In a remarkable departure from the norm, creating a comprehensive atlas of all cell types within an organism typically requires extensive multinational collaboration and hefty budgets. However, there are exceptions. According to a recent report on the Nature website, three researchers led by Chengxiang Qiu and Beth K. Martin from the Department of Genome Sciences at the University of Washington in Seattle, along with Ian C. Welsh from the Jackson Laboratory in the United States, accomplished this feat in just one year and at a remarkably low cost.
The trio managed to create a time-lapse atlas of the developing mouse embryo over 10 days, spending only about $370,000. This achievement marks the largest mouse embryo atlas to date, offering insights into how stem cells differentiate into specific cell types, how organs develop, and how an organism changes after birth.
The process of mouse embryo development from day eight (top left) to birth (bottom right). Image source: C. Qiu et al./Nature
Division of labor in embryo "tree-mapping"
The aforementioned study was published in the journal Nature in February of this year. Chengxiang Qiu, Beth K. Martin, and Ian C. Welsh are the co-first authors of the paper, leading the research and completing the primary work.
Chengxiang Qiu is also the corresponding author of the paper, and Jay Shendure, a geneticist at the University of Washington in Seattle, who is the mentor of Qiu's laboratory, is also a corresponding author. Other authors of the paper contributed to relevant assistance work.
Qiu and Martin, members of Jay Shendure's laboratory, utilized a novel single-cell transcriptomics technique known as sci-RNA-seq3. This technology, developed in Shendure's lab renowned for molecular biology innovations, enables high-throughput, multi-sample processing at a lower budget, facilitating the examination of mRNA assembly within individual cells.
To demonstrate the utility of this technique, Qiu and Martin embarked on profiling the single-cell transcriptome of embryonic mice at approximately 19 days gestation. Given the challenge of maintaining intact cells throughout the study, they opted to disaggregate entire mouse embryos, isolating their nuclei. These nuclei were then distributed into various culture dishes and labeled with distinct molecular tags on their mRNA.
Subsequently, the nuclei were merged, re-separated, and subjected to additional labeling in each dish, repeating this process iteratively. Ultimately, each nucleus acquired a unique set of tags—molecular barcodes—that researchers could leverage to decipher the transcriptome of individual cells.
Using this approach, mRNA from these cells was sequenced, and a tree-like model was constructed to simulate how one cell type transitions into another. Initially, Qiu and Martin collected embryos every 24 hours over a five-day period. However, the substantial transcriptional variations between different time points made it challenging to track how stem cells differentiate into specific cell types, likened by Shendure to missing frames in a video, rendering it more akin to stop-motion animation than continuous smooth progression.
To address this issue, Qiu and Martin collaborated with Welsh from the Jackson Laboratory, a biomedical research and mouse breeding institution based in Bar Harbor, Maine, renowned for its comprehensive mouse facilities. Welsh, with his expertise in mouse research, collected mouse embryos every 2 to 6 hours during a 10-day gestation period, meticulously gathering 83 embryos, promptly freezing them, and dispatching them to Seattle.
In Seattle, Martin collected single-cell transcriptomes, which Qiu subsequently mapped onto the tree-like structure model, illustrating when and how 190 cell types (such as hepatocytes or bone marrow cells) originated from the embryo.
To further enrich this "tree," researchers integrated data from multiple sources, including their own starting from day 8 of gestation, existing work from the Shendure team, and other studies mapping transcriptomic landscapes of these embryos and younger ones. This integration added 110,000 cells.
These data formed the "root" of the tree-like model, enabling researchers to trace how early stem cells differentiate into specific types observed in older embryos.
The data-driven tree encompasses cell types throughout the entire mouse development process from fertilized egg to pup. Image source: referenced paper.
Fortunate discoveries, future prospects
This discovery is of significant importance.
The resulting atlas contains transcriptomes of mice at 45 time points, available for developmental biologists to delve deeper. It comprises 12.4 million cells, the largest mouse embryo atlas to date.
According to Bertie G?ttgens, a stem cell biologist at the University of Cambridge, this research is "impressive on many levels, both in terms of its scale of achievement and in how it fulfills its research goals."
Sarah Teichmann, a cell geneticist at the Wellcome Sanger Institute and co-founder of the Human Cell Atlas, believes this achievement will advance various studies, including the ability to compare mouse and human development.
Yonatan Stelzer, an epigeneticist at the Weizmann Institute of Science in Israel, notes that this research will be helpful in future efforts to map the cells of individual organs or tissues.
Simultaneously, this discovery is serendipitous.
Researchers noted two phenomena. First, the most significant changes in the transcriptome occurred within the first hour after mouse birth. Shendure referred to it as "the most stressful moment of life." Some changes were expected, such as lung and fat cells altering activity to adapt to the external uterine environment.
Pure luck led to another discovery. Welsh usually used cesarean section to deliver mice. However, one day, upon returning from lunch, he discovered an unexpected litter of newborn pups. Upon analyzing these mice, significant differences were found in their transcriptomes compared to cesarean-born mice. Researchers suggest that these differences may explain the impact of birth mode on health outcomes.
However, there are areas for improvement in this study.
Teichmann points out that there is still work to be done on the mouse atlas. For example, some time points have more complete transcriptomes than others, and researchers have not yet separated mice by sex to observe these differences.
Reportedly, Shendure's team's next plan is to create single-cell atlases of juvenile and adult mice from conception to death.
Stelzer states that future embryonic research will delve further, not only studying how cells develop over time but also tracking them in 3D space to understand how they divide and move to form a complete mouse.
Stelzer also adds that future research can clarify some questions, such as how two cells with similar transcriptomes determine whether they develop into the left or right eye.
"However, we are still far from solving the entire embryo puzzle," Stelzer says.
References:
