Genomic Imprinting

Core Idea

Genomic imprinting is an epigenetic phenomenon in which certain genes are expressed from only one parental allele — either the maternal or paternal copy — while the other is silenced through DNA methylation established in the parental germline. Approximately 100-200 imprinted genes have been identified in mammals, many clustered in imprinting control regions. Imprinting is established during gametogenesis (differently in sperm versus eggs), maintained through development, and erased and re-established each generation in the germline. Many imprinted genes regulate growth: paternally expressed genes (like Igf2) tend to promote growth, while maternally expressed genes (like H19, Igf2r) tend to restrict it — consistent with the parental conflict hypothesis. Imprinting disorders (Prader-Willi, Angelman, Beckwith-Wiedemann syndromes) demonstrate the clinical consequences of disrupted mono-allelic expression.

Explainer

In standard Mendelian genetics, it does not matter whether an allele comes from the mother or the father — both are expected to function equally. Genomic imprinting violates this assumption. For imprinted genes, only one parental allele is active; the other is silenced by epigenetic modifications (primarily DNA methylation) established during gamete formation. This means cells are functionally hemizygous at imprinted loci — a loss-of-function mutation or deletion of the active allele cannot be compensated by the silent allele from the other parent.

The molecular mechanism involves differentially methylated regions (DMRs) — DNA sequences that are methylated on one parental allele but not the other. These methylation differences are established during gametogenesis: oocytes and sperm methylate different imprinting control regions. After fertilization, the differential methylation is faithfully maintained through development by DNMT1 (maintenance methyltransferase), which copies methylation patterns to the newly synthesized DNA strand during replication. The methylation state of the imprinting control region determines which allele's genes are expressed — through mechanisms including blocking CTCF insulator binding (Igf2/H19 locus), directing antisense RNA expression that silences nearby genes, or directly recruiting repressive chromatin complexes.

The parental conflict hypothesis (Haig, 1993) provides an evolutionary explanation for why imprinting exists. In mammals, the mother invests heavily in each offspring through pregnancy and lactation, and her evolutionary interest is to distribute resources among all her offspring (present and future). The father's evolutionary interest is to maximize resource extraction for his offspring specifically (since future offspring may have a different father). Consistent with this prediction, many paternally expressed imprinted genes promote fetal growth (Igf2 — insulin-like growth factor 2), while many maternally expressed genes restrict growth (Igf2r, H19, p57). Imprinting thus represents a molecular tug-of-war between parental genomes over resource allocation to offspring.

Imprinting disorders illustrate the clinical consequences. Prader-Willi syndrome (loss of paternally expressed genes at 15q11-q13) causes insatiable appetite, obesity, and intellectual disability. Angelman syndrome (loss of the maternally expressed UBE3A gene at the same locus) causes severe intellectual disability, seizures, and characteristic happy demeanor. Beckwith-Wiedemann syndrome (overexpression of Igf2 due to imprinting defects at 11p15) causes overgrowth and increased cancer risk. These disorders demonstrate that normal development requires precisely one active copy of imprinted genes — neither zero (when the active allele is lost) nor two (when the silent allele is inappropriately activated). Imprinting adds a layer of regulation beyond the DNA sequence, making the parental history of each chromosome developmentally relevant.