genetic mechanisms of intelligence

Chapter 1General IntroduCtIon

General introduction

13

intelligence is one of the most studied complex behavioral traits and has been a research focus for more than a century. intelligence shows consistent and strong associations with important life outcomes such as educational and occupational success, social mobility, health, illness and survival (for review see (deary et al., 2012)). In addition, mutations in genes involved in the development of intellectual abilities have been shown to be associated with mild mental disability (Zechner et al., 2001), indicating that at least some forms of mild mental disability may be the low extreme of the normal distribution of intelligence (Plomin & Kovas, 2005). In addition, intellectual dysfunction is a prominent feature of multiple mental disorders, including autism, dementia and schizophrenia. Studying the causes of individual differences in normal intellectual function is thus also of importance to understanding the causes of intellectual dysfunction, and is of broad clinical and societal interest. Although intelligence is often seen as a typically human trait, other species also show intelligent behavior. In addition, biological structures thought to be involved in human intelligence are also present in other species and have been shown to be involved in similar, intelligent behavior across species. For a more detailed description of the biology of intelligence and its importance see Box 1.

ChAPter 1

14

BOX 1- INTELLIGENCE: IMPORTANCE

BIoloGY oF IntellIGenCe

intelligence is most widely studied in humans, but has also been observed in other animals. there are some arguments about the use of the term “intelligence” for non-humans, but there is no doubt that cognitive abilities related to intelligence such as learning from observation/experience, memory, perception and problem solving are important adaptive skills for numerous animals. In 1960, Van Lawick-Goodall observed for the first time that chimpanzees can use a wooden stick as a tool to fish for termites in the ground (Van Lawick-Goodall, 1973). Later it was reported that birds can learn to eat non toxic monarchs by experience (Brower & Glazier, 1975) and that other animals, besides bird and chimpanzees, are also capable of learning from experience and use tools to overcome some daily challenges (Shettleworth, 2009, Van Lawick-Goodall et al., 1971). Intelligence, both in humans and other animals, is a useful asset in different life aspects, but especially when adaptation to a new environment is necessary, when the life challenges are novel and complex and situations are ambiguous, changing, or unpredictable (Gottfredson, 1997b, Shettleworth, 2009, Van Lawick-Goodall et al., 1971). Gottfredson et al (Gottfredson & Deary, 2004) reported that individuals with high intelligence test scores have an increased chance to have a good education, a good occupation and consequently a higher income and that on the other hand, individuals with lower intelligence test scores have more chance to poverty, incarceration and chronic welfare use. recent anatomical and behavioral studies showed that, even though there are differences in detail, the anatomical and functional organization of the brain medial temporal-lobe system is similar in humans, nonhumans primates and mice (Squire, 1992; Mayford et al., 1996). Studies with functional brain imaging revealed that the human frontal lobes were implicated in abstract reasoning (duncan et al., 1996, Gray et al., 2003, Piercy, 1964) and that posterior brain lesions often cause substantial decreases in intelligence test scores (Luerding et al., 2004). Most recently, studies also revealed a common pattern of activity of some brain regions, such as the frontal and parietal cortex, in response to different types of cognitive challenges (Duncan, 2010, Duncan & Owen, 2000), such as during working memory and problem solving tasks.

NOte: hIGhLIGhteD IN CIrCLeS Are the reGIONS WIth COMMON PAtterN OF ACtIVItY DUrING DIFFereNt tYPeS OF COGNItIVe tASKS.

Although controversies exist, the generally accepted definition of human intelligence is that intelligence is ‘a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience’ (Gottfredson, 1997a). the


15

more practical definition is that intelligence is simply ‘what the intelligence test measures’(Boring, 1923). Psychometric intelligence tests such as the Wechsler Adult Intelligence Scale (WAIS (Wechsler, 1997)) and Wechsler Intelligence Scale for children (WISC (Wechsler, 1991)) usually consist of a number of component subtests that taken together are used to infer a general intelligence test score: the IQ (Intelligence Quotient) test score. the standardized distribution of IQ test scores follows a Gaussian distribution, and ranges from ~50 to ~150, with a standardized mean of 100. IQ test scores are generally presented as Full Scale IQ, Verbal IQ and Performance IQ, although different tests may adhere to different IQ dimensions. For a more detailed description of the IQ test used in this thesis please see Box 2.

BOX 2- INTELLIGENCE: MEASUREMENT

IQ TEST Intelligence tests usually consist of a number of component subtests that taken together are used to infer a general IQ score. Intelligence tests such as the WAIS (Wechsler, 1997) and WISC (Wechsler, 1991) provide an index of both general IQ and primary abilities such as verbal comprehension, working memory, perceptual speed, and perceptual organisation. In addition to the Wechsler scales, the Groninger Intelligence test GIt (Luteijn & Ploeg, 1983) is the most frequently used intelligence test of formal IQ in the Netherlands and, although analogous to the Wechsler Adult Intelligence Scale, relies less on the verbal abilities. Below is a schematic representation of the building blocks of one of the tests used in this thesis, the WAIS test.

IQ DISTRIBUTION In a general population, large inter-individual differences in IQ test scores can be observed. the distribution of IQ scores in a general population fits a normal bell-shaped curve (Gaussian distribution). Ninety-five percent of the population fall within IQ scores of 70 and 130 and sixty-eight percent of the general population scores between 85 and 115 (reynolds et al 1987, Gottfredson, 1997, hermstein & Murray, 1994 ). In the general population, 2% has an IQ under 70 and this is generally considered as the benchmark for "intellectual disability", a condition of limited mental ability in that it produces difficulty in adapting to the demands of life (Curry et al., 1997; roeleveld et al., 1997; Croen et al., 2001). In the other extreme (above 140) are the highly intelligent/gifted, representing approximately 0.25% of the population (1 in 400) (reynolds et al 1987, Gottfredson, 1997, hermstein & Murray, 1994 ).

ChAPter 1

16

individuals from the same family tend to resemble each other more with respect to their IQ test scores, than unrelated people. these differences between unrelated persons and the resemblance between relatives are due to genetic factors, shared environmental factors, cultural transmission from one generation to the next, social interactions between family members, or a combination of these mechanisms. Below, I provide a general introduction into how genetic variation in the human genome can lead to individual differences in trait values.

Genetic variants and individual trait differencesGenetic mutations can create new alleles and sometimes can lead to diseases such as Microcephaly, where mutations in a single gene ASPM (abnormal spindle-like microcephaly) cause a brain developmental disorder, altering neuronal migration and cortical layering process (for a review see (Bond et al., 2002)). the combination of genetic mutations and variants is called haplotype. haplotypes that occur more often in a disease group are high-risk haplotypes. For example, in a schizophrenia family based study it was found that a certain combination of 8 allele calls in the DTNBP1 (dystrobrevin binding protein 1) gene only occurred in the disease group (Van Den Oord et al., 2003). human individuals differ from one another by about one base pair per thousand. If these differences occur within coding regions (regions that code proteins) or regulatory regions, phenotypic variation in a trait may result. It is estimated that humans have more than 5.9 million sites where genetic variation exists (durbin et al., 2010). Genetic variation among humans occurs on many different scales, ranging from point mutations to large copy number variation. Box 3 explains how human genetic informations is stored.


17

BOX 3 -GENETIC INFORMATION

the genetic information for all living things is found in DNA (Deoxyribonucleic acid). this genetic material encodes the information that, together with environmental factors, ultimately shapes what an organism is, controls how it grows and develops, and influences how it behaves. Selective activation and inactivation of genes encoded in DNA enables an organism to produce specific proteins in particular amounts over time and in response to the internal and external factors. the production of protein takes place in two steps. In the first step, called “transcription”, the DNA message is copied into a messenger rNA (mrNA) by an enzyme (rNA polymerase). In this second step, called “translation,” the mrNA is then used to make a matching protein sequence. the human genetic information is stored on 23 chromosome pairs (total of 46) located in the nucleus plus the small mitochondrial DNA as shown in the figure below.

A chromosome is an organized structure of DNA and protein found in each cell. each individual has two strands of the DNA and the information within a particular gene or DNA region (locus/loci) is not always the same between individuals: Different individuals may carry different copies of a locus. Since different copies of a gene give different instructions to the cell, genetic variation may lead to phenotypic variation or diseases when a copy of an allele leads to incorrect instructions for the cell. each unique form of a single gene is called an allele. For example, two sequenced DNA fragments from two different individuals, ttCGGAtAA to ttCGAAtAA, contain a difference in a single nucleotide (sequence letter). In this case we say that there are two alleles of this DNA fragment: G and A. three nucleotides form a codon, which codes for an amino acid and multiple amino acids form proteins.

Single nucleotide polymorphisms (SNPs) are the most common type of point mutation. Changes in a single nucleotide in the coding region can either lead to no changes at the amino acid level (synonymous SNP), or changes in the coded amino acid level (non-synonymous SNP). even when the polymorphism remains silent at the protein level, as in the case of synonymous mutations, the effect of the polymorphism can be assayed at the level of mrna. this is based on the fact that when a polymorphism occurs in coding dna it will be present in the transcribed mrNA, allowing the relative abundance of the two alleles in cells from an individual heterozygous for a given SNP to be assayed. Polymorphisms occurring in promoter regions upstream of genes (non coding regions) may potentially affect the process

ChAPter 1

18

of transcription, whereby rNA polymerase is recruited to the gene and the nascent mrNA synthesized. Variation in the DNA sequence may potentially alter the affinities of existing protein–DNA interactions or, indeed, recruit new proteins to bind to the DNA, altering the specificity and kinetics of the transcriptional process (Andolfatto, 2005, Farzaneh-Far et al., 2001, Knight et al., 1999a, tournamille et al., 1995). Copy number variations (CNVs) are variations in the number of copies in the dna. it was generally thought that genes were almost always present in two copies in a genome. However, recent discoveries have revealed that segments of dna in the general population, can vary in number of copies. the length of CNVs can range from thousands of base pairs to millions of base pairs. these genetic variants can be caused by structural rearrangements of the genome such as deletions, duplications, inversions, and translocations. these structural changes in the genome can interfere with the gene dosage and may cause perturbation in particular pathways. CNVs have been implicated in several types of neurodevelopmental diseases to different degrees, including autism and intellectual disability (Knight et al., 1999b, Sebat et al., 2007). the past decades have seen major technological advances that allow investigating the DNA more thoroughly and on a larger scale. this has also resulted in multiple efforts in identifying the genetic causes of individual intellectual abilities. Below I will provide a general overview of the past efforts for identifying genetic variants for intelligence and I will then continue with explaining the rationale behind the research strategies for the current thesis.

Dissecting the heritability of intelligence: current state of affairs at the start of my PhD project in 2007the heritability of intelligence ranges from around 30% in young children to more than 80% in late adulthood (Bouchard & Mcgue, 1981, Deary et al., 2006, Vinkhuyzen et al., 2011) and it is generally assumed to be influenced by many genes of small effect that in turn may interact with each other and with the environment (Davies et al., 2011, Plomin & Spinath, 2004, Savitz et al., 2006). Given its polygenic nature and the small genetic effect, detection of putative genes underlying intelligence proves to be a difficult task. in the past thirty years, considerable progress has been made in genotyping technologies, which facilitates the detection of the actual genetic variants that underlie the high heritability of intelligence. In the 1980’s - 1990’s, genetic linkage studies were widely used to detect genomic regions of interest to complex traits, such as intelligence. In the 1990’s candidate gene studies rapidly replaced linkage studies, only to be replaced themselves around 2005 by hypothesis free genome wide association.


19

Linkage studiesthe basic idea of linkage analysis is to compare resemblance between subjects at a genetic marker with resemblance at a measured trait (e.g., intelligence) (for a more detailed explanation of linkage analysis please see Appendix 1 of this thesis). relatives who resemble each other more phenotypically will also resemble each other more genetically if the trait under study is heritable. Linkage studies have provided evidence for a genetic contribution to intelligence. the first whole genome linkage scan for intelligence was published in 2005 (Posthuma et al., 2005) and reported two significant areas of linkage to intelligence in a total of 634 sibling pairs; one on the long arm of chromosome 2 (2q) and one on the short arm of chromosome 6 (6p). these findings were in convergence with linkage findings in clinical disorders that are characterized by cognitive disabilities. For instance, the chromosome 2 area had been implicated in linkage scans for autism, dyslexia and schizophrenia (Addington et al., 2005, Buxbaum et al., 2001, Fagerheim et al., 1999, Imgsac, 2001, Kaminen et al., 2003, Ylisaukko-Oja et al., 2005), while the chromosome 6 region is the main linkage area for reading ability and dyslexia (Cardon et al., 1994, Fisher et al., 1999, Grigorenko et al., 1997). Four linkage studies for intelligence have been published since 2005 (Buyske et al., 2006, Dick et al., 2006, Luciano et al., 2006, Wainwright et al., 2006). two studies with a partly overlapping sample confirmed the importance of the areas on chromosomes 2 and 6 for specific aspects of intelligence (Luciano et al., 2006) as well as for academic achievement, which is highly correlated with IQ scores (Wainwright et al., 2006). the study by Luciano and others (Luciano et al., 2006) performed a genome wide linkage scan of 795 microsatellite markers in 361 families, comprising 2-5 siblings. In this study, they showed that both word recognition and IQ test score were linked to chromosome 2, corroborating the notion that the same genes influence different aspects of cognitive ability (Plomin & Kovas, 2005). Dick and others (Dick et al., 2006) also confirmed linkage for intelligence to the chromosome 6 area (total of 201 families), and a second scan on the same dataset (Buyske et al., 2006) showed strong evidence for linkage of specific cognitive abilities on chromosome 14, which had shown suggestive linkage in the two previous linkage studies (Dick et al., 2006, Luciano et al., 2006). Converging evidence from these five whole genome linkage studies provided support for the involvement of three different chromosomal regions in human intelligence: 2q24.1-31.1, 6p25-21.2, and 14q11.2-12 (see table 1). Significant genomic regions that result from a linkage study tend to be relatively large (10-20Mb), harboring many potential candidate genes the obvious next step after finding the main linkage areas for intelligence is therefore to detect the relevant genetic factors in these linkage regions, using a candidate gene/loci association approach.

ChAPter 1

20

Table 1. Genomic loci associated with intelligence through linkage studies

Locus Refs. Previous population

2q24.1-31.1 Posthuma et al. (2005) Family study, NL and AUS, N = 950

Luciano et al. (2006) Family study, NL and AUS, N = 836

2q31.3 Butcher et al. (2008) Family study, UK, N = 3,195

6p25-21.2 Posthuma et al. (2005) Family study, NL and AUS, N = 950

Luciano et al. (2006) Family study, NL and AUS, N = 836

7q32.1 Butcher et al. (2008) Family study, UK, N = 3,195

14q11.2-12 Buyske et al. (2006) COGA, US, N = 1,115

16p13.3 Butcher et al. (2008) Family study, UK, N = 3,195

Note: COGA = Collaborative Studies on Genetics of Alcoholism; N indicates sample size.

Candidate gene/loci approachthe candidate gene/loci approach involves testing for association of pre-selected genes or genetic regions with a disease or trait. the statistical test for association involves a simple comparison of allele-frequencies across cases and controls (dichotomous traits) or a linear regression of genotype on trait level (quantitative traits). the selection of suitable and promising candidate genes, however, requires prior knowledge, e.g. from previous linkage findings or from putative underlying biological mechanisms (or pathways) that might be relevant to intelligence, such as neurotransmission and neurodevelopment. candidate genes might also be selected because they are associated with a more extreme form of a trait. It has been hypothesized (i.e. the “Generalist Genes hypothesis”) that genes affecting intellectual abilities are to some extent the same genes that affect intellectual disabilities (Plomin & Kovas, 2005). In 2007, a handful of candidate genes had been associated with intelligence (see table 2), however, just a couple of genes showed replication in independent studies (see (Posthuma, 2006)) (for more details about the candidate gene approach and genes previously associated with intelligence please see appendix 1 of this thesis). Candidate gene studies can be performed relatively quickly and inexpensively and may allow identification of genes with relatively small effects. however, the candidate gene approach is limited by preselection of genes and relies heavily on how much is already known about the biology or genomic location of the trait being investigated. Without prior knowledge, exploratory, hypothesis free methods like genome wide association analysis might be more suitable.


21

Table 2. Previous genetic association findings on intelligence

Gene Chr Gene size SNP Type P-value Population

DNAJC13 3 121371 rs1378810 intron 0.0007 Family study, UK, N=3,1951

TBC1D7 6 35001 rs2496143 intron 0.037Family study, UK,

N = 3,1951

DTNBP1 6 140233 rs1018381 intron 0.008Case-Control schizophrenia, USA,

N = 3392

ALDH5A1 6 42238 rs2760118coding-non-synonymous

0.001Population sample, USA,

N = 5943

IGF2R 6 137452 rs3832385 mrna-utr 0.02Population sample, USA,

N = 1024

CHRM2 7 148372 rs8191992 mrna-utr <0.017 Population sample, USA, N=8285

rs8191992 mrna-utr 0.036 COGA, USA, N = 1,1136

rs1378650 — 0.028 COGA, USA, N = 11136

rs1424548 — 0.037 COGA, USA, N = 11136

rs2350780 intron 0.016 COGA, USA, N = 11136




CHRM2 7 148372 rs324640 intron <0.001 Family study, NL, N = 6677

rs324650 intron <0.01 Family study, NL, N = 6677

rs2061174 intron <0.01 Family study, NL, N = 7628


BDNF 11 66856 rs6265coding-non-synonymous

0.046 Population sample, ChINA, N = 1149

rs6265coding-non-synonymous

0.001Population sample, SCOtLAND,

N = 90410

CTSD 11 11237 rs17571coding-non-synonymous

0.01Population sample, USA,

N = 76711

FADS3 11 91903 rs174455 intron 0.013 Family study, UK, N = 31951

DRD2 11 65564 rs2075654 intron 0.05 Family study, NL, N = 76212

KL 13 49708 rs9536314coding-non-synonymous


N = 91513

APOE 19 3611 rs28931577coding-non-synonymous


N = 46614

rs769455coding-non-synonymous


N = 46614

SNAP25 20 88588 rs362602 — 0.005 Family study, NL, N = 76215

rs363039 intron 0.001 Family study, NL, N = 76215

rs363050 intron 0.0002 Family study, NL, N = 76215

PRNP 20 15437 rs1799990coding-non-synonymous


N = 91516

CBS 21 23120 rs5742905coding-non-synonymous

0.02 Population sample, US, N = 20217

COMT 22 27221 rs4680coding-non-synonymous

0.05 Family study, NL, N = 76212

Note: COGA = Collaborative Studies on Genetics of Alcoholism family study; N = sample size (individuals). references in this table: 1Butcher et al.(2008); 2Burdick et al. (2006a, b); 3Plomin et al. (2004); 4chorney et al. (1998); 5Comings et al. (2003); 6Dick et al. (2007); 7Gosso et al. (2006b); 8Gosso et al. (2007); 9tsai et al. (2004); 10harris et al. (2006); 11Payton et al. (2006); 12Gosso et al. (2008a); 13Deary et al. (2005); 14deary et al. (2002); 15Gosso et al. (2006a); 16Kachiwala et al. (2005); 17Barbaux et al. (2000)

ChAPter 1

22

Genome wide associationGenome wide association (GWA) analysis became feasible around 2005. GWA allows testing for association between the trait of interest and genetic variants genotyped across the entire genome. GWA studies are dependent on the number of genetic variants genotyped. More genotyped variants means better genome coverage which increases the chance to find the causal variants (for a more detailed description of GWA approach and genetic terminologies please see Appendix 1 of this thesis). Plomin and coworkers (Butcher et al., 2008, Butcher et al., 2005, Plomin, 1999, Plomin et al., 2004) conducted several early, whole genome association studies (using 10K markers). they showed significant association of a functional polymorphism in ALDH5A1 (aldehyde dehydrogenase 5 family) on chromosome 6p with intelligence. Later, the same group, using more genotype markers (500K) and consequently better genome coverage, identified three novel genes through a whole genome scan using a DNA pooling approach for low and high IQ groups involved in intelligence; DNAJC13 (dnaJ (Hsp40) homolog, subfamily c, member 13), FADS3 (fatty acid desaturase 3) and TBC1D7 (tBC1 domain family, member 7) (Butcher et al., 2008). the previously mentioned chromosome 6 linkage area lies close to, but a bit further downstream of, the association that was reported in the genome wide allelic association study by the same group in 2005 (Butcher et al., 2005). In 2007, both linkage and association studies had thus provided valuable insights into the genetics of intelligence. however these findings were largely not replicated. Of the few replicated findings none showed sufficient evidence for causation. therefore, there was a strong need for new and combined strategies to unveil the genetic mechanisms underlying intelligence.

Aims and outline of this thesisAt the start of my PhD project in 2007 the field of the genetics of intelligence could be characterized by three main statements: 1. Intelligence is heritable; 2. Genes for intelligence most likely reside on chromosomes 2 and 6; and 3. there are a couple of named candidate genes (e.g. SNAP25, CHRM2, APOE, DRD2) that are likely involved in intelligence. As these statements were still very general, the general aim of this thesis was to look beyond these statements and aid in detecting genetic mechanisms that underlie inter individual variation in intelligence using a combination of different strategies. the delineation and selection of the gene finding strategy to conduct is difficult and ponderous. Below I present the strategies used in this thesis and the rationale behind them.


23

Candidate gene/loci human linkage studies are the method of choice when searching for Mendelian disease or polygenic traits influenced by rare alleles, however linkage is less powerful than association studies for detecting genes of minor to moderate effect that underlie complex traits such as intelligence. Despite its limitations, previous whole genome linkage scans for intelligence have provided good hints, harboring hundreds of possible candidate genes. these candidate genes should then be followed up to test whether they are a causative factor for the trait. In chapter 2 of this thesis I explored these previous linkage regions and made use of an in silico candidate loci association approach.

Gene-environment interaction and brain expressionIn the candidate gene/loci association studies, genes and regions are selected based on prior knowledge of biological pathways, which are known or predicted to be involved in the variance of the trait. Single variant or single gene analyses generally account for only a small proportion of the phenotypic variation in complex traits such as intelligence. Alternatively, gene set or pathway association analyses are playing an increasingly important role in uncovering genetic architectures of complex traits through the identification of systematic genetic and environmental interactions. there are, in general, two dominant lines of approches for genetic association: Association analyses based on genetic variants and those based on gene expression profiles. however, genetic influence on the complex trait can manifest in many ways, such as alterations of gene expression, single point mutations, variation of repetitive sequences, genetic structural rearrangements and gene-environment interaction; thus, an integrative approach combining multiple forms of evidence can more accurately and comprehensively capture pathway associations. In chapter 3, I investigated the genetic variants of genes in the fatty acid pathway, which is known to play an important role in cognitive function. In this study I used an integrative strategy combining gene-enviroment interaction with brain expression to identify genetic mechanisms underlying intelligence.

Generalist gene hypothesis and extreme phenotypesIn 2005, it was proposed that most genetic effects for learning abilities and disabilities are general rather than specific, implicating that the same genetic factors are largely responsible for both upper and lower extremes of the intellectual ability trait (Plomin & Kovas, 2005). In Chapter 4, I put this ‘Generalist Genes hypothesis’ to a test and investigated whether a gene that was previously reported to influence intelligence in the normal range, is also associated to the more extreme forms of intelligence. i

ChAPter 1

24

used two selected samples of extreme phenotypes (intellectual disability versus High IQ) to increase the statistical power of the analysis. Such an ‘extreme phenotypes’ design is more informative about alleles for disease susceptibility and can yield a reduction in genotyping requirements of several orders of magnitude compared with that required for randomly selected samples.

Gene-based testWithout prior candidate gene or loci, methods like genome wide association analysis (GWA) might be more suitable. In the past decade, the scale of genotyping and genetic association studies has increased rapidly, from single-locus analysis to GWA studies that allow the screening of 500,000–1,000,000 SNPs, covering 65%–95% of the human genome, depending on the array of choice. this has proven a successful method for identifying common variants with a relatively large effect but has been less effective when rare variants of large effect are of importance or when identifying genes of small effect for complex traits such as intelligence. In this thesis we decided to combine GWA and a gene-based association approach to find genes associated with education attainment, a trait that is highly correlated with intelligence (0.8) (deary et al., 2007). the gene-based approach tests for a correlation between all genetic variants (normally SNPs) within a gene and surrounding regions (normally ±50 Kb) and a complex trait, rather than each marker individually. Gene-based approaches combine the effects of all SNPs in a gene into a test-statistic and also correct for linkage disequilibrium (LD) between variants. the advantage of a gene-based approach is that, variations in protein-coding (e.g. within the gene) and adjacent regulatory regions (e.g. promoter regions) are easier to determine the biological function than intergenic regions (e.g. “gene desert” regions). Second, gene-based tests allow for direct comparison between different populations, despite the potential for different linkage disequilibrium (LD) patterns and/or functional alleles. third, these analyses can account for multiple independent functional variants within a gene, with the potential to greatly increase the power to identify disease/trait-associated genes with a lower number of statistical tests than the SNP-based GWA approach. In Chapter 5, I used a gene-based association approach to find genes associated with the highest level of education completed. In addition, I tested whether previously reported candidate genes for intelligence show a significant association with education attainment. Finally, the results of all four studies from my thesis are summarized and discussed in Chapter 6, together with a view on future approaches for elucidating the role of genes in intelligence.


25

REFERENCESAddington, A.M., Gornick, M., Duckworth, J., Sporn, A., Gogtay, N., Bobb, A., Greenstein, D., Lenane, M.,

Gochman, P., Baker, N., Balkissoon, r., Vakkalanka, r.K., Weinberger, D.r., rapoport, J.L. & Straub, r.e. (2005) GAD1 (2q31.1), which encodes glutamic acid decarboxylase (GAD67), is associated with childhood-onset schizophrenia and cortical gray matter volume loss. Mol Psychiatry, 10, 581-588.

Andolfatto, P. (2005) Adaptive evolution of non-coding DNA in Drosophila. Nature, 437, 1149-1152.

Bond, J., roberts, e., Mochida, G.h., hampshire, D.J., Scott, S., Askham, J.M., Springell, K., Mahadevan, M., Crow, Y.J., Markham, A.F., Walsh, C.A. & Woods, C.G. (2002) ASPM is a major determinant of cerebral cortical size. Nat Genet, 32, 316-320.

Boring, e. (1923) Intelligence as the tests test It. New Republic, 35-37.

Bouchard, t.J., Jr. & McGue, M. (1981) Familial studies of intelligence: a review. Science, 212, 1055-1059.

Brower, L.P. & Glazier, S.C. (1975) Localization of heart poisons in the monarch butterfly. Science, 188, 19-25.

Butcher, L.M., Davis, O.S., Craig, I.W. & Plomin, r. (2008) Genome wide quantitative trait locus association scan of general cognitive ability using pooled DNA and 500K single nucleotide polymorphism microarrays. Genes Brain Behav, 7, 435-446.

Butcher, L.M., Meaburn, e., Knight, J., Sham, P.C., Schalkwyk, L.C., Craig, I.W. & Plomin, r. (2005) SNPs, microarrays and pooled DNA: identification of four loci associated with mild mental impairment in a sample of 6000 children. Hum Mol Genet, 14, 1315-1325.

Buxbaum, J.D., Silverman, J.M., Smith, C.J., Kilifarski, M., reichert, J., hollander, e., Lawlor, B.A., Fitzgerald, M., Greenberg, D.A. & Davis, K.L. (2001) evidence for a susceptibility gene for autism on chromosome 2 and for genetic heterogeneity. Am J Hum Genet, 68, 1514-1520.

Buyske, S., Bates, M.e., Gharani, N., Matise, t.C., tischfield, J.A. & Manowitz, P. (2006) Cognitive traits link to human chromosomal regions. Behav Genet, 36, 65-76.

Cardon, L.r., Smith, S.D., Fulker, D.W., Kimberling, W.J., Pennington, B.F. & DeFries, J.C. (1994) Quantitative trait locus for reading disability on chromosome 6. Science, 266, 276-279.

Davies, G., tenesa, A., Payton, A., Yang, J., harris, S.e., Liewald, D., Ke, X., Le hellard, S., Christoforou, A., Luciano, M., McGhee, K., Lopez, L., Gow, A.J., Corley, J., redmond, P., Fox, H.c., Haggarty, P., Whalley, L.J., McNeill, G., Goddard, M.e., espeseth, t., Lundervold, A.J., reinvang, I., Pickles, A., Steen, V.M., Ollier, W., Porteous, D.J., horan, M., Starr, J.M., Pendleton, N., Visscher, P.M. & Deary, I.J.

(2011) Genome wide association studies establish that human intelligence is highly heritable and polygenic. Mol Psychiatry, 16, 996-1005.

Deary, I.J., Spinath, F.M. & Bates, t.C. (2006) Genetics of intelligence. Eur J Hum Genet, 14, 690-700.

Deary, I.J., Strand, S., Smith, P. & Fernandes, C. (2007) Intelligence and educational achievement. Intelligence, 35, 13-21.

Deary, I.J., Yang, J., Davies, G., harris, S.e., tenesa, A., Liewald, D., Luciano, M., Lopez, L.M., Gow, A.J., corley, J., redmond, P., Fox, h.C., rowe, S.J., haggarty, P., McNeill, G., Goddard, M.e., Porteous, D.J., Whalley, L.J., Starr, J.M. & Visscher, P.M. (2012) Genetic contributions to stability and change in intelligence from childhood to old age. Nature, 482, 212-215.

Dick, D.M., Aliev, F., Bierut, L., Goate, A., rice, J., hinrichs, A., Bertelsen, S., Wang, J.C., Dunn, G., Kuperman, S., Schuckit, M., Nurnberger, J., Jr., Porjesz, B., Beglieter, h., Kramer, J. & hesselbrock, V. (2006) Linkage analyses of IQ in the collaborative study on the genetics of alcoholism (COGA) sample. Behav Genet, 36, 77-86.

Duncan, J. (2010) the multiple-demand (MD) system of the primate brain: mental programs for intelligent behaviour. Trends Cogn Sci, 14, 172-179.

Duncan, J., emslie, h., Williams, P., Johnson, r. & Freer, C. (1996) Intelligence and the frontal lobe: the

ChAPter 1

26

organization of goal-directed behavior. Cogn Psychol, 30, 257-303.

Duncan, J. & Owen, A.M. (2000) Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci, 23, 475-483.

Durbin, r.M., Abecasis, G.r., Altshuler, D.L., Auton, A., Brooks, L.D., Gibbs, r.A., hurles, M.e. & McVean, G.A. (2010) A map of human genome variation from population-scale sequencing. Nature, 467, 1061-1073.

Fagerheim, t., raeymaekers, P., tonnessen, F.e., Pedersen, M., tranebjaerg, L. & Lubs, h.A. (1999) A new gene (DYX3) for dyslexia is located on chromosome 2. J Med Genet, 36, 664-669.

Farzaneh-Far, A., Davies, J.D., Braam, L.A., Spronk, h.M., Proudfoot, D., Chan, S.W., O’Shaughnessy, K.M., Weissberg, P.L., Vermeer, C. & Shanahan, C.M. (2001) A polymorphism of the human matrix gamma-carboxyglutamic acid protein promoter alters binding of an activating protein-1 complex and is associated with altered transcription and serum levels. J Biol Chem, 276, 32466-32473.

Fisher, S.e., Stein, J.F. & Monaco, A.P. (1999) A genome wide search strategy for identifying quantitative trait loci involved in reading and spelling disability (developmental dyslexia). Eur Child Adolesc Psychiatry, 8 Suppl 3, 47-51.

Gottfredson, L.S. (1997a) Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography (reprinted from the Wall Street Journal, 1994). Intelligence, 24, 13-23.

Gottfredson, L.S. (1997b) Why g Matters: the Complexity of everyday Life. Intelligence, 24, 79-132.

Gottfredson, L.S. & Deary, I.J. (2004) Intelligence Predicts health and Longevity, but Why? Current Directions in Psychological Science, 13, 1-4.

Gray, J.r., Chabris, C.F. & Braver, t.S. (2003) Neural mechanisms of general fluid intelligence. Nat Neurosci, 6, 316-322.

Grigorenko, e.L., Wood, F.B., Meyer, M.S., hart, L.A., Speed, W.C., Shuster, A. & Pauls, D.L. (1997) Susceptibility loci for distinct components of developmental dyslexia on chromosomes 6 and 15. Am J Hum Genet, 60, 27-39.

IMGSAC (2001) A genomewide screen for autism: strong evidence for linkage to chromosomes 2q, 7q, and 16p. Am J Hum Genet, 69, 570-581.

Kaminen, N., hannula-Jouppi, K., Kestila, M., Lahermo, P., Muller, K., Kaaranen, M., Myllyluoma, B., Voutilainen, A., Lyytinen, h., Nopola-hemmi, J. & Kere, J. (2003) A genome scan for developmental dyslexia confirms linkage to chromosome 2p11 and suggests a new locus on 7q32. J Med Genet, 40, 340-345.

Knight, J.C., Udalova, I., hill, A.V.S., Greenwood, B.M., Peshu, N., Marsh, K. & Kwiatkowski, D. (1999a) A polymorphism that affects OCt-1 binding to the tNF promoter region is associated with severe malaria. Nat Genet, 22, 145-150.

Knight, S.J., regan, r., Nicod, A., horsley, S.W., Kearney, L., homfray, t., Winter, r.M., Bolton, P. & Flint, J. (1999b) Subtle chromosomal rearrangements in children with unexplained mental retardation. Lancet, 354, 1676-1681.

Luciano, M., Wright, M.J., Duffy, D.L., Wainwright, M.A., Zhu, G., evans, D.M., Geffen, G.M., Montgomery,

G.W. & Martin, N.G. (2006) Genome wide scan of IQ finds significant linkage to a quantitative trait locus on 2q. Behav Genet, 36, 45-55.

Luerding, r., Boesebeck, F. & ebner, A. (2004) Cognitive changes after epilepsy surgery in the posterior cortex. J Neurol Neurosurg Psychiatry, 75, 583-587.

Luteijn, F. & Ploeg, F.A.e. (1983) GIT: Groninger Intelligentie Test : handleiding, Swets & Zeitlinger bv.

Piercy, M. (1964) the effects of Cerebral Lesions on Intellectual Function: A review of Current research trends. Br J Psychiatry, 110, 310-352.

Plomin, r. (1999) Genetics and general cognitive ability. Nature, 402, C25-29.


27

Plomin, r. & Kovas, Y. (2005) Generalist genes and learning disabilities. Psychol Bull, 131, 592-617.

Plomin, r. & Spinath, F.M. (2004) Intelligence: genetics, genes, and genomics. J Pers Soc Psychol, 86, 112-129.

Plomin, r., turic, D.M., hill, L., turic, D.e., Stephens, M., Williams, J., Owen, M.J. & O’Donovan, M.C. (2004) A functional polymorphism in the succinate-semialdehyde dehydrogenase (aldehyde dehydrogenase 5 family, member A1) gene is associated with cognitive ability. Mol Psychiatry, 9, 582-586.

Posthuma, D., de Geus,e.J.C. (2006) Progress in the Molecular-Genetic Study of Intelligence. Current Directions in Psychological Science, 15, 151-155.

Posthuma, D., Luciano, M., Geus, e.J., Wright, M.J., Slagboom, P.e., Montgomery, G.W., Boomsma, D.I. & Martin, N.G. (2005) A genomewide scan for intelligence identifies quantitative trait loci on 2q and 6p. Am J Hum Genet, 77, 318-326.

Savitz, J., Solms, M. & ramesar, r. (2006) the molecular genetics of cognition: dopamine, COMt and BDNF. Genes Brain Behav, 5, 311-328.

Sebat, J., Lakshmi, B., Malhotra, D., troge, J., Lese-Martin, C., Walsh, t., Yamrom, B., Yoon, S., Krasnitz, A., Kendall, J., Leotta, A., Pai, D., Zhang, r., Lee, Y.h., hicks, J., Spence, S.J., Lee, A.t., Puura, K., Lehtimaki, t., Ledbetter, D., Gregersen, P.K., Bregman, J., Sutcliffe, J.S., Jobanputra, V., Chung, W., Warburton, D., King, M.C., Skuse, D., Geschwind, D.h., Gilliam, t.C., Ye, K. & Wigler, M. (2007) Strong association of de novo copy number mutations with autism. Science, 316, 445-449.

Shettleworth, S.J. (2009) Cognition, Evolution, and Behavior, Oxford University Press, USA.

tournamille, C., Colin, Y., Cartron, J.P. & Le Van Kim, C. (1995) Disruption of a GAtA motif in the Duffy gene promoter abolishes erythroid gene expression in Duffy-negative individuals. Nat Genet, 10, 224-228.

van den Oord, e.J., Sullivan, P.F., Jiang, Y., Walsh, D., O’Neill, F.A., Kendler, K.S. & riley, B.P. (2003) Identification of a high-risk haplotype for the dystrobrevin binding protein 1 (DtNBP1) gene in the Irish study of high-density schizophrenia families. Mol Psychiatry, 8, 499-510.

Van Lawick-Goodall, J. (1973) the behavior of chimpanzees in their natural habitat. Am J Psychiatry, 130, 1-12.

Van Lawick-Goodall, J., Daniel S. Lehrman, r.A.h. & evelyn, S. (1971) tool-Using in Primates and Other Vertebrates. Advances in the Study of Behavior. Academic Press, pp. 195-249.

Vinkhuyzen, A.A., van der Sluis, S. & Posthuma, D. (2011) Life events moderate variation in cognitive ability (g) in adults. Mol Psychiatry, 16, 4-6.

Wainwright, M.A., Wright, M.J., Luciano, M., Montgomery, G.W., Geffen, G.M. & Martin, N.G. (2006) A linkage study of academic skills defined by the Queensland core skills test. Behav Genet, 36, 56-64.

Wechsler, D. (1991) Examiner’s manual, Wechsler intelligence scale for children, Psychological Corporation, New York.

Wechsler, D. (1997) WAIS-III Wechsler Adult Intelligence Scale, Psychological Corporation, San Antonio.

Ylisaukko-Oja, t., Peyrard-Janvid, M., Lindgren, C.M., rehnstrom, K., Vanhala, r., Peltonen, L., Jarvela, I. & Kere, J. (2005) Family-based association study of DYX1C1 variants in autism. Eur J Hum Genet, 13, 127-130.

Zechner, U., Wilda, M., Kehrer-Sawatzki, h., Vogel, W., Fundele, r. & hameister, h. (2001) A high density of X-linked genes for general cognitive ability: a run-away process shaping human evolution? Trends Genet, 17, 697-701.

genetic mechanisms of intelligence

Documents