Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
Media
Part of Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
- Title
- Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
- extracted text
-
Social Epigenetics: Incorporating
Epigenetic Effects as Social Cause
and Consequence
DOUGLAS L. ANDERTON and KATHLEEN F. ARCARO
Abstract
Epigenetics is a field of study that invites an interdisciplinary interaction of the social
and biological sciences. This collaboration has, in fact, led to a blossoming research
community over the past two decades, which is using new data, methods, and conceptual frameworks to address a host of old and emergent research questions. A
recent (2014) search of PubMed found over a thousand articles on social, behavioral, and cognitive epigenetics. If one includes epidemiological epigenetic studies
that incorporate either social causes or consequences in their research, the number
expands nearly threefold. Yet, social epigenetics is a still nascent field, marginalized
and misunderstood in social science. In this essay, we attempt to review basic epigenetic concepts and the way in which epigenetics has, and can be, of use to social and
behavioral scientists in addressing some of the most fundamental sorts of questions
their disciplines raise.
INTRODUCTION
Reference to epigenetics, and the use of epigenetic concepts, has grown over
the recent past in many social and behavioral science disciplines (Goodman,
Heath, & Lindee, 2003; Landecker & Panofsky, 2013) and has an already
extensive history within cognitive psychology (Lester et al., 2011; Miller,
2010). However, a long history of the abuse of genetic concepts in the social
sciences has also been followed by a subsequent disregard by many for the
potential role of genetics in phenomena of behavioral and social interest.
Much of the abuse of genetic constructs in the social sciences came from
simplistic presumptions that complex socially expressed or constructed
traits, such as race, intelligence, class, and so on, could be traced to specific
fixed underlying genetic expression and destiny, an approach fueled by
racism, classism, and sexism. While genetic causation retained a following among more biologically inclined social sciences (e.g., demography,
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.
1
2
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
neuropsychology, and epidemiology) the relationship of genetics to behavior and broader social constructs has been largely shunned until recent
times. Newer generations have been more willing to investigate the social
consequences and causes of genetic influences, especially as embodied in the
notion of a mutable, even environmentally socially responsive, epigenetic
influence on phenotypes, rather than an emphasis on, in the short-run
fixed or immutable, DNA genotypes imposed as dominant upon the social
landscape. The potential for “social epigenetics,” embracing the integration
of epigenetic measures and mechanisms into social and behavioral science
research, as cause, consequence, or mechanism, is an exciting and potentially
very fruitful line of inquiry.
FOUNDATIONAL RESEARCH: WHAT IS EPIGENETICS?
The origins and uses of the term “epigenetic” are varied (Haig, 2004).
Usage vaguely familiar to that of modern molecular biologists dates at
least to Waddington’s (1942) reference to underlying mechanisms by which
“genes of the genotype bring about phenotypic effects.” For at least the past
two decades it has, more or less, had a familiar meaning defined (in the
negative) as “the study of mitotically and/or meiotically heritable changes
in gene function that cannot be explained by changes in DNA sequence”
(Riggs, Martienssen, & Russo, 1996). The notion that epigenetic changes to
gene function are heritable dates back to the mid-1960s. However, other
definitions remain in use, and the centrality of heritability to a definition of
epigenetics is not universally accepted. Bird, for example, defines epigenetics more broadly and as “the structural adaptation of chromosomal regions
so as to register, signal or perpetuate altered activity” (2007). And, for much
of the work relevant to social scientists, this broad definition is appropriate
and appealing. This definition emphasizes, as did Waddington’s, epigenetic processes that alter gene expression, including DNA methylation as
well as other DNA and chromatin modifications that have not yet been
demonstrated to be heritable.
At the risk of over-abstraction, it is perhaps helpful to pause and present
things in a simplified form. Consider a gene, or DNA, is largely fixed where
each gene is associated with an expression of genetic information or “trait.”
Here, trait simply means a specific molecular genetic expression, or the way
in which information from a gene is used on a molecular level (not trait in the
abusive sense of race, intelligence, etc.). The gene’s expression can be altered
by various molecular processes which are by (negative) definition epigenetic,
changes to the gene from outside. These epigenetic influences on gene expression are in contrast to the strictly genetic influences that include mutations or
polymorphisms in the DNA sequence. Some of these processes are, in turn,
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
3
influenced even by the wider environment and social phenomena such as
diet, stress, or environmental exposures. Thus, the social world can influence the expression of genes in a molecular manner. In turn, many believe
this epigenetically altered genetic expression can have a broader influence
on phenomena of social interest, ranging, for example, from specific genetically related diseases, endocrine function, and bodily reactions to stress, to
even antisocial behavior or suicidal tendencies. There is evidence that some
epigenetic changes can occur in utero, such that the environment of one generation can influence the epigenetic state of the next generation. And, there
is evidence of truly heritable epigenetic transmission from one generation to
the next through sexual reproduction, implying potentially intergenerational
effects from, as a hypothetical example, one generation’s childhood dietary
behavior to the next generation’s childhood health. Lastly, it is worth noting that some discussions often refer to the cumulative epigenetic changes in
gene expression as the “epigenome” or a “ghost gene” although this collection of epigenetic changes does not exist, in any cohesive molecular sense,
independent of the gene.
At the molecular level, epigenetic processes involve the placement of chemical tags directly on the DNA or on the proteins that help organize the DNA
(Rodríguez-Paredes & Esteller, 2011). These chemical tags are critical in controlling the expression of the gene and in maintaining genetic stability and
can be passed on to daughter cells at cell division. By far the most studied tag
is the methyl group, a carbon atom with three hydrogen atoms (CH3 ) that can
be attached at specific locations on our DNA. DNA methylation is only one of
several epigenetic mechanisms. However, because methylation studies are so
prevalent, including among the examples we cite, methylation merits a brief
explanation. In humans, methyl groups can be attached to only one of the
four DNA bases, cytosine (C), and only when the cytosine is followed by guanine (G) in the DNA sequence. Because the cytosine and guanine on the same
strand are joined by a phosphate (p) molecule, the potential DNA methylation sites are referred to as CpGs. CpGs are not evenly distributed across our
genome. Regions of our DNA with a high density of CpGs are referred to as
CpG islands. Borrowing from oceanography, CpGs occurring a short distance
from an island are on the shore, and moving further out, the shelf followed
by the deep sea. The density and location of CpGs in our DNA sequence
is important. CpG islands occur more frequently within genes, in regions
responsible for turning the gene on and off (promoter regions), whereas as
low-density CpGs occur more frequently in DNA that is between genes and
is responsible for maintaining genetic stability. Biologists are busy cataloging
the location and density of methylated CpGs in various tissues from an array
of developmental, exposure, and diseased states; however, much remains to
be learned about the biological effects of methylation at specific CpGs.
4
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
CUTTING-EDGE RESEARCH: USING EPIGENETIC DATA IN THE
SOCIAL SCIENCES
Just as one does not have to be an oncologist to study the social correlates of
cancer, for example, one does not have to be a molecular biologist to consider
the role of epigenetics in social and behavioral models. In most instances,
research addressing epigenetic influences is inherently collaborative and
increasingly in the model of “big science” involving a team of, perhaps,
molecular biologists, epidemiologists, behavioral experts, disease specialists, research physicians, cognitive psychologists, laboratory technicians,
and so on, all of whom will be involved in formulating the central research
problem on which social and behavioral scientists collaborate, identifying
gene regions or targets of epigenetic interest, defining likely epigenetic
covariates, supervising appropriate biological assays, and so on. This is true
regardless of whether the driving research question is one generated from
within the social sciences (e.g., environmental justice research on differential
exposure of minority populations to potential epigenetic damage) or one
generated by any other discipline that simply relies upon expertise from
the social and behavioral sciences for the context in which those disciplines
would view the research problem at hand.
To facilitate a somewhat pragmatic discussion of epigenetic research and
social science in this big science context, let us start with a word about operationalization and measurement. Again, although methylation is only one
form of epigenetic change, it has had attention from molecular biologists
and social scientists in particular because of potential epigenetic effects on
prominent diseases (e.g., various cancers), behaviorally related body processes (e.g., cortisol response) and evidence of potentially direct cognitive
pathways (e.g., suicidal tendencies). Quantitative measures, usually percentage, of methylation in specific gene regions or sites of interest are the most
common measurements analyzed. Processing cell samples (e.g., saliva, tissue
samples, epithelial cells, etc.) to obtain laboratory measurements of methylation for various regions or sites of the gene is still carried out in many individual laboratories but is increasingly a big business and a matter of contracting
out for measurements using biological assays collected by researchers. For
this reason, we will not spend time here on various laboratory methods and
technical procedures. Instead, the focus here is on the potential to be achieved
from integrating such measures into social and behavioral scientific research
and the overarching issues of research design and data analysis surrounding
epigenetic work of importance to the social and behavioral sciences.
One distinction, of relevance, is between measurements from deductively
targeted genetic sites of interest (e.g., pyrosequencing for measurement of
methylation in specific promoter regions) and the increasing use of inductive
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
5
or exploratory measurements across broad swaths of the genome (e.g., global
array analysis and genome-wide sequencing of bisulfite-modified DNA,
although these methods can also be used to obtain targeted site information). The challenges these different approaches present to statistical analysis
can be very difficult, ranging from small sample size issues for precious
biological tissue samples, to the large amount of data generated by global
arrays and genome sequencing and complex statistical methods required
for analysis. Commercial laboratories providing methylation measurement
services, especially array measurements, also tend to provide basic software
for analyzing the data and complementary routines are available in public
software, largely contributed routines in the R-software suite. Again, the
complexity of statistical analysis for much epigenetic data often calls for a
big science approach with a knowledgeable statistical analyst.
Thus, without trivializing the complexities of methylation analysis, it is not
unreasonable to say the social scientists will usually encounter epigenetics as
a database, not unlike many they are familiar with, of quantitative measurements of methylation for either target regions of the gene or for large sections
of the genome. Analytical approaches may be deductive and targeted at specific genetic sites using familiar analytical techniques or inductive and requiring high-throughput analysis, often using proprietary laboratory-supplied
software. In the remainder of this essay, we do not dwell on laboratory or
statistical methodology details. Instead, as suggested, we take an approach
in which epigenetic measures are operationalized and available as variables
for a specific research problem and focus on the conceptual or functional use
of epigenetics as a construct and framework within the social sciences. We do
this by focusing, in turn, on recent examples of work, and some speculation
as to the future of this work, in which epigenetic measures are used by social
scientists as independent causal effects, dependent variable outcome measures, intervening variables representing epigenetic mechanisms, durational
effects carrying causation from one point in time to a later manifestation of
outcomes and intergenerational variables potentially carrying the effects of
gene–environment interactions across generations.
KEY ISSUES FOR FUTURE RESEARCH: THE UTILITY AND ROLES
OF EPIGENETIC DATA IN THE SOCIAL SCIENCES
EPIGENETIC VARIATION AS A DEPENDENT OUTCOME
The most common existing use of epigenetic variation to date is that of a
biological assay, used as a dependent variable, or biomarker, of health outcomes in studies to identify exposures responsible for the epigenetic change.
6
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
Epidemiologists, demographers, health and medical researchers, anthropologists, neuropsychologists, and many others, have all shown interest in the
potential for incorporating epigenetic assays and measurements as potentially more direct indicators of environmental effects than abstract measures
of allostatic health, body condition, or future disease outcomes. Epigenetic
variations provide a measure of induced change in genetic fortunes. These
may be studied simply as indicators of environmental exposure or be targeted because of known associations with eventual disease outcomes. When
studying rare diseases with complex lifelong pathways, such as most cancers,
the prospect of studying outcome measures that may be more immediate,
but indicative of both environmental causation and thereby induced predisposition toward eventual disease outcomes, is appealing. Although there are
many studies that have employed epigenetic variation as a direct outcome
measure in this manner, two examples may be of particular interest in an
interdisciplinary context: dietary studies and epigenetic toxicology.
An early interest in the effect of social behaviors on epigenetic change
emerged in studies of the relationships between diet and methylation (Van
den Veyver, 2002) and of diet to other epigenetic mechanisms (Romagnolo,
Dashwood, & Ziegler, 2012). These interests have fueled an emerging science of nutrigenomics. Not surprisingly, most evidence that dietary intake
affects epigenetics exists for folate, which supplies building blocks of DNA
methylation. However, a long list of other dietary components have been
studied, including alcohol, flavonoids, polyphenols, lycopene, methionine,
B-vitamins, resveratrol, choline, genistein, and sulforaphane, among others,
and such common foods containing some of these compounds as green
tea, spinach, sunflower seeds, baker’s yeast, soy, broccoli sprouts, fiber, or
cruciferous vegetables (Hardy & Tollefsbol, 2011). One area of particular
attention in dietary studies is in prenatal diets where animal studies suggest
folate or choline deficits can cause lifelong hypomethylation. Another great
interest in dietary studies concerns the potential treatment aspects of diets
which contain dietary compounds that modulate epigenetic programming
and can be used to target methylated promoters known to contribute
to cancers. Studies of various cancers sometimes consider both overall
genome-wide reduction in DNA methylation (global hypomethylation)
resulting in chromosomal instability and hypermethylation within the
CpG islands of a specific gene promoter, often tumor suppressing genes,
as an outcome variable. The prospects of identifying hypermethylated,
cancer-related, promoters and targeting them with dietary demethylating
agents before disease develops drives a great deal of interest in studies of
diet and methylation.
Another interest in the use of epigenetic change as outcome variables
comes from communities who are in search of sensitive, widely detectable,
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
7
indications of environmental or toxicological exposure. The use of epigenetic
variations as direct outcomes in examining toxicological or environmental
effects has emerged as a new field of considerable potential (Sahu, 2012;
Szyf, 2011). Researchers who examine diffuse environmental hazards and
issues of environmental justice, for example, have long struggled with the
fact that outcome measures such as disease incidence rates measure only the
potentially rare events that exist at the end of a long and complex pathway
that may nonetheless be impacted by environmental exposures and bodily
predisposition to eventual outcomes. Use of measures such as epigenetic
variations that are more immediate mechanistic effects of environmental
exposures is of great value in such circumstances. Focusing on epigenetic
mechanisms may explicate an actual pathway of ultimate harm, while
addressing the research problems posed by disease outcomes that may
remain both rare in exposed populations and may not manifest for years
hence. This interest has again given rise to a host of studies of specific environmental toxicants such as endocrine disruptors, heavy metals, and other
persistent and common environmental pollutants in our air, food, and water
supplies. Most environmental epigenetic studies have utilized methylation
measures. However, the still emerging and new field of environmental
epigenomics has grown to encompass studies across the entire range of
epigenetic mechanisms (Godfrey, 2012; Hou, Zhang, & Baccarelli, 2011).
EPIGENETIC VARIATION AS A CAUSAL INFLUENCE
While most studies have emphasized and utilized epigenetic change as a
dependent variable or outcome of environmental circumstances, other profitable areas of epigenetic research have focused on using epigenetic changes
or biomarkers as an independent variable responsible for direct, or eventual,
disease or behavioral outcomes. Some studies have gone even further and
explored both causes and consequences of epigenetic variation in a single
study, bringing epigenetic variation into the more realistic role of an intervening variable. Again, two examples may suffice to illustrate the uses and
potential of this epigenetic approach: neurocognitive/behavioral epigenetics
and epigenetic cancer studies.
Within the social and behavioral sciences, neurocognitive and behavioral
epigenetics is likely the most familiar application of epigenetics to many
and has been the subject of most futuristic speculation. Most, not all, of the
research in neurocognitive epigenetics has focused on identifying epigenetic
variations associated with cognitive or behavioral outcomes. Epigenetic
variation has been considered, for example, as a potential explanation in
processes of memory, learning, age-related cognitive changes, depression,
aggression, mental illness, and other cognitive outcomes of interest. Studies
8
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
have also examined epigenetic effects on behaviorally related endocrine
response systems and for specific mental illness outcomes. Epigenetic
variations have been implicated in the neurodegenerative disorders such as
Alzheimer’s and Parkinson’s disease, the most common form of inherited
mental retardation (fragile X syndrome), Rett syndrome, and more recently,
as a potential influence on autism spectrum disorders.
Given the difficulty of brain tissue sampling, one of the most notable studies of this sort is a cadaver study of suicide victims (McGowan et al., 2009).
This study replicated earlier animal model work supporting a plausible association between methylation of glucocorticoid receptors and later life abilities
to cope with stress. By comparative analysis of those who suffered childhood
abuse and those who did not, the authors also suggest abusive stress may be
a source of DNA methylation, bringing methylation into the role of an intervening variable affecting later suicide among child abuse victims. This study
is only one of many suggesting a link between stress and later life cognition
and health. Miller (2010) summarizes this research and several threads of
more recent, specifically cognitive, epigenetic research including changes in
gene expression through histone modification that are related to subsequent
learning and memory, as well as age-related changes in memory, across the
lifespan.
Undoubtedly, the greatest strides in studying epigenetic change as a direct
cause of socially important outcomes have been in the epidemiological study
of methylation and disease outcomes, especially cancer. After nearly two
decades of numerous studies seeking to identify specific misregulation in
epigenetic mechanisms (including DNA methylation, histone modification,
nucleosome remodeling, and RNA-mediated targeting) that can culminate in
cancer, there is little doubt that such connections are abundant (see Dawson &
Kouzarides, 2012, for one recent summary). However, despite these numerous studies implicating specific epigenetic misregulation with both tumor
suppression and cancer outcomes, there are not, as yet, definitive epigenetic
biomarkers identified as suitable for routine clinical practice. One forefront of
epigenetic cancer research in the coming decades will be to advance research
to the point of contributing to clinical diagnosis and treatment of specific cancers. Of course, identification of specific pathways is also potentially critical
to preventive measures. While much has been achieved, even in this most
prolific area of research, epigenetics is a nascent science.
With a growing number of epidemiological studies, the potential role of
methylation, as not only a cause but also an intervening effect between
social behaviors and disease or behavioral outcomes is naturally receiving
considerable attention. Research in developmental psychology by Frances
Champagne and others (see Champagne, 2009 for an overview), for example,
provides examples from animal model research suggesting early exposures
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
9
to stress could epigenetically modify later life stress response mechanisms;
or that juvenile isolation could influence alter epigenetic pathways that
contribute in turn to isolation syndrome among adults; and other similar
examples of species-specific social behavior with epigenetic intervening
effects that subsequently impact social behavior. Discussion of mechanisms
of socially mediated epigenetic misregulation in human populations (e.g.,
excessive alcohol consumption that reduces folates related to epigenetic
changes) coupled with studies linking such epigenetic changes to disease
outcomes (e.g., liver cancer), suggest a fuller understanding of the relation
of social behaviors to disease, or behavior, through specific epigenetic
pathways is within reach. For many diseases such as breast cancer, where
the major correlates include not only heredity (e.g., BRCA1) but significant
influences that are largely social in character (e.g., alcohol, diet, obesity,
childbearing, and breastfeeding, etc.), epigenetics may provide a critical
missing link in the pathways between social behavior and socially significant
disease and behavioral outcomes.
Again, even in the most advanced areas of epigenetic research, our understanding of the complex pathways linking environments to epigenetics to
outcomes is only now emerging. The potential of such research excites the
imagination. Yet, such speculation is also warranted to the extent that it
stimulates new studies and explorations of problems that have long defied
the simple bifurcated heredity plus behavior equation. Particularly from a
social science or public health perspective, the environmentally interactive
and intervening role of epigenetic change may be critically important
for understanding both cause and consequence of socially important
“epidemic” conditions (e.g., obesity or autism) and ill-defined syndromes
(e.g., chronic fatigue, environmental sensitivity, fibromyalgia) that have
been largely socially and symptomatically defined, with, as yet, poorly
understood biological pathways. With such high-value targets in sight, and
a new paradigm with some seeming promise, the occasional speculative, or
even wildly speculative, hypothesis is perhaps deserving of some latitude.
EPIGENETICS AND DURATIONAL EFFECTS
For social epigenetics, one of the most interesting and useful elements of
the conceptual framework afforded is the durational aspects of epigenetic
effects. As in much of the research cited earlier regarding social behaviors
(e.g., Champagne, 2009), epigenetic influences such as an environmental
exposure may occur early in the lifespan, or in utero, and those relatively
durable epigenetic changes carry consequences forward in time such
that they then manifest in an impact on disease or biological and cognitive outcomes in later life. This durational model, carrying the effects
10
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
of environmental exposures far removed in time to ultimately observed
probabilistic outcomes, provides a mechanism suitable to numerous social
and epidemiological inquiries. Animal models demonstrating that maternal
crack cocaine exposure alters epigenetic profiles, resulting in dysfunctional
social interactions among offspring throughout their life course, have raised
awareness of the plausibly groundbreaking potential for such studies in
human populations. There is, as yet, more speculation than research in this
regard. Kuzawa and Sweet (2009), for example, argue the correlated racial
disparities in birth weight differences and cardiovascular diseases across
groups offer a rationale for considering epigenetic links between early-life
environmental factors, such as maternal stress during pregnancy and adult
racial disparities in cardiovascular diseases. Similarly, Thornburg, Shannon,
Thuillier, and Turker (2010) argue that such differences, along with research
linking rapid growth in the womb to metabolic disease and obesity and also
to breast and lung cancers, calls for research to determine the epigenetic
processes underlying these linkages.
Demographers and epidemiologists have also long been interested in
durational processes in which in utero, or early life-course, events and
stress have resulted in long-term adult morbidity and mortality in historical
populations. Such durational effects are often simply regarded as a “black
box,” increasing frailty among the exposed populations and the likelihood of
later morbidity and mortality. Epigenetics provides a conceptual framework
that may well lead us to more specific pathways by which distinctive early
life-course exposures and events influence the likelihood of particular
morbidity outcomes over the life course. The potential for such research is
significant. However, the implementation of longitudinal life-long epigenetic
studies of early life-course effects on epigenetic change and probabilistic
outcome variables, such as morbidity, is anything but simple compared to a
short-term clinical trial, for example, to examine the effects of stress or diet
on epigenetic misregulation. An increasing recognition for the big science
model needed to address such questions, including the linkage of large
longitudinal population studies to genetic and epigenetic data, provides
some hope for significant advances in such research despite the many
challenges faced in such work.
TRANSGENERATIONAL EPIGENETICS
Without question, the most controversial and one of the most exciting
prospects regarding epigenetics has been some initial evidence for transgenerational transmission of epigenetic effects (Kaiser, 2014). This research goes
beyond the durational model, which supports possible mutigenerational
(but not transgenerational) effects. In a multigenerational model early-life
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
11
or in utero conditions may create epigenetic changes that are manifest in
later life course outcomes of offspring but exposure of those offspring
occurred after sexual reproduction and is not transgenerational or inherited.
However, some epigeneticists argue, and some evidence from animal model
experiments suggests, that induced epigenetic changes can be transmitted
meiotically, or through sexual reproduction, to the next generation or even
several generations hence. This research is still in very early stages, with
a variety of protocols, some supporting studies, and some failing to find
support. Difficulties in conducting such research are vastly greater, yet
again, than those faced by life-long studies of epigenetic effects. Research
has been exclusively confined to animal models. In addition, a biomolecular
mechanism for this intergenerational transmission epigenetic misregulation
has also not yet been elaborated. So, as with all truly cutting-edge research,
the jury is still deliberating not simply the promise but the reality of
transgenerational epigenetics.
Yet, while scholars are still divided and there are skeptics, there are also
sufficient converts to transgenerational epigenetics, and enough cumulating
supporting evidence, to ensure that this line of research will continue. Transgenerational epigenetic change will ultimately either be refuted or, perhaps
more likely, grow into in a new, and paradigm-shifting, field of future study.
If transgenerational epigenetic hypotheses are confirmed, the processes to be
studied will not be a simple replicate of studies into heritability among future
generations. Even among those who strongly support the notion of transgenerational epigenetic change, this claim is often accompanied by the appropriate qualification that continuing methylation and demethylation influences
attenuate the force of transmission across generations in ways that will need
to be better understood.
The prospect of transgenerational epigenetics for social science is revolutionary to say the least and one which fuels enthusiastic speculation.
The idea, for example, that social circumstance of one generation may
contribute to epigenetic changes which are then transmitted to, and impact
the life-course prospects, of successive generations, could help to explain
the intergenerational effects of circumstance on social life that have been
found persistent, even in the face of intergenerationally changing social
environments and circumstance.
LIMITS AND FURTHER CONSIDERATIONS
There are many limits to the growing enthusiasm for epigenetic research and
its expansion into other disciplines including the social sciences. Many current studies, for example, rely ultimately on a selection by the dependent
variable and introduce the possibility of reverse causation. If we select tumor
12
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
samples, for example, to bioassay, how do we ensure that epigenetic changes
found in patients with tumors, and not controls, are not symptomatic rather
than causal. One emerging answer to this problem is the statistical use of
a Mendelian randomization approach (Relton & Smith, 2010) equivalent to
an instrumental variables model with genotype acting as an instrument for
the exposure of interest. Another problem in conducting such research is
simply the expense and difficulties of obtaining biological assays and the
big science funding model required of a truly interdisciplinary epigenetic
study. However, there has been an explosion of large data resources combining social and genome-wide array data (e.g., Add Health, the Health and
Retirement Study, Wisconsin Longitudinal Study, Framingham Heart Study)
and similar resources for genome-wide methylation studies are emerging.
The linkage of tissue banks and large-scale longitudinal population surveys
provide another potential avenue for social epigenetic research. A growing
number of epigenetic databases with linked methylation and survey data are
available for public use and efforts by funding agencies to support standardized assays, held in common for future research, have resulted in a growing
body of secondary data resources.
Another substantial hurdle lies in the lack of standardized protocols for
complex hypotheses and the most challenging studies such as transgenerational epigenetic research. There is no magic shortcut to a mature epigenetic
science and only continued study and careful replication will resolve these
issues over the near future.
REFERENCES
Bird, A. (2007). Perceptions of epigenetics. Nature, 447(7143), 396–398.
Champagne, F. (2009). Epigenetic influence of social experiences across the lifespan.
Developmental Psychobiology, 52, 299–311.
Dawson, M. A., & Kouzarides, T. (2012). Cancer epigenetics: From mechanism to
therapy. Cell, 150(1), 12–27.
Godfrey, A. (2012). Minisymposium brings epigenetic experts to NIEHS. Environmental
Factor, (February 2012). NC: NIEHS Research Triangle Park.
Goodman, A. H., Heath, D., & Lindee, M. S. (Eds.) (2003). Genetic nature/culture:
Anthropology and science beyond the two-culture divide. Berkeley: University of California Press.
Haig, D. (2004). The (dual) origin of epigenetics. Cold Spring Harbor Symposium, Quantitative Biology, 69, 67–70.
Hardy, T., & Tollefsbol, T. O. (2011). Epigenetic diet: Impact on the epigenome and
cancer. Epigenomics, 3(4), 503–508.
Hou, L. X., Zhang, D. W., & Baccarelli, A. (2012). Environmental chemical exposures
and human epigenetics. International Journal of Epidemiology, 41(1), 79–105.
Kaiser, J. (2014). The epigenetic heretic. Science, 24(343), 361–363.
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
13
Kuzawa, C., & Sweet, E. (2009). Epigenetics and the embodiment of race: Developmental origins of US racial disparities in cardiovascular health. American Journal
of Human Biology, 21(1), 2–15.
Landecker, H., & Panofsky, A. (2013). From social structure to gene regulation and
back: A critical introduction to environmental epigenetics for sociology. Annual
Review of Sociology, 39, 333–357.
Lester, B. M., Tronick, E., Nestler, E., Abel, T., Kosofsky, B., Kuzawa, C. W., … , Wood,
M. A. (2011). Behavioral epigenetics. Annals of the New York Academy of Sciences
1226: 14–33.
Miller, G. (2010). The seductive allure of behavioral epigenetics. Science, 2(329),
24–27.
McGowan, P. O., Sasaki, A., D’Alessio, A.C., Dymov, S., Labonte, B., Szyf, M., … ,
Meaney, M. J. (2009) Epigenetic regulation of the glucocorticoid receptor in human
brain associates with childhood abuse. Nature Neuroscience 12(3-March): 342–348.
Relton, C. C., & Smith, G. D. (2010). Epigenetic epidemiology of common complex disease: Prospects for prediction, prevention, and treatment. PLoS Med, 7(10),
e1000356.
Riggs, A. D., Martienssen, R. A., & Russo, V. E. A. (1996). Introduction. In V. E. A.
Russo, R. A. Martienssen & A. D. Riggs (Eds.), Epigenetic mechanisms of gene regulation. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Rodríguez-Paredes, M., & Esteller, M. (2011). Cancer epigenetics reaches mainstream
oncology. Nature America, 17(3), 330–339.
Romagnolo, D. F., Dashwood, R., & Ziegler, T. R. (2012). Nutritional regulation of
epigenetic changes. Advances in Nutrition, 3(5), 749–750.
Sahu, S. C. (2012). Toxicology and epigenetics. West Sussex, England: J. C. Wiley Ltd.
Szyf, M. (2011). The implications of DNA methylation for toxicology: Toward toxicomethylomics, the toxicology of DNA methylation. Toxicological Sciences, 120(2),
235–255.
Thornburg, K., Shannon, J., Thuillier, P., & Turker, M. (2010). In utero life and epigenetic predisposition for disease. Advances in Genetics, 10(71), 57–78.
Van den Veyver, I. B. (2002). Genetic effects of methylation diets. Annual Review of
Nutrition, 22, 255–282.
Waddington, C. H. (1942). The epigenotype. Endeavour, 1, 18.
FURTHER READING
Landecker, H., & Panofsky, A. (2013). From social structure to gene regulation and
back: A critical introduction to environmental epigenetics for sociology. Annual
Review of Sociology, 39, 333–357.
Sahu, S. C. (2012). Toxicology and epigenetics. West Sussex, England: J. C. Wiley Ltd.
WIKIPEDIA ANNOTATION AND GLOSSARY
Allostatic Health, http://en.wikipedia.org/wiki/Allostatic_loaD
BRCA1, http://en.wikipedia.org/wiki/BRCA1
14
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
DNA Mutations, Polymorphisms, http://en.wikipedia.org/wiki/Mutation; http://
en.wikipedia.org/wiki/Polymorphic_DNA
Epigenetic Mechanisms, Methylation, Chromatin Modifications, Histone Modification, Nucleosome Remodeling, RNA-mediated targeting, http://en.wikipedia.
org/wiki/Epigenetics#Mechanisms
Epigenomics, Methylation, Hypomethylation, Hypermethylation, Pryosequencing, Global Array Analysis, Genome-wide sequencing, Bisulfite modification,
http://en.wikipedia.org/wiki/Epigenomics
Endocrine disruptor, http://en.wikipedia.org/wiki/Endocrine_disruptor
Glucocorticoid Receptor, http://en.wikipedia.org/wiki/Glucocorticoid_receptor
Mitosis, Mitotically, Meiosis, Meiotically, http://simple.wikipedia.org/wiki/
Mitosis; http://simple.wikipedia.org/wiki/Meiosis
Mendelian Randomization Approach, Instrumental Variables Model, http://
en.wikipedia.org/wiki/Mendelian_randomization
DOUGLAS L. ANDERTON SHORT BIOGRAPHY
Douglas L. Anderton is Distinguished Professor and Chair of Sociology at
the University of South Carolina. He researches population–environment
interactions, social epigenetics, and historical population health. He is
author of over 60 journal articles as well as Demography: Study of Human
Populations (2007), Population of the United States (1998), Fertility on
the Frontier (1993); and edited Public Sociology (2006), and Readings in
Population Research Methodology (1997). He is completing a historical
monograph, Manufacturing Grammars of Death, and his recent social
epigenetic research is with Dr. Kathleen Arcaro, addressing environmental
contaminants, methylation, and breast cancer using breast milk as a bioassay.
KATHLEEN F. ARCARO SHORT BIOGRAPHY
Kathleen F. Arcaro is an Associate Professor in the Department of Veterinary
Sciences at the University of Massachusetts-Amherst. Her research is focused
on discovering epigenetic biomarkers of breast cancer risk and understanding the epigenetic mechanisms underlying drug resistance in the treatment
of breast cancer. Her long-term study of breast milk is aimed at determining
the extent to which the exfoliated epithelial cells in milk can be used to assess
breast cancer risk.
RELATED ESSAYS
Telomeres (Psychology), Nancy Adler and Aoife O’Donovan
The Sexual Division of Labor (Anthropology), Rebecca Bliege Bird and Brian
F. Codding
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
15
Genetics and the Life Course (Sociology), Evan Charney
Sexual Behavior (Anthropology), Melissa Emery Thompson
Genetic and Environmental Approaches to Political Science (Political Science),
Zoltán Fazekas and Peter K. Hatemi
Evolutionary Approaches to Understanding Children’s Academic Achievement (Psychology), David C. Geary and Daniel B. Berch
Genetics and Social Behavior (Anthropology), Henry Harpending and Gregory Cochran
An Evolutionary Perspective on Developmental Plasticity (Psychology), Sarah
Hartman and Jay Belsky
Genetic Foundations of Attitude Formation (Political Science), Christian Kandler et al.
Niche Construction: Implications for Human Sciences (Anthropology), Kevin
N. Laland and Michael O’Brien
Evolutionary Perspectives on Animal and Human Personality (Anthropology), Joseph H. Manson and Lynn A. Fairbanks
Behavioral Heterochrony (Anthropology), Victoria Wobber and Brian Hare
-
Social Epigenetics: Incorporating
Epigenetic Effects as Social Cause
and Consequence
DOUGLAS L. ANDERTON and KATHLEEN F. ARCARO
Abstract
Epigenetics is a field of study that invites an interdisciplinary interaction of the social
and biological sciences. This collaboration has, in fact, led to a blossoming research
community over the past two decades, which is using new data, methods, and conceptual frameworks to address a host of old and emergent research questions. A
recent (2014) search of PubMed found over a thousand articles on social, behavioral, and cognitive epigenetics. If one includes epidemiological epigenetic studies
that incorporate either social causes or consequences in their research, the number
expands nearly threefold. Yet, social epigenetics is a still nascent field, marginalized
and misunderstood in social science. In this essay, we attempt to review basic epigenetic concepts and the way in which epigenetics has, and can be, of use to social and
behavioral scientists in addressing some of the most fundamental sorts of questions
their disciplines raise.
INTRODUCTION
Reference to epigenetics, and the use of epigenetic concepts, has grown over
the recent past in many social and behavioral science disciplines (Goodman,
Heath, & Lindee, 2003; Landecker & Panofsky, 2013) and has an already
extensive history within cognitive psychology (Lester et al., 2011; Miller,
2010). However, a long history of the abuse of genetic concepts in the social
sciences has also been followed by a subsequent disregard by many for the
potential role of genetics in phenomena of behavioral and social interest.
Much of the abuse of genetic constructs in the social sciences came from
simplistic presumptions that complex socially expressed or constructed
traits, such as race, intelligence, class, and so on, could be traced to specific
fixed underlying genetic expression and destiny, an approach fueled by
racism, classism, and sexism. While genetic causation retained a following among more biologically inclined social sciences (e.g., demography,
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.
1
2
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
neuropsychology, and epidemiology) the relationship of genetics to behavior and broader social constructs has been largely shunned until recent
times. Newer generations have been more willing to investigate the social
consequences and causes of genetic influences, especially as embodied in the
notion of a mutable, even environmentally socially responsive, epigenetic
influence on phenotypes, rather than an emphasis on, in the short-run
fixed or immutable, DNA genotypes imposed as dominant upon the social
landscape. The potential for “social epigenetics,” embracing the integration
of epigenetic measures and mechanisms into social and behavioral science
research, as cause, consequence, or mechanism, is an exciting and potentially
very fruitful line of inquiry.
FOUNDATIONAL RESEARCH: WHAT IS EPIGENETICS?
The origins and uses of the term “epigenetic” are varied (Haig, 2004).
Usage vaguely familiar to that of modern molecular biologists dates at
least to Waddington’s (1942) reference to underlying mechanisms by which
“genes of the genotype bring about phenotypic effects.” For at least the past
two decades it has, more or less, had a familiar meaning defined (in the
negative) as “the study of mitotically and/or meiotically heritable changes
in gene function that cannot be explained by changes in DNA sequence”
(Riggs, Martienssen, & Russo, 1996). The notion that epigenetic changes to
gene function are heritable dates back to the mid-1960s. However, other
definitions remain in use, and the centrality of heritability to a definition of
epigenetics is not universally accepted. Bird, for example, defines epigenetics more broadly and as “the structural adaptation of chromosomal regions
so as to register, signal or perpetuate altered activity” (2007). And, for much
of the work relevant to social scientists, this broad definition is appropriate
and appealing. This definition emphasizes, as did Waddington’s, epigenetic processes that alter gene expression, including DNA methylation as
well as other DNA and chromatin modifications that have not yet been
demonstrated to be heritable.
At the risk of over-abstraction, it is perhaps helpful to pause and present
things in a simplified form. Consider a gene, or DNA, is largely fixed where
each gene is associated with an expression of genetic information or “trait.”
Here, trait simply means a specific molecular genetic expression, or the way
in which information from a gene is used on a molecular level (not trait in the
abusive sense of race, intelligence, etc.). The gene’s expression can be altered
by various molecular processes which are by (negative) definition epigenetic,
changes to the gene from outside. These epigenetic influences on gene expression are in contrast to the strictly genetic influences that include mutations or
polymorphisms in the DNA sequence. Some of these processes are, in turn,
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
3
influenced even by the wider environment and social phenomena such as
diet, stress, or environmental exposures. Thus, the social world can influence the expression of genes in a molecular manner. In turn, many believe
this epigenetically altered genetic expression can have a broader influence
on phenomena of social interest, ranging, for example, from specific genetically related diseases, endocrine function, and bodily reactions to stress, to
even antisocial behavior or suicidal tendencies. There is evidence that some
epigenetic changes can occur in utero, such that the environment of one generation can influence the epigenetic state of the next generation. And, there
is evidence of truly heritable epigenetic transmission from one generation to
the next through sexual reproduction, implying potentially intergenerational
effects from, as a hypothetical example, one generation’s childhood dietary
behavior to the next generation’s childhood health. Lastly, it is worth noting that some discussions often refer to the cumulative epigenetic changes in
gene expression as the “epigenome” or a “ghost gene” although this collection of epigenetic changes does not exist, in any cohesive molecular sense,
independent of the gene.
At the molecular level, epigenetic processes involve the placement of chemical tags directly on the DNA or on the proteins that help organize the DNA
(Rodríguez-Paredes & Esteller, 2011). These chemical tags are critical in controlling the expression of the gene and in maintaining genetic stability and
can be passed on to daughter cells at cell division. By far the most studied tag
is the methyl group, a carbon atom with three hydrogen atoms (CH3 ) that can
be attached at specific locations on our DNA. DNA methylation is only one of
several epigenetic mechanisms. However, because methylation studies are so
prevalent, including among the examples we cite, methylation merits a brief
explanation. In humans, methyl groups can be attached to only one of the
four DNA bases, cytosine (C), and only when the cytosine is followed by guanine (G) in the DNA sequence. Because the cytosine and guanine on the same
strand are joined by a phosphate (p) molecule, the potential DNA methylation sites are referred to as CpGs. CpGs are not evenly distributed across our
genome. Regions of our DNA with a high density of CpGs are referred to as
CpG islands. Borrowing from oceanography, CpGs occurring a short distance
from an island are on the shore, and moving further out, the shelf followed
by the deep sea. The density and location of CpGs in our DNA sequence
is important. CpG islands occur more frequently within genes, in regions
responsible for turning the gene on and off (promoter regions), whereas as
low-density CpGs occur more frequently in DNA that is between genes and
is responsible for maintaining genetic stability. Biologists are busy cataloging
the location and density of methylated CpGs in various tissues from an array
of developmental, exposure, and diseased states; however, much remains to
be learned about the biological effects of methylation at specific CpGs.
4
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
CUTTING-EDGE RESEARCH: USING EPIGENETIC DATA IN THE
SOCIAL SCIENCES
Just as one does not have to be an oncologist to study the social correlates of
cancer, for example, one does not have to be a molecular biologist to consider
the role of epigenetics in social and behavioral models. In most instances,
research addressing epigenetic influences is inherently collaborative and
increasingly in the model of “big science” involving a team of, perhaps,
molecular biologists, epidemiologists, behavioral experts, disease specialists, research physicians, cognitive psychologists, laboratory technicians,
and so on, all of whom will be involved in formulating the central research
problem on which social and behavioral scientists collaborate, identifying
gene regions or targets of epigenetic interest, defining likely epigenetic
covariates, supervising appropriate biological assays, and so on. This is true
regardless of whether the driving research question is one generated from
within the social sciences (e.g., environmental justice research on differential
exposure of minority populations to potential epigenetic damage) or one
generated by any other discipline that simply relies upon expertise from
the social and behavioral sciences for the context in which those disciplines
would view the research problem at hand.
To facilitate a somewhat pragmatic discussion of epigenetic research and
social science in this big science context, let us start with a word about operationalization and measurement. Again, although methylation is only one
form of epigenetic change, it has had attention from molecular biologists
and social scientists in particular because of potential epigenetic effects on
prominent diseases (e.g., various cancers), behaviorally related body processes (e.g., cortisol response) and evidence of potentially direct cognitive
pathways (e.g., suicidal tendencies). Quantitative measures, usually percentage, of methylation in specific gene regions or sites of interest are the most
common measurements analyzed. Processing cell samples (e.g., saliva, tissue
samples, epithelial cells, etc.) to obtain laboratory measurements of methylation for various regions or sites of the gene is still carried out in many individual laboratories but is increasingly a big business and a matter of contracting
out for measurements using biological assays collected by researchers. For
this reason, we will not spend time here on various laboratory methods and
technical procedures. Instead, the focus here is on the potential to be achieved
from integrating such measures into social and behavioral scientific research
and the overarching issues of research design and data analysis surrounding
epigenetic work of importance to the social and behavioral sciences.
One distinction, of relevance, is between measurements from deductively
targeted genetic sites of interest (e.g., pyrosequencing for measurement of
methylation in specific promoter regions) and the increasing use of inductive
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
5
or exploratory measurements across broad swaths of the genome (e.g., global
array analysis and genome-wide sequencing of bisulfite-modified DNA,
although these methods can also be used to obtain targeted site information). The challenges these different approaches present to statistical analysis
can be very difficult, ranging from small sample size issues for precious
biological tissue samples, to the large amount of data generated by global
arrays and genome sequencing and complex statistical methods required
for analysis. Commercial laboratories providing methylation measurement
services, especially array measurements, also tend to provide basic software
for analyzing the data and complementary routines are available in public
software, largely contributed routines in the R-software suite. Again, the
complexity of statistical analysis for much epigenetic data often calls for a
big science approach with a knowledgeable statistical analyst.
Thus, without trivializing the complexities of methylation analysis, it is not
unreasonable to say the social scientists will usually encounter epigenetics as
a database, not unlike many they are familiar with, of quantitative measurements of methylation for either target regions of the gene or for large sections
of the genome. Analytical approaches may be deductive and targeted at specific genetic sites using familiar analytical techniques or inductive and requiring high-throughput analysis, often using proprietary laboratory-supplied
software. In the remainder of this essay, we do not dwell on laboratory or
statistical methodology details. Instead, as suggested, we take an approach
in which epigenetic measures are operationalized and available as variables
for a specific research problem and focus on the conceptual or functional use
of epigenetics as a construct and framework within the social sciences. We do
this by focusing, in turn, on recent examples of work, and some speculation
as to the future of this work, in which epigenetic measures are used by social
scientists as independent causal effects, dependent variable outcome measures, intervening variables representing epigenetic mechanisms, durational
effects carrying causation from one point in time to a later manifestation of
outcomes and intergenerational variables potentially carrying the effects of
gene–environment interactions across generations.
KEY ISSUES FOR FUTURE RESEARCH: THE UTILITY AND ROLES
OF EPIGENETIC DATA IN THE SOCIAL SCIENCES
EPIGENETIC VARIATION AS A DEPENDENT OUTCOME
The most common existing use of epigenetic variation to date is that of a
biological assay, used as a dependent variable, or biomarker, of health outcomes in studies to identify exposures responsible for the epigenetic change.
6
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
Epidemiologists, demographers, health and medical researchers, anthropologists, neuropsychologists, and many others, have all shown interest in the
potential for incorporating epigenetic assays and measurements as potentially more direct indicators of environmental effects than abstract measures
of allostatic health, body condition, or future disease outcomes. Epigenetic
variations provide a measure of induced change in genetic fortunes. These
may be studied simply as indicators of environmental exposure or be targeted because of known associations with eventual disease outcomes. When
studying rare diseases with complex lifelong pathways, such as most cancers,
the prospect of studying outcome measures that may be more immediate,
but indicative of both environmental causation and thereby induced predisposition toward eventual disease outcomes, is appealing. Although there are
many studies that have employed epigenetic variation as a direct outcome
measure in this manner, two examples may be of particular interest in an
interdisciplinary context: dietary studies and epigenetic toxicology.
An early interest in the effect of social behaviors on epigenetic change
emerged in studies of the relationships between diet and methylation (Van
den Veyver, 2002) and of diet to other epigenetic mechanisms (Romagnolo,
Dashwood, & Ziegler, 2012). These interests have fueled an emerging science of nutrigenomics. Not surprisingly, most evidence that dietary intake
affects epigenetics exists for folate, which supplies building blocks of DNA
methylation. However, a long list of other dietary components have been
studied, including alcohol, flavonoids, polyphenols, lycopene, methionine,
B-vitamins, resveratrol, choline, genistein, and sulforaphane, among others,
and such common foods containing some of these compounds as green
tea, spinach, sunflower seeds, baker’s yeast, soy, broccoli sprouts, fiber, or
cruciferous vegetables (Hardy & Tollefsbol, 2011). One area of particular
attention in dietary studies is in prenatal diets where animal studies suggest
folate or choline deficits can cause lifelong hypomethylation. Another great
interest in dietary studies concerns the potential treatment aspects of diets
which contain dietary compounds that modulate epigenetic programming
and can be used to target methylated promoters known to contribute
to cancers. Studies of various cancers sometimes consider both overall
genome-wide reduction in DNA methylation (global hypomethylation)
resulting in chromosomal instability and hypermethylation within the
CpG islands of a specific gene promoter, often tumor suppressing genes,
as an outcome variable. The prospects of identifying hypermethylated,
cancer-related, promoters and targeting them with dietary demethylating
agents before disease develops drives a great deal of interest in studies of
diet and methylation.
Another interest in the use of epigenetic change as outcome variables
comes from communities who are in search of sensitive, widely detectable,
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
7
indications of environmental or toxicological exposure. The use of epigenetic
variations as direct outcomes in examining toxicological or environmental
effects has emerged as a new field of considerable potential (Sahu, 2012;
Szyf, 2011). Researchers who examine diffuse environmental hazards and
issues of environmental justice, for example, have long struggled with the
fact that outcome measures such as disease incidence rates measure only the
potentially rare events that exist at the end of a long and complex pathway
that may nonetheless be impacted by environmental exposures and bodily
predisposition to eventual outcomes. Use of measures such as epigenetic
variations that are more immediate mechanistic effects of environmental
exposures is of great value in such circumstances. Focusing on epigenetic
mechanisms may explicate an actual pathway of ultimate harm, while
addressing the research problems posed by disease outcomes that may
remain both rare in exposed populations and may not manifest for years
hence. This interest has again given rise to a host of studies of specific environmental toxicants such as endocrine disruptors, heavy metals, and other
persistent and common environmental pollutants in our air, food, and water
supplies. Most environmental epigenetic studies have utilized methylation
measures. However, the still emerging and new field of environmental
epigenomics has grown to encompass studies across the entire range of
epigenetic mechanisms (Godfrey, 2012; Hou, Zhang, & Baccarelli, 2011).
EPIGENETIC VARIATION AS A CAUSAL INFLUENCE
While most studies have emphasized and utilized epigenetic change as a
dependent variable or outcome of environmental circumstances, other profitable areas of epigenetic research have focused on using epigenetic changes
or biomarkers as an independent variable responsible for direct, or eventual,
disease or behavioral outcomes. Some studies have gone even further and
explored both causes and consequences of epigenetic variation in a single
study, bringing epigenetic variation into the more realistic role of an intervening variable. Again, two examples may suffice to illustrate the uses and
potential of this epigenetic approach: neurocognitive/behavioral epigenetics
and epigenetic cancer studies.
Within the social and behavioral sciences, neurocognitive and behavioral
epigenetics is likely the most familiar application of epigenetics to many
and has been the subject of most futuristic speculation. Most, not all, of the
research in neurocognitive epigenetics has focused on identifying epigenetic
variations associated with cognitive or behavioral outcomes. Epigenetic
variation has been considered, for example, as a potential explanation in
processes of memory, learning, age-related cognitive changes, depression,
aggression, mental illness, and other cognitive outcomes of interest. Studies
8
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
have also examined epigenetic effects on behaviorally related endocrine
response systems and for specific mental illness outcomes. Epigenetic
variations have been implicated in the neurodegenerative disorders such as
Alzheimer’s and Parkinson’s disease, the most common form of inherited
mental retardation (fragile X syndrome), Rett syndrome, and more recently,
as a potential influence on autism spectrum disorders.
Given the difficulty of brain tissue sampling, one of the most notable studies of this sort is a cadaver study of suicide victims (McGowan et al., 2009).
This study replicated earlier animal model work supporting a plausible association between methylation of glucocorticoid receptors and later life abilities
to cope with stress. By comparative analysis of those who suffered childhood
abuse and those who did not, the authors also suggest abusive stress may be
a source of DNA methylation, bringing methylation into the role of an intervening variable affecting later suicide among child abuse victims. This study
is only one of many suggesting a link between stress and later life cognition
and health. Miller (2010) summarizes this research and several threads of
more recent, specifically cognitive, epigenetic research including changes in
gene expression through histone modification that are related to subsequent
learning and memory, as well as age-related changes in memory, across the
lifespan.
Undoubtedly, the greatest strides in studying epigenetic change as a direct
cause of socially important outcomes have been in the epidemiological study
of methylation and disease outcomes, especially cancer. After nearly two
decades of numerous studies seeking to identify specific misregulation in
epigenetic mechanisms (including DNA methylation, histone modification,
nucleosome remodeling, and RNA-mediated targeting) that can culminate in
cancer, there is little doubt that such connections are abundant (see Dawson &
Kouzarides, 2012, for one recent summary). However, despite these numerous studies implicating specific epigenetic misregulation with both tumor
suppression and cancer outcomes, there are not, as yet, definitive epigenetic
biomarkers identified as suitable for routine clinical practice. One forefront of
epigenetic cancer research in the coming decades will be to advance research
to the point of contributing to clinical diagnosis and treatment of specific cancers. Of course, identification of specific pathways is also potentially critical
to preventive measures. While much has been achieved, even in this most
prolific area of research, epigenetics is a nascent science.
With a growing number of epidemiological studies, the potential role of
methylation, as not only a cause but also an intervening effect between
social behaviors and disease or behavioral outcomes is naturally receiving
considerable attention. Research in developmental psychology by Frances
Champagne and others (see Champagne, 2009 for an overview), for example,
provides examples from animal model research suggesting early exposures
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
9
to stress could epigenetically modify later life stress response mechanisms;
or that juvenile isolation could influence alter epigenetic pathways that
contribute in turn to isolation syndrome among adults; and other similar
examples of species-specific social behavior with epigenetic intervening
effects that subsequently impact social behavior. Discussion of mechanisms
of socially mediated epigenetic misregulation in human populations (e.g.,
excessive alcohol consumption that reduces folates related to epigenetic
changes) coupled with studies linking such epigenetic changes to disease
outcomes (e.g., liver cancer), suggest a fuller understanding of the relation
of social behaviors to disease, or behavior, through specific epigenetic
pathways is within reach. For many diseases such as breast cancer, where
the major correlates include not only heredity (e.g., BRCA1) but significant
influences that are largely social in character (e.g., alcohol, diet, obesity,
childbearing, and breastfeeding, etc.), epigenetics may provide a critical
missing link in the pathways between social behavior and socially significant
disease and behavioral outcomes.
Again, even in the most advanced areas of epigenetic research, our understanding of the complex pathways linking environments to epigenetics to
outcomes is only now emerging. The potential of such research excites the
imagination. Yet, such speculation is also warranted to the extent that it
stimulates new studies and explorations of problems that have long defied
the simple bifurcated heredity plus behavior equation. Particularly from a
social science or public health perspective, the environmentally interactive
and intervening role of epigenetic change may be critically important
for understanding both cause and consequence of socially important
“epidemic” conditions (e.g., obesity or autism) and ill-defined syndromes
(e.g., chronic fatigue, environmental sensitivity, fibromyalgia) that have
been largely socially and symptomatically defined, with, as yet, poorly
understood biological pathways. With such high-value targets in sight, and
a new paradigm with some seeming promise, the occasional speculative, or
even wildly speculative, hypothesis is perhaps deserving of some latitude.
EPIGENETICS AND DURATIONAL EFFECTS
For social epigenetics, one of the most interesting and useful elements of
the conceptual framework afforded is the durational aspects of epigenetic
effects. As in much of the research cited earlier regarding social behaviors
(e.g., Champagne, 2009), epigenetic influences such as an environmental
exposure may occur early in the lifespan, or in utero, and those relatively
durable epigenetic changes carry consequences forward in time such
that they then manifest in an impact on disease or biological and cognitive outcomes in later life. This durational model, carrying the effects
10
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
of environmental exposures far removed in time to ultimately observed
probabilistic outcomes, provides a mechanism suitable to numerous social
and epidemiological inquiries. Animal models demonstrating that maternal
crack cocaine exposure alters epigenetic profiles, resulting in dysfunctional
social interactions among offspring throughout their life course, have raised
awareness of the plausibly groundbreaking potential for such studies in
human populations. There is, as yet, more speculation than research in this
regard. Kuzawa and Sweet (2009), for example, argue the correlated racial
disparities in birth weight differences and cardiovascular diseases across
groups offer a rationale for considering epigenetic links between early-life
environmental factors, such as maternal stress during pregnancy and adult
racial disparities in cardiovascular diseases. Similarly, Thornburg, Shannon,
Thuillier, and Turker (2010) argue that such differences, along with research
linking rapid growth in the womb to metabolic disease and obesity and also
to breast and lung cancers, calls for research to determine the epigenetic
processes underlying these linkages.
Demographers and epidemiologists have also long been interested in
durational processes in which in utero, or early life-course, events and
stress have resulted in long-term adult morbidity and mortality in historical
populations. Such durational effects are often simply regarded as a “black
box,” increasing frailty among the exposed populations and the likelihood of
later morbidity and mortality. Epigenetics provides a conceptual framework
that may well lead us to more specific pathways by which distinctive early
life-course exposures and events influence the likelihood of particular
morbidity outcomes over the life course. The potential for such research is
significant. However, the implementation of longitudinal life-long epigenetic
studies of early life-course effects on epigenetic change and probabilistic
outcome variables, such as morbidity, is anything but simple compared to a
short-term clinical trial, for example, to examine the effects of stress or diet
on epigenetic misregulation. An increasing recognition for the big science
model needed to address such questions, including the linkage of large
longitudinal population studies to genetic and epigenetic data, provides
some hope for significant advances in such research despite the many
challenges faced in such work.
TRANSGENERATIONAL EPIGENETICS
Without question, the most controversial and one of the most exciting
prospects regarding epigenetics has been some initial evidence for transgenerational transmission of epigenetic effects (Kaiser, 2014). This research goes
beyond the durational model, which supports possible mutigenerational
(but not transgenerational) effects. In a multigenerational model early-life
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
11
or in utero conditions may create epigenetic changes that are manifest in
later life course outcomes of offspring but exposure of those offspring
occurred after sexual reproduction and is not transgenerational or inherited.
However, some epigeneticists argue, and some evidence from animal model
experiments suggests, that induced epigenetic changes can be transmitted
meiotically, or through sexual reproduction, to the next generation or even
several generations hence. This research is still in very early stages, with
a variety of protocols, some supporting studies, and some failing to find
support. Difficulties in conducting such research are vastly greater, yet
again, than those faced by life-long studies of epigenetic effects. Research
has been exclusively confined to animal models. In addition, a biomolecular
mechanism for this intergenerational transmission epigenetic misregulation
has also not yet been elaborated. So, as with all truly cutting-edge research,
the jury is still deliberating not simply the promise but the reality of
transgenerational epigenetics.
Yet, while scholars are still divided and there are skeptics, there are also
sufficient converts to transgenerational epigenetics, and enough cumulating
supporting evidence, to ensure that this line of research will continue. Transgenerational epigenetic change will ultimately either be refuted or, perhaps
more likely, grow into in a new, and paradigm-shifting, field of future study.
If transgenerational epigenetic hypotheses are confirmed, the processes to be
studied will not be a simple replicate of studies into heritability among future
generations. Even among those who strongly support the notion of transgenerational epigenetic change, this claim is often accompanied by the appropriate qualification that continuing methylation and demethylation influences
attenuate the force of transmission across generations in ways that will need
to be better understood.
The prospect of transgenerational epigenetics for social science is revolutionary to say the least and one which fuels enthusiastic speculation.
The idea, for example, that social circumstance of one generation may
contribute to epigenetic changes which are then transmitted to, and impact
the life-course prospects, of successive generations, could help to explain
the intergenerational effects of circumstance on social life that have been
found persistent, even in the face of intergenerationally changing social
environments and circumstance.
LIMITS AND FURTHER CONSIDERATIONS
There are many limits to the growing enthusiasm for epigenetic research and
its expansion into other disciplines including the social sciences. Many current studies, for example, rely ultimately on a selection by the dependent
variable and introduce the possibility of reverse causation. If we select tumor
12
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
samples, for example, to bioassay, how do we ensure that epigenetic changes
found in patients with tumors, and not controls, are not symptomatic rather
than causal. One emerging answer to this problem is the statistical use of
a Mendelian randomization approach (Relton & Smith, 2010) equivalent to
an instrumental variables model with genotype acting as an instrument for
the exposure of interest. Another problem in conducting such research is
simply the expense and difficulties of obtaining biological assays and the
big science funding model required of a truly interdisciplinary epigenetic
study. However, there has been an explosion of large data resources combining social and genome-wide array data (e.g., Add Health, the Health and
Retirement Study, Wisconsin Longitudinal Study, Framingham Heart Study)
and similar resources for genome-wide methylation studies are emerging.
The linkage of tissue banks and large-scale longitudinal population surveys
provide another potential avenue for social epigenetic research. A growing
number of epigenetic databases with linked methylation and survey data are
available for public use and efforts by funding agencies to support standardized assays, held in common for future research, have resulted in a growing
body of secondary data resources.
Another substantial hurdle lies in the lack of standardized protocols for
complex hypotheses and the most challenging studies such as transgenerational epigenetic research. There is no magic shortcut to a mature epigenetic
science and only continued study and careful replication will resolve these
issues over the near future.
REFERENCES
Bird, A. (2007). Perceptions of epigenetics. Nature, 447(7143), 396–398.
Champagne, F. (2009). Epigenetic influence of social experiences across the lifespan.
Developmental Psychobiology, 52, 299–311.
Dawson, M. A., & Kouzarides, T. (2012). Cancer epigenetics: From mechanism to
therapy. Cell, 150(1), 12–27.
Godfrey, A. (2012). Minisymposium brings epigenetic experts to NIEHS. Environmental
Factor, (February 2012). NC: NIEHS Research Triangle Park.
Goodman, A. H., Heath, D., & Lindee, M. S. (Eds.) (2003). Genetic nature/culture:
Anthropology and science beyond the two-culture divide. Berkeley: University of California Press.
Haig, D. (2004). The (dual) origin of epigenetics. Cold Spring Harbor Symposium, Quantitative Biology, 69, 67–70.
Hardy, T., & Tollefsbol, T. O. (2011). Epigenetic diet: Impact on the epigenome and
cancer. Epigenomics, 3(4), 503–508.
Hou, L. X., Zhang, D. W., & Baccarelli, A. (2012). Environmental chemical exposures
and human epigenetics. International Journal of Epidemiology, 41(1), 79–105.
Kaiser, J. (2014). The epigenetic heretic. Science, 24(343), 361–363.
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
13
Kuzawa, C., & Sweet, E. (2009). Epigenetics and the embodiment of race: Developmental origins of US racial disparities in cardiovascular health. American Journal
of Human Biology, 21(1), 2–15.
Landecker, H., & Panofsky, A. (2013). From social structure to gene regulation and
back: A critical introduction to environmental epigenetics for sociology. Annual
Review of Sociology, 39, 333–357.
Lester, B. M., Tronick, E., Nestler, E., Abel, T., Kosofsky, B., Kuzawa, C. W., … , Wood,
M. A. (2011). Behavioral epigenetics. Annals of the New York Academy of Sciences
1226: 14–33.
Miller, G. (2010). The seductive allure of behavioral epigenetics. Science, 2(329),
24–27.
McGowan, P. O., Sasaki, A., D’Alessio, A.C., Dymov, S., Labonte, B., Szyf, M., … ,
Meaney, M. J. (2009) Epigenetic regulation of the glucocorticoid receptor in human
brain associates with childhood abuse. Nature Neuroscience 12(3-March): 342–348.
Relton, C. C., & Smith, G. D. (2010). Epigenetic epidemiology of common complex disease: Prospects for prediction, prevention, and treatment. PLoS Med, 7(10),
e1000356.
Riggs, A. D., Martienssen, R. A., & Russo, V. E. A. (1996). Introduction. In V. E. A.
Russo, R. A. Martienssen & A. D. Riggs (Eds.), Epigenetic mechanisms of gene regulation. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Rodríguez-Paredes, M., & Esteller, M. (2011). Cancer epigenetics reaches mainstream
oncology. Nature America, 17(3), 330–339.
Romagnolo, D. F., Dashwood, R., & Ziegler, T. R. (2012). Nutritional regulation of
epigenetic changes. Advances in Nutrition, 3(5), 749–750.
Sahu, S. C. (2012). Toxicology and epigenetics. West Sussex, England: J. C. Wiley Ltd.
Szyf, M. (2011). The implications of DNA methylation for toxicology: Toward toxicomethylomics, the toxicology of DNA methylation. Toxicological Sciences, 120(2),
235–255.
Thornburg, K., Shannon, J., Thuillier, P., & Turker, M. (2010). In utero life and epigenetic predisposition for disease. Advances in Genetics, 10(71), 57–78.
Van den Veyver, I. B. (2002). Genetic effects of methylation diets. Annual Review of
Nutrition, 22, 255–282.
Waddington, C. H. (1942). The epigenotype. Endeavour, 1, 18.
FURTHER READING
Landecker, H., & Panofsky, A. (2013). From social structure to gene regulation and
back: A critical introduction to environmental epigenetics for sociology. Annual
Review of Sociology, 39, 333–357.
Sahu, S. C. (2012). Toxicology and epigenetics. West Sussex, England: J. C. Wiley Ltd.
WIKIPEDIA ANNOTATION AND GLOSSARY
Allostatic Health, http://en.wikipedia.org/wiki/Allostatic_loaD
BRCA1, http://en.wikipedia.org/wiki/BRCA1
14
EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES
DNA Mutations, Polymorphisms, http://en.wikipedia.org/wiki/Mutation; http://
en.wikipedia.org/wiki/Polymorphic_DNA
Epigenetic Mechanisms, Methylation, Chromatin Modifications, Histone Modification, Nucleosome Remodeling, RNA-mediated targeting, http://en.wikipedia.
org/wiki/Epigenetics#Mechanisms
Epigenomics, Methylation, Hypomethylation, Hypermethylation, Pryosequencing, Global Array Analysis, Genome-wide sequencing, Bisulfite modification,
http://en.wikipedia.org/wiki/Epigenomics
Endocrine disruptor, http://en.wikipedia.org/wiki/Endocrine_disruptor
Glucocorticoid Receptor, http://en.wikipedia.org/wiki/Glucocorticoid_receptor
Mitosis, Mitotically, Meiosis, Meiotically, http://simple.wikipedia.org/wiki/
Mitosis; http://simple.wikipedia.org/wiki/Meiosis
Mendelian Randomization Approach, Instrumental Variables Model, http://
en.wikipedia.org/wiki/Mendelian_randomization
DOUGLAS L. ANDERTON SHORT BIOGRAPHY
Douglas L. Anderton is Distinguished Professor and Chair of Sociology at
the University of South Carolina. He researches population–environment
interactions, social epigenetics, and historical population health. He is
author of over 60 journal articles as well as Demography: Study of Human
Populations (2007), Population of the United States (1998), Fertility on
the Frontier (1993); and edited Public Sociology (2006), and Readings in
Population Research Methodology (1997). He is completing a historical
monograph, Manufacturing Grammars of Death, and his recent social
epigenetic research is with Dr. Kathleen Arcaro, addressing environmental
contaminants, methylation, and breast cancer using breast milk as a bioassay.
KATHLEEN F. ARCARO SHORT BIOGRAPHY
Kathleen F. Arcaro is an Associate Professor in the Department of Veterinary
Sciences at the University of Massachusetts-Amherst. Her research is focused
on discovering epigenetic biomarkers of breast cancer risk and understanding the epigenetic mechanisms underlying drug resistance in the treatment
of breast cancer. Her long-term study of breast milk is aimed at determining
the extent to which the exfoliated epithelial cells in milk can be used to assess
breast cancer risk.
RELATED ESSAYS
Telomeres (Psychology), Nancy Adler and Aoife O’Donovan
The Sexual Division of Labor (Anthropology), Rebecca Bliege Bird and Brian
F. Codding
Social Epigenetics: Incorporating Epigenetic Effects as Social Cause and Consequence
15
Genetics and the Life Course (Sociology), Evan Charney
Sexual Behavior (Anthropology), Melissa Emery Thompson
Genetic and Environmental Approaches to Political Science (Political Science),
Zoltán Fazekas and Peter K. Hatemi
Evolutionary Approaches to Understanding Children’s Academic Achievement (Psychology), David C. Geary and Daniel B. Berch
Genetics and Social Behavior (Anthropology), Henry Harpending and Gregory Cochran
An Evolutionary Perspective on Developmental Plasticity (Psychology), Sarah
Hartman and Jay Belsky
Genetic Foundations of Attitude Formation (Political Science), Christian Kandler et al.
Niche Construction: Implications for Human Sciences (Anthropology), Kevin
N. Laland and Michael O’Brien
Evolutionary Perspectives on Animal and Human Personality (Anthropology), Joseph H. Manson and Lynn A. Fairbanks
Behavioral Heterochrony (Anthropology), Victoria Wobber and Brian Hare