Adhesion and integration of tissue engineered cartilage to porous polyethylene for composite ear reconstruction Niamh A. O’Sullivan,1 Shinji Kobayashi,1 Mitun P. Ranka,1 Katherine L. Zaleski,1 Michael J. Yaremchuk,1 Lawrence J. Bonassar,2 Mark A. Randolph1,3 1
Division of Plastic and Reconstructive Surgery, Massachusetts General Hospital, Boston, Massachusetts Department of Biomedical Engineering and Sibley School of Mechanical and Aerospace Engineering, Cornell University, Ithaca, New York 3 Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts 2
Received 11 March 2014; revised 7 July 2014; accepted 8 August 2014 Published online 6 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33269 Abstract: The objective of this study was to assess the ability of tissue engineered cartilage to adhere to and integrate with porous polyethylene (PPE) in vivo and to evaluate the biomechanical integrity of the bond formed at the interface. Porcine auricular, articular, and costal chondrocytes were suspended in fibrin gel polymer and placed between discs of PPE to form tri-layer constructs. Controls consisted of fibroblasts suspended in gel or gel alone between the discs. Constructs were implanted into nude mice for 6, 12, and 18 weeks. Upon harvest, specimens were evaluated for neocartilage formation and integration into the PPE, using histological, dimensional (mass, thickness, diameter), and biomechanical (adhesion strength, interfacial stiffness, failure energy and failure strain) analyses. Neotissue was formed in all experimental constructs, consisting mostly of neocartilage integrating with
discs of PPE. Control samples contained only fibrous tissue. Biomechanical analyses demonstrated that adhesion strength, interfacial stiffness, and failure energy were all significantly higher in the chondrocyte-seeded samples than in fibroblast-seeded controls, with the exception of costal constructs at 12 weeks, which were not significantly greater than controls. In general, failure strains did not vary between groups. In conclusion, porous polyethylene supported the growth of neocartilage that formed mechanically functional C 2014 Wiley Periodicals, Inc. J Biomed Mater bonds with the PPE. V Res Part B: Appl Biomater, 103B: 983–991, 2015.
Key Words: cartilage tissue engineering, integrative repair, biomechanical analysis, adhesion, porous polyethylene, tissue engineered ear
How to cite this article: O’Sullivan NA, Kobayashi S, Ranka MP, Zaleski KL, Yaremchuk MJ, Bonassar LJ, Randolph MA 2015. Adhesion and integration of tissue engineered cartilage to porous polyethylene for composite ear reconstruction. J Biomed Mater Res Part B 2015:103B:983–991.
INTRODUCTION
Auricular reconstruction remains one of the greatest challenges for the plastic and reconstructive surgeon and requires great technical expertise. Autologous costal cartilage grafts remain the gold standard for auricular reconstruction. However, limitations of this method include donor site morbidity, lack of sufficient donor tissue for the reconstruction, difficulty in shaping, resorption and warping of the graft materials, and the necessity for multi-stage operations.1,2 One alternative is the use of alloplastic implants, such as porous polyethylene (MEDPORTM). Despite initial reluctance, alloplasts have enjoyed increased popularity in both nasal and auricular reconstruction. Porous polyethylene has been successfully used in facial surgery for over 20 years and overcomes many of the limitations of autologous tissues.3–7 Porous polyethylene is a commercially available biocompatible, and strong material that can be easily manipulated, carved, contoured fixated, and further customized to obtain
a precise three-dimensional construct. Such properties have resulted in its popularity in aesthetic surgery. However, it is not without limitations, the foremost being the risk of extrusion and infection of the implant when placed beneath a thin layer of skin.8 Tissue engineered cartilage (TEC) could provide an alternative strategy for auricular reconstruction. Indeed, early studies have been promising where previous work has shown that it is possible to engineer neocartilage in the size and shape of human auricles and nasal tip.9–15 A number of different polymers have been combined with both xenogenic and human chondrocytes to produce neocartilage. Limitations of current experimental approaches include the lack of sufficient cells for large construct formation, difficulties with regard to thickness of TEC, and problems with the maintenance of the size and complex three dimensional shapes over time.1,10–12,14–16 A number of studies have investigated the use of PPE as an endoskeleton for TEC.11,17–21 This study goes a step
Correspondence to: M.A. Randolph; e-mail: [email protected]
C 2014 WILEY PERIODICALS, INC. V
983
further to quantify the adhesion and integration of engineered cartilage at the porous polyethylene interface. Integration at this interface would be critical for the success of a hybrid construct, as failure of the neocartilage to adhere to and integrate with the porous polyethylene could result in construct distortion and/or failure. The study also evaluates the ability of porous polyethylene to support the growth of tissue engineered cartilage from different anatomical chondrocyte sources. The ability of porcine auricular, costal, and articular chondrocytes to produce new cartilaginous matrix in vivo in the presence of PPE will be assessed both histologically and as a function of the temporal variation in dimensions of tissue engineered cartilage and porous polyethylene hybrid constructs. The ability of the neocartilage to adhere to and integrate with the PPE will also be determined through the quantification of the biomechanical integrity of the bond formed at the interface of the tissue and the material. MATERIALS AND METHODS
Overview Auricular, costal, and articular chondrocytes were harvested from swine. Chondrocytes were combined with fibrin gel polymer and placed between discs of porous polyethylene (60 constructs were made for each group) to form tri-layer constructs. Control constructs consisted of human dermal fibroblasts suspended in fibrin gel polymer or acellular fibrin gel polymer between the discs. All tri-layer constructs were implanted into nude mice for 6, 12, and 18 weeks. The mass, thickness, and diameter of the specimens were measured at the time of explantation. The specimens were then subjected to histological and biomechanical analyses. A schematic representing the methodology used is shown in Figure 1. Chondrocyte isolation and preparation The experimental protocol was conducted under Massachusetts General Hospital Institutional Animal Care and Use Committee-approved protocols in accordance with local regulations. Immediately following euthanasia, auricular, articular, and costal cartilages were harvested from 3 to 6 month old Yorkshire pigs. The unopened knee joints were prepared with a 10% povidone-iodine solution and wiped down with sterile gauze, before opening the joints and removing the articular cartilage. Porcine auricles were rinsed and scrubbed with betadine-soaked scrubbing brushes to remove any external contaminants. Costal cartilage samples were placed in baths of povidone-iodine and soaked for 30 minutes. Following removal, the cartilage was transferred to a sterile field and dissected from the overlying skin, muscle, soft tissue and perichondrium, and placed into sterile culture dishes filled with sterile phosphate buffered saline (PBS). The cartilage was washed twice with sterile PBS supplemented with antibiotic-antimycotic solution (10,000 U penicillin, 10 mg streptomycin, and 25 mg amphotericin B per milliliter in 0.9% sodium chloride). The cartilage was minced into 1 mm3 pieces using sterile razor blades.
984
O’SULLIVAN ET AL.
FIGURE 1. Schematic to summarise the experimental design. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
The chondrocytes were isolated enzymatically using the Klagsbrun method.22 A sterile 0.1% collagenase solution was prepared by mixing type II collagenase (Worthington Biochemical, Freehold, NJ) solution with Hams F-12 medium (GIBCO, Grand Island, NY) at a concentration of 1 mg/mL. The collagenase solution was filtered using a 0.2 mm filter and the 1 mm3 cartilage pieces were incubated in the collagenase solution for 14–16 hours at 37 C. Following digestion, the solution was filtered through a sterile 100 mm filter to remove undigested fragments. The filtered solution was centrifuged, and the resultant chondrocyte pellet was washed three times in PBS. The cells were re-suspended in 20 mL of standard chondrocyte media containing Ham’s F12 supplemented with 0.05 % L-ascorbic acid (Sigma, St. Louis, MO), 1 % non-essential amino acids (GIBCO, Grand Island, NY), L-glutamine (GIBCO, Grand Island, NY), antibiotic-antimycotic solution and 10% fetal bovine serum (Sigma Chemical Co, St. Louis, MO). Cell numbers were assessed using a haemocytometer and cell viability ascertained via the trypan blue dye exclusion test. The cells were re-suspended in chondrocyte media (as above) at a concentration of 40 million cells/mL. Human dermal fibroblast preparation Human dermal fibroblasts served as the control cell type. They were obtained from frozen samples of cells (cat No. CRL-1489; American Type Culture Collection). The cells were thawed according to the directions from the supplier, washed, and plated into culture flasks. After cell expansion, the fibroblasts were enzymatically isolated with 0.5% trypsin and re-suspended in media at a concentration of 40 million cells/mL. Construct preparation and assembly A sterile brass punch was used to cut 6 mm diameter discs from porous polyethylene 0.85 mm in thickness. The discs were placed in sterile culture plates and orientated such that their rougher surface would be in contact with the cells
POROUS POLYETHYLENE FOR COMPOSITE EAR RECONSTRUCTION
ORIGINAL RESEARCH REPORT
and fibrin mixture. Following cell isolation, the chondrocytes and fibroblasts were then suspended in purified porcine fibrinogen (Sigma Chemical Co., St. Louis, MO) and 0.9 % sodium chloride to reach a fibrinogen concentration of 80 mg/mL and a cellular concentration of 80 3 106 cells/ mL. Topical thrombin (Jones Pharma Incorporated, Bristol, VA) was diluted in 40 nM calcium chloride to produce a final concentration of 50 U/mL. Equal volumes of both the cell-fibrinogen suspension and the thrombin solution were combined to produce a cell-fibrin gel polymer suspension with a final concentration of 40 3 106 cells/mL. This concentration has been previously shown to produce quality neocartilage when using this fibrin-based polymer.23 Control constructs consisted of fibroblasts suspended in fibrin gel or accellular fibrin gel placed between the discs. Construct implantation Athymic male nude mice (Massachusetts General Hospital, Boston, MA) were obtained at 5–6 weeks of age and allowed to acclimate for 2 weeks. Pre-operatively the mice were anesthetized with tribromoethanol (250 mg/kg), and their skin was prepared with a povidone-iodine solution. All constructs were implanted under sterile surgical conditions. A 2–4 cm linear midline incision was made on the dorsum of each animal, blunt dissection was used to create subcutaneous pockets and a construct was placed into each pocket (four disc constructs per animal). Surgical staples were used to close the incisions, and the staples were removed on the tenth post-operative day. Construct harvest Constructs were harvested at 6, 12, and 18 weeks, photographed, and their appearance documented. Diameter and thickness were measured using a micrometer, and mass was measured and recorded. The constructs were subjected to gross, histological, and biomechanical analyses. Histological evaluation Three samples per group from each time point were randomly selected for histological evaluation. Specimens were fixed in 10 % phosphate-buffered formalin for 24 hours. After paraffin embedding, serial sections (5 lm thickness) were obtained and stained with hematoxylin and eosin and either Toluidine blue or Safranin-O and then examined with transmitted light microscopy on a Nikon Eclipse E600 microscope (Nikon). Three independent, blinded reviewers assessed the slides for presence or absence of neocartilage formation, integration with substrate and presence or absence of foreign body reactions, as assessed by presence of giant cells and lymphocytic infiltrate. Following biomechanical testing, random constructs from the experimental and control groups were fixed and stained as above and examined to determine the precise location of specimen fracture. All images were recorded with a Photometrics Coolsnap colour digital camera (Roper Scientific). Mechanical testing Following removal of the fibrous capsule surrounding the constructs, attempts were made to pull the discs apart both
manually and using forceps as described by Peretti et al.24 The mechanical integrity of the bond formed between the neocartilage and the PPE discs was then evaluated using an Enduratec EL2100 testing frame (Bose, Eden Prarie, MN) employing a biomechanical analysis protocol that has been previously reported.25 Immediately following harvest, a minimum of eight samples per group from each time point were frozen and stored at 280 C. Samples were thawed prior to biomechanical testing and the fibrotic capsule surrounding the construct was carefully removed. The thickness and diameter of the constructs was measured. The flat ends of the constructs were attached to plexiglass rods using quick-setting cyanoacrylate glue. The rods were mounted in the jaws of the Enduratec testing platform. A small volume (
Division of Plastic and Reconstructive Surgery, Massachusetts General Hospital, Boston, Massachusetts Department of Biomedical Engineering and Sibley School of Mechanical and Aerospace Engineering, Cornell University, Ithaca, New York 3 Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts 2
Received 11 March 2014; revised 7 July 2014; accepted 8 August 2014 Published online 6 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33269 Abstract: The objective of this study was to assess the ability of tissue engineered cartilage to adhere to and integrate with porous polyethylene (PPE) in vivo and to evaluate the biomechanical integrity of the bond formed at the interface. Porcine auricular, articular, and costal chondrocytes were suspended in fibrin gel polymer and placed between discs of PPE to form tri-layer constructs. Controls consisted of fibroblasts suspended in gel or gel alone between the discs. Constructs were implanted into nude mice for 6, 12, and 18 weeks. Upon harvest, specimens were evaluated for neocartilage formation and integration into the PPE, using histological, dimensional (mass, thickness, diameter), and biomechanical (adhesion strength, interfacial stiffness, failure energy and failure strain) analyses. Neotissue was formed in all experimental constructs, consisting mostly of neocartilage integrating with
discs of PPE. Control samples contained only fibrous tissue. Biomechanical analyses demonstrated that adhesion strength, interfacial stiffness, and failure energy were all significantly higher in the chondrocyte-seeded samples than in fibroblast-seeded controls, with the exception of costal constructs at 12 weeks, which were not significantly greater than controls. In general, failure strains did not vary between groups. In conclusion, porous polyethylene supported the growth of neocartilage that formed mechanically functional C 2014 Wiley Periodicals, Inc. J Biomed Mater bonds with the PPE. V Res Part B: Appl Biomater, 103B: 983–991, 2015.
Key Words: cartilage tissue engineering, integrative repair, biomechanical analysis, adhesion, porous polyethylene, tissue engineered ear
How to cite this article: O’Sullivan NA, Kobayashi S, Ranka MP, Zaleski KL, Yaremchuk MJ, Bonassar LJ, Randolph MA 2015. Adhesion and integration of tissue engineered cartilage to porous polyethylene for composite ear reconstruction. J Biomed Mater Res Part B 2015:103B:983–991.
INTRODUCTION
Auricular reconstruction remains one of the greatest challenges for the plastic and reconstructive surgeon and requires great technical expertise. Autologous costal cartilage grafts remain the gold standard for auricular reconstruction. However, limitations of this method include donor site morbidity, lack of sufficient donor tissue for the reconstruction, difficulty in shaping, resorption and warping of the graft materials, and the necessity for multi-stage operations.1,2 One alternative is the use of alloplastic implants, such as porous polyethylene (MEDPORTM). Despite initial reluctance, alloplasts have enjoyed increased popularity in both nasal and auricular reconstruction. Porous polyethylene has been successfully used in facial surgery for over 20 years and overcomes many of the limitations of autologous tissues.3–7 Porous polyethylene is a commercially available biocompatible, and strong material that can be easily manipulated, carved, contoured fixated, and further customized to obtain
a precise three-dimensional construct. Such properties have resulted in its popularity in aesthetic surgery. However, it is not without limitations, the foremost being the risk of extrusion and infection of the implant when placed beneath a thin layer of skin.8 Tissue engineered cartilage (TEC) could provide an alternative strategy for auricular reconstruction. Indeed, early studies have been promising where previous work has shown that it is possible to engineer neocartilage in the size and shape of human auricles and nasal tip.9–15 A number of different polymers have been combined with both xenogenic and human chondrocytes to produce neocartilage. Limitations of current experimental approaches include the lack of sufficient cells for large construct formation, difficulties with regard to thickness of TEC, and problems with the maintenance of the size and complex three dimensional shapes over time.1,10–12,14–16 A number of studies have investigated the use of PPE as an endoskeleton for TEC.11,17–21 This study goes a step
Correspondence to: M.A. Randolph; e-mail: [email protected]
C 2014 WILEY PERIODICALS, INC. V
983
further to quantify the adhesion and integration of engineered cartilage at the porous polyethylene interface. Integration at this interface would be critical for the success of a hybrid construct, as failure of the neocartilage to adhere to and integrate with the porous polyethylene could result in construct distortion and/or failure. The study also evaluates the ability of porous polyethylene to support the growth of tissue engineered cartilage from different anatomical chondrocyte sources. The ability of porcine auricular, costal, and articular chondrocytes to produce new cartilaginous matrix in vivo in the presence of PPE will be assessed both histologically and as a function of the temporal variation in dimensions of tissue engineered cartilage and porous polyethylene hybrid constructs. The ability of the neocartilage to adhere to and integrate with the PPE will also be determined through the quantification of the biomechanical integrity of the bond formed at the interface of the tissue and the material. MATERIALS AND METHODS
Overview Auricular, costal, and articular chondrocytes were harvested from swine. Chondrocytes were combined with fibrin gel polymer and placed between discs of porous polyethylene (60 constructs were made for each group) to form tri-layer constructs. Control constructs consisted of human dermal fibroblasts suspended in fibrin gel polymer or acellular fibrin gel polymer between the discs. All tri-layer constructs were implanted into nude mice for 6, 12, and 18 weeks. The mass, thickness, and diameter of the specimens were measured at the time of explantation. The specimens were then subjected to histological and biomechanical analyses. A schematic representing the methodology used is shown in Figure 1. Chondrocyte isolation and preparation The experimental protocol was conducted under Massachusetts General Hospital Institutional Animal Care and Use Committee-approved protocols in accordance with local regulations. Immediately following euthanasia, auricular, articular, and costal cartilages were harvested from 3 to 6 month old Yorkshire pigs. The unopened knee joints were prepared with a 10% povidone-iodine solution and wiped down with sterile gauze, before opening the joints and removing the articular cartilage. Porcine auricles were rinsed and scrubbed with betadine-soaked scrubbing brushes to remove any external contaminants. Costal cartilage samples were placed in baths of povidone-iodine and soaked for 30 minutes. Following removal, the cartilage was transferred to a sterile field and dissected from the overlying skin, muscle, soft tissue and perichondrium, and placed into sterile culture dishes filled with sterile phosphate buffered saline (PBS). The cartilage was washed twice with sterile PBS supplemented with antibiotic-antimycotic solution (10,000 U penicillin, 10 mg streptomycin, and 25 mg amphotericin B per milliliter in 0.9% sodium chloride). The cartilage was minced into 1 mm3 pieces using sterile razor blades.
984
O’SULLIVAN ET AL.
FIGURE 1. Schematic to summarise the experimental design. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
The chondrocytes were isolated enzymatically using the Klagsbrun method.22 A sterile 0.1% collagenase solution was prepared by mixing type II collagenase (Worthington Biochemical, Freehold, NJ) solution with Hams F-12 medium (GIBCO, Grand Island, NY) at a concentration of 1 mg/mL. The collagenase solution was filtered using a 0.2 mm filter and the 1 mm3 cartilage pieces were incubated in the collagenase solution for 14–16 hours at 37 C. Following digestion, the solution was filtered through a sterile 100 mm filter to remove undigested fragments. The filtered solution was centrifuged, and the resultant chondrocyte pellet was washed three times in PBS. The cells were re-suspended in 20 mL of standard chondrocyte media containing Ham’s F12 supplemented with 0.05 % L-ascorbic acid (Sigma, St. Louis, MO), 1 % non-essential amino acids (GIBCO, Grand Island, NY), L-glutamine (GIBCO, Grand Island, NY), antibiotic-antimycotic solution and 10% fetal bovine serum (Sigma Chemical Co, St. Louis, MO). Cell numbers were assessed using a haemocytometer and cell viability ascertained via the trypan blue dye exclusion test. The cells were re-suspended in chondrocyte media (as above) at a concentration of 40 million cells/mL. Human dermal fibroblast preparation Human dermal fibroblasts served as the control cell type. They were obtained from frozen samples of cells (cat No. CRL-1489; American Type Culture Collection). The cells were thawed according to the directions from the supplier, washed, and plated into culture flasks. After cell expansion, the fibroblasts were enzymatically isolated with 0.5% trypsin and re-suspended in media at a concentration of 40 million cells/mL. Construct preparation and assembly A sterile brass punch was used to cut 6 mm diameter discs from porous polyethylene 0.85 mm in thickness. The discs were placed in sterile culture plates and orientated such that their rougher surface would be in contact with the cells
POROUS POLYETHYLENE FOR COMPOSITE EAR RECONSTRUCTION
ORIGINAL RESEARCH REPORT
and fibrin mixture. Following cell isolation, the chondrocytes and fibroblasts were then suspended in purified porcine fibrinogen (Sigma Chemical Co., St. Louis, MO) and 0.9 % sodium chloride to reach a fibrinogen concentration of 80 mg/mL and a cellular concentration of 80 3 106 cells/ mL. Topical thrombin (Jones Pharma Incorporated, Bristol, VA) was diluted in 40 nM calcium chloride to produce a final concentration of 50 U/mL. Equal volumes of both the cell-fibrinogen suspension and the thrombin solution were combined to produce a cell-fibrin gel polymer suspension with a final concentration of 40 3 106 cells/mL. This concentration has been previously shown to produce quality neocartilage when using this fibrin-based polymer.23 Control constructs consisted of fibroblasts suspended in fibrin gel or accellular fibrin gel placed between the discs. Construct implantation Athymic male nude mice (Massachusetts General Hospital, Boston, MA) were obtained at 5–6 weeks of age and allowed to acclimate for 2 weeks. Pre-operatively the mice were anesthetized with tribromoethanol (250 mg/kg), and their skin was prepared with a povidone-iodine solution. All constructs were implanted under sterile surgical conditions. A 2–4 cm linear midline incision was made on the dorsum of each animal, blunt dissection was used to create subcutaneous pockets and a construct was placed into each pocket (four disc constructs per animal). Surgical staples were used to close the incisions, and the staples were removed on the tenth post-operative day. Construct harvest Constructs were harvested at 6, 12, and 18 weeks, photographed, and their appearance documented. Diameter and thickness were measured using a micrometer, and mass was measured and recorded. The constructs were subjected to gross, histological, and biomechanical analyses. Histological evaluation Three samples per group from each time point were randomly selected for histological evaluation. Specimens were fixed in 10 % phosphate-buffered formalin for 24 hours. After paraffin embedding, serial sections (5 lm thickness) were obtained and stained with hematoxylin and eosin and either Toluidine blue or Safranin-O and then examined with transmitted light microscopy on a Nikon Eclipse E600 microscope (Nikon). Three independent, blinded reviewers assessed the slides for presence or absence of neocartilage formation, integration with substrate and presence or absence of foreign body reactions, as assessed by presence of giant cells and lymphocytic infiltrate. Following biomechanical testing, random constructs from the experimental and control groups were fixed and stained as above and examined to determine the precise location of specimen fracture. All images were recorded with a Photometrics Coolsnap colour digital camera (Roper Scientific). Mechanical testing Following removal of the fibrous capsule surrounding the constructs, attempts were made to pull the discs apart both
manually and using forceps as described by Peretti et al.24 The mechanical integrity of the bond formed between the neocartilage and the PPE discs was then evaluated using an Enduratec EL2100 testing frame (Bose, Eden Prarie, MN) employing a biomechanical analysis protocol that has been previously reported.25 Immediately following harvest, a minimum of eight samples per group from each time point were frozen and stored at 280 C. Samples were thawed prior to biomechanical testing and the fibrotic capsule surrounding the construct was carefully removed. The thickness and diameter of the constructs was measured. The flat ends of the constructs were attached to plexiglass rods using quick-setting cyanoacrylate glue. The rods were mounted in the jaws of the Enduratec testing platform. A small volume (
