4 March 2009. Disturbances in glutamate signaling caused by disruptions in the N-methyl-D-aspartate receptor (NMDA-R) pathway have been implicated in numerous psychiatric disorders, particularly schizophrenia (see Current Hypotheses by B. Moghaddam and D. Javitt). Though much is known about this pathway, especially the role of the calcium/calmodulin-dependent protein kinase II (CaMKII) in synaptic plasticity, the mechanisms linking NMDA-R dysfunction to the cognitive and behavioral phenotypes seen in schizophrenia are largely unknown. A new study led by Jannet Kocerha of The Scripps Research Institute in Florida and published online on 5 February in Proceedings of the National Academy of Sciences USA suggests that a microRNA (miRNA), miR-219, may be one such link.
miRNAs, snippets of RNA about 20 nucleotides long that block gene expression by binding to mRNA (which either gums up protein translation or targets the mRNA strand for destruction), are attractive candidates for neuropsychiatric research. As Joseph Coyle of Harvard Medical School points out in a commentary accompanying the new PNAS article, of the several hundred miRNAs identified in the 15 years since their discovery in the nematode C. elegans, “half . . . are expressed predominantly or exclusively in brain,” where they are believed to play crucial roles in orchestrating neural development and regulating plasticity (see SRF related news story).
Putting brakes on a CaM
After injecting mice with dizocilpine, an NMDA-R antagonist similar to phencyclidine (PCP) that produces schizophrenia-like behaviors in both rodents and humans, Kocerha’s team used microarrays to measure miRNA expression levels in the prefrontal cortex, and they found a significant reduction in miR-219, a brain-specific miRNA. Using another model of schizophrenia, mice carrying a hypomorphic mutation in Grin1, which codes for the NR1 subunit of the NMDA-R, the researchers found that miR-219 levels were similarly reduced in both prefrontal cortex and hippocampus.
The antipsychotic drugs haloperidol and clozapine are known to reverse the behavioral phenotypes seen in dizocilpine-treated or Grin1-mutant mice. When the scientists administered either drug, the hyperlocomotion and stereotypy characteristic of these mouse models was sharply reduced, and miR-219 expression was returned to baseline levels.
Having established that pharmacological or genetic disruption of the NMDA-R down-regulates miR-219, Kocerha and colleagues next used bioinformatic techniques to identify potential targets for miR-219. Out of six hybridization candidates for miR-219, the team settled on the gamma subunit of CaMKII (CaMKIIγ) since this kinase is involved in NMDA-R trafficking and rapid, transient expression of the receptor in neural dendrites to regulate plasticity. In vitro analyses with neuron-like cultured cells confirmed that miR-219 suppresses CaMKIIγ expression, an effect that was reversed when the cells were transfected with an antisense inhibitor of miR-219. These results were confirmed in cells from mouse prefrontal cortex. In complementary in vivo experiments, an antisense inhibitor of miR-219 was continually infused into the third ventricle of mice for seven days, and CaMKIIγ expression levels again increased. Moreover, infusion of the antisense inhibitor significantly attenuated the behavioral effects of dizocilpine. All told, the authors write, “these data support the hypothesis that miR-219 represents an integral component of NMDA-R signaling.”
In his commentary, Coyle identifies several “inconsistencies” in the data. First, the authors report that reductions in miR-219 levels in dizocilpine-treated mice were restricted to prefrontal cortex, but “the pathology of schizophrenia is widespread in the cortex,” writes Coyle (see also Coyle, 2006). Although miR-219 levels declined after acute administration of dizocilpine, this decline was not observed with chronic administration of the drug, which Coyle, citing the work of Mohn and colleagues (Mohn et al., 1999), says “is considered to be a better model of schizophrenia.” Moreover, in the Grin1-mutant mice—a genetic analogue to chronic dizocilpine administration—miR-219 was reduced in the hippocampus as well as the prefrontal cortex. Finally, though low miR-219 levels are implicated in the increased locomotion seen after acute dizocilpine treatment, the fact that this behavior was reduced when the actions of miR-219 were blocked with an antisense inhibitor is “counterintuitive,” says Coyle.
On the positive side of the ledger, Coyle notes that the gene coding for miR-219 is located at 6p21, which has been identified as a risk locus for schizophrenia (Roig et al., 2007). He also points out that miR-219 transcription is regulated by a complex including the circadian rhythm protein CLOCK, which has been implicated in bipolar disorder (see SRF related news story). "Thus, miR-219 provides a nexus for 2 risk pathways for serious mental illness: (i) psychosis via hypofunction of NMDA receptors through a downstream effect on CaMKIIγ, and (ii) mood instability by disruption of CLOCK–BMAL1 function, which has been implicated in bipolar disorder," he writes.—Pete Farley.
Kocerha J, Faghihi MA, Lopez-Toledano MA, Huang J, Ramsey AJ, Caron MG, Sales N,
Willoughby D, Elmen J, Hansen HF, Orum H, Kauppinen S, Kenny PJ, Wahlestedt C. MicroRNA-219 modulates NMDA receptor-mediated neurobehavioral dysfunction.
Proc Natl Acad Sci USA. 2009 Feb 5. Abstract