Engineering osteochondral tissues with human adipose tissue derived stem cells under precise biomechanical and biochemical in vitro environments

Tese de doutoramento em Engenharia BiomédicaWith the continuous increase in life expectancy, health problems tend to arise, whereas joint ailments are among the most prevalent. The knee joint, in particular, is subjected to serious illnesses that range from articular cartilage injuries, osteochondral lesions and osteoarthritis. Joint pathologies globally affect population of all ages and gender, resulting in reduced quality of life due to limited activity and increased expenditures, either directly (treatment, private expenditure), or indirectly (lost productivity, lost earnings). In this context, the need for new therapies and products that could satisfy the growing clinical requirements will rise in the future. Tissue engineering and regenerative medicine, as a rapidly emerging field, is expected to provide valuable solutions. Osteochondral lesions are articular cartilage lesions where the underlying bone is also damaged. A tissue engineering approach needs to address the cartilage component, composed by a hydrated soft tissue layer, and the bone compartment, composed by a stiff, complex and vascularized tissue. In this thesis we propose complementary approaches to evolve current state of the art, and move a step further in the development of functional solutions for cartilage, bone and/or osteochondral regeneration. Our rationale was based on the use of human adipose stem cells (hASC) as a single cell source to engineer all tissue compartments, by taking advantage of its characteristics: 1) hASC have intrinsic capacity to differentiate into the chondrogenic, osteogenic and endothelial lineages; 2) hASC can be obtained from adipose tissue collected by dedicated or non-dedicated liposuction procedures, repeatedly and abundantly; 3) hASC are isolated by enzymatic digestion, yielding high cell number, which may avoid or minimize cell expansion. Furthermore, we explored the use of precise biomechanical and biochemical environments specific for each engineered tissue, in order to improve cell differentiation and matrix deposition. Specifically, hydrostatic pressure stimulation was employed for cartilage engineering, while flow perfusion and inherent shear stress were investigated in bone development context. In order to vascularize engineered bone, a specific spatio-temporal regulation of growth factors and cells were thoroughly explored. To this end, five experimental studies were performed. The first studied focused on the development of two bioreactor devices aimed at generating hydrostatic pressure (HP) for dynamic culturing of cartilage tissue. We hypothesized that the formation of engineered cartilage could be augmented by applying such physiologic stimuli to chondrogenic cells (human nasal chondrocytes - HNC) or stem cells (human adipose stem cells - hASC), cultured in gellan gum hydrogels, by varying both frequency and magnitude of loading. In the HNC study, the best tissue development was achieved for pulsatile HP regimen, while in the hASC study, the best cartilage outcomes were obtained for physiologic loading (5 MPa), as evidenced by gene expression of aggrecan, collagen type II and sox-9, metachromatic staining of cartilage matrix and immunolocalization of collagens. The next step aimed to evaluate the effects of scaffold architecture and biomechanics, in order to optimize silk scaffolds for bone tissue engineering. Silk scaffolds were fabricated using different solvents (aqueous vs. hexafluoro-2-propanol - HFIP), pore sizes (250-500μm vs. 500- 1000μm) and structures (lamellar vs. spherical pores). Given the great potential of hASC for cell-based therapies and tissue engineering, in particular bone tissue, silk scaffold and hASCs are two promising components, which have not been previously investigated in combination. The porous HFIP silk scaffold with 400-600 μm pores performed better than any other scaffold, while the lamellar scaffolds performed better than spherical-pore scaffolds. We further used this HFIP-silk scaffold as cell support for dynamic culturing studies, where the effects of pulsatile perfusion on in vitro bone expression by human adipose stem cells (hASCs) was assessed. We hypothesized that the formation of engineered bone could be augmented by replicating physiologic stimuli – pulsatile interstitial flow - to cells cultured in porous scaffolds using bioreactors with medium perfusion. This was confirmed, once the best tissue development was achieved for the sequence of 2 weeks of steady flow and 3 weeks of pulsatile flow, as evidenced by gene expression, construct compositions, histomorphologies and biomechanical properties. Further challenge was to vascularize the engineered bone grafts. Even more demanding was to use the same cell source – hASC. From a clinical perspective, it would be ideal to engineer vascularized bone grafts starting from one single cell harvest obtained from the patient. We hypothesized that a sequential application of osteogenic and endothelial growth factors to hASC cultured on biomaterial scaffolds (HFIP-silk scaffold), with different timing of addition of fresh cells could support the development of bone-like tissue containing an integrated vascular To this end, five experimental studies were performed. The first studied focused on the development of two bioreactor devices aimed at generating hydrostatic pressure (HP) for dynamic culturing of cartilage tissue. We hypothesized that the formation of engineered cartilage could be augmented by applying such physiologic stimuli to chondrogenic cells (human nasal chondrocytes - HNC) or stem cells (human adipose stem cells - hASC), cultured in gellan gum hydrogels, by varying both frequency and magnitude of loading. In the HNC study, the best tissue development was achieved for pulsatile HP regimen, while in the hASC study, the best cartilage outcomes were obtained for physiologic loading (5 MPa), as evidenced by gene expression of aggrecan, collagen type II and sox-9, metachromatic staining of cartilage matrix and immunolocalization of collagens. The next step aimed to evaluate the effects of scaffold architecture and biomechanics, in order to optimize silk scaffolds for bone tissue engineering. Silk scaffolds were fabricated using different solvents (aqueous vs. hexafluoro-2-propanol - HFIP), pore sizes (250-500μm vs. 500- 1000μm) and structures (lamellar vs. spherical pores). Given the great potential of hASC for cell-based therapies and tissue engineering, in particular bone tissue, silk scaffold and hASCs are two promising components, which have not been previously investigated in combination. The porous HFIP silk scaffold with 400-600 μm pores performed better than any other scaffold, while the lamellar scaffolds performed better than spherical-pore scaffolds. We further used this HFIP-silk scaffold as cell support for dynamic culturing studies, where the effects of pulsatile perfusion on in vitro bone expression by human adipose stem cells (hASCs) was assessed. We hypothesized that the formation of engineered bone could be augmented by replicating physiologic stimuli – pulsatile interstitial flow - to cells cultured in porous scaffolds using bioreactors with medium perfusion. This was confirmed, once the best tissue development was achieved for the sequence of 2 weeks of steady flow and 3 weeks of pulsatile flow, as evidenced by gene expression, construct compositions, histomorphologies and biomechanical properties. Further challenge was to vascularize the engineered bone grafts. Even more demanding was to use the same cell source – hASC. From a clinical perspective, it would be ideal to engineer vascularized bone grafts starting from one single cell harvest obtained from the patient. We hypothesized that a sequential application of osteogenic and endothelial growth factors to hASC cultured on biomaterial scaffolds (HFIP-silk scaffold), with different timing of addition of fresh cells could support the development of bone-like tissue containing an integrated vascularTo this end, five experimental studies were performed. The first studied focused on the development of two bioreactor devices aimed at generating hydrostatic pressure (HP) for dynamic culturing of cartilage tissue. We hypothesized that the formation of engineered cartilage could be augmented by applying such physiologic stimuli to chondrogenic cells (human nasal chondrocytes - HNC) or stem cells (human adipose stem cells - hASC), cultured in gellan gum hydrogels, by varying both frequency and magnitude of loading. In the HNC study, the best tissue development was achieved for pulsatile HP regimen, while in the hASC study, the best cartilage outcomes were obtained for physiologic loading (5 MPa), as evidenced by gene expression of aggrecan, collagen type II and sox-9, metachromatic staining of cartilage matrix and immunolocalization of collagens. The next step aimed to evaluate the effects of scaffold architecture and biomechanics, in order to optimize silk scaffolds for bone tissue engineering. Silk scaffolds were fabricated using different solvents (aqueous vs. hexafluoro-2-propanol - HFIP), pore sizes (250-500μm vs. 500- 1000μm) and structures (lamellar vs. spherical pores). Given the great potential of hASC for cell-based therapies and tissue engineering, in particular bone tissue, silk scaffold and hASCs are two promising components, which have not been previously investigated in combination. The porous HFIP silk scaffold with 400-600 μm pores performed better than any other scaffold, while the lamellar scaffolds performed better than spherical-pore scaffolds. We further used this HFIP-silk scaffold as cell support for dynamic culturing studies, where the effects of pulsatile perfusion on in vitro bone expression by human adipose stem cells (hASCs) was assessed. We hypothesized that the formation of engineered bone could be augmented by replicating physiologic stimuli – pulsatile interstitial flow - to cells cultured in porous scaffolds using bioreactors with medium perfusion. This was confirmed, once the best tissue development was achieved for the sequence of 2 weeks of steady flow and 3 weeks of pulsatile flow, as evidenced by gene expression, construct compositions, histomorphologies and biomechanical properties. Further challenge was to vascularize the engineered bone grafts. Even more demanding was to use the same cell source – hASC. From a clinical perspective, it would be ideal to engineer vascularized bone grafts starting from one single cell harvest obtained from the patient. We hypothesized that a sequential application of osteogenic and endothelial growth factors to hASC cultured on biomaterial scaffolds (HFIP-silk scaffold), with different timing of addition of fresh cells could support the development of bone-like tissue containing an integrated vascular network. Three strategies were evaluated by changing spatio-temporal cues, but only one of the combinations, in particular the osteo-induction of hASC seeded to silk scaffold for 3 weeks, followed by addition of fibrin-encapsulated hASC to which vasculogenic cues were provided for 2 weeks, resulted in the most promising outcomes towards vascularized bone grafts. The final experimental design focused on the development of an in vitro model for studies of heterotypic cellular interactions that couple blood vessel formation with osteogenesis by using human umbilical vein endothelial cells (HUVECs) and human bone marrow mesenchymal stem cells (hMSCs). In this study, we hypothesized that the sequential application of growth factors, to firstly induce the formation of stable vasculature and subsequently initiate osteogenic differentiation, could provide a biologically-inspired in vitro model of bone vascularization. Two important findings resulted from these studies: (i) vascular development needs to be induced prior to osteogenesis, and (ii) the addition of additional hMSCs at the osteogenic induction stage improves both tissue outcomes, and anastomosis of vascular networks with the host (SCID mice) vasculature. Taken altogether, the results obtained during the accomplishment and completion of this thesis prove the successful use of human adipose stem cells for osteochondral tissue engineering, as mechanically responsive cell source, which, in combination with appropriate growth factors, generate both cartilage and bone compartments. Besides the success obtained with dynamic culturing, it was proven that hASC have great potential to be used as single cell source for the development of vascularized bone grafts.network. Three strategies were evaluated by changing spatio-temporal cues, but only one of the combinations, in particular the osteo-induction of hASC seeded to silk scaffold for 3 weeks, followed by addition of fibrin-encapsulated hASC to which vasculogenic cues were provided for 2 weeks, resulted in the most promising outcomes towards vascularized bone grafts. The final experimental design focused on the development of an in vitro model for studies of heterotypic cellular interactions that couple blood vessel formation with osteogenesis by using human umbilical vein endothelial cells (HUVECs) and human bone marrow mesenchymal stem cells (hMSCs). In this study, we hypothesized that the sequential application of growth factors, to firstly induce the formation of stable vasculature and subsequently initiate osteogenic differentiation, could provide a biologically-inspired in vitro model of bone vascularization. Two important findings resulted from these studies: (i) vascular development needs to be induced prior to osteogenesis, and (ii) the addition of additional hMSCs at the osteogenic induction stage improves both tissue outcomes, and anastomosis of vascular networks with the host (SCID mice) vasculature. Taken altogether, the results obtained during the accomplishment and completion of this thesis prove the successful use of human adipose stem cells for osteochondral tissue engineering, as mechanically responsive cell source, which, in combination with appropriate growth factors, generate both cartilage and bone compartments. Besides the success obtained with dynamic culturing, it was proven that hASC have great potential to be used as single cell source for the development of vascularized bone grafts. investigados no contexto do desenvolvimento ósseo. A fim de vascularizar o enxerto ósseo desenvolvido, foram exploradas condições específicas de regulação espaço-temporal de células e fatores de crescimento. Para este propósito, cinco estudos experimentais foram realizados. O primeiro estudo focado no desenvolvimento de dois bioreatores visou gerar pressão hidrostática (HP) para a cultura dinâmica do tecido cartilaginoso. Consideramos como hipótese que a formação de cartilagem poderia ser aumentada através da aplicação de tais estímulos fisiológicos, tanto em células primárias (condrócitos) (septo nasal humano - HNC) ou células-estaminais (células estaminais do tecido adiposo humano - hASC), cultivadas em hidrogéis de goma gelana, variando a frequência e magnitude de carga. No estudo HNC, o melhor desenvolvimento de tecido foi conseguido para o regime de HP pulsátil, enquanto que no estudo hASC, os melhores resultados foram obtidos através da aplicação de níveis fisiológicos de HP (5 MPa), como evidenciado pela expressão genética de agrecano, colagénio tipo II e sox-9, assim como através da coloração metacromática da matriz da cartilagem e imunolocalização de colagénios. O próximo passo pretendia avaliar os efeitos da arquitectura e propriedades biomecânicas de suportes, a fim de otimizar suportes de seda para a engenharia de tecido ósseo. Suportes de seda foram fabricados usando diferentes solventes (aquoso vs hexafluoro-2-propanol - HFIP), tamanhos de poros (250-500μm versus 500-1000μm) e estruturas (poros lamelar versus esférica). Dado o grande potencial de hASC para terapias celulares e engenharia de tecidos, em especial do tecido ósseo, os suportes de seda e hASC constituem dois componentes promissores, que ainda não foram previamente investigados em combinação. O suporte composto por seda-HFIP, com poros esféricos de 400-600 μm, demonstrou melhor desempenho do que qualquer outro suporte, enquanto os suportes com estrutura lamelar demonstraram melhor desempenho do que os suportes com poros esféricos. Este suportes de seda-HFIP foram posteriormente utilizados para os estudos de cultura dinâmica, onde foram avaliados os efeitos da perfusão pulsátil sobre a expressão óssea por hASC. Consideramos como hipótese que a formação de osso poderia ser aumentada através da replicação de estímulos fisiológicos - fluxo intersticial pulsátil - para as células cultivadas em suportes porosos utilizando bioreatores com perfusão do meio de cultura. Esta hipótese foi confirmada, uma vez que o melhor desenvolvimento de tecido foi obtido para a sequência de 2 semanas de fluxo constante e 3 semanas de fluxo pulsátil, como evidenciado pela expressão genética, composição do enxerto, histomorfologias e propriedades biomecânicas. O desafio seguinte consistia em vascularizar os enxertos ósseos desenvolvidos. Mais exigente foi o uso da mesma fonte celular - hASC. Numa perspectiva clínica, o ideal seria desenvolver enxertos ósseos vascularizados a partir de uma colheita celular única obtida do paciente. Consideramos como hipótese que uma aplicação sequencial de fatores de crescimento endoteliais e osteogénicos em hASC cultivadas em biomateriais (suportes de seda-HFIP), com diferentes períodos de adição de células frescas, poderia beneficiar o desenvolvimento de osso contendo uma rede vascular integrada. Três estratégias foram avaliadas mas apenas uma das combinações forneceu resultados promissores para o desenvolvimento de enxerto ósseo vascularizado, em particular a osteo-indução das hASC aderidas ao suporte de seda durante 3 semanas, seguido pela adição de hASC encapsuladas em fibrina, às quais foram fornecidas fatores vasculogénicos durante 2 semanas. O desenho experimental final concentrou-se no desenvolvimento de um modelo in vitro para estudo de interações celulares heterotípicas que combinem formação de vasos sanguíneos com osteogénese, usando células endoteliais do cordão umbilical humano (HUVECs) e células-estaminais mesenquimatosas da medula óssea humana (hMSCs). Neste estudo, testamos a hipótese da aplicação sequencial de fatores de crescimento, em primeiro lugar, induzir a formação de vasos estáveis e, posteriormente, iniciar a diferenciação osteogénica, poder fornecer um modelo in vitro de vascularização óssea, biologicamente inspirado. Duas conclusões importantes resultaram deste estudo: (i) o desenvolvimento vascular necessita de ser induzida antes da osteogénese, e (ii) a adição de hMSCs adicionais na fase de indução osteogénica melhora os resultados do tecido, assim como a anastomose das redes vasculares com a vasculatura do animal recetor (ratinhos SCID). Analisados em conjunto, os resultados obtidos durante a realização e conclusão desta tese provam o sucesso do uso de células estaminais do tecido adiposo humano para engenharia de tecidos osteocondrais, como fonte de células mecanicamente sensíveis às distintas forças de estimulação biomecânica, que, em combinação com os fatores de crescimento adequados, desenvolveram os compartimentos cartilagíneo e ósseo. Além do sucesso obtido com a cultura dinâmica, foi comprovado que as hASC apresentam grande potencial para serem utilizadas como fonte celular única para o crescimento de enxertos ósseos vascularizados.Financially, I want to thank the Portuguese Foundation for Science and Technology (FCT) for providing me the PhD grant (SFRH/BD/42316/2007), as well as funding from the American National Institute of Health (NIH) for GVN’s lab (DE161525 and EB02520), which allowed all this to happen