University of Minnesota
Nathan Springer is out to unearth the epigenetic factors in different lines of corn.
Above and beyond the genes
How stealthy 'epigenetic' factors are changing the study of heredity
By Deane Morrison
Since identical twins share all the same genes, how come sometimes only one gets a heritable disease?
And how can genetically identical cells in a leaf be different colors?
Those were just two mysteries that hinted at factors beyond the DNA sequence having a hand in determining traits. Those factors turn out to be stealthy chemical agents that turn genes on or off and sometimes modify the degree to which genes are expressed as traits.
Welcome to the realm of epigenetics, a red-hot frontier in the modern science of heredity.
"Epigenetics is the study of heritable information not contained in the genetic sequence," says Nathan Springer, an associate professor of plant biology at the University of Minnesota. "The term 'epigenetics' literally means 'above the genes' and describes information superimposed upon genetic sequences."
So important is the field that the National Science Foundation just awarded $1.6 million to Springer and Matthew Vaughn of the University of Texas at Austin to study the epigenetic control of traits in corn. Springer's work, along with studies of human epigenetics by U of M researchers Irving Gottesman, Heather Nelson, and others, illustrates the immense reach of epigenetics as a door to understanding heredity.
From cradle to grave
It's a good thing evolution has provided a means of switching genes in a cell nucleus on and off, either wholly or partly. Consider an embryo: Various genes must be turned on and off in different tissues and at different times during development. Otherwise, we'd all end up as big balls of identical cells with nothing to differentiate brain from liver from muscle. Similarly, genes are turned on and off after birth to guide growth, reproduction, and other functions.
Mother of all tags
In most mammals, one of the two X chromosomes in females is epigenetically inactivated and turned into a dense structure called a Barr body. Since males have only one X chromosome, this keeps the "dose" of genes on the X chromosome even for the two sexes. In other words, because humans have evolved to need only one copy of those genes, an extra copy would cause problems—and so females eliminate it.
The simplest form of epigenetic control is a "tagging" process called methylation, in which an enzyme attaches small molecules called methyl groups to specific sites on DNA. With enough methyl groups, a tag will foil the cellular machinery that activates genes.
Some tags can come and go throughout an organism's life. This process can start in the earliest stages of development and can leave identical twins with substantial differences in the patterns of their tags. This may in turn lead to only one having a disease or other genetically influenced condition.
Epigenetic tagging often occurs in a seemingly random fashion. This can cause headaches for plant breeders, because if different patterns of tagging occur in genetically identical seeds, it can reduce their ability to predict traits. Removing some of that surprise for corn breeders is the aim of Springer's NSF grant.
He and his colleagues are looking for epigenetic tagging patterns that can affect crop quality in various lines of corn and ways to test plants for their presence.
For instance, he wants to know if there are epigenetic changes whose presence isn't associated with any particular DNA sequence. ("Such epigenetic changes may contain important, previously hidden, information," says Springer.) And if there are, where—that is, at which locations on which chromosomes—do they occur?
Finally, if such differences are found, how heritable or stable are they?
"Right now we can't predict how many generations an epigenetic change will last," Springer says. For example, if an epigenetic tag inactivates an undesirable gene, it would be good to know if that tag is likely to be lost sometime in the next few breeding cycles.
Several human genes act to suppress tumors, and others—oncogenes—promote cancer when overactive. Epigenetic tags can orchestrate the behavior of both, says Nelson, an associate professor in the Masonic Cancer Center and Division of Epidemiology and Community Health.
"Aberrant methylation is equivalent to mutation events," she explains. "For example, tumor suppressor genes. You want them [active] in a cell." But methylation, as well as a mutation, could inactivate them.
The same goes for an oncogene, she says. A loss of methylation can leave it permanently "turned on."
"We've known for a long time that globally, tumors are [under]-methylated," Nelson says. But, she adds, no one knows yet whether the low level of methylation is a cause or effect of tumor growth.
She and her colleagues have recently submitted for publication a study comparing lung tumors with normal lung tissue and identifying key tumor suppressor genes that become methylated during tumor formation.
"We know that this process of proper methylation is important," she says. One of her main goals is to find a reliable test for whether a person's cells are competent to keep the right regions of their genome methylated—and silent.
Over his long career, Gottesman, a senior fellow in the Department of Psychology and the Bernstein Professor in Adult Psychiatry, has become convinced that epigenetic factors are at least part of the reason why IQ and schizophrenia, though "highly heritable," may vary between identical twins.
In one test, he and his colleagues looked at two sets of identical twins: one in which both twins had schizophrenia and one in which only one twin had it. In all four subjects they compared the degree to which a gene related to mood and brain functioning was methylated.
In that respect, the schizophrenic person with a normal twin was closer to the two schizophrenic twins than to his own sibling.
"This was a proof of principle," says Gottesman, meaning that the study showed that identical twins can differ in their epigenetic profiles in ways that may affect their health outcomes.
Gottesman also sees epigenetics as a likely bridge that allows environmental factors to affect the functioning of genes. The effects of environment on intelligence are a case in point.
"Intelligence is 70 percent heritable. That means that 70 percent of the variance in intelligence among individuals in a population is explainable by inheritance," Gottesman says. Except, he found, when poverty enters the picture.
In a landmark study, he and his colleagues at the University of Virginia examined many sets of twin children, both affluent and poor. In the affluent, IQ heritability was high. But in the poor children, they found very little evidence for a genetic component to intelligence; in other words, its heritability was near zero. Thus, a poor environment can swamp genetic influences.
"We can explain this as how having a good environment lets genes 'do their own thing,'" says Gottesman. "But in a poor environment, genes are too busy dealing with the privations of a poor environment."
Poor pre- and perinatal conditions may lead to any number of biochemical effects in children. These in turn could ultimately cause altered patterns of epigenetic tagging that lead to lowered IQ.
Gottesman would like to see a full-scale effort, involving scientists with a variety of skills, to discover the hidden story of how epigenetic tagging, gene function, and the environment interact to produce traits. The complexity of such an undertaking would rival NASA's drive to put men on the moon.
"We really need an Apollo Program for epigenetics," he says.