UC Berkeley

Human pluripotent stem cells (hPSCs), including both embryonic stem cells (hESCs) and induced pluripotent stem cells (hiPSCs), are undifferentiated cells that can self-renew indefinitely and have the potential to differentiate into cells of all three germ layers of the body. As such, hPSCs hold promise in understanding human development, drug screening, tissue engineering, and regenerative medicine. On one hand, current hPSC models of development offer the potential to understand and investigate the molecular mechanisms and morphogenic events that underly early human embryogenesis. However, current hPSC developmental models are limited by lack of spatiotemporal control of morphogen signaling, irreproducibility, and are typically confined to 2D with little physiological relevance. Additionally, during developmental events such as gastrulation and neurulation, it is unclear whether prescribed positional cues and boundary formation is the result of exogenous morphogen signaling gradients or the intrinsic self-organizational capacity of the epiblast or neurectoderm respectively. On the other hand, hPSCs represent a promising source to enable cell replacement therapy and treat diseases of degeneration. However, the scalable expansion and differentiation of therapeutic candidates from hPSCs is severely limited by current 2D culture methods and reliance on animal derived components. To address these limitations, we aimed to develop engineered pluripotent stem cell technologies to enable controlled and physiologically relevant hPSC models of development, including gastrulation and neurulation, as well as the scalable manufacturing of hPSC-derived cellular therapies.

In chapter 2, we combined optogenetic stimulation and single-cell imaging approaches to study self-organization during early differentiation of human pluripotent stem cells, in a 2D model of early embryonic patterning. Precise control of morphogen signal dynamics, achieved through activation of canonical Wnt/β-catenin signaling over a high dynamic range (>500-fold) using an optoWnt optogenetic system, drove broad transcriptional changes and mesendoderm differentiation of human ESCs at high efficiency (>95% cells). Furthermore, activating Wnt signaling in subpopulations of cultured ESCs in 2D induced cell self-organization and mosaic patterning of the ectoderm and mesendoderm, mediated by changes in cell migration, epithelial to mesenchymal transition (EMT), and downstream activation of TGF-β signaling. We further demonstrated that optogenetic activation of cell subpopulations can be used to uncover signaling feedback mechanisms between neighboring cell types. These findings revealed an instructive role for Wnt in directing cell segregation and established a 2D hESC model system for investigating feedback mechanisms relevant to early human embryogenesis.

In chapter 3, we investigated the role of selective Wnt activation during human gastrulation in a 3D multicellular system using optogenetics as a novel method to instruct “salt-and-pepper” morphogen activation. We aimed to evaluate the intrinsic cellular self-organization mechanisms present within the epiblast like-cells, in the absence of exogenous asymmetric signal activation. We demonstrated selective Wnt activation in a subset of hESCs is sufficient to trigger symmetry breaking and dynamic hemispherical polarization, through an accompanying EMT-like event and cadherin switching. We also demonstrated the difference between continuous vs transient TGFβ activation within a subset of the mesendoderm subsequently instructed the divergent formation of the mesoderm and endoderm, resulting in germ layer emergence and spatial patterning within the 3D gastruloids. These findings supported the hypothesis hPSCs have the intrinsic ability to self-organize and may inform the role of the epiblast population in the developing human embryo prior to gastrulation. Additionally, these findings suggested that selective Wnt activation is sufficient for symmetry breaking and germ layer emergence. Overall, the 3D optoWnt gastruloid model may be useful for further probing the molecular mechanisms underlying successful human gastrulation.

In chapter 4, we investigated the spatiotemporal role of controlled and selective Wnt activation underlying human neurulation using optogenetics. We reported a novel approach to model neural tube patterning and boundary formation through optogenetic stimulation of canonical Wnt signal activation in hPSCs. We demonstrated the dose dependent nature of optoWnt activation in response to discrete light intensity, and the ability to differentially instruct neural tube progenitor cells of the A-P neural tube axis using only blue light stimulation. We successfully derived NPCs from the forebrain, midbrain, and hindbrain through dose-dependent light activation of optoWnt signaling in a dynamic range of 0.0-1.5 µW/mm2, correlating to a similar range observed with 0-5 μm of a chemical Wnt agonist. This dynamic range may be useful for photopatterned gradients or non-uniform projections, to study the spatiotemporal role of Wnt signaling underlying neural tube patterning. Finally, we reported the presence of self-organizational mechanisms underlying A-P axis patterning and potential MHB establishment in 2D and 3D mixed hPSC co-culture models of selective optoWnt activation. Together, these results represented proof-of-concept studies into developing hPSC models of neural tube patterning leveraging optogenetic control.

In chapter 5, we moved beyond hPSC developmental models into enabling technologies for manufacturing of hPSC-derived therapeutics. We reported a fully defined, reproducible, and tunable thermoreversible hydrogel platform for high quality, scalable 3D cell production. Our novel synthesis method allowed for precise gelation temperature control (24-32°C), hydrogel stiffness tuning (100-4000 Pa), and improved reproducibility via prevention of any unintended covalent crosslinking. After material optimization, we demonstrated the cell production potential of the hydrogel for hPSC expansion, pluripotency maintenance, and differentiation into cells of numerous lineages. Finally, we demonstrated the ability to functionalize the polymer, to conjugate bioactive components for local presentation and control. Overall, the novel 3D thermoreversible hydrogel platform has broad applications in scalable, high quality cell production to overcome the biomanufacturing burden of stem cell therapy as well as enabling investigations into the role of the ECM in hPSC tissue models.

In summary, we engineered optogenetic technologies to develop advanced hPSC models of human gastrulation and neurulation and biomaterial technologies to enable the scalable production of hPSC-derived therapeutics. These advances may promote future studies into the mechanisms underlying human development as well as the clinical and commercial impact of stem cell therapies.