Bioengineered in vitro Model for Peri-implantation Human Embryogenesis

Abstract

Implantation is a critical developmental milestone for early human embryogenesis and successful pregnancy. During implantation, the pluripotent epiblast gives rise to the squamous amnion and the columnar embryonic disc, which together enclose the amniotic cavity to form an asymmetric cystic structure called the amniotic sac. Amniogenesis - the development of amnion - marks the first differentiation of the epiblast during implantation; in parallel, the formation of the amniotic sac delineates the morphology of the human embryo and prepares it for subsequent gastrulation. Despite its fundamental and clinical importance, the development of the amnion and the amniotic sac in humans remains mysterious due to the scarcity and restricted availability of in vivo human embryo specimens and the lack of in vitro models. In this dissertation, we report the first in vitro model for peri-implantation human amniogenesis and amniotic sac development, by culturing human pluripotent stem cells (hPSCs) in a bioengineered niche that mimics the physical microenvironment experienced by human embryos during implantation. Specifically, we find that hPSCs cultured in such implantation-like niche undergo self-organized development and form three dimensional (3D) structures in vitro that molecularly and morphologically resemble human amnion and human amniotic sac in vivo. We further show that implantation-like physical niche cues - a soft tissue bed and a 3D extracellular matrix (ECM) - are both necessary and sufficient for triggering the amniogenic development in an otherwise self-renewal-permissive biochemical context. We also demonstrate that the hPSC-derived amniotic sac embryoid can model human amniotic sac development beyond the peri-implantation stage with primitive streak-like development. Additionally, we unveil an endogenous BMP-SMAD signaling underlying peri-implantation human amniogenesis and amniotic sac patterning. Furthermore, we demonstrate an integrative mechanotransduction that can induce amniogenesis of hPSCs by integrating multiple upstream signals including substrate rigidity, cell cluster size, and cell spreading. Enabled by such new knowledge, we have developed a novel micro-engineered array of hPSC-derived amnion-like tissues, as a potential high-throughput screening platform for regenerative medicine. Together, findings in this dissertation provide innovative in vitro systems for investigating early human embryogenesis during implantation and early gastrulation, thereby helping advance human embryology and reproductive and regenerative medicine.