For 9 months, doctors can at best make educated guesses about a growing fetus’ future as a healthy human. Those first fuzzy black-and-white ultrasound images can provide a mother with the peace of mind that she has a boy or girl with a beating heart, but many developmental maladies leave no visible fingerprint. Doctors can diagnose some genetic disorders by means of amniocentesis, in which they examine whole fetal cells extracted from the amniotic fluid that surrounds a baby in the womb. But the risk of complications from this invasive procedure, which requires inserting a long needle into a woman’s abdomen, can exceed the risk that something was wrong to begin with.
In the late 1990s, Dennis Lo, now at the Chinese University of Hong Kong, was studying a possible low-risk way to get a window on a fetus’ health by searching for fetal cells in pregnant women’s blood. But “the problem with fetal cells is that they are incredibly rare,” says Lo. One milliliter of a pregnant woman’s blood, which contains 6 million cells, might yield only one or two fetal cells.
But then a series of papers caught Lo’s eye. The papers examined how cancer tumors shed not just whole cells but also pieces of loose DNA, called cellfree DNA (cfDNA), into a patient’s blood. He wondered whether a developing fetus—which, like a tumor, is parasitic on its host and consists of rapidly dividing cells—might also shed bits of DNA. He ultimately proved that it did by showing that he could detect fragments of Y chromosomes, which only males have, in the blood of pregnant women carrying boys.
That research and the work that followed led to the rapid development, by independent companies, of tests that use cfDNA to determine a fetus’ sex and blood type. Doctors now use these tests, which give accurate results weeks before an ultrasound exam can, to screen fetal sex in women carrying fetuses at risk of sex-linked genetic disorders such as hemophilia and muscular dystrophy (SN: 7/22/06, p. 56). Scientists have even discovered that, by measuring levels of cfDNA in a mother’s blood, they can predict premature births and preeclampsia, a disease that causes dangerously high blood pressure in pregnant women. But the tests have limits: Scientists can be sure they are detecting a fetal gene only if they know that the mother doesn’t possess the same gene herself.
Researchers seeking to refine and expand this kind of testing have now turned their attention to another part of a fetus’ genetic material, called messenger RNA (mRNA). By looking for bits of mRNA in a mother’s blood, scientists are learning what genes a fetus expresses as it grows and develops. This new perspective, they hope, will lead to prenatal tests for a plethora of developmental disorders.
The 9-month picture
In the first few days of pregnancy, some cells split from the fetus to form the placenta, the interface between mother and fetus. But cells constantly detach from the fringes of the placenta, says Lo. As they break apart, these cells release DNA, which can pass through to the mother’s side of the placenta and enter her bloodstream. The cfDNA lasts about 15 minutes before it degrades.
The biggest limitation to prenatal testing based on fetal cfDNA is the difficulty of distinguishing fetal cfDNA from maternal cfDNA. Scientists can focus on genes that the fetus inherited from its father, which wouldn’t normally be found in the mother, or they can look for the so-called silenced forms of specific genes that are expressed in a normal adult but not in a fetus.
For these reasons, fetal-cfDNA tests can diagnose only a few genetic diseases. For example, a fetus will develop cystic fibrosis if it inherits from both parents mutations in a single gene. The parental mutations need not be the same, however. If mother and father carry different mutations, finding the father’s version of the gene in the mother’s blood means that the fetus has inherited the paternal flaw. But for most genetic diseases, a fetus needs to inherit identical mutations from both parents. This means that the mother would already have that mutated DNA circulating in her blood, making it hard to distinguish the fetal DNA from her own.
“For most genes, you’ve got a problem because the mother has the gene and the baby has that same gene too,” says Lo. One way around this difficulty is to look for mRNA instead of DNA. Cells generate mRNA from a gene’s DNA only when they need to manufacture proteins. Many genes involved in fetal development aren’t functional in an adult, so mRNA from them would never show up in adult cells. For example, most genes that tell the brain how to grow are needed only as a fetus develops, so mRNA for these developmental genes wouldn’t be circulating in an adult.
In the October Journal of Clinical Investigation, Diana Bianchi and Jill Maron of the Tufts–New England Medical Center in Boston describe experiments in which they identified more than 100 fragments of fetal mRNA in maternal blood. The researchers looked for mRNA in blood samples taken from nine women before and after they gave birth in order to pinpoint genes that were in the mothers’ blood before delivery but not afterward. To be sure of the origin of the mRNA, they then checked what they found against mRNA in blood taken from the newborn babies.
Many of the genes expressed just before birth, Bianchi and Maron found, mediate development of the neural system. Others allow vision and other senses to function. For a baby about to open its eyes to light for the first time, this makes sense, says Maron. At the point of being born, a baby needs to ramp up the production of everything that will help it face the world.
The researchers hope to pursue this strategy by creating a timeline of normal fetal-gene expression through the whole 9 months of pregnancy. They would then compare this benchmark with fetal-gene expression throughout the pregnancies of women whose babies are known to have certain genetic disorders.
“Since we’ve started to develop what a normal fetus looks like at term, what if we look at an abnormal fetus at term? Or what if we move back to the second term?” says Maron. “We want to see what’s possible. You could take a mother’s blood and see how a fetus’ brain is developing, or its eyes.”
If doctors can track the activity of genes during fetal growth, Bianchi says, they will know, simply by testing a drop of blood, as soon as something abnormal shows up, rather than having to wait until they can see something physically wrong on an ultrasound scan.
“The ultimate goal, of course, is fetal treatment,” says Bianchi.
Michael Katz, senior vice president of research and global programs at the March of Dimes in White Plains, N.Y., says that this technique holds immeasurable promise for prenatal diagnosis of rare conditions that doctors currently have no tests for. But Katz cautions against using new tests clinically before they are as accurate as amniocentesis.
“They need a vertical expansion of their technique—more precision—and they need horizontal expansion—more things it can test for,” says Katz. “It’s long term. I don’t mean 20 years, but maybe 5.”
Scientists have long been stymied in their attempts to find a noninvasive way to detect Down syndrome, which affects about 1 in 700 live births worldwide. Caused by the presence of a third copy of chromosome 21—a phenomenon known as trisomy-Down syndrome can be definitively diagnosed only by amniocentesis or by the slightly riskier chorionic villus sampling, which involves examining cells taken from the placenta.
“Initially, many people suggested we could never crack trisomy with cellfree DNA,” says Lo. “Diagnosing a trisomy, by definition, is counting the number of chromosomes in a cell. And if we don’t have the entire cell, just DNA swimming outside it, then how can you do that?”
Lo found a solution, however, when he turned to cellfree mRNA instead of DNA. His results appeared in the February Nature Medicine.
If a fetus has three copies of chromosome 21, Lo figured, it must have acquired two copies from one parent and one from the other. Lo identified a gene, PLAC4, on chromosome 21 that is expressed only in placental tissue—so that any PLAC4 mRNA in a pregnant woman’s blood must come from the placenta. What gives PLAC4 its diagnostic value, however, is that it comes in two distinct forms that differ at just one location on the gene. About half the population has one form of PLAC4, half the other.
Putting together these ingredients, Lo devised a way to test for Down syndrome by examining fetal mRNA in maternal blood. If a mother and father have different versions of PLAC4, then finding equal amounts of the corresponding mRNAs in the mother’s blood would mean that the fetus is normal, with two copies of chromosome 21. But if there is twice as much of one version as of the other, the fetus must have three copies of chromosome 21—and Down syndrome.
Lo tried this test on 67 pregnant women. Ten of the women carried fetuses already known to have Down syndrome while the rest were known to have healthy fetuses. The blood-based test detected 9 of the 10 cases of Down syndrome.
Although this test doesn’t work for fetuses whose parents both have the same version of PLAC4, future adaptations of the test could take advantage of other single-spot variations in the PLAC4 gene. Lo says that a test combining many such hot spots would not only extend its applicability to a larger proportion of the population but would also more accurately diagnose Down syndrome and other syndromes related to extra chromosomes.
Katz, however, worries that such a test still would miss out on disorders that arise when a fetus has an extra part of a chromosome but not the whole chromosome. Six percent of Down syndrome cases, for example, are caused by chromosome 21 being duplicated only in part.
“There’s a whole range of other rare birth defects caused by chromosomal abnormalities that depend on repeats,” Katz says. Lo responds that as long as a gene like PLAC4 can be found on the part of chromosome 21 that’s duplicated, the test would still work.
The hope, then, is that as the science advances, increasingly sophisticated blood-based tests will allow doctors to peer through the veil of mystery that obscures the developing fetus. While some bits of information may ruin surprises for moms and dads, other tests will confirm the health of their unborn baby—something every parent wants.
Questioning the Value of Knowledge
New tests raise ethical concerns
The idea that a finger prick can reveal whether a fetus is a boy or a girl and whether it is afflicted with any genetic maladies doesn’t appeal just to parents and doctors. A number of companies see a commercial future in simple, at-home genetic tests.
Ethicists, though, worry that these companies tend to stretch the limits of science. Baby Gender Mentor, marketed by Acu-Gen Biolab, Inc. in Lowell, Mass., is a blood-based cfDNA test to determine fetal sex. The company calls its product “the most accurate DNA gender test” and boasts 99.9% accuracy. But more than 100 women are involved in a class action lawsuit, stemming from false results, that accuses Acu-Gen Biolab of fraud and misrepresentations.
Bioethicist Gail H. Javitt of the Johns Hopkins University says that the problem with products such as this is that they’re not regulated enough. Baby Gender Mentor and other tests like it aren’t subject to regulation by the U.S. Food and Drug Administration, so it’s possible that these types of tests can be developed into commercial products before the science behind them has been fully validated.
“There’s so much information coming out of the human genome with the potential to be beneficial to disease treatment and prevention,” Javitt says. “But the regulatory system just hasn’t kept up with this dramatic change.”
As scientists explore the wide horizon of mRNA-based prenatal tests, these regulatory challenges are bound to emerge repeatedly. Although the new generation of tests offers the possibility of accurately diagnosing a broad range of developmental disorders, it also brings forth challenging ethical questions on what information expectant parents ought to know—and what to do with this information.