June 5, 2014. Disrupted connections between the thalamus and auditory cortex are found in a mouse model of 22q11.2 deletion syndrome (22q11DS), a strong genetic risk factor for schizophrenia, reports a new study published online in Science on June 5. Led by Stanislav Zakharenko, St. Jude Children’s Research Hospital, Memphis, Tennessee, the researchers find that this reduction in thalamo-cortical synaptic transmission results from elevated expression of dopamine D2 receptors in the thalamus and can be reversed by antipsychotic medications. They also link haploinsufficiency of a specific gene, DGCR8, within the deleted region to the elevated D2 receptors and thalamo-cortical dysfunction.
The study is “a significant advance” because it shows how the 22q11.2 deletion leads to a relatively specific phenotype, said Christoph Kellendonk of Columbia University, New York City, who was not associated with the study. “It’s not that all projections are altered; it’s really only this specific [thalamo-cortical] one,” he added.
Rob Sweet, University of Pittsburgh, Pennsylvania, also praised the study for its diverse array of techniques and methodological rigor. However, he cautioned against drawing a link between the current findings and schizophrenia, given that human studies have not found a similar disruption of thalamo-cortical function in the disorder.
Of 22q11.2 mice
Based on the understanding that antipsychotics treat positive symptoms of schizophrenia such as auditory hallucinations primarily through blockade of D2 receptors, the researchers set out “to find an auditory cortex circuit that’s not only abnormal in schizophrenia, but also is rescued by antipsychotics,” Zakharenko said.
To model schizophrenia, first author Sungkun Chun and colleagues turned to 22q11DS, the strongest genetic risk factor for schizophrenia that has been identified to date (see SRF related news report). Approximately one-quarter of deletion carriers eventually develop the disease. The researchers used the Df(16)1/+ mouse, a model of 22q11DS lacking part of the mouse homologue of the human deleted region.
Given the strong risk for schizophrenia in 22q11.2 deletion carriers, the Df(16)1/+ mouse model is “certainly one of the [schizophrenia] models that is most worthwhile to study,” said Kellendonk.
Based on the emergence of positive symptoms in the time period from late adolescence to early adulthood in schizophrenia, the authors used slices from mature four- to five-month-old mice. When they measured evoked AMPA receptor-mediated excitatory postsynaptic currents (EPSCs) from pyramidal neurons in layers 3 and 4 of the auditory cortex, the main recipients of projections originating in the auditory thalamus (the ventral medial geniculate nucleus, or MGv), Chun and colleagues observed a reduction in synaptic transmission in slice preparations from Df(16)1/+ mice compared to wild-types. Measured as the slope of the EPSC (or the amount of current over time), this deficit seemed to be presynaptic as well as specific to thalamo-cortical projections: There were no alterations in the synaptic transmission of cortico-cortical, cortico-thalamic, or hippocampal projections in Df(16)1/+ slices.
Treatment with the antipsychotics haloperidol and clozapine reversed the thalamo-cortical synaptic deficits in a dose-dependent fashion in Df(16)1/+ slices, whereas there was no effect of D2 blockade in recordings from wild-type mice. This antipsychotic effect was also specific to thalamo-cortical projections to the auditory cortex, as it was not found in cortico-cortical projections or in thalamo-cortical projections to the somatosensory or visual cortices.
The impaired thalamo-cortical projections seemed to indeed be mediated by D2 receptors. A D2 antagonist raised the EPSC slope in Df(16)1/+ but not wild-type slices; a D2 agonist had no effect on either genotype. In addition, D2 receptor transcript and protein levels were greater in the thalamus (but not the cortex, hippocampus, or striatum) of Df(16)1/+ animals. A significant, though smaller, surplus of D2 protein was observed in postmortem MGv tissue from 13 schizophrenia subjects and 13 controls.
“We found the mechanism in the mouse and tested it in the human, so [the postmortem findings] were a very nice verification,” said Zakharenko.
Kellendonk was less convinced by the human expression data. The majority of the 13 patients with schizophrenia were taking antipsychotics, and this chronic blockade of D2 receptors “could theoretically lead to the upregulation,” he said.
In addition, the disrupted synaptic transmission between the thalamus and the auditory cortex that Chun and colleagues observed in the mouse model seems at odds with the literature, Kellendonk added. Functional connectivity studies performed in patients have instead shown an increased connectivity between the thalamus and sensory cortex, he said.
To identify the specific gene in the deleted region that might be responsible for the thalamo-cortical synaptic deficits, Chun and colleagues screened 12 mouse strains with deletions in gene clusters or single genes located within the 22q11.2 homologous region.
Haloperidol sensitivity was detected in one strain: mice missing one copy of the microRNA processing gene DGCR8 (Dgcr8+/-) that has previously been implicated in several of the phenotypes of 22q11 deletion mouse models (see SRF related news report; SRF news report). Similar to the Df(16)1/+ mice, the Dgcr8+/- mice had elevated D2 receptor mRNA levels in the MGv but not the cortex. They also displayed deficits in prepulse inhibition (PPI) of acoustic startle, a behavioral abnormality found in schizophrenia and observed in some models of the illness, which could be reversed with haloperidol.
Hemizygous deletion of Dgcr8 in the MGv recapitulated the sensitivity of thalamo-cortical projections to antipsychotics. In addition, small interfering RNA (siRNA)-mediated knockdown of the D2 receptor abolished the antipsychotic sensitivity in Df(16)1/+ mice. Overexpression of the D2 receptor in MGv thalamo-cortical projection neurons resulted in antipsychotic sensitivity and PPI deficits in wild-types. Taken together, these data suggest that overexpression of D2 receptors through haploinsufficiency in Dgcr8 drives the thalamo-cortical abnormalities.
Pittsburgh's Sweet said that the mouse cortex finding in the study—that there was no loss of thalamo-cortical terminals—is consistent with his group's postmortem study of auditory cortex in schizophrenia (Moyer et al. 2013). However, he also noted that “there is not yet compelling evidence that auditory thalamo-cortical projections are abnormal in schizophrenia patients.” Moreover, the few functional MRI studies that have looked for this have not established a role for auditory thalamus in auditory hallucinations in schizophrenia, Sweet added.—Allison A. Curley.
Chun S, Westmoreland JJ, Bayazitov IT, Eddins D, Pani AK, Smeyne RJ, Yu J, Blundon JA, Zakharenko SS. Specific disruption of thalamic inputs to the auditory cortex in schizophrenia models. Science 2014;344(6188):1178-1182. Abstract