Discussion Post Help

Genes in Context Gene–Environment Interplay and the Origins of Individual Differences in Behavior Frances A. Champagne and Rahia Mashoodh

Columbia University

ABSTRACT—Interactions between genes and the environ-

ment are a critical feature of development. Insights into the

dynamic interplay between these factors have come from

laboratory studies exploring experience-dependent chan-

ges in gene function, which illustrate the importance of

environmental factors in determining activity of the genome.

These studies have implications for our understanding of

the origins of individual differences in behavior and may

provide new ways of thinking about the transmission of

traits across generations. Here we will highlight how these

new findings illustrate the importance of putting genes in

context.

KEYWORDS—epigenetic; gene–environment; DNA methyl-

ation; inheritance; individual differences

Historically, the question of the origins of individual differences

in personality, aptitudes, and even physical features has led

to debates over nature versus nurture. However, it is becoming

increasingly clear that creating a division between genes and

environment limits our understanding of the complex biological

processes through which individual differences are achieved.

The reality that the interaction between genes and environment

is a critical feature of development is emerging as a central

theme in laboratory studies and longitudinal analyses in human

populations. However, appreciating the existence of this inter-

action is simply the first step in broadening our theoretical ap-

proach to the study of behavior. To move forward, we must ask

‘‘What do genes do?’’ and ‘‘How do genes and environments

interact?’’ Recent studies combining molecular biology with the

study of behavior may provide insight into these issues and

perhaps even call into question our current understanding of

mechanisms involved in the transmission of traits across gen-

erations. Here we will highlight these new findings and illustrate

the importance of putting genes in context.

LABORATORY AND LONGITUDINAL APPROACHES TO

GENE–ENVIRONMENT INTERACTIONS

Though recentadvances in our ability to detect genetic variations

have led to rapid progress in the study of gene-by-environment

(G � E) effects, clues that G � E was critical in considering the origins of behavior have been available for a long time. In 1958,

Cooper and Zubek published a report in which rats selectively

bred to be either ‘‘maze-dull’’ or ‘‘maze-bright’’ were reared after

weaning in either ‘‘enriched’’ environments containing increased

sensory stimuli or ‘‘impoverished’’ environments containing

limited sensory stimuli (Cooper & Zubek, 1958). In the rats

reared under standard conditions, stable and heritable group

differences in cognitive ability were observed in adulthood.

However, maze-dull animals reared in an enriched environment

showed a significant improvement in learning ability, and maze-

bright animals reared under impoverished conditions showed a

significant decline in performance. This study provides evidence

that, even when considering a genetically derived characteristic,

our prediction of behavior must incorporate knowledge of the

environmental context of development.

A more recent example of G � E comes from the Dunedin longitudinal study (Caspi et al., 2003), which explored the roles

of variation in a gene that alters serotonin levels and exposure to

stressful life events across a 20-year period in determining risk of

depression. Levels of serotonin within neural circuits are altered

by the number of serotonin transporter proteins, and in humans

there aregeneticvariations that leadtoeitherhigh or lowlevelsof

the serotonin transporter. The serotonin system has been impli-

cated in variations in mood, and this system is the target of most

pharmacological interventions in the treatment of depression.

Among individuals within the Dunedin study, risk of depression

was predicted by the interaction of serotonin transporter geno-

type andthe numberofstressful life eventsexperienced.Thus, no

differences in risk of depression emerged as a function of geno-

Address correspondence to Frances A. Champagne, Columbia Uni- versity, Department of Psychology, 406 Schermerhorn Hall, 1190 Amsterdam Avenue, New York, NY 10027; e-mail: fac2105@colum- bia.edu.

CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE

Volume 18—Number 3 127Copyright r 2009 Association for Psychological Science

type when the number of stressful life events was low. However,

when an individual had experienced a high frequency of stressful

events, genotype effects were observed, with individuals pos-

sessing the low-serotonin-transporter-level gene variant being at

greater risk of depression. Though certain genetic variations can

lead to risk or resilience to psychological disorder (see Kim-

Cohen & Gold, 2009, this issue), this ‘‘potential’’ may not be

observed unless variation in the environment is considered.

CONTEXTUAL DETERMINANTS OF GENE FUNCTION

Empirical findings from G � E studies raise an important question: ‘‘If the effects of genetic variation can vary depending

on characteristics of the environment, then what are environ-

ments doing to genes to alter their impact?’’ To address this

question, we must first address the following question: ‘‘What do

genes do?’’ Historically, gene was a term used to describe a unit

of heritable material. Since the discovery of DNA, the study of

genetics has come to mean the study of DNA, with gene defined

as a particular sequence of DNA. Due to the complex nature of

DNA, it is perhaps easier to employ an analogy that conveys the

basic notions of gene function. Think of an individual’s DNA as

books in a library that have been ordered and arranged very

precisely by a meticulous librarian. These books contain a

wealth of knowledge and the potential to inspire whoever should

choose to read them. Asking what DNA does is like asking what a

book in this library does. Books sit on a shelf waiting to be read.

Once read, the information in those books can have limitless

consequences. Likewise, DNA sits in our cells and waits to

be read by an enzyme called RNA polymerase, leading to the

production of messenger RNA (mRNA)—a process referred to as

transcription (Fig. 1a). The mRNA transcript is a copy of the

DNA sequence that can further be ‘‘translated’’ into protein. The

reading, or expression, of DNA can, like the books in our library,

have limitless consequences. However, without the active pro-

cess that triggers such expression, this potential may never

be realized. Importantly, it is the environment around the DNA

that contains those critical factors that make it possible to read

the DNA (Fig. 1b; also see Cole, 2009, this issue, for extended

discussion of the regulation of gene expression).

The control of gene expression is ultimately determined by

how accessible the sequence of DNA is to factors within the cell

that are involved in transcription. Influences that determine the

expression of DNA without altering the sequence of DNA are

referred to as epigenetic, meaning ‘‘in addition to genetic.’’ One

particular epigenetic mechanism that may have consequences

for long-term changes in gene activity is DNA methylation (Fig.

1c). DNA can become modified through the addition of a methyl

chemical group to particular sites within the gene sequence.

DNA methylation typically reduces the accessibility of DNA and

can lead to ‘‘silencing’’ of the gene (Razin, 1998). In the library

analogy, one can think of multiple factors that will influence the

likelihood a book will or will not be read. Even books containing

very valuable information may sit undisturbed and unread,

gradually collecting dust. This may be particularly true if the

book is hard to get to. It may be located on a shelf that is par-

ticularly difficult to reach or blocked by some piece of furniture.

DNA methylation reduces the likelihood of transcription much

in the same way that shifting furniture in a library can reduce the

likelihood that a book will be read. The gene is there, but sits

unread, collecting dust.

ENVIRONMENTAL INFLUENCES ON GENE ACTIVITY

A recent breakthrough in our understanding of gene–environ-

ment interplay comes from studies exploring the epigenetic

processes that are altered by an individual’s experiences during

development. Based primarily on studies in rodents, these par-

adigms address the question raised by G � E research: ‘‘What are environments doing to genes to alter their impact?’’ In ro-

dents, variations in maternal care lead to individual differences

in the expression of genes that alter the stress response. Low

levels of glucocorticoid receptors (GR) within the hippocampus,

a brain region critical for learning and memory, result in a pro-

longed response to stress. Analysis of DNA methylation within

the regulatory region of the GR gene indicates that low levels of

ACTAGGCTAGATTCAGGATCTTAG “Active” Gene

ACTAGGCTAGATTCAGGATCTTAG

M

M

M

M M

M

Transcription a

b

c

ACTAGGCTAGATTCAGGATCTTAG MM M M M x “Silent”

Gene

Fig. 1. Illustration of the epigenetic control of gene expression and the environmental context of DNA. As shown in the top panel (A), genes consist of a sequence of DNA consisting of ‘‘C,’’ ‘‘T,’’ ‘‘A,’’ and ‘‘G’’ nucleotides preceded by a promotor region of DNA (the black bar). The promoter region responds to factors that control the likelihood of transcription (reading of the DNA). In order for transcription to occur, enzymes that ‘‘read’’ the DNA (the gray oval) must bind to the promotor region of the gene. When this occurs, the gene is ‘‘active’’ and can alter the function of the cell. The environmental context of the gene, shown in the middle panel (B), includes factors that increase gene activity (i.e., enzymes that read the DNA, shown as gray ovals) and factors that decrease gene activity (i.e., methyl groups, illustrated as circles labeled ‘‘M’’); these factors will determine the likelihood that a gene will be expressed. When a methyl chemical group attaches to the promotor region, as shown in the bottom panel (C), the enzymes that transcribe DNA are blocked and the gene becomes ‘‘silent’’; this is referred to as DNA methylation.

128 Volume 18—Number 3

Genes in Context

maternal care are associated with elevated levels of DNA

methylation, which epigenetically silence this gene (Weaver

et al., 2004). Moreover, the epigenetic status of the GR gene can

be targeted pharmacologically in adulthood. Treatment with a

drug that promotes increases in accessibility of DNA results in

decreased GR methylation and a dramatic shift in the phenotype

of adult offspring who received low levels of maternal care

(Weaver et al., 2004). Conversely, when adult offspring who

experienced high levels of care are treated with a drug that in-

creases the availability of methyl groups within the brain, they

become indistinguishable from offspringwho received low levels

of maternal care (Weaver et al., 2005). These dynamic altera-

tions in DNA methylation in adulthood have also been observed

in studies of learning and memory (Miller & Sweatt, 2007). The

experience of learning is associated with rapid changes in

methylation of genes within the hippocampus, and if DNA

methylation is inhibited there will be impairment in memory for

the experience. These studies illustrate the role of epigenetic

mechanisms in shaping the activity of the genome in response

to environmental cues and demonstrate the plasticity that is

possible through shifts in DNA methylation.

The prenatal period is characterized by rapid changes in brain

development and is thus a sensitive time during which the

quality of the environment can exert sustained effects on func-

tioning. In rodents, exposure to chronic variable stress during

the first trimester is associated with increased methylation of the

regulatory region of the GR gene (Mueller & Bale, 2008). This

effect could potentially be mediated by (a) stress-induced de-

creases in postnatal maternal behavior (Champagne & Meaney,

2006), (b) alterations to gene expression in the placenta (Mueller

& Bale, 2008) that may restrict access of the fetus to maternal

resources, or (c) a direct influence of maternal stress hormone on

fetal gene expression. Modification to the fetal ‘‘epigenome’’ can

also be achieved through variations in maternal diet during

pregnancy. A striking example of this phenomenon comes from

work with a mouse model in which a mutation of the Agouti gene

leads to alterations in coat color and metabolism. The severity of

the effects of this mutation depends on the level of DNA meth-

ylation of the Agouti gene; high levels of DNA methylation will

epigenetically silence this mutation and induce a ‘‘pseudo-

agouti’’ mouse that is comparable in phenotype to a mouse

without the mutation. When pregnant female mice with the

Agouti mutation are placed on a diet that is rich in methyl groups,

the methylation status of this gene is altered such that offspring

develop a pseudoagouti phenotype (Dolinoy, 2008). Thus, ex-

perience-dependent change in the epigenetic status of genes is

not limited to the postnatal period.

IMPLICATIONS OF GENE–ENVIRONMENT INTERPLAY

FOR PSYCHOLOGICAL FUNCTIONING

The molecular processes described in laboratory studies may

also be critical in understanding the origins of individual

differences in humans. Analyses of DNA methylation in cells

extracted from fetal cord blood suggest that antenatal maternal

depression and anxiety during the third trimester can lead to

increased levels of DNA methylation of the GR gene promotor

region, having consequences for the stress response of infants at

3 years of age (Oberlander et al., 2008). These effects emerge

even in the absence of depression-induced decreases in post-

natal mother–infant interactions. The stability of DNA methyl-

ation also permits analysis of the epigenetic status of genes in

postmortem brain tissue, which can be correlated to life expe-

riences and psychological functioning. In a recent study, DNA

methylation of ribosomal genes in hippocampal tissue of suicide

victims with a history of abuse and neglect was compared to that

of controls. Elevated levels of methylation were detected in

ribosomal RNA genes among suicide victims (McGowan et al.,

2008), and this effect was found to be specific to the hippo-

campus. Ribosomes are critical for the production of proteins

and thus serve as a critical link between the expression of genes

and the level of protein created.

Studies of monozygotic (MZ) twins also provide important

insights into epigenetic effects in humans. Comparison of the

gene expression of 3-year-old and 50-year-old MZ twins indi-

cates a higher level of discordance in patterns of gene expression

among older twins that is associated with increasing differences

in DNA methylation in older compared to younger twins (Fraga

et al., 2005). Though it is unknown whether concordance in

young twins is due to germ-line (the cells that transmit genetic

material across generations) or prenatal factors and whether the

emerging discordance is random or driven by specific environ-

mental events, there is evidence that epigenetic variation in MZ

twins may account for differential risk of mental illness. Analysis

of methylation patterns within the catechol-O-methyltransferase

(COMT) gene in tissue samples from 5-year-old MZ twins indi-

cates varying degrees of discordance, with some MZ twin pairs

showing a high degree of discordance and others being very

similar in epigenetic status (Mill et al., 2006). COMT is an en-

zyme involved in the inactivation of neurotransmitters such as

dopamine and norepinephrine, and disruptions in these neuro-

transmitter systems have been implicated in many forms of

psychopathology. The divergence in methylation of the COMT

gene within these twin pairs may predict differential risk of

neurodevelopmental disorder in later life. Incorporating epige-

netic analysis into twin studies represents a novel approach to

the study of the origins of individual differences.

TRANSMISSION OF TRAITS ACROSS GENERATIONS:

RETHINKING INHERITANCE

In addition to shaping developmental trajectories within an

individual’s life span, DNA methylation may also have impli-

cations for the transmission of traits from one generation to the

next. There are two distinct pathways through which this

transmission can occur: (a) the behavioral transmission of traits

Volume 18—Number 3 129

Frances A. Champagne and Rahia Mashoodh

through experience-dependent changes in the methylation of

genes, and (b) environmental effects that change DNA methyl-

ation in germ cells and are thus transmitted through the germ

line of subsequent generations. An example of the first pathway

comes from studies of the transmission of maternal care across

generations. Variations in maternal care in rodents have been

demonstrated to alter the epigenetic status of hypothalamic es-

trogen receptors of female offspring (Champagne et al., 2006).

These receptors are critical in regulating maternal behavior and

coordinate the sensitivity of females to hormonal cues. Experi-

ence of low levels of maternal care in infancy is associated with

increased estrogen receptor promotor methylation, decreased

receptor expression, and subsequent decreases in the adult

maternal behavior of these offspring. Thus, there is a behavioral

transmission of individual differences in maternal care across

generations. Interestingly, the quality of environmental condi-

tions experienced by these females at later periods in develop-

ment can alter this transgenerational inheritance. Prolonged

social isolation from peers and prenatal stress can lead to re-

ductions in maternal care that are passed on to subsequent

generations (Champagne & Meaney, 2006, 2007). These studies,

which are conducted in rodents that have limited genetic vari-

ability, suggest that similarities in traits between parental and

offspring generations involve far more than the inheritance

of genes.

Though epigenetic characteristics of DNA are dynamic in

response to environmental cues, these modifications are also

stable and heritable. Thus, both genetic and epigenetic factors

are transmitted down cell lineages with consequences for the

activity of genes within these lineages. However, when consid-

ering the question of inheritance at the level of an individual, we

must know whether epigenetic patterns within the germ line are

correlated to those patterns found within the developing or-

ganism. In rodents, prenatal exposure to endocrine disruptors

lead to abnormal methylation patterns in sperm cells that are

observed several generations beyond the point of initial expo-

sure (Anway, Cupp, Uzumcu, & Skinner, 2005). This germ-line

epigenetic inheritance of environmentally induced effects pro-

vides further support for the notion that the transmission of traits

across generations is not limited in scope to the inheritance of

DNA.

CONCLUSION

Just as a library is more than a collection of books, the genome is

more than just DNA. The challenge for the field of epigenetics is

to determine the origins of the ‘‘uniqueness’’ of each individual’s

library by exploring the relationship between genetic and epi-

genetic variation. Though there are many basic questions to be

addressed regarding the pathways whereby specific experiences

target particular genes, this field of research certainly has

promise in uncovering the nature of experience-dependent

changes in development both within and across generations.

Advances in tools available to study these effects in humans will

be critically important in further exploring the role of epige-

netics within the broad field of psychological science.

Recommended Reading Champagne, F.A. (2008). Epigenetic mechanisms and the transgener-

ational effects of maternal care. Frontiers of Neuroendocrinology, 29, 386–397. Provides a thorough review of the potential role of epigenetic factors in mediating the effects of maternal care within

and across generations.

Jirtle, R.L., & Skinner, M.K. (2007). Environmental epigenomics and

disease susceptibility. Nature Reviews Genetics, 8, 253–262. A review of our current understanding of environmentally induced

epigenetic changes and the influence of these processes on indi-

vidual risk of disease.

Maher, B. (2008). Personal genomes: The case of the missing herita-

bility. Nature, 456, 18–21. An interesting commentary on the relationship between heritability estimates and the biological

processes that determine the relationship between genes and

behavior.

Meaney, M.J. (2001). Maternal care, gene expression, and the trans-

mission of individual differences in stress reactivity across gen-

erations. Annual Review of Neuroscience, 24, 1161–1192. A review of the profound influence of maternal care on gene expression and

behavior of offspring.

REFERENCES

Anway, M.D., Cupp, A.S., Uzumcu, M., & Skinner, M.K. (2005). Epi-

genetic transgenerational actions of endocrine disruptors and male

fertility. Science, 308, 1466–1469.

Caspi, A., Sugden, K., Moffitt, T.E., Taylor, A., Craig, I.W.,

Harrington, H., et al. (2003). Influence of life stress on depression:

Moderation by a polymorphism in the 5-HTT gene. Science, 301, 386–389.

Champagne, F.A., & Meaney, M.J. (2006). Stress during gestation alters

postpartum maternal care and the development of the offspring in a

rodent model. Biological Psychiatry, 59, 1227–1235.

Champagne, F.A., & Meaney, M.J. (2007). Transgenerational effects of

social environment on variations in maternal care and behavioral

response to novelty. Behavioral Neuroscience, 121, 1353–1363.

Champagne, F.A., Weaver, I.C., Diorio, J., Dymov, S., Szyf, M., &

Meaney, M.J. (2006). Maternal care associated with methylation of

the estrogen receptor-alpha1b promoter and estrogen receptor-

alpha expression in the medial preoptic area of female offspring.

Endocrinology, 147, 2909–2915.

Cole, S.W. (2009). Social regulation of human gene expression. Current Directions in Psychological Science, 18, 132–137.

Cooper, R.M., & Zubek, J.P. (1958). Effects of enriched and restricted

early environments on the learning ability of bright and dull rats.

Canadian Journal of Psychology, 12, 159–164.

Dolinoy, D.C. (2008). The agouti mouse model: An epigenetic biosensor

for nutritional and environmental alterations on the fetal epigen-

ome. Nutrition Reviews, 66(Suppl 1), S7–S11.

Fraga, M.F., Ballestar, E., Paz, M.F., Ropero, S., Setien, F., Ballestar,

M.L., et al. (2005). Epigenetic differences arise during the lifetime

130 Volume 18—Number 3

Genes in Context

of monozygotic twins. Proceedings of the National Academy of Sciences, USA, 102, 10604–10609.

Kim-Cohen, J., & Gold, A.L. (2009). Measured gene–environment

interactions and mechanisms promoting resilient development.

Current Directions in Psychological Science, 18, 138–142.

McGowan, P.O., Sasaki, A., Huang, T.C., Unterberger, A., Suderman,

M., Ernst, C., et al. (2008). Promoter-wide hypermethylation of the

ribosomal RNA gene promoter in the suicide brain. PLoS ONE, 3, e2085.

Mill, J., Dempster, E., Caspi, A., Williams, B., Moffitt, T., & Craig, I.

(2006). Evidence for monozygotic twin (MZ) discordance in

methylation level at two CpG sites in the promoter region of

the catechol-O-methyltransferase (COMT) gene. American Jour- nal of Medical Genetics B: Neuropsychiatric Genetics, 141, B421– B425.

Miller, C.A., & Sweatt, J.D. (2007). Covalent modification of DNA

regulates memory formation. Neuron, 53, 857–869.

Mueller, B.R., & Bale, T.L. (2008). Sex-specific programming of off-

spring emotionality after stress early in pregnancy. Journal of Neuroscience, 28, 9055–9065.

Oberlander, T.F., Weinberg, J., Papsdorf, M., Grunau, R., Misri, S., &

Devlin, A.M. (2008). Prenatal exposure to maternal depression,

neonatal methylation of human glucocorticoid receptor gene (NR3C1)

and infant cortisol stress responses. Epigenetics, 3, 97–106.

Razin, A. (1998). CpG methylation, chromatin structure and gene si-

lencing-a three-way connection. EMBO Journal, 17, 4905–4908.

Weaver, I.C., Cervoni, N., Champagne, F.A., D’Alessio, A.C., Sharma,

S., Seckl, J.R., et al. (2004). Epigenetic programming by maternal

behavior. Nature Neuroscience, 7, 847–854.

Weaver, I.C., Champagne, F.A., Brown, S.E., Dymov, S., Sharma, S.,

Meaney, M.J., et al. (2005). Reversal of maternal programming of

stress responses in adult offspring through methyl supplementa-

tion: Altering epigenetic marking later in life. Journal of Neu- roscience, 25, 11045–11054.

Volume 18—Number 3 131

Frances A. Champagne and Rahia Mashoodh