‘Dux’ and Cover: an enigmatic protein’s potential role in immune evasion

From the Tapscott Lab, Human Biology Division

If there’s anything we know about Mother Nature, it’s that she’s usually very efficient. When it comes to proteins, this efficiency can manifest as multifunctionality: one protein that carries out multiple, sometimes quite diverse tasks. The list of multifunctional proteins expands by the day, but this time around, all eyes are on DUX4, a quirky member of a family of transcription factors which are extensively studied by the Tapscott Lab in the Human Biology Division at Fred Hutch. A recent study out of the Tapscott Lab, spearheaded by graduate student-turned-postdoc Dr. Amy E. Spens, proposes a new mechanism by which DUX4—in addition to regulating developmental gene expression—suppresses innate immune responses by interacting with the well-known immune modulator STAT1.

But first, let’s take a step back. DUX4 is one member of a family of DoUble homeoboX (DUX) transcription factors which are uniquely expressed in placental mammals. Its canonical function is to regulate gene expression; specifically, DUX4 is expressed very early in embryogenesis, where it helps initiate expression of the nascent zygotic genome, a process termed zygotic genome activation. To put it a different way, if your genome is a party, then DUX4 is the popular kid who shows up first, brings all of their friends, and turns on the music. Unfortunately, DUX4 is also notorious as the cause of a devastating disease called FacioScapuloHumeral muscular Dystrophy (FSHD). In this disease, inherited mutations cause the misexpression of DUX4 in skeletal muscle tissue, where it activates an early embryonic gene expression program which eventually causes muscle cell loss.

For a while, research efforts focused on tracking down the transcriptional targets of DUX4 and investigating their roles as targets for potential FSHD treatments. Over the years, however, the Tapscott Lab started collecting pieces of a puzzle which formed a slightly different scene—this one about DUX4 and the immune system. The first piece of the puzzle was reported back in 2012, when former Tapscott Lab postdoc Dr. Linda Geng made the curious observation that cells transduced with lentivirus containing DUX4 didn’t mount an innate immune response, as cells normally do when infected with lentivirus. Another piece of the puzzle was revealed in 2019, when a collaboration between the Tapscott and Robert Bradley labs yielded the discovery that DUX4 is re-expressed in a diverse collection of cancers, where it appeared to stifle immune responses by inhibiting the major histocompatibility complex (MHC) class I, the system by which our cells alert the immune system of potential threats. So, by the time an enthusiastic graduate student named Amy Spens joined the Tapscott Lab, it was clear that DUX4 had a potential role in innate immune suppression in addition to its canonical role as a transcription factor. The juicy open question which remained, however, was how DUX4 accomplished this task.

To start tackling this question, Spens wanted to know how far innate immune suppression by DUX4 could reach: DUX4 clearly had a role when it came to lentiviruses or a specific inflammatory cytokine called interferon gamma (IFNγ) in cancer, but would it suppress immune responses to other stimuli as well? To test this, Spens cultured DUX4-expressing human muscle cells and challenged them with an assortment of molecules (including double-stranded RNA, a DNA-sensing pathway molecule called cGAMP, and a different type of interferon, IFNβ) known to induce a diverse array of innate immune sensing pathways. In non-DUX4 expressing cells, these stimuli induce transcription of a collection of interferon-stimulated genes, or ISGs. In all cases, DUX4 suppressed this ISG induction, suggesting that it functions as a broad suppressor of innate immune pathways in response to diverse triggers.
 

A cartoon illustration depicting a portion of the DUX4 protein wrapping around a strand of DNA, with a two-color background dividing the image in half vertically to illustrate 'two sides' of DUX4 function.
An artistic rendering featuring the two homeodomains of human DUX4 bound to DNA, representing two distinct sides of DUX4 function. Structure from PDB (5Z2T), image edited by D Sokolov.

So, DUX4 appears to be a precocious regulator of innate immunity—but how does this new function relate to its well-described, canonical function as a transcription factor? To tease this apart, Dr. Spens turned to classic molecular biology. One end of the DUX4 protein—the amino, or N-terminus—contains specific features called homeodomains which allow it to bind DNA and regulate transcription. Knowing this, Spens created DUX4 constructs with mutated or missing homeodomains and tested their ability to suppress ISG induction following IFNγ stimulation in cells. Surprisingly, these truncated proteins were still able to suppress ISGs, indicating that DUX4’s canonical role as a transcription factor was dispensable for its suppression of innate immune signaling.

If DUX4 wasn’t using its homeodomains to suppress ISG expression, then how was it accomplishing this task? This time, Dr. Spens turned her attention to the other end of the DUX4 protein—the carboxy, or C-terminus—which contains two small motifs known only by their amino acid composition: LLxxLL. DUX4’s LLxxLL motifs are not as well understood as its homeodomains, but they are evolutionarily conserved throughout the DUXC/DUX4 family, suggesting some important function. Spens created versions of DUX4 lacking these motifs, and again tested their ability to suppress ISGs following interferon stimulation. Shockingly, deleting these two small motifs abolishes DUX4’s ISG suppression capability! So, the two LLxxLL motifs in the DUX4 C-terminal domain are necessary for ISG suppression, but are they sufficient? Gratifyingly, Spens demonstrated that an 85-amino acid C-terminal fragment of DUX4—containing both LLxxLL motifs—maintained the majority of its ISG suppression capability. This capped off a satisfying result: the two faces of DUX4—one of a transcription factor, and another of an immune modulator—are functionally and structurally separable.

At this point, Spens and colleagues were thoroughly convinced of DUX4’s additional role in suppressing innate immune responses, but the mechanistic question of how exactly it accomplished this task remained. To track down DUX4’s partners in crime, the team had to go fishing, figuratively speaking: a research technician in the lab, Nicholas Sutliff, expressed the DUX4 C-terminal domain (DUX4-CTD) in human myoblast cells, treated them with IFNγ, used an antibody to purify out DUX4-CTD, and then used mass spectrometry to identify any proteins that came along with it. “I think that sometimes, big data approaches like mass spec give you more questions than answers,” notes Spens, “But in this case, we were lucky that at the top of the list of DUX4 interactors was STAT1, a super well-known modulator of innate immune signaling!” Indeed, a flurry of follow up experiments confirmed that the CTD of DUX4 interacted with STAT1, that this interaction probably depends on phosphorylation of a specific amino acid on STAT1, and that DUX4-CTD binding to STAT1 prevented it from accessing the promoters of the ISGs which it regulates in response to interferon. Underscoring the potential significance of this novel function, Spens also showed that Dux—the mouse ortholog of DUX4—retained ISG suppression capabilities, suggesting that immune modulation is a conserved function of the DUXC family of proteins.

All told, Spens and colleagues present a new, major piece of the DUX4 puzzle—a puzzle which is not yet complete, but which could have far-reaching implications for better understanding FSHD, cancer, or even development. As she puts it, “There are still so many other interactors to follow up on, and other details of the DUX4-STAT1 interaction waiting to be fleshed out, but I have to remind myself that science happens one step at a time.” In all, Dr. Spens’ study reminds us that—even when we think we know what a protein does—there is sometimes another side of the coin just waiting to be discovered (or, should we say, a ‘DUX side of the moon’?).
 


The spotlighted research was funded by the National Institutes of Health, the Friends of FSH Research, and the Chris Carrino Foundation for FSHD.

Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium member Dr. Stephen Tapscott contributed to this study.

Spens, A. E., Sutliff, N. A., Bennett, S. R., Campbell, A. E., & Tapscott, S. J. 2023. Human DUX4 and mouse Dux interact with STAT1 and broadly inhibit interferon-stimulated gene induction. eLife, 12, e82057.