfected porcine bile duct cells successfully

It was observed that the bile duct epithelial cells showed a round and fusiform morphology, and the cells were small, the arrangement was regular, the nucleus was obvious, the cells were closely arranged and uniform (Fig. 2A). The expression of transfected porcine bile duct epithelial cells was observed by fluorescence microscope. The results showed that the green fluorescent protein (GFP) still expressed in the 20th passage and the morphology of cells did not change significantly (Fig. 2B). The results of CCK8 showed that the proliferation curve of the transfected porcine bile duct epithelial cells was not significantly different from that of the primordial cells (Fig. 2C). These results indicated that lentivirus-mediated GFP gene was successfully transfected into porcine bile duct epithelial cells.

Fig. 2.

Lentivirus-mediated GFP gene transfection into porcine bile duct epithelial cells. A Bile duct epithelial cells observed under the microscope. (scale bar = 50 μm). B Expression of GFP in transfected porcine bile duct epithelial cells observed by fluorescence microscope. GFP: green fluorescent protein. (scale bar = 50 μm). C The proliferation curve of normal porcine bile duct epithelial cells (a). and transfected porcine bile duct epithelial cells (b)

Successful co-culture of HAAM and porcine bile duct epithelial cells

The results of HE staining, frozen sections, scanning electron microscopy, CK-19 immunofluorescence staining, and DAPI staining suggests that in the first 4 weeks, the number of porcine bile duct epithelial cells on the human acellular artery matrix gradually increased, and gradually fused and joined to form a sheet (Figs. 3, 4). It confirmed that the co-culture of the human acellular arterial matrix and porcine bile duct epithelial cells was successful.

Fig. 3.

RHAAM. A HE staining of RHAAM in 1 w, 2 w, 3 w, and 4 w. B Frozen section of RHAAM in 1 w, 2 w, 3 w, and 4 w. C Scanning electron microscope images of RHAAM in 1 w, 2 w, 3 w, and 4 w. RHAAM: recellularized human acellular arterial matrix; HE staining: hematoxylin–eosin staining; w week (s). (scale bar = 200 μm)

Fig. 4.

CK-19 immunofluorescence staining and DAPI staining of RHAAM in 1 w, 2 w, 3 w, and 4 w. CK-19 immunofluorescence staining: cytokeratin 19 immunofluorescence staining; DAPI staining: 4′,6-diamidino-2-phenylindole staining; w week (s). (scale bar = 200 μm)

Reconstruction of porcine biliary tract defect with HAAM

Survival of experimental animals in different groups

After reconstruction of porcine biliary tract defect with HAAM (Fig. 5), the animals of the RHAAM group were in good condition, and ate normally on the first day after surgery. All of the pigs gained weight and survived until the predetermined time. In the Control group, one pig died due to respiratory arrest caused by excessive anesthesia. The other animals passed through the operation successfully. However, three pigs we observed which suffered from poor spirits and poor diet since the 7th week, died in week 8. It was found that scar hyperplasia of common bile duct resulted in complete blockage of the bile duct, severe dilation of the proximal bile duct, hepatic cholestasis, and bile duct infection. The remaining 8 animals naturally survived to a predetermined time point.

Fig. 5.

RHAAM repairs the segmental defect of the porcine bile duct. A The pictures of the bright field after reconstruction of segmental defect of porcine bile duct with RHAAM. B The pictures of the bright field of RHAAM. RHAAM: recellularized human acellular arterial matrix. (scale bar = 1.5 cm)

Liver function and blood routine are normal

Table 1 shows the values of hemoglobin (HGB), white blood cell (WBC), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), glutamyl transpeptidase (GGT), total bilirubin (TBil), direct bilirubin (DBil), and total bile acid (TBA) at 1 w, 2 w, 4 w, and 8 w after surgery. Among them, ALT and AST were used to monitor the liver function after the operation. ALP, GGT, TBil, DBil, and TBA reveals the incidence of obstructive jaundice. Concurrently, WBC is an indicator of infection, and HGB was usually used to monitor anemia. The results showed that there was no obstructive jaundice and liver function injury in the RHAAM group, and the blood routine was also within normal range (Table 2).

Table 2.

Changes in liver enzymes, bilirubin, and blood routine

	1 w		2 w		4 w		8 w
	RHAAM group (N = 3)	Control group (N = 2)	RHAAM group (N = 3)	Control group (N = 3)	RHAAM group (N = 3)	Control group (N = 3)	RHAAM group (N = 3)	Control group
ALT (U/L)	63.67 ± 9.29	135.50 ± 52.5	58.67 ± 3.79	366.00 ± 307.94	87.67 ± 5.69	243.33 ± 245.04	71.50 ± 13.44	±
AST (U/L)	64.33 ± 12.50	491.00 ± 335.00	118.67 ± 35.39	369.67 ± 466.68	101.67 ± 62.50	227.33 ± 285.57	109.00 ± 91.92	±
ALP (U/L)	74.33 ± 15.57*	347.50 ± 145.5	55.67 ± 6.81	345.00 ± 181.58	59.67 ± 7.09	233.00 ± 165.93	103.50 ± 54.45	±
GGT (U/L)	74.67 ± 22.50*	459.50 ± 82.50	87.00 ± 38.12	463.67 ± 243.70	68.67 ± 52.79	153.67 ± 159.66	90.00 ± 2.83	±
TBIL (umol/L)	1.93 ± 0.98*	13.35 ± 2.35	0.40 ± 0.26*	137.27 ± 17.46	0.75 ± 0.25*	3.03 ± 0.90	0.45 ± 0.07	±
DBIL (umol/L)	1.50 ± 1.22*	9.10 ± 0.80	0.17 ± 0.06*	103.00 ± 12.24	0.40 ± 0.10*	1.45 ± 0.45	0.30 ± 0.00	±
ALB (g/L)	29.37 ± 14.17	11.70 ± 0.80	31.87 ± 18.45	13.07 ± 1.12	30.83 ± 13.45	14.17 ± 1.97	42.95 ± 6.29	±
HGB (g/L)	161.00 ± 18.03	138.00 ± 10.00	192.00 ± 32.91	157.33 ± 9.07	201.67 ± 18.34*	132.67 ± 17.04	200.50 ± 24.75	±
WBC (109/L)	19.69 ± 2.96	21.97 ± 1.07	15.92 ± 4.54	12.59 ± 3.71	13.31 ± 2.47	16.88 ± 4.39	21.77 ± 3.18	±

Scarring is less and bile duct is smooth

In the Control group, from week 1 to week 8, the acellular arterial matrix melted gradually and the scar increased gradually. The common bile duct was completely blocked and the proximal bile duct was dilated severely in the 8th w. In the RHAAM group, from the 1st week to the 7th week, the RHAAM was fused with the surrounding tissues gradually, the scar proliferation decreased gradually, and the common bile duct was completely unobstructed. The RHAAM was completely fused with surrounding tissues with no difference between them. Only a residual non-absorbable suture can be seen, there is a small amount of scar tissue around the non-absorbable suture, and the bile duct is completely unobstructed (Fig. 6).

Fig. 6.

Morphological changes of the bile duct in the RHAAM group and Control group at 1 w, 2 w, 4 w, and 8 w. RHAAM: recellularized human acellular arterial matrix; w week (s). (n = 12/group, scale bar = 1.5 cm)

After the repair of RHAAM, no dilation in the bile duct

From week 1 to week 4, there was no obvious dilation in the bile duct, the acellular matrix segment was not developed, and the bile duct was well developed in the Control group. At the 2nd week, magnetic resonance cholangiopancreatography (MRCP) showed mild dilation of the bile duct, the acellular matrix segment was not developed and the lower end of the bile duct was normally displayed. In the 4th week, there was a serious dilatation above the acellular matrix, the gallbladder was seriously enlarged, the acellular matrix segment and the lower end of the common bile duct were not developed. In the 8th week, the MRCP examination was not performed in the Control group due to the death of all the animals. For the RHAAM group, from the 1st week to the 8th week, MRCP showed that there was no obvious dilation in the bile duct, the acellular matrix segment was developed, and the bile duct was well developed (Fig. 7).

Fig. 7.

MRCP inspection of the Control group and RHAAM group in 1 w, 2 w, 4 w, and 8 w. MRCP: magnetic resonance cholangiopancreatography; RHAAM: recellularized human acellular arterial matrix; w week (s). (n = 12/group, scale bar = 1.5 cm)

Repair of porcine bile duct

From the first week to the 7th week of the Control group, there was a gradual increase of receptor cells in the HAAM, inflammatory cell infiltration gradually increased, and the HAAM dissolved gradually. DAPI staining and CK19 immunofluorescence staining showed more red-stained CK19 + cells, but no glandular structure (Fig. 8). In the RHAAM group, from the first week to the 7th week, it was observed that the receptor cells gradually increased in RHAAM, inflammatory cell infiltration gradually reduced, and the glandular structure gradually formed. DAPI staining and immunofluorescence staining of CK19 and GFP showed that the glandular structure was mostly composed of red-stained CK19 + and green-stained GFP + cells. At the 8th week, the HAAM was completely degraded and a large number of regular glands were formed in the RHAAM group, which was no different from the normal bile duct tissue (Fig. 9).

Fig. 8.

HE staining, DAPI staining and CK19 immunofluorescence staining in bile duct tissue of Control group and RHAAM group in 1 w, 2 w, 4 w, and 8 w. HE staining: hematoxylin–eosin staining; DAPI staining: 4′,6-diamidino-2-phenylindole staining; CK19 immunofluorescence staining: cytokeratin 19 immunofluorescence staining; w week (s). (n = 12/group, scale bar = 200 μm)

Fig. 9.

HE staining, DAPI staining, CK19 and GFP immunofluorescence staining in bile duct tissue of RHAAM group in 1 w, 2 w, 4 w, and 8 w. HE staining: hematoxylin–eosin staining; DAPI staining: 4′,6-diamidino-2-phenylindole staining; CK19 immunofluorescence staining: cytokeratin 19 immunofluorescence staining; GFP immunofluorescence staining: green fluorescent protein immunofluorescence staining; w week (s). (n = 12/group, scale bar = 200 μm)

WB Detection and Real-time quantitative PCR Detection

The expression of GGT, CK19, CK7 protein and its mRNA in bile duct epithelial cells increased gradually over time. Compared with the Control group, there was a significant difference (Fig. 10). At week 8, all the experimental animals died in the control group, and the data was not available.

Fig. 10.

Detection of GGT, CK19 and CK7 expression in bile duct epithelial cells. A Protein bands OF GGT, CK19, and CK7. B The relative intensity of CK7, CK19, and GGT. C Proliferation index of CK7, CK19, GGT1. (n = 12/group, *p < 0.05)

Discussion

In this study, the lentivirus-mediated GFP gene was successfully transfected into porcine bile duct epithelial cells. It was clearly expressed in the transfected porcine bile duct epithelial cells and continued more than 20 passages. After transfection, the morphology of cells did not change significantly. The results of CCK8 showed that the proliferation curve of the transfected porcine bile duct epithelial cells was not obviously different from that of the primordial cells. It lays a solid foundation for the marking of seed cells and follow-up research. Clinically, gallbladder tissue is often discarded after liver transplantation and cholecystectomy. We can not only get bile duct epithelial cells from the discarded gallbladder, but also we can use minimally invasive surgery such as endoscopic retrograde cholangiopancreatography (ERCP) to enter the common bile duct and get bile duct cells by cell brushes. Furthermore, through the use of three-dimensional culture system, we can get the seed cells that we need [16]. Our results proved that porcine bile duct epithelial cells can grow stably in HAAM, it also can proliferate normally and fuse with an acellular arterial matrix.

Our data showed that in 8 weeks after the operation, all the pigs in the RHAAM group survived successfully, and that there was no obvious biliary obstruction in pigs. MRCP showed no obvious bile leakage or biliary stricture. Our acellular vascular stent successfully drained bile to the duodenum. After the GFP-containing bile duct cells were implanted into the acellular arterial matrix, GFP-positive bile duct cells could still be detected by histology and immunofluorescence assay for 8 weeks. This indicates that our bile duct cells are successfully implanted, and the implanted bile duct cells can successfully resist the invasion of bile and thus protect the acellular arterial matrix until the newborn bile duct is formed. In the Control group, mild to severe obstructive jaundice, cholangitis and extrahepatic bile duct dilatation were observed in the experimental animals from the 2nd week. This may be due to the acellular matrix being immersed in bile and degrading from the 2nd week. After degradation, bile corrodes the surrounding tissue to cause scar hyperplasia and obstruction, resulting in a series of consequences such as obstructive jaundice, infection, and so on.

With the increase in the number of liver transplants, donor arteries are more readily available. The diameter of the human artery is diverse, which can meet the needs of different diameter tissues. In the process of decellularization, the human artery produces a 3-D structure, which is basically composed of collagen fibers, so it is well tolerated in immunology [17]. Its loose intima and media are more favorable to seed cell implantation and crawling, which can repair the corresponding tissue damage. The acellular matrix of blood vessels contains more elastic fibers, mainly composed of elastin [18]. Contact with the lumen cells of the blood vessels (i.e. endothelium) or extrahepatic bile duct (i.e. the bile duct epithelium) is the basement membrane. Although the vascular basement membrane may have a specific composition different from that of the extrahepatic bile duct, the general structure should be comparable, including laminin, IV collagen, and fibronectin [18–20]. Therefore, in our opinion, cells should be able to interact with the basement membrane and transplant there, even if they come from different individuals. Other components found in the acellular matrix are adhesion molecules (such as fibronectin and laminin), proteoglycans, glycoproteins and various growth factors (e.g., transformation growth factor-b, basic fibroblast growth factor, and vascular endothelial growth factor) [21]. It has been shown that some of these growth factors maintain their biological activity even after final sterilization and long-term storage [22]. These growth factors promote cell adhesion and differentiation that repair injured tissue and prevent scar formation.

It is worth noting that the acellular arterial matrix of replanted bile cells ensures smooth drainage of bile from the intrahepatic bile duct to the duodenum within 8 weeks after transplantation, and obtains a function and shape similar to that of the primary bile duct. However, the long-term effect and the repair structure of common bile duct long segment defect are still in need of further experimental study.

Author contributions

Wei Liu drafting the article, a substantial contribution to the conception and design of the study. Sheng-Ning Zhang, Zong-Qiang Hu, Shi-Ming Feng, Zhen-Hui Li, Shu-Feng Xiao, and Hong-Shu Wang performed the research and contribution to acquisition, interpretation, and analysis of data. Li Li revising the draft critically and final approval of the version to be published.

Funding

The study was supported by National Natural Science Foundation of China (No. 81560089) and Yunnan Province Health Science and Technology Project (No.2014NS198).

Compliance with ethical standards

Conflicts of interest

The authors have no financial conflicts of interest.

Ethical statement

The study protocol was approved by the institutional review board of Ganmei Hospital affiliated to Kunming Medical University. Informed consent was confirmed by the Ganmei Hospital affiliated to Kunming Medical University. The animal studies were performed after receiving approval of the Ethics Committee of Ganmei Hospital affiliated to Kunming Medical University (2015-01).