Schizophrenia is one of the most common, devastating, and least understood neuropsychiatric illnesses present in the human population. Despite decades of research involving neurochemical, neuroanatomical, neuropathologic, neurodevelopmental, neuropsychological, and genetic approaches, no clear etiopathophysiology has been elucidated. Among the most robust findings, however, is the contribution of genetics to disease development. Statistical models suggest that susceptibility to the disorder is governed by the effects of multiple genes, coupled with environmental and stochastic factors. This review briefly summarizes recent etiopathologic findings and hypotheses, with special attention to genetics.
Schizophrenia is one of the most common, devastating, and least understood neuropsychiatric illnesses afflicting the human population. Although first defined by Emil Kraepelin and Eugen Bleuler at the turn of the 20th century, symptoms of schizophrenia have been described throughout history and across diverse cultures, geographic locations, and socioeconomic categories. Thus, schizophrenia is “part of the common baggage of humanity.”
1A worldwide prevalence of nearly 1% makes schizophrenia one of the most widespread disorders known, and its impact on affected individuals, family members, and society as a whole would be difficult to overstate. Primarily characterized as a disorder of disrupted cognitive and emotional processing, the disease strikes at those characteristics that make us human. Its usual onset in early adulthood tends to derail social relationships and interrupt educational or career plans. Coupled with the chronic nature of the cognitive and emotional debilitation, affected individuals most often experience lifelong occupational disability and disrupted social relationships. In the United States alone, the cost of disease in terms of treatment expense and lost productivity was estimated at more than $30 billion in 1990.
2As a further measurement of the impact of disease, most estimates indicate that as many as a third of homeless Americans have schizophrenia.
Despite decades of research involving neurochemical, neuroanatomical, neuropathologic, neurodevelopmental, neuropsychological, and genetic approaches, no clear etiopathophysiology of schizophrenia has been elucidated. Prevailing hypotheses involve subtle neurodevelopmental defects that alter brain circuitry and result in aberrant information and emotional processing. A genetic contribution to disease development is evident; however, these effects are unlikely to be the result of a single major gene. Rather, the interplay of multiple genes, each with small effects, is more likely to underlie disease susceptibility. Additional factors, including prenatal and postnatal environmental insults and random effects, are presumed to be operative. Further genetic complexity is added by the fact that susceptibility genes and interacting environmental factors likely will vary among different population subgroups. Nonetheless, with use of a variety of genetic epidemiological and molecular genetic approaches, the identification of susceptibility genes should be possible. Both genome-wide gene searches and targeted studies of plausible susceptibility genes should continue. The choice of candidate genes can be guided in part through the findings provided by neurobiological investigations of schizophrenia. A short review of neurobiological findings and a summary of genetic findings is provided herein, along with a brief description of promising directions in the search for susceptibility genes.
DEFINITION AND TREATMENT
Schizophrenia is a severe and chronic neuropsychiatry disease that affects cognition, emotional processing, and behavior. Although termed a disease, clinical heterogeneity is marked; thus, schizophrenia is probably best described as a symptom complex. Characteristic clinical features of schizophrenia can be classified into 3 symptom clusters: (1) positive or psychotic symptoms of hallucinations, delusions (including unusual thoughts and suspiciousness), and distorted perceptions; (2) negative symptoms of flat or blunted affect and emotions, amotivation, avolition, anhedonia, or alogia; and (3) disorganized symptoms of confused thinking, incoherence or looseness of associations in thought and speech, and odd or bizarre behavior.
4Although affected individuals may predominantly display signs and symptoms of 1 cluster type, the occurrence of other symptom types is not precluded. Regardless of predominant symptom type, a general decline in cognitive functions, including attention, executive functions, and working memory, is central to the behavioral disturbance and functional disability.
5As better clinical characterizations are developed, diagnostic criteria for the disorder are updated in the Diagnostic and Statistical Manual of Psychiatric Disorders (DSM).
Subtypes of schizophrenia (catatonic, disorganized, paranoid, undifferentiated, and residual) have been described in the DSM criteria.
6However, these subtypes have been shown “to have little diagnostic validity … only modest stability over time and do not remain consistent within familial groups.”
7Thus, the utility of such subtyping for clinical, research, or other purposes is unclear.
The lifetime morbid risk of schizophrenia is approximately 1% in both men and women. However, sex differences have been reported for age at onset, premorbid functioning, clinical course, response to antipsychotic drugs (APDs), and other disease features. Among men, the median age at onset is in the early to mid-20s, whereas onset is 3 to 5 years later in women. Recent data also suggest a second, smaller peak in women near menopause at ages 40 to 45 years.
8Onset may be abrupt, although most individuals have an insidious onset with a variety of premorbid symptoms. These symptoms, including social withdrawal, diminution or loss of interest in school or work, deterioration in hygiene and grooming, and unusual behaviors, may develop gradually over years. In general, premorbid functioning and social achievement have been reported to be better in women.
13Men have been found to have more negative or deficit symptoms than women.
16Among the myriad of environmental, temporal, and other factors in vestigated for disease association, an excess of winter births has been shown, but only for a subset of patients.
4Additionally, correlation between the occurrence of influenza epidemics and subsequent increases in rates of schizophrenia has been shown, especially when the exposure to influenza occurred during the second trimester of gestation when both cortical and limbic brain development are pronounced.
4However, direct evidence of viral causation or contribution to disease development is lacking.
Because of the chronic nature of schizophrenia, often lifelong treatment is required. Before the advent in the early 1950s of the APD chlorpromazine, affected individuals were sequestered by hospitalization and underwent insulin coma therapy, prefrontal lobotomy, and other dramatic, but not necessarily efficacious, treatments. From about 1950 to the mid-1980s, several chlorpromazine-related drugs and other drugs were introduced. Although mechanisms of action are not completely elucidated, these drugs appear to function primarily through the binding of dopamine D2 receptors and are referred to as typical APDs. The blockade of receptors in the nigrostriatal and corticolimbic dopaminergic pathways frequently results in extrapyramidal adverse effects, parkinsonism, or irreversible tardive dyskinesia (Figure 1).
When treated with typical APDs, approximately 30% of individuals have been reported to be refractory to treatment. Even among responsive patients in whom overt psychotic symptoms have been ameliorated, negative disease symptoms (eg, alogia, avolition, anhedonia) often do not abate and occasionally appear to worsen. Whether these negative symptoms simply are unmasked by treatment of the positive symptoms or whether some are exacerbated or even caused by the adverse effects of treatment is not completely known.
In 1990, clozapine was introduced in the United States as the prototypic atypical APD because of its reportedly greater affinity for various subtypes of serotonin receptors and for D4-type, rather than D2-type, dopamine receptors. Since 1990, additional atypical APDs (eg, risperidone, olanzapine, quetiapine, ziprasidone) have been marketed in the United States, with many more in various stages of clinical testing. Although these drugs hold promise to be both more tolerable to patients (eg, no extrapyramidal adverse effects) and more effective for the amelioration of negative and cognitive deficit symptoms in addition to positive symptoms, these advantages remain to be shown clearly. Atypical APDs have been associated with anticholinergic, antihistamine, and other adverse effects (eg, excessive weight gain) and, in the case of clozapine, include a potentially fatal agranulocytosis. Controversy remains about the optimal treatment of first episode cases,
18but consensus appears to have been reached with regard to treatment-refractory patients. For these individuals, clozapine offers the most benefit.
Thus, the pharmacological treatment of schizophrenia remains empiric, with the clinician aiming to balance an amelioration of symptoms with the occurrence of adverse effects. The choice of APD, the decision to use any adjuvant treatments (eg, lithium, carbamazepine, benzodiazepines), and other treatment decisions often are based on adverse effect profiles, patient's history of response and nonresponse, history of compliance, and other case-by-case considerations.
There is a rapidly increasing breadth and depth of data on the brain in patients with schizophrenia, including findings from structural and functional imaging technologies to molecular gene expression studies. A definitive pathophysiology for the disorder is far from established, with questions remaining about whether some findings are primary or secondary, isolated or part of a larger process, general or specific to particular (as yet undefined) subforms of the disorder, etc. The more robust findings and various hypotheses that have been developed based on an integration of these findings are described subsequently. Integrative hypotheses suggest that schizophrenia is a disease of disordered neural circuitry involving multiple anatomical brain regions (eg, cortex, thalamus, basal ganglia, medial temporal lobe) and their modulation by neurotransmitter-specific projection systems.
Neuroanatomical, Neurofunctional, and Neurocognitive Studies
Computed tomography and magnetic resonance imaging consistently show increased volumes of the lateral and third ventricles in schizophrenic patients. These studies also generally show reductions in overall brain volumes of schizophrenic patients and specific reductions in the size of medial temporal lobe structures, such as the amygdala and hippocampus.
21Moreover, studies have reported size decreases of the thalamus and abnormalities in midline developmental regions.
22None of these changes are specific for schizophrenia, although some have been shown to be present in patients with a first disease episode and no previous medication use.
Functional neuroimaging techniques, such as positron emission tomography (PET), provide in vivo measurements of regional glucose metabolism or cerebral blood flow, both of which reflect regional neuronal activity.
23These techniques have allowed investigations of specific regional metabolic abnormalities in the brains of living schizophrenic patients either at rest or during performance of relevant functional tasks. Most of these studies have detected changes in activity in the prefrontal cortex, basal ganglia structures, temporolimbic regions, and thalamus, suggesting disturbed functioning of corticostriato-thalamo-cortical circuits.
24Decreased activity in the prefrontal cortex of schizophrenic patients is often observed during tasks of cognitive activation and working memory.
25During active auditory hallucinations, abnormal activation of the thalamus, striatum, limbic (especially hippocampus), and paralimbic regions has been detected.
26In a study
27requiring recall of complex narrative material, schizophrenic patients displayed abnormalities in prefrontal, thalamic, and cerebellar sites, suggesting disruption in pontine-cerebellar-thalamic-frontal circuitry.
Historically, based on the observation that clinically therapeutic APDs worked by binding dopamine receptors, the dopaminergic system has been explored extensively (Figure 1). Findings suggest that a complex dopamine dysregulation occurs, with hyperdopaminergic activity in the mesencephalic projections to the limbic striatum and hypodopaminergic activity in the neocortex.
5Evidence for hyperdopaminergic activity has included correlation between the efficacy of dopamine receptor-binding drugs and reduction in positive symptoms as well as increased D2 receptor levels in postmortem and PET studies. Recent studies
5have suggested that various positive symptoms correlate with abnormalities in presynaptic dopamine storage, release, transport, and reuptake in mesolimbic systems. Hypoactivity of dopamine systems is suggested by findings of decreased dopamine turnover in patients with negative symptoms,
28and in some studies
29dopamine agonists have been shown to improve negative symptoms. Functional imaging studies
31also suggest that hypofrontality is more pronounced in patients with negative symptoms.
Serotonergic, glutamatergic, and other neurotransmitter systems (eg, y-aminobutyric acid [GABA]) have been investigated in schizophrenia, especially in reference to interaction with dopaminergic systems. For example, lesioning of rats with the excitotoxin kainic acid produces limbiccortical damage, including hippocampal cell loss, decreases in kainate receptors, decreases in dopamine release in the nucleus accumbens, increases in D2-like dopamine receptors, and an increased behavioral sensitivity to stress.
32In studies of GABAergic systems, decreases in glutamic acid decarboxylase, the GABA-synthesizing enzyme, have been observed in the prefrontal cortex of schizophrenic patients,
34and alterations in subtypes of GABAergic neurons have been reported.
Deficits in functional measures of sensory gating and inhibition have been consistently reported in schizophrenic patients and their family members.
37One of the neuronal mechanisms responsible for such inhibitory gating involves activation of cholinergic nicotinic receptors in the hippocampus. Stimulation of these receptors by nicotine transiently restores inhibitory gating function in schizophrenic patients,
38which could account for the extremely high rate of heavy smoking observed among schizophrenic patients (estimates ranging from 74% to 92% of patients compared with 30% to 35% for the general population).
39Cholinergic nicotinic receptor levels are also diminished in postmortem hippocampal tissue from some schizophrenic patients, making this system yet another candidate for neurochemical defects contributing to the symptoms of schizophrenia.
The opioid system has also been considered a possible candidate for involvement in schizophrenia, based mainly on similarities between the pharmacological effects of opioids and psychiatric symptoms (particularly psychotic symptoms such as hallucinations).
41Hypotheses have been proposed for both increased and decreased levels of various opioid peptides as underlying factors leading to schizophrenic symptoms. However, clinical studies based on these hypotheses have often produced variable results. For example, some studies
42have found reductions in the cerebrospinal fluid levels of enkephalin (from proenkephalin A) in patients with chronic schizophrenia vs controls, whereas other studies
43have reported higher levels of this peptide in the cerebrospinal fluid of patients who received antipsychotic medications.
Neurodevelopmental vs Neurodegenerative Findings
Several neuropathologic findings support a role for neurodevelopmental defects in schizophrenia.
44Gliosis, which is normally associated with adult-onset brain injuries and neurodegenerative conditions, is absent in the brains of schizophrenic patients, indicating a developmental origin for any abnormalities. In addition, cortical cytoarchitectural anomalies have been detected in schizophrenic patients, suggesting abnormal neuronal migration during the second trimester of gestation.
Andreasen et al
45reviewed data from magnetic resonance imaging and PET scans in schizophrenic patients and developed an integrated theory involving dysfunction of cortical-subcortical-cerebellar circuitry as causative in the disorder. Anatomical localization of specific disease symptoms has been attempted. The general conclusions from these data are that negative symptoms may derive from aberrations in the prefrontal cortex, auditory hallucinations may derive from dysfunction in the temporal lobes, and generalized thought disorder may arise from irregularities in the planum temporale.
45Integration of these and other findings has led to the hypothesis that the “diverse symptoms of schizophrenia reflect abnormalities in connectivity in the circuitry that links prefrontal and thalamic regions and where cerebro-cerebellar connectivity may also be disrupted.”
45The abnormality is presumed to be neurodevelopmental in origin, involving perhaps aberrations in neuronal migration, cellular alignment, apoptosis, dendrite and spine formation or pruning, or synapse formation or deletion. Such abnormalities may be inborn (ie, genetic), acquired (ie, environmental), or a combination of genes and environment.
Deranged corticolimbic circuitry, resulting from alterations in brain morphologic findings, such as abnormal arrangement and/or density of neurons in key corticolimbic regions, also has been hypothesized as an underlying cause of schizophrenia.
46More specifically, a shifting of cortical dopamine afferents from pyramidal to nonpyramidal neurons in a specific brain region (anterior cingulate cortex layer II [ACCx-II]) of schizophrenic patients is proposed as contributing to disease symptoms. Dopamine projections to the ACCx-II would be increased with respect to GABAergic interneurons in this model. An inhibitory effect of dopamine on GABAergic cells would result, with downstream derangements in inhibitory and disinhibitory signals. This abnormal circuitry could play a partial role in schizophrenia pathophysiology.
In an additional model hypothesizing limbic-cortical neuronal systems as key to schizophrenia, Csernansky and Bardgett
32state that symptoms of psychosis and disorganized behavior could arise from reduced excitatory glutamatergic inputs from the hippocampus and other limbic structures to the ventral striatum, whereas negative symptoms could be the result of abnormal functioning of frontal lobe structures with connections from limbic structures (eg, dorsolateral prefrontal cortex). Data from animals with limbic-cortical lesions support a decrease in glutamatergic input to the nucleus accumbens and resulting decreases in presynaptic dopamine release and increases in density of dopamine D2-like receptors. Either neuro-developmental abnormalities or later progressive lesions could be the cause of the limbic-cortical abnormalities observed in at least a subgroup of schizophrenic patients.
32O'Donnell and Grace
30summarized data supporting involvement of multiple interrelated systems, including the prefrontal cortex, the medial temporal lobe (including the hippocampus), the ventral striatum, and the mesolimbic dopamine system, in schizophrenia. The authors contend that the varied symptoms of schizophrenia could be the result of “aberrant information processing in subcortical structures that ultimately results in abnormal regulation of cortical activity,” independent of the location of any primary lesion. Dysregulation of thalamocortical activity, in which the thalamus provides a final common pathway for interactions between the basal ganglia and the cortex, is postulated as fundamental in the disease process. Abnormalities in thalamic volumes in schizophrenic patients have been reported, although not consistently.
Weickert and Kleinman
51also extensively reviewed neuroanatomical (including functional studies) and neurochemical data in schizophrenia, concluding that “abnormalities in neural circuits with synaptic pathology involving the hippocampus/entorhinal cortex, the dorsolateral prefrontal cortex, and the striatum/nucleus accumbens” may underlie the pathophysiology of schizophrenia. Furthermore, the accumulated data suggest that “early onset, nonprogressive, structural abnormalities in the temporal lobe and prefrontal cortex, which are linked via pyramidal glutamate neurons, somehow relate to a peripubertal onset of subcortical dopaminergic abnormalities, psychosis, and cognitive deficits.”
51Peripubertal changes in neurotransmitter systems, including dopamine, serotonin, norepi-nephrine, glutamate, GABA, and acetylcholine, may contribute. Reorganization of cortical connections also occurs during the peripubertal period, with substantial regression (pruning) of synapses. Aberrant pruning, perhaps excessive in some areas and inadequate in others, has been hypothesized to underlie schizophrenia.
54Some alterations in correlates of synaptic connectivity (eg, levels of particular synaptic proteins such as synaptophysin) in the brains of schizophrenic patients have been reported.
Emphasizing the variety of structural abnormalities in the brains of schizophrenic patients and the increased rates of obstetric complications and aberrant psychological and neurologic functioning during childhood, Weickert and Weinberger
56suggest that derangements in neurodevelopmental processes contribute to disease pathophysiology. Abnormalities in neuronal cell proliferation, migration, or connectivity, including axonal outgrowth, survival, synaptic regression, and myelination, may be involved. Because these neurodevelopmental processes are fundamentally controlled by neurodevelopmental molecules, Weickert and Weinberger propose that the relevant genes coding for such molecules be examined. Data on aberrant expression or function of developmentally important genes should be integrated with findings from continued studies of structural brain abnormalities, derangements in functional circuitry, and alterations in individual and interrelated neurotransmitter systems. Among the gene families of importance in neurodevelopmental processes are those involved in early pattern formation (HOX, POU family genes), cell proliferation (FGF, EGF family genes), cell migration (immunoglobulin family genes such as neural cell adhesion molecule, reelin), axonal outgrowth (limbic-associated membrane protein, growth-associated protein 43), survival of connections (NGF family, nerve growth factor), programmed cell death (BCL2 family, p53, cyclin D), myelination (myelin basic protein, myelin-promoting factors), and pubertal changes (estrogen and androgen receptors).
57purports that it is “widely accepted that schizophrenia originates from abnormalities occurring during the early stages of neural development,” but argues that data on response to therapy suggest a limited neurodegenerative process in the progression of disease. Reflecting on the fact that most patients respond to treatment during a first episode but later may develop resistance, Lieberman
57suggests that “ongoing deterioration” is brought about by an underlying “degenerative process [that] operates during the active psychotic phase of the illness.” The degenerative process may be incremental in addition to a fundamental neurodevelopmental deficit. Individuals with the neurodevelopmental sign of enlarged ventricular volume are both slower to respond and less responsive to therapy during first episode treatment compared with individuals with normal ventricular size.
58Likewise, patients with a history of obstetric complications have been shown to have a poorer response to first episode therapy.
59Some investigators have found an association between enlarged ventricles and history of obstetric complications, presumably reflecting a common underlying cause. Thus, individuals who become nonresponsive to therapy over time may represent a subset of patients who have both neurodevelopmental insults and continuing neurodegenerative processes.
Thus, clues and hypotheses regarding the pathophysiology of schizophrenia abound, but a clear and consistent picture of underlying neuroanatomical, neurodevelopmental, and neurochemical disease mechanisms has yet to emerge. Likewise, specific causes of the disease remain obscure, with most evidence indicating some complex combination of genetic and environmental effects. Although these various complex factors remain extremely difficult to disentangle, the overwhelming evidence for a substantial genetic contribution to schizophrenia presents an important path forward for our understanding of the disease.
Family, Adoption, and Twin Studies
The classic genetic epidemiological approaches of family, twin, and adoption studies have convincingly established a significant role for genetic inheritance in the etiology of schizophrenia.
60The combined results of numerous European family studies published from 1921 to 1987, as summarized by Gottesman,
61clearly show an increased risk for developing the disease in first-degree relatives of schizophrenic patients. Compared with a lifetime risk for schizophrenia in the general population of approximately 1%, the risk for a sibling or child of a schizophrenic parent is approximately 10%. The risk increases further with the number of affected relatives. For example, an individual has a 16% risk of developing schizophrenia when both a parent and a sibling are affected and a 46% risk when both parents are affected. Increased risk, although still present, decreases sharply for second- and third-degree relatives (5% and 2%, respectively). Many of the early studies have been criticized for methodological problems, such as nonsystematic sampling methods, lack of standardized diagnostic criteria, absence of proper controls, and unblinded assessments.
62More recent studies
64that specifically address these concerns, however, have continued to confirm these risk estimates and show remarkably little variation across differing samples.
Although these studies have decisively shown the familial aggregation of schizophrenia, twin and adoption studies have provided compelling evidence that genetic factors rather than shared family environment account for most of this aggregation. Twin studies rely on comparison of concordance rates (extent to which both members of pairs of twins either do or do not express a trait) between monozygotic and dizygotic twins. Monozygotic twins share 100% of their genetic material, whereas dizygotic twins share on average 50% (as is the case for nontwin siblings), yet both types of twins share environmental influences to approximately the same extent. Therefore, a significantly higher rate of concordance for monozygotic than dizygotic twins indicates genetic transmission of the disorder.
All major twin studies of schizophrenia have confirmed this trend. Both older work, as summarized by Gottesman and Shields,
65and more recent studies using standardized diagnostic criteria
67consistently show 45% to 50% monozygotic concordance compared with 10% to 15% dizygotic concordance rates. Furthermore, studies of disease occurrence among the offspring of nondiscordant monozygotic twins have shown an equally elevated risk of disease development compared with the general population.
68The finding that the offspring of the affected monozygotic twins would have an increased risk of disease was expected. However, the demonstration that the offspring of the unaffected cotwins had similarly elevated risks illustrates the phenomenon of genetic nonpenetrance in which individuals shown or presumed to have deleterious gene(s) do not have the disease phenotype.
Several major adoption studies
73reported increased rates of schizophrenia in individuals adopted from schizophrenic biological parents by healthy families compared with control adoptees from healthy biological parents (approximately 10% vs 1%). These studies also report increased rates of disease in biological relatives of adopted schizophrenic patients compared with adoptive relatives and relatives of control adoptees.
Taken together, these data clearly indicate a substantial genetic contribution to the etiology of schizophrenia, with estimates for the overall heritability of the disease ranging from 63% to 85%.
74The mode of inheritance, however, remains much less clear. Obviously, the overall risks for developing schizophrenia in different classes of relatives do not follow a pattern predicted by simple singlegene mendelian transmission. An autosomal dominant (mendelian) mode of transmission seems to fit certain subgroups of families with multiple affected members.
75Overall, however, most single-gene models for schizophrenia do not adequately fit the data from twin and family studies.
The most widely accepted model for the transmission of schizophrenia, known as the polygenic threshold model,
77describes the inheritance of a predisposition to develop the disorder. According to this theory, the liability to develop the disorder is normally distributed in the population, and this distribution reflects the additive effects of several different genes plus environmental factors. Only those individuals who exceed a certain threshold of liability develop the disease. Relatives of schizophrenic patients have on average an increased liability compared with the general population because of inherited predisposing genetic factors, causing more of these relatives to be beyond the threshold for manifesting the disorder. This concept fits the observed patterns of inheritance and provides explanations for several puzzling features of schizophrenia genetics. However, this theory remains difficult to prove in the absence of any direct links between schizophrenia and specific genes or environmental risk factors. A mixed model of inheritance, in which 1 gene of major effect acts in combination with a background of polygenes each having a small effect, is also not excluded by the data.
Regardless of the exact mode(s) of inheritance and role of environmental factors in schizophrenia (or in other multifactorial disorders such as hypertension, cancer, and diabetes), the identification of susceptibility genes, each presumably of small effect, presents a daunting task. In a recent review,
78a variety of study paradigms, including linkage analysis and positional cloning, case-control association studies, and linkage disequilibrium approaches, are described in terms of their optimal design, feasibility, and statistical power. Descriptions of various study approaches and examples of their application to the study of schizophrenia are described briefly in the following sections.
Linkage Analysis and Chromosomal Studies
Even in the absence of a known mode of inheritance, investigators have attempted to locate genes involved in predisposition to disease using a variety of linkage approaches. Briefly, genes on the same chromosome are linked if inherited together more often than would be expected by chance. Among other factors, increasing physical distance between gene positions along the chromosome (ie, gene loci) increases the probability of crossover events occurring between them and thus decreases the chance of their coordinated inheritance. Calculating the probability of linkage between known chromosomal markers and disease allows the mapping of disease-related genes to specific chromosomal regions. Positional cloning techniques then narrow the region and eventually isolate the gene in question.
Linkage analysis has proved to be a powerful approach for identifying an unknown causative gene by localizing it among markers of known chromosomal position and inheritance patterns. Markers may be of several different classes (eg, tandem repeats, single nucleotide changes) but must be sufficiently varied between individuals so that the pattern of inheritance can be tracked through families. As a simple example, if all individuals with disease in an extended, large family showed a particular marker pattern, whereas all unaffected individuals showed a different pattern, a candidate region for a causative gene would be in the vicinity of the markers shared by affected individuals. Linkage analysis yields a probability estimate that the observed pattern of marker sharing would have occurred by chance alone. Estimates are presented as the logarithm of the odds ratio (LOD) score. Thus, a LOD score of 3.0 signifies 1000:1 odds in favor of linkage between an unknown gene and the marker(s), whereas a LOD score of −2.0 signifies a 100:1 odds against linkage.
Traditional methods of linkage analysis for isolating disease-related genes have been extensively applied to schizophrenia.
79These methods, however, were originally devised and work best for single-gene mendelian disorders. Linkage studies have low power to detect genes that play only a small part in the transmission of disease, adding to the difficulties of applying these methods to complex diseases such as schizophrenia.
80Linkage studies in schizophrenia have used a variety of sampling strategies (eg, affected sibling pairs, nuclear families), sampling pools (eg, geographically and ethnically isolated populations, general population), statistical analyses (eg, parametric in which the mode of inheritance is postulated, nonparametric in which no underlying model is assumed), disease definitions (eg, schizophrenia only, schizophrenia and schizoaffective disorder), disease-associated trait end points (eg, eye tracking abnormalities, evoked auditory potentials), and target genomic regions (eg, genome-wide, candidate chromosomal region). Although significant LOD scores and/or suggestive LOD scores have been reported, in general these have not been replicated consistently. Large collaborative efforts have been undertaken in the hope of generating more conclusive results. As reviewed by Bray and Owen
81and Riley and Williamson,
82linkages with statistically significant LOD scores have been reported from large international collaborative studies for 22q11-q13, 6p24-p22, 8p22-p21, and 6q. Regions for which data are suggestive but not yet fully convincing include 1q21-q22, 5q21-q31, 10p15-p11, and 13q14.1-q32. Additionally, some investigations have shown evidence for shared susceptibility regions (eg, 10p, 10q, 18p) between schizophrenia and affective disorders such as bipolar disorder, whereas others have not.
Chromosomal aberrations cosegregating with schizophrenia have been studied extensively in search of causative genes. A Scottish family cosegregating for a balanced chromosomal translocation (1;11)(q42.1;q14.3) with schizophrenia and related psychiatric disorders has been reported.
87Two novel genes on chromosome 1 were found to be disrupted by the translocation and have been provisionally named disrupted-in-schizophrenia 1 and 2 (DISC1 and DISC2). DISC1 encodes a large protein without significant sequence homology to other known proteins. Based on predicted structure, the novel protein appears to contain a globular N-terminal domain(s) and helical C-terminal domain, a structure that has the potential to form a coiled coil by interaction with other proteins. As reported by the authors, similar structures may be present in a variety of unrelated proteins that function in the nervous system. Interestingly, the second novel gene, DISC2, appears to be a noncoding RNA molecule that is antisense (ie, the complementary sequence) of DISC1 and has been hypothesized to be involved in the regulation of expression of DISC1. The authors concluded that DISC1 and DISC2 may represent candidate genes for susceptibility to psychiatric illness. Whether these genes are in fact susceptibility loci for schizophrenia needs further validation.
Genes mapping to 22q11-q13 have been of interest based on the association of psychotic symptoms in almost 30% of individuals with the velocardiofacial syndrome. The velocardiofacial syndrome is caused by small interstitial deletions in the 22q11 region in 80% to 85% of individuals.
88At least 1% of schizophrenic patients have been found to have deletions in this region.
90Genes mapping to the broader region that have been investigated as candidates in schizophrenia include catechol-O-methyltransferase (COMT),
92ubiquitin fusion degradation 1 protein,
93tissue inhibitor of metalloproteinase 3,
94G protein a subunit gene,
95and synapsin III.
96Some supportive yet inconclusive data have been generated. The most widely studied gene has been COMT and particularly a G to A nucleotide change that results in the substitution of a methionine for valine at codon 158 in the protein product (Val158Met). COMT represents a major degradative pathway for dopamine, and the enzyme produced with methionine has one quarter of the activity of the enzyme containing valine. A multitude of investigations have been conducted to determine whether the variant form of COMT is associated with schizophrenia, subsets of schizophrenia, response to APDs, or any other psychiatric phenotype or pharmacological response, but results have been inconclusive.
A missense change in the WKL1 gene was reported to be associated with periodic catatonia, a familial subtype of catatonic schizophrenia.
92WKL1 maps to 22q13.33 and is a putative, nonselective cation channel expressed exclusively in the brain. In the single pedigree examined, there was a strong segregation of the mutant form of the gene with the periodic catatonia phenotype, despite the fact that some individuals in the family carried the mutation but remained unaffected. Catatonia, a psychomotor disturbance that may be akinetic and stuporous or hyperkinetic and excited, is not specific to schizophrenia and has been observed in other psychiatric disorders.
99Although the finding of a missense change that cosegregates with a familial subtype of schizophrenia is exciting, it also shows the difficulty of identifying genes, especially those of small effect, in a clinically and genetically heterogeneous disorder.
An examination of 9 candidate genes within a 1.5 million-base pair (megabase) region of DNA at 22q11 has resulted in the identification of 2 genes with evidence of association to rare early-onset schizophrenia (before the age of13 years) or schizophrenia with early-onset features (deviant behaviors before the age of 10 years and schizophrenia onset before the age of 18 years).
100Proline dehydrogenase (PRODH2) and the DiGeorge syndrome critical region 6 genes were shown to have marker patterns that were disease associated and replicated in different patient and control samples. Several missense changes that may prevent synthesis of a fully functional PRODH2 enzyme were identified. These data are intriguing and have prompted further analyses of these genes in additional patient groups.
Candidate Gene Association Studies
The candidate gene-based association study approach represents a complementary approach to linkage-based analyses and is particularly well suited to detect genes of small effect.
101These studies simply compare the frequency of specific alleles (1 of several alternative forms of a gene or DNA sequence at a specific chromosomal position or locus) in a sample of unrelated patients with ethnically matched controls. The approach provides an additional advantage in not requiring the availability of affected family members. Association studies may be of 2 distinct forms, often referred to as direct or indirect studies.
102In direct studies, candidate genes are screened for deleterious sequence changes, which then are tested for association with disease by comparing frequencies in affected and unaffected individuals. The candidate gene must first be examined extensively for sequence changes, or other polymorphism data must be available. The potential effect of identified sequence changes on gene function must be inferred (eg, amino acid substitutions in critical protein regions, mutations that cause premature truncation of the protein product, mutations that likely disrupt gene processing) or shown in experimental systems.
In contrast, indirect studies rely on linkage disequilibrium between markers in particular candidate genes or gene regions and disease. Linkage disequilibrium refers to the situation in which alleles occur together more often than can be accounted for by chance, denoting that the alleles are physically close on the same chromosome. The concept is similar to that used in linkage analysis but with the important difference that unrelated cases and controls may be used in linkage disequilibrium studies. If the frequency of a particular marker allele is increased in affected individuals compared with unaffected individuals, the chromosomal region around the marker(s) is more finely investigated to narrow the region and identify causative gene mutations that cosegregate with the marker(s).
Indirect studies have been problematic because of their dependence both on linkage of the unknown disease gene to a particular marker locus and presence of strong linkage disequilibrium between causative mutation(s) in the disease gene and 1 particular marker allele. If multiple independent mutations cause disease, as is often the case for X-linked and autosomal dominant diseases and some autosomal recessive diseases such as phenylketonuria, a single marker allele may not always be associated with disease. This is because some mutations may occur by chance in chromosomes containing a particular marker allele, whereas other mutations may occur in the chromosomes with the other allele(s). Thus, this type of marker-based study may produce negative results even if a correct candidate gene has been chosen. Additionally, because marker allele frequencies may differ significantly among individuals of different ethnic backgrounds, false-positive results can occur due to confounding,
103also referred to as population stratification.
Although linkage disequilibrium-based association studies that rely on a limited set of markers have many potential pitfalls, the concept still holds promise. The key to success may be in more narrowly defining shared alleles on the basis of multiple, densely spaced single nucleotide polymorphisms.
105Single nucleotide polymorphisms are highly abundant in the human genome, with an estimated 1 per 1000 base pairs of DNA. These biallelic markers may be used to construct narrow haplotypes (the pattern of a particular DNA sequence on 1 chromosome that is inherited as a unit) more often shared by affected individuals than unaffected individuals. If dense maps are developed, linkage disequilibrium association analyses may become possible in the general population, not just in isolated, inbred populations. These dense maps are under construction.
For both direct and indirect association studies, candidate genes may be selected from chromosomal regions for which suggestive linkages have been identified, as aforementioned with reference to the chromosome 22 q11-13 region. Genes may also be selected based on known or presumed function, pharmacological relevance, or other biologically based criteria. Obviously, the greatest drawback of this approach is the difficulty of selecting the most likely candidate genes for a disorder whose pathophysiology is poorly understood. Nonetheless, the various hypotheses regarding the biological mechanisms of schizophrenia have provided clues that guide the selection of numerous genes as potential candidates for association with disease. Unfortunately, but not surprisingly, such candidate gene studies for schizophrenia have thus far yielded only negative, contradictory, or unreplicated results.
Genes encoding components of the dopamine system have historically been at the top of the list of prime candi dates for association with schizophrenia. However, numerous studies
108of all 5 dopamine receptor genes, as well as other genes involved in dopamine transport or metabolism, have failed to show any clear associations with schizophrenia. One of the more promising results has stemmed from reports of a small increase in susceptibility to schizophrenia associated with homozygosity for a Ser-9-Gly polymorphism in the D3 receptor gene.
109However, more recent work has failed to replicate this finding.
Genes involved in other neurotransmitter systems known to interact with dopamine pathways or otherwise implicated in the biochemistry of schizophrenia, as well as several genes important in neurodevelopment, also have been explored in association studies. Among these various genes tested, a few promising leads have been uncovered. For example, weak association with disease has been shown for the nerve growth factor neurotrophin 3 in some studies,
111although not others.
112A C to T silent polymorphism in the serotonin receptor gene (most likely in linkage disequilibrium with an unidentified functional variant) has also been linked to disease in some reports.
113Interestingly, this polymorphism seems to predict response to clozapine,
115although this finding also awaits replication. Genes encoding various subunits of glutamate receptors also have been examined for association with disease or pharmacological response, without major positive findings.
119Other genes that have been investigated extensively but for which strong disease associations have not been consistently found include developmentally related genes such as NOTCH4,
123ion channel genes such as hSKCa3, KCNN3, GLRA2, and the a-7 nicotinic receptor,
128as well as a variety of genes in the major histocompatibility complex.
129Many of the investigated genes contain short repetitive DNA sequences, known as trinucleotide repeats, as described more fully in the following section.
Trinucleotide Repeat Analyses
An additional model for the genetic transmission of schizophrenia has arisen from the identification of a new class of human mutation-heritable unstable DNA. In the human genome, repetitive DNA sequences are abundant. Certain types of these repetitive sequences (eg, those containing multiple repeats of a 3 DNA base [trinucleotide] unit such as CAG or CCG) are inherently unstable, and both expansions and contractions of the number of repeating 3 base units can occur. The significant expansion of specific trinucleotide repeat sequences has been discovered as the underlying cause of several neurologic or neurodegenerative diseases, including Huntington disease, myotonic dystrophy, and types of spinocerebellar ataxia and fragile X syndrome. Importantly, this trinucleotide repeat expansion was proved to be the biological basis of the clinically observed phenomenon known as genetic anticipation, which is characterized by an earlier age at onset and increased severity of illness in successive generations of affected individuals.
Anticipation was originally described for families with psychosis and mental retardation at the turn of the century
131and was subsequently observed in other neurologic diseases, such as Huntington disease and myotonic dystrophy. This clinical feature was dismissed for many decades, however, as a statistical artifact resulting from ascertainment bias.
132Ascertainment biases that lead to incorrect conclusions of anticipation can arise for several reasons. For example, reduced reproductive fitness in individuals with severe early-onset illness results in the preferential selection of parents with later disease onset. Children with younger age at onset can also be overrepresented in the sample because these more severe cases are more likely to be treated and identified for study. The recruitment of affected parent-offspring pairs during a limited time frame favors the selection of pairs with parents with later-onset disease and offspring with early-onset disease because the offspring may not have had time to develop late-onset disease.
Nevertheless, the advent of modern molecular biology techniques provided conclusive evidence for the occurrence of true genetic anticipation in certain neurologic diseases. These methods uncovered the molecular equivalent of anticipation in the form of trinucleotide repeat expansions. For diseases linked to genes containing these repeat expansions, the degree of expansion varies across generations of affected families and correlates with clinical symptoms. An increase in the number of trinucleotide repeats correlates with an increase in the severity of the disorder and a decrease in the age at onset in younger generations. This finding has rekindled interest in reports of anticipation observed in families showing strong inheritance of neuropsychiatric diseases, such as schizophrenia and bipolar disorder.
Reevaluation of the inheritance patterns of major psychoses has suggested that the unstable DNA model represents a good fit for the twin and family epidemiological data of schizophrenia and affective disorders.
133In fact, this model competes well with the traditional multifactorial polygenic theory, and the nonmendelian behavior of unstable DNA could explain some of the complex, unresolved facets of psychiatric genetics. For example, unstable DNA can exist in a premutation form, in which the number of repeats is higher than normal but just below the disease threshold, and can undergo somatic mutation in early embryogenesis. These factors could account for the puzzling observation of identical rates of schizophrenia in the offspring of discordant monozygotic twins.
68If the initial inherited repeat number was just below the threshold for disease, then unequal amounts of expansion could lead to disease in 1 twin but not the other. Further repeat expansion in the offspring of the unaffected twin could then lead to the expression of the disease in that individual as well.
However, the complex nature of schizophrenia and its inheritance have impeded the task of definitively demonstrating the occurrence of true genetic anticipation in this disease. Nonetheless, evidence for anticipation in some families displaying strong aggregation of schizophrenia continues to mount.
134Several recent studies of families with multiple affected members have been conducted to test for anticipation while specifically addressing the major sources of ascertainment bias. For example, 1 study
135investigated 8 Eastern Canadian families (186 total members) with many schizophrenic relatives for the presence of anticipation. This study found significant decreases in age at onset and increased severity across generations. Figure 2 shows a representative pedigree for one of these families. Overall, the study reported a mean age at first hospitalization of 41 years for the grandparental generation, 34 years for the parental generation, and 26 years for the index generation. An increase in the severity of disease was also observed based on rates of hospitalization for psychosis and the proportion of such hospitalizations compared with schizotypal conditions across successive generations.
These results have been confirmed by various additional studies analyzing large numbers of European and American families and using similar methods to limit ascertainment biases.
137A reanalysis of archival data from the extensive familial mental illness study by Penrose has also given weight to the argument for true anticipation. This study was able to control for ascertainment bias by taking advantage of a large data set (7339 individuals from 3109 families) ascertained during an extended time frame (70 years) and including aunt/uncle-niece/nephew pairs of affected individuals in addition to parent-offspring pairs.
139Therefore, the growing consensus that true genetic anticipation occurs in at least some highly familial forms of schizophrenia has heightened interest in the unstable DNA hypothesis. According to this hypothesis, a still unidentified trinucleotide repeat expansion, either singly or in combination with other genetic factors, may account for the inheritance of schizophrenia. This adds yet another approach to searching for specific genes associated with this disease.
Finally, evidence for genetic anticipation in families of schizophrenic patients has led to the development and implementation of novel molecular approaches for identifying genes associated with schizophrenia. The long stretches of trinucleotide repeats associated with anticipation provide a specific feature of DNA with defined sequences for which to look. This gives methods targeted toward these sequences distinct advantages over standard linkage mapping techniques. Furthermore, searching specifically for genes containing trinucleotide repeat expansions would require no a priori knowledge of pathophysiology, in contrast to candidate gene-based approaches. If the underlying hypothesis is correct, then methods aimed at detecting these long repetitive sequences in the genome, isolating the genes containing these sequences, and assessing size expansion in association with disease may shed new light on genetic mechanisms of schizophrenia.
Preliminary support for the presence of expanded trinucleotide repeat sequences in the DNA of schizophrenic patients has been reported. With use of a method referred to as repeat expansion detection analysis,
140significant shifts in the distribution of CTG repeats have been shown in groups of schizophrenic patients compared with controls, with the patient group having evidence of expansions.
143However, increases were modest and overlapped with the size of repeats found in unaffected individuals. Also, although the repeat expansion detection technique is able to detect the presence of expanded repeats, no information is provided on the underlying gene(s) in which the expansion occurred or the gene(s) chromosomal location.
With a view toward identifying genes containing trinucleotide repeat expansions, other methods have been developed and implemented in large screening efforts. Lists of novel CAG repeat-containing genes have been compiled as potential candidates for association studies.
146However, these studies are limited in their specificity for expanded vs normal repeats, in the type of repeat that can be detected (ie, CAGs), and in their ability to detect repeats located within coding regions of genes only. Repeat expansions within the nonprotein coding portions of the gene (introns) have been shown to cause disease (eg, GAA repeat expansion in intron 1 of the frataxin gene causing Friedreich ataxia).
To overcome some of these difficulties, a method was developed to directly detect and then identify the gene containing repeat expansions directly from the DNA of patients. This method, termed DIRECT for direct identification of repeat expansion and cloning technique, was used to successfully identify a CAG repeat expansion causing spinocerebellar ataxia type 2.
148The DIRECT technique could potentially be applied to the detection of other pathologic repeat expansions. The approach has definite advantages for schizophrenia studies in that it requires no prior knowledge of chromosomal locations or pathologic functions of causative genes. Limitations include the amount of sample DNA needed for adequate sensitivity and a variable ability to unequivocally identify disease-specific repeat expansions from amid multiple repeat sequences found in both healthy and diseased individuals. Thus, the ability to adequately resolve the deleterious, disease-related expansions depends on multiple experimental design factors, the sizes of repeat expansions, and other parameters.
Gene Expression Studies
A recently developed genetic strategy is that of gene expression analysis. Through expression analysis, a profile of which genes are activated at which time and at what levels can ultimately be obtained. If differences in gene expression levels are identified between diseased and healthy individuals, the involved genes may represent candidates for disease causation, progression, response to therapy, or other factors, depending on the experimental design. A variety of techniques are available for the analysis of gene expression patterns, including arrays in which the messenger RNA (mRNA) for tens of thousands of genes may be simultaneously measured from target tissue. Messenger RNA is the single-stranded template that is produced (transcribed) from a gene's DNA as an intermediate step for those genes coding for protein products (translation). However, expressed genes include not only those that are transcribed and then translated into protein but also those that are transcribed into RNA but not translated into protein (eg, transfer and ribosomal RNAs).
In studies of schizophrenia, postmortem brain tissue from particular regions (eg, hippocampus, thalamus, prefrontal cortex) of diseased and healthy individuals (usually matched for age, sex, time from death to tissue analysis, tissue preparation techniques, and other variables) has been analyzed. In a recently published study
149using prefrontal cortex samples from matched pairs of patients with schizophrenia and control subjects, mRNA transcripts encoding proteins involved in the regulation of presynaptic function were decreased in all patients with schizophrenia. These genes also displayed a different combination of decreased expression in these patients. When other genes not involved in presynaptic function were evaluated, no alterations in expression were found. Selected presynaptic function gene microarray observations were verified by other experiments that used alternate approaches. Two genes demonstrating the most consistently altered mRNA expression levels were N-ethylmaleimide-sensitive factor and synapsin II. Both are involved in presynaptic functioning, and the authors concluded that the combined data suggested that patients with schizophrenia share a common abnormality in presynaptic function. Increased expression of genes encoding synaptic glutamatergic transporters in thalami from schizophrenic patients compared with controls also has been reported.
150In other studies of thalamic nuclei in schizophrenic patients and controls, altered ionotropic glutamate receptor binding and reduced mRNA expression levels in the thalamus were reported for various subunit genes, including NMDAR1, NMDAR2B, NMDAR2C, gluR1, gluR3, and KA2. The authors noted that the differences were most prominent in nuclei with reciprocal projections to limbic regions.
151The observed abnormalities in N-methyl-D-aspartate, 2-amino-3-hydroxy-5-methyl-4-isoxazole proprionic acid, and kainate receptor expression in limbic thalamus were suggestive of an N- methyl-D-aspartate receptor hypoactivity in schizophrenia and were consistent with diminished glutamatergic activity in the thalamus. Alternatively, the results could suggest abnormal glutamatergic innervation in afferent and/or efferent regions. In an analysis of prefrontal cortex tissue, up-regulation of apolipoprotein genes L1, L2, and L4 was shown and replicated in additional samples from schizophrenic patients from distinct populations (Japan and New Zealand).
152The apolipoprotein gene cluster (L1-L6) has been mapped to chromosome 22q12.3, a region previously linked in some studies of schizophrenia and near the deletion region in VCFS (22q11).
Expression analysis, although extremely valuable as yet another approach to identifying candidate genes, is limited by the availability of appropriate target tissue, its temporal nature (genes expressed at the time of death), genes represented on the array, and other factors, including the possible effects of antipsychotic medication on gene expression. The application of gene expression analysis to animal models of aberrant behavior, in which multiple experimental parameters may be controlled, may yield insights into patterns of gene expression during development and puberty, antipsychotic treatment, stressful conditions, and other paradigms. Not only may particular candidate genes be identified but also entire pathways of sequential and coordinated gene expression may be elucidated for intensive study in schizophrenic patients.
SUMMARY AND FUTURE DIRECTIONS
Despite clear evidence for a genetic contribution to the development of schizophrenia, previously tested hypotheses and current methods have thus far failed to convincingly identify specific genetic factors that cause or predispose individuals to the disorder. We hope that active research in neuroanatomy, functional neuroanatomy, neuropathology, neurochemistry, neuropsychology, neurodevelopment, molecular neuroscience, and molecular epidemiology will provide clues about underlying pathophysiology and gene systems of interest. Heterogeneity of the underlying genetic and etiologic factors is the expectation, with neurodevelopmental and possibly neurodegenerative processes involved. Suspect candidate genes, either currently known or yet to be identified, may play a role not only in disease development and progression but also in the mediation of response to therapy.
- Cambridge University Press, Cambridge, England1996: 35 The History of Mental Symptoms: Descriptive Psychopathology Since the Nineteenth Century.
- The costs of schizophrenia: assessing the burden.Psychiatr Clin North Am. 1993; 16: 413-423
- What we know about homelessness among mentally ill persons: an analytical review and commentary.in: Lamb HR Bachrach LL Kass FI Treating the Homeless Mentally Ill. American Psychiatric Association, Washington, DC1992: 13-40
- Schizophrenia.in: Tasman A Kay J Lieberman JA Psychiatry. Vol 2. WB Saunders Co, Philadelphia, Pa1997: 927-965
- Catching up on schizophrenia: natural history and neurobiology.Neuron. 2000; 28: 325-334
- American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. 4th ed. American Psychiatric Association, Washington, DC1994
- Schizophrenia.in: Rubenstein E Federman DD Scientific American Medicine. Vol 3. Chapter 13, Section VII. Scientific American, Inc, New York, NY2000
- Epidemiology of schizophrenia.Can J Psychiatry. 1997; 42: 139-151
- Gender differences in affective, schizoaffective, and schizophrenic disorders.Schizophr Bull. 1990; 16: 319-329
- Gender differences in the clinical expression of schizophrenia.Schizophr Res. 1992; 7: 225-231
- Effects of occupational stress on the physical and psychological health of women in a microelectronics plant.Soc Sci Med. 1992; 34: 1377-1383
- Gender differences in schizophrenia.J Nerv Ment Dis. 1995; 183: 522-528
- First-episode schizophrenia, I: early course parameters.Schizophr Bull. 1996; 22: 241-256
- Gender differences in onset of illness, treatment response, course, and biologic indexes in first-episode schizophrenic patients.Am J Psychiatry. 1995; 152: 698-703
- Sex differences in neuroanatomical and clinical correlations in schizophrenia.Am J Psychiatry. 1996; 153: 799-805
- Schizophrenia throughout life: sex differences in severity and profile of symptoms.Schizophr Res. 1996; 21: 1-12
- Atypical antipsychotics in the treatment of schizophrenia: systematic overview and meta-regression analysis.BMJ. 2000; 321: 1371-1376
- Atypical antipsychotics [editorial].BMJ. 2000; 321: 1360-1361
- Schizophrenia: improving outcome.Harv Rev Psychiatry. 1999; 6: 229-240
- Schizophrenia.Med Clin North Am. 2001; 85: 663-689
- Brain abnormality in schizophrenia: a systematic and quantitative review of volumetric magnetic resonance imaging studies.Br J Psychiatry. 1998; 172: 110-120
- The role of the thalamus in schizophrenia.Can J Psychiatry. 1997; 42: 27-33
- Positron-emission tomography and regional brain metabolism in schizophrenia research.in: Volkow ND Wolf AP Positron-Emission Tomography in Schizophrenia Research. American Psychiatric Press, Washington, DC1991: 27-46
- Cortical-striatal-thalamic circuits and brain glucose metabolic activity in 70 unmedicated male schizophrenic patients.Am J Psychiatry. 1993; 150: 1325-1336
- Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia, I: regional cerebral blood flow evidence.Arch Gen Psychiatry. 1986; 43: 114-124
- A functional neuroanatomy of hallucinations in schizophrenia.Nature. 1995; 378: 176-179
- Schizophrenia and cognitive dysmetria: a positron-emission tomography study of dysfunctional prefrontal-thalamic-cerebellar circuitry.Proc Natl Acad Sci U S A. 1996; 93: 9985-9990
- Biochemical findings of negative symptoms in schizophrenia and their putative relevance to pharmacologic treatment: a review.Neuropsychobiology. 1994; 30: 160-172
- Amphetamine and negative symptoms of schizophrenia.Psychopharmacology (Berl). 1996; 123: 211-214
- Dysfunctions in multiple interrelated systems as the neurobiological bases of schizophrenic symptom clusters.Schizophr Bull. 1998; 24: 267-283
- A separate disease within the syndrome of schizophrenia.Arch Gen Psychiatry. 2001; 58: 165-171
- Limbic-cortical neuronal damage and the pathophysiology of schizophrenia.Schizophr Bull. 1998; 24: 231-248
- Decreased glutamic acid decarboxylase67 messenger RNA expression in a subset of prefrontal cortical gamma-aminobutyric acid neurons in subjects with schizophrenia.Arch Gen Psychiatry. 2000; 57: 237-245
- Reelin and GAD67 downregulation and psychosis vulnerability.Biol Psychiatry. 2000; 47: S68
- Neurochemical correlates of cortical GABAergic deficits in schizophrenia: selective losses of calcium binding protein immunoreactivity.Brain Res Bull. 2001; 55: 579-584
- Alterations in chandelier neuron axon terminals in the prefrontal cortex of schizophrenic subjects.Am J Psychiatry. 1999; 156: 1709-1719
- Gating and habituation deficits in the schizophrenia disorders.Clin Neurosci. 1995; 3: 131-139
- Normalization by nicotine of deficient auditory sensory gating in the relatives of schizophrenics.Biol Psychiatry. 1992; 32: 607-616
- Cigarette smoking in schizophrenia: relationship to psychopathology and medication side effects.Am J Psychiatry. 1992; 149: 1189-1194
- Schizophrenia and nicotinic receptors.Harv Rev Psychiatry. 1994; 2: 179-192
- Endorphins and schizophrenia.Prog Brain Res. 1992; 93: 433-453
- Met-enkephalin level in the cerebrospinal fluid of schizophrenic patients.Clin Chim Acta. 1983; 128: 367-371
- Methionine(Met)-enkephalin and leucine(Leu)-enkephalin in CSF of chronic schizophrenics: correlation with psychopathology.Protides Biological Fluids. 1984; 32: 175-178
- From neuropathology to neurodevelopment.Lancet. 1995; 346: 552-557
- “Cognitive dysmetria” as an integrative theory of schizophrenia: a dysfunction in cortical-subcortical-cerebellar circuitry?.Schizophr Bull. 1998; 24: 203-218
- Model generation and testing to probe neural circuitry in the cingulate cortex of postmortem schizophrenic brain.Schizophr Bull. 1998; 24: 219-230
- Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenics.Arch Gen Psychiatry. 1990; 47: 1023-1028
- Thalamic abnormalities in schizophrenia visualized through magnetic resonance image averaging.Science. 1994; 266: 294-298
- Cortical development and thalamic pathology in schizophrenia.Schizophr Bull. 1997; 23: 483-501
- Structural magnetic resonance image averaging in schizophrenia.Am J Psychiatry. 1998; 155: 1064-1073
- The neuroanatomy and neurochemistry of schizophrenia.Psychiatr Clin North Am. 1998; 21: 57-75
- Synaptic density in human frontal cortex-developmental changes and effects of aging.Brain Res. 1979; 163: 195-205
- Schizophrenia: caused by a fault in programmed synaptic elimination during adolescence?.J Psychiatr Res. 1982-83; 17: 319-334
- Changes in caudate volume with neuroleptic treatment [letter].Lancet. 1994; 344: 1434
- Development of the prefrontal cortex during adolescence: insights into vulnerable neural circuits in schizophrenia.Neuropsychopharmacology. 1997; 16: 385-398
- A candidate molecule approach to defining developmental pathology in schizophrenia.Schizophr Bull. 1998; 24: 303-316
- Pathophysiologic mechanisms in the pathogenesis and clinical course of schizophrenia.J Clin Psychiatry. 1999; 60: 9-12
- Time course and biologic correlates of treatment response in first-episode schizophrenia.Arch Gen Psychiatry. 1993; 50: 369-376
- Obstetric complications predict treatment response in first-episode schizophrenia.Psychol Med. 1999; 29: 621-627
- Genetic basis of schizophrenia.Lancet. 1995; 346: 678-682
- Schizophrenia Genesis: The Origins of Madness. WH Freeman, New York, NY1991
- The genetics of schizophrenia: a current, genetic-epidemiologic perspective.Schizophr Bull. 1993; 19: 261-285
- Psychiatric illness in first-degree relatives of schizophrenic and surgical control patients: a family study using DSM-III criteria.Arch Gen Psychiatry. 1985; 42: 770-779
- The Roscommon Family Study I: methods, diagnosis of probands, and risk of schizophrenia in relatives.Arch Gen Psychiatry. 1993; 50: 527-540
- Schizophrenia, the Epigenetic Puzzle. Cambridge University Press, Cambridge, England1982
- Twin concordance and DSM-III schizophrenia: scrutinizing the validity of the definition.Arch Gen Psychiatry. 1987; 44: 634-641
- Twin concordance for DSM-III-R schizophrenia.Acta Psychiatr Scand. 1991; 83: 395-401
- Confirming unexpressed genotypes for schizophrenia: risks in the offspring of Fischer's Danish identical and fraternal discordant twins.Arch Gen Psychiatry. 1989; 46: 867-872
- Psychiatric disorders in foster home reared children of schizophrenic mothers.Br J Psychiatry. 1966; 112: 819-825
- Mental illness in the biological and adoptive relatives of schizophrenic adoptees: findings relevant to genetic and environmental factors in etiology.Am J Psychiatry. 1983; 140: 720-727
- The inheritance of schizophrenia spectrum disorders: a reanalysis of the Danish adoptee study data.Am J Psychiatry. 1983; 140: 1167-1171
- Mental illness in the biological and adoptive relatives of schizophrenic adoptees: replication of the Copenhagen Study in the rest of Denmark.Arch Gen Psychiatry. 1994; 51: 442-455
- Interaction between genetic vulnerability and family environment: the Finnish adoptive family study of schizophrenia.Acta Psychiatr Scand. 1991; 84: 460-465
- The strength of the genetic effect: is there room for an environmental influence in the aetiology of schizophrenia?.Br J Psychiatry. 1994; 164: 593-599
- Does early onset characterize autosomal dominant schizophrenia? [abstract].Schizophr Res. 1993; 9: 118-119
- Genetic linkage and complex diseases, with special reference to psychiatric disorders.Genet Epidemiol. 1990; 7: 3-16
- A polygenic theory of schizophrenia.Proc Natl Acad Sci U S A. 1967; 58: 199-205
- Searching for genetic determinants in the new millennium.Nature. 2000; 405: 847-856
- The maddening hunt for madness genes.Nat Genet. 1997; 17: 127-129
- The molecular genetics of schizophrenia: progress so far.Mol Med Today. 1997; 3: 124-130
- Searching for schizophrenia genes.Trends Mol Med. 2001; 7: 169-174
- Sane genetics for schizophrenia.Nat Med. 2000; 6: 253-255
- Do schizophrenia and affective disorder share susceptibility genes?.Schizophr Res. 1999; 39: 107-111
- No support for linkage to the bipolar regions on chromosomes 4p, 18p, or 18q in 43 schizophrenia pedigrees.Am J Med Genet. 2000; 96: 224-227
- A genome survey indicates a possible susceptibility locus for bipolar disorder on chromosome 22.Proc Natl Acad Sci U S A. 2001; 98: 585-590
- Are schizophrenic and bipolar disorders related? a review of family and molecular studies.Biol Psychiatry. 2000; 48: 531-538
- Disruption of two novel genes by a translocation co-segregating with schizophrenia.Hum Mol Genet. 2000; 9: 1415-1423
- Searching for susceptibility genes in schizophrenia.Eur Neuropsychopharmacol. 2001; 11: 395-398
- Screening for 22q11 deletions in a schizophrenia population.Schizophr Res. 2001; 52: 167-170
- Velo-cardio-facial syndrome: a model for understanding the genetics and pathogenesis of schizophrenia.Br J Psychiatry. 2001; 179: 397-402
- Association study of a functional catechol-O-methyltransferase gene polymorphism in Japanese schizophrenics.Neurosci Lett. 1998; 243: 109-112
- A missense mutation in a novel gene encoding a putative cation channel is associated with catatonic schizophrenia in a large pedigree.Mol Psychiatry. 2001; 6: 302-306
- Association study of a promoter polymorphism of UFD1L gene with schizophrenia.Am J Med Genet. 2001; 105: 529-533
- Systematic search for mutations in the human tissue inhibitor of metalloproteinases-3 (TIMP-3) gene on chromosome 22 and association study with schizophrenia.Am J Med Genet. 2001; 105: 275-278
- Analysis of GNAZ gene polymorphism in bipolar affective disorder.Am J Med Genet. 1999; 88: 324-328
- A third member of the synapsin gene family.Proc Natl Acad Sci U S A. 1998; 95: 4667-4672
- Identification of sequence variants and analysis of the role of the catechol-O-methyltransferase gene in schizophrenia susceptibility.Biol Psychiatry. 1998; 43: 425-431
- Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia.Proc Natl Acad Sci U S A. 2001; 98: 6917-6922
- Genes and subtypes of schizophrenia.Trends Mol Med. 2001; 7: 281-283
- Genetic variation at the 22q11 PRODH2/DGCR6 locus presents an unusual pattern and increases susceptibility to schizophrenia.Proc Natl Acad Sci U S A. 2002; 99: 3717-3722
- Novel association approach for determining the genetic predisposition to schizophrenia: case-control resource and testing of a candidate gene.Am J Med Genet. 1993; 48: 28-35
- Variations on a theme: cataloging human DNA sequence variation.Science. 1997; 278: 1580-1581
- Case-control Studies: Design, Conduct, Analysis. Oxford University Press, New York, NY1982
- The new genomics: global views of biology.Science. 1996; 274: 536-539
- Prospects for whole-genome linkage disequilibrium mapping of common disease genes.Nat Genet. 1999; 22: 139-144
- No structural mutation in the dopamine D2 receptor gene in alcoholism or schizophrenia: analysis using denaturing gradient gel electrophoresis.JAMA. 1994; 271: 204-208
- Overview: a current perspective on twin studies of schizophrenia.Am J Psychiatry. 1983; 140: 1413-1425
- Genetics and schizophrenia.in: Hirsh SR Weinberger DR Schizophrenia. Blackwell Science, Oxford, England1995: 253-274
- Allelic association between a Ser-9-Gly polymorphism in the dopamine D3 receptor gene and schizophrenia.Hum Genet. 1996; 97: 714-719
- The dopamine D3 receptor (DRD3) Ser9Gly polymorphism and schizophrenia: a haplotype relative risk study and association with clozapine response.Mol Psychiatry. 1998; 3: 72-75
- Schizophrenia and neurotrophin-3 alleles.Acta Psychiatr Scand. 1997; 95: 414-419
- Neurotrophin-3 gene polymorphisms and schizophrenia: no evidence for linkage or association.Psychiatr Genet. 1996; 6: 183-186
- Association between schizophrenia and T102C polymorphism of the 5-hydroxytryptamine type 2a-receptor gene.Lancet. 1996; 347: 1294-1296
- Association between clozapine response and allelic variation in 5-HT2A receptor gene.Lancet. 1995; 346: 281-282
- Evidence for association between polymorphisms in the promoter and coding regions of the 5-HT2A receptor gene and response to clozapine.Mol Psychiatry. 1998; 3: 61-66
- Identification of single nucleotide polymorphisms (SNPs) and other sequence changes and estimation of nucleotide diversity in coding and flanking regions of the NMDAR1 receptor gene in schizophrenic patients.Mol Psychiatry. 2001; 6: 274-284
- Association analysis for NMDA receptor subunit 2B (GRIN2B) genetic variants and psychopathology and clozapine response in schizophrenia.Psychiatr Genet. 2001; 11: 219-222
- Mutation screening of the metabotropic glutamate receptor mGluR4 (GRM4) gene in patients with schizophrenia.Psychiatr Genet. 2001; 11: 79-83
- Mutation analysis of the N-methyl-D-aspartate receptor NR1 subunit gene (GRIN1) in schizophrenia.Neurosci Lett. 2000; 296: 168-170
- A family-based and case-control association study of the NOTCH4 gene and schizophrenia.Mol Psychiatry. 2002; 7: 100-103
- Association analysis of NOTCH4 loci in schizophrenia using family and population-based controls.Nat Genet. 2001; 28: 126-128
- Failure to confirm NOTCH4 association with schizophrenia in a large population-based sample from Scotland.Nat Genet. 2001; 28: 128-129
- The NOTCH4 locus is associated with susceptibility to schizophrenia.Nat Genet. 2000; 25: 376-377
- Nuclear localization and dominant-negative suppression by a mutant SKCa3 N-terminal channel fragment identified in a patient with schizophrenia.J Biol Chem. 2001; 276: 27753-27756
- Association study of CAG repeats in the KCNN3 gene in Japanese patients with schizophrenia, schizoaffective disorder and bipolar disorder.Psychiatry Res. 2001; 101: 203-207
- Systematic screening for mutations in the glycine receptor alpha2 subunit gene (GLRA2) in patients with schizophrenia and other psychiatric diseases.Psychiatr Genet. 2001; 11: 45-48
- Inhibitory neurophysiological deficit as a phenotype for genetic investigation of schizophrenia.Am J Med Genet. 2000; 97: 58-64
- Linkage disequilibrium for schizophrenia at the chromosome 15q13-14 locus of the alpha7-nicotinic acetylcholine receptor subunit gene (CHRNA7).Am J Med Genet. 2001; 105: 20-22
- Schizophrenia and HLA: a review.Schizophr Res. 2001; 47: 1-12
- DNA secondary structure: a common and causative factor for expansion in human disease.Proc Natl Acad Sci U S A. 1999; 96: 1823-1825
- The Huxley lecture on hereditary aspects of nervous and mental diseases.BMJ. 1910; 2: 1013-1020
- The problem of anticipation in pedigrees of dystrophia myotonica.Ann Eugenics. 1948; 14: 125-132
- Unstable genes—unstable mind?.Am J Psychiatry. 1995; 152: 164-172
- Anticipation in schizophrenia: biology or bias?.Am J Med Genet. 1997; 74: 275-280
- Evidence for anticipation in schizophrenia.Am J Hum Genet. 1994; 54: 864-870
- Age of onset in sporadic and familial schizophrenia [abstract].Schizophr Res. 1994; 11: 101-102
- Further epidemiological evidence for anticipation in schizophrenia.Biomed Pharmacother. 1997; 51: 376-380
- Survey of cases of familial mental illness.Eur Arch Psychiatry Clin Neurosci. 1991; 240: 315-324
- Anticipation or ascertainment bias in schizophrenia? Penrose's familial mental illness sample.Am J Hum Genet. 1997; 60: 630-637
- Direct detection of novel expanded trinucleotide repeats in the human genome.Nat Genet. 1993; 4: 135-139
- CAG repeat expansions and schizophrenia: association with disease in females and with early age-at-onset.Hum Mol Genet. 1995; 4: 1957-1961
- Expanded CAG repeats in schizophrenia and bipolar disorder.Nat Genet. 1995; 10: 380-381
- Confirmation of association between expanded CAG/CTG repeats and both schizophrenia and bipolar disorder.Psychol Med. 1996; 26: 1145-1153
- Novel triplet repeat containing genes in human brain: cloning, expression, and length polymorphisms.Genomics. 1993; 16: 572-579
- Survey of CAG/CTG repeats in human cDNAs representing new genes: candidates for inherited neurological disorders.Hum Mol Genet. 1996; 5: 1001-1009
- CAG/CTG and CGG/GCC repeats in human brain reference cDNAs: outcome in searching for new dynamic mutations.Genomics. 1998; 47: 414-418
- Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion.Science. 1996; 271: 1423-1427
- Identification of the spinocerebellar ataxia type 2 gene using a direct identification of repeat expansion and cloning technique, DIRECT.Nat Genet. 1996; 14: 277-284
- Molecular characterization of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex.Neuron. 2000; 28: 53-67
- Vesicular glutamate transporter transcript expression in the thalamus in schizophrenia.Neuroreport. 2001; 12: 2885-2887
- Ionotropic glutamate receptor binding and subunit mRNA expression in thalamic nuclei in schizophrenia.Am J Psychiatry. 2000; 157: 1811-1823
- Gene expression analysis in schizophrenia: reproducible up-regulation of several members of the apolipoprotein L family located in a high-susceptibility locus for schizophrenia on chromosome 22.Proc Natl Acad Sci U S A. 2002; 99: 4680-4685
© 2002 Mayo Foundation for Medical Education and Research. Published by Elsevier Inc. All rights reserved.