The effects of a neuregulin 1 variant on white matter density and integrity

Article metrics


Theories of abnormal anatomical and functional connectivity in schizophrenia and bipolar disorder are supported by evidence from functional magnetic resonance imaging (MRI), structural MRI and diffusion tensor imaging (DTI). The presence of similar abnormalities in unaffected relatives suggests such disconnectivity is genetically mediated, albeit through unspecified loci. Neuregulin 1 (NRG1) is a psychosis susceptibility gene with effects on neuronal migration, axon guidance and myelination that could potentially explain these findings. In the current study, unaffected subjects were genotyped at the NRG1 single nucleotide polymorphism (SNP) rs6994992 (SNP8NRG243177) locus, previously associated with increased risk for psychosis, and the effect of genetic variation at this locus on white matter density (T1-weighted MRI) and integrity (DTI) was ascertained. Subjects with the risk-associated TT genotype had reduced white matter density in the anterior limb of the internal capsule and evidence of reduced structural connectivity in the same region using DTI. We therefore provide the first imaging evidence that genetic variation in NRG1 is associated with reduced white matter density and integrity in human subjects. This finding is discussed in the context of NRG1 effects on neuronal migration, axon guidance and myelination.


Schizophrenia and bipolar disorder are severe mental disorders that affect approximately 2% of the population worldwide and, in spite of ameliorative treatment, show a tendency to become recurrent or persistent. Originally described as separate syndromes, it is generally recognized that there is no symptom, treatment or outcome specific to either disorder.1 This suggests that there are commonalities in the pathophysiology of schizophrenia and bipolar disorder that transcend conventional diagnostic boundaries.

There is strong evidence of white matter abnormality in both schizophrenia and bipolar disorder. Patients with both conditions show white matter density reductions in the anterior limb of the internal capsule (ALIC) and in prefrontal subgyral white matter using T1-weighted magnetic resonance imaging (MRI).2, 3 These abnormalities are shared by unaffected relatives in proportion to their proximity to an affected subject.4 These findings suggest that white matter density reductions are likely to be associated with shared genetic factors, although no specific genetic variants have yet been found which account for these findings.

Diffusion tensor MRI (DTI) is an imaging technique that is able to index the integrity of axonal connections more specifically by measuring the extent to which water motion is constrained. Unlike T1-weighted MRI, it is theoretically less susceptible to the signal generated by non-neuronal tissue components and evidence from studies of primary demyelinating conditions suggests that it is both valid and a more sensitive technique than conventional MRI.5

DTI studies of patients with schizophrenia and bipolar disorder have generally confirmed the results from T1-weighted MRI.6 Deficits in the ALIC and frontal white matter have been confirmed in both schizophrenia and bipolar disorder by studying patients within the same study,7 supporting the idea that these deficits may be related to the shared pathophysiology of both conditions. These deficits are also found in individuals at high risk of schizophrenia for genetic reasons,8 suggesting that they are likely to be associated with genetic factors potentially related to the aetiology of psychosis.

Schizophrenia and bipolar disorder are highly familial disorders with several genes implicated in their aetiology. Evidence from family studies suggests a degree of familial co-aggregation and linkage studies have identified several chromosomal regions common to both disorders.9 These findings suggest the presence of susceptibility genes that increase the risk of both disorders.

Neuregulin 1 (NRG1) is a gene located on chromosome 8p that was first associated with schizophrenia in the Icelandic population.10 The majority of subsequent association studies, including two in Scottish populations,11, 12 have shown association of NRG1 with both schizophrenia and bipolar disorder.13 While there has been heterogeneity between studies, recent meta-analysis has confirmed association of the original Icelandic haplotype (HapICE) with schizophrenia.14 NRG1 is a complex gene with several potential transcripts classified by the 5′ terminal exon. Generally these can be grouped into two types: those that contain an immunoglobulin (Ig) moiety (Ig-NRG, including type I, II and IV) and those that have a cysteine-rich domain (CRD-NRG, including type III). Ig-NRG activates the ErbB receptors, in particular ErbB4, by diffusion whereas CRD-NRG is expressed on the neuronal membrane and relies on juxtacrine signalling to activate ErbB.

Recent evidence has focussed on a particular polymorphism, rs6994992 (SNP8NRG243177) in the type-IV promoter region of NRG1. This single nucleotide polymorphisms (SNP) is part of the original risk-associated HapICE and is associated with altered transcription factor binding, including myelin transcription factor-1, and increased levels of type-IV NRG1 mRNA in post-mortem tissue.15 Furthermore, genetic variation at rs6994992 (SNP8NRG243177) has been shown to alter type IV promoter activity in in vitro receptor assays.16 We have shown that SNP8NRG243177 is associated with altered frontotemporal brain function and psychotic symptoms in individuals at high risk of schizophrenia for genetic reasons17 and others have also found an association with cognitive function.18 Disconnectivity between these regions has been proposed to account for the symptoms of schizophrenia.19 These convergent lines of evidence suggest that SNP rs6994992 (SNP8NRG243177) exerts a functional effect through alterations in NRG1 expression.

Recent evidence suggests that NRG1 risk haplotypes may affect the trajectory of white matter development in childhood-onset schizophrenia.20 Given NRG1's role in axonal guidance21 and myelination,22 and the presence of apparently genetically determined white matter abnormalities in psychosis, we sought to test whether NRG1 SNP rs6994992 (SNP8NRG243177) has effects on white matter density and integrity. Unmedicated controls with no personal or family history of psychosis were recruited from the established Scottish population. White matter density and integrity were investigated using T1-weighted MRI and DTI, respectively.

Materials and methods

Participants and genotyping

Eighty-seven screened healthy control subjects with no personal or close family history of mental illness were identified from our previous studies. All subjects underwent T1-weighted imaging and 43 subjects provided additional DTI data. A lifetime absence of mental disorder was confirmed at interview with a psychiatrist. SNP rs6994992 (SNP8NRG243177) genotype was determined from DNA extracted from whole blood using PCR (TaqMan, Assay-by-Design). The study was approved by the local research ethics committee and all participants provided written informed consent.

T1-weighted white matter imaging

All participants were scanned on a 1.5 T GE MRI scanner. Midline sagittal localization was followed by two further sequences to image the whole brain. The first sequence was a transverse spin-echo scan which acquired both T2- and proton-density-weighted images of the brain. These images were subsequently clinically reported by a consultant in neuroradiology. The third sequence was a coronal gradient echo sequence with magnetization preparation that produced 128 coronal high-resolution T1-weighted images that were used for structural image analysis (TI=600 ms, TE=3.4 ms, flip angle=15°, field-of-view, FOV=22, slice thickness=1.7 mm, matrix=256 × 192).

In an adaptation of the optimized methodology proposed by Good et al.,23 we used non-linear warping of extracted (skull-stripped) brains to an extracted study-specific template.24 All scans were segmented in normalized space using study-specific a priori tissue maps.23, 25 White matter images were then smoothed using a 12 mm Gaussian kernel prior to statistical analysis. Subjects homozygous for the T allele were compared in two contrasts to (1) individuals homozygous for the C allele (CC genotype) and (2) subjects carrying the C allele (CC plus CT genotypes). Analyses were conducted using a one-way analysis of variance (ANOVA) followed by a contrast of parameter estimates as outlined above. Consistent with our previous studies,2, 4 the T1 white matter analysis was restricted to frontal subgyral white matter and the ALIC using a small volume correction (SVC) image, constructed by tracing an image constructed from the average of all white matter segments. The extent and location of between-group differences were illustrated using statistical maps, thresholded at a significance level of P<0.001 uncorrected. All analyses were conducted in SPM99. The results presented are significant at P<0.05 (corrected).

Diffusion tensor white matter imaging

Using a single-shot pulsed gradient spin-echo echo-planar (EP) imaging sequence, whole-brain DTI data were generated from sets of axial EP images (b=0 and 1000 s mm−2) with diffusion gradients applied sequentially along 51 non-collinear directions arranged uniformly in space. Forty-eight contiguous axial slice locations were imaged with (FOV=220 × 220 mm, slice thickness=2.8 mm, matrix=96 × 96 (zero filled to 128 × 128), TR=17 s, TE=93.4 ms). In addition to the 51 diffusion-weighted EPI volumes, three T2-weighted baseline echoplanar (EPI) volumes were also acquired.

All DICOM format magnitude images were transferred from the scanner to a Sun Blade 2000 workstation (Sun Microsystems, Mountain View, CA, USA) and converted into Analyze format (Mayo Foundation, Rochester, MN, USA). Using the FLIRT toolbox (, a three-dimensional computational image alignment program, bulk patient motion and eddy current induced artefacts were removed from the DTI data by registering all EPI volumes to the first T2-weighted EPI volume. The apparent diffusion tensor of water (D) was calculated in each voxel from the signal intensities in the component EP images. Maps of fractional anisotropy (FA) for every subject were generated on a voxel-by-voxel basis from the sorted eigenvalues of D and converted into Analyze format, resulting in a series of skull-stripped FA and averaged T2-weighted volumes for further analysis.

Voxel-based analysis was conducted using SPM software. The T2-weighted images from each subject were normalized to the skull-stripped MNI template using optimized segmentation and normalization parameters. Normalization of the FA maps was achieved by applying the same transformation parameters (38). Both normalized T2-weighted images and FA maps were smoothed at 12 mm full-width at half-maximum (39).

FA was compared between groups, defined on the basis of rs6994992 (SNP8NRG243177) status, using a one-way ANOVA followed by a contrast of parameter estimates as outlined above. Statistical parametric maps were thresholded at P=0.01 (uncorrected) using a white matter mask and an SVC was then applied to restrict the search for between-group differences to the regions identified by the white matter T1-weighted image analysis. Differences between groups were considered to be significant when the corrected P-value was less than 0.05 for search volumes of 1 resel or more, or P<0.05 uncorrected for search volumes of less than 1 resel.


T1-weighted imaging

The number of subjects in each genotype group did not significantly depart from Hardy–Weinberg equilibrium (TT=14, CT=32, CC=41) and there were no significant differences between the groups in terms of age (mean age (s.d.): TT=30.9 (9.0), CT=30.5 (10.0), CC=34.8 (12.0)) or the proportion of male subjects (proportion male: TT=11/14, CT=16/32, CC=18/41).

Individuals homozygous for the C allele had higher white matter density in the tip of the right anterior internal capsule compared to T homozygotes (Figures 1 and 2, peak 1: MNI coordinates=(28, 28, 10), t=4.07, P-corrected=0.041; peak 2: MNI coordinates=(28, 24, 12), t=4.01, P-corrected=0.049). Subjects with the TT genotype also had significantly reduced white matter density as assessed by voxel-based morphometry compared to the C allele carriers at the tip of the right anterior capsule (peak 1: MNI coordinates=(29, 30, 7), t=4.41, P-corrected=0.028; peak 2: MNI coordinates=(29, 26, 8), t=4.38, P-corrected=0.016; peak 3: MNI coordinates=(29, 23, 12), t=4.26, P-corrected=0.023). The associated cluster extended both medially and laterally into frontal white matter (Figure 1). Reductions in ALIC white matter density were also found in the left anterior internal capsule, although these did not reach statistical significance.

Figure 1

Figure shows regions of reduced white matter density in individuals homozygous for the T allele compared to carriers of the C allele.

Figure 2

Figure shows mean white matter density (WMD) estimates at voxels of significant difference (shown on x axis) in T1-weighted MRI analysis. Error bars represent ±1 s.e.

Diffusion tensor imaging

The number of subjects from each genotype group for whom DTI data was available did not significantly depart from Hardy–Weinberg equilibrium (TT=6, CT=22, CC=15). There were no significant differences between the groups in terms of age (age: mean (s.d.): TT=35.8 (11.1), CT=35.5 (10.9), CC=37.1 (12.3)), or gender (proportion male: TT=5/6, CT=8/22, CC=10/15).

Based on the results above, we restricted the search volume to the bilateral ALIC. Subjects homozygous for the T allele showed reduced FA in the right ALIC compared to C allele homozygotes (Figures 3 and 4, peak MNI coordinates=(22, 14, 8), t=2.65, P=0.006, search volume=0.7 resels; Figure 2). Reductions were also found in the genu of the left ALIC and in cerebellar white matter, although these did not reach statistical significance.

Figure 3

Figure shows regions of reduced FA in individuals homozygous for the T allele compared to carriers of the C allele.

Figure 4

Figure shows mean fractional anisotropy (FA) estimates at peak voxel of significant difference. Error bars represent ±1 s.e.


Here we provide the first evidence that the risk allele at the NRG1 variant rs6994992 (SNP8NRG243177) is associated with reduced white matter density and integrity in the ALIC in otherwise normal subjects. This suggests that NRG1 may increase susceptibility to psychosis by altering connections between prefrontal cortex and other brain regions.

Using T1-weighted MRI, and DTI, abnormalities of white matter density and integrity have been shown in the ALIC in schizophrenia,2, 3 bipolar disorder2, 26 and their unaffected relatives.2 These findings have been associated with increased genetic liability to psychosis, albeit to unidentified loci4 and by functional imaging studies in patients with schizophrenia27, 28 and those at high risk for genetic reasons.29 These studies further implicate white matter disconnectivity in psychosis, although they cannot determine whether this is because of reduced axonal number, increased axonal disorganization or dysmyelination.

Steffanson et al.10, 11 provided the first evidence of an association between NRG1 and schizophrenia in an Icelandic population, an association that was later extended to a Scottish population12 and to patients with bipolar disorder.13 Recent work from our group has found that rs6994992 (SNP8NRG243177) is associated with abnormal cortical function and psychotic symptoms in people at increased risk of schizophrenia for genetic reasons,17 supporting a functional role for this polymorphism in the development of psychosis. Together with evidence of altered white matter developmental trajectories in childhood-onset schizophrenia, this suggests that NRG1 may, at least in part, exert its effects on psychosis susceptibility and brain function through effects on white matter.

NRG1 has multiple roles in the brain and is translated into several different isoforms classified according to their 5′ terminus into types I–VI. SNP rs6994992 (SNP8NRG243177) lies within the promoter region of type-IV NRG1 and is associated with altered transcription factor binding, including myelin-associated transcription factor-I, and altered NRG1 mRNA expression.15 Although type-III NRG1 is currently understood to be more closely involved in axonal myelination, the role of type-IV NRG1 remains to be fully elucidated.30 Recent studies have shown that Ig-NRG isoforms, of which Type-IV neuregulin is an example, are involved in neuronal migration and axon guidance.30 One possibility therefore is that alterations in the expression of type-IV NRG1 influence white matter by perturbing the development of white matter tracts. It however also remains possible that rs6994992 (SNP8NRG243177) is marking variation elsewhere in the deCODE haplotype, or that it is in linkage disequilibrium with another as yet unidentified functional variant.

Ig-NRG1 has been implicated in the guidance of thalamocortical axons from the dorsomedial thalamus, through the stratum to the ventral telencephalon in mice.21 Since the ALIC contains tracts connecting the prefrontal cortex to the thalamus, striatum and other brain regions, these findings are of potential relevance to the current study, although no direct evidence of this kind exists in human subjects. Reduced axonal guidance could however lead to reduced axonal numbers, or increased axonal disorganization within in the internal capsule, and consequently to reduced density and integrity in the same region.

A number of potential limitations to the current study should be considered. First, the sample size is relatively small in comparison to most genetic association studies. This limitation is however minimized by the examination of a specific hypothesis and the use of two imaging modalities with mutually supportive findings. Second, all subjects were unaffected, none had a known family history of psychosis and none are expected to develop psychosis in future years. Although we argue that the effects of genetic variation in NRG1 on white matter may contribute to disease pathogenesis, it is likely that the impact of this effect on the development of symptoms is dependant on the presence of other interacting genetic or environmental factors. Nevertheless, studying unaffected controls has considerable advantages in terms of identifying the phenotypic effects of NRG1 variation in a situation unconfounded by factors associated with the disease state, such as medication.

In conclusion, we have shown that variation in the neuregulin gene has demonstrable effects on white matter structure and connectivity. This provides support for the involvement of NRG1 across diagnostic boundaries through actions on neuronal migration, axon guidance or myelination.


  1. 1

    Kendell RE . The Role of Diagnosis in Psychiatry. Blackwell: Edinburgh, 1975.

  2. 2

    McIntosh AM, Job DE, Moorhead TW, Harrison LK, Lawrie SM, Johnstone EC . White matter density in patients with schizophrenia, bipolar disorder and their unaffected relatives. Biol Psychiatry 2005; 58: 254–257.

  3. 3

    Zhou S, Suzuki M, Hagino H, Takahashi T, Kawasaki Y, Nohara S et al. Decreased volume and increased asymmetry of the anterior limb of the internal capsule in patients with schizophrenia. Biol Psychiatry 2003; 54: 427–436.

  4. 4

    McIntosh AM, Job D, Moorhead TWJ, Harrison LK, Whalley HC, Johnstone EC et al. Genetic liability to schizophrenia or bipolar disorder and its relationship to brain structure. Am J Med Genet 2006; 141B: 76–83.

  5. 5

    Guo AC, MacFall JR, Provenzale JM . Multiple sclerosis: diffusion tensor MR imaging for evaluation of normal-appearing white matter. Radiology 2002; 222: 729–736.

  6. 6

    Shenton ME, Dickey CC, Frumin M, McCarley RW . A review of MRI findings in schizophrenia. Schizophr Res 2001; 49: 1–52.

  7. 7

    Sussmann JE, Lymer GKS, McKirdy J, Moorhead TWJ, Maniega SM, Job D et al. White matter abnormalities in bipolar disorder detected using diffusion tensor magnetic resonance imaging. (in review) 2007.

  8. 8

    Maniega SM, Lymer GKS, Bastin ME, Marjoram D, Job DE, Moorhead TWJ et al. A DT-MRI study of subjects at high genetic risk of schizophrenia. In review 2007.

  9. 9

    Craddock N, O'Donovan MC, Owen MJ . Genes for schizophrenia and bipolar disorder? Implications for psychiatric Nosology. Schizophr Bull 2006; 32: 9–16.

  10. 10

    Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, Sigmundsson T, Brynjolfsson J et al. Neuregulin 1 and susceptibility to schizophrenia. Am J Hum Genet 2002; 71: 185.

  11. 11

    Stefansson H, Sarginson J, Kong A, Yates P, Steinthorsdottir V, Gudfinnsson E et al. Association of neuregulin 1 with schizophrenia confirmed in a Scottish population. Am J Hum Genet 2003; 72: 83–87.

  12. 12

    Thomson PA, Christoforou A, Morris SW, Adie E, Pickard BS, Porteous DJ et al. Association of neuregulin 1 with schizophrenia and bipolar disorder in a second cohort from the Scottish population. Mol Psychiatry 2007; 12: 94–104.

  13. 13

    Green EK, Raybould R, MacGregor S, Gordon-Smith K, Heron J, Hyde S et al. Operation of the schizophrenia susceptibility gene, neuregulin 1, across traditional diagnostic boundaries to increase risk for bipolar disorder. Arch Gen Psychiatry 2005; 62: 642–648.

  14. 14

    Li D, Collier DA, He L . Meta-analysis shows strong positive association of the neuregulin 1 (NRG1) gene with schizophrenia. Hum Mol Genet 2006; 15: 1995–2002.

  15. 15

    Law AJ, Lipska BK, Weickert CS, Hyde TM, Straub RE, Hashimoto R et al. Neuregulin 1 transcripts are differentially expressed in schizophrenia and regulated by 5′ SNPs associated with the disease. Proc Natl Acad Sci USA 2006; 103: 6747–6752.

  16. 16

    Tan W, Wang Y, Gold B, Chen J, Dean M, Harrison PJ et al. Molecular cloning of a brain-specific, developmentally regulated neuregulin 1 (NRG1) isoform and identification of a functional promoter variant associated with schizophrenia. J Biol Chem 2007; 282: 24343–24351.

  17. 17

    Hall J, Whalley HC, Job DE, Baig BJ, McIntosh AM, Evans KL et al. A neuregulin 1 variant associated with abnormal cortical function and psychotic symptoms. Nat Neurosci 2006; 9: 1477–1478.

  18. 18

    Stefanis NC, Trikalinos TA, Avramopoulos D, Smyrnis N, Evdokimidis I, Ntzani EE et al. Impact of schizophrenia candidate genes on schizotypy and cognitive endophenotypes at the population level. Biol Psychiatry 2007; 62: 784–792.

  19. 19

    Friston KJ, Frith CD . Schizophrenia: a disconnection syndrome? Clin Neurosci 1995; 3: 89–97.

  20. 20

    Addington AM, Gornick MC, Shaw P, Seal J, Gogtay N, Greenstein D et al. Neuregulin 1 (8p12) and childhood-onset schizophrenia: susceptibility haplotypes for diagnosis and brain developmental trajectories. Mol Psychiatry 2007; 12: 195–205.

  21. 21

    Lopez-Bendito G, Cautinat A, Sanchez JA, Bielle F, Flames N, Garratt AN et al. Tangential neuronal migration controls axon guidance: a role for neuregulin-1 in thalamocortical axon navigation. Cell 2006; 125: 127–142.

  22. 22

    Corfas G, Roy K, Buxbaum J . Neuregulin 1-erbB signaling and the molecular/cellular basis of schizophrenia. Nat Neurosci 2004; 7: 575–580.

  23. 23

    Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS . A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001; 14 (Part 1): t-36.

  24. 24

    Spencer MD, Moorhead TW, McIntosh AM, Stanfield AC, Muir WJ, Hoare P et al. Grey matter correlates of early psychotic symptoms in adolescents at enhanced risk of psychosis: a voxel-based study. Neuroimage 2007; 35: 1181–1191.

  25. 25

    Moorhead TWJ, Job DE, Whalley HC, Sanderson TL, Johnstone EC, Lawrie SM . Voxel-based morphometry of comorbid schizophrenia and learning disability: analyses in normalized and native spaces using parametric and nonparametric statistical methods. Neuroimage 2004; 22: 188–202.

  26. 26

    McIntosh AM, Job DE, Moorhead WJ, Harrison LK, Whalley HC, Johnstone EC et al. Genetic liability to schizophrenia or bipolar disorder and its relationship to brain structure. Am J Med Genet B Neuropsychiatr Genet 2006; 141: 76–83.

  27. 27

    Fletcher P, McKenna PJ, Friston KJ, Frith CD, Dolan RJ . Abnormal cingulate modulation of fronto-temporal connectivity in schizophrenia. Neuroimage 1999; 9: 337–342.

  28. 28

    Lawrie SM, Buechel C, Whalley HC, Frith CD, Friston KJ, Johnstone EC . Reduced frontotemporal functional connectivity in schizophrenia associated with auditory hallucinations. Biol Psychiatry 2002; 51: 1008–1011.

  29. 29

    Whalley HC, Simonotto E, Marshall I, Owens DG, Goddard NH, Johnstone EC et al. Functional disconnectivity in subjects at high genetic risk of schizophrenia. Brain 2005; 128 (Part 9): 2097–2108.

  30. 30

    Harrison PJ, Law AJ . Neuregulin 1 and schizophrenia: genetics, gene expression, and neurobiology. Biol Psychiatry 2006; 60: 132–140.

Download references


We thank all of the participants without whom this study would not have been possible. This study was supported by a Chief Scientist Office Project Grant (CZB/4/434) to AMM and colleagues, two MRC Programme Grants (G9226254/G9825423) to ECJ and colleagues, the Sir Mortimer and Theresa Sackler Foundation and two MRC Clinical Training Fellowships (G84/5699 to Dr McIntosh and G0600429 to Dr Hall). McIntosh is currently supported by the Health Foundation. We also thank the Scottish Funding Council Brain Imaging Research Centre and Wellcome Trust Clinical Research Facility for the image acquisition and genetic analyses respectively.

Author information

Correspondence to A M McIntosh.

Additional information

Conflicts of interest

None of the authors have any conflict of interest to declare.

Rights and permissions

Reprints and Permissions

About this article


  • schizophrenia
  • bipolar disorder
  • MRI
  • tensor-based morphometry
  • cohort study

Further reading