Table 3.
Childhood adversity and studies leveraging epigenome-wide methylation arrays in samples of children.
| Author | Sample | Age(s) at methylation assessment | Gender | Ancestry | Tissue type | Adversity measurement | Adversity-related findings |
|---|---|---|---|---|---|---|---|
| Cecil et al.,166 |
N = 124; 68% exposed to CM Characterization of sample: Inner city youth in London |
16–24 years | 53% female | 49% white, 33% black, 18% other | Buccal cells | Physical, sexual, emotional abuse, emotional and physical neglect via the CTQ | Physical and sexual abuse, and physical neglect associated with epigenetic variation; no significant association between emotional abuse and neglect and epigenetic variation; top loci (e.g., GABBR1, GRIN2D, CACNA2D4, PSEN2) include markers implicated in stress-sensitive outcomes; gene ontology supports some common epigenetic signatures related to growth and neural development |
| Cicchetti et al.,78 |
N = 548; 54% exposed to CM Characterization of sample: Low-income |
School age (M = 9 years) | 48% female |
68% black, 21% white, 12% biracial or other; 21% Latino |
Saliva | Physical, sexual, emotional abuse, supervisory and physical neglect via child protection case files coded with the MCS | Maltreated children had higher levels of methylation at sites where methylation was generally low, and lower methylation at sites where methylation generally high, compared to non-maltreated children; Additional site-specific analyses using a candidate approach examined AKDH2, ANKK1, and NR3C1; maltreated girls had lower methylation of ALDH2 than non-maltreated girls, and maltreated boys had higher ALDH2 methylation than non-maltreated boys; boys with early, but not recent, maltreatment had higher methylation of ALDH2 than non-maltreated boys; early maltreated children also had higher ANKK1 methylation than non-maltreated children |
| Dunn et al.,79 |
N = 774; 67% exposed to adversity Characterization of sample: Enrolled in ARIES study nested within ALSPAC |
7 years | 50% female | Not reported | Cord blood at birth and peripheral blood at age 7 | Physical, sexual, emotional abuse; maternal psychopathology; single parent; family instability; financial stress/poverty; neighborhood disadvantage measured via maternal report from a single item or from psychometrically validated standardized measures | Nearly all types of adversity exposure in early childhood (age 3 or younger) associated with methylation differences at age 7; only physical and sexual abuse in middle childhood associated with methylation differences at age 7; accumulation and recency of adversity did not explain variance in methylation; findings supported gene ontology pathways involving growth, axon development, and neuron apoptotic processes |
| Esposito et al.,167 | N = 83; 60% institutionalized in Eastern Europe or Russia and adopted in the U.S.; 40% raised with biological families in the U.S. | 12–18 years in adopted group (M = 15.7 years); 13–17 years in non-adopted group (M = 15.4 years) | 52% female | 100% European descent | Peripheral blood mononuclear cells | 46 stressful life events within the past year via the Life Events Checklist—Child/Adolescent version; history of institutionalization | Children adopted from institutions of abandoned or orphaned children in Eastern Europe or Russia had higher levels of methylation compared to children of similar ancestry raised by biological parents in the United States; although after correction for cell type and multiple covariates, adoption groups were not associated with methylation, observed leftward skew supported a separate step of filtration to remove sites with no variation and genes with differentially methylated sites included TMEM200C, PPP1R3G, GLYATL2, CYP1A1, and miR-324.; however differential methylation in CYP1A1 may be due to adopted children having had higher early life exposures to cigarette smoke, or current smoking |
| Jovanovic et al.,81 | N = 101 |
6–13 years (M = 9.7 years) |
55% female | 100% African American | Saliva | 9 items capturing direct violence exposure and 13 items measuring witnessing violent events via the Violence Exposure Scale for Children-Revised, a cartoon-based self-report interview of children’s lifetime exposure to violence | Exposure to direct violence was positively associated with epigenetic age acceleration (the residual between DNA methylation age and chronological age); the group with the most age acceleration had the most trauma |
| Kumsta et al.,168 |
N = 49; n = 16 with extended institutional deprivation; n = 17 with limited institutional deprivation; n = 16 control with no institutional deprivation Characterization of sample: Enrolled in ERA study; children adopted from Romania into families residing in England |
11–15 years | Not reported | 100% Romanian in institutional deprivation groups; 100% UK nationals in control group | Buccal cells | Duration of institutional deprivation categorized as none, less than, or greater than six months | Children exposed to more than 6 months in Romanian institutions (extended institutional deprivation) showed elevated methylation; methylation of early adopted Romanian children (limited institutional deprivation) was not significantly different from control group with no institutional deprivation |
| Marini et al.,169 | N = 973; 13% exposed to physical abuse, 49% exposed to family instability Characterization of sample: Enrolled in ALSPAC | 7.5 years | 50% female | 97% white | Whole blood | Physical, sexual, emotional abuse; maternal psychopathology; single parent; family instability; financial stress/poverty; neighborhood disadvantage measured via maternal report from a single item or from psychometrically validated standardized measures | Financial hardship associated with epigenetic age acceleration; using Hannum’s epigenetic clock, sexual or physical abuse at 3.5 years associated with older epigenetic age, financial hardship and neighborhood disadvantage at 7 years associated with acceleration in epigenetic aging; no associations emerged using Horvath’s epigenetic clock |
| Naumova et al.,170 |
N = 28; n = 14 raised since birth in institutional care, n = 14 raised by biological parents Characterization of sample: Participants resided in northwest region of the Russian Federation |
7–10 years (M = 8.1 years in institutionalized group; M = 8.4 years in control group) | 36% female in institutionalized group; 29% female in control group | Predominantly Slavic | Whole blood |
General trauma, physical punishment, emotional and sexual abuse prior to age 18 assessed using the Early Trauma Inventory Self Report-Short Form |
The institutionalized group demonstrated overall proportionally greater methylation relative to the controls; analysis of pathway enrichment supported enrichment of the upmethylated (increased methylation) genomes of institutionalized children in regions related to cellular signaling and immune response; functional annotation implicated methylation profile differences in genes related to brain function (including genes in the dopaminergic system, glucocorticoid and steroid biosynthesis, serotonergic system, etc.) |
| Papale et al.,171 | N = 22; 50% exposed to high stress |
9–12 years (M = 10.9 years) |
100% female | 50% Caucasian | Saliva | Life stress measured by the Youth Life Stress Interview | High stress exposure was associated with variability in methylation |
| Sumner et al.,80 |
N = 247 Characterization of sample: 27% below poverty line |
8–16 years (M = 12.7 years) |
48% female | 39% white, 28% black, 21% other, 12% Hispanic | Saliva | Physical, sexual, and emotional abuse; domestic violence exposure; exposure to other interpersonal violence; emotional neglect; food insecurity via child interview and self-report measures | Exposure to threat associated with accelerated DNA methylation age and advanced pubertal stage, but exposure to deprivation was not; threat exposure affected depressive symptoms through DNA methylation age |
| Weder et al.,172 | N = 190; 50% exposed to CM | 5–14 years (M = 10.2 years) | 58% female | 17% European American, 38% Hispanic, 30% African American, 15% biracial | Saliva | CM assessed via child protection records, parent and child reports of trauma using the KSADS interview, child reports on the CTQ, maternal reports of domestic violence on the Partner Violence Inventory | Differences in methylation between children with CM and those with no CM were observed in methylation sites in the regions of BDNF, FKBP5, NR3C1 |
| Yang et al.,173 | N = 192; 50% exposed to CM | 5–14 years (M = 10.2 years) | 58% female | 17% European American, 38% Hispanic, 30% African American, 15% biracial | Saliva | CM assessed via child protection records, parent and child reports of trauma using the KSADS interview, child reports on the CTQ, maternal reports of domestic violence on the Partner Violence Inventory | Differential methylation at 2868 CpG sites between children with CM and those with no CM; children with CM had higher methylation at CpG sites with low to mid-range methylation, and lower methylation at sites with high methylation, than children with no CM |
CM childhood maltreatment, CTQ Childhood Trauma Questionnaire, MCS Maltreatment Classification System, ALSPAC Avon Longitudinal Study of Parents and Children, KSADS Kiddie-SADS—Lifetime Version.