The study, published in PLOS Genetics by Christopher J. Johnson, Joshua Kavaler, Timethea Slade, Kayla Wilson, Christina D. Cota, and Alberto Stolfi, focuses on the tunicate Ciona, one of the most tractable experimental models for studying chordate development. Tunicates occupy a pivotal position in the tree of life: they are the sister group to vertebrates within the chordate phylum, meaning they share a more recent common ancestor with us than any other invertebrate lineage does. Yet unlike vertebrates, tunicates evolved a biphasic lifecycle, alternating between a motile larval stage and a sedentary adult stage. The papillae are the key organ governing the transition between these two phases, making them a fascinating case study in how evolution repurposes embryonic machinery for new anatomical innovations.

The papillae of most tunicate larvae are arranged as a group of three morphologically identical organs. During embryogenesis, they arise from the anterior neural plate border, a region of the developing embryo nestled between the ventral epidermis and the more dorsal and posterior progenitors of the neural tube. This embryonic address matters. Based on its location and its molecular signature, the anterior neural plate border in tunicates has been evolutionarily linked to the vertebrate placodes, the thickened patches of embryonic ectoderm that give rise to sensory structures such as the inner ear, the nose, and the lens of the eye. Understanding how the papillae are patterned therefore offers a window into how sensory organs were built in the common ancestor of all chordates.

Previous work had established that a signaling pathway called BMP, short for bone morphogenetic protein, is essential for the specification, patterning, and morphogenesis of the papilla region in the Ciona embryo. BMP signaling is one of the great workhorses of developmental biology, used across the animal kingdom to draw boundaries between tissues and to instruct cells about their fates. But knowing that BMP is required is only the beginning of the story. Signaling pathways are blunt instruments; they must be interpreted by transcription factors, the proteins that switch specific genes on or off in specific cells. The downstream mechanisms that translate BMP activity into the actual anatomy of the papillae remained poorly understood, and that is the gap the new study set out to close.

The researchers identified a transcription factor called NK4, the Ciona ortholog of the vertebrate NKX2-3, NKX2-5, and NKX2-6 genes, as a central player in this process. NK4 belongs to the NKX2 family of homeobox transcription factors, a group of DNA-binding proteins with deep evolutionary roots and well-documented roles in organ development across the animal kingdom. In vertebrates, the closest relatives of NK4 are famous for their roles in building the heart and the thyroid, among other structures. The new work shows that in the tunicate embryo, this ancient gene family has been recruited to a very different task: carving a single field of cells into three discrete sensory organs.

The evidence places NK4 squarely downstream of BMP signaling. The team found that NK4 is activated by BMP in the embryonic region that will give rise to the papillae, meaning that BMP does not act alone but instead works through this transcription factor to execute its patterning program. This kind of pathway-to-transcription-factor relay is a recurring theme in development, allowing a broadly active signal to be converted into precise, cell-type-specific gene expression. In this case, the relay allows a diffuse gradient of BMP activity to be transformed into the sharply defined anatomy of three separate papillae.

What does NK4 actually do once it is switched on? The study provides a compelling answer: it acts as a transcriptional repressor, and its principal target appears to be a regulatory gene called Foxg. In normal embryos, Foxg is expressed in exactly three cell clusters, and those clusters are the ones that give rise to the three papillae. NK4’s job is to restrict Foxg expression to those three spots. The logic is elegant: rather than activating the papilla program directly in three places, the embryo uses NK4 to repress the program everywhere except where other influences allow it to persist. Repression, in developmental biology, is often as creative as activation, and this system is a textbook example.

The functional tests drive the point home with striking clarity. When the researchers disrupted NK4 function, the boundary control was lost and the embryo formed a single large papilla instead of three separate ones. The papilla program, freed from repression, spread across the entire region and merged into one organ. The opposite experiment produced the mirror-image result: when NK4 was overexpressed, its repressive power overwhelmed the system, Foxg was shut down everywhere, and the papillae were eliminated entirely. Together, these gain-of-function and loss-of-function experiments show that the level of NK4 activity is precisely calibrated, and that the tripartite anatomy of the organ depends on keeping that calibration within narrow limits.

How does the embryo achieve that calibration? The study points to two opposing influences that fine-tune NK4’s reach. On one side, the expression of NK4 is restricted dorsally by Chordin, a well-known antagonist of BMP signaling. Chordin binds BMP molecules and prevents them from activating their receptors, so in regions where Chordin is present, the BMP-to-NK4 relay is interrupted. On the other side, the researchers found that Msx, a transcription factor expressed on the ventral side of the embryo, can alleviate the repressive effect of NK4. In other words, where Msx is present, NK4’s grip on Foxg is loosened, allowing the papilla program to proceed. The interplay between Chordin’s dorsal restriction and Msx’s ventral counteraction potentially explains the characteristic tripartite patterning of the papillae: three narrow zones where the balance of these forces tips in favor of Foxg expression, surrounded by territory where NK4 keeps the program silenced.

Beyond its immediate findings, the study raises broader questions about the evolution of sensory organs in chordates. The anterior neural plate border that produces the papillae is thought to correspond to the territory that, in vertebrates, generates placodes and the neural crest, two innovations that underpin the elaborate sensory and cranial structures of animals with backbones. If a BMP-dependent, NKX2-family-mediated repressive circuit helps carve sensory territories out of this border region in tunicates, similar logic may have operated in the last common ancestor of tunicates and vertebrates, or may have been independently redeployed in each lineage from an ancient toolkit. Either way, the work demonstrates that the difference between one organ and three can hinge on a single repressor and its antagonists, a reminder that the grand architecture of animal bodies is built from molecular switches of exquisite sensitivity. For a larva whose survival depends on landing in the right place at the right time, that precision is everything.

Subject of Research: BMP-dependent NKX2-family transcriptional regulation patterning the tripartite sensory-adhesive papillae of tunicate larvae

Article Title: An NKX2-3/5/6 homolog is required downstream of BMP signaling to pattern the sensory-adhesive organ of a tunicate larva

Article References: Johnson, C. J., Kavaler, J., Slade, T., Wilson, K., Cota, C. D., & Stolfi, A. (2026). An NKX2-3/5/6 homolog is required downstream of BMP signaling to pattern the sensory-adhesive organ of a tunicate larva. PLOS Genetics, 22(10), e1012334. https://doi.org/10.1371/journal.pgen.1012334

Image Credits: AI Generated

DOI: 10.1371/journal.pgen.1012334

Keywords: tunicates, Ciona, BMP signaling, NKX2 transcription factors, NK4, Foxg, Chordin, Msx, papillae, neural plate border, embryonic patterning, chordate evolution