Model Stem and Progenitor Cell Systems

"All models are wrong, but some are useful." –George Box

The most commonly used experimental models for human biology are serum-derived cancer cell lines, such as HeLa cells. While convenient, these cell lines often bear little resemblance to their tissue or cancer of origin due to extensive adaptation to typical lab growth conditions (e.g., non-physiological levels of nutrients and signaling molecules, aberrant growth on artificial substrates, etc.). Therefore, their applicability to mammalian biology is questionable. As a result, we have moved away from their use. Instead, we experiment with stem and progenitor cell types that naturally divide and proliferate during mammalian development or tissue homeostasis. Now, many stem and progenitor cell types can be enticed to grow in the lab using defined two- or three-dimensional "niches" that allow maintenance of their cell identity, molecular signatures, and regulatory networks. Over the years we have worked with four different stem/progenitor cell systems on a variety of biological questions. These include: embryonic stem cells, human hematopoietic stem and progenitor cells, neural progenitors, and tumor-derived cancer stem-like cells.

Embryonic stem cells (ESCs)

Embryonic stem cells (ESCs) are cell lines derived from the inner cell mass (ICM) of blastocyst stage mammalian embryos. They can grow indefinitely in culture and give rise to cells of all three embryonic germ layers as well as germ cells. For these reasons ES cells hold great promise for regenerative medicine. While many of molecular details of the ESC self-renewal network have emerged, more in depth knowledge will be required to facilitate future ESC-based clinical applications. To date, we have performed multiple functional genomic screens for genes and pathways required for exit from the pluripotent state in mouse ESCs (Schaniel et al., Stem Cells 2009; Beschinger et al., Cell 2013). Recently, we have extended these studies to human "naïve" ESCs with Dr. Hannele Ruohola-Baker's group at University of Washington using a genome-wide CRISPR-Cas9 library and focused retest library.

Human hematopoietic stem and progenitor cells (HSPCs)

Human hematopoietic stem and progenitor cells (HSPCs) can be routinely isolated as mixed CD34+ progenitor pools from peripheral human blood or bone marrow in sufficient quantities for ex vivo lineage studies and experimentation. Through a collaborative effort with Dr. Beverly Torok-Storb (Clinical Research Division) we have characterized factors that promote progenitor outgrowth (Chen et al., Genes and Dev. 2013). We are currently focused on understanding stem cell commitment to erythroid and megakaryotic lineages, which give rise to red blood cells and platelets. This includes a genome-wide CRISPR-Cas9 screen for genes affecting erythropoiesis. One clinical goal is to direct lineage choice to help treat post-HSC transplantation cytopenias, a goal we pursue in collaboration with Dr. Beverly Torok-Storb (Clinical Research Division, Fred Hutch).

Human neural progenitors (hNPCs)

Human neural progenitors (hNPCs) can now be isolated and expanded in laboratory conditions resembling their in vivo niches in the brain and during development (e.g., the perivascular niche). We have recently performed genome-wide RNAi and CRISPR-Cas9 screens for growth limiting and growth promoting genes in two hNPC isolates (Toledo et al., Cell Reports 2015). We are currently characterizing hits from these screens that regulate quiescent-like state, which hNPCs adopt each cell cycle that limit their expansion. Many of these genes are candidate tumor suppressors found mutated across a large number of cancers.

Glioblastoma multiforme stem-like cells (GSCs)

We routinely isolate and culture tumor-initiating glioblastoma multiforme stem-like cells (GSCs) from brain tumor samples in the same defined conditions that permit NPC outgrowth in the lab. Under these conditions, GSCs remarkably retain the development potential and specific genetic alterations found in the patient’s tumor. We use GSC isolates in conjunction with our functional genomic tools to identify, characterize, and develop novel candidate therapeutic targets for Glioblastoma and other deadly cancers [described further below] (Ding et al., Cancer Discovery 2013; Hubert et al. Genes and Dev. 2013; Toledo et al., Dev. Cell 2014; Toledo et al., Cell Reports 2015).