Population Health: Behavioral and Social Science Insights

By W. Thomas Boyce

Abstract

Given what we know, even now, about the interactive roles of genetic and environmental variation in pathogenesis, the advent of epigenomics, which illuminates the physical, molecular points of connection between genes and contexts, will almost certainly revolutionize theories of disease causation. Because social environments are among the most powerful determinants of health and illness, such theories will also increasingly blur conceptual distinctions between the social and biological sciences. Coming breakthroughs in epigenetic research are poised, as well, to elucidate fundamental enigmas of gene-environment interplay, such as the role of critical periods and developmental time, individual differences in susceptibility to environmental exposures, and the possible role of intergeneration transmission of risks. Solving such mysteries will profoundly influence the practice of medicine, the development of new preventive interventions, and the course of future research.

Introduction

Attention has been repeatedly called to the gaping conceptual and epistemic divisions between the biological and social sciences¹ or, more generally and historically, between the sciences and the humanities.² By contrast, this chapter presents an argument for the existence of a molecular convergence between the biological and social sciences—a convergence that will define and illuminate the common ground between nature and nurture. The nascent field of epigenetics is revealing a mechanism by which environmental variation influences gene expression through alterations in chromatin biology, thereby reconciling the organismic and contextual accounts for differences in health, disease, and human development.

The Human Genome Project advanced a revolutionary understanding of how human destiny is shaped by individual and population level differences in genetic sequences.^3,4 This sequencing of the human genome launched an exhilarating but often preoccupying debate over the roles played by common and rare genetic variants in the etiology of complex human diseases.^5,6 Concurrently, the burgeoning fields of social epidemiology, health psychology, and behavioral medicine have systematically revealed profound effects of socioemotional and cultural contexts on trajectories of development and risks of disease.^7-9 Genetically informed study designs have documented both genetic and contextual contributions to the origins of many, if not most, disease phenotypes, but the proportion of phenotypic variation attributable to differences in gene sequences remains, in most cases, substantially smaller than published heritability estimates. ¹⁰ preemptive and catalytic epigenomic science has now proposed that such "missing heritability" may be most readily discoverable in the epigenetic embedding of environmental exposures onto the background DNA sequence.^11,12 The obsolescent, genetic, and environmental forms of determinism spawned by each side of this scientific divide have given way to a new vision of how genetic biology and social environments physically intersect in the epigenetic events and processes shaping development and destiny.

Conrad Waddington first used the term "epigenetic" in relation to human development in a 1942 paper examining the processes and events by which genotypes are functionally linked to adult phenotypes.¹³ The now flourishing science of epigenetics (from the Greek root epi, meaning upon or over) has been defined as the study of "structural adaptations of chromosomal regions so as to register, signal, or perpetuate altered [gene] activity states."¹⁴ Epigenetic mechanisms change gene activity or expression by altering DNA organization without modifying the genetic code of the DNA.¹¹ As illustrated in Figure 1, chromatin is the physical packaging of DNA within chromosomes in the form of linear sequences of genes wrapped around groups of histone proteins, together resembling "beads on a string."

Figure 1: The chromatin packaging of DNA wound around histone protein octamers

Source: National Human Genome Research Institute. Available at https://www.genome.gov/Glossary/index.cfm?id=32.

The multiple chemical tags—or epigenetic "marks"—placed on DNA (specifically, on the cytosine nucleotides within a cytosine-guanine, or CpG, sequence) or histone proteins as a consequence of environmental exposures and experiences, regulate the density of chromatin conformation, thereby modulating physical access of RNA polymerases, the decoding enzymes, to the DNA sequence. Epigenetic marks also determine the emergence of stable differences in cell types during embryogenesis, thus creating an "epigenetic paradox" in which histological stability over time and dynamic, moment-to-moment changes in response to experience are both encoded in the same epigenetic patterns governing gene transcription.¹⁵ The presence or absence of these chemical marks, such as DNA methylation and histone protein acetylation, methylation, ubiquitination, or phosphorylation, determines the organization of chromatin structure, which ultimately controls transcription. Transcription, in turn, governs the emergence of phenotypes through the influence of its protein products on, for example, the development of brain function and structure. Epigenetic marks are reversible through the activity of histone deacetylases (HDACs) and other demethylating enzymes, rendering chromatin restructuring processes potentially modifiable and accessible to pharmacological interventions that target such enzymes.¹⁶ Patterns of epigenetic tagging of DNA and histone proteins thus constitute physical points of connection between genes and environments, not unlike the synapses that establish a physical nexus between neurons, allowing neuron-to-neuron communication and the formation of neural circuits.

Although patterns of epigenomic variation are clearly influenced by aspects of physical environmental exposure, such as diet and physical toxins,^17,18 it also has become clear that perturbations in socioemotional contexts can result in long-term changes in the human and animal epigenomes.^19-21 Epidemiologic research by Borghol et al.,²² Essex et al.,18and Kobor and colleagues^23,24 has detected longitudinal associations between childhood disadvantage and genome-wide promoter methylation in mid-life, between parental stress in infancy and differential methylation in adolescence, and between early socioeconomic status (SES) and inflammatory gene expression in leukocyte transcriptomes. Essex and colleagues¹⁸ found that systematic, long-term differences in DNA methylation in buccal epithelial cells were present in mid-adolescence among youth whose parents had reported high levels of stress and adversity in the youths' infancy and preschool years (Figure 2). Further, the pattern of maternal and paternal differential methylation was commensurate with the known timing of parent-specific influences on development, i.e., greater maternal stressor effects in infancy, followed by an emergence of paternal effects in the preschool years. Such findings, of whole genome differences in DNA methylation status, are also consistent with the known cellular effects of stress hormones, such as glucocorticoids, which regulate expression of approximately 10 percent of the genome.²⁵ They are supported, as well, by reports of real time, dynamic changes in both promoter and exon region gene methylation among individuals exposed to laboratory-based, stressful challenges²⁶ and evidence of reduced expression of HDAC and proinflammatory genes under meditation conditions.²⁷ Epigenetic processes may also be implicated in the emergence of resilience²⁸ and in the beneficial developmental effects of enriched early environments.²⁹

Chromatin Modifications in Gene-Environment Interplay

Modifications of chromatin also appear likely to underlie and, in some cases, mechanistically explain the gene x environment (GxE) interactions documented at an accelerating pace over the past decade among human samples. The reports of Caspi, Moffitt, and colleagues^30,31 from the Dunedin Multidisciplinary Health and Development Study revealed statistical interactions between early environmental conditions (e.g., child maltreatment and stressful life events) and functional gene polymorphisms (e.g., the MAOA, monoamine oxidase A, and 5HTT, serotonin transporter, genes) in the prediction of antisocial behavior and depression/suicidality. Although legitimate concerns have been raised for the replicability of GxE interaction reports,³² some such effects have been replicated in independent samples,³³ and new evidence continues to accrue. GxE interactions have been observed and recorded in both human^34,35 and animal³⁶ species, leading some to assert that, in effect, virtually all disorders of health and development may be both genetic and environmental in origin, in the sense that all result from the mutually interactive influences of both.³⁷ The search for reliable GxE interactions may be abetted by the development of both empirical evidence that polygenic risk scores associated with developmental phenotypes can discern promising new single nucleotide polymorphism (SNP) targets³⁸ and computational models suggesting that genome-wide association study (GWAS) approaches to GxE discovery may be more promising than candidate SNP by SNP searches.³⁹ In fields such as psychiatric genomics, the way forward appears to lie in new knowledge of how multiple genes with additive or multiplicative effects, each with incremental influences, are assembled into functional genetic networks that interact with social environmental conditions to produce important phenotypic disorders.⁴⁰

Figure 2: Differential methylation of buccal epithelial cell CpG sites in mid-adolescence, by parental reports of stress and adversity in children's infancy and preschool periods

Source: Essex MJ, Boyce WT, Hertzman C, et al. Epigenetic vestiges of early developmental adversity: hildhood stress exposure and DNA methylation in adolescence. Child Dev 2013;84(1):58-75. Used with permission.

Most recently, work by Binder and colleagues⁴¹ has revealed a molecular process by which an epidemiologically observed interaction between a functional polymorphism in the gene coding for FK506 binding protein 5 (FKBP5, an intracellular regulator of the hypothalamic-pituitary-adrenocortical (HPA) stress response system) and childhood trauma predict symptoms of post-traumatic stress disorder (PTSD) in adulthood. Specifically, this work shows that the GxE interaction effect is mediated through demethylation of DNA in the glucocorticoid response elements of FKBP5. This observation is the first to show how chromatin modification and epigenetic marks may constitute the molecular mechanism for at least some GxE interactions. The work suggests that, rather than GxE interactions and epigenomic events being two distinctive forms of gene-environment interplay, epigenetic mechanisms may underlie most or all statistically documented GxE interactions.

Gene-environment interplay in the pathogenesis of disease and maladaptive development is important because the burden of such disorders is not randomly distributed within human populations. A disproportionate share of morbidities falls upon and afflicts a small subset of individuals—usually about 15-20 percent of both children and adults—who sustain well over half of the total illness and disability.⁴² Much of this excessively burdened subpopulation is made up of those who live in conditions of poverty and, as a consequence, sustain far greater exposures to adversity and stress, as well as to environmental toxins, food insecurity and dietary constraints, and inadequate medical care. The work of Evans and colleagues⁴³ has shown how children from impoverished communities experience greater housing density, ambient noise, family chaos, and neighborhood violence than their counterparts from middle income homes. There is now strong idence that such adversity—especially early in life— is associated with disturbances of childhood mental health, more disordered developmental trajectories, poorer educational achievements, and lifelong risks of chronic disorders of health and well-being.^41,44,45 Further, beyond childhood, the studies of Anda⁴⁶ and Felitti⁴⁷ have revealed linkages between adverse childhood events and long-term risks of disorders of both mental and physical health in adulthood. Striking social class differences in incident psychiatric and biomedical disorders are believed to be attributable to lasting differences in phenotype, stemming from adversity-related perturbations in epigenetic processes.²²

Following upon the work of Meaney and colleagues demonstrating epigenetically mediated links between maternal deprivation in rat pups and lifelong stress reactivity,^11,48 human epidemiologic research by Borghol et al,²¹ and Kobor and colleagues^22,23 has detected longitudinal associations between childhood disadvantage and genome-wide promoter methylation in mid-life and between early SES and inflammatory gene expression in leukocyte transcriptomes. Prenatal exposures to famine and adversity during the 19^44-45 Dutch Hunger Winter have also been associated with differential methylation in a variety of other developmentally and immunologically active genes.^49,50 Similarly, institutionalized children from the former Russian Federation have been reported to show whole genome hypermethylation, compared to parent-reared controls.⁵¹ Oberlander and colleagues⁵² reported increased methylation of the glucocorticoid receptor (GR) gene in infants born to mothers with prenatal depression, and Radtke et al,⁵³ derived similar findings from the leukocytes of adolescents whose mothers were exposed to intimate partner violence during pregnancy. By decreasing GR expression through epigenetic marks, the adrenocortical stress response system may be perennially up-regulated, resulting in maladaptive response to stress and trauma. Such findings, of both whole genome and gene-specific differences in DNA methylation status, are consistent with the cellular effects of stress hormones, such as glucocorticoids.^24,54 Taken together, these observations offer substantial evidence for an extensive interplay among genes, epigenomes, and social environments in the genesis of perturbed or problematic social and biological development.

Individual Differences in Biological Responses to Adversity

Biological and psychological responses to early life stressors are not uniform across individuals, and there is abundant and emerging evidence for substantial individual differences in susceptibility to environmental influences, both positive and negative. The collective work of Boyce and Ellis,^55,56 Belsky,⁵⁷ and van IJzendoorn and colleagues58 has documented, at levels of analysis ranging from genetic variation and physiological stress reactivity to temperamental characteristics and individual behavior, the presence of childhood subgroups, usually comprising approximately one in five children, who sustain either the worst or best developmental and health outcomes, conditional upon the character of the social environments in which they are being reared. Children with high levels of laboratory-based, autonomic nervous system reactivity to challenge show either the highest or lowest levels of externalizing behavior problems, depending on their exposure to significant marital conflict within their families,⁵⁹ and Bush et al,⁶⁰ have recently shown that the BDNF Val66Met polymorphism confers an increased neuroendocrine sensitivity to socioeconomic context, with Met-carriers having the highest or lowest cortisol expression levels, depending on SES. This differential susceptibility to environmental influences has been demonstrated in both non-human primates^35,61 and human children⁶² and detected at the levels of allelic variation,^59,63 autonomic and adrenocortical stress responsivity,^64,65 and individual differences in temperament and behavior.⁵⁶

Differential susceptibility has also been shown to occur in relation to epigenetic marks and chromatin modifications. Beach et al,⁶⁶ in a sample of African American youth from working poor communities, found that cumulative socioeconomic adversity and the S- allele of the 5HTT, serotonin transporter gene interactively predicted promoter region methylation within a group of 200+ depression-related genes. Youth with the S-allele had either the highest or lowest levels of depression-related gene methylation, depending on the intensity of exposures to poverty-related stress. Strunk and colleagues⁶⁷ have argued that the increased susceptibility of infants to infectious disease agents may be due to differential methylation of immune-regulating and other developmentally salient genes. Further, the Binder laboratory⁴⁰ has demonstrated a differential susceptibility of individuals bearing the AG/AA "risk" allele of the FKBP5 gene (Figure 3). Such individuals have either the highest or lowest rates of adult PTSD, conditional upon childhood exposures to sexual and/or physical abuse. Further, the molecular process by which this epidemiologically-observed interaction occurs is mediated through DNA demethylation in the glucocorticoid response elements of FKBP5. These observations are among the first to show how chromatin modification and epigenetic marks might constitute the actual molecular mechanisms for a differential susceptibility to environmental conditions.

Figure 3: The interaction of early trauma exposure and FKBP5 risk allele predicting lifetime post-traumatic stress disorder (PTSD)

Source: Modified from Klengel T, Mehta D, Anacker C, et al. Allele-specific FKBP5 DNA demethylation mediates gene-childhood trauma interactions. Nat Neurosci 2013;16(1):33-41. Used with permission.
Note: Consistent with the differential susceptibility hypothesis, the "risk" allele is protective in the absence of trauma.

Critical Periods in the Epigenetic Regulation of Neurodevelopment

These epigenetically-mediated differences in sensitivity to early environments also emerge within a temporal framework. For example, unilateral blindness occurs if an occlusion of vision in one eye (due to strabismus, cataracts, or other causes) is not corrected during the early period of visual cortical development (i.e., birth to 6 or 7 years of age). This constitutes a so-called critical period, defined as a developmental interval in which the presence or absence of important experiences or exposures result in irreversible change in brain circuitry.⁶⁸ Critical periods have been described in the Bucharest Early Intervention Project, in which institutionalized Romanian children were randomized to either orphanage care-as-usual or placement into foster care.⁶⁹ As shown in Figure 4, teacher-rated social skills at 8 years of age were critically influenced by whether foster placement occurred before or after the age of 20 months; foster care children placed before 20 months had skills comparable to their never-institutionalized peers, while those placed after 20 months had the same deficits as children who remained in orphanages.⁷⁰

Figure 4: Teacher-rated social skills of Romanian orphans and never-institutionalized (NIG) Romanian children at 8 years of age, by assignment to a care as usual group (CAUG) versus a foster care group (FCG) before and later than 20 months of age

Source: Almas A, Degnan K, Radulescu A, et al. The effects of early intervention and the moderating effects of brain activity on institutionalized children's social skills at age 8. Proc Nat Acad Sci U S A 2012;109(Suppl 2):17228-31. Used with permission.

There is new evidence that such critical periodicity in the acquisition of developmental skills may be mediated by epigenetic processes. Takesian and Hensch⁷¹ have shown how molecular "triggers" and "brakes" initiate and constrain plasticity in the brain over time and how the molecular events underlying critical period onset and offset are epigenetic in origin. Critical period onset appears guided and timed by the maturation of excitatory-inhibitory (E-I) circuit balance, and epigenetic factors regulate the expression of the GAD67 gene that codes for the GABA inhibitory neurotransmitter involved in E-I equilibration. Various pharmacological agents targeting epigenetic processes—drugs such as valproate, a histone deacetylase (HDAC) inhibitor—can shift the timing of critical period onset. The closure of the critical period for ocular dominance acquisition, on the other hand, involves the down-regulation of vision-dependent histone acetylation and phosphorylation. Valproate has been shown, moreover, to reopen the critical window for the acquisition of absolute pitch,⁷² and E-I circuitry imbalance and critical period timing errors have been recognized within mouse models of autism spectrum disorder.⁷³ Thus, there is at least provisional evidence that much of the molecular machinery underlying critical period onset and offset is epigenetic in origin.⁷⁴

Summary: What's Known and Unknown

Dramatic advancements in understanding and manipulating developmental events—those that catalyze the amplification or diminution of risks to mental or physical health—are often accompanied by a wholesale attribution of as yet unexplained biological enigmas to the newly discovered events. Thus, as knowledge of the epigenetic origins of key developmental processes has come into view, chromatin modifications have become the mechanism du jour for explaining and exploiting molecular biology and gene-environment interplay. In such a scientific setting, it may therefore be useful to take stock of what we now actually know and what we do not yet understand.

What We Know

A reasonable summary of what we do know, partially or in full, would include the following points:

• Socially partitioned childhood adversities have potent and pervasive linkages to health and development, with some persisting over the lifetime of the individual; these effects are at least partially mediated by epigenetic processes guiding the adaptive differential expression of adversity-responsive genes.
• The effects of stress and adversity are also highly variable, both in terms of physiological and behavioral responses and in consequent physical and mental morbidities; a subset of childhood populations shows evidence of exquisite sensitivities to the character of both aversive and supportive social environments—that is, a differential susceptibility to socioemotional environmental influence.
• The resultant, high level of individual differences in health and development is thus an interactive product of biological susceptibility and environmental conditions; since genetic variation plays a role in the acquisition of environmental sensitivity, most disorders of maladaptive development will likely be attributable to the epigenetic co-operation of genetic and environmental variation.
• Epigenetic modifications of chromatin structure thus appear, at least to some degree, to mediate biological and psychological responses to adversity, to be the basis for differences in susceptibility to context, and to underlie the inception and termination of developmentally salient, critical periods.

Such discoveries hold great potential promise for understanding the striking individual differences in developmental trajectories and health within human populations. Nonetheless, this emerging field of developmental epigenetics presently remains, at best, in its infancy.

What We Don't Know

What is not yet known about chromatin modification, its origins and consequences, far outstrips that which we know. Among the research questions still incompletely explored or completely unexplored are the following:

• What are the molecular events by which GxE interactions occur? How are experiences and exposures molecularly transduced into the chemical marks and other processes that constitute the epigenome and regulate gene transcription?
• How can dynamic epigenetic change be measured against the backdrop of systemic, tissue- specific differences in epigenetic profiles? What (if anything) can be learned about early adversity-related epigenetic modifications in hippocampal neurons from those found in buccal epithelial cells (BECs)?
• What are the timeframes for dynamic epigenetic change? Are different forms of chromatin modification more or less rapid or slow, stable or evanescent? Can epigenetic interventions modify the windows of critical periodicity for psychosocial exposures, as well as those for physical exposures?
• What epigenetic basis may there be for the inter-generational character of severe adversity effects? How are risks for psychopathology transmitted from one generation to the next, sometimes despite an absence of exposure in the latter?
• Is there an epigenetic account for the remarkable heterogeneity in disease phenotypic effects of genetic (i.e., allelic) variation? How do assemblies of common and rare genetic variants operate conjointly with environmentally mediated epigenetic change to produce risks for complex disorders of human health?

Implications for Practice

In fact, this heterogeneity in disease phenotype, environmental exposures, genetic susceptibilities, and responsivity to treatment is among the most challenging dilemmas in the medical care of human populations. Addressing such complexity is both the promise and the profound difficulty of achieving a new "precision medicine," defined as "the tailoring of medical treatment to the individual characteristics of each patient."⁷⁵ Such a vision lies notably beyond the (relatively) facile challenges of pharmacogenetics, which would select medications based on the biogenomic characteristics of an individual patient. Precision medicine, by contrast, would discover, fabricate, and select medications based not only on diagnosis and pathogenic agent, but also on sensitivities known and unknown to the patient, host allelic variants with implications for drug metabolism and clearance, and genetic features of both agent and neoplasm. A full rendition of precision medicine would also demand and create a new taxonomy of human disease, involving the molecular disaggregation of old, organ-focused categories of morbidity and the use of vast new data sets to discover currently unknown cross-disease and cross-system commonalities in mechanism.

The science of epigenetics could play a key, catalytic role in the discernment of such a new disease taxonomy. Because most disorders will ultimately be shown to have both environmental and gene- based etiologic components, examination of the molecular interface between such interactive components, i.e., the epigenome, will almost certainly become a standard diagnostic procedure. Assessment of epigenetic profiles is indeed already a standard of high level care in oncology and psychiatry, where chromatin modifications serve as biomarkers of disease states and progression.⁷⁶

The implications of using epigenetic science to effect a shift to a new disease taxonomy would be legion. The late 19th century identification of the tubercle bacillus as the pathogenic agent in tuberculosis (TB) led to an unanticipated recombination of scrofula (lymphatic TB), Pott disease (TB of the spine), pulmonary "consumption," and tuberculous meningitis—previously regarded as distinctive forms of morbidity—into a single disease category, i.e., the multiple forms and manifestations of infection with Mycobacterium tuberculosis. Similarly, insights into the epigenetic commonalities among once separate diagnoses, each with its own corresponding therapeutic approach, could shift, nearly overnight, the way that health care providers think about diagnostic categories, approach laboratory confirmation of a given diagnosis, and engage therapeutic strategies.

Implications for Research

Given the complexities that novel diagnostic and therapeutic practices would entail, the research enterprise that would guide and sustain such a shift in medical practice would require a new level of commitment to and investment in the emerging science of complexity and dynamic systems.⁷⁷ The computational models being constructed as the methodological platform of that science incorporate key features only modestly (and insufficiently) represented in current, conventional research and biostatistical methods: non-linearity of associations, multilevel causation, changes over developmental time, and interrelations among multiple, causal covariates. Given the scale and complexity of the epigenetic processes at work in human health and disease, the advent of new complex modeling approaches may be not only welcome but essential. Using such models to simulate experimental manipulations of causal factors, some of the epigenetic puzzles that might be addressable are the possible transgenerational inheritance of epigenetic marks, the origins and plasticity of critical and sensitive developmental periods, and the emergence, within populations, of individuals with either exceptional sensitivity or exceptional resilience to the privations of early life in impoverished, high adversity settings.

When viewed from the perspective of future generations of scientists and scholars, the most profound and momentous residual of this epigenetic era in the history of science may well be its dissolution of the boundary between the biological and social sciences. What began as two wholly different views of human nature has given way to a singular, new vision of how genetic and social environmental variation work together on risks for disease and maladaptive development. What began as two alternative or even antagonistic sets of methods, models, and theories—one biological and the other psychosocial in focus—have been forcibly and brilliantly reconciled by the scientific needs of each for the perspective of the other. It is almost certainly safe to claim that human biology and the social sciences will never be the same.

Acknowledgments

The research upon which this report is based was supported by the National Institute of Mental Health (grant award R24MH081797), the MacArthur Foundation Research Network on Psychopathology and Development, the Sunny Hill Health Centre/British Columbia Leadership Chair in Child Development at the University of British Columbia, and the Child and Brain Development Program of the Canadian Institute for Advanced Research. The opinions presented in this chapter are those of the author and do not necessarily represent the position of the Agency for Healthcare Research and Quality, the National Institutes of Health, or the U.S. Department of Health and Human Services.

Author's Affiliation

Departments of Pediatrics and Psychiatry, University of California, San Francisco; Canadian Institute for Advanced Research, Toronto, Canada.

Address correspondence to: W. Thomas Boyce, MD, UCSF Division of Developmental Medicine, Department of Pediatrics, 550 16th Street, 4th Floor, San Francisco, CA 94158; email: tom.boyce@ucsf.edu.

References

1. Weingart P, Mitchell SD, Richerson PJ, et al. Human by nature: between biology and the social sciences. Mahwah, NJ: Lawrence Ehrlbaum; 2013.
2. Snow CP. The two cultures. Cambridge, UK: Cambridge University Press; 1959.
3. Bailey JN, Pericak-Vance MA, Haines JL. (2014). The impact of the human genome project on complex disease. Genes (Basel) 2014;5(3):518- 35.
4. Lander ES. Initial impact of the sequencing of the human genome. Nature 2011;470(7333):187-97.
5. Gibson G. Hints of hidden heritability in GWAS. Nat Genet 2010;42(7):558-60.
6. McClellan J, King MC. Genomic analysis of mental illness: a changing landscape. JAMA 2010;303(24):2523-4.
7. Adler NE, Rehkopf DH. U.S. disparities in health: descriptions, causes, and mechanisms. Annu Rev Public Health 2008;29:235-252.
8. Berkman L, Epstein AM. Beyond health care—socioeconomic status and health. N Engl J Med 2008;358(23):2509- 10.
9. Syme SL. Historical perspective: the social determinants of disease - some roots of the movement. Epidemiol Perspect Innov 2005;2(1):2.
10. Zaitlen N, Kraft P. Heritability in the genome-wide association era. Hum Genet 2012;131(10):1655-64.
11. Meaney, MJ. Epigenetics and the biological definition of gene x environment interactions. Child Dev 2010;81(1):41-79.
12. Nelson VR, Nadeau JH. Transgenerational genetic effects. Epigenomics 2010;2(6):797- 806.
13. Waddington CH. The epidenotype. Endeavour 1942;1:18-20.
14. Bird A. Perceptions of epigenetics. Nature 2007;447(7143):396-8.
15. Sweatt JD, Meaney MJ, Nestler EJ, et al. Epigenetic regulation in the nervous system: basic mechanisms and clinical impact. London: Elsevier; 2013.
16. Allis CD, Jenuwein T, Reinberg D, et al. Epigenetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 2007.
17. Drummond EM, Gibney ER. Epigenetic regulation in obesity. Curr Opin Clin Nutr Metab Care 2013;16(4):392-7.
18. LaSalle JM. A genomic point-of-view on environmental factors influencing the human brain methylome. Epigenetics 2011;6(7):862 -9.
19. Essex MJ, Boyce WT, Hertzman C, et al. Epigenetic vestiges of early developmental adversity: childhood stress exposure and DNA methylation in adolescence. Child Dev 2013;84(1):58-75.
20. Provencal N, Suderman MJ, Guillemin C, et al. The signature of maternal rearing in the methylome in rhesus macaque prefrontal cortex and T cells. J Neurosci 2012;32(44):15626-42.
21. Tung, J., Barreiro, L. B., Johnson, Z. P., et al. Social environment is associated with gene regulatory variation in the rhesus macaque immune system. Proc Natl Acad Sci U S A 2012;109(17):6490- 5.
22. Borghol N, Suderman M, McArdle W, et al. Associations with early life socio-economic position in adult DNA methylation. Int J Epidemiol 2012;41(1):62- 74.
23. Lam LL, Emberly E, Fraser HB, et al. Factors underlying variable DNA methylation in a human community cohort. Proc Natl Acad Sci U S A 2012;109(Suppl 2):17253-60.
24. Powell ND, Sloan EK, Bailey MT, et al. Social stress up-regulates inflammatory gene expression in the leukocyte transcriptome via beta-adrenergic induction of myelopoiesis. Proc Natl Acad Sci U S A 2013;110(41):16574- 9.
25. Matthews SG, Phillips DI. Transgenerational inheritance of stress pathology. Exp Neurol 2012;233(1):95-101.
26. Unternaehrer E, Luers P, Mill J, et al. Dynamic changes in DNA methylation of stress-associated genes (OXTR, BDNF) after acute psychosocial stress. Transl Psychiatry 2012;2:e150.
27. Kaliman P, Alvarez-Lopez MJ, Cosin-Tomas M, et al. Rapid changes in histone deacetylases and inflammatory gene expression in expert meditators. Psychoneuroendocrinology 2014;40:96- 107.
28. Zannas AS, West AE. Epigenetics and the regulation of stress vulnerability and resilience. Neuroscience 2014;264:157-70.
29. Sweatt JD. Experience-dependent epigenetic modifications in the central nervous system. Biol Psychiatry 2009;65(3):191 -7.
30. Caspi A, McClay J, Moffitt TE, et al. Role of genotype in the cycle of violence in maltreated children. Science, 2002;297(5582):851-4.
31. Caspi A, Sugden K, Moffitt TE, et al. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003;301(5631):386-9.
32. Duncan LE, Keller MC. A critical review of the first 10 years of candidate gene-by-environment interaction research in psychiatry. Am J Psychiatry 2011;168(10):1041-9.
33. van Winkel M, Peeters F, van Winkel R, et al. Impact of variation in the BDNF gene on social stress sensitivity and the buffering impact of positive emotions: replication and extension of a gene- environment interaction. Eur Neuropsychopharmacol 2014;24 (6):930-8.
34. Drury SS, Theall KP, Smyke AT, et al. Modification of depression by COMT val158met polymorphism in children exposed to early severe psychosocial deprivation. Child Abuse Negl 2010;34(6):387-95.
35. Zohsel K, Buchmann AF, Blomeyer D, et al. Mothers' prenatal stress and their children's antisocial outcomes–a moderating role for the Dopamine D4 Receptor (DRD4) gene. J Child Psychol Psychiatry 2014;55(1):69-76.
36. Barr CS, Newman TK, Lindell S, et al. Interaction between serotonin transporter gene variation and rearing condition in alcohol preference and consumption in female primates. Arch Gen Psychiatry 2004;61(11):1146- 52.
37. Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health, 2005;95(Suppl 1):S144-50.
38. Rietveld CA, Medland SE, Derringer J, et al. GWAS of 126,559 individuals identifies genetic variants associated with educational attainment. Science 2013;340(6139):1467-71.
39. Marigorta UM, Gibson G. A simulation study of gene- by-environment interactions in GWAS implies ample hidden effects. Front Genet 2014;5:225.
40. Gratten J, Wray NR, Keller MC, et al. Large- scale genomics unveils the genetic architecture of psychiatric disorders. Nat Neurosci 2014;17(6):782-90.
41. Klengel T, Mehta D, Anacker C, et al. Allele-specific FKBP5 DNA demethylation mediates gene-childhood trauma interactions. Nat Neurosci 2013;16 (1):33-41.
42. Hertzman C, Boyce WT. (2010). How experience gets under the skin to create gradients in developmental health. Annu Rev Public Health 31:329-47 3p following 347.
43. Evans GW, Chen E, Miller G, et al. How poverty gets under the skin: a life-course perspective. In Maholmes V, King RB (Eds), The Oxford handbook of poverty and child development (pp. 13-36). Oxford: Oxford University Press; 2012.
44. Boyce WT, Sokolowski MB, Robinson GE. Toward a new biology of social adversity. Proc Natl Acad Sci U S A 2012;109(Suppl 2):17143-8.
45. Shonkoff JP, Boyce WT, McEwen BS. Neuroscience, molecular biology, and the childhood roots of health disparities: building a new framework for health promotion and disease prevention. JAMA 2009;301 (21):2252-9.
46. Anda RF, Croft JB, Felitt, VJ, et al. Adverse childhood experiences and smoking during adolescence and adulthood. JAMA 1999;282(17):1652- 8.
47. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med 1998;14(4):245-58.
48. Weaver IC, Meaney MJ, Szyf M. Maternal care effects on the hippocampal transcriptome and anxiety- mediated behaviors in the offspring that are reversible in adulthood. Proc Natl Acad Sci U S A 2006;103(9):3480- 5.
49. Heijmans BT, Tobi EW, Stein AD, et al. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A 2008;105(44):17046-9.
50. Tobi EW, Lumey LH, Talens RP, et al. DNA methylation differences after exposure to prenatal famine are common and timing- and sex-specific. Hum Mol Genet 2009;18(21):4046-53.
51. Naumova OY, Lee M, Koposov R, et al. Differential patterns of whole-genome DNA methylation in institutionalized children and children raised by their biological parents. Dev Psychopathol 2012;24(1):143- 55.
52. Oberlander TF, Weinberg J, Papsdorf M, et al. Prenatal exposure to maternal depression, neonatal methylation of human glucocorticoid receptor gene (NR3C1) and infant cortisol stress response. Epigenetics 2008;3(2):97- 106.
53. Radtke KM, Ruf M, Gunter HM, et al. Transgenerational impact of intimate partner violence on methylation in the promoter of the glucocorticoid receptor. Transl Psychiatry 2011;1:e21.
54. Hostinar CE, Sullivan RM, Gunnar MR. Psychobiological mechanisms underlying the social buffering of the hypothalamic-pituitary-adrenocortical axis: a review of animal models and human studies across development. Psychol Bull 2014;140(1):256-82.
55. Boyce WT, Ellis BJ. Biological sensitivity to context. I. An evolutionary-developmental theory of the origins and functions of stress reactivity. Dev Psychopathol 2005;17(2):271-301.
56. Ellis BJ, Essex MJ, Boyce WT. Biological sensitivity to context. II. Empirical explorations of an evolutionary-developmental hypothesis. Dev Psychopathol 2005;17(2):303-28.
57. Belsky J. Differential susceptibility to rearing influence: an evolutionary hypothesis and some evidence. In Ellis BJ, Bjorklund DF (Eds), Origins of the social mind: evolutionary psychology and child development (pp. 139-163). New York: Guilford; 2005.
58. Bakermans-Kranenburg MJ, van Ijzendoorn MH. (2011). Differential susceptibility to rearing environment depending on dopamine-related genes: new evidence and a meta-analysis. Dev Psychopathol 2011;23 (01):39-52.
59. Obradovic J, Bush NR, Boyce WT. The interactive effect of marital conflict and stress reactivity on externalizing and internalizing symptoms: the role of laboratory stressors. Dev Psychopathol 2011;23(1):101-14.
60. Bush N, Guendelman M, Adler N, et al. BDNF allelic variants moderate social disparities in children's basal cortisol expression. 2014; submitted.
61. Boyce WT, O'Neill-Wagner P, Price CS, et al. Crowding stress and violent injuries among behaviorally inhibited rhesus macaques. Health Psychol 1998;17(3):285-9.
62. Ellis BJ, Boyce WT, Belsky J, et al. Differential susceptibility to the environment: an evolutionary–neurodevelopmental theory. Dev Psychopathol 2011;23(1):7-28.
63. Guerra S, Martinez FD. Asthma genetics: from linear to multifactoril approaches. Annu Rev Med 2008;59:327- 41
64. Alkon A, Boyce WT, Davis NV, et al. Developmental changes in autonomic nervous system resting and reactivity measures in Latino children from 6 to 60 months of age. J Dev Behav Pediatr 2011;32(9):668- 77.
65. Manuck SB, Craig AE, Flory JD, et al. Reported early family environment covaries with menarcheal age as a function of polymorphic variation in estrogen receptor-alpha. Dev Psychopathol 2011;23(1):69- 83.
66. Beach SR, Brody GH, Lei MK, et al. Is serotonin transporter genotype associated with epigenetic susceptibility or vulnerability? Examination of the impact of socioeconomic status risk on African American youth. Dev Psychopathol 2014;26 (2):289- 304.
67. Strunk T, Jamieson SE, Burgner D. Genetic and epigenetic susceptibility to early life infection. Curr Opin Infect Dis 2013;26(3):241- 7.
68. Fox SE, Levitt P, Nelson CA. How the timing and quality of early experiences influence the development of brain architecture. Child Dev 2010;81(1):28- 40.
69. Nelson CA, Fox NA, Zeanah C. Romania's abandoned children. Cambridge, MA: Harvard University Press; 2014.
70. Almas A, Degnan K, Radulescu A, et al. The effects of early intervention and the moderating effects of brain activity on institutionalized children's social skills at age 8. Proc Nat Acad Sci U S A 2012;109 (Suppl 2):17228-31.
71. Takesian AE, Hensch TK. Balancing plasticity/ stability across brain development. Prog Brain Res 2013;207:3-34.
72. Gervain J, Vines BW, Chen LM, et al. Valproate reopens critical-period learning of absolute pitch. Front Syst Neurosci 2013;7:102.
73. Gogolla N, Leblanc JJ, Quast KB, et al. Common circuit defect of excitatory-inhibitory balance in mouse models of autism. J Neurodev Disord 2009;1(2):172-81.
74. Fagiolini M, Jensen CL, Champagne FA. Epigenetic influences on brain development and plasticity. Curr Opin Neurobiol 2009;19(2):207- 12.
75. National Research Council, Committee on a Framework for Development a New Taxonomy of Disease. Toward precision medicine: building a knowledge network for biomedical research and a new taxonomy of disease. Washington, DC: National Academies Press; 2011.
76. Brookes E, Shi Y. Diverse epigenetic mechanisms of human disease. Annu Rev Genet 2014;48:237-68.
77. Galea S, Riddle M, Kaplan GA. Causal thinking and complex system approaches in epidemiology. Int J Epidemiol 2010;39(1):97-106.

W. Thomas Boyce, MD, is Distinguished Professor of Pediatrics and Psychiatry and heads the Division of Developmental Medicine in the Department of Pediatrics, University of California, San Francisco. Previously, he was Professor of Pediatrics and the Sunny Hill Health Centre-BC Leadership Chair in Child Development at the University of British Columbia, in the Human Early Learning Partnership, and at the Child and Family Research Institute of BC Children's Hospital. Dr. Boyce has served as a member of Harvard University's National Scientific Council on the Developing Child, UC Berkeley's Institute of Human Development, and as a founding co-Director of the Robert Wood Johnson Foundation Health & Society Scholars Program at Berkeley and UCSF. He co-directs the Child and Brain Development Program for the Canadian Institute for Advanced Research and serves on the Board on Children, Youth, and Families of the National Academy of Sciences. He was elected in 2011 to the Institute of Medicine.