Chapter 5 - Heritability of Structural Brain Traits: An Endophenotype Approach to Deconstruct Schizophrenia

https://doi.org/10.1016/S0074-7742(09)89005-3Get rights and content

Structural brain phenotypes are quantitative traits showing considerable variation in human populations. Quantitative structural brain abnormalities are also repeatedly reported in patients with psychiatric disorders such as schizophrenia. Studying the genetic and environmental causes of these differences might therefore highlight biological mechanisms underlying neuroanatomical phenotypes and directly result in the identification of risk factors for schizophrenia. Heritability estimates indicate a strong genetic component contributing to neuroanatomical phenotypes. Brain structure volumes have substantial heritability rates ranging from high (70–95%) for total brain volume, cerebellar, gray and white matter, and corpus callosum, to moderate (40–70%) determined for the hippocampus, the four lobes (frontal, temporal, occipital, and parietal lobe), temporal horn volume, brain parenchyma, white matter hyperintensity, and planum temporal asymmetry. Middle structures of the brain show high heritability scores for the deeper structures (ontogenetically earlier formed) and moderate heritability scores for the surface structures. Structures formed earlier in development show consistently higher heritability rates than brain structures formed later in development, for example, surface structures, which seem to be influenced by environmental factors. Even higher heritability reaching 0.99 for total brain volume are estimated in nonhuman primate (NHP) models employing inbred extended pedigree and highly uniform rearing conditions, reducing the effects of environmental factors. Applying highly heritable structural brain phenotypes may serve as an endophenotype for gene mapping studies and lead to identification of genes that are involved in the regulation of human brain volume and the biological mechanisms involved in the causal mechanisms of psychiatric disorders.

Introduction

Schizophrenia is a highly heritable disorder (Cannon et al., 1998, Cardno and Gottesman, 2000, Sullivan et al., 2003) characterized by delusions, hallucinations, disorganized speech (frequent derailment or incoherence), grossly disorganized or catatonic behavior, and negative symptoms, that is, affective flattening, alogia (poverty of speech) and avolition (general lack of desire, interest or motivation to pursue meaningful goals) (DSM-IV-Tr, 2000). It is believed that this complex phenotype of schizophrenia is a result of underlying complex genetic architecture involving interactions between multiple loci and environmental factors (van Os et al., 2008). Another source of difficulty is the reliability of phenotypic assessments. Schizophrenia diagnostics employ a patient's self-reports rather than biochemical, electrophysiological, or other reliably measurable biomarkers. The validity of schizophrenia as a diagnostic entity has therefore been criticized by a number of scientists and clinicians as lacking in scientific validity and diagnostic reliability (Bentall et al., 2004, Boyle, 2002, Kendell and Jablensky, 2003, Linscott et al., 2009, van Os,, 2009, van Os and Murray, 2008). Due to the diagnostic heterogeneity and the genetic and phenotypic complexity of schizophrenia, as well as other complex brain-related disorders, identification of a genetic basis of this disease is a very challenging task.

The endophenotypic approach is one of several strategies that have been adopted to deal with the complexities of these disorders and identify underlying genes (Gottesman and Gould, 2003). The principle of this approach is to reduce complex phenotypes into components (neurophysiological, biochemical, endocrine, cognitive, neuropsychological, or neuroanatomical phenotypes) that could be reliably quantified and are under close regulation by genetic variation. The endophenotype (or intermediate phenotype) is therefore a kind of biomarker that, unlike disease diagnosis, has a clearer underlying biological and genetic basis. They are expected to be genetically correlated with disease liability, and can be measured in both affected and unaffected individuals. Since quantitative endophenotypes are generally closer to the action of the gene and are measurable in both unaffected and controls, they exhibit higher genetic signal-to-noise ratios and are anticipated to provide a greater power to localize and identify disease-related quantitative trait loci (QTLs) than affection status alone (Blangero et al., 2003).

Human studies often use endophenotypes that are related to evolutionary conserved traits and therefore can be extended to animal models.

Volumes of various brain structures are quantitative phenotypes that can be reliably measured on MRI scans using either automated (voxel-based morphometry method, VBM; deformation-based morphometry; surface-based morphometry) or manual tracing (region and volume of interest, ROI and VOI method) (Ashburner and Friston, 2000, Good et al., 2001a; Thompson and Toga, 1997, Thompson et al., 1996a, Thompson et al., 1996b). Brain structure measurements show considerable interindividual variation in general populations and differences between healthy and affected persons, and are therefore considered promising candidate endophenotypes that might facilitate investigations of a genetic basis into the natural variation of brain-related traits in populations and for psychiatric disorders such as schizophrenia. When measuring brain structure and the heritability of these structures, normal neuroanatomical variations in the human brain should be taken into account (Allen et al., 2002). These neuroanatomical variations are under the influence of variables such as gender (Cosgrove et al., 2007, Giedd, 2008) and age (Carne et al., 2006, Good et al., 2001b), or other components of genetic variety such as height (Kappelman, 1993). Intelligence seems to be related to brain volume as well. In humans, brain volume is a quantitative trait with high heritability (Posthuma et al., 2002b; Thompson et al., 2001, Thompson et al., 2002). Previous studies have shown that brain volume is also correlated with general intelligence, working memory, perceptual organization, and processing speed, which are also highly heritable (Posthuma et al., 2003, Roth and Dicke, 2005). In recent years, it has been discovered that the structure of the adult human brain changes when new cognitive or motor skills, including vocabulary, are learned (Lee et al., 2007), and that this structural neuroplasticity (increased gray matter volume), after 3 months of training a visual motor skill, seems to last for at least 3 months without further practicing (Driemeyer et al., 2008, Ilg et al., 2008). Another example is the impact of handedness, which has a significant genetic component (McManus and Bryden, 1991, Annett et al., 1974) and is strongly related with cerebral asymmetry (Alexander and Annett, 1996, Geschwind and Galaburda, 1985). To what extend the cerebral asymmetry is also heritable remains unclear (Geschwind et al., 2002). The neuroanatomical phenotype and the heritability of brain structures varies among individuals with a neuropsychiatric disorder, meaning that the influence the disease state has on the heritability is such that the morphology of brain structures can be altered by the disease state and these morphologically altered brain structures can be inherited. For example, there are studies showing small hippocampal volumes in adult humans with recurrent major depression (Bremner et al., 2000, Sheline et al., 1999) and posttraumatic stress disorder (Bremner et al., 1995). Hippocampal morphology is also altered by stress in carefully controlled studies with rodents (Gould and Tanapat, 1999; Magarinos et al., 1995; Sapolsky, 2000). These smaller hippocampi have been taken as evidence that stress-related disorders induce hippocampal volume loss (Sapolsky, 2000, Sheline et al., 1999). One possible pathway by which altered brain morphology changes might be inherited is that small hippocampal volumes are inherited and act as predisposing factors toward the development of psychiatric disorders that are triggered by stress (Gurvits et al., 1996, Sapolsky, 2000).

When measuring brain structure volumes and calculating heritability estimates, these variables, having a strong genetic component themselves, should be taken into account.

(to study genetics of both natural and disease-related variation)

Twin and adoption studies have shown substantial genetic influences are involved in the risk of developing schizophrenia (Cannon et al., 1998, Cardno et al., 2002, Kendler and Diehl, 1993, McGuffin et al., 1984). The identification of predisposing genes has been hampered by difficulties in detecting nonpenetrant carriers and by uncertainties concerning the extent of locus of heterogeneity (McDonald and Murphy, 2003). By studying the inheritance of endophenotypes, we can increase the power to detect the genes involved in clarifying the pathways leading from genetic predisposition to clinical disorder (Gottesman and Gould, 2003). One of the most extensively studied endophenotypes in schizophrenia research is the study of structural brain abnormalities. These structural changes are well established in schizophrenia. The most robust findings from meta-analyses of these structural brain abnormalities is increased total ventricular volume, reduction of whole brain and intracranial volume, reduced hippocampal and amygdalar volume (Lawrie and Abukmeil, 1998, Ward et al., 1996, Wright et al., 2000). These brain abnormalities are also present in unaffected family members of patients with schizophrenia, to a less degree, but significantly more than in controls according to a meta-analysis reporting on brain changes in unaffected family members of patients with schizophrenia (Boos et al., 2007, van Haren et al., 2004).

One hypothesis in using these structural abnormalities as endophenotypes, in order to get closer to the action of the genes, suggests that there might be an overlap in the neurodevelopmental genes responsible for both the volume change and the development of the illness, that is, the existence of a genetic correlation. Schizophrenia may involve genetically determined pathological processes of early brain development which continue to unfold as the brain matures through neuronal loss and synaptic pruning during adolescence. Neurodevelopmental abnormalities then lead to the activation of pathological neural circuits, which respond to environmental stressors, leading to the emergence of symptoms (Deutch, 1993). This hypothesis is supported by the finding that MRI abnormalities are present at the onset of the illness, and progress very slowly if at all (Weinberger, 1995). A contrasting view on the etiology of schizophrenia is that the relationship between schizophrenia and brain volumes might be environmental in nature. For example, perinatal trauma has been shown to be an important determinant of some brain structure abnormalities in schizophrenia (Cannon et al., 2003, McNeil, 2000, Verdoux et al., 1997).

Defining to what degree the development of these brain structures is of genetic origin, that is, the heritability of brain structures, and using these brain structures as an endophenotype may be helpful in better identifying the action of these neurodevelopmental genes. Heritability estimates for brain structures in healthy persons and persons with schizophrenia will help to differentiate between disease related alterations in brain morphology and genetic influences (heritability).

Twin studies are considered important in investigating genetic influences on variation in human brain morphology in healthy individuals and those with schizophrenia. The twin model is particularly helpful in determining the relative contribution of genetic, common, and unique environmental influences on variation in brain structures (Posthuma and Boomsma, 2000). Heritability estimates of brain structure are usually based on data from monozygotic twins (MZ, who are nearly genetically identical) and dizygotic twins (DZ, who share on average 50% of their segregating genes). If heritability estimates are based on the assumption that for a certain brain measure, MZ twin pairs will resemble each other more closely than DZ twin pairs, it can be inferred that variation of the brain measure is heritable. However, in addition to genetic influences, common (or shared) environmental influences may play a role in explaining resemblances. The effects of shared environmental factors may be suggested when correlations in DZ twins are larger than half of the MZ correlation (Boomsma et al., 2002). The effects of unique environmental factors are obtained from the extent to which MZ twins do not resemble each other. However, the twin method has been criticized for its nongeneralizability due to differences in intrauterine and family environment differences between the twins and compared with singletons (Price, 1950).

Heritability estimates and the search for genes in diseased and nondiseased humans are difficult due to the impact of environmental factors. Studies in nonhuman primates (NHPs) might be helpful in elucidating the impact of environmental factors.

Obstacles in investigating complex traits in human subjects involve the hard-to-control and extensive effects of environment, personal development, and medical history including administrated medications. To a large extent, these problems can be overcome by applying an animal model. Particularly useful are NHPs due to the high conservation of anatomical, neurophysiological and cognitive traits as well genetic sequence with humans. NHP pedigrees have helped studies of various complex traits including the neruoanatomical due mostly to the availability of pedigree facilitating genetic studies, established genetic relationships among pedigree animals, well-documented medical and developmental history and uniform rearing conditions reducing the influence of environmental factors.

To disentangle the possible mechanisms by which genetic and environmental factors contribute to the morphological abnormalities in the brain animal models may be used, specifically NHPs, for example, monkey brains, in which the influence of environment is reduced. The evolution of cognitive function in hominoids, for example, depends largely on our understanding of the organization of the frontal lobes in extant humans and apes (Semendeferi et al., 1997). There are studies with findings showing similarities to human brains, such as the possibility of microstructural plasticity in the NHP hippocampus (Sapolsky et al., 1990, Siegel et al., 1993, Uno et al., 1989). Other studies show the impact of environmental factors on the volumes of different brain structures, such as the affect of differential rearing on the corpus callosum size in rhesus monkeys (Sánchez et al., 1998). Studies on brain characteristics and the genetic variation of regional brain morphologies of NHPs will help us understand the meaning of the brain characteristics in humans as well. They will also help us determine whether certain assumptions, such as those about certain brain characteristics being strictly unique to humans, are correct. For example, it is often claimed that the frontal lobe is disproportionately larger in humans than in other species, but conflicting reports exist on this issue. Results of a study by Semendeferi et al. (1997) indicate that although the absolute volume of the brain and the frontal lobe is the largest in humans, the relative size of the frontal lobe is similar across hominoids and humans do not have a larger frontal lobe than expected from a primate brain of human size. Other important functional parts of the brain, like area 10 in cortical areas of the frontal lobe, which is involved in higher cognitive functions, also seem to form the frontal pole in chimpanzee, bonobo, orangutan, macaque, and gibbon brains. Area 10 has similar cytoarchitectonic features in the hominoid brain, but aspects of its organization vary slightly across species, including relative width of its cortical layers and the space available for connections (Semendeferi et al., 2001). Although some features of the human brain, like the asymmetric Broca's area and the planum temporal in Wernicke's posterior receptive language area in the left hemisphere, were thought to be unique to humans, studies in great apes (Cantalupo and Hopkins, 2001) and chimpanzees (Gannon et al., 1998) show the same anatomic hemispheric asymmetry in the Broca's area and planum temporal, respectively.

Section snippets

Heritability of Brain Structure Phenotypes

Heritability estimates allow for determining which neuroanatomical measures are traits with marked genetic components and could therefore be interesting both from the perspective of natural variation in brain related processes in population and as an attractive endopehnotype for schizophrenia studies. Investigating heritability in various aspects in healthy individuals, in relation to schizophrenia and in the NHP model, may answer two essential questions: (1) to what degree are brain volumes

Genes for Brain Structures in Healthy and Persons with Schizophrenia

The dramatically increased brain volume plays an important role in the origin of our species. In humans, brain volume is a quantitative trait with high heritability (Posthuma et al., 2002b; Thompson et al., 2001). Disturbances in brain development is one of the explanations for neuropsychiatric disorders such as schizophrenia (Weinberger, 1995), suggesting that developmental genes involved in the development of the brain may overlap with the causal genes, for example, potential candidate genes,

Limitations of Neuroimaging Studies

A number of limitations in imaging studies need to be acknowledged, such as the validity and reliability of measures (e.g., using measurements that rely on a representative slice vs. whole volume or do not correct for total brain volume). Measurements of brain volumes reveal differences between affected subjects and healthy controls, but the magnitude of these differences is moderate and there is always a substantial overlap in the distributions of the comparison groups. However, looking at the

Acknowledgments

This research was supported by the Netherlands Organization for Scientific Research (NWO) under project number 017.002.048. We also thank A. Jasinska, Ph.D., UCLA/Los Angeles/USA, for her advices.

References (177)

  • J. Guidry et al.

    New genetic hypothesis of schizophrenia

    Med. Hypotheses

    (1999)
  • T. Hata et al.

    Possible effect of the APOE epsilon 4 allele on the hippocampal volume and asymmetry in schizophrenia

    Am. J. Med. Genet

    (2002)
  • H.E. Hulshoff Pol et al.

    Genetic contributions to human brain morphology and intelligence

    J. Neurosci.

    (2006)
  • S.M. Lawrie et al.

    Brain abnormality in schizophrenia. A systematic and quantitative review of volumetric magnetic resonance imaging studies

    Br. J. Psychiatry

    (1998)
  • M.C. Mahaney et al.

    Quantitative genetics of relative organ weight variation in captive baboons

    Hum. Biol

    (1993)
  • C. McDonald et al.

    The new genetics of schizophrenia

    Psychiatry Clin. North Am.

    (2003)
  • P. McGuffin et al.

    Twin concordance for operationally defined schizophrenia. Confirmation of familiality and heritability

    Arch. Gen. Psychiatry

    (1984)
  • A.M. McIntosh et al.

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

    Mol. Psychiatry

    (2008)
  • G.H. Nguyen et al.

    DNA stability and schizophrenia in twins

    Am. J. Med. Genet. B Neuropsychiatr. Genet.

    (2003)
  • G.D. Pearlson et al.

    Structural and functional magnetic resonance imaging in psychiatric disorders

    Can J. Psychiatry

    (2007)
  • A.M. Addington et al.

    GAD1 (2q31.1), which encodes glutamic acid decarboxylase (GAD67), is associated with childhood-onset schizophrenia and cortical gray matter volume loss

    Mol. Psychiatry

    (2005)
  • A.M. Addington et al.

    Neuregulin 1 (8p12) and childhood-onset schizophrenia: Susceptibility haplotypes for diagnosis and brain developmental trajectories

    Mol. Psychiatry

    (2007)
  • I. Agartz et al.

    BDNF gene variants and brain morphology in schizophrenia

    Am. J. Med. Genet. B Neuropsychiatr. Genet

    (2006)
  • J.S. Allen et al.

    Normal neuroanatomical variation in the human brain: An MRI-volumetric study

    Am. J. Phys. Anthropol.

    (2002)
  • F. Angelucci et al.

    BDNF in schizophrenia, depression and corresponding animal models

    Mol. Psychiatry

    (2005)
  • M. Annett

    Handedness in the children of two left-handed parents

    Br. J. Psychol.

    (1974)
  • L.D. Atwood et al.

    Genetic variation in white matter hyperintensity volume in the Framingham Study

    Stroke

    (2004)
  • W.F. Baaré et al.

    Quantitative genetic modeling of variation in human brain morphology

    Cereb Cortex

    (2001)
  • W.F. Baaré et al.

    Volumes of brain structures in twins discordant for schizophrenia

    Arch. Gen. Psychiatry

    (2001)
  • A.J. Bartley et al.

    Genetic variability of human brain size and cortical gyral patterns

    Brain

    (1997)
  • R.P. Bentall et al.

    Models of Madness: Psychological, Social and Biological Approaches to Schizophrenia

    (2004)
  • A. Biondi et al.

    Are the brains of monozygotic twins similar? A three-dimensional MR study

    AJNR Am. J. Neuroradiol.

    (1998)
  • M. Boyle

    Schizophrenia: A Scientific Delusion?

    (2002)
  • H.B. Boos et al.

    Brain volumes in relatives of patients with schizophrenia: A meta-analysis

    Arch. Gen. Psychiatry

    (2007)
  • D. Boomsma et al.

    Classical twin studies and beyond

    Nat. Rev. Genet.

    (2002)
  • R.G. Brans et al.

    Heritability of changes in brain volume over time in twin pairs discordant for schizophrenia

    Arch. Gen. Psychiatry

    (2008)
  • J.D. Bremner et al.

    Functional neuroanatomical correlates of the effects of stress on memory

    J. Trauma Stress

    (1995)
  • J.D. Bremner et al.

    Hippocampal volume reduction in major depression

    Am. J. Psychiatry

    (2000)
  • C.M. Bruckner et al.

    Symmetric differences squared and analysis of variance procedures for estimating genetic and environmental variances and covariances for beef cattle weaning weight: (a) Comparison via simulation (b) Estimates from a data set

    J. Anim. Sci.

    (1986)
  • J.W. Buckholtz et al.

    Allelic variation in RGS4 impacts functional and structural connectivity in the human brain

    J. Neurosci.

    (2007)
  • W. Cahn et al.

    Brain morphology in antipsychotic-naïve schizophrenia: A study of multiple brain structures

    Br. J. Psychiatry.

    (2001)
  • J.H. Callicott et al.

    Variation in DISC1 affects hippocampal structure and function and increases risk for schizophrenia

    Proc. Natl Acad. Sci. USA

    (2005)
  • A.G. Cardno et al.

    Twin studies of schizophrenia: From bow-and-arrow concordances to star wars Mx and functional genomics

    Am. J. Med. Genet

    (2000)
  • A.G. Cardno et al.

    Clinical variables and genetic loading for schizophrenia: Analysis of published Danish adoption study data

    Schizophr. Bull.

    (2002)
  • D. Carmelli et al.

    Evidence for genetic variance in white matter hyperintensity volume in normal elderly male twins

    Stroke

    (1998)
  • T.D. Cannon et al.

    The genetic epidemiology of schizophrenia in a Finnish twin cohort. A population-based modeling study

    Arch. Gen. Psychiatry

    (1998)
  • T.D. Cannon et al.

    Cortex mapping reveals regionally specific patterns of genetic and disease-specific gray-matter deficits in twins discordant for schizophrenia

    Proc. Natl Acad. Sci. USA

    (2002)
  • T.D. Cannon et al.

    Early and late neurodevelopmental influences in the prodrome to schizophrenia: Contributions of genes, environment, and their interactions

    Schizophr. Bull.

    (2003)
  • T.D. Cannon et al.

    Association of DISC1/TRAX haplotypes with schizophrenia, reduced prefrontal gray matter, and impaired short- and long-term memory

    Arch. Gen. Psychiatry

    (2005)
  • C. Cantalupo et al.

    Asymmetric Broca's area in great apes

    Nature

    (2001)
  • Cited by (30)

    • Heritability and reliability of automatically segmented human hippocampal formation subregions

      2016, NeuroImage
      Citation Excerpt :

      However, structural variance within the whole hippocampus may be less heritable in elderly adults (h2 = 0.4–0.65) (DeStefano et al., 2009; Mather et al., 2015; Sullivan et al., 2001), possibly due to environmental stressors (Hedges and Woon, 2010), alterations in testosterone levels (Panizzon et al., 2012) or other endogenous biological factors. Similarly, total hippocampal volume is only moderately heritable in schizophrenia (h2 = 0.36–0.73) (Kaymaz and Os, 2009; Roalf et al., 2015). Thus, while the heritability of total hippocampal volume is well established across many populations, the heritability of structural variations in individual subregions has yet to be delineated.

    • Schizophrenia risk variants modulate white matter volume across the psychosis spectrum: Evidence from two independent cohorts

      2015, NeuroImage: Clinical
      Citation Excerpt :

      Neurobiological models of schizophrenia (SZ) and bipolar disorder (BD) implicate altered interregional connectivity and suggest that some of the changes in white matter structure may be heritable (Kaymaz and van Os, 2009; McDonald et al., 2004; McIntosh et al., 2006). This hypothesis is supported by studies that show that decreased white matter volume may be an indicator of genetic liability to both schizophrenia (Francis et al., 2013; Kaymaz and van Os, 2009; McDonald et al., 2004; McIntosh et al., 2006; Oertel-Knochel et al., 2012; van Haren et al., 2012) and bipolar disorder (Emsell and McDonald, 2009; Hasler et al., 2006; Matsuo et al., 2012). Genome wide association studies (GWAS) are identifying an increasing number of single-nucleotide polymorphism (SNPs), significantly associated with risk for schizophrenia and/or bipolar disorder (Cardno and Owen, 2014; GWAS Consortium, 2011; Psychiatric GWAS Consortium Bipolar Disorder Working Group, 2011; Schulze et al., 2014; Smoller et al., 2013), although the functional mechanisms by which these variants confer risk for psychosis remain largely unknown.

    • Dopamine Receptor Imaging in Schizophrenia: Focus on Genetic Vulnerability

      2013, Imaging of the Human Brain in Health and Disease
    View all citing articles on Scopus
    View full text