Walking through Dr. Michael Kyba’s latest mouse model of FSHD

In 2008 and 2009, Friends of FSH Research awarded Dr. Michael Kyba a $100,000 grant to begin developing a mouse model for FSHD focusing on the gene, DUX4 (See D4Z4/DUX4 Induced Pathologies in Mice ). It is important to note that in 2008/09 there was still no consensus that DUX4 was the culprit for causing FSHD. It took the foresight of researchers like Dr. Kyba to begin to develop these models even as the debate was ongoing. Donations like yours allowed us to invest in these high-risk experiments. The preliminary data that Dr. Kyba generated using seed funding from Friends of FSH Research and the FSH Society contributed to him acquiring a large $2 million dollar grant from the National Institutes of Health.

The final results of over 5 years of research were published this week in the journal Cell Reports in the paper entitled Dominant Lethal Pathologies in Male Mice Engineered to Contain an X-Linked DUX4 Transgene. What the study found is that even a small amount of DUX4 introduced into an organism has very bad consequences. Some of these consequences parallel the human disease and teach us about the human condition. Some of the consequences may be specific to mice – but that is also useful to understand.

It is important to stay realistic about why we invest in mouse models of FSHD, what we hope to learn from them, and how we think they will be useful in the future. FSHD is a remarkably “human” disease. This means that DUX4 does not exist in mice, and when we introduce the gene into mice we have to be mindful that it may not behave the same way as in humans. Contrast that to diabetes, where mice and humans metabolize sugar in much the same way. Both organisms have a pancreas with islets containing beta-cells that can sense glucose and respond by secreting insulin. Diabetes in a mouse looks a lot like diabetes in a human. This is not the case with FSHD, nor should we expect it to be.

If DUX4 pathology is not going to be the same in human and mouse, then why do we even bother? It seems like wishful thinking to expect mice to develop a muscular dystrophy that affects the face and shoulder girdles when we introduce DUX4. We pursue this because if we ever want to develop a therapeutic strategy for FSHD, then we need to test safety and efficacy in an animal model to lay the foundation for clinical trials. Therefore, the goal is not to recreate FSHD in a mouse, but rather to understand the consequence of introducing FSHD-causing genes. Although similar phenotypes would be useful, if however DUX4 makes a mouse stand on its hind legs and sing the national anthem, then that is what we have to reverse with our therapeutic strategy.

In this paper, Dr. Kyba’s lab developed a creative approach to introducing DUX4 into mice. The tactic was meant to address a key problem: introducing DUX4 into any cell is very toxic – and chances are if you were to try make a mouse that produced DUX4 it wouldn’t last past the embryo stage. Therefore Dr. Kyba adapted a technique he developed previously in mouse cell culture function on the whole organism. He introduced DUX4 into the cells in such a way that they would only turn on DUX4 when they were fed the antibiotic, doxycycline. As an extra precaution, they introduced the gene into the X chromosome. Females have two X chromosomes, and males have one. Therefore, if DUX4 had a nasty effect, females should still survive because of their ability to shut off one of the X chromosomes entirely, and they would be able to learn about the effects of DUX4 in the males.

There were a number of key findings in this paper that are important. First, they describe that DUX4, even in the absence of the doxycycline trigger, is very toxic. This means that sub-detectable levels of the protein can still wreak havoc, not only on muscle, but also on other organ systems such as the skin and blood vessels. Also importantly, this is the first paper to connect a DUX4-induced activity to the disorganization of blood vessels in the retina, which is a common feature of FSHD in patients.

Another very interesting aspect of this study is the effect of DUX4 on skeletal muscle. Since DUX4 is toxic to all cells, it should be no surprise that DUX4 had toxic effects on muscle in the mouse – but there is still a burning question in the field: precisely when does DUX4 cause muscle damage? Is it the muscle stem cells that are the problem? Is it the muscle fibers? Is it muscle repair that has gone awry? To try and address these questions, the lab activated DUX4 in isolated myoblasts from the mice. Activating the cells to express DUX4 caused them to die, and at low levels impaired their ability to form muscle fibers. They next isolated muscle cells from the DUX4-expressing mice and transplanted them into another mouse. They found that injection of DUX4 expressing cells impaired regeneration of the entire muscle. So the answer to how does DUX4 induce muscle damage? is “yes.” It seems that activation of DUX4 at any stage of the development of muscle can impair its formation, maintenance and regeneration. Whether all of these processes are occurring in patients is unknown.

Mouse models of FSHD are critical for the drug development process. A few years ago, our field had no mouse models and now we have a number of them that each have their own features. Our goal at Friends is to support mouse models that in turn compliment our investments in drug discovery. For example, we have recently funded the Kyba lab to look for direct inhibitors of DUX4. If he identifies compounds that directly interfere with DUX4, he now has a mouse model to follow it up with. The mouse may not have FSHD; but we will be able to tell whether or not the compound he identifies has any effect.

As always, please feel free to contact me if you have any questions about this study.