Whole-organ tissue engineering: Decellularization and recellularization of three-dimensional matrix liver scaffolds Shabnam Sabetkish,1 Abdol-Mohammad Kajbafzadeh,1 Nastaran Sabetkish,1 Reza Khorramirouz,1 Aram Akbarzadeh,1 Sanam Ladi Seyedian,1 Parvin Pasalar,2 Saghar Orangian,1 Reza Seyyed Hossein Beigi,1 Zahra Aryan,1 Hesam Akbari,1 Seyyed Mohammad Tavangar3 1
Pediatric Urology Research Center, Section of Tissue Engineering and Stem Cells Therapy, Department of Pediatric Urology, Children’s Hospital Medical Center, Tehran, Iran (IRI) 2 Department of Medical Biochemistry, Tehran University of Medical Sciences, Tehran, Iran (IRI) 3 Department of Pathology, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran (IRI) Received 21 June 2014; accepted 3 July 2014 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35291 Abstract: To report the results of whole liver decellularization by two different methods. To present the results of grafting rat and sheep decellularized liver matrix (DLM) into the normal rat liver and compare natural cell seeding process in homo/xenograft of DLM. To compare the results of in vitro whole liver recellularization with rats’ neonatal green fluorescent protein (GFP)-positive hepatic cells with outcomes of in vivo recellularization process. Whole liver of 8 rats and 4 sheep were resected and cannulated via the hepatic vein and perfused with sodium dodecyl sulfate (SDS) or Triton 1 SDS. Several examinations were performed to compare the efficacy of these two decellularization procedures. In vivo recellularization of sheep and rat DLMs was performed following transplantation of multiple pieces of both scaffolds in the subhepatic area of four rats. To compare the efficacy of different scaffolds in autologous cell seeding, biopsies of homograft and xenograft were assessed 8 weeks postoperatively. Whole DLMs of 4 rats were also recellularized in vitro by perfusion of rat’s fetal GFP-positive hepatic cells with pulsatile bioreactor. Histological evaluation and enzymatic assay were
performed for both in vivo and in vitro recellularized samples. The results of this study demonstrated that the triton method was a promising decellularization approach for preserving the three-dimensional structure of liver. In vitro recellularized DLMs were more similar to natural ones compared with in vivo recellularized livers. However, homografts showed better characteristics with more organized structure compared with xenografts. In vitro recellularization of liver scaffolds with autologous cells represents an attractive prospective for regeneration of liver as one of the most compound organs. In vivo cell seeding on the scaffold of the same species may have more satisfactory outcomes when compared with the results of xenotransplantation. This study theoretically may pave the road for in situ liver regeneration probably by implantation of homologous DLM or in vitro C 2014 recellularized scaffolds into the diseased host liver. V Wiley Periodicals, Inc. J Biomed Mater Res Part A: 00A:000–000, 2014.
Key Words: liver, extra cellular matrix, decellular, tissue engineering, transplantation
How to cite this article: Sabetkish S, Kajbafzadeh A-M, Sabetkish N Khorramirouz R, Akbarzadeh A, Seyedian SL, Pasalar P, Orangian S, Beigi RSH, Aryan Z, Akbari H, Tavangar SM. 2014. Whole-organ tissue engineering: Decellularization and recellularization of three-dimensional matrix liver scaffolds. J Biomed Mater Res Part A 2014:00A:000–000.
INTRODUCTION
End stage liver diseases including cirrhosis, chronic viral hepatitis, hepatocellular carcinoma, injuries from alcohol abuse, or even inborn metabolic disorders, often lead to demands for organ transplantation.1,2 As a result, many of these patients may die before an appropriate organ becomes achievable.2 Due to increasing number of patients on the waiting list for transplantation, short donor organ, and a need for life-long immunosuppressive remedy, tissue engineering has been developed as an alternative and established therapy.3 Tissue engineering approaches have the
ability to develop extracellular matrix (ECM) for cell adhesion, proliferation, maturation and differentiation support, enhancement of cell engraftment, and cell-to-cell contact. Cell transplantation therapy has been considered as a technique of choice in most of the severe ill patients due to its noninvasive characteristic besides its cost efficiency and less risk compared with organ transplantation. However, its application has been limited because of severe shortage of donor cells and insufficient long-term efficiency.4 Biological scaffolds composed of ECM, are used in different reconstructive medicine strategies for tissue and organ
Additional Supporting Information may be found in the online version of this article. None of the authors has direct or indirect commercial financial incentive associating with publishing the article and does not have any conflict of interest, and will sign the Disclosure Form. Correspondence to: Professor A.M. Kajbafzadeh; e-mail: [email protected]
C 2014 WILEY PERIODICALS, INC. V
1
replacement.5 The intact nature of the scaffold and the existence of growth factors are crucial points for developing an ideal tissue-engineered liver. It has been demonstrated that the interactions between cells and surrounded ECM can promote the migration of cells and tissue-specific gene expression.6 An effective decellularization procedure has the potential to preserve ECM complex composition, threedimensional ultrastructure, and bioactivity.5 Most of the decellularization techniques culminate in architecture disruption or loss of vital components due to the extra usage of disruptive agents and ineffective methods. A feasible decellularization protocol for a three-dimensional organ can create a scaffold with a functional vascular network.7 Recently, ECM of porcine and rat liver has been created using various chemicals and techniques.2,8 However, challenges in the decellularization methods including the potential of detergents to destroy ECM or their induced toxicity must be overcome before presenting decellularized organs to the clinical setting.9–11 The objective of this study was to propose a new feasible and appropriate liver decellularization technique and evaluate the efficiency of DLM for in vivo and in vitro cell seeding. We also aimed to compare the efficacy of in vivo liver recellularization with homograft and xenograft of DLMs. MATERIALS AND METHODS
Liver harvesting The animals were acclimated and had free access to standard food and water before participating in the study. All animal protocols were conducted under guidelines approved by the local ethical committee of the Tehran University of Medical Sciences. Prior to harvesting the liver, rats, and sheep were sacrificed via following systemic heparinization (5000 IU bolus dose before surgery and 25 IU/kg/h IV infusion intraoperatively in sheep and 500 IU bolus dose before surgery with the same infusion rate for rat). Then, the abdominal cavity was exposed through a midline incision under sterile condition. The portal vein and the common hepatic artery were dissected and cannulated as afferents. For exposing the suprahepatic vena cava, a midline sternotomy was done. Then the inferior vena cava was cannulated as an efferent in order to prove an antegrade perfusion (artery to vein). All accessory vessels and the bile duct were ligated. Fat, attached connective tissues, and diaphragmatic muscles were dissected free and the livers were resected. The afferent cannula was then attached to a peristaltic pump. Natural livers were served as the control samples; while other livers were decellularized with two different methods (TX 1 SDS or SDS) and served as experimental groups. Decellularization procedure Complete livers of eight Sprague Dawley rats weighing 0.25–0.3 kg, were aseptically obtained and immediately perfused with 0.2 L of distilled water containing 0.01% heparin to remove blood and clots. Then, the harvested livers were divided into two groups based on two decellulariza-
2
SABETKISH ET AL.
tion methods. All the solutions were autoclaved to have germ-free scaffolds for further cell seeding process. Method 1 (TX 1 SDS). After perfusion of distilled water, the organ was perfused with 0.2 L of 1% Triton X-100 for 1 h to dissolve cell membranes. In the next step, the livers were again washed with 0.2 L of distilled water for the next 1 h. The decellularization process continued with portal perfusion of 0.05% sodium dodecyl sulfate (SDS) as a biological detergent. About 0.4 L of SDS was infused to remove cytoplasmic proteins and nuclear remnants. The livers were then washed with 0.2 L of distilled water to remove residual SDS. Method 2 (SDS). All parts of the decellularization process were the same as method 1, except the usage of Triton X-100, which was omitted. These two methods were applied for decellularization of sheep livers with some modifications (1% SDS and 5% Triton X-100 with an approximate quadruple increase in the period of decellularization). All the previous stages were performed with a perfusion rate of 10 mL/ min under continuous shaking, and all solutions contained 1% antibiotics and antimycotic (i.e., penicillin, streptomycin, and amphotericin; Invitrogen, Carlsbad, CA). At the end of the procedure, all livers were stored in phosphate-buffered saline (PBS) in 4 C to be divided into sections and submitted for further investigations. Histological and immunohistochemical analyses To evaluate the efficacy of decellularization procedure and tissue structure integrity, DLMs and natural sections from the liver matrices of both sheep and rat scaffolds were fixed in 10% buffered formalin (pH 5 7.2) for 24 h at room temperature and embedded in paraffin with serial sections of 5 mm thickness. After the fixation, the adjacent samples were stained with hematoxylin and eosin (H&E). The reticulin and trichrome staining were also performed to visualize the reticular fibers and ECM structure, respectively. To stain nuclei of cells and evaluate the presence of nuclear materials in DLMs, 40 , 6-diamidino-2-phenylindole (DAPI) was used. Sections were incubated in xylene for 10 min, rehydrated in ethanol series, and equilibrated in Mcllvaine’s buffer for 5 min. DAPI staining solution was applied on slides (200 lL) and then incubated in a dark room for 15 min. Slides were imaged using IX71 Olympus microscope. Digital images were then processed with Photoshop 6.0 (Adobe Systems). Biochemical analysis (hydroxyproline content) The collagen content of the native and decellularized livers of both scaffolds was assessed by determining the amount of hydroxyproline. The hydroxyproline was quantified according to a previously described method.12–14 Tissue samples were hydrolyzed in 5 mL of 6N HCl at 110 C for 14–16 h. Then, 0.5 mL of Chloramine-T (EMD Chemicals, Gibbstown, NJ) solution (0.14 g Chloramine T, 2 mL distilled H2O, 8 mL citrate/acetate buffer) was added to each sample and the specimens were incubated at room temperature for
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
20 min. Then, 2 mL of distilled water and 8 mL of citrate/ acetate buffer was added to each sample and the samples were incubated at room temperature for 20 min. One milliliter of Ehrlich’s reagent (Sigma, Poole, UK) was then added, and the resulting mixture was incubated at 65 C for 20 min. Absorbance at 543 nm was measured on a spectrophotometer (using a BioTek Synergy HT1 micro plate reader), and collagen content in experimental samples was determined with hydroxyproline standard. Results were evaluated by analysis of variance (ANOVA) and Duncan tests to build up homogeneity groups, which show significance among differences at a 95% level. Scanning electron microscopy (SEM) Samples of native and decellularized sheep and rat livers (segments of 1 cm length) were fixed with 2.5% glutaraldehyde. Then, samples were dehydrated in an ethanol with concentrations of 30, 50, 70, and 90% and processed under a critical point dryer (Autosamdri-814; Tousimis) for 15 min. Specimens were coated with 3.5 mm of gold– palladium to obtain electrical conductivity. The samples were then analyzed using SEM (S3500N; Hitachi High Technologies America) at voltage of 5 kV to ascertain the efficacy of current procedures in cell removal and ECM preservation. Tensile test A tensile-test device, compromising a force transducer, digital meter, and a motorized stage, (Zwick/Roell, Model: Hct 400/ 25, Germany) was used to measure the rigidity of decellularized livers with method 1 and 2 and compare them with the natural tissue. For this purpose, natural and decellularized scaffolds were cut in 2 3 2 cm sections and were subjected to mounting uniaxial force with the acceleration rate of 4 mm/min until the disappearance of load was demonstrated by the device. Mechanical tests were performed in room temperature (23 6 1 C). Finally, the curve of strength–stress was drawn, the maximal point of which indicates the maximum pressure tolerance of the decellularized and natural tissues. Magnetic resonance imaging (MRI) and CT angiography We used this technique to evaluate vascular preservation and 3D structure of the DLMs and compare them with native livers. The sample tissues were engulfed into dimegluminegadopentetate 469 mg that was fined with normal saline for 8 h prior to MRI analysis. Images were taken and processed within 2 min of injection. Super conductive MRI t1 3D sequence SPGR (spoiled grass echo) with very thin slice, 3D reconstruction on MRI scanner (GE SIGNA 0.5 T), and Multi planner reformation were performed to obtain view and reformat the images. The protocol parameter for pulse sequences was TR 5 30–50 ms, TE 5 6 ms, MATRIX 5 320*224 NSA 5 4, and flip angle 30 d. To have better images, manual conduction was applied for prescanning and analyzing signal frequency. The volume rendering software (GE advantage workstation 4.2) was applied for downloading the data.
DNA quantification In this procedure, the cell walls/membranes are disrupted to isolate DNA. Lipids, proteins, and sugars are also separated from nucleic acid. This method was conducted according to the method described by Laird et al.15 For DNA quantification, 1 mg samples from DLMs of both scaffolds were obtained and homogenized in a solution containing 0.25% trypsin and 1 mM EDTA in deionized water. The homogenate was incubated with invariable stirring for 3 h at 37 C. In the next step, the cell lysis was continued with a solution containing 2% SDS, 5 mM EDTA, 200 mM NaCl, and 100 mM TRIS-HCl, pH 8.5 for 24 h at 55 C. The DNA extraction was performed in isopropanol and later dissolved in a solution of 10 mM Tris-HCl, 0.1 mM EDTA, and pH 5 7.5. Spectrophotometric analysis at 260 nm was applied for determining the amount of DNA. In vivo implantation Both homo and xenograft were performed by implanting small pieces of rat and sheep DLMs to the subhepatic area of four rats to evaluate the biocompatibility of the scaffolds and compare the role of different scaffolds with analogous and dissimilar origins, in recellularization procedure. Samples of sheep DLMs as well as rat DLMs were cut into pieces of 0.5 cm3. Four rats were anesthetized with intramuscular injection of Ketamine (150 mg/kg) and Xylazine (15 mg/kg). Then, right upper quadrant incision was made to access the infrahepatic space. Then, the pieces of homo and xeno grafts were attached to the right lobe of rat liver by four simple sutures at each corner with Prolene 3-0. Following the placement of the implanted tissues, the incision was sutured with Prolene 3-0. Until the first follow-up, the experimental animals were monitored preciously for any postoperative complication. The rats were sacrificed 8 weeks postoperatively to assess the in vivo immunoreactions and efficacy of cell seeding process. The explants were sectioned and stained with H&E and IHC staining. Cell culture Liver tissues were aseptically obtained from 4 green fluorescent protein (GFP) positive newborn rats and rinsed with fresh, sterile balanced salt solution (BSS). Unwanted tissues were gently dissected and the remained tissue was chopped into about 1-mm3 cubes with crossed scalpels. The tissue was then transferred to a centrifuge tube and washed with BSS for three times. After transferring the pieces to a 25 cm2 flask, the BSS was drained off and 4.5 mL of growth medium and 0.5 mL of crude collagenase were added. It was subsequently incubated at 37 C for 4 h without agitation. In the next step, the cell suspension was centrifuged at 753g for 3 min. After discarding the supernatant medium, the pellet was re-suspended in 5 mL of medium and was seeded in a 25 cm2 flask. The medium was replaced after 48 h. In vitro recellularization DLMs of four rats were transferred to a closed-system bioreactor that was placed in an incubator for the purpose of liver cell seeding. In this system, we applied a 100-mL glass bottle, which was connected to a pump with two holds on the cap; one for hepatic vein cannula and one for medium
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
3
FIGURE 1. Decellularization steps: (A) Normal cannulated Liver for decellularization process and (B) Second step of liver decellularization with triton and SDS. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
circulation. An 18-gauge I.V. cannula was used for preserving the scaffold in a hanging position in the bottle. The bioreactor was designed to circulate the medium 2 min in each hour with 58 min of resting time without medium circulation to enhance the cell adherence to the scaffold. For the first two days, 70 mL medium containing a total of 18 million cells was transferred through the portal vein cannula into the scaffold with a flow rate of 5 mL/min at 37 C in a cell culture incubator with 5% CO2. The cell seeding on DLMs was continued for 15 consecutive days; changing the medium with 3 days intervals. All the steps mentioned above, were performed using a shaker to prevent any cell adherence to the container wall. Afterwards, recellularized livers were carefully removed from the bottle within a biosafety cabinet for further histological and enzymatic examinations. Figure 1 shows a rat liver before and after decellularization process, which is ready for recellularization
procedure. Figure 2 depicts the simple bioreactor system in which the liver scaffold is being recellularized by GFPpositive hepatic cells. For immunohistochemical examinations, the previous method described by De Spiegelaere et al. 2010 was applied.16 Sections from paraffin-embedded samples were prepared and heated 2.5 and 10 min at 850 and 160 watt in antigen retrieval Citra solution (Biogenex, Klinipath, Belgium), respectively. Slides were submerged in 30% (v/v) rabbit serum in PBS for 30 min at room temperature and incubated with primary antibodies, which were rinsed in a PBS solution containing 2% (v/v) bovine serum albumin, in a humidified chamber for 1 h at 37 C. In immunohistochemical staining, Hepatocyte Specific Antigen Antibodies (Santa Cruz Biotechnology) was applied. This marker is also called Hepatocyte Paraffin 1 or Hep Par 1, which is localized to the mitochondria of hepatocytes.
FIGURE 2. Recellularization steps: (A) Perfusion of medium and cells, (B) GFP-positive hepatic cells. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
4
SABETKISH ET AL.
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
FIGURE 3. Evaluation of sheep DLMs: (A–C) H&E staining of normal and decellularized scaffolds with method 1 and 2, (D–F) DAPI staining of natural and decellularized livers with method 1 and 2, and (G–I) Tensile test of natural and processed livers with method 1 and 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Enzymatic evaluations A piece of natural, both transplanted livers (in vivo xeno and homo recellularized grafts), and in vitro recellularized liver tissues; weighing about 5 mg, was kept in PBS (pH 5 7.2) and squeezed in a mortar. The mashed tissues were then centrifuged at 12,000 rpm for 15 min. The supernatant was removed and assessed for determination of several hepatic enzymes including SGOP, SGPT, alkaline phosphatase, LDH, and Gamma GT.
RESULT
The results of decellularization process of sheep and rat scaffolds demonstrated that the livers treated with SDS 1 TX became whitish in appearance and translucent; however, the livers treated with SDS were less lucid. In both methods, the scaffolds maintained their gross appearance and size. Examination of native liver and sheep DLMs by H&E staining revealed that an intact liver architecture with complete
lobulation was obtained in method 1 in both sheep and rat scaffolds. Intact Glisson’s capsule, central veins and portal triads were also detected without remaining endothelial cells. However, a mildly distorted architecture without cellularity was discerned in method 2. H&E staining of normal and decellularized scaffolds with method 1 and 2 demonstrated a vast difference in ECM composition of the two scaffolds, which shows the superiority of method 1 in ECM preservation. These findings suggest the deterioration of ECM components in method 2 [Fig. 3(A–C)]. The parameters of the tensile test showed no significant changes in the mechanical properties of matrices with no detectable collagen loss when compared with the native liver. However, the scaffolds treated with SDS 1 TX method were more similar to natural tissue in maximal load parameter. All DLMs underwent tensile test for evaluation of resisting biomechanical forces. No significant change was detected in the mechanical properties of matrices treated with protocol 1 (maximal force 5 268 N) when compared with the native
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
5
FIGURE 4. Scanning electron microscope: (A–C) Natural tissue, (D–F) Decellularized liver with protocol 1, and (G–I) Decellularized scaffold with protocol 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
conduits (maximal force 5 249 N). The results verified no detectable extracellular elements loss due to the decellularization protocol. In fact, maximal load was approximately similar in both native and DLMs with protocol 1; suggesting its probable usefulness for further transplantation. However, biophysical properties of DLMs treated with protocol 2 (maximal force 5 315 N) was less similar to the native tissues [Fig. 3(D–I)]. SEM analysis of the sheep DLM showed a well preserved ECM in the scaffolds obtained from TX 1 SDS method, which were more similar to the native liver. However, some degree of deterioration was observed in the scaffolds treated with method 2 (Fig. 4). Rat DLMs were also analyzed by several histological examinations including H&E, reticulin, and trichrome staining as well as SEM analysis. The results were similar to sheep scaffolds in which the matrices treated with TX 1 SDS were better preserved and completely decellularized (Fig. 5). ECM quantification authenticated that 64.2 6 2.3% and 56.1 6 1.2% of the dry weight of DLM treated with protocol 1 and 2 were consisted of collagen, respectively. However, collagen content of DLMs obtained from protocol 1 was significantly higher (p < 0.05) compared with fresh liver tissue (51.5 6 1.4%). Glycosaminoglycans (GAGs) content was 24.2 6 1.9% and 18.3 6 1.4% of the dry weight of DLMs treated with protocol 1 and 2, respectively, as compared with natural liver 15.9 6 2.1%. The amount of GAG was sig-
6
SABETKISH ET AL.
nificantly higher (p < 0.05) in the scaffolds treated with TX 1 SDS. The higher amounts of collagen and GAG content in DLMs of method 1 as compared with natural livers were justifiable considering the fact that the cellular components were removed from the scaffolds. However, these amounts were lower in DLMs obtained from method 2; that suggests the ECM deterioration during the decellularization process. The cell survival after decellularization procedure was calculated by counting the cells in eight random microscopic fields, using the H&E slides. The residual cells were counted and the average number of cells was obtained (Table I). The hydroxyproline content of native liver was 0.57 6 0.52 lg/mg versus 0.55 6 0.38 lg/mg and 0.59 6 0.78 lg/mg for method 2 and 1, respectively. The concentrations of hydroxyproline per gram of wet liver tissues before and after decellularization were not significantly different in DLMs obtained from method 2 (p > 0.05). However, hydroxyproline content was significantly higher (p < 0.05) in the scaffolds treated with TX 1 SDS (Fig. 6). These findings were compatible with the results obtained from collagen and GAG evaluations. DNA quantification was 0.05) according to 1-way ANOVA. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
7
FIGURE 7. Histological examination of Sheep DLM before and after the transplantation: (A) Complete cell removal with preservation of basic structure frame was seen in the liver that was decellularized with method 1. The wall of the portal vein was thicker in method 1 with more connective tissue in the structure of portal vein wall, (B) More lysis reactions were found in method 2, and (C,D) Xenogeneic transplantation of DLMs with fibrotic changes in hepatocyte seeded scaffolds that were obtained from method 1 and 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
showed that the structure was noticeably similar to the natural liver with more organized hepatocytes as compared with in vivo seeded DLMs. The results of IF microscopy authenticated that GFP-positive cells were successfully seeded on the empty DLM after 15 days in a bioreactor. Moreover, hepatocyte staining showed that hepatocytes were seeded around the vessels as well as repopulated the surrounding parenchymal area (Fig. 9). DISCUSSION
In this study, an intact biological scaffold with complete microvasculature preservation was produced. The current study provided efficient three-dimensional architecture with preservation of collagen and ECM integrity, which could be an appropriate substrate for cell attachment and differentiation. In vitro recellularization was more satisfactory compared with in vivo experiment. However, homograft was more successful as compared with the xenograft. Liver transplantation still remains the only feasible treatment for liver damages; especially in end-stage liver diseases. Donor organ shortage, immunogenicity, and poor survival rate as the main driving forces for liver transplantation have made the development of liver tissue engineering more necessary. Tissue engineering and regenerative medicine have been considered as alternative emerging technologies in damaged livers to improve survival rate and quality of life in affected patients.
Cell differentiation and an acceptable function totally depend on the preservation of three-dimensional organ architecture. Over the years, many methods of decellularization have been applied for organ reconstruction.17 In the study of Crapo et al., a large volume of 0.25–0.5% SDS was used to remove nuclear remnants and cytoplasmic proteins without significant loss of three-dimensional architecture.5 In our previous study, five different liver decellularization protocols were compared. The results demonstrated that the decellularization protocol with ammonium hydroxide and triton X-100 was the most effective technique in preserving the vascular network and three-dimensional architecture as well as complete cell removal.18 Whole organ decellularization has been previously performed by portal vein perfusion using two methods to evaluate the effect of Triton X-100, as a nonionic surfactant for the recovery of membrane components19 before using SDS as an anionic detergent that lyses cells and dissolves cytoplasmic components and membrane lipids.20 Triton X-100 is usually used to solubilize cellular membranes; while SDS is used to clear the remaining nuclear remnants from the matrix. The results of the present study showed the effect of Triton X-100 when it was used before SDS (method 1) when compared with SDS alone (method 2). This may be due to the fact that Triton X-100 as the first detergent is efficient for disrupting lipid-lipid and lipid-protein interactions to dissolve cell membranes. The order of detergents used in the decellularization procedure may have an effect in protecting
TABLE II. Hepatic Enzymes Level
Normal Liver In Vitro Recellularized Liver In Vivo Recellularized Liver (Sheep to Rat) In Vivo Recellularized Liver (Rat to Rat)
8
SABETKISH ET AL.
SGOT (IU/L)
SGPT (IU/L)
Alkaline Phosphatase (U/L)
LDH (U/L)
Gamma GT (IU/L)
18,962 11,394 402 1011
1368 924 33 222
261 194 11 69
8694 5968 45 531
407 256 59 91
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
FIGURE 8. Histological evaluation of homografts and xenograft: (A,B) H&E and hepatocyte staining of rat to rat transplantation (C&D) H&E and hepatocyte staining of sheep to rat transplantation. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the architecture. However, further studies are needed to compare the effect of detergent usage in different orders. Post decellularization MRI and angiography showed almost intact micro vascular network with complete endothelial cell removal as a conduit for distribution of host cells to provide cell seeding and differentiation. Histological evaluation of the decellularized matrix confirmed ECM preservation and intact architecture with complete elimination of cells and nuclear components as well as venous endothelial cells removal in both methods. However, parenchymal or endothelial cells removal in SDS method was not as complete as Triton method. The decellularization protocols proposed here produced successful results with H&E staining, SEM, and DAPI staining. These results were confirmed by DNA quantification as a more sensitive method. The findings of some studies have suggested that
Pediatric Urology Research Center, Section of Tissue Engineering and Stem Cells Therapy, Department of Pediatric Urology, Children’s Hospital Medical Center, Tehran, Iran (IRI) 2 Department of Medical Biochemistry, Tehran University of Medical Sciences, Tehran, Iran (IRI) 3 Department of Pathology, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran (IRI) Received 21 June 2014; accepted 3 July 2014 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35291 Abstract: To report the results of whole liver decellularization by two different methods. To present the results of grafting rat and sheep decellularized liver matrix (DLM) into the normal rat liver and compare natural cell seeding process in homo/xenograft of DLM. To compare the results of in vitro whole liver recellularization with rats’ neonatal green fluorescent protein (GFP)-positive hepatic cells with outcomes of in vivo recellularization process. Whole liver of 8 rats and 4 sheep were resected and cannulated via the hepatic vein and perfused with sodium dodecyl sulfate (SDS) or Triton 1 SDS. Several examinations were performed to compare the efficacy of these two decellularization procedures. In vivo recellularization of sheep and rat DLMs was performed following transplantation of multiple pieces of both scaffolds in the subhepatic area of four rats. To compare the efficacy of different scaffolds in autologous cell seeding, biopsies of homograft and xenograft were assessed 8 weeks postoperatively. Whole DLMs of 4 rats were also recellularized in vitro by perfusion of rat’s fetal GFP-positive hepatic cells with pulsatile bioreactor. Histological evaluation and enzymatic assay were
performed for both in vivo and in vitro recellularized samples. The results of this study demonstrated that the triton method was a promising decellularization approach for preserving the three-dimensional structure of liver. In vitro recellularized DLMs were more similar to natural ones compared with in vivo recellularized livers. However, homografts showed better characteristics with more organized structure compared with xenografts. In vitro recellularization of liver scaffolds with autologous cells represents an attractive prospective for regeneration of liver as one of the most compound organs. In vivo cell seeding on the scaffold of the same species may have more satisfactory outcomes when compared with the results of xenotransplantation. This study theoretically may pave the road for in situ liver regeneration probably by implantation of homologous DLM or in vitro C 2014 recellularized scaffolds into the diseased host liver. V Wiley Periodicals, Inc. J Biomed Mater Res Part A: 00A:000–000, 2014.
Key Words: liver, extra cellular matrix, decellular, tissue engineering, transplantation
How to cite this article: Sabetkish S, Kajbafzadeh A-M, Sabetkish N Khorramirouz R, Akbarzadeh A, Seyedian SL, Pasalar P, Orangian S, Beigi RSH, Aryan Z, Akbari H, Tavangar SM. 2014. Whole-organ tissue engineering: Decellularization and recellularization of three-dimensional matrix liver scaffolds. J Biomed Mater Res Part A 2014:00A:000–000.
INTRODUCTION
End stage liver diseases including cirrhosis, chronic viral hepatitis, hepatocellular carcinoma, injuries from alcohol abuse, or even inborn metabolic disorders, often lead to demands for organ transplantation.1,2 As a result, many of these patients may die before an appropriate organ becomes achievable.2 Due to increasing number of patients on the waiting list for transplantation, short donor organ, and a need for life-long immunosuppressive remedy, tissue engineering has been developed as an alternative and established therapy.3 Tissue engineering approaches have the
ability to develop extracellular matrix (ECM) for cell adhesion, proliferation, maturation and differentiation support, enhancement of cell engraftment, and cell-to-cell contact. Cell transplantation therapy has been considered as a technique of choice in most of the severe ill patients due to its noninvasive characteristic besides its cost efficiency and less risk compared with organ transplantation. However, its application has been limited because of severe shortage of donor cells and insufficient long-term efficiency.4 Biological scaffolds composed of ECM, are used in different reconstructive medicine strategies for tissue and organ
Additional Supporting Information may be found in the online version of this article. None of the authors has direct or indirect commercial financial incentive associating with publishing the article and does not have any conflict of interest, and will sign the Disclosure Form. Correspondence to: Professor A.M. Kajbafzadeh; e-mail: [email protected]
C 2014 WILEY PERIODICALS, INC. V
1
replacement.5 The intact nature of the scaffold and the existence of growth factors are crucial points for developing an ideal tissue-engineered liver. It has been demonstrated that the interactions between cells and surrounded ECM can promote the migration of cells and tissue-specific gene expression.6 An effective decellularization procedure has the potential to preserve ECM complex composition, threedimensional ultrastructure, and bioactivity.5 Most of the decellularization techniques culminate in architecture disruption or loss of vital components due to the extra usage of disruptive agents and ineffective methods. A feasible decellularization protocol for a three-dimensional organ can create a scaffold with a functional vascular network.7 Recently, ECM of porcine and rat liver has been created using various chemicals and techniques.2,8 However, challenges in the decellularization methods including the potential of detergents to destroy ECM or their induced toxicity must be overcome before presenting decellularized organs to the clinical setting.9–11 The objective of this study was to propose a new feasible and appropriate liver decellularization technique and evaluate the efficiency of DLM for in vivo and in vitro cell seeding. We also aimed to compare the efficacy of in vivo liver recellularization with homograft and xenograft of DLMs. MATERIALS AND METHODS
Liver harvesting The animals were acclimated and had free access to standard food and water before participating in the study. All animal protocols were conducted under guidelines approved by the local ethical committee of the Tehran University of Medical Sciences. Prior to harvesting the liver, rats, and sheep were sacrificed via following systemic heparinization (5000 IU bolus dose before surgery and 25 IU/kg/h IV infusion intraoperatively in sheep and 500 IU bolus dose before surgery with the same infusion rate for rat). Then, the abdominal cavity was exposed through a midline incision under sterile condition. The portal vein and the common hepatic artery were dissected and cannulated as afferents. For exposing the suprahepatic vena cava, a midline sternotomy was done. Then the inferior vena cava was cannulated as an efferent in order to prove an antegrade perfusion (artery to vein). All accessory vessels and the bile duct were ligated. Fat, attached connective tissues, and diaphragmatic muscles were dissected free and the livers were resected. The afferent cannula was then attached to a peristaltic pump. Natural livers were served as the control samples; while other livers were decellularized with two different methods (TX 1 SDS or SDS) and served as experimental groups. Decellularization procedure Complete livers of eight Sprague Dawley rats weighing 0.25–0.3 kg, were aseptically obtained and immediately perfused with 0.2 L of distilled water containing 0.01% heparin to remove blood and clots. Then, the harvested livers were divided into two groups based on two decellulariza-
2
SABETKISH ET AL.
tion methods. All the solutions were autoclaved to have germ-free scaffolds for further cell seeding process. Method 1 (TX 1 SDS). After perfusion of distilled water, the organ was perfused with 0.2 L of 1% Triton X-100 for 1 h to dissolve cell membranes. In the next step, the livers were again washed with 0.2 L of distilled water for the next 1 h. The decellularization process continued with portal perfusion of 0.05% sodium dodecyl sulfate (SDS) as a biological detergent. About 0.4 L of SDS was infused to remove cytoplasmic proteins and nuclear remnants. The livers were then washed with 0.2 L of distilled water to remove residual SDS. Method 2 (SDS). All parts of the decellularization process were the same as method 1, except the usage of Triton X-100, which was omitted. These two methods were applied for decellularization of sheep livers with some modifications (1% SDS and 5% Triton X-100 with an approximate quadruple increase in the period of decellularization). All the previous stages were performed with a perfusion rate of 10 mL/ min under continuous shaking, and all solutions contained 1% antibiotics and antimycotic (i.e., penicillin, streptomycin, and amphotericin; Invitrogen, Carlsbad, CA). At the end of the procedure, all livers were stored in phosphate-buffered saline (PBS) in 4 C to be divided into sections and submitted for further investigations. Histological and immunohistochemical analyses To evaluate the efficacy of decellularization procedure and tissue structure integrity, DLMs and natural sections from the liver matrices of both sheep and rat scaffolds were fixed in 10% buffered formalin (pH 5 7.2) for 24 h at room temperature and embedded in paraffin with serial sections of 5 mm thickness. After the fixation, the adjacent samples were stained with hematoxylin and eosin (H&E). The reticulin and trichrome staining were also performed to visualize the reticular fibers and ECM structure, respectively. To stain nuclei of cells and evaluate the presence of nuclear materials in DLMs, 40 , 6-diamidino-2-phenylindole (DAPI) was used. Sections were incubated in xylene for 10 min, rehydrated in ethanol series, and equilibrated in Mcllvaine’s buffer for 5 min. DAPI staining solution was applied on slides (200 lL) and then incubated in a dark room for 15 min. Slides were imaged using IX71 Olympus microscope. Digital images were then processed with Photoshop 6.0 (Adobe Systems). Biochemical analysis (hydroxyproline content) The collagen content of the native and decellularized livers of both scaffolds was assessed by determining the amount of hydroxyproline. The hydroxyproline was quantified according to a previously described method.12–14 Tissue samples were hydrolyzed in 5 mL of 6N HCl at 110 C for 14–16 h. Then, 0.5 mL of Chloramine-T (EMD Chemicals, Gibbstown, NJ) solution (0.14 g Chloramine T, 2 mL distilled H2O, 8 mL citrate/acetate buffer) was added to each sample and the specimens were incubated at room temperature for
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
20 min. Then, 2 mL of distilled water and 8 mL of citrate/ acetate buffer was added to each sample and the samples were incubated at room temperature for 20 min. One milliliter of Ehrlich’s reagent (Sigma, Poole, UK) was then added, and the resulting mixture was incubated at 65 C for 20 min. Absorbance at 543 nm was measured on a spectrophotometer (using a BioTek Synergy HT1 micro plate reader), and collagen content in experimental samples was determined with hydroxyproline standard. Results were evaluated by analysis of variance (ANOVA) and Duncan tests to build up homogeneity groups, which show significance among differences at a 95% level. Scanning electron microscopy (SEM) Samples of native and decellularized sheep and rat livers (segments of 1 cm length) were fixed with 2.5% glutaraldehyde. Then, samples were dehydrated in an ethanol with concentrations of 30, 50, 70, and 90% and processed under a critical point dryer (Autosamdri-814; Tousimis) for 15 min. Specimens were coated with 3.5 mm of gold– palladium to obtain electrical conductivity. The samples were then analyzed using SEM (S3500N; Hitachi High Technologies America) at voltage of 5 kV to ascertain the efficacy of current procedures in cell removal and ECM preservation. Tensile test A tensile-test device, compromising a force transducer, digital meter, and a motorized stage, (Zwick/Roell, Model: Hct 400/ 25, Germany) was used to measure the rigidity of decellularized livers with method 1 and 2 and compare them with the natural tissue. For this purpose, natural and decellularized scaffolds were cut in 2 3 2 cm sections and were subjected to mounting uniaxial force with the acceleration rate of 4 mm/min until the disappearance of load was demonstrated by the device. Mechanical tests were performed in room temperature (23 6 1 C). Finally, the curve of strength–stress was drawn, the maximal point of which indicates the maximum pressure tolerance of the decellularized and natural tissues. Magnetic resonance imaging (MRI) and CT angiography We used this technique to evaluate vascular preservation and 3D structure of the DLMs and compare them with native livers. The sample tissues were engulfed into dimegluminegadopentetate 469 mg that was fined with normal saline for 8 h prior to MRI analysis. Images were taken and processed within 2 min of injection. Super conductive MRI t1 3D sequence SPGR (spoiled grass echo) with very thin slice, 3D reconstruction on MRI scanner (GE SIGNA 0.5 T), and Multi planner reformation were performed to obtain view and reformat the images. The protocol parameter for pulse sequences was TR 5 30–50 ms, TE 5 6 ms, MATRIX 5 320*224 NSA 5 4, and flip angle 30 d. To have better images, manual conduction was applied for prescanning and analyzing signal frequency. The volume rendering software (GE advantage workstation 4.2) was applied for downloading the data.
DNA quantification In this procedure, the cell walls/membranes are disrupted to isolate DNA. Lipids, proteins, and sugars are also separated from nucleic acid. This method was conducted according to the method described by Laird et al.15 For DNA quantification, 1 mg samples from DLMs of both scaffolds were obtained and homogenized in a solution containing 0.25% trypsin and 1 mM EDTA in deionized water. The homogenate was incubated with invariable stirring for 3 h at 37 C. In the next step, the cell lysis was continued with a solution containing 2% SDS, 5 mM EDTA, 200 mM NaCl, and 100 mM TRIS-HCl, pH 8.5 for 24 h at 55 C. The DNA extraction was performed in isopropanol and later dissolved in a solution of 10 mM Tris-HCl, 0.1 mM EDTA, and pH 5 7.5. Spectrophotometric analysis at 260 nm was applied for determining the amount of DNA. In vivo implantation Both homo and xenograft were performed by implanting small pieces of rat and sheep DLMs to the subhepatic area of four rats to evaluate the biocompatibility of the scaffolds and compare the role of different scaffolds with analogous and dissimilar origins, in recellularization procedure. Samples of sheep DLMs as well as rat DLMs were cut into pieces of 0.5 cm3. Four rats were anesthetized with intramuscular injection of Ketamine (150 mg/kg) and Xylazine (15 mg/kg). Then, right upper quadrant incision was made to access the infrahepatic space. Then, the pieces of homo and xeno grafts were attached to the right lobe of rat liver by four simple sutures at each corner with Prolene 3-0. Following the placement of the implanted tissues, the incision was sutured with Prolene 3-0. Until the first follow-up, the experimental animals were monitored preciously for any postoperative complication. The rats were sacrificed 8 weeks postoperatively to assess the in vivo immunoreactions and efficacy of cell seeding process. The explants were sectioned and stained with H&E and IHC staining. Cell culture Liver tissues were aseptically obtained from 4 green fluorescent protein (GFP) positive newborn rats and rinsed with fresh, sterile balanced salt solution (BSS). Unwanted tissues were gently dissected and the remained tissue was chopped into about 1-mm3 cubes with crossed scalpels. The tissue was then transferred to a centrifuge tube and washed with BSS for three times. After transferring the pieces to a 25 cm2 flask, the BSS was drained off and 4.5 mL of growth medium and 0.5 mL of crude collagenase were added. It was subsequently incubated at 37 C for 4 h without agitation. In the next step, the cell suspension was centrifuged at 753g for 3 min. After discarding the supernatant medium, the pellet was re-suspended in 5 mL of medium and was seeded in a 25 cm2 flask. The medium was replaced after 48 h. In vitro recellularization DLMs of four rats were transferred to a closed-system bioreactor that was placed in an incubator for the purpose of liver cell seeding. In this system, we applied a 100-mL glass bottle, which was connected to a pump with two holds on the cap; one for hepatic vein cannula and one for medium
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
3
FIGURE 1. Decellularization steps: (A) Normal cannulated Liver for decellularization process and (B) Second step of liver decellularization with triton and SDS. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
circulation. An 18-gauge I.V. cannula was used for preserving the scaffold in a hanging position in the bottle. The bioreactor was designed to circulate the medium 2 min in each hour with 58 min of resting time without medium circulation to enhance the cell adherence to the scaffold. For the first two days, 70 mL medium containing a total of 18 million cells was transferred through the portal vein cannula into the scaffold with a flow rate of 5 mL/min at 37 C in a cell culture incubator with 5% CO2. The cell seeding on DLMs was continued for 15 consecutive days; changing the medium with 3 days intervals. All the steps mentioned above, were performed using a shaker to prevent any cell adherence to the container wall. Afterwards, recellularized livers were carefully removed from the bottle within a biosafety cabinet for further histological and enzymatic examinations. Figure 1 shows a rat liver before and after decellularization process, which is ready for recellularization
procedure. Figure 2 depicts the simple bioreactor system in which the liver scaffold is being recellularized by GFPpositive hepatic cells. For immunohistochemical examinations, the previous method described by De Spiegelaere et al. 2010 was applied.16 Sections from paraffin-embedded samples were prepared and heated 2.5 and 10 min at 850 and 160 watt in antigen retrieval Citra solution (Biogenex, Klinipath, Belgium), respectively. Slides were submerged in 30% (v/v) rabbit serum in PBS for 30 min at room temperature and incubated with primary antibodies, which were rinsed in a PBS solution containing 2% (v/v) bovine serum albumin, in a humidified chamber for 1 h at 37 C. In immunohistochemical staining, Hepatocyte Specific Antigen Antibodies (Santa Cruz Biotechnology) was applied. This marker is also called Hepatocyte Paraffin 1 or Hep Par 1, which is localized to the mitochondria of hepatocytes.
FIGURE 2. Recellularization steps: (A) Perfusion of medium and cells, (B) GFP-positive hepatic cells. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
4
SABETKISH ET AL.
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
FIGURE 3. Evaluation of sheep DLMs: (A–C) H&E staining of normal and decellularized scaffolds with method 1 and 2, (D–F) DAPI staining of natural and decellularized livers with method 1 and 2, and (G–I) Tensile test of natural and processed livers with method 1 and 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Enzymatic evaluations A piece of natural, both transplanted livers (in vivo xeno and homo recellularized grafts), and in vitro recellularized liver tissues; weighing about 5 mg, was kept in PBS (pH 5 7.2) and squeezed in a mortar. The mashed tissues were then centrifuged at 12,000 rpm for 15 min. The supernatant was removed and assessed for determination of several hepatic enzymes including SGOP, SGPT, alkaline phosphatase, LDH, and Gamma GT.
RESULT
The results of decellularization process of sheep and rat scaffolds demonstrated that the livers treated with SDS 1 TX became whitish in appearance and translucent; however, the livers treated with SDS were less lucid. In both methods, the scaffolds maintained their gross appearance and size. Examination of native liver and sheep DLMs by H&E staining revealed that an intact liver architecture with complete
lobulation was obtained in method 1 in both sheep and rat scaffolds. Intact Glisson’s capsule, central veins and portal triads were also detected without remaining endothelial cells. However, a mildly distorted architecture without cellularity was discerned in method 2. H&E staining of normal and decellularized scaffolds with method 1 and 2 demonstrated a vast difference in ECM composition of the two scaffolds, which shows the superiority of method 1 in ECM preservation. These findings suggest the deterioration of ECM components in method 2 [Fig. 3(A–C)]. The parameters of the tensile test showed no significant changes in the mechanical properties of matrices with no detectable collagen loss when compared with the native liver. However, the scaffolds treated with SDS 1 TX method were more similar to natural tissue in maximal load parameter. All DLMs underwent tensile test for evaluation of resisting biomechanical forces. No significant change was detected in the mechanical properties of matrices treated with protocol 1 (maximal force 5 268 N) when compared with the native
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
5
FIGURE 4. Scanning electron microscope: (A–C) Natural tissue, (D–F) Decellularized liver with protocol 1, and (G–I) Decellularized scaffold with protocol 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
conduits (maximal force 5 249 N). The results verified no detectable extracellular elements loss due to the decellularization protocol. In fact, maximal load was approximately similar in both native and DLMs with protocol 1; suggesting its probable usefulness for further transplantation. However, biophysical properties of DLMs treated with protocol 2 (maximal force 5 315 N) was less similar to the native tissues [Fig. 3(D–I)]. SEM analysis of the sheep DLM showed a well preserved ECM in the scaffolds obtained from TX 1 SDS method, which were more similar to the native liver. However, some degree of deterioration was observed in the scaffolds treated with method 2 (Fig. 4). Rat DLMs were also analyzed by several histological examinations including H&E, reticulin, and trichrome staining as well as SEM analysis. The results were similar to sheep scaffolds in which the matrices treated with TX 1 SDS were better preserved and completely decellularized (Fig. 5). ECM quantification authenticated that 64.2 6 2.3% and 56.1 6 1.2% of the dry weight of DLM treated with protocol 1 and 2 were consisted of collagen, respectively. However, collagen content of DLMs obtained from protocol 1 was significantly higher (p < 0.05) compared with fresh liver tissue (51.5 6 1.4%). Glycosaminoglycans (GAGs) content was 24.2 6 1.9% and 18.3 6 1.4% of the dry weight of DLMs treated with protocol 1 and 2, respectively, as compared with natural liver 15.9 6 2.1%. The amount of GAG was sig-
6
SABETKISH ET AL.
nificantly higher (p < 0.05) in the scaffolds treated with TX 1 SDS. The higher amounts of collagen and GAG content in DLMs of method 1 as compared with natural livers were justifiable considering the fact that the cellular components were removed from the scaffolds. However, these amounts were lower in DLMs obtained from method 2; that suggests the ECM deterioration during the decellularization process. The cell survival after decellularization procedure was calculated by counting the cells in eight random microscopic fields, using the H&E slides. The residual cells were counted and the average number of cells was obtained (Table I). The hydroxyproline content of native liver was 0.57 6 0.52 lg/mg versus 0.55 6 0.38 lg/mg and 0.59 6 0.78 lg/mg for method 2 and 1, respectively. The concentrations of hydroxyproline per gram of wet liver tissues before and after decellularization were not significantly different in DLMs obtained from method 2 (p > 0.05). However, hydroxyproline content was significantly higher (p < 0.05) in the scaffolds treated with TX 1 SDS (Fig. 6). These findings were compatible with the results obtained from collagen and GAG evaluations. DNA quantification was 0.05) according to 1-way ANOVA. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00
7
FIGURE 7. Histological examination of Sheep DLM before and after the transplantation: (A) Complete cell removal with preservation of basic structure frame was seen in the liver that was decellularized with method 1. The wall of the portal vein was thicker in method 1 with more connective tissue in the structure of portal vein wall, (B) More lysis reactions were found in method 2, and (C,D) Xenogeneic transplantation of DLMs with fibrotic changes in hepatocyte seeded scaffolds that were obtained from method 1 and 2. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
showed that the structure was noticeably similar to the natural liver with more organized hepatocytes as compared with in vivo seeded DLMs. The results of IF microscopy authenticated that GFP-positive cells were successfully seeded on the empty DLM after 15 days in a bioreactor. Moreover, hepatocyte staining showed that hepatocytes were seeded around the vessels as well as repopulated the surrounding parenchymal area (Fig. 9). DISCUSSION
In this study, an intact biological scaffold with complete microvasculature preservation was produced. The current study provided efficient three-dimensional architecture with preservation of collagen and ECM integrity, which could be an appropriate substrate for cell attachment and differentiation. In vitro recellularization was more satisfactory compared with in vivo experiment. However, homograft was more successful as compared with the xenograft. Liver transplantation still remains the only feasible treatment for liver damages; especially in end-stage liver diseases. Donor organ shortage, immunogenicity, and poor survival rate as the main driving forces for liver transplantation have made the development of liver tissue engineering more necessary. Tissue engineering and regenerative medicine have been considered as alternative emerging technologies in damaged livers to improve survival rate and quality of life in affected patients.
Cell differentiation and an acceptable function totally depend on the preservation of three-dimensional organ architecture. Over the years, many methods of decellularization have been applied for organ reconstruction.17 In the study of Crapo et al., a large volume of 0.25–0.5% SDS was used to remove nuclear remnants and cytoplasmic proteins without significant loss of three-dimensional architecture.5 In our previous study, five different liver decellularization protocols were compared. The results demonstrated that the decellularization protocol with ammonium hydroxide and triton X-100 was the most effective technique in preserving the vascular network and three-dimensional architecture as well as complete cell removal.18 Whole organ decellularization has been previously performed by portal vein perfusion using two methods to evaluate the effect of Triton X-100, as a nonionic surfactant for the recovery of membrane components19 before using SDS as an anionic detergent that lyses cells and dissolves cytoplasmic components and membrane lipids.20 Triton X-100 is usually used to solubilize cellular membranes; while SDS is used to clear the remaining nuclear remnants from the matrix. The results of the present study showed the effect of Triton X-100 when it was used before SDS (method 1) when compared with SDS alone (method 2). This may be due to the fact that Triton X-100 as the first detergent is efficient for disrupting lipid-lipid and lipid-protein interactions to dissolve cell membranes. The order of detergents used in the decellularization procedure may have an effect in protecting
TABLE II. Hepatic Enzymes Level
Normal Liver In Vitro Recellularized Liver In Vivo Recellularized Liver (Sheep to Rat) In Vivo Recellularized Liver (Rat to Rat)
8
SABETKISH ET AL.
SGOT (IU/L)
SGPT (IU/L)
Alkaline Phosphatase (U/L)
LDH (U/L)
Gamma GT (IU/L)
18,962 11,394 402 1011
1368 924 33 222
261 194 11 69
8694 5968 45 531
407 256 59 91
WHOLE LIVER TISSUE ENGINEERING
ORIGINAL ARTICLE
FIGURE 8. Histological evaluation of homografts and xenograft: (A,B) H&E and hepatocyte staining of rat to rat transplantation (C&D) H&E and hepatocyte staining of sheep to rat transplantation. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the architecture. However, further studies are needed to compare the effect of detergent usage in different orders. Post decellularization MRI and angiography showed almost intact micro vascular network with complete endothelial cell removal as a conduit for distribution of host cells to provide cell seeding and differentiation. Histological evaluation of the decellularized matrix confirmed ECM preservation and intact architecture with complete elimination of cells and nuclear components as well as venous endothelial cells removal in both methods. However, parenchymal or endothelial cells removal in SDS method was not as complete as Triton method. The decellularization protocols proposed here produced successful results with H&E staining, SEM, and DAPI staining. These results were confirmed by DNA quantification as a more sensitive method. The findings of some studies have suggested that
