The findings, reported in the April 29 issue of PNAS Early Edition, help explain social interactions among female animals, and may shed light on social phobias and disorders in humans.
The researchers, led by Rockefeller professor Donald Pfaff, Ph.D., show that strains of female mice that lack the genes for oxytocin and the estrogen receptors alpha and beta fail to recognize normal female mice after repeated instances when the normal animal was placed in the same space. The gene "knockout" mice also failed to investigate a new "intruder mouse" under circumstances where a genetically normal mouse would do so.
As a result the Rockefeller team was able to infer for the first time the roles of oxytocin and estrogen receptor genes in social recognition in female mice.
"We have linked four genes known to be involved in social interaction into one model to explain the behavior known as social recognition," says first author Elena Choleris, Ph.D. "Social recognition is a crucial behavior in animals including humans that live in groups, because in order to know who your friends, bosses and enemies are, you need to not only recognize individuals but remember who they are."
"The effect of knocking out three separate genes and producing similar changes in behavior "formed a parallelism that fits into a very large literature having to do with molecular biology and social behavior, which we've condensed into a single hypothesis," says Pfaff, who heads the Laboratory of Neurobiology and Behavior at Rockefeller.
"The notion that cognition having to do with social life is different from other forms of cognition as managed by the brain is becoming widely accepted among scientists who study brain and behavior," adds Pfaff. "So 'smartness' about social life is different from smartness about SAT scores."
According to Thomas R. Insel, M.D., director of the National Institute of Mental Health, the branch of the National Institutes of Health that supported this research, "Don Pfaff and his colleagues continue to demonstrate interesting and important relationships between genes, brain and behavior. In this paper, they show that knockouts of different genes result in remarkably similar deficits. The implication is that these systems compose part of a 'micronet' necessary for complex behavior.
"What I find exciting about this work is that Pfaff's group has taken on a behavior as complex as social recognition, one that might not seem amenable to a single gene approach," adds Insel. "What they demonstrate is that this ostensibly complex behavior can be completely abolished by not only one, but several single gene 'knockouts.' Thus, what may seem complex and scientifically unapproachable to us may have a relatively simple and highly accessible molecular basis."
In the new research, Choleris, Pfaff and their colleagues studied changes in social recognition in female mice when each gene coding for the neuropeptide oxytocin, estrogen receptor alpha or estrogen receptor beta, was removed from the social recognition pathway in mice. Estrogen receptors are large proteins, which bind the sex hormone estradiol. The scientists created "knockouts," or mice genetically altered to lack certain genes, for oxytocin and the estrogen receptors. (Although oxytocin receptor knockout mice do not exist -- this particular gene is necessary for survival -- other researchers have shown that blocking the oxytocin receptor in rats impairs social recognition.)
Scientists evaluate social recognition in mice using a "social habituation/dishabituation" test, in which an intruder mouse is presented to another mouse in its home cage several times so that the resident mouse becomes familiar with the intruder mouse.
"If you have a mouse in its home cage, and you introduce a second mouse, the resident will sniff and investigate," says Choleris. The time that the resident mouse spends investigating is measured. The intruder mouse is removed from the cage and replaced after 15 minutes. The researchers repeated this four times, and the amount of time the resident mouse spent investigating the intruder decreased. This is called the habituation phase. The researchers then introduce a different intruder mouse, and normally the resident mouse's sniffing and investigating increases again. This is called dishabituation.
When the researchers introduced an intruder mouse to a resident mouse that lacked the genes for oxytocin or the estrogen receptors, the scientists did not observe habituation -- the resident mouse's investigation of the intruder did not decrease as a normal test mouse's would after repeated introductions of the intruder.
"The gene 'knockout' resident mice acted as if the same mouse was always new," says Choleris. "They showed the same level of investigation as if it were a new mouse every time."
The researchers hypothesize that a basal forebrain system brings together estrogen-responsive neurons and oxytocin-producing neurons to regulate and modulate social recognition. For mice, this function begins in the nose. Animals detect chemical cues though the main olfactory system, which processes odors, and the accessory, or vomeronasal, olfactory system, which identifies pheromones. Nerve cells from both systems project from the nose's epithelium lining back to the amygdala, the almond-shaped brain region that houses the body's center for emotions, including fear and anxiety.
While the amygdala is processing information from the nose, another brain region, the hypothalamus, is busy producing oxytocin, which is transported to the amygdala where it binds to oxytocin receptors. At the same time, the ovaries are producing estrogen, which enters the blood stream and binds to estrogen receptor alpha in the amygdala and estrogen receptor beta in the hypothalamus.
"It was previously known that estrogen affects oxytocin production, and now we think that estrogen affects oxytocin synthesis and release through estrogen receptor beta," says Choleris. "And we know that estrogen receptor alpha is present in the amygdala and it is necessary for induction of the gene for the oxytocin receptor.
"So, here we have all four genes coming into play. If you interrupt this micronet at any of these levels, you can predict what we see in terms of impairment of social recognition." For example, Choleris says, if the induction of the oxytocin receptor is blocked, oxytocin will not be able to regulate social recognition.
The researchers say that these findings could yield new insights into certain aspects of human social behavior. Oxytocin, for example, is clearly involved in the maternal bond found in humans. This "peptide of affiliation" also is known to regulate anxiety, so it may play a role in social phobias. And there is evidence in certain pathologies, like autism, that oxytocin is deficient.
"The manner in which these findings comprise a mouse-based theory is through the olfactory input - social investigation follows from sniffing, so our idea depends upon the prominence of pheromonal and other olfactory inputs to the amygdala," says Pfaff.
As Pfaff explains, in the human brain, the amygdala is tied to social and emotional behaviors. "The relationship between social recognition, social memory and aggression is probably a dynamic that is not limited to mice, even though this specific circuit and these specific olfactory cues most certainly are."
In addition to Choleris and Pfaff, other authors of the paper are Sonoko Ogawa, Ph.D., at Rockefeller; Jan-Åke Gustaffson, M.D., Ph.D., at the Karolinska Institute, Sweden; Kenneth S. Korach, Ph.D., at National Institute of Environmental Health Sciences, Research Park Triangle, N.C.; and Louis Muglia, M.D., Ph.D., Washington University, School of Medicine, St. Louis, Mo.