DEVELOPMENT OF A HETEROGENEOUS PRO ANGIOGENIC PROTEIN MIXTURE ENCAPSULATION SYSTEM By SARAH TONELLO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
2 Â© 2014 Sarah Tonello
3 T o my family, Luca and all my friends
4 ACKNOWLEDGMENTS I would like to acknowledge the J. Crayton Pruitt Family Department of Biomedical Engineering at the University of Florida a nd the department of Biomedical Engineering at Politecnico di Milano for giving me the opportunity to take part in the formative Atlantis CRISP dual degree program. In particular, I acknowledge Prof essor Van Oostrom, Professor Baselli, Professor Redaelli, Profe ssor Ferrigno, and everyone who took part and made possible the creation of the Atlantis double degree program. I would like to thank my ad visor, Professor Peter McFetridge, for supporting me througho ut the development of the project. He has constantly provided me with guidance and motivation, giving me the opportunit y to become a more independent student and to improve my approach to the research in the interesting field of Tissue Engineering. I would also like to express my gratitude to the member of my disse rtation committee, Professor Jon P. Dobson and Professor Blanka Sharma for having accepted to examine this work and for their suggestions and mentoring throughout the course of the work. I would like to thank my advisor in Politecnico of Milano, Professor Candiani, for his suggestions and help. I would like to thank all my lab mates Alice Cambiaghi, Ben Goldberg, Leslie Goldberg, Mediha Gurel, Andrea Matuska, Marc Moore, Cassandra Juran, Cl audia Siverino and Aurore Van de Walle for being s o helpful every time I needed, for teaching me all the technique I have applied in my research and for making the lab such a nice environment. I would also like to thank Manelle Merad, Enrico Opri , Rosama ria Tricarico, Aniruddh Ravindran, Kathi Jung, Yiqi Gao who supported me and contribut ed to making
5 my life in Gainesville such a unique and pleasant experience . A special thanks to Ben and Aniruddh for reviewing my thesis and for their help with the writin g. I would li ke to thank Roberta Filippini, Roberta Savoldi, Alessandro Gaffurini , Davide Brancato, Libera Maria Sacco , Carlotta Mondadori who w ere always ready to support and motivate me from far away . A special thanks to my boyfriend Luca, for his help, suggestions, support and love he was able to give me all the time. Most of all, I am grateful to my parents Cesare and Chiara who hav e provided me with love , motivation, and guidance. I could not have done all of this without their support.
6 TABLE OF CONT ENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Overview and Rational ................................ ................................ ............................ 14 Specific Aims ................................ ................................ ................................ .......... 17 Specific Aim 1: Characterization of induction of angiogenesis in HUVECs 2D culture usin g human Placental Matrix ................................ ...................... 18 Specific Aim 2: hPM encapsulation in PLGA microparticles and optimization of MPs features and in vitro protein release rate. ................................ .......... 18 Specific Aim 3: Evaluation of the angiogenic effect of hPM controlled release from PLGA microparticles in an Alginate based 3D culture of Human Umbilical Vein Endothelial Cells (HUVECs) ................................ ...... 18 2 BACKGROUND ................................ ................................ ................................ ...... 20 Origin of Blood Vessels: Vasculogenesis and Angiogenesis ................................ .. 20 Mechanisms of angiogenesis ................................ ................................ ........... 20 Angiogenic stimuli ................................ ................................ ............................ 24 Angiogenesis in tissue engineering ................................ ................................ .. 30 Angiogene sis quantification ................................ ................................ .............. 33 Perinatal Tissues ................................ ................................ ................................ .... 35 Perinatal Tissues properties ................................ ................................ ............. 35 Human Placenta: anatomy and main properties ................................ ............... 37 Characterization of human Placental Matrix (hPM) ................................ .......... 40 Biomaterials for Tissue Engineering Applications ................................ ................... 42 Protein controlled release methods ................................ ................................ .. 42 Biomaterials for 3D cell cultures and angiogenic assay ................................ ... 49 3 GENERAL MATERIALS AND METHODS ................................ .............................. 51 Experimental Methods ................................ ................................ ............................ 51 Isolation of Vascular Endothelial Cells from the Human Umbilical Vein ........... 51
7 Freezing of Vascular Endothelial Cells from the Human Umbilical Vein ........... 53 Derivation of human Placental Matrix (hPM) ................................ .................... 54 Analytical Methods ................................ ................................ ................................ .. 56 Calcein AM staining ................................ ................................ .......................... 56 Angiogenesis quantification ................................ ................................ .............. 56 Statistics ................................ ................................ ................................ ........... 58 4 INDUCTION OF ANGIOGENESIS IN HUVECs 2D CULTURE USING HUMA N PLACENTAL MATRIX ................................ ................................ ............................ 60 Background and Rational ................................ ................................ ........................ 60 Materials and Methods ................................ ................................ ............................ 61 Materials ................................ ................................ ................................ ........... 61 Assessment of the correlation between hPM pro angiogenic effect during time and its protein content. ................................ ................................ .......... 62 Effect of multiple inoculations of hPM in a HUVECs culture. ............................ 63 Effect of different modality of hPM delivery on HUVECs 2D culture ................. 64 Resu lts ................................ ................................ ................................ .................... 64 Assessment of the correlation between hPM pro angiogenic effect during time and its protein content. ................................ ................................ .......... 64 Effect of multiple inoculations of hPM in a HUVECs culture ............................. 66 Effect of different modality of hPM delivery on HUVECs 2D culture ................. 67 Discussion ................................ ................................ ................................ .............. 68 5 DEVELOPMENT OF A HETEROGENEOUS PROTEIN MIXTURE ENCAPSULATION METHOD USING PLGA MICROPARTICLES ......................... 76 Background and Rational ................................ ................................ ........................ 76 Material and Methods ................................ ................................ ............................. 77 Materials ................................ ................................ ................................ ........... 77 Preparation of hPM loaded PLGA Micropa rticles ................................ ............. 77 Protocol Optimization ................................ ................................ ....................... 78 PLGA microparticles features characterization: size and surface evaluation ... 79 Loading efficiency ................................ ................................ ............................. 80 In vitro protein release from hPM loaded PLGA microparticles ........................ 81 Sodium Dodecyl Sulfate Polyacrylamide gel electrophoresis (SDS PAGE) ... 82 Results ................................ ................................ ................................ .................... 83 Preparation of hPM loaded PLGA Micropartic les ................................ ............. 83 Protocol Optimization ................................ ................................ ....................... 84 Sodium Dodecyl Sulfate Polyacrylamide gel Electrophoresis (SDS PAGE) .. 87 Discussion ................................ ................................ ................................ .............. 87 6 DEVELOPMENT OF AN ANGIOGENIC ASSAY USING ALGINATE HYDROGEL AND HPM LOADED PLGA MICROPARTICLES ............................... 97 Background and Rational ................................ ................................ ........................ 97 Materials and Methods ................................ ................................ ............................ 98
8 Materials ................................ ................................ ................................ ........... 98 Optimization of 3D culture parameters ................................ ............................. 98 Angiogenic assay: Alginate hydrogel based 3D culture of HUVECs .............. 100 Re sults ................................ ................................ ................................ .................. 101 Optimization of 3D culture parameters ................................ ........................... 101 Angiogenic assay: Alginate hydrogel based 3D culture of HUVECs .............. 1 02 Discussion ................................ ................................ ................................ ............ 104 7 CONCLUSION AND FUTURE WORKS ................................ ............................... 109 Summary ................................ ................................ ................................ .............. 109 Future Works ................................ ................................ ................................ ........ 111 Characterization of the protein released from PLGA microparticles ............... 111 Optimization of matrix composition for the angiogenic assay ......................... 111 Realization of a smart controlled release using iron oxide nanoparticles (IO NPs) ................................ ................................ ................................ ............ 112 LIST OF REFERENCES ................................ ................................ ............................. 113 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 122
9 LIST OF TABLES Table page 4 1 Parameters for angiogenesis quant ification of HUVECs at different time points from hPM inoculation ................................ ................................ ............... 65 4 2 Parameters evaluated for angiogenesis quantification of HUVECs inoculated with different inoculations of hPM. ................................ ................................ ...... 66 4 3 Parameters for angiogenesis quantification of HUVECs comparing coated hPM with inoculated hPM ................................ ................................ ................... 68
10 LIST OF FIGURES Figure page 2 1 Mechanisms of angiogenesis. ................................ ................................ ............ 22 2 2 Gross structure and function of human Placenta ................................ ................ 37 2 3 Detailed an atomy of human Placenta ................................ ................................ . 38 2 4 Schematic representation of distinct uses of particles in the TE field according to size ................................ ................................ ................................ . 43 2 5 Mec hanisms for protein release ................................ ................................ .......... 46 3 1 Angiogenesis quantification . ................................ ................................ ............... 58 4 1 Assessment of the correlation between hPM effect and its pr otein content:. ...... 71 4 2 Morphological and topographic features of angiogenic network ......................... 72 4 3 Effect of multiple inoculation of hPM o n different densities of HUVECs:. ........... 73 4 4 Morphological and top ographic features of angiogenic network. ........................ 74 4 5 Effect of different hPL delivery methods . ................................ ............................ 75 5 1 Results from original protocol hPM encapsulation. ................................ ............. 90 5 2 Effect of BSA co encapsulation on PLGA micro particles size.. .......................... 91 5 3 Effect of BSA co encapsulation on PLGA microparticles release . ...................... 92 5 4 Microparticles size evaluation. ................................ ................................ ............ 93 5 5 Release rate evaluation. ................................ ................................ ..................... 94 5 6 SEM characterization.. ................................ ................................ ....................... 95 5 7 SD S page analysis of microparticles supernatant. ................................ ............. 96 6 1 Optimization of 3D culture matrix preparation ................................ ................... 106 6 2 3D Alginate based hydr ogel c ulture system : 7,14, 21 days . ............................. 107 6 3 3D Alginate based hydrogel c ulture system : 2,4,6,8,10 days . .......................... 108
11 LIST OF ABBREVIATIONS BP Branch ing Point CLS Capillary like Structures DDS Drug Delivery System EC Endothelial Cell ECM Extra Cellular Matrix EPC Endothelial Progenitor Cell FDPE Freeze Dried Placental Extract FGF Fibroblast Growth Factor HME Hot Melt Extrusion HUVEC Human Umbelical Vein Endothelial Cell PBS Phosphate Buffer Saline PDECGF Platelet Derived Growth Factor PLGA Poly (lactic co glycolic acid ) hPM human Placental Matrix SEM Scanning Electron Microscopy TBS Tris Buffer Saline TGF Transfo rming Growth Factor TSP 1 Thrombospondin 1 VEGF Vascular Endothelial Growth Factor W/O/W Water in Oil in Water
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DEVELOPMENT OF A HETEROGENEOUS PRO ANGIOGENIC PROTEIN MIXTURE ENCAPSULATION SYSTEM By Sarah Tonello August 2014 Chair: Peter S. McFetridge Major: Biomedical Engineering Tissue engineering aims to deve lop functional organs and tissues, using a therapeutic combination of exogenous cells, 3D scaffolds and growth factors. One of the most significant hurdles limiting organ regeneration is the development of an effective vasculature supply of oxygen and nutr ients to cells seeded in the engineered construct. Current approaches have remained unsatisfactory. As a possible solution to this problem, our laboratory developed a pro angiogenic extract from the human placenta, referred to as Human Placental Matrix (hP M). Despite the observation that the extract induces and modulates the initial stages of angiogenesis, the newly formed networks degrade 5 days after initial treatments. Multiple applications of hPM at discrete time points promoted the formation of a more mature and stable capillary network, thus the use of a controlled release method was hypothesized stabilize network formation over extended periods. In these investigations, hPM was encapsulated using poly(lactic co glycolic acid) (PLGA) microparticles to extend the release period, without the use of crosslinking agents. Following optimization of the microparticle preparation phase, microparticle
13 morphological features (size, encapsulation efficiency, porosity) were characterized and the associated protein release profiled. Subsequently, the cellular response to hPM delivered via PLGA microparticles was assessed using 2D and 3D Alginate based hydrogel culture system with Human Umbilical Vein Endothelial Cells (HUVECs), to assess the angiogenic response. Res ults from the optimized encapsulation process showed microparticles with an average size of 91.82 Â± 2.92 Âµm, with an encapsulation efficiency of 75%, and a release profile extending over 30 days. 3D angiogenic assay with hPM loaded PLGA microparticles embe dded showed initial stimulation of tubular angiogenic structures after 14 days and formation of a more mature angiogenic network after 21 days of culture. Although optimization is required, a sustained angiogenic response over an extended period of 21 days was observed within the 3D hydrogel culture system. This confirmed the effectiveness of the controlled hPM release approach to guide formation and maintain capillary networks.
14 CHAPTER 1 INTRODUCTION Overview and R ational Tissue engineering encompasses a promising field of biomedical engineering to overcome the problems of biocompatibility, such as the undesired inflammatory and immunogenic responses ass ociated with the implant of synthetic devices , or xenographic and allographic tissues. Langer R 1 defines the discipline of tissue engineering as "an interdisciplinary field that applies the principles of engineering and life sciences toward the development of biological substitutes that restore, maintain, or improve tissue functi on or a whole organ . " A more recent definition is given by MacArthur BD 2 , w here the final clinical application is stressed , defining tissue engineering as "understanding the principles of tissue growth, and applying this to produc e functional replacement tissue for clinical The main aim of tissue engineering is therefore t o realize a functional part of tissue or organ utilizing a biocompatible and biodegradable scaffold , oftentimes seeded with autologous and/or stem cells. Wi th proper chemical and mechanical conditioning, the final product ought to be an engineered graft able to be integrated into the body, with the aim of effectively restoring the function of the injured (or malfunctioning) organ . One of the most important is sue s related to realizing a functional engineered tissue is ensuring that the introduced cellularized scaffold receives proper nourishment and gas exchange . In every human tissue this function is carried out b y the blood vasculature, which permit s the exch ange of oxygen and of nutrients in all the cells . An interesting study from Francis et al., 1997, estimated that the domain that a single cell can effectively communicate in, receive nutrients and oxygen is 250 Âµm in size.
15 Since an effective communication between neighboring cell s is essential for tiss ue function, this results have funda mental implications for engineering tissue function ex vivo , limiting the thickness of constructs 3 . T herefore, one of the most pervasive contempo rary challenge s across the field of tissue engineering is to properly guide the formation and maintenance of capillary networks within engineered tissue constructs. This could help to overcome the limitation related to the thickness and the size of the eng ineered constructs, providing nutrients and oxygen to cells. Vasculogenesis and angiogenesis represent two complex processes referring to the de novo formation of capillary vessels from endothelial cells or from pre exi sting vessels, respectively. Both proc esses take place during embryonic development, when the vasculature of the entire fetus must be formed , and in adult s, particularly in the processes of wound healing and tumor formation . It is believed that a complex molecular array of growth factors and other signaling molecules work synergistically with a variety of cell types to collectively stimulate, regulate, and maintain the process o f angiogenesis. Several approaches have been evaluated to induce angiogenesis in engineered constructs, with the aim of improving their long term function: the direct or sustained delivery of several angiogenic growth factors ( notably, VEGF), alone and in combination 4 , 5 ; the use of gene therapies as a tool to enable sustained releas e of these angiogenic proteins; the use of a cell based therapy using endothelial cells to create microvascular networks ; modifications in scaffold design and chem ical composition 6 , 7 .
16 To date, none of these approaches have proved clinically efficacious. Translation into the clinic has been prevented due to two critical issues: the lack of stability over time and the poor biocompatibility of animal derived pro angiogenic matrices (e.g. Matrigel from Eng elbreth Holm Swarm mouse sarcoma) 8 . To overcome these issues, we focused our attention on the human placenta, the organ that acts as an interface between mother and fetus, which i s both richly vascularized and readily available. Through a p articular p rocess, the protocol for the derivation of a pro angiogenic extract from the human placenta was optimized. The mixture, referred to as human Placental Matrix (hPM), has been determined to contain more than 2600 proteins, among which are several important angiogenic growth factors. Both in vitro and in vivo studies of effect on cells were conducted . HUVECs were seeded onto an hPM coat ed 96 well plate , and following 24 hours of cul ture, they started to organize i n to an angiogenic capillary network. Some angiogenesis related proteins not present in the extract, including VEGFA, were up regulated in the HUVECs when seeded onto hPM. Animal studies showed that the incubation of a scaffo ld in hPM before the implantation in a rat model, compared with Matrigel or with PBS (control), improved capillary network formation and reduced fibrotic capsule formation 9 . E ven with hPM shown to modulate the initial stages of a ngiogenesis , and the formation of a mature angiogenic network after a few days of cell culture, it appeared to start degrading after 5 days. This finding suggested that a controlled release of hPM could help to form a more stable capillary network.
17 Among FDA a pproved PLA copolymers, PLGA is one of the most used to achieve the controlled release of growth factors (e.g. BMP, VEGF, bFGF) 10 , 11 , 12 for tissue engineering application s. Mu ltiple studies investigated how to optimize the rele ase of a single protein, some evaluate d strategies to obtain dual release , embedding microparticles in a polymeric matrix or constructing microparticles with multiple shells , 13 and a few studies have evaluated the possibility of co encapsulat ing two different protein s in PLGA microparticles 14 15 .To date, to the best of our knowledge, this is the first study in which a complex , heterogeneous mixture of proteins was encapsulate d using PLGA microparticles . In light of this, the present study aims to optimize a suitable controlled release method us ing PLGA microparticles to deliver hPM for a longer period of time, in order to allow the cells to organize in to a more mature, stable and long lasting vascular network. The final aim w ill be to assess the effect of hPM controlled released from PLGA microparticles on the stability of the angiogenic network, using a 3D hydrogel ba sed cell culture of endothelial cells. Specific Aims The specific aim o f this project is to optimize a protocol for encapsulation of hPM in PLGA microparticle s and to evaluate the angiogen ic effect of this protein mixture if controlled release in a n endo thelial cells culture. In this project, we aim to evaluate associate total protein release rate over time, and the effect of the proteins released from PLGA microparticles on endothelial cultures to assess the usefulness of this method to induce a more mature network of capillaries.
18 Specific Aim 1: Characterization of i nduction of angiogenesis in HUVEC s 2D cultur e using human Placental Matrix The first aim of this project was to characterize the evolution of the angiogenic network of HUVECs in response to hPM under different conditions. The first experiment was related to the evaluation of HUVECs response to hPM in which denaturation of proteins was induced, to confirm a correlation between protein function and angiogenic ne twork formation.The second experiment aimed to assess the need of a controlled release method to improve network stability. The use of multiple inoculations of hPM was addressed to mimic a controlled release during time. Specific Aim 2: hPM encapsulation i n PLGA microparticles and optimization of MPs features and in vitro protein release rate. The second aim of the project was to assess the feasibility of the encapsulation of the complex protein mixture in PLGA microparticles evaluating encapsulation effici ency, and release rate. After that, the protocol was optimized to get to the desired results for our specific cell culture application, in term of size, encapsulation efficiency and release rate. Specific Aim 3: Evaluation of the a ngiogenic effect of hPM controlled release from PLGA microparticles in an Alginate based 3D culture of Human Umbilical Vein Endothelial Cells (HUVECs) The third aim of this project was to assess the effect of hPM controlled released using PLGA microparticl es on HUVECs. An optimization of the 3D culture preparation was performed to obtain the desired features of Alginate matrix, suitable to realize the angiogenic assay. A 2D culture of HUVECs cells was realized to evaluate differences in HUVECs organization when seeded onto hPM, when hPM was inoculated in the angiogenic media and when hPM was released from PLGA microparticles. After that, an angiogenic assay was realized using a 3D Alginate based hydrogel culture system ,
19 comparing in a long term culture (21 d ays) the stability of angiogenic network induced by pure hPM to the one induced by hPM delivered by PLGA microparticles.
20 CHAPTER 2 BACKGROUND Origin of Blood Vessels: Vasculogenesis and A ngiogenesis An overview of the main molecular mechanisms and factor s involved in the angiogenic process will be discussed. Angiogenic induction in tissue engineering will be covered as well. Mechanisms of angiogenesis The cardiovascular system is the first to be developed in the embryo, since it is the only one able to gu arantee the survival of all the cells in the body, providing them with nourishment and oxygen. The first recor ded scientific insight into the understanding of the mechanism of formation of new vessels dates back to the Scottish anatomist and surgeon John H unter. His observations (summarized in his Treatise published in 1794) suggested that proportionality between vascularity and metabolic requirements occurs in both health and disease. Although the word angiogenesis never appears in his writings, he was the first to show interest for this field investigating the mechanisms of angiogenesis, better explained in 1970 from Judah Folkman, who gave rise to the modern history of angiogenesis, mainly with his hypothesis related to the angiogenesis dependent nature o f tumor growth 16 . Recognition that control of angiogenesis could have a therapeutic value has stimulated great scientific interest during the past 40 years. One of the main goals to improve results of research areas such as cel l therapy, regenerative medicine and tissue engineering is to gain a deeper understanding of the molecular mechanism of angiogenesis and of the complex arrays of chemical factor involved in the process.
21 The formation of new vessels starts from particular c hanges involving circulating endothelial cells progenitors in the embryo and the derived differentiated cells in adults. Although initially this process was thought to occur only during the embryo growth, further studies have demonstrated that it takes pl ace in the adult age as well, particularly during wound healing process and during the transition from a benign to a malignant form of tumor. Endothelial cells in their mature stage are therefore able to build channels that efficiently distribute blood in the various part of the body: even if they are elongated, thin and fragile cells, their inability to collapse derives from their capability to sense changes both in the mechanics of blood pressure and in the chemical composition of ECM 17 . The formation of new blood vessels can be classified in two processes: vasculogenesis and angiogenesis. Sometimes neovascularization has been used as a synonym for angiogenesis, but it specifically refers to the formation of any blood vessel in t he adult age, regardless of its size or type: it therefore includes other types of vascular growth like arteriogenesis, venogenesis and lymphangiogenesis. Vasculogenesis refers to the formation of new vessels starting from endothelial cell s precursors . It is a dynamic process which involves cell cell and cell matrix interactions, temporarily and spatially directed by a complex of growth and morpho genic factors. Even if it takes place mainly during the embryo development, it has been shown that it can be ini tiated even in adults from circulating endothelial progenitor cells (EPC) (derivatives of stem cells) in specific conditions such as tumor growth, revascularization after ischem ia and in case of endometriosis 18 .
22 Vasculogenesis is initiated by angioblasts, derived from lateral mesoderm, which in the bone marrow differentiate into Endothelial cells (EC). Once they differentiate, they start to organize primordial vessels, due to cell cell contact, which evolve after that in an endo thelial cell tube. Finally , a primary vascular network is originated from these different tubes and then stabilized due to pericytes and vascular smooth muscle cells. Differently from vasculogenesis, angiogenesis refers to the growth of vessels from alread y existing ones, both in the embryo development and in adults. Two different processes can lead to angiogenesis: the most known one is named sprouting angiogenesis, discovered about 200 years age, and the less known discovered by Butti et al. only a couple of decades ago, is call ed intussusceptive angiogenesis 19 . Figure 2 1 : Mechanisms of angiogenesis.The arrows indicate the direction of propagation of the angiogenic sti muli. (Source: http://educationportal.com/academy/lesson/ what is angiogenesis --definition factors quiz.ht mL ) As the name implies, sprouting angiogenesis is characterized by sprouts composed of endothelial cells, growing toward a specific angiogenic stimulus from the wall of an already existing blood vessel. The process of sprouting angiogenesis is
23 usually initiated in poorly oxygenated tissues. Parenchymal cells can sense a level of hypoxia, since it demands the formation of new vessels to satisfy their metabolic requirements. Their response to the low level of oxygen is the secretion of a key proangiogenetic factor called vascular endothelial cells growth factor (VEGF A). Not all the endothelial cells can initiate the process of sprou ting angiogenesis; a particular ponsible to guide the capillary sprouting through the extracellular matrix toward an angiogenic stimulus. Six main phases can be described in the basic sprouting angiogenesis 19 , 20 : (1) a pa rticular endothelial cells exposed to the angiogenic factor (VEGF A) become tip cells, owing to a specific signaling cascade induced by VEGF ; (2) t he tip cells lead the developing sprout by extending numerous filopodia; (3) t he sprouts elongate due to the prolifera tion of endothelial stalk cells; (4) o nce the sprouts are long enough, the tip cells from two developing s prouts fuse and create a lumen; (5) b lood which starts to flow, oxygenate the tissue and decrease the amount of angiogenic factor ; and (6) fi nally p ericytes, ECM and mechanical factor stabilize the newly developed capillary. Intussusceptive angiogenesis, also called splitting angiogenesis, refers to a different angiogen ic process which does not start from the migration and proliferation of a ti p cells, but which is initiated by opposite ECs into the capillary lumen that protrude forming an endothelial junction. After that, the endothelial bilayer and the basement membrane are perfora ted allowing growth factors to enter and finally fibroblast and pericytes stabilized the newly formed vessels. Since this process requires only a reorganization of endothelial cells and does not involve the long processes related with the proliferation and migration of endothelial cells, it is thought to be faster and more
24 efficient compared with the sprouting one. Beca use of its recent discover, the mechanism of intussusceptive angiogenesis is still poorly understood and the research in this particular branch is currently particularly active. Angiogenic stimuli A comp lex array of mechanical and chemical factors is known to play an important role in the processes of vasculogenesis and angiogenesis. Mechanical stimuli have not been fully characterized yet. There is a significant amount of controversy with regards to shea r stress acting on capillaries to cause angiogenesis, although current knowledge suggests that increased muscle contractions may increase angiogenesis. This may be due to an increase in the production of nitric oxide during exercise , which results in vasod ilation of blood vessels 21 . Chemical stimuli are predominant to regulate angiogenesis: when endothelial cells are subjected to hypoxia, they respond to the low level of oxygen producing an increasing amount of chemical growth fac tors (such as VEGF) which induce the angiogenetic process and stabilize the capillary network. Different families of growth factors are involved in the control of the angiogenetic process. The main pro angiogenic factors identified are Vascular Endothelial Growth Factor (VEGF), basic Fibroblast Growth Factor (bFGF), Platelet derived Growth Factor (PDGF), Transforming Growth Fac tor process such as Keratinocyte Growth Factor (KGF) , Hepatocyte Growth Factor and Angiogenin 22 . Angiostatic factors play an important contrasting role in the complex array of chemical factors which regulate the delicate balance of the angiogenetic process. These c hemical signal s which inhibit the growth of new blood vessels can be
25 endogenous , produced normally within the body (for example angiostatin, endostatin or thrombospondin ) or exogenous, introduced through drugs or diet 23 . A deeper u nderstanding of the anti angiogenetic effect of these molecules, could help to provide useful treatments applicable to many ty pes of cancer, macular degeneration in the eye and several other diseases which involve proliferation of abnormal vessels 24 , 25 . Pro a ngiogenic factors: Vascular Endothelial cells Growth Factors ( VEGF ) VEGF represents the most known and investigated regulator of neo vascularization, whose importance is due to the ability to initiate alone t he angiogenic process, then completed with the co operation of other growth factor families. This sulfide bo nded dimeric glycoprotein is involved in the regulation of multiple steps of angiogenesis 4 , from the activation and regulation of migration and prol iferation of endothelial cells, to the induction of tubular like formations and vascular hyperpermeability 26 . Different VEGF forms exist, and one of the peculiarities of VEGF family is the largely non redundant role of its membe rs, each one with specific role in angiogenesis, embryonic angiogenesis, vasculogenesis or lymphoangiogenesis. VEGF production is stimulated in endothelial cells by a transcription factor, called hypoxia inducible factor (HIF) , which is expressed when cell s are not receiving enough oxygen. The mechanism used by VEGF to stimulate cellular response and induce angiogenesis is by binding to tyrosine kinase receptors (the VEGFRs) on endothelial
26 cells surface, causing them to dimerize and become activated through transphosphorylation. The main component of the VEGF family is VEGF A; it stimulates angiogenesis in health and dis ease by signaling mainly through VEGF receptor 2 ( VEGFR 2, also known as FLT1), and through other co receptors, called neuropilins ( NRP1 an d NRP2 ), which enhance the activity of VEGFR 2 but also signal independently . The response of endothelial cells to a VEGF gradient is highly organized: it induces a specific gene expression cascade with the final aim to ensure th at the specific EC stimula ted becomes a tip cell and takes the lead to initiate the angiogenic process . Two different isoforms of VEGF factors exist, with defined functions: soluble VEGF factors, which promote vessel enlargement, and matrix bound, which stimulate branching. Anothe r important difference has been noticed between paracrine and autocrine VEGF: the first one essentially released by tumo r s , increases vessel branching and makes tumor vessels abnormal, whereas the second , released by endo thelial cells, maintains vascular h omeostasis . The presence of mutation in VEGF growth factor and of its receptors results in vascular tumors and in pathological angiogenesis, while the deficiency of them aborts vascular development, because of the expression of angiopoietin 2 27 . The research for an effective anti VEGF therapy to stop the development of vessels in pathologies such as various types of cancers and in macular degenera tion , represent a really active field. Even if in 2008 it was demonstrated that ant i VEGF drugs show therapeutic efficacy in mouse models of cancer and in an increasing number of human cancers, the transitory
27 benefits and the restoration of tumor growth and progression in many cases enhanced the need of further investigations 28 . Fibroblast Growth Factors ( FGF ) Fibroblast Growth Factors (FGF) family represent s one of the first discovered and investigated among angiogenic factors. Its members are involved in a wide range of biological functions, such as angiogene sis, arteriogenesis, wound healing, cell and tissue proliferation and differentiation, embryonic development and various endocrine signaling pathways. FGF induces angiogenesis by binding heparin domain and through interactions with cell surface associated heparan sulfate proteoglycans. FGFs can activate directly the specific receptors (FGFRs) on endothelial cells or indirectly stimulate angiogenesis by inducing the release of angiogenic factors from other cell types. FGF1 and FGF2 are the specific factors involved in the control of the angiogen ic process . They are able to induce endothelial cell proliferation and their physical organization into tube like structures, promoting the growth of new blood vessels from the pre existing vasculature. They have bee n demonstra ted to be able to induce revascularization of ischemic tissues, more than VEGF and Platelet derived Growth Factor (PDGF) 29 , 30 . Because of the partial redundancy of FGF family and since the absence of FGF was shown to not produce vascular defects, the d evelopment of specific FGF or FGFR inhibitors for blocking angiogenesis is lagging behind 27 . Platelet derived Growth Factor (PDGF) Platelet derived growth factor ( PDGF) is one of the main and firstly discovere d 31 indirect pro angiogenic factors, initially isolated from platelets, but then identified in
28 other different cell types under certain conditions. PDGF is known to be a potent mitog en for cells of mesenchymal origin, including smooth muscle cells and glial cells , but it plays an important role in the angiogenetic process as well. The interaction between PDGF BB and its specific receptor on endothelial cells (PDGF tion, sprouting and proliferation in vitro 4 . M oreover, since several studies showed that a deficiency for PDGF B or PDGF receptors cause malformation or incomplete development of vessels wall, this growth factor is essential to achieve v essel stabilization and to avoid vessel leakage, tortuosity, micro aneurysm formation and bleeding 11 . Three different isoforms of PDGF have been identifie d (PDGF AA, PDGF BB, PDGF AB ) and two diffe rent types of receptors: alpha type and beta type PDGFRs . PDGF induce angiogenic response with a mechanism similar to VEGF . On the side of therapies against tumor angiogenesis, PDGFR inhibition diminishes tu mor growth by causing pericyte detachment, leadin g to immature vessels that are prone to regressi on. Overall, further studies are required to explore the be nefits and risks of PDGF inhibition for the treatment of cancer 27 . Transforming Growth Factor Beta (TGF ) Tran sforming growth factor ) is a polypeptide, secreted as inac tive precursors, which need to be activated b y proteases, low pH or heat to act in the angiogenetic process. In addition to important functions related to the control of EC prolifer ation and differentiation , it is a type of cytokine which plays a role in establishing and maintaining the vessel wall integrity. TGF iostatic or angiogenic molecule, depending on the quantity circulating: low doses of TGF stimulate
29 p roliferation of endothelial cells and tube formation in vitro , whereas higher doses have contrary effects. TGF beta is produced by many different cell types, including macrophages, and it performs its action on angiogenetic process recruiting inflammatory cells, which in turn r elease pro angiogenic cytokines . It contributes to the stabilization of vessel wall regulating the proliferation of smooth muscle cells and pericytes 23 , 27 . Due to its context dep endent pro and anti angiogenic effect, it is considered a promising factor to realize the suitable mechanisms to stop the proliferation of epithelial cells at the early stages of oncogenesis or to block the stabilization of vessel walls 32 , 27 . Angiopoietin Angiopoietins and their receptor on EC (Tie2) represent another fundamental system involved in the morphogenesis and stabilization of interaction between endothelial cells and surrounding tissues. Fou r different paracrine growth factors (Ang1, Ang2, Ang3, Ang4) are the members of this growth factor family; even if they all bind to Tie2 receptor on EC, they can have an opposite eff ect and the result in angiogen ic development would depend on this balance 19 . Although they are not capable of inducing proliferation or tube formation in endothelial cells in vitro , these cytokines are involved in the control of many function s in vivo (microvascular permeability, vasodilation, and va soconstriction) and they are able to promote sprouting of endothelial cells 33 . Angiostatic factors Endogenous inhibitors of angiogenesis include various anti angiogenic peptides,
30 hormone metabolites, and apoptosis modulators, wh ich can be classified mainly in two big groups: matrix derived and non matrix derived . Among the matrix derived a ngiostatic factors the main ones are En dostatin and Thrombospondin 1. Endostatin is derived from collagen type XVIII and it performs its anti angiogenic action by interfering with the transduction of pro angiogenic signals (FGF,VEGF), by blocking endothelial cell motility, by inducing apoptosis and inhibiting cells proliferation. Thrombospondin 1 was the first one to be recognized as an inhibito r of angiogenesis. It acts on the angiogenetic process by regulating cell adhesion, proliferation and survival, transforming gr owth factor specific proteases 34 . Among the non m atrix derived, the main one is A ngiostatin, derived from a cleavage of plasminogen, which prevent neo vessels formation inhibiting endothelial cell proliferation and migration 35 . Angiogenesis in tissue engineering The research for an effective method to induce angiogenesis in engineered constructs nowadays represents the biggest challenge in the field of tissue engine ering. I nducing the formation of a stable capillary network would help to overcome the limitation in size and thickness of the construct, due to oxygen inability to diffuse deeper than 200 Âµ m. A better understanding of the molecular mechanisms involved in the angiogenetic process and recent advances in the technology of biomaterials for scaffold design and for sustained growth factors release, stimulated several approaches to induce successful angiogenesis in the transplanted grafts 22 . The predominant pro angiogenic effect of particular growth factors (especially VEGF)
31 suggested the approach based on their direct injection to induce revascularization in the engineered construct. The results obtained from therapeutic application in vivo , showed that the enhanced vascularization and collateral vessel formation was characterized by a disordered growth of blood vessels and it was followed by rapid degradation and an incomplete angiogenic process 22 . Since a successful angiogenic outcome could not be obtained with a single factor, these results suggested that a combination of them could improve the stability of the different angiogen ic steps, and this was investigated by injecting VEGF combined with other gro wth factor such as FGF or Ang1 36 , 37 . T he rapid clearance and degradation of the protein from the administration site encouraged the use of biomaterials or stable scaffold that enable slow and sustaine d growth factor release as a useful strategy to get to a functional vessel network 22 . Different groups have investigated the influence on ECs behavior of the chemical properties of specific materials used for scaffold preparatio n. The use of natural derived materials such as Matrigel or ECM mimic matrixes (collagen, laminin and other typical ECM component) showed an enhanced angiogenic response of the cells after the in vivo implantation 38 . Another ap proach focused on the modification of scaffold material by immobilizing directly the growth factors before processing the polymer/natural material, showing initial stages of microvascular formation in vitro 13 , 39 . Similarly, a really promising technique is based on the pre encapsulation of the growth factors in nanoparticles or microparticles, successively embedded in the 3D scaffold. The ability of the microspheres to slowly release the factors and to protect the bioactivity of the
32 proteins and the combination with specific materials as 3D matrix, allowed to elicit strong enhancement to vascular sprouting both in vitro and in vivo 40 . Since in the last decade numerous improvem ents have characterized the field of cell and gene therapies, their usefulness to induce angiogenesis in tissue engineered graft has been evaluated. The stimulation of vessel growth using gene transfer delivering a plasmid encoding VEGF to the cells incorp orated to the scaffold 41 , showed increased blood vessel density relative to control. Similar results were obtained both with viral and non viral vectors in promoting the tissue growth an d the remodeling of the vessel network, ov ercoming limitation imposed by the release kinetics of a controlled release method to deliver the required growth factors 42 . To overcome the limitation s due to lack of the different cell types involved in the angiogenic process in vitro , the use of stem cells demonstrated the possibility to give rise to endothelial cells, pericytes and vascular smooth muscle cells, seeding in the scaffold Endothelial Progenitors Cells, derivable from bone marrow, fat tissue or peripheral blood 43 , 44 . Promising results were finally shown combining a multiple growth factors releasing scaffold to st em cell therapy, with an induction of complete angiogenesis in vivo , with more efficient vessel st abilization 45 . The results obtained up to date enhance the need of further research regarding the specific molecular mechanism of the angiogenic process and the differences between ECs behavior in vitro and the in vivo environme nt. Main future directions point toward a combinatorial growth factors stimulation specifically after the construct implantation, an optimization of the ECM material (to get to in vitro environment able to
33 mimic the in vivo ones) and finally to the use of endothelial cells progenitors to guarantee the availability of all the cells types involved in the various angiogenic steps 7 . Angiogenesis quantification The choice of a suitable method to evaluate and quantify each step of the an giogenic process is required to assess the efficiency of any pro or anti angiogenic therapy, both for clinical or tissue engineering applications. Multiple different angiogenic assays have been developed and optimized to try to monitor and evaluate the beh avior of cells during the different steps of angiogenesis, to understand the specific role of each factor in inducing ECs organization, to assess the effect and the required quantity of novel pharmaceutical agents in drug testing. Despite notable improveme nts up to date, no assay responds to the requirements needed to completely characterize all the steps of the angiogenetic process in all its molecular and biological aspects. The ideal angiogenetic assay should be robust, cheap, time saving, reliable, easy to reproduce, include positive and negative control and be ready to report results useful for clinical applications 46 . It should be able to provide a quantification of the release rate, spatial and temporal distribution of the angiogenic factor designing a dose response curve. Moreover, it should provide a quantitative measure of the newly tubule like structure in terms of vascular length, number of branching point , number of tubules, areas and volume. Finally, all the results o btained in vitro should be confirmed in vivo , with a long term and possibly not invasive monitoring 47 . in vitro in vivo assays.
34 A common criterion to validate an in vitro angiogenic assay is the presence of a lumen within the capillary like structure in which endothelial cells organized when cultured. The way in which ECs organize leads to a classification of the in vitro assays between two dimensional assays, when c ells develop tubular structure on the surface of the substrate, or three dimensional, if the cells invade the ECM like structure 48 . More in detail, specific in vitro assays can be used to investigate different characteristics of endothelial cells behavior. ECs proliferation rate is usually evaluated using MMT or DNA quantification. Their migration through the use of particular transfilter assay, usually a modification of Boyden chamber, to visualize the pseudopodia extension and d iscriminate between chemotaxis (directional migration) and chemokinesis (random motility). Finally, ECs differentiation forming a capillary like network is evaluated through the use of 3D gel structure (collagen, Matrigel) in which the cells develop tube f orming tight junctions. Since ECs are not the only cell type involved in the angiogenic process, some in vitro assay aims to study the interaction between them and other cell types (mural cells) during the steps of angiogenesis. Even if in vitro assays are an easy and cost effective method to gain a good molecular understanding of all the angiogenic steps and to provide an early validation for screening purpose or for a drug testing, their results have to be considered carefully because of the great gap exi sting between the in vitro and the in vivo conditions 47 . First of all, ECs cultured in vitro are not in the same quiescent condition as in vivo ; secondarily after proliferation in vitro , ECs tend to lose their phenotype; final ly their behavior is strongly affected by the environment and by the interaction with other cell types (SMC, perycites), conditions which cannot be reproduced in vitro .
35 A more realistic understanding of the angiogenic process and quantification of the neov ascularization can be obtained using in vivo assays, which mainly can be prepared following three approaches: microcirculatory preparation in animals, usually in a chamber created in the ear, back or cheek of rodents; vascularization of an implanted bio compatible polymer matrix; excision of vascularized tissues from animal or human 47 . Though the in vivo approaches resemble more the physiological conditions, they are expensive, time consuming and several difficulties have be en encountered to reliably assess the temporal and spatial response of ECs and to evaluate the dose response curves. To conclude, seen pros and cons of both the approaches, a complete interpretation of the efficacy of any therapeutic strategy on angiogenes is induction or inhibition can be achieved only with a combination of multiple in vitro assay, which results should then be tested in vivo , to provide more realistic results necessary for the translation into clinic 46 . Perinatal T issues One of the most vascula rized organ in the body is the Human Placenta. Following a brief overview of the main properties of perinatal tissues with an insight on human placenta and on the properties that make it an attractive material for tissue en gineering application s . Perinatal Tissues properties One of the main issues related to the improvement of tissue engineering is to find a safe, easily available and appropriate source of cells and scaffold materials able to
36 mimic the physiological environm ent to promote the same processes then in vivo . I n the past years, uses of both embryonic and adult stem c ells have been investigated, but several disadvantages for both of them arose, related essentially to the ethical issues for embryonic and limit in di fferentiation potential for the adult ones . Perinatal materials represented a promising alternative source of stem cells and of scaffolds t o overcome these limits : umbilical cord, amniotic and chorionic membrane, amniotic fluid and placenta villi represent cell and scaffold sources which have recently come under close scrutiny for clinical and tissue engineering applications 49 . The easy and unlimited availability of this sources and the minimal ethical and legal issues associated with their use , represent other important advantages which make t hese sources so interesting . Moreover, they are characterized by some impressive properties relevant for clinical and tissue engineering applications : anti inflammatory and anti microbial pr operties, low antigenicity, richness in extracellular matrix, interesting mechanical properties and pro angiogenic potential. Regarding their anti inflammatory properties, since inflammation is a central consequence of every kind of injury and devices imp lant , often resulting in permanent scarring and fibrosis, a reduction in inflammation by stem cell therapy would be an important goal to repair injured tissues and to allow the integration of engineered grafts 50 51 . The lack or very low expression of antigens on the surfaces of cells isolated from these tissues could help to avoid the use of immunosuppressant drugs , since an immunogenic reaction would be prevented 52 . Moreo ver, the mech anical characteristics and the richness in extracellular matrix and basement membrane make these tissues an ideal source to p rovide support for cells cultures and to realize innovative technique for
37 tissue engineering applicatio n, overcoming t he main limitations of field like small diameters vessels engineering 53 54 . Pro angiogenic properties of these tissues, especially of the villous placenta, make them promising to be applied to realize re liable angiogenic assay and for tissue engineering applications 55 . Human Placenta : anatomy and main properties H uman placenta, literally meaning s the main organ of interface between the mother and the fetus, providing nutrients and oxygen during the embryonic development. The first and most comp rehensive definition was given in 1937 fetal tissue for the purposes of phy siological Figure 2 2 : Gross structure and function of human Placenta (Source:http://www.walgreens.com/marketing/library/graphics//images/en/170 10.jpg) The main functions of this organ are therefore to maximiz e oxygen and nutrient acquisition from the mother, (working as lung, digestive, excretory and endocrine
38 system for the fetus) but at the same time to minimize immunological rejection by the maternal immune system 56 . A mature placenta is composed of a fetal and a maternal side. The fetal side presents an outer membrane, called amnion , which cover s an inner surface, called chorion , in which it is inserted the umbilical cord , containing two arteries and one vein embedded in aternal side, usually called decidua basalis, contained around 15 30 groped lobules called cotyledons, each one consisting of a main stem of a chorionic villus as well as its branches and sub branches. Chorionic villi main role is to maximize the surface u seful for the nutrient and oxygen exchange between the mother and the fetus and they represent the main functional unit of the human placenta as a vascular organ. Chorionic villi receive blood supplies from both mother and fetus circulatory systems, which remain separated. Figure 2 1 : Detailed anatomy of human Placenta (Source: http://embryology.med.unsw.edu.au/embryology/index.php?title=File: Gray0039.jpg )
39 Feto placental circulation is characterized by umbilical cord vein and arteri es, which branch into cotyledon vessels and capillaries inside the cotyledons villi. The utero placental circulation starts with the maternal blood flow into the intervillous space through decidual spiral arteries. Since cotyledons are surround ed by matern al blood , e xchange of oxygen and nutrients take place in the intervillous space , as the oxygenated maternal blood flows around terminal villi where deoxygenated fetal blood flows in the cotyledons capillaries. The development of a normal fetus strongly de pend upon a suitable placental perfusion, therefore the development of the placental vascular network is a tightly controlled vasculogenic and angiogenic process as gestation progresses. A complex in terplay between physical and chemical factors including o xygen, growth factors and growth inhibitors produced by placental cells (trophoblasts, Hofbauer cells, pericytes, and endothelial cells) play an important role in the control of vasculogenesis and angiogenesis 57 . The main factor s involved in these processes are the following: basic Fibroblast Growth Factor (bFGF) involved in the recruitment of haemangiogenic progenitor cells; Vascular Endothelial Growth F actor A (VEGF A or VEGF), responsible for the commitmen t, growth and aggreg ation of EC precursors for the formation of the haemangiogenic cords, highly expressed in early pregnancy; Placenta growth factor (PlGF) involved toward s the end of pregnancy, when branching angiogenesis is replaced by non branching angiogenic mechanism . A ngiopoietin 1 and 2, whose balance and interaction with receptor Tie 2 control the stability of the outer part of vessel walls 58 . In addition to the previous traditional growth factor, multiple st udies investigated the role in th e placental vascular development of another protein, Angiogenin . This potent angiogenetic stimulator is known to induce angiogenesis by activating vessel
40 endothelial and smooth muscle cells and triggering a number of biological processes, including cell mi gration, invasion, proliferation, and formation of tubular structures 59 . V illous cytotrophoblasts, one of the main placental cell types , have been shown to express and secrete angiogenin. T he biological activity recorded in vitr o , after cytotrophoblast isolation and culture , suggested that an giogenin in the human placenta is involved not only in the angiogen ic process but also in vascular and tissue homeostasis, immunogenic camouflage of the fetus, and host defenses 60 , 61 . The high number of cytokine and growth factors, together with its anti inflammatory, low immunogenicity, pro angiogenic properties and richness in ECM, make human placenta an attractive material to mimic t he in vivo environment and realize reliable in vitro angiogenic assays. Characterization of human Placental Matrix (hPM) A strong interest towards h uman placenta as a rich source of biologically active components with impress ive properties has been shown s tarting from the beginning of the 20th century when the preparation of a first curative placenta extract was realized 62 . From that moment on, various extract s have been derived from human placenta, both for c linical and research purposes. For clinical applications, various placenta derived extracts have shown positive results in term of anti inflammatory and analgesic effect, encouraging possible application for drugs or in surgical applications 50 . Othe r uses were investigated in combination with a cell therapy or alone to improve the current technique regarding reconstructive surgery 63 and wound healing 64 , 65 , where advers e effects such as inflammation represent the main challenge. Another area regard s finally the promotion of cells differentiation and proliferation for regenerative medicine application 66 , 67 , with a pr omising application for liver regeneration. Differently from the
41 clinical perspective, for research purposes human placenta has mainly been used as a source for in vitro synthe sis of several growth factors 68 , 69 in order to investigate their function, and gain a better understanding of the molecular mechanisms involved in particular diseases, tumor growth and formation and tissue regeneration. Up to date, even if pro angiogenic properties of all the perin atal tissues have been recognized and the growth factor expression in placenta investigated, no other laboratory has purposed the extraction from human placenta of a pro angiogenic protein mixture useful to induce angiogenesis in the transplanted graft in tissue engineering application. Qualitatively the mixture, named human Placental Matrix (hPM), is a tra nslucent and viscous compound, rich of pro teins and growth factors. It is obtained from full term human placentas following the protocol described in C hapter 3 . The characterization of hPM , including broad spec trophotometric analysis, total protein content w ith Sodium Dodecyl Sulfate P olyacrylamide gel electrophoresis ( SDS page ) , cytokine and gene analysis, has been performed in our laboratory . Either a bsorption spectroscopy or total protein co ntent analysis confirmed the high protein content of hPM. S pectra obtained from spectroscopic analysis showed a significant absorba nce peak at 280 nm, wavelength to which the proteins in solution, mainly constitute d by aromatic amino acids s uch as tryptophan and tyrosine 70 , are most commonly measured. Total protein content analysis revealed the protein conce ntration mL . Data obtained from cytokine array analysis detect ed a total of 54 cytokines in hPM. The highest chemilumi nescent intensity is shown by angiogenin, follow ed by Acrp30Ag, IGFBP 1, NAP 2 and Fas/TNFGSF6 which are all known to be
42 pro angiogenic cytokines. Gene a nalysis of HUVECs seeded with hPM showed up regulation of many angiogenic genes (including VEGF A) in comparison to HUVECS seeded onto cell culture plates. Biomaterials for Tissue Engineering A pplications One of the main areas of interest for tissue engineering application focuses on the use of smart and modified biomaterials to improve the delivery of useful growth factors or to mimic physiological extracellular matrix (ECM) characteristics. Encapsulation techniques to ac hieve a sustained and modulated release of bioactive proteins have been deeply evaluated in the past century, and the main parameters which could affect the stability of the protein and its release rate delineated 71 72 73 . The realization of smart controlled delivery system, with the use of particular processes of encapsulation and with remotely activated s mart materials, represent s one of the most prom ising research area, aiming to customized release profile s 74 75 . The development of matrices able to mimic physiological ECM has become one of the most active areas of tissue engineering. Synthetic and natural materials has be en evaluated and several chemical mod ification investigated , trying to induce particular behavior in cells thanks to specific cell matrix interaction in the scaffold 76 . Protein c ontrolled release methods A large number of reco mbinant proteins have been investigated for t herapeutic applications in the last decade. A common problem is represented by their instability, caused by their short half lives. multiple , which makes necessary multiple inoculations to obtain t he desired ther apeutic effects. Methods for controlled release, distinguished among microparticles , microspheres, nanoparticles, represent the so lution to improve the therapeut ic effect of proteins and drugs in several applications 71 . Several c riteria
43 can be used to classify and characterize controlled release methods. The size of particles rep resents the first criterion which allow to distinguish microparticles (from 1 to 100 0 um) from nanoparticles (1 to 100 0 nm). Figure 2 2 : Schematic representation of distinct uses of particles in the TE field according to size : (A) Nanoparticles can be used for release of bioactive agents to the ce ll culture medium or for cell internalization (B) Incorporation of microparticles in 3D systems for enhancement of the matrix properties. (C) Use of microparticles for the delivery of bioactive agents to control cell behavior. (D) Hollow capsules obtained by LbL which, after core leaching, may contain encapsulated bioactive agents for c ontrolled release or liquefied medium with cells. (E) Microparticles in the range of 100 1,000 Âµm can be used: (a) in combination with cells to obtain cell induced aggregation; (b) to allow the formation of scaffolds with interconnected porosity after part icle agglomeration by sintering or solubilization methods; (c) as cell microcarriers for cell expansion. (F) Use of hydrogel particles with encapsulated cells for organ printing and mesoscaleself assembly. (G) Fabrication of 3D porous constructs obtained b y a LbL strategy. (source 77 ) Particle size represents a parameters that affect strongly the protein release ra te and the initial burst . Usually smaller particles dissolve faster than larger ones. Moreover, it has been obs erved t hat initial release burst decrease with increasing microparticles
44 diameter, likely due to the increase d surface to volume ratio of the smaller microparticles. Therefore a combination of different size of particles could make possible to obtain various cust omized release rates 78 . For tissue engineering application, both nanoparticles and microparticles have be en investigated, alone or embedded in specific 3D matrices. For the specific aims of single or combined release of growth fa ctors, microparticles of various range from 5 to 100 um 79 11 80 71 have been investigated. Another criteria used to classify micro/nanoparticles is the method of preparation. Several consideration s about the manufacturing and the modification of microparticles useful for tissue engineering and regenerative med icine application are required. A control over particle size, shape, surface characterist ics, and porosity should be provided. Moreover, t he method should also ideally allow the production of large quantities of particles with a narrow size distribution. Finally the technique has to maintain the activity of the protein while microparticles are processed. The most c ommonly used techniques are 77 81 : HOT MELT EXTRUSION ( HME ) . It i nvolves the compaction and conversion of blends from a powder or a granular mix into a product of uniform shape; el evated temperatures required. SPRAY DRYING . It t medium; high temperatures required. ELECTRO SPRAYING . I t represents a slightly modified form of th e electrospinning process, and applies an electric field to a polymeric solution extruded from a syringe. GELATION . It use s a polymeric solution, extruded and dropped in a hardening bath containing a slowly stirred solution responsible for the crosslinking of the polymer. Technique compatible with encapsulation of cells or delicate molecules and with the use of different combinations of polymers in the initial liquid formulation, including stimuli -
45 responsive macromolecules allows particular designs, including layer by layer (LbL) methodology, permitting to produce liquid core shells for cell encapsulation 82 . SUPERHYDROPHOBIC SUR FACES . This meth od is used to prepare microparticles depositing drops of liquid precursors containing the polymer and other sub stances onto the surface, rolling of water drops over superhydrophobic surfaces. COACERVATION . T he solubili ty of the polymeric solution is decreased by the introduction of a contrasting component. Two distinct phases are obtained: one containing the coacer vate phase and other containing the supernatant, allowing the encapsulation of both hydrophilic and hydrophobic drugs. S OLVENT REMOVAL OR EM ULSIFICATION . It is also k nown as double emulsion method, including two different water in oil in water (W/O/W) and oil in wate r in oil (O/W/O) emulsions , which allows microparticles to be formed by removal of the organic solvent from the polymer phase . Among all the methods described, W/O/W double emulsion method represents the most widely used because of its relatively simple and mild process, convenience in controlling process parameters, and ability to produce with inexpen sive instrument. It consists of dispersed oil globules containing smaller aqueous droplets; it is obtained firstly homogenizing an aqueous protein solution in an organic solution or in an oil and then dispersing this primary W/O emulsion into a large volum e of water containing an emulsifier to form a double emulsion. The final formation of microparticles is obtained by removal of the organic solvent from the polymer phase by solvent extraction or solvent evaporation 83 84 . The mechanism of protein release represents another important criterion to classify controlled release methods . Both hydr ophilicity and biodegradability are properties based on which release mechanism of entrapped p roteins and drugs can change. Three main mechanism s can be observed 85 :
46 1) Diffusion: hydrophilic matrices release the contained drug b y diffusion phenomena due to the swelling of the polymer upon contact with fluids. 2) Biodegradability: for biodegradable matrices the release of the drug is controlled by the rate of degradation in the physiological environment, mainly because of hydrolysis reactions. 3) Enhanced controlled release: external stimuli (such as magnetic field, change of temperature, of pH) can be given to modify matrices properties and enhance the drug release. Figure 2 3 : Mechanisms for protein release (Source: 85 ) The material used to prepare particles represent s another important element which strongly characterize s microparticles features 86 . M ost commonly used natural materials are gelatin, collagen, alginate, chitosan 87 88 . Despite they prese nt obvious advantages related with the biocomp atibility with human body, it is difficult to control their protein release over a period of weeks or months. Because of their chemical structure similar to human body constituents , the degradation rate of natu ral material based microparticles appears to be fast (around one week) 88 , 87 , 89 , preventing the protein release to be sustained for a long period without use of c hemical or photo chemical crosslin king to slow down the degradation of the microparticles. 88 , 90 . Despite
47 positive results showed using glutaraldehyde or UV crosslinking, all these methods present the risk to a ffect the bioactivity of the proteins. Biocompatib le synthetic materials represent a widely investigated alternative to natural materials. PLA copolymers represent biocompatible and FDA approved synthetic materials normally used for protein encapsulation f or biomedical application 71 , 91 , 92 . These polymers provide a long lasting controlled release of proteins, thanks to their chemical structure and to the possibility to realize composites and multi layered microparticles 15, 92 , 93 .Among those , PLGA or poly (lactic co glycolic acid) is one of the most used to achieve the controlled release of specific growth factors (for examp le BMP, VEGF, bFGF) 71 , 72 , 11 , 10 , 12 for tissue engineering application s . Despite the fact that since t he early 90s P LGA microparticles have been largely explored and their practical importance assessed, the mechanisms related with physical and chemical mass transport phenomena involved with the control of p rotein release have not been fully understood. One of the main reasons is related to the complexity and the different variables involved in the mass transport mechanisms. Once in contact with aqueous media, protein release is due to the hydrolytical cleavage of the ester bond caused by the penetration of water in the microparticle structure. A complex array of elements is involved in the achievement of a sustained and complete release of the protein in its native form from PLGA microparticles 94 . The most important factors are the pH and che mical composition of the release medium, the different methods used for protein sampling, the physico chemical properties of the encapsula ted proteins and their reciprocal interaction, the geometry and size of the particle, the charge of the polymer and it s interaction with encapsulated proteins.
48 Several studies have been performed to investigate the main parameters influencing proteins release from PLGA. Changes in PLGA microparticles sizes were shown to influence mainly the initial burst release, increa sing proportionally to the decrease in size, but not signific antly the release rate trend 72 , 95 . Further evaluation s made to assess the stability of the protein encapsulated showed that there was no ap parent effect of the microparticles preparation on the stru ctural integrity of the protein. However, after one week of incubation the protein released, it started to hydrolyze to smaller fragments, probably due to a decreased pH for the accumulation of deg raded PLGA. 73 Although encapsulation of a single protein using PLGA microparticles has been widely evaluated and assessed in the past literature, currently a great effort is involved in the study of the parameters affecting enc apsulation and release of multiple proteins, especially to achieve an effective combined growth factors delivery for tissue engineering application 96 . Several studies evaluated how to realize polymeric systems for dual growth fa ctors release, embedding microparticles in a polymeric matrix or realizing microparticles with multiple shells 13 . A few studies investigated the possibility to encapsulate two different protein in PLGA microparticles 14 15 , as far as we know up to date none has investigated the possibility to encapsulate a complex mixture of different pro teins using PLGA microparticles. Further evaluations to achieve an effective release of com plex mixtures need to be performed, since differences in molecular weight, charge, properties of the proteins contained in the heterogeneous mixture could interact with each ot her and with the polymer itself.
49 Biomaterials for 3D cell cultures and angiogeni c assay A dvances in cell and tissue engineering and in the understanding of molec ular mechanisms related to cell behavior and cell response to drugs, stimulated the research to find a suitable way to test growth factors and protein delivery on ce lls cultur es . A great development therefore had recently characterized the field of biomaterials for 3D cell cultures to mimic the extracellular matrix properties. One of the most challenging areas of tissue engineering research is r elated to the development of new therapeutic approaches that aim to help the body exert in natural mechanisms for vascularized tissue growth. For the specific application of efficient angiogenic assay, to evaluate the behavior of e ndothelial cells when pro angiogenic growth factor s alone or in combination are delivered during time, the most investigated materials have been collagen, laminin, fibrin, alginate, hyaluronan and chitosan for the natural materials, and PLGA and PEG hydrogel for the synthetic materials 76 . Collagen represents one of the most investigated materials for 3D cultures realization. It is one of the most abundant components of extracellular matrix. It is an insoluble f ibrous protein that plays an important role in angiogenesis, inducing EC t o organize in a capillary like network 97 . Like fibrin and other proteins, collag en type I matrices can be used to create a high ly porous network , with favorable properties for cell adhesion and migration 76 . More recent and advanced studies are trying to mimic ECM physiological composition combining col lagen with other proteins , l ike fibrin or fibronectin, to improve EC organiza tion in microcapillary networks 98 .
50 Alginate , thanks to its excellent biocompatibility and biodegradability, is another material extensively used for different application in tissue engineering as hydrogel synthetic ECM and as growth factors controlled deliver method 99 , 100 It is an anionic polysaccharide which forms a hydrocolloid gel in presence of divalent cations. Opinion s about its properties as a bulk material are mixed: from one side stimulation of inflammatory cells 76 and limited ability to interact with mammalian cells 99 has been registered, but from the other side the mild gelation conditions and the almost independency from temperature make it a good candidate for growth fact ors slow release and cell entrapment. The possibility to be easily processed and modified to produce 3D scaffolding materials with specific tunable properties 100 represents a further important advantage . Several studies hav e been conducted to investigate the properties of Alginate as synthetic ECM both in the specific application of on vitro 3D angiogenetic assay and in vivo as scaffold for cells embedding. In vivo studies with implantation of growth factor loaded alginate c onstructs showed an increase in capillary density but without vascular stabilization 101 . Investigat i ons of angiogenesis molecular mechanisms, suggested that the d elivery of multiple factors might represent the solution to th is instability. On another side , cell delivery approaches focus on stimulating vascularization either via cell release of soluble factors, cell proliferation and incorporation into new vessels or alginate pre vascularization prior to implantation.
51 CHAPT ER 3 GENERAL MATERIALS AND METHODS Experimental M ethods Isolation of Vascular Endothelial Cells from the Human Umbilical Vein Different techniques have been optimized to isolate v ascular endothelial cells from hum an umbilical vein (HUVECs). The following p aragraph describes the procedure adopted in this study, adapted from the method reported by Jaffe 102 . Materials : Metal luer adapters (autoclaved), 1 scalpel blade / scalpel handle (autoclaved), 1 scalpel blade / scalpel hand le (not sterile), metal ring stand (ethanol) and clamp (autoclaved), several bags of paper towel (autoclaved), plastic zip ties / zip gun (autoclave ties; ethanol zip gun), pipette gun and 10 mL / 25 mL pipettes, 60 mL syringes (1 per cord and additional 2 0 mL syringes for filter sterilization), Costar 0.22 mL collagenase), 50 mL Falcon tubes, T75 flask (1 per cord), The reagents needed were prepared depending on the number of cords we were using. Per each cord we used 50 mL of sterile PBS (autoclaved), 1 mg/ mL filter sterilized bovine collagenase (Gibco, Invitrogen, NY, USA) in 15 mL of sterile PBS, 15 mL + 8 mL of fully prepared VascuLife VEGF media (VascuLife VEGF Medium Complete Kit, Lifeline, MD, USA); Methods: An appropriate volume (15 mL per umbilical cord) of 1 mg/ mL bovine collagenase in sterile PBS was prepared and filte r sterilize in hood using 0.22 Âµm filter and sterile 60 mL syringe. The collagenase solution was preheated to 37Â°C in a warm water bath 10 15 minutes prior to usage to maintain its activity. The workbench was prepared under a laminar flow hood by laying out paper towels on the metal tray, spraying them with plenty of with ethanol and taping them to its
52 edges. After that the metal ring stand, (cleaned before with dish soap and then with ethanol), all the autoclaved tools and reagents were placed under hood (paying attention to spray everything with ethanol to avoid contamination) Full term human placenta was collected from the delivery suite at Shands Hospital (Gaines ville, FL, USA) up to 24 48 hours after delivery. Umbilical cord was cut from the placenta directly at the base. Depending on the cords length, in some cases they were cut into multiple lengths to avoid blood clots. To avoid any damage to the cells, we pa id attention not to massage or to rinse the cord, but to simply clean it off using ethanol soaked paper towel before placing it under the hood. After cutting them from the placenta, umbilical cords were laid onto the tray and clear cuts at about 3 cm from cord edges were made using a scalpel, to prevent any bacterial contamination. For each cord, after choosing the luer adapter size depending on the cord dimensions, it was plug into a syringe, then inserted into one end of the cord vein and secured tightly with a zip tie to prevent any leakage. T he umbilical cord was then draped over ring stand to have the open end above the waste be aker; 50 mL of sterile, warm (preheated) PBS were poured directly into syringe and then gently injected through the cord in o rd er to remove all the blood. After washing the cord, plunger was removed from the syringe and the cord open end was tie off with a zip tie. Warm collagenase was obtained from the water bath, and holding the syringe so that the cord is suspended in the air, it was poured until able to fill the vein and a bit extra, paying attention to massage out air bubbles. After changing the plunger, the collag enase solution was injected slowly into the cord to pressurize the vein. Cords containing collagenase enzyme wer e allowed to incubate under the hood
53 for about 25 minutes. After cords were incubated, the end of the cord was pinched off just above zip tie and the suspension inside the cord was allowed to drain into a sterile 50 mL Falcon tube, and an equal volume of c ulture media was added to inactivate the collagenase. The cells suspension was centrifuged (Allegra X 12R Centrifuge, Beckman Coulter, Inc., CA, USA) for 5 minutes at 1000 rpm, the supernatant removed, the pellet re suspended in 8 mL Vasculife media and pi petted into a T75 flask for culture. Cells were left to attach to the flask for at least 2 hours or overnight at 37Â°C in a humidified 6% CO2 incubator, and then they were washed with sterile, warm (37Â°C) PBS to remove non adherent cells and debris. Fresh , warm (37Â°C) media was added to the flask and this latter was changed every two days. Cell growth was evaluated using an optical microscope every time the media was changed, and when the cells were passaged as they reached confluence using A c cutase enzyme to detach them from the bottom of the flask. Cell passage number between 1 and 3 were used for every experiment included in this study. F reezing of Vascular Endothelial Cells from the Human Umbilical Vein As soon as the cells reached passage 3, they were frozen to keep them ready to be used in following experiments. All the media was removed from the flask, and 5 mL of fresh media was added and removed after 5 minutes. 5 mL of ac c utase was added to the flask and left incubating under the hood for 10 or more minutes. Ac c utase with cells was then removed from the flask and centrifuged for 10minutes at 1000 rpm. The supernatant was removed and the pellet was suspended with a suitable volume of Endothelial Cells Media (1.75 mL per each flask) . 1.75 mL of media w/ce lls was pipetted using a micropipette into the cryogenic vials and 10% (of volume) of DMSO (Dimethyl
54 Sulfoxide, Fisher Scientifc Inc., USA) was added into each vial. The vials were closed and label ed with the concentration of cells / mL , after counting them with a haemocytometer. Finally they were progressively frozen at a rate of 1Â°C/min to 86Â°C and then stored in liquid nitrogen. De rivation of human Placental Matrix (hPM) The human Placental Matrix (hPM) is a mixture of proteins and growth factors obtain ed from a f ull term human placenta. The derivation method used in this study was developed in our laboratory 9 and it is adapted from Matrigel Protocol 103 . Full term human placentas were collected f rom the delivery suite at Shands Hospital (Gainesville, FL, USA) within 12 hours after birth. Umbilical cord, chorionic and amniotic membranes were removed and the remaining placenta was dissected into cubes of 2x2 cm and frozen to 86Â°C. The chopped tissu e was weighted and separated in several plastic bags each one containing 200g of tissue, each one stored at 86 Â° C until needed for further preparation. The described process is suitable per 200g of chopped placenta. Twelve hours after freezing, the dissec ted placental cubes were transported to a cold room and maintained at 4Â°C where the rest of the placental extraction process was completed. Using a blender, placenta chunks were minced in cold 3.4 M NaCl buffer (198.5 g of NaCl, 12.5 mL of 2M tris, 1.5 g o f EDTA, and 0.25 g of NEM in 1 liter of distilled water). This NaCl buffer/ placenta mix was then homogenized into a paste using a Tissuetek Homogenizer (Homogenizer Model No SDT, Serial No 99025, Tekm ar Company, OH, USA) . 250 mL plastic bottles suitable fo r ultracentrifuge were filled with the obtained compound, balanced and spun at 7000 rpm for 15 min using an ultracentrifuge (Sorvall RC6+ Centrifuge, Thermo Scientific, NC, USA). The supernatant
55 was discarded; the placental extract was homogenized in 150 m L of new cold 3.4 M NaCl buffer and the pa ste was newly centrifuged at 7000 RPM. This process was repeated 2 more times, until the p ellets were free of blood, cell and serum derived proteins. Placenta pellets were homogenized in 100 mL of cold urea buffer . Accordingly to the protocol optimized in our laboratory, 4M urea buffer (240 g of urea, 6 g of tris base, and 9 g of NaCl in 1 liter of dist illed water) was used to prepare the pro angiogenic mixture . The proteins were extracted and solubilized by stirri ng continuously the homogenized placenta pellets on a magnetic stir plate for 24 hours at 4Â°C. The urea placenta mixture was then poured in distinct plastic bottles suitable for ultracentrifugation; these latter were balanced and spun at 14000 rpm for 20 min using an ultracentrifuge (Sorvall RC6+ Centrifuge, Thermo Scientific, NC, USA). The pellet was discarded an d the supernatant was dialyzed using 8000 MW dialysis tubing (MWCO 8,000; Spectrum Laboratories, Inc., CA, USA). These latter were placed in 1 l iter glass cylinder filled with 1 liter of cold TBS (6 g of tris base and 9 g of Nacl in 1 liter of distilled water) and 2.5 mL of chloroform for sterilization. After at least 2 hours dialysis, the TBS solution was discarded and replaced with 1 l fresh TBS . TBS buffer was changed 4 more times, at 2 hours intervals, to be sure that small undesired molecules, urea, and chloroform we re remove d from the extracts. Finally, under a laminar flow hood, dialysis tubes (sterile inside) were opened and the viscous content collected into sterile 50 mL Falcon and centrifuged (Allegra X 12R Centrifuge, Beckman Coulter, Inc., CA, USA) at 4000 rpm for 15 minutes to remove polymerized proteins. After centrifugation, the supernatants were collected and split up in aliquots of 2 mL into sterile 15 mL Falcon tubes . The final biomaterial, a pink viscous
56 extract, was stored at 86Â°C rea dy for use. Analytical M ethods Calcein AM staining Calcein AM is a non fluorescent cell permeant dye that can be used to determine cell viabil ity in most eukaryotic cells. In live cells the non fluorescent calcein AM is converted to green fluorescent calcein, after acetoxymethyl ester hydrolysis by intracellular esterases. A Live/Dead Assay (Invitrogen Life Technologies, NY, USA) Calcein AM bas ed was used to stain HUVECs . Calcein AM was dilute d using EC media to a final concentration of 2 Âµg/ mL and then pipetted in each well of the multiwells plates were HUVECs were cultured for each specific experiments. Plates were incubated for 30 minutes in a humidified 6% CO 2 incubator at 37 Â°C and then observed using the fluorescence microscope. Angiogenesis quantification Angiog enesis quantification was performed using an inverted fluorescence microscope (Zeiss Axiovert 200 Inverted Fluorescence Microscop e) after cells Calcein AM staining has been performed as previously described . After cell staining, selected fields of view wer e photographed for each sample using a color digital camera attached to the fluorescence microscope, using the suitable fluoresc ence filter and with a 5x magnification (field of view ) . The acquired image s were saved as TIFF files and analyzed using the free software ImageJ 1.45s (Wayne, Ra sband National Institutes of Health, USA http://imagej.nih.gov/ij/ ). Several semiquantitat ive and quantitative methods have been used in previous studies to evaluate outcomes from in vitro angiogenic assays 104 , 105 , 106 . In this study b oth topological (n umber of meshes and number of branching points ) and
57 morphological (mean tubule length) parameters (see Figure 3 1) have been used since they allow the characterization of t he spatial organization of the ECs in the capillary like network. The number of me shes, identified by avascular zones surrounded b y hexagonally arranged vessels 107 , represents an important indicator to discriminate a mature capillary network from an immature or degraded one. A mature capillary network sho uld present a lower number of wide meshes, compared to an immature capillary like structure where meshes appear to be more but smaller or less and incomplete due to the degradation of surrounding tubules. After the TIFF image was opened, the number of mesh es was manually counted in each image for a triplicate number of samples for each condition. The result was then normalized for the area of the field of view to get to a result express in term of meshes/mm 2 . The number of branching points, defined as nodes where branches meet or from where tubules sprout, represent another useful parameter to quantify cell response to an angiogenic stimulus. Usually the initial stages of angiogenesis are characterized by a high number of branching points, due to the develop ment of sprouts. When the network stabilizes and becomes more mature, the number of branching points is reduced 16 . Branching points were manually highlighted an d counted using the multi point selection tool available in the Ima geJ software. The n umber of branching points was counted in each image for a triplicate number of samples for each condition. The result was then normalized for the area of the field of view to get to a result express in term of BPs/mm 2 . Length of the tubu lar structure represents a significant parameter to characterize the maturity of an angiogenic network. In a mature capillary network, tubules appear to
58 be longer then in an immature or degraded network. After opening the image, the scale was set (Analyze > Set Scale) by referring to original image . Tubule length was assessed by dra wing a line along each tubule and the measure of that line was automatically calculated by the software (A nalyze > Measure) and written ou Since branching points can be positioned along the tubule, the width of the tubule was taken into account. Each tubule was considered as continuous until a morphological change in tubule width was caused by a real branch. Figure 3 1 . Angiogenesis quantification : the image showed the element considered in the analysis of images to quantify angiogenesis output from angiogenic assay. A,B,C indicate meshes, D,E branching points, the arrow point a tubule like structure. Statistics Expe riments were performed in triplicate. All graphical and tabulated data were displa yed as mean Â± mean standard error . Data analysis was performed using Excels (Microsoft Office) and the free software Minitab 15 . Significance tests were calculated using ANO VA tests were more than two conditions were evaluated, and then the specific differences evaluated using post hoc
59 tests . When only two conditions were compared unpaired, two Test with unequal variance were used . Significance levels were set at * p < 0.05.
60 CHAPTER 4 INDUC TION OF ANGIOGENESIS IN HUVECs 2 D CULTURE USING HUMAN PLACENTAL MATRIX Background and Rational As discussed in the background section, one of the main challenges of tissue engineering research is to find a suitable metho d to induce angiogenesis in engineered tissues, to provide cells with nutrients and oxygen. All the described approaches evaluated to promote in vitro angiogenesis were based on the attempt to create an environment surrounding cells able to mimic ECM propr ieties . However, to date no one of them succeeded to be translated into clinic, especially because of the use of non human compounds. One of the clearest example comes from Matrigel, (named GeltrexTM BMM by Invitrogen I nc) a complex protein mixture derived from Engelbreth Holm Swarm mouse sarcoma . Due to its derivation, it could not be translated into clinic, even if results in term of ECM mimic and of angiogenesis induction were satisfactory 8 . The derivation of an angiogenic extra ct using human based materials represents basic prerequisite to allow any clinical applications. growth factors, cytokines and ECM components has attracted attention from the research as a promising material for angiogenic assays. In light of these reasons, the rational for the idea developed in our laboratory was built: the derivation of a pro angiogenic extract, the human Placental Matrix (hPM) from the human placenta. As described in the introduction, previous in vitro a nd in vivo studies performed in our lab showed then e ven if the extract is able to induce and modulate the initial stages of angiogenesis, after 5 days the newly formed capi llary network starts degrading . The lack of stability of newly formed capillary net work and the research of suitable materials to achieve a long lasting capillary network formation is a common challenge for all the
61 approaches using growth factors delivery to induce angiogenesis. To improve these results w e suggested that a controlled de l ivery of hPM over time could be useful to create a more stable and long lasting capillary network. The first aim of this study was therefore to validate this hypothesis investigating the response of HUVECs to hPM. First, an evaluation of the correlation b etween hPM effect on HUVECs and its protein content was assessed. After that, a comparison of HUVEC s behavior when hPM was added at a single or at multiple time points was performed. Finally, a comparison between angiogenic response of HUVECs when seeded o n hPM coating and when hPM was added in solution with a ngiogenic media was evaluated. This last experiment aimed to bring preliminary results to bett er evaluate the response expected in the 3D culture, since one of the main differences compared to angiogen esis induced in 2D culture is represented by the different method with which cells receive hPM. Materials and M ethods Materials HUVECs, pre isolated from human umbilical veins and pre cultured as previously described in General Material and Methods , wer e detached from the flask using A c cutase and centrifuged at 1000 rpm for 5 minutes. The cell pellet was resuspended in Angiogenic Media . This last was prepared adding 25 mL of glutamine, 0.5 mL of hydrocortisone, 0.5 mL of ascorbic acid, 10 mL of FBS, 1.25 of VEGF, and 1.25 o f bFGF to 500 mL of VascuLife Basal Media. HUVECs were counted using a hemo cytometer to be ready for the seeding. hPM was derived from human placentas following the protocol described in Chapter 3 and it was stored at 86 Â°C read y for use.
62 Assessment of the correlation between hPM pro angiogenic effect during time and its protein content . In light of the results obtained from the cytochine analysis, pro angiogenic effect of hPM is expected to be due to the different proteins conte nt. As described in the literature, proteins function depends upon their tertiary structure and it can be affected by protein denaturation. One common procedure followed to denature proteins (for example for SDS page analysis) is to heat them. When heated at 90 100 Â°C for 5 minutes, proteins can be completely denatured and linearized. For our specific aim, to confirm a direct correlation between hPM effect and its protein content, we evaluated the difference on HUVECs between hPM normally thawed at room te mperature and hPM heated at 90 100 Â°C to induce proteins denaturation. Two aliquots of hPM were thawed at room temperature. One of them was then heated in a baker containing water at 100 Â°C. The temperature was maintained stable on a hot plate and checked with a thermometer. After 5 minutes, the tube with hPM was removed from the baker and was left to cool down at room temperature. 980 of hPM (corresponding to 100 ÂµL /cm2) were pipetted in each well of three 6 wells plates: 3 wells coated with hPM thawed at room temperature and 3 with the denatured one. The plates were put on a shaker plate at 30 rpm for 1 minutes to allow hPM to coat on the bottom of each well and then in a humid ified 6% CO 2 incubator at 37Â°C to warm up. After that, HUVECs suspended in 1 .9 mL of angiogenic media (corresponding to 200 /cm2) were seeded at a density of 8 x 10 4 cells/cm2 onto hPM in each well. Plates were incubated for different periods of time: 1 day, 3 days, 5 days. Each one was stained using Calcein AM, imaged with a fluorescence microscope and
63 analyzed as described in Chapter 3. Effect of multiple inoculations of hPM in a HUVECs culture. 190 (corresponding to 100 / cm2) of thawed human Placental Matrix were pipetted and evenly coated onto the bottom of each well of a 24 wells plate using an orbital shaker at 3 0 rpm for 1 min ute. The plate was then incubated at 37Â° C for 30 minutes to allow the hPM to warm up. HUVECs were suspended in Angiogenic media to get to the desired cells density. of cell suspension was pipetted on the top of the hPM and then the s ame amount of Angiogenic Medium (corresponding to 200 / cm2 ) was added to each prepared sample of the plate . To evaluate the influence of the cell density on the angiogenic organization of HUVECs 12 wells were seeded with 20000 cells/cm 2 and 12 with 60000 cells/cm 2 . Three different 24 wells plates were prepared with the previous procedure and placed in a humidified 6% CO2 incubator at 37Â°C. One 24 wells plate was used as control: cells were directly seeded at the same densities used for the other plates, without hPM coating, and Angiogenic media was replaced every two days. The amount of hPM and Angiogenic media was optimized in our laborator y with previous experiment to evaluate the activity of hPM and the changes of its behavior over time . Cells seeded onto hPM coating received different profile of hPM inoculations: in one plate hPM was added only on day 1 (day of seeding), in another on day 1 and 3 and in the last one three on day 1,3 and 5. All cells were cultured for 7 days and angiogenic media was replaced on day 3 and 5. All the plates were stained on day 7, imaged with a fluorescence microscope and the images analyzed as described in Chapter 3.
64 Effect of different modality of hPM delivery on HUVECs 2D culture 12 wells plate s with six wells for ea ch of the two condition evaluated were seeded as following . For the first condit of hPM (100 /cm 2 ) was pipetted and evenly coated on the bottom of each well on a shaker plate for 1 minute at 30 rpm, and then heated in the incubator for 30 min utes. After that, cells at different densities (20000 cells/cm 2 , 40000 cells/cm 2 , 80000 cells/cm 2 ) were pipetted onto the coated hPM suspended in 760 of angiogenic media. For the second condition, cells at the same densities were directed plated on eac h well, and after that the same amount of hPM (380 ) was added to the media. Cells were stained after 3 days of culture, time points at which a more mature capillary network is usually formed. Results Assessment of the correlation between hPM pro angioge nic effect during time and its protein content . A significantly different effect in the behavior of HUVECs seeded onto the two differently treated hPM could be registered as shown in Figure 4 1 . Cells seeded onto hPM thawed at room temperature appeared to organize in an angiogenic network. Differently, cells seeded onto hPM heated at 90 100 Â°C showed only an initial attempt to organize in angiogenic meshes, as suggested from empty spaces observed a t day 1, but then on day 3 and 5 they appeared to proliferat e without organizing any mature network. Average length of the tubules, the number of meshes and the number of branching points are summarized in Table 4 1 and graphically showed in Figure 4 2 .
65 Results from one way ANOVA analysis showed a statistically sig nificant difference between the different time points of tubule length (p value<0.001), of number of meshes per mm 2 (p value=0.003) and of number of branching points per mm 2 (p value =0.004). Post Hoc test analysis conducted with un paired T test with Bon ferroni correction (significant p value set as 0.016) showed a statistically significant difference between tubule length of every time points (p values < 0.001), number of meshes between day 1 and day 5 (p value = 0.003) and number of branching points bet ween day 1 and day 5 (p values = 0.011). Table 4 1 Parameters for angiogenesis quantification of HUVECs at different time points from hPM inoculation Parameters Day 1 Day 3 Day 5 Average tubule length (Âµm) 78.38 Â± 2.89 183.10 Â± 6 .19 133.22 Â± 4.75 Average number of meshes per mmÂ² 27.75 Â± 2.80 19.12 Â± 1.25 13.39 Â± 2.22 Average number of branching points per mmÂ² 61.52 Â± 1.80 44.44 Â± 5.60 32.76 Â± 6.12 A s shown in Figure 4 1 , it was clearly detectable a maturation of the networ k after 3 days from the seeding and an initial degradation at day 5, shown by the decreasing in tubule length and by a reduced number of meshes. HUVECs seeded onto hPM in which proteins denaturation was induced showed two opposite behavior: in some areas cells proliferation stop and no alive cells could be stained, whereas in other areas HUVECs proliferated without a modulation of the initial steps of the angiogenic process. Only some empty spaces noticed at day 1 in HUVEC
66 organization suggest ed the atte mpt of an organization in meshes but no define d network formation was then observed. Effect of multiple inoculations of hPM in a HUVECs culture Angiogenic organization of HUVECs is affected by the number of inoculation of hPM received and by the density of cells seed ed onto it (as shown in Figure 4 3 ). The average tubule length, number of meshes and of branching point for each condition is summarized in Table 4 2 and graphically showed in Figure 4 4 . Table 4 2 . Parameters evaluated f or angiogenesis quantification of HUVECs inoculated with different inoculations of hPM. Parameters Cells density (cells/cmÂ²) 1 inoculation (day 1) 2 inoculations (day 1,3) 3 inoculations (days 1,3,5) Average tubule length (Âµm) 20000 95.05 Â± 3..20 152.73 Â± 5.95 168.99 Â± 7.28 60000 96.69 Â± 3.13 184.46 Â± 8.07 213.79 Â± 8.70 Average number of meshes per mmÂ² 20000 1.39 Â± 0.11 2.41 Â± 0.40 2.46 Â± 0.15 60000 8.57 Â± 1.52 5 Â± 1.62 3.12 Â± 0.53 Average number of branching points per mmÂ² 20000 23.43 Â± 4.61 17.58 Â± 2.91 19.13 Â± 1.69 60000 33.59 Â± 4.07 21.90 Â± 5.83 18.09 Â± 2.39 Results from one way ANOVA showed a statistically significant difference between the three different conditions for tubule length (P value<0.001). Post Hoc analysis conducted using unpaire d T test with Bonferroni correction (significative p value set as 0.016) shown a statistically significant increase in tubule length from HUVECs that received only 1 inoculation to HUVECs that received 2 or 3 (p value <0.001) for both cells densities. Cell s seeded with a density of 60000 cells/cm 2 showed a significant increase in tubule length also between 2 and 3 inoculations (p value =
67 0.014). The number of meshes and branching points appeared to be significantly different only for cells seeded with a den sity of 60000 cells/cm 2 (p value for meshes = 0.002; p value for branching points = 0.028) and not for cells seeded at 20000 cells/cm 2 (p value respectively 0.128 and 0.477). P ost hoc tests for cells at 60000 cells/cm 2 using un paired t test with Bonferron i correction showed no significant decrease in the average number of meshes and branching points between HUVEC which received only 1 inoculation of hPM and the ones which received multiple inoculations since p values were all higher than the p value set as threshold (0.016). Un paired T test performed between the parameters of cells seeded at different densities showed a significant increase of tubule length from HUVECs with a seeding density 20000 cells/cm 2 to the ones seeded at 60000 cells/cm 2 with 2 ino culations (p value = 0.002) and with 3 inoculations (p value < 0.001). No statistically significant difference was found in the number of meshes and branching points between different cells seeding densities. Effect of different modality of hPM delivery o n HUVECs 2D culture A faster response and a more mature network was formed by HUVECs seeded onto hPM rather than when hPM w as added in the media ( Figure 4 5 ). From a quantitative analysis of the micro capillary network, we could notice a statistical diffe rence between length of tubules in the two first conditions at all the different cell densities. (p values < 0.001). The number of branching points was increased for all cell densities when hPM was inoculated while the number of meshes appeared to decrease for cells seeded at 20000 and 40000 cells/cm 2 and to increase for cells seeded at 60000 cells/cm 2 . No statistically significant difference could be detected, except for cells seeded with a cell density of 60000 cell/cm 2 (p value=0.015
68 branching points; p value=0.048 meshes). Quantitative parameters are summarized in Table 4 3 . Table 4 3 . Parameters for angiogenesis quantification of HUVECs comparing coated hPM with inoculated hPM Parameters Delivering method 20000 cells/cmÂ² 40000 c ells/cmÂ² 60000 cells/cmÂ² Average tubule length (Âµm) hPL coated 241.78 Â± 12.67 253.70 Â± 19.46 274.83 Â± 16.38 hPL inoculated 131.84 Â± 8.01 170.77 Â± 12.07 142.45 Â± 7.86 Average number of meshes per mmÂ² hPL coated 11.95 Â± 3.65 12.38 Â± 5.83 18.09 Â± 11.22 hPL inoculated 16.34 Â± 1.12 12.69 Â± 0.89 38.17 Â± 5.76 Average number of branching points per mmÂ² hPL coated 2.96 Â± 1.20 1.90 Â± 0.44 5.87 Â± 2.91 hPL inoculated 1.42 Â± 0.22 1.26 Â± 0.44 10.3 Â± 1.12 Discussion One of the main challenges of tissue enginee ring is to provide cells in the engineered graft with nutrients and oxygen. Induce endothelial cells to organize in a microvascular network when seeded in a 2D culture represents the starting point required to achieve the desired effect in a 3D culture. U p to date no successful strategies has been optimized, therefore the pro angiogenic effect of human Placental Matrix was shown to represent a really promising alternative to animal derived pro angiogenic materials. Several strategies evaluated from differ ent research groups are based on the use of direct or sustained delivery of angiogenic growth factor to induce the formation of a
69 capillary network. Recently one of the most promising future directions purposed to reach the realization of a stable capillar y network was the development of a system able to deliver a combination of different growth factor in order to recreate the complex physiological mixture cells usually undergo in vivo . Human Placental Matrix effect was therefore hypothesized to be due to i ts heterogeneous composition and to a combined effect of different angiogenic factors. The fir st experiment here described aimed to confirm this hypothesis, knowing that if its effect is due to a combination of proteins, after their denaturation the effect derived from their function should be affected. The result of the performed experiment confirmed that inducing denaturation in the protein mixture using heat, changes dramatically the response of HUVECs inoculated with hPM. This suggested the idea that th e main strength of the extract is due to the combination of proteins contained. The evaluation of changes in the network at different time points after the seeding onto hPM confirmed what was enhanced in previous studies. The dynamic of the angiogenic netw ork formation is characterized by a rapid organization in a capillary like structure after 24 hours, a maturation of the network in 3 days and a regression of the stability of the structure after 5 days from the first application of hPM (see Figure 4 1 ). A s described in the background section , different mechanisms can bring to the formation of capillary like structure, depending upon the environment, the concentration of growth factor, the density of cells. In the second experiment, the influence of a diffe rent cell seeding density and of a different concentration of hPM over time was evaluated. Cells density was shown to affect in particular the initial phase of network formation: as shown in Figure 4 3 , initial
70 stages are characterized by a sprouting angio genesis when a lower cells density is used and by an intussusceptive angiogenesis with a higher density. This is confirmed from the significantly higher number of meshes which could be counted with a density of 60000 cells/cm 2 ( Figure 4 3 ) and of 80000 cell s/cm 2 ( Figure 4 1 ) respect to a cell density of 20000 cells/cm 2 ( Figure 4 3 ). The evaluation of the effect of multiple inoculation of hPM on the stability of the newly formed network confirmed that multiple inoculations of hPM at discrete time points (day 1, 3, and 5) could bring to a more mature network formation. In particular, tubules length was significantly increased in cell culture which received hPM at multiple time points compared to the ones which received of hPM only once, the number of meshes was r educed and they appeared wider and finally less branching points could be counted because of the increased maturity of the network. These findings suggested that a suitable controlled release of hPM could represent the solution to improve the stability of the newly formed capillary like HUVECs network.
71 Figure 4 1 . Assessment of the correlation between hPM effect and its protein content: Cell morphologies of microvessel tubules formed by HUVECs seeded onto hPM and cultured for different periods of time da ys compared with HUVECS seeded onto denatured hPM . From top to bottom: A,C and E represent HUVECs response to denatured hPM respectively at days 1,3,5; B,D and F represent HUVEC response when seeded onto hPM. Cell seeded 8 x 10 4 cells/cm 2 .
7 2 Figure 4 2 . Morphological and topographic features of angiogenic network formed on hPM by HUVECs seeded at 8 x 10 4 cells/cm 2 . From top: average tubule length, average number of meshes and of branching points per mm 2 formed by HUVECs incubated for different period o f time (1,3,5 days). Symbols on the top of columns indicate a statistical difference between the two groups determined p erforming ANOVA experiment followed by a double tailed t test with unequal variance corrected with Bonferroni correction.
73 Figure 4 3 . Effect of multiple inoculation of hPM on different densities of HUVECs: Cell morphologies of microvessel tubules formed by HUVECs seeded onto hPM and cultured for 7 days compared with HUVECS seeded onto a tissue culture plate, s hown as a control. From top to bottom : Figure A and B controls; Figure C and D only one inoculation of hPM on day 1, Figure E and F two inoculations on day 1 and 3 and Figure G and H three inoculations on day 1,3 and 5. From left to right: Figure s A,C,E and G cell density of 20 000 cells/cm 2 and Figure s B,D,F and H cell density of 60000 cells/cm 2 .
74 Figure 4 4 . Morphological and top ographic features of angiogenic network formed on hPM by HUVECs which received 1,2 or 3 inoculation . From top to bottom: average tubule length, average number of BPs and of meshes for mm 2 formed by HUVECs of PE (A,B and C respectively). Asterisks indicates a statistical difference between the two groups determined performing ANOVA experiment followed by a double tailed t test with unequal varianc e corrected with Bonferroni correction. Average tubule length, number of BPs and of meshes were assessed after 7 days of culture.
75 Figure 4 5 . Effect of different hPL delivery methods : e valuation of the difference between HUVECs response to hPM coated ( A,B,C) and hPM added in s olution with media (D,E,F) .
76 CHAPTER 5 DEVELOPMENT OF A HETEROGENEOUS PROTEIN MIXTURE ENCAPSULATION METHOD USING PLGA MICROPARTICLES Background and Rational As hypothesized in the d iscussion of Chapter 4 , a controlled release meth od suitable for protein encapsulation could represent a solution to achieve a slow release of hPM and to obtain a more stable and long lasting capillary networ k. Differently from drug encapsulation, protein encapsulation represent s a challenging procedure since it requires specific attention to maintain the specific activity of the protein is strongly affected by its folding which must be protected during particles preparation. As descr ibed in Chapter 2 , several studies have investigated the use of micropar t icles to encapsulate growth factors to ind uce angiogenesis. In particular, PLA copolymers represent the most used materials allowing a controlled protein release over week s and even month. Moreover, the release profile can be optimized changing morphologi cal parameters such as size, protein encapsulated, and composition of the polymer. PLGA represent s one of the most investigated among those materials : microparticles o f a size ranging from 5 to 120 Âµm have been wi dely used for tissue engineering applicatio ns, obtaining controlled release over a month. T o date no studies have been performed trying to encapsulate such a complex protein mixtur e as the human Placental Matrix, and this represents one of the main novelties of this study. P ossibl e issues related t o the encapsulation of protein with different charges and properties have been investigated: interactions between different proteins and between proteins and polymers could prevent the complete release of the proteins from the microp articles or affect size and loading efficiency of microparticles 94 .
77 Our aim is to optimize a protocol to obtain microparticles of an average size lower than 100 Âµm, with a lo ading efficiency higher than 60 % and to obtain a release rate characterized by an initial burst lower than the 30 % and a con trolled release over one month. This results would represent the desired features described in the literature as compatible with microparticles use in 3D cells cultures 77 . Consider ing all this desired properties , start ing from a protocol suitable for a single model protein encapsulation (BSA) an optimization process was performe d , modify ing specific steps of the protocol after a careful analysis of each described parameters. Materia l and M ethods Materials Polyvinil alcohol ( molecular wt 30000 70000, 87 90% hydrolyzed, Sigma P8136 ), BSA (Sigma A9647 ), Poly(DL lactide co glycolide) ( Lactel, absorbable polymers, product number B6010 4 ), Chloroform, human Placental Matrix (see General Material a nd Methods for derivation protocol) . P reparation of hPM loaded PLGA Microparticles PLGA microparticles were prepared using a water in oil in water emulsion. The first water solution (W1) was represented by the protein mixture, hPM, prepared ac cording to the protocol optimized in our laboratory. The second water solution (W2) was prepared by dissolving 2g of polyvinyl alcohol in 100 mL of DI water. The oil solution (O) was obtained by dissolving 90 mg of PLGA in 3 mL of Chloroform until the solut ion appeared clear. W1 was added to O and homogenized at 2 x 10 4 rpm for 1 minute. Using a micropipette, the obtained primary emulsion was added dropwise in W2 while stirring at 300 rpm. When all the primary emulsion was added, the resulting secondary emu lsion was covered loosely with an aluminum foil and left to stir (300 rpm) overnight
78 in a fume hood to let the solvent evaporate. The day after, the secondary emulsion was centrifuged at 1000 rpm for 10 minutes, the supernatant removed and the microparticl es were washed two more time s with DI water. After that, hardened microparticles were suspended in DI water, free ze dried for 48 hours and then stored at 4C until needed. Protocol Optimization The heterogeneou s nature of hPM represents a challenging elem ent to get to PLGA microparticles with the desired features. Starting from th e comparison between PLGA m icroparticles loaded with a single model protein (BSA) and PLGA microparticles hPM loaded using the protocol previously described, we proceed ed with an evaluation of the variables that could be modified in order to obtain the desired results. Modifications of the original protocol during the optimization process are following described . a. BSA co encapsulation (PROTOCOL 2) Firstly, a modification of the water protein phase was evaluated, to assess the effect of BSA co encapsulation on size and release. 3 mg of BSA were dissolved directly in 1 mL of hPM , while preparing the protein solution. b. M odification in the homogenization of the primary emulsion (PR OTOCOL 3) The second modification was related to the duration of primary emulsion homogenization.Since different studies in the literature had addressed to the duration of the homogenization steps as important to modify the size of microparticles 94 , we m odified this phase from one to two minutes. c. Secondary emulsion homogenization (PROTOCOL 4) One of the main differences between micro and nanoparticles preparation protocols is represented by the presence of a second homogeniz ation phase after the
79 secondary emulsion has been created. This suggested that introducing a further homogenization phase in the protocol could reduce the size of the microparticles. Therefore , the protocol was modified adding an homogenization of the seco ndary emulsion at 2 x 10 4 rpm for 1 minute. d. Modifi cation on the homogenization of the secondary emulsion (PROTOCOL 5) The last modification of homogenization timing was related to the modification of the secondary emulsion homogenization duration. Respe ct to protocol 4, the only difference was the decreasing of the duration from 1 minute to 20 seconds. Before proceeding in the opt imization process, microparticles morphologic al feature and the associated release rate after every modification were evaluate d and compared with the desired results. PLGA microparticles features characterization: size and surface evaluation Morphologic al features of PLGA microparticles were evaluated using different microscopy techniques. Microparticles size was estimated th rough image analysis, using a color digital camera attached to an optical microscope (Leica DM IL LED, Leica Mycrosystems Inc., IL, USA). The acquired images were saved as TIFF files a nd analyzed using the free software ImageJ 1.45s. After setting the scal e (Analyze > Set Scale), microparticles diameter wa s determined by drawing a line along each particle. M easure ments were a utomatically calculated by the software (Ana lyze sheet. Surface morphology and porosity of the microparticles were characterized using a Scan ning Electronic Microscope ( S 4000 FE SEM, Hitachi High Technologies, TX, USA) . F reeze dried microspheres were mounted onto aluminum stub using double -
80 sided graphite tape and then sputter coated with a thin lay er of gold and palladium using a scatter (deskV, Denton Vacuum) before examination. C oated samples were then examined under the microscope at an acceleration voltage of 10 kV a nd photographed using a lower magnif ication to evaluate dimension an d shape of m icroparticles an d using a higher one to evaluate the change in porosity and morphology features while proceeding with the optimization protocol. Loading efficiency L oading efficiency of hPM in PLGA microparticles was analyzed c omparing two different method s. The first one consists in the direct measurement of proteins encapsulated after microparticles dissolution. The protocol used was an hydrolysis technique described in several studies using PLGA microparticles 108 , 73 . Briefly, 15 mg of lyophilized microspheres were digested with 5 mL of 0.1 M NaOH containing 5% w/v SDS and hydrolyzed on a shaker for 15 h at ambient temperature until a clear solution was obtained. Sodium hydroxide catalyzes the hy drolysis of the polymer and SDS ensures the complete solubilization of the protein during the polymer hydrolysis. The resulting clear solution was then neutralized to pH 7 by addition of 1 M HCl and centrifuged at 5000 rpm for 10 min. Protein concentration in the supernatant was then analyzed in triplicate for each protocol using Spectr ophotometer , with a BCA standard Pierce protein assay . The second method was represented by an indirect measurement of microparticles protein encapsulated by quantifying the amount of protein left in the supernatant, not encapsulated during the process. Since the concentration of proteins
81 to be detected was really low, a micro BCA Pierce protein assay was preferred, since the standard assay could not detect a protein concentra tion under 25 Âµg/ mL . The encapsulation efficiency was expressed as the ra tio of encapsulated protein to proteins added during microparticles preparation. In vitro protein release from hPM loaded PLGA microparticles For this purpose, 10 mg of hPM loaded microparticles were suspended in 1 mL of Phospate Buffer Saline (PBS). Three samples we re evaluated for each protocol. Suspended microparticles were incubated at 37 Â°C in a shaker incubator to avoid the microparticles to sink onto the bottom of the tube. Every two days the tube s were centrifuged at 5000 rpm for 5 minutes. The supernatant was collected and restored with the same amount of fresh PBS. The period for in vitro release rate evaluation was fixed at 21 days, as a starting point to evaluate hPM re lease over a period compatible with cell culture. Protein concentration was quantified by measuring the absorbance at 562 nm with a BioTek microplate reader, using a protein assay ( Micro BCA Protein Assay Kit Pierce ) . 150 of release buffer was loade d in wells of a 96 well plate. Protein c oncentrations were compared to freshly prepared standards ranging from 200 to 0.5 Âµg/ mL . The linear working range of the assay was 1 20 0 Âµg/ mL and the detection limit 0.1 Âµg/ mL . Average absorbance of standard PBS was subtracted from all measurements . The amount of proteins in each sample was summed with the amount at each previous time point to build a cumulative release curve. These results were integrated with results from the encapsulation efficiency to obtai n a p ercentage cumulative releas e.
82 Sodium Dodecyl Sulfate Polyacrylamide gel electrophoresis (SDS PAGE) Since hPM represents a complex protein mixture, which include s growth factors characterized by various charges and properties, we were interested in unders tanding if the different properties of the proteins contained in the extract could affect the encapsulation process. A Sodium Dodecyl Sulfate Polyacrylamide gel electrophoresis (SDS page) analysis was performed on the supernatant collected after the over night stirring. Four different solution were analyzed: the first and the second one prepared dissolving respectively BSA and hPM in PVA with the same concentration used during mi roparticles preparation . 10 mg of BSA were dilute d in 50 mL of DI water to pre pare the first solution while 100 of hPM were dilute d in 5 mL of DI water to prepare the second one. The third and th e forth solutions were represented by the supernatant removed while collecting respectively BSA loaded and hPM loaded microparticles. Protein st andards where used to comp are molecular weight of the detected proteins. SDS page was performed using BIO RAD electrophoresis system. Firstly, running buffer was prepared diluting 100 mL of 10x running buffer stock with 900 mL of DI water. Samples were prepared mixing 40 the d iluted sample , 38 of 2x Laemm 0 li sample buffer and 2 mercaptoethanol, and then heated at 90 100 Â°C for 5 minutes. After that a Mini PROTEAN Â® Bio Rad ) was prepared and the electrophoresis cell was assemb led. The inner buffer chamber was filled with 200 mL of 1x running buffer and the outer buffer chambers with 550 mL . 4 0 of samples were carefully loaded in each well using a fine pipette tip. When the gel was loaded with all the samples , the power suppl y was connected and electrophoresis was performed with a fixed voltage of 200 V until the color ed band reached the bottom line of the precast gel.
83 After electroph oresis was complete, the power supply was turned off and the electrical leads disconnected. T he gel cassettes was popped up, the gel was removed by floating it off the plate into water and then stained with Coomassie Blue for one hour under continuous shaking. Gel destaining was performed using a 4:1:5 Methanol: Acetic Acid: DI water solution. Fin al gel was washed in DI water and then phptographed. Results P reparation of hPM loaded PLGA Microparticles PLGA microparticles resulting from the initial protocol for hPM encapsulation ( Figure 5 1 ) showed an average size of 447 Â± 32 Âµm and a normal dist ribution of sizes from 100 to 1000 Âµm. The loading efficiency of hPM loaded PLGA microparticles quantified with an indirect method was 62 Â± 4% and with the direct method was 64 Â± 4%. From an in vitro release analysis a low initial burst was released, corr esponding to 10% of the total amount of protein encapsulated, (21.38 Â± 0.83 Âµg/ mL ), and the 51.23 Â± 14.07 % (134.56 Â± 36.96 Âµg/ mL ) of the protein initially encapsulated was released from PLGA microparticles after 21 days. The release rate profile appeared to be linear through the period evaluated, but the concentration released at the end of the evaluated period was really lower comp ared with the concentration optimized to induce angiogenesis with inoculation of pure hPM. Control PLGA microparticles prepare d with the same protocol but loaded wit h a single model protein ( BSA ) showed an average size of 239.60 Â± 9.45 Âµm and a normal distribution of sizes fro m 50 to 400 Âµm, with the 70% of microparticles in the range between 150 and 350 Âµm. The loading efficien cy of BSA loaded PLGA microparticles quantified with an indirect method was 72 Â± 5% and with the direct meth od 76 Â± 3 %.
84 From an in vitro release analysis a higher initial burst compared to hPM loaded microparticles was released, corresponding to 20% of th e total amount of BSA encapsulated, ( 51.52 Â± 1.76 Âµg/ mL ), and the 66.7 Â± 9.55 % ( 166.74 Â± 23.8 Âµg/ mL ) of the protein initially encapsulated was released from PLGA microparticles after 21 days. The release rate profile appeared to be linear and similar to hPM ones in the period evaluated. The different results observed between hPM and BSA loaded (see Figure 5 1 ,13 and 14) PLGA microparticles suggested tha t an optimization process was required. Protocol Optimization a. BSA co encapsulation (PROTOCOL 2) S ince BSA loaded microparticles prepared with the same protocol showed a statistically significant smaller average diameter (p value<0.001) then hPM loaded ones, the first attempt to reduce the size of microparticles was to change the composition of the pro tein solution adding BSA as an emulsifier (PROTOCOL 2). Results from the size quantification ( Figure 5 2 ) showed a significantly smaller average diameter for the microparticles encapsulating both hPM and BSA. Average diameter was 250.83 Â± 12.68 Âµm. One way ANOVA performed between sizes of hPM , BSA and hPM +BSA loaded microparticles showed a statistically significant p value (p<0.0001), and post hoc tests showed a statistical significant difference bet ween PLGA microparticles hPM loaded and BSA loaded, and b etween the one hPM +BSA loaded and hPM loaded . No significant diff erence was assessed between BSA loaded and hPM +BSA loaded. From an in vitro release analysis ( Figure 5 3 ) a higher initial burst compared to hPM loaded microparticles was released, correspond ing to 17 Â± 2.24 % of the total amount of BSA encapsulated, ( 67.11 Â± 8.80 Âµg/ mL ), and the 98.93 Â± 23.43%
85 (388.81 Â± 92.02 Âµg/ mL ) of the protein initially encapsulated was released from PLGA microparticles after 21 days. In vitro release profile appeared to follow a different trend from release profile of hPM and BSA loaded microparticles. Loading efficiency resulted equal to 55 Â± 2% and 53 Â± 4% respectively for the direct and indirect method ( Figure 5 3 ) . SEM analysis confirmed the average size resulting f rom the optical microscopy. The porosity appeared to be uniform and the shape of the microparticles appeared to be disc shaped (see Figure 5 6 ) . Despite the significant decreasing in sizes, the resulting microparticles were still bigger than the range usua lly suitable for tissue engineering application and the completely different trend of release suggested us to pursue the optimization process with an optimization of the emulsification steps, in term of duration, without affecting hPM composition. b. Modif ication of the primary emulsion homogenization (PROTOCOL 3) Results from the modification of primary emulsion homogenization showed microparticles with a significantly smaller size, wi th an average diameter of 276.78 Â± 13.41 Âµm ( Figure 5 4 ) . The initial bu rst registered in the in vitro release evaluation was not significantly different from the original protocol: from the 8.14 Â± 0.31 % of protocol 1 (21.38 Â± 0.83 Âµg/ mL ) it became 6.42 Â± 2.07 % in protocol 3 (18.88 Â± 6.11 Âµg/ mL ). Differently the cumulative r elease after 21 days increased significantly from 51.23 % to 64.09 Â± 24.87 % (188.43 Â± 73.13 Âµg/ mL ). Loading efficiency was shown to increase to 68 Â± 6% and 73 Â± 4% for the indirect and direct methods respectively . Results of in vitro release and loading efficiency summarized in Figure 5 5 .
86 SE M results showed uniform porosity and a spherical shape ( Figure 5 6 ) . c. Secondary emulsion homogenization (PROTOCOL 4) Results obtained introducing one minute of homogenization of the secondary emulsion produced micr oparticles with an average size of 37.79 Â± 1.30 Âµm ( Figure 5 4 ). A significant increase both in burst release and in percentage release after 21 days could be noticed , associated with the significant reduction in size: the 23.75 Â± 0.40 % of the total prote in encapsulated (60.82 Â± 1.03 Âµg/ mL ) was released within the first day, while 98.62 Â± 27.98 % (252.48 Â± 71.64 Âµg/ mL ) during 21 days. Loading efficiency to decrease to 60 Â± 5% and 63 Â± 3% for indirect a nd direct method respectively . Results of in vitro rele ase and loading efficiency summarized in Figure 5 5 . SEM results showed a surface with uniform porosity and a spherical shape ( Figure 5 6 ) . In some images, fragments of incomplete microparticles could be noticed. d. Modification of the secondary emulsion h omogenization (PROTOCOL 5) Reducing the homogenization time of the se condary emulsion to 20 seconds resulted in microparticles with an average size of 91.85 Â± 2.92 Âµm, ranging from 5 to 200 Âµm, with the 70% between 20 and 100 Âµm ( Figure 5 4 ) . 27.72 Â± 3.73 % (106.33 Â± 14.33 Âµg/ mL ) of the protein encapsulated were released during the first 24 hours,and then a controlled release with a linear trend co uld be registered with the 87.60 Â± 1 4.02 % (335.98 Â± 53.77 Âµg/ mL ) of all the protein encapsula ted released afte r 21 days . Further analyse s over a longer period of time were performed to compare protocol P4 and P5 . A fter 30 days, microparticles from protocol 4 appear ed to be completely degraded and to have released all the amount of proteins encapsulated , while micr o particles from protocol 5 released the 93.12 Â± 1 4.3 2 % (357 Â± 54.95 Âµg/ mL ) of the total proteins encapsulated. The encapsulation efficiency
87 was the highest obtained, 70 Â± 8 % and 79 Â± 9 % for the indirect and direct methods respectively. Results of in vit ro release and loading efficiency summarized in Figure 5 5 . SE M results showed a uniform porosity and a spherical shape ( Figure 5 6 ) . Sodium Dodecyl Sulfate Polyacrylamide gel Electrophoresis (SDS PAGE) Results from SDS page ( Figure 5 7 ) showed a unifor m encapsulation of hPM in PLGA microparticles , without a selective encapsulation for specific type of proteins. The analysis on the supernatant collected from hPM loaded microparticles showed a similar protein content than the solution of pure hPM but with lower concentration. A similar result could be observed for the analysis of supernatant collected from BSA loaded microparticles, used as a control. The broad number of band detected in hPM electrophoresis analysis confirmed what previously assessed in ou r laboratory during hPM characterization. Discussion Several studies addressed PLGA as a suitable material to obtain a controlled delivery of growth factors for tissue engineering applications. In this study, we used a water in oil in water technique to en capsulate the pro angiogenic mixture in PLGA microparticles. The final aim was to obtain a controlled and sustained release of hPM over a long period of time, to stabilize HUVEC response while creating a capillary like structure. The first element enhanced in this study was the influence of the protein mixture for a single protein (BSA) encapsulation, resulting hPM loaded microparticles showed an average size significantl y bigger than BSA loaded ones. (p value < 0.001). This suggested that the reason could be due to the different interaction the proteins
88 contained in the protein mixture were able to create with the polymer, differently from BSA. Bovine Serum Albumin is a p rotein which has been used widely in several different microparticles protocol both as model protein to evaluate the feasibility of the encapsulation process and as an emulsifier agent to stabilize the first water in oil emulsion when a single p rotein had to be encapsulated 109 . C o encapsulation of BSA with hPM was evaluated to reduce the size and increase the loading efficiency. Despite the significant reduction of the size, microparticles size was still bigger than the desire d range and present a trend different from other protocols. Because of the difficulties in understanding the complex interactions between BSA and hPM proteins and of the different release kinetics, the optimization process was continued without affecting t he composition of the protein mixture. Literature findings show that both the speed and the duration of the homogenizing phase strongly affect the size of the microparticles 92 94 . One of the main dif ference s between microparticles and nanoparticles production protocol is a different modality to obtain a higher homogenization of the primary and of the secondary emulsion. Homogenization of the secondary emulsion turned out to represent the key point to be modified in the process in order to get to hPM loaded the desired range. Despite microparticles from protocol 4 possessed all the desired characteristics to represent the optimized controlled release method for our aims, the load ing efficiency was significantly reduced. Encapsulation efficiency measurement were repeated three times to assess the reliability of results. A reduction in the duration of the timing of the homogenization phase of the secondary emulsion produced
89 micropar ticles slightly bigger but in the range suitable for tissue engineering applications, and the encapsulation increased up to 78% of the total amount of proteins loaded in the protein solution. The encapsulation effici ency did not appear to be cor related li nearly with the size of the microparticles, but a too long homogenization of the secondary emulsion appeared to affect the loading efficiency, presumably due to the formation of fragmented incomplete particles. On the opposite, results in term of in vitro release rate evaluation of the different batches obtained with the different protocols duri ng the optimization process were strongly correlated to the size ( Figure 5 5 ), in agreement with literature findings 95 72 Results from a comparison between the release kinetic of the different protocols obtained with a modification of the homogenization steps showed an overall similar trend, except for some differences. In protocol 4, mostly due to the lower enca psulation efficiency, the release was faster and characterized by a saturation in the release profile from day 17 on. In protocol 3, a particular trend was registered, with a flat rate around day 7: possible reasons to explain this could be related to a pa rticular interaction between proteins and polymer at the specific order of magnitude which characterize the protocol, or to specific degradation kinematic of PLGA . A deeper understanding of this mechanisms would help to understand the release dynamics 94 . Microparticles from the optimized protocol showed a linear release, characterized by a profile similar to protocol 1 but with higher protein release controlled over 30 days, smaller dimension (< 100 Âµ m) and higher efficiency (> 60 %). Results from SDS page, evaluated only for the final protocol, showed a homogeneous not selective
90 encapsulation of the protein mixture, suggesting that the protein composition of hPM was not affected during the encapsulation process. All these features , suggested the use of these optimized controlled release method to develop an Alginate based angiogenic assay to assess HUVECs response to the controlled release of hPM over an extended period of time. Figure 5 1 . Results from original protocol hPM enc apsulation. From the top left: Optical Micros cope image of hPM loaded MPs; size distribution; Re lease rate profile over 21 days. Results from the encapsulation process (in the table) showed a fair encapsulation efficiency, a slow controlled release and a size which appears to be bigger than the commonly used particles for growth factors delivery,
91 Figure 5 2 . Effect of BSA co encapsulation on PLGA microparticles size. From top: histogram comparing average diameter size of BSA loaded, hPM loaded and BSA + hPM loaded microparticles; Optical images of microparticles from left BSA, hPM , BSA + hPM loaded . Scale bars 500 Âµm.
92 Figure 5 3 . Effect of BSA co encapsulation on PLGA microparticles release : t he release from hPM loaded , BSA loaded and hPM + BSA loade d microparticles was evaluated. From the top: Comparison between percentual cumulative release over 21 days, loading efficiency as ratio between (mg of proteins encapsulated)/ (mg of protei ns in hPM + 2 mg of BSA)
93 Figure 5 4 . Microparticles size eva luation. From the top to bottom on the left optical images and on the right the correspondent histogram of size distribution for all the different protocol during the optimization process.
94 Figure 5 5 . Release rate evaluation: the curves represented in the graph s represent the cumulative release rate up to 21 days for all the protocol during the optimization process; to assess the protein release 10 mg of microparticles were incubated in 1 mL of PBS, and the supernatant was collected and analyzed using a Pierce BSA protein assay kit. Every collection was conducted triplicate. From the top: cumulative release of microparticles in Âµg, percentual cumulative release, loading efficiency both measured with an indirect method (measuring protein non encapsulated in the supernatant after microparticles collection) and dissolving microparticles in an NaOH SDS page solution.
95 Figure 5 6 . SEM characterization. Scanning Electronic Microscope imaging of PLGA microparticles loaded with hPM for every protocol through th e optimization process. From top to bottom SEM evaluation of all the protocol through the optimization process, on the lieft microparticle aspect and shape evaluation (A, C, E, G) on the right surface porosity evaluation with a higher magnification (B, D, F, H).
96 Figure 5 7 . SDS page analysis of microparticles supernatant: photograph of gel after staining showing qualitative protein content in supernatant of BSA loaded and hPM loaded microparticles and of the relative initial content loaded during micropar ticles preparation. A uniform encapsulation seems to be suggested for hPM, even if the low concentration of proteins in the supernatant prevent a clear detection of protein which had a lower concentration even in the initial mixture.
97 CHAPTER 6 DEVELO PMENT OF AN ANGIOGENIC ASSAY USING ALGINATE HYDROGEL AND HPM LOADED PLGA MICROPARTICLES Back g round and Rational As discussed in Chapter 2, in vitro angiogenic assays represent the fastest and cheapest way to evaluate and quantify angiogenic organization of endothelial cells. A three dimensional space could represent a useful method to mimic the physiological environment in which cells usually perform their functions. In Chapter 5 , a suitable technique for hPM encapsulation was optimized. In the light of thi s achievement, the aim of the exper iments here described was to show that the controlled delivery of the pro angiogenic mixture from PLGA microparticles could prevent the degradation of the capillary network and stabilize the formation of a more mature ang iogenic structure. Alginate has been evaluated as a material able to sustain endothelial cells organization in an angiogenic network and therefore suitable as a matrix for in vitro angiogenesis assay 76 , 101 . An alginate based hydrogel was selected for these experiments, to realize a time and cost effective test. Different methods to encapsulate cells in an Alginate based matrix has been described in the literature: beads preparation with controlled release methods encapsulated, preparation of a homogeneous matrix before cell seeding, formation of a matrix with cells suspended 99 , 110 . The first aim was therefore to optimize all the parameters necessary to create the best 3D culture to allow a long term viability
98 of cells. The second aim was to combine the effect of a 3D matrix and of the controlled delivery of hPM from PLGA microparticles to evaluate the effect on the network stability. Materi als and M ethods Materials HUVECs were isolated as described in Chapter 3. They were detached from the flask using Accutase, suspended in angiogenic media and finally counted with a hemocytometer to obtain the density required for the different experiments. Angiogenic media was obtained by adding the following supplements to 500 mL of VascuLife Basal Media: 25 mL of glutamine, 0.5 mL of hydrocortisone, 0.5 mL of ascorbic acid, 10 mL of FBS, 1.25 of VE GF, and 1.25 of bFGF. Human Placental Matrix was prepared following protocol described in Chapter 3. PLGA microparticles hPM loaded were prepared following protocol 5. Materials used to prepare the 3D matrix were Sodium Alginate (Sigma Aldrich) and Calcium Chloride (Fisher Scientific). Optimization of 3D culture parameters An optimization of the parameters required to obtain a suitable matrix for cells viability was performed. Different Alginate concentrations (0.5%, 1%, and 1.5%) and Calcium Ch loride concentrations (0.025 M, 0.054 M, 0.128 M) were compared to evaluate the effect on matrix properties. Different thicknesses of the substrate and different composition of the solutions were evaluated to assess the behavior of cells cultured for 10 da ys in Alginate gel. Finally, different protocols to prepare the Alginate hydrogel were evaluated.
99 Sodium alginate solutions were sterilized using filtration method, to avoid the inevitable enhancement of polymer breakdown caused by heating 111 . Calcium Chloride solutions was autoclaved. In a 12 wells plate 3 different conditions were compared, four wells for each condition. For the first condition 800 of Sodium Alginate solution was pipetted in each well and then 800 of Calcium Chloride was carefully pipetted on the top of the previous layer, paying attention to distribute it homogeneously to avoid differences in matrix thickness and features. The plate was left still for 10 minutes to allow polymerization of the gel, and then HUVECs suspended in Angiogenic media were pipetted on the top of the matrix at a density of 60000 cells/cm 2 . For the second condition, HUVECs were suspended in Alginate solution, slowly mixed with a pipette to create a homogeneous mixture. After that 800 of solution was pipetted in each well, 800 of Calcium Chloride was carefully pipetted on top of cells suspension and the plate left still for 15 minutes to allow polymerization. For the last condition, Alginate beads were realized. 2 mL of Calci um Chloride was pipetted in each well and HUVECs were suspended in Alginate solution. This suspension was dropped using a pipette from 10 cm above the plate and then the plate was left still for 15 minutes to allow the formation of the beads. For both the last two conditions, after polymerization, exceeding Calcium Chloride was removed, hydrogel washed twice with PBS and 760 ul of angiogenic media was added (200 ul/cm2). Two plates were prepared under the same conditions. They were stained with
100 Calcein AM a t two different time points (5 and 10 days) to evaluate the distribution and the viability of cells in the matrix. Angiogenic assay: Alginate hydrogel based 3D culture of HUVECs Three different conditions were compared: a control group with HUVECs embedded in Alginate matrix, a group with HUVECs embedded in Alginate matrix with pure hPM mixed to the solution, and a group where HUVECs were suspended in Alginate matrix with hPM loaded MPs embedded. 12 wells plate were used for every condition. For the control group HUVECs cells were suspended in 2.6 mL of media and then slowly mixed to 7 mL of Alginate 1.5 %. For pure hPM loaded matrix, 3 mL of hPM was slowly mixed to 6 mL of Alginate 1.5 % and finally cells were suspended with 1 mL of media were slowly adde d and mixed to the suspension. For the angiogenic assay with hPM loaded microparticles embedded in the Alginate matrix 72 mg of microparticles (6 mg per well) were suspended in 4 mL of PBS. MPs suspended were added to 6 mL of Alginate 1.5 % solution and mi xed to get to a homogeneous compound. HUVECs were suspended in 1 mL of Angiogenic media and then gently mixed to the alginate + MPs suspension. For all the three conditions, 800 of alginate solution with the previously described compositions were pipet ted in each well, 800 of Calcium Chloride were carefully pipetted on top and then the plate was left 15 minutes still to allow polymerization of the hydrogel. After polymerization occurred, 760 of angiogenic media was added to each plate, and they we re put at 37Â°C in a humidified 6% CO2 incubator.
101 Two different investigations were performed. One to evaluate the long term effect of MPs compared with the long term effect of pure hPM, staining one plate for each condition at day 7, 14 and 21 with Calcein AM. Another analysis aimed to evaluate a comparison of the response of HUVECs during time to pure and controlled release hPM. Six wells for each condition were stained on day 2,4,6,8 and 10. The spatial organization of the HUVECs in capillary like network was evaluated quantifying morphological and topological parameters as described in Chapter 3. Results Optimization of 3D culture parameters Different conclusions were derived from evaluation of different combinations of Alginate and Calcium Chloride conce ntrations. 0.5% Alginate was not suitable to suspend cells, because they tended to sink and attached to the bottom of the well while polymerizing was taking place. 2% Alginate solution, on the opposite, produced a thick matrix where cells showed a lower le vel of viability after 10 days (results not shown). 1.5 % Alginate produced best results in term of viability and cell distribution. Between the different concentrations of Calcium Chloride the highest (0.128 M) showed to produce too compact and thick mat rixes, while the lowest (0.025 M) increased the time needed to allow polymerization, affecting cells viability. 0.054 M Calcium Chloride gave best results DI water and PBS were compared to evaluate differences in cells viability: after 10 days cells plated with Alginate solution prepared with water did not show a high viability while cells plated under the same conditions with Alginate solution prepared with PBS showed a higher percentage of viability.
102 Thus, Alginate 1.5% prepared with PBS and Calcium Chlor ide 0.054M were used to optimize the protocol for 3D culture realization ( Figure 6 1 ) . HUVECs plated on the top shown a low degree of migration in the matrix. They appeared to adhere to the upper surface, with a behavior comparable with a 2D culture. The p rotocol to prepare Alginate based homogeneous matrix shown to polymerize after 15 minutes and when washed with PBS it showed a compact matrix of a thickness of around 2 mm. HUVECs embedded appear circular shaped, except for some of them which seemed to adh ere more and assume an elongated shape. The distribution of cells in the three dimensions was homogeneous. The level of viability of cells embedded was maintained after 10 days, suggesting that the thickness of the matrix was suitable for nutrient and gase s providing, even if not a high proliferation could be observed. Resulting Alginate beads from the third protocol had a white compact aspect and a size around 4 mm. Cells embedded appeared to be more concentrated and homogenously distributed. Angiogenic as say: Alginate hydrogel based 3D culture of HUVECs Results from the angiogenic assay ( Figure s 6 2 and 6 3 ) confirmed the hypothesis that a controlled release of hPM could improve the stability of the capillary like structures formed from HUVECs. After 7 days of culture ( Figure 6 2 ) only a few sprouts and some tubular structures could be observed in the wells were cells were embedded together with MPs. No mature meshes could be counted. Oppositely, HUVECs embedded in Alginate matrix containing pure hPM formed an initial angiogenic network, characterized by an
103 average length of tubules of 207.90 Â± 15.31 Âµm, with 1.27 Â± 0.84 meshes and 9.60 Â± 0.70 branching points per mm 2 . Control wells showed circular shaped endothelial cells. After 14 days ( Figure 6 2 ) , HUVECs embedded with hPM loaded MPs started to organize initial tubules with an average length of 137.41 Â± 9.77 Âµ m and 21 Â± 1.5 branching points per mm 2 . Only a few meshes could be counted. At the same time point, the network formed by HUVECs cultured with pure hPM appeared to be almost completely degraded, with no meshes observed but just isolated tubules with an average length of 226.58 Â± 25.64 Âµm. At 21 days ( Figure 6 2 ) , while the angiogenic network in alginate with pure hPM was completely degraded, initial o rganization of more stable tubular like structures could be observed in HUVECs cultured with hPM loaded microparticles. The average tubule length was of 204.07 Â± 12.75 Âµm an d an average number of branching points per mm 2 of 11 Â± 1.25. Only few immature mes hes could be observed. Results from the second experiment ( Figure 6 3 ) to compare HUVECs response to pure and controlled released hPM over time showed different trends.HUVECs response to pure hPM appeared to be faster, since at day 4 some tubular structure s could be observed and a real angiogenic network could be appreciate at day 6. It was c haracterized by an average length of tubules of 207.90 Â± 15.31 Âµm, with 1.27 Â± 0.84 meshes and 9.60 Â± 0.70 branching points per mm 2 . From day 8 it was shown to start de grading and on day 10, no network could be detected but only isolated tubules. HUVECs response to hPM controlled delivered from PLGA microparticles appeared to be slower. Up to day 6 it was characterized by isolated sprouts and tubules. At day 8 it
104 started to organize some initial network formation, even if they appear to be not homogeneously distributed and no formation of complete meshes could be detected. Discussion The optimization of some culture parameters performed with the first experiment together with evaluation of Chapter 4 aimed to get to an environment able to properly guide the formation of a capillary network within the gel. The influence of the protocol for alginate matrix preparation, of the concentration of hPM, of the density of cells and of the amount of microparticles on cellular response were evaluated. In light of the results obtained, we proceeded with an in vitro angiogenic assay to evaluate HUVECs response to hPM pure and controlled delivered from PLGA microparticles. Outcomes after 21 days of culture showed evidence of tubule like structures formation within the gel where PLGA microparticles hPM loaded were embedded. Despite this, the angiogenic network observed did not appear to be as mature as the one HUVECs were shown to organize after 3 days when cultured onto hPM in 2D culture. However, results from the comparison between HUVECs response to pure and controlled delivered hPM, confirmed PLGA microparticles as a promising method to sustain HUVECs organization in tubular structure f or an extended period of time compared with a dose of pure hPM directly mixed in the matrix. While with pure hPM cells in the matrix cells appears to respond faster, but to create an angiogenic network with a limited stability, cellular response to hPM con trolled delivered seems to be slower but to increase during time, maintaining stability of the structure previously created. The reasons why the final capillary like structure l ack of maturity and of complete angiogenic meshes could be addressed to differe nt elements involved in the culture.
105 First of all, the total amount of proteins delivered after 21 days from PLGA microparticles embedded in the matrix (300 Âµ g per well of a 12 wells plate) was significantly lower than the concentration of proteins corresp onding to the amount of hPM optimized in our laboratory in previous works ( 1 mg per well of a 12 wells plate). The solution could be to increase the amount of microparticles embedded in the matrix, to get to a concentration of hPM comparable with the on e used with pure hPM. However, two main issues could be related with this. In the first place, increasing the amount of microparticles in the matrix will risk to increase the thickness of the construct, affecting the viability of cells. Secondarily, if the thickness is maintained low, the higher amount of microparticles embedded in the matrix could affect cells migration and angiogenic response. The material used for the angiogenic assay is another important parameter which could influence HUVECs response t o hPM pro angiogenic effect. Although Alginate has been recently evaluated as a useful material for 3D culture and specifically for neo vascularization purposes 101 , contrasting opinion have been given about his use to mimic E CM mainly because of its different composition from the components of physiological ECM and for the limited interaction with mammalian cells 99 . As discussed in several reviews 76 112 , different materials are available to develop in vitro angiogenic assay and a comparison between outcomes of the experiments performed in this work using other matrix components (e.g. collagen, fibrin, laminin) could be useful to gain a bett er understanding of the role played by the matrix on HUVECs response to hPM.
106 Figure 6 1 . Optimization of 3D culture matrix preparation : HUVECs cutured on the top of Alginate matrix(A,B) , HUVECs embedded in Alginate homogeneous matrix (C,D) and HUVECs embedded in Alginate beads (E,F) were compared at 2 time points (5 and 10 days) to assess the response of HUVEC to the different conditions.
107 Figure 6 2 . 3D Alginate based hydrogel culture system: 7,14, 21 days (an giogenic assay): Evaluat ion of the differ ence in HUVECs long term response to hPM controlled release; Three condition compared at time points 7,14,21 days. From top: HUVECs in Alginate based matrix with PLGA microparticles embedded; HUVECs in Alginate based matrix with pre hPM added; HUVECs in A lginate based matrix.
108 Figure 6 3 . 3D Alginate based hydrogel culture system: 2,4,6,8,10 days (a ngiogenic assay): Comparison of short term response of HUVECs between pure hPM (second column) and controlled release from PLGA microparticles (first column) e valuate at frequent time points to estimate the dynamic of HUVECs angiogenic response to hPM .
109 CHAPTER 7 CONCLUSION AND FUTURE WORKS Summary The work presented in this dissertation focuses on understanding and modulating cellular response to the controlled rel e ase of a pro angiogenic mixture, referred as human Placental Matrix ( hPM ). Specifically, these studies have assessed a suitable encapsulation method of hPM using PLGA mcroparticles to achieve a controlled release in 3D cell culture. The first goal wa s the development and optimization of a proper protocol for the encapsulation of the heterogeneous mixture. The second goal was to evaluate the angiogenic effect of the controlled release of hPM developing an angiogenic assay using Alginate based 3D cultur e of Human Umbilical Vein Endothelial Cells (HUVECs). I n Chapter 4 , investigations aimed to evaluate the angiogenic effect of hPM on HUVECs 3 D culture. The prevalent role proteins play in determine the angiogenic effect of hPM was confirmed from the evalua tion of cellular response to hPM in which denaturation of the proteins had been induced compared to the response to normally thawed and warmed up hPM . The quantification of angiogenesis over time at specific time points enhanced changes in the network form ation: an increase in stability was registered during the first 3 days, followed by an initial degradation of the network starting from 5 days after the first day of seeding onto hPM. In light of this, experiments comparing different conditions of hPM inoc ulation in the cells culture confirmed an increased stability of the network for cells that received hPM at multiple regular time points. These findings, suggested that the use of a controlled release method to sustain hPM release
110 over an extended period o f time could improve the stability and maintenance of the network. In Chapter 5, starting from a standard protocol suitable for a single protein encapsulation, a proper encapsulation for hPM was optimized. Protein solution composition, duration of the homo genization time, secondary emulsion homogenization were investigated as key point to achieve the suitable size, morphologic feature and associate protein release suitable for our aims. After different protocols were investigated, the optimization process r esulted in PLGA microparticles with a mean diameter size of 91.82 Âµm, an encapsulation efficiency of 75% , and a complete releas e profile over 30 days. An overall qualitative assessment of protein encapsulation performed using SDS page analysis showed a uni form encapsulation of the heterogeneous mixture, without a selective encapsulation for specific proteins. In Chapter 6, an angiogenic assay was prepared, by embedding hPM loaded PLGA microparticles in an Alginate based matrix, to evaluate HUVECs response t o the controlled release of hPM. An optimization of the procedure for matrix preparation and for cell density and hPM amount was performed. From a long term study, evaluating cells response at three specific time points (7, 14, and 21 days) the formation o f an initial angiogenic structure could be noticed within the gel with hPM loaded PLGA microparticles embedded. Oppositely, the angiogenic network which could be noticed at day 7 within the gel with pure hPM appeared to be degraded at day 14 and 21. In or der to compare the different changes in network formation between pure and controlled delivered hPM, an evaluation at more frequent time points was performed. It showed the most mature angiogenic network to be formed around day 6 for the gel with
111 pure hPM mixed, followed by a degradation phase. A constant increase in network stability could be noticed when hPM was controlled released from PLGA microparticles. These studies have detailed a proper method for hPM encapsulation and, even if several parameters required to be optimized to improve stability and maturity of the angiogenic network formed within the gel, initial stages of angiogenic structures formation could be observed with the Alginate based angiogenic assay. Overall, the work presented in this di ssertation has confirmed the usefulness of a controlled release method to sustain the formation and maintenance of an endothelial cells capillary network. Future W orks Characterization of the protein released from PLGA microparticles Human Placental Matrix represents a mixture of several proteins with completely different properties. This could affect the associate release, due to protein protein and protein polymer interaction. A SDS page analysis on the supernatant collected at the different time points d uring the in vitro release studies, could give significant information about the specific types of proteins released over time and help understanding cells response to controlled released hPM. Optimization of matrix composition for the angiogenic assay In this study, an Alginate based hydrogel was used to develop the angiogenic assay. Even if alginate has been defined as a promising material as synthetic ECM both in the specific application of on vitro 3D angiogenetic assay and in vivo as scaffold for cells embedding , its properties in term of degradation rate and chemical composition are very different compared with components of the physiological ECM. An evaluation of the same angiogenic assay performed in this study using other materials to realize
112 the EC M environment (e.g. collagen, fibrin, laminin) could be useful to understand the effect of cell matrix interaction in improving and properly guide the formation and maintenance of an angiogenic network. Realization of a smart controlled release using iron oxide nanoparticles (IO NPs) The increased stability of the angiogenic network using multiple inoculations at discrete time points and the successful encapsulation of hPM in PLGA MPs, suggested the idea to realize a smart controlled release method able to enhance and customize the delivery of the pro angiogenic mixture. Iron Oxide nanoparticles (IO NPs) have recently attracted attention from the research for biomedical applications of smart controlled release methods, thanks to their biocompatibility and su perparamagnetic properties. When an external alternating magnetic field is applied, each NPs is subjected to a force that leads to an acceleration in the direction of increasing field strength, inducing change s in the local magnetic field. For frequencies between 100 kHz and several MHz, NPs m agnetic dipole cannot follow external magnetic field changes without a t ime lag . This leads to loss of magnetic energy, heating up the NPs a nd the surrounding environment 113 . A possible f uture development of this work, could be to embed IO NPs in PLGA MPs, and use the described induced hyperthermia to increase the porosity of the temperature sensitive polymer (PLGA), thus enhancing hPM release 74 , 114 . This could realize a customized release profile with discrete spikes, comparable with multiple inoculations described in Chapter 4, possibly improving angiogenic network maintenance and stability.
113 LIST OF REFERENCES 1. Langer, R. & Vacanti, J. P. Tissue engineering. Science 260, 920 926 (1993). 2. MacArthur, B. D. & Oreffo, R. O. C. Bridging the gap. Nature 433, (2005). 3. Francis, K. & Palsson, B. O. Effective intercellular communication distances are determined by the relative time constants for cyto/chemokine secretion Proceedings of the National Academy of Sciences 94, 12258 12262 (1997). 4. Chu, H. & Wang, Y. Therapeutic angiogenesis: controlled delivery of angiogenic factors. The r Deliv 3, (2012). 5. Xie, J, Wang H, Wang Y, Ren F et al. Induction of angiogenesis by controlled delivery of vascular endothelial growth factor using nanoparticles. Cardiovasc Ther 31, e12 18 (2013). 6. Chung, J. & Shum Tim, D. Neovascularization in Tiss ue Engineering. Cells 1, 1246 1260 (2012). 7. Laschke, M. W . , Harder Y . , Amon M . , Martin I . et al. Angiogenesis in tissue engineering: breathing life into constructed tissue substitutes. Tissue Eng 12, 2093 2104 (2006). 8. Hughes, C. S., Postovit, L. M. & Lajoie, G. A. Matrigel: a complex protein mixture required for optimal growth of cell culture. Proteomics 10, 1886 1890 (2010). 9. Moore, M. C. Modulation of nutrient deficiences occuring in engineered ex vivo tissue scaffolds. (2013). 10. Si mÃ³n Yarza, T, Formiga F . R . , Tamayo E, Pelacho B. et al. PEGylated PLGA microparticles containing VEGF for long term drug delivery. International Journal of Pharmaceutics 440, 13 18 (2013). 11. Kirby G . T . S . , White L . J . , Rahman C . V . , Cox H . C . et al. PLGA Based Micro particles for the Sustained Release of BMP 2. Polymers (2011). doi:3(1):571 586. 12. Wang, Y. , Liu X.C., Zhao J., Kong X.R. et al. Degradable PLGA scaffolds with basic fibroblast growth factor: experimental studies in myocardial revascularization. Tex Heart Inst J 36, (2009). 13. Richardson, T. P., Peters, M. C., Ennett, A. B. & Mooney, D. J. Polymeric system for dual growth factor delivery. Nat Biotech 19, 1029 1034 (2001). 14. RomÃ¡n, B. S. , Irache J.M., GÃ³mez S., Tsapis N. et al. Co encapsulatio n of an antigen and CpG oligonucleotides into PLGA microparticles by TROMS technology. European Journal of Pharmaceutics and Biopharmaceutics 70, 98 108 (2008).
114 15. Acharya, A. P., Lewis, J. S. & Keselowsky, B. G. Combinatorial co encapsulation of hydro phobic molecules in poly(lactide co glycolide) microparticles. Biomaterials 34, 3422 3430 (2013). 16. Adair, T. H. Angiogenesis . (Morgan & Claypool Publishers, 2010). at 17. Carmeliet, P. Angiogenesis i n health and disease. Nat Med 9, 653 660 (2003). 18. Sullivan, D. C. & Bicknell, R. New molecular pathways in angiogenesis. Br J Cancer 89, 228 231 (2003). 19. Karamysheva, A. F. Mechanisms of angiogenesis. Biochemistry (Mosc) 73, 751 762 (2008). 20. De Smet, F., Segura, I., De Bock, K., Hohensinner, P. J. & Carmeliet, P. Mechanisms of vessel branching: filopodia on endothelial tip cells lead the way. Arterioscler Thromb Vasc Biol 29, 639 649 (2009). 21. Prior, B. M., Yang, H. T. & Terjung, R. L . What makes vessels grow with exercise training? J Appl Physiol (1985) 97, 1119 1128 (2004). 22. Largo , R.A., Ramakrishnan V., Ehrbar , M., Ziogas , A . et al. Angiogenesis and Vascularity for Tissue Engineering Applications, Regenerative Medicine and Ti ssue Engineering Cells and Biomaterials, Prof. Daniel Eberli. InTech (2011). doi:10.5772/25141 23. Distler, J. H. W. , Hirth, A., Kurowska Stolarska, M., Gay, R.E., et al. Angiogenic and angiostatic factors in the molecular control of angiogenesis. Q J Nucl Med 47, 149 161 (2003). 24. Ng, E. W. M. & Adamis, A. P. Targeting angiogenesis, the underlying disorder in neovascular age related macular degeneration. Can J Ophthalmol 40, 352 368 (2005). 25. Nature 458, 686 687 (2009). 2 6. Iruela Arispe, M. & Dvorak, H. Angiogenesis: a dynamic balance of stimulators and inhibitors. Thromb Haemost 78, 672 677 (1997). 27. Carmeliet, P. & Jain, R. K. Molecular mechanisms and clinical applications of angiog enesis. Nature 473, (2011). 28. Bergers, G. & Hanahan, D. Modes of resistance to anti angiogenic therapy. Nat Rev Cancer 8, 592 603 (2008). 29. Cao, R. , BrÃ¥kenhielm, E., Pawliuk, R., Wariaro, D., et al. Angiogenic synergism, vascular stability and im provement of hind limb ischemia by a combination of PDGF BB and FGF 2. Nat Med 9, 604 613 (2003).
115 30. Stegmann, T. J. New approaches to coronary heart disease: induction of neovascularisation by growth factors. BioDrugs 11, 301 308 (1999). 31. Paul, D., Lipton, A. & Klinger, I. Serum factor requirements of normal and simian virus 40 transformed 3T3 mouse fibroplasts. Proc Natl Acad Sci U S A 68, 645 652 (1971). 32. Hill, J. J. , Tremblay, T.L., Cantin, C., O'Connor McCourt, M. et al. Glycoproteomic analysis of two mouse mammary cell lines during transforming growth factor (TGF) beta induced epithelial to mesenchymal transition. Proteome Sci 7, (2009). 33. Gilbert, P. S. & Gilbert, S. F. Developmental Biology (Loose Leaf) . (Sinauer Associates Incor porated, 2010). at 34. Nyberg, P., Xie, L. & Kalluri, R. Endogenous inhibitors of angiogenesis. Cancer Res 65, 3967 3979 (2005). 35. , Holmgren, L., Shing, Y., Chen, C., et al. Angiostati n: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 79, 315 328 (1994). 36. Wilcke, I. , Lohmeyer, J.A., Liu, S., Condurache, A., et al. VEGF(165) and bFGF protein based therapy in a slow release s ystem to improve angiogenesis in a bioartificial dermal substitute in vitro and in vivo. Langenbecks Arch Surg 392, 305 314 (2007). 37. Zacchigna, S., Tasciotti, E., Kusmic, C., Arsic, N. et al. In vivo imaging shows abnormal function of vascular endoth elial growth factor induced vasculature. Hum Gene Ther 18, 515 524 (2007). 38. Norrby, K. In vivo models of angiogenesis. J Cell Mol Med 10, (2006). 39. Jansen, J. A., Vehof, J.W., RuhÃ©, P.Q., Kroeze Deutman, H. et al. Growth factor loaded scaffolds for bone engineering. J Control Release 101, 127 136 (2005). 40. Des Rieux, A. , Ucakar, B., Mupendwa, B.P., Colau, D. et al. 3D systems delivering VEGF to promote angiogenesis for tissue engineering. J Control Release 150, 272 278 (2011). 41. Jang, J . H., Rives, C. B. & Shea, L. D. Plasmid delivery in vivo from porous tissue engineering scaffolds: transgene expression and cellular transfection. Mol Ther 12, 475 483 (2005). 42. Heyde, M., Partridge, K.A., Oreffo, R.O., Howdle, S.M., et al. Gene ther apy used for tissue engineering applications. J Pharm Pharmacol 59, 329 350 (2007). 43. Yamashita, J. Itoh, H., Hirashima, M., Ogawa, M. et al. Flk1 positive cells derived from embryonic stem cells serve as vascular progenitors. Nature 408, 92 96 (2000) .
116 44. Kaushal, S. , Amiel, G.E., Guleserian, K.J., Shapira, O.M. et al. Functional small diameter neovessels created using endothelial progenitor cells expanded ex vivo. Nat Med 7, 1035 1040 (2001). 45. Saif, J. , Schwarz, T.M., Chau, D.Y., Henstock, J . et al. Combination of injectable multiple growth factor releasing scaffolds and cell therapy as an advanced modality to enhance tissue neovascularization. Arterioscler Thromb Vasc Biol 30, 1897 1904 (2010). 46. Staton, C. A., Reed, M. W. R. & Brown, N . J. A critical analysis of current in vitro and in vivo angiogenesis assays. Int J Exp Pathol 90, (2009). 47. Jain, R. K., Schlenger, K., Hockel, M. & Yuan, F. Quantitative angiogenesis assays: progress and problems. Nat Med 3, 1203 1208 (1997). 48. Vailhe, B., Vittet, D. & Feige, J. J. In Vitro Models of Vasculogenesis and Angiogenesis. Lab Invest 81, 439 452 49. Matikainen, T. & Laine, J. Placenta an alternative source of stem cells. Toxicology and Applied Pharmacology 207, 544 549 (2005). 50. KH, L. Anti inflammatory and analgesic effects of human placenta extract. 25, (2011). 51. Kawakatsu, M., Urata, Y., Goto, S., Ono, Y. & Li, T. S. Placental extract protects bone marrow derived stem/progenitor cells against radiation injury through ant i inflammatory activity. J Radiat Res 54, 268 276 (2013). 52. Manochantr, S. , U pratya, Y., Kheolamai, P., Rojphisan, S. et al. Immunosuppressive properties of mesenchymal stromal cells derived from Intern Med J 43, 430 9 (2013). 53. Hopper, R. A., Woodhouse, K. & Semple, J. L. Acellularization of human placenta with preservation of the basement membrane: a potential matrix for tissue engineering. Annals of plastic surgery 51, 598 602 (2003). 54. Ilancheran, S., Moodley, Y. & Manuelpillai, U. Human fetal membranes: a source of stem cells for tissue regeneration and repair? Placenta 30, 2 10 (2009). 55. Athanassiades, A. & Lala, P. K. Role of placenta growth factor (PIGF) in human extravillous trophoblast proliferation, migration and invasiveness. Placenta 19, 465 473 (1998). 56. Wooding, P. & Burton, G. Comparative Placentation: Structures, Functions and Evolution . (Springer Berlin Heidelberg, 2008). 57. Wang, Y. & Zhao, S. Vascular Biolo gy of the Placenta. Chapter 6, Vasculogenesis and Angiogenesis of Human Placenta. (San Rafael (CA): Morgan & Claypool Life Sciences, 2010).
117 58. Charnock Jones, D. S. Aspects of human fetoplacental vasculogenesis and angiogenesis. I. Molecular regulation . 25, (Placenta, 2004). 59. Gao, X. & Xu, Z. Mechanisms of action of angiogenin. Acta Biochim Biophys Sin (Shanghai) 40, 619 624 (2008). 60. Pavlov, N. , Hatzi, E., Bassaglia, Y., Frendo, J.L., et al. Angiogenin distribution in human term placenta, an d expression by cultured trophoblastic cells. Angiogenesis 6, 317 330 (2003). 61. Rajashekhar, G., Loganath, A., Roy, A. C., Chong, S. S. & Wong, Y. C. Extracellular matrix dependent regulation of angiogenin expression in human placenta. J Cell Biochem 96, (2005). 62. Gupta, S. P. On the use of placental extracts. Indian J Ophthalmol 3, 1:61 (1953). 63. Oh, E. J., Kim, T. K., Shin, J. H., Choi, J. H. & Chung, H. Y. Biologic filler using human fibroblasts and placenta extracts. J Craniofac Surg 22, 1557 1560 (2011). 64. De, D., Chakraborty, P. D. & Bhattacharyya, D. Regulation of trypsin activity by peptide fraction of an aqueous extract of human placenta used as wound healer. J Cell Physiol 226, 2033 2040 (2011). 65. Hong, J. W., Lee, W. J., Hahn, S. B., Kim , B. J. & Lew, D. H. The effect of human placenta extract in a wound healing model. Ann Plast Surg 65, (2010). 66. Jung, J. , Lee, H.J., Lee, J.M., Na, K.H., et al. Placenta extract promote liver regeneration in CCl4 injured liver rat model. Int Immunopharm acol 11, 976 984 (2011). 67. Shin, K. S., Lee, H. J., Jung, J., Cha, D. H. & Kim, G. J. Culture and in vitro hepatogenic differentiation of placenta derived stem cells, using placental extract as an alternative to serum. Cell Prolif 43, 435 444 (2010). 68 . Frolik, C. Purification and initial characterization of a type beta transforming growth factor from human placenta. (Proceedings of the National Academy of Sciences of the United States of America, 1983). 69. Stromberg, K. Human term placenta contains tr ansforming growth factors. 106, (Biochemical and Biophysical Research Communications, 1982). 70. Aitken, A. & Learmonth, M. in The Protein Protocols Handbook (Walker, J.) 3 6 (Humana Press, 2002). at 71. Cohen, S., Yoshioka, T., Lucarelli, M., Hwang, L. & Langer, R. Controlled Delivery Systems for Proteins Based on Poly(Lactic/Glycolic Acid) Microspheres. Pharm Res 8, 713 720 (1991).
118 72. Acharya, G. , Shin, C.S., Vedantham, K., McDermott, M., et al. A study of dr ug release from homogeneous PLGA microstructures. Journal of Controlled Release 146, 201 206 (2010). 73. Igartua, M., HernÃ¡ndez , R.M, Esquisabel , A., GascÃ³n , A.R et al. Stab ility of BSA encapsulated into PLGA microspheres using PAGE and capillary electroph oresis. International Journal of Pharmaceutics 169, 45 54 (1998). 74. Alvarez Lorenzo, C., Concheiro, A., Schneider, H. J., Royal Society of Chemistry (Great Britain) & Shahinpoor, M. Smart Materials for Drug Delivery . (Royal Society of Chem istry, 2013). at 75. Ariga, K., Lvov, Y. M., Kawakami, K., Ji, Q. & Hill, J. P. Layer by layer self assembled shells for drug delivery. Advanced Drug Delivery Reviews 63, 762 771 (2011). 76. Zisch, A. H., Lut olf, M. P. & Hubbell, J. A. Biopolymeric delivery matrices for angiogenic growth factors. Cardiovasc Pathol 12, (2003). 77. Oliveira, M. B. & Mano, J. F. Polymer based microparticles in tissue engineering and regenerative medicine. Biotechnol Prog 27, (201 1). 78. Berkland, C., King, M., Cox, A., Kim, K. & Pack, D. W. Precise control of PLG microsphere size provides enhanced control of drug release rate. Journal of Controlled Release 82, 137 147 (2002). 79. Formiga, F. R., Pelacho, B., Garbayo, E., Imbuluzqu eta, I. et al. Controlled delivery of fibroblast growth factor 1 and neuregulin 1 from biodegradable microparticles promotes cardiac repair in a rat myocardial infarction model through activation of endogenous regeneration. Journal of Controlled Release 17 3, 132 139 (2014). 80. Patel, Z. S., Ueda, H., Yamamoto, M., Tabata, Y. & Mikos, A. G. In vitro and in vivo release of vascular endothelial growth factor from gelatin microparticles and biodegradable composite scaffolds. Pharm Res 25, 2370 2378 (2008). 81. Chau, D. Y. S., Agashi, K. & Shakesheff, K. M. Microparticles as tissue engineering scaffolds: manufacture, modification and manipulation. Materials Science and Technology 24, 1031 1044 (2008). 82. Kreft, O., Prevot, M., MÃ¶hwald, H. & Sukhorukov, G. B. Sh ell in Shell Microcapsules: A Novel Tool for Integrated, Spatially Confined Enzymatic Reactions. Angewandte Chemie International Edition 46, 5605 5608 (2007). 83. Ye, M., Kim, S. & Park, K. Issues in long term protein delivery using biodegradable micropart icles. J Control Release 146, 241 260 (2010). 84. Mun, S. , Choi, Y., Rho, S.J., Kang, C.G. et al. Preparation and characterization of water/oil/water emulsions stabilized by polyglycerol polyricinoleate and whey protein isolate. J Food Sci 75, E116 125 (20 10).
119 85. Balmayor, E. R., Azevedo, H. S. & Reis, R. L. Controlled delivery systems: from pharmaceuticals to cells and genes. Pharm Res 28, 1241 1258 (2011). 86. Burdick, J. A. & Mauck, R. L. Biomaterials for Tissue Engineering Applications: A Review of the Past and Future Trends . (Springer, 2010). at 87. Jay, S. M. & Saltzman, W. M. Controlled delivery of VEGF via modulation of alginate microparticle ionic crosslinking. Journal of Controlled Release 134, 26 34 (2009). 88. Chan, O. C. M., So, K. F. & Chan, B. P. Fabrication of nano fibrous collagen microspheres for protein delivery and effects of photochemical crosslinking on release kinetics. Journal of Controlled Release 129, 135 143 (2008). 89. Jay, S. M., Sh epherd, B. R., Bertram, J. P., Pober, J. S. & Saltzman, W. M. Engineering of multifunctional gels integrating highly efficient growth factor delivery with endothelial cell transplantation. FASEB J 22, 2949 2956 (2008). 90. Chan, B. P. , Hui, T.Y ., Chan, O.C ., So, K.F . et al. Photochemical cross linking for collagen based scaffolds: a study on optical properties, mechanical properties, stability, and hematocompatibility. Tissue Eng 13, (2007). 91. Lassalle, V. & Ferreira, M. L. PLA nano and microparticles fo r drug delivery: an overview of the methods of preparation. Macromol Biosci 7, 767 783 (2007). 92. Lee, W. L. , She, Y.C., Widjaja, E., Chong, H.C. et al. Fabrication and drug release study of double layered microparticles of various sizes. J Pharm Sci 101, 2787 2797 (2012). 93. Lee, W. L., Shi, W. X., Low, Z. Y., Li, S. & Loo, S. C. J. Modeling of drug release from biodegradable triple layered microparticles. J Biomed Mater Res A 100, 3353 3362 (2012). 94. Giteau, A., Venier Julienne, M. C., Aubert PouÃ«ssel , A. & Benoit, J. P. How to achieve sustained and complete protein release from PLGA based microparticles? International Journal of Pharmaceutics 350, 14 26 (2008). 95. Klose, D., Siepmann, F., Elkharraz, K. & Siepmann, J. PLGA based drug delivery systems: Importance of the type of drug and device geometry. International Journal of Pharmaceutics 354, 95 103 (2008). 96. Santo, V. E., Gomes, M. E., Mano, J. F. & Reis, R. L. Controlled release strategies for bone, cartilage, and osteochondral engineering -Part II: challenges on the evolution from single to multiple bioactive factor delivery. Tissue Eng Part B Rev 19, 327 352 (2013). 97. Whelan, M. C. & Senger, D. R. Collagen I initiates endothelial cell morphogenesis by inducing actin polymerization through sup pression of cyclic AMP and protein kinase A. J Biol Chem 278, 327 334 (2003).
120 98. Hielscher, A. C. & Gerecht, S. Engineering approaches for investigating tumor angiogenesis: exploiting the role of the extracellular matrix. Cancer Res 72, 6089 6096 (2012). 99. Rowley, J. A., Madlambayan, G. & Mooney, D. J. Alginate hydrogels as synthetic extracellular matrix materials. Biomaterials 20, 45 53 (1999). 100. Sun, J. & Tan, H. Alginate based biomaterials for regenerative medicine applications. Materials 6, (201 3). 101. Gandhi, J. K., Opara, E. C. & Brey, E. M. Alginate based strategies for therapeutic vascularization. Ther Deliv 4, 327 341 (2013). 102. Jaffe, E. A., Nachman, R. L., Becker, C. G. & Minick, C. R. Culture of human endothelial cells derived from umb ilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52, 2745 2756 (1973). 103. Kibbey, M. Maintenance of the EHS sarcoma and Matrigel preparation. Journal of Tissue Culture Methods 16, 227 230 (1994). 104. Brey, E. M. , King, T.W., Johnston, C., McIntire, L.V., et al. A technique for quantitative three dimensional analysis of microvascular structure. Microvasc Res 63, 279 294 (2002). 105. Nguyen, M., Shing, Y. & Folkman, J. Quantitation of angiogenesis and antiangiogenesis in the chick embryo chorioallantoic membrane. Microvasc Res 47, (1994). 106. Guidolin, D., Vacca, A., Nussdorfer, G. G. & Ribatti, D. A new image analysis method based on topological and fractal parameters to evaluate the angiostatic activity of docetaxel by using the Matrigel assay in vitro. Microvascular Research 67, 117 124 (2004). 107. Drake, C. & Little, C. in Vascular Morphogenesis: In vivo , In vitro , In Mente (Little, C., Mironov, V. & Sage, E. H.) 3 19 (BirkhÃ¤user Boston, 1996). at 108. Ravi, S. , Peh, K.K., Darwis, Y., Murthy, B.K. et al. Development and characterization of polymeric microspheres for controlled release protein loaded drug delivery system . 70, (2008). 109. Bouyer, E., Mekhloufi, G., Rosilio , V., Grossiord, J. L. & Agnely, F. Proteins, polysaccharides, and their complexes used as stabilizers for emulsions: alternatives to synthetic surfactants in the pharmaceutical field? Int J Pharm 436, 359 378 (2012). 110. Ashton, R. S., Banerjee, A., Puny ani, S., Schaffer, D. V. & Kane, R. S. Scaffolds based on degradable alginate hydrogels and poly(lactide co glycolide) microspheres for stem cell culture. Biomaterials 28, 5518 5525 (2007).
121 111. Draget, K. I., Myhre, S., SkjÃ¥kBr k, G. & Ã˜stgaard, K. Regeneration, Cultivation and Differentiation of Plant Protoplasts Immobilized in Ca alginate Beads. Journal of Plant Physiology 132, 552 556 (1988). 112. Goodwin, A. M. In vitro assays of angiogenesis for assessment of angiogenic an d anti angiogenic agents. Microvascular Research 74, 172 183 (2007). 113. Hofmann Amtenbrink, M., von Rechenberg, B. & Hofmann, H. Superparamagnetic nanoparticles for biomedical applications . (Transworld Research Network Kerala,, India, 2009). 114. Kumar, C. S. S. R. & Mohammad, F. Magnetic nanomaterials for hyperthermia based therapy and controlled drug delivery. Advanced Drug Delivery Reviews 63, 789 808 (2011).
122 BIOGRAPHICAL SKETCH Sarah Tonello was born in Brescia, Italy in 1990. She graduated cum la ude from Politecnico di Milano in 2012 with a Bachelor of Scie nce in Biomedical Engineering. She continued her postgraduate study at Politecnico di Milano, focusing on her work on cell and tissue engineering and biotechnology in Biomedical Engineering. In summer 2013, a s a recipient of the A tlantis CRISP dual degree grant, she first took part to a three week Summer School in Strasbourg and then she moved to the University of Florida to join the J. Crayton Pruitt Family Depart ment of Biomedical Engineering to complete her Master of Science in Biomedical Engineering under the supervision of Dr. Peter McFetridge. Upon competition of her studies at both Politecnico di Milano and University of Florida , she plans to pursue her career in the research field.