Genomic imprinting in mammals was recognized around 25 years ago from two very different experimen tal approaches in the mouse. From nuclear transplantation experiments, it was shown that both a paternal genome and a maternal genome were required in order for normal development of an embryo to birth1,2. At the same time, sophisticated genetic techniques were being used to generate mice that were chromo somally balanced, but had inherited both copies of a single chromosome or part of a chromosome from one parent only. Uniparental inheritance of some chromosome regions (called imprinting regions), but not others, resulted in the birth of mice with abnormal phenotypes3. The finding that phenotypic effects could be localized to particular chromosome regions implied that specific genes were involved, and this led to the startling conclusion that, contrary to Mendel's principles, some mammalian genes functioned differ ently depending on whether they had been inherited from the mother or the father. From further genetic experiments, 13 imprinting regions on eight autosomes were identified. The developmental phenotypes associated with uniparental inheritance of these regions ranged from pre and postnatal growth defects, reduced viability as well as postnatal behavioural deficits (Figure 1). Furthermore, the genetic experiments in the mouse indicated that imprinting may affect human disease and a number of human disorders with developmental defects involving growth and/or neurological anomalies resulting from mutations in im printed genes or affecting epigenetic imprinting marks have been identified.

This content is only available as a PDF.