Cut, paste, grow: Study reveals surprising totipotency in early worm embryos

The embryo was no bigger than a grain of sand. Under the microscope, it had divided just twice; four cells arranged in a tidy cluster, each carrying instructions for a different piece of the animal it was supposed to become. The two smaller cells, the micromeres, would build the brain, throat, and most of the outer skin. The two larger ones, the macromeres, were slated for gut and stem cells and little else. That was the plan according to millions of years of evolution.

Then, Amber Rock took one of those macromeres and cut it free.

Rock, a PhD candidate in the Department of Organismic and Evolutionary Biology at Harvard, and her advisor, Professor Mansi Srivastava, Curator of Invertebrate Zoology in the Museum of Comparative Zoology, were astonished.

“My assumption was that if I deleted a single blastomere, the entire embryo couldn’t develop because each cell contains specific information needed to build an entire animal,” said Rock. Yet, at both the 2- and 4-cell stage, the cells developed into a full animal on their own.

The discovery, reported in Current Biology, represents the first documented case in which a cell produced by an asymmetric, fate-specifying division retains full totipotency — the ability to generate an entire organism from scratch. This is particularly surprising because these worms, Hofstenia miamia, follow an "invariant" cleavage program, where the fate of each cell is thought to be fixed almost immediately.

H. miamia, or three-banded panther worms, are tiny acoel worms that grow to about 500 micrometers and can regenerate their entire bodies as adults by relying on extremely flexible stem cells called neoblasts. Remove the tail and it grows another. Cut the worm into pieces and each fragment becomes a complete worm within weeks.

Srivastava, who has studied these worms for decades, and Rock set out to find if this cellular plasticity is unique to adults or if embryonic cells could also show plasticity. After early trials with 2- and 4-cell embryos, Rock determined the new cells were not from the same source as neoblast, but the mechanism was still unknown.

The study used transgenic panther worms, Tuba::Kaede, developed by Srivastava and former postdoctoral fellow, Lorenzo Ricci, in a 2021 Developmental Cell study. Tuba::Kaede carry cells that fluoresce green and can be photoconverted to red with a laser, allowing lineage tracing across generations. Rock performed blastomere isolations by manually dissecting embryos with an eyelash tool at the 2-, 4-, and 8-cell stage. She then swapped in transgenic cells and recombined the embryos into new structures. With transplanted cells glowing green, they could track both native cells and the introduced cells reestablished in nontraditional positions.

At this stage, cells normally produce specific tissues – neurons or muscle – and  occupy fixed locations. But when rearranged, they were able to make almost every single cell type in the adult worm. “It’s amazing that you can completely tear apart the embryo, put it back together, and somehow it develops,” said Rock.

Srivastava and former student, Julian Kimura (PhD ‘22) showed in a 2022 Cell study that beginning at the 4-cell stage, cells undergo “fate assignment,” meaning they continue development by producing only their designated tissues. However, by isolating and reconfiguring the cells, Rock and Srivastava have shown that these same cells retain totipotency and can produce an adult worm. Survival rates for these embryos were comparable to undisturbed embryos. The worms not only grew to healthy adults, they reproduced and regenerated normally when cut.

The plasticity of the Hofstenia cells after reaggregation surprised Rock. “You can take apart and randomly reaggregate sea urchin embryos, but they will sort themselves out,” she said. “They immediately realize they are out of order and then self-segregate and rearrange. But Hofstenia did not self-organize, it changed fate to build the adult worm.”

One challenge was removing the embryo’s protective shell while keeping them alive. Rock tested methods used in other marine invertebrates before successfully landing on dechorionating – chemically stripping the outer membrane without damaging the embryo. Another challenging was the transcriptome, previously based largely on adult RNA. Srivastava and Rock updated it with the embryonic data, creating a more complete genetic reference.

The molecular basis for this remarkable feat remains unknown. Using nascent RNA labeling, the researchers found that the transition from maternal to zygotic control likely occurs 48 to 72 hours after laying, well after the 4-cell stage. This timing suggests that totipotency in the macromeres is supported by maternal contributions rather than newly activated genes.

“There are so many different strategies of how an animal develops,” Rock said. “One isn’t better than the other, but it was so surprising to see this embryo take a little bit from both sides — where it has stereotyped cleavage, but it also maintains plasticity. It’s interesting to think about how pre-planned early development can go from being rigid to full totipotency.”

Srivastava added, “This finding is a humbling reminder of how much more biology remains to be discovered, and that our understanding of the natural world is limited by what we have studied so far. I had been skeptical of Amber’s proposal to study the Hofstenia embryo in this way, I was convinced the cells would behave in a predictable way. But they didn’t.”

####