Schizophrenia Research Forum - A Catalyst for Creative Thinking

ErbB4 Deletion Models Aspects of Schizophrenia

October 16, 2013. A conditional knockout of ErbB4 from parvalbumin-expressing interneurons reproduces some features of schizophrenia, according to a new study published September 18, 2013 in Neuron. Led by Oscar Marín and Beatriz Rico of CSIC-University Miguel Hernández in Alicante, Spain, the study finds that the deletion produces synaptic deficits, increases cortical excitability and oscillatory activity, and results in impaired social behavior and cognitive function.

The most surprising aspect of the study, said Lin Mei of Georgia Regents University in Augusta, was that “The behavioral consequences and the electrophysiological disruptions seem much more dramatic than the actual amount of synapse reduction [found] in morphological analysis.” He added that the findings suggest that a “subtle change of GABAergic function during development is very damaging.” Mei was not involved in the study.

The tyrosine kinase receptor ErbB4 came to the interest of schizophrenia researchers because its ligand neuregulin 1 (Nrg1) is on the "traditional" short list of schizophrenia susceptibility genes, determined to have strong evidence in the era before genomewide association studies (see SRF related news story). ErbB4 is highly expressed in the parvalbumin (PV) subclass of GABAergic interneurons, and plays a critical role in the development and maturation of GABA circuitry (see SRF related news story). Research showing that PV cells are crucial for γ oscillatory activity and cognition, and the hypothesis that alterations in these cells underlie some of the cognitive deficits of schizophrenia, has led to the suggestion that Nrg1-ErbB4 signaling may be at least partially to blame for some cognitive symptoms (Rico and Marín, 2011).

However, postmortem studies of the Nrg1-ErbB4 pathway have thus far been confusing – both increased and decreased levels of ErbB4 and Nrg1 signaling been reported in the illness. Perhaps consistent with these findings, both over- and under-expression of the NRG-ErbB4 pathway in mice mimic features of schizophrenia (see SRF related news story). This suggests that any imbalance, either up or down, is detrimental, and has led Mei and others to hypothesize that the pathway may be increased in a subset of patients but decreased in others.

Subtle synaptic defects
In the current study, first author Isabel del Pino and colleagues followed the loss-of-function avenue. They generated conditional ErbB4 mutant mice with lower levels of the receptor in PV cells expressing the transcription factor Lhx6. Using retroviral labeling, the researchers performed separate analyses of the two populations of PV neurons: chandelier and basket cells. Consistent with prior studies, they observed fewer excitatory axon terminals (as measured by VGlut1 immunoreactivity) contacting the soma and dendrites of both ErbB4-deficient PV chandelier and basket cells in the cortex. Corresponding decreases were observed in PV cell clusters expressing the postsynaptic marker PSD95, as well as in the frequency of hippocampal miniature excitatory postsynaptic currents (EPSCs) recorded from these cells, providing further evidence that ErbB4 is required in order to establish a normal complement of excitatory synapses onto PV cells.

The researchers next turned their attention to the inhibitory synapses made by PV cells synapsing onto pyramidal cells. Chandelier cells derive their name from the distinctive shape of their axons (which form vertical arrays that resemble candlesticks) and synapse exclusively onto the axon initial segment (AIS) of pyramidal neurons. Similar to postmortem findings from schizophrenia (Woo et al., 1998), the density of candlestick presynaptic terminals was lower in ErbB4 mutants than controls. There were also fewer α2 subunit-containing GABAA receptor clusters found postsynaptically at the AIS in the ErbB4 mutants, although curiously this finding is opposite of what has been observed in the illness (Volk et al., 2002). ErbB4 function appeared to only be required for the development and maintenance of chandelier cell inhibitory synapses, as there were largely no alterations in PV basket cell to pyramidal cell synapses in the ErbB4 mutants.

Looking at both populations of PV cells combined, the researchers found that total cortical protein levels of GAD67 and PV were reduced in the knockouts, with no change in GAD65, similar to postmortem schizophrenia findings (see SRF related news story). Also reminiscent of the illness, the number but not the length of dendritic spines was reduced in hippocampal pyramidal cells.

Big functional consequences
To examine the effect of fewer synapses in the ErbB4 mutants on hippocampal network activity, del Pino and colleagues recorded spontaneous EPSCs in pyramidal cells. Consistent with the molecular alterations indicating a loss of excitatory synapses on PV cells as well as deficits in inhibitory signaling, pyramidal cells received more excitatory drive in ErbB4 mutants than in controls (as evidenced by an increase in spontaneous EPSCs). Spontaneous EPSC frequency was also elevated in PV cells, suggesting that these inhibitory interneurons also received enhanced excitatory drive.

Interestingly, despite the fact that the conditional knockdown of ErbB4 in PV neurons produced subtle defects in connectivity (as evidenced by the lack of change in the tangential migration, organization, and density of PV neurons), major changes in hippocampal network activity were observed. In vivo local field potential recordings in anesthetized mice revealed an increase in resting-state oscillatory power in the dentate gyrus and CA1 region in the ErbB4 mutants, as well as increases in CA1 θ, α, β, and γ band power. In contrast to the large perturbations found in the hippocampus, local field potential changes in the cortex were less dramatic, with an increase in γ band power found in the infralimbic cortex. In addition to these local network changes, the researchers also found a reduction in the long-range synchrony between the hippocampus and prefrontal cortex, as well as reduced hippocampal-prefrontal coherence in the θ band.

The most impressive experiments in the paper, said Andres Buonanno, of the National Institute of Child Health & Human Development in Bethesda, Maryland, were the examination of hippocampal rhythms in awake, freely-moving mice exploring an open field. Consistent with the data from anesthetized animals, the researchers found that CA1 local field potentials were increased in the ErbB4 mutants in a more physiologically relevant setting, as was the power of γ oscillations.

Consistent with other mouse of models of restricted ErbB4 expression, the current model also produced a variety of behavioral alterations: increased locomotor activity, evidence of reduced anxiety, impaired sociability, and deficits in sensorimotor gating (Shamir et al., 2012).

Using the Y maze spontaneous alternation test, the researchers also reported deficits in working memory. This finding could appear at odds with data demonstrating that γ oscillations are a neural substrate underlying working memory and other cognitive processes. “If the knockouts have more γ power, why do they perform less well on the Y maze?” commented Buonanno.

These data are also at odds with what has been observed in schizophrenia, Buonanno said. Although the literature on γ oscillations in the illness is complex, in general, most studies have reported that reduced γ power is associated with cognitive deficits (see SRF related news story). The data may not align perfectly with schizophrenia, but the current findings nevertheless point to a connection between the Nrg1-ErbB4 signaling pathway, GABAergic circuitry assembly and function, and the cognitive symptoms of the illness.—Allison A. Curley

Del Pino I, García-Frigola C, Dehorter N, Brotons-Mas JR, Alvarez-Salvado E, Martínez de Lagrán M, Ciceri G, Gabaldón MV, Moratal D, Dierssen M, Canals S, Marín O, Rico B. Erbb4 Deletion from Fast-Spiking Interneurons Causes Schizophrenia-like Phenotypes. Neuron. 2013 Sep 18. Abstract

Comments on News and Primary Papers
Comment by:  Beatriz RicoOscar Marin
Submitted 30 October 2013
Posted 5 November 2013

We would like to provide an answer to the question raised by Andrés Buonanno: “If the knockouts have more γ power, why do they perform less well on the Y maze?” As explained in the manuscript, the abnormal increase in γ power observed in conditional ErbB4 mutants would not necessarily lead to better performance, because interneurons are not pacing pyramidal cells at the proper/normal rhythm. In addition, local hypersynchrony seems to affect long-range functional connectivity: We showed a prominent decoupling between the hippocampus and prefrontal cortex. The increase in excitability and synchrony, and the decoupling between the hippocampus and prefrontal cortex, are likely the cause of the behavioral deficits in cognitive function.

In line with this, we respectfully disagree with Buonanno's next comment that “these data are also at odds with what has been observed in schizophrenia.” Indeed, as we mentioned in the manuscript, recent studies indicate that medication-naive, first-episode, and chronic patients with schizophrenia show elevated γ-band power in resting state. Baseline increases in γ oscillations are consistent with increases in the excitatory/inhibitory ratio of cortical neurons. Thus, cortical rhythm abnormalities in schizophrenia seem to include both abnormal increases in baseline power—as we observed in conditional ErbB4 mutants—as well as deficits in task-related oscillations (Uhlhaas and Singer, 2012).


Uhlhaas PJ, and Singer W. (2012). Neuronal dynamics and neuropsychiatric disorders: toward a translational paradigm for dysfunctional large-scale net- works. Neuron 75, 963–980. Abstract

View all comments by Beatriz Rico
View all comments by Oscar Marin

Comments on Related Papers

Related Paper: Reciprocal alterations in pre- and postsynaptic inhibitory markers at chandelier cell inputs to pyramidal neurons in schizophrenia.

Comment by:  David Lewis, SRF Advisor
Submitted 27 October 2005
Posted 27 October 2005

The authors demonstrated that the presynaptic reduction of the GABA transporter in chandelier axon terminals was associated with an upregulation of postsynaptic α2-containing GABAA receptors in the axon initial segments of pyramidal neurons in subjects with schizophrenia.

View all comments by David Lewis
Comments on Related News

Related News: Asynchrony and the Brain—Gamma Deficits Linked to Poor Cognitive Control

Comment by:  Richard Deth
Submitted 14 December 2006
Posted 15 December 2006

Schizophrenia is associated with dopaminergic dysfunction, impaired gamma synchronization and impaired methylation. It is therefore of interest that the D4 dopamine receptor is involved in gamma synchronization (Demiralp et al., 2006) and that the D4 dopamine receptor uniquely carries out methylation of membrane phospholipids (Sharma et al., 1999). A reasonable and unifying hypothesis would be that schizophrenia results from a failure of methylation to adequately support dopamine-stimulated phospholipid methylation, leading to impaired gamma synchronization. Synchronization in response to dopamine can provide a molecular mechanism for attention, as information in participating neural networks is able to bind together to create cognitive experience involving multiple brain regions.

View all comments by Richard Deth

Related News: Asynchrony and the Brain—Gamma Deficits Linked to Poor Cognitive Control

Comment by:  Fred Sabb
Submitted 12 January 2007
Posted 12 January 2007
  I recommend the Primary Papers

Cho and colleagues find patients with schizophrenia showed a reduction in induced gamma band activity in the dorsolateral prefrontal cortex compared to healthy control subjects during a behavioral task that is known to challenge cognitive control processes. Importantly, the induced gamma band activity was correlated with better performance in healthy subjects, and negatively correlated with higher disorganization symptoms in patients with schizophrenia. These findings help explain previous post-mortem evidence of disruptions in thalamofrontocortical circuits in these patients.

These findings tie together several different previously identified phenotypes into a unifying story. The ability to link phenotypes across translational research domains is paramount to understanding complex neuropsychiatric diseases like schizophrenia. Cho and colleagues provide an excellent example for connecting evidence from symptom rating scales with behavioral, neural systems and neurophysiological data. Although not specifically addressed by the authors, these data may have important implications for understanding the neural basis of thought disorder as well. Hopefully, these findings will provide a frame-work for examining more informed and specific phenotypes relevant to schizophrenia.

View all comments by Fred Sabb

Related News: Research Roundup: PV Interneurons and Neural Circuit (Dys)Function

Comment by:  Takao Hensch
Submitted 15 February 2014
Posted 15 February 2014

Our Conte Center is focused on the transcriptome, connectome, and plasticity of PV cells as the neurodevelopmental basis for mental illness. Their maturational state dictates the degree of plasticity in developmental critical periods, and now we know, from Donato et al., in adult learning. Once plasticity is opened by PV cells' function, it closes when they mature ("high PV" state), including the tightening of the perineuronal nets (PNNs) around them.

In schizophrenia, PV cells may remain in the "low PV" weak PNN state for some time longer than normal, suggesting, interestingly, that developmental plasticity may be prolonged (i.e., neural circuits fail to stabilize when they normally should). PV cell maturation may potentially be controlled by Otx2 secreted from the choroid plexus, which would link enlarged ventricles to impaired PV cells in the brain in schizophrenia (see Spatazza et al., 2013).


Spatazza J, Lee HH, Di Nardo AA, Tibaldi L, Joliot A, Hensch TK, Prochiantz A. Choroid-plexus-derived Otx2 homeoprotein constrains adult cortical plasticity. Cell Rep. 2013 Jun 27;3(6):1815-23. Abstract

View all comments by Takao Hensch