47 terms

G and B: Imaging genetics

lecture slides

Terms in this set (...)

Sample exam question
1. How are genetic imaging studies carried out?

2. Describe how genetic research using neuroimaging might contribute to the
understanding of psychiatric illness.

3. Discuss how the use of imaging in genetic research has provided us with
new insights into how genetic risk might operate, with examples of specific
Imaging studies
Imaging studies involve the use of techniques that allow the observation of structure and online activity in the brain.

This can be a useful method within genetic research as though there are phenotypes; expressions of particular genetic sequence that are external to the body and are easily discerned with the eye, there are many that are internal, known as endophenotypes that require the use of specific equipment to monitor individual differences inside of the body.
These differences can be endocrinological, biochemical, cognitive, neuroanatomical and neurophysical and brain imaging can be used to observe the neurophysical variations.

Endophenotype is a genetic epidemiology term which is used to parse behavioral symptoms into more stable phenotypes with a clear genetic connection.
• Endocrinological
• Biochemical
• Cognitive
• Neuroanatomical
• Neurophysiological (Electrophysiology, Positron emission tomography (PET) and Functional magnetic resonance imaging (fMRI))
What is deoxyribonucleic acid?
A nucleic acid containing the genetic instructions used in the development and functioning of all known living organisms (with the exception of RNA viruses). The DNA segments carrying this genetic information are called genes. Likewise, other DNA sequences have structural purposes, or are involved in regulating the use of this genetic information. Along with RNA and proteins, DNA is one of the three major macromolecules that are essential for all known forms of life.
What does DNA do?
• Deoxyribonucleic acid (DNA)
• Ribonucleic acid (RNA)
• Translator mechanism (ribosome)
• Amino acid sequence
• Protein folding

WIKI: DNA consists of two long polymers of simple units called nucleotides, with backbones made of sugars and phosphate groups joined by ester bonds. These two strands run in opposite directions to each other and are therefore anti-parallel. Attached to each sugar is one of four types of molecules called nucleobases (informally, bases). It is the sequence of these four nucleobases along the backbone that encodes information. This information is read using the genetic code, which specifies the sequence of the amino acids within proteins. The code is read by copying stretches of DNA into the related nucleic acid RNA in a process called transcription.

Within cells DNA is organized into long structures called chromosomes. During cell division these chromosomes are duplicated in the process of DNA replication, providing each cell its own complete set of chromosomes. Eukaryotic organisms (animals, plants, fungi, and protists) store most of their DNA inside the cell nucleus and some of their DNA in organelles, such as mitochondria or chloroplasts. In contrast, prokaryotes (bacteria and archaea) store their DNA only in the cytoplasm. Within the chromosomes, chromatin proteins such as histones compact and organize DNA. These compact structures guide the interactions between DNA and other proteins, helping control which parts of the DNA are transcribed.
What are we measuring in DNA?
• GATC code controls protein structure
• An area coding for a protein is called a gene
• This code can vary between individuals (polymorphism)
• Variants are called alleles (many in each gene). We have two copies of almost every gene (which may be different alleles)
• Alleles can cause:
-Different shaped proteins
-Different amounts of protein
Why are we all so similar? Why are
we all so different?
• More than 99.9 % of human DNA sequence is the same across the population
• The 0.1% accounts for individual differences
• Variations in the genome (such as a single nucleotide polymorphisms - SNPs) are
found about every 1000th base pair on average
• These polymorphisms (multiform) account for much of individual differences in the risk for psychiatric disorders
Such variations are attributed to small discrepancies in the genotypes of an individual and it appears that only 0.1% of the DNA sequence accounts for the differences between humans (The 1000 Genomes Project Consortium, 2010).

The sequence of a gene may only differ in one nucleotide base pair - a single nucleotide polymorphism, creating a variant of the gene - an allele that will shape the subsequent protein and thus can alter how it functions. As approximately 70% of all genes are expressed in the brain, and as many of those have polymorphisms, these could potentially account for individual differences in the way the brain processes information, in addition to identifying susceptibility to psychiatric diseases.
Schizophrenia has high heritability
Why large-scale studies of psychiatric genetics have failed
• Polygenic, heterogeneous disorders
• Generally weak risk effects
• Epistasis and gene-environment interactions
• Rare, highly influential alleles not captured by GWAS
• Genome-wide association study: examination of many common genetic variants in different individuals to see if any variant is associated with a trait. GWAS typically focus on associations between single-nucleotide polymorphisms (SNPs) and traits like major diseases.
In comparison to standard behavioural genetic studies
In comparison to standard behavioural genetic studies that often utilise the classic twin design, where monozygotic and dizygotic twins are evaluated on their similarity in task performance to examine genetic and environmental contribution to a particular behaviour, imaging genetics can go further by correlating this behaviour to specific brain areas through the knowledge of neurotransmitter systems and networks that are known to be involved.

This information can be used to then map systems in the brain to the genes that control their development.
Quantitative trait loci model
The commonly accepted method of mapping behaviour to genes is the quantitative trait loci model. The model recognises that variation within a trait is likely to lie on a continuum that carries the bell-shape curve of a normal distribution, rather than the trait falling into a discrete classification.

The 'cut-off point' to state what is considered to be abnormal is then imposed and it is assumed that a number of polymorphisms will be more abundant in the abnormal group compared to the controls. Thus, when attempting to map psychological function to genotype, the ability to observe brain function is a helpful intermediate stage.
The quantitative trait locus (QTL) model for common complex disorders
Why put the brain in between?
• 70% of all genes expressed in the brain, many of these genes are polymorphic
• These functional polymorphisms can
account for individual differences in how the brain processes information
Imaging genetics: candidate gene
association approach
• 'Candidate gene association approach' - relationship between a phenotype and specific allele of a gene
• 'Imaging genetics' is just another type
of genetic association analysis
• Phenotype is the structure or physiological response of the brain
MRI links genes and behaviour
Heritability in neurobiology
Why should imaging genetics be more successful than other approaches?
• Traditionally, the impact of genetic polymorphisms on psychiatric well-being has been examined using behavioural ratings
• Results often weak and inconsistent
• Large samples required
• People can manifest the same behaviour for several underlying reasons
• Brain regions subserving specific cognitive and emotional processes are more objectively measurable than the subjective rating of the same process
• Genetic effects may be more easily apparent at the level of brain - more 'signal to noise'
Advantages of imaging studies
• Require fewer participants due to more 'signal to noise' - this is good as imaging is
• Can examine specificity of gene effects to multiple functional systems (e.g. prefrontal, striatal, limbic)
• 'in vivo' study of functional genetic variation
The protocol for imaging genetics Step 1
Having discussed why this brain imaging method is useful, particular protocol must be carried out to ensure that the data too is useful.

Step 1: Selection of candidate genes:
• Well-defined functional polymorphisms,
associated with specific physiological
effects at the cellular level
• Genes containing allele variants with likely functional implications involving
circumscribed neuroanatomical systems

The genes and alleles that are suspected to be causing a specific variation, based on its protein output. To select a candidate gene, the polymorphisms must be well established and connected with specific physiological effects that are visible at a cellular level.

Also, the gene must have alleles that can potentially cause alternative functioning within neuroanatomical systems. An example of a candidate gene is discussed by Winterer and Weinberger (2004) with regards to the catechol-O-methyltransferase (COMT) gene as it is recognised as coding for the important COMT enzyme that controls the removal of dopamine in the prefrontal cortex.

Variation in this gene has been related to differences in prefrontal cortex function and a higher risk for schizophrenia, where dopamine plays an integral role.
The protocol for imaging genetics Step 2
Control for non-genetic factors
• Any gene affecting brain response will not explain all of the variance in the outcome measure
• Systematic differences between genotype groups could either obscure a true gene effect or masquerade for one
• Age, gender, IQ, population stratification
• Environmental factors such as illness, injury, or substance abuse
• Task performance
-Linked with BOLD response
-Match or consider variability

A single gene that affects brain response will not be able to account for all of the variance in the outcome measure. Other regular dissimilarities between genotype groups have the potential to cloud or cover true genetic effect.

This can include sex and age, environmental factors like injury or substance abuse or task performance before the study. To avoid this, participants should be matched or in the least, the validity of the study should be considered.
The protocol for imaging genetics: Step 3
Task selection
Imaging tasks must maximize sensitivity and inferential value, as the interpretation of potential gene effects depends on the validity of the information processing paradigm
• Engage circumscribed brain circuits
• Produce robust signals
• Show variance across subjects

Tasks must should be of maximum sensitivity and have great inferential value as the reading of possible gene effects depends on the validity of the information that was processed.

In a study by MacDonald, Thermenos, Barch and Seidman (2009), a study that also considers schizophrenia, the discrepancies between results from previous research is suggested to be from a lack of consistency in the tasks that participants were being told to complete and the cognitive demands that were required for each task.
Replication of any finding of extreme importance
Intact protocol
With the protocol intact, experimenters then need to select the technique that they find most appropriate to image the brain. Positron Emission Tomography (PET), electrophysiology and function magnetic resonance imaging (fMRI) are all methods that can be used.

A study by Wallace et al. (2006) that investigated the heritability in neurobiology and used electroencephalography (EEG) to measure certain neural connections. EEG involves a number of electrodes placed around the scalp that record electrical signals that occur in certain areas of the brain. This technique displays the neural oscillations at each place in the brain and the experimenter can observe abnormalities or spikes of activity this way, although this method is only suitable for surface structures in the brain as it is not particularly intrusive.
Though there is more than one measure available, many if not most experimenters tend to elect fMRI. This is likely due to fMRI having a higher temporal resolution than PET whilst also displaying functionality on an image of a brain, compared to the labelled but dissociated lines of EEG.
Allele examples
Allele examples: Serotonin transporter gene
• Recycles serotonin
• Long and short alleles (l and s)
• l allele is more efficient at producing serotonin transporter protein

A well explored illustration of fMRI use is within the investigation of a serotonin transporter (5-HTTLPR) gene. Lesch et al. (1996) found that the gene was involved with the recycling of serotonin, but that there were variations of this gene - long (la) and short (s) alleles.

The long allele was found to be more efficient at reusing serotonin whereas short allele carriers appeared to be at a greater risk of anxiety and depression. Hariri et al (2002) found that short allele carriers also showed higher resting-state brain activity and greater response to fearful faces in the amygdala. As the ability to replicate is an important factor in all psychological findings, this study and its results have been replicated on several occasions.
Serotonin transporter gene and
s allele carriers are at greater risk
for anxiety and depression (gene x environment interaction)
Serotonin transporter gene and
the amygdala
• s allele carriers show increased resting-state brain activity in amygdala
• s allele carriers show increased response to fearful faces in amygdala
Serotonin transporter gene and
the amygdala - replication 1
Serotonin transporter gene and
the amygdala - replication 2
Serotonin transporter gene and
the amygdala - replication 3
(social phobics)
In addition to this Furmark et al (2004) demonstrated the same effect in social phobics, who are more likely to have the s allele and Danlowski et al (in press) extended the effect to even heterozygote carriers, who have both one la allele and one s allele and found that s/l carriers were had a greater response to sad stimuli.
Serotonin transporter gene and
the amygdala - replication 4 (sad
Serotonin transporter gene and
the amygdala
• Increased amygdala activity observed in s
allele carriers
• Hyper-responsiveness of amygdala to emotional environmental stimuli (Munafo
et al 2008)
• Does amygdala activity relate to harm-avoidance?

Munafo, Brown and Hariri (2008) again observed hyper-responsiveness in that amygdala to emotional environmental stimuli, which could perhaps suggest that amygdala activity is associated to the avoidance of harm.
Amygdala reactivity and harm
No correlation between amygdala reactivity and harm avoidance

However, Hariri et al (2005) already noted that there seemed to be no correlation between activity in the amygdala and harm avoidance in either hemisphere. Wood and Grafman (2003) explain that the amygdala is part of a network and that it does not work in isolation. It is connected with the inferior temporal visual association areas, the ventromedial prefrontal cortex and most relevantly, with the anterior cingulate cortex (ACC).
But the amygdala is not alone...
Amygdala-anterior cingulate coupling predicts harm avoidance
Reduced coupling of the amygdala and anterior cingulate in S allele carriers
Pezawas et al (2005) noted that there was less coupling and interaction between the amygdala and the ACC in s allele carriers, and that this interaction is a better predictor of harm avoidance.
5-HTTLPR gene and other cognitive functions
Moreover, Roiser et al. (2009) extend this pattern to other cognitive functions. As the 5-HTTLPR gene is also associated with less prefrontal regulatory control, it was thought that these regions would overlap with those that are involved in decision making biases. The study showed that those in the s/s group were significantly more likely to exercise the framing effect; a bias dependent on contextual cues and ambiguity where participants would tend towards the certain monetary option when the text mentioned 'gaining' and would tend towards the gambling option when the text mentioned 'losing'.

This effect was also measured by simultaneously collecting fMRI data which again showed greater amygdala activity when using the those in the s/s group were using the framing effect, a pattern not seen in the la/la group. And when choosing to counter the effect, the long allele group showed significant interaction between the amygdala and ACC opposed to when they were following the bias, a pattern not observed in the short allele group.
Conclusion: Serotonin transporter gene
5-HTTLPR genotype affects prefrontal
control over amygdala reactivity, leading to altered emotional experience and temperament

Hamann (2005) concludes that the 5-HTTLPR genotype influences the control of the prefrontal lobes over amygdala reactivity which consequently produces an adjusted emotional experience and temperament.
other genetic polymorphisms on other psychological functions
Whilst there has been much focus on the use of fMRI in the examination of the 5-HTTLPR gene, fMRI has been used to uncover the effects of other genetic polymorphisms on other psychological functions. Montag, Reuter, Newport, Elger and Weber (2008) use fMRI to monitor the startle reflex concept in participants viewing emotional stimuli. Earlier studies have shown that the brain development neurotrophic factor (BDNF) gene is an important component in synaptic plasticity, and is likely to be involved in some psychiatric illnesses such as depression and anxiety disorders. The BDNF Vall66Met polymorphism impacts the secretion of BDNF which thus affects the signalling at TrkB receptors that influence the growth of dendrites in neurones.

Structural MRI studies previously revealed that smaller hippocampi is related to the 66Met variant which would suggest that it could be an allele the presents more risk. In this experiment, those with the 66Met variant showed a stronger response in amygdala activation when viewing emotional stimuli, leading to its association with a higher anxiety trait.
Similarly, MacDonald et al. (2008) also considered increased activation in the prefrontal cortex and parietal cortex in relation to schizophrenia, but only in the right hemisphere. In their study that contrasted schizophrenic patients with their healthy, nonpsychotic relatives, fMRI data showed that schizophrenic patients showed much abnormal activity in two thirds of brain area contrasts and was spread over many areas of the brain including the cerebellum, dorsal and ventral prefrontal cortex, parietal cortices, and the thalamus. Although this imaging study could not associate these effects with a gene, the authors suggest that this methodology is essential for understanding the effect of unexpressed genetic liability to schizophrenia and that due to the many areas that are affected, no single abnormal gene, brain region or mechanism has yet emerged.
Tryptophan hydroxylase 2 (TPH2) gene
Reuter, Kuepper and Hennig (2007) investigated attention, working memory and the tryptophan hydroxylase 2 (TPH2) gene that regulates the synthesis of serotonin and has two variants - the G and T polymorphisms. Previous imaging studies have illustrated prefrontal and parietal regions are activated during tasks that demand the use of working memory. Their study found more prefrontal and parietal activation in the more demanding task in all participants, which was expected, but did not find any difference in performance between genotype groups. But although performance did not differ, there were still differences in brain activity between the groups; the homozygote T carriers seemed to show higher levels of activity in those brain areas than those who were homozygote G carriers, or heterozygote carriers. These findings were in accordance with previous evidence and support the hypothesis that homozygote T carriers may need to compensate for deficits in executive control function by with increased brain activity.
Twin design and genetic imaging
Given that the previous examples have all shown a similar approach in addressing their investigation, this next study is interesting in the aspect that it combines imaging techniques with a conventional approach to behavioural genetics; the twin design. Overall, Peper et al. found through structural MRI, that frontal lobe volume was highly heritable. The authors identify previous studies that looked at the association between brain volume and intelligence in accordance to genetics.
Twin Designs: Intelligence
Hulshoff Pol et al (2006) claim that intelligence is connected with the grey matter of the frontal and occipital lobes, the parahippocampus and the white matter joining them, because these are influenced by the genes that are common to both intelligence and brain structure.

However, the use of a twin design allowed the study to be sensitive to environmental contributions, which seemed to be present in elderly twins. There is uncertainty as to whether the genes influence the shape of the brain or whether they influence the environment that the individual chooses to be in, that in turn impacts brain structure.

The study also goes on to provide more instances where imaging genetics has been proven useful. It has helped in understanding the apolipoprotein (ApoE) gene which is believed to aid the growth of cells and the regeneration of nerves, and where the E4 allele that is associated with high risk for Alzheimer's disease. Imaging studies have shown that the E4 allele was associated with smaller hippocampi in healthy elderly participants and generally among women, demonstrating how even those who are not clinically diagnosed could still be impacted. Furthermore, with the monoamine oxidase A-gene that is associated with anti-social behaviour, brain imaging has shown volume reduction in the cingulate gyrus, amygdala, insula and hypothalamus in the low-expression variant of the gene, providing implications of the other functional effects it may have.
What are the next steps?
• Longitudinal studies needed!
• Developing neural circuitry of particular interest
• Taking into account variability in environmental stressors as well as genetic make-up
-System before and after developmental insults
-System before and after hormonal changes
-The effect of a polymorphism in childhood/adolescence
may be very different to that in adulthood!
• Behavioural outcome

Though, imaging genetics has already presented a great amount of new information, Sebastian et al (2010) encourage the field to begin considering longitudinal studies and take interest in developing neural circuitry which will assist in forming a grand picture. And though further research utilising adapted methodology is necessary, it is clear the genetic imaging research is able to contribute significantly to behavioural genetic research by helping to bridge assumptions made when only the genotype and its final psychological outcome are visible.
• Neuroimaging methods allow us to analyze brain structure and function
• Some of these measures are highly heritable
• Genes known to affect neuronal function are associated with psychiatric illness
• Recent neuroimaging genetic studies
suggest plausible neurobiological mechanisms underlying some of these