
(© adimas - stock.adobe.com)
Scientists say the origins of the ‘Hippo pathway’ date back to single-celled organisms
In A Nutshell
- The Hippo signaling pathway that helps control tissue size in animals and malfunctions in cancers already existed before animals evolved, operating in single-celled organisms.
- Researchers used breakthrough CRISPR techniques to delete genes in choanoflagellates, revealing these organisms use the Hippo pathway to control colony size by regulating extracellular matrix production rather than cell division.
- When scientists removed the warts gene, choanoflagellate colonies doubled in size, growing from 11 to 21 cells on average, with some reaching 60 cells by overproducing the biological glue that holds them together.
- The discovery shows the Hippo pathway was likely repurposed during evolution: managing cellular glue production in early ancestors, then adapted by animals to control tissue growth by limiting cell division.
Genetic instructions controlling how big your organs grow date back to long before animals even existed, according to research conducted on microscopic organisms considered our closest single-celled relatives.
Scientists studying choanoflagellates discovered these organisms use the very same molecular machinery to control colony size that animals later adapted to regulate organ growth. Dubbed the Hippo signaling pathway, the system was already operational before the first animals evolved. Today it helps control tissue size in animals and malfunctions in many cancers.
Thibaut Brunet, an evolutionary biologist at Institut Pasteur in Paris who led the research, compared the discovery to finding out that a smartphone’s operating system was originally written for a completely different device. The findings appear in Cell Reports.
How The Hippo Pathway Holds Clues to Animal Evolution
Choanoflagellates are single-celled organisms that occasionally form multicellular colonies called rosettes. These colonies resemble the earliest developmental stage of animal embryos, making choanoflagellates a living window into how our ancestors made the leap from single cells to multicellular life.
The research team used CRISPR gene editing to delete genes in the Hippo pathway, including one called warts. Without this gene, the choanoflagellate colonies grew to twice their normal size, with rosettes averaging about 21 cells instead of the usual 11. Some mutant colonies ballooned to 60 cells, more than twice the maximum size ever seen in normal colonies.
The mechanism behind this growth spurt turned out to be surprising. In animals, the Hippo pathway often controls tissue size by regulating cell division. But in choanoflagellates, the pathway doesn’t control how fast cells divide. Instead, it regulates the production of extracellular matrix, the molecular scaffolding that holds colonies together.
How Ancient Organisms Stuck Together
When the researchers examined the giant colonies under microscopes, they found dramatically more extracellular matrix material in the mutants compared to normal colonies. The matrix formed elaborate branching structures, creating more attachment points for cells to stick together.
RNA sequencing revealed that disabling the Hippo pathway activated genes for matrix production, including one called couscous that helps build the glycosylated proteins forming the colony’s core, along with fibrillar collagen and C-type lectins. Without Hippo pathway regulation, cells overproduced the biological glue holding them together, allowing colonies to grow larger before splitting apart.
This discovery suggests the Hippo pathway may have had an ancestral role in managing extracellular matrix in early organisms. Animals appear to have repurposed this ancient system for a related but different job: controlling tissue size by regulating cell division.
From Single Cells to Cancer Research
The connection between choanoflagellates and animals isn’t just an academic curiosity. The Hippo pathway malfunctions in many human cancers, allowing tumors to grow unchecked. Understanding how this system worked in our single-celled ancestors could offer insights into what goes wrong in disease.
The pathway’s core components (Hippo, Warts, and Yorkie) exist in choanoflagellates, in animals, and in filastereans, another close relative of animals. The genes encoding these proteins have been faithfully copied and passed down through hundreds of millions of years of evolution, accumulating modifications but never disappearing entirely. These are true homologs inherited from a common ancestor that lived before animals appeared.
The research was made possible by a new gene-editing technique the team developed specifically for choanoflagellates. Previous methods for deleting genes in these organisms were laborious and unpredictable, with success rates as low as 0.3 percent. The new approach boosted efficiency to between 40 and 100 percent among antibiotic-resistant clones, making such experiments practical for the first time.
The technique inserts an antibiotic resistance gene into the target location, allowing researchers to use antibiotics to select only the cells where gene editing succeeded. This eliminated the need to isolate and test hundreds of individual cells to find the rare successful edits. Using this method, the team successfully knocked out five of the six genes they targeted.
The findings add nuance to how the Hippo pathway evolved. The work in choanoflagellates suggests size control through extracellular matrix management predates animals, complicating earlier ideas that the pathway’s growth-control function arose only within animals.
In filastereans, the Hippo pathway controls yet another function: cell shape and contractility rather than proliferation or matrix production. This patchwork of functions across different organisms paints a picture of a versatile genetic toolkit that evolution has repeatedly adapted for new purposes.
When your organs stop growing at the right size, or when that regulation fails and cancer develops, you’re experiencing the consequences of genetic software written before the first animals ever existed.
Paper Summary
Methodology
The researchers developed a new CRISPR-Cas9 gene knockout method for the choanoflagellate Salpingoeca rosetta. They designed repair templates containing a puromycin antibiotic resistance cassette flanked by sequences matching the target gene location. After introducing CRISPR components and repair templates into cells via nucleofection (a type of electroporation), they selected successfully edited cells using puromycin. The team tested two approaches: inserting the resistance cassette to interrupt the gene, or deleting the entire gene and replacing it with the cassette. They targeted six genes total: rosetteless, couscous, and jumble (known developmental regulators) plus warts, yorkie, and hippo (Hippo pathway components). Edited cells were isolated by limiting dilution and genotyped by PCR and Sanger sequencing. Phenotypes were assessed through microscopy after inducing rosette formation with Algoriphagus bacteria, and gene expression was analyzed using RNA sequencing with at least three biological replicates per condition.
Results
The new knockout method achieved 40-100% efficiency among antibiotic-resistant clones, a dramatic improvement over previous methods (0.3-16.5%). The team successfully generated knockouts for five of six targeted genes. Warts kinase knockout resulted in rosettes containing twice as many cells as wild-type (21 versus 11 cells on average), with some mutant colonies reaching 60 cells. Confocal microscopy revealed these giant rosettes had more abundant and extensively branched extracellular matrix. RNA sequencing identified 22 genes consistently upregulated in both warts insertion and deletion mutants, including seven extracellular matrix components (fibrillar collagen, three C-type lectins, couscous) and four glycosyltransferases. Yorkie knockout had no effect on rosette size or gene expression, though the authors noted alternative start codons might have produced partially functional truncated proteins. Warts-regulated genes were downregulated in yorkie deletion mutants, confirming the antagonistic relationship between Warts and Yorkie seen in animals.
Limitations
The insertion method may not fully inactivate genes encoding cytoplasmic or nuclear proteins if alternative translation start sites exist downstream of the insertion point. This appeared to occur with hippo and yorkie insertions, where transcripts suggested production of truncated but potentially functional proteins. The deletion method addresses this limitation but proved more technically challenging. The team could not generate hippo deletion mutants despite three attempts, suggesting the gene may be essential. RNA sequencing experiments lacked time-course data to track dynamic gene expression changes during rosette development. The study focused on one choanoflagellate species, and findings may not generalize to other species with different multicellular behaviors. The episomal maintenance of repair templates in template-only controls (without CRISPR components) was unexpected and its mechanism remains unclear. The method currently allows knockout of only one gene at a time due to the single selectable marker available.
Funding and Disclosures
The research was supported by Institut Pasteur (G5 package), the European Research Council Starting Grant EvoMorphoCell (grant agreement ID: 101040745), the Bert L. & N. Kuggie Vallee Foundation, ANR-23-CE13-0031, and the CNRS (UMR 3691). Mylan Ansel was supported by a Contrat Doctoral Spécifique Normalien from the École Normale Supérieure de Lyon and is a student in the FIRE PhD program funded by the Bettencourt Schueller Foundation and the EURIP graduate program (ANR-17-EURE-0012). The Biomics Platform at Institut Pasteur is supported by France Génomique (ANR-10-INBS-09) and IBISA. The authors declared no competing interests.
Publication Details
Combredet, C., Ansel, M., and Brunet, T. (2025). “A selection-based knockout approach for a choanoflagellate reveals regulation of multicellular development by Hippo signaling,” was published October 28, 2025 in Cell Reports 44, 116345. doi:10.1016/j.celrep.2025.116345







