Genomic imprinting is the mechanism for transgenerational transmission of epigenetic adaptations to changing environmental conditions and the requirements for sex differences in reproductive success. Imprinting involves differential DNA methylation in the egg and sperm (Figure 1). Upon fertilization, these marks can be reprogramed in an XX- versus XY-biased manner, which then tailors the epigenome for future life as an XX or XY individual (6). The regulation of imprinted genes is governed by a complex interplay of DNA methylation, histone modifications, noncoding RNAs (microRNAs and long noncoding RNAs [lncRNAs]), and chromatin structure, with imprinted control regions (ICRs) playing pivotal roles in the establishment and preservation of imprinted marks (7).

Mechanisms of transgenerational epigenetics. Transgenerational inheritance of epigenetic modifications are influenced by hormonal endocrine deregulators/disruptors and assisted fertility procedures and environmental exposures (chemicals, pollutants, toxins and pathogens), lifestyle factors (sedentary vs. physical activity, diet, alcohol, drug and nicotine use), maternal and paternal stressors (emotional, physical, psychological, and relationship dynamics) that can be passed on to subsequent generations: from parent (F0) to fetus (F1), to fetal gametes (F2), and so on. Such epigenetic modifications are known to alter the imprinting status of various genes (DLK1-MEG3, PEG1/MEST, UBE3A, CDKN1C, IGF2, H19) that manifest in imprinting disorders including: Prader-Willi Syndrome, and Angelman syndrome, and Beckwith-Wiedemann syndrome, among others. These syndromes are affiliated with cellular growth abnormalities predisposing the affected individual to an array of cancers including Wilms’ tumor, neuroblastoma, hepatoblastoma, and breast, uterine, ovarian and prostate cancers.

During gametogenesis, specifically in primordial germ cells (PGCs), epigenetic marks are erased through global demethylation. This is followed by sex-specific DNA methylation patterns in sperm and eggs (8). Upon fertilization, the pronuclei of the egg and sperm merge, forming the zygote, which undergoes extensive epigenetic reprogramming of DNA demethylation and histone modification marks.

Transgenerational transfer of imprinted genes is maintained through multiple mechanisms, including ATP-dependent SWI/SNF and ISWI chromatin-remodeling complexes. These complexes play critical roles in maintaining the chromatin structure at imprinted loci (9). DNA methylation and histone modifications at imprinted loci are also preserved across generations of cells and individuals. In addition to DNA methyltransferase (DNMT) and histone-modifying enzyme activity, noncoding RNAs, which may be produced by the imprinted genes themselves, also participate in feedback loops and regulatory networks involved in maintaining imprinted genes (6, 7).

Imprinted genes play essential roles in embryonic growth, maternal-placental interactions, nutrient transfer, organogenesis, morphogenesis, and postnatal metabolism (8). The importance of imprinting is well illustrated by the pathological consequences of imprinting disorders (IDs). Anomalous DNA methylation patterns and loss of imprinting at specific genomic loci are associated with a range of developmental abnormalities and diseases, including Angelman syndrome (AS), Prader-Willi syndrome (PWS), Beckwith-Wiedemann syndrome (BWS) (7, 8), and Silver-Russell syndrome (6).

PWS and AS result in developmental and cognitive impairments that manifest along with multiple other syndrome-specific features. Both syndromes result from multiple mechanisms, including IDs involving chromosome 15q11–q13. Which syndrome occurs depends on whether there is loss of maternal expression of maternally expressed genes (MEGs) and UBE3A (in AS) or loss of paternal expression of paternally expressed genes (PEGs) (in PWS) (10). Sex differences in expression of MEGs and PEGs differs widely in a tissue-specific manner, with different sex-specific and shared tissues exhibiting either MEG- or PEG- dominant expression (11). It is important to note that sex differences in gene and protein expression are not required for sex differences in gene and protein activation and action (12–15). In murine models and human studies, the absence of sex differences in gene and protein expression was still associated with substantial differences in their action due to sex differences in chromatin accessibility, gene-regulatory networks, and intracellular signaling pathway regulation.

In BWS, the ID involves chromosome 11 and demethylation of the maternal IGF2, LIT1, KvDMR gene region (which regulates a cluster of genes) or methylation of the H19DMR region (also known as imprinting center 1 [IC1]) (16). IGF2 is an essential growth promoter in early fetal life, and H19DMR is an important negative regulator of its function. Normally, maternal IGF2 is imprinted and silenced, while paternal H19 is imprinted and silenced. This antagonism between maternal and paternal imprints is essential for normal growth. In BWS, there is unopposed IGF2 function, resulting in an overgrowth syndrome with hemihypertrophy, hyperinsulinism, and a 10% increase in risk of childhood cancers such as multifocal bilateral Wilms’ tumor, hepatoblastomas, and neuroblastomas (17). In a rodent embryonic brain analysis, there was evidence for sex differences in IGF2 and H19 expression (18).

In addition to cancers that complicate BWS, variant ID methylation is associated with neuroblastoma (involving the DLK1-MEG3 imprinted domain) (6); acute myeloblastic leukemia (due to hypermethylation of the imprinted NNAT locus) (7); uterine leiomyoma (due to overexpression of PEG1 [also known as MEST]) (8); colorectal cancer (due to hypomethylation of H19 and IGF2, or IGF2 DMR0 hypomethylation) (7); breast cancer (due to PEG1 loss of imprinting) (6); and ovarian cancer (due to epigenetic alterations in the IGF2/H19 gene cluster or downregulation of ARHI and PEG3, whose products have tumor-suppressor function) (6, 7).

There are no reported robust or consistent sex differences in PWS phenotypes (19). There are, however, sex differences in the frequency of AS and PWS. These two syndromes can arise from nondisjunction during gametogenesis, resulting in uniparental disomy of pathogenic regions of chromosome 15. When this occurs during oogenesis, the offspring inherit two maternally imprinted copies of chromosome 15 and no paternally expressed copy of the gene and develop PWS. If the nondisjunction occurs during spermatogenesis, offspring inherit two paternally imprinted copies of chromosome 15 and develop AS. Because nondisjunction occurs more frequently during oogenesis than spermatogenesis, maternal uniparental disomy causing PWS is more common than paternal uniparental disomy causing AS (10). Thus, AS, PWS, and BWS all illustrate the presence of powerful sex differences in imprinting and the importance of balance between sex-adapted imprints for normal development, reproduction, disease risk, and long-term health.

In addition to supporting normal development, imprinting provides a mechanism for sex chromosome complement-adapted writing, erasing, and reading of DNA methylation marks for the transmission of positive and negative effects of the prior generations’ environmental stresses. Striking examples of sex differences in the transgenerational effects of stress are found in the metabolic reprogramming that has followed multiple famines, such as the Dutch famine of 1944–1945, the Great Chinese Famine of 1959–1961, as well as in Swedish famine cohorts (6, 7).

A number of sex differences in the Dutch famine effects have been identified. The first was a flip in the female-to-male birth ratio, from 47:53 before the famine to 52–51:48–49 in the affected cohort. Females exposed to famine in utero had higher rates of cardiovascular disease and cancer, with increased mortality from these causes than females born before the famine (20). Males exposed to in utero starvation had smaller intracranial volumes and on functional MRI (fMRI) studies appeared to have brains older than their chronological age, increased depression and anxiety, as well as inferior physical performance abilities (21). The children of individuals exposed to famine early in life also exhibit altered rates of obesity, hyperglycemia, type 2 diabetes, renal dysfunction/chronic kidney disease, and cardiovascular disorders (22). Interestingly, individuals in the Dutch famine cohort exhibit alterations in IGF2, but not H19, imprinting compared with unaffected siblings, underscoring the potential action of imprinting on transgenerational consequences of changing environmental stress (23). The Överkalix famine cohort and the Uppsala Birth Cohort Multigenerational Study (24–26) report similar sex differences in the transgenerational effects of famine.

We can expect highly personalized effects of environmental exposures and maternal/paternal stress on the programming and reprogramming of imprinted loci. Nutrition, lifestyle, stress, and exposure to chemicals and toxins impact the maintenance of imprinted alleles by affecting the activity of epigenetic regulators. Moreover, it is important to recognize that imprinting provides an established biological mechanism by which transgenerational gender stress can become ineluctably entangled with chromosomal and gonadal sex.