Journal Logo

Clinical and Translational Research

Generation and Transplantation of an Autologous Vascularized Bioartificial Human Tissue

Mertsching, Heike1; Schanz, Johanna1; Steger, Volker2; Schandar, Markus1; Schenk, Martin3; Hansmann, Jan1; Dally, Iris1; Friedel, Godehard2; Walles, Thorsten1,2,4

Author Information
doi: 10.1097/TP.0b013e3181ac15e1


In many surgical subspecialties, therapeutic options for surgical reconstruction of organ defects are limited and, regrettably, no dependable synthetic or biologic reconstruction materials have yet been found (1, 2). Here, tissue engineering gives the opportunity to generate living substitutes for tissues and organs, which may overcome the drawbacks of classical tissue reconstruction: lacking quality and quantity of autologous grafts, immunogenicity of allogenic grafts, and loosening of alloplastic implants (3). Although this area has been the focus of considerable research, many challenges remain, including the isolation and expansion of appropriate cell types, the arrangement of assorted cells into correct spatial organization, and the creation of the optimal microenvironment for cell growth and differentiation. The integration of in vitro-generated tissues with the host is an area of intense interest, as it requires not only a metabolic and structural match to the in vivo environment but also functional vasculature to sustain function after implantation (4). Until now, successes have been restricted to relatively thin or avascular structures (e.g., cartilage and skin), where postimplantation vascularization from the host is sufficient to meet the implant’s demand for oxygen and nutrients. Vascularization remains a critical obstacle in engineering thicker, metabolically demanding organs of sufficient size (5). Acknowledging these limitations, we focused on the generation of an acellular biological scaffold that provides the natural mircroenvironment for bioartificial vascular network engineering. On the basis of clinically established heterotopic transplantation of an autologous jejunal segment for reconstruction of the digestive tract, we developed methods for the generation of a bioartificial vascularized scaffold (BioVaSc) for tissue engineering that affords vascular anastomosis of any bioartificial construct to the recipient blood supply (6).

Here, we report the generation and clinical implementation of a bioartificial human tissue with an innate vascularization that may be suitable as starting basis for various sophisticated bioengineered tissue and organ replacements in future clinical applications.


Decellularization of the Porcine Scaffold

A porcine jejunal segment was obtained from a 3-month-old pig for scaffold generation. All animals received human care in compliance with the Guide for Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No. 85-23, revised 1996) after approval from our institutional animal protection board. Our experimental protocol to obtain a porcine jejunal segment has been published previously (6) and is described in detail online (Supplemental Digital Content 1, For chemical decellularization, we connected the arterial pedicle to a recirculating perfusion circuit of a bioreactor and perfused it for 75 min with 4% sodium desoxychylate monohydrate solution (Sigma Aldrich, Hamburg, Germany) at a rate of 2.3 mL/min (Fig. 1A). To remove cell residues and sodium desoxychylate, we perfused the decellularized porcine jejunal segment for 75 min with phosphate-buffered saline solution (Sigma Aldrich) at a rate of 3.9 mL/min and a pressure of 80 to 100 mm Hg. Then, the decellularized porcine jejunal segment was incubated in 150 mL DNase I-solution (200 U/mL) (Roche, Penzberg, Germany) at room temperature over night. For removal of all chemical residues, the porcine scaffold was incubated 7 times for 4 hr in 4°C cold phosphate-buffered saline solution. To effectuate tissue sterility, the scaffold was γ-irradiated with 25 kGy over night (BBF Sterilizationsservice GmbH, Rommelshausen, Germany).

Biotechnological approach to generate human tissue with an innate vascularization and main operative steps of the implant procedure. (A) Tissue decellularization and reseeding takes place in a bioreactor connected to a computer controlled perfusion circuit. (B) Chemicals, cells and cell culture media are injected via the arterial pedicle, circulate through the graft and are recovered via the venous pedicle. (C) The acellular vascular structures of the decellularized jejunal segment are reseeded with endothelial cells. The graft surface and the scaffold lumen are coated with autologous fibroblasts and muscle cells (for clarity only denoted in the diagram). (D) The left upper arm was chosen in the right-handed patient because of easy accessibility. (E) The bioartificial autologous tissue was connected to the brachial vascular bundle. (F) The BioVaSc connected to the recipient blood supply. Notice the cross-linked capillary network within the BioVaSc and the hematoma (*) depicting localized vascular damage. (G) The BioVaSc is enveloped with a synthetic fleece to preclude secondary vascularization of the bioartificial transplant by the surrounding tissue.

Histologic Characterization of the Acellular Porcine Jejunal Matrix

Segments of the acellular porcine jejunal matrix were incubated for 1 hr in Bouin’s reagent (formaldehyde and picric acid) (Sigma, Munich, Germany) and embedded in paraffin for histologic and immunohistochemical analysis. Human skin served as control. Hematoxylin-Eosin staining was performed according to standard protocols. For immunohistochemical characterization, the tissue was stained using the EnVision technique by the presence of CD 31 (MCA 1746, Serotec, Düsseldorf, Germany), Flk-1 (sc-6251, Santa Cruz, Heidelberg, Deutschland), and VE-cadherin (sc-6458, Santa Cruz). Endogenous peroxidase was blocked with peroxidase blocking solution (Dako, Hamburg, Germany), and unspecific binding was blocked by the use of antibody diluent (Dako). For detection, a EnVison System+horseradish peroxidase (Dako) was used. The staining reaction was performed with diaminobenzidine, and slides were counterstained with hematoxylin (Sigma). The Feulgen-reaction was used as a qualitative marker to detect DNA residues in the decellularized porcine tissue. For this, the tissue was processed with a commercial DNA staining kit according to the manufacturer’s protocols (Merck, Darmstadt, Germany). For all stainings, isotype controls were performed. Only stainings with negative isotype controls were analyzed.

Proteinbiochemical Characterization of the Acellular Porcine Jejunal Matrix

The water content and dry weight of the native and the decellularized porcine tissue were determined (n=9). For this, the specimen was frozen at −80°C, thawed at room temperature (10 min), and minced in 2 to 4 mm large segments, which were transferred into 2-mL Eppendorf containers (Eppendorf, Hamburg, Germany). After moist mass identification, the segments were dried at 36.2°C in a cabinet desiccator for 96 hr and reweighed. For protein content determination, the tissue segments were dissolved in 0.5 M acetic acid (Merck) at room temperature over night on a horizontal shaker (Greiner Bio-One, Kremsmünster, Austria) at 200 rpm. Thereafter, collagens, glycosaminoglycans and proteoglycans, and elastin were quantified using the Sicrol soluble collagen, the Blyscan sulfated glycosaminoglycan, and the Fastin elastin assays (all Biocolor, Newtownabbey, UK), respectively. The individual working steps are listed in Supplemental Table 1 (see Table, Supplemental Digital Content 2, Protein concentration was analyzed by triplicate photometric measurements (Genequant II, Pharmacia Biotech [Biochrom], Cambridge, UK) in a 96-well plate at 540 nm, 656 nm, and 513 nm wave length. Purified rat tail collagen solution (6 mg/mL; Fraunhofer IGB, Stuttgart, Germany) was used as positive control, and reagent solution without tissue probe was used as negative control. For DNA quantification, a laboratory kit was used (DNeasy Blood & Tissue, Qiagen, Hilden, Germany). Native and acellular porcine jejunal segments were lyophilized, and their dry weight was determined (n=3). Then, 2-mg native and 10-mg acellular jejunal segment were minced and incubated with 180 μL ATL buffer (Qiagen) and 20 μL proteinase K at 56°C (water bath) over night. Two hundred microliter AL buffer (Qiagen) and 200 μL ethanol were added, and the assay was spun for 1 min with 6,000 rpm in a DNeasy Mini spin column. After two washing steps, the DNA was eluted with 400 μL AE buffer (Qiagen) and spun (6,000 rpm). The elute was analyzed photospectrometrically at 260 nm wave length. DNA content was normalized with the weighed in amount of tissue.

Generation of the Bioartificial Vascularized Scaffold

The vascular remainings within the decellularized jejunal segment were reseeded with human (1) bone marrow-derived mesenchymal stem cells (bmMSC), (2) cutaneous microvascular endothelial cells (mECs), or (3) peripheral blood mononuclear cells (PBMCs) using a specially designed recirculating perfusion set up (Fig. 1B). The cell preparation steps were published previously (6). For endothelial differentiation, bmMSC, mEC, and PBMC, were maintained in the endothelial cell basal medium (EBM)-2 supplemented with hydrocortisone (100 μg/500 mL culture medium solution), fetal calf serum (50 mL/500 mL), porcine vascular endothelial growth factor (0.25 μg/500 mL), human basic fibroblast growth factor (5 μg/500 mL), human epidermal growth factor (2.5 μg/500 mL), insulin-like growth factor (R3 IGF, 10 μg/500 mL), ascorbic acid, penicillin (100 IE/mL), streptomycin (100 μg/mL), and essential amino acids (500 mg/500 mL). Performing a two-staged reseeding procedure, 5 mL EBM-2 containing (1) 10.5×106 bmMSC, (2) 4.5×106 mEC, or (3) 15×106 PBMC and 2000 IE heparin/mL were injected into the arterial BioVaSc pedicle on two consecutive days. Cells were allowed to adhere to the matrix for 30 min (day 1) and 3.5 hr (day 2) before a pulsatile perfusion at a rate of 1.26 mL/min was implemented. For the first 2 days, the BioVaSc was perfused at a pressure of 20 to 60 mm Hg and a frequency of 1 Hz. Starting on day 3, the perfusion pressure was increased to 80 to 120 mm Hg. The BioVaSc lumen was perfused accordingly with EBM-2 in a separate open circuit. Tissue culture was performed in a 5% CO2 athmosphere providing 95% humidity and 37°C and maintained for 14 days. To safeguard ideal culture conditions, temperature, gas exchange, pump activity (Ismatec, Wertheim-Mondfeld, Germany), and systemic pressure levels in the bioreactor were controlled computer aided (Hewlett-Packard, Böblingen, Germany; Software: Measure foundry, Data Translation, Bietigheim-Bissingen, Germany) and adjusted automatically when necessary. Medium samples were obtained for sterile control daily.

Histologic and Functional BioVaSc Characterization

For histologic characterization of the BioVaSc, hematoxylin-eosin (H&E) staining, and CD31, Flk-1 and VE-cadherin immunostaining were carried out as described earlier. To characterize the tissue viability, a life-dead staining was performed (LIVE/Dead Viability/CytotoxicityKit, Invitrogen, Karlsruhe, Germany). Therefore, 5×5 mm biopsies of the BioVaSc were incubated in 1 mL dye solution, consisting of 20 μL ethidium homodimer (4 μmol) and 5 μL calcein AM (Invitrogen, Karlsruhe, Germany) (2 μmol) dissolved in cell culture medium. The specimen was incubated for 30 min at 37°C. After a triplicate washing step (each: 2 mL phosphate-buffered saline for 10 min), the biopsies were analyzed by fluorescence microscopy (Axiovert 200 M, Zeiss, Göttingen, Germany) using 492 nm and 517 nm filters. To determine the cellular growth based on metabolic activity, a colorimetric assay using 3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyl tetrazolium bromide (MTT) was applied. Here, the yellow MTT dye is reduced by dehydrogenase in living cells to produce purple MTT formazan, which can be solubilized and read visually or quantified by spectrophotometric measurement at 570 nm (7, 8). For this, a biopsy was obtained from the midportion of BioVaSc and incubated in 2 mL MTT dye (3 mg/mL) and 4 mL EBM-2 for 90 min. Then, the reagent solution was transferred into a 96-well plate for photometric measurements.

Clinical Implementation

To evaluate the BioVaScs capabilities to function in a human subject by providing a functional vascular network for bioartificial organ supply, we conducted a pilot trial in a patient awaiting tracheoesophageal reconstruction of an extensive tracheoesophageal defect with an autologous bioartificial transplant. Informed consent of the patient was obtained beforehand and ethical committee approval sought from the Eberhard Karls University Tuebingen. In addition, permission for compassionate use was obtained from the medical association Nordwürttemberg. For implant generation, we isolated autologous PBMC from the patient’s blood and autologous muscle cells and fibroblasts from a myocutaneous biopsy of the left thigh. The clinical biopsy procedure is described in detail online (Supplemental Digital Content 3,

The BioVaSc was generated as described earlier. To increase tissue stability and minimize adverse immune reactions, the transplant surfaces were coated with autologous fibroblasts and muscle cells, which had to be expanded for 4 weeks in collagen-I solution (1 mg/mL) to obtain a sufficient high cell number. For co-culture implementation, 5 mL collagen I solution containing 4×105 fibroblasts were transferred (1) onto the scaffold surface and (2) into the BioVaSc lumen on time of vascular reendothelialization (Fig. 1C). To prevent cell loss, the scaffold outlets were occluded with bulldog clamps. Medium samples were obtained for sterile control daily. Before tissue implantation, the tissue culture medium was removed, and the BioVaSc was perfused for 2 hr with 50 mL serum-free EBM (4°C cold) containing 100 μL heparin, thereby cooling down the tissue to prevent tissue ischemia.

The bioartificial tissue was implanted into the left upper arm of the patient and connected end-to-side to the patient’s left brachial artery and vein after systemic administration of 5,000 IE heparin (Fig. 1D-F). The microanastomoses were performed using an operation microscope (OPMI Pentero, Carl Zeiss Meditec, Jena, Germany). The BioVaSc implant procedure can be viewed online (see Video, Supplemental Digital Content 4, The bioartificial implant was enveloped in a Gore-Tex polytetrafluoroethylene surgical membrane (W.L. Gore & Associates, Inc., Flagstaff, AZ) to prevent secondary vascularization from the surrounding soft tissue by sprouting vessels (9) (Fig. 1G). To safeguard the vascular microanastomosis and the bioartificial capillary network within the transplant, we treated the patient with high-dose anticoagulation (partial thromoboplastin time >50 sec) and aspirin (300 mg/day). Implant duration was limited to 1 week. After explantation, the bioartifical tissue was stored in 4°C cold saline solution and immediately transferred to the tissue culture laboratory for histopathologic analysis. The left brachial artery and vein were reconstructed directly without the need of repair tissues.

Characterization of the Explanted BioVaSc Transplant

For histologic and functional characterization of the explanted BioVaSc, H&E staining, CD31 and VE-cadherin immunostaining, and life-dead staining were carried out as described earlier. The objectives were (1) viability of the explanted tissue and (2) presence or absence of vascular thrombus formation within the bioartificial vessel structures. For the delineation of thrombus formation, the BioVaScs vascular tree was subdivided into five zones: (1) anastomosis area, (2) area of branching vessels, (3) microvessels, (4) transition microvessel-capillary bed, and (5) capillary bed. H&E sections of each zone were prepared, and a Ladewig staining was performed to stain organized intravascular thrombus (blue staining). Heretofore, histologic sections were incubated in a descending ethanol row (96%, 70%, and 50%) for 2 min each and washed in distilled water. Then, sections were stained with Weigert’s ferreous-hämatoxylin (2 min), phosphotungstic acid (5 min), and Ladewig dyeing (aniline blue, acid fuchsin, and helianthine) (5 min) with distilled water lavages in between. Finally, the specimen was dehydrogenized in an ascending ethanol row (96%, 99%, isopropanol I, isopropanol II). To specify the histologic findings, five representative sections of every zone (100-fold magnification) were plotted analyzing (1) vessel number, (2) vessel diameter, and (3) presence of thrombus per field of vision. For this, an AxioVert 200 M microscope (Zeiss, Jena, Germany) and related software(Axio Vision Version 4.5) were used.


Generation of an Acellular Biological Scaffold

Histologic workup confirmed the thorough decellularization of the porcine jejunal segment including its feeding artery and draining vein (Fig. 2). Cellular and DNA residues were detectable at the end of the decellularization procedure by H&E staining and the Feulgen reaction. However, subsequent tissue incubation with DNase solution resulted in their complete removal. The DNA quantification in native and decellularized tissue samples supported the histologic findings: Although the native intestinal tissue contained 26.4±1.4 μg/mg DNA, our decellularization procedure resulted in 0.5±0.01 μg/mg DNA (Fig. 3A). The decellularization - and sterilization procedure resulted in a 58% decrease in tissue dry weight (P<0.01) and a 9% increase in its water content (Fig. 3B). The extracellular matrix of the decellularized porcine jejunal segment is still composed of a high proportion elastin and collagen fibers (14% and 16%, respectively) (Fig. 3C); however, the predated tissue processing resulted in a 17% increase of nonspecific protein fibers.

Histological characterization of the decellularized porcine jejunal segment. Left column: Native porcine jejunal segment. Middle column: Porcine jejunal segment following decellularization by sodium-desoxy-cholat perfusion. Right column: Porcine jejunal segment following additional enzymatic DNA digestion. Upper row: Macroscopic appearance. Middle row: H&E staining. Lower row: Feulgen reaction denoting DNA residues.
Proteinbiochemical characterization of the acellular biological scaffold. (A) Quantitative determination of the DNA amount in the native tissue and following decellularization and DNA digestion. Rat tail collagen was used as control. (B) Dry weight and water content of the native tissue and the acellular biological scaffold following decellularization and DNA digestion. (C) Extracellular matrix composition of the native tissue and the BioVaSc following decellularization and DNA digestion.

Characterization of the Reseeded BioVaSc

The injected vascular cells (bmMSC, mEC, or PBMC) repopulated the decellularized vascular structures and differentiated into an endothelial lining expressing the characteristic vascular surface markers CD 31, VE-cadherin, and Flk-1 (Fig. 4A-C). Life-dead staining confirmed cell viability within the scaffold throughout the culture period (Fig. 4D, E). The metabolic activity of the reseeded scaffold biopsies over time illustrated a continued increase in cell number (Fig. 4F).

Histological and functional characterization of the bioartificial vascularized scaffold (BioVaSc) reseeded with human microvascular endothelial cells. (A) H&E staining. (B) CD31 immunostaining. (C) VE-cadherin immunostaining. (D, E) Life-dead staining. Viable endothelial cells are green, dead cells appear red. The tangent section displays branching capillaries lined with endothelial cells. (F) Metabolic activity of seeded BioVaSc by colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) determination over time as an indirect marker for cell growth.

Clinical Implementation

The tissue stability of the bioartificial autologous transplant allowed for standard surgical handling. Arterial and venous microanastomosis was realized using a 7-0 monofilament suture. The aggressive anticoagulation protocol resulted in hematoma formation at the operation site necessitating removal on the second postoperative day. Here, several capillaries at both margins of the BioVaSc were identified as the sole source of bleeding. The bleed was staunched by electrocautery. At explantation after 1 week, the bioengineered tissue was viable, and the embedded vascular network was patent. On bioartificial tissue explantation, we detected a hematoma surrounding the bioartificial transplant, but no active bleeding. The transplant itself appeared cell rich, fleshy, and viable. We detected a pulsation in the arterial pedicle and a venous backflow from the bioartificial transplant after disconnection of the venous anastomosis.

Characterization of the Recovered Bioartificial Transplant

Life-dead staining of the explanted BioVaSc confirmed tissue viability (Fig. 5A). The vascular endothelium within the BioVaSc still expressed endothelial cell-specific markers CD 31, VE-cadherin, and Flk-1 (Fig. 5B, C). For a meticulous histologic characterization of the transplant’s vascular patency, the recovered graft was subdivided into five zones (Fig. 5D). The applied H&E and Ladewig-staining confirmed the clinical finding of a patent vascular network. Although no evidence of thrombus formation was detectable in zones I-III, some microvessels and capillaries were occluded in zone IV and V (Fig. 5E-H). Quantitative analysis of vessel numbers and diameters per zone showed a continuous shift of singular large diameter vessels (>200 μm) at the anastomosis site toward multitudinous microvessels (50-200 μm) and capillaries (<50 μm) in the graft periphery (Fig. 6A). Overall vessel patency was 93% (Fig. 6B).

Characterization of the recovered human bioartificial implant. (A) Life-dead staining. Viable endothelial cells are green, dead cells appear red. (B) CD31 immunostaining. (C) Flk-1 immunostaining. (D) Macroscopic appearance of the recovered bioartificial transplant with the labeling of our arbitrary vascular zoning. Notice the lingering hematoma in the BioVaSc wall (*). (E, F) Representative sections of zone V (H&E and Ladewig staining). While vessel patency is not definable by H&E staining, vascular thrombus is stained blue by Ladewig staining (arrows). (G, H) Representative sections of zone III (H&E and Ladewig staining) showing the vascular diameters identified for vessel characterization.
Quantitative analysis of vessel diameters and vascular patency. (A) Allocation of capillaries (<50 μm), microvessels (50-200 μm), and major vascular structures (>200 μm) inside the BioVaSc. (B) Few intracapillary thrombus formation was detected in the graft periphery (patterned columns (left): number of open vessels per field of vision; blank columns (right): number of closed vessels).


Tissue engineering was developed originally as an alternative therapy for the treatment of tissue loss or end-stage organ failure resolving the shortage in tissues and organs for transplantation therapy (10, 11). The advances in cell and tissue culture techniques and the progress in materials research led to the development of uncounted bioartificial tissues for experimental and clinical applications. The multitudinous transplantation studies of the past displayed the elementary importance of tissue vascularization for graft function (12). Notably, the clinically realized tissue engineering applications represent simple tissues that are supplied by diffusion but not by vascularization. Therefore, one of the major obstacles for the generation of clinically applicable bioartificial tissues for reconstructive surgery and organ replacements is to keep the construct viable in vitro (during cultivation and formation of the tissue) as well as in vivo (on implantation) (13). In vivo, the construct must be vascularized immediately to allow for its survival and later integration because the host’s vascularization is not sufficient to feed the implant (13, 14). Recently, we reported the development of a primary vascularized biological scaffold (BioVaSc) in a porcine model (6), which provides a vascular tree including a capillary network for the engineered transplant that afford’s vascular anastomosis to the recipient blood supply. In continuation of these studies, we modulated our techniques for engineering vascularized human transplants.

A porcine jejunal segment was decellularized and freed from all cellular residues. The vascular structures within the remaining extracellular matrix were reseeded with human endothelial cells differentiated or isolated from (1) bmMSC, (2) PBMC, or (3) cutaneous mEC. Histologic workup confirmed the formation of a viable endothelium. Clinical short-term implantation substantiated vessel function and tissue viability.

One critical issue in the use of biological primarily xenogenic scaffolds for bioartificial tissue and organ generation is safe guarding the absence of cellular residues, which would elicit a hyperacute graft rejection in the clinical setting (15). Our findings indicate that sole decellularization procedures applying surface-active agents may not be sufficient to prevent tissue inflammation and rejection (Figs. 2 and 3). However, additional DNA digestion reliably eliminated cell residues.

Our reseeding protocol of human endothelial cells resulted in the formation of a vascular endothelium expressing the characteristic surface markers (Fig. 4) (16). Previously, we could demonstrate functional integrity of the endothelial vascular lining and its distribution throughout the entire graft by 2-[18F]-fluoro-2′-desoxy-glucose positron emission tomography (6). Now, the colorimetric identification of the BioVaScs increasing metabolic activity over time is an indirect marker for endothelial cell proliferation in the bioartificial vascular bed inside the BioVaSc and can be applied for monitoring tissue maturation in vitro (Fig. 4F). The availability of three different cell sources (bmMSC, PBMC, and mEC) for the engineering of human vascular structures suggests the possibility to tailor a bioartificial transplant to the patient’s needs, thereby minimizing the invasiveness of the used biopsy procedures (i.e., bone marrow biopsy).

The motivation for our research program is the generation of bioartificial tissue substitutes for the surgical reconstruction of complex airway and tracheoesophageal defects. Previously, we implemented less sophisticated bioartificial tissues for the treatment of a tracheobronchial insufficiency (17). We could show that the bioartificial tissue integrates into the airway and undergoes a substantial remodeling process within the first 3 months after transplantation (18). Therefore, we planned to operate a patient with 6×2-cm large tracheoesophageal fistula. The reconstruction of this extensive lesion necessitated a large bioartificial substitute with an innate vascularization to assure graft survival. The BioVaSc had been applied successfully in short-term animal experiments (19); however, no data existed whether the bioartificial vascular network would function in a human recipient and whether it would withstand arterial blood pressure in vivo. Hence, we avoided the risk of graft failure after transplantation into the thoracic cavity with consecutive bleeding or mediastinitis. Therefore, we generated an identical graft needed for tracheoesophageal reconstruction and implanted it into the left upper arm of the patient, using one half of the cells isolated from the tissue biopsies. The graft was recovered after 1 week, because tissue necrosis due to vascular obliteration would be clearly perceptible by that time on one hand, but tissue disintegration would be at its beginning—minimizing the potential associated risks like bleeding or wound infection for the patient. After analysis of the recovered implant, we estimated its function being suitable and the tracheoesophageal reconstruction was conducted the following week (20). No additional biopsy procedures were necessary for implant generation.

The transplant surfaces were seeded with fibromuscular autologous cells before implantation because the host immune response toward α-Gal (Galα1,3-Galβ1-4GlcNAc-R) oligosaccharide (Gal epitope) expressed by the acellular extracellular matrix has not been clarified yet (15, 21, 22). It has been speculated that the Gal epitope may be accountable for early bioprosthetic heart valve degeneration and calcification in younger patients, and the failure of acellular porcine heart valve scaffolds due host response induction (23, 24). In contrast, it has been hypothesized that the induced host reaction contributes to a low-level inflammatory process that aids in gradual bioartificial transplant integration (21). However, because Raeder et al. (25) have shown that the presence of Gal-epitope delays the constructive remodeling of extracellular matrix xenografts and own former studies identified a lack of remodeling and early calcification in acellular grafts, we tried to minimize adverse reactions to our transplant (26).

The objective of the clinical BioVaSc implementation was to evaluate the stability and thrombogenicity of the transplant’s innate vascular network and tissue viability after transplantation as proof of concept. Although the immunologic reaction toward bioartificial transplants is of high interest and barely investigated (4, 15, 27), this was not the focus of our study. The transplant was enveloped in a synthetic fleece to prevent vessel ingrowth from the surrounding tissue, which in theory, might have falsified the evaluation of the actual contribution of the bioartificial vascular structures to support the transplant (28). Furthermore, the short implantation duration of only 1 week makes a secondary vascularization of the transplant unlikely (12, 29). Concerning the host reaction to the bioartificial tissue, we have no histologic evidence of an immune response beyond the acute inflammatory reaction that is present after every surgical tissue injury during the first week (30, 31) (Fig. 5). Furthermore, it has to be assumed that this acute inflammatory response is triggered by both, the bioartificial transplant and the polytetrafluoroethylene surgical membrane (31). Therefore, no sound conclusions can be made regarding the transplant-mediated immune reactions. This issue should be addressed, however, in lengthier implantations by tissue biopsies, which can be obtained by bronchoscopy in the setting of bioartificial airway reconstructions (18). Independently, our histologic workup of the recovered transplant confirmed the tissue viability and the persisting tissue differentiation. Moreover, it supported the clinical finding of vessel patency at time of explantation (Fig. 6).

To conclude, we have met one important milestone that is required to engineer bioartificial organ replacements of significant size for human application: generating a functional vascular network to support the bioartificial tissue. These results may open the door for the clinical application of various sophisticated bioartificial tissue substitutes and organ replacements.


The authors thank Brigitte Hoehl for her excellent technical assistance and Nadine Reith and Antje Nagler for their support in characterizing the acellular matrix. Beate Rothe, Robert-Bosch-Hospital, Department of Laboratory Medicine, prepared the buffy-coats. The OPMI Pentero surgical microscope used for the clinical transplantations was kindly provided by Carl Zeiss Meditec (Jena, Germany).


1. Kwan MD, Longaker MT. Regenerative medicine: The next frontier. Transplantation 2008; 86: 206.
2. Birchall M, Macchiarini P. Airway transplantation: A debate worth having? Transplantation 2008; 85: 1075.
3. Vacanti JP, Langer R. Tissue engineering: The design and fabrication of living replacement devices for surgical reconstruction and transplantation. Lancet 1999; 354(suppl 1): SI32.
4. Sefton MV, Babensee JE, Woodhouse KA. Innate and adptive immune responses in tissue engineering. Semin Immunol 2008; 20: 83.
5. Jain RK, Au P, Tam J, et al. Engineering vascularized tissue. Nat Biotechnol 2005; 23: 821.
6. Mertsching H, Walles T, Hofmann M, et al. Engineering of a vascularized scaffold for artificial tissue and organ generation. Biomaterials 2005; 26: 6610.
7. Hansen MB, Nielsen SE, Berg K. Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J Immunol Methods 1989; 119: 203.
8. Thom SM, Horobin RW, Seidler E, et al. Factors affecting the selection and use of tetrazolium salts as cytochemical indicators of microbial viability and activity. J Appl Bacteriol 1993; 74: 433.
9. Loebe M, Alexi-Meskhishvili V, Wenig Y, et al. Use of polytetrafluoroethylene surgical membrane as a pericardial substitute in the correction of congenital heart defects. Tex Heart Inst J 1993; 20: 213.
10. Fuchs JR, Nasseri BA, Vacanti JP. Tissue engineering: A 21st century solution to surgical reconstruction. Ann Thorac Surg 2001; 72: 577.
11. Atala A. Tissue engineering and regenerative medicine: Concepts for clinical application. Rejuvenation Res 2004; 7: 15.
12. Walles T, Mertsching H. Strategies for scaffold vascularization in tissue engineering. In: Kendall JB, ed. Biomaterials research advances. New York, Nova Science Publishers 2007, pp 183.
13. Levenberg S, Langer R. Advances in tissue engineering. Curr Top Dev Biol 2004; 61: 113.
14. Walles T, Herden T, Haverich A, et al. Influence of scaffold thickness and scaffold composition on bioartificial graft survival. Biomaterials 2003; 24: 1233.
15. Badylak SF, Gilbert TW. Immune response to biologic scaffold materials. Semin Immunol 2008; 20: 109.
16. Schanz JE. Aufbau eines vaskuluarisierten Lebertestsystems [Generation of a vascularized liver test system]. Saarbrücken, VDM 2008; ISBN 3639065824.
17. Macchiarini P, Walles T, Biancosino C, et al. First human transplantation of a bioengineered airway tissue. J Thorac Cardiovasc Surg 2004; 128: 638.
18. Walles T, Biancosino C, Zardo P, et al. Tissue remodelling in a bioartificial fibromuscular patch following transplantation in a human. Transplantation 2005; 80: 284.
19. Schultheiss D, Gabouev AI, Cebotari S, et al. Biological vascularized matrix for bladder tissue engineering: Matrix preparation, reseeding technique and short-term implantation in a porcine model. J Urol 2005; 173: 276.
20. Walles T, Steger V, Wurst H, et al. Pumpless extracorporeal gas exchange aiding central airway surgery. J Thorac Cardiovasc Surg 2008; 136: 1372.
21. Stone KR, Abdel-Motal UM, Walgenbach AW, et al. Replacement of human anterior cruciate ligaments with pig ligaments: A model for anti-non-gal antibody response in long-term xenotransplantation. Transplantation 2007; 83: 211.
22. McPherson TB, Liang H, Record RD, et al. Galalpha(1,3)Gal epitope in porcine small intestinal submucosa. Tissue Eng 2000; 6: 233.
23. Konakci KZ, Bohle B, Blumer R, et al. Alpha-Gal on bioprostheses: Xenograft immune response in cardiac surgery. Eur J Clin Invest 2005; 35: 17.
24. Reider E, Seebacher G, Kasimir MT, et al. Tissue engineering of heart valves. Decellularized porcine and human valve scaffolds differ importantly in residual potential to attract moncytic cells. Circulation 2005; 111: 2792.
25. Raeder RH, Badylak SF, Sheehan C, et al. Natural anti-galactose alpha 1,3 galactose antibodies delay, but do not prevent the acceptance of extracellular matrix xenografts. Transpl Immunol 2002; 10: 15.
26. Walles T, Puschmann C, Haverich A, et al. Acellular implantation—No alternative to tissue engineering. Int J Artif Org 2003; 26: 225.
27. Methe H, Hess S, Edelman ER. The effect of three-dimensional matrix-embedding of endothelial cells on the humoral and cellular immune response. Semin Immunol 2008; 20: 117.
28. Eiselt P, Kim BS, Chacko B, et al. Development of technologies aiding large-tissue engineering. Biotechnol Prog 1998; 14: 134.
29. Anderson JM. Biological responses to materials. Ann Rev Mater Res 2001; 31: 81.
30. Luttikhuizen DT, Harmsen MC, Van Luyn MJ. Cellular and molecular dynamics in the foreign body reaction. Tissue Eng 2006; 12: 1955.
31. Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2008; 20: 86.

Clinical transplantation; Tissue engineering; Vascularization

Supplemental Digital Content

© 2009 by Lippincott Williams & Wilkins