Human Genetic Variation

As mentioned above, genetic epidemiological studies can only indicate the presence of an aggregate genetic effect but not which type and how many variations contribute to the effect. This next section will give an overview of the types of variation that occur in the human genome and will provide examples for implications of each of them for psychiatric disorders (see also Figure 3-2).

FIGURE 3-2. Chromosomes, genes, and genetic variation.

FIGURE 3-2. Chromosomes, genes, and genetic variation.

Panel A shows a representation of chromosome 6 (Chr6), which spans 170 megabases of DNA. Panel B shows a zoomed-in representation of the area highlighted in red in panel A. This region contains 31 genes within 2 megabases. Genes can be transcribed from both strands of the DNA. The arrows indicate the direction of transcription and translation of the respective genes. In the region shown, three copy number variants (CNVs) have been identified. All three span several genes. Panel C shows a zoomed-in representation of the gene for FK506 binding protein 5 (FKBP5) highlighted with a red frame in panel B. The gene spans 115 kilobases and is composed of 11 exons (translated into protein). The intervening introns are transcribed into RNA but are spliced out to form the mature mRNA that serves as the template for translation. The transcription start is at exon 1. In this gene, more than 60 SNPs have been genotyped within the HapMap Project. Their positions are indicated by the triangles. Panel D shows sequence examples for three common polymorphisms.

Panel A shows a representation of chromosome 6 (Chr6), which spans 170 megabases of DNA. Panel B shows a zoomed-in representation of the area highlighted in red in panel A. This region contains 31 genes within 2 megabases. Genes can be transcribed from both strands of the DNA. The arrows indicate the direction of transcription and translation of the respective genes. In the region shown, three copy number variants (CNVs) have been identified. All three span several genes. Panel C shows a zoomed-in representation of the gene for FK506 binding protein 5 (FKBP5) highlighted with a red frame in panel B. The gene spans 115 kilobases and is composed of 11 exons (translated into protein). The intervening introns are transcribed into RNA but are spliced out to form the mature mRNA that serves as the template for translation. The transcription start is at exon 1. In this gene, more than 60 SNPs have been genotyped within the HapMap Project. Their positions are indicated by the triangles. Panel D shows sequence examples for three common polymorphisms.

Source. Representations from

Variation on a Chromosomal Scale Variation in Chromosomal Number

The human genome has approximately 3 billion bases that are distributed over 23 chromosome pairs, with 22 pairs of autosomes and 1 pair of sex chromosomes, X and Y. The most obvious genetic variations can be observed at the light microscope level in the karyotype. This approach visualizes metaphase chromosomes using histological procedures, allowing identification of each specific pair of chromosomes and variations in the total number of chromosomes, such as unisomies and trisomies. Several of the known variations of total chromosome number have an associated psychiatric phenotype. For example, Down syndrome is a complex neurodevelopmental disorder that results in variable levels of mental retardation, and in old age, dementia strikingly similar to Alzheimer's disease (Visootsak and Sherman 2007). Down syndrome results from trisomy 21 (i.e., inheritance of three copies of chromosome 21, due to meiotic nondysjunction during oogenesis. Turner syndrome, in which there is only a single X chromosome (i.e., an XO karyotype), is associated with nonverbal learning disabilities, particularly in arithmetic, select visuospatial skills, and processing speed (Sybert and McCauley 2004).


Karyotypic examination and other cytogenetic techniques such as fluorescent in situ hybridization (FISH) can reveal additional large-scale chromosomal abnormalities, such as translocations, deletions, or duplications of large regions of chromosomes. In a large Scottish pedigree, a balanced translocation between chromosomes 1 and 11 appears causally linked to a series of major psychiatric disorders, including schizophrenia, bipolar disorder, recurrent major depression, and conduct disorder (St. Clair et al. 1990). This balanced translocation (which exchanged parts of chromosome 1 with parts of chromosome 11 to produce two abnormal chromosomes, but no net loss of chromosomal material) disrupts two genes at the translocation breakpoint on chromosome 1, termed "disrupted in schizophrenia" (DISC) 1 and 2 (Millar et al. 2000, 2001). Subsequent molecular analysis has provided strong evidence that variation in DISC1 can alter the risk for schizophrenia; the locus is presently considered by most a "confirmed" schizophrenia locus (Porteous et al. 2006).


Microdeletions occurring on the long arm of chromosome 22 have received considerable attention as cytogenetic risk factors for the development of schizophrenia (Karayiorgou and Gogos 2004). The 22q11 deletion syndrome (DS), in which 1.5-3 million base pairs of DNA are missing on one copy of 22q, includes a spectrum of disorders affecting structures associated with development of the fourth branchial arch and migration of neural crest cells (e.g., the great vessels of the heart, the oropharynx, facial midline, and thymus and parathyroid glands). Originally described as distinct disease syndromes prior to the elucidation of their common molecular etiology, 22q11DS includes velocardiofacial syndrome (VCFS), DiGeorge syndrome, and conotruncal anomaly face syndrome. Following an initial report of early-onset psychosis in patients with VCFS (Shprintzen et al. 1992), Pulver and colleagues examined psychiatric symptoms in adults with VCFS (Pulver et al. 1994) and in a cohort of patients ascertained for schizophrenia (Karayiorgou et al. 1995). The latter study identified two previously undiagnosed cases in 200 patients, verified by fluorescent in situ hybridization to carry 22q11 deletions. These findings, together with earlier reports of suggestive linkage of 22q11-22q12 (Gill et al. 1996; Pulver et al. 2000), strongly suggested that a gene or genes in the 22q11DS region could contribute to risk for schizophrenia.


Duplications of the long arm of chromosome 15 (15q11-13) are the most frequent cytogenetic anomalies in autism spectrum disorders, occurring in approximately 1%-2% of cases (Cook 2001). This duplication syndrome cannot be clinically differentiated from idiopathic autism spectrum disorders (Veenstra-VanderWeele and Cook 2004), indicating that a complete workup of autism should include testing for this cytogenetic abnormality, as well as for several others (Martin and Ledbetter 2007). Interestingly, deletion of this same region of 15q is associated with Angelman syndrome when the deletion occurs on the maternal copy of chromosome 15, and with Prader-Willi syndrome when the deletion occurs on the paternal chromosome (or more rarely, when two maternal copies of chromosome 15 are present, and the paternal chromosome is missing entirely, a condition known as maternal disomy). Both syndromes manifest as quite distinct but dramatic neurobehavioral disorders (Nicholls and Knepper 2001; Vogels and Fryns 2002).

Molecular Variation in the Genome

The majority of genetic and genomic studies in neuropsychiatry conducted to date have examined variation at the molecular level, which would be undetectable with methods appropriate for the kinds of variation described above. To introduce this section, we provide basic definitions of terms.

Definition of Alleles, Genotypes, Haplotypes

The definition of alleles, genotypes, and haplotypes is common to all the types of polymorphisms discussed below. An allele is a variation in DNA sequence that occurs at a particular polymorphic site on one chromosome. Every individual with a normal set of chromosomes has two alleles for each polymorphism on the autosomes (nonsex chromosomes, numbers 1-22). On the sex chromosomes, men have only one allele each for all polymorphisms located on the X and Y chromosomes, whereas women carry two copies of each X-linked allele. A genotype is the combined description for the variation at a particular corresponding point on homologous chromosomes and is expressed as two alleles. When the alleles on both chromosomes are the same, it is a homozygous genotype. When the alleles differ, it is a heterozygous genotype. A haplotype, a term derived from abbreviation of "haploid genotype," is the sequence of alleles along an adjacent series of polymorphic sites on a single chromosome. When genotypic data are available from three generations, haplotypes in the third generation can be unambiguously deduced. In the absence of sufficient family-based data (e.g., in case-control studies of unrelated individuals), some haplotypes are ambiguous because the combination of genotypes at the polymorphic sites under study can be explained by more than one set of possible chromosomal arrangements of the component alleles. In such cases, methods such as estimation maximization (EM) can be used to infer the most likely haplotype (Hawley and Kidd 1995; Long et al. 1995).

Copy Number Variation

Genome-scale investigations enabled by the sequencing of the human genome and the advent of microarray-based comparative genomic hybridization have recently revealed a previously unappreciated form of polymorphic variation in the human genome: chromosomal regions containing one or more genes can sometimes be deleted or, alternatively, occur in multiple copies, with the number of copies differing among individuals (Nadeau and Lee 2006; Sebat et al. 2004). Such copy number variants (CNVs) occur normally in human populations, and a preliminary map of such variants is now available (Redon et al. 2006). They have recently been associated with marked differences in gene expression (Stranger et al. 2007). CNVs can also be associated with predisposition to disease, including neurobehavioral disorders such as autism and schizophrenia (International Schizophrenia Consortium 2008; Sebat et al. 2007; Stefansson et al. 2008). Research in this exciting new area is in its infancy but has already contributed importantly to the genetics of psychiatric disorders (Cook and Scherer 2008).

Copy number variation of the cytochrome P450 gene CYP2D6, which is important for the metabolism of many antidepressants, neuroleptics, and mood stabilizers (Kirchheiner et al. 2004), provides a prominent example of the importance of CNVs to pharmacogenetics. The presence in the genome of copy-number variation at this locus was inferred through biochemical-genetic studies predating the molecular era and was subsequently confirmed by molecular studies. The reported range of copy numbers of CYP2D6 is from 0 to 13. The number of functional CYP2D6 gene copies directly correlates with plasma levels of metabolized drugs, such as the TCA nortriptyline (Bertilsson et al. 2002). Patients with 0 or 1 functional copy of the gene attain therapeutic plasma levels of nortriptyline with very low doses and can easily reach potentially toxic concentrations with typical or high doses. Patients with 2-4 copies, on the other hand, would require high-normal doses to even reach therapeutic plasma levels (Kirchheiner et al. 2001). In the case of the one reported patient with 13 gene copies, even high-normal doses did not produce therapeutic plasma concentrations (Dalen et al. 1998).

Insertion/Deletion Polymorphisms

Microscopic insertions and deletions (much smaller than CNVs—on the order of one to hundreds of base pairs [bp]) are another important type of genetic variation. The most famous insertion/deletion polymorphism in psychiatric genetics is a common functional polymorphism in the promoter region of the serotonin transporter gene SLC6A4, referred to as the 5-HT transporter gene-linked polymorphic region (5-HTTLPR). It consists of a repetitive region containing 16 imperfect repeat units of 22 bp, located approximately 1,000 bp upstream of the transcriptional start site (Heils et al. 1996; Lesch et al. 1996). The 5-HTTLPR is polymorphic because of the insertion/deletion of the repeat units 6-8 (of the 16 repeats), which produces a short (S) allele that is 44 bp shorter than the long (L) allele. Although the 5-HTTLPR was originally described as biallelic, rare (<<5%) very-long and extra-long alleles have been described in Japanese and African Americans (Gelernter et al. 1999). Numerous additional variants within the repetitive region also occur (Nakamura et al. 2000). Thus, although most studies continue to treat this complex region as biallelic, this is an oversimplification that may be hiding additional genetic information. The 5-HTTLPR has been associated with different basal activity of the transporter, most likely related to differential transcriptional activity (Heils et al. 1996; Lesch et al. 1996). The long variant (L allele) of this polymorphism has been shown to lead to a higher serotonin reuptake by the transporter in vitro. It is also noteworthy that the function of this insertion/deletion polymorphism may be influenced by a single nucleotide polymorphism (SNP) that occurs with the L allele (Hu et al. 2006). However a positron emission tomography study could not identify differences in serotonin transporter binding potential by 5-HTTLPR genotype, even when including the information of the additional SNP, in healthy control subjects or patients with major depression (Parsey et al. 2006). This polymorphism has shown associations with a multitude of psychiatric disorders and related phenotypes. The best established are an association with response to treatment with SSRI (Serretti et al. 2007) and the moderation of the influence of life events on the development of depression (Caspi et al. 2003; Kaufman et al. 2004; Kendler et al. 2005; Surtees et al. 2006; Wilhelm et al. 2006; Zalsman et al. 2006).

Microsatellites: STRs and VNTRs

A very important class of polymorphisms, upon which molecular linkage studies and some association studies were based until very recently, is microsatellite markers, also called short tandem repeats (STRs) or variable number of tandem repeats (VNTRs). The polymorphisms consist of simple sequences, such as GT or GATA, that are repeated a variable number of times. An individual may thus have 5 GT repeats at a specific locus on one chromosome and 7 such repeats on the other. When these regions are amplified by polymerase chain reaction (PCR), a technique for producing many copies of a specific small portion of the genome based on DNA polymerase activity directed by specific DNA sequences, the difference in number of repeats results in differences in the length of the amplified fragments (a difference of 4 bp in the current example), allowing efficient genotyping of these polymorphisms by gel electrophoresis, which separates DNA fragments according to length. About 30,000 of these polymorphisms are now known in the human genome (Kawashima et al. 2006; Tamiya et al. 2005), and they have served as markers for genomewide linkage analysis (see "Linkage Studies" subsection below for more detail). VNTRs, however, not only serve as genetic markers for linkage analysis but also may produce functional variation within genes. An important example of such functional variation is a VNTR in the 3' untranslated region (UTR) of the dopamine transporter gene (DAT). The repeat element consists of a 40-bp sequence that can occur with 3-11 repeats, with 9 and 10 repeats being the most common (Vandenbergh et al. 1992). Different effects of the 9 or 10 repeats on gene expression and DAT binding using single photon emission computed tomography (SPECT) in humans have been reported, although the direction of these differences is controversial (Greenwood and Kelsoe 2003; Inoue-Murayama et al. 2002; Martinez et al. 2001; Mill et al. 2002, 2005; van Dyck et al. 2005; VanNess et al. 2005), with the 9 or the 10 repeat alleles respectively associated with higher expression of the DAT gene and higher DAT binding in different studies. Another example of a functional VNTR (48-bp repeat) is a polymorphism in the third exon of the DRD4 locus (the locus encoding the D4 dopamine receptor), which results in a variable number of glutamine residues in the third intracellular loop of the dopamine D4 receptor protein (Van Tol et al. 1992). The allelomorphic proteins differ in their ligand-binding affinities, but all couple to G proteins (Van Tol et al. 1992). Both polymorphisms have been associated with a multitude of psychiatric and behavioral phenotypes (many of which require further investigation before their validity can be established with confidence).

Single Nucleotide Polymorphisms

The polymorphisms that have revolutionized (not only) psychiatric genetics are SNPs (Altshuler et al. 2000; Sachidanandam et al. 2001). SNPs consist of a single-base difference at a particular site in the genome—in two-thirds of cases, a cytosine (C)-to-thymidine (T) exchange. Although theoretically the presence of all four different bases at an SNP is possible, the vast majority of SNPs have only two alleles, although SNPs with three or four different alleles have occasionally been reported. As of June 2008, close to 10 million SNPs have been catalogued in public databases (such as dbSNP []). SNPs are so far the most common type of genetic variation and may represent up to 90% of all genetic variations (although this estimate may need revision as knowledge about CNVs accumulates). Besides being very common (SNPs occur, on average, every 300 bases), SNPs are also amenable to high-throughput genotyping methods (Kim and Misra 2007; Kwok 2000). Since SNPs are common (essentially every gene of interest has a number of known SNPs) and cheap to genotype, they have become the markers of choice for psychiatric genetic studies. SNPs are now replacing STRs in genomewide linkage studies, as more genetic information can be gleaned from 5,000-6,000 common SNPs across the genome than from the 300-400 markers typical of STR-based genome scans. SNPs are also the backbone of genetic association analysis.

Besides serving as genetic markers for chromosomal loci in association studies, SNPs can also have functional relevance themselves. SNPs in regulatory regions can alter the transcriptional regulation of a gene, SNPs in regions relevant for mRNA splicing can alter splice sites, and SNPs in protein-coding exons can encode differences in primary amino acid sequence. Interestingly, when one considers the occurrence of SNPs across the genome, one observes that SNPs are most dense in intergenic and intronic regions, while they are scarcer in putative regulatory regions and exons, suggesting selection against putatively functional variants (Altshuler et al. 2000; Sachidanandam et al. 2001). Within exons, SNPs that will not cause an amino acid exchange due to the degenerate code are called synonymous SNPs as opposed to nonsynonymous SNPs, which lead to amino acid substitution. One nonsynonymous SNP that has repeatedly been associated with psychiatric phenotypes is a valine-to-methionine exchange at amino acid position 108/158 in the soluble (S) or membrane-bound (MB) form of the catechol-O-methyltransferase (COMT) peptide sequence due to a G-to-A exchange at position 472 of exon 4. This amino acid exchange dramatically affects the temperature lability of this enzyme, with the methionine allelomorphic protein having only one-fourth of the enzyme activity at 37°C as the valine allelomorph (Lachman et al. 1996; Lotta et al. 1995). It is presumed that individuals with the Val/Val genotype have a more rapid inactivation of centrally released dopamine than individuals with the other genotypes, following an additive genetic model. This polymorphism seems to impact dopamine transmission, particularly in the frontal cortex, where expression of the dopamine transporter protein is relatively low, and has been associated with differences in executive cognition and functional activity of the prefrontal cortex during working memory tasks (Egan et al. 2001; Goldberg et al. 2003; Joober et al. 2002; Mattay et al. 2003). While it was hypothesized that this polymorphism could also alter the risk for schizophrenia (particularly since COMT resides in the 22q11 region, the deletion of which has been associated with schizophrenia), larger association studies and meta-analyses do not fully support that conclusion (Fan et al. 2005).

Anxiety and Depression 101

Anxiety and Depression 101

Everything you ever wanted to know about. We have been discussing depression and anxiety and how different information that is out on the market only seems to target one particular cure for these two common conditions that seem to walk hand in hand.

Get My Free Ebook

Post a comment