Featured Articles

Diet During Pregnancy May Have Effects Lasting Into Adulthood

By Sharon Begley, Wall Street Journal

22 August 2003: As mysteries go, these don't seem to have much in common: A child born underweight has a higher than normal risk of developing heart disease, diabetes, obesity and hypertension as an adult.

One identical twin develops schizophrenia, which studies of families show is a genetic disease, but the other twin is spared.

Last week's column 1looked at scientists' growing realization that, when it comes to important changes in the genome -- if I may corrupt an old political mantra -- "It's not just the sequence, stupid."

Mice with identical genes for fur color can be brownish or yellow, depending on whether their gene for fur color has been silenced by what their mother ate during pregnancy. It's beginning to look as if such "epigenetic" changes, defined as those having no effect on the sequence of molecules that make up a genome, may be major players in determining traits and disease risk.

"The completion of the human genome project is a monumental event, but there's still an enormous amount that we have not yet fleshed out," says psychiatrist James Potash of Johns Hopkins University School of Medicine, Baltimore. "Epigenetic variation is one."

Take the enigma of fetal programming, in which nutrition during gestation seems to affect the risk of disease decades later. At a June conference on the subject, attended by some 700 scientists, "What came shining through is that birth weight affects the risk of diabetes, coronary heart disease, obesity, hypertension and breast or prostate cancers," says David Barker of England's University of Southampton. Scrawny newborns, in general, grow up to have a higher incidence of the first four; chubby ones, a higher risk of the latter.

At first, scientists thought the reason was physiology, not genetics. For example, newborns who are small for their length probably have fewer kidney cells than they should. Since the kidneys regulate blood pressure, undersized kidneys can increase later risk of hypertension and thus heart disease, explains Dr. Barker.

But fetal programming "almost certainly" reflects epigenetic changes, too, says Craig Cooney of the University of Arkansas for Medical Sciences. That's because, much as in the mice whose color reflects what mom ate while pregnant, nutrients reaching the human fetus can include more or fewer of the molecules that silence or activate genes. Maybe too few nutrients during gestation might mean not enough of the molecules that silence heart-disease-causing genes.

"The nutrition an embryo receives at crucial stages of development can have important and lasting effects on the expression of various genes, including those involved in health and disease," says Randy Jirtle of Duke University Medical Center, Durham, N.C.

One target of such silencing must have Gregor Mendel turning over in his grave. The Austrian monk, regarded as the founder of genetics, concluded that which parent a gene comes from is irrelevant. True, we carry two copies of every gene (except those on the Y chromosome), one from mom and one from dad. But dozens of genes in sperm or ova are tagged with the biochemical equivalent of "don't mind me." Throughout life, those genes are silenced, or "imprinted." If mom's gene is imprinted, only dad's counts; if dad's is imprinted, only mom's counts.

The gene sequence hasn't changed, so imprinting is epigenetic -- and something you don't want to mess up. When the gene for insulin-like growth factor 2 (IGF2) loses its imprinting, for instance, the once-silenced copy is activated, loosing a flood of growth factor that promotes childhood and adult cancers. Yet if you were to sequence that IGF2 gene, it would look just fine.

Such imprinting mistakes may be affecting some test-tube babies. The incidence of a rare genetic disease called Beckwith-Wiedemann syndrome was six times as high as in children conceived the traditional way, according to a study published in January. This syndrome occurs when IGF2 loses its "keep quiet" marker.

"There is reason to believe, from animal studies, that assisted-reproductive technology can lead to more frequent imprinting errors," says Hopkins geneticist Andrew Feinberg. One suspect: the broth in which ova and embryos grow before being implanted in the mother's womb. It may somehow unsilence imprinted genes.

Epigenetics might also solve the puzzle of identical twins who do not have the same "genetic" diseases, especially psychiatric ones. "You wonder if the difference might be that something causes a gene related to mental illness to be silenced in one twin but not the other," says Dr. Potash.

In the old joke, a drunk searches for his lost keys under a streetlight, not because he dropped them there, but because the light is good. The search for genetic variants -- differences in DNA sequences -- underlying complex diseases is starting to look like that. Sequence variants are easy to find; the light's good there, so scientists have found more than a million sequence variants. But they don't correspond too well with genetically based complex diseases. No wonder the spotlight is turning from genetics to epigenetics, the pattern of gene silencing and activation.