This is another important study supporting the glial growth factors deficiency and synaptic destabilization hypothesis of schizophrenia we proposed in 2002 (Moises et al., 2002). The glial synaptic destabilization hypothesis is based on the landmark 1997 paper by Pfrieger and Barres and the tripartite synapse model suggested by Philip Haydon and coworkers (Araque et al., 1999; Pascual et al., 2005). In reference to its underlying principle, the glial growth factors deficiency and synaptic destabilization hypothesis might also more conveniently and briefly be designated as the weakened tripartite-synapse hypothesis of schizophrenia.

References:
Moises HW, Zoega T, Gottesman II. The glial growth factors deficiency and synaptic destabilization hypothesis of schizophrenia. BMC Psychiatry. 2002;2:8. Abstract

Moises HW, Gottesman II. Does glial asthenia predispose to schizophrenia? Arch Gen Psychiatry 2004; 61:1170. Abstract

Pfrieger FW, Barres BA. Synaptic efficacy enhanced by glial cells in vitro. Science. 1997;277:1684-7. Abstract

Araque A, Parpura V, Sanzgiri RP, Haydon PG. Tripartite synapses: glia, the unacknowledged partner. Trends Neurosci. 1999; 22:208-15. Abstract

Pascual O, Casper KB, Kubera C, Zhang J, Revilla-Sanchez R, Sul JY, Takano H, Moss SJ, McCarthy K, Haydon PG. Astrocytic purinergic signaling coordinates synaptic networks. Science 2005; 310: 113-6. Abstract

View all comments by Hans W. Moises

Peirce's paper is an exciting addition to the white matter hypothesis in schizophrenia. (Note: many of the authors of this paper are colleagues of ours at the Conte Center investigating white matter in schizophrenia at Mount Sinai.) As noted in the news story, findings from a number of different areas are beginning to come together in support of the white matter hypothesis in schizophrenia. Genetic findings in myelin-related genes, as outlined and referenced above, are demonstrating increased susceptibility to schizophrenia. Imaging findings from diffusion tensor studies are demonstrating abnormalities across multiple brain areas (reviewed in Kubicki et al., 2005), with more recent studies showing that specific white matter tracts are not only abnormal in schizophrenia, but are associated with symptomatology and cognitive deficits (Kubicki et al., 2002;   Read more

Peirce's paper is an exciting addition to the white matter hypothesis in schizophrenia. (Note: many of the authors of this paper are colleagues of ours at the Conte Center investigating white matter in schizophrenia at Mount Sinai.) As noted in the news story, findings from a number of different areas are beginning to come together in support of the white matter hypothesis in schizophrenia. Genetic findings in myelin-related genes, as outlined and referenced above, are demonstrating increased susceptibility to schizophrenia. Imaging findings from diffusion tensor studies are demonstrating abnormalities across multiple brain areas (reviewed in Kubicki et al., 2005), with more recent studies showing that specific white matter tracts are not only abnormal in schizophrenia, but are associated with symptomatology and cognitive deficits (Kubicki et al., 2002; Kubicki et al., 2003; Nestor et al., 2004). Postmortem examination is revealing that oligodendrocytes are decreased in number and abnormally spaced in patients with schizophrenia (Hof et al., 2002; Hof et al., 2003). These converging data argue strongly for the involvement of myelin, oligodendrocytes, and white matter in schizophrenia.

We continue to examine various aspects of white matter involvement in schizophrenia with the hope of providing both translational data (i.e., the relationship between symptom severity or independent living and white matter coherence) and further basic science data that may shed some light on upstream events that contribute to myelin and oligodendrocyte deficits. These new data by the Owen and O'Donovan group are a valuable contribution.

References:
Hof PR, Haroutunian V, Copland C, Davis KL, Buxbaum JD. Molecular and cellular evidence for an oligodendrocyte abnormality in schizophrenia. Neurochem Res. 2002 Oct;27(10):1193-200. Abstract

Hof PR, Haroutunian V, Friedrich VL Jr, Byne W, Buitron C, Perl DP, Davis KL. Loss and altered spatial distribution of oligodendrocytes in the superior frontal gyrus in schizophrenia. Biol Psychiatry. 2003 Jun 15;53(12):1075-85. Abstract

Kubicki M, McCarley R, Westin CF, Park HJ, Maier S, Kikinis R, Jolesz FA, Shenton ME. A review of diffusion tensor imaging studies in schizophrenia. J Psychiatr Res. 2005 Jul 13; [Epub ahead of print] Abstract

Kubicki M, Westin CF, Maier SE, Frumin M, Nestor PG, Salisbury DF, Kikinis R, Jolesz FA, McCarley RW, Shenton ME. Uncinate fasciculus findings in schizophrenia: a magnetic resonance diffusion tensor imaging study. Am J Psychiatry. 2002 May;159(5):813-20. Abstract

Kubicki M, Westin CF, Nestor PG, Wible CG, Frumin M, Maier SE, Kikinis R, Jolesz FA, McCarley RW, Shenton ME. Cingulate fasciculus integrity disruption in schizophrenia: a magnetic resonance diffusion tensor imaging study. Biol Psychiatry. 2003 Dec 1;54(11):1171-80. Erratum in: Biol Psychiatry. 2004 Mar 15;55(6):661. Abstract

Nestor PG, Kubicki M, Gurrera RJ, Niznikiewicz M, Frumin M, McCarley RW, Shenton ME. Neuropsychological correlates of diffusion tensor imaging in schizophrenia. Neuropsychology. 2004 Oct;18(4):629-37. Abstract

The Peirce et al. paper represents an important contribution to understanding the possible mechanisms through which genetic risk factors could contribute to the pathophysiology of schizophrenia. Studies of SNPs in candidate genes for schizophrenia are most clearly related to mechanism when the SNP changes amino acid sequence (rarely), or when the SNP changes mRNA expression (commonly postulated, but less often demonstrated). Studies combining SNP and mRNA analyses are challenging, and Peirce et al. provide a novel approach—by measuring the relative amount of mRNA expressed from the variant and the wild-type alleles in brain tissue from heterozygotes. They demonstrated relatively reduced expression from the variant allele. It must be noted however, that these studies were carried out in brain tissue from individuals described as being “free from psychiatric or neurological disorder at time of death” (not schizophrenia samples as suggested by the SRF news story [Editor's note: since corrected]), and the total expression of CNP mRNA was not determined. While CNP mRNA...  Read more

The Peirce et al. paper represents an important contribution to understanding the possible mechanisms through which genetic risk factors could contribute to the pathophysiology of schizophrenia. Studies of SNPs in candidate genes for schizophrenia are most clearly related to mechanism when the SNP changes amino acid sequence (rarely), or when the SNP changes mRNA expression (commonly postulated, but less often demonstrated). Studies combining SNP and mRNA analyses are challenging, and Peirce et al. provide a novel approach—by measuring the relative amount of mRNA expressed from the variant and the wild-type alleles in brain tissue from heterozygotes. They demonstrated relatively reduced expression from the variant allele. It must be noted however, that these studies were carried out in brain tissue from individuals described as being “free from psychiatric or neurological disorder at time of death” (not schizophrenia samples as suggested by the SRF news story [Editor's note: since corrected]), and the total expression of CNP mRNA was not determined. While CNP mRNA expression is reported to be lower in schizophrenia, and Peirce et al. demonstrate the variant allele is a risk factor for schizophrenia in studies of genetic association, it remains uncertain to what extent the lower CNP mRNA expression in schizophrenia is related to genetic variation or to other factors. CNP mRNA differences in expression between schizophrenia and control samples appear to be of different magnitude in different brain regions from the same cases (Katsel et al., 2005). This could represent non-genetic effects. However, genetic variation in CNP could also be more or less likely to be expressed in different brain regions. In this regard, the samples used in the Peirce et al. study were mixed, coming from frontal, parietal, or temporal cortex. Studies with larger sample sizes, and of schizophrenia as well as control tissues, will be needed to test these possibilities.

This paper demonstrates several important shifts in research strategies for schizophrenia. Many previous studies of candidate genes in the illness have chosen their targets based on concepts of the mechanism of action of antipsychotic drugs, or by virtue of the proximity of a gene to a genetic linkage site defined with anonymous markers. The choice of candidate gene here is based on a wide range of neurobiological evidence, including studies of gene expression and protein levels. As well, the authors do not limit their study to one gene; instead, they expand their investigation to include plausibly interacting gene targets. Analysis of complex disorders will likely need more than simple models, and the approach here is worth noting.

The gap still remains between the DNA-mRNA approaches and protein analysis. Gene expression is one factor determining mRNA levels. However, especially in human brain tissue samples, many other antemortem and postmortem factors contribute to the measured level of mRNA. The meaning of gene expression measures obtained for...  Read more

This paper demonstrates several important shifts in research strategies for schizophrenia. Many previous studies of candidate genes in the illness have chosen their targets based on concepts of the mechanism of action of antipsychotic drugs, or by virtue of the proximity of a gene to a genetic linkage site defined with anonymous markers. The choice of candidate gene here is based on a wide range of neurobiological evidence, including studies of gene expression and protein levels. As well, the authors do not limit their study to one gene; instead, they expand their investigation to include plausibly interacting gene targets. Analysis of complex disorders will likely need more than simple models, and the approach here is worth noting.

The gap still remains between the DNA-mRNA approaches and protein analysis. Gene expression is one factor determining mRNA levels. However, especially in human brain tissue samples, many other antemortem and postmortem factors contribute to the measured level of mRNA. The meaning of gene expression measures obtained for oligodendrocyte/myelination-related genes from samples comprising largely gray matter is not entirely certain. The role of oligodendrocytes in gray matter may deserve more attention.

The genetic evidence presented here for an interaction between OLIG2 and ErbB4 is intriguing. A recent paper from Steve Arnold’s group indicated the neuregulin-1−ErbB4 signaling pathway appears to be overactive in schizophrenia, with consequences for NMDA receptor function (Hahn et al., 2006). Although the analysis in their paper focused on neurons, if a similarly overactive pathway was operative in oligodendrocytes, inhibition of myelination might be a predicted outcome (Sussman et al., 2005), with downregulation of a host of oligodendrocyte/myelination-related genes as a consequence. Of further interest, NMDA receptors may have important roles in oligodendrocytes as well as in neurons (Matute, 2006).

References:

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. Altered neuregulin 1-erbB4 signaling contributes to NMDA> receptor hypofunction in schizophrenia. Nat Med. 2006 Jul;12(7):824-8. Epub 2006 Jun 11. Abstract

Sussman CR, Vartanian T, Miller RH. The ErbB4 neuregulin receptor mediates suppression of oligodendrocyte maturation. J Neurosci. 2005 Jun 15;25(24):5757-62. Abstract

Matute C. Oligodendrocyte NMDA receptors: a novel therapeutic target. Trends Mol Med. 2006 Jul;12(7):289-92. Epub 2006 Jun 5. Abstract

Comment by Daniel Stewart and Kenneth Davis
The Corfas results are intriguing. Their findings confirm much of what we have either found or suspect in schizophrenia relating to white matter involvement. Demonstrations of OLIG2 interactions with ErbB4 in the cortex and with CNP in the striatum in schizophrenia from our team (Georgieva et al., 2006) fit well with this investigation in providing evidence for a link between a variety of potential etiologic oligodendrocyte-related mechanisms in schizophrenia. While in our study, we did not find interaction with NRG1 and OLIG2, it is important to note that differential expression of NRG1 might be found only at certain points in the timeline of disease development. Other recent support from our team for white matter involvement in schizophrenia comes from an investigation in which an SNP associated with CNP was found to be significantly correlated with schizophrenia (Peirce et al., 2006). Interestingly, Corfas’s group reports that...  Read more

Comment by Daniel Stewart and Kenneth Davis
The Corfas results are intriguing. Their findings confirm much of what we have either found or suspect in schizophrenia relating to white matter involvement. Demonstrations of OLIG2 interactions with ErbB4 in the cortex and with CNP in the striatum in schizophrenia from our team (Georgieva et al., 2006) fit well with this investigation in providing evidence for a link between a variety of potential etiologic oligodendrocyte-related mechanisms in schizophrenia. While in our study, we did not find interaction with NRG1 and OLIG2, it is important to note that differential expression of NRG1 might be found only at certain points in the timeline of disease development. Other recent support from our team for white matter involvement in schizophrenia comes from an investigation in which an SNP associated with CNP was found to be significantly correlated with schizophrenia (Peirce et al., 2006). Interestingly, Corfas’s group reports that when ErbB signaling is abolished in oligodendrocytes, myelin structure appears normal, but the myelin sheath is significantly thinner. This is in line with some of the ultrastructural findings of Uranova’s group and in rodent studies looking at MAG-deficient mice (both reviewed in Davis et al., 2003)—another downregulated myelin-related gene found in brains of schizophrenia patients.

Reductions in oligodendrocyte number on the order of 20 percent have been demonstrated in the brains of schizophrenia patients (Hof et al., 2002). Although this finding does not precisely parallel the findings in this investigation, the authors’ adroitly point out that this may be because the abnormalities they induced were during early oligodendrocyte and myelin expression, while it is possible that the abnormalities seen in the brains of schizophrenia patients occur relatively later in development, more likely during the second large wave of cortical myelination at the end of the second decade of life. The authors also point out that “defects in ErbB signaling in different cell types may contribute to different aspects of psychiatric symptoms.” This might also be the case in schizophrenia, giving rise to the myriad presentations of the disease, as might the fact that expression of both NRG1 and ErbB4 are susceptible to environmental insult.

Other important similarities between the authors’ findings and schizophrenia include that, even in light of these myelin abnormalities, gross brain volumes, as well as several other measures, remained normal. This buttresses the idea that in schizophrenia, myelin abnormalities might be at the root of the often unimpressive brain changes noted in schizophrenia on gross structural imaging. And finally, although speculative, the authors do note an intriguing set of behavioral abnormalities, some of which could mimic the social isolation and poor relatedness of schizophrenia, which is particularly remarkable given the increased susceptibility to amphetamines and the trends seen in DAT, D1, and D2 expression in this investigation.

Corfas’s findings are indeed exciting, and we commend his team on an eloquently designed and implemented investigation.

Neuregulin1 (NRG1) is the most promising risk factor for schizophrenia, and the study of the signaling of NRG1 and its receptor ErbB4 is very important in understanding the pathophysiology of the disease. Like other promising risk factors for schizophrenia, NRG1/ErbB4 is multifunctional with many molecular isoforms. NRG1/ErbB signaling plays a role both before and after birth. Furthermore, ErbB4 is expressed not only in neurons but also in other types of cells, such as oligodendrocytes.

To address context-dependent functions one by one, dominant-negative transgenic mice can be very useful. The advantage of dominant-negative transgenics is that we can knock down the endogenous function of our target molecules (in this work, ErbB4) in a temporally and spatially specific manner by utilizing a well-characterized promoter. In this outstanding study by Corfas and colleagues, they used the CNP promoter that confirms dominant-negative ErbB4 selectively in oligodendrocytes (but not in astrocytes and neurons) only after birth. This approach will be very useful in schizophrenia...  Read more

Neuregulin1 (NRG1) is the most promising risk factor for schizophrenia, and the study of the signaling of NRG1 and its receptor ErbB4 is very important in understanding the pathophysiology of the disease. Like other promising risk factors for schizophrenia, NRG1/ErbB4 is multifunctional with many molecular isoforms. NRG1/ErbB signaling plays a role both before and after birth. Furthermore, ErbB4 is expressed not only in neurons but also in other types of cells, such as oligodendrocytes.

To address context-dependent functions one by one, dominant-negative transgenic mice can be very useful. The advantage of dominant-negative transgenics is that we can knock down the endogenous function of our target molecules (in this work, ErbB4) in a temporally and spatially specific manner by utilizing a well-characterized promoter. In this outstanding study by Corfas and colleagues, they used the CNP promoter that confirms dominant-negative ErbB4 selectively in oligodendrocytes (but not in astrocytes and neurons) only after birth. This approach will be very useful in schizophrenia research.

The remarkable finding is that they observed alterations in dopamine-mediated neuronal networks and associated behaviors by disturbing NRG1/ErbB4 selectively in cells of oligodendrocyte lineage. Three important paradigms for schizophrenia (white matter pathology, dopamine, and a susceptibility gene) converge in this paper, and in this sense, I find it very exciting.

Does the effect of NRG1/ErbB4 signaling on myelination occur downstream of purinergic signaling? Fields suggests that adenosine is of primary importance in regulating early development of OPCs, where it stimulates differentiation and myelination (Fields, 2006). It's of interest that cAMP stimulates expression of neuregulin and cAMP levels in the lung are decreased in A2A adenosine receptor (22q11.2)-deficient mice (Tokita et al., 2001; Nadeem et al., 2007). Do you see reduced neuregulin levels in 22q11 deletion syndrome? Of particular interest is the study by Desai and colleagues reporting that signaling via the adenosine A2A receptor downregulates thrombospondin 1 (Desai et al., 2005). Perhaps overexpression of thrombospondin 1 may help explain the occular abnormalities in this syndrome (Wu et al., 2006; Forbes et al., 2007; Stalmans, 2005). Thrombospondins are also involved in synaptogenesis (Christopherson et al., 2005).

References:

Fields RD. Nerve impulses regulate myelination through purinergic signalling. Novartis Found Symp. 2006;276:148-58; discussion 158-61, 233-7, 275-81.

Tokita Y, Keino H, Matsui F, Aono S, Ishiguro H, Higashiyama S, Oohira A. Regulation of neuregulin expression in the injured rat brain and cultured astrocytes. J Neurosci. 2001 Feb 15;21(4):1257-64.

Nadeem A, Fan M, Ansari HR, Ledent C, Mustafa SJ. Enhanced airway reactivity and inflammation in A2A adenosine receptor deficient allergic mice. Am J Physiol Lung Cell Mol Physiol. 2007 Feb 9; [Epub ahead of print]

Desai A, Victor-Vega C, Gadangi S, Montesinos MC, Chu CC, Cronstein BN. Adenosine A2A receptor stimulation increases angiogenesis by down-regulating production of the antiangiogenic matrix protein thrombospondin 1. Mol Pharmacol. 2005 May;67(5):1406-13. Epub 2005 Jan 26. Comment in: Mol Pharmacol. 2005 May;67(5):1385-7.

Wu Z, Wang S, Sorenson CM, Sheibani N. Attenuation of retinal vascular development and neovascularization in transgenic mice over-expressing thrombospondin-1 in the lens. Dev Dyn. 2006 Jul;235(7):1908-20.

Forbes BJ, Binenbaum G, Edmond JC, Delarato N, McDonald-McGinn DM, Zackai EH. Ocular findings in the chromosome 22q11.2 deletion syndrome. J AAPOS. 2007 Apr;11(2):179-182. Epub 2006 Nov 30.

Stalmans I. Role of the vascular endothelial growth factor isoforms in retinal angiogenesis and DiGeorge syndrome. Verh K Acad Geneeskd Belg. 2005;67(4):229-76.

Christopherson KS, Ullian EM, Stokes CC, Mullowney CE, Hell JW, Agah A, Lawler J, Mosher DF, Bornstein P, Barres BA. Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell. 2005 Feb 11;120(3):421-33. Comment in: Cell. 2005 Feb 11;120(3):292-3.

View all comments by Mary Reid

Several lines of evidence indicate that oligodendrocytes and myelin are disturbed in schizophrenia (Davis et al., 2003; Segal et al., 2007). However, the relationship of these alterations to the pathogenesis of schizophrenia is still unclear. A recent paper by Budel et al. proposes one possible link between oligodendrocyte and myelin pathology and schizophrenia pathogenesis. The gene for Nogo-66 receptor 1 (RTN4R) is located within the 22q11.2 locus where a hemizygous microdeletion (1.5 Mb) occurs at a frequency of one in 5,000. Twenty to 30 percent of individuals with the deletion develop schizophrenia. Several candidate genes for the schizophrenia phenotype within this locus have been characterized for genetic association, and common variants of the Nogo-66 receptor 1 gene have shown association in one study (Liu et al., 2002), but replication studies have not confirmed the findings using different cohorts (  Read more

Several lines of evidence indicate that oligodendrocytes and myelin are disturbed in schizophrenia (Davis et al., 2003; Segal et al., 2007). However, the relationship of these alterations to the pathogenesis of schizophrenia is still unclear. A recent paper by Budel et al. proposes one possible link between oligodendrocyte and myelin pathology and schizophrenia pathogenesis. The gene for Nogo-66 receptor 1 (RTN4R) is located within the 22q11.2 locus where a hemizygous microdeletion (1.5 Mb) occurs at a frequency of one in 5,000. Twenty to 30 percent of individuals with the deletion develop schizophrenia. Several candidate genes for the schizophrenia phenotype within this locus have been characterized for genetic association, and common variants of the Nogo-66 receptor 1 gene have shown association in one study (Liu et al., 2002), but replication studies have not confirmed the findings using different cohorts (Meng et al., 2007; Hsu et al., 2007). Rare variants for the Nogo-66 receptor 1 gene not found in controls have also been identified in schizophrenia cases (Hsu et al., 2007; Sinibaldi et al., 2004), but their functional or pathological association with schizophrenia has not been demonstrated.

Budel et al. first performed a genetic association study of common variants of Nogo-66 receptor 1 gene and confirmed previous findings in their independent cohort. Because these are intronic variants whose effects on Nogo-66 receptor 1 are not clear, they also searched for rare variants of the gene by direct sequencing of 542 DNA samples from individuals with schizophrenia. They found four previously unreported rare variants that are non-synonymous, giving a rate of non-synonymous rare variants of ~1 percent (eight in 870). These researchers also sequenced 650 control DNA samples and found eight rare non-synonymous variants, which together with previous other studies makes the total number of sequenced control samples 1,250. The overall incidence of coding region variants in controls and schizophrenia does not differ between the two groups. However, a bioinformatic analysis suggested that four out of eight rare variants found in schizophrenia are potentially detrimental to protein function, whereas none of eight rare variants found in controls are, indicating a higher incidence of potentially deleterious variations in the Nogo-66 receptor 1 gene in schizophrenia.

Nogo-66 is a myelin-associated outgrowth inhibitor that is responsible for the inhibition of regeneration of CNS axons. It binds to Nogo-66 receptor 1 which induces repulsive responses from neurons. On this basis, Budel et al. investigated whether these rare variants found in schizophrenia are functional. They found that both the R377Q and R377W variants in the signaling domain could not induce Nogo-66-mediated repulsive response from neurons when they were introduced into neurons that do not express the Nogo receptor. Interestingly, these variants could suppress the repulsive response from neurons that express endogenous Nogo-66 receptor 1, suggesting that they could work as dominant-negative forms. They also showed that Nogo-66 binding domain variants R119W and R196H both showed reduced binding to Nogo-66. Their results demonstrate that some of the rare variants found in schizophrenia (and not in controls) are functional variants, some of which may work as dominant negatives.

They further analyzed Nogo-66 receptor 1 knockout mice for behavioral characteristics relevant to schizophrenia. They found that Nogo-66 receptor 1 knockout mice show impairment in spatial working memory, a promising endophenotype in schizophrenia. This was specific to working memory as no deficits in the radial arm water maze test or passive avoidance test were observed. Nogo-66 receptor 1 knockout mice did not, however, show any changes in PPI, another important phenotype relevant for schizophrenia mouse models.

The authors had shown previously that Nogo-66 receptor-mediated axon growth inhibition is crucial for formation of ocular dominance in the visual cortex in mice, suggesting myelin involvement in brain wiring refinement and restriction of neuronal plasticity (McGee et al., 2005). The prefrontal cortex—believed to be impaired in schizophrenia—completes its myelination during late adolescence and early adulthood, when symptoms of schizophrenia emerge (Benes, 1989). Therefore, their study supports the idea that abnormal myelination may be a risk factor for schizophrenia. Interestingly, another Nogo receptor, pirB, has been identified (Atwal et al., 2008), and PirB is also shown to be involved in ocular dominance formation in the visual cortex (Syken et al., 2006). It would be interesting to look into PirB involvement in myelination and oligodendrocyte differentiation as well as in schizophrenia.

The Budel et al. study leaves a number of issues open. First, the incidence of each rare variant is one in 870 in schizophrenia, and to prove that they are not found in controls would require screening in a much larger number of controls for a reliable genetic study. This is always the problem when we characterize rare variants in association with disorders such as schizophrenia. Second, Hsu et al. also performed a behavioral analysis of their Nogo-66 receptor 1 knockout mice and did not find the working memory deficits that were seen in this study. They both used T maze-based delayed spatial working memory paradigm, but with subtle difference in their protocols. Also, Budel et al. used very strict inclusion/exclusion criteria of animals for the final testing based on animals’ ability during training sessions. Furthermore, as Hsu et al. suggested, there may be differences in mouse genetic background and age, which is not described in this study; Hsu et al. used a C57Blx129 mixed genetic background. Third, as Budel et al. mentioned, a recent study showed that Nogo-66 receptor 1 has synaptic functions, playing a role in glutamate receptor modulation (Lee et al., 2008), and as such it is distinctly possible that the Nogo-66 receptor 1 involvement in pathogenesis of schizophrenia is not related to myelin, but to synaptic deficits. Nevertheless, this study clearly demonstrated that rare variants in Nogo-66 receptor 1 found in individuals with schizophrenia show defects in myelin-mediated neuronal function, which could explain in part the working memory deficits observed in Nogo-66 receptor 1 knockout mice, and may provide a possible causal link between defects in oligodendrocyte function, myelination, and pathogenesis of schizophrenia.

References:

Davis, K.L., et al. (2003) White matter changes in schizophrenia: evidence for myelin-related dysfunction. Arch Gen Psychiatry 60, 443-456. Abstract

Segal, D., et al. (2007) Oligodendrocyte pathophysiology: a new view of schizophrenia. Int J Neuropsychopharmacol 10, 503-511. Abstract

Liu, H., et al. (2002) Genetic variation in the 22q11 locus and susceptibility to schizophrenia. Proc Natl Acad Sci U S A 99, 16859-16864. Abstract

Meng, J., et al. (2007) No association between the genetic polymorphisms in the RTN4R gene and schizophrenia in the Chinese population. J Neural Transm 114, 249-254. Abstract

Hsu, R., et al. (2007) Nogo Receptor 1 (RTN4R) as a candidate gene for schizophrenia: analysis using human and mouse genetic approaches. PLoS ONE 2, e1234. Abstract

Sinibaldi, L., et al. (2004) Mutations of the Nogo-66 receptor (RTN4R) gene in schizophrenia. Hum Mutat 24, 534-535. Abstract

McGee, A.W., et al. (2005) Experience-driven plasticity of visual cortex limited by myelin and Nogo receptor. Science 309, 2222-2226. Abstract Abstract

Atwal, J.K., et al. (2008) PirB is a functional receptor for myelin inhibitors of axonal regeneration. Science 322, 967-970. Abstract

Syken, J., et al. (2006) PirB restricts ocular-dominance plasticity in visual cortex. Science 313, 1795-1800. Abstract

Lee, H., et al. (2008) Synaptic function for the Nogo-66 receptor NgR1: regulation of dendritic spine morphology and activity-dependent synaptic strength. J Neurosci 28, 2753-2765. Abstract

Individuals with hemizygous microdeletions at the 22q11.2 locus display a range of cognitive and behavioral deficits, and compared to the general population these individuals have a greatly increased risk of developing schizophrenia (Karayiorgou et al., 1995). A number of candidate schizophrenia susceptibility genes have been identified within the 22q11.2 region (Mukai et al., 2004; Paterlini et al., 2005; Paylor et al., 2006; Stark et al., 2008). In our paper (Hsu et al., 2007), we evaluated RTN4R (NgR1), one of the genes in the 22q11.2 region, as a schizophrenia susceptibility gene using a variety of approaches including human association analyses as well as mouse behavioral and anatomical assays. We evaluated common RTN4R variants in a large Afrikaner family sample and found RTN4R polymorphisms which...  Read more

Individuals with hemizygous microdeletions at the 22q11.2 locus display a range of cognitive and behavioral deficits, and compared to the general population these individuals have a greatly increased risk of developing schizophrenia (Karayiorgou et al., 1995). A number of candidate schizophrenia susceptibility genes have been identified within the 22q11.2 region (Mukai et al., 2004; Paterlini et al., 2005; Paylor et al., 2006; Stark et al., 2008). In our paper (Hsu et al., 2007), we evaluated RTN4R (NgR1), one of the genes in the 22q11.2 region, as a schizophrenia susceptibility gene using a variety of approaches including human association analyses as well as mouse behavioral and anatomical assays. We evaluated common RTN4R variants in a large Afrikaner family sample and found RTN4R polymorphisms which showed weak sex-specific association with schizophrenia. We also sequenced for rare RTN4R coding variants in an independent sample and identified two novel nonconservative RTN4R coding variants that were found in individuals with schizophrenia but not in a control population. Finally, we observed that Rtn4r-deficient mice had locomotory defects but showed normal behavior in a panel of schizophrenia-related tasks, including working memory.

In a more recently published study, Budel et al. performed similar experiments using independent schizophrenia and control populations. This group identified several nonconservative RTN4R coding variants in schizophrenia patients consistent with previous studies (Sinibaldi et al., 2004; Hsu et al., 2007); as with our study, the screening of a large number of additional patients and controls would be necessary to demonstrate an unequivocal association with schizophrenia. However, the authors provided functional evidence which indicates that some of these variants fail to transduce myelin-derived inhibitory signals in vitro.

Budel and colleagues also evaluated an independently-generated Rtn4r-deficient mouse strain for schizophrenia-related endophenotypes. As was previously reported (Kim et al., 2004), these mice exhibited locomotory deficits including hypoactivity in an open field test, similar to observations by our own group. Interestingly, both our group and the Budel group measured spatial working memory using the delayed alternation T maze assay, but with differing results. While we found no significant differences, Budel et al. reported a deficit in the Rtn4r-deficient mice. There are a number of issues that are raised by this discrepancy as well as several possible explanations.

First, differences in the assay protocols could account for the discrepancy. In our experiments, both the acquisition and retention of working memory were sequentially assayed following the shaping (training) period, whereas Budel et al. averaged data across the 30 days following training, which does not clearly distinguish between the acquisition of the task and the retention of working memory. In addition, Budel et al. used a 2.5 second delay between trials, which would place a smaller load on working memory compared to the 5 and 20 second delays used in our experiments.

Second, the design of the targeting constructs could potentially have an influence on behavior. In our study, we chose a self-excisable selection cassette containing the neo selection marker (Bunting et al., 1999). Excision of the neo gene following germline transmission ensures that any observed phenotype is due to the deletion rather than any long-range transcriptional effects of the selection cassette (Olson et al., 1996). This is particularly important in the case of RTN4R given that the RTN4R gene lies adjacent to ZDHHC8 and PRODH, two other genes in the 22q11.2 region which have been implicated in schizophrenia (Paterlini et al., 2005; Mukai et al., 2008).

Finally, we conducted our experiments in wild-type, heterozygous, and homozygous Rtn4r knockout animals, whereas Budel et al., using smaller cohorts, reported results from only wild-type and homozygous animals in the T maze assay. In the radial-arm water maze, on the other hand, Budel et al. included heterozygotes but omitted the wild-type controls. Comparisons between all three genotypes—homozygous, heterozygous, and wild-type—are particularly relevant in this context given that schizophrenia is strongly associated with a hemizygous deletion of the 22q11.2 region. Furthermore, Budel et al. used pure strain mice for the T maze assay while mixed background mice were used for the water maze; it is unclear to what degree strain effects could play a role (Crawley, 1996). The use of accurate and reliable mouse models is invaluable in analyzing the effect of genetic loci predisposing complex psychiatric disorders (Arguello and Gogos, 2006; Kvajo et al., 2008), and the results discussed here highlight some of the caveats that must be taken into account when interpreting the effect of single gene deletions on complex behaviors such as working memory, particularly when evaluated as an endophenotype of schizophrenia.

References:

Arguello PA, Gogos JA. Modeling madness in mice: one piece at a time. Neuron. 2006. 52(1):179-96. Abstract

Budel S, Padukkavidana T, Liu BP, et al. Genetic variants of Nogo-66 receptor with possible association to schizophrenia block myelin inhibition of axon growth. J Neurosci. 2008. 28(49):13161-72. Abstract

Bunting M, Bernstein KE, Greer JM, et al. Targeting genes for self-excision in the germ line. Genes Dev. 1999. 13(12):1524-8. Abstract

Crawley, JN. Unusual behavioral phenotypes of inbred mouse strains. Trends Neurosci. 1996. 19(5):181-2. Abstract

Hsu R, Woodroffe A, Lai WS, et al. Nogo Receptor 1 (RTN4R) as a candidate gene for schizophrenia: analysis using human and mouse genetic approaches. PLoS ONE. 2007. 2(11):e1234. Abstract

Karayiorgou M, Morris MA, Morrow B, et al. Schizophrenia susceptibility associated with interstitial deletions of chromosome 22q11. Proc Natl Acad Sci U S A. 1995. 92(17):7612-6. Abstract

Kim JE, Liu BP, Park JH, et al. Nogo-66 receptor prevents raphespinal and rubrospinal axon regeneration and limits functional recovery from spinal cord injury. Neuron. 2004. 44(3):439-51. Abstract

Kvajo M, McKellar H, Arguello PA, et al. 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. 105(19):7076-81. Abstract

Mukai J, Dhilla A, Drew LJ, et al. Palmitoylation-dependent neurodevelopmental deficits in a mouse model of 22q11 microdeletion. Nat Neurosci. 2008. 11(11):1302-10. Abstract

Mukai J, Liu H, Burt RA, et al. Evidence that the gene encoding ZDHHC8 contributes to the risk of schizophrenia. Nat Genet. 2004. 36(7):725-31. Abstract

Olson EN, Arnold HH, Rigby PW, et al. Know your neighbors: three phenotypes in null mutants of the myogenic bHLH gene MRF4. Cell. 1996. 85(1):1-4. Abstract

Paterlini M, Zakharenko SS, Lai WS. Transcriptional and behavioral interaction between 22q11.2 orthologs modulates schizophrenia-related phenotypes in mice. Nat Neurosci. 2005. 8(11):1586-94. Abstract

Paylor R, Glaser B, Mupo A, et al. Tbx1 haploinsufficiency is linked to behavioral disorders in mice and humans: implications for 22q11 deletion syndrome. Proceedings of the National Academy of Sciences of the United States of America. 2006. 103(20):7729-34. Abstract

Sinibaldi L, De Luca A, Bellacchio E, et al. Mutations of the Nogo-66 receptor (RTN4R) gene in schizophrenia. Hum Mutat. 2004. 24(6):534-5. Abstract

Stark KL, Xu B, Bagchi A, et al. Altered brain microRNA biogenesis contributes to phenotypic deficits in a 22q11-deletion mouse model. Nat Genet. 2008. 40(6):751-60. Abstract

This is a thorough and generally well-written manuscript that provides further evidence to the hypothesis that schizophrenia may be viewed as a disconnectivity syndrome (Frith, 1996; Davis et al., 2003) due to disturbances in myelination.

Even though the authors examined a large sample consisting of 3 different populations (Caucasians, African-Americans and Chinese Han trio sample), they do not provide details regarding the age-ratio of these populations, nor do they report the treatment of these patients. Hence, there is a growing body of evidence of age-related changes in the human brain (Allen et al., 2005).We consider that the authors of this study fail to investigate of how the effects of age are expressed. It can not be ruled out whether there is any effect of the medication in the observed results. Even though medication is not implicated in the observed alterations in gene expression in schizophrenia in several studies (  Read more

This is a thorough and generally well-written manuscript that provides further evidence to the hypothesis that schizophrenia may be viewed as a disconnectivity syndrome (Frith, 1996; Davis et al., 2003) due to disturbances in myelination.

Even though the authors examined a large sample consisting of 3 different populations (Caucasians, African-Americans and Chinese Han trio sample), they do not provide details regarding the age-ratio of these populations, nor do they report the treatment of these patients. Hence, there is a growing body of evidence of age-related changes in the human brain (Allen et al., 2005).We consider that the authors of this study fail to investigate of how the effects of age are expressed. It can not be ruled out whether there is any effect of the medication in the observed results. Even though medication is not implicated in the observed alterations in gene expression in schizophrenia in several studies (Hakak et al., 2001; Pongrac et al., 2002), the possibility that the expression levels of some genes may be modulated by medication (Pongrac et al., 2002) cannot be disregarded. Ultimately, it is essential to take into consideration that schizophrenia is a syndrome, so less broad diagnostic categories are crucial. These issues must be investigated more intensely, in order to obtain statistically sound results.

Nonetheless, Budel et al. present evidence that there is a positive association of NGR variation in schizophrenic patients which may be relevant to the pathophysiology of the disorder. Their results support the hypothesis that oligodendroglial dysfunction with subsequent abnormalities in myelin maintenance and repair contribute to the schizophrenic syndrome (Davis et al., 2003). Therefore they shed more light on the aetiopathogenesis of schizophrenia and the understanding of its seemingly disparate genetic aspects.

References:

Frith C. Neuropsychology of schizophrenia, what are the implications of intellectual and experiential abnormalities for the neurobiology of schizophrenia? Br Med Bull 1996; 52: 618-626. Abstract

Davis KL, Stewart DG, Friedman JI, Buchsbaum M, Harvey PD, Hof PR et al. White matter changes in schizophrenia: evidence for myelin-related dysfunction. Arch Gen Psychiatry 2003; 60: 443–456. Abstract

Allen JS, Bruss J, Brown CK, Damasio H et al. Normal neuroanatomical variation due to age: The major lobes and a parcellation of the temporal region. Neurobiology of Aging 2005; 26: 1245–1260. Abstract

Hakak Y, Walker JR, Li C, Wong WH, Davis KL, Buxbaum JD et al. Genome-wide expression analysis reveals dysregulation of myelination- related genes in chronic schizophrenia. Proc Natl Acad Sci USA 2001; 98: 4746–4751. Abstract

Pongrac J, Middleton FA, Lewis DA, Levitt P, Mirnics K. Gene expression profiling with DNA microarrays: advancing our understanding of psychiatric disorders. Neurochem Res 2002; 27: 1049–1063. Abstract