NEW YORK – Five US research groups are employing gene editing, protein depletion, and epigenomic silencing technologies in a pilot study to characterize the functions of approximately 1,000 genes essential for processes such as cell survival, development, and lineage determination, with the goal to eventually characterize the function of every human gene.
The teams, participating in Phase 1 of the National Human Genome Research Institute (NHGRI)-funded Molecular Phenotypes of Null Alleles in Cells (MorPhiC) initiative, aim to determine the best strategies for studying the phenotypes of null alleles, or versions of genes that do not make functional proteins.
Winning strategies will be used to scale up the program in later phases.
Paul Robson and Bill Skarnes, researchers at the Jackson Laboratory, are comparing three approaches for creating null alleles in extra-embryonic or neuroectodermal human cell lineages, which are related to placenta and brain development.
"Our approach is to first look at the systems that may differ between human and mouse," Robson said. "Our two lineages were chosen because they are the most divergent between species."
Robson said that the brain lineage was also attractive for understanding the genetic bases of neurodevelopmental disorders. In this system, the Jackson Lab researchers plan to knock out 125 transcription factors at different developmental time points, beginning with the earliest.
"What we're trying to capture is the primary molecular consequences of loss of a gene," Robson said.
The three approaches the Jackson Lab team will evaluate consist of using CRISPR-Cas9 systems to alter every reading frame within their genes of interest by inserting a single base to create a stop codon, knocking out essential exons, and knocking out whole genes.
The effects of gene knockouts will be evaluated both in cultured cells and in three-dimensional organoids.
Throughout their work on this project, the Jackson Lab team will also continue to develop gene targeting methods and test protein depletion via degron systems, which induce the degradation of a protein.
The auxin-inducible degron provides a way to deplete proteins in vivo to study their function. The addition of the plant hormone auxin conditionally induces the degradation of target proteins by the proteasome, enabling researchers to observe the effects of that loss in a living system.
The team will initially test those strategies in 24 genes before settling on one long-term approach. By the end of the grant's five-year funding period, they aim to have determined the molecular phenotype of 250 genes.
Similar to the Jackson Lab project and also using degron systems, Mazhar Adli of the Northwestern University Feinberg School of Medicine aims to characterize 250 essential human genes in induced pluripotent stem cells (iPSCs), with more extensive characterization to follow in cardiomyocytes.
The genes Adli's lab will study are essential in that cells require them for survival. Because cells die without them, characterizing their function has proven difficult.
"We cannot really study these genes with typical knockout or knockdown strategies," Adli said.
"Most of [the genes] are transcription factors or chromatin regulators," he said, "and we're going to rapidly deplete these genes and study how they affect gene expression and chromatin accessibility."
Changes will be tracked through allele-specific barcodes and ATAC-seq at defined time points after protein depletion.
"If you have a unique barcode per cell per allele," Adli said, "then we can combine many, many cells and cell lines together. This will allow us to do all sorts of screening projects."
Adli, for instance, wants to use this barcoded cell system to understand which genes control a cell's "stemness," or its ability to enter or maintain a stem cell state.
This system will also be useful in conducting synthetic lethal screens. Identifying new tumor suppressors and synthetic lethal drug targets in cancer is a key research focus of Adli's lab.
Luke Gilbert, a researcher at the University of California, San Francisco, plans to study cellular lineage commitment in iPSC-derived three-dimensional culture models called organoids. In contrast to the other teams, Gilbert and his colleagues will silence their target genes using the CRISPRoff epigenomic editor, a technology that Gilbert and Jonathan Weissman of the Massachusetts Institute of Technology pioneered last year in which a single dead Cas9 fusion protein is used to transiently methylate DNA and trigger repressive histone modifications.
This also avoids the need to characterize gene function in clonal cell lines, which is often prone to certain artifacts.
Furthermore, Gilbert explained that because most of the body's cells are genetically identical, "we don't need to manipulate the genome to characterize gene function. Rather, we just need robust ways of turning genes off the way our bodies naturally do."
"We're going to perturb key genes involved in lineage commitment like transcription factors and epigenetic regulators, and then use multiomics approaches to characterize how those genetic perturbations drive cell fate decisions," Gilbert said.
After analyzing the loss of target gene function at various time points, Gilbert and his team will "zoom in" on select candidates in an effort to characterize cellular spatial decisions and cell-cell interactions.
Danwei Huangfu, an investigator at the Sloan Kettering Institute, which is part of Memorial Sloan Kettering Cancer Center, is curating roughly 100 iPSC and ESC lines from diverse ancestral populations, focusing mainly on genes that affect neurodevelopmental and metabolic disorders, such as autism and diabetes.
Huangfu's group plans to knock out mainly genes essential in cell fate decision, as well as some disease-related genes, largely pertaining to diabetes, which is a particular interest of her group.
Knockdowns will be done by CRISPR-Cas9, in three distinct multicellular systems: a micropattern-based gastruloid model for early tri-germ-layer differentiation, a neuro-glial tri-culture system, and a three-dimensional pancreatic islet-like organoid culture for studies into diabetes.
Additionally, Huangfu and her colleagues will carry out phenotyping assays in primary human islet cells to test the generalizability of their approach beyond stem cell systems.
Huangfu's group will characterize phenotypes via single-cell RNA-seq, an approach that she describes as straightforward, despite not being simple.
"We think single-cell RNA-seq is a method that's generalizable," she said. "Obviously, it could be very nice to incorporate additional methods because the phenotype of a cell is not just reflected in the transcriptome, but I think the transcriptome is a good way to start."
By analyzing gene function beginning with those that are essential for cellular development, Huangfu hopes to eventually move beyond single-gene functional characterization toward evaluating gene groups as functional units.
Ideally, she said, "you'll be able to say, 'Hey, this is a group of genes that tend to function in similar ways, and these are the [transcriptome] modules that they impact.'"
Finally, Stephan Schürer, a researcher at the University of Miami School of Medicine, is in the process of establishing the MorPhiC Data Resource and Administrative Coordinating Center (DRACC), tasked with supporting the MorPhiC research consortium by analyzing, annotating, and disseminating MorPhiC data, integrating external data, and serving as an administrative and coordination center.
As part of that mission, the DRACC will also create a MorPhiC web portal that will provide global access to the consortium's data.
"[It's] very exciting to me that NIH is thinking about these things," Adli said, "because our understanding of genes is extremely biased. And I think with this project, hopefully that bias is going to narrow down a little bit."