Function of Imprinted Genes: Implications for Placental Development and Growth (PDF - 2.19 MB)
Dr. Benjamin Tycko, Columbia University, Department of Pathology, New York, New York
Genomic imprinting leads to parent-of-origin dependent allelic silencing of a small subset of mammalian genes. The earliest experiments identifying this non-Mendelian phenomenon involved the creation of uniparental conceptuses in mice, and immediately highlighted a role for imprinted genes in controlling the placenta. Androgenetic conceptuses showed relative overgrowth of placental tissues, while gynogenetic conceptuses had stunted placentas and yolk sacs. Since then, there has been much interest in the biological rationale for imprinting. The intergenomic conflict hypothesis was proposed 13 years ago by Haig and colleagues motivated by data from mice with knockouts of two prototypical imprinted genes, Igf2 and Igf2r. Igf2 is imprinted such that the paternal allele is expressed and the maternal allele silenced, while Igf2r, encoding a clearance receptor for circulating Igf2 peptide, is imprinted in the opposite pattern. The two knockouts produced opposite phenotypes: placental and fetal growth retardation in the Igf2-null conceptuses, and overgrowth in the Igf2r-null conceptuses. The conflict hypothesis explained these observations by positing opposite maternal versus paternal drives to control allocation of maternal resources to each conceptus. In settings of multiple paternity, it is argued that the father will propagate his genome most efficiently if his germline imprints genes to promote the growth of his offspring. The mother, by contrast, is postulated to propagate her genome most efficiently by imprinting genes to prevent undue metabolic demands on her resources by any single conceptus, that is, to restrain pre- and postnatal growth. Recently, we reviewed information from gene knockout experiments and naturally occurring mutations in imprinted genes, including data for 25 imprinted loci . We concluded that there is much evidence supporting the Haig hypothesis, that there are no in vivo data contradicting it, but that the results for some imprinted loci remain neutral or difficult to interpret.
A functioning placenta is necessary to allow alleles to be passed to the next generation, but the downside to the mother is the parasitic aspect of this organ. In addition to shunting maternal resources to the fetus, the late-stage placenta consumes up to 30 percent of calories delivered by the mother, simply to maintain its own function. Among the imprinted genes for which in vivo data are available, at least eight regulate the size and/or development of this organ. These include Igf2, Mash2, p57 Kip2, Ipl, Esx1, Peg1, Peg3, and an incompletely characterized imprinted locus on mouse chromosome 12. Based on knockout data for Ipl and Esx1, and on genetic data for the chromosome 12 imprinted locus, we have suggested a modification of the conflict hypothesis, in which maternal drive to restrain placental growth is by itself sufficient to select for imprinting . This scenario invokes conflict between the mother and the extraembryonic tissues of her offspring. By limiting placental size, the mother may produce a greater number of offspring over her lifetime, thereby more effectively propagating her alleles in the population.
What is the medical importance of this line of work? Poorly understood placental defects are thought to underlie some cases of IUGR and maternal preeclampsia, two common and costly medical conditions. But pathologists have no firm understanding of the etiology of either of these disorders. Placental overgrowth in humans, as occurs in the Beckwith-Wiedemann syndrome, has a well-defined epigenetic basis, involving the dysregulation of imprinted genes. Could the more common problems of placental insufficiency and IUGR also be linked to pathological imprinting?