AN IN VITRO MODEL OF TISSUE-ENGINEERED SKIN SUBSTITUTE
WITH INTEGRATED FLOW NETWORKS IN A PERFUSION
BIOREACTOR
by
WAN-HSIANG LIANG
Submitted in partial fulfillment of the requirements
For the degree of Doctor of Philosophy
Dissertation Advisor: Dr. Harihara Baskaran
Department of Chemical Engineering
CASE WESTERN RESERVE UNIVERSITY
May, 2011 CASE WESTERN RESERVE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
We hereby approve the thesis/dissertation of
Wan-Hsiang______Liang
candidate for the ______Ph.D. degree *.
(signed)______Prof. Harihara Baskaran (chair of the committee)
______Prof. Jean Welter
______Prof. Chung-Chiun Liu
______Prof. Heidi Martin
______
______
(date) March ______25th, 2011
*We also certify that written approval has been obtained for any proprietary material contained therein. TABLE OF CONTENTS
Table of Contents ...... iii!
List of Figures ...... ix!
Preface ...... xiii!
Acknowldgements ...... xv
Abstract ...... xvii
1. Introduction ...... 1!
1.1! Purpose and Nature of the Study ...... 1!
1.2! Significance and Justification ...... 2!
1.2.1! Tissue Engineering ...... 2!
1.2.2! TE Skin Substitutes ...... 3!
1.2.3! Challenges for Tissue-Engineered (TE) Skin Substitutes ...... 4!
1.3! Guiding Research Goals ...... 4!
1.4! Previous Studies ...... 6!
1.4.1! Design Optimal Flow Networks for TE Skin Substitutes ...... 6!
1.4.2! Lithography Technique for Micropatterning of Collagen Membranes ...... 8!
1.5! Summary ...... 9!
1.6! References ...... 10!
2. Background ...... 11!
iii 2.1! Human Skin Tissue ...... 11!
2.1.1! Epidermis ...... 13!
2.1.2! Dermis ...... 16!
2.1.3! Dermo-Epidermal Junction ...... 17!
2.2! Wound Healing ...... 18!
2.2.1! Hemostasis ...... 20!
2.2.2! Inflammation ...... 20!
2.2.3! Proliferation and Reepithelialization ...... 21!
2.2.4! Remodeling ...... 22!
2.3! Tissue Engineering ...... 23!
2.4! Skin Tissue Engineering ...... 27!
2.4.1! Epidermal Substitutes ...... 29!
2.4.2! Dermal Substitutes ...... 30!
2.4.3! Composite Substitutes ...... 32!
2.5! Challenges in Skin Tissue Engineering ...... 34!
2.6! References ...... 38!
3. Development of Concentrated Collagen-Chondroitin Sulfate Scaffolds with Tunable Properties for Tissue Engineering Applications ...... 48!
3.1! Introduction ...... 48!
3.2! Materials and Methods ...... 53!
iv 3.2.1! Preparation of CG Solution ...... 53!
3.2.2! Preparation of Concentrated CG Solution and Scaffolds ...... 54!
3.2.3! Characterization of Concentrated CG Solution and Scaffolds ...... 55!
3.2.3.1! Water Uptake ...... 55! 3.2.3.2! Chondroitin Sulfate Content ...... 55! 3.2.3.3! Imaging and Image Analysis ...... 56! 3.2.3.4! Mechanical Strength Testing ...... 57!
3.2.4! CG scaffolds for TE Skin Substitutes ...... 59!
3.2.4.1! Medium Preparation: ...... 59! 3.2.4.2! Keratinocyte Isolation: ...... 59! 3.2.4.3! Skin Substitutes Preparation ...... 60! 3.2.4.4! Histoloy ...... 61!
3.2.5! Statistical Analysis ...... 61!
3.3! Results ...... 61!
3.3.1! General Scaffold Characterization ...... 62!
3.3.2! Pore Structure ...... 67!
3.3.3! Mechanical Properties ...... 68!
3.3.4! Tissue Engineering Potential ...... 71!
3.4! Discussion ...... 72!
3.5! Conclusions ...... 78!
3.6! References ...... 78!
4. Design of a Perfusion Bioreactor for Skin Tissue Engineering ...... 83!
v 4.1! Introduction ...... 83!
4.2! Materials and Methods ...... 86!
4.3! Results and Discussion ...... 92!
4.4! Conclusions ...... 96!
4.5! References ...... 96!
5. Development of an in Vitro Tissue-Engineered Skin Substitute with Integrated Flow Networks in a Perfusion Bioreactor ...... 100!
5.1! Introduction ...... 100!
5.2! Materials and Methods ...... 104!
5.2.1! Preparation of Dermal Substitutes with Integrated Flow Networks ...... 104!
5.2.1.1! Fabrication of Silicon Substrate with Optimal Flow Network Design 106! 5.2.1.2! Fabrication of CG Scaffolds with Flow Networks ...... 108! 5.2.1.3! Fabrication of PDMS Molds ...... 110! 5.2.1.4! Submerged Culture ...... 110! 5.2.1.5! Air-liquid Interface Culture ...... 111!
5.2.2! Assessment of Cell Survival on CG Scaffolds ...... 113!
5.2.3! Histological Analysis ...... 114!
5.2.4! Immunohistochemical Analysis ...... 114!
5.2.5! Cell Viability ...... 115!
5.3! Results ...... 116!
5.3.1! Micropatterning on CG Scaffolds ...... 116!
vi 5.3.2! The Effect of Seeding Density on Short-term Cell Survival ...... 116!
5.3.3! The Effect of Surface Modification on Short-term Cell Survival ...... 118!
5.3.4! The Effect of Flow Rate on Epidermis Formation ...... 120!
5.3.5! Effect of Flow Rate on Keratinocyte Viability ...... 128!
5.4! Discussion ...... 128!
5.5! Conclusions ...... 135!
5.6! References ...... 136!
6. An Adhesion Technique Based on Albumin/Glutaraldehyde Bioadhesives for Collagen Scaffolds ...... 141!
6.1! Introduction ...... 141!
6.2! Materials and Methods ...... 145!
6.2.1! Preparation of CG Scaffolds ...... 145!
6.2.2! Adhesion Technique ...... 146!
6.2.3! Mechanical testing ...... 147!
6.2.4! Cytotoxicity of BSA/GTA Bioadhesive in vitro ...... 149!
6.2.4.1! Quantification of Released GTA ...... 149! 6.2.4.2! Cytotoxicity In vitro ...... 149!
6.2.5! Fabrication of Composite CG Scaffolds with Flow Networks ...... 151!
6.2.6! MTT Interference ...... 152!
6.2.7! Statistical Analysis ...... 153!
6.3! Results ...... 153!
vii 6.3.1! Bonding Strengths of Bioadhesives ...... 153!
6.3.2! Cytotoxicity of BSA/GTA Bioadhesive in vitro ...... 155!
6.3.2.1! Release of GTA Quantification ...... 155! 6.3.2.2! In vitro Cytotoxicity of GTA Released from Bioadhesives ...... 156!
6.3.3! Microfluidics within CG Composite ...... 158!
6.3.4! MTT Interference ...... 159!
6.4! Discussion ...... 160!
6.5! Conclusions ...... 165!
6.6! References ...... 166!
7. Conclusions ...... 170!
7.1! Conclusions ...... 170!
7.2! Future Directions ...... 173!
7.3! References ...... 174!
Bibliography ...... 170!
viii LIST OF FIGURES
1. Introduction ...... 1
Figure 1-1 Tissue-engineered (TE) skin substitute with integrated vascularization ...... 2!
Figure 1-2 (a) Basic network designs (b) Rectangular and (c) square duct network designs for porosity equal to 0.4 (d) Rectangular duct designs for number of generations equal to 4 (e) Rectangular duct designs for porosity equal to 0.5 ...... 7!
Figure 1-3 (a) Schematic of casting technique for CG membranes. (b) SEM images of flow networks cast onto (A) CG membranes using micropatterning. (B) is SEM images of the smallest “island” on CG scaffolds. (C) is corresponding AutoCAD design of flow networks ...... 9!
2. Background ...... 11!
Figure 2-1 The structure of human skin...... 12!
Figure 2-2 (a) Schematic of epidermis and (b) histology of human epidermis...... 14!
Figure 2-3 Three general stages of wound healing process ...... 19!
Figure 2-4 The concept of tissue engineering...... 25!
Figure 2-5 Key developments in TE skin substitutes...... 34!
3. Development of Concentrated Collagen-Chondroitin Sulfate Scaffolds with Tunable Properties for Tissue Engineering Applications ...... 48! Figure 3-1 Schematic of preparation of concentrated CG scaffolds...... 54!
Figure 3-2 Digital images of CG solution (a) before and (b) after centrifugation...... 55!
Figure 3-3 The dogbone-shaped specimen for tensile tests...... 58!
ix Figure 3-4 The relationship between nominal collagen density and removed percentage of supernatants...... 63!
Figure 3-5 The relationship between nominal collagen density and GAG content (%, GAG weight/total scaffold weight) of the scaffolds obtained from concentrated CG solutions ...... 64!
Figure 3-6 Water uptake (a) per volume and (b) per dry scaffold weight of scaffolds obtained from concentrated CG solution ...... 66!
Figure 3-7 SEM images of (a) unconcentrated and (b) concentrated scaffolds. (c) The relationship between the pore mean diameter and nominal collagen density ...... 68!
Figure 3-8 (a) Modulus of elasticity of the scaffolds. (b) Ultimate stress of the scaffolds at failure from concentrate ...... 70!
Figure 3-9 Compression modulus of the scaffolds ...... 71!
Figure 3-10 (a) H&E staining image of TE skin substitutes obtained by using 0.8% w/v concentrated CG scaffolds in static cultures...... 72!
4. Design of a Perfusion Bioreactor for Skin Tissue Engineering ...... 83!
Figure 4-1 Digital image of the bioreactor from (a) top view and (b) side view...... 87!
Figure 4-2 Schematic diagram of the bioreactor design ...... 88!
Figure 4-3 Digital image of a bioreactor in an air-liquid interface mode using a syringe pump to perfuse medium through TE skin substitutes...... 89!
Figure 4-4 Schematic diagrams of a TE skin substitute for computational modeling...... 91!
Figure 4-5 Glucose concentration profiles in the interface (a) between the dermal analog and the flow networks and (b) between the dermal analog and the epidermal analog...... 95!
x 5. Development of an in Vitro Tissue-Engineered Skin Substitute with Integrated Flow Networks in a Perfusion Bioreactor ...... 100! Figure 5-1 Schematic of preparation of TE skin substitute in a perfusion bioreactor. ... 105!
Figure 5-2 Schematic of standard UV light lithography ...... 108!
Figure 5-3 Schematic of the fabrication method for CG sponges with flow networks. . 109!
Figure 5-4 (a) Digital picture of the patterned CG scaffold in the PDMS mold bonded by silicone adhesive. (b) Micro-CT image of the flow networks...... 111!
Figure 5-5 SEM images of flow networks cast onto (a) CG scaffolds using modified- micropatterning technique. (b) SEM image of the smallest “island” on CG scaffolds. (C) Corresponding AutoCAD design of flow networks ...... 116!
Figure 5-6 Fluorescent images of keratinocyte proliferation with seeding densities of 1 X 105 cells/cm2 and 5 X 105 cells/cm2 ...... 118!
Figure 5-7 Fluorescent images of keratinocyte proliferation on unmodified CG scaffolds and type IV collagen modified CG scaffolds ...... 120!
Figure 5-8 H&E stained sections of TE skin substitutes after 14 days of air-liquid interface culture in (a) static and perfusion culture at (b) 100, (c) 500, and (d) 1000 µL/hr perfusion rate ...... 121!
Figure 5-9 (a) Schematics of CG scaffold with flow networks. H&E stained sections of TE skin substitutes cultured at (b) 100 µL/hr and (c) 1000 µL/hr perfusion rates for 14 days...... 124!
Figure 5-10 DAPI stained sections of TE skin substitutes after 14 days of air-liquid interface culture in (a) static and perfusion culture at (b) 100, (c) 500, and (d) 1000 µL/hr ...... 125!
Figure 5-11 Immunostaining of TE skin substitutes after 14 days of air-liquid interface culture for cell nuclei (blue), CK5 (red) (a, c, e, g), and CK14 (red) (b, d, f, h) in (a)
xi and (b) static and perfusion culture at (c) and (d) 100, (e) and (f) 500, and (g) and (h) 1000 µL/hr perfusion rate ...... 127!
Figure 5-12 Effect of perfusion rate on viability of TE skin substitutes cultured for 7 days ...... 128!
6. An Adhesion Technique Based on Albumin/Glutaraldehyde Bioadhesives for Collagen Scaffolds ...... 141! Figure 6-1 Two commonly performed mechanical tests for testing bonding strength of bioadhesives ...... 145!
Figure 6-2 Schematic diagrams of bonded CG scaffolds for peel andtensile tests...... 148!
Figure 6-3 Digital images of bonded CG scaffolds undergoing peel and tensile tests in a universal testing machine...... 148!
Figure 6-4 Bonding strength of bioadhesives obtained from various concentrations of BSA and GTA ...... 154!
Figure 6-5 The release of GTA from various bioadhesive after bonding ...... 155!
Figure 6-6 The release of GTA from various bioadhesive after washing thoroughly with PBS for cell viability testing ...... 156!
Figure 6-7 SEM images of attached cells after 5 and 24 h of seeding with various bioadhesives...... 158!
Figure 6-8 Micro-CT images of microfluidics within (a) a CG composite bonded by BSA/GTA bioadhesives and (b) a CG scaffold with flow networks and a PDMS slab bonded by a silicone adhesive...... 159!
Figure 6-9 The effect of BSA or FBS/GTA on MTT reduction in the absence of cells . 160!
xii PREFACE
Each year, millions of Americans suffer tissue lose or end-stage organ failure.
Approximately a quarter of patients in need of organ transplants die while waiting for suitable donors. The current demands for transplant organs and tissues are far surpassing the supply. Tissue engineering has been proposed as an alternative to treat organ failure.
In 1987, Tissue engineering was defined by the National Science Foundation bioengineering panel, referring to the application of the principles and methods of engineering and the life sciences toward the development of biological substitutes to restore, maintain or improve function. In order to accomplish this, tissue engineering requires the interaction of three basic biologic elements: cells, biomaterial scaffolds, and in vitro cultivation. In general, functional tissue-engineered (TE) constructs are achieved by in vitro culture of biomaterial scaffolds containing cells in well-defined environments, typically achieved by the use of bioreactors.
Among all TE constructs, TE skin substitutes have enjoyed success in the treatment of both acute and chronic cutaneous wounds. TE skin substitutes typically contain two components, a dermal component and an epidermal component, mimicking the two prominent layers of the native skin. However, there are several challenges encountered by current TE substitutes such as inadequate biomaterial scaffolds to serve as dermal substitutes and lack of vasculature in vitro. The ultimate goal of this dissertation is to develop an in vitro model of TE skin substitutes with integrated vasculature in a perfusion bioreactor to overcome these limitations, presenting by a four-branched approach.
xiii In Chapter 1, we describe the purpose and nature of this study, along with a brief introduction to tissue engineering and TE skin substitutes. In Chapter 2, we describe the relationship between native human skin, wound healing process, and skin tissue engineering. We also provide information about commercially available TE skin substitutes and discuss both advantages and disadvantages of those skin substitutes. This leads to a description of current challenges for TE skin substitutes.
In Chapter 3, we describe an approach based on centrifugation to fabricate porous collagen scaffolds with optimal properties for TE skin substitutes. Physical properties are all significantly improved compared to scaffolds currently in use. These concentrated collagen scaffolds serve as dermal substitutes for the following researches in this study.
In Chapter 4 and 5, we develop an in vitro model of TE skin substitute with integrated flow networks in a perfusion bioreactor. We hypothesize the perfusion flow will improve the epidermis formation, which is confirmed through histology, immunostaining, and tissue viability. One major obstacle for fabricating micro-scale microfluidic devices within porous collagen scaffolds is no adequate bioadhesive to bond collagen scaffolds.
Chapter 6 presents an adhesion technique based on albumin/glutaraldehyde bioadhesives to resolve this obstacle.
Finally, we discuss the future directions of developing TE skin substitutes with integrated vasculature, which is accomplished by endothelialization of the flow networks. While there are still many challenges remained, we sincerely feel that each tiny step is making a huge progress in tissue engineering and believe eventually we will revolutionize healthcare.
xiv ACKNOWLEDGEMENTS
I would like to acknowledge everyone who has helped me complete this dissertation during the past five years. I would like to thank my advisor, Dr. Harihara Baskaran, for giving me research directions and guiding me towards the successful researches. Under his guidance, I have found the passion for conducting researches and truly enjoyed every moment when I discussed my idea and results with him. I would like to thank Dr. Jean F.
Welter, who has provided me many valuable suggestions and biological knowledge.
Without his help, I would not be able to finish this journey. Additionally, I would like to thank Dr. Heidi B. Marti, who is not only the member of my committee to help me finish this dissertation, but also the coordinator of graduate program in Chemical Engineering
Department to help me finish my Ph.D. Finally, I would especially like to thank Dr.
Chung-Chiun Liu, who is a wonderful mentor and provides many help not only in my campus life but also in my personal life. I will always remember our first meeting in
Taiwan back to 2006, which opens the door for me and let me experience such a wonderful life here.
I would like to acknowledge those people who have helped me make this journey success: Jim Berilla from Department of Civil Engineering, who built the bioreactors for me and helped me with experimental setup and instrumentation; Nancy Edgehouse from
Histology Core Facility, who helped me with histology process and worked with me to solve our issues; Reza Shaghi-Moshtaghin from Department of Materials Science and
Engineering, who is a truly expert of SEM and taught me how to take wonderful SEM pictures; Shu-Bin Yu from Electronics Design Center (EDC), who has been very helpful
xv for microfabrication process. I would also like to thank Dr. Joseph Mansor from
Department of Mechanical and Aerospace Engineering for mechanical testing, Adam
Kresak from Immunohistochemistry Core Facility for immunohistochemical analysis, and Kitsie Penick from Skeletal Research Center.
I would regret if I did not express my gratitude for all members of Biotransport group, both former and current, including Dr. Vijayakumar Janakiraman, Dr. Saheli Sarkar, Dr.
Brian Kienitz, Alexander L. Rivera, Chih-Ling Chou, Randall Toy, Ken Zhao, and
Hsiang-Yu Chan. I would like to thank Vijay, Saheli, and Lee. Saheli and Lee have been working with me during the past five years. We all joined our group at the same time and have been seeing each other almost every single day. I could not imagine how boring my research life will be without their company.
Finally, I would like to thank my family for supporting me and always believing that I can finish this journey. I also want to take this opportunity to thank all my best friends,
Chia-Wei Soong, Vera Chan, Pin-Ann Lin, Fang-Chia Chang, and Chang-Wen Hsieh, who are my second family and make my life in Cleveland a memorable one. Thank you and I love you all.
xvi An in Vitro Model of Tissue-Engineered Skin Substitute with
Integrated Flow Networks in a Perfusion Bioreactor
Abstract
by
WAN-HSIANG LIANG
Lack of vascularization has been suggested as one of the major limitations for current tissue-engineered (TE) skin substitutes. Until vascularization of the implanted tissue occurs, nutrients delivery and waste removal still rely primarily on passive diffusion. As a result, it is now widely accepted that TE skin substitutes should contain integrated vasculature before transplantation to improve their survival in vivo.
In this dissertation, an in vitro model of TE skin substitute was developed with integrated flow networks in a perfusion bioreactor. First, a new approach was suggested based on centrifugation for obtaining highly concentrated yet porous collagen-glycosaminoglycan scaffolds. Water uptake, morphology, mechanical properties, and tissue-engineering potential of the concentrated scaffolds were investigated. The results show that the new approach can lead to scaffolds containing four times as much as collagen as that in conventional unconcentrated scaffolds. Water uptake and mechanical properties were
xvii significantly improved. In addition, well-stratified TE skin substitutes were obtained using the concentrated scaffolds under static culture conditions.
A perfusion bioreactor system was designed for TE skin substitutes with integrated flow networks. The perfusion bioreactor provided not only a submerged culture mode for keratinocyte proliferation but also an air-liquid interface culture mode for keratinocyte differentiation. Utilizing the perfusion bioreactor, TE skin substitutes with integrated flow networks were successfully fabricated in vitro. The effect of flow on the epidermis formation was assessed through histology, immunostaining, and tissue viability. TE skin substitutes cultured at 1000 µL/h perfusion rate showed a well-stratified epidermis, along with anatomy comparable with that of control but thicker stratum spinosum and stratum corneum.
Finally, to obtain large 3D organ/tissue substitutes, an adhesion technique was developed based on albumin/glutaraldehyde bioadhesives to bond collagen scaffolds. Mechanical properties of adhesion were investigated, and the results suggest that the bioadhesives bonded collagen scaffolds very well. Biocompatibility of the bioadhesives was demonstrated in vitro. It is feasible to use the bioadhesives to obtained closed flow networks in a collagen composite. During this investigation, interference in a commonly used cell viability assay due to adhesive components was discovered.
xviii Chapter 1
INTRODUCTION
1.1 Purpose and Nature of the Study
Of all tissue engineering applications, tissue-engineered (TE) skin substitutes have achieved great success in the treatment of both acute and chronic wounds. However, lack of vascularization has been suggested as one of the major limitations for current TE skin substitutes. Vascularization allows nutrients and oxygen delivery, and waste removal between the host and the implanted skin substitutes. The overall goal of our research was to develop TE skin substitutes with integrated vasculature in a perfusion bioreactor system in vitro to eliminate the critical limitation [Figure 1-1]. This was accomplished by using a four-branched approach, which included: (i) design and characterization of optimal planar flow networks with maximum transport efficiency, (ii) fabrication of scaffolds for TE skin substitutes and micropatterning, (iii) development of a perfusion bioreactor system and culture of TE skin substitutes with integrated flow networks, and
(iv) endothelialization of the flow networks for vascularization. The first part of the approach had previously been done by our group and described elsewhere (Janakiraman et al. 2007a; Janakiraman et al. 2007b). The objective of this dissertation was to achieve the second and third parts of the approach.
1
Figure 1-1 Tissue-engineered (TE) skin substitute with integrated vascularization, which consists of a dermal analog and an epidermal analog. The epidermal analog is differentiated from keratinocytes, while the dermal analog is made up by a collagen scaffold. A flow network is integrated into dermal analog and endothelialized subsequently to obtain vascularization.
1.2 Significance and Justification
1.2.1 Tissue Engineering
Langer and Vacanti defined tissue engineering as “an interdisciplinary field that applies the principles of engineering and the life sciences toward the development of biological substitutes that restore, maintain, or improve tissue function” (Langer and Vacanti 1993).
In simple terms, the objective of tissue engineering is to regenerate functional tissues in vitro. In order to accomplish this objective, tissue engineering requires the interaction of three basic biologic elements: cells, biomaterial scaffolds, and in vitro cultivation.
2 Among these, living cells form the critical component. Sources of the cells include autologous cells from the patient, allogeneic cells from a human donor who is not immunologically identical to the patient, and xenogeneic cells from a different species.
All native tissues within the body contain biomaterials such as collagen. Biomaterial scaffolds provide three-dimensional structures for cell migration, proliferation, and differentiation. They also act as an external environment for providing signals to the resident cells (Hubbell 1995). Biomaterial scaffolds can be made of natural biomaterials or synthetic polymers. They can be formed as hydrogels, porous scaffolds, or fiber mats.
Often, functional TE constructs are achieved by in vitro culture of scaffolds containing cells in well-defined environments, typically achieved in a bioreactor.
1.2.2 TE Skin Substitutes
TE skin substitutes have enjoyed success in the treatment of both acute and chronic cutaneous wounds. TE skin substitutes typically contain two components, a dermal component and an epidermal component. This composite arrangement mimics the two prominent layers of the native skin. The dermal analog is typically an acellular biomaterial scaffold or incorporated with living cells such as fibroblasts that are constituent of the native tissue. The epidermal analog is typically obtained by the differentiation of a single cell type, keratinocytes. There are many types of TE skin substitutes. They can be classified according to many factors such as the presence or absence of fibroblasts, the use of autologous, allogeneic, or xenogenic cells or tissues, or the use of natural or synthetic biomaterials.
3 1.2.3 Challenges for Tissue-Engineered (TE) Skin Substitutes
Despite the potential to revolutionize healthcare, current TE skin substitutes and TE constructs, in general, encounter limitations such as inadequate clinical scalability of biomaterial scaffolds and inadequate mass transport for clinically useful-sized products
(Griffith and Naughton 2002). To improve nutrient transfer within TE skin substitutes, porous scaffolds are generally utilized instead of solid matrices (Freyman et al. 2001).
Collagen-based porous scaffolds, for example, are the most commonly used biomaterial scaffolds in skin tissue engineering. However, these scaffolds are still not optimized; for example, pore structure and mechanical properties of the scaffolds have not been optimized. In addition, delayed or absence of vascularization has been suggested as one of the major reasons for implanted skin failure (Boyce 1996). Without proper vascularization, implanted TE skin substitutes only rely on the host vasculature by the passive diffusion of nutrients after transplantation (Langer and Vacanti 1999). As a result, it is now widely accepted that TE skin substitutes should contain integrated vasculature before transplantation to improve their survival in vivo (Fidkowski et al. 2005).
1.3 Guiding Research Goals
The strategic goal of this research was to develop TE skin substitutes with integrated vasculature in a perfusion bioreactor. A very recent paper has shown that the survival chances of TE skin substitutes after transplantation were significantly improved due to
‘pre-existing’ capillary-like structures of TE skin substitutes, which not only developed functional anastomoses with the host’s blood vessels, but also promoted rapid and complete vascularization of the implanted tissues (Gibot et al. 2010). However, there are
4 several drawbacks in this system, such as the absence of biomaterial scaffolds to provide an adequate environment for guiding cell growth. Lack of scalability due to culturing TE skin substitutes in static conditions is another disadvantage. Most importantly, although capillary-like structures were developed in their skin substitutes, in vitro vasculature was not achieved, so the effect of vasculature on the growth of skin substitutes in vitro was not clear.
The objective of this dissertation was to develop an in vitro model of TE skin substitute with integrated flow networks and to assess the effect of such networks on the growth of
TE skin substitutes. We first developed an approach based on centrifugation to fabricate porous collagen scaffolds with optimal properties for TE skin substitutes. We demonstrated that water uptake and mechanical properties were significantly improved compared to other scaffolds currently in use. The pores in the modified scaffolds were slightly smaller compared to unmodified scaffolds, and pore walls were made up of continuous sheet of collagen fibers. We further showed that well-stratified TE skin substitutes could be obtained using the modified scaffolds under static culture conditions.
Next, we developed a perfusion bioreactor system and then cultured TE skin substitutes in the bioreactor system. We assessed the effect of different medium flow rates on the differentiation and proliferation of TE skin substitutes through histology, immunostaining, and tissue viability. TE skin substitutes cultured at highest perfusion rate (1000 µL/h) show a well-stratified epidermis, along with anatomy comparable with that of control but thicker stratum spinosum and stratum corneum.
5 Finally, to obtain large 3D organ/tissue substitutes, we developed an adhesion technique based on albumin/glutaraldehyde bioadhesives to bond collagen scaffolds. We investigated the mechanical properties of adhesion. A disadvantage of the albumin/glutaraldehyde bioadhesives is their potential toxic effects: a recipient may be exposed to the bioadhesives themselves or their residues. We therefore tested the adhesion of human keratinocytes using collagen scaffolds with the bioadhesives, and the results suggest that the bioadhesives does not have evidenced cytotoxic effect in vitro.
During this investigation, we discovered that a commonly used tissue viability method suffers from interference by the use of the adhesives. Using the adhesives, we were able to develop collagen-based microfluidic devices.
1.4 Previous Studies
The following previous studies relevant to this research, were completed in our laboratory and published in Journal of Medical Devices, 1(3);233-237, 2007 and Annals of Biomedical Engineering, 35(3):337-347, 2007 (Janakiraman et al. 2007a; Janakiraman et al. 2007b).
1.4.1 Design Optimal Flow Networks for TE Skin Substitutes
In order to obtain TE skin substitutes with integrated vasculature, bifurcating capillary flow networks with optimal transport characteristics were designed for a given pressure drop (Janakiraman et al. 2007b). A two-dimensional design was chosen due to limitations of current microfabrication techniques. The general design consists of an upstream arteriole section, where at any generation a single vessel branches into two daughter vessels, and a downstream venule section, which is a mirror image of the arteriole section
6 [Figure 1-2 (a)]. Two types of duct flow, rectangular [Figure 1-2 (b)] and square [Figure
1-2 (c)], were incorporated, and optimal designs were determined by using a generalized reduced gradient algorithm [Figure 1-2 (d) and (e)]. The results indicate that the rectangular ducts have superior mass transport characteristics compared to the square ducts. The mass transport efficiency (surface area/volume) was significantly enhanced compared to published results. This study provided the first step towards the rational design of vascular flow networks with mass transport characteristics similar to that of native tissues, through principles of optimization and mathematical modeling.
Figure 1-2 (a) Basic network designs (b) Rectangular and (c) square duct network designs for porosity equal to 0.4 and number of generations equal to 6. (d) Rectangular duct designs for number of generations equal to 4 with porosities: A = 0.45, B = 0.55, C = 0.6.
7 (e) Rectangular duct designs for porosity equal to 0.5 with number of generations: A= 2, B = 3, C = 4 (Janakiraman et al. 2007b).
1.4.2 Lithography Technique for Micropatterning of Collagen Membranes
A micropatterening technique was developed to fabricate rationally designed flow networks on biomaterial scaffolds (Janakiraman et al. 2007a). This technique included three steps: scaffold surface dissolution, feature resolution, and feature stabilization
[Figure 1-3 (a)]. Briefly, thin (~500 µm) type I collagen-GAG membranes were obtained via vacuum filtering. Acetic acid was applied to the surface briefly for selective dissolution of collagen molecules. A silicon substrate, containing a negative template of the desired features made up from photoresist, was then applied on the partially dissolved membrane surface. The formed features in the collagen membrane were then stabilized in a glutaraldehyde solution. The embedded networks were shown to be very stable with a feature resolution on the order of 2-3 µm [Figure 1-3 (b)]. Establishment of microfluidics and subsequent endothelialization of flow networks are expected to form functional vascular networks. In addition, these patterned membranes can be used as templates for topographically directed cell growth or as a model system to study various microvascular disorders where feature scales are important. Most importantly, this pressure-casting technique is extremely amenable for application in other biomaterial scaffolds and can be used for the development of complex three-dimensional TE constructs with built-in flow networks and surface topography.
8
Figure 1-3 (a) Schematic of casting technique for CG membranes. (b) SEM images of flow networks cast onto (A) CG membranes using micropatterning. (B) is SEM images of the smallest “island” on CG scaffolds. (C) is corresponding AutoCAD design of flow networks (Janakiraman et al. 2007a).
1.5 Summary
Despite TE skin substitutes have been successfully served as alternative to split-thickness skin grafts, current TE skin substitutes still face several limitations such as inadequate clinical scalability of biomaterial scaffolds and inadequate mass transport for clinically useful-sized products. Among all biomaterial scaffolds, collagen-based porous scaffolds are the most commonly used biomaterial scaffolds in skin tissue engineering. However, physical properties of these scaffolds are still not optimized. On the other hand, delayed or absence of vascularization has been suggested as one of the major reasons for
9 implanted skin failure. This dissertation aimed to overcome these critical limitations for current TE skin substitutes.
1.6 References
Boyce ST. 1996. Cultured skin substitutes: A review. Tissue Engineering 2(4):255-266.
Fidkowski C, Kaazempur-Mofrad MR, Borenstein J, Vacanti JP, Langer R, Wang YD. 2005. Endothelialized microvasculature based on a biodegradable elastomer. Tissue Engineering 11(1-2):302-309.
Freyman TM, Yannas IV, Gibson LJ. 2001. Cellular materials as porous scaffolds for tissue engineering. Progress in Materials Science 46(3-4):273-282.
Gibot L, Galbraith T, Huot J, Auger FA. 2010. A preexisting microvascular network benefits in vivo revascularization of a microvascularized tissue-engineered skin substitute. Tissue Engineering Part A 16(10):3199-3206.
Griffith LG, Naughton G. 2002. Tissue engineering - Current challenges and expanding opportunities. Science 295(5557):1009-+.
Hubbell JA. 1995. Biomaterials in tissue engineering. Nature Biotechnology 13(6):565- 576.
Janakiraman V, Kienitz B, Baskaran H. 2007a. Lithography technique for topographical micropatterning of collagen-glycosaminoglycan membranes for tissue engineering applications. Journal of Medical Devices 1(3):233-7.
Janakiraman V, Mathur K, Baskaran H. 2007b. Optimal planar flow network designs for tissue engineered constructs with built-in vasculature. Annals of Biomedical Engineering 35(3):337-347.
Langer R, Vacanti JP. 1993. Tissue engineering. Science 260(5110):920-926.
Langer RS, Vacanti JP. 1999. Tissue engineering: The challenges ahead. Scientific American 280(4):86-89.
Vacanti JP, Langer R. 1999. Tissue engineering: the design and fabrication of living replacement devices for surgical reconstruction and transplantation. Lancet 354:SI32-SI34.
10 Chapter 2
BACKGROUND
2.1 Human Skin Tissue
Skin, the largest organ of human body, is a vital organ that covers the entire outside of the body. Human skin is composed of two primary layers: a well-stratified epidermis and a dermis consisting mainly of connective tissue. Undulating between the epidermis and dermis is a dermo-epidermal junction, while below the dermis is a fatty layer, which is usually known as subcutaneous tissues or hypodermis [Figure 2-1]. The epidermis prevents body from any infection, whereas the dermis serves as a location for the appendages of the skin and is mechanically active. The dermo-epidermal junction forms mechanical support for the epidermis and also acts as a partial barrier against exchange of cells and large molecules.
11
Figure 2-1 The structure of human skin. [Source: http://www.meb.uni-bonn.de/cancer.gov/]
The skin is the largest immunologically active organ in the body. The primary functions of the skin are to: (1) act as a protective barrier between the body and the environment,
(2) arrest the penetration of pathogenic microbial agents or destructive chemicals, (3) absorb radiation from the sun, (4) serve as a storage center for water, lipids, and vitamin, and (5) prevent the loss of fluids. In addition, the skin plays an important role in regulation of body temperature, and responds to external mechanical forces. For example, the epidermis has a degree of mechanical strength to withstand damage and the ability to repair itself if damaged, and the dermis provides elasticity in response to mechanical insults.
12 2.1.1 Epidermis (Breathnach 1975; Odland 1991)
The epidermis is an avascular, well-stratified tissue mainly composed of layers of keratinocytes scattered with other cell types including melanocytes, Langerhans cells, and
Merkel cells. Keratinocytes, constituting about 95% of the epidermis, form several well- defined layers during epithelialization. These layers include stratum germinativum, stratum spinosum, stratum granulosum, and stratum corneum [Figure 2-2]. The stratum germinativum, also known as stratum basale or basal layer, is the deepest layer of the 4 layers and often described as only one cell thick. The basal cells of the stratum germinativum, which are small and cuboidal (10-14 µm), form a single cell layer and lie with their long axis perpendicular to the dermo-epidermal junction. The basal cells have large and dark-staining nuclei, dense cytoplasm containing many ribosomes, and dense tonofilament bundles. Above the stratum germinativum, the epibasal keratinocytes proliferate to form the stratum spinosum, also referred to as the spinous or prickle-cell layer. The polyhedral cells of the stratum spinosum overlie the basal cell layer and usually form a mosaic 5-10 layers thick. They become flattened with their long axis arranged parallel to the skin surface. Above the stratum spinosum is the stratum granulosum where the granular cells are flattened and their cytoplasm is filled with keratohyaline granules, deeply basophilic and irregular in size and shape. The thickness of the stratum granulosum is generally proportional to the thickness of the stratum corneum: it is, for example, only 1-3 layers thick in areas in which the stratum corneum is thin but up to 10 layers thick in areas with thick stratum corneum such as palms and soles.
The outermost layer of the epidermis is the stratum corneum, also known as horny layer, where the cells are anucleate, or dead. The thickness of the stratum corneum varies
13 depending on the amount of protection required by a region of the body. For example, the stratum corneum is thicker in palms since they are used to grip objects (Eady et al. 1998;
Murphy 2004).
Figure 2-2 (a) Schematic of epidermis and (b) histology of human epidermis. There are five layers in epidermis (from bottom to top): stratum basale, stratum spinosum, stratum granulosum, stratum lucidum, and stratum corneum.
[Sources: (a) https://lcsdanatomyphysiology.wikispaces.com/ (b) http://missinglink.ucsf.edu/lm/DermatologyGlossary/epidermis.html]
Generally, the main function of the stratum spinosum is to prevent evaporation of water and allows the skin to retain moisture due to organized layers of bipolar lipids manufactured by spinous cells. Keratin, a strong fibrous protein, is also synthesized in the stratum spinosum and forms a mesh that holds water and aids in retaining moisture within the skin. On the other hand, the major function of the stratum corneum is its capability to
14 respond to mechanical or chemical changes in the environment. In addition, due to its low permeability, the stratum corneum not only effectively retards loss of water from the inner hydrated layers, but also protects against damage from the environment (Winsor and Burch 1944).
Within the epidermis, proliferation is largely confined to the stratum basale. There are two types of proliferating keratinocytes in the stratum basale: stem cells, which have unlimited self-renewal capacity, and transit amplifying cells, which are daughters of stem cells and have a lower capacity for self-renewal (about three rounds of division) and higher probability of undergoing terminal differentiation (Hall and Watt 1989; Potten
1974; Potten 1981; Lavker and Sun 1982). Stem cell divisions give rise both to stem cells and to transit amplifying cells. Daughter of transit amplifying cells withdraw from the cell cycle and undergo terminal differentiation as they move upwards to the surface of the epidermis. Most cells in the stratum spinosum have lost their ability to divide, but they are metabolically active and larger than basal cells. They are also inevitably committed to differentiation. In the stratum granulosum, granular cells occupy an intermediate position between the spinous cells and the metabolically inactive cells in the stratum corneum.
The cells in the stratum corneum have lost their nuclei and comprise of keratin filaments, surrounded by an insoluble protein envelope (the cornified envelope) that is closely apposed to the inner face of the plasma membrane. The stage of terminal differentiation is correlated with both cell size and position (Watt 1989). For example, involucrin, a marker of terminal differentiation, was only synthesized by large suprabasal cells, suggesting that the terminal differentiation is accompanied by a decrease in substrate adhesiveness (Watt and Green 1982). In summary, keratinization includes changes in
15 keratins, envelope proteins, plasma-membrane glycoproteins, intercellular lipids, desmosomes, and other intercellular adhesion proteins (Breathnach 1975).
2.1.2 Dermis (Jarrett 1974; Montagna and Parakkal 1974)
The dermis is a supporting matrix or ground substance in which polysaccharides and protein are linked to produce macromolecules with a capability to withhold water in their domain. Collagen, which has great tensile strength and forms the major constituent of the dermis, and elastin, which makes up a small proportion of the dermis, are the main mechanically active components in the dermis. There are several cell types in the dermis, including fibroblasts, endothelial cells, mast cells, monocytes, and macrophages. The dermis also contains nerve, lymphatic vessels, and blood vessels. In human adults, the whole mass of the dermis constitutes 15-20% of total body weight.
Collagen makes up 75% of the dry weight and 18-30% of the volume of the dermis.
Collagen fibers are soft and flexible, but also strong and inelastic. In the dermis, more then 70% collagen is type I collagen and 15% type III collagen. Besides collagen, connective tissue such as dermis also contains various proteoglycans and adhesive glycoproteins, all interconnecting with each other to form an extracellular matrix (ECM) with differing combinations in different tissues. Elastin fibers in the dermis form an extensive network that also intermeshes with collagen fibers. Elastin fibers can be deformed by a small force and recover their original dimensions when the stress is relieved. Other adhesive glycoproteins in the dermis include fibronectin (Oldberg and
Ruoslahti 1982; Ruoslahti et al. 1982), vitronectin, and thrombospondin (Wikner et al.
1987). Glycosaminoglycan (GAG) is another important component in the dermis that acts
16 as a linker between certain cell-surface receptors and ECM components such as collagen, fibronectin or fibrin (Lander 1993).
Under physiological conditions, fibroblasts are the most numerous cells in the dermis.
Fibroblasts originate from the mesenchymal connective tissue, which also gives rise to chondroblasts and osteoblasts. Recently, it has been suggested that they may also arise from other mesenchymal elements such as vascular endothelium (Iwano et al. 2002). The appearance of fibroblasts varies according to their state of activity. When active, fibroblasts contain a large prominent nucleus with one or more equally prominent nucleoli. Otherwise, they are long and slender and have a nucleus that nearly fills the cells when inactive. The main function of fibroblasts is the production of collagen
(Layman et al. 1971) and elastin (Ayer 1964). Moreover, they stimulate the differentiation of endothelial cells to form vasculature by the secretion of vascular endothelial growth factor (VEGF) (Velazquez et al. 2002).
2.1.3 Dermo-Epidermal Junction (Breathnach 1964; Briggaman and Wheeler 1975;
Eady 1988)
The dermo-epidermal junction, an interface between the dermis and epidermis, is one of the largest epithelial-mesenchymal junctions in the body. The dermo-epidermal junction plays a crucial role in a wide range of epithelial-mesenchymal interactions that promote the morphogenesis and homeostasis of human skin (Eady 1988). It also involves in signaling pathway between the ECM and the basal keratinocytes, and serves as a barrier.
Basement membrane, the major constituent of the dermo-epidermal junction, has been shown to affect keratinocyte differentiation, survival, and growth (Alonso and Fuchs
17 2003). The basement membrane consists of a number of collagenous and non- collagenous macromolecules. These macromolecules are capable of binding to each other to form a matrix that facilitates the main functions of the basement membrane. Type IV collagen is the main structural constituent of the basement membrane (Yaoita et al. 1978).
Researchers have shown that the presence of type IV collagen can support the early stage of basement membrane maturation and lead to normalization of tissue organization and optimization of cell growth and survival (Fleischmajer et al. 1998; Segal et al. 2008).
2.2 Wound Healing
Cutaneous wounds, both acute and chronic, are very a common disease. Each year, approximately 500,000 people receive medical attention to address thermal injuries,
40,000 individuals are hospitalized and over 4,000 patients die because of burns in the
United States (Inc 1993). Cutaneous wound can arise from traumas such as burns, surgical treatments such as skin graft harvesting or melanoma excision, or chronic disease related to pathophysiological phenomena such as diabetic ulcers (Auger et al.
2009). Wound healing is a dynamic process that involves complex interactions between blood cells, cutaneous parenchymal cells, soluble factors, and ECM (Epstein et al. 1999).
Understanding wound healing can benefit researchers when developing and designing successful TE skin substitutes. Wound healing is generally divided into four stages: hemostasis, inflammation, reepithelialization, and remodeling, although some authors include hemostasis into inflammation stage [Figure 2-3].
18
Figure 2-3 There are three general stages of wound healing process: (a) inflammation, (b) new tissue formation, and (c) remodeling.
19 2.2.1 Hemostasis
Hemostasis, the first stage of wound healing, begins immediately after the injury has occurred unless there is a severe arterial hemorrhage. Hemostasis involves platelet aggregation, fibrin clot formation, and coagulation. To stop blood loss, injured blood vessels undergo vasospasm and release platelets, which adhere to the injured site and release growth factors such as platelet-derived growth factor (PDGF) that participate in the wound healing process. After adherence, aggregated platelets form a clot and create a surface to stop bleeding (Heldin and Westermark 1999). This blood clot not only provides a barrier to microbial invasion, but also serves as a provisional matrix for invading cells and a reservoir of growth factors. Furthermore, activated platelets secrete multiple growth factors and cytokines including PDGF and transforming growth factor
(TGF-!), which recruit and stimulate neutrophils, monocytes, and fibroblasts in the inflammation stage (Ghosh and Clark 2007). On the other hand, coagulation is activated by the release of tissue factors from endothelial cells (Slupsky et al. 1998). The coagulation involves a series of enzyme cascades in which components are activated by proteolysis. Fibrin, which forms the backbone of the clot, is the end result of the coagulation. ECM proteins such as fibronectin, vitronectin, and thrombospondin also form during the coagulation.
2.2.2 Inflammation
Inflammation, which results from the release of multiple inflammatory mediators, occurs within hours days after injury (Henry and Garner 2003). Inflammation begins when the neutrophils migrate and adhere to the injured site. The adherence and proliferation of
20 neutrophils into the injured site via the injured blood vessels are controlled through adhesion molecules that are expressed by endothelial cells. Mediators including PDGF,
TGF-!, epidermal growth factor (EGF), and fibroblast growth factor (FGF) released by platelets activate neutrophils (Arturson 1996).
The main purpose of the migration of neutrophils is to destroy any bacteria, foreign debris, or necrotic tissue in the wound area. After the area is cleaned from foreign particulates, the fibrin clot synthesized during the hemostasis provides a three- dimensional environment (Bosman and Stamenkovic 2003) where the neutrophils secrete chemoattractants that stimulate monocytes to invade the wound site and attach to ECM proteins (Brown 1995). In addition, the fibrin clot serves as a protein reservoir by binding cytokines and growth factors and amplifies their chemotactic properties by increasing local mediator concentration (Tran et al. 2004). These cell-matrix interactions then stimulate differentiation of monocytes into macrophages, which then secrete a plethora of additional cytokines into the wound (Riches 1996). Afterward, macrophages become the predominanat leukocyte in the wound after 3 days of injury.
2.2.3 Proliferation and Reepithelialization
Proliferation is triggered by the factors secreted during 4-21 days after injury. During this stage, fibroblasts are stimulated by PDGF and FGF and further migrate to the wound site to produce new ECM components including collagen, elastin, and glycosaminoglycans.
Endothelial cell migration and proliferation are stimulated by FGF from fibroblasts along with VEGF from platelets and neutrophils. Eventually, under the influence of all factors interacting with the cells, new blood vessels are formed and the capillary sprouts
21 interplay to each other to form capillary networks, a process called angiogenesis, at the wound site (Ghosh and Clark 2007).
One of the most important stages of wound healing is reepithelialization, which is accomplished by the migration of epidermal cells, keratinoctyes, into the wound site to restore the barrier function of the skin (Coulombe 1997). It is defined as the reconstitution of an organized, stratified, and squamous epithelium that permanently covers the wound area. While fibroblasts and endothelial cells invade the wound site from its bed, well-differentiated keratinocytes must detach from their neighboring cells and basement membrane, migrate over the wound site, proliferate, and finally differentiate. To facilitate migration, ECM-degrading enzymes such as collagenase, that helps to pave a pathway between the collagen-rich dermis and the fibrin-rich clot, are secreted and expressed from keratinocytes (Ghosh and Clark 2007). The reepithelialization process ultimately results in the production of a highly-organized, zipper-like basement membrane (Clark et al. 1982) and well-stratified layers of the epidermis After reepithelialization is completed, keratinocytes return to their original phenotype and become firmly attached to the underlying dermis.
2.2.4 Remodeling
The last step of wound healing is remodeling, a highly organized process that includes regulation of both the production and degradation of a new ECM. After about 3 weeks of injury, the production of collagen is matched by its degradation, and the ECM components are modified by the balanced mechanisms of proteolysis and new matrix secretion. While collagen matures, the number of intramolecular and intermolecular
22 crosslinkers increases, which improves strength and stability of collagen (Hornstra et al.
2003). The wound also gains its strength via the replacement of immature type III collagen by mature type I collagen (Johnstone and Farley 2005). The complete maturation may take months or even years, depending on the size and various factors of the initial wound.
In normal tissues, collagen fibers form a highly organized basket-weave architecture.
However, the collagen fibers in scar tissues are smaller, thinner, and randomly arranged, making the scar tissues weaker than the normal tissues. In addition, excessive contraction in large wounds can lead to contractures, while excess formation of the scar tissues can result in hypertrophic scars or keloid scars if the scar tissue extend into the healthy tissues around the wound. Depending on the wound, contraction starts around 6 days after injury and can last well after the reepithelialization stage. Myofibroblasts, differentiated from the fibroblasts during the early and middle stages of wound healing, is the major cell responsible for contraction.
2.3 Tissue Engineering
Each year, millions of Americans suffer tissue loss or end-stage organ failure. Although physicians can perform organ transplantation, surgical reconstruction, or use mechanical devices such as kidney dialyzers (Skeggs and Leonards 1948) to treat tissue loss or dysfunction, approximately a quarter of patients in need of organ transplants still die while waiting for suitable donors (1990; Lanza et al. 1997). The demand for transplant organs and tissues are far exceeds the supply, and reports indicate that this gap will continue to expand (Cohen et al. 1993; Wynn and Alexander 2011). Tissue engineering
23 has therefore been suggested as an alternative solution to treat organ failure (Langer and
Vacanti 1993; Nerem and Sambanis 1995).
Tissue engineering, defined by the National Science Foundation bioengineering panel in
1987, refers to the application of the principles and methods of engineering and the life sciences toward the development of biological substitutes to restore, maintain or improve function [Figure 2-4]. For some organs such as bioartificial livers (Rozga et al. 1994;
Shatford et al. 1992), the first step may involve the development of extracorporeal devices; however, the long-term goal of tissue engineering is to develop the functional tissues and organs to fully eliminate the need for transplant organs. As such, tissue engineering can be defined as the development of devices or processes that utilize living cells and biomaterial scaffolds to generate the functional tissues or organs in vitro for subsequent implantation. Newer areas of gene therapy, regenerative medicine, and stem cell-based organogenesis are explicitly excluded in this study. Based on our definition, tissue engineering generally requires three components: cells, biomaterial scaffolds, and in vitro cultivation.
24
Figure 2-4 The concept of tissue engineering (Fuchs et al. 2001).
If cells are the key to the tissue engineering, the advent of mammalian cell culture technology such as growing mammalian cells outside of the body opened the door for this field. Sources of cells include autologous cells from the patient, allogeneic cells from a human donor who is not immunologically identical to the patient, and xenogeneic cells from a different species. In the case of skin, researchers have shown that the presence of autologous cells promote the wound healing process, especially when spontaneous epithelialization is impossible or too long to be completed (Gallico et al. 1984).
Allogeneic cells have been used successfully to treat cutaneous wounds, especially chronic venous ulcers (Falanga et al. 1998; Leigh et al. 1987). Although the prospect of using xenogeneic cells in tissue engineering remains controversial due to the potential for transmitting animal pathogens to humans, xenogeneic cells could be used as temporarily solutions while waiting for a suitable donor or until the tissue repairs itself.
25 Biomaterial scaffolds, on the other hand, have been developed to mimic the characteristics of the native tissues, such as eliciting specific cellular functions and directing cell-cell interactions (Hubbell 1995). When designing the biomaterial scaffolds, some requirements have to be met. For example, they should be biocompatible both in bulk and degraded form, have appropriate mechanical properties to support cell growth, and be porous and permeable to permit cells encapsulation and nutrients diffusion.
Furthermore, they are required to be both chemically and structurally stable, not just in terms of their handling characteristics, but in terms of keeping cells alive and functional after implantation (Nerem and Sambanis 1995).
The biomaterial scaffolds can be made of natural extracellular matrix molecules such as collagen (Freyman et al. 2001), chitosan (Madihally and Matthew 1999), hyaluronic acid
(Aigner et al. 1998), and gelatin (Kang et al. 1999). Although natural biomaterials may most closely simulate the native cellular environment, large batch-to-batch variations upon isolation from biological tissues is one of the drawbacks of natural biomaterials
(Angelova and Hunkeler 1999). Poor mechanical properties of scaffolds made from natural biomaterials also limit their applications (Lavik and Langer 2004). Synthetic materials such as poly(glycolic acid) (PGA) (Freed et. al 1994), poly(ethylene glycol)
(PEG) (Burdick and Anseth 2002), and poly(lactic-co-glycolic acid) (PLGA) (Li et al.
2002) have also been developed. Most synthetic polymers are degraded via chemical hydrolysis and insensitive to enzymatic processes; their degradation, therefore, does not vary from patient to patient (Lutolf and Hubbell 2005).
The biomaterial scaffolds for TE constructs can be classified into two different categories: hydrogels and porous scaffolds. The hydrogels have been used in numerous
26 applications in tissue engineering to study cell-cell and cell-matrix interactions due to their structural similarity to the macromolecular-based components in the body (Lee and
Mooney 2001). However, the hydrogels are rarely used as scaffolds for TE skin substitutes due to lack of adequate mechanical properties, which are not only necessary to support cell growth, but play an important role during surgeries. The other category of biomaterial scaffolds is porous scaffolds. Because of their large surface area, porous scaffolds can provide appropriate spatial organization for cell attachment and growth.
Another advantage of porous scaffolds is their ability to allow diffusion of nutrients and removal of waste between host and implanted tissues. This is very important for regeneration of highly metabolic organs such as liver and pancreas (Yang et al. 2001).
2.4 Skin Tissue Engineering
Decades ago, it is impossible to imagine growing TE skin substitutes to repair human skin. Initially, to treat injuries requiring substitution of skin, the only options were either autologous or allogeneic split or full-thickness skin grafts, tissue flaps, and free-tissue transfers (Pomaha! et al. 1998). Over the last 35 years, however, several TE skin substitutes have been developed. Although autografts are still the best solution to any injury requiring substitution of skin, TE skin substitutes may serve as alternatives to the autografts, especially when patients suffer extensive burns (>60% total body surface area), and provide better healing (Eaglstein et al. 1995). Moreover, they can be used to as an experimental model in dermatology and toxicology (Goldberg and Frazier 1989).
Successful TE skin substitutes would benefit patients with either acute or chronic wounds.
In burn victims, especially those suffering from an acute life-threatening situation with
27 large and deep burn wounds, little healthy skin is left for split-thickness skin grafts. In these cases, TE skin substitutes will become the only treatment for permanent healing.
On the other hand, although on demand production of TE skin substitutes is not important to the patients with chronic or elective wounds, matching size, texture, and color may be considered and designed carefully before transplantation. From a clinical perspective, there are several factors that should be considered when developing TE skin substitutes, including time required to prepare TE skin substitutes, time required for implanted tissue to achieve vascularization, the risk of immune response and microbial contamination, mechanical stability of implanted tissue, and finally the cost.
TE skin substitutes can be divided into three categories: epidermal substitutes, dermal substitutes, or composite substitutes. Successful epidermal substitutes must provide barrier functions, prevent transepidermal water loss, and minimize microbial contamination. Successful dermal substitutes, on the other hand, must provide mechanical support to epidermis, allow for rapid vascularization, and provide overall mechanical stability and durability of skin. From the tissue engineering point of view, TE skin substitutes refer to skin substitutes made of cells only, biomaterial scaffolds only, or combinations of cells and scaffolds. There are many types of TE skin substitutes, and they can be generally classified to (1) cellular or acellular skin substitutes, (2) autologous, allogeneic or xenogenic cells or tissues, (3) epidermal substitutes, dermal substitutes or composite skin substitutes.
28 2.4.1 Epidermal Substitutes
Most frequently, the epidermal substitutes consist of cultured keratinocytes. Billingham et al. first recognized the potential for transplantation of a “sheet of pure epidermis” to cover large full-thickness skin defects (Billingham and Reynolds 1952). However, the major breakthrough in skin tissue engineering was made when Rheinwald and Green developed a technique to culture human keratinocytes on a layer of irradiated 3T3 murine fibroblast in vitro and obtain the first stratified colonies of keratinocytes from a single cell (Rheinwald and Green 1975). Keratinocytes can undergo about 50 to 60 population doublings if cultured from neonatal foreskin using this method. Due to successful of culturing keratinocytes, autologous keratinocyte sheets were used in burn patients
(O'Connor et al. 1981) and tested worldwide (Cuono et al. 1986; Gallico et al. 1984;
Munster 1996; Ronfard et al. 2000). Researchers have shown that keratinocytes synthesize basal-lamina proteins in vitro and restore the basement membrane in vivo
(Gallico et al. 1984). In addition, the sheets provide rapid coverage of the wound, good pain relief, and functional and cosmetic outcomes. There are several disadvantages of the epidermal substitutes such as grafts instability and wound contraction due to the absence of the a dermal component (Hafemann et al. 1999). Moreover, 3 weeks are required for the graft cultivations.
Cultured epidermal autografts (EpicelTM) can provide permanent coverage of large area from a skin biopsy (Carsin et al. 2000; Hefton et al. 1986; O'Connor et al. 1981). They have been used to treat burns (Gallico et al. 1984; O'Connor et al. 1981), cutaneous wounds, and chronic leg ulcers (Hefton et al. 1986). However, a lag period of 3 weeks is the major drawback of Epicel. Conversely, cultured epidermal allografts are available
29 immediately and no biopsy is required. They can be grown in advance and cryopreserved
(Bolivar-Flores et al. 1990). They promote granulation formation and stimulate epithelialization from wound edges and from adnexal structures in the dermis of superficial wounds (Phillips 1993). The downside of the cultured epidermal allografts is that they do not survive permanently on the wound.
2.4.2 Dermal Substitutes
In 1981 Bell et al. developed the first skin-equivalent consisting of fibroblasts cast in collagen lattices and successfully tested it in an animal model (Bell et al. 1981). Since then, several dermal substitutes have been developed to overcome some of the above- mentioned shortcomings when using epidermal substitutes. The dermal substitutes consist of biomaterials scaffolds with or without cultured fibroblasts. There are several advantages of the dermal substitutes. They provide mechanical support as well as elasticity and tensile strength. They also act as an anchor for epithelial glands and keratinizing appendage structures. Very importantly, the presence of dermal substitutes enhances reepithelialization, inhibits wound contraction (Branski et al. 2007), and improves esthetic outcome (Kangesu et al. 1993). The biomaterial scaffolds in the dermal substitute can be made of synthetic or natural biomaterials. Among all natural biomaterials, collagen, the most abundant protein in human body, is one of the most commonly used natural biomaterials in skin tissue engineering. Hyaluronic acid has also shown potential as a dermal analog due to its involvement in scar-less wound healing in the fetus (Adzick and Longaker 1992). Among synthetic materials, PGA/PLA grafts have shown lower microbial contamination potential than collagenous grafts (Hansbrough et al.
1992), although cell viability in the grafts decreases with time due to release of acidic
30 monomers during degradation. Cultured fibroblasts are not essential to regenerate TE skin substitutes; however, they have been shown to promote establishment of the dermo- epidermal junction and epithelial-mesenchymal interactions.
An acellular dermal allograft with an intact basement membrane (Alloderm") was developed by removing the epidermis and cellular materials from fresh cadaver skin
(Callcut et al. 2006; Wainwright 1995) and was approved by the FDA for the treatment of burns in 1992. Wainwright et al. demonstrated that wound healing from using acellular dermal allografts with thinner split-thickness autografts were equivalent to that using thicker split-thickness autografts alone (Wainwright et al. 1996). The acellular dermal allografts have been used in the treatment of facial soft tissue defects with excellent results (Achauer et al. 1998). The advantages of the acellular dermal allografts include:
(1) they are immunologically inert due to the absence of the living cells; (2) they provide a template with a natural dermal structure for regeneration including the presence of the an intact basement membrane; and (3) they allow for the use of thinner autografts.
However, the risk of transmitting infectious diseases and requirement of two surgical procedures limit their applications. Another acellular dermal substitute composed of collagen and chondroitin-6-sulfate (dermal analog) with silicone backing (epidermal analog) (Integra") was developed to regenerate neodermis (Branski et al. 2007;
Moiemen et al. 2006; Stiefel et al. 2010). After the implanted dermal analog is vascularized, the disposable epidermal analog is removed and replaced with split thickness skin grafts. Again, Integra requires two separates surgical procedures.
Dermagraft", made up of a bioabsorbable polyglactin mesh scaffold seeded with allogeneic neonatal fibroblasts, is another commercially available dermal substitutes
31 (Cooper et al. 1991). Dermagraft is considered to facilitate the healing process due to secretion of growth factors and deposition of dermal matrix proteins (Pham et al. 2007;
Wong et al. 2007). It also provides good resistance to tearing. Dermagraft is generally used for chronic venous or diabetic foot ulcers (Wong et al. 2007).
Either epidermal substitutes alone or dermal substitutes alone would not provide permanent solutions to full-thickness skin defects. Without dermal substitutes, grafts would not obtain enough support for epidermis growth and provide mechanical support, while without epidermal substitutes reepithelialization would not initiate. An autologous composite substitute, composed of a stratified epithelium on a fibroblast-containing matrix of collagen and glycosaminoglycan, was first proposed in the late 1980s. It is now widely accepted that a combination of either autologous or allogeneic keratinocytes with acellular biomaterial matrices is the minimum essential requirement to treat skin defects successfully.
2.4.3 Composite Substitutes
Next major progress was in TE skin was the development of a bilayer artificial skin composed of a temporary Silastic epidermis and an acellular porous collagen-condroitin-
6-sulfate dermis (Burke et al. 1981). The results showed that the implanted dermis was populated with the host fibroblasts and blood vessels after grafting, and the anatomic structure of the implanted dermis was the same as that of the normal dermis. This acellular skin substitute is now commercially available as Integra", consisting of an artificial dermal analog (matrix of bovine collagen and chondroitin-6-sulfate) and a disposable epidermal analog (silicone sheet). The invention of Integra and the appealing
32 idea of combining cultured keratinocytes with Integra were certainly a major step in skin tissue engineering since it finally provides patients with a dermis off-the-shelf and a laboratory grown epidermis. However, reality has shown that cultured autologous keratinocytes do not take well on the dermis produced by Integra (Pandya et al. 1998). As a result, a thin split-thickness autograft is required for Integra to be used in the treatment of large wounds.
On the other hand, based on this technique developed by Bell et al., Boyce et al. developed TE skin substitutes from autologous fibroblasts and keratinocytes in collagen- glycosaminoglycan substrates and compared them to split-thickness autografts in the treatment of massive burns. The results showed that there was no difference between TE skin substitutes and split-thickness autografts (Boyce et al. 1995). This technique was then transformed into Apligraf" (also know as ‘Graftskin’). Apligraf, a bilayer substitute containing allogeneic keratinocytes, fibroblasts, and bovine type I collagen matrix, was approved by the FDA for the treatment of venous ulcers. Dermal fibroblasts are inosculated in a collagen matrix followed by keratinocytes seeded on top and subsequently exposed to an air-liquid interface to trigger differentiation and form functional skin substitutes (Wilkins et al. 1994). The advantages of Apligraf include ease of application, able to apply the graft as an outpatient procedure, and avoidance of a surgical procedure that leaves the wound site open. However, high cost and short shelf life are the main drawbacks of Apligraf.
Because of the effort to establish dermo-epidermal skin substitutes, researchers have revealed the importance of interactions between fibroblasts and keratinocytes. Although the presence of living fibroblasts is not necessary in the dermal substitutes, it benefits
33 wound healing process by regulation of matrix deposition (type I, type IV collagen, elastin, and laminin) (Demarchez et al. 1992; Marks et al. 1991; Okamoto and Kitano
1993), epidermal differentiation (El-Ghalbzouri et al. 2002) and dermal regeneration
(Ayer 1964; Layman et al. 1971), and also providing FGF (Delvoye et al. 1988; Konig and Bruckner-Tuderman 1991).
Figure 2-5 Key developments in TE skin substitutes.
2.5 Challenges in Skin Tissue Engineering
34 Although TE skin substitutes have shown success in the treatment of burns, soft tissue trauma, cutaneous wounds, and chronic ulcers, there are still some major challenges in skin tissue engineering. The first challenge is the source of cells. As mentioned before, sources of cells for TE skin substitutes include autologous, allogeneic, and xenogeneic cells. Although autologous cells are the best option, it may take up to 3 weeks before TE skin substitutes are ready for implantation. Conversely, allogeneic cells may be isolated and cryopreserved in advance, thereby overcome the limitation from autologous cells.
However, the risk of immune response against the implanted tissues is a major concern when applying allogeneic cells. Regardless of the origin of the cells, whether autologous or allogeneic, the major drawback is that the primary cells obtained from human donors are often restricted by low expansion capabilities. In addition, their phenotypes may adapt to in vitro condition, and further limit their ability to form new tissues (Schnabel et al.
2002).
Other alternative sources of cells for TE skin substitutes are adult and embryonic stem
(ES) cells. Both of them have shown great promises for treating patients, especially for those who have limited donor sites, such as burn victims with more than 60% of total body surface area injured. ES cells have great potential because they can be expanded from an undifferentiated state in vitro and further induced to form different cell types or tissues. However, since they are primarily obtained from the inner cell mass of a week- old embryo (Thomson et al. 1998), use of ES is controversial. The next option is adult stem cells such as epithelial stem cells and bone marrow stem cells. In skin tissue engineering, epithelial stem cells have been identified in the hair follicle bulge, which is at the base of the epithelial compartment. Bulge cells possess stem cell characteristics,
35 including multipotency, high proliferative potential, and their cardinal feature of quiescence. Although bulge cells only contribute to the epidermis during wound healing process, they can regenerate new hair follicles, epidermis, and sebaceous glands when combined with neonatal dermal cells. Bulge cells provide an alternative cell source in skin tissue engineering (Cotsarelis 2006). Autologous bulge cells, however, still face the same limitations as autologous keratinocytes.
Another challenge in skin tissue engineering is the optimization of biomaterial scaffold properties. There are several requirements that have to be considered properly in the design of scaffolds for tissue engineering. The material and its architecture, such as hydrogels and porous sponges, are important consideration. The geometric shape often affects the behavior and properties of the scaffolds, thereby should be optimized. For example, a thicker dermal analog will have superior mechanical strength but keratinocytes will have less nutrient availability. Moreover, new strategies in skin tissue engineering should enable the control of the microenvironment at the nano-, micro-, and macroscale (Nelson and Bissell 2005). The scaffolds can be either fabricated with highly organized or random architectures. Interestingly, although nanopores are too small to direct cell migration and infiltration, researchers have shown that nanopores can promote gas and nutrient diffusion, and further support cell survival (Muschler et al. 2004).
Regardless of whether the scaffolds are hydrogels or porous sponges, mechanical properties have to match the native tissues to provide the correct stress environment for the implanted tissue. The mechanical properties are mainly determined by the characteristics of the material that make up the bulk of the scaffolds and their architecture and geometry. For skin tissue engineering, the scaffolds with the comparable mechanical
36 properties to the native skin tissues can facilitate the differentiation of the cells, mediate cellular activity, and further promote the wound healing process (Muschler et al. 2004).
The Young’s modulus of the in vivo human skin has been measured between 0.4 and 0.8
MPa (Agache et al. 1980). However, the Young’s modulus of the pure collagen-based scaffolds has been estimated only about 25 kPa (Roeder et al. 2002). Therefore, the mechanical properties of collagen-based scaffolds should be improved to provide better wound healing process. Other factors that need to be considered include: scaffold surface structure and chemistry for cell attachment, its degradation rate, the effects of the degradation productions on short- and long-term cytotoxicity, the rate at which the scaffolds lose their mechanical stability, and the ability to encapsulate cells and proteins
(Kohn 2004; Yang et al. 2001).
Among all challenges in skin tissue engineering, inadequate mass transfer has been suggested as one of the major reasons for implanted skin failure. This is a critical problem that affects almost all TE constructs. In vivo, skin microvasculature provides excellent support for mass transport. After transplantation, vasculariziation is too slow in current TE skin substitutes to provide enough nutrients to the epidermis when the dermal analogs reach a threshold thickness. Hence, most commercially available dermal substitutes thicker than 1 mm require two-step surgery to avoid the implanted epidermis necrosis. Tremblay et al. have shown that inosculation of preexisting capillaries in TE skin substitutes with the host’s vasculature only takes less than 4 days (Tremblay et al.
2005). In comparison, 14 days are required to achieve a similar result when using TE skin substitutes without preexisting capillaries. Recently, Gibot et al. have shown that a preexisting capillary-like structures improved revascularization of TE skin substitutes in
37 vivo (Gibot et al. 2010). Collectively, these results also suggest that pre-vascularization is necessary to overcome the mass transfer limitation and further improve the survival chance of implanted skin substitutes.
2.6 References
1990. U.S. Scientific Registry for Organ Transplantation and the Organ Procurement and Transplant Network. Annual report. Richmond VA.
Achauer BM, VanderKam VM, Celikoz B, Jacobson DG. 1998. Augmentation of facial soft-tissue defects with Alloderm dermal graft. Annals of Plastic Surgery 41(5):503-507.
Adzick NS, Longaker MT. 1992. Scarless fetal healing - Therapeutic implications. Annals of Surgery 215(1):3-7.
Agache PG, Monneur C, Leveque JL, Rigal J. 1980. Mechanical properties and Young's modulus of human skin in vivo. Archives of Dermatological Research 269(3):221- 232.
"#$%&'! ()! *&$&+&'! ()! ,-./+&'! 0)! 1234566#2! 7)! 028&9#5! ")! ,233&'! 1)! :29.&%;2-&'! <)! =2-32%%! ">! ?@@A>! 12'.#+2$&! .#99-&! &%$#%&&'#%$! B#.C! %58&+! %5%B58&%! 9.'-6.-'&D! ;#532.&'#2+! ;29&D! 5%! CE2+-'5%#6! 26#D! ;&%/E+! &9.&'>! (5-'%2+! 5F! G#53&D#62+!H2.&'#2+9!I&9&2'6C!JKLKMN?OKP?A?>
Alonso L, Fuchs E. 2003. Stem cells of the skin epithelium. Proceedings of the National Academy of Sciences of the United States of America 100(Suppl 1):11830-11835.
"%$&+582!=)!,-%Q&+&'!7>!?@@@>!I2.#5%2+#/#%$!.C&!D&9#$%!5F!45+E3&'#6!;#532.&'#2+9>! *'&%D9!#%!G#5.&6C%5+5$E!?OL?RMNJR@PJK?>
Arturson G. 1996. Pathophysiology of the burn wound and pharmacological treatment. The Rudi Hermans lecture, 1995. Burns 22(4):255-274.
Auger FA, Lacroix D, Germain L. 2009. Skin substitutes and wound healing. Skin Pharmacology and Physiology 22(2):94-102.
Ayer JP. 1964. Elastic tissue. In: Hall DA, editor. International Review of Connective Tissue Research. New York: Academic Press. p 33-100.
Bell E, Ehrlich HP, Buttle DJ, Nakatsuji T. 1981. Living tissue formed in vitro and accepted as skin-equivalent tissue of full thickness. Science 211(4486):1052-4.
38 Billingham RE, Reynolds J. 1952. Transplantation studies on sheets of pure epidermal epithelium and on epidermal cell suspensions. British Journal of Plastic Surgery 5(1):25-36.
Bolivar-Flores J, Poumian E, Marsch-Moreno M, de Oca GM, Kuri-Harcuch W. 1990. Use of cultured human epidermal keratinocytes for allografting burns and conditions for temporary banking of the cultured allografts. Burns 16(1):3-8.
Bosman FT, Stamenkovic I. 2003. Preface to extracellular matrix and disease. Journal of Pathology 200(4):421-422.
Boyce ST, Goretsky MJ, Greenhalgh DG, Kagan RJ, Rieman MT, Warden GD. 1995. Comparative assessment of cultured skin substitutes and native skin autograft for treatment of full-thickness burns. Annals of Surgery 222(6):743-752.
Branski LK, Herndon DN, Pereira C, Mlcak RP, Celis MM, Lee JO, Sanford AP, Norbury WB, Zhang X-J, Jeschke MG. 2007. Longitudinal assessment of Integra in primary burn management: A randomized pediatric clinical trial *. Critical Care Medicine 35(11):2615-2623.
Breathnach AS. 1964. The dermo-epidermal junction. In: Rook A, Champion RH, editors. Progress in the Biological Sciences in Relation to Dermatology. Cambridge: Cambridge University Press. p 415-425.
Breathnach AS. 1975. Aspects of epidermal ultrastructure. Journal of Investigative Dermatology 65(1):2-15.
Briggaman RA, Wheeler CE, Jr. 1975. The epidermal-dermal junction. Journal of Investigative Dermatology 65(1):71-84.
Brown EJ. 1995. Phagocytosis. Bioessays 17(2):109-117.
G-'D#6Q!(")!"%9&.C!:S>!KRRK>!0C5.5&%6249-+2.#5%!5F!59.&5;+29.9!#%!#%T&6.2;+&!IU7P 35D#F#&D! 0
Burke JF, Yannas IV, Quinby WCJ, Bondoc CC, Jung WK. 1981. Successful use of a physiologically acceptable artificial skin in the treatment of extensive burn injury. Annals of Surgery 194(4):413-428.
Callcut RA, Schurr MJ, Sloan M, Faucher LD. 2006. Clinical experience with Alloderm: A one-staged composite dermal/epidermal replacement utilizing processed cadaver dermis and thin autografts. Burns 32(5):583-588.
Carsin H, Ainaud P, Le Bever H, Rives JM, Lakhel A, Stephanazzi J, Lambert F, Perrot J. 2000. Cultured epithelial autografts in extensive burn coverage of severely traumatized patients: a five year single-center experience with 30 patients. Burns 26(4):379-387.
39 Clark RAF, Lanigan JM, Dellapelle P, Manseau E, Dvorak HF, Colvin RB. 1982. Fibronectin and fibrin provide a provisional matrix for epidermal cell migration during wound reepithelialization. Journal of Investigative Dermatology 79(5):264-269.
Cohen S, Bano MC, Cima LG, Allcock HR, Vacanti JP, Vacanti CA, Langer R. 1993. Design of synthetic polymeric structures for cell transplantation and tissue engineering. Clinical Materials 13(1-4):3-10.
Cooper ML, Hansbrough JF, Spielvogel RL, Cohen R, Bartel RL, Naughton G. 1991. In vivo optimization of a living dermal substitute employing cultured human fibroblasts on a biodegradable polyglycolic acid or polyglactin mesh. Biomaterials 12(2):243-248.
Cotsarelis G. 2006. Epithelial stem cells: A folliculocentric view. Journal of Investigative Dermatology 126(7):1459-1468.
Coulombe PA. 1997. Towards a molecular definition of keratinocyte activation after acute injury to stratified epithelia. Biochemical and Biophysical Research Communications 236(2):231-238.
Cuono C, Langdon R, McGuire J. 1986. Use of cultured epidermalvautografts and dermal allografts as skin replacement after burn injury. Lancet 1(8490):1123-1124.
Delvoye P, Pierard D, Noel A, Nusgens B, Foidart JM, Lapiere CM. 1988. Fibroblasts induce the assembly of the macromolecules of the basement membrane. Journal of Investigative Dermatology 90(3):276-282.
Demarchez M, Hartmann DJ, Regnier M, Asselineau D. 1992. The role of fibroblasts in dermal vascularization and remodeling of reconstructed human skin after transplantation onto the nude mouse. Transplantation 54(2):317-326.
Eady RAJ. 1988. The basement membrane: interface between the epithelium and the dermis: structural features. Archives of Dermatology 124(5):709-712.
Eady RAJ, Leigh IM, Pope FM. 1998. Anatomy and organization of human skin. In: Champion RH, Burton JL, Burns DA, Breathnach SM, editors. Rook/Wilkinson/Ebling Textbook of Dermatology. Oxford: Blackwell Science Ltd. p 37-112.
Eaglstein WH, Iriondo M, Laszlo K. 1995. A composite skin substitute (graftskin) for surgical wounds: a clinical experience. Dermatologic Surgery 21(10):839-843.
El-Ghalbzouri A, Gibbs S, Lamme E, Van Blitterswijk CA, Ponec M. 2002. Effect of fibroblasts on epidermal regeneration. British Journal of Dermatology 147(2):230-243.
40 Epstein FH, Singer AJ, Clark RAF. 1999. Cutaneous wound healing. New England Journal of Medicine 341(10):738-746.
Falanga V, Margolis D, Alvarez O, Auletta M, Maggiacomo F, Altman M, Jensen J, Sabolinski M, Hardin-Young J, and the Human Skin Equivalent Investigators Group. 1998. Rapid healing of venous ulcers and lack of clinical rejection with an allogeneic cultured human skin equivalent. Archives of Dermatology 134(3):293- 300.
Fleischmajer R, Utani A, MacDonald Ii ED, Perlish JS, Pan TC, Chu ML, Nomizu M, Ninomiya Y, Yamada Y. 1998. Initiation of skin basement membrane formation at the epidermo-dermal interface involves assembly of laminins through binding to cell membrane receptors. Journal of Cell Science 111(14):1929-1940.
X'&&D!Y<)!Z-%T2QP=582Q58#6!U)!G#'5%!I()!<2$+&9!7G)!Y&9%5E!71)!G2'+5B!S:)!Y2%$&'!I>! ?@@J>! G#5D&$'2D2;+&! 45+E3&'! 962FF5+D9! F5'! .#99-&! &%$#%&&'#%$>! =2.-'&! G#5.&6C%5+5$E!?KLOMN[A@P[@V!
X'&E32%!*H)!\2%%29!]Z)!U#;95%!Y(>!KRR?>!1&++-+2'!32.&'#2+9!29!45'5-9!962FF5+D9!F5'! .#99-&!&%$#%&&'#%$>!0'5$'&99!#%!H2.&'#2+9!S6#&%6&!J[LVPJMNKOVPKAK>
Fuchs JR, Nasseri BA, Vacanti JP. 2001. Tissue engineering: a 21st century solution to surgical reconstruction. The Annals of Thoracic Surgery 72(2):577-591.
Gallico GG, Oconnor NE, Compton CC, Kehinde O, Green H. 1984. Permanent coverage of large burn wounds with autologous cultured human epithelium. New England Journal of Medicine 311(7):448-451.
Ghosh K, Clark RAF. 2007. Wound repair. In: Lanza R, Langer R, Vacanti JP, editors. Principles of Tissue Engineering. 3 ed. Amsterdam: Elsevier. p 1149-1166.
Gibot L, Galbraith T, Huot J, Auger FA. 2010. A preexisting microvascular network benefits in vivo revascularization of a microvascularized tissue-engineered skin substitute. Tissue Engineering Part A 16(10):3199-3206.
Goldberg AM, Frazier JM. 1989. Alternatives to animals in toxicity testing. Scientific American 261(2):24-30.
Hafemann B, Ensslen S, Erdmann C, Niedballa R, Zuhlke A, Ghofrani K, Kirkpatrick CJ. 1999. Use of a collagen/elastin-membrane for the tissue engineering of dermis. Burns 25(5):373-384.
,2++! 0")! ^2..! XH>! ?@A@>! S.&3! 6&++9N! *C&! $&%&'2.#5%! 2%D! 32#%.&%2%6&! 5F! 6&++-+2'! D#8&'9#.E>!7&8&+543&%.!?R[LJMN[?@P[VV>
Hansbrough JF, Cooper ML, Cohen R, Spielvogel R, Greenleaf G, Bartel RL, Naughton G. 1992. Evaluation of a biodegradable matrix containing cultured human
41 fibroblasts as a dermal replacement beneath meshed skin grafts on athymic mice. Surgery 111(4):438-446.
Hefton JM, Caldwell D, Biozes DG, Balin AK, Carter DM. 1986. Grafting of skin ulcers with cultured autologous epidermal cells. Journal of the American Academy of Dermatology 14(3):399-405.
Heldin CH, Westermark B. 1999. Mechanism of action and in vivo role of platelet- derived growth factor. Physiological Reviews 79(4):1283-1316.
Henry G, Garner WL. 2003. Inflammatory mediators in wound healing. Surgical Clinics of North America 83:483-+.
Hornstra IK, Birge S, Starcher B, Bailey AJ, Mecham RP, Shapiro SD. 2003. Lysl oxidase is required for vascular and diaphragmatic development in mice. Journal of Biological Chemistry 278(16):14387-14393.
Hubbell JA. 1995. Biomaterials in tissue engineering. Nature Biotechnology 13(6):565- 576.
Medical Data International Inc. 1993. Wound Care in the US: Emerging Trends, Management and New Product Development. Irvine.
Iwano M, Plieth D, Danoff TM, Xue C, Okada H, Neilson EG. 2002. Evidence that fibroblasts derive from epithelium during tissue fibrosis. Journal of Clinical Investigation 110(3):341-350.
Jarrett A. 1974. The physiology and pathophysiology of the skin. In: Jarrett A, editor. The Dermis and Dendrocytes. London: Academic Press.
Johnstone CC, Farley A. 2005. The physiological basics of wound healing. Nurs Stand 19(43):59-65; quiz 66.
:2%$!,^)!*2;2.2!\)!]Q2D2!\>!?@@@>!X2;'#62.#5%!5F!45'5-9!$&+2.#%!962FF5+D9!F5'!.#99-&! &%$#%&&'#%$>!G#532.&'#2+9!KRL?JMN?VV@P?VJJ>
Kangesu T, Navsaria HA, Manek S, Fryer PR, Leigh IM, Green CJ. 1993. Kerato-dermal grafts: the importance of dermis for the in vivo growth of cultured keratinocytes. British Journal of Plastic Surgery 46(5):401-409.
Kohn J. 2004. New approaches to biomaterials design. Nature Materials 3(11):745-747.
Konig A, Bruckner-Tuderman L. 1991. Epithelial-mesenchymal interactions enhance expression of collagen VII in vitro. Journal of Investigative Dermatology 96(6):803-808.
42 Lander AD. 1993. Proteoglycans. In: Kreis T, Vale R, editors. Guidebook to the Extracellular Matrix and Adhesion Proteins. Oxford: Oxford University Press. p 12-16.
Langer R, Vacanti JP. 1993. Tissue engineering. Science 260(5110):920-926.
Lanza RP, Langer R, Chick WL. 1997. Principles of tissue engineering. Austin, TX: Academic Press.
Y28#Q!<)!Y2%$&'!I>!KRRJ>!*#99-&!&%$#%&&'#%$N!6-''&%.!9.2.&!2%D!4&'94&6.#8&9>!"44+#&D! H#6'5;#5+5$E!2%D!G#5.&6C%5+5$E![WL?MN?PA>!
Y28Q&'! I)! S-%! *>! ?@AK>! ,&.&'5$&%&#.E! #%! &4#D&'32+! ;292+! Q&'2.#%56E.&9N! 35'4C5+5$#62+!2%D!F-%6.#5%2+!65''&+2.#5%9>!S6#&%6&!K?WLJWVOMN?KV@P?KJ?>
Layman DL, McGoodwin EB, Martin GR. 1971. The nature of the collagen synthesized by cultured human fibroblasts. Proceedings of the National Academy of Sciences of the United States of America 68(2):454-458.
Lee KY, Mooney DJ. 2001. Hydrogels for tissue engineering. Chemical Reviews 101(7):1869-1880.
Leigh IM, Purkis PE, Navsaria HA, Phillips TJ. 1987. Treatment of chronic venous ulcers with sheets of cultured allogenic keratinocytes. British Journal of Dermatology 117(5):591-597.
Y#!^P()!Y2-'&%6#%!1*)!12.&'95%!<()!*-2%!IS)!:5!X:>!KRRK>!<+&6.'594-%!%2%5F#;'5-9! 9.'-6.-'&N! "! %58&+! 962FF5+D! F5'! .#99-&! &%$#%&&'#%$>! (5-'%2+! 5F! G#53&D#62+! H2.&'#2+9!I&9&2'6C![RLJMN[?VP[K?>!
Y-.5+F! H0)! ,-;;&++! (">! KRRW>! SE%.C&.#6! ;#532.&'#2+9! 29! #%9.'-6.#8&! &_.'26&++-+2'! 3#6'5&%8#'5%3&%.9! F5'! 35'4C5$&%&9#9! #%! .#99-&! &%$#%&&'#%$>! =2.-'&! G#5.&6C%5+5$E!KVL?MNJOPWW>!
H2D#C2++E!SZ)!H2..C&B!,^*>!?@@@>!05'5-9!6C#.592%!962FF5+D9!F5'!.#99-&!&%$#%&&'#%$>! G#532.&'#2+9!KRL?KMN??VVP??JK>
Marks MG, Doillon C, Silver FH. 1991. Effects of fibroblasts and basic fibroblast growth factor on facilitation of dermal wound healing by type I collagen matrices. Journal of Biomedical Materials Research 25(5):683-696.
Moiemen NS, Vlachou E, Staiano JJ, Thawy Y, Frame JD. 2006. Reconstructive surgery with Integra dermal regeneration template: Histologic study, clinical evaluation, and current practice. Plastic and Reconstructive Surgery 117(7S):160S-174S 10.1097/01.prs.0000222609.40461.68.
Montagna W, Parakkal PF. 1974. The Structure and Function of Skin. New York: Academic Press.
43 Munster AM. 1996. Cultured skin for massive burns - A prospective, controlled trial. Annals of Surgery 224(3):372-375.
Murphy GE. 2004. Histology of the skin. In: Elder DE, Elenitsas R, Johnson BL, Murphy GE, editors. Lever's Histopathology of the Skin. 9 ed. Philadelphia: Lippincott Williams & Wilkins. p 9-58.
Muschler GF, Nakamoto C, Griffith LG. 2004. Engineering principles of clinical cell- based tissue engineering. Journal of Bone and Joint Surgery 86(7):1541-1558.
Nelson CM, Bissell MJ. 2005. Modeling dynamic reciprocity: Engineering three- dimensional culture models of breast architecture, function, and neoplastic transformation. Seminars in Cancer Biology 15(5):342-352.
Nerem RM, Sambanis A. 1995. Tissue engineering: from biology to biological substitutes. Tissue Engineering 1(1):3-13.
O'Connor NE, Mulliken JB, Banksschlegel S, Kehinde O, Green H. 1981. Grafting of burns with cultured epithelium prepared from autologous epidermal cells. Lancet 1(8211):75-78.
Odland GF. 1991. Structure of the skin. In: Goldsmith LA, editor. Physiology, Biochemistry, and Molecular Biology of the Skin. New York: Oxford University Press. p 3-62.
Okamoto E, Kitano Y. 1993. Expression of basement membrane components in skin equivalents: Influence of dermal fibroblasts. Journal of Dermatological Science 5(2):81-88.
Oldberg A, Ruoslahti E. 1982. Interactions between chondroitin sulfate proteoglycan, fibronectin, and collagen. Journal of Biological Chemistry 257(9):4859-4863.
Pandya AN, Woodward B, Parkhouse N. 1998. The use of cultured autologous keratinocytes with integra in the resurfacing of acute burns. Plastic and Reconstructive Surgery 102(3):825-828.
Pham C, Greenwood J, Cleland H, Woodruff P, Maddern G. 2007. Bioengineered skin substitutes for the management of burns: A systematic review. Burns 33(8):946- 957.
Phillips TJ. 1993. Biologic skin substitutes. Journal of Dermatologic Surgery and Oncology 19(8):794-800.
Pomaha! B, Svensjö T, Yao F, Brown H, Eriksson E. 1998. Tissue engineering of skin. Critical Reviews in Oral Biology & Medicine 9(3):333-344.
05..&%! 1S>! ?@OJ>! *C&! &4#D&'32+! 4'5+#F&'2.#8&! -%#.N! .C&! 4599#;+&! '5+&! 5F! .C&! 6&%.'2+! ;292+!6&++>!1&++!0'5+#F&'2.#5%!OL?MNOOPAA>!
44 05..&%!1S>!?@A?>!1&++!'&4+26&3&%.!#%!&4#D&'3#9!LQ&'2.545#&9#9M!8#2!D#96'&.&!-%#.9!5F! 4'5+#F&'2.#5%>!]%.&'%2.#5%2+!'&8#&B!5F!6E.5+5$E![@NKO?PV?A>
Rheinwald JG, Green H. 1975. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6(3):331-344.
Riches DWH. 1996. Macrophage involvement in wound repair, remodeling and fibrosis. In: Clark RAF, editor. The Molecular and Cellular Biology of Wound Repair. 2 ed. New York: Plenum Press. p 95-141.
Roeder BA, Kokini K, Sturgis JE, Robinson JP, Voytik-Harbin SL. 2002. Tensile mechanical properties of three-dimensional type I collagen extracellular matrices with varied microstructure. Journal of Biomechanical Engineering 124(2):214- 222.
Ronfard V, Rives JM, Neveux Y, Carsin H, Barrandon Y. 2000. Long-term regeneration of human epidermis on third degree burns transplanted with autologous cultured epithelium grown on a fibrin matrix. Transplantation 70(11):1588-1598.
Rozga J, Morsiani E, Lepage E, Moscioni AD, Giorgio T, Demetriou AA. 1994. Isolated hepatocytes in a bioartificial liver: a single group view and experience. Biotechnology and Bioengineering 43(7):645-653.
Ruoslahti E, Pierschbacher M, Engvall E, Oldberg A, Hayman EG. 1982. Molecular and biological interactions of fibronectin. Journal of Investigative Dermatology 79:S65-S68.
Schnabel M, Marlovits S, Eckhoff G, Fichtel I, Gotzen L, Vecsei V, Schlegel J. 2002. Dedifferentiation-associated changes in morphology and gene expression in primary human articular chondrocytes in cell culture. Osteoarthritis and Cartilage 10(1):62-70.
Segal N, Andriani F, Pfeiffer L, Kamath P, Lin N, Satyamurthy K, Egles C, Garlick JA. 2008. The basement membrane microenvironment directs the normalization and survival of bioengineered human skin equivalents. Matrix Biology 27(3):163-170.
Shatford RA, Nyberg SL, Meier SJ, White JG, Payne WD, Hu WS, Cerra FB. 1992. Hepatocyte function in a hollow fiber bioreactor: a potential bioartificial liver. Journal of Surgical Research 53(6):549-557.
Skeggs LT, Leonards JR. 1948. Studies on an artificial kidney: I. preliminary results with a new type of continuous dialyzer. Science 108(2800):212-213.
Slupsky JR, Kalbas M, Willuweit A, Henn V, Kroczek RA, Muller-Berghaus G. 1998. Activated platelets induce tissue factor expression on human umbilical vein endothelial cells by ligation of CD40. Thrombosis and Haemostasis 80(6):1008- 1014.
45 Stiefel D, Schiestl C, Meuli M. 2010. Integra Artificial Skin® for burn scar revision in adolescents and children. Burns 36(1):114-120.
Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. 1998. Embryonic stem cell lines derived from human blastocysts. Science 282(5391):1145-1147.
Tran KT, Griffith L, Wells A. 2004. Extracellular matrix signaling through growth factor receptors during wound healing. Wound Repair and Regeneration 12(3):262-268.
Tremblay P-L, Hudon V, Berthod F, Germain L, Auger FA. 2005. Inosculation of tissue- engineered capillaries with the host's vasculature in a reconstructed skin transplanted on mice. American Journal of Transplantation 5(5):1002-1010.
Velazquez OC, Snyder R, Liu Z-J, Fairman RM, Herlyn M. 2002. Fibroblast-dependent differentiation of human microvascular endothelial cells into capillary-like, three- dimensional networks. The FASEB Journal.
Wainwright D, Madden M, Luterman A, Hunt J, Monafo W, Heimbach D, Kagan R, Sittig K, Dimick A, Herndon D. 1996. Clinical evaluation of an acellular allograft dermal matrix in full-thickness burns. Journal of Burn Care & Research 17(2):124-136.
Wainwright DJ. 1995. Use of an acellular allograft dermal matrix (AlloDerm) in the management of full-thickness burns. Burns 21(4):243-248.
^2..! XH>! ?@A@>! *&'3#%2+! D#FF&'&%.#2.#5%! 5F! &4#D&'32+! Q&'2.#%56E.&9>! 1-''&%.! `4#%#5%!#%!1&++!G#5+5$E!?L[MN??ROP?W>!
^2..! XH)! U'&&%! ,>! ?@AK>! S.'2.#F#62.#5%! 2%D! .&'3#%2+! D#FF&'&%.#2.#5%! 5F! 6-+.-'&D! &4#D&'32+!6&++9>!=2.-'&!K@WLWAJAMNJVJPJV[>
Wikner NE, Dixit VM, Frazier WA, Clark RAF. 1987. Human keratinocytes synthesize and secrete the extracellular matrix protein, thrombospondin. Journal of Investigative Dermatology 88(2):207-211.
Wilkins LM, Watson SR, Prosky SJ, Meunier SF, Parenteau NL. 1994. Development of a bilayered living skin construct for clinical applications. Biotechnology and Bioengineering 43(8):747-756.
Winsor T, Burch GE. 1944. Differential roles of layers of human epigastric skin on diffusion rate of water. Archives of Internal Medicine 74(6):428-436.
Wong T, McGrath JA, Navsaria H. 2007. The role of fibroblasts in tissue engineering and regeneration. British Journal of Dermatology 156(6):1149-1155.
Wynn JJ, Alexander CE. 2011. Increasing organ donation and transplantation: the U.S. experience over the past decade. Transplant International 24(4):324-332.
46 Yang SF, Leong KF, Du ZH, Chua CK. 2001. The design of scaffolds for use in tissue engineering. Part 1. Traditional factors. Tissue Engineering 7(6):679-689.
Yaoita H, Foidart J-M, Katz SI. 1978. Localization of the collagenous component in skin basement membrane. Journal of Investigative Dermatology 70(4):191-193.
47 Chapter 3
DEVELOPMENT OF CONCENTRATED COLLAGEN-
CHONDROITIN SULFATE SCAFFOLDS WITH TUNABLE
PROPERTIES FOR TISSUE ENGINEERING APPLICATIONS
This work is published in Journal of Biomedical Materials Research Part A, 94A(4):
1050-1060, 2010
3.1 Introduction
The basic premise of tissue engineering is to combine the appropriate cells with a material under conditions that lead to tissue formation. Different cells interact with complex networks of extracellular matrix (ECM), which are macromolecules comprised of proteins and polysaccharides secreted by the cells themselves, to form different tissues.
The ECM, in addition to contributing to mechanical integrity, has important signaling and regulator functions such as cell adhesion, migration, proliferation, differentiation, and metastasis, in the development, maintenance, and regeneration of tissues (Daley et al.
2008; Furth et al. 2007). Therefore, to repair dysfunctional or damage tissue, it is necessary to fabricate an artificial environment, which promotes and facilitates the inherent abilities of cell adhesion, migration, proliferation, and differentiation (Lavik and
Langer 2004). Such an artificial environment, which is commonly referred to as a scaffold in tissue engineering and made up of a biocompatible and biodegradable material, serves as a support and the ECM during different processes of tissue formation.
48 Scaffold biomaterials for tissue engineering applications can be classified into different categories: synthetic and natural materials. Synthetic biomaterials such as poly(L-lactic acid) (PLLA) and poly (L-glycolic acid) (PLGA) have received considerable attention for tissue engineering applications and have shown promise in preclinical animal studies and some early human clinical trials. The synthetic biomaterials have provided predictable and producible mechanical and physical properties and can be fabricated with great precision. Another advantage of synthetic biomaterials is that most of them are biodegraded via chemical hydrolysis and insensitive to enzymatic process so that their degradation does not vary from patient to patient. However, synthetic biomaterials tend to elicit a foreign material type of response in the host, specifically, a fibrous connective tissue deposition leading to formation of dense scars and fibrosis.
Natural biomaterials such as alginate, Matrigel, collagen, and fibrin have been investigated as alternatives to synthetic biomaterials (Ruszczak 2003). Collagen has been used successfully as a scaffold for wound healing and multiple scaffolds based on it are currently available for clinical use (Glowacki and Mizuno 2008; Ruszczak 2003).
Matrigel, which is solubilized basement membrane preparations extracted from mouse tumors that contains several components of basement membranes enriched with laminin, is commercially available from BD Biosciences (San Jose, CA, USA) and has been used for cardiac tissue engineering (Morritt et al. 2007; Sieminski et al. 2002). Alginate, a polysaccharide derived from brown seaweed, has been used as a scaffold for the encapsulation and immunoprotection of transplanted cells (Wong 2004). Successful and stable cultures of different cells in alginate gels have been proposed for the bone and cartilage tissue engineering (Marijnissen et al. 2002). Although natural biomaterials have
49 shown promises, there are certain limitations and concerns regarding their use. First, it is difficult to control their mechanical properties and degradation rates over a wide range
(Lee et al. 2001). Second, there is a possibility that the naturally derived biomaterials may provoke a serious immune response (Schmidt and Baier 2000). Proper characterization and screening of the natural biomaterials may eliminate concerns, but overcoming limitations of the biomaterials is still challenging.
Among all natural biomaterials, collagen-based biomaterials have tremendous potential as scaffold materials for tissue engineering (Glowacki and Mizuno 2008). Type I collagen, the most abundant protein in the human body, is the primary constituent of the
ECM in most tissues. Collagen-based biomaterials have been used in clinical trials in the treatment of defects, especially in wound healing (Eaglstein and Falanga 1997).
Experimental and clinical studies have shown that the wound healing process was improved using collagen-based scaffolds. Moreover, xenogenic collagen scaffolds did not cause any foreign-body reaction nor any immune rejection when applying them to the wound (Soo et al. 1993).
Collagen-based biomaterials used for tissue engineering applications can generally be classified into two different categories: hydrogels and porous scaffolds. Collagen hydrogels are made as collagen molecules in aqueous solution, and under appropriate conditions, spontaneously assemble/gel to form a semisolid structure. Such gels are widely used to study cell-cell and cell-matrix interactions in a variety of biological processes as cells can be incorporated in the solution prior to gelling (Rosenblatt et al.
1994; Wallace and Rosenblatt 2003). In addition, with support, these gels are also used in a variety of tissue engineering applications (Glowacki and Mizuno 2008; Huynh et al.
50 1999; Wallace and Rosenblatt 2003). A downside to the use of gels for tissue engineering is their lack of mechanical stability; collagen gels are difficult to handle in a clinical setting. Although techniques such as cross-linking, other chemical modifications, and cell-mediated contraction can be used to make the gels stronger (Chen et al. 2007; Duan and Sheardown 2006; Han et al. 2003; Orban et al. 2004; Saito et al. 2007; van Wachem et al. 2001), gels are still mechanically inferior to dry scaffolds.
The other category of biomaterials is dry porous scaffolds. Among collagen-based copolymers, porous collagen-glycosaminoglycan (CG) scaffolds hold great potential for use in tissue engineered products (Yannas et al. 1980). CG scaffolds are produced by lyophilizing a CG-acetic acid solution, which results in a continuous dry porous CG scaffold suitable for cell seeding. Such lyophilized scaffolds are mechanically superior to collagen gels. They also have several advantageous properties compared to scaffolds made from other materials: CG scaffolds are made from a natural material common to the extracellular matrix of most tissues; they possess haemostatic properties, low antigenicity, and appropriate mechanical characteristics for use in tissue engineering applications.
They have also shown an excellent ability to promote cell attachment and proliferation in a wide variety of tissue engineering applications (Buijtenhuijs et al. 2004; Han et al.
2003; Kuberka et al. 2002).
The physical and chemical properties of the bulk scaffold material determine the theoretical upper limit of the scaffold mechanical properties such as tensile and compressive properties. The actual mechanical properties of a useful scaffold product will be lower, and depend largely on the scaffold structure, itself the result of a compromise design to meet biological and mass-transport needs. For collagen-based
51 scaffolds, the modulus of pure collagen has been estimated at between ~3 and 9 GPa
(Sasaki and Odajima 1996), which leaves ample room for compromise when designing for most tissues.
Despite the potential of porous CG scaffolds, their physical, chemical, and biological properties must be optimized for the targeted tissue. The pore size and architecture of CG scaffolds greatly affects cells adhesion (O'Brien et al. 2004; O'Brien et al. 2005) and mass-transport characteristics. The surface area/volume ratio depends on the density and average pore diameter. However, the cell diameter dictates the minimum pore size, which varies from cell to cell. Depending on the applications, pore size must be carefully controlled. In addition, the pore size and volume fraction also affect the mass-transport of nutrients and waste to and from the cell in the scaffold, as well as the transport of signaling molecules between cells. A more densely packed scaffold increases the resistance to mass transfer of nutrients and waste, but a looser scaffold increases the number of cells that can attach. This suggests the existence of an optimal scaffold composition that balances the number of cells that can be attached to the scaffold with the transport of nutrients and waste to and from these cells required for cell viability and growth. Constructs with tunable properties of volume fraction, pore size distribution and mechanical strength, therefore, are needed to satisfy the cell adhesion and mass transport demands placed on them by different tissue engineered systems. Previously, the average pore size in CG scaffolds could be varied by changing the lyophilization process parameters (O'Brien et al. 2005). The final freeze-drying temperature has been shown to control the pore size of the eventual scaffolds to some extent; a lower freeze-drying
52 temperature led to a smaller mean pore size, and a range of pore sizes, from 150 to 95 micrometers, were obtained (O'Brien et al. 2005).
Despite these advances, the base solution concentration used to obtain CG scaffolds was limited by the collagen solubility; in most studies, the collagen concentration used was about 0.25 % w/v. This limits the final collagen density that can be obtained in the scaffolds, which subsequently impacts the physicochemical properties of the scaffolds.
We developed a new, simpler method that allows for greater control over collagen concentration in the scaffolds using solution centrifugation. We found that centrifugation and subsequent removal of supernatant of the CG solution before lyophilization resulted in scaffolds that have a range of collagen concentration, mechanical properties and water uptake characteristics. We further showed evidence for the potential use of these scaffolds as the dermal analoh in TE skin substitutes.
3.2 Materials and Methods
3.2.1 Preparation of CG Solution
The base CG solutions were made by using a method adopted from Yannas, et al.
(Yannas et al. 1980). 2.2 g of Type I collagen from bovine Achilles tendon (Sigma, St.
Louis, MO) was added to 800 ml of 0.5% v/v acetic acid solution (Sigma). The solution was then blended in an ice-bath cooled vessel using an overhead homogenizer (IKA
Works, Wilmington, NC) at a speed of 13,500 rpm. After 20 minutes of blending, 0.22 g
(10% of weight of collagen) of chondroitin 6-sulfate from shark cartilage (Sigma), dissolved in 40 ml of 0.5% v/v acetic acid solution, was added drop-wise to the collagen solution. The solution was homogenized for an additional 20 minutes at 13,500 rpm. The
53 final concentrations of collagen and chondroitin sulfate in the CG solutions were about
2.6 and 0.26 g/liter, respectively.
3.2.2 Preparation of Concentrated CG Solution and Scaffolds
A batch of 840 ml of CG solution was prepared using normal techniques. 40 ml CG solution was aliquoted to Oak Ridge centrifugal tube (Nalge, Rochester, NY) for centrifugation. Each aliquot was centrifuged for 30 min at 38720 " g (r.c.f.) with an automatic superspeed refrigerated centrifuge (RC-5C, Sorvall Instruments, DuPont,
Wilmington, DE) [Figure 3-2]. After centrifugation, a percentage of the solution was removed from the top of each aliquot. The remaining solution and the collagen pellet were mixed thoroughly, frozen from 4°C to -40°C in 120 min and kept at -40°C overnight, and then lyophilized for 24 h using a lyophilizer (VirTis AdVantage EL, SP
Industries, Inc., Warminster, PA) [Figure 3-1]. All CG scaffolds fabricated in this study underwent a dehydrothermal process after freeze-drying to strengthen the collagen network. CG scaffolds were dehydrated for 24 h at 120°C under vacuum (25 mTorr).
Figure 3-1 Schematic of preparation of concentrated CG scaffolds.
54
Figure 3-2 Digital images of CG solution (a) before and (b) after centrifugation.
3.2.3 Characterization of Concentrated CG Solution and Scaffolds
3.2.3.1 Water Uptake
Water-uptake experiments were performed by obtaining 10 mm diameter disks of scaffolds using a biopsy punch (Acuderm, Inc., Ft. Lauderdale, FL), weighing the dried scaffolds, and then weighing the hydrated scaffolds after they were hydrated thoroughly for 4 days (n = 6).
3.2.3.2 Chondroitin Sulfate Content
Scaffolds obtained from various concentrated solutions were used to quantify retained chondroitin sulfate (CS). After the water-uptake experiment, the same 12 mm diameter disks of hydrated scaffolds were placed in 2.0 ml Eppendorf tubes and dried using a
Speed-vac (Savant, Thermo-Fisher, Waltham, MA). The CS amount in the scaffolds was quantified by a GAG assay performed as described previously, with some modifications
(Carrino et al. 1991; Penick et al. 2005; Welter et al. 2007). Scaffolds were digested with
55 1.0 ml of papain buffer (25 #g/ml papain; 2 mM cysteine; 50 mM NaH2PO4; 2 mM
EDTA; pH 6.5) (Sigma) (Ponticiello et al. 2000) at 65°C. After the digestion was complete, the scaffold extracts were transferred to a 5 ml tube and an additional 1.0 ml of papain buffer was added for dilution. A 0.45 #m pore-size nitrocellulose membrane was prewetted with dH2O and placed into a dot-blot apparatus. A 250 #l aliquot of 0.02%
Safranin O in 50 mM sodium acetate (pH 4.8) was pipetted into each well, followed by
25 #l aliquots of the papain-digested extracts or standards (shark-cartilage CS C)
(Carrino et al. 1991). After vacuuming, the wells were rinsed with dH2O, and the filter was removed from the apparatus. Individual dots were cut out, transferred to microcentrifuge tubes and eluted in 10% cetylpyridinium chloride at 37°C. The absorbance of the eluates was read at 530 nm in a microplate reader.
3.2.3.3 Imaging and Image Analysis
The microstructures of the lyophilized vacuum dried scaffolds in the cross-section were observed under scanning electron microscopy (SEM) (XL30, Philip, Eindhoven,
Netherlands). The CG scaffolds were dehydrated using the method described by Breat et al. Briefly, the samples were sequentially dehydrated in solutions of increasing concentrations of ethanol (30%, 50%, 70%, 80%, 90%, 95% and 100%) for at least 1 hour in each solution. The 100% ethanol was changed twice to ensure all water in the sample was exchanged with ethanol. Samples were then dried with a critical point drying method. This method is used to prevent the gas-liquid interface from causing any artifacts on the scaffold surface while drying. Prior to critical point drying, the samples were subjected to at least 2 washes to exchange all the ethanol with liquid CO2 in a critical
56 point dryer. Dried samples were Pd-sputter-coated using a Desk sputter coater (Denton
Vacuum, Moorestown, NJ), and then observed under SEM.
Some samples were fixed in 1% glutaraldehyde (Sigma) at room temperature overnight, embedded in ethylene glycol-methacrylate (Electron Microscopy Sciences, Hatfield, PA), and sectioned. The mean pore size of the scaffolds was determined by analyzing SEM images. Three SEM images of each scaffold (n = 6) were selected and at least 15 apparent pores were measured from from image with image analysis software (Image Pro
Plus 6.2, Media Cybernetics, Inc., Bethesda, MD). The mean pore size of each pore was calculated averaging the major and minor axes lengths (Carrino et al. 1991; Martin et al.
2004; Shi et al. 2008; Yeong et al. 2007).
3.2.3.4 Mechanical Strength Testing
Tensile tests: The tensile properties of dry CG scaffolds were assessed via tensile testing
(n = 7 for each condition). CG scaffolds were cut into dogbone-shaped specimens with a gauge length of 22 mm and width of 5 mm [Figure 3-3]. They were then tested to failure using an university testing machine (Instron model 5565, Canton, MA) at a strain rate of
1.4 mm/min. The elasticity modulus (ab) was determined by fitting the experimental data of stress (#) and strain ($) with the following model (Veronda and Westmann 1970) consisting of two parameters a and b:
a eb" 1 ! Tension = ( # )
57
Figure 3-3 The dogbone-shaped specimen for tensile tests.
Compression tests: Disc-shaped CG scaffolds were obtained by using a 10-mm biopsy punch (Acuderm, Inc., Ft. Lauderdale, FL). The thickness and mass of the samples were measured. The compressive modulus of each disc was determined in a Rheometrics
Solids Analyzer II compression device (Piscataway, NJ). Each disc was compressed for 3 s at a strain rate of 5% s-1. Compression tests were performed on six samples per condition, so that an average compressive modulus could be ascertained. A stress-strain curve was then recorded by a computer data acquisition system, and transferred to Excel®
(Microsoft, Redmond, WA) to determine an average compressive modulus of elasticity (%) from equation (Harley et al. 2007):
! Compression = "# where # is stress and $ is strain obtained from compression tests.
58 3.2.4 CG scaffolds for TE Skin Substitutes
3.2.4.1 Medium Preparation:
Derived media for culturing composite skin substitutes were keratinocyte seeding medium (KSM), keratinocyte priming medium (KPM), and air-liquid interface medium
(ALIM). The KSM composition is: (3:1) Dulbecco’s Modified Eagle’s Medium (DMEM) high glucose/Ham’s F12 medium, 1% fetal bovine serum (FBS), penicillin (100 IU/mL)- streptomycin (100 #g/mL) (Invitrogen, Carlsbad, CA), 50 #g/mL ascorbic acid, 0.2
#g/mL hydrocortisone, 5 #g /mL insulin (Sigma), 10-10 M cholera toxin (CalBiochem,
EMD Chemicals, Gibbstown, NJ). The KPM composition is 24 #M bovine serum albumin (BSA), 1 mM L-serine, 10 #M L-carnitine, and fatty acid cocktail: 25 µM oleic acid, 15 µM linoleic acid, 7 µM arachidonic acid, 25 µM palmitic acid (Sigma) in KSM.
The ALIM composition is 1 ng/mL epidermal growth factor (Sigma) in serum-free KPM.
3.2.4.2 Keratinocyte Isolation:
Keratinocytes were isolated from discarded neonatal human foreskins by the method of dispase digestion followed by trypsin treatment as described in Papini, et al. (Papini et al.
2003). The protocol was approved by the University Hospitals of Cleveland Institutional
Review Board. Neonatal human foreskin samples were collected from Rainbows Hospital
(Cleveland, OH) and maintained at 4 °C for 6 h in defined keratinocyte serum-free medium (defined-KSFM, Invitrogen), containing 10 #g /mL gentamicin and 5X antibiotic-antimycotic (Invitrogen). The samples were cut into approximate 5 x 5 mm pieces, and connective tissue was removed carefully without destroying the dermis. The samples were then stored in a 50 mL BD Falcon centrifugal tube (BD Biosciences,
59 Sparks, MD) with 0.5% dispase (Invitrogen) at 4°C for 12-18 h. Each enzyme-treated piece was washed with phosphate-buffered saline (PBS) (Invitrogen) at pH 7.4 and the epidermis was separated from the dermis. The epidermis was then treated with 5 mL of
o 0.05% trypsin in an incubator at 37 C and at 5% CO2 in humidified air environment for 5 min, and the enzyme activity was then terminated with 5 mL of bovine serum
(Invitrogen). The cells and epidermis were transferred to a 50 mL centrifugal tube and centrifuged for 8 min at 100g. The supernatant was carefully removed, and the cells and epidermis were gently resuspended in keratinocyte serum free medium (KSFM)
(Invitrogen). The cells (keratinocytes) and KSFM were then transferred to a T-75 flask
(Falcon). The medium was changed every three days until cells reaching 90% confluence.
3.2.4.3 Skin Substitutes Preparation
The cells were cultured in KSFM with medium changed every three days, and the cells were passaged when they reached 90% confluence. To prepare TE skin substitutes, we used passage 2 cells. Skin substitutes were prepared using CG scaffolds as described by
Medalie et al. (Medalie et al. 1996). CG scaffolds were prepared by the above method, and then cut into 1 x 1 cm pieces. CG scaffolds were sterilized with 1X antibiotic- antimycotic in PBS overnight at 4°C, and then washed with PBS three times before culturing. Before seeding keratinocytes, CG scaffolds were treated with 100 µg/mL human fibronectin (Invitrogen) for 2 h in an incubator. Keratinocytes were seeded at a density of 2.5 " 105 cells/cm2 on CG scaffolds using KSM in 6-well culture plates
(Falcon). The scaffolds with cells were cultured in a submerged mode with KSM for 1 day, then with KPM for 2 days. The epidermal portion of the constructs was then exposed to air and then cultured in ALIM for 7 days, with medium replaced every 2 days.
60 3.2.4.4 Histoloy
For histology, cultured skin substitutes were fixed with 4% paraformaldehyde overnight in a refrigerator. Samples for histological process were dehydrated in solutions of increasing concentrations of ethanol, 30%, 50%, 70%, 80%, 90%, 95% and 100%, for at least 1 hour each in each solution, and then exchanged ethanol with xylene twice for
Paraplast (Fisher Scientific, Pittsburgh, PA) embedding. Paraplast was melted at 56°C, and samples were filled up with molten Paraplast. Fresh molten Paraplast was replaced three times to ensure all xylene in the sample was exchanged with Paraplast. Samples were then embedded into an embedding mold with filled molten Paraplast, by placing them in the appropriate position and then solidified on the cold plate. Samples were removed from mold and sectioned with a Microtome (Leica, Germany). Adjacent 5 µm- thick sections were stained with Hematoxylin and Eosin (H&E) for histology. H&E stained sections were mounted in permount.
3.2.5 Statistical Analysis
For all quantitative results, a one-way analysis of variance (ANOVA) and Tukey-Kramer multiple comparison procedures were performed using Origin (OriginLab, Northampton,
MA) to compare data groups. A p value less than 0.05 was used to determine statistical significance.
3.3 Results
61 3.3.1 General Scaffold Characterization
Figure 3-4 shows the effect of centrifugation on the collagen concentration. By removing different amounts of supernatants, collagen solutions of different concentrations were obtained. The average collagen-CS concentration of scaffolds (dry mass of scaffold/volume of scaffold, %w/v) was plotted as a function of the percentage of the total solution removed as supernatant. The results show that we can obtain scaffolds that have dry collagen-CS content ranging from 2 (65% removal) to 5 (85% removal) times that of scaffolds from unconcentrated solution. Unconcentrated scaffolds had a nominal collagen density of 0.26 % w/v. Clearly, increasing the amount of supernatant removed increased the concentration of the collagen-CS in the scaffolds (p<0.05). Removing supernatant amounts >85% of total solution was difficult as this led to the removal of the collagen precipitate. Further, the centrifugation time had very little effect on the concentrated scaffolds (data not shown); a centrifugation time of 20 minutes was sufficient to achieve the final concentration.
62
Figure 3-4 The relationship between nominal collagen density and removed percentage of supernatants. *represents statistical significance (t-test, p=0.05) relative to base solution (0% removed).
To understand the effect of concentration on the CS retention, we performed GAG assays on scaffolds obtained by these different procedures. The results [Figure 3-5] show that
CS was retained in the concentrated scaffolds at levels similar or greater than the levels in the unconcentrated scaffolds. The dashed line in Figure 2 indicates the level of CS added to the solution before homogenization.
63
Figure 3-5 The relationship between nominal collagen density and GAG content (%, GAG weight/total scaffold weight) of the scaffolds (n=6 each) obtained from concentrated CG solutions. Dashed line represents the amount of GAG added to the solution. *represents statistical significance (t-test, p=0.05) relative to base solution (0% removed).
Water uptake is an important indicator used to determine the suitability of a biomaterial for tissue engineering. Further, for a CG scaffold, water uptake is dependent not only on the amount of collagen in the scaffold but also its pore characteristics. Figure 3-6 shows water uptake of scaffolds obtained from concentrated collagen-CS solution when hydrated for 4 days. On a per volume basis [Figure 3-6 (a)] all concentrated scaffolds have significantly increased water uptake when compared to unconcentrated scaffolds.
On a per dry scaffold weight [Figure 3-6 (b)], however, although water uptake of
64 scaffolds from concentrated solutions seemed to decrease, the results were not statistically significant when compared to water uptake of scaffolds from unconcentrated solution.
65
Figure 3-6 Water uptake (a) per volume and (b) per dry scaffold weight of scaffolds obtained from concentrated CG solution. *represents statistical significance (t-test, p=0.05) relative to base solution.
66 3.3.2 Pore Structure
Figure 3-7 shows SEM images of scaffolds that were obtained from unconcentrated
[0.3% w/v, Figure 3-7 (a)] and concentrated [1.2% w/v, Figure 3-7 (b)] collagen-CS solutions. The unconcentrated scaffold pores appear circular and slightly larger. Further, the walls of the pores were made up of porous collagen fibers [insert in Figure 3-7 (a)].
On the other hand, the pores of concentrated scaffolds while circular were slightly smaller. More importantly, the walls of the pores were made of continuous sheet of collagen fibers [insert in Figure 3-7 (b)]. The pores appear larger in unconcentrated samples whereas they appear smaller and more in number in the concentrated samples.
Figure 3-7 (c) shows that the pore mean diameter decreased by as much as 20% as the collagen concentration increased (p<0.05). However, overall porosity ($) for both concentrated and unconcentrated scaffolds, estimated from the equation:
m ! = 1" # where m is the nominal density of the collagen scaffold and & is the density of anhydrous collagen (1.3 g/mL), were very high and ranged from 0.98 (concentrated) to 0.995
(unconcentrated).
67
Figure 3-7 SEM images of (a) unconcentrated and (b) concentrated scaffolds. (c) The relationship between the pore mean diameter and nominal collagen density. *represents statistical significance (t-test, p=0.05) relative to base solution (0% removed).
3.3.3 Mechanical Properties
To determine whether concentrating collagen-CS solution leads to scaffolds with improved mechanical properties, we performed tensile and compressive loading tests on
68 dry scaffolds that were not chemically fixed. Tensile loading was applied to samples until failure. Figure 3-8 (a) shows modulus of elasticity values for various scaffolds obtained from concentrated collagen-CS solutions. All concentrated scaffolds were significantly stronger and had modulus of elasticity as much as 33 times greater when compared with unconcentrated scaffolds (nominal collagen density: 0.3% w/v). The improvements were disproportional to the collagen density; tripling the collagen density increased the modulus of elasticity by 30-fold. The modulus of elasticity in tension decreased by about
30% when the nominal collagen increased from 1.1 to 1.2% w/v. Figure 3-8 (b) shows ultimate stress values of the scaffolds at failure. Again, the concentrated scaffolds performed better, and had ultimate stress as much as 12 times greater than those of unconcentrated scaffolds. The increases were disproportionately larger than the increases in collagen content of the scaffolds. The ultimate stress also decreased by about 20% when the nominal collagen increased from 1.1 to 1.2% w/v. Further, to understand whether these results also indicate the scaffolds ability to resist compression, we subjected the samples to compression tests. Figure 3-9 shows compressive modulus of elasticity as a function of nominal collagen density. The results are similar to those under tension; the concentrated scaffolds had elastic moduli that are as much as 56 times greater than that of unconcentrated scaffolds. The increases in the modulus of elasticity values were achieved with relatively smaller increase (3-4 fold) in collagen density.
69
Figure 3-8 (a) Modulus of elasticity of the scaffolds. (b) Ultimate stress of the scaffolds at failure from concentrate. *represents statistical significance (t-test, p=0.05) relative to base solution
70
Figure 3-9 Compression modulus of the scaffolds. *represents statistical significance (t- test, p=0.05) relative to base solution.
3.3.4 Tissue Engineering Potential
To evaluate the potential for tissue engineering applications, we used the concentrated scaffolds (nominal collagen density: 0.8% w/v) to culture skin tissue constructs in vitro.
We used 0.8% w/v scaffold, based on its properties such as water uptake, and moduli of elasticity, it was averaged and represented the concentrated scaffolds. H&E-stained cross- sections of the cultured skin substitutes show that well-stratified epidermis were formed, along with an organized granular layer [Figure 3-10].
71
Figure 3-10 (a) H&E staining image of TE skin substitutes obtained by using 0.8% w/v concentrated CG scaffolds in static cultures. A well-stratified epidermis was formed and its histological organization was comparable to that of human skin (b).
3.4 Discussion
In this work, we developed a technique to obtain porous collagen-glycosaminoglycan scaffolds of varying collagen concentrations for tissue engineering application. We showed that these scaffolds have concentration-dependent water-uptake rates, mechanical properties and pore geometry. In general, increasing the collagen density increases water uptake per volume of the scaffold, leads to smaller pore sizes, and to an increase in compressive and tensile moduli. We further showed that these scaffolds are biocompatible for skin tissue engineering.
Researchers typically use the base CG solution (collagen 0.26 % w/v) for tissue engineering applications. The base concentration is limited by the collagen solubility.
72 This further limits the final collagen density that can be obtained in the scaffolds, and then leads to a porous scaffold that is mechanically difficult to handle, and often collapses when exposed to water. By centrifuging at relatively high centrifugal force, we can obtain a collagen concentration as high as 5 times greater than the commonly-used
CG scaffolds (Yannas et al. 1980). Further, by diluting this highly concentrated collagen, we can obtain a range of concentrations corresponding to scaffolds with varying properties. The process of centrifugation is rapid (5-20 minutes) as the collagen solution sediments rapidly due to the relatively large centrifugal force applied. Cross et al. also prepared high weight percent CS solution (0.8, 1.5, 2, and 3% w/v) by directly suspending dried collagen in 0.1% acetic acid (Cross et al. 2010). To dissolve collagen, the suspensions were agitated approximately 3 times a day by manually inverting the tubes repeatedly for approximately 30 s, and collagen was fully dissolved after 5 days of the daily agitation. Although they were able to prepare higher concentration collagen solution; however, the process was time-consuming. To our knowledge, our technique of centrifugation of collagen solution to obtain scaffolds of tunable properties, while simple, is also new. Varying the freezing rate during the lyophilization process has been shown to result in collagen scaffolds of tunable properties such as the pore size (O'Brien et al.
2004). This technique, based on the physics of the lyophilization process, however, is technically more difficult to implement, and it is not clear whether the resulting benefits extend to better control over the mechanical properties of the scaffolds.
Morphometric analysis of scaffolds imaged through SEM shows that scaffolds obtained from concentrated CG solutions have more but smaller pores compared to scaffolds obtained from unconcentrated CG solution. Although increased amount of collagen in the
73 scaffolds obtained from concentrated solutions can lead to smaller overall void volume, this does not directly indicate smaller pores. The smaller pores indicate that the initial CG solution is highly homogeneous. Despite the change in pore size, the overall porosity, however, remains high >98%, possibly because of complimentary change in the number of pores. This suggests that we can use the scaffolds obtained from concentrated CG solutions for tissue engineering applications requiring high cell-scaffold interactions as well as cell-cell interactions. Further, since the scaffold porosity is similar to that of scaffolds obtained from unconcentrated CG solution, the amount of cells seeded in the former scaffolds will not be affected. The latter scaffolds, however, have inferior tensile and compressive properties, collapse during seeding, and are difficult to handle. Our results also suggest a range of pore sizes that can be obtained from our method.
Researchers traditionally use freezing temperature and other lyophilization parameters to control the pore size of such scaffolds (O'Brien et al. 2005). Our technique presents a much simpler alternative, which not only can control the pore size but also can increase the mechanical properties of scaffolds.
An important property of CG scaffolds is that CS plays a critical role in stabilizing scaffold structure by inducing efficient “stress transfer” in collagen fibrils (Garg et al.
1989). In addition, CS affects important biological properties such as cell adhesion and proliferation, and in general, cell signaling (Yannas et al. 1980). Therefore, CS retention in the concentrated scaffolds is an important parameter. Our results demonstrate that the concentrated scaffolds retain the same ratio of GAG to collagen (1:10 by mass) thus suggesting that the new process does not diminish the beneficial effect of GAG on the general scaffold stability and function.
74 Water retention by scaffolds is an important property that has implications in cell seeding and mass transport. Our results indicate that the scaffolds made from concentrated CG solution take up more water per volume than scaffolds made from unconcentrated CG solution. In general, water absorbed by porous scaffolds is characterized as: substrate- bound water and free water. The amount of bound water is on the order of 0.35 g/g of collagen (Nomura et al. 1977). As the amount of water uptake in our scaffolds is substantially greater (30-150 g/g) than this value, most of the water in the scaffolds is free.
Our results show that the water uptake in concentrated scaffolds is not a result of an increased amount of collagen but due to the maintenance of three-dimensional structure, two important factors that determine the amount of free water in porous scaffolds.
Another important characteristic of TE scaffolds is to provide an adequate mechanical environment until matrix production by the cells takes control of this process. The optimal environment depends on the targeted tissue. For example, typical compressive modulus of knee articular cartilage is about 7 MPa (Shepherd and Seedhom 1999) and tensile modulus of excised adult human skin is about 50 MPa (Edwards and Marks 1995).
Therefore, it is important to have scaffolds with tunable mechanical properties.
Depending on the type of tissue, performance under either tension, compression or both stresses can affect the ultimate function of the product in vivo. Although there are scaffolds made from synthetic materials that have tunable mechanical properties
(PEGDA, alginate, etc.), tuning mechanical properties of collagen-based scaffolds is under-investigated. Our results suggest that a concentrated CG solution can lead to scaffolds that have tunable mechanical properties over a wide range. For cartilage tissue- engineering application, concentrated scaffolds could be made with a compressive
75 modulus of elasticity an order of magnitude less than that of the native tissue, which is an improvement compared to other similar scaffolds (Harley et al. 2007) and a huge improvement over unconcentrated scaffolds. For skin tissue engineering, concentrated scaffolds could be made with a tensile modulus of elasticity that is an order of magnitude less than that of the native tissue but still better than that of similar scaffolds fabricated by other researchers (Harley et al. 2007). It should be mentioned that while the relative increases observed in dry scaffolds will also be maintained in wet scaffolds, in general, wet scaffolds have inferior mechanical strength when compared to dry scaffolds.
The increases observed in compressive and tensile moduli of elasticity in the concentrated CG solution scaffolds are both due to increased collagen density and changes in pore geometry and distribution. Increasing the collagen density in the scaffold is expected to increase the mechanical strength of the scaffold, as the elastic modulus is proportional to the square of the nominal collagen density (Harley et al. 2007). In our scaffolds, the observed increases in tensile and compressive moduli, however, cannot be explained by the increase collagen density alone. The increases are disproportionally higher: 30- and 56-fold increases in tensile and compressive moduli, respectively, for a fourfold increase in collagen density. It is possible that the base pore structure of the scaffold itself is changed as the collagen density is increased. Smaller pores and pore walls that are comprised of a continuous sheet of collagen in scaffolds made from concentrated CG solution improve the mechanical properties of the scaffold. A contradiction in our results is that the tensile tests showed that modulus of elasticity and ultimate stress decreased when the nominal collagen density increased from 1.1 to 1.2 % w/v. This is probably due to the accuracy with which we can delineate the collagen
76 density in the highly concentrated scaffolds. Indeed, scaffolds obtained after removing
85% solution as supernatant have a wider range of nominal collagen density as shown in
Figure 3-1 and could lead to the disparity in the results. In designing scaffolds with tunable mechanical properties, a compromise must be struck between providing adequate handling properties and ensuring that we do not stress-shield the cells to the point that external beneficial mechanical stimulation is compromised. In addition, improvement in mechanical properties by increasing collagen concentration should be balanced with the degradation rate required for the tissue targeted, as degradation rate decreases with increases in collagen density of the scaffolds.
From a practical tissue engineering perspective, these scaffolds performed quite well. The scaffolds could be stored, handled, cut to size, and sterilized without difficulty. The excellent seeding characteristics of the scaffolds used in the chondrogenesis experiments suggest that they are suitable for a variety of tissue engineering applications. After 3 weeks of bioreactor culture, the cell and GAG contents and the collagen distributions were indistinguishable from those obtained in small-scale scaffold-free aggregate culture
(Welter et al. 2007), suggesting that the scaffold had no deleterious effect on the chondrogenic differentiation process. Further, the scaffolds can also be used as a dermal analog; a differentiated epidermal structure can be obtained by culturing keratinocytes on top of the scaffolds. The new method improves upon the method that was originally developed by Yannas et al (Yannas et al. 1980) and Burke et al. (Burke et al. 1981) to use porous dry substrates of collagen-GAG as dermal equivalents. The scaffolds from the new method can also be used in the development of composite cultured skin equivalents
(Eaglstein and Falanga 1997).
77 3.5 Conclusions
In summary, we describe a simple and rapid centrifugation method to fabricate collagen-
GAG scaffolds with various collagen densities. We show that scaffolds with various collagen densities. We show that scaffolds with various collagen concentrations have differing properties. In general, increasing the collagen density increases water uptake per volume of the scaffold, leads to smaller pore sizes, and to an increase in compressive and tensile moduli. We further demonstrate the biocompatibility of the concentrated collagen scaffolds by tissue engineering cartilage and cultured skin equivalents in vitro. The tunable properties of these scaffolds, as well as the adjustable collagen density which can lead to different degradations rates in vivo, showed them useful for a variety of biomedical applications in regenerative medicine.
3.6 References
Banks-Schlegel S, Green H. 1981. Involucrin synthesis and tissue assembly by keratinocytes in natural and cultured human epithelia. The Journal of Cell Biology 90(3):732-737.
Buijtenhuijs P, Buttafoco L, Poot AA, Daamen WF, van Kuppevelt TH, Dijkstra PJ, de Vos RAI, Sterk LMT, Geelkerken BRH, Feijen J and others. 2004. Tissue engineering of blood vessels: characterization of smooth-muscle cells for culturing on collagen-and-elastin-based scaffolds. Biotechnology and Applied Biochemistry 39(Pt 2):141-149.
Burke JF, Yannas IV, Quinby WC, Bondoc CC, Jung WK. 1981. Successful use of a physiologically acceptable artificial skin in the treatment of extensive burn injury. Annals of Surgery 194(4):413-428.
Carrino DA, Arias JL, Caplan AI. 1991. A spectrophotometric modification of a sensitive densitometric safranin-O assay for glycosaminoglycans. Biochemistry International 24(3):485-495.
78 Chen D-C, Lai Y-L, Lee S-Y, Hung S-L, Chen H-L. 2007. Osteoblastic response to collagen scaffolds varied in freezing temperature and glutaraldehyde crosslinking. Journal of Biomedical Materials Research Part A 80A(2):399-409.
Cross VL, Zheng Y, Won Choi N, Verbridge SS, Sutermaster BA, Bonassar LJ, Fischbach C, Stroock AD. 2010. Dense type I collagen matrices that support cellular remodeling and microfabrication for studies of tumor angiogenesis and vasculogenesis in vitro. Biomaterials 31(33):8596-8607.
Daley WP, Peters SB, Larsen M. 2008. Extracellular matrix dynamics in development and regenerative medicine. Journal of Cell Science 121(3):255-264.
Duan X, Sheardown H. 2006. Dendrimer crosslinked collagen as a corneal tissue engineering scaffold: Mechanical properties and corneal epithelial cell interactions. Biomaterials 27(26):4608-4617.
Eaglstein WH, Falanga V. 1997. Tissue engineering and the development of Apligraf®, a human skin equivalent. Clinical Therapeutics 19(5):894-905.
Edwards C, Marks R. 1995. Evaluation of biomechanical properties of human skin. Clinics in Dermatology 13(4):375-380.
Furth ME, Atala A, Van Dyke ME. 2007. Smart biomaterials design for tissue engineering and regenerative medicine. Biomaterials 28(34):5068-5073.
Garg AK, Berg RA, Silver FH, Garg HG. 1989. Effect of proteoglycans on type I collagen fibre formation. Biomaterials 10(6):413-419.
Glowacki J, Mizuno S. 2008. Collagen scaffolds for tissue engineering. Biopolymers 89(5):338-344.
Han B, Jaurequi J, Tang BW, Nimni ME. 2003. Proanthocyanidin: A natural crosslinking reagent for stabilizing collagen matrices. Journal of Biomedical Materials Research Part A 65A(1):118-124.
Harley BA, Leung JH, Silva ECCM, Gibson LJ. 2007. Mechanical characterization of collagen-glycosaminoglycan scaffolds. Acta Biomaterialia 3(4):463-474.
Huynh T, Abraham G, Murray J, Brockbank K, Hagen PO, Sullivan S. 1999. Remodeling of an acellular collagen graft into a physiologically responsive neovessel. Nature Biotechnology 17(11):1083-1086.
Kuberka M, Heschel I, Glasmacher B. 2002. Preparation of collagen scaffolds and their applications in tissue engineering. Biomedizinische Technik/Biomedical Engineering 47(s1a):485-487.
Lavik E, Langer R. 2004. Tissue engineering: current state and perspectives. Applied Microbiology and Biotechnology 65(1):1-8.
79 Lee CH, Singla A, Lee Y. 2001. Biomedical applications of collagen. International Journal of Pharmaceutics 221(1-2):1-22.
Marijnissen WJCM, van Osch GJVM, Aigner J, van der Veen SW, Hollander AP, Verwoerd-Verhoef HL, Verhaar JAN. 2002. Alginate as a chondrocyte-delivery substance in combination with a non-woven scaffold for cartilage tissue engineering. Biomaterials 23(6):1511-1517.
Martin I, Wendt D, Heberer M. 2004. The role of bioreactors in tissue engineering. Trends in Biotechnology 22(2):80-86.
Medalie DA, Eming SA, Tompkins RG, Yarmush ML, Krueger GG, Morgan JR. 1996. Evaluation of human skin reconstituted from composite grafts of cultured keratinocytes and human acellular dermis transplanted to athymic mice. Journal of Investigative Dermatology 107(1):121-127.
Morritt AN, Bortolotto SK, Dilley RJ, Han XL, Kompa AR, McCombe D, Wright CE, Itescu S, Angus JA, Morrison WA. 2007. Cardiac tissue engineering in an in vivo vascularized chamber. Circulation 115(3):353-360.
Nomura S, Hiltner A, Lando JB, Baer E. 1977. Interaction of water with native collagen. Biopolymers 16(2):231-246.
O'Brien FJ, Harley BA, Yannas IV, Gibson L. 2004. Influence of freezing rate on pore structure in freeze-dried collagen-GAG scaffolds. Biomaterials 25(6):1077-1086.
O'Brien FJ, Harley BA, Yannas IV, Gibson LJ. 2005. The effect of pore size on cell adhesion in collagen-GAG scaffolds. Biomaterials 26(4):433-441.
Orban JM, Wilson LB, Kofroth JA, El-Kurdi MS, Maul TM, Vorp DA. 2004. Crosslinking of collagen gels by transglutaminase. Journal of Biomedical Materials Research Part A 68A(4):756-762.
Papini S, Cecchetti D, Campani D, Fitzgerald W, Grivel JC, Chen S, Margolis L, Revoltella RP. 2003. Isolation and clonal analysis of human epidermal keratinocyte stem cells in long-term culture. Stem Cells (Miamisburg) 21(4):481- 494.
Penick KJ, Solchaga LA, Welter JF. 2005. High-throughput aggregate culture system to assess the chondrogenic potential of mesenchymal stem cells. Biotechniques 39(5):687-691.
Ponticiello MS, Schinagl RM, Kadiyala S, Barry FP. 2000. Gelatin-based resorbable sponge as a carrier matrix for human mesenchymal stem cells in cartilage regeneration therapy. Journal of Biomedical Materials Research 52(2):246-255.
Rosenblatt J, Devereux B, Wallace DG. 1994. Injectable collagen as a pH-sensitive hydrogel Biomaterials 15(12):985-995.
80 Ruszczak Z. 2003. Effect of collagen matrices on dermal wound healing. Advanced Drug Delivery Reviews 55(12):1595-1611.
Saito H, Taguchi T, Aoki H, Murabayashi S, Mitamura Y, Tanaka J, Tateishi T. 2007. pH-responsive crosslinkers swelling behavior of collagen gels prepared by novel based on naturally derived di- or tricarboxylic acids. Acta Biomaterialia 3(1):89- 94.
Sasaki N, Odajima S. 1996. Stress-strain curve and Young's modulus of a collagen molecule as determined by the X-ray diffraction technique. Journal of Biomechanics 29(5):655-658.
Schmidt CE, Baier JM. 2000. Acellular vascular tissues: natural biomaterials for tissue repair and tissue engineering. Biomaterials 21(22):2215-2231.
Shepherd DET, Seedhom BB. 1999. The 'instantaneous' compressive modulus of human articular cartilage in joints of the lower limb. Rheumatology 38(2):124-132.
Shi HF, Han CM, Mao ZW, Ma L, Gao CY. 2008. Enhanced angiogenesis in porous collagen-chitosan scaffolds loaded with angiogenin. Tissue Engineering Part A 14(11):1775-1785.
Sieminski AL, Padera RF, Blunk T, Gooch KJ. 2002. Systemic delivery of human growth hormone using genetically modified tissue-engineered microvascular networks: Prolonged delivery and endothelial survival with inclusion of nonendothelial cells. Tissue Engineering 8(6):1057-1069.
Soo C, Rahbar G, Moy RL. 1993. The immunogenicity of bovine collagen implants. Journal of Dermatologic Surgery and Oncology 19(5):431-434. van Wachem PB, Plantinga JA, Wissink MJB, Beernink R, Poot AA, Engbers GHM, Beugeling T, van Aken WG, Feijen J, van Luyn MJA. 2001. In vivo biocompatibility of carbodiimide-crosslinked collagen matrices: Effects of crosslink density, heparin immobilization, and bFGF loading. Journal of Biomedical Materials Research 55(3):368-378.
Veronda DR, Westmann RA. 1970. Mechanical characterization of skin-finite deformations. Journal of Biomechanics 3(1):111-124.
Wallace DG, Rosenblatt J. 2003. Collagen gel systems for sustained delivery and tissue engineering. Advanced Drug Delivery Reviews 55(12):1631-1649.
Welter JF, Solchaga LA, Penick KJ. 2007. Simplification of aggregate culture of human mesenchymal stem cells as a chondrogenic screening assay. Biotechniques 42(6):732-+.
Wong M. 2004. Alginates in tissue engineering. Biopolymer Methods in Tissue Engineering 238:77-86.
81 Yannas IV, Burke JF, Gordon PL, Huang C, Rubenstein RH. 1980. Design of an artificial skin. II. Control of chemical composition. Journal of Biomedical Materials Research 14(2):107-132.
Yeong WY, Chua CK, Leong KF, Chandrasekaran M, Lee MW. 2007. Comparison of drying methods in the fabrication of collagen scaffold via indirect rapid prototyping. Journal of Biomedical Materials Research Part B-Applied Biomaterials 82B(1):260-266.
82 Chapter 4
DESIGN OF A PERFUSION BIOREACTOR FOR SKIN TISSUE
ENGINEERING
4.1 Introduction
Bioreactors are defined as devices in which biological and/or biochemical processes can be performed under highly controlled environmental and operating conditions such as pH, temperature, pressure, and nutrient supply (Martin et al. 2004). Although bioreactors are well established for the cultivation of microbes or mammalian cells on an industrial scale, the development of bioreactors for tissue engineering applications is still ongoing. The major challenge for bioreactors in tissue engineering is the complexity of different types of TE constructs, and each type of products will require an individualized design
(Ratcliffe and Niklason 2002). Therefore, tissue-specific bioreactors should be designed based on comprehensive understanding of biological and engineering aspects of target tissues. Bioreactors provide not only reliable and reproducible techniques to reveal the biological, chemical, or physical effects for specific tissues, but also the potential to improve the quality of TE constructs (Martin et al. 2004; Martin and Vermette 2005).
In tissue engineering, bioreactors are used for different purposes such as cell proliferation
(Kino-Oka et al. 2005), generation of three-dimensional tissue products from isolated cells (Li et al. 2008; Wang et al. 2005), or maintenance of living tissues (Ladd et al.
2009). For instance, bioreactors have become more important in organ transplantations since researchers can utilize them to develop a wide variety of TE constructs in vitro.
83 There are many factors that need to be considered when designing bioreactors. For example, they should have successful and adequate cell seeding manners to yield uniform cell distributions within large scaffolds. Culturing conditions such as medium composition and mechanical conditions can also have significantly impact on tissue growth. Mass transfer of oxygen and waste products is one of the most important factors limiting large tissue culture. Last, zero contamination risk, scalability, and ease of manufacturing, handling and storage all need to be considered when designing bioreactors. As a result, design of bioreactors requires collaboration between engineers and biologists.
As mentioned before, each TE construct will require individualized bioreactor design. In cartilage tissue engineering, custom-designed bioreactors were developed to culture cell- seeded scaffolds under dynamic unconfined compression. Cartilage produced by articular chondrocytes on polyglycolic acid scaffolds or agarose disks in the bioreactors show superior mechanical properties compared to cartilage cultured in static conditions
(Dunkelman et al. 1995; Gooch et al. 2001; Mauck et al. 2000). More recently, Penick et al. had developed a low-volume, modular, parallel plate bioreactor for the culture of TE cartilage (Penick et al. 2005b). The design included gas permeable surfaces and a biocompatible Delrin® substrate. In bone tissue engineering, a rotating wall vessel bioreactor (RMVB), which generates very low shear stress levels on cells, has been developed for culturing osteoblasts with bio-derived bone scaffolds for three-dimensional bone grafts (Sikavitsas et al. 2002; Song et al. 2008). The results indicate that the cells expanded in the bone grafts using RMVB were improved by as much as five times compared to that developed using T-flask and spinner flask. Also, there were
84 significantly more collagen fibers mineralized nodules and new osteoid tissue formed than those in T-flask and spinner flask. In cardiovascular tissue engineering, functional autologous arteries were cultured on tubular scaffolds in a perfusion bioreactor, which apply pulsatile radial distensions at a controlled frequency and radial strain to develop compliant vessels (Niklason et al. 1999). Vessel wall thickness, collagen content, and suture retention strengths were all improved significantly after 8 weeks of pulsatile culture. On the other hand, TE heart valves cultured using a dynamic bioreactor also show improved cell penetration into scaffolds due to mechanical stimuli to which cells were subjected (Colton 1995; Gandaglia et al. 2010; Shachar and Cohen 2003).
In skin tissue engineering, Kerator, a computer controlled bioreactor for automated culture and harvest of HKs, has been developed by Prenosil et al. (Kalyanaraman and
Boyce 2007; Kalyanaraman and Boyce 2009; Kino-Oka and Prenosil 2000; Prenosil and
Kino-Oka 1999; Prenosil and Villeneuve 1998). Although the Kerator reduced labor and material requirements and increase availability of HKs for TE skin substitutes, it only provides a method for culturing HKs in a submerged culture mode. On the other hand,
Sun et al. developed a closed bioreactor system for culturing TE skin at an air-liquid interface (Sun et al. 2005). In their system, scaffold and bioreactor sterilization, cell seeding, and medium perfusion were all performed using a peristaltic pump. While these above reactor systems have certain features that are advantageous, none of them can be used to culture differentiated TE skin substitutes with built-in convective channels.
In tissue engineering, researchers have been showing that the problems associated with poor diffusion can be eliminated using perfusion bioreactors (Cartmell et al. 2003).
Moreover, one of the advantages of a perfusion bioreactor is delivery of controlled
85 mechanical stimuli such as flow-induced shear stresses, scaffold strains, or hydrostatic pressures to tissue products (Engbers-Buijtenhuijs et al. 2006; Jeong et al. 2005; Price et al. 2010; Sikavitsas et al. 2003). Several perfusion bioreactors have been developed for bone and cartilage tissue engineering (Alves da Silva et al. 2010). On the other hand, there are not many perfusion bioreactors for skin tissue engineering. Kalyanaraman et al. had developed a perfusion culture system for TE skin substitutes and discussed the effect of medium flow rate on viability and barrier function of skin substitutes (Kalyanaraman et al. 2008). However, the scaffolds did not have integrated perfusion networks. The goal of this study was to develop a perfusion bioreactor system for TE skin substitutes with integrated flow networks. We used computational models to show that glucose concentration profiles are improved in the bioreactor system.
4.2 Materials and Methods
The bioreactor consists of a solid chamber made by Dupont Delrin® (McMaster-Carr,
Aurora, OH) and the dimensions of the inner chamber are 6.5 x 3.5 x 0.8 cm (length x width x depth), as illustrated in Figure 4-1 and Figure 4-2. One side of the bioreactor is covered by a O2 and CO2 permeable fluorinated ethylene propylene (FEP®, Dupont,
McMaster-Carr) membrane. The FEP membrane and the Delrin reactor block is sealed by a silicone rubber gasket and the entire assembly is held in place with an aluminum block, six screws (size: 6/32), and wing nuts (McMaster-Carr). The reactor contains five polyvinylidene fluoride (PVDF) ports (Value Plastics, Inc., Fort Collins, CO) which are permanently fixed on the bioreactors and connected to Luer connectors (McMaster-Carr) at the other end. Three of them are located on the side of the bioreactors and are used as an inlet, outlet, and for adjusting the medium level during the submerged mode. The
86 other two ports are on the bottom of the bioreactor and can be used to supply culture medium during the air-liquid interface mode. During the submerged mode, about 15 mL fresh medium is injected manually into the culture chamber to submerge TE skin substitutes for keratinocyte proliferation, while fresh medium is perfused to the integrated flow networks within TE skin substitutes for keratinocyte differentiation using Harvard
Apparatus syringe pump (PHD 2000, Harvard Apparatus, MA, USA) during the air- liquid interface mode. Waste is aspirated from culture chamber during the submerged mode whenever necessary, while waste is collected continuously by using a 50 mL
Falcon centrifugal tube [Figure 4-3]. All parts of the bioreactor can be steam sterilized in an autoclave.
Figure 4-1 Digital image of the bioreactor from (a) top view and (b) side view. 4-1(a) shows the reactor chamber with the gas permeable FEP membrane. The three ports on the side allow maintenance of TE skin during submerged mode. The two openings inside the reactor chamber correspond to the ports for feeding medium during TE skin culture in air-liquid interface mode. 4-1(b) shows the side view of the reactors with these ports at the bottom of the reactor.
87
Figure 4-2 Schematic diagram of the bioreactor design. The system consists of a culture chamber (1), PDMS flat membrane for support (2), CG membrane with built-in microchannels (3), microchannels for internal flow (4), and a gas permeable FEP membrane (5). For submerged culture, medium are delivered into chamber with (6) syringe and collected in a container (9) through an inlet (7) and an outlet (8). For air/liquid interface culture, medium are pumped into chamber with a syringe pump (10) and drained to another container (14) through an inlet (12) and an outlet (13).
88
Figure 4-3 Picture of a bioreactor in an air-liquid interface mode using a syringe pump to perfuse medium through TE skin substitutes. Waste was collected continuously by using a 50 mL Falcon centrifugal tube.
To better understand the glucose concentration profiles in the bioreactor and further prove that more uniform glucose concentration profiles can be obtained in the bioreactor system, we used computational models to predict the glucose concentration profiles for different flow rates at two surfaces in the TE skin substitutes during bioreactor operation: the interface between the dermal analog and the flow networks [dotted line in Figure 4-4
(a)], and the interface between the dermal analog and the epidermal analog [dashed line in Figure 4-4 (a)]. The thickness of the dermal analog is 500 µm. The epidermis is modeled as a reactive surface on top of the dermis. The depth of the flow networks is 100
µm. The thickness of dermal analog above the flow network is 100 µm and below the flow network is 300 µm. By combining the convection, diffusion, and reaction, glucose concentration profiles in the bioreactors were calculated using finite-element models
89 (COMSOL Multiphysics 3.2, COMSOL Inc., Burlington, MA). Due to symmetry, only half the domain was used for simulations as this helped with efficient use of memory to obtain accurate results. The domain was meshed to obtain about 46,000 elements.
The Navier-Stokes equations were first solved to obtain steady-state velocity profiles within the flow networks. The boundary conditions included specification of the: average velocity at the flow inlet (58 #m/s), the pressure at the outlet (101.3 kPa), and the assumption of no-slip condition at all other boundaries. These results were then used with species conservation laws to determine the glucose concentration profiles in the system.
The boundary conditions for the latter included specification of the: glucose concentration (6 mol/m3) at inlet, and surface glucose consumption by keratinocytes at the epidermal-dermal interface [Figure 4-4]. No flux conditions were assumed at all other boundaries. The consumption rate of glucose by keratinocytes is expressed by the
Vmax [S] Michaelis-Menten equation: v = , where Vmax is the maximum rate of conversion Km + [S]
-5 2 (3.1"10 mol/m /sec), [S] is substrate concentration, and Km is the Michaelis constant (6 mol/m3) (Spravchikov et al. 2001). Diffusivities of glucose in dermal analog and medium are 6.4"10-6 and 6.7"10-6 cm2/sec, respectively.
90
Figure 4-4 Schematic diagrams of a TE skin substitute for computational modeling. (a) A TE skin substitute consists of an epidermal analog, modeled as a reactive surface on top of the dermal analog, and a dermal analog (500 µm) on the bottom. Flow networks (100 µm) is embedded within the dermal analog. The thickness of dermal analog above the flow network is 100 µm and below the flow network is 300 µm. Two surfaces of glucose concentration profiles for different flow rates were predicted: the interface between the dermal analog and the flow networks (dotted line), and the interface between the dermal analog and the epidermal analog (dashed line). (b) Half design of flow network. (c) Boundary conditions for species conservation laws: maximum glucose concentration at inlet (6 mol/m3) and surface glucose consumption by keratinocytes at the epidermal- dermal interface.
91 4.3 Results and Discussion
In order to decrease the risk for contamination, the materials for constructing bioreactors should be non-toxic and sterilizable. We chose Dupont Delrin for the bioreactors, PVDF for the ports, and silicone tubing for perfusing medium as they are non-toxic and steam sterilizable. Delrin is an engineering resin with suitable mechanical properties and is an excellent material for fabricating of laboratory tools such as bioreactors. It is a polymer of formaldehyde and contains stabilizers and/or fillers, which potentially could be released into the medium as formaldehyde or other thermal breakdown products upon repeated autoclaving. Penick et al. showed that there is no negative effect on the proliferation and chondrogenic differentiation of mesenchymal stem cells (MSCs) due to long-term exposure to Delrin after culturing MSCs in Delrin bioreactors for up to 3 weeks (Penick et al. 2005a). Our results with TE skin substitutes verify these findings.
It is important that the bioreactors can be operated with minimal human intervention.
This is essential not only for controlled, reproducible, and statistically relevant basic studies, but also for future routine manufacturing of skin tissue constructs for clinical use
(Kalyanaraman et al. 2008). In our system, a syringe pump is used to perfuse medium.
Each pump can hold up to ten syringes, and since the pump is used only during air-liquid interface mode, up to ten reactors can be perfused by a single pump. In addition, the syringe pump allows automatic dispensing of medium either by controlling the rate or the volume of medium dispensed. Luer connecters provide easy access to the inside of the bioreactor, while maintaining its isolation, thus minimizing any risk of contamination.
Culturing TE skin substitutes using HKs requires a submerged culture mode for proliferation of HKs and then an air/liquid interface culture mode for differentiation to
92 develop well-differentiated epidermis. In our system, the liquid handling ports outside the substitutes help form the submerged mode of culture, whereas an air/liquid interface culture mode will be obtained by perfusing the integrated flow network in the construct with medium using two separate ports.
It has been widely accepted that the supply of oxygen and soluble nutrients becomes critically limiting for culturing three-dimensional TE products. Oxygen, which is an important molecule in all aerobic metabolic cycles, is one of the most important nutrients for cells. In most conditions, mass transfer of oxygen from the gaseous phase to the culture medium control the rate of metabolism of the cells. Due to poor solubility in water, oxygen often becomes the limiting nutrient for successful tissue growth in vitro.
Normally, the oxygen solubility in the culture medium is thermodynamically limited to about 0.2 mM when atmospheric oxygen is used (Glacken et al. 1983). Therefore, mass transfer of oxygen to and from tissues is a key factor in most bioreactor designs.
Overall, the molar flux of oxygen to the cells in the bioreactors must be sufficient to maintain a normal respiratory rate of the cells and to prevent cell death and tissue necrosis. This overall minimal flux can be estimated by experiments on freely suspended cells. By assuming that the rate of respiration of cells is identical in tissues and in freely suspended cells, we can approximate the total molar flux necessary to maintain normal cell metabolism relative to oxygen. After obtaining the overall molar flux, the bioreactors need to be designed properly to achieve the overall molar flux. To achieve this, FEP membranes were utilized to exchange oxygen in the culture chamber from the incubator.
We chose FEP membranes because of the following reasons: (1) the membranes are transparent and thus can serve as the top window to observe the TE skin substitute inside
93 the bioreactor; (2) the membranes are O2 and CO2 gas permeable (Gabridge and Gladd
1984; Martin et al. 2004); (3) the membranes are not affected by the two most common forms of sterilization, autoclaving and gamma irradiation; and (4) the membranes are chemically inert, flexible, and weldable (Jeong et al. 2005). Pathi et al. also showed that
FEP membranes are suitable for providing oxygen supply (Pathi et al. 2005). It should be mentioned that oxygen supply is not expected to be an issue in our system as keratinocytes are at the surface of TE skin substitutes and form a thickness (~100 µm) that is well within the diffusional thickness of oxygen.
Glucose is another important nutrient that supports cell growth. Although glucose has much higher solubility in the culture medium compared to oxygen, it is also crucial to maintain uniform glucose concentration in the bioreactors. During an air-liquid interface mode, the only source of glucose is the medium convected through the integrated flow networks from which glucose diffuses through the dermal analog to the epidermal analog, where HKs consume glucose. However, the dermal analog can form key resistance to glucose transporting to the cells.
The simulation results show that the glucose levels in the middle of flow networks increase significantly when flow rates increase from 100 µL to 5 mL/h [Figure 4-5]. In our model, the maximum glucose concentration is 6 mol/m3, which is the same as that in
Dulbecco’s Modified Eagle Medium (low glucose) (D-MEM, Invitrogen). The glucose concentrations increase from 2 mol/m3 to 5 mol/m3 at the interface between the dermal analog and the flow networks [Figure 4-5 (a)], while the glucose concentrations increase from 1 mol/m3 to 2 mol/m3 at the interface between the dermal analog and the epidermal analog. The decreases of glucose concentrations between two interfaces are due to
94 resistance to glucose transporting to the cells from the dermal analog. However, 2 mol/m3 is sufficient for keratinocyte proliferation and differentiation. Interestingly, Spravchikov et al. have shown that glucose transport rates increased when keratinocytes were cultured at low glucose concentration (2 mol/m3) (Spravchikov et al. 2001). Despite glucose concentrations increase in the middle of flow networks, glucose concentration profiles are also more uniform throughout the construct at higher flow rates. It should be mentioned that although increasing flow rates will lead to uniform glucose concentration profiles, it will also lead to increased medium utilization. These results should be used together with histology and other experimental results to obtain optimal flow rates.
Figure 4-5 Glucose concentration profiles in the interface (a) between the dermal analog and the flow networks and (b) between the dermal analog and the epidermal analog. The numbers underneath the figure indicate the flow rates in µL/hr.
95 4.4 Conclusions
In summary, we design a perfusion bioreactor for TE skin substitutes with integrated flow networks. The perfusion bioreactor provides not only a submerged culture mode for keratinocyte proliferation but also an air-liquid interface culture mode for keratinocyte differentiation. We discuss the effect of perfusion rates on glucose concentration profiles in the bioreactor by using computational models. The optimal perfusion rate can be obtained by cooperating these results with experimental considerations.
4.5 References
Alves da Silva ML, Martins A, Costa-Pinto AR, Correlo VM, Sol P, Bhattacharya M, Faria S, Reis RL, Neves NM. 2010. Chondrogenic differentiation of human bone marrow mesenchymal stem cells in chitosan-based scaffolds using a flow- perfusion bioreactor. Journal of Tissue Engineering and Regenerative Medicine.
Cartmell SH, Porter BD, García AJ, Guldberg RE. 2003. Effects of medium perfusion rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Engineering 9(6):1197-1203.
Colton CK. 1995. Implantable biohybrid artificial organs. Cell Transplantation 4(4):415- 436.
Dunkelman NS, Zimber MP, LeBaron RG, Pavelec R, Kwan M, Purchio AF. 1995. Cartilage production by rabbit articular chondrocytes on polyglycolic acid scaffolds in a closed bioreactor system. Biotechnology and Bioengineering 46(4):299-305.
Engbers-Buijtenhuijs P, Buttafoco L, Poot AA, Dijkstra PJ, de Vos RAI, Sterk LMT, Geelkerken RH, Vermes I, Feijen J. 2006. Biological characterisation of vascular grafts cultured in a bioreactor. Biomaterials 27(11):2390-2397.
Gabridge MG, Gladd MF. 1984. Gaseous oxide toxicity evaluated with cell monolayers on collagen-coated, gas-permeable teflon membranes. Environmental Health Perspectives 54:347-352.
Gandaglia A, Bagno A, Naso F, Spina M, Gerosa G. 2010. Cells, scaffolds and bioreactors for tissue-engineered heart valves: a journey from basic concepts to
96 contemporary developmental innovations. European Journal of Cardio-Thoracic Surgery In Press, Corrected Proof.
Glacken MW, Fleischaker RJ, Sinskey AJ. 1983. Large-scale production of mammalian cells and their products: engineering principles and barriers to scale-up. Annals of the New York Academy of Sciences 413(1):355-372.
Gooch KJ, Blunk T, Courter DL, Sieminski AL, Bursac PM, Vunjak-Novakovic G, Freed LE. 2001. IGF-I and mechanical environment interact to modulate engineered cartilage development. Biochemical and Biophysical Research Communications 286(5):909-915.
Jeong SI, Kwon JH, Lim JI, Cho S-W, Jung Y, Sung WJ, Kim SH, Kim YH, Lee YM, Kim B-S and others. 2005. Mechano-active tissue engineering of vascular smooth muscle using pulsatile perfusion bioreactors and elastic PLCL scaffolds. Biomaterials 26(12):1405-1411.
Kalyanaraman B, Boyce ST. 2007. Assessment of an automated bioreactor to propagate and harvest keratinocytes for fabrication of engineered skin substitutes. Tissue Engineering 13(5):983-993.
Kalyanaraman B, Boyce ST. 2009. Wound healing on athymic mice with engineered skin substitutes fabricated with keratinocytes harvested from an automated bioreactor. Journal of Surgical Research 152(2):296-302.
Kalyanaraman B, Supp DM, Boyce ST. 2008. Medium flow rate regulates viability and barrier function of engineered skin substitutes in perfusion culture. Tissue Engineering Part A 14(5):583-593.
Kino-Oka M, Ogawa N, Umegaki R, Taya M. 2005. Bioreactor design for successive culture of anchorage-dependent cells operated in an automated manner. Tissue Engineering 11(3-4):535-545.
Kino-Oka M, Prenosil JE. 2000. Development of an on-line monitoring system of human keratinocyte growth by image analysis and its application to bioreactor culture. Biotechnology and Bioengineering 67(2):234-239.
Ladd MR, Lee SJ, Atala A, Yoo JJ. 2009. Bioreactor maintained living skin matrix. Tissue Engineering Part A 15(4):861-868.
Li W-J, Jiang YJ, Tuan RS. 2008. Cell-nanofiber-based cartilage tissue engineering using improved cell seeding, growth factor, and bioreactor technologies. Tissue Engineering Part A 14(5):639-648.
Martin I, Wendt D, Heberer M. 2004. The role of bioreactors in tissue engineering. Trends in Biotechnology 22(2):80-86.
97 Martin Y, Vermette P. 2005. Bioreactors for tissue mass culture: Design, characterization, and recent advances. Biomaterials 26(35):7481-7503.
Mauck RL, Soltz MA, Wang CCB, Wong DD, Chao P-HG, Valhmu WB, Hung CT, Ateshian GA. 2000. Functional tissue engineering of articular cartilage through dynamic loading of xhondrocyte-seeded agarose gels. Journal of Biomechanical Engineering 122(3):252-260.
Niklason LE, Gao J, Abbott WM, Hirschi KK, Houser S, Marini R, Langer R. 1999. Functional arteries grown in vitro. Science 284(5413):489-493.
Pathi P, Ma T, Locke BR. 2005. Role of nutrient supply on cell growth in bioreactor design for tissue engineering of hematopoietic cells. Biotechnology and Bioengineering 89(7):743-758.
Penick KJ, Solchaga LA, Berilla JA, Welter JF. 2005a. Performance of polyoxymethylene plastic (POM) as a component of a tissue engineering bioreactor. Journal of Biomedical Materials Research Part A 75A(1):168-174.
Penick KJ, Solchaga LA, Welter JF. 2005b. High-throughput aggregate culture system to assess the chondrogenic potential of mesenchymal stem cells. Biotechniques 39(5):687-691.
Prenosil JE, Kino-Oka M. 1999. Computer controlled bioreactor for large-scale production of cultured skin grafts. Annals of the New York Academy of Sciences 875(1):386-397.
Prenosil JE, Villeneuve PE. 1998. Automated production of cultured epidermal autografts and sub-confluent epidermal autografts in a computer controlled bioreactor. Biotechnology and Bioengineering 59(6):679-683.
Price GM, Wong KHK, Truslow JG, Leung AD, Acharya C, Tien J. 2010. Effect of mechanical factors on the function of engineered human blood microvessels in microfluidic collagen gels. Biomaterials 31(24):6182-6189.
Ratcliffe A, Niklason LE. 2002. Bioreactors and bioprocessing for tissue engineering. Annals of the New York Academy of Sciences 961(1):210-215.
Shachar M, Cohen S. 2003. Cardiac tissue engineering, ex-vivo: design principles in biomaterials and bioreactors. Heart Failure Reviews 8(3):271-276.
Sikavitsas VI, Bancroft GN, Holtorf HL, Jansen JA, Mikos AG. 2003. Mineralized matrix deposition by marrow stromal osteoblasts in 3D perfusion culture increases with increasing fluid shear forces. Proceedings of the National Academy of Sciences of the United States of America 100(25):14683-14688.
Sikavitsas VI, Bancroft GN, Mikos AG. 2002. Formation of three-dimensional cell/polymer constructs for bone tissue engineering in a spinner flask and a
98 rotating wall vessel bioreactor. Journal of Biomedical Materials Research 62(1):136-148.
Song K, Liu T, Cui Z, Li X, Ma X. 2008. Three-dimensional fabrication of engineered bone with human bio-derived bone scaffolds in a rotating wall vessel bioreactor. Journal of Biomedical Materials Research Part A 86A(2):323-332.
Spravchikov N, Sizyakov G, Gartsbein M, Accili D, Tennenbaum T, Wertheimer E. 2001. Glucose effects on skin keratinocytes. Implications for diabetes skin complications. Diabetes 50(7):1627-1635.
Sun T, Norton D, Haycock JW, Ryan AJ, MacNeil S. 2005. Development of a closed bioreactor system for culture of tissue-engineered skin at an air-liquid interface. Tissue Engineering 11(11-12):1824-1831.
Wang D, Liu W, Han B, Xu R. 2005. The bioreactor: a powerful tool for large-scale culture of animal cells. Current Pharmaceutical Biotechnology 6(5):397-403.
99 Chapter 5
DEVELOPMENT OF AN IN VITRO TISSUE-ENGINEERED
SKIN SUBSTITUTE WITH INTEGRATED FLOW NETWORKS
IN A PERFUSION BIOREACTOR
5.1 Introduction
Tissue-engineered (TE) skin substitutes have enjoyed moderate success in the clinics.
The gold standard of treatment for cutaneous wounds requiring skin transplantation is still autologous split-thickness skin grafts. Composite TE skin substitutes, containing dermal and epidermal analogs, however, may be the only choice for patients with large area wounds and limited donor sites. Unfortunately, several drawbacks limit the potential of TE skin substitutes for clinical applications. Lack of vascularization has been suggested as one of the major limitations for current TE skin substitutes (Boyce 1996).
For a given TE skin substitute to attach promptly and remain alive, rapid vascularization is required after transplantation. Until vascularization of the implanted tissue occurs, nutrients delivery and waste removal still rely primarily on passive diffusion.
To address mass transfer limitations, porous scaffolds are utilized instead of solid matrices. The porosity of these scaffolds must be designed to fulfill the requirements for vascularization and diffusion of nutrients (Freyman et al. 2001). Certain scaffold modifications have recently been proposed to overcome this limitation. For example, angiogenin, a protein inducer of angiogenesis, was loaded in porous collagen-chitosan scaffolds to promote angiogenesis in dermis. The results showed that the presence of
100 angiogenin enhanced the angiogenesis of implanted substitutes (Shi et al. 2008). Supp et al. developed TE skin substitutes containing genetically modified keratinocytes that overexpress vascular endothelial growth factor. The modified grafts exhibited increased numbers of blood vessels and decreased time to achieve vascularization compared to unmodified control grafts. The results indicate that genetic modification of keratinocytes in TE skin substitutes could enhance vascularization after transplantation (Supp et al.
2000). However, gene therapy has many risks in a clinical setting. More importantly, mass transport limitation will still occur due to passive diffusion of substrate until TE skin substitutes are vascularized in vivo.
Young et al. have proposed two mechanisms for vascularization of TE skin substitutes: inosculation, the anastomosis of the grafts vessels to the host vasculature, and neovascularization, the growth of completely new vessels from the host vasculature
(Young et al. 1996). Tremblay et al. showed that the inosculation was achieved less than
4 days after transplantation when using TE skin substitutes with capillary-like structures formed from seeded endothelial cells, while a 14-day period was required to achieve similar results for neovascularization (Tremblay et al. 2005). Recently, Gibot et al. developed TE skin substitutes from xenogeneic keratinocytes, fibroblasts, and endothelial cells without any exogenous angiogenic growth factors or scaffolds (Gibot et al. 2010).
The results showed that the endothelial cells formed capillary-like structures and may have played a role in early inosculation between the grafts and the host vasculature 4 days after transplantation in athymic mice. These results suggest that TE skin substitutes with capillary-like structures can form functional anastomoses with the host blood vessels and promote rapid and complete vascularization of the implanted tissues by exerting an
101 angiogenic effect. Despite the recent advances, to our knowledge, there have been no TE skin substitutes with integrated vascular networks prior to implantation.
BioMEMS, commonly referred to as biomedical and biological microelectro-mechanical systems, has become increasingly intriguing research areas in a wide variety of applications such as biosensors, cell biology, drug delivery, and tissue engineering. In general, BioMEMS can be defined as “devices or systems constructed using micro- or nanoscale fabrication technique for investigations in biomedical and biological research.
They frequently involve silicon-based micromachining to fabricate microscale stamps and microfluidic devices. They provide several unique features for biomedical research, including analyte sensitivity, electrical responsiveness, temporal control, and feature sizes similar to cells, tissues, and organs (Grayson et al. 2004). Soft lithography is an alternative microfabrication technique based on replication molding. Typically, an elastomer is patterned by curing on a micromachined mold to create precise microscale features. The advantages of soft lithography include the capacity for rapid prototyping, easy fabrication with low cost, and adaptable process parameters (Unger et al. 2000).
Shin et al. developed a methodology based on BioMEMS and soft lithography to create endothelialized networks with a vascular geometry using biocompatible poly(dimethyl siloxane) (PDMS) matrices (Shin et al. 2004). A new soft lithography technique was established by our group recently to cast microscale networks onto collagen- glycosaminoglycan (CG) membranes (Janakiraman et al. 2007a). The embedded networks are very stable with a feature resolution on the order of 2-3 µm, while the technique has versatility and can be applied to other soft biomaterial matrices. Previously, bifurcating capillary flow networks with optimal transport characteristics were designed
102 by our group using human skin as a model (Janakiraman et al. 2007b). Endothelialization of these networks in CG membranes can be a platform for developing TE skin substitutes with integrated vascular networks.
Perfusion bioreactors provide dynamic fluid flow and can simulate the physiological conditions in vitro (Minuth et al. 2000). In general, the bioreactor can modulate both biochemical cues, including regulating cytokines and growth factors and providing appropriate signaling environments, and biophysical cues, including conducting shear stress, hydrostatic and dynamic pressures to cells and tissues. It can also facilitate mass transfer of nutrients and wastes between cells and environments (Pathi et al. 2005). In addition to enhancing mass transport within TE constructs, cell proliferation and differentiation, and extracellular matrix production and composition can be improved significantly in perfusion culture compared to static culture (Arano et al. 2010; Minuth et al. 2000; Zhao and Ma 2005). Studies have shown that lower flow rates lead to better results on the growth of TE constructs (Cartmell et al. 2003; Kalyanaraman et al. 2008).
This phenomenon may result from several factors such as the magnitude of shear stress, autocrine/paracrine signaling effect, and the concentration of dissolved oxygen (Dan et al.
2010). Cartmell et al. have demonstrated that lower flow rates promoted cell proliferation while higher rates accelerated osteoblast differentiation but also let to increased levels of cell death (Cartmell et al. 2003). Conversely, Grayson et al. have found that higher flow rates did not decrease human mesenchymal stem cells (hMSCs) viability, but rather improved tissue distribution within scaffolds (Grayson et al. 2009). These results all indicated that flow rates may regulate cellular behavior directly within TE constructs and
103 should be optimized based on architecture and porosity of scaffolds, cell and tissue types, medium composition and viscosity, and also bioreactor systems (Grayson et al. 2010).
In this study, we developed an in vitro model of TE skin substitute with integrated flow networks cultured in a perfusion bioreactor. We hypothesized the perfusion culture could influence the growth of TE skin substitutes compared to static culture. The effect of flow rates on epidermis formation and cell viability were investigated through histological analysis, immunohistochemical analysis, and viability assay.
5.2 Materials and Methods
5.2.1 Preparation of Dermal Substitutes with Integrated Flow Networks
The process of preparation of dermal substitutes with flow networks consists of the following strategy [Figure 5-1]. First, silicon templates with the networks were microfabricated. The template designs were then transferred to CG scaffolds to obtain an open network on one side and a flat surface for cell seeding on the other side. The scaffold was attached to a PDMS mold with the open network facing to the PDMS mold using a double-sided silicone adhesive. Keratinocytes were seeded on the flat surface of the scaffold and cultured in a submerged mode. During air-liquid interface culture mode operation, the TE skin substitute (CG scaffolds with keratinocytes) was transferred to another PDMS mold that was attached to a perfusion bioreactor and contained another double-sided silicone adhesive for anchoring the substitute. This last step converted the open network into a closed network, and allowed supply of air-liquid interface (ALIM) medium through the network incorporated in the TE skin substitute. Below, we describe the individual process in detail.
104
Figure 5-1 Schematic of preparation of TE skin substitute in a perfusion bioreactor. (a) A silicon substrate is microfabricated to obtain a silicon template with networks. (b) A CG scaffold is micropatterned using the silicon substrate to obtain an open network on one side and a flat surface for cell seeding on the other side. (c) A PDMS mold for submerged culture mode is prepared and a silicone double-sided adhesive is put to PDMS mold. The CG scaffold with an open network is attached to the PDMS mold with network facing to PDMS mold. During submerged culture mode, keratinocytes are seeded on the top of flat surface and cultured for 3 days. (d) For air-liquid interface culture mode, another PDMS mold with the silicone adhesive is prepared and holes for inlet and outlet ports that correspond to the ports in the bioreactor are punched out. The PDMS mold is attached to a perfusion bioreactor. After submerged culture mode, the TE skin substitute (CG scaffolds with keratinocytes) is transferred to the PDMS mold. This last step converts the open network into a closed network. ALIM is perfused through the network to obtain air- liquid interface culture mode.
105 5.2.1.1 Fabrication of Silicon Substrate with Optimal Flow Network
Design
Prior to integrating flow networks into CG scaffolds as dermal analogs, silicon substrates with flow networks were fabricated utilizing SU-8 2075 negative photoresist (PR)
(MicroChem Corp., MA, USA) by standard UV lithography. First, photomasks, in which the network design features are transparent while the surroundings are opaque, were obtained from Advance Reproductions Corp., MA. The photomasks were made by printing AutoCAD designs of flow networks on transparent photofilms in a high resolution (>3000 dpi) laser printer. The plastic photomasks provide low-cost alternatives to traditional chrome masks and have resolution in the order of ~2 µm. Standard UV lithography was then carried out in Electronics Design Center at CWRU to obtain silicon templates for subsequent soft lithography process. The process for fabricating the negative templates of PR in a silicon substrate is shown in Figure 5-2 and is described in detail below.
(a) SU-8 PR spin coating: A uniform film of PR was coated on the silicon substrate
(University Wafer) using a spin coater (Laurell Technologies Co., PA). The thickness of coated PR film primarily depends on spin speed; spin conditions, therefore, must be optimized to achieve the desired thickness. PR was dispensed carefully onto the substrate with volume ratio of 1 mL for 1 inch of substrate without generating bubbles followed by spin cycle. There are two steps in the spinning cycle, including spinning at 500 rpm for
10 s with acceleration of 100 rpm/s to spread PR on the substrate, and spinning at 1000 rpm for 30 s with acceleration of 200 rpm/s to obtain the desired thickness. The final thickness of coated PR film is about 200 µm using this spin cycle.
106 (b) Soft bake: After spin coating, soft baking was performed to evaporate solvent and to achieve PR self-planarization by using a level hotplate. During the process, the gravity force can affect the flatness of the PR film. Hence, the hotplate was carefully adjusted to a horizontal position prior to the baking process. Baking time and conditions were determined based on the thickness of PR. For 200 µm PR film, the temperature was held at 75oC for 7 min and then ramped gradually to 95oC and held for 45 min. The substrate was then cooled down to room temperature gradually by leaving the substrates on the hotplate for 1 h after turning off hotplate.
(c) UV lithography: To obtain vertical sidewalls, UV radiation below 350 nm was eliminated using a long pass filter from Omega Optical, VT. The UV lithography was processed using a mask aligner (ABM, Inc., CA) with exposure intensity of 1.4 mW/cm2.
The total exposure dose was determined based on the thickness of PR film. With optimal exposure, a visible latent image will be seen in the film within 5-15 s after placing on the hotplate for post exposure bake (PEB). For 200 µm PR film, the total exposure dose was about 700 mJ/ cm2.
(d) PEB: After exposure, PEB was carried out directly using a level hotplate to selectively cross-link the exposed portions of the film. During PEB, cross-linking of PR can result in a highly stressed film. As a result, a two-step process followed by slow cooling was used to minimize stress. For 200 µm PR film, the temperature was held at
75oC for 5 min and then ramped gradually to 95oC and held for 15 min. The substrates were then cooled down to room temperature gradually by leaving the substrates on the hotplate for 1 h after turning off hotplate.
107 (e) Development. The substrates were developed using SU-8 developer (MicroChem
Corp., MA, USA) with gentle agitation for 12 min. This process dissolved uncross-linked
PR. Finally the substrates were rinsed with isopropyl ethanol and dried with nitrogen.
Figure 5-2 Schematic of standard UV light lithography
5.2.1.2 Fabrication of CG Scaffolds with Flow Networks
CG scaffolds were fabricated using the method as described in Chapter 3 (Section 3.2.2).
In this study, 30 mL of supernatant was removed from each centrifugal tube after centrifugation. The remaining solution underwent the standard lyophilization process followed by 24 h dehydrothermal cross-linking treatment at 120°C. All CG scaffolds were stored at 4°C prior to use.
CG scaffolds were patterned using a modified lithography technique described in Chapter
1, Section 1.4.2. A single bifurcation design was chosen for these studies as it is a lot easier to fabricate and our primary goal is to demonstrate the efficacy of the integrated flow networks on TE skin growth. Prior to patterning, silicon wafer templates with the flow networks were cut to obtain individual flow patterns (2.5 x 6 cm). A CG scaffold of
108 size 2.5 (width) x 6 (length) x 0.5 (depth) cm was placed in a P-150 petri dish. 0.5 mL of acetic acid solution (0.5 v/v%) was then spread on top of the scaffold to initial selective dissolution. A cut silicon template with a single flow network was placed on top of the scaffold with pattern side down and in contact with the scaffold. To achieve close uniform contact and good resolution, a glass slide was placed on top of the silicon template followed by an aluminum bars (1 x 3 x 2 in). Carbodiimide (EDC) solution (14 mM 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride and 5.5 mM N- hydroxysuccinimide; Sigma) (Lee et al. 2001) was subsequently added to submerge the scaffolds to stabilize and cross-link the dissolved collagen network around the PR features. After 48 h, CG scaffolds were removed, cross-linked with fresh EDC solution for an additional 24 h at 4°C [Figure 5-3]. Excess EDC was rinsed from the scaffolds using PBS, and the CG scaffolds were stored in PBS at 4°C until use.
Figure 5-3 Schematic of the fabrication method for CG sponges with flow networks.
109 5.2.1.3 Fabrication of PDMS Molds
PDMS molds with a recessed portion of dimensions 1 x 2 x 0.5 in were prepared by pouring uncured liquid elastomer with 10 % curing reagent (w/w, Dow Corning, MI) on top of an aluminum slab (1 x 2 x 0.5 in) placed on a weigh boat (Xia and Whitesides
1998). Air bubbles were removed from the elastomer by degassing under vacuum, and the elastomer was cured in an oven at 80°C for 4 h. The cured PDMS was then cut and removed from the weight boat. This led to a PDMS mold with recessed portion for skin tissue culture. Two of such molds were fabricated. One of them was used in the submerged culture mode. A double-sided adhesive (FT 3012, Avery Dennison, OH), cut to the size of the PDMS mold (1 x 2 in), was attached to the recessed surface of the mold with the backing layer in place. The other mold was used in the air-liquid culture mode.
For the latter, holes for inlet and outlet ports that correspond to the ports in the bioreactor were punched out using a biopsy punch. Another double-sided, cut to the size of the
PDMS mold (1 x 2 in) and punched two holes that correspond to the holes in the PDMS mold, was attached to the recessed surface of the mold with the backing layer in place.
5.2.1.4 Submerged Culture
KSM, KPM and ALIM were prepared using the method as described in Chapter 3,
Section 3.2.4.1. Human keratinocytes were isolated using the method as described in
Chapter 3, Section 3.2.4.2. The CG scaffold with integrated flow networks prepared as per procedure in Section 5.2.1.2 was sterilized using a 1X antibiotic-antimycotic PBS solution (Invitrogen) at 4°C overnight and then washed thoroughly with PBS. Prior to seeding cells, the CG scaffold was treated with either 100 µg/mL human fibronectin
(Invitrogen) or 100 µg/mL type IV collagen (Sigma) for 2 h at 37°C. The surface with
110 open network flow channels of CG scaffolds was air dried in the biosafety cabinet, and subsequently placed to the adhesive in the PDMS mold (prepared for submerged culture mode as per the procedure described above) for submerged culture [Figure 5-4].
Keratinocytes were seeded at densities of 1, 2.5, or 5 " 105 cells/cm2 on top of the CG scaffold using 2 mL of the KSM and incubated for 1 day. KSM was then replaced by
KPM for an additional 2 days.
Figure 5-4 (a) Digital picture of the patterned CG scaffold in the PDMS mold bonded by silicone adhesive. (b) Micro-CT image of the flow networks.
5.2.1.5 Air-liquid Interface Culture
The PDMS mold prepared for air-liquid interface culture as per procedure described in
Section 5.2.1.3 was attached to the bioreactor by connecting the ports that protruded from the reactor surface with the corresponding holes in the PDMS mold. The holes were slightly smaller than the outside diameter of the ports and thus formed a liquid tight seal.
111 The bioreactor with the PDMS mold and the associated components, and silicone tubing with Luer connectors were all steam sterilized using an autoclave (Tuttnauer, NY).
After 3 days of submerged culture (1 day in KSM and 2 days in KPM), the CG scaffold was removed from the submerged culture PDMS mold using a sterile forceps, dried the surface of the flow network using sterile wipes, and attached to the PDMS mold inside the bioreactor by carefully aligning the scaffold with the mold. The air-liquid interface medium was manually injected into the flow network through one of the PVDF ports using a syringe with silicone tubing and Luer connecter. Once medium occupied the flow network, the Luer connecter was disconnected from the PVDF port and reconnected to the other PVDF port. The CG scaffold was slightly lifted from the connected port side using sterile forceps to release air when feeding medium. After the release of trapped air, a continuous flow could be obtained by injecting medium through the network. ALIM was added to the bioreactor in the space between the Delrin wall and the PDMS mold to prevent the CG scaffold from drying. The bioreactor was then sealed with the silicone gasket, FEP membrane, and aluminum frame. The entire assembly was placed in an incubator at 37°C and at 5% CO2 in humidified air environment. TE skin substitutes were cultured in the air-liquid interface for 14 days. Perfusion of ALIM to the flow networks was accomplished by using a syringe pump described in Section 4.2. The flow loop was open with the outlet media collected in a centrifuge tube and not recycled. Medium perfusion rates of 100, 500, and 1000 µL/hr were used to investigate epidermis formation in vitro.
112 5.2.2 Assessment of Cell Survival on CG Scaffolds
After submerged cultures, effects of cell density and surface modification on short-term cell survival were investigated using a fluorescent cell staining technique (Vybrant®
CFDA SE cell tracer kit, Invitrogen) For these purposes, keratinocytes were stained using
CFDA SE before seeding to CG scaffolds. First, 500 µg of CFDA SE was dissolved in 90
µL of DMSO (both include in the kit) to obtain 10 mM CFDA SE stock solution. 10 µM
CFDA SE working solution was then obtained by adding 5 µL of stock solution to 5 mL of PBS (Invitrogen) just prior to use.
After obtaining confluent keratinocytes in a T-150 flask, the flask was washed with PBS twice. 10 mL of 0.05% trypsin (Invitrogen) was added to the flask and placed in an incubator at 37°C and at 5% CO2 in humidified air environment for 5 min. After 5 min,
10 mL of bovine serum (Invitrogen) was added to terminate the trypsin reaction. The solution, containing 10 mL of trypsin and 10 mL of bovine serum with cells, was transferred to a 50 mL centrifugal tube (Falcon) and centrifuged to obtain a cell pellet.
The supernatant was aspirate and the cells were resuspended gently in 5 mL prewarmed
CFDA SE working solution. The centrifugal tube was placed in an incubator at 37°C and at 5% CO2 in humidified air environment for 5 min. After 5 min, the cells were centrifuged, resuspended in KSM, and seeded on the unmodified CG scaffold, CG scaffold modified with 500 µL of type IV collagen (100 µg/mL), and CG scaffold modified with 500 µL of fibronectin (100 µg/mL) at different seeding densities (1, 2.5, or
5 " 105 cells/cm2). The cells were then cultured under submerged culture mode for 3 days
113 as per procedure described in Section 5.2.1.1 for 3 days. At day 3, samples were collected and imaged in a fluorescent microscope (Olympus).
5.2.3 Histological Analysis
At day 14, samples were collected for histological and immunohistohemical analysis. TE skin substitutes were fixed with 4% buffered-paraformaldehyde solution (Ted Pella, Inc.,
Redding CA) overnight at 4°C. Before general paraffin embedding, the samples were pre-embedded with 3% agar (Sigma). Agar has been used for embedding tissues prior to paraffin processing when fixed tissues need to be arranged in a particular orientation
(Jones and Calabresi 2007). Based on this technique, we pre-embedded the samples prior to paraffin processing to prevent epidermis coming off from the dermis during dehydration. They were then embedded in paraffin, sectioned (8 µm thick), stained with hematoxylin and eosin (H&E), and examined using a bright-field microscope as described in Section 3.2.4.4.
5.2.4 Immunohistochemical Analysis
For immunostaining, TE skin substitutes were fixed, paraffin embedded, and sectioned as described above. The sections were baked at 60°C and rehydrated as follows: they were treated with xylene three times for 5 min each, 100% ethanol twice for 3 min each, 95% ethanol twice for 3 min each, 70% ethanol for 3 min, and water for 5 min followed by rinsing in PBS three times for 10 min each. Individual samples in the sections were isolated for staining by using a Pap pen (Research Products International Corp., Mount
Prospect, IL). After rehydration, the sections were blocked using 10% donkey serum
(Santa Cruz Biotechnology, Inc., Santa Cruz, CA) in PBS for 30 min at room temperature.
114 The primary antibody (cytokeratin 5 (CK5) and 14 (CK14), Novus Biologicals, Littleton,
CO) was applied at 1:400 dilution for CK5 and 1:50 for CK14 and the sections were incubated for 2 h at room temperature. After rinsing in PBS twice for 10 min each, a fluorescein isothiocyanate (FITC)-conjugated donkey anti-mouse IgG secondary antibody (Novus Biologicals) was used at 1:500 dilution for 45 min at room temperature.
After rinsing in PBS twice for 10 min each, 4’, 6-diamidino-2-phenylindole (DAPI) was applied at 1:10000 dilution for 5 min at room temperature. Stained sections were mounted in 5% n-propyl gallate in glycerol and imaged. Imaging of the sections were carried in an upright fluorescent microscope (Leica).
5.2.5 Cell Viability
At day 7 after air-liquid interface culture, the TE skin substitutes were removed from the bioreactors for cell viability tests using a standard MTT assay (n = 2 for each condition)
(Mosmann 1983). The MTT solution, containing 0.5 mg/mL (3-(4,5-dimethylthiazol-2- yl)-2,5-diphenyltetrazolium bromide) (Invitrogen), was prepared in PBS freshly and sterilized using a syringe filter. TE skin substitute samples at different locations (inlet, center, outlet, and outside of the flow network) were obtained by using a 6 mm diameter biopsy punch. The individual samples were then placed in a 12-well plate, and incubated with 3 mL of sterile MTT solution at 37°C in a 5% CO2 containing humidified air environment for 4 h. After incubation, the MTT solution was aspirated and the samples were incubated with 1 mL of 2-methoxyethanol (Sigma) at room temperature for 3 h with mixing. The amount of MTT-formazan in the solution was determined by measuring the absorbance at 570 nm with a spectrophotometer (SpectraMax M2, Molecular Devices,
Sunnyvale, CA).
115 5.3 Results
5.3.1 Micropatterning on CG Scaffolds
Figure 5-5 shows SEM images of the surface of the flow networks cast onto CG scaffolds and a corresponding AutoCAD design of flow networks. The flow networks [Figure 5-5
(a)] and an island [Figure 5-5 (b)] have sharp edges, and the original design [Figure 5-5
(c)] is reproduced on CG scaffolds.
Figure 5-5 SEM images of flow networks cast onto (a) CG scaffolds using modified- micropatterning technique. (b) SEM image of the smallest “island” on CG scaffolds. (C) Corresponding AutoCAD design of flow networks
5.3.2 The Effect of Seeding Density on Short-term Cell Survival
We first investigated the effect of seeding density on short-term cell survival on CG scaffolds using a fluorescent dye, CFDA-SE. CFDA-SE is a cell permeable dye generally
116 used for cell proliferation research. CFDA-SE enters cells by diffusion and is cleaved by intracellular esterases to yield highly fluorescent, amine-reactive carboxyfluorescein succinimidyl ester, which reacts with intracellular amines and forms fluorescent conjugates. The dye level is maintained relatively stable in non-dividing cells and is not transferred to adjacent cells in a population. If a stained cell divides, the dye level is inherited equally between two daughter cells, resulting in both daughter cells having a
CFDA-SE concentration approximately 50% of that of the mother cell. A cell stained with CFDA-SE can be detected following up to 8 successive cell divisions (Lyons and
Parish 1994).
Figure 5-6 (a) and (c) illustrate cell survival at seeding density of 1 " 105 cells/cm2 at day
1 and 3, respectively, and (b) and (d) represent the proliferation at seeding density of 5 "
105 cells/cm2 at day 1 and 3. The results show that keratinocytes identified as fluorescent cells reached subconfluence at day 3 with seeding densities of 1 " 105 cells/cm2, while keratinocytes reached about 90% confluence at day 1 with seeding densities of 5 " 105 cells/cm2.
117
Figure 5-6 Fluorescent images of TE skin showing keratinocytes proliferation with seeding densities of 1 X 105 cells/cm2 at (a) day 1 and (b) day 3, and 5 X 105 cells/cm2 at (c) day 1 and (d) day 3. Scale bar is 100 µm.
5.3.3 The Effect of Surface Modification on Short-term Cell Survival
Researchers have shown that type IV collagen, a major component in basement membrane, and fibronectin may regulate keratinocyte attachment, migration, proliferation, and differentiation (Bush and Pins 2010; Segal et al. 2008). As a result, we anticipated that surface modification with type IV collagen or fibronectin would improve keratinocyte survival. Figure 5-7 shows fluorescent images of keratinocyte proliferation
118 at day 1 and 3 using unmodified CG scaffolds [(a) and (b)] and type IV collagen modified
CG scaffolds [(c) and (d)]. The results suggested that type IV collagen enhanced keratinocyte attachment and proliferation, showing confluence after culture 3 days. On the other hand, histology of TE skin substitutes results revealed that keratinocyte differentiation was improved in the presence of fibronectin but not type IV collagen (data not shown). Fibronectin was therefore chosen to treat CG scaffolds for following experiments.
119
Figure 5-7 Fluorescent images of keratinocyte proliferation on unmodified CG scaffolds at (a) day 1 and (b) day 3, and type IV collagen modified CG scaffolds at (c) day 1 and (d) day 3. Scale bar is 100 µm.
5.3.4 The Effect of Flow Rate on Epidermis Formation
Images of H&E stained sections of TE skin substitutes after 14 days of air-liquid interface culture in static and perfusion culture at 100, 500, and 1000 µL/hr are shown in
Figure 5-8 (a), (b), (c), and (d), respectively. TE skin substitutes cultured under static conditions showed a well-stratified epidermis, containing partially- organized stratum basale but well-organized stratum spinosum, granulosum and corneum. TE skin
120 substitutes cultured at 100 µL/hr perfusion rate [Figure 5-8 (b)] also showed a well- stratified epidermis, containing well-organized stratum basale, spinosum, granulosum, and corneum. The sublayers were thicker in perfusion culture conditions. The difference between static and perfusion culture conditions was more notable for 1000 µL/hr, which led to highly-organized thicker sublayers. At 500 µL/hr perfusion rate [Figure 5-8 (c)], however, slightly disorganized epidermis was obtained with a thin but clear stratum corneum.
Figure 5-8 H&E stained sections of TE skin substitutes after 14 days of air-liquid interface culture in (a) static and perfusion culture at (b) 100, (c) 500, and (d) 1000 µL/hr
121 perfusion rate. Well-stratified epidermis, containing well-organized stratum basale, spinosum, granulosum, and corneum, were obtained at 100 and 1000 µL/hr perfusion rates and static cultures. Scale bar is 25 µm.
Figure 5-9 illustrated H&E stained sections of TE skin substitutes cultured at 100 and
1000 µL/h perfusion rates for 14 days. For TE skin substitutes cultured at 100 µL/h perfusion rate [Figure 5-9 (b)], the results show that the sections of scaffolds that were exposed to internal medium flow contained well-differentiated epidermis [(B) in Figure
5-9 (b)], whereas the sections of scaffolds that were not exposed to flow had no cells [(C) in Figure 5-9 (b)]. For TE skin substitutes cultured at 1000 µL/h perfusion rate [Figure 5-
9 (c)], the results show that both sections of scaffolds that were exposed [(B) in Figure 5-
9 (c)] and not exposed [(C) in Figure 5-9 (b)] to internal medium flow contained well- differentiated epidermis. This demonstrates clearly the effect of flow was necessary to sustain epidermal differentiation. This internal control is also the best way to test the efficacy of the integrated flow networks.
122
123
Figure 5-9 (a) Schematics of CG scaffold with flow networks. (A) and (B) are top view schematics and (C) is a cross-sectional schematic of dashed line in (B). H&E stained sections of TE skin substitutes cultured at (b) 100 µL/hr and (c) 1000 µL/hr perfusion rates for 14 days. (A) shows an image of H&E stained section of TE skin substitute that was not exposed to internal medium flow. (B) shows an image of H&E stained section of TE skin substitute that was exposed to internal medium flow. Scale bar is 25 µm.
124 Figure 5-10 shows DAPI stained sections of TE skin substitutes after 14 days of air-liquid interface culture in static (a) and perfusion culture at 100 (b), 500 (c), and 1000 (d) µL/hr perfusion rate conditions. Number of cells in TE skin substitutes cultured at 1000 µL/hr perfusion rate [Figure 5-10 (d)] was significantly greater than that of other conditions, while cell number of TE skin substitutes cultured at 100 µL/hr perfusion rate [Figure 5-
10 (b)] was comparable with that of control [Figure 5-10 (b)]. At 500 µL/hr perfusion rate [Figure 5-10 (c)], scattered nuclei was observed with the lowest cell number.
Figure 5-10 DAPI stained sections of TE skin substitutes after 14 days of air-liquid interface culture in (a) static and perfusion culture at (b) 100, (c) 500, and (d) 1000 µL/hr. Scale bar is 100 µm.
125 Figure 5-11 shows immunohistochemical staining results. Images of CK5 [left panel,
Figure 5-11 (a, c, e, g)] and CK14 [right panel, Figure 5-11 (b, d, f, h], two basal proliferative cell markers, -stained sections of TE skin substitutes after 14 days of air- liquid interface culture in static [Figure 5-11 (a), (b)] and perfusion culture at 100 [Figure
5-11 (c), (d)], 500 [Figure 5-11 (e), (f)], and 1000 [Figure 5-11 (g), (h)] µL/hr perfusion rate conditions are shown. Both CK5 and CK14 were expressed in every condition, suggesting that basal cells were present in every condition.
126
Figure 5-11 Immunostaining of TE skin substitutes after 14 days of air-liquid interface culture for cell nuclei (blue), CK5 (red) (a, c, e, g), and CK14 (red) (b, d, f, h) in (a) and (b) static and perfusion culture at (c) and (d) 100, (e) and (f) 500, and (g) and (h) 1000 µL/hr perfusion rate. Dashed line indicates the dermo-epidermal junction. Scale bar is 25 µm
127 5.3.5 Effect of Flow Rate on Keratinocyte Viability
Figure 5-12 shows MTT assay results of TE skin substitutes obtained at day 7 during air- liquid interface culture. The results showed that viability increased with perfusion rates.
More importantly, viability of TE skin substitutes cultured in the perfusion bioreactors was significantly greater (about 4-6 times) than those cultured in static conditions.
However, it should be mentioned that the results were not statistical significant due to insufficient data (n=2).
Figure 5-12 Effect of perfusion rate on viability of TE skin substitutes cultured for 7 days (n=2 each). (a) Absorbance values of samples normalized to that of static culture as a function of perfusion rate. (b) Absorbance values of samples obtained in locations of flow normalized to that of samples obtained from same TE skin substitute but at locations with no flow networks. Error bars represent SEM.
5.4 Discussion
The aim of this study is to investigate the efficacy of integrated flow networks on epidermal differentiation in TE skin substitutes. Previous studies have suggested that pre-
128 existing vasculature could lead to improved function of TE products in vivo. In this study, we developed an in vitro model of TE skin substitutes with integrated flow networks with optimal transport efficiency, and investigated the effect of flow networks on the growth and differentiation of keratinocytes. The integrated flow networks were designed to have maximum mass transport characteristics (Janakiraman et al. 2007b). For a single generation bifurcating network used in our studies, the network has the maximum transport area per unit volume while the pressure drop is kept at a much lower value than the physiological pressure drop for a comparable tissue volume in vivo. Thus the network used in the system is expected to yield maximum nutrient transport and lead to a TE skin substitute with improved function. Our results indicated that keratinocytes differentiation in the epidermis was improved when using TE skin substitutes with an integrated flow network. Increasing flow rates resulted in thicker and better-stratified epidermis, along with well-organized stratum basal (except 500 µL/hr perfusion rate) [Figure 5-8]. Very importantly, in our system, we cultured TE skin substitutes for only 14 days. The results, viz., the thickness and stratification of the epidermis are better than those cultured for 21 days by other researchers (Ebersole et al. 2010; Kalyanaraman et al. 2008; Sun et al.
2005). This demonstrates not only the biological efficacy of the integrated flow networks but also the accelerated tissue formation in the TE skin substitutes. The latter is a critical hurdle in the application of tissue engineering to solve clinical problems.
Seeding density is an important factor guiding keratinocyte differentiation. Poumay et al. have shown that development of intimate cell-cell contacts and probably cell-matrix or other cellular interactions associated with keratinocyte confluence in monolayer culture serves as a crucial signal to induce terminal differentiation of kerationcytes (Poumay and
129 Pittelkow 1995). We assessed the effect of seeding density on keratinocyte proliferation after submerge culture as the first step in this study. We chose seeding density values of 1,
2.5, or 5 " 105 cells/cm2, as these are used by other researchers. The results showed that confluence was not obtained with density of 1 " 105 cells/cm2, while 5 " 105 cells/cm2 formed confluence at day 1. A seeding density of 2.5 " 105 cells/cm2 led to confluence at day 3 (beginning of air-liquid interface culture) and was chosen for subsequent experiments. Histology and immohistochemistry results from TE skin substitutes cultured using this density confirmed our choice; a stratified epidermis was formed when using this density.
In general, keratinocytes in the stratum basale express CK5 and CK14, two basal proliferative cell markers. A well-differentiated epidermis should contain a uniform layer of stratum basale overlaying on the dermis. As shown in Figure 5-11, CK5 and CK14 were both expressed in TE skin substitutes obtained from static and perfusion culture, suggesting that basal cells were present in every condition. However, CK5 and CK14 were also shown in other layers except stratum corneum. This is probably due to two reasons. First, at the beginning of epidermis formation, proliferative keratinocytes, which express CK5 and CK14, form a confluent layer on the dermal analog (stratum basale).
They then migrate upward and start to differentiate to form stratum spinosum and granulosum. However, some proliferative keratinocytes may not be triggered to differentiate, resulting in expression of CK5 and CK14 in the stratum spinosum and granulosum. Second, epidermis formation in vivo generally takes around 4 weeks, while it only takes 14 days in our in vitro experiment. As a result, hyperproliferation of keratinocytes may occur in TE skin substitutes, resulting in CK5 and CK14 left in
130 spinous and granular cells. To localize differentiated keratinocytes, other markers such as
CK1 and CK10, which are only expressed in differentiated keratinocytes, should be used to perform immunohistochemistry.
In this study, MTT was used to assess cell viabilities of TE skin substitutes obtained from static culture and perfusion culture. MTT, a yellow water soluble tetrazolium salt (3-(4,
5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide), has been widely used as a means of measuring cell viability (Mosmann 1983). In general, succinate dehydrogenase system of active mitochondrial in live cells reduces MTT to a water-insoluble purple formazan, which can be dissolved using a suitable solvent and quantified spectrophotometrically. The amount of formazan is proportional to the number of metabolically active cells. As shown in Figure 5-12 (a), cell viabilities were increased when TE skin substitutes were cultured in the perfusion bioreactor. TE skin substitutes cultured at 1000 µL/hr perfusion rate had the highest viability, while TE skin substitutes cultured at 100 and 500 µL/hr perfusion rates had similar viabilities. Moreover, cell viabilities of TE skin substitutes obtained from inlet, center, and outlet were superior than that of TE skin substitutes obtained from outside of the flow network [Figure 5-12 (b)], which clearly shows the effect of flow on epidermal differentiation. Corresponding to histology results, the effect of flow was necessary to sustain epidermal differentiation.
Perfusion culture of constructs is designed to enhance mass transport within TE constructs, which has been shown to improve the quality of TE constructs (Goldstein et al. 2001). Perfusion can also be used to achieve homogenous development of tissues in vitro and has been proposed as a means to overcome size constraint (Grayson et al. 2008;
Martin et al. 2009). During the air-liquid interface static culture of TE skin substitutes
131 except for cells near the edge of the scaffolds, the primary source of nutrients for the cells is the bulk fluid that is located outside the dermis. The nutrient molecule has to diffuse through the dermal layer to reach the cells. If the thickness of the dermal layer reaches a threshold value, limited nutrient diffusion between cells in the epidermal substitutes and medium would lead to either poorly differentiated epidermis or premature differentiation followed by death (MacNeil 2007). In static culture of skin TE substitutes, it has been shown that the depletion of nutrients and accumulation of toxic waste are both associated with poor proliferation and differentiation. As a result, it is widely accepted that the perfusion culture would provide a better culture environment than the static culture; which is corresponded to our results. At 100 [Figure 5-8 (b)] and 1000 µL/hr perfusion rates [Figure 5-8 (d)], a well-stratified epidermis was formed, along with well-organized stratum basale, spinosum, granulosum, and corneum. Increasing perfusion rates resulted in thicker epidermis, and anatomy was comparable with static culture. Moreover, for TE skin substitutes cultured at 100 µL/hr perfusion rate [Figure 5-9 (b)], a well-stratified epidermis was obtained when the scaffold was exposed to internal medium flow [(B) in
Figure 5-9 (b)], while immature or no epidermis was obtained when the scaffold was not exposed to internal medium flow [(A) in Figure 5-9 (b)]. When increasing perfusion rate to 1000 µL/hr [Figure 5-9 (c)], a well-stratified epidermis was obtained in whole scaffold, including exposed [(B) Figure 5-9 (c)] and unexposed [(C) Figure 5-9 (c)] scaffold.
Corresponding to the simulation results obtained in Chapter 4, Section 4.3, histology results clearly demonstrate that increasing perfusion rates leads to higher glucose concentrations in the middle of flow networks, which can diffuse to the scaffold that is not exposed to internal medium flow and further result in a well-stratified epidermis.
132 Besides mass transport, perfusion also has important biological roles. Shear stress, for example, is induced by flow and plays a vital role in both intracellular and cellular levels.
While in vivo the endothelial cells are the predominant cell type that is exposed to fluid shear, other cell types such osteoblasts in bone, and tenocytes in tendon also undergo shear-based deformation. In these cases, it has been shown that optimal levels of shear stress improve cell proliferation, differentiation, and adhesion (Dan et al. 2010). In the case of TE bone, the shear stress accelerates osteogenic differentiation and increases mineralized matrix deposition in a dose-depend manner (Li et al. 2009; Sikavitsas et al.
2003). Interestingly, Cartmell et al. have shown that although the flow may benefit the development of TE bone, higher flow rates was detrimental to the viability of cells due to flow-induced shear stress (Cartmell et al. 2003). In our system, however, the direct effect of flow-induced shear stress is minimal on the cells; porous scaffold material separated the cells from flow. As a result, a wall shear stress in the flow network did not detrimentally affect the growth of the epidermal substitutes.
Although the perfusion bioreactors have been well developed for several TE constructs
(Bancroft et al. 2003; Pazzano et al. 2000), there have been only a few very designs for
TE skin substitutes. This is possible due to the unique culture condition for TE skin substitutes; an air-liquid interface culture is commonly used for inducing keratinocyte differentiation (Parenteau et al. 1992; Prunieras et al. 1983). Submerged culture mode leads to only partially stratified epithelium with approximately three cell layers thick, including a basal cell layer, a suprabasal cell layer, and a squamous layer. On the other hand, all sub-layers are formed after exposing kerationcytes to air. In bioreactor systems of Sun et al. (Sun et al. 2005) and Kalyanaraman et al. (Kalyanaraman et al. 2008) to
133 culture TE skin substitutes, since the medium was pumped directly to the system containing TE skin substitute on a platform, any small changes in medium flow rates can cause the location of the air-liquid interface to change and can negatively affect the quality of differentiation. This is especially important when high flow rates are used which can directly expose the epidermis to flow. Moreover, autocrine/paracrine cell signals can be flushed away by flow before they exert influence. The problem of dilution of beneficial factors may also explain why the cell viability and keratinocyte proliferation were all diminished at the highest flow rate in the above studies. In our design, there is no direct exposure of epidermis to flow. As there is a thin layer of scaffold between the flow and the cells, even at high flow rates, autocrine/paracrine factors released by cells will have a chance to exert influence on neighboring cells.
Studies have shown that the presence of fibroblasts accelerated wound healing by regulation of matrix deposition including type I, type IV collagen and elastin (Ayer 1964;
Demarchez et al. 1992; Layman et al. 1971; Marks et al. 1991) and epidermal and dermal regeneration (El-Ghalbzouri et al. 2002; Murphy et al. 1990). Similarity, dermal fibroblasts play a central role in regeneration of TE skin substitutes in vitro.
Keratinocytes poorly attached to the basement membrane and were disorganized in the absence of fibroblasts (Ralston et al. 1997). El-Ghalbzouri et al. showed that in the absence of fibroblasts, partially stratified epidermis with only three or four viable cell layers was formed, while keratinocyte proliferation and differentiation were all improved in the presence of fibroblasts (El-Ghalbzouri et al. 2002). Erdag et al. demonstrated that thicker well-stratified epidermis was formed in the presence of fibroblasts (Erdag and
Sheridan 2004). In our system, even though fibroblasts were not used, a thick well-
134 stratified epidermis was still formed. This could be due to the conjugation between EDC and fibronectin; which was demonstrated to enhance epidermis thickness and keratinocyte proliferation compared to unmodified dermal substitutes (Bush and Pins
2010). Indeed, our results show that partially stratified epidermis was obtained when using type IV collagen (data not shown), a major protein in the basement membrane and was found to improve keratinocyte differentiation when using type I collagen gel inosculated with human fibroblasts (Segal et al. 2008), modified CG scaffolds. In addition, the concentrated CG scaffolds, which were developed in Chapter 3 (Liang et al.
2010), may also improve keratinocyte proliferation and differentiation. One of the mechanisms through which fibroblasts improve keratinocyte proliferation and differentiation is by the secretion of ECM, which supports keratinocyte growth. Although fibroblasts were absent in our system, the concentrated CG scaffolds, which contain three times as much collagen as that in conventional unconcentrated scaffolds, could provide enough collagen to support growth and differentiation of keratinocytes. Regardless of the mechanism, our results indicate that fibroblasts are not necessary to achieve a well- differentiated epidermis in composite TE skin substitutes. This simplifies the culture system for TE skin substitutes considerably.
5.5 Conclusions
In summary, we have developed an in vitro model of TE skin substitute with integrated flow networks in a perfusion bioreactor. The effect of flow networks and flow rate on the epidermis formation and keratinocyte proliferation were assessed. We show that appropriate seeding density and surface modification enhanced keratinocyte attachment and proliferation. Under optimal conditions of perfusion, we show that TE skin
135 substitutes cultured in the perfusion bioreactor were biologically and morphologically superior to skin substitutes cultured in the static condition. Increasing flow rates led to thicker and well-organized epidermis. Overall, we successfully demonstrate the efficacy of integrated convective flow networks on the growth of TE skin substitutes.
Endothelialization of these flow networks can lead to prevascularized TE skin substitutes and will pave the way for translational research.
5.6 References
Arano T, Sato T, Matsuzaka K, Ikada Y, Yoshinari M. 2010. Osteoblastic cell proliferation with uniform distribution in a large scaffold using radial-flow bioreactor. Tissue Engineering Part C-Methods 16(6):1387-1398.
Ayer JP. 1964. Elastic tissue. In: Hall DA, editor. International Review of Connective Tissue Research. New York: Academic Press. p 33-100.
Bancroft GN, Sikavitsas VI, Mikos AG. 2003. Design of a flow perfusion bioreactor system for bone tissue-engineering applications. Tissue Engineering 9(3):549-554.
Boyce ST. 1996. Cultured skin substitutes: A review. Tissue Engineering 2(4):255-266.
Bush KA, Pins GD. 2010. Carbodiimide conjugation of fibronectin on collagen basal lamina analogs enhances cellular binding domains and epithelialization. Tissue Engineering Part A 16(3):829-838.
Cartmell SH, Porter BD, García AJ, Guldberg RE. 2003. Effects of medium perfusion rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Engineering 9(6):1197-1203.
Dan L, Chua C-K, Leong K-F. 2010. Fibroblast response to interstitial flow: A state-of- the-art review. Biotechnology and Bioengineering 107(1):1-10.
Demarchez M, Hartmann DJ, Regnier M, Asselineau D. 1992. The role of fibroblasts in dermal vascularization and remodeling of reconstructed human skin after transplantation onto the nude mouse. Transplantation 54(2):317-326.
Ebersole GC, Anderson PM, Powell HM. 2010. Epidermal differentiation governs engineered skin biomechanics. Journal of Biomechanics 43(16):3183-3190.
136 El-Ghalbzouri A, Gibbs S, Lamme E, Van Blitterswijk CA, Ponec M. 2002. Effect of fibroblasts on epidermal regeneration. British Journal of Dermatology 147(2):230-243.
Erdag G, Sheridan RL. 2004. Fibroblasts improve performance of cultured composite skin substitutes on athymic mice. Burns 30(4):322-328.
Freyman TM, Yannas IV, Gibson LJ. 2001. Cellular materials as porous scaffolds for tissue engineering. Progress in Materials Science 46(3-4):273-282.
Gibot L, Galbraith T, Huot J, Auger FA. 2010. A preexisting microvascular network benefits in vivo revascularization of a microvascularized tissue-engineered skin substitute. Tissue Engineering Part A 16(10):3199-3206.
Goldstein AS, Juarez TM, Helmke CD, Gustin MC, Mikos AG. 2001. Effect of convection on osteoblastic cell growth and function in biodegradable polymer foam scaffolds. Biomaterials 22(11):1279-1288.
Grayson ACR, Shawgo RS, Johnson AM, Flynn NT, Yawen LI, Cima MJ, Langer R. 2004. A BioMEMS review: MEMS technology for physiologically integrated devices. Proceedings of the IEEE 92(1):6-21.
Grayson WL, Bhumiratana S, Cannizzaro C, Chao PHG, Lennon DP, Caplan AI, Vunjak- Novakovic G. 2008. Effects of initial seeding density and fluid perfusion rate on formation of tissue-engineered bone. Tissue Engineering Part A 14(11):1809- 1820.
Grayson WL, Fröhlich M, Yeager K, Bhumiratana S, Chan ME, Cannizzaro C, Wan LQ, Liu XS, Guo XE, Vunjak-Novakovic G. 2009. Engineering anatomically shaped human bone grafts. Proceedings of the National Academy of Sciences.
Grayson WL, Marolt D, Bhumiratana S, Fröhlich M, Guo XE, Vunjak-Novakovic G. 2010. Optimizing the medium perfusion rate in bone tissue engineering bioreactors. Biotechnology and Bioengineering:n/a-n/a.
Janakiraman V, Kienitz B, Baskaran H. 2007a. Lithography technique for topographical micropatterning of collagen-glycosaminoglycan membranes for tissue engineering applications. Journal of Medical Devices 1(3):233-7.
Janakiraman V, Mathur K, Baskaran H. 2007b. Optimal planar flow network designs for tissue engineered constructs with built-in vasculature. Annals of Biomedical Engineering 35(3):337-347.
Jones MV, Calabresi PA. 2007. Agar-gelatin for embedding tissues prior to paraffin processing. Biotechniques 42(5):569-570.
137 Kalyanaraman B, Supp DM, Boyce ST. 2008. Medium flow rate regulates viability and barrier function of engineered skin substitutes in perfusion culture. Tissue Engineering Part A 14(5):583-593.
Layman DL, McGoodwin EB, Martin GR. 1971. The nature of the collagen synthesized by cultured human fibroblasts. Proceedings of the National Academy of Sciences of the United States of America 68(2):454-458.
Lee CR, Grodzinsky AJ, Spector M. 2001. The effects of cross-linking of collagen- glycosaminoglycan scaffolds on compressive stiffness, chondrocyte-mediated contraction, proliferation and biosynthesis. Biomaterials 22(23):3145-3154.
Li D, Tang T, Lu J, Dai K. 2009. Effects of flow shear stress and mass transport on the construction of a large-scale tissue-engineered bone in a perfusion bioreactor. Tissue Engineering Part A 15(10):2773-2783.
Liang W-H, Kienitz BL, Penick KJ, Welter JF, Zawodzinski TA, Baskaran H. 2010. Concentrated collagen-chondroitin sulfate scaffolds for tissue engineering applications. Journal of Biomedical Materials Research Part A 94A(4):1050-1060.
Lyons AB, Parish CR. 1994. Determination of lymphocyte division by flow cytometry. Journal of Immunological Methods 171(1):131-137.
MacNeil S. 2007. Progress and opportunities for tissue-engineered skin. Nature 445(7130):874-880.
Marks MG, Doillon C, Silver FH. 1991. Effects of fibroblasts and basic fibroblast growth factor on facilitation of dermal wound healing by type I collagen matrices. Journal of Biomedical Materials Research 25(5):683-696.
Martin I, Smith T, Wendt D. 2009. Bioreactor-based roadmap for the translation of tissue engineering strategies into clinical products. Trends in Biotechnology 27(9):495- 502.
Minuth WW, Schumacher K, Strehl R, Kloth S. 2000. Physiological and cell biological aspects of perfusion culture technique employed to generate differentiated tissues for long term biomaterial testing and tissue engineering. Journal of Biomaterials Science-Polymer Edition 11(5):495-522.
Mosmann T. 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65(1- 2):55-63.
Murphy GF, Orgill DP, Yannas IV. 1990. Partial dermal regeneration is induced by biodegradable collagen glycosaminoglycan grafts. Laboratory Investigation 62(3):305-313.
138 Parenteau NL, Bilbo P, Nolte CJM, Mason VS, Rosenberg M. 1992. The organotypic culture of human skin keratinocytes and fibroblasts to achieve form and function. Cytotechnology 9(1-3):163-171.
Pathi P, Ma T, Locke BR. 2005. Role of nutrient supply on cell growth in bioreactor design for tissue engineering of hematopoietic cells. Biotechnology and Bioengineering 89(7):743-758.
Pazzano D, Mercier KA, Moran JM, Fong SS, DiBiasio DD, Rulfs JX, Kohles SS, Bonassar LJ. 2000. Comparison of chondrogensis in static and perfused bioreactor culture. Biotechnology Progress 16(5):893-896.
Poumay Y, Pittelkow MR. 1995. Cell density and culture factors regulate keratinocyte commitment to differentiation and expression of suprabasal K1/K10 keratin. Journal of Investigative Dermatology 104(2):271-276.
Prunieras M, Regnier M, Woodley D. 1983. Methods for cultivation of keratinocytes with an air-liquid interface. Journal of Investigative Dermatology 81(s1):28s-33s.
Ralston DR, Laytont C, Dalley AJ, Boyce SG, Freedlander E, MacNeil S. 1997. Keratinocytes contract human dermal extracellular matrix and reduce soluble fibronectin production by fibroblasts in a skin composite model. British Journal of Plastic Surgery 50(6):408-415.
Segal N, Andriani F, Pfeiffer L, Kamath P, Lin N, Satyamurthy K, Egles C, Garlick JA. 2008. The basement membrane microenvironment directs the normalization and survival of bioengineered human skin equivalents. Matrix Biology 27(3):163-170.
Shi H, Han C, Mao Z, Ma L, Gao C. 2008. Enhanced angiogenesis in porous collagen- chitosan scaffolds loaded with angiogenin. Tissue Engineering Part A 14(11):1775-1785.
Shin M, Matsuda K, Ishii O, Terai H, Kaazempur-Mofrad M, Borenstein J, Detmar M, Vacanti JP. 2004. Endothelialized networks with a vascular geometry in microfabricated poly(dimethyl siloxane). Biomedical Microdevices 6(4):269-278.
Sikavitsas VI, Bancroft GN, Holtorf HL, Jansen JA, Mikos AG. 2003. Mineralized matrix deposition by marrow stromal osteoblasts in 3D perfusion culture increases with increasing fluid shear forces. Proceedings of the National Academy of Sciences of the United States of America 100(25):14683-14688.
Sun T, Norton D, Haycock JW, Ryan AJ, MacNeil S. 2005. Development of a closed bioreactor system for culture of tissue-engineered skin at an air-liquid interface. Tissue Engineering 11(11-12):1824-1831.
Supp DM, Supp AP, Bell SM, Boyce ST. 2000. Enhanced vascularization of cultured skin substitutes genetically modified to overexpress vascular endothelial growth factor. Journal of Investigative Dermatology 114(1):5-13.
139 Tremblay P-L, Hudon V, Berthod F, Germain L, Auger FA. 2005. Inosculation of tissue- engineered capillaries with the host's vasculature in a reconstructed skin transplanted on mice. American Journal of Transplantation 5(5):1002-1010.
Unger MA, Chou H-P, Thorsen T, Scherer A, Quake SR. 2000. Monolithic microfabricated valves and pumps by multilayer soft lithography. Science 288(5463):113-116.
Xia YN, Whitesides GM. 1998. Soft lithography. Annual Review of Materials Science 28:153-184.
Young DM, Greulich KM, Weier HG. 1996. Species-specific in situ hybridization with fluorochrome-labeled DNA probes to study vascularization of human skin grafts on athymic mice. Journal of Burn Care and Rehabilitation 17(4):305-310.
Zhao F, Ma T. 2005. Perfusion bioreactor system for human mesenchymal stem cell tissue engineering: Dynamic cell seeding and construct development. Biotechnology and Bioengineering 91(4):482-493.
140 Chapter 6
AN ADHESION TECHNIQUE BASED ON
ALBUMIN/GLUTARALDEHYDE BIOADHESIVES FOR
COLLAGEN SCAFFOLDS
6.1 Introduction
Collagen-based scaffolds have several applications in tissue engineering. We have shown that these scaffolds are amenable for microfabrication; micro-scale flow networks can be fabricated in such scaffolds with high precision (Janakiraman et al. 2007). These networks, however, are planar in nature. Although, currently, there is no fabrication technique that allows for true three-dimensional fabrication of defined networks in biomaterials such as collagen with micron-scale resolution, it will be greatly beneficial if multiple planar networks can be stacked to form a three-dimensional flow network. In this study, we investigated a bioadhesive for bonding CG scaffolds with each other.
A bioadhesive, defined as any substance with characteristics that allow for polymerization, must either hold tissues together or serve as a barrier to leakage. Otani et al. suggests that an ideal bioadhesive should have reasonable bonding strength to tissue and the bonding strength should not be affected by the presence of water (Otani et al.
1996a; Otani et al. 1996b). Bioadhesives should be liquid before curing and solidify immediately after applying. They should also have an appropriate viscosity before curing, so that they can be easily applied to a specific tissue site without flowing everywhere.
Finally, bioadhesives should be non-cytotoxic and be able to degrade in vivo.
141 On the basis of bioadhesives composition, they can be classified as natural, synthetic, or combined preparations and further divided into fibrin glues, cyanoacrylates, albumin- based compounds, collagen-based compounds, and hydrogels. Fibrin glues have been widely used as bioadhesives and are the most commonly used bioadhesives in plastic and reconstructive surgery (Currie et al. 2001). They have been shown to improve wound healing in various applications (Adant et al. 1993; Vedung and Hedlund 1993) and skin graft take, especially at difficult grafting sites (Vibe and Pless 1983). They have little inflammatory reaction. Moreover, the fibrin glues have been used as delivery systems for exogenous growth factors for angiogenesis (Brown et al. 1996; Koolwijk et al. 1996).
However, low bonding strength and a risk of infection due to isolation from human blood limit the use of the fibrin glues in biomedical field (Sekine et al. 2001).
Cyanoacrylates, synthesized by condensation of cyanoacetate with formaldehyde in the presence of a catalyst, are commonly used bioadhesives in medical, dental, and commercial applications (Leggat et al. 2004). However, cured tissues by using the cyanoacrylates are much stiffer than normal tissues and thus their potential as a successful bioadhesive has been limited. In addition, they have been shown to be cytotoxic due to the release of formaldehyde upon degradation (Tseng et al. 1990).
Hydrogels, composed of gelatin and poly(L-glutamic acid) (PLGA) (Otani et al. 1996a;
Otani et al. 1996b) or polyethylene glycol (PEG) (Ranger et al. 1997), are commonly used biological glues for soft tissues. However, bonding strength between the hydrogel and the host tissue is still inferior (Brigham et al. 2009). Researchers have developed hydrogels cross-linked with carbodiimides (Sung et al. 1999). The results have shown
142 that firm adherence between mouse skin and other soft tissues with a higher bonding strength occurred compared to fibrin glues. In addition, inflammatory reactions around hydrogels are mild after implantation, and hydrogels are gradually absorbed by tissues.
Collagen-based bioadhesives, made from combining collagen and thrombin, are relatively new, but have shown significant potential. Two collagen-based bioadhesives are currently approved for clinical use in the United States. They differ in delivery based on their specialty use and therefore have slightly different applicators. They both work by providing a matrix for the clot and then assist with coagulation by delivering fibrinogen to the area (Chapman et al. 2001; Oz et al. 2000; Renkens et al. 2001; Sakon et al. 1989).
However, like collagen-based biomaterial scaffolds, they degrade at relatively rapid rate.
Another group of bioadhesives is based on combining albumin with cross-linking agents.
BioGlue (CryoLife, Inc., Kennesaw, GA), composed of stoichiometrically equivalent doses of bovine serum albumin (BSA, 45% solution) and glutaraldehyde (GTA, 10% solution), is the only albumin-based bioadhesive approved in the United States for limited clinical use. It was developed not only to act as a sealant, but also as an agent to strengthen friable tissues, particularly in acute aortic dissection. Prior to use, BSA and
GTA are kept separate, and mixed just before they are delivered. The mechanism of adhesion is based on a chemical reaction between aldehydes and amines. The amines, provided by the $-amino groups of lysine residues in albumin and extracellular matrix proteins from the tissues to be bonded, are covalently cross-linked by GTA. Therefore,
BSA/GTA bioadhesives guarantee strong adherence to tissues and synthetic materials due to the permanent covalent bonds formed with the tissue One disadvantage of albumin/glutaraldehyde bioadhesives is its potential toxic effects: cells may be exposed
143 to (i) the bioadhesive directly, (ii) residues of the bioadhesive such as excess GTA, or
(iii) components released or degraded from the bioadhesive. Although the manufacturer of BioGlue claims that is not locally toxic as long as unmixed GTA does not contact tissues directly, Fürst et al. have shown that released GTA from polymerized BioGlue induced cytotoxic effects both in vitro and in vivo (Fürst and Banerjee 2005). However, due to the type of application, the washing procedure in this study involved using saline solution for 1 min only. Optimizing the compositions of GTA and modifying the washing procedure can lead to reduced cytotoxicity of BioGlue.
In order for bioadhesives to be used to secure an inert structure, it must be shown to possess adequate mechanical strength and stability over time. Measuring the mechanical strength of the bonded structure may be the most direct way to quantify its performance.
There is no universally agreed test or parameter to determine the bonding strength; there are, however, two commonly performed tests: tensile and peel tests (Lenaerts and Gurny
1990). They are popular because they are relatively quick and simple and can be used to screen several candidates. In tensile tests, forces are perpendicular to the plane of the joint and distribute uniformly over the entire joint. Also, all bioadhesives are put to work at the same time. Stress at a tearing point or failure is one of the parameters obtained from tensile tests. On the other hand, forces are limited to a very fine line at the edge of the joint in peel tests. Peel tests measure the ability of the joint to resist peeling forces.
The work of adhesion (WOA), the force to peel one bonded component from the other component, and the maximum force of detachment (Fmax), the maximum force to separate two components, are the most preferred bioadhesive parameters from peel tests (Zaman et al. 2010).
144
Figure 6-1 Two commonly performed mechanical tests for testing bonding strength of bioadhesives: (a) tensile and (b) peel tests.
In the present study, we developed an adhesion technique based on BSA/GTA bioadhesives to bond collagen scaffolds. The bioadhesives were made from various combinations of concentrations of BSA (8 and 12 % w/v) and GTA (4 and 8 % v/v) in this study. We investigated the mechanical properties of bioadhesives through tensile and peel tests. We further fabricated a CG composite with a closed network using the bioadhesives. Finally, we examined the potential release of GTA from the polymerized bioadhesives and its cytotoxic potential in vitro. During this investigation, we discovered that a commonly used cell viability method suffered from interference by the use of the bioadhesives.
6.2 Materials and Method
6.2.1 Preparation of CG Scaffolds
CG scaffolds of size 2.5 (width) x 6 (length) x 0.5 (depth) cm were fabricated using the method as described in Chapter 3, Section 3.2.2 and stored at 4°C prior to use. Since our
145 goal is to stack CG scaffolds with flow networks to form three-dimensional structures, we sham-treated the scaffolds for collagen lithography. The CG scaffold was then flattened using a modified lithography approach. First, the CG scaffold was placed in a P-150 petri dish and a standard glass slide was placed on the scaffold. An aluminum bar (1 x 3 x 2 in) was immediately placed on the glass slide and aligned carefully with the glass slide. EDC solution (Section) was subsequently added to cross-link the scaffold. After 48 h, the CG scaffold (approximately 300 µm thick) was removed and cross-linked with fresh EDC solution for an additional 24 h at 4°C. Excess EDC was removed from the scaffold by using PBS, and the CG scaffold was stored in PBS at 4°C until use. 1 x 2.5 cm
Rectangular CG scaffolds of size 1 (width) x 2.5 (length) x 0.3 (depth) cm were cut with a scalpel for use in mechanical tests, and 12 mm diameter discs were obtained by using a biopsy punch for use in cytotoxicity experiments.
6.2.2 Adhesion Technique
4% and 8% v/v GTA were diluted from 50% GTA (Sigma) with PBS and stored at 4°C prior to use. 80 mg and 120 mg of BSA (Sigma) were added to 1 mL of PBS at pH 7.4 to obtain 8% and 12% w/v BSA, respectively. Two rectangular samples of CG scaffolds of the same size were placed in a P-100 petri dish. Kimwipes were used to remove moisture on the bonding surfaces prior to bonding. 100 µL each of GTA and BSA solutions were added to a glass slide and thoroughly mixed by using a micropipette tip. Four different combinations of GTA and BSA solutions (4% GTA & 8% BSA, 4% GTA & 12% BSA,
8% GTA & 8 % BSA, 8% GTA & 12% BSA) were tested. Within 1 min after mixing, with the help of a micropipette tip the mixture was spread evenly on the dry surface of
146 one of the samples, and the samples were placed on top of each other to initiate bonding.
The assembled sample was air-dried in a laminar hood. The bonding process was complete when the sample started drying, which occurred within 1 hr. After the bonding process was complete, and the bonded samples were stored in PBS at 4°C until used in mechanical tests.
6.2.3 Mechanical testing
Bonding strength was determined through peel and tensile tests (n = 4 for each condition) by using an university testing machine (Instron). For peel tests [Figure 6-2 (a)], the bonded area was 1 x 1.5 cm, and for tensile tests [Figure 6-2 (b)], the bonded area was 1 x 0.5 cm. A larger bonded area was chosen for the peel test because peel tests are not as quantitative as tensile tests and a larger area leads to a larger data set and more accuracy.
In tensile tests, the samples were tested to failure at a strain rate of 0.1 mm/s. Ultimate stress at failure was determined. In the peel tests, WOA and Fmax were determined
[Figure 6-3]
147
Figure 6-2 Schematic diagrams of bonded CG scaffolds for (a) peel and (b) tensile tests.
Figure 6-3 Digital images of bonded CG scaffolds undergoing (a) peel and (b) tensile tests in a universal testing machine.
148 6.2.4 Cytotoxicity of BSA/GTA Bioadhesive in vitro
6.2.4.1 Quantification of Released GTA
All BSA and GTA solutions were sterilized by filtering the solutions through 0.2 µm syringe filters. 12 mm diameter disc-shaped CG scaffolds were bonded by using the method described above. Immediately after bonding, 3 mL of PBS was added to each sample and the mixture was agitated for 1 h at room temperature in a shaker. A 1 mL aliquot of solution from the mixture was harvested and immediately frozen at -20C for
GTA release analysis. The bonded samples went through two more washes of similar nature with fresh PBS. The samples were then stored in 1X antibiotic-antimycotic solution in PBS at 4°C overnight. Excess antibiotic-antimycotic was removed from the samples through a series of three washes with PBS. Each scaffold was then stored in 3 mL PBS at 4°C overnight for use in cytotoxicity experiments.
Released GTA was quantitated by using a colorimetric micromethod (Boratynski and Zal
1990; Fürst and Banerjee 2005) (n = 3 for each condition). Frozen PBS samples obtained during the first hour of wash of bonded CG scaffolds were thawed and 200 µL of the samples or GTA standard solutions were mixed with 1 mL of phenol reagent (40 µL of
5% phenol in water in 10 mL of 70% perchloric acid; Sigma) and incubated at room temperature for 15 min. Absorbance was measured at 479 nm with a spectrophotometer
(SpectraMax M2, Molecular Devices, Sunnyvale, CA).
6.2.4.2 Cytotoxicity In vitro
A custom 4-well culture plate was fabricated by bonding a PDMS slab (2 x 5 x 1 cm) containing four 12 mm diameter holesto a glass slide in the presence of oxygen plasma
149 bonding (Chaudhury and Whitesides 1991; Fakes et al. 1988; Owen and Smith 1994) in a plasma etcher (Structure Probe, Inc., West Chester, PA). The multi-well culture plate was steam sterilized in an autoclave and placed in a P-100 petri dish for cytotoxicity studies.
Human keratinocytes were isolated from discarded neonatal human foreskins by the method described in Section 3.2.4.2. The cells were cultured in KSFM, and passage 2-4 cells were used in the experiments. 100,000 cells were seeded onto the bonded CG scaffold using 500 µL KSFM and cultured for 24 h.
Cell cytotoxicity was investigated at 5 and 24 h after seeding cells by using a standard
MTT assay (n = 3 for each condition). The MTT solution, containing 0.5 mg/mL (3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Invitrogen), was prepared in
PBS freshly and sterilized using a syringe filter. After seeding cells for 5 and 24 h, the individual bonded CG scaffolds with cells were transferred to a 12-well plate, and incubated with 3 mL of sterile MTT solution at 37°C in a 5% CO2 containing humidified air environment for 4 h. After incubation, the MTT solution was aspirated and the samples were incubated with 1 mL of 2-methoxyethanol (Sigma) at room temperature for
3 h with mixing. The amount of MTT-formazan in the solution was determined by measuring the absorbance at 570 nm with a spectrophotometer.
Cell cytotoxicity was also investigated under scanning electron microscopy (SEM)
(Quanta 200 3D, FEI, Hillsboro, Oregon). After seeding cells for 5 and 24 h, the bonded
CG scaffolds with cells were fixed with 4% buffered-paraformaldehyde solution for 1 h at room temperature. They were then dehydrated in solutions of increasing concentrations of ethanol in water, 30%, 50%, 70%, 80%, 90%, 95% and 100%, for at least 1 hour in each solution. The 100% ethanol was changed twice to ensure all water in the samples
150 was exchanged with ethanol. Samples were then dried using hexamethyldisilazane
(HMDS) (Nation 1983). After dehydration, the samples were immersed in HMDS for 5 min at room temperature, and air dried in a fume hood. Dried samples were Pd-sputter- coated using a sputter coater, and then observed under SEM.
6.2.5 Fabrication of Composite CG Scaffolds with Flow Networks
Two CG scaffolds of size 2.5 (width) x 6 (length) x 0.5 (depth) cm were fabricated using the method as described in Chapter 3, Section 3.2.2. One scaffold was patterned to obtain an open network using the method as described in Chapter 5, Section 5.2.1.2, and the other one was flattened using the method described in Chapter 56, Section 6.2.1. A
PDMS slab of size 3 (width) x 6.5 (length) x 0.5 (depth) cm was prepared by using soft- lithography and attached to a bioreactor using double-sided silicone adhesive (FT 3012,
Avery Dennison). The flattened CG scaffold was then bonded to the PDMS slab using silicone adhesive. The PDMS slab served as a stiff backing for the soft collagen scaffold so that medium can be injected into the scaffold. After removing moisture on the surface of the flattened CG scaffold using Kimwipes, 500 µL bioadhesive obtained from 250 µL of 4% GTA and 250 µL of 12% BSA solutions were applied to the surface. The patterned
CG scaffold was subsequently placed on the bioadhesive with the open network facing the flattened CG scaffold to obtain a composite with a closed network. The composite was air-dried in a laminar hood for 1 h to ensure the bonding is complete. The closed flow network was examined under a micro-CT instrument (Inveon PET-CT, Siemens,
Hoffman Estates, IL). Iodinated 50 nm-liposomal particles (a gift from Dr. Ketan
151 Ghaghada of University of Texas Health Science Center) were injected manually into the closed flow network, and the entire assembly was imaged.
6.2.6 MTT Interference
MTT reduction by the bonded CG scaffold with cells was found to be greater than that by the untreated CG scaffold with cells. As a result, MTT reduction by BSA and GTA was investigated to the influence of BSA and GTA on MTT reduction in the absence of cells.
Different concentrations of solutions of BSA (0.5, 1, 4, and 8%), FBS (10 and 20%)
(BSA concentrations in FBS: 2 - 3.6 g/100mL, Invitrogen) and GTA (0, 0.5, 1, and 4%) were prepared. MTT solution (0.5 mg/mL) was prepared in PBS freshly prior to experiments. All solutions weresterilized by passing them through 0.2 µm syringe filters.
100 µL of GTA was added to each well of a 48-well plate followed by 100 µL of BSA or
FBS to obtain the bioadhesives (final volume =200 µL). In controls, only BSA, FBS or
GTA solution was used. When testing the reduction, 300 µL of MTT solution was added to each well and the solutions were incubated at 37°C in a 5% CO2 containing humidified air environment for 4 h. After incubation, 100 µL of 2-methoxyethanol was added and the contents agitated at room temperature for 3 h. Before measuring the absorbance, 100
µL of PBS was added to the wells that only contained one reagent (BSA, FBS, or GTA) to obtain the final volume of 200 µL. The amount of MTT-formazan in the solution was determined by measuring the absorbance at 570 nm with a spectrophotometer.
152 6.2.7 Statistical Analysis
For all quantitative results, a one-way analysis of variance (ANOVA) and Tukey-Kramer multiple comparison procedures were performed using Origin (OriginLab, Northampton,
MA) to compare data groups. A p value less than 0.05 was used to determine statistical significance.
6.3 Results
6.3.1 Bonding Strengths of Bioadhesives
To evaluate the bonding strength of the bioadhesives, we performed peel and tensile tests on the bonded CG scaffolds. Figure 6-4 (a) shows WOA, the average force to peel one bonded component from the other component, and Figure 6-4 (b) shows Fmax, the maximum force to separate two components, for various bioadhesives obtained from different concentrations of BSA and GTA. As illustrated in the figure, the bioadhesives obtained from 4%GTA/12%BSA had the greatest bonding strength among all bioadhesives. When using 4% GTA, the bonding strength was significantly enhanced by increasing the BSA concentration, while the bonding strength was not affected by increasing the BSA concentration when using 8% GTA. Similarity, when using 8% BSA, the bonding strength was significantly enhanced by increasing the GTA concentration, while the bonding strength was unaffected by increasing the GTA concentration when using 12% BSA. Figure 6-4 (c) shows the maximum stress at failure in tensile tests.
Although the maximum stress seemed to increase with increasing BSA or GTA concentrations, the results were not statistically significant.
153
Figure 6-4 Bonding strength of bioadhesives obtained from various concentrations of BSA and GTA. (a) shows work of adhesion (WOA) and (b) shows maximum force of detachment (Fmax) obtained from peel tests. (c) shows maximum stress at failure obtained from tensile tests. Error bars represent SEM. * represents statistical significance relative to bioadhesives obtained from lower concentrations.
154 6.3.2 Cytotoxicity of BSA/GTA Bioadhesive in vitro
6.3.2.1 Release of GTA Quantification
To evaluate the cytotoxicity of GTA released from bioadhesive, GTA content in washing solutions was quantified after bonding and before seeding cells by using a colorimetric micromethod. The results [Figure 6-5] indicate that increasing the GTA concentration in the bioadhesives led to an increased release of GTA. However, it should be noticed that most of the GTA had reacted with BSA to form the bioadhesive, and only a small amount of GTA was released during washing with PBS. Figure 6-6 shows the effect of washing procedure on the amount of GTA released over 1 h. Before seeding cells, the amount of
GTA released over 1 h was as low as 0.01% in all bioadhesives.
Figure 6-5 The release of GTA from various bioadhesive after bonding. Error bars represent SEM. * represents statistical significance relative to GTA released after bonding.
155
Figure 6-6 The release of GTA from various bioadhesive after washing thoroughly with PBS for cell viability testing. Error bars represent SEM. * represents statistical significance relative to GTA released before washing.
6.3.2.2 In vitro Cytotoxicity of GTA Released from Bioadhesives
The in vitro cytotoxicity of released GTA was investigated from the attachment and growth of keratinocytes seeded on the bonded CG scaffolds. Figure 6-7 shows SEM images of cells attached to the bonded CG scaffolds obtained from various bioadhesives after 5 and 24 h of seeding. As illustrated from the images, the morphology of cells was similar to proliferating cells in a cell culture flask. Moreover, filaments were observed in every condition, suggesting cells were able to attach to the bonded CG scaffolds and spread. Samples of bioadhesives made with 4% GTA and 8% BSA, and 8% GTA and
12% BSA solutions show better cell morphology compared to samples from other bioadhesives.
156 157
Figure 6-7 SEM images of attached cells after 5 and 24 h of seeding with various bioadhesives.
6.3.3 Microfluidics within CG Composite
Figure 6-8 (a) shows a micro-CT image of microfluidics within the CG composite with the closed network bonded by the bioadhesives. Although there were leakages through the flow network, and the flow was not as clear as that obtained from a composite of a
CG scaffold and a PDMS slab bonded by the silicone adhesive from Avery Dennison
[Figure 6-8 (b)], we were still able to obtain a closed network and create flow between two CG scaffolds by using the bioadhesives in the bioreactor.
158
Figure 6-8 Micro-CT images of microfluidics within (a) a CG composite bonded by BSA/GTA bioadhesives and (b) a CG scaffold with flow networks and a PDMS slab bonded by a silicone adhesive.
6.3.4 MTT Interference
Figure 6-9 shows MTT reduction at various concentrations of BSA (0.5, 1, 4, and 8%) only, FBS (10 and 20%) only, and BSA or FBS with GTA (0, 0.5, 1, and 4%). When there was only BSA or FBS, only small amount of MTT reduction took place, while no reduction (within the resolution of the spectrophotometer) was observed when there was only GTA (data not shown). Even in the presence of a small amount of GTA (0.5%),
BSA or FBS reduced MTT to formazan significantly. Any further increase in MTT- reduction was achieved by increasing BSA/FBS levels but not by GTA, except for 4%
BSA and 4% GTA solution [Figure 6-9 (b)].
159
Figure 6-9 The effect of BSA or FBS/GTA on MTT reduction in the absence of cells. Error bars represent SEM. All data are statistical significance relative to no GTA.
6.4 Discussion
The aim of this study was to develop an adhesion technique for bonding porous collagen scaffolds. This will allow us to fabricate three-dimensional constructs. A commercial
160 adhesive (BioGlue), composed of bovine serum albumin (45%) and glutaraldehyde
(10%), is currently being used as a hemostatic agent for cardiac and vascular surgery.
BioGlue bonds to tissue covalently and reinforces friable tissue into a tougher and more workable consistency. Although its potential as a hemostatic agent has been investigated widely, no report has investigated its use to bond collagen scaffolds. In this study, we developed an adhesion technique based on the components of BioGlue to bond CG scaffolds. Due to concerns about toxicity of GTA and also the property of BSA preventing cell adhesion (Whitesides et al. 2001), significantly reduced concentrations of both components were used in this study (4 and 8% for GTA and 8 and 12% for BSA).
We tested the bonding strength and potential cytotoxicity of bioadhesives obtained from various combinations of concentrations of BSA and GTA. In a feasibility study, we utilized the adhesive to fabricate a CG composite with integrated flow networks. We chose 4%BSA and12% GTA for this purpose because the bioadhesive obtained from this composition has the best mechanical property as measured from peel tests. The results demonstrate that the adhesive can be used to obtain a closed flow network in CG scaffolds. This new approach provides a direction for developing layered CG scaffolds.
In this study, we first investigated the effect of BSA and GTA concentrations on bonding strength [Figure 6-4]. Our observation that increasing the BSA concentration from 8 to
12% led to a significant increase in bonding strength at the lower GTA concentration
(4%) but not at the higher GTA concentration (8%) suggest a stoichiometric relationship between GTA and BSA in the order of 4% to 12%. At 4% GTA, increasing BSA concentration from 8% to 12% led to increased cross-linking as GTA is present in excess and addition of substrate for crosslinking improved the bonding and subsequently the
161 mechanical strength (Taguchi et al. 2007). Further increases in GTA, on the other hand, led to stoichiometrically excess GTA that did not react with BSA. However, it should be mentioned that the released GTA was much smaller than GTA used in the bioadhesives.
It is possible that excess GTA bonded to amino group in CG scaffolds after applying to
CG scaffolds.
Similarity, increasing the GTA concentration from 4 to 8% led to a significant increase in bonding strength at the lower BSA concentration (8%) but not at the higher BSA concentration (12%), suggesting a stoichiometric relationship between BSA and GTA in the order of 8% to 12%. Active aldehyde groups of GTA can react with amino groups of proteins to form amide bond (Migneault et al. 2004). AT 8% BSA, 8% GTA provided more active aldehyde groups to react with BSA compared to 4% GTA, resulting in increased bonding strength. However, bonding strength was not improved at 12% BSA due to insufficient aldehyde groups to react with all amino groups provided by 12% BSA.
Interestingly, although aldehyde groups in 4% GTA was less than 8% GTA, bonding strength of 12%BSA/4%GTA was comparable with 12%BSA/8%GTA. This is probably because of higher cross-linking densities obtained from 12%BSA/4%GTA.
Fürst et al. have shown that release of glutaraldehyde from BioGlue caused significant in vitro and in vivo toxicity (Fürst and Banerjee 2005). They applied BioGlue to cellulose sheets following manufacture’s instruction, and submerged the sheets with 5 mL saline solution and incubated for 1 min. Supernatants were removed from samples, and released
GTA was quantified. They further cultured human embryo lung fibroblasts and mouse myoblasts with medium supplemented with 10% supernatants from BioGlue incubation.
The results suggested that the supernatants obtained from overlaying polymerized
162 BioGlue with saline solution were cytotoxic to both fibroblasts and myoblasts.
Conversely, although the attachments were not as good as that when using CG scaffolds without bioadhesives, our results demonstrated that human keratinocytes were able to attach and grow on the bonded CG scaffolds after washing thoroughly with PBS. It is probably due to concentrations of GTA used in our bioadhesives were lesser than that in
BioGlue (4 and 8% vs. 10%). Most importantly, released GTA was significantly reduced after washing procedure in our study, and released GTA in washing solutions in our study was also smaller than that in their study (20 vs. 100 µg/mL). Indeed, it has been suggested that potential cytotoxicity of GTA was not evidenced at lower percentage of
0.25% in collagen-based scaffolds (Ma et al. 2003). On the other hand, despite BSA is widely used to prevent cell attachment, our results demonstrated that the cells were able to adhere to all regions of CG scaffolds.
One remained challenge for developing a well-organized microvascularization with TE skin substitutes is lack of appropriate approach to bond collagen scaffolds. Although capillary flow networks were spontaneously formed after inosculating endothelial cells within porous collagen scaffolds (Tremblay et al. 2005), the architecture of microvascularization cannot be controlled. As a result, if we can fabricate CG composites with integrated micro-scale flow networks, a well-organized microvascularization can be obtained after performing endothelialization of flow networks. Although microfluidics results from the micro-CT image were not perfect and there were leakages, this was very likely due to our spreading technique and not due to inferior bonding of the scaffolds.
Indeed, except for initial leakage that occurred during flow initiation, there were no other leakages suggesting that this technique has tremendous potential. To our knowledge,
163 there has been no report that successfully demonstrated flow through defined structures within porous collagen scaffolds.
MTT, a yellow water soluble tetrazolim salt (3-(4, 5-dimethylthiazolyl-2)-2, 5- diphenyltetrazolium bromide), has been widely used as a means of measuring cell viability. In general, succinate dehydrogenase system of active mitochondria in live cells reduces MTT to yield water insoluble purple formazan. The formazan can be dissolved by using a suitable solvent and the resulting solution is analyzed spectrophotometrically.
The amount of formazan has been shown to be proportional to the number of metabolically active cells. As a result, MTT has been used in many applications to assess cell proliferation and cytotoxicity (Liu et al. 1997). However, researchers have shown that MTT is also reduced by other components such as ascorbic acid or flavonoids
(Bernas and Dobrucki 2002; Chakrabarti et al. 2000; Talorete et al. 2006). Chakrabarti et al. have shown that vitamin A (retinol) acts as an enzyme that catalyzes MTT reduction by vitamin C (ascorbic acid) (Chakrabarti et al. 2000), resulting in the formation of formazan in the absence of cells in M199 medium, containing both retinol and ascorbic acid, but not in RPMI 1640 medium. However, M199 is not commonly used medium for cell culture, while DMEM, the most frequently used medium, does not contain retinol and ascorbic acid. Moreover, both retinol and ascorbic acid are not essential for proliferation of most cell types. Thus, MTT reduction by these compounds may not affect most cell viability testing.
Initially, when we attempted to measure cell viability characteristics of the bioadhesives, huge amount of formazan was formed immediately (within 5 min) after the addition of
MTT to bonded CG scaffolds but not to control scaffolds. We, therefore, deduced that
164 components of the bioadhesive may reduce MTT to formazan in the absence of cells.
MTT reduction by only BSA or GTA was then investigated in the absence of cells.
Because most GTA would react with BSA in the adhesive, we used lower concentrations:
0.5, 1, and 4% GTA instead of 4 and 8%. Our observations that no formazan was formed when adding MTT to GTA only (data not shown), and only minor reduction was observed that too after a long incubation time (~4 h) when adding MTT to BSA only suggest that both components need to be present for MTT reduction. . Even the presence of very small amounts of GTA was sufficient for BSA to reduce MTT. This suggests that
GTA may act as a co-factor on BSA’s reaction with MTT. However, more experiments need to be performed to understand the mechanism.
Since FBS normally contains BSA (2 - 3.6 w/v %, Invitrogen) and is essential in many cell culture systems, we investigated the effect of FBS on MTT reduction with or without
GTA, and the results were similar to that of BSA. Other researchers have shown that concentrations of serum affected MTT reduction (Chakrabarti et al. 2000; Talorete et al.
2006; Zhang and Cox 1996). Our results indicate that albumin in the serum is the primary factor that causes the interference. As a result, we suggest it will be more accurate to use other viability assays in the presence of BSA or FBS with GTA.
6.5 Conclusions
In summary, we have developed an adhesion technique based on albumin/glutaraldehyde bioadhesives to bond collagen-based scaffolds. We show that bioadhesives bond collagen scaffolds very well. We demonstrate that cells were able to attach and grow on bonded
CG scaffolds, suggesting that the bioadhesives do not have evidenced cytotoxic effect in
165 vitro. We further show that it is feasible to use the bioadhesive to obtained closed flow networks in a CG composite. Finally, we have discovered interference in a commonly used cell viability assay due to adhesive components.
6.6 References
Adant JP, Detroz B, Dsilva M, Natowitz L, Ledoux M, Pestiaux B, Leclercq P. 1993. Skin-grafting with fibrin glue in burns. European Journal of Plastic Surgery 16(6):292-297.
Bernas T, Dobrucki J. 2002. Mitochondrial and nonmitochondrial reduction of MTT: Interaction of MTT with TMRE, JC-1, and NAO mitochondrial fluorescent probes. Cytometry 47(4):236-242.
Boratynski J, Zal T. 1990. Colorimetric micromethods for glutaraldehyde determination by means of phenol and sulfuric acid or phenol and perchloric acid. Analytical Biochemistry 184(2):259-262.
G'#$C23! H7)! G#6Q! ")! Y5! <)! G&%D2+#! ")! G-'D#6Q! (")! :C2D&3C599&#%#! ">! KRR@>! H&6C2%#62++E! '5;-9.! 2%D! ;#52DC&9#8&! 65++2$&%! 2%D! 4C5.56'599+#%Q2;+&! CE2+-'5%#6! 26#D! 9&3#P#%.&'4&%&.'2.#%$! %&.B5'Q9>! *#99-&! <%$#%&&'#%$! 02'.! "! ?WLOMN?[JWP?[WV>
Brown KJ, Maynes SF, Bezos A, Maguire DJ, Ford MD, Parish CR. 1996. A novel in vitro assay for human angiogenesis. Laboratory Investigation 75(4):539-555.
Chakrabarti R, Kundu S, Kumar S. 2000. Vitamin A as an enzyme that catalyzes the reduction of MTT to formazan by vitamin c. Journal of Cellular Biochemistry 80(1):133-138.
Chao HH, Torchiana DF. 2003. BioGlue: Alburnin/glutaraldehyde sealant in cardiac surgery. Journal of Cardiac Surgery 18(6):500-503.
Chapman WC, Sherman R, Boyce S, Malawer M, Hill A, Buncke G, Block JE, Fung CJ, Clavien P, Lee KF and others. 2001. A novel collagen-based composite offers effective hemostasis for multiple surgical indications: Results of a randomized controlled trial. Surgery 129(4):445-450.
Chaudhury MK, Whitesides GM. 1991. Direct measurement of interfacial interactions between semispherical lenses and flat sheets of poly(dimethylsiloxane) and their chemical derivatives. Langmuir 7(5):1013-1025.
166 Currie LJ, Sharpe JR, Martin R. 2001. The use of fibrin glue in skin grafts and tissue- engineered skin replacements: A review. Plastic and Reconstructive Surgery 108(6):1713-1726.
Fakes DW, Davies MC, Brown A, Newton JM. 1988. The surface-analysis of a plasma modified contact-lens surface by ssims. Surface and Interface Analysis 13(4):233- 236.
Fürst W, Banerjee A. 2005. Release of glutaraldehyde from an albumin-glutaraldehyde tissue adhesive causes significant in vitro and in vivo toxicity. Annals of Thoracic Surgery 79(5):1522-1529.
Hewitt CW, Marra SW, Kann BR, Tran HS, Puc MM, Chrzanowski FA, Tran JLV, Lenz SD, Cilley JH, Simonetti VA and others. 2001. BioGlue surgical adhesive for thoracic aortic repair during coagulopathy: Efficacy and histopathology. Annals of Thoracic Surgery 71(5):1609-1612.
Janakiraman V, Kienitz B, Baskaran H. 2007. Lithography technique for topographical micropatterning of collagen-glycosaminoglycan membranes for tissue engineering applications. Journal of Medical Devices 1(3):233-7.
Koolwijk P, vanErck MGM, deVree WJA, Vermeer MA, Weich HA, Hanemaaijer R, vanHinsbergh VWM. 1996. Cooperative effect of TNF alpha, bFGF, and VEGF on the formation of tubular structures of human microvascular endothelial cells in a fibrin matrix. Role of urokinase activity. Journal of Cell Biology 132(6):1177- 1188.
Kumar A, Maartens NF, Kaye AH. 2003. Evaluation of the use of BioGlue (R) in neurosurgical procedures. Journal of Clinical Neuroscience 10(6):661-664.
Leggat PA, Kedjarune U, Smith DR. 2004. Toxicity of cyanoacrylate adhesives and their occupational impacts for dental staff. Industrial Health 42(2):207-211.
Lenaerts V, Gurny R. 1990. Bioadhesive drug delivery systems. Boca Raton, Fla.: CRC Press. [7], 231 p. p.
Liu Y, Peterson DA, Kimura H, Schubert D. 1997. Mechanism of cellular 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction. Journal of Neurochemistry 69(2):581-593.
Ma L, Gao C, Mao Z, Zhou J, Shen J, Hu X, Han C. 2003. Collagen/chitosan porous scaffolds with improved biostability for skin tissue engineering. Biomaterials 24(26):4833-4841.
Migneault I, Dartiguenave C, Bertrand MJ, Waldron KC. 2004. Glutaraldehyde: behavior in aqueous solution, reaction with proteins, and application to enzyme crosslinking. Biotechniques 37(5):790-802.
167 Mosmann T. 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65(1- 2):55-63.
Nation JL. 1983. A new method using hexamethyldisilazane for preparation of soft insect tissues for scanning electron microscopy. Stain Technology 58(6):347-351.
Otani Y, Tabata Y, Ikada Y. 1996a. A new biological glue from gelatin and poly(L- glutamic acid). Journal of Biomedical Materials Research 31(2):157-166.
Otani Y, Tabata Y, Ikada Y. 1996b. Rapidly curable biological glue composed of gelatin and poly(L-glutamic acid). Biomaterials 17(14):1387-1391.
Owen MJ, Smith PJ. 1994. Plasma Treatment of Polydimethylsiloxane. Journal of Adhesion Science and Technology 8(10):1063-1075.
Oz MC, Cosgrove DM, III, Badduke BR, Hill JD, Flannery MR, Palumbo R, Topic N. 2000. Controlled clinical trial of a novel hemostatic agent in cardiac surgery. Annals of Thoracic Surgery 69(5):1376-1382.
Ranger WR, Halpin D, Sawhney AS, Lyman M, Locicero J. 1997. Pneumostasis of experimental air leaks with a new photopolymerized synthetic tissue sealant. American surgeon 63(9):788-795.
Renkens KL, Payner TD, Leipzig TJ, Feuer H, Morone MA, Koers JM, Lawson KJ, Lentz R, Shuey H, Conaway GL and others. 2001. A multicenter, prospective, randomized trial evaluating a new hemostatic agent for spinal surgery. Spine 26(15):1645-1650.
Sakon M, Monden M, Gotoh M, Kobayashi K, Kambayashi J, Mori T, Okamura J. 1989. Use of microcrystalline collagen powder and fibrinogen tissue adhesive for hemostasis and prevention of rebleeding in patients with hepatocellular carcinoma associated with cirrhosis of the liver. Surgery Gynecology & Obstetrics 168(5):453-454.
S-%$! ,^)! ,-2%$! 7H)! 1C2%$! ^,)! ,-2%$! I=)! ,9-! (1>! ?@@@>! <82+-2.#5%! 5F! $&+2.#%! CED'5$&+! 6'599+#%Q&D! B#.C! 82'#5-9! 6'599+#%Q#%$! 2$&%.9! 29! ;#52DC&9#8&9N! ]%! 8#.'5!9.-DE>!(5-'%2+!5F!G#53&D#62+!H2.&'#2+9!I&9&2'6C!J[LJMNWKRPWVR>
S&Q#%&! *)! =2Q23-'2! *)! SC#3#/-! \)! a&D2! ,)! H2.9-35.5! :)! *2Q#35.5! \)! :#E5.2%#! *>! KRR?>! "! %&B! .E4&! 5F! 9-'$#62+! 2DC&9#8&! 32D&! F'53! 45'6#%&! 65++2$&%! 2%D! 45+E$+-.23#6!26#D>!(5-'%2+!5F!G#53&D#62+!H2.&'#2+9!I&9&2'6C!WJLKMNVRWPV?R>
Taguchi T, Saito H, Iwasashi M, Sakane M, Kakinoki S, Ochiai N, Tanaka J. 2007. Development of a novel glue consisting of naturally-derived biomolecules: Citric acid and human serum albumin. Journal of Nanoscience and Nanotechnology 7(3):742-747.
168 Talorete TPN, Bouaziz M, Sayadi S, Isoda H. 2006. Influence of medium type and serum on MTT reduction by flavonoids in the absence of cells. Cytotechnology 52(3):189-198.
Tremblay P-L, Hudon V, Berthod F, Germain L, Auger FA. 2005. Inosculation of tissue- engineered capillaries with the host's vasculature in a reconstructed skin transplanted on mice. American Journal of Transplantation 5(5):1002-1010.
Tseng YC, Tabata Y, Hyon SH, Ikada Y. 1990. In vitro toxicity test of 2-cyanoacrylate polymers by cell culture method. Journal of Biomedical Materials Research 24(10):1355-1367.
Vedung S, Hedlund A. 1993. Fibrin glue: Its use for skin grafting of contaminated burn wounds in areas difficult to immobilize. Journal of Burn Care and Rehabilitation 14(3):356-358.
Vibe P, Pless J. 1983. A new method of skin-graft adhesion. Scandinavian Journal of Plastic and Reconstructive Surgery and Hand Surgery 17(3):263-264.
Whitesides GM, Ostuni E, Takayama S, Jiang XY, Ingber DE. 2001. Soft lithography in biology and biochemistry. Annual Review of Biomedical Engineering 3:335-373.
Zaman MA, Martin GP, Rees GD. 2010. Bioadhesion and retention of non-aqueous delivery systems in a dental hard tissue model. Journal of Dentistry 38(9):757-764.
Zhang ZD, Cox G. 1996. MTT assay overestimates human airway smooth muscle cell number in culture. Biochemistry and Molecular Biology International 38(3):431- 436.
169 Chapter 7
CONCLUSIONS
7.1 Conclusions
Decades ago, it was impossible to imagine growing TE skin substitutes to repair human skin. Initially, the only options to treat injuries requiring skin substitution were either autologous or allogeneic split or full-thickness skin grafts. Over the last 35 years, TE skin substitutes have developed and progressed at a very rapid rate. Although autografts are still the best solution to any injury requiring skin substitution, TE skin substitutes serve as the best alternatives to the autografts, especially when patients suffer extensive burns
(>60% total body surface area). Today, more than 10 TE skin substitutes have been approved by the US Food and Drug Administration (FDA) and have been used in clinical applications. However, current TE skin substitutes still encounter several challenges.
Delayed or absence of vascularization has been suggested as one of the major reasons for implanted skin failure (Boyce 1996).
In this dissertation, we have developed an in vitro model of TE skin substitute with integrated flow networks in a perfusion bioreactor. We hypothesize that this model could resolve the current limitations suffered by TE skin substitutes and provide a better model for clinical applications. We describe a four-branched approach, which included: (i) design and characterization of optimal planar flow networks with maximum transport efficiency, (ii) fabrication of scaffolds for TE skin substitutes and micropatterning, (iii) development of a perfusion bioreactor system and culture of TE skin substitutes with
170 integrated flow networks, and (iv) endothelialization of the flow networks for vascularization. The first part of the approach has been done previously by our group and described elsewhere (Janakiraman et al. 2007a; Janakiraman et al. 2007b). The objective of this dissertation is to achieve the second and third approaches and is described in detail.
(i) Fabrication of scaffolds for TE skin substitutes.
The first challenge in skin tissue engineering is the optimization of biomaterial scaffold properties. To improve nutrient transfer within TE skin substitutes, porous scaffolds are generally utilized instead of solid matrices. However, these scaffolds have not been optimized. We have developed a simple and rapid approach based on centrifugation to fabricate highly concentrated yet porous collagen scaffolds for TE skin substitutes (Liang et al. 2010). We show that scaffolds obtained from concentrated solutions have various collagen densities and properties. In general, increasing the collagen density increases water uptake per volume, leads to smaller pore sizes, and improves mechanical properties including compressive and tensile moduli compared to scaffolds currently in use. We further demonstrate a well-stratified TE skin substitute obtained by utilizing the concentrated scaffolds in vitro.
(ii) Development of a perfusion bioreactor system and culture of TE skin substitutes with integrated flow networks.
We have designed a perfusion bioreactor system for TE skin substitutes with integrated flow networks. The perfusion bioreactor not only provides a submerged culture mode for keratinocyte proliferation but also an air-liquid interface culture mode for keratinocyte
171 differentiation. We discuss the effect of perfusion rates on glucose concentration profiles in the bioreactor by using computational models.
Next, we have developed an in vitro model of TE skin substitute with integrated flow networks in the perfusion bioreactor by utilizing the concentrated scaffolds. The effects of flow networks and perfusion rates on the epidermis formation and keratinocyte proliferation were assessed. We show that an appropriate seeding density and surface modification improve keratinocyte attachment and short-term survival. We further show that TE skin substitutes cultured in the perfusion bioreactor were morphologically comparable with skin substitutes cultured in the static condition. Increasing perfusion rates led to thicker and well-organized epidermis. Most importantly, epidermis formation benefited by utilizing concentrated CG scaffolds in the absence of human fibroblasts.
Overall, we successfully demonstrate the effect of mass transport on the growth of TE skin substitutes.
(iii) Development of an adhesion technique based on albumin/glutaraldehyde adhesives for bonding collagen scaffolds.
Finally, to obtain microvascularization within porous collagen scaffolds, we have developed an adhesion technique based on albumin/glutaraldehyde bioadhesives to bond collagen scaffolds. We investigate the mechanical properties of the adhesion. We demonstrate that cells were able to attach and grow on bonded CG scaffolds, suggesting that the bioadhesives do not have evidenced cytotoxic effect in vitro. We further show that it is feasible to use the bioadhesive to obtained closed flow networks in a CG
172 composite. Finally, we discover that BSA significantly reduces MTT, a commonly used cell viability assay, in the presence of a small amount of GTA.
7.2 Future Directions
In this dissertation, we have successfully developed an in vitro model of TE skin substitute with integrated flow networks in a perfusion bioreactor. Single bifurcation of flow network was utilized in this study. To fully understand the effect of mass transport efficiency on the epidermis formation, higher generations of flow networks should be utilized.
Researchers have shown that perfusion rates could affect cell proliferation and differentiation in TE bone and cell culture systems by many means, including flow- induced shear stress (Cartmell et al. 2003), oxygen concentrations (Pathi et al. 2005), and autocrine/paracrine factor secretion (Dan et al. 2010). However, since medium is perfused through the flow networks and cells are not exposed directly to medium in our system, the role of perfusion rates is not clear. To obtain an optimal perfusion rate, it is essential to investigate how the perfusion rates affect epidermis formation.
Finally, to fully eliminate the mass transfer limitation encountered by current TE skin substitutes, endothelialization must be performed in the integrated flow networks to obtain vasculature in vitro. To accomplish this, endothelial cells will be seeded in the flow networks embedded within collagen scaffolds to form vasculature. Although we have developed an adhesion technique to bond collagen scaffolds and tested the short- term cytotoxicity in vitro, there are still some challenges that have to be investigated such as leakages and long-term cytotoxicity. After obtaining vasculature, keratinocytes will be
173 co-cultured with endothelial cells to achieve our ultimate goal: TE skin substitutes with integrated vasculature in vitro, and the performance of TE skin substitutes with integrated vasculature in vivo will be assessed.
7.3 References
Boyce ST. 1996. Cultured skin substitutes: A review. Tissue Engineering 2(4):255-266.
Cartmell SH, Porter BD, García AJ, Guldberg RE. 2003. Effects of medium perfusion rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Engineering 9(6):1197-1203.
Dan L, Chua C-K, Leong K-F. 2010. Fibroblast response to interstitial flow: A state-of- the-art review. Biotechnology and Bioengineering 107(1):1-10.
Janakiraman V, Kienitz B, Baskaran H. 2007a. Lithography technique for topographical micropatterning of collagen-glycosaminoglycan membranes for tissue engineering applications. Journal of Medical Devices 1(3):233-7.
Janakiraman V, Mathur K, Baskaran H. 2007b. Optimal planar flow network designs for tissue engineered constructs with built-in vasculature. Annals of Biomedical Engineering 35(3):337-347.
Liang W-H, Kienitz BL, Penick KJ, Welter JF, Zawodzinski TA, Baskaran H. 2010. Concentrated collagen-chondroitin sulfate scaffolds for tissue engineering applications. Journal of Biomedical Materials Research Part A 94A(4):1050-1060.
Pathi P, Ma T, Locke BR. 2005. Role of nutrient supply on cell growth in bioreactor design for tissue engineering of hematopoietic cells. Biotechnology and Bioengineering 89(7):743-758.
174 BIBLIOGRAPHY
1990. U.S. Scientific Registry for Organ Transplantation and the Organ Procurement and Transplant Network. Annual report. Richmond VA.
Achauer BM, VanderKam VM, Celikoz B, Jacobson DG. 1998. Augmentation of facial soft-tissue defects with Alloderm dermal graft. Annals of Plastic Surgery 41(5):503-507.
Adant JP, Detroz B, Dsilva M, Natowitz L, Ledoux M, Pestiaux B, Leclercq P. 1993. Skin-grafting with fibrin glue in burns. European Journal of Plastic Surgery 16(6):292-297.
Adzick NS, Longaker MT. 1992. Scarless fetal healing - Therapeutic implications. Annals of Surgery 215(1):3-7.
Agache PG, Monneur C, Leveque JL, Rigal J. 1980. Mechanical properties and Young's modulus of human skin in vivo. Archives of Dermatological Research 269(3):221-232.
Aigner J, Tegeler J, Hutzler P, Campoccia D, Pavesio A, Hammer C, Kastenbauer E, Naumann A. 1998. Cartilage tissue engineering with novel nonwoven structured biomaterial based on hyaluronic acid benzyl ester. Journal of Biomedical Materials Research 42(2):172-181.
Alonso L, Fuchs E. 2003. Stem cells of the skin epithelium. Proceedings of the National Academy of Sciences of the United States of America 100(Suppl 1):11830-11835.
Alves da Silva ML, Martins A, Costa-Pinto AR, Correlo VM, Sol P, Bhattacharya M, Faria S, Reis RL, Neves NM. 2010. Chondrogenic differentiation of human bone marrow mesenchymal stem cells in chitosan-based scaffolds using a flow- perfusion bioreactor. Journal of Tissue Engineering and Regenerative Medicine.
Angelova N, Hunkeler D. 1999. Rationalizing the design of polymeric biomaterials. Trends in Biotechnology 17(10):409-421.
Arano T, Sato T, Matsuzaka K, Ikada Y, Yoshinari M. 2010. Osteoblastic cell proliferation with uniform distribution in a large scaffold using radial-flow bioreactor. Tissue Engineering Part C-Methods 16(6):1387-1398.
Arturson G. 1996. Pathophysiology of the burn wound and pharmacological treatment. The Rudi Hermans lecture, 1995. Burns 22(4):255-274.
Auger FA, Lacroix D, Germain L. 2009. Skin substitutes and wound healing. Skin Pharmacology and Physiology 22(2):94-102.
175 Ayer JP. 1964. Elastic tissue. In: Hall DA, editor. International Review of Connective Tissue Research. New York: Academic Press. p 33-100.
Bancroft GN, Sikavitsas VI, Mikos AG. 2003. Design of a flow perfusion bioreactor system for bone tissue-engineering applications. Tissue Engineering 9(3):549-554.
Banks-Schlegel S, Green H. 1981. Involucrin synthesis and tissue assembly by keratinocytes in natural and cultured human epithelia. The Journal of Cell Biology 90(3):732-737.
Battler A, Leor J, Singer A, Simon M. 2006. Wound healing and skin substitutes. Stem Cell and Gene-Based Therapy: Springer London. p 375-393.
Bell E, Ehrlich HP, Buttle DJ, Nakatsuji T. 1981. Living tissue formed in vitro and accepted as skin-equivalent tissue of full thickness. Science 211(4486):1052-4.
Bernas T, Dobrucki J. 2002. Mitochondrial and nonmitochondrial reduction of MTT: Interaction of MTT with TMRE, JC-1, and NAO mitochondrial fluorescent probes. Cytometry 47(4):236-242.
Bilbo PR, Nolte CJM, Oleson MA, Mason VS, Parenteau NL. 1993. Skin in complex culture - The transition from culture phenotype to organotypic phenotype. Journal of Toxicology-Cutaneous and Ocular Toxicology 12(2):183-196.
Billingham RE, Reynolds J. 1952. Transplantation studies on sheets of pure epidermal epithelium and on epidermal cell suspensions. British Journal of Plastic Surgery 5(1):25-36.
Blank IH. 1953. Further observations on factors which influence the water content of the stratum corneum. Journal of Investigative Dermatology 21(4):259-271.
Bolivar-Flores J, Poumian E, Marsch-Moreno M, de Oca GM, Kuri-Harcuch W. 1990. Use of cultured human epidermal keratinocytes for allografting burns and conditions for temporary banking of the cultured allografts. Burns 16(1):3-8.
Boratynski J, Zal T. 1990. Colorimetric micromethods for glutaraldehyde determination by means of phenol and sulfuric acid or phenol and perchloric acid. Analytical Biochemistry 184(2):259-262.
Bosman FT, Stamenkovic I. 2003. Preface to extracellular matrix and disease. Journal of Pathology 200(4):421-422.
Bottcher-Haberzeth S, Biedermann T, Reichmann E. 2010. Tissue engineering of skin. Burns 36(4):450-460.
Boyce ST. 1996. Cultured skin substitutes: A review. Tissue Engineering 2(4):255-266.
176 Boyce ST, Goretsky MJ, Greenhalgh DG, Kagan RJ, Rieman MT, Warden GD. 1995. Comparative assessment of cultured skin substitutes and native skin autograft for treatment of full-thickness burns. Annals of Surgery 222(6):743-752.
Branski LK, Herndon DN, Pereira C, Mlcak RP, Celis MM, Lee JO, Sanford AP, Norbury WB, Zhang X-J, Jeschke MG. 2007. Longitudinal assessment of Integra in primary burn management: A randomized pediatric clinical trial *. Critical Care Medicine 35(11):2615-2623.
Breathnach AS. 1964. The dermo-epidermal junction. In: Rook A, Champion RH, editors. Progress in the Biological Sciences in Relation to Dermatology. Cambridge: Cambridge University Press. p 415-425.
Breathnach AS. 1975. Aspects of epidermal ultrastructure. Journal of Investigative Dermatology 65(1):2-15.
Briggaman RA, Wheeler CE, Jr. 1975. The epidermal-dermal junction. Journal of Investigative Dermatology 65(1):71-84.
Brigham MD, Bick A, Lo E, Bendali A, Burdick JA, Khademhosseini A. 2009. Mechanically robust and bioadhesive collagen and photocrosslinkable hyaluronic acid semi-interpenetrating networks. Tissue Engineering Part A 15(7):1645-1653.
Brown EJ. 1995. Phagocytosis. Bioessays 17(2):109-117.
Brown KJ, Maynes SF, Bezos A, Maguire DJ, Ford MD, Parish CR. 1996. A novel in vitro assay for human angiogenesis. Laboratory Investigation 75(4):539-555.
Buijtenhuijs P, Buttafoco L, Poot AA, Daamen WF, van Kuppevelt TH, Dijkstra PJ, de Vos RAI, Sterk LMT, Geelkerken BRH, Feijen J and others. 2004. Tissue engineering of blood vessels: characterization of smooth-muscle cells for culturing on collagen-and-elastin-based scaffolds. Biotechnology and Applied Biochemistry 39(Pt 2):141-149.
Burdick JA, Anseth KS. 2002. Photoencapsulation of osteoblasts in injectable RGD- modified PEG hydrogels for bone tissue engineering. Biomaterials 23(22):4315- 4323.
Burke JF, Yannas IV, Quinby WCJ, Bondoc CC, Jung WK. 1981. Successful use of a physiologically acceptable artificial skin in the treatment of extensive burn injury. Annals of Surgery 194(4):413-428.
Bush KA, Pins GD. 2010. Carbodiimide conjugation of fibronectin on collagen basal lamina analogs enhances cellular binding domains and epithelialization. Tissue Engineering Part A 16(3):829-838.
177 Callcut RA, Schurr MJ, Sloan M, Faucher LD. 2006. Clinical experience with Alloderm: A one-staged composite dermal/epidermal replacement utilizing processed cadaver dermis and thin autografts. Burns 32(5):583-588.
Carrino DA, Arias JL, Caplan AI. 1991. A spectrophotometric modification of a sensitive densitometric safranin-O assay for glycosaminoglycans. Biochemistry International 24(3):485-495.
Carsin H, Ainaud P, Le Bever H, Rives JM, Lakhel A, Stephanazzi J, Lambert F, Perrot J. 2000. Cultured epithelial autografts in extensive burn coverage of severely traumatized patients: a five year single-center experience with 30 patients. Burns 26(4):379-387.
Cartmell SH, Porter BD, García AJ, Guldberg RE. 2003. Effects of medium perfusion rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Engineering 9(6):1197-1203.
Chakrabarti R, Kundu S, Kumar S. 2000. Vitamin A as an enzyme that catalyzes the reduction of MTT to formazan by vitamin c. Journal of Cellular Biochemistry 80(1):133-138.
Chao HH, Torchiana DF. 2003. BioGlue: Alburnin/glutaraldehyde sealant in cardiac surgery. Journal of Cardiac Surgery 18(6):500-503.
Chapman WC, Sherman R, Boyce S, Malawer M, Hill A, Buncke G, Block JE, Fung CJ, Clavien P, Lee KF and others. 2001. A novel collagen-based composite offers effective hemostasis for multiple surgical indications: Results of a randomized controlled trial. Surgery 129(4):445-450.
Chaudhury MK, Whitesides GM. 1991. Direct measurement of interfacial interactions between semispherical lenses and flat sheets of poly(dimethylsiloxane) and their chemical derivatives. Langmuir 7(5):1013-1025.
Chen D-C, Lai Y-L, Lee S-Y, Hung S-L, Chen H-L. 2007. Osteoblastic response to collagen scaffolds varied in freezing temperature and glutaraldehyde crosslinking. Journal of Biomedical Materials Research Part A 80A(2):399-409.
Cherubino P, Grassi F, Bulggheroni P, Ronga M. 2003. Autologous chondrocyte implantation using a bilayer collagen membrane: a preliminary report. Journal of Orthopaedic Surgery (Hong Kong) 11(1):10-15.
Clark RAF, Lanigan JM, Dellapelle P, Manseau E, Dvorak HF, Colvin RB. 1982. Fibronectin and fibrin provide a provisional matrix for epidermal cell migration during wound reepithelialization. Journal of Investigative Dermatology 79(5):264-269.
178 Cohen S, Bano MC, Cima LG, Allcock HR, Vacanti JP, Vacanti CA, Langer R. 1993. Design of synthetic polymeric structures for cell transplantation and tissue engineering. Clinical Materials 13(1-4):3-10.
Colton CK. 1995. Implantable biohybrid artificial organs. Cell Transplantation 4(4):415- 436.
Cooper ML, Hansbrough JF, Spielvogel RL, Cohen R, Bartel RL, Naughton G. 1991. In vivo optimization of a living dermal substitute employing cultured human fibroblasts on a biodegradable polyglycolic acid or polyglactin mesh. Biomaterials 12(2):243-248.
Cotsarelis G. 2006. Epithelial stem cells: A folliculocentric view. Journal of Investigative Dermatology 126(7):1459-1468.
Coulombe PA. 1997. Towards a molecular definition of keratinocyte activation after acute injury to stratified epithelia. Biochemical and Biophysical Research Communications 236(2):231-238.
Cross VL, Zheng Y, Won Choi N, Verbridge SS, Sutermaster BA, Bonassar LJ, Fischbach C, Stroock AD. 2010. Dense type I collagen matrices that support cellular remodeling and microfabrication for studies of tumor angiogenesis and vasculogenesis in vitro. Biomaterials 31(33):8596-8607.
Cuono C, Langdon R, McGuire J. 1986. Use of cultured epidermalvautografts and dermal allografts as skin replacement after burn injury. Lancet 1(8490):1123-1124.
Currie LJ, Sharpe JR, Martin R. 2001. The use of fibrin glue in skin grafts and tissue- engineered skin replacements: A review. Plastic and Reconstructive Surgery 108(6):1713-1726.
Daley WP, Peters SB, Larsen M. 2008. Extracellular matrix dynamics in development and regenerative medicine. Journal of Cell Science 121(3):255-264.
Damour O, Gueugniaud PY, Berthinmaghit M, Rousselle P, Berthod F, Sahuc F, Collombel C. 1994. A dermal substrate made of collagen-GAG-chitosan for deep burn coverage: first clinical uses. Clinical Materials 15(4):273-276.
Dan L, Chua C-K, Leong K-F. 2010. Fibroblast response to interstitial flow: A state-of- the-art review. Biotechnology and Bioengineering 107(1):1-10.
Delvoye P, Pierard D, Noel A, Nusgens B, Foidart JM, Lapiere CM. 1988. Fibroblasts induce the assembly of the macromolecules of the basement membrane. Journal of Investigative Dermatology 90(3):276-282.
Demarchez M, Hartmann DJ, Regnier M, Asselineau D. 1992. The role of fibroblasts in dermal vascularization and remodeling of reconstructed human skin after transplantation onto the nude mouse. Transplantation 54(2):317-326.
179 Duan X, Sheardown H. 2006. Dendrimer crosslinked collagen as a corneal tissue engineering scaffold: Mechanical properties and corneal epithelial cell interactions. Biomaterials 27(26):4608-4617.
Dudas M, Wysocki A, Gelpi B, Tuan TL. 2008. Memory encoded throughout our bodies: Molecular and cellular basis of tissue regeneration. Pediatric Research 63(5):502- 512.
Dunkelman NS, Zimber MP, LeBaron RG, Pavelec R, Kwan M, Purchio AF. 1995. Cartilage production by rabbit articular chondrocytes on polyglycolic acid scaffolds in a closed bioreactor system. Biotechnology and Bioengineering 46(4):299-305.
Eady RAJ. 1988. The basement membrane: interface between the epithelium and the dermis: structural features. Archives of Dermatology 124(5):709-712.
Eady RAJ, Leigh IM, Pope FM. 1998. Anatomy and organization of human skin. In: Champion RH, Burton JL, Burns DA, Breathnach SM, editors. Rook/Wilkinson/Ebling Textbook of Dermatology. Oxford: Blackwell Science Ltd. p 37-112.
Eaglstein WH, Falanga V. 1997. Tissue engineering and the development of Apligraf®, a human skin equivalent. Clinical Therapeutics 19(5):894-905.
Eaglstein WH, Iriondo M, Laszlo K. 1995. A composite skin substitute (graftskin) for surgical wounds: a clinical experience. Dermatologic Surgery 21(10):839-843.
Ebersole GC, Anderson PM, Powell HM. 2010. Epidermal differentiation governs engineered skin biomechanics. Journal of Biomechanics 43(16):3183-3190.
Edwards C, Marks R. 1995. Evaluation of biomechanical properties of human skin. Clinics in Dermatology 13(4):375-380.
El-Ghalbzouri A, Gibbs S, Lamme E, Van Blitterswijk CA, Ponec M. 2002. Effect of fibroblasts on epidermal regeneration. British Journal of Dermatology 147(2):230-243.
El-Kassaby AW, Retik AB, Yoo JJ, Atala A. 2003. Urethral stricture repair with an off- the-shelf collagen matrix. The Journal of Urology 169(1):170-173.
Engbers-Buijtenhuijs P, Buttafoco L, Poot AA, Dijkstra PJ, de Vos RAI, Sterk LMT, Geelkerken RH, Vermes I, Feijen J. 2006. Biological characterisation of vascular grafts cultured in a bioreactor. Biomaterials 27(11):2390-2397.
Epstein FH, Singer AJ, Clark RAF. 1999. Cutaneous wound healing. New England Journal of Medicine 341(10):738-746.
180 Erdag G, Sheridan RL. 2004. Fibroblasts improve performance of cultured composite skin substitutes on athymic mice. Burns 30(4):322-328.
Fakes DW, Davies MC, Brown A, Newton JM. 1988. The surface-analysis of a plasma modified contact-lens surface by ssims. Surface and Interface Analysis 13(4):233- 236.
Falanga V, Margolis D, Alvarez O, Auletta M, Maggiacomo F, Altman M, Jensen J, Sabolinski M, Hardin-Young J, and the Human Skin Equivalent Investigators Group. 1998. Rapid healing of venous ulcers and lack of clinical rejection with an allogeneic cultured human skin equivalent. Archives of Dermatology 134(3):293- 300.
Fidkowski C, Kaazempur-Mofrad MR, Borenstein J, Vacanti JP, Langer R, Wang YD. 2005. Endothelialized microvasculature based on a biodegradable elastomer. Tissue Engineering 11(1-2):302-309.
Fleischmajer R, Utani A, MacDonald Ii ED, Perlish JS, Pan TC, Chu ML, Nomizu M, Ninomiya Y, Yamada Y. 1998. Initiation of skin basement membrane formation at the epidermo-dermal interface involves assembly of laminins through binding to cell membrane receptors. Journal of Cell Science 111(14):1929-1940.
Freed LE, Vunjak-Novakovic G, Biron RJ, Eagles DB, Lesnoy DC, Barlow SK, Langer R. 1994. Biodegradable polymer scaffolds for tissue engineering. Nature Biotechnology 12(7):689-693.
Freyman TM, Yannas IV, Gibson LJ. 2001. Cellular materials as porous scaffolds for tissue engineering. Progress in Materials Science 46(3-4):273-282.
Fuchs JR, Nasseri BA, Vacanti JP. 2001. Tissue engineering: a 21st century solution to surgical reconstruction. The Annals of Thoracic Surgery 72(2):577-591.
Fürst W, Banerjee A. 2005. Release of glutaraldehyde from an albumin-glutaraldehyde tissue adhesive causes significant in vitro and in vivo toxicity. Annals of Thoracic Surgery 79(5):1522-1529.
Furth ME, Atala A, Van Dyke ME. 2007. Smart biomaterials design for tissue engineering and regenerative medicine. Biomaterials 28(34):5068-5073.
Fusenig NE. 1994. Epithelial-mesenchymal interactions regulate keratinocyte growth and differentiation in vitro. In: Leigh IM, Lane B, Watt F, editors. The Keratinocyte Handbook. Cambridge: Cambridge University Press.
Gabridge MG, Gladd MF. 1984. Gaseous oxide toxicity evaluated with cell monolayers on collagen-coated, gas-permeable teflon membranes. Environmental Health Perspectives 54:347-352.
181 Gallico GG, Oconnor NE, Compton CC, Kehinde O, Green H. 1984. Permanent coverage of large burn wounds with autologous cultured human epithelium. New England Journal of Medicine 311(7):448-451.
Gandaglia A, Bagno A, Naso F, Spina M, Gerosa G. 2010. Cells, scaffolds and bioreactors for tissue-engineered heart valves: a journey from basic concepts to contemporary developmental innovations. European Journal of Cardio-Thoracic Surgery In Press, Corrected Proof.
Garg AK, Berg RA, Silver FH, Garg HG. 1989. Effect of proteoglycans on type I collagen fibre formation. Biomaterials 10(6):413-419.
Garner J, Brown R. 2002. Recent advances in topical agents for prehospital haemostasis. Trauma 4(4):203-209.
Ghosh K, Clark RAF. 2007. Wound repair. In: Lanza R, Langer R, Vacanti JP, editors. Principles of Tissue Engineering. 3 ed. Amsterdam: Elsevier. p 1149-1166.
Gibot L, Galbraith T, Huot J, Auger FA. 2010. A preexisting microvascular network benefits in vivo revascularization of a microvascularized tissue-engineered skin substitute. Tissue Engineering Part A 16(10):3199-3206.
Glacken MW, Fleischaker RJ, Sinskey AJ. 1983. Large-scale production of mammalian cells and their products: engineering principles and barriers to scale-up. Annals of the New York Academy of Sciences 413(1):355-372.
Glowacki J, Mizuno S. 2008. Collagen scaffolds for tissue engineering. Biopolymers 89(5):338-344.
Goldberg AM, Frazier JM. 1989. Alternatives to animals in toxicity testing. Scientific American 261(2):24-30.
Goldstein AS, Juarez TM, Helmke CD, Gustin MC, Mikos AG. 2001. Effect of convection on osteoblastic cell growth and function in biodegradable polymer foam scaffolds. Biomaterials 22(11):1279-1288.
GonÁalves A, Costa P, Rodrigues MrT, Dias IR, Reis RL, Gomes ME. 2010. Effect of flow perfusion conditions in the chondrogenic differentiation of bone marrow stromal cells cultured onto starch based biodegradable scaffolds. Acta Biomaterialia In Press, Corrected Proof.
Gooch KJ, Blunk T, Courter DL, Sieminski AL, Bursac PM, Vunjak-Novakovic G, Freed LE. 2001. IGF-I and mechanical environment interact to modulate engineered cartilage development. Biochemical and Biophysical Research Communications 286(5):909-915.
182 Grayson ACR, Shawgo RS, Johnson AM, Flynn NT, Yawen LI, Cima MJ, Langer R. 2004. A BioMEMS review: MEMS technology for physiologically integrated devices. Proceedings of the IEEE 92(1):6-21.
Grayson WL, Bhumiratana S, Cannizzaro C, Chao PHG, Lennon DP, Caplan AI, Vunjak- Novakovic G. 2008. Effects of initial seeding density and fluid perfusion rate on formation of tissue-engineered bone. Tissue Engineering Part A 14(11):1809- 1820.
Grayson WL, Fröhlich M, Yeager K, Bhumiratana S, Chan ME, Cannizzaro C, Wan LQ, Liu XS, Guo XE, Vunjak-Novakovic G. 2009. Engineering anatomically shaped human bone grafts. Proceedings of the National Academy of Sciences.
Grayson WL, Marolt D, Bhumiratana S, Fröhlich M, Guo XE, Vunjak-Novakovic G. 2010. Optimizing the medium perfusion rate in bone tissue engineering bioreactors. Biotechnology and Bioengineering:n/a-n/a.
Griffith LG, Naughton G. 2002. Tissue engineering - Current challenges and expanding opportunities. Science 295(5557):1009-+.
Hafemann B, Ensslen S, Erdmann C, Niedballa R, Zuhlke A, Ghofrani K, Kirkpatrick CJ. 1999. Use of a collagen/elastin-membrane for the tissue engineering of dermis. Burns 25(5):373-384.
Hall PA, Watt FM. 1989. Stem cells: The generation and maintenance of cellular diversity. Development 106(4):619-633.
Han B, Jaurequi J, Tang BW, Nimni ME. 2003. Proanthocyanidin: A natural crosslinking reagent for stabilizing collagen matrices. Journal of Biomedical Materials Research Part A 65A(1):118-124.
Hansbrough JF, Cooper ML, Cohen R, Spielvogel R, Greenleaf G, Bartel RL, Naughton G. 1992. Evaluation of a biodegradable matrix containing cultured human fibroblasts as a dermal replacement beneath meshed skin grafts on athymic mice. Surgery 111(4):438-446.
Harley BA, Leung JH, Silva ECCM, Gibson LJ. 2007. Mechanical characterization of collagen-glycosaminoglycan scaffolds. Acta Biomaterialia 3(4):463-474.
Haynesworth SE, Goshima J, Goldberg VM, Caplan AI. 1992. Characterization of cells with osteogenic potential from human marrow. Bone 13(1):81-88.
Hefton JM, Caldwell D, Biozes DG, Balin AK, Carter DM. 1986. Grafting of skin ulcers with cultured autologous epidermal cells. Journal of the American Academy of Dermatology 14(3):399-405.
Heldin CH, Westermark B. 1999. Mechanism of action and in vivo role of platelet- derived growth factor. Physiological Reviews 79(4):1283-1316.
183 Henry G, Garner WL. 2003. Inflammatory mediators in wound healing. Surgical Clinics of North America 83:483-+.
Hewitt CW, Marra SW, Kann BR, Tran HS, Puc MM, Chrzanowski FA, Tran JLV, Lenz SD, Cilley JH, Simonetti VA and others. 2001. BioGlue surgical adhesive for thoracic aortic repair during coagulopathy: Efficacy and histopathology. Annals of Thoracic Surgery 71(5):1609-1612.
Hornstra IK, Birge S, Starcher B, Bailey AJ, Mecham RP, Shapiro SD. 2003. Lysl oxidase is required for vascular and diaphragmatic development in mice. Journal of Biological Chemistry 278(16):14387-14393.
Hubbell JA. 1995. Biomaterials in tissue engineering. Nature Biotechnology 13(6):565- 576.
Huynh T, Abraham G, Murray J, Brockbank K, Hagen PO, Sullivan S. 1999. Remodeling of an acellular collagen graft into a physiologically responsive neovessel. Nature Biotechnology 17(11):1083-1086.
Inc MDI. 1993. Wound Care in the US: Emerging Trends, Management and New Product Development. Irvine: Medical Data International Inc.
Iwano M, Plieth D, Danoff TM, Xue C, Okada H, Neilson EG. 2002. Evidence that fibroblasts derive from epithelium during tissue fibrosis. Journal of Clinical Investigation 110(3):341-350.
Janakiraman V, Case Western Reserve University. Dept. of Chemical Engineering. 2008. Design, fabrication and characterization of bifurcating microfluidic networks for tissue-engineered products with built-in microvasculature. p 263 p.
Janakiraman V, Kienitz B, Baskaran H. 2007a. Lithography technique for topographical micropatterning of collagen-glycosaminoglycan membranes for tissue engineering applications. Journal of Medical Devices 1(3):233-7.
Janakiraman V, Mathur K, Baskaran H. 2007b. Optimal planar flow network designs for tissue engineered constructs with built-in vasculature. Annals of Biomedical Engineering 35(3):337-347.
Jarrett A. 1974. The physiology and pathophysiology of the skin. In: Jarrett A, editor. The Dermis and Dendrocytes. London: Academic Press.
Jensen MD, Wallach DFH, Lin PS. 1974. Comparative growth characteristics of vero cells on gas-permeable and conventional supports. Experimental Cell Research 84(1-2):271-281.
Jeong SI, Kwon JH, Lim JI, Cho S-W, Jung Y, Sung WJ, Kim SH, Kim YH, Lee YM, Kim B-S and others. 2005. Mechano-active tissue engineering of vascular smooth
184 muscle using pulsatile perfusion bioreactors and elastic PLCL scaffolds. Biomaterials 26(12):1405-1411.
Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. 1998. In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Experimental Cell Research 238(1):265-272.
Johnstone CC, Farley A. 2005. The physiological basics of wound healing. Nurs Stand 19(43):59-65; quiz 66.
Jones MV, Calabresi PA. 2007. Agar-gelatin for embedding tissues prior to paraffin processing. Biotechniques 42(5):569-570.
Kalyanaraman B, Boyce ST. 2007. Assessment of an automated bioreactor to propagate and harvest keratinocytes for fabrication of engineered skin substitutes. Tissue Engineering 13(5):983-993.
Kalyanaraman B, Boyce ST. 2009. Wound healing on athymic mice with engineered skin substitutes fabricated with keratinocytes harvested from an automated bioreactor. Journal of Surgical Research 152(2):296-302.
Kalyanaraman B, Supp DM, Boyce ST. 2008. Medium flow rate regulates viability and barrier function of engineered skin substitutes in perfusion culture. Tissue Engineering Part A 14(5):583-593.
Kang HW, Tabata Y, Ikada Y. 1999. Fabrication of porous gelatin scaffolds for tissue engineering. Biomaterials 20(14):1339-1344.
Kangesu T, Navsaria HA, Manek S, Fryer PR, Leigh IM, Green CJ. 1993. Kerato-dermal grafts: the importance of dermis for the in vivo growth of cultured keratinocytes. British Journal of Plastic Surgery 46(5):401-409.
Kino-Oka M, Ogawa N, Umegaki R, Taya M. 2005. Bioreactor design for successive culture of anchorage-dependent cells operated in an automated manner. Tissue Engineering 11(3-4):535-545.
Kino-Oka M, Prenosil JE. 2000. Development of an on-line monitoring system of human keratinocyte growth by image analysis and its application to bioreactor culture. Biotechnology and Bioengineering 67(2):234-239.
Kohn J. 2004. New approaches to biomaterials design. Nature Materials 3(11):745-747.
Konig A, Bruckner-Tuderman L. 1991. Epithelial-mesenchymal interactions enhance expression of collagen VII in vitro. Journal of Investigative Dermatology 96(6):803-808.
Koolwijk P, vanErck MGM, deVree WJA, Vermeer MA, Weich HA, Hanemaaijer R, vanHinsbergh VWM. 1996. Cooperative effect of TNF alpha, bFGF, and VEGF
185 on the formation of tubular structures of human microvascular endothelial cells in a fibrin matrix. Role of urokinase activity. Journal of Cell Biology 132(6):1177- 1188.
Kuberka M, Heschel I, Glasmacher B. 2002. Preparation of collagen scaffolds and their applications in tissue engineering. Biomedizinische Technik/Biomedical Engineering 47(s1a):485-487.
Kumar A, Maartens NF, Kaye AH. 2003. Evaluation of the use of BioGlue (R) in neurosurgical procedures. Journal of Clinical Neuroscience 10(6):661-664.
Ladd MR, Lee SJ, Atala A, Yoo JJ. 2009. Bioreactor maintained living skin matrix. Tissue Engineering Part A 15(4):861-868.
Lander AD. 1993. Proteoglycans. In: Kreis T, Vale R, editors. Guidebook to the Extracellular Matrix and Adhesion Proteins. Oxford: Oxford University Press. p 12-16.
Langer R, Vacanti JP. 1993. Tissue engineering. Science 260(5110):920-926.
Langer R, Vacanti JP. 1999. Tissue engineering: The challenges ahead. Scientific American 280(4):86-89.
Lanza RP, Langer R, Chick WL. 1997. Principles of tissue engineering. Austin, TX: Academic Press.
Larouche D, Tong XM, Fradette J, Coulombe PA, Germain L. 2008. Vibrissa hair bulge houses two populations of skin epithelial stem cells distinct by their keratin profile. Faseb Journal 22(5):1404-1415.
Lavik E, Langer R. 2004. Tissue engineering: current state and perspectives. Applied Microbiology and Biotechnology 65(1):1-8.
Lavker R, Sun T. 1982. Heterogeneity in epidermal basal keratinocytes: morphological and functional correlations. Science 215(4537):1239-1241.
Lee CH, Singla A, Lee Y. 2001a. Biomedical applications of collagen. International Journal of Pharmaceutics 221(1-2):1-22.
Lee CR, Grodzinsky AJ, Spector M. 2001b. The effects of cross-linking of collagen- glycosaminoglycan scaffolds on compressive stiffness, chondrocyte-mediated contraction, proliferation and biosynthesis. Biomaterials 22(23):3145-3154.
Lee KY, Mooney DJ. 2001. Hydrogels for tissue engineering. Chemical Reviews 101(7):1869-1880.
Leggat PA, Kedjarune U, Smith DR. 2004. Toxicity of cyanoacrylate adhesives and their occupational impacts for dental staff. Industrial Health 42(2):207-211.
186 Leigh IM, Purkis PE, Navsaria HA, Phillips TJ. 1987. Treatment of chronic venous ulcers with sheets of cultured allogenic keratinocytes. British Journal of Dermatology 117(5):591-597.
Lenaerts V, Gurny R. 1990. Bioadhesive drug delivery systems. Boca Raton, Fla.: CRC Press. [7], 231 p. p.
Lennon DP, Haynesworth SE, Young RG, Dennis JE, Caplan AI. 1995. A chemically defined medium supports in vitro proliferation and maintains the osteochondral potential of rat marrow-derived mesenchymal stem cells. Experimental Cell Research 219(1):211-222.
Li D, Tang T, Lu J, Dai K. 2009. Effects of flow shear stress and mass transport on the construction of a large-scale tissue-engineered bone in a perfusion bioreactor. Tissue Engineering Part A 15(10):2773-2783.
Li W-J, Jiang YJ, Tuan RS. 2008. Cell-nanofiber-based cartilage tissue engineering using improved cell seeding, growth factor, and bioreactor technologies. Tissue Engineering Part A 14(5):639-648.
Li W-J, Laurencin CT, Caterson EJ, Tuan RS, Ko FK. 2002. Electrospun nanofibrous structure: A novel scaffold for tissue engineering. Journal of Biomedical Materials Research 60(4):613-621.
Liang W-H, Kienitz BL, Penick KJ, Welter JF, Zawodzinski TA, Baskaran H. 2010. Concentrated collagen-chondroitin sulfate scaffolds for tissue engineering applications. Journal of Biomedical Materials Research Part A 94A(4):1050-1060.
Liu Y, Peterson DA, Kimura H, Schubert D. 1997. Mechanism of cellular 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction. Journal of Neurochemistry 69(2):581-593.
Loss M, Wedler V, Kunzi W, Meuli-Simmen C, Meyer VE. 2000. Artificial skin, split- thickness autograft and cultured autologous keratinocytes combined to treat a severe burn injury of 93% of TBSA. Burns 26(7):644-652.
Lutolf MP, Hubbell JA. 2005. Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nature Biotechnology 23(1):47-55.
Lyons AB, Parish CR. 1994. Determination of lymphocyte division by flow cytometry. Journal of Immunological Methods 171(1):131-137.
Ma L, Gao C, Mao Z, Zhou J, Shen J, Hu X, Han C. 2003. Collagen/chitosan porous scaffolds with improved biostability for skin tissue engineering. Biomaterials 24(26):4833-4841.
187 MacNeil S. 2007. Progress and opportunities for tissue-engineered skin. Nature 445(7130):874-880.
Madihally SV, Matthew HWT. 1999. Porous chitosan scaffolds for tissue engineering. Biomaterials 20(12):1133-1142.
Marijnissen WJCM, van Osch GJVM, Aigner J, van der Veen SW, Hollander AP, Verwoerd-Verhoef HL, Verhaar JAN. 2002. Alginate as a chondrocyte-delivery substance in combination with a non-woven scaffold for cartilage tissue engineering. Biomaterials 23(6):1511-1517.
Marks MG, Doillon C, Silver FH. 1991. Effects of fibroblasts and basic fibroblast growth factor on facilitation of dermal wound healing by type I collagen matrices. Journal of Biomedical Materials Research 25(5):683-696.
Martin I, Smith T, Wendt D. 2009. Bioreactor-based roadmap for the translation of tissue engineering strategies into clinical products. Trends in Biotechnology 27(9):495- 502.
Martin I, Wendt D, Heberer M. 2004. The role of bioreactors in tissue engineering. Trends in Biotechnology 22(2):80-86.
Martin Y, Vermette P. 2005. Bioreactors for tissue mass culture: Design, characterization, and recent advances. Biomaterials 26(35):7481-7503.
Mauck RL, Soltz MA, Wang CCB, Wong DD, Chao P-HG, Valhmu WB, Hung CT, Ateshian GA. 2000. Functional tissue engineering of articular cartilage through dynamic loading of xhondrocyte-seeded agarose gels. Journal of Biomechanical Engineering 122(3):252-260.
Medalie DA, Eming SA, Tompkins RG, Yarmush ML, Krueger GG, Morgan JR. 1996. Evaluation of human skin reconstituted from composite grafts of cultured keratinocytes and human acellular dermis transplanted to athymic mice. Journal of Investigative Dermatology 107(1):121-127.
Meinel L, Karageorgiou V, Fajardo R, Snyder B, Shinde-Patil V, Zichner L, Kaplan D, Langer R, Vunjak-Novakovic G. 2004. Bone tissue engineering using human mesenchymal stem cells: effects of scaffold material and medium flow. Annals of Biomedical Engineering 32(1):112-122.
Middelkoop E, van den Bogaerdt A, van Zuijlen P, Lamme E, Ulrich M. Living Skin Equivalents: Fibroblasts. European Tissue Repair Society.
Migneault I, Dartiguenave C, Bertrand MJ, Waldron KC. 2004. Glutaraldehyde: behavior in aqueous solution, reaction with proteins, and application to enzyme crosslinking. Biotechniques 37(5):790-802.
188 Minuth WW, Schumacher K, Strehl R, Kloth S. 2000. Physiological and cell biological aspects of perfusion culture technique employed to generate differentiated tissues for long term biomaterial testing and tissue engineering. Journal of Biomaterials Science-Polymer Edition 11(5):495-522.
Moiemen NS, Vlachou E, Staiano JJ, Thawy Y, Frame JD. 2006. Reconstructive surgery with Integra dermal regeneration template: Histologic study, clinical evaluation, and current practice. Plastic and Reconstructive Surgery 117(7S):160S-174S 10.1097/01.prs.0000222609.40461.68.
Montagna W, Parakkal PF. 1974. The Structure and Function of Skin. New York: Academic Press.
Morritt AN, Bortolotto SK, Dilley RJ, Han XL, Kompa AR, McCombe D, Wright CE, Itescu S, Angus JA, Morrison WA. 2007. Cardiac tissue engineering in an in vivo vascularized chamber. Circulation 115(3):353-360.
Mosmann T. 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65(1- 2):55-63.
Munster AM. 1996. Cultured skin for massive burns - A prospective, controlled trial. Annals of Surgery 224(3):372-375.
Murphy GE. 2004. Histology of the skin. In: Elder DE, Elenitsas R, Johnson BL, Murphy GE, editors. Lever's Histopathology of the Skin. 9 ed. Philadelphia: Lippincott Williams & Wilkins. p 9-58.
Murphy GF, Orgill DP, Yannas IV. 1990. Partial dermal regeneration is induced by biodegradable collagen glycosaminoglycan grafts. Laboratory Investigation 62(3):305-313.
Muschler GF, Nakamoto C, Griffith LG. 2004. Engineering principles of clinical cell- based tissue engineering. Journal of Bone and Joint Surgery 86(7):1541-1558.
Nation JL. 1983. A new method using hexamethyldisilazane for preparation of soft insect tissues for scanning electron microscopy. Stain Technology 58(6):347-351.
Nelson CM, Bissell MJ. 2005. Modeling dynamic reciprocity: Engineering three- dimensional culture models of breast architecture, function, and neoplastic transformation. Seminars in Cancer Biology 15(5):342-352.
Nelson CM, Bissell MJ. 2006. Of extracellular matrix, scaffolds, and signaling: Tissue architecture regulates development, homeostasis, and cancer. Annual Review of Cell and Developmental Biology 22:287-309.
Nerem RM, Sambanis A. 1995. Tissue engineering: from biology to biological substitutes. Tissue Engineering 1(1):3-13.
189 Niklason LE, Gao J, Abbott WM, Hirschi KK, Houser S, Marini R, Langer R. 1999. Functional arteries grown in vitro. Science 284(5413):489-493.
Nomura S, Hiltner A, Lando JB, Baer E. 1977. Interaction of water with native collagen. Biopolymers 16(2):231-246.
O'Brien FJ, Harley BA, Yannas IV, Gibson L. 2004. Influence of freezing rate on pore structure in freeze-dried collagen-GAG scaffolds. Biomaterials 25(6):1077-1086.
O'Brien FJ, Harley BA, Yannas IV, Gibson LJ. 2005. The effect of pore size on cell adhesion in collagen-GAG scaffolds. Biomaterials 26(4):433-441.
O'Connor NE, Mulliken JB, Banksschlegel S, Kehinde O, Green H. 1981. Grafting of burns with cultured epithelium prepared from autologous epidermal cells. Lancet 1(8211):75-78.
Odland GF. 1991. Structure of the skin. In: Goldsmith LA, editor. Physiology, Biochemistry, and Molecular Biology of the Skin. New York: Oxford University Press. p 3-62.
Okamoto E, Kitano Y. 1993. Expression of basement membrane components in skin equivalents: Influence of dermal fibroblasts. Journal of Dermatological Science 5(2):81-88.
Oldberg A, Ruoslahti E. 1982. Interactions between chondroitin sulfate proteoglycan, fibronectin, and collagen. Journal of Biological Chemistry 257(9):4859-4863.
Orban JM, Wilson LB, Kofroth JA, El-Kurdi MS, Maul TM, Vorp DA. 2004. Crosslinking of collagen gels by transglutaminase. Journal of Biomedical Materials Research Part A 68A(4):756-762.
Otani Y, Tabata Y, Ikada Y. 1996a. A new biological glue from gelatin and poly(L- glutamic acid). Journal of Biomedical Materials Research 31(2):157-166.
Otani Y, Tabata Y, Ikada Y. 1996b. Rapidly curable biological glue composed of gelatin and poly(L-glutamic acid). Biomaterials 17(14):1387-1391.
Owen MJ, Smith PJ. 1994. Plasma Treatment of Polydimethylsiloxane. Journal of Adhesion Science and Technology 8(10):1063-1075.
Oz MC, Cosgrove DM, III, Badduke BR, Hill JD, Flannery MR, Palumbo R, Topic N. 2000. Controlled clinical trial of a novel hemostatic agent in cardiac surgery. Annals of Thoracic Surgery 69(5):1376-1382.
Pandya AN, Woodward B, Parkhouse N. 1998. The use of cultured autologous keratinocytes with integra in the resurfacing of acute burns. Plastic and Reconstructive Surgery 102(3):825-828.
190 Papini S, Cecchetti D, Campani D, Fitzgerald W, Grivel JC, Chen S, Margolis L, Revoltella RP. 2003. Isolation and clonal analysis of human epidermal keratinocyte stem cells in long-term culture. Stem Cells (Miamisburg) 21(4):481- 494.
Parenteau NL, Bilbo P, Nolte CJM, Mason VS, Rosenberg M. 1992. The organotypic culture of human skin keratinocytes and fibroblasts to achieve form and function. Cytotechnology 9(1-3):163-171.
Pathi P, Ma T, Locke BR. 2005. Role of nutrient supply on cell growth in bioreactor design for tissue engineering of hematopoietic cells. Biotechnology and Bioengineering 89(7):743-758.
Pazzano D, Mercier KA, Moran JM, Fong SS, DiBiasio DD, Rulfs JX, Kohles SS, Bonassar LJ. 2000. Comparison of chondrogensis in static and perfused bioreactor culture. Biotechnology Progress 16(5):893-896.
Pellegrini G, Ranno R, Stracuzzi G, Bondanza S, Guerra L, Zambruno G, Micali G, De Luca M. 1999. The control of epidermal stem cells (holoclones) in the treatment of massive full-thickness burns with autologous keratinocytes cultured on fibrin. Transplantation 68(6):868-879.
Penick KJ, Solchaga LA, Berilla JA, Welter JF. 2005a. Performance of polyoxymethylene plastic (POM) as a component of a tissue engineering bioreactor. Journal of Biomedical Materials Research Part A 75A(1):168-174.
Penick KJ, Solchaga LA, Welter JF. 2005b. High-throughput aggregate culture system to assess the chondrogenic potential of mesenchymal stem cells. Biotechniques 39(5):687-691.
Pham C, Greenwood J, Cleland H, Woodruff P, Maddern G. 2007. Bioengineered skin substitutes for the management of burns: A systematic review. Burns 33(8):946- 957.
Phillips TJ. 1993. Biologic skin substitutes. Journal of Dermatologic Surgery and Oncology 19(8):794-800.
Phillips TJ, Gilchrest BA. 1991. Cultured epidermal allografts as biological wound dressings. Progress in Clinical and Biological Research 365:77-94.
Pomaha! B, Svensjö T, Yao F, Brown H, Eriksson E. 1998. Tissue engineering of skin. Critical Reviews in Oral Biology & Medicine 9(3):333-344.
Ponticiello MS, Schinagl RM, Kadiyala S, Barry FP. 2000. Gelatin-based resorbable sponge as a carrier matrix for human mesenchymal stem cells in cartilage regeneration therapy. Journal of Biomedical Materials Research 52(2):246-255.
191 Pörtner R, Nagel-Heyer S, Goepfert C, Adamietz P, Meenen NM. 2005. Bioreactor design for tissue engineering. Journal of Bioscience and Bioengineering 100(3):235-245.
Potten CS. 1974. The epidermal proliferative unit: the possible role of the central basal cell. Cell Proliferation 7(1):77-88.
Potten CS. 1981. Cell replacement in epidermis (keratopoiesis) via discrete units of proliferation. International review of cytology 69:271-318.
Poumay Y, Pittelkow MR. 1995. Cell density and culture factors regulate keratinocyte commitment to differentiation and expression of suprabasal K1/K10 keratin. Journal of Investigative Dermatology 104(2):271-276.
Prenosil JE, Kino-Oka M. 1999. Computer controlled bioreactor for large-scale production of cultured skin grafts. Annals of the New York Academy of Sciences 875(1):386-397.
Prenosil JE, Villeneuve PE. 1998. Automated production of cultured epidermal autografts and sub-confluent epidermal autografts in a computer controlled bioreactor. Biotechnology and Bioengineering 59(6):679-683.
Price GM, Wong KHK, Truslow JG, Leung AD, Acharya C, Tien J. 2010. Effect of mechanical factors on the function of engineered human blood microvessels in microfluidic collagen gels. Biomaterials 31(24):6182-6189.
Prunieras M, Regnier M, Woodley D. 1983. Methods for cultivation of keratinocytes with an air-liquid interface. Journal of Investigative Dermatology 81(s1):28s-33s.
Ralston DR, Laytont C, Dalley AJ, Boyce SG, Freedlander E, MacNeil S. 1997. Keratinocytes contract human dermal extracellular matrix and reduce soluble fibronectin production by fibroblasts in a skin composite model. British Journal of Plastic Surgery 50(6):408-415.
Ranger WR, Halpin D, Sawhney AS, Lyman M, Locicero J. 1997. Pneumostasis of experimental air leaks with a new photopolymerized synthetic tissue sealant. American surgeon 63(9):788-795.
Ratcliffe A, Niklason LE. 2002. Bioreactors and bioprocessing for tissue engineering. Annals of the New York Academy of Sciences 961(1):210-215.
Reece TB, Maxey TS, Kron IL. 2001. A prospectus on tissue adhesives. American Journal of Surgery 182(2):40S-44S.
Renkens KL, Payner TD, Leipzig TJ, Feuer H, Morone MA, Koers JM, Lawson KJ, Lentz R, Shuey H, Conaway GL and others. 2001. A multicenter, prospective, randomized trial evaluating a new hemostatic agent for spinal surgery. Spine 26(15):1645-1650.
192 Rheinwald JG, Green H. 1975. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6(3):331-344.
Riches DWH. 1996. Macrophage involvement in wound repair, remodeling and fibrosis. In: Clark RAF, editor. The Molecular and Cellular Biology of Wound Repair. 2 ed. New York: Plenum Press. p 95-141.
Rodkey WG, Steadman JR, Li ST. 1999. A clinical study of collagen meniscus implants to restore the injured meniscus. Clinical Orthopaedics and Related Research(367):S281-S292.
Roeder BA, Kokini K, Sturgis JE, Robinson JP, Voytik-Harbin SL. 2002. Tensile mechanical properties of three-dimensional type I collagen extracellular matrices with varied microstructure. Journal of Biomechanical Engineering 124(2):214- 222.
Ronfard V, Rives JM, Neveux Y, Carsin H, Barrandon Y. 2000. Long-term regeneration of human epidermis on third degree burns transplanted with autologous cultured epithelium grown on a fibrin matrix. Transplantation 70(11):1588-1598.
Rosenblatt J, Devereux B, Wallace DG. 1994. Injectable collagen as a pH-sensitive hydrogel Biomaterials 15(12):985-995.
Rozga J, Morsiani E, Lepage E, Moscioni AD, Giorgio T, Demetriou AA. 1994. Isolated hepatocytes in a bioartificial liver: a single group view and experience. Biotechnology and Bioengineering 43(7):645-653.
Rue LW, Cioffi WG, McManus WF, Pruitt BA, Phillips LG, Meyer AA, Tompkins RG, Saffle JR. 1993. Wound closure and outcome in extensively burned patients treated with cultured autologous keratinocytes. Journal of Trauma-Injury Infection and Critical Care 34(5):662-668.
Ruoslahti E, Pierschbacher M, Engvall E, Oldberg A, Hayman EG. 1982. Molecular and biological interactions of fibronectin. Journal of Investigative Dermatology 79:S65-S68.
Ruszczak Z. 2003. Effect of collagen matrices on dermal wound healing. Advanced Drug Delivery Reviews 55(12):1595-1611.
Saito H, Taguchi T, Aoki H, Murabayashi S, Mitamura Y, Tanaka J, Tateishi T. 2007. pH-responsive crosslinkers swelling behavior of collagen gels prepared by novel based on naturally derived di- or tricarboxylic acids. Acta Biomaterialia 3(1):89- 94.
Sakon M, Monden M, Gotoh M, Kobayashi K, Kambayashi J, Mori T, Okamura J. 1989. Use of microcrystalline collagen powder and fibrinogen tissue adhesive for hemostasis and prevention of rebleeding in patients with hepatocellular carcinoma
193 associated with cirrhosis of the liver. Surgery Gynecology & Obstetrics 168(5):453-454.
Sasaki N, Odajima S. 1996. Stress-strain curve and Young's modulus of a collagen molecule as determined by the X-ray diffraction technique. Journal of Biomechanics 29(5):655-658.
Schmidt CE, Baier JM. 2000. Acellular vascular tissues: natural biomaterials for tissue repair and tissue engineering. Biomaterials 21(22):2215-2231.
Schnabel M, Marlovits S, Eckhoff G, Fichtel I, Gotzen L, Vecsei V, Schlegel J. 2002. Dedifferentiation-associated changes in morphology and gene expression in primary human articular chondrocytes in cell culture. Osteoarthritis and Cartilage 10(1):62-70.
Segal N, Andriani F, Pfeiffer L, Kamath P, Lin N, Satyamurthy K, Egles C, Garlick JA. 2008. The basement membrane microenvironment directs the normalization and survival of bioengineered human skin equivalents. Matrix Biology 27(3):163-170.
Sekine T, Nakamura T, Shimizu Y, Ueda H, Matsumoto K, Takimoto Y, Kiyotani T. 2001. A new type of surgical adhesive made from porcine collagen and polyglutamic acid. Journal of Biomedical Materials Research 54(2):305-310.
Shachar M, Cohen S. 2003. Cardiac tissue engineering, ex-vivo: design principles in biomaterials and bioreactors. Heart Failure Reviews 8(3):271-276.
Shastri VP. 2009. In Vvvo engineering of tissues: biological considerations, challenges, strategies, and future Directions. Advanced Materials 21(32-33):3246-3254.
Shatford RA, Nyberg SL, Meier SJ, White JG, Payne WD, Hu WS, Cerra FB. 1992. Hepatocyte function in a hollow fiber bioreactor: a potential bioartificial liver. Journal of Surgical Research 53(6):549-557.
Shepherd DET, Seedhom BB. 1999. The 'instantaneous' compressive modulus of human articular cartilage in joints of the lower limb. Rheumatology 38(2):124-132.
Shi H, Han C, Mao Z, Ma L, Gao C. 2008. Enhanced angiogenesis in porous collagen- chitosan scaffolds loaded with angiogenin. Tissue Engineering Part A 14(11):1775-1785.
Shimizu M, Takakuwa Y, Nitta S. 1995. Study of stimulation-secretion coupling in a flow culture system: periodic secretion of hepatocyte growth factor by interleukin-1 alpha-stimulated human embryonic lung fibroblasts. Biochimica Et Biophysica Acta 1244(2-3):357-362.
Shin M, Matsuda K, Ishii O, Terai H, Kaazempur-Mofrad M, Borenstein J, Detmar M, Vacanti JP. 2004. Endothelialized networks with a vascular geometry in microfabricated poly(dimethyl siloxane). Biomedical Microdevices 6(4):269-278.
194 Sieminski AL, Padera RF, Blunk T, Gooch KJ. 2002. Systemic delivery of human growth hormone using genetically modified tissue-engineered microvascular networks: Prolonged delivery and endothelial survival with inclusion of nonendothelial cells. Tissue Engineering 8(6):1057-1069.
Sikavitsas VI, Bancroft GN, Holtorf HL, Jansen JA, Mikos AG. 2003. Mineralized matrix deposition by marrow stromal osteoblasts in 3D perfusion culture increases with increasing fluid shear forces. Proceedings of the National Academy of Sciences of the United States of America 100(25):14683-14688.
Sikavitsas VI, Bancroft GN, Mikos AG. 2002. Formation of three-dimensional cell/polymer constructs for bone tissue engineering in a spinner flask and a rotating wall vessel bioreactor. Journal of Biomedical Materials Research 62(1):136-148.
Skeggs LT, Leonards JR. 1948. Studies on an artificial kidney: I. preliminary results with a new type of continuous dialyzer. Science 108(2800):212-213.
Slupsky JR, Kalbas M, Willuweit A, Henn V, Kroczek RA, Muller-Berghaus G. 1998. Activated platelets induce tissue factor expression on human umbilical vein endothelial cells by ligation of CD40. Thrombosis and Haemostasis 80(6):1008- 1014.
Solchaga LA, Goldberg VM, Mishra R, Caplan AI, Welter JF. 2004a. FGF-2 modifies the gene expression profile of bone marrow-derived human mesenchymal stem cells. Transactions of the Orthopaedic Research Society 29:777.
Solchaga LA, Penick KJ, Welter JF. 2004b. A "manual" mosaicking approach to generating large, high-resolution digital images of histological sections. Proceedings of the Royal Microscopical Society 39:313-320.
Solchaga LA, Tognana E, Penick K, Baskaran H, Goldberg VM, Caplan AI, Welter JF. 2006. A rapid seeding technique for the assembly of large cell/scaffold composite constructs. Tissue Engineering 12(7):1851-1863.
Solchaga LA, Welter JF, Lennon DP, Caplan AI. 2004c. Generation of pluripotent stem cells and their differentiation to the chondrocytic phenotype. Cartilage and Osteoarthritis, Vol 1: Cellular and Molecular Tools 100:53-67.
Song K, Liu T, Cui Z, Li X, Ma X. 2008. Three-dimensional fabrication of engineered bone with human bio-derived bone scaffolds in a rotating wall vessel bioreactor. Journal of Biomedical Materials Research Part A 86A(2):323-332.
Soo C, Rahbar G, Moy RL. 1993. The immunogenicity of bovine collagen implants. Journal of Dermatologic Surgery and Oncology 19(5):431-434.
195 Spravchikov N, Sizyakov G, Gartsbein M, Accili D, Tennenbaum T, Wertheimer E. 2001. Glucose effects on skin keratinocytes. Implications for diabetes skin complications. Diabetes 50(7):1627-1635.
Springer IN, Nocini PF, Schlegel KA, De Santis D, Park J, Warnke PH, Terheyden H, Zimmermann R, Chiarini L, Gardner K and others. 2006. Two techniques for the preparation of cell-scaffold constructs suitable for sinus augmentation: Steps into clinical application. Tissue Engineering 12(9):2649-2656.
Stiefel D, Schiestl C, Meuli M. 2010. Integra Artificial Skin® for burn scar revision in adolescents and children. Burns 36(1):114-120.
Strom SC, Chowdhury JR, Fox IJ. 1999. Hepatocyte transplantation for the treatment of human disease. Seminars in Liver Disease 19(1):39-48.
Sun T, Norton D, Haycock JW, Ryan AJ, MacNeil S. 2005. Development of a closed bioreactor system for culture of tissue-engineered skin at an air-liquid interface. Tissue Engineering 11(11-12):1824-1831.
Sung HW, Huang DM, Chang WH, Huang RN, Hsu JC. 1999. Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhesives: In vitro study. Journal of Biomedical Materials Research 46(4):520-530.
Supp DM, Supp AP, Bell SM, Boyce ST. 2000. Enhanced vascularization of cultured skin substitutes genetically modified to overexpress vascular endothelial growth factor. Journal of Investigative Dermatology 114(1):5-13.
Taguchi T, Saito H, Iwasashi M, Sakane M, Kakinoki S, Ochiai N, Tanaka J. 2007. Development of a novel glue consisting of naturally-derived biomolecules: Citric acid and human serum albumin. Journal of Nanoscience and Nanotechnology 7(3):742-747.
Talorete TPN, Bouaziz M, Sayadi S, Isoda H. 2006. Influence of medium type and serum on MTT reduction by flavonoids in the absence of cells. Cytotechnology 52(3):189-198.
Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. 1998. Embryonic stem cell lines derived from human blastocysts. Science 282(5391):1145-1147.
Tran KT, Griffith L, Wells A. 2004. Extracellular matrix signaling through growth factor receptors during wound healing. Wound Repair and Regeneration 12(3):262-268.
Tremblay P-L, Hudon V, Berthod F, Germain L, Auger FA. 2005. Inosculation of tissue- engineered capillaries with the host's vasculature in a reconstructed skin transplanted on mice. American Journal of Transplantation 5(5):1002-1010.
196 Tseng YC, Tabata Y, Hyon SH, Ikada Y. 1990. In vitro toxicity test of 2-cyanoacrylate polymers by cell culture method. Journal of Biomedical Materials Research 24(10):1355-1367.
Unger MA, Chou H-P, Thorsen T, Scherer A, Quake SR. 2000. Monolithic microfabricated valves and pumps by multilayer soft lithography. Science 288(5463):113-116.
Vacanti JP, Langer R. 1999. Tissue engineering: the design and fabrication of living replacement devices for surgical reconstruction and transplantation. Lancet 354:SI32-SI34. van Wachem PB, Plantinga JA, Wissink MJB, Beernink R, Poot AA, Engbers GHM, Beugeling T, van Aken WG, Feijen J, van Luyn MJA. 2001. In vivo biocompatibility of carbodiimide-crosslinked collagen matrices: Effects of crosslink density, heparin immobilization, and bFGF loading. Journal of Biomedical Materials Research 55(3):368-378.
Vedung S, Hedlund A. 1993. Fibrin glue: Its use for skin grafting of contaminated burn wounds in areas difficult to immobilize. Journal of Burn Care and Rehabilitation 14(3):356-358.
Velazquez OC, Snyder R, Liu Z-J, Fairman RM, Herlyn M. 2002. Fibroblast-dependent differentiation of human microvascular endothelial cells into capillary-like, three- dimensional networks. The FASEB Journal.
Veronda DR, Westmann RA. 1970. Mechanical characterization of skin-finite deformations. Journal of Biomechanics 3(1):111-124.
Vibe P, Pless J. 1983. A new method of skin-graft adhesion. Scandinavian Journal of Plastic and Reconstructive Surgery and Hand Surgery 17(3):263-264.
Wainwright D, Madden M, Luterman A, Hunt J, Monafo W, Heimbach D, Kagan R, Sittig K, Dimick A, Herndon D. 1996. Clinical evaluation of an acellular allograft dermal matrix in full-thickness burns. Journal of Burn Care & Research 17(2):124-136.
Wainwright DJ. 1995. Use of an acellular allograft dermal matrix (AlloDerm) in the management of full-thickness burns. Burns 21(4):243-248.
Wallace DG, Rosenblatt J. 2003. Collagen gel systems for sustained delivery and tissue engineering. Advanced Drug Delivery Reviews 55(12):1631-1649.
Wang D, Liu W, Han B, Xu R. 2005. The bioreactor: a powerful tool for large-scale culture of animal cells. Current Pharmaceutical Biotechnology 6(5):397-403.
Watt FM. 1989. Terminal differentiation of epidermal keratinocytes. Current Opinion in Cell Biology 1(6):1107-15.
197 Watt FM, Green H. 1982. Stratification and terminal differentiation of cultured epidermal cells. Nature 295(5848):434-436.
Welter JF, Solchaga LA, Penick K, Baskaran H, Goldberg VM, Caplan AI, Berilla J. 2005. A modular bioreactor system for cartilage tissue engineering. Transactions of the Orthopaedic Research Society 30:1771.
Welter JF, Solchaga LA, Penick KJ. 2007. Simplification of aggregate culture of human mesenchymal stem cells as a chondrogenic screening assay. Biotechniques 42(6):732-+.
Werner S, Grose R. 2003. Regulation of wound healing by growth factors and cytokines. Physiological Reviews 83(3):835-870.
Whitesides GM, Ostuni E, Takayama S, Jiang XY, Ingber DE. 2001. Soft lithography in biology and biochemistry. Annual Review of Biomedical Engineering 3:335-373.
Wikner NE, Dixit VM, Frazier WA, Clark RAF. 1987. Human keratinocytes synthesize and secrete the extracellular matrix protein, thrombospondin. Journal of Investigative Dermatology 88(2):207-211.
Wilkins LM, Watson SR, Prosky SJ, Meunier SF, Parenteau NL. 1994. Development of a bilayered living skin construct for clinical applications. Biotechnology and Bioengineering 43(8):747-756.
Winsor T, Burch GE. 1944. Differential roles of layers of human epigastric skin on diffusion rate of water. Archives of Internal Medicine 74(6):428-436.
Wong M. 2004. Alginates in tissue engineering. Biopolymer Methods in Tissue Engineering 238:77-86.
Wong T, McGrath JA, Navsaria H. 2007. The role of fibroblasts in tissue engineering and regeneration. British Journal of Dermatology 156(6):1149-1155.
Wynn JJ, Alexander CE. 2011. Increasing organ donation and transplantation: the U.S. experience over the past decade. Transplant International 24(4):324-332.
Xia YN, Whitesides GM. 1998. Soft lithography. Annual Review of Materials Science 28:153-184.
Yang SF, Leong KF, Du ZH, Chua CK. 2001. The design of scaffolds for use in tissue engineering. Part 1. Traditional factors. Tissue Engineering 7(6):679-689.
Yannas IV. 2005. Similarities and differences between induced organ regeneration in adults and early foetal regeneration. Journal of the Royal Society Interface 2(5):403-417.
198 Yannas IV, Burke JF, Gordon PL, Huang C, Rubenstein RH. 1980. Design of an artificial skin. II. Control of chemical composition. Journal of Biomedical Materials Research 14(2):107-132.
Yaoita H, Foidart J-M, Katz SI. 1978. Localization of the collagenous component in skin basement membrane. Journal of Investigative Dermatology 70(4):191-193.
Yasumizu R, Sugiura K, Iwai H, Inaba M, Makino S, Ida T, Imura H, Hamashima Y, Good RA, Ikehara S. 1987. Treatment of type 1 diabetes mellitus in non-obese diabetic mice by transplantation of allogeneic bone marrow and pancreatic tissue. Proceedings of the National Academy of Sciences of the United States of America 84(18):6555-6557.
Yeong WY, Chua CK, Leong KF, Chandrasekaran M, Lee MW. 2007. Comparison of drying methods in the fabrication of collagen scaffold via indirect rapid prototyping. Journal of Biomedical Materials Research Part B-Applied Biomaterials 82B(1):260-266.
Young DM, Greulich KM, Weier HG. 1996. Species-specific in situ hybridization with fluorochrome-labeled DNA probes to study vascularization of human skin grafts on athymic mice. Journal of Burn Care and Rehabilitation 17(4):305-310.
Zaman MA, Martin GP, Rees GD. 2010. Bioadhesion and retention of non-aqueous delivery systems in a dental hard tissue model. Journal of Dentistry 38(9):757-764.
Zhang ZD, Cox G. 1996. MTT assay overestimates human airway smooth muscle cell number in culture. Biochemistry and Molecular Biology International 38(3):431- 436.
Zhao F, Ma T. 2005. Perfusion bioreactor system for human mesenchymal stem cell tissue engineering: Dynamic cell seeding and construct development. Biotechnology and Bioengineering 91(4):482-493.
199