Immune cell type and DNA methylation vary with reproductive status in women: possible pathways for costs of reproduction

Abstract

Background Consistent with evolutionarily-theorized costs of reproduction (CoR), reproductive history in women is associated with life expectancy and susceptibility to certain cancers, autoimmune disorders, and metabolic disease. Immunological changes originating during reproduction may help explain some of these relationships. Method ology: To explore the potential role of the immune system in female CoR, we characterized leukocyte composition and regulatory processes using DNA methylation (DNAm) in a cross-sectional cohort of young (20-22 year-old) women differing in reproductive status. Results Compared to nulliparity, pregnancy was characterized by differential methylation at 828 sites, 96% of which were hypomethylated and enriched for genes associated with T-cell activation, innate immunity, pre-eclampsia, and neoplasia. Breastfeeding was associated with differential methylation at 1107 sites (71% hypermethylated), enriched for genes involved in metabolism, immune self-recognition, and neurogenesis. There were no significant differences in DNAm between nulliparous and parous women. However, compared to nullipara, pregnant women had lower proportions of B, CD4T, CD8T, and natural killer (NK) cells, and higher proportions of granulocytes and monocytes. Monocyte counts were lower and NK counts higher among breastfeeding women, and remained so among parous women. Implications Our findings point to widespread differences in DNAm during pregnancy and lactation. These effects appear largely transient, but may accumulate with gravidity become detectable as women age. Nulliparous and parous women differed in leukocyte composition, consistent with more persistent effects of reproduction on cell type. These findings support transient (leukocyte DNAm) and persistent (cell composition) changes associated with reproduction in women, illuminating potential pathways contributing to CoR. LAY SUMMARY Evolutionary theory and epidemiology support costs of reproduction (CoR) to women’s health that may involve changes in immune function. We report differences in immune cell composition and gene regulation during pregnancy and breastfeeding. While many of these differences appear transient, immune cell composition may remain, suggesting mechanisms for female CoR.

Pregnancy is linked to faster epigenetic aging in young women

“…The metabolic, immunological, and inflammatory profiles prevalent in high-income, Western settings, where these clocks were trained, often differ from those observed in the CLHNS and other non-Western contexts ( 51 , 52 ). More importantly, all of our follow-up samples were taken from pregnant women, whose metabolic, physiological, and immunological profiles—as well as methylomes—are quite different from the largely nonpregnant population used in the training datasets ( 53 – 55 ). While it is unclear to what extent the physiological, metabolic, and immunological changes that accompany pregnancy are qualitatively distinct from variation in these states among nonpregnant people ( 55 , 56 ), clocks built around these measures may be particularly sensitive to reproductive status.…”

“…More importantly, all of our follow-up samples were taken from pregnant women, whose metabolic, physiological, and immunological profiles—as well as methylomes—are quite different from the largely nonpregnant population used in the training datasets ( 53 – 55 ). While it is unclear to what extent the physiological, metabolic, and immunological changes that accompany pregnancy are qualitatively distinct from variation in these states among nonpregnant people ( 55 , 56 ), clocks built around these measures may be particularly sensitive to reproductive status. Because they are trained on chronological age, which is precisely measured and has similar meaning in all contexts, Horvath and Hannum clocks may be more robust to differences in biomarkers that vary between populations and with reproductive status.…”

“…Second, we restricted our measure of reproductive effort to gravidity. While many of the metabolic, physiological, and immunological changes associated with pregnancy overlap with those that accompany aging ( 55 , 56 ), breastfeeding and parental care are also thought to contribute to the long-term costs of reproduction. More work characterizing these other forms of reproductive investment—which in the case of breastfeeding requires detailed records of the frequency, duration, and intensity of breastfeeding for each child—and estimating their impact on biological aging is needed ( 22 , 58 ).…”

Maternal epigenetic clocks measured during pregnancy do not predict gestational age at delivery or offspring birth outcomes: a replication study in metropolitan Cebu, Philippines

“…Another limitation was that our blood samples were not taken at the same time during pregnancy for each woman. This may be important because prior work has demonstrated that DNAm in general and epigenetic age specifically, and their relationship with birth outcomes, can change during pregnancy [ 36 – 38 ]. Nevertheless, our blood sampling occurred within a relatively narrow range of 23–41 weeks, and we adjusted all clock measures for gestational age at measurement.…”

The emerging role of TET enzymes in the immune microenvironment at the maternal-fetal interface during decidualization and early pregnancy

“…Studies have shown that the genes affected by decreased methylation during decidualization were mainly associated with immune response regulation ( 19 , 20 ). Furthermore, DNA methylation also plays a vital role in immune cell development and maturation, which contributes to decidual immune homeostasis ( 21 – 23 ). Altogether, we performed a comprehensive literature review concerning the roles of TET enzymes in the microenvironment at the maternal-fetal interface during decidualization and early pregnancy.…”