What makes a skin cell a skin cell or a brain cell a brain cell? Every cell in our body shares the same DNA, but their purpose differs based on which parts of their genetic code are accessible to be expressed and at what time. This is strongly impacted by the structure of DNA, which is wrapped around protein complexes called histones and organized in a complex three-dimensional form called chromatin. Modifying the way chromatin is packaged by histones can program (or re-program) cells for a specific function.
Viruses are excellent at re-programming cells; they need this so that the cell becomes a virus making factory instead of whatever it was the cell was originally intended for. To do so, viral proteins frequently interact with host histones to control the cell’s response to infection. Viruses can subdue the cell’s innate defenses, steal histones to promote expression of their own genes, or, in at least one case, change chromatin to manipulate the physical structure of the nucleus in a way that benefits viral egress (read our coverage of this story here).
One virus that manipulates host chromatin in a dramatic fashion is human cytomegalovirus (HCMV), a herpesvirus that has devastating impact on immunocompromised people, including bone marrow transplant patients. During active replication in skin cells, HCMV remodels the normally round nucleus into a bean-shaped one that spins around a cytoplasmic structure called the viral-induced assembly compartment (vIAC). The vIAC is important for more than just giving cool pictures and videos; it is where viral capsids go to mature before budding through the host cell membrane.
But what kind of chromatin hijacking is necessary for this dramatic nuclear reorganization? That’s a specialty of Dr. Daphne Avgousti, who was faculty in the Human Biology Division for several years before moving to the University of Miami recently. During her lab’s tenure in HB, Dr. Laurel Kelnhofer-Millevolte—a MSTP student now returned to UW Medical School—tackled this question. Her first-author publication was recently published in Nature Communications.
In this study, the team focused on macroH2A1, which Dr. Avgousti calls “an enigmatic histone.” It is “both up-regulated and down-regulated in multiple scenarios, including many cancers, making it very challenging in the field to understand how it functions to support genome integrity of the host cell,” she explains.
“We showed in 2023 that [herpes simplex virus (HSV)] takes advantage of macroH2A1-associated heterochromatin as a highway to escape the nucleus,” Dr. Avgousti continues.
Dr. Kelnhofer-Millevolte was curious if HCMV, a cousin to HSV, might also use macroH2A1 to optimize infection. Indeed, she observed that infectious viral progeny are reduced 30-fold when produced from skin cells lacking macroH2A1. The nuclear structure also is impacted: instead of a pronounced bean shape with a large vIAC, infected macroH2A1-knockout fibroblasts have smaller vIAC and mis-formed nuclei. Moreover, the virions produced from macroH2A1-knockout cells are also different: they are less infectious due to capsid malformation.
This data suggests that viral maturation is impacted by macroH2A1. To understand how, the team performed RNA sequencing of infected macroH2A1-knockout fibroblasts compared to wild-type. They found that expression of many neuronal genes was lost upon macroH2A1 knockout.
Wait a second. Neuronal genes? Why would axon development or neuronal tracking be expressed in skin cells? Well, in uninfected fibroblasts, they aren’t. But during lytic HCMV infection, neuronal pathways are somehow induced. What’s more, this neuronal gene program is dependent on macroH2A1, because when macroH2A1 is gone, this gene expression is lost.
“We show for the first time, to our knowledge, that a virus activates specific dormant neuronal genes to promote viral maturation” explains Dr. Avgousti. “This has not been shown before as far as I know in any capacity.”
Is this just a happy accident (for the virus), or is there something else at work?
To understand this, the team decided to profile known histone-hijacker: the viral protein IE1. Through binding other histones, IE1 has been shown to drive gene expression of host genes to benefit HCMV.
Intriguingly, the team found that IE1 does indeed interact with macroH2A1. IE1 binds a patch of macroH2A1 that is shared with other histones, which makes sense given its role in general chromatin hijacking. However, IE1’s interaction with macroH2A1 specifically appears to be crucially important for expressing neuronal genes during fibroblast infection.
“IE1 specifically targets histone variant macroH2A … to de-repress hundreds of genes that otherwise have no business in this cell type (neuronal trafficking genes in fibroblasts),” Dr. Avgousti explains.
Why does it do this? It turns out, several of these neuronal genes appear to have a role in viral maturation and spread. Dr. Kelnhofer-Millevolte tested this via knockdown a panel of the most highly induced neuronal genes and found that viral plaque size was significantly reduced in many cases.
“It means that HCMV ‘knew’ that these neuronal genes were there, silenced and dormant, and that some of those genes could allow progeny viruses to be more infectious,” she continues. “So it evolved a way to activate these genes, by targeting a histone that is not rapidly evolving, such that viruses are 30x more infectious than without this mechanism,” Dr. Avgousti says.
Future work will focus on how clinical isolates of HCMV may hijack macroH2A1 or other host chromatin proteins. This could inform therapeutic strategies against HCMV using FDA-approved compounds.
Beyond just HCMV, Dr. Avgousti is especially interested if “macroH2A targeting is conserved across evolution of other types of virus,” she says. “Amazingly, two different herpesviruses target this vulnerable histone variant in opposite ways … Making it clear that chromatin vulnerabilities exploited by viruses is a thing!”
Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Member Dr. Adam Geballe contributed to this research.
The spotlighted research was funded by the National Institutes of Health, the University of Washington Magnuson Scholarship, and National Research Foundation of Korea. This research was also supported by start-up Funds from the Fred Hutch, the University of Miami, and the RNA Bioscience Initiative at the University of Colorado School of Medicine.
Kelnhofer-Millevolte LE, Smith JR, Nguyen DH, Wilson LS, Lewis HC, Arnold EA, Brinkley MR, Shin K, Ahn JH, Kim ET, Kulej K, Geballe AP, Ramachandran S, & Avgousti DC. 2025. Human cytomegalovirus induces neuronal gene expression through IE1 for viral maturation. Nature communications. https://doi.org/10.1038/s41467-025-61915-7
Hannah Lewis is a postdoctoral research fellow with Jim Boonyaratanakornkit’s group in the Vaccine and Infectious Disease Division (VIDD). She is developing screens to find rare B cells that produce protective antibodies against human herpesviruses. She obtained her PhD in molecular and cellular biology from the University of Washington.
For the Media