| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:1794-1801
© 2005 The Society of Thoracic Surgeons
a Department of Cardiovascular Surgery, Kobe University Graduate School of Medicine, Kobe
b Department of Cardiovascular Surgery, Osaka University Graduate School of Medicine, Osaka
c Terumo, Ltd, Kanagawa, Japan
d Medtronic, Ltd, Santa Ana, California
Accepted for publication May 9, 2005.
* Address correspondence to Dr Sawa, Department of Surgery (E1), Osaka University Graduate School of Medicine, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan (Email: sawa{at}surg1.med.osaka-u.ac.jp).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
METHODS: Porcine aortic valves were decellularized using detergent. Fibronectin-hepatocyte growth factor was introduced into the decellularized valves. The decellularized valves with fibronectin-hepatocyte growth factor were implanted into the pulmonary arterial trunk of dogs (F group: n = 15). As controls, decellularized valves without the growth factor (C group: n = 12), and with hepatocyte growth factor (H group: n = 12) were implanted in the same manner. Histologic examinations were performed 1 week and 1 month after implantation.
RESULTS: One week after implantation, endothelial cells partially covered the surface of the graft in the F group but not the C and H groups. Although the C and H groups had inadequate recellularization 1 month after implantation, the F group showed a monolayer of endothelial cells, underneath which were areas of additional cell layers, which were vimentin positive. Quantitative evaluation demonstrated the amount of vimentin in the F group was 71% of the native control, and it was much lower in the other groups (C, 2.8%; H, 16.8%) 1 month after implantation.
CONCLUSIONS: This study demonstrated that fibronectin-hepatocyte growth factor enhanced early in situ recellularization in decellularized valves.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
| Doctor Kitajima discloses a financial relationship with Terumo, Ltd; and Drs Ueda and Coppin with Medtronic, Ltd.
|
Valve replacement using mechanical or tissue valves is the most common method of treating advanced dysfunction of cardiac valves. However, mechanical prostheses have some limitations, such as the risk of infection and thromboembolism and anticoagulant-related hemorrhage [1]. Then, tissue valves and xenografts are considered superior in hemodynamics and relatively resistant to infection, and they do not require anticoagulation treatment. However, these grafts have limited durability [2, 3], which may be due to their mechanical fatigue of the leaflet tissue and their immunogenic potential. Another limitation of the existing valve substitutes is that current prostheses do not grow, which leads to the need for repeated valve replacement in pediatric patients. These problems may be due to the lack of living cells and tissue remodeling in the matrix [4].
To overcome these limitations, tissue-engineered valves have been developed, and the concept of "acellular grafts" has also appeared recently [5, 6]. One of the methods involves decellularization of the xenograft heart valve. The cellular components are removed from the xenograft by detergent treatment, leaving the extracellular matrix (ECM) in the acellular graft. The decellularized valve induces a lower immunologic response to the implantation because of the reduction in antigen [5], and a recellularization with mesenchymal cells that could allow regeneration of the extracellular matrix.
Some reports suggest that ex-vivo cell seeding using a bioreactor is necessary to promote the recellularization of decellularized tissues [7, 8]. However, these cases have been associated with problems and limitations such as increased time and cost, impracticality of harvesting autologous cells in emergent cases. Therefore, tissue valves with ex-vivo cell seeding may be difficult to apply in general medical treatment. To overcome these limitations, one important strategy is to introduce a growth factor that will induce optimal in situ recellularization.
We developed a new decellularized heart valve containing fibronectin-hepatocyte growth factor (Fn-HGF), a fusion protein in which fibronectin and hepatocyte growth factor are combined by a covalent linkage. Fibronectin is the cell adhesive factor for binding HGF to the graft, and HGF attracts and stimulates reendothelialization by endothelial cells [9, 10]. In this study, we implanted our new decellularized valve containing Fn-HGF into dogs, and verified the hypothesis that the introduction of Fn-HGF into decellularized tissues accelerated the early in situ recellularization.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Chemicals that were used for each processes were as follows: sodium chloride (Sigma, St. Louis, MO), sodium azide (Sigma), ethylenediaminetetraacetic acid (Sigma), HEPES (Sigma), protease inhibitor cocktail (Sigma), sodium lauryl sulfate (Invitrogen), and Triton X-100 (ICN Biomedicals, Irvine, CA).
Creation of the Fusion Protein of Fibronectin and HGF
The human fibronectin collagen-binding domain (FNCBD) was fused to the amino terminus of mature human HGF with an inserted amino acid sequence using the baculovirus technique, as described [11].
Recombinant Human Hepatocyte Growth Factor
Recombinant human HGF (r-HGF) was purified from the conditioned medium of CHO cells transfected with an expression vector containing the cDNA for HGF with 5 amino acids deleted, as described [12–16].
Combination of Fn-HGF (or r-HGF) With Decellularized Porcine Aortic Root
Decellularized porcine aortic roots were washed in phosphate-buffered saline containing antibiotics (100 U penicillin, 0.1 mg streptomycin, 0.25 µg/mL amphotericin B; Invitrogen, Carlsbad, CA) for 72 hours (500 mL/valve). The wash solution was changed every 24 hours, then decanted.
Fetal bovine serum 20 µl (ICN Biomedicals) was added to the Fn-HGF 1 mL (50 µg/mL) and the mixture was stirred at 37°C for 15minutes. During this process, the single chains of Fn-HGF form heterodimers.
The rinsed valves were then placed into a Fn-HGF solution (Fn-HGF 50 µg, PBS 50 mL; 50 mL/valve), put into a 500 KPa pressurized container (LMP-100 air pump; AS ONE, Osaka, Japan; TM5SRV; Unicontrols, Tokyo, Japan) for 12 hours, and stirred afterward at room temperature for 12 hours. The Fn-HGF solution was decanted, and the valves were rinsed with 500 mL PBS for 15minutes.
In the case of r-HGF, the same procedure as above was performed: r-HGF 1 mL (50 µg/mL), r-HGF solution (r-HGF 50 µg, PBS 50 mL; 50 mL/valve).
Implantation Into Dogs
The breed of dogs used in this study is beagle dogs (body weight 8 to 10 kg, all female). The decellularized valves containing Fn-HGF were implanted into the pulmonary arterial trunk of dogs (F group: n = 15). As a control, decellularized valves without Fn-HGF were implanted (C group: n = 12). Valves containing recombinant HGF were also implanted (H group: n = 12). Anesthesia was induced with 5 mg/kg ketamine and 5 mg/kg pentobarbital, and maintained by inhalational sevofluorane. The heart was exposed through left thoracotomy at the third intercostal space. Systemic anticoagulation was induced with heparin 100 IU/kg. The monocusp patch consists of a leaflet, and an aortic wall was made by cutting the whole decellularized valve into three equal parts. The patch plasty (1.5 x 1.5 cm) was performed with a side-bit partial clamp of the main pulmonary artery. The suture line was made with a 5-0 continuous polypropylene suture. Heparin was reversed with 100 IU of protamine per kilogram after patch plasty. Histologic examination of the grafts was performed 1 week and 1 month after implantation. Regarding the explantation procedure, the implanted graft was exposed with the same anesthesia and approach of the implantation procedure. Then potassium chloride (20 mL per dog, 1 mEq/mL; Terumo) was injected for sacrifice.
All animals received humane care in compliance with the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (NIH publication No.85-23, revised 1985).
Histology and Immunohistochemistry
Cryosections (5-µm thick) of tissues were stained with hematoxylin and eosin and visualized in bright field using an Olympus IX70 microscope. Immunohistochemical staining was performed using the monoclonal antibody of factor VIII-related antigen (DAKO, Kyoto, Japan), vimentin (DAKO), and HGF (DAKO). The immunoreaction was detected with 3, 3-diaminobenzidine or fluorescein isothiocyanate. The vimentin is positive at immature mesenchymal cells, and the factor VIII is at endothelial cells.
Quantification of Recellularization of Leaflets
The cell count of leaflets was performed in 1 month samples using an microscope. The number of cells per square millimeter was counted in five different fields in each group.
Quantitative Evaluation of HGF
The amount of HGF in samples was determined by HGF EIA (Immunis; Institute of Immunology, Tokyo, Japan).
Quantitative Reverse Transcription–Polymerase Chain Reaction
The vimentin mRNA level was analyzed by reverse transcription–polymerase chain reaction (RT-PCR). Samples were stored in RNAlater (Qiagen, Valencia, CA). The RNA was isolated using a RNeasy mini kit (Qiagen) and treated with a RNase Free Dnase Set (Qiagen). Then 1 µg total RNA was reverse-transcribed with Omniscript Reverse Transcriptase (Qiagen). The volume, 20 µL, of the resultant cDNA mixture was adjusted to 100 µL by adding double-distilled water. The RT-PCR was performed using the ABI Prism 7700 Sequence Detection System with TaqMan probe (Applied Biosystems Inc, Foster City, CA). TaqMan probe was labeled with a reporter dye (FAM) at the 5'-end and a quencher dye (TAMRA) at the 3'-end. Thermal cycling parameters for real-time RT-PCR were at 50°C for 2 minutes and at 95°C for 10 minutes, followed by 40 cycles of shuttle heating at 95°C for 15 seconds and at 60°C for 1 minute.
The measurement for each cDNA is performed in tripricate. Relative quantification was calculated as the ratio between the amount of target and GAPDH (a housekeeping gene) within the same sample. The sequences of the specific primers were as follows: GAPDH(sense),GTGATGCTGGTGCTGAGTATGTTGT; GAPDH(antisense),TGGCTAGAGGAGCCAAGCAGTT; vimentin(sense),ACTAATGAGTCTCTGGAACGCCA; vimentin(antisense),CTTAACATTCAGCAGGTCCTGGATC.
Statistical Analysis
All values are expressed as the mean ± SEM. The statistical differences in all data were determined by a Student's t test. A value of less than 0.05 was considered to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Histology and Immunohistochemistry
The histologic examination 1 week after implantation showed a monolayer of cells partially covering the surface of the graft tissue in the F group (Fig 1A). This monolayer was positive for immunohistochemical staining for factor VIII (Fig 1D). In contrast, recellularization was not observed in the C and H groups (Fig 1B, 1C, 1E, and 1F). One month after implantation, one layer of homogeneous cells fully covered the surface in the F group (Fig 2A). In the H group, there was a partial monolayer of cells on the tissue (Fig 2B), and in the C group there were fragmentary regions of a monolayer, far less than in the H group (Fig 2C). These cells were factor-VIII positive (Fig 2D, 2E, and 2F). At the cusp region, recellularization was observed in the F group much more than in the other groups (Fig 3A, 3B, and 3C). The cusps have a tendency to recellularize more easily than the vascular wall according to histologic preparations. In addition, at 1 month, in the F group, multilayered cells were observed on parts of the tissue. These cells consisted of the monolayer cells, which were factor-VIII positive, in the outer layer and some additional layers of cells that were vimentin positive (Fig 4A, 4B, and 4C). This is the most important matter in this study.
|
|
|
|
|
Quantification of Recellularization of Leaflets
In the F group, the cell count was 220.4 ± 53.3/mm2. It was 75.2 ± 19.2/mm2 in the H group (p < 0.001) and 26.0 ± 9.7/mm2 in the C group (p < 0.001).
Quantitative Evaluation of HGF
Hepatocyte growth factor was not detected in the native porcine aortic root. At the preimplantation point, the HGF concentration was 342.3 ± 42.3 ng/g dry weight in the F group and 161.6 ± 24.8 ng/g dry weight in the H group (p = 0.01). At 1 week, it was 12.9 ± 1.6 ng/g dry weight in the F group, and 2.1 ± 0.7 ng/g dry weight in the H group (p < 0.001; Fig 6). Hepatocyte growth factor was not detected biochemically at the 1-month time point in either group (data not shown).
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Fibronectin collagen-binding domain is extracted from fibronectin, and has just only the ability of binding to collagen and gelatin. We established a collagen-binding growth factor consisting of HGF and FNCBD [11, 18]. This fusion protein has a high affinity for collagen, and exerted its growth factor activity even after collagen binding. When FNCBD connects with these collagen ligands, it forms an efficient attachment of HGF to the graft that is stable over the long term. This adhesion and stability were demonstrated by HGF quantification and staining. In the preimplant graft, attachment of HGF to the decellularized tissue was higher in the Fn-HGF group than in the r-HGF-treatment group. In the 1-week samples, Fn-HGF remained along the surface of the graft. Consequently, the HGF was functional around the graft in situ, and the neointima built up more promptly. Generally, it takes more than 1 or 3 months for the decellularized scaffold to be recellularized with endothelial cells [19, 20]. On the other hand, in the graft with Fn-HGF, the endothelial cells matured as a monolayer in the brief period of only 1 week, but did not in the controls. Consequently, the vascular regeneration may have been accelerated even 1 month later. At 1 month, there were some areas where vimentin-positive cells were layered underneath the monolayer of endothelial cells. The appearance of these cells suggested the invasion of mesenchymal cells. At the cusp region, the cusps had a tendency to recellularize more easily than the vascular wall according to histologic preparations. Sixty percent of the cusp of our decellularized valve consists of collagen, and 20% of the vascular wall (data not shown). The difference of collagen content brings about the tendency that the cusps were recellularized more easily than the vascular wall. The recellularizetion of a leaflet is significantly better in the Fn-HGF valves among three groups, based on the results of the number of cells in leaflets. Quantitative RT-PCR of the 1-month samples demonstrated that the quantity of vimentin was significantly higher in the Fn-HGF valves than in those containing HGF alone. The vimentin is positive at immature mesenchymal cells. In this study, there are possibilities that their cells were branched from endothelial cells or were induced from mesenchymal progenitor cells in the blood, depending on the function of endothelial cells, not only the migration from the neighboring tissue, because those cells were found as skip lesions. That means that components of the vascular wall may have proliferated owing to the functions of the endothelial cells. Therefore, the introduction of Fn-HGF may accelerate reendothelialization of the decellularized heart valve in situ.
Originally, HGF was identified as a potent mitogen stimulating hepatocyte growth [21]. Since then, many recent findings have suggested that HGF also stimulates various other cells, such as vascular endothelial cells [22, 23]. In vivo, HGF is a heparin-binding glycoprotein that is secreted as a single-chain, biologically inert precursor. Under the appropriate conditions, such as tissue damage, single-chain HGF is converted to its bioactive form (heterodimer) at a specific site within the molecule. This conversion may be mediated by urokinase plasminogen activator or by a protease homologous to factor XII [24]. Hepatocyte growth factor is a mesenchyme-derived pleiotropic factor that regulates the growth, motility, and morphogenesis of various types of cells. Remarkable biological activities associated with HGF include a potent mitogenic effect on endothelial cells, stimulation of angiogenesis, and acceleration of the regeneration of endothelial cells [9, 10]. In contrast, HGF does not stimulate the growth of vascular smooth muscle cells (VSMCs), but does stimulate their migration [25, 26]. In the present study, HGF might have played a role in accelerating the reendothelialization of decellularized valves.
Endothelial cells are known to secrete various vasoactive factors. Endothelial cells may modulate vascular growth, because they secrete many antiproliferative factors, such as nitric oxide and vascular natriuretic peptides [27, 28]. Meanwhile, VSMC growth is controlled by a balance of growth inhibitors and promoters, and endothelial cell dysfunction destroys this balance and leads to VSMC proliferation [29, 30]. Endothelial cell dysfunction is well documented in hypertensive patients [31, 32]. The loss of the substances produced from endothelial cells might be related to the development and progression of atherosclerosis in hypertensive patients [29, 30]. Therefore, a lack of endothelial cells might promote abnormal vascular growth, such as thrombosis, hyperplasia, and obstruction. Conversely, the presence of normal endothelial cells might lead to the normal growth of VSMCs, and normal vascular regeneration in the long run. The precellularization of valves using a bioreactor seems effective for improving vascular regeneration. However, almost all the reseeded endothelial cells become detached after the implantation of heart valves, because of the change in hemodynamic circumstances and aging of the endothelial cells. Therefore, early reendothelialization is very important for adaptation of the decellularized heart valve after implantation.
In the decellularized tissues, the cellular components, including endothelial cells, are removed from the allograft, leaving the ECM in the acellular graft. The decellularized valve may induce a lower immunologic response due to antigen reduction and to the superior recellularization on this scaffold [33]. These are the best advantages and chief purpose of using decellularized tissues. As noted above, early recellularization with normal endothelial cells is critical for the healthy vascular regeneration of decellularized tissues.
There are several limitations to this study. First, we adopted the patch plasty technique using a dog model in terms of cost, patch size, and simplifying operations. Some articles have been reported about successful recellularization of decellularized tissues and heart valves using dog models [34, 35]. In the present study, however, the valve function was not evaluated. A whole valve implantation using an ovine model or a primate model is necessary to evaluate the valve function and the long-term recellularization, including the evaluation of the future differentiation of vimentin-positive cells as the next phase. Second, the ability of the decellularized tissue to grow is unproved at present. A tissue-engineered valve that lacks potential to grow is not useful for pediatric applications. Long-term analysis is needed to test the ability of our decellularized tissue to grow. Third, indeed, massive immunologic response did not occurred in this study. However, there is a possibility that the scaffold that consists of untreated porcine tissues has some persisting immunogenicity. Detailed immunologic evaluations, including T-lymphocyte activation, will be necessary as a next step.
In conclusion, this study demonstrated that early in situ recellularization of the decellularized valve was made possible by adding Fn-HGF fusion protein to the valve. The goal of developing our decellularized graft is to create a graft that is nonimmunogenic, noncalcified, nonthrombogenic, resistant to infections, and moreover, has a normal biological ability to grow. The conclusion of this study is that this approach is a promising first step toward our goal.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Ota, T. W. Gilbert, D. Schwartzman, C. F. McTiernan, T. Kitajima, Y. Ito, Y. Sawa, S. F. Badylak, and M. A. Zenati A fusion protein of hepatocyte growth factor enhances reconstruction of myocardium in a cardiac patch derived from porcine urinary bladder matrix. J. Thorac. Cardiovasc. Surg., November 1, 2008; 136(5): 1309 - 1317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Schoen Evolving Concepts of Cardiac Valve Dynamics: The Continuum of Development, Functional Structure, Pathobiology, and Tissue Engineering Circulation, October 28, 2008; 118(18): 1864 - 1880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ota, S. Taketani, S. Iwai, S. Miyagawa, M. Furuta, M. Hara, E. Uchimura, Y. Okita, and Y. Sawa Novel Method of Decellularization of Porcine Valves Using Polyethylene Glycol and Gamma Irradiation Ann. Thorac. Surg., April 1, 2007; 83(4): 1501 - 1507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Gulbins and A. Pritisanac Invited commentary Ann. Thorac. Surg., November 1, 2005; 80(5): 1801 - 1802. [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 |
