Genome Sequences Riddled With Errors

The human genome is difficult enough to decipher when its DNA sequences are transcribed properly and nearly impossible to decode when they’re not. Peter Forster, a geneticist at the University of Cambridge, says mistakes happen a lot—between 60 percent and 70 percent of published studies of sequences of human mitochondrial DNA contain significant errors. Sometimes a single letter is wrong; sometimes entire columns have been transposed. “And those are just the errors that I can prove without getting sued—where I actually have the confirmation of the original author that indeed this was an error,” says Forster. “Obviously, that’s a low estimate because not everybody has gotten back to me, and there may well be a lot of errors that I haven’t picked up.”

Forster slogged through more than 120 publications and 23,000 individual DNA sequences searching for mistakes. He ran the data through a program that “mathematically reconstructs the evolutionary tree of the DNA types alive today.” Certain “non-treelike structures” alert the program to sequences that are biologically impossible. Sequences that are biologically possible but erroneous can sneak through, which is why Forster says his estimates are, if anything, low. The method is well suited to examining studies with mitochondrial DNA because the genes are passed straight from mother to child. “With most other bits of our genome, which are passed down from both parents and then recombined, you can’t do this kind of error-detection analysis,” says Forster. “And I’m worried that many other studies on other parts of the genome that are medically relevant might have a similar error rate. There’s no reason to suppose they shouldn’t.” The mistakes could waylay researchers looking for disease-causing mutations and affect court cases in which DNA is used as evidence.

The good news is that most of the errors occur when researchers interpret the data for submission to a journal, not during actual experiments in the lab. So Forster recommends an old-fashioned but effective solution: “Double-checking. The journals should warn authors that they may be asked to show some of their original data. The rate of error does warrant spot checks at the very least.” 

—Michael Abrams

Human Embryo Experimentation Sparks Outcry

A quarter century ago, British physicians Robert Edwards and Patrick Steptoe faced condemnation when they added an egg to sperm in a petri dish and created an embryo that became Louise Brown. That technique of in vitro fertilization now looks positively tame compared with the newest genetic manipulations. At the July meeting of the European Society for Human Reproduction and Embryology in Madrid, scientists were horrified—and transfixed—by two presentations: one that explored adding cells to developing embryos and another that outlined a process of growing egg cells from aborted human fetuses.

The first procedure is eerily like fusing two separate human beings. Norbert Gleicher, a gynecologist and the chairman of the Center for Human Reproduction in New York City, took 21 human female embryos and added male cells to each one. In 12 cases the cells were successfully integrated. Gleicher says he decided to use human embryos because many embryology techniques do not work in animal models. He inserted male cells because the Y chromosome is a genetic marker that can be tracked, making it easy to see where and how the injected cells integrate into the recipient female embryos. He describes his results as a small step down the road to treatments for genetic diseases and insists that ethical objections are ridiculous at such an early stage. “We simply wanted to determine whether transplantation in human embryos is possible,” he says. “We’re years from doing this kind of procedure in a clinical setting.” Nonetheless, many scientists aren’t waiting until late-phase trials to voice concerns.

“The scientific and ethical basis of these experiments is totally flawed,” says reproductive biologist Lynn Fraser, a past chairman of the European Society. “If you modify an embryo this way, you have no control over the relative proportions of the genetically defective cells and the normal cells in any baby that might result. You can’t direct the normal cells to go to the right organs to correct a particular genetic defect.”

Tal Biron-Shental, a gynecologist at Meir Hospital in Israel, also faced intense criticism when she described her team’s attempts to cultivate ovarian tissue from 22- to 33-week-old aborted fetuses. Some of the ovarian cells showed indications of maturing into eggs after a month but then stopped developing. Biron-Shental says the chemicals in which the cells were cultured proved to be the blocking agent. To take her research further, she will search for the right chemicals. Because donor eggs for infertile couples are in short supply, there is keen interest in eggs that could be grown in a lab; however, the implications of growing eggs from aborted tissue loom large. While Biron-Shental thinks the procedure will be accepted, some colleagues disagree. “The use of fetal ovarian tissue for this purpose raises many social, ethical, legal, and scientific concerns,” says Suzi Leather, the chairwoman of Britain’s Human Fertilization and Embryology Authority, a government agency that monitors the activities of fertility clinics. “It would be difficult for any child to come to terms with having been created by aborted fetuses.” The use of fetal ovarian tissue for fertility purposes has been banned in the United Kingdom since 1994.

Gleicher thinks acceptance is simply a matter of time. “Whenever anything new happens in this field, it’s always considered unethical at first,” he says.

—Elizabeth Svoboda

Zoologists Announce Aging Surprise

For a bird, Leach’s storm petrel lives a very long time—up to 30 years. The offshore bird’s secret, revealed for the first time in May by zoologists at Iowa State University, is in the storm petrel’s telomeres, repetitive bits of DNA that sit on the ends of the chromosomes in each cell like protective caps. Each time a cell divides in most animals, its chromosomes make a copy for the new cell. But not all of the telomere is copied each time; telomere caps tend to be shorter in the copies. Scientists suspect that each new generation gets a shorter telomere cap until finally the cells can divide no more. The process, they believe, results in aging. The storm petrel is different—its telomeres actually lengthen with age. “No other species we’ve looked at shows lengthening telomeres,” says Carol Vleck, an associate professor in the university’s department of ecology, evolution, and organismal biology and leader of the team. “We have a correlation here, but we don’t know cause and effect. We’d like to think that telomere lengthening does facilitate long life, but it’s probably just one factor. We’d like to study the albatross, another very long-lived bird.”

The researchers are continuing to study a storm petrel population of known age on Kent Island, New Brunswick. They hope to learn more about the relationship between the birds’ immune systems and an enzyme called telomerase, which maintains telomere length, extending the chromosome’s ability to replicate. Although telomerase could somehow slow aging, it is also found in most tumor cells, where it aids the uncontrolled growth that characterizes cancer.             

—Michael W. Robbins

Depression Gene Leads to Questions About Prozac

Links between genetics and mental illness led to several major discoveries this year: Researchers at Myriad Genetics in Salt Lake City reported finding a gene related to depression, and researchers at the University of California at San Diego found a flawed gene linked to manic depression. At the University of Pittsburgh, scientists located several chromosomal regions that may increase susceptibility to depression and addictive behavior. A large long-term study at New Zealand’s Dunedin School of Medicine demonstrated a genetic basis for the different ways that people respond to emotional stress—and in the process raised new questions about how antidepressant drugs actually work.

The Dunedin researchers, with scientists from the University of Wisconsin and King’s College London, tracked 847 New Zealanders from birth to age 26. Each underwent genetic testing that focused on how the 5-HTT gene, which instructs the body to create a protein to transport serotonin within the brain, was structured in different individuals. They found a link between 5-HTT and the likelihood of developing depression in response to such stressful events as long-term unemployment, homelessness, divorce, and physical or sexual abuse. The gene comes in pairs and in two sizes—long and short. Of people who had experienced four or more stressful events, those with long genes were half as likely to suffer depression as those with at least one short gene.

The findings contained a surprise about antidepressants like Prozac, Paxil, and Zoloft. The drugs are presumed to work by boosting serotonin levels. But short genes correspond with both higher levels of serotonin and depression. “This implies that these drugs actually don’t work by increasing serotonin levels, as we originally thought,” says Terrie Moffitt, a researcher on the study and a professor of psychology at the University of Wisconsin and King’s College London. “They may work by bringing about gradual changes in the structure of the [network of] neurons.”

This finding may help explain why antidepressants work only about half the time, and why they work for some people and not others. Still, Moffitt points out that genes aren’t the only factor in depression. She herself has two short versions of 5-HTT but so far has suffered no significant depression.  The reason? “No significant stress.”

—Annette Foglino

Biologists Reexamine Cause of the Black Death

What wiped out half the population of Europe in 1348? The Black Death, of course. But what exactly was the Black Death? Biologists have assumed for more than a century that the culprit was Yersinia pestis, or plague. Three years ago researchers at the University of the Mediterranean in France thought they had settled the matter when they examined three Black Death victims and found segments of Yersinia DNA in their teeth.

But this year Alan Cooper, head of the Ancient Biomolecules Centre at Oxford University, showed that the teeth were most likely contaminated with a modern—not a medieval—strain of Yersinia.

“It’s incredibly easy,” Cooper says, to test a long-dead corpse and find plague. “In fact, it’s almost impossible not to get a positive result when doing ancient DNA work because there’s so much contamination around. It’s incredibly difficult to get an authentic result.” Part of the problem, in Cooper’s view, is that the researchers are microbiologists, not “proper DNA researchers. They don’t have a laboratory set up appropriately, on the whole, and the very pathogen that they’re most interested in is probably already present in large amounts around the building.”

It’s not as if the French researchers hadn’t taken precautions. The biologists looked for a different segment of the Yersinia genome every time they examined a new tooth. “The main problem,” Cooper says, “is if you actually have Yersinia DNA in the building from any previous work in the lab—even 5, 10, 15 years before—it doesn’t matter what target you go for because the DNA from the actual bug itself will have all the genes that you might want to go looking for.”

To test his skepticism, Cooper examined 121 teeth from 66 Black Death victims. He used a technique in which the tooth is coated in silicone before it is sent to the lab. The coating, says Cooper, prevents “any nasties from getting off the tooth surface” and contaminating the interior when a sample is taken.

Cooper found no trace of Yersinia DNA. That doesn’t mean the bacteria didn’t cause the Black Death, however. “I’m still a traditionalist,” says Cooper. “I think it was Yersinia.” He simply doesn’t think the French have proved it.                

—Michael Abrams

73

Bygone Bull Birthed From Frozen DNA

When the calf struggled to his feet on April 1 and bellowed, “we all cheered like proud parents,” says geneticist Robert Lanza. That’s because the 40-pound male banteng is an animal like no other. It is the first healthy, viable clone of an endangered species and the first successful birth from a cross-species nuclear transfer.

Paradoxically, the reason for the cloning was to increase genetic diversity among the dwindling pool of surviving bantengs, a species of large, wild oxen once common in the forests and jungles of Southeast Asia. Lanza, vice president of medical and scientific development at Advanced Cell Technology in Worcester, Massachusetts, is one of the geneticists who proposed the idea. The genes came from the Zoological Society of San Diego’s Frozen Zoo, a collection of 35,000 preserved animal tissue samples, including skin cells from a male banteng that had been killed in a fight with another animal at the zoo in 1980.

“We didn’t know whether cells that old would even work,” says Lanza. Researchers transferred the banteng DNA to Angus cow eggs from which the nuclei had been removed. Using an activation protocol that required bathing the eggs in chemicals, thereby imitating the fertilization process, the team was able to generate 45 embryos. The embryos were sent overnight to Trans Ova Genetics in Sioux Center, Iowa, where they were implanted in 30 Angus cows. Of 16 pregnancies, two went full term, with cesarean births in early April. The first calf was normal; the second was born almost twice the normal size and euthanized after being judged unlikely to survive.

The surviving calf was transferred to the San Diego Zoo Wild Animal Park, where the staff has been hand-rearing it. Now weighing more than 350 pounds, it is nearing adulthood. At sexual maturity, it will be able to mate with female bantengs in the zoo. The offspring of that union, Lanza says, will be “100 percent banteng and not a hybrid at all.”

—Michael W. Robbins

Good Runners Have Good Proteins

What gives a great athlete superhuman speed? Genetics may make a crucial difference. Scientists at the Institute of Neuromuscular Research at Children’s Hospital in Sydney, Australia, have discovered a variant of a gene called alpha-actinin-3 that appears to help muscles contract faster and with greater force by making a protein called actinin in muscle fibers. Researchers took DNA samples of more than 300 athletes who represented Australia in the Olympics or other major competitions. Their DNA was compared with the genetic profiles of 400 randomly chosen people. Among elite sprinters, 95 percent were found to have the gene variant versus about 82 percent of the general population.

Long-distance runners seem to have a different kind of genetic edge. Athletes who excel in endurance sports are more likely to have another version of the gene—this one called the X variant—that does not produce actinin. Kathryn North, a neurologist and clinical geneticist who led the study, suspects that its absence may make muscles contract more slowly and absorb nutrients at a lower speed, which is better for endurance.

North cautions that genetics are only part of what makes a successful athlete. “It is still highly contentious whether we can use genetic markers to predict athletic performance,” she says. “There are any number of critical nongenetic factors that play a key role: coaching, equipment, competition, and serendipity. But every variable counts. And genetic information may help athletes make better-informed decisions about the likelihood of success in a particular sport.”

—Annette Foglino

Horse Racing May Never Be the Same

A horse was born this year with a most unusual pedigree. Cesare Galli and his colleagues at the Laboratory of Reproductive Technology in Cremona, Italy, announced the delivery in July of a 79-pound foal that has the distinction of being her mother’s twin sister. Prometea is not only the first cloned horse but also the first widely reported case of a cloned mammal whose genetic donor is her surrogate mother. Biologists had previously assumed that a mother’s immune system would reject a genetically identical fetus.

If cloning a horse were scored for difficulty, the feat would rank well above the more familiar mouse, rabbit, sheep, or pig. And the difficulty of cloning a mule might rank even higher. Horses usually release only one egg during ovulation, and they have an 11-month gestation. Mules, from the union of a horse and a donkey, are usually sterile because of the mismatched chromosomes they inherit. But in May Gordon Woods, a professor of animal and veterinary science at the University of Idaho, reported the birth of Idaho Gem, the world’s first cloned mule.

Racing fans are excited by the arrival of both Prometea and Idaho Gem. Don Jacklin, president of the American Mule Racing Association, who helped finance Woods’s research team, hopes cloning will reproduce prizewinning animals. Still, if the poor health and short life of Dolly the sheep is an indication, cloned equines will have the odds stacked against them.

—Eric Levin

Decaffing  Bean Genes

Nature puts caffeine in coffee beans, then humans go to great lengths to take it out. Before being roasted, green beans must be soaked in poisonous solvents, carbon dioxide, or repeated water baths to remove their jolt. But research teams around the world have been trying, through selective breeding or genetic manipulation, to coax the coffee plant into omitting the caffeine in the first place.

Last June researchers led by Shinjiro Ojita at the Nara Institute of Science and Technology in Japan announced they had created the first coffee plants genetically engineered to do just that. In a normal coffee plant, enzymes add methyl groups to a chemical called xanthosine, converting it into caffeine. Using a technique called RNA interference, Ojita’s team constructed transgenic coffee plants in which the gene that governs production of one of those enzymes is, in essence, turned off. The resulting plants have about 60 percent less caffeine. Beans from the plants are expected to show a similar decrease, but researchers won’t know for sure until this first crop matures and beans are harvested in about four years. Whether they will make a decent double chocolate latte is a different question entirely.

—Michael W. Robbins