Cell division in every eukaryotic organism – from small, unicellular yeast to complex, multicellular orangutans – depends on proper separation of genetic information into the newly divided cells. To ensure each cell receives the correct genetic information, replicated DNA molecules are tightly condensed, each with its own centromere, a specialized region of the chromosomes where mitotic spindles grab onto sister chromatids so that they can be pulled apart when the cell divides. Centromere formation is mediated by the deposition of CENPA, a centromere-specific histone H3 variant. In fungi and mammals, specialized histone chaperone proteins containing scm3 domains recognize regions of centromeric DNA, deliver CENPA, and facilitate CENPA-containing nucleosome assembly.
The presence of proteins with scm3 domains in yeast and humans suggests these proteins have ancient evolutionary origins. This, combined with CENPA's essential role in chromosome segregation, led scientists to suspect that scm3 domain-containing orthologs would be conserved across evolution. However, initial attempts to identify homologs in many animal species were unsuccessful. Without readily identifiable homologs, researchers instead hypothesized that animals could have evolved unique chaperones for centromeric proteins or entirely different centromere assembly processes. Dr. Jeremy Hollis, a recent PhD graduate co-mentored by Drs. Harmit Malik and Melody Campbell, wasn’t so sure.
Many DNA-binding proteins (like those with scm3 domains) are disordered, meaning that they do not assume one rigid conformation. This leaves room for rapid divergence in the DNA and protein sequences for homologs of the same protein over the course of evolution. Based on previous work, scientists think that only one region of scm3 domain-containing proteins assumes a rigid conformation to deliver CENPA to the centromere, allowing for rapid sequence divergence in the rest of the protein. Standard homology detection methods like BLAST searches or 3D structural comparisons often perform poorly for proteins with low homology across species. Hollis suspected that a new approach integrating sequence comparisons and structural predictions would reveal previously undetected scm3 domain-containing proteins in many animal species.
Researchers had not identified proteins with scm3 domains in ray-finned fish such as salmon or cod, so Hollis decided to start his search there. He used an iterative BLAST search approach starting with the sequences of scm3 domain-containing homologs in humans and the western clawed frog to identify homologs in distant species. He was unable to identify homologs in ray-finned fish after these initial searches, but did identify one potential hit in a lobe-finned fish species. When he added this sequence to his BLAST searches, he identified scm3 domain-containing proteins in several species without reported homologs, including ray-finned fish, supporting the notion that at least all vertebrates share a common scm3 domain-containing ancestor.
Sequence similarity alone isn’t enough to definitively prove that the homologs they identified have the same function as previously identified scm3 domain proteins. To confirm this, Jason Stonick from Cecilia Moens’ lab functionally interrogated the zebrafish homolog, a type of ray-finned fish. They expressed GFP-tagged CENPA and mCherry-tagged homolog in zebrafish zygotes and used microscopy to determine if the proteins colocalized in the nucleus. They found that the two proteins colocalized only during interphase of the cell cycle, an expression pattern that mirrors that of human scm3. Next, they hypothesized that the new scm3 domain-containing homolog would be necessary for early zebrafish development given its fundamental role in CENPA deposition. They used CRISPR to knock out the homolog in zebrafish embryos and saw that knockout embryos were misshapen and dying 24 hours after fertilization. These data confirm that the protein identified in their homology search was indeed a true homolog of the human scm3 domain-containing protein.
Proteins with scm3 domains were also entirely missing from nematodes like C. elegans prior to this study. C. elegans has long been used to study fundamental aspects of the cell cycle, and most essential centromere-binding proteins, including CENPA, have been identified in this species. Given how useful C. elegans has been for understanding the cell cycle, the absence of an scm3 domain-containing homolog in this species was surprising. Hollis again used an iterative BLAST search to identify an scm3 domain-containing protein in nematodes. After several searches using putative homologs from distantly related nematode species, the team found a homolog in C. elegans called nefr-1. This result was initially puzzling because nefr-1 was previously thought to be important in neuronal function, but when Hollis used AlphaFold to model interactions with nefr-1, CENPA, and another histone, he saw that the architecture closely resembled that of the human homolog. Nefr-1 also contained an scm3 domain, so they teamed up with Irini Topalidou in the Lehrbach lab to create a nefr-1 knockout in C. elegans. They found that nefr-1 knockout animals were nonviable, indicating that nefr-1 is essential. They next examined colocalization between NEFR-1 and CENPA and found that the two proteins colocalized in the nucleus during cell division. Further biochemical analyses showed that there is a direct interaction between NEFR-1 and CENPA, indicating that NEFR-1 is indeed the long-missing scm3 domain-containing ortholog in C. elegans.
A third lineage in which the status of the CENPA chaperone was ambiguous was in insects, including Drosophila, where a CAL1 chaperone had been identified but its relationship with other animal chaperones left unclear whether it represented a divergent homolog or a reinvention. Buoyed by the dual successes in fishes and nematodes, Jeremy traced the evolution of the CAL1 CENPA chaperone in flies, and showed that it indeed is a divergent homolog of the chaperone found in most animals. With this work, they filled the evolutionary gap of missing proteins with scm3 domains in most animal species, solving a longstanding mystery in evolutionary biology. These results are certainly exciting, but for Hollis, the best part was collaborating with his colleagues in the Basic Sciences Division. “This was a collaboration of a bunch of different labs in Basic Sciences, and it was so fun to be able to walk down the hall and chat with my friends and also be working on this project together,” he explained. Hollis hopes that similar strategies can be used to find several of these “missing genes” and continue to fill longstanding evolutionary gaps.
This work was supported by funding from the National Science Foundation, Fred Hutchinson Cancer Center, a Pew Charitable Trusts Biomedical Scholars award, the National Institutes of Health, and the Howard Hughes Medical Institute.
Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Members Drs. Cecilia Moens, Nicolas Lehrbach, Melody Campbell, and Harmit Malik contributed to this research.
Hollis JA, Stonick JA, Topalidou I, Young JM, Moens CB, Lehrbach NJ, Campbell MG, Malik HS. 2026. Remote homology and functional genetics unmask deeply preserved Scm3/HJURP orthologs in metazoans. Sci Adv. 2026 Jul 3;12(27):eaeg5356. doi: 10.1126/sciadv.aeg5356.
Kelsey Woodruff is a PhD candidate in the Termini Lab at Fred Hutch Cancer Center. She studies how acute myeloid leukemia cells remodel the sugars on their membranes to reprogram cancer cell signaling. Originally from Indiana, she holds a bachelor's degree in Biochemistry from Ball State University. Outside of lab, you can find her crocheting and enjoying the Seattle summers.