Epigenetics

DNA methylation adds a –CH3 group to cytosine bases at CpG dinucleotides; in gene promoters this typically silences transcription by blocking transcription factor binding and recruiting repressive complexes. It is written by DNMTs, erased by TETs, and was long thought to be the most stable epigenetic mark. Histone modification refers to a wide range of chemical additions (acetylation, methylation, phosphorylation, ubiquitination) to the N-terminal tails of histone proteins. Histone acetylation (by HATs) loosens chromatin and activates transcription; deacetylation (by HDACs) compacts and represses. Histone methylation has context-dependent effects (H3K4me3 activates; H3K27me3 represses). Together, these marks constitute a "histone code" that determines whether a gene is accessible to the transcription machinery.

A gene-environment interaction occurs when the effect of an environmental exposure on an outcome depends on an individual's genotype — or equivalently, when genetic effects on a trait differ across environments. The Caspi et al. (2003) finding was important because it was among the first candidate GxE studies with a clear molecular mechanism: the short allele of 5-HTTLPR reduces serotonin transporter expression, lowering synaptic serotonin clearance. In combination with childhood maltreatment (which activates the HPA axis and influences serotonergic systems), this conferred substantially elevated depression risk in adulthood — while the same maltreatment had much less effect in long-allele carriers. The paper launched the gene-environment interaction paradigm and provided biological plausibility for why adverse childhood experiences have lasting effects on mental health in some individuals but not others.

The diathesis-stress model (Zubin & Spring, 1977) proposes that genetic or biological vulnerabilities (diatheses) combine with environmental stressors to produce disorder — the genetic variant increases risk in adverse conditions but has no effect in benign ones. Differential susceptibility (Belsky, 2005) proposes something stronger: that "susceptibility" alleles increase responsiveness to the environment in both directions — making carriers more harmed by adversity but also more benefited by supportive conditions than less-sensitive individuals. The metaphor is orchids vs dandelions: dandelions grow acceptably in any environment; orchids fare badly in poor conditions but spectacularly in optimal ones. Meta-analytic evidence supports differential susceptibility effects for several GxE pairs, suggesting that evolutionary maintenance of "risk" alleles reflects their benefit in advantageous environments.

Transgenerational epigenetic inheritance is well-established in plants and some invertebrates. In mammals, the evidence is more limited: most methylation marks are erased during gametogenesis and early embryonic reprogramming. However, some marks — particularly at imprinted loci and in germ cells — can escape erasure. Human epidemiological studies (e.g., the Dutch Hunger Winter, Överkalix cohort) suggest that nutritional and psychological exposures in one generation influence health outcomes in grandchildren, though the epigenetic mechanism is not definitively established. Meaney and colleagues showed that stress-induced epigenetic changes in glucocorticoid receptor methylation in rats influence the maternal behaviour of the offspring themselves, propagating the phenotype across generations via behaviour rather than direct germline epigenetic transmission — a compelling but mechanistically distinct pathway.