Pepper & Nettle (P&N) have delivered an important contribution to the field by hypothesising that people exhibit behaviourally adaptive responses to deprivation. Further, we are pleased that the authors acknowledge that DNA methylation may play a role in the behavioural constellation of deprivation (BCD). Caution is needed, as most human behavioural epigenetics research (among healthy individuals) is correlational or blood- or saliva-based and therefore does not permit an understanding of the underlying functional molecular physiology. Inflammation is only one logical possibility, as suggested by P&N, but epigenetic profiles involved in the BCD likely vary by age and tissue type. For example, DNA methylation, a well-studied epigenetic alteration, varies by age, sex, tissue, cell type, and isoform, making blood-based correlational studies potentially misleading. Work by Horvath (Reference Horvath2013) on the epigenetic clock indicates that, regardless of tissue type, chronical age can be predicted with high accuracy from DNA methylation profiles. In the case of pathological tissue (e.g., cancerous tumours), epigenetic age is advanced relative to healthy tissue (Horvath Reference Horvath2013; Reference Horvath2015). We argue that the precision of the epigenetic clock indicates specific life-history adaptations that are system-wide. In the case of impoverished environments, we suspect from an evolutionary and epigenetic perspective that adaptive responses to stress by young people (e.g., present-oriented behaviours) should be qualitatively distinct from the adaptive responses to stress by older people (e.g., future-oriented behaviours, such as altruism and giving social support). Indeed, we hypothesise that if older people were to maintain a BCD strategy suggested by P&N, they would be at an inclusive fitness disadvantage compared to those who employ a life-history-dependent shift (e.g., caring for grandchildren, community-based altruism). From an epigenetic-profiling perspective, we expect a shift in profiles as we age to correlate with changes in the behavioural adaptive responses to deprivation. For example, the behavioural strategy that works best for a 21-year-old facing deprivation would be costly for a 60-year-old. There is an important analogy to be made between the emphasis on mating for younger people and the shift toward parenting among older people. Brown (Reference Brown2015) suggests that genes important for growth (e.g., development of secondary sexual characteristics, such as big muscles in young men) could become costly later in life (e.g., development of cancers). Such effects are consistent with antagonistic pleiotropy perspectives on senescence, the proposal by Williams (Reference Williams1957) that a gene controlling one trait could be beneficial to the organism's fitness at younger ages and detrimental to the organism's fitness later in life. We believe it is important for P&N to extend the BCD model to incorporate epigenetic adaptive responses to antagonistic pleiotropy, which may be more marked in deprived environments.
A second related issue raised by P&N is the importance of studying pathological responses to deprivation. Specifically, they propose in the target article abstract that they “emphasise the idea that the present-oriented behaviours of the constellation are a contextually appropriate response to structural and ecological factors rather than a pathology.” Experimental work on the epigenetics of stress in humans and other organisms suggests an inverted U or J relation between stress and adaptive responses (Bernal et al. Reference Bernal, Dolinoy, Huang, Skaar, Weinhouse and Jirtle2013; Park et al. Reference Park, Skaar, Jirtle and Hoyo2017). This nonlinear relation between stress and adaptive responses can be called a “hormesis effect” (Chalk & Brown Reference Chalk and Brown2014). In the case of deprivation and epigenetics, evidence suggests that early exposure to extreme stress changes our epigenome in ways that place us at increased risk of disease. Swartz et al. (Reference Swartz, Hariri and Williamson2017) show how DNA methylation is related to risk-linked amygdala activity, which would be adaptive in harsh environments, a finding that is consistent with P&N. However, at extreme levels, these epigenetic changes may lead to depression or possibly suicide, which would likely be maladaptive, especially among older people, where community-based altruism may be beneficial to health in impoverished or low SES communities (Brown et al. Reference Brown, Consedine and Magai2005; Martinez et al. Reference Martinez, Frick, Glass, Carlson, Tanner, Ricks and Fried2006).
In conclusion, present-oriented behaviours are adaptive in younger people for reproduction purposes; however, from an inclusive fitness perspective, grandparental investment is beneficial at older ages. This may help explain why even in older, deprived communities, altruism is positively correlated with health (Brown et al. Reference Brown, Consedine and Magai2005). Extreme present-oriented selfish behaviours may be strongly selected against among older individuals in deprived communities due to inclusive fitness losses or the costs of competing against younger individuals. Finally, experimental work on epigenetics in rodents (Bernal et al. Reference Bernal, Dolinoy, Huang, Skaar, Weinhouse and Jirtle2013; Waterland & Jirtle Reference Waterland and Jirtle2003) clearly shows that high levels of stress in utero cause negative outcomes, whilst moderate levels of stress are beneficial. We do believe that a molecular epigenetic approach provides a useful extension to P&N's BCD approach. Age-dependent BCD can be tested using measures of developmental disruption at the molecular epigenetic (DNA methylation dysregulation as a barometer of stress, e.g., Horvath Reference Horvath2013; Park et al. Reference Park, Skaar, Jirtle and Hoyo2017; Waterland & Jirtle Reference Waterland and Jirtle2003) and behavioural levels of analysis (e.g., prolonged sociopathy in older people). Manifestations of present-oriented behaviours in older people from deprived communities would be an example of pathology. In contrast, future-oriented altruism and giving social support (e.g., community volunteerism, especially among older people in impoverished communities; Brown et al. Reference Brown, Consedine and Magai2005) are expected to be more viable condition-dependent strategies relative to present-oriented behaviours.
Pepper & Nettle (P&N) have delivered an important contribution to the field by hypothesising that people exhibit behaviourally adaptive responses to deprivation. Further, we are pleased that the authors acknowledge that DNA methylation may play a role in the behavioural constellation of deprivation (BCD). Caution is needed, as most human behavioural epigenetics research (among healthy individuals) is correlational or blood- or saliva-based and therefore does not permit an understanding of the underlying functional molecular physiology. Inflammation is only one logical possibility, as suggested by P&N, but epigenetic profiles involved in the BCD likely vary by age and tissue type. For example, DNA methylation, a well-studied epigenetic alteration, varies by age, sex, tissue, cell type, and isoform, making blood-based correlational studies potentially misleading. Work by Horvath (Reference Horvath2013) on the epigenetic clock indicates that, regardless of tissue type, chronical age can be predicted with high accuracy from DNA methylation profiles. In the case of pathological tissue (e.g., cancerous tumours), epigenetic age is advanced relative to healthy tissue (Horvath Reference Horvath2013; Reference Horvath2015). We argue that the precision of the epigenetic clock indicates specific life-history adaptations that are system-wide. In the case of impoverished environments, we suspect from an evolutionary and epigenetic perspective that adaptive responses to stress by young people (e.g., present-oriented behaviours) should be qualitatively distinct from the adaptive responses to stress by older people (e.g., future-oriented behaviours, such as altruism and giving social support). Indeed, we hypothesise that if older people were to maintain a BCD strategy suggested by P&N, they would be at an inclusive fitness disadvantage compared to those who employ a life-history-dependent shift (e.g., caring for grandchildren, community-based altruism). From an epigenetic-profiling perspective, we expect a shift in profiles as we age to correlate with changes in the behavioural adaptive responses to deprivation. For example, the behavioural strategy that works best for a 21-year-old facing deprivation would be costly for a 60-year-old. There is an important analogy to be made between the emphasis on mating for younger people and the shift toward parenting among older people. Brown (Reference Brown2015) suggests that genes important for growth (e.g., development of secondary sexual characteristics, such as big muscles in young men) could become costly later in life (e.g., development of cancers). Such effects are consistent with antagonistic pleiotropy perspectives on senescence, the proposal by Williams (Reference Williams1957) that a gene controlling one trait could be beneficial to the organism's fitness at younger ages and detrimental to the organism's fitness later in life. We believe it is important for P&N to extend the BCD model to incorporate epigenetic adaptive responses to antagonistic pleiotropy, which may be more marked in deprived environments.
A second related issue raised by P&N is the importance of studying pathological responses to deprivation. Specifically, they propose in the target article abstract that they “emphasise the idea that the present-oriented behaviours of the constellation are a contextually appropriate response to structural and ecological factors rather than a pathology.” Experimental work on the epigenetics of stress in humans and other organisms suggests an inverted U or J relation between stress and adaptive responses (Bernal et al. Reference Bernal, Dolinoy, Huang, Skaar, Weinhouse and Jirtle2013; Park et al. Reference Park, Skaar, Jirtle and Hoyo2017). This nonlinear relation between stress and adaptive responses can be called a “hormesis effect” (Chalk & Brown Reference Chalk and Brown2014). In the case of deprivation and epigenetics, evidence suggests that early exposure to extreme stress changes our epigenome in ways that place us at increased risk of disease. Swartz et al. (Reference Swartz, Hariri and Williamson2017) show how DNA methylation is related to risk-linked amygdala activity, which would be adaptive in harsh environments, a finding that is consistent with P&N. However, at extreme levels, these epigenetic changes may lead to depression or possibly suicide, which would likely be maladaptive, especially among older people, where community-based altruism may be beneficial to health in impoverished or low SES communities (Brown et al. Reference Brown, Consedine and Magai2005; Martinez et al. Reference Martinez, Frick, Glass, Carlson, Tanner, Ricks and Fried2006).
In conclusion, present-oriented behaviours are adaptive in younger people for reproduction purposes; however, from an inclusive fitness perspective, grandparental investment is beneficial at older ages. This may help explain why even in older, deprived communities, altruism is positively correlated with health (Brown et al. Reference Brown, Consedine and Magai2005). Extreme present-oriented selfish behaviours may be strongly selected against among older individuals in deprived communities due to inclusive fitness losses or the costs of competing against younger individuals. Finally, experimental work on epigenetics in rodents (Bernal et al. Reference Bernal, Dolinoy, Huang, Skaar, Weinhouse and Jirtle2013; Waterland & Jirtle Reference Waterland and Jirtle2003) clearly shows that high levels of stress in utero cause negative outcomes, whilst moderate levels of stress are beneficial. We do believe that a molecular epigenetic approach provides a useful extension to P&N's BCD approach. Age-dependent BCD can be tested using measures of developmental disruption at the molecular epigenetic (DNA methylation dysregulation as a barometer of stress, e.g., Horvath Reference Horvath2013; Park et al. Reference Park, Skaar, Jirtle and Hoyo2017; Waterland & Jirtle Reference Waterland and Jirtle2003) and behavioural levels of analysis (e.g., prolonged sociopathy in older people). Manifestations of present-oriented behaviours in older people from deprived communities would be an example of pathology. In contrast, future-oriented altruism and giving social support (e.g., community volunteerism, especially among older people in impoverished communities; Brown et al. Reference Brown, Consedine and Magai2005) are expected to be more viable condition-dependent strategies relative to present-oriented behaviours.