NEW YORK – A pair of new studies are underlining the potential of applying single-cell transcriptomic strategies to track cellular and molecular phenotypes in developing model organism embryos containing specific genetic alterations.
For the first of the studies, published in Nature on Wednesday, researchers from the University of Lübeck and Kiel University, the Max Planck Institute for Molecular Genetics, the University of Washington, and elsewhere used single-cell RNA sequencing to profile nuclei from more than 1.6 million individual cells from 101 mouse embryos grown up to embryonic day 13.5. The embryos represented four wild-type genotypes, as well as 22 mutant genotypes linked to developmental phenotypes ranging from mild to severe.
"In 2011, the International Mouse Phenotyping Consortium set out to drive towards the 'functionalization' of every protein-coding gene in the mouse, by generating thousands of [knock out] mouse lines," the authors wrote. "In principle, the whole-embryo scRNA-seq phenotyping approach presented here could be extended to all Mendelian genes or even to all 20,000 mouse gene [knock outs]."
With the help of several analytical algorithms, the team harnessed single-cell transcriptomic features from across the developing mouse embryos to track cell type-specific gene expression and developmental trajectories across dozens of cell types in mutant mouse genotypes, analyzing them alongside data on more than 1.3 million cells from 61 wild-type mouse embryos grown to embryonic day 9.5 to 13.5 for a mouse organogenesis cell atlas they previously developed.
"In our current manuscript we apply a completely unbiased approach by profiling entire embryos using single-cell RNA sequencing [of] over 100 embryos of 25 different genetic backgrounds," co-senior and co-corresponding author Malte Spielmann, director of the University of Lübeck and Kiel University's Institute of Human Genetics, explained in an email.
He noted that conventional phenotyping methods such as microscopy and histology have not kept pace with new gene editing or gene knockout strategies that make it possible to rapidly and accurately generate genetic variation in mice.
In contrast, the team's whole-embryo phenotyping approach does make it possible to measure the diverse, cross-organ consequences of removing or altering individual mouse genes — pleiotropic effects that might have been missed with more plodding approaches in the past.
Indeed, Spielmann and his colleagues reported that the "diverse mutants analyzed yielded a variety of results that speak to the utility of whole-embryo scRNA-seq for phenotyping."
While the trajectories for dozens of cell types were upended in a subset of the mutant embryos profiled, for example, the team's analyses suggested that other mutant embryos were marked by more subtle shifts that involved a limited subgroup of cell types.
Likewise, the investigators noted that the between-embryo differences they detected were not limited to the mutant mouse genotypes. On the contrary, their results highlighted variation between embryos from distinct wild-type genotypes, while spelling out phenotypic features linked to loss-of-function alterations, gain-of-function changes, deletions, and other variants.
"Overall," the researchers reported, "our findings show how single-cell profiling of whole embryos can enable the systematic molecular and cellular phenotypic characterization of mouse mutants with unprecedented breadth and resolution."
More broadly, Spielmann suggested, a similar approach may be deployed in the future to tease out subtler cellular changes that have not yet been found, while flagging possible treatment targets.
In a related Nature study, a team at UW, the Allen Discovery Center for Cell Lineage Tracing, and the Fred Hutchinson Cancer Center used an embryo barcoding, multiplexing, and single-nucleus RNA-seq strategy to assess developing zebrafish models, identifying 99 cell types and 156 cell subtypes across 33 main tissue types.
"Together, our scalable approach is flexible, comprehensive, cost-effective, and more uniform than conventional phenotyping strategies," co-corresponding authors Cole Trapnell and David Kimelman, both genome sciences researchers at UW, and their colleagues wrote. "We anticipate that this new experimental and analytical workflow will enable rapid, high-resolution phenotyping of whole animals to better understand the genetic dependencies of cell types in a developing organism."
That team's "zebrafish single-cell atlas of perturbed embryos" effort encompassed some 3.2 million cells from 1,812 zebrafish embryos sampled at 19 embryonic or early larval development time points and included both wild-type embryos and embryos containing 23 distinct genetic alterations.
"The high degree of replication in our study (eight or more embryos per condition) enables us to estimate the variance in cell type abundance organisms-wide and to detect perturbation-dependent deviance in cell type composition relative to wild-type embryos," the authors explained.
By digging into the single-cell transcriptomic profiles, differentially abundant cell type data, differentially expressed gene sets, and other features found in the embryos, the team got a glimpse at the consequences of the genetic perturbations considered, while unearthing new clues about related developmental process.
"As our field accumulates a catalog of whole-embryo, single-cell transcriptional phenotypes," authors of that study suggested, "the potential for discovering mechanisms through which the vertebrate genome controls development using computational and statistical tools will only grow."