1 KIDNEY TISSUE ENGINEERING USING DECELLULARIZED ORGANS AND PLURIPOTENT STEM CELLS By MATTHEW JAMES WILLIAMS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIRE MENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Matthew James Williams
3 To Jennifer, Connor, and Caden
4 ACKNOWLEDGMENTS This tissue engineering research project is exactl y what I had envisioned when I enrolled at the University of Florida, and through it I have had the chance to meet and interact with extraordinary people, without which success would not have been possible. First, I need to thank my mentor, Dr. Chris Batic h, for introducing this project and for his tremendous patience and unwavering encouragement and support. I would also extend gratitude to the rest of my supervisory committee, Dr. Ed Ross for his enthusiasm and commitment to success, Dr. Nao Terada for hi s pragmatic advice, and Dr. Rich Dickinson and Dr. Ben Keselowsky for their ideas and constructive dialog. Others that have been instrumental to the project include Dr. Ellison and Dr. Adin for expert instruction in how to work with animals, Dr. Clapp for patient explanation of histology interpretation, and Dr. Petersen for collaboration in liver applications. Charlie Hamazaki taught me cell culture and analysis. Linda Archer not only assisted with animal surgeries but was an amazing facilitator in the labo ratory. Marda Jorgensen was an invaluable resource for histology processing. Nakato, Taili, Tom, Brad, Albina, Maribel, Matt, Laurent, and Alan all made a significant contribution this project. Ed Botchwey from the University of Virginia and Steve Miller f rom RTI Biologics have both been very supportive of my work, and I give thanks to them. Finally, I would like to acknowledge Jan Ryerse and the Saint Louis University Research Microscopy Core for use of their microscope. This project was partially funded b y the University of Florida Office of Research through the Research Opportunity Seed Fund. Throughout this journey, I have been incredibly fortunate to establish friendships with those involved in the project, in the laboratories, and in the classrooms. I give thanks for their camaraderie and encouragement. I also want to recognize my fraternity brothers for their companionship and extraordinary friends Chris, JP, Kent, Matt, Tom, and Vishal for good times
5 and unconditional support. Finally, I could not hav e completed this challenging research without the love of my parents, my brother Brian, the LaPlante family, and the foundation of my strength, my beautiful and devoted wife Jennifer and my precious boys Connor and Caden. The data reported here have been s upplied by the United States Renal Data System (USRDS). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy or interpretation of the U.S. government. This work was suppor ted in part by Health Resources and Services Administration contract 234 2005 37011C. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does men tion of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ........... 4 LIST OF TABLES ................................ ................................ ................................ ...................... 9 LIST OF FIGURES ................................ ................................ ................................ .................. 10 LIST OF ABBREVIATIONS ................................ ................................ ................................ .... 12 ABSTRACT ................................ ................................ ................................ ............................. 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ............. 17 2 BACKGROUND ................................ ................................ ................................ ............... 21 End Stage Renal Disease and Renal Failure ................................ ................................ ........ 21 The Kidney ................................ ................................ ................................ ........................ 21 Anatomy and Physiology ................................ ................................ ............................. 22 Extracellular Matrix ................................ ................................ ................................ .... 22 Organogenesis ................................ ................................ ................................ ............. 25 Kidney Repair ................................ ................................ ................................ ............. 29 Current Approaches to Renal Tissue Engineering ................................ ............................... 31 Kidney Xenotransplantation ................................ ................................ ........................ 31 Cell based Therapies ................................ ................................ ................................ ... 32 Tissue Structure Generation ................................ ................................ ........................ 34 Implantable Artificial Constructs ................................ ................................ ................. 35 Tissue Engineering Scaffolds ................................ ................................ ............................. 37 Extracellular Matrix Scaffolds ................................ ................................ ..................... 39 Tissue Decellularization ................................ ................................ .............................. 40 Cells for Kidney Tissue Engineering ................................ ................................ .................. 41 Adult Stem Cells ................................ ................................ ................................ ......... 42 Fetal Stem Cells ................................ ................................ ................................ .......... 43 Pluripotent Stem Cells ................................ ................................ ................................ 44 3 KIDN EY DECELLULARIZATION ................................ ................................ .................. 48 Experimental Plan ................................ ................................ ................................ .............. 48 Introduction ................................ ................................ ................................ ................. 48 Organ Harvest ................................ ................................ ................................ ............. 50 Kidney Decellularization ................................ ................................ ............................. 50 Scaffold Analysis ................................ ................................ ................................ ........ 51 Results ................................ ................................ ................................ ............................... 52 Scaffold Production ................................ ................................ ................................ ..... 52
7 Scaffold Evaluation ................................ ................................ ................................ ..... 52 Procedure Improve ment ................................ ................................ ................................ ..... 54 Elevated Reservoir Perfusion Apparatus ................................ ................................ ...... 54 Decellularization Protocol Improvement ................................ ................................ ..... 55 Supplementary Scaffold Analysis ................................ ................................ ....................... 62 Scanning E lectronic M icroscopy ................................ ................................ ................. 62 ECM Immunohistochemistry ................................ ................................ ....................... 62 Cell Removal Immunohistochemistry ................................ ................................ .......... 64 Discussion ................................ ................................ ................................ .......................... 65 Tissue Engineering Scaff old ................................ ................................ ........................ 65 Acellular Kidney Analysis ................................ ................................ ........................... 66 Decellularization Protocol ................................ ................................ ........................... 76 C onclusion ................................ ................................ ................................ .................. 90 4 SCAFFOLD CULTURE WITH PLURIPOTENT STEM CELLS ................................ ...... 92 Experimental Plan ................................ ................................ ................................ .............. 92 Introduction ................................ ................................ ................................ ................. 92 Cell Delivery and Scaffold Culture ................................ ................................ .............. 95 Cultured Scaffold Analysis ................................ ................................ .......................... 96 Results ................................ ................................ ................................ ............................... 97 Whole Scaffold Culture ................................ ................................ ............................... 97 Perfused Scaffold Culture ................................ ................................ ............................ 99 Sectioned Scaffold Culture ................................ ................................ ........................ 101 Embryoid Body Culture ................................ ................................ ............................ 112 Discussion ................................ ................................ ................................ ........................ 114 Tissue Engineering Scaffold Cultured with Embryonic Stem Cells ............................ 114 Cell Distribution ................................ ................................ ................................ ........ 115 Cel l Proliferation ................................ ................................ ................................ ....... 118 Cell Differentiation ................................ ................................ ................................ ... 119 Embryoid Body Comparison ................................ ................................ ..................... 122 Endothelial Differentiation ................................ ................................ ........................ 124 Perfusion Culture ................................ ................................ ................................ ...... 125 Culture Media Modifications ................................ ................................ ..................... 127 Conclusion ................................ ................................ ................................ ................ 130 5 CONCLUSION AND FUTURE STUDIES ................................ ................................ ...... 132 Research Summary ................................ ................................ ................................ ........... 132 Promoting Differentiation ................................ ................................ ................................ 133 Scaffold Culture ................................ ................................ ................................ ............... 134 Cell A lternatives ................................ ................................ ................................ .............. 135 Xenogenic Scaffolds ................................ ................................ ................................ ........ 136 6 MATERIALS AND METHODS ................................ ................................ ...................... 139 Organ Harvest ................................ ................................ ................................ .................. 139
8 Organ Decellularization ................................ ................................ ................................ ... 140 Organ Perfusion ................................ ................................ ................................ ........ 140 Decellularization Protocols ................................ ................................ ........................ 141 In Vitro Embryonic Stem Cell Engraftment and Culture ................................ ................... 148 Embryonic Stem Cells ................................ ................................ ............................... 148 Cell Engraftment and Culture ................................ ................................ .................... 149 Embryoid Bodies ................................ ................................ ................................ .............. 150 Histochemistry ................................ ................................ ................................ ................. 151 Tissue Preparation ................................ ................................ ................................ ..... 151 Hematoxylin and Eosin Staining ................................ ................................ ................ 151 Immunohistochemistry ................................ ................................ .............................. 153 Scanning Electron Microscopy ................................ ................................ ......................... 154 APPENDIX A DECELLULARIZATION AND PERFUSION CULTURE APPARATUSES .................. 156 B SU PPLEMENTAL HISTOLOGY ................................ ................................ .................... 160 LIST OF REFERENCES ................................ ................................ ................................ ........ 163 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ... 180
9 LIST OF TABLES Table pa ge 3 1 Decellularization quality of kidney scaffolds produced with the elevated reservoir perfusion apparatus and detergent based protocols ................................ ......................... 56 3 2 Kidney cellular proteins removed in decellularization process ................................ ....... 65 4 1 Immunohistochemical detection of embryoid body differentiation compared to seeded acellular kidney scaffolds. ................................ ................................ ................ 114 6 1 VBM decellularization protocol ................................ ................................ ................... 143 6 2 Kidney Mod 1 decellularization protocol ................................ ................................ ..... 143 6 3 Kidney Mod 2 decellularization protocol ................................ ................................ ..... 144 6 4 Graduated 3% decellularization protocol ................................ ................................ ..... 145 6 5 Graduated 10% decellularization protocol ................................ ................................ .... 145 6 6 Kidney Extended decellularization protocol ................................ ................................ 146 6 7 Kidney Extended 2 decell ularization protocol ................................ .............................. 146 6 8 Kidney NaDC decellularization protocol ................................ ................................ ..... 147 6 9 Kidney SDS decellularization protocol ................................ ................................ ........ 147 6 10 Kidney 1 Day SDS decellularization protocol ................................ .............................. 148 6 11 Hematoxylin and Eosin Staining Protocol ................................ ................................ .... 152 6 12 Immunohistochemical Antibodies ................................ ................................ ................ 153 6 13 Slide Pretreatment for Immunohistochemistry ................................ ............................. 154
10 LIST OF FIGURES Figure page 2 1 Basement membrane of the adult kidney nephron is a rich tapestry of proteins, divided into functional segments ................................ ................................ .................... 24 2 2 Development of the nep hron ................................ ................................ .......................... 26 2 3 Renal developmental lineage map ................................ ................................ .................. 28 2 4 Tissue engineered renal units ................................ ................................ ......................... 36 3 1 Organ harvest and perfusion based decellularization ................................ ...................... 53 3 2 Improved decellularization protocols for removal of cell debris ................................ ..... 57 3 3 Supp lementary analysis of decellularized scaffolds ................................ ........................ 63 4 1 Whole scaffold static culture: kidney scaffolds produced with the Graduated 10% decellularization protocol and cultured with arterially delivere d B5/EGFP ES murine embryonic stem cells ................................ ................................ ................................ ..... 98 4 2 Perfused organ culture: murine B5/EGFP ES cells injected arterially into intact kidney scaffold and cultured with perfused media system ................................ ............ 102 4 3 Sectioned scaffold culture: typical distribution patterns of GFP labeled murine ES cells delivered into the renal artery of decellularized kidney scaffolds and cultured for 14 days ................................ ................................ ................................ ................... 104 4 4 Sectioned scaffold culture: morphological patterns of B5/EGFP ES cell growth in acellular kidney scaffolds ................................ ................................ ............................ 105 4 5 Sectioned scaffold culture: cultured sca ffold after ureteral retrograde seeding of B5/EGFP cells ................................ ................................ ................................ ............. 107 4 6 Patterns of ES cell division and apoptosis on section cultured scaffolds reflected with KI 67 and Cleaved Caspase 3 (CC3) IHC ................................ ................................ .... 109 4 7 Retention of transcription factor Oct 4, regulator of pluripotency, in B5/EGFP ES cells section cultured on acellular kidney scaffolds produced with the Kidney SDS protocol ................................ ................................ ................................ ....................... 110 4 8 Pan cytokeratin IHC reveals epithelial differentiation of seeded B5/EGFP ES cells at 10 days of sectioned scaffold culture ................................ ................................ ........... 111 4 9 Immunohistochemical analy sis of Pax 2 expression in ES cells section cultured on acellular kidney scaffolds produced using Kidney SDS protocol ................................ ... 113
11 A 1 Decellularization apparatus diagram ................................ ................................ ............ 156 A 2 Pressure based organ perfusion apparatus for decellularization ................................ .... 157 A 3 Perfusion culture apparatus diagram ................................ ................................ ............ 158 A 4 Organ perfusion bioreactor ................................ ................................ .......................... 159 B 1 Reference immunohistochemical comparison of normal rodent kidney tissue .............. 160 B 2 Local ization patterns of unremoved cellular proteins post Kidney NaDC decellularization using IHC ................................ ................................ ......................... 161 B 3 Detection of key proteins for cell division, apoptosis, and differentiation in B5/EGFP ES cells c ultured as embryoid bodies ................................ ................................ ........... 162
12 LIST OF ABBREVIATIONS AFS CELLS Amniotic fluid stem cells BM Basement membrane BMP Bone morphogenetic protein B RD U Bromodeoxyuridine BrdU Bromodeoxyuridine B S LB4 Bandeiraea simplicifolia le ctin B 4 BUN Blood urea nitrogen CC3 Cleaved caspace 3 CHAPS 3 [(3 cholamidopropyl)dimethylammonio] 1 propanesulfonate CK Cytokeratin CMA Cranial mesenteric artery CMC Critical micelle concentration DAB 3,3' Diaminobenzidine DAPI diamidino 2 phenylindo le DMEM Dulbecco' s modified Eagle's m edium DNA Deoxyribonucleic acid DN ASE Deoxyribonuclease EB S Embryoid bodies ECM Extracellular matrix EDTA Ethylenediaminetetraacetic acid ELISA Enzyme linked immunosorbent assay EPC S Endothelial progenitor cells ES CEL LS Embryonic stem cells ESRD End stage renal disease
13 FBS Fetal bovine serum GBM Glomerular basement membrane G DNF Glial cell line derived neurotrophic factor GFP Green fluorescent protein GSG Glycosaminoglycan H&E Hematoxylin and eosin HGF Hepatocyte growt h factor HSPG Heparan sulfated proteoglycan HUVEC S Human umbilical vein endothelial cells IHC Immunohistochemistry IMDM Iscove's modified Dulbecco's m edium I PSC S Induced pluripotent stem cells I RAD Implantable Renal Assist Device LIF Leukemia inhibitory fa ctor mES cells Murine embryonic stem cells MET Mesenchymal epithelial transition MHC Major histocompatibility complex M RNA Messenger ribonucleic acid MSC S Multipotent stromal cells N A DC Sodium deoxycholate PBS Phosphate buffered saline PCL polycaprolactone PECAM 1 Platelet endothelial cell adhesion molecule PLGA poly(lactic co glycolic acid) PMSF Phenylmethylsulfonyl fluoride
14 Q RT PCR Quantitative reverse transcriptase polymer chain reaction RAD Renal Assist Device R ET Receptor tyrosine kinase SCNT Somatic c ell nuclear transfer SDS Sodium dodecyl sulfate SEM Scanning electron microscopy TBS T Tris buffered saline + 0.1% Tween TEM Transmission electron microscopy U NAG Urinary N acetyl beta D glucosaminidase VBM Vascular basement membrane V I WAK Vicenza wearabl e artificial kidney WAK Wearable artificial kidney XNA S Xenoreactive natural antibodies
15 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doct or of Philosophy KIDNEY TISSUE ENGINEERING USING DECELLULARIZED ORGANS AND PLURIPOTENT STEM CELLS By Matthew James Williams August 2013 Chair: Christopher Batich Major: Biomedical Engineering Each year, there is a severe shortage of kidneys for organ t ransplant. One approach to alleviate this demand is to reconstruct kidneys using tissue engineering scaffolds. Such scaffolds may be produced through the decellularization of whole organs. Ideally, the resulting extracellular matrices would retain the orga n specific architecture and chemical composition to guide implanted cells into functional structures and eventually working organs. Pluripotent embryonic stem cells are highly proliferative, capable of differentiating into all cells of the kidney, and resp ond to tissue specific extracellular matrix proteins. Examining the ability of these cells to proliferate and differentiate in a whole kidney extracellular matrix scaffold would serve as a valuable baseline for future studies in producing a functioning org an. Decellularization of whole rat kidneys was achieved by perfusion of detergent based solutions through the renal artery. The most effective 5 day protocol utilized Triton X 100, sodium dodecyl sulfate (SDS), salts, and deoxyribonuclease (DNase) and was determined by modifying detergent concentration and perfusion duration and testing an alternative ionic detergent. The resultant organ shaped structure was spongy and translucent. Hematoxylin and eosin histochemical analysis and electron microscopy showed preserved glomerular, vascular and tubular structures without observed nuclear material and only trace amounts of residual
16 cellular debris. In addition, immunohistochemistry ( IHC ) showed removal of cellular proteins and preservation of essential basement m embrane (BM) proteins collagen IV and laminin Murine embryonic stem cells were seeded into the decellularized scaffolds and cultured up to 14 days with one of three protocols: whole scaffold, perfused scaffold, and sectioned scaffold culture. Cells delive red into the renal artery showed initial glomerular localization with subsequent expansion into adjacent vasculature, interstitium, and tubules. Over time, histological analysis revealed distinct cell morphologies, patterns of cell division and apoptosis, and ce ll arrangements with Pax 2 and c ytokeratin positivity. Cytokeratin is an epithelial cell marker and Pax 2 is necessary for kidney development. The evidence of cellular change of mouse stem cells in response to a rat scaffold is a promising step towar d establishment of a xenogenic scaffold source for engineered kidneys.
17 CHAPTER 1 INTRODUCTION Once an organ such as the kidney reaches a state of irreversible degradation, organ function must be replaced to ensure survival of the patient Currently, this is best accomplished through organ transplantation. In the United States, as of November 16, 2012, a total of 116,447 people were candidates for organ transplant, and of those, 81% were waiting for kidney transplantation. However, donors are in short supp ly. In 2011, only 16,800 candidates received a transplanted kidney ( Based on OPTN data as of November 16, 2012 ). Renal replacement therapies such as hemodialysis and peritoneal dialysis extend life expe ctancy and may bridge the time gap before transplantable organs becomes available, but at the cost of significant associated complications and increased rate of mortality over time. 1 Advancements in these technologies and associated treatment strategies steadily improve patient outcome, but they are still not adequate replacements for a functioning organ. Fortunately, advances in the new field of regenerative medicine give hope that a lternate sources of renal tissue may be created. A definition of regenerative medicine as given in an editorial by Mason and Dunnill or establish normal func based therapies with optional gene modification, biomaterials, and methods from molecular medicine. Stem cells are likely to play a key role. 2 One such approach to address the kidney shortage is to use a t issue scaffold to guide implanted cells into a functional replacement organ. However the kidney is a complex organ with over 30 cell types arranged in intricate structures 3 It is the precise arrangement of these cells on the renal extracellular matrix (ECM) that allows the kidney to perform its critical functions. These functions include blood filtration, waste removal, electrolyte regulation, blood
18 volume control, and stimulati on of red blood cell productio n. 4 Anthony Atal a stated that renal 5 Current techniques of fabricating cellular scaffolds include electrospinni ng, which has been used to create meshes of biopolymer with extracellular matrix proteins or bound bioactive peptides, and gas based techniques, which generate biomaterials with interconnecting pores for improved nutrient diffusion and cellular waste remov al 6 9 Although noteworthy, these techniques cannot yet mimic the advanced histoarchitecture of native renal ECM, and are unlikely to promote functional arrangeme nt of seeded cells. Extracellular matrix is a three dimensional network comprised of proteins, proteoglycans, and glycoproteins that supports cell attachment and arrangement, acts as a regulator and reservoir for growth factors, and modulates cell signalin g. Kidney ECM is crucial to renal function as it is a primary component in glomerular filtration and supports each segment of distinct epithelial cells with unique underlying matrix composition 10 The extraction of ECM fro m a kidney may provide a sufficient scaffold for seeded cells to form renal structures and eventuall y a whole functional kidney; this can be accomplished with a process called decellularization. Decellularization is the process that removes cells from tiss ue leaving extracellular matrix. Methods for decellularization include mechanical agitation and shearing, freeze thawing, sonication, and wash ing with acid or detergent The overall goal is to successfully remove any immunogenic cell remnants while preserv ing beneficial extracellular matrix proteins. 11 Extracellular matrices from both human and porcine sources have already been implanted clinically with or without growth factors to treat a variety of conditions. These scaffolds have been d erived from bone, skin, bladder, nerve, heart valve, and small intestine. 12 17 They typically rely on influx and differentiation of host cells to effect repair. En gineering a functional kidney,
19 however, may require intact whole organ ECM structure with population and arrangement of the multiple cell types in vitro Extracellular matrices for the purpose of organ reconstruction have been prepared from whole heart, lu ng, live r and pancreas 18 27 In most of these cases, the primary method of decellularization is perfusion of detergents through the vasculature. 28 A promising cell source for scaffold based kidney generation is pluripotent stem cells 29 The se cells have a high capacity for proliferation and the potential to differentiate into all cell types in the kidney. In addition, extracellular matrix isolated from a tissue has been shown to influence differentiation of pluripotent cells toward that tiss 20 30 Pluripotent cell lines have been established from embryonic stem cells, amniotic flu id stem cells, and induced pluripotent stem cells. 31 33 Other potential cell sources such as adult multipotent stem cells have been identified in adult kidneys, bu t they have not been shown to account for all cell types in the kidney 34 37 Extrarenal stem cell sources such as hematopoietic stem cells, endothelial stem cells and multipotent stromal cells (MSCs ) are another option; however, given testing on current injury repair models, they have not readily contributed to nephrogenic cell lineages 38 39 The hypothesis for this research is that an extracellular matrix tissue scaffold can be produced with perfusion based organ de ce llularization and with unique renal specifi c histoarchitecture best support the multitude of cell types necessary for proper kidney function and ultimately the generation of a replacement organ. While proving this hypothesis will likely take decades of research, the testing of two specific aims wil l provide fundamental knowledge for advancement of kidney organ engineering. The first will determine if cells and cellular debris may be removed from a whole rat kidney while preserving extracellular structures and matrix
20 proteins. The second will test th e ability of the ECM scaffold to effect attachment, proliferation and differentiation of injected pluripotent murine embryonic stem cells. The develo pment of a reliable process for kidney specific whole organ scaffold production would be a significant ste p toward the eventual generation of a new organ. Such a scaffold would allow the in vitro examination of cells in the outline of functional structures and in contact with retained extracellular molecules. Optimal conditions for cell growth and site specifi c differentiation may eventually be interpreted. Testing the response of pluripotent stem cells to the scaffold will serve as a baseline for further studies in scaffold preparation and modification, cell source alternatives, and culture variations with the goal of functional structure formation. As the removal of cellular debris removes a significant component of destructive host response, this research may lead to new organs formed from non donor matched kidneys, kidneys with cellular disease yet with pres erved matrix, or xenogenic sources 40 42 Additionally, pluripotent stem cells can now be derived from patient specific, non invasive cell sources, so manipulation of these cells into functional structures may be clinically relevant 43 44 This dissertation is divided into five additional chapters including a survey of relevant research, development of whole kidney decellularization, response of embryonic stem cells to renal extracellular matrix scaffolds, an exploration of options for further research, and details of the proced ures used in this research. Appendices include further details of the constructed apparatuses utilized in these studies and supplementary histology images.
21 CHAPTER 2 BACKGROUND End Stage Renal Disease and Renal Failure O ne of the most prominent healthcar e concerns in the country is renal failure, or the inability of the kidney to remove wastes and maintain electrolyte balance T he non reversible condition that will inevitably lead to kidney failure is end stage renal disease (ESRD) Acute renal failure an d chronic kidney disease can both progress to ESRD In 2010, almost 600,000 people were treated for ESRD in the U.S., with associated healthcare costs estimated at $47.5 billion. A majority of ESRD patients have a primary diagnosis of hypertension and /or d iabetes. The treatment for ESRD is renal replacement therapy, mainly hemodialysis, peritoneal dialysis, and ultimately kidney transplant. Dialysis treatments are used to extend life until a transplantable kidney is obtained, but they are costly, and the 5 year survival rate for those on dialysis is only 35% Kidney transplant markedly improves long term mortality, with 5 year survival improved to 73%, but th e severe shortage of procured organs considerably limits availability of this treatment option 1 In 2012, November, 94,556 people were waiting for a kidney transplant, yet in 2011 only 16,813 patients received transplant s ( Based o n OPTN data as of November 16, 2012 ). Production of functional, implantable tissue engineered kidney s would significantly impact the population of transplant candidates. Further more as the estimated primary cause of long term graft failure is antibody me diated rejection, a kidney designed to be non immunogenic would improve long term survivability of transplant recipients 45 The Kidney Knowledge of renal struc ture and function, including natural extracellular architecture, organ development, and response to injury will contribute to the creation of functional kidney tissue.
22 Anatomy and Physiology The kidney is a complex bean shaped organ that filters blood, rem oves waste, balances electrolytes, mediates acid base balance, controls blood volume and stimulates red blood cell production It contains primary functional units called nephrons. Each nephron contains a renal corpuscle which sits at the head of a multi s egment uriniferous tubule. The renal corpuscle capsule. Almost 25% of the blood pumped from the heart enters the kidney through the renal artery and then proceeds th rough branching arteries to the glomerulus. Blood filtrate passes through the fenestrated endothelium of the glomerulus, the glomerular basement membrane, and visceral epithelial cells or podocytes. The then captures the filtrate and direc t s it down the tubule. The blood c ontinues to move through a peri tubular capillary network before returning to circulation through the venous system. In the tubule, a sequence of segment specific, cell mediated absorption and reclamation processes allow wa ter and necessary metabolites such as glucose and amino acids to pass back into circul ation through the adjacent peri tubular capillary network. The remaining filtrate then passes through the collecting ducts, converges into the papillary ducts of Bellini, empties into the calyces, and departs through the ureters. The parenchyma of the kidney consists of two gross structures in radial alignment, the superficial cortex and deeper medulla. The medulla is further divided into an outer and inner zone. The cortex mainly contains the glomeruli and convoluted sections of the tubules. The medulla contains the straight portions of the tubules and collecting ducts. 4 Extracellular Matrix The extracellular matrix exists as a network of proteins and polysaccharides that is produced by an d supports cells in a tissue. S pecialized sheets of extracellular matrix called basement membranes underlie epithelial cells and connect them to the interstitial matrix.
23 Basement membranes are necessary for normal tissue function and development and defect s cause diminished function. They also play a key role in glomerular filtration 10 They regulate cell adhesion, polarity, migration, growth and differentiation 46 The adult kidney has distinct ECM composition with in its proximal tubule, loop of Henle, distal tubule, collecting duct, and the mesangial matrix (Figure 2 1). Component proteins of the basement membrane include collagen IV, laminin, entactin/nidogen, and sulfated proteoglycans. Collagen IV provides the covalently bonded latticework that structurally supports the basement membrane. The polypeptide chains wh ich assemble to form a collagen IV triple helix distal tubular, and col lecting duct basement membranes. 10 Laminins are ubiquitous basement membrane glycoproteins which self associate, form sheets, and commonly connect with Collagen IV through entactin/nidogen, perlecan, and fibronectin 47 48 Further, laminins act as the primary binding sites for cellular integrins and play a major role in cell attachm ent, differentiation, and preservation of phenotype; maintenance of tubular and vascular structures; a nd numerous nephrogenic processes 10 46 49 50 The isoforms of laminin differ with tissue type and developmental stage. The primary laminins in vascular basement membranes are 411, 421, 511, and 521, with 111, 211, 511, and 521 in the remainder of the adult kidney basement membranes 10 46 Proteoglycans are found in all basement membranes, and consist of protein cores with the glycosaminoglycan side chains hyaluronan, heparan sulfate, chondroitin sulfate, and/or dermatan sulfate. These glycosaminoglycans stabilize the basem ent membrane through binding with
24 laminin, collagen IV, and entactin/nidogen and are important mediators of cell behavior. They are necessary for growth and development in the kidne y. 48 51 Proteoglycans in the kidney include agrin, perlecan, bamacan, and collagen XVII. 10 Another function of proteoglycans is to bind and Figure 2 1. Basement membrane of the adult kidney nephron is a rich tapestry of proteins, divided into functional segments distal tubule; GBM, glomerular basemen t membrane; LH, loop of Henle; MM, mesangial matrix; PT, proximal tubule. Reprinted by permission from Macmillan Publishers Ltd: Kidney International. Miner JH. Renal basement membra ne components. vol. 56, pg. 2017, (Figure 1), copyright 1999
25 regulate g rowth factor delivery 48 49 52 53 Growth factors such as VEGF, PDGF, HGF, FGF, TGF 7, and Wnt 4 are all involved in renal processes 52 54 5 6 Renal extracellular matrix constituents play a role in development 10 51 There is a distinct spatiotempora l expression profile; at various stages of development, there are shifts in renal extracellular molecules, including alterations in collagen alpha chain and laminin isoform expression 57 Change of normal extracellular matrix e xpression patterns can result in diminished or loss of function in the kidney 10 48 51 For example, entactin/nidogen is required for tubulogen e sis and collagen XVIII is required for ureteric bud branching 48 Organogenesis The processes involved in transformation of embryonic cells into an adult organ can give insight into potential stem cell sources, tissue patterning, or the differentiation of stem cells i nto functional adult phenotypes. The origin of kidney development can be traced back to the inner cell mass of the blastocyst; it forms a single sheet of cells called the epiblast The cells of the epiblast, in the process of gastrulation, migrate along th e primitive streak and develop into embryonic endoderm, mesoderm, and ectoderm. Mesodermal tissue then exhibits a patterning whereby the lateral axis segregates into intermed iate and lateral plate mesoderm. From the intermediate mesoderm, the nephric or Wo lffian ducts form, in which rudimentary pronephric and mesonephric epithelial kidney tubules branch 58 However, in higher vertebrates, these structures are supplanted by metanephric kidney development, which is initiated on the caudal en d of the nephric ducts (Figure 2 2). In this pr ocess, an outgrowth of the duct called the ureteric bud extends into metanephric mesenchyme and induces aggregation of cells around the tip 59 This occurs at E11 (days after conception in mouse). These cells, known as cap mesenchyme, are considered stem cell precursors for all epithelial cells in the nephron except those in th e
26 collecting duct 60 Conversion of these cells into epithelium is known as mesenchymal epithelial transition (MET) The signals generated during the formation of aggregates cause the bud to branch and extend more bud tips Meanwhile, cap mesenchyme cells polarize to form renal vesicles, which cleft once to become comma shaped bodies, and again to form s shaped bodies. One end of these s shaped bodies connects back to the branching ureteric system and the other A B C D Figure 2 2. Development of the nephron A) Branching ureteric buds extend into the metanephric mesenchyme, inducing condensation o f the cells at the tip into cap mesenchyme. These are stem cell precursors to all e pithelial cells in the nephron. B) Mesenchymal cells that do not aggregate form stromal cells that are precursors of vascular pericytes and cells in the interstitium and mes angium. Cap mesenchymal cells then polarize to form renal vesicles. C) These vesicles form S shaped bodies, merge with the collecting system, recruit vasculature, and D) develop and elongate into mature nephrons. Key molecular markers expressed at each sta ge are indicated Adapted with permission: Development. Dressler GR. Advances in early kidney specification, development and patterning. 2009; 136(23), pg. 3867, (Figure 4)
27 recruits end othelial cells to form the glomerular capillaries. Over time, and through a series of intricate molecular interactions, the nephron matures and the cells differentiate into at least 26 different cell types (based on morphology, location, and function) asso ciated with an adult kidney 58 61 All cells in the adult kidney originate from the intermediate mesoderm. Fr om that source, distinct transitional cell populations identified by unique combinations of expressed molecular markers, progressively differentiate and interact over time to form the necessary adult somatic cell types. A basic lineage map summarizing the acknowledged renal cell populations and pathways of differentiation is given in Figure 2 3. 62 This map, with associated population markers, can be used to establish renal differentiation of stem cells. The Osr1 gene is the earliest known marker of intermediate mesoderm It is required for the expression of other factors involved in kidn ey formation such as Eya1, Si x2, Pax2, Sall1 and Gdnf (glial cell line derived neurotrophic factor) 63 Lim1 (lhx1) is also expressed in the intermediate mesoderm, and later in the nephric duct, and many of the structures of the developing metanephros A ll epithelial patter ning in the intermediate mesoderm requires Lim1 64 In addition to lim1 and odd1, pax2 and pax8 are also expressed in more mature intermediate mesoderm 58 65 Hox genes mediate mesodermal patterning; Hox11 is required to distinguish metanephric mesenchyme 59 Two genes directly involved with regulating ureteric bud outgrowth are Gdnf which is secreted by the metanephric mesenchyme, and its receptor, R et (receptor tyrosine kinase), in the ureteric bud epithelium. S ecretion of Wnt9b from the ureteric bud tip then initiates MET 60 Condensed mesenchyme or cap mese nchyme is distinguished by Six2, Cited1, and Gdnf 59 60 The release of Wnt9b also upregulates Fgf8 and Wnt4 the latter activating renal vesicle formation and
28 polarization of epithelia. Foxd1 identifies s tromal progenitors such as interstitial or mesangial cells, al so derived from the metanephric mesenchyme 60 O ther significan t genes involved in the molecular regulation of nephrogenesis include Pax 2, WT 1, Eya 1 and Sall1. Pax2 a critical factor in branching morphogenesis is expressed Figure 2 3. Re nal developmental lineage map Two primary cell populations, nephric duct and metanephric mesenchyme are derived from intermediate mesoderm. The nephric duct gives rise to the renal collecting system, whereas the metanephric mesenchyme is a precursor to al l other cells in the nephron as well as vascular and stromal populations. Cells in the cap mesenchyme are a precursor to all epithelial cells in the nephron. Ovals indicate anchor marker genes, unique to the specific population Reprinted by permission: Cold Spring Harb Perspect Biol 2012;4:a008300
29 in ureteric bud metanephric mesenchyme, and cap mesenchyme. 65 68 Pax2 is also necessary for selective differentiation of cap mesenchyme and F oxd1+ stromal stem cells into more mature phenotypes 38 WT 1 is a transcription factor that represses Pax 2 during normal development, allowing terminal dif ferentiation in the metanephros. 69 70 Eya 1 a specifier of metanephric mesenchyme, plays a go verning role in Gdnf regulation and is essentia l for nephrogenesis 71 Sall1 is yet another gene required for ureteric bud invasion and indu ction of metanephric mesenchyme. 72 Many of the genes and proteins listed are also implicated in adult rep arative pathways, such as Pax 2 73 Kidney Repair The kidney has innate mechanisms for damage repair. Benigni et al. (2010) provide an informative summary but a complete picture has not been established Nevertheless sources of repair have been reported to include differentiat ed renal cell types, extra renal cells, and kidney stem cells. The primary contributor to replacement of apoptotic or denuded cells caused by ischemic injury has been shown to be adult renal epithelial cells 38 One proposed mechanism is dedifferentiation migration, proliferation and re differentiation of the epithelial cells 73 74 Furthermore, i t has been suggested that epithelial cells gain plasticity and revert to functionally active p rogenitor s through epithelial to mesenchymal transition 38 Extra renal cells s uch as bone marrow derived stem cells, endothelial stem cells and mesenchymal stromal cells do not normally contribute to nephrogenic lineages 39 However, they have been shown to have a supp ortive role in renal repair through mobilization to site of injury and paracrine signaling 38 Resident epithelial cells and macrophages have also been shown to ha ve a localized paracrine effect. 38 75 For example, macrophages, which are adept at invading tissue, release Wnt7B into injured kidney tissue, promoting basement membrane repair, reduction of apoptosis, and tubule repopulation 75
30 Stem cells have been isolated from the adult kidney, but their role in repair has not been entirely elucidated. They have been shown to activate upon injury, enter into cell division, and mobilize to sites of repair 39 They have also been shown to engraft into developing kidney cell populations yet their strength may be as another supportive paracrine influence. 36 37 76 Suspected renal stem cells have been identified by expression of recognized stem cell ma rkers, lack of differentiated cell markers, prolonged cellular quiescence, enhanced proliferation capacity, or mobilization upon injury 34 35 37 38 In addition, these cells are typically tested for ability to differentiate into renal epithelial cells, integrat e into a developing kidney, and differentiate into unrelated cell types to affirm multipotent character 34 35 37 38 Adult kidney stem cells have been isolated from glomerular parietal epithelium, tu bules, collecting ducts, papillae, and related interstitium 35 38 74 76 77 One example, called label retaining tubular cells, was identified using bromodeoxyuridine (BrdU) labeling Peritoneal cells were exposed over time to BrdU, which incorporate d into replica ting cell DNA. When exposure was finished, rapidly dividing cells lost BrdU, leaving slowly dividing cells with retained label. These labelled cells integrated into developing metanephric kidneys, proliferated after ischemic injury, and responded to growth factors known to effect kidney repair 36 These cells may correspond with cells identified by Gupta et al. (2006), which showed expression of stem cell (mesenchy mal, embryonic, and developing kidney) markers vimentin, CD90, Oct4, and Pax 2 and lacked expression of differentiated cell markers MHC class 1 o r 2 or c ytokeratin 34 Two other markers that have been used for kidney stem cell identif ication are CD24 and CD133 74 76 77 Presence of CD24 and CD133 and absence of PDX, a marker of podocyte and endothelial cells, was used to identify the renal progenitors located in the urinary pole adjacent to the tubule opening These cells contributed to both the podocyte and tubu le cell populations after
31 renal injury 77 Unfortunately, these tissue specific stem cells, as well as other innate mechanisms for kidney repair, are unable to re generate lost tissue or compensate for effects of acute renal failure or chronic kidney disease. Current Approaches to Renal Tissue Engineering There are several proposed avenues to address the need for functional kidneys: the improvement of current transp lant technology, the use of cellular engineering techniques to provide reparative cells to damaged tissue, generation of tissue rudiments, and construction of implantable bio artificial renal structures. Kidney Xenotransplantation The primary concern in ki dney transplantation is the limited number of viable donor organs. Therefore, xenotransplantation is being pursued as an alternative Organs from other animal species, such as pig, would solve the demand but the severe immune response resulting in graft r ejection would need to be mitigated. The current state of kidney xenotransplantation is reviewed by Ramackers et al. (2012) 78 Transplantation of a xenogen ei c organ provokes a hyperacute rejection response, whi ch is caused from activation of xenoreactive natural antibodies (XNA s ) in the host These XNA s are present without immunization and target xenogenic antigens. The primary target of XNA s is alpha gal epitope which is expressed on all animals except human a nd humanoid primates. Identified targets of this and other XNAs may be resolved by gene knockout in the donor tissue, bu t acute vascular rejection still remains a concern. Acute vascular rejection develops from an adverse interaction between human blood an d xenogenic endothelium resulting in coagulation and subsequent graft failure Progress has been made with transgenic expression of protective and anti thrombotic proteins, but a viable xenogen eic transplant is still not likely in the near future O ne aven ue to mitigate rejection may be to implant primordial xenogen e ic kidneys (renal metanephroi) and allow them to develop in situ as
32 xenoreactive antigens are not present in renal tissue at an early stage of development. The vasculature can be recruited from the host, in essence removing a significant source of graft rejection. These experiments are in early stages however, and pig to primate implantation still needs to be performed 78 In addition, other concerns such as viral transmission and determination if developed organs will meet physiologic demand need to be pursued 79 Cell based Therapies A popular strategy for the repair of tissue is to deliver cells that will replace damaged cells, stimulate localized repair, or even generate new tissue. Epithelial cells, adult kidney stem cells, ex tra renal stem cells, and even amniotic fluid stem cells have all been used in this context. In one effort to provide a possible regenerative treatment for chronic kidney disease, a tubular cell enriched population of primary renal cells embedded in a gela tin hydrogel was injected into a healthy rodent kidney. The cells showed evidence of neo kidney tissue formation with structures resembling early glomeruli and tubules with little evidence of cell necrosis or destructive inflammatory or fibrotic response. However, long term cell integration and functional impro vement have yet to be evaluated. 80 pathways. Isolation and delivery or mobilization of these cells into the damaged area is a logical step in promoting kidney repair. In current models of kidney repair, the selected stem c ells are usually delivered arterially or injected under the kidney ca psule. Significant achievements include integration of cells into injured parenchymal tissue and expression of differentiated renal epithelial markers. Cells may also self assemble into new structures. Ultimately though, success is determined by a measure of functional recovery. Adult kidney stem cells have been used to repair renal tissue One example is the proximal tubule stem cell identified by Gupta et al. (2006) It showed up to 10% integration into
33 renal epithelial tissue in a rodent ischemia reperfu sion injury model. However, there was no improvement in renal function as measured by serum creatinine or creatinine clearance 34 Similarly, another study injected proximal tubule stem cells under the capsule after ischemia reperfusio n and though it showed no variation of serum creatinine or blood urea nitrogen (BUN) it did result in a reduction of tubule impairment as indicated by urinary N acetyl beta D glucosaminidase (U NAG) measurement 35 Interestingly, this cell line spontaneously formed tubule like structures when implanted into muscle. In a model of tubulonecrosis, CD133+ human tubule stem cells were injected intravenously into mice and showed localization and increased integration into site of injury whereas this response was minimal in uninjured control animals 76 Other kidney stem cell s were tested in models of kidney repair. Papillary stem cells have been shown to integrate into renal me dulla during transient ischemia. 81 Glomerular parietal ce lls selected for CD24+ mesenchymal stem cell marker were implanted in a clot under the capsule and showed signs of new glomerular and tubular tissue formation with recruitment of vasculature and expression of mature markers; however, cells only activated u nder compensatory response from contralateral nephrectomy. 74 Unfortunately, in all these adult kidney stem cell repair models, there was no functional improvement demonstrated. However, in one study a reduction in proteinuria was achieved with repeated intravenous injection s into mice with Adriamycin induced nephropathy. These cells could be expanded in vitro and differentiated into either podocytes or tubular cells. In addition, injected cells contributed to these populations in vivo after injury 77 Extrarenal adult stem cells have also been investigated for cell based kidney repair. It was shown that an infusion of bone marrow cells reduced mortality, improved renal function, and reduced macrophage invasion in a rat glomerulosclerosis model. Higher doses of bone
34 marrow cells induced regenera tion of mesangial and glomerular endothe lial cells. 82 In 2007, it was reported that an intrarenal arterial injection of multipote nt stromal cells (MSCs) reduced glomerular adhesion formation proteinuria, and serum ur ea nitrogen in a rat model of glomerulonephritis. 83 However, in the long term (60+ days) there was histological evidence of sclerosis and adipocyte maldifferentia tion. The most beneficial aspect of mesenchymal stromal cell therapy may actually be enhancement of their postulated role in renal repair: angiogenesis, reduction of inflammation, and stimulation of resident stem cell populati ons through paracrine signalin g. 84 As of 2012, clinical trials for use of MSCs in ischemia reperf usion injury are being conducted. 85 An alternate source of cells for r enal repair is amniotic fluid stem cells. Human amniotic fluid cells selected for CD117+ have been shown to increase tubular proliferation, reduce apoptosis, relieve tubular necrosis through cell protection and recover function in immune deficient mice. T hese cells do not induce tumor formation, a particular con cern for pluripotent cell lines. 32 Therapies based on amniotic fluid stem cells have pote ntial to treat r enal injuries, h owever, very few have shown functional recovery let alone long term efficacy. In addition, these studies have mainly shown repair through cell protective mechanisms with minor evidence of engraftment. Some renal disease s cause structural da mage to the extracellular matrix, to which there is insofar no evidence of repair. Tissue Structure Generation Certain cell based therapies have been shown to generate neo kidney tissue when implanted in to a kidney or ectopic site 35 80 However, strategies are also being pursued to generate renal structures using developmental tissues. Qiao et al. (1999) showed th at ureteric bud tissue could be separated and induced to branch in vitro using only soluble media factors. Further, implantation of branched ureteric bud into a pool of metanephric mesenchyme induced
35 formation of nephron like structures. It was suggested t hat developmental rudiments could be isolated, expanded and combined to form complex kidney structures 86 In a later study it was shown that such neo kidney structures would recruit vasculature when implanted under the kidne y capsule 87 Recent studies demonstrated how autologous multipotent stromal cells could be placed in a xenoembryo at location of a ureteric bed before branch ini tialization and generate nephron like structures. These structures could then be implanted back into donor omentum, recruit vasculature, an d produce urine. 88 Such approaches are still at the proo f of concept stage and require the use of xenoembryos, but isolation of the factors and developmental steps required to differentiate structures from autologous cells may lead to a completely in vitro procedure for generation of functional renal tissue. Im plantable Artificial Constructs Renal dialysis is a time consuming process To improve patient independence and compliance, there is sustained effort to produce miniaturized portable dialysis systems. Ronco et al. (2011) summarizes recent advancements. Not able device designs include the WAK (wearable artificial kidney) and ViWAK (Vicenza wearable artificial kidney). The former uses low flow continuous perfusion, while the latter employs a daily process of loading the peritoneal cavity with 2 liters of fresh solution followed by equilibrium circulation, ultrafiltration, and fluid exchange. In order to lengthen the time between treatments, methods for glucose and bicarbonate regulation and small molecule removal are being investigated. Reduction in size and i ncreased function ality and efficiency will correlate with improvements in nanoscale materials micromechanics and microfluidics 89 Parallel efforts aim to use see ded renal cells on artificial constructs in order to reproduce the resorptive, homeostatic, metabolic, and endocrinologic functions of the kidney. One notable study from Lanza et al. (2002) used cells from a cloned bovine metanephros. These cells were seed ed onto collagen I coated polycarbonate membranes
36 and self associated into renal units with connected glomerular and tubular structures (Figure 2 4 ). Implantable constructs assembled from these units recruited vasculature and produced urine like fluid whe n implanted subcutaneously 90 The Renal Assist Device (RAD) is another notable bioartificial device using renal cells. It pairs a conventional hollow fiber dialyze r which acts as a hemofilter with a second dialyzer containing a confluent layer o f renal tubule epithelial cells. 91 This improves the absorption and reclamation functions of the system, and has been shown to benefit pa tients with acute renal failure. 92 Currently, this type of system remains extra corporeal. However, in order to convert this type of system into a self contained persistent, implantable artificial kidney, design challenges and necessary technological improvements such as in vivo compartme ntalization and membrane filtration efficiency, thickness, uniform porosity, biocompatibility and mechanical integrity are being addressed 93 94 Conditions have already been described for a renal epithelial c ell bioreactor 94 The culmination of these efforts to produce an implantable artificial kidney has resulted in the implantable Renal Assist Device (iRA D) project F igure 2 4 Tissue engineered renal units Cloned bovine kidney cells self assemble around collagen coated polycarbonate tubes which drain into a polyethylene reservoir through silastic catheters. A). Constructs produce urine after three months of subcutaneous implantation. B) Histochemical analysis shows continuity between glomerulus, tubules, and polycarbonate membrane Adapted by pe rmission from Macmillan Publishers Ltd : Nature Biotechnology. Lanza RP et al Generation of histocompatible tissues using nuclear transplantation. vol. 20, pg. 692, (Figures 3 & 4), c opyright 2002
37 In 2012, in order to find an expedient solution for ESRD, the FDA announced that devices such as the iRA D and WAK would be fast tracked. 95 Tissue Engineering Scaffolds Biomaterials have been used to deliver cells to a site of intended kidney repair ; support cell containment, attachment, and polarization ; mediate destructiv e host responses ; and even promote cell differentiation and structure assembly 74 80 90 The overall goal is to provide a protected bioactive environment that will promote cell viability and eventual assembly into functional tissue structures. Biomaterials may be synthetic or tissue derived. A simple example of bio material use i s the delivery of mouse CD24+ parietal epithelial cells into kidney parenchyma using clot gathered from an autologous tail vein bleed. The clot a simple biomaterial, served to keep the cells localized. 74 In contrast, another study implanted an assortment of biomaterial constructs into a kidney in vivo to test histological response and determine the most suitable material for an implantable cell mater ial construct. Tested b iomaterials included combinations of gelatin, hyaluronic acid, polycaprolactone (PCL) and poly(lactic co glycolic acid) ( PLGA ) Gelatin hydrogel was the most promising material to use in the construct a ccording to measures of histop athology such as inflammation, fibrosis, necrosis, and mineralization and measures of biocompatibility including degradation, neovascularization, and neotissue formation. The resulting primary renal cell/gelatin hydrogel implant promoted cell assembly int o new renal structures. 80 In an alternate biomaterials application, a layer of collagen I was used to anchor developmental kidney epithelial cells to polycarbonate membrane s in the construction of a small renal implant that produced urine like fluid 90 These examples of biomaterials used in renal tissue engineering are relatively sim ple in structure h owever, given the architectural complexity of renal tissue, a tissue scaffold may be more effective in guiding kidney tissue formation.
38 A tissue scaffold has been described as optimally having four functions or features: architecture, me chanical stability, cell and tissue compatibility, and bioactivity. 96 Tissue appropriate architecture is the structure that has space and shape for functional cell arrangement, vascularization, and nutrient and metabolite transfer. Mechanically, the structure needs to withstand expected stresses at site of implantation while maintaining architectural stability. Additionally, as mechanical properties such as tissue st iffness can affect differentiation tendencies, it is desirable to match the scaffold and target tissue. Compatibility is the ability of the scaffold to integrate into recipient tissue and al low for cell attachment, growth, and differentiation without c ausi ng adverse tissue response. Bioactivity is the active regulation of cellular behavior. This may be an intrinsic feature of the scaffold or bestowed through modification For example, ligands may be embedded in a scaffold to promote cell proliferation and d ifferentiation. Complex b iomaterial scaffolds have been fabricated for use in tissue engineering. The goal was to mimic the normal extracellular environment of a tissue Intricate porous structures have been created from polymers using gas based techniques that include foaming, CO 2 water emulsion templating, and use of high pressure CO 2 during crosslinking. They have been shown to allow for cellular infiltration, vascular invasion, and nutrient and waste movement. 6 Another current fabrication technique for artificial scaffold production is electrospinning. It uses electrical charge to drive a polymer solution into nanoscale fibers that may be layered to form a mesh. These m eshes have been produced from synthetic polymers and ECM proteins such as laminin or PCL and collagen, and have also been embedded with bioactive peptides 7 9 Desp ite these advances, artificial scaffolds cannot yet mimic the heterogeneous structures and elaborate chemical landscape s that are contained in natural ECM, and are therefore unlikely to reproduce
39 the in vivo microenvironment necessary for cell behavior app ropriate to the tissue. 97 Fortunately, various methods exist to isolate intact extracellular matrices from tissues for use as cellular scaffolds. Extracellular Matrix Scaffolds Natural extracellular matrices, given their tissue specific structures, present as ready made tissue scaffolds. They can be isolated from tis sues usin g a cell removal process called dec ellularization 11 This process generally diminishes bioactivity and structural integrity fo r the sake of biocompatibility. It has been shown, in regard to cell growth and directed stem cell differentiati on, that culture with intact whole ECM is superior to heterogeneous mixture s of ECM proteins, which is in turn superior to individual matrix components 20 30 Implanted ECM has been shown to promote angiogenesis, recruitment of circulating progenitor cells, and constructive tissue remodeling. 49 Non donor matched extracellular matrix scaffolds from both human and xenogeneic sources have been implanted clinically for the sake of tissue repair. Extracellular matrix constituents are gen erally conserved across species and it is possible to prepare xenografts that are immunologically tolerated that do not trigger cell mediated implant rejection. 40 97 98 The ability to implant n on human ECM grafts sign ificantly i ncreases the pool of available scaffold material Scaffolds derived from human and porcine tissues have been implanted clinically. These include bone skin, bladder nerve heart valve, and small intestine. 12 17 Acellular matrices have also been produced from esophagus, ureter, tendon, ligament, skeletal muscle, artery, and even amniotic membrane. 99 107 A recent trend is the decellularization of organs for the intended purpose of whole organ regeneration ; these scaffolds have been created from heart, lung, liver, and pancreas. 18 27
40 Tissue Decellularization The removal of cells from a tissue, or decellularization, leaves a residual extracellular matrix. This matrix can then be used as a scaffold for cell ular repopulation. The process of decellularization can involve physic al methods such as deformation pressurization agitation, sonication, or freeze thaw cycles; chemical methods such as acid, base, alcohol or detergent exposure, chelation treatments, or use of osmotic stress; or en zymatic methods to degrade cellular material or inhibit endogenous ECM degrading protease s 11 108 Each step i n a decellularization protocol can produce a unique result dependent on tissue, location, species, and age. 49 The overall goal is creation of a scaffold that evokes maximum cellular response with minimum immunogenicity. 109 This is achieved with removal of cellular debris, including DNA and other cellular antigens, and preservation of extracellular matrix proteins and structure. 108 The primary challenge is that each step in the cell removal process can have a harmful effect on native ECM, such as denatura tion of cell adhesion peptides, so the procedure must be carefully balanced between cell removal and the destruction or removal of native ECM proteins. The most prevalent method for tissue decellularization is treatment with biological detergents 11 108 Detergents disrupt lipid lipid and lipid protein interactions and are used to break apart cell membranes and ca rry away cellular material. Detergent properties are derived from their amphiphilic nature. The charge of the head group defines the detergent as non ionic, cationic, anionic, or zwitterionic. The measure of detergency, or detergent effectiveness, is depen dent on the critical micelle concentration (CMC). The CMC is the concentration at which a micelle will form. Below the CMC detergent molecules will incorporate into a cell membrane, but detergent self association including lipid detergent and protein dete rgent complexes will not form unless the CMC is exceeded. Detergents with low CMCs form more stable micelles with slow molecule incorporation. The CMC varies with temperature, pH, strength of counterions,
41 and presence of proteins, lipids, and other surfact ants. 110 Ionic detergents can modify protein structure and disrupt protein aggregates to a greater extent than non ion ic or zwitterionic detergents. T hey are therefore more effective at cellular debris removal but may also cause more damage to basement membrane proteins. Factors su ch as head group, concentration, and duration of exposure all affect the outcome of a decellularization. 11 Detergent based protocols have been particularly effective at decellularizing whole organs 18 27 108 The solutions are typically perfused through the vasculature, efficiently delivering to all a reas of the organ. 28 The establishment of an effective detergent based decellularization protocol for the production of kidney whole organ scaffolds would be a significant step toward th e possibility of a functional tissue engineered kidney. Cells for Kidney Tissue Engineering Once a tissue engineering scaffold is produced, the greatest challenge is to repopulate it with the necessary cells in functional arrangement. There at least 26 ide ntified cell types in an adult kidney. 58 Potential sources include adult kidney somatic cells, adult renal stem cells, extrarenal stem cells, pluripotent stem cells and cells involved in kidney organogenesis. In practical application, the selection will depend on not only which cell s will respond to the particular scaffold and growth environment, but which will be tolerated by the host upon implantation. In addition, methods for efficient isolation, purification, expansion, and differentiation are nece ssary to achieving sustainable a nd reproducible outcomes. Adult kidney parenchymal cells, after injection into a kidney, have shown promise by self assembling into kidney like structures. 80 These structur es have not been functionally tested, but it would be worthwhile to see if this behavior could be reproduced in a controlled in vitro environment with a kidney specific scaffold. The primary drawback to this approach is that it would likely require destruc tion of a human kidney to harvest the cells. In addition, the cells
42 would need to be HLA matched and the maintenance and balance of the many cell types over more than short term would likely not be feasible for routine application. The alternative approach is to use stem cells. Adult Stem Cells Stem cells have a high capacity for proliferation and the potential to differentiate into multiple mature cell types. The derivation of maintenance culture conditions and establishment of stem cell lines may provide an abundant cell source for organ regeneration. The challenge is to identify the stem cell or limited combination of stem cells which have the potential to differentiate into all cell types necessary for kidney function. A review of promising stem cell sou rces is provided by Harari Steinberg et al. (2011). 39 Several adult renal stem cell candidates have been identified, typically through expression of mature and embryonic markers, reduced or prolific cell division, activation during injury repair, and ability to integrate into a developing kidney. 34 37 Locations of proposed stem cells have included prox imal, distal and collecting tubules, interstitium, glomerular parietal epithelium, and papillae. 34 38 77 These multipotent cells have been shown to differentiate into adult kidney cells and contribute to renal tissue after injection into developing metanephros. 34 36 37 In addition, they have been induced into non renal lineages. 34 37 Extrarenal stem cell sources such as hematopoietic stem cells and MSCs are another option for scaffold repopulation. Examination of current injury repair models reveals they do not normally contribute to nephrogenic cell lineages. 38 39 However, similar to adult kidney stem cells, they have been ind uced to differentiate into various alternate tissue cell types. 111 In additi on, it was shown that human MSCs will differentiate and contribute to glomerular epith elial, tubular epithelial, and interstitial cells if implanted in a cultured embryonic rodent kidney. 112 Unfortunately, culture conditions to achieve this in vitr o have not been defined. Unless these adult stem cells can be implanted without eliciting a rejection response, they will
43 have to be extracted from the host or donor matched. Hematopoietic stem cells and mesenchymal stromal cells have the advantage of bein g isolated from peripheral blood. 111 The attraction of adult stem cell use for kidney tissue engineering is the opportunity for autologous sourcing, yet there is no assuring that any combination of these cells can account for the totality of adult renal somatic cells. Fetal Stem Cells All cells in the kidney are derived from the in termediate mesoderm. From there branching network s of transitional cell populations p rogress, interacting and differentiating until all cells in a mature kidney are accounted fo r (Figure 2 3) These precursor cell populations, able to differentiate into multiple mature renal cell types, may be candidates for use in scaffold based tissue en gineering. One cell line, 7.1.1, has been derived from rodent metanephric mesenchyme and will differentiate into all renal cells except for those originating from the ureteric bud epithelium. 81 Other possibilities include the Six2+ Cited1+ stem cells from the cap mesenchyme, from which all the epithelial cells in the nephron originate, or the Foxd1+ stem cells, precursors to the stromal population 38 113 These potential stem cell sources still need defined culture conditions that would allow in vitro maintenance w ith preservation of stemness. Cells from dissociated metanephros have been shown to self assemble into glomerular and tubular like structures. 90 Further, it has been shown that cells from a fetal organ will infiltrate an acellular scaffold when co cultured adjacently. 114 An advantage of using cells from a fetal kidney is their inherent propensity to differentiate into mature kidney cells. 39 In an effort to create a histocompatible source for these embryonic kidney cells, somatic cell nuclear transfer (SCNT) was utilized (Lanza, 2002). This technique, also considered therapeut ic c loning, involves insertion of the nucleus of a somatic cell from the host into a donor egg, initiation of
44 cell division, and implantation of the resulting embryo into a surrogate. The developing organ of interest may then be used at the target time point Realistically, it is unlikely that human fetal cell sources or SCNT methods will be employed. First, the manipulation of human fetuses for the purposes of cell harvest has strong ethical considerations. Second, SCNT has been largely superseded by the creat ion of induced pluripotent stem cells ( IPSC s) autologous cells modified to have similar properties to embryonic stem cells. 33 Pluripotent Stem Cells Pluripoten t stem cells have the ability to differentiate into any cell derived from the three germ layers, endoderm, mesoderm, and ectoderm. This makes them attractive for scaffold based tissue engineering as one cell source may provide all necessary cell types for organ function. 29 Given this plasticity, the primary challenge is to direct cells to intended lineage without maldifferentiation or teratoma formation. 39 Pluripotent cells include embryonic stem cells, amniotic fluid stem cells, and induced pluripotent stem cells. They each can form embryoid bodies, three dimensional aggregates capable of in vitro tissue formation. 31 33 Also, each have been directed toward renal lineages using prescribed cell culture conditions. 32 115 Another advantage to pluripotent stem cells is they likely have a proliferative capacity that exceeds adult or fetal stem cells. 39 Embryonic stem cells. Embryonic stem (ES) cells are pluripotent cells derived from the inner cell mass of blastocyst stage embryos. 116 117 They are easy to maintain, have a high doubling capacity, and can be passaged indefinitely. 31 116 118 There are established cell lines from human and mouse, among others, and there is an extensive body of knowledge in handling, differentiation, and application. 31 In regard to renal differentiation, teratomas devel oped from implanted ES cells were shown to spontaneously form branching ducts with neph rogenic
45 morphology and express kidney specific markers. 119 Also, ES cell s in culture have been directed to differentiate toward mesoderm, intermediate mesoderm, metanephric mesenchyme, and more mature tubule cell types through addition of growth factors and other additives to the medium; these include BMP2 (bone morphogenetic protein), BMP4, BMP7, GDNF, HGF (hepatocyte growth factor), Activin and retinoic acid. 120 123 Establishment of human ES cell lines requires manipulation of embry os, which gives rise to ethical implications. Initially, establishment of ES cell lines required embryo destruction, yet it has also been accomplished using a non destructive blastomere biopsy. This technique removes a cell at the 6 8 cell stage of an embr yo and is used by fertility clinics for pre implantation genetic diagnosis and selection. 124 Unfortunately, differences between biopsied and unalt ered embryos have been identified and long term effects of the surviving embryos are uncertain. 125 126 Curr ently, at least two hundred human ES cell lines are available fo r use 127 Amniotic fluid stem cells. In order to test for chromosomal abnormalities or infection during pregnancy, fluid may be withdrawn from the amniotic sac in a procedure called amniocentesis. 128 Roughly one percent of the cells in the extracted fluid have been identified as stem cells with either c Kit (CD117) or Oct4 positivit y. 32 129 These amniotic fluid stem (AFS) cells are highly proliferative, have been differentiated into cell s from all three germ layers, and if positive for both Oct4 and c Kit form embryoid bodies. Amniotic fluid stem cells have cell surface markers and gene expression patterns that are distinct from ES cells. 32 In addition, they do not develop spontaneous tumors when implanted into immunodeficient mice. 129 There is no curre nt AFS cell therapy, but these cells have been shown to integrate into adult tubules and early nephron structures. 32 130 Also, using growth factors, AFS cells have been differentiated into mesenchymal and epithelial cells using growth factors, with one derivation showing expression
46 of a podocyte marker. 32 Despite these achievements, there is still little research in directed differentiation of AFS cells into diverse renal lineages. Finally, one specific drawback of routine use of amniotic fluid stem cells is that there is a minor risk of miscarriage associated with amniocentesis. 128 Induced pluripotent stem cells. Embryonic stem cells are an attractive source for tissue engineering applications because of pluripotency and high proliferative capacity. However, potential immune response and ethical considerations in obtaining the se cells are substantial drawbacks. One solution is to reprogram adult autologous cells into a pluripotent stem cell state. This has been accomplished with human dermal fibroblasts by retroviral transduction of four transcription factors, Oct3/4, Sox2, Klf 4, and c Myc. The resulting cells, dubbed induced pluripotent ste m cells mimic embryonic stem cells with similar morphology, proliferative capacity, gene expression, surface antigens, and telomerase activity. 33 They can be differentiated into mesodermal, endodermal, and ectodermal lineages through embryoid body formation, and have been differentiated to cells such as neurons, cardiomyocytes, adipocytes, and panc islet cells. 122 131 132 The primary foci of IPSC research has since been identification of new accessible cell sources and devising reprogramming protocols for minimal modification and removal of the genetic imprint. 133 Once the cells reach the embryonic state, lingering modified expression patterns can inhibit or influence differentiation potential. 122 Recent efforts of nuclear reprogramming include factor substitution or process efficiency improvement with use of small molecules, direct delivery of reprogramming factors, and transfection of synthesized mRNA. 133 Induced pluripotent stem cells have also been generated from bone marrow cells, kidney mesangial cells, cord blo od stem cells mesenchymal cells in cord matrix and placental amniotic membrane, peripheral blood cells and exfoliated tubule cells excreted in urine. 43 44 134 136 The
47 primary advantage to IPSC as a cell source is that reprogrammed autologous cells can be used to generate tissue with out immunogenic rejection. 137 Further, cell sources can now be easily acquired from urine or peripheral blood without need for tissue biopsy 44 138 One caution is that adult cells may be susceptible to age, mutation, and disease. 135 As an alternative, cord blood, umbilical cord matrix, and placental amniotic membrane can be banked with little ethical concern for donor matched applications. 134 135 Embryonic stem cells and induced pluripotent stem cells have been induced to differentiate to similar renal lineages. In one study, both achieved mesoderm, metanephric mesenchyme, and tubule differenti ation. However, ES cells were more responsive to differentiation factors and IPSC s had delayed loss of Oct3/4 expression. This may be related to the method of cell reprogramming and continued expression of transduced transcription factors. 122 Relatedly, it has also been shown that IPSC s have a predisposition to differentiate toward their source tissue type due to epigenetic patterning (histone modification or DN A methylation), and are more resistant than native ES cells to progress to intended lineages. 133 This predisposi tion may be partially mitigated by progressing the cells to intended phenotype and then reverting; however, this may not be practical in application. 139 Finally, as with ES cells, given high proliferation and diverse differentiation potential, use of these cells carries a risk of teratoma formation.
48 CHAPTER 3 KIDNEY DECELLULARIZATION Experimental Plan Introduction The production of tissue engineering scaffolds throu gh the decellularization of tissue has been repeatedly shown to be successful with an assortment of tissue types. These scaffolds provide a biologically active tissue appropriate template for cellular repopulation with the ultimate goal of regenerating a f unctioning organ. A whole organ preparation is especially valuable as it retains organ shaped three dimensional extracellular histoarchitecture. It has been postulated that only an intact organ extracellular matrix best represents the conjoined tapestry of microenvironments in vivo and is therefore best able to facilitate the generation of whole functioning complex tissues. 140 This is particularly relevant to the kidney, as its vital filtration function requires a precisely coordin ated arrangement of numerous cell types with heterogeneous underlying extracellular matrices. There are many avenues for decellularizing tissue, including a combination of mechanical agitation or strain, freeze thawing, osmotic disruption, enzyme exposure, and acid, base and detergent wash. 11 The selection of the optimal protocol is dependent on the type of tissue, and factors to consider are cellularity, density, lipid content, and size 108 In the past several years, decellularization protocols have been successful in producing intact whole organ extracellular matrix scaffolds. The primary mechan ism for cell removal is perfusion of detergent based solutions through the innate vascular network via cannulation of the arterial supply. This approach for fluid delivery takes advantage of the intrinsic system for oxygen and nutrient distribution and not only maximizes exposure of the tissue to the decellulariz ation solutions but provides an avenue for resulting cellular debris removal. Typically, once the organ is
49 decellularized, the vascular network remains intact and may then be used as convenient syst em for cell delivery and distribution. Whole organ scaffolds have been successfully produced from lung, heart and liver using perfusion based decellularization. They each showed preservation of key extracellular matrix proteins and effective removal of DN A. Further, each has been used as a scaffold for organ specific cellular repopulation. 18 20 27 A protocol for rabbit whole organ kidney decellularization was published in 1979 by Brendel et al Detergents and enzymes were circulated into the renal artery in a specific sequence over two days to produce a translucent achromatized kidney. 141 However, the purpose of the protocol was for the isolation and study of the renal extracellular matrix; specifically the glomerular basement membrane (GBM) The study showed that the produced renal ECM had a preserved b asement membrane ultrastructure and retained a degree of glomerular charge and size selectivity Perfusion of a red blood cell solution throu gh the preserved vascular bas ement membrane network showed no leakage of cells across the GBM and into the tubules; in addition a collectible urinary filtrate was produced. Given the intention of the protocol and that it was developed before the advent of tissue engineering and regenerative medicine, it was never assessed as a potential cellular scaffold. In t he following year, that same group released a paper that contained a simpler vascular basement membrane (VBM) preparation protocol 142 That protocol was designed to be used on multiple tissues including rat kidney and it is t hat protocol which was first evaluated in this research for production of a whole kidney tissue engineering scaffold. The remainder of this chapter will detail initial experimental plans and results, protocol improvement strategies and subsequent results, a discussion of the approaches, and considerations for future work in regard to the production an acellular kidney scaffold with
50 potential to support renal tissue engineering applications and the eventual generation or repair of functional renal tissue. De tailed methods and materials are described in Chapter 6 Organ Harvest Kidneys for use in the decellularization experiments were ha rvested from Sprague Dawley rats using a variation of the protocol described in Adin et al. (2005). This procedure allowed fo r the isolation and direct cannulation of the right renal artery in situ 143 In addition, administration of an anticoagulant and vasodilator minimized clotting and vasoconstriction during kidney harvest. Once the cannula was secure, the organ w as rinsed with physiologic saline, extracted, and immediately attached to the decellularization system. The organ was suspended by the cannulated renal artery and connected to the perfusion system via a L uer fitting. Optionally, the ureter may have been cannulated before kidney removal to measure urinary filtrate. During the procedure, t he animal w as under full anesthesia and once the organ was extracted it was euthanized. All animal work w as done humanely and according to guidelines supplied by the Institute for Lab Animal Research Guide for the Care and Use of Laboratory Animals and approved by the University of Florida (UF) Institutional Animal Care and Use Committee. Kidney Decellularization Decellularization of the kidney was performed with the LifeSustainer 1000 automated organ perfusion workstation. It was introduced in Sadri et al. (1997) and used for organ perfusion experiments in Adin et al. (2005). 143 This system perfuse s a small animal organ with constant flow rate or pressure at a desired temperature. 144 In addition, sensors for pH, oxygenation, and carbon dioxide saturation allow this system to be used for any subsequent perfused organ culture experiments. In the following experiments, harvested kidneys were perfused in a pulsatile manner within rat physiologic tolerances (~120/80 mmHg). The perfusate
51 was drawn from a reservoir and once passed through the kidney either recirculated through a filter or collected for disposal or analysis. The decellu la rization protocol described by Brendel and Meeza n (1980) is a stepwise perfused organ protocol that was designed to lyse cells and progressively remove material without causing blockage; this is especially relevant in the miniscule tubular network of the kidney. 142 The protocol, VBM is detailed in Chapter 6 ( Table 6 1). Briefly, 2% Triton X 100 was perfused through the organ fo r 48 hours to solubilize hydrophobic cell membranes, lyse cells, and wash cytoplasmic contents; followed by buffered DNase to digest any bulky, viscous nuclear material; and then more Triton X 100 to assist in further cellular debris removal. Interspersed were de ionized water flushes to osmotically challenge remaining cells and rinse solutions. Scaffold Analysis Initial evaluation of the decellularized kidneys was accomplished with routine histology. The prepared scaffolds were fixed in 10% neutral buffere d formalin, embedded in paraffin, sectioned, stained with hematoxylin and eosin (H&E), and viewed using bright field microscopy. This preparation method has a high degree of tissue structure preservation and hematoxylin highlight ed any DNA remaining in the scaffold. DNA and other cell debris can hamper cell growth and trigger destructive immune response in an implanted extracellular matrix scaffold. 40 145 146 The sections of tissue were interpreted for removal of cell debris and retention of extracellular morphology. Specifically, the functional regions or zones of the kidney, the cortex and the outer and inner medulla, were inspected for intact cell or nuclear membranes, DNA remnants, and other cellular debris. In addition, the vascular, glomerular, and tubular basement membranes were inspected for expected morp hology and continuity. Further analyses such as electron microscopy and protein IHC supplemented successful outcomes.
52 Results Scaffold Production Kidneys from the SD rats were successfully cannulated, secured, and rinsed (Figure 3 1). The goal of the surge ry was a complete, low resistance organ flush with no visible areas of blood pooling; it was thought that such would be most amenable to the proceeding organ perfusion steps. Factors that contributed to success were speed of the surgery, control of hemorrh age, minimal organ manipulation, and secur ing the cannula without leakage. Once the harvested and rinsed kidneys were secured to the perfusion system (Figure 3 1), the VBM decellularization protocol was initiated. However, it became readily apparent that t he pressure control algorithms were having difficulty adjusting to the changes in both resistance and fluid viscosity experienced during a decellularization sequence. The result was periodic spikes in pressure as reported by the inline transducer. A switch to flow control showed some, yet not significant improvement. In addition, over time, miniature air bubbles appeared within the tubing and were not effectively removed in the air trap. In cases, it appeared these air bubbles worked into the perfusing kidn eys and likely embolized many of the internal vessels. Efforts were made to control any air intrusion, but to no avail; the tubing fixtures and fluid switches in the system were likely ill equipped to adapt to long term detergent perfusion. Regardless, the protocols were completed and decellularized kidneys were produced. Scaffold Evaluation Decellularized kidneys appeared white and semi transparent and of a similar size and shape to the harvested kidneys (Figure 3 1). The organs were compressed and exhibit ed sponge like re expansion. Histological evaluation of the organs showed recognizable structures, but with very large cavernous defects and a great deal of proteinaceous debris. The cause of the defects was likely pressure spikes resulting from a combinat ion of air emboli in the kidney vasculature
53 A B F E D C Figure 3 1. Organ harvest and perfusion based decellularization A) Kidney harvest with in situ saline perfusion, B) harvested kidney with both arterial and ureteral cannu lation, C) LifeSystem 1000 organ perfusion sys tem, D) decellularized kidney, and H&E stain (10x magnification, bars denote 200m) of E) normal kidney and F) organ decellularized with machine driven organ perfusion system; note the damage to ECM ( VBM Mod 1 protocol).
54 and rapid release of cellular lysates. In order to slow the initial release of cellular material during decellularization, a modified protocol, VBM Mod 1 was devised which decreased the detergent concentrati on of the first perfusate solution. This change was inspired by an earlier paper from Brendel et al. (1979) that targeted the preparation of an acellular rabbit kidney. 141 This change resulted in marginal improvement but the considerable damage to the ECM remained (Figure 3 1). Interestingly despite this damage, it appeared most of the cells were lysed with only a few isolated patches of intact cellular or nuclear membranes. Also there were no identifiable areas of hematoxylin positivity, suggesting all of the DNA had been removed or was diffused within the visually eosinophilic proteina ceous debris. Procedure Improvement Elevated Reservoir Perfusion Apparatus The initial results have shown that a structurally intact decellularized kidney could not be achieved with the LifeSustainer 1000 automated organ perfusion workstation. A system red esign to include retrofitted tubing fixtures, a mechanism to relieve pressure spikes, and modification of the control algorithms for improved pressure control would have been prohibitive in time and cost, especially given the system and control algorithms were proprietary and would require substantial reverse engineering. Therefore, a new, simpler apparatus was designed to perfuse kidneys. The driving principle behind the new organ perfusion system was pressure control based on height of an elevated perfusa te reservoir. Tubing from the reservoir directed the decellularization solutions through a vented inline filter, and then down through a split that allowed two organs to be attached. Perfusate that passed through the kidney was collected in a common beaker and then returned by peristaltic pump to the elevated reservoir. A transducer was placed at the level of the kidneys to assure desired pressure. Even though the principle is based
55 on a static fluid assumption, the pressure remained stable and was maintain ed as long as there were no leaks in the system or gross ruptures in the kidney Design and images of the apparatus are detailed in the appendices (Figure s A 1 and Figure A 2 ). This simple system filtered debris, minimized air entrapment, and provided a co mmon solution source for two decellularizing organs This apparatus was used for all subsequent scaffold preparations. Decellularization Protocol Improvement The Brendel and Meezan VBM decellularization protocol (1980), implemented with the LifeSustainer 1 000 automated organ perfusion workstation produced kidney scaffolds with large extracellular matrix defects and substantial proteinaceous debris, yet successfully lysed most of the cells. To address the gross ECM defects, a new organ perfusion system with better pressure control was devised and constructed. However, further modifications to the decellularization protocol were also thought necessary to reduce residual cellular debris. Therefore, iterative adjustments were made such as increased detergent con centration and duration, the addition of ionic detergent and detergent additives, and the substitution of detergents. These changes and results are described below and summarized in Table 3 1. The protocols are fully detailed in Chapter 6 (Materials and Me thods). Table 3 1 also includes the number of kidneys that were decellularized by each of these protocols, many of which were involved in cell implantation experiments described in the following chapter. It should be noted that the overall goal of this res earch was to produce an acellular kidney scaffold that could be evaluated as a cellular scaffold for tissue engineering, and ultimately success would be judged by cellular response. It was not the primary goal to necessarily optimize a protocol for deterge nt times, concentrations, and precise sequence. However, modifications were tested in order to identify potential parameters or arrive at a protocol which would result in a dramatic improvement in debris removal, as such a produced kidney was thought to be a superior baseline construct from which to evaluate.
56 Table 3 1 describes a basic system for categorization of decellularization quality. Determinants include presence of intact nuclear envelopes and amount of residual cell debris in three areas of the ki dney: cortex, inner medulla, and outer medulla. Pictorial examples of each category are represented in Figure 3 2. Note that these categories are based on an objective classification of the group as a whole, and do not assess the level of variation within the scaffolds produced by a specific protocol. Acellular perfused kidney protocol. Kidney Mod 1 was the first decellularization prot ocol that was compiled and used on the new elevated reservoir perfusion apparatus. It protocol (1979). 141 That protocol was optimized for rabbit k idney and preceded their subsequently published VBM protocol ; their VBM protocol was the one used in these initial Table 3 1. Decellularization quality of kidney scaffolds produced with the elevated reservoir perfusion apparatus and detergent based protocols Decellularization Protocol # DQ Modification Kidney Mod 1 2 G M Kidney protocol with addition of ionic NaDC and additional DNase step Kidney Mod 2 3 G Kidney Mod 1 with lengthened detergent perfusion durations Graduated 3% 2 G Gradually increasing detergent conce ntrations Graduated 10% 5 VG Similar to Graduated 3% with higher concentrations Kidney Extended 4 VG Similar to Kidney Mod 2 with longer detergent perfusion durations Kidney NaDC 16 G Similar to Kidney Extended with reduced detergent durations and addit ional detergent step Kidney SDS 19 E Equivalent to Kidney NaDC with substitution of SDS for NaDC #: Number of scaffolds produced with given protocol; DQ: typical decellularization quality; G: good, glomeruli free of debris, patchy zones of debris in cort ex, most debris located in outer medulla, diffuse debris patterns in inner medulla, possible isolated areas of intact cellular or nuclear membranes; VG: very good, all areas of kidney improved, minimal debris in cortex and inner medulla, some debris still in outer medulla, no intact cell remnants; E: excellent, only trace debris in entire kidney; NaDC sodium deoxycholate detergent; SDS, sodium dodecyl sulfate detergent.
57 E C D A B F Figure 3 2. Improved decellularization protocols for removal of cell debris A D) H&E histochemistry (10x magnification, bars denote 200m). A) Normal kidney. B D) Protocols showing increasing decellularization quality, B) Kid ney Mod 1 : good decellularization quality, C) Graduated 10% : very good decellularization quality with reduced debris in the scaffold, D) Kidney SDS : excellent quality with only trace debris in the scaffold. E) Kidney SDS : image of whole decellularized kidn ey, note the increased translucency. F) Kidney SDS : Bright field image showing intact vascular network (4x magnification).
58 decellularization experiments (Brendel, 1980). 142 Even though the VBM protocol was considered an updated version, the acellular perfused kidney protocol included steps that appeared to be more aggressive and had the potential to be more effective at cell debris removal and removal of cellular membranes. In com parison to the previous VBM protocol, this protocol includes the detergent additives CaCL 2 and MgSO 4 to promote cell permeability and encourage activation of endogenous nucleases, NaCl to osmotically assist in nuclear membrane rupture, and the ionic deterg ent sodium deoxycholate ( NaDC ) to solubilize previously insoluble protein aggregates and improve cell debris removal. In addition, an intermediate DNase treatment was added to dissuade accumulation of cumbersome nuclear material, which even though not appa rent at the conclusion of the previous VBM protocol (via H&E stain), may have still aggregated and blocked protein removal during the protocol. In an act of caution in contrast to these more aggressive steps, the durations of solution exposure was kept at a minimum. Kidneys produced with the Kidney Mod 1 protocol and with the elevated reservoir perfusion apparatus contained no ruptures of the extracellular matrix, and glomerular, tubular, and vascular basement membranes appeared intact and with expected mor phology (Figure 3 2). All of the cells and a majority of eosinophilic cellular debris was removed from the cortex. However, cell debris was spread throughout the outer and inner medulla, with areas of intact nuclear envelopes. Lack of hematoxylin contrast suggested that the DNA had been successfully remov ed or perhaps that the remaining nucleic elements were diffused in the proteinaceous debris so as not to be apparent. The next iteration of protocol adjustment, Kidney Mod 2 involved the lengthening of mos t of the detergent steps in Kidney Mod 1 so that the total protocol duration was almost doubled, similar to the VBM protocol. Oddly, this did not yield notable improvement.
59 High detergent concentration. Two kidney decellularization protocols were developed which involved gradually increasing nonionic detergent (Triton X 100) concentrations, one with a peak concentration of 3% and one with 10%. This was done in an effort to temper any sudden release of cell lysates so as to not overwhelm the ability of the c irculating perfusate to effectively wash away cellular material. The inclusion of DNase, detergent additives, and sodium deoxycholate steps were incorporated as in previous protocols, although these two protocols only contain one DNase step. The previous p rotocols only included a maximum 3% nonionic detergent concentration; the G raduated 10% protocol magnifies this significantly to explore the effects of higher detergent concentrations. Additionally, the G raduated 10% protocol included more steps with the a ddition of sodium chloride to improve lysing of nuclear membranes. Histological analysis of acellular scaffolds produced with the Graduated 3% protocol showed no meaningful improvement over previous protocols. However, the Graduated 10% protocol yielded a scaffold with significantly reduced cellular debris and an absence of identifiable cellular remnants (Figure 3 2) Extended duration. It was determined that a graduated increase of detergent level to a high concentration produced acellular kidneys with less cellular debris and a sharp reduction in identifiable cell remnants such as nuclear membranes. The next set of protocols tested the effects of exaggerated detergent exposure without the increase in concentration. Total protocol duration was extended to 7 days, with almost 6 days of Triton X 100 treatment with calcium chloride ( CaCL 2 ) magnesium sulfate ( MgSO 4 ) and sodium chloride ( NaCl ). In addition, the NaDC duration was increased to 20 hours, at least double that of the previous procedures. The two proto cols, Kidney Extended and Kidney Extended 2 only contained one difference; Kidney Extended had a DNase treatment interrupting the mentioned Triton X 100 step, whereas Kidney
60 Extended 2 did not. Fortunately, the acellular kidneys produced using either proto col showed a very good decellularization, with characteristics similar to the Graduated 10% protocol. In addition, there was little noticeable difference between the two protocols, calling into question the necessity of the intermediate and costly DNase st ep. Convenient timing. Tested protocols have suggested that high detergent concentration and prolonged exposure positively affect cellular debris removal during kidney decellularization. The next protocol was developed to combine the more aggressive deterg ent steps of the acellular perfused kidney protocol (Brendel, 1979) with the convenient 24 hour based timing of the VBM protocol (Brendel, 1980). 141 142 The resulting protocol is shorter in duration than the Kidney Extended protocols, yet retains the long sodium deoxycholate exposure. This was done to see if ionic detergent exposure was a strong factor for ef fective debris removal. Other changes were an extension of the DNase step and removal of the initial 0. 5% Triton X 100 treatment. The kidney scaffolds generated from the Kidney NaDC protocol unfortunately were not as good as the ones generated from the hig h concentration and extended duration protocols. However, at this time, scaffold cell seed experiments were progressing and showed promising results on this preparation. Therefore, many of these scaffolds were produced to provide an unwavering baseline for the cell seed experiments. The alterations in the decellularization protocol was not completely frozen, however, as a shift in ionic detergent proved to be a very successful as described in the next protocol. Ionic detergent change. The two published proc edures which inspired the decellularization protocols used in this research (Brendel, 1979; Brendel, 1980) were pursued because they had been designed for and tested on the kidney. 141 142 However, the procedures were not developed for the purpose of production of a scaffold for tissue engineering n or
61 optimized for debris removal. A survey of contemporary publ ished protocols for decellularizing tissue suggested a change in detergents may have yield ed more promising results. Sodium dodecyl su lfate (SDS) has been used alone or in conjunction with other detergents or agents to decellularize tissue. It is an ionic detergent and in comparison to NaDC SDS is reported to be superior in removal of nuclear remnants and cytoplasmic proteins. However, it has also been described as disruptive to native tissue structure and collagen integrity and results in a reduction of g lycosaminoglycan (GAG) concentration; though on the whole less than NaDC 11 Sodium dodecyl sulfate has been used to decellularize artery and vein, lung, liver, and heart. 20 23 25 26 101 103 147 In order to evaluate SDS, a simple replacement of the ionic detergent NaDC in the Kidney NaDC protocol with an equivalent concentration of SDS was dictated ( Kidney SDS ). As all other parameters wou ld remain fixed, a direct comparison could be made. The decellularization protocol utilizing the SDS ionic detergent was a substantial improvement and produced kidneys with no identifiable intact cells and only trace eosinophilic debris. The renal extracel lular matrices all appeared intact and with expected morphologies. The generated scaffolds as a whole appeared highly translucent, with easily recognizable vascular networks (Figure 3 2). Consequently, this protocol was used for the generation of numerous acellular kidneys for use in scaffold seeding experiments. Further, the response of cells to scaffolds produced using two different ionic detergents could be directly evaluated. Finally, given the success of the use of SDS in decellularizing kidneys, and t hat SDS is effective at DNA removal, a simple preliminary experiment was conceived which subjected two rat kidneys to a one day perfusion of SDS followed by a deionized water flush (1 Day SDS). 26 The resulting scaffolds achieved debris removal similar to the full Kidney SDS protocol, indicating significant potential for time and step reduction.
62 Supplementary Scaffold Analysis Scanning E lectronic M icroscopy The primar y criteria for determination of success of a produced decellularized scaffold in regard to the discussed series of protocols has been extracellular matrix morphology preservation and removal of cell lysates as interpreted by H&E histochemistry. However, ot her key factors for consideration were preservation of continuous three dimensional structure and retention of essential extracellular matrix proteins. Scanning electron microscopy (SEM) is an imaging modality that provides a surface scan of a material at submicron resolution. It has been used on whole organ ECM scaffolds to make a general assessment of three dimensional topographical patterning, including basement membrane continuity, and to verify removal of cellular debris. 24 27 141 148 Therefore, an SEM image was taken of a transverse kidney section produced with a high detergent exposure protocol ( Graduated 10% ). Despite this aggressive treatment, the scaffold showed remarkable preservation of the visualized extracellular matrix, and only sparse are as of protein debris (Figure 3 3). ECM Immunohistochemistry One concern with decellularization treatments, especially those with high detergent concentrations, excessive detergent exposures, and ionic detergent use is damage to the extracellular matrix pro teins that are necessary for constructive cellular response. 11 20 22 108 149 Immunohistochemistry is a histological technique that employs antibodies for contrasting target proteins on a tissue slide and it c an be used to verify retention of extracellular proteins. Collagen IV and laminin are ECM proteins that are primary structural components for renal basement membranes. 10 Collagen IV, specifically, is essential for glomerul ar filtration and has been shown to be important to laminin mediated attachment and migration of murine ES cells. 150 Laminin is a proteoglycan that is involved in a multitude of signal transduction pathways, has been linked to
63 nephrogenic processes, and has been shown to play an important role in the in vivo fate of embryonic stem cells. 10 51 Collagen and laminin IHC was performed on scaffolds generated using the Kidney NaDC and Kidney SDS decellularization protocols. As expected, collagen IV and laminin was ubiquitous and co ntinuous over the entire renal scaffold (Figure 3 3). Reference IHC for normal tissue is included in the appendices Figure B 1. Interestingly, in the scaffolds A B C D Figure 3 3. Supplementary analysis of decellularized scaffolds Scanning electron micrographs of acellular kidneys produ ced with the Graduated 10% protocol, A) glomerulus and adjacent tubules shows continuous basement membrane archi tecture arrow identifies cellular debris (300x magnification, bar denotes 100m), and B) inner medulla (65x magnification, bar denotes 500m). C D) IHC stains verifying retention of key basement membrane proteins, C) collagen IV in the cortical region of scaffold and D) laminin in the medulla ( Kidney SDS protocol, 40x objective magnification).
64 produced using the Kidney NaDC protocol, the heretofore eosinophilic cell debris appeared highl y stained for laminin yet clear of collagen IV (data not included). Cell Removal Immunohistochemistry In addition to assuring retention of extracellular matrix proteins, immunohistochemistry has also been utilized to verify the removal of cellular proteins 20 22 24 109 It offers an additional check beyond the visualized removal of eosin o philic debris through hematoxylin and eosin histochemistry. Further, antibodies to proteins specific to distinct cell populations in the kidney allow identification of those cel ls or cell lysates which resist removal. In the following chapter, embryonic stem cells implanted and cultured in decellularized scaffolds were tested against four antibodies, pan c ytokeratin, WT1, Pax 2, and Ksp cadherin to determine potential renal diffe rentiation. The target proteins are also present in the adult kidney including glomerular, tubular, and collecting duct epithelial cells. 151 155 Table 3 2 compares the effectiveness of the two most utilized decellularization protocols, Kidney NaDC and Kidney SDS at removing any trace of the indicated cellular proteins. According to immunohistochemical observation, the improved Kidney SDS protocol was successful at r emoving all traces of these proteins. The Kidney NaDC protocol, on the other hand, showed retained WT1 localized to the glomeruli; this is consistent with normal podocyte expression. Pax 2 and Ksp cadherin immunohistochemistry showed positivity in the unwa shed cellular debris, which could indicate intact cellular proteins. However, an alternate protein detection technique should be employed to exclude nonspecific binding of the antibodies to the debris Examples of these two cases, residual cellular protein s in expected location and in cellular debris are displayed in Figure B 2 in the appendices
65 Discussion Tissue Engineering Scaffold Rat kidneys have been successfully decellularized. This has been achieved through careful organ harvest, design and construction of a perfusion apparatus for decellularization, and the development of a protocol that produced intact acellular organs. The optimal solution sequenc e was based on the ionic detergent sodium dodecyl sulfate. The utility of the produced acellular kidneys as a renal specific tissue engineering scaffold, however, remains to be evaluated. The ideal tissue engineering scaffold is a structure that, once impl anted, will not result in a destructive inflammation or immune response and will direct resident or engrafted cells to functional configurations. In the case of such a scaffold being generated from organ decellularization, key parameters for success includ e the removal of all antigenic or inflammatory elements, the preservation of three dimensional structures, and the retention of biomolecules which support the desired cellular behavior. Unfortunately, most of the steps in a typical decellularization proces s modify the native organ so as to improve one parameter at the Table 3 2. Kidney cellular proteins removed in decellularization process Protocol pan CK WT1 Pax 2 Ksp cadherin Kidney NaDC + D D Kidney SDS Based on immunohistochemical observation, the Kidney SDS decellularization protocol is more effective than the Kidney NaDC protocol at removing cellular proteins in adult rat kidneys. Cytokeratins (pan CK) are intermediate filament proteins present in most renal e pithelial cells; WT1 is a transcription factor found in glomerular podocytes; Pax 2 is a transcription factor primarily localized to cells in collecting ducts and descending limbs in loops of Henle; Ksp cadherin is a kidney specific transmembrane adhesion protein located in all cells in the nephron except for those in the glomeruli, blood vessels, and papillary epithelium; it is especially pronounced in the distal convoluted tubules and collecting ducts. Key: +, visualized residual proteins in proximity to expected location in normal kidney; no visualized proteins; D, no visualized proteins in expected location, coloration indicates possible diffuse presence in unremoved proteinaceous debris.
66 sacrifice of another. For example, stronger detergents are more effective at cell lysis and debris removal but at the cost of increased solubilization or damage of extracellular matrix protein s. 11 20 22 109 This research utilizes the removal of cellular and nuclear debris, retention of collagen IV and laminin, and preservation of extracellular matrix morphology and ultrastructure by histology and SEM to gauge success. These parameters have been show n in analogous contemporary protocols to improve implanted cell viability, proliferation, and encourage site specific differentiation. 20 40 108 114 However, defined criteria for specific evaluation of whole organ kidney scaffolds in respect to renal orien ted cellular responses have not yet been established. This would require an intensive cyclical process of decellularization protocol modification, acellular kidney production, scaffold characterization, measurement of implanted cell response, and further e ducated revision of the decellularization protocol. Cell response to the scaffolds generated in this research is discussed in Chapter 4. However, it was beyond the current scope of the project to modify the scaffold generation protocol based on cell respon se feedback. Nevertheless, considering the results of the decellularization experiments, there are scaffold analysis techniques and protocol modifications that may be considered to improve outcome. Acellular Kidney Analysis The characterization of produced renal tissue scaffolds is central to making educated improvements to the decellularization process and for estimating the determinants of engrafted cell behavior. Specific target areas include verification of cell lysis and cell remnant removal, retention of biologically active extracellular matrix and matrix associated molecules, and preservation of a continuous three dimensional extracellular tissue architecture. Cell debris removal. One of the primary advantages of using decellularized tissue for an imp lantable tissue engineering scaffold is the sharp reduction of destructive host immune response as compared to a whole organ transplant. 41 42 It has been suggested that the cause is
67 removal of cellular antigens and the conservation of ECM protein structures within and across multiple species. 11 Further, it has been shown that the decellularization process will allow implants from xenograft scaffold sources, easing reliance on human donors. 98 156 Figure 3 2 demonstrates a progressive removal of cellular debris from the kidney as o b served with H&E histochemistry. In particular all treatment variants show complete reduction of nuclear hematoxylin positive staining. Hematoxylin and eosin histochemistry has been used to verify removal of DNA in other decellularization studies ; it has the advantage of showing localized areas of poor decellularization in reference to intact extracellular anatomy 21 114 Alternately, some groups diamidino 2 phenylindole (DAPI) fluorescence histochemistry 20 25 27 DAPI directly interacts with nucleic acids and can be used on hi stological slides or live cells Quantitative methods using DNA isolation and spectrophotometry are a lso popular for analyzing removal of cellular debris but they are normalized to total tissue mass or total protein and lack the capability for convenient location mapping. 22 24 27 G iven the importance of DNA removal for graft acceptance Crapo et al. (2011) suggests that H&E alone is comparatively insensitive a nd that a combination of criteria should be established for removal verification: <50 ng dsDNA per mg ECM dry weight <200 bp DNA fragment length lack of visible nuclear material in tissue sections stained with DAPI or H&E These criteria are offered as a ba seline for constructive host remodeling, yet will need to be verified, modified, and expanded according to specific tissues and implant locations. 108 The decellularization protocols developed in this research have resulted in a progressive reduction of eosinophilic protein debris in produced scaffolds as visualized by routine H&E histology (Figure 3 2). It has bee n shown elsewhere that an overall reduction in residual cell ular
68 material results in improved cell growth on implanted scaffolds. 40 In many cases, the tissue surro unding the glomeruli was clean, while the outer medulla was congested with protein aggregation. It could be interpreted that during decellularization the cell lysates from the glomerular capsules and proximal tubules travelled downstream at a rate that exc eeded the ability of the more distal tubular pathways to clear. In response, the decellularization protocols were adjusted in attempt to reduce release rate of cellular lysates or improve solubility of aggregated proteins. One caveat to the use of H&E hist ochemistry is that there is no distinction between cellular or extracellular proteins. Therefore, other contemporary protocols have incorporated techniques to recognize cell specific antigens for verification of cell debris removal. 20 22 24 109 These technique s include IHC protein immunoblot (western blot), and enzyme linked immunosorbent assay (ELISA). Immunohistochemistry allows visualization with associated structures while western blot and ELISA provide a level of quantitation. ELISA is the most accurate y et comparatively costly. 157 In this research, IHC analysis was performed on decellularized scaffolds with a small handful of antibodies. These antibodies covered an assortment of cells, including all epithelial cells in the nephron. It showed that the Kidney SDS decellularization protocol was clearly superior to Kidney NaDC in terms of cell debris removal. The antibodies used also provided some degree of localization information. For example, detection of Ksp cadherin without Pax 2 may indicate retained cellular proteins from the In other decellularization research, the proteins that have been used to verify cell removal include MHC1 (major histocompatibility complex 1) actin as generic, ubiquitous cellular proteins and CD31 or PECAM 1 (platelet endothelial cell adhesion molecule) as endothelial cell markers. 20 24 MHC2, a protein that specifically activates host immune response, has been targeted as well. 24 Others have also used markers that
69 identify the cells specific to the organ. 22 109 Given the diverse arrangement of cells in the kidney, it may be beneficial during decellularization protocol optimization to use a set of markers that uniquely identifies each of the diverse cell populations. Yabuki et al. (201 1) suggest all cell types on the kidney can be identified by the binding patterns of a particular group of lectins. Lectins are highly specific sugar binding proteins involved in biological recognition phenomena. 158 Another paper uses a combination of intermediate filament proteins such as individual cytokeratins and vimentin to distinguish normal tissue from renal epithelial tumors; these could also be used to identify each group of kidney cells. 153 The use of these types of marker sets will prove that cell phenotypes ha ve been removed, or if not, which should be targeted. Finally, the SEM images acquired in this study were taken to show a continuous preserved extracellular matrix; however, protein aggregates were also easily recognized (Figure 3 3). Scanning electron mic roscopy has been to verify cell material removal in other studies, but similar to H&E histochemistry, it lacks ability to distinguish between cellular and basement membrane proteins. 24 ECM retention. Vascular, glomerular, tubular, and collecting duct basement membranes are specialized extracellular matrices that support adhesion, growth, polarization, and differentiation of the numerous types of epithelial cells t hat are in the developing and adult kidne y 10 51 Many of the constituent molecules are directly involved in nep hrogenesis and have been shown to direct stem cell differentiation. 30 51 159 163 Further, it is thought that native extracellular protein architecture is optimal for organ specific cell behavior and is the most suited environment for differentiation of stem cells into a targeted tissue phenotype. 18 30 159 164 Pr imary component proteins of renal basement membranes include collagen IV, laminins, entactin/nidogen, and an assortment of proteoglycans. 10 46 51 Methods for protein localization or
70 quantification can be used to assess the preservation of these essential proteins in decellularized kidney tissue. Laminin and colla gen IV immunohistochemistry was performed on decellularized kidneys (Figure 3 3), and as expected both of these antibodies showed complete and continuous coverage in the tissue scaffold. 10 47 165 Interestingly, scaffolds produced with decellularization protocols utilizing NaDC showed the proteinaceous debris as hea vily positive for laminin and negative for collagen IV (data not shown). This may indicate that the NaDC protocols cause sloughing of laminin during perfusion, whereas SDS based protocols do not. Alternatively, it may imply that SDS based treatments, in wh ich scaffolds are mostly devoid of debris, are actually more effective at solubilization of basement membrane proteins. This data should be considered preliminary, however, until alternative methods of protein quantification can be used for validation, and to rule out unspecific laminin antibody binding or entrapment in the proteinaceous debris. The use of IHC for extracellular protein detection is commonplace in analysis of acellular tissue; however, there is usually little effort expended to identify spec ific collagen or laminin isoforms. 18 20 22 23 25 27 40 114 Given the heterogeneous structure of kidney ECM, preservation of each of the isoforms in their native locations may be crucial in establishing functioning tissue and showing this would be a viable continuing step in this research. 10 In addition, methods for quantification of protein loss during decellularization would be valuable. Even though IHC provides location based information, and even with densitometric analysis, total quantities are difficult to interpret. 166 Some research teams have used a colorimetric assay for hydroxyproline concentration, which would approximately quantify loss of collagen. 24 26 103 Others have used western blot. 18 22 ELISA may also be empl oyed 157 These techniques are not location specific, yet perhaps m ultiple samples of various regions of renal tissue through careful
71 dissection would yield more focused results. These quantitative measures may be normalized through lyophilized weight or an estimation of total protein. 24 Recently proteoglycans, or associated glycosaminoglycans (GAGs), have become more common targets of acellular scaffold analyses. 18 24 26 27 1 14 Proteoglycans are involved in cell adhesion, growth factor binding and delivery regulation, and are necessary for a multitude of nephrogenic processes. 48 49 52 54 56 57 167 They have been shown to be necessary for renal growth and development, and the removal of GAGs and other adhesive proteins slow cellular migration and reduce cell growth on acellular tissue. 51 105 149 Key proteoglycans in renal basement membranes are perlecan, agrin, bamacan, and col lagen XVIII. All those listed are heparan sulfated proteoglycans (HSPGs) and carry primarily heparan sulfate GAG chains, with the exception of bamacan, a chondroitin sulfate proteoglycan. Heparan sulfate proteoglycans are required for mesenchymal condensat ion and initiation of branching morphogenesis. 53 167 Perlecan is in all adult renal basement membranes, impro ves stability under mechanical stress, and is present in ureteric bud basement membrane. 50 51 Agrin has a stron g presence in GBM, 50 51 168 Bamacan is present in all adult renal BMs except GBM, although it is a lso seen in the developing glomerulus. 10 Collagen XVIII plays a role in mechanical integrity of the glomerular and tubular basement membranes, is implicated in Wnt signaling pathways, and is required for ureteric bud branching. 51 1 69 The identification and quantification of proteoglycans or specific glycosaminoglycans after the decellularization process would be an informative future addition to this research. One method for GAG quantification is a spectropho to metric approach that uses dimethylene blue dye. 24 26 This approach can distinguish between chondroitin, dermatan, heparan and ker atan sulfates using polysaccharidases. 170 On the other hand, less quantitative yet location specific in formation can be
72 obtained with IHC Alcian blue is a cationic dye that has been used on decellularized constructs to highlight GAGs. 24 Alternatively, Nakayama et al. (2010) used heparan sulfate proteoglycan IHC to observe decellularization induced changes. 114 Any of these methods would further improve the characterization of acellular kidney scaffolds. As mentioned, extracellular matrix pr oteoglycans may sequester growth factors, serve as a delivery reservoir, and act in their regulation. Growth factors stimulate cell proliferation and differentiation and serve roles in ECM protein regulation. 52 54 55 Unfortunately, the process of decellularization, especially protocols containing ionic detergen t, may cause a sharp reduction in ECM bound growth factors. 11 70 However, this needs to be verified with the ki dney scaffold, especially if GAGs survive the treatment. De Kock et al. (2011) detected surviving VEGF in a decellularized liver after both nonionic and ionic detergent perfusion. 21 Nakayama et al. (2010) has shown with IHC that HSPGs while reduced, may survive decellularization of renal tissue. 114 Prepared acellular tissue that is growth factor depleted, yet retains preserved GAGs, may be exposed to growth factors in solution for re association. Lai et al. (2006) has used this technique with bovine pericardia and basic fibroblast growth factor. gradual growth factor release in vitro and improved angio genesis in vivo 171 Growth factors that bind to extracellular proteoglycans and play a role in renal function or development include, among others, VEGF, PDGF, HGF, FGF, TGF 7, and Wnt 4. 52 54 56 VEGF binds to perlecan and is responsible for rena l vasculogenesis, glom erulogene sis and tubulogenesis as well as endothelial cell differentiation. 56 172 PDGF modulates endotheli al functions and enhances ECM protein synthesis. 52 TGF 55 173 HGF is expressed in the metanephric mesen chyme and regulates tubulogenesis. 52 FGFs interact with the GAG chains of syndecan and perlecan, mediate activation of WT1, are involved in matrix
73 product ion, are critical to growth and patterning of all renal lineages, and play roles in early and late stages of development. 174 175 BMP7 binds to the ECM protein fibrillin 1, plays a role in branching morphogenesis, and mitigates secretion of pro inflammatory cytokines by tubular endothelial cells. 176 177 In addition, BMP7 has been specifically shown to differentiate embryoid bodies into intermediate mesoderm according to increased Pax 2 expression and ability of cel ls to integrate into developing tubular epithelia. 121 Wnt proteins are lipid modified glycoproteins that are stabilized by HSPGs. 54 Wnt 4, specifically, facilitates tubular regeneration in the adult kidney and is required for tubulogenesis through induction of mesenchymal to epithelial transition. 138 178 Also, WNT4 with enhancement from HGF and Activin A can differentiate murine embryonic stem ( mES ) cell into renal tubular cells. 179 The identification of growth factors in decellularized tissue or in modified tissue is typically a ccomplished with an ELISA assay 149 157 171 Immunohistochemistry may also be employed. The preservation or manipula tion of growth factors in the decellularized kidney scaffold may be key to directing regeneration and should be in strong consideration for any proceedings studies. Structural preservation. The kidney has elegant and moderately tortuous three dimensional s tructures which efficiently direct blood distribution and filtration; metabolite, ion, and water reabsorption; and collection and removal of urinary filtrate. The preservation of native kidney extracellular architectures may play a key role in the generati on of newly functional tissue, facilitating the correct anatomical arrangement of populating cells. Other decellularized organ matrices have been used to generate functional organs such as a beating heart, a lung capable of oxygen and carbon dioxide exchan ge, and a liver that secretes albumin. 22 24 26 Although each of these tissue engineered organs is still being devel oped, they are arguably the most advanced solutions yet devised, and have insofar the greatest potential to become clinically
74 relevant. All of the decellularized organs used to generate these structures have demonstrated intact native extracellular archite cture, including functional vascular networks. The kidney scaffolds produced with this research have retained their overall shape with a visually identifiable vascular tree. In addition, all relevant glomerular, tubular and collecting duct structures remai ned intact and appeared continuous with expected sizes and segmentation. It has been suggested that preserved morphology improves cell growth in decellularized scaffolds. 40 The morphology of the kidney scaffold was inspected using gross visualization, stained tissue slides, and SEM; however, other avenues for characterization exist that may enhance elucidation. Validation of vascular system integrity could be perfor med with injected Evans blue dye or contrast for microCT. 23 24 Considering histology, identification of disconti nuities could be improved using specialized dyes that enhance basement membrane contrast such as Periodic acid 180 Scanning electron microscopy was used to examine surface topography of the kidney scaffolds. The s caffold selected was produced with a high detergent concentration decellularization protocol with NaDC as the ionic detergent. This is arguably one of the more aggressive treatments with most potential for extracellular matrix damage and it was assumed tha t this would be a worst case representation. However, the SDS based protocol, which was superior in debris removal, was never successfully tested so it may be prudent in the future to validate with additional SEM studies. TEM (transmission electron microsc opy) is another imaging modality that is typically used to inspect cell morphology and cell cell and cell matrix interaction in tissue, often with the use of heavy metal immunolabeling. It has also been used, though, to examine ECM morphology and ultrastru cture. For example, TEM has been used to examine lung ECM, verify elastin and collagen fibers in explanted acellular tissue engineered constructs, and inspect collagen bundle size, grouping, and spacing in decellularized porcine
75 pulmonary valves. 24 181 182 Brendel et al. (1979) used TEM and perfused pa rticles to examine glomerular filtration properties of decellularized kidneys 141 I t may be beneficial to repeat these experiments with the SDS derived scaffold to assure that filtration is not diminished and that basement membrane ultrastructure is minimally damaged. The following chapter details the injection of embryonic stem cells in to the kidney scaffold through the renal artery. Initially, the cells remained confined to the glomeruli, likely occupying only the vascular capillaries. It may be, even though there are other routes of cell delivery, that a degree of basement membrane por osity is desired to improve migration of seeded cells into all areas of the kidney. This would require mechanical stress, excessive detergent treatment, or exposure to an ECM degrading enzyme. The tradeoff is such treatment would likely damage the ECM and constituent molecules, reducing effectiveness as a biological scaffold. A phenomenon that occurred in SDS derived kidney scaffolds seeded with cells is shrinking of overall structure over incubation time (Chapter 4). This did not occur in the scaffolds des igned with NaDC ionic detergent. It may be that incomplete removal of cell debris in the NaDC scaffolds provided resistance to scaffold deformation. It may also be that the SDS based procedures caused a change in ECM composition or component binding. Corti ella et al. (2010) saw shrinkage in seeded and incubated lung scaffolds after decellularization with SDS. 20 Nakayama et al. (2010), in which a static fluid prep aration was used, saw size reduction in their Rhesus monkey kidney scaffolds after treatment with SDS. 114 It may be that perfusion served to delay the effect for the scaffolds presented earlier this chapter. A potential solution is the decellularization of kidneys at low temperature (4C), which prevented the scaffold shrinkage in the Nakayama et al. protocol. 114
76 The kidney scaffolds have endured median physiologic fluid pressures during decellularization wi thout disruption of the ECM. This does not, however, simulate probable in vivo conditions, which include compression, shear, and exaggerated fluid pressures. Nakayama et al. (2010) showed a significant reduction in compressive modulus in a decellularized k idney after their treatment. 114 In the matter of vasculature, rat iliac arteries were decellularized using SDS and NaDC yet retained roughly 80% of native vessel stiffness, much better than artificial constructs. 101 Similar mechanical properties in engrafted vessels play a key role in graft success. Mechanical testing should be considered before scaffold implanta tion, with or without engrafted cells. Decellularization Protocol Organ decellularization is typically achieved with perfusion of solutions through the arterial supply. Success is determined based on the effectiveness of the fluid delivery apparatus and th e sequence, composition, and exposure duration of the perfusates. This process should ideally be robust to anatomical variation and provide scaffolds that have minimal residual cell material, do not illicit immune response, and support functional tissue gr owth. Crapo et al. (2011) provides a focused review of perfusion based organ decellularization protocols. 108 C irculated solutions may include detergents or acids to solubilize cell membranes and intracellular contents, chelating agents to disrupt cell attachment, or enzymes such as proteases or nucleases for cell digestion. In addition the solutions may be hypoton ic or hypertonic to provide osmotic stress for cell membrane rupture. The act of fluid perfusion naturally applies shear stress to intact cells and provides a means of washing away cell debris. Moreover, supplemental methods for cell disruption have been e mployed such as freeze thaw cycles, agitation, and sonication. 11 20 27 108 Highlighting the advantages of perfused solutions for scaffold production, one protocol achieved lung decellularization with arterial and tracheal
77 perfusion of SDS in 2 3 hours, w hereas decellularization in a rotating chamber with freeze/thaw cycles and DNase/RNase treatments required 5 weeks of detergent exposure. 20 22 The kidney decellularization protocols in this research primarily utilized non ionic and ionic perfused detergents with variations in concentration, perfusion duration, sequence, and total number of solutions. Osmotic stress was achieved with distilled water and high salt concentrations. The digestion of nuclear material was facilitated with deoxyribonuclease and the addition of divalent cations for activation of endogenous nucleases. A series of progressive protocol modifica tions have yielded a decellularized kidney with minimal cell debris, preserved morphology, and retention of primary ECM proteins. However, the process has not been optimized for preservation of other significant basement membrane molecules or reduction in total time, chemical exposure, or number of steps. Such optimization has the potential to improve throughput and reliability and yield a more biologically responsive scaffold. Organ perfusion. Two distinct perfusion systems were utilized for these decellul arization experiments. The first was the LifeSustainer 1000 automa ted organ perfusion workstation with which the perfusate was driven directly into the organ by a peristaltic pump with either pressure or flow rate control. In pressure control mode, the flo w rate of the peristaltic pump was directed by an algorithm with input from a pressure transducer. Unfortunately, t wo phenomena confounded this control system with our rat kidney decellularization model First, any air bubbles in the fluid lines, which wer e difficult to mitigate with the detergent solutions and multiple tubing fixtures, would cause fluctuations in the pressure transducer reading. In addition, these air bubbles had the potential to work their way into the kidney causing multiple emboli. A de cellularizing kidney even without these emboli would have increases and decreases in resistance depending on the rate of cell rupture, debris aggregation, and removal of material.
78 These alone or in synergy were too difficult for the propriety pressure cont rol algorithm to manage without damaging pressure spikes. Unfortunately, even with a constant flow rate setting, air emboli and cell debris caused increases in fluid resistance through the organ and corresponding damaging increases in pressure. The second perfusion system was an improvement over the first because of the simplification in plumbing and the passive gravity based pressure control. There were very few areas for bubble generation or entrapment, and the scaffolds did not show any gross damage from excess pressure. A similar system but with fluid height controlled by a glass column was used for kidney decellularization by Brendel et al. (1979). 141 Constant pressure decellularization has also been used for heart and lung. 22 23 Heart perfusion typically employs the Langendorff perfusion technique and the system can be designed with pressure or flow rate control. 183 Pressure control for this technique can also be regulated using a column or elevated reservoir. This comes at the expense of necessary excess fluid volume to achieve the effect. It may be possible to perform decellularization of the kidney using flow rate control if development of air emboli is controlled. It has been successfully used in the liver by a few groups 21 24 25 27 However, as the kidney has more complicated tubular structures it may require careful monitoring and recording of pressures to assure that a certain flow rate does not cause damage. One of the benefits of perfusion decellularization, aside from solution delivery and debris removal, is the increased mechanical stress on the cells caused from escalated pressure and shear. Of note, the elevated reservoir perfusion system developed by our lab does not deliver cyclic pulsations to the organs. This is because the peristaltic pump is only used to return solutions to the reservoir, not to drive them through the organ. This was done to minimize potential damage to the ECM that was seen in the previous system. However, it is possible the mechanical
79 pulsati ons would be beneficial for improved cell lysis, debris removal, and prevention or disruption of protein aggregation. To test this, the perfusion system may be redesigned to move the peristaltic pump into a fluid driving configuration. A bypass or tube spl it may be placed The magnitude of the pulsations delivered to the tissue will be dependent on the type of pump and the proximity to the organ. This type of system was used for the perfused organ culture detailed in Chapter 6 with a diagram included in Appendix A, Figure A 3 It may be considered that given a constant pressure, the flow rate of perfusates through a decellularizing organ will change depending on the deg ree of cell removal, the size of the kidney, and the type of solution. Characteristics such as viscosity, detergency, concentration, and charge may each play a role. Alternatively, the fluid pressure in a decellularizing organ will change over time with a fixed flow rate. While the LifeSustainer 1000 perfusion workstation was in use, it was observed that during constant pressure operation, the flow rate of a perfused detergent tended to increase to a plateau over several hours. Identification of these plate aus for each perfused solution may be a means to optimize duration of exposure. Over the course of the decellularization protocols, the flow rates, measured by drip from the organ, measurably shifted as the solutions changed. Aside from progressive increas e in flow as a result of cell removal, the flow rates were generally highest with deionized water, followed by nonionic detergent and then ionic detergent. Increasing concentration decreased flow rate, as expected. The decellularization of a kidney follow ed an expected path of flow rate changes; however, more data needs to be collected before it can be used to predict outcome. Insofar, confounding factors may be individual variants in rat kidney anatomy, wide variation in age, and improvement of surgical t echnique including cannula sutur ing for leak reduction.
80 The perfusing organs from both systems were open to the lab environment. A 0. 2 micron inline solution filter, addition of the sodium azide antimicrobial to detergent perfusates, and treatment of decel lularized kidney with an antimycotic/antibiotic treatment were sufficient to avoid contamination during organ culture. However, a re design of the perfusion system to assure the components are modular and sterilizable with an isolated organ chamber would f urther reduce the likelihood of contamination. Decellularization solutions. Kidney scaffolds have been produced by perfusing a whole organ using a prescribed sequence of solutions. The scaffolds have preserved extracellular matrix structures and basic comp onent proteins with almost complete removal of cellular debris. The predominant measure of success, though, is the ability of the scaffold to influence cells to form functional renal structures. The next chapter details the effects of the kidney scaffold o n implanted murine embryonic stem cells. These results may be used as a baseline to gauge further changes to the decellularization protocol. We have shown that increased concentration and exposure duration of perfused detergents or detergent substitutions improve cell debris removal. However, many of these modifications may actually diminish the biological effectiveness of the scaffold by degrading extracellular matrix components and stripping associated growth factors. An optimal scaffold will likely be pr oduced by determining the least aggressive chemical treatments to achieve removal of cell debris, DNA, and other cellular antigens. Ideally, each step of the solution protocol should be evaluated using scaffold characterization and basic response to cell g rowth. Detergent selection and perfusion sequence play a key role in the success of a decellularization protocol. In this research, the nonionic detergent Triton X 100 was initially used, with the addition of the ionic detergents NaDC and SDS Nonionic det ergents typically
81 disrupt lipid lipid and lipid protein interactions with preservation of protein protein interactions. They are effective at solubilizing cytoplasmic membranes and poor at disrupting nuclear membranes. Ionic detergents solubilize both cell membranes but will denature proteins with increased exposure. 11 Many organ decellularization protocols utilize both types of detergents. For kidney decellularization, most of the cytoplasmic and nuclear membranes were disrupted with the use of Triton X 100 with hypotonic and hypertonic solutions. Sodium deoxycholate was added primarily to improve removal of accumulated cell debris. However, the substitution of NaDC with SDS proved to be a much more effective means of debris removal. Trito n X 100 has been used to decellularize such tissues as heart valve, coronary artery, and liver, but usually with the addition of proteases or hypotonic solutions. 1 8 106 184 It has been paired with NaDC to decellularize kidney, nerve and heart. 27 141 142 185 It has been paired with SDS for esophagus, h eart and liver decellularization. 21 23 25 99 In many cases, SDS is called upon to remove nuclei left by a Triton X 100 treatment. 21 25 99 In other cases Triton X 100 it is used as a means to wash SDS from the organ. 23 Some treatments of lung, kidney, vein and amniotic membrane use only SDS, while SDS and NaDC were used fo r artery and heart valve decellularization. 20 101 1 05 114 147 186 187 Sodium deox ycholate has been used as the sole detergent for liver decellularization. 188 Triton X 100, SDS, and NaDC have all been used in alone and in combination to decellu larize a variety of tissue. O ther detergents o r decellularization agents have contributed to success in scaffold preparation and may be considered to improve the kidney scaffold protocol s developed herein CHAPS (3 [(3 cholamidopropyl)dimethylammonio] 1 p ropanesulfonate) is a zwitterionic detergent. A zwitterionic detergent is less denaturing than an ionic detergent but has greater solubility than nonionic detergent. 11 It has been used to decellularize lung with >99% DNA
82 reduction, remova l of MHC 1, MHC 2 actin, moderate reduction in elastin (40%) and severe loss of GAGs (10% remaining). 24 Human umbilical artery has been decellularize d with CHAPS and SDS an d retained good mechanical integrity, yet with a higher tensile modulus or less compliance than native artery. 103 Triton X 200 is an anionic detergent and has been used in nerve decellularization. It was superior to SDS overall, particularly in basic morphology and basil lamina preservation, but not as effective in cell removal. In that same study, Triton X 200 was combined with sulfobetaine 10 and sulfobetaine 16. T his theoretically allowed a tighter packing of micelles with improved solubilization and reduction of protein aggregation, ultimately improving non destructive removal of cell debris. Of the detergent combinations tested this last one proved optimum for t he creation of a nerve scaffold. 109 Tributyl phosphate is a denaturing organic solvent, yet has been shown to have minimal effect on certain ECMs. 11 When utilized for tendon decellularization, it proved to be superior to a sequence of 1% SDS and 1% Triton X 100 detergents in regard to collagen preservation and cell removal. 189 The selection of which detergents and perfusion sequence are optimal for the generation of a cellular scaffold is likely specific to the organ A particular example is the use of SDS for tissue decellularization. The reported risk o f SDS use is damage to collagen and removal of GAGs yet this may be mitigated by tissue type. 11 21 23 114 149 Reing et al. (2010) showed marked reduction in growth factors and diminished cell growth after a ddition of SDS in dermis decellularization. 149 DeKock et al. (2010) used SDS to remove nuclei in liver but it resulted in significant loss of VEGF. 21 On the other hand, Ott et al. (2008) showed no loss of GAG after decellularization of heart with SDS. 23 Further, Nakayama et al. (2010) showed not only that SDS caused no quali tative reduction in HSPG s but that overall morphology of the renal scaffold was superior to that produced with Triton X 100. 114 From this, it clear that tissues have unique
83 responses to detergents, thus highlighting the need for o rgan specific testing and characterization. Further, as in this research, the use of an alternate detergent may yield unexpected and beneficial results. Aside from substitution or trial of alternate detergents, other steps may be taken to improve the devel oped kidney decellularization protocol. The final tested protocol iterations used nonionic and ionic detergents with enzymatic and osmotic treatments. Reduction in detergent concentration or exposure duration, complete removal of processing steps, or other modifications have the potential to produce a scaffold with superior cell response. The Brendel and Meeza n protocols on which these kidney experiments were based used Triton X 100 concentrations of 0 .5% 3% v/v with suggested sodium deoxycholate concentr ation of 4% w/v. 141 142 The protocols in t his research primarily adhered to these limits, with the except ion of a few trials at up to 10% Triton X 100 concentration. Also, SDS was set to 4% w/v to compare to NaDC at equivalent ionic detergent concentration. It was demonstrated that increased concentration of perfused nonionic detergent (Triton X 100) improved removal of cell debris. Be that as it may, it was further shown that use of SDS improved cell removal more dramatically. Insofar, the detergent exposure for the most effective kidney decellularization protocol Kidney SDS is 71 hours for Triton X 100 and 2 1 hours for SDS. If the use of SDS is the primary determinant of success, then there may be opportunities to reduce both concentration and duration of both detergents in the decellularization protocol. De Kock et al. (2011) impressively improved upon a liv er decellularization protocol by reducing 72 hours of detergent flush to 1 hour at equivalent concentrations (1% Triton X 100 and 1% SDS). Aside from efficiency in production this change yielded greater preservation of laminin. 21 26 Shupe et al. (2010) was able to remove residual nuclear membranes in liver with only 0 .1% SDS following a graded Triton X 100 series. 25 27 In
84 lung, Cortiella et al. in a non perfusion protocol achieved decellularization with a 1% SDS solutio n, whereas 2% caused excessive ECM degradation. 20 Using perfusion, another protocol successfully decellularized lung with 0 .1% SDS at 2 hours duration, although with some residual nuclear material in associated cartilaginous tissue; 0 .5% caused ECM degradation 22 It may be that lower concentration is necessary in a perfused protocol in addition to shortened duration. Whole heart was decellularized using only 1% SDS for 12 hours with a short Triton X 100 rinse. The preparation showed no appreciable reduction in associated GAGs 23 Nakayama et al. without use of perfusion, decell ularized a kidney in 10 days using 1% SDS. 114 With the use of perfusion, and given the success of reduced concentration and duration in other organs, opportunities exist to refine this kidney protocol and minimize time and solution exposure. One paper suggests that a method to compare detergents in a decellularization protocol is to compare equivalent levels of detergency through the critical micelle concentration (CMC). 109 The CMC i s the minimum concentration required for micelle formation. In practice, this value is affected by factors such as salt concentration, temperature, pH, and presence of other surfactants. Also, solubility in ionic detergent is influenced by presence of counterions. 110 Therefore, while a direct comparison between nonionic and ionic detergents based on CMC alone may not be entirely precise, it still serves as a basic reference for compar ison. Further, CMC or multiplier may be used as a baseline to test for the lowest effective concentration of detergent. The CMC for SDS is 0 .24% w/v and for NaDC is between 0.083 and 0.249% w/v in aqueous solutions. Given the CMC values, the comparison of equivalent concentrations of NaDC and SDS for purposes of decellularization was a feasible approximation. Of note, effective concentrations of SDS for decellularization in other works cited earlier were below 0 .24% w/v Sodium dodecyl
85 sulfate exhibits a re cruiting and cooperative behavior at interfaces, likely allowing it to remain effective at concentrations lower than the CMC value. The decellularization protocols developed in this research typically required between 7 and 11 solutions steps. Many of thes e steps have been retained from the foundation kidney decellularization research by Brendel and Meezan (1979, 1980). 141 142 They designed their original protocol s to lyse and remove most of the cell ular material without the use of ionic detergent. It included steps to account for detergent gradation, enzymatic DNA removal, additional rinsing, and additives fo r increasing cell permeability, exerting osmotic pressure, and activation of endogenous nucleases. The theory was that with the long, convoluted tubular pathways in the kidney, the sudden release of cellular lysate and resulting protein aggregation should be avoided; a carefully stepped protocol would mitigate this phenomenon. However, the use of ionic detergent with its improved solubility may obviate these additional steps, as suggested by the preliminary success of the 1 Day SDS protocol. Successful redu ction of decellularization steps will have the advantage of ease of processing, lower cost, and reduction of inconsistencies. In liver decellularization research, two protocols used a gradation of detergents for whole organ decellularization. Shupe et al. (2010) used a series of 1%, 2%, 3% Triton X 100 followed by 1% SDS and Uygan et al. (2010) used .01%, 0 .1%, 1% SDS with a 1% Triton X 100 rinse. 25 26 This was later improved by De Kock et al. (2011) to 1% SDS followed by 1% Triton X 100. 21 This suggests a gradation of concen trations may not be necessary with ionic detergent use. In addition, all these protocols accomplished DNA reduction without the use of DNase. In fact, it has been shown that ionic detergent is effective at DNA reduction, even at a concentration of 0 .1% SDS as confirmed by DAPI and dsDNA quantification. 25 149 This is with the mentioned caveat of potential growth
86 fa ctor destruction with ionic detergent use. 149 One lung protocol required use of nucleases following an extended 5 week SDS treatment, but this may have been attri buted to use of non perfused organ methodology. 20 If DNA reduction is necessary for a particular decellularization protocol, and if the scaffold is being subseq uently conditioned for cell seeding or implantation, DNA may be effectively reduced by circulation of serum with its associated nucleases. 24 103 However, this may run the risk of also introducing proteases that are harmful to the matrix. 190 The complete removal of DN ase from the kidney decellularization protocols in this research would be a considerable cost saving as well as nullify the risk of lingering enzyme in the produced scaffolds. The use of deionized water to both exert hypotonic stress on cells and rinse an ECM scaffold is commonplace, but the use of salts to create hypertonic stress is rare in contemporary protocols. The addition of NaCl to detergent solutions in the kidney decellularization protocol by Brendel et al. was intended to aid in disruption of nuc lear membranes. 141 However, given the effectiveness of the ionic detergent SDS to d isrupt cell membranes, this additive and associated steps may not be necessary in the Kidney SDS protocol. Altogether, given the use of the ionic detergent SDS, 8 steps in the Kidney SDS protocol are likely not necessary. Recently, whole lung and heart we re successfully d ecellularized using only one ionic and one nonionic detergent p erfusion step followed by a pair of rinses, without addition of enzymes or salts. 22 23 Further, in other research, kidneys were decellularized in a 10 day non perfused organ process with only SDS and a rinse. 114 This supports the preliminary results of the 1 Day SDS e xperiment conducted herein, as perfusion based decellularization protocols have been considerably less in duration compared to non perfused counterparts.
87 Other protocol modifications may improve the kidney decellularization protocols. It has been shown in a nerve decellularization model that use of deionized water caused 30% swelling in the nerve, which was relieved by substitution with phosphate buffered saline (PBS). 109 All of the solutions and rinses in the kidney protocols in this research are based on deionized water, but it may be beneficial to substitute PBS. In other research, l iver, lung, and nerve hav e been successfully decellularized using only isotonic solutions. 24 25 109 147 It has been theorized that endogenous cellular proteases are a culprit in the variable results of decellularization protocols reported in the literature. 105 Once cells are lysed, these proteases may immediately degrad e the ECM. Therefore, some protocols have added protease inhibitors to counter this effect. Typically, ethylenediaminetetraacetic acid (EDTA) is used to inhibit matrix metalloproteinase activity through chelation of metallic divalent cations. 24 27 99 103 105 149 Caution must be taken though, as EDTA can also sequester divalent cations interfering with cell cell cadherin bonding and cell matrix integri n mediated attachment. Premature release of intact cells would likely clog the fluid pathways in the kidney. Decellularization protocols have achieved success by adding EDTA to the detergent solutions. 24 27 103 105 149 Bhrany et al. (2006), on the other hand, only used protease inhibitors during an initial hypertonic cell lysis step; although since they were applied to esophagus, there was no risk of clogging through rapid cell material release. In addit ion to 5mM EDTA, that group added other protease inhibitors including 0.2 mM phenylmethylsulfonyl flu ori de (PMSF), 10 M leupeptin, and 1.5 M pepstatin. 99 Relatedl y, if DNase or other enzyme is being utilized for decellularization, consideration should be taken as the buffer ions may also activate lysed proteases. Nakayama et al. (2010) performed an informative experiment, testing equivalent decellularization condit ions at two different temperatures. They showed that in kidney
88 decellularization, use of either Triton X 100 or SDS at 4C resulted in better preserved ECM morphology than treatment at 37C. Further, the scaffolds created using SDS at 37C exhibited compac tion, signifying loss of ECM integrity. 114 All perfusions used in the research described herein were performed at room temperature, approximately 25C. The scaffolds produced using SDS did not initially appear compacted. However, a fter seeded cell incubation (Chapter 4) the scaffold became noticeably compact; likely due to the stress of cell invasion. Lowering perfusion temperature may provide a solution to this behavior. Other decellularization protocols have used solutions at 37 C, in some cases to enhance enzyme activity, yet collagen IV has been documented to denature at or slightly below 37C. 20 21 27 191 Lowering the temperature of decellularization from 37C to 4C significantly improved preservation of collagen triple he l ical structure as tested by circular dichroism spectroscopy. 192 The final steps in any decellularization protocol is the rinsing of cytotoxic chemicals and ste rilization so that the scaffold may be directly implanted or used for in vitro cell seeding and culture. The process of rinsing and sterilization often overlap. In this scaffold research, the decellularizing kidneys were exposed to ambient air. However, 0 5% sodium azide was added to the detergent solutions as an antimicrobial and the perfusate passed through a 0 .2 micron filter. Further, since the kidney experienced consistent flow, the likelihood of microbe contamination was reduced Once finished with de cellularization, the scaffold was moved into a sterile container and treated under sterile conditions. If earmarked for cell seeding experiments, the scaffold was perfused or pre treated with serum free medium. Initial culture medium contained the antibiot ics penicillin and streptomycin and a treatment dose of the antimycotic amphotericin B; the latter was reduced to a maintenance dose upon media refresh. Overall, this procedure eliminated contamination for tissue cultures and allowed cells to proliferate, but it would likely
89 need to be revised for any implantation studies. In recent literature, the rinsing of organ scaffolds that have been subsequently implanted or cultured has spanned from 1 hour to 5 days. 23 26 Typically, this rinse is PBS 20 22 27 114 Residual decellularization chemicals lingering in the scaffold would be cytotoxic to cells. In this research, the length of the final wash was inherited from previous proto cols which were not intended for cell implantation. However, the survival of implanted cells in a prepared kidney scaffold showed the rinse to be sufficient. A straightforward way to optimize this time was demonstrated by Lichtenberg et al. (2006), in whic h time samples of the perfusing rinse were collected and tested in a cellular metabolic assay, MTS. 186 This type of assay exposes cultured cells to the samples and quantifies resulting cell viability and prol iferation. Results should indicate the least amount of rinsing time required for complete reduction of toxicity. On a related note, Reing et al. (2010) cautions chemicals lingering in the scaffold may interfere with characterization tests. Specifically, un rinsed SDS likely disrupts accurate GAG quantification using the dimethylene blue assay. 149 To improve SDS removal, protocols have followed SDS treatment with non ionic Triton X 100. 23 Sterilization techniques used for acellular tissue scaffolds vary considerably, and until recently there has been no direct comparison reported. 3 28 108 Common sterilization techniques for medical devices tend to denature proteins and generate cytotoxic residue. 28 Notwithstanding, gamma radiation (1.0 1.5 Mrad dose) has been used to sterilize kidney and liver scaffolds before implantation. 3 18 Typically, antibiotics or antimycotics are used for disinfection in the final organ rinse, in storage solutions, or in a media perfusion precedin g cell implantation. 20 22 26 114 Some protocols have ad ded an antimicrobial such as sodium azide or ammonium hydroxide to the perfusing solutions. 18 27 141 142 Other protocols have incorporated a 0 .1% peracetic acid or 4% ethanol flush. 26 27 105 Scaffolds treated with peracetic acid have shown preserved mechanical
90 integrity and good retention of ECM protein s and bound growth factors. 28 In a study of kidney decellularization, Sullivan et al. (2012) compared contemporary sterilization methods incl uding gamma radiation and per fusion of 70% ethanol, 0.2% per acetic acid, or SporGon reagent (7.35% hydrogen peroxide, 0.53% peracetic acid). They found that gamma radiation at 10 Mrad dose was the only method that assured prolonged sterility of porcine sc affolds (out to 6 months) and allowed cell growth. 3 Any reduction in cell proliferation as a result of scaffold irradiation, however, was not assessed. And it remains to be investigated how adversely this st erilization treatment affects somatic cell maintenance or ability to support stem cell differentiation. Conclusion A procedure has been developed that used perfusion based organ decellularization to produce a whole kidney extracellular matrix tissue engine ering scaffold. Based on a protocol from Brendel and Meezan developed in 1980, the perfused solutions have been substituted and modified in duration, concentration, and sequence to achieve an acellular kidney with preservation of renal specific ECM structu res, retention of basic extracellular molecules, and thorough removal of cellular debris. This work provides a baseline for further studies in renal scaffold characterization and modification based on in vitro cell seed or scaffold implantation studies. Th e primary challenge is balancing removal of tissue elements that inhibit cell response and retention of elements that promote viability, differentiation, and functional organization. It is likely that the least aggressive protocol to achieve DNA and cell p rotein removal will yield scaffolds with the greatest potential to direct cells A straightforward approach is to reduce solutions steps and lower concentration and duration of perfused detergents. Measuring the quantity and distribution of retained GAGs a nd associated growth factors during this optimization may serve as an intermediate gauge of success. Specifically, promising next steps may include:
91 1. Incorporation of analysis for residual DNA localization and quantification 2. Measurement of glycosaminoglycan retention 3. Identification of growth factors such as VEGF, PDGF, BMP 7, and Wnt 4 in scaffold 4. Modification of the decellularization protocol to include only SDS at 0 .1% to 1%, with Triton X 100 or PBS rinse at 4C Overall, the nature and parameters of the d ecellularization protocols will evolve as more determinants of cell behavior are recognized and measured and it is ascertained how they are affected by the various techniques and solutions involved in organ decellularization. Further, as these protocols ar e ultimately intended to produce a medical product, they will likely be influenced by donor species, sterilization technique, recipient species, patient immune status, specific nature of clinical treatment, intended post process or pre implant augmentation federal regulations, industrial process optimization, and market pressures. In the short term, however, and to provide a basic assessment of in vitro scaffold functionality, rat scaffolds produced in this research are seeded with mouse embryonic stem cel ls and cultured. Details of cell response are discussed in the next chapter.
92 CHAPTER 4 SCAFFOLD CULTURE WITH PLURIPOTENT STEM CELLS Experimental Plan Introduction Whole organ extracellular matrices have been isolated from kidneys through perfusion of det ergent based solutions. It is hypothesized that these matrices will best support the cells and complex tissue structures in the kidney as synthetic cellular scaffolds cannot yet mimic the histoarchitecture of natural intact renal extracellular matrix. The prepared matrices have been characterized by removal of cellular debris, preservation of renal structures, and retention of basement membrane proteins as these parameters have been shown in other models to influence seeded cell growth, differentiation, and organ specific arrangement. Ultimately, though, utility of a tissue scaffold must be determined through direct examination with a relevant cell population. Kidney ECM has distinct structure and protein composition with in the vasculature, glomeruli, and va rious segments in the tubules and collecting ducts that support a multitude of epithelial cell populations. 10 51 Further, it has been postulated that this unique heterogeneous composition plays a role in specificity of the numerous cells in the kidney. 10 Maeshima et al (2006) have demonstrated that adult tubule stem cells cultured with ECM components laminin, collagen I, collagen IV, or fibronectin showed distinct morphologies, attachment tendencies, and grouping patterns. 36 Extracellular matrix produced from a particular organ improves maintenance of resident cells in culture, amplifies growth rate, reduces apoptosis, and guides tissue specific differentiation. 20 24 30 159 164 Decellula rized organ ECM has been used on several accounts as a cellular scaffold with intention to reform organs. Notable efforts include liver, lung, and heart. The resulting cell matrix constructs each gained a degree of functionality such as contraction in the heart, gas
93 exchange in the lung, and albumin secretion, urea synthesis, and cytochrome P450 expression ( a key metabolic enzyme) in the liver. 18 22 24 26 28 However, in these models, all but the liver used fetal or neonatal organs as a cell source. The liver construct was able to use primary hepatocytes, which make up a majority of the mass of a normal liver and carry inherent regenerative potential. The use of same species fetal and neonatal cells offers proof that developing cells of an organ respond to adult ECM bu t these sources are not ethically viable as a long term solution. Pluripotent stem cells m ay serve as an alternate source as they may be derived from autologous adult tissue. The first step is to test the ability of seeded stem cells to respond to the dece llularized kidney scaffold; this includes attachment, proliferation differentiation and alignment of the cells with the preserved structural boundaries of the ECM Pluripotent stem cells have the ability to form any cell in the kidney and are highly proli ferative. 31 33 Embryonic stem cells, in particular, are readily available from established cell lines, easily maintained, and have an extensive history of differen tiation research. Further, differentiation tendencies of ES cells are likely translatable to patient derived IPSC s for future applications in the clinic. 122 In regard to kidney differentiation, implanted ES cells, through embryoid body intermediaries, spontaneously formed teratomas with rudimentary ductal structures; the cells in these structures expressed markers observed during kidney development. 119 Whole organ kidney extracellular matrix may guide these cells into more advanced renal structures. It has been shown that differentiation of ES cells is influenced by E CM composition. 160 Moreover, ECM derived from an organ will promote organ specific lineage differentiation 18 20 30 159 In renal organogenesis, critical steps are modulated by renal extracellular matrix proteins. 10 51 57 162 These proteins may also promote differentiation events in renal structures derived from
94 ES cells. After decellularization treatments, the kidney extracellular matrix scaffold showed preserved collag en IV and laminin. Laminin plays a role in branching morphogenesis, polarization of epithelia, and other developmental events and collagen IV is important to laminin mediated attachment and migration of ES cells. 51 57 150 Insofar, there is no recorded case of ES cell differentiation on ECM specifically derived fro m the kidney. In potential future clinical application, whole organ decellularized scaffolds would be produced from those human kidneys unsuitable for transplant, but with intact extracellular matrix. However, successful use of xeno graft ECM, such as from pig, would drastically reduce the high demand for human donor kidneys. Certain cell free preparations of non kidney xenografts have already been successfully implanted generating a constructive modeling response without eliciting cell mediated implant reje ction. 11 16 40 98 102 Alternatively, xenogenic whole organ liver ECM was used in vitro to maintain functional hepatocytes over 45 days. 193 Biocompatibility and cellular response to xeno derived scaffolds m ay be due to the high level of extracellular matrix protein conservation across species. 11 Examination and classification of the response of murine ES cells cultured in rat decellularized kidneys will provide valuable insight into the bas eline potential of xenogenic whole organ extracellular matrix to guide proliferation and differentiation of pluripotent stem cells. I n the remainder of this chapter details of these experiments a presentation and interpretation of results and promising avenues for continued research are provided. Specific highlights include static and perfusion culture methods, a survey of cell morphology, proliferation, and apoptosis patterns in relation to basement membrane structures, and detection of differentiation markers in comparison with embryoid bodies cultured without exogenous ECM contact Detailed methods and materials are described in Chapter 6
95 Cell Delivery and Scaffold Culture In a published decellularized liver model, the whole organ ECM scaffolds were e ffectively populated with hepatocytes via portal vein injection. 26 The hepatocytes initially surrounded the vasculature, but after two days of culture were shown di stributed throughout the matrix. Decellularized kidneys from this research have also retained their overall organ shape with intact vascular networks, and so in an effort to repopulate the kidney scaffold and gauge response of mouse pluripotent stem cells to the rat kidney ECM, ES cells were injected into the renal artery and cultured. The murine ES cells (B5/EGFP) injected into the kidney scaffolds were gifted from Andras Nagy. These cells were designed to express green fluorescent protein (GFP) regardless of differentiated phenotype. 194 This allowed continuous observation during culture relative to the translucent kidney scaffold. The mES cells were mainta ined in an undifferentiated state using mouse leukemia inhibitory factor (LIF). Before injection, these cells were suspended in a basic differentiation media formulation including 20% fetal bovine serum (FBS), an antibiotic, and an antimycotic. In other re search, embryoid bodies differentiated in media with FBS have spontaneously expressed genes essential for kidney development. 119 195 In attempt to isolate cell response to ECM exposure, no other agents previously proven to promote renal specific differentiation were added. Once decellularization treatments were complete, the scaffolds intended for injection w ere briefly perfused with serum free media and transferred to a biological hood. Then, a suspension of 2x10 6 mES cells in differentiation media was carefully injected into each renal artery. After delivery, the arterial cannulae were detached and the seede d organs were submerged in conical 50ml tubes filled with differentiation media. The organs were cultured up to 14 days, with media refresh every 2 3 days. That duration was chosen to capture significant milestones if differentiation progressed in time alo ng normal developmental pathways. Murine
96 embryonic stem cells are typically extracted from the blastocyst at E3.5 (days post coitum). 116 117 Murine renal development starts at E7.5 with metanephric development initiating at E11, and fully differentiated structures appearing at E17. 48 196 197 Finally, it was assumed that there would be insufficient cells at 14 days of culture to cause issue with nutrient or metabol ic waste diffusion, something that would be a concern with a full thickness organ. 198 The selection of which prepared scaffold and associated decellularization protocol was used for these cell seeding experiments was dependent upon the stage of concurrent kidney decellularization trials. Finally, a s these experiments progressed, alternate methods of scaffold culture and cell delivery were performed as described in the forthcoming results. Cultured Scaffold Analysis The techniques for analysis of the cultured scaffolds were designed to assess cell di stribution and reveal any morphological changes associated with extracellular matrix contact. Throughout culture, gross localization and grouping patterns were actively monitored with fluorescence microscopy. After completion of the culture period, the cel l matrix constructs were fixed, paraffin embedded and ex amined using H&E histochemistry Compared to frozen tissue embedding, paraffin embedding is superior for preservation of extracellular structures. Typically, a pair of kidneys were produced in each d ecellularization run, with one seeded and another serving as an acellular control. Upon observing the histology slides, localization of cells in relation to the observable structures such as glomer tubules, and interstiti al space s were noted. In addition any morphological cell patterns in relation to those structures was also described. After initial assessments, subsequent seeded scaffold experiments were conducted, and basic patterns of cell division, apoptosis, and diff erentiation were examined to provide a baseline analysis of mES cell interaction with whole kidney extracellular matrix scaffolds.
97 Results Whole Scaffold Culture Cells that were injected into the renal artery of whole acellular scaffolds achieved significa nt dispersal with radial streaks, scattered clumps, and appreciable accumulation in the glomeruli (Figure 4 1). Specific localization outside of the glomeruli was difficult to interpret given the uniform appearance of the tubular cavities. At ten days of c ulture, fluorescent images showed the outline of apparent intact tubular networks. Histology at this time point revealed cells in the glomeruli, within arteries, and interstitial spaces, especially prevalent near the surface of the scaffold. Arteries were identified by the presence of a thicker basement membrane. The decellularized kidney displayed in the figure at culture day 10 received an anomalously poor decellularization for the Graduated 10% protocol. However, as the eosinophilic cell debris is most l ikely aggregated in the tubules, it aided in highlighting the cells as localized to the interstitium. This assertion supports the appearance of the fluorescence images, which shows thinner GFP positive outlines of larger non cell occupied spaces, the tubul es or collecting ducts. Further, if the vascular basement membranes remained intact, a portion of the injected cells were likely forced into the interstitial capillary system. Also noticeable at the day 10 time point is the apparent presence of apoptotic c ells, primarily in the glomeruli. They are recognized by nucleus fragmentation, small scattered apoptotic bodies, and loss of hematoxylin stain. One last analysis performed on this sample was use of anti GFP IHC to prove that only implanted cells remained in the scaffold (data not shown). All recognizable intact cells were positive for GFP. One subsequent experimental alteration, after consideration of prominent apoptotic cell presence, was to integrate a perfused media circuit for improved cell viability.
98 A B F E D C g v i F igure 4 1. Whole scaffol d static culture: kidney scaffolds produced with the Graduated 10% decellularization protocol and cultured with arterially delivered B5/EGFP ES murine embryonic stem cells Cells at 1 day of culture show A) initial cortical distribution and B) filling of g lomeruli (H& E stains). C D) Strong GFP signal from scaffold at 10 days culture suggesting profile of tubular network. E F) Cells at 10 days of culture shows cells in (v)essels, (g)lomeruli and (t)ubular interstitium (H&E stains ) Magnifications: A, D, and E: 10x; B and F: 40x; C: 4x. Scale bars: A and E: 200m ; B and F: 50 m
99 Perfused Scaffold Culture It is commonly accepted that optimal conditions for generation of an in vitro organ or tissue are those that best approximate the natural, healthy in vivo environment. 198 In recent literature, there are several notable examples of moving beyond simple static culture to g row organ structures, and the primary mechanism is perfusion of media through the organ. 18 22 24 26 This has the advant ages of improving gas diffusion and nutrient delivery and removal of cellular waste 198 In addition, the perfusion mechanism can be designed to deliver physiologic flow conditions including a pulsatile pressure waveform that mimics normal c ardiac output. 23 The simulation of organ based fluid flow has resulted in improved endothelial and epithelial cell adhesion, viability, alignment and morphological appearance, and improved proliferation and differentiation of organ specific functional cell t ypes 22 24 199 The decellularized kidneys in this research retained intact vascular networks, and were therefore w ell suited to enact media perfusion through the renal artery. If diffusional barriers existed in whole kidney static culture, improved nutrient and oxygen delivery and waste removal had the potential to reduce apoptosis of the seeded embryonic stem cells. 23 198 Perfused media also has been shown to significantly increase the volume of viable cell structures in seeded scaffolds. 23 Another primary reason for implementing a perfusion culture system is better approximation of the in vivo renal environment, as mechanotransduction plays a significant role in kidney function. For example, fluid based shear stress plays a part in ion movement regulation, stability of fluid reabsorption, secretion of calcium and potassium, and release of nitric oxide for vasodilation. 200 Fluid flow al ters epithelial cell gene expression, growth factor synthesis in endothelial cells, prompts realignment of cytoskeletal elements, and modulates tight junction formation. 201 Further, it promotes prolife ration and differentiation of stem cells, including ES cells. 201 202 One specific example is ES cells t hat express the vascular endothelial
100 growth factor receptor, Flk 1, are induced to endothelial differentiation with applied shear stress. 202 Perfusion bioreact or design. A system was designed to maintain perfusing media through a cell seeded acellular kidney in an isolated and sterile environment. A diagram and images of the apparatus are given in Appendix A (Figures A 3 and A 4). Media was driven by a peristalt ic pump. A pressure limiting circuit allowed the pump to be set at a range of speeds without increase in pressure. This was achieved by allowing a majority of the perfusing media to bypass the kidney and travel up a vertical column. The adjustable height o f this column set median pressure in the kidney, analogous to the elevated reservoir used in the decellularization apparatus. In addition, increases in resistance caused by cell proliferation would not cause increases in pressure but a compensatory reducti on in flow. Close proximity of the pump to the kidney preserved transmission of the pump generated pulsatile waveform in the media, approximating natural circulation. A mean pressure of 100 mmHg with a 20 30 mmHg pulse range and a periodicity of 270 to 300 beats/min could be achieved, an approximation of normal adult rat circulatory pressure and pulse. Organ and media reservoir chambers were contained in an incubator to regulate temperature. Flow of biomedical grade 5% CO 2 gas through the kidney drip chambe r and reservoir chamber was designed to improve regulation of dissolved gases in media. The tubing, connectors, and bottles were all reusable and sterilizable, and most of the plumbing could be assembled in a laminar flow hood. Only a couple of connections were required outside the hood, minimizing the chance for contamination. Perfused culture results. Decellularized rat kidneys were injected with 2x10 6 mES cells and sealed within the organ drip chamber of the perfusion bioreactor. After 12 hours to allow for cell adhesion, circulation was initiated. Pressure was only increased to 40/20 mmHg as fetal
101 aortic blood pressure is significantly reduced compared to normal adult blood pressure. 203 At 3 days of culture, cells were primarily localized to the glomeruli, with limited radial streaks in the medulla (Figure 4 2); total cell numbers were remarkably reduced as compared with a static seed at 1 day (Figure 4 1). The cells at perfusion day 3, however, did not appear inordinately apoptotic, therefore two phenomena may have contributed to the sparse cellularity. First, cells were dislodged from the scaffold once circulation was initiated and pushed through the venous sy stem or if the glomerular basement membrane was inconsistent, passed out the tubules and resulting urinary outflow. Second, the application of shear stress to the mES cells may have served to inhibit proliferation. At day 6 of circulation (Figure 4 2), flu orescence micrographs reveal three distinct patterns in the scaffold, one resembling a glomerular tuft with adjacent vasculature, one an interstitial pattern similar to that observed in whole organ culture (Figure 4 1), and one a pattern resembling filled convoluted tubules. Histology shows glomerular cells with adjacent eosinophilic debris and what appears to be clusters of filled tubules. It is likely that the tubule pattern noted in the fluorescence images represent the tubule clusters observed in the H& E images. Perfusion of medium may assist the cells in localizing into the tubule spaces. Overall, in comparison with whole organ static culture, perfusion of media through a seeded scaffold results in a sharp reduction of mES cells, but likely improved mov ement of cells into the tubules. Sectioned Scaffold Culture One of the goals of this research was to assess patterns of proliferation, apoptosis, and differentiation of the mES cells cultured on acellular kidney scaffolds. However, until development of the Kidney SDS decellu la rization protocol, quality of cell removal had a greater degree of organ to organ variability. In order to best assess the time course of cellular response, a third method of culture was established. Once a whole organ scaffold was see ded arterially, it
102 E C D A B F G Figure 4 2. Perfused organ culture: murine B5/EGFP ES cells injected arterially into intact kidney scaffold and cultured with perfused media system ( Kidney NaDC decellularization protocol) Media was continuously delivered through the renal artery. At 3 days, cells are A) dispersed th roughout glomeruli with B) some streaks in the medulla at overall reduced number compared to static culture (H&E stain s) Fluorescen ce observation at day 6 shows C) glomerular profile with adjacent vessels, D) two distinct patterns, one interstitial and one tubular, and E) an extensive tubular network. H &E stains at day 6 showing F) glomerular cells with increased eosinophilic debris a nd G) tubules adjacent to glomeruli filled with cells Magnifications: A, B, F, and G: 40x; C and D: 15x; E: 4x. Scale bars: A, B, F, and G: 50 m
1 03 was placed in static culture for a day before being divided in to 4 6 thick transverse sec tions. These sections were cultured collectively in a multi well dish, with individual sections removed at prescribed time points. The primary a dvantage to this culture method is exposure of mES cells to identical scaffolds for multiple time points and each section had enough depth to preserve 3D architecture. Acellular kidneys produced with both the Kidney NaDC and Kidney SDS protocols were cultu red using the sectioned scaffold method and there were many similarities in overall cell distribution and morphology. However, given that superior scaffolds were prepared using the Kidney SDS protocol, it is the culture of those that will be the focus of t he remaining results. Nevertheless, any differences between the two types will be briefly noted in context. Cell distribution and morphological patterns. One advantage of using sectioned scaffold culture is the ability to observe a transverse view of the k idney. This allows visualization of seeded GFP positive cells relative to the cortex and medulla as opposed to a simple surface view. As such, over the course of a 14 day culture period, there were more recognizable patterns in gross cell distribution (Fig ure 4 3). A day after injection into a Kidney SDS produced acellular kidney, fluorescence microscopy showed cells heavily localized to the glomeruli. At day 3 and 4 they expanded into adjacent vasculature and progressed into the medulla. After day 6, cells were apparent in the larger vessels. At later time points, day 10 and 14, the cortex appeared increasingly crowded with large clusters of GFP positive signal. In addition, the scaffolds appeared to become more compact. In contrast, seeded scaffolds produc ed with the Kidney NaDC protocol retained a predominantly interstitial cell occupied pattern as seen in previous whole and perfused organ culture experiments and these scaffolds resisted compaction.
104 Once cultures were completed, H&E histochemistry was us ed to verify cell localization and examine condition and morphology relative to adjacent ECM, including glomerular, tubular, B A C D E F G H I Figure 4 3. Sectioned scaffold culture: typical distribution patterns of GFP la beled murine ES cells delivered into the renal artery of decellularized kidney scaffolds and cultured for 14 days Seeded scaffolds were sectioned into 4 6 transverse segments after one day of whole organ culture. A H) Scaffold produced using Kidney SDS pr otocol. A B) Overlapping images of ES cells at 1 day culture showing initial glomerular localization, A) bright field micrograph and B) fluorescence micrograph. C) Outgrowth of cells into adjacent vasculature at 3 days, and D) joining clusters of glomeruli at day 4. E F) Movement of cells into larger vessels at day 6. G H) Increasing surface fluorescence and graft compaction at days 10 and 14. I) In comparison, cells cultured in scaffolds produced using Kidney NaDC protocol showed similar progression yet wi th appearance of more fluorescence in interstitial reticular patterning as shown at day 10. Additionally, there is no compaction in these samples. Magnifications: A, B, C, E, G, H, and I: 4x; D and F: 10x.
105 vascular, and interstitial basement membranes (Figure 4 4). With some variations in cell dispersal and culture time, a typical chr onology of cell interaction was observed. In scaffolds A B F E D C I H G Figure 4 4. Sectioned scaffold culture: morphological patterns of B5/EGFP ES cell growth in acellular kidney scaffolds H&E stains. A H) Scaffolds produced with the Kidney SDS decellularization protocol. A) At day 4, cells are primarily localized to the glomeruli. B) At day 6, cells pack into the larger vessels. C) Example of intact cells at day 8 in the glomeruli, interstitium, and larger vessels; note the eosinophilic cell remnants in one vessel interior. At day 10, D) most cells are degraded in glomeruli, yet others line E) l arge vessels and F) expand into tubules. The tubules show patches of deteriorating cells with fractioned nuclei. G) At day 14, cells may show a less crowded spacing in glomeruli and interstitium. H ) A severe example of cell overgrowth and scaffold compacti on that may be seen with scaffolds produced using the Kidney SDS protocol. I ) Cells cultured on a scaffold generated with the Kidney NaDC protocol at day 10 show more restrained growth in tubules with morpholog y suggesting luminal apoptosis. Magnifications : A, B, and H : 10x; C, D E, F, G, and I: 40x.
106 produced with the Kidney SDS decellularization protocol, injected mES cells were initially captured in the glomeruli. At 3 or 4 days of culture the cells expanded into adjacent vessels, including perit ubular interstitium, and were present in groups as deep as the papilla. Cell presence in tubules was rare at these time points. At days 6 and 8, cells packed into larger vessels, spread on the capsule, and occupied some tubules. Signs of apoptosis were app arent in all locations, but especially so in the glomeruli. At days 10 and 14, cells at various locations, interstitium, tubules, and vessels continued to expand while the scaffold experienced overall contraction. Cells spread out in larger vessels and ass umed a squamous morphology. The capsular cell layer thickened and protruded away from the scaffold. Evidence of apoptosis remained, markedly so on the interior of larger packed spaces, such as within vessels or tubules; most glomeruli at this point were fi lled with eosinophilic debris. However, in these and other areas of accumulated apoptotic cell remnants such as interstitium and tubules, some cells survived in less crowded configurations. In comparison, cells cultured on Kidney NaDC produced scaffolds sh owed more restrained cell expansion, greater confinement within extracellular matrix boundaries, and less penetration of the interior. Also, the scaffold compression at later time points was minimal. Retrograde seeding. Delivery through the renal artery of decellularized kidney scaffolds exposed the cells to glomerular, tubular, and vascular basement membranes as well as interstitium in the cortex and medulla. However, population of the tubules was an uncommon occurrence, and usually was not apparent until several days of culture. Retrograde seeding for the kidney involves injection of cells into the ureter. The ureter dilates into the renal pelvis, an open chamber on the interior of the kidney. 4 The pelvis divides into calyces. Calyces surround the apices of renal papillae pyramidal structures with numerous pores from which collecting ducts
107 empty. The intention of retrograde seeding was for cells to be driven into the calyces and find their way into collecting ducts and adjoining tubules, improving frequency and distributi on of cell delivery to these locations. Murine ES cells were injected into four acellular kidneys produced with the Kidney SDS protocol and cultured 1, 6 and 10 days using the sectioned scaffold method. Only 55% (SD=23%) of 2x10 6 cells were retained after injection, as opposed to 97% (SD=4%) for arterial injected scaffolds. After a day of culture, it was apparent that distribution of cells throughout the scaffold was patchy and inconsistent. However, cells were present in both medulla and cortex, and in so me instances reached the renal corpuscles (Figure 4 5). Interestingly, a majority of the cells seemed to be located in the interstitium, with some suspected to be in tubules. Cells also were noted on the calyceal boundaries. At days 6 and 10 of culture, ce lls filled tubules and interstitial spaces. In the calyces, cells spread and formed a layer with squamous morphology that thickened over time. No cells were present at any time within the renal corpuscles or arterioles. Overall, this seeding route did deli ver a greater relative portion of cells into the A B i t Figure 4 5. Sectioned scaffold culture: cultured scaffold after ureteral retrograde seeding of B5/EGFP cells Kidney SDS decellularization protocol, H&E stains. A) 1 day after seeding, injected ce lls primarily localized to inter tubule spaces (40x magnification). B) Cells after 6 days of culture fill both (i)nterstitial space and (t)ubules (20x magnification).
108 tubules, yet the overall quantity and dispersion of cells penetrating into the cortex and medulla was significantly reduced. Cell Division and Apoptosis. According to H&E histochemistry, over a 14 day cultur e period, mES cells seeded into decellularized kidneys appeared to undergo rapid cell division and apoptosis in patterns throughout the scaffold. KI 67 and cleaved caspase 3 (CC3) i mmunohistochemistry was used to verify these observations and discern speci fic relationships to glomerular, vascular, tubular, and interstitial extracellular matrix T he KI 67 protein is present in cells undergoing any stage of mitosis whereas CC3 is present in cells that have initiated apoptosis. 204 205 At day 3 of cell culture in a Kidney SDS scaffold there was a strong presence of KI 67 in glomeruli, interstitium, and arterioles, y et CC3 was also showing in these areas (Figure 4 6). At day 6 of culture, cell division was present in all areas as noted at day 3, as well is in tubules. Interestingly, KI 67 appeared strongest on cells adjacent to arteriole basement membrane, within expa nding tubule cavities, and spreading on the capsule. Cleaved caspase 3 was also present at all locations, yet reduced in the vessel interior and on the capsule. Glomeruli at this time were heavily apoptotic, and positivity persisted in these locations beyo nd destruction of cells. At later time points, such as day 14, the marker for cell division remained in scattered cells. These cells were not in aggregated groups as at earlier culture times. In addition these cells thrived even in proximity to large field s of apoptotic residue. Also notable at this culture time are cells which appear quiescent, with no indication of cell division or apoptosis. Differentiation. In order for stem cells to effectively repopulate kidney scaffolds, it is necessary for the scaff olds and culture conditions to promote differentiation of those cells into functional renal phenotypes. In these experiments, the cell culture medium has not been specifically designed to promote renal differentiation so the ability of the scaffolds to pro mote
109 renal differentiation may be targeted. The following analyses were performed on scaffolds produced from the Kidney SDS decellularization protocol. Initially, the cultured scaffolds were Ki 67 CC3 Day 3 Day 14 Day 6 A B C D E F Figure 4 6. Patterns of ES cell divi sion and apoptosis on section cultured scaffolds reflected with KI 67 and Cleaved Caspase 3 (CC3) IHC ( 40x magnification, bars denote 50m ) Scaffolds decellularized using Kidney SDS protocol. A, B ) At day 3, cell division is strong in glomeruli and inters titium (left), yet there is also significant apoptotic signal arising in those areas (right). C, D) By day 6, serial sections show cells adjacent to vessel walls remain strong in proliferation, as well as cells on the capsule and in scattered locations inc luding interstitium and filled tubule cavities (left). In contrast, there is pronounced signal for apoptotic residue in glomeruli, and some positivity in smaller vessels, interior of larger cell filled areas, and interstitium (right). E, F ) Over longer culture times, 14 days, scattered cells are still Ki 67 positive (left), and a diverse landscape of CC3 demonstrates a dynamic history of cell turnover (right). At these culture times, many intact cells do not show positivity for CC3 or Ki 67 indicating a degree of stability.
110 examined using Oct 4 immunohistochemistry (Figure 4 7). Oct 4 is a key factor in regulation of ES cell pluripotency I t has been used as a marker of undifferentiated stem cells with loss of Oct 4 signifying differentiation. 34 122 179 Oct 4 is present in undifferentiated B5/EGFP mES cells. At day 4 of culture in the scaffold, Oct 4 is indiscriminately retained in approx imately half of all cells. This is slightly diminished at day 6, with cells in the interstitium and vessel lining containing a greater portion of positive cells. At day 14, scattered groups throughout the scaffold are still Oct 4 positive. Cytokeratins ar e intermediate filament proteins that commonly mark epithelial differentiation. They are present in many adult kidney cells including those in the proximal tubule, distal convoluted tubule, straight portion of distal tubule, and collecting duct. 153 They are extensively present in the developing metanephros yet are absent from undifferentiated metanephric mesenchyme. 206 Figure 4 8 shows c ytokeratin positivi ty at day 10 of seeded scaffold culture. Cells were positive in groups throughout the scaffold, and readily identifiable in some assuming a squamous morphology, w ere positive. There may have been c ytokeratin positivity in cells located in the tubules, but interpretation of location was imprecise due to A B C Figure 4 7. Retention of transcription factor Oct 4, regulator of pluripotency, in B5/EGFP ES cells section cultured on acellular kidney scaffolds produced with the Kidney SDS protocol (40x magnification, bars denote 50m). A) Most cells are positive at day 4. B) At day 6, a significant number of cells retain positivity in the glomeruli, interstitium, and vessels C) At day 14 of culture, scattered groups of cells remain positive.
111 scaffold deformation. Cytokeratins were not detected in undifferentia ted embryoid bodies. Reference c ytokeratin de tection in an embryonic kidney is shown in the appendices, Figure B 1. Given overall drop in Oct 4 expression and presence of epithelial differentiation, immunohistochemistry was performed on cultured scaffolds for three key proteins involved in kidney org anogenesis, Pax 2, WT1, and Ksp cadherin. Pax 2 identifies a handful of cell populations in the developing metanephros, including inter mediate mesoderm, ureteric bud and cap mesenchyme. It is a key regulator of branching morphogenesis and MET, and is nece ssary for differentiation of early stem cell populations such as cap mesenchyme and Foxd1+ stromal cells. 38 58 65 68 WT1 is present in the metanephric mesenchyme and proximal renal vesicles during development. 60 69 70 WT1 promotes differentiation to renal vesicle cells by suppression of Pax 2. 70 Pa x 2 and WT 1 proteins are both transcription factors located in the cell nucleus. 151 152 Ksp cadherin is a tr ansmembrane cell adhesion protein found exclusively in the kidney. 119 154 155 It can found on the basolateral membranes of epithelial cells in the ureteric bud and related progeny. It is also present in non glomerular epithelial cells of the nephron after polarization, which typically occurs during onse t of active glomerular filtration. 155 Dete ction of these three proteins (Pax 2, WT1, and Ksp cadherin) on mES cells cultured on decellularized kidney A B C Figure 4 8 Pan cytok eratin IHC reveals epithelial differentiation of seeded B5/EGFP ES cells at 10 days of sectioned scaffold culture (40x magnification bars denote 50m ) Scaffolds were produced using the Kidney SDS decellularization protocol. Positive cells were identified in the A) glomeruli, B) interstitium, and C) lining vascular basement membranes
112 scaffolds may indicate differentiation to renal lineages. After IHC was performed, Pax 2 was found present in diverse locations at all tested time points, although no evidence of WT1 was discovered. Ksp cadherin was absent at any location at multiple time points except for one anomalous cluster of cell s, which shall be considered inconsequential as it was not repeatable. None of these proteins were found present in undifferentiated B5/EGFP cells. Pax 2 expression persisted from day 3 to day 14 of culture (Figure 4 9). At day 3, a majority of the intact cells were positive in all populated areas including glomeruli, arterioles, and interstitium. At day 6, most of the intact cells remained positive, including cells now lining calyces, large vessels, and on the capsules. At later time points, 10 or 14 days, P ax 2 positivity was not as consistent in intact cells as previou s time points; many cells were P ax 2 negative. However, immunopositive cells still remained at all mentioned locations with the addition of cells that were scattered within tubules and those have progressed into less co mpact groupings show P ax 2 positivity. One example is cells that have remained viable after most of the original glomerular seeded cells have become ap optotic Reference IHC images for Pax 2 and Ksp cadherin on normal adult and embryonic kidney tissue is available in Appendix B (Figure B 1). Embryoid Body Culture One method to initiate differentiation of ES cells and promote generation of simple tissue structures is to culture the cells as embryoid bodies (EBs), or suspended cell aggregates. 116 117 Comparison of EB culture with seeded scaffold culture allows analysis of ES cell differentiation without the influence of kidney extracellular matrix composition and structure. Therefore, B5/EGFP ES cell EBs were generated and cultured using the hanging drop method. 207 Media composition was equivalent to that used in the cultured scaffold experiments. Once cultures were completed, immunohistochemistry was performed on these EBs; a summary of differentiation
113 markers in comparison with seeded scaffold cult ure is given in Table 4 1. Figure B 3 in the appendices displays relevant histological images. Embryoid bodies show a dynamic tapestry of cell di vision and apoptosis. The small densely aggregated ES cells with high nucleus to cytoplasm ratio progressed i nto alternate cell morphologies and less compact arrangements, with A B C D E F G H I Figure 4 9. Immunohistochemical analysis of Pax 2 expression in ES cells section cultured on acellular kidney scaffolds produced using Kidney SDS protocol (40x magnification). Cells showed widespread positivity by day 3 at all locations including A) glomeruli and arterioles, B) interstitium, and C) inner medulla. At day 6, cells were immunopositive D) in the glomeruli and adjacent interstitium, E) i n the inner medulla, and F) lining large vessels and on the capsule. G) At day 10, despite previous apoptosis, pax 2 positive cells remained in the glomeruli; H) cells were also found lightly scattered throughout packed tubules and lining the capsule. I) A t 14 day of culture, pax 2 cells persisted in various locations, including interstitium, in
114 each embryoid body transforming into unique heterogeneous cell collections. There was also a continuous loss of Oct 4 expression. As stated earlier, at day 0, there was no evidence of c yto keratin s Pax 2, WT1, or Ksp cadherin positivity. At day 10 of culture, c ytokeratins were present on the periphery of the bodies and adjacent to interior spaces. Pax 2 immunopositivity was visible in a significant quantity of cells. Ksp cadherin was not de tected. WT1 IHC was not available at this time point. Discussion Tissue Engineering Scaffold Cultured with Embryonic Stem Cells In this research, kidneys have been decellularized to produce whole organ extracellular matrices. These matrices were designed to be used as tissue engineering scaffolds for the generation of implantable kidneys. Other groups have used whole organ ECM scaffolds from heart, liver, and lung to create functional constructs. 18 22 24 26 These types of scaffolds, if properly prepared, can promote cell attachment, prol iferation, differentiation, and organization into tissue specific structures. 20 24 One particular challeng e in kidney scaffold based tissue Table 4 1. Immunohistochemical detection of embryoid body differentiation compared to seeded acellular kidney scaffolds. Tissue pan CK WT1 Pax 2 Ksp cadherin EBs Day 0 EBs Day 10 + ND + Ki dney Scaffold Day 10 + + Embryonic stem cells seeded into scaffolds decellularized with the Kidney SDS protocol or cultured as embryoid bodies (EBs) are both positive for key proteins in renal organogenesis. Cytokeratins (pan CK) are extensively prese nt in epithelial cells of the developing metanephros, but absent from undifferentiated mesenchyme. WT1 is a transcription factor present in metanephric mesenchyme and proximal renal vesicle of developing nephron. Pax 2 is a transcription factor present in cell populations of intermediate mesoderm, ureteric bud, and cap mesenchyme. Ksp cadherin is present in ureteric bud epithelium and appears after non glomerular epithelial cells in the nephrogenic zone have polarized. Key: protein not detected; +, prot ein detected in multiple locations in proximity of implanted cells; ND, no data
115 engineering, given the complex tubular architecture of renal ECM, is repopulation of the scaffold with the requisite epithelial cell types to reinstate filtration and reabsorption functionality. One potential solution, giv en that ECM components can influence differentiation of pluripotent stem cells, is to deliver these cells into the scaffold and allow them to selectively differentiate into the required somatic cell types in response to the underlying heterogeneous ECM. 20 30 160 To test the response of murine ES cells to whole d ecellularized rat kidney, cells were delivered into the renal arteries and cultured up to 14 days. Initial distribution of cells and subsequent expansion patterns were observed with fluorescence microscopy. Hematoxylin and eosin histochemistry was then use d to inspect cell localization and morphological changes in relation to the underlying basement membrane. Immunohistochemistry was employed to detect patterns in cell division, apoptosis, and differentiation. The antibodies used in the differentiation stud ies targeted proteins expressed in epithelial cells and early stage nephrogenic progenitor cell populations. Also examined was a pluripotent stem cell marker, the loss of which signified differentiation. To compare response of these ES cells to those uninf luenced by ECM, embryoid bodies were cultured to equivalent time points and differences in cell proliferation, apoptosis, and differentiation were noted. Also during this study a perfusion culture method was examined. Finally, to improve delivery of cells to the tubules and collecting ducts, ureteral seeding was explored. Cell Distribution In order to effectively repopulate a tissue scaffold, cells must either be delivered to all areas of the scaffold or be able to proliferate and migrate to fill voids. On ce mES cells were injected into the renal artery of a rat kidney scaffold produced with the Kidney SDS decellularization protocol, a majority were captured in the glomerular capillary tufts. These seeded glomeruli were effectively distributed throughout th e cortex and outer medulla of the
116 kidney. In a model of glomerulonephritis repair, delivery of MSCs through the renal artery also localized to the glomeruli, achieving a 70% distribution. 16 83 Assuring that cells are delivered to the glomeruli is critical given the cell requirement for normal glomerular filtration. After delivery, the most probable scenario is that du ring the first 6 8 days in section culture, cells remained confined to an intact vascular system, progressing outward from the glomeruli into the efferent arteriole and peritubular capillary system (including vasa recta) or back into the afferent arteriole and larger arteries. The peritubular capillary system is located within the interstitium in both cortex and medulla; this could explain cell presence in both locations. At 6 8 days of e space, lining the calyces, and on the exterior kidney capsule. It may be, however, that appearance of cells in these alternate locations is due in part to the section culture method; cells could have migrated outside the boundary of the section and retur ned in an adjacent location. In whole kidney static culture, cell appearance outside of vascular locations was markedly reduced, with the caveat that those culture samples were produced with different decellularization protocols; those protocols may have h ad a different effect on ECM composition and condition, which effects the ability of the cells to migrate. Delivery of cells through the renal artery of the kidney scaffold was an effective means of dispersing cells. However, a robust vascular basement mem brane may be inhibiting the ES cells from migrating to other areas of the kidney, such as the tubules or collecting ducts. Based on smooth muscle cell infiltration of arterial matrix, it has been suggested that migration is dependent on density of the matr ix. 101 103 Other cells have been successful in passing through the vasculature. Uygun et al ( 2010 ) used the vascular supply to repopulate an SDS produced whole liver scaffold with primary hepatocytes. 26 In a model of tubulonecrosis repair, an injection of
117 adult renal ste m cells successfully passed through the vessels and localized to damaged tubules. 76 However, this movement may have been motivated by chemoattractants released by the injured cells. Another potential route for delivery of cells into the kidney scaffold is through the intact ureter. Injected cells have the potential to reach past collecting ducts and tubules all the way to the epithelial side of the glomeruli in the ureters of the kidney scaffolds, cells did reach the glomeruli but for the most part were unexpectedly localized to the interstitium (Figure 4 5). In addition, this delivery method was ineffective and distributing cells through the cortex and medulla; only sparse radial streaks or wedge shaped sections were observed. A majority of the cells were retained in the calyceal spaces or passed through the scaffold altogether. It may be that delivery press ure of the cell solution exceeded mechanical tolerances of the basement membranes causing ruptures and emptying or rerouting of cells, or that the anatomy of the rat kidneys is not entirely conducive to retrograde cell delivery. On a positive note, many ce lls did arrive in the tubules and over 10 days of culture filled those spaces. Despite its potential, this route of cell delivery was not pursed because of low reliability and inconsistency in cell placement. In addition, scaffold culture permitted cells d elivered into the renal artery to move into tubules over time, allowing assessment of those areas. In the future, however, revisiting and improving ureteral cell delivery would be beneficial, as a culture method that retains the whole kidney scaffold is ne cessary to achieve a functioning organ. One method of improving cell seeding in whole organs is to divide delivery of a single cell suspension into multiple injections and space them at an even time interval. 26 If ureteral delivery is refractory to improvements, another option is direct microinjection of cells into the body of the kidney. 80 130
118 Precise identification of cell location in the scaffolds was challenging and open to interpretation, especially in the Kidney SDS scaffolds, as all of the cell features typically used for labeling in normal kidney histology had been washed away. In this research, interstitial spaces were identified as the narrow basement membrane spaces surrounding the larger spaces. The larger spaces were likely proximal or distal tubules or collecting ducts and housed the bulk of the epithelial cell population. In early iterations of the kidney decellularization protocol, cellular debris was located within these larger spaces, supporting this assertion. The primary caveat to using ECM spacing to identi fy interstitium is that at later time points in Kidney SDS scaffold cultures, severe scaffold distortion made this method impractical. In the future, precise labeling of ECM could be achieved with IHC, such as with combinations of individual laminins or co llagen chains. 10 Cell Proliferation An effective tissue scaffold promotes viability of implanted cells and supports proliferation to assure ECM coverage. Other factors play a role in this such as growth media constituents and overall cell culture environment. To evaluate the Kidney SDS scaffold, an IHC study was performed to assess any tendencies toward seeded ES cell proliferation or apoptosis. It was immediately clear that ES cells maintained a high level of cell prolifer ation and apoptosis in all areas of the kidney. Significant variability made attempt to correlate these events with adjacent extracellular matrix challenging, yet a few trends did present themselves. After 3 6 days of culture, a bulk of the cells in the gl omeruli strongly expressed the protein marker for apoptosis, CC3. This may be due to crowding and a lack of nutrient/waste diffusion at that location or a lack of sufficient intact basement membrane proteins in the glomerular capillary to inhibit ES cells apoptosis. It may also be a result of overall limited cell survivability in the scaffold under the culture conditions; many of the glomeruli were initially seeded and would
119 likely be the first to become apoptotic. Cells lining the calyces, capsule, and ope n vascular basement membrane had a tendency to resist apoptosis and express the marker for cell division, KI 67. In contrast, cells on the interior of crowded vessels, or on the interior of large ES cell aggregates have shown increased CC3 protein expressi on (Figure 4 6). Embryonic stem cells are, by nature, very dynamic. Upon observation of standard EB development, ES cells that are adjacent to ECM proteins will proliferate while unattached cells become apoptotic. 208 This is a basic part of intended cavity formation and tissue structuring. 207 In other research, fetal kidney cells were allowed to migrate into decellularized kidney ECM, but both the c ell explant and scaffold cultured cells showed evidence of apoptosis after 5 days. 114 It may be that the duration of cell survivability in this research is comparable to other contemporary cell seeding protocols. During the first 4 6 days of cell culture on Kidney SDS scaffolds, seeded ES cells proliferated to the limits of the ECM boundaries and in most instances distorted the matrix. This matrix distortion is exacerbated by days 10 14, often resulting in a dramatic reduction in to tal graft volume. Cells cultured on scaffolds that retained cell debris, such as produced with the Kidney DOC and Graduated 10% protocols filled the ECM spaces without distortion. This may be due to residual cellular material stabilizing the matrix. Anothe r consideration is that SDS may be causing structural changes to the ECM, as the protocols that left residual cell debris utilized NaDC. Cortiella et al (2010) saw scaffold shrinkage after implanting ES cells into an acellular lung. 20 The lung was decellularized using SDS. Nakayama et al (2010) saw acellular kidney shrinkage after use of SDS, but before cell seeding. 114 Those scaffolds, however, were produced at an elevated temperature of 37C. Cell Differentiation Fundamental to the use of ES cells to repopulate an acellular kidney scaffold is differentiation based on ECM contact. Extracellular matrix has been shown to regulate gene
120 exp ression, cell phenotype, and promote differentiation of ES cells. 20 30 160 In addition, 3D matrices have been shown to improve differentiation of ES cells through increased cell cell and cell matrix association. 209 The purpose of the scaffold culture experiments conducted herein was to establish a baseline cell response without post decellularization scaffold modification or addition of culture media factors that promote renal specific differentiation. Therefore, as an initial histological assessment, mES cells injected into Kidney SDS ECM scaffolds were cultured and observed for morphological changes. Initially, the cells appeared den se and crowded with a high nuclear:cytoplasmic ratio, and many of these cells seemed to remain as such throughout the duration of the 14 day culture. After day 6, some cells adjacent to vascular or tubular basement membranes assumed a more cuboidal morphol ogy. Alternatively after that time point, cells that were uncrowded and attached to the ECM of larger open spaces such as arterioles, arteries, and calyces displayed a squamous morphology. Some cells in the glomeruli and in interstitium at days 10 and 14 o f culture assumed a more relaxed spacing; interestingly, this appeared to correlate to areas with high levels of apoptotic cell residue. One possibility is that if the apoptotic cells were confined to the vasculature, then these cells were located outside the capillaries such as on the epithelial side of the GBMs or exterior to the peritubular capillaries yet still contained in the interstitium. In an attempt to elucidate the observed morphological evidence of differentiation, IHC was conducted. Oct 4 is a marker of undifferentiated ES cells, and loss of expression has commonly signified differentiation. 122 Over the course of 14 days of culture, Oct 4 expression, although diminished over time, still had significant level of expression in non apoptotic cells (Figure 4 7). However, cytokeratin IHC clearly showed epithelialization of cells in groups throughout the scaffold, including in the glomeruli, on the capsule, in the interstitium, and lining
121 the large vessels (Figure 4 8). The next three antibodies detected proteins that are expressed in early nephrogenesis, Pax 2, WT1, and Ksp cadherin. Combinations of these three proteins are expressed in all early kidney pro genitor cell populations. They are also present in adult kidney cells, but the ES cells are likely to pass through a progenitor population before progressing to terminally differentiated adult kidney cells. According to gene expression, differentiation of ES cells in embryoid bodies typically follow traditional developmental pathways. 122 Pax 2 was present in a majority of the seeded mES cells throughout the dura tion of the culture (Figure 4 9). Pax 2 is present in late stage intermediate mesoderm, ureteric bud and epithelial progeny, induced metanephric mesenchyme, and the cap mesenchyme stem cell population. 59 60 121 It is c ommonly used to identify kidney speci fic progenitor cell populations 34 114 119 122 WT1 is present in metanephric mesenchyme and cap mesenchyme and Ksp cadherin is present in ureteric bud and related progeny 60 69 70 119 154 155 210 Therefore, the only remaining possibility for Pax 2 positive cells is late stage intermediate mesoderm. This does not guarantee renal specific lineage however, as Pax 2 has also been detected in devel opment of mid hindbrain, optic nerve, inner ear, and female genital organs. 67 211 Proving that the Pax 2 po sitive cells belong to the intermediate mesoderm cell population would require detected expression of other protei ns such as Osr1, Lhx1, and Emx1. 60 63 Pax 2 is also present in later stages of kidney development, such as in pretubular aggregates derived from cap mesenchyme or adult proximal tubules cells, but these cells would also be positive for WT 1 or Ksp cadh erin. 119 Cytokeratin IHC was not performed in conjunction with the Pax 2 IHC, so co localization could not be used to further isolate the Pax 2 cell population In addition, the cytokeratin IHC employed detected a broad spectrum of cytokeratins and therefore could not be used to mark specific kidney epithelial cells. It may be that the cytokeratin expressing cells and Pax 2 expressing cells are separate cell
122 pop ulations. Also, with absence of Ksp cadherin and WT1, cytokeratin positive epithelial cells are not specific to the kidney. Finally as Pax 2 and cytokeratin expression was not limited to specific basement membranes types, preferential differentiation could not be recognized. In the future, it may be beneficial to complement IHC with qRT PCR (quantitative reverse transcriptase polymer chain reaction ) analysis for detection of gene expression markers. Immunohistochemistry analysis is based on localized protei n concentration and visual interpretation of the observer, while qRT PCR quantifies total mRNA in a sample. 212 Given the heterogeneity of the kidney ECM, however, a method for targeting specific areas in the tissue sample such as laser microdissection may be required for location based com parison. 119 Use of qRT PCR may support IHC results or offer alternative data for cell identification. The primary caution in comparing the two results is that there are numerous opportunities for variability during processing of both techniques and total protein may not correlate with total mRNA 212 The markers selected for identification of early renal progenitors, Pax 2, WT1, and Ksp cadherin are not unique to any particular nephrogenic cell p opulation. Additional markers would serve to clarify differentiated phenotypes. One valuable resource in the selection of these markers is a collaborative database of genitourinary development, GUDMAP. It contains known gene expression profiles and matched IHC for every identified nephrogenic cell population. It also identifies anchor genes, which are unique to a particular cell population 196 197 Embryoid Body Comparison One method to promote ES cell differentiation and tissue formation is to allow the cells to form three dimensional aggregates called embryoid bodies. Embryoid bodies develop into heterogeneou s structures and cell differentiation paths may be effected by media, culture environment, size of EBs, and composition of adjacent ECM. 207 In this research, EBs were formed from the mES cells and cultured up to 14 days in the same media used for scaffold
123 culture. In order to examine the effects of kidney ECM, EBs were compared to the mES cells cultured in the scaffold. CC3 and KI 67 immunohistochemistry showed interior cells undergoing apoptosis with a majority of the remaining cells undergoin g cell division (Figure B 1). This behavior is typical of EB development and cavity formation; cells near the exterior produce ECM, notably collagen IV and laminin, which in turn promotes the survival of those cells. 207 208 Similar behavior was observed in scaffold culture whereby cells attached to vascular basement membrane had a tendency to express KI 6 7 while unattached cells in the lumen expressed the apoptotic marker CC3. At day 0 of culture, none of the EBs showed immunopositivity for the markers of differentiation pan CK, WT1, Pax 2,or Ksp cadherin (Table 4 1). After 10 days of culture, both the EBs and cells in the scaffold showed positivity for pan CK and Pax 2 and absence of Ksp cadherin. In the EBs, Pax 2 was expressed in most cells with CKs present on exterior borders and developing interior spaces. This distribution of proteins is similar to ce lls cultured in the scaffold. There is no data for WT 1 beyond day 0. In other EB differentiation research, mES cell lines EB3, D3, R1, and 129 sv ES all showed Pax 2 mRNA expression by day 6 of culture, Ksp cadherin by day 7, and WT1 by day7. 119 122 195 However, there was variation in the expression pro files. For example, qRT PCR analysis of the EB3 culture showed initial KSP expression at day 7 with sharp increases at day 12 and 18, whereas a densitometric RT PCR analysis of the D3 culture showed Ksp cadherin presence at day 1 with gradual increase to d ay 14. 122 195 Both of these results contrast with the absence of Ksp cadherin in this research. However, comparison between all these results may not be reasonable as the cell lines, media conditions, and detection methods were all different. On the other hand, there may be a case for additional analysis of both the cultured scaffolds and EBs with an mRNA det ection method such as qRT PCR.
124 Given the limited IHC comparison of B5/EGFP mES cells cultured in acellular kidney scaffolds or as EBs, there is insofar no evidence that the Kidney SDS scaffold promoted differentiation of the ES cells beyond inherent tende ncy. However, as described, the analysis is limited, and further examination may reveal evidence of improved differentiation. Traditional IHC is exceptional at pairing protein expression with tissue location, but it is not a quantitative analysis. QRT PCR or ELISA could be employed to quantify any differences in mRNA or protein expression, with tissue segments isolated by microdissection. One consideration is that the timing of gene expression in the seeded cells may lag behind EBs, as the seeded cells need time to expand into the matrix and respond to extracellular signals. A better comparison may be to examine the response of mES cells cultured in two different types of acellular matrices. Another consideration is that ES cells cultured in the Kidney SDS s caffolds showed strong tendency for proliferation, often crowding into spaces. This may affect overall trends in differentiation as a reduced number of cells are in contact with the ECM. Endothelial Differentiation Based on histological observation many of the injected mES cells were assumed to be retained in the kidney scaffold vasculature. Therefore an opportunity existed to test the cells for endothelial differentiation driven by contact with intact vascular basement membrane. Analysis of the cultured sc affold samples was performed at the lab of Dale Abrahamson at the University of Kansas Medical Center. Using IHC, they found that the cultured cells were positive for both Bandeiraea simplicifolia lectin B 4 (BsLB4) and VEGFR2 (Flk 1). 213 BsLB4 lectin is a highly selective carbohydrate binding protein that marks cells of endothelial lineage, including mesoderm derived angioblasts. 214 VEGFR2 is the receptor for VEGF growt h factor and signaling between the two is necessary for vascular development. 215 In other work, it was suggested that one source of endothelial cells in the kidney are Flk 1+ cells in the metanephric mesenchyme. 62
125 However, lack of WT 1 protein detection in this research implied the mES cells cultured in the kidney scaffold did not transition through a metanephric mesenchyme phenotype. An informative experiment would be to analyze the cultured scaffold for any VEGF retained in the decellu larized kidney. One group decellularized whole liver using Triton X 100 and SDS and found VEGF (which can bind to ECM) present in the scaffold. 21 It was shown tha t VEGF can promote human ES cells into functional endothelial cells. 216 It was also shown that the application of shear stress will induce differentiation of Flk 1+ ES cells into vascular endothelial cells in vitro 202 Perfusion Culture A system was designed to perfuse media through the cell seeded kidney scaffold in a pulsatile manner approximating rat cardiac output. Vascular perfusion of media improves oxygen and nutrient delivery to cells, removes products of cellular metabolism, and allows thicker tissue growth. 198 In addition, as fluid flow plays a role in regulation of renal cell epithelium, perfusion of media allows closer simulation of the in vivo cellular environment 200 201 Simulation of in vivo microe nvironment improves tissue specific cell growth, differentiation, and structure formation. 22 24 201 Other grou ps have used perfusion bioreactors to grow lung, heart, and liver tissue. 22 24 26 In this research, mES cells were seeded into acellular kidney scaffolds and given 12 hours to adhere before perfusion of media was initiated. As in static culture experiments, the initial injection contained 2x10 6 cells. However, by day 3, there was a significant reduction in cells as com pared to static scaffold cultures. The number of cells retained in all scaffolds after a seed injection was fairly consistent, so one possibility for the cell reduction is that once flow commenced non adherent cells were dislodged and passed out of the kid ney. At day 6 of culture,
126 associated glomerular capillaries. An explanation for this is that stress from media circulation caused a rupture in the GBM, allowing cells to pass into the tubules. This may also account for the increased tubule seeding associated with perfused culture. The other possibility for overall cell reduction is that media conditions or shear stress reduced proliferation of the ES cells. The cells as ob served at day 3 of culture were uncharacteristically isolated so reduction of proliferation may at least be a contributing factor. All scaffolds used in perfused culture experiments were produced from the Kidney NaDC protocol, and as a consequence, retaine d cellular debris. This debris would have affected media flow patterns and stress distribution, possibly influencing condition of the cells or contributing to ECM rupture. This is compounded by the fact that a majority of cells were delivered to the glomer ular capillaries, and since they were not flattened endothelial cells, served to block fluid flow. Conducting future experiments with a debris free Kidney SDS scaffold would be a worthwhile experiment. Further, once the technique for ureteral seeding is im proved, shifting the cell injection to a ureteral seed would allow unimpeded media flow through the glomeruli. This may enhance growth and epithelial differentiation of cells in tubules and collecting ducts. One reason the perfusion culture experiments wer e not pursued for the duration of this research study was because of the overall reduction in cells compared to static culture methods. Two other reasons were comparative cost of the experiments and lack of ability to compare one cell seed/scaffold combina tion over multiple time points, which was accomplished with section scaffold culture. Therefore, analyses of cell proliferation, apoptosis, and differentiation were not conducted. Despite the overall reduction in cells, there may have been positive shifts in cell response such as increased expression of endothelial or epithelial markers. Yamamoto et al (2005) showed that fluid shear stress promoted differentiation of Flk 1 positive ES cell s to
127 vascular endothelial cells in vitro 202 The Flk 1+ ES cells were induced by culture on collagen IV. Despite the reasons the perfusion culture experiments were temporarily sidelined, use of a bioreactor that mimics in vivo circulation, as shown in other studies, remains the most promising system to culture cells in a kidney scaffold. In addition, while section scaffold culture is effective for analysis of cell response in the short term, given the disruption of the whole int act matrix, it is not a method that will lead to an implantable kidney construct. In order to continue with perfusion culture, two design flaws that may lead to an adverse cell response need to be addressed. Pressure driven gas flow was designed to improve diffusion in the media, however, it was completely dehumidified. The media reservoir was significant (500ml) but the gas may have pulled enough moisture from the media to adversely change the osmolality of the culture medium. Also, the regulation of CO 2 and therefore culture pH, was challenging. It was discovered that the media tubing was highly permeable to gas, especially CO 2 Since a significant portion of the tubing passed outside of the incubator, there was opportunity for appreciable CO 2 loss. To co mpensate in the short term, CO 2 concentration in the incubator was increased to drive diffusion of the gas back into the media. Solutions to these problems are replacement of the tubing with low gas permeability and the addition of a cylinder washer for ga s humidification. 22 Other potential improvements to this initial design include use of a membrane oxygenator for improved gas diffusion, use of a heat exchanger inste ad of an incubator to control temperature, and replacement of the vertical pressure regulation circuit with a sealed, pressurized chamber. 22 183 Culture Media Modifications One goal of this research was to assess the ability of the decellularized kidney scaffolds to promote renal differentiation of mES cells using a basic growth media without the influence of
128 addition al growth factors or other pro differentiation media additives. In future work, however, modification of the base media or addition of these factors could be used to enhance the innate potential of the scaffold to control the fate of seeded cells or even p romote the cells to a fully differentiated phenotype. In the base media, each ingredient has potential to affect cell fate. For example, the antimycotic added to the differentiation media in this research, Fungizone (Amphotericin B), had potential to effe ct cellular gene expression and modify results. 217 One potential avenue for improved differentiation is substitution of the base media with other formulations that have been used to differentiate ES cells into renal lineage. Morizane et al (2009) used DMEM, 10% FBS, and 0.1 mM 2 mercaptoethanol while testing specific growth factors in their studies of differentiating both ES cells and IPSCs into renal lineages. 122 The Morizane et al study also used less FBS than used in this research. 122 Fetal bovine serum, a primary ingredient in most growth media re cipes, contains a high content of embryonic growth factors and other components that have been shown to support cell metabolism. It is commonly used in ES cell media to promote differentiation. The drawback to FBS use in media is that not all components ar e identified and batch to batch variability. 31 This variability can influence potential to differentiate cells along a certain lineage. To find improved base growth media formulations it may be worth trying different batches of FBS, or reducing FBS r equirement altogether. Other potential sources for base media recipes are those used for metanephric organ culture. These formulations are designed to maintain all cells in a developing kidney. Nakayama et al (2010) used DMEM/F12, 25 ng/mL prostaglandin E 1% insulin/transferrin/selenium, 10% FBS, and 1% penicillin/streptomycin to support growth of fetal kidney cells from whole explanted metanephros into adjacent decellularized kidney matrix. 114 Giuliani et al (2008) was able to e
129 15 base medium, 1% penicillin/streptomycin, and 2% FBS. Reduction of FBS to 2% in this case actually reduced necrosis over span of the culture period while supporting nor mal nephrogenic development. 218 In addition to base media changes, other additives such as growth factors may be used to promote renal differentiation of the mE S cells. These may be applied to ES cultures before injection to pre differentiate cells and allow use of marker specific cell sorting for population purification, after injection to allow differentiation enhancement by the ECM, or both. 219 Promising targets for factor mediated mES cell differentiation are one of the kidney progenitor populations. These include intermediate mesoderm, the precursor to all cells in the kidney; ureteric bud epithelium, the precursor to all collecting duct cells; or metanephric mesenchyme, the precursor to all cells in the nephron including vascular and stromal cells. 62 Morizane et al (2009) claims 4 days of 10 ng/ml Activin exposure induces mesoderm differentiation, as shown by Brachyury expression. 122 Two other protocols use 10 ng/ml Activin and 0.1 M retinoic acid for 5 days of culture to advance cells to an intermediate mesoderm state, although one of the protocols adds 50 ng/ml of BMP7 to reduce expression of the neuronal marker Synapsin 1. 121 123 Morizane et al also showed that once mesoderm was induced with Activin that metanephric mesenchyme could be achieved per increased WT1 and Pax 2 expression with exposure to 150 ng/ml GDNF or 15 ng/ml BMP7 for an additional 10 days. 122 These markers may also indicate a more differentiated population such as cap mesenchyme or pretubular aggregates. Altern atively, continued exposure of Activin out to 14 days enhances Pax 2 and Ksp cadherin expression, possibly indicative of ureteric bud, collecting duct epithelium, or tubular epithelium. 122 Other studies suggest hepatocyte growth factor (HGF) may promote tubular differentiation. 36 179 Wnt 4 plays a role in aggregation and polarization of epithelia, and may be another useful culture
130 additive. 179 In regard to the kidney scaffold culture medium, it would be useful to test any of the above growth factor pro grams. The addition of these factors could encourage kidney specific structure formation in an ECM scaffold. A simple first pass would be addition of 10 ng/ml Activin to the growth medium for the duration of the culture period to see if it induces Ksp cadh erin expression. Conclusion The first important step after creation of a tissue scaffold is to characterize the response of implanted cells. Once this is achieved, scaffold preparation and growth conditions can be modified to enhance functional tissue form ation. In this research mES cells were seeded into decellularized whole kidneys and cultured up to 14 days. Cell delivered through the renal artery localized to the glomeruli and adjacent vasculature, but in section culture were to migrate to the tubules, capsule, and calyces over time. These cells were widely dispersed in the organ. Cells delivered into the ureter showed increased tubule or collecting duct localization, but dispersal through the organ was poor. Cells lining the arteries, in the calyces and on the capsule tended to express the marker for cell division. Cells delivered into the glomeruli or unattached to ECM tended to show increased apoptosis. Pockets of cytokeratin positivity indicated epithelial differentiation. Also widespread were markers of endothelial lineage commitment, VEGFR2 and BsLB4. Most cells expressed Pax 2, but there was an absence of WT1 and Ksp cadherin, suggesting possible mesoderm differentiation. Limited IHC analysis of EB culture without kidney ECM contact also showed expr ession of Pax 2 without Ksp cadherin expression, suggesting the kidney ECM scaffold may not have had a strong influence on renal specific differentiation. Scaffold culture with perfused media resulted in reduction of cells, but an analysis of differentiati on was not performed. Once ureteral seeding has been achieved, re engaging the perfusion culture experiments with a debris free Kidney SDS scaffold and the
131 addition of renal inducing growth factors may be a promising step toward generation of a functional kidney.
132 CHAPTER 5 CONCLUSION AND FUTURE STUDIES Research Summary Extracellular matrix tissue scaffolds were created by the decellularization of rat kidneys. The decellularization process involved arterial perfusion of detergent based solutions. The prot ocol utilizing SDS successfully removed cellular proteins and DNA while preserving ECM proteins collagen IV and laminin. Extracellular structures including a complete vascular network remained intact. In order to test the ability of the scaffold to support cell growth, mouse ES cells were injected into the renal artery. Embryonic stem cells have the potential to differentiate into all cells of the adult kidney. Proliferation and differentiation of these cells in response to the ECM would serve as a valuable step toward the generation of a functional kidney. Once injected, the cells primarily localized to glomerular tufts and adjacent capillaries. Unless the ECM was disrupted, a majority of these cells seemed to remain in the vasculature during culture. Conta ct with ECM promoted cell division and lack of contact promoted apoptosis. Immunohistochemical studies showed patches of epithelialization throughout the scaffolds. Of the three markers of kidney progenitor cell populations, only Pax 2 was detected. Ksp ca dherin and WT1 were absent from the cultured cells. Markers of endothelial lineage VEGFR2 and BsLB4 were both detected. Through this research, a whole organ kidney scaffold was created and the response of seeded mES cells to the scaffold was examined. Give n the specific decellularization protocols, cells, seeding method, and scaffold culture conditions, long term cell growth into renal structures was unlikely and kidney specific differentiation was limited. However, this research revealed multiple avenues n ot only for improving this outcome but for progressing toward a functioning implantable kidney.
133 Promoting Differentiation There are two primary approaches for improving renal differentiation of the mES c ultured in the kidney scaffolds: modifi cation of scaf fold production and addition of soluble factors to the growth media. The decellularization process involves a tradeoff between cellular material removal and preservation of basement membrane components. The best protocol for cell debris removal developed i n this research was Kidney SDS However, there is opportunity to reduce exposure to the decellularization chemicals while maintaining cell debris removal. This should improve retention of ECM embedded growth factors and growth factor associated proteins, u ltimately leading to stronger influence over seeded cells. Extracellular matrix associated growth factors that play a role in kidney development include VEGF, PDGF, HGF, FGF, TGF 7, and Wnt 4. 52 54 56 A prevalent kidney ECM protein that sequesters growth factors is heparan sulfate proteoglycan. 56 Once preserved growth factors are identified, a program of complementing growth factors or other additives to stimulate differentiation may be added to the scaffo ld growth media. If HSPG or other growth factor associated ECM proteins are preserved, loading growth factors into the scaffold before cell injection is an option. 171 One consideration in planning scaffold modifications or growth media additives is that ECM protein composition shifts during development. 10 52 Another is that growth factor distribution and dose in an adult matrix is likely related to homeostasis of adult renal function and not entirely suitable to differentiate embryonic cells. A third consideration is that mous e cells are implanted in a rat matrix, so embedded rat growth factors may have reduced potency to effect mouse cell differentiation. These considerations may be partially addressed by supplementing the scaffold or media with mouse growth factors and increa sing dose of growth factors shown to promote nephrogenic lineages, such as Activin, GDNF, and BMP7. 122 Qiao et al (1999) showed that
134 soluble cytokines are all that is necessary to initiate nephrogenesis from renal progenitor populations. 86 Scaffold Culture It was estimated that most of the mES cells that were injected into the renal artery of the decellularized kidney scaffolds re mained confined to the vasculature, but were able to eventually migrate into other areas due matrix disruption caused by the section scaffold culture method. In order to create whole organ tissue, however, the integrity of the scaffold must be preserved. T herefore, in order to access the tubules and collecting ducts, methods for improved ureteral seeding must be explored. These may include changes in injection pressure, solution volume, and total number of injections. It may also be that rat kidney is resis tant to retro grade perfusion and another animal model must be used. One advantage to retrograde seeding of ES cells is that the cells are exposed to tubular and collecting duct ECM and associate d growth factors from the onset, whereas the cells delivered i nto the vasculature may have been induced toward a vascular lineage and less responsive to other ECM after migration. Another advantage to ureteral seeding is that the vascular system may be populated with endothelial cells instead of reliance on endotheli al differentiation of ES cells. Endothelialization of decellularized vasculature protects it from thrombosis, allowing the repopulated kidney to be implanted. 18 23 101 103 147 199 Other cell seeded scaffolds, such as from liver, heart, and lung have been endothelialized and implanted and have shown basic organ function. 18 22 24 26 Sources of endothelial cells include human umbilical vein endothelial cells (HUVECs) and autologous harvest, such as from saphenous artery or vein. 18 22 181 199 Endothelial progenitor cells (EPCs), which can be isolated from peripheral blood, may also be used to line the vasculature 181 220 The culture perfusion system designed in this research caused a reduction of mES cells in the scaffold, especially the glomeruli. This may have been caused from excessive pressure
135 disrupting the scaffold and allowing the cells to escape. This may also have been caused from fluid induced shear stress reducing cell division. In any case, the cells localized to the tubules or collecting ducts did thrive. Therefore, once improved ureteral seeding of mES cells has been achi eved, perfusion culture through the vasculature may be a valuable next step. Perfusion bioreactors have been shown in other seeded whole organ scaffolds to improve cell growth and epithelial differentiation. 22 24 201 In regard to vascular endothelialization, perfusion culture also improves cell attachment, alignment, elongation and overall patency rate. 24 199 Cell A lternatives Once the scaffold and culture conditions have been optimized for renal differentiation of mES cells, testing the response of other cell sources may be worthwhile. Various adult kidney stem cells have shown ability to integrate into developing kidneys, but unless such cell lines were developed to be immune tolerant, they would require biopsy for ext racti on and growth in vitro for subsequent autologous transplantation. In addition, it is doubtful that cells extracted from a diseased or damaged kidney would be suitable for new tissue growth. Multipotent stromal cells could be isolated from autologous b one marrow or donor matched from a tissue bank. Although they have primarily demonstrated a supporting role in ki dney repair, Yokoo et al (2005 ) has shown that MSCs contribute to epithelial and interstitial cells when implanted into embryonic kidneys. 112 Amniotic fluid stem cells are another cell source that could take advantage of tissue banking. They are pluripotent and capable of renal specific differentiation and structure formation. 32 130 The most promising alternate cell to test, however, is iPSCs. They are a plurip otent cell source typically reprogrammed from adult autologous cells by forced expression of genes associated with embryonic cells. 33 Induced pluripotent stem c ells have been differentiated into renal lineages and respond to the same inductive factors that influence ES cells 122 Also, they have been reprogrammed from cell sources that do not require biopsy such as
136 mononuclear cells isolated from peripheral blood or tubule epithelial cells excreted in urine. 43 44 Considerations for using those cells however are that peripheral blood cells are susceptible to age and mutation and tubule cells may be affected by kidney disease. 135 Alternatively, iPSCs may be derived from tissue banked cord blood, umbilical cord matrix, or placental amniotic membrane 134 135 Compared to ES cells, based on epigenetic patterning, iPSCs are less likely to progress along developmental pathways and are more likely to differentiate back into source tissue. 133 139 This may be an advantage if the source tissue is healthy kidney. In a study of renal differentiation, fibrobla st derived iPSCs lagged behind ES cells when exposed to equivalent inductive factors. It was suggested that continued expression of the embryonic vectors in these iPSCs suppressed differentiation. One possible resolution is to use an alternate method for r eprogramming such as synthetic mRNA. 133 Finally, instead of an embryonic state, it may be possible to reprogram cells directly into a kidney precursor population. 131 Xenogenic Scaffolds In this research mouse cells were implanted into rat extracellular matrix scaffolds and with culture showed markers for endothelial progenitors, epithelial cells, and what is likely intermediate mesoderm. It should be considered, however, that this cell res ponse is to a xenogenic scaffold and may be diminished compared to what would have been seen with an allogenic scaffold under equivalent processing and culture conditions. Given this, it may be worthwhile to test cell response with a mouse acellular kidney In the end though, the only significant way to increase the pool of scaffold tissue for growing replacement kidneys aside from considerable scaffold fabrication technology improvements, is to use xenograft sources. Therefore, the most viable path to cli nical significance is to optimize xenogenic scaffold production and culture conditions for best renal specific cell response. Further more as optimal scaffold decellularization protocols and culture conditions are likely to shift depending on cell
137 and tiss ue sources, early success of this rodent based research should prompt advancement to human cells on porcine derived scaffolds, as that pairing is the most likely to be clinically successful. It has been shown that xenogenic tissue can be decellularized to reduce immunogenicity and allow constructi ve remodeling upon implantation 98 102 156 This remodeling or specific host cell infiltration and basement membrane production and is considered a positive interaction of graft tissue and host cells. 147 221 As an extension of this research, the decellularized rat kidneys implanted with mES cells were examined for mouse ECM productio n. The analysis showed presence of mouse specific laminin. 213 This is a positive sign of xenograft remodeling with one caveat. Composition of deposited ECM is refl ected by the current phenotype and environment of the cells. 222 If cells of early nephrogenic lineages produce ECM, it may shield signaling cues from the adult matrix, inhibiting further d ifferentiation. In this case, there may be more reliance on culture based signals. One solution may be to reduce the level of growth factors that promote ECM production, such as TGF 55 173 Once cells in the kidney scaffolds have been sufficiently d ifferentiated and arranged in the extracellular scaffold, research can progress to in vitro function testing and in vivo implantation. One method used by other groups to test organ constructs with intact vasculature is by circulation of blood. 22 26 In the case of populated kidney scaffolds, the filtrate collected in the ureter could be examined for glucose, amino acids electrolytes and other molecules to test filtration specificity of the glomerulus and resorptive function of the epithelial cells. The final step in functional analysis is to implant the kidneys by anastomosing to available vasculature, similar to studie s of seeded scaffolds that have been genera ted from heart, liver, and lung 22 24 26
138 28 Demonstration of renal function recovery would be a significant advancement for those suffering from renal disease.
139 CHAPTER 6 MATERIALS AND METHODS Organ Harvest All kidneys used in the establishment of a decellularization protocol for renal tissue engineering or any subsequent cell seeding and culture experiments were harvested from male Sprague Dawley rats of at least 2 months of age and 250g weight The rats were obtained from Harlan Laboratories Inc. (Indianapolis, IN) and housed in the VM 19 Metabolic Building barrier facility at the University of Florida Veterinary Medical Complex under the care of Animal Care Services (Gainesville, FL). All animal use and associated procedures were performed in accordance with the guidelines su pplied by the Institute for Lab Animal Research Guide for the Care and Use of Laboratory Animals, regulated by the Public Health Service Policy on Humane Care and Use of Laboratory Animals, and approved by the Institutional Animal Care and Use Committee at the University of Florida (UF). Rat kidneys were harvested according to a modified form of the non survival procedure described in Aden et al. (2005). 143 Animals were anesthetized by chamber induction with 5% isoflurane in 100% oxygen, and then maintained by mask administration of 2 3% isoflurane in 100% oxygen. This method of anesthesia was selected to minimized animal handling and stress prior to the harvest proc edure, as well as to better control any discomfort or pain experienced during the procedure. Depth of anesthesia was checked according to righting reflex, reaction to toe pinch as well as overall disposition of the animal (breathing and heart rates). Tempe rature of the animal was maintained with a circulating water warming pad. After anesthesia was induced, the abdomen was shaved, cleaned, and wiped with isopropyl alcohol Once the animal was secured in dorsal recumbency, a ventral midline laparotomy was pe rformed and the viscera exteriorized to expose the right kidney. Then, after blunt and sharp dissection, sutures were pre
140 placed around the distal cranial mesenteric artery (CMA), and the aorta and vena cava superior to the CMA and inferior to the renal ar tery. These served to isolate the vascular supply and avoid hemorrhage during cannulation. Before tightening the sutures, 1400 units/kg of heparin sodium was delivered into the peritoneal cavity to reduce coagulation in the organ. Once the vessels were lig ated, an ophthalmic 19G Park Irrigating Cannula (Bausch & Lomb Surgical, Inc., Rancho Cucamonga, CA) was quickly introduced into the cranial mesenteric artery and directed across the aorta and into the right renal artery. Sutures were then used to secure t he cannula to the renal artery. Once anchored, the renal vein was nicked and the kidney was flushed with 40ml of a 0. 9% saline solution containing sodium nitroprusside, a vasodilator; during the flush, the kidney was observed for blanching to assure a comp lete rinse of blood. After the rinse, the attaching vessels and ureter were cut and the kidney was removed en bloc from the retroperitoneal space. Then, while under anesthesia, the rat was euthanized with an intracardiac injection of 120 mg/kg sodium pento barbital with confirmation by thoracotomy. On occasion, variant anatomy or damage of the CMA necessitated an alternate route of cannulation, and it has been achieved with aortic cannulation both superior and inferior and directed into to the renal artery. Another procedural variation was the addition of a ureteral cannula, PE 10 tubing, which was used for filtrate collection or retrograde cell delivery. Organ Decellularization Organ Perfusion Two perfusion systems were used to decellularize kidneys. In both cases, the harvested kidney, suspended by the renal artery and attached cannula, were affixed to the system by a luer fitting. Prescribed decellularization solutions were then perfused into the kidney through the renal artery and drained from the renal ve in or ureter into a collection receptacle for disposal or
141 recirculation. Perfusates were delivered at room temperature with a pressure maintained between 90 and 110 mmHg, within rat physiologic limits. The first perfusion system used was the LifeSustainer 1000 automated organ perfusion workstation (LifeSystems, Redmond, WA). 144 It was a system designed to maintain viability of an organ ex vivo for scientific research and monit ors pH, fluid pressure, as well as oxygen and carbon dioxide saturation. In addition, it has controls for regulation of temperature and perfusion with flow rate or pressure regulation. Once the kidneys were fixed to the machine, perfusate solutions were pu lled from prepared bottles and circulated according to the prescribed decellularization protocol. The second perfusion system was designed and const ructed i n house and drastically improved pressure regulation and air entrapment in fluid lines. In addition it included a large debris entrapment (Whatman, Inc., Piscataway, NJ), and capacity for decellularization of two kidneys simultaneously. A diagram and apparatus i mages are included in Appendix A (Figures A 1 and A 2) Decellularization Protocols Harvested kidneys, once attached to a perfusion system, were subject to a series of solutions delivered into the vasculature with the express goal of removing all cellular material in a progressive manner without causing blockage within the intricate vascular and tubular networks or disruption of the extracellular matrix. The following protocols were based on earlier works from Brendel, Meezan, and Nagle (1979 1980 ), who de sired to isolate the basement membrane of the kidney and other organs for study of morphology, permeability, and chemical composition. 141 142 It was the purpose of this work to evaluate and modify the original decellularization protocols so as to produce an acellular whole organ kidney amenable to cell
142 seeding and other tissue engineering or regenerative appl ications. The protocols include d use of the nonionic detergent Triton X 100 (Sigma Aldrich, St. Louis, MO) for solubilization of hydrophobic cell membranes, cell lysis, and washing of resulting debris. Additives were used such as sodium chloride to osmotic ally disrupt nuclear membranes or calcium chloride (Fisher Scientific, Hampton, NH) and magnesium sulfate (Fisher) to improve membrane permeability and activate endogenous nucleases for digestion of bulky nuclear material. DNA was enzymatically degraded wi th a buffered solution of 2000 Kunitz units/mg deoxyribonuclease 1 (DNase) from bovine pancreas (Sigma). The buffer constituents sodium acetate, magnesium chloride, and sodium chloride were all acquired from Fisher Scientific. Ionic detergents such as sodi um deoxycholate ( NaDC ) and sodium dodecyl sulfate (SDS) were optionally used to remove any debris or non soluble cellular material. Finally, deionized water was used as an intermediate or final rinsing step and may have exerted osmotic stress on cell membr anes. Each of the detergent solutions contained .05% sodium azide as an antimicrobial and all solutions were aqueous After a final rinse, if a kidney was intended for cell inj ection and culture, 50cc of Dulbecco's modified Eagle's m edium ( DMEM ) was perfus ed and the cannula was sprayed with 70% ethanol before removal from the decellularization apparatus. The following tables describe the decellularization protocols used for preparing an acellular kidney tissue engineering scaffold. Appended to each is a de scription of modifications. Decellularization protocol: VBM. This protocol ( Table 6 1) was based on Brendel and VBM preparation (Brendel, 1980). 142 Decellularization protocol: VBM Mod 1. This protocol was identical to VBM ( Table 6 1) except with reduction of the Triton X 100 detergent concentration in step 1 to 0. 5%. This
143 specific protocol (1979), was thought to minimize sudden release of proteinaceous debris and subsequent clogging of tubules. 141 Decellularization protocol: Kidney Mod 1. This was the first protocol used with the elevated reservoir perfusion apparatus (Appendix A 1). Steps were incorporated from the acellular perfused kidney protocol (Brendel, 1979) to improve debris removal and rupture of intact nuclear envelopes ( Table 6 2). This inc luded changing perfu sion durations, the addition of calcium chloride, magnesium sulfate, and sodium chloride to the detergent solutions, and use of Table 6 1. VBM decellularization protocol Step Dur Rec Solution 1 24h N 2% v/v Triton X 100 2 24h Y 2% v/v Triton X 100 3 2h N De ionized water 4 15m N 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 4h Y 2% v/v Triton X 100 6 1h N De ionized water 7 2h N 20% v/v ethanol 8 1h N De ionized water Total duration: 58h. Key: dur, length of perfusion; rec, perfusate is recirculated and filt ered; solution, perfusate recipe. Table 6 2. Kidney Mod 1 decellularization protocol Step Dur Rec Solution 1 3h N 0.5% v/v Triton X 100 2 3h Y 0.5% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 3 2h N De ionized water 4 4h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 3h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 6 3h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 7 2h N De ionized water 8 4h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 9 2h N De i onized water 10 2h Y 4% w/v sodium deoxycholate 11 4h N De ionized water Total duration: 58h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solution, perfusate recipe.
144 the ionic detergent sodium deoxycholate as an extra cleaning step. Also, a second DNase step was added to re duce any buildup of viscous DNA. Finally, there was removal of ethanol to avoid any fixation. Decellularization protocol: Kidney Mod 2. The duration of detergent perfusion steps, both nonionic and ionic, were increased from Kidney Mod 1 to test for better cell debris removal ( Table 6 3). Specifically, step 6, the most aggressive nonionic dete rgent step, was significantly increased to a duration commensurate with the acellular perfused kidney protocol (Brendel, 1979). 141 Decellularization protocol: Graduated 3%. This protocol was developed to offer a slow, graduated increase in detergent concentration with late addition of additives (calcium chloride, magnesium sulfate and s odium chloride). There was also only one DNase step ( Table 6 4). Decellularization protocol: Graduated 10%. This protocol was similar to Graduated 3% except for more aggressive detergent concentrations ( Table 6 5). Table 6 3. Kidney Mod 2 decellularization protocol Step Dur Rec Solution 1 3h N 0 .5% v/v Triton X 100 2 9h Y 0 .5% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 3 2h N De ionized water 4 4h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 6h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 6 16h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 7 2h N De ionized water 8 4h Y 50mM sodium acetate 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 9 2h N De ionized water 10 8h Y 4% w/v sodium deoxycholate 11 6h N De ionized water Total duration: 62h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solution, perfusate rec ipe.
145 Decellularization protocol: Kidney Extended. This protocol resemble d Kidney Mod 2 yet with a very long Triton X 100 with additives detergent step interrupted by a DNase step ( Table 6 6). Decellularization protocol: Kidney Extended 2. This protocol was identical to Kidney Extended except for the remo val of the first DNase step and rinse, and a lengthening of the Triton X 100 with additives step to gain an equivalent total duration ( Table 6 7). Table 6 4. Graduated 3% decellularizatio n protocol Step Dur Rec Solution 1 3h N 0 .1% v/v Triton X 100 2 6h Y 0 .1% v/v Triton X 100 3 9h Y 0 .5% v/v Triton X 100 4 9h Y 3% v/v Triton X 100 5 9h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 6 9h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 7 2h N De ionized water 8 6h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 9 3h N De ionized water 10 9h Y 4% w/v sodium deoxycholate 11 4h N De ionized water Total duration: 69h. Key: dur, length of perfusion; rec, perfus ate is recirculated and filtered; solution, perfusate recipe. Table 6 5. Graduated 10% decellularization protocol Step Dur Rec Solution 1 3h N 0 .5% v/v Triton X 100 2 7h Y 0 .5% v/v Triton X 100 3 10h Y 3% v/v Triton X 100 4 10h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 5 10h Y 6% v/v Triton X 100, 5mM CaC l 2 5mM MgSO 4 1M NaCl 6 10h Y 10% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 7 2h N De ionized water 8 6h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 9 2h N De ionized water 10 10h Y 4% w/v sodium deoxycholate 11 4h N D e ionized water Total duration: 74h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solution, perfusate recipe.
146 Decellularization protocol: Kidney NaDC This protocol was another combination of the Brendel et al. decellularization protocol s (1979, 1980) that incorporated the detergent additives and ionic detergent from the former and the con venient 24 hour based timing of the second protocol ( Table 6 8). 141 142 Also, in contrast to the previous protocols, the initial non recirculating step was changed to recirculating due to reliance on the inline filter to capture cell debris. Table 6 6. Kidney Extended decellularization protocol Step Dur Rec Solution 1 3h N 0 .5% v/v Triton X 100 2 24h Y 3% v/v Triton X 100 3 40h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 4 6h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 2h N De ionized water 6 64h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 7 6h Y 5 0mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 8 2h N De ionized water 9 20h Y 4% w/v sodium deoxycholate 10 4h N De ionized water Total duration: 171h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solu tion, perfusate recipe. Table 6 7. Kidney Extended 2 decellularization protocol Step Dur Rec Solution 1 3h N 0 .5% v/v Triton X 100 2 24h Y 3 % v/v Triton X 100 3 112h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 4 6h Y 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 2h N De ionized water 6 20h Y 4% w/v sodium deoxycholate 7 4h N De ionized water Total duration : 171h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solution, perfusate recipe.
147 Decellularization protocol: Kidney SDS. This protocol was identical to the Kidney NaDC protocol except for replacement of ionic detergent with an equivalent concentration of SDS ( Table 6 9). Decellularization protocol: 1 Day SDS. This protocol tested if the ionic detergent sodium dodecyl sulfate was the primary determinant of decellularization success by removing all steps in the Kidney SDS protoco l that precede SDS treatment. ( Table 6 10) Table 6 8. Kidney NaDC decellularization protocol Step Dur Rec Solution 1 21h Y 3% v/v Triton X 100 2 27h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 3 2h N De ionized water 4 22h N 50mM sodium acetate, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 1h N De ionized water 6 23h N 3% v/v Triton X 100 7 21h Y 4% w/v sodium deoxycholate 8 3.5h N De ionized water Total duration: 120.5h. Key: dur, length of perfusion; rec, perfusate is recirculat ed and filtered; solution, perfusate recipe. Table 6 9. Kidney SDS decellularization protocol Step Dur Rec Solution 1 21h Y 3% v/v Triton X 100 2 27h Y 3% v/v Triton X 100, 5mM CaCl 2 5mM MgSO 4 1M NaCl 3 2h N De ionized water 4 22h N 50mM sodium aceta te, 10mM MgCl 2 100mM NaCl, .0025% DNase (pH 5.0) 5 1h N De ionized water 6 23h N 3% v/v Triton X 100 7 21h Y 4% w/v sodium dodecyl sulfate 8 3.5h N De ionized water Total duration: 120.5h. Key: dur, length of perfusion; rec, perfusate is recirculated and filtered; solution, perfusate recipe.
148 In V itro Embryonic Stem Cell Engraftment and Culture Embryonic Stem Cells Green fluorescent murine embryonic stem (mES) cells (B5/EGFP, courtesy of Andras Nagy) were used to test the ability of the acellular kidney scaffolds to direct differentiation of implanted pluripotent stem cells. 194 These cells have been well characterized and the persiste nce of fluorescence through cell division and differentiation allowed visualization not only of initial cell seed distribution but the subsequent cell migration and grouping patterns over the length of the culture period. The murine B5/EGFP ES cells were c ultured according to Singh et al. (2007). 223 The mES cells were maintained in an undifferentiated state on gelatin co ated dishes in Knock out DMEM (Life Technolog ies BRL, Grand Island, NY) containing 10% knockout serum replacement (Life Technologies BRL), 1% fetal bovine serum (Atlanta Biologicals, Norcross, GA), 2 mM L Technologies BRL), 1000 units/ml recombinant mouse LIF (ESGRO Chemicon, Temecula, CA). To prepare the cells for injection, the undifferentiated ES cells were dissociated using 0.25% trypsin/EDTA (Life Technologies BRL). Then, they were su spended in a solution devoid of LIF, which was also used as the growth media that would permit differentiation in all three of the scaffold incubation protocols described below: Iscove's modified Dulbecco's m edium ( IMDM ) supplemented with 20% fetal bovine serum (Atlanta Biologicals), 2 m M L Table 6 10. Kidney 1 Day SDS decellularization protocol Step Dur Rec Solution 7 24h Y 4% w/v sodium dodecyl sulfate 8 3.5h N De ionized water Total duration: 27.5h. Key: dur, length of perfusion; rec, perfusa te is recirculated and filtered; solution, perfusate recipe.
149 monothioglycerol. All cultures in this study were maintained at 37 C in 5% CO 2 The B5/EGFP ES cell line was maintained by Dr. Takashi Hamazaki in the laboratory of Dr. Naohiro Terada. Cell Engraftment and Culture Immediately following decellularization, kidneys that were allocated for mES cell seeding were perfused with 20ml of DMEM (Fisher Scientific) and placed in a sterile container for transport. Then under sterile conditions, 2x10 6 ES cells suspended in growth medium were slowly injected into the arterial or ureteral cannula so as not to cause an EC M rupture from pressure buildup One of three scaf fold culture methods was then followed, whole kidney static culture, whole kidney perfused culture, or sectioned scaffold culture as described below. Scaffolds were cultured for durations of 6, 10, and 14 days, with additional time points taken at 3, 4, or 8 days as tissue was available. This range of time points was selected to capture any transient differentiation patterns that may parallel normal murine nephrogenic development. Frequency of media change is described within each culture method; however, a t first change in scaffold culture, gross cell localization, migration, and grouping tendencies were monitored through GFP expression by fluorescence microscopy ( IX70; Olympus, Tokyo, Japan). Whole scaffold static culture. The initial culture method used was simple submersion of the seeded scaffold into a 50ml conical culture tube. Growth media was refreshed every 2 3 days. Perfused organ culture. In an effort to i mprove viability and provide a more suitable microenvironment, seeded acellular scaffolds were placed in a perfused culture system. This sealed system was specifically designed to allow independent adjustment of mean pressure and pulse rate, with a pulsati le pressure differential of between 20 and 30 mmHg. This allowed
150 gradual increase of mean pressure over time with maintenance of periodicity. Typically, the scaffold s were injected with cells and affixed to the apparatus with submersion in media, but circu lation of the culture was not initiated for 12 hours. This method gave the cells time to adhere to the ECM before being exposed to shear stress. Medical grade sterile gases (5% CO 2 air mix, Praxair, Inc., Danbury, CT) were delivered to the culture chamber and media reservoir, to maintain media at pH 7.4. Given the high volume of the circulated media reservoir, growth media was only changed every 6 7 days. An apparatus diagram and images are shown in the appendices (Figure A 3 and A 4 ). Sectioned scaffold cu lture. The final seeded scaffold culture employed involves sectioning the kidney in thick sections so that multiple time points can be accomplished with a single decellularized organ preparation. In addition, it was assumed that sectioning of the kidney wo uld lessen any diffusional barriers that developed after continued cell growth. Initially, a fresh scaffold was injected with cells and left to sit in a deep cell culture dish (60cm) for 24 hours to let the cells adhere. Then the scaffold was sectioned tra nsversely in 2 3 mm increments, with samples placed in a 12 well cell culture dish. Sections which retained the primary functional areas of the kidney, cortex, outer medulla, and inner medulla were used for data time points Embryoid Bodies Embryoid bodies were cultured at analogous time points to the seeded kidney scaffold cultures. This allowed ES cells to multiply and differentiate without influence of contacting extracellular matrix. This experimental control was useful in immunohistochemical and gene e xpression comparisons. Using the differentiation medium described above, ES cells were cultured for 2 days by the hanging drop method (2 10 3 ES cells per each 30 l drop). Afterward, embryoid bodies in hanging drops were transferred to suspension culture in 100 mm petri dishes and incubated for 3,6,10 and 14 days. At the designated time point EBs were
151 of embryoid bodies was performed by Dr. Takashi Hamazaki i n the laboratory of Dr. Naohiro Terada. Histochemistry Tissue Preparation Most of the freshly prepared decellularized kidneys and cell seeded and cultured scaffolds were prepared for histological evaluation. First, the kidneys were submerged in 10% neutral buffered formalin for 24 hours. The NBF was Richard Allan Scientific brand and distributed by Fisher Scientific. Afterward, the kidney scaffolds were transferred to phosphate buffered saline and submitted to the Molecular Pathology & Immunology Core at th e University of Florida for paraffin embeddi midline, warmed to 42 C in a water b ath, placed on Superf rost Plus positively charged slides (Fisher Scientific), and drie d overnight at room temperature using a ductless enclosure. Slides wer e then earmarked for H&E staining or IHC; t hose protocols are described below. Completed histology slides were observed with an Olympus BX51 Microscope and photographed using an Optronic Digital Camera with Magnafire 3.1 software. Hematoxylin and Eosin Sta ining The primary method for tissue evaluation was microscopic inspection of hematoxylin and eosin (H&E) stained paraffin embedded tissue. The slide mounted tissue sections were deparaffinized, rehydrated, stained, and dehydrated according to Table 6 10. S teps include use of Hematoxylin 7211 for blue staining of cell nuclei, Clarifier 1 to remove residual hematoxylin bound to the slide, Bluing Reagent to enhance the blue color, and Eosin Y to stain proteins pi nk. These reagents were all Richard Allan Scientific brand. Once the tissue was stained, coverslips were applied to the slides using Cytoseal XYL (Richard Allan Scientific). All
152 hemato xylin and eosin histological staining was performed at the UF Stem Cell Program, Core Histology Laboratory. For the purpose of scaffold evaluation, each of the samples was scrutinized for preservation of extracellular matrix morphology in the cortex, outer medulla, and inner medulla of the kidney scaffold. In addition, it was noted if there were intact cells, nuclei, stained DNA aggregates, or other un cleared cellular debris. For analysis of seeded and cultured kidney scaffolds, the slides were inspected f or cell distribution, grouping, morphology, ECM association, and appearance of apoptosis. Table 6 11 Hematoxylin and Eosin Staining P rotocol Step Duration Solution 1 5m Xylene 2 5m Xylene 3 2m 100% ethanol 4 2m 100% ethanol 5 1m 95% ethanol 6 1m 95% ethanol 7 1m Distilled water 8 2m 15s Hematoxylin 9 1m Distilled water 10 1m Distilled water 11 1m Clarifier 12 1m Distilled w ater 13 1m Bluing Reagent 14 1m Distilled water 15 1m 80% ethanol 16 1m 30s Eosin Y 17 1m 95% ethanol 18 1m 95% ethanol 19 1m 100% ethanol 20 1m 100% ethanol 21 1m xylene 22 1m xylene 23 1m xylene Slides were spaced on a rack and sequentially s ubmerged in a series of dishes containing the solution as indicated above
153 Immunohistochemistry Immunohistochemistry was used to detect retention of extracellular proteins in rat decellularized kidney scaffolds and to identify patterns of cell division, apoptosis, and differentiation of implanted murine embryonic stem cells. In addition, embryoid bodies and embryonic kidneys (E17.5) were used to provide experimental controls. As cell implantation was on a xenogenic scaffold, both rat and mouse tissue was used to test for antibody cross reactivity or non specific b inding. All IHC, with exception of the antibodies indicated in Table 6 11, was performed at the UF Stem Cell Program, Core Histology Laboratory (Gainesville, FL). The stains were implemented using the Vectastain Elite ABC Kit (Vector Laboratories, Inc., Bu rlingame, CA) according to manufac an enzyme based (immunoperoxidase) reaction and the chromogen 3,3' Diaminobenzidine (DAB) to develop color. To initiate the process, tissue slides were deparaffinized and hydrated accordin g to Table Table 6 12 Immunohistochemical Antibodies Target Protein Species Company Cat # Dilution Retrieval Cleaved Caspase 3 Rabbit Cell Signaling Tech 9661 1:50 Dako Coll agen IV Rabbit Abcam AB13966 1:500 Prot K GFP Rabbit Abcam 290 1:500 Citrate Ki67 Rabbit Leica NCL KI67P 1:2000 Dako Ksp cadherin Mouse Invitrogen 35 4700 1:200 None Laminin Rabbit Abcam AB11575 1:25 Prot K Oct4 Rabbit Abcam AB20650 1:1000 Hi pH Pan cytokeratin* Mouse Ventana 760 2135 Pax 2 Rabbit Invitrogen 71 6000 1:200 Citrate WT1* *IHC performed by Shands Medical Laboratories at Rocky Point (Gainesville, FL) Retrieval methods: citrate, sodium citrate buffer (10mM Sodium Citrate, 0.05% Tween 20, pH 6.0) for 25 mins at RT; Dako, Dako Target Retrieval Solution (Dako North America, Inc., Carpinteria, CA) for 20 mins at 95C and 20 mins at RT; Hi pH, distilled water (pH 10.0) for 25 mins at 95C and 10 mins at RT; Prot K, proteinase K enzyma tic solution Company details: Abcam, Inc. (Cambridge, MA); BD Pharmingen (San Diego, CA); Cell Signaling Technology, Inc. (Danvers, MA); Invitrogen Corporation (Carlsbad, CA); Leica Microsystems Inc. (Buffalo Grove, IL); Ventana Medical Systems, Inc. (Tu cson, Arizona)
154 6 12. Included was a step to quench endogenous peroxidase activity (3% hydrogen peroxide). Next, an antigen unm asking or retrieval step was performed, the specific method indicated and described on Table 6 11. The steps proceeded according to instruction, with TBS T (Tris buffered saline + 0.1% Tween) buffer and the use of an avidin/biotin blocking step (Avidin/Bio tin Blocking Kit, Vector Laboratories). Primary antibodies ( Table 6 11) with listed dilutions were incubated overnight at 4C, except for GFP, which was incubated for 1 hour at 4C. Secondary antibodies were incubated for 30 minutes. Finally, after the sli des were developed and counterstained with hematoxylin (Hematoxylin 7211, Richard Allan Scientific), the slides were covered using Cytoseal XYL (Richard Allan Scientific) mountant. Scanning Electron Microscopy Scaffold morphology, including quality of dece llularization and basement membrane continuity, was evaluated using the scanning electron microscopy facilities at the Major Analytical Instrumentation Center, University of Florida. Sample preparation and equipment operation was performed by Dr. Brad Will enberg. The freshly decellularized kidneys were fixed overnight in 2.5% glutaraldehyde at room temperature, freeze dried, then fractured on the transverse, midline plane. The samples were mounted on an aluminum stub (SPI Supplies, West Chester, PA) and spu tter coated with carbon (Ion Equipment Corp., Santa Clara, CA). The Table 6 13 Slide Pretreatment for Immunohistochemistry Step Duration Solution 1 5m Xylene 2 5m Xylene 3 2m 100% ethanol 4 2m 100% ethanol 5 10m 3% hydrogen peroxide in methanol 6 3m 95% ethanol 7 1m 70% ethanol 8 1m Distilled water 9 1m Distilled water Slides were spaced on a rack and sequentially submerged in a series of dishes containing the solution as indicated above.
155 prepared samples was analyzed using a JEOL JSM 6400 SEM (JEOL USA, Peabody, MA) equipped with an Oxford energy dispersive spectroscope (EDS) system and a LINK ISIS software package version 3.35 (Oxford Instruments USA, Concord, MA).
156 APPENDIX A DECELLULARIZATION AND PERFUSION CULTURE APPARATUSES Figure A 1. Decellularization apparatus diagram This system was designed to perfuse an organ at constant pressure. The pressure was determined b y the height of the solution in an elevated reservoir relative to the organ. The pressure was adjusted by raising or lowering the reservoir bottle. The decellularization solution in the reservoir bottle (1) passed through a .2/.8 micron inline fluid filter (2) and was delivered into the cannulated renal artery of the kidney (3). The solution perfused the kidney, emptied out of the ureter or renal vein, and was collected in a beaker (4). A peristaltic pump (5) then drew out the perfused fluid and either dire cted it to a waste container or recirculated it back to the elevated reservoir bottle. Pressure was monitored (6) at the arterial cannula (7). Leaks in the descending fluid line or an insecure arterial cannula caused increased flow and corresponding pressu re drop. Flow through the kidney was also affected by type of solution and decellularization progress. Progressive removal of cellular material during decellularization lowered fluid resistance. Despite these factors, and the variable reservoir fluid heigh ts, pressure was typically maintained to within +/ 10% of 100 mmHg. 7 1 4 6 2 3 5
157 Figure A 2. Pressure based organ perfusion apparatus for decellularization A) Perfusate reservoir (1) with descending fluid line that passes through a .2/.8 filter (2). B) Continuatio n of descending fluid line terminating into a Luer fitting (3) that allows both organ attachment and pressure measurement (4). A beaker (5) collects perfusate that has passed through the organs and optionally returns it to the elevated reservoir with a per istaltic pump (6). The pump controller is also pictured (7). As pictured, there is no kidney being decellularized and the reservoir bottle return line is not attached. Tygon 3350 tubing was used for minim al extractables and ultra smooth inner bore for redu ced potential for particle entrapment.
158 Figure A 3. Perfusion culture apparatus diagram Primary high flow media circuit: media was drawn from a reservoir (1), passed through a .2/.8 micron filter (2), pulled through a variable speed peristaltic pump (3 ), and driven through both a tubing splitter (4) and simple elevated assembly (5 and 6) before returning to the reservoir (1). The elevated assembly, a tubing connector (5) vented to the air through a filter (6), allow ed media pushed to a prescribed height to mix with the air and passively fall to the reservoir. The tubing splitter adjacent to the p ump (4) let media travel through the kidney (7) at pressure determined by the height of the vertical media column. Media was not solely forced into the kidney so there was little risk of ECM rupture ; the flow rate through the kidney was therefore dependent on resistance in the kidney Th e pressure waveform was monitored (8) in close proximity to the kidney (diagram is not precise in arrangement o r scale). Finally, media that was pa ssed through the kidney collected in the media reservoir (1). To maintain sterility and improve gas diffusion in media, tank delivered (9) sterile biomedical grade gas (5% CO 2 in air mix) was gently passed through the organ chamber (7) an d media rese rvoir (1). An air filter (10) was used on exhaust to assure sterility. To control temperature, t he orga n chamber and media reservoir were placed within an incubator (11 ). 4 5 6 1 3 8 9 10 2 7 1 1
159 m f s Figure A 4. Organ perfusion bioreactor This system was designed to circulate media through a seeded acellular organ. Tubing, connectors, and bottles were all reusable and sterilizable. Most of the plumbing could be assembled in a biological hood, and only a couple of connections outside the hood were necessary. A) Incubator with adjacent biomedical gas tank. B) In terior of incubator containing an upper scaffold perfusion bottle (s) a lower media reservoir (m) and a media filter (f) The media filter prevented cells from traveling though circulation. Also visible are the white disc air filters. C) Elevated assembly that controlled pressure to k idney. D) Pump and controller located behind incubator. Fluid lines were passed through the ventilation port on the back of the incubator (arrow) E) A tubing split connector (hidden) granted fluid access to a pressure transducer in proximity to the kidney The disposable TruWave pressure transducers were gifted from Edwards Lifesciences (Irvine, CA). Pressure waveforms were observed using a donated monitor from Clinical Engineering at Shands Hospital (Gainesville, FL). F) Placement of seeded acellular kidn ey into the perfusion chamber
160 APPENDIX B SUPPLEMENTAL HISTOLOGY A B C D E F G H I Figure B 1. Reference immunohistochemical comparison of normal rodent kidney tissue (40x magnification, bars denote 50m). Pax 2 in A) adult rat, B) adult mouse, and C) embryonic mouse at 17.5 days post conception; Ksp cadherin in D) adult rat, E) adult mouse, and F) embryonic mouse at 17.5 days post conception; G) Collagen IV in adult rat; H ) laminin in adult rat; and I ) pan Cytokeratin in embryonic mouse at 17.5 days post conception.
161 A B Figure B 2. Localization patterns of unremoved cellular proteins post Kidney NaDC decellularization using IHC (40x magnification, bars denote 50m). A) WT1 in expected glomerular location, and B) possible diffuse Ksp cadherin protein in unremoved proteinaceous d ebris.
162 A B C D E F G H I J K L Pan CK Pax 2 Ksp cadherin KI 67 Oct 4 CC3 Figure B 3. Detection of key proteins for cell division, apoptosis, and differentiation in B5/EGFP ES cells cultured as embryoid bodies (40x magnification, bars den ote 50m) Ksp cadherin IHC slides shown in vertical pairs, the top at day 0 and the bottom at day 10 or 14. A, D) KI 67 positivity shows most cells undergoing cell division at day 0 and many intact cells at day 14. B, E) CC3, a marker of apoptosis, is onl y present in a few scattered cells at day 0, but significantly larger areas at day 14. C, F) Partial retention of Oct 4 at day 10 shows remaining presence of undifferentiated stem cells. G, J) Pan cytokeratin marks epithelial differentiation and is absent at day 0 and present at day 10. H, K) Pax 2 is present in intermediate mesoderm, ureteric bud, and cap mesenchyme populations and day 10 shows heavy positivity while day 0 is negative. I, L) Lack of KSP cadherin at day 0 and 10 indicates no spontaneous dif ferentiation into ureteric bud epithelium or polarized renal epithelium.
163 LIST OF REFERENCES 1. U. S. Renal Data System USRDS 2012 Annual Data Report: Atlas of Chronic Kidney Disease and End Stage Renal Disease in the United States. Bethesda (MD): National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2012. Ava ilable at http://www.usrds.org/adr.aspx 2. Mason C, Dunnill P. A brief definition of regenerative medicine. Regen Med 2008; 3:1 5. 3. Sullivan DC, Mirmalek Sani SH, Deegan DB, Baptista PM, Aboushwareb T, Atala A, et al. Decellularization methods of porcine kidneys for whole organ engineering using a high throughput system. Biomaterials 2012; 33:7756 64. 4. Columbus: McGraw Hill M edical, 2012. 5. Atala A. Regenerative medicine strategies. J Pediatr Surg 2012; 47:17 28. 6. Dehghani F, Annabi N. Engineering porous scaffolds using gas based techniques. Curr Opin Biotechnol 2011; 22:661 6. 7. Hartman O, Zhang C, Adams EL, Farach Carson MC, Petrelli NJ, Chase BD, et al. Biofunctionalization of electrospun PCL based scaffolds with perlecan domain IV peptide to create a 3 D pharmacokinetic cancer model. Biomaterials 2010; 31:5700 18. 8. Lee SJ, Liu J, Oh SH, Soker S, Atala A, Yoo JJ. Devel opment of a composite vascular scaffolding system that withstands physiological vascular conditions. Biomaterials 2008; 29:2891 8. 9. Neal RA, McClugage SG, Link MC, Sefcik LS, Ogle RC, Botchwey EA. Laminin nanofiber meshes that mimic morphological propert ies and bioactivity of basement membranes. Tissue Eng Part C Methods 2009; 15:11 21. 10. Miner JH. Renal basement membrane components. Kidney Int 1999; 56:2016 24. 11. Gilbert TW, Sellaro TL, Badylak SF. Decellularization of tissues and organs. Biomaterial s 2006; 27:3675 83. 12. Brooks DN, Weber RV, Chao JD, Rinker BD, Zoldos J, Robichaux MR, et al. Processed nerve allografts for peripheral nerve reconstruction: a multicenter study of utilization and outcomes in sensory, mixed, and motor nerve reconstructio ns. Microsurgery 2012; 32:1 14. 13. Brown JW, Elkins RC, Clarke DR, Tweddell JS, Huddleston CB, Doty JR, et al. Performance of the CryoValve SG human decellularized pulmonary valve in 342 patients relative to the conventional CryoValve at a mean follow up of four years. J Thorac Cardiovasc Surg 2010; 139:339 48.
164 14. Buinewicz B, Rosen B. Acellular cadaveric dermis (AlloDerm): a new alternative for abdominal hernia repair. Ann Plast Surg 2004; 52:188 94. 15. El Kassaby AW, Retik AB, Yoo JJ, Atala A. Urethral stricture repair with an off the shelf collagen matrix. J Urol 2003; 169:170 3; discussion 3. 16. Pu LL, Plastic Surgery Educational Foundation DC. Small intestinal submucosa (Surgisis) as a bioactive prosthetic material for repair of abdominal wall fasci al defect. Plast Reconstr Surg 2005; 115:2127 31. 17. Wang JC, Alanay A, Mark D, Kanim LE, Campbell PA, Dawson EG, et al. A comparison of commercially available demineralized bone matrix for spinal fusion. Eur Spine J 2007; 16:1233 40. 18. Baptista PM, Sid diqui MM, Lozier G, Rodriguez SR, Atala A, Soker S. The use of whole organ decellularization for the generation of a vascularized liver organoid. Hepatology 2011; 53:604 17. 19. Claudius C, Schuetz C, Clippinger B, Vacanti J, Markmann J, Ott H. Bio enginee red endocrine pancreas based on decellularized pancreatic matrix and mesenchymal stem cell/islet cell coculture. J Am Coll Surg 2010; 211:S62. 20. Cortiella J, Niles J, Cantu A, Brettler A, Pham A, Vargas G, et al. Influence of acellular natural lung matri x on murine embryonic stem cell differentiation and tissue formation. Tissue Eng Part A 2010; 16:2565 80. 21. De Kock J, Ceelen L, De Spiegelaere W, Casteleyn C, Claes P, Vanhaecke T, et al. Simple and quick method for whole liver decellularization: a nove l in vitro three dimensional bioengineering tool? Arch Toxicol 2011; 85:607 12. 22. Ott HC, Clippinger B, Conrad C, Schuetz C, Pomerantseva I, Ikonomou L, et al. Regeneration and orthotopic transplantation of a bioartificial lung. Nat Med 2010; 16:927 33. 23. Ott HC, Matthiesen TS, Goh SK, Black LD, Kren SM, Netoff TI, et al. Perfusion decellularized matrix: using nature's platform to engineer a bioartificial heart. Nat Med 2008; 14:213 21. 24. Petersen TH, Calle EA, Zhao L, Lee EJ, Gui L, Raredon MB, et al Tissue engineered lungs for in vivo implantation. Science 2010; 329:538 41. 25. Shupe T, Williams M, Brown A, Willenberg B, Petersen BE. Method for the decellularization of intact rat liver. Organogenesis 2010; 6:134 6. 26. Uygun BE, Soto Gutierrez A, Ya gi H, Izamis ML, Guzzardi MA, Shulman C, et al. Organ reengineering through development of a transplantable recellularized liver graft using decellularized liver matrix. Nat Med 2010; 16:814 20.
165 27. Wainwright JM, Czajka CA, Patel UB, Freytes DO, Tobita K, Gilbert TW, et al. Preparation of cardiac extracellular matrix from an intact porcine heart. Tissue Eng Part C Methods 2010; 16:525 32. 28. Song JJ, Ott HC. Organ engineering based on decellularized matrix scaffolds. Trends Mol Med 2011; 17:424 32. 29. Pl eniceanu O, Harari Steinberg O, Dekel B. Concise review: Kidney stem/progenitor cells: differentiate, sort out, or reprogram? Stem Cells 2010; 28:1649 60. 30. Philp D, Chen SS, Fitzgerald W, Orenstein J, Margolis L, Kleinman HK. Complex extracellular matri ces promote tissue specific stem cell differentiation. Stem Cells 2005; 23:288 96. 31. Keller G. Embryonic stem cell differentiation: emergence of a new era in biology and medicine. Genes Dev 2005; 19:1129 55. 32. Rosner M, Schipany K, Gundacker C, Shanmug asundaram B, Li K, Fuchs C, et al. Renal differentiation of amniotic fluid stem cells: perspectives for clinical application and for studies on specific human genetic diseases. Eur J Clin Invest 2012; 42:677 84. 33. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 2007; 131:861 72. 34. Gupta S, Verfaillie C, Chmielewski D, Kren S, Eidman K, Connaire J, et al. Isolation and characterization of kid ney derived stem cells. J Am Soc Nephrol 2006; 17:3028 40. 35. Kitamura S, Yamasaki Y, Kinomura M, Sugaya T, Sugiyama H, Maeshima Y, et al. Establishment and characterization of renal progenitor like cells from S3 segment of nephron in rat adult kidney. FA SEB J 2005; 19:1789 97. 36. Maeshima A, Sakurai H, Nigam SK. Adult kidney tubular cell population showing phenotypic plasticity, tubulogenic capacity, and integration capability into developing kidney. J Am Soc Nephrol 2006; 17:188 98. 37. Oliver JA, Maaro uf O, Cheema FH, Martens TP, Al Awqati Q. The renal papilla is a niche for adult kidney stem cells. J Clin Invest 2004; 114:795 804. 38. Benigni A, Morigi M, Remuzzi G. Kidney regeneration. Lancet 2010; 375:1310 7. 39. Harari Steinberg O, Pleniceanu O, Dek el B. Selecting the optimal cell for kidney regeneration: fetal, adult or reprogrammed stem cells. Organogenesis 2011; 7:123 34. 40. Hudson TW, Zawko S, Deister C, Lundy S, Hu CY, Lee K, et al. Optimized acellular nerve graft is immunologically tolerated a nd supports regeneration. Tissue Eng 2004; 10:1641 51.
166 41. Rieder E, Nigisch A, Dekan B, Kasimir MT, Muhlbacher F, Wolner E, et al. Granulocyte based immune response against decellularized or glutaraldehyde cross linked vascular tissue. Biomaterials 2006; 27:5634 42. 42. Rieder E, Seebacher G, Kasimir MT, Eichmair E, Winter B, Dekan B, et al. Tissue engineering of heart valves: decellularized porcine and human valve scaffolds differ importantly in residual potential to attract monocytic cells. Circulation 2 005; 111:2792 7. 43. Staerk J, Dawlaty MM, Gao Q, Maetzel D, Hanna J, Sommer CA, et al. Reprogramming of human peripheral blood cells to induced pluripotent stem cells. Cell Stem Cell 2010; 7:20 4. 44. Zhou T, Benda C, Duzinger S, Huang Y, Li X, Li Y, et a l. Generation of induced pluripotent stem cells from urine. J Am Soc Nephrol 2011; 22:1221 8. 45. Sellares J, de Freitas DG, Mengel M, Reeve J, Einecke G, Sis B, et al. Understanding the causes of kidney transplant failure: the dominant role of antibody me diated rejection and nonadherence. Am J Transplant 2012; 12:388 99. 46. Kruegel J, Miosge N. Basement membrane components are key players in specialized extracellular matrices. Cell Mol Life Sci 2010; 67:2879 95. 47. Kizuka Y, Kobayashi K, Kakuda S, Nakaji ma Y, Itoh S, Kawasaki N, et al. Laminin 1 is a novel carrier glycoprotein for the nonsulfated HNK 1 epitope in mouse kidney. Glycobiology 2008; 18:331 8. 48. Muller U, Brandli AW. Cell adhesion molecules and extracellular matrix constituents in kidney dev elopment and disease. J Cell Sci 1999; 112 ( Pt 22):3855 67. 49. Badylak SF. The extracellular matrix as a scaffold for tissue reconstruction. Semin Cell Dev Biol 2002; 13:377 83. 50. Davis GE, Senger DR. Endothelial extracellular matrix: biosynthesis, rem odeling, and functions during vascular morphogenesis and neovessel stabilization. Circ Res 2005; 97:1093 107. 51. Lelongt B, Ronco P. Role of extracellular matrix in kidney development and repair. Pediatr Nephrol 2003; 18:731 42. 52. Kanwar YS, Carone FA, Kumar A, Wada J, Ota K, Wallner EI. Role of extracellular matrix, growth factors and proto oncogenes in metanephric development. Kidney Int 1997; 52:589 606. 53. Steer DL, Nigam SK. Developmental approaches to kidney tissue engineering. Am J Physiol Renal Physiol 2004; 286:F1 7. 54. Fuerer C, Habib SJ, Nusse R. A study on the interactions between heparan sulfate proteoglycans and Wnt proteins. Dev Dyn 2010; 239:184 90.
167 55. Ruoslahti E, Yamaguchi Y. Proteoglycans as modulators of growth factor activities. Ce ll 1991; 64:867 9. 56. Whitelock JM, Iozzo RV. Heparan sulfate: a complex polymer charged with biological activity. Chem Rev 2005; 105:2745 64. 57. Kanwar YS, Wada J, Lin S, Danesh FR, Chugh SS, Yang Q, et al. Update of extracellular matrix, its receptors, and cell adhesion molecules in mammalian nephrogenesis. Am J Physiol Renal Physiol 2004; 286:F202 15. 58. Dressler GR. The cellular basis of kidney development. Annu Rev Cell Dev Biol 2006; 22:509 29. 59. Dressler GR. Advances in early kidney specificatio n, development and patterning. Development 2009; 136:3863 74. 60. Little MH. Renal organogenesis: what can it tell us about renal repair and regeneration? Organogenesis 2011; 7:229 41. 61. Saxen L. Organogenesis of the Kidney. Cambridge: Cambridge Universi ty Press, 1987. 62. Little MH, McMahon AP. Mammalian kidney development: principles, progress, and projections. Cold Spring Harb Perspect Biol 2012; 4. 63. James RG, Kamei CN, Wang Q, Jiang R, Schultheiss TM. Odd skipped related 1 is required for developme nt of the metanephric kidney and regulates formation and differentiation of kidney precursor cells. Development 2006; 133:2995 3004. 64. Fujii T, Pichel JG, Taira M, Toyama R, Dawid IB, Westphal H. Expression patterns of the murine LIM class homeobox gene lim1 in the developing brain and excretory system. Dev Dyn 1994; 199:73 83. 65. Narlis M, Grote D, Gaitan Y, Boualia SK, Bouchard M. Pax2 and pax8 regulate branching morphogenesis and nephron differentiation in the developing kidney. J Am Soc Nephrol 2007; 18:1121 9. 66. Brophy PD, Ostrom L, Lang KM, Dressler GR. Regulation of ureteric bud outgrowth by Pax2 dependent activation of the glial derived neurotrophic factor gene. Development 2001; 128:4747 56. 67. Favor J, Sandulache R, Neuhauser Klaus A, Pretsch W, Chatterjee B, Senft E, et al. The mouse Pax2(1Neu) mutation is identical to a human PAX2 mutation in a family with renal coloboma syndrome and results in developmental defects of the brain, ear, eye, and kidney. Proc Natl Acad Sci U S A 1996; 93:13870 5. 68. Rothenpieler UW, Dressler GR. Pax 2 is required for mesenchyme to epithelium conversion during kidney development. Development 1993; 119:711 20.
168 69. Kreidberg JA, Sariola H, Loring JM, Maeda M, Pelletier J, Housman D, et al. WT 1 is required for ear ly kidney development. Cell 1993; 74:679 91. 70. Ryan G, Steele Perkins V, Morris JF, Rauscher FJ, 3rd, Dressler GR. Repression of Pax 2 by WT1 during normal kidney development. Development 1995; 121:867 75. 71. Sajithlal G, Zou D, Silvius D, Xu PX. Eya 1 acts as a critical regulator for specifying the metanephric mesenchyme. Dev Biol 2005; 284:323 36. 72. Nishinakamura R, Matsumoto Y, Nakao K, Nakamura K, Sato A, Copeland NG, et al. Murine homolog of SALL1 is essential for ureteric bud invasion in kidney d evelopment. Development 2001; 128:3105 15. 73. Imgrund M, Grone E, Grone HJ, Kretzler M, Holzman L, Schlondorff D, et al. Re expression of the developmental gene Pax 2 during experimental acute tubular necrosis in mice 1. Kidney Int 1999; 56:1423 31. 74. S wetha G, Chandra V, Phadnis S, Bhonde R. Glomerular parietal epithelial cells of adult murine kidney undergo EMT to generate cells with traits of renal progenitors. J Cell Mol Med 2011; 15:396 413. 75. Lin SL, Li B, Rao S, Yeo EJ, Hudson TE, Nowlin BT, et al. Macrophage Wnt7b is critical for kidney repair and regeneration. Proc Natl Acad Sci U S A 2010; 107:4194 9. 76. Bussolati B, Bruno S, Grange C, Buttiglieri S, Deregibus MC, Cantino D, et al. Isolation of renal progenitor cells from adult human kidney. Am J Pathol 2005; 166:545 55. 77. Ronconi E, Sagrinati C, Angelotti ML, Lazzeri E, Mazzinghi B, Ballerini L, et al. Regeneration of glomerular podocytes by human renal progenitors. J Am Soc Nephrol 2009; 20:322 32. 78. Ramackers W, Klose J, Winkler M. Xeno kidney transplantation: from idea to reality. Transplant Proc 2012; 44:1773 5. 79. Cascalho M, Ogle BM, Platt JL. Xenotransplantation and the future of renal replacement. J Am Soc Nephrol 2004; 15:1106 12. 80. Basu J, Genheimer CW, Rivera EA, Payne R, Mih alko K, Guthrie K, et al. Functional evaluation of primary renal cell/biomaterial neo kidney augment prototypes for renal tissue engineering. Cell Transplant 2011; 20:1771 90. 81. Oliver JA, Barasch J, Yang J, Herzlinger D, Al Awqati Q. Metanephric mesench yme contains embryonic renal stem cells. Am J Physiol Renal Physiol 2002; 283:F799 809. 82. Li B, Morioka T, Uchiyama M, Oite T. Bone marrow cell infusion ameliorates progressive glomerulosclerosis in an experimental rat model. Kidney Int 2006; 69:323 30.
169 83. Kunter U, Rong S, Boor P, Eitner F, Muller Newen G, Djuric Z, et al. Mesenchymal stem cells prevent progressive experimental renal failure but maldifferentiate into glomerular adipocytes. J Am Soc Nephrol 2007; 18:1754 64. 84. Basu J, Ludlow JW. Develo pmental engineering the kidney: leveraging principles of morphogenesis for renal regeneration. Birth Defects Res C Embryo Today 2012; 96:30 8. 85. de Vries DK, Schaapherder AF, Reinders ME. Mesenchymal stromal cells in renal ischemia/reperfusion injury. Fr ont Immunol 2012; 3:162. 86. Qiao J, Sakurai H, Nigam SK. Branching morphogenesis independent of mesenchymal epithelial contact in the developing kidney. Proc Natl Acad Sci U S A 1999; 96:7330 5. 87. Rosines E, Sampogna RV, Johkura K, Vaughn DA, Choi Y, Sa kurai H, et al. Staged in vitro reconstitution and implantation of engineered rat kidney tissue. Proc Natl Acad Sci U S A 2007; 104:20938 43. 88. Yokoo T, Kawamura T. Xenobiotic kidney organogenesis: a new avenue for renal transplantation. J Nephrol 2009; 22:312 7. 89. Ronco C, Davenport A, Gura V. The future of the artificial kidney: moving towards wearable and miniaturized devices. Nefrologia 2011; 31:9 16. 90. Lanza RP, Chung HY, Yoo JJ, Wettstein PJ, Blackwell C, Borson N, et al. Generation of histocomp atible tissues using nuclear transplantation. Nat Biotechnol 2002; 20:689 96. 91. Humes HD, Fissell WH, Weitzel WF. The bioartificial kidney in the treatment of acute renal failure. Kidney Int Suppl 2002:121 5. 92. Humes HD, Weitzel WF, Bartlett RH, Swanik er FC, Paganini EP, Luderer JR, et al. Initial clinical results of the bioartificial kidney containing human cells in ICU patients with acute renal failure. Kidney Int 2004; 66:1578 88. 93. Fissell WH, Manley S, Westover A, Humes HD, Fleischman AJ, Roy S. Differentiated growth of human renal tubule cells on thin film and nanostructured materials. ASAIO J 2006; 52:221 7. 94. Fissell WH, Roy S. The implantable artificial kidney. Semin Dial 2009; 22:665 70. 95. Jefferson E. FDA announces plans to pilot end sta ge kidney disease technology in new program. Silver Springs (MD): U.S. Food and Drug Administration, 2012. 96. Chan BP, Leong KW. Scaffolding in tissue engineering: general approaches and tissue specific considerations. Eur Spine J 2008; 17 Suppl 4:467 79. 97. Hubbell JA. Materials as morphogenetic guides in tissue engineering. Curr Opin Biotechnol 2003; 14:551 8.
170 98. Allman AJ, McPherson TB, Badylak SF, Merrill LC, Kallakury B, Sheehan C, et al. Xenogeneic extracellular matrix grafts elicit a TH2 restricte d immune response. Transplantation 2001; 71:1631 40. 99. Bhrany AD, Beckstead BL, Lang TC, Farwell DG, Giachelli CM, Ratner BD. Development of an esophagus acellular matrix tissue scaffold. Tissue Eng 2006; 12:319 30. 100. Borschel GH, Dennis RG, Kuzon WM, Jr. Contractile skeletal muscle tissue engineered on an acellular scaffold. Plast Reconstr Surg 2004; 113:595 602; discussion 3 4. 101. Borschel GH, Huang YC, Calve S, Arruda EM, Lynch JB, Dow DE, et al. Tissue engineering of recellularized small diameter vascular grafts. Tissue Eng 2005; 11:778 86. 102. Derham C, Yow H, Ingram J, Fisher J, Ingham E, Korrosis SA, et al. Tissue engineering small diameter vascular grafts: preparation of a biocompatible porcine ureteric scaffold. Tissue Eng Part A 2008; 14:18 71 82. 103. Gui L, Muto A, Chan SA, Breuer CK, Niklason LE. Development of decellularized human umbilical arteries as small diameter vascular grafts. Tissue Eng Part A 2009; 15:2665 76. 104. Raghavan SS, Woon CY, Kraus A, Megerle K, Choi MS, Pridgen BC, et al. Human flexor tendon tissue engineering: decellularization of human flexor tendons reduces immunogenicity in vivo. Tissue Eng Part A 2012; 18:796 805. 105. Wilshaw SP, Kearney JN, Fisher J, Ingham E. Production of an acellular amniotic membrane matrix for use in tissue engineering. Tissue Eng 2006; 12:2117 29. 106. Wilson GJ, Courtman DW, Klement P, Lee JM, Yeger H. Acellular matrix: a biomaterials approach for coronary artery bypass and heart valve replacement. Ann Thorac Surg 1995; 60:S353 8. 107. Yos hida R, Vavken P, Murray MM. Decellularization of bovine anterior cruciate ligament tissues minimizes immunogenic reactions to alpha gal epitopes by human peripheral blood mononuclear cells. Knee 2012; 19:672 5. 108. Crapo PM, Gilbert TW, Badylak SF. An ov erview of tissue and whole organ decellularization processes. Biomaterials 2011; 32:3233 43. 109. Hudson TW, Liu SY, Schmidt CE. Engineering an improved acellular nerve graft via optimized chemical processing. Tissue Eng 2004; 10:1346 58. 110. Seddon AM, C urnow P, Booth PJ. Membrane proteins, lipids and detergents: not just a soap opera. Biochim Biophys Acta 2004; 1666:105 17. 111. Chong PP, Selvaratnam L, Abbas AA, Kamarul T. Human peripheral blood derived mesenchymal stem cells demonstrate similar charact eristics and chondrogenic differentiation potential to bone marrow derived mesenchymal stem cells. J Orthop Res 2012; 30:634 42.
171 112. Yokoo T, Ohashi T, Shen JS, Sakurai K, Miyazaki Y, Utsunomiya Y, et al. Human mesenchymal stem cells in rodent whole embry o culture are reprogrammed to contribute to kidney tissues. Proc Natl Acad Sci U S A 2005; 102:3296 300. 113. Kobayashi A, Valerius MT, Mugford JW, Carroll TJ, Self M, Oliver G, et al. Six2 defines and regulates a multipotent self renewing nephron progenit or population throughout mammalian kidney development. Cell Stem Cell 2008; 3:169 81. 114. Nakayama KH, Batchelder CA, Lee CI, Tarantal AF. Decellularized rhesus monkey kidney as a three dimensional scaffold for renal tissue engineering. Tissue Eng Part A 2010; 16:2207 16. 115. Osafune K. In vitro regeneration of kidney from pluripotent stem cells. Exp Cell Res 2010; 316:2571 7. 116. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981; 292:154 6. 117. Marti n GR. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A 1981; 78:7634 8. 118. Smith AG. Embryo derived stem cells: of mice and men. Annu Rev Cell Dev Biol 20 01; 17:435 62. 119. Yamamoto M, Cui L, Johkura K, Asanuma K, Okouchi Y, Ogiwara N, et al. Branching ducts similar to mesonephric ducts or ureteric buds in teratomas originating from mouse embryonic stem cells. Am J Physiol Renal Physiol 2006; 290:F52 60. 1 20. Bruce SJ, Rea RW, Steptoe AL, Busslinger M, Bertram JF, Perkins AC. In vitro differentiation of murine embryonic stem cells toward a renal lineage. Differentiation 2007; 75:337 49. 121. Kim D, Dressler GR. Nephrogenic factors promote differentiation of mouse embryonic stem cells into renal epithelia. J Am Soc Nephrol 2005; 16:3527 34. 122. Morizane R, Monkawa T, Itoh H. Differentiation of murine embryonic stem and induced pluripotent stem cells to renal lineage in vitro. Biochem Biophys Res Commun 2009; 390:1334 9. 123. Ren X, Zhang J, Gong X, Niu X, Zhang X, Chen P, et al. Differentiation of murine embryonic stem cells toward renal lineages by conditioned medium from ureteric bud cells in vitro. Acta Biochim Biophys Sin (Shanghai) 2010; 42:464 71. 124. Chung Y, Klimanskaya I, Becker S, Li T, Maserati M, Lu SJ, et al. Human embryonic stem cell lines generated without embryo destruction. Cell Stem Cell 2008; 2:113 7.
172 125. Kirkegaard K, Hindkjaer JJ, Ingerslev HJ. Human embryonic development after blastomer e removal: a time lapse analysis. Hum Reprod 2012; 27:97 105. 126. Yu Y, Wu J, Fan Y, Lv Z, Guo X, Zhao C, et al. Evaluation of blastomere biopsy using a mouse model indicates the potential high risk of neurodegenerative disorders in the offspring. Mol Cel l Proteomics 2009; 8:1490 500. 127. In Stem Cell Information [World Wide Web Site] Bethesda (MD): National Institutes of Health, U.S. Department of Health and Human Services, 2010 [cited July 25, 2013] Available at http://stemcells.nih.gov/research/registry/Pages/Default.aspx 128. Enzensberger C, Pulvermacher C, Degenhardt J, Kawacki A, Germer U, Gembruch U, et al. Fetal loss rate and associated risk factors after amniocentesis chorionic villus sampling and fetal blood sampling. Ultraschall Med 2012; 33:E75 9. 129. De Coppi P, Bartsch G, Jr., Siddiqui MM, Xu T, Santos CC, Perin L, et al. Isolation of amniotic stem cell lines with potential for therapy. Nat Biotechnol 2007; 25:1 00 6. 130. Perin L, Giuliani S, Jin D, Sedrakyan S, Carraro G, Habibian R, et al. Renal differentiation of amniotic fluid stem cells. Cell Prolif 2007; 40:936 48. 131. Hendry CE, Little MH. Reprogramming the kidney: a novel approach for regeneration. Kidne y Int 2012; 82:138 46. 132. Taura D, Noguchi M, Sone M, Hosoda K, Mori E, Okada Y, et al. Adipogenic differentiation of human induced pluripotent stem cells: comparison with that of human embryonic stem cells. FEBS Lett 2009; 583:1029 33. 133. Muchkaeva IA Dashinimaev EB, Terskikh VV, Sukhanov YV, Vasiliev AV. Molecular mechanisms of induced pluripotency. Acta Naturae 2012; 4:12 22. 134. Cai J, Li W, Su H, Qin D, Yang J, Zhu F, et al. Generation of human induced pluripotent stem cells from umbilical cord m atrix and amniotic membrane mesenchymal cells. J Biol Chem 2010; 285:11227 34. 135. Giorgetti A, Montserrat N, Rodriguez Piza I, Azqueta C, Veiga A, Izpisua Belmonte JC. Generation of induced pluripotent stem cells from human cord blood cells with only two factors: Oct4 and Sox2. Nat Protoc 2010; 5:811 20. 136. Sancho Bru P, Roelandt P, Narain N, Pauwelyn K, Notelaers T, Shimizu T, et al. Directed differentiation of murine induced pluripotent stem cells to functional hepatocyte like cells. J Hepatol 2011; 5 4:98 107. 137. Guha P, Morgan JW, Mostoslavsky G, Rodrigues NP, Boyd AS. Lack of immune response to differentiated cells derived from syngeneic induced pluripotent stem cells. Cell Stem Cell 2013; 12:407 12.
173 138. Stark K, Vainio S, Vassileva G, McMahon AP. Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt 4. Nature 1994; 372:679 83. 139. Kim K, Doi A, Wen B, Ng K, Zhao R, Cahan P, et al. Epigenetic memory in induced pluripotent stem cells. Nature 2010; 467:285 90. 140. Koh CJ, Atala A. Tissue engineering, stem cells, and cloning: opportunities for regenerative medicine. J Am Soc Nephrol 2004; 15:1113 25. 141. Brendel K, Meezan E, Nagle RB. The acellular perfused kidney: a model for basement membrane permeability. I n: Kefalides NA, ed. Biology and chemistry of basement membranes. New York: Academic Press, 1979:101 17. 142. Brendel K, Meezan E. Vascular basement membranes: preparation and properties of material isolated with the use of detergents. Adv Exp Med Biol 198 0; 131:89 103. 143. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia reperfusion injury in the isolated, perfused rat kidney. Am J Physiol Renal Physiol 2005; 288:F778 84. 144. Sadri F, Loebel NG, Bhagwandin BD, Karuna K aran A. A fully automated organ perfusion workstation: the LifeSustainer 1000. Transplant Proc 1997; 29:3588 90. 145. Brown BN, Valentin JE, Stewart Akers AM, McCabe GP, Badylak SF. Macrophage phenotype and remodeling outcomes in response to biologic scaff olds with and without a cellular component. Biomaterials 2009; 30:1482 91. 146. Nagata S, Hanayama R, Kawane K. Autoimmunity and the clearance of dead cells. Cell 2010; 140:619 30. 147. Martin ND, Schaner PJ, Tulenko TN, Shapiro IM, Dimatteo CA, Williams T K, et al. In vivo behavior of decellularized vein allograft. J Surg Res 2005; 129:17 23. 148. Barnes CA, Brison J, Michel R, Brown BN, Castner DG, Badylak SF, et al. The surface molecular functionality of decellularized extracellular matrices. Biomaterials 2011; 32:137 43. 149. Reing JE, Brown BN, Daly KA, Freund JM, Gilbert TW, Hsiong SX, et al. The effects of processing methods upon mechanical and biologic properties of porcine dermal extracellular matrix scaffolds. Biomaterials 2010; 31:8626 33. 150. Li HY, Liao CY, Lee KH, Chang HC, Chen YJ, Chao KC, et al. Collagen IV significantly enhances migration and transplantation of embryonic stem cells: involvement of alpha2beta1 integrin mediated actin remodeling. Cell Transplant 2011; 20:893 907. 151. Cai Q, D mitrieva NI, Ferraris JD, Brooks HL, van Balkom BW, Burg M. Pax2 expression occurs in renal medullary epithelial cells in vivo and in cell culture, is osmoregulated, and promotes osmotic tolerance. Proc Natl Acad Sci U S A 2005; 102:503 8.
174 152. Mundlos S, Pelletier J, Darveau A, Bachmann M, Winterpacht A, Zabel B. Nuclear localization of the protein encoded by the Wilms' tumor gene WT1 in embryonic and adult tissues. Development 1993; 119:1329 41. 153. Skinnider BF, Folpe AL, Hennigar RA, Lim SD, Cohen C, T amboli P, et al. Distribution of cytokeratins and vimentin in adult renal neoplasms and normal renal tissue: potential utility of a cytokeratin antibody panel in the differential diagnosis of renal tumors. Am J Surg Pathol 2005; 29:747 54. 154. Thedieck C, Kuczyk M, Klingel K, Steiert I, Muller CA, Klein G. Expression of Ksp cadherin during kidney development and in renal cell carcinoma. Br J Cancer 2005; 92:2010 7. 155. Thomson RB, Aronson PS. Immunolocalization of Ksp cadherin in the adult and developing rabbit kidney. Am J Physiol 1999; 277:F146 56. 156. Allman AJ, McPherson TB, Merrill LC, Badylak SF, Metzger DW. The Th2 restricted immune response to xenogeneic small intestinal submucosa does not influence systemic protective immunity to viral and bacter ial pathogens. Tissue Eng 2002; 8:53 62. 157. Engvall E, Perlmann P. Enzyme linked immunosorbent assay (ELISA). Quantitative assay of immunoglobulin G. Immunochemistry 1971; 8:871 4. 158. Yabuki A, Mitani S, Mizukami K, Yamato O. Nephron segment identifica tion in the normal canine kidney by using lectin histochemistry. Res Vet Sci 2012; 93:560 4. 159. Baharvand H, Azarnia M, Parivar K, Ashtiani SK. The effect of extracellular matrix on embryonic stem cell derived cardiomyocytes. J Mol Cell Cardiol 2005; 38: 495 503. 160. Chen SS, Fitzgerald W, Zimmerberg J, Kleinman HK, Margolis L. Cell cell and cell extracellular matrix interactions regulate embryonic stem cell differentiation. Stem Cells 2007; 25:553 61. 161. Czyz J, Wobus A. Embryonic stem cell differentia tion: the role of extracellular factors. Differentiation 2001; 68:167 74. 162. Little MH. Regrow or repair: potential regenerative therapies for the kidney. J Am Soc Nephrol 2006; 17:2390 401. 163. Taqvi S, Roy K. Influence of scaffold physical properties and stromal cell coculture on hematopoietic differentiation of mouse embryonic stem cells. Biomaterials 2006; 27:6024 31. 164. Wicha MS, Lowrie G, Kohn E, Bagavandoss P, Mahn T. Extracellular matrix promotes mammary epithelial growth and differentiation in vitro. Proc Natl Acad Sci U S A 1982; 79:3213 7.
175 165. Mima A, Ohshiro Y, Kitada M, Matsumoto M, Geraldes P, Li C, et al. Glomerular specific protein kinase C beta induced insulin receptor substrate 1 dysfunction and insulin resistance in rat models of dia betes and obesity. Kidney Int 2011; 79:883 96. 166. Walker RA. Quantification of immunohistochemistry -issues concerning methods, utility and semiquantitative assessment I. Histopathology 2006; 49:406 10. 167. Bullock SL, Fletcher JM, Beddington RS, Wilson VA. Renal agenesis in mice homozygous for a gene trap mutation in the gene encoding heparan sulfate 2 sulfotransferase. Genes Dev 1998; 12:1894 906. 168. Goldberg S, Harvey SJ, Cunningham J, Tryggvason K, Miner JH. Glomerular filtration is normal in the a bsence of both agrin and perlecan heparan sulfate from the glomerular basement membrane. Nephrol Dial Transplant 2009; 24:2044 51. 169. Kinnunen AI, Sormunen R, Elamaa H, Seppinen L, Miller RT, Ninomiya Y, et al. Lack of collagen XVIII long isoforms affect s kidney podocytes, whereas the short form is needed in the proximal tubular basement membrane. J Biol Chem 2011; 286:7755 64. 170. Farndale RW, Buttle DJ, Barrett AJ. Improved quantitation and discrimination of sulphated glycosaminoglycans by use of dimet hylmethylene blue. Biochim Biophys Acta 1986; 883:173 7. 171. Lai PH, Chang Y, Chen SC, Wang CC, Liang HC, Chang WC, et al. Acellular biological tissues containing inherent glycosaminoglycans for loading basic fibroblast growth factor promote angiogenesis and tissue regeneration. Tissue Eng 2006; 12:2499 508. 172. Tufro A, Norwood VF, Carey RM, Gomez RA. Vascular endothelial growth factor induces nephrogenesis and vasculogenesis. J Am Soc Nephrol 1999; 10:2125 34. 173. Furness PN. Extracellular matrix and t he kidney. J Clin Pathol 1996; 49:355 9. 174. Bates CM. Role of fibroblast growth factor receptor signaling in kidney development. Pediatr Nephrol 2007; 22:343 9. 175. Strutz F, Zeisberg M, Hemmerlein B, Sattler B, Hummel K, Becker V, et al. Basic fibrobla st growth factor expression is increased in human renal fibrogenesis and may mediate autocrine fibroblast proliferation. Kidney Int 2000; 57:1521 38. 176. Gregory KE, Ono RN, Charbonneau NL, Kuo CL, Keene DR, Bachinger HP, et al. The prodomain of BMP 7 tar gets the BMP 7 complex to the extracellular matrix. J Biol Chem 2005; 280:27970 80. 177. Zeisberg M. Bone morphogenic protein 7 and the kidney: current concepts and open questions. Nephrol Dial Transplant 2006; 21:568 73.
176 178. Terada Y, Tanaka H, Okado T, Shimamura H, Inoshita S, Kuwahara M, et al. Expression and function of the developmental gene Wnt 4 during experimental acute renal failure in rats. J Am Soc Nephrol 2003; 14:1223 33. 179. Kobayashi T, Tanaka H, Kuwana H, Inoshita S, Teraoka H, Sasaki S, e t al. Wnt4 transformed mouse embryonic stem cells differentiate into renal tubular cells. Biochem Biophys Res Commun 2005; 336:585 95. 180. Jones DB. Nephrotic glomerulonephritis. Am J Pathol 1957; 33:313 29. 181. Quint C, Kondo Y, Manson RJ, Lawson JH, Da rdik A, Niklason LE. Decellularized tissue engineered blood vessel as an arterial conduit. Proc Natl Acad Sci U S A 2011; 108:9214 9. 182. Sarathchandra P, Smolenski RT, Yuen AH, Chester AH, Goldstein S, Heacox AE, et al. Impact of gamma irradiation on ext racellular matrix of porcine pulmonary valves. J Surg Res 2012; 176:376 85. 183. Bell RM, Mocanu MM, Yellon DM. Retrograde heart perfusion: the Langendorff technique of isolated heart perfusion. J Mol Cell Cardiol 2011; 50:940 50. 184. Grauss RW, Hazekamp MG, Oppenhuizen F, van Munsteren CJ, Gittenberger de Groot AC, DeRuiter MC. Histological evaluation of decellularised porcine aortic valves: matrix changes due to different decellularisation methods. Eur J Cardiothorac Surg 2005; 27:566 71. 185. Sondell M, Lundborg G, Kanje M. Regeneration of the rat sciatic nerve into allografts made acellular through chemical extraction. Brain Res 1998; 795:44 54. 186. Lichtenberg A, Tudorache I, Cebotari S, Ringes Lichtenberg S, Sturz G, Hoeffler K, et al. In vitro re en dothelialization of detergent decellularized heart valves under simulated physiological dynamic conditions. Biomaterials 2006; 27:4221 9. 187. Schaner PJ, Martin ND, Tulenko TN, Shapiro IM, Tarola NA, Leichter RF, et al. Decellularized vein as a potential scaffold for vascular tissue engineering. J Vasc Surg 2004; 40:146 53. 188. Burra P, Tomat S, Conconi MT, Macchi C, Russo FP, Parnigotto PP, et al. Acellular liver matrix improves the survival and functions of isolated rat hepatocytes cultured in vitro. In t J Mol Med 2004; 14:511 5. 189. Deeken CR, White AK, Bachman SL, Ramshaw BJ, Cleveland DS, Loy TS, et al. Method of preparing a decellularized porcine tendon using tributyl phosphate. J Biomed Mater Res B Appl Biomater 2011; 96:199 206. 190. Thrailkill KM Moreau CS, Cockrell G, Simpson P, Goel R, North P, et al. Physiological matrix metalloproteinase concentrations in serum during childhood and adolescence, using Luminex Multiplex technology. Clin Chem Lab Med 2005; 43:1392 9.
177 191. Shimizu A, Kawai K, Yan agino M, Wakiyama T, Machida M, Kameyama K, et al. Characteristics of type IV collagen unfolding under various pH conditions as a model of pathological disorder in tissue. J Biochem 2007; 142:33 40. 192. Negishi J, Funamoto S, Kimura T, Nam K, Higami T, Ki shida A. Effect of treatment temperature on collagen structures of the decellularized carotid artery using high hydrostatic pressure. J Artif Organs 2011; 14:223 31. 193. Lin P, Chan WC, Badylak SF, Bhatia SN. Assessing porcine liver derived biomatrix for hepatic tissue engineering. Tissue Eng 2004; 10:1046 53. 194. Hadjantonakis AK, Gertsenstein M, Ikawa M, Okabe M, Nagy A. Generating green fluorescent mice by germline transmission of green fluorescent ES cells. Mech Dev 1998; 76:79 90. 195. Kramer J, Stei nhoff J, Klinger M, Fricke L, Rohwedel J. Cells differentiated from mouse embryonic stem cells via embryoid bodies express renal marker molecules. Differentiation 2006; 74:91 104. 196. Harding SD, Armit C, Armstrong J, Brennan J, Cheng Y, Haggarty B, et al The GUDMAP database -an online resource for genitourinary research. Development 2011; 138:2845 53. 197. McMahon AP, Aronow BJ, Davidson DR, Davies JA, Gaido KW, Grimmond S, et al. GUDMAP: the genitourinary developmental molecular anatomy project. J Am So c Nephrol 2008; 19:667 71. 198. Chung S, King MW. Design concepts and strategies for tissue engineering scaffolds. Biotechnol Appl Biochem 2011; 58:423 38. 199. Yang D, Guo T, Nie C, Morris SF. Tissue engineered blood vessel graft produced by self derived cells and allogenic acellular matrix: a functional performance and histologic study. Ann Plast Surg 2009; 62:297 303. 200. Weinbaum S, Duan Y, Satlin LM, Wang T, Weinstein AM. Mechanotransduction in the renal tubule. Am J Physiol Renal Physiol 2010; 299:F1 220 36. 201. Frohlich EM, Zhang X, Charest JL. The use of controlled surface topography and flow induced shear stress to influence renal epithelial cell function. Integr Biol (Camb) 2012; 4:75 83. 202. Yamamoto K, Sokabe T, Watabe T, Miyazono K, Yamashita JK, Obi S, et al. Fluid shear stress induces differentiation of Flk 1 positive embryonic stem cells into vascular endothelial cells in vitro. Am J Physiol Heart Circ Physiol 2005; 288:H1915 24. 203. Struijk PC, Mathews VJ, Loupas T, Stewart PA, Clark EB, S teegers EA, et al. Blood pressure estimation in the human fetal descending aorta. Ultrasound Obstet Gynecol 2008; 32:673 81.
178 204. Gown AM, Willingham MC. Improved detection of apoptotic cells in archival paraffin sections: immunohistochemistry using antibo dies to cleaved caspase 3. J Histochem Cytochem 2002; 50:449 54. 205. Scholzen T, Gerdes J. The Ki 67 protein: from the known and the unknown. J Cell Physiol 2000; 182:311 22. 206. Carev D, Saraga M, Saraga Babic M. Expression of intermediate filaments, EG F and TGF alpha in early human kidney development. J Mol Histol 2008; 39:227 35. 207. Bratt Leal AM, Carpenedo RL, McDevitt TC. Engineering the embryoid body microenvironment to direct embryonic stem cell differentiation. Biotechnol Prog 2009; 25:43 51. 20 8. Battista S, Guarnieri D, Borselli C, Zeppetelli S, Borzacchiello A, Mayol L, et al. The effect of matrix composition of 3D constructs on embryonic stem cell differentiation. Biomaterials 2005; 26:6194 207. 209. Liu H, Roy K. Biomimetic three dimensional cultures significantly increase hematopoietic differentiation efficacy of embryonic stem cells. Tissue Eng 2005; 11:319 30. 210. Brunskill EW, Aronow BJ, Georgas K, Rumballe B, Valerius MT, Aronow J, et al. Atlas of gene expression in the developing kidne y at microanatomic resolution. Dev Cell 2008; 15:781 91. 211. Kahraman K, Kiremitci S, Taskin S, Kankaya D, Sertcelik A, Ortac F. Expression pattern of PAX2 in hyperplastic and malignant endometrium. Arch Gynecol Obstet 2012; 286:173 8. 212. Badve S. Equiv alency of RT PCR and immunohistochemistry: fact or factoid. Breast Cancer Res Treat 2009; 116:145 7. 213. Ross EA, Abrahamson DR, St John P, Clapp WL, Williams MJ, Terada N, et al. Mouse stem cells seeded into decellularized rat kidney scaffolds endothelia lize and remodel basement membranes. Organogenesis 2012; 8:49 55. 214. Coffin JD, Harrison J, Schwartz S, Heimark R. Angioblast differentiation and morphogenesis of the vascular endothelium in the mouse embryo. Dev Biol 1991; 148:51 62. 215. Abrahamson DR. Development of kidney glomerular endothelial cells and their role in basement membrane assembly. Organogenesis 2009; 5:275 87. 216. Nourse MB, Halpin DE, Scatena M, Mortisen DJ, Tulloch NL, Hauch KD, et al. VEGF induces differentiation of functional endot helium from human embryonic stem cells: implications for tissue engineering. Arterioscler Thromb Vasc Biol 2010; 30:80 9.
179 217. Lin ZY, Chuang WL, Chuang YH. Amphotericin B up regulates angiogenic genes in hepatocellular carcinoma cell lines. Eur J Clin Inv est 2009; 39:239 45. 218. Giuliani S, Perin L, Sedrakyan S, Kokorowski P, Jin D, De Filippo R. Ex vivo whole embryonic kidney culture: a novel method for research in development, regeneration and transplantation. J Urol 2008; 179:365 70. 219. Vigneau C, Po lgar K, Striker G, Elliott J, Hyink D, Weber O, et al. Mouse embryonic stem cell derived embryoid bodies generate progenitors that integrate long term into renal proximal tubules in vivo. J Am Soc Nephrol 2007; 18:1709 20. 220. Kaushal S, Amiel GE, Guleser ian KJ, Shapira OM, Perry T, Sutherland FW, et al. Functional small diameter neovessels created using endothelial progenitor cells expanded ex vivo. Nat Med 2001; 7:1035 40. 221. Wilhelmi M, Giere B, Harder M. Interaction of cells with decellularized biolo gical materials. Adv Biochem Eng Biotechnol 2012; 126:105 16. 222. Badylak SF, Taylor D, Uygun K. Whole organ tissue engineering: decellularization and recellularization of three dimensional matrix scaffolds. Annu Rev Biomed Eng 2011; 13:27 53. 223. Singh AM, Hamazaki T, Hankowski KE, Terada N. A heterogeneous expression pattern for Nanog in embryonic stem cells. Stem Cells 2007; 25:2534 42.
180 BIOGRAPHICAL SKETC H Matthew James Williams was raised in Michigan and graduated from Holland High School in 1993. He received a Bache lor of Science and Engineering in Mechanical Engineering from the University of Michigan in 1998. During his undergraduate years he was employed in car parts, furniture, and HVAC systems manufacturing. In his senior year, through a coope rative engineering program, he spent two semesters in automotive climate control test engineering at the DENSO International America Tech Center (Southfield, MI). He was also a member of the marching band and band service fraternity Kappa Kappa Psi. After graduation, he was hired as a project management consultant at ACM, Inc. (Ann Arbor, MI), and spent two years advising Ford Motor Company in program timing and resource management for EPA certification. In 2000, in order to make a direct impact on healthca re, the author relocated to study Biomedical Engineering for work in medical image processing under the guidance of Dr. Frank Bova. He programmed routines to impro ve segmentation of CT based spine models for image guided surgery. Afterward, he transitioned to doctoral studies with a tissue engineering focus. Mentored by Dr. Chris Batich, his research tested the response of mouse embryonic stem cells to kidney extrac ellular matrix scaffolds for the purpose of organ generation. In 2009, he received an internship at RTI Biologics (Alachua, FL) for research and development of orthopedic implants. In 2011, the author spent a year as laboratory supervisor for Dr. Ed Botchw ey at the University of Virginia. The Botchwey Lab pursued projects in regenerative medicine with focus on vascular remodeling, immune regulation, and endogenous stem cell recruitment. In 2013, the author completed his studies and was awarded a Ph.D. in Bi omedical Engineering from the University of Florida. Matthew currently resides in St. Louis, MO, with his wife Jennifer, and two children Connor and Caden.