Abstract

Neuropathological and brain imaging studies suggest thatschizophrenia may result from neurodevelopmental defects.Cytoarchitectural studies indicate cellular abnormalities suggestive ofa disruption in neuronal connectivity in schizophrenia, particularly inthe dorsolateral prefrontal cortex. Yet, the molecular mechanismsunderlying these findings remain unclear. To identify molecularsubstrates associated with schizophrenia, DNA microarray analysis wasused to assay gene expression levels in postmortem dorsolateralprefrontal cortex of schizophrenic and control patients. Genesdetermined to have altered expression levels in schizophrenics relativeto controls are involved in a number of biological processes, includingsynaptic plasticity, neuronal development, neurotransmission, andsignal transduction. Most notable was the differential expression ofmyelination-related genes suggesting a disruption in oligodendrocytefunction in schizophrenia.

Methods

Microarray Procedure.

Microarray analysis was performed essentially as previously described(8). Briefly, 8 μg total RNA was used to synthesize cDNA that wasthen used as a template to generate biotinylated cRNA. cRNA wasfragmented and added to Affymetrix (Santa Clara, CA) HuGeneFL chips(contains probes for over 6,000 human genes) as described in thestandard protocol outlined in the Gene Chip Expression AnalysisTechnical Manual (Affymetrix). After sample hybridization, microarrayswere washed and scanned with a laser scanner (Agilent, Palo Alto, CA).Each sample was profiled in duplicate, with cRNA prepared separatelyfrom total RNA. RNA quality was also assessed by examination of the 3′to 5′ ratios for human actin and glyceraldehyde-3-phosphatedehydrogenase (GAPDH) oligonucleotides on Affymetrix Test 2 chips.Relative quantitative reverse transcription (RT)-PCR by usingQuantumRNA 18S Internal Standards (Ambion, Austin, TX) was performed asdescribed in the manufacturer's protocol to validate the changes ingene expression detected by microarray analysis. Seven of the 89 geneslisted in Table1 were tested andconfirmed to have similar changes in expression levels, as determinedby microarray analysis (data not shown).

Table 1.

Genes differentially expressed in chronicschizophrenia Fold-ΔGeneAcc.no.PvalueFold-ΔGeneAcc.no.P value

Open in a new tab

 Genes with altered expression levels inschizophrenia samples in comparison with control samples. Genes areclustered into groups by biological function. Genes not classified arealso listed. The mean fold-change of each gene in schizophrenic samplesrelative to control samples as well as theP value andGenBank accession number are indicated. Genes with decreases inexpression levels are highlighted in red. 

Results

The dorsolateral prefrontal cortex (DLPFC) has been implicated bya number of studies in the pathology of schizophrenia (12,13). DNAmicroarray analysis was performed on postmortem tissue from the DLPFCof 12 chronic schizophrenic and 12 control patients to identify changesin gene transcription associated with the disease. Selection of tissuefrom those patients who had chronic intractable schizophrenia wasintended to reduce the heterogeneity associated with schizophrenia andto maximize disease-associated gene expression differences. Amodel-based metaanalysis approach was used to compute the geneexpression values and confidence interval for the fold-change of eachgene in the profiled tissue (9). Classification of genes having alteredexpression levels in schizophrenics relative to controls was based on anumber of criteria, including fold-change and statistical analyses.Genes found to have altered expression levels in schizophrenia wereclustered into groups by biological function. Identification offunctionally grouped genes may imply that abnormal regulation ofparticular biological processes, biochemical pathways, or cell types isprevalent in the disease state. Table1 lists 89 genes found to havealtered expression levels in schizophrenic samples in comparison withcontrol samples.

A striking finding was the identification of five genes whoseexpression is enriched in myelin-forming oligodendrocytes, all of whichwere transcriptionally down-regulated in schizophrenia. In contrast,genes in all of the other groups were predominantly up-regulated inexpression. These five genes have been implicated in the formation andmaintenance of myelin sheaths, which are critical for efficient axonalsignal propagation and provide extrinsic trophic signals that affectthe development and long-term survival of axons (14). MAL, myelin andlymphocyte protein, and the actin-capping protein gelsolin areexpressed in oligodendrocytes and localized in compact myelin (15,16).Altered expression of 2′,3′-cyclic nucleotide 3′-phosphodiesterase,myelin-associated glycoprotein (MAG), and transferrin has been shown toresult in aberrant oligodendrocyte-mediated myelination and development(17–19). MAG is also believed to regulate the interaction betweenmyelinating cells and axons (20). Dysregulation of these genes suggestsa disruption in normal oligodendrocyte function. The neuregulinreceptor Her3 (ErbB3), which is involved in Schwann cell developmentand myelination (21), was also found to be down-regulated inschizophrenia. Down-regulation of Her3 may likewise affectoligodendrocytes as neuregulin promotes oligodendrocyte survival andthe proliferation of their precursor cells in culture (22).

Thirty-five transcripts that can be detected by the microarrays wereclassified as being involved in myelination, six of which had decreasedexpression levels in schizophrenia. Given that there are2³⁵ − 1 subsets of myelination-related genes,the possibility exists that the set of six shown to be differentiallyexpressed are false-positives (i.e., type-1 errors). The probability ofthe six myelination-related genes passing the filtering criteria bychance was therefore assessed by using a permutation-based analysis.The analysis was performed by randomly partitioning the 24 patient andcontrol samples into two groups of 12 samples. The permutation of thegrouping was performed 200,000 times. This sample size is large enoughto yield a standard error of 0.00021 for the estimatedPvalue. For each permutation, the 35 myelination-related genes wereanalyzed based on the filtering criteria. The sum oft-statistics of the genes in the filtered set (−17.60 forthe original six genes) is used as the statistic to assess thesignificance of the filtered set. AP value of 0.00852 wascalculated based on these results, which is highly suggestive of theinvolvement of myelination-related genes in schizophrenia. Thisanalysis took into account the possibility that myelination-relatedgenes may be coregulated in a given pathway instead of being regulatedindependently of one another. Such an approach provides a much moreconservative estimation of the probability than would be obtained bysimply considering the genes to be only regulated independently.

Evidence from a number of studies has led to the hypothesis thatschizophrenia is a neurodevelopmental disorder with synaptic pathologythat could become most apparent in early adulthood (23). Several genesinvolved in neuronal development and plasticity were also up-regulatedin schizophrenics compared with controls. Myristoylated alanine-rich Ckinase substrate (MARCKS), growth-associated protein-43 (GAP-43),superior cervical ganglia-10 (SCG-10), and neuroserpin are involvedin various aspects of neuronal development. In addition, these genesmay also play a functional role beyond early development by modulatingsynaptic plasticity (24,25). The results are consistent with previousreports of increased GAP-43 protein levels in the frontal cortices ofschizophrenics (26). The light chain of kinesin could likewise beclassified in this group for its role in the transport of variousaxonal components that might modulate neuronal plasticity (27).

Other groups of functionally clustered genes are involved inneurotransmission. Four genes regulating the γ-aminobutyric acid(GABA) neurotransmission pathway were found up-regulated inschizophrenia. The localization of GABA in diverse neuronal populationswhose numbers may be differentially altered in schizophrenia (28)suggests that the transcriptional changes detected here reflect thedisruption of GABAergic neurotransmission in different DLPFC cellgroups. Several neuropeptides were also found to have alteredexpression levels. Each of these neuropeptides has been reported tohave a broad range of functions, which makes it difficult to understandthe effects of their dysregulation. In addition, genes that may beinvolved in neuropeptide processing and release had changed inexpression levels. These results support previous reports ofdisruptions in various neurotransmission systems in schizophrenia (3,29).

Dopamine is known to bind to G protein-coupled receptors, therebyactivating various downstream protein kinase and phosphatase signalingpathways (30). A number of genes involved in these pathways weredysregulated in schizophrenics. These include genes in the proteinkinase A pathway (protein kinase A RII subunit, adenylylcyclase-associated protein 2) and protein kinase C (beta isozyme).Various calcium-regulated components of these pathways such ascalmodulin, calcineurin A, and the calcineurin regulatorZAKI-4 were also up-regulated in schizophrenics, as was the calcineurinsubstrate inositol 1,4,5-triphosphosphate receptor. Both cAMP-dependentprotein kinase A (PKA) and calcineurin-related pathways converge on theregulation of dopamine- and cAMP-regulated phosphoprotein ofM_r 32,000 (DARPP-32) phosphorylation,a key regulator of protein phosphatase 1 (30). Although not present onthe microarray for detection at the mRNA level in the present study,DARPP-32 protein has been shown to be decreased in the prefrontalcortex of schizophrenics (P. Greengard, personal communication).Dysregulation of genes involved in dopamine receptor-mediated signalingis therefore consistent with the hypothesis that schizophrenia may be afunctional disorder of hyperdopaminergic activity (31).

Interestingly, several genes differentially expressed in the braintissue of the schizophrenics are involved in the regulation of thecytoskeleton. NF-L and NF-M are components of neurofilaments, whereasprofilin II regulates actin filament formation. Gelsolin, MARCKS,GAP-43, and SCG10 may also regulate the cytoskeleton by directing andstabilizing axons and dendrites as well as cell shape, size, andpolarity. Alterations in cellular architecture by modulatory elementsof the cytoskeleton could explain some of the morphometric changesreported in different schizophrenic brain regions (2).

The relative expression levels of the 89 genes in each sample aredisplayed in Fig.1. Genes are clusteredby their relative expression pattern over the 24 patient samples, withtheir correlations depicted in the dendrogram. Overall, there appearsto be a distinction in the gene expression pattern between control andschizophrenic samples. As expected, genes found to be down-regulated inschizophrenia tissue relative to control tissue cluster within aseparate branch from the genes found to be up-regulated in the diseasestate. Five of the six myelination-related genes cluster near oneanother, suggesting a similarity in their transcriptional regulation.Other gene expression relationships are also evident from this clusteranalysis. For example, glutamate decarboxylase-65 (GAD65) and GAD67, aswell as somatostatin I and neuropeptide Y, cluster closely to oneanother, as would be expected based on known metabolic or anatomicalrelationships. None of the 89 genes are differentially expressedconsistently in all schizophrenic samples in comparison to all controlsamples. Linear discriminant analysis (LDA; ref.32) was applied toassess the distinction between the schizophrenic and control samplesbased on their expression profiles. This method seeks the linearcombination of variables that maximizes the ratio of between-groupvariance and within-group variance by using the grouping information.Implicitly, the linear weights used by LDA depend on how a geneseparates in the two groups and how this gene correlates with the othergenes. LDA was applied to the data matrix of 24 (samples) by 89 (probesets). The results show that schizophrenic and control samples can bereasonably segregated into two distinct groups based on the pattern oftheir expression profiles (Fig.2A). LDA was also performedfor the six myelination-related genes alone (Fig.2B). A comparable degree of separation of control andschizophrenic samples was found relative to the analysis performed byusing all 89 genes. LDA was repeated by using all 35 of themyelination-related genes that can be detected by the microarray. Thisanalysis takes an unbiased approach because the genes used to fit theLDA are not derived by the selection criteria, but by the virtue oftheir functional relation to the six myelination-related genesidentified in the screen. Using all 35 myelination-related genesmarkedly improved the capacity of the analysis to discriminate the twopatient groups (Fig.2C). Improvement in the LDA by theaddition of other myelination-related genes suggests that theirexpression levels also correlate with the disease state. LDA usingother functional groups identified in the screen resulted in poorergroup segregation (data not shown). These results, together with thepermutation-based analysis, clearly provide evidence of the involvementof the myelination pathway in schizophrenia.

The schizophrenic patients in this study were being treated withneuroleptic (anti-psychotic) medications until the time of death.Neuroleptics are known to inhibit activation of dopaminereceptor-mediated signal transduction. It is possible that treatmentwith neuroleptics may result in the observed changes in geneexpression, such as those involved in dopamine signaling. To addressthis possibility, DLPFC samples from four additional schizophrenicpatients that were medication free for 6 weeks or more were profiled.Statistical analysis (P < 0.05, Student'st test) was performed on the genes listed in Table1 tocompare the expression levels in schizophrenics off medication withthose on medication. None of the 89 genes were found to havestatistically significant differences in expression levels incomparisons between schizophrenics on and off medications. Thus,medication may not account for the observed gene expression changes inschizophrenia. The low sample number of schizophrenics off medication,however, may have limited the analysis of these genes. Therefore, wecannot rule out the possibility that the expression levels of somegenes may be modulated by neuroleptics.

Discussion

We have identified a number of functionally clustered groups ofgenes with altered expression levels in schizophrenics. The mostnotable of these groups of genes is involved in central nervous systemmyelination and points to the possibility of oligodendrocytes as aspecific cell type that is functionally deficient in schizophrenia.Dysregulation of the myelination-related genes is unlikely to be due toneuroleptic medications. Animal studies of the effects of neuroleptictreatments on glial cells have reported discrete increases inproliferation, presumably of astrocytes and microglia (33). Yet, themyelination-related genes identified here are predominantly expressedin oligodendrocytes. Furthermore, dysregulation of several of thesegenes has been shown to inhibit oligodendrocyte development andfunction, rather than promote cellular proliferation (17–19). Inaddition, none of the 89 genes had statistically significantdifferences in expression levels in comparisons between schizophrenicson and off medications.

While this manuscript was in preparation, Mirnicset al.(34) reported a microarray analysis of the prefrontal cortex of sixschizophrenic patients. This report indicated an alteration in theexpression levels of genes involved in the regulation of presynapticfunction. Although we did observe changes in the expression levels ofsome genes involved in presynaptic function, a few reasons may accountfor the discrepancies in the results between the two reports. First,the patient cohort studied here represents a population of elderlyschizophrenics with such profound impairment as to necessitate lifelonghospitalization, whereas the Mirnicset al. cohort werecommunity-dwelling schizophrenics dying at young or middle age.Furthermore, although the Mirnicset al. cohort wasrepresentative of the population who on death were referred to theMedical examiner office, the cohort described here representindividuals who had been residents of long-term hospital units and whodied in old age of natural causes. In addition, the microarray typesused by Mirnicset al. contain probes for only one of thesix myelination genes identified in this study and therefore would nothave detected the changes in gene expression observed here. Becauseschizophrenia is a heterogenous disorder with a varying clinicalpresentation over the lifetime of the patient, studies performed atboth middle and old age have equal relevance in determining theunderlying etiology.

Oligodendrocytes increase neuronal conduction velocity through theirinsulating properties and provide extrinsic trophic factors thatpromote neuronal maturation and axonal survival (35). Given thatmyelinating glia have been reported to regulate the axonal cytoskeleton(35), we hypothesize that alterations in oligodendrocyte–axoninteractions may underlie the subtle cytoarchitectural changes found inschizophrenia (2). This hypothesis is consistent with previous reportsthat suggest a possible role for myelination in schizophrenia.Myelination of the prefrontal cortex has been observed to occur in lateadolescence and early adulthood, which is typically the age of onset ofschizophrenia (36,37). Brain imaging studies have found subtle whitematter abnormalities in schizophrenics (38,39), whereas metachromaticleukodystrophy, a demyelination disorder, is associated with aschizophrenic-like psychoses (40). Immunohistochemical staining of theschizophrenic brain tissues may help to verify a disruption in neuronalmyelination. It will also be of interest to perform studies ondemyelination mouse models with mild phenotypes to look for aschizophrenic-like behavior (e.g., prepulse inhibition). In addition,microarray analyses of other brain regions from schizophrenic patients,as well as tissues from patients with different psychiatric conditions(e.g., depression and anxiety), are warranted to further validate theresults. The finding of a group of genes involved in oligodendrocytefunction having differential expression in the DLPFC of schizophrenicsprovides a new area for future studies and demonstrates the utility ofgenome-scale expression profiling for the discovery of the etiologicalbasis of neuropsychiatric disorders.

Supplementary Material

Acknowledgments

We thank D. Lockhart, J. Hogenesch, and M. Cooke for helpfuldiscussions, and S. Kay and H. C. Hemmings, Jr. for comments onthe manuscript. The Schizophrenia Brain Bank and some of the studiesreported were supported by Merit Review and Mental Illness Research,Education and Clinical Centers (MIRECC; Department of Veterans Affairs)awards (to V.H.) and by U.S. Public Health Service Grant MH45212 (toK.L.D.).

Abbreviations

PMI

postmortem interval

DLPFC

dorsolateralprefrontal cortex

LDA

linear discriminant analysis

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.