Differentiation of Human Mesenchymal Stem/Stromal Cells Into Myogenic Cells for Urethral Sphincter Muscle Engineering Biotechnol

Differentiation of Human Mesenchymal Stem/Stromal Cells into Myogenic Cells for Urethral Sphincter Muscle Engineering

Sara Ferreira Martins Gomes

Thesis to obtain the Master of Science Degree in Biotechnology

Supervisors: Professor Cláudia Alexandra Martins Lobato da Silva Professor Joaquim Manuel Sampaio Cabral

Examination Committee Chairperson: Professor Arsénio do Carmo Sales Mendes Fialho Supervisor: Professor Cláudia Alexandra Martins Lobato da Silva Member of the Committee: Professor Gabriel António Amaro Monteiro

December 2015

I dedicate this thesis to the memory of my father, Henrique Martins Gomes (November 4th, 1946 - May 11th, 2015)

iii

iv Acknowledgments

First, I would like to thank Professor Joaquim Cabral and Professor Cláudia Silva for making it possible for me to develop my dissertation work at the Stem Cell Bioengineering and Regenerative Medicine Laboratory (SCBL-RM), at IBB-BERG-IST, and for accepting to be my Supervisors. I am truly grateful for the opportunity to work in this field of study, which I have been passionate about since I was in high school. I also want to thank Professor Cláudia for all the guidance and encouragement and for all the trust put into my work and my ideas.

Second, I owe a big thank you to Irina Simões, who guided me since the beginning, taught me everything I needed to know about this project and was always accessible to help me even at thousands of kilometres away in Zurich. I could not have achieved the results I present here without her knowledge and direction, and I could not have been attributed a better mentor.

Third, I want to thank my colleagues at the SCBL-RM laboratory, who were always available to help me and give me guidance when I needed: Ana Fernandes, Márcia Mata, Francisco Moreira, Diogo Pinto, Raquel Cunha, Marta Costa, João Silva, Cláudia Miranda, Tiago Dias and Carlos Rodrigues. You all had an important part in the development of this project and of my laboratory skills.

I want to give a warm thank you to Alexandra Salvado, Ângela Neves and Mafalda Cavalheiro for being my partners in this journey, and for all the support and laughter that we shared over the last two years. Without you, group assignments would have been extremely dull and definitely not as productive as they were.

Last but definitely not least, I want to thank my family, especially the three most important gentlemen in my life: my boyfriend/best friend, my brother, and my father. Rui, you are my rock: you were there for me through the toughest times, holding me together and giving me a reason to fight for. My dear Mano, I want to thank you for all the wise words, support and guidance, especially over the last 6 months, and for encouraging me to always do my best. Dad, thank you for telling me how proud of me you were; those words are what keep me going in the hardest moments. I hope I keep making you proud. I deeply miss you.

vi Abstract

Stress urinary incontinence (SUI) is a medical condition that requires novel alternative therapies aiming to restore and maintain the integrity and function of the urethral sphincter, the muscle layer responsible for the normal continence mechanism. This MSc project targeted the establishment of effective myogenic differentiation protocols for Mesenchymal Stem/Stromal Cells (MSCs) for urethral sphincter engineering, with particular focus on exploring the myogenic potential of the Stromal Vascular Fraction (SVF) of the adipose tissue. The ability of MSCs to differentiate into smooth muscle cells (SMCs) and skeletal myofibers, the main constituents of the sphincter, has already been demonstrated in few in vivo and in vitro studies, but with little translation of this knowledge into clinical settings. The effects of 5-aza-2’-deoxycytidine (5-AZAd) and PD98059, chemical inducers of skeletal muscle and smooth muscle differentiation in MSCs, respectively, were tested herein. MSC differentiation into both cell types was evaluated by the detection of smooth and skeletal muscle lineage-specific markers by flow cytometry and immunofluorescence techniques at different timepoints in early cell passages (P<3). The expression of skeletal muscle markers was assessed in magnetically sorted and unsorted SVF cells coated on distinct substrates. Myogenesis-committed cells were identified in uncultured SVF, but no myoblast-like cells were isolated. Still, SVF-derived cells were shown to possess intrinsic myogenic potential that was enhanced when combined with culture substrates (in particular, gelatin coating), thus holding great potential for skeletal muscle engineering applications. Conversely, 5-AZAd supplementation failed to induce myogenesis and triggered severe cytotoxic effects, while PD98059 did not provide enough stimuli to sustain smooth muscle differentiation, which likely requires the use of 3-D culture conditions and/or biomechanical stimulation.

Keywords Stress Urinary Incontinence Mesenchymal Stem/Stromal Cells Stromal Vascular Fraction Myogenic Differentiation Smooth Muscle Cells Skeletal muscle Cells

vii

viii Resumo

A incontinência urinária de esforço é uma condição que requer novas terapias que visem restaurar e manter a integridade e função do esfíncter uretral, a camada de músculo responsável pelo mecanismo de continência normal. Este Projeto de Mestrado teve como objectivo o desenvolvimento de protocolos que permitam a diferenciação eficaz de células estaminais/estromais mesenquimais (ou mesenquimatosas) - CEMs - em linhagens miogénicas para reconstituição do esfíncter uretral, com especial foco no estudo do potencial miogénico da fracção vascular estromal (FVE) do tecido adiposo. Neste sentido, foi testado o efeito dos compostos 5-aza-2’-desoxicitidina (5-AZAd) e PD98059, indutores químicos de diferenciação em músculo esquelético e liso, respectivamente, em CEMs. A diferenciação de CEMs em ambos os tipos de músculo foi avaliada através da detecção de marcadores específicos de cada linhagem, por técnicas de citometria de fluxo e imunofluorescência em tempos de cultura distintos e em passagens celulares baixas (P<3). A expressão de marcadores de músculo esquelético foi avaliada também em células da FVE minimamente processada e isoladas magneticamente, plaqueadas sob superfícies revestidas com componentes da matriz extracelular (gelatina e fibronectina). Foram detectadas células miogénicas na FVE minimamente processada, no entanto não foram isoladas células com morfologia mioblástica. Ainda assim, foi mostrado que as células obtidas a partir da FVE são dotadas de um potencial miogénico intrínseco, reforçado na presença de substratos, sendo potencialmente exequível a aplicação destas células em estratégias de reconstituição de músculo esquelético. Por outro lado, a adição de 5-AZAd não levou à indução de miogénese e conduziu a efeitos citotóxicos, enquanto que a adição de PD98059 não foi suficiente para sustentar a diferenciação de CEMs em músculo liso, que provavelmente requer a implementação de condições de cultura em 3-D e/ou estimulação biomecânica.

Palavras-Chave Incontinência Urinária de Esforço Células Estaminais/Estromais Mesenquimais (ou Mesenquimatosas) Fração Vascular Estromal Diferenciação Miogénica Células do Músculo Liso Células do Músculo Esquelético

x Index

ACKNOWLEDGMENTS ...... V ABSTRACT ...... VII RESUMO ...... IX LIST OF FIGURES ...... XIII LIST OF TABLES ...... XVI ABBREVIATIONS LIST ...... XVII I. INTRODUCTION ...... 1 I.1 BACKGROUND ...... 1 I.1.1 Urinary Incontinence ...... 1 I.1.2 Regenerative Medicine ...... 3 I.1.3 Muscle Cells ...... 8 I.1.4 Mesenchymal Stem/Stromal Cells ...... 11 I.1.5 Tissue Engineering-derived Urethral Constructs in Clinical Trials ...... 17 I.2 MYOGENIC DIFFERENTIATION OF MSCS: CURRENT STATUS ...... 18 I.2.1 Differentiation of MSCs into Skeletal Muscle Cells ...... 18 I.2.2 Differentiation of MSCs into SMCs ...... 22 I.3 AIM OF STUDIES...... 26 II. MATERIALS AND METHODS ...... 27 II.1 HUMAN-DERIVED SAMPLES ...... 27 II.2 CULTURE MEDIA ...... 27 II.2.1 Ex-vivo expansion of ADSCs, BM-MSCs and SVF-derived cells ...... 27 II.2.2 Thawing and cryopreservation of SVF cells, ADSCs and BM-MSCs ...... 27 II.3 MULTILINEAGE DIFFERENTIATION ASSAYS...... 28 II.4 IMMUNOPHENOTYPE CHARACTERIZATION ...... 28 II.5 5-AZA-2’-DEOXYCYTIDINE TREATMENT ...... 29 II.5.1 Effect of 5-aza-2’-deoxycytidine on cell viability...... 29 II.5.2 Effect of 5-aza-2’-deoxycytidine on cell apoptosis and morphology ...... 29 II.6 CD34+ SVF-DERIVED CELL SORTING AND MYOGENESIS INDUCTION ...... 30 II.6.1 CD34/CD56 decay assessment in normal and ultralow attachment plates ...... 30 II.7 SMOOTH MUSCLE DIFFERENTIATION INDUCTION ...... 31 II.8 IMMUNOFLUORESCENCE STAINING ...... 31 II.9 INTRACELLULAR STAINING FOR FLOW CYTOMETRY ...... 32 II.10 DECELLULARIZATION OF PORCINE URETHRAS ...... 32 III. RESULTS AND DISCUSSION ...... 33 III.1 MESENCHYMAL STEM/STROMAL CELL ISOLATION...... 33 III.1.1 Mesenchymal Stem/Stromal Cell Characterization ...... 35 III.2 DIFFERENTIATION OF MESENCHYMAL STEM/STROMAL CELLS INTO SKELETAL MUSCLE CELLS ..... 38 III.2.1 Myogenic induction using 5-aza-2’deoxycytidine ...... 38 III.2.2 Myogenic induction of CD34+ cells from the Stromal Vascular Fraction ...... 46 III.3 DIFFERENTIATION OF MESENCHYMAL STEM/STROMAL CELLS INTO SMOOTH MUSCLE CELLS ...... 63 III.4 DECELLULARIZATION OF PORCINE URETHRAS ...... 70 IV. FUTURE TRENDS AND CONCLUSIONS ...... 73 V. REFERENCES ...... 75

xii List of Figures

Figure I.1. The lower urinary tract in women and men. From Fry et al., 2009...... 1 Figure I.2. Cross-section of the upper portion of a female sphincter. From Heesakkers and Gerretsen, 2004...... 2 Figure I.3. The two main strategies involving stem cells in Regenerative Medicine-related clinical applications. From Schmitt et al., 2012...... 4 Figure I.4. The fertilized egg and its immediate progeny (until the 8-cell embryo) are totipotent cells. The cells from the inner mass of the blastocyst are pluripotent, as they can only give rise to the 3 germ layers of the embryo. Once cultured in vitro, pluripotent stem cells are capable of differentiating into neural cells (ectoderm), cardiac muscle or blood cells (mesoderm), among others. From the website of Division of Biology and Medicine from Brown University (http://biomed.brown.edu).28 ...... 5 Figure I.5. Transcription factors involved in the regulation of myogenic lineage progression. While Pax3/7, Myf5 and MyoD are crucial for the initial specification and commitment of the progenitor cells, MyoG and Mrf4 are determinant for the early and late differentiation steps, leading to mature myotubes/myofibers. From Bentzinger et al., 2012...... 9 Figure I.6. The Mesengenic Process: MSCs proliferate and their progeny can be induced to enter one of several mesenchymal lineage pathways, including marrow stroma, osteogenesis, chondrogenesis, tendogenesis, myogenesis and adipogenesis. Adapted from Caplan and Correa, 2011...... 12 Figure I.7. The MSC perivascular niche hypothesis. Within the niche, O2 tension is variable. MSC in contact with blood vessels (BV) would interact with (1) various other differentiated cells (DC1, DC2, etc.) by means of cell-adhesion molecules, (2) ECM deposited by the niche cells (mediated by integrin receptors), and (3) signalling molecules. From Kolf et al., 2007...... 16 Figure I.8. FACS vs. MACS. A) A fluorescence-activated sorter can be used as a preparative tool to separate fluorescently-labelled cells from a heterogeneous cell suspension. B) Through magnetic labelling, desired cells are firstly separated from contaminants and then eluted from the column. Left image from the website of Midlands Technical College (www.midlandstech.edu).116 Right image adapted from the website of Humboldt University of Berlin (http://edoc.hu-berlin.de/)117 . 20 Figure I.9. Structures of cytidine and its analogs, 5-AZA and 5-AZAd. R=ribose. dR=deoxyribose. Adapted from Christman, 2002...... 21 Figure I.10. The ERK/MAPK signalling pathway. Its activation triggers the sequential phosphorylation of three kinases (MAPKKK,MAPKK and MAPK), resulting in the translocation of ERK to the nucleus and the activation of transcription factors (e.g. Elk-1 and c-Myc) which control the expression of genes that are required for cell growth, differentiation and survival. From Kim and Bar-Sagi, 2004...... 24 Figure III.1. P0 SVF-derived cells cultured for 6 days in DMEM medium supplemented with 10% of (A) MSC-qualified FBS and (B) standard FBS. Arrow points to a pericyte-like cell...... 34 Figure III.2. P0 SVF-derived cells cultured for 15 days in DMEM+3%FBS (A) and 11 days in DMEM+10%FBS (B). Some cells begin appearing more extended and flattened with increased culture time, possibly due to senescence...... 34 Figure III.3. Multilineage differentiation characterization of BM-MSCs, ADSCs and AT-SVF cells. BM- MSCs and ADSCs were isolated and plated in DMEM+10%FBS (MSC-qualified), while AT-SVF cells were isolated in DMEM media supplemented with 20%FBS (not MSC-qualified), dexamethasone and bFGF, before confluence was reached and the medium was replaced to the respective differentiation medium. Scale bar = 100 μm...... 36 Figure III.4. Effect of 5-AZAd on cell morphology and viability. ADSCs were cultured for 7 days in DMEM+2%HS, harvested and placed on a haemocytometer for the trypan blue exclusion test. A) Control cells displayed typical round shapes; B) Cells exposed to 10 μM of 5-AZAd displayed a very thin elongated morphology and stained positively for trypan blue...... 39 Figure III.5. Cell viability assay for ADSCs cultured in DMEM+2%HS previously exposed to 5-AZAd. Cells were treated with various concentrations of 5-AZAd for 24 hours (2, 5, 7.5 and 10 μM) and cell viability was measured by counting cell numbers at 1, 3, 5, 7, 12 and 14 days thereafter. Two cell densities were tested: 3000 cells/cm2 (left) and 5000 cells/cm2 (right). No cell bars indicate zero cell viability, except for 5 μM...... 39 Figure III.6 Effect of 5-AZAd on cell viability...... 41

xiii Figure III.7. Flow cytometry dot plot charts of ADSCs cultured for 2 days in DMEM+2%HS after 5- AZAd treatment. A) control ADSCs; B) ADSCs treated with 10 μM 5-AZAd...... 41 Figure III.8. Alterations in cell morphology and confluence of ADSCs cultured in DMEM+2%HS after two weeks of 5-AZAd treatment. (A) Control ADSCs and ADSCs exposed to 2 μM (not shown) reached high confluence levels and maintained a fibroblast-like shape. (B) ADSCs exposed to 10 μM and 7.5 μM (not shown) of 5-AZAd achieved low confluence levels and two types of cell shapes could be discerned: thin elongated cells (black arrow) and broad irregular cells (white arrow)...... 42 Figure III.9. Effect of 5-AZAd on cell morphology of ADSCs. Control ADSCs and ADSCs treated with 2 μM 5-AZAd were positive for phalloidin-stained F-actin (red), and DAPI-stained nuclei (blue) at days 2 (top), 7 (middle) and 14 (bottom). ADSCs treated with 10 μM 5-AZAd stained positively at day 2, but no fluorescence was detected thereafter. Scale bar = 100 μm...... 43 Figure III.10. 5-AZAd effect on surface marker expression after 2, 7 and 14 days of treatment in ADSCs cultured in DMEM+2%HS...... 44 Figure III.11. Evaluation of Pax7 expression in an uncultured SVF sample. (A) Gated flow cytometry dot plot chart in which 3 subpopulations could be discerned; (B) Pax7 expression values and number of Pax7-expressing cells in a total of 49.6x105 viable cells (C) Pax7 expression histograms for subpopulations 1, 2 and 3 (left to right). Red lines = control; Blue lines = sample...... 46 Figure III.12. Flow cytometry dot plot charts of CD34-enriched (A) and CD34-depleted (B) fractions acquired after MACS for sample SVF-2. (C) Expression of CD34 and CD56 (in percentage and MFI) for both fractions...... 48 Figure III.13. Flow cytometry dot plot charts of unsorted SVF cells (A), and cells from the CD34- enriched (B) and CD34-depleted (C) fractions obtained after CD34 sorting for sample SVF-3. Two subpopulations (1 and 2) could be discerned in the CD34-enriched and CD34-depleted fractions. (D) Expression of CD34 and CD56 (in percentage and MFI). (E) Extracellular marker characterization of cells from the CD34-depleted fraction...... 48 Figure III.14. Flow cytometry dot plot charts of unsorted SVF cells (A) CD34-enriched (B) and CD34- depleted (C) fractions obtained after CD34 sorting of a cell pool of SVF cells. Two subpopulations (1 and 2) could be discerned in the CD34-enriched fraction. (D) Expression of CD34 and CD56 (in percentage and MFI)...... 49 Figure III.15. Evaluation of Pax7 expression by flow cytometry in an uncultured SVF sample (SVF-4) and in the enriched and depleted fractions obtained after CD34 MACS. (A) Gated flow cytometry charts; (B) Flow cytometry histograms of Pax7 expression in the unsorted SVF (left) and the CD34-enriched fraction (right); (C) Frequency, cell number and expression for each gated population and number of Pax7-expressing cells in a total of 48.6x105 viable SVF cells, 5.7x105 CD34 positive cells and 3.3x105 CD34 negative cells...... 50 Figure III.16. Cells cultured on fibronectin-coated (A) and non-coated (B) plates, using LG DMEM media, derived from the CD34-enriched fraction of the SVF, acquired by MACS. Low confluence areas (left) and high confluence areas (right) could be obtained in all conditions, except for unsorted SVF cells, which were 100% confluent. Black arrows point to pericyte-like cells...... 51 Figure III.17. Flow cytometry analysis of CD34 and CD56 expression for SVF and cells cultured on non-coated plates, after 11 days in culture, and cells cultured on gelatin and fibronectin-coated plates, after 12 days in culture. Top: Flow cytometry dot plot charts. Bottom: Expression percentages and MFI values for CD34 and CD56 for each condition. MP = Main population (higher SSH and FSH gate), SP = Subpopulation (lower SSH and FSH gate)...... 52 Figure III.18. Expression of myogenic markers in P0 cells cultured on gelatin- and fibronectin-coated plates (HG) after 12 days in culture. Flow cytometry dot plot charts and histograms of Pax7, MyoD and myogenin (left to right) of the main population gated of cells cultured on fibronectin- (HG) (A) and gelatin-coated plates (B). (C) Myogenic marker expression (in percentage and MFI values) of the gated main populations and subpopulations of both conditions...... 53 Figure III.19. Cells displaying distinct morphologies after 8 days of induction, detected by light microscopy (top) and fluorescence microscopy (bottom). (A) Cells cultured on fibronectin-coated plates (LG), in which aligned spindle-like and broad flattened cells can be discerned; (B) SVF cells, in which the development of several vacuoles is visible; (C) cells cultured on gelatin-coated plates and (D) cells cultured on non-coated plates; Broad flattened cells with enlarged nuclei (frequently binucleated) could be seen in all conditions, as well as smaller spindle-like cells. Nuclei were stained with DAPI and the cytoskeleton (F-actin filaments) was stained with phalloidin-TRITC...... 54

xiv Figure III.20. Multinucleated cells (P1) present in a very low confluence area of a gelatin-coated well, cultured for 17 days, detected by DAPI and Phalloidin-TRITC staining...... 55 Figure III.21. MyoD expression assessed by immunofluorescence after 2, 4 and 8 days of medium replacement for all conditions (secondary antibody: goat anti-rabbit Alexa 546). From day 4 to day 8 a decrease in the fluorescence intensity and the number of fluorescent cells was seen. Phalloidin-TRITC and DAPI staining (bottom row) allowed the observation of aligned nuclei and some fused cells. No Pax7 and myogenin expression was observed through the assay for all the conditions. Scale bar = 50 μm...... 56 Figure III.22. Flow cytometry analysis of CD34 and CD56 expression after 2 and 3 days in culture for cells isolated from unsorted SVF and magnetically sorted cells from the CD34-enriched fraction...... 58 Figure III.23. Aggregate adherence to gelatin-coated plates after 24 hours in culture. Left to right: unsorted SVF cells in expansion media; unsorted SVF cells in skeletal muscle cell growth media; CD34-enriched cell fraction in expansion medium; CD34-enriched cell fraction cell in skeletal muscle cell growth media. Scale bar = 100 μm...... 59 Figure III.24. Bladder SMCs (passaged 6) plated in bladder smooth muscle cell growth media after 8 days (A) and 21 days (B) of culture...... 64 Figure III.25. Smooth muscle marker expression in cultured SMCs. A) Flow cytometry dot plot chart of bladder SMCs (P5) after 14 days in culture and B) percentage values of expression of smooth muscle markers. Secondary antibody: goat anti-mouse Alexa488. C) Immunofluorescence for markers α-SMA, SM-MHC, calponin and desmin detected in P6 bladder SMCs after 21 days in culture. Scale bar = 50 μm. Secondary antibodies: goat anti-mouse Alexa488 for α-SMA, calponin and SM-MHC; goat-anti rabbit Alexa 546 for desmin...... 64 Figure III.26. BM-MSCs (A) and SVF-derived cells (B) plated in DMEM+10%FBS, cultured for 3 and 9 days, respectively...... 65 Figure III.27. Smooth muscle marker expression in AT-SVF cells and BM-MSCs after 2, 4 and 7 days of differentiation induction with PD98059, assessed by flow cytometry and immunofluorescence. A) Intracellular flow cytometry for α-SMA, calponin, SM-MHC and desmin. Secondary antibodies used: goat anti-mouse Alexa488 (in BM-MSCs assay); goat anti-mouse PE for α-SMA, calponin and SM-MHC and goat anti-rabbit Alexa 488 for desmin (in AT-SVF assay). B) Immunofluorescence after 7 days of induction for calponin (top) and desmin (bottom) in AT-SVF cells. C) Immunofluorescence after 7 days of induction for calponin in BM-MSCs. Secondary antibodies: goat anti-mouse Alexa488 for calponin and goat-anti rabbit Alexa 546 for desmin. Scale bar = 50 μm...... 67 Figure III.28. (A) Schematics of the decellularization apparatus based on mechanochemical action. Briefly, the detergent solution (or distilled water) is placed inside the (a) vessel where the (b) cannulated urethra is submerged. The solution is pumped by a (c) peristaltic pump and re- circularized for 24 hours at a rate of 40 ml/min. Simultaneously, through a (d and e) magnetic stirrer plate the solution was agitated at 340 rpm. After 48 hours, the direction of perfusion was changed (from the black to the grey arrows). Adapted from Simões et al., 2015 (under revision). (B) Length and protocol duration for each decellularized urethra. (C) Main steps of the decellularization process and their outcomes for urethra #4. Biological tissue preparation includes AT and distal extremity removal, catheter placement and immobilization with sutures (all performed in day 0). Gradual whitening of the tissue over perfusion time occurred, indicating efficient cell removal...... 71

xv List of Tables

Table II.1. Panel of mouse anti-human monoclonal antibodies used to characterize MSCs and SVF- derived cells, their commercial brands, conjugated fluorophores and isotypes...... 29 Table II.2. Panel of anti-human primary monoclonal antibodies and respective fluorophore-conjugated secondary antibodies used to stain cells for markers relevant for myogenic differentiation assessment, dilution used, commercial brand and isotype...... 31 Table III.1. Age and gender of the AT donors and the assays performed with each sample...... 33 Table III.2. Expression of the markers used in the identification of SVF-derived cells and MSCs, according to Bourin et al. and Dominici et al., by several P0 SFV-derived plate-adherent populations cultured in distinct media, and by P0 BM-MSCs. For AT-SVF cells in expansion medium n=3 (except CD14 and HLA-DR, for which n=1). FBS*= MSC-qualified FBS ...... 35 Table III.3. Effect of 5-AZAd on cell viability. PI/Annexin V flow cytometry analysis for ADSCs cultured in DMEM+2%HS, treated with 2 and 10 μM of 5-AZAd, at days 2, 7 and 14 after treatment. The expression of annexin and PI served as a measure of cell viability, apoptosis and necrosis...... 41 Table III.4. 5-AZAd effect on surface marker expression after 2, 7 and 14 days of treatment in ADSCs cultured in DMEM+2%HS...... 44 Table III.5. Experiments performed with cells from different SVF donors. *Due to a shortage in SVF sample numbers, a cell pool combining samples SVF-4 and SVF-5 had to be performed...... 47 Table III.6. MACS yield for samples subjected to MACS. Total cell numbers of the SVF samples and the CD34 enriched and depleted fractions acquired after MACS were calculated by viable cell counts using the trypan blue exclusion test. The numbers of CD34+ cells within the samples and/or the positive fractions were calculated using values acquired by flow cytometry...... 48 Table III.7. Expression of myogenic markers (Pax7, MyoD, myogenin and skeletal MHC) in P1 cells after 2, 4 and 8 days of differentiation induction, assessed by flow cytometry. Secondary antibodies: Goat-anti mouse/rabbit Alexa 488 for myogenin (day 2) and MyoD (days 2 and 4), respectively; goat-anti mouse/rabbit PE for myogenin (days 4 and 8) and MyoD (day 8)...... 55

xvi Abbreviations List

ADSCs Adipose derived-Stromal/Stem Cells AT Adipose Tissue bFGF Basic Fibroblast Growth Factor BM Bone Marrow BM-MSCs Bone Marrow-derived Mesenchymal Stem/Stromal Cells CD Cluster of Differentiation DMEM Dulbecco's modified Eagle's Medium DNA Deoxyribonucleic acid CpG C-phosphate-G ECM Extracellular Matrix EGF Epidermal Growth Factor EMA European Medicines Agency ERK Extracellular-signal-Regulated Kinase FACS Fluorescence-Activated Cell Sorting FDA Food and Drug Administration FITC Fluorescein FSH Forward Scatter GSK-3β Glycogen Synthase Kinase-3β GvHD Graft versus Host Disease HDAC Histone Deacetylase HG High Glucose HGF Hepatocyte Growth Factor HLA Human Leukocyte Antigen HS Horse Serum HSCs Hematopoietic Stem Cells IGF Insulin-like Growth Factor iPSCs Induced Pluripotent Stem Cells ISCT International Society for Cellular Therapy ITS Insulin-Transferrin-Selenium LG Low Glucose MACS Magnetic-Activated Cell Sorting MAPK Mitogen-Activated Protein Kinase MSCs Mesenchymal Stem/Stromal Cells MHC Myosin Heavy Chain MRFs Myogenic Regulatory Factors Muse Multilineage-differentiating Stress-Enduring MyoG Myogenin PE Phyroerythrin PCR Polymerase Chain Reaction RNA Ribonucleic Acid SDS Sodium Dodecyl Sulfate SMCs Smooth Muscle Cells SM-MHC Smooth Muscle Myosin Heavy Chain SSH Side Scatter SRF Serum Response Factor SUI Stress Urinary Incontinence

xvii SVF Stromal Vascular Fraction TGF-β Transforming Growth Factor-β UCM-MSCs Umbilical Cord Matrix Stem Cells or Umbilical Cord-derived MSCs UI Urinary Incontinence 3-D Three-Dimensional 5-AZA 5- azacytidine 5-AZAd 5-aza-2’-deoxycytidine α-SMA Smooth Muscle Alpha Actin

xviii

I. INTRODUCTION I.1 Background I.1.1 Urinary Incontinence

Urinary incontinence (UI) affects more than 200 million people worldwide.1,2 Although not life-threatening, it is a common medical and social condition that dramatically reduces the quality of life of patients. In the United States of America, the total annual cost of UI and associated conditions was estimated to range up to 32 billion dollars in 2006.2,3 Stress urinary incontinence or SUI is the most prevalent type of UI.4 According to the International Continence Society, SUI is the complaint of involuntary leakage of urine from the urethra on effort or exertion, or on sneezing or coughing.1,5 SUI affects between 16% and 35% of adult women and the occurrence of this condition rises with increasing age.4,6,7 Although more prevalent in woman, SUI can affect men, mostly due to prostate surgery.2 Still, many incontinent patients do not report symptoms for various reasons, including embarrassment or fear of treatment, which might indicate that the prevalence of UI may be much higher than documented.1,6 Disguised UI is the cause of decreased self-esteem, neuroses, depression states and difficulties in sex life, leading to a profound impact on the psychological condition of the patient.8 The risk factors for the development of SUI include age, vaginal deliveries, genetic load (collagen deficiency), estrogens deficiency, diabetes mellitus, pelvic surgery, chronic constipation, chronic cough, obesity and heavy physical work.7,8

UI emerges from a malfunction of the lower urinary tract. The lower urinary tract comprises two regions: the urinary bladder and the outflow tract (bladder neck, or base, and urethra - Figure I.1).9

Figure I.1. The lower urinary tract in women and men. From Fry et al., 2009.

Urine is produced in the kidneys and stored in the bladder while the lumen of the urethra is tightly shut, occasionally opening for the voluntary elimination of urine through the urethra.2 Structurally, the bladder is composed mainly by smooth muscle and connective tissue.10 During urine

storage, the smooth muscle - detrusor - is relaxed, making the bladder a low pressure, high volume reservoir. The reverse happens during micturition, when the muscle layer becomes active, making the bladder a high pressure, contracting vesicle for urine expulsion.10

The urethra plays a key role in the continence mechanism. The urethra is a multilayered tubular structure that consists of smooth muscle with intrafascicular connective tissue, a rich submucosa (or lamina propria) with collagen fibers and microvascularization, and a lining epithelium (or urothelium).11,12 Additionally, the presence of a circular omega-shaped striated muscle layer that surrounds the smooth muscle, named rhabdosphincter or external urethral sphincter, in the mid-urethra in humans, allows the normal urethral closure mechanism (Figure I.2).11

Figure I.2. Cross-section of the upper portion of a female sphincter. From Heesakkers and Gerretsen, 2004.

The combined actions of the mentioned tissues serve to create wall tension that compresses the lumen closed, though the rhabdosphincter contributes the most for the continence mechanism.2,11 During urine storage, this layer is tonically active, becoming dynamically active during increases in abdominal pressure and inactive during normal micturition.10 Similarly, urethral smooth muscle cells (SMCs) exhibit tonic spontaneous contractile activity that allows urine to remain in the bladder against the forces of gravity, while micturition is assured by the relaxation of the urethral smooth muscle.13 SUI may arise from malfunctions of the rhabdosphincter, when bladder pressure exceeds urethral pressure, in the setting of sudden increases of intra-abdominal forces.1 In this way, leakage of urine from the urethra is caused by functional and morphological defects of the rhabdosphincter (e.g. a decrease in the density of the sphincter), which can arise, as mentioned before, from age-dependent spontaneous apoptosis of the striated muscle cells or from childbirth, maternal injury, or surgical interventions.14

The current treatments for SUI include non-invasive measures such as lifestyle changes (fluid intake management), pelvic floor muscle training and pharmaceutical agents.11 However, since SUI is fundamentally an anatomical problem, none of these measures fully corrects the underlying pathology, nor are these satisfactorily efficient. Moreover, the unwanted side-effects of some drugs

(e.g. cardiovascular issues and orthostatic hypotension) are not adequate for the elderly population and the expenses of new drugs often lead to discontinuation.1,6 On the other hand, the surgical approach to SUI involves suspension operations, placement of slings to support the urethra and periurethral and transurethral injection therapy, by injecting biomaterials (bulking agents) to increase urethral coaptation and outlet obstruction.15 Still, these procedures give rise to several complications and do not always lead to the correction of UI as they do not treat its pathophysiologic causes.6,14,16

Due to the limitations of current treatments, novel alternative approaches have been considered, such as Regenerative Medicine strategies, with the objective of restoring and maintaining the integrity and function of damaged/insufficient urethral muscle tissue.17 Furthermore, one of the objectives of Regenerative Medicine therapy for SUI is the reconstruction of the sphincter muscle itself, by resorting to stem cells.16 Clinical trials using injectable fibroblasts and muscle cells have been somewhat successful, with the purpose of treating atrophy of the urethral submucosa and defects of the rhabdosphincter, respectively.14 However, fibroblasts only act as a tissue bulking agent, through the production of extracellular matrix (ECM), and the ex vivo preparation of muscle precursor cells drastically reduces their myogenic potential in vivo.16,18 Animal studies suggest that mesenchymal stem/stromal cells (MSCs) – focused in Section I.1.4 – display the potential to regenerate both muscle and ganglion components of the sphincter, thus holding great potential for sphincter reconstruction.6,19 Importantly, a Phase II clinical trial aiming to test the therapeutic effects of the intralesional application of ex-vivo expanded adipose-derived cells for female SUI was already initiated.20

Artificial sphincters have been tested on patients; however, these devices are associated with a multitude of complications, including urethral edema and atrophy and erosion of the sphincter.6,21 Therefore, a combined strategy using cell therapy and proper scaffolds should be the ideal approach to reconstruct a functional/contractile urethral sphincter. Apart from its therapeutic potential, such construct would serve as a representative in vitro model of SUI, allowing researchers to test novel therapies, including drugs (i.e. drug screening) or cell therapy, and their effects in terms of toxicology and efficacy, before proceeding into clinical trials. Additionally, it would be important to better understand how specific damages in urethral support give rise to improper functioning of the urethral sphincter, consequently leading to incontinence, and the development of such research tool would allow to answer several questions that remain unanswered about SUI.11

I.1.2 Regenerative Medicine

In 1992, the term “Regenerative Medicine” was introduced by Leland Kaiser, who predicted that a new branch of medicine would develop to attempt to change the course of chronic diseases and in many instances would regenerate tired and failing organ systems.22 Regenerative Medicine comprises two main branches: cell therapy and tissue engineering (Figure I.3). Cell therapy aims to treat diseases or regenerate injured tissues through the injection of a specific cell population.23 The cells can either be isolated from the patient (autologous setting) or from donors (allogeneic setting). On the other hand, tissue engineering aims to construct tissues or organs for implantation, through the

combined use of stem/progenitor/adult cells, three-dimensional (3-D) scaffolds and/or soluble factors.23

Figure I.3. The two main strategies involving stem cells in Regenerative Medicine-related clinical applications. From Schmitt et al., 2012.

I.1.2.1 Stem Cells

Stem cells are undifferentiated cells that are defined by their functional capacity to self-renew as well as generate differentiated progeny.24,25 One example of a truly totipotent cell, capable of generating all cell types of a complete organism (including the extraembryonic tissues) is the fertilized egg, but also the cells that result from its first divisions.24 As the fertilized egg starts to asymmetrically divide, giving rise to the blastocyst, the cells from the epyblast (the inner mass of the blastocyst) are no longer totipotent, as they can only differentiate into cells from the three germ layers: mesoderm, ectoderm and endoderm (Figure I.4).24 These embryonic stem cells are classified as pluripotent, together with other cell types such as embryonic germ cells and embryonal carcinoma cells (isolated from teratocarcinomas) and, more recently, induced pluripotent stem cells (iPSCs).25 As their denomination suggests, iPSCs are obtained from committed cells that were artificially reprogrammed into a pluripotent state. The original method, developed by Yamanaka and co-workers, consisted in the transduction of skin cells with four genes (the pluripotency keepers Oct 3/4 and Sox2 and the oncogenes c-Myc and Klf4).26 The following term in the cell potency hierarchy is “multipotent”, and it is attributed to cells capable of differentiating into a limited range of cell types, within the same germ layer (e.g. hematopoietic stem cells or HSCs).24,25 These are also denominated as progenitor cells. Lastly, precursor cells such as angioblasts, hepatoblasts or myoblasts, are regarded as unipotent, as they can only sustain one cell lineage or cell type.25 When referring to stem cells obtained from any postnatal organism, the term “adult stem cell” is commonly applied in order to distinguish them from embryonic stem cells.24 For instances, MSCs (further discussed in Section I.1.4) are multipotent cells included in the adult stem cell group.

Stem cells in adult animals are extremely important to maintain tissue and organ mass during normal cellular turnover, with each tissue having a resident population of stem cells.24 It is still not clear if adult stem cells retain some plasticity and are able to go through a phenomenon called “transdifferentiation”, which is described to occur when stem cells derived from one adult tissue change into cellular phenotypes not normally found in that tissue.27 Although the legitimacy of this phenomenon has been highly debated, some authors claim that adult stem cells from mature tissues can be reprogrammed (“transdifferentiated”) to change into a completely distinct phenotype from that normally found in that tissue.27

Figure I.4. The fertilized egg and its immediate progeny (until the 8-cell embryo) are totipotent cells. The cells from the inner mass of the blastocyst are pluripotent, as they can only give rise to the 3 germ layers of the embryo. Once cultured in vitro, pluripotent stem cells are capable of differentiating into neural cells (ectoderm), cardiac muscle or blood cells (mesoderm), among others. From the website of Division of Biology and Medicine from Brown University (http://biomed.brown.edu).28

I.1.2.2 (Stem) Cell Therapy

Stem cells represent a promising cell therapy tool for Regenerative Medicine largely because of their ability to self-renew and differentiate into many functional cell types.29 Ideally, a stem cell therapy approach would comprise validated methods of stem cell isolation, proliferation and differentiation. Once isolated, and under the appropriate experimental conditions, stem cells could proliferate in order to attain clinically relevant numbers of undifferentiated cells.30 Subsequently, these undifferentiated cells could be also induced into lineage-specific differentiation through the implementation of defined in vitro differentiation protocols, generating the desired cell type.30 This approach allows the use of autologous stem cells, which is advantageous in the clinical setting, as

they do not induce immune responses and/or rejection.31 However, tissue-resident stem cells are a very small subpopulation of cells and they are difficult to isolate from most differentiated somatic cells present in postnatal tissues and organs.31

Currently, the use of stem cells for cell therapy as a well established and validated procedure is only performed in the treatment of hemato-oncological diseases. The transplantation of HSCs, one of the most successful examples of cell therapy, has been practiced for decades for this purpose.23 In bone marrow (BM) transplantation, HSCs are injected into the blood circulation of the recipient and find their way to the BM by a cytokine-mediated phenomenon termed “homing”.23 In the last decade, the use of cells derived from the umbilical cord blood ascended as a potential alternative cell source to BM transplantation due to advantages such as painless extraction procedure and lower incidence of graft-versus-host disease (GvHD).32

Besides BM transplantation that is fully implemented worldwide, some examples of approved cell therapy-based products include Prochymal (human adult MSCs expanded ex-vivo) and Holoclar (limbal epithelial stem cells). Prochymal was approved in Canada and New Zealand for paediatric use in acute GvHD33 - discussed in Section I.1.4.4 -, while Holoclar was approved by the European Medicines Agency (EMA) for the treatment of corneal diseases34.

Aside from ethical issues, the main reason why embryonic stem cell-derived therapy products have not been approved by regulatory agencies such as the United States Food and Drug Administration (FDA) is due to concerns related with the safety of transplanting embryonic stem cell-derived products.35 Cell products derived from pluripotent stem cells entail the possibility of spontaneous teratoma formation (i.e. residual undifferentiated stem cells) and the presence of unwanted cell types.35 Furthermore, (stem) cell therapy implicates other risks such as biological or donor-to-donor variability, lack of genetic, epigenetic and overall stability (a major hurdle when ensuring lot-to-lot consistency), microbiological contamination and immunological responses to alloantigens.35,36 Ultimately, until standards and methodologies for preclinical safety and efficacy evaluation - namely potency assays -37, product characterization, and process validation and control are fully developed, the approval and implementation of (stem) cell therapy is hindered.36

I.1.2.3 Tissue Engineering

Cell therapy alone is not sufficient to regenerate large tissue defects or even replace whole organs. In this regard, the approach of “tissue engineering” is the more promising strategy.23 Two main approaches are used in tissue engineering. The acellular approach involves the use of natural or synthetic matrices (often termed scaffolds) alone, while the cellular approach involves the use of cells, which are processed and seeded onto a scaffold (cell-seeded scaffold approach) before implantation.17

Tissue generation is a complex procedure comprising several steps, beginning with the isolation of cells from a biopsy, the expansion/differentiation of cells in vitro, the generation and maturation of a 3-D construct, and subsequently, the use of the construct as a test system or graft.38 One of the first successful cases of a tissue-engineered organ transplant occurred in 2008, in which a patient from Spain received a re-engineered trachea. Briefly, a donor trachea was first decellularized using a detergent (without denaturing the native ECM) and the resulting scaffold was re-cellularized in a rotating bioreactor using autologous epithelial cells and mesenchymal stem cell-derived chondrocytes, where cells were exposed to both air and liquid phases to mimic the environment of the organ.39 3-D shaped biomaterials - scaffolds - are of extreme importance for tissue engineering applications, since the properties of the substrate on which the cells are seeded influence cellular responses and cell fate.30 Furthermore, scaffolds should present various requirements, such as the presence of interconnected micropores (allowing cell seeding and proliferation, as well as vascular formation and waste transport), optimal porosity with adequate surface area and mechanical strength.40 Although synthetic/biosynthetic biomaterials and, most recently, nano-biomaterials, are used for stem cell differentiation, naturally-derived scaffolds are the most appropriate for developing tissue engineering constructs, mainly due to their biocompatibility. For instances, biomaterials like agarose, alginate, hyaluronic acid, gelatin (porous denaturated collagen), fibrin glue, collagen derivatives, and decellularized tissue matrices can be used as supporting scaffolds.30 In particular, decellularized matrices are biodegradable and bioresorbable (supporting the reconstruction of a completely normal tissue without significant inflammation) and provide efficient scaffolds for cell seeding, as well as inherent bioactivity and mechanical similarity to native ECM.12,30

After seeding the expanded cell population onto a tissue-specific scaffold (tissue generation phase), comes the maturation phase, when cells can adapt to the 3-D environment, produce ECM, establish cell-cell interactions and spread along and through the construct. During this phase, tissue development is guided by various cues, such as cell-matrix interactions, soluble factors and physical stimuli (mechanical, electrical and electromagnetic cues).38 The finalized tissue construct can be further used as an implant or as a test system (e.g. for drug screening or toxicology testing). One of the most relevant hurdles inherent to tissue engineering is tissue neovascularization, which is essential to supply oxygen and nutrients to the cells in the 3-D constructs. It is virtually impossible to expect the neovascularization throughout a cell–scaffold construct in the case of in vitro tissue engineering.40 As a consequence, the construction of entire vascularized organs in vitro is precluded by the complexity of generating new vessels. As of today, only cartilage-composed organs (e.g. ear)41 or relatively simple organs (e.g. trachea)39 have been constructed through tissue engineering approaches. Still, some tissue engineering-derived products are currently on the market, including autologous cultured chondrocytes and allogeneic fibroblast products (i.e. Carticel and Dermagraft, for cartilage and skin replacement, respectively).35

I.1.3 Muscle Cells

As mentioned previously, the urethral sphincter is populated by (i) SMCs on the inner layer and (ii) skeletal myofibers on the outer layer. Given the importance of the urethra musculature, in order to construct an engineered urethral tissue, to use as a tool for urethral cell therapy, functional layers of both smooth and skeletal muscle cells must be generated via tissue engineering-related approaches.

I.1.3.1 Smooth Muscle Cells

Bladder smooth muscle fibers have a fusiform shape, are 300-400 m in length and are of the single unit (visceral) type (i.e. become excited and contract as a single unit).42 SMCs in general can be derived from eight different origins: neural crest, secondary heart field, somites, various types of stem cells, mesangioblasts, proepicardium, splanchnic mesoderm and mesothelium.43,44 Bladder smooth muscle differentiation is regulated by epithelial-mesenchymal interactions and a variety of factors and signalling pathways.42 An epithelial signal is necessary for the induction of smooth muscle differentiation from the bladder mesenchyme, which seems to govern the pattern of this muscle.42 Among the factors that regulate SMC differentiation, sonic hedgehog (Shh), expressed by the bladder urothelium, is one of the most important. Transforming growth factor-β (TGF-), expressed in the developing bladder, and serum response factor (SRF), a key controller of muscle-specific gene expression, as well as some DNA-binding proteins, are also involved in bladder smooth muscle differentiation.42

In contrast to terminally differentiated skeletal myocytes (mentioned in Section I.1.3.2), SMCs retain remarkable plasticity and can undergo profound and reversible changes in their phenotype in response to changes in local environmental cues.45 This plasticity is based on the ability of SMCs to exhibit either a “synthetic” or a “contractile” phenotype. By switching to a “synthetic” phenotype - a process that can be dependent on arterial injury, growth factors and arteriogenesis - SMCs change their phenotypic markers and start secreting ECM proteins.45,46 Generally, differentiated SMCs (displaying a “contractile” phenotype) exhibit low proliferation rates, low synthetic activity, and express a unique repertoire of contractile proteins, ion channels, and signalling molecules required for the contractile function of these cells.43 Some of these molecules, especially contractile proteins, can therefore be used as markers for SMCs. Smooth muscle alpha actin (-SMA) is considered a marker of early SMC differentiation and the most general marker of the SMC lineage, although it is also expressed by myofibroblasts and endothelial cells under certain conditions.42,43,46 Smoothelin, a cytoskeletal protein found in mature SMCs, is absent in myofibroblast cells, providing a potential marker to distinguish between SMCs and fibroblasts.46 Calponin and SM22 (an actin-associated protein) are considered intermediate markers while smooth muscle myosin heavy chain (SM-MHC) is a marker of fully differentiated SMCs, thus being considered the late and most specific maker of differentiated SMCs.42

I.1.3.2 Skeletal Muscle Cells

The process of generating skeletal muscle – myogenesis – is used to describe de novo embryonic muscle formation (embryonic myogenesis) and the regeneration of injured muscle (adult myogenesis), which are very similar processes in terms of the regulatory molecules involved.47,48 Embryonic myogenesis is a highly controlled mechanism induced in the mesoderm of the embryo, implicating signalling pathways (mainly Notch and Wnt) and transcription factors, including Six1/4, Pax3/7 and the so-called myogenic regulatory factors (MRFs) MyoD, Myf5, myogenin (MyoG) and MRF4 (also known as Myf6), which are collectively expressed in the skeletal muscle lineage.47,48,49 During embryonic myogenesis, some embryonic progenitors specialize into satellite stem cells, while others skip this quiescent (i.e. not mitotically active) stage and directly differentiate into myoblasts.47 At this early stage in the lineage progression, cell fate is mainly regulated by transcription factors Six1/4, Pax3/7, Myf5 and MyoD (Figure I.5).

Figure I.5. Transcription factors involved in the regulation of myogenic lineage progression. While Pax3/7, Myf5 and MyoD are crucial for the initial specification and commitment of the progenitor cells, MyoG and Mrf4 are determinant for the early and late differentiation steps, leading to mature myotubes/myofibers. From Bentzinger et al., 2012.

Both quiescent and activated satellite cells express Pax7 (which is considered an evolutionary conserved specific marker), while Pax3 is only expressed in some muscle types.47,48,50 Non-committed satellite cells exhibit a Pax7+Myf5- phenotype whereas committed cells display Pax7+Myf5+.48,50,51 Upon activation, those committed but still quiescent satellite cells (Pax7+MyoD-) simultaneously upregulate MyoD, enter the cell cycle, and proliferate as Pax7+MyoD+ myoblasts. By downregulating Pax7 and upregulating MyoG and myosin heavy chain (MHC), myoblasts enter the differentiation

program (Pax7-MyoD+MyoG+).50,51 This last stage involves myoblasts aligning, giving rise to myocytes, and fusing homotypically to form multinucleated myofibers or myotubes.47,48 Once embryonic myogenesis is completed, the remaining satellite stem cells will reside in a specialized environment within skeletal muscle termed “satellite cell niche”, which serves as a pool of myogenic progenitor stem-like cells. The formation of this niche ensures muscle regeneration throughout the adult lifetime, by supporting the self-renewal of satellite cells (in response to stimuli such as injury) and preventing their differentiation.51,52 Interestingly, satellite cells have also been described as multipotent stem cells potentially capable of differentiating into other cell types belonging to the mesenchymal lineage, including adipocytes and osteocytes.53

Within the niche, satellite cells surround myofibers, lying between the plasmalemma and the basal membrane.47 Moreover, 95% of satellite cells are subjacent to a vessel and separated from it by the basement membrane, which allows their function to be modulated by other cell types not in direct contact with them.48,51,54 The differentiation of satellite cells in adult myogenesis, which occurs upon muscle injury, is highly dependent on extrinsic regulatory factors. Signalling molecules such as fibroblast growth factor (FGF), members of the TGF-β family, insulin-like growth factor-I and II (IGF-I/II), hepatocyte growth factor (HGF) and interleukins take part in muscle regeneration.47,48 Quiescent satellite cells express CD34, a surface marker implicated in cell adhesion and signalling; however, it is downregulated when cells are activated.48,51,55 CD56 (also known as neural cell adhesion molecule-1) is considered one of the most reliable markers of satellite and myoblast cells in human muscle (although it is not lineage specific), since it is expressed in quiescent cells and is further maintained during proliferation and differentiation.52,56

Although satellite cells exhibit tremendous regeneration potential in vivo, they highly depend on their niche environment to keep their proliferation ability.47,54 This niche “addiction” is a major hurdle when large cell numbers are required, for instances, for cell therapy purposes (generally 1-5 millions of cells/kg of body weight)57. After isolation, satellite cells can only divide a small number of times and rapidly differentiate into myoblasts irreversibly expressing Myf5, MyoD and MyoG.44,48,49 Other disadvantages of using these cells in cell therapy or tissue engineering settings include poor accessibility and low numbers of satellite cells obtained from muscle biopsies.47 Although myoblasts may seem a good alternative due to their proliferative capacity, which makes their expansion in vitro possible, this capacity is largely reduced as the number of rounds of cultivation (i.e. passages) increases, and so does their ability to fuse and produce myofibers.59 Allogeneic myoblasts transplanted into Duchenne muscular dystrophy patients have been, in early trials, largely inefficient owing to immune rejection, rapid death (presumed to be caused by anoikis induction, i.e. apoptotic death caused by reduced contact with ECM components)60, and limited intramuscular migration.52,58 Finally, the use of myofibers as therapeutic agents is precluded by their barely detectable postmitotic turnover in healthy young muscle and, consequently, lack of turnover in vitro.51 Therefore, an alternative cell type with both proliferative capacity and stability in vitro, as well

as myogenic differentiation potential, must be selected for cell therapy and tissue engineering applications involving skeletal muscle cells.

I.1.4 Mesenchymal Stem/Stromal Cells

In the early 1990s, Arnold Caplan at Case Western University used the term “Mesenchymal Stem Cell” to describe BM-derived cells with the ability to both self renew and differentiate along classical mesenchymal lineages, including adipocytes, chondrocytes, myocytes, osteoblasts, and tenocytes.61,62 The term “Mesenchymal Stromal Cells” is also valid, since MSCs belong to the marrow stroma, providing support and nourishment to HSCs. These cells were first reported by Friedenstein and colleagues in 1970, who identified the presence of fibroblast-like cells that could be obtained from the adult BM, form colonies on plastic and differentiate into mature cells of mesenchymal lineages, referring to them as Colony-Forming Unit-Fibroblasts (CFU-Fs).63,64

I.1.4.1 Definition and In Vitro Characteristics

According to the International Society for Cellular Therapy (ISCT), MSCs are defined by three criteria: (1) adherence to plastic; (2) specific surface antigen expression (positive for CD73, CD90 and CD105 in greater than 95% of cells in culture, but negative for hematopoietic markers CD14, CD34 and CD45, as well as CD11b, CD79a and human leukocyte antigen-D related (HLA-DR) surface markers); and (3) multipotent differentiation potential in vitro (capable of osteogenesis, adipogenesis and chondrogenesis) demonstrated by staining.65 Currently, these criteria are unanimously considered as extremely minimalist, as they presuppose that practically every non-clonal culture of cells from any connective tissue could potentially be classified as a MSC culture under the right culture conditions.66 Although not officially recognized by the ISCT, besides being able to differentiate into stromal cells, osteocytes, chondrocytes and adipocytes, MSCs are also capable of differentiating into myotubes in vitro (Figure I.6). 67–69

MSCs belong to the mesodermal lineage, the germ layer that includes blood, bone, BM, muscle, adipose tissue (AT) and connective tissues.24 Although MSCs are currently defined by their surface antigen expression, there is no cell surface marker that specifically and uniquely identifies MSCs. Other expressed markers include Stro-1 (which is associated with the capacity to form colonies, and, therefore, is gradually lost during culture expansion), CD200 (OX-2), adhesion molecules such as CD54, CD102 and CD50 (or ICAM-1,2 and 3, respectively), several growth factors/cytokine receptors and integrins.70–72

Remarkably, MSCs display a unique immune phenotype, described as MHC I+, MHC II-, CD40-, CD80-, CD86- and regarded as non-immunogenic.73 As mentioned above, MSCs are negative for CD45 (leukocyte common antigen) and HLA-DR antigens, suggesting that these cells can be immunoprivileged. In fact, MSCs appear to use an array of mechanisms to minimize rejection by the

host including modulation of dendritic and T cell function, as well as the creation of a suppressive microenvironment.74 Clinically, this immune phenotype represents a great advantage, as it theoretically allows the transplantation of allogeneic MSCs into a host without the need of immunosuppression. As stem cells, MSCs have the ability to proliferate without spontaneously differentiating, while maintaining the ability to differentiate into multiple cell types in vitro.75,76 These characteristics allow multiple in vitro passages of expansion to be performed, reaching high cell numbers of undifferentiated MSCs (e.g. 50 to 375 million cells generated by passage 2 from a 10 ml marrow aspirate)75.

Figure I.6. The Mesengenic Process: MSCs proliferate and their progeny can be induced to enter one of several mesenchymal lineage pathways, including marrow stroma, osteogenesis, chondrogenesis, tendogenesis, myogenesis and adipogenesis. Adapted from Caplan and Correa, 2011.

I.1.4.2 MSCs Sources

One other important feature that allows the use of MSCs in Regenerative Medicine relies on their accessibility and ease of harvest from various tissues. MSCs have been isolated from the BM, peripheral blood, cord blood, cord Wharton’s jelly, AT, amniotic fluid, compact bone, periosteum, synovial membrane and synovial fluid, articular cartilage and fetal tissues.64 MSCs have been further identified in skeletal muscle, hair-follicle and urine.77 However, differences in culture morphology,

growth rates, proliferation potential, and differentiation capacity can occur depending on the tissue from which MSCs are isolated.64

The most promising MSC sources in terms of significance of the number of cells obtained and the amount of descriptive data in the literature correspond to BM (BM-derived MSCs – BM-MSCs), AT (adipose tissue derived-stromal/stem cells – ADSCs) and umbilical cord (umbilical cord matrix stem cells or umbilical cord derived MSCs – UCM-MSCs). Although BM is considered the traditional source of MSCs, these cells constitute only a small percentage of about 0.01%-0.001% of the total population of nucleated cells in the marrow, and the harvesting process can involve an invasive and painful procedure.32,75 BM-MSCs are isolated from the mononuclear layer of BM aspirates after separation by density gradient centrifugation. Once cultured in vitro, MSCs adhere to the plastic surface of culture plates, while hematopoietic progenitors and other non-adherent cells are removed by medium changes.75 Human AT can be routinely obtained as excised surgical specimens or as lipoaspirates (often resulting in more than 1 L of tissue), and subsequently digested with collagenase and centrifuged.62 It has been described that a single millilitre of human lipoaspirate can yield between 0.25 to 0.375x106 ADSCs, capable of differentiating into adipocyte, chondrocyte and osteoblast lineages in vitro.62 UCM-MSCs are collected from the umbilical cord by explant culture or by enzymatic digestion and filtration, achieving yields of 8.6x105 of adherent cells/cm.78 UCM-MSCs present a greater proliferation capacity and faster growth rate in comparison to BM-MSCs, as well as a lower expression of HLA-class I.32,78 The umbilical cord presents other advantages such as accessibility, painless procedures to donors, the possibility of autologous and allogeneic cell therapy and lower risk of viral contamination.78 However, the number of cells obtained from one single cord unit is scarce for clinical applications, thereby requiring ex-vivo expansion strategies.

I.1.4.3 In vivo Roles of MSCs

It has been suggested that cells with mesenchymal characteristics reside in virtually all postnatal organs and tissues.73 The fact that MSCs seem to reside in almost every tissue of the human body incites questions about the in vivo role of these cells, apart from their differentiation to mesenchymal lineages (allowing the growth and natural turnover of mesenchymal tissues). In fact, MSCs are known to secrete a great variety of bioactive factors with paracrine and autocrine effects (referred to as trophic factors/effects) such as enhanced angiogenesis and tissue regeneration, inhibition of fibrosis (scar formation) and apoptosis, and stimulation of mitosis and differentiation of tissue-intrinsic progenitor cells.79,80 It should be noted that these trophic factors/effects do not lead to the differentiation of the secreting MSCs, but rather primarily affect their neighbouring cells. Moreover, MSCs act by displaying immunomodulatory activities (including the suppression of the local immune system), recruiting endogenous MSCs to the injury site and possibly transferring mitochondria or vesicular components containing mRNA, microRNA and proteins.81

MSCs have the ability to migrate chemotactically into tissues showing inflammation and injury, possibly in response to signals that are upregulated under injury conditions73. After being mobilized to the injury site, MSCs are presumed to act as curative agents by becoming activated and establishing a regenerative microenvironment, through the secretion of bioactive molecules and the regulation of the local immune response.80 In conclusion, the currently proposed mechanisms of action for MSCs in vivo include not only differentiation but also paracrine/trophic effects, immunomodulatory effects, and transfer of vesicular components.81 While the differentiation capacity of MSCs benefits tissue engineering-related applications, the capacity to secrete bioactive factors with regenerative and immunomodulatory effects could be beneficial for cell therapy applications, such as tissue repair.

I.1.4.4 MSCs in Clinical Trials

The success of MSCs in modulating immune responses and promoting tissue repair in preclinical studies has prompted the exploration of MSCs in clinical settings.29 MSCs have been explored to treat myocardial infarction, multiple system atrophy, liver failure and cirrhosis, diabetic critical limb ischemia and foot ulcer and GvHD, among other diseases, with no severe side affects.29 In particular, GvHD, a severe reaction followed by allogeneic HSCs/bone marrow transplants, in which the transplanted cells “reject” the host’s own cells, can be attenuated by MSC co-transplantation.82 Little is known about the mechanisms of suppression of GvHD, however, the release of soluble factors, induction of regulatory T cells, and repair of damaged target organs are possible mechanisms.82 As mentioned in Section I.1.2.2., a MSC-based product (i.e. ex-vivo expanded cells) named Prochymal was already approved for paediatric use in GvHD, in Canada and New Zealand.

MSCs do not show tumorigenic activity, as opposed to pluripotent stem cells (mentioned in Section I.1.2.2), which makes them an extremely encouraging stem cell type for Regenerative Medicine proposes. Other clinical MSC applications include tissue engineering constructs or direct MSC injection for bone and cartilage repair. For autologous cartilage repair, natural matrices (e.g. alginate and collagen), as well as synthetic polymers are available, with the design of the ECM being of extreme importance, since it must lead to proper differentiation of MSCs into chondrocytes.23,83 Autologous MSCs were successfully implanted in patients to enhance fracture/osteotomy healing, fill bone defects, treat pseudarthrosis, bone cysts, osteonecrosis, or enhance spinal fusion.23 Besides their astonishing osteogenic differentiation capacity, MSCs are capable of secreting paracrine factors that enhance angiogenesis during fraction healing.23 These angiogenic properties give great hopes for MSCs as therapeutic agents for diseases caused by limited angiogenesis, such as peripheral artery disease (ischemia), myocardial infarction and cerebral ischemia/stroke.81 However, the exploitation of the trophic properties of MSCs for cell therapy may come with one disadvantage: the shifting from local delivery of MSC to systemic administration, although less invasive and more convenient, causes only a small percentage of the infused MSC (often <1%) to

reach the target tissue with cell entrapment commonly observed in capillaries within the liver, spleen and lung.84 Furthermore, the absence of a specific marker of MSCs as well as the heterogeneity/inconsistency of MSCs isolation and expansion techniques are still major hurdles, which make it difficult to compare pre-clinical and clinical results from different research groups.81

I.1.4.5 Emerging Issues: Muse cells and the Perivascular Niche Hypothesis

Although there is a general agreement on the multipotency capability of MSCs (capability of differentiating into most mesodermal cell lineages, excluding the hematopoietic cell lineage), some reports have emerged in the past decade suggesting that MSCs might be pluripotent (capable of differentiating into mesodermal, ectodermal and endodermal lineages) or, at least, able to transdifferentiate.27 However, most of these results have been widely debated due to their non-reproducibility, which discredits their reliability and authenticity.

Nonetheless, a supposedly pluripotent stem cell type, residing among MSCs, was recently found by Kuroda and colleagues: multilineage-differentiating stress-enduring (Muse) cells.85 As their denomination suggests, Muse cells are resistant to stress conditions, namely 16-hour incubation with collagenase or trypsin, low temperatures, low serum and hypoxia.85,86 This subpopulation of cells expresses CD105, CD90 and CD29, markers expressed by MSCs, but also pluripotency markers (SSEA-3, Nanog, Oct3/4, and Sox2).85,86 These cells are able to differentiate into endodermal, ectodermal, and mesodermal cells both in vitro and in vivo when stimulated by certain combinations of cytokines and trophic factors.85 However, they only comprise 1% of BM-MSCs grown in adherent culture, and are thought to exist in a dormant, or quiescent state under normal physiological circumstances within the cellular niche.85 Muse cells have been efficiently isolated from BM85, AT86 and dermis85, as well as from commercially available cultured fibroblasts and ADSCs87,88, and were further characterized by different research groups (namely, a research team at the University of California, Los Angeles)86, which suggests that the existence of these pluripotent MSC-like cells is possible. Still, the existence of pluripotent cells in adult mesenchymal tissues is a very controversial subject, and more studies confirming the characterization and differentiation potential of these cells are required in order to replicate and credit the abovementioned data. Notably, Muse cells do not seem to show tumorigenicity when transplanted in vivo and do not require genetic modification strategies to differentiate, which enables their use for clinical cell therapy applications.85,89 In fact, the therapeutic potential of these cells was very recently explored in animal models of cerebral ischemia85 and of diabetic skin ulcers86. Through the selection of SSEA-3 positive cells from the BM and AT by fluorescence-activated cell sorting (FACS) and magnetic-activated cell sorting (MACS) (mentioned in Section I.2.1.2.), the sorted Muse-enriched and Muse-depleted populations were transplanted into the mentioned animal models.90,91 In the first study, only the Muse-enriched population showed engraftment capacity into ischemic tissue, while in the second study this population displayed greater therapeutic effects to accelerate impaired wound healing associated with type 1 diabetes.90,91

The reported presence of quiescent Muse cells in vivo raises questions about the existence of a MSC niche, where more primitive MSC-like cells might be maintained under severe stress conditions, such as hypoxia. In fact, some data suggest that hypoxia enhances not only the proliferative capacity but also the plasticity of MSCs.70,92,93 However, the presence of a true MSC niche has not been confirmed nor identified. Nevertheless, a theory suggesting that the MSC niche is of perivascular nature is currently under development. At the same time, a major unresolved question relates to the identity of MSCs within their native tissues. There is growing evidence that pericytes (also referred as Rouget cells or mural cells), may be the native cells of the ex vivo MSCs.64 MSCs express CD106, or VCAM-1 (vascular cell adhesion molecule-1), also expressed on blood vessel endothelial and adjacent cells, and have been identified in perivascular locations.70,94,95 Furthermore, MSCs and pericytes share remarkable similarities such as their differentiation potential (into osteoblasts, chondrocytes, and adipocytes), the ability to maintain multipotency at high passage numbers in culture and common marker expression.71,94 These similarities, in particular the fact that pericytes express cell surface markers that are identical to those expressed by isolated MSCs94, led Caplan to speculate if all MSC are, in fact, pericytes.96 However, there is controversy around the connotation of pericytes as “stem cells”, since they are not know to proliferate and differentiate into mature cell phenotypes in vivo following injury.64

Ultimately, the possibility that MSCs reside in a perivascular niche throughout the body would give them easy access to all tissues in response to systemic influences related to injury, lending credence to the notion that MSCs are integral to the healing of many different tissues.70,71 In conclusion, the existence of a perivascular MSC niche is plausible, where MSCs would interact with other cell types, ECM and signalling molecules (Figure I.7) and periodically migrate to the bloodstream, in situations of injury or disease. Additionally, as the O2 tension within the niche is presumed to be variable70,97, it can be hypothesized that oxygen-dependent cell-specific compartments are likely to exist, and that hypoxia conditions can be eventually associated with the presence of primitive/progenitor MSCs (presumably Muse cells).

Figure I.7. The MSC perivascular niche hypothesis. Within the niche, O2 tension is variable. MSC in contact with blood vessels (BV) would interact with (1) various other differentiated cells (DC1, DC2, etc.) by means of cell-adhesion molecules, (2) ECM deposited by the niche cells (mediated by integrin receptors), and (3) signalling molecules. From Kolf et al., 2007.

I.1.5 Tissue Engineering-derived Urethral Constructs in Clinical Trials

Few cases of tissue engineering-derived urethral constructs have been developed and tested on a pre-clinical and clinical basis for the treatment of SUI, contrasting with cell therapy approaches, which have been more assessed in the literature. Still, a pre-clinical study tested the feasibility of seeding autologous urothelial cells and SMCs, harvested from bladder biopsies, on tubularized acellular collagen scaffolds for urethral replacement in canine models, when compared to tubularized matrices without cells.98 The results showed that the urothelial cells formed a multilayered epithelial lining on the luminal side of the scaffold, while the muscle cells formed a layer on the outer surface.98 By contrast, the animals that received the tubularized matrix without the cells developed urethral strictures, urine extravasation, and hematoma and fistula formation.98 These results show the importance of including cellularized scaffolds in the treatment of incontinence, as the inclusion of scaffolds alone is not enough for complete tissue healing, and may lead to complications.

The same type of approach was implemented in a clinical study where human patients with urethral defects underwent urethral reconstruction with tissue-engineered tubularized urethras.99 Bladder biopsies were performed in order to collect SMCs and urothelial cells from the patients, which were seeded onto tubular scaffolds made from polyglycolic acid.99 Although most patients became continent after urethral reconstruction, one of the patients, who suffered from SUI caused by a pelvic disruption that involved the sphincter, needed a pubovesical sling after surgery.99 Thus, in cases of SUI where an intrinsic sphincter deficiency is what mainly causes incontinence, alternative clinical procedures to reconstruct the sphincter are required. Furthermore, the need to perform biopsies to collect autologous cells carries additional risks to the patient, as this surgical procedure may cause further complications in the genitourinary tract. Therefore, novel autologous cell sources should be explored.

To the extent of the reviewed literature presented herein, no studies using MSCs as a cell source for tissue engineered-urethra constructs, to be applied at a clinical level, were found. Still, MSCs have been used in clinical studies where stem cell therapy approaches were implemented to treat SUI. One particular clinical study aimed to treat male SUI (caused by urethral sphincter deficiency) by injecting autologous ADSCs into the patients’ rhabdosphincter.100 At 6 months after the injection, UI improved in terms of leakage volume and the function of the urethral sphincter was improved in all cases.100

MSCs have been considered one of the most promising types of cells to be used in SUI therapies.2,101 Nevertheless, a major challenge in using MSCs for muscle tissue engineering is to robustly direct their differentiation towards the desired cell type and having the cells retaining that specific phenotype in vitro and, importantly, upon implantation in vivo.

I.2 Myogenic Differentiation of MSCs: Current Status I.2.1 Differentiation of MSCs into Skeletal Muscle Cells

The myogenic capability of MSCs has been demonstrated both in vivo, through the transplantation of MSCs in animal models, and in vitro, through the use of myogenic differentiation induction strategies. An in vivo animal study demonstrated the survival of BM-MSCs transplanted in the urethral wall of urethral sphincter injured rats, as well as their differentiation into striated muscles and nerves.19 The cells were identified within rhabdosphincters on day 14 after transplantation, and were positive for both desmin and skeletal MHC. Additionally, some injected cells were positive for PGP9.5, a neural specific protein, suggesting that a small fraction of MSCs differentiated into nerves.19 Several in vitro approaches have been developed in order to differentiate MSCs into myoblasts. The most common strategies found in the literature rely on the addition of specific soluble factors and/or DNA demethylation agents, MSC co-culture with myoblasts and the use of patterned surfaces. Dulbecco's modified Eagle's medium (DMEM) supplemented with Horse Serum (HS), among other factors, is a commonly used myogenic induction medium used in the literature.69 The supplementation of DMEM with 5% HS and corticosteroids (dexamethasone and hydrocortisone) to UCM-MSCs was shown to lead to the sequential expression of MyoD, MyoG and MHC, in accordance to the myogenic differentiation pattern.102

Regarding the co-culture procedures found in the literature, Gunetti and colleagues concluded that BM-MSCs do not fuse with striated muscle cells in both in vitro (co-culture with C2C12 myoblasts) and in vivo (transplantation into rat pelvic muscles) conditions.103 However, when seeded onto a laminin matrix and plated in the presence of Insulin-Transferrin-Selenium (ITS) supplement and Epidermal Growth Factor (EGF), MSCs expressed myogenic markers, although myotube formation was not observed.103 Beier and co-workers observed that MSCs showed enhanced ability to fuse and contribute to myotube formation when cultured with primary myoblasts under the stimulus of bFGF and dexamethasone.104 Interestingly, in both studies the authors demonstrated that MSCs constitutively express certain muscle-specific markers. Additionally, Garza-Rodea and colleagues reported that human MSCs derived from AT, BM and synovial membrane in co-culture with murine C2C12 myoblasts and plated in the presence of ITS formed hybrid myotubes at low frequencies.105 Notably, when transplanted into an animal model of muscle damage, AT-derived MSCs contributed the most for myoregeneration in vivo.105 These results suggest that co-culture with myoblasts may not be sufficient to fully induce MSCs to differentiate into skeletal muscle and that the combination of specific growth factors/hormones, some known to promote myogenic differentiation106, may be essential.

Finally, the use of matrices with the purpose of inducing myogenic differentiation of MSCs has also been investigated, with some successful case studies. Recently, efficient commitment of BM-MSCs to the myogenic lineage was accomplished through plating on a biomimetic polyurethane acrylate resin replicating myoblast topography.107 The resin, fabricated through a nano-imprint lithography technique, induced morphology changes and higher expression of muscle-specific genes in cultured MSCs (namely MyoD, MyoG and skeletal MHC), when compared to flat substrates.107

Analogously, ADSCs were shown to undergo stiffness-induced lineage commitment and fusion (albeit at low percentages) when plated in a muscle-mimicking matrix with a stiffness value of 10 kPa, without any other form of external stimuli.68 In these conditions, ADSCs surpassed BM-MSCs, which never underwent stiffness-mediated fusion, thus suggesting that the AT may be a source of cells with enhanced myogenic potential when compared to cells from BM.

I.2.1.1 Stromal Vascular Fraction: The Myogenesis-committed Subpopulation

The stromal vascular fraction (SVF) of the white AT consists of a heterogeneous population of cells comprising adipose stromal cells (15-30%), hematopoietic-lineage cells (25-45%), endothelial cells (10-20%), pericytes (3-5%) and fibroblasts, among others.108,109 According to the ISCT, stromal SVF cells are identified phenotypically by the following phenotype: CD45-CD235a-CD31-CD34+. Through plating, the plastic-adherent population (e.g. ADSCs) is isolated, and these cells are known to retain MSC markers, including CD90, CD73 and CD105 and remain negative for CD45 and CD31.108

Cells derived from the SVF cultured in DMEM supplemented with 5% HS alone have been reported to spontaneously differentiate into skeletal muscle cells without any additional external stimuli.110 Di Rocco and co-workers reported that this rare subpopulation of myogenesis-committed cells only corresponded to 0.001% of total plated cells.110 Concordantly, Shan and colleagues reported that the myogenic differentiation potential of the white AT SVF-derived cells resides in the Myf5-lineage progenitors.111 The Myf5-lineage progenitor pool contains subpopulations of adipogenic and myogenic progenitors that could be isolated in transgenic mice based on stem cell antigen-1 (Sca1) expression (positive and negative expression, respectively).111 When transplanted into an in vivo model, the Myf5+Sca1- SVF cells expressed MyoD and Pax7 and differentiated into myofibers expressing low MHC.111 Still, hydrocortisone and/or dexamethasone supplementation together with HS seem to drive SVF-derived cells to undergo myogenesis (assessed by multinucleation, skeletal MHC and MyoD detection), at average frequencies around 12% and 15%, without resorting to any kind of cell separation methodology.69,112 A subpopulation of myogenesis-committed cells was also detected in semi-solid medium experiments, where clusters of myogenic cells spontaneously emerged from cultures of crude SVF.113 Still, further research regarding the characterization and ideal isolation and expansion of such myogenic cells from the SVF (namely, the detection of its unique markers) is required in order to conclude if the achievement of sufficient numbers of skeletal muscle cells from the SVF is possible.

I.2.1.2 Stem Cell Separation Technologies: FACS vs. MACS

The isolation of pure stem cell populations from a heterogeneous suspension (e.g. potential myogenesis-committed progenitor cells from crude SVF) is a fundamental aspect of stem cell clinical application and basic research.114 Besides purity, several criteria such as resolution capabilities and scalability of the process, final cell viability and the achievement of clinically relevant cell numbers are

important to consider when choosing the ideal stem cell separation technique, which ultimately depends on the final application of the cellular product. Stem cell separation methods can be classified into two categories: techniques based on physical properties (size/density), and techniques based on affinity.114 For the purpose of the work developed herein, only affinity-based techniques will be considered, particularly FACS and MACS. FACS is a powerful technique, with high precision, resolution and sensitivity that allows highly pure cell populations (over 95%) to be obtained.114 This technique is composed by three major components: the fluidics, optics and electronics.115 The separation process relies on antibodies tagged with fluorescent dyes that bind to surface antigens either in the desired or the undesired cell fraction (in case of positive or negative selection, respectively) in mixed cell suspensions. Through the use of mechanical vibrations, the input liquid stream is separated into small droplets, allowing the cells to be analysed individually.115 Each cell is then charged positively or negatively and deflected into sampling tubes, depending on whether they fluoresce or not (Figure I.8). Still, FACS has limited throughput (107 cells/hour) and long processing times (3h-6h) and as an alternative, separation using antibody-coated magnetic beads can be used (Figure I.8).114,115 After mixing the cell suspension and the beads, the mixture passes through a column under the influence of a strong magnetic field. Bead-carrying cells are retained while the unbound contaminants are washed from the column. The final step of the process consists on turning the magnetic field off and eluting the desired cells.

A B 1. Magnetic labelling with MACS Microbeads

2. Positive selection using MACS column

3. Elution of positively selected cells

Positive fraction

Figure I.8. FACS vs. MACS. A) A fluorescence-activated sorter can be used as a preparative tool to separate fluorescently-labelled cells from a heterogeneous cell suspension. B) Through magnetic labelling, desired cells are firstly separated from contaminants and then eluted from the column. Left image from the website of Midlands Technical College (www.midlandstech.edu).116 Right image adapted from the website of Humboldt University of Berlin (http://edoc.hu-berlin.de/)117

One of the advantages of MACS when compared to FACS is the existence of less expensive commercial systems, while FACS requires specialised technicians to operate the machine. In particular, for clinical applications requiring automated cell separation on a large scale, a closed and sterile system (CliniMACS® system by Miltenyi Biotec) is already available.115,118 FACS, on the other hand, may not be as adequate for clinical-grade applications due to possible contamination risks and the high costs associated with the equipment and reagents.114,115 Regarding cell viability, MACS is believed to induce lower shear stress to the cells,119 although cell lysis is likely to occur when eluting the desired cells from the MACS column. Still, the possibility of cell interference caused by the magnetic particles themselves is described as the main drawback of using MACS.114,115

I.2.1.3 5-Azacytidine to Induce Myogenesis

5-Azacytidine (5-AZA), a cytidine analogue, was first synthetized in 1964.120 5-AZA was originally developed and tested as a nucleoside antimetabolite with clinical specificity for acute leukemias.121 Antimetabolites are chemical compounds that are structurally similar to purine and pyrimidine nucleotides, becoming readily incorporated into either DNA or RNA, and consequently interfere with normal cell functions, particularly gene expression.122 Although the genotoxic effect of these compounds can theoretically occur upon all cells undergoing synthesis of new DNA, they mostly affect fast-diving cells such as tumour cells, which is the reason why antimetabolites are used as chemotherapeutic agents.122 Currently, 5-AZA is marketed under the trade name Vidaza, as well as its deoxy derivative, 5-aza-2’-deoxycytidine (5-AZAd) or Decitabine, for the treatment of myelodysplastic syndrome (Figure I.9).123 While 5-AZA is incorporated into both DNA and RNA, 5-AZAd affects only DNA, which makes this compound approximately 10-fold more cytotoxic.123

Figure I.9. Structures of cytidine and its analogs, 5-AZA and 5-AZAd. R=ribose. dR=deoxyribose. Adapted from Christman, 2002.

5-AZA acts by covalently inhibiting DNA cytosine methyltransferases, which become irreversibly bound to 5-AZA residues present in DNA.123 As methyltransferases are responsible for DNA methylation (i.e. suppression of specific genes), their inhibition results in passive loss of methylation, thereby allowing the expression of formerly silenced genes.123 Since the methylation target for the mammalian maintenance methyltransferase (Dnmt1) is C-phosphate-G (CpG) dinucleotide pairs in hemi-methylated sites, 5-AZA should indirectly reactivate the expression of genes with promoter CpG islands.123

In 1995, Wakitani and colleagues demonstrated that immortalized rat BM-MSCs exposed to 5-AZAd differentiated into myogenic phenotypes in vitro.67 The cells exhibited a multinucleated morphology, contracted spontaneously and when exposed to acetylcholine, and expressed skeletal MHC.67 Since then, several research groups have claimed that 5-AZA-treated MSCs were able to differentiate to skeletal muscle-like cells124–127, cardiomyocyte-like cells128,129 and osteocytes130.

In one pre-clinical study, 5-AZA-exposed rat ADSCs formed myoblast-like cells that were injected into the urethral sphincter of animal models of SUI.131 According to the authors, the myoblast-like cells induced from ADSCs played a positive role in improving the urine-controlling ability of SUI animal models.131 Still, in vitro, Drost and colleagues concluded that independently of 5-AZA exposure, MSCs express smooth and striated muscle antigens and that treatment with 5-AZA only moderately and transiently increases the myogenic character of the cells.126 Burlacu and co-workers applied several 5-AZA pulses to MSCs and determined that MSCs retained their multipotent capacity after one pulse with 5-AZA, whereas additional pulses resulted in a restricted differentiation potential with simultaneous increased ability to accomplish chondrogenic commitment.132 In this study, the authors stated that the increased expression of muscle-specific genes observed after 5-AZA treatment was not sufficient for the induction of differentiation into muscle cells such as cardiomyocytes. Concordantly, Martin-Rendon and collaborators reported that only a small proportion of BM-MSCs could generate cardiomyocyte-like cells in vitro after treatment with 5-AZA, while Liu and colleagues demonstrated that BM-MSCs could not be induced to differentiate into cardiomyocytes.133,134 Still, 5-AZA-treated murine-immortalized BM-MSCs were shown to differentiate into cardiomyocyte-like cells and display contractile activity,128 while human BM-MSCs induced with 5-AZA were reported to form myotube-like structures positive for myocardium-specific α-actin, sarcomeric β-MHC and troponin-T, although no spontaneous beat was detected.129 Contrasting with all these studies, Zhou and colleagues reported that 5-AZA treatment facilitated osteogenic differentiation, which was accompanied by hypomethylation of genomic DNA and increased osteogenic gene expression.130 It is possible to conclude that the results present in the literature are, in a general matter, not consistent and even contradictory regarding the effect of 5-AZA on MSC differentiation.

I.2.2 Differentiation of MSCs into SMCs

Great attention has been put into researching the capacity of MSCs to differentiate into SMCs, particularly vascular SMCs, with the purpose of overcoming the primary limiting factors in vascular tissue engineering, namely the construction of blood vessels with adequate SMC layers.135 Moreover, SMCs are major components of hollow visceral organs, such as the urethra. However, the transition from a stem cell to a SMC is a complex biologic process that is not yet fully understood. Still, cultured SMCs and BM-MSCs show similarities such as the expression of -SMA, SM22, calponin and smoothelin, and transplanted BM-MSCs have been shown to be capable of differentiating into a

vascular SMCs phenotype in vivo.136,137 These encouraging findings suggest that it may be possible to establish a protocol capable of driving BM-MSC into the SMC lineage in vitro.

I.2.2.1 Soluble Factors

TGF-β plays an important role in cell differentiation and vascular remodelling.42 There is considerable evidence that TGF-β may play a dual role in SMC differentiation, being important for both initial commitment and further differentiation to SMCs.138 Therefore, several studies have tested the effect of this soluble factor in MSC differentiation, frequently in association with platelet derived growth factor-BB (PDGF-BB).139–142 TGF-β was shown to enhance the expression of -SMA in BM-MSCs, as opposed to PDGF-BB which lowered its expression.139 However, when cultured in the presence of both these soluble factors, or TGF-β alone, BM-MSCs could be induced to differentiate in a relatively homogeneous way into SMCs expressing SM22, -SMA, calponin and SM-MHC.46 Wang and collaborators showed that TGF-β induced cell morphology changes as well as an increase in actin fibers in BM-MSCs.143 According to them, TGF-β coordinates the expression of gelsolin (an actin-binding protein that is a key regulator of actin filament assembly and disassembly) and -SMA, to promote the assembly of -actin and actin filaments in MSCs.143 Human iPSCs were successfully differentiated into functional SMCs through an intermediate stage of multipotent MSCs (capable of adipogenesis, osteogenesis, chondrogenesis and myogenesis).144 At this stage, the cells were exposed to differentiation medium containing TGF-β and heparin, and a significant increase in -SMA, calponin, and SM-MHC (early, middle and late SMC marker genes, respectively) was observed, as well as the exhibition of a robust fibrillar organization, which suggested the development of a contractile phenotype. Moreover, tissue constructs prepared from these cells exhibited high levels of contractility in response to receptor- and non-receptor-mediated agonists.144 BM-MSCs have been successfully used as a source of SMCs for vessel engineering. In a particular study, BM-MSCs were differentiated into SMCs through the addition of TGF-β and seeded onto a biomimetic culture system in vitro.141 The engineered vessel walls were further optimized by the alteration of conditions such as matrix proteins, soluble factors, and cyclic strain, and the constructs were found to be substantially similar to native vessels.141

I.2.2.2 ERK/MAPK Signalling Pathway

Few studies have shown that the extracellular-signal-regulated kinase (ERK)/ mitogen-activated protein kinase (MAPK) signalling pathway exerts an anti-myogenic effect in SMC differentiation. This pathway plays an important role in the regulation of cell proliferation, differentiation and migration, and is under the control of extracellular ligands related to extracellular stimuli such as growth factors and environmental stresses.145

Once the ERK/MAPK pathway is activated, the terminal component of this pathway (MAPK) translocates from the cytoplasm to the nucleus, in order to induce specific changes in gene expression in response to extracellular signals.145 These signals come from receptor tyrosine kinases (RTKs) and are transduced by the activation of the ERK/MAPK cascade (Figure I.10)146. Initially, a protein complex consisting of the RAS exchange factor son-of-sevenless (SOS) and the growth-factor-receptor bound protein-2 (GRB2) is recruited to a docking site on the RTKs.146 This results in the activation of the small GTPase RAS, which triggers the sequential phosphorylation of three kinases: MAPK kinase kinase (MAPKKK), MAPK kinase (MAPKK) and MAPK.146 Subsequently, the phosphorylated isoform of the classical MAPK (known as ERK) translocates to the nucleus and phosphorylates and activates transcription factors such as Elk-1 and c-Myc, which control the expression of genes required for cell growth, differentiation and survival.146 The ERK/MARK pathway is associated with the repression of SRF, a critical regulator of the differentiation of vascular SMC.140,147 The activity of SRF is regulated by single interactions with myocardin or Elk-1.147 Myocardin functions as a very potent myogenic factor for vascular SMC and induces the expression of SMC gene markers, including SM-MHC.148 Elk-1, on the other hand, competes for the SRF binding site with myocardin and antagonizes myocardin, thus supposedly disabling the induction of SMC gene markers and acting as a myogenic repressor.147

Figure I.10. The ERK/MAPK signalling pathway. Its activation triggers the sequential phosphorylation of three kinases (MAPKKK,MAPKK and MAPK), resulting in the translocation of ERK to the nucleus and the activation of transcription factors (e.g. Elk-1 and c-Myc) which control the expression of genes that are required for cell growth, differentiation and survival. From Kim and Bar-Sagi, 2004.

Taking these findings into consideration, it has been suggested that by blocking ERK/MAPK signalling through the use of a MEK inhibitor, the activation of SRF by myocardin would be allowed,

hence possibly driving the differentiation of BM-MSCs into SMCs in vitro.140 Using the small molecule modulators PD98059 and U0126, which inhibit ERK/MEK kinases, Tamama and colleagues were able to block the ERK/MAPK signalling pathway and upregulate SMC marker gene expression in BM-MSCs, likely through induction of myocardin expression.140 Furthermore, the authors tested the ability of the MEK-inhibitor treated BM-MSCs to contract or relax in response to stimuli, and observed that the cells contracted a collagen gel in response to endothelin, the most potent vasoconstrictive mediator, as oppose to the non-treated BM-MSCs.140 These promising results suggest that the use of MEK inhibitors might be an effective approach to differentiate MSCs into SMCs.

I.2.2.3 Other approaches

A recent study demonstrated that high cell density promotes MSC differentiation into contractile SMCs (measured by the expression of SMC gene and protein markers), when cultured in growth/myogenic differentiation medium containing bFGF, due to the formation of adherens junctions.149 This process was found to be mediated through cadherin-11, which acted as a regulator of myogenesis, increasing the expression of TGF-β and ultimately regulating the expression of SRF.149 Direct cell-to-cell contact was shown to induce MSCs to differentiate into SMCs, through direct co-culture of BM-MSCs with SMCs.150 By contrast, indirect co-culture and conditioned media did not induce the expression of SMC gene markers.150 Additionally, mechanical straining and patterns of substrates were also tested to differentiate MSCs without using biochemical reagents.151 Higher level straining (10%) and grooved surfaces were shown to positively influence the differentiation of MSCs into SMC-like cells.151 Conversely, coating with bladder-derived ECM and 3-D dynamic culture conditions were shown be insufficient to induce SMC marker expression in BM-MSCs in the absence of biochemical stimuli.142 In this study, PDGF-BB and TGF-β alone induced the expression of smooth muscle genes and proteins, including SM-MHC.142

Recently, sodium butyrate (NaB), a histone deacetylase (HDAC) inhibitor was used in a strategy to promote rat MSCs to differentiate into SMCs.152 Concentrations of 1.0 and 1.5 mmol/L of NaB enhanced the mRNA expression of calponin (more than 1-fold), SM-MHC (approximately 2-fold) and -SMA (less than 1-fold). Moreover, NaB significantly enhanced MSC differentiation into SMCs in a co-culture system with bladder SMCs and the NaB-induced cells were proven to be functional as they were sensitive to depolarization.152 However, much like 5-AZA, this compound may act in an unspecific way or even lead to toxicity, which is a disadvantage when compared to the use of growth factors and/or matrices to induce differentiation.

Lastly, the use of highly porous microcarriers, produced using thermally induced phase separation, have been developed as a method aiming the expansion and differentiation of ADSCs into smooth muscle-like cells.153 The authors induced the cell’s differentiation by using a differentiation medium containing TGF-β, which resulted in a significant increase in the mRNA expression of cell contractile apparatus components caldesmon (approximately 5-fold), calponin (4.5-fold), and SM-MHC

(approximately 3.5-fold) at 14 days culture.153 This approach has the advantages of using microcarriers, which allow cell expansion and differentiation to occur as a one step process and can potentially act as a delivery vehicle for the attached cells for therapeutic applications.

I.3 Aim of Studies

The aim of this MSc project was to develop strategies to direct MSCs to differentiate into skeletal muscle and SMCs for urethral sphincter engineering. To achieve this goal, the effects of compounds that are described in the literature as myogenic inducers in MSCs were tested, and the myogenic potential of cells derived from the SVF was explored. Ultimately, the discovery of the appropriate biochemical stimuli required to establish a laboratory-based myogenic differentiation protocol (i.e. the ideal formulation of a potential skeletal/smooth muscle induction media) was targeted. The following specific objectives were pursued: (i) Studying the effects of 5-AZAd in the myogenic differentiation of ADSCs; (ii) Exploring the potential of SVF-derived cells to undergo myogenesis without the use of demethylation compounds, co-culture strategies or in vivo engraftment; (iii) Developing strategies for the isolation of a cell subpopulation from SVF with potentially increased myogenic potential; (iv) Studying the effects of compounds that act upon the activation or repression of specific signalling pathways important in smooth muscle differentiation (e.g. small molecules such as PD98059) in MSCs.

Finally, a decellularization protocol based on mechanochemical action was tested in porcine urethras, with the aim of potentially combining this method with the knowledge gathered in this thesis to further develop urethral sphincter engineering strategies.

II. Materials and Methods

II.1 Human-derived samples

AT samples (lipoaspirates) were obtained following donor informed consent from the abdomen and thigh areas of both female and male healthy donors with ages ranging from 30 to 52 years old, at Clínica de Todos-os-Santos, Lisbon, Portugal. A BM sample was obtained following donor informed consent from a 36-year-old male healthy donor at Instituto Português de Oncologia, Lisboa, Portugal.

II.2 Culture Media II.2.1 Ex-vivo expansion of ADSCs, BM-MSCs and SVF-derived cells

All cells were maintained in culture using low glucose (LG) (1g/L) Dulbecco’s modified Eagle’s Medium (DMEM), supplemented with 10% fetal bovine serum (FBS, Gibco®, Carlsbad, CA) and antibiotics (0.025μg/mL penicillin and 0.025U/mL streptomycin). The FBS used for ADSCs and BM-MSCs culture was MSC-qualified, which means it was certified for MSC research. On the other hand, the connotation “AT-SVF cells” is referred to cells derived from AT that were isolated and cultured in FBS that was not MSC-qualified (standard FBS). These cells were not labelled here as MSCs, as the SVF is composed by several distinct cell populations.108 The medium was replaced every 3 to 4 days to diminish toxic metabolites and provide fresh nutrients, thus promoting cell growth. Cell expansion was performed in static culture systems (T-flasks, BD FalconTM Biosciences©, Franklin

Lakes, NJ) in an incubator at 37ºC with a humidified atmosphere of 5% CO2. Images of the cell’s morphology (in bright field microscopy) were captured with Leica Camera CD350F Software in a CK40 Microscope (Olympus®).

II.2.2 Thawing and cryopreservation of SVF cells, ADSCs and BM-MSCs

Cryopreserved SVF cells (previously isolated by an enzymatically-based method154) were thawed and recovered by dilution in a proportion of 1:5 in DMEM+20%FBS supplemented with DNase (1:100 dilution of 1mg/mL aliquots, Roche Applied Science, Hague Road, IN) while ADSCs and BM-MSCs were recovered in DMEM+20%FBS. The cells were then centrifuged for 7 minutes at 1250 rpm, resuspended and plated in the appropriate culture media at cell densities between 3000 and 5000 cells/cm2 for ADSCs and BM-MSCs or at 10 000 cells/cm2 for SVF cells. When 80-90% of confluence was reached, the cells were harvested with accutase (Sigma®, St. Louis, MO) or trypsin 0.05% (Gibco®) supplemented with 1 mM EDTA (Gibco®) for 7 minutes. Enzymatic digestion was stopped with DMEM+10%FBS in a proportion of 3:1 and the cells were centrifuged for 7 minutes at 1250 rpm. Viable cells were counted under an optical microscope using the trypan blue dye exclusion method (Gibco®), and part of the cells were cryopreserved. In order to protect the cells from damage during freezing or thawing, dimethylsulfoxide (DMSO), a cryopreservative compound, was used.155 The cells were resuspended in RecoveryTM cell culture freezing medium (Gibco®) (which contains DMSO) or in a solution of 10%DMSO (Merck Millipore©, Billerica, Massachusetts, USA)+90%FBS and

distributed into freezing vials (1mL/vial). The cells were then stored overnight in a -80ºC freezer and transferred to a liquid/vapour phase nitrogen container for long-term storage.

II.3 Multilineage differentiation assays

SVF-derived cells, ADSCs and BM-MSCs were plated on 24-well plates with an initial density of 3 000 cells/cm2 for adipogenesis and osteogenesis. When 100% confluence was reached, the medium was replaced by the respective differentiation medium (osteogenic (Gibco®) or adipogenic (Gibco®)). For chondrogenesis, a pellet of 0.3-0.5x106 cells was resuspended and droplets of this suspension were plated in a 24-well low attachment plate in chondrogenic medium (Gibco®). The medium was changed every 3 to 4 days.

II.3.1.1 Adipogenesis Staining

The cells were washed with PBS (Gibco®) and fixed with 2% paraformaldehyde (PFA) solution (Sigma®) for 30 minutes at room temperature (RT). After being washed with distilled water, the cells were stained with 0.3% Oil Red-O solution (Sigma®) for 1 hour at RT. Lastly, the cells were washed twice in distilled water and observed under a microscope.

II.3.1.2 Osteogenesis Staining

The cells were washed in PBS and fixed with 2% PFA for 5 minutes at RT. Afterwards, the cells were washed and kept in distilled water for 15 minutes. Then cells were stained and incubated with reagent X, containing 0.1M Tris-HCl solution (Sigma®), the substrate Naphthol AS MX-PO4 (0.1mg/mL) (Sigma®) dissolved in dimethylformamide (Fischer Scientific®) for alkaline phosphatase staining, and 0.6mg/mL Red Violet LB salt (Sigma®) for 45 minutes at RT. After incubation, the cells were washed three times, kept in distilled water and observed under the microscope. The Von Kossa staining protocol was then performed. The cells were washed with PBS and stained with 2.5% (w/w) silver nitrate (Sigma®) for 30 minutes at RT. The cells were washed three times with distilled water and observed under the microscope.

II.3.1.3 Chondrogenesis Staining

The cells were washed in PBS and fixed in 4%PFA for 30 minutes at RT. Then the cells were washed twice and incubated with Alcian Blue 1% (Gibco®) for 1 hour at RT. After being washed with PBS and kept in distilled water, the cells were observed under the microscope.

II.4 Immunophenotype characterization

Once isolated, ADSCs, BM-MSCs and SVF-derived cells were characterized immunophenotypically according to a panel of extracellular markers.65 BM-MSCs were additionally tested for CD14 to evaluate monocyte contamination, as well as HLA-DR, and SVF-derived cells were tested for satellite cell markers (CD34 and CD56). The cells were treated with 0.05% trypsin, accutase

or TrypLETM 1X (Gibco®) when stated, centrifuged for 7 minutes at 1250 rpm and resuspended in PBS. The cells were distributed into FACS tubes (BD FalconTM) and incubated for 15 minutes at RT in the dark with the following mouse anti-human monoclonal antibodies (1:10 dilution):

Table II.1. Panel of mouse anti-human monoclonal antibodies used to characterize MSCs and SVF-derived cells, their commercial brands, conjugated fluorophores and isotypes.

Surface Markers Brand Conjugated Fluorophore Isotype CD14 BioLegend® PE IgG1 CD31 BioLegend® PE IgG1 CD34 BioLegend® FITC IgG1 CD45 BioLegend® PE IgG1 CD56 BD Biosciences© PE IgG1 CD73 BioLegend® PE IgG1 CD90 BioLegend® FITC IgG1 CD105 BioLegend® PE IgG1 HLA-DR BioLegend® PE IgG2a IgG2a BD Biosciences© PE - IgG1/IgG1 BD Biosciences© FITC/PE -

Control groups were incubated with FITC-/ PE- conjugated mouse IgG1 isotype antibodies (1:10 dilution). After incubation, 2 mL of PBS was added to each FACS tube, and the cells were then centrifuged for 5 minutes at 1500 rpm and resuspended in 1% PFA. Quantitative analysis was performed using a FACSCalibur flow cytometer (Becton Dickinson®, San Jose, CA) and with FlowJo software (Version 8.8.6, Tree Star Inc., Ashland, Oregon, USA). MFI values were calculated by dividing the geometrical mean fluorescence values of the sample by the ones of the isotype alone.104

II.5 5-aza-2’-deoxycytidine treatment

II.5.1 Effect of 5-aza-2’-deoxycytidine on cell viability

ADSCs were seeded into 24-well plates (BD FalconTM) at cell densities of 3000 and 5000 cells/cm2 in DMEM+10%FBS (MSC-qualified). The cells were allowed to adhere to tissue culture flasks for 24 hours, after which the medium was replaced by myogenic differentiation medium composed by DMEM+2% HS (Sigma®) supplemented with 0, 2, 5, 7.5 or 10 μM of 5-AZAd (Sigma®).126 After 24 hours of 5-AZAd exposure, the medium was replaced by DMEM+2%HS alone. The cells were cultured for 2 weeks with medium changes every 3 to 4 days. At days 3, 5, 7, 12 and 14 after the 24-hour 5-AZA pulse, cell viability was assessed using the trypan blue exclusion test. The cells were washed with PBS and harvested with accutase for 7 minutes, or with 0.05% or 0.25% trypsin when stated. The cells were centrifuged for 7 minutes at 1250 rpm and resuspended in PBS.

II.5.2 Effect of 5-aza-2’-deoxycytidine on cell apoptosis and morphology

ADSCs were plated at a cell density of 5000 cells/cm3 on T75 flasks and 12-well plates (BD FalconTM) in DMEM+10%FBS (MSC-qualified). Myogenic induction was performed as mentioned

above, with cell exposure to 0, 2 or 10 μM of 5-AZAd in DMEM+2%HS. To quantify the percentage of apoptotic and necrotic cells after 2, 7 and 14 days of 5-AZAd exposure, the cells were stained with FITC-conjugated anti-Annexin V and PE-conjugated anti-PI antibodies (Kit from Life Technologies®, BD Biosciences©). The cells were analysed by flow cytometry within 1 hour. A cytoskeleton morphology assay was performed in order to assess possible morphology changes induced by 5-AZAd. Cells plated in 12-well plates were stained for F-actin with Phalloidin-Tetramethylrhodamine (TRITC) (1 μg/mL, Sigma®).

II.6 CD34+ SVF-derived cell sorting and myogenesis induction

Human SVF-derived CD34+ cells were isolated using CD34 MicroBeads (kit from Miltenyi Biotec®), according to the manufacturer’s instructions. Crude SVF cells and magnetically sorted cells from the enriched and depleted fractions were analysed by flow cytometry and plated in T25 flasks at 10 000 cells/cm2 in myogenic expansion medium156,157, composed by DMEM + 20%FBS + 10-6M dexamethasone (Sigma®) + 2.5 ng/mL bFGF (Sigma®). The effects of gelatin (0.2%, Sigma®) and fibronectin (50 μg/mL, Sigma®) coatings, as well as the use of high glucose (HG) (4.5 g/L) DMEM medium (Gibco®) (versus DMEM-LG, 1 g/l Glucose), were tested for the CD34-enriched and CD34-depleted fractions. Once the cells reached confluence, they were harvested with 0.05% trypsin for 7 minutes, tested for CD34 and CD56 expression by flow cytometry and plated at 4500 cells/cm2 in T25 flasks and at 3000 cells/cm2 in 12-well plates, in the same conditions. Myogenic induction was performed when the cells reached confluence, by the addition of DMEM-HG + 2%HS for 8 days with medium changes every 3 to 4 days. After 2, 4 and 8 days of medium replacement, the cells were tested for the expression of Pax7, MyoD, myogenin and skeletal MHC (when stated) by flow cytometry and immunofluorescence. Cells plated on 12-well plates were also stained with Phalloidin-TRITC (1 μg/mL, Sigma®) at day 8 of induction to assess cytoskeleton morphology and nuclei alignment and fusion.

II.6.1 CD34/CD56 decay assessment in normal and ultralow attachment plates

Fresh SVF cells were thawed and subjected to MACS sorting for CD34+ cell enrichment. Unsorted cells and cells from the positively selected fraction were plated at 10 000 cells/cm2 in 6-well plates and in 24-well ultralow attachment plates, in myogenic expansion medium and in skeletal muscle cell growth medium (Zen-Bio, Inc., Research Triangle Park, NC). After 3 and 4 days of culture, the cells plated in 6-well plates were harvested (0.05% trypsin for 7 minutes) and the expression of CD34 and CD56 was assessed by flow cytometry. For cells plated in ultralow attachment plates, a medium change was performed once by medium collection, centrifugation at 12500 rpm for 7 minutes and cell resuspension in fresh medium. After 6 days in culture, the medium was collected and cells in suspension were plated on gelatin-coated T12.5 plates. Aggregate adhesion was observed and after 4 days in culture the cells were harvested (0.25% trypsin for 10 minutes) and the expression of CD34 and CD56 was assessed by flow cytometry.

II.7 Smooth muscle differentiation induction

After isolating and expanding AT-SVF cells and BM-MSCs in DMEM+10%FBS (standard and MSC-qualified FBS, respectively), the cells were plated at 5000 cells/cm2 in T25 flasks and 12-well plates in the same conditions. When confluence was reached, the cells were exposed to smooth muscle induction medium, composed by DMEM-LG supplemented with 3%FBS (standard FBS) and 1, 10 and 30 μM of PD98059 (Calbiochem®, Merck Millipore©, 25 mM stock solution in DMSO) for 7 days. DMEM-HG supplemented with 3%FBS plus 10 μM of PD98059 was also tested in AT-SVF cells. Cells plated in DMEM+3%FBS, DMEM+10%FBS and DMEM-HG+10%FBS alone were tested as control groups. The expression of smooth muscle markers -SMA, calponin, SM-MHC and desmin (when stated) was evaluated after 2, 4 and 7 days of medium replacement by intracellular flow cytometry and immunofluorescence. Bladder SMCs plated in bladder SMC growth medium were used as positive control (both from Zen-Bio, Inc.). The expression of the previously mentioned markers was evaluated by flow cytometry after 14 days of culture in P5 SMCs and by immunofluorescence after 21 days of culture in P6 SMCs.

II.8 Immunofluorescence staining

Cells cultured on 12-well plates were washed with PBS and fixed with 4% PFA for 20-30 minutes at RT, followed by rehydration with PBS. Cell permeabilization was performed with Blocking Solution (10% NGS+ 0.1%Triton X100 (Gibco®)+PBS) for 15 minutes at RT. The cells were washed with PBS and incubated with the primary antibody in Staining Solution (5%NGS+0.1%Triton X100+PBS) at a concentration of 1μL/mL for 2 hours at RT. After being washed with PBS, the cells were incubated with the respective secondary antibody (1:500 dilution) for 1 hour at RT (see Table II.2). The cells were washed and incubated in a 1.5 μL/mL solution of 4',6-diamidino-2-phenylindole (DAPI, Sigma®) in PBS for 2 minutes for nuclei staining. The stained cells were washed with PBS and viewed with a CK40 Microscope equipped with DAPI, TRITC and FITC filters. Images were edited with Image-J (Version 1.48, NIH, Maryland, USA).

Table II.2. Panel of anti-human primary monoclonal antibodies and respective fluorophore-conjugated secondary antibodies used to stain cells for markers relevant for myogenic differentiation assessment, dilution used, commercial brand and isotype.

Primary Secondary Dilution Brand Isotype Dilution Brand Antibody Antibody Pax7 1:100 SCBT® Goat anti-mouse FITC IgG2a 1:500 Abcam®

MyoD 1:100 Abcam® Goat anti-rabbit Alexa 488/PE IgG 1:500 Invitrogen™ Myogenin 1:200 BD Biosciences© Goat anti-mouse Alexa 488/PE IgG 1:500 Invitrogen™ -SMA 1:40 SCBT® Goat anti-mouse Alexa 488/PE IgG 1:500 Invitrogen™ Calponin 1:100 Abcam® Goat anti-mouse Alexa 488/PE IgG 1:500 Invitrogen™ Desmin 1:200 Abcam® Goat anti-rabbit Alexa 488/546 IgG 1:500 Invitrogen™ SM-MHC 1:40 SCBT® Goat anti-mouse Alexa 488/PE IgG 1:500 Invitrogen™

II.9 Intracellular staining for flow cytometry

Intracellular staining was performed in order to assess the expression of skeletal muscle-specific markers (Pax7, Myod, MyoG and skeletal MHC), smooth muscle-specific markers (α-SMA, calponin and SM-MHC) and desmin. Cells cultured in T25 flasks were harvested with 0.05% trypsin for 7 minutes, centrifuged for 7 minutes at 1250 rpm and resuspended in PBS. The cells were permeabilized and fixed in FACS tubes using the BD Cytofix/Cytoperm™ Fixation/Permeabilization

Solution Kit (BD Biosciences©) according to the manufacturer's instructions. After the fixation/permeabilization steps, the cells were incubated with a primary monoclonal antibody, washed with PBS and permeabilization buffer, and then labelled with the respective secondary antibody (Table

II.2). Control tubes were incubated only with the secondary antibody. The cells were resuspended in PBS and analysed by flow cytometry within 1 hour using a FACSCalibur flow cytometer and FlowJo software. At least 1000 gated cells were considered for the flow cytometry analysis. MFI values were calculated as previously mentioned.

II.10 Decellularization of Porcine Urethras

Four urethras from euthanized F1 Large White X Landrace female pigs, weighting between 28 and 30 kg, were obtained from Faculdade de Medicina Veterinária da Universidade de Lisboa, in accordance with the Direcção Geral de Alimentação e Veterinária guidelines. The lower urinary tract (bladder and urethra) was harvested immediately after slaughter, frozen and transported to the laboratory in appropriate containers, and left overnight at -20ºC. Excess AT was removed with tweezers and the bladders were cut transversely 2 to 4 cm above the bladder neck. The urethras were cannulated from the bladder side (Jelco® I.V. Catheter, 14G, Smiths Medical, USA) and ligated with sutures (Ethicon™ A185 sutures, Johnson & Johnson, USA). The decellularization process was performed in a dynamic system. A catheter was connected to a peristaltic pump (Ecoline VC-380, Ismatec®, Wertheim, Germany) set at 40 mL/minute, through L/S 14 G tubing (Masterflex, Gelsenkirchen, Germany) and an adaptor. The urethras were placed inside a vessel with constant agitation (set at 340 rpm). Prior to the decellularization process, the urethras were submerged and perfused with distilled water for 24 hours. Decellularization was initiated once the water solution was replaced by 0.5% sodium dodecyl sulfate (SDS) solution (Sigma®) and constant perfusion of the urethra was allowed for at least two days, after which the direction of the perfusion was switched. The detergent solution was renewed everyday (24-hour cycles). By the end of at least 5 days, all 0.5% SDS solution was removed and replaced by distilled water for 60 minutes. Fresh distilled water was then pumped for 24 hours to remove detergent residues before the urethras were recovered and thawed at -20ºC.

III. Results and Discussion

In this section, the results regarding the isolation and characterization (Subsection III.1) of the cell populations that were subsequently used in skeletal and smooth muscle differentiation studies (Subsections III.2 and III.3, respectively) will be presented and discussed. Additionally, a brief description of a decellularization protocol adopted for the decellularization of porcine urethras and its results and efficacy are included (Subsection III.4).

III.1 Mesenchymal Stem/Stromal Cell isolation

Lipoaspirates obtained from healthy donors were previously processed and SVF cells were isolated according to the enzymatic isolation protocol from Shah and colleagues.154 The SVF cells were cryopreserved in vials and were subsequently used for the isolation of MSCs (when the culture medium included MSC-qualified FBS) or AT-SVF cells (when standard FBS, e.g. not MSC-qualified, was used) (Table III.1). The culture medium was changed every 3 to 4 days, allowing the non-adherent cells to be removed, until 80-90% confluence was reached.

Table III.1. Age and gender of the AT donors and the assays performed with each sample.

Donor Information Assays Performed SVF Donor Samples Age of Donor Gender of Skeletal Muscle Smooth Muscle Reference (years) Donor Differentiation Differentiation SVF-1 - Female Yes Yes SVF-2 38 Female Yes - SVF-3 30 Male Yes - SVF-4 52 Female Yes Yes SVF-5 36 Female Yes -

A BM sample collected from a 36-year-old male healthy donor was isolated in DMEM+10%FBS and the adherent cells displayed typical fibroblastic shapes. These cells were further used for a smooth muscle differentiation assay (Section III.3).

SVF cells plated in different medium formulations showed distinct morphology and proliferation characteristics. In particular, the use of MSC-qualified FBS or standard FBS affected the morphology of the isolated cells. When MSC-qualified FBS was used, the cells adhered after 24 hours, allowing the identification of typical MSC-shaped cells (displaying a “fibroblast-like” shape) after 4-5 days in culture (Figure III.1A). By contrast, when standard FBS was used, few adhered cells could be observed in the first 2-3 days of isolation, and their shape was more heterogeneous. Early in culture, some adherent cells looked roundish while others were more elongated. Elongated cells usually displayed long cytoplasmic edges. As the days in culture increased, the cells started elongating and a heterogeneous cell population was obtained. While most cells had fibroblastic shapes, some cells displayed pericyte-like morphologies with several cytoplasmic extensions or protrusions (Figure III.1B). Pericyte-like cells were obtained when three distinct medium formulations

supplemented with standard FBS were used (DMEM+10%FBS, DMEM+3%FBS and DMEM+20%FBS supplemented with dexamethasone and bFGF, or myogenic expansion media) but were not observed in SVF cells isolated with MSC-qualified FBS. These data may indicate that different cell populations are obtained from the SVF depending on the type of serum that is used in the isolation step, or that the MSC-like culture obtained with standard FBS is more heterogeneous.

A B

Figure III.1. P0 SVF-derived cells cultured for 6 days in DMEM medium supplemented with 10% of (A) MSC-qualified FBS and (B) standard FBS. Arrow points to a pericyte-like cell.

Among the cells isolated with standard FBS, the three medium formulations used in the studies performed herein also affected cell proliferation rate, shape and size. These changes were more noticeable with increased culture time. Cells plated with DMEM+3%FBS and DMEM+10%FBS tended to suffer from contact inhibition, usually reaching 90% confluence levels at most, and adopted a wider and flattened shape with prolonged culture time, which may be a sign of senescence. Conversely, cells plated in myogenic expansion medium sustained a highly proliferative state, frequently reaching 100% confluence, and maintained an elongated and spindle-like morphology. This differences in cell proliferation and morphology can be explained by the presence of potent mitogens (namely bFGF) in this medium formulation, which promote cell proliferation.

A B

Figure III.2. P0 SVF-derived cells cultured for 15 days in DMEM+3%FBS (A) and 11 days in DMEM+10%FBS (B). Some cells begin appearing more extended and flattened with increased culture time, possibly due to senescence.

III.1.1 Mesenchymal Stem/Stromal Cell Characterization III.1.1.1 Immunophenotypical Characterization

After reaching confluence, the cells isolated from the SVF and BM were harvested and characterized immunophenotypically using the surface antigens indicated by Bourin and co-workers108 and Dominici and colleagues,65 including: CD73, CD90 and CD105 (all expressed by MSCs), CD14 (expressed in monocytes), CD34 (hematopoietic progenitor/endothelial antigen also expressed in satellite cells), CD31 (platelet endothelial adhesion molecule), CD45 (lymphocyte common antigen, expressed in all hematopoietic cells) and HLA-DR (MHC class II cell surface receptor, not expressed in MSCs). According to the obtained flow cytometry results, the isolated populations expressed MSC markers at proportions indicated by the ISCT that allow MSC classification (>95% expression for CD73, CD90 and CD105 and <2% expression for CD45, CD31 and HLA-DR). The expression of CD14 was higher than the indicated by Dominici and colleagues (less than 2%) for the cells tested for this marker. However, this probably occurred because of the antibody clone used (HCD14), which was reported to stain cross-reactive MSC populations.158 CD34 expression was higher for AT-SVF cells. This marker is expressed in crude SVF and at variable levels in plated SVF-derived cells, depending on the culture conditions.108 The presence of low serum conditions seemed to enhance CD34 expression in AT-SVF cells when compared to cells plated in myogenic expansion medium (Table III.2).

Table III.2. Expression of the markers used in the identification of SVF-derived cells and MSCs, according to Bourin et al. and Dominici et al., by several P0 SFV-derived plate-adherent populations cultured in distinct media, and by P0 BM-MSCs. For AT-SVF cells in expansion medium n=3 (except CD14 and HLA-DR, for which n=1). FBS*= MSC-qualified FBS

Expression (%) AT-SVF cells AT-SVF cells AT-SVF cells ADSCs BM-MSCs Expansion DMEM+3%FBS DMEM+3%FBS DMEM+10%FBS* DMEM+10%FBS* Media 15 day culture 25 day culture CD14 5.5 11 - - 24 CD31 - 0.10±0.2 0.7 0 - CD34 3.8 3.6±2.1 32 20 1.0 CD45 0.9 0±0 0.50 0 2.0 CD73 99 99±0.4 - 98 99 CD90 99 97±2.3 97 99 98 CD105 99 98±0.5 - 93 99 HLA-DR 0.30 0 - 0 0

III.1.1.2 Multilineage Differentiation Assays

The isolated ADSCs, AT-SVF cells and BM-MSCs were characterized in terms of multilineage differentiation ability, i.e. the ability of MSCs to differentiate into adipocytes, chondrocytes and osteocytes under specific in vitro tissue culture-differentiating conditions. This is also one criteria for defining MSCs, according to the Dominici’s criteria.65 As shown in Figure III.3, the induction of chondrogenesis resulted in the formation of spheroids, which marked positively for Alcian Blue,

indicating the presence of cartilage mucopolysaccharides namely glycosaminoglycans. On the other hand, lipid droplets could be detected after the induction of adipogenic differentiation by Oil Red-O staining. Finally, osteoblasts were detected by alkaline phosphatase and Von Kossa stainings. Alkaline phosphatase stained pre-osteoblasts, while the Von Kossa staining allowed the visualization of calcium deposits, indicative of matrix mineralization. The results obtained below indicate that cells isolated from the BM and from the adipose SVF (whether standard FBS or MSC-qualified FBS was used) are multipotent.

BM-MSCs ADSCs AT-SVF cells

Osteogenic

Differentiation

Adipogenic

Differentiation

Chondrogenic

Differentiation

Figure III.3. Multilineage differentiation characterization of BM-MSCs, ADSCs and AT-SVF cells. BM-MSCs and ADSCs were isolated and plated in DMEM+10%FBS (MSC-qualified), while AT-SVF cells were isolated in DMEM media supplemented with 20%FBS (not MSC-qualified), dexamethasone and bFGF, before confluence was reached and the medium was replaced to the respective differentiation medium. Scale bar = 100 μm.

There is some controversy in the literature regarding the currently used criteria for defining MSC populations in vitro, mainly because of the fact that no exclusive marker for MSC identification is presently known. Several authors have questioned the specificity and reliability of the relatively minimal criteria in the ISCT definition of MSC, as these criteria refer to widely shared properties of cells from connective tissues in general, and thus do not to allow to say, without ambiguity, that cells meeting the criteria are only MSC and not some other cell type.66,159 Since then, the ISCT has published a joint statement together with the International Federation for Adipose Therapeutics and Science (IFATS) with several criteria that can be used to identify SVF cells and the respective stromal

adherent cell fraction (e.g. ADSCs).108 The SVF harbours a highly heterogeneous cell population, composed by pericytes (3-5%), endothelial cells (10-20%), stromal cells (15-30%) and hematopoietic-lineage cells (25-45%).108 The degree of heterogeneity depends, in part, on the AT depot site and the digestion protocol.108 Although not mentioned in this official statement, other adherent cell subpopulations such as SMC-like cells and myotube-forming cells have been successfully isolated from the SVF.95,110,159 Thus, it is possible that the SVF includes small percentages of myogenic-committed cells that can be useful in therapeutic approaches targeting muscle cells. The true in vivo identity of AT progenitors has also been a cause of debate in the literature. ADSC location within human AT was reported to be essentially perivascular, which led some authors to propose that ADSC are vascular precursor (stem) cells - rather than “fat stem cells” - at various stages of differentiation capable of differentiating in situ into host tissue-specific cell types.95 Analogously, the discovery of a subpopulation of perivascular cells in multiple human organs that expressed both MSC and pericyte markers led Arnold Caplan to speculate if all MSCs are, in fact, pericytes.94,96

Given the inherent heterogeneity of the SVF, the isolation of morphologically and phenotypically distinct populations is somewhat expected, especially when growth conditions not optimized for MSCs are employed. The immunophenotypic characterization assay performed herein suggests that stromal cells are the main components of the isolated SVF-derived cells in all the medium formulations tested (see Table III.2). Moreover, CD31 expression levels were lower than 1%, indicating that the adherent cell fraction obtained was not contaminated with endothelial cells (which are present at 10-20% in the SVF). Still, given the phenotypic similarities between MSCs and other cell types, particularly pericytes and fibroblasts94,160, non-stromal cells are likely to have been isolated when not-MSC qualified FBS was used, resulting in the appearance of pericyte-like cells. Despite the morphology differences observed, lineage differentiation studies indicate that all the isolated populations (denominated here as BM-MSCs, AT-SVF cells and ADSCs) are capable of mesenchymal trilineage differentiation (Figure III.3).

Cell morphology is an important parameter not only to evaluate cell isolation and differentiation but also senescence. Here, changes in cell morphology with increasing culture time were observed in medium not supplemented with mitogenic factors. A study in which Wharton’s Jelly-derived MSCs and ADSCs were compared morphologically and in terms of proliferation capacity over long-term cultures identified three main cell morphologies: (a) spindle-shaped, elongated; (b) polygonal, with few, small membrane protrusions and (c) flattened, with irregular shape and large nuclei.161 Increasing passaging of ADSCs resulted in gradual acquisition of the last mentioned type of morphology together with increases in cell size, which the authors concluded to be signs of senescence, as cells with those characteristics were 훽-galactosidase positive.161 One other study reported that independently of the plating density and cell passage, the proportion of flat cells increases over culture time in

BM-MSCs.162 However, flat clones were able to give rise to spindle-like cells, indicating that flat MSCs are not terminally differentiated and still hold some clonogenic potential.162 Although most researchers believe that spindle-like cells represent the true multipotent subpopulation, no correlation between cell shape and “stemness” was yet established.162 This type of information would be of extreme importance for the development of protocols aiming the myogenic differentiation of MSC populations, especially since flat cells are thought to be more mature cells and thus possess diminished differentiation potential.

III.2 Differentiation of Mesenchymal Stem/Stromal Cells into Skeletal Muscle Cells

III.2.1 Myogenic induction using 5-aza-2’deoxycytidine

5-azacytidine (5-AZA) and its deoxy analogue 5-aza-2’deoxycytidine (5-AZAd) are described in the literature as myogenic inducers in BM-MSCs, ADSCs and UCM-MSCs in both in vitro and in vivo studies at concentrations ranging from 0.1 to 10 μM. However, since these compounds indirectly induce alterations in the chromatin conformation of DNA, potentially leading to mutations or cytotoxic effects, cell viability assays (which are currently lacking in the literature in myogenic differentiation studies involving these two compounds) should be performed. For myogenic induction, the protocol used by Drost and colleagues was adopted.126 Briefly, ADSCs were allowed to adhere to the plastic surface of tissue culture plates for 24 hours, after which 5-AZAd diluted in DMEM+2%HS was added to cells for 24 hours (day 0), followed by medium replacement to DMEM+2%HS alone (day 1). Thus, day 2 after 5-AZAd exposure (or treatment) refers to 24 hours in the presence of the demethylating agent plus 24 hours without the agent. The same rationale was considered for other time points (i.e. 7 and 14 days of treatment). No 5-AZAd was added to control cells.

III.2.1.1 Cell Viability Assay

A preliminary study was conducted with P2 ADSCs in order to determine if a 24-hour pulse of 5-AZAd at concentrations of 2, 5, 7.5 and 10 μM led to cytotoxic effects, through a cell counting assay. Cell morphology was also evaluated by optical microscopy. Additionally, two cell densities were tested (3000 and 5000 cells/cm2). ADSCs treated with 7.5 and 10 μM of 5-AZAd developed extremely thin and elongated morphologies, as opposed to control ADSCs, which maintained their fibroblastic shape (Section III.2.1.3). Additionally, when harvested with accutase and viewed on a haemocytometer under an optical microscope, two important observations were made: firstly, the cells exposed to 7.5 and 10 μM were still plate-adherent after accutase treatment; secondly, the unattached cells exposed to 7.5 and 10 μM of 5-AZAd displayed a very thin and elongated morphology, instead of the typical round shape adopted by MSCs upon harvesting and myoblasts after enzymatic digestion treatment, and stained positively for trypan blue (Figure III.4). Both observations represent abnormal cell behaviours, suggesting that 5-AZAd at 7.5 and 10 μM induced some cell cytotoxicity.

A B

Figure III.4. Effect of 5-AZAd on cell morphology and viability. ADSCs were cultured for 7 days in DMEM+2%HS, harvested and placed on a haemocytometer for the trypan blue exclusion test. A) Control cells displayed typical round shapes; B) Cells exposed to 10 μM of 5-AZAd displayed a very thin elongated morphology and stained positively for trypan blue.

At day 7 the cells were harvested with 0.05% trypsin for 5 minutes to evaluate if a harsher enzymatic digestion protocol would efficiently detach the cells treated with 7.5 and 10 μM of 5-AZAd from the plates. No changes in cell detachment efficiency were observed, as these cells remained adherent to the plates. At day 14, 0.25% trypsin was tested but with the same outcome. It was concluded that for future studies, a harsher enzymatic formulation should be used to detach 5-AZAd treated cells, such as TrypLETM (1X). Interestingly, accutase and trypsin were used for cell detachment in several reviewed studies and there was no reference in the literature to unusual effects of 5-AZAd in cell adherence to plastic or trypsin resistance. Thus, it is possible that increased cell attachment could be a consequence of unspecific effects induced by the demethylation agent used herein. A pronounced decrease in cell viability was confirmed by cell counting for 5-AZAd at 7.5 and 10 μM, for which no viable cells were observed after day 1 of 5-AZAd treatment. In fact, in these conditions, all cells stained positively for trypan blue at days 3, 5, 7, 12 and 14 after 5-AZAd exposure (Figure III.5).

Figure III.5. Cell viability assay for ADSCs cultured in DMEM+2%HS previously exposed to 5-AZAd. Cells were treated with various concentrations of 5-AZAd for 24 hours (2, 5, 7.5 and 10 μM) and cell viability was measured by counting cell numbers at 1, 3, 5, 7, 12 and 14 days thereafter. Two cell densities were tested: 3000 cells/cm2 (left) and 5000 cells/cm2 (right). No cell bars indicate zero cell viability, except for 5 μM.

No plated cells could be observed for the 5 μM of 5-AZAd at days 3, 5, 7, 12 and 14, probably due to an error in the initial plating of the cells, therefore, the cell viability for this concentration could not be evaluated. Control cells and cells exposed to 2 μM of 5-AZA were similar in terms of cell morphology and viability. Regarding the two cell densities tested, the results showed that with 5000 cells/cm2 higher cell viabilities were achieved in comparison with 3000 cells/cm2 (Figure III.5). From the results obtained in this assay, two hypotheses were formulated: (i) 5-AZAd at concentrations above 7.5 μM induced early cell apoptosis, between 24 and 72 hours after 5-AZAd treatment; (ii) 5-AZAd at concentrations above 7.5 μM induced abnormal cytoskeleton alterations that affected cell morphology. These two hypotheses were tested through a cell apoptosis assay and a phalloidin staining assay, respectively.

III.2.1.2 Cell Apoptosis Assay

P1 ADSCs were plated on 12-well plates, exposed to 2 and 10 μM of 5-AZAd for 24 hours and subjected to a propidium iodide (PI)/Annexin V flow cytometry analysis to prove 5-AZAd at 10 μM induced early cell apoptosis. This assay gives more information than the trypan blue exclusion test (which only detects viable and necrotic cells), since the PI/Annexin V flow cytometry assay allows viable, apoptotic and necrotic cells to be distinguished through the use of FITC-labelled annexin and PI molecules. When cells start becoming apoptotic, phosphatidyl serine phospholips flip to the extracellular surface of the cell, and annexin can bind to them, thus detecting early apoptotic cells.163 PI, a fluorescent dye that binds to DNA, enters the cell and binds to DNA when the cell is late apoptotic and the membrane is no longer intact, thus allowing the detection of necrotic cells.163

After 2, 7 and 14 days of 5-AZAd treatment the cells were analysed and it was possible to discriminate between viable (Annexin V-/PI-), apoptotic (PI-/Annexin+) and necrotic (PI+/Annexin-, PI+/Annexin+) cells (Table III.3 and Figure III.6).163

After day 2 of 5-AZAd treatment, the cells exposed to 10 μM were necrotic (92%), while most of the cells exposed to 2 μM were viable (89%), as well as the control cells (94%). These results prove that 10 μM of 5-AZAd induces early cell apoptosis even before 48 hours after 5-AZAd treatment, as the cells were already necrotic at this timepoint. Moreover, since at this concentration only 8% of the cells were viable, cell apoptosis was not assessed for this 5-AZAd concentration at days 7 and 14, since at least 1 000 cells (gated) are required for an accurate flow cytometry analysis to be performed.

The number of cells rose as the number of days in culture increased, indicating that the cells proliferated throughout the assay. Cells treated with 10 μM 5-AZAd did not achieve the minimal number of gated events for flow cytometry analysis (Table III.3). Furthermore, the dot plot chart obtained for this population of cells indicated that it was mostly composed by cell debris, since this population was located on very low values of forward scatter (FSH, i.e. measure of cell size) and side scatter (SSH, i.e. cell complexity) (Figure III.7). This was also an indicator that the cells were no longer viable.

Table III.3. Effect of 5-AZAd on cell viability. PI/Annexin V flow cytometry analysis for ADSCs cultured in DMEM+2%HS, treated with 2 and 10 μM of 5-AZAd, at days 2, 7 and 14 after treatment. The expression of annexin and PI served as a measure of cell viability, apoptosis and necrosis.

Day 2 Day 7 Day 14

Control 2 μM 10 μM Control 2 μM Control 2 μM

Necrotic (%) 5.7 5.2 92 4.0 4.9 2.0 1.8 Apoptotic (%) 0.67 6.1 0 0.72 0.84 0.29 0.46 Viable (%) 94 89 8.0 96 94 98 98 Gated Cells 9564 9417 88 9387 9389 8656 8496

Figure III.6 Effect of 5-AZAd on cell viability.

A B

Figure III.7. Flow cytometry dot plot charts of ADSCs cultured for 2 days in DMEM+2%HS after 5-AZAd treatment. A) control ADSCs; B) ADSCs treated with 10 μM 5-AZAd.

These results indicate that it is not viable to treat MSCs with concentrations of 5-AZAd equal or above 10 μM, since it leads to early cell necrosis. Few studies regarding the effects of 5-AZA on cell viability were found in the literature. The 50% inhibitory concentration of 5-AZA after 24 hours of induction was reported to be around 40 μM, with little decrease in viability at concentrations below 10 μM.130 The fact that cells treated with 10 μM of 5-AZA have been successfully used in the literature in flow cytometry analysis also suggests that 5-AZA might be significantly less toxic than the analogue used herein. Still, 5-AZA can be cytotoxic, as Tomita and colleagues reported that BM-MSCs

incubated with concentrations of 5-AZA above 20 μM not only appeared abnormal but also more than 50% necrosed.127 Additionally, the authors observed damage in BM-MSCs cultured with 5 and 10 μM 5-AZA for 24 hours.127

III.2.1.3 Microscopic Analysis of Cell Morphology

Control ADSCs (Figure III.8A) and ADSCs exposed to 2 μM of 5-AZAd maintained a fibroblast-like shape, while ADSCs exposed to 10 μM appeared unhealthy and displayed very heterogeneous morphologies. Essentially two types of cells were observed: very thin and elongated cells and broad, irregular cells, sometimes with multiple cytoplasmic projections (Figure III.8B). Some multinucleated cells could be observed in all conditions.

A B

Figure III.8. Alterations in cell morphology and confluence of ADSCs cultured in DMEM+2%HS after two weeks of 5-AZAd treatment. (A) Control ADSCs and ADSCs exposed to 2 μM (not shown) reached high confluence levels and maintained a fibroblast-like shape. (B) ADSCs exposed to 10 μM and 7.5 μM (not shown) of 5-AZAd achieved low confluence levels and two types of cell shapes could be discerned: thin elongated cells (black arrow) and broad irregular cells (white arrow).

It has been described that 5-AZA-treated BM-MSCs (3, 5 and 10 μM of 5-AZA) show more heterogeneous morphologies and alterations in size and growth properties.133 In particular, the presence of two types of cells, very thin and elongated vs. broad and irregular, is in accordance with this previous report.133 Although the authors observed that the thin elongated cells were able to replicate faster and form colonies, no relation between these two types of morphologies and cell function, growth or differentiation has been established.

III.2.1.4 Phalloidin Immunofluorescence Staining

In order to visualize possible alterations in the cytoskeleton morphology of cells exposed to 5-AZAd, FITC-phalloidin and DAPI were used to stain their cytoskeleton (F-actin) and nuclei, respectively. Control cells and cells treated with 2 μM 5-Azad (Figure III.9) were similar in terms of morphology and confluence throughout the culture time. Very thin cells could be observed 2 days after 5-AZAd treatment with 10 μM, which is in accordance with the observed cell morphology under the

optical microscope. No fluorescence was detected at days 7 and 14 after 5-AZAd treatment (10 μM), probably due to cell necrosis, as shown in the cell apoptosis assay. These results indicate that after 48 hours of 10 μM 5-AZAd treatment severe cell morphology alterations are induced.

Figure III.9. Effect of 5-AZAd on cell morphology of ADSCs. Control ADSCs and ADSCs treated with 2 μM 5-AZAd were positive for phalloidin-stained F-actin (red), and DAPI-stained nuclei (blue) at days 2 (top), 7 (middle) and 14 (bottom). ADSCs treated with 10 μM 5-AZAd stained positively at day 2, but no fluorescence was detected thereafter. Scale bar = 100 μm.

III.2.1.5 Surface Marker Expression

The expression of surface markers was assessed by flow cytometry at days 2, 7 and 14 after 5-AZAd treatment. Cells treated with 10 μM of 5-AZAd did not achieve sufficient cell numbers for a reliable flow cytometry analysis at day 2 after treatment (Table III.4). As the cell apoptosis assay indicated that the cells exposed to this concentration of 5-AZAd were no longer viable, their surface marker expression was not assessed at days 7 and 14. The expression of the following markers was evaluated (Table III.4 and Figure III.10): CD73, CD90, CD105, CD34, CD45 and CD56 (neural adhesion molecule, expressed in skeletal muscle). CD56 was evaluated as a marker of myogenesis, in order to assess if 5-AZAd induced myogenesis.164

Table III.4. 5-AZAd effect on surface marker expression after 2, 7 and 14 days of treatment in ADSCs cultured in DMEM+2%HS.

Control 2 μM 10 μM

Day 2 (%) Day 7 (%) Day 14 (%) Day 2 (%) Day 7 (%) Day 14 (%) Day 2 (%)

CD73 100 100 100 100 100 100 83 CD90 98 90 92 99 87 72 51 CD105 100 93 96 100 72 92 79 CD34 11 8.4 4.5 10 6.2 3.8 0.89 CD56 0.37 0.19 0.10 0.13 0.11 0 0 CD31 0.23 0.16 0 0.82 0.29 0.12 0 CD45 0.32 0 0 0 0 0.09 17

Figure III.10. 5-AZAd effect on surface marker expression after 2, 7 and 14 days of treatment in ADSCs cultured in DMEM+2%HS.

5-AZAd-treated cells decreased the expression of MSC markers CD73, CD90 and CD105. This decrease was observed 2 days after treatment for ADSCs treated with 10 μM of 5-AZAd, but only after day 7 for cells treated with 2 μM of 5-AZAd. Moreover, a significant increase in CD45 expression was detected in ADSCs treated with 10 μM of 5-AZAd (17%). One study showed that MSCs expressed CD45 after 5-AZAd induction (1 μM) through immunohistochemical staining and real time PCR techniques.165 However, this increase in CD45 is probably not associated with haematopoietic lineage differentiation but rather with apoptosis, since CD45 is thought to have a role in apoptosis regulation.166 Control ADSCs and ADSCs treated with 2 μM of 5-AZAd expressed CD34 at day 2 after 5-AZAd treatment, and the expression of this marker decreased as the culture days increased. Importantly, no CD56 expression was detected, suggesting that 5-AZAd did not induce myogenesis. Still, many reports claim that 5-AZA induces MSCs to differentiate into myogenic lineages, although contradictory conclusions were found amongst the studies present in the literature (see Section I.2.1.3 – 5-Azacytidine to Induce Myogenesis).67,124–132,134 The fact that 5-AZA is being studied for both cardiomyocyte and myogenic differentiation proves this lack of consistency. Furthermore, the results are difficult to compare, as different research groups use different culture media compositions and distinct markers to induce and prove myogenic differentiation, respectively.

Importantly, although described in the literature as being effective in inducing myogenesis, the treatment of MSCs with 5-AZA presents one great disadvantage: the unspecific mechanism of action of this antimetabolite. In fact, the mechanism through which 5-AZA induces myogenesis is not yet understood, although some studies have suggested potential mechanisms for its action, such as inducing the demethylation and activation of a myogenic determination locus that is usually transcriptionally inactive.167

Wakitani and co-workers were the first to report that an immortalized line of rat BM-MSCs could be differentiated into contractile muscle cells by using 5-AZA (concentrations ranging from 3 to 50 μM) or 5-AZAd (from 0.1 to 10 μM). Interestingly, of the concentrations tested, 0.3 μM of 5-AZAd gave the highest incidence of myogenic conversion, according to the authors.67 Analogously, in 1999, Makino and colleagues successfully differentiated immortalized rat BM-MSCs into contractile cardiomyogenic cells expressing desmin, α-MHC and α-actin using 3 μM 5-AZA.128 Since then, 5-AZA and its deoxy analogue have been widely tested in immortalized and primary cell lines. Liu and collaborators concluded that rat BM-MSCs could not be induced to differentiate into cardiomyocytes using 5-AZA unless the cells were immortalized.133 This is a plausible hypothesis, since immortalized cell lines are more prone of overcoming the toxic effects of molecules such as 5-AZA than primary cell lines, and could eventually be reprogrammed into myogenic differentiation in response to 5-AZA. It is also possible that the cells differentiate due to a random reprogramming of gene expression induced by the drug, rather than to the cell’s ability towards myogenic differentiation.133 As a cytidine analogue, 5-AZA is capable of being incorporated into the DNA, subsequently acting by covalently inhibiting DNA cytosine methyltransferases. From a regulatory perspective, it would be very challenging to approve a treatment modality involving the use of cells exposed to a molecule that can potentially give rise to unpredictable effects, such as toxicity or tumorigenicity. This demethylation agent may cause an unregulated expression of various genes, disabling its use in therapeutic contexts, in which the cells cannot present any kind of unexpected behaviours.168 Consequently, research in the field of muscle engineering using MSCs as a source of muscle-like cells should focus on the establishment of novel protocols adequate for clinical applications.

In conclusion, in the experiments presented herein, 5-AZA did not induce myogenesis and at 10 μM led to early cell necrosis and immunophenotypical and cell morphology alterations. The study presented herein has limitations such as lack of statistical relevance, as only one population of ADSCs was tested immunophenotypically, and the fact that only one marker was used to detect myogenesis (CD56 expression). Still, the results presented in this thesis give a new insight about the effects of 5-AZAd on cell apoptosis and morphology. Moreover, it is possible that 5-AZAd affects the expression of cell adhesion molecules, as standard enzymatic digestion procedures were not sufficient to detach 5-AZAd-treated cells from culture plates, a phenomenon that was not described in the reviewed literature.

III.2.2 Myogenic induction of CD34+ cells from the Stromal Vascular Fraction

One particular study confirmed the presence of cultured Pax7-expressing AT-SVF cells that spontaneously differentiate into myotube-like cells in the presence of myogenic differentiation medium alone.110 However, their percentages within the SVF were reported to be very low (0.001% of total plated cells) and their characterization concerning marker expression is not known, making it difficult to isolated them. Moreover, an extensive literature search indicated that myogenic-committed cells have not yet been identified in uncultured SVF. Thus, with the aim of assessing if myogenic-committed cells exist in the human white AT, intracellular flow cytometry staining for Pax7 was performed in a fresh SVF sample in order to identify Pax7-expressing cells (Figure III.11).

B SVF Subpopulation

1 2 3

Gate Frequency (%) 41.9 8.57 11.7 Gated events 5912 1210 1653 Pax7 expression (%) 2.0 7.7 5.3 Pax7-expressing 1.05 Cells (total, 105)

C 100 100 100

80 80 80

x x a

a 60 60 a 60

M M

f f

o o

% % 40 40 % 40

20 20 20 9.34 2.42 6.23

0 0 0 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 Pax7 (FITC) Pax7 (FITC) Pax7 (FITC)

Figure III.11. Evaluation of Pax7 expression in an uncultured SVF sample. (A) Gated flow cytometry dot plot chart in which 3 subpopulations could be discerned; (B) Pax7 expression values and number of Pax7-expressing cells in a total of 49.6x105 viable cells (C) Pax7 expression histograms for subpopulations 1, 2 and 3 (left to right). Red lines = control; Blue lines = sample.

Three subpopulations were identified in the dot plot chart (Figure III.11A), which were not evenly enriched with Pax7+ cells. Subpopulation 2, placed at low FSH and intermediate SSH values, corresponding to cells with low size but with some complexity, was the most enriched population (7.7%), followed by subpopulation 3 (5.3%) and subpopulation 1 (2%). The identification of a considerable percentage of Pax7-expressing cells in the SVF, which translated into numbers corresponded to 1.05x105 cells in a total of 49.6x105, distributed into 3 gated subpopulations, remarkably indicates that myogenic-committed cells do exist in vivo in the adult AT and that these cells can potentially be isolated. Therefore, a method for further isolating Pax7+ cells from the SVF was aimed. However, as it is not possible to isolate/sort cells based on the expression of intracellular

markers, these cells could not be sorted according to Pax7 expression. An extracellular maker was thus needed for this purpose.

Several protocols for satellite cell/skeletal muscle stem cell isolation have been successfully established. Among them, satellite cell sorting through CD34/CD56 positive selection is a common method that has allowed the isolation of satellite cells with different commitment levels and differentiation potentials.156,157 In particular, it has been shown that CD34-CD56+ satellite cells are purely myogenic, while CD34+CD56+ satellite cells are also capable of adipogenic differentiation.157 Based on this information, enzymatically processed SVF cells were submitted to CD34 cell sorting through MACS with the aim of potentially isolating Pax7-expressing cells. This methodology was repeated and the obtained cells were used for several assays (Table III.5). The cells obtained after plating will be designated here as AT-SVF cells, as standard FBS was used for their isolation.

Table III.5. Experiments performed with cells from different SVF donors. *Due to a shortage in SVF sample numbers, a cell pool combining samples SVF-4 and SVF-5 had to be performed.

Experiments Performed SVF Sample CD34/CD56 Low attachment Myogenic Differentiation Pax7 expression MACS Reference Expression plate cultivation Assay SVF-1 - Yes - - - SVF-2 Yes - Yes - - SVF-3 Yes - Yes - Yes SVF-4 - Yes Yes Yes - Cell pool* Yes - Yes - Yes

III.2.2.1 Expression of myogenic markers in SVF cells before and after MACS

Fresh SVF cells were subjected to MACS in order to sort cells according to CD34 expression (Table III.6). CD34 and CD56 expression was then determined for the CD34 positive and negative fractions through flow cytometry analysis. The term “CD34-enriched” and “CD34-depleted” refers to the initial status of fresh cells (and not to cells after passages in culture). The term “CD34+ cells” refers to cells that effectively expressed CD34 (confirmed by flow cytometry). Of notice, CD34 expression decreased substantially in the first passages in culture.

The yield of the sorting process could be obtained for two of the sorted SVF samples. This yield represents the number of CD34+ cells present in the CD34-enriched fraction divided by the number of CD34+ cells present in the initial SVF sample. Both values were calculated by multiplying the total number of cells by the corresponding percentage of CD34+ cells (obtained by flow cytometry analysis). The yield values obtained (33% and 58%) seem reasonable, when compared to recovery values obtained for the same target molecule and MACS system in the literature (78% after one cycle and 52% after two cycles, for umbilical cord blood cells).169

Table III.6. MACS yield for samples subjected to MACS. Total cell numbers of the SVF samples and the CD34 enriched and depleted fractions acquired after MACS were calculated by viable cell counts using the trypan blue exclusion test. The numbers of CD34+ cells within the samples and/or the positive fractions were calculated using values acquired by flow cytometry.

Cells (106) SVF Sample Total Cells (106) CD34+ cells (106) Cells (106) CD34+ cells (106) Purity MACS CD34- Reference Before MACS Before MACS CD34+ fraction After MACS (%) Yield (%) fraction SVF-2 8.6 - 1.6 1.5 94 3.4 - SVF-3 5.2 3.0 1.3 1.0 77 2.0 33 SVF-4 4.8 - 0.57 - - 0.33 - Cell pool 4.1 2.0 1.5 1.2 80 0.60 58

After the sorting process, CD34 and CD56 expression was assessed in the CD34-enriched and depleted fractions obtained for samples SVF-2, SVF-3 and the cell pool (Figures III.12-III.14).

1000 1000 CD34-enriched CD34-depleted CD34-enriched CD34-depleted A B C 800 800

% MFI % MFI r r

e e t t

t 600 t 600 a a

c c CD34 98 90 23 3.4

S S

e e d d

i 400 i 400

S S CD56 8.6 2.0 0.20 1.0

200 200 Gated 16 802 21 112 66 22.1 events 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Forward Scatter Foward Scatter

Figure III.12. Flow cytometry dot plot charts of CD34-enriched (A) and CD34-depleted (B) fractions acquired after MACS for sample SVF-2. (C) Expression of CD34 and CD56 (in percentage and MFI) for both fractions.

1000 1000 1000 A Unsorted SVF B CD34-enriched C CD34-depleted

800 800 800

r r r

e e e

t t t t

t 600 t 600 600

a a a

c c c

S S S

e e e

d d d i i 400 i 400 400

2 13.2 11.42

S S S

200 200 200 49.51 1 45.6

14.9 0 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 Forward Scatter Forward Scatter Forward Scatter

CD34-depleted D E 1 (%) 2 (%) Unsorted CD34-enriched CD34-depleted D SVF 1 2 1 2 CD90 3.1 14.1

% MFI % MFI % MFI % % CD73 7.2 3.8

CD34 77 26 97 75 89 32 1.6 14 CD31 32.3 20.4 CD56 32 3.4 94 11 33 5.3 11 0.70 CD45 96.2 36.9 Gated 931 6449 1713 1208 918 CD105 1.4 1.5 events

Figure III.13. Flow cytometry dot plot charts of unsorted SVF cells (A), and cells from the CD34-enriched (B) and CD34-depleted (C) fractions obtained after CD34 sorting for sample SVF-3. Two subpopulations (1 and 2) could be discerned in the CD34-enriched and CD34-depleted fractions. (D) Expression of CD34 and CD56 (in percentage and MFI). (E) Extracellular marker characterization of cells from the CD34-depleted fraction.

1000 1000 1000 A Unsorted SVF B CD34-enriched C CD34-depleted

800 800 800

r r r

e e e t t t

t 600 t 600 t 600 a a a

c c 2 c

S S S

e e e d d d

i 400 i 400 20.3 i 400 S S S

200 200 42.2 200 45.8 1 21.7 0 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 Forward Scatter Forward Scatter Forward Scatter

CD34-enriched Unsorted CD34- SVF depleted 1 2 Figure III.14. Flow cytometry dot plot charts of unsorted SVF cells % MFI % % MFI % MFI (A) CD34-enriched (B) and CD34-depleted (C) fractions obtained after CD34 sorting of a cell pool of SVF cells. Two subpopulations CD34 75 28 6.6 93 80 85 30 (1 and 2) could be discerned in the CD34-enriched fraction. (D) CD56 0.50 1.8 0.40 6.8 2.6 3.1 1.7 Expression of CD34 and CD56 (in percentage and MFI). Gated events 5217 5390 6327 3047

Over 70% of cells were CD34+ in the two samples of unsorted SVF analysed by flow cytometry, while CD56 expression was more divergent (32% and 0.50%). CD34 expression in the enriched cell fractions reached values around 90%. Some CD34 expression was detected in the negatively sorted fractions, which may have lowered the yield of the separation process. Still, it is possible that these cells are CD34dim (i.e. expression slightly increased when compared to the negative control). Since the magnetic sorting process depends on the number of magnetically-labelled antibody molecules bound to the desired target protein (which translates into stronger or weaker retentions in the column), cells with fewer CD34 molecules on their surface are more prone to be washed from the column. The results shown in Figure III.13E indicate that hematopoietic cells are the main constituents of the CD34 negative fraction. In one of the positively sorted samples, a subpopulation of cells positive for both CD34 and CD56 was obtained (Subpopulation 1, Figure III.13B). To the extent of the literature reviewed herein, besides satellite cells, only myoendothelial cells are known to simultaneously express CD34 and CD56.170,171 These results indicate that a population with myogenic potential may have been sorted by performing CD34 MACS. However, the results show clear donor-to-donor variability in regard to the obtained populations and their levels of marker expression. Based on these results, and with the aim of determining if sorting CD34 positive cells is, in fact, a method of enriching the percentage of Pax7-expressing cells from the SVF, MACS was performed and the levels of Pax7 were assessed before and after the process, for both CD34-enriched and depleted fractions (Figure III.15). The results indicate that sorting CD34-positive cells allowed the enrichment of Pax7-expressing cells: the expression of this phenotype increased over 7-fold in the CD34 positive fraction (4.3%) in comparison to the initial SVF sample (0.56%). On the other hand, no significant Pax7 expression was detected in the CD34 negative fraction, which suggests that most Pax7-expressing cells present in the tested SVF sample express CD34, indicating that they may hold an immature satellite cell-like phenotype (CD34+CD56+).

1000 1000 1000 A Unsorted SVF CD34-enriched CD34-depleted

800 800 800

r r r

e e e

t t t t

t 600 t 600 600

a a a

c c c

S S S

e e e d

d d 2 i

i 400 i 400 400

S S S 19.5 200 200 200 1 39.1 53.6 35.9 0 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 Forward Scatter Forward Scatter Forward Scatter

C Unsorted CD34- CD34-depleted 100 SVF Enriched 100 1 2 B Unsorted SVF CD34-enriched 80 80 Gate 39.1 53.6 35.9 19.5

x Frequency (%)

a 60 a 60

M M

f f

o Gated events 6232 10381 2224 4098 o

% 40 % 40 Pax7 0.56 4.2 0 0.26 20 20 Expression (%) 1.46 5 Pax7-expressing 0 0 4 2.72 2.39 - 0.02 100 101 102 103 104 100 101 102 103 104 cells (10 ) Pax7 (FITC) Pax7 (FITC)

Figure III.15. Evaluation of Pax7 expression by flow cytometry in an uncultured SVF sample (SVF-4) and in the enriched and depleted fractions obtained after CD34 MACS. (A) Gated flow cytometry charts; (B) Flow cytometry histograms of Pax7 expression in the unsorted SVF (left) and the CD34-enriched fraction (right); (C) Frequency, cell number and expression for each gated population and number of Pax7-expressing cells in a total of 48.6x105 viable SVF cells, 5.7x105 CD34 positive cells and 3.3x105 CD34 negative cells.

III.2.2.2 Plating and isolation of CD34-enriched and CD34-depleted fractions

The cell fractions obtained previously were plated onto tissue culture flasks with the aim of assessing if distinct cell populations could be obtained in the two fractions and if the levels of CD34 and CD56 were maintained after plating. Additionally, cells from both fractions were plated in the presence of gelatin and fibronectin substrates (i.e. coating of the plastic culture surface), to promote the adhesion of myogenic-committed cells, and in the presence of high glucose DMEM (DMEM-HG) medium, to test if increased glucose concentrations promote cell differentiation.172 The cells were plated into the following conditions: a) non-coated dishes, b) gelatin-coated dishes, c) fibronectin-coated dishes and d) fibronectin-coated dishes in the presence of DMEM-HG medium. Unsorted SVF cells were plated onto non-coated dishes. An expansion step was performed in the presence of myogenic expansion medium (DMEM supplemented with standard FBS, not MSC-qualified, dexamethasone and bFGF)157 and the levels of CD34 and CD56 were determined after the first passage.

Surprisingly, no cells derived from the CD34-depleted fraction adhered to tissue culture plates, even in the presence of a gelatin and fibronectin coating. 48 hours after plating, only floating cells were observed and after 7 days in culture no cells were adherent, and so the cells were discarded. The assay was repeated to account for biological variability and the same outcome was obtained. Moreover, plated cells from the CD34-enriched fraction displayed similar sizes and morphologies when compared to unsorted SVF cells. Based on both these observations, it was hypothesized that

only the CD34+ cells within the unsorted SVF were able to adhere and proliferate in culture. This hypothesis was further tested in Section III.2.2.4.

The results presented below (including in Section III.2.2.3) correspond to cells initially belonging to a cell pool of SVF cells (Figure III.14). After 10 days in culture, unsorted SVF cells were completely confluent, while cells from the CD34-enriched fraction were not evenly spread across the plates and grew as dense colonies (Figure III.16). Since unsorted SVF cells were not submitted to the MACS pre-separation filter, which has 30 μm diameter pores, large cell aggregates could have been initially plated in this condition, which may have contributed to the achievement of high confluence levels earlier. In low confluence areas, pericyte-like cells could be observed (Figure III.16). The expression of CD34 and CD56 was determined after the first passage (Figure III.17). For all conditions, a main population and a subpopulation of cells (placed at lower SSH and FSH values) could be identified. The results suggest that the presence of coating had great influence on CD56 expression, since cells cultured on gelatin and fibronectin-coated plates presented CD56 expression levels around 80%, while SVF (unsorted cells) and non-coated cells presented negligible levels of CD56 expression. Conversely, CD34 expression was more variable between the tested conditions. Cells cultured on fibronectin-coated plates exposed to DMEM-HG medium presented the lowest levels of CD34 expression (0.8%), while non-coated cells presented the highest (21%). The high levels of CD56 expression (and contrastingly lower CD34 expression) might indicate that the cells adopted a phenotype similar to activated satellite cells, which tend to lose CD34 expression and are characterized phenotypically as CD34-CD56+. This experiment was repeated with cells from a different donor (SVF-2), but significantly lower levels of CD34 and CD56 expression were obtained. One of the factors that might have influenced these results is cell confluence and/or plating density, since different confluence levels were achieved in the two assays. Thus, the assay needs to be repeated to account for these factors.

Figure III.16. Cells cultured on fibronectin-coated (A) and non-coated (B) plates, using LG DMEM media, derived from the CD34-enriched fraction of the SVF, acquired by MACS. Low confluence areas (left) and high confluence areas (right) could be obtained in all conditions, except for unsorted SVF cells, which were 100% confluent. Black arrows point to pericyte-like cells.

1000 1000 1000 1000 1000 SVF No coating Gelatin Fibronectin Fibronectin (HG) 800 800 800 800 (LG) 800

r r r r

e e e e

t t

600 t t

t t t 600 t 600 600 600

a a a a

c c c c

S S S S

e e e e

e 72.8

d d d d

i i i i

i 400

S S S 400 S 400 400 400

S 68.2 14.4 11.3 13.6 8.49 11.3 67 70.2 200 200 200 200 200 65.9

0 0 0 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 Forward Scatter Forward Scatter Forward Scatter Forward Scatter Forward Scatter

CD34-enriched fraction-derived cells SVF No coating Gelatin Fibronectin LG Fibronectin HG

% MFI % MFI % MFI % MFI % MFI

CD34 (MP) 7.9 2.0 21 2.9 9.3 1.8 14 2.3 0.80 1.2 CD56 (MP) 0 - 0.20 1.7 90 7.1 84 20 80 5.1 Gated events 3317 7103 6085 2285 2835 CD34 (SP) 21 2.4 17 4.1 5.2 1.0 0 - 0 - CD56 (SP) 0.60 1.2 0 - 54 4.4 20 4.8 4.8 4.0 Gated events 414 1112 656 367 657 Total cells (106) 2.48 0.57 1.83 0.44 1.76

Figure III.17. Flow cytometry analysis of CD34 and CD56 expression for SVF and cells cultured on non-coated plates, after 11 days in culture, and cells cultured on gelatin and fibronectin-coated plates, after 12 days in culture. Top: Flow cytometry dot plot charts. Bottom: Expression percentages and MFI values for CD34 and CD56 for each condition. MP = Main population (higher SSH and FSH gate), SP = Subpopulation (lower SSH and FSH gate).

III.2.2.3 Myogenic Differentiation Assay

Myogenic induction was performed by replacing the plating medium by DMEM-HG supplemented with 2% HS, a medium commonly used to induce myogenic induction in MSCs. To determine at which stage of myogenesis the cells were, four stage-specific markers expressed in the myogenic differentiation process were used: Pax7 (expressed in satellite cells), MyoD (expressed in proliferating and differentiating myoblasts), myogenin (expressed only in differentiating myoblasts) and skeletal muscle MHC (expressed in differentiated myotubes). After 2, 4 and 8 days of medium replacement, the expression of these markers was evaluated through intracellular flow cytometry and immunostaining methodologies. The expression of myogenic markers before medium replacement was evaluated on cells cultivated on gelatin- and fibronectin-coated (HG) plates, the only two conditions for which enough cell numbers were obtained to perform intracellular staining for flow cytometry analysis (Figure III.18). Surprisingly, the results indicate that cells cultured on gelatin- and fibronectin-coated (HG) plates expressed MyoD and myogenin (over 97% and 84% of the gated main population, respectively) before the medium replacement. Moreover, the MFI values show that MyoD was being significantly expressed, as well as myogenin, although at lower levels.

1000 100 100 100 A 800 80 80 80

x x x

a a a

t 600 60 60 60

M M M

f f f

o o o

% % %

S 400 40 40 40

200 11.2 20 20 20 58.4 33.3 99.5 91.5

0 0 0 0 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 200 400 600 800 1000 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 Forward Scatter Pax7 (FITC) MyoD (FITC) Myogenin (FITC) Ungated fibroHG_beforeinduction.0011000 100 100 100 Event Count: 11625

B 800 80 80 80

x x x

a a a t 60 60 60 t 600

M M M

f f f

o o o

i % % 40 % 40 40

S 400

20 20 20 200 4.68 7.66 98.3 85.2 78.1

0 0 0 0 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 200 400 600 800 1000 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 Forward Scatter Pax7 (FITC) MyoD (FITC) Myogenin (FITC)

Main Population Subpopulation C Gelatin Fibronectin (HG) Gelatin Fibronectin (HG)

% MFI % MFI % MFI % MFI

Pax7 7.3 1.7 32 2.2 0.10 1.1 1.4 1.3 MyoD 98 89 99 92 76 14 92 20 Myogenin 84 6.8 90 9.0 0 - 0.30 2.0

Figure III.18. Expression of myogenic markers in P0 cells cultured on gelatin- and fibronectin-coated plates (HG) after 12 days in culture. Flow cytometry dot plot charts and histograms of Pax7, MyoD and myogenin (left to right) of the main population gated of cells cultured on fibronectin- (HG) (A) and gelatin-coated plates (B). (C) Myogenic marker expression (in percentage and MFI values) of the gated main populations and subpopulations of both conditions.

After induction, the morphology of some cells was changed. These cells became broad and flattened, and kept increasing in proportion with culture time. After 8 days of medium replacement, three types of cells could be visualized: (a) spindle-like cells, usually arranged in an aligned fashion, (b) spindle-like cells with several vacuoles and (c) broad, flattened cells with enlarged nuclei (Figure III.19A). The presence of enlarged cells, with significantly decreased nucleocytoplasmic ratios when compared to spindle-like cells, was confirmed by immunofluorescence (Figure III.19A and B). Importantly, MyoD expression was reduced in cells with enlarged nuclei and was more intense around confluent spindle-like cells, which suggests that flattened enlarged cells lost myogenic marker expression. The accumulation of vacuolar structures similar to lipid droplets with increased culture time might be related to spontaneous adipogenic differentiation, which has been correlated with plating ADSCs and satellite cells in DMEM-HG (i.e. 4.5 g/l glucose).173,174 The frequency of these structures depended on the plating conditions (approximately 40% for SVF cells, 25% for cells cultured on non-coated plates, 10-15% on fibronectin-coated (LG), 5% on fibronectin-coated (HG) and 5-10% on gelatin-coated plates).

A B

C D

Figure III.19. Cells displaying distinct morphologies after 8 days of induction, detected by light microscopy (top) and fluorescence microscopy (bottom). (A) Cells cultured on fibronectin-coated plates (LG), in which aligned spindle-like and broad flattened cells can be discerned; (B) SVF cells, in which the development of several vacuoles is visible; (C) cells cultured on gelatin-coated plates and (D) cells cultured on non-coated plates; Broad flattened cells with enlarged nuclei (frequently binucleated) could be seen in all conditions, as well as smaller spindle-like cells. Nuclei were stained with DAPI and the cytoskeleton (F-actin filaments) was stained with phalloidin-TRITC.

Similar dot plots were obtained for all conditions after medium replacement. However, the marker expression results indicate that changing the plating medium to DMEM+2%HS induced drastic alterations in myogenin and MyoD expression (Table III.7). At day 2, although MyoD expression was maintained, myogenin expression decreased drastically for cells on gelatin and fibronectin-coated (HG) plates when compared to the expression before induction (Figure III.18). These results were confirmed by immunofluorescence, in which MyoD expression was detected in all tested conditions, while myogenin and Pax7 were not. Surprisingly, MyoD was detected not only in the nucleus but also in the cytoplasm of the cells (Figure III.20). At day 4 after medium replacement, the results regarding marker expression were similar to day 2. However, the flow cytometry results obtained for day 8 show a drastic reduction in MyoD percentage levels. A reduction in MyoD-expressing cells was also observed in the immunofluorescence results (Figure III.21). It should be noted that due to a stock rupture, at day 4 the secondary antibody used for myogenin had to be switched (from Alexa 488 to PE), and at day 8 for MyoD (also from Alexa 488 to PE). Therefore, it is possible that this switch in the fluorophore used influenced the results. At day 8 skeletal MHC was introduced in the panel in order to identify terminally differentiated myoblasts. The results indicate that gelatin coating was the condition that promoted myogenesis the most, since almost 34% of cells expressed skeletal MHC. Surprisingly, SVF cells cultured on non-coated plates displayed the second highest level of MHC expression (6.5%), followed by

positively sorted cells cultured on uncoated plates (3.3%). In accordance, phalloidin staining allowed the visualization of aligned nuclei and some fused cells (Figure III.21). Interestingly, round multinucleated cells (P1) were identified by DAPI and phalloidin staining in a very low confluence area of a gelatin-coated well. No Pax7 and myogenin expression were observed by immunofluorescence throughout the assay.

Table III.7. Expression of myogenic markers (Pax7, MyoD, myogenin and skeletal MHC) in P1 cells after 2, 4 and 8 days of differentiation induction, assessed by flow cytometry. Secondary antibodies: Goat-anti mouse/rabbit Alexa 488 for myogenin (day 2) and MyoD (days 2 and 4), respectively; goat-anti mouse/rabbit PE for myogenin (days 4 and 8) and MyoD (day 8).

Day 2

SVF No coating Gelatin Fibronectin (LG) Fibronectin (HG)

% MFI % MFI % MFI % MFI % MFI

Pax7 3.6 1.3 0.70 1.3 1.8 1.4 0.20 1.2 0.7 1.4 MyoD 99 18 99 44 99 24 99 24 99 28 Myogenin 2.8 1.7 2.8 1.8 6.8 1.8 2.4 1.6 1.1 1.6 Day 4

SVF No coating Gelatin Fibronectin (LG) Fibronectin (HG)

% MFI % MFI % MFI % MFI % MFI

Pax7 0.50 1.3 0.10 1.1 0.70 1.2 0.70 1.3 2.4 1.4 MyoD 98 19 99 20 99 23 99 20 100 21 Myogenin 0.70 1.5 0.60 1.4 0.70 1.3 0.8 1.4 0 -

Day 8 (%) SVF No coating Gelatin Fibronectin (LG) Fibronectin (HG)

Pax7 0 0.10 0 0.20 0.10 MyoD 10 7.4 33 6.8 35 Myogenin 1.0 0.10 0 0.10 0.60 Skeletal MHC 6.5 3.3 34 1.2 2.4

Figure III.20. Multinucleated cells (P1) present in a very low confluence area of a gelatin-coated well, cultured for 17 days, detected by DAPI and Phalloidin-TRITC staining.

SVF No Coating Gelatin Fibronectin (LG) Fibronectin (HG)

D2 MyoD

D4 MyoD

D8 MyoD

Actin

D8 F

Figure III.21. MyoD expression assessed by immunofluorescence after 2, 4 and 8 days of medium replacement for all conditions (secondary antibody: goat anti-rabbit Alexa 546). From day 4 to day 8 a decrease in the fluorescence intensity and the number of fluorescent cells was seen. Phalloidin-TRITC and DAPI staining (bottom row) allowed the observation of aligned nuclei and some fused cells. No Pax7 and myogenin expression was observed through the assay for all the conditions. Scale bar = 50 μm.

III.2.2.4 CD34/CD56 decay

The results obtained for the experiments performed in the sections above suggest that both unsorted SVF cells and CD34-enriched cells obtained by MACS tend to lose CD34 expression once plated. It was hypothesized that this loss of expression was due to downregulation of CD34 expression rather than death of CD34+ cells. To confirm this hypothesis, cells from unsorted SVF and magnetically-sorted CD34 cells were plated at 10 000 cells/cm2 in expansion medium and in a commercially available formulation of skeletal muscle growth medium, to test if a skeletal-specific medium would help the maintenance of a CD34+CD56+ phenotype. CD34 and CD56 expression was then measured by flow cytometry 2 and 3 days after the cells were plated (Figure III.22). For this assay, the number of sorted cells in the CD34 positive fraction was not sufficient to test both media, therefore, the cells were only plated in expansion medium. At day 2, only 330 events in total were obtained for CD34-sorted cells. Two populations could be discerned (Subpopulation 1: 98% CD34+, 0.84% CD56+; Subpopulation 2: 54% CD34+, 4.8% CD56+), however; this data was not taken into consideration since it was not possible to perform a statistically relevant analysis.

At day 2 after plating, a population of cells placed at low FSH and SSH levels was obtained for unsorted SVF cells. This population was positive for CD34 in both tested medias (over 89% and 86% of cells were CD34+ when plated in skeletal and expansion medium, respectively). CD56 expression, on the other hand, was more dependent on the plating medium, as 36% of the cells plated in skeletal media were CD56+, while only 9.3% of cells plated in expansion medium were CD56+. At day 3 after plating, the formation of two distinct cell populations was evident in the flow cytometry dot plots for unsorted SVF cells (plated in both media) and sorted cells (plated in expansion medium). Both populations presented different phenotypes especially in terms of CD56 expression. The newly formed cell population, placed at higher FSH and SSH values (Population 1), was positive for CD34 and presented some CD56 expression, while Population 2 presented lower CD34 expression and was negative for CD56 (0% expression for all conditions). It can be hypothesized that between day 2 and 3 of culture the adhered cells started elongating, and thus a new population of cells with higher complexity and size developed. Surprisingly, this new population of cells (Population 1) presented increased CD34 expression when compared to cells plated for 2 days, for unsorted SVF cells. At day 3, slight differences between cells plated in the two different media could be detected, for cells belonging to this population. In comparison to cells plated in expansion medium, cells cultured using skeletal muscle medium had lower CD34 expression and higher CD56 expression. Cells derived from the CD34-enriched fraction had higher CD34 and CD56 expression percentages than unsorted SVF cells plated in both media. The role of CD34 in SVF cells and ADSCs is currently unknown.175 These results indicate that SVF cells that initially adhere to tissue culture plates are predominantly CD34+. However, AT-SVF cells seem to progressively lose CD34 expression when they start proliferating, or as a result of increasing confluence and cell-to-cell contact. This hypothesis would explain why significantly lower CD34 expression was detected for unsorted SVF cells after 11 days in culture (Figure III.17)

SVF – Day 2 SVF D2 SVF D2 Skeletal Muscle Expansion Medium Expansion Medium Skeletal Muscle Medium Medium % MFI % MFI CD34 89 45 86 47

CD56 35 8.6 9.3 8.3 Gated events 3196 4191

SVF – Day 3 CD34-enriched – Day 3

Skeletal Muscle Medium Expansion Medium Expansion Medium

Subpopulation 1 Subpopulation 2 Subpopulation 1 Subpopulation 2 Subpopulation 1 Subpopulation 2

% MFI % MFI % MFI % MFI % MFI % MFI

CD34 91 36 45 4.8 94 42 82 11 98 42 82 11 CD56 15 2.8 0 - 8.9 3.3 0 - 19 3.0 0 - Gated events 1911 1255 1205 1361 3093 4008

SVF D3 SVF D3 CD34-enriched D3 Skeletal Muscle Medium Expansion Medium Expansion Medium

1 2

2 2 1

Figure III.22. Flow cytometry analysis of CD34 and CD56 expression after 2 and 3 days in culture for cells isolated from unsorted SVF and magnetically sorted cells from the CD34-enriched fraction.

III.2.2.5 Culture of SVF-derived cells on low attachment plates

Satellite cells usually require a substrate to adhere to, given the fact that they adhere slower to culture dishes when compared to other cell types such as MSCs.110 Once satellite cells adhere, they quickly begin differentiating into myoblasts, instead of proliferating in an undifferentiated state. MSCs, on the other hand, start proliferating greatly once they adhere, making it difficult to isolate more quiescent cells that may exist in the SVF (e.g. satellite cells). Therefore, in an attempt to isolate myogenesis-committed cells from the SVF, cells derived from fresh SVF and cells from the CD34-enriched fraction sorted by MACS were plated onto low-attachment plates. The use of low-attachment plates for satellite cell expansion has been implemented for mice Pax7+ cells (after cell isolation through FACS) with success, as Pax7+ cells were able to proliferate and form aggregates in suspension.176 Two types of medium were tested: the expansion medium mentioned above and a commercially available skeletal muscle cell growth medium formulation. After 6 days of culture on low attachment plates, aggregates in suspension could be observed. The medium was then collected and cells in suspension were plated on gelatin-coated plates. Adhesion of the aggregates to the culture

plate was observed after 24 hours in culture. Aggregates were formed by typical AT-SVF cells that were able to proliferate in suspension, and no satellite/myoblast-like cells were observed (Figure III.23). The cells were harvested after 4 days and the expression of CD34 and CD56 was assessed by flow cytometry. No CD56 expression was obtained in any of the conditions, suggesting that myogenic-committed cells were not isolated.

Figure III.23. Aggregate adherence to gelatin-coated plates after 24 hours in culture. Left to right: unsorted SVF cells in expansion media; unsorted SVF cells in skeletal muscle cell growth media; CD34-enriched cell fraction in expansion medium; CD34-enriched cell fraction cell in skeletal muscle cell growth media. Scale bar = 100 μm.

III.2.2.6 Discussion

Most of the data in the literature indicate that MSC myogenic differentiation requires either DNA demethylation treatment, co-culture with myoblasts or in vivo muscle engraftment. As demonstrated in Section III.2.1, DNA demethylation treatment, under the conditions tested in this thesis, did not reveal to be a plausible option. Conversely, few reports of spontaneous myogenic differentiation of cells derived from the SVF, without any of the previously mentioned stimuli, have emerged.69,110–113 Some of these studies demonstrated that the SVF contains subpopulations of cells that exhibit myogenic phenotypes in vitro and that are capable of differentiating into myotubes.110,111 When reading the existing literature on this matter, one important question can be raised: does the myogenic potential of the AT relies, in fact, on the presence of a low percentage of myogenic-committed cells, as concluded by these authors110,111, or is the multipotent adherent fraction of the SVF (e.g. ADSCs) capable of myogenic potential per se in the presence of adequate stimuli? The results obtained herein give evidence that both hypotheses might be correct.

To the extent of the literature reviewed herein, this is the first time that Pax7-expressing cells are identified in uncultured fresh human SVF samples. Thus, it is reasonable to claim that myogenic-committed cells do exist in vivo in the adult AT, as Pax7 is the most reliable marker for identifying skeletal muscle progeny. Still, this assay should be repeated in order to account for biological variability and to assess the average percentage of Pax7+ cells in the SVF. Additionally, it would be interesting to determine if MyoD- and myogenin-expressing cells exist in fresh uncultured SVF, in order to conclude if myogenic cells at different stages of commitment can be obtained from the AT. Although the isolation of Pax7+ cells was not successful, even after the implementation of two different strategies, these results open a new window of clinical applications for the AT. For instances, if sufficient numbers of myogenic cells are accessible from the AT and an efficient isolation method is

developed, the SVF could become a reliable alternative source of skeletal muscle cells, thus eliminating the need of invasive muscle biopsies. Moreover, it has been hypothesized that the immunosuppressive properties of undifferentiated ADSCs increase the feasibility of transplantation procedures in damaged muscle, which represents an important advantage in a clinical therapy perspective.177

Although CD34-cell sorting through MACS enhanced the content of Pax7+ cells by 7-fold, other markers should be used for isolating these cells, as the results obtained herein indicate that over 80% of SVF cells that adhere to tissue culture plates in the first 48 hours of culture are CD34+. Hence, for future studies, CD56 should be a promising marker for sorting myogenic cells from the SVF, as AT-derived MSCs do not express this marker.178 FACS could have also been used as sorting methodology, instead of MACS. Indeed, FACS allows the achievement of highly pure fractions, and holds greater precision, resolution and sensitivity, which are important qualities when the isolation of rare subpopulations is aimed. Importantly, an anti-CD56 antibody attached to a bright fluorophore (e.g. PerCP) should be chosen, in order to allow the isolation of cells expressing different amounts of the antigen (e.g. CD56bright and CD56dim cells). The use of antibodies with inferior relative brightness, such as FITC-conjugated antibodies, may result in less efficient cell separations, especially if the desired cell population does not significantly overexpress the chosen antigen.179 According to the ISCT, uncultured SVF cells express CD34 at moderate levels (>20%), and lose CD34 expression over few cell passages108. Three uncultured SVF samples were assessed herein by flow cytometry and considerably high levels of initial CD34 expression were obtained (75, 77 and 98%). Moreover, in the first 48 hours of culture, CD34+ cells represented over 80% of total cells (depending on the culture medium used), and even higher values of expression were reached at 72 hours of culture. These results are in accordance with the observation that the CD34-depleted cell fraction obtained by MACS lacks proliferative potential. One similar study was performed by Maumus and colleagues, who sorted CD34+ cells from human AT by MACS and concluded that only CD34+ cells possessed colony-forming activity.180 These results indicate that the proliferative potential of the SVF relies on CD34+ cells, which deeply contrasts with UC-derived MSCs and BM-MSCs that do not express CD34. Additionally, the maintenance of CD34 and CD56 expression seemed to greatly depend on the culture medium and the presence of a substrate (i.e. coating). In Table III.2, it was observed that low serum medium stimulated CD34 expression in P0 AT-SVF cells. From the results obtained in Figure III.17, no clear correlation between cell sorting, presence/absence of substrate and CD34 expression could be made. However, CD56 expression was clearly augmented in the presence of substrates (gelatin or fibronectin).

Cells plated on a coated plastic surface adopted a phenotype similar to activated satellite cells (CD34-CD56+) which is consistent with the high MyoD and myogenin levels observed for cells cultured on gelatin- and fibronectin- (LG) coated plates before cell re-plating. These results suggest that the expansion medium in which the cells were isolated might have positively influenced the cells’ commitment to the myogenic lineage. Moreover, it can be proposed that AT-SVF cells possess some

myogenic potential per se and begin expressing myogenesis-related proteins in the presence of adequate biochemical stimuli (e.g. dexamethasone and bFGF). In fact, dexamethasone supplementation is known to promote myogenic differentiation in BM-MSC, leading to the formation of desmin-positive myotubes in a dose-dependent manner, and enhance the proliferation and differentiation of myoblasts.106,181,182 Concordantly, a study showed that significantly higher numbers of MSC nuclei were involved in myotube formation when bFGF and dexamethasone were added to co-cultures with myoblasts when compared to DMEM medium supplemented with HS alone, thus suggesting a possible role of dexamethasone and bFGF as pro-myogenic factors.104 However, it should be noted that dexamethasone is not a specific pro-myogenic factor, as it also promotes adipogenic, osteogenic and chondrogenic differentiation in MSCs.170 In future studies, the cells should be expanded in medium free of these factors, in order to allow the proliferation/maintenance of the cells in an undifferentiated state and avoid early myogenesis commitment.

The presence of gelatin coating was reported to promote myoblast differentiation presumably by collagen interactions mediated by a myoblast cell surface glycoprotein named gp46, which is reduced in differentiation-defective mutants of rat skeletal myoblasts.183 The expression of gp46 remains constant during myoblast replication and fusion but decreases markedly postfusion, which is in accordance with the pattern of collagen synthesis in L6 myoblasts (peaking just as cells align and dropping after fusion).183,184 Through the use of a compound that specifically inhibits collagen hydroxylation, it was concluded that collagen-deficient L6 myoblasts were unable to align and fuse if the compound was added before myoblast alignment.185 These evidence suggest that collagen is essential at early stages of myogenic commitment but dispensable for the subsequent steps of myogenic differentiation. Still, in a more recent study in which the role of fibronectin and gelatin coating on myogenesis was explored, it was concluded that fibronectin promoted myoblast migration and more organized collective alignment, and gave rise to a larger number of myotubes, with more nuclei per myotube, when compared to gelatin.186 Conversely, other authors observed no significant differences in myotube development in myoblasts cultured on gelatin and fibronectin coated plates in low serum conditions.187 Moreover, these authors concluded that exogenous matrix substrates are not essential for serum free cultures if matrix secretion is triggered and the culture surface supports the retention of secreted matrix proteins.187 In the conditions of the experiments presented herein, the presence of coating had impact on myogenic marker expression, as gelatin coating increased skeletal MHC expression by 10-fold after 8 days of induction (34% versus 3.3% for cells in the absence of coating). Ultimately, it is possible that the presence of different ECM molecules influences the progression of myogenesis by the exertion of distinct effects at different stages (initial alignment and commitment versus myoblast fusion and myotube formation). Thus, it would be interesting to attempt to mimic the skeletal muscle ECM composition pattern and its dynamic changes when muscle regeneration occurs, in order to optimize myogenesis in vitro.

Myogenin and MyoD expression were not maintained for long-term culture periods, and the expression of both markers was drastically reduced after day 2 (gelatin and fibronectin HG conditions) and 8 of media replacement, respectively. One particular study involving UC-derived MSCs showed that, in the presence of dexamethasone and hydrocortisone, MSCs expressed MyoD at day 3 of culture and myogenin at day 7, but both markers were not expressed after 2 weeks.102 It should be noted that a switch in the secondary antibody used for MyoD at day 8 might have influenced the results obtained, since MyoD expression was still observed by immunofluorescence at day 8, albeit at lower intensities when compared to day 4. In order to avoid possible interferences caused by the secondary antibodies, the use of labelled primary antibodies should be preferred in future studies. Still, the drastic decrease in MyoD expression, which was not followed by myogenin upregulation, might have had two possible causes. Firstly, MyoD decreases may be a consequence of the increasing proportion of broad flattened cells with culture time, as lower fluorescence intensity was observed in these cells when compared to high confluent areas filled with spindle-like cells. Secondly, the myogenic induction medium used herein might not have sustained and promoted the last steps of myogenic differentiation (cell fusion and myotube formation). Consequently, it is possible that important factors might have been lacking in the medium formulation used. Common myogenic differentiation medium formulations include DMEM supplemented with insulin and transferrin, or hydrocortisone, dexamethasone and FBS, apart from HS, which is usually used in ranges from 2 to 5%.31,105,112,157 Importantly, DMEM-HG medium has been reported to not only promote adipogenic differentiation in skeletal muscle and ADSCs but also inhibit myogenesis in C2C12 myoblasts, while DMEM-HG and high insulin activate myoblast differentiation.173,188

Unsorted SVF cells demonstrated the same MyoD expression levels as sorted and coated cells, and the same type of population was obtained for all conditions. Thus, CD34 sorting is probably a dispensable step, which is a positive aspect since cell sorting is an expensive operation that often leads to cell losses. In future studies, the myogenic marker expression of unsorted SVF cells cultured on coated plates should be also tested. One surprising result was the presence of two roundish multinucleated cells in a low confluence area of a gelatin-coated well. Since skeletal muscle are the only kind of multinucleated cells existing in the human organism, it can be postulated that myotube formation occurred in P0 (initial cell passage, upon initial cell seeding after isolation), and that these myotubes were unable to elongate once re-plated. This atypical morphology could have been caused by cytoskeletal arrestment caused by cell trypsinization and inadequate plating conditions (e.g. plating density and/or medium composition). In fact, myotubes are known to undergo anoikis, a response induced by either a loss of cell anchorage, anchorage to inappropriate ECM proteins or disrupted cell adhesion to the ECM, consequently exhibiting spherical morphologies and eventually detaching from culture plates.189 In particular, laminin–211 was reported to be important in preventing this phenomenon.189 The possibility of myotube formation accentuates the notion that myogenesis progressed dynamically and rapidly at early stages in culture, which contrasts with the current established myogenic differentiation protocols, that can last up to 6 weeks in order to significantly detect skeletal MHC expression.69,102,112

Under forced MyoD expression, MyoD-expressing ADSCs were shown to have the capacity to fuse with dystrophic myoblasts, restore dystrophin expression and to form fibers.190 In a very recent study, MyoD expression was induced in rat ADSCs through the transfection of small activating RNA targeting MyoD gene promoters, in a process known as RNA activation.191 Desmin expression could be detected after 14 days of transfection, and was even more evident after 21 and 28 days.191 Herein, natural MyoD expression was accomplished, which represents a major advantage when compared to genetically modified or transfected cells for future clinical applications. MyoD expression was detected not only in the nucleus but also in cytoplasmic areas. Transcription factors are synthetized in the cytosol, and many are known to interact with cytosolic proteins and translocate to the nucleus, where they exert biological effects. Therefore, it is plausible to propose that MyoD is either (i) being arrested in the cytosol by unknown interactions with other proteins, and that its translocation to the nucleus is only triggered once the cells receive enough stimuli to proceed to myogenic differentiation or (ii) cytoplasmic arrestment is being caused due to MyoD turn over (degradation) by the ubiquitin proteasome system.192

Lastly, to further enhance myotube formation, specific biomaterials or substrates with different stiffnesses could be tested. ADSCs were shown to undergo stiffness-induced lineage commitment and fuse (albeit at low percentages – 2%) when plated in a muscle-mimicking extracellular matrix (10 kPa) surpassing BM-MSCs, which never underwent stiffness-mediated fusion.68 The same group of researchers developed a mechanically-patterned hydrogel that led to increased fusion rates and fusing cell numbers in comparison with unpatterned matrices.193 Thus, the importance of physical cues in the form of ECM composition, stiffness and patterning, should not be underestimated.

III.3 Differentiation of Mesenchymal Stem/Stromal Cells into Smooth Muscle Cells

As already mentioned, human adult SMCs have limited proliferation capacity, unless the cells are immortalized, which makes the ex vivo expansion of these cells for tissue engineering applications very unviable.135 With the aim of achieving smooth muscle differentiation of MSCs from two different sources (SVF and BM), PD98059 (a compound already successfully used in one other study for this purpose)140 was tested. This compound belongs to the small molecules category, which are becoming increasingly popular in stem cell self-renewal and differentiation studies, mainly because of their ability to specifically act on proteins part of signalling pathways.194 PD98059 is a potent and selective inhibitor of MAPKK, a key protein in the ERK/MAPK pathway. This pathway is believed to play an important part in inhibiting the activity of SRF, a critical regulator of vascular SMC differentiation.140

A commercially available line of human bladder SMCs was used as positive control. These cells were initially spindle-like but became larger and more rectangular in shape as time in culture increased (Figure III.24). The expression of smooth muscle markers was tested in control bladder SMCs by flow cytometry and immunofluorescence (Figure III.25).

A B

Figure III.24. Bladder SMCs (passaged 6) plated in bladder smooth muscle cell growth media after 8 days (A) and 21 days (B) of culture.

1000 Population A B 800 1 (%) 2 (%)

r 1

t 47.4 α-SMA 32 13 t 600

Calponin 51 28

i 400

S SM-MHC 4.5 2.7 2 17.6 200

0 0 200 400 600 800 1000 Forward Scatter

α-SMA Calponin Desmin Merge

SM-MHC Calponin Desmin Merge

Figure III.25. Smooth muscle marker expression in cultured SMCs. A) Flow cytometry dot plot chart of bladder SMCs (P5) after 14 days in culture and B) percentage values of expression of smooth muscle markers. Secondary antibody: goat anti-mouse Alexa488. C) Immunofluorescence for markers α-SMA, SM-MHC, calponin and desmin detected in P6 bladder SMCs after 21 days in culture. Scale bar = 50 μm. Secondary antibodies: goat anti-mouse Alexa488 for α-SMA, calponin and SM-MHC; goat-anti rabbit Alexa 546 for desmin.

The flow cytometry results obtained for marker expression in control cells surprisingly show poor expression levels for SM-MHC (Figure III.25). Calponin and α-SMA expression was higher but did

not come close to 90-100% (51% and 32% for Population 1, respectively), which was the expected expression range. Thus, desmin was included in the marker panel in immunofluorescence testing. Although this intermediate filament is not smooth muscle specific, as it is expressed also in skeletal muscle and cardiomyocytes, it can be used as an indicator of myogenic lineage differentiation. Low fluorescence intensity could be detected for α-SMA and SM-MHC, while most cells clearly expressed calponin and desmin.

P1 AT-SVF cells and P3 BM-MSCs isolated and plated in DMEM+10%FBS (standard FBS and MSC-qualified FBS, respectively) were induced with DMEM+3%FBS supplemented with different concentrations of PD98059, after reaching confluence. The cells displayed spindle-like morphologies (Figure III.26) and were in culture for 6 days (BM-MSCs) and 9 days (AT-SVF cells) before induction.

A B

Figure III.26. BM-MSCs (A) and SVF-derived cells (B) plated in DMEM+10%FBS, cultured for 3 and 9 days, respectively.

During BM-MSC expansion, P2 cells were tested for monocyte contamination by flow cytometry with monocyte-specific marker CD14. The low percentage of expression obtained for this marker (0.18 %) indicated that a monocyte-free population was achieved.

Cell morphology and viability did not suffer any significant changes from PD98059 supplementation. The effects of PD98059 in the expression of smooth muscle markers α-SMA, calponin, SM-MHC and desmin were tested in the abovementioned cells after 2, 4 and 7 days of induction. Different conditions were tested in the two assays. For BM-MSCs, two controls (DMEM+10%FBS and DMEM+3%FBS) were implemented, and 1 μM, 10 μM and 30 μM of PD98059 were tested, based on the work of Tamama and co-workers.140 For AT-SVF cells, DMEM-HG medium supplemented with 10 μM of PD98059 was added as a test condition, and DMEM-HG supplemented with 3%FBS was the only control used. A medium change was performed at day 4. The flow cytometry results obtained for BM-MSCs indicate that the maximum expression values for α-SMA and calponin were reached at day 2 of induction, and the expression of these markers gradually dropped. However, the expression levels obtained for 30 μM at day 2 were close to the ones of the positive control (29 for α-SMA and 34 for calponin versus 32 and 51, respectively, in

the positive control). For SM-MHC, little expression was obtained for 10 μM, 30 μM and DMEM+10%FBS alone at day 2 (1.9, 1.3 and 1.0, respectively), which decreased to values close to 0 at days 4 and 7. Surprisingly, for AT-SVF cells no expression was obtained for α-SMA, calponin and SM-MHC. It should be noted that due to a stock rupture the secondary antibody used for the AT-SVF assay was different (in terms of the conjugated fluorophore) from the one used in the BM-MSCs assay. Desmin expression, on the other hand, achieved its maximum at day 4 for most conditions. Still, for the condition using 30 μM of PD98059, 22% desmin expression was reached at day 7. In general, PD98059 concentration was not directly correlated with increased marker expression, apart from BM-MSCs at day 2. Additionally, cells from the control conditions frequently presented expression levels higher than cells exposed to PD98059.

Regarding the immunofluorescence results obtained, calponin- and desmin-expressing cells could be observed in all conditions (Figure III.27). Importantly, fluorescent cells were only observed near the edges of the wells, which suggests that 3-D-like spatial stimuli might promote myogenic differentiation. Few cells were positive for both desmin and calponin. Calponin-expressing cells presented broad polygonal morphologies and increased sizes, as opposed to desmin-expressing cells, which tended to be smaller and more spindle-like. In BM-MSCs, very low intensity fluorescence could be observed for α-SMA and SM-MHC at day 2 in cells near the edges of the wells for all conditions. Conversely, no fluorescence was detected for α-SMA and SM-MHC in any condition for AT-SVF cells. In general, higher fluorescence intensities were observed for calponin in cells exposed to PD98059, however, the same correlation was not valid for desmin.

As already mentioned, smooth muscle cells are of great importance in tissue engineering applications in general, as these cells line the vessels of the vascular system and are major components of hollow visceral organs. Neovascularization is essential to supply oxygen and nutrients to the cells in 3-D constructs, making this one of the biggest challenges that the tissue engineering field needs to overcome.40 This challenge is hindered by the fact that smooth muscle cells do not possess enough proliferative capacity to allow for significant cell numbers to be obtained for Regenerative Medicine purposes.135 Moreover, SMC extraction from patients involves invasive tissue biopsies. Consequently, other cell types with increased availability and expansion capabilities in vitro, such as MSCs, have been strategically chosen as cell sources for differentiation assays directed towards the smooth muscle lineage for tissue engineering applications.141,195 MSCs are the ideal cell source given the phenotypic similarities between these two cell types.139 In fact, two decades ago, cultured BM-MSCs were shown to express SMC markers such as the ones used herein (α-SMA, calponin and SM-MHC) in long-term (7 week) confluent cell cultures.136 Since then, several studies have successfully implemented different strategies to induce MSCs to differentiate into SMCs.

BM-MSCs AT-SVF cells A Day 2 Day 2

Control Control Control HG 1 μM 10 μM 30 μM 1 μM 10 μM 10 μM (HG) 30 μM 3%FBS 10%FBS 3%FBS α-SMA 8.2 5.8 29 7.5 25 0 0 0.20 0.10 0.10 Calponin 18 29 34 21 28 0 0 0 0 0 SM-MHC 0 1.9 1.3 0.5 1.0 0 0.20 0 0 0 Desmin - - - - - 4.5 3.8 3.0 3.5 4.5 Day 4 Day 4

Control Control Control HG 1 μM 10 μM 30 μM 1 μM 10 μM 10 μM (HG) 30 μM 3%FBS 10%FBS 3%FBS α-SMA 4.2 1.9 1.2 7.2 11 0.70 0.80 0 0.10 0.20 Calponin 20 14 14 13 15 0.30 0.50 0 0 0.30 SM-MHC 0 0 0 0.20 0 0 0 0 0 0

Desmin - - - - - 6.3 11 5.6 3.5 8.5 Day 7 Day 7

Control Control Control HG 1 μM 10 μM 30 μM 1 μM 10 μM 10 μM (HG) 30 μM 3%FBS 10%FBS 3%FBS α-SMA 0.10 0.50 0.50 1.0 0.20 0.10 0.50 0.50 1.0 0.20

Calponin 0 0.30 1.0 0.20 0.30 0 0.30 1.0 0.20 0.30

SM-MHC 0 0 0 0.10 0 0 0 0 0.10 0 Desmin - - - - - 3.6 4.2 4.2 22 8.5

B AT-SVF cells Day 7 1 μM 10 μM 10 μM (HG) 30 μM Control (HG)

C BM-MSCs Day 7 1 μM 10 μM 30 μM Control (3%FBS) Control (10%FBS)

Figure III.27. Smooth muscle marker expression in AT-SVF cells and BM-MSCs after 2, 4 and 7 days of differentiation induction with PD98059, assessed by flow cytometry and immunofluorescence. A) Intracellular flow cytometry for α-SMA, calponin, SM-MHC and desmin. Secondary antibodies used: goat anti-mouse Alexa488 (in BM-MSCs assay); goat anti-mouse PE for α-SMA, calponin and SM-MHC and goat anti-rabbit Alexa 488 for desmin (in AT-SVF assay). B) Immunofluorescence after 7 days of induction for calponin (top) and desmin (bottom) in AT-SVF cells. C) Immunofluorescence after 7 days of induction for calponin in BM-MSCs. Secondary antibodies: goat anti-mouse Alexa4 88 for calponin and goat-anti rabbit Alexa 546 for desmin. Scale bar = 50 μm.

The SMC line used herein as positive control showed surprisingly low levels of SM-MHC expression and less than 50% expression for α-SMA (assessed by flow cytometry). Still, in the company’s website it is claimed that each lot is tested by flow cytometry for the smooth muscle markers calponin and α-SMA.196 Hence, it is possible that the lower expression levels obtained here are due to the primary antibodies used. Importantly, since SMCs show phenotypic plasticity, it is also possible that the cells were not cultured for sufficient time periods to allow for a complete phenotype switch into the contractile-phenotype (characterized by higher expression of contractile proteins such as α-SMA, calponin and SM-MHC).197 Therefore, it can be hypothesized that most cultured SMCs still retained a synthetic-phenotype (e.g. proliferative phenotype) at the time of marker expression assessment, which ultimately resulted in the exhibition of low expression levels for contractile proteins.

Serum alone (3% and 10% FBS) seemed to induce the expression of SMC markers in BM-MSCs and AT-SVF cells, possibly because of proteins and factors existing in the serum that might stimulate smooth muscle-lineage commitment. α-SMA is described as one of the earliest SMC markers expressed in undifferentiated BM-MSC culture, under normal and reduced serum conditions, as well as in AT-SVF cells.136,139,198 Besides α-SMA, all isoforms of calponin and desmin were shown to be expressed in undifferentiated BM-MSCs.199 Herein, this phenomenon was more evident for calponin and desmin, for which positive cells were detected by immunofluorescence in all tested conditions, including the control cells. Therefore, the only marker that can truly be considered as “smooth muscle-specific” is SM-MHC, which is not spontaneously expressed by MSCs. Interestingly, calponin-expressing cells tended to have broad polygonal shapes, resembling SMCs in confluent culture.

In the conditions of the present study, PD98059 did not provide enough stimuli to significantly and consistently enhance the expression of smooth muscle markers. No expression was detected by flow cytometry for AT-SVF cells, for all markers except desmin. Since calponin-expressing cells were observed at all timepoints in the immunofluorescence assay, it is possible that the secondary antibody used in the flow cytometry analysis influenced the results. Again, fluorophore-conjugated primary antibodies should be used in order to avoid possible fluorophore-derived variability in the results. Several PD98059 pulses might be necessary to fully induce smooth muscle differentiation (e.g. medium replacement every day for 5 days). Tamama and co-workers achieved the upregulation of smooth muscle markers SM-MHC, caldesmon and α-SMA in BM-MSCs with PD98059. However, the authors of this study did not mention how many pulses/medium changes were performed over the induction time, which lasted 5 to 7 days according to the authors.140 No other studies regarding the effect of PD98059 in MSC differentiation were found in the literature, except for one report in which it was concluded that this molecule blocks osteogenic differentiation and promotes adipogenic differentiation in a dose-dependent manner in MSCs.200 Conversely, several studies have demonstrated that TGF-β is successful in inducing MSCs to the SMC lineage, and a correlation between TGF-β production and embryonic SMC differentiation has been established.46,138,139,142,143,195,201 Therefore, TGF-β might be a promising candidate to test in future

studies since several data are already available in the literature, which potentially facilitates the establishment of a reproducible and efficient differentiation protocol. Other strategies may include the use of commercially available smooth muscle differentiation medium, which was reported to induce the expression of calponin, caldesmon, SM-MHC, SM22 and smoothelin.198

The results obtained herein for AT-SVF cells indicate that DMEM-HG did not enhance smooth muscle differentiation. Interestingly, in a study where several medium formulations were tested for AT-SVF expansion, it was concluded that cells isolated and expanded in DMEM-HG supplemented with standard FBS more closely resembled SMCs than MSCs.159 Conversely, cells isolated in smooth muscle cell medium (basal medium) presented a significantly higher frequency of adipogenic progenitors.159 Taken together, these data reinforce the importance of medium formulation selection when isolating and expanding AT-SVF, especially if the enrichment of non-stromal cell populations is aimed. Thus, in future studies, AT-SVF isolation with DMEM-HG medium can be tested in order to promote SMC enrichment.

Importantly, marker expression by immunofluorescence was only observed in cells near the edges of the wells, which indicates that physical cues namely 3-D orientation should be vital for directing the differentiation of MSCs into myogenic lineages. In a study where the effects of biomechanical straining and surface patterns of substrates were tested on MSCs (without using any biochemical reagents for cell proliferation and differentiation), it was concluded that higher level straining and a grooved surface (rather than a flat surface) positively influenced the differentiation of MSCs into smooth muscle-like cells.151 These results are in accordance with the abovementioned observation. In one other study, the combination of mechanical (uniaxial strain) and biochemical (TGF-β addition) stimuli was tested, and the results indicated that higher expression of a smooth muscle phenotype at a transcriptional level was obtained in the presence of both types of stimuli.202 Lastly, in one study where three mechanical forces relevant to the cardiovascular system (cyclic stretch, cyclic pressure, and laminar shear stress) were applied independently to BM-MSCs, cells exposed to cyclic stretch expressed SMC proteins with increasing levels of differentiation with increasing magnitudes of stretch.203 These studies show that surface patterning and mechanical stimulation also play an important part in inducing myogenic differentiation and that biochemical cues namely medium composition should not be the only factor taken into consideration when developing a differentiation protocol. Additionally, the effect of ECM molecules such as gelatin, fibronectin or collagen coatings should be also tested, as the presence of these ECM molecules was described to further contribute to the stimulation of SMC differentiation in BM-MSCs.141 Finally, as the primary function of SMCs is to contract and relax, cell contraction assays (e.g. measure of carbachol response in collagen gels) should be performed after assessing smooth muscle marker expression (in case of positive results).

III.4 Decellularization of Porcine Urethras

The development of strategies for urethral sphincter engineering requires the selection of appropriate scaffolds for cell seeding, to be used once proper in vitro myogenic differentiation protocols are established. In Section I.1.2.3 several potential scaffold materials were mentioned; however, decellularized tissue matrices present advantages such as biological inherent bioactivity and mechanical similarity to native ECM. Thus, protocols targeting efficient urethral tissue decellularization for the generation of acellular urethral bioscaffold should be developed. To this end, four urethras removed from euthanized female pigs were decellularized through a mechanochemical process based on a protocol developed by Simões and colleagues (Simões et al., under revision 2015). A dynamic system was implemented in which cell removal was promoted by the constant flow of a mild detergent solution inside of the urethral tissue through perfusion for 4 days. Simultaneously, agitation (340 rpm) on the outer surface of the tissue was induced. Distilled water was perfused for 24 hours in the beginning and at the end of the process, which lasted for 6 days (Figure III.28). The protocol adopted herein allowed efficient cell removal, which could be visualized by the gradual whitening of the tissue. Cell removal was not completely successful in urethra #3, in which some brown coloured tissue could be observed at the distal end (external extremity) of the urethra. This was most likely caused by the urethra size, since the urethra was too long when compared to the catheter to allow for successful perfusion in the mentioned extremity. The protocol duration was exceptionally increased to 10 consecutive days in order to allow maximum cell removal without compromising the tissue and ECM structure. In general, this protocol proved to be robust, reproducible and inexpensive, since only one reagent (SDS) is required, and peristaltic pumps are common equipment in research facilities that work with bioreactors.

Decellularization is an attractive technique for urethral scaffold generation not only because of the possibility of retaining the architecture of the native tissue, including the vasculature, but also due to the presence of biofactors and appropriate mechanical structure in the ECM that are required for cell adhesion and proliferation.12,30 However, the field of urinary tract organ tissue engineering has been poorly studied, in particular in areas such as the development of urethral decellularization protocols for scaffold production, which are currently lacking in the literature. Still, bladder decellularization strategies have been established for tissue engineering purposes204–206. These can eventually be useful for clinical trials or studies aiming the treatment of bladder-related incontinence, such as overactive bladder. SUI, in particular, requires the development of strategies aiming urethra - and not bladder - decellularization, such as the protocol tested herein.

Severe incontinence cases, caused by damages in the urethral tissue, would greatly benefit from studies performed in adequate in vitro urethral models, in particular for the study of the rhabdosphincter, which is essential for the continence mechanism. In this sense, this protocol has

several applications, the most straightforward being the development of in vitro models of SUI for studies involving drug testing, cell seeding strategies, cell-based therapies and urethral engineering.

A B Length Protocol Urethra # (cm) duration (days) 1 2 6 2 5 6 3 8.5 10

4 9 6

Figure III.28. (A) Schematics of the decellularization apparatus based on mechanochemical action. Briefly, the detergent solution (or distilled water) is placed inside the (a) vessel where the (b) cannulated urethra is submerged. The solution is pumped by a (c) peristaltic pump and re-circularized for 24 hours at a rate of 40 ml/min. Simultaneously, through a (d and e) magnetic stirrer plate the solution was agitated at 340 rpm. After 48 hours, the direction of perfusion was changed (from the black to the grey arrows). Adapted from Simões et al., 2015 (under revision). (B) Length and protocol duration for each decellularized urethra. (C) Main steps of the decellularization process and their outcomes for urethra #4. Biological tissue preparation includes AT and distal extremity removal, catheter placement and immobilization with sutures (all performed in day 0). Gradual whitening of the tissue over perfusion time occurred, indicating efficient cell removal.

IV. Future Trends and Conclusions

Skeletal muscle engineering has suffered from several setbacks that have precluded clinical translation of research in this field. These setbacks arise from the inability to expand and maintain skeletal muscle progenitors (satellite cells and myoblasts) in an undifferentiated state in vitro. This MSc project targeted the establishment of effective myogenic differentiation protocols for MSCs for urethral sphincter reconstruction, with particular focus on exploring the myogenic potential of the SVF from AT. AT-derived cells are the most promising autologous cell source candidate for SUI cell therapy or tissue engineering approaches owing to the following reasons: (i) ability to proliferate in vitro, allowing for clinically relevant cell numbers to be obtained; (ii) clinical safety, with no reports of tumour formation in the literature; (iii) Abundance and ease of harvest, without the need of resorting to urinary tract surgery/biopsies; (iv) Intrinsic potential for committing to myogenic lineages, as proven herein and by few other studies.69,110–113

When reviewing the current literature on MSC myogenic differentiation induction, the most frequently adopted approach is, by far, 5-AZA supplementation to culture medium. However, as shown here, the cytotoxic effects caused by this DNA demethylating agent should exclude its application in Regenerative Medicine approaches. Moreover, contrasting data on its effect on cell differentiation have been published, and no myogenesis induction was detected with the means used herein. Thus, research on this field should progress towards selecting and expanding cells with intrinsic myogenic potential. The results obtained here contribute to the development of this field of research, as they indicate that (i) myogenic-committed cells are present in uncultured SVF and (ii) even if this subpopulation is not efficiently selected, plated AT-SVF cells show intrinsic myogenic potential in the presence of adequate stimuli (e.g. dexamethasone and bFGF), that can be enhanced in the presence of gelatin coating, as indicated by the higher skeletal MHC expression levels obtained. These results should be repeated, to account for biological variability, and the following alterations should be implemented: (i) marker expression should be assessed by mRNA transcript quantification by RT-PCR, in order to confirm the results by an antibody-independent method; (ii) primary antibodies conjugated to fluorophores should be preferred; (iii) skeletal MHC expression should be included in the marker panel and be analysed earlier in culture; (iv) the formulation of the myogenic induction medium used should be optimized (e.g. DMEM-LG supplemented with HS, dexamethasone and hydrocortisone or DMEM-HG supplemented with HS, insulin and transferrin); (v) marker expression should be evaluated at different cell confluence levels to establish a possible correlation between cell confluence and myogenesis induction. Furthermore, to attempt the isolation of Pax7+ cells from the SVF, FACS for CD56 should be employed, as already discussed. Finally, the use of substrates with stiffness values similar to skeletal muscle (10 kPa) might also be an interesting approach to further stimulate myogenesis without resorting to chemical supplementation or serum.

Regarding the results obtained for the smooth muscle differentiation assays, these indicate that chemical induction with PD98059 was not sufficient to promote a consistent increase in smooth muscle marker expression in AT-SVF cells and BM-MSCs. In fact, the observations made in these assays suggest that 3-D-like spatial organization might have promoted desmin and calponin expression even in non-chemically stimulated cells. Therefore, 3-D stimuli in the form of patterned substrates and ECM mimicking through the presence of coating should be explored, as well as the influence of biomechanical cues (e.g. mechanical forces involved in the functioning of the internal smooth muscle layer of the urethral sphincter). Although little is still known about smooth muscle progenitors/precursors and the mechanisms that lead to the differentiation of bladder SMCs, TGF-β is assumed to be important in the embryonic development of bladder SMCs.42 On the other hand, consistent data on the literature point to this molecule as an effective smooth muscle differentiation inducer in MSCs.45,131–133,144,188 Hence, TGF-β might be a more promising candidate to study in further studies.

Finally, by combining this knowledge with cell expansion, urethra decellularization and cell seeding techniques, an in vitro model to study SUI can potentially be developed, which would serve as an important connecting tool between in vitro and clinical research.

V. References

1. Norton, P. & Brubaker, L. Urinary incontinence in women. Lancet 367, 57–67 (2006). 2. Lin, C. & Lue, T. Stem cell therapy for stress urinary incontinence: a critical review. Stem Cells Dev. 21, 834–843 (2012). 3. Levy, R. & Muller, N. Urinary incontinence: economic burden and new choices in pharmaceutical treatment. Adv. Ther. 23, 556–573 (2006). 4. Cerruto, M., D’Elia, C., Aloisi, A., Fabrello, M. & Artibani, W. Prevalence, Incidence and Obstetric Factors’ Impact on Female Urinary Incontinence in Europe: A Systematic. Urol. Int. 90, 1–9 (2012). 5. Abrams, P. et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol. Urodyn. 21, 167–178 (2002). 6. Thaker, H. & Sharma, A. K. Regenerative medicine based applications to combat stress urinary incontinence. World J. Stem Cells 5, 112–123 (2013). 7. Stothers, L. & Friedman, B. Risk factors for the development of stress urinary incontinence in women. Curr. Urol. Rep. 12, 363–369 (2011). 8. Iwanowicz-Palus, G., Stadnicka, G. & Włoszczak-Szubzda, A. Medical and psychosocial factors conditioning development of stress urinary incontinence (SUI). Ann. Agric. Environ. Med. 20, 135–139 (2012). 9. Fry, C. H., Meng, E. & Young, J. S. The physiological function of lower urinary tract smooth muscle. Auton. Neurosci. 154, 3–13 (2010). 10. Thor, K. B. & Donatucci, C. Central nervous system control of the lower urinary tract: new pharmacological approaches to stress urinary incontinence in women. J. Urol. 172, 27–33 (2004). 11. DeLancey, J. Why do women have stress urinary incontinence? Neurourol. Urodyn. 29, S13–S17 (2010). 12. Orabi, H. et al. Tissue engineering of urinary bladder and urethra: advances from bench to patients. Sci. World J. 2013, (2013). 13. Brading, a F. Spontaneous activity of lower urinary tract smooth muscles: correlation between ion channels and tissue function. J. Physiol. 570, 13–22 (2006). 14. Mitterberger, M. et al. Adult stem cell therapy of female stress urinary incontinence. Eur. Urol. 53, 169–75 (2008). 15. Heesakkers, J. P. F. a & Gerretsen, R. R. R. Urinary incontinence: sphincter functioning from a urological perspective. Digestion 69, 93–101 (2004). 16. Bae, J. H. & Yoo, J. J. Cell-based therapy for urinary incontinence. Korean J. Urol. 51, 1–7 (2010). 17. Shokeir, A. a, Harraz, A. M. & El-Din, A. B. S. Tissue engineering and stem cells: basic principles and applications in urology. Int. J. Urol. 17, 964–973 (2010). 18. Smythe, G. M. & Grounds, M. D. Exposure to tissue culture conditions can adversely affect myoblast behavior in vivo in whole muscle grafts: implications for myoblast transfer therapy. Cell Transpl. 9, 379– 393 (2000). 19. Kinebuchi, Y. et al. Autologous bone-marrow-derived mesenchymal stem cell transplantation into injured rat urethral sphincter. Int. J. Urol. 17, 359–368 (2010). 20. Gregorio, S., García-Arranz, M., Olmo, D. & Pedrero, F. Phase II Clinical Trial to Study the Feasibility and Safety of the Expanded Autologous Mesenchymal Stem Cells Use Derived From Adipose Tissue (ASC) for the Local Feminine Stress Urinary Incontinence. (2012). 21. Hussain, M., Greenwell, T. J., Venn, S. N. & Mundy, A. R. The current role of the artificial urinary sphincter for the treatment of urinary incontinence. J. Urol. 174, 418–424 (2005). 22. Kaiser, L. R. The future of multihospital systems. Top. Health Care Financ. 18, 32–45 (1992). 23. Schmitt, A., van Griensven, M., Imhoff, A. B. & Buchmann, S. Application of stem cells in orthopedics. Stem Cells Int. 2012, (2012). 24. Fortier, L. a. Stem Cells: Classifications, Controversies, and Clinical Applications. Vet. Surg. 34, 415–423 (2005). 25. Hui, H. et al. in Stem Cells in Clinic and Research (ed. Gholamrezanezhad, A.) 3–15 (InTech, 2011), Available at http://www.intechopen.com/books/stem-cells-in-clinic-and-research/stem-cells- general- features-and-characteristics (Accessed on January 2015) 26. Takahashi, K. & Yamanaka, S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676 (2006).

27. Sanchez-Ramos, J. R. Neural cells derived from adult bone marrow and umbilical cord blood. J. Neurosci. Res. 69, 880–893 (2002). 28. Pluripotent Stem Cells. Division of Biology and Medicine, Brown University, Available at http://biomed.brown.edu/Courses/BI108/BI108_2005_Groups/07/PluripotentStemCells.jpg (Accessed on November 2015) 29. Ren, G., Chen, X., Dong, F. & Li, W. Concise Review : Mesenchymal Stem Cells and Translational Medicine : Emerging Issues. Stem Cells Transl. Med. 1, 51–58 (2012). 30. Hwang, N. S., Varghese, S. & Elisseeff, J. Controlled differentiation of stem cells. Adv. Drug Deliv. Rev. 60, 199–214 (2008). 31. Bharadwaj, S. et al. Multipotential differentiation of human urine-derived stem cells: potential for therapeutic applications in urology. Stem Cells 31, 1840–56 (2013). 32. Chao, Y., et al. Umbilical cord-derived mesenchymal stem cells for hematopoietic stem cell transplantation. J. Biomed. Biotechnol. 2012, (2012). 33. Prochymal. Available at http://www.osiris.com/therapeutics.php# (Accessed on November 2015) 34. Bakker, A. & Langer, C. Cell-based therapies - an innovative therapeutic option in ophthalmology : Treating corneal diseases with stem cells. Fed. Heal. J. – Heal. Res. – Heal. Prot. 58, 1259–1264 (2015). 35. Carpenter, M. K., Frey-Vasconcells, J. & Rao, M. S. Developing safe therapies from human pluripotent stem cells. Nat. Biotechnol. 27, 606–613 (2009). 36. Kirouac, D. C. & Zandstra, P. W. The systematic production of cells for cell therapies. Cell Stem Cell 3, 369–81 (2008). 37. Bravery, C. a. et al. Potency assay development for cellular therapy products: An ISCT* review of the requirements and experiences in the industry. Cytotherapy 15, 9–19 (2013). 38. Hansmann, J., Groeber, F., Kahlig, A., Kleinhans, C. & Walles, H. Bioreactors in tissue engineering - principles, applications and commercial constraints. Biotechnol. J. 8, 298–307 (2013). 39. Macchiarini, P. et al. Clinical transplantation of a tissue-engineered airway. Lancet 372, 2023–2030 (2008). 40. Ikada, Y. Challenges in tissue engineering. J. R. Soc. Interface 3, 589–601 (2006). 41. Haisch, A., Kläring, S., Gröger, A., Gebert, C. & Sittinger, M. A tissue-engineering model for the manufacture of auricular-shaped cartilage implants. Eur. Arch. Otorhinolaryngol. 3, 316–321 (2002). 42. Tasian, G., Cunha, G. & Baskin, L. Smooth muscle differentiation and patterning in the urinary bladder. Differentiation. 80, 106–117 (2010). 43. Oostrom, O., Fledderus, J., de Kleijn, D., Pasterkamp, G. & Verhaar, M. Smooth Muscle Progenitor Cells: Friend or Foe in Vascular Disease? Curr. Stem Cell Res. Ther. 4, 131–140 (2009). 44. Groot, A. G. Smooth muscle cell origin and its relation to heterogeneity in development and disease. Arterioscler. Thromb. Vasc. Biol. 19, 1589–1594 (1999). 45. Rzucidlo, E. M., Martin, K. A. & Powell, R. J. Regulation of vascular smooth muscle cell differentiation. J Vasc Surg. 45, A25–A32 (2007). 46. Ross, J., Hong, Z., Willenbring, B. & Zeng, L. Cytokine-induced differentiation of multipotent adult progenitor cells into functional smooth muscle cells. J Clin Invest. 116, 3139–3149 (2006). 47. Bentzinger, C. F., Wang, Y. X. & Rudnicki, M. a. Building muscle: molecular regulation of myogenesis. Cold Spring Harb. Perspect. Biol. 4, 1–16 (2012). 48. Tajbakhsh, S. Skeletal muscle stem cells in developmental versus regenerative myogenesis. J. Intern. Med. 266, 372–389 (2009). 49. Tajbakhsh, S., Borello, U. & Vivarelli, E. Differential activation of Myf5 and MyoD by different Wnts in explants of mouse paraxial mesoderm and the later activation of myogenesis in the absence of Myf5. Development. 125, 4155–4162 (1998). 50. Caplan, A. Mesenchymal stem cells. J. Orthop. Res. 9, 641–50 (1991). 51. Bentzinger, C. F., Wang, Y. X., von Maltzahn, J. & Rudnicki, M. a. The emerging biology of muscle stem cells: implications for cell-based therapies. Bioessays 35, 231–241 (2013). 52. Péault, B. et al. Stem and progenitor cells in skeletal muscle development, maintenance, and therapy. Mol. Ther. 15, 867–877 (2007). 53. Asakura, A., Komaki, M. & Rudnicki, M. Muscle satellite cells are multipotential stem cells that exhibit myogenic, osteogenic, and adipogenic differentiation. Differentiation 68, 245–253 (2001). 54. Boonen, K. J. M. & Post, M. J. The muscle stem cell niche: regulation of satellite cells during regeneration. Tissue Eng. Part B. Rev. 14, 419–431 (2008).

55. Beauchamp, J. R. et al. Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J. Cell Biol. 151, 1221–1233 (2000). 56. Capkovic, K. L., Stevenson, S., Johnson, M. C., Thelen, J. J. & Cornelison, D. D. W. Neural cell adhesion molecule (NCAM) marks adult myogenic cells committed to differentiation. Exp. Cell Res. 314, 1553– 1565 (2008). 57. Lazarus, H. M. et al. Cotransplantation of HLA-identical sibling culture-expanded mesenchymal stem cells and hematopoietic stem cells in hematologic malignancy patients. Biol. Blood Marrow Transplant. 11, 389–398 (2005). 58. Farini, A., Razini, P., Erratico, S., Torrente, Y. & Meregalli, M. Cell based therapy for Duchenne muscular dystrophy. J. Cell. Physiol. 221, 526–534 (2009). 59. Deschênes, I., Chahine, M., Tremblay, J., Paulin, D. & Puymirat, J. Increase in the proliferative capacity of human myoblasts by using the T antigen under the vimentin promoter control. Muscle and Nerve 20, 437–445 (1997). 60. Bouchentouf, M., Benabdallah, B. F., Rousseau, J., Schwartz, L. M. & Tremblay, J. P. Induction of Anoikis following myoblast transplantation into SCID mouse muscles requires the Bit1 and FADD pathways. Am. J. Transplant. 7, 1491–1505 (2007). 61. Caplan, A. Mesenchymal stem cells. J. Orthop. Res. 9, 641–650 (1991). 62. Gimble, J. M. et al. Adipose-derived stromal/stem cells: a primer. Organogenesis 9, 3–10 (2013). 63. Friedenstein, A., Chailakhjan, R. & Lalykina, K. The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells. Cell Tissue Kinet. 3, 393–403 (1970). 64. Augello, A., Kurth, T. & Bari, C. De. Mesenchymal stem cells: a perspective from in vitro cultures to in vivo migration and niches. Eur Cell Mater 20, 121–133 (2010). 65. Dominici, M. et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8, 315–317 (2006). 66. Bianco, P., Cao, X., Frenette, P. & Mao, J. The meaning, the sense and the significance: Translating the science of mesenchymal stem cells into medicine. Nat. Med. 19, 35–42 (2013). 67. Wakitani, S., Saito, T. & Caplan, A. I. Myogenic cells derived from rat bone marrow mesenchymal stem cells exposed to 5-azacytidine. Muscle Nerve 18, 1417–1426 (1995). 68. Choi, Y., Vincent, L. G., Lee, A. R., Dobke, M. K. & Engler, A. J. Mechanical derivation of functional myotubes from adipose-derived stem cells. Biomaterials 33, 2482–2491 (2012). 69. Zuk, P., Zhu, M. & Mizuno, H. Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng. 7, 211–228 (2001). 70. Kolf, C. M., Cho, E. & Tuan, R. S. Mesenchymal stromal cells. Biology of adult mesenchymal stem cells: regulation of niche, self-renewal and differentiation. Arthritis Res. Ther. 9, 204–214 (2007). 71. Kuhn, N. Z. & Tuan, R. S. Regulation of stemness and stem cell niche of mesenchymal stem cells: implications in tumorigenesis and metastasis. J. Cell. Physiol. 222, 268–677 (2010). 72. Deans, R. J. & Moseley, A. B. Mesenchymal stem cells : Biology and potential clinical uses. Exp. Hematol. 28, 875–884 (2000). 73. Chamberlain, G., Fox, J., Ashton, B. & Middleton, J. Concise review: mesenchymal stem cells: their phenotype, differentiation capacity, immunological features, and potential for homing. Stem Cells 25, 2739–49 (2007). 74. Ryan, J., Barry, F., Murphy, J. & Mahon, B. Mesenchymal stem cells avoid allogeneic rejection. J. Inflamm. 2, (2005). 75. Pittenger, M. F. Multilineage Potential of Adult Human Mesenchymal Stem Cells. Science (80-. ). 284, 143–147 (1999). 76. Sekiya, I., Larson, B. L., Smith, J. R. & Pochampally, R. Expansion of Human Adult Stem Cells from Bone Marrow Stroma: Conditions that Maximize the Yields of Early Progenitors and Evaluate Their Quality. Stem Cells 20, 530–541 (2002). 77. Yang, B., Zheng, J. & Zhang, Y. Myogenic differentiation of mesenchymal stem cells for muscle regeneration in urinary tract. Chin. Med. J. (Engl). 126, 2952–2959 (2013). 78. Lu, L., Liu, Y., Yang, S. & Zhao, Q. Isolation and characterization of human umbilical cord mesenchymal stem cells with hematopoiesis-supportive function and other potentials. Haematol. Hematol. J. 91, 1017– 1026 (2006). 79. Caplan, A. I. & Dennis, J. E. Mesenchymal stem cells as trophic mediators. J. Cell. Biochem. 98, 1076–84 (2006). 80. Caplan, A. & Correa, D. The MSC : An Injury Drugstore. Cell Stem Cell 9, 11–15 (2011).

81. Bronckaers, A. et al. Mesenchymal stem / stromal cells as a pharmacological and therapeutic approach to accelerate angiogenesis. Pharmacol. Ther. 143, 181–196 (2014). 82. Blanc, K. Le, Frassoni, F., Ball, L. & Locatelli, F. Mesenchymal stem cells for treatment of steroid- resistant, severe, acute graft-versus-host disease: a phase II study. Lancet 371, 1579–86 (2008). 83. Seo, S. & Na, K. Mesenchymal stem cell-based tissue engineering for chondrogenesis. J. Biomed. Biotechnol. 2011, (2011). 84. Ankrum, J. & Karp, J. M. Mesenchymal stem cell therapy: Two steps forward, one step back. Trends Mol. Med. 16, 203–209 (2010). 85. Kuroda, Y., Kitada, M., Wakao, S. & Nishikawa, K. Unique multipotent cells in adult human mesenchymal cell populations. 107, 8639–8643 (2010). 86. Heneidi, S. et al. Awakened by Cellular Stress : Isolation and Characterization of a Novel Population of Pluripotent Stem Cells Derived from Human Adipose Tissue. PLoS One 8, (2013). 87. Wakao, S. et al. Multilineage-differentiating stress-enduring (Muse) cells are a primary source of induced pluripotent stem cells in human fibroblasts. Proc. Natl. Acad. Sci. U. S. A. 108, 9875–9880 (2011). 88. Ogura, F. et al. Human adipose tissue possesses a unique population of pluripotent stem cells with nontumorigenic and low telomerase activities: potential implications in regenerative medicine. Stem Cells Dev. 23, 717–728 (2014). 89. Kuroda, Y. & Dezawa, M. Mesenchymal Stem Cells and Their Subpopulation , Pluripotent Muse Cells , in Basic Research and Regenerative Medicine. Anat. Rec. 110, 98–110 (2014). 90. Yamauchi, T. et al. Therapeutic Effects of Human Multilineage-Differentiating Stress Enduring (MUSE) Cell Transplantation into Infarct Brain of Mice. PLoS One 10, (2015). 91. Kinoshita, K., Kuno, S., Ishimine, H. & Aoi, N. Therapeutic Potential of Adipose-Derived SSEA-3-Positive Muse Cells for Treating Diabetic Skin Ulcers. Stem Cells Transl. Med. 4, 146–155 (2015). 92. Grayson, W., Zhao, F., Izadpanah, R., Bunnell, B. & Ma, T. Effects of Hypoxia on Human Mesenchymal Stem Cell Expansion and Plasticity in 3D Constructs. J. Cell. Physiol. 207, 331–339 (2006). 93. Dos Santos, F. et al. Ex vivo expansion of human mesenchymal stem cells: A more effective cell proliferation kinetics and metabolism under hypoxia. J. Cell. Physiol. 223, 27–35 (2010). 94. Crisan, M. et al. A perivascular origin for mesenchymal stem cells in multiple human organs. Cell Stem Cell 3, 301–13 (2008). 95. Lin, G. et al. Defining Stem and Progenitor Cells within Adipose Tissue. Stem Cells Dev. 17, 1053–1063 (2008). 96. Caplan, A. I. All MSC are Pericytes? Cell Stem Cell 3, 229–230 (2008). 97. Mohyeldin, A., Garzón-Muvdi, T. & Quiñones-Hinojosa, A. Oxygen in stem cell biology: A critical component of the stem cell niche. Cell Stem Cell 7, 150–161 (2010). 98. Orabi, H., AbouShwareb, T., Zhang, Y., Yoo, J. J. & Atala, A. Cell-seeded tubularized scaffolds for reconstruction of long urethral defects: a preclinical study. Eur. Urol. 63, 531–8 (2013). 99. Raya-Rivera, A. et al. Tissue-engineered autologous urethras for patients who need reconstruction: an observational study. Lancet 377, 1175–82 (2011). 100. Yamamoto, T. et al. Periurethral injection of autologous adipose-derived regenerative cells for the treatment of male stress urinary incontinence: Report of three initial cases. Int. J. Urol. 19, 652–659 (2012). 101. Aicher, W. et al. Towards a Treatment of Stress Urinary Incontinence: Application of Mesenchymal Stromal Cells for Regeneration of the Sphincter Muscle. J. Clin. Med. 3, 197–215 (2014). 102. Gang, E., Jeong, J., Hong, S. & Hwang, S. Skeletal Myogenic Differentiation of Mesenchymal Stem Cells Isolated from Human Umbilical Cord Blood. Stem Cells 22, 617–624 (2004). 103. Gunetti, M. et al. Myogenic potential of whole bone marrow mesenchymal stem cells in vitro and in vivo for usage in urinary incontinence. PLoS One 7, (2012). 104. Beier, J. P. et al. Myogenic differentiation of mesenchymal stem cells co-cultured with primary myoblasts. Cell Biol. Int. 35, 397–406 (2011). 105. Garza-Rodea, A. S. De et al. Myogenic Properties of Human Mesenchymal Stem Cells Derived From Three Different Sources. Cell Transplant. 21, 153–173 (2012). 106. Belanto, J. J. et al. Dexamethasone induces dysferlin in myoblasts and enhances their myogenic differentiation. Neuromuscul Disord. 20, 111–121 (2010). 107. Lee, E. a., Im, S.-G. & Hwang, N. S. Efficient myogenic commitment of human mesenchymal stem cells on biomimetic materials replicating myoblast topography. Biotechnol. J. 9, 1604–1612 (2014).

108. Bourin, P. et al. Stromal cells from the adipose tissue-derived stromal vascular fraction and culture expanded adipose tissue-derived stromal/stem cells: A joint statement of the International Federation for Adipose Therapeutics and Science (IFATS) and the International So. Cytotherapy 15, 641–648 (2013). 109. Bourin, P. et al. Myogenic Differentiation by Human Processed Lipoaspirate Cells. Cytotherapy 15, 641– 648 (2013). 110. Di Rocco, G. et al. Myogenic potential of adipose-tissue-derived cells. J. Cell Sci. 119, 2945–2952 (2006). 111. Shan, T. et al. Distinct populations of adipogenic and myogenic Myf5-lineage progenitors in white adipose tissues. J. Lipid Res. 54, 2214–24 (2013). 112. Mizuno, H. et al. Myogenic Differentiation by Human Processed Lipoaspirate Cells. Plast Reconstr Surg. 109, 199–209 (2002). 113. Dromard, C. et al. Mouse adipose tissue stromal cells give rise to skeletal and cardiomyogenic cell subpopulations. Front. cell Dev. Biol. 2, (2014). 114. Zhu, B. & Murthy, S. K. Stem cell separation technologies. Curr. Opin. Chem. Eng. 2, 3–7 (2013). 115. Diogo, M. M., da Silva, C. L. & Cabral, J. M. S. Separation technologies for stem cell bioprocessing. Biotechnol. Bioeng. 109, 2699–2709 (2012). 116. FACS. Midlands Technical College Webpage. Available at http://classes.midlandstech.edu/carterp/Courses/bio225/chap18/figure_18_12_labeled.jpg (Accessed on November 2015) 117. MACS. Humboldt University of Berlin Webpage. Available at http://edoc.hu-berlin.de/dissertationen/hajkova-petra-2002-09-16/HTML/hajkova_html_m271ca532.gif (Accessed on November 2015)

118. Miltenyi Biotec Web Page - CliniMACS® System. Available at http://www.miltenyibiotec.com/en/clinical- applications/clinimacs-system.aspx (Acessed on November 2015) 119. Grützkau, A. & Radbruch, A. Small but mighty: How the MACS1-technology based on nanosized superparamagnetic particles has helped to analyze the immune system within the last 20 years. Cytom. Part A 77, 643–647 (2010). 120. Pliml, J. & Šorm, F. Synthesis of a 2-deoxy-d-ribofuranosyl-s-azacytosine. Collect. Czechoslov. Chem. Commun. 29, 2576–2578 (1964). 121. Čihák, A., Veselý, J. & Šorm, F. Incorporation of 5-azacytidine into liver ribonucleic acids of leukemic mice sensitive and resistant to 5-azacytidine. Biochim. Biophys. Acta 108, 516–518 (1965). 122. Périgaud, C., Gosselin, G. & Imbach, J. L. Nucleosides and Nucleotides Nucleoside Analogues as Chemotherapeutic Agents : A Review. Nucleosides and Nucleotides 11, 903–945 (1992). 123. Christman, J. K. 5-Azacytidine and 5-aza-2 ’-deoxycytidine as inhibitors of DNA methylation : mechanistic studies and their implications for cancer therapy. Oncogene 21, 5483–5495 (2002). 124. Meligy, F. Y. et al. The efficiency of in vitro isolation and myogenic differentiation of MSCs derived from adipose connective tissue, bone marrow, and skeletal muscle tissue. Vitr. Cell. Dev. Biol. 48, 203–215 (2012). 125. Song, M., Kim, H., Choi, Y., Kim, K. & Chung, C. Skeletal myogenic differentiation of human periodontal ligament stromal cells isolated from orthodontically extracted premolars. Korean J. Orthod. 42, 249–254 (2012). 126. Drost, A. C. et al. In vitro Myogenic Differentiation of Human Bone Marrow – Derived Mesenchymal Stem Cells as a Potential Treatment for Urethral Sphincter Muscle Repair. Ann. N. Y. Acad. Sci. 1176, 135–143 (2009). 127. Tomita, S., Li, R., Weisel, R. D., Mickle, D. A. G. & Kim, E. Autologous Transplantation of Bone Marrow Cells Improves Damaged Heart Function. Circulation 100, 247–256 (1999). 128. Makino, S. et al. Cardiomyocytes can be generated from marrow stromal cells in vitro. J. Clin. Invest. 103, 697–705 (1999). 129. Zhang, Y., Chu, Y., Shen, W. & Dou, Z. Effect of 5-azacytidine induction duration on differentiation of human first-trimester fetal mesenchymal stem cells towards cardiomyocyte-like cells. Interact. Cardiovasc. Thorac. Surg. 9, 943–946 (2009). 130. Zhou, G.-S., Zhang, X.-L., Wu, J.-P. & Zhang, R.-P. 5-Azacytidine facilitates osteogenic gene expression and differentiation of mesenchymal stem cells by alteration in DNA methylation. Cytotechnology 60, 11– 22 (2009). 131. Fu, Q., Song, X.-F., Liao, G.-L., Deng, C.-L. & Cui, L. Myoblasts differentiated from adipose-derived stem cells to treat stress urinary incontinence. Urology 75, 718–723 (2010). 132. Rosca, A. & Burlacu, A. Effect of 5-Azacytidine : Evidence for Alteration of the Multipotent Ability of Mesenchymal Stem Cells. Stem Cells Dev. 20, 1213–1221 (2011).

133. Liu, Y. et al. Growth and differentiation of rat bone marrow stromal cells : does 5-azacytidine trigger their cardiomyogenic differentiation ? Cardiovasc. Res. 58, 460–468 (2003). 134. Martin-Rendon, E. et al. 5-Azacytidine-treated human mesenchymal stem/progenitor cells derived from umbilical cord, cord blood and bone marrow do not generate cardiomyocytes in vitro at high frequencies. Vox Sang. 95, 137–148 (2008). 135. McKee, J. A. et al. Human arteries engineered in vitro. EMBO Rep. 4, 633–638 (2003). 136. Galmiche, B. M. C. et al. Stromal Cells From Human Long-Term Marrow Cultures Are Mesenchymal Cells That Differentiate Following a Vascular Smooth Muscle Differentiation Pathway. Blood 82, 66–76 (1993). 137. Iwase, T. et al. Comparison of angiogenic potency between mesenchymal stem cells and mononuclear cells in a rat model of hindlimb ischemia. Cardiovasc. Res. 66, 543–551 (2005). 138. Sinha, S. et al. Transforming growth factor- beta 1 signaling contributes to development of smooth muscle cells from embryonic stem cells. Am. J. Physiol. cell Physiol. 22908, 1560–1568 (2004). 139. Kinner, B., Zaleskas, J. M. & Spector, M. Regulation of Smooth Muscle Actin Expression and Contraction in Adult Human Mesenchymal Stem Cells. Exp. Cell Res. 278, 72–83 (2002). 140. Tamama, K., Sen, C. K. & Wells, A. Differentiation of bone marrow mesenchymal stem cells into the smooth muscle lineage by blocking ERK/MAPK signaling pathway. Stem Cells Dev. 17, 897–908 (2008). 141. Gong, Z. & Niklason, L. E. Small-diameter human vessel wall engineered from bone marrow-derived mesenchymal stem cells (hMSCs). FASEB J. 22, 1635–48 (2008). 142. Tian, H. et al. Myogenic differentiation of human bone marrow mesenchymal stem cells on a 3D nano fibrous scaffold for bladder tissue engineering. Biomaterials 31, 870–877 (2010). 143. Wang, D. et al. Proteomic profiling of bone marrow mesenchymal stem cells upon transforming growth factor beta1 stimulation. J. Biol. Chem. 279, 43725–34 (2004). 144. Bajpai, V. K., Mistriotis, P., Loh, Y.-H., Daley, G. Q. & Andreadis, S. T. Functional vascular smooth muscle cells derived from human induced pluripotent stem cells via mesenchymal stem cell intermediates. Cardiovasc. Res. 96, 391–400 (2012). 145. Nishimoto, S. & Nishida, E. MAPK signalling: ERK5 versus ERK1/2. EMBO Rep. 7, 782–786 (2006). 146. Kim, H. J. & Bar-Sagi, D. Modulation of signalling by Sprouty: a developing story. Nat. Rev. Mol. Cell Biol. 5, 441–450 (2004). 147. Wang, Z. et al. Myocardin and ternary complex factors compete for SRF to control smooth muscle gene expression. Nature 428, 185–189 (2004). 148. Van Tuyn, J. et al. Activation of cardiac and smooth muscle-specific genes in primary human cells after forced expression of human myocardin. Cardiovasc. Res. 67, 245–255 (2005). 149. Alimperti, S., You, H., George, T., Agarwal, S. K. & Andreadis, S. T. Cadherin-11 regulates both mesenchymal stem cell differentiation into smooth muscle cells and the development of contractile function in vivo. J. Cell Sci. 127, 2627–38 (2014). 150. Wang, T., Xu, Z., Jiang, W. & Ma, A. Cell-to-cell contact induces mesenchymal stem cell to differentiate into cardiomyocyte and smooth muscle cell. Int. J. Cardiol. 109, 74–81 (2006). 151. Jang, J., et al. Combined effects of surface morphology and mechanical straining magnitudes on the differentiation of mesenchymal stem cells without using biochemical reagents. J. Biomed. Biotechnol. 2011, 1–9 (2011). 152. Liu, J., Wang, Y., Wu, Y., Ni, B. & Liang, Z. Sodium Butyrate Promotes the Differentiation of Rat Bone Marrow Mesenchymal Stem Cells to Smooth Muscle Cells through Histone Acetylation. PLoS One 9, (2014). 153. Parmar, N., Ahmadi, R. & Day, R. M. A Novel Method for Differentiation of Human Mesenchymal Stem Cells into Smooth Muscle-Like Cells on Clinically Deliverable Thermally Induced Phase Separation Microspheres. Tissue Eng. Part C. Methods 00, 1–9 (2014). 154. Shah, F. S., Wu, X., Dietrich, M., Rood, J. & Gimble, J. M. A non-enzymatic method for isolating human adipose tissue-derived stromal stem cells. Cytotherapy 15, 979–985 (2013). 155. Butler, M. Animal Cell Culture and Technology. (Garland Science/BIOS Scientific Publishers, 2005). 156. Pisani, D. F. et al. Hierarchization of myogenic and adipogenic progenitors within human skeletal muscle. Stem Cells 28, 2182–2194 (2010). 157. Pisani, D. F. et al. Isolation of a Highly Myogenic CD34-Negative Subset of Human Skeletal Muscle Cells Free of Adipogenic Potential. Stem Cells 28, 753–764 (2010). 158. Pilz, G. a et al. Human mesenchymal stromal cells express CD14 cross-reactive epitopes. Cytometry. A 79, 635–45 (2011).

159. Basu, J. et al. Expansion of the human adipose-derived stromal vascular cell fraction yields a population of smooth muscle-like cells with markedly distinct phenotypic and functional properties relative to mesenchymal stem cells. Tissue Eng. Part C. Methods 17, 843–860 (2011). 160. Kundrotas, G. Surface markers distinguishing mesenchymal stem cells from fibroblasts. Acta medica Litu. 19, 75–79 (2012). 161. Christodoulou, I., Kolisis, F. N., Papaevangeliou, D. & Zoumpourlis, V. Comparative Evaluation of Human Mesenchymal Stem Cells of Fetal (Wharton’s Jelly) and Adult (Adipose Tissue) Origin during Prolonged In Vitro Expansion: Considerations for Cytotherapy. Stem Cells Int. 2013, (2013). 162. Neuhuber, B., Swanger, S. A., Howard, L., Mackay, A. & Fischer, I. Effects of plating density and culture time on bone marrow stromal cell characteristics. Exp. Hematol. 36, 1176–1185 (2008). 163. Vermes, I., Haanen, C., Steffens-Nakken, H. & Reutelingsperger, C. A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J. Immunol. Methods 184, 39–51 (1995). 164. Sinanan, A. C. M., Hunt, N. P. & Lewis, M. P. Human adult craniofacial muscle-derived cells: neural-cell adhesion-molecule (NCAM; CD56)-expressing cells appear to contain multipotential stem cells. Biotechnol. Appl. Biochem. 40, 25–34 (2004). 165. Yeh, S.-P. et al. Induction of CD45 expression on bone marrow-derived mesenchymal stem cells. Leukemia 20, 894–896 (2006). 166. Dupéré-Minier, G., Desharnais, P. & Bernier, J. Involvement of tyrosine phosphatase CD45 in apoptosis. Apoptosis 15, 1–13 (2010). 167. Konieczny, S. F. & Emerson, C. P. Induction of Stable Mesodermal Stem Cell Lineages from 10T1/2 Cells : Evidence for Regulatory Genes Controlling Determination. Cell 38, 791–800 (1984). 168. Bel, A. et al. Transplantation of autologous fresh bone marrow into infarcted myocardium: a word of caution. Circulation 108 Suppl , II247–52 (2003). 169. Andrade, P. Z., Lobato Da Silva, C., Dos Santos, F., Almeida-Porada, G. & Cabral, J. M. S. Initial CD34 + cell-enrichment of cord blood determines hematopoietic stem/progenitor cell yield upon ex vivo expansion. J. Cell. Biochem. 112, 1822–1831 (2011). 170. Zheng, B. et al. Prospective identification of myogenic endothelial cells in human skeletal muscle. Nat. Biotechnol. 25, 1025–1034 (2007). 171. Chen, C.-W., Corselli, M., Péault, B. & Huard, J. Human Blood-Vessel-Derived Stem Cells for Tissue Repair and Regeneration. J. Biomed. Biotechnol. 2012, 1–9 (2012). 172. Li, Y.-M. et al. Effects of high glucose on mesenchymal stem cell proliferation and differentiation. Biochem. Biophys. Res. Commun. 363, 209–215 (2007). 173. Aguiari, P. et al. High glucose induces adipogenic differentiation of muscle-derived stem cells. Proc. Natl. Acad. Sci. U. S. A. 105, 1226–31 (2008). 174. Vettor, R. et al. The origin of intermuscular adipose tissue and its pathophysiological implications. Am J Physiol Endocrinol Metab 987–998 (2009). doi:10.1152/ajpendo.00229.2009 175. Scherberich, A., Di Maggio, N. & McNagny, K. M. A familiar stranger: CD34 expression and putative functions in SVF cells of adipose tissue. World J Stem Cells 5, 1–8 (2013). 176. Bosnakovski, D., Xu, Z., Li, W. & Thet, S. Prospective Isolation of Skeletal Muscle Stem Cells with a Pax7 Reporter. Stem Cells 26, 3194–3204 (2008). 177. Sirabella, D., De Angelis, L. & Berghella, L. Sources for skeletal muscle repair: From satellite cells to reprogramming. J. Cachexia. Sarcopenia Muscle 4, 125–136 (2013). 178. Gimble, J. M., Katz, A. J. & Bunnell, B. A. Adipose-Derived Stem Cells for Regenerative Medicine. Circ. Res. 100, 1249–1260 (2007). 179. Baumgarth, N. & Roederer, M. A practical approach to multicolor flow cytometry for immunophenotyping. J. Immunol. Methods 243, 77–97 (2000). 180. Maumus, M. et al. Native human adipose stromal cells: localization, morphology and phenotype. Int. J. Obes. (Lond). 35, 1141–1153 (2011). 181. Grigoriadis, A. E., Heersche, J. N. M. & Aubin, J. E. Differentiation of muscle, fat, cartilage, and bone from progenitor cells present in a bone-derived clonal cell population: Effect of dexamethasone. J. Cell Biol. 106, 2139–2151 (1988). 182. Guerriero, V. & Florini, J. R. Dexamethasone effects on myoblast proliferation and differentiation. Endocrinology 106, 1198–1202 (1980). 183. Gates, G., Nandan, D., Brickenden, A. & Sanwal, B. Differentiation defective mutants of skeletal myoblasts altered in a gelatin-binding glycoprotein. Biochem Cell Biol. 65, 767–775 (1987).

184. Nusgens, B., Delain, D., Senechal, H. & Winand, R. Metabolic changes in the extracellular matrix during differentiation of myoblasts of the L6 line and of a myo- non-fusing mutant. Exp Cell Res. 161, 51–62 (1896). 185. Nandan, D., Clarke, E. P., Ball, E. & Sanwal, B. Ethyl-3,4-dihydroxybenzoate Inhibits Myoblast Differentiation: Evidence for an Essential Role of Collagen. Cell 110, 1673–1679 (1990). 186. Vaz, R., Martins, G. G., Thorsteinsdóttir, S. & Rodrigues, G. Fibronectin promotes migration, alignment and fusion in an in vitro myoblast cell model. Cell Tissue Res. 348, 569–578 (2012). 187. Chaturvedi, V. et al. Interactions between Skeletal Muscle Myoblasts and their Extracellular Matrix Revealed by a Serum Free Culture System. PLoS One 10, (2015). 188. Grzelkowska-Kowalczyk, K., Wieteska-Skrzeczyńska, W., Grabiec, K. & Tokarska, J. High glucose- mediated alterations of mechanisms important in myogenesis of mouse C2C12 myoblasts. Cell Biol. Int. 37, 29–35 (2013). 189. Munoz, J., Zhou, Y. & Jarrett, H. W. LG4-5 domains of laminin-211 binds alpha-dystroglycan to allow myotube attachment and prevent anoikis. J. Cell. Physiol. 222, 111–119 (2010). 190. Goudenege, S. et al. Enhancement of Myogenic and Muscle Repair Capacities of Human Adipose– derived Stem Cells With Forced Expression of MyoD. Mol. Ther. 17, 1064–1072 (2009). 191. Wang, C. et al. Small activating RNA induces myogenic differentiation of rat adipose-derived stem cells by upregulating MyoD. Int. braz j urol 41, 764–772 (2015). 192. Lingbeck, J. M., Trausch-Azar, J. S., Ciechanover, A. & Schwartz, A. L. Determinants of nuclear and cytoplasmic ubiquitin-mediated degradation of MyoD. J. Biol. Chem. 278, 1817–1823 (2003). 193. Choi, Y. S. et al. The alignment and fusion assembly of adipose-derived stem cells on mechanically patterned matrices. Biomaterials 33, 6943–51 (2012). 194. Schugar, R. C., Robbins, P. D. & Deasy, B. M. Small molecules in stem cell self-renewal and differentiation. Gene Ther. 15, 126–135 (2008). 195. Harris, L., Abdollahi, H., Zhang, P. & McIlhenny, S. Differentiation of Adult Stem Cells into Smooth Muscle for Vascular Tissue Engineering. J Surg Res 168, 306–314 (2011). 196. Zen-Bio Webpage. Available at http://www.zen-bio.com/products/cells/human-bladder-cells.php (Accessed on October 2015) 197. Beamish, J. A., He, P., Kottke-Marchant, K. & Marchant, R. E. Molecular Regulation of Contractile Smooth Muscle Cell Phenotype: Implications for Vascular Tissue Engineering. Tissue Eng. Part B Rev. 16, 467–491 (2010). 198. Rodríguez, L. V et al. Clonogenic multipotent stem cells in human adipose tissue differentiate into functional smooth muscle cells. Proc. Natl. Acad. Sci. U. S. A. 103, 12167–72 (2006). 199. Liu, Y., Deng, B., Zhao, Y., Xie, S. & Nie, R. Differentiated markers in undifferentiated cells: Expression of smooth muscle contractile proteins in multipotent bone marrow mesenchymal stem cells. Dev. Growth Differ. 55, 591–605 (2013). 200. Jaiswal, R., Jaiswal, N., Bruder, S. & Mbalaviele, G. Adult Human Mesenchymal Stem Cell Differentiation to the Osteogenic or Adipogenic Lineage Is Regulated by Mitogen-activated Protein Kinase. J. Cell. Physiol. 275, 9645–9652 (2000). 201. Kurpinski, K. et al. Transforming growth factor-beta and notch signaling mediate stem cell differentiation into smooth muscle cells. Stem Cells 28, 734–42 (2010). 202. Khani, M.-M., Tafazzoli-Shadpour, M., Goli-Malekabadi, Z. & Haghighipour, N. Mechanical characterization of human mesenchymal stem cells subjected to cyclic uniaxial strain and TGF-β1. J. Mech. Behav. Biomed. Mater. 43, 18–25 (2015). 203. Maul, T. M., Chew, D. W., Nieponice, A. & Vorp, D. A. Mechanical stimuli differentially control stem cell behavior: morphology, proliferation, and differentiation. Biomech. Model. Mechanobiol. 10, 939–953 (2011). 204. Rosario, D. J. et al. Decellularization and sterilization of porcine urinary bladder matrix for tissue engineering in the lower urinary tract. Regen. Med. 3, 145–156 (2008). 205. Consolo, F. et al. A dynamic distention protocol for whole-organ bladder decellularization: histological and biomechanical characterization of the acellular matrix. J. Tissue Eng. Regen. Med. 3, 145–156 (2013). 206. Liao, W. et al. Construction of ureteral grafts by seeding bone marrow mesenchymal stem cells and smooth muscle cells into bladder acellular matrix. Transplant. Proc. 45, 730–734 (2013).