Until recently, mechanistic studies of development and disease have relied on animal model systems. With the discovery of human embryonic stem cells (hESCs), then induced pluripotent stem cells (iPSCs), and finally organoids, mechanistic discovery-based research on human tissues is now routine. Cells and tissues derived from hESCs and iPSCs, collectively called human pluripotent stem cells (hPSCs), have many advantages for research. They are scalable and can be genetically manipulated, derived from genetically diverse populations and used to model patient-specific pathology. Moreover, hPSCs can be differentiated into individual cell types, or they can be used to engineer highly complex, functional tissues. Finally, hPSCs are human and allow for studies of human development and diseases in ways that were previously impossible. Here, we discuss three preprints that take advantage of the experimental strengths of hPSC-derived tissues to investigate new mechanisms in development and disease. In one preprint (Wong et al., 2022 preprint), the authors have developed an hPSC model to investigate the mechanisms by which a rare and transient population of cells, the ventral foregut endoderm, becomes committed to specific endoderm lineages. We also discuss two preprints (Victor et al., 2022 preprint; Zhao et al., 2022 preprint) describing how hPSCs are being used to identify which cell types in the brain are driving the pathology of a genetically caused form of Alzheimer's disease.
Human development
Organ development across species involves reiterating cycles of expansion of proliferative progenitor cells followed by lineage commitment and differentiation. Modeling these processes during mammalian development is technically challenging given the inaccessibility of mammalian embryos for live imaging in utero and the paucity of material for biochemical and molecular studies of mechanisms. One example of a rare and experimentally challenging population of progenitor cells are foregut endoderm cells, which give rise to the thyroid, lungs and liver, as well as a portion of the esophagus, stomach, pancreas and duodenum. Past studies of the mammalian ventral foregut have relied on tedious microdissection (Deutsch et al., 2001; Havrilak et al., 2017; Jennings et al., 2017; Jung et al., 1999; Rossi et al., 2001; Wells and Melton, 2000) or complicated cell lineage and genetic analyses (Lawson et al., 1986; Spence et al., 2009; Tremblay and Zaret, 2005). Given these challenges, relatively few mechanistic studies of early endoderm patterning and organ specification have been performed in mammals. One technically courageous approach used dissected mouse foregut and concluded that epigenetic pioneer factors act to prime genes that regulate regional identity, which may subsequently drive lineage commitment (Iwafuchi-Doi et al., 2016; Reizel et al., 2021; Xu et al., 2011; Zaret et al., 2008). However, relatively few labs are technically able or willing to perform such difficult experiments.
A recent preprint by Wong et al. (2022 preprint) perfectly exemplifies how hPSC culture systems have breathed new life into experimentally intractable questions, such as the mechanisms that drive organ development from the ventral foregut. Wong et al. discovered that hPSC-derived definitive endoderm progenitors (EPs) expanded in culture were developmentally similar to ventral foregut cells. A single-cell RNA-sequencing comparison of EPs with human foregut showed a high degree of similarity with ventral foregut, including expression of HHEX, a canonical marker of ventral foregut across vertebrates. Expanded EPs in culture robustly formed ventral foregut derivatives, including pancreas and liver and, as such, were renamed ventral foregut (VFG). These culture systems were used to study how the expansion of VFG progenitors was involved in specific gene enhancer modifications that were driving organ lineage commitment.
They identified classes of enhancer elements that became poised and activated in the VFG that were similar to tissues isolated from human embryos (Gerrard et al., 2020; Jennings et al., 2017). Some classes of poised enhancers were either pancreatic or hepatic, as they were activated when VFG cells were directed into pancreatic or liver fate decisions, respectively. Other classes of enhancers associated with mesoderm or ectoderm were simultaneously repressed. Consistent with previous reports (Iwafuchi-Doi et al., 2016; Reizel et al., 2021; Xu et al., 2011), the FOXA family of pioneering transcription factors were functionally required to establish the proper epigenetic landscape (prepattern) for robust activation of pancreas and liver enhancers in the human foregut.
One particularly important feature of this work is how the authors consistently compare in vitro findings to data from human embryonic cells and tissues. Benchmarking hPSC-derived systems to embryos is essential to build confidence that these new in vitro systems are accurately modeling what happens in vivo. Although this and other studies are just beginning to reveal how VFG organs form, there are many mechanistic questions that can be explored using this new experimentally tractable system.
Human disease modeling
iPSC technologies provide unprecedented opportunities to study the genetic basis of human diseases in primary human cells and tissues. For example, the apolipoprotein E ε4 allele (APOE4) is the greatest known genetic risk factor for developing late-onset Alzheimer's disease (AD); however, human APOE4 transgenic mouse models (in which murine Apoe is replaced with human APOE4) are not sufficient to recapitulate AD-like pathology without being crossed with familiar Alzheimer disease (FAD) transgenic mice (Jankowsky and Zheng, 2017; Sanchez-Varo et al., 2022). Here, we highlight two recent preprints (Victor et al., 2022 preprint; Zhao et al., 2022 preprint) using human iPSC-derived models to show that defects in APOE4-mediated lipid metabolism in non-neural glial cells cause impaired neuronal development and may contribute to the pathology observed in AD patients.
To assess the causal relationship between disease-associated genotypes and phenotypes, both Victor et al. and Zhao et al. created isogenic iPSCs lines of APOE3/3, APOE4/4 and APOE knockouts using CRISPR/Cas9 genome-editing. Whereas APOE4 alters tau phosphorylation and Aβ production in iPSC-derived neurons (Lin et al., 2018; Wang et al., 2018), iPSC-derived microglial cells carrying APOE4/4 accumulated lipids and diminished cholesterol secretion compared with those with APOE3/3. This possibility of cell type-specific roles of APOE was investigated in the preprint by Zhao et al. (2022 preprint). This work used unbiased single-cell transcriptomics analysis of iPSC-derived cerebral organoids with distinct cell diversity, including radial glial cells, excitatory neurons, inhibitory neurons, and astrocytes. Unexpectedly, the authors found APOE deficiency alters brain cell composition by shifting neurogenesis to astrogenesis, a pathology that could be reversed using the EIF2 signaling pathway inhibitor ISRIB. Importantly, this work confirmed cell type-specific cholesterol metabolism dysregulation in iPSC-derived cerebral organoids carrying APOE4/4, enhanced cholesterol biosynthesis in excitatory neurons and excessive lipid accumulation in astrocytes.
The brain is a complex mix of neurons, astrocytes and microglia. The role of microglia in driving AD pathology in these APOE patients is unclear. By replacing the disease-neutral allele APOE3 with the disease-causing allele APOE4 specifically in microglia, Victor et al. (2022 preprint) showed that APOE4 induces an impaired lipid metabolic program in iPSC-derived microglia, rendering APOE4 microglia weakly responsive to neuronal activity. The authors further demonstrated the deranged microglial lipolysis could exacerbate pro-inflammatory signals in APOE4 microglia. Importantly, APOE4 microglia impaired neurotrophic functions, for example by disrupting coordinated neural network activities when co-cultured with conditioned-media-treated neurons and neural spheroids. This work suggested that AD disease pathology might involve glial–neuronal crosstalk.
In summary, these preprints are exemplars of the unique experimental strengths of iPSC technologies to study human development and disease. Wong et al. used iPSC-derived VFG endoderm cell culture to provide new insight into the molecular events that link progenitor cell proliferation to organ lineage commitment. Victor et al. and Zhao et al. used AD patient-derived and CRISPR-corrected isogenic iPSC lines to dissect how different cell types in the brain are impacted by APOE mutations and how they in turn contribute to AD pathologies. Moving forward, these in vitro systems will enable new studies of molecular mechanisms, dissection of cell–cell crosstalk and further our understanding of how early developmental pathologies can predispose patients to AD later in life. Lastly, these systems could be used to identify potential therapeutic targets to halt or slow down disease progression.
Note added in proof
Victor et al., 2022 has now been published as: Victor, M. B., Leary, N., Luna, X., Meharena, H. S., Scannail, A. N., Bozzelli, P. L., Samaan, G., Murdock, M. H., von Maydell, D., Effenberger, A. H., et al. (2022). Lipid accumulation induced by APOE4 impairs microglial surveillance of neuronal-network activity. Cell Stem Cell. 29, 1197-1212.e8. doi:10.1016/j.stem.2022.07.005.