1 September 2009. Genetically engineered mice without the brain protein kalirin show sparse dendritic spines and decreased glutamatergic signaling in the frontal cortex, according to a study in the August 4 PNAS. The mice also show changes in cognition and behavior that are reminiscent of those seen in some people with schizophrenia. Furthermore, some of the deficits emerged only as the knockout mice matured, echoing developmental aspects of the disease. The work, from Peter Penzes of Northwestern University in Chicago, Illinois, and colleagues, fits with the known role of kalirin in dendritic spine formation, and raises the intriguing possibility that lower kalirin levels seen in brain tissue of people with schizophrenia could play a part in the disease.
In the brain, most excitatory nerve impulses land on dendritic spines, postsynaptic linchpins of learning and memory (see SRF related news story; SRF news story; and SRF news story). Research on neurons, and more recently, on mice, suggests that dendritic spine formation depends on kalirin (for reviews, see Penzes and Jones, 2008; Rabiner et al., 2005; Sommer and Budreck, 2009). The protein, a guanine-nucleotide exchange factor, prompts Rac1, a GTPase from the Rho family, to bind GTP. This triggers remodeling of actin, which forms a framework for the spines.
The schizophrenia connection unfolds from observations that the prefrontal cortex of subjects with the disease contains too few dendritic spines (see, e.g., Glantz and Lewis, 2000). This may go hand-in-hand with their shortage of kalirin-7, the most plentiful kalirin isoform in normal adult brains (see SRF related news story).
To shed light on how kalirin might affect phenotypes tied to neuropsychiatric disease, first author Michael Cahill and his coauthors developed mice without the gene KALRN, which encodes kalirin. “Surprisingly, we found that the knockout of KALRN produces striking, yet selective, age-dependent, anatomical, functional, and behavioral deficits, which resemble disease-related phenotypes,” they write.
To the naked eye, the mice looked normal, despite the absence of kalirin. However, their prefrontal cortex showed less Rac1 activity than that of wild-type mice. Since Rac1 activity in the hippocampus remained normal, the researchers think that Rac guanine-nucleotide exchange factors other than kalirin may keep communication channels in the hippocampus open. The cortex, on the other hand, may need kalirin to keep neurons talking.
In contrast to the Rac1 findings, the cortex and hippocampus of the knockout mice showed normal levels of Cdc42, a Rho-like GTPase not controlled by kalirin. “This further indicates that kalirin loss produces a specific deficit in cortical Rac1-GTP levels,” write Cahill and colleagues.
Quiet gene stymies signaling
Next, Cahill and colleagues explored how silencing KALRN affects the dendrites. In the frontal cortex, but not the hippocampus, they found only scattered dendritic spines in the knockout mice. This low density of dendritic spines appeared age-dependent: The researchers saw it only at age 12 weeks and not at three weeks, suggesting a parallel between the development of dysfunctional spine formation and the course of schizophrenia, which typically manifests in adolescence or later.
In mature cortical neurons in culture, overexpression of kalirin-7 reversed the deficits, further implicating kalirin in spine formation. The protein may also help slip the AMPA type of glutamate receptors into spines and control neural signaling involving these receptors (Xie et al., 2007). In fact, when Cahill and colleagues examined layer 5 pyramidal neurons from the frontal cortex, they found reduced neurotransmission mediated by AMPA receptors in the knockout mice.
Fleshing out the phenotype
When the researchers turned to cognition and behavior, they found changes in the knockout mice that echo some of those found in subjects with schizophrenia. For instance, at age 12 weeks, the kalirin-deficient mice performed poorly on tests of working memory, despite having tested normal at three weeks. In contrast, their reference memory, which involves associations across trials, seemed to match that of their wild-type peers. To explain these findings, Cahill and colleagues note that working memory taps connections between the prefrontal cortex and the hippocampus, whereas reference memory stems from the hippocampus and does not rely on the prefrontal cortex.
Besides working memory impairments, the knockout mice also showed flawed sensory-motor gating in a prepulse inhibition paradigm. In addition, they lacked a healthy interest in socializing. At age 12 weeks, they walked excessively in their cages, which they had not done when they were younger. The atypical antipsychotic clozapine curbed their hyperactivity but did little for their memory.
Four-legged schizophrenia models?
Clearly, this study sheds light on kalirin’s in vivo effects. It says much about the protein’s ability to promote Rac1 activity, spine building, and glutamatergic signaling, with consequences for cognition and behavior. In addition, Cahill and colleagues present some evidence that it influences only selective outcomes, bolstering the case that it may play a unique role in cortical and behavioral deficits of interest in schizophrenia.
“The age-dependent delayed functional and behavioral impairments in schizophrenia have been difficult to explain in humans and to model in animals,” write the researchers. Time will tell whether kalirin-knockout mice can help with those tasks.—Victoria L. Wilcox.
Cahill ME, Xie Z, Day M, Barbolina MV, Miller CA, Weiss C, Radulovic J, Sweatt JD, Disterhoft JF, Surmeier J, Penzes P. Kalirin regulates cortical spine morphogenesis and disease-related behavioral phenotypes. PNAS. 2009 Aug 4;106(31):13058-13063. Abstract