Micropattern technology applied to the in vitro study of the early embryonic development

Micropattern technology, which enables control of cell and tissue architecture in vitro has been demonstrated as a useful and efficient tool for modeling the microenvironments at different scales and complexities. In the last 20 years, scientists have benefited a lot in revealing and dissecting the mechanism of communication between cells and the surrounding tissues and leading to the function from the breakthroughs in micropattern technology. Moreover, micropattern technology allows users to culture cells under well-defined geometric confinement by controlling cell shape, size, position, or multi-layered architecture. From the study of cell biology and developmental biology, we know that both geometric and mechanical cues present in the microenvironment affect cell behavior a lot. However, there is no possibility that we can test both these cues under standard tissue culture. Nowadays, many micropattern methods are available to address this problem at various scales. Generally speaking, these new methods provide a powerful platform for asking fundamental and mechanism questions in cell biology and tissue engineering, including cell survival, proliferation, differentiation, cell migration, cytokinesis, and cell polarity. Human pluripotent stem cells (hPSCs), including embryonic stem cells (hESCs) and induced pluripotent stem cells (hiPSCs), are widely used in regenerative medicine as well as experimental model of normal and diseased organogenesis because of their nearly pluripotent differentiation potential into all cell lineages of all the three germ layers: endoderm, ectoderm and mesoderm germ layers. As we know, the differentiation fates of hPSCs are highly sensitive to local environmental factors that can modulate autocrine or paracrine signaling as well as mechanotransduction processes mediated by physical cues. Cell micropatterning encompasses a set of technical strategies that have been developed to spatially organize the geometry and location of a cell population with the purpose to control the local cellular microenvironment, such as cell-cell and cell-matrix interactions. In the context of hPSCs, cell micropattern has been employed to gain significant insights into how geometric and chemical cues modulate cell fates decision and cell organization into early embryonic differentiation patterns. At the same time, 2D and 3D micropatterned hPSCs have been used to control the colony size of multicellular patterns, which in turn will influence differentiation decisions into three germ layers. In recent years, numerous cell micropattern methods have been established and developed, but only very few, such as microcontact printing, micro-well culture, photo-patterning, and micro-stencil, have been successfully applied to micropatterned hPSCs. The challenge with micropatterned hPSCs lies in their fragility and a most stringent requirement of the microenvironment which include the specific extracellular matrix (ECM) and growth conditions for cell adhesion and survival. To date, the use of micropattern has shown that the self-organization of hESCs can be influenced by both geometric and chemical cues and generate several ring-like cell populations of different cell-fates, similar to those observed at gastrulation. These self-organizing patterns emerge as a consequence of the interaction between receptor localization and the production of the BMP-inhibitor NOGGIN. This system represents an in vitro model ideally suited to reveal the complex interaction between signaling, fate, and shape, as well as explore symmetric-breaking events and the self-organization properties of pluripotent stem cells. In response to specific factors, for example, dual-Smad inhibitor and WNT inhibitor, the micropatterned hESCs can be differentiated into neural progenitors and primitive streak-like populations respectively. Interestingly, micropattern technology can also be applied to regenerative medicine, for example, the micropatterned organizer cells can be transplanted into the chicken embryo and subsequently induces a secondary axis which later initiates a neural fate in the host. As a conclusion, micropattern technology applied to the in vitro study can help us understand and reveal the secret of human embryonic development in various ways. In the meantime, more and more tissue engineering methods, include both 2-dimensional and 3-dimensional, will be established and combined with micropattern technology to develop a well-defined microenvironment, which will help people to generate more complex in vitro model. Here in this thesis, we have established our micropattern technique by applying a fast and convenient surface functionalization procedure. By using different photomasks, we can make the micropatterns in various shapes, and sizes ranged from 50μm to 1000μm. Then a two-steps of surface Poly-L-lysine and ECM coating is necessary to generate the cell culture substrate. For the micropatterned cell culture, we tested and modified the protocol, and now we can harvest stable and well-formed cell colonies in culture for more than 8 days. With the purpose to investigate how the neural induction will be affected when under geometric confinement, we performed a micropatterned neural induction experiment by using a dual-Smad inhibitor neural induction protocol we have developed previously. In standard cell culture, it has been reported and demonstrated that the dual inhibitions of Smad signaling is highly efficient in the neural conversion of both hESCs and hiPSCs. The synergistic action of two inhibitors, SB431542 and NOGGIN, is sufficient to induce rapid (~6days) and complete (>80%) neural conversion under adherent culture conditions. Also in the same work, the future date suggested that the cell density, which means the initial seeding density, influences the outcome of cell fates of neural induction significantly: High seeding density promoted cell fate presents central nervous system while low seeding density promoted neural crest cell fate. So, we hypothesis that, our micropatterned neural induction platform could be useful to generate different cell fates located individually along the colony axis, and this different allocation could be an in vitro model to mimic the patterning of ectoderm. As we expected, we found that in the micropatterned neural induction colony, cells self-organized themselves into 3 main populations from inner to outer. In the center of the colony, cells showed a relatively low density and expressed both AP-2α and P75 (neural crest markers). On the contrary, cells outside expressed NESTIN, SOX1, and PAX6 (neural progenitor markers) and distributed as a ring structure between the center and border. This population presents a central nervous system cell fate. By comparing the cell density distributed form the center to the border, we found that the low density promotes neural crest cell fate in the center while high density promotes central nervous system cell fate outside, and the cells at the border had a more compact morphology and highly expressed only NESTIN, this cell population presents the surface ectoderm cell fate. General speaking, our micropatterned neural induction model can be used as an in vitro platform to mimic the human ectodermal patterning. Later on, we extended our experimental model to establish an in vitro co-culture system with the purpose to investigate how the 3 germ layers communicate with each other during embryogenesis. During this period, a new hESCs-GFP cell line was established by Lentivirus infection, and it was used in the co-culture system to mark the cells of pre/sub seeded. To mimic the co-cultured mesodermal and endodermal cells, we developed a meso-endoderm differentiation protocol under standard cell culture condition, and this meso-endoderm population can be seeded on top of the neuroectoderm cell population to simulate the in vivo architecture. Interestingly, we found that the vast majority subsequently seeded meso-endoderm cells can only adhere to the border of neuroectoderm fate, and they arranged into a ring-like structure close to PAX6+ cells. This adhesion property may be determined by the intrinsic differences between specific cell fates of ectoderm. Moreover, we also found that when the meso-endoderm cells were co-cultured with neuroectoderm, a new cell population can be generated from sub-seeded cells, and they all co-expressed PAX6 but are not exist in meso-endoderm cell culture. Additionally, with 3 days’ co-culture, we surprisingly found some cells located above the PAX6+ neuroectoderm cells, and they showed a totally different cell morphology form other cell populations. It seems these cells were self-organized into a linear morphology and became the connecting cross-structure in this co-culture system. At all events, this co-culture system we developed has been demonstrated a robust platform in the study of interaction and communication between different germ layers. To investigate the self-organization ability of subsequently seeded meso-endoderm, we next performed a micropatterned meso-endoderm cell culture. Briefly, we initiated meso-endoderm induction by applying the same protocol with 24 hours and then seeded this mixed population on the same micropatterns. To simulate the same condition, we repeated all the same seeding density, medium, and medium change for 3 days. Unexpectedly, the micropatterned meso-endoderm cells differentiated into distinct different cell populations with diverse morphologies. Generally, cells in the colony center showed a compact and multi-layered structure while cells at the border differentiated into many linear structures. As a conclusion, it is evident that the meso-endoderm cell fate has intrinsic self-organization property under geometric confinement, but how the organization is affected when co-cultured with neuroectoderm are still not clear. From this perspective, we are facing the opportunity and challenging at the same time, and more research activities are necessary for the coming future.