Schizophrenia Research Forum - A Catalyst for Creative Thinking

Convergence Zone: NRG1 Signaling Linked to DISC1 Expression

16 March 2010. A new study reveals an intriguing link between two top suspects in schizophrenia, neuregulin-1 (NRG1) and disrupted in schizophrenia-1 (DISC1). Published in PNAS online on March 8, the study reports two surprises: NRG1 signaling boosts levels of an isoform of DISC1 protein, and DISC1 itself is found inside multiple types of glial cells in addition to its better known location inside neurons. These findings suggest a web of complex interactions—both between molecules and between different cell types—in the disorder.

The result of a four-way collaboration among the laboratories of Akira Sawa and Philip Wong, both at Johns Hopkins University, Baltimore, Maryland; Eva Anton at the University of North Carolina, Chapel Hill; and Carsten Korth at Heinrich Heine University of Düsseldorf, Germany, the study outlines a chain of events leading from NRG1 to DISC1. In doing so, they also find a bonus link with BACE1 (β-site amyloid precursor protein cleaving enzyme), an Alzheimer disease-related protease that is of interest to schizophrenia researchers because it cleaves NRG1 (see SRF related news story).

Although variations in the genes for NRG1 and for DISC1 are well-known risk factor candidates for schizophrenia, the abnormal behaviors of transgenic mice led first author Saurav Seshadri and colleagues to explore a relationship between the two. Mice deficient in NRG1, DISC1, or BACE1 all have problems with working memory and prepulse inhibition, behavioral abnormalities also found in schizophrenia. These behavioral overlaps suggested that the molecules themselves may be somehow connected—something not too hard to imagine given that NRG1 is a nexus of molecular interaction. When cleaved by BACE1, NRG1 is freed to bind with its receptors, the ErbB receptors. These activated ErbB proteins trigger intracellular signaling cascades, which, among other things, promote cell growth, proliferation, and differentiation—all important elements for building a brain.

The researchers began by examining DISC1 protein expression in vitro. When NRG1 or its cousin NRG2 was applied to immature neurons taken from rat brains and maintained in culture, levels of a 130-kDa isoform of DISC1 went up; treatment with NRG3, by contrast, had no effect. The effect was most pronounced for NRG1, amounting to a 1.77-fold increase in DISC1 over that found in untreated cells. This stemmed mainly from increases in DISC1 in neurites, the immature processes branching out of neurons that ultimately wire them up to other neurons.

Next, the researchers probed which ErbB receptors mediated this effect, using RNAi to systematically knock down levels of ErbB2, ErbB3, or ErbB4 in vitro. Neuron cultures without ErbB2 or ErbB3 could not translate treatment with NRG1 into an increase in the DISC1 isoform, indicating that these receptors were required to boost its expression. NRG1 treatment did not raise the DISC1 isoform's expression when the PI3K/Akt intracellular signaling pathway was blocked with a drug, or when transcription was halted with actinomycin D. Together these findings suggest that NRG1 elevates this DISC1 isoform's expression by binding ErbB2 and ErbB3 receptors, which in turn activate the PI3K/Akt pathway downstream and, somewhere along the line, transcription occurs.

More evidence for this NRG1-DISC1 connection came from mice with low levels of NRG1. NRG1-deficient mice had substantial decreases in the 130-kDa DISC1 isoform: a mouse heterozygous for a null mutation in NRG1 had a 23.6 percent decrease in the isoform compared to wild-type controls, and homozygotes, which do not survive beyond embryonic day 11, also had marked reductions in the isoform relative to controls. When the researchers looked at BACE1 knockouts—mice originally designed to study Alzheimer disease—they confirmed that there was less cleaved NRG1 around to bind ErbB receptors. These animals had a 44.1 percent decrease of the DISC1 isoform in the cortex compared to controls, due mainly to a lack of the protein in neurites.

Because NRG1 signaling occurs both in neurons and glia, the researchers looked for DISC1 in glia. Using an antibody to human DISC1, not only did they find DISC1 protein in neurons of postmortem human brain tissue, they also found it in multiple types of glial cells, as measured by staining for DISC1 that colocalized with markers for astrocytes, oligodendrocytes, and microglia. The same applied for rat cells in culture, and these findings widen the scope of DISC1's possible sites of action in the brain.

Though the researchers were able to home in on the 130-kDa DISC1 isoform using two specific antibodies and RNAi, further studies will have to elucidate the exact nature of this isoform. The study also redirects attention to ErbB2 and ErbB3 receptors in schizophrenia, which have not been as tantalizing as ErbB4 receptors (see Chong et al., 2008). The authors note that ErbB2 lacks an extracellular binding site, and ErbB3 is missing the intracellular tyrosine kinase site necessary for activation. Together in a heterodimer, these molecules may successfully transduce NRG1 binding. Overall, the study demonstrates the potential dividends of looking beyond a pet molecule or cell type, particularly at other disease gene candidates.—Michele Solis.

Seshadri S, Kamiya A, Yokota Y, Prikulis I, Kano SI, Hayashi-Takagi A, Stanco A, Eom TY, Rao S, Ishizuka K, Wong P, Korth C, Anton ES, Sawa A. Disrupted-in-Schizophrenia-1 expression is regulated by beta-site amyloid precursor protein cleaving enzyme-1-neuregulin cascade. Proc Natl Acad Sci USA. 2010 March. Abstract

Comments on News and Primary Papers
Comment by:  Amanda Jayne Law, SRF Advisor
Submitted 19 April 2010
Posted 19 April 2010

The study of Seshadri, Sawa, and colleagues presents novel evidence of a potential biological link between two lead schizophrenia susceptibility genes, NRG1 and DISC1. The principal finding of the study is that NRG1 (EGFβ) regulates expression of a specific isoform of DISC1, mediated via ErbB2/3 but not ErbB4. The influence of NRG1 on expression of the DISC1 isoform was confirmed in a variety of in-vitro and in-vivo models. Specifically, the authors report (using Western blotting with the DISC1 antibodies: D27 and mExon3), that treatment with NRG1 (and NRG2), but not NRG3, increases levels of DISC1 immunoreactivity at 130 kDa in immature and mature rat primary neuron cultures. Interestingly, NRG1 (or NRG2) had no effect on expression of the previously reported full-length DISC1 immunoreactive bands of 100-105 kDa. Convincingly, reduction of the 130 kDa DISC1 band was observed in BACE1 -/- and NRG1 +/- mice, both of which have reduced NRG1 signaling. Taken together, these findings suggest that NRG1 signaling regulates expression of a unique 130 kDa DISC1 protein.

This is an important and thoughtful paper, but there are some details that raise questions about the interpretation of the results. Interestingly, two previous studies that characterized the D27 (and mExon3) antibody in mouse brain (Schurov et al., 2004; Ishizuka et al., 2007) failed to report the 130 kDa band described here. Ishizuka et al. reported that immunoprecipitation with the mExon3 antibody followed by detection with the D27 antibody recognized two primary signals (100 and 105 kDa), thought to correspond to full-length DISC1. In contrast, in the present study the authors report that immunoprecipitation of neuronal lysates using mExon3, followed by Western blotting with D27, consistently identifies an additional 130 kDa band (Fig. S2B), which is also present in the P0 mouse cortex (Fig 3C). Whilst it is not clear what accounts for these apparent differences in signal detection of the 130 kDa band using the same antibodies, factors such as species specificity (rat vs. mouse), tissue type, and developmental stage are likely relevant. Such factors are important considerations for future work. Similarly, it will be crucial to determine whether the 130 kDa band is present in human brain and how it relates to risk for schizophrenia. Of final note, the authors performed extensive experimentation in an attempt to confirm the identity of the 130 kDa band (including successful knockdown by a previously characterized RNAi to DISC1), but interestingly they fail to identify any DISC1 sequence in the 130 kDa signal using mass spectrometry (see discussion). In light of this, it is paramount that future studies determine exactly what the 130 kDa proposed DISC1 band represents (i.e., a novel splice isoform, post-transcriptionally modified protein, etc.), given that NRG1’s effects are specifically related to this variant.

In conclusion, this study provides intriguing evidence of a potential molecular link between NRG1 and DISC1, but at present, the interpretation of the results rests on an immunoblot band of unknown identify.


Schurov IL, Handford EJ, Brandon NJ, Whiting PJ. Expression of disrupted in schizophrenia 1 (DISC1) protein in the adult and developing mouse brain indicates its role in neurodevelopment. Mol Psychiatry . 2004 Dec 1 ; 9(12):1100-10. Abstract

Ishizuka K, Chen J, Taya S, Li W, Millar JK, Xu Y, Clapcote SJ, Hookway C, Morita M, Kamiya A, Tomoda T, Lipska BK, Roder JC, Pletnikov M, Porteous D, Silva AJ, Cannon TD, Kaibuchi K, Brandon NJ, Weinberger DR, Sawa A. Evidence that many of the DISC1 isoforms in C57BL/6J mice are also expressed in 129S6/SvEv mice. Mol Psychiatry . 2007 Oct ; 12(10):897-9. Abstract

View all comments by Amanda Jayne LawComment by:  Alexander Arguello
Submitted 3 May 2010
Posted 3 May 2010

This paper raises an interesting issue. It is unclear how an immuno band that has no DISC1 sequences can result from "alternative splicing or post-translational modification." Could someone provide a mechanistic account, at the molecular level, of how this may be possible? To support that this band is DISC1, at least some DISC1 sequence should have been detected. This issue could be related to the non-specific cross-reactivity of many DISC1 antibodies (see Kvajo et al., 2008 for a discussion) and now also raises the possibility of off-target effects of DISC1 RNAi.

Resolving these issues will be paramount for making meaningful insights into how variations in DISC1 contribute to psychotic disorders.


Kvajo M, McKellar H, Arguello PA, Drew LJ, Moore H, MacDermott AB, Karayiorgou M, Gogos JA. A mutation in mouse Disc1 that models a schizophrenia risk allele leads to specific alterations in neuronal architecture and cognition. Proc Natl Acad Sci U S A. 2008 May 13;105(19):7076-81. Abstract

View all comments by Alexander ArguelloComment by:  Saurav SeshadriAtsushi KamiyaEva AntonAkira Sawa (SRF Advisor)
Submitted 4 May 2010
Posted 4 May 2010

We are very glad to see Dr. Law’s thoughtful and very supportive comments on the work by Seshadri et al. We share the recognition, as we pointed out in the discussion of the paper, that identification of 130 kDa signal at the molecular level is an important future question. To confirm the authenticity of immunoreactivity, we tested if the 130 kDa signal is immunoprecipitated and immunoblotted by different DISC1 antibodies. Similar immunoreactive approaches have been used earlier to distinguish DISC1 isoforms, including a 71 kDa isoform in association with PDE4 (Millar et al., 2005; Chubb et al., 2008). Knockout mice deficient in DISC1 that we have recently generated (unpublished) were used for evaluating the specificity of several antibodies against DISC1 (Schurov et al., 2004; Ishizuka et al., 2007; Duan et al., 2007; Koike et al., 2006). Loss of this immunoreactivity by authentic shRNAs further supports this idea. The sequences of shRNAs are the same as those used in the study by Mao et al. (Mao et al., 2009) to demonstrate that DISC1 may be involved in progenitor cell proliferation.

Of note, mass spectrometry cannot be an ultimate confirmation, because with this technique it is hard to distinguish the signals from two adjacent or overlapped bands in Western blots of 1D gels, one of which is real and the other not. Therefore, regardless of our initial mass spectrometry analysis (even if one finds sequences of the target protein), validation with both immunoprecipitation and RNAi is required to draw a conclusion on the identity of 130 kDa signal. In the study by Seshadri et al., these two ways of validation were successfully made.

Furthermore, whether or not this 130 kDa isoform is also expressed in humans is a critical question. It is also very important to consider context-dependent expression of unique isoforms of genetic susceptibility factors. This unique form (130 kDa) is likely to be in that category; thus, as Dr. Law suggested, comparative analysis is very useful. Further analysis of the genesis, function, and processing of various DISC1 isoforms in the brain will be a worthy pursuit in the context of schizophrenia.


Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR, Malloy MP, Chubb JE, Huston E, Baillie GS, Thomson PA, Hill EV, Brandon NJ, Rain JC, Camargo LM, Whiting PJ, Houslay MD, Blackwood DH, Muir WJ, Porteous DJ. DISC1 and PDE4B are interacting genetic factors in schizophrenia that regulate cAMP signaling. Science. 2005 Nov 18 ; 310(5751):1187-91. Abstract

Chubb JE, Bradshaw NJ, Soares DC, Porteous DJ, Millar JK. The DISC locus in psychiatric illness. Mol Psychiatry. 2008 Jan 1 ; 13(1):36-64. Abstract

Schurov IL, Handford EJ, Brandon NJ, Whiting PJ. Expression of disrupted in schizophrenia 1 (DISC1) protein in the adult and developing mouse brain indicates its role in neurodevelopment. Mol Psychiatry. 2004 Dec 1 ; 9(12):1100-10. Abstract

Ishizuka K, Chen J, Taya S, Li W, Millar JK, Xu Y, Clapcote SJ, Hookway C, Morita M, Kamiya A, Tomoda T, Lipska BK, Roder JC, Pletnikov M, Porteous D, Silva AJ, Cannon TD, Kaibuchi K, Brandon NJ, Weinberger DR, Sawa A. Evidence that many of the DISC1 isoforms in C57BL/6J mice are also expressed in 129S6/SvEv mice. Mol Psychiatry. 2007 Oct 1 ; 12(10):897-9. Abstract

Duan X, Chang JH, Ge S, Faulkner RL, Kim JY, Kitabatake Y, Liu XB, Yang CH, Jordan JD, Ma DK, Liu CY, Ganesan S, Cheng HJ, Ming GL, Lu B, Song H. Disrupted-In-Schizophrenia 1 regulates integration of newly generated neurons in the adult brain. Cell. 2007 Sep 21 ; 130(6):1146-58. Abstract

Koike H, Arguello PA, Kvajo M, Karayiorgou M, Gogos JA. Disc1 is mutated in the 129S6/SvEv strain and modulates working memory in mice. Proc Natl Acad Sci U S A. 2006 Mar 7 ; 103(10):3693-7. Abstract

Mao Y, Ge X, Frank CL, Madison JM, Koehler AN, Doud MK, Tassa C, Berry EM, Soda T, Singh KK, Biechele T, Petryshen TL, Moon RT, Haggarty SJ, Tsai LH. Disrupted in schizophrenia 1 regulates neuronal progenitor proliferation via modulation of GSK3beta/beta-catenin signaling. Cell. 2009 Mar 20 ; 136(6):1017-31. Abstract

View all comments by Saurav Seshadri
View all comments by Atsushi Kamiya
View all comments by Eva Anton
View all comments by Akira Sawa

Comments on Related News

Related News: Down to BACE-ics—Old Mouse a New Model for Schizophrenia?

Comment by:  Victor ChongCynthia Shannon Weickert (SRF Advisor)
Submitted 23 May 2008
Posted 23 May 2008

The findings of Savonenko et al. (2008) are an impressive addition to the growing evidence supporting a role for neuregulin-1 (NRG1) in schizophrenia pathology. The authors not only revealed a novel relationship between schizophrenia-like behavior and the loss of BACE1 proteolytic function, but also showed that this association results from disruption of BACE1-mediated NRG1 cleavage. These observations support the notion that aberrant processing of NRG1 may contribute to the development of schizophrenia-like phenotypes, providing a basis for examining other NRG1-cleaving pathways in the context of schizophrenia. Savonenko et al. were thorough in their behavioral assessment of the BACE1 mutant mice, convincingly showing that these animals exhibit schizophrenia-related behaviors that could be exacerbated by psychostimulants and improved by antipsychotic drug treatment.

What remains unclear, however, is the relationship between the NRG1/ErbB4 protein findings in the BACE1 mutant mouse brain and those previously reported in the schizophrenic human brain. For example, the authors reported reductions in ErbB4-PSD95 coupling in the BACE1 mutant mouse, whereas Hahn et al. (2006) demonstrated increased ErbB4-PSD95 interaction in the prefrontal cortices of schizophrenic patients. In addition, our recent investigation found elevated prefrontal cortical levels of both NRG1 C-terminal fragment (ICD) and full-length ErbB4 protein in schizophrenic subjects (Chong et al., 2008), while Savonenko et al. showed decreased NRG1 C-terminal fragment levels with no alterations in ErbB4 protein in the BACE1 mutant mouse cortex. On the other hand, the lack of variations in overall cortical ErbB4 in these mice may correspond to the findings of Hahn et al. (2006) who reported no alterations in prefrontal cortical ErbB4 protein levels in schizophrenic subjects.

These seemingly conflicting results could suggest that any imbalance in cortical NRG1 signaling, whether increased or diminished, may lead to schizophrenia. Indeed, studies have suggested that improper tuning of other cortical signaling systems, particularly those of dopamine, can contribute to cognitive deficits associated with this disease (Vijayraghavan et. al, 2007). Optimal synaptic function may display “inverted-U” shaped response to NRG1-ErbB4 activity as proposed by Role and Talmage (2007). Alternatively, the authors speculated that some of the discrepancies between the findings in the BACE1 mutant mice and those observed in the schizophrenic humans may be due to differences in the duration of NRG1 signaling modification between the animals and the patients, who had a lifetime of mental illness. One way to examine the validity of this suggestion is to look at cortical ErbB4-PSD95 coupling and NRG1/ErbB4 protein levels in the BACE1 mutant mice at different developmental and adult time points. This approach could test whether these animals at later stages in life display alterations in cortical ErbB4-PSD95 interactions and/or in NRG1/ErbB4 protein levels comparable to those seen in schizophrenic subjects of the human studies, which primarily consisted of adults beyond middle age. Also of interest would be to create NRG1 and ErbB4 gain-of-function mutants where the timing of over-expression could be controlled.

Given the significance of NRG1 signaling/cleavage in the BACE1 mutant mouse schizophrenia-like phenotypes, it may also be important to consider pathways leading to changes in ErbB4 C-terminal fragment levels in schizophrenia etiology. A recent paper by Walsh et al. (2008) demonstrated that at least one schizophrenic patient in their study has a gene deletion encompassing the C-terminal intracellular kinase domain of ErbB4, and we have found decreases in ErbB4 C-terminal fragments relative to full-length ErbB4 in the frontal cortex of schizophrenic subjects (Chong et al., 2008). These observations together with those of Savonenko et al. raise interesting questions regarding how molecular alterations in NRG1 signaling and cleavage may impact ErbB4 signaling and cleavage and whether changes in NRG1 and/or ErbB4 could be primary or secondary to the schizophrenia disease process.

In summary, Savonenko et al. have provided a novel avenue to probe NRG1 function and processing in relation to schizophrenia pathology. They have also introduced BACE1 as a potentially important schizophrenia susceptibility molecule that to our knowledge has not been directly investigated in subjects with schizophrenia and may be worth studying in the brain tissues of these patients. In addition, it would be interesting to examine how the schizophrenia-related traits of the BACE1 mutant mice compare with those of other NRG1 mutant mice such as the heterozygous NRG1 transmembrane knock-out mice (Stefansson et al., 2002). Such an investigation could provide insight into whether similar NRG1 signaling deficiencies underlie the schizophrenia-like phenotypes of these animal models.


Hahn CG, Wang HY, Cho DS, Talbot K, Gur RE, Berrettini WH, Bakshi K, Kamins J, Borgmann-Winter KE, Siegel SJ, Gallop RJ, Arnold SE. (2006) Altered neuregulin 1-erbB4 signaling contributes to NMDA receptor hypofunction in schizophrenia. Nat Med. 12:824-8. Abstract

Chong VZ, Thompson M, Beltaifa S, Webster MJ, Law AJ, Weickert CS. (2008) Elevated neuregulin-1 and ErbB4 protein in the prefrontal cortex of schizophrenic patients. Schizophr Res. 100:270-80. Abstract

Vijayraghavan S, Wang M, Birnbaum SG, Williams GV, Arnsten AF. (2007) Inverted-U dopamine D1 receptor actions on prefrontal neurons engaged in working memory. Nat Neurosci. 10:376-84. Abstract

Role LW, Talmage DA (2007) Neurobiology: new order for thought disorders. Nature. 448:263-5. Abstract

Walsh T, McClellan JM, McCarthy SE, Addington AM, Pierce SB, Cooper GM, Nord AS, Kusenda M, Malhotra D, Bhandari A, Stray SM, Rippey CF, Roccanova P, Makarov V, Lakshmi B, Findling RL, Sikich L, Stromberg T, Merriman B, Gogtay N, Butler P, Eckstrand K, Noory L, Gochman P, Long R, Chen Z, Davis S, Baker C, Eichler EE, Meltzer PS, Nelson SF, Singleton AB, Lee MK, Rapoport JL, King MC, Sebat J. (2008) Rare structural variants disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science. 320:539-43. Abstract

Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, Sigmundsson T, Ghosh S, Brynjolfsson J, Gunnarsdottir S, Ivarsson O, Chou TT, Hjaltason O, Birgisdottir B, Jonsson H, Gudnadottir VG, Gudmundsdottir E, Bjornsson A, Ingvarsson B, Ingason A, Sigfusson S, Hardardottir H, Harvey RP, Lai D, Zhou M, Brunner D, Mutel V, Gonzalo A, Lemke G, Sainz J, Johannesson G, Andresson T, Gudbjartsson D, Manolescu A, Frigge ML, Gurney ME, Kong A, Gulcher JR, Petursson H, Stefansson K. (2002) Neuregulin 1 and susceptibility to schizophrenia. Am J Hum Genet. 71:877-92. Abstract

View all comments by Victor Chong
View all comments by Cynthia Shannon Weickert