Social order of mice 5 January 2013

Like humans, mice quickly develop social order

    Adult mice arrange themselves into leaders and followers, but not if they have autism-like traits, suggesting new directions for research.
  • icon_zoom.png
    In their Weizmann lab with large mouse enclosures are, clockwise from left, Tali Kimchi with Aharon Weissbrod, Alexander Shapiro and Molly Dayan In their Weizmann lab with large mouse enclosures are, clockwise from left, Tali Kimchi with Aharon Weissbrod, Alexander Shapiro and Molly Dayan Copyright: Weizmann Institute of Science
    By Avigayil Kadesh
    Put five unacquainted adult males together in a house, and within hours one will take his place as the alpha male and another as the beta. Add a couple of females, and you can bet the dominant dude and his sidekick will claim them.
    What’s true for humans – if you’ve ever watched an episode of the reality show Big Brother, this scenario will seem familiar – is also true for mice. A team led by Israeli neurobiologist Dr. Tali Kimchi has done the world’s first deep, systematic study of the mechanisms regulating the social behavior of these little mammals, whose brains and genetic makeup are close to that of people.
    The results, published in June 2013 in Nature Communications, provide insight into normative social behavior. And because the researchers also watched the behavior of mice engineered to display typical symptoms of autism, the study sheds new light on societal aspects of social disorders.
    Using a proprietary automated system, the scientists identified dozens of individual social behaviors, such as seeking out specific companions for activities or rest, avoiding certain individuals and attacking others. They were able to isolate and identify typical behaviors of individuals, pairs and groups. The system even predicted, with near-perfect accuracy, the mating patterns of the mice.
    Ruler of the roost 
    “We took mice that were strangers to each other, all the same age and strain,” Kimchi explains. “In the first stage, we put five males in the cage. We could see clearly, every time we did this experiment, that in a matter of a few hours there was a dominant male established. He ruled the enclosure and chased the others, and decided if others could approach the food and water chambers. Later, he mated with the females we added, and chased away competitors.” 
    One mouse always took the No. 2 position in the social order. “We could see that he always waited for a chance to replace the dominant male, and they had the most interaction,” says Kimchi. The other three mice remained subordinate to the two top honchos throughout each week-and-a-half study. 
    On day seven, two females were added, causing an immediate social imbalance. “Every mouse in the system became significantly more active and social,” Kimchi reports. “There was some resulting instability, but in a few hours everything went back to the way it was before, and the alpha reclaimed his position.”  
    What determines which mice will be dominant, subordinate or submissive? How much of this is controlled by genes, how much by environment and how much is the result of the interaction between the two?  
    “We don’t know this yet, but hopefully we will use our device to figure it out,” Kimchi says.
    A mouse in the house
    Previous social studies of mice involved only a pair of the rodents, in a typically small lab cage that offered limited ability to interact. Moreover, researchers needed to wade through hours of footage to quantify10 minutes of social behavior. 
    Kimchi’s experiments at the Weizmann Institute of Science in Rehovot were structured in a revolutionary way, based on methods she had devised over four and a half years of post-doc research at Harvard University. 
    “When I came to Weizmann in 2008, I set a goal to build an automated system to allow the study of complex social interaction with multiple individuals under semi-natural conditions mimicking as far as possible behavior of mice in the field,” she says. “Nobody had ever done this before, and it took us about three years to build the setup.” 
    This was accomplished by Kimchi’s research team – including Aharon Weissbrod, Genady Wasserman and Alex Shapiro -- together with Ofer Feinerman of the Institute’s Physics of Complex Systems Department. 
    They built an enclosure of about three feet by three feet -- 50 times larger than the standard cage for studies of mouse behavior. Since mice, like humans, are social creatures, groups of five or more were studied at a time.
    Inside the sophisticated enclosure free of human intervention, the position of each mouse was tracked via a microchip telemetry system similar to that planted under the skin of pets for ID purposes, fused with video-tracking software with infrared capabilities for the dark.  
    All components were invented, patented and built at the Weizmann, and allowed the researchers to track the position of each individual, 30 times per second, in a spatial resolution of less than a centimeter. From this rich recorded data, the scientists extracted detailed social profiles of each mouse.  
    Autistic mice 
    In order to see the interactional differences between “normal” mice and those with the social difficulties and repetitive behaviors typical of people with autism spectrum disorders, Kimchi’s team put mice bred to have autistic behaviors into the enclosure and watched what happened. 
    The autistic mice groups rarely established a dominant male. If they did, the social order did not remain stable. This indicates, says Kimchi, “that if you don’t know who’s in front of you, you cannot form relationships.” 
    Thanks to the experiments and the ability to share recorded data with researchers anywhere, Kimchi’s team has effectively invented a new tool to deepen the understanding of social, aggressive, sexual and maternal behaviors critical to the survival of every species – as well as abnormal social behaviors.  
    “We could use this for neuropsychotic research by using different types of mouse models, and maybe as a tool to evaluate drug treatments for various disorders,” says Kimchi. 
    “What is striking to me is that you can see the same principles, the same instinctive behaviors, in mice and humans. As a behaviorist, I see that you can use a mouse model very nicely to study instinctive behaviors because 98% of genes correlate between mice and humans, as well as brains and hormones.” 
    Her lab is now studying additional models to look for similar features in mice with different neuropsychiatric disabilities. “This will allow us, perhaps, to screen for novel genes related to those disorders,” she says.