Early life adversity (ELA) has been associated with an increased risk for diseases in which the immune system plays a critical role. The ELA immune phenotype is characterized by inflammation, impaired cellular immunity, and immunosenescence. However, data on cell-specific immune effects are largely absent. Additionally, stress systems and health behaviors are altered in ELA, which may contribute to the generation of the ELA immune phenotype. The present investigation tested cell-specific immune differences in relationship to the ELA immune phenotype, altered stress parameters, and health behaviors in individuals with ELA ( = 42) and those without a history of ELA (control, = 73). Relative number and activation status (CD25, CD69, HLA-DR, CD11a, CD11b) of monocytes, NK cells, B cells, T cells, and their main subsets were assessed by flow cytometry. ELA was associated with significantly reduced numbers of CD69CD8 T cells ( = 0.022), increased numbers of HLA-DR CD4 and HLA-DR CD8 T cells ( < 0.001), as well as increased numbers of CD25CD8 T cells ( = 0.036). ELA also showed a trend toward higher numbers of CCR4CXCR3CCR6 CD4 T cells. Taken together, our data suggest an elevated state of immune activation in ELA, in which particularly T cells are affected. Although several aspects of the ELA immune phenotype were related to increased activation markers, neither stress nor health-risk behaviors explained the observed group differences. Thus, the state of immune activation in ELA does not seem to be secondary to alterations in the stress system or health-risk behaviors, but rather a primary effect of early life programming on immune cells.

Early life adversity (ELA) increases the risk for multiple age-related diseases, such as diabetes type 2 and cardiovascular disease. As prevalence is high, ELA poses a major and global public health problem. Immunosenescence, or aging of the immune system, has been proposed to underlie the association between ELA and long-term health consequences. However, it is unclear what drives ELA-associated immunosenescence and which cells are primarily affected. We investigated different biomarkers of immunosenescence in a healthy subset of the EpiPath cohort. Participants were either parent-reared (Ctrl, n = 59) or had experienced separation from their parents in early childhood and were subsequently adopted (ELA, n = 18). No difference was observed in telomere length or in methylation levels of age-related CpGs in whole blood, containing a heterogeneous mixture of immune cells. However, when specifically investigating T cells, we found a higher expression of senescence markers (CD57) in ELA. In addition, senescent T cells (CD57+) in ELA had an increased cytolytic potential compared to senescent cells in controls. With a mediation analysis we demonstrated that cytomegalovirus (CMV) infection, which is an important driving force of immunosenescence, largely accounted for elevated CD57 expression observed in ELA. Leukocyte telomere length may obscure cell-specific immunosenescence; here, we demonstrated that the use of cell surface markers of senescence can be more informative. Our data suggest that ELA may increase the risk of CMV infection in early childhood, thereby mediating the effect of ELA on T cell-specific immunosenescence. Thus, future studies should include CMV as a confounder or selectively investigate CMV seronegative cohorts.

Early life adversity (ELA) has been associated with inflammation and immunosenescence, as well as hyporeactivity of the HPA axis. Because the immune system and the HPA axis are tightly intertwined around the glucocorticoid receptor (GR), we examined peripheral GR functionality in the EpiPath cohort among participants who either had been exposed to ELA (separation from parents and/or institutionalization followed by adoption; n = 40) or had been reared by their biological parents (n = 72). Expression of the strict GR target genes FKBP5 and GILZ as well as total and 1F and 1H GR transcripts were similar between groups. Furthermore, there were no differences in GR sensitivity, examined by the effects of dexamethasone on IL6 production in LPS-stimulated whole blood. Although we did not find differences in methylation at the GR 1F exon or promoter region, we identified a region of the GR 1H promoter (CpG 1-9) that showed lower methylation levels in ELA. Our results suggest that peripheral GR signaling was unperturbed in our cohort and the observed immune phenotype does not appear to be secondary to an altered GR response to the perturbed HPA axis and glucocorticoid (GC) profile, although we are limited in our measures of GR activity and time points.