Teratomas Reported to Be More Useful Model than Organoids

Bioengineers at the University of California San Diego report that teratomas, tumors made up of different tissues such as bone, brain, hair and muscle, are a useful model to study human development—from the early embryonic stage onward. Teratomas form when a mass of stem cells differentiates uncontrollably, forming all types of tissues found in the body; they are considered to be an undesired byproduct of stem cell research.

"We've been fascinated with the teratoma for quite a while," said Prashant Mali, senior author on a paper published in Cell today. "Not only is the teratoma an intriguing tumor to look at in terms of the diversity of cell types, but it also has regions of organized tissue-like structures. This prompted us to explore its utility in both cell science and cell engineering contexts."

"There's no other model like it. In just one tumor, you can study all of these different lineages, all of these different organs, at the same time," said Daniella McDonald, co-first author of the study. "Plus, it's a vascularized model, it has a three-dimensional structure and it's human-specific tissue, making it the ideal model for recreating the context in which human development happens."

The team used teratomas grown from human stem cells injected under the skin of immunodeficient mice. Using single-cell RNA sequencing, the team was able to map 20 cell types, or "human lineages" (brain, gut, muscle, skin, etc.) that were consistently present in all the teratomas they analyzed. They then used CRISPR-Cas9 to screen and knock out 24 genes known to regulate development. They found multiple genes that play roles in the development of multiple lineages.

"What's remarkable about this study is that we could use the teratoma to discover things in a much faster way. We can study all of these genes on all of these human lineages in a single experiment," said co-first author Yan Wu. "With other models, like organoids, that separately model one lineage at a time, we would have had to run many different experiments to come up with the same results as we did here."