|
|
||||||||
J Thorac Cardiovasc Surg 2004;127:399-405
© 2004 The American Association for Thoracic Surgery
Surgery for acquired cardiovascular disease |
a Department of Cardiothoracic Surgery and Ludwig-Boltzmann-Institute for Cardiosurgical Research, University of Vienna, Vienna, Austria
Received for publication March 22, 2003; revisions received May 1, 2003; accepted for publication June 23, 2003.
* Address for reprints: Guenter Weigel, MD, Department of Cardiothoracic Surgery, University of Vienna, AKH, Waehringer Guertel 18-20, 1090 Vienna, Austria
guenter.weigel{at}akh-wien.ac.at
| Abstract |
|---|
|
|
|---|
METHODS: Porcine aortic and pulmonary roots were treated with trypsin, sodium-dodecyl-sulphate, or a new method using 0.25% tert-octylphenyl-polyoxyethylen in combination with sodium-deoxycholate. After a subsequent ribonuclease digestion, specimens were seeded with in vitro expanded human saphenous vein endothelial cells and myofibroblasts.
RESULTS: After treatment with trypsin and subsequent ribonuclease digestion, endothelial attachment took place; however, xenogenic cells were still visible within the matrix. Unexpectedly, when human cells were seeded onto specimens that had been decellularized with sodium-dodecyl-sulphate, the matrices were surrounded by nonviable endothelial cell fragments, indicating a toxic influence of the ionic detergent; 0.25% tert-octylphenyl-polyoxyethylen together with sodium-deoxycholate completely removed porcine cells and enabled host recellularization.
CONCLUSION: Compared with trypsin and sodium-dodecyl-sulphate involving decellularization procedures, reported to be effective in cell removal and susceptible to recellularization with human cells, only the porcine matrix treated with a new detergent-based decellularization method using 0.25% tert-octylphenyl-polyoxyethylen/sodium-deoxycholate followed by nuclease digestion presented an excellent scaffold for recellularization with human cells.
Human allografts eliminate the need of anticoagulation and have the best hemodynamic properties, but their availability is limited. None of these currently used heart valve substitutes shows any growth potential or regeneration capability, which would play an important role especially in the treatment of pediatric patients.
Tissue engineering of heart valves is a multidisciplinary approach to overcome these limitations. The created autologous living replacement structure should have the ability to self-repair, remodel, and thereby increase durability without the risk of rejection and ideally even grow.
The initial and crucial step is to obtain an optimal valve matrix.4,5 We have chosen the concept of using a decellularized xenogenous heart valve.
The aim of this study was to test different decellularization protocols for their efficiency of complete cell removal and ability to produce a matrix with excellent recellularization susceptibility with human cells.
| Material and methods |
|---|
|
|
|---|
Three different decellularization protocols were used for comparison, of which 2 have been described5-8 and the third was developed in our laboratory, using 2 detergents formerly used for cell extraction9 in combination with ribonuclease digestion and to our knowledge has not been published before.
In the first group the valve conduits (PV = 5, AV = 4) were treated with 0.1% trypsin (Gibco, Paisley, Scotland, UK) with 0.02% ethylenediaminetetraacetic acid (EDTA; Merck, Darmstadt, Germany) in phosphate-buffered saline solution without Ca2+ and Mg2+ (PBS-/-, Gibco) for 48 hours. In the second group valve conduits (PV = 4, AV = 3) were treated with 0.1% sodium-dodecyl-sulphate (SDS; BioRad, Hercules, Calif) in PBS-/- for 24 hours. Treatment in group 3 (PV = 7, AV = 7) is a detergent-based decellularization mixture consisting of a 24-hour incubation period with 0.25% tert-octylphenyl-polyoxyethylen (Triton X-100, BioRad) plus 0.25% sodium-deoxycholate (Merck) in PBS-/- at 37°C, followed by a intensive wash cycle with M-199 medium (Gibco) for 72 hours at 4°C to remove residual substances. All specimens were subsequently treated by a nuclease digestion with ribonuclease (100 µg/mL; RNase, Roche Diagnostics Gmbh, Mannheim, Germany) and deoxyribonuclease (150 IU/mL; DNase, Sigma, St Louis, Mo) with 50 mmol/L MgCl2 in PBS-/- for 24 hours at 37°C to remove remaining nucleic remnants. Samples were again washed with M-199 medium for 24 hours at 4°C. All steps were conducted under continuous shaking.
Cell separation and expansion
Human saphenous vein endothelial cells (HSVEC) were harvested from discarded segments of vena saphena magna of patients undergoing coronary artery bypass surgery. In brief, the veins were cannulated and flushed with sterile PBS-/- under sterile conditions. The lumen was then filled with 0.2% collagenase type II (Gibco) solution and left at room temperature for 15 minutes. To dislodge the loosened endothelial cells the vein was gently massaged and flushed with M-199 medium. The HSVEC were pelleted by centrifugation for 5 minutes at 300g and thereafter resuspended in culture medium consisting of M-199 medium, 20% pooled human serum, 100 µg/mL heparin (Sigma), 100 U/mL penicillin, and 100 µg/mL streptomycin (Gibco) and 30 mg/L endothelial cell growth supplement (Upstate, Lake Placid, NY).10 The cell suspension was transferred to a 25-cm2 culture flask. Endothelial cells were maintained in a humidified incubator at 37°C and 5% CO2 and the culture medium was changed every fourth or fifth day. Confluent HSVEC were detached using 0.05% trypsin/EDTA in PBS-/- and subcultured. Purity of cells was evaluated by factor VIII (FVIII: vWF) staining,11 transmission electron microscopy (presence of Weibel-Pallade bodies), and expression of CD62E (E-selectin). No contamination by myocytes or fibroblasts was detected.
The remaining vessel was minced and pieces of 1 to 2 mm2 were placed into a 25-cm2 culture flask with myofibroblast culture medium containing Dulbeco's modified Eagle's medium (BioWhittaker Europe, Verviers, Belgium), 100 U/mL penicillin, and 100 µg/mL streptomycin until saphenous vein media cells grew out onto the surface of the culture flask. Confluent cell cultures were subcultured with 0.05% trypsin/EDTA in PBS-/-.
Cell seeding
The seeding susceptibility of the 3 differently decellularized porcine heart valve matrices was compared without any pretreatment of the matrices. Each valve conduit was divided and leaflets were dissected. Two 100-mm2 specimens of the pulmonary or aortic root, as well as 2 leaflets of each conduit, were placed into a culture dish and seeded under static conditions with either 1 x 104/mm2 HSVEC or 2 x 104/mm2 saphenous vein myofibroblasts (only HSVEC were seeded onto SDS-treated specimens). Afterward, the specimens were incubated in a humidified environment at 37°C and 5% CO2 either for 5 days (HSVEC) or 10 days (myofibroblasts).
Biocompatibility of SDS treatment
To examine a possible toxic influence of SDS released from decellularized specimens, increasing concentrations (10%-30% v/v) of PBS-/-, used to store the SDS-treated root specimens, were added to culture medium of in vitro cultured endothelial cells (EC) and compared with the influence of PBS-/- that was not used for storing specimens.
Morphological analysis
The specimens were either fixed by immersion in 2.5% glutaraldehyde for at least 24 hours and further processed for SEM or embedded in Tissue Tek OCT Compound (Sakura, Zoeterwoude, The Netherlands). Five cryostat sections (10 µm) of each specimen were stained for the production of von Willebrand factor (anti-vWf, Serotec, Düsseldorf, Germany) to analyze the viability of the transplanted HSVEC. The DNA-specific dye TO-PRO 3 (Molecular Probes, Leiden, The Netherlands) was used to assess the removal of xenogenous cells or nuclei in the porcine matrix.
The sections were also stained with an anti-porcine collagen type I and III antibody (Monosan, Uden, The Netherlands), shown to be cross-reactive with human type I/III collagen, and with an anti-elastin antibody (monoclonal anti-elastin, Clone BA-4, Sigma) to examine structural alterations of the matrix fibers.
Cryostat section was scanned for remaining cell remnants with a Zeiss LSM 510 laser scanning microscope (Zeiss, Jena, Germany) using the 3-dimensional analysis mode and representative micrographs are shown.
| Results |
|---|
|
|
|---|
|
|
Increasing concentrations (10%-30%) of PBS-/-, used to store the SDS-treated root-specimen, added to culture medium of in vitro cultured HSVEC revealed a decreasing amount of attached and viable endothelial cells in the culture flasks within 24 hours. In contrast, the same amount of PBS-/- that was not used for storage of SDS-treated specimens did not affect in vitro cell growth (Figure 3, A to D). This test for the biocompatibility of SDS treatment indicates that not even a prolonged washout of SDS removes residual ionic-detergent substances within root matrices.
|
The TO-PRO 37 staining in the sections examined demonstrated no visible cells within the porcine pulmonary (n = 28) as well as aortic valve matrix (n = 28) (Figure 2, D). Confocal microscopic examination of specimens seeded with media cells demonstrated the thin, bright green collagen-specific area surrounding the patchy attached human myofibroblasts illustrating viable and actively collagen producing cells. A representative micrograph is shown in Figure 4.
|
| Discussion |
|---|
|
|
|---|
Currently used glutaraldehyde cross-linked valves are mostly of porcine origin,2 though different reports according xenogeneic rejection patterns12-14 will need further investigation. Viral-associated risks15-17 should be considered when xenogenous tissue is used for tissue engineering. This was a reason to involve ribonuclease digestion within a decellularization procedure. Due to the limited availability of human autologous heart valves the use of xenogenic valves is still of obvious advantage. Because vascular grafts placed in humans do not spontaneously form an endothelial monolayer, whereas they do in animals,18 the importance of a confluent and viable endothelial cell layer on valve substitutes prior to implantation seems essential compared with other approaches.14,19 Seeding of vascular grafts with human autologous endothelial cells improves their long-term patency20,21 and reduces the thrombogenicity of cardiovascular implants.22
The recipient saphenous vein is an easily accessible cell source for seeding decellularized valves prior to implantation. Early experiments comparing allogenic and autologous cells favored the latter.23 The use of artery segments24 is unlikely to be transferable to the clinical situation.
Several decellularization methods involving detergents and/or enzymatic digestion have been described previously with some success but only very few reports have dealt with their susceptibility to seeding leaflet as well as root matrices.
In this study we demonstrate that the decellularization method of xenogenous heart valves is a crucial feature to obtain the optimal matrix for a tissue-engineered valve substitute. We found considerable differences in regard to the efficiency of complete xenogenous cell removal and the susceptibility to human cell attachment.
We as others found using trypsin5 or a nonionic detergent like Triton X-100 followed by ribonuclease digestion25 revealed only an incomplete removal of cells. Bader and colleagues5 described that a longer incubation with trypsin might be more effective in cell removal but causes a disruption of the matrix bundles. However, residual cells and cell remnants within the matrix might lead to calcification.12 As the naturally occurring alpha-gal antibodies in humans are interacting with alpha-gal epitopes26 on porcine cells, the efficient extraction of xenogenous cells is crucial. Otherwise an acute immunogenic response and early graft failure will follow.19
SDS treatment has previously been used to obtain complete acellularity7 but was also reported to destabilize the collagen triple helical domain and to swell the elastin network.9,27 However, these studies did not describe the seeding susceptibility. SDS is still described to be efficient for leaflet decellularization8 and was recently presented to be compatible for seeding of leaflets with endothelial cells.28 Although in our study SDS produced a biological matrix free of any visible xenogenous cells, it was impossible to seed the porcine aortic or pulmonary root specimens with human endothelial cells or myofibroblasts, as massive cell lysis occurred within 24 hours of incubation. This controversial finding might be due to probably higher levels of residual SDS remaining in the vascular wall. When increasing amounts of PBS-/-, which was used to store the SDS decellularized valve matrix after several days of washing, were added to culture medium of HSVEC, cell lysis occurred.
The findings of Kim and colleagues28 with respect to the seeding of leaflets might be explainable by the fact that the washing procedure is more effective when treating the much thinner valve leaflets. Because in a clinical setting the complete valve conduit will likely be used, the observation made by us seems crucial with respect to further studies involving SDS.
Our decellularization procedure, involving the combination of Triton X-100 and sodium-deoxycholate followed by a washing process with M-199 medium to remove any residual detergents and an enzymatic digestion with DNAse/RNAse to remove remaining nucleic acids, results in an excellent scaffold, free of any histologically detectable xenogenous cells or cell remnants, which is well suited for seeding with human cells. Transplanted endothelial cells and myofibroblasts remained viable. We could even demonstrate active collagen synthesis by human myofibroblasts attached to the xenogenous matrix. Although Triton and cholate treatment was presented not to affect the structural integrity of either elastin and collagen,9 further studies concerning mechanical testing are currently under investigation in our laboratory. Cell seeding under flow conditions should not have impact on the biocompatibility of xenogenous matrices, but cultivation under shear stress will provide the ability of proper adherence of seeded cells to acellular matrix surface and is also subject of current experiments.
We conclude from our findings that decellularization procedures used differ considerably in their efficiency of cell removal and susceptibility of the matrix to recellularization. Our newly developed method produces an excellently preserved xenogenic scaffold that fulfills both requirements, acellularity and high susceptibility to recellularization with human vascular cells.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. M. Dohmen and W. Konertz Decellularization of xenogenic biologic tissue. Ann. Thorac. Surg., June 1, 2008; 85(6): 2163 - 2163. [Full Text] [PDF] |
||||
![]() |
F. Juthier, A. Vincentelli, J. Gaudric, D. Corseaux, O. Fouquet, C. Calet, T. L. Tourneau, V. Soenen, C. Zawadzki, O. Fabre, et al. Decellularized heart valve as a scaffold for in vivo recellularization: Deleterious effects of granulocyte colony-stimulating factor J. Thorac. Cardiovasc. Surg., April 1, 2006; 131(4): 843 - 852. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Ota, Y. Sawa, S. Iwai, T. Kitajima, Y. Ueda, C. Coppin, H. Matsuda, and Y. Okita Fibronectin-Hepatocyte Growth Factor Enhances Reendothelialization in Tissue-Engineered Heart Valve Ann. Thorac. Surg., November 1, 2005; 80(5): 1794 - 1801. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Vesely Heart Valve Tissue Engineering Circ. Res., October 14, 2005; 97(8): 743 - 755. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Ketchedjian, A. L. Jones, P. Krueger, E. Robinson, K. Crouch, L. Wolfinbarger Jr, and R. Hopkins Recellularization of Decellularized Allograft Scaffolds in Ovine Great Vessel Reconstructions Ann. Thorac. Surg., March 1, 2005; 79(3): 888 - 896. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |