Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels

Cell-laden micropatterns using selfassembled cellECM microtissues in soft pectin hydrogels, for skin regeneration Fábio Jorge OliveiraRangel Mestrado em Biologia Molecular e Celular Departamento de Biologia 2014-2015 Orientador Pedro Granja, Ph.D, Professor Auxiliar, FEUP Coorientador Aureliana Sousa, Ph.D, INEB Todas as correções determinadas pelo júri, e só essas, foram efetuadas. O Presidente do Júri, Porto, ______/______/_________ FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels i Dissertação de candidatura ao grau de Mestre em Biologia Celular e Molecular submetida à Faculdade de Ciências da Universidade do Porto. O presente trabalho foi desenvolvido sob a orientação científica do Professor Doutor Pedro Granja, com coorientação pela Doutora Aureliana Sousa, no INEB (Instituto Nacional de Engenharia Biomédica), I3S (Instituto de Investigação e Inovação em Saúde). Dissertation for applying to a Master’s Degree in Cell and Molecular Biology, submitted to the Faculty of Sciences of the University of Porto. The present work was developed under the scientific supervision of Professor Bruno Silva-Santos, co-supervised by Doutora Auraliana Filipa and was done at INEB (Instituto Nacional de Engenharia Biomédica), I3S (Instituto de Investigação e Inovação em Saúde FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels ii Aknowledgements First of all, I would like to thank Professor Doutor Pedro Granja. His passion for this work drew me to the field with the feeling that I always belonged there. Thank you for pushing me to do my best. To the person that contributed the most to my growth, Doutora Aureliana Sousa, thank you for all the knowledge you passed on to me on both the scientifical and personal fields. Thank you for allowing me to fall on my face. To my collegues in INEB, who welcomed me into the family and helped me whenever I needed. To my friends, Helena Brigas and Miguel Rocha, who stayed with me and shared their joys and sorows. To Romeu Catarino, Daniela Gonzáles, André Resende, Bárbara Andrade, João Teixeira, André Silva, André Resende e Paulo Neves, who got me out of my house so I can clear my head. Finally, and foremost, to my family. Thanks to my parents Fatima and Jorge Rangel to always supported me at the economic, emotional and intelectual levels.To my girlfriend, Marta Monteiro who, even when away, gave me strenght to carry on. To my sister Catarina and Daniel who bugged me when I needed to. To Sr. José and Dona Isabel who offered me home and distraction when I needed. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels iii Abstract Advanced skin regeneration therapies can combine biomaterials, cells, growth factors and advanced biomanufacturing techniques for the fabrication of constructs that will ultimately mimic native skin anatomy. Regardless of the specific tissue-engineering approach for in vitro artificial skin substitute production, to engineer functional skin, the formation of an efficient vascular network is required. Aiming to develop a strategy to improve constructs microvascularization with fibroblasts support endothelial cells in the formation of self-assembled vascular structures, this study allowed the dissection of human umbilical vein endothelial cells (HUVEC) and neonatal human dermal fibroblasts (NHDFs) behavior in a 3D microenvironment. We addressed for the first time the effect of several culture parameters on cells behavior when embedded on RGD-grafted soft pectin hydrogels. Conditions such as media composition, cell density, cell type to type ratio and polymer concentration were optimized on standard 2D culture conditions. The results obtained allowed us to choose the best conditions to proceed into a 3D experimental setup. A 3:1 ratio of M199 to DMEM media was selected for HUVEC:NHDFs co-cultures and we also determined that low HUVEC to NHDFs ratios, in 2D environments led to NHDFs spreading in detriment of HUVEC proliferation while higher ratios sustained a controlled environment where HUVECs were able to grow and assemble in spider weblike structures. In a three dimensional context, Cell behavior parameters displayed better outcomes for lower hydrogel formulations (1.5% w/v) and higher cell densities (1.5x107 cells.mL-1).. Fibroblasts formed spheroidss and contracted the matrix, while maintaining the metabolic activity, in a matrix and cell density-dependent way, with 1.5% (w/v) pectin hydrogels embedded with 1x107cells.mL-1 demonstrating microtissues formation. Based on combination of NHDFs and HUVECs, a cocuture systems were developed in soft pectin hydrogel matrices. Within these, HUVEC survival was increased, and fibroiblast spheroids formation was observed. Although further investigation is needed, we developed a a three-dimensional co-culture system in RGD-grafted soft pectin hydrogel in which fibroblasts support endothelial cells, and established this techniques as a promising strategy for in vitro microvascularization towards skin regeneration therapies. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels iv Resumo As terapias avançadas de regeneração da pele combinam biomateriais, células, fatores de crescimento e técnicas avançadas de biofabrico de estruturas que, em última análise, visam mimetizar a anatomia da pele. Independentemente da abordagem in vitro usada em engenharia de tecidos para regeneração de pele artificial, para produzir uma pele funcional, é necessária a formação de uma rede vascular eficiente. Com o objetivo de desenvolver uma estratégia para melhorar a microvascularização in vitro, este estudo visou dissecar o comportamento de células endoteliais da veia umbilical humana (HUVECs) e fibroblastos dérmicos humanos neonatais (HDFns) num microambiente 3D. Abordamos, pela primeira vez, os efeitos de vários parâmetros de cultura no comportamento das células de em cultura em matrizes macias de hidrogéis de pectina modificados com RGD. Condições como a composição do meio, a densidade celular e a proporção entre os tipos de células foram optimizadas em condições de cultura 2D padrão. Os resultados obtidos permitiram-nos escolher as melhores condições para proceder às experiencias em ambientes 3D. Um meio composto por um rácio de 3:1 de M199 para DMEM, foi selecionado para a a cocultura de HUVEC:HDFns. Determinamos também que, em condições de cultura 2D, um baixo rácio de HUVEC para HDFns levou à proliferação de HDFns em detrimento do crescimento das HUVEC enquanto rácios mais elevados sustentaram um ambiente onde as HUVECs foram capazes de crescer e estabelecer estruturas numa formação semelhante a teias de aranha. Os parâmetros de comportamento celular sobre os quais nos debruçamos exibiram melhores resultados para formulações de hidrogéis com concentrações de pectina menores (1.5% w/v) e concentrações altas de células (1.5x107 celulas.mL-1). Os fibroblastos, demonstraramse capazes de formar esferóides e contrair a matriz, mantendo a atividade metabólica, de uma forma dependente da densidade celular e da matriz, verificando-se que, aquando do aprisionamento de 1x107 celulas.mL-1 em hidrogéis de pectina com uma concentração de 1.5% (w/v), ocorreu a formação de microtecidos. Com base na combinação de NHDFs e HUVECs, foram desenvolvidos dois sistemas de cocultura em hidrogéis de pectina macia. Nestes sistemas, a sobrevivência das HUVECs foi aumentada e a formação de esferóides foi observada nos fibroblastos. Embora seja necessária uma investigação mais aprofundada, desenvolvemos um sistema de cocultura tridimensional em hidrogéis macios de pectina transformada com RGD no qual os fibroblastos suportam as células endoteliais. A técnica neste trabalho FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels v estabelecida apresenta-se assim como uma estratégia promissora para a a microvascularização in vitro tendo em vista terapias de regeneração da pele. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels vi Table of contents Aknowledgements......................................................................................................................... ii Abstract ........................................................................................................................................ iii Resumo ..........................................................................................................................................iv List of Abbreviations ..................................................................................................................... xii 1. Introduction .......................................................................................................................... 1 1.1. Skin ................................................................................................................................ 2 1.1.1 Skin lesions and regenerative medicine ................................................................ 3 1.2. Vascularization .............................................................................................................. 6 1.2.1 Endothelial cells............................................................................................................ 7 1.2.2 Vascularization strategies ............................................................................................ 8 1.3. Extracellular matrix ..................................................................................................... 11 1.3.1 Hydrogels ............................................................................................................. 12 1.4. Main Goals................................................................................................................... 17 2. Materials and Methods ........................................................................................................... 18 2.1. Cell Culture .................................................................................................................. 19 2.1.1 Routine passaging ............................................................................................... 19 2.1.2 Cell thawing ......................................................................................................... 20 2.1.3. Co-culture media selection ....................................................................................... 20 2.1.4. HUVECs and FBs density optimization ..................................................................... 21 2.2. Pectin hydrogel............................................................................................................ 21 2.2.1. Pectin purification ..................................................................................................... 21 2.2.2. Carbodiimide RGD-grafting ....................................................................................... 22 2.3. 3D in vitro cell characterization .................................................................................. 23 2.3.1 Characterization of HUVECs and FBs monocultures behavior within 3D RGD-grafted soft pectin hydrogels ........................................................................................................... 23 2.3.2. HUVEC and Fibroblasts 3D monocultures performance under different culture media ................................................................................................................................... 24 2.3.3. 3D HUVEC:FB co-culture in soft pectin hydrogels ..................................................... 24 2.4. Phenotype characterization ........................................................................................ 26 2.4.1. Cell metabolic activity ............................................................................................... 26 2.4.2. Total dsDNA quantification ....................................................................................... 27 FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels vii 2.4.3 2D co-culture readouts........................................................................................ 27 2.4.4. HUVECs and FBs 3D monocultures and co-culture morphology and spatial distribution .......................................................................................................................... 28 2.5. Data treatment ............................................................................................................ 29 2.5.1. Statistical analysis ...................................................................................................... 29 2.5.2. Image treatment ....................................................................................................... 30 3. Results ..................................................................................................................................... 31 3.1 Preparation of 3D biofuncional RGD-grafted pectin ......................................................... 32 3.2. Determination of 2D optimal HUVEC/FB culture media composition ............................. 33 3.3. Determination of 2D optimal in vitro HUVEC/FB ratio .................................................... 36 3.4. Analysis of HUVEC and FB monocultures’ behavior in 3D-culture ................................... 40 3.4.1 HUVEC behavioral analysis on 3D soft pectin hydrogels............................................ 41 3.4.2. FBs behavioral analysis on 3D soft pectin hydrogels FBs .......................................... 42 3.5 HUVEC:FB co-culture establishment in 3D soft pectin hydrogels ..................................... 53 3.5.1. Characterization of the influence of M 3:1 supplementation on HUVECs or FB monocultures in 3D soft pectin hydrogels .......................................................................... 53 3.5.2. Characterization of HUVEC:FB co-culture behavior in a 3D soft pectin hydrogel ..... 54 3.5.3 Micropatterning ......................................................................................................... 57 4. Discussion ................................................................................................................................ 59 4.1. 2D characterization of an HUVEC:FB co-culture .............................................................. 60 4.1.1. Characterization of HUVEC and FB 2D monocultures under under different supplementation conditions ............................................................................................... 61 4.1.2. Characterization of HUVEC:FB co-culture behavior in a 2D environment, under different seeding ratios ....................................................................................................... 62 4.2 HUVEC and FB monocultures’ behaviour in 3D-culture .................................................... 64 4.3 HUVEC:FB co-culture establishment in 3D soft pectin hydrogels ..................................... 70 4.4 3D HUVEC:FB co-culture spatial patterning: Microinjected HUVEC-laden soft pectin on a FB-ladden soft pectin bed ....................................................................................................... 73 5. Conclusions and Future Remarks ............................................................................................ 75 6. References ............................................................................................................................... 78 7. Annexes ................................................................................................................................... 97 List of Figures FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 1 1. Introduction Introduction FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 2 1.1. Skin Skin is the largest organ of the human body, representing roughly one tenth of the body mass (Metcalfe & Ferguson 2006; Groeber et al., 2011) performing very important functions besides its obvious aesthetical function. Skin performs several functions: acts as a protective barrier, preventing dehydration, limiting organism invasion by potentially noxious agents (e.g. toxins, virus, UV radiation) also by impermeabilizing the body, helps in the thermoregulation of the body, works as a cushion, among others (Metcalfe & Ferguson 2007; Yildirimer et al., 2012; Pereira et al., 2013). Figure 1. A schematic of the structure of skin. Image from Naturally Healthy Skin (http://www.naturallyhealthyskin.org/anatomy-of-the-skin/the-dermis/dermis-anatomy-of-the-skin/ The skin is composed of three layers: epidermis, dermis and hypodermis (Figure 1) (hypodermis (Groeber et al., 2011; Pereira et al., 2013). The epidermis is thin and totally cellular, mainly composed of keratinocytes but also containing other cell types, such as Langerhans cells and melanocytes. Due to the constant exposure, homeostasis is achieved by constant substitution of the environment-exposed cells by cell migration from the basal layers, which, in turn, are composed of epidermal stem cells able of self-renewal and repair (Alonso et al., 2003; Chunmeng & Tianmin, 2004; Metcalfe & Ferguson 2007; Pereira et al., 2013). In addition, the skin appendages (e.g hair, nails, sweat glands and sebaceous glands) are derived from and linked to the FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 3 epidermal layer presenting however deep projections into the dermal layer (Martin, 1997). Situated directly below the epidermis is the dermis. This layer constitutes the bulk of the skin, providing support and nourishment. It contains vascularized extracellular matrix (ECM) rich in collagen, elastin and glycosaminoglycans (GAGs), being responsible for the elasticity and mechanical integrity (Jones et al., 2002; Metcalfe & Ferguson 2007; Groeber et al., 2011). These properties are modulated by fibroblasts, the main cell type in the dermal layer and the the main source of ECM (Berthod et al., 2006). Furthermore, fibroblasts also produce remodeling enzymes, such as proteases and collagenases, playing an important role in wound healing (Ratner et al., 2004). Present in this layer, but in lesser amounts, are also endothelial cells and smooth muscle cells, composing a vascular system, mast cells, which are part of the immune system being responsible for the early recognition of pathogens and cutaneous sensory nerves that pass through dermis into the epidermal layer (Metcalfe & Ferguson 2007; Urb & Sheppard 2012; Pereira et al., 2013). The third layer, the hypodermis, is a well vascularized area mostly composed of adipose tissue, contributing for the mechanical and thermoregulatory properties of the skin as well as acting as an energy source (Metcalfe & Ferguson 2007; Yildirimer et al., 2012; Pereira et al., 2013). 1.1.1 Skin lesions and regenerative medicine Skin lesions, whether caused by physical/chemical factors (e.g. burns, lacerations, ulcers, acute wounds, surgery, among others) or by chronicle diseases are fairly common (Martin, 1997; Groeber et al., 2011). Upon injury that leads to the disruption of the structure and function of natural tissue, under certain physiological circumstances, skin displays a complex and continuous natural process, overlapping events of hemostasis, inflammation, migration, proliferation and differentiation. These occur due to a constant environmental change that exposes cells to complex molecular patterns which sets off a series of metabolic cascades, propelling the wound through the phases of healing, overlapping events of hemostasis, inflammation, migration, proliferation and differentiation (Figure 2) (Mutsaers et al., 1997; Martin, 1997; Guo & DiPietro, 2010; Häggström et al., 2010). FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 4 Figure 2. Chronological representation of the phases of wound healing. Adapted from Häggström et al., 2010. However, depending on the lesion extent, wound environmental exposure can pose a high infection risk that can lead to deeper skin damage, tissue necrosis or, ultimately, death. As such, skin lesions must be treated as a critical issue in healthcare (Zöller et al., 2014). In current medical treatments, clinical strategies rely on the use of closure materials that may act solely as a barrier while natural wound healing occurs or actively contribute for the restoration of the epidermal function while becoming incorporated into the healing wound. Nowadays, it is possible to find several solutions for skin wound treatment (Guo & DiPietro, 2010). Depending on the wound type, depth, extension and the patient, several strategies can be applied. For superficial lesions (mainly affecting the epidermis), creams and ointments are used for disinfection, cleaning, debridement or to help the wound healing process. Although still used, due to their properties their limited permanency in the human body, these solutions have been substituted for more advanced strategies (Boateng et al., 2008). Wound dressings have been widely used due to their low cost and effectiveness. This medical strategy consists in the application of natural or synthetic material over the wound protecting it from the environment. Traditional wound dressings (e.g. bandages, cotton wool, lint and gauzes), covered the wound, keeping the wound dry and preventing the entry of pathogens into the wound (Boateng et al., 2008; Pereira et al., 2013). Nowadays, accompanying the evolution in the science and technological fields, wound dressings present more advanced solutions for wound healing. Obtained from natural or synthetic sources, modern wound dressings are available as films, foams or gels (Boateng et al., 2008). Based on the concept of creating an optimal environment, which includes an exudation control allowing a moist, non-detrimental environment, effective oxygen circulation aiding the regeneration process, good adhesion to the lesion surface and low pathogen penetration, while minimizing maceration and scar formation, (Stephen-Haynes et al., 2014), several modern wound dressings were developed, as reviewed by Pereira et al. (2013). Moreover, some dressings can even act as drug delivery systems, incorporating the therapeutic agent releasing it in the wound bed (Elsner & Zilberman, 2010; Pereira et al., 2013; Boateng et al., 2015; Momoh et al., 2015). However, due to the complexity of the healing process and the wide variety of FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 5 skin wounds existent, no single dress is able to fulfill the requirements for full skin recovery. Notwithstanding the importance of the referred methods for skin regeneration therapies, in cases of severe lesions in the dermis or hypodermis, a complex treatment is required. At the present day, autografts, surgical reconstruction using the patient own skin, are the ―gold standard‖ procedure (Goldberg, 1992; MacNeil, 2007). This strategy however presents limitations depending on the lesion extension and due to the creation of additional surgical sites (Goldberg, 1992). Another solution is the use of allografts, surgical reconstruction using another patient skin. This, however, can pose complication at both ethical and medical levels, as another patient is exposed to a risk situation while also subjecting the wounded patient to a graft that can potentially carry a disease or suffer immunological rejection (Goldberg, 1992). A potential solution to this problem is to approach this from a tissue engineering–based standpoint for de novo organogenesis, using biomaterial scaffolds and a person’s own cells to grow or fabricate skin substitutes (Cuono et al., 1986; Zöller et al., 2014). To date, there are several clinically available skin substitutes, with these being divided into epidermal, dermal, and dermo-epidermal tissue-engineered constructs. As mimicking the extracellular matrix (ECM) structural integrity and function is of key importance, several strategies are revolving around collagen-based matrices (Boyd et al., 2007; Johnen et al., 2008; Cen et al., 2008). Other skin substitute biomaterials used as matrices are chitosan (Mao et al., 2003; Mohd et al., 2013), hyaluronic acid (Park et al., 2004; Wang et al., 2006), among others. Despite recent developments wound healing, the techniques and biomaterials available present significant limitation for skin regeneration. To our knowledge, at the present time, there are no models of skin substitutes that fulfill all the criteria, replicating the anatomical and physiological requirements for biological stability at epidermal and/or dermal. Additionally, available skin substitutes suffer from poor integration, scarring and lack of differentiated structures (e.g. hair and sebaceous glands), contrasting with the aesthetics of uninjured skin (Boateng et al., 2008). Advanced skin regeneration therapies already combine biomaterials, cells, growth factors and advanced biomanufacturing techniques for the fabrication of constructs that mimic skin anatomy. Recently, several methods have been developed to spatially encode local properties to 3D materials-based culture systems. These biofabrication techniques are capable of constructing micropatterned materials, with a high degree of control, by finely tuning and defining material geometries, localization of biomolecular cues, and other mechanical properties, enabling a precise control over the bulk material properties (Nichol & Khademhosseini, 2009; Nikkhah et al., 2012; Pataky et al., 2012; Culver et al., 2012). These are designated bottom up approaches and consist on the formulation of tissue building FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 6 blocks with specific microarchitectural features for modular assembly, in an attempt to replicate the heterogeneous nature of endogenous tissues and organs. Another used approach to is to use tissue engineering strategies typically that employs a ―top-down‖ These consist on seeding cells into biomaterial matrices capable of recreating biomimetic structures, exploiting the innate abilities of cells to sense their local environment through cell–cell and cell–extracellular matrix (ECM), self-assembling into complex networks (Dean et al., 2007; Seidlits et al., 2011; Maia et al., 2014). This strategy relies on the ability of the cells to reconstruct the intricate microarchitectural and functional features of natural microenvironments to achieve the desired biological effect. However, for a given skin substitute to attach promptly, a vascularized wound bed is required. Deep wounds that affect the dermal layer constitute a problem. If the skin substitute surpasses a certain thickness nutrient diffusion is limited and the vascularization process is too slow, resulting in necrosis and graft loss. As such, any tissue-engineering constructs that aims to mimic natural tissues and, ultimately, organs, must ideally conjugate all the key components – cells, extracellular matrix (ECM), and vasculature – in precise geometries (Auger et al., 2013; Battiston et al, 2014). 1.2. Vascularization From the various obstacles for tissue engineered skin substitutes, the inability of the grafts to acquire proper vascularization has been proposed as the most likely reason for deleterious effect on epidermal survival human tissue-engineered skin constructs. The inability to properly assemble a vascular structure within the graft, leads to necrosis at the tissue core, and poor survival due to ischemic injury (Rivron et al., 2008; Auger et al., 2013). Regardless of the specific tissue-engineering approach to create artificial skin any construct that involves living cells needs to fulfill the conditions in which cells are able survive and redeem their biological functions. Reconstructed tissues need to be able to access to oxygen and nutrients, as well as elimination of carbon dioxide and other cellular waste products (Folkman & Hochberg, 1973; Novosel et al., 2011; Auger et al., 2013). It is, therefore, paramount, for the successful transplantation of human tissue-engineered constructs, the formation of a vascular network. At both the stage of in vitro growth and assemble and after the patient implantation of the graft (Rivron et al., 2008; Novosel et al., 2011; Auger et al., 2013). FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 7 1.2.1 Endothelial cells Blood vessels are a multi-cellular system composed of vascular smooth muscle cells (VSMCs), fibroblasts and endothelial cells (ECs) (Ratner et al., 2004). Vascular networks, ranging from large sized vessels, as are arteries and veins, to the microsized vasculature networks formed within organs, are lined with a single layer of endothelial cells (ECs), on which one part of the surface defines the lumen while the other is in contact with a highly specialized EC, the basement membrane. In order to sustain their tubular architecture and allow a contractile behavior in these structures, ECs are enveloped by mural cells (e.g. pericytes, VSMCs). EC formation occurs mainly through mesodermal precursor’s differentiation of hemangioblasts and/or angioblasts, a critical process in embryogenesis and tumor formation (Augustin et al., 1994; Mani et al., 2008). These cells form a barrier that, due to their capacity of extravasation and high surface-to-volume ratio are capable of actively transport small molecules, macromolecules and hormones, while also performing multiple functions depending on the location and size of the blood vessel that they are lining (Ruoslahti & Rajotte, 2000, Bouis et al., 2001; Pinkney et al., 1997). As such, ECs play an important role mediating many physiological functions such as hemostasis maintenance, vasomotor tone, blood cell trafficking, permeability, proliferation, survival, and innate and adaptive immunity (Aird, 2007). There are two processes from which neovascularization can take place: angiogenesis, a process through which new blood vessels are formed from preexisting ones, and vasculogenesis, the generation of a new vascular network from endothelial progenitor cells (EPCs) in the absence of preexisting blood vessels (Luttun et al., 2002). These capillary generation events involve a complex sequence of events, which cell adhesion, migration, alignment, protease secretion, and tubule formation. Throughout these, ECs must be exposed to growth factors interaction and mechanical cues as well as cell-cell and cell-EMC interactions all of which must be precisely timed and with the correct concentrations (Yamamoto et al., 2003; Lokmic et al., 2008; Arnaoutova et al., 2009). ECs can be isolated from different endothelium. With proper specific medium supplementation, several ECs population like human umbilical vein endothelial cells (HUVECs) or human dermal microvascular endothelial cells (HDMECs) can be isolated and cultured in vitro. Although they possess several common characteristics like cellcell contact inhibition when confluent, similar morphology and identical expression of cellular markers, choosing the source of ECs is of critical issue. Due to the endothelium FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 8 heterogeneity, site specific properties of the ECs could be translated in vitro, originating different outcomes when exposed to the same factors. Throughout the years, the phenotypic heterogeneity of the endothelium has been characterized and described recurring lectin staining, immunohistochemistry, in situ hybridization and real-time intravital microscopy, being, nowadays possible to select the most appropriate EC type for each design (Boius, 2001; Aird, 2003; Aird, 2012). Among these, human umbilical vein endothelial cells (HUVECs) have been of critical importance, largely contributing for scientific knowledge breakthroughs in molecular medicine providing insights over ECs embryogenesis, angiogenesis, vasculogenesis and pathology, at both cellular and molecular levels (Nakatsu et al., 2003; Poliseno et al., 2006; Anand et al., 2010). HUVECs are easily available, free from any pathological process and they are physiologically more relevant than many established cell lines (Cooper & Sefton, 2011).Initial passages of these cells, maintain nearly all of the features of native vascular endothelial cells expressing several endothelial cell specific markers such as: von Willebrand factor a large adhesive glycoprotein that, in the blood, serves as a stabilizing factor for Factor VIII (Zanetta et al., 2000); platelet endothelial cell adhesion molecule-1 (PECAM or CD31), an endothelial specific adhesion molecule (Goldberger et al., 1994); VE-cadherin (CD144), a cadherin expressed in the tight junctions (Esser et al., 1998); and specific signaling pathways receptors markers for vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) (Esser et al., 1998; Salcedo et al., 1999). HUVECs have an average life span of 10 serial passages, time after which the cells enter senescence, tending to stop proliferation, form giant multicellular aggregates and dye (Jaffe et al.,1973). Although recovered from a major vessel, HUVECs have been proven capable of forming microvascular structures (Kenneth et al., 2006; Sorrell et al., 2007; Zheng et al., 2012). All summed up, HUVECs 3D culture presents itself as a promising strategy for in vitro microvasculature formation and characterization. To recover functional endothelial cell self-assembled into microvascular structures could presents itself as a major advance in biomanufacturing techniques forthcoming the construction of functional grafts for patient transplantation. 1.2.2 Vascularization strategies New vessel formation is essential for wound healing. As such, to culture cells under 3D conditions using a material that can mimic the ECM, and recapitulate some key aspects of the native cellular microenvironment is paramount. Although several 2D FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 9 strategies were conducted, in 1983, Montesano et al. (1983) evidenced the importance of culture ECs in a three dimensional environment. Bidimensional environments fail to mimic several cues necessary for the creation of a specific cellular organization. Nowadays, 3D cell cultures stand as essential models for the study of cell biology, as well as support matrices that can incorporate mechanical and biochemical stimuli directly conveyed by the ECM. As such, in vitro 3D microvascularization is highly dependent on the composition and properties of biomaterial matrix along with the presence of precisely timed delivery of angiogenic growth factors (Montesano et al., 1983; Nakatsu et al., 2003; Sieminski et al., 2004; Ghajar et al., 2008), being necessary for any attempt that intends to mimic this process, a fine tune of the conditions to which ECs will be exposed. Since the perception that angiogenesis could be achieved, several in vitro (Folkman & Haudenschild, 1980), several studies attempted to mimic the natural conditions necessary for this process to occur. Although some single component matrices (e.g. collagen, Matrigel and fibrin), when coupled with specialized growth factors, were able to support tube formation (Montesano et al., 1983; Montesano et al., 1986; Chalupowicz et al., 1995; Bach et al., 1998; Dai et al., 2004; Kleinman & Martin, 2005), attempts to monoculturing ECs on biomaterial matrices for microvasculature formation has not been an effective strategy. In monocultures, ECs seem unable to survive and proliferate and, subsequently, selfassembly into tube-like structures is not archived (Janvier et al., 1997). In addition, it is important, that newly formed structures mature and form stable structures. This implies that the interconnected capillary structures are self-sustained after the initial conditions are not present, which is hard to achieve in monoculture as capillary structural sustainability is dependent on the formation of highly specific bonds between ECs and ECM as well as the envelopment of these cells by mural cells (Ribatti et al., 2011). Although during angiogenesis ECs migrate and make sprouts without mural cells’ perivascular cells (PCs) are among the first cells responsible for the invasion of newly vascularized tissues, determining the location of sprout formation and guiding newly formed vessels by interaction with EC via paracrine communication (Ribatti et al., 2011). FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 10 Figure 3. Schematic representation of the dynamics of a co-culture system. Adapted from Battiston et al., 2014 To reproduce this complexity, co-culture systems can be developed to mimic the natural conditions. These involve the culture of two or more types of cells within the same matrix (Battiston et al., 2014). This strategy takes advantage of both the natural cell-ECM interaction and the natural crosstalk between cells, through soluble factors and/or cell-cell interaction and cell-cell contact (Figure 3) (Seghezzi et al., 1998; Grinnel et al., 2000; Saito et al., 2005; Wenger et al., 2005). Co-culture systems are often used with the intent of using one cell type to provide a desired stimulus to a second cell type, presenting a natural, cost-effective strategy for tissue regeneration. This strategy as proven itself effective for ECs tube-like structures formation, as coculturing ECs with fibroblasts (Wenger et al., 2005; Sorrell et al., 2005; Li et al., 2013; Guerreiro et al., 2014; Costa-Almeida et al, 2015), osteoblasts (Hoffman et al., 2008; Grellier et al., 2009; Ghanaati et al., 2011), mesenchymal cells (Wu et al., 2007; Kolbe et al., 2011) and smooth muscle cells (Melero-Martin et al., 2007; Foubert et al., 2008) provides the necessary stimulation for increased ECs survival, proliferation and capillary-like structures assembly that resemble the normal ECs alignment. Human fibroblasts are abundant in the dermis, being the main source of ECM components (e.g. collagen, fibronectin and proteoglycans) and, therefore, modulating mechanical extracellular microenvironment which is critical for vasculogenesis (Berthod et al., 2006). Furthermore, these cells are strongly related to angiogenesis as they infer over the EC behavior through fibroblast-derived proteins (e.g. fibroblast growth factor-2 (FGF-2) and vascular endothelial growth factor (VEGF), the latter a key modulator of normal vessel generation (Seghezzi et al., 1998; Saito et al., 2005), cell-cell dynamics (Wenger et al., 2005) and mechanical extracellular microenvironment contraction (Grinnel et al., 2000), all of which are necessary to modulate EC sprouting and the expansion of capillary-like network (Neufeld et al., 1999; Velazquez et al., 2002; Yamamoto et al., 2003). However, to build a co-culture system, the physico-chemical properties must be carefully considered as biomaterial will serve as support in the initial stages of the FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 17 pectin gels present a higher cytocompatibility, cell adhesion and proliferation, improving al the subsequent cellular functions (Munarin et al., 2011; Munarin et al., 2012; Neves e al., 2015). Furthermore, in addition to the structural resemblance between pectin and alginate, allowing into present the same numerous benefits of alginate, pectin stands out as it presents an interesting degradation profile under simulated physiological conditions (Munarin et al., 2012). Finally, more recently, our group explored the potential of the pectin hydrogels crosslinking by internal ionotropic gelation using the slow-gelling calcium carbonate/D-glucono-d-lactone (CaCO3/GDL) system. Neves e al., (2015) addressed, for the first time, the use of in situ-forming pectin hydrogels as skin cell carriers for tissue engineering, providing an ionotropic internal gelation scheme suitable for in situ gelling systems. Although much still remains to be elucidated about this polymer as a biomaterial, the studies found about the easy tunability of this biomaterial for tissue regeneration (Morra et al., 2004; Bussy et al., 2008; Nagel et al., 2008; Munarin et al., 2011; Munarin et al., 2012; Neves et al., 2015), evidence the promising capabilities of pectin hydrogels as a powerful material system for cell delivery, tissue engineering and regenerative medicine applications. 1.4. Main Goals The incorporation of microvascular networks within the in vitro tissue-engineered skin before its transplantation into a patient would be a major contribution, surpassing the need of relying only on the host’s system ability to promote vascularization. Although encouraging developments have been made in the field (Rivron et al., 2008; Place et al., 2009), in vitro vascularization remains a challenge. In this work, we intend to use a combined approach using the tunable characteristics of soft pectin hydrogel, cells and growth factors to mimic the natural mechanisms involved in the formation of a microvascular network. We aim to construct a three-dimensional, RGD-grafted, soft pectin hydrogel in which fibroblasts support endothelial cells in the formation of selfassembled vascular structures for skin regeneration therapies, while also providing new insights on the biomimetic properties of soft pectin hydrogel’s for future tissueengineering strategies. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 18 2. Materials and Methods Materials and Methods FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 19 2.1. Cell Culture 2.1.1 Routine maintenance Commercial human umbilical vein endothelial cells (HUVECs) (LONZA) and neonatal human dermal fibroblasts (NHDFs, hereafter referred as FBs) (Corriel Institute) were used. HUVECs were cultured in T75 culture flasks, coated with 0.2% (w/v) gelatin from porcine skin (30 minutes at 37 °C, Fluka), with M199 medium (Sigma) supplemented with 10% v/v of inactivated fetal bovine serum (FBS), 1% of antibiotic solution composed of penicillin and streptomycin (Pen/Strep, Gibco) and 0.1 mg.mL-1 of heparin (Sigma-Aldrich) with every-other-day medium exchange. Fibroblasts were cultured in T75 culture flasks with Dulbecco’s Modified Eagle Medium (DMEM; Gibco) supplemented with 10% (v/v) non heat inactivated FBS (Gibco), 1% (v/v) 1% pen/strep (Gibco) and 1% of antimycotic Amphotericin B solution (Sigma) with no medium changes necessary. The cells were incubated at 37 °C, under a humidified atmosphere of 5% v/v CO2 in air. Entrapped cells in pectin discs, when in monoculture, were also cultured in the same conditions, with the media being renewed every three days. After reaching confluence, the cells were trypsinized. For HUVEC trypsinization, the culture medium was removed and the T75s were washed with 5 mL of PBS (NaCl 137 mM, KCl 2.7 mM, NaHPO4.2H2O 10 mM, KH2PO4 1.8 mM, pH 7.4). The HUVECs were incubated with 2 mL of Trypsin/EDTA in PBS (Trypsin 0.05 % w/v, Sigma; EDTA 0.5 mM, Sigma; pH 7.5) for 5 minutes at 37 °C. The T75s were gently tapped to loosen the cells and 2 mL of M199 were added to inactivate the enzyme. The cells were recovered into a single T75, resuspended to avoid aggregates and 10 µL of the solution were loaded into a Neubauer chamber, where the cells were counted under a microscope. HUVECs where seeded in 0.2% (w/v) gelatin-coated T75 at a density of 6x105 cells/T75 and supplemented with 12 mL M199 with 0.03 mg.mL-1 of ECGS. For FBs trypsinization the culture medium was removed and the T75s were washed with 5 mL of PBS. The cells were incubated with 1 mL of Trypsin/EDTA in PBS (Trypsin 0.25% w/v, Sigma; EDTA 2.21 mM, Sigma; pH 7.5) for 5 minutes at 37 °C, after which the flasks were gently tapped to loosen the cells. Neutralization of the trypsin was archived by adding 1 mL of DMEM to each T75 and cells were recovered to a single flask. FBs were resuspended, and cells were counted under the microscope using a Neubauer chamber. FBs were seeded at 5 x 105 cells/T75 and supplemented with 8 mL of DMEM. Both cell types were incubated in the previously described conditions. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 20 For each experiment, HUVECs were used at passages 6-10 and fibroblasts were used at passages 5-10. 2.1.2 Cell thawing HUVECs and FBs cryovials containing 1x106 cells.mL-1 in 10% v/v DMSO in medium, stored in liquid nitrogen, were thawed by immediately placing them in a 37 °C water bath for 1 minute. To the cryovials containing HUVECs or FBs, 1 mL of, respectively, M199 or DMEM was added and a mild up and down was carried out to resuspend the cells. HUVECs were seeded at 6x105 cells/T75 and on supplemented with 12 mL of M199 with 0.03 mg.mL-1 of endothelial cell growth supplement (ECGS, Corning), whereas FBs were seeded at 5x105 cells/T75 and supplemented with 8 mL of DMEM. The cells were incubated at 37 °C under a humidified atmosphere of 5% v/v CO2 in air. Each medium was the next day, for DMSO removal. 2.1.3. Co-culture media selection In order to select a medium that provides a performance close to the ideal for both cell types, HUVECs and FBs behavior was evaluated in several media, including M199, DMEM (both supplemented as previously described) and a combination of the two in three different ratios of M199:DMEM: 3:1 (M3:1), 1:1 (M1:1) and 1:3 (M1:3). Monocultures of both HUVECs and FBs were carried out on 12-well plates with seeding densities of 3.0x104 (D1) and 6.1x104 (D2), which corresponds, respectively, to the relative seeding density per surface area of a T75 and twice as much cells per surface area. For each medium composition, HUVECs were seeded on 0.2% (w/v) gelatin-coated 12-well plates, whereas 1 mL FBs were seeded in uncoated 12-well plates. For HUVECs, each medium composition was supplemented with 0.03 mg.mL-1 of ECGS. To evaluate the effect of the different media on cell behavior, three time points were selected (24h, 72h and 120h) and metabolic activity and total doublestranded DNA quantification assays were carried out. Three replicates were conducted for each time point. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 21 2.1.4. HUVECs and FBs density optimization FBs and HUVECs co-cultures were established with four different cell ratios of 1:1 (R 1:1), 2:1 (R 2:1), 3:1 (R 3:1) and 5:1 (R 5:1) (HUVECs:FBs), at the two seeding densities of D1 and D2. Cells were obtained from T75s cultures following the previously described trypsinization methods for each cell type (Section 2.1.1.). After cell count, the different cell ratios were established. Cells were seeded on 0.2% (w/v) gelatin-coated 12-well plates with M 3:1 supplemented with 0.03 mg.mL-1 of ECGS. To evaluate the effect of the different ratios on cell behavior, at 24h, 72h and 120h, metabolic activity and total double-stranded DNA quantification assays were carried out. Three replicates were conducted for each time point. At each time point, cells were fixed with 1 mL of 4% (v/v) paraformaldehyde (Merk) in PBS for 20 minutes at room temperature (RT) for phenotype characterization. 2.2. Pectin hydrogel 2.2.1. Pectin purification Low methoxyl (LM) citrus pectin (Classic CU701), 86% and a DM of 37%, kindly provided by Herbstreith & Fox (Neuenbürg, Germany), hereafter known as RawPec, was purified based on the protocol as described in Neves et al. (2015). A 1% (w/v) RawPec solution was prepared in ultrapure water (18 MU, Milli-Q UltraPure Water System, Millipore). Following complete pectin dissolution, the pH of the solution was measured and adjusted to 6. The 1% (w/v) RawPec solution was submitted to a sequential filtration through decreasing pore diameter filters, namely, 0.80 µm, 0.45 µm, and 0.22 µm filter membranes (mixed cellulose esters, MCE, Millipore). After filtration, activated charcoal (Norit, Sigma-Aldrich, 2% (w/w) were added to the solution, which was stirred for 1 hour at RT. The suspension was centrifuged for 1h at 27 000 rcf at RT. The supernatant was carefully recovered, and submitted to a new centrifugation with the same parameters, to remove the activated charcoal. The supernatant was recovered and submitted to filtration through a 0.22 µm filter membranes. Pectin was then lyophilized and stored at -20 °C until further use. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 22 2.2.2. Carbodiimide RGD-grafting To surpass the cell-anchorage difficulties imposed by the hydrophobic nature of the hydrogels, biofuncional chemically modified pectin has to be obtained. Pectin was covalently modified with the oligopeptide (Glycine)4-Arginine-Glycine-Aspartic acidSerine-Proline (G4RGDSP) (GenScript), using aqueous carbodiimide chemistry (Rowley et al. 1998), based on the methods previously described for pectin (Munarin et al. 2011; Munarin et al. 2012; Neves et al. 2015). To minimize carbodiimide chemistry side reactions and provide maximum reaction efficiency, a purified pectin solution (1% (w/v)) was prepared in 2-(N-morpholino) ethanesulfonic (MES) acid buffer (0.1 M MES buffering salt, Sigma, and 0.3 M NaCl), with the pH adjusted to 6.5 using 1 M NaOH at RT, overnight. The solution was divided in two in order to obtain RGD-grafted purified pectin (RGDPec) and a control of unmodified pectin (BLKPec). The covalent pectinRGD bond is then achieved by adding, water-soluble carbodiimide, 1-ethyl- (dimethylaminopropyl) carbodiimide (EDC), which is used to form amide linkages between amine containing molecules and the carboxylate moieties on the polymer backbone, N-hydroxy-sulfosuccinimide (sulfo-NHS), a co-reactant which stabilizes the reactive EDC-intermediate form against a competing hydrolysis reaction and G4RGDSP, our RGD-containing oligopeptide. To increase efficiency of the amide bond formation, by minimizing COOH on the RGD reaction, these components were added quickly and following the order: sulfo-NHS (N-Hydroxysulfosuccinimide) (Pierce Chemical, 27.40 mg) and EDC (N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide) (Sigma, 48.42 mg), at a molar ratio of 1:2, followed by addition of the oligopeptide (G4RGDSP) (16.70 mg) to the RGDPec, and both solutions were allowed to react for 20 h under constant stirring. The reactions were quenched with hydroxylamine hydrochloride (Sigma, 18 mg per gram of pectin), and dialyzed against decreasing concentrations of NaCl (30 g, 25 g, 20 g, 15 g, 10 g, 5 g) in ultrapure water for the first 2 days and against ultrapure water with 0 g of NaCl on the last day. The membranes stayed in each solution at least 4 hours, remaining at least one night in the ultrapure water. Both RGDPec and BLKPec were treated with activated charcoal (Norit, SigmaAldrich, 2% (w/w)) for 1h, at RT with stirring. The suspension was centrifuged for 1h at 27 000 rcf at RT. The supernatant was carefully recovered and submitted to filtration through a 0.22 µm filter membranes. Pectin was then lyophilized and stored at -20 °C until further use. The success of the immobilization was further confirmed via UV spectra analysis of RGD-pectin. The amount of covalently modified peptide was estimated by successive FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 23 dilutions of RGD in 1% (w/v) pectin solutions, measured in a microplate reader (Biotek Synergy MX) with Ex/Em at 530/590 nm 2.3. 3D in vitro cell characterization 2.3.1 Characterization of HUVECs and FBs monocultures behavior within 3D RGD-grafted soft pectin hydrogels In order to verify the optimal set up for HUVECs and FBs monocultures on soft pectin matrices, six experimental set-ups were tested, consisting on three different cell densities (D3 = 5 x 106 cells.mL-1; D4 = 1 x 107 cells.mL-1; D5 = 1.5 x 107 cells.mL-1) and two RGD-grafted pectin concentrations (1.5% and 2.5% (w/v)), adjusted to 200 µM of RGD). 2.3.1.1. Cell entrapment The 3D matrices will be prepared by calcium-induced gelation, following the previously described method by Neves et al. (2015). To obtain the final pectin concentrations of 1.5% and 2.5% (w/v), lyophilized sterilefiltered (0.22 mm) RGDPec was dissolved in 0.9 wt% NaCl (in ultrapure water) at 3% and 4% (w/v), respectively. These precursor solutions were adjusted with sterile-filtered (0.22 mm) BLKPec at the same concentrations in order to obtain a final RGD concentration of 200 µM. To trigger hydrogel formation, based on the stoichiometric considerations from Neves et al. (2015), CaCO3 dissolved in 0.9 wt% NaCl (in ultrapure water) was added and carefully mixed with the pectin solution, followed by the addition of D-glucono-d-lactone (GDL, Sigma). Each type of cell, previously cultured in T75 was trypsinized, centrifuged and resuspended in 0.9 wt% NaCl (in ultrapure water) and, at the considered densities, mixed with the pectin solution. For the preparation of cylindrical pectin matrices with a height = 0.5 mm, 20 µL of the cell-laden hydrogel was cast onto a teflon plate. 0.5 mm spacers were used and a second teflon plate was applied over the hydrogel. A humidified chamber was prepared and gelation was allowed to occur at 37 °C for 1 hour under a humidified atmosphere of 5% v/v CO2 in air. After the crosslinking reaction occurred, the cell-laden matrices were transferred to a 24-well culture plate coated with pHEMA (Folkman & Moscona, 1978) and 500 µL of fresh medium was added. To evaluate both the effects of pectin concentration and FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 24 entrapping density on cell behavior, four time points were selected (24h, 48h, 96 and 144h) and metabolic activity and total double-stranded DNA quantification assays were carried out. Three replicas were casted for each formulation and time point. At each time point, cell-laden soft pectin matrices were recovered and fixed in a 4% v/v PFA in TBS-Ca, for a phenotype analysis. 2.3.2. HUVEC and Fibroblasts 3D monocultures performance under different culture media To evaluate the effect of different media on HUVECs and FBs monoculture in a RGDgrafted soft pectin 3D hydrogels, the formulations E3 and F3 were supplemented with the medium that presented the closest to ideal performance maintenance in a 2D environment. The embeddings for these formulations were carried out as described in ―Characterization of HUVECs and FBs monocultures behavior within 3D RGD-grafted soft pectin hydrogels‖ and the cell-laden matrices recovered into a 24-well culture plate coated with pHEMA (Folkman & Moscona, 1978) and 500 µL of fresh M3:1 was added. The HUVEC-laden matrices were also supplemented with 0.03 mg.mL-1 of ECGS. At each time point (24h, 48h, 96 and 144h) the cell-laden pectin matrices were recovered and fixated in a 4% v/v PFA in TBS-Ca and metabolic activity and total double-stranded DNA quantification assays were carried out. Three replicates were conducted for each formulation and time point. 2.3.3. 3D HUVEC:FB co-culture in soft pectin hydrogels Co-cultures were established by entrapping the two different cell types (HUVECs and FBs) at a cell ratio of 3:1 (HUVEC:FB) using the formulation: 1.5% (w/v) Pectin with 1.5 x 107 cells.mL-1 (hereafter described as CC1). Both cell types, cultured in T75, were individually trypsinized and the ratio was established in a 50 mL Falcon. The cells were centrifuged at 1 200 RPM for 5 minutes at RT and the embedding was carried out as described in ―Characterization of HUVECs and FBs monocultures behavior within 3D RGD-grafted soft pectin hydrogels‖. The resulting cell-laden matrices were transferred to a 24-well culture plate coated with pHEMA (Folkman & Moscona, 1978) and 500 µL the selected medium, supplemented with 0.03 mg.mL-1 of ECGS, was added. At each time point (24h, 48h, 96 and 144h) the cell-laden pectin matrices were recovered and fixated in a 4% v/v PFA in TBS-Ca and metabolic activity and total double-stranded FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 25 DNA quantification assays were carried out. Three replicates were conducted for each formulation and time point. 2.3.3.1. 3D HUVEC:FB co-culture spatial patterning: Microinjected HUVEC-laden soft pectin on a FBS-laden soft pectin bed As reviewed by Battiston et al., 2014 an initial spatial patterning of the cells onto the matrix may lead to different outcomes in the co-culture. In this work, a micropatterning technique was used for HUVEC:FB co-culture. Due to time limitations we were only able to carry out a pilot assay. This micropattern was designed to incorporate a HUVEC-laden soft pectin island in the center of an FB-laden soft pectin matrix. Two different independent cell-laden pectin formulations, namely 1.5% (w/v) Pectin with 1.5 x 107 HUVECs.mL-1 and 1.5% (w/v) Pectin with 1 x 107 FBs.mL-1 were simultaneously carried out, using the previously described method. When ready, the spatially patterned matrix was constructed by casting 40 µL of F2 onto a Teflon plate followed by a rapid addition of 10 µL of E3 in the center of the F2 matrix using a 10 µL gel micropipette (See figure 6). 0.5 mm spacers were used and a second Teflon plate was applied over the pectin hydrogel. A humidified chamber was prepared and gelation was left to occur at 37 °C for 1 hour under a humidified atmosphere of 5% v/v CO2 in air. After the crosslinking the cell-laden matrices were transferred to a 24-well culture plate coated with pHEMA (Folkman & Moscona, 1978) and 500 µL of fresh M3:1, supplemented with 0.03 mg.mL-1 of ECGS, was added. Cell-laden matrices were maintained at 37 °C under a humidified atmosphere of 5% v/v CO2 in air, with the media being substituted by the third day. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 26 Figure 6. Schematic representation of the 3D HUVEC:FB co-culture spatial patterning embedding process. 2.4. Phenotype characterization 2.4.1. Cell metabolic activity Cell metabolic activity assessment was performed using a resazurin assay. This assay consists on the bioreduction of the resazurin, which is the oxidized form, to resofurin, which, accordingly, increases the fluorescence. For the 2D resazurin assays, resazurin solution was prepared by dissolution of 10 mg of resazurin (Sigma) in 100 mL of PBS (NaCl 137 mM, KCl 2.7 mM, NaHPO4.2H2O 10 mM, KH2PO4 1.8 mM, pH 7.4), whereas for 3D assays the solution was prepared in 100 mL of TBS (Tris-buffered saline, 50 mM Tris-Cal, 150 mM NaCl, pH 7,5). After complete dissolution, the solution was sterilized by filtration through a 0.22 µm filter and stored at -20 °C, protected from light until further use. At the designated time points for each experiment, namely, 24h, 72h and 120h for 2D and 24h, 48h, 96h and 144h for 3D, the medium was carefully removed cells were incubated with of 20% v/v of the stock resazurin solution (0.1 mg.mL-1, Sigma) in medium for 3 h at 37 °C. Three replicates were monitored for each condition, with three cellular-less replicates serving as control. After the incubation time, 100 µL of the supernatant was transferred to a 96-well black plate with clear bottom (Greiner). Three replicates per well were performed. Fluorescence measurements were carried out using a microplate reader (Biotek Synergy MX) with Ex/Em at 530/590 nm and final results were plotted on Graph Pad Prism 6 software (PRISM). FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 33 Final approximate RGD content is 18 mg RGD per 1 g of pectin. 3.2. Determination of 2D optimal HUVEC/FB culture media composition As reviewed by Battiston et al. (2014), a co-culture system involves multiple interactions, presenting several challenges. A correct selection of the base medium is then necessary to optimize growth and cell phenotype. Endothelial cells and fibroblasts will be cultured separately in 24-well culture plates using five different medium compositions. These included M199, DMEM (both supplemented as previously described) and a combination of the two in three different ratios of M199:DMEM: 3:1 (M3:1), 1:1 (M1:1) and 1:3 (M1:3). This experiment was carried out to evaluate the effect of the different media on cell behavior, allowing the selection of the best-suited media for the co-culture of endothelial cells and fibroblasts. At day 1, HUVECs (Figure 8), at both initial seeding densities (C1 and C2) present similar dsDNA yields (Figure 8, a and b). From the start it is possible to observe that despite the dsDNA profiles are not significantly different, significant differences can be found in the metabolic activity (Figure 8, c and d). Three distinct profiles can be perceived whereas M3:1 presents a similar profile to the optimal medium, M199. M1:1 and M1:3 possess similar metabolic behaviors but are already significantly different from the optimal medium. Finally, DMEM presents the lowest metabolic rate. From day 1 to day 3, an increase in the DNA content is observed, accompanied by an increase in the total metabolic activity in all but DMEM supplemented cultures, where these profiles seem to be maintained or decrease. Finally, from day 3 to day 5, we observe a decrease in the total dsDNA and metabolic activity. Despite these fluctuations, the three profiles are maintained and significantly different between them. Throughout the experiment it is possible to observe that the metabolic activity (normalized with the total dsDNA) of HUVECs is maintained. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 34 Figure 8. Effect of cell density and medium composition on metabolic activity and proliferation of HUVECs in 2D during 5 days in culture a) and b) total dsDNA (PicoGreen assay), c) and d) metabolic activity (resazurin assay) and e) and f) metabolic activity per nanogram of dsDNA of HUVECs. * denotes statistically significant differences (p < 0.05). Regarding FBs (Figure 9), at day 1, it is possible to see that the different seeding densities are translated into different dsDNA yields (Figure 9, a and b). This difference is also represented in the total metabolic activity (Figure 9, c and d) whereas C1 presents a lower total metabolic activity than C2. At day 3, it is observed an increase for both seeding densities in the total DNA content and metabolic activity. By day 5, the total DNA content seems to remain unaltered except for the M199, M3:1 and M1:1 media conditions of C1, where it increases. Here, we observe a metabolic activity increase for C1 densities, whereas for C2 the values remain unaltered for M 3:1, M 1:1 and M 1:3, increasing in DMEM and M199. When normalized, these results point to a constant metabolic activity FBs for seeded at 6.1x104 cell/well throughout the culture FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 35 time, whereas for lower seeding concentrations (namely 3.0x104 cell/well) it is possible to observe a decrease followed by a slight increase, stabilizing at similar values to C2. Figure 9. Effect of cell density and medium composition on metabolic activity and proliferation of FBs in 2D within a 5 days culture period. a) and b) total dsDNA (PicoGreen assay), c) and d) metabolic activity (resazurin assay) and e) and f) metabolic activity per nanogram of dsDNA of NHDFs. * denotes statistically significant differences (p < 0.05). Given that, for HUVECs, there are no significant alterations in the metabolic and proliferation profiles alterations when using M3:1 (when compared to M199) and that the same happens to FBs (when compared to DMEM), M3:1 is a suitable candidate for supplementation in a co-culture system. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 36 3.3. Determination of 2D optimal in vitro HUVEC/FB ratio The majority of natural tissues consist of multi-cellular systems of two or more cell types which interact with each other to facilitate viability, proliferation and differentiation (Nam et al., 2011; Schubert et al., 2008; Traphagen et al., 2013). Blood vessels are a multi-cellular system composed of endothelial cells (ECs), vascular smooth muscle cells (VSMCs), and fibroblasts (Ratner et al., 1996). In tissue engineering, co-culture systems have been increasingly used as it provides simulation of the in vivo physical and biological properties. Furthermore, published data indicate that co-culture of ECs with FBs can provide the complex mixture of growth factors, ECM and cell-cell contacts necessary to potentiate tubulogenesis (Berthod et al., 2006; Sorrell et al., 2007; Auger et al., 2013). However, as in natural environments cell distribution is not uniform, for in vitro optimal experimental outcomes in co-culture assays it is necessary an optimization of the cell seeding number and ratio. In order to evaluate the effect of FBs on HUVECs capillary-like self-assembly, four our different cell ratios were established, namely, R 1:1, R 2:1, R 3:1 and R 5:1 (HUVECs:FBs) were tested at two seeding densities of 3.04x104 (D1) and 6.08x104 (D2). Cells were seeded on a 0.2% gelatincoated 12-well plate with the selected medium (M 3:1) supplemented with ECGS for five days. At day 1, the different seeding densities present different DNA content, with higher values for C2, result of the different densities seeded. Regarding the total metabolic activity, for C1 all ratios present similar values whereas for C2 it is possible to observe a pattern where R 5:1 > R 3:1 > R 2:1 > R 1:1, hence presenting higher activities in ratios containing more HUVECs (Figure 10, d), which can also be verified when the values are normalized (using the total dsDNA) (Figure 10, f). By day 3, there are observed increases in the total DNA for C1 and C2. However, whereas for C1 there are no noticeable differences between the different ratios, for C2, R 1:1 presents higher values than the remainder ratios. This difference is translated in the total metabolic activity where is possible to verify that R 1:1 presented significantly higher metabolic values when compared to the other ratios. Although it is possible to see a decrease in the metabolic activity of the cells (normalized with the total dsDNA) (Figure 10, e and f) it is verified a maintenance of the higher values for endothelial rich ratios (except for R1:1 and R2:1 in C1 seeding conditions, which can be due to an deficient recovery of the dsDNA, as these also present lower values for total DNA). It is however important to FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 37 Figure 10. Effect of cell density and cell ratio on metabolic activity and proliferation of HUVEC:FB co-culture in 2D within a 5 days culture period. a) and b) total dsDNA (PicoGreen assay), c) and d) metabolic activity (resazurin assay) and e) and f) metabolic activity per nanogram of dsDNA of HUVEC: * denotes statistically significant differences (p < 0.05). point out that despite this decrease, the cells’ metabolic activity is 5 to 6-fold higher than any of the individual cultures. At day 5, the total DNA increases for both seeding conditions, presenting, in all ratios, values superiors to those verified in the monocultures. For C1 and C2, in the R 1:1, the ratio containing the highest initial fibroblasts number (presenting however half of those verified in the monocultures), the total DNA content is 3 and 4-fold higher, respectively, when compared to FBs monocultures whereas for R 5:1, an endothelial rich ratio, the total DNA content is 3fold higher than for HUVEC monocultures. However, regarding the metabolic activity, despite an increase for the total metabolic activity, normalized values show a decrease in the activity of the cells, as it already occurred from day 1 to 3. Nonetheless, the metabolic values are 2-fold higher than individual cell cultures. Putting all together, for a 2D HUVEC:FB co-culture, lower ratios seem to favor proliferation whereas higher ratios seem to favor higher metabolic activities per cell (metabolic activity normalized FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 38 Figure 11. Pictures of HUVEC:FB co-cultures prepared at 6.08x104 cells/well at the different ratios of 1:1, 2:1, 3:1 and 5:1. Images were obtained at the first and last day of 5-days. Scale bars, 200 µm. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 39 Figure 12. Pictures of HUVEC:FB co-cultures prepared at 6.08x104 cells/well at the different ratios of 1:1, 2:1, 3:1 and 5:1. Images were obtained at the first and last day of a 5. Scale bars, 200 µm. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 40 with total dsDNA). For a 5 days culture, the different seeding densities do not seem to influence the final outcome of the culture, affecting however the development of the culture throughout the culture days, as it was also seen for individual cultures. However, to our purpose, to obtain capillary like structures, metabolic and proliferation studies are not sufficient. In order to properly choose for an adequate ratio, the cell morphology and spatial distribution must be taken into account. To analyze this, we’ve fixated the different conditions at the different time points and took pictures of the structures formed using an inverted light microscope. The images reveal evident differences between cells organization in different ratios used for HUVEC:FB cocultures, as can be depicted in the Figures 11 and 12. At lower HUVEC:FB ratios, namely R 1:1 and R 2:1 both cells appeared widely distributed at the cell culture surface throughout the five days, presenting their respective fibroblastic-like and cobblestone-like shapes. At day 5, however, it is observed that these cultures present similar structures to the ones observed in FB monocultures at confluence, hence demonstrating FB dominance in the cell culture surface. At higher ratios, namely R 3:1 and R 5:1, both cell types also archive fibroblastic-like and cobblestone-like shapes. With these ratios, at day 5, it is possible to verify that cells were rearranged in a significantly different manner featuring web-like structures characteristics of tubular-like cellular network formation. All together, high densities with endothelial rich ratios seem more appropriate as 2D co-culture formulations for HUVEC development and capillarylike self assembly. 3.4. Analysis of HUVEC and FB monocultures’ behavior in 3Dculture As depicted in previously, RGD-grafted pectin hydrogels present cytocompatibility, cell adhesion and proliferation characteristics for the improvement of cellular functions (Munarin et al., 2011; Munarin et al., 2012; Neves e al., 2015). As such, pectin stands as a particularly appealing biomimetic material for 3D hydrogel formation for cell culture. For hydrogels, the mechanical performance of a matrix depends on the polymer and crosslinker characteristics, concentration, gelling conditions (e.g. temperature, pH, and gelation time), swelling, and degradation. In turn, these will have an impact in the biomaterial mechanical properties, including elasticity, compressibility, viscoelastic behavior, tensile strength, and failure strain (Anseth et al., 1996). With this in mind, we have followed an already described gelation scheme thoroughly characterized by Neves et al (2015), varying the concentration of the pectin used in the FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 41 hydrogels between 1.5% and 2.5% (w/v). These concentrations where used to determine the effects matrix characteristics (e.g. mesh size, mechanical compliance) over the ability of cell to regulate their metabolic activity, alter their morphology, migrate and establish cell-cell contact. Furthermore, to evaluate the effect of cell density on the behavior of cells in 3D environments, cells were entrapped within pectin matrices at different entrapping densities, namely: D3 = 5x106 cells.mL-1; D4 = 1x107 cells.mL-1 and D5 = 1.5x107 cells.mL-1. Cell metabolic activity (Resazurin assay), total dsDNA (PicoGreen assay) and morphology (Confocal microscope images) assays were carried out at days 1, 2, 4 and 6. 3.4.1 HUVEC behavioral analysis on 3D soft pectin hydrogels At day 1, the dsDNA content of the different formulations was proportional to their respective original cell density, presenting however higher values for a 2.5% (w/v) pectin concentration (Figure 13). It is also important to note that the total dsDNA for lowest entrapping density in 1.5% (w/v) pectin hydrogels is nearly inexistent. Regarding the metabolic activity, HUVECs presented higher total metabolic activities in agreement with the previous observations for the total dsDNA. The highest metabolic activity registered was for the 1.5x107 HUVECs.mL-1. When normalized (with total dsDNA), the metabolic results show that for 1.5% (w/v) pectin concentrations all entrapping densities present the same cellular activity, whereas for 2.5% (w/v) lower densities seem to present higher cellular activities. For all the formulations tested (except for 2.5% (w/v) hydrogels with 1x107 HUVECs.mL-1, possibly due to deficient manipulation throughout the experiment), the total dsDNA content gradually decreased along the period of culture, being accompanied by the decrease in the metabolic activity. By day 4, the metabolic activity assay demonstrate close or equal to 0, independently of the content of DNA, which were maintained till day 6. Regarding the spatial behavior (Figures 14 and 15), for the lowest entrapping density, namely, 5x106 HUVECs.mL-1, the images were not presented as they did not help to elucidate what was happening within the hydrogels. As for 1x107 HUVECs.mL-1 and 1.5x107 HUVECs.mL-1 entrapping densities, independently of the pectin concentration, cells seem widely distributed within the matrix. Cells maintained a globular structure, indicative of the lack of adhesion, throughout the experiment. As the total dsDNA and metabolic assays shown very low DNA content and metabolic activity for day 4 and 6, no stains were carried, hence no images are presented. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 42 All together, these results indicate that, independently of the pectin concentration and entrapment densities, monoculturing HUVECs on pectin hydrogels is not an efficient strategy for HUVEC self-assembled tubulogenesis or even maintenance. Figure 13. Effect of initial cell entrapment density and pectin concentration on metabolic activity and proliferation of HUVEC in a 3D soft pectin hydrogel within a 6 days culture period. a) and b) total dsDNA (PicoGreen assay), c) and d) metabolic activity (resazurin assay) and e) and f) metabolic activity per nanogram of dsDNA of HUVEC * denotes statistically significant differences (p < 0.05). 3.4.2. FBs behavioral analysis on 3D soft pectin hydrogels FBs As regards to FBs, two significantly different profiles can be observed for different pectin concentrations (Figure 16). For 1.5% (w/v), at day 1, it is possible to verify that the cell-loaded matrix present total dsDNA values in accordance to the initial number of cells entrapped. By day 2, a slight increase in the total dsDNA is observed for the entrapping densities of 5x106 FBs.mL-1 and 1x107 FBs.mL-1, which, attending to the remaining profile, could be the result of the total DNA assay manipulation. At day 4, dsDNA seems to slightly decrease, maintaining its values at day 6. Relatively to the FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 49 Figure 20. Effect of the initial entrapping density on a 2.5% pectin 3D hydrogel on FBs’ conformation within a 6 days culture period. FBs were stained for F-actin (Green) and nuclei (Blue). Scale bars, 100 µm. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 50 Regarding the spatial behavior, accompanying the tendency presented by cell metabolic activity, 1.5% (w/v) pectin hydrogels presented better results when compared to 2.5% (w/v) pectin hydrogels. Within these, as seen in the Figures 17 and 18, cells were widely distributed, being able to adhere, acquiring a fibroblast-like shape. Throughout time, these cells were able to establish cell-cell contacts, forming FB spheroids which increased in number and size, as depicted in Figure 21 and 22 Table 1. These structures presented significantly larger spheroids when compared to the ones formed in 2.5% (w/v) pectin hydrogels. It is also demonstrated that, depending on the initial entrapping densities in 1.5% (w/v) pectin hydrogels, higher entrapping densities favor a significantly larger size of the spheroids (Table 1). However, as stated before, due to degradation of the matrices at 1.5x107 FBs.mL-1 (the discs broke and started to decompose). Regarding 2.5% (w/v) pectin hydrogels, independently of the Figure 21. Effect of Initial cell entrapping and pectin concentration over FBs spheroid size. a) and b) spheroids average size throughout the 6 days of culture. c).and d) relative frequency of spheroid size at day 6. * denotes statistically significant differences (p < 0.05) ) between different entrapping densities on different pectin concentrations. c) d) Area (µm2) Area (µm2) FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 51 Table 1. Effect of initial cell entrapping densities and pectin concentration over FBs spheroid size (µm2) and number. Area means and standard deviation is presented in micrometers. N stands for total number of spheroids. * denotes statistically significant differences (p < 0.05) between the same entrapping density on different pectin concentrations. α denotes statistically significant differences (p < 0.05) between different entrapping densities on different pectin concentrations. Day 1 Day 2 Day 4 Day 6 1.5% Pectin Area n Area n Area n Area n 5x106 FBs.mL-1 761±288* 37 960±358* 176 1032±486* 261 1017±676* 206 1x107 FBs.mL-1 927±287*α 170 1051±422* 183 1121±481* 291 1244±580*α 251 2.5% Pectin 5x106 FBs.mL-1 660±120 34 652±105 29 441±90 19 489±77 26 1x107 FBs.mL-1 705±203 87 735±212 60 500±140 53 528±150 34 1.5x107 FBs.mL-1 895±232 α 153 915±294 α 83 889±235 α 51 758±136 α 59 entrapping density, cells exhibit a globular-like shape, with some cells presenting a fibroblast-like shape, mostly observed at the surface (Figures 19 and 20). Although there is possible to observe some spheroids, the number and size is significantly lower than what is observed for 1.5% (w/v) pectin hydrogels. Throughout the period of the culture, 2.5% (w/v) pectin matrices loaded with 1.5x107 FBs.mL-1 present significant larger spheroids when compared to the other entrapping densities. However, for all densities, spheroid number and size decreases with time, accompanying the decrease in metabolic activity previously described for these conditions. Figure 22. Effect of pectin concentration over the ability of FBs to contract the matrix. Macroscopic differences of 1.5% and 2.5% pectin hydrogels seeded with 1x107 FBs.mL-1. Images were obtained at the first and last day of a 6-days culture using an inverted microscope using a magnification of 16.3 x. a) and b) correspond to 1.5% (w/v) pectin hydrogels at day 1 and 6 respectively. c) and d) correspond to 2.5% (w/v) pectin hydrogels at day 1 and 6 respectively. e) represents the relative size of the pectin matrices when compared to day 1 FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 52 Table 2. Effect of pectin concentration over the ability of FBs to contract the matrix. Macroscopic differences of 1.5% and 2.5% pectin hydrogels entrapped with 1x107 FBs.mL-1. Areas means and standard deviation is measured in square millimeters 1x107 cells.mL-1 Day 1 Day 2 Day 4 Day 6 1.5% Pectin 31.95 ± 1.09 21.40 ± 0.04 18.81 ± 0.35 17.58 ± 0.57 2.5% Pectin 38.34 ± 1,09 33.72 ± 0.42 31.46 ± 1.32 29.81 ± 0.12 Another important feature of the gels is the mechanical compliance, which will impact over the cells ability to exert forces to deform the matrices. To test the influence of pectin concentration over the ability of FBs monocultures to promote matrix contraction, macroscopic images of 1.5% and 2.5% (w/v) pectin hydrogels matrices loaded with 1x107 FBs.mL-1, the highest performing comparable density (1.5x107 FBs.mL-1 was excluded as for 1.5% (w/v) pectin matrices were unsuitable to handle), were recovered at day 1, 2, 4 and 6 (Table 2 and Figure 22). At day 1, 1.5% (w/v) pectin hydrogels presented smaller areas than 2.5% (w/v) pectin hydrogels. As these matrices were produced with the same volume (20 µL), this difference indicates that after the initial swelling (verified while manipulating but not measured), from day 0 to day 1 matrices start to contract at lower pectin concentrations. By day 2, 1.5% (w/v) pectin hydrogels presented a 33% matrix contraction when compared to day 1 and by day 6 and 45%. Regarding 2.5% (w/v) pectin hydrogels, for these days, we observe a matrix contraction of 12% and 22% respectively, when compared to day 1. Furthermore, by day 6, 2.5% (w/v) pectin hydrogels are 1.70x larger than 1.5% (w/v) pectin hydrogels. Summing up, for both pectin concentrations, total dsDNA values remained essentially constant throughout the period of culture, suggesting that the entrapped cells were not able to proliferate, independently of the original cell density. The main differences were found at the metabolic activity and cellular spatial distribution. In 1.5% (w/v) pectin hydrogel matrices, cells were able to adhere to the matrix, acquiring fibroblast-like shape, and establish cell-cell interactions, forming FBs spheroids, while maintaining a steady-state of metabolic activity. In these conditions, higher entrapping densities favored a higher spheroid formation with significantly higher areas. Moreover, for 1x107 FBs.mL-1, macroscopic images of the cell-laden matrices allowed to observe alterations of the construct size, presenting at day 6 to almost half of the size of the one observed at day 1. However, regarding 2.5% (w/v) pectin hydrogels, cells essentially maintained FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 53 a globular like shape, exhibiting a fibroblast-like shape mostly at the surface. These results were in accordance to what was expected as 2.5% (w/v) pectin hydrogels are less compliant and provide less free space for cells to spread when compared to 1.5% (w/v) pectin hydrogels (Neves et al., 2015). FBs spheroids were also observed in this construct but with significantly inferior numbers and size. Finally, macroscopic images demonstrate that these matrices maintain a more robust aspect possessing not only, at day 6, 80% of the size when compared to day 1 but also almost a 2-fold increase when compared to 1.5% (w/v) pectin hydrogels. Results obtained by us indicate that cells might be constrained by the polymeric network at higher pectin concentrations, favoring the use of more compliant matrices for FBs culture. Within these (1.5% (w/v) pectin hydrogels), higher entrapping densities seem to stimulate cell-cell contacts and the formation spheroids. 3.5 HUVEC:FB co-culture establishment in 3D soft pectin hydrogels As previously stated, coculturing ECs with FBs can potentiate tubulogenesis (Berthod et al., 2006; Sorrell et al., 2007; Auger et al., 2013). Nonetheless, great differences can be observed among different matrices, cell type’s combinations, entrapping densities and entrapping ratios. As such, it is of utmost importance to optimize these factors for 3D cultures in each specific application design. As documented throughout this thesis, several optimization steps were taken into account in order to choose the coculturing conditions, namely: medium supplementation (through metabolic activity and total dsDNA) and cell ratio (through metabolic activity, total dsDNA and optical microscopy analyzes), in 2D experiments, and polymer concentration and cell entrapping densities (through metabolic activity, total dsDNA and optical microscopy analyzes), in 3D experiments. Aiming to promote the capillary self-assembly of HUVECs in a 3D pectin hydrogel, these were co-culture with FB at a ratio of 3:1 (HUVEC:FB), with an entrapping density of 1.5x107 cells.mL-1. These were cultured in pectin hydrogels with a 1.5% pectin concentration and supplemented with M 3:1 with 0.03 mg.mL-1 of endothelial cell growth supplement (ECGS). 3.5.1. Characterization of the influence of M 3:1 supplementation on HUVECs or FB monocultures in 3D soft pectin hydrogels Preceding the HUVEC:FB co-culture, in order to evaluate the medium influence on the 3D cells behavior, both HUVECs and FBs were loaded in 1.5% (w/v) pectin hydrogels FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 54 at the chosen entrapment density namely, 1.5x107 cell.mL-1. The behavior was analyzed recurring to metabolic and total dsDNA assays and, as depicted by Figure 23. Throughout the culture period, M3:1 does not present any significant difference displaying very similar profiles when compared to the optimal medium for each cell type, with exceptions made to the metabolic activity of FBs at day 1 for M 3:1 supplementation and dsDNA content for FBs in M3:1 at day 6. The first can be explained due to a faster manipulation of the cell in the entrapment process, leading to a slower loss of activity by the FBs whereas the second, given the metabolic results for the same condition, should be a DNA manipulation error. With this experiment, M 3:1 was successfully tested as a potential candidate for HUVEC:FB co-culture in both 2D and 3D environments. Figure 23. Effect medium composition on metabolic activity and proliferation of HUVECs and FBs in a 3D pectin hydrogel within a 6 days culture period. a) total dsDNA (PicoGreen assay), b) metabolic activity (resazurin assay) and c) metabolic activity per nanogram of dsDNA. * denotes statistically significant differences (p < 0.05). 3.5.2. Characterization of HUVEC:FB co-culture behavior in a 3D soft pectin hydrogel HUVEC:FB co-culture were established by simultaneous entrapment of both cell types in 3D pectin hydrogel, with a entrapping density of 1.5x107 cells.mL-1, at a ratio of 3:1 (HUVEC:FB). To evaluate the behavior of the co-culture, at days 1, 2, 4 and 6, we proceeded to measure the total dsDNA and metabolic activity. These values where then compared to the ones obtained for FBs and HUVECs monocultures in the same conditions. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 55 Figure 24. Co-culture of HUVECs and FBs in a ratio of 3:1 (HUVEC:FB) in a 3D pectin hydrogel within a 6 days culture period. a) total dsDNA (PicoGreen assay), b) metabolic activity (resazurin assay) and c) metabolic activity per nanogram of dsDNA As Figure 24 shows, throughout the culture period co-cultures present a steady decrease in the total dsDNA. This decrease is accompanied by a decrease in metabolic activity. When compared to HUVEC monocultures, at all time points, cocultures presented higher dsDNA and metabolic values. However, the dsDNA decreasing profile it is similar for both conditions. This could imply that the metabolic differences observed can be a result of a higher cell activity of the cells in a co-culture environment, which is verified throughout the culture period when normalizing the metabolic results with dsDNA content, and/or by the activity of FBs within the matrix. Nonetheless, these results showed that despite the dsDNA loss with time (which represents a cell loss) the culture was able to maintain activity until day 6, which did not happened for HUVEC monoculture. Cell morphology and re-arrangement within the matrices was analyzed by confocal microscopy. Cell were stained against vWF (Red), an endothelial specific cell marker, α-SMA (Gray) a protein expressed by cell with contractile abilities, F-actin (Green) and nuclei were counterstained with DAPI (Blue) and the merged images were assembled with ImageJ. At day 1 it is possible to observe a wide distribution of vWF-positive cell, representing HUVECs and vWF-negative cell, which represent FBs (Figure 25 a and d). At this time point images do not show α-sma presence in the cells. Moreover, few FB spheroids are observed, which is probably a consequence of the low number of fibroblasts entrapped FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 56 (we previously verified that lower entrapment densities led to fewer and smaller FB spheroids size). Due to experimental problems, as it had already occurred for FB monocultures in 2.5% (w/v) pectin hydrogels, actin staining was not successful, therefore not elucidatory of the shape of the cells. By day 4, it was possible to observe an increase in the number of aggregates in the co-culture, maintaining however widely distributed, unorganized profile (Figure 25 b and e). At this time point there is possible to observe that vWF-negative cells start to express α-sma, which could be indicative of a differentiation of fibroblasts into myofibroblasts. More interestingly vWF-positive cell, namely HUVECs, seem to also be expressing α-sma. Although uncommon, these phenomenon is described in the literature (Lu et al., 2004; Cevallos et al., 2006), indicating a possible differentiation of HUVECs into smooth muscle cells that, in conjugation with myofibroblasts, would fulfill the role of perivascular cells. Figure 25. Effect of cell type to type 3:1 ratio on HUVEC:FB co-culture cell morphology and spatial distribution for a 6 days culture period. Cells were stained against vWF (Red) and α-SMA (Gray), for F-actin (Green) and nuclei (Blue). Scale bars, 100 µm By day 6 (Figure 25 c and f), the expression of α-sma in vWF-positive cell became more evident. Cells remain spread through the matrix, not acquiring any specific organization. All together, these results demonstrate that, in a 6-days culture period, HUVEC self-assembly into tubular-like structure is not favored by the ratio used. Moreover, this ratio seems to promote HUVECs expression of α-sma, promoting their differentiation into smooth muscle cells b) a) c) d) e) f) FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 57 3.5.3 Micropatterning For a different approach on the problem, a micropatterning technique was tested. Micropatterning can be used to control the area on which cell-cell interaction can occur as well as the cell populations that are allowed to interact with one another. Using micropatterning, a HUVEC island was inserted within a matrix of 3D cultured fibroblasts. However, under the light of the previous results for the co-culture experiments, and given the technical implications of this process in situ protocol adjustments were performed. As such, the final experimental conditions presented an HUVEC:FB co-culture with a ratio of, approximately 1:3 (more precisely 1:2.66) with an entrapping density of 1.1x107 cells.mL-1. This condition was evaluated at the days 1, 2 and 4, using a confocal microscope. For differentiation between the co-cultures, R 3:1 HUVEC:FB co-culture will be depicted as CoHUVECs and the R 1:3 HUVEC:FB coculture will be depicted as MiHUVECs. When observed by confocal microscopy, at day 1, under these conditions, it is possible to observe the formation of larger FBs aggregates when compared to the CoHUVECs, with these already express α-sma (Figure 26 a and d). HUVECs where positioned essentially in the center. By day two, it is possible to see the HUVECs rich island from which it is possible to observe that cells are migrating towards the periphery, the FB rich zones. This HUVEC rich island presents several vWF-positive cells that do not express α-sma. In fact, α-sma seems to be expressed in the contact zones between HUVECs and FBs (Figure 26 b). By day 4, HUVECs seem to be more widely distributed, with a mixture of vWF-positive cells that express α-sma and vWF-positive cells that do not. However, at this time point, images are not elucidative on the ability of cells self-assemble into capillary-like structures or even acquiring a spiderweb-like spatial arrangement (Figure 26 c and f), indicative of possible tubular-like formations. Due to the experimental difficulties and time restrains, it was not possible to evaluate this culture behavior at day 6, which could be more elucidative on MiHUVECs ability to promote self-assembled tubular structures. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 58 Figure 26. Effect of cell type to type 1:3 ratio on HUVEC:FB co-culture cell morphology and spatial distribution for a 6 days culture period. Cells were stained against vWF (Red) and α-SMA (Gray), for F-actin (Green) and nuclei (Blue). Scale bars, 100 µm Finally, it is important to note that the in situ alterations performed while carrying out the experiment imply that in this construction present three variables when compared to the HUVEC:FB co-cultures previously described: entrapping density, cellular ratio and spatial patterning. Each of these variations should be isolated to properly attend to its influence on HUVECs self-assembly. . b) a) c) d) e) f) FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 65 Engler et al., 2006; Li et al., 2010; Munarin et al., 2011; Fonseca et al., 2011; Munarin et al., 2012; Fonseca et al., 2013; Neves et al., 2015). As such, to bypass the adhesiveness issue of the polymer, a system based on the combination of RGDmodified pectin was attempted by grafting RGD-containing oligopeptides into the pectin backbone using aqueous carbodiimide chemistry (Rowley et al. 1998), based on the methods previously described for pectin (Munarin et al. 2011; Munarin et al. 2012; Neves et al. 2015). Another factor taken into account was diffusion. Diffusion is crucial to attend the cell nutritional requirements, metabolic wastes, and soluble molecules, being a critical issue in microscale designs. Diffusion is dependent on the distance that a molecule as to travel. As such, smaller matrices facilitate diffusion as the center of the matrix is within a more reachable distance, easing the cell’s effort to obtain nutrients and deplete wastes. Furthermore, pectin microsphere studies demonstrated that diameters ranging from 300-500 μm were suitable to convey the incoming flow of oxygen and the outgoing of catabolites, as cells maintained their viability for 29 days (Munarin et al., 2011). Attending to these factors, in this work we’ve used 500 μm cylindrical pectin hydrogel matrices modified with controlled final RGD concentration of 200 μM. Within these, the mechanical influence was tested through the use of different polymer concentrations (1.5% and 2.5% (w/v)). Moreover, in 3D matrices cell-cell interactions are directly influence by the cell entrapment density (Maia et al., 2014). To address this, different entrapment densities were also tested to fibroblasts and endothelial cells do study their cell-cell implications on the behavioral development within the hydrogels. To achieve the desired in a 3D environment studies suggest that are required strategies to overcome the physical impediments posed by the matrices, which include: growth arrest; stimulating cell senescence and favoring a quiescent-like state (Bott et al., 2010). As evidenced for HUVEC monocultures (Figure 6), all the conditions established, the total dsDNA content gradually decreased along the period of culture, being accompanied by the decrease in the metabolic activity. Endothelial cells are typically ―quiescent‖ (the average lifespan of an EC is more than 1 year) (Aird, 2007). As such, and given the favoring of a quiescent-like state by 3D matrices, proliferation was not expected. However, in addition to the lack of proliferation, cell loss is observed throughout the experiment, with lower dsDNA values being obtained in 1.5% (w/v) pectin hydrogels in the first 2 days. A possible explanation for this is the pore size of the matrices. As depicted by Neves et al. (2015), the initial mesh size of pectin hydrogels is higher for the 1.5% (w/v) hydrogels when compared to the 2.5% (w/v) hydrogels (707 nm vs 380 nm). Experimental observations show that, after medium addition to the matrices, pectin hydrogels swell, as other authors already verified FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 66 (Sriamornsak et al., 2007). As water migrates to the matrix, pores and channels are created (Liu et al., 2003). Given that HUVECs are cells with measured sizes of 14-15 µm, cells might be escaping the matrix (or even dying) before being able to establish the necessary integrin-RGD adhesive interactions, being the loss faster in bigger size pore matrices. Furthermore, this loss might also impact over the already adhered cells as cellular critical densities values for cellular development might not be achieved and subsequently promote cell detachment. In fact, for 3D environments, at lower entrapment densities, cell–cell interactions may be easily hindered, and, as consequence, a decrease in the biological performance can be noted (Cukierman et al., 2001). Among others factors, paracrine signals, as vascular endothelial growth factor (VEGF) (Silva & Mooney, 2010) and basic fibroblast growth factor (bFGF) (Edelman et al., 1991) have been linked with better proliferation, migration and viability results for ECs. The lack of these soluble factors in out constructs can be another of the limiting factor. As such, although described in the literature that lower stiffness materials favor endothelial self-assembly and tubulogenesis (Saunders & Hammer, 2010; Bidarra et al., 2011; Maia et al., 2014), for pectin, in these conditions HUVEC monoculture is not sustainable. Any further attempt at HUVEC monoculture designs using pectin hydrogels must attend to the initial pectin swelling and soluble factors presence issues. Possible solutions for these must be attended through chemical modifications of the polysaccharide and/or different gelation strategies (Liu et al., 2003) and addition of the factors through exogenous addition or co-culture systems (Eckermann et al., 2011). Furthermore, despite these issues, aiming microvasculature formation in vitro, 1.5% (w/v) pectin hydrogels use should be prioritized as, as demonstrated by Saunders & Hammer (2010), more compliant matrices (Young modulus of 140-1000 Pa) promote network formation, whereas stiffer matrices (Young modulus > 1000 Pa) do not. Regarding FB monocultures, the present results showed that entrapped FBs were metabolically active throughout culture time within all pectin hydrogels, although no stimulation of cell proliferation is observed (Figure 8). Better results were however observed in 1.5% (w/v) hydrogels were cells presented a steady-state of metabolic activity throughout the 6 days, whereas for for 2.5% (w/v) hydrogels it was verified a decrease along time. In 1.5% (w/v) hydrogels, fibroblasts were able to spread and establish cell-to-cell contacts inside the RGD-grafted pectin hydrogels, leading to the formation of multicellular aggregates, which was not observed for 2.5% (w/v) hydrogels where cells remained round and dispersed. As previous 1.5 % (w/v) pectin hydrogel mesh size is larger, forming a polymeric network that is less dense that those presented by 2.5% (w/v) matrices. Pore size is of critical importance, affecting not only FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 67 the diffusion process and well as the cell migration, cell differentiation and cell distribution, by altering perception of the environment by the cells (Lee et al., 2008; Choi et al., 2010; Bergmeister et al., 2013). As such, within the first days (before the matrix contract), less dense matrices may lead to enhancement of the biological performance, a factor which was already proposed by other authors (Zhang et al., 2011; Bidarra et al., 2013; Neves et al., 2015). Moreover, 1.5 % (w/v) pectin hydrogel present softer microenvironments. Coupled with the pore size, these matrices present more compliant environments that allow cells to exert ―tracking‖ forces to deform the surrounding matrix, and, thus, migrate and aggregate. Moreover these microscopic observations (aggregates formation, Figure 9), when coupled with the macroscopic evidences (matrix contraction, Figure 11) suggest tissue formation, which is evidenced to be dependent on mechanical input to the cells, as also observed in other studies (Butler et al., 2000; Drury et al., 2003; Reinhart-King et al., 2008; Reinhart-King, 2011). To note that, for 2.5% (w/v) hydrogels, although the documented internal behavior presented mostly round cells, a similar effect to the one documented by Maia et al. (2014) for 2 (w/v)% alginate hydrogels when compared to 1% (w/v) for hMSCs. However, in our matrix surface appeared populated by stretched cells, able to establish cell-cell contacts and forming cellular networks. (See annexes Figure 1.). These results show that, independently of the selected pectin concentration, by taking advantages of the mechanical compliance posed by pectin, FBs were able to migrate outwards the matrix, populating the surface and forming aggregates within the hydrogels, which was already demonstrated for Neves et al. (2015) regarding hMSCs. Given that the dsDNA analysis demonstrated steady values for the period of culture, one can speculate that these were formed through self-assembly rather than clonal expansion. Nonetheless, to provide an insight over the origin of these aggregates, a KI-67 test should be conducted. Cell density directly influences cell–cell signaling. Therefore, the initial seeding densities will impact on the 3D cellular behavior presented by the cells in a co-culture. In 2D environments, cellular densities are well studied. In these, higher cell densities are known to promotes cell-to-cell contact, which can lead to contact inhibition (cell cycle arrest and thus proliferation inhibition) (Puliafito et al., 2012), stimulate cellular differentiation (Hohn et al., 1996), among others. For 3D environments, these effects were also verified, whit higher cell densities promoting clusters formation (Zhang et al., 2011), differentiation (Mudera et al., 2010) and ECM production (Huang et al., 2008; Talukdar et al., 2011; Maia et al., 2014; Neves et al., 2015). In this work, higher entrapment densities promoted FB clusters formation, with significant differences in relation to lower densities. It is however important to note that 1.5% (w/v) pectin FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 68 hydrogel with 1.5 x 107 FBs.mL-1 construct were not able to sustain FB development due to matrix decomposition. In one hand, as exposed by Neves et al. (2015), the crosslinking times are slightly affected when cells are present within the hydrogel (~5 minutes for 1,5% (w/v) and ~3 minutes for 2.5% (w/v)), more specifically with 8x106 cells.mL-1. In our work, the highest used density was almost 2-fold higher, being 1.5 x 107 FBs.mL-1. Therefore, the cross-linking reaction could have been significantly altered leading to a more sensitive matrix. In addition, the high porosity present in 1,5% (w/v) matrices sacrifices mechanical properties by reducing the amount of material present in the matrix (Lee et al., 2008). On the other hand, in 3D cell culture, the matrix must withstand cell attachment forces (Lee et al., 2008), which at higher cell densities are expected to be greater. All together, we suggest that the matrix degradation might be a physical impediment of the matrix, inefficient cross-linking reaction and/or high porosity, when exposed to high densities contractile cell embedding. Finally, although our work only establishes a comparison between polymer concentrations, other studies demonstrated that the final maximum force produced is dependent on the number of cells within the matrix (Eastwood et al., 1994). To sum up, these results demonstrate a matrix and cell density-dependent 3D cell behavior, suggesting a potentiated response for lower stiffness and higher cell entrapment, where FBs tend to aggregate into ―tissue-like‖ structures. The metabolic results, coupled microscopic (aggregates formation) and macroscopic (matrix contraction) evidences, suggests microtissue formation, with possible ECM deposition and angiogenic growth factors production, in lower pectin hydrogels concentrations (1.5% (w/v)), being the best result, for FBs, the 1.5% (w/v) pectin matrices with 1 x 107 FBs.mL-1 construct. Fibroblast microtissue formation is of utmost importance for the optimization of a coculture system. In skin, FBs are the main responsible for ECM modulation. FBs have the ability of remodeling the elasticity and mechanical integrity of matrix, by producing enzymes, such as proteases and collagenases (Ratner et al., 2004). This ability is also observed in in vitro conditions were FBs, under controlled conditions, are able to produce natural ECM proteins such as FN, collagen, GAGs, tenascin-C, and others (Dzamba & Peters, 1991; Korducki et al., 1992; Berthod et al., 2006; Soucy & Romer, 2009; Costa-Almeida et al., 2015). Among these FN stands out as it is significantly stronger than than the RGD peptide alone. FN presents a native tertiary structure that favors the specific angiogenesis-dependent integrin interactions (Petrie et al., 2006). Moreover, additional peptide sequences present in the polymer (e.g. PHRSN) enhance the α5β1 integrin binding to the FN-RGD motif, which are also liked to be a requirement for angiogenesis (Aota et al., 1994; Laurens et al., 2009). Besides matrix FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 69 deposition, fibroblasts are also strongly related to angiogenesis as they infer over the EC behavior through fibroblast-derived proteins, namely fibroblast growth factor-2 (FGF-2) and vascular endothelial growth factor (VEGF), the latter a key modulator of normal vessel generation (Seghezzi et al., 1998; Saito et al., 2005). Furthermore, deformations of the matrix suggest that fibroblasts may be producing smooth muscle αactin (α-SMA), a fibroblasts contractile marker. This indicates that fibroblasts are going through differentiation into myofibroblasts, which are involved in wound contraction and remodeling wound healing processes, on which vascularization is potentiated (Darby et al., 1990; Arora et al., 1999). ECs in compliant matrices are shown to communicate through mechanical signals, perceiving and reacting to tension dependent stresses of neighboring cells (Reinhart-King et al., 2008; Reinhart-King, 2011). Through traction forces, interactions between endothelial cells and ECM, for example, regulate bFGF and subsequently, capillary development (Ingber & Folkman, 1989). All together, the use of a co-culture system using ECs and fibroblasts to test biomaterials biocompatibility and their influence in in vitro angiogenesis assays is the logical step to follow. In the present study, to achieve in vitro microvascularization, we purposed to integrate cellular, biochemical, and biophysical cues biomaterials, taking advantage of the natural crosstalk between cells, through soluble factors and/or cell-cell interaction and the mechanical compliance demonstrated in 1.5%(w/v) pectin hydrogels. This strategy as proven itself effective for ECs tube-like structures formation, as coculturing ECs with fibroblasts (Wenger et al., 2005; Sorrell et al., 2005; Li et al., 2013; Guerreiro et al., 2014; Costa-Almeida et al., 2015). Here, we intend to take advantage of the FBs documented ability to produce natural ECM (Costa-Almeida et al., 2015; Berthod et al., 2006) and soluble factors (VEGF and bFGF) (Seghezzi et al., 1998; Saito et al., 2005) to, through a natural and cost-effective way , potentiate angiogenesis. Furthermore, as critical cell density was showed to be needed for network formation (Saunders & Hammer, 2010), to surpass the HUVEC loss by our results evidenced, the highest cell density was chosen. Finally before proceeding to the 3D HUVEC:FB co-culture, M 3:1 efficiency for HUVEC and FB supplementation was tested. Therefore, two additional conditions were tested using identical monoculture conditions to those chosen for coculture (medium type, biomechanical stimulation, cell densities, culture substrate, among others). With this, we intend to address the data interpretation issue, distinguish between the relative contributions of cell-cell interactions versus cell-biomaterial interactions, aiding to determine effects of the cells compared to how they are stimulated by the biomaterial substrate alone. As demonstrated by Figure 12, the monocultures cultured in the selected medium, M 3:1, presented identical profiles to FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 70 the ones cultured in the optimal medium. This result indicates that this medium is suitable for both 2D and 3D HUVEC and FB co-cultures. These results will serve as a standpoint for co-culture comparison, and, if differences exist, this indicates that they will be most likely due to cell-cell interactions. 4.3 HUVEC:FB co-culture establishment in 3D soft pectin hydrogels Tissue architecture and function are closely interrelated. Blood vessels consist of multicellular system with 3 distinct layers of endothelium, smooth muscle, and connective tissue. More precisely, these are lined by a longitudinally oriented single layer of ECs, which defined the lumen, followed by circularly oriented smooth muscle layer (pericytes) and outer connective tissue layers (Ratner et al., 2004). As such, the formation of mature and functional vascular networks requires the cooperation of endothelial cells (ECs) and perivascular cells, with cell-cell direct contact and paracrine signaling interactions being of utmost importance for the sustainability of mature microvasculature (Auger et al., 2013; Battiston et al., 2014). As previously pointed out, until this point, this study focused on the use optimization of the conditions for the establishment of a HUVEC:FB co-culture. We are specifically interested in addressing this subject in context of microvascularization inclusion in in vitro self-assembled artificial skin for skin regeneration therapies. As described throughout this work, in vitro self-assembled capillary formation is dependent on EC lining for lumen formation (Ratner et al., 1996; Ratner et al., 2004; Koh et al., 2008). For this, several cues provided by ECM-cell, cell-cell and cell-growth factor interactions play key roles in this process, all of which were demonstrated, by our and other works, to be present in FB cultures capable of forming tissue like structures (Seghezzi et al., 1998; Saito et al., 2005; Reinhart-King et al., 2008; Soucy & Romer, 2009; Reinhart-King, 2011). As such, through weighted selection based on observed result and current literature, a coculture was established under the following conditions: 1.5% (w/v) pectin hydrogels seeded with 1.5x107 cells.mL-1 with a 3:1 ratio favoring HUVECs. This was supplemented with a mixture medium composed of M199:DMEM in a 3:1 ratio. The results were evaluated through metabolic and total dsDNA assays as well as staining against anti-α-SMA and anti-vWF, respectively, a contractile marker characteristic (but not exclusive) of myofibroblasts and a endothelial-specific marker. As depicted by (Figure 13), throughout the culture period co-cultures present a steady decrease in the total dsDNA, accompanied by a decrease in metabolic activity. This profile is similar to HUVEC monocultures. However by day 4 and until day 6, unlike FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 71 monocultures, co-cultures maintained metabolic activity. Furthermore, when normalized (by the total dsDNA) these results show an increased cell metabolic activity with time (Figure 13). As such, the establish co-culture seem to present a mix of the characteristics presented by HUVEC and FB monocultures, namely, a HUVEC-like cell loss profile (total dsDNA and metabolic activity decrease) and a FB-like positive response to the mechanical compliance of the matrix. However these results do not positively distinguish between cell types. As such, to provide more insight on the implications of the co-culture in HUVECs behavior, spatial arrangement was analyzed through immunostaining techniques. As showed in Figure 14, the main objective was not achieved as no tube-like structures are observed. Nonetheless, by 6 it is still possible to verify a prominent demarcation of HUVECs by anti-vWF (Figure 14. Image c), indicating FB co-culture had a positive impact over HUVECs, increasing their survival. However, as mentioned, our HUVEC:FB did not support capillary which was already verified in other studies (Wenger et al., 2005; Sorrell et al., 2005; Soucy et al., 2009; Eckermann et al., 2012; Guerreiro et al., 2014). As it is known VEGF, a fibroblast-derived protein, is a potent and key mediator for angiogenesis (Seghezzi et al., 1998; Korff et al., 2001; Saito et al., 2005; Silva & Mooney, 2010; Eckermann et al., 2011). The effects of this growth factor over EC are, dose-dependent (Conn et al., 1990), gradient dependent (Gerhardt et al., 2003) and time-dependent (Silva & Mooney, 2010), with the best results optimal results being obtained with high VEGF levels (50 ng.mL-1) at early time points and constant presence over time (Nakatsu et al., 2003 ;Silva & Mooney, 2010). Moreover VEGF is also linked to present a potent synergetic effect with bFGF (also known as FGF-2) for angiogenesis induction (Pepper 1992). In fact, bFGF, whether released by fibroblasts or EC (Schweigerer, 1987), induces VEGF expression in endothelial cells, leading to capillary formation (Seghezzi et al., 1998). However, we constructed a co-culture with a low FB density in the culture which could result in low growth factor concentrations values and, consequently, weaker stimulation of the HUVECs. In fact, low FB density could suitable, yet untested, explanation for our results as another density-dependent phenomena might not occur. For example, higher ECM deposition rates within cellular aggregates, which have been linked with higher cell densities (Maia et al., 2014). Among the proteins secreted releases FN, stands out as it is significantly stronger than than the RGD peptide alone, by promoting multiple integrin potentiated in interaction with FN-RGD (Aota et al., 1994; Laurens et al., 2009; Petrie et al., 2006; Soucy & Romer, 2009). In our co-culture, as Figure 14, shows it is verified few FB clusters are observed with relatively small sizes. As such, these may not have reached the critical conditions for a physiologically relevant matrix deposition for HUVEC self-assembled capillary formation. Finally, the FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 72 mechanical modulation impact on ECs behavior cannot be disregarded as it has the potential to promote endothelial cell expression of bFGF and promote angiogenesis (Ingber & Folkman 1989; Berthod et al., 2006; Reinhart-King et al., 2008; Reinhart-King, 2011). The magnitude of the exerted forces is not only dependent on matrix stiffness (Sieminski et al., 2004) but is also modulated by cell density. (Eastwood et al., 1994). Our results show that, despite the low FBs densities, FBs were able to express α-SMA by the 4th day, suggesting that myofibroblasts differentiation occurred and that a contractile phenotype was achieved. However these in these densities, the mechanical tension forces applied showed might have not been able to reach the critical values for angiogenesis stimulation. All in all, although no certain confirmation was obtained in this study, we postulate that one of the possible reasons for HUVEC microvascular unsuccessful self-assembly was the low FB density. Furthermore, it is important to note that HUVECs phenotype suffers alterations through time. At day 1, HUVECs are characterized by a red phenotype indicating the presence of vWF, an endothelial specific marker (Zanetta et al., 2000). However, throughout our culture period, the cells progress from red to orange, being this difference more pronounced by day 6. This indicates a progressive expression of α-SMA, which, as we already noted, is a contractile marker. Although, to our knowledge, this phenotype is not commonly expressed in ECs, in some cases it might be observed. Through Jagged1-Notch interaction, endothelial cells are shown to undergo endothelialmesenchymal transdifferentiation, leading to the expression of α-SMA (Noseda et al., 2004; Noseda et al., 2006). Cevallos et al. (2006) demonstrated that cyclic strain induces expression of specific smooth muscle cell markers in human endothelial cells. This is however a situation that requires a deeper insight on the mater. In the future would be interesting to stain HUVEC monocultures against anti-α-SMA, allowing the distinction between whether this phenotype is a consequence of cell-cell or cell-ECM interactions. Furthermore stain of anti-α-SMA against HUVEC monocultures should be carried out against M199 or m 3.1 supplemented conditions, evaluating the effect of the media/serum on this phenotype, which were already demonstrated to impact over several cell biological functions (Shahdadfar et al., 2005; Kunz-Schughart et al., 2006); Eckermann et al. 2011; Traphagen et al., 2013). FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 73 4.4 3D HUVEC:FB co-culture spatial patterning: Microinjected HUVEC-laden soft pectin on a FB-laden soft pectin bed Attending to the issues verified in the co-cultures a spatial patterning approach was attempted. As HUVEC cell loss was one of the main concerns, we embedded a HUVEC island in the center of a FB-laded 1.5(w/v) pectin matrix. By doing so, we intended to use peripheral FB-containing matrix as a net, trying to prevent HUVEC loss. Furthermore, do protocol adjustment on the course of the experiment, allowed us to increase the initial FB entrapment density from the 3.75x106 cells.mL-1 verified in the 3:1 (HUVEC:FB) ratio to 8x106 cells.mL-1, altering the ratio to approximately 1:3 (HUVEC:FB). This FBs density was closer to the optimal FB density for microtissues formation (1x107 cells.mL-1). From day 1, FB presented more clusters with larger sizes. Furthermore, α-SMA is expressed in the fibroblasts from day 1, which was not observed in our initial co-cultures (Figure 15 a and d). By day 2, HUVECs already presented a phenotype similar to the one in day 6 of our co-culture, expressing both αSMA and vWF (Figure 15 b and e). More importantly, these images may provide an insight on why the HUVECs are expressing α-SMA. In the Figure 15 b) it is possible to observe the HUVEC island, at the right side of the image, and the fibroblast rich zones. The orange phenotype resulting of the simultaneous expression of α-SMA and vWF occurs mainly in the fibroblast contact zones. As previously mentioned, Notch signaling has been implicated in the transdifferentiation of ECs to smooth muscle cells, leading to the production α-SMA through a Jagged1-Notch interaction (Noseda et al., 2004; Noseda et al., 2006). Notch receptors are membrane-tethered receptor that mediates cell-cell receptor-ligand interactions. As such, in these conditions, we hypothesize that, through cell-cell direct contact mechanisms, FBs are recruiting HUVECs to a transdifferentiation into smooth muscle cells through a Jagged1-Notch interaction. Nonetheless, as aforementioned, media supplementation and cell-ECM influences on HUVEC phenotype should be performed to better infer on this phenomenon. Finally, by day 4 Figure 15 c) show a mixture of FBs, α-SMA positive HUVECs and α-SMA negative HUVECs, which seem to be acquiring a spider web-like pattern. The variations imposed between or 3:1 (HUVEC:FB) co-culture and our spatially patterned co-culture seem to favor spatial arrangement in HUVECs suggesting tubular formation. However, to confirm HUVEC self-assembly in these conditions, a prolonged experience time in needed, which, due to the technical difficulties imposed during this embedding protocol and time limitations, was not possible. Furthermore, matrix FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 74 deposition and growth factor detection assays should be conducted for FBs monocultures and both co-cultures for an accurate description of the FB environmental remodeling, providing better insights on data interpretation. All in all, notwithstanding the fact that no clear evidence for tubular formation was presented, our results offer promising expectations for future experiments under similar conditions. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 81 Braccini, I., Grasso, R. P., & Pérez, S. (1999). Conformational and configurational features of acidic polysaccharides and their interactions with calcium ions: a molecular modeling investigation. Carbohydrate Research, 317(1), 119-130. Bussy, C., Verhoef, R., Haeger, A., Morra, M., Duval, J. L., Vigneron, P., ... & Nagel, M. D. (2008). Modulating in vitro bone cell and macrophage behavior by immobilized enzymatically tailored pectins. Journal of Biomedical Materials Research Part A, 86(3), 597-606. Butler, David L., Steven A. Goldstein, and Farshid Guilak. "Functional tissue engineering: the role of biomechanics." Journal of biomechanical engineering122.6 (2000): 570-575. Cen, L., Liu, W., Cui, L., Zhang, W., & Cao, Y. (2008). Collagen tissue engineering: development of novel biomaterials and applications. Pediatric research, 63(5), 492496. Cevallos, M., Riha, G. M., Wang, X., Yang, H., Yan, S., Li, M., ... & Chen, C. (2006). Cyclic strain induces expression of specific smooth muscle cell markers in human endothelial cells. Differentiation, 74(9‐10), 552-561. Chalupowicz, D. G., Chowdhury, Z. A., Bach, T. L., Barsigian, C., & Martinez, J. (1995). Fibrin II induces endothelial cell capillary tube formation. The Journal of cell biology, 130(1), 207-215. Chiang, B., Essick, E., Ehringer, W., Murphree, S., Hauck, M. A., Li, M., & Chien, S. (2007). Enhancing skin wound healing by direct delivery of intracellular adenosine triphosphate. The American journal of surgery, 193(2), 213-218.. Choi, S. W., Zhang, Y., & Xia, Y. (2010). Three-dimensional scaffolds for tissue engineering: the importance of uniformity in pore size and structure. Langmuir,26(24), 19001-19006. Chrobak, K. M., Potter, D. R., & Tien, J. (2006). Formation of perfused, functional microvascular tubes in vitro. Microvascular research, 71(3), 185-196. Chunmeng, S., & Tianmin, C. (2004). Skin: a promising reservoir for adult stem cell populations. Medical hypotheses, 62(5), 683-688. Conn, G., Soderman, D. D., Schaeffer, M. T., Wile, M., Hatcher, V. B., & Thomas, K. A. (1990). Purification of a glycoprotein vascular endothelial cell mitogen from a rat glioma-derived cell line. Proceedings of the National Academy of Sciences, 87(4), 1323-1327. Cooper, T. P., & Sefton, M. V. (2011). Fibronectin coating of collagen modules increases in vivo HUVEC survival and vessel formation in SCID mice. Acta biomaterialia, 7(3), 1072-1083. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 82 Costa-Almeida, R., Gomez-Lazaro, M., Ramalho, C., Granja, P. L., Soares, R., & Guerreiro, S. G. (2014). Fibroblast-endothelial partners for vascularization strategies in tissue engineering. Tissue Engineering Part A, 21(5-6), 1055-1065. Cukierman, E., Pankov, R., & Yamada, K. M. (2002). Cell interactions with threedimensional matrices. Current opinion in cell biology, 14(5), 633-640. Cukierman, E., Pankov, R., Stevens, D. R., & Yamada, K. M. (2001). Taking cell-matrix adhesions to the third dimension. Science, 294(5547), 1708-1712. Culver, J. C., Hoffmann, J. C., Poché, R. A., Slater, J. H., West, J. L., & Dickinson, M. E. (2012). Three‐Dimensional Biomimetic Patterning in Hydrogels to Guide Cellular Organization. Advanced materials, 24(17), 2344-2348. Cuono, C., Langdon, R., & McGuire, J. (1986). Use of cultured epidermal autografts and dermal allografts as skin replacement after burn injury. The Lancet, 327(8490), 1123-1124. Dai, N. T., Williamson, M. R., Khammo, N., Adams, E. F., & Coombes, A. G. A. (2004). Composite cell support membranes based on collagen and polycaprolactone for tissue engineering of skin. Biomaterials, 25(18), 4263-4271. Daley, W. P., Peters, S. B., & Larsen, M. (2008). Extracellular matrix dynamics in development and regenerative medicine. Journal of cell science,121(3), 255-264. Darby, I., Skalli, O., & Gabbiani, G. (1990). a-Smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing.Lab Invest, 63(1), 2129. Dean, D. M., Napolitano, A. P., Youssef, J., & Morgan, J. R. (2007). Rods, tori, and honeycombs: the directed self-assembly of microtissues with prescribed microscale geometries. The FASEB Journal, 21(14), 4005-4012. Dee, K. C., Andersen, T. T., & Bizios, R. (1999). Osteoblast population migration characteristics on substrates modified with immobilized adhesive peptides. Biomaterials, 20(3), 221-227. DeVolder, R., & Kong, H. J. (2012). Hydrogels for in vivo‐like three‐dimensional cellular studies. Wiley Interdisciplinary Reviews: Systems Biology and Medicine,4(4), 351-365. Discher, D. E., Janmey, P., & Wang, Y. L. (2005). Tissue cells feel and respond to the stiffness of their substrate. Science, 310(5751), 1139-1143. Drury, J. L., & Mooney, D. J. (2003). Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials, 24(24), 4337-4351. Durand, D., Bertrand, C., Clark, A. H., & Lips, A. (1990). Calcium-induced gelation of low methoxy pectin solutions—Thermodynamic and rheological considerations. International journal of biological macromolecules, 12(1), 14-18. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 83 Dzamba, B. J., & Peters, D. M. (1991). Arrangement of cellular fibronectin in noncollagenous fibrils in human fibroblast cultures. Journal of cell science,100(3), 605612. Eastwood, M., McGrouther, D. A., & Brown, R. A. (1994). A culture force monitor for measurement of contraction forces generated in human dermal fibroblast cultures: evidence for cell-matrix mechanical signalling. Biochimica et Biophysica Acta (BBA)- General Subjects, 1201(2), 186-192. Eckermann, C. W., Lehle, K., Schmid, S. A., Wheatley, D. N., & Kunz‐Schughart, L. A. (2011). Characterization and modulation of fibroblast/endothelial cell co‐cultures for the in vitro preformation of three‐dimensional tubular networks. Cell biology international, 35(11), 1097-1110. Edelman, E. R., Mathiowitz, E., Langer, R., & Klagsbrun, M. (1991). Controlled and modulated release of basic fibroblast growth factor. Biomaterials, 12(7), 619-626. Ehrbar, M., Sala, A., Lienemann, P., Ranga, A., Mosiewicz, K., Bittermann, A., ... & Lutolf, M. P. (2011). Elucidating the role of matrix stiffness in 3D cell migration and remodeling. Biophysical journal, 100(2), 284-293. Elsner, J. J., & Zilberman, M. (2010). Novel antibiotic-eluting wound dressings: An in vitro study and engineering aspects in the dressing's design. Journal of tissue viability, 19(2), 54-66. Engler, A. J., Sen, S., Sweeney, H. L., & Discher, D. E. (2006). Matrix elasticity directs stem cell lineage specification. Cell, 126(4), 677-689. Esser, S., Lampugnani, M. G., Corada, M., Dejana, E., & Risau, W. (1998). Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells. Journal of cell science, 111(13), 1853-1865. Fang, Y., Al-Assaf, S., Phillips, G. O., Nishinari, K., Funami, T., & Williams, P. A. (2008). Binding behavior of calcium to polyuronates: comparison of pectin with alginate. Carbohydrate Polymers, 72(2), 334-341. Folkman, J., & D'Amore, P. A. (1996). Blood vessel formation: what is its molecular basis?. Cell, 87(7), 1153-1155. Folkman, J., & Haudenschild, C. (1980). Angiogenesis in vitro. Folkman, J., & Hochberg, M. (1973). Self-regulation of growth in three dimensions. The Journal of experimental medicine, 138(4), 745-753. Fonseca, K. B., Bidarra, S. J., Oliveira, M. J., Granja, P. L., & Barrias, C. C. (2011). Molecularly designed alginate hydrogels susceptible to local proteolysis as threedimensional cellular microenvironments. Acta biomaterialia, 7(4), 1674-1682. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 84 Fonseca, K. B., Gomes, D. B., Lee, K., Santos, S. G., Sousa, A., Silva, E. A., ... & Barrias, C. C. (2013). Injectable MMP-sensitive alginate hydrogels as hMSC delivery systems. Biomacromolecules, 15(1), 380-390. Foubert, P., Matrone, G., Souttou, B., Leré-Déan, C., Barateau, V., Plouët, J., ... & Tobelem, G. (2008). Coadministration of endothelial and smooth muscle progenitor cells enhances the efficiency of proangiogenic cell-based therapy.Circulation research, 103(7), 751-760. Fraeye, I., Duvetter, T., Doungla, E., Van Loey, A., & Hendrickx, M. (2010). Fine-tuning the properties of pectin–calcium gels by control of pectin fine structure, gel composition and environmental conditions. Trends in food science & technology, 21(5), 219-228. Fuchs, S., Hofmann, A., & Kirkpatrick, C. J. (2007). Microvessel-like structures from outgrowth endothelial cells from human peripheral blood in 2-dimensional and 3dimensional co-cultures with osteoblastic lineage cells. Tissue engineering, 13(10), 2577-2588. Gerhardt, H., Golding, M., Fruttiger, M., Ruhrberg, C., Lundkvist, A., Abramsson, A., ... & Betsholtz, C. (2003). VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. The Journal of cell biology, 161(6), 1163-1177. Ghajar, C. M., Chen, X., Harris, J. W., Suresh, V., Hughes, C. C., Jeon, N. L., ... & George, S. C. (2008). The effect of matrix density on the regulation of 3-D capillary morphogenesis. Biophysical journal, 94(5), 1930-1941. Ghanaati, S., Fuchs, S., Webber, M. J., Orth, C., Barbeck, M., Gomes, M. E., ... & James Kirkpatrick, C. (2011). Rapid vascularization of starch–poly (caprolactone) in vivo by outgrowth endothelial cells in co‐culture with primary osteoblasts. Journal of tissue engineering and regenerative medicine, 5(6), e136-e143. Giancotti, F. G., & Ruoslahti, E. (1999). Integrin signaling. Science, 285(5430), 10281033. Goldberg, V. M. (1992). Natural history of autografts and allografts. In Bone implant grafting (pp. 9-12). Springer London. Goldberger, A., Middleton, K. A., Oliver, J. A., Paddock, C., Yan, H. C., DeLisser, H. M., ... & Newman, P. J. (1994). Biosynthesis and processing of the cell adhesion molecule PECAM-1 includes production of a soluble form.Journal of Biological Chemistry, 269(25), 17183-17191. Green, D. W., Mann, S., & Oreffo, R. O. (2006). Mineralized polysaccharide capsules as biomimetic microenvironments for cell, gene and growth factor delivery in tissue engineering. Soft Matter, 2(9), 732-737. Grellier, M., Ferreira-Tojais, N., Bourget, C., Bareille, R., Guillemot, F., & Amédée, J. (2009). Role of vascular endothelial growth factor in the communication between FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 85 human osteoprogenitors and endothelial cells. Journal of cellular biochemistry, 106(3), 390-398. Grellier, M., Granja, P. L., Fricain, J. C., Bidarra, S. J., Renard, M., Bareille, R., ... & Barbosa, M. A. (2009). The effect of the co-immobilization of human osteoprogenitors and endothelial cells within alginate microspheres on mineralization in a bone defect. Biomaterials, 30(19), 3271-3278. Grinnell, F. (2000). Fibroblast–collagen-matrix contraction: growth-factor signalling and mechanical loading. Trends in cell biology, 10(9), 362-365. Grinnell, F., & Petroll, W. M. (2010). Cell motility and mechanics in three-dimensional collagen matrices. Annual review of cell and developmental biology, 26, 335-361. Groeber, F., Holeiter, M., Hampel, M., Hinderer, S., & Schenke-Layland, K. (2011). Skin tissue engineering—in vivo and in vitro applications. Advanced drug delivery reviews, 63(4), 352-366. Grzesik, W. J., & Robey, P. G. (1994). Bone matrix RGD glycoproteins: immunolocalization and interaction with human primary osteoblastic bone cells in vitro. Journal of Bone and Mineral Research, 9(4), 487-496. Guerreiro, S. G., Oliveira, M. J., Barbosa, M. A., Soares, R., & Granja, P. L. (2014). Neonatal Human Dermal Fibroblasts Immobilized in RGD‐Alginate Induce Angiogenesis. Cell transplantation, 23(8), 945-957. Guo, S., & DiPietro, L. A. (2010). Factors affecting wound healing. Journal of dental research, 89(3), 219-229. Hall, H. (2007). Modified fibrin hydrogel matrices: both, 3D-scaffolds and local and controlled release systems to stimulate angiogenesis. Current pharmaceutical design, 13(35), 3597-3607. Harholt, J., Suttangkakul, A., & Scheller, H. V. (2010). Biosynthesis of pectin. Plant physiology, 153(2), 384-395. Hoffman, A. S. (2012). Hydrogels for biomedical applications. Advanced drug delivery reviews, 64, 18-23. Hofmann, A., Ritz, U., Verrier, S., Eglin, D., Alini, M., Fuchs, S., ... & Rommens, P. M. (2008). The effect of human osteoblasts on proliferation and neo-vessel formation of human umbilical vein endothelial cells in a long-term 3D co-culture on polyurethane scaffolds. Biomaterials, 29(31), 4217-4226. Hohn, H. P., Grümmer, R., Bosserhoff, S., Graf-Lingnau, S., Reuss, B., Bäcker, C., & Denker, H. W. (1996). The role of matrix contact and of cell-cell interactions in choriocarcinoma cell differentiation. European journal of cell biology, 69(1), 76-85. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 86 Huang, A. H., Yeger-McKeever, M., Stein, A., & Mauck, R. L. (2008). Tensile properties of engineered cartilage formed from chondrocyte-and MSC-laden hydrogels. Osteoarthritis and Cartilage, 16(9), 1074-1082. Ingber, D. E. (1990). Fibronectin controls capillary endothelial cell growth by modulating cell shape. Proceedings of the National Academy of Sciences,87(9), 35793583. Ingber, D. E., & Folkman, J. (1989). Mechanochemical switching between growth and differentiation during fibroblast growth factor-stimulated angiogenesis in vitro: role of extracellular matrix. The Journal of cell biology, 109(1), 317-330. Inngjerdingen, K. T., Patel, T. R., Chen, X., Kenne, L., Allen, S., Morris, G. A., ... & Paulsen, B. S. (2007). Immunological and structural properties of a pectic polymer from Glinus oppositifolius. Glycobiology, 17(12), 1299-1310. Jackson, C. L., Dreaden, T. M., Theobald, L. K., Tran, N. M., Beal, T. L., Eid, M., ... & Mohnen, D. (2007). Pectin induces apoptosis in human prostate cancer cells: correlation of apoptotic function with pectin structure.Glycobiology, 17(8), 805-819. Jaffe, E. A., Nachman, R. L., Becker, C. G., & Minick, C. R. (1973). Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. Journal of Clinical Investigation, 52(11), 2745. Janvier, R., Sourla, A., Koutsilieris, M., & Doillon, C. J. (1996). Stromal fibroblasts are required for PC-3 human prostate cancer cells to produce capillary-like formation of endothelial cells in a three-dimensional co-culture system. Anticancer research, 17(3A), 1551-1557. Johnen, C., Steffen, I., Beichelt, D., Bräutigam, K., Witascheck, T., Toman, N., ... & Gerlach, J. C. (2008). Culture of subconfluent human fibroblasts and keratinocytes using biodegradable transfer membranes. Burns, 34(5), 655-663. Jones, I., Currie, L., & Martin, R. (2002). A guide to biological skin substitutes. British journal of plastic surgery, 55(3), 185-193. Kleinman, H. K., & Martin, G. R. (2005, October). Matrigel: basement membrane matrix with biological activity. In Seminars in cancer biology (Vol. 15, No. 5, pp. 378-386). Academic Press. Kogan, G., Šoltés, L., Stern, R., & Gemeiner, P. (2007). Hyaluronic acid: a natural biopolymer with a broad range of biomedical and industrial applications. Biotechnology letters, 29(1), 17-25. Koh, W., Stratman, A. N., Sacharidou, A., & Davis, G. E. (2008). In vitro three dimensional collagen matrix models of endothelial lumen formation during vasculogenesis and angiogenesis. Methods in enzymology, 443, 83-101. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 87 Kolbe, M., Xiang, Z., Dohle, E., Tonak, M., Kirkpatrick, C. J., & Fuchs, S. (2011). Paracrine effects influenced by cell culture medium and consequences on microvessellike structures in Co-cultures of mesenchymal stem cells and outgrowth endothelial cells. Tissue Engineering Part A, 17(17-18), 2199-2212. Korducki, J. M., Loftus, S. J., & Bahn, R. S. (1992). Stimulation of glycosaminoglycan production in cultured human retroocular fibroblasts.Investigative ophthalmology & visual science, 33(6), 2037-2042. Korff, T., Kimmina, S., Martiny-Baron, George., & Augustin, H. G. (2001). Blood vessel maturation in a 3-dimensional spheroidal Co-culture model: direct contact with smooth muscle cells regulates endothelial cell quiescence and abrogates VEGF responsiveness. The FASEB Journal, 15(2), 447-457. Kunz-Schughart, L. A., Schroeder, J. A., Wondrak, M., van Rey, F., Lehle, K., Hofstaedter, F., & Wheatley, D. N. (2006). Potential of fibroblasts to regulate the formation of three-dimensional vessel-like structures from endothelial cells in vitro. American Journal of Physiology-Cell Physiology, 290(5), C1385-C1398. Labat-Robert, J., Bihari-Varga, M., & Robert, L. (1990). Extracellular matrix.FEBS letters, 268(2), 386-393. Laurens, N., Engelse, M. A., Jungerius, C., Löwik, C. W., van Hinsbergh, V. W., & Koolwijk, P. (2009). Single and combined effects of αvβ3-and α5β1-integrins on capillary tube formation in a human fibrinous matrix. Lee, J., Cuddihy, M. J., & Kotov, N. A. (2008). Three-dimensional cell culture matrices: state of the art. Tissue Engineering Part B: Reviews, 14(1), 61-86. Li, X. Y., Wang, T., Jiang, X. J., Lin, T., Wu, D. Q., Zhang, X. Z., ... & Yuan, M. J. (2010). Injectable hydrogel helps bone marrow-derived mononuclear cells restore infarcted myocardium. Cardiology, 115(3), 194-199. Liu, L., Fishman, M. L., Kost, J., & Hicks, K. B. (2003). Pectin-based systems for colonspecific drug delivery via oral route. Biomaterials, 24(19), 3333-3343. Lokmic, Z., & Mitchell, G. M. (2008). Engineering the microcirculation. Tissue Engineering Part B: Reviews, 14(1), 87-103. Lu, X., Dunn, J., Dickinson, A. M., Gillespie, J. I., & Baudouin, S. V. (2004). Smooth muscle α-actin expression in endothelial cells derived from CD34+ human cord blood cells. Stem cells and development, 13(5), 521-527. Luttun, A., Carmeliet, G., & Carmeliet, P. (2002). Vascular progenitors: from biology to treatment. Trends in cardiovascular medicine, 12(2), 88-96. MacNeil, S. (2007). Progress and opportunities for tissue-engineered skin.Nature, 445(7130), 874-880. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 88 Maia, F. R., Fonseca, K. B., Rodrigues, G., Granja, P. L., & Barrias, C. C. (2014). Matrix-driven formation of mesenchymal stem cell–extracellular matrix microtissues on soft alginate hydrogels. Acta biomaterialia, 10(7), 3197-3208. Mani, S. A., Guo, W., Liao, M. J., Eaton, E. N., Ayyanan, A., Zhou, A. Y., ... & Weinberg, R. A. (2008). The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell, 133(4), 704-715. Mao, J., Zhao, L., de Yao, K., Shang, Q., Yang, G., & Cao, Y. (2003). Study of novel chitosan‐gelatin artificial skin in vitro. Journal of Biomedical Materials Research Part A, 64(2), 301-308. Martin, P. (1997). Wound healing--aiming for perfect skin regeneration. Science,276(5309), 75-81. Maxwell, E. G., Belshaw, N. J., Waldron, K. W., & Morris, V. J. (2012). Pectin–an emerging new bioactive food polysaccharide. Trends in Food Science & Technology, 24(2), 64-73. Melero-Martin, J. M., Khan, Z. A., Picard, A., Wu, X., Paruchuri, S., & Bischoff, J. (2007). In vivo vasculogenic potential of human blood-derived endothelial progenitor cells. Blood, 109(11), 4761-4768. Metcalfe, A. D., & Ferguson, M. W. (2007). Bioengineering skin using mechanisms of regeneration and repair. Biomaterials, 28(34), 5100-5113. Mohd Hilmi, A. B., Halim, A. S., Jaafar, H., Asiah, A. B., & Hassan, A. (2013). Chitosan dermal substitute and chitosan skin substitute contribute to accelerated full-thickness wound healing in irradiated rats. BioMed research international, 2013. Mohnen, D. (2008). Pectin structure and biosynthesis. Current opinion in plant biology, 11(3), 266-277. Momoh, F. U., Boateng, J. S., Richardson, S. C., Chowdhry, B. Z., & Mitchell, J. C. (2015). Development and functional characterization of alginate dressing as potential protein delivery system for wound healing. International journal of biological macromolecules, 81, 137-150. Montesano, R., Orci, L., & Vassalli, P. (1983). In vitro rapid organization of endothelial cells into capillary-like networks is promoted by collagen matrices.The Journal of cell biology, 97(5), 1648-1652. Montesano, R., Orci, L., and Vassalli, P. (1983) In vitro rapid organization of endothelial cells into capillary-like networks is promoted by collagen matrices, J Cell Biol 97, 16481652. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 89 Montesano, R., Vassalli, J. D., Baird, A., Guillemin, R., & Orci, L. (1986). Basic fibroblast growth factor induces angiogenesis in vitro. Proceedings of the National Academy of Sciences, 83(19), 7297-7301. Morita, T., & Kourembanas, S. (1995). Endothelial cell expression of vasoconstrictors and growth factors is regulated by smooth muscle cell-derived carbon monoxide. Journal of Clinical Investigation, 96(6), 2676. Morra, M., Cassinelli, C., Cascardo, G., Nagel, M. D., Della Volpe, C., Siboni, S., ... & Ulvskov, P. (2004). Effects on interfacial properties and cell adhesion of surface modification by pectic hairy regions. Biomacromolecules, 5(6), 2094-2104. Mudera, V., Smith, A. S. T., Brady, M. A., & Lewis, M. P. (2010). The effect of cell density on the maturation and contractile ability of muscle derived cells in a 3D tissue‐engineered skeletal muscle model and determination of the cellular and mechanical stimuli required for the synthesis of a postural phenotype.Journal of cellular physiology, 225(3), 646-653. Munarin, F., Giuliano, L., Bozzini, S., Tanzi, M. C., & Petrini, P. (2010). Mineral phase deposition on pectin microspheres. Materials Science and Engineering: C, 30(3), 491496. Munarin, F., Guerreiro, S. G., Grellier, M. A., Tanzi, M. C., Barbosa, M. A., Petrini, P., & Granja, P. L. (2011). Pectin-based injectable biomaterials for bone tissue engineering. Biomacromolecules, 12(3), 568-577. Munarin, F., Petrini, P., Farè, S., & Tanzi, M. C. (2010). Structural properties of polysaccharide-based microcapsules for soft tissue regeneration. Journal of Materials Science: Materials in Medicine, 21(1), 365-375. Munarin, F., Petrini, P., Tanzi, M. C., Barbosa, M. A., & Granja, P. L. (2012). Biofunctional chemically modified pectin for cell delivery. Soft Matter, 8(17), 4731-4739. Mutsaers, S. E., Bishop, J. E., McGrouther, G., & Laurent, G. J. (1997). Mechanisms of tissue repair: from wound healing to fibrosis. The international journal of biochemistry & cell biology, 29(1), 5-17. Nagel, M. D., Verhoef, R., Schols, H., Morra, M., Knox, J. P., Ceccone, G., ... & Siboni, S. (2008). Enzymatically-tailored pectins differentially influence the morphology, adhesion, cell cycle progression and survival of fibroblasts.Biochimica et Biophysica Acta (BBA)-General Subjects, 1780(7), 995-1003. Nakatsu, M. N., Sainson, R. C., Aoto, J. N., Taylor, K. L., Aitkenhead, M., Pérez-delPulgar, S., ... & Hughes, C. C. (2003). Angiogenic sprouting and capillary lumen formation modeled by human umbilical vein endothelial cells (HUVEC) in fibrin gels: the role of fibroblasts and Angiopoietin-1☆.Microvascular research, 66(2), 102-112. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 90 Nakatsu, M. N., Sainson, R. C., Pérez-del-Pulgar, S., Aoto, J. N., Aitkenhead, M., Taylor, K. L., ... & Hughes, C. C. (2003). VEGF121 and VEGF165 regulate blood vessel diameter through vascular endothelial growth factor receptor 2 in an in vitro angiogenesis model. Laboratory investigation, 83(12), 1873-1885. Nam, M. H., Lee, H. S., Seomun, Y., Lee, Y., & Lee, K. W. (2011). Monocyteendothelium-smooth muscle cell interaction in co-culture: proliferation and cytokine productions in response to advanced glycation end products.Biochimica et Biophysica Acta (BBA)-General Subjects, 1810(9), 907-912. Nehls, V., Herrmann, R., Hühnken, M., & Palmetshofer, A. (1998). Contact-dependent inhibition of angiogenesis by cardiac fibroblasts in three-dimensional fibrin gels in vitro: implications for microvascular network remodeling and coronary collateral formation. Cell and tissue research, 293(3), 479-488. Neufeld, G., Cohen, T., Gengrinovitch, S., & Poltorak, Z. (1999). Vascular endothelial growth factor (VEGF) and its receptors. The FASEB Journal, 13(1), 9-22. Neves, S. C., Gomes, D. B., Sousa, A., Bidarra, S. J., Petrini, P., Moroni, L., ... & Granja, P. L. (2015). Biofunctionalized pectin hydrogels as 3D cellular microenvironments. Journal of Materials Chemistry B, 3(10), 2096-2108. Nichol, J. W., & Khademhosseini, A. (2009). Modular tissue engineering: engineering biological tissues from the bottom up. Soft matter, 5(7), 1312-1319. Nikkhah, M., Eshak, N., Zorlutuna, P., Annabi, N., Castello, M., Kim, K., ... & Khademhosseini, A. (2012). Directed endothelial cell morphogenesis in micropatterned gelatin methacrylate hydrogels. Biomaterials, 33(35), 9009-9018. Noseda, M., Fu, Y., Niessen, K., Wong, F., Chang, L., McLean, G., & Karsan, A. (2006). Smooth muscle α-actin is a direct target of Notch/CSL. Circulation research, 98(12), 1468-1470. Noseda, M., McLean, G., Niessen, K., Chang, L., Pollet, I., Montpetit, R., ... & Karsan, A. (2004). Notch activation results in phenotypic and functional changes consistent with endothelial-to-mesenchymal transformation. Circulation research, 94(7), 910-917. Novosel, E. C., Kleinhans, C., & Kluger, P. J. (2011). Vascularization is the key challenge in tissue engineering. Advanced drug delivery reviews, 63(4), 300-311. Palamà, I. E., D'Amone, S., Coluccia, A. M., Biasiucci, M., & Gigli, G. (2012). Cell selfpatterning on uniform PDMS-surfaces with controlled mechanical cues.Integrative Biology, 4(2), 228-236. Park, S. N., Kim, J. K., & Suh, H. (2004). Evaluation of antibiotic-loaded collagenhyaluronic acid matrix as a skin substitute. Biomaterials, 25(17), 3689-3698. FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 97 7. Annexes Annexes FCUP Cell-laden micropatterns using self-assembled cell-ECM microtissues in soft pectin hydrogels 98 Figure 27. Effect of the initial seeding density on a 2.5% pectin 3D hydrogel on FBs’ spatial distibution within a 6 days culture period. Pectin surface view. FBs were stained for F-actin with Alexa Fluor 488 phalloidin (Green) and nuclei were counterstained with DAPI (Blue). Scale bars, 100 µm.