Concepts familiar from grade-school algebra have broad ramifications in computer science.
Everyone inherits two sets of genes, one from each parent. In most cases, this means we have two functional copies of each gene.
However, almost a decade ago mouse geneticists showed that maternal and paternal genomes do not always act in the same way. In certain cases, either the paternal or maternal copy of a gene is specifically inactivated, or silenced, by a process called "genomic imprinting."
Interest in genomic imprinting has grown in recent years as a result of clinical studies showing that it may be important in many human diseases. Abnormalities in imprinting have been implicated in two inherited forms of mental retardation, Angelman's syndrome and Prader-Willi syndrome, as well as several forms of inherited cancer, including Wilms' tumor.
Scientists have been working for some time to understand how imprinting operates. Now researchers at the Whitehead Institute for Biomedical Research report the first in vivo evidence that genomic imprinting is associated with a specific chemical change in the DNA.
In the November 25 issue of Nature, Dr. Rudolf Jaenisch, Dr. En Li, and Caroline Beard show that genomic imprinting is maintained by a process called DNA methylation (the chemical attachment of methyl groups, CH3-, to certain regions of the DNA). Previous laboratory studies had suggested that DNA methylation might play an important role in the regulation of gene expression during development, but the evidence was not conclusive.
To learn more, the Jaenisch group created an animal model. They used new gene transfer technologies to produce a mouse strain with one very specific defect: the mice carry a mutation in the gene encoding the enzyme required to add methyl groups to DNA (DNA methyltransferase). Animals with one copy of the mutant gene are physiologically normal, but when two such animals are mated, some of the embryos-embryos with two copies of the mutant gene-die in the womb. The timing of death is particularly interesting. The embryos die between days 9 and 11, when cells normally begin to differentiate into primitive tissues and organ systems.
To understand the cause of death, the Whitehead scientists began searching for genes affected by the mutation in DNA methyltransferase. They discovered that inactivation of DNA methyltransferase dramatically altered the activity of one particular set of genes-regulatory genes known to be imprinted in early development.
For example, in normal embryos the maternal copy of the regulatory gene H19 is active and the paternal copy is silent. In the mutant embryos, both copies were switched on at approximately the same levels. These results indicated that DNA methylation is essential for maintaining the inactive state of the paternal H19 gene.
In future studies, the Whitehead group will begin searching for signals in the genome that trigger imprinting. Meanwhile, the new mouse model will provide a system for exploring other processes believed to be associated with DNA methylation, including X inactivation (in the cells of female mammals, one of the two X chromosomes is inactivated), tumor formation, and aging.
Dr. Jaenisch is a member of the Whitehead Institute and professor of biology at MIT. Dr. Li, formerly a postdoctoral fellow in the Jaenisch laboratory, is now an assistant in genetics at the Cardiovascular Research Center at Massachusetts General Hospital. Ms. Beard is a technical associate in the Jaenisch lab.
A version of this article appeared in the December 8, 1993 issue of MIT Tech Talk (Volume 38, Number 17).