30 Mar 2010
31 March 2010. A genomic deletion associated with increased risk for schizophrenia disrupts communication between regions of the brain involved in working memory in mice, according to a study published in Nature on March 31. Mice engineered to mimic the human chromosome 22q11.2 deletion showed impaired signaling between the hippocampus, crucial for memory, and the prefrontal cortex, which is involved in complex mental processes such as decision-making and planning.
Although only 30 percent of people with the 22q11.2 deletion also have schizophrenia, the deletion provides a toehold for researchers trying to understand how this specific genetic defect translates into abnormal behaviors. In 2008, Joseph Gogos and Maria Karayiorgou at Columbia University, New York, introduced a mouse model of the 22q11.2 deletion that lacked a 1.5 Mb region of mouse chromosome 16 that contains the same genes lost in the human deletion. These mice display impaired learning and prepulse inhibition—features that, though not diagnostic of schizophrenia, are commonly found in the disorder. Their neurons also have altered production of certain microRNAs (see SRF related news story) and lack post-translational add-ons called palmitates that anchor proteins to their proper location (see SRF related news story).
In the new study, Gogos and Karayiorgou team up with Joshua Gordon, also of Columbia University, to examine the electrical signals of the brain while the mice learn and perform a working memory task. Although disruptions in how parts of the brain talk to each other have previously turned up in schizophrenia, pinpointing the precise neural circuits involved has been limited by the resolution of human brain imaging. The new study steps into this void by recording activity from single neurons and groups of neurons in a mouse model that carries a clear genetic risk factor for schizophrenia.
Out of synch
Using electrodes implanted in the hippocampus and the prefrontal cortex, first author Torfi Sigurdsson and colleagues simultaneously monitored electrical activity in both regions to measure how synchronized they were—an indicator of how well connected they are—while the mice learned a simple T-maze task. In the sample phase of the task, mice ran down the track to a T-intersection, where they were directed to go into one of the two arms. In the following choice phase, the mice had to remember where they had been and go to the arm that they hadn't visited during the previous sample phase. The researchers focused on the neural activity preceding the mouse's entrance into one of the arms.
The electrodes picked up the action potentials of single neurons as well as field potentials stemming from the summed activity of large populations of neurons. Field potentials oscillate at various frequencies, and one of particular interest is the theta rhythm, which clocks in at 4-12 Hz and is associated with learning and memory. The theta rhythm drives spiking in cells elsewhere in the brain, including the prefrontal cortex, and the synchronous activity that normally results between the two regions indicates a robust connection between them.
The researchers found that, although signals from the hippocampus and prefrontal cortex appeared normal in the 22q11.2DS mice, their timing was off. Compared to wild-type littermates, the theta rhythms in the hippocampus and the prefrontal cortex did not oscillate together as much, and single neurons in the prefrontal cortex did not fire as consistently at the same phase of the hippocampal theta rhythm. This reduced synchrony was apparent during both the sample and choice phases of the task, though like the wild-type mice, synchrony was slightly higher during the choice phase, which requires memory retrieval.
The researchers next explored whether out-of-synch signaling between the hippocampus and prefrontal cortex could explain the longer time it took for the 22q11DS mice to learn the T-maze task. When they measured the theta rhythm synchrony between these two regions that occurred before the mice even began training on the task, they found that mice with weaker synchrony ended up taking longer to learn the task than mice with stronger synchrony. Theta rhythm synchrony grew as the mice—both wild-type and mutants—learned the task. These correlations suggest that reduced synchrony between the hippocampus and prefrontal cortex contributes to a working memory impairment.
The study shows how subtle changes in timing can disrupt information flow in the brain, which in turn can profoundly influence behavior. The authors note that future studies will have to examine the detailed anatomy of the axons and synapses that connect, both directly and indirectly, the prefrontal cortex to the hippocampus in these mice. It will also be important to identify which individual genes or combination of genes within the 22q11.2 deletion contribute to neural synchrony. In the meantime, this study's focus on neural activity and its timing elucidates another important step along the winding route that leads from a specific genetic defect to behavior.—Michele Solis.
Sigurdsson T, Stark KL, Karayiorgou M, Gogos JA, Gordon JA. Impaired hippocampal-prefrontal synchrony in a genetic mouse model of schizophrenia. Nature 2010 April 1.