Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering
Pankaj Godara
A Thesis Submitted for the Degree of Doctor of Philosophy
at the
University of New South Wales
Faculty of Engineering
Graduate School of Biomedical Engineering
2010 Originality Statement
‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substan- tial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowl- edgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.’
Signed ......
Date ...... Copyright Statement
‘I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University li- braries in all forms of media, now or hereafter known, subject to the provisions of the
Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the abstract of my thesis in Dissertations Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted
I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.’
Signed ......
Date ...... Authenticity Statement
‘I certify that the Library deposit digital copy is a direct equivalent of the final offi- cially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’
Signed ......
Date ...... Acknowledgements
First and foremost I would like to thank my family, my mother Saroj, my father Lal, and my brother Vikas for their constant encouragement, endless love, and unwavering support. Thank you for always believing in me and for making me the person I am today. I survived and the sky did not fall.
This body of work would not have been possible without kind assistance from my supervisors.
Thank you to Bruce Milthorpe and the Graduate School of Biomedical Engineering for giving me the opportunity to undertake this project, and to Clive McFarland for contributions.
I would like to express my most sincere heartfelt gratitude to Robert Nordon for his patient supervision and guidance in seeing this project through to the end. Thank you for your help in the preparation of this manuscript, for taking such an inspiring interest in my work, and for teaching me so much. What an experience.
John Whitelock is thanked for advice and friendship, especially during the latter stages of this project. Research students Joost vab den Berg and Carl Gabel are acknowledged for assistance with experimental work, and Andrew Sims is thanked for help with all things Latex, Matlab and CAD.
NIH 3T3 cells were kindly provided by iNano, University of Aarhus, Aarhus, Denmark.
MS-5 cells were kindly donated by Alla Dolnikov of the Children’s Cancer Institute
Australia, and equine bone marrow was kindly donated by Chris O’Sullivan of the
Randwick Equine Centre. To all the GSBmE L4 technical staff, thank you for your advice and guidance.
To all the past and present research students at GSBmE, thank you for enriching this journey.
To my dear friends Gina Katsiolis, Richa Gupta and Thania Kearns - Thank you for always lending an ear to listen; for your patience and kindness; for being the voice of reason when there was none; and for reminding me of life beyond this thesis.
A clay pot sitting in the sun will always be a clay pot.
It has to go through the white heat of the furnace to become porcelain.
Mildred W. Struven Publications and presentations arising from this thesis
Journal Papers
1. Godara, P., Nordon, R.E., McFarland, C.D. Mesenchymal stem cells in tissue en-
gineering. Journal of Chemical Technology & Biotechnology, Volume 83, Number
4, April 2008 , pp. 397-407(11 pages)
2. Godara, P., McFarland, C.D., Nordon, R.E. Design of bioreactors for mesenchy-
mal stem cell tissue engineering. Journal of Chemical Technology & Biotechnol-
ogy, Volume 83, Number 4, April 2008 , pp. 408-420(13 pages)
Conference Papers
1. Nordon, R. E., The, O., Godara, P., You, S., Van Den Berg, J. and Rosengarten,
G. Membrane Culture System for Manufacture of Cells for Transplantation: From
Lab to Clinic. 19th Annual Conference of Australasian Society for Biomaterials
and Tissue Engineering, 21-23 January 2009, Sydney, Australia.
2. Godara, P., Ko, K., You, S., Soon, L., Foong, F., Rosengarten, G., Nordon, R. In
Vitro Methods for Investigation of Cell-Matrix Interactions. Matrix Biology So-
ciety of Australia and New Zealand Annual Meeting, 13-16 October 2008, Mantra
Ettalong Beach, Australia. 3. Godara, P., Nordon, R.E., Ko, KH., Milthorpe, B.K. and McFarland, C.D. Iden-
tification of Adult Stem Cells for Tissue Engineering Applications. ISAC Samuel
A. Latt Meeting: Stem Cells in the Age of Fluorescence Technology, Australian
Stem Cell Centre Conference, 6-9 November 2005, Gold Coast, Australia.
4. Godara, P., Nordon, R.E., Ko, KH., Milthorpe, B.K. and McFarland, C.D. Iden-
tification of Adult Stem Cells for Tissue Engineering Applications. Stem Cell and
Tissue Engineering Common Interest Group Mini Symposium, August 18 2005,
Sydney Australia.
5. Godara, P., Nordon, R.E., Ko, KH., Milthorpe, B.K. and McFarland, C.D. Iden-
tification of Adult Stem Cells for Tissue Engineering Applications. Sydney Tissue
Engineering and Matrix Group IXth Symposium, May 12 2005, Sydney, Australia.
6. Godara, P., Nordon, R.E., Ko, KH., Milthorpe, B.K. and McFarland, C.D. Iden-
tification of Adult Stem Cells for Tissue Engineering Applications. Australian
Society for Biomaterials Conference, March 31-April 2 2005, Victor Harbour,
Australia. Abstract
The ex vivo manufacture of functional organs and tissues for implantation can impact on therapeutic needs arising from an ageing population. The rapid expansion of the field of tissue engineering has arisen in response to this clinical need. In order for mesenchymal stem cells (MSC) to be useful clinically, sufficient numbers must be obtained. It was hypothesised that MSC could be isolated from a heterogeneous population of cells, and that hollow fibre bioreactor systems may be a scalable technology for expansion.
In order to manufacture functional tissues, appropriate bioreactor devices must be developed to direct cellular differentiation. It was hypothesised that long-term ex vivo tissue gradients could be established.
A substrate (Aldefluor) for aldehyde dehydrogenase (ALDH) activity was employed to label cells from human umbilical cord blood (hUCB) and cells isolated from rat, murine, porcine and equine origins. Growth of anchorage dependent cells was carried out in cuprophan hollow fibres coated with a novel recombinant protein, with and without extracapillary co-culture. Micro bioreactors were manufactured using the techniques of lithography from polydimethylsiloxane, and device biocompatibility was assessed using the anchorage dependent cell line NIH 3T3.
While cells isolated from rat, murine, porcine and equine origins showed an ALDH bright population, MSC could not be purified from hUCB. Cellular attachment to cel- lulose based substrates coated with the recombinant protein occurred within 2 hours.
Anchorage dependent cells could not be maintained in the hollow fibres. Stable con- centration gradient profiles were generated experimentally in two micro-devices with differing geometries. In the second device, the concentration gradient was maximal in the region of flow stasis, and cells were found to remain viable inside the cell chamber belowaflowrateof4μL/min for 72 hours.
Aldefluor substrate was not useful in the prospective isolation of MSC from a heteroge- neous population. Coating of cellulose substrates with a novel recombinant protein was found to be necessary for cell attachment. Growth in the hollow fibres was found to be suboptimal compared to conventional methods. Cell chemotaxis and tissue morphogen- esis could be studied using the second micro-device developed without the confounding effect of fluid shear stress. Table of Contents
Abbreviations and Symbols x
List of Tables xiii
List of Figures xiv
1 Introduction 1
1.1 Research Motives ...... 1
1.2ThesisAims...... 3
1.3ThesisLayout...... 4
2 Literature Review 5
2.1Introduction...... 5
2.2TissueEngineering...... 5
2.3StemCells...... 8
2.3.1 Definitionsandterminology...... 8
2.3.2 BiologyofMSC...... 10
2.3.3 MSCmarkers...... 15
2.4MSCinTissueEngineering...... 17
2.4.1 Bone...... 18
2.4.2 Cartilage...... 18
i Table of Contents Table of Contents
2.4.3 OsteochondralJunction...... 19
2.5Bioreactors...... 19
2.6MSCMetabolismandPhysiologicalRequirements...... 20
2.6.1 Growthfactorsandhormones...... 20
2.6.2 Oxidative microenvironment ...... 22
2.6.3 Strainandshearstress...... 23
2.6.4 Oxygen and glucose uptake ...... 23
2.7MassTransport...... 24
2.7.1 Theroleofdiffusionandconvection...... 24
2.7.2 Selectiveexchangeusingmembranesystems...... 26
2.8BioreactorDesigns...... 27
2.8.1 Micro-carriersandscaffolds...... 29
2.8.2 Mixed vessels ...... 30
2.8.3 Perfusionbioreactors...... 31
3 Cell identification and isolation 37
3.1Introduction...... 37
3.1.1 IsolationTechniques...... 38
3.1.2 CellMarkers...... 39
3.1.3 ApproachandAims...... 41
3.2MaterialsandMethods...... 43
3.2.1 Materials...... 43
3.2.2 PurificationofhUCB...... 43
3.2.3 Purificationofporcinebonemarrow...... 44
3.2.4 Purificationofmousebonemarrow...... 45
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering ii Table of Contents Table of Contents
3.2.5 Aldefluorlabeling...... 46
3.2.6 Cordbloodfibroblastcolonyformingassay...... 46
3.2.7 Ratbonemarrowfibroblastcolonyformingassay...... 47
3.2.8 Cellantibodylabeling...... 47
3.2.9 AnalysisandSorting...... 47
3.3Results...... 48
3.3.1 IsolationofHSCandMSCfromcordblood...... 48
3.3.2 IsolationofMSCfromRat...... 62
3.3.3 IsolationofMSCbyplasticadherencefromotherspecies.... 67
3.3.4 ALDH staining profile of marrow and MSC from other species . 70
3.4Discussion...... 82
3.4.1 MSCincordblood...... 83
3.4.2 ALDHvariationwithpassage...... 84
3.4.3 ALDHvariationwithspecies...... 85
3.5Conclusion...... 86
4 Hollow fibre expansion of anchorage dependent cells 87
4.1Introduction...... 87
4.1.1 MSCexpansion...... 87
4.1.2 Cell surface receptors - Integrin binding ...... 89
4.1.3 ApproachandAims...... 89
4.2MaterialsandMethods...... 91
4.2.1 Materials...... 91
4.2.2 Methods...... 92
4.3Results...... 95
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering iii Table of Contents Table of Contents
4.3.1 AttachmentEffectsofCBDR...... 95
4.3.2 In situ cellstaining...... 99
4.3.3 CellcultureinHFModules...... 106
4.4Discussion...... 114
4.4.1 In situ cellstaining...... 115
4.4.2 CellcultureinHFmodules...... 117
4.5Conclusion...... 120
5 Multi-channel bioreactor for engineering composite tissues from MSC 121
5.1Introduction...... 121
5.1.1 ClinicalneedforMSCdifferentiatedproducts...... 121
5.1.2 Engineeringofosteochondralgrafts...... 124
5.1.3 Bioreactorstogeneratetissuegradients...... 125
5.1.4 ApproachandAims...... 126
5.2Materials...... 127
5.3Cross-flowdiffusionchamber...... 128
5.3.1 Introduction...... 128
5.3.2 Construction and assembly ...... 131
5.3.3 Evaluationofdesign...... 132
5.3.4 ResultsandDiscussion...... 134
5.4Hollowfibrediffusionchamber...... 136
5.4.1 Introduction...... 136
5.4.2 Construction and assembly ...... 137
5.4.3 GelRemoval...... 138
5.4.4 Results...... 139
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering iv Table of Contents Table of Contents
5.4.5 Discussion...... 141
5.5Conclusion...... 142
6 Microfluidic Bioreactors 143
6.1Introduction...... 143
6.1.1 MicrofluidicAdvantages...... 143
6.1.2 MicroscaleEffects...... 145
6.1.3 MicrofluidicGradientGeneration...... 149
6.1.4 ApproachandAims...... 150
6.2MaterialsandMethods...... 151
6.2.1 MaterialsandEquipment...... 151
6.2.2 Methods...... 151
6.3PyramidalGradientGeneratingDevice...... 162
6.3.1 Descriptionofdevice...... 162
6.3.2 Results...... 164
6.3.3 Discussion...... 172
6.4CounterFlowDiffusionChamber...... 174
6.4.1 DescriptionofDevice...... 174
6.4.2 Results...... 175
6.4.3 Discussion...... 183
6.5Conclusion...... 184
7 Biocompatibility and perfusion culture using the counter flow diffusion chamber 185
7.1Introduction...... 185
7.1.1 Perfusionsystems...... 188
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering v Table of Contents Table of Contents
7.1.2 ApproachandAims...... 188
7.2MethodsandMaterials...... 190
7.2.1 Materials...... 190
7.2.2 Devicepreparationforcellularuse...... 190
7.2.3 CellattachmenttoPETmembrane...... 191
7.2.4 Hoechst 33342 staining ...... 191
7.2.5 PluroniccoatingofPETmembrane...... 192
7.2.6 Directcontactcytotoxicityassay...... 192
7.2.7 Cellgrowthinhibitionassay...... 193
7.2.8 Cytotoxicityofassembledsystem...... 193
7.2.9 Cytotoxicityoftubing...... 194
7.2.10Cytotoxicityofsyringes...... 195
7.3Results...... 197
7.3.1 CellattachmenttoPETmembrane...... 197
7.3.2 Hoechst 33342 staining ...... 197
7.3.3 Inoculationdensity...... 198
7.3.4 Biocompatibility and material surface interaction assessment . . 206
7.3.5 Cell growth inside counter flow diffusion chamber using Tygon tubing...... 218
7.4Discussion...... 221
7.4.1 SystemOptimisation...... 221
7.4.2 Biocompatibilty...... 221
7.5Conclusion...... 225
8 Summary and Recommendations 226
8.1Cellidentificationandisolation...... 226
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering vi Table of Contents Table of Contents
8.2 Hollow fibre expansion of anchoragedependentcells...... 227
8.3 Multi-channel bioreactor for engineering composite tissuesfromMSC...... 228
8.4MicrofluidicBioreactors...... 229
8.5 Biocompatibility and perfusion culture using the counterflowdiffusionchamber...... 231
References 233
Appendix 262
A Materials and General Methods 262
A.1Materials...... 262
A.1.1CellCultureMaterialsandReagents...... 262
A.1.2Surgical...... 263
A.1.3Histology...... 264
A.1.4Equipment...... 264
A.2Methods...... 264
A.2.1CellCultureReagentPreparation...... 264
A.2.2BoneMarrowHarvest...... 268
A.2.3PrimaryCellPurificationforCulturing...... 270
A.2.4MaintenanceofPrimaryCells...... 271
A.2.5 Maintenance of L929 Cell line ...... 273
A.2.6MaintenanceofMS-5Cellline...... 273
A.2.7MaintenanceofNIH3T3(eGFP)Cellline...... 274
A.3HistologicalMethods...... 274
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering vii Table of Contents Table of Contents
A.3.1StainPreparation...... 274
A.3.2EmbeddingandSectioning...... 275
A.3.3WellPlateStaining...... 275
A.3.4SectionedSlideStaining...... 275
B Development of CBD-Fibronectin fragment chimaeras 277
C Cross-flow Bioreactor Design Drawings 284
D Cross-flow Bioreactor Design 2 Drawings 287
E Fabrication of SU-8 master moulds 291
F Dissipation of gradient along diffusion chamber for the pyramidal gra- dient generating device 295
G Estimation of concentration gradient generated by the counter flow diffusion chamber 300
H Elementary hydrodynamic model for counter flow diffusion chamber302
I Matlab code 306
I.1 GetGain...... 306
I.2 GainCal...... 310
I.3 Coordinates...... 311
I.4 Collage...... 314
I.5 AllignInlet ...... 317
I.6 AllignCollage ...... 319
I.7 AllignCollage2 ...... 321
I.8 ExitAngles...... 323
I.9 GraphExitVModel...... 326
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering viii Table of Contents Table of Contents
I.10 TreeCalFix ...... 330
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering ix Abbreviations and Symbols
ALDH Aldehyde Dehydrogenase ALP Alkaline Phosphatase α-MEM Minimum Essential Medium - Alpha Medium BAAA BODIPY aminoacetaldehyde BMP Bone Morphogenetic Protein BMSC Bone Marrow Stromal Cell BMSSC Bone Marrow Stromal Stem Cells BSA Albumin from Bovine Serum C Molar concentration Cal-AM Calcein-AM CBDR Cellulose Binding Domain Retronectin CFU-F Colony Forming Unit Fibroblastic CFSE 5-(and-6)-carboxyfluorescein diacetate, succinimidyl ester (5(6)-CFDA, SE)
CO2 Carbon Dioxide D Mass Diffusion coefficient Da Dalton DAAA Dansly-aminoacetaldehyde DMEM - LG Dulbecco’s Modified Eagle Medium (D-MEM) - Low Glucose DMSO Dimethyl Sulfoxide DPBS Dulbecco’s Phosphate Buffered Saline EDTA Ethylenediaminetetra Acetic Acid EMEM Minimum Essential Medium Eagle FACS Fluorescence Activated Cell Sorting FBS Fetal Bovine Serum
x Abbreviations and Symbols
FCS Fetal Calf Serum FDA US Food and Drug Administration FGF Fibroblast Growth Factor FITC Fluorescein Isothiocyanate FSC Forward Scatter FTL-3 Fetal Liver Tyrokinase 3 Ligand G-CSF Granulocyte Colony Stimulating Factor GMP Good Manufacturing Practice HA Hydroxyapatite HAc Glacial Acetic Acid H&E Hematoxylin and Eosin HCl Hydrochloric Acid HEPES N-[2-Hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid] HF Hollow Fibre HSC Haemopoietic Stem Cell hUCB Human Umbilical Cord Blood ICM Inner Cell Mass IM Intramuscular IMDM Iscove’s Modified Dulbecco’s Medium IP Intraperitoneal I.U. International Units IV Intraventricular or Intravenous J Flux of particles k Boltzmann’s constant MAPC Multipotent Adult Progenitor Cell MoAb Monoclonal Antibody MPC Mesodermal Progenitor Cell MSC Mesenchymal Stem Cell MW Molecular Weight NaCl Sodium Chloride
NaHCO3 Sodium Bicarbonate NaOH Sodium Hydroxide
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xi Abbreviations and Symbols
PDMS Poly(dimethylsiloxane) PE Phycoerythrin Pe P´eclet number PerCP-Cy5.5 Peridinin Chlorophyll Protein - Cyanin - 5.5 PI Propidium Iodide P/S Penicillin G Streptomycin Sulphate ROS Reactive Oxidative Species Re Reynolds Number RS Recycling Stem Cells Runx Runt Homology Domain Transcription Factor SCF Stem Cell Factor SEM Standard Error of the Mean SF Silk Fibroin SPC Stromal precursor cell SSC Side Scatter T Temperature TCP Tissue Culture Plastic TE Tissue Engineering TPO Thrombopoeitin TGF-β Transforming Growth Factor Beta UV Ultra Violet v Velocity WFI Water for Irrigation μ Viscosity ρ Density τ Shear Stress
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xii List of Tables
2.1 Specific uptake of oxygen and glucose (mol/spercell)...... 24
3.1CFU-FassayofhUCBcells...... 51
3.2 hUCB cells labeled with Aldefluor substrate, Aldefluor substrate and CD34,orCD45...... 60
3.3CFU-Fassayofwholeratbonemarrowcells...... 62
3.4MeanfluorescenceratioofAldefluorsubstratelabeledratcells...... 67
3.5MeanfluorescenceratioofAldefluorlabeledporcinecells...... 77
3.6MeanfluorescenceratioofAldefluorlabeledequinecells...... 80
5.1Currentsurgicaltreatmentsforcartilagedegeneration...... 122
5.2 Cell and scaffold strategy combinations for osteochondral junction repair. 124
6.1 Shear stress and P´eclet numbers in the diffusion chamber of the pyrami- dalgradientgeneratingdevice...... 164
6.2 Shear stress in the cell chamber of the counter flow diffusion chamber . 175
7.1Cellcultureandgrowthinmicrofluidicdevices...... 187
7.2Summaryofconditions...... 205
7.3Directcontactcytotoxicityassay...... 208
7.4 Reactivity Grades ...... 209
A.1PassageDefinitions...... 271
xiii List of Figures
2.1Humanstemcellscategorisedbyorigin...... 10
2.2 Influence of soluble and mechanical factors on MSC renewal and differ- entiationintochondrogenicandosteogeniclineages...... 21
2.3Classificationofbioreactordevices...... 27
2.4Bioreactorgeometryandmasstransfer...... 28
3.1MononuclearhUCBcellslabeledwithAldefluorsubstrate...... 49
3.2 CFU-F assay of hUCB cells, representative cell colony ...... 51
3.3CFU-FassayofhUCBcells,platingefficiencyvsplatingdensity.... 52
3.4hUCBlabeledwithAldefluorsubstrate:flowcytometricanalysis.... 53
3.5hUCBlabeledwithCD34PE:flowcytometricanalysis...... 54
3.6 hUCB labeled with Aldefluor substrate and CD45 PE: flow cytometric analysis...... 55
3.7hUCBlabeledwithCD45PE:flowcytometricanalysis...... 55
3.8 hUCB labeled with Aldefluor substrate and CD34 PE: flow cytometric analysis...... 56
3.9hUCBlabeledwithAldefluorsubstrate:flowcytometricanalysis.... 57
3.10 hUCB labeled with Aldefluor substrate and CD34 PE: flow cytometric analysis...... 58
3.11hUCBlabeledwithCD45PE:flowcytometricanalysis...... 59
3.12 Percentage of CD45 PE labeled hUCB: fluorescence histogram ..... 59
xiv List of Figures List of Figures
3.13 Isolated hUCB ALDHbr CD34neg fractionation cultured under mesenchy- maloptimisedmediaconditions...... 61
3.14 CFU assay of whole rat bone marrow cells, representative colonies . . . 63
3.15 Aldefluor substrate labeled rat bone marrow: flow cytometric analysis . 64
3.16 Percentage of Aldefluor substrate labeled rat bone marrow cells: fluores- cencehistogram...... 64
3.17Aldefluorsubstratelabeledratcells:flowcytometricanalysis...... 66
3.18 Aldefluor substrate labeled mouse bone marrow: flow cytometric analysis 70
3.19 Percentage of Aldefluor substrate labeled mouse bone marrow cells: flu- orescence histogram ...... 71
3.20 Aldefluor substrate labeled mouse L929 fibroblasts: flow cytometric anal- ysis...... 72
3.21 Percentage of Aldefluor substrate labeled mouse L929 fibroblasts cells: fluorescence histogram ...... 72
3.22 Aldefluor substrate labeled porcine bone marrow: flow cytometric analysis 74
3.23 Percentage of Aldefluor substrate labeled porcine bone marrow cells: fluorescence histogram ...... 74
3.24Aldefluorsubstratelabeledporcinecells:flowcytometricanalysis.... 76
3.25 Aldefluor substrate labeled equine bone marrow: flow cytometric analysis 78
3.26 Percentage of Aldefluor substrate labeled equine bone marrow cells: flu- orescence histogram ...... 78
3.27Aldefluorsubstratelabeledequinecells:flowcytometricanalysis.... 80
4.1Singlehollowfibremodule...... 93
4.2 Cell attachment to bacteriological grade plates coated with CBDR . . . 95
4.3CellattachmenttoacellulosemembranecoatedwithCBDR...... 96
4.4CellattachmenttoacellulosehollowfibrecoatedwithCBDR...... 98
4.5 The effect of staining with DRAQ5 on MS-5 cells in HF modules coated withCBDR...... 100
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xv List of Figures List of Figures
4.6 The effect of staining with Calcein-AM, Hoechst 33342 and PI on MS-5 cellsinHFmodulescoatedwithCBDR...... 102
4.7 Auto-fluorescence of hollow fibres ...... 102
4.8 The effect of staining with DAPI, rhodamine phalloidin and PI on MS-5 cellsinHFmodulescoatedwithCBDR...... 104
4.9 The effect of staining with DAPI and rhodamine phalloidin on MS-5 cells inHFmodulescoatedwithCBDR...... 105
4.10 H&E staining of MS-5 cells in a HF module coated with CBDR, embed- ded and sectioned ...... 106
4.11 Rat bone marrow cell growth in CBDR coated hollow fibre modules . . 108
4.12 Images of rat bone marrow cell growth in CBDR coated hollow fibre modules...... 109
4.13 NIH 3T3 (eGFP) cell growth in CBDR coated hollow fibre modules . . 110
4.14 Images of NIH 3T3 (eGFP) cell growth in CBDR coated hollow fibre modules...... 111
4.15 Fold expansion of NIH 3T3 (eGFP) cells in CBDR coated hollow fibre modules, with extracapillary KG1a co-culture ...... 112
4.16 Fold expansion of MS-5 cells in CBDR coated hollow fibre modules . . . 113
5.1Currentapproachestoosteochondralgraftfabrication...... 124
5.2Cross-flowbioreactordesign...... 129
5.3Cross-flowbioreactordesign,detailedchamberview...... 130
5.4 Cross-flow bioreactor design schematic representation of fluid flow . . . 130
5.5Cross-flowbioreactordesign,sideview...... 131
5.6Cross-flowbioreactordesign,topview...... 131
5.7Cross-flowbioreactordesign2...... 133
5.8Cross-flowbioreactordesign2,channelview...... 134
5.9 Cross-flow bioreactor design 2, assembled view ...... 134
5.10Hollowfibrediffusionchamber...... 136
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xvi List of Figures List of Figures
5.11Constructionofhollowfibrediffusionchamber(1)...... 137
5.12Constructionofhollowfibrediffusionchamber(2)...... 138
5.13 Representative image of a slide hollow fibre bioreator with a collagen gel 139
5.14Effectofcelldensity...... 140
6.1Schematicdrawingofthepyramidalgradientgeneratingdevice..... 153
6.2Schematicdrawingofthecounterflowdiffusionchamber...... 154
6.3 Set up of assembly used to punch holes in PDMS...... 154
6.4Falsecolourgradientmap...... 155
6.5Uncorrectedtestimage...... 156
6.6 Uncorrected test image shown using false colour gradient mapping . . . 156
6.7 Mean pixel intensity vs Concentration. Mean gain vs Concentration. . . 158
6.8 Mean pixel intensity vs Exposure time. Mean gain vs Exposure time . . 159
6.9Correctedtestimage...... 160
6.10Correctedtestimageshownusingfalsecolourgradientmapping.....160
6.11 Photograph of a gradient generated by the pyramidal gradient generating device...... 162
6.12 Example of diffusive mixing taking place inside the serpentine channels ofthepyramidalgradientgenerator...... 163
6.13 Dissipation of concentration gradient along diffusion chamber of pyra- midalgradientgeneratingdevicegradientmodel...... 165
6.14 Effect of P´eclet number on concentration produced inside the diffusion chamberofthepyramidalgradientgeneratingdevice...... 166
6.15 Concentration profile at diffusion chamber inlet of pyramidal gradient generatingdevice...... 168
6.16 Concentration profile at diffusion chamber exit of pyramidal gradient generatingdevice...... 168
6.17 Concentration profiles for pyramidal gradient generating device at diffu- sionchamberinlet...... 169
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xvii List of Figures List of Figures
6.18 Concentration profiles for pyramidal gradient generating device along diffusionchamber...... 169
6.19 Effect of flow rate and position along diffusion chamber on concentration gradient at the mid-line of pyramidal gradient generating device diffusion chamber...... 170
6.20 Concentration gradient experimental data at mid-line of pyramidal gra- dient generating device diffusion chamber superimposed upon theoreti- callymodeledconcentrationgradient...... 170
6.21 Visualisation of diffusive mixing at the diffusion chamber inlet of the pyramidalgradientgeneratingdevice...... 171
6.22 Photograph of a gradient generated by the counter flow diffusion chamber174
6.23 Concentration gradient for counter flow diffusion chamber as predicted bytheoreticalmodel...... 176
6.24 Elementary hydrodynamic model. View is of longitudinal section through thedevice...... 177
6.25 Flow profiles predicted by the hydrodynamic model for dimensionless flow in the upper cell chamber of the counter flow diffusion chamber . . 177
6.26 Counter flow diffusion chamber flow profiles for fluorescein, with fluores- centdrainageexitopen...... 179
6.27 Counter flow diffusion chamber flow profiles for fluorescein, with water drainageexitopen...... 179
6.28 Counter flow diffusion chamber flow profiles for FITC20, with fluorescent drainageexitopen...... 180
6.29 Counter flow diffusion chamber flow profiles for FITC20, with water drainageexitopen...... 180
6.30 Counter flow diffusion chamber flow profiles for FITC150, with fluores- centdrainageexitopen...... 181
6.31 Counter flow diffusion chamber flow profiles for FITC150, with water drainageexitopen...... 181
7.1CellattachmenttoPETmembrane...... 197
7.2 Hoechst 33342 staining of cells on PET membrane ...... 198
7.3Highdensitycellsincounterflowdiffusionchamber...... 200
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xviii List of Figures List of Figures
7.4 Low density cells in counter flow diffusion chamber, with overnight flow 202
7.5 Confluent density cells in counter flow diffusion chamber, with overnight flow...... 203
7.6 Confluent density cells in counter flow diffusion chamber, without overnight flow...... 204
7.7Cellgrowthinhibitionassay...... 210
7.8Metalneedle...... 211
7.9Effectofmetalneedleoncelllayer...... 211
7.10 Cell growth inhibition assay of disposable plastic needles using NIH 3T3 (eGFP)cells...... 212
7.11PluroniccoatingofPETmembrane...... 213
7.12 Counter flow diffusion chamber connected with silicone tubing and plas- ticneedles...... 214
7.13 Syringe Extraction 4℃ ...... 215
7.14 Silicone, PTFE and polyethylene tubing extraction 4℃ ...... 216
7.15 Albumin and pluronic coated PTFE, and uncoated Tygon tubing ex- traction 4℃ ...... 217
7.16CFDCusingTygontubingmediumextraction...... 218
7.17 NIH 3T3 (eGFP) cells in counter flow diffusion chamber using Tygon tubing...... 220
C.1Cross-flowbioreactordesigngeneralview...... 285
C.2Cross-flowbioreactordesigndetailedchannelview...... 286
D.1Cross-flowbioreactordesign2topview...... 287
D.2Cross-flowbioreactordesign2frontview...... 288
D.3Cross-flowbioreactordesign2bottomview...... 288
D.4 Cross-flow bioreactor design 2 section view ...... 289
D.5Cross-flowbioreactordesign2isometricview...... 289
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xix List of Figures List of Figures
D.6Cross-flowbioreactordesign2rightview...... 290
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering xx Chapter 1
Introduction
1.1 Research Motives
The lack of available organs and tissues for implantation and the increasing demand for replacement tissues by ageing populations, has driven the rapid expansion of the interdisciplinary field of tissue engineering, which utilises combinations of cells, delivery systems, molecular signals, and specialised microenvironments to produce viable tissue that can be used either in vivo or extra-corporeally to restore, maintain, or improve the function of damaged or injured tissues [1, 2].
Mesenchymal stem cells (MSC) offer great promise in therapies aimed at repairing, replacing or regenerating damaged or diseased tissues and organs. This potential is due to their capacity for self-renewal, ability to differentiate down a range or lineages, and the potential in autologous therapies, free from major ethical concerns. Key issues facing widespread MSC therapeutic use include both the scarcity in adult tissues and the current lack of a simple unambiguous identifying marker.
The clinical potential of MSC culture systems will be realised by development of cost- effective and safe processes for large-scale expansion of MSC and derivative tissue types.
The development of these therapeutic products has stalled, mainly because flask or
1 Chapter 1. Introduction 1.1. Research Motives pellet cultures employed to study biological processes are not appropriate methods for clinical delivery. The manual passaging of large-volume MSC cultures is labour- intensive and relies on the skill of the lab worker to prevent contamination of cultures.
The expansion of MSC strongly depends on medium supplemented with FBS, a problem which is coming under increasing scrutiny by regulatory authorities, with the risk of transmission of infectious agents via serum. The generation of solid tissues such as cartilage and bone will require technologies for biomimicry of essential tissue elements, microvascular architecture for exchange of metabolites, and regulation of differentiation by cell-cell interactions, extracellular matrix molecules and growth factors.
It will be necessary to move away from the tissue culture flask and pipette for clin- ical process development. Bioreactor technologies offer the promise of transforming laboratory-based techniques into scalable cell and tissue production processes with stan- dards of safety and efficacy similar to those established for pharmaceutical therapeu- tics. Perfusion systems replace the vascular system and can support high-density cell growth. Selective exchange of gases and metabolites is possible when semi-permeable membranes are combined with perfusion systems. Soft lithography and microfluidic design offer extreme precision to define tissue geometry, leading to the next generation of bioreactor devices for tissue engineering.
The challenge will be to develop cost-effective bioreactor processes for the growth of primary cells and tissues in a manner which is controllable, standardize-able, and scaleable [3]. Their specialized growth factor requirements and sensitivity to microen- vironment will increase the cost of production in comparison to mammalian cell pro- duction of biopharmaceuticals which have less stringent growth factor requirements.
This research will focus on addressing the issues of cell isolation, expansion, and the development of a suitable bioreactor for the generation of stable controllable gradients in which tissue differentiation can take place.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 2 Chapter 1. Introduction 1.2. Thesis Aims
1.2 Thesis Aims
It is hypothesised that MSC can be isolated from a heterogeneous population of cells using a prospective isolation method that is not based on adherence to tissue culture plastic (TCP). Currently functional separation methods used perturb the original bio- chemical state of the cells.
It is hypothesised that hollow fibre bioreactor systems may be a scalable technology for
MSC expansion. Flask culture of MSC at the clinical scale is expensive and cumber- some. For clinically relevant numbers to be achieved, MSC expansion must take place in a tailored, controllable, standardize-able and cost effective manner.
It is hypothesised that long-term tissue gradients can be established ex vivo.Chemo- taxis and tissue morphogenesis can be studied via the use of specially designed biore- actor devices. Whilst there are reports of devices that can set up such a gradient, a void exists for devices that can generate such a gradient in thick tissue constructs.
With these hypotheses in view, the thesis has the following specific aims to:
1. Isolate a homogeneous population of precursor cells from a hetergeneous popula-
tion;
2. Determine initial feasibility of expanding anchorage dependent cell lines including
MSC using a scalable in vitro (hollow fibre) bioreactor technology;
3. Develop a bioreactor device capable of producing a stable controllable gradient;
4. Demonstrate cell growth inside a gradient generating chamber.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 3 Chapter 1. Introduction 1.3. Thesis Layout
1.3 Thesis Layout
Chapter 2 Presents a literature review of mesenchymal stem cells and bioreactors in the context of tissue engineering.
Chapter 3 Explores the application of a metabolic marker used for primitive haemopoi- etic precursor cell isolation to cells of mesenchymal origin.
Chapter 4 Investigates the use of a hollow fibre bioreactor device for the growth of anchorage dependent cells.
Chapter 5 Evaluates the design of multi-channeled single chambered bioreactor de- vices for the ability to produce stable controllable gradients.
Chapter 6 Presents experimentally produced stable controllable gradients inside mi- crofluidic bioreactor devices, and compares experimental results to theoretical predic- tions.
Chapter 7 Assess the biocompatibility and characterises cellular response of devices presented in Chapter 6.
Chapter 8 Summaries the work presented in this thesis, discussing conclusions and achievements. Recommendations for future work are also presented.
Appendix A Presents the general materials and methods used throughout the thesis.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 4 Chapter 2
Literature Review
2.1 Introduction
In this chapter a literature review on mesenchymal stem cells and bioreactors in the context of tissue engineering is presented.
2.2 Tissue Engineering
The lack of organs and tissues available for implantation is now being described as a crisis [4] and is a crucial problem faced by every major field of transplantation and reconstructive surgery. In addition to the social consequences of this unmet clinical need, the burden to the economy is significant: globally, the direct healthcare costs of organ replacement are about $US350 billion (about 8% of healthcare spending).
Furthermore, ageing populations in the Western world are increasing the demand for replacement tissues. In recent years this steadily increasing demand has driven the rapid expansion of the new field of tissue engineering (TE), which has been defined as
‘an interdisciplinary field that applies the principles of engineering and the life sciences toward the development of biological substitutes that restore, maintain, or improve tissue function’ [5]. Tissue engineering shares many common elements with the closely
5 Chapter 2. Literature Review 2.2. Tissue Engineering related fields of regenerative medicine and cell therapy. It utilizes combinations of cells, delivery systems, molecular signals, and specialized microenvironments to produce vi- able tissue that can be used either in vivo or extra-corporeally to restore, maintain, or improve the function of damaged or injured tissues [1, 2]. While TE is an inter- disciplinary field, it has also been described as lying at an ‘uncomfortable intersection between basic science and applied medicine’ [6]. These authors suggest that effort is often exerted at the ends of each field, rather than in a truly interdisciplinary form.
While new insights and discoveries mean that the number of organs or tissues that can potentially be engineered is ever-growing, only a few have reached clinical utility and these are generally where there has been a ‘long standing insight into stem cell biol- ogy’ [7]. Conversely, knowledge gained in the process of developing engineered tissues can be used to further our understanding in areas such as normal tissue function or drug development.
Clearly, tissue manufacture does not come without risk, and given the complexity and expense of therapies based on TE, alternative strategies are also being pursued. Auto- grafts do not have tissue immune rejection problems, but involve trauma to the donor site and the quality and quantity of the autograft tissue may be compromised by preex- isting disease. While allografts have appropriate biological and mechanical properties, they carry the risks of infection and host rejection, requiring lifelong immunosuppres- sion. Xenografts potentially offer an unlimited supply of new tissues but currently perform poorly, requiring immunosuppression and carrying the risk of inducing strong inflammatory responses and cross-species viral transmission (zoonosis). Somatic cell nuclear transfer, or therapeutic cloning, is another potential avenue for the procure- ment of cells and subsequent production of the desired tissue; though at this stage there is still only proof-of-concept in animal models. While this overcomes many of the supply issues associated with TE, a wide range of ethical and scientific issues still remain to be resolved, including long-term chromosomal stability, tumour (teratoma) formation and ethical standards for harvesting of ova for these applications. Another important recent development in this area, as outlined in the section on Dedifferen- tiation, is the reprogramming of fibroblasts to undifferentiated pluripotent stem cells
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 6 Chapter 2. Literature Review 2.2. Tissue Engineering using transcription factors.
In recent years considerable research effort has been devoted to the use of mesenchy- mal stem cells (MSC) in TE because of their capacity to differentiate down multiple lineages. MSC do not raise the same level of ethical concern as that surrounding em- bryonic stem cells [8]. Furthermore, human embryonic stem cells require a feeder layer for both isolation and expansion. The feeder layer is often of mouse origin and consists of a mitotically inactive layer of fibroblasts, which enhances self-renewal and prevents karyotypic abnormalities during expansion of the stem cells [9]. For therapeutic ap- plications, the presence of non-human cells in the production process is of concern to regulatory authorities. Another attraction of using adult stem cells is that they can be autologous, taken from the patient, manipulated as required and then implanted back into the same individual, without eliciting an immune response.
MSC have been shown to differentiate both in vitro and in vivo. The effects of some molecular signals, such as from hormones, vitamins, growth factors and cytokines that drive in vitro MSC proliferation and differentiation down various lineages have been identified [10] (see the section on Differentiation). MSC can also take on tissue-specific characteristics of mesenchyme derived tissues such as cartilage, bone and tendon when co-cultured with specialised cell types or when exposed to tissue extracts in vitro, as recently reviewed by de Silva Meirelles et al. [11]. It is also becoming apparent that MSC have an important in vivo immunosuppressive effect. Treatment with bone marrow derived cells for graft-versus-host disease after allogeneic cell transplantation has shown a reduction in symptoms [12]. Thus the combination of MSC with TE has great potential in addressing a wide variety of clinical applications.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 7 Chapter 2. Literature Review 2.3. Stem Cells
2.3 Stem Cells
2.3.1 Definitions and terminology
As with many rapidly developing areas of study, the nomenclature surrounding stem cell research is continually evolving, which can lead to confusion in terminology. This is particularly true in the context of stem cells, which vary in both differentiation potential and origin. When described by their ability to differentiate into different tissue types, stem cells have been traditionally classified as either totipotent, pluripotent or multipotent. Totipotent stem cells have the ability to give rise to all possible cell types of an organism, including the extraembryonic tissues, and form a complete organism.
Pluripotent stem cells have less capacity for differentiation than totipotent stem cells.
While they possess the ability to give rise to most cell types, they cannot form a complete organism, or give rise to the extraembryonic tissues. The distinction between the differentiation capacity of totipotent and pluripotent cells has recently become blurred, with authors now suggesting that cells once defined as pluripotent can indeed produce some or all of the extraembryonic tissues [13]. Multipotent stem cells have a more limited differentiation potential. These cells can produce multiple lineages that form an entire tissue or tissues. The distinction between pluripotent and multipotent cells has also recently become less clear. While it was thought for some time that stem cells found in adult tissues were only capable of developing into cells of that particular tissue type, it has now been shown that stem cells found in several adult tissues are capable of producing cells that do not belong to the same lineage as that of the originating tissue [14]. MSC have been shown to differentiate into non-mesodermal tissue types such as ectodermal astrocytes [15] and endodermal hepatocytes [16]. Stem cells with the capacity to produce two or more lineages within a tissue are described as oligopotent, and those cells with the ability to only produce one lineage are described as unipotent.
Progenitor cells possess less potential to differentiate than stem cells, and lie somewhere in the differentiation pathway between a stem cell and a terminally differentiated cell.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 8 Chapter 2. Literature Review 2.3. Stem Cells
The term precursor is often used interchangeably with progenitor, but some argue that precursor cells possess less potential than progenitor cells, while not being terminally differentiated. Cell plasticity refers to its differentiation ability: the more plastic a cell, the greater its ability to differentiate into different tissue types.
As shown in Figure 2.1, human stem cells when categorised by origin are generally described as embryonic, foetal, umbilical or adult. Embryo-derived pluripotent cells [9] are found in the inner cell mass (ICM) of the 5 to 6 day-old blastocyst. Embryonic germ cells are found in the gonadal ridge of the 5 to 9 week-old foetus. Foetal stem cells are mulitpotent, with foetal MSC found in a number of tissues including blood, liver, bone marrow, lung, kidney, and pancreas [17, 18]. Umbilical stem cells are multipotent [19], and are generally isolated from either the umbilical cord [20] or the Wharton’s Jelly, a mucoid connective tissue surrounding the blood vessels inside the umbilical cord [21].
Adult stem cells are further classified as either germline or somatic. Germline stem cells are unipotent, and only give rise to oocytes in the female and spermatozoa in the male, which are both totipotent stem cells. Adult stem cells are also referred to as somatic cells as they arise from the body or soma, and are multipotent. These cells have a more limited capacity for differentiation, with the ability to develop down multiple lineages that form a particular tissue.
Mesenchymal stem cells are one particular adult somatic cell type that has generated a great amount of interest for TE applications because of their differentiation capacity.
Once again, the rapid evolution in understanding of MSC biology has led to a confusing nomenclature, with no strict consensus as to what these cells are capable of. Thus MSC are also referred to as BMSC (bone marrow stromal cells), BMSSC (bone marrow stromal stem cells), CFU-F (colony forming unit-fibroblastic), SPC (stromal precursor cells), MAPC (multipotent adult progenitor cells), MPC (mesodermal progenitor cells), and RS (recycling stem cells) [22]. This review will endeavour to provide an overview of our current understanding of MSC, and describe their increasingly important role in
TE applications.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 9 Chapter 2. Literature Review 2.3. Stem Cells
Human Stem Cells
Embryonic Foetal Umbilicus Adult
Blastocyst Gonadal ridge Abortus Umbilical Wharton’s Germline Somatic (5-7 days) (6 weeks) (Foetal tissues) cord blood Jelly
Embryonic Embryonic Foetal Umbilical cord Umbilical cord Spermatogonia Oogonia Hemopoietic Mesenchymal Other* stem cells germ cells stem cells blood stem cells matrix stem cells
* Includes Epidermal (skin, hair), Bone marrow Peripheral Bone marrow Eye, Gut, Liver, Neuronal blood stroma
Figure 2.1: Human stem cells categorised by origin, modified from Bongso and Lee [23].
2.3.2 Biology of MSC
There are numerous recent reviews on MSC [24–31]. The term MSC was first used in the early 1990s [32] to describe cells initially identified by Friedenstein, Owens and col- leagues more than 20 years ago, centering around the adherence of fibroblastic-like bone marrow cells to tissue culture plastic and subsequent monolayer cloning occurring from a single cell into discrete colonies [33]. These adherent spindle shaped cells morpholog- ically resembling fibroblasts [34] were termed colony forming unit fibroblastic (CFU-F) cells, and were initially shown to differentiate down an osteogenic pathway [35].
2.3.2.1 Self-renewal
Stem cells are present in all tissues and act as a demand-driven reservoir for replenishing differentiated cells and repopulating tissue. They are defined by certain characteristics that set them apart from all other cells. Stem cells have the ability to remain in an undifferentiated, unspecialised state that does not display any specific characteristics of the surrounding tissue. They have the ability to self-renew, and do so at each division to maintain their numbers. This characteristic is often utilised in TE applications where large numbers of undifferentiated cells are required prior to commitment to a specific lineage. They also have the ability to differentiate into a specific cell type required by the host tissue. It is this unique property of having two developmental options
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 10 Chapter 2. Literature Review 2.3. Stem Cells available at each division, depending on which cytokine or hormonal cues are received, that makes these cells extremely important.
A cell’s phenotype will be maintained or increased in number if on average, at least one daughter cell has the same phenotype as the parent. This process is referred to as self-renewal [36]. This mode of division results in repopulation of tissue with stem cells after injury for tissue repair, or during development. Asymmetric division generates two different daughter cells. One is identical to the mother, and one is committed to a particular differentiation pathway. This type of division also maintains stem cell numbers in the tissue, while replenishing differentiated cells. All stem cells appear to have this intrinsic ability to either self-renew or differentiate and to switch between the two types of division as required, but the cell microenvironment provides the signals that define the direction the cell takes. This unique capacity for cell division is maintained throughout the lifespan of the cell [37].
The ability to continually self-renew and proliferate is linked to telomere length, which in adult somatic cells is shortened with each division, thus reflecting cellular age. The enzyme telomerase prevents shortening of telomere length by extending length at the end of the chromosome. Initially immortal cell lines, malignant tumour cells and em- bryonic stem cells [38] were the only cells thought to display telomerase activity, but it has now been shown that some somatic cells during ex vivo expansion also possess low levels of the enzyme [39]. While telomerase activity has been detected in adult stem cells from skin, gut and the haemopoietic system, so far this is not the case for human
MSC [40]. Expanding MSC causes loss in telomere length and a study by Baxter et al. highlighted the severe ageing during expansion using present protocols [41].
2.3.2.2 Source/niche
While these cells were initially isolated from mammalian bone marrow, they have now been found in a range of adult tissues including periosteum, trabecular bone, adipose tissue, synovium, skeletalmuscle, deciduous teeth [42], brain, spleen, liver, kidney, lung,
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 11 Chapter 2. Literature Review 2.3. Stem Cells thymus, pancreas [11], dermis [43], blood [44, 45], tendon and ligament. MSC have a wide species distribution, having been found in a number of species including human, mouse, rat, baboon, sheep, dog, pig, cow and horse as recently reviewed by Dazzi et al. [46].
The concept of a stem cell niche was proposed by Schofield in 1978 [47]. It has two components; anatomical and functional, that enable stem cell reproduction and self- renewal [48]. In a recent review, Kolf et al. examined the cellular, soluble and extracel- lular matrix components of the MSC niche and discussed the niche-related phenomena of stem cell homing to sites of injury [49].
Mammalian bone marrow is thought to contain three distinct cellular systems: haemopoi- etic, endothelial and stromal. MSC are most commonly harvested from the stromal compartment of bone marrow and are hence often referred to as bone marrow stromal cells. The stromal cell system was first described by Owen in 1985 [50, 51]. Stromal cells were defined as being non-haemopoietic and mesenchymal in origin. The stromal cell system comprises marrow derived stromal cells, MSC and all their progeny [51].
It has now been widely reported [52–54] that the postnatal mammalian bone marrow microenvironment does indeed sustain two main phenotypically and functionally dis- tinct populations of precursor cells; MSC, which give rise to the musculoskeletal tissues of the body, and haemopoietic stem cells (HSC), which give rise to blood cells and other elements found in the blood. There is an important interaction between these two cel- lular groups. Cells of mesenchymal origin structurally and functionally support and maintain hematopoiesis [54, 55]. This is achieved through providing a microenviron- ment that delivers required growth factors and promotes cell-to-cell and cell-to-matrix interactions [52, 54, 55].
2.3.2.3 Differentiation
The differentiation capability of adherent bone marrow stromal cells into mesodermal connective tissue lineages such as bone (osteoblasts) [56], cartilage (chondrocytes) [57,
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 12 Chapter 2. Literature Review 2.3. Stem Cells
58], and fat (adipocytes) [59–61] has been well described [62]. Successful differentiation into cells of the central nervous system, skeletal muscle, liver, and heart has also been reported [11].
Minguell et al. give a detailed summary of a range of in vitro molecular signals that have been identified in driving MSC differentiation down various lineages [63]. Growth factors and cytokines such as fibroblast growth factor (FGF), transforming growth factor-beta (TGF-β1), and bone morphogenetic protein (BMP), hormones such as dex- amethasone, vitamins such as ascorbic acid, and chemicals such as β-glycerophosphate each play an important role in achieving the desired cellular response. The complex interactions occurring from the major signalling pathways such as Wnt (secreted lipid- modified proteins, whose dysfunction results in oncogenic effects), Notch, BMP (se- creted proteins that belong to the TGF-β1 family), and transcriptional regulation of major factors such as runt homology domain transcription factor (Runx) 2 determine cell fate in a context dependent manner [64]. Kolf et al. have recently reviewed lineage- restrictive molecular regulators and their mechanisms of action [49].
As discussed earlier, adult stem cells now appear to have a broader differentiation ca- pacity than earlier work suggested. This is also the case for a group of adult progenitor cells found in bone marrow MSC cultures that were shown to convert to endothelium, ectoderm and endoderm both in vitro and in vivo [65]. These cells were also shown to contribute functionally to most somatic tissues. Differentiation into all three germ layer lineages: mesodermal, ectodermal and endodermal [11] lends weight to the argu- ment that these cells may indeed be pluripotent. This also raises questions as to the appropriateness of describing them as mesenchymal.
Despite the growing amount of data available, there still exists considerable debate on the subject of MSC plasticity. Some results, such as from Jiang et al. [65]. have been difficult to reproduce, and many in the field consider that they do not meet stringent enough criteria and may represent experimental artefacts. The controversy surrounding the differentiation capability of these cells is further propelled by no absolute widely accepted definition of a MSC, only operational definitions. As a result, only enrichment
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 13 Chapter 2. Literature Review 2.3. Stem Cells of various cell properties can be used to describe them.
The ability to differentiate down a range of lineages is particularly important in TE applications, where large numbers of cells are generally required. The use of differen- tiated cells for TE applications often faces challenges of cell isolation, expansion and maintenance of the differentiated phenotype. In this situation, the ability to take cells such as MSC that are amenable to harvesting and expansion and then induce them to undergo differentiation would be extremely valuable.
2.3.2.4 Dedifferentiation
Another possible future avenue available for the acquisition of large numbers of ho- mogeneous cells is the use of dedifferentiation. Traditionally, this process has been associated with organisms such as salamanders which exhibit extensive regenerative capacity, whereby differentiated cells are able to revert to a precursor cell type. More recently, this concept has been applied to mammals, such as humans, which have a more limited ability to regenerate tissue [66]. While much of this work is aimed at stimulating in vivo dedifferentiation in order to induce a regenerative response, it is equally applicable to the in vitro formation of progenitor cells from fully differenti- ated cells. A recent example is the dedifferentiation of neonatal pancreatic cells into precursor cells which can then expand and redifferentiate down both pancreatic and neural lineages [67]. In this regard, there are now indications that MSC are able to transdifferentiate into non-mesenchymal lineages, though the underlying mechanisms are poorly understood [68]. Similarly, it has recently been reported that the monolayer culturing of human articular chondrocytes led to dedifferentiated cells which exhibited reversion to a more primitive cell type [69].
If it proves possible to subsequently differentiate this type of cell down different lineages, the potential benefit to TE is obvious. It has recently been reported that germline- competent cells exhibiting the morphology and growth properties of embryonic stem cells and expressing embryonic stem cell marker genes can be generated from mouse
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 14 Chapter 2. Literature Review 2.3. Stem Cells
fibroblasts [70–72]. This work has now been extended to human cells [73, 74], offering an alternative strategy for the supply of cells for TE. While there are still many issues to be resolved, such as the long-term risk of teratoma formation, this is a promising and rapidly advancing area as indicated by the recent discovery that pluripotency can be induced without the use of the cancer-inducing c-Myc gene [75]. The ability to harvest an easily accessible and plentiful differentiated cell type such as blood or epithelial cells, dedifferentiate them to a more plastic cell type and then redifferentiate down a specific lineage would greatly enhance progress in this challenging area. This is an area of intense research activity, with many questions still to be answered, such as the stability of the redifferentiated phenotype following implantation.
2.3.3 MSC markers
Ideally there would be a unique property or simple set of characteristics that could be used to isolate and characterise MSC, but the reality is more complex. Furthermore, there is no widely accepted absolute definition of what constitutes an MSC, only opera- tional definitions. While the ability of marrow stromal cells to adhere to tissue culture plastic has been utilised as a discriminator to separate these cells from most of those with a haemopoietic origin [76], selection on this adherence basis results in a heteroge- neous population. From this heterogeneous population of retrospectively isolated cells, only a minor proportion will have clonogenic potential, and this method enriches the progeny of the colony-forming cells, but not the actual CFU-F [77].
To date, there is no simple technique available for prospective isolation of MSC from a heterogeneous population such as adult bone marrow, where they are present in low numbers ranging from 1 in 1x105 to1in1x106 mononuclear cells (reviewed by Baksh et al. [22]). Researchers generally use a panel of markers, including monoclonal antibody
(MoAb) labelling to identify, characterize, and isolate particular populations of cells from a heterogeneous sample, employing characteristic cell surface proteins. These cell markers can be used to identify and isolate homogeneous populations based either on the presence of characteristic cell surface proteins or the absence of lineage defining
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 15 Chapter 2. Literature Review 2.3. Stem Cells markers [78].
There are markers available that bind to specific antigens present on the cell surface.
One such marker developed by Simmons and Torok-Storb is STRO-1. STRO-1 is a murine IgMMoAb that can be used to isolate stromal precursors in freshly aspirated human bone marrow suspensions [79]. While STRO-1 does not bind to hematopoietic progenitor cells, it does bind to approximately 10% of bone marrow cells that comprise mainly glycophorin A - positive nucleated erythroid precursors [79], and as a result is not sufficient to totally purify a bone marrow population [77]. The use of vascular cell adhesion molecule-1 (VCAM-1/CD106) when used in conjunction with STRO-1 on human bone marrow has been shown to isolate a highly enriched population of cells with clonogenic potential [77]. While STRO-1 has shown some crossreactivity with mouse it is generally only used with human cells, limiting its usefulness as a cross-species marker.
As a result, panels of markers for expressed surface antigens are used in an attempt to select purified populations of MSC. Currently, there are a number of reported combi- nations, recently reviewed by Sethe et al. [80]. These include SH2, SH3, CD29, CD44,
CD71, CD90, CD106, CD120a, CD124 which were all observed to be uniformly positive for culture-expanded human adherent bone marrow cells [62]. The same cultures were found to be lacking expression of CD14, CD34 and CD45. Minguell et al.havegiven a detailed account of MSC characteristic expression of specific antigens, cytokines and growth factors, cytokine and growth factor receptors, adhesion molecules and extra- cellular matrix components [63]. Despite considerable research effort in this area, as yet there is no consensus view as to the ideal panel of markers for MSC isolation or characterisation. Characteristics such as the ‘side population’ (a metachromatic shift seen with H33342 staining), presence of high telomerase and aldehyde dehydrogenase
(ALDH) activity are now being considered as viable options to isolate stem cells [81]. If the results of such studies indicate that cells from different tissue origins possess these common characteristics, as is the case for embryonic and haemopoietic stem cells, then the argument for the existence of a set of universal stem cell markers is strengthened.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 16 Chapter 2. Literature Review 2.4. MSC in Tissue Engineering
2.4 MSC in Tissue Engineering
There is growing general consensus that in order to be useful for TE applications stem cells must possess certain reproducible characteristics. Stem cells should differentiate as directed into the required cell type and then integrate with the surrounding tissue after implantation. Ideally, stem cells should be found in abundant quantities with the ability to proliferate extensively. If small tissue samples contained large numbers of stem cells, then the strict and difficult requirements on cell expansion without differentiation could be easily overcome. The trauma or morbidity involved in cell harvesting should be justified in terms of patient outcomes [82].
The main shortcomings of adult stem cell based therapies are the rarity of these cells in tissues, their limited capacity for self-renewal and the difficulty with isolation. Cur- rent isolation methods used to obtain MSC for TE include adherence to tissue culture plastic [32], the use of magnetic beads [83], fibrin microbeads [84], density gradient centrifugation [62,85], size sieving [86], cell sorting based on surface protein expression, or combinations [87] of these. Many of the existing stem cell isolation procedures are also relatively expensive. They require lengthy multiple steps and reagents, and do not translate well from the laboratory setting into a production environment, an important criterion for widespread clinical utility.
The culture expansion process typically results in further dilution of stem cell numbers, as lineage committed transit amplifying cells grow more rapidly. Researchers are trying to find paths around this problem. One approach is to source cells from tissues that have low MSC density but are available in large volumes, such as placenta [88–90], or adipose tissue from liposuctions [82], so more cells can be harvested.
The plasticity of MSC makes them promising candidates for a number of clinical TE applications [91]. The number of citations in the literature for the use of these cells for various diseases is ever growing.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 17 Chapter 2. Literature Review 2.4. MSC in Tissue Engineering
2.4.1 Bone
One of the first and now most widespread applications of MSC is their use in bone defect repair. The first proof that bone defects in large animals could be repaired by MSC was provided by Bruder et al. who loaded autologous culture expanded bone marrow cells onto porous ceramic cylinders and implanted them into critical sized segmental defects in dog femora for 16 weeks. It was found that both woven and lamellar bone had filled the pores of the cell loaded implants and that the amount of bone was significantly greater than in the cell-free implants [92,93]. This is now widely accepted, and the use of MSC in the TE of bone has been covered by many authors [94–96]. Comprehensive reviews of all areas of bone TE are also available [97–99].
2.4.2 Cartilage
Adult cartilage tissue has limited capacity to repair. None of the current resurfacing techniques provide totally adequate repair of the articulating surface for a number of reasons such as the avascular nature of cartilage, and the limited potential of chon- drocytes to proliferate [100]. As a consequence of these poor results, new strategies are required to combat the ever growing need to replace articulating surfaces. In the
USA alone, the combined effect of both direct traditional medical costs and the cost of indirect economic wage loss from arthritis is greater than $65 billion annually [101], where as many as 21 million people are reported to suffer from severe joint pain and related dysfunction [102]. These figures are set to rise with population ageing and in- creasing prevalence of obesity, thus making successful joint repair therapies extremely
financially lucrative.
The formation of in vitro cartilage, while most often approached with isolated artic- ular chondrocytes [103] can be carried out with MSC, with a number of advantages.
Harvesting autologous chondrocytes causes trauma at the donor site, and only a small number of cells are obtainable. There is also a limited ability of the harvested cells to proliferate and undergo differentiation, as their functional state does not favour re-
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 18 Chapter 2. Literature Review 2.5. Bioreactors generation [104]. Not only are appropriate mechanical properties of cartilage difficult to replicate in vitro, but also without a supporting matrix it can be difficult to deliver target cells to the desired site and then maintain required retention. Unattached cells also do not proliferate and produce matrix [105].
Wakitani et al. implanted autologous culture expanded iliac crest bone marrow cells embedded in porcine derived collagen gels into articular cartilage defects in the medial femoral condyle as part of the repair strategy of osteoarthritic knee joints. While they reported no significant difference between cell-loaded and cell-free control gels, the arthroscopic and histological grading score was better in the cell-loaded group [106].
2.4.3 Osteochondral Junction
One problem facing the implantation of TE cartilage is fixing the new tissue to the subchondral bone in the joint. The aim of producing a solid join between the manu- factured cartilage and underlying bone has been a driving force behind osteochondral graft TE.
The osteochondral junction is a particularly difficult tissue to replicate in vitro. While
MSC can be differentiated down both bone and cartilage lineages, it is a tissue that is often approached as two separate pieces that are later put together as one. In a recent review Martin et al. discuss all the different combinations that are currently used to produce osteochondral grafts [107]. No group has yet reported producing a heterogenous graded construct that is manufactured with the same base material for both chondral and sub-chondral bone formation, seeded with the same cells.
2.5 Bioreactors
Bioreactors and fermentation processes have provided the production needs of the food and biopharmaceutical industry over many years. Many of the principles established by these applications can be translated into MSC production. However, there are several
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 19 Chapter 2. Literature Review2.6. MSC Metabolism and Physiological Requirements important differences between biopharmaceutical and stem cell production processes that are relevant for the design of bioreactors. The most obvious is that the product is a cell, rather than a protein. Accurate characterization of the cell product and develop- ment of regulatory guidelines for cell and tissue production are formidable challenges.
MSC growth and differentiation are complex and tightly regulated, requiring addition of scarce or expensive components to media (growth factors and serum components).
Pharmaceutical-like economy of scale will be difficult to realize, particularly if MSC products need to be individualized and tissue matched to avoid graft rejection.
2.6 MSC Metabolism and Physiological Requirements
Fetal calf serum (FCS) provides many of the key elements of culture media, including the proteins responsible for transporting essential metabolic substrates such as lipids, iron and fat-soluble vitamins. Growth and differentiation of MSC are regulated by growth factors, extracellular matrix proteins, proteoglycans and hormones. The key to development of fully defined culture media for clinical application will be to define the components within serum that are responsible for MSC expansion, and to understand the microenvironmental factors that regulate differentiation. The influences of soluble and mechanical factors on MSC growth and differentiation to cartilage and bone are summarized in Figure 2.2 [108].
2.6.1 Growth factors and hormones
Extracellular matrix proteins such as fibronectin, laminin and vitronectin are adsorbed from serum onto tissue culture grade polystyrene plasticware and may provide the necessary adhesive requirements for MSC proliferation [109]. The lysophospholipid sphingosylphosphorylcholine increases the proliferation of MSC in a dose-dependent fashion, via G-protein signalling and the c-Jun N-terminal kinase pathway [110]. In another study FGF-4 was shown to increase the in vitro expansion rate of human
MSC [111].
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 20 Chapter 2. Literature Review2.6. MSC Metabolism and Physiological Requirements
+ sphingosylphosphorylcholine Low oxygen FGF-4 - Integrin signaling Shear stress
MSC
Dexamethasone Dexamethasone ascorbic acid -2-phosphate TGF-b, BMP-2, FGF-2 b-glycerophosphate
+ + Cyclic strain Shear stress
Chondrogenic Osteogenic
Figure 2.2: Influence of soluble and mechanical factors on MSC renewal and differentiation into chondrogenic and osteogenic lineages. The components of serum that promote MSC renewal, and those that comprise chondrogenic or osteogenic media, are shown over the arrows. Oxygen concentration, cyclic strain and shear stress can have positive (+) or negative (-) effects on growth and differentiation [108].
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 21 Chapter 2. Literature Review2.6. MSC Metabolism and Physiological Requirements
Johnstone et al. described the in vitro chondrogenesis of MSC by pellet culture [57].
Inclusionof10−7 mol/L dexamethasone to the culture medium was sufficient to induce chondrogenic differentiation, though efficiency could be greatly enhanced with the ad- dition of transforming growth factor-β1(TGF-β1) with or without dexamethasone.
Shirasawa et al. showed that addition of BMP2 to TGF-β1 and dexamethasone dra- matically increased cartilage pellet size and the synthesis of cartilage matrix by MSC isolated from synovium [112]. Fibroblast growth factor-2 (FGF-2) and dexamethasone have also been shown to increase the production of hyaluronan in two-dimensional cul- ture of elastic cartilage-derived cells [113]. Osteogenic differentiation from MSC was optimized by Jaiswal et al. by adding dexamethasone (100 nmol/L), L-ascorbic acid-2- phosphate (0.05 mmol/L) and β-glycerophosphate (10 mmol/L) to culture media [56].
2.6.2 Oxidative microenvironment
Oxygen tension plays an important role in regulation of MSC renewal and differentia- tion. Fehrer et al. studied the replicative senescence of MSC during long-term culture and showed that low oxygen tension (3% atm) prolonged proliferative lifespan by about
10 population doublings. The in vitro differentiation capacity of MSC was restricted by low oxygen during long-term culture, resulting in reduced adipogenic and osteogenic potential. These effects could be reversed by cultivating cells at 20% pO2 [114]. Oxygen concentration changes have been shown to induce phenotypic changes in chondrocytes, such as a switch in collagen production from type II to type I, which can then lead to the formation of cartilage tissue with inferior biomechanical properties [115].
Oxidative species generated by culture media components [116] could also influence in- tracellular redox potential and differentiation. Changes in intracellular redox potential can modify cellular signalling pathways, transcription factors and gene expression [117].
Reyes et al. used autofluorescent spectroscopy to study the change in intracellular redox potential of MSC differentiated using osteogenic medium [118]. The ratio of reduced pyridine nucleotides to oxidized flavoproteins was used to estimate intracellular redox potential. Differentiation down the osteogenic path was associated with a reduced ratio
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 22 Chapter 2. Literature Review2.6. MSC Metabolism and Physiological Requirements
(oxidation). These authors also showed that cell density and contact could also reduce the ratio and intracellular redox potential.
2.6.3 Strain and shear stress
Mechanical factors have also been reported to influence chondrocyte metabolism. Cycli- cal compression or strain has been shown to alter the biosynthetic activity of chondro- cytes in tissue-engineered cartilage constructs. Depending on the magnitude of strain, high frequencies are associated with greater biosynthetic activity [119–121]. Chondro- cytes embedded in agarose have a frequency-dependent response in both proliferation and proteoglycan (GAG) synthesis, with optimal GAG synthesis at 1 Hz for a strain amplitude of 15% [121].
MSC expansion and osteogenic differentiation are sensitive to fluid shear stress and strain. Using elastomeric membranes with parallel grooves, Kurpinksy et al.showed that MSC align parallel to the strain axis (5% strain at 1 Hz over 2.4 days). Mechan- otransduction was associated with increased MSC proliferation [122]. Zhao et al. [123] showed that MSC proliferation as determined by colony-forming units-fibroblast (CFU-
F) was enhanced in highly porous poly(ethylene terephthalate) matrices at low flow rates (0.1 versus 1.5 mL/min) over a 20-day period of culture. The higher flow rate
(1.5 mL/min) upregulated osteogenic differentiation potential. The authors concluded that metabolic factors alone could not account for enhanced proliferation, and that shears in the range 10−5 − 10−4 Pa were responsible for the shift towards osteogene- sis. Differentiation of osteocytes has been shown to be enhanced in the range 0.5-1.5
Pa [124].
2.6.4 Oxygen and glucose uptake
The specific uptakes of oxygen and glucose for MSC and chondrocytes reported in the literature are shown in Table 2.1. Chondrocytes may have lower oxygen requirements and increased glucose consumption, though direct comparison is often not possible due
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 23 Chapter 2. Literature Review 2.7. Mass Transport to differences in experimental design, and no studies have directly compared MSC and chondrocytes.
Table 2.1: Specific uptake of oxygen and glucose (mol/s per cell).
Cell Type Metabolite MSC Ref. Chondrocytes Ref. Oxygen 2.5 ±1.3 x 10−17 [125] 2-4 x 10−17 [126] 3.3 x 10−17 [127] 0-1.5 x 10−18 [128] Glucose 1.4 x 10−16 [129] 3.2 x 10−15 [130] 9.3 x 10−16 [131]
2.7 Mass Transport
The role of mass transport is to ensure that cell metabolism is kept within a physi- ological range by provision of metabolic substrates and removal of toxic degradation products. Understanding how device geometry is related to convective or diffusive transport limitations is therefore a key element of bioreactor design.
2.7.1 The role of diffusion and convection
There is a range of substrates that need to be supplied to the biomass. The rate of diffusion is proportional to their concentration gradient, the constant of proportionality being the diffusion coefficient. The Stokes-Einstein equation relates the radius of the diffusing particle and temperature to the diffusion coefficient:
kT D = (2.1) 6πR where k is Boltzmann’s constant, T is temperature and R is the particle radius [132].
The volume of a sphere is proportional to the cube of its radius, (V = 4/3πr3), so the diffusion coefficient is approximately inversely related to the cube root of molecular
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 24 Chapter 2. Literature Review 2.7. Mass Transport weight. Larger substrates including low-density lipoproteins, free fatty acids, trans- ferrin and growth factors will have a lower diffusion coefficient. Their specific cellular uptake (mol/spercell)ismanyordersofmagnitudelowerthanoxygen,asisthemolar concentration required for binding to cell surface receptors (nano- to picomolar range).
The diffusion of growth factors is also limiting at low concentration, particularly if high density culture is not supplemented with additional growth factors.
The mass flux is also related to the gradient that can be generated at the cell/medium interface (Fick’s law of diffusion). For a stagnant boundary around a cell the mass transfer is by diffusion alone, and is limited by the thickness of the stagnant layer and the concentration at the boundary of the stagnant layer. Mass transfer is increased by reducing the thickness of the stagnant boundary layer surrounding cells. The metabolite with the lowest solubility relative to specific uptake (Table 2.1) is oxygen: approxi- mately 0.2 mmol/L in room air (partial pressure of oxygen is 0.2 atm) at 37℃. Flask culture systems rely primarily on diffusion, and to a lesser extent natural convection, for transport of oxygen to cells. The depth of media in the flask limits the supply of oxygen from the gas phase [133].
Within a polymer scaffold (e.g., polygycolic acid), such as those used to generate car- tilage [134], the material properties of the system are spatially and temporally het- erogeneous. Galban et al. considered a seeded polymer scaffold as consisting of two phases [135, 136]: a void phase (β), which contains the nutrient fluid and some poly- mer matrix, and the cell phase (γ), which includes the cells, nutrient fluid, extracellular matrix and some polymer matrix. Diffusive transport within the two phase system is modelled by coupling boundary conditions and diffusion equations for both phases (β,
γ). The volume-averaging method was utilized to derive a single averaged nutrient con- tinuity equation that allows calculation of effective diffusion coefficients as a function of cell volume fraction and time. Leddy et al. measured the diffusivity of tissue-engineered cartilage as a function of scaffold material, culture conditions and time in culture [137].
Diffusivity in these constructs was much greater than in native cartilage. A decrease in diffusivity over time was most likely related to new matrix synthesis and matrix contraction by cells in the fibrin and gelatin scaffolds.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 25 Chapter 2. Literature Review 2.7. Mass Transport
The role of convection is to reduce diffusion limitations imposed by stagnant boundary layers surrounding cells. Hydrodynamic modelling can be used to predict the flow field for various geometries by solution of the Navier-Stokes equations. This is only trivial for simple geometries that generate well-defined flow fields such as parallel plates [138,139], membranes [140] or hollow-fibre bioreactors [141].
Improving local perfusion of thick tissue constructs remains a significant challenge for scaffold-based devices. Static culture of cell-seeded 3D scaffolds typically produces thin tissue growth localized to the construct periphery [142]. Scaffolds can be perfused by housing them within a flow-through column [143–145], or by suspending them within rotary culture devices or spinner flasks [146]. While flow rate can be used to modulate media exchange, pore sizes, connectivity and anisotropy can impart vastly different rates of media exchange and shear stress on cells within the construct.
Calculating flow-mediated shear stress within a porous scaffold is not a trivial exercise.
Porter et al. utilized computational fluid dynamics to model the flow of media through scaffolds [142]. Micro-computed tomography was used to define scaffold geometry, with generation of a detailed simulation of the flow field within pores. The authors were able to estimate average shear stress within the scaffold and to estimate the fluid shear stresses (5 x 105 Pa) that correlate with increased osteogenic proliferation. Zhao et al. extended this numerical approach to include terms for the diffusion and consumption of oxygen within the scaffold [123].
2.7.2 Selective exchange using membrane systems
Gas-permeable or dialysis membranes offer independent control of respiratory gases or small molecules. The quantity of oxygen that can be delivered to the biomass is limited by its solubility in water. Media perfusion rates can be drastically reduced if oxygen is supplied from a silicone membrane in close proximity to cells [138].
A dialysis membrane partitions proteins and growth factors from metabolites of less than 8000 Da. Cells and proteins are dialysed against media without growth factors.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 26 Chapter 2. Literature Review 2.8. Bioreactor Designs
Cells are grown in direct contact with the membrane at zero shear, while culture media are exchanged by a perfusion system on the other side of the membrane. Membrane perfusion systems can support cell densities that approach those found in tissues [147].
2.8 Bioreactor Designs
A bioreactor is defined as any device that provides the physiological requirements of the cell (e.g., nutrients, growth factors and mechanical environment) for study of cel- lular function or scaling production of cells and their products. Figure 2.3 shows a classification of mammalian cell culture methods that is based on culture device geom- etry. Figure 2.4 indicates how convection is used to facilitate mass transfer for various bioreactor geometries. The simplest bioreactors are static culture devices (flasks, bags or dishes) consisting of a single unstirred compartment where nutrients diffuse to cells.
Gas exchange (oxygen and carbon dioxide) occurs at the media/gas interface. Mixing will reduce the stagnant boundary layer surrounding cells, scaffolds or micro-carriers, as well as creating a more homogeneous cellular microenvironment. Mixed vessels can be
‘fed’ batchwise, or if cells are immobilized onto scaffolds or micro-carriers, using a con- tinuous perfusion system. A perfusion system attempts to replace the function of the microcirculation so that cells can be grown near tissue density. In the following sections we will review the application of these devices to MSC growth and differentiation.
Static Mixed Perfusion
Stirred Column Flasks
Rotary Parallel plate
Dishes Wave Hollow Fiber
± Scaffolds Bags Micro-fluidic or micro-carriers
Figure 2.3: Classification of bioreactor devices
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 27 Chapter 2. Literature Review 2.8. Bioreactor Designs
A. Spinner flask B. Rotating wall C. Wave
D. Column E. Parallel plate F. Hollow fibre
Figure 2.4: Bioreactor geometry and mass transfer. The diagrams indicate how convection facilitates mass transfer for various bioreactor geometries: (A) media are mixed by a magnetic impeller; (B) media inside a rotating-wall bioreactor have the same angular velocity, and particles are suspended in ‘free-fall’; (C) bag culture on a rocker generates a wave motion which mixes cells and media; (D) ‘plug’flow through a column generates a uniform perfusion, though media may not pene- trate the interstices of the porous matrix; (E) the parallel plate design generates a parabolic velocity profile with shear stress directly proportional to flow rate. A silicone membrane (upper plate) is used for exchange of oxygen and carbon dioxide; (F) adherent cells growing on the inside of a dialysis fibre with media flowing over the outside of the fibre.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 28 Chapter 2. Literature Review 2.8. Bioreactor Designs
2.8.1 Micro-carriers and scaffolds
Porous micro-carriers were developed for mammalian cell recombinant protein produc- tion in stirred vessels. The beads generally range between 100 and 400μmindiameter, providing a large surface area for attachment of cells. Beads increase attachment sur- face area and are less likely to foul filtration devices used to separate cells from media and secreted products. Granet et al. investigated the use of Cytodex™ and Biosion™ for use in osteoblastic culture [148]. Cytodex™ beads are made of a thin layer of denatured type I collagen, chemically coupled to a matrix of cross-linked dextran. Biosion™ beads are small, globular, negatively charged plastic particles. Micro-carrier beads have also been used for MSC expansion [149].
The rationale for the use of scaffolds in the manufacture of bone is to mimic closely the anatomical organization of bone and its tissue matrix. Hydroxyapatite (HA), which in different forms has been approved by the US Food and Drug Administration, is currently used clinically as a bone void filler, and is a logical choice for development of tissue-engineered bone grafts because of its osteoinductive properties. Porous HA is manufactured using a variety of processes including hydrothermal conversion of coral, deproteinization of bovine bone, foaming of HA slurries and porogen leaching. It is also possible to manufacture HA structures with well-defined porous geometry with bubble jet printing or robotic deposition [150,151]. There are numerous studies examining the in vitro growth of MSC on HA scaffolds [152–163]. Pre-clinical proof of principle is provided by the sheep metatarsal model of bone healing, where composite HA-MSC transplants have been shown to have almost the same osteogenic potential as autologous bone grafts in terms of the amount of newly formed bone present at 4 months [164].
This approach is gaining wider clinical acceptance as an alternative to autologous bone grafting.
Resorbable polymer scaffolds for bone regeneration include those manufactured from silk [165–169], chitin [154, 163, 170], collagen [170, 171] and polyglycolic acids [172–
174]. Porous three-dimensional silk scaffolds have been made using three fabrication techniques: freeze-drying, salt leaching and gas foaming. Silk fibroin (SF) is a biocom-
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 29 Chapter 2. Literature Review 2.8. Bioreactor Designs patible, enzymatically degradable material, which can be processed into a bone-like structure, for generation of bone [165, 166]. The resorbable polymer scaffolds that have been investigated for the generation of cartilage include polylactic and polylactic- glycolic acid [175–177], alginate [176, 178], silk [179, 180], polycaprolactone [181, 182] and devitalized menisci [183].
2.8.2 Mixed vessels
2.8.2.1 Spinner flasks
Spinner flasks are glass or plastic vessels with a central magnetic stirrer shaft and side arms for the addition and removal of cells and medium, and gassing with CO2- enriched air (Figure 2.4(A)). Spinners increase the efficiency of scaffold cell seeding and survival in comparison to static culture. Scaffolds are suspended within the spinner flask by threading onto long needles embedded in the side arm stoppers or other support structures. Scaffold and spinner flasks have been used for cultivation of MSC with osteogenic differentiation [152, 165, 166, 184].
2.8.2.2 Rotating-wall bioreactors
The rotating-wall reactor (Figure 2.4(B)) was originally designed to simulate micrograv- ity effects similar to those achieved for cell growth experiments performed by NASA on the space shuttle [185]. A horizontally rotating cylinder which is completely filled with culture medium (no gas-liquid interface) rotates liquid inside at the same angu- lar rate as the wall. If the velocity of the rotating fluid is equal and opposite to the sedimentation rate of cells or micro-carriers, the cell suspension will be maintained in a state of ‘free-fall’. Viscous flow around the falling particles reduces the thickness of the boundary layer surrounding cells, resulting in a shorter diffusion distance from the bulk media to the cell. Oxygenation of the media is provided by a silicone mem- brane wrapped around a central cylinder. Cell-seeded scaffolds can be grown in an
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 30 Chapter 2. Literature Review 2.8. Bioreactor Designs environment with low shear stresses.
The rotating-wall reactor has been used to suspend micro-carriers [148], porous scaf- folds [153,186] or hollow micro-spheres [187] for osteogenic differentiation with superior results when compared to static culture. Cartilage engineering is also particularly fea- sible using the rotating-wall vessel. Chondrocytes have been generated on beads [188], meshes [189] and novel porous biopolymers [190]. Most notably, Ohyabu et al.were able to generate large cartilage cylinders (1.25 cm × 0.6 cm, height × diameter) from rabbit marrow cells which formed spontaneously without a scaffold [191].
2.8.2.3 Wave bioreactors
The wave bioreactor system (Figure 2.4(C)) provides a gentle wave motion for mixing, provides higher oxygen transfer than in spinner flasks, and has been shown to perform comparably with stirred-tank bioreactors for working volumes between 1 and 100 L
[192]. Gas-permeable bags are simply placed on a rocker, which induces the wave motion. Cholesterol adsorption onto low-density polyethylene bags sold commercially for this application can be a problem, but can be overcome by pre-treating the bags or replacing them with inert fluorinated ethylene propylene [193]. There have been no reported applications to MSC growth and differentiation, but the disposable contained bag system has obvious advantages for clinical applications.
2.8.3 Perfusion bioreactors
A perfusion system distributes media via a network of channels to the biomass and increases mass transfer by continual exchange of media. Various bioreactor configura- tions for this purpose include scaffolds packed within a column, hollow fibre arrays or
‘printed ’ networks of micro-channels (microfluidics). The fluid path must be confined so that perfusion is equally distributed to cells. Mass transfer is further enhanced if the distances between immobilized cells and the perfusion channel are short. Regular geometries such as hollow fibre arrays and microfluidic designs have a more predictable
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 31 Chapter 2. Literature Review 2.8. Bioreactor Designs
flow distribution compared to porous networks.
2.8.3.1 Column bioreactors
Hydrodynamic principles that govern the flow pattern inside columns are applied to achieve uniform (plug) flow (Figure 2.4(D)) through the column [141]. The flow dis- tribution will depend on the hydraulic resistance of the column bed and the design of headers that distribute flow to the column. If the hydraulic permeability of the ma- trix is non-uniform, flow will be unequally distributed. If the pressure drop along the column is much greater than the dynamic pressure at the inlet header, the flow should mainly depend on the hydraulic permeability of the scaffold, and will be uniform if the scaffold porosity is uniform. Bancroft et al. developed perfusion bioreactors for bone tissue engineering applications [143, 194]. The scaffold is held in a cylindrical cassette sandwiched between two ‘O ’ rings which seal the chamber. The fluid path is confined so that it passes through the scaffold rather than around it.
Fluid flow increases oxygen transport [127], osteogenesis and mineralization in a dose- dependent manner [143, 195–198]. There have been some studies showing that os- teogenic development is enhanced in perfusion bioreactors in comparison to spinner
flask and rotating-wall bioreactors [146, 199]. Non-uniform seeding of the scaffold has been a problem in static and perfusion systems. Investigators have partially overcome this by oscillatory flow [199] or low pressure [200].
The beneficial effects of perfusion have also been verified for cartilage production [175,
201–204], though it is probably fair to comment that the material produced is ‘cartilage- like ’ because it does not have the mechanical strength of explanted hyaline cartilage, limiting application to cosmetic rather than load-bearing applications. One of the problems may be related to poor retention of secreted matrix molecules such as collagen
II and various glycosaminoglycans, by the perfusion system. The strength of cartilage is also related to the orientation of collagen fibrils [205]. The anisotropic properties of cartilage may be realized by use of oriented nano-fibrous scaffolds [206].
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 32 Chapter 2. Literature Review 2.8. Bioreactor Designs
2.8.3.2 Parallel plates
A radial perfusion bioreactor design was developed for the support of human haematopoiesis on a stromal feeder layer [207,208]. The bioreactor consists of two primary compartments: a gas compartment that is separated from the bottom compartment by a gas-permeable, liquid-impermeable membrane, and the liquid-filled bottom compartment with a tissue culture plastic surface for support of anchorage-dependent cells (Figure 2.4(E)). Fresh medium enters the liquid compartment at the centre and flows radially outward over cells before exiting into the waste container. The device has been integrated into a
Good Manufacturing Practice (GMP) cell production system, which consists of an in- cubator unit and a processor unit to inoculate and harvest cells. Each bioreactor is a disposable cell cassette which is loaded into processor or incubator platforms [209]. The parallel plate perfusion device was used to significantly expand CFU-F and progenitors with osteogenic potential from bone marrow mononuclear cells [139].
2.8.3.3 Hollow-fibre bioreactor
The clinical use of hollow-fibre modules for cell expansion was described 20 years ago by
Knazek et al [210]. Human tumour-infiltrating lymphocytes isolated from patients with metastatic melanoma were expanded by 2-3 logs to produce around 1010 lymphocytes.
Since that time hollow-fibre systems have been applied to mammalian cell culture processes including lymphocyte expansion [158, 211], gene transfer to hematopoietic cells [212], production of recombinant proteins and viruses [178, 213–217], hepatocyte culture and extracorporeal hepatic assist devices [182, 218–222]. Hollow-fibre biore- actors have been used to study cartilage development. The distribution of collagen, proteoglycan and glycosaminoglycan content within the bioreactor was determined by
Fourier-transform infrared imaging spectroscopy [223, 224].
A hollow-fibre bioreactor is a two-compartment system consisting of intracapillary and extracapillary spaces. Intracapillary flow is distributed by headers to a hollow-fibre bundle that is potted in resin. Modules can be designed so that flow distributes equally
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 33 Chapter 2. Literature Review 2.8. Bioreactor Designs to each hollow fibre in the bundle [141]. Flow within the lumen of each fibre has a parabolic velocity profile, so that cells attached to the inside surface are subject to a uniform shear stress which is directly proportional to the intracapillary flow rate. The hollow-fibre bundle is encased in a cylindrical shell with ports for flow of media around hollow fibres. The hollow-fibre membrane is semi-permeable, the pore size determining which molecular species are rejected.
The most widespread biomedical application is renal dialysis, where blood is passed through the intracapillary side of a hollow-fibre membrane dialyser with a low molecular weight cut-off (<10 000 Da). The dialysate is perfused through the extracapillary side of the module. Analogously, cells have been grown on the inside of the fibre (Figure
2.4(F)), with perfusion of media on the outside of fibres [147, 218, 221]. Alternatively, cells are inoculated into the extracellular space [210,211] with intracapillary perfusion.
A macroporous membrane will allow proteins to cross compartments, whereas a dialysis membrane will retain proteins within the cell growth compartment. The hollow-fibre membrane may be modified with ligands for cell separation or attachment of anchorage- dependent cell types [225, 226].
Elevated upstream pressure in hollow fibres can drive intracapillary medium out of the upstream section of the fibre, which is then driven back into the fibre downstream.
This secondary flow is termed Starling flow [227] and tends to drive cells towards the downstream end of the module. Starling flow is not a significant problem for hollow-
fibre dialysis modules, which have a much lower hydraulic permeability compared to macroporous membranes [141].
2.8.3.4 Microfluidic bioreactors
Microfluidic devices are fabricated using soft lithographic techniques originally devel- oped by Whitesides and colleagues [228–231]. Poly(dimethylsiloxane) (PDMS), which is biocompatible [232], optically transparent, permeable to respiratory gases and elas- tomeric, is cast onto silicon wafers that have been patterned and profiled by pho-
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 34 Chapter 2. Literature Review 2.8. Bioreactor Designs tolithography [233]. The silicon mould is manufactured by borrowing well-established lithographic methods developed for the semiconductor industry. Two or more PDMS layers can be sandwiched together to form individually addressable reaction chambers with controlling micro-valves, multiplexers and micro-peristaltic pumps [234–238].
Soft lithography is a relatively simple and cheap approach providing great flexibility for manufacturing a wide range of bioreactor device geometries with extreme precision.
Diffusion of nutrients is across micron distances, so the mass transfer requirements of cells are easily met. Well-defined geometry means that viscous flow and mass transfer can be estimated by relatively simple calculations. Soft lithography therefore provides a versatile approach in studying the metabolic and physiological requirements of cells.
Pioneering studies by Leclerc et al. [232] established the early cell culture applications of soft lithography. A reactor consisting of 10 PDMS layers stacked together with four cell culture chambers, and a chamber dedicated for oxygen supply, established proof of concept for the scalability of this bioreactor manufacturing approach [239]. The device was used to support the growth of the hepatic cell line Hep G2. Photolithographic methods were then tested with photosensitive biodegradable polymers which could be used for tissue-engineering applications. Initial biocompatibility data were provided by the attachment and growth of a range of cell lines on microstructures created with a new photosensitive biodegradable polymer [240]. PDMS micro-devices with a 3D microstructured channel network were also used to study the influence of shear stress on osteoblast growth and differentiation [241]. PDMS devices for culturing cells under conditions of cyclic strain have also been developed. Micro-grooved PDMS sheets have provided a relatively simple method for studying the mechanosensing properties of
MSC [122].
Toh et al. micro-fabricated a PDMS device for support of MSC and hepatocytes. Cell- cell interactions were facilitated by perfusion-seeding cells through a pillar array which concentrated cells within the bioreactor. Cells were then stabilized by a laminar flow complex coacervation reaction of polyelectrolytes (cationic methylated collagen and an- ionic terpolymer of hydroxylethylmethacrylate-methylmethacrylate-methylacrylic acid)
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 35 Chapter 2. Literature Review 2.8. Bioreactor Designs
[242]. Histological staining of bioreactors seeded with MSC showed that there were cal- cium deposits after one week of osteogenic induction.
The concept of ‘lab-on-a-chip ’ analysis has been applied to cellular analysis by Gomez-
Sjoberg et al [235]. They have developed a microfluidic chip that creates arbitrary culture media formulations in 96 independent culture chambers and maintains cell vi- ability for weeks. This system was used to automate analysis of MSC proliferation, motility and osteogenic differentiation in response to a range of cell culture regimes.
Time-lapse imaging revealed that overall cell motility decreased in chambers where cells were stimulated with osteogenic medium. The results also showed that human
MSC need a minimum of 4 days’ stimulation with osteogenic medium to fully com- mit to differentiation into the osteogenic lineage. The study provides an example of how microfluidic systems can rapidly optimize culture conditions for tissue engineering applications.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 36 Chapter 3
Cell identification and isolation
3.1 Introduction
Currently many of the existing stem cell isolation procedures are relatively expensive.
They require lengthy multiple steps and reagents, and do not translate well from the laboratory setting into a clinical one, an important criterion for widespread clinical us- ability. For research applications, one requires a prospective isolation technique that can be applied across the multiple species used for mesenchymal stem cell research. Alde- hyde dehydrogenase (ALDH) substrates have been used for cross-species haemopoietic progenitor isolation. The utility of an ALDH substrate as a MSC marker is therefore investigated.
Accurate precursor cell identification from a heterogeneous population remains an im- portant issue if precursor cells are to be utilized for their differentiation potential in tissue engineering applications. For clinical utility, a cheap large scale cell separation method is required. Experimental analysis of MSC function and regulation may require more specific methods. The cell separation technologies most suited for the application will depend on the tissue source (e.g. bone marrow, placenta, fat), the volume of tissue to be processed, and the intended use (clinical versus experimental).
37 Chapter 3. Cell identification and isolation 3.1. Introduction
It has been widely reported [52–54] that the postnatal mammalian bone marrow mi- croenvironment sustains two main phenotypically and functionally distinct populations of precursor cells, mesenchymal stem cells, which give rise to the musculoskeletal tis- sues of the body; and haemopoietic stem cells (HSC), which give rise to blood cells and other elements found in the blood. Therefore an initial separation would need to separate haemopoietic from mesenchymal cellular elements.
Human umbilical cord blood (hUCB) can, and has been used as an alternate source of
HSC to bone marrow. Cord blood has been shown to contain HSC [243,244], and for a number of years now has been considered a viable alternative to bone marrow as a source for transplantable HSC [245,246]. Recently there has been considerable debate amongst researchers if it also contains MSC. Some authors have failed to isolate MSC from hUCB [247–252] whilst others are reporting that they have been successful [20,253–262].
To resolve this open question, the frequency of colony forming unit fibroblastic (CFU-F) in cord blood will be determined.
3.1.1 Isolation Techniques
Stem cells are isolated on the premise that they possess properties that can be used to distinguish them from other cells. These properties are pluri- or multi-potency, quies- cence (adult stem cells), and proliferative potential. Unfortunately it is only possible to assay these properties retrospectively, because in vitro or in vivo growth is required to measure these properties directly. An alternate approach is to define biochemical mark- ers that correlate with the cardinal properties of stem cells. Common characteristics such as the presence of high telomerase and ALDH activity are now being considered as viable options to isolate cells [81], as is the ‘side population’ seen with Hoechst 33342 dye, discovered by Goodell et al. [263] over 15 years ago.
Current methodologies for the isolation of MSC from a heterogeneous bone marrow pop- ulation are based on the initial work of Friedenstein, Owens, and colleagues, centering around the adherence of fibroblastic like bone marrow cells to tissue culture plastic and
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 38 Chapter 3. Cell identification and isolation 3.1. Introduction subsequent monolayer cloning occurring from a single cell into discrete colonies [33].
These adherent spindle shaped cells morphologically resembling fibroblasts [34] were termed colony forming unit fibroblastic (CFU-F) cells, and were initially shown to dif- ferentiate down an osteogenic pathway [35]. Adherent bone marrow stromal cells have now been demonstrated to possess the ability to differentiate into various mesenchymal tissue types [62].
This is a functional separation method, which perturbs the original biochemical state of MSC. Whilst practical for clinical applications and most experimental approaches, plastic adherence may significantly alter the biological properties of MSC.
Whilst the ability of marrow stromal cells to adhere to tissue culture plastic has been utilized as a discriminator to separate these cells from most of those with a hematopoi- etic origin [76], selection on this adherence basis does not result in a homogeneous starting population of adherent bone marrow cells. From this heterogeneous popula- tion of retrospectively isolated cells, only a minor proportion will be progenitors, and this method selects for the progeny of the colony forming cells, but not for the actual
CFU-F [77].
From a rigorous biological perspective, a prospective isolation method is required to capture the scarce precursor cells, thought to present in adult bone marrow in low numbers ranging from 1 in 1 x 105 to 1 in 1 x 106 mononuclear cells, as reviewed by
Baksh et al [22].
3.1.2 Cell Markers
To date, there is no simple technique available for the isolation of MSC from a heteroge- neous population. Researchers generally use a panel of markers, including monoclonal antibody (MoAb) labeling to identify, characterize, and isolate particular populations of cells from a heterogeneous sample, employing characteristic cell surface proteins.
These cell markers can be used to identify and isolate homogeneous populations based either on the presence of characteristic cell surface proteins or the absence of lineage
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 39 Chapter 3. Cell identification and isolation 3.1. Introduction defining markers [78].
There are markers available that bind to specific antigens present on the cell surface.
One such developed marker by Simmons and Torok-Storb is STRO-1. STRO-1 is a murine IgM MoAb that can be used to isolate stromal precursors in freshly aspirated bone marrow suspensions [79].
Whilst STRO-1 does not bind to hematopoietic progenitor cells, it does bind to ap- proximately 10% of bone marrow cells that comprise mainly glycophorin A - positive nucleated erythroid precursors [79], and as a result is not sufficient to totally purify a bone marrow population [77]. Another drawback being that STRO-1 is species specific
(human), thus limiting its usefulness as an across species marker.
Aldehyde dehydrogenase activity has been used to identify and isolate haemopoietic progenitor cells. The proof of principle for the idea of employing a fluorescent sub- strate for assessment of ALDH activity was first presented by Jones et al [264, 265], reporting the development of dansyl-aminoacetaldehyde (DAAA) as a means of isolat- ing haemopoietic progenitor cells in mice.
The human ALDH gene superfamily contains 19 functional and 3 non functional genes
[266]. ALDH activity has been found in over 50 animals, fungi and bacteria [267] and is known to be involved with the oxidation of a variety of endogenous and exogenous aldehydes, including retinol (vitamin A) [268].
ALDH appears to be involved in the mechanisms responsible for resistance to alkylating agents cyclophosphamide and 4-hydroperoxycyclophosphamide [269], and is found in a variety of tissue types, including red blood cells, and in various subcellular locations, such as the cytosol, mitochondria, or microsome [267]. Haemopoietic stem cells are relatively resistant to alkylating reagents, leading to the hypothesis that ALDH activity can be used to identify haemopoietic stem cells. It is the activity of the cytosolic intracellular protein that is of most interest here. Thus it is postulated that other stem cell types, such as MSC, may also retain resistance to alkylating agents, and have similar ALDH staining profiles to HSC.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 40 Chapter 3. Cell identification and isolation 3.1. Introduction
BODIPY aminoacetaldehyde (BAAA), was developed [270], also for isolating haemopoi- etic progenitor cells that express ALDH. BAAA consists of an aminoacetaldehyde moi- ety bonded to the BODIPY fluorochrome and has been commercialized (Aldefluor®).
Upon incubation, BAAA freely diffuses into cells, where the enzyme converts it into a carboxylate ion, BODIPY amino acetate (BAA) [270]. The BAA is constrained intracellularly due to its net negative charge, and cells with high levels of ALDH present can be identified by fluorescence and flow cytometry.
The substrate employs a simple inexpensive method where the fluorochrome is excited by visible light, and the emission spectrum of BAA does not overlap significantly with other fluorochromes making it easily used with other supplementary markers [270].
The substrate is non toxic [78], with BAA not entering the nucleus or binding DNA, and rapidly flowing out, once the calcium ion efflux inhibitor used to slow down the rate at which the substrate is pumped out of the labeled cells is removed [270].
Another potential advantage for Aldefluor is its cross-species applicability. The sub- strate is metabolised by most mammalian cells. Most MoAbs are species specific lim- iting their application as universal markers of stem cell function.
3.1.3 Approach and Aims
Whilst most cells have some ALDH activity, haemopoietic progenitor cells can be identi-
fied by flow cytometry from their lack of granularity and relatively high ALDH activity.
The subset containing haemopoietic progenitors is easily identified from the bivariate distribution of Aldefluor fluorescence and side scatter (see Section 3.3.1.3) [270, 271].
Therefore the approach is to replicate this methodology for detection of HSC in cord blood before testing its utility for MSC identification and sorting. The CD34 and CD45 antigens were used to independently identify haemopoietic progenitor cells.
The CD45 antigen is a tyrosine phosphatase, also known as the leukocyte common antigen (LCA). It is present on all human cells of haemopoietic origin, except erythroid
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 41 Chapter 3. Cell identification and isolation 3.1. Introduction cells, platelets and their precursor cells. The CD34 antigen has traditionally been used as a marker for immature haemopoietic cells [272]. It is expressed on the surface of a heterogeneous population of haemopoietic progenitors in terms of both lineage commitment and degree of differentiation [273], and as such, along with the CD45 antigen can be used to eliminate cells of haemopoietic origin.
The goal for this chapter is to develop robust methodology for isolation and detection of MSC from multiple species. The specific aims for this chapter are to:
• Replicate methodology for detecting HSC in cord blood using Aldefluor.
• Replicate methodology for assessing mesenchymal stem cell frequency using the
CFU-F assay.
• Estimate the frequency of CFU-F in cord blood.
• Isolate MSC from multiple species.
• Establish potential utility of Aldefluor substrate as a MSC marker across multiple
species.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 42 Chapter 3. Cell identification and isolation 3.2. Materials and Methods
3.2 Materials and Methods
3.2.1 Materials
Verapamil hydrochloride (062K0325), ammonium chloride (NH4Cl, A-4514), and Histo- paque1077 (10771-100ML) were obtained from Sigma-Aldrich, Sydney Australia. Alde-
fluor substrate and proprietary assay buffer were part of a kit manufactured by StemCo
Biomedical, Durham NC, and were obtained from StemCell Technologies (01700) through
Thermo Electron, VIC, in Australia.
Ethylenediaminetetra acetic acid (EDTA, 800682) was obtained from ICN Biomedicals,
Ohio USA, and Iscove’s Modified Dulbecco’s Medium (IMDM, 21056-023 (for cell label- ing), 12200-036 (for haemopoietic culture)) were obtained from Invitrogen Corporation,
Sydney Australia.
Monoclonal antibodies were obtained from Becton Dickinson Biosciences, Sydney Aus- tralia. CD34 (348057), CD45 (555483) and isotype control Mouse IgG1 (349043) were phycoerythrin [PE] conjugated, with CD45 (340953) and isotype control Mouse IgG1
(347202) peridinin chlorophyll protein - cyanin - 5.5 [PerCP-Cy5.5] conjugated.
Thrombopoeitin (TPO, 288-TPN), Stem Cell Factor (SCF, 255-SC-050), Fetal Liver
Tyrokinase 3 Ligand (FTL-3 Ligand, 308-FKN-025) and, Granulocyte Colony Stimu- lating Factor (G-CSF, 214-CS-025) were obtained from Bioscientific Pty Ltd, Gymea
Australia.
All other chemicals, reagents and tissue culture products were prepared and supplied as described in Appendix A.
3.2.2 Purification of hUCB
Whole hUCB samples donated by healthy consenting donors from the Australian Cord
Blood Bank were enriched for mononuclear cells by density gradient centrifugation and
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 43 Chapter 3. Cell identification and isolation 3.2. Materials and Methods ammonium chloride red cell lysis as follows.
10mL of blood was aliquoted in to four 50mL tubes and 18mL of room temperature
DPBS supplemented with 2mM EDTA, pH7.2 (washing buffer) was added to each tube and mixed well by inversion. Using a 10mL pipette, 12mL of room temperature
Histopaque1077 was layered under the diluted cord blood of each tube, and centrifuged for 35 minutes at 373g without break. The upper layer was aspirated without disturbing the buffy coat, and discarded. The buffy coat was collected from all tubes into a single
50mL tube. 20 - 25mL washing buffer was added then mixed and centrifuged for 10 minutes at 230g. The resulting pellet was resuspended with 40mL freshly made cold ammonium chloride solution (cold WFI was added to a 10 times stock solution of ammonium chloride immediately prior to use), and incubated on ice in the dark for 15 minutes. A washing step was repeated twice where cells were centrifuged for 10 minutes at 230g, and the resulting pellet was resuspended in 40mL washing buffer each time.
Cells were centrifuged a final time and resuspended in 10mL washing buffer and counted with a haemocytometer. The required number of cells was then used immediately.
3.2.3 Purification of porcine bone marrow
Porcine marrow was harvested from pigs as described in Appendix A.2.2.3, and enriched for mononuclear cells by density gradient centrifugation and ammonium chloride red cell lysis as follows.
Harvested sample (diluted bone marrow in saline and heparin) was aliquoted in to two
50mLtubesanddilutedwithDPBSsuchthataratioof1:2actualbonemarrow:DPBS was obtained and mixed well by inversion. Using a 10mL pipette, an equal amount of room temperature Histopaque1077 to bone marrow was layered under the diluted bone marrow of each tube, and centrifuged for 35 minutes at 373g without break. The upper layer was aspirated without disturbing the buffy coat, and discarded. The buffy coat was collected from both tubes into a single 50mL tube. 30 - 35mL washing buffer was added then mixed and centrifuged for 10 minutes at 230g. The resulting pellet was
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 44 Chapter 3. Cell identification and isolation 3.2. Materials and Methods resuspended with 40mL freshly made cold ammonium chloride solution (cold WFI was added to a 10 times stock solution of ammonium chloride immediately prior to use), and incubated on ice in the dark for 15 minutes. A washing step was repeated twice where cells were centrifuged for 10 minutes at 230g, and the resulting pellet was resuspended in 40mL washing buffer each time. Cells were centrifuged a last time and resuspended in 10mL washing buffer and counted with a haemocytometer. The required number of cells was then used immediately.
3.2.4 Purification of mouse bone marrow
Mouse marrow was harvested from mice as described in Appendix A.2.2.2, and enriched for mononuclear cells by density gradient centrifugation from a method modified from
[274] as follows.
The harvested sample from all twelve bones (six femur and six tibia) was combined and suspended in 5mL EMEM supplemented with 20% FBS in a 50mL tube. 25mL of
DPBS supplemented with 0.1% BSA was added and mixed with the marrow sample.
Using a 10mL pipette, approximately 14mL of room temperature Hisptopaque1077 was layered under the diluted bone marrow, and centrifuged for 35 minutes at 373g without brake.
A density gradient was not produced, so the top half of the sample (Tube 1) was pipetted into another 50mL tube (Tube 2), such that Tube 1 now contained the bottom half of the original sample, and Tube 2 contained the top half. 25mL of DPBS supplemented with 0.1% BSA was added into each tube and centrifuged at 373g for 15 minutes to pellet cells.
The resulting pellet in Tube 2 was white/grey in colour and formed in a neat capsule at the bottom of the tube. The pellet in Tube 1 was red in colour with many cells dispersed adhering to the side of the tube. All cells were taken from Tube 2 and only the cells in the centre were taken from Tube 1. Cells were placed in a 15mL tube and centrifuged at 190g for 10 minutes. The resulting pellet was suspended in 5mL
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 45 Chapter 3. Cell identification and isolation 3.2. Materials and Methods
EMEM supplemented with 20% FBS, and counted (ignoring red blood cells) with a haemocytometer. The required number of cells was then used immediately.
3.2.5 Aldefluor labeling
Dry Aldefluor reagent was activated as per manufacturers instructions, and frozen in either 5μLor10μL aliquots and stored in the dark at -20℃.
Frozen substrate aliquots were defrosted on ice in the dark immediately prior to use.
Aldefluor substrate at 0.122μg/mL was added to 1 x 106 cells/mL suspended in ei- ther proprietary Aldefluor assay buffer or IMDM containing 200μmol/L Verapamil and incubated in the dark at 37℃ for 30 to 60 minutes, unless otherwise stated. Excess substrate was removed after incubation by centrifuging cells at 190g for 10 minutes, and resuspending in either Aldefluor assay buffer, or EMEM supplemented with 20%
FBS. Analysis was performed immediately.
For each experiment, a control sample of unlabeled cells was used to establish auto
fluorescence. Aldefluor substrate was excited at 488nm, with gates used to exclude non-viable cells and debris.
3.2.6 Cord blood fibroblast colony forming assay
Human umbilical cord blood was purified as described in Section 3.2.2, and cells were plated out at the required densities in 35mm tissue culture plastic Petri dishes in 2mL ℃ of EMEM supplemented with 20% FBS in triplicates at 37 ,5%CO2 for a period of 15 days. Medium was replaced with fresh medium after 5 days, and then every 4 days until the cultures were fixed and stained with crystal violet as described in Appendix
A.3.3.1 on day 15. The resultant colonies were scored at a magnification of 7x, using an Olympus SZX12 stereomicroscope.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 46 Chapter 3. Cell identification and isolation 3.2. Materials and Methods
3.2.7 Rat bone marrow fibroblast colony forming assay
Rat bone marrow cells were harvested as described in Section A.2.2.1 from both femurs of three rats, to reduce inter animal variability, and isolated as described in Section
A.2.3.1. Equal numbers of cells from each of the 6 femurs were plated at densities of
2.5 x 106 and 5 x 106 in 100mm tissue culture plastic Petri dishes in 10mL of EMEM supplemented with 20% FBS in triplicate at 37℃,5%CO2 for a period of 15 days. Medium was replaced with fresh medium after 4 days. On the fifth day, medium was replaced with EMEM supplemented with 10% FBS, and then replaced again every
4 days. The cultures were fixed and stained with Alkaline Phosphatase (ALP) as described in Appendix A.3.3.2 on day 15. The resultant colonies were scanned using a HP Scanjet 4470c scanner with HP PrecisionScan Pro 3.14 software. The digitally scanned colonies were scored visually on ALP brightness.
3.2.8 Cell antibody labeling
Cells were labeled with PE-conjugated antibodies to CD34 and CD45, and PerCP-
Cy5.5-conjugated antibodies to CD45. Cells were centrifuged at 230g for 10 minutes, incubated on ice in the dark for 30 minutes with 20μLantibodyper1x106 cells. Cells were washed with 2mL DPBS supplemented with 2% FBS twice by centrifugation at 230g for 10 minutes, resuspended in Aldefluor assay buffer and kept on ice until required. Cells were further labeled with Aldefluor, if required as per Section 3.2.5.
Appropriate isotype controls were used to determine the negative fraction
3.2.9 Analysis and Sorting
Labeled cells were either analysed on a FACSort™ (BDIS), equipped with a 15mW argon laser tuned to 488nm, or sorted on a MoFlo sorter equipped with a 20mW air cooled argon laser tuned to 488nm. Both machines had standard sets of filters for red and green fluorescence. Data analysis was performed using Cytomation Summit.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 47 Chapter 3. Cell identification and isolation 3.3. Results
3.3 Results
3.3.1 Isolation of HSC and MSC from cord blood
3.3.1.1 Aldefluor substrate titration
A titration of Aldefluor was conducted with varying concentrations of substrate to determine the optimal labeling concentration in our system.
1x106 mononuclear hUCB cells/mL were labeled with concentrations ranging from
0.122μg/mL to 1.22μg/mL Aldefluor substrate as described in Section 3.2.5. Cord blood was purified as described in Section 3.2.2 and the resulting flow cytometric plots are shown in Figure 3.1.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 48 Chapter 3. Cell identification and isolation 3.3. Results
(a) 0.122μg/mL Aldefluor substrate (b) 0.610μg/mL Aldefluor substrate
(c) 0.244μg/mL Aldefluor substrate (d) 0.915μg/mL Aldefluor substrate
(e) 0.488μg/mL Aldefluor substrate (f) 1.22μg/mL Aldefluor substrate
Figure 3.1: Mononuclear hUCB cells labeled with Aldefluor substrate. Aldefluor intensity vs side scatter.
The low orthogonal light scattering and bright fluorescence region, defined as SSClo
ALDHbr [270] has characteristic populations for hUCB as seen in Figure 3.1. It was also observed (data not shown) that with increasing substrate concentrations, the univariate histogram showing Aldefluor fluorescence does not change shape, but rather only results in a shift towards the right along the Aldefluor axis.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 49 Chapter 3. Cell identification and isolation 3.3. Results
It may be inferred from this that at these levels Aldefluor substrate does not appear to be immediately toxic as there was no noticeable change in cell size or granularity, and a substrate concentration of 0.122μg/mL is sufficient to label cells, and consequentially has been used for all further labeling, unless otherwise stated.
3.3.1.2 Incidence of CFU-F in cord blood
A colony forming unit study was carried out on mononuclear human umbilical cord blood as described in Section 3.2.6 using the densities of 1 x 105,5x105,1x106 and
5x106 cells per plate.
After 15 days of culture, cell seeding densities of 1 x 105,5x105 and 1 x 106 were found to be too low, as they did not produce any colonies. The cell seeding density of 5 x 106 was found to be too high, with cells that initially formed discrete colonies spreading into neighbouring colonies by day 15. As a result of this merging of colonies, it was difficult to distinguish between the original discrete clusters.
The CFU-F study was repeated with mononuclear hUCB cells and plated out at den- sities of 1 x 106,1.5x106,2x106,3x106 and 4 x 106 as described in Section 3.2.6, with the variation that medium was replaced with fresh medium after 6 days, and then every 3 days for the duration of the culture period. Results are shown in Table 3.1, with a representative colony shown in Figure 3.2. Figure 3.3 shows the effect of plating density on plating efficiency.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 50 Chapter 3. Cell identification and isolation 3.3. Results
Table 3.1: CFU-F assay of mononuclear hUCB cells grown for 15 days in EMEM supplemented with 20% FBS in 35mm dishes. Medium was replaced with fresh medium after 6 days, and then replaced every 3 days. Cultures were fixed and stained with crystal violet at day 15, and colonies were scored at a magnification of 7x.
Plating density Average Colony Number (cells/plate) (standard deviation), per 3 plates 1x106 0.67 (0.47) 1.5 x 106 0.33 (0.47) 2x106 6.67 (3.4) 3x106 23.33 (8.06)1 4x106 25.00 (5.66)1
1Due to the close proximity of colonies, a slight error in counting may have occurred.
Figure 3.2: Representative mononuclear hUCB cell colony at day 15, 40x. Cells were seeded at a density of 2 x 106 in 35mm dishes and grown for 15 days in EMEM supplemented with 20% FBS. Medium was replaced with fresh medium after 6 days, and then replaced every 3 days. Cultures were fixed and stained with crystal violet at day 15.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 51 Chapter 3. Cell identification and isolation 3.3. Results
8
7
6 ) -6
5
4
3 Plating efficiency (x10 2
1
0 0 0.5 1 1.5 2 2.5 3 3.5 4 Plating density (x106 cells/plate)
Figure 3.3: CFU-F assay of mononuclear hUCB cells grown for 15 days in EMEM supple- mented with 20% FBS in 35mm dishes. Plating efficiency vs plating density. Medium was replaced with fresh medium after 6 days, and then replaced every 3 days. Cultures were fixed and stained with crystal violet at day 15, and colonies were scored at a magnification of 7x.
Table 3.1 illustrates that upon being plated at the correct density, mononuclear hUCB cells formed discrete adherent colonies. If the plating density was too low, colonies were not observed, and if the plating density was too high, then the resultant colonies merged into one another and were difficult to count. Plating out at an even higher density resulted in cells adhering and proliferating in more of a monolayer like manner, rather than with the formation of discrete colonies.
Figure 3.2 indicates that these colonies are comprised of cells that are somewhat spindle shaped. These results indicate that mononuclear hUCB comprises a population of somewhat spindle shaped cells that are both adherent and colony forming in nature, both characteristic of mesenchymal precursor cells though it would be necessary to confirm that these cell can form bone, cartilage and fat to confirm this hypothesis.
Figure 3.3 demonstrates that as the plated cell number increases, so does the plating efficiency.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 52 Chapter 3. Cell identification and isolation 3.3. Results
3.3.1.3 Identification of haemopoietic progenitor cell markers within SSClo
ALDHbr subset
Figure 3.4 shows the gate used to identify putative haemopoietic progenitor cells.
HUCB cells were purified as described in Section 3.2.2, labelled with Aldefluor sub- strate as described in Section 3.2.5 and analysed as described in Section 3.2.9.
A B
Figure 3.4: Mononuclear hUCB labeled with Aldefluor substrate. Left hand plot shows forward vs side scatter, right hand plot shows Aldefluor intensity vs side scatter.
The circled region of R1 seen in Figure 3.4B is the characteristic SSClo ALDHbr pop- ulation previously reported in literature [270, 271] and comprises 0.89% of the total number of events. This region of interest is back gated in Figure 3.4A (pink events) to indicate the size (forward scatter) and granularity (side scatter) of the population being investigated for progenitor cells of mesenchymal origin. It should be noted that this is a very small fraction of the entire population.
The incidence of CD34+ cells in small mononuclear cells (R1) is shown in Figure 3.5A and B.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 53 Chapter 3. Cell identification and isolation 3.3. Results
A B
Figure 3.5: Mononuclear hUCB labeled with CD34 PE. Left hand plot shows forward vs side scatter, right hand plot shows the gated region in A with forward scatter vs CD34 intensity.
The gated region of R1 in Figure 3.5A is 67.1% of the entire cell population. Only events from this gated region are shown in Figure 3.5B, with 1.7% being CD34+ (region
R3) and 98.3% CD34neg (region R5). It can be seen that there is only a very small percentage of cells that are CD34+.
HUCB cells were purified as described in Section 3.2.2, labeled with a monoclonal antibody directed against CD45 as described in Section 3.2.8, labeled with Aldefluor substrate as described in Section 3.2.5, and analysed as described in Section 3.2.9. The resulting plots are shown in Figure 3.6, with the isotype control for CD45 shown in
Figure 3.7.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 54 Chapter 3. Cell identification and isolation 3.3. Results
A B
Figure 3.6: Mononuclear hUCB labeled with Aldefluor substrate and CD45 PE. Left hand plot shows Aldefluor intensity vs side scatter, right hand plot shows the gated region in A with Aldefluor vs CD45 intensity.
A B
Figure 3.7: Mononuclear hUCB labeled with CD45 PE. Left hand plot shows forward vs side scatter, right hand plot shows the gated region in A with forward scatter vs CD45 intensity.
The gated region of R1 in Figure 3.6A is 1.1% of the entire cell population. Only events from this gated region are shown in Figure 3.6B, with 99.6% being ALDHbr CD45+
(region R3) and 0.4% ALDHbr CD45neg (region R5).
MSC would be ALDHbr but CD45neg (region R5 Figure 3.6B). If these cells are also
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 55 Chapter 3. Cell identification and isolation 3.3. Results
CD34neg, then they potentially may be mesenchymal in origin, as they express high levels of ALDH.
HUCB cells were purified as described in Section 3.2.2, labeled with a monoclonal antibody directed against CD34 as described in Section 3.2.8, labeled with Aldefluor substrate as described in Section 3.2.5, and analysed as described in Section 3.2.9. The resulting plots are shown in Figure 3.8.
A B
Figure 3.8: Mononuclear hUCB labeled with Aldefluor substrate and CD34 PE. Left hand plot shows Aldefluor intensity vs side scatter, right hand plot shows the gated region in A with Aldefluor vs CD34 intensity.
The gated region of R1 in Figure 3.8A is 1.2% of the entire cell population. Only events from this gated region are shown in Figure 3.8B, with 49.1% being ALDHbr CD34+
(region R3) and 50.9% ALDHbr CD34neg (region R5).
To summarise, the SSClo ALDHbr cells of cord blood comprises at least two subpopu- lations as distinguished by CD34 positivity and CD45 antigen expression. The CD45 negative subfraction has a very low incidence (<0.4%).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 56 Chapter 3. Cell identification and isolation 3.3. Results
3.3.1.4 Haemopoietic progenitors but not MSC found in SSClo ALDHbr sub-
set
HUCB cells were purified as described in Section 3.2.2. These purified cells were then either labeled with a monoclonal antibody directed against CD34 as described in Sec- tion 3.2.8, and then labeled with Aldefluor substrate as described in Section 3.2.5; or labeled with a monoclonal antibody directed against CD45 as described in Section 3.2.8.
Labeled cells purified using a MoFlo high speed sorter (Section 3.2.9) to derive highly purified subsets of ALDHbr SSClo cells. These were ALDHbr, ALDHbr CD34+, ALDHbr
CD34neg,CD45neg,andCD45farneg populations. The sorting strategy is illustrated in
Figures 3.9 to 3.12.
Isolating ALDHbr fraction
A B
Figure 3.9: Mononuclear hUCB labeled with Aldefluor substrate. Left hand plot shows forward scatter vs side scatter, right hand plot shows the gated region R1 in plot A with Aldefluor (FL1) vs side scatter.
The gated region of R1 in Figure 3.9A is 93.4% of the entire cell population. Events with a forward scatter much lower than 64 have been excluded, as they are most likely non-viable cells or debris. Events from the gated region R1 in Figure 3.9A are shown in
Figure 3.9B, with 2.0% being ALDHbr (region R2). R1 and R2 are combined to define
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 57 Chapter 3. Cell identification and isolation 3.3. Results the SSClo ALDHbr population to be sorted with respect to CD45 and CD34 antigen expression.
Isolating ALDHbr CD34+ and ALDHbr CD34neg fraction
Figure 3.10: Mononuclear hUCB labeled with Aldefluor substrate and CD34 PE. Aldefluor (FL1) vs CD34 (FL2) intensity.
11 x 103 ALDHbr CD34neg cells were collected from region R6 in Figure 3.10, and 28 x
103 ALDHbr CD34+ cells from region R4 in Figure 3.10.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 58 Chapter 3. Cell identification and isolation 3.3. Results
Isolating CD45neg CD45farneg fraction
A B
Figure 3.11: Mononuclear hUCB labeled with CD45 PE. Left hand plot shows forward scatter vs side scatter, right hand plot shows forward scatter vs CD45 (FL2) intensity.
0.2% 4.4%
Figure 3.12: Fluorescence histogram of CD45 PE (FL2) showing percentage of CD45 PE labeled hUCB.
The gated region of R1 in Figure 3.11A is 89.5% of the entire cell population. Events with a forward scatter much lower than 64 have been excluded, as they are most likely non-viable cells or debris. Events from the gated region R1 in Figure 3.11A are shown in Figure 3.11B, with 99.9% being CD45+ (region R4), and 0.1% being CD45neg (region
R6). Figure 3.12 shows that 4.4% of the events in Figure 3.11A region R1 are CD45neg
(region R8) with 401 x 103 cells being collected, and 0.2% are CD45farneg (region R7) with 4 x 103 cells being collected.
All cell fractions are summarised in Table 3.2.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 59 Chapter 3. Cell identification and isolation 3.3. Results
Table 3.2: Isolated mononuclear hUCB cells labeled with Aldefluor substrate, Aldefluor sub- strate and CD34, or CD45.
Fraction Number of cells Figure isolated ALDHbr 55 x 106 Region R2 Figure 3.9B ALDHbr CD34+ 28 x 106 Region R4 Figure 3.10 ALDHbr CD34neg 11 x 106 Region R6 Figure 3.10 CD45neg 401 x 106 Region R8 Figure 3.12 CD45farneg 4x106 Region R7 Figure 3.12
The fractionated cells from Section 3.3.1.4 were cultured under both mesenchymal and haemopoietic optimised medium conditions. Mesenchymal medium conditions con- sisted of EMEM supplemented with 20% FBS, and haemopoietic medium conditions consisted of IMDM supplemented with 20% FBS with 100ng/mL of TPO, SCF, FTL-3
Ligand and, G-CSF. It should be noted that a haemopoietic colony forming assay was not conducted in semisolid media, but rather a cytokine-driven haemopoietic expansion culture was carried out2.
Each group of fractionated cells were split into two sets, with one half being cultured under mesenchymal conditions, and the other under haemopoietic conditions. All frac- tionated cells were plated out into a single 35mm tissue culture dish, with the exception of the ALDHbr fraction which was plated out into two 35mm tissue culture dishes, with
2mL of medium at 37℃,5%CO2 for a period of 15 days.
Mesenchymal medium was replaced with fresh medium after 6 days, and then every
3 days until the cultures were fixed and stained with crystal violet as described in
Appendix A.3.3.1 on day 15. Any resulting mesenchymal colonies were scored at a magnification of 7x, using an Olympus SZX12 stereomicroscope. For the haemopoi- etic cultures, medium was not changed during the 15 days of growth due to the low number of cells. The haemopoietic cells of interest are non-adherent in nature and a medium change would require centrifuging, resulting in loss of cells from an already
2Kap-Hyoun Ko, PhD candidate, GSBmE, UNSW
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 60 Chapter 3. Cell identification and isolation 3.3. Results low population.
After mesenchymal cultures were stained with crystal violet, very few if any adherent cells were visible. No colonies were found at a magnification of 7x viewed with an
Olympus SZX12 stereomicroscope in of any of the mesenchymal culture fractions.
As anticipated, fractions thought to contain haemopoietic progenitor cells, ALDHbr and ALDHbr CD34+ did well under haemopoietic culture conditions, as did cells from the CD45neg fraction, indicating that this region was still picking up cells that were
CD45+, which was the basis for including a fractionation of the CD45farneg region, to capture truly CD45neg cells.
A: 5 d B: 8 d C: 12 d
Figure 3.13: Isolated mononuclear hUCB ALDHbr CD34neg fractionation cultured under mes- enchymal optimised media conditions in 35mm dishes, 20x. Picture A is taken at 5 days, picture B at 8 days and picture C at 12 days.
Figure 3.13 shows that fractionated cells remained viable after isolation, but gradually deteriorated over the growing time and at day 15 no significant cell numbers were present. This trend of cell death with increasing time was observed for all fractions, not just the ALDHbr CD34neg group presented in Figure 3.13.
These findings suggests that on this particular occasion, progenitor cells of mesenchy- mal origin, if isolated from mononuclear hUCB using Aldefluor substrate as an indicator of ALDH activity, were unable to be cultured successfully under the given media con- ditions. If present in cord blood, MSC were not co-purified with HSC in the SSClo
ALDHbr subsets.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 61 Chapter 3. Cell identification and isolation 3.3. Results
3.3.2 Isolation of MSC from Rat
3.3.2.1 CFU-F Assay
A colony forming unit study was carried out as per Section 3.2.7 on rat bone marrow cells to verify that the isolation procedure was capturing a portion of mesenchymal progenitor cells present in bone marrow. These progenitor cells are clonogenic in nature and should produce colonies if plated at the correct density.
Average colony numbers are given in Table 3.3, with representative colonies for both plating densities are given in Figure 3.14.
Table 3.3: CFU-F assay of whole rat bone marrow cells grown for 15 days. Cells were plated out in EMEM supplemented with 20% FBS in 35mm dishes. Medium was replaced with fresh medium after 4 days, and on the fifth day was replaced with EMEM supplemented with 10% FBS, and replaced every 4 days. Cultures were fixed and stained with Alkaline Phosphatase at day 15, and colonies digitally scanned and scored.
Plating density Average Colony Number (cells/plate) (standard deviation), per 3 plates 2.5 x 106 52 (9.29) 5x106 58 (4.04)
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 62 Chapter 3. Cell identification and isolation 3.3. Results
A B
Figure 3.14: Digital images of representative plates of whole rat bone marrow cell colonies at day 15 expressing ALP. Image A plated at a density of 2.5 x 106, image B plated at a density of 5.5 x 106 cells.
Whole bone marrow cells isolated from rat femurs successfully formed colonies when plated out at densities of 2.5 x 106 and 5 x 106 cells in 100mm Petri dishes, reflecting that the isolation procedure (of flushing the bone marrow from the femurs and centrifuging as per Appendix A.2.3.1) is capturing progenitor cells of mesenchymal origin.
3.3.2.2 Aldefluor labeling of rat bone marrow
CFU-F were not detected in Aldefluor substrate labeled fractions sorted from cord blood. The starting frequency of CFU-F is only around 1 : 1 x 106 cord blood cells, and it is likely that sorting did not provide enough CFU-F to be detected (Table 3.2).
Therefore rat bone marrow was studied because it has an approximately 25-fold higher plating efficiency (Table 3.3).
Rat bone marrow was harvested as described in Section A.2.2.1, labeled with Aldefluor substrate as described in Section 3.2.5, with the cells being incubated in 0.5mL of buffer, and analysed as described in Section 3.2.9. The resulting flow cytometric plots of labeled cells are shown in Figures 3.15 and 3.16.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 63 Chapter 3. Cell identification and isolation 3.3. Results
(a) Unlabeled rat whole bone mar- (b) Unlabeled rat whole bone mar- row cells row cells
(c) Labeled rat whole bone marrow (d) Labeled rat whole bone marrow cells cells
Figure 3.15: Flow cytometric analysis of rat whole bone marrow cells labeled with Aldefluor substrate. Left hand panels show forward vs side scatter, right hand panels show Aldefluor intensity vs side scatter. Plots A and B are unlabeled cells, plots C and D are labeled with Aldefluor.
A: Unlabeled 1.2%
B: Labeled 98.5%
Figure 3.16: Fluorescence histogram of Aldefluor showing percentage of Aldefluor substrate labeled rat whole bone marrow cells.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 64 Chapter 3. Cell identification and isolation 3.3. Results
It can be seen from Figure 3.15 plots A and C that labeling with Aldefluor substrate does not affect cell size or granularity as there is no change in either the forward scatter or side scatter distribution of cells. Labeling does however produce a shift along the
Aldefluor axis as illustrated in Figure 3.15 plots B and D. It is noted that passaged plastic adherent marrow cells were also Aldefluor substrate positive, but had a markedly different Aldefluor staining profile compared to fresh rat marrow (Figure 3.17).
The gated region of R1 in Figure 3.16 shows the percentage of events which are Aldefluor positive. Region R1 is set by comparing the plots of Aldefluor substrate labeled cells with unlabeled cells. 98.5% of rat whole bone marrow was found to be ALDH+ as seen in Figure 3.16 region R1.
A cluster of SSClo ALDHbr cells (the circled region in Figure 3.15d) can be seen in rat whole bone marrow, but this population is not as distinct a subset as seen in cord blood, when labeled with Aldefluor substrate.
3.3.2.3 Aldefluor labeling of plastic adherent rat bone marrow cells and
the effect of passage
The effect of passage on ALDH activity was investigated using tissue culture plastic adherent rat bone marrow cells.
Rat bone marrow was harvested as described in Section A.2.2.1, isolated as described in Section A.2.3.1, and maintained as described in Section A.2.4.2 until the desired passage number was reached. The resulting passage 0, 1, 2, and 3 cells were labeled with Aldefluor substrate as described in Section 3.2.5, with the cells being incubated in either 0.5 or 1mL of buffer, and analysed as described in Section 3.2.9. The resulting
flow cytometric plots of labeled cells are shown in Figure 3.17.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 65 Chapter 3. Cell identification and isolation 3.3. Results
(a) Unlabeled rat passage 0 cells (b) Labeled rat passage 0 cells
(c) Unlabeled rat passage 1 cells (d) Labeled rat passage 1 cells
(e) Unlabeled rat passage 2 cells (f) Labeled rat passage 2 cells
(g) Unlabeled rat passage 3 cells (h) Labeled rat passage 3 cells
Figure 3.17: Flow cytometric analysis of rat passage 0, 1, 2 and 3 tissue culture plastic ad- herent bone marrow cells labeled with Aldefluor substrate. Aldefluor intensity vs side scatter. Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 66 Chapter 3. Cell identification and isolation 3.3. Results
Figures 3.15 and 3.17 illustrate that rat whole bone marrow, passage 0, 1, 2 and 3 cells all contain a population of ALDH+ cells that were identified with Aldefluor substrate.
The SSClo ALDHbr cluster seen with Aldefluor labeling of rat whole bone marrow
(Figure 3.15d) has disappeared with the passaged cells (Figure 3.17).
Mean fluorescence ratios were calculated from fluorescence histograms (data not shown) by dividing mean fluorescence of labeled cells by mean fluorescence of unlabeled cells, with the results shown in Table 3.4. In order to determine the effect of passage on
Aldefluor labeling, analysis of variance (ANOVA) was carried out on mean fluorescence ratios from passage 0 to passage 3. No significant difference in staining between passages was observed (p>0.05) for rat cells. Mean and standard error of mean fluorescence ratios for passage 0 to passage 3 were 6.17 and 0.69 respectively. The staining of bone marrow was much greater than plastic adherent cells (35.44 versus 6.17 ± 0.69).
Table 3.4: Mean fluorescence ratio of Aldefluor substrate labeled rat cells.
Cell Type Mean Fluorescence Ratio Bone marrow 35.44 Passage 0 4.73 Passage 1 6.67 Passage 2 5.46 Passage 3 7.86
3.3.3 Isolation of MSC by plastic adherence from other species
3.3.3.1 Mouse Marrow
Mouse bone marrow was harvested as described in Appendix A.2.2.2, purified as de- scribed in Section 3.2.4, labeled with Aldefluor substrate as described in Section 3.2.5, and analysed as described in Section 3.2.9. The resulting flow cytometric plots of labeled cells are shown in Figures 3.18 and 3.19.
Several attempts were also made to grow mouse marrow on tissue culture plastic as
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 67 Chapter 3. Cell identification and isolation 3.3. Results described in Appendix A.2.4.2 for rat, but the marrow failed to adhere and proliferate successfully. As a consequence, Aldefluor labeling was only carried out on mouse bone marrow.
3.3.3.2 Mouse L929 Fibroblasts
Mouse L929 fibroblasts were used to establish a negative control. These cells are derived from an immortalised cell line and should not express Aldefluor substrate.
Mouse L929 Fibroblasts were maintained as described in Appendix A.2.5, labeled with
Aldefluor substrate as described in Section 3.2.5, and analysed as described in Section
3.2.9. The resulting flow cytometric plots of labeled cells are shown in Figure 3.20 and
3.21.
3.3.3.3 Porcine
The effect of passage on ALDH activity was investigated using porcine bone marrow and tissue culture plastic adherent porcine bone marrow cells.
Porcine bone marrow was harvested as described in Appendix A.2.2.3, purified as de- scribed in Section 3.2.3, labeled with Aldefluor substrate as described in section 3.2.5, and analysed as described in Section 3.2.9.
Porcine bone marrow was harvested as described in Appendix A.2.2.3, isolated as de- scribed in Appendix A.2.3.3, and maintained as described in Appendix A.2.4 until the desired passage number was reached. The resulting passage 0, 1, and 2 cells were la- beled with Aldefluor substrate as described in Section 3.2.5 and analysed as described in Section 3.2.9.
The resulting flow cytometric plots of labeled cells are shown in Figures 3.22 to 3.24.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 68 Chapter 3. Cell identification and isolation 3.3. Results
3.3.3.4 Equine
The effect of passage on ALDH activity was investigated using equine bone marrow and tissue culture plastic adherent equine bone marrow cells.
Equine bone marrow was harvested as described in Appendix A.2.2.4, labeled with
Aldefluor substrate as described in section 3.2.5, and analysed as described in Section
3.2.9.
Equine bone marrow was harvested as described in Appendix A.2.2.4, isolated as de- scribed in Appendix A.2.3.3, and maintained as described in Appendix A.2.4 until the desired passage number was reached. The resulting passage 1 and 3 cells were labeled with Aldefluor substrate as described in Section 3.2.5, and analysed as described in
Section 3.2.9.
The resulting flow cytometric plots of labeled cells are shown in Figures 3.25 to 3.27.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 69 Chapter 3. Cell identification and isolation 3.3. Results
3.3.4 ALDH staining profile of marrow and MSC from other species
3.3.4.1 Mouse
(a) Unlabeled mouse whole bone (b) Unlabeled mouse whole bone marrow cells marrow cells
(c) Labeled mouse whole bone mar- (d) Labeled mouse whole bone mar- row cells row cells
Figure 3.18: Flow cytometric analysis of mouse whole bone marrow cells labeled with Alde- fluor substrate. Left hand panels show forward vs side scatter, right hand panels show Aldefluor intensity vs side scatter. Plots A and B are unlabeled cells, plots C and D are labeled with Aldefluor.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 70 Chapter 3. Cell identification and isolation 3.3. Results
A: Unlabeled 3.4%
B: Labeled 39.5%
Figure 3.19: Fluorescence histogram of Aldefluor showing percentage of Aldefluor substrate labeled mouse whole bone marrow cells.
As with the rat data, it can be seen from Figure 3.18 plots A and C that labeling with
Aldefluorsubstratedoesnotaffectcellsizeorgranularityasthereisnochangeineither the forward scatter or side scatter distribution of cells. Labeling does however produce a shift along the Aldefluor axis as illustrated in Figure 3.18 plots B and D.
39.5% of mouse whole bone marrow was found to be ALDHbr as seen in Figure 3.19 region R1, set by comparing the plots of Aldefluor substrate labeled cells with unlabeled cells (mean fluorescent ratio of labeled cells to unlabled cells was calculated as 5.15).
When compared to 98.5% obtained for rat marrow, this may indicate that animal age could be a factor affecting both ALDH activity levels, and the abundance of progenitor cells, seeing as the rat bone marrow was harvested from 4-5 week old rats, whilst mouse marrow was harvested from several month old mice. Progenitor cell numbers may also vary with species.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 71 Chapter 3. Cell identification and isolation 3.3. Results
3.3.4.2 Mouse L929 Fibroblasts
(a) Unlabeled mouse L929 fibrob- (b) Unlabeled mouse L929 fibrob- lasts lasts
(c) Labeled mouse L929 fibroblasts (d) Labeled mouse L929 fibroblasts
Figure 3.20: Flow cytometric analysis of mouse L929 fibroblasts labeled with Aldefluor sub- strate. Left hand panels show forward vs side scatter, right hand panels show Aldefluor intensity vs side scatter. Plots A and B are unlabeled cells, plots C and D are labeled with Aldefluor.
A: Unlabeled 0.8%
B: Labeled 2.3%
Figure 3.21: Fluorescence histogram of Aldefluor showing percentage of Aldefluor substrate labeled mouse L929 fibroblasts.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 72 Chapter 3. Cell identification and isolation 3.3. Results
As already established for rat and mouse, comparing plots A and C from Figure 3.20 illustrate that labeling with Aldefluor substrate does not affect cell size or granularity as there is no change in either the forward scatter or side scatter distribution of cells.
However, unlike the previously presented data, labeling with Aldefluor substrate does not produce a shift along the Aldefluor axis in plots B and C of Figure 3.20.
The small change in region R1 of Figure 3.21, upon comparing labeled to unlabeled events, is attributed to background staining (mean fluorescent ratio of labeled cells to unlabled cells was calculated as 1.37), not the presence of ALDHbr cells in the population. From the lack of ALDHbr cells in both Figures 3.20 and 3.21, it can be presumed that mouse L929 fibroblasts do not express ALDH.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 73 Chapter 3. Cell identification and isolation 3.3. Results
3.3.4.3 Porcine
(a) Unlabeled porcine mononuclear (b) Unlabeled porcine mononuclear bone marrow cells bone marrow cells
(c) Labeled porcine mononuclear (d) Labeled porcine mononuclear bone marrow cells bone marrow cells
Figure 3.22: Flow cytometric analysis of porcine mononuclear bone marrow cells labeled with Aldefluor substrate. Left hand panels show forward vs side scatter, right hand panels show Aldefluor intensity vs side scatter. Plots A and B are unlabeled cells, plots C and D are labeled with Aldefluor.
A: Unlabeled 1.2%
B: Labeled 99.9%
Figure 3.23: Fluorescence histogram of Aldefluor showing percentage of Aldefluor substrate labeled porcine mononuclear bone marrow cells.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 74 Chapter 3. Cell identification and isolation 3.3. Results
Labeling with Aldefluor substrate produces a shift along the Aldefluor axis as illustrated in Figure 3.22 plots B and D. Both these trends continued for porcine passage 0, passage
1 and passage 2 (Figure 3.24) cell samples.
99.9% of porcine mononuclear bone marrow was found to be ALDHbr as illustrated in
Figure 3.23, region R1, set by comparing the plots of Aldefluor substrate labeled cells with unlabeled cells. This is virtually the entire population that was labeled, and unlike the rat bone marrow sample which was not depleted of red blood cell contamination, the porcine marrow had both a density gradient separation and red blood cell purification with ammonium chloride performed.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 75 Chapter 3. Cell identification and isolation 3.3. Results
(a) Unlabeled porcine passage 0 cells (b) Labeled porcine passage 0 cells
(c) Unlabeled porcine passage 1 cells (d) Labeled porcine passage 1 cells
(e) Unlabeled porcine passage 2 cells (f) Labeled porcine passage 2 cells
Figure 3.24: Flow cytometric analysis of porcine passage 0, 1, and 2 tissue culture plastic adherent bone marrow cells labeled with Aldefluor substrate. Aldefluor intensity vs side scatter.
Mean fluorescence ratios were calculated from fluorescence histograms (data not shown), with the results shown in Table 3.5. In order to determine the effect of passage on
Aldefluor labeling, analysis of variance was carried out on mean fluorescence ratios from passage 0 to passage 2. No significant difference in staining between passages was observed (p>0.05) for porcine cells. Mean and standard error of mean fluorescence
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 76 Chapter 3. Cell identification and isolation 3.3. Results ratios for passage 0 to passage 2 were 4.38 and 0.87 respectively. The staining of bone marrow was much greater than plastic adherent cells (50.81 versus 4.38 ± 0.87).
Table 3.5: Mean fluorescence ratio of Aldefluor labeled porcine cells.
Cell Type Mean Fluorescence Ratio Bone marrow 50.81 Passage 0 3.49 Passage 1 6.13 Passage 2 3.51
It can be seen from Figures 3.22 and 3.24 that porcine mononuclear bone marrow, passage 0, 1 and 2 cells all contain a population of ALDH+ cells that were identified with Aldefluor substrate. As previously seen with rat whole bone marrow, passaged cells have a lower potential for labeling with Aldefluor substrate than the mononuclear bone marrow, as highlighted in Table 3.5.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 77 Chapter 3. Cell identification and isolation 3.3. Results
3.3.4.4 Equine
(a) Unlabeled equine bone marrow (b) Unlabeled equine bone marrow cells cells
(c) Labeled equine bone marrow (d) Labeled equine bone marrow cells cells
Figure 3.25: Flow cytometric analysis of equine mononuclear bone marrow cells labeled with Aldefluor substrate. Left hand panels show forward vs side scatter, right hand panels show Aldefluor intensity vs side scatter. Plots A and B are unlabeled cells, plots C and D are labeled with Aldefluor.
A: Unlabeled 1.9%
B: Labeled 24.7%
Figure 3.26: Fluorescence histogram of Aldefluor showing percentage of Aldefluor substrate labeled equine whole bone marrow.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 78 Chapter 3. Cell identification and isolation 3.3. Results
Figure 3.25 quite clearly illustrates that there is a discrepancy between plots A and C.
This indicates that there has been a change in cell size and granularity as a result of labeling with Aldefluor substrate. This change has not been observed with any other species data set, and it is suggested that this an outlying data set that is not truly representative of the effect of labeling with Aldefluor substrate, especially given that this effect is not present with the equine tissue culture plastic selected passage 1 and 3 cell samples. There should be minimal or little variation in scatter morphology resulting from Aldefluor labeling. Due to the unavailability of primary cell samples, Aldefluor labeling on equine marrow could not be repeated to verify this result. It should be noted that marrow sample was not enriched for mononuclear cells by density gradient centrifugation and ammonium chloride red cell lysis.
Labeling does produce the previously seen shift along the Aldefluor axis as illustrated in Figure 3.25 plots B and D. This trend is continued for equine passage 1 and passage
3 (Figure 3.27) cell samples.
The gated region of R1, set by comparing the plots of Aldefluor substrate labeled cells with unlabeled cells in Figure 3.26 shows 24.7% of events are Aldefluor substrate bright.
This proportion is similar to that obtained for mouse marrow.
The small spike seen at the beginning of the Aldefluor spectrum in Figure 3.26, is part of the main group of cells in Figure 3.25, falling in the main distribution and is not a separate distinct population.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 79 Chapter 3. Cell identification and isolation 3.3. Results
(a) Unlabeled equine passage 1 cells (b) Labeled equine passage 1 cells
(c) Unlabeled equine passage 3 cells (d) Labeled equine passage 3 cells
Figure 3.27: Flow cytometric analysis of equine passage 1 and 3 tissue culture plastic adherent bone marrow cells labeled with Aldefluor substrate. Aldefluor intensity vs side scatter.
Mean fluorescence ratios were calculated from fluorescence histograms (data not shown), with the results shown in Table 3.5. In order to determine the effect of passage on
Aldefluor labeling, analysis of variance was carried out on mean fluorescence ratios of passage 1 and passage 3. No significant difference in staining between passages was ob- served (p>0.05) for equine cells. Mean and standard error of mean fluorescence ratios for passages 1 and 3 were 9.15 and 0.08 respectively. The staining of bone marrow was lower than plastic adherent cells (2.56 versus 9.15 ± 0.08).
Table 3.6: Mean fluorescence ratio of Aldefluor labeled equine cells.
Cell Type Mean Fluorescence Ratio Bone marrow 2.56 Passage 1 9.06 Passage 3 9.22
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 80 Chapter 3. Cell identification and isolation 3.3. Results
It can be seen from Figures 3.25 to 3.27 and Table 3.6 that equine bone marrow, passage1 and 3 cells all contain a population of ALDHbr cells that were identified with
Aldefluor substrate.
The bone marrow sample did not contain as many ALDHbr cells as compared to the rat and porcine data, and both tissue culture plastic adherent passage 1 and 3 samples contained larger proportions of ALDHbr cells. It is entirely plausible that this is related to the variation between plots A and C in Figure 3.25, with perhaps that the bone marrow sample was not representative of the results that could be obtained from equine bone marrow. Due to the unavailability of primary cell samples, Aldefluor labeling on equine marrow could not be repeated to verify this surprising result.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 81 Chapter 3. Cell identification and isolation 3.4. Discussion
3.4 Discussion
One of the main driving aims of this work was to study a potential method for prospec- tive isolation of progenitor cells, rather than the currently utilised retrospective isolation techniques used for mesenchymal stem cells. Tissue culture plastic adherence is one such method where adherent cells are deemed to be mesenchymal in origin, but as a culture method, is likely to alter MSC biology.
Cell selection based on the use of monoclonal antibodies for MSC is focused around both positive and negative selection. This method fractionates out all known cells and then assumes that what are left are mesenchymal progenitor cells.
BODIPY aminoacetaldehyde was developed by Storms et al [270] as a means of iso- lating haemopoietic progenitor cells that express aldehyde dehydrogenase. Substrate functionality was initially tested using human umbilical cord blood, chosen by the au- thors due its rising potential as an alternative source of transplantable HSC to bone marrow. Storms et al demonstrated that a population of cells with low orthogonal light scattering properties were present in hUCB that stained brightly with BAAA when combined with an efflux inhibitor on the basis of ALDH activity and were conse- quentially shown to be enriched for primitive haemopoietic progenitor cells.
The established performance link between ALDH activity and Aldefluor substrate in hUCB haemopoietic progenitor cells was used as a starting point for this study, to firstly determine if the assay performs as expected in an already known system, before moving forward with investigating the potential of ALDH activity as a marker for mesenchymal progenitor cells in hUCB and bone marrow cells of several mammalian species.
Employing intracellular metabolic activity as a means of identifying a progenitor pop- ulation is a novel approach that can avoid the need to raise monoclonal antibodies for each new species that is investigated. Whilst Aldefluor substrate provides a positive selection technique, it requires another marker such as for low side scatter in order to have enough specificity to identify haemopoietic progenitor subpopulations. We were
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 82 Chapter 3. Cell identification and isolation 3.4. Discussion unable to demonstrate the co-purification of MSC with HSC from cord blood using a
SSClo ALDHbr gate. Furthermore it was shown that MSC isolated by plastic adherence were Aldefluor positive, but had high SSC.
3.4.1 MSC in cord blood
The presence of MSC in human umbilical cord blood is currently contentious. Some groups are reporting finding MSC in cord blood, with others claiming no success. In this study, the lowest plating density to produce colonies was found to be 2 x 106 cells/plate, at a frequency of 3.3 CFU-F per 1 x 106 cells plated. These cells formed discrete adherent colonies, and were somewhat spindle shaped (Figure 3.2), characteristic of mesenchymal progenitor cells. A differentiation study down the mesenchymal pathways of bone, carilage and fat would be required to confirm if these cells were mesenchymal progenitor cells.
This study aimed to isolate mesenchymal progenitor cells in hUCB based on ALDH activity with Aldefluor substrate. Aldefluor substrate can be used to isolated haemopoi- etic progenitor cells from hUCB, which are present in the low orthogonal light scattering and bright fluorescence region. This region was investigated for the presence of mes- enchymal progenitor cells. Whilst on this occasion a population of ALDHbr CD34neg cells was believed to have been isolated from hUCB, the isolated cells were unable to be cultured successfully to demonstrate the clonal properties that distinguish progenitor cells from more differentiated cells. This inability to sustain viability amongst the frac- tionated populations in mesenchymal optimised conditions may be a result of several factors.
The low cell numbers from the cell fractionation may have played a part, as plating efficiency is not independent of the plated cell number (Table 3.1). As the plated cell number is increased, the plating efficiency increases (Figure 3.3).
It is possible that the cells may have been damaged by the sorting process and conse- quentially were unable to proliferate in culture. Although if this were the case, then
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 83 Chapter 3. Cell identification and isolation 3.4. Discussion the cells cultured under haemopoietic conditions would not have proliferated either.
Cells from the ALDHbr, ALDHbr CD34+,andCD45neg fractions all proliferated under haemopoietic conditions.
The number of CFU-F that should be present in each sorted fraction was approximated from the lowest incidence reported in literature occurring in bone marrow (1 in 1 x 106 cells) (Table 3.2). Cells from the ALDHbr fraction were plated at a density of 13.75 x106 cells per 35mm dish. Based on the results from the CFU-F assay of unsorted hUCB (Table 3.1), this should have been a sufficient plating density for the formation of colonies. Cells from the ALDHbr CD34neg fraction were plated at an approximate density of 5 x 106 cell per 35mm dish, which also should have been sufficient for the formation of colonies.
These results suggest that MSC were not co-purified with HSC in the low orthogonal light scattering region. It is likely that MSC would be found in the high orthogonal light scattering region, and consequently, Aldefluor substrate was not useful in isolating mesenchymal progenitor cells from hUCB.
3.4.2 ALDH variation with passage
It was found that passaged cells stain poorly compared to marrow with Aldefluor sub- strate. Data sets for rat and porcine marrow were well outside the 95% confidence level calculated from the mean fluorescent ratios of the passaged cells. Whilst the equine marrow sample when labeled with Aldefluor substrate did contain a population of ALDHbr cells, the mean fluorescent ratio was found to be lower for the marrow than for passaged cells. A change in cell size and granularity was seen between unlabled and labeled populations of equine bone marrow (Figure 3.25, plots A and C). This change was not observed for any other species studied, and was also not seen for equine tissue culture plastic selected passage 1 and 3 cells. It was thought to be an outlying data set not truly representative of the effect of Aldefluor substrate labeling, and due to the unavailability of equine marrow samples, could not be verified.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 84 Chapter 3. Cell identification and isolation 3.4. Discussion
One possible explanation for the decrease of ALDHbr cells in the passaged populations compared to the bone marrow, is that the primitive haemopoietic compartment, which stains brightly, is depleted by plastic adherence and passage. Also, down regulation of the ALDH genes may be occurring due to the effects of passage and time spent on tissue culture plastic.
No significant effect of passage was seen with Aldefluor substrate staining for both the rat and porcine cells.
3.4.3 ALDH variation with species
If Aldefluor substrate has use in a variety of species, not just for human cells, which the product was initially developed for, it will easily expand the possibility of prospectively isolating progenitor cells. Unlike monoclonal antibodies that must be raised against species specific antigens, the substrate would have much wider reaching application.
This study found that mammalian marrow cells and tissue culture plastic selected cells from rat, murine, porcine and equine origins all showed an ALDHbr population.
Rat whole bone marrow when labeled with Aldefluor substrate had a cluster of ALDHbr
SSClo cells (circled region in Figure 3.15d). This population was not as distinct as that seen for cord blood and was not present for any other species.
A change in proportion of the ALDHbr population was seen with species. Porcine marrow had the highest mean fluorescence ratio (50.81), followed by rat (35.44), mouse
(5.15) and then equine (2.56). Equine tissue culture plastic selected cells had the highest average mean fluorescence ratio (9.15), followed by rat (6.17) and then porcine (4.38).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 85 Chapter 3. Cell identification and isolation 3.5. Conclusion
3.5 Conclusion
The methodology for detecting HSC in human umbilical cord blood using Aldefluor substrate was successfully replicated using a concentration of 0.122μg/mL.
The lowest plating density to produce discrete adherent colonies comprising of some- what spindle shaped cells was found to be 2 x 106 cells/plate, at a frequency of 3.3
CFU-F per 1 x 106 cells plated. Plating efficiency was found to be dependent on the plated cell number, and as the plated cell number increased, so did the plating efficiency.
MSC were not co-purified with HSC in the low orthogonal light scattering region when labeled with Aldefluor substrate. It is likely that MSC would be found in the high orthogonal light scattering region, and consequently, Aldefluor substrate was not useful in isolating mesenchymal progenitor cells from hUCB.
The methodology for assessing mesenchymal stem cell frequency using a CFU-F assay was replicated using rat bone marrow cells. A plating density of 2.5 x 106 cells/plate produced a frequency of 21 CFU-F per 1 x 106 cells plated.
Cells were successfully isolated from multiple species (rat, murine, porcine and equine), and all showed an ALDHbr population when labeled with Aldefluor substrate. It was found that passaged cells stained poorly when compared to marrow, and no significant effect was observed with passage number. A change in proportion of the ALDHbr population was seen with species, with rat whole bone marrow being the only sample to show a small cluster of ALDHbr SSClo cells.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 86 Chapter 4
Hollow fibre expansion of anchorage dependent cells
4.1 Introduction
Bioreactors have traditionally been used in industrial applications such as fermenta- tion, wastewater, and food processing, along with pharmaceutical and recombinant protein production [275], which are now well established processes. Large numbers of mammalian cells have been generated for biopharmaceutical production via the use of traditional bioreactor technologies (such as roller bottles, stirred tanks, micro-carriers and hollow fibres), but this has mainly been with immortalised cell types that do not differentiate in culture or have such specialist growth environment requirements as pri- mary progenitor cells. Translation of these technologies to rare primary cell expansion remains a challenge.
4.1.1 MSC expansion
The low frequency of MSC in bone marrow (1: 104) makes expansion a prerequisite for
MSC therapies [276]. For clinical applications where MSC will be used as a transfusion
87 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.1. Introduction product (graft-versus-host disease, renal failure, Crohn’s disease, myocardial ischaemia, etc.), the optimal dose and frequency of administration can only be determined by clinical trials. The time-consuming and labour-intensive nature of conventional flask culture has restricted upper target doses in clinical trials to about 108 cells per patient
[276, 277]. Therapeutic efficacy is likely to lie beyond this range.
For flask culture systems, MSC reach confluence at around 5 × 103 cells cm−2 and max- imal expansion relies on seeding at relatively low density (1-50 cells cm−2). Bartmann et al. developed a two-step culture protocol in which 1.5 × 108 cells were expanded from 10mL bone marrow over a 4-week period [278]. The first and second passages required 0.2m2 and 2.5m2 of culture surface area, respectively. Plastic adherent cells from bone marrow mononuclear cells were cultured over the first 10-13 days using 8-
10 225cm2 tissue culture flasks. The secondary large-scale culture was achieved using
10 four-layered tissue culture flasks. While this approach has the advantage of easy translation from laboratory-scale studies that use tissue culture ware, it is not easily scaled.
MSC are anchorage dependent cells. The success of MSC growth and expansion relies upon the adherence of these cells to an appropriate growing substrate [27]. Surface chemistry and topography have major effects on cell adherence and interaction. Typ- ically, anchorage dependent cells are grown on a specially treated polystyrene surface, rendered wettable by oxidation. The oxidation process increases the adhesiveness of the surface to the cells.
Specifically engineered or treated surfaces may be used to either prevent or enhance cell growth. Surfaces treated with extracellular matrix components have the ability to facilitate adhesion of anchorage dependent cells, through integrin mediated ligand binding.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 88 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.1. Introduction
4.1.2 Cell surface receptors - Integrin binding
Integrins are a family of heterodimeric cell surface receptors [279]. They consist of two noncovalently linked polypeptide chains, designated alpha and beta [280], that mediate cell-to-cell and cell-to-extracellular matrix interactions through ligand binding. Inte- grins mediate cell adhesion to the extracellular matrix proteins fibronectin, laminin and collagen [281], whilst also connecting to the cytoskeleton inside the cell, and regulating intracellular signaling pathways [282].
Cell adhesive interactions perform fundamental roles during many normal physiological processes. Integrin mediated adhesive interactions are involved in the regulation of cellular functions such as embryonic development, tumor cell growth and metastasis, programmed cell death, hemostasis, leukocyte homing and activation, bone resorption, clot retraction, and the response of cells to mechanical stress [283]. Integrins also perform vital functions in directing cellular migration, proliferation, and differentiation.
4.1.3 Approach and Aims
Hollow fibre systems offer a number of advantages over conventional culture methods.
These include a reduction in culture space, as the large surface area to volume ratio means a compact device can be used in place of an incubator full of tissue culture flasks, and these systems can be connected to closed flow systems for automated inoculation, culture and harvesting [147, 284]. This chapter investigates a hollow fibre module, containing a single cellulose hollow fibre [225].
Theaimsofthischapterareto:
• Investigate the potential of a recombinant fibronectin protein in facilitating cell
attachment to cellulose based surfaces.
• Develop an in situ method for the viewing and counting of anchorage dependent
cells inside a hollow fibre.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 89 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.1. Introduction
• Examine the feasibility of expanding anchorage dependent cells inside a hollow
fibre.
• Investigate the effect of extracapillary co-culture on cell proliferation inside a
hollow fibre.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 90 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.2. Materials and Methods
4.2 Materials and Methods
4.2.1 Materials
Cuprophan (regenerated cellulose) dialyzer fibres with an external diamter of 200μm and a wall thickness of 7μm were obtained from Membrana GmbH, Wuppertal, Ger- many. Flat sheet cellulose membranes were removed from a Lundia®10 parallel flow dialyzer (5N) obtained from Gambro. Retronectin, a recombinant protein that contains the cell-binding domain (type III repeat, 8,9,10), a high affinity heparin-binding domain
II (type III repeat, 12,13,14), and CS1 site within the alternatively spliced IIICS region of human fibronectin fused with a cellulose binding domain (CBDR) was synthesised as per Appendix B. Sylgard® 184 Silicone Elastomer (main component polydimethyl- siloxane (PDMS), Base (3097366-1004), Curer (3097358-1004)), Silastic laboratory tub- ing (0.51mm inner diameter, 0.94mm outer diameter, 508-002) were obtained from Dow
Corning, MI USA. Hypodermic needles were obtained from Terumo Medical, Sydney
Australia. BIOSTATUS DRAQ5 (BOS-889-001-R200) was obtained from Alexis Bio- chemicals, San Diego USA. Hoechst 33342 (trihydrochloride, trihydrate H1399) and
4’,6-diamidino-2-phenylindole, dilactate (DAPI, dilactate, D3571) were obtained from
Invitrogen Molecular Probes, Melbourne Australia. Rhodamine Phalloidin (R-415) was obtained from Molecular Probes, Eugene Oregon USA. Calcein-acetoxymethyl es- ter (calcein-AM, 354217) was obtained from Becton Dickinson Biosciences, Sydney
Australia. Propidium Iodide (PI, P4170) and Paraformaldehyde (P-6148) were ob- tained from Sigma-Aldrich, Sydney Australia. Triton-X100 (6008083) was obtained from Packard Instrument Company, IL USA. Murine MS-5 stromal cell line was kindly donated by Dr. Alla Dolnikov of the Children’s Cancer Institute Australia. Mouse em- bryonic fibroblast cell line, NIH 3T3 transduced with enhanced green fluorescent protein
(eGFP), was kindly provided by iNano, University of Aarhus, Aarhus, Denmark.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 91 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.2. Materials and Methods
4.2.2 Methods
4.2.2.1 Construction of single hollow fibre modules
Single hollow fibre modules (Figure 4.1) were constructed from Cuprophan hollow fibres as previously described1 [225]. Two opposing 1mm holes were drilled into the side of a 35mm tissue culture polystyrene Petri dish using a hand held drill just above the bottom rim of the dish. Injection ports made from 10mm lengths of 23-gauge stainless steel tubing (outer diameter, 0.8mm) were cut from 23-gauge hypodermic disposable needles. The injection ports were covered with a 5mm segment of silastic tubing, and inserted into the wall of the dish, with the silastic tubing creating a seal between the Petri dish and the stainless steel tubing. A Cuprophan hollow fibre was carefully threaded through one injection port, through the dish and out the other injection port, whilst ensuring that the lumen of the hollow fibre did not collapse during the process.
Epoxy cement was applied to the outside of the dish to secure the injection ports in place. Once dry, approximately 0.2mL of PDMS (9 parts base mixed with 1 part curing agent for 15 minutes) was applied to the inner side of the dish at the injection port, to create a seal around the hollow fibre, stainless steel tube and silastic tubing joint. The
PDMS was cured at 60℃ overnight, and the process repeated for the other joint. The hollow fibre was cut flush with the edge of the stainless steel tubing on both injection ports. The sharp end of a 23-gauge hypodermic disposable needle was filed off, and inserted into 5cm of silastic tubing. The free end of the tubing was attached to the injection ports.
1A.J.J. van den Berg, Masters student, Biomedical Engineering - Biomaterials, University of Gronin- gen, Groningen, The Netherlands, is acknowledged for assistance provided in module manufacture.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 92 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.2. Materials and Methods
Figure 4.1: Single hollow fibre module
4.2.2.2 Sterlisation of single hollow fibre modules
Hollow fibre modules were placed under UV light for approximately 1 hour in a biolog- ical flow hood. The 35mm dish housing the fibre was washed with 2-3mL of 70/80% ethanol, whilst the fibre itself was flushed with 0.3mL of the ethanol solution using a
1mL syringe. This washing process was repeated with water for irrigation (WFI) and
Dulbecco’s phosphate buffered saline (DPBS). The modules were left with DPBS to ensure no drying out of the fibre occurred.
4.2.2.3 CBDR coating of single hollow fibre modules
A needle injection port was connected to the leur connector at the entry of the hollow
fibre. 0.2mL of either 75μg/mL in DPBS or neat CBDR was injected into the fibre using a 1mL syringe. The needle and syringe were left attached at the entry whilst a needle injection port was carefully attached to the leur connector at the exit of the
fibre. The needle and attached syringe at the entry were gently removed, ensuring that no air entered the hollow fibre. Modules were incubated overnight at 4℃,withDPBS remaining in dish during incubation.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 93 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.2. Materials and Methods
4.2.2.4 Cell inoculation of single hollow fibre modules
Cell samples were prepared at the required concentration and loaded into a 1mL syringe using a 21-gauge hypodermic needle. The needle injection port at the exit of the fibre was removed, and approximately 100μL of the cell sample was injected into the fibre, ensuring that no air bubbles entered the hollow fibre. The needle and syringe used for loading were left attached at the entry whilst the exit needle injection port was connected, and the tubing at both the exit and entry were clamped with metal clamps.
Medium was placed in the dish, and the devices placed in a humidified in a humidified atmosphere of 95% air with 5% CO2 at 37℃.
4.2.2.5 Preparation of CBDR coated cellulose dishes
0.5mL of PDMS (9 parts base mixed with 1 part curing agent for 15 minutes) was placed in a 35mm Petri dish. After bubbles were dissolved under a vacuum, a cellulose membrane was overlayed and the dish baked at 60℃ overnight. After washing with
DPBS, the dish was incubated with 2mL of 75μg/mL of CBDR at 4℃ overnight.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 94 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
4.3 Results
4.3.1 Attachment Effects of CBDR
Cell attachment onto a variety of cell culture substrates was tested using a recombinant protein with a cellulose binding domain.
Bacteriological grade plates were coated with neat CBDR, and left overnight at 4℃.The plates were rinsed with DPBS and inoculated with 20 x 103 MS-5 cells in 2mL α-MEM supplemented with 10% FBS. After 2 hours incubation in a humidified atmosphere of
95% air with 5% CO2 at 37℃, attachment had occurred to the coated plate, as seen in Figure 4.2a.
(a) CBDR coated plate (b) Uncoated control plate
Figure 4.2: MS-5 cell attachment after 2 hours incubation in a humidified atmosphere of 95% air with 5% CO2 at 37℃ to CBDR coated and uncoated bacteriological grade plates. Representative images, captured under phase contrast, using a 10x objective.
100 x 103 NIH 3T3 (eGFP) cells in 2mL IMDM supplemented with 10% FBS were seeded onto a cellulose membrane coated with CBDR, prepared as per Section 4.2.2.5.
After 1 hour of incubation in a humidified atmosphere of 95% air with 5% CO2 at 37℃ cell attachment was beginning to occur onto the coated cellulose membrane, as can be seen in Figure 4.3a.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 95 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) CBDR coated cellulose (b) CBDR coated cellulose (c) CBDR coated cellulose membrane, 1 hour membrane, 2 hours membrane, 3 hours
(d) Uncoated cellulose mem- (e) Uncoated cellulose mem- (f) Uncoated cellulose mem- brane, 1 hour brane, 2 hours brane, 3 hours
(g) TCP, 1 hour (h) TCP, 2 hours (i) TCP, 3 hours
(j) Bacteriological grade plate, 1 (k) Bacteriological grade plate, (l) Bacteriological grade plate, 3 hour 2hours hours
Figure 4.3: NIH 3T3 (eGFP) (100 x 103) cell attachment to CBDR coated and uncoated cellulose membranes. Attachment after 1, 2 and 3 hours incubation in a humidified atmosphere of 95% air with 5% CO2 at 37℃. Representative images, captured under phase contrast, using a 10x objective.
Hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
Modules were coated with CBDR as per Section 4.2.2.3 and inoculated with a 5 x 106
MS-5 cells/mL solution as per Section 4.2.2.4. 2mL of α-MEM supplemented with 10%
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 96 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
FBS was added to each module, and cell attachment to the coated hollow fibre had taken place within 2 hours (Figure 4.4c).
It can be seen from Figure 4.4 that within the presence of CBDR, cell attachment and spreading occurs quite quickly inside the cuprophan hollow fibres. Without the CBDR, the cells do not attach at all, making the CBDR coating an absolute requirement for growth of NIH 3T3 (eGFP) cells inside the cuprophan hollow fibres.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 97 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) CBDR coated HF immedi- (b) Uncoated HF immediately ately after inoculation after inoculation
(c) CBDR coated HF, 2 hours (d) Uncoated HF, 2 hours
(e) CBDR coated HF, 3 hours (f) Uncoated HF, 3 hours
(g) CBDR coated HF, 24 hours (h) Uncoated HF, 24 hours
(i) CBDR coated HF, 48 hours (j) Uncoated HF, 48 hours
Figure 4.4: MS-5 cell attachment to CBDR coated cellulose hollow fibres. 5 x 106 MS-5 cell- s/mL were inoculated into CBDR coated hollow fibre modules. Attachment after incubation in a humidified atmosphere of 95% air with 5% CO2 at 37℃.Repre- sentative images, captured under phase contrast, using a 10x or 20x objective. Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 98 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
4.3.2 In situ cell staining
The establishment of an in situ staining method would allow for both the assessment of cellular viability and proliferation. A number of staining methods were investigated for these purposes.
4.3.2.1 Vital cell staining
Hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
Modules were coated with CBDR as per Section 4.2.2.3 and inoculated with a 10 x
106 MS-5 cells/mL solution as per Section 4.2.2.4. 2mL α-MEM supplemented with
10% FBS was added to each module, and modules were incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃. After 24 hours incubation, growth medium was removed, modules were washed twice with DPBS, incubated with 2.5mL of 10μM
DRAQ5 (in DPBS) solution at 37℃ for 1-3 minutes, and viewed. The effect of in situ staining with DRAQ5 on CBDR coated and uncoated hollow fibre modules is shown in
Figure 4.5.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 99 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) CBDR coated hollow fibre
(b) Uncoated hollow fibre
Figure 4.5: The effect of staining with DRAQ5 on MS-5 cells in HF modules coated with CBDR. 10 x 106 MS-5 cells/mL were inoculated into a CBDR coated hollow fibre module. After 24 hours, cells were stained with a 10μM DRAQ5 solution and viewed with an Olympus DP70 in light mode, using a 10x objective.
Hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
Modules were coated with CBDR as per Section 4.2.2.3, inoculated with a 10 x 106 MS-
5 cells/mL solution as per Section 4.2.2.4. 2mL α-MEM supplemented with 10% FBS was added to each module, and modules were incubated in a humidified atmosphere of
95% air with 5% CO2 at 37℃. Control plates (35mm TCP Petri dishes) were inoculated with 250 x 106 MS-5 cells. After 24 hours incubation, modules and control plates were stained with either calcein-AM (2.5mL of 1μg/mL Cal-AM in DPBS at 37℃ for 20 minutes in the dark), Hoechst 33342 (2.5mL of 10μg/mL Hoechst 33342 in DPBS at
37℃ for 20 minutes), PI (2.5mL of 1μg/mL PI in DPBS at 37℃ for 1 minute in the
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 100 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results dark) or a combination. All modules and plates were rinsed twice with DPBS before and after incubation with a stain. Modules and plates were treated with a 70% methanol solution prior to staining with PI, which resulted in detachment of cells from the fibre.
Theresultsforin situ cell staining in CBDR coated hollow fibre modules for Calcein-
AM, Hoechst 33342, and PI are shown in Figures 4.6a, 4.6d and 4.6g respectively, and staining results for TCP dishes are shown in Figures 4.6b, 4.6e and 4.6h respectively.
It was observed that after 1-2 hours, the calcein retained in the cell cytoplasm was being adsorbed by the fibre, resulting in the fibre fluorescing (Figure 4.6c). The fibre did not adsorb Calcein-AM upon initial staining, as was observed for Hoechst 33342 and PI (Figures 4.6f and 4.6i respectively), where it was found that the dyes were binding to the hollow fibre without the presence of cells or CBDR. The hollow fibre was also found to auto-fluoresce when excited at a wavelength of 355nm (Figure 4.7), the same wavelength used for Hoechst 33342.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 101 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) Calcein-AM stained cells in (b) Calcein-AM stained cells on (c) Calcein-AM transfer from HF coated with CBDR TCP cells to HF
(d) Hoechst 33342 stained cells (e) Hoechst 33342 stained cells (f) Hoechst 33342 binding to HF in HF coated with CBDR on TCP
(g) PI stained cells in HF coated (h) PI stained cells on TCP (i) PI binding to HF with CBDR
Figure 4.6: The effect of staining with Calcein-AM, Hoechst 33342 and PI on MS-5 cells in HF modules coated with CBDR. Representative images, captured using fluorescence microscopy, with a 10x objective.
Figure 4.7: Auto-fluorescence of hollow fibres when excited at a wavelength of 355nm. Rep- resentative image, captured using fluorescence microscopy, with a 10x objective.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 102 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
4.3.2.2 Non-vital cell staining
Hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
Modules were coated with CBDR as per Section 4.2.2.3, inoculated with a 10 x 106 MS-
5 cells/mL solution as per Section 4.2.2.4. 2mL α-MEM supplemented with 10% FBS was added to each module, and modules were incubated in a humidified atmosphere of
95% air with 5% CO2 at 37℃. Control plates (35mm TCP Petri dishes) were inoculated with 250 x 103 MS-5 cells. After 24 hours incubation, modules were fixed with a 4% paraformaldehyde solution (10 minutes at room temperature) and stained with either
DAPI (2mL of 1μg/mL DAPI in DPBS solution for 10 minutes at room temperature in the dark), rhodamine phalloidin (2mL of 1 U/mL rhodamine phalloidin in DPBS solution for 30 minutes at room temperture in the dark) or PI (2.5mL of 1μg/mL PI in DPBS solution at room temperature for 1 minute in the dark). All modules were rinsed twice with DPBS before and after incubation with a stain.
The results for in situ cell staining in CBDR coated hollow fibre modules for DAPI, rhodamine phalloidin, and PI are shown in Figures 4.8a, 4.8d and 4.8g respectively. Due to poor uptake of the dyes by the fixed cells, the cell membranes were permeabilised
(using a 0.1% Triton X-100 solution), and the staining process repeated. Greater uptake of the dyes is now seen (Figures 4.8b, 4.8e and 4.8h). The staining effects on
TCP control dishes where cells were both fixed and permeabilised prior to staining are shown in Figures 4.8c and 4.8f. Figure 4.9 shows the effect of staining with both DAPI and rhodamine phalloidin, for fixed and permeabilised cells in a HF module (Figure
4.9a) and on TCP (Figure 4.9b).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 103 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) Fixed cells in CBDR coated (b) Fixed and permeabilised (c) Fixed and permeabilised HF stained with DAPI cells in CBDR coated HF cells on TCP stained with stained with DAPI DAPI
(d) Fixed cells in CBDR coated (e) Fixed and permeabilised (f) Fixed and permeabilised HF stained with rhodamine cells in CBDR coated HF cells on TCP stained with phalloidin stained with rhodamine rhodamine phalloidin phalloidin
(g) Fixed cells in CBDR coated (h) Fixed and permeabilised HF stained with PI cells in CBDR coated HF stainedwithPI
Figure 4.8: The effect of staining with DAPI, rhodamine phalloidin and PI on MS-5 cells in HF modules coated with CBDR. Representative images, captured using fluorescence microscopy, with a 10x objective.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 104 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) Fixed and permeabilised cells in CBDR (b) Fixed and permeabilised cells on TCP coated HF stained with DAPI and rhodamine stained with DAPI and rhodamine phalloidin phalloidin
Figure 4.9: The effect of staining with DAPI and rhodamine phalloidin on MS-5 cells in HF modules coated with CBDR. Representative images, captured using fluorescence microscopy, with a 10x objective.
In order to ascertain cross sectional information, cells inside the hollow fibre modules were fixed with a 10% neutral buffered formalin solution (10 minutes), the hollow fibre was encased with agarose, which once set was cut out of the module. The sample was embedded in paraffin, sectioned as per Appendix A.3.2, and stained with H&E as per
Appendix A.3.4. Figure 4.10 shows that during the embedding and sectioning process, the attached cells come away from the fibre, and internal structure is not entirely conserved. Staining artifacts can also be seen.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 105 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
Figure 4.10: H&E staining of MS-5 cells in a HF module coated with CBDR, embedded and sectioned. Representative section, image captured using light microscopy, with a 10x objective.
In situ cell staining summary
In situ vital staining was found to be feasible using the nuclear stain DRAQ5. The methods applied for Calcein-AM and Hoechst 33342 did not result in the development of a useful vital staining procedure due to transfer of the dye to the fibre from the cells, and non specific binding of the dye to the fibre respectively.
Both cytoskeletal and nuclear staining using DAPI and rhodamine phalloidin was ob- tained by fixation and permeabilisation of the cells, whereby providing a means of assessing in situ cellular proliferation.
4.3.3 Cell culture in HF Modules
The growth of rat bone marrow cells isolated by adherence to tissue culture plastic, NIH
3T3 (eGFP) and MS-5 cells were investigated inside hollow fibre modules coated with
CBDR retronectin. Where cell counts are performed, only cells attached to the hollow
fibre surface were counted due to difficulty in distinguishing between dividing cells,
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 106 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
floating cells and apoptotic cells, which all displayed a similar rounded morphology.
Fold Expansion was calculated by dividing the cell number at a given time point by the number of cells inoculated into the fibre.
4.3.3.1 Primary rat cells
16 hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and
4.2.2.2. All modules were coated with a 75μg/mL CBDR solution as per Section 4.2.2.3, and inoculated with rat passage 4 bone marrow cells in EMEM supplemented with 20%
FBS as per Section 4.2.2.4. 8 modules were inoculated with a 100 x 103 cells/mL solu- tion, and the remaining 8 modules were inoculated with a 500 x 103 cells/mL solution.
2mL EMEM supplemented with 20% FBS was added to each module, and modules were incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃. 2 replicates per time point were fixed with a 4% paraformaldehyde solution (10 minutes at room temperature), permeabilised (using a 0.1% Triton X-100 solution for 5 minutes at room temperature) and then stained with DAPI (2mL of 1μg/mL DAPI in DPBS solution for 10 minutes at room temperature in the dark) and rhodamine phalloidin (2mL of 1
U/mL rhodamine phalloidin in DPBS solution for 30 minutes at room temperture in the dark). All cells inside the hollow fibre were counted using fluorescence microscopy, with resulting cell counts shown in Figure 4.11 and corresponding representative images in Figure 4.12.
From Figure 4.11 it can be seen that when the CBDR coated hollow fibre modules were inoculated at densities of 100 x 103 and 500 x 103 cells/mL with rat passage 4 bone marrow cells, cellular proliferation did not occur. Some viability was maintained for
5 days, as indicated by the morphology of cells seen in Figures 4.12c and 4.12d, but could not be maintained under these conditions.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 107 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
100
80
60
40 Cells per fibre per Cells 20
0 012345678910 Time (Days)
(a) Inoculation concentration 100 x 103 cells/mL.
500
400
300
200 Cells per fibre per Cells 100
0 012345678910 Time (Days)
(b) Inoculation concentration 500 x 103 cells/mL.
Figure 4.11: Rat bone marrow cell growth in CBDR coated hollow fibre modules. Rat passage 4 bone marrow cells were inoculated at densities of 100 and 500 x 103 cells/mL into hollow fibre modules coated with a 75μg/mL CBDR solution. 2 replicate modules were stained with DAPI and rhodamine phalloidin per time point, and all cells counted inside the hollow fibre using fluorescence microscopy.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 108 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) Inoculation concentration (b) Inoculation concentration 100 x 103 cells/mL, day 1 500 x 103 cells/mL, day 1
(c) Inoculation concentration (d) Inoculation concentration 100 x 103 cells/mL, day 5 500 x 103 cells/mL, day 5
(e) Inoculation concentration (f) Inoculation concentration of 100 x 103 cells/mL, day 9 500 x 103 cells/mL, day 9, module 1
(g) Inoculation concentration 500 x 103 cells/mL, day 9, module 2
Figure 4.12: Rat bone marrow cell growth in CBDR coated hollow fibre modules. Rat pas- sage 4 bone marrow cells were inoculated at concentrations of 100 and 500 x 103 cells/mL into hollow fibre modules coated with a 75μg/mL CBDR solu- tion. 2 replicate modules stained with DAPI and rhodamine phalloidin per time point. Representative images, captured using fluorescence microscopy, using a 10x objective.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 109 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
4.3.3.2 NIH 3T3 (eGFP) cells
3 hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
All modules were coated with neat CBDR solution as per Section 4.2.2.3, and inoculated with a solution of 500 x 103 NIH 3T3 (eGFP) cells/mL as per Section 4.2.2.4. 2mL
IMDM supplemented with 10% FBS was added to each module, and modules were incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃. Culture medium in the module was changed every 4 days. All cells inside the hollow fibre were counted for each time point using fluorescence microscopy, with resulting cell counts shown in
Figure 4.13, and corresponding representative images for days 14 and 16 in Figure 4.14.
From Figure 4.13 it can be seen that when CBDR coated hollow fibres were inoculated at a density of 500 x 103 cells/mL with NIH 3T3 (eGFP) cells, cellular proliferation did occur, with maximum expansion was observed at day 8. Deterioration of cells by day 14 and 16 can be seen in Figure 4.14.
700 Module 1 Module 2 600 Module 3 500
400
300
Cells per fibre 200
100
0 0 2 4 6 8 10 12 14 16 Time (Days)
Figure 4.13: NIH 3T3 cell growth in CBDR coated hollow fibre modules. NIH 3T3 (eGFP) cells were inoculated at a concentration of 500 x 103 cells/mL into 3 hollow fibre modules coated with neat CBDR. All cells were counted inside each hollow fibre using fluorescence microscopy for each time point.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 110 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
(a) Module 1, day 14 (b) Module 2 day 14
(c) Module 1, day 16 (d) Module 2, day 16
Figure 4.14: NIH 3T3 (eGFP) cell growth in CBDR coated hollow fibre modules. NIH 3T3 cells (eGFP) were inoculated at a concentration of 500 x 103 cells/mL into 3 hollow fibre modules coated with neat CBDR. Representative images, captured using fluorescence microscopy, using a 10x objective at days 14 and 16 of culture.
8 hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
All modules were coated with neat CBDR solution as per Section 4.2.2.3, and inoculated with NIH 3T3 (eGFP) cells/mL as per Section 4.2.2.4. 4 modules were inoculated with a 500 x 103 cells/mL solution, and the remaining 4 modules were inoculated with a
1x106 cells/mL solution. 2mL IMDM supplemented with 10% FBS was added to each module, and modules were incubated in a humidified atmosphere of 95% air with
5% CO2 at 37℃. The acute myelogenous leukemia cell line, KG1a, was added to the extracapillary medium of 4 modules (2 from each concentration group) in order to in order to investigate the effect of co-culture on growth of NIH 3T3 (eGFP) cells. All cells inside the hollow fibre were counted for each time point using fluorescence microscopy, with resulting fold expansion shown in Figure 4.15.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 111 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results
Greatest fold expansion (4.7) was achieved after 3 days of culture with the inoculation density of 1 x 106 cells/mL with an extracapillary co-culture of KG1a cells.
7 500x10e3 cells 500x10e3 cells with extracapillary KG1a cell co-culture 6 1x10e6 cells 5 1x10e6 cells with extracapillary KG1a cell co-culture
4
3 l pasi
2
1
0 0135 Time (Days)
Figure 4.15: Fold expansion of NIH 3T3 (eGFP) cells in CBDR coated hollow fibre modules, with extracapillary KG1a co-culture. NIH 3T3 (eGFP) cells were inoculated at a concentration of 500 x 103 and 1 x 106 cells/mL into 8 hollow fibre modules coated with neat CBDR. KG1a cells were added to 2 modules of each concentra- tion. Bars denote mean fold expansion for 2 modules ± one standard deviation.
4.3.3.3 MS-5 cells
6 hollow fibre modules were constructed and sterilised as per Sections 4.2.2.1 and 4.2.2.2.
Modules were coated with a 75μg/mL CBDR solution as per Section 4.2.2.3, and in- oculated with a solution of 500 x 103 MS-5 cells/mL as per Section 4.2.2.4. 2mL of
α-MEM supplemented with 10% FBS was added to each module. As culture medium deteriorates rapidly at 37℃ without cell contact, and in order to circumvent any ef- fects caused by potential media degradation, supporting cells were cultured outside the hollow fibre in the module. 3 modules received an extracapillary co-culture of MS-5 cells, and as a result 100 x 103 cells were seeded into the bottom of the module, and all modules were incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃. All cells inside the hollow fibres were counted for each time point under phase
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 112 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.3. Results contrast using light microscopy, and medium was changed every 3 days2. After 19 days culture, modules were fixed with a 4% paraformaldehyde solution (10 minutes at room temperature), permeabilised (using a 0.1% Triton X-100 solution for 5 minutes at room temperature) and then stained with DAPI (2mL of 1μg/mL DAPI in DPBS solution for 10 minutes at room temperature in the dark) and rhodamine phalloidin (2mL of 1
U/mL rhodamine phalloidin in DPBS solution for 30 minutes at room temperture in the dark). Fold expansion for each group is given in Figure 4.16.
Fold expansion after 10 days culture was found to be more than double for modules with an extracapillary co-culture, compared to modules without.
4.5 Modules with extracapillary MS-5 cell co-culture 4 Control modules
3.5
3
2.5
2
Fold Expansion 1.5
1
0.5
0 14710 Time (Days)
Figure 4.16: Fold expansion of MS-5 cells in CBDR coated hollow fibre modules. MS-5 cells were inoculated at a concentration of 500 x 103 cells/mL into hollow fibre mod- ules coated with a 75μg/mL CBDR solution. An extracapillary co-culture of MS-5 cells was added to half of the modules. Cells inside the hollow fibres were counted for each time point under phase contrast, using light microscopy. Bars denote mean fold expansion for 3 modules ± one standard deviation.
2Modules were prepared, inoculated and counted by A.J.J. van den Berg, Masters student, Biomed- ical Engineering - Biomaterials, University of Groningen, Groningen, The Netherlands, 2008.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 113 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion
4.4 Discussion
For MSC therapies to translate from laboratory-scale studies into clinically useful rapid consecutive perfusion treatments, expansion to sufficient numbers is a prerequisite.
The expense associated with growing cells in monolayer, occurring from both the labour component involved in manual inoculation and passaging, and the physical space re- quired makes bioreactor technologies not only attractive but necessary. Not only do they provide a high degree of reproducibility, control and automation [275], whilst de- livering a suitable in vitro environment, the reduction in human interference results in reduced error.
Current techniques for MSC expansion depend on addition of FBS to culture media.
FBS is an undesirable source of xenogeneic antigens; it has the potential to transmit animal viruses and prions and has problems with batch-to-batch variability. The most stringent regulators may require that culture medium follow pharmaceutical-like spec- ification requirements, and as such should be composed of highly purified and well characterized (e.g. recombinant) proteins. Another advantage of a fully defined media specification would be to improve the reproducibility of the expansion process, without the need to test serum batches individually. There is some confusion with regard to the constituents of ‘serum-free’ or ‘serum replacement’ products. These proprietary formu- lations may be better characterized but still contain serum components, perhaps from animal sources, which may not be highly purified or suitable for human use [285, 286].
Autologous serum is a viable alternative to FBS supplemented media and minimizes the risk of disease transmission. Mizuno et al. developed a closed-bag system to separate serum, and obtained superior human MSC expansion in comparison to FBS supple- mented media [287]. A platelet lysate obtained from human thrombocyte concentrates enhanced human MSC output compared to FBS, and was proposed as a clinically appli- cable animal serum replacement [288]. The limited supply and cost of blood products are important scale-up considerations.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 114 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion
The use of a recombinant protein (fragment of fibronectin), with a cellulose binding domain, for cell attachment to cellulose substrates was investigated. The longer term implication of which would lie in potentially using a recombinant attachment factor as a viable alternative to serum in culture medium. In this study, serum was included in all experiments. Soluble components in serum are also required for MSC growth (i.e. the lysophospholipid sphingosylphosphorylcholine). Development of a fully formulated serum free media for MSC expansion was beyond the scope of this thesis.
It was found that the cellulose binding domain retronectin was very effective in fa- cilitating cell attachment to cellulose based cell culture substrates and bacteriological grade (non-charged) polystyrene. Cell attachment occurred within 2 hours onto CBDR coated bacteriological grade plates (Figure 4.2), with similar results seen for attach- ment to flat cellulose membranes (Figure 4.3) and cellulose hollow fibres (Figure 4.4) coated with the recombinant protein. Without the recombinant protein, attachment did not occur to the cellulose based membranes or hollow fibres, even with the presence of serum in the culture medium.
4.4.1 In situ cell staining
In situ cell staining methods provide a means of assessing cellular proliferation inside the hollow fibre devices whilst leaving the developing tissue structure intact. Methods which involve the detachment and subsequent removal of adhered cells for the purposes of counting do not give insight into the formation and development of cell structures.
Vital staining methods would allow cellular assessment within the hollow fibres at any given time point during the culture period without requiring the termination of the experiment, and as a result were firstly investigated. The vital DNA dye, DRAQ5 was used to stain cell nuclei in both recombinant protein coated and uncoated hollow fibre modules (Figure 4.5). DRAQ5 can be added to either buffer or culture medium and does not require washing prior to imaging. It should identify nuclei and have high penetration through cell layers making it ideal for this application. Both fluorescence
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 115 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion and bright field microscopy could be used to visualize DRAQ staining. A dramatic difference was observed between the coated (Figure 4.5a) and uncoated fibre (Figure
4.5b). DRAQ5 is a nuclear intercalating stain and further studies would be required to establish that it does not inhibit growth.
The vital stains calcein-Am and Hoechst 33342 were employed in an attempt to visualise cell morphology and provide a means of assessing proliferation.
Calcein-AM is widely used for determining cell viability as the non-fluorescent, cell permeant compound is converted by intracellular esterases into calcein by live cells and retained in the cytoplasm. Whilst calcein-AM provides both morphological and functional information of viable cells, it can not be used as an effective method to count cells inside the hollow fibres due to difficulty in distinguishing between individual cells, particularly at high cell densities when cells form clumps (Figure 4.6a). Interestingly, it was also found that after a period of 1-2 hours post cell staining, the calcein retained in the cell cytoplasm was being transferred to the hollow fibre, and the fibre fluorescing
(Figure 4.6c).
Hoechst 33342 can be used on both live and fixed cells, and as it binds to DNA provides a means of counting cells by the number of nuclei present. The staining of cells inside the hollow fibre devices showed that whilst nuclei are visible (Figure 4.6d), there is non specific binding of the dye that occurs to the uncoated hollow fibres (Figure 4.6f), and that the fibres also auto-fluoresce (Figure 4.7) when excited at a wavelength of
355nm, the same wavelength as that required for Hoechst 33342. These factors explain the strong background seen in Figure 4.6d, making it difficult to count nuclei. The detachment and clumping of cells seen in Figure 4.6d is a result of the Hoechst 33342 staining procedure being carried out in PBS rather than cell culture medium.
PI was used as a means of assessing cell death, with Figure 4.6g illustrating that non- viable cells are readily visible. It was also found that there was a small amount of uptake of PI by the uncoated fibre (Figure 4.6i). This amount did not however affect assessment of viability by PI.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 116 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion
As vital staining methods did not reveal a means of accurately assessing cellular prolif- eration, the cells were fixed, permeabilised and staining was carried out with DAPI and rhodamine phalloidin. DAPI binds strongly to DNA, and provides a means of counting cell nuclei. Permeabilisation of the cell membranes allowed more dye to enter the cells, consequently having a marketed effect on the increase of cellular staining. Figure 4.8a shows fixed cells and Figure 4.8b highlighting the effect of fixation and permeabilisation for staining with DAPI. The same effect is seen for rhodamine phalloidin staining, used to stain the actin filaments due to its bonding to F-actin, allowing visualisation of the actin cytoskeleton. Figure 4.8d shows the result of staining on fixed cells and Figure
4.8e showing the improved effect of permeabilisation on rhodamine phalloidin staining.
The combination of staining with DAPI and rhodamine phalloidin provides a method of assessing in situ cellular proliferation for the hollow fibre modules (Figure 4.9a).
4.4.2 Cell culture in HF modules
Whilst CBDR facilitates cellular attachment to cellulose hollow fibres through integrin mediated adhesion, maintaining cellular viability and encouraging proliferation in this micro-environment remains a challenge. Sensitivity to the micro-environment, espe- cially at low densities, must be overcome for long term culture, for clinically relevant yields to be achieved from low cell numbers.
A variety of adherent cells were cultured in the hollow fibre modules, and the effects of initial seeding density, and supporting cell culture were investigated.
Rat tissue culture plastic adherent bone marrow cells were found to be particularly sensitive to inoculation density. When inoculated into the hollow fibre modules at a densities of 100 x 103 and 500 x 103 cells/mL, results did not indicate proliferation of the cells seeded (Figure 4.11). Visual inspection indicated that the cells seeded into the hollow fibres attached to the cellulose surface. Some cells remained viable for 5 days
(Figures 4.12c and 4.12d for inoculation densities of 100 x 103 and 500 x 103 cells/mL respectively), but proliferation was not observed and viability could not be maintained
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 117 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion under these culture conditions for more than a few days. The viable cells observed in Figure 4.12g appears to be a result of a variation in the inoculation process with a particularly high number of cells being inoculated into the device.
It was found that the inoculation procedure produced a degree of variability in the final number of cells seeded into the hollow fibres. This variability at time of inoculation, especially for low densities, had a marketed influence on the viability of cells. It is also possible that only a minority of bone marrow adherent cells are proliferative.
As primary cells can be quite sensitive and particular to the culture conditions, the inoculation density of 500 x 103 cells/mL were tested on the mouse fibroblast cell line
NIH 3T3 (eGFP). As these cells were transduced with a retroviral vector expressing the enhanced green fluorescent protein, it would not be necessary to stain the cells in order to conduct a cell count using fluorescent microscopy, allowing individual modules to be monitored over a time period without effecting the developing culture. The main draw- back with using a staining method which requires cellular fixation and permeabilisation in such a study is that the cell growth must be halted in order to obtain information on proliferation.
Maximum expansion was observed at day 8 of culture (Figure 4.13) for 2 modules when
NIH 3T3 (eGFP) cells were inoculated at a density at 500 x 103 cells/mL. Interestingly, even when the modules started culture with a similar number of cells, there was great variation in proliferation (Figure 4.13), illustrating how sensitive the cells are to the culture conditions.
There is at least one order of magnitude difference between the ratio of cells to culture medium in the hollow fibre modules compared to conventional flask culture, resulting in medium effects in the modules not normally impacting on conventional flask culture.
Without cell contact, culture medium deteriorates rapidly at 37℃. The formation of free radicals and reactive oxygen species in the medium also have the potential to induce cell damage. In order to reduce these effects, and potentially increase cell proliferation, supporting cells were cultured outside the hollow fibre in the module.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 118 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.4. Discussion
The impact of the suspension cell line KG1a (acute myelogenous leukemia) on cellular proliferation and viability were investigated. A suspension cell line was chosen due to ease of maintenance, allowing greater control of the supporting culture.
The culture of NIH 3T3 (eGFP) cells at densities of 500 x 103 and 1 x 106 cells/mL with an extracapillary co-culture of KG1a cells (Figure 4.15) revealed that the greatest fold expansion was achieved after 3 days of culture with the higher inoculation density with a supporting culture. However, there were not enough replicate experiments to test the statistical significance of this observation. Interestingly, at day 5, the fold expansion seen for the lower density (both with and with out supporting culture) was greater than that for the higher density and almost equal to that seen on day 3 for the higher inoculation density with a supporting culture. By day 5, the cells in the hollow
fibre for the higher cell density cultured with supporting cells, were not countable as the cells were clustering and apoptosing. This same result was seen for the remaining modules by day 7. NIH 3T3 (eGFP) cells can be quite space restrictive, and tend to deteriorate in over-confluent culture conditions.
The growth of the murine MS-5 stromal cell line was investigated inside the hollow fibre modules. Being a stromal cell line, MS-5 cells are not as sensitive to over-confluence as NIH 3T3 (eGFP) cells. Cells were inoculated at a density of 500 x 103 cells/mL and the effect of supporting MS-5 cell extracapillary culture was investigated. Fold expansion after 10 days culture was found to be more than double for modules with an extracapillary culture, compared to modules without (Figure 4.16). Cells within the modules could not be counted after 10 days in culture, due to difficulty in distinguishing individual cells. After 19 days of culture, the MS-5 cells (with supporting extracapil- lary culture) had formed an almost confluent tube around the fibre, illustrating that with an appropriate inoculation density, and optimisation of culture conditions, cell proliferation was feasible inside the hollow fibre modules coated with a recombinant protein with a cellulose binding domain.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 119 Chapter 4. Hollow fibre expansion of anchorage dependent cells 4.5. Conclusion
4.5 Conclusion
A recombinant protein with a cellulose binding domain was found to effectively facilitate adhesion through ligand binding on cellulose based surfaces. Fully defined media will require provision bound adhesion signals for support of anchorage dependent growth of
MSC.
An in situ method for the viewing and counting of anchorage dependent cells inside a hollow fibre was developed. It was found that fixation and permeabilisation of the cells inside the fibres facilitates uptake of the stains DAPI and rhodamine phalloidin.
The proliferation of rat tissue culture adherent bone marrow, NIH 3T3 (eGFP) and
MS-5 cells were investigated inside hollow fibre modules coated with a recombinant protein with a cellulose binding domain. Growth was suboptimal compared to tissue culture flasks. Hollow fibre modules co-cultured with KG1a or MS-5 cells appear to show an increased fold expansion which may be related to a reduction in oxidative degradation of culture media.
Future studies, such as those conducted by Gomez-Sjoberg et al. [235] using lab-on- a-chip technology, customising individual culture conditions in terms of cell seeding density, composition of culture medium, and feeding schedule, should be conducted.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 120 Chapter 5
Multi-channel bioreactor for engineering composite tissues from MSC
5.1 Introduction
The role of bioreactors is to transform laboratory-based experimental approaches into scalable cell and tissue production processes with standards of safety and efficacy similar to those established for pharmaceutical therapeutics. Whilst the transfer of laboratory developed technology to a clinical setting remains a large problem [289], from a clinical perspective, the ability to provide cell-based therapy as an ‘off the shelf option’, tailored to the specific needs of the patient and the biology of the lesion, is attractive and should be pursued [290].
5.1.1 Clinical need for MSC differentiated products
The differentiation capabilities of mesenchymal stem cells can be utilised for clinical applications where musculoskeletal tissue replacement or repair is required. Tissue
121 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.1. Introduction engineering of an osteochondral junction could enhance current surgical treatments, which include replacement of the entire joint, resurfacing of the articulating surface, replacement of the damaged tissue, or promotion of repair of the articular surface (Table
5.1).
The clinical goal centers around providing symptomatic relief and improved joint func- tion to the patient. Ideally, the production of repair tissue that has the same funtional and mechanical properties as that of hyaline articular cartilage is desired. Many surgi- cal treatments fail to prevent the future degeneration of the repair tissue and also the surrounding host tissue. The future of osteochondral junction repair lies in biodegrad- able scaffolds and autologous cells to provide a mechanically functional hyaline repair tissue, that integrates well with the wound edges [291].
Table 5.1: Current surgical treatments for cartilage degeneration.
Surgical Treatment Description Total Arthroplasty Replacement of entire joint, generally carried out on the largest and most loaded joints of the body, the hips and knees. Two main categories of prosthetic implants: synthetic and biological. Syn- thetic implants can be metallic, polymeric, ceramic or a combina- tion. Biological implants are either Auto-/Allo-/or Xeno-grafts. Arthroscopic Lavage The irrigation of a joint with solutions of sodium chloride, Ringer or Ringer and lactate to remove any fluid or loose debris [292]. Shaving Mechanical removal of diseased chondral tissue [293]. Debridement The removal of a greater portion of chondral tissue than that is done in the shaving procedure [294]. Arthroscopic Abrasion A multiple tissue debridement technique developed by Johnson Arthroplasty during the 1980s, where superficial abrasions are only carried out on the exposed bone, not on any area of intact degenerative artic- ular cartilage, to remove dead osteones, exposing viable bone and the surface vascularity [295].
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 122 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.1. Introduction
Surgical Treatment Description Drilling/ Subchondral The drilling of holes (2.0-2.5mm in diameter) into subchondral penetration bone under regions of damaged articular cartilage, conceived by Pridie in 1959 [296], to allow the release of marrow elements, pro- viding an enriched environment for new spontaneous tissue for- mation repairing damage. Microfracture Modification of the drilling technique by Steadman and colleagues where perforactions into the underlying bone are made as close together as possible, usually 3 to 4 mm apart, up to 4mm deep and with a diamter of 0.5-1.0mm [297]. Perichondral/ Periosteal The chondrogenic potential of periosteum, arising from its cam- graft bium layer which contains chondrocyte precursor cells that form cartilageduringlimbdevelopmentandgrowthinutero,doesso once again during fracture healing, making it useful for the repair of damaged cartilage [298]. Autologous osteochondral A technique developed by Hangody in 1991 where multiple small transplantation: Mosaic- cylindrical plugs (ranging from 2.7 to 8.5 mm) of bone with at- plasty tached cartilage surface are harvested from the relatively less weight bearing periphery of the patellofemoral joint being treated are placed into the defect allowing a 90-100% filling rate. Packing closely together allows reformation of cartilage surface [299]. Autologous chondrocyte Transplantation of terminally differentiated transplanted chondro- transplantation cytes; chondrocyte percursor cells from the grafted periosteum; or MSC from the subchondral bone marrow space; first described by Brittberg et al. in 1994 [300], often combined with a periosteal membrane which acts as a source of cells, a scaffold for delivering and retaining them, and a source of local growth factors [301]. Allogeneic Osteochondral The substitution of healthy articular cartilage, usually derived grafts from cadavers. Chondroectomy/ Chon- Not particularly popular, clinically or experimentally, due to a dral grafting lack of suitable donor sites and difficulty that arises from ade- quately securing purely chondral tissue transplants [302].
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 123 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.1. Introduction
5.1.2 Engineering of osteochondral grafts
Currently, approaches to many gradient requiring tissue grafts, such as the osteochon- dral junction, are combinations of cell and scaffold strategies. Table 5.2 illustrates com- mon combinations currently used in the production of osteochondral junction grafts, which is shown schematically in Figure 5.1.
Table 5.2: Cell and scaffold strategy combinations for osteochondral junction repair.
Scaffold options Cell options (A) A scaffold for the bone component, whilst (I) The scaffold is loaded with a single cell the cartilage component remains scaffold free. source having chondrogenic capacity.
(B) Different scaffolds for both the bone (II) The bone component is loaded with and cartilage components, which are com- cells having osteogenic capacities, and the bined at the time of implantation. cartilage component is loaded with cells having chondrogenic capacities. (C) A single scaffold is produced for the graft, but is a heterogeneous/bilayered com- (III) The entire scaffold is loaded with posite where the bone component differs to a single cell source having both chondrogenic the cartilage component. and osteogenic differentiation capacity.
(D) A single homogeneous scaffold for (IV) Cell-free approach. both components.
A B C D No Scaffold Scaffold structure Bone layer scaffold Cartilage layer scaffold Same scaffold for both layers I II III IV Cell free scaffold Cell Chondrogenic cells arrangement Osteogenic cells Common cell source
Figure 5.1: Schematic representation of currently used approaches for the preparation and fabrication of osteochondral grafts, modified from Martin et al. [107].
It is quite common to use different scaffold bases for the bone and cartilage components, due to the differing nature and regenerative potential of the two tissue types. A variety of scaffolds are used in the TE of bone and cartilage [303–306], with a recent review by
Keeney and Pandit [307] summarising synthetic and natural scaffolds used in bilayered
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 124 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.1. Introduction approaches [308] (scaffold structure C in Figure 5.1). The number of groups reporting the use of MSC in such scaffolds [309–315] is rising. Ideally, for osteochondral repair, a graft combining D and III would be most preferable, as the osteochondral junction is a natural gradient. Two distinct tissue types come together with an integrated junction, rather than an abrupt change in tissue and cell phenotype, as is the case with biphasic scaffolds. Structural integrity upon implantation becomes an issue, due to mechanical loading.
5.1.3 Bioreactors to generate tissue gradients
The establishment of in vitro gradients makes possible the study of biological problems in the areas of migration, proliferation, development, and differentiation that, in vivo, occur along a gradient.
Whilst gradient generation plays an important role in the study of chemotaxis, a pro- cess of directed cell migration in chemoattractant (soluble molecule) gradients funda- mentally important in cancer metastasis [316], embryogenesis [317], and wound heal- ing [318], the TE implications lie in directed differentiation.
It is becoming recognised that the development of appropriate multi-parametric in vitro culture systems for the formation of clinically applicable hierarchically organised and functional custom-designed tissue grafts before their in situ implantation is required
[319]. The application of bioreactors for the purpose of producing a tissue engineered osteochondral graft has been highlighted by several authors [289,320–323], with several groups now reporting development of specific devices for this purpose. Approaches include the use of a double chamber geometry to produce the two tissue types [324,325], and the use of a bilayered scaffolds (scaffold structure C, cell arrangement II, in Figure
5.1), with recirculating medium [145].
The goal of this thesis is to develop a single chamber bioreactor, with a controllable gradient, capable of producing a graded graft from a single scaffold source, using a single progenitor cell source (scaffold structure D, cell arrangement III, in Figure 5.1).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 125 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.1. Introduction
5.1.4 Approach and Aims
Theaimsofthischapterareto:
• Evaluate the design of multi-channel single chambered bioreactor devices.
• Test the ability of these devices to produce a stable controllable gradient inside
a chamber suitable for cell maintenance and growth.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 126 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.2. Materials
5.2 Materials
Polycarbonate perspex sheeting (800mm thick) was obtained from Allplastics Engi- neering, Sydney Australia. Glass slides (7101-BP) were obtained from Livingstone
International, Sydney Australia. RV Silicone adhesive (MED-1000) was obtained from
Nusil Silicone Technology, USA. Winged infusion sets (1SV*19BLS, 3SV*19BLK) were obtained from Terumo Medical, Sydney Australia. Hollow fibres were cut from an
Asahi Plasmasflo plasmapheresis module, with a molecular weight cut off of 3 x 106
Da, obtained from Asahi Kasei Kuraray Medical, Japan. Silicone gaskets were cut from Silastic Sheeting (Subdermal Implant Material, 500-7), obtained from Dow Corn- ing, Michigan USA. Membranes were cut from Durapore ® Membrane Filters (0.2μm
GVPP, SAIJ065H7), obtained from Millipore, Sydney Australia. Masterflex Pharmed pump tubing (06458-14) manufactured by Saint Gobain, Ismatec Tygon R-3603 3 col- lar pump tubing (95608-10, 95608-12), Tygon R-3603 laboratory tubing (06408-62),
Miniature Barbed Polypropylene Fitting (connectors, 6365-90), Masterflex pump con- sole drive (model 7521-57), 12 channel 8 Roller Pump Head (07519-25), and small cartridges (07519-85) were all obtained from Cole-Palmer Instrument Company, IL
USA through Extech Equipment, Melbourne Australia.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 127 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
5.3 Cross-flow diffusion chamber
5.3.1 Introduction
A cross-flow bioreactor was developed where a gradient could be set up between zones perfused with different culture media constituents. The bioreactor was comprised of two ploycarbonate blocks, two porous membranes, and a gasket. Flow channels were milled into the polycarbonate blocks, and fluid flow was separated from the cell chamber by the porous membrane. The cell chamber, encased by a gasket, was sandwiched between an upper and lower membrane which permitted medium to freely perfuse through, yet not allow the contents of the cell chamber to be washed away, as illustrated in Figure
5.2. The graft thickness is dictated by the gasket depth. Medium flow both above and below the graft, allows medium to perfuse into the graft from both sides, permitting the development of a thicker graft.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 128 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
Figure 5.2: Cross-flow bioreactor design.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 129 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
Figure 5.3: Cross-flow bioreactor design, detailed chamber view. A solid boundary exists between the two medium zones, comprised of 4 and 2 channels.
Four chambers were constructed in parallel, to allow the simultaneous production of four grafts. Each upper and lower block consists of six channels, four channels to accommodate one particular medium type, and two channels for another, thereby es- tablishing a gradient inside the cell chamber. Medium is supplied from a single source for all channels of the same type in each flow chamber, with branching occurring to provide flow down each individual channel. Medium passes from one chamber to the next as seen in Figure 5.4.
Flow chamber 1
Graded zone
Flow chamber 2
Device 1 Device 2 Device 3 Device 4
Figure 5.4: Cross-flow bioreactor design schematic representation of fluid flow in both top and bottom channels encasing cell chamber. Arrows indicate fluid flow from one device to the next. The ridges between the two medium zones, and between the flow chambers sits on the porous membrane. Exaggerated view, not drawn to scale.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 130 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
Detailed drawings can be found in Appendix C.
This set up allows the use of different medium types in both the upper and lower chamber. For example, the upper chamber can be used to supply grow factors and can
flow at a slower rate, while the lower chamber can be used with basal growth factor free medium, and can flow at a faster rate. In theory, flow from both sides can provide sufficient nutrient diffusion into the cell chamber and prevent the development of a necrotic centre.
5.3.2 Construction and assembly
Fluid flow channels were machined into the polycarbonate block on a milling machine using parallel traversing tables with a 1mm cutter in house. Inlet and outlet holes to each chamber were also cut using the milling machine. The cell chamber gasket and membranes were cut to size using a scalpel. The metal shafts of 19G disposable needles and Tygon R-3603 laboratory tubing were used to connect each parallel flow chamber, as seen in Figures 5.5 and 5.6.
Figure 5.5: Cross-flow bioreactor design, side view.
Figure 5.6: Cross-flow bioreactor design, top view.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 131 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
5.3.3 Evaluation of design
A number of design issues were highlighted with the manufacture of this initial device.
The device is held together with many screws, which has several implications. Not only are there ergonomic issues with assembling the device, but it is difficult to assemble without disturbing, damaging or contaminating the scaffold inside the cell chamber.
Even though the device was firmly clamped together with screws at approximately 1 cm spacing, there was leakage of media along the plain of the gasket.
The scaffold in the cell chamber was required to compress the porous membrane against the milled flow channels. When this does not occur, the membrane comes away from the channels and the fluid takes the path of least resistance, shorting across the channels to the outlet. As a result, medium exchange did not occur in the entire chamber. A well defined gradient was not established between the parallel cross flow perfusion zones.
The cross-flow diffusion chamber bioreactor was simplified in order to exert greater control over the fluid path. This was achieved by concentrating on a single device with separate parallel flow paths and a gasket that would create a better seal around the porous membrane. The same principle of a cell chamber encased in a gasket, layered between porous membranes and ploycarbonate blocks with flow channels was followed, as is illustrated in Figure 5.7.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 132 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
Figure 5.7: Cross-flow bioreactor design 2.
The fluid flow path was altered such that flow was individually controlled and restricted to each specific channel. The was achieved by placing an input and exit port at the beginning and end of each channel. Further views can be found in Appendix D.
A 1.2mm slitting saw was used to cut 5 fluid channels into the polycarbonate blocks.
Each slit was 18mm long, 1mm deep, 1.2mm wide and a 0.5mm gap was left between each slit. Vertical holes were drilled into the polycarbonate block at the beginning and end of each channel, with one side having 0.6mm diameter holes and the other with
0.8mm diameter holes. This diameter difference was to accommodate either a 23G or
21G needle or winged infusion set, hence causing a difference in flow rates in the top and bottom block. G clamps were used to further tighten the assembled bioreactor and prevent leakage.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 133 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber
Figure 5.8: Cross-flow bioreactor design 2, view of channels from inner side.
Figure 5.9: Cross-flow bioreactor design 2, assembled view with membrane and gasket.
5.3.4 Results and Discussion
A cell growth inhibition assay (as described in Section 7.2.7) was carried out on all components used in the multi-channel bioreactor device (data not shown). Testing with mouse L929 fibroblasts did not show cell growth inhibition greater than 30% (the usual measure of cytotoxicity for this test) on any of the materials used. As a result, further toxicity testing was carried out on the assembled device. Briefly, the device was sterilised via ethylene oxide and once ready for use was flushed with a 10% BSA in
PBS solution. Cell growth medium was pumped through the device, and samples were collected over a 3 day time period. Samples were frozen until all samples had been collected and a cell growth inhibition assay (data not shown) was conducted using
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 134 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.3. Cross-flow diffusion chamber passage 2 primary rat cells. No inhibition greater than 30% was observed.
As both the upper and lower block both contain five fluid channels, using the current design set up, a multi-channel pump head is required to run each device.
The large bolts on either side of the cell compartment and fluid channels require align- ment for assembly, and again the possibility of disrupting, damaging or contaminating the cell chamber occurs from assembly.
Whilst the assembled device is held together with G clamps, which prevented leakage from the fluid channels and cell chamber, there was incomplete sealing at inlet and outlet flow ports. The channel ports were designed such that the hole had a slightly smaller diameter than the shaft to be inserted, ensuring a tight fit, and preventing leakage. However, loosening of the tight fit of the shaft into the entry hole over time lead to media leakage at the join. A number of gluing methods were used to attempt to seal this leak, but were unsuccessful.
This design has five separate flow channels that can be grouped together and used for the establishment of a gradient in ratios of 1:4 or 2:3. Even though each parallel flow path had separate inlet and outlet ports, there was flow between parallel flow paths because the scaffold did not exert a sufficient compressive force to prevent separation between parallel polycarbonate channels and the porous membrane. Without sufficient support from the membrane, the flow path is disrupted and fluid exchanges freely across the channels, making the generation of a defined controllable steady gradient difficult.
This design is still reliant upon the presence of a scaffold inside the cell chamber, and is limited in its applications. There is no option of developing a scaffold free cell graft inside the chamber.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 135 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
5.4 Hollow fibre diffusion chamber
5.4.1 Introduction
An approach using hollow fibres, rather than a membrane, to separate fluid flow from the cellular construct and to generate a gradient inside a single chamber was adopted.
The design was based on a bioreactor described by Sauer et al. [326,327] where two glass slides encasing semipermeable hollow fibres are glued together with silicone beading.
The hollow fibres are the fluid flow channels.
The device described by Sauer et al. was originally developed to provide a means of assessing cell growth at any time during the culture period. Thus the culture device was transparent and could be mounted on a microscope stage. This design was modified so that a chemical gradient could be established between two independent hollow fibre bundles, and is shown in Figure 5.10.
Figure 5.10: Hollow fibre diffusion chamber.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 136 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
5.4.2 Construction and assembly
Bioreactors were constructed following a method outlined by Sauer et al. [326]. Glass slides were used in order to keep cells in the desired scaffold rather than encourage migration out of the scaffold and adherence to the glass. The inner cell chamber was formed by applying a bead of silicone around its periphery (Figure 5.11). The cell seeding ports and 24 hollow fibres with a gap in the middle, were mounted within the wet silicone, which was left for 2 hours at room temperature to set.
Figure 5.11: Silicone beading applied to form the inner cell chamber of bioreactor, with seed- ing port tubing and hollow fibres lain in to the wet beading.
The hollow fibres were cut such that they extend approximately 3 mm past the dried beading forming inner compartment. Silicone beading was then applied along the edges of the slide to form the medium inlet and outlet chambers. Silicone tubing was laid into the wet beading, giving access to the medium chambers, as shown in Figure 5.12, and left to dry for 2 hours at room temperature.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 137 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
Figure 5.12: Silicone beading is applied to the outer edges of the slide to form the medium inlet and outlet chambers.
Silicone beading was applied over the top of all existing beading, both the inner cell compartment, and the medium compartments, and over the tubing. Another glass slide was gently placed on top, and held in place for a few seconds to allow the beading to attach, and left to dry overnight at room temperature. Once dry, 19 gauge winged infusion sets were attached to the medium compartment inlets and outlets, by passing the needle into the tubing comprising the compartment ports, leaving the luer lock end free. Bioreactors were tested for functionality, with any leaks fixed with silicone beading, and sterilised via ethylene oxide sterilisation by the Sterilisation Unit, at the
Prince of Wales Hospital, Randwick.
5.4.3 Gel Removal
Bioreactors were cut open along all silicone edges with a scalpel. The top slide was carefully removed, whilst leaving the gel intact. The gel was then cut free and lifted out with the hollow fibres. The gel was embedded and sectioned as described in Appendix
A Section A.3.2.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 138 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
5.4.4 Results
A cell density study was carried out to determine the effect of seeding density on histological processing. Passage 7 rat bone marrow stromal cells were seeded into hollow
fibre bioreactors at densities of 75 x 103, 200 x 103, 500 x 103,and1x106 cells/mL in a 12mg/mL kangaroo tail collagen (manufactured as per Appendix A.2.1.2) and 25mM
HEPES supplemented EMEM solution free of FBS. A representative device is shown in Figure 5.13. Bioreactors were placed at 37℃ overnight (without any perfusion) to allow the collagen to crosslink to form a gel and were then fixed with 10% formalin overnight. The gel was removed from the bioreactors as per Section 5.4.3, and embedded in agarose and sectioned as per Section A.3.2. Sections were cut at a thickness of 4μm and 10μm, and were stained with H&E as per Section A.3.4.3. The 10μm sections appeared brighter and showed clearer detail than the thinner 4μm sections (result not shown). Differential interphase contrast (DIC) images were captured using an Olympus
DP70 digital camera attached to an Olympus BX51 microscope under a x10 objective of both the cells prior to fixing whilst still in the bioreactor, and the 10μm H&E stained sections. Results obtained are shown in Figure 5.14.
Figure 5.13: A representative image of a slide hollow fibre bioreactor with a collagen gel.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 139 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
A B
C D
E F
G H
Figure 5.14: Effect of cell density on histolgical processing. Passage 7 rat bone marrow stro- mal cells were seeded into hollow fibre bioreactors at densities of 75 x 103 (A and B), 200 x 103 (C and D), 500 x 103 (E and F), and 1 x 106 (G and H) cell- s/mL in a 12mg/mL collagen and 25mM HEPES supplemented EMEM solution free of FBS. Bioreactors were incubated overnight at 37℃ andthenfixedwith 10% formalin overnight. The gel was removed from the bioreactors, embedded in agarose, sectioned (10μm), and stained with H&E. Representative DIC im- ages captured with a x10 objective of the cells prior to fixing whilst still in the bioreactor are shown in A, C, E and G; and the corresponding H&E stained cross-sections of the fibres and cells in the extra capillary space are shown in B, D, F and H, with the pink circles the fibre lumen. Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 140 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.4. Hollow fibre diffusion chamber
5.4.5 Discussion
At the time of manufacture, it was noticed that the tubing did not adhere well to the wet silicone beading, and would often come lose once the beading was dry. Issues of sterility were raised from the tubing repeatedly coming away from the silicone beading.
Silicone beading was reapplied over the tubing ports in an attempt to hold the tubing in place, but this approach was quite messy (as can be seen in Figure 5.13) and not completely satisfactory. The issue of loss of sterility would possibly be a problem in long term culture.
A significant shortcoming of the manual manufacturing process was imprecise device geometry.
From the initial cell density study, a number of issues with this design were highlighted.
An uneven distribution of cells in the collagen gel solution, leads to an uneven distri- bution of cells in the bioreactor device. With high seeding densities, such as 1 x 106 cells/mL, this effect is more pronounced and noticeable, but occurs with all seeding densities. The initial distribution of cells in a scaffold effects the final distribution of cells in the formed tissue [275]. If the gel and media solution are mixed too rigorously prior to injection, small amounts of air becomes entrapped, leading to small air bubbles inthegelwhichcannotberemovedlaterasthecollagenstartstosetsquiterapidlyat room temperature. It is preferable to avoid the introduction of air bubbles as this will disrupt the geometry of the cell culture chamber, with a non linear chemical gradient.
From Figure 5.13 the presence of large air pockets or bubbles can be seen quite clearly in the cell chamber. Air pockets such as these were present in all of the devices, and once present are difficult to remove as this set up provides no means of degassing the chamber once loaded.
The H&E stained sections of Figure 5.14 illustrate that even when cells are loaded into the chamber in a well mixed collagen gel, it is difficult to obtain a uniform cell distribution throughout the cell chamber. The same stained sections also show that
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 141 Chapter 5. Multi-channel bioreactor for engineering composite tissues from MSC 5.5. Conclusion when sections are cut, even after embedding in agarose, the collagen gel contracts, and pulls away from the fibres.
5.5 Conclusion
Many natural gradients exist in biology, and the generation of in vitro gradients allows the study of these often fundamental biological phenomena that occur along a gradient.
Bioreactor devices capable of providing an in vitro gradient facilitate both the explo- ration of these phenomena and also allow the formation of tissue grafts which contain multiple distinct tissue types, such as the osteochondral junction, providing a graded region between the tissue types.
The single chambered devices presented in this chapter do not have the ability to generate precisely defined and controllable steady gradients inside the cell chamber.
The cross-flow diffusion chamber devices presented employed a membrane to separate
fluid flow from the cellular construct. Flow channels were milled in polycarbonate blocks, and the devices were affected by leakage. Devices were reliant upon both internal and external compression for function, with leakage also occurring at fluid entry and exit ports. Reliance upon a scaffold inside the cell chamber limited application.
The hollow fibre diffusion chamber device presented employed hollow fibres to separate
fluid flow from the cellular construct. The manual manufacturing process resulted in imprecise device geometry and issues of maintaining sterility for long term culture were also raised.
A more robust manufacturing method is required for the production of such devices for research purposes, if controllable steady gradients are to be achieved. This manufac- turing method should be relatively simple, produce devices with a well defined precise geometry and also allow for the upscale of these devices. The field of microfluidics has the potential to satisfy all these criteria.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 142 Chapter 6
Microfluidic Bioreactors
6.1 Introduction
The field of microfluidics is emerging as a valuable research tool for the production and study of cellular micro-environments [328]. Whilst the use of microfluidic devices in TE is still in its infancy, contribution is expected to occur on two fronts: the production of larger complex tissue structures through the provision of continuous gas and nutrient supply and waste removal; and the opportunity to develop in vitro physiological systems for studying fundamental biological phenomena [329].
6.1.1 Microfluidic Advantages
Microfluidic based methods offer a variety of advantages over conventional methods
[330]. With the increased flexibility now available from soft lithography techniques
[228, 331–333], designing and building micro-scale devices has become possible in al- most any academic research laboratory as advanced expensive clean room facilities are not required for device fabrication [334]. PDMS is inexpensive, readily available as an off the shelf two component kit [335], biocompatible [232], elastomeric, permeable to respiratory gases [336], optically transparent and has low autofluorescence (when com-
143 Chapter 6. Microfluidic Bioreactors 6.1. Introduction pared with many plastics used in microfabrication) [337]. It is this feature that makes microfluidic devices particularly compatible with optical imaging and microscopy tech- niques [338] including phase contrast, differential interference contrast (DIC), epifluo- rescence and confocal microscopy [335]. The unique ability to provide live, real time microscopic observation [339–343] without destroying the device or interrupting the ex- periment in progress is an important step forward for studying biological phenomena, as traditionally measurements could only be taken at the end of the culture period.
The most obvious advantage of microfluidic devices lies with reducing the scale of the culture, and thus reducing sample reagent volumes. The average volume per cell can be reduced from one order of magnitude [344] to three orders of magnitude [345] smaller than in conventional culture systems. The use of nano-litre volumes [345] results in both a reduction of costs associated with expensive reagents such as growth factors and cytokines, and also leads to a reduction in waste production [338] and space.
Culture medium and reagents supplied to the device can flow from a cool source, such as a refrigerator, slowing degradation.
With a reduction in size comes the ability to upscale and array microfluidic devices, which facilitates high-throughput experimentation [338]. Whilst the ability to con- duct multiple parallel investigations is valuable [333, 346, 347], the ability to control parameters such as cell seeding density, composition of culture medium and feeding schedule of each individual cell growth chamber [235], is where the future of this tech- nology really lies. Inexpensive process automation with computerisation of pumps and valves [348, 349] on integrated microfluidic devices can perform rapid and repro- ducible measurements with small sample volumes while eliminating the need for labour- intensive and potentially error-prone laboratory manipulations [334].
Culture miniaturization also provides the ability to precisely control and manipulate the cellular micro-environment [342, 344, 350–352]. This can be achieved through various techniques such as surface patterning [230,330,335,353–355] which modifies the physical surface and results in various contact guidance manipulations [335]; the creation of mechanical strain, through shear, in the appropriate physiological range [334, 356]; or
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 144 Chapter 6. Microfluidic Bioreactors 6.1. Introduction from the generation of chemotactic gradients from the precise well-defined geometries of the devices [350].
6.1.2 Microscale Effects
There are a number of physical effects that take place on the microscale that are not evident in conventional culture systems where cells are grown in relatively large volumes at low density (<2 million cells/mL). Microfluidic devices aim to mimic certain aspects of the in vivo cellular micro-environment; short distances between cells, supply and removal of metabolites by the micro-circulation, local shear stress, and morphogenic gradients required for tissue development. There are a number of mechanical, transport and material effects that must be considered in order to achieve successful cellular culture.
6.1.2.1 Mechanical forces
Shear Stress
In microfluidic devices it is common to flow medium past cells adhered within the microchannel [232]. The fluid flow generates shear effects on the cells. The relationship between shear stress and the velocity gradient of fluid flow in a channel is given by
Newtown’s law of viscosity: dv τ −μ (6.1) = dx where τ is the shear stress, μ is the fluid viscosity, v is the fluid velocity, and x is the position within the channel [357]. From equation 6.1 it follows that for a given maximum fluid velocity, shear stress in a smaller channel will be greater than in larger channel, as in the smaller channel there is a larger change in velocity over a smaller x, giving a larger value for the ‘shear rate’ dv/dx. The shear induces both a drag and torque on deposited or adherent cells. For a perfect sphere in contact with a wall, the drag and torque are directly proportional to the wall shear stress, and the square or
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 145 Chapter 6. Microfluidic Bioreactors 6.1. Introduction cube of the radius (a) respectively [358].
drag = 1.7009(6)πa2τ (6.2)
torque = 0.94399(8)πa3τ (6.3)
Thus fluid shear can be used to transport cells within the device by controlling flow rate through channels with precisely defined geometry. Additionally, because Reynolds numbers are low, the flow field is predominantly determined by viscous forces. Shear stress and pressure drops in microchannels scale linearly with channel flow rate, greatly simplifying the design of microfluidic networks.
Surface tension, bubble formation and dissolution
The introduction of bubbles to microfluidic channels must be avoided, because a channel is effectively blocked by an air liquid interface. Flow around bubbles disrupts the well- defined flow field generated by device geometry. The pressure (p) required to displace a bubble from a channel is related to the contact angle between the fluid and the channel material (θ), the liquid-air surface tension (γ), and the radius of the channel (r) [357]:
2γ cos θ p (6.4) = r
Thus the required pressure to clear the channel of bubbles is inversely related to the radius of the channel. Culture aeration is generally not required for microfluidic devices if gas permeable polymers such as PDMS are used, as bubbles rapidly dissolve if the channel pressure is greater than the pressure on the outside of the device (atmospheric pressure).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 146 Chapter 6. Microfluidic Bioreactors 6.1. Introduction
6.1.2.2 Transport phenomena
Laminar Flow
Reynolds number (Re) of a flow can be calculated by:
ρDv Re = (6.5) μ where ρ is the fluid density, μ is the fluid viscosity, D is the diameter or channel depth, and v is the average fluid velocity [359]. Re <2300 characterises laminar flow, whereas
Re >4000 characterises turbulent flow. For a Reynolds number less than 1 (creeping
flow), viscous forces are greater than inertial forces. In laminar flow, there is no bulk movement of fluid between streamlines. Flow in microscale devices is generally laminar and viscous forces dominate.
Diffusion
The process of diffusion occurs when particles or molecules spread from areas of high concentration to areas of low concentration and can be calculated from Fick’s first law of diffusion: dC J −D (6.6) = dx where J is the flux of particles, D is the mass diffusion coefficient or mass diffusivity,
C is the molar concentration and x is the position [360]. The minus sign indicates that transport is from high to low concentrations. This form of Fick’s law is acceptable if the concentration varies only in one dimension (x-direction) and the solute is dilute.
Fick’s law is also expressed in three dimensions by the vector equation:
J = −D∇C (6.7)
Diffusion becomes the dominant transport mechanism only at long time scales and/or
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 147 Chapter 6. Microfluidic Bioreactors 6.1. Introduction short distances as illustrated by the Einstein-Smoluchowski equation:
d2 =2Dt (6.8) which relates the mean distance, d, a particle with diffusion coefficient, D,candiffuse in time t [361]. The Stokes-Einstein equation relates diffusion coefficient to molecular radius: kT D = (6.9) 6πμr where k is Boltzmann’s constant, T is temperature, μ is fluid viscosity and r is the particle radius [360]. Thus small molecules diffuse faster than large molecules because their viscous drag in the fluid is lower.
The effect of diffusion is more prominent in microscale devices, than in macroscale cell culture systems where convection, the flow of matter due to bulk flow of a fluid or directed flow, is the dominant transport mechanism. The ratio between mass transport due to convection and that due to diffusion is given by the dimensionless P´ecletnumber
(Pe): vl Pe = (6.10) D where v is the average fluid velocity, l is the characteristic length, and D is the diffusion coefficient of the particle [361]. Large P´eclet numbers (Pe 1) occur when convec- tion dominates, and small P´eclet numbers (Pe 1) occur when diffusion dominates.
Generally, the length and width of a microchannel, the average speed of the working
fluid as well as the diffusion constants of the molecules of interest are known, allowing calculation of the P´eclet number and determination of the effect of diffusion compared to directed flow.
6.1.2.3 Material interfaces
The miniaturization of culture devices can highlight phenomena that are not generally present in macroscale cell culture. Micro-devices have a large surface area to volume
(SAV) ratio, which can lead to surface adsorption issues, mainly with proteins. The
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 148 Chapter 6. Microfluidic Bioreactors 6.1. Introduction adsorption of proteins is a particular problem, as this depletes the culture medium of components necessary for successful cell maintenance and proliferation. Factors se- creted by the cells for normal cellular function are also depleted. Many of the materials used in device manufacture are hydrophobic, which can contribute to surface adsorption effects [362]. Depletion of protein concentrations affects subsequent culture conditions.
Growth of cells at very low density is also a problem. For a single hollow fibre in a culture well (see Section 4.3.3), the cells were grown at an effective concentration of around 200 cells/mL. Cells grown at low density are more susceptible to toxic compo- nents caused by oxidative degradation of culture media components [116]. Microfluidics can deal with this as culture perfusion rates can be scaled down to levels similar to those found in tissues.
6.1.3 Microfluidic Gradient Generation
With the use of microfluidics, gradient generation and control relevant to biological cell size (10 - 100 μm) has become an easier undertaking. A variety of microfluidic devices with various geometries have been developed for the generation of cellular gradients
[333, 363–368].
Kim et al. [363] describe a micro device which provides a 3 dimensional micro-environment suitable for cell culture. A linear concentration gradient profile due to molecular diffu- sion is formed across the cell chamber. The development of this device was however not undertaken with the aim of tissue manufacture, but rather as a potential application for cell-based screening and assays. As gradient generation relies solely upon molecular diffusion through the cell construct, the application of this device for the purposes of thick tissue graft development is limited as cells can be washed away. There is no membrane separating the cell layer from flow.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 149 Chapter 6. Microfluidic Bioreactors 6.1. Introduction
6.1.4 Approach and Aims
This chapter presents two microfluidic devices for generating chemical gradients; a pyramidal gradient generating device which has previously been described in literature
[369], and counter flow diffusion chamber which uses a novel design.
Both micro devices presented allow the formation of a stable concentration gradient across a single cell culture chamber. The micro devices have a precise geometry and allow for real time image capture. The devices are PDMS based, with replicate devices manufactured easily. These devices offer a number of improvements over the devices presented in Chapter 5.
Theaimsofthischapterareto:
• Experimentally set up and measure the concentration gradient produced by a
pyramidal gradient generating device composed of serpentine channels reported
in literature, and to determine the effect of flow rate on gradient generation.
• Experimentally set up a concentration gradient inside a novel diffusion chamber
produced by counter flow and to determine the effect of flow rate on gradient
generation.
• Determine the effect of molecular weight on experimentally obtained flow rate
profiles for the counter flow diffusion chamber.
• Conduct experimental evaluation of the predictions of a simple model to describe
counterflow separation.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 150 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
6.2 Materials and Methods
6.2.1 Materials and Equipment
Sylgard® 184 Silicone Elastomer (main component polydimethylsiloxane (PDMS),
Base (3097366-1004), Curer (3097358-1004)) was obtained from Dow Corning, MI USA.
0.4μm PET membrane (PIHT30R48) was obtained from Millipore, Sydney Australia.
0.75mm Harris Uni-Core™ punch (T980-075) was obtained from ProSciTech, QLD Aus- tralia. Fluorescein (F-6377), FITC 20 (Fluorescein isothiocyanate-Dextran, FD-20s) and FITC 150 (FD-150s) were all obtained from Sigma Aldrich, Sydney Australia.
Laboratory corona treater (BD-20ACV) was obtained from Electro-Technic Products,
IL USA. Model ‘11’ plus dual syringe pump with serial communication (MA1 70-
2212) manufactured by Harvard Apparatus, MA USA, was supplied by SDR Clinical
Technology, Sydney Australia. Fluorescence microscope AXIOSKOP2 MAT mot, with colour camera AxioCam mRc5, and 2.5x Plan-NEOFLUAR objective were all obtained from Zeiss, Germany.
6.2.2 Methods
6.2.2.1 Master Mould Manufacture
Chromium masks were produced from AutoCAD drawing by the Bandwidth Foundary,
Sydney, Australia. Development of the master mould by SU8 lithography techniques
(exposure of light sensitive photoresist SU8 spun onto on a silicon wafer to UV light using the photomask) was performed1 at the UNSW Semiconductor Nanofabrication
Facility, Sydney Australia as per the method outlined in Appendix E. The master moulds were a thickness of 300μm, as measured by a Dektak 3030 surface profiler.
1by Owen The (staff, Graduate School of Biomedical Engineering, University of New South Wales) and Huaying Chen (PhD candidate, Graduate School of Biomedical Engineering, University of New South Wales)
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 151 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
6.2.2.2 Corona treating of PDMS surfaces
In order to bond PDMS to PDMS, surfaces were treated with discharge from a corona treater using an established method [370]. Briefly, the surfaces to be bonded were cleaned with isopropanol and placed in a fume hood on top of glass slides, bonding side up. All metal objects and flammable solutions were removed from the immediate vicinity. The output voltage of the corona treater was adjusted such that a soft and stable discharge (purple in colour) was produced from the electrode. The electrode was passed back and forth approximately 1cm above each bonding surface for thirty seconds. Bonding surfaces were gently pressed together, whilst ensuring that no air was trapped in between. Surfaces were left under weights overnight.
6.2.2.3 Pyramidal Gradient Generating Device Manufacture and Assembly
40 mL of PDMS (9 parts base mixed with 1 part curing agent for 15 minutes) was poured onto the master mould with the required microfluidic gradient generator pattern. After bubbles were dissolved under a vacuum, the mould was baked at 60℃ overnight. An imprint free layer of PDMS was also cast. Once set, the imprint (approximately 5 mm thick) was removed from the wafer and cut to size. Holes were cut at the fluid entry and exit ports and also the cell inlets. A cover of equal size was cut from the imprint free casting. Surfaces to be bonded were treated with discharge from the corona treater as per Section 6.2.2.2. The plain PDMS was placed over the patterned surface and left to bond overnight under weights.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 152 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
Cell inlet Diffusion Diffusion Chamber Chamber Outlet Inlet
Inlet
Diffusion chamber
Cell inlet
Figure 6.1: Schematic drawing of the pyramidal gradient generating device. Flow channels are 350μm wide and 300μm deep. The cell diffusion chamber is 1680μmwide.
6.2.2.4 Counter Flow Diffusion Chamber Manufacture and Assembly
30 mL of PDMS (9 parts base mixed with 1 part curing agent for 15 minutes) was poured onto the master mould with the required microfluidic pattern of the cell chamber and drainage system. After bubbles were dissolved under a vacuum, the mould was baked at 60℃ overnight. Once set, the imprints (approximately 4 mm thick) were removed from the wafer and cut to size. The device was assembled in a 4 part process. The cell chamber was firstly reinforced with a thick (6-7 mm) layer of PDMS. The non-patterned side of the cell chamber and the reinforcing layer were both treated with discharge from the corona treater as per Section 6.2.2.2, assembled and left to bond overnight under weights. Once bonded, holes were cut through both the reinforcing layer and the cell chamber to create fluid entry and exit ports. The inner surfaces of the device with the cell chamber and drainage pattern were treated with corona discharge. A membrane, cut to size was aligned between the cell chamber and drainage coil. The assembled device was left to bond overnight under weights.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 153 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
Figure 6.2: Schematic drawing of the counter flow diffusion chamber. All flow channels are 200μm wide and 300μm deep. The cell chamber is 8000μm long, 3500μmwide and 300μm deep.
The set up shown in Figure 6.3 was used to ensure that all holes were punched as vertically and cleanly as possible without causing any tears in the PDMS, which would lead to leaking around the steel tubing once the system was fully pressurized.
Figure 6.3: Set up of assembly used to punch holes in PDMS.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 154 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
Once the device was assembled and all tubing connected, a pressure head was connected to the exit to ensure the degassing of bubbles from inside the chamber and outlet coils.
6.2.2.5 Quantification of concentration gradients by fluorescence microscopy
Concentration gradients were quantified using standard test solutions of the fluorescent dyes fluorescein, FITC20 and FITC150. A correcting gain matrix was calculated from
16 bit grey scale test images, using the method outlined below, and applied to all images taken.
The fluorescence microscope used for image capture displayed non-uniform pixel gain.
The effect, whilst present at all magnifications, was particularly pronounced at lower magnification levels, such as with a 2.5x objective. False colour gradient mapping (using
Photoshop, Figure 6.4) was applied to test the uniformity of 16 bit grey scale image intensity using a well mixed solution of fluorescein (Figure 6.5). It can be clearly seen that the lower right corner of the image has significantly higher high intensity signal as detected by the CCD camera (Figure 6.6).
Figure 6.4: False colour gradient mapping was used to remap greyscale images to colour, in order to enhance the visual perception of contrast in small variations in fluorescent intensity.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 155 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
Figure 6.5: Uncorrected test image of Figure 6.6: The uncorrected test image of 5μg/mL fluorescein exposed for 5μg/mL fluorescein exposed for 39.9ms, 2.5x. 39.9ms in Figure 6.5 shown using the false colour gradient map of Figure 6.4.
Such an effect is usually not of significant concern for the capture of biological images, where fluorescent intensity is generally not being quantitatively measured across the
field of view. For the intended purpose in this application, where images are required to provide precise measurement on the gradient generating capabilities of the micro-device, imaging techniques are required to detect small variations in concentration over a large
field of view. As a result, uniform signal gain across the field of view is paramount for the accurate measurement of fluorescent intensity.
The cause of non-uniformity of image intensity is either non-uniform illumination of the sample by mercury arc lamp optics, or non-uniform pixel gain of the CCD camera.
Once non-uniformity of illumination was minimised by realignment of the mercury arc lamp, software was developed to correct signal gain using standard concentrations of
fluorescent dyes.
The image intensity (or concentration) was examined using a set of standard curve solutions for fluorescein (1 to 10μg/mL), FITC20 (0.2 to 0.9mg/mL) and FITC150
(0.2 to 0.9mg/mL). The effect of CCD camera exposure time was examined using stan- dard test solutions (5μg/mL for fluorescein, 0.5mg/mL for FITC20, and 0.5mg/mL for
FITC150). Exposure times were long enough to obtain adequate signal to background
(no fluorescent dye) ratio, whilst not photo bleaching the fluorescent dye solution. For
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 156 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods each image captured, a 3mL volume was placed in a 35mm round Petri dish. All images were taken at the centre of the dish, to ensure an even field of fluorescence, and elim- inate any additional light being reflected from the edge of the dish. A 2.5x objective was used.
Multiplication of an image by the gain matrix corrects for non-uniformities introduced by the camera. Images obtained were processed in Matlab2. The GetGain function
(Appendix I.1) calculates a gain matrix from the images taken for both varying concen- trations and exposure times. The GainCal function (Appendix I.2) is used to perform this correction, using the matrix generated by the GetGain function. The results for concentration and exposure time are presented in Figures 6.7 and 6.8 respectively.
2Matlab functions written by Carl Gabel, MBiomedE student, GSBME, UNSW 2008
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 157 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
4 x 10 1.1 6
5 1.05 4 1
3 Gain
Mean Intensity 2 0.95 1
0 0.9 0 2 4 6 8 10 0 2 4 6 8 10 Fluorescein Concentration (µg/mL) Fluorescein Concentration (µg/mL)
4 x 10 1.1 5
4 1.05
3 1 Gain 2 Mean Intensity 0.95 1
0 0.9 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 FITC20 Concentration (mg/mL) FITC20 Concentration (mg/mL)
4 x 10 5 1.1
4 1.05
3 1 Gain 2 Mean Intensity 0.95 1
0 0.9 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 FITC150 Concentration (mg/mL) FITC150 Concentration (mg/mL)
Figure 6.7: Left hand panel: Mean pixel intensity vs Concentration for fluorescein, FITC20 and FITC150. Error bars denote ± one standard deviation. Right hand panel: Mean gain vs Concentration for fluorescein, FITC20 and FITC150. Error bars denote ± one standard deviation.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 158 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods
4 x 10 6 1.1
5 1.05 4
1
3 Gain
Mean Intensity 2 0.95 1
0 0.9 0 20 40 60 80 100 0 20 40 60 80 100 Fluorescein 5µg/mL Exposure Time (ms) Fluorescein 5µg/mL Exposure Time (ms)
4 x 10 6 1.1
5 1.05 4
3 1 Gain
2 Mean Intensity 0.95 1
0 0.9 0 50 100 150 200 0 50 100 150 200 FITC20 0.5mg/mL Exposure Time (ms) FITC20 0.5mg/mL Exposure Time (ms)
4 x 10 5 1.1
4 1.05
3 1 Gain 2 Mean Intensity 0.95 1
0 0.9 0 50 100 150 200 0 50 100 150 200 FITC150 0.5 mg/mL Exposure Time (ms) FITC150 0.5 mg/mL Exposure Time (ms)
Figure 6.8: Left hand panel: Mean pixel intensity vs Exposure time for fluorescein, FITC20 and FITC150. Error bars denote ± one standard deviation. Right hand panel: Mean gain vs Exposure time for fluorescein, FITC20 and FITC150. Error bars denote ± one standard deviation.
It can be seen quite clearly that a linear relationship exists between both mean pixel intensity and solution concentration, and between mean pixel intensity and exposure time, for all the molecular weight solutions. Therefore, a change in pixel intensity directly correlates to a change in concentration or exposure time. The relationship
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 159 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods between camera mean gain and both concentration and exposure time is constant across all molecular weight solutions. As a result, the gain matrix calculated from the exposure time images was used for image correction.
It should be noted that values for a 0% concentration solution were omitted from the calculation of pixel gain. Noise arising from the camera resulted in small but significant signal at zero concentration. Values for very small exposure times were also omitted
(data not shown), as the camera shutter had not fully opened within this time.
The result of image correction using the gain matrix is presented in Figures 6.9 and
6.10, where the entire field of view is now even and the lower right portion of the image in no longer disproportionately bright. All images taken were corrected using this process.
Figure 6.9: Image of 5μg/mL fluorescein ex- Figure 6.10: The corrected image of 5μg/mL posed for 39.9ms in Figure 6.5 af- fluorescein exposed for 39.9ms in ter correction. Figure 6.9 shown using the false colour gradient map of Figure 6.4.
6.2.2.6 Image Alignment and Collation
Even at the lowest magnification, with the 2.5x objective, the regions of interest for the microdevices studied did not fit in the field of view for one image, and as a result multiple images had to be taken. These images were then required to be stitched together into a single image, providing the entire field of view required. This was achieved by determining the point of overlap, aligning the images, and collating into a
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 160 Chapter 6. Microfluidic Bioreactors 6.2. Materials and Methods single image. Image alignment and collation for the Counter Flow Diffusion Chamber
(CFDC) employed the Matlab functions 3 outlined in Appendix I.3 and I.4, whilst the pyramidal gradient generating device employed the functions outlined in Appendix I.5 to I.10 to achieve this.
3written by Carl Gabel, MBiomedE student, GSBME, UNSW 2008
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 161 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3 Pyramidal Gradient Generating Device
6.3.1 Description of device
Jeon et al. [369] describe a gradient generating micro-device capable of forming a con- centration gradient using up to three different fluid input streams. A network of serpen- tine flow channels allows three parallel laminar streams to progressively separate and recombine, with segments for mixing by diffusion. The streams converge into a single outlet. A concentration gradient is developed perpendicular to the bulk flow direction.
The concentration profile is controlled by modifying the concentration and flow rate at inlet ports. Linear and nonlinear concentration gradients that could be controlled temporally and spatially were generated [364], whilst sawtooth profiles, periodic profiles and overlapping gradients of different solutes were also generated [371].
Human neutrophil migration has been successfully tested in a number of chemotactic gradients [350] using this design. Neural stem cells from the developing cerebral cortex were cultured for more than 1 week by exposing cells to a concentration gradient of growth factors under continuous flow [351].
Figure 6.11: Photograph of a gradient generated by the pyramidal gradient generator using red and blue dyes, at a flow rate 1μL/min. There is no input flow from the middle inlet.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 162 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
Figure 6.12: Illustration of diffusive mixing taking place inside the serpentine channels of the pyramidal gradient generator, at a flow rate of 0.25μL/min using a 0.1μg/mL fluorescein solution with the application of a false colour gradient map.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 163 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3.2 Results
6.3.2.1 Calculation of shear stress and P´eclet numbers
Wall fluid shear stress inside the cell chamber was calculated for laminar flow between parallel-plates for a Newtonian fluid using [372]:
3Qμ τ = (6.11) 2WB2 where τ is the shear stress, Q the flow rate, μ the fluid viscosity, W the width of the channel and B is half the plate separation (μ = 0.01 poise, W = 0.168cm, B = 0.015cm).
The diffusion coefficient for fluorescein in water was calculated using the Stokes-Einstein equation (Equation 6.9), where k = 1.3780 x 10−16, T = 310◦K, μ = 0.01 dyne- second/cm2,andr =33x10−8 cm [373].
The P´eclet number was calculated using Equation 6.10, where v = flow rate/cross sectional area of flow chamber (flow rate / 0.168cm x 0.03cm), l = 0.168cm, and D =
4.76 x 10−6 cm2/sec.
Table 6.1: Shear stress and P´ecletnumbers in the diffusion chamber of the pyramidal gradient generating device.
Flow rate (μL/min) Shear stress (dynes/cm2) P´eclet number 4 26.5 x 10−3 466.57 2 13.2 x 10−3 233.29 1 6.61 x 10−3 116.65 0.5 3.31 x 10−3 58.32 0.25 1.65 x 10−3 29.16
Cells in interstitial tissue are subject to shear stresses in the range 0-20 (dynes/cm2)
[333], illustrating that the shear developed inside the cell chamber is extremely low in comparison, even at the highest flow rate.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 164 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3.2.2 Theoretical gradient and P´eclet number
A theoretical model which predicts the concentration gradient generated inside the dif- fusion chamber of the pyramidal gradient generating device was developed4 (Appendix
F). The device is modeled as a single flow cell, with the boundary conditions shown in
Figure 6.13. The process is modeled assuming that there is uniform plug flow along the flow cell; the walls of the flow cell are impermeable to mass and there is zero
flux normal to the wall; the diffusive flux along the direction of the flow (x-direction) and between the plates of the flow cell (z-direction) is negligible compared to the flux across the streamlines (y-direction); and there is a linear concentration gradient at the entry to the flow cell. The model defines dimensionless variables for C (concentration normalised with respect to maximum concentration), X (distance along direction of
flow, normalised with respect to width of chamber), Y (distance normal to the flow, normalised with respect to width of chamber), and P´ecletnumber. The concentration gradient at the mid-line of the chamber, Y = 0.5, was calculated and is shown in Figure
6.20. The effect of P´eclet number on concentration is shown in Figure 6.14 (simulated data).
CX ,1 0 Y
Y 1
V x CYY0, CXY ,
Y 0 X 0
CX,0 0 Y
Figure 6.13: Dissipation of concentration gradient along diffusion chamber of pyramidal gra- dient generating device gradient model.
4by Dr. Robert Nordon, Graduate School of Biomedical Engineering, University of New South Wales
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 165 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
Pe=1 0 0.8 0.6
Y 0.5 0.4 0.2 1 0 2 4 6 8 10
Pe=10 0 0.8 0.6
Y 0.5 0.4 0.2 1 0 2 4 6 8 10
Pe=100 0 0.8 0.6
Y 0.5 0.4 0.2 1 0 2 4 6 8 10 X
Figure 6.14: Effect of P´eclet number on concentration produced inside the diffusion cham- ber of the pyramidal gradient generating device. Flow is occurring inside the diffusion chamber along the x-direction, and Y values of 0 and 1 represent the opposing edges of the chamber.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 166 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3.2.3 Experimentally obtained concentration profile
Whilst the pyramidal gradient generating device allows the formation of a gradient using up to three different fluid input streams, in this instance, the device was only tested using two different input streams. The first input port was supplied with a
0.1μg/mL fluorescein solution, the middle input port was left closed, while the third input port was supplied with water at the same flow rate as the port supplied with
fluorescein. The device was tested at flow rates of 4, 2, 1, 0.5 and 0.25μL/min. The resulting steady state non-dimensional concentration profiles at the cell chamber inlet and along the cell chamber are shown in Figures 6.15 and 6.16. These images show that a concentration gradient has been successfully established across the cell chamber of the pyramidal gradient generating device. All images were corrected as per Section
6.2.2.5, and aligned and collated as per Section 6.2.2.6. The resulting concentration profiles are shown in Figures 6.17 and 6.18. The concentration gradient is dissipated along the flow cell. This rate of dissipation is inversely related to the flow rate.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 167 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
(a) 4μL/min (a) 4μL/min
(b) 2μL/min (b) 2μL/min
(c) 1μL/min (c) 1μL/min
(d) 0.5μL/min (d) 0.5μL/min
(e) 0.25μL/min (e) 0.25μL/min
Figure 6.15: Concentration profile at dif- Figure 6.16: Concentration profile at dif- fusion chamber inlet using a fusion chamber exit using a 0.1μg/mL fluorescein solution in 0.1μg/mL fluorescein solution in the first port, water in the third the first port, water in the third port, and closed middle port. port, and closed middle port. Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 168 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
0.1 0.25 µL/min 0.5 µL/min 0.09 1 µL/min 0.08 2 µL/min 0.07 4 µL/min Top Edge 0.06 Bottom Edge 0.05
Concentration 0.04 0.03 0.02 0.01 0 0 500 1000 1500 2000 2500 Transverse position (µm)
Figure 6.17: Concentration profiles for pyramidal gradient generating device at diffusion chamber inlet for flow rates of 4, 2, 1, 0.5 and 0.25μL/min. A 0.1μg/mL fluo- rescein solution was used for the first input port, the middle input port was left closed, and the third input port was supplied with water at the same flow rate as the port supplied with fluorescein.
µ 0.1 0.25 L/min 0.5 µL/min 0.09 1 µL/min 0.08 2 µL/min 0.07 4 µL/min 0.06 0.05
Concentration 0.04 0.03 0.02 0.01 0 0 200 400 600 800 1000 1200 1400 1600 1800 Transverse position (µm)
Figure 6.18: Concentration profiles for pyramidal gradient generating device along diffusion chamber for flow rates of 4, 2, 1, 0.5 and 0.25μL/min. A 0.1μg/mL fluorescein solution was used for the first input port, the middle input port was left closed, and the third input was supplied with water at the same flow rate as the port supplied with fluorescein.
Comparing Figures 6.17 and 6.18, it can be seen that a change in concentration gradient profile takes place from the position where the cells are to be injected into the device, at the cell inlet, to a position further downstream towards the diffusion chamber exit. This
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 169 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device change, quantified at the mid-line of the diffusion chamber, is shown in Figure 6.19.
The comparison between experimental data and theoretical model (Section 6.3.2.2) is shown in Figure 6.20.
The theoretical model provides a reasonable approximation of the decrease in con- centration gradient along the flow cell. The dimensionless concentration gradient is calculated by dividing the concentration gradient at the outlet by the concentration gradient at the inlet.
Flow rate (µL/min) 0 0.5 1 1.5 2 2.5 3 3.5 4 0.0E+0
Inlet -2.0E-5 Outlet
-4.0E-5
-6.0E-5
-8.0E-5
-1.0E-4
Concentration gradient at midline, Y=0.5 -1.2E-4
Figure 6.19: Effect of flow rate and position along diffusion chamber on concentration gradi- ent at the mid-line (Y = 0.5) of pyramidal gradient generating device diffusion chamber.
1.2 Theoretical gradient at midline 1 Experimental data
0.8
0.6
Gradient at Y=0.5 0.4
0.2
0 0 0.2 0.4 0.6 0.8 1 1.2 X/Pe
Figure 6.20: Concentration gradient experimental data at mid-line of pyramidal gradient gen- erating device diffusion chamber superimposed upon theoretically modeled con- centration gradient. Theoretical gradient at mid-line produced from model de- scribed in Appendix F.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 170 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3.2.4 Visualisation of diffusive mixing at the diffusion chamber inlet
False colour mapping can be used to visualise mixing by diffusion at the inlet of the diffusion chamber of the pyramidal gradient generating device. It can be seen in Figure
6.21 that the blue and yellow streams diverge, whilst the red/orange/purple streams converge. This is the effect of diffusion of fluorescein across streamlines.
Figure 6.21: Visualisation of diffusive mixing at the diffusion chamber inlet of the pyramidal gradient generating device. The cell chamber inlet of the pyramidal gradient gen- erating device at a flow rate of 1μL/min, using a 0.1μg/mL fluorescein solution, with false colour mapping applied to highlight changes in pixel intensity.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 171 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device
6.3.3 Discussion
A microfluidic PDMS based device comprised of a network of serpentine flow channels
[369] has been used to demonstrate a concentration gradient profile consistent with literature [351]. The gradient generating device is capable of forming a concentration gradient, perpendicular to the direction of the laminar fluid flow. It was demonstrated that the single gradient shape formed in the outlet could be controlled by input fluid
flow rate, consistent with literature [369].
High fluid flow rate gave a high P´eclet number (Table 6.1). High P´eclet number when modeled showed a steep gradient inside the chamber (Figure 6.14), and the experimental data showed steep gradients for high flow rates at both the cell chamber inlet and outlet
(Figures 6.15a and 6.16a respectively).
Even though the device can be designed such that fluid shear stress is almost negligible, cells when detached would still roll along the flow cell. Adhesive factors are then required to prevent egress of cells from the device. The ideal design would create a concentration gradient without any shear stress. The counterflow design has the advantage of generating the steepest gradient at the flow stasis point (zero shear, see
Section 6.4.2.2).
Cellular attachment to PDMS will not occur without a surface protein coating. Chung et al. [351] demonstrated the proliferation and differentiation of human neural cells for more than 1 week, whilst exposing the cells to a concentration gradient of growth factors under continuous flow conditions. In this case, PDMS imprinted with the flow channels was laid over a glass slide, such that the glass slide formed the bottom of the channels.
The glass slide was treated with the protein laminin, to facilitate cellular attachment.
Similarly, the protein fibronectin has also been used to allow cellular attachment to
PDMS surfaces [235].
The potential of this design for use in the establishment of thick tissue grafts is restricted due to a number of geometrical device parameters. The serpentine flow channels con-
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 172 Chapter 6. Microfluidic Bioreactors 6.3. Pyramidal Gradient Generating Device verge to a single outlet with a gradient established across the width of the outlet. Cells are to be cultured along this outlet, and as the culture medium flows directly over the top of the cells, the thickness of the graft produced is limited. This design relies upon mass transfer through the cell layer, while convective flow occurs over the top of the cell layer. In a perfusion system, where the flow passes through the cell construct layer, culture medium and respiratory gases are supplied to the entire depth of the cell construct, whilst also ensuring adequate waste removal.
The device can be used to study responses to a chemotactic gradient, such as migration, proliferation, development and differentiation, but generally only for a single monolayer of cells, as this set up is reliant upon cell adhesion to the cell chamber. The capacity to form particular tissue types such as cartilage, where a high density cell pellet is generally required for the differentiation of progenitor cells into chondrocytes is limited.
As culture medium flows over the cell construct, any cells not anchored will be washed away with the flow. The geometry of the cell chamber will produce a long thin strip of tissue, dimensions not particularly useful or ideal for an implant from a clinical perspective.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 173 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
6.4 Counter Flow Diffusion Chamber
6.4.1 Description of Device
A novel micro-device capable of forming a steady concentration gradient in a zone with zero fluid shear stress was developed. The device is comprised of two PDMS castings and a porous membrane. The first casting has the imprint of a single cell culture chamber, and the second casting is imprinted with a drainage coil. The castings are separated by the membrane. There are two fluid input streams, from opposing sides of the cell chamber. Flow enters the cell chamber from each side, is pushed through the cell chamber, the membrane, and into the drainage coils. A concentration gradient forms at the zone of flow convergence where wall fluid shear stress approaches zero.
Cells will in theory accumulate in this region, if there is no attachment.
Figure 6.22: Photograph of a gradient generated by the counter flow diffusion chamber using red and blue dyes, at a flow rate 1μL/min.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 174 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
6.4.2 Results
6.4.2.1 Shear stress
Wall fluid shear stress at the entry to the parallel plate flow chamber was calculated using Equation 6.11, where W = 0.35cm, with all other parameters remaining constant.
Shear stress decreases towards the stasis point in the middle section of the diffusion chamber.
Table 6.2: Shear stress in the cell chamber of the counter flow diffusion chamber.
Flow rate (μL/min) Shear stress (dynes/cm2) 4 12.7 x 10−3 2 6.3 x 10−3 1 3.2 x 10−3 0.5 1.6 x 10−3 0.25 0.8 x 10−3
The shear developed at the entry to the counter flow diffusion chamber is similar to that generated inside the pyramidal gradient generator, but decreases towards the middle of the parallel plate flow cell.
6.4.2.2 Theoretical profile of interface
A theoretical model which predicts the concentration gradient generated inside the counter flow diffusion chamber was developed5 (Appendix G). This model assumes that there is no diffusion in the z-direction, that is from the upper cell chamber to the lower drainage chamber. The model also assumes that flow along the parallel plate flow cells is plug flow, and that the velocity decreases linearly. There is a stasis point in the middle of the flow cell with reversal in x-velocity direction. The model defines dimensionless variables for C (concentration normalised with respect to maximum concentration), and relates the P´ecletnumber linearly to the square of the distance from the origin (x
= 0). The concentration gradient as predicted by the theoretical model is shown in
5by Dr. Robert Nordon, Graduate School of Biomedical Engineering, University of New South Wales
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 175 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
Figure 6.23.
Fluid velocity Concentration gradient
9H9HX eX 2 /2 VX Pe CX 1 :I: I CX :1erf: II , 2;: ;J2 JI 2 1 4 0.9 0.8 3 0.7 concentration gradient 2 0.6 concentration 0.5 fluid velocity 1 0.4 0.3 0 0.2 0.1 -1 X Pe 0 -3-2-10123-0.1 -2 -0.2
-0.3 -3 -0.4 -0.5 -4
Figure 6.23: Concentration gradient for counter flow diffusion chamber as predicted by theo- retical model.
Assuming that there is no diffusion in the z-direction, linear flow gradient at stasis point generates a Gaussian concentration gradient with the steepest gradient at the zone with zero shear. This is an advantage over the pyramidal gradient generating device.
A theoretical model which predicts the hydrodynamic characteristics of the counter flow diffusion chamber was developed 6 (Appendix H). The device is modeled as two parallel plate flow cells separated by a porous membrane, with the boundary conditions shown in Figure 6.24, depicting a longitudinal section through the device. The model assumes opposing and equal flows q, enter the top flow cell, pass through a porous membrane
6by Dr. Robert Nordon, Graduate School of Biomedical Engineering, University of New South Wales
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 176 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber with hydraulic permeability κ, and exit the bottom flow cell through one outlet with
flow rate -2q. The model also assumes that the parallel plate flow cells have infinite width, so there is no pressure gradient in the z-direction (out of page) and that there is parabolic laminar flow between the two parallel plate flow cells. The non-dimensional
flow resistance to flow for the upper and lower flow cells (A1 and A2) are inversely proportional to the cube of the plate separation, directly proportional to the length of the flow cell, and directly proportional to the hydraulic permeability of the membrane.
As a result, the flow profiles generated by this model are independent of fluid viscosity, but dependent on plate separation, length of the flow cell and membrane permeability.
Flow profiles generated are shown in Figure 6.25.
BC1:qq 0 BC2 : ql q 1 vxFV px px 1 dp x m HX 1 rq x 12 11 b dx 2 1
dp x 2b 2 rq x 2 dx 22 qq ql BC4 :2 0 2 BC3:2 0
Figure 6.24: Elementary hydrodynamic model. View is of longitudinal section through the device.
A1 = 0.1 A1 = 1 A1 = 10 A1 = 100 1 1 1 1 A2 = 0.1 A2 = 0.1 A2 = 0.1 A2 = 0.1 0.8 0.8 0.8 0.8 A2 = 1 A2 = 1 A2 = 1 A2 = 1 0.6 A2 = 10 0.6 A2 = 10 0.6 A2 = 10 0.6 A2 = 10 A2 = 100 A2 = 100 A2 = 100 A2 = 100 0.4 0.4 0.4 0.4
0.2 0.2 0.2 0.2
Q 0 Q 0 Q 0 Q 0
-0.2 -0.2 -0.2 -0.2
-0.4 -0.4 -0.4 -0.4
-0.6 -0.6 -0.6 -0.6
-0.8 -0.8 -0.8 -0.8
-1 -1 -1 -1 0 0.5 1 0 0.5 1 0 0.5 1 0 0.5 1 X X X X
Figure 6.25: Flow profiles predicted by the hydrodynamic model for dimensionless flow in the upper cell chamber of the counter flow diffusion chamber. A1 and A2 are non- dimensional flow resistance variables (upper and lower flow cells respectively).
A number of effects can be observed from Figure 6.25. The flow rate gradient is linear if non-dimensional resistance A1 and A2 are less than 1; the interface moves towards
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 177 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber the outlet if A2>A1; and flow gradient at the mid-line is decreased for large A1, which means that the non-dimensional resistance can be used to control the size of the ‘stasis zone’.
6.4.2.3 Experimentally obtained profile of interface
In order to generate a concentration gradient in the cell chamber, one input was supplied with a fluorescent dye solution (fluorescein 10μg/mL, FITC20 1mg/mL and FITC150
1mg/mL) whilst the other input was supplied with water. Only one of the exit ports on the drainage coil was left open at a time, with the other blocked closed. The exit closest to the fluorescent dye input was termed the fluorescent drainage exit, whilst the exit closest to the water input was termed the water drainage exit. The device was tested at flow rates of 4, 2, 1, 0.5 and 0.25μL/min, with firstly the fluorescent exit open and then water exit open. All images were corrected as per Section 6.2.2.5, and aligned and collated as per Section 6.2.2.6, with the resulting steady state flow profiles shown in Figures 6.26 to 6.31.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 178 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
(a) 4μL/min (a) 4μL/min
(b) 2μL/min (b) 2μL/min
(c) 1μL/min (c) 1μL/min
(d) 0.5μL/min (d) 0.5μL/min
(e) 0.25μL/min (e) 0.25μL/min
Figure 6.26: Counter flow diffusion chamber Figure 6.27: Counter flow diffusion chamber flow profiles using a 10μg/mL flow profiles using a 10μg/mL fluorescein solution with the flu- fluorescein solution with the wa- orescent drainage exit open for ter drainage exit open for flow flow rates from 4 to 0.25μL/min. rates from 4 to 0.25μL/min.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 179 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
(a) 4μL/min (a) 4μL/min
(b) 2μL/min (b) 2μL/min
(c) 1μL/min (c) 1μL/min
(d) 0.5μL/min (d) 0.5μL/min
(e) 0.25μL/min (e) 0.25μL/min
Figure 6.28: Counter flow diffusion chamber Figure 6.29: Counter flow diffusion chamber flow profiles using a 1mg/mL flow profiles using a 1mg/mL FITC20 solution with the flu- FITC20 solution with the water orescent drainage exit open for drainage exit open for flow rates flow rates from 4 to 0.25μL/min. from 4 to 0.25μL/min.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 180 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
(a) 4μL/min (a) 4μL/min
(b) 2μL/min (b) 2μL/min
(c) 1μL/min (c) 1μL/min
(d) 0.5μL/min (d) 0.5μL/min
(e) 0.25μL/min (e) 0.25μL/min
Figure 6.30: Counter flow diffusion chamber Figure 6.31: Counter flow diffusion chamber flow profiles using a 1mg/mL flow profiles using a 1mg/mL FITC150 solution with the flu- FITC150 solution with the water orescent drainage exit open for drainage exit open for flow rates flow rates from 4 to 0.25μL/min. from 4 to 0.25μL/min.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 181 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
The flow profiles of three fluorescent dyes of varying molecular weights (fluorescein
(370 Da), FITC (20 kDa and 150 kDa)) were experimentally generated. It has been demonstrated that the concentration gradient profile is influenced by device flow rate, molecular weight of the fluorescent marker, and the direction.
There is obvious diffusion between the upper and lower chambers. The model ignores diffusion in the z-direction, but it can be seen that diffusion across the porous mem- brane is a significant factor influencing the concentration profile in the device. The concentration gradient is lower if the water drainage exit is left open. Thus it appears that there is diffusion of marker back into the upper chamber from the drainage coil in the downstream drainage region. This effect is more marked at low flow rates (see Fig- ures 6.26 versus 6.27, 6.28 versus 6.29, 6.30 versus 6.31). This effect is more important for lower molecular weight markers which have a higher diffusion coefficient (compare
Figures 6.27 versus 6.29 versus 6.31). The concentration profile for the highest molec- ular weight marker (150 kDa, lowest diffusion coefficient) was virtually unaffected by the drainage direction, indicating that this is indeed a diffusion related phenomenon.
If the fluorescent drainage exit is left open, a relatively high concentration gradient interface is generated, though there was some inconsistency with the position of the interface at lower flow rates (see Figure 6.30e).
It was found that the resulting flow profile was stable over time (overnight) at low flow rates 4 − 0.025μL/min (data not shown).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 182 Chapter 6. Microfluidic Bioreactors 6.4. Counter Flow Diffusion Chamber
6.4.3 Discussion
A novel micro-device capable of forming a steady concentration gradient was developed.
Flow profiles were experimentally obtained for three fluorescent dyes of varying molecu- lar weight. The main advantage of this configuration is that the concentration gradient is maximal in the region of flow stasis. Thus cell chemotaxis and tissue morphogenesis can be studied without the confounding effect of fluid shear stress.
This micro-device offers a number of advantages over the pyramidal gradient generating device. The cell culture chamber is not reliant upon cell adherence for operation. As a porous membrane separates the cell chamber from the drainage coils, the cells are trapped inside the chamber, ensuring that they will not be washed away with convective
fluid flow.
The device is not restricted to monolayer cell culture because cells can be concentrated at the stasis zone or along drainage channels (Chapter 7). Tissue types that are reliant on a high density of cells, such as cartilage, could be formed inside the chamber. The differentiation of progenitor cells into chondrocytes requires a high density pellet, the culture parameters of which this device is capable of providing.
As the width of the graded zone in the middle of the device can be controlled by device geometry, this device offers the ability to control the graded zone produced at the centre of the cell chamber. This would be achieved by increasing the non-dimensional hydraulic resistance of the upper (cell chamber), either by increasing the hydraulic permeability of the membrane, or by reducing the upper chamber plate separation
(Figure 6.25).
The model developed to predict the concentration gradient profile inside the diffusion chamber assumes that there is no diffusion in the z-direction, that is from the upper to the lower chamber. This assumption is clearly not correct, as diffusion into the lower chamber when flow direction is reversed is observed.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 183 Chapter 6. Microfluidic Bioreactors 6.5. Conclusion
6.5 Conclusion
A method was developed to accurately image concentration gradients at micron reso- lution.
Stable concentration gradient profiles of biological cell size have been experimentally generated in two micro-devices with differing geometries. It was demonstrated that the concentration gradient profiles and shear stress developed inside the cell chambers of both devices could be controlled by varying the input flow rate. The theoretical models developed were useful in predicting device behaviour, and in the case of the pyramidal diffusion chamber, experimental results were found to closely follow the model.
The optically transparent nature of PDMS used to manufacture the devices allowed for real time imaging. The second device has the advantage that it allows generation of a steep gradient in a region of flow stasis.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 184 Chapter 7
Biocompatibility and perfusion culture using the counter flow diffusion chamber
7.1 Introduction
There has been little change in conventional cell culture techniques over the last several decades. As in vivo cells respond to spatially and temporally organized signals in the surrounding micro-environment [344] the application of microfluidic devices for in vitro micro-environment development [329,334,338,374] for both basic biological phenomena investigation and tissue formation is being recognised as possessing great potential for
TE.
Microfluidic devices have already successfully been employed to culture a range of cells, as detailed in Table 7.1. They have also been utilized to generate a variety of different culture conditions, from driving cell growth and proliferation [333,341,342,345,351,367], to cell differentiation [341, 344, 348, 351], cellular behaviour [342, 349, 356, 368, 375–378] and cell function maintenance [241, 346, 379–382].
185 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.1. Introduction
Stem cell fates such as self-renewal and differentiation are particularly sensitive to the micro-environment [383, 384], and microfluidics offers potential opportunities of micro-environment manipulation and subsequent enhancement of stem cell cultures not possible using traditional techniques [367]. Microfluidic cell culture platforms have been used to differentiate progenitor cells [235, 385] and also provide the ability to co-culture cells [342, 346, 386, 387].
Microfluidic devices offer the ability to maintain long-term cultures, with a number of groups demonstrating systems suitable for cell culture in the range of 2 [342, 344, 379,
381], 3 [348] and 4 [346] week time periods. Reliable fluidic control in long term culture allows not only the study of biochemically delicate differentiation process, which take several days to complete, but also gives insight into other basic biological phenomena.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 186 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.1. Introduction
Table 7.1: Cell culture and growth in microfluidic devices
Cell type Description Ref cell line C2C12 murine skeletal muscle cells [341, 344, 348] HeLa human cervical cancer cells [333, 339, 347, 352, 376, 388] OC-2 ovarian endometrioid adenocarcinoma cells [343] MDA-MB 231 metastatic human breast cancer cells [330] MCF7 human breast cancer cells [242] Colo-205 colorectal adenocarcinoma cells [388] Caco-2 caucasian colon adenocarcinoma cells [388] NIH 3T3 murine fibroblasts [330, 333, 367] NIH 3T3-J2 fibroblasts [346, 386, 387] WT NR6 fibroblasts (3T3 derivative) [356] MC3T3-E1 mouse calvarial osteoblasts [241, 341] 3T3-L1 adipocytes [389] SY5Y human neuroblastoma [333] HepG2 human hepatocytes [242, 333, 380, 389, 390] C3A human hepatocytes [389, 391] L2 rat lung TypeII epithelial cells [389] A549 human lung adenocarcinoma epithelial cells [391] HMEC-1 human microvasuclar endothelial cells [381] RCL-2583 mouse endothelial cells, GFP [392] CHO DG44 cells [354] mouse (strain R1) embryonic stem cells, GFP [393] HuT 78 T human lymphoma cells [392] RPMI 8226 B human myeloma cells [392] primary human bone marrow stromal cells [235, 385] rat bone marrow stromal cells [242, 391] rabbit bone marrow stromal cells [335] ABJ1 mouse embryonic stem cells [367] mouse embryonic fibroblasts [393] human neutrophils [365] bovine endothelial cells [333] bovine aortic endothelial cells [377] HUVEC human umbilical vein endothelial cells [330, 378] HMVEC human dermal microvascular endothelial cells [349, 394] rat hepatocytes [242, 346, 382, 386, 387, 395, 396] human neural stem cells [351] rat embryonic neurons [342] E18 rat cortical neurons [330] mouse islet of Langerhans [375]
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 187 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.1. Introduction
7.1.1 Perfusion systems
Perfusion systems attempt to replace the function of microcirculation, permitting cells to be grown near tissue density. They allow the culture of high density cells, and the formation of 3D tissue structures with nutrients supplied to all parts of the construct.
Currently, many of the microfluidic devices classified as perfusion reactors [235,241,330,
343–346,348,349,352,356,363,367,375,376,382,385,386,396] generally have geometries such that flow occurs over the top of cells, and does not pass through the cell layer.
Devices specifically designed for 3 dimensional cell culture [391–394, 397] employ a variety of different geometries to deliver medium to the cell construct. High density culture devices which rely on diffusion from bulk flow across a porous barrier to the cell layer [340, 380] limit the thickness of the cell layer to the depth that nutrients can penetrate via diffusion.
When designing a device to facilitate thick tissue production and culture, such as those required for TE applications, convective flow through the cell construct will ensure gas and nutrient supply and waste removal from the entire construct, preventing the establishment of a necrotic centre where medium has not reached via diffusion. If the geometry of the device is altered such that the cells are seeded onto a microporous barrier or membrane [395], then medium can flow through the entire tissue construct, with the membrane preventing cell loss.
7.1.2 Approach and Aims
This system provides a scaffold free environment for the development of 3 dimensional tissue structures. The device allows perfusion of medium to the entire cell construct through the use of a microporous membrane.
Theaimsofthischapterareto:
• Asses device biocompatibility.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 188 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.1. Introduction
• Investigate any surface adsorption and/or protein depletion effects.
• Characterise cellular response to the counter flow diffusion chamber and flow
circuitry.
• Determine parameters necessary for cell viability in the counter flow diffusion
chamber.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 189 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials
7.2 Methods and Materials
7.2.1 Materials
Silastic Laboratory Tubing (thin silicone tubing, 508-002), Silastic Laboratory Tub- ing (thick silicone tubing, 508-004), Dow Corning silastic tubing (control silastic tub- ing, 601-365), Sylgard® 184 Silicone Elastomer (main component polydimethylsilox- ane (PDMS), Base (3097366-1004), Curer (3097358-1004)) were obtained from Dow
Corning, MI USA. Tygon tubing (R-3603) and Miniature Barbed Polypropylene Fit- ting (connectors, 6365-90) were obtained from Cole-Parmer Instrument Company, IL
USA. Needles (plastic) were obtained from Terumo Medical, Sydney Australia. Needles
(metal, New ‘Solila’ Surgical Hypodermic Needles) were obtained from The Amalga- mated Dental Company, Sydney Australia. 0.4μm PET membrane (PIHT30R48) was obtained from Millipore, Sydney Australia. Pluronic F-127 (P2443) was obtained from
Sigma-Aldrich, Sydney Australia. Vi-CELL XR Cell Viability Analyser was obtained from Beckman Coulter, CA USA. NIH 3T3 (eGFP) and MC 3T3 cells were kindly provided by iNano, University of Aarhus, Aarhus, Denmark.
All other chemicals, reagents and tissue culture products were prepared and supplied as described in Appendix A.
7.2.2 Device preparation for cellular use
Flow circuitry and connectors were assembled and flushed with 80% ethanol. All com- ponents were connected to the actual cell chamber and the entire assembled system was
filled with 80% ethanol for sterilisation. After 1 hour, the entire system was irrigated with 0.9% sodium chloride solution. The system was then flushed with IMDM sup- plemented with 10% FBS and the outlet connected to a sterilsed (autoclaved) waste collection bottle. The waste bottle was connected to a pressure head to ensure de- gassing of bubbles from inside the chamber and outlet coils. The entire set up was transfered to a humidified atmosphere of 95% air with 5% CO2 at 37℃. Fluid flow was
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 190 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials set to 4μL/min into the cell chamber from both sides and the chamber left to degas overnight.
7.2.3 Cell attachment to PET membrane
The PET membrane was cut out of the cell well insert housing and cut in half. One half was placed cell side up (the side that was originally facing up in the insert) and the other half placed cell side down into 2 35mm dishes. To ensure the option of cell attachment onto the membrane, a concentrated NIH 3T3 (eGFP) cell solution totaling
2x105 cells in 50μL was directly pipetted onto each of the membranes. Cultures were placed in a humidified atmosphere of 95% air with 5% CO2 at 37℃ for 1.5 hours. After this time, 3mL of IMDM supplemented with 10% FBS was added and the membranes incubated for 24 hours.
7.2.4 Hoechst 33342 staining
2mLofa1x105 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with
10% FBS was seeded into the wells of a 6 well plate, both directly on the tissue culture plastic and into PET membrane inserts. An additional 5mL of media was added to the wells containing the PET membrane inserts to ensure that the bottom surface of the membrane was exposed to media and that the cell suspension inside the insert did not pass through the membrane, leaving the cell surface dry. The well plates were placed in a humidified atmosphere of 95% air with 5% CO2 at 37℃ for 24 hours. After 24 hours, media was removed from all wells and 2mL of a 10μg/mL Hoechst 33342 solution was added and incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃ for 1 hour. The stain solution was then removed, all wells washed twice with DPBS and wells immediately viewed using fluorescent microscopy.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 191 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials
7.2.5 Pluronic coating of PET membrane
2mL of 0.2% w/w pluronic, a surfactant used to prevent protein adsorption and foaming in fermentation processes, in PBS was used to coat the surface of a PET membrane insert and the well of a 6 well plate and left for 1 hour at room temperature. The
Pluronic was removed and the membrane insert and well washed twice with cell culture medium. 2mL of 2 x 105 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with 10% FBS was seeded into the treated PET membrane insert and the TC well. An additional 5mL of media was added to the wells containing the PET membrane inserts to ensure that the bottom surface of the membrane was exposed to media and that the cell suspension inside the insert did not pass through the membrane, leaving the cell surface dry. The well plates were placed in a humidified atmosphere of 95% air with
5% CO2 at 37℃ for 24 hours. After 24 hours, media was removed from all wells and immediately viewed.
7.2.6 Direct contact cytotoxicity assay
2mL of a 50 x 103 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with 10%
FBS was seeded into 35mm dishes and incubated in a humidified atmosphere of 95% air with 5% CO2 at 37℃ for 24 hours. After 24 hours, medium was aspirated from the dishes and replaced with 0.8mL fresh medium. Test samples were then carefully placed directly on top of the cell layer and incubated for a further 24 hours. All samples were evaluated in triplicate. Prior to removal of the test material from the monolayer surface the boundary positions of the samples were marked out on the bottom of the plate.
For each test dish, the medium was aspirated from the dish and sufficient 0.4% Trypan
Blue in PBS to cover the dish was added for a period of approximately 1 minute. The stain was poured off and the dish gently irrigated with 2 washes of DPBS. Cells were examined using both bright field and phase contrast microscopy and graded from 0 to
4 for cytotoxic response to the sample material.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 192 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials
7.2.7 Cell growth inhibition assay
2mL of a 25 x 103 NIH 3T3 (eGFP) cells/mL or MC 3T3 cells/mL solution in IMDM supplemented with 10% FBS was seeded into 35mm dishes and incubated in a humidi-
fied atmosphere of 95% air with 5% CO2 at 37℃ for 24 hours. Extracts were prepared in one of two ways. When the test solution was a liquid, it was diluted 1:3 with IMDM supplemented with 10% FBS. Solid materials were placed in extraction vials with a so- lution of sodium chloride (NaCl), 0.9% w/v and extracted at 121℃ for 60 minutes. A ratio of 15cm2 surface area or 1g of material in 2.5mL NaCl was used. Once extracted, the solution was diluted 1:3 with IMDM supplemented with 10% FBS. The cell plate medium was aspirated and replaced with 1mL of the prepared test or control solutions and incubated for 48 hours. After 48 hours, the plates were inspected for abnormalities and/or microbial infection. The medium was then carefully poured off the plates and the plates washed with DPBS. 2mL solution of 1:1 trypsin enzyme:DPBS was added to the plates for 3 to 4 mintues to detach the cells from the plate surface. The cells were agitated until detached from the surface of the plate. 1mL of cell suspension was then counted using the Vi-CELL XR.
7.2.8 Cytotoxicity of assembled system
Cytotoxicity of the entire assembled system was assessed using the following method.
2mL of a 25 x 103 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with
10% FBS was seeded into 35mm dishes and incubated in a humidified atmosphere of
95% air with 5% CO2 at 37℃ for 24 hours.
Extracts were prepared using the follow method: Counter flow diffusion chamber tubing and connectors were assembled and flushed with 80% ethanol. All components were connected to the cell chamber and the entire assembled system was filled with 80% ethanol for sterilisation. After 1 hour, the system was irrigated with 0.9% sodium chlo- ride solution. Devices that required coating were then filled with the coating solution
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 193 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials
(pluronic or BSA) and left to incubate at room temperature for 1 hour. Assembled devices where then flushed with 0.9% sodium chloride solution. Once all the coating solution had been removed, devices to be tested with media were flushed with IMDM supplemented with 10% FBS. The outlet was then connected to a sterilised bottle for medium collection, and 10mL syringes with either 0.9% sodium chloride solution or
IMDM supplemented with 10% FBS were connected. The entire set up was transfered to either a 4℃ refridgerated environment or a humidified atmosphere of 95% air with
5% CO2 at 37℃. Flow rate was set at 4μL/min and the system run overnight until enough medium for extraction was collected.
Once enough extract solutions had been obtained, the cell plate medium was aspirated and replaced with 1mL of either the extracted media (undiluted) or 0.9% sodium chlo- ride solution diluted 1:3 with IMDM supplemented with 10% FBS and incubated for
48 hours. After 48 hours, the plates were inspected for abnormalities and/or microbial infection. The medium was then carefully poured off the plates and the plates washed with DPBS. 2mL solution of 1:1 trypsin enzyme:DPBS was added to the plates for 3 to 4 mintues to detach the cells from the plate surface. The cells were agitated until detached from the surface of the plate. 1mL of cell suspension was then counted using the Vi-CELL XR.
7.2.9 Cytotoxicity of tubing
2mL of a 25 x 103 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with
10% FBS was seeded into 35mm dishes and incubated in a humidified atmosphere of
95% air with 5% CO2 at 37℃ for 24 hours.
Extracts were prepared using the follow method: 1m of tubing was flushed with 80% ethanol for sterilisation. After 1 hour, the tubing was irrigated with 0.9% sodium chloride solution. Tubing that required coating were then filled with the coating solution
(pluronic or BSA) and left to incubate at room temperature for 1 hour and then flushed with 0.9% sodium chloride solution. Once all the coating solution had been removed,
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 194 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials tubing to be tested with media was flushed with IMDM supplemented with 10% FBS.
The tubing was then connected to a sterilised bottle for medium collection, and 10mL syringes with either 0.9% sodium chloride solution or IMDM supplemented with 10%
FBS were connected. If the lumen of the tubing to be tested was too small to fit over the bottle and syringe attaching connectors, a small bridging portion of silicone tubing was used. The entire set up was transfered to either a 4℃ refridgerated environment or a humidified atmosphere of 95% air with 5% CO2 at 37℃. Flow rate was set at 4μL/min and the system run overnight until enough medium for extraction was collected.
Once enough extract solutions had been obtained, the cell plate medium was aspirated and replaced with 1mL of either the extracted media (undiluted) or 0.9% sodium chlo- ride solution diluted 1:3 with IMDM supplemented with 10% FBS and incubated for
48 hours. After 48 hours, the plates were inspected for abnormalities and/or microbial infection. The medium was then carefully poured off the plates and the plates washed with DPBS. 2mL solution of 1:1 trypsin enzyme:DPBS was added to the plates for 3 to 4 mintues to detach the cells from the plate surface. The cells were agitated until detached from the surface of the plate. 1mL of cell suspension was then counted using the Vi-CELL XR.
7.2.10 Cytotoxicity of syringes
2mL of a 25 x 103 NIH 3T3 (eGFP) cells/mL solution in IMDM supplemented with
10% FBS was seeded into 35mm dishes and incubated in a humidified atmosphere of
95% air with 5% CO2 at 37℃ for 24 hours.
Extracts were prepared using the follow method: Syringes to be coated were filled with the coating solution and left for 1 hour at room temperature. After coating was complete, the coating solution was expelled and 10mL of IMDM supplemented with
10% FBS was drawn up. Syringes were generally not flushed prior to test media being drawn up, except in the case of the pluronic coated syringe, which was flushed with
100mL of 0.9% sodium chloride solution prior to the test media being drawn up. In
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 195 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.2. Methods and Materials the case of uncoated syringes, the test solution was simply drawn up. Syringes were capped with sterile needles and transfered to a 4℃ refridgerated environment for 24 hours.
After 24 hours, the cell plate medium was aspirated and replaced with 1mL of either the test media (undiluted) or in the case of the FBS supplemented PBS or BSA the test solution was diluted 1:3 with IMDM supplemented with 10% FBS and incubated for
48 hours. After 48 hours, the plates were inspected for abnormalities and/or microbial infection. The medium was then carefully poured off the plates and the plates washed with DPBS. 2mL solution of 1:1 trypsin enzyme:DPBS was added to the plates for 3 to 4 mintues to detach the cells from the plate surface. The cells were agitated until detached from the surface of the plate. 1mL of cell suspension was then counted using the Vi-CELL XR.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 196 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
7.3 Results
7.3.1 Cell attachment to PET membrane
A difference in cell attachment properties of the two sides of the PET membrane could drastically affect the adherence of cells inside the counter flow diffusion chamber. There- fore the cell attachment potential of the PET membrane was investigated by seeding cells onto both sides of the membrane as per section 7.2.3 and the result is presented in Figure 7.1.
(a) Cell side up (b) Cell side down
Figure 7.1: NIH 3T3 (eGFP) attachment after 24 hours to the PET membrane, both the side originally facing up in the insert and the underneath. Representative images captured using fluorescence microscopy, with a 10x objective.
Figure 7.1 illustrates that there is no noticeable difference in cell attachment to the side originally facing up in the insert and the underneath, allowing the membrane to be used either way in the device.
7.3.2 Hoechst 33342 staining
Cells attached to the PET membrane were stained with Hoechst 33342 as per section
7.2.4. Figure 7.2 illustrates that Hoechst 33342 stains the nuclei well with no non- specific binding of the dye to the membrane, providing an effective method to count nuclei in-situ. The blurriness of the image is a result of the insert being inverted and
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 197 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results cells being captured through the membrane. Due to the geometry of the insert, the cell layer was not within the focal length of the lens in the upright position.
(a) PET membrane (b) TCP
(c) PET membrane (d) TCP
Figure 7.2: NIH 3T3 (eGFP) cells attached to the PET membrane and TCP stained with 10μg/mL Hoechst 33342. Representative images captured using fluorescence mi- croscopy, with a 10x objective. Both a and b are taken under conditions capturing Hoechst 33342 and green fluorescence. Images c and d are taken under conditions capturing Hoechst 33342.
7.3.3 Inoculation density
7.3.3.1 High density inoculation for pellet formation
The counter flow diffusion chamber was connected to flow circuitry using silicone tubing and metal needles as per section 7.2.2, with the drainage coils on top, and the cell chamber below. Fluid flow was set to 4μL/min into the cell chamber from both sides and the chamber left to degas overnight. Upon inoculation, it was anticipated that the
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 198 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results cells would deposit onto the PDMS surface and would be concentrated at the stasis zone interface. Unexpectedly, cells were deposited onto the roof of the cell chamber and accumulated over the flow channels (drainage coils). This dramatic effect is illustrated in Figure 7.3. It can be seen that the cells deteriorated rapidly under these culture conditions.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 199 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
(a) Representative image immediately after inoculation
(b) Representative image 24 hours after inoculation
(c) Representative image 48 hours after inoculation
Figure 7.3: 6x106 NIH 3T3 (eGFP) cells in a volume of 0.5mL were injected into the counter flow diffusion chamber under a flow of 100μ L . After innoculation at room tem- perature, the system was transferred into a humidified atmosphere of 95% air with 5% CO2 at 37℃ and a flow of 4μL/min into the cell chamber from both sides. Images were captured using fluorescence microscopy, with a 10x objective. Images are representative and not taken at the same position in the chamber. The physical set up of the system is such that the drainage coils are on top, and the cell chamber is below.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 200 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
7.3.3.2 Low density inoculation
The cell inoculation procedure was altered in an attempt to prevent the alignment of cells along the drainage coil. Several changes where made to the flow circuitry to achieve this. Firstly, flow to the chamber was turned off. Secondly, an additional
fluid exit was made available at the other side of the chamber to enable flow to proceed across the chamber and out, and not be forced through the membrane into the drainage coils. Thirdly, the drainage flow circuitry was closed, forcing excess inoculation flow out the newly created chamber exit, rather than through the membrane into the drainage system.
Cell attachment was found to be concentration dependent. Cells did not attach and spread at low density. Figure 7.4 demonstrates that attachment did not occur at the cell inoculation density of 88.8 x 103 cells/mL. Typically a confluent 75cm2 tissue culture
flask will hold 4 x 106 cells. For confluence in 4 days, generally 1/20th of the cells are transfered to a new flask, i.e. 200 x 103 cells. The internal cell growing surface area of the cell chamber was calculated to be 0.28cm2 (from a width of 8000μm and length of
35000μm). Seeding the chamber at the same density for confluence in 4 days, entails seeding 746 cells. The volume of the cell chamber was calculated to be 8.4μL (from a depth of 300μm), giving a cell inoculation concentration 88.8 x 103 cells/mL. Figure
7.5 demonstrates that when the cell inoculation density is increased to that which is scaled to a confluent T75 flask, there is cell attachment occurring within 2 hours. Once again cells do not remain viable after 24 hours under a fluid flow of 4μL/min. Thus it appears that cell attachment or viability is concentration dependent. Figure 7.6 demonstrates that a detrimental effect from flow is not the cause of loss of viability, as even when flow is turned off, cells do not remain viable after a culture period of 24 hours. Table 7.2 summaries culture conditions tested so far, with effects noted from cell seeding density and inoculation flow conditions. The inability to maintain cell viability with the current set up indicates that there is a biocompatibility or material surface interaction issue with a particular component or the system as a whole.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 201 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
(a) Representative image immediately after inoculation
(b) Representative image 2.5 hours after inoculation
(c) Representative image 24 hours after inoculation
Figure 7.4: 88.8 x 103 NIH 3T3 (eGFP) cells/mL was injected into the counter flow diffusion chamber without any flow. Inoculation fluid was allowed to flow through the chamber and out an exit port, rather than pushed through the membrane into the drainage coils. After inoculation at room temperature, the system was transferred into a humidified atmosphere of 95% air with 5% CO2 at 37℃ and a flow of 4μL/min into the cell chamber from both sides. Images were captured using fluorescence microscopy, with a 10x objective. Images are representative and not taken at the same position in the chamber. The physical set up of the system is such that the drainage coils are on top, and the cell chamber is below.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 202 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
(a) Representative image immediately after inoculation
(b) Representative image 2 hours after inoculation
(c) Representative image 24 hours after inoculation
Figure 7.5: 1.7 x 106 NIH 3T3 (eGFP) cells/mL was injected into the counter flow diffusion chamber without any flow. Inoculation fluid was allowed to flow through the chamber and out an exit port, rather than pushed through the membrane into the drainage coils. After inoculation at room temperature, the system was transferred into a humidified atmosphere of 95% air with 5% CO2 at 37℃ and a flow of 4μL/min into the cell chamber from both sides. Images were captured using fluorescence microscopy, with a 10x objective. Images are representative and not taken at the same position in the chamber. The physical set up of the system is such that the drainage coils are on top, and the cell chamber is below.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 203 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
(a) Representative image immediately after inoculation
(b) Representative image 2 hours after inoculation
(c) Representative image 24 hours after inoculation
Figure 7.6: 1.7 x 106 NIH 3T3 (eGFP) cells/mL was injected into the counter flow diffusion chamber without any flow. Inoculation fluid was allowed to flow through the chamber and out an exit port, rather than pushed through the membrane into the drainage coils. After inoculation at room temperature, the system was transferred into a humidified atmosphere of 95% air with 5% CO2 at 37℃ without any flow. Images were captured using fluorescence microscopy, with a 10x objective. Images are representative and not taken at the same position in the chamber. The physical set up of the system is such that the drainage coils are on top, and the cell chamber is below.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 204 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results 24hrs Not viable Not viable Not viable Not viable - Viable Viable Viable 2-2.5hrs Cell Viability Viable Viable Viable Not viable Inoculation On On On Off Overnight On Off Off Off Flow Condition Inoculation Summary of conditions Across Across Across chamber chamber chamber Through membrane Drainage Configuration Table 7.2: Low High Density Seeding Confluence Confluence Silicone tubing in media delivery circuit All silicone tubing All silicone tubing All silicone tubing Components Metal Needles Tygon tubing at exit Metal Needles Metal Needles Plastic Needles
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 205 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
7.3.4 Biocompatibility and material surface interaction assessment
Initial experiments with the counter flow diffusion chamber highlighted the need to do a thorough analysis of biocompatibility.
All materials used in the counter flow diffusion chamber and flow circuitry were common laboratory materials often used in the manufacture of bioreactor devices with no obvious biocompatibility or known surface interaction concerns. Silicone tubing was chosen to permit gas exchange within the incubator, allowing the culture medium to reach the required pH, and be sufficiently gassed prior to delivery to the cells. Metal needles were chosen for ease of sterilisation, the chamber and flow circuitry could be assembled on the bench and the entire set up had the potential to be autoclaved. Replacement of metal needles with disposable plastic headed ones did not alter cell viability.
A variety of biocompatibility and material surface interaction testing was conducted with the following aims:
• to determine if any components used in the system are cytotoxic;
• to establish if there is any material surface interaction leading to a media depletion
effect resulting from the large surface area to volume ratio, and if so, can this
adsorption issue be remediated through the use of various surface coatings such
as pluronic or BSA; and
• to determine if media degradation was occurring independently of direct cytotox-
icity or material surface interactions.
7.3.4.1 Direct contact cytotox assay
A direct contact cytotoxicity assay was conducted on all materials used in the assembly of the counter flow diffusion chamber and flow circuitry as per section 7.2.6. A sample is considered to have cytotoxic effects if it produced any changes in cell morphology or cell density. The incorporation of Trypan Blue stain into the cytoplasm of cells is a
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 206 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results clear indication of disrupted cell plasma membranes and is the prime indicator of cell death in this assay.
Initially samples were not washed, rinsed or treated in any way prior to use in the direct contact assay, in order to exclude possible effects of sterilisation on material cytotoxicity. The assay was repeated with all samples rinsed in an 80% v/v ethanol solution and then in a 0.9% NaCl solution. The results presented in Table 7.3 appear to indicate that all materials used displayed a degree of toxicity, with the NIH 3T3 (eGFP) cells being slightly more sensitive than the MC 3T3 cells. There was little difference in rinsed and unwashed samples. Surprisingly, the silastic tubing used as a negative control did not cause any cell death when tested with MC 3T3 cells, but all other tubing did, even that made of silastic. PDMS, known to be reasonably biocompatible, elicited a far higher than expected degree of reactivity, indicating that perhaps the contact assay did not reflect the cytotoxicity of the perfusion system.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 207 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
Table 7.3: Direct contact cytotoxicity assay
Reactivity Replicate Sample Description MC 3T3 NIH 3T3 (eGFP) 1 Rinsed PDMS 2 2 Rinsed PDMS 2 3 Rinsed PDMS 3 1 Unwashed PDMS 2 4 2 Unwashed PDMS 3 4 3 Unwashed PDMS 2 4 1 Rinsed Thin Silicone Tubing 4 2 Rinsed Thin Silicone Tubing 4 3 Rinsed Thin Silicone Tubing 4 1 Unwashed Thin Silicone Tubing 4 4 2 Unwashed Thin Silicone Tubing 4 4 3 Unwashed Thin Silicone Tubing 4 4 1 Rinsed Thick Silicone Tubing 4 2 Rinsed Thick Silicone Tubing 4 3 Rinsed Thick Silicone Tubing 4 1 Unwashed Thick Silicone Tubing 4 4 2 Unwashed Thick Silicone Tubing 3 4 3 Unwashed Thick Silicone Tubing 4 4 1 Rinsed Tygon Tubing 2 2 Rinsed Tygon Tubing 2 3 Rinsed Tygon Tubing 3 1 Unwashed Tygon Tubing 3 4 2 Unwashed Tygon Tubing 3 3 3 Unwashed Tygon Tubing 3 4 1 Rinsed Connectors 2 2 Rinsed Connectors 2 3 Rinsed Connectors 2 1 Unwashed Connectors 4 2 2 Unwashed Connectors 4 2 3 Unwashed Connectors 4 2 1 Rinsed Metal Needles 4 2 Rinsed Metal Needles 4 3 Rinsed Metal Needles 3 1 Unwashed Metal Needles 4 4 2 Unwashed Metal Needles 4 4 3 Unwashed Metal Needles 4 3 1 Rinsed Plastic Needles 4 2 Rinsed Plastic Needles 3 3 Rinsed Plastic Needles 4 1 Unwashed Plastic Needles 1 plastic, 2 metal 4 2 Unwashed Plastic Needles 1 plastic, 2 metal 4 3 Unwashed Plastic Needles 1 plastic, 2 metal 4
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 208 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
Reactivity Replicate Sample Description MC 3T3 NIH 3T3 (eGFP) 1 Rinsed control Latex 4 2 Rinsed control Latex 4 3 Rinsed control Latex 4 1 Unwashed control Latex 4 4 2 Unwashed control Latex 4 4 3 Unwashed control Latex 4 4 1 Rinsed control Silastic Tubing 2 2 Rinsed control Silastic Tubing 2 3 Rinsed control Silastic Tubing 4 1 Unwashed control Silastic Tubing 0 4 2 Unwashed control Silastic Tubing 0 4 3 Unwashed control Silastic Tubing 0 3 1 Untreated null 0 0 2 Untreated null 0 0 3 Untreated null 0 0
Table 7.4: Reactivity Grades
Grade Reactivity Description of Reactivity Zone 0 None No detectable zone under or around sample 1 Slight Zone limited to under the sample 2 Mild Zone extends <0.5cm 3 Moderate Zone between 0.5 and 1.0 cm 4 Severe Zone extends >1.0cm beyond sample
7.3.4.2 Cell growth inhibition assay
A cell growth inhibition assay was conducted on all materials used in the counter flow diffusion chamber and flow circuitry as per section 7.2.7 using both MC 3T3 and NIH
3T3 (eGFP) cells, with the results presented in Figure 7.7. A material which causes cell growth inhibition above 30% is considered toxic, and it can be seen from Figure 7.7 that cell growth inhibition occurred from both the metal and plastic needles. Results for both cell lines are similar.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 209 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
170
160 MC3T3
150 NIH3T3
140
130
120
110
100
90
80
70 % Inhibition
60
50
40
30
20
10
0
-10 Null Latex Empty PDMS Silastic Connectors Ethanol 4% Ethanol 5% Metal Needle Ethanol 7.5% Tygon Tubing Plastic Needle Extraction Control 1 Extraction Control 2 Thin Silicone Tubing Thick Silicone Tubing Controls
Figure 7.7: Cell growth inhibition assay using NIH 3T3 (eGFP) and MC 3T3 cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
The metal needles and disposable plastic needles were the only components that showed cell growth inhibition greater than 30% and as a result, both were investigated further.
7.3.4.3 Needle Investigation
A whole metal needle was placed on a layer of NIH 3T3 (eGFP) cells (4 x 105 cells seeded into a 100mm dish and left overnight). The needle was sterilised in an 80% ethanol solution for 1 hour and rinsed in DPBS before being placed on the cells. After
24 hours a chemical reaction had occured at the junction of the needle shaft and the head as seen in Figure 7.8. The metal needles were deemed cytotoxic and were not used again in the flow circuitry.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 210 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
Figure 7.8: Metal needle placed on a layer of cells in media for 24 hours, 10x.
(a) Representative image near needle head (b) Representative image near reactive junction
(c) Representative image far from needle in same (d) Representative image control dish dish
Figure 7.9: Effect of metal needle on NIH 3T3 (eGFP) cell layer after 24 hours of incuba- tion. Representative images captured using fluorescence microscopy, with a 10x objective.
A cell growth inhibition assay was conducted on the Terumo disposable plastic needles.
In order to determine if any particular part of the needle was predominantly responsible for the cell growth inhibition seen in Figure 7.7, needles were cut into the three distinct
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 211 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results components: the metal shaft, the plastic head, and the shaft and plastic junction, to determine any effect of the glue used. Once 1 gram of each part had been weighed, samples were tested as per section 7.2.7 using NIH 3T3 (eGFP) cells, with the results presented in Figure 7.10.
150
130
110
90
70 ibi i
50
30
10
-10 Null Latex Empty Silastic EtoH 4% EtoH 5% (plastic) EtoH 7.5% glue) Needle Head Needle Middle (metal, plastic and Extraction Control 1 Extraction Control 2 Needle Shaft (metal)
Controls
Figure 7.10: Cell growth inhibition assay of disposable plastic needles using NIH 3T3 (eGFP) cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
The needle head and junction both showed cell growth inhibition greater than 30%.
The metal component was below 30%. This is a particulary useful finding as it allows the continued use of the metal shaft for connection of the flow circuitry to the chamber.
As needles were also being used to connect tubing to the media delivering syringes, an alternate method will need to be sought for this purpose.
7.3.4.4 Pluronic coating of PET membrane
Pluronic F127 is a nonionic copolymer surfactant that can be used to passivate surfaces to prevent adsorption of cell culture medium proteins. Whilst pluronic is relatively non toxic, cellular response to coated PET membranes was tested prior to use in the counter
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 212 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
flow diffusion chamber and flow circuitry. PET membranes and TCP were coated with pluronic as per section 7.2.5 and Figure 7.11 illustrates that coating with pluronic does not affect cell attachment to the PET membrane. This indicates, that if deemed necessary, the chamber surfaces may be passivated with pluronic.
(a) Pluronic coated PET mem- (b) Pluronic coated TCP 3.5 (c) TCP 3.5 hours brane 3.5 hours hours
(d) Pluronic coated PET mem- (e) Pluronic coated TCP 22 (f) TCP 22 hours brane 22 hours hours
(g) Pluronic coated PET mem- (h) Pluronic coated TCP 52.5 (i) TCP 52.5 hours brane 52.5 hours hours
Figure 7.11: NIH 3T3 (eGFP) cell growth on 0.2% w/w Pluronic in PBS coated PET mem- brane and TCP. Representative images captured using fluorescence microscopy, with a 10x objective.
7.3.4.5 Counter flow diffusion chamber with silicone tubing connected sys-
tem
The results from Figure 7.7 did not show cell growth inhibition greater than 30% for the silicone tubing or PDMS. As a result, the chamber was connected to flow circuitry
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 213 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results comprising of silicone tubing, with plastic needles connecting to the syringes. A cell growth inhibition assay was conducted using both medium and saline pumped through the system at 4℃ as per section 7.2.8 using NIH 3T3 (eGFP) cells, in order to determine if either a media depletion effect was occuring, or if there was toxicity present in the system. Figure 7.12 illustrates inhibition of growth for both saline and media extracts, it could not be remediated through coating with pluronic or BSA. The cell grow inhibition present with the saline indicates that toxicity was present in the system and altering the length of the silicone exit tubing from 150cm to 30cm had little impact.
170 Long Exit Tubing, Media Short Exit Tubing, Media 150 Short Exit Tubing, Saline
130
110
90 ibi i 70
50
30
10
-10 Empty Uncoated Pluronic Coated Albumin Coated
Figure 7.12: Cell growth inhibition assay of medium and saline extracted through the counter flow diffusion chamber connected to flow circuitry comprising of silicone tubing, with plastic needles connecting to the syringes, using NIH 3T3 (eGFP) cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
7.3.4.6 Syringe Extraction 4 degrees
In order to determine if the disposable syringes were the component responsible for the toxicity present in the system, a comprehensive a cell growth inhibition assay exploring
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 214 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results both toxicity and surface protein adsorption was conducted as per section 7.2.10 with various coatings using NIH 3T3 (eGFP) cells, with the results presented in Figure 7.13.
50 ncoated 5 Coated
0
35
30
5
ibi i 0
15
10
5
0
-5 Null Saline Empty l umin luronic l umin 2mg mL Media Media Media 10% S 20% S 30% S 20mg mL S 10% S S 20% S S 30% S S
Figure 7.13: Cell growth inhibition assay of various plastic syringe coatings and protein solu- tions at 4℃, using NIH 3T3 (eGFP) cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
No cell growth inhibition due to syringe toxicity was observed in Figure 7.13. There were modest inhibition levels for coated PBS supplemented with 30% FBS and Albumin.
7.3.4.7 Tubing Extraction 4 degrees
As the syringes were not the cause of the toxicity, the tubing was further investigated.
A cell growth inhibition assay to test both toxicity and adsorption was conducted using the medium that was pumped through silicone, PTFE and polyethylene tubing as per section 7.2.9 using NIH 3T3 (eGFP) cells, with the results presented in Figure 7.14. No needles were used in this set up. A small piece of silicone tubing was used to connect the PTFE and polyethylene tubing to the syringe and collection bottles.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 215 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
150 Media Saline 130
110
90
70 ibi i
50
30
10
-10 ull Empty Tubing Syringe Polyethylene Polyethylene PT E Tubing PT E Syringe Silicone Tubing Silicone Syringe
Figure 7.14: Cell growth inhibition assay of medium and saline extracted through silicone, PTFE and polyethylene tubing at 4℃, using NIH 3T3 (eGFP) cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
Figure 7.14 illustrates clearly that the silicone tubing has an elevated cell growth inhi- bition with the saline, revealing that the tubing is indeed toxic. This toxicity was not observed with the polyethylene or PTFE tubing. A media depletion effect was however observed in all three types of tubing.
7.3.4.8 Pluronic and Albumin coated PTFE tubing, Uncoated Tygon tub-
ing
In order to determine if the media depletion effect observed occurring with the PTFE tubing could be remediated, a cell growth inhibition assay was conducted using the medium extracted through pluronic coated and BSA coated PTFE tubing as per section
7.2.9 using NIH 3T3 (eGFP) cells, with the results presented in Figure 7.15. Uncoated
Tygon tubing was also tested. Due to time constraints, only one type of tubing could be tested further, and PTFE tubing was chosen over polyethylene for its biocompatibility and wider use in in vitro applications.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 216 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
110
90
70
50 ibi i
30
10
-10 ull Empty Media Saline Tygon Syringe Tygon Syringe Pluronic Coated Albumin Coated Pluronic Coated Albumin Coated PT E Tubing Media Tygon Tubing Media Tygon Tubing Saline PT E Tubing Media PT E Syringe Media PT E Syringe Media
Figure 7.15: Cell growth inhibition assay of medium extracted through albumin and pluronic coated PTFE, and saline extracted through uncoated Tygon Tubing, at 4℃, using NIH 3T3 (eGFP) cells. Dotted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
It can be seen quite clearly that the media depletion effects of the PTFE tubing could not be remediated through pluronic or BSA coating. Surprisingly, very low levels of cell growth inhibition were observed for the Tygon tubing, indicating that there was no toxicity present, or any surface protein adsorption effects.
7.3.4.9 Counter flow diffusion chamber connected system using Tygon tub-
ing
The counter flow diffusion chamber was connected to flow circuitry comprising of only
Tygon tubing. The metal shafts of plastic needles were used to connect the chamber to the flow circuitry. A cell growth inhibition assay was conducted using medium as per section 7.2.8 at both 4 and 37 ℃ using NIH 3T3 (eGFP) cells. From Figure 7.16 it can be seen that at 4 ℃ cell growth inhibition is well below 30%. However at 37 ℃
, the elevated cell growth inhibition present is most like a result of media degradation rather than surface protein adsorption effects.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 217 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
70 C 37 C 0
50
0
30 ibi i
0
10
0
-10 CC Empty Syringe
Figure 7.16: Cell growth inhibition assay of medium extracted through the CFDC using Ty- gon tubing flow circuitry in a 4℃ refridgerated environment and a humidified atmosphere of 95% air with 5% CO2 at 37℃, using NIH 3T3 (eGFP) cells. Dot- ted line at 30% inhibition. Bars denote mean of triplicate test plates ± SEM.
7.3.5 Cell growth inside counter flow diffusion chamber using Tygon tubing
NIH 3T3 (eGFP) cells were loaded into a counter flow diffusion chamber assembled with only Tygon tubing and connectors as per section 7.2.2. The physical set up of the system was such that the cell chamber was on top, and the drainage coils were below. The setup was run with bicarbonate free IMDM supplemented with 10% FBS.
The resulting images are shown in Figure 7.17. The appearance of Figure 7.17a is a result of compliance in the larger diameter Tygon tubing. Even though the exits were tied off quite tightly, the increased volume in the tubing had sufficient compliance to allow inoculation fluid to flow through the membrane and out the drainage coils, resulting in the appearance of the stripped cellular pattern first observed in Figure 7.3, with the cells again accumulating on the membrane over the drainage coils. 3 hours post inoculation, it can be seen that the cells are attaching to the surface of the PET membrane, Figure 7.17b and spreading into the free space, indicating that there was
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 218 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results no damage due to the inoculation process. At this point, flow was turned on. After
24 hours, the cells retain the morphology of viable cells, which is maintained 72 hours post inoculation.
(a) Representative image immediately after inoculation
(b) Representative image 3 hours after inoculation
(c) Representative image 24 hours after inoculation
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 219 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.3. Results
(d) Representative image 48 hours after inoculation
(e) Representative image 72 hours after inoculation
Figure 7.17: 1.7 x 106 NIH3T3(eGFP)cells/mLwereinjectedintothecounterflowdiffu- sion chamber without any flow. Inoculation fluid was allowed to flow through the chamber and out an exit port, rather than pushed through the membrane into the drainage coils. After inoculation at room temperature, the system was transferred into a humidified atmosphere of 95% air with 5% CO2 at 37℃ and a flow of 4μL/min into the cell chamber from both sides. Images were captured using fluorescence microscopy, with a 10x objective. Images are representative and not taken at the same position in the chamber. The physical set up of the system is such that the cell chamber is on top, and the drainage coils are below.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 220 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.4. Discussion
7.4 Discussion
Currently, there are no reports in the literature of micro-devices implementing the use of a membrane to create a converging flow field and concentration gradient in the way which has been described. The design of the microfluidic device presented offers a number of advantages over existing micro-devices. The implementation of a microporous membrane has the following practical advantages: a) cells are concentrated within the device; b) a steady concentration gradient is established within a zone of low shear stress; and c) the device could potentially be used for longer term culture, for the development of composite tissue structures from a multipotent progenitor cell source.
7.4.1 System Optimisation
As with any novel system, expected operation can can differ from actual performance.
For the counter flow diffusion chamber, device drainage geometry and inoculation den- sity were found to affect cell viability.
It was not possible to maintain the viability of a NIH 3T3 (eGFP) cell pellet. NIH 3T3
(eGFP) cells do not grow in pellet culture, which may explain why their viability was compromised at high density. Furthermore, excessive suction onto the membrane may have resulted in mechanical damage to this fastidious cell line. The geometry of this device may be altered to reduce the level of suction at the membrane. A reduction in the width of the drainage channels would result in a more even distribution of permeation over the membrane surface. This may prevent the banding pattern, with concentration of cells within the stasis zone.
7.4.2 Biocompatibilty
Little is known about surface adsorption effects, material cytotoxicity and media degra- dation processes at the microscale. Biocompatibility can become a central issue in
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 221 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.4. Discussion microscale culture because flow rates are so low, and trace impurities are not diluted compared to large volume culture. Additionally a high surface area to volume ratio can result in physical adsorption of proteins particularly when present at low concentrations
(e.g., growth factors). Additionally, in the absence of antioxidants or cells, media is rapidly degradation at 37℃ [116].
The direct cytotoxicty of metal needles was most likely from chromium plating; heavy metals are very cytotoxic, even at trace levels (Figure 7.9). A cell growth inhibition assay conducted on the various parts of the disposable plastic needles revealed that the metal shaft was below 30% inhibition, where as the plastic head and junction of plastic and metal were above (Figure 7.10). This finding allowed the use of the metal shafts as connectors, a means of attaching the flow tubing to the actual cell chamber.
Disposable syringes were found to be not toxic (Figure 7.13).
These results indicated that the source of toxicity in the system was either the tubing, or the actual cell chamber. Initially, the flow circuitry was assembled with silicone tubing. Silicone tubing is often used in cell culture systems for gas exchange, having permeability to CO2 and O2. Culture medium pH adjustment and oxygenation to nec- essary cellular levels occurs via the tubing before the medium reaches the cell chamber.
Silicone tubing is also used widely for its biocompatibility properties, and this particu- lar tubing is reported to have no peroxide by-products, being platinum cured [398,399].
It is however possible that platinum cured silicon tubing is directly cytotoxic, as was evidenced in this study.
Tubing was tested on its own, without being connected to the cell chamber. Polyethy- lene and PTFE tubing were also investigated along with the silicone tubing via culture medium and saline extraction studies. The silicone tubing when tested with saline, demonstrated high cell growth inhibition, indicating clearly that it was cytotoxic (Fig- ure 7.14). Loss of culture medium bioactivity at 4℃ was also found to be occurring in all types of tubing. This was indicated by the lower cell growth inhibition associated with the saline extraction, compared to the higher cell growth inhibition exhibited for cell culture medium extraction. This unexpected and highly surprising result indicates
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 222 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.4. Discussion that the reduced flow rates, and the large surface area to volume ratio, are having a marked effect at the microscale.
An attempt was made to passivate the surface of the PTFE tubing with both pluronic
F127 and BSA (Figure 7.15), with neither agent being successful. Gomez-Sjoberg et al. [235] suggested surface passivation with pluronic F127 as necessary in their PDMS based micro-device, and as a result this method was applied to this system. The concept of surface passivation was also extended to include BSA. One of the primary roles of albumin in culture media is to remove or prevent the production of damaging reactive oxidative species (ROS) [116]. Tygon tubing was also tested with culture medium and saline, and showed neither a loss of culture medium bioactivity at 4℃, or cytotoxicity. As a result, the counter flow diffusion chamber was connected to flow circuitry comprising only of tygon tubing. Medium was extracted at both 4 and 37℃
(Figure 7.16). The elevated cell growth inhibition seen at 37℃ may be a result of media degradation occurring in the tubing, or may also be related to increased protein adsorption at 37℃. This culture medium effect was not significant enough to inhibit the growth of NIH3T3 in the device (Figure 7.17).
The generation of reactive oxidative species by culture components and cells themselves limits culture productivity. ROS are produced or metabolised by the intracellular pathway responsible for regulating cellular redox potential. Alterations of cellular redox potential will affect signal transduction, DNA and RNA synthesis, protein synthesis, enzyme activation, and regulation of the cell cycle [117]. The biological effect of these transcriptional changes can be differentiation, senescence or apoptosis. Reactive oxygen intermediates derived from mature phagocytic cells have been shown to limit progenitor cell self-renewal [400]. The rate of ROS reaction is increased at higher temperatures
(4℃ versus 37℃) and by the surface area of culture media that is directly exposed to oxygen gas. There is also light-independent formation of ROS by auto-oxidising media components such as thiols, metal ions and lipids.
The studies conducted in this chapter highlight the need to carefully assess the contri- bution of direct material cytotoxicity, surface adsorption and media degradation effects
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 223 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.4. Discussion at the microscale. Standard tests do not always reveal cytotoxicity, context sensitive toxicity needs to be considered, especially when such large surface area to volume ratios are present in the slow flowing micro-devices.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 224 Chapter 7. Biocompatibility and perfusion culture using the counter flow diffusion chamber 7.5. Conclusion
7.5 Conclusion
In this chapter, it has been shown that NIH 3T3 (eGFP) cells can remain viable inside the cell chamber below a flow rate of 4μL/min for 72 hours. Biocompatibility is a major issue at the microscale. Future work should continue refinement of this bioreactor design to minimise the effects of direct and indirect cytotoxicity caused by media degradation effects.
Micro-devices hold great potential for many biological applications. The micro-device tested has the potential to establish ex-vivo gradients, allowing the study of chemotaxis and tissue morphogensis that naturally occur along an in vivo gradient. The device has the potential to allow investigation of basic biological phenomena, whilst also holding the potential to generate long-term tissue gradients in thick tissue constructs.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 225 Chapter 8
Summary and Recommedations
This thesis is concerned with studying various aspects of tissue engineering includ- ing the isolation of a homogeneous population of precursor cells from a heterogeneous population; the ability to expand anchorage dependent progenitor cells; and the devel- opment of bioreactor devices capable of supporting cell growth under an in vitro stable controllable gradient.
8.1 Cell identification and isolation
The aim of this study was to develop a robust methodology for the prospective isolation and detection of mesenchymal progenitor cells from multiple species. The established performance link between ALDH activity and Aldefluor substrate in human umbilical cord blood haemopoietic progenitor cells was used as a starting point for this study.
The methodology for assessing mesenchymal stem cell frequency using the CFU-F assay was successfully replicated for both hUCB and rat bone marrow. The frequency of
CFU-F in rat marrow was found to be higher than in hUCB. Whilst the methodology for detecting HSC in hUCB using Aldefluor substrate was successfully replicated, MSC were not co-purified with HSC in the low orthogonal light scattering region when labeled with Aldefluor substrate.
226 8.2. Hollow fibre expansion of Chapter 8. Summary and Recommendations anchorage dependent cells
Cells were successfully isolated from rat, murine, porcine and equine origins, and all showed an ALDH bright population when labeled with Aldefluor substrate. Rat whole bone marrow was the only sample to show a small cluster of cells in the ALDH bright, low orthogonal light scattering region.
Future recommendations for this work would include replication of the cord blood sorting study with bone marrow, either rat or human, which would have a much higher incidence of CFU-F. This would conclusively establish that CFU-F do not co-purify in the ALDH bright, low orthogonal light scattering subset. Also, the ALDH bright, high orthogonal light scattering subset requires investigation, to determine if CFU-F are present in this subset.
The toxicity of the substrate on mesenchymal progenitor cells has not been evaluated as part of this study. Bone marrow cells, once incubated with substrate can be plated out under mesenchymal optimised conditions with growth compared to non labeled bone marrow cells. The level at which the substrate becomes harmful to mesenchymal progenitor cells should also be determined.
8.2 Hollow fibre expansion of anchorage dependent cells
The aim of this chapter was to determine the feasibility of expanding anchorage de- pendent cells using scalable hollow fibre bioreactor technology which offers a number of advantages over conventional culture methods, originally optimised for fibroblast, not progenitor cell growth.
This chapter investigated a hollow fibre module, containing a single cellulose hollow fibre
[225] for the growth of anchorage dependent cells. It was determined that cellulose based surfaces required protein coating for cellular attachment to occur. A novel recombinant protein with a cellulose binding domain was found to effectively facilitate adhesion through ligand binding on cellulose based surfaces.
An in situ methodology was developed for viewing and counting anchorage dependent
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 227 8.3. Multi-channel bioreactor for engineering composite Chapter 8. Summary and Recommendations tissues from MSC cells inside hollow fibres, and the proliferation of rat tissue culture adherent bone mar- row, NIH 3T3 (eGFP) and MS-5 cells were investigated. Growth was found to be sub- optimal compared to conventional methods. The influence of extracapillary co-culture was investigated and appeared to show an increased fold expansion.
Whilst development of a fully formulated serum free media for MSC expansion was beyond the scope of this thesis, future recommendations for this work would include the investigation of developing such a cell culture medium with the novel recombinant protein presented. Incorporation of the CBDR, along with the inclusion of soluble components found in serum such as the lysophospholipid sphingosylphosphorylcholine, could offer an alternative to current MSC culture media requiring the addition of animal serum components. Fully formulated serum free media are important for the translation of MSC based therapies from a laboratory setting to clinically relevant practices.
Future recommendations for this work also include determining the scalability of hollow
fibre systems for MSC expansion. The device presented is limited to a single fibre, and is not indicative of a system which would be used for the expansion of progenitor cells in a clinical setting. To obtain clinically relevant cell numbers, devices comprising of multiple fibres would be required. Future studies, such as those conducted using lab- on-a-chip technology [235], are required to optimise the culture conditions. Cell seeding density, composition of culture medium, and feeding schedule should be determined to in order to obtain maximum fold expansion.
8.3 Multi-channel bioreactor for engineering composite
tissues from MSC
The aim of this study was to evaluate the design of multi-channeled single chambered bioreactor devices for the ability to produce stable controllable gradients inside a cham- ber suitable for cell maintenance and growth. The generation of in vitro gradients allows the study of fundamental biological phenomena that occur along a gradient whilst also making possible the manufacture of tissue grafts containing multiple tissue types.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 228 Chapter 8. Summary and Recommendations 8.4. Microfluidic Bioreactors
Two sets of single chambered device designs were constructed and presented, with neither design having the ability to generate precisely defined and controllable steady gradients inside the cell chamber.
The first set of devices presented employed a membrane to separate fluid flow from the cellular construct and were built using polycarbonate blocks. Devices were reliant upon both internal and external compression, and the presence of a scaffold inside the cell chamber, for function. These devices were affected by leakage. The second device design presented employed hollow fibres to separate fluid flow from the cellular construct. Manual manufacturing processes resulted in imprecise device geometry.
A more robust manufacturing method was deemed necessary for the production of such devices, and as a result the methods from the field of microfluidics were employed.
8.4 Microfluidic Bioreactors
The aims of this chapter were to study concentration gradient formation in micro- devices manufactured using the techniques of lithography from polydimethylsiloxane, and determine the effect of flow rate on gradient generation.
Stable concentration gradient profiles of biological cell size were experimentally gener- ated in two micro-devices with differing geometries.
The first device, previously described in literature [369], was comprised of a network of serpentine flow channels which allow three adjacent laminar input streams to flow to- gether until full diffusive mixing can take place in the single converged outlet, producing a steady concentration gradient perpendicular to the direction of fluid flow.
The second device, which used a novel design, had converging fluid streams with dif- fering morphogen concentrations entering the cell chamber. Cells were to be grown on a porous membrane which forms the floor of the chamber. A steady concentration gra- dient was established along the direction of fluid flow in the region of flow convergence.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 229 Chapter 8. Summary and Recommendations 8.4. Microfluidic Bioreactors
This configuration provided the advantage of generating a steep concentration gradient in the region of flow stasis. As a result, cell chemotaxis and tissue morphogenesis could be studied without the confounding effect of fluid shear stress.
The optically transparent nature of PDMS used to manufacture the devices allowed for real time imaging, and a method was developed to accurately image concentration gradients at micron resolution.
It was demonstrated that the concentration gradient profiles and shear stress developed inside the cell chambers of both devices could be controlled by varying the input flow rate. Theoretical models were developed, and were found to be useful in predicting device behaviour.
Future work should model diffusion and fluid dynamics in 3-D using a numerical solver.
The comprehensive modeling package will take into account device geometry, and gen- erate the flow field and concentration profile in the x, y and z directions.
Recommendations for future work for the second micro-device presented involve varying the input conditions and studying the effect on gradient generation. In this study, the opposing input flow rates were kept equal. The effect on concentration gradient generation of unequal flow rates should be determined. It may be possible to use unequal flow rates to alter and manipulate the gradient as desired.
Also, alterations to the pattern for the drain of the second micro-device may allow ma- nipulation of the cell pattern achieved inside the cell chamber. For example, changing the drain pattern from the current multiple coil configuration, to a single coil, may achieve a single crisp cell gradient in the middle of cell chamber. This would be useful for the formation of thick tissues.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 230 8.5. Biocompatibility and perfusion culture using the Chapter 8. Summary and Recommendations counter flow diffusion chamber
8.5 Biocompatibility and perfusion culture using the counter
flow diffusion chamber
The aim of this study were to asses the biocompatibility of the second micro-device presented in Chapter 6.
It was determined that biocompatibility was a major issue at the microscale. Compo- nents not exhibiting toxicity at the macroscale, such as silicone tubing, were found to exhibit toxicity at the microscale, and the importance of context sensitive biocompati- bility was highlighted.
NIH 3T3 (eGFP) cells were shown to remain viable inside the cell chamber below a
flow rate of 4μL/min for 72 hours.
The design of this micro-device lends itself to a number of tissue engineering applica- tions. Whilst the counter flow diffusion chamber does not require the presence of a cell scaffold for operation, the design does not preclude the use of a scaffold which could be injected into the cell chamber. The gradient generating capabilities also allow the study of cellular response to chemotactic gradients, whilst the optically transparent nature of PDMS allows real time imaging, permitting insight into tissue development and cellular progression.
The determination of cellular proliferation and viability monitoring inside the cell cham- ber can be easily carried out with in situ histological staining methods, such as with use of the dye Hoechst 33342, where it has already been established that non-specific binding does not occur onto the porous membrane. Hoechst 33342 can be added to the culture medium and provides a relatively simple method for counting nuclei. Live cell imaging may be used for more detailed mapping of cell viability, migration and differentiation.
The effect of perfusion rate on cell viability inside the cell chamber require further investigation, to determine if the nutrient supply and waste removal is sufficient. The
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 231 8.5. Biocompatibility and perfusion culture using the Chapter 8. Summary and Recommendations counter flow diffusion chamber relationship between required flow rate and cell density inside the cell chamber may change with cell proliferation and subsequent increase in cell numbers.
The geometry of the counter flow diffusion chamber can allow for the co-culture of particular cells types that occur naturally in vivo, such as HSC and MSC. It is possible to set up and mimic a 3 dimensional bone marrow like micro-environment using this device. The device can be used for the differentiation of multiple tissue types, and the establishment of a cellular gradient inside the chamber will render this device most useful in a number of applications. Not only can tissues with a naturally occurring graded zone, such as with the osteochondral junction be manufactured, but under such a chemotactic gradient, environments for the study of neurite outgrowth can also be set up.
The device can also be used to study early patterning that occurs in the embryo.
Nodal/BMP gradients that result in early embryonic structure [401] can be established in such a device using the opposing input flow streams.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 232 References
[1] F. Berthiaume and M. L. Yarmush. Tissue engineering. In J. D. Bronzino, editor, The Biomedical Engineering Handbook, volume 2 of The Electrical Engineering Handbook Series, pages 109–1 109–12. CRC Press, 2 edition, 2000.
[2] L. G. Griffith and G. Naughton. Tissue engineering–current challenges and ex- panding opportunities. Science, 295(5557):1009–14, 2002.
[3] Y. Martin and P. Vermette. Bioreactors for tissue mass culture: design, charac- terization, and recent advances. Biomaterials, 26(35):7481–503, 2005.
[4] S. V. McDiarmid. Current status of liver transplantation in children. Pediatr Clin North Am, 50(6):1335–74, 2003.
[5] R. Langer and J. P. Vacanti. Tissue engineering. Science, 260(5110):920–6, 1993.
[6] D. E. Ingber, V. C. Mow, D. Butler, L. Niklason, J. Huard, J. Mao, I. Yannas, D. Kaplan, and G. Vunjak-Novakovic. Tissue engineering and developmental biology: Going biomimetic. Tissue Eng, 12(12):3265–3283, 2006.
[7] P. Bianco and P. G. Robey. Stem cells in tissue engineering. Nature, 414(6859):118–21, 2001.
[8] F. P. Barry. Biology and clinical applications of mesenchymal stem cells. Birth Defects Res Part C Embryo Today, 69(3):250–6, 2003.
[9] P. J. Donovan and J. Gearhart. The end of the beginning for pluripotent stem cells. Nature, 414(6859):92–7, 2001.
[10] H. K. Vaananen. Mesenchymal stem cells. Ann Med, 37(7):469–79, 2005.
[11] L. da Silva Meirelles, P. C. Chagastelles, and N. B. Nardi. Mesenchymal stem cells reside in virtually all post-natal organs and tissues. J Cell Sci, 119(Pt 11):2204–13, 2006.
[12] K. Le Blanc, I. Rasmusson, B. Sundberg, C. Gotherstrom, M. Hassan, M. Uzunel, and O. Ringden. Treatment of severe acute graft-versus-host disease with third party haploidentical mesenchymal stem cells. Lancet, 363(9419):1439–41, 2004.
[13] A. Smith. A glossary for stem-cell biology. Nature, 441(29):1060, 2006.
[14] D. Clarke and J. Frisen. Differentiation potential of adult stem cells. Curr Opin Genet Dev, 11(5):575–80, 2001.
233 References References
[15] G. C. Kopen, D. J. Prockop, and D. G. Phinney. Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains. Proc Natl Acad Sci U S A, 96(19):10711–6, 1999.
[16] Y. Sato, H. Araki, J. Kato, K. Nakamura, Y. Kawano, M. Kobune, T. Sato, K. Miyanishi, T. Takayama, M. Takahashi, R. Takimoto, S. Iyama, T. Matsunaga, S. Ohtani, A. Matsuura, H. Hamada, and Y. Niitsu. Human mesenchymal stem cells xenografted directly to rat liver are differentiated into human hepatocytes without fusion. Blood, 106(2):756–63, 2005.
[17] K. O’Donoghue and N. M. Fisk. Fetal stem cells. Best Pract Res Clin Obstet Gynaecol, 18(6):853–75, 2004.
[18] C. Campagnoli, I. A. Roberts, S. Kumar, P. R. Bennett, I. Bellantuono, and N. M. Fisk. Identification of mesenchymal stem/progenitor cells in human first- trimester fetal blood, liver, and bone marrow. Blood, 98(8):2396–402, 2001.
[19] I. Rogers and R. F. Casper. Umbilical cord blood stem cells. Best Pract Res Clin Obstet Gynaecol, 18(6):893–908, 2004.
[20] K. Bieback, S. Kern, H. Kluter, and H. Eichler. Critical parameters for the isola- tion of mesenchymal stem cells from umbilical cord blood. Stem Cells, 22(4):625– 34, 2004.
[21] H. S. Wang, S. C. Hung, S. T. Peng, C. C. Huang, H. M. Wei, Y. J. Guo, Y. S. Fu, M. C. Lai, and C. C. Chen. Mesenchymal stem cells in the wharton’s jelly of the human umbilical cord. Stem Cells, 22(7):1330–7, 2004.
[22] D. Baksh, L. Song, and R. S. Tuan. Adult mesenchymal stem cells: characteri- zation, differentiation, and application in cell and gene therapy. J Cell Mol Med, 8(3):301–16, 2004.
[23] A. Bongso and E. H. Lee. Stem Cells: Their Definition Classification and Sources in Stem Cells From Bench to Bedside. World Scientific Publishing, Singapore, 2005.
[24] P. Bianco and P. Gehron Robey. Marrow stromal stem cells. J Clin Invest, 105(12):1663–8, 2000.
[25] E. H. Javazon, K. J. Beggs, and A. W. Flake. Mesenchymal stem cells: paradoxes of passaging. Exp Hematol, 32(5):414–25, 2004.
[26] A. I. Caplan and S. P. Bruder. Mesenchymal stem cells: building blocks for molecular medicine in the 21st century. Trends Mol Med, 7(6):259–64, 2001.
[27] M. Kassem, M. Kristiansen, and B. M. Abdallah. Mesenchymal stem cells: cell biology and potential use in therapy. Basic Clin Pharmacol Toxicol, 95(5):209–14, 2004.
[28] C. A. Roufosse, N. C. Direkze, W. R. Otto, and N. A. Wright. Circulating mesenchymal stem cells. Int J Biochem Cell Biol, 36(4):585–97, 2004.
[29] O. N. Koc and H. M. Lazarus. Mesenchymal stem cells: heading into the clinic. Bone Marrow Transplant, 27(3):235–9, 2001.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 234 References References
[30] J. J. Minguell, P. Conget, and A. Erices. Biology and clinical utilization of mesenchymal progenitor cells. Braz J Med Biol Res, 33(8):881–7, 2000.
[31] C. A. Vacanti and J. P. Vacanti. The science of tissue engineering. Orthop Clin North Am, 31(3):351–6, 2000.
[32] A. I. Caplan. Mesenchymal stem cells. J Orthop Res, 9(5):641–50, 1991.
[33] A. J. Friedenstein, R. K. Chailakhjan, and K. S. Lalykina. The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells. Cell Tissue Kinet, 3(4):393–403, 1970.
[34] H. Castro-Malaspina, R. E. Gay, G. Resnick, N. Kapoor, P. Meyers, D. Chiarieri, S. McKenzie, H. E. Broxmeyer, and M. A. Moore. Characterization of human bone marrow fibroblast colony-forming cells (cfu-f) and their progeny. Blood, 56(2):289–301, 1980.
[35] M. Owen and A. J. Friedenstein. Stromal stem cells: marrow-derived osteogenic precursors. Ciba Found Symp, 136:42–60, 1988.
[36] S. J. Morrison and J. Kimble. Asymmetric and symmetric stem-cell divisions in development and cancer. Nature, 441(7097):1068–74, 2006.
[37] G. Bianchi, A. Muraglia, A. Daga, G. Corte, R. Cancedda, and R. Quarto. Mi- croenvironment and stem properties of bone marrow-derived mesenchymal cells. Wound Repair Regen, 9(6):460–6, 2001.
[38] C. Fehrer and G. Lepperdinger. Mesenchymal stem cell aging. Exp Gerontol, 40(12):926–30, 2005.
[39] D. Parsch, J. Fellenberg, T. H. Brummendorf, A. M. Eschlbeck, and W. Richter. Telomere length and telomerase activity during expansion and differentiation of human mesenchymal stem cells and chondrocytes. JMolMed, 82(1):49–55, 2004.
[40] S. Zimmermann, M. Voss, S. Kaiser, U. Kapp, C. F. Waller, and U. M. Martens. Lack of telomerase activity in human mesenchymal stem cells. Leukemia, 17(6):1146–9, 2003.
[41] M. A. Baxter, R. F. Wynn, S. N. Jowitt, J. E. Wraith, L. J. Fairbairn, and I. Bellantuono. Study of telomere length reveals rapid aging of human marrow stromal cells following in vitro expansion. Stem Cells, 22(5):675–82, 2004.
[42] F. P. Barry and J. M. Murphy. Mesenchymal stem cells: clinical applications and biological characterization. Int J Biochem Cell Biol, 36(4):568–84, 2004.
[43] H. E. Young, T. A. Steele, R. A. Bray, J. Hudson, J. A. Floyd, K. Hawkins, K. Thomas, T. Austin, C. Edwards, J. Cuzzourt, M. Duenzl, P. A. Lucas, and Jr. Black, A. C. Human reserve pluripotent mesenchymal stem cells are present in the connective tissues of skeletal muscle and dermis derived from fetal, adult, and geriatric donors. Anat Rec, 264(1):51–62, 2001.
[44] N. J. Zvaifler, L. Marinova-Mutafchieva, G. Adams, C. J. Edwards, J. Moss, J. A. Burger, and R. N. Maini. Mesenchymal precursor cells in the blood of normal individuals. Arthritis Res, 2(6):477–88, 2000.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 235 References References
[45] Q. He, C. Wan, and G. Li. Concise review: multipotent mesenchymal stromal cells in blood. Stem Cells, 25(1):69–77, 2007.
[46] F. Dazzi, R. Ramasamy, S. Glennie, S. P. Jones, and I. Roberts. The role of mesenchymal stem cells in haemopoiesis. Blood Rev, 20(3):161–71, 2006.
[47] R. Schofield. The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells, 4(1-2):7–25, 1978.
[48] D. T. Scadden. The stem-cell niche as an entity of action. Nature, 441(7097):1075– 9, 2006.
[49] C. M. Kolf, E. Cho, and R. S. Tuan. Mesenchymal stromal cells. biology of adult mesenchymal stem cells: regulation of niche, self-renewal and differentia- tion. Arthritis Res Ther, 9(1):204, 2007.
[50] M. Owen. Marrow stromal stem cells. J Cell Sci Suppl, 10:63–76, 1988.
[51] R. J. Deans and A. B. Moseley. Mesenchymal stem cells: biology and potential clinical uses. Exp Hematol, 28(8):875–84, 2000.
[52] J. R. Mauney, V. Volloch, and D. L. Kaplan. Role of adult mesenchymal stem cells in bone tissue engineering applications: current status and future prospects. Tissue Eng, 11(5-6):787–802, 2005.
[53] P. Bianco, M. Riminucci, S. Gronthos, and P. G. Robey. Bone marrow stromal stem cells: nature, biology, and potential applications. Stem Cells, 19(3):180–92, 2001.
[54] B. Short, N. Brouard, T. Occhiodoro-Scott, A. Ramakrishnan, and P. J. Simmons. Mesenchymal stem cells. Arch Med Res, 34(6):565–71, 2003.
[55] S. Gronthos, P. J. Simmons, S. E. Graves, and P. G. Robey. Integrin-mediated interactions between human bone marrow stromal precursor cells and the extra- cellular matrix. Bone, 28(2):174–81, 2001.
[56] N. Jaiswal, S. E. Haynesworth, A. I. Caplan, and S. P. Bruder. Osteogenic differentiation of purified, culture-expanded human mesenchymal stem cells in vitro. J Cell Biochem, 64(2):295–312, 1997.
[57] B. Johnstone, T. M. Hering, A. I. Caplan, V. M. Goldberg, and J. U. Yoo. In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Exp Cell Res, 238(1):265–72, 1998.
[58] J. U. Yoo, T. S. Barthel, K. Nishimura, L. Solchaga, A. I. Caplan, V. M. Goldberg, and B. Johnstone. The chondrogenic potential of human bone-marrow-derived mesenchymal progenitor cells. JBoneJointSurgAm, 80(12):1745–57, 1998.
[59] J. H. Bennett, C. J. Joyner, J. T. Triffitt, and M. E. Owen. Adipocytic cells cultured from marrow have osteogenic potential. J Cell Sci, 99 ( Pt 1):131–9, 1991.
[60] J. Justesen, K. Stenderup, E. F. Eriksen, and M. Kassem. Maintenance of os- teoblastic and adipocytic differentiation potential with age and osteoporosis in human marrow stromal cell cultures. Calcif Tissue Int, 71(1):36–44, 2002.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 236 References References
[61] S. R. Park, R. O. Oreffo, and J. T. Triffitt. Interconversion potential of cloned human marrow adipocytes in vitro. Bone, 24(6):549–54, 1999.
[62] M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, and D. R. Marshak. Multilin- eage potential of adult human mesenchymal stem cells. Science, 284(5411):143–7, 1999.
[63] J. J. Minguell, A. Erices, and P. Conget. Mesenchymal stem cells. Exp Biol Med (Maywood), 226(6):507–20, 2001.
[64] N. K. Satija, G. U. Gurudutta, S. Sharma, F. Afrin, P. Gupta, Y. K. Verma, V. K. Singh, and R. P. Tripathi. Mesenchymal stem cells: molecular targets for tissue engineering. Stem Cells Dev, 16(1):7–23, 2007.
[65] Y. Jiang, B. N. Jahagirdar, R. L. Reinhardt, R. E. Schwartz, C. D. Keene, X. R. Ortiz-Gonzalez, M. Reyes, T. Lenvik, T. Lund, M. Blackstad, J. Du, S. Aldrich, A. Lisberg, W. C. Low, D. A. Largaespada, and C. M. Verfaillie. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature, 418(6893):41–9, 2002.
[66] S. J. Odelberg. Inducing cellular dedifferentiation: a potential method for enhanc- ing endogenous regeneration in mammals. Semin Cell Dev Biol, 13(5):335–43, 2002.
[67] W. Zhao, T. Hirose, M. Ishikawa, Y. Oshima, S. Hirai, S. Ohno, and H. Taniguchi. Neonatal pancreatic cells redifferentiate into both neural and pancreatic lineages. Biochem Biophys Res Commun, 352(1):84–90, 2007.
[68] L. Jackson, D. R. Jones, P. Scotting, and V. Sottile. Adult mesenchymal stem cells: Differentiation potential and therapeutic applications. JPostgradMed, 53(2):121–127, 2007.
[69] J. Diaz-Romero, J. P. Gaillard, S. P. Grogan, D. Nesic, T. Trub, and P. Mainil- Varlet. Immunophenotypic analysis of human articular chondrocytes: changes in surface markers associated with cell expansion in monolayer culture. J Cell Physiol, 202(3):731–42, 2005.
[70] Y. Li, J. McClintick, L. Zhong, H. J. Edenberg, M. C. Yoder, and R. J. Chan. Murine embryonic stem cell differentiation is promoted by socs-3 and inhibited by the zinc finger transcription factor klf4. Blood, 105(2):635–7, 2005.
[71] K. Takahashi and S. Yamanaka. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126(4):663–76, 2006.
[72] K. Okita, T. Ichisaka, and S. Yamanaka. Generation of germline-competent in- duced pluripotent stem cells. Nature, 448(7151):313–7, 2007.
[73] K. Takahashi, K. Tanabe, M. Ohnuki, M. Narita, T. Ichisaka, K. Tomoda, and S. Yamanaka. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131(5):861–72, 2007.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 237 References References
[74] J. Yu, M. A. Vodyanik, K. Smuga-Otto, J. Antosiewicz-Bourget, J. L. Frane, S. Tian, J. Nie, G. A. Jonsdottir, V. Ruotti, R. Stewart, II Slukvin, and J. A. Thomson. Induced pluripotent stem cell lines derived from human somatic cells. Science, 318(5858):1917–20, 2007.
[75] M. Nakagawa, M. Koyanagi, K. Tanabe, K. Takahashi, T. Ichisaka, T. Aoi, K. Okita, Y. Mochiduki, N. Takizawa, and S. Yamanaka. Generation of in- duced pluripotent stem cells without myc from mouse and human fibroblasts. Nat Biotechnol, 26(1):101–6, 2008.
[76] D. J. Prockop. Marrow stromal cells as stem cells for nonhematopoietic tissues. Science, 276(5309):71–4, 1997.
[77] S. Gronthos, A. C. Zannettino, S. J. Hay, S. Shi, S. E. Graves, A. Kortesidis, and P. J. Simmons. Molecular and cellular characterisation of highly purified stromal stem cells derived from human bone marrow. J Cell Sci, 116(Pt 9):1827–35, 2003.
[78] J. Cai, M. L. Weiss, and M. S. Rao. In search of ”stemness”. Exp Hematol, 32(7):585–98, 2004.
[79] P. J. Simmons and B. Torok-Storb. Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody, stro-1. Blood, 78(1):55–62, 1991.
[80] S. Sethe, A. Scutt, and A. Stolzing. Aging of mesenchymal stem cells. Ageing Res Rev, 5(1):91–116, 2006.
[81] J. Cai, A. Cheng, Y. Luo, C. Lu, M. P. Mattson, M. S. Rao, and K. Furukawa. Membrane properties of rat embryonic multipotent neural stem cells. J Neu- rochem, 88(1):212–26, 2004.
[82] J. M. Gimble, A. J. Katz, and B. A. Bunnell. Adipose-derived stem cells for regenerative medicine. Circ Res, 100(9):1249–60, 2007.
[83] M. K. Majumdar, M. Keane-Moore, D. Buyaner, W. B. Hardy, M. A. Moorman, K. R. McIntosh, and J. D. Mosca. Characterization and functionality of cell surface molecules on human mesenchymal stem cells. JBiomedSci, 10(2):228– 41, 2003.
[84] L. Zangi, R. Rivkin, I. Kassis, L. Levdansky, G. Marx, and R. Gorodetsky. High- yield isolation, expansion, and differentiation of rat bone marrow-derived mes- enchymal stem cells with fibrin microbeads. Tissue Eng, 12(8):2343–54, 2006.
[85] A. Alhadlaq and J. J. Mao. Mesenchymal stem cells: isolation and therapeutics. Stem Cells Dev, 13(4):436–48, 2004.
[86] S. C. Hung, N. J. Chen, S. L. Hsieh, H. Li, H. L. Ma, and W. H. Lo. Isolation and characterization of size-sieved stem cells from human bone marrow. Stem Cells, 20(3):249–58, 2002.
[87] E. A. Jones, S. E. Kinsey, A. English, R. A. Jones, L. Straszynski, D. M. Meredith, A. F. Markham, A. Jack, P. Emery, and D. McGonagle. Isolation and charac- terization of bone marrow multipotential mesenchymal progenitor cells. Arthritis Rheum, 46(12):3349–60, 2002.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 238 References References
[88] V. L. Battula, P. M. Bareiss, S. Treml, S. Conrad, I. Albert, S. Hojak, H. Abele, B. Schewe, L. Just, T. Skutella, and H. J. Buhring. Human placenta and bone marrow derived msc cultured in serum-free, b-fgf-containing medium express cell surface frizzled-9 and ssea-4 and give rise to multilineage differentiation. Differ- entiation, 75(4):279–91, 2007. [89] Y. Fukuchi, H. Nakajima, D. Sugiyama, I. Hirose, T. Kitamura, and K. Tsuji. Human placenta-derived cells have mesenchymal stem/progenitor cell potential. Stem Cells, 22(5):649–58, 2004. [90] Z. Miao, J. Jin, L. Chen, J. Zhu, W. Huang, J. Zhao, H. Qian, and X. Zhang. Iso- lation of mesenchymal stem cells from human placenta: comparison with human bone marrow mesenchymal stem cells. Cell Biol Int, 30(9):681–7, 2006. [91] J. E. Grove, E. Bruscia, and D. S. Krause. Plasticity of bone marrow-derived stem cells. Stem Cells, 22(4):487–500, 2004. [92] S. P. Bruder, K. H. Kraus, V. M. Goldberg, and S. Kadiyala. The effect of implants loaded with autologous mesenchymal stem cells on the healing of canine segmental bone defects. JBoneJointSurgAm, 80(7):985–96, 1998. [93] S. Kadiyala, R. G. Young, M. A. Thiede, and S. P. Bruder. Culture expanded canine mesenchymal stem cells possess osteochondrogenic potential in vivo and in vitro. Cell Transplant, 6(2):125–34, 1997. [94] A. R. Derubeis and R. Cancedda. Bone marrow stromal cells (bmscs) in bone engineering: limitations and recent advances. Ann Biomed Eng, 32(1):160–5, 2004. [95] S. P. Bruder and B. S. Fox. Tissue engineering of bone. cell based strategies. Clin Orthop, (367 Suppl):S68–83, 1999. [96] A. J. Salgado, O. P. Coutinho, and R. L. Reis. Bone tissue engineering: state of the art and future trends. Macromol Biosci, 4(8):743–65, 2004. [97] U. Meyer, U. Joos, and H. P. Wiesmann. Biological and biophysical principles in extracorporal bone tissue engineering. part i. Int J Oral Maxillofac Surg, 33(4):325–32, 2004. [98] H. P. Wiesmann, U. Joos, and U. Meyer. Biological and biophysical principles in extracorporal bone tissue engineering. part ii. Int J Oral Maxillofac Surg, 33(6):523–30, 2004. [99] U. Meyer, U. Joos, and H. P. Wiesmann. Biological and biophysical principles in extracorporal bone tissue engineering. part iii. Int J Oral Maxillofac Surg, 33(7):635–41, 2004. [100] J. Glowacki. In vitro engineering of cartilage. J Rehabil Res Dev, 37(2):171–7, 2000. [101] D. W. Jackson, T. M. Simon, and H. M. Aberman. Symptomatic articular cartilage degeneration: the impact in the new millennium. Clin Orthop, (391 Suppl):S14–25, 2001. [102] C. K. Kuo, W. J. Li, R. L. Mauck, and R. S. Tuan. Cartilage tissue engineering: its potential and uses. Curr Opin Rheumatol, 18(1):64–73, 2006.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 239 References References
[103] S. D. Waldman, M. D. Grynpas, R. M. Pilliar, and R. A. Kandel. The use of specific chondrocyte populations to modulate the properties of tissue-engineered cartilage. Journal of Orthopaedic Research, 21(1):132–8, 2003.
[104] M. V. Risbud and M. Sittinger. Tissue engineering: advances in in vitro cartilage generation. Trends Biotechnol, 20(8):351–6, 2002.
[105] R. D. Coutts, R. M. Healey, R. Ostrander, R. L. Sah, R. Goomer, and D. Amiel. Matrices for cartilage repair. Clin Orthop, (391 Suppl):S271–9, 2001.
[106] S. Wakitani, K. Imoto, T. Yamamoto, M. Saito, N. Murata, and M. Yoneda. Hu- man autologous culture expanded bone marrow mesenchymal cell transplantation for repair of cartilage defects in osteoarthritic knees. Osteoarthritis Cartilage, 10(3):199–206, 2002.
[107] I. Martin, S. Miot, A. Barbero, M. Jakob, and D. Wendt. Osteochondral tissue engineering. JBiomech, 40(4):750–65, 2007.
[108] P. Godara, C. D. McFarland, and R. E. Nordon. Design of bioreactors for mesenchymal stem cell tissue engineering. Journal of Chemical Technology and Biotechnology, 83(4):408–420, 2008.
[109] J. W. Lee, Y. H. Kim, K. D. Park, K. S. Jee, J. W. Shin, and S. B. Hahn. Importance of integrin beta1-mediated cell adhesion on biodegradable polymers under serum depletion in mesenchymal stem cells and chondrocytes. Biomaterials, 25(10):1901–9, 2004.
[110] E. S. Jeon, H. Y. Song, M. R. Kim, H. J. Moon, Y. C. Bae, J. S. Jung, and J. H. Kim. Sphingosylphosphorylcholine induces proliferation of human adipose tissue- derived mesenchymal stem cells via activation of jnk. J Lipid Res, 47(3):653–64, 2006.
[111] J. Farre, S. Roura, C. Prat-Vidal, C. Soler-Botija, A. Llach, C. E. Molina, L. Hove-Madsen, J. J. Cairo, F. Godia, R. Bragos, J. Cinca, and A. Bayes-Genis. Fgf-4 increases in vitro expansion rate of human adult bone marrow-derived mes- enchymal stem cells. Growth Factors, 25(2):71–6, 2007.
[112] S. Shirasawa, I. Sekiya, Y. Sakaguchi, K. Yagishita, S. Ichinose, and T. Muneta. In vitro chondrogenesis of human synovium-derived mesenchymal stem cells: op- timal condition and comparison with bone marrow-derived cells. J Cell Biochem, 97(1):84–97, 2006.
[113] A. Yokoyama, T. Muneta, A. Nimura, H. Koga, T. Mochizuki, Y. Hata, and I. Sekiya. Fgf2 and dexamethasone increase the production of hyaluronan in two- dimensional culture of elastic cartilage-derived cells: in vitro analyses and in vivo cartilage formation. Cell Tissue Res, 329(3):469–78, 2007.
[114] C. Fehrer, R. Brunauer, G. Laschober, H. Unterluggauer, S. Reitinger, F. Kloss, C. Gully, R. Gassner, and G. Lepperdinger. Reduced oxygen tension attenu- ates differentiation capacity of human mesenchymal stem cells and prolongs their lifespan. Aging Cell, 6(6):745–57, 2007.
[115] C. L. Murphy and J. M. Polak. Control of human articular chondrocyte differen- tiation by reduced oxygen tension. J Cell Physiol, 199(3):451–9, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 240 References References
[116] A. Grzelak, B. Rychlik, and G. Bartosz. Light-dependent generation of reactive oxygen species in cell culture media. Free Radic Biol Med, 30(12):1418–25, 2001. [117] F. Q. Schafer and G. R. Buettner. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic Biol Med, 30(11):1191–212, 2001. [118] J. M. Reyes, S. Fermanian, F. Yang, S. Y. Zhou, S. Herretes, D. B. Murphy, J. H. Elisseeff, and R. S. Chuck. Metabolic changes in mesenchymal stem cells in osteogenic medium measured by autofluorescence spectroscopy. Stem Cells, 24(5):1213–7, 2006. [119] T. T. Chowdhury, D. L. Bader, J. C. Shelton, and D. A. Lee. Temporal reg- ulation of chondrocyte metabolism in agarose constructs subjected to dynamic compression. Arch Biochem Biophys, 417(1):105–11, 2003. [120] C. R. Lee, A. J. Grodzinsky, and M. Spector. Biosynthetic response of passaged chondrocytes in a type ii collagen scaffold to mechanical compression. JBiomed Mater Res A, 64(3):560–9, 2003. [121] D. A. Lee, T. Noguchi, S. P. Frean, P. Lees, and D. L. Bader. The influence of mechanical loading on isolated chondrocytes seeded in agarose constructs. Biorhe- ology, 37(1-2):149–61, 2000. [122] K. Kurpinski, J. Chu, C. Hashi, and S. Li. Anisotropic mechanosensing by mes- enchymal stem cells. Proc Natl Acad Sci U S A, 103(44):16095–100, 2006. [123] F. Zhao, R. Chella, and T. Ma. Effects of shear stress on 3-d human mesenchymal stem cell construct development in a perfusion bioreactor system: Experiments and hydrodynamic modeling. Biotechnol Bioeng, 96(3):584–95, 2007. [124] G. L. Jiang, C. R. White, H. Y. Stevens, and J. A. Frangos. Temporal gradients in shear stimulate osteoblastic proliferation via erk1/2 and retinoblastoma protein. Am J Physiol Endocrinol Metab, 283(2):E383–9, 2002. [125] D. von Heimburg, K. Hemmrich, S. Zachariah, H. Staiger, and N. Pallua. Oxygen consumption in undifferentiated versus differentiated adipogenic mesenchymal precursor cells. Respir Physiol Neurobiol, 146(2-3):107–16, 2005. [126] J. Malda, P. van den Brink, P. Meeuwse, M. Grojec, D. E. Martens, J. Tramper, J. Riesle, and C. A. van Blitterswijk. Effect of oxygen tension on adult articu- lar chondrocytes in microcarrier bioreactor culture. Tissue Eng, 10(7-8):987–94, 2004. [127] F. Zhao, P. Pathi, W. Grayson, Q. Xing, B. R. Locke, and T. Ma. Effects of oxygen transport on 3-d human mesenchymal stem cell metabolic activity in perfusion and static cultures: experiments and mathematical model. Biotechnol Prog, 21(4):1269–80, 2005. [128] M. C. Lewis, B. D. Macarthur, J. Malda, G. Pettet, and C. P. Please. Hetero- geneous proliferation within engineered cartilaginous tissue: the role of oxygen tension. Biotechnol Bioeng, 91(5):607–15, 2005. [129] J. F. Markusen, C. Mason, D. A. Hull, M. A. Town, A. B. Tabor, M. Clements, C. H. Boshoff, and P. Dunnill. Behavior of adult human mesenchymal stem cells entrapped in alginate-grgdy beads. Tissue Eng, 12(4):821–30, 2006.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 241 References References
[130] M. Tomita, E. F. Sato, M. Nishikawa, Y. Yamano, and M. Inoue. Nitric ox- ide regulates mitochondrial respiration and functions of articular chondrocytes. Arthritis Rheum, 44(1):96–104, 2001.
[131] R. A. Windhaber, R. J. Wilkins, and D. Meredith. Functional characterisation of glucose transport in bovine articular chondrocytes. Pflugers Arch, 446(5):572–7, 2003.
[132] B. R. Bird, W. E. Steward, and E. N. Lightfoot. Transport Phenomena. Wiley, New York, 1960.
[133] R. E. Spier and B. Griffiths. An examination of the data and concepts germane to the oxygenation of cultured animal cells. Dev Biol Stand, 55:81–92, 1983.
[134] L. E. Freed, G. Vunjak-Novakovic, and R. Langer. Cultivation of cell-polymer cartilage implants in bioreactors. J Cell Biochem, 51(3):257–64, 1993.
[135] C. J. Galban and B. R. Locke. Analysis of cell growth kinetics and substrate diffusion in a polymer scaffold. Biotechnol Bioeng, 65(2):121–32, 1999.
[136] C. J. Galban and B. R. Locke. Effects of spatial variation of cells and nutrient and product concentrations coupled with product inhibition on cell growth in a polymer scaffold. Biotechnol Bioeng, 64(6):633–43, 1999.
[137] H. A. Leddy, H. A. Awad, and F. Guilak. Molecular diffusion in tissue-engineered cartilage constructs: effects of scaffold material, time, and culture conditions. J Biomed Mater Res B Appl Biomater, 70(2):397–406, 2004.
[138] C. A. Peng and B. O. Palsson. Determination of specific oxygen uptake rates in human hematopoietic cultures and implications for bioreactor design. Ann Biomed Eng, 24(3):373–81, 1996.
[139] J. E. Dennis, K. Esterly, A. Awadallah, C. R. Parrish, G. M. Poynter, and K. L. Goltry. Clinical-scale expansion of a mixed population of bone-marrow-derived stem and progenitor cells for potential use in bone-tissue regeneration. Stem Cells, 25(10):2575–82, 2007.
[140] T. J. Klein and R. L. Sah. Modulation of depth-dependent properties in tissue- engineered cartilage with a semi-permeable membrane and perfusion: a contin- uum model of matrix metabolism and transport. Biomech Model Mechanobiol, 6(1-2):21–32, 2007.
[141] R. E. Nordon and K. Schindhelm. Design of hollow fiber modules for uniform shear elution affinity cell separation. Artif Organs, 21(2):107–15, 1997.
[142] B. Porter, R. Zauel, H. Stockman, R. Guldberg, and D. Fyhrie. 3-d computational modeling of media flow through scaffolds in a perfusion bioreactor. JBiomech, 38(3):543–9, 2005.
[143] G. N. Bancroft, V. I. Sikavitsas, J. van den Dolder, T. L. Sheffield, C. G. Ambrose, J. A. Jansen, and A. G. Mikos. Fluid flow increases mineralized matrix deposi- tion in 3d perfusion culture of marrow stromal osteoblasts in a dose-dependent manner. Proc Natl Acad Sci U S A, 99(20):12600–5, 2002.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 242 References References
[144] J. Glowacki, S. Mizuno, and J. S. Greenberger. Perfusion enhances functions of bone marrow stromal cells in three-dimensional culture. Cell Transplant, 7(3):319–26, 1998.
[145] N. Mahmoudifar and P. M. Doran. Tissue engineering of human cartilage and osteochondral composites using recirculation bioreactors. Biomaterials, 26(34):7012–24, 2005.
[146] A. S. Goldstein, T. M. Juarez, C. D. Helmke, M. C. Gustin, and A. G. Mikos. Effect of convection on osteoblastic cell growth and function in biodegradable polymer foam scaffolds. Biomaterials, 22(11):1279–88, 2001.
[147] R. E. Nordon. Method and apparatus for culturing cells, 1999.
[148] C. Granet, N. Laroche, L. Vico, C. Alexandre, and M. H. Lafage-Proust. Rotating-wall vessels, promising bioreactors for osteoblastic cell culture: com- parison with other 3d conditions. Med Biol Eng Comput, 36(4):513–9, 1998.
[149] Y. Yang, F. M. Rossi, and E. E. Putnins. Ex vivo expansion of rat bone marrow mesenchymal stromal cells on microcarrier beads in spin culture. Biomaterials, 28(20):3110–20, 2007.
[150] E. Charriere, J. Lemaitre, and P. Zysset. Hydroxyapatite cement scaffolds with controlled macroporosity: fabrication protocol and mechanical properties. Bio- materials, 24(5):809–17, 2003.
[151] J. G. Dellinger, 3rd Cesarano, J., and R. D. Jamison. Robotic deposition of model hydroxyapatite scaffolds with multiple architectures and multiscale porosity for bone tissue engineering. JBiomedMaterResA, 82(2):383–94, 2007.
[152] T. Mygind, M. Stiehler, A. Baatrup, H. Li, X. Zou, A. Flyvbjerg, M. Kassem, and C. Bunger. Mesenchymal stem cell ingrowth and differentiation on coralline hydroxyapatite scaffolds. Biomaterials, 28(6):1036–47, 2007.
[153] D. Turhani, E. Watzinger, M. Weissenbock, B. Cvikl, D. Thurnher, G. Wittwer, K. Yerit, and R. Ewers. Analysis of cell-seeded 3-dimensional bone constructs manufactured in vitro with hydroxyapatite granules obtained from red algae. J Oral Maxillofac Surg, 63(5):673–81, 2005.
[154] Z. Ge, S. Baguenard, L. Y. Lim, A. Wee, and E. Khor. Hydroxyapatite-chitin ma- terials as potential tissue engineered bone substitutes. Biomaterials, 25(6):1049– 58, 2004.
[155] C. T. Harris and L. F. Cooper. Comparison of bone graft matrices for human mesenchymal stem cell-directed osteogenesis. JBiomedMaterResA, 68(4):747– 55, 2004.
[156] N. Kotobuki, K. Ioku, D. Kawagoe, H. Fujimori, S. Goto, and H. Ohgushi. Ob- servation of osteogenic differentiation cascade of living mesenchymal stem cells on transparent hydroxyapatite ceramics. Biomaterials, 26(7):779–85, 2005.
[157] T. Morishita, K. Honoki, H. Ohgushi, N. Kotobuki, A. Matsushima, and Y. Takakura. Tissue engineering approach to the treatment of bone tumors: three cases of cultured bone grafts derived from patients’ mesenchymal stem cells. Artif Organs, 30(2):115–8, 2006.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 243 References References
[158] L. De Bartolo, A. Piscioneri, G. Cotroneo, S. Salerno, F. Tasselli, C. Campana, S. Morelli, M. Rende, M. C. Caroleo, M. Bossio, and E. Drioli. Human lymphocyte peek-wc hollow fiber membrane bioreactor. J Biotechnol, 132(1):65–74, 2007.
[159] M. Nishikawa, A. Myoui, H. Ohgushi, M. Ikeuchi, N. Tamai, and H. Yoshikawa. Bone tissue engineering using novel interconnected porous hydroxyapatite ceram- ics combined with marrow mesenchymal cells: quantitative and three-dimensional image analysis. Cell Transplant, 13(4):367–76, 2004.
[160] A. A. Sawyer, K. M. Hennessy, and S. L. Bellis. The effect of adsorbed serum proteins, rgd and proteoglycan-binding peptides on the adhesion of mesenchymal stem cells to hydroxyapatite. Biomaterials, 28(3):383–92, 2007.
[161] A. A. Sawyer, D. M. Weeks, S. S. Kelpke, M. S. McCracken, and S. L. Bellis. The effect of the addition of a polyglutamate motif to rgd on peptide tethering to hydroxyapatite and the promotion of mesenchymal stem cell adhesion. Bio- materials, 26(34):7046–56, 2005.
[162] H. Yoshikawa and A. Myoui. Bone tissue engineering with porous hydroxyapatite ceramics. J Artif Organs, 8(3):131–6, 2005.
[163] F. Zhao, W. L. Grayson, T. Ma, B. Bunnell, and W. W. Lu. Effects of hydroxya- patite in 3-d chitosan-gelatin polymer network on human mesenchymal stem cell construct development. Biomaterials, 27(9):1859–67, 2006.
[164] W. Bensaid, K. Oudina, V. Viateau, E. Potier, V. Bousson, C. Blanchat, L. Sedel, G. Guillemin, and H. Petite. De novo reconstruction of functional bone by tissue engineering in the metatarsal sheep model. Tissue Eng, 11(5-6):814–24, 2005.
[165] S. Hofmann, H. Hagenmuller, A. M. Koch, R. Muller, G. Vunjak-Novakovic, D. L. Kaplan, H. P. Merkle, and L. Meinel. Control of in vitro tissue-engineered bone- like structures using human mesenchymal stem cells and porous silk scaffolds. Biomaterials, 28(6):1152–62, 2007.
[166] H. J. Kim, U. J. Kim, G. G. Leisk, C. Bayan, I. Georgakoudi, and D. L. Kaplan. Bone regeneration on macroporous aqueous-derived silk 3-d scaffolds. Macromol Biosci, 7(5):643–55, 2007.
[167] C. Li, C. Vepari, H. J. Jin, H. J. Kim, and D. L. Kaplan. Electrospun silk-bmp-2 scaffolds for bone tissue engineering. Biomaterials, 27(16):3115–24, 2006.
[168] L. Meinel, R. Fajardo, S. Hofmann, R. Langer, J. Chen, B. Snyder, G. Vunjak- Novakovic, and D. Kaplan. Silk implants for the healing of critical size bone defects. Bone, 37(5):688–98, 2005.
[169] R. Nazarov, H. J. Jin, and D. L. Kaplan. Porous 3-d scaffolds from regenerated silk fibroin. Biomacromolecules, 5(3):718–26, 2004.
[170] X. Li, Q. Feng, W. Wang, and F. Cui. Chemical characteristics and cytocom- patibility of collagen-based scaffold reinforced by chitin fibers for bone tissue engineering. J Biomed Mater Res B Appl Biomater, 77(2):219–26, 2006.
[171] X. B. Yang, R. S. Bhatnagar, S. Li, and R. O. Oreffo. Biomimetic collagen scaf- folds for human bone cell growth and differentiation. Tissue Eng, 10(7-8):1148–59, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 244 References References
[172] J. M. Karp, M. S. Shoichet, and J. E. Davies. Bone formation on two-dimensional poly(dl-lactide-co-glycolide) (plga) films and three-dimensional plga tissue engi- neering scaffolds in vitro. JBiomedMaterResA, 64(2):388–96, 2003.
[173] W. Huang, B. Carlsen, I. Wulur, G. Rudkin, K. Ishida, B. Wu, D. T. Yamaguchi, and T. A. Miller. Bmp-2 exerts differential effects on differentiation of rabbit bone marrow stromal cells grown in two-dimensional and three-dimensional systems and is required for in vitro bone formation in a plga scaffold. Exp Cell Res, 299(2):325–34, 2004.
[174] W. S. Kim and H. K. Kim. Tissue engineered vascularized bone formation using in vivo implanted osteoblast-polyglycolic acid scaffold. J Korean Med Sci, 20(3):479– 82, 2005.
[175] N. Mahmoudifar and P. M. Doran. Tissue engineering of human cartilage in bioreactors using single and composite cell-seeded scaffolds. Biotechnol Bioeng, 91(3):338–55, 2005.
[176] J. S. Wayne, C. L. McDowell, K. J. Shields, and R. S. Tuan. In vivo response of polylactic acid-alginate scaffolds and bone marrow-derived cells for cartilage tissue engineering. Tissue Eng, 11(5-6):953–63, 2005.
[177] K. Uematsu, K. Hattori, Y. Ishimoto, J. Yamauchi, T. Habata, Y. Takakura, H. Ohgushi, T. Fukuchi, and M. Sato. Cartilage regeneration using mesenchy- mal stem cells and a three-dimensional poly-lactic-glycolic acid (plga) scaffold. Biomaterials, 26(20):4273–9, 2005.
[178] V. Dewar, P. Voet, F. Denamur, and J. Smal. Industrial implementation of in vitro production of monoclonal antibodies. Ilar J, 46(3):307–13, 2005.
[179] S. Hofmann, S. Knecht, R. Langer, D. L. Kaplan, G. Vunjak-Novakovic, H. P. Merkle, and L. Meinel. Cartilage-like tissue engineering using silk scaffolds and mesenchymal stem cells. Tissue Eng, 12(10):2729–38, 2006.
[180] Y. Wang, U. J. Kim, D. J. Blasioli, H. J. Kim, and D. L. Kaplan. In vitro cartilage tissue engineering with 3d porous aqueous-derived silk scaffolds and mesenchymal stem cells. Biomaterials, 26(34):7082–94, 2005.
[181] B. M. Baker and R. L. Mauck. The effect of nanofiber alignment on the matura- tion of engineered meniscus constructs. Biomaterials, 28(11):1967–77, 2007.
[182] Z. Chen, Y. Ding, and G. Li. Configuration of a new bioartificial liver support system and in vitro evaluation of its functions. AnnClinLabSci, 35(1):7–14, 2005.
[183] T. Yamasaki, M. Deie, R. Shinomiya, Y. Izuta, Y. Yasunaga, S. Yanada, P. Shar- man, and M. Ochi. Meniscal regeneration using tissue engineering with a scaffold derived from a rat meniscus and mesenchymal stromal cells derived from rat bone marrow. JBiomedMaterResA, 75(1):23–30, 2005.
[184] L. Meinel, V. Karageorgiou, R. Fajardo, B. Snyder, V. Shinde-Patil, L. Zichner, D. Kaplan, R. Langer, and G. Vunjak-Novakovic. Bone tissue engineering using human mesenchymal stem cells: effects of scaffold material and medium flow. Ann Biomed Eng, 32(1):112–22, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 245 References References
[185] R. P. Schwarz, T. J. Goodwin, and D. A. Wolf. Cell culture for three-dimensional modeling in rotating-wall vessels: an application of simulated microgravity. J Tissue Cult Methods, 14(2):51–7, 1992.
[186] K. Song, Z. Yang, T. Liu, W. Zhi, X. Li, L. Deng, Z. Cui, and X. Ma. Fabrication and detection of tissue-engineered bones with bio-derived scaffolds in a rotating bioreactor. Biotechnol Appl Biochem, 45(Pt 2):65–74, 2006.
[187] Q. Q. Qiu, P. Ducheyne, and P. S. Ayyaswamy. Fabrication, characterization and evaluation of bioceramic hollow microspheres used as microcarriers for 3-d bone tissue formation in rotating bioreactors. Biomaterials, 20(11):989–1001, 1999.
[188] J. Duke, E. Daane, J. Arizpe, and D. Montufar-Solis. Chondrogenesis in aggre- gates of embryonic limb cells grown in a rotating wall vessel. Adv Space Res, 17(6-7):289–93, 1996.
[189] J. C. Hu and K. A. Athanasiou. Low-density cultures of bovine chondrocytes: effects of scaffold material and culture system. Biomaterials, 26(14):2001–12, 2005.
[190] D. L. Nettles, S. H. Elder, and J. A. Gilbert. Potential use of chitosan as a cell scaffold material for cartilage tissue engineering. Tissue Eng, 8(6):1009–16, 2002.
[191] Y. Ohyabu, N. Kida, H. Kojima, T. Taguchi, J. Tanaka, and T. Uemura. Carti- laginous tissue formation from bone marrow cells using rotating wall vessel (rwv) bioreactor. Biotechnol Bioeng, 95(5):1003–8, 2006.
[192] V. Singh. Disposable bioreactor for cell culture using wave-induced agitation. Cytotechnology, 30(1-3):149–58, 1999.
[193] J. Kadarusman, R. Bhatia, J. McLaughlin, and W. R. Lin. Growing cholesterol- dependent ns0 myeloma cell line in the wave bioreactor system: overcoming cholesterol-polymer interaction by using pretreated polymer or inert fluorinated ethylene propylene. Biotechnol Prog, 21(4):1341–6, 2005.
[194] G. N. Bancroft, V. I. Sikavitsas, and A. G. Mikos. Design of a flow perfusion bioreactor system for bone tissue-engineering applications. Tissue Eng, 9(3):549– 54, 2003.
[195] S. M. Mueller, S. Mizuno, L. C. Gerstenfeld, and J. Glowacki. Medium perfusion enhances osteogenesis by murine osteosarcoma cells in three-dimensional collagen sponges. JBoneMinerRes, 14(12):2118–26, 1999.
[196] H. Hosseinkhani, Y. Inatsugu, Y. Hiraoka, S. Inoue, and Y. Tabata. Perfusion cul- ture enhances osteogenic differentiation of rat mesenchymal stem cells in collagen sponge reinforced with poly(glycolic acid) fiber. Tissue Eng, 11(9-10):1476–88, 2005.
[197] A. Braccini, D. Wendt, C. Jaquiery, M. Jakob, M. Heberer, L. Kenins, A. Wodnar- Filipowicz, R. Quarto, and I. Martin. Three-dimensional perfusion culture of human bone marrow cells and generation of osteoinductive grafts. Stem Cells, 23(8):1066–72, 2005.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 246 References References
[198] Y. Wang, T. Uemura, J. Dong, H. Kojima, J. Tanaka, and T. Tateishi. Appli- cation of perfusion culture system improves in vitro and in vivo osteogenesis of bone marrow-derived osteoblastic cells in porous ceramic materials. Tissue Eng, 9(6):1205–14, 2003.
[199] D. Wendt, A. Marsano, M. Jakob, M. Heberer, and I. Martin. Oscillating perfu- sion of cell suspensions through three-dimensional scaffolds enhances cell seeding efficiency and uniformity. Biotechnol Bioeng, 84(2):205–14, 2003.
[200] J. Wang, Y. Asou, I. Sekiya, S. Sotome, H. Orii, and K. Shinomiya. Enhancement of tissue engineered bone formation by a low pressure system improving cell seeding and medium perfusion into a porous scaffold. Biomaterials, 27(13):2738– 46, 2006.
[201] T. Davisson, R. L. Sah, and A. Ratcliffe. Perfusion increases cell content and ma- trix synthesis in chondrocyte three-dimensional cultures. Tissue Eng, 8(5):807–16, 2002.
[202] D. Pazzano, K. A. Mercier, J. M. Moran, S. S. Fong, D. D. DiBiasio, J. X. Rulfs, S. S. Kohles, and L. J. Bonassar. Comparison of chondrogensis in static and perfused bioreactor culture. Biotechnol Prog, 16(5):893–6, 2000.
[203] N. Mahmoudifar and P. M. Doran. Effect of seeding and bioreactor culture con- ditions on the development of human tissue-engineered cartilage. Tissue Eng, 12(6):1675–85, 2006.
[204] N. S. Dunkelman, M. P. Zimber, R. G. Lebaron, R. Pavelec, M. Kwan, and A. F. Purchio. Cartilage production by rabbit articular chondrocytes on polyglycolic acid scaffolds in a closed bioreactor system. Biotechnol Bioeng, 46(4):299–305, 1995.
[205] C. R. Roberts, J. K. Rains, P. D. Pare, D. C. Walker, B. Wiggs, and J. L. Bert. Ultrastructure and tensile properties of human tracheal cartilage. JBiomech, 31(1):81–6, 1998.
[206] W. J. Li, R. Tuli, C. Okafor, A. Derfoul, K. G. Danielson, D. J. Hall, and R. S. Tuan. A three-dimensional nanofibrous scaffold for cartilage tissue engineering using human mesenchymal stem cells. Biomaterials, 26(6):599–609, 2005.
[207] M. R. Koller, S. G. Emerson, and B. O. Palsson. Large-scale expansion of human stem and progenitor cells from bone marrow mononuclear cells in continuous perfusion cultures. Blood, 82(2):378–84, 1993.
[208] C. A. Peng and B. O. Palsson. Cell growth and differentiation on feeder layers is predicted to be influenced by bioreactor geometry. Biotechnol Bioeng, 50(5):479– 92, 1996.
[209] Koller M. Smith A. Mandalam, R. Ex vivo hematopoietic cell expansion for bone marrow transplantation. In Nordon R. Schindhelm, K., editor, Ex Vivo Cell Therapy, pages 273–291. Academic Press, San Diego, CA, 1999.
[210] R. A. Knazek, Y. W. Wu, P. M. Aebersold, and S. A. Rosenberg. Culture of human tumor infiltrating lymphocytes in hollow fiber bioreactors. J Immunol Methods, 127(1):29–37, 1990.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 247 References References
[211] C. C. Malone, P. M. Schiltz, A. D. Mackintosh, L. D. Beutel, F. S. Heinemann, and R. O. Dillman. Characterization of human tumor-infiltrating lymphocytes expanded in hollow-fiber bioreactors for immunotherapy of cancer. Cancer Bio- ther Radiopharm, 16(5):381–90, 2001. [212] D. Pan, D. F. Stroncek, and C. B. Whitley. Improved gene transfer and nor- malized enzyme levels in primitive hematopoietic progenitors from patients with mucopolysaccharidosis type i using a bioreactor. JGeneMed, 6(12):1293–303, 2004. [213] D. Pan and C. B. Whitley. Closed hollow-fiber bioreactor: a new approach to retroviral vector production. JGeneMed, 1(6):433–40, 1999. [214] T. A. Gardner, S. C. Ko, L. Yang, J. J. Cadwell, L. W. Chung, and C. Kao. Serum-free recombinant production of adenovirus using a hollow fiber capillary system. Biotechniques, 30(2):422–7, 2001. [215] M. J. Gramer and J. Maas. Optimal ns0 cell growth in a hollow fiber bioreactor requires increased serum concentration or a cholesterol supplement on the cell side of the fiber. Biotechnol Prog, 19(6):1762–6, 2003. [216] B. Kalbfuss, Y. Genzel, M. Wolff, A. Zimmermann, R. Morenweiser, and U. Re- ichl. Harvesting and concentration of human influenza a virus produced in serum- free mammalian cell culture for the production of vaccines. Biotechnol Bioeng, 97(1):73–85, 2007. [217] J. Kuhn, K. Molle, T. Brinkmann, C. Gotting, and K. Kleesiek. High-density tissue-like cultivation of jar choriocarcinoma cells for the in vitro production of human xylosyltransferase. J Biotechnol, 103(2):191–6, 2003. [218] S. F. Abu-Absi, G. Seth, R. A. Narayanan, K. Groehler, P. Lai, M. L. Anderson, T. Sielaff, and W. S. Hu. Characterization of a hollow fiber bioartificial liver device. Artif Organs, 29(5):419–22, 2005. [219] E. Curcio, L. De Bartolo, G. Barbieri, M. Rende, L. Giorno, S. Morelli, and E. Drioli. Diffusive and convective transport through hollow fiber membranes for liver cell culture. J Biotechnol, 117(3):309–21, 2005. [220] H. F. Lu, W. S. Lim, P. C. Zhang, S. M. Chia, H. Yu, H. Q. Mao, and K. W. Leong. Galactosylated poly(vinylidene difluoride) hollow fiber bioreactor for hepatocyte culture. Tissue Eng, 11(11-12):1667–77, 2005. [221] Q. Meng, G. Zhang, and D. Wu. Hepatocyte culture in bioartificial livers with different membrane characteristics. Biotechnol Lett, 26(18):1407–12, 2004. [222] Q. Shi, R. Mitteregger, D. Falkenhagen, and Y. T. Yu. A novel configuration of bioartificial liver support system based on circulating microcarrier culture. Artif Cells Blood Substit Immobil Biotechnol, 28(4):273–91, 2000. [223] K. Potter, L. H. Kidder, I. W. Levin, E. N. Lewis, and R. G. Spencer. Imaging of collagen and proteoglycan in cartilage sections using fourier transform infrared spectral imaging. Arthritis Rheum, 44(4):846–55, 2001. [224] M. Kim, X. Bi, W. E. Horton, R. G. Spencer, and N. P. Camacho. Fourier transform infrared imaging spectroscopic analysis of tissue engineered cartilage: histologic and biochemical correlations. JBiomedOpt, 10(3):031105, 2005.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 248 References References
[225] R. E. Nordon, A. Shu, F. Camacho, and B. K. Milthorpe. Hollow-fiber assay for ligand-mediated cell adhesion. Cytometry A, 57(1):39–44, 2004.
[226] R. E. Nordon, B. K. Milthorpe, K. Schindhelm, and P. R. Slowiaczek. An exper- imental model of affinity cell separation. Cytometry, 16(1):25–33, 1994.
[227] C. Heath, G. Belfort, B. E. Hammer, S. D. Mirer, and J. M. Pimbley. Magnetic resonance imaging and modeling of flow in hollow-fiber bioreactors. Journal of American Institute of Chemical Engineers, 36(4):547–558, 1990.
[228] Y. Xia and G. M. Whitesides. Soft lithography. Annual Review of Materials Science, 28:153–184, 1998.
[229] Y. Xia and G. M. Whitesides. Soft lithography. Angewandte Chemie International Edition, 37:550–575, 1998.
[230] R. S. Kane, S. Takayama, E. Ostuni, D. E. Ingber, and G. M. Whitesides. Pat- terning proteins and cells using soft lithography. Biomaterials, 20(23-24):2363–76, 1999.
[231] E. J. Kim, Y. Xia, and G. M. Whitesides. Polymer microstructures formed by moulding in capillaries. Nature, 376:581–584, 1995.
[232] E. Leclerc, Y. Sakai, and T. Fujii. Cell culture in 3-dimensional microfluidic structure of pdms (polydimethylsiloxane). Biomedical Microdevices, 5(2):109– 114, 2003.
[233] T. H. Park and M. L. Shuler. Integration of cell culture and microfabrication technology. Biotechnol Prog, 19(2):243–53, 2003.
[234] N. R. Beer, B. J. Hindson, E. K. Wheeler, S. B. Hall, K. A. Rose, I. M. Kennedy, and B. W. Colston. On-chip, real-time, single-copy polymerase chain reaction in picoliter droplets. Anal Chem, 79(22):8471–5, 2007.
[235] R. Gomez-Sjoberg, A. A. Leyrat, D. M. Pirone, C. S. Chen, and S. R. Quake. Ver- satile, fully automated, microfluidic cell culture system. Anal Chem, 79(22):8557– 63, 2007.
[236] M. A. Unger, H. P. Chou, T. Thorsen, A. Scherer, and S. R. Quake. Mono- lithic microfabricated valves and pumps by multilayer soft lithography. Science, 288(5463):113–6, 2000.
[237] A. R. Wheeler, W. R. Throndset, R. J. Whelan, A. M. Leach, R. N. Zare, Y. H. Liao, K. Farrell, I. D. Manger, and A. Daridon. Microfluidic device for single-cell analysis. Anal Chem, 75(14):3581–6, 2003.
[238] H. Yin, N. Pattrick, X. Zhang, N. Klauke, H. C. Cordingley, S. J. Haswell, and J. M. Cooper. Quantitative comparison between microfluidic and microtiter plate formats for cell-based assays. Anal Chem, 80(1):179–85, 2008.
[239] E. Leclerc, Y. Sakai, and T. Fujii. Microfluidic pdms (polydimethylsiloxane) bioreactor for large-scale culture of hepatocytes. Biotechnol Prog, 20(3):750–5, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 249 References References
[240] E. Leclerc, K. S. Furukawa, F. Miyata, Y. Sakai, T. Ushida, and T. Fujii. Fab- rication of microstructures in photosensitive biodegradable polymers for tissue engineering applications. Biomaterials, 25(19):4683–90, 2004.
[241] E. Leclerc, B. David, L. Griscom, B. Lepioufle, T. Fujii, P. Layrolle, and C. Legal- laisa. Study of osteoblastic cells in a microfluidic environment. Biomaterials, 27(4):586–95, 2006.
[242] Y. C. Toh, C. Zhang, J. Zhang, Y. M. Khong, S. Chang, V. D. Samper, D. van Noort, D. W. Hutmacher, and H. Yu. A novel 3d mammalian cell perfusion- culture system in microfluidic channels. Lab Chip, 7(3):302–9, 2007.
[243] H. Mayani and P. M. Lansdorp. Biology of human umbilical cord blood-derived hematopoietic stem/progenitor cells. Stem Cells, 16(3):153–65, 1998.
[244] T. Nakahata and M. Ogawa. Hemopoietic colony-forming cells in umbilical cord blood with extensive capability to generate mono- and multipotential hemopoietic progenitors. J Clin Invest, 70(6):1324–8, 1982.
[245] H. E. Broxmeyer, G. W. Douglas, G. Hangoc, S. Cooper, J. Bard, D. English, M. Arny, L. Thomas, and E. A. Boyse. Human umbilical cord blood as a potential source of transplantable hematopoietic stem/progenitor cells. Proc Natl Acad Sci USA, 86(10):3828–32, 1989.
[246] H. Mayani, J. A. Alvarado-Moreno, and P. Flores-Guzman. Biology of human hematopoietic stem and progenitor cells present in circulation. Arch Med Res, 34(6):476–88, 2003.
[247] M. Gutierrez-Rodriguez, E. Reyes-Maldonado, and H. Mayani. Characterization of the adherent cells developed in dexter-type long-term cultures from human umbilical cord blood. Stem Cells, 18(1):46–52, 2000.
[248] J. M. Hows, B. A. Bradley, J. C. Marsh, T. Luft, L. Coutinho, N. G. Testa, and T. M. Dexter. Growth of human umbilical-cord blood in longterm haemopoietic cultures. Lancet, 340(8811):73–6, 1992.
[249] K. Mareschi, E. Biasin, W. Piacibello, M. Aglietta, E. Madon, and F. Fagioli. Isolation of human mesenchymal stem cells: bone marrow versus umbilical cord blood. Haematologica, 86(10):1099–100, 2001.
[250] H. Mayani, M. Gutierrez-Rodriguez, L. Espinoza, E. Lopez-Chalini, A. Huerta- Zepeda, E. Flores, E. Sanchez-Valle, F. Luna-Bautista, I. Valencia, and O. T. Ramirez. Kinetics of hematopoiesis in dexter-type long-term cultures established from human umbilical cord blood cells. Stem Cells, 16(2):127–35, 1998.
[251] S. A. Wexler, C. Donaldson, P. Denning-Kendall, C. Rice, B. Bradley, and J. M. Hows. Adult bone marrow is a rich source of human mesenchymal ’stem’ cells but umbilical cord and mobilized adult blood are not. Br J Haematol, 121(2):368–74, 2003.
[252] M. Yu, Z. Xiao, L. Shen, and L. Li. Mid-trimester fetal blood-derived adher- ent cells share characteristics similar to mesenchymal stem cells but full-term umbilical cord blood does not. Br J Haematol, 124(5):666–75, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 250 References References
[253] A. Erices, P. Conget, and J. J. Minguell. Mesenchymal progenitor cells in human umbilical cord blood. Br J Haematol, 109(1):235–42, 2000.
[254] E. J. Gang, S. H. Hong, J. A. Jeong, S. H. Hwang, S. W. Kim, I. H. Yang, C. Ahn, H. Han, and H. Kim. In vitro mesengenic potential of human umbilical cord blood-derived mesenchymal stem cells. Biochem Biophys Res Commun, 321(1):102–8, 2004.
[255] E. J. Gang, J. A. Jeong, S. H. Hong, S. H. Hwang, S. W. Kim, I. H. Yang, C. Ahn, H. Han, and H. Kim. Skeletal myogenic differentiation of mesenchymal stem cells isolated from human umbilical cord blood. Stem Cells, 22(4):617–24, 2004.
[256] H. S. Goodwin, A. R. Bicknese, S. N. Chien, B. D. Bogucki, C. O. Quinn, and D. A. Wall. Multilineage differentiation activity by cells isolated from umbilical cord blood: expression of bone, fat, and neural markers. Biol Blood Marrow Transplant, 7(11):581–8, 2001.
[257] J. A. Jeong, E. J. Gang, S. H. Hong, S. H. Hwang, S. W. Kim, I. H. Yang, C. Ahn, H. Han, and H. Kim. Rapid neural differentiation of human cord blood-derived mesenchymal stem cells. Neuroreport, 15(11):1731–4, 2004.
[258] T. J. Kang, J. E. Yeom, H. J. Lee, S. H. Rho, H. Han, and G. T. Chae. Growth kinetics of human mesenchymal stem cells from bone marrow and umbilical cord blood. Acta Haematol, 112(4):230–3, 2004.
[259] M. W. Lee, J. Choi, M. S. Yang, Y. J. Moon, J. S. Park, H. C. Kim, and Y. J. Kim. Mesenchymal stem cells from cryopreserved human umbilical cord blood. Biochem Biophys Res Commun, 320(1):273–8, 2004.
[260] O. K. Lee, T. K. Kuo, W. M. Chen, K. D. Lee, S. L. Hsieh, and T. H. Chen. Isolation of multipotent mesenchymal stem cells from umbilical cord blood. Blood, 103(5):1669–75, 2004.
[261] S. E. Yang, C. W. Ha, M. Jung, H. J. Jin, M. Lee, H. Song, S. Choi, W. Oh, and Y. S. Yang. Mesenchymal stem/progenitor cells developed in cultures from uc blood. Cytotherapy, 6(5):476–86, 2004.
[262] Z. Q. Ye, J. K. Burkholder, P. Qiu, J. C. Schultz, N. T. Shahidi, and N. S. Yang. Establishment of an adherent cell feeder layer from human umbilical cord blood for support of long-term hematopoietic progenitor cell growth. Proc Natl Acad Sci U S A, 91(25):12140–4, 1994.
[263] M. A. Goodell, K. Brose, G. Paradis, A. S. Conner, and R. C. Mulligan. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. JExpMed, 183(4):1797–806, 1996.
[264] R. J. Jones, J. P. Barber, M. S. Vala, M. I. Collector, S. H. Kaufmann, S. M. Ludeman, O. M. Colvin, and J. Hilton. Assessment of aldehyde dehydrogenase in viable cells. Blood, 85(10):2742–6, 1995.
[265] R. J. Jones, M. I. Collector, J. P. Barber, M. S. Vala, M. J. Fackler, W. S. May, C. A. Griffin, A. L. Hawkins, B. A. Zehnbauer, J. Hilton, O. M. Colvin, and S. J. Sharkis. Characterization of mouse lymphohematopoietic stem cells lacking spleen colony-forming activity. Blood, 88(2):487–91, 1996.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 251 References References
[266] V. Vasiliou and D. W. Nebert. Analysis and update of the human aldehyde dehydrogenase (aldh) gene family. Hum Genomics, 2(2):138–143, 2005.
[267] A. Yoshida, A. Rzhetsky, L. C. Hsu, and C. Chang. Human aldehyde dehydro- genase gene family. Eur J Biochem, 251(3):549–57, 1998.
[268] V. Vasiliou, A. Pappa, and D. R. Petersen. Role of aldehyde dehydrogenases in endogenous and xenobiotic metabolism. Chem Biol Interact, 129(1-2):1–19, 2000.
[269] L. Sreerama and N. E. Sladek. Cellular levels of class 1 and class 3 aldehyde dehydrogenases and certain other drug-metabolizing enzymes in human breast malignancies. Clin Cancer Res, 3(11):1901–14, 1997.
[270] R. W. Storms, A. P. Trujillo, J. B. Springer, L. Shah, O. M. Colvin, S. M. Ludeman, and C. Smith. Isolation of primitive human hematopoietic progenitors on the basis of aldehyde dehydrogenase activity. Proc Natl Acad Sci U S A, 96(16):9118–23, 1999.
[271] R. W. Storms, P. D. Green, K. M. Safford, D. Niedzwiecki, C. R. Cogle, O. M. Colvin, N. J. Chao, H. E. Rice, and C. A. Smith. Distinct hematopoietic progen- itor compartments are delineated by the expression of aldehyde dehydrogenase and cd34. Blood, 106(1):95–102, 2005.
[272] C. P. McGuckin, D. Pearce, N. Forraz, J. A. Tooze, S. M. Watt, and R. Pettengell. Multiparametric analysis of immature cell populations in umbilical cord blood and bone marrow. Eur J Haematol, 71(5):341–50, 2003.
[273] G. Gaipa, E. Coustan-Smith, E. Todisco, O. Maglia, A. Biondi, and D. Campana. Characterization of cd34+, cd13+, cd33- cells, a rare subset of immature human hematopoietic cells. Haematologica, 87(4):347–56, 2002.
[274] A. M. de Kreuk, A. Zevenbergen, J. W. van Oostveen, G. J. Schuurhuis, P. C. Huijgens, and A. R. Jonkhoff. A single-step colony-forming unit assay for unsep- arated mobilized peripheral blood, cord blood, and bone marrow. J Hematother Stem Cell Res, 10(6):795–806, 2001.
[275] I. Martin, D. Wendt, and M. Heberer. The role of bioreactors in tissue engineering. Trends Biotechnol, 22(2):80–6, 2004.
[276] O. Ringden, M. Uzunel, I. Rasmusson, M. Remberger, B. Sundberg, H. Lon- nies, H. U. Marschall, A. Dlugosz, A. Szakos, Z. Hassan, B. Omazic, J. Aschan, L. Barkholt, and K. Le Blanc. Mesenchymal stem cells for treatment of therapy- resistant graft-versus-host disease. Transplantation, 81(10):1390–7, 2006.
[277] H. M. Lazarus, O. N. Koc, S. M. Devine, P. Curtin, R. T. Maziarz, H. K. Holland, E. J. Shpall, P. McCarthy, K. Atkinson, B. W. Cooper, S. L. Gerson, M. J. Laugh- lin, Jr. Loberiza, F. R., A. B. Moseley, and A. Bacigalupo. Cotransplantation of hla-identical sibling culture-expanded mesenchymal stem cells and hematopoietic stem cells in hematologic malignancy patients. Biol Blood Marrow Transplant, 11(5):389–98, 2005.
[278] C. Bartmann, E. Rohde, K. Schallmoser, P. Purstner, G. Lanzer, W. Linkesch, and D. Strunk. Two steps to functional mesenchymal stromal cells for clinical application. Transfusion, 47(8):1426–35, 2007.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 252 References References
[279] J. Heino and J. Kapyla. Cellular receptors of extracellular matrix molecules. Curr Pharm Des, 15(12):1309–17, 2009.
[280] S. K. Akiyama. Integrins in cell adhesion and signaling. Hum Cell, 9(3):181–6, 1996.
[281] M. Leiss, K. Beckmann, A. Giros, M. Costell, and R. Fassler. The role of inte- grin binding sites in fibronectin matrix assembly in vivo. Curr Opin Cell Biol, 20(5):502–7, 2008.
[282] M. A. Schwartz and D. W. DeSimone. Cell adhesion receptors in mechanotrans- duction. Curr Opin Cell Biol, 20(5):551–6, 2008.
[283] E. A. Clark and J. S. Brugge. Integrins and signal transduction pathways: the road taken. Science, 268(5208):233–9, 1995.
[284] R. E. Nordon, B. K. Milthorpe, P. R. Slowiaczek, and K. Schindhelm. Cell separation device, 1998.
[285] M. Korhonen. Culture of human mesenchymal stem cells in serum-free conditions: no breakthroughs yet. Eur J Haematol, 78(2):167; author reply 168, 2007.
[286] N. Meuleman, T. Tondreau, A. Delforge, M. Dejeneffe, M. Massy, M. Liber- talis, D. Bron, and L. Lagneaux. Human marrow mesenchymal stem cell culture: serum-free medium allows better expansion than classical alpha-mem medium. Eur J Haematol, 76(4):309–16, 2006.
[287] N. Mizuno, H. Shiba, Y. Ozeki, Y. Mouri, M. Niitani, T. Inui, H. Hayashi, K. Suzuki, S. Tanaka, H. Kawaguchi, and H. Kurihara. Human autologous serum obtained using a completely closed bag system as a substitute for foetal calf serum in human mesenchymal stem cell cultures. Cell Biol Int, 30(6):521–4, 2006.
[288] C. Lange, F. Cakiroglu, A. N. Spiess, H. Cappallo-Obermann, J. Dierlamm, and A. R. Zander. Accelerated and safe expansion of human mesenchymal stromal cells in animal serum-free medium for transplantation and regenerative medicine. J Cell Physiol, 213(1):18–26, 2007.
[289] J. S. Temenoff and A. G. Mikos. Review: tissue engineering for regeneration of articular cartilage. Biomaterials, 21(5):431–40, 2000.
[290] A. Getgood, R. Brooks, L. Fortier, and N. Rushton. Articular cartilage tis- sue engineering: today’s research, tomorrow’s practice? JBoneJointSurgBr, 91(5):565–76, 2009.
[291] S. N. Redman, S. F. Oldfield, and C. W. Archer. Current strategies for articular cartilage repair. Eur Cell Mater, 9:23–32; discussion 23–32, 2005.
[292] P. J. Livesley, M. Doherty, M. Needoff, and A. Moulton. Arthroscopic lavage of osteoarthritic knees. JBoneJointSurgBr, 73(6):922–6, 1991.
[293] H. K. Kim, M. E. Moran, and R. B. Salter. The potential for regeneration of ar- ticular cartilage in defects created by chondral shaving and subchondral abrasion. an experimental investigation in rabbits. J Bone Joint Surg Am, 73(9):1301–15, 1991.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 253 References References
[294] H. D. Clarke and W. N. Scott. The role of debridement: through small portals. J Arthroplasty, 18(3 Suppl 1):10–3, 2003. [295] L. L. Johnson. Arthroscopic abrasion arthroplasty: a review. Clin Orthop, (391 Suppl):S306–17, 2001. [296] B. Muller and D. Kohn. [indication for and performance of articular cartilage drilling using the pridie method]. Orthopade, 28(1):4–10, 1999. [297] J. R. Steadman, W. G. Rodkey, and J. J. Rodrigo. Microfracture: surgical tech- nique and rehabilitation to treat chondral defects. Clin Orthop, (391 Suppl):S362– 9, 2001. [298] S. W. O’Driscoll. Articular cartilage regeneration using periosteum. Clin Orthop Relat Res, (367 Suppl):S186–203, 1999. [299] L. Hangody and P. Fules. Autologous osteochondral mosaicplasty for the treat- ment of full-thickness defects of weight-bearing joints: ten years of experimental and clinical experience. JBoneJointSurgAm, 85-A Suppl 2:25–32, 2003. [300] M. Brittberg, A. Lindahl, A. Nilsson, C. Ohlsson, O. Isaksson, and L. Peterson. Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. NEnglJMed, 331(14):889–95, 1994. [301] M. Brittberg, T. Tallheden, B. Sjogren-Jansson, A. Lindahl, and L. Peterson. Autologous chondrocytes used for articular cartilage repair: an update. Clin Orthop, (391 Suppl):S337–48, 2001. [302] E. B. Hunziker. Articular cartilage repair: basic science and clinical progress. a review of the current status and prospects. Osteoarthritis Cartilage, 10(6):432–63, 2002. [303] D. W. Hutmacher. Scaffolds in tissue engineering bone and cartilage. Biomate- rials, 21(24):2529–43, 2000. [304] R. Stoop. Smart biomaterials for tissue engineering of cartilage. Injury, 39 Suppl 1:S77–87, 2008. [305] B. P. Chan and K. W. Leong. Scaffolding in tissue engineering: general ap- proaches and tissue-specific considerations. Eur Spine J, 17 Suppl 4:467–79, 2008. [306] J. M. Oliveira, M. T. Rodrigues, S. S. Silva, P. B. Malafaya, M. E. Gomes, C. A. Viegas, I. R. Dias, J. T. Azevedo, J. F. Mano, and R. L. Reis. Novel hydroxyapatite/chitosan bilayered scaffold for osteochondral tissue-engineering applications: Scaffold design and its performance when seeded with goat bone marrow stromal cells. Biomaterials, 27(36):6123–37, 2006. [307] M. Keeney and A. Pandit. The osteochondral junction and its repair via bi-phasic tissue engineering scaffolds. Tissue Eng Part B Rev, 15(1):55–73, 2009. [308] T. M. O’Shea and X. Miao. Bilayered scaffolds for osteochondral tissue engineer- ing. Tissue Eng Part B Rev, 14(4):447–64, 2008. [309] A. Alhadlaq, J. H. Elisseeff, L. Hong, C. G. Williams, A. I. Caplan, B. Sharma, R. A. Kopher, S. Tomkoria, D. P. Lennon, A. Lopez, and J. J. Mao. Adult stem cell driven genesis of human-shaped articular condyle. Ann Biomed Eng, 32(7):911–23, 2004.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 254 References References
[310] A. Alhadlaq and J. J. Mao. Tissue-engineered osteochondral constructs in the shape of an articular condyle. JBoneJointSurgAm, 87(5):936–44, 2005.
[311] P. Angele, R. Kujat, M. Nerlich, J. Yoo, V. Goldberg, and B. Johnstone. Engi- neering of osteochondral tissue with bone marrow mesenchymal progenitor cells in a derivatized hyaluronan-gelatin composite sponge. Tissue Eng, 5(6):545–54, 1999.
[312] J. Gao, J. E. Dennis, L. A. Solchaga, A. S. Awadallah, V. M. Goldberg, and A. I. Caplan. Tissue-engineered fabrication of an osteochondral composite graft using rat bone marrow-derived mesenchymal stem cells. Tissue Eng, 7(4):363–71, 2001.
[313] J. Gao, J. E. Dennis, L. A. Solchaga, V. M. Goldberg, and A. I. Caplan. Repair of osteochondral defect with tissue-engineered two-phase composite material of injectable calcium phosphate and hyaluronan sponge. Tissue Eng, 8(5):827–37, 2002.
[314] A. Tampieri, M. Sandri, E. Landi, D. Pressato, S. Francioli, R. Quarto, and I. Martin. Design of graded biomimetic osteochondral composite scaffolds. Bio- materials, 29(26):3539–46, 2008.
[315] R. Tuli, S. Nandi, W. J. Li, S. Tuli, X. Huang, P. A. Manner, P. Laquerriere, U. Noth, D. J. Hall, and R. S. Tuan. Human mesenchymal progenitor cell-based tissue engineering of a single-unit osteochondral construct. Tissue Eng, 10(7- 8):1169–79, 2004.
[316] A. Muller, B. Homey, H. Soto, N. Ge, D. Catron, M. E. Buchanan, T. McClana- han, E. Murphy, W. Yuan, S. N. Wagner, J. L. Barrera, A. Mohar, E. Verastegui, and A. Zlotnik. Involvement of chemokine receptors in breast cancer metastasis. Nature, 410(6824):50–6, 2001.
[317] T. N. Behar, A. E. Schaffner, C. A. Colton, R. Somogyi, Z. Olah, C. Lehel, and J. L. Barker. Gaba-induced chemokinesis and ngf-induced chemotaxis of embryonic spinal cord neurons. J Neurosci, 14(1):29–38, 1994.
[318] P. C. Wilkinson and J. M. Lackie. The adhesion, migration and chemotaxis of leucocytes in inflammation. Curr Top Pathol, 68:47–88, 1979.
[319] W. L. Grayson, P. H. Chao, D. Marolt, D. L. Kaplan, and G. Vunjak-Novakovic. Engineering custom-designed osteochondral tissue grafts. Trends Biotechnol, 26(4):181–9, 2008.
[320] G. Vunjak-Novakovic, L. Meinel, G. Altman, and D. Kaplan. Bioreactor cultiva- tion of osteochondral grafts. Orthod Craniofac Res, 8(3):209–18, 2005.
[321] D. Wendt, M. Jakob, and I. Martin. Bioreactor-based engineering of osteo- chondral grafts: from model systems to tissue manufacturing. J Biosci Bioeng, 100(5):489–94, 2005.
[322] J. F. Mano and R. L. Reis. Osteochondral defects: present situation and tissue engineering approaches. J Tissue Eng Regen Med, 1(4):261–73, 2007.
[323] C. Haasper, J. Zeichen, R. Meister, C. Krettek, and M. Jagodzinski. Tissue engineering of osteochondral constructs in vitro using bioreactors. Injury,39 Suppl 1:S66–76, 2008.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 255 References References
[324] C. H. Chang, C. C. Lin, C. H. Chou, F. H. Lin, and H. C. Liu. Novel bioreac- tors for osteochondral tissue engineering. Biomedical Engineering - Applications, Basis and Communications, 17(1):38–43, 2005. [325] P. B. Malafaya and R. L. Reis. Bilayered chitosan-based scaffolds for osteochon- dral tissue engineering: Influence of hydroxyapatite on in vitro cytotoxicity and dynamic bioactivity studies in a specific double-chamber bioreactor. Acta Bio- mater, 2008. [326] I. M. Sauer, R. Schwartlander, J. Schmid, E. Efimova, F. W. Vondran, D. Kehr, G. Pless, A. Spinelli, B. Brandenburg, E. Hildt, and P. Neuhaus. The slidereactor– a simple hollow fiber based bioreactor suitable for light microscopy. Artif Organs, 29(3):264–7, 2005. [327] R. Schwartlander, J. Schmid, B. Brandenburg, E. Katenz, F. W. Vondran, G. Pless, X. Cheng, A. Pascher, P. Neuhaus, and I. M. Sauer. Continuously microscopically observed and process-controlled cell culture within the slidereac- tor: proof of a new concept for cell characterization. Tissue Eng, 13(1):187–96, 2007. [328] S. A. Vanapalli, M. H. G. Duits, and F. Mugele. Microfluidics as a functional tool for cell mechanics. Biomicrofluidics, 3(1):012006, 2009. [329] H. Andersson and A. van den Berg. Microfabrication and microfluidics for tissue engineering: state of the art and future opportunities. Lab Chip, 4(2):98–103, 2004. [330] S. W. Rhee, A. M. Taylor, C. H. Tu, D. H. Cribbs, C. W. Cotman, and N. L. Jeon. Patterned cell culture inside microfluidic devices. Lab Chip, 5(1):102–7, 2005. [331] G. M. Walker, H. C. Zeringue, and D. J. Beebe. Microenvironment design con- siderations for cellular scale studies. Lab Chip, 4(2):91–7, 2004. [332] G. M. Whitesides, E. Ostuni, S. Takayama, X. Jiang, and D. E. Ingber. Soft lithography in biology and biochemistry. Annu Rev Biomed Eng, 3:335–73, 2001. [333] P. J. Lee, P. J. Hung, V. M. Rao, and L. P. Lee. Nanoliter scale microbioreactor array for quantitative cell biology. Biotechnol Bioeng, 94(1):5–14, 2006. [334] J. El-Ali, P. K. Sorger, and K. F. Jensen. Cells on chips. Nature, 442(7101):403– 11, 2006. [335] E. T. Peterson and I. Papautsky. Microtextured polydimethylsiloxane substrates for culturing mesenchymal stem cells. Methods Mol Biol, 321:179–97, 2006. [336] T. C. Merkel, V. I. Bondar, K. Nagai, B. D. Freeman, and I. Pinnau. Gas sorption, diffusion, and permeation in poly(dimethylsiloxane). J Polym Sci, 38:415–434, 2000. [337] A. Piruska, I. Nikcevic, S. H. Lee, C. Ahn, W. R. Heineman, P. A. Limbach, and C. J. Seliskar. The autofluorescence of plastic materials and chips measured under laser irradiation. Lab Chip, 5(12):1348–54, 2005. [338] D. N. Breslauer, P. J. Lee, and L. P. Lee. Microfluidics-based systems biology. Mol Biosyst, 2(2):97–112, 2006.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 256 References References
[339] P. J. Lee, T. A. Gaige, and P. J. Hung. Dynamic cell culture: a microfluidic function generator for live cell microscopy. Lab Chip, 9(1):164–6, 2009.
[340] P. J. Lee, T. A. Gaige, N. Ghorashian, and P. J. Hung. Microfluidic tissue model for live cell screening. Biotechnol Prog, 2007.
[341] N. Futai, W. Gu, J. W. Song, and S. Takayama. Handheld recirculation system and customized media for microfluidic cell culture. Lab Chip, 6(1):149–54, 2006.
[342] A. M. Taylor, M. Blurton-Jones, S. W. Rhee, D. H. Cribbs, C. W. Cotman, and N. L. Jeon. A microfluidic culture platform for cns axonal injury, regeneration and transport. Nat Methods, 2(8):599–605, 2005.
[343] C. C. Hsieh, S. B. Huang, P. C. Wu, D. B. Shieh, and G. B. Lee. A microfluidic cell culture platform for real-time cellular imaging. Biomed Microdevices, 2009.
[344] A. Tourovskaia, X. Figueroa-Masot, and A. Folch. Differentiation-on-a-chip: a microfluidic platform for long-term cell culture studies. Lab Chip, 5(1):14–9, 2005.
[345] A. Prokop, Z. Prokop, D. Schaffer, E. Kozlov, J. Wikswo, D. Cliffel, and F. Bau- denbacher. Nanoliterbioreactor: long-term mammalian cell culture at nanofabri- cated scale. Biomed Microdevices, 6(4):325–39, 2004.
[346] B. J. Kane, M. J. Zinner, M. L. Yarmush, and M. Toner. Liver-specific functional studies in a microfluidic array of primary mammalian hepatocytes. Anal Chem, 78(13):4291–8, 2006.
[347] P. J. Hung, P. J. Lee, P. Sabounchi, N. Aghdam, R. Lin, and L. P. Lee. A novel high aspect ratio microfluidic design to provide a stable and uniform microenvi- ronment for cell growth in a high throughput mammalian cell culture array. Lab Chip, 5(1):44–8, 2005.
[348] W. Gu, X. Zhu, N. Futai, B. S. Cho, and S. Takayama. Computerized microfluidic cell culture using elastomeric channels and braille displays. Proc Natl Acad Sci USA, 101(45):15861–6, 2004.
[349] J. W. Song, W. Gu, N. Futai, K. A. Warner, J. E. Nor, and S. Takayama. Computer-controlled microcirculatory support system for endothelial cell culture and shearing. Anal Chem, 77(13):3993–9, 2005.
[350] N. Li Jeon, H. Baskaran, S. K. Dertinger, G. M. Whitesides, L. Van de Water, and M. Toner. Neutrophil chemotaxis in linear and complex gradients of interleukin-8 formed in a microfabricated device. Nat Biotechnol, 20(8):826–30, 2002.
[351] B. G. Chung, L. A. Flanagan, S. W. Rhee, P. H. Schwartz, A. P. Lee, E. S. Monuki, and N. L. Jeon. Human neural stem cell growth and differentiation in a gradient-generating microfluidic device. Lab Chip, 5(4):401–6, 2005.
[352] P. J. Hung, P. J. Lee, P. Sabounchi, R. Lin, and L. P. Lee. Continuous perfusion microfluidic cell culture array for high-throughput cell-based assays. Biotechnol Bioeng, 89(1):1–8, 2005.
[353] V. Dhir, A. Natarajan, M. Stancescu, A. Chunder, N. Bhargava, M. Das, L. Zhai, and P. Molnar. Patterning of diverse mammalian cell types in serum free medium with photoablation. Biotechnol Prog, 25(2):594–603, 2009.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 257 References References
[354] K. S. Yun and E. Yoon. Micro/nanofluidic device for single-cell-based assay. Biomed Microdevices, 7(1):35–40, 2005.
[355] D. Falconnet, G. Csucs, H. M. Grandin, and M. Textor. Surface engineer- ing approaches to micropattern surfaces for cell-based assays. Biomaterials, 27(16):3044–63, 2006.
[356] H. Lu, L. Y. Koo, W. M. Wang, D. A. Lauffenburger, L. G. Griffith, and K. F. Jensen. Microfluidic shear devices for quantitative analysis of cell adhesion. Anal Chem, 76(18):5257–64, 2004.
[357] M. C. Potter and D. C. Wiggert. Mechanics of Fluids. Prentice Hall, Upper Saddle River, 1997.
[358] M.E. O’Neill. Sphere in contact with a plane wall in a slow linear shear flow. Chemical Engineering Science, 23:1293–1298, 1968.
[359] D. J. Beebe, G. A. Mensing, and G. M. Walker. Physics and applications of microfluidics in biology. Annu Rev Biomed Eng, 4:261–86, 2002.
[360] L. A. Hines and R. N. Maddox. Mass Transfer Fundamentals and Applications. Prentice Hall International Series in the Physical and Chemical Engineering Sci- ences. Prentice Hall, Upper Saddle River, 1985.
[361] J. P. Kutter and H. Klank. Microfluidics - theoretical aspects. In O. Geschke, H. Klank, and P. Telleman, editors, Microsystem Engineering of Lab-on-a-chip Devices, pages 13–37. Wiley, Weinheim, 2008.
[362] H. Xu and M. L. Shuler. Quantification of chemical-polymer surface interactions in microfluidic cell culture devices. Biotechnology Progress, 25(2):543–551, 2009.
[363] M. S. Kim, J. H. Yeon, and J. K. Park. A microfluidic platform for 3-dimensional cell culture and cell-based assays. Biomed Microdevices, 9(1):25–34, 2007.
[364] F. Lin, W. Saadi, S. W. Rhee, S. J. Wang, S. Mittal, and N. L. Jeon. Generation of dynamic temporal and spatial concentration gradients using microfluidic devices. Lab Chip, 4(3):164–7, 2004.
[365] W. Saadi, S. W. Rhee, F. Lin, B. Vahidi, B. G. Chung, and N. L. Jeon. Generation of stable concentration gradients in 2d and 3d environments using a microfluidic ladder chamber. Biomed Microdevices, 9(5):627–35, 2007.
[366] C. W. Li, R. Chen, and M. Yang. Generation of linear and non-linear concen- tration gradients along microfluidic channel by microtunnel controlled stepwise addition of sample solution. Lab Chip, 7(10):1371–3, 2007.
[367] L. Kim, M. D. Vahey, H. Y. Lee, and J. Voldman. Microfluidic arrays for loga- rithmically perfused embryonic stem cell culture. Lab Chip, 6(3):394–406, 2006.
[368] B. G. Chung, A. Manbachi, W. Saadi, F. Lin, N. L. Jeon, and A. Khademhosseini. A gradient-generating microfluidic device for cell biology. JVisExp, (7):271, 2007.
[369] N. Li Jeon, S. K. Dertinger, D. T. Chiu, I. S. Choi, A. D. Stroock, and G. M. Whitesides. Generation of solution and surface gradients using microfluidic sys- tems. Langmuir, 16(22):8311–8316, 2000.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 258 References References
[370] K. Haubert, T. Drier, and D. Beebe. PDMS bonding by means of a portable, low-cost corona system. Lab Chip, 6(12):1548–1549, 2006.
[371] S. K. Dertinger, D. T. Chiu, N. Li Jeon, and G. M. Whitesides. Generation of gradients having complex shapes using microfluidic networks. Anal Chem, 73(6):1240–1246, 2001.
[372] R. E. Nordon. Uniform shear elution affinity cell separation. PhD thesis, Uni- versity of New South Wales, 1994.
[373] Sigma-Aldrich. Product information fluorescein isothiocyanate-dextran. Techni- cal report, 1997.
[374] A. Khademhosseini, R. Langer, J. Borenstein, and J. P. Vacanti. Microscale technologies for tissue engineering and biology. Proc Natl Acad Sci U S A, 103(8):2480–7, 2006.
[375] J. G. Shackman, G. M. Dahlgren, J. L. Peters, and R. T. Kennedy. Perfusion and chemical monitoring of living cells on a microfluidic chip. Lab Chip, 5(1):56–63, 2005.
[376] D. M. Thompson, K. R. King, K. J. Wieder, M. Toner, M. L. Yarmush, and A. Jayaraman. Dynamic gene expression profiling using a microfabricated living cell array. Anal Chem, 76(14):4098–103, 2004.
[377] B. L. Gray, D.K. Liew, S.D. Collins, R.L. Smith, and A.I. Barakat. Microchan- nel platform for the study of endothelial cell shape and function. Biomedical Microdevices, 4(1):9–16, 2002.
[378] E. Tkachenko, E. Gutierrez, M. H. Ginsberg, and A. Groisman. An easy to assem- ble microfluidic perfusion device with a magnetic clamp. Lab Chip, 9(8):1085–95, 2009.
[379] M. J. Powers, D. M. Janigian, K. E. Wack, C. S. Baker, D. Beer Stolz, and L. G. Griffith. Functional behavior of primary rat liver cells in a three-dimensional perfused microarray bioreactor. Tissue Eng, 8(3):499–513, 2002.
[380] M. Y. Zhang, P. J. Lee, P. J. Hung, T. Johnson, L. P. Lee, and M. R. Mofrad. Mi- crofluidic environment for high density hepatocyte culture. Biomed Microdevices, 10(1):117–21, 2008.
[381] M. Shin, K. Matsuda, O. Ishii, H. Terai, M. Kaazempur-Mofrad, J. Borenstein, M. Detmar, and J.P. Vacanti. Endothelialized networks with a vascular geometry in microfabricated poly(dimethyl siloxane). Biomedical Microdevices, 6(4):269– 278, 2004.
[382] Y. Tanaka, K. Sato, M. Yamato, T. Okano, and T. Kitamori. Cell culture and life support system for microbioreactor and bioassay. J Chromatogr A, 1111(2):233–7, 2006.
[383] S. Viswanathan, T. Benatar, M. Mileikovsky, D. A. Lauffenburger, A. Nagy, and P. W. Zandstra. Supplementation-dependent differences in the rates of embry- onic stem cell self-renewal, differentiation, and apoptosis. Biotechnol Bioeng, 84(5):505–17, 2003.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 259 References References
[384] W. Prudhomme, G. Q. Daley, P. Zandstra, and D. A. Lauffenburger. Multivariate proteomic analysis of murine embryonic stem cell self-renewal versus differentia- tion signaling. Proc Natl Acad Sci U S A, 101(9):2900–5, 2004. [385] X. Ju, D. Li, N. Gao, Q. Shi, and H. Hou. Hepatogenic differentiation of mes- enchymal stem cells using microfluidic chips. Biotechnol J, 3(3):383–91, 2008. [386] J. W. Allen, S. R. Khetani, and S. N. Bhatia. In vitro zonation and toxicity in a hepatocyte bioreactor. Toxicol Sci, 84(1):110–9, 2005. [387] A. W. Tilles, H. Baskaran, P. Roy, M. L. Yarmush, and M. Toner. Effects of oxygenation and flow on the viability and function of rat hepatocytes cocultured in a microchannel flat-plate bioreactor. Biotechnol Bioeng, 73(5):379–89, 2001. [388] M. Stangegaard, S. Petronis, A. M. Jorgensen, C. B. Christensen, and M. Dufva. A biocompatible micro cell culture chamber (microccc) for the culturing and on-line monitoring of eukaryote cells. Lab Chip, 6(8):1045–51, 2006. [389] K. Viravaidya and M. L. Shuler. Incorporation of 3t3-l1 cells to mimic bioaccu- mulation in a microscale cell culture analog device for toxicity studies. Biotechnol Prog, 20(2):590–7, 2004. [390] Y. Tanaka, M. Yamato, T. Okano, T. Kitamori, and K. Sato. Evaluation of effects of shear stress on hepatocytes by a microchip-based system. Measurement Science and Technology, 17:3167–3170, 2006. [391] S. M. Ong, C. Zhang, Y. C. Toh, S. H. Kim, H. L. Foo, C. H. Tan, D. van Noort, S. Park, and H. Yu. A gel-free 3d microfluidic cell culture system. Biomaterials, 29(22):3237–44, 2008. [392] K. Liu, R. Pitchimani, D. Dang, K. Bayer, T. Harrington, and D. Pappas. Cell culture chip using low-shear mass transport. Langmuir, 24(11):5955–60, 2008. [393] J. Lii, W. J. Hsu, H. Parsa, A. Das, R. Rouse, and S. K. Sia. Real-time microflu- idic system for studying mammalian cells in 3d microenvironments. Anal Chem, 80(10):3640–7, 2008. [394] V. Vickerman, J. Blundo, S. Chung, and R. Kamm. Design, fabrication and implementation of a novel multi-parameter control microfluidic platform for three- dimensional cell culture and real-time imaging. Lab Chip, 8(9):1468–77, 2008. [395] M. J. Powers, K. Domansky, M. R. Kaazempur-Mofrad, A. Kalezi, A. Capitano, A. Upadhyaya, P. Kurzawski, K. E. Wack, D. B. Stolz, R. Kamm, and L. G. Grif- fith. A microfabricated array bioreactor for perfused 3d liver culture. Biotechnol Bioeng, 78(3):257–69, 2002. [396] P. Roy, H. Baskaran, A. W. Tilles, M. L. Yarmush, and M. Toner. Analysis of oxygen transport to hepatocytes in a flat-plate microchannel bioreactor. Ann Biomed Eng, 29(11):947–55, 2001. [397] M. H. Wu, S. B. Huang, Z. Cui, Z. Cui, and G. B. Lee. A high throughput perfusion-based microbioreactor platform integrated with pneumatic micropumps for three-dimensional cell culture. Biomed Microdevices, 10(2):309–19, 2008. [398] Dow-Corning. Silastic laboratory tubing, reference number 51-0924e-01. Techni- cal report, 2005.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 260 References References
[399] A. Colas, R. Malczewski, and K. Ulman. Silicone tubing for pharamceutical processing. Technical report, Dow Corning, Life Sciences, 2004.
[400] R. C. Meagher, A. J. Salvado, and D. G. Wright. An analysis of the multilin- eage production of human hematopoietic progenitors in long-term bone marrow culture: evidence that reactive oxygen intermediates derived from mature phago- cytic cells have a role in limiting progenitor cell self-renewal. Blood, 72(1):273–81, 1988.
[401] M. Iwasaki, A. X. Le, and J. A. Helms. Expression of indian hedgehog, bone morphogenetic protein 6 and gli during skeletal morphogenesis. Mech Dev, 69(1- 2):197–202, 1997.
[402] H. K. Kleinman, D. H. Rorbach, V. P. Terranova, H. H. Varner, A. T. Hewitt, C. M. Grotenddorst, C. M. Wilkes, G. R. Martin, H. Seppa, and E. Schiffmann. Collagenous matrices as determinants of cell function. In H. Furthmayr, editor, Immunochemistry of the extracellular matrix, pages 151–173. CRC Press, Boca Raton, 1982.
[403] S. Bazin and A. Delaunay. Preparation of acid and citrate soluble collagen. In D. A. Hall, editor, The methodology of connective tissue research, pages 13–17. Joynson and Bruvvers, Oxford, 1976.
[404] Z. Deyl and M. Adam. Preparation of insoluble collagen. In D. A. Hall, editor, The methodology of connective tissue research, pages 1–17. Joynson and Bruvvers, Oxford, 1976.
[405] A. M. Fielding. Preparation of neutral salt soluble collagen. In D. A. Hall, editor, The methodology of connective tissue research, pages 9–12. Joynson and Bruvvers, Oxford, 1976.
[406] R. Reed. The reconstitution of native-type fibres from solutions of collagen. In D. A. Hall, editor, The methodology of connective tissue research, pages 53–61. Joynson and Bruvvers, Oxford, 1976.
[407] F. S. Steven. Preparation of macromolecular collagens. In D. A. Hall, editor, The methodology of connective tissue research, pages 19–27. Joynson and Bruvvers, Oxford, 1976.
[408] C. Modin, A. L. Stranne, M. Foss, M. Duch, J. Justesen, J. Chevallier, L. K. Andersen, A. G. Hemmersam, F. S. Pedersen, and F. Besenbacher. Qcm-d studies of attachment and differential spreading of pre-osteoblastic cells on ta and cr surfaces. Biomaterials, 27(8):1346–54, 2006.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 261 Appendix A
Materials and General Methods
A.1 Materials
A.1.1 Cell Culture Materials and Reagents
Dulbecco’s Phosphate Buffered Saline, without calcium chloride and magnesium chlo- ride (DPBS - CaMg, D5652-50L), Minimum Essential Medium Eagle (EMEM, M-0643),
N-[2-Hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid] (HEPES, H3375 - 500G), NaHCO3 (S-5761 - 500G), trypan blue (T8154-100ML), dexamethasone (D2915-100MG), glyc- erol phosphate calcium salt (G6626-100G), fluorescein isothiocyanate-dextran (average molecular weight 150,000, FITC, FD150S), albumin from bovine serum (BSA, A7906-
500g), pepsin from porcine gastric mucosa (P7000), and dialysis membrane (cellulose membrane, D-9652) were all obtained from Sigma-Aldrich, Sydney Australia. Trypsin-
EDTA (59430C-100ML) was obtained from SAFC Biosciences through Sigma-Aldrich,
Sydney Australia, and penicillin G 5000U/mL streptomycin sulphate 5000mcg/mL
(P/S, 59620-100ml) was obtained from JRH Biosciences through Sigma-Aldrich, Sydney
Australia. Fetal bovine serum (FBS, 10099-141), Dulbecco’s Modified Eagle Medium
(D-MEM), powder (low glucose, 31100-035; high glucose, 12100-046), Medium 199, powder (M199, 31600-034), Iscove’s Modified Dulbecco’s Medium, powder (IMDM,
12200-028) and Minimum Essential Medium - Alpha Medium, powder (α-MEM, 11900-
262 Appendix A. Materials and General Methods A.1. Materials
016) were obtained from Gibco Invitrogen Corporation, Melbourne Australia. Recom- binant TGF-β1 (PHP134) was obtained from Serotec, through Australian Laboratory
Services, Sydney Australia, and ascorbic acid-2-phosphate magnesium salt (358610050) from Acros Organics, through Ajax Finechem, Sydney Australia. Water for Irriga- tion (WFI, AHF7114) was obtained from Baxter, Sydney Australia. 0.22μm syringe driven filter units (Millex GP SLGP033RS), 0.22μm large filter units (Sterivex GP
SVGPB1010) and triple distilled H2O were obtained from Millipore, MA USA. Fresh kangaroo tails were obtained from Southern Game Meat, Sydney Australia, and were frozen and stored at -20℃ until required. Both sodium chloride (NaCl, A465-5KG) and sodium hydroxide (NaOH, A482-500G) were obtained from Ajax Finechem, Sydney
Australia. Tris-hydrochloride (Tris-HCl, 816100) was obtained from MP Biomedicals,
Sydney Australia.
Glacial acetic acid (HAc, 100015N), hydrochloric acid (HCl, A1367-2.5L GL) and
Dimethyl sulphoxide (DMSO, 10323) AnalR were obtained from BDH. Ethanol ab- solute extra pure (4.10050.0500) was obtained from Merck, Melbourne Australia.
Murine stroma 5 (MS-5) cell line was donated by Dr. Alla Dolnikov from the Chil- dren’s Cancer Institute Australia, Sydney Australia. Mouse embryonic fibroblast cell line, NIH 3T3 (eGFP), was kindly provided by iNano, University of Aarhus, Aarhus,
Denmark.
A.1.2 Surgical
Sodium chloride injection BP 0.9%, 90mg in 10mL, was obtained from Pharmacia &
Upjohn, Sydney Australia; 10% w/v povidone-iodine (Betadine) from Faulding Phar- maceuticals, Adelaide Australia; heparin sodium (porcine mucous) 5000I.U./mL (Hep- arin, 463733) from Mayne Pharma, Melbourne Australia; ketamine hydrochloride (Ke- tamine) 100mg/mL from Parnell Laboratories, Sydney Australia; and pentobarbitone sodium (Lethabarb) 325mg/mL from Virbac, Sydney Australia. Hypodermic needles and SurFlo IV catheters were obtained from Terumo Medical, Sydney Australia; and surgical blades from Swann-Morton, through Smith And Nephew, Melbourne Australia.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 263 Appendix A. Materials and General Methods A.2. Methods
Cordless drill (BD1020AR) was obtained from, Ryobi, Tokyo Japan.
A.1.3 Histology
Crystal violet (9700) GURR, silver nitrate (10233) AnalaR, and xylene (10293.6H)
AnalaR were obtained from BDH. Alkaline phosphatase (86C-1KT) was obtained from
Sigma-Aldrich, Sydney Australia, and Sodium Thiosulphate (J87) from Ajax Finechem,
Sydney Australia. 10% Formalin solution, and 3% agarose solution were obtained from the Histology and Microscopy Unit, UNSW Sydney Australia. Mountant (7961) was obtained from Merck, Melbourne Australia.
A.1.4 Equipment
For all tissue culture a Heraeus Labofuge 400R with rotor no.8179 and buckets no.8172 was used. For collagen preparation a Beckman Coulter™ ultracentrifuge, Avanti J-
20XP Series fixed-angle rotor, and JLA-8.1000 canisters were used. Collagen was dried on a Labconco, MO USA, freeze drier, and pH was measured used a TPS 901-PH pH-mV-Temperature meter, Brisbane Australia.
A.2 Methods
A.2.1 Cell Culture Reagent Preparation
Unless otherwise stated, EMEM was used for all cell culture and pH was adjusted to between 7.2 and 7.4 with 1M NaOH or 1M HCl. Media were sterilised through a 0.22μm
filter and frozen in aliquots at -20℃ until required.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 264 Appendix A. Materials and General Methods A.2. Methods
A.2.1.1 Basal Media Preparation
To prepare EMEM, 3.6 L of WFI was measured into a 4L beaker. 34.9g of EMEM powder, 19.04g Hepes and 3g NaHCO3 was added and stirred covered until all the powders had dissolved. pH was measured and adjusted with 1M NaCl or 1M HCl solution to 7.30. 180mL aliquots were defrosted and 2mL of penicillin streptomycin added, to give a concentration of 1% (50U/mL penicillin, 50μg/mL streptomycin) and either 20 or 40mL of FBS were added depending if a 10% or 20% serum concentration was required.
To prepare α-MEM,1LofWFIwasmeasuredintoa1Lbeaker.10.2gofα-MEM powder and 2.2g NaHCO3 was added and stirred covered until all the powders had dissolved.
To prepare Medium 199, 1 L of WFI was measured into a 1L beaker. 9.5g of Medium
199 powder and 2.2g NaHCO3 was added and stirred covered until all the powders had dissolved.
To prepare DMEM-LG, 1 L of WFI was measured into a 1L beaker. 10.0g of DMEM-
LG powder and 3.7g NaHCO3 was added and stirred covered until all the powders had dissolved.
To prepare DMEM-HG, 1 L of WFI was measured into a 1L beaker. 10.0g of DMEM-
HG powder and 3.7g NaHCO3 was added and stirred covered until all the powders had dissolved.
To prepare IMDM, 17.7g of IMDM powder was used per litre of medium to be prepared.
The powder was dissolved in a small volume of WFI, 3.024g NaHCO3 was added, and then the entire amount measured with enough WFI added to make up to the whole litre. Medium was stirred covered until all powders had dissolved.
To prepare DPBS, 9.55g of DPBS powder was used per litre of WFI. The powder was dissolved in a small volume of WFI and then the entire amount measured with enough
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 265 Appendix A. Materials and General Methods A.2. Methods
WFI added to make up to the whole litre. DPBS was filter sterilised (0.22μm) into 200 or 400mL bottles.
A.2.1.2 Kangaroo Tail Type I Collagen
A salt pepsin digestion method modified from [402–407] was used to produce Type I collagen from kangaroo tail tendons as follows.
Freshly frozen kangaroo tails were stored at -20℃ until required, then were thawed at
4℃ overnight. Once thawed, the fur and skin were removed with a scalpel, and then the tail was cut into approximately 20cm long pieces. The tendons were pulled out of the tail with forceps and placed on ice. The tendons were frozen at -20℃ until processed further. 80g net weight of kangaroo tail tendon was finely chopped with a scalpel and ground in a mortar and pestle and dissolved in 32L of 0.5M glacial acetic acid (HAc).
It was allowed to stand, with stirring, for 30 minutes at 4℃ to allow swelling of the tendons. Crystallised and lyophilised pepsin powder was dissolved in 10mM HCL, and added to the tendon solution at 10% wet weight of the tendon, and pH adjusted to 2.5
-3.0. The solution was covered and left at 4℃ for 4 days, whilst being stirred slowly with a magnetic stirrer. After 4 days, the solution was centrifuged at 1200g for 30 minutes at 4℃.
The supernatant was decanted and kept cold. It was brought to 0.7M NaCl, using a stock solution of 4.5M NaCl. The stock solution was cold and was added very slowly, with constant stirring. A heavy, white dispersion of collagen formed upon the addition of the salt solution. This solution was allowed to stand with gentle stirring for 2 hours, and then centrifuged at 2000g for 30 minutes at 4℃. The supernatant was discarded and the pellet redissolved in 20L of 0.2M Tris-HCl with a pH of 7.4. Once the collagen was all redissolved, the pH of the whole solution was adjusted to 7.0 with a 1M NaOH solution. The solution was then brought up to 1.6M NaCl using a 4.5M
NaCl stock solution which was added slowly, but with vigorous stirring of the solution whilst avoiding air bubbles. It was left to then stand for 3 hours with slow stirring then
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 266 Appendix A. Materials and General Methods A.2. Methods centrifuged at 5000g for 30 minutes at 4℃ to produce a clear supernatant which was decanted and adjusted to 2.5M NaCl, resulting in the formation of a Type I collagen precipitate upon standing. The solution was left for 2 hours, and then centrifuged and the pellet retained. The collagen pellet was dialysed twice against 30L of fresh cold
50mM HAc at 4℃ for a duration of 12 hours each iteration.
The resulting collagen solution was then placed inside either conical freeze drying flasks or Teflon coated trays and frozen in a -70℃ freezer. These frozen solutions were then freeze dried overnight. The freeze dried collagen was stored at 4℃ until required. When required, the freeze dried collagen was reconstituted to 4mg/mL with 1M HAc and stirred overnight at 4℃. Dialysis membranes were softened in warm H2Oandweighted. The dissolved collagen solution was placed inside the membranes, the ends were knotted and the sample weighed. The samples were dialysed against 30L of 10mM HAc for a duration of 24 hours at 4℃, to sterilise the collagen. The concentration of the collagen was increased from 4mg/mL to 20-50mg/mL by pegging dialysis bags in a laminar flow hood at room temperature for several days until the desired concentration was reached as judged by weighing each dialysis bag. Given an initial collagen concentration, initial weight and final weight, the final collagen concentration can be calculated as follows:
M1 × C1 C2 = (A.1) M2
Where:
M1 = initial mass of collagen per bag
M2 = final mass of collagen per bag
C1 = initial collagen concentration i.e. 4mg/mL
C2 = final collagen concentration achieved.
The collagen was then dialysed twice (in the same dialysis membrane) against fresh
0.15M NaCl in 8L of 10mM HAc for 3 hours at 4℃. The dialysate was kept cold and was well stirred to thoroughly dissolve any salt crystals prior to the addition of the dialysis bags containing the collagen. The collagen was then dialysed twice against
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 267 Appendix A. Materials and General Methods A.2. Methods
8L of cold fresh 0.15M NaCl for 3 hours at 4℃. The dialysed collagen was aseptically transferred to screw cap septum vials, and stored at 4℃ until required for gelling.
A.2.2 Bone Marrow Harvest
A.2.2.1 Rat
Bone marrow was obtained from euthanized syngeneic inbred PVG/c male 5-6 week old rats weighing approximately 100g. Animals were initially sedated with 0.2mL of
100mg/mL ketamine hydrochloride subcutaneously. 1000 I.U. Heparin was adminis- tered subcutaneously 10 minutes post sedation. Animals were euthanized 10 minutes later with 0.3mL of 325mg/mL sodium pentobarbitone via cardiac puncture. Inside a laminar flow hood, both lower limbs were doused with 10% povidone-iodine solution.
The left and right femur, and the left and right tibia were aseptically removed and soft tissues were detached. Each bone was placed in a sterile Petri dish and covered, to maintain sterility.
A.2.2.2 Mouse
Bone marrow was harvested from euthanized 6 month old Balb/C mice weighing ap- proximately 30g. Animals were euthanized by CO2 asphyxiation. Long bones were removed as described in section A.2.2.1.
A.2.2.3 Porcine
Porcine marrow was aspirated from live healthy juvenile farm (30 - 40 kg) pigs being used as part of a closed chest porcine model of ischemia reperfusion, under which no significant haemo dynamic consequences or blood loss were observed. Oxygen levels were normal throughout the procedure with no evidence of hypoxia. All animals, bar one, from which marrow was harvested, had heart attacks prior to harvest.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 268 Appendix A. Materials and General Methods A.2. Methods
All animals were initially given 1.8mL Ketamine Hydrochloride 100mg/mL IM (180mg),
0.6mL Xylazine 20mg/mL IM (12mg) and 150mg Aspirin oral. All animals inhaled
Isoflurane for the duration of the procedure, and also received approximately 100mL of iodinated contrast, whilst fluid maintenance was with saline. All animals were intra- venously given varying amounts of Heparin (5,000U to 20,000U), Metroprolol 1mg/mL
(12.5mL to 30mL), Lignocaine (3.5mL to 9.5mL). One animal was given 7mL Levo- vist, one animal had a VF shock of 360J, and two animals received an experimental drug, designed to target the gene Egr-1, using the liposomal transfection agent Fu-
Gene6 (Roche), with no adverse effect or change in bone marrow being observed by the varying treatments.
The area around the knee was shaved using a razor and soapy water. The shaved area was doused with a 10% povidone-iodine solution. Under sterile conditions, the area was draped and an incision was made through the skin at the medial proximal end of the tibia. Any bleeding arteries and veins were promptly diathermied. A hole was drilled approximately 20mm deep into the tibia using a hand held drill, with a 3.47 mm bit.
A 14G catheter with a 10mL syringe containing 1000 I.U. heparin (in 1mL) and 4mL saline was placed into the hole, and marrow was aspirated out by drawing back on the syringe. The catheter and syringe were removed from the tibia, and the syringe was capped with a sterile needle to ensure the marrow remained sterile.
A.2.2.4 Equine
Bone marrow was aspirated from the hip of an anesthetised horse undergoing hip surgery at the Randwick Equine Centre, Sydney Australia. Two samples of bone mar- row in heparin, total solution volume of 4mL each were delivered on ice by Dr Chris
O’Sullivan shortly after the aspiration.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 269 Appendix A. Materials and General Methods A.2. Methods
A.2.3 Primary Cell Purification for Culturing
A.2.3.1 Rat
Inside a tissue culture hood, the bones were washed with 10mL of sterile DPBS. Meta- physes from both ends were resected with bone choppers whilst the bone was held with forceps. Bone marrow cells were collected by flushing the diaphysis with EMEM containing 10% FBS. This was done using a 19 gauge syringe needle for femur and a
21 gauge needle for tibia, attached to a 10mL syringe, filled with 10mL medium, with the marrow being collected into a new 10cm Petri dish.
A suspension of bone marrow cells was obtained by repeated aspiration of the marrow plug through a 21 gauge needle for femur and a 23 gauge needle for tibia. The cell preparation was then transferred into a 15mL tube and centrifuged for 10 minutes at
190g. The resulting pellet was suspended in 4mL of EMEM containing 20% FBS and plated into a T25 vented tissue culture flask and cultured in a humidified atmosphere of
95% air with 5% CO2 at 37℃. After 48 hours, the flask was gently agitated to dislodge any non adherent cells, and the culture medium was removed. 5mL of new medium was added. The medium was changed two or three times a week. After approximately
10 days of culture, as the culture approached confluence, the cells were passaged by trypsinization (0.05% trypsin/EDTA solution) and plated into a T75 vented tissue cul- ture flask with 10mL of medium. After approximately 7 days of culture, as the culture approached confluence, cells were again passaged and plated into a T225 vented tissue culture flask with 30mL of medium. As the culture reached almost complete confluence, cells were passaged and used as required in experiments.
A.2.3.2 Mouse
Inside a tissue culture hood, the bones were washed with 10mL of sterile DPBS. The distal end of the bone was resected with bone choppers whilst the bone was held with forceps. The bone was then placed distal end down in a 26 gauge needle. The needle containing the bone was placed in a 1.5mL tube, sharp end down, containing 250μL
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 270 Appendix A. Materials and General Methods A.2. Methods of EMEM containing 20% FBS, which was then capped. The tube was centrifuged for 2 minutes at 760g. Bone marrow was collected from the bottom of the tube. Any remaining marrow inside the needle was also flushed out with medium. Mouse marrow was used immediately and not cultured.
A.2.3.3 Porcine and Equine
Inside a tissue culture hood, the marrow sample was diluted with EMEM containing
10% FBS and placed in 15ml tube. A suspension of bone marrow cells was obtained by repeated aspiration of any marrow plugs through a 21 gauge needle. The cell prepa- ration was then centrifuged for 10 minutes at 190g. If no defined pellet was observed, the cell suspension was mixed thoroughly, diluted further with medium, transferred into multiple 15ml tubes, and centrifuged again for 10 minutes at 190g. The pellet was resuspended and cultured as described in section A.2.3.1.
A.2.4 Maintenance of Primary Cells
A.2.4.1 Passage Definitions
Table A.1: Passage Definitions
Passage # Cell details Trypsin contact Primary cells Cells straight from the animal 0 0 Suspension of primary cells which have been 1 plated onto tissue culture plastic, grown up and thenremovedwithtrypsin 1 Suspension of passage 0 cells that have been re 2 plated onto tissue culture plastic, grown up and thenremovedwithtrypsin n Passage (n-1) cells that have been plated onto n+1 tissue culture plastic and then removed with trypsin
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 271 Appendix A. Materials and General Methods A.2. Methods
A.2.4.2 Maintenance and Passage of Cells
When cells were 80-90% confluent, the old media was removed from the flask and the cells were gently washed with DPBS. Trypsin was added, 1.5ml for T25 flask, 3ml for T75 flask and 10ml for T225 flask, and the flask placed in the incubator for 3-5 minutes, or until cells dislodged with gentle agitation. EMEM supplemented with 20%
FBS, greater than amount of trypsin used, was added to the flask and mixed well to ensure that all the cells had lifted off the flask surface. The cell suspension was transferred to either a 15ml tube for T25 and T75 flasks, or to a 50ml tube for T225
flasks, and centrifuged for 8 minutes at 68g. The supernatant was discarded and the cell pellet resuspend in EMEM supplemented with 20% FBS. The cell suspension was plated into the required flask at the required density. Medium was changed 2-3 times a week.
A.2.4.3 Freezing Cells
Once cells were passaged and counted, the cell suspension was centrifuged for 8 minutes at 68g. The cell pellet was suspended in EMEM containing 20% FBS and 10% DMSO, and immediately transferred into cryovials at one T75 flask per tube. The cryovials were transferred to a cold cryogenic container containing isopropanol and placed in a-70℃ freezer. After 24 hours, the cryovials were transferred to a dewar containing liquid nitrogen.
A.2.4.4 Thawing cells
The desired cryovial was removed from liquid nitrogen and thawed immediately in a
37℃ water bath. Once defrosted, the cells were transferred to a 15ml tube containing
10ml of EMEM supplemented with 20% serum and centrifuged for 8 minutes at 68g.
The supernatant was discarded and the cell pellet resuspended in culture medium and transferred into the appropriate culture flask, usually a T25 if cells were frozen from a
T75 flask.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 272 Appendix A. Materials and General Methods A.2. Methods
A.2.4.5 Cell Counting and Viability
Cell suspensions were counted either using a haemocytometer, with viability confirmed with equal parts cell suspension and trypan blue, or a Vi-CELL XR Cell Viability
Analyser (Beckman Coulter) which automates the widely accepted trypan blue dye exclusion protocol. The Vi-CELL XR is a video imaging system that mixes a sample of suspended cells with trypan blue and delivers it to a flow cell. 100 images of each sample were taken, from which the software determined parameters such as total cell count/mL, viable cell count/mL, cell diameter, and cell circularity.
A.2.5 Maintenance of L929 Cell line
A.2.5.1 Maintenance and Passage
When cells were 80-90% confluent, the old media was removed from the flask and the cells were gently washed with DPBS. 1.5ml trypsin was added per T25 flask, and the flask placed in the incubator for 3-5 minutes, or until cells dislodged with gentle agitation. EMEM supplemented with 10% FBS, greater than amount of trypsin used, was added to the flask and mixed well to ensure that all the cells had lifted off the
flask surface. The cell suspension was transferred to either a 15ml tube and centrifuged for 3 minutes at 190g. The supernatant was discarded and the cell pellet resuspend in
EMEM supplemented with 10% FBS. The cell suspension was plated into the required
flask at the required density.
A.2.6 Maintenance of MS-5 Cell line
A.2.6.1 Maintenance and Passage
MS-5 cells were passaged as per section A.2.5.1 α-MEM supplemented with 10% serum was used.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 273 Appendix A. Materials and General Methods A.3. Histological Methods
A.2.7 Maintenance of NIH 3T3 (eGFP) Cell line
NIH 3T3 cell line was stably transduced by Modin and colleagues with a retroviral vector expressing the enhanced green fluorescent protein (eGFP) [408].
A.2.7.1 Maintenance and Passage
NIH 3T3 cells were passaged as per section A.2.5.1 IMDM supplemented with 10% serum was used.
A.2.7.2 Thawing cells
The desired cryovial was removed from liquid nitrogen and thawed immediately in a
37℃ water bath. Once defrosted, the cells were transferred to a 15ml tube, and 2 ml of
FBS was slowly added. 10ml of media supplemented with 10% serum was then added and centrifuged for 3 minutes at 190g. The supernatant was discarded, the cell pellet resuspended in 10ml culture medium and centrifuged again. This washing process was repeated once more. The cell suspension was plated into the required flask at the required density. Medium was changed after 24hrs.
A.3 Histological Methods
A.3.1 Stain Preparation
A.3.1.1 Crystal violet
A crystal violet solution was made by dissolving 1.5g of crystal violet in 100ml ethanol.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 274 Appendix A. Materials and General Methods A.3. Histological Methods
A.3.2 Embedding and Sectioning
After fixation with 10% formalin for at least overnight, samples were embedded in 3% agarose and processed on a Tissue-Tek VIP, Sakura Japan, and embedded in paraffin by the Histology and Microscopy Unit, UNSW. 10μm sections were cut on a Leica
Microtome, Sydney Australia, were deparaffinised in xylene and rehydrated through a series of alcohols ranging between 100% to 70%. Sections were stained as required and then dehydrated in alcohol, 70% - 100% and cleared in xylene and cover-slipped.
A.3.3 Well Plate Staining
A.3.3.1 Crystal violet for 35mm Petri dishes
Culture medium was removed and the dish washed with DPBS. Approximately 1ml of crystal violet stain was added to each dish and after 10 minutes, the dish was washed with DPBS until the crystal violet was removed. As the stain was made in alcohol, the staining process also fixed the cells.
A.3.3.2 Alkaline phosphatase
Alkaline phosphatase activity was determined using kit 86C-1KT (Sigma-Aldrich) as per the manufacturer’s instructions.
A.3.4 Sectioned Slide Staining
A.3.4.1 Deparaffinising and Rehydrating
Sections were deparaffinised in two baths of xylene for 5 minutes, and rehydrated in three baths of 100% alcohol for 2 minutes, one bath of 70% alcohol for 2 minutes, and one bath of triple distilled H2O for 2 minutes.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 275 Appendix A. Materials and General Methods A.3. Histological Methods
A.3.4.2 Dehydrating and Mounting
Sections were dehydrated for mounting by one bath of 70% alcohol for 2 minutes, three baths of 100% alcohol for 2 minutes, and cleared in two baths of xylene for 2 minutes. Xylene was streaked over the cover slip and the slide was gently placed over the mountant at 45° upside down. Both slide and cover slip were inverted and left to dry on a flat surface overnight.
A.3.4.3 Hematoxylin and Eosin
Required sections were stained with hematoxylin and eosin (H&E) by the Histology and Microscopy Unit, UNSW.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 276 Appendix B
Development of CBD-Fibronectin fragment chimaeras
The following protocol was used by Dr Kanda Sangthongpitag and Dr Frances Foong
(School of Biotechnology and Biomolecular Sciences, University of New South Wales) in the synthesis of CBD-Fn R277-R271.
Takara et al. Retronectin (TM) US Patents No. 5198423, No. 5686278, No. 6033907 and European Patent No. 399806.
Tissue culture grade polystyrene culture ware have charged surface modifications for the passive adsorption of fibronectin and vitronectin from serum, which when bound, support the growth of anchorage dependent cell lines. Cellulose hollow fibres are uncharged, and do not adsorb enough of these adhesion molecules for cell attachment and growth. Recombinant DNA techniques were use to clone and express fusion protein chimaeras between the cellulose binding domain (CBD) isolated from clostridium species and binding domains isolated from human fibronectin molecule. The CBD would facilitate binding between the fibronectin fragment and cellulose membrane. The various fibronectin fragments created as CBD fusion chimaeras are shown below.
277 Appendix B. Development of CBD-Fibronectin fragment chimaeras
Table 1 Cellulose binding domain fibronectin domain fusion proteins Name Source Fibronectin domain Cognateligand CBD-Fn R277 ATCC Cell binding domain 1 VLA-5 (RGDS) CBD-Fn R271 ATCC & CS1 and heparin VLA-4 endothelial cell binding domain II Heparin line (EC-RF24) CBD-Fn R277-R271 ATCC Cell binding domain 1, VLA-4, VLA-5, CS1 and heparin glycoproteins binding domain
The pET34b+ vector (Novagen) contains the sequence for the clostridium cellulovorans CBD under the control of the T7 promoter. Downstream of the CBD sequence is a thrombin protease cleavage site followed by a multiple cloning site (MCS) and a poly histidine coding region. This allows for the production of CBD fusion protein with a poly-his tag when the insert of interest is cloned into the MCS. The thrombin cleavage site allows for the removal of the CBD from the fusion partner. These recombinant proteins were expressed in E. coli and purified from insoluble cell extracts by nickel metal ion chromatography.
Construction of vectors1 Fibronectins are high molecular weight glycoproteins found in plasma and extracellular matrices. The primary structure of fibronectins are similar but not identical polypeptides. This is due to the variability of internal primary sequences which are the result of alternative splicing in the extra domain region (ED) and connecting strand region (IIICS) of the pre-mRNA. The fibronectin molecule consists of three different types of homologous amino acid repeats (Figure 1). They are type I (~45 amino acid residues long with 2 disulfide bonds), type II (60 amino acid residues long with 2 disulfide bonds) and type III (90 residues long). The remaining amino acids of the fibronectin molecule make up a non type III connecting domain. Units 8-10 of type III and non type III (CS1 region) consists of the cell binding domain 1, containing a RGDS motif, which binds to integrin VLA-5 and cell binding domain 2 which binds to integrin VLA-4. The units 12-14 contain a heparin binding domain proteoglycan region which mediates DNA/RNA attachment to the fibronectin molecule.
1 Constructed by Dr Kanda Sangthongpitag and Dr Frances Foong. School of Biotechnology and Biomolecular Sciences, UNSW
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 278 Appendix B. Development of CBD-Fibronectin fragment chimaeras
RGDS
8 9 10 11 12 13 14 NH2 COOH
Fibrin I/Heparin I Collagen Cell I Heparin II Cell II Fibrin II
Type I Type II Type III Non type III (IIICS)
Figure 1 Structure of the fibronectin molecule.
The cloning strategy of CBD-three functional parts of fibronectin is outlined below:
Z Construction of CBD-fnC277
The recombinant plasmid pFH 154 (Table 2) carrying a 2.5 kb cDNA fragment of fibronectin 1 was obtained from ATCC. Physical mapping of the 2.5 kb cDNA indicated that it contained cell binding domain 1 (RGDS or VLA-5 binding ligand) and part of heparin (Heparin 2) binding domain of fibronectin 1. The fnC277 fragment containing the spaning region of cell biding domain was amplified from the digested HindIII pFH154 by PCR. PCR was performed in a volume of 50 l containing standard PCR buffer, 2.5 mM MgCl2, 200 M of dNTP, 150 ng of forward and reverse fnC277 primers (Table 3), 50 ng of DNA and 1 U a mixture of Taq polymerase and Pwo polymerase (Roche). Amplification conditions consisted of an initial denaturation step at 94oC for 3 min, then 25 cycles of 30 sec at 94oC, 30 sec 55oC and 90 sec at 72oC followed by one additional cycle of 7 min at 72oC. The PCR C277 fragment was ligated to pGEMT-easy. The correct recombinant plasmid pGEMT-easy-fnC277 was analysed by DNA sequencing and physical mapping. This recombinant plasmid was designated pRN024 (Table 2). Plasmid pRN024 will be digested by SalI and HindIII restriction enzymes. The SalI/HindIII fnC277 fragment will be inserted into the expression vectors pET23b and pET34b at SalI and HindIII sites (Figure 2) and subsequently expressed in E.coliBL21(DE3)pLys.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 279 Appendix B. Development of CBD-Fibronectin fragment chimaeras
Table 2 Cloned plasmids Plasmid Relevant characteristics pGEMT-easy Apr, lacZ, multiple cloning site (ApaI-NsiI), T-overhang, 3018 bp
pGEM11 Apr, lacZ, multiple cloning site (SfiI-HindIII), 3223 bp
pET23b Apr, T7 promoter, ribosome binding site (rbs), T7-Tag, multiple cloning site (BamHI-XhoI), His-Tag, T7 terminator, 3665 bp
pET23d Apr, T7 promoter, rbs, T7-Tag, multiple cloning site (BamHI-XhoI), His-Tag, T7 terminator, 3663bp
pET34b Kmr, T7 promoter, lac operator, rbs, CBDclos-Tag, S-Tag, multiple cloning site (SrfI-XhoI), His-Tag, T7 terminator, 5918 bp
pMCO Apr, Lac promoter, ompA & pelB periplasmic targeting domains, c-myc Tag, amber stop codon, geneIII protein, 3776 bp
pFH154 Apr, carrying 2.5 kb cDNA of fibronectin 1, 6.5 kb
pRN001 Apr, insertion of Fab BB9 fragment into pMCO plasmid, 5067 bp
pRN002 Apr, ligation of scFv BB9 fragment to pGEMT-easy, 3801 bp
pRN003 Apr, ligation of scFv BB9 fragment to pGEMT-easy, 3801 bp
pRN004 Apr, insertion of scFv BB9 into pET23d, 4446 bp
pRN005 Apr, insertion of Fab 7E10 fragment into pMCO plasmid, 5076 bp
pRN007 Apr, insertion of Fab KF8 fragment into pMCO plasmid
pRN014 Apr, ligation of scFv KF8 fragment to pGEMT-easy, 3786 bp
pRN022 Apr, insertion of scFv KF8 fragment into pET23b, 4433 bp
pRN023 Apr, insertion of scFv 7E10 fragment into pGEM11, 3967 bp
pRN024 Apr, ligation of fnC277 fragment to pGEMT-easy, 3858 bp
pRN025 Apr, insertion of scFv BB9 into pET34b, 6701 bp
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 280 Appendix B. Development of CBD-Fibronectin fragment chimaeras
Table 3 Primer sequences Primers Sequence
Antibodies primer R Hc 5’ –AGGCTTACTAGTACAATCCCTGGGCACAAT- 3’ F Hc 5’ –AGGTCCAACTGCTCGAGTCTGG F Lc5 (BB9 &KF8) 5’ – CCAGATGTGAGCTCGTGATGACCCAGACTCCA-3’ F Lc7 (7E10) 5’ –CCAGTTCCGAGCTCGTGATGACACAGTCTCCA-3’ R Lc 5’ –GCGCCGTCTAGAATTAACACTCATTCCTGTTGAA-3’ F VH+L BB9-23 5’ –CTCGAAAGCTTGGCTGATCTGGCAAAACC- 3’ F VH+L BB9-34 5’ –CTCAAGCTTGGGGCTGATCTGGC- 3’ R VH+L BB9 5’ –GCAGCATCCTCGAGTTTGATTTCC- 3’
F VL KF8 5’ –CAG ATG TGA GCT CGT GAT GAC- 3’ R VL KF8 5’ –GGT AGG GAG CTC GTC AAT TGT- 3’
F VH KF8 5’ –GTG GAG CTG AAT TCG TCT GGC C- 3’ R VH KF8 5’ –CTG TGG GGG GTA CCC AGC TGG GTG AC-3’
F VL 7E10 5’-GGA TCC GAA TTC AGT GAT GAC ACA GTC- 3’ R VL 7E10 5’-GGA TGG GAG CTC CAT GGA TAC AGT TGG TG- 3’
F VH 7E10 5’ –TTG AGT CTG AGC TCG CGC TGA TGA AG- 3’ R VH 7E10 5’ –GGA TAG ACA GAC TCG AGT GTC GTT TTG- 3’
F Fn C277 5’ –ATC ATA GCA TGC GTC GAC CCT CCC ACT G- 3’ R Fn C277 5’ –CAT GCA TCT GAA GCT TGG ATG GTT TGT C- 3’
F Fn H271 5’ –CCA TGC ATA AGC TTA TGG CTA TTC CTG C- 3’ R Fn H271 5’-CTT ATG ACT GCA GCG GCC GCT GTG GAA GGA- 3’
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 281 Appendix B. Development of CBD-Fibronectin fragment chimaeras
Z Construction of CBD-FnH271
Because plasmid pFH154 lacked some part of the heparin binding domain (DNA biding domain) and CS1 (VLA-4 binding site) which is part of the connecting strand (IIICS) region, RT-PCR was carried out to obtain these two regions. Total RNA was isolated from the endothelial cell (EC-RF24) and converted into first strand cDNA. The first strand cDNA fragment encoding heparin binding domain and CS1 was amplified by PCR. PCR was performed in a volume of 25 l containing standard PCR buffer, 2.5 mM MgCl2, 200 M of dNTP, 150 ng of forward and reverse fnH271 primers (Table1), 5 l of first strand cDNA and 1 U AmpliTaq polymerase (Perkin Elmer). Amplification conditions consisted of an initial denaturation step at 94oC for 3 min, then 30 cycles of 30 sec at 94oC, 30 sec 50oC and 2 min at 72oC followed by one additional cycle of 10 min at 72oC. The fnH271 fragment will be ligated to pGEMT-easy. The correct recombinant plasmid will be analysed by DNA sequencing and physical mapping. The fnH271 fragment will subsequently be cloned into the expression vectors pET23b and pET34b at HindIII and NotI sites (Figure 2).
Z Construction of CBD-fnC277-fnH271
fnC277 PCR fragment will be fused to fnH271 PCR fragment at the HindIII site. Amplification of fnC277-fnH271 fragment will be carried out using forward fnC277 primer and reverse fnH271 primer (Table 3) and ligated to pGEMT-easy. Cloning fnC277-fnH271 fragment was performed as previously described.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 282 Appendix B. Development of CBD-Fibronectin fragment chimaeras
A. pET23-fnC277or fnH271 or fnC277-H271
His Tag CDS T7 terminator f1 origin C277 or H271
T7 Tag CDS (207-239) T7 promoter (303-319) pET23-fnC277 or fnH271 T7 transcription start (302) ApR
pBR ori
B. pET34b-fnC277 or fnH271 or fnC277-H271
His.Tag T7 terminator f1 origin fnC277 or fnH271
S.Tag KmR CBD clos
T7 promoter pET34b-fnC277 or fnH271
pBR322 lacI
Figure 2 The model of an expression plasmid of fibronectin containing high affinity of binding domains: RGDS, heparin 2 and CS1
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 283 Appendix C
Cross-flow Bioreactor Design Drawings
284 Appendix C. Cross-flow Bioreactor Design Drawings
Figure C.1: Cross-flow bioreactor design general view, all dimensions are in mm.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 285 Appendix C. Cross-flow Bioreactor Design Drawings
Figure C.2: Cross-flow bioreactor design detailed channel view, all dimensions are in mm.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 286 Appendix D
Cross-flow Bioreactor Design 2 Drawings
Figure D.1: Cross-flow bioreactor design 2 top view, scale: 1:1.
287 Appendix D. Cross-flow Bioreactor Design 2 Drawings
Figure D.2: Cross-flow bioreactor design 2 front view.
Figure D.3: Cross-flow bioreactor design 2 bottom view.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 288 Appendix D. Cross-flow Bioreactor Design 2 Drawings
Figure D.4: Cross-flow bioreactor design 2 section view.
Figure D.5: Cross-flow bioreactor design 2 isometric view.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 289 Appendix D. Cross-flow Bioreactor Design 2 Drawings
Figure D.6: Cross-flow bioreactor design 2 right view.
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 290 Appendix E
Fabrication of SU-8 master moulds
The following protocol was used by Owen The (Staff, Graduate School of Biomedi- cal Engineering, University of New South Wales) and Huaying Chen (PhD candidate,
Graduate School of Biomedical Engineering, University of New South Wales) in the development of master moulds by SU8 lithography techniques at the UNSW Semicon- ductor Nanofabrication Facility, Sydney Australia.
SU-8 2100 (Y111075 0500L1GL), SU-8 2025 (Y111069 0500L1GL) and developer (Y020100
4000L1PE) were obtained from MicroChem Corporation, MA USA.
Wafer Preparation
Wash in Acetone, and rinse in isopropanol Dry with compressed air Dehydration bake for 3 min at 180°C
291 Appendix E. Fabrication of SU-8 master moulds
Spin Coat
Align and centre the wafer on the vacuum chuck Apply the SU-8 onto the wafer on the vacuum chuck o When using highly viscous resists, it is easier to pour a small amount straight from the bottle onto a wafer spinning at 100-200RPM to ensure that the blob of SU-8 is centred on the wafer. o Let the SU-8 sit for a couple of minutes to allow any bubbles to rise to the surface Spin the wafer to 500RPM at a ramp of 100RPM/s and hold for 3s. Ramp to the desired speed at 500RPM/s and hold at the speed for 30s. See Figure 1: SU-8 2100 (a), 2025(b) Spin Speed vs Thickness Let the wafer sit on a level surface for a few minutes o Due to the viscosity of the SU-8, there may be some bubbles and unevenness on film. The best way to remove this is to let the wafer rest
The range of SU-8 thicknesses is 20-80µm and 100-250 µm for SU-8 2025 and 2100, respectively.
(a) SU-8 2100 (b) SU-8 2025 Figure 1: SU-8 2100 (a), 2025(b) Spin Speed vs Thickness
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 292 Appendix E. Fabrication of SU-8 master moulds
Soft bake (Total amount of time in this step is approximate155 mins)
From Room Temperature, ramp to 40°C at 120°C/hr Hold at 40°C for 5 minutes Ramp to 60°C at 120°C/hr Hold at 60° C for 5 minutes Ramp to 120°C at 240°C/hr Hold at 120°C for 60 minutes Ramp to Room temperature at 120°C/hr To ensure that the wafer has been baked to completion, test by pressing an edge of the wafer with tweezers. The photo resist should not be soft, and only leave a depression after considerable force has been applied
(a) SU-8 2100 (b) SU-8 2025 Figure 2 Recommended soft bake times for SU-8 2100, 2025
Exposure
Expose using Mask Aligner at the appropriate dosage for the desired thickness. See Figure 3 expos. Adjust the time to suit the mask aligner power (the Quintel Mask aligner at SNF has a power of 10mW/cm2).
(a) SU-8 2100 (b) SU-8 2025 Figure 3 Exposure dosages for SU-8 2100, 2025
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 293 Appendix E. Fabrication of SU-8 master moulds
Post Exposure Bake (Total amount of time in this step is around 180 hours)
From Room Temperature, ramp up to 80°C at 80°C/hr Hold at 80°C for 90 minutes Ramp down to Room temperature at 80°C/hr The pattern should be visible after this step.
(a)SU-8 2100 (b) SU-8 2025 Figure 4 Recommended post exposure bake times for SU-8 2100, 2025
Development
Develop in either SU-8 Developer (preferred) or Acetone. Development time increases with thickness and bake times. See Figure 5. To speed up the development process, agitate the sample as it is being developed. Once completed, rinse with clean developer/acetone for 10 seconds. Rinse with isopropanol. Isopropanol will wash off the developer/acetone with the dissolved SU8 and halt the development process. If white marks present, develop further. Rinse in DI water for 5 minutes and dry with compressed air
(a) SU-8 2100 (b) SU-8 2025 Figure 5 Development times for SU-8 2100, 2025
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 294 Appendix F
Dissipation of gradient along diffusion chamber for the pyramidal gradient generating device
The following model describing dissipation of gradient along the diffusion chamber for the pyramidal gradient generating device was written by Dr Robert Nordon, Senior
Lecturer, Graduate School of Biomedical Engineering, University of New South Wales,
2009.
295 Appendix F. Dissipation of gradient along diffusion chamber for the pyramidal gradient generating device
The pyramidal device consists of a network of channels that generate a concentration gradient. The concentration gradient entering the flow cell shown in Figure 1 is dissipated by diffusion. The reduction in the gradient along the diffusion chamber is modelled by applying the continuity equation for mass and Fick’s law of diffusion. The equation of continuity expressing conservation of mass is c .JJ R (1) t diff conv
Jdiff Dc . (2)
where c is concentration, . is the gradient operator, Jdiff is mass flux due to diffusion
(a vector), Jconv is convective flux, R is a source term and D is the diffusion coefficient of the solute in water. CX ,1 0 Y
Y 1
V x CYY0, CXY ,
Y 0 X 0
CX ,0 0 Y
Figure 1 Dissipation of concentration gradient along diffusion chamber This process is modelled with the following assumptions.
1. There is uniform plug flow along the flow cell. The fluid velocity is Vx 2. The walls of the flow cell are impermeable to mass and there is zero flux normal to the wall at y 0 and yw i.e., cxy,0 y . yw0, 3. The diffusive flux along the direction of flow ( x direction) and between the plates of the flow cell ( z direction) is negligible compared to the flux across streamlines ( y direction).
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 296 Appendix F. Dissipation of gradient along diffusion chamber for the pyramidal gradient generating device
4. There is a linear concentration gradient at the entry to the flow cell i.e., cxy, ky. x0 Defining dimensionless variables cxyVw CXYPe;;;x (3) cwwD max The governing equation at steady state ct 0 is 2C C 0 Pe (4) YX2 With boundary conditions C CYY0, ; 0; (5) Y Y 0,1 The Fourier series solution to this PDE is found by separation of variables. 2 14 1 FV21i 2 X CXY,c B osFV2i1 YexpGW (6) 2 2 HX Pe 2 i1 21i HXGW 1 The concentration gradient at Y is 2 2 CXY, 41 FV21i 2 X B exp GW (7) YiP e Y 12 i1 21 HXGW
The following Matlab code which solves the steady state concentration and calculates the concentration gradient at the midline for the pyramidal gradient generating device was written by Dr Robert Nordon, Senior Lecturer, Graduate School of Biomedical
Engineering, University of New South Wales, 2009.
function [C,grad]=diffchamber1(X,Y,Pe,n) %% Solves the diffchamber (Pyramidal gradient generating device) % steady state concentration % Pe Peclet number %x− dimensionless length along chamber (0..infinity) %y− dimensionless transverse position (0..1) % assumes linear inlet concentration C(0,y)=y % assumes zero flux at y=0 and 1 % expresses solution as Fourier cosine series %n− number of terms % solution is C(x ,y )=
% 1/2−4/Piˆ2*sum(1/(2*i−1)ˆ2*cos((2*i−1)*Pi*y )*exp(−(2*i−1)ˆ2*Piˆ2*x /Pe),i=1..n)
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 297 Appendix F. Dissipation of gradient along diffusion chamber for the pyramidal gradient generating device
X=X(:); Y=Y(:); Z=zeros(length(Y),length(X)); %evaluate coefficients a=zeros(n,1); i=0:n−1; c=(2*i(2:n)−1); c=repmat(c,length(Y),1); y=repmat(Y,1,n−1);
Cy=1./c.ˆ2.*cos(c.*y*pi); for k=1:length(Pe) for j=1:length(X) x=repmat(X(j),size(y));
Cx=exp(−c.ˆ2*piˆ2.*x/Pe(k)); Z(:,j)=1/2−4/piˆ2*sum(Cy.*Cx,2); end subplot(length(Pe),1,k); imagesc(X,Y,Z); colormap(hot); colorbar; ylabel('Y'); title(['Pe=' num2str(Pe(k))]); C{k}=Z; if k==length(Pe) xlabel('X'); end end %% Also calculates concentration gradient at midline % grad =
% −4/Piˆ2*sum(−1/(2*i−1)*sin(1/2*(2*i−1)*Pi)*Pi*exp(−(2*i−1)ˆ2*Piˆ2*x /Pe), %i=1..n) x=0:0.01:1; grad(:,1)=x; x=x'; x=repmat(x,1,n); a=2*(1:n)−1; a=repmat(a,length(x),1);
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 298 Appendix F. Dissipation of gradient along diffusion chamber for the pyramidal gradient generating device
grad(:,2)=−4/piˆ2*sum(−1./a.*sin(1/2*a*pi)*pi.*exp(−a.ˆ2*piˆ2.*x),2); figure; plot(grad(:,1),grad(:,2)); xlabel('X/Pe'); ylabel('gradient at Y=0.5');
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 299 Appendix G
Estimation of concentration gradient generated by the counter flow diffusion chamber
The following model estimating the concentration gradient generated by the counter
flow diffusion chamber was written by Dr Robert Nordon, Senior Lecturer, Graduate
School of Biomedical Engineering, University of New South Wales, 2009.
300 Appendix G. Estimation of concentration gradient generated by the counter flow diffusion chamber
Assume there is a linear converging flow field Vx kx (1) where x is position, k is the velocity gradient, and Vx is the fluid velocity. Note that the velocity is negative (to the left) if x is positive and positive (to the right) if x is negative. The governing convection diffusion equation in one dimension is dc2 dc 0 Dkx (2) dx2 dx where c is concentration and D is the diffusion coefficient of the solute in water. The boundary conditions are lim lim cc; c0 (3) xx max Defining dimensionless variables: Vxx kx2 k c Pe ;;X x Pe C ; (4) DD D c max Thus the Peclet number is linearly related to the square of the distance from the origin x 0 . The governing equation and boundary conditions are then dC2 dC 0;0 xC ; C 1; (5) dX2 dX The solution for this non-linear ordinary differential equation is 9H9HX CX 1 : : II :1erf: II (6) 2;J: ;J2 I The concentration gradient is eX 2 /2 CX (7) 2
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 301 Appendix H
Elementary hydrodynamic model for counter flow diffusion chamber
The following elementary hydrodynamic model for the counter flow diffusion chamber was written by Dr Robert Nordon, Senior Lecturer, Graduate School of Biomedical
Engineering, University of New South Wales, 2009.
302 Appendix H. Elementary hydrodynamic model for counter flow diffusion chamber
Elementary hydrodynamic model for counterflow gradient generator The device consists of two parallel plate flow cells separated by a porous membrane. Figure 1 depicts a longitudinal section through the device.
BC1:qq 0 BC2 : ql q 1 vxFV px px 1 dp x m HX12 1 rq x dx 11 b 2 1
dp x 2b 2 rq x 2 dx 22 qq BC3:ql 0 BC4 :2 0 2 2
Figure 1 Elementary hydrodynamic model As a first approximation of the hydrodynamic characteristics of the device the following assumptions are made: 1. The parallel plate flow cells are assumed to have infinite width, so there is no pressure gradient in the z direction (out of page). dp x dp x 2. The pressure gradients 1 and 2 in the x direction are linearly dx dx qx qx related to the cross-sectional flow rates 1 and 2
dp x dp x rq x 1 ; rq x 2 (1) 11 dx 2 2 dx r r 3. The coefficients 1 and 2 are calculated by assuming parabolic laminar flow between parallel plates with separation 2B , width W , and viscosity .
33 rr; (2) 12B 33WBW 2212 4. The porous membrane has hydraulic permeability , so the permeation flux across the membrane is given by Darcy’s Law
vxFV px px (3) m HX12 5. The flows into the upper chamber with length l are equal but opposite in direction:
qqqlq 110; (4) 6. There is only one outlet in the lower chamber, so qqql 2202; 0 (5)
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 303 Appendix H. Elementary hydrodynamic model for counter flow diffusion chamber
Defining dimensionless variables that are related to the plate separation of the top and bb l bottom chamber 12, , hydraulic permeability and the length of the flow cells . 3l 223 l AA; (6) 12bb3 3 2212 The dimensionless volume flow rates and length are qqx QQX12; (7) 12qql whereq is the inlet flow rate for the upper chamber (lower chamber outlet flow rate is 2q ). The dimensionless governing equation and boundary conditions are dQ2 X 1 AQ X AQ X dX 2 11 2 2 (8) dQ2 X 2 AQ X AQ X dX 2 22 11 QQ01;1 1 11 (9) QQ 2202;10 The general solution is A 1 FVcX11 cX cX cX2 QXAe e c A e e 2 (10) 112 c HXGW 2 ce1 c cAA with 12. Note that for this model, the flow profile is independent of fluid viscosity, but dependent on membrane permeability, flow cell length and plate separation.
The following Matlab code for the elementary hydrodynamic model for the counter
flow diffusion chamber was written by Dr Robert Nordon, Senior Lecturer, Graduate
School of Biomedical Engineering, University of New South Wales, 2009.
function [X,Q]=hydromodel(A1,A2) %% hydrodynamic model for countercurrent diffusion chamber [A1,A2]=meshgrid(A1,A2); C=A1+A2; X=0:0.01:1; A1=repmat(A1,[1,1,length(X)]); A2=repmat(A2,[1,1,length(X)]); C=repmat(C,[1,1,length(X)]); X=repmat(X(:),[1,size(A1,1),size(A2,2)]); X=permute(X,[2,3,1]); C=sqrt(C);
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 304 Appendix H. Elementary hydrodynamic model for counter flow diffusion chamber
Q=1./C.ˆ2./(1−exp(2*C)); Q=Q.*(A1.*(−exp(−C.*(X−1))+exp(C.*(X+1)))+(A1+2.*A2).*(−exp(−C.*(X−2))+exp(C.*X))); Q=Q−A2./(A1+A2); [m,n,p]=size(A1); for i=1:n subplot(1,n,i); plot(squeeze(X(1,1,:)),squeeze(Q(:,i,:))); grid on; legval=num2str(squeeze(A2(:,1,1))); Alab=repmat('A2=',m,1); legval=[Alab legval]; legend(legval); title(['A1='num2str(A1(1,i,1))]); xlabel('X'); ylabel('Q'); end
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 305 Appendix I
Matlab code
The following Matlab code was written by Carl Gabel, MBiomedE student, Graduate
School of Biomedical Engineering, University of New South Wales, 2008.
The electrical interference or electrical gain produced by the camera’s photosensors is referred to as the camera gain.
I.1 GetGain
This program produces a gain matrix which is used to correct for the non uniformity introduced by the camera.
Images captured displayed a non uniform brightness of static uniform solutions. In order correct for the anomolous electrical gain introduced by the camera’s photosensor, the gain of each pixel must be calculated. Each image is 1040 by 1388 pixels. The gain was calculated by determining the deviation of each pixel from the overall image mean intensity. A unity matrix (1040 by 1388 of 1) is divided by the gain of the pixels.
This result is effectively the inverse of the anomalous gain applied by the camera’s electronics and is called the gain matrix.
306 Appendix I. Matlab code I.1. GetGain
The GetGain program illustrates a relationship between an independent variable and the camera gain. It also serves to illustrate the relationship between pixel intensity and an independent variable.
A relationship between exposure time and gain, and concentration and gain was in- vestigated. The independent variables tested in this case were exposure time and concentration. It was found that the camera gain was independent of both exposure time and concentration.
As there is no relationship, to minimise any noise or random errors in the system, the
final gain matrix is calculated as the mean of the individual gain matrices of each of the calibration images. This yields two gain matrices, one for exposure time and one for concentration. As there was an insignificant difference between these two matrices, the exposure time matrix was used as the gain matrix. This gain matrix was produced for fluorescein, FITC20 and FITC50.
Due to the linear relationship between both concentration and pixel intensity, and for exposure time and pixel intensity, it is very simple to calculate a concentration given an exposure time and pixel intensity.
There are inactive or hyper-active photo receptors in the camera which produce a very dark or excessively bright pixel respectively. Upon analysis of the images it was found that these inactive or hyper-active photo receptors demonstrated a dynamic response, appearing as a dark or bright pixel randomly on images. The intensity of these dark or bright pixels had a dominating effect when determining the multi-image mean pixel gain. To correct for this, an anti-aliasing filter was generated and then applied to the individual images used to calculate the mean gain matrix.
The output of this function is the gain matrix. Multiplying an image by the gain matrix corrects for the non uniformity introduced by the camera.
function GainOut = GetGain(indep, varname)
[FileName, PathName] = uigetfile('*.tif','Select the tiff−file', ...
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 307 Appendix I. Matlab code I.1. GetGain
'MultiSelect', 'on'); %indep = [012345678910]; for i = 1:length(FileName) img = imread([PathName FileName{i}]); if isa(img, 'uint16')==1 calimg(:,:,i) = img; imstat(:,i) = reshape(img,1443520,1); elseif isa(img, 'uint8')==1
calimg(:,:,i) = uint16(img)*255; imstat(:,i) = reshape(uint16(img)*255,1443520,1); else msgbox(strcat('The file ',FileName(i), ... ' is not a valid target.'), 'Invalid Image File', 'error'); break end end calmean = mean(double(imstat)); calstd = std(double(imstat)); p = polyfit(indep,calmean,1); py = indep.*p(1) + p(2); for i = 1:length(FileName)
gain(:,:,i) = ones(1040,1388).*calmean(i)./double(calimg(:,:,i)); gainstat(:,i) = reshape(gain(:,:,i),1443520,1); end gainmean = mean(gainstat); gainstd = std(gainstat); figure('Name', strcat('Mean Intensity vs. ', varname)); errorbar(indep,calmean,calstd, 'x'); hold on; plot(indep, py, 'r'); xlabel(varname);ylabel('Mean Intensity') hold off; figure('Name', strcat('Gain vs. ', varname)) ... ;errorbar(indep,gainmean,gainstd);
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 308 Appendix I. Matlab code I.1. GetGain
xlabel(varname);ylabel('Gain') cin = mean(double(gain),3); a0 = 1/sqrt(2*pi); a1 = 1/sqrt(2*pi)*exp(−1/2); a2 = 1/sqrt(2*pi)*exp(−1/sqrt(2)); a3 = a0+4*a1+4*a2; gaincalibration = cin; imax = 1040; jmax = 1388; for i = 2:imax−1 for j = 2:jmax−1
gaincalibration(i,j) = (a0*cin(i,j) + a1*(cin(i−1,j) ... +cin(i+1,j)+cin(i,j−1)+cin(i,j+1))+a2*(cin(i−1,j−1) ... +cin(i+1,j−1)+cin(i−1,j+1)+cin(i+1,j+1)))/a3; end end
GainOut = {gaincalibration, p}; end
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 309 Appendix I. Matlab code I.2. GainCal
I.2 GainCal
This program is used to calibrate images using the gain matrix generated by GetGain.
When the program is run, images are selected to be calibrated, which are then calibrated and then saved as new images.
function y = GainCal(gain)
[FileName,PathName] = uigetfile('*.tif','Select the tiff−file', ... 'MultiSelect', 'on'); %gain = gaincalibration; clear calimg; for i = 1:length(FileName) calimg = imread([PathName FileName{i}]); if isa(calimg, 'uint16')==1 fixedcalimg = calimg; elseif isa(calimg, 'uint8')==1
fixedcalimg = uint16(calimg) .* 256; else msgbox(['The file ' [PathName FileName{i}] ... ' is not a valid target.'], 'Invalid Image File', 'error'); break end
imwrite(uint16(round(gain.*double(fixedcalimg))), ... [PathName 'Cal ' FileName{i}]); end end
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 310 Appendix I. Matlab code I.3. Coordinates
I.3 Coordinates
This data file contains the co ordinates of the top left corner of each image to be collated for the Counter Flow Diffusion Chamber.
FITC20EFO{1}= [1138,2538;250,1399;1207,1387;2156,1378;263,152; ... 1204,141;2159,129]; FITC20EFO{2}= [1095,2521;220,1365;1186,1353;2116,1343;218,136; ... 1195,125;2122,116]; FITC20EFO{3}= [1432,2547;547,1417;1510,1409;2445,1399;526,167; ... 1520,155;2451,145]; FITC20EFO{4}= [1134,3178;226,2010;1189,2000;2133,1990;219,767; ... 1160,757;2145,747]; FITC20EFO{5}= [1480,3188;590,1954;1557,1942;2496,1932;571,741; ... 1557,729;2496,719]; FITC20EFO{6}= [1474,4033;574,2892;1532,2882;2476,2872;566,1657; ... 1532,1649;2476,1636;542,1049;1583,1041;2470,1032];
FITC20EHO{1}= [1479,4204;603,3167;1485,3250;2495,3238;593,1939; ... 1558,1928;2515,1917;577,695;1573,683;2509,674]; FITC20EHO{2}= [1512,4466;620,3365;1586,3356;2540,3344;609,2130; ... 1577,2120;2546,2108;588,873;1596,862;2534,851]; FITC20EHO{3}= [1460,4241;578,3207;1554,3196;2506,3187;562,1957; ... 1552,1947;2496,1934;556,700;1584,690;2481,681]; FITC20EHO{4}= [1288,4306;436,3249;1402,3238;2354,3227;422,2005; ... 1406,1995;2361,1986;420,727;1402,717;2349,706]; FITC20EHO{5}= [1351,4253;483,3198;1451,3188;2399,3176;477,1955; ... 1449,1945;2393,1935;463,696;1467,684;2395,674];
FITC150EFO{1}= [1462,2511;574,1486;1548,1476;2473,1466;566,277; ... 1525,267;2468,257]; FITC150EFO{2}= [1498,2536;616,1453;1581,1442;2502,1431;591,244; ... 1564,233;2502,244]; FITC150EFO{3}= [1545,2548;627,1537;1588,1529;2521,1519;613,302; ... 1588,290;2525,281]; FITC150EFO{4}= [1303,3936;415,2823;1373,2812;2317,2800;392,1592; ... 1373,1580;2307,1569;380,948;1303,929;2299,925];
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 311 Appendix I. Matlab code I.3. Coordinates
FITC150EFO{5}= [1465,4122;579,3022;1543,3012;2474,3002;557,1794; ... 1552,1782;2472,1774;512,605;1495,594;2364,595]; FITC150EFO{6}= [1455,3854;567,2971;1533,2959;2458,2951;541,1746; ... 1535,1742;2458,1732;609,1164;1581,1158;2470,1140]; FITC150EFO{7}= [1340,2374;443,1386;1408,1376;2332,1366;429,453; ... 1413,447;2332,439;];
FITC150EHO{1}= [1433,4178;547,3119;1494,3109;2453,3099;543,1894; ... 1502,1883;2451,1871;543,634;1532,623;2443,613]; FITC150EHO{2}= [1459,4190;581,3213;1512,3203;2467,3193;557,1991; ... 1525,1980;2461,1969;545,721;1524,710;2444,700]; FITC150EHO{3}= [1464,4106;570,3187;1547,3179;2474,3167;566,1946; ... 1534,1940;2470,1930;566,681;1592,671;2464,661]; FITC150EHO{4}= [1336,4004;451,3104;1416,3092;2349,3082;439,1897; ... 1399,1887;2347,1878;441,610;1475,599;2331,590]; FITC150EHO{5}= [1486,4214;577,3166;1551,3154;2492,3143;597,1954; ... 1565,1942;2488,1931;579,684;1592,672;2480,663];
FLUOROEFO{1}= [1389,3665;472,2674;1427,2666;2368,2658;451,1447; ... 1431,1440;2368,1428]; FLUOROEFO{2}= [1020,3823;132,2903;1085,2890;2035,2879;114,1700; ... 1096,1691;2028,1682]; FLUOROEFO{3}= [1043,2439;124,1562;1075,1554;2003,1542;100,315; ... 1062,305;2017,295]; FLUOROEFO{4}= [1487,2545;580,1613;1517,1603;2466,1591;542,357; ... 1517,349;2472,340]; FLUOROEFO{5}= [1068,2413;152,1489;1108,1478;2045,1468;128,286; ... 1109,275;2053,265]; FLUOROEFO{6}= [1480,2612;562,1680;1521,1670;2471,1660;546,513; ... 1521,505;2471,493]; FLUOROEFO{7}= [1420,2612;496,1667;1450,1658;2399,1645;475,500; ... 1434,489;2395,477];
FLUOROEHO{1}= [1413,4508;535,3358;1483,3350;2426,3340;508,2132; ... 1483,2123;2428,2119;492,850;1503,839;2414,832]; FLUOROEHO{2}= [1360,4305;462,3192;1440,3180;2378,3169;470,1904; ... 1440,1895;2375,1884;440,617;1450,609;2353,599]; FLUOROEHO{3}= [1353,4151;440,3205;1401,3195;2341,3184;440,1974; ...
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 312 Appendix I. Matlab code I.3. Coordinates
1401,1962;2336,1951;418,693;1458,683;2321,672]; FLUOROEHO{4}= [1323,4141;387,3083;1339,3074;2281,3064;363,1871; ... 1340,1860;2275,1849;357,598;1364,588;2256,578]; FLUOROEHO{5}= [1386,4177;498,3212;1456,3203;2398,3193;488,1997; ... 1454,1987;2397,1976;468,719;1420,709;2387,699]; FLUOROEHO{6}= [1396,3544;477,2651;1445,2640;2378,2557;460,1375; ... 1430,1365;2362,1352;445,96;1445,85;2353,77];
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 313 Appendix I. Matlab code I.4. Collage
I.4 Collage
This program collates images of the Counter Flow Diffusion Chamber into a single image. This program allows the user to input a matrix which specifies the geometrical arrangement of the images to be collated.
function out = Collage(order,x) %order should be the order that the images should be collated in %note that 1 does not mean the first image, but instead equates to %Cal 1.tif and 2 equates to Cal 2.tif. Blanks should be set to 0. %###############################% %% % 8520 % % 9631 % % 10740 % %% %###############################%
%x should be 2D vector with the first column listing the y co−ordinates %and the second column listing the x co−ordinates
%order = collage10; %x = FITC20EFO{6}; iw = 1388; ih = 1040;
[FileName, PathName] = uigetfile('*.tif','Select the tiff−file', ... 'MultiSelect', 'on'); col = zeros(4000,6000); for i = 1:length(FileName) if isnan(str2double(FileName{i}(6))) == 0
fileid = str2double(FileName{i}(5))*10 ... + str2double(FileName{i}(6));
Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 314 Appendix I. Matlab code I.4. Collage
else fileid = str2double(FileName{i}(5)); end img{fileid} = imread([PathName, FileName{i}]); end for i = 1:size(order,1) for j = 1:size(order,2) %:−1:1
if order(i,j) =0 col(x(order(i,j),1):x(order(i,j),1)+ih−1, ... x(order(i,j),2):x(order(i,j),2)+iw−1) = img{order(i,j)}; end %col(x(i,1):x(i,1)+ih−1,x(i,2):x(i,2)+iw−1) = img{i}; end end for i = 1:size(order,1)−1 for j = 1:size(order,2)
if order(i,j) =0 && order(i+1,j) =0 %height, width h(i,j) = ih−abs(x(order(i,j),1)−x(order(i+1,j),1))−1; w(i,j) = iw−abs(x(order(i,j),2)−x(order(i+1,j),2))−1;
g{i,j} = zeros(h(i,j)+1,w(i,j)+1); for m = 0:w(i,j) for n = 0:h(i,j) g{i,j}(n+1,m+1) = (h(i,j)−n)/h(i,j); end end if x(order(i,j),2) =g{i,j}.*double(img{order(i,j)}(ih−h(i,j) ... :ih,iw−w(i,j):iw))+(1−g{i,j}) ... .*double(img{order(i+1,j)}(1:1+h(i,j),1:1+w(i,j))); else col(x(order(i+1,j),1):x(order(i+1,j),1)+h(i,j), ... x(order(i,j),2):x(order(i,j),2)+w(i,j)) ... Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 315 Appendix I. Matlab code I.4. Collage =g{i,j}.*double(img{order(i,j)}(ih−h(i,j) ... :ih,1:1+w(i,j)))+(1−g{i,j})... .*double(img{order(i+1,j)}(1:1+h(i,j),iw−w(i,j):iw)); end end end end imwrite(uint16(col), [PathName, 'Collage.tif']) %figure;imshow(uint16(col)) end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 316 Appendix I. Matlab code I.5. AllignInlet I.5 AllignInlet This program aligns the inlet images of the different flow rates such that features occur at the same (x,y) coordinates and determines the pixel intensity of the fluid at the device inlet (i.e. the fluid being pumped into the device). This program detects the horizontal and vertical edges of features in the image and aligns them accordingly to each other. Once this is achieved, the mean pixel intensity of a pre-determined region near the inlet is calculated. Knowing what the concentra- tion is in this pre-determined region, it is therefore possible to determine the pixel intensity value for the input concentration. Given the concentration curves calculated in GetGain, and the pixel intensity value at the input, it is then possible to determine the concentration at other points in the device based on the pixel intensity. PathName = strcat('D:\Docs\Thesis\Pics\Fluorescein 06 10 08\'); FileName{1} = strcat('0 5ul min'); FileName{2} = strcat('0 25ul ml'); FileName{3} = strcat('1ul min'); FileName{4} = strcat('2ul min'); FileName{5} = strcat('4ul min'); clear img; clear stat; clear stat2; clear temp; %res is equivelent to the leading black space, for images that are too far %to the right/bottom they will have a leading black of les than res. res %must at least as large as as the largest positive offset of images that %are too far to the right/bottom res = 50; for i = 1:5 img{i,1} = imread([PathName, FileName{i}, '\Cal inlet.tif']); stat{i} = mean(double(img{i,1}(:,550:600)),2)/48000*5; for j=1:1000 Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 317 Appendix I. Matlab code I.5. AllignInlet if stat{i}(j)>1 break end end temp = zeros(2000, 1388); diff = 25−j; temp(1+diff+res:1040+diff+res,:) = img{i,1}; img{i,2} = temp(1:1200,:); end res = 220; for i = 1:5 temp = zeros(1200, 1700); stat2{i} = mean(double(img{i,2}(400:450,:)),1)/48000*5; for j=1200:−1:1 if stat2{i}(j)>1 break end end diff = 950−j; temp(:,1+diff+res:1388+diff+res) = img{i,2}; %Change this to 1:5200 instead of 1:4000 for Collage2 img{i,3} = temp(:,1:1500); end for i=1:5 %figure;imshow(uint16(img{i,3})); %imwrite(uint16(img{i,3}), [PathName, FileName{i}, ... %'\Cal inlet Alligned.tif']) pix(i) = mean(mean(img{i,3}(140:200,1070:1120))); end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 318 Appendix I. Matlab code I.6. AllignCollage I.6 AllignCollage This program is used to align the collages of the varying flow rates such that features in those collages occur at the same (x,y) coordinates. This initial program was used to align the simple collages. PathName = strcat('D:\Docs\Thesis\Pics\Fluorescein 06 10 08\'); FileName{1} = strcat('0 5ul min'); FileName{2} = strcat('0 25ul ml'); FileName{3} = strcat('1ul min'); FileName{4} = strcat('2ul min'); FileName{5} = strcat('4ul min'); clear img; clear stat; clear stat2; clear temp; const maxwidth = 3500; const maxheight = 3500; for i = 1:5 img{i} = imread([PathName, FileName{i}, '\Collage3.tif']); stat{i} = mean(double(img{i}(:,550:600)),2)/48000*5; for j=1:1000 if stat{i}(j)>1 break end end temp = zeros(maxheight*1.1, maxwidth); diff = 660−j; if diff > 0 temp(1+diff:min(maxheight,length(img{i})), 1:size(img{i},2)) ... = img{i}(1:min(maxheight,length(img{i}))−diff,:); else temp(1:length(img{i}(1−diff:min(maxheight−diff, ... length(img{i})),:)), 1:size(img{i},2)) ... Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 319 Appendix I. Matlab code I.6. AllignCollage = img{i}(1−diff:min(maxheight−diff,length(img{i})),:); end img2{i} = temp(1:maxheight,1:maxwidth); end clear img; clear temp; clear stat; for i = 1:5 temp = zeros(maxheight, maxwidth); stat2{i} = mean(double(img2{i}(200:250,:)),1)/48000*5; for j=1:1500 if stat2{i}(j)>1 break end end diff = 450−j; if diff > 0 temp(:,1+diff:1500+diff) = img2{i}; else temp(:,1:1500+diff) = img2{i}(:,1−diff:1500); end %Change this to 1:5200 instead of 1:4000 for Collage2 img3{i} = temp(1:maxheight,1:maxwidth); end for i=1:5 imshow(uint16(img3{i})); %imwrite(uint16(img{i,3}), [PathName, FileName{i}, ... %'\Collage3 Alligned.tif']) end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 320 Appendix I. Matlab code I.7. AllignCollage2 I.7 AllignCollage2 This program is used to align the collages of the varying flow rates such that features in those collages occur at the same (x,y) coordinates. This program improves upon AllignCollage and is used for collating the more complex collages. PathName = strcat('D:\Docs\Thesis\Pics\Fluorescein 06 10 08\'); FileName{1} = strcat('0 25ul ml'); FileName{2} = strcat('0 5ul min'); FileName{3} = strcat('1ul min'); FileName{4} = strcat('2ul min'); FileName{5} = strcat('4ul min'); clear img; clear stat; clear stat2; clear temp; % buffer acts to allow images that have a positive displacement to be % shifted to the left or upwards. Ensure that buffer is larger the the % maximum positive displacement. xbuffer = 300; ybuffer = 150; maxwidth = 3500; maxheight = 3500; %xscan is a vertical range (on the y axis) for which the prominent feature %on the x axis is likely to be found within. xscan = 600:610; %xallign is the point at which the prominent feature will be alligned to. xallign = 450; yscan = 300:310; yallign = 650; for i = 1:5 img{i} = imread([PathName, FileName{i}, '\Collage3.tif']); Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 321 Appendix I. Matlab code I.7. AllignCollage2 stat{i} = mean(double(img{i}(:,yscan)),2)/48000*5; for j=1:1000 if stat{i}(j)>0.5 break end end diff = yallign−j; temp = uint16(zeros(maxheight, maxwidth)); temp(1+ybuffer+diff:ybuffer+diff+size(img{i},1),1:size(img{i},2)) ... = img{i}; img{i} = temp(1:maxheight, 1:maxwidth); end clear stat; for i = 1:5 stat{i} = mean(double(img{i}(xscan+ybuffer,:)),1)/48000*5; for j=1:1000 if stat{i}(j)>0.8 break end end diff = xallign−j; temp = uint16(zeros(maxheight, maxwidth)); temp(1:size(img{i},1),1+xbuffer+diff:xbuffer+diff+size(img{i},2)) ... = img{i}; img{i} = temp(1:maxheight, 1:maxwidth); %figure;imshow(uint16(img{i})); end for i=1:5 %figure;imshow(uint16(img{i})); imwrite(uint16(img{i}(190:2300,300:2920)), [PathName, FileName{i}, ... '\Collage3 Alligned.tif']) end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 322 Appendix I. Matlab code I.8. ExitAngles I.8 ExitAngles This program rotates the channels of the pyramidal gradient generator so they are aligned horizontally. This allows comparison at the same position in different channels. It is used to identify the best position in the channel for analysis. It is desirable to exclude bubbles and non uniform regions which this program allows the user to do. %%%ExitAngles clear chan; clear img; PathName = 'D:\Docs\Thesis\Pics\Fluorescein 06 10 08\'; FileName{1} = strcat('0 25ul ml'); FileName{2} = strcat('0 5ul min'); FileName{3} = strcat('1ul min'); FileName{4} = strcat('2ul min'); FileName{5} = strcat('4ul min'); c = [1267 330 1660 119; 1296 529 2202 159; 1336 696 2460 386; ... 1454 862 2478 720; 1454 1028 2436 1032; 1347 1181 2422 1340; ... 1322 1362 2411 1637]; ang = atan((c(:,4)−c(:,2))./(c(:,3)−c(:,1))); L = round(sqrt((c(:,3)−c(:,1)).ˆ2+(c(:,4)−c(:,2)).ˆ2)); UpChannel = [920 1020; 2450 2500; 2880 2930; L(4)−100 L(4); ... L(5)−100 L(5); L(6)−100 L(6); L(7)−100 L(7)]−80; Mid = [200 700; 650 1600; 650 1500; 1200 2200; 1000 2000; ... 1000 2000; 1000 2000]−80; colour{1} = 'b'; colour{2} = 'c'; colour{3} = 'g'; colour{4} = 'm'; colour{5} = 'r'; Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 323 Appendix I. Matlab code I.8. ExitAngles for i = 1:5 img{i} = imread([PathName, FileName{i}, '\Collage3 Alligned.tif']); end for im = 1:5 for k = 1:7 if k==2 Lmax = L(k)+50; elseif k==3 Lmax = L(k)−20; else Lmax = L(k)−50; end for i = 1:Lmax for j = 1:146 m = sqrt(iˆ2+jˆ2); adang = atan(j/i); chan{k,im}(j,i) = img{im}(round(c(k,2)+m*sin(ang(k) ... −adang)),round(c(k,1)+m*cos(ang(k)−adang))); end end chan{k,im} = flipud(chan{k,im}); end end for k = 1:7 figure(k); hold on; for im = 1:5 stat = mean(chan{k,im},1); xax = (0:length(stat)−1)/0.4+80; plot(xax, stat/pix(im)*0.1, colour{im}) ustat(k, im) = mean(stat(round(UpChannel(k,1) ... *0.4:UpChannel(k,2)*0.4))); ustd(k,im) = std(stat(round(UpChannel(k,1) ... *0.4:UpChannel(k,2)*0.4))); Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 324 Appendix I. Matlab code I.8. ExitAngles mstat(k,im) = mean(stat(round(Mid(k,1)*0.4:Mid(k,2)*0.4))); mstd(k,im) = std(stat(round(Mid(k,1)*0.4:Mid(k,2)*0.4))); end xlabel('Distance from channel exit (\mu m)'); ylabel('Equivalent Concentration (\mu g/mL)'); if k<3 axis([80 max(xax) 0.055 0.11]); elseif k==3 axis([80 max(xax) 0.05 0.1]); elseif k==4 axis([80 max(xax) 0.04 0.08]); elseif k==5 axis([80 max(xax) 0.02 0.06]); elseif k==6 axis([80 max(xax) 0.01 0.03]); elseif k==7 axis([80 max(xax) 0 0.01]); end hold off; savefig(['D:\Docs\Thesis\NewLatex\7 results\figures\ExitGraph Chan', ... num2str(k)], k, 'pdf', '−cmyk') end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 325 Appendix I. Matlab code I.9. GraphExitVModel I.9 GraphExitVModel This program generates a concentration gradient for the nine exit channels. After the position has been determined using ExitAngles, the concentration at each of those positions is calculated. The concentration in each of the exit channels can be set to a specifically desired value by adjusting the inlet parameters such as which of the three inlets the fluorescent dye is connected to, the flow rate of the dye and the dye concentration. By knowing the relationship between the concentration in each of the exit channels and the concentration gradient across the region that encompasses the entire outlet channel width at the position where the cell inlet is, then the desired concentration gradient at the region that encompasses the entire outlet channel width at the position where the cell inlet is can be achieved. %%GraphExitVModel upad = zeros(2,5); colour{1} = 'b'; colour{2} = 'c'; colour{3} = 'g'; colour{4} = 'm'; colour{5} = 'r'; padustat = [ustat; upad]*0.1; padustd = [ustd; upad]*0.1; figure(1); hold on; for i = 1:5 padustat(:,i) = padustat(:,i)/pix(i); plot(padustat(:,i), [colour{i}, '+−']) %errorbar(padustat(:,i), padustd(:,i), [colour{i},'x−']) end plot(fliplr(con(1,:)), 'k+−') Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 326 Appendix I. Matlab code I.9. GraphExitVModel xlabel('Channel Number') ylabel('Concentration (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\UpChan Grad', ... 1, 'pdf', '−cmyk') padmstat = [mstat; upad]*0.1; padmstd = [mstd; upad]*0.1; figure(2); hold on; for i = 1:5 padmstat(:,i) = padmstat(:,i)/pix(i); plot(padmstat(:,i), [colour{i}, '+−']) %errorbar(padmstat(:,i), padmstd(:,i), [colour{i},'x−']) end plot(fliplr(con(1,:)), 'k+−') xlabel('Channel Number') ylabel('Concentration (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\MidChan Grad', ... 2, 'pdf', '−cmyk') fl = flow(1,:) * 60e9; totconm = zeros(5,1); totconu = zeros(5,1); for i = 1:5 for j = 1:7 totconm(i) = totconm(i) + mstat(j,i)/pix(i)*0.1*fl(j); totconu(i) = totconu(i) + ustat(j,i)/pix(i)*0.1*fl(j); end end ufact = ones(9,5); mfact = ones(9,5); for i = 1:9 ufact(i,:) = ufact(i,:).*0.1./transpose(totconu); mfact(i,:) = mfact(i,:).*0.1./transpose(totconm); end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 327 Appendix I. Matlab code I.9. GraphExitVModel figure(3); padustat2 = padustat.*ufact; hold on; for i = 1:5 plot(padustat2(:,i), [colour{i}, '+−']) %errorbar(padustat(:,i), padustd(:,i), [colour{i},'x−']) end plot(fliplr(con(1,:)), 'k+−') xlabel('Channel Number') ylabel('Concentration (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\UpChan GradNorm', ... 3, 'pdf', '−cmyk') figure(4); padmstat2 = padmstat.*mfact; hold on; for i = 1:5 plot(padmstat2(:,i), [colour{i}, '+−']) end plot(fliplr(con(1,:)), 'k+−') xlabel('Channel Number') ylabel('Concentration (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\MidChan GradNorm', ... 4, 'pdf', '−cmyk') temp = ones(9,5); for i = 1:5 temp(:,i) = fliplr(con(1,:)); end erroru = (temp − padustat2); errorm = (temp − padmstat2); figure(5); hold on; for i = 1:5 Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 328 Appendix I. Matlab code I.9. GraphExitVModel plot(erroru(:,i), [colour{i}, '+−']) end xlabel('Channel Number') ylabel('Absolute Error (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\UpChan Error', ... 5, 'pdf', '−cmyk') figure(6); hold on; for i = 1:5 plot(errorm(:,i), [colour{i}, '+−']) end xlabel('Channel Number') ylabel('Absolute Error (\mu g/mL)') hold off; savefig('D:\Docs\Thesis\NewLatex\7 results\figures\MidChan Error', ... 6, 'pdf', '−cmyk') Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 329 Appendix I. Matlab code I.10. TreeCalFix I.10 TreeCalFix This program collates pyramidal gradient generator images within the middle of the tree, and at the endpoint into a single image. %function y = TreeCalFix(gain) gain = Gain{1}; gain = ones(1040, 1388); FileName1 = {'Cal 1.tif', 'Cal 2.tif', 'Cal 3.tif', 'Cal 4.tif'}; FileName2 = {'Cal 5.tif', 'Cal 6.tif', 'Cal 7.tif', 'Cal 8.tif', ... 'Cal 9.tif', 'Cal 10.tif'}; FileName3 = {'Cal 11.tif', 'Cal 12.tif', 'Cal 13.tif'}; PathName = 'H:\Microscope Pictures\Gradients\Xmas Tree\Fluorescein 06 10 08\0 25ul ml\'; clear calimg1;clear g1; clear calimg2;clear g2; clear calimg3;clear g3; collage1 = uint16(zeros(4000,1500)); collage2 = uint16(zeros(5500,1500)); collage3 = uint16(zeros(3000,1500)); c1 = [55 45 36 27; 16 851 1687 2528]; c2 = [90 69 61 53 44 15; 22 864 1712 2557 3395 4171]; c3 = [1118 1110 50; 59 901 480]; gainfix = ones(1040,1388); for i = 1:1388 for j = 1:1040 %gainfix(j,i) = (255−(9*j/1040))/255; end end for i = 1:length(FileName1) calimg1{i} = uint16(round(gain.*gainfix. ... *double(imread([PathName FileName1{i}])))); Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 330 Appendix I. Matlab code I.10. TreeCalFix end for i = 1:length(FileName2) calimg2{i} = uint16(round(gain.*gainfix. ... *double(imread([PathName FileName2{i}])))); end for i = 1:length(FileName3) calimg3{i} = uint16(round(gain.*gainfix. ... *double(imread([PathName FileName3{i}])))); end for i = 1:3; collage1(c1(2,i):c1(2,i)+1039,c1(1,i):c1(1,i)+1387) = calimg1{i}; collage2(c2(2,i):c2(2,i)+1039,c2(1,i):c2(1,i)+1387) = calimg2{i}; collage3(c3(2,i):c3(2,i)+1039,c3(1,i):c3(1,i)+1387) = calimg3{i}; x1(i,1) = 1040−abs(c1(2,i+1)−c1(2,i)); %height x1(i,2) = 1388−abs(c1(1,i+1)−c1(1,i)); %width x2(i,1) = 1040−abs(c2(2,i+1)−c2(2,i)); %height x2(i,2) = 1388−abs(c2(1,i+1)−c2(1,i)); %width for j = 1:x1(i,1) for k = 1:x1(i,2) g1{i}(j,k) = (x1(i,1)−j)/(x1(i,1)); end end for j = 1:x2(i,1) for k = 1:x2(i,2) g2{i}(j,k) = (x2(i,1)−j)/(x2(i,1)); end end end for i = 4:5 x2(i,1) = 1040−abs(c2(2,i+1)−c2(2,i)); %height x2(i,2) = 1388−abs(c2(1,i+1)−c2(1,i)); %width for j = 1:x2(i,1) for k = 1:x2(i,2) g2{i}(j,k) = (x2(i,1)−j)/(x2(i,1)); end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 331 Appendix I. Matlab code I.10. TreeCalFix end end collage1(c1(2,4):c1(2,4)+1039,c1(1,4):c1(1,4)+1387) = calimg1{4}; collage2(c2(2,4):c2(2,4)+1039,c2(1,4):c2(1,4)+1387) = calimg2{4}; collage2(c2(2,5):c2(2,5)+1039,c2(1,5):c2(1,5)+1387) = calimg2{5}; collage2(c2(2,6):c2(2,6)+1039,c2(1,6):c2(1,6)+1387) = calimg2{6}; x3(1,1) = 1040−abs(c3(2,2)−c3(2,1)); %height x3(1,2) = 1388−abs(c3(1,2)−c3(1,1)); %width x3(2,1) = 1040−abs(c3(2,3)−c3(2,1)); %height x3(2,2) = 1388−abs(c3(1,3)−c3(1,1)); %width x3(3,1) = 1040−abs(c3(2,3)−c3(2,2)); %height x3(3,2) = 1388−abs(c3(1,3)−c3(1,2)); %width x3(4,1) = 1040−abs(c3(2,2)−c3(2,1)); %height x3(4,2) = 1388−abs(max(c3(1,1),c3(1,2))−c3(1,3)); %width for j = 1:x3(1,1) for k = 1:x3(1,2) g3{1}(j,k) = (x3(1,1)−j)/(x3(1,1)); end end for j = 1:x3(2,1) for k = 1:x3(2,2) %g3{2}(j,k) = (x3(2,1)−j+x3(2,2)−k)/(x3(2,1)+x3(2,2)); g3{2}(j,k) = ((x3(2,1)−j)/x3(2,1)+(k)/x3(2,2))/2; end end for j = 1:x3(3,1) for k = 1:x3(3,2) g3{3}(j,k) = ((j)/x3(2,1)+(k)/x3(2,2))/2; end end for j = 1:x3(4,1) for k = 1:x3(4,2) %g3{4}(j,k) = (x3(4,1)−abs(x3(4,1)−2j)+x3(4,2)−k) ... /(x3(4,1)+x3(4,2)); end end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 332 Appendix I. Matlab code I.10. TreeCalFix collage3(c3(2,2):c3(2,1)+1039,max(c3(1,1),c3(1,2)) ... :min(c3(1,1),c3(1,2))+1387) = uint16(g3{1} .* ... double(calimg3{1}((1041−x3(1,1)):1040,1:x3(1,2)))+ (1−g3{1}).* ... double(calimg3{2}((1:x3(1,1)),(1389−x3(1,2)):1388))); collage3(c3(2,3):c3(2,1)+1039,c3(1,1):c3(1,3)+1387) = uint16(g3{2} .* ... double(calimg3{1}((1041−x3(2,1)):1040,1:x3(2,2)))+ (1−g3{2}).* ... double(calimg3{3}((1:x3(2,1)),(1389−x3(2,2)):1388))); collage3(c3(2,2):c3(2,3)+1039,c3(1,2):c3(1,3)+1387) = uint16(g3{3} .* ... double(calimg3{2}((1:x3(3,1)),1:x3(3,2)))+ (1−g3{3}).* ... double(calimg3{3}((1041−x3(3,1):1040),(1389−x3(3,2)):1388))); %collage3(c3(2,i+1):c3(2,i)+1039,c3(1,i):c3(1,i+1)+1387) = %collage3(c3(2,i+1):c3(2,i)+1039,c3(1,i):c3(1,i+1)+1387) = ... %uint16(g3{i} .* double(calimg3{i}((1041−x3(i,1)) ... %:1040,1:x3(i,2)))+ (1−g3{i}).* ... %double(calimg3{i+1}((1:x3(i,1)),(1389−x3(i,2)):1388))); for i = 1:3 collage1(c1(2,i+1):c1(2,i)+1039,c1(1,i):c1(1,i+1)+1387) ... = uint16(g1{i} .* double(calimg1{i}((1041−x1(i,1)) ... :1040,1:x1(i,2)))+ (1−g1{i}).* ... double(calimg1{i+1}((1:x1(i,1)),(1389−x1(i,2)):1388))); collage2(c2(2,i+1):c2(2,i)+1039,c2(1,i):c2(1,i+1)+1387) ... = uint16(g2{i} .* double(calimg2{i}((1041−x2(i,1)) ... :1040,1:x2(i,2)))+ (1−g2{i}).* ... double(calimg2{i+1}((1:x2(i,1)),(1389−x2(i,2)):1388))); end for i = 4:5 collage2(c2(2,i+1):c2(2,i)+1039,c2(1,i):c2(1,i+1)+1387) ... = uint16(g2{i} .* double(calimg2{i}((1041−x2(i,1)) ... :1040,1:x2(i,2)))+ (1−g2{i}).* ... double(calimg2{i+1}((1:x2(i,1)),(1389−x2(i,2)):1388))); end %collage2(c2(2,5):c2(2,4)+1039,c2(1,4):c2(1,5)+1387) ... %= uint16(g2{4} .* double(calimg2{4}((1041−x2(4,1)) ... % :1040,1:x2(4,2)))+ (1−g2{4}).* ... Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 333 Appendix I. Matlab code I.10. TreeCalFix % double(calimg2{5}((1:x2(4,1)),(1389−x2(4,2)):1388))); imwrite(collage1, [PathName 'Collage.tif'], 'tif'); imwrite(collage2, [PathName 'Collage2.tif'], 'tif'); imwrite(collage3, [PathName 'Collage3.tif'], 'tif'); %end Design of Bioreactors for Mesenchymal Stem Cell Tissue Engineering 334