Epigenetic memory is the ability of certain cells to pass on stable DNA modifications like methylation via mitotic and sometimes meiotic divisions. This causes alterations in the descended cell’s properties and behaviors. It is possible that the “lived” experience of the cells—physiological and environmental influences during the cell’s lifetime—can confer these epigenetic marks.
Recently, dos Santos et al. investigated the relationship between epigenetic memory and lactation.1 To do this, the team implanted nulliparous and uniparous mice with slow-release hormone pellets (estrogen/progesterone) that mimic mouse pregnancy. They harvested the animals’ mammary glands 6 and 12 days after implantation for histological analysis, sequencing and RNA quantification. For the latter, the team relied upon real-time polymerase chain reaction (PCR) using a 7900 Real-Time PCR System and gene-specific primers created with Primer Express software (both Thermo Scientific).
Histologically, both nulliparous and parous mice exhibited ductal branching consistent with pseudo-pregnancy. Here the parous animals showed a more rapid response to the hormone cocktail, with a greater number of ductal structures at each time point when compared to nulliparous animals. Using antibody staining, the team observed that while both nulliparous and parous mice produced similar levels of milk proteins 12 days after implantation, the mammary cells of parous mice displayed much stronger staining at day 6. This indicates delayed milk production for nulliparous animals.
The team turned to whole genome sequencing, focusing on methylation as the potential mechanism for epigenetic alterations. Hierarchical clustering of the methylation profiles of nulliparous mammary cells and non-mammary cells (embryonic stem cells, brain cells, blood cells, sperm and intestinal cells) showed a distinct conserved epigenetic signature for mammary cells. These cells further split into two methylation profiles, corresponding to the luminal and basal compartments.
The researchers report that all animals retained compartmental identity in their mammary cells, but within each compartment, parous mice exhibited significant differences from nulliparous animals. In general, the luminal compartment experienced greater impact than did the basal compartment. The team reports that fewer than 10 regions of the basal compartment exhibited simultaneous changed methylation status in more than one cell type. This contrasts sharply with the luminal compartment, where approximately 800 regions showed hypomethylation in parous mice and 50 regions showed increased methylation as a result of parity. This makes sense, since luminal cells are more abundant and more dynamic in the mammary gland during pregnancy when compared with basal cells.
Overall, the altered methylome of luminal cells related to the development of ductal structures and alveoli during pregnancy, milk production during lactation, and remodeling post-lactation. Gene ontology associated regions that lost methylation as a result of pregnancy with genes that participate in cell proliferation, cell-cell adhesion and cell death. Further, the researchers note that whole genome analysis as applied here dramatically increased the identification of differentially methylated regions (DMRs) when compared with previous studies.
The team also analyzed a Stat5a ChIP-seq data set from a mammary gland in lactation2 after noting strong enrichment for STAT family motifs. The Stat5a/b family of transcription regulators have been linked to alveolar development and milk production. The researchers found that 63% of the peaks occurred in hypomethylated regions (HMRs) present in both nulliparous and parous methylomes, 17% only in parous methylomes, and 1% only in nulliparous methylomes. The remaining 19% did not overlap HMRs but still presented as less methylated in parous samples. The team indicates that during pregnancy, Stat5a activity results in hypomethylation at its binding sites and that this state persists after pregnancy even after Stat5a activity declines to its baseline.
Finally, dos Santos et al. looked at how enduring changes in DNA methylation patterns prime pregnancy-associated genes to mount a rapid response to future pregnancies. They analyzed 47 genes, 33 of them linked with parous-specific HMRs and 1 (Ccnd1) that methylates post-pregnancy. They found that within the luminal compartment, most parous-specific HMRs retained the state of low methylation throughout most of the mouse’s reproductive life even though the luminal compartment experiences continuous cellular turnover.
Overall, the team confirmed that pregnancy produces an epigenetic memory that alters the mammary gland, resulting in a more rapid response to subsequent pregnancies. They note that an early pregnancy confers life-long protection against breast cancer and hypothesize that this phenomenon may rely on similar mechanisms, representing fertile areas for further study into long-term epigenetic changes.
1. dos Santos, C., et al. (2015) “An epigenetic memory of pregnancy in the mouse mammary gland,” Cell Reports, 11(7) (pp. 1102–1109), doi: 10.1016/j.celrep.2015.04.01.
2. Kang, K., et al. (2014) “Mammary-specific gene activation is defined by progressive recruitment of STAT5 during pregnancy and the establishment of H3K4me3 marks,” Molecular Cellular Biology, 34(3) (464–473), doi: 10.1128/MCB.00988-13.