AP-2 transcription factors, which control sleep in flies and worms, are confirmed to do the same in mammals.

Humans are not alone in their deep need for sleep. Almost all animals, even tiny nematode worms and fruit flies, suffer when deprived of their Z’s, but little is known about how sleep is controlled. New work published in GENETICS advances our understanding of this mysterious physiological state by pinpointing a key gene family that affects sleep architecture.

Two research groups independently conducted parallel studies on neural crest-derived AP-2 transcription factors in mice. Their work demonstrates that the associated genes play a conserved role in mammalian sleep, though there is some evolutionary divergence and added complexity in mammals compared to invertebrates.

From Worms to Mice

Henrik Bringmann leads the Max Planck Research Group on Sleeping and Waking. In 2013, his laboratory screened C. elegans worms for mutations affecting sleep. AP-2 deletion mutants showed no detectable sleep, suggesting these transcription factors are important regulators of the process. In 2016, Bringmann led another study showing that AP-2 transcription factors are also needed for sleep in Drosophila

“Sleep appears to have evolved at least 600 million years ago and has been conserved since then, which suggests that many fundamental principles of sleep regulation are conserved,” said Bringmann. “This said, sleep is more complex in humans than in worms or flies, so we need mammalian models to understand additional levels of complexity.”

The researchers’ next step was to establish whether a similar sleep-regulating role is played by related genes in mammals, the TFAP2 genes. In their most recent study, the team generated mice that were heterozygous knockouts for either Tfap2a or Tfap2b and compared each mutant to wild-type littermates. In addition to analyzing sleep duration and brain waves, they examined symptoms of sleep deprivation such as memory loss and stress resistance.

Since the AP-2 transcription factors promote sleep in invertebrates, researchers initially expected that the equivalent genes would also promote sleep in mice. They were therefore surprised to find that while losing Tfap2bfunction reduced both sleep quality and quantity, Tfap2a mutants slept for the typical amount of time and the quality of their sleep was actually higher than wild-type. This result suggests that the function of AP-2 transcription factors has diverged over the course of evolution, perhaps to allow sleep quality to be fine-tuned in either direction.

Working Backwards from Human Disorders

Meanwhile, another research group also published an article in the same issue of GENETICS examining the effects of two specific TFAP2B mutations in mice. Yu Hayashi of Kyoto University  and University of Tsukuba said his lab was inspired to pursue this avenue of research by a paper in PNAS. That study described multiple human families with mutations in TFAP2B that are associated with a rare disorder known as Char syndrome. Two of these families exhibited symptoms of disordered sleep, including sleepwalking and extremely shortened sleep.

“We were astonished to read about a family that slept just two to three hours per night and had no signs of fatigue. We thought that maybe analyzing this gene can help us understand what sleep is for in the first place,” said Hayashi. “We even thought that maybe this mutation might somehow substitute for the function of sleep. Could there be a way to help humans need less sleep?”

To answer these questions, Hayashi’s team set out to replicate those families’ specific TFAP2B point mutations in mouse models. They measured both how long the mice stayed awake and the duration of different sleep phases compared to wild-type mice and heterozygous knockouts. The results showed that TFAP2B helps determine the amount of nonrapid eye movement sleep (NREMS). However, the effects of the point mutations in mice did not match the symptoms observed in humans—the mice did not sleepwalk or show dramatically shortened sleep.

In the mice that carried the same mutation as the human family with short sleep, female mice showed fragmented sleep, while male mice did not. This result was unforeseen because there was no reported gender difference in the human family. “It was surprising for me to see the gender difference,” said lead author Ayaka Nakai, a graduate student in the Hayashi lab. It is possible that future research may uncover sex differences in sleep regulation.

In terms of generating a model of fatigue-free short sleep, the results did not match their initial hopes—sleep was reduced overall in heterozygous mutant mice, but it was generally fragmented rather than shortened. However, the results clearly established TFAP2B’s important role in sleep architecture and laid the groundwork for learning more about how sleep works.

Next Steps

Ultimately, said Bringmann, “it was satisfying to hear that both approaches converged on the same conclusions regarding TFAP2B’s role in sleep.” Together, these two papers establish that AP-2 transcription factors contribute to sleep control in mammals—just as they do in flies and worms.

The gene’s evolutionary conservation is a key lesson from the study, said Nakai.

For Nakai and Yu, the next research step is creating knockdown mice that have Tfap2b downregulated only in the nervous system. This will allow them to observe the neural effect of a homozygous loss-of-function mutation, which is developmentally lethal if the knockout is genome-wide. Studying a homozygous knockdown may offer clearer insights into how the gene affects neuron specification and activity.

Next, Bringmann is interested in examining other mammalian AP-2 paralogs, but he says invertebrates will continue to be important for identifying other genes that contribute to sleep regulation. “Going back and forth between different models will be the future of sleep research for the next years, and this will be facilitated by looking at homologous genes and conserved principles.”


Functional Divergence of Mammalian TFAP2a and TFAP2b Transcription Factors for Bidirectional Sleep Control

Yang Hu, Alejandra Korovaichuk,  Mariana Astiz, Henning Schroeder, Rezaul Islam, Jon Barrenetxea, Andre Fischer, Henrik Oster and Henrik Bringmann

GENETICS 2020 216: 735-752. 


Sleep Architecture in Mice Is Shaped by the Transcription Factor AP-2β

Ayaka Nakai, Tomoyuki Fujiyama, Nanae Nagata, Mitsuaki Kashiwagi, Aya Ikkyu, Marina Takagi, Chika Tatsuzawa, Kaeko Tanaka, Miyo Kakizaki, Mika Kanuka, Taizo Kawano, Seiya Mizuno, Fumihiro Sugiyama, Satoru Takahashi, Hiromasa Funato, Takeshi Sakurai, Masashi Yanagisawa and Yu Hayashi

GENETICS 2020 216:753-764


Grace Niewijk is a freelance science writer and editor from Chicago. Her writing has appeared in Genetic Engineering & Biotechnology News, Clinical OMICs, Proto Magazine, and elsewhere. She holds a degree in molecular biophysics and biochemistry from Yale.

View all posts by Grace Niewijk »